**Part 3**

**Understanding Immunological Aspects of HIV in an Infected Person** 

148 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Yurin O., Kravchenko A., Afonina L., et al. (2001). Nikavir – a novel domestic antiretroviral

Zakharova N., Kuptsov D., Mishustina Y., et al. (2008). Experience of nikavir (phosphazide)

application in chemoprevetion of the vertical HIV transmission. *Infectious Diseases*,

agent. Epidemiology and Infectious diseases, № 1, p. 43-45.

Anthology, St.Petersburg, 2007, p. 93-98.

**9** 

*Russia* 

**Natural Catalytic Antibodies** 

Georgy A. Nevinsky

 **in Norm and in HIV-Infected Patients** 

*Institute of Chemical Biology and Fundamental Medicine of SD of RAS* 

Antibodies (Abs) have been first characterized as proteins which are produced by the immune system and have a sole function of binding other molecules, called antigens, with the goal of eliciting an immune response. In this classical conception, Abs act similarly to enzymes in specific binding other molecules. However, in contrast to enzymes they do not have the ability to catalyze chemical conversions of their bound partners. For the vast majority of Abs, this observation is correct. However, in a 1946 consideration of enzyme function, Linus Pauling first hypothesized that the active center of an enzyme is closely juxtaposed to a "strained configuration" of its substrate (that is, targeted against the structure of the transition state) rather than to the native conformation of the substrate molecule (Pauling, 1946). This idea led Jencks in 1969 to propose that Abs generated in an anti-hapten immune response against chemically stable analogs of the transition-state of a

reaction of interest could potentially display an enzymatic activity (Jencks, 1969).

anti-hapten catalytic Abs were termed abzymes (derived from **a**nti**b**ody en**zyme**).

activity (Barbas et al., 1997; Janda et al., 1997; Keinan, 2005, and refs therein).

In 1985, a general method for generating catalytic monoclonal Abzs against transition state analogs, and a way to use those Abs to accelerate chemical reactions, was first described (Schochetman & Massey, 1985). One year later two groups were able to produce the first monoclonal Abs with catalytic properties, which were generated against hapten analogs of the transition states for *p*-nitrophenylphosphorylcholine (Pollack et al., 1986) or for monoaryl phosphonate esters (Tramontano et al., 1986a, 1986b). The artificial monoclonal

The evolution of the technology of artificial Abzs during the last two decades has led not only to the rapid development of direct approaches for the generation of Abs with specified properties, but also to the creation of strategies to revise the targeting specificity of individual Abzs. Such modifications of antigen binding specificity can be achieved genetically *in vitro*, by application of the site-directed mutagenesis, or genetic selection or screening (using approaches such as phage display). Alternatively, modification can be induced directly on purified antibody, *via* selective chemical modification by direct introduction of catalytic groups into the Ab combining site. Some studies describing these approaches include (Keinan, 2005, and refs therein). The employment of the approaches have demonstrated that the substrate specificity (and/or the specific activity) of some artificial Abzs is comparable to or even higher than that of enzymes with the same catalytic

**1. Introduction** 

## **Natural Catalytic Antibodies in Norm and in HIV-Infected Patients**

Georgy A. Nevinsky *Institute of Chemical Biology and Fundamental Medicine of SD of RAS Russia* 

## **1. Introduction**

Antibodies (Abs) have been first characterized as proteins which are produced by the immune system and have a sole function of binding other molecules, called antigens, with the goal of eliciting an immune response. In this classical conception, Abs act similarly to enzymes in specific binding other molecules. However, in contrast to enzymes they do not have the ability to catalyze chemical conversions of their bound partners. For the vast majority of Abs, this observation is correct. However, in a 1946 consideration of enzyme function, Linus Pauling first hypothesized that the active center of an enzyme is closely juxtaposed to a "strained configuration" of its substrate (that is, targeted against the structure of the transition state) rather than to the native conformation of the substrate molecule (Pauling, 1946). This idea led Jencks in 1969 to propose that Abs generated in an anti-hapten immune response against chemically stable analogs of the transition-state of a reaction of interest could potentially display an enzymatic activity (Jencks, 1969).

In 1985, a general method for generating catalytic monoclonal Abzs against transition state analogs, and a way to use those Abs to accelerate chemical reactions, was first described (Schochetman & Massey, 1985). One year later two groups were able to produce the first monoclonal Abs with catalytic properties, which were generated against hapten analogs of the transition states for *p*-nitrophenylphosphorylcholine (Pollack et al., 1986) or for monoaryl phosphonate esters (Tramontano et al., 1986a, 1986b). The artificial monoclonal anti-hapten catalytic Abs were termed abzymes (derived from **a**nti**b**ody en**zyme**).

The evolution of the technology of artificial Abzs during the last two decades has led not only to the rapid development of direct approaches for the generation of Abs with specified properties, but also to the creation of strategies to revise the targeting specificity of individual Abzs. Such modifications of antigen binding specificity can be achieved genetically *in vitro*, by application of the site-directed mutagenesis, or genetic selection or screening (using approaches such as phage display). Alternatively, modification can be induced directly on purified antibody, *via* selective chemical modification by direct introduction of catalytic groups into the Ab combining site. Some studies describing these approaches include (Keinan, 2005, and refs therein). The employment of the approaches have demonstrated that the substrate specificity (and/or the specific activity) of some artificial Abzs is comparable to or even higher than that of enzymes with the same catalytic activity (Barbas et al., 1997; Janda et al., 1997; Keinan, 2005, and refs therein).

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 153

HIV-1 is the etiologic agent of an extremely dangerous human disease, AIDS (Fauci et al., 2008, and refs therein). The association of immune dysfunction in patients with HIV infection and AIDS and the development of AI diseases are very interesting. At this moment the spectrum of reported autoimmune phenomena in AIDS patients is increasing (for review see Zandman-Goddard & Shoenfeld, 2002). A special feature of ADs is high concentrations of auto-Abs (Abs to many different endogenous antigens) (Zouali, 2001; Pisetsky, 2001). The development of AI diseases is characterized by spontaneous generation of primary Abs to proteins, nucleic acids and their complexes, polysaccharides, nucleotides etc. (Earnshaw & Rothfield, 1985; Raptis & Menard, 1980). Later the secondary idiotypic and then antiidiotypic Abs to the primary ones are usually generated, etc. Immunization of animals with DNA or RNA and especially their complexes with proteins leads to the production of

During frank loss of immunocompetence, AI diseases that are predominantly T cell subtype CD8 driven predominate. There is evidence for B cell stimulation and many auto-Abs are reported in HIV patients. HIV-dependent activation of B lymphocytes leads to the production of auto-Abs not only to different viral proteins including HIV reverse transcriptase (RT) and integrase (IN), but also to human cell components, and various immune complexes including anti-cardiolipin, anti-beta2 GPI, anti-DNA, anti-small nuclear ribonucleoproteins, anti-thyroglobulin, anti-thyroid peroxidase, anti-myosin, and antierythropoietin and possibly other human cell and blood components (Fauci et al., 2008; Zandman-Goddard & Shoenfeld, 2002)*.* The list of reported autoimmune diseases in HIV/AIDS includes SLE, anti-phospholipid syndrome, vasculitis, primary biliary cirrhosis, polymyosits, Graves' disease, and idiopathic thrombocytopenic purpura (Zandman-Goddard & Shoenfeld, 2002). The presence of AI phenomena and production of auto-Abs in chronic bacterial and viral infections including HIV could be related to molecular mimicry between microbial or viral and host antigens (Zandman-Goddard & Shoenfeld, 2002; Hentati et al., 1994; Ternynck et al., 1991), altered self, abnormal expression of

**2. Features of the immune status of patients with AIDS, bacterial, and** 

anti-DNA and anti-RNA Abs (Gottieb & Shwartz, 1972; Mitsuhashi et al., 1978).

immunoregulatory molecules, and the anti-idiotypic network (Barzilai et al., 2008).

Sato et al., 1994; Hirokawa et al., 2002; Ikeda et al., 2003).

Interestingly, some other viral infections similarly to ADs can also demonstrate AI reactions leading to the formation of Abs to various human and viral antigens. Compared with healthy donors, concentrations of DNA and anti-DNA Abs are higher not only in patients with SLE (36% of SLE patients), but also in multiple sclerosis (17-18%), primary Sjogren's syndrome (18%), Hashimoto's thyroiditis (23%), myasthenia gravis (6%), rheumatoid arthritis (7%), autoimmune hepatitis (Shoenfeld et al., 1988, 1989), and also in lymphoproliferative (Kozyr et al., 1998) and some viral diseases including viral hepatitis, AIDS (Gololobov et al., 1994), and TBE (Garmashova et al., 2004). In the sera of patients with several AI diseases, RNA and anti-RNA Abs were also detected (Blanco et al., 1991;

AIDS, TBE, and hepatitis demonstrating strong reorganization of immune system have some similarities with typical AI diseases such as SLE (HT, and others) which is a systemic AI polyetiologic diffuse disease that is characterized by disorganization of conjunctive tissues with the paramount damage to skin and visceral capillaries (Hhachn, 1996). All known AI and viral diseases like AIDS, viral hepatitis, and TBE are characterized by significant disturb the immune status of the patients accompanied by humoral and cellular

**autoimmune diseases** 

Artificial Abzs against transition chemical states of different reactions have been studied intensively (Thayer et al., 1999). Mechanistic basis for the activity of such Abzs is becoming well understood (Janda et al., 1997; Thayer et al., 1999; Keinan, 2005, and refs therein). The field of artificial Abzs has been amply reviewed recently (Martin & Schultz, 1999; Suzuki, 1994; Keinan, 2005, and refs therein), for more detailed description of the relevant reactions.

During last two decades it has become clear that auto-Abs from the sera of patients with different autoimmune (AI) diseases can possess enzymic activities (Suzuki, 1994; Keinan, 2005; Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). The first example of a natural Abz was an IgG found in bronchial asthma patients, which hydrolyzes intestinal vasoactive peptide (VIP) (Paul et al., 1989), the second was an IgG with DNase activity in SLE (Shuster, et al., 1992), and the third was an IgG with RNase activity in SLE (Buneva et al., 1994). Later, different natural catalytic IgG and/or IgA, IgM hydrolyzing oligopeptides, proteins, DNA, RNA, nucleotides, and polysaccharides were detected in the sera of patients with several autoimmune (AI) and viral pathologies, and Abzs with these and other different activities were discovered in healthy human milk (for review see Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b).

The phenomenon of catalysis by auto-Abzs is extremely interesting and potentially applicable in many different fields including new types of efficient catalysts, new generation of drugs, evaluation of the functional roles of Abzs in innate and adaptive immunity, and understanding of certain aspects of self-tolerance and of the destructive or positive responses in AI diseases. The field of monoclonal Abzs with immunotherapeutic potential has recently been reviewed (Wentworth et al., 1998; Tellier, 2002; Zhou et al., 2002; Nishi, 2003; Stockwin and Holmes, 2003; Hanson et al., 2005; Gabibov et al., 2006; Planque et al., 2008; Wójcik & Kieć-Kononowicz, 2008). Some general possibilities of present and future therapeutic Abs and Abzs application were discussed in (Stockwin & Holmes, 2003). Abs and Abzs can be used to neutralize pathogens, toxins and endogenous mediators of pathology. As cell-targeting reagents, Abs can be used to modulate cytoplasmic cascades or to tag specific cells for complement- or effector-mediated lysis. Abs can also be modified to deliver toxic or modulatory payloads (small molecules, radionuclides and enzymes) and engineered to bind multiple epitopes or even to have novel catalytic activity. The modular structure of Igs and the availability of Ab fragment libraries also make it possible to produce variable-domain therapeutics (Fab, single-chain and domain of Abs). Although exhibiting less favorable kinetics *in vivo*, these fragments are straightforward to express and easily penetrate tissues, making them especially useful as neutralizing or delivery agents. The number of approved Abzs is expected to increase in the near term, as the platform is adopted as a viable alternative to small molecule discovery (Stockwin & Holmes, 2003). The Abzs strategy can be employed for new methods of drug synthesis, as well as for in vivo therapies. Catalytic antibodies seem to be a promising tool for therapeutic purposes, because of their specificity and stereoselectivity. For instance, cocaine-hydrolyzing Abzs have been developed, and may provide a novel approach to the problems of drug addiction (Hanson et al., 2005). Possible application of Abzs for prodrug activation and their potential utility in clinical oncology was also discussed (Nishi, 2003). Abzs have two distinct advantages over canonical enzymes: first, they can be selected to perform reactions not catalyzed by endogenous enzymes, and second, they can be humanized to minimize their immunogenicity (Stockwin & Holmes, 2003).

Artificial Abzs against transition chemical states of different reactions have been studied intensively (Thayer et al., 1999). Mechanistic basis for the activity of such Abzs is becoming well understood (Janda et al., 1997; Thayer et al., 1999; Keinan, 2005, and refs therein). The field of artificial Abzs has been amply reviewed recently (Martin & Schultz, 1999; Suzuki, 1994; Keinan, 2005, and refs therein), for more detailed description of the

During last two decades it has become clear that auto-Abs from the sera of patients with different autoimmune (AI) diseases can possess enzymic activities (Suzuki, 1994; Keinan, 2005; Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). The first example of a natural Abz was an IgG found in bronchial asthma patients, which hydrolyzes intestinal vasoactive peptide (VIP) (Paul et al., 1989), the second was an IgG with DNase activity in SLE (Shuster, et al., 1992), and the third was an IgG with RNase activity in SLE (Buneva et al., 1994). Later, different natural catalytic IgG and/or IgA, IgM hydrolyzing oligopeptides, proteins, DNA, RNA, nucleotides, and polysaccharides were detected in the sera of patients with several autoimmune (AI) and viral pathologies, and Abzs with these and other different activities were discovered in healthy human milk (for review see

The phenomenon of catalysis by auto-Abzs is extremely interesting and potentially applicable in many different fields including new types of efficient catalysts, new generation of drugs, evaluation of the functional roles of Abzs in innate and adaptive immunity, and understanding of certain aspects of self-tolerance and of the destructive or positive responses in AI diseases. The field of monoclonal Abzs with immunotherapeutic potential has recently been reviewed (Wentworth et al., 1998; Tellier, 2002; Zhou et al., 2002; Nishi, 2003; Stockwin and Holmes, 2003; Hanson et al., 2005; Gabibov et al., 2006; Planque et al., 2008; Wójcik & Kieć-Kononowicz, 2008). Some general possibilities of present and future therapeutic Abs and Abzs application were discussed in (Stockwin & Holmes, 2003). Abs and Abzs can be used to neutralize pathogens, toxins and endogenous mediators of pathology. As cell-targeting reagents, Abs can be used to modulate cytoplasmic cascades or to tag specific cells for complement- or effector-mediated lysis. Abs can also be modified to deliver toxic or modulatory payloads (small molecules, radionuclides and enzymes) and engineered to bind multiple epitopes or even to have novel catalytic activity. The modular structure of Igs and the availability of Ab fragment libraries also make it possible to produce variable-domain therapeutics (Fab, single-chain and domain of Abs). Although exhibiting less favorable kinetics *in vivo*, these fragments are straightforward to express and easily penetrate tissues, making them especially useful as neutralizing or delivery agents. The number of approved Abzs is expected to increase in the near term, as the platform is adopted as a viable alternative to small molecule discovery (Stockwin & Holmes, 2003). The Abzs strategy can be employed for new methods of drug synthesis, as well as for in vivo therapies. Catalytic antibodies seem to be a promising tool for therapeutic purposes, because of their specificity and stereoselectivity. For instance, cocaine-hydrolyzing Abzs have been developed, and may provide a novel approach to the problems of drug addiction (Hanson et al., 2005). Possible application of Abzs for prodrug activation and their potential utility in clinical oncology was also discussed (Nishi, 2003). Abzs have two distinct advantages over canonical enzymes: first, they can be selected to perform reactions not catalyzed by endogenous enzymes, and second, they can be humanized to minimize their

relevant reactions.

Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b).

immunogenicity (Stockwin & Holmes, 2003).

## **2. Features of the immune status of patients with AIDS, bacterial, and autoimmune diseases**

HIV-1 is the etiologic agent of an extremely dangerous human disease, AIDS (Fauci et al., 2008, and refs therein). The association of immune dysfunction in patients with HIV infection and AIDS and the development of AI diseases are very interesting. At this moment the spectrum of reported autoimmune phenomena in AIDS patients is increasing (for review see Zandman-Goddard & Shoenfeld, 2002). A special feature of ADs is high concentrations of auto-Abs (Abs to many different endogenous antigens) (Zouali, 2001; Pisetsky, 2001). The development of AI diseases is characterized by spontaneous generation of primary Abs to proteins, nucleic acids and their complexes, polysaccharides, nucleotides etc. (Earnshaw & Rothfield, 1985; Raptis & Menard, 1980). Later the secondary idiotypic and then antiidiotypic Abs to the primary ones are usually generated, etc. Immunization of animals with DNA or RNA and especially their complexes with proteins leads to the production of anti-DNA and anti-RNA Abs (Gottieb & Shwartz, 1972; Mitsuhashi et al., 1978).

During frank loss of immunocompetence, AI diseases that are predominantly T cell subtype CD8 driven predominate. There is evidence for B cell stimulation and many auto-Abs are reported in HIV patients. HIV-dependent activation of B lymphocytes leads to the production of auto-Abs not only to different viral proteins including HIV reverse transcriptase (RT) and integrase (IN), but also to human cell components, and various immune complexes including anti-cardiolipin, anti-beta2 GPI, anti-DNA, anti-small nuclear ribonucleoproteins, anti-thyroglobulin, anti-thyroid peroxidase, anti-myosin, and antierythropoietin and possibly other human cell and blood components (Fauci et al., 2008; Zandman-Goddard & Shoenfeld, 2002)*.* The list of reported autoimmune diseases in HIV/AIDS includes SLE, anti-phospholipid syndrome, vasculitis, primary biliary cirrhosis, polymyosits, Graves' disease, and idiopathic thrombocytopenic purpura (Zandman-Goddard & Shoenfeld, 2002). The presence of AI phenomena and production of auto-Abs in chronic bacterial and viral infections including HIV could be related to molecular mimicry between microbial or viral and host antigens (Zandman-Goddard & Shoenfeld, 2002; Hentati et al., 1994; Ternynck et al., 1991), altered self, abnormal expression of immunoregulatory molecules, and the anti-idiotypic network (Barzilai et al., 2008).

Interestingly, some other viral infections similarly to ADs can also demonstrate AI reactions leading to the formation of Abs to various human and viral antigens. Compared with healthy donors, concentrations of DNA and anti-DNA Abs are higher not only in patients with SLE (36% of SLE patients), but also in multiple sclerosis (17-18%), primary Sjogren's syndrome (18%), Hashimoto's thyroiditis (23%), myasthenia gravis (6%), rheumatoid arthritis (7%), autoimmune hepatitis (Shoenfeld et al., 1988, 1989), and also in lymphoproliferative (Kozyr et al., 1998) and some viral diseases including viral hepatitis, AIDS (Gololobov et al., 1994), and TBE (Garmashova et al., 2004). In the sera of patients with several AI diseases, RNA and anti-RNA Abs were also detected (Blanco et al., 1991; Sato et al., 1994; Hirokawa et al., 2002; Ikeda et al., 2003).

AIDS, TBE, and hepatitis demonstrating strong reorganization of immune system have some similarities with typical AI diseases such as SLE (HT, and others) which is a systemic AI polyetiologic diffuse disease that is characterized by disorganization of conjunctive tissues with the paramount damage to skin and visceral capillaries (Hhachn, 1996). All known AI and viral diseases like AIDS, viral hepatitis, and TBE are characterized by significant disturb the immune status of the patients accompanied by humoral and cellular

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 155

The origin of natural Abzs in different AI, viral and bacterial diseases may be complex. Similarly to artificial Abzs against analogs of transition states of catalytic reactions, naturally occurring Abzs with DNase and RNase activities may be Abs raised directly against free DNA and RNA or these nucleic acids acting as haptens bound to different proteins and resembling transition states of catalytic reactions (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein). Immunization of rabbits with pure DNA and RNA generated Abs interacting with DNA and possessing weak DNase and RNase activities (Krasnorutskii et al., 2008a, 2008b). Many SLE anti-DNA Abs are directed against histone-DNA nucleosomal complexes appearing as a result of internucleosomal cleavage during apoptosis (Founel & Muller, 2002). Apoptotic cells are the primary source of antigens and immunogens in SLE, and certain features in recognition, processing, and/or presentation of apoptotic autoantigens by antigen-presenting cells can trigger AI processes (Founel & Muller, 2002). Anti-DNA–protein and anti-RNA–protein complexes and other antinuclear antibodies were found in the sera of patients with multisystem connective tissue disease (Gottieb & Shwartz, 1972). Therefore, we have emulated such natural complexes using complexes of DNA and RNA with methylated bovine serum albumin (mBSA). Immunization of rabbits with complex of DNA and RNA with mBSA elicited production of 10-50-fold more active DNase and RNase IgGs, while pIgGs from animals immunized with mBSA were catalytically inactive (Krasnorutskii et al., 2008a, 2008b). Immunization of healthy rabbits with DNase I, DNase II, and pancreatic RNase A also produced anti-idiotypic IgGs with intrinsic DNase and RNase activities (Krasnorutskii et al., 2008c, 2008d, 2009). Thus, DNase and RNase Abzs in different AI diseases may be a cocktail of Abs against complexes of DNA and RNA with proteins and antiidiotypic Abzs to very different DNA- and RNA-

Healthy humans and patients with many diseases with insignificant AI reactions usually lack Abzs or develop Abzs with very low catalytic activities, often on a borderline of the sensitivity of detection methods (Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Natural Abzs hydrolyzing DNA and RNA are described from the sera of patients with several AI (SLE; Shuster, et al., 1992; Buneva et al., 1994; Andrievskaya et al., 2000, 2002; Vlassov et al., 1998), Hashimoto's thyroiditis and polyarthritis (Vlasov et al., 1998), multiple sclerosis (MS) (Baranovskii et., 1998, 2001; Nevinsky et al., 2001), asthma (Galvita et al., 2007), and viral and bacterial diseases: viral hepatitis (Baranovskii et al., 1997; Vlasov et al., 1999), tick bone encephalitis (TBE; Parkhomenko et al., 2010), AIDS (Odintsova et al., 2006a), and several diseases caused by different bacterial infections (Parkhomenko et al., 2009). It was shown, that like in the case of AI-patients, IgGs with DNase activity from autoimmune mice are the earliest and statistically significant markers of pathology and these activities are detectable at the pre-disease stage, when there are no visible markers of SLE pathology or significant proteinuria, and anti-DNA titres are within the typical ranges of these indicators for healthy mice (Dubrovskaya et al., 2003; Andryushkova et al., 2007;

Using different approaches convincing evidence was provided that, similarly to Abzs from SLE and MS patients (Savel'ev et al., 2003; Ivanen et al., 2002, 2004), amylase activity is intrinsic to autoimmune mouse polyclonal IgGs (Andryushkova et al., 2006, 2007). It was shown that the relative activities of IgGs from MRL-lpr/lpr mice in the hydrolysis of DNA, ATP, and polysaccharides correlate very well with some visible (pink spots, baldness of the head and parts of the back, general health deterioration, etc.) and biochemical (proteinuria, Ab titers to native and denatured DNA) markers of AI pathologies during various stages of

hydrolyzing enzymes.

Kuznetsova et al., 2007).

AI reactions, with detectable tissue-specific and organ-nonspecific Abs (Bigazzi, 1983; Sugiama & Yamamota, 1996; Nevinsky, 2010b). At the same time, microbial and viral infections expose the human organism to different components from the parasite's cells and viral particles, including protein, DNA, RNA, lipids, and polysaccharides. The sera of mice infected with different microbe's bacterial pathogens contain a variety of Abs to the parasite's antigens and to human lipids, proteins, and nuclear components, including anti-DNA Abs (Ternynck et al., 1991; Unterkircher et al., 1993; Matsiota-Bernard et al., 1993, 1996; Hentati et al., 1994; Boekel et al., 1997; Wun et al., 2001). The origin of anti-DNA Abs in the infections remains speculative; some of them may arise inadvertently in the course of a normal immune response due to the induction by Abs that bear structures (mimotopes) mimicking DNA (Wun et al., 2001). The immunoregulatory effect of the infection seems to be related, at least partially, to the increase in a particular population of Abs, the polyreactive Abs (Matsiota-Bernard et al., 1996). It has been proposed that bacterial and viral infectious agents can act in some cases through the mechanism of molecular mimicry and stimulate development of different AI diseases. For instance, the agents responsible for molecular mimicry in multiple sclerosis include measles, hepatitis B, herpes simplex, influenza, papilloma, and Epstein-Bar viruses (Steinman, 2001). Thus, AI reactions in different AI, viral and bacterial infection diseases are very similar and may be strongly associated.

#### **3. The origin of artificial and natural abzymes**

Artificial Abzs can be obtained by immunization of animals with chemically stable analogs of transition states of chemical reactions (reviewed in Martin & Schultz, 1999; Nevinsky et al., 2000a; Tanaka, 2002; Tanaka and Barbas, 2002; Dias et al., 2002; Keinan, 2005). On the other hand, artificial antiidiotypic Abs can also possess catalytic activity (Barbas et al., 1997; Wentworth et al., 1998). Building on earlier observations on the existence of idiotypic determinants related to the antigen, Jerne proposed that the immune system is selfregulated by a network of idiotype–anti-idiotype interactions (Jerne, 1974). The simplified model of this network may be schematically presented as follows:

Antibodies 1 and 2 are termed idiotype and anti-idiotype, respectively, etc. There is convincing evidence that such idiotype–anti-idiotype networks are actually present in the body. The presence of blood serum Ab4 (in the notation shown in the scheme) has been confirmed in experimental animals (Jerne, 1974).

If the active site of an enzyme plays the role of antigen triggering this anti-idiotypic chain, it is logical to suggest that the secondary anti-idiotypic Ab2 may possess the structure, a part of which represents an "internal image" or "mould" of the active site of this enzyme, and, consequently, these Abs may possess some properties of this enzyme. This remarkable property of idiotypic mimicry has been exploited to raise monoclonal antiidiotypic Abzs with several different catalytic activities (reviewed in Keinan, 2005; Nevinsky et al., 2005).

AI reactions, with detectable tissue-specific and organ-nonspecific Abs (Bigazzi, 1983; Sugiama & Yamamota, 1996; Nevinsky, 2010b). At the same time, microbial and viral infections expose the human organism to different components from the parasite's cells and viral particles, including protein, DNA, RNA, lipids, and polysaccharides. The sera of mice infected with different microbe's bacterial pathogens contain a variety of Abs to the parasite's antigens and to human lipids, proteins, and nuclear components, including anti-DNA Abs (Ternynck et al., 1991; Unterkircher et al., 1993; Matsiota-Bernard et al., 1993, 1996; Hentati et al., 1994; Boekel et al., 1997; Wun et al., 2001). The origin of anti-DNA Abs in the infections remains speculative; some of them may arise inadvertently in the course of a normal immune response due to the induction by Abs that bear structures (mimotopes) mimicking DNA (Wun et al., 2001). The immunoregulatory effect of the infection seems to be related, at least partially, to the increase in a particular population of Abs, the polyreactive Abs (Matsiota-Bernard et al., 1996). It has been proposed that bacterial and viral infectious agents can act in some cases through the mechanism of molecular mimicry and stimulate development of different AI diseases. For instance, the agents responsible for molecular mimicry in multiple sclerosis include measles, hepatitis B, herpes simplex, influenza, papilloma, and Epstein-Bar viruses (Steinman, 2001). Thus, AI reactions in different AI, viral and bacterial infection diseases are very similar and may be strongly associated.

Artificial Abzs can be obtained by immunization of animals with chemically stable analogs of transition states of chemical reactions (reviewed in Martin & Schultz, 1999; Nevinsky et al., 2000a; Tanaka, 2002; Tanaka and Barbas, 2002; Dias et al., 2002; Keinan, 2005). On the other hand, artificial antiidiotypic Abs can also possess catalytic activity (Barbas et al., 1997; Wentworth et al., 1998). Building on earlier observations on the existence of idiotypic determinants related to the antigen, Jerne proposed that the immune system is selfregulated by a network of idiotype–anti-idiotype interactions (Jerne, 1974). The simplified

Antibodies 1 and 2 are termed idiotype and anti-idiotype, respectively, etc. There is convincing evidence that such idiotype–anti-idiotype networks are actually present in the body. The presence of blood serum Ab4 (in the notation shown in the scheme) has been

If the active site of an enzyme plays the role of antigen triggering this anti-idiotypic chain, it is logical to suggest that the secondary anti-idiotypic Ab2 may possess the structure, a part of which represents an "internal image" or "mould" of the active site of this enzyme, and, consequently, these Abs may possess some properties of this enzyme. This remarkable property of idiotypic mimicry has been exploited to raise monoclonal antiidiotypic Abzs with several different catalytic activities (reviewed in Keinan, 2005; Nevinsky et al., 2005).

**3. The origin of artificial and natural abzymes** 

confirmed in experimental animals (Jerne, 1974).

model of this network may be schematically presented as follows:

The origin of natural Abzs in different AI, viral and bacterial diseases may be complex. Similarly to artificial Abzs against analogs of transition states of catalytic reactions, naturally occurring Abzs with DNase and RNase activities may be Abs raised directly against free DNA and RNA or these nucleic acids acting as haptens bound to different proteins and resembling transition states of catalytic reactions (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein). Immunization of rabbits with pure DNA and RNA generated Abs interacting with DNA and possessing weak DNase and RNase activities (Krasnorutskii et al., 2008a, 2008b). Many SLE anti-DNA Abs are directed against histone-DNA nucleosomal complexes appearing as a result of internucleosomal cleavage during apoptosis (Founel & Muller, 2002). Apoptotic cells are the primary source of antigens and immunogens in SLE, and certain features in recognition, processing, and/or presentation of apoptotic autoantigens by antigen-presenting cells can trigger AI processes (Founel & Muller, 2002). Anti-DNA–protein and anti-RNA–protein complexes and other antinuclear antibodies were found in the sera of patients with multisystem connective tissue disease (Gottieb & Shwartz, 1972). Therefore, we have emulated such natural complexes using complexes of DNA and RNA with methylated bovine serum albumin (mBSA). Immunization of rabbits with complex of DNA and RNA with mBSA elicited production of 10-50-fold more active DNase and RNase IgGs, while pIgGs from animals immunized with mBSA were catalytically inactive (Krasnorutskii et al., 2008a, 2008b). Immunization of healthy rabbits with DNase I, DNase II, and pancreatic RNase A also produced anti-idiotypic IgGs with intrinsic DNase and RNase activities (Krasnorutskii et al., 2008c, 2008d, 2009). Thus, DNase and RNase Abzs in different AI diseases may be a cocktail of Abs against complexes of DNA and RNA with proteins and antiidiotypic Abzs to very different DNA- and RNAhydrolyzing enzymes.

Healthy humans and patients with many diseases with insignificant AI reactions usually lack Abzs or develop Abzs with very low catalytic activities, often on a borderline of the sensitivity of detection methods (Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Natural Abzs hydrolyzing DNA and RNA are described from the sera of patients with several AI (SLE; Shuster, et al., 1992; Buneva et al., 1994; Andrievskaya et al., 2000, 2002; Vlassov et al., 1998), Hashimoto's thyroiditis and polyarthritis (Vlasov et al., 1998), multiple sclerosis (MS) (Baranovskii et., 1998, 2001; Nevinsky et al., 2001), asthma (Galvita et al., 2007), and viral and bacterial diseases: viral hepatitis (Baranovskii et al., 1997; Vlasov et al., 1999), tick bone encephalitis (TBE; Parkhomenko et al., 2010), AIDS (Odintsova et al., 2006a), and several diseases caused by different bacterial infections (Parkhomenko et al., 2009). It was shown, that like in the case of AI-patients, IgGs with DNase activity from autoimmune mice are the earliest and statistically significant markers of pathology and these activities are detectable at the pre-disease stage, when there are no visible markers of SLE pathology or significant proteinuria, and anti-DNA titres are within the typical ranges of these indicators for healthy mice (Dubrovskaya et al., 2003; Andryushkova et al., 2007; Kuznetsova et al., 2007).

Using different approaches convincing evidence was provided that, similarly to Abzs from SLE and MS patients (Savel'ev et al., 2003; Ivanen et al., 2002, 2004), amylase activity is intrinsic to autoimmune mouse polyclonal IgGs (Andryushkova et al., 2006, 2007). It was shown that the relative activities of IgGs from MRL-lpr/lpr mice in the hydrolysis of DNA, ATP, and polysaccharides correlate very well with some visible (pink spots, baldness of the head and parts of the back, general health deterioration, etc.) and biochemical (proteinuria, Ab titers to native and denatured DNA) markers of AI pathologies during various stages of

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 157

proteinuria, but also with a significant additional change in the profile of HSC differentiation. This change seems to be the most important factor in the irreversible switching of the mouse immune system to an AI mode, since the changes in cell proliferation and apoptosis in different organs occur mainly on transition from healthy to pre-diseased mice and the observed differences in these indices between pre-diseased and diseased mice are insignificant (Andryushkova et al., 2006, 2007, 2009). Immunization of healthy young AI mice leads to the highest increase in urine protein, titers of anti-DNA Abs as well as DNase, amylase and ATPase Abz activities, occurring in parallel with a significant decrease in apoptosis, especially in bone marrow, thymus and spleen. However, the profile of HSC differentiation in immunized mice is quite different from the pre-diseased and spontaneously diseased mice, but comparable with that for young healthy animals. It was shown, that in contrast to spontaneously diseased AI mice, immunization with DNA does not remarkably affect bone marrow stem cells; the increased levels of anti-DNA Abs and Abzs in immunized mice may be mainly provided by an activation of lymphocyte differentiation and proliferation in different organs, first of all in the spleen, with a concomitant decrease in apoptosis. A significant decrease in apoptosis in the immunized mice may be an important factor providing the increased number of specific lymphocytes producing auto-Abs and Abzs, which are normally eliminated. Very high urine protein concentration and visible markers of SLE demonstrated by the immunized mice may be a result of kidney and spleen dysfunction (Andryushkova et al., 2006, 2007, 2009). Overall, in contrast to immunization of healthy mice an appearance of Abzs and increase in their activity is associated with changes in differentiation and proliferation of mice bone marrow HSC. At the onset of AI diseases (pre-disease condition), Abs are usually contain catalytic Abzs produced by a single clone, or at least a relatively narrow repertoire of Abzs with relatively low relative activities. In the course of chronic AI pathology development, the repertoire of Abzs constantly widens and Abs with significantly higher RAs can be found. In addition, the number of Abzs with high RAs usually increases during exacerbation of AI pathologies (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein). It should be mentioned, that the detection of Abzs was shown to be the earliest and statistically significant indicator of development of different autoimmune diseases in humans (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein)

Natural Abzs from sera of patients are usually polyclonal in origin and are products of different immuno-competent cells (Nevinsky et al., 2000b, 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Natural Abz purification is one of the most complicated aspects of

In study Abzs from the sera of HIV-infected patients, electrophoretically and immunologically homogeneous AIDS Ab fraction (pIgG+pIgM+pIgA) was first purified by chromatography of the serum proteins on Protein A-Sepharose under conditions that remove non-specifically bound proteins (Odintsova et al., 2006a, 2006b; Baranova et al., 2009, 2010). pIgMs were separated from pIgAs and pIgGs by FPLC gel filtration of the total Ab fraction on a Superdex 200 (Baranova et al., 2009, 2010). The homogeneity of the 150 kDa IgG was confirmed by SDS-PAGE with silver staining, which showed a single band under nonreducing conditions and two bands corresponding to the H and L chains after reduction

and animals (Andryushkova et al., 2006, 2007, 2009).

**4.1 Purification of natural abzymes** 

**4. Catalytic antibodies of HIV-infected patients** 

their study; it was discussed in detail in review (Nevinsky et al., 2000b).

mouse SLE (Andryushkova et al., 2006, 2007, 2009). Similarly to Abzs with DNase and RNase activities, catalytic Abs with polysaccharide-hydrolysing activity can be Abs directly against polysaccharides and their complexes with proteins and enzymes and second antiidiotypic Igs against different enzymes hydrolyzing polysaccharides (Andryushkova et al., 2006, 2007, 2009; Nevinsky et al., 2005, 2010a, 2010b).

IgGs and/or IgMs and IgAs hydrolyzing different peptides and proteins were also found in AI and other diseases: vasoactive intestinal peptide (VIP) in asthma (Paul et al., 1989), thyroglobulin in HT and rheumatoid arthritis (Li et al., 1995; Kalaga et al., 1995), prothrombin in multiple myeloma (Thiagarajan et al., 2000), protein factor VIII in haemophilia A (Lacroix-Desmazes et al., 1999), and myelin basic protein (MBP) in MS (Polosukhina et al., 2004, 2005, 2006, 2006; Legostaeva et al., 2010). Some healthy humans produce Abzs with low VIP- (Paul et al., 1989), and thyroglobulin-hydrolyzing activities (Kalaga et al., 1995), but usually healthy volunteers and patients with many diseases with insignificant autoimmune reactions lack Abzs with proteolytic activity (Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Since immunization of AI mice results in a dramatically higher incidence of Abzs with a higher activity than in conventionally used normal mouse strains (Tawfik et al., 2002; Nishi, 2002), the formation of Abzs in AI and some viral diseases may be much more profuse. The question is why autoimmunization of AI patients and mice results in a dramatically higher incidence of catalytically inactive Abs and Abzs with enzyme properties as compared with healthy humans and animals.

MRL-lpr/lpr mice spontaneously developing a SLE-like disorder are a very promising model to study the mechanisms of natural Abzs generation and their role in the pathogenesis of pronounced AI disturbances. A mutation in the *lpr* gene of these mice leads to a deficit in functional Fas ligand and dysregulation of apoptosis in homozygotes (Watanabe-Fukunada et al., 1992; Nagata & Suda, 1995). As a result, the mice develop SLElike phenotype, including accumulation of double-negative T cells (CD4– CD8– B220+ TCR+) in the peripheral lymphoid organs.

Recently we have carried out the first analysis of possible correlations between the relative activities (RAs) of mouse IgGs in the hydrolysis of DNA, ATP, and polysaccharides with several clinical and biochemical markers of AI pathologies (proteinuria, Ab titers to native and denatured DNA) at various stages of mouse SLE (Andryushkova et al., 2006, 2007, 2009). An ever-growing number of observations suggested that AI diseases may originate from defects in hematopoietic stem cells (Ikehara et al., 1990). Therefore, lymphocyte proliferation and apoptosis at different stages of the AI disorder development in MRL-lpr/lpr mice were also studied. It was shown that IgGs from the sera of 2-7 month-old control non-autoimmune (CBAxC57BL)F1 and BALB/c mice and 2-3 months-old MRL-lpr/lpr mice (conditionally healthy mice) are catalytically inactive (Andryushkova et al., 2006, 2007, 2009). During spontaneous development of deep SLE-like pathology a specific reorganization of immune system of these mice leads to conditions associated with a production of IgGs hydrolyzing DNA, ATP, and polysaccharides with low catalytic activities (conditionally pre-diseased mice) (Andryushkova et al., 2006, 2007, 2009). First significant changes in differentiation and proliferation of mice bone marrow hematopoietic stem cells (HSC; granulocyticmacrophagic colony-forming unit; erythroid burst-forming unit, and granulocyticerythroid-megacaryocytic-macrophagic colony-forming unit) in pre-diseased in comparison with healthy mice are most likely only temporary, since a transition from the pre-diseased to diseased mice is associated not only with an increase in the RAs of different Abzs and

mouse SLE (Andryushkova et al., 2006, 2007, 2009). Similarly to Abzs with DNase and RNase activities, catalytic Abs with polysaccharide-hydrolysing activity can be Abs directly against polysaccharides and their complexes with proteins and enzymes and second antiidiotypic Igs against different enzymes hydrolyzing polysaccharides (Andryushkova et

IgGs and/or IgMs and IgAs hydrolyzing different peptides and proteins were also found in AI and other diseases: vasoactive intestinal peptide (VIP) in asthma (Paul et al., 1989), thyroglobulin in HT and rheumatoid arthritis (Li et al., 1995; Kalaga et al., 1995), prothrombin in multiple myeloma (Thiagarajan et al., 2000), protein factor VIII in haemophilia A (Lacroix-Desmazes et al., 1999), and myelin basic protein (MBP) in MS (Polosukhina et al., 2004, 2005, 2006, 2006; Legostaeva et al., 2010). Some healthy humans produce Abzs with low VIP- (Paul et al., 1989), and thyroglobulin-hydrolyzing activities (Kalaga et al., 1995), but usually healthy volunteers and patients with many diseases with insignificant autoimmune reactions lack Abzs with proteolytic activity (Nevinsky et al., 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Since immunization of AI mice results in a dramatically higher incidence of Abzs with a higher activity than in conventionally used normal mouse strains (Tawfik et al., 2002; Nishi, 2002), the formation of Abzs in AI and some viral diseases may be much more profuse. The question is why autoimmunization of AI patients and mice results in a dramatically higher incidence of catalytically inactive Abs and Abzs with enzyme properties as compared with healthy

MRL-lpr/lpr mice spontaneously developing a SLE-like disorder are a very promising model to study the mechanisms of natural Abzs generation and their role in the pathogenesis of pronounced AI disturbances. A mutation in the *lpr* gene of these mice leads to a deficit in functional Fas ligand and dysregulation of apoptosis in homozygotes (Watanabe-Fukunada et al., 1992; Nagata & Suda, 1995). As a result, the mice develop SLElike phenotype, including accumulation of double-negative T cells (CD4– CD8– B220+ TCR+)

Recently we have carried out the first analysis of possible correlations between the relative activities (RAs) of mouse IgGs in the hydrolysis of DNA, ATP, and polysaccharides with several clinical and biochemical markers of AI pathologies (proteinuria, Ab titers to native and denatured DNA) at various stages of mouse SLE (Andryushkova et al., 2006, 2007, 2009). An ever-growing number of observations suggested that AI diseases may originate from defects in hematopoietic stem cells (Ikehara et al., 1990). Therefore, lymphocyte proliferation and apoptosis at different stages of the AI disorder development in MRL-lpr/lpr mice were also studied. It was shown that IgGs from the sera of 2-7 month-old control non-autoimmune (CBAxC57BL)F1 and BALB/c mice and 2-3 months-old MRL-lpr/lpr mice (conditionally healthy mice) are catalytically inactive (Andryushkova et al., 2006, 2007, 2009). During spontaneous development of deep SLE-like pathology a specific reorganization of immune system of these mice leads to conditions associated with a production of IgGs hydrolyzing DNA, ATP, and polysaccharides with low catalytic activities (conditionally pre-diseased mice) (Andryushkova et al., 2006, 2007, 2009). First significant changes in differentiation and proliferation of mice bone marrow hematopoietic stem cells (HSC; granulocyticmacrophagic colony-forming unit; erythroid burst-forming unit, and granulocyticerythroid-megacaryocytic-macrophagic colony-forming unit) in pre-diseased in comparison with healthy mice are most likely only temporary, since a transition from the pre-diseased to diseased mice is associated not only with an increase in the RAs of different Abzs and

al., 2006, 2007, 2009; Nevinsky et al., 2005, 2010a, 2010b).

humans and animals.

in the peripheral lymphoid organs.

proteinuria, but also with a significant additional change in the profile of HSC differentiation. This change seems to be the most important factor in the irreversible switching of the mouse immune system to an AI mode, since the changes in cell proliferation and apoptosis in different organs occur mainly on transition from healthy to pre-diseased mice and the observed differences in these indices between pre-diseased and diseased mice are insignificant (Andryushkova et al., 2006, 2007, 2009). Immunization of healthy young AI mice leads to the highest increase in urine protein, titers of anti-DNA Abs as well as DNase, amylase and ATPase Abz activities, occurring in parallel with a significant decrease in apoptosis, especially in bone marrow, thymus and spleen. However, the profile of HSC differentiation in immunized mice is quite different from the pre-diseased and spontaneously diseased mice, but comparable with that for young healthy animals. It was shown, that in contrast to spontaneously diseased AI mice, immunization with DNA does not remarkably affect bone marrow stem cells; the increased levels of anti-DNA Abs and Abzs in immunized mice may be mainly provided by an activation of lymphocyte differentiation and proliferation in different organs, first of all in the spleen, with a concomitant decrease in apoptosis. A significant decrease in apoptosis in the immunized mice may be an important factor providing the increased number of specific lymphocytes producing auto-Abs and Abzs, which are normally eliminated. Very high urine protein concentration and visible markers of SLE demonstrated by the immunized mice may be a result of kidney and spleen dysfunction (Andryushkova et al., 2006, 2007, 2009). Overall, in contrast to immunization of healthy mice an appearance of Abzs and increase in their activity is associated with changes in differentiation and proliferation of mice bone marrow HSC. At the onset of AI diseases (pre-disease condition), Abs are usually contain catalytic Abzs produced by a single clone, or at least a relatively narrow repertoire of Abzs with relatively low relative activities. In the course of chronic AI pathology development, the repertoire of Abzs constantly widens and Abs with significantly higher RAs can be found. In addition, the number of Abzs with high RAs usually increases during exacerbation of AI pathologies (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein). It should be mentioned, that the detection of Abzs was shown to be the earliest and statistically significant indicator of development of different autoimmune diseases in humans (Nevinsky et al., 2003, 2005, 2010a, 2010b, and refs therein) and animals (Andryushkova et al., 2006, 2007, 2009).

## **4. Catalytic antibodies of HIV-infected patients**

#### **4.1 Purification of natural abzymes**

Natural Abzs from sera of patients are usually polyclonal in origin and are products of different immuno-competent cells (Nevinsky et al., 2000b, 2002a, 2002b, 2003, 2005, 2010a, 2010b, and refs therein). Natural Abz purification is one of the most complicated aspects of their study; it was discussed in detail in review (Nevinsky et al., 2000b).

In study Abzs from the sera of HIV-infected patients, electrophoretically and immunologically homogeneous AIDS Ab fraction (pIgG+pIgM+pIgA) was first purified by chromatography of the serum proteins on Protein A-Sepharose under conditions that remove non-specifically bound proteins (Odintsova et al., 2006a, 2006b; Baranova et al., 2009, 2010). pIgMs were separated from pIgAs and pIgGs by FPLC gel filtration of the total Ab fraction on a Superdex 200 (Baranova et al., 2009, 2010). The homogeneity of the 150 kDa IgG was confirmed by SDS-PAGE with silver staining, which showed a single band under nonreducing conditions and two bands corresponding to the H and L chains after reduction

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 159

a typical example for AIDS IgGs (lane 1); there was no hydrolysis of DNA by control Abs from healthy donors (lane 2). Control human urine (lane 3) and bovine pancreatic DNase I (lane 4) also cleaved DNA, but produced bands in the position well below (33–36 kDa) the intact pIgGs (150 kDa). Since SDS dissociates all protein complexes, the detection of the DNase activity in the gel zone corresponding only to intact IgGs together with the absence of any other activity band or protein band (Fig. 2B), provides direct evidence that AIDS pIgGs hydrolyze DNA and are not contaminated by canonical DNases. In addition, after incubation of IgGs with DTT only light chains of AIDS IgGs demonstrated DNase activity

**A B C** 

Fig. 2. *In situ* SDS-PAGE analysis of DNase activity of AIDS intact IgGs (lane 1), healthy humans (lane 2) (B); human urine (lane 3) and bovine pancreatic DNase I (lane 4, B) in nonreducing conditions. IgGs were analyzed in reducing conditions after Abs incubation with DTT (C). After electrophoresis, the gel containing 3 g/ml thymus DNA (B and C) was incubated under special conditions for protein refolding and hydrolysis of nucleic acids; the nuclease activity was visualized by ethidium bromide staining of the gels (B and C). The longitudinal slices of the same gel were used for Coomassie R250 staining to reveal the positions of IgG bands (lane 2, A) and protein molecular mass markers (lane 1, A).

AIDS IgGs and IgMs were separated by SDS-PAGE respectively under nonreducing and reducing conditions and their proteolytic activity was detected after the extraction of proteins from excised gel slices (Baranova et. al., 2009, 2010). The detection of INhydrolyzing activity in the gel region corresponding only to IgG, together with the absence of any other bands of the activity or protein, provided direct evidence that IgG possesses INhydrolyzing activity. Similar results were obtained for AIDS IgGs hydrolyzing HIV RT,

As mentioned above, pIgMs cannot enter the gel. The absence of IN-hydrolyzing activity from all gel zones corresponding to the intact pIgMs under nonreducing conditions (data not shown), together with hydrolysis of IN only with separated heavy and light chains of IgMs under reducing conditions and the absence of any other bands of the activity (Fig. 3)

It was shown (Odintsova et. al., 2006b), that in contrast to known different nonspecific proteases hydrolyzing many proteins, AIDS pIgGs non-fractionated by affinity chromatography on Sepharoses bearing specific immobilized proteins efficiently hydrolyze only: human casein > HIV-1 RT > HSA but not many other tested proteins. Later it was shown that AIDS IgGs and IgMs also hydrolyze HIV-1 integrase (for example, Fig. 4)

provides a direct evidence that IgM possesses IN-hydrolyzing activity.

**IgG**

**H chain**

**L chain**

**1 2 1 2 3 4** 

(Fig. 2C).

**kDa**

**150**

**14**

HSA, and human casein (Odintsova et. al., 2006b).

(Baranova et. al., 2009, 2010).

(Fig. 1). Since IgM has a very high molecular mass (~970 kDa), it cannot enter SDS-PAGE gels under nonreducing conditions (Fig. 1A, lane 2). Two bands corresponding to the H and L chains of pIgMs were evident after Ab reduction (Fig. 1B, lane 1). The absence of any protein bands in the gel corresponding to pIgMs under nonreducing conditions (Fig. 1A, lane 2) and the presence of only two bands corresponding to the heavy and light chains under reducing conditions (Fig. 1B, lane 1) demonstrates the absence of protein contaminations in the pIgM preparations.

Fig. 1. SDS-PAGE in a nonreducing 4–15% gradient gel followed by silver staining of control pIgGs (lane 1) and pIgMs (lane 2) after affinity chromatography on protein A-Sepharose and FPLC gel filtration on a Superdex 200 column (A). SDS-PAGE of pIgGs (lane 2) and pIgMs (lane 1) in a reducing 12% gel (B).

#### **4.2 Criteria to establish that catalytic activity is intrinsic to antibodies**

The application of rigid criteria allowed the authors of the first article concerning natural Abzs (Paul et al., 1989) to conclude that VIP-hydrolyzing activity is an intrinsic property of Abs from the sera of patients with asthma. Later several additional criteria were proposed (for review see Nevinsky et al., 2000a, 2002a, 2002b, 2003, 2005).

It was shown that non-fractionated on affinity sorbents bearing immobilized DNA or specific protein substrates, pIgGs and pIgMs from HIV infected patients effectively hydrolyze DNA (Odintsova et. al., 2006a), HIV-1 RT, human casein, human serum albumin (HAS; Odintsova et. al., 2006b), and HIV-1 IN (Baranova et. al., 2009, 2010) but not many other tested proteins. We applied a set of strict criteria worked out previously (Paul et al., 1989, Nevinsky et al., 2000a, 2002a) for an analysis of DNase and proteolytic activity as an intrinsic property of AIDS IgGs and/or IgMs. The most important of these are: i) electrophoretic homogeneity of pIgGs and pIgMs (Fig. 1); ii) the complete absorption of AIDS IgGs and IgMs with the DNase or proteolytic activities by Sepharose bearing immobilized anti-light chain of human Abs leading to a disappearance of the activity from the solution and recover following its elution with an acidic buffer (pH 2.6); iii) FPLC gel filtration of IgGs using an acidic buffer (pH 2.6) did not lead to a disappearance of the activity, which tracked exactly with IgGs or IgMs. The fulfilment of these criteria was observed for Abzs with all activities mentioned above (Odintsova et. al., 2006a, 2006b; Baranova et. al., 2009, 2010).

To exclude possible artefacts due to hypothetical traces of contaminating enzymes, pIgGs were subjected to SDS-PAGE in a gel co-polymerized with calf thymus DNA, and their DNase activity was detected by incubating the gel in the standard reaction buffer (Fig. 2). Ethidium bromide staining of the gels after the electrophoresis and refolding of IgGs revealed sharp dark bands against a fluorescent background of DNA. Fig. 2B demonstrates

(Fig. 1). Since IgM has a very high molecular mass (~970 kDa), it cannot enter SDS-PAGE gels under nonreducing conditions (Fig. 1A, lane 2). Two bands corresponding to the H and L chains of pIgMs were evident after Ab reduction (Fig. 1B, lane 1). The absence of any protein bands in the gel corresponding to pIgMs under nonreducing conditions (Fig. 1A, lane 2) and the presence of only two bands corresponding to the heavy and light chains under reducing conditions (Fig. 1B, lane 1) demonstrates the absence of protein

> **H(IgM) H(IgG)**

> > **L**

**42**

**67 94 kDa**

> **29 14**

**42 94 150**

**A B**

Fig. 1. SDS-PAGE in a nonreducing 4–15% gradient gel followed by silver staining of control pIgGs (lane 1) and pIgMs (lane 2) after affinity chromatography on protein A-Sepharose and FPLC gel filtration on a Superdex 200 column (A). SDS-PAGE of pIgGs (lane 2) and pIgMs

The application of rigid criteria allowed the authors of the first article concerning natural Abzs (Paul et al., 1989) to conclude that VIP-hydrolyzing activity is an intrinsic property of Abs from the sera of patients with asthma. Later several additional criteria were proposed

It was shown that non-fractionated on affinity sorbents bearing immobilized DNA or specific protein substrates, pIgGs and pIgMs from HIV infected patients effectively hydrolyze DNA (Odintsova et. al., 2006a), HIV-1 RT, human casein, human serum albumin (HAS; Odintsova et. al., 2006b), and HIV-1 IN (Baranova et. al., 2009, 2010) but not many other tested proteins. We applied a set of strict criteria worked out previously (Paul et al., 1989, Nevinsky et al., 2000a, 2002a) for an analysis of DNase and proteolytic activity as an intrinsic property of AIDS IgGs and/or IgMs. The most important of these are: i) electrophoretic homogeneity of pIgGs and pIgMs (Fig. 1); ii) the complete absorption of AIDS IgGs and IgMs with the DNase or proteolytic activities by Sepharose bearing immobilized anti-light chain of human Abs leading to a disappearance of the activity from the solution and recover following its elution with an acidic buffer (pH 2.6); iii) FPLC gel filtration of IgGs using an acidic buffer (pH 2.6) did not lead to a disappearance of the activity, which tracked exactly with IgGs or IgMs. The fulfilment of these criteria was observed for Abzs with all activities mentioned above (Odintsova et. al., 2006a, 2006b;

To exclude possible artefacts due to hypothetical traces of contaminating enzymes, pIgGs were subjected to SDS-PAGE in a gel co-polymerized with calf thymus DNA, and their DNase activity was detected by incubating the gel in the standard reaction buffer (Fig. 2). Ethidium bromide staining of the gels after the electrophoresis and refolding of IgGs revealed sharp dark bands against a fluorescent background of DNA. Fig. 2B demonstrates

**IgM**

**IgG**

**4.2 Criteria to establish that catalytic activity is intrinsic to antibodies** 

(for review see Nevinsky et al., 2000a, 2002a, 2002b, 2003, 2005).

contaminations in the pIgM preparations.

(lane 1) in a reducing 12% gel (B).

Baranova et. al., 2009, 2010).

a typical example for AIDS IgGs (lane 1); there was no hydrolysis of DNA by control Abs from healthy donors (lane 2). Control human urine (lane 3) and bovine pancreatic DNase I (lane 4) also cleaved DNA, but produced bands in the position well below (33–36 kDa) the intact pIgGs (150 kDa). Since SDS dissociates all protein complexes, the detection of the DNase activity in the gel zone corresponding only to intact IgGs together with the absence of any other activity band or protein band (Fig. 2B), provides direct evidence that AIDS pIgGs hydrolyze DNA and are not contaminated by canonical DNases. In addition, after incubation of IgGs with DTT only light chains of AIDS IgGs demonstrated DNase activity (Fig. 2C).

Fig. 2. *In situ* SDS-PAGE analysis of DNase activity of AIDS intact IgGs (lane 1), healthy humans (lane 2) (B); human urine (lane 3) and bovine pancreatic DNase I (lane 4, B) in nonreducing conditions. IgGs were analyzed in reducing conditions after Abs incubation with DTT (C). After electrophoresis, the gel containing 3 g/ml thymus DNA (B and C) was incubated under special conditions for protein refolding and hydrolysis of nucleic acids; the nuclease activity was visualized by ethidium bromide staining of the gels (B and C). The longitudinal slices of the same gel were used for Coomassie R250 staining to reveal the positions of IgG bands (lane 2, A) and protein molecular mass markers (lane 1, A).

AIDS IgGs and IgMs were separated by SDS-PAGE respectively under nonreducing and reducing conditions and their proteolytic activity was detected after the extraction of proteins from excised gel slices (Baranova et. al., 2009, 2010). The detection of INhydrolyzing activity in the gel region corresponding only to IgG, together with the absence of any other bands of the activity or protein, provided direct evidence that IgG possesses INhydrolyzing activity. Similar results were obtained for AIDS IgGs hydrolyzing HIV RT, HSA, and human casein (Odintsova et. al., 2006b).

As mentioned above, pIgMs cannot enter the gel. The absence of IN-hydrolyzing activity from all gel zones corresponding to the intact pIgMs under nonreducing conditions (data not shown), together with hydrolysis of IN only with separated heavy and light chains of IgMs under reducing conditions and the absence of any other bands of the activity (Fig. 3) provides a direct evidence that IgM possesses IN-hydrolyzing activity.

It was shown (Odintsova et. al., 2006b), that in contrast to known different nonspecific proteases hydrolyzing many proteins, AIDS pIgGs non-fractionated by affinity chromatography on Sepharoses bearing specific immobilized proteins efficiently hydrolyze only: human casein > HIV-1 RT > HSA but not many other tested proteins. Later it was shown that AIDS IgGs and IgMs also hydrolyze HIV-1 integrase (for example, Fig. 4) (Baranova et. al., 2009, 2010).

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 161

 (Odintsova et al., 2006b). It means that within pools of AIDs pIgGs and pIgMs only specific anti-IN Abzs are able to hydrolyze intact globular molecules of viral integrase (Baranoba et al., 2009, 2010), while specific anti-HSA, anti-casein, and anti-RT IgGs hydrolyze their

**4.3 Comparison of the relative catalytic activity of Abs from different AIDS patients**  Sera of ten healthy donors, 110 HIV-infected patients (18–40-years-old; men and women) including 65 at the stage of pre-AIDS and 45 at the stage of generalized lymphadenopathy (GL) according to the classification of the Center of Disease Control and Prevention were used to analyze the catalytic activities of IgGs and IgMs (Odintsova et al., 2006a, 2006b; Baranova et al., 2009, 2010). Patients with pre-AIDS stage were characterized by a decrease in their body mass up to 10%, fungal, bacterial and viral lesions of skin and mucosal surfaces, shingles, repeating pharyngitis, sinusitis, otitis, and frequent acute respiratory infection. Polyclonal IgGs from 10 healthy controls were inactive in DNA hydrolysis (Odintsova et al., 2006a). Similar results were obtained earlier for several groups of 10-20 healthy humans used as controls in the studies of DNase Abs from the sera of patients with AI diseases (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein).The type of plasmid supercoiled (sc) DNA hydrolysis by AIDS pIgGs did not depend on the Ab concentration and the rate of the hydrolysis linearly increased with the increase in IgG concentration and time of incubation (Odintsova et al., 2006a). Fig. 5 illustrates a cleavage of plasmid DNA (14 g/ml) by Abs (0.3 mg/ml) from several AIDS patients after 4 h of incubation. One can see that in this period some Abs cause only single breaks in one strand of supercoiled DNA (lanes 1-3), whereas others cause multiple breaks and as a result the formation of linear DNA (lanes 4-6). The most active Abs hydrolyze DNA into short and medium length ODNs (lanes 7-10). It should be mentioned that Fig. 5 in principle illustrates a range of possible changes of the relative DNase activities for patients with not only AIDS but also with different AI diseases and viral pathologies previously analyzed (Nevinsky *et al.,* 2000a, 2003, 2005, 2010a, 2010b and refs therein). When passing from one pathology to another only the values of a relative

percent of patients with low, middle and high DNase activities is usually changed.

Fig. 5. DNase activities of catalytic AIDS IgGs from different patients in the cleavage of supercoiled (sc) and nicked (n) plasmid DNA (14 g/ml). Lanes 1–10, IgGs (0.3 mg/ml; incubation for 4 h) from the sera of 10 different patients; C1, scDNA incubated alone; C2 and

The efficiency of DNA cleavage was calculated from the increase of DNA in the band of scDNA (10-40 % of initial DNA hydrolysis); the measured RAs (%) for IgGs were normalized to standard conditions (0.1 mg/ml Abs, 2 h) and a complete transition of scDNA to its relaxed form was taken for 100%. While four AIDS IgG preparations were completely inactive, the remaining 106 IgG samples (96 %) demonstrated the RAs from 5 to 100% of

C3, scDNA incubated with Ab from the sera of two healthy donors.

scDNA hydrolysis (Table 1).

specific target proteins (Odintsova et. al., 2006b).

Fig. 3. SDS-PAGE analysis of IN-hydrolyzing activity of AIDS IgMs. After reducing SDS-PAGE of purified AIDS IgMs the gel was incubated under special conditions for renaturation of Abs. The relative IN-hydrolyzing activity (RA, %) was quantified using the extracts of fragments (2–3-mm each) of one longitudinal slice of the gel corresponding to 3 individual IgMs: (■), IgM1; (□), IgM2; (∆), IgM3.

It was shown that immunogenic VIP (Paul et al., 1989), human MBP (Ponomarenko et al., 2006; Legostaseva at al., 2010), human milk casein (Odintsova et al., 2011), stimulate formation of Abzs which in contrast to canonical proteases efficiently hydrolyze only antigen-proteins, but not many other proteins tested. To analyze the "average" proteolytic activity of AIDS Abzs, two mixtures of equal amounts of electrophoretically homogeneous pIgGs (pIgGmix) and pIgMs (pIgMmix) with different relatively high and average activities from the sera of seven patients were prepared. After purification of anti-IN Abs on IN-Sepharose these Abzs hydrolyzed only IN and cannot hydrolyze other proteins including viral RT (Fig. 4). In addition, it was shown that AIDS pIgGs and pIgMs after their purification on sorbents bearing immobilized HIV RT (Fig. 4), human casein or HSA specifically hydrolyzed only cognate protein, but not many other proteins including HIV IN

Fig. 4. SDS-PAGE analysis of protein hydrolysis by AIDS IgGs and IgMs. A, HIV-1 IN was incubated for alone (1) or in the presence of pIgGs from two healthy donors (2 and 3) and pIgGs from two different AIDS patients (4 and 5), or with AIDS IgGs purified on IN-Sepharose (6). B, AIDS IgMs purified on RT-Sepharose were incubated with different proteins without (odd numbers) and with Abs (even numbers): HIV-1 IN (1 and 2); p66 HIV RT (3 and 4); HSA (5 and 6); myelin basic protein (7 and 8). Lane 9 corresponds to mixture of standard protein markers with known molecular masses.

**205**

**kDa**

individual IgMs: (■), IgM1; (□), IgM2; (∆), IgM3.

**RA, %**

**150**

PAGE of purified AIDS IgMs the gel was incubated under special conditions for

**97**

Fig. 3. SDS-PAGE analysis of IN-hydrolyzing activity of AIDS IgMs. After reducing SDS-

renaturation of Abs. The relative IN-hydrolyzing activity (RA, %) was quantified using the extracts of fragments (2–3-mm each) of one longitudinal slice of the gel corresponding to 3

It was shown that immunogenic VIP (Paul et al., 1989), human MBP (Ponomarenko et al., 2006; Legostaseva at al., 2010), human milk casein (Odintsova et al., 2011), stimulate formation of Abzs which in contrast to canonical proteases efficiently hydrolyze only antigen-proteins, but not many other proteins tested. To analyze the "average" proteolytic activity of AIDS Abzs, two mixtures of equal amounts of electrophoretically homogeneous pIgGs (pIgGmix) and pIgMs (pIgMmix) with different relatively high and average activities from the sera of seven patients were prepared. After purification of anti-IN Abs on IN-Sepharose these Abzs hydrolyzed only IN and cannot hydrolyze other proteins including viral RT (Fig. 4). In addition, it was shown that AIDS pIgGs and pIgMs after their purification on sorbents bearing immobilized HIV RT (Fig. 4), human casein or HSA specifically hydrolyzed only cognate protein, but not many other proteins including HIV IN

**1 2 3 4 5 6 1 2 3 4 5 6 7 8 9**

**kDa kDa**

**67 43 30**

**14.4**

Fig. 4. SDS-PAGE analysis of protein hydrolysis by AIDS IgGs and IgMs. A, HIV-1 IN was incubated for alone (1) or in the presence of pIgGs from two healthy donors (2 and 3) and pIgGs from two different AIDS patients (4 and 5), or with AIDS IgGs purified on IN-Sepharose (6). B, AIDS IgMs purified on RT-Sepharose were incubated with different proteins without (odd numbers) and with Abs (even numbers): HIV-1 IN (1 and 2); p66 HIV RT (3 and 4); HSA (5 and 6); myelin basic protein (7 and 8). Lane 9 corresponds to mixture of

**20**

standard protein markers with known molecular masses.

**A B**

**43**

**0 3 6 9 12 15 18 21 24**

**Fragment number**

**30**

**18**

 (Odintsova et al., 2006b). It means that within pools of AIDs pIgGs and pIgMs only specific anti-IN Abzs are able to hydrolyze intact globular molecules of viral integrase (Baranoba et al., 2009, 2010), while specific anti-HSA, anti-casein, and anti-RT IgGs hydrolyze their specific target proteins (Odintsova et. al., 2006b).

## **4.3 Comparison of the relative catalytic activity of Abs from different AIDS patients**

Sera of ten healthy donors, 110 HIV-infected patients (18–40-years-old; men and women) including 65 at the stage of pre-AIDS and 45 at the stage of generalized lymphadenopathy (GL) according to the classification of the Center of Disease Control and Prevention were used to analyze the catalytic activities of IgGs and IgMs (Odintsova et al., 2006a, 2006b; Baranova et al., 2009, 2010). Patients with pre-AIDS stage were characterized by a decrease in their body mass up to 10%, fungal, bacterial and viral lesions of skin and mucosal surfaces, shingles, repeating pharyngitis, sinusitis, otitis, and frequent acute respiratory infection.

Polyclonal IgGs from 10 healthy controls were inactive in DNA hydrolysis (Odintsova et al., 2006a). Similar results were obtained earlier for several groups of 10-20 healthy humans used as controls in the studies of DNase Abs from the sera of patients with AI diseases (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein).The type of plasmid supercoiled (sc) DNA hydrolysis by AIDS pIgGs did not depend on the Ab concentration and the rate of the hydrolysis linearly increased with the increase in IgG concentration and time of incubation (Odintsova et al., 2006a). Fig. 5 illustrates a cleavage of plasmid DNA (14 g/ml) by Abs (0.3 mg/ml) from several AIDS patients after 4 h of incubation. One can see that in this period some Abs cause only single breaks in one strand of supercoiled DNA (lanes 1-3), whereas others cause multiple breaks and as a result the formation of linear DNA (lanes 4-6). The most active Abs hydrolyze DNA into short and medium length ODNs (lanes 7-10). It should be mentioned that Fig. 5 in principle illustrates a range of possible changes of the relative DNase activities for patients with not only AIDS but also with different AI diseases and viral pathologies previously analyzed (Nevinsky *et al.,* 2000a, 2003, 2005, 2010a, 2010b and refs therein). When passing from one pathology to another only the values of a relative percent of patients with low, middle and high DNase activities is usually changed.

Fig. 5. DNase activities of catalytic AIDS IgGs from different patients in the cleavage of supercoiled (sc) and nicked (n) plasmid DNA (14 g/ml). Lanes 1–10, IgGs (0.3 mg/ml; incubation for 4 h) from the sera of 10 different patients; C1, scDNA incubated alone; C2 and C3, scDNA incubated with Ab from the sera of two healthy donors.

The efficiency of DNA cleavage was calculated from the increase of DNA in the band of scDNA (10-40 % of initial DNA hydrolysis); the measured RAs (%) for IgGs were normalized to standard conditions (0.1 mg/ml Abs, 2 h) and a complete transition of scDNA to its relaxed form was taken for 100%. While four AIDS IgG preparations were completely inactive, the remaining 106 IgG samples (96 %) demonstrated the RAs from 5 to 100% of scDNA hydrolysis (Table 1).

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 163

two subgroups with either rapid or slow progression of the disease. According to the recommendations of the Center of Disease Control and Prevention, the transition time from GL to pre-AIDS stage (≤ 2 and > 2 years, respectively) was used as the measure of the disease rate of progression. The number of GL patients demonstrating a detectable proteolytic activity was comparable in the case of rapid (41.7 %) and slow (47.8%) progression, while DNase activity was observed in 41.3 % of the patients with rapid progression and only in 29.5% of the patients with slow progression. In the pre-AIDS group, the number of IgGs with DNase and proteolytic activity was slightly higher in patients with rapid progression (55.8 and 58.1 %, respectively) than in patients with slow (50.7 and 51.8 %)

On first glance, high activity of IgGs from the blood of AIDS patients in hydrolysis of βcasein (which is not a typical component of human blood) is unexpected. However, it was recently shown by 2D electrophoresis that six of nine sera from AIDS patients contained Abs against casein, and five against human milk lactalbumin (Goldfarb, 2001). Thereby, the activation of β-casein synthesis in AIDS patients driven by not yet understood factor can not be excluded. It is interesting that mRNA corresponding to the gene encoding for β-casein is produced in mouse T-killer cells (also for unknown reason) (Grusby et al., 1990). In this way, it can not be excluded that genes, encoding for β-casein, as well as this protein itself, can play a special (but not yet known) role in the HIV virus life cycle, its replication, or

At the same time, so far Abzs hydrolyzing HSA were found only in AIDS patients. A possible reason of production of HSA-hydrolyzing Abs in AIDS patients is also not known. Later the RAs of AIDS IgGs (Baranova et al., 2009) and IgMs (Baranova et al., 2010) in the hydrolysis of IN were analyzed. Sera of 19 HIV-infected patients (18-40 yr old; men and women) including 13 at the stage of pre-AIDS and 6 at the stage of GL were used to study IN-hydrolyzing activity of IgGs, while 18 Ab preparations corresponding to pre-AIDS stage

pIgGs and pIgMs from ten control healthy donors were unable to catalyze IN hydrolysis. Interestingly, 11 of 13 IgGs from patients with pre-AIDS (84.6 %) and 6 of 6 (100 %) with GL demonstrated detectable of high IN-hydrolyzing activity (Baranova et al., 2009). There was no statistically significant difference in the IgG RAs between the two groups of patients; average values of IgG IN-hydrolyzing RAs were 1.99 ± 1.68 for pre-AIDS and 3.4 ± 1.31 (M IN/1h)/mg of Abs for GL patients. All 16 IgMs purified from patients with pre-AIDS (100%) and 6 of 8 IgMs (75%) from patients with GL demonstrated high or detectable INhydrolyzing activity (Baranova et al., 2010). There was no statistically significant difference (*p* = 0.71) in the IgM RAs between the two groups of patients; the average values of IgM INhydrolyzing RAs were 3.8 ± 2.2 M IN per hour per mg of Abs (range 0.3–7.3 M IN per hour per mg of Abs) for pre-AIDS and 3.3 ± 2.6 M IN per hour per mg of Abs (range 0–8.1

Overall, in the case of Abs with DNase, casein-, and IN-hydrolyzing activity we have found only a negligible difference in the RAs of Abs from HIV-infected patients at the GL and pre-AIDS stages. However, it is not surprising and agrees with the published data that a detection of Abzs is the earliest indicator of the development of many AI diseases in humans and animals (Andryushkova et al., 2006, 2007, 2009; Nevinsky et al., 2005, 2010a, 2010b, and refs therein). According to our data, various catalytic activities of Abzs are usually very easily detectable at the onset of AI diseases when the total concentrations of non-catalytic

progression of the disease.

development of AI reactions in AIDS patients.

M IN per hour per mg of Abs) for GL.

and 8 preparations to GL were used to analyze RAs of IgMs.

On the fist step we have analyzed the RAs of proteolytic activity of AIDS pIgGs in the hydrolysis of many different proteins including HIV RT, HSA, and human -casein (Odintsova et al., 2006b). It was shown that among all proteins used, IgGs from different HIV-infected patients with detectable or high rate hydrolyze only HIV-1 reverse transcriptase (RT), human serum albumin (HSA), and human -casein. Interestingly, the highest rate of the hydrolysis was observed for -casein (Odintsova et al., 2006b). Therefore, the RA of 110 AIDS IgGs in the hydrolysis of casein was analyzed. The efficiency of casein cleavage was analysed by SDS-PAGE and calculated from the decrease in the band of nonhydrolyzed casein taking into account the control reaction incubated in the absence of IgGs.


\*Percent of total number of patients (110) is given in parenthesis.

\*\*A complete transition of 14 g/ml scDNA to its relaxed form (0.1 mg/ml IgGs, 2 h) was taken for 100%. \*\*\*A complete hydrolysis of 0.1 mg/ml human -casein (0.1 mg/ml IgGs, 2 h) was taken for 100%.

Table 1. The relative catalytic activities of AIDS IgGs in the hydrolysis of scDNA and casein in the case of total group and patients with different stages of diseases development.

Purified pIgGs from ten healthy donors were unable to catalyze casein hydrolysis, whereas 105 of 110 IgGs (95 %) demonstrated high or detectable casein-hydrolyzing activity (Table 1). With the development of the disease at the stage of pre-AIDS, IgGs from some patients demonstrated a high DNase activity (≥80 % of DNA hydrolysis) but the average RA was 44.7 % 21.3 % (Table 1). At the stage of generalized lymphadenopathy, IgGs from 4 of 45 patients (8.8 %) did not possess detectable DNase activity. However, the average RA value of DNase activity for this group of patients (40.2 % 26.3 %) was comparable with that for the pre-AIDS group. At the same time, the number of IgG preparations with very high activities (≥80 % of DNA hydrolysis) was significantly lower in the GL group. At the GL stage, 5 of 45 (11 %) Abs were completely inactive, while several IgGs demonstrated high RA in the hydrolysis of casein (up to 86.7%); the average RA for all 45 patients was 53.6 % 22.6 % (Table 1). All 65 IgGs from patients with pre-AIDS were catalytically active, but the average RA (51.5 % 25.9 %) was comparable with that for IgGs from patients with GL (Table 1). Each group of patients corresponding to GL and pre-AIDS stage was divided into

On the fist step we have analyzed the RAs of proteolytic activity of AIDS pIgGs in the hydrolysis of many different proteins including HIV RT, HSA, and human -casein (Odintsova et al., 2006b). It was shown that among all proteins used, IgGs from different HIV-infected patients with detectable or high rate hydrolyze only HIV-1 reverse transcriptase (RT), human serum albumin (HSA), and human -casein. Interestingly, the highest rate of the hydrolysis was observed for -casein (Odintsova et al., 2006b). Therefore, the RA of 110 AIDS IgGs in the hydrolysis of casein was analyzed. The efficiency of casein cleavage was analysed by SDS-PAGE and calculated from the decrease in the band of nonhydrolyzed casein taking into account the control reaction incubated in the absence of IgGs.

> Number of IgG preparations with DNase activity\*\*

lymphadenopathy (45) 40.2 % ± 26.3 % 53.6 % 22.6 % pre-AIDS (65) 44.7 % ± 21.3 % 51.5 % 25.9 %

\*\*A complete transition of 14 g/ml scDNA to its relaxed form (0.1 mg/ml IgGs, 2 h) was taken for 100%. \*\*\*A complete hydrolysis of 0.1 mg/ml human -casein (0.1 mg/ml IgGs, 2 h) was taken for 100%. Table 1. The relative catalytic activities of AIDS IgGs in the hydrolysis of scDNA and casein in the case of total group and patients with different stages of diseases development. Purified pIgGs from ten healthy donors were unable to catalyze casein hydrolysis, whereas 105 of 110 IgGs (95 %) demonstrated high or detectable casein-hydrolyzing activity (Table 1). With the development of the disease at the stage of pre-AIDS, IgGs from some patients demonstrated a high DNase activity (≥80 % of DNA hydrolysis) but the average RA was 44.7 % 21.3 % (Table 1). At the stage of generalized lymphadenopathy, IgGs from 4 of 45 patients (8.8 %) did not possess detectable DNase activity. However, the average RA value of DNase activity for this group of patients (40.2 % 26.3 %) was comparable with that for the pre-AIDS group. At the same time, the number of IgG preparations with very high activities (≥80 % of DNA hydrolysis) was significantly lower in the GL group. At the GL stage, 5 of 45 (11 %) Abs were completely inactive, while several IgGs demonstrated high RA in the hydrolysis of casein (up to 86.7%); the average RA for all 45 patients was 53.6 % 22.6 % (Table 1). All 65 IgGs from patients with pre-AIDS were catalytically active, but the average RA (51.5 % 25.9 %) was comparable with that for IgGs from patients with GL (Table 1). Each group of patients corresponding to GL and pre-AIDS stage was divided into

\*Percent of total number of patients (110) is given in parenthesis.

0 4 (4 %)\* 5 (5 %) 5 – 20 21 (19 %) 11 (10 %) 21 – 40 33 (30 %) 24 (21.4 %) 41 – 60 18 (16.4 %) 25 (22.7 %) 61 – 80 7 (6.4 %) 20 (18.2 %) 81 – 100 17 (15.5 %) 25 (22.7 %)

Average DNase RA Average protease RA

Number of IgG preparations with casein-hydrolyzing

activity\*\*\*

Relative activity, % of scDNA hydrolysis

Number of preparations and

stage of disease

Generalized

two subgroups with either rapid or slow progression of the disease. According to the recommendations of the Center of Disease Control and Prevention, the transition time from GL to pre-AIDS stage (≤ 2 and > 2 years, respectively) was used as the measure of the disease rate of progression. The number of GL patients demonstrating a detectable proteolytic activity was comparable in the case of rapid (41.7 %) and slow (47.8%) progression, while DNase activity was observed in 41.3 % of the patients with rapid progression and only in 29.5% of the patients with slow progression. In the pre-AIDS group, the number of IgGs with DNase and proteolytic activity was slightly higher in patients with rapid progression (55.8 and 58.1 %, respectively) than in patients with slow (50.7 and 51.8 %) progression of the disease.

On first glance, high activity of IgGs from the blood of AIDS patients in hydrolysis of βcasein (which is not a typical component of human blood) is unexpected. However, it was recently shown by 2D electrophoresis that six of nine sera from AIDS patients contained Abs against casein, and five against human milk lactalbumin (Goldfarb, 2001). Thereby, the activation of β-casein synthesis in AIDS patients driven by not yet understood factor can not be excluded. It is interesting that mRNA corresponding to the gene encoding for β-casein is produced in mouse T-killer cells (also for unknown reason) (Grusby et al., 1990). In this way, it can not be excluded that genes, encoding for β-casein, as well as this protein itself, can play a special (but not yet known) role in the HIV virus life cycle, its replication, or development of AI reactions in AIDS patients.

At the same time, so far Abzs hydrolyzing HSA were found only in AIDS patients. A possible reason of production of HSA-hydrolyzing Abs in AIDS patients is also not known.

Later the RAs of AIDS IgGs (Baranova et al., 2009) and IgMs (Baranova et al., 2010) in the hydrolysis of IN were analyzed. Sera of 19 HIV-infected patients (18-40 yr old; men and women) including 13 at the stage of pre-AIDS and 6 at the stage of GL were used to study IN-hydrolyzing activity of IgGs, while 18 Ab preparations corresponding to pre-AIDS stage and 8 preparations to GL were used to analyze RAs of IgMs.

pIgGs and pIgMs from ten control healthy donors were unable to catalyze IN hydrolysis. Interestingly, 11 of 13 IgGs from patients with pre-AIDS (84.6 %) and 6 of 6 (100 %) with GL demonstrated detectable of high IN-hydrolyzing activity (Baranova et al., 2009). There was no statistically significant difference in the IgG RAs between the two groups of patients; average values of IgG IN-hydrolyzing RAs were 1.99 ± 1.68 for pre-AIDS and 3.4 ± 1.31 (M IN/1h)/mg of Abs for GL patients. All 16 IgMs purified from patients with pre-AIDS (100%) and 6 of 8 IgMs (75%) from patients with GL demonstrated high or detectable INhydrolyzing activity (Baranova et al., 2010). There was no statistically significant difference (*p* = 0.71) in the IgM RAs between the two groups of patients; the average values of IgM INhydrolyzing RAs were 3.8 ± 2.2 M IN per hour per mg of Abs (range 0.3–7.3 M IN per hour per mg of Abs) for pre-AIDS and 3.3 ± 2.6 M IN per hour per mg of Abs (range 0–8.1 M IN per hour per mg of Abs) for GL.

Overall, in the case of Abs with DNase, casein-, and IN-hydrolyzing activity we have found only a negligible difference in the RAs of Abs from HIV-infected patients at the GL and pre-AIDS stages. However, it is not surprising and agrees with the published data that a detection of Abzs is the earliest indicator of the development of many AI diseases in humans and animals (Andryushkova et al., 2006, 2007, 2009; Nevinsky et al., 2005, 2010a, 2010b, and refs therein). According to our data, various catalytic activities of Abzs are usually very easily detectable at the onset of AI diseases when the total concentrations of non-catalytic

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 165

(Andryushkova et al., 2006, 2007, 2009). Immunization of healthy mice with DNA also leads to production of DNase Abzs; however, it is associated only with increased lymphocyte proliferation and suppression of apoptosis of lymphocytes in different organs (especially spleen), but not with a change in differentiation of the bone marrow cells. Immune processes after immunization of mammals with bacterial DNA, proteins, polysaccharides during many infectious diseases may be considered similar to those after immunization of healthy mice with DNA, different proteins and enzymes. According to theoretical analysis, the adaptive improvement of the catalytic turnover is limited by the rate of B cell receptor signal transduction, as rapid release of antigen fragments from catalytic B cell receptors aborts clonal selection, but production of catalysts can occur at increased levels under conditions of rapid B cell signaling in AI disease (Paul et al., 2006). In addition, the RAs of DNase Abs increased with the progress of the AI pathology, while the time course of immunization associated with some infections is usually not so long as compared with AI diseases, which can have chronic character. In contrast to AI diseases, treatment and recovery of patients with bacterial infections usually eliminates Abzs with various activities. In addition, Abzs may have protective functions in patients with bacterial infections. It was shown that the presence of IgG endowed with serine protease-like activity in the plasma of patients with

sepsis strongly correlates with their survival (Lacroix-Desmazes et al., 2005).

these bacterial and viral components.

**5. Extreme diversity of AIDS abzymes 5.1 Structural diversity of AIDS abzymes** 

AI processes.

In contrast to DNase abzymes, the polysaccharide-hydrolyzing Abs are usually present even in the sera of healthy humans and their activity remarkably increases in the sera of patients and animals with different AI diseases (Andryushkova et al., 2006; 2007) and especially with pathologies caused by viral infections (Buneva V.N., personal communication). Formation of specific Abs against DNA and other components of bacteria and some viruses during infections of healthy mammals suggests that the specific catalytic Abs can mostly hydrolyze

It is possible that co-action of Abzs with proteolytic and polysaccharide-hydrolyzing activities can at least partially degrade bacterial cell walls and viral particles and facilitate the entry of Abzs into the bacterial cells and hydrolysis of bacterial DNA, proteins and other components. This cooperative action of abzymes with different catalytic activities may have a protective effect against diseases caused by bacterial and viral infections. At the same time, in contrast to many viral and bacterial infections, HIV-infection stimulates AI reactions. Therefore, at the first stage of AIDS development catalytic Abzs against different viral components can protect humans, similarly to the situation in bacterial and viral infections that do not stimulate AI reactions. Later, due to molecular mimicry between viral and host antigens, viral antigens can affect hematopoietic stem cells and trigger the development of

DNase, RNase, ATPase, amylase, and protease Abzs may show very different contributions of variable domains of H- and L-chains to their active centers. Chromatographically separated light chains of IgGs from the sera of asthma patients were found to be active in the hydrolysis of VIP (Sun et al., 1994; 1997). The light chain of the VIP Abz was expressed in bacteria, purified, and found to possess an intrinsic catalytic activity (Tyutyulkova et al., 1996). The Abz-dependent hydrolysis of DNA and RNA by isolated light chains of IgGs from SLE, MS, asthma, and other AI patients, as well as from MRL-lpr/lpr mice, is more

Abs to specific auto-antigens have not yet increased significantly and correspond to their ranges for healthy donors. At the early stages of AI diseases, the repertoire of Abzs is usually relatively small but it greatly increases with the progress of the disease, leading to the generation of catalytically diverse abzymes with different activities and functions (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein). In addition, AI reactions in the case of some viral diseases including AIDS patients are in some extent similar to AI diseases. At the same time, immunization of AI mice produces an unexpectedly high increase in the number of clones secreting various auto-Abs, including Abzs, in comparison with normal mice (Nishi, 2002; Tawfik et al., 2002).

HIV-1 RT- and IN-hydrolzying pIgGs and IgMs from HIV-infected patients were the first examples of catalytic Abzs produced in humans against viral proteins after a viral infection (Odintsova et al., 2006b; Baranova et al., 2009, 2010). In addition, it was shown for the first time that HIV infection stimulates autoimmune reactions leading to the formation of Abzs that hydrolyze at least two human proteins, HSA and casein. It is known that HIV infection stimulates the development of many AI diseases (Zandman-Goddard et al., 2002). One can suppose that in some other viral and bacterial infections may induce similar processes to some extent.

During many infections, the human organism is exposed to different bacterial components including protein, DNA, RNA, and polysaccharides (Ternynck et al., 1991; Unterkircher et al., 1993; Matsiota-Bernard et al., 1993, 1996; Hentati et al., 1994; Boekel et al., 1997; Wun et al., 2001). Because of their ability to bind a variety of exogenous antigens, including bacterial and viral ones, natural Abs play a major role in the primary line of defense against infections. Some results suggest that the synthesis of auto-Abs and Abs directed against bacterial antigens at least partially follow distinct pathways, but with the existing experimental data it is impossible to determine unambiguously whether these two Ab populations are produced by the same or distinct B-cell subpopulations (Matsiota-Bernard et al., 1993). Recently, DNase activity in the patients with diseases caused by several bacterial infections has been analyzed (Parkhomenko et al., 2008). The catalytic activities were significantly lower than in patients with different AI pathologies and increased in the following order: streptococcal infection (erysipelas) < urogenital chlamydiosis associated with arthritis (Reiter's disease) < meningococcal meningitis < shigellosis < suppurative surgical infections caused by *Staphylococcus aureus* < suppurative surgical infections caused by epidermal staphylococci < urogenital ureaplasmosis associated with reactive arthritis.

In addition, DNA-hydrolyzing IgGs was found in the sera of patients with hepatitis (Baranovskii et al., 1997) and tick bone encephalitis (Parkhomenko et al., 2010-11). Interestingly, TBE like HIV infection of humans stimulate formation of Abzs with several proteolytic activities (Parkhomenko T., personal communication). The RAs of IgGs in the hydrolysis of DNA increased in the following order: diabetes ≤ bacterial infections ≤ viral hepatitis < polyarthritis < Hashimoto's thyroiditis < AIDS ≤ MS < SLE (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein).

Taking these observations together, we suggest that the specific activity of polyclonal Abs from the sera of patients with diseases caused by bacterial infections are usually lower than those for typical AI diseases and most probably they can differ in their biological functions. It was shown that the specific reorganization of immune system during the spontaneous development of a profound SLE-like pathology in MRL-lpr/lpr mice is associated with changes in the differentiation profile and the level of proliferation of bone marrow hematopoietic stem cells and with production of DNase, ATPase, and amylase Abzs

Abs to specific auto-antigens have not yet increased significantly and correspond to their ranges for healthy donors. At the early stages of AI diseases, the repertoire of Abzs is usually relatively small but it greatly increases with the progress of the disease, leading to the generation of catalytically diverse abzymes with different activities and functions (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein). In addition, AI reactions in the case of some viral diseases including AIDS patients are in some extent similar to AI diseases. At the same time, immunization of AI mice produces an unexpectedly high increase in the number of clones secreting various auto-Abs, including Abzs, in comparison with normal

HIV-1 RT- and IN-hydrolzying pIgGs and IgMs from HIV-infected patients were the first examples of catalytic Abzs produced in humans against viral proteins after a viral infection (Odintsova et al., 2006b; Baranova et al., 2009, 2010). In addition, it was shown for the first time that HIV infection stimulates autoimmune reactions leading to the formation of Abzs that hydrolyze at least two human proteins, HSA and casein. It is known that HIV infection stimulates the development of many AI diseases (Zandman-Goddard et al., 2002). One can suppose that in some other viral and bacterial infections may induce similar processes to

During many infections, the human organism is exposed to different bacterial components including protein, DNA, RNA, and polysaccharides (Ternynck et al., 1991; Unterkircher et al., 1993; Matsiota-Bernard et al., 1993, 1996; Hentati et al., 1994; Boekel et al., 1997; Wun et al., 2001). Because of their ability to bind a variety of exogenous antigens, including bacterial and viral ones, natural Abs play a major role in the primary line of defense against infections. Some results suggest that the synthesis of auto-Abs and Abs directed against bacterial antigens at least partially follow distinct pathways, but with the existing experimental data it is impossible to determine unambiguously whether these two Ab populations are produced by the same or distinct B-cell subpopulations (Matsiota-Bernard et al., 1993). Recently, DNase activity in the patients with diseases caused by several bacterial infections has been analyzed (Parkhomenko et al., 2008). The catalytic activities were significantly lower than in patients with different AI pathologies and increased in the following order: streptococcal infection (erysipelas) < urogenital chlamydiosis associated with arthritis (Reiter's disease) < meningococcal meningitis < shigellosis < suppurative surgical infections caused by *Staphylococcus aureus* < suppurative surgical infections caused by epidermal staphylococci < urogenital ureaplasmosis associated with reactive arthritis. In addition, DNA-hydrolyzing IgGs was found in the sera of patients with hepatitis (Baranovskii et al., 1997) and tick bone encephalitis (Parkhomenko et al., 2010-11). Interestingly, TBE like HIV infection of humans stimulate formation of Abzs with several proteolytic activities (Parkhomenko T., personal communication). The RAs of IgGs in the hydrolysis of DNA increased in the following order: diabetes ≤ bacterial infections ≤ viral hepatitis < polyarthritis < Hashimoto's thyroiditis < AIDS ≤ MS < SLE (Nevinsky *et al.,* 2003,

Taking these observations together, we suggest that the specific activity of polyclonal Abs from the sera of patients with diseases caused by bacterial infections are usually lower than those for typical AI diseases and most probably they can differ in their biological functions. It was shown that the specific reorganization of immune system during the spontaneous development of a profound SLE-like pathology in MRL-lpr/lpr mice is associated with changes in the differentiation profile and the level of proliferation of bone marrow hematopoietic stem cells and with production of DNase, ATPase, and amylase Abzs

mice (Nishi, 2002; Tawfik et al., 2002).

2005, 2010a, 2010b and refs therein).

some extent.

(Andryushkova et al., 2006, 2007, 2009). Immunization of healthy mice with DNA also leads to production of DNase Abzs; however, it is associated only with increased lymphocyte proliferation and suppression of apoptosis of lymphocytes in different organs (especially spleen), but not with a change in differentiation of the bone marrow cells. Immune processes after immunization of mammals with bacterial DNA, proteins, polysaccharides during many infectious diseases may be considered similar to those after immunization of healthy mice with DNA, different proteins and enzymes. According to theoretical analysis, the adaptive improvement of the catalytic turnover is limited by the rate of B cell receptor signal transduction, as rapid release of antigen fragments from catalytic B cell receptors aborts clonal selection, but production of catalysts can occur at increased levels under conditions of rapid B cell signaling in AI disease (Paul et al., 2006). In addition, the RAs of DNase Abs increased with the progress of the AI pathology, while the time course of immunization associated with some infections is usually not so long as compared with AI diseases, which can have chronic character. In contrast to AI diseases, treatment and recovery of patients with bacterial infections usually eliminates Abzs with various activities. In addition, Abzs may have protective functions in patients with bacterial infections. It was shown that the presence of IgG endowed with serine protease-like activity in the plasma of patients with sepsis strongly correlates with their survival (Lacroix-Desmazes et al., 2005).

In contrast to DNase abzymes, the polysaccharide-hydrolyzing Abs are usually present even in the sera of healthy humans and their activity remarkably increases in the sera of patients and animals with different AI diseases (Andryushkova et al., 2006; 2007) and especially with pathologies caused by viral infections (Buneva V.N., personal communication). Formation of specific Abs against DNA and other components of bacteria and some viruses during infections of healthy mammals suggests that the specific catalytic Abs can mostly hydrolyze these bacterial and viral components.

It is possible that co-action of Abzs with proteolytic and polysaccharide-hydrolyzing activities can at least partially degrade bacterial cell walls and viral particles and facilitate the entry of Abzs into the bacterial cells and hydrolysis of bacterial DNA, proteins and other components. This cooperative action of abzymes with different catalytic activities may have a protective effect against diseases caused by bacterial and viral infections. At the same time, in contrast to many viral and bacterial infections, HIV-infection stimulates AI reactions. Therefore, at the first stage of AIDS development catalytic Abzs against different viral components can protect humans, similarly to the situation in bacterial and viral infections that do not stimulate AI reactions. Later, due to molecular mimicry between viral and host antigens, viral antigens can affect hematopoietic stem cells and trigger the development of AI processes.

## **5. Extreme diversity of AIDS abzymes**

#### **5.1 Structural diversity of AIDS abzymes**

DNase, RNase, ATPase, amylase, and protease Abzs may show very different contributions of variable domains of H- and L-chains to their active centers. Chromatographically separated light chains of IgGs from the sera of asthma patients were found to be active in the hydrolysis of VIP (Sun et al., 1994; 1997). The light chain of the VIP Abz was expressed in bacteria, purified, and found to possess an intrinsic catalytic activity (Tyutyulkova et al., 1996). The Abz-dependent hydrolysis of DNA and RNA by isolated light chains of IgGs from SLE, MS, asthma, and other AI patients, as well as from MRL-lpr/lpr mice, is more

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 167

iodoacetamide (a specific inhibitor of thiol proteases) usually has no remarkable effect on their proteolytic activity (Paul et al., 1989; Li et al., 1995; Kalaga et al., 1995; Odintsova et al., 2006; 2011; Polosukhina et al., 2004, 2005, 2006). Surprisingly, however, IN-hydrolyzing activity of all 10 IgGs was inhibited by iodoacetamide by 12-98 % (Baranova et al., 2009).

**1 2 3 4 5 6 7**

Fig. 6. SDS-PAGE analysis of a decrease in the intensity of protein band after its incubation with pIgGs from AIDS patients in different conditions. IN was incubated alone (1); in the presence of AIDS pIgGs from the sera of six different AIDS patients (IgG-12 – IgG19) and in

Similar surprising situation was observed for ten AIDS pIgMs with IN-hydrolyzing activity (Table 2) (Baranova et al., 2010). Leupeptin, significantly inhibited the proteolytic activity of only two of ten individual AIDS IgMs, and very weak inhibition was observed for one more preparation (Table 2). A specific inhibitor of acidic proteases, pepstatin A, significantly inhibited the IN-hydrolyzing activity of only three pIgMs and demonstrated weak inhibition of two preparations (Table 2). Surprisingly, a significant inhibition of serine protease-like activity by AEBSF was found only for two and weak suppression of the activity for another of ten AIDS IgMs. Proteolytic activity of five of ten AIDS IgMs was inhibited by 33–91% after incubating the IgMs with 0.01 M EDTA, while this chelating reagent at 0.1 M concentration decreased the relative activity of six preparations by 64–98% and inhibited three more preparations for ~8-10% (Table 2). As for AIDS IgGs, iodoacetamide inhibited the IN-hydrolyzing activity of all ten of ten Abs by 30–99% (Table 2). The inhibition of AIDS IgGs and IgMs with EDTA was comparable with that for IgGs from patients with MS

Iodoacetamide, a specific inhibitor of thiol proteases, usually does not significantly affect the activity of proteolytic Abzs (≤ 3–7% inhibition) (see above). Therefore, it was surprising that the IN-hydrolyzing activity of AIDS IgGs was suppressed by iodoacetamide in all 100% preparations by 12–98% (average value 65.7 ± 20.6%) in a stark contrast with other known proteolytic Abzs. A similar result was observed for AIDS IgMs; iodoacetamide suppressed the IN-hydrolyzing activity by 30–99% (average value 75.6 ± 21.2%) in all ten Ab preparations. Interestingly, there was no statistically significant difference in the inhibition of AIDS pIgGs and pIgMs by iodoacetamide (*p* = 0.2). Our findings support the idea that the pools of pIgGs and IgMs of AIDS patients can contain IN-hydrolyzing Abzs of four types

the absence of other components (2) or in the presence of AEBSF (3), leupeptin (4), iodoacetamide (5), EDTA (6), and pepstatin A (lane 7). Arrows indicate the positions of

IN

**IgG12**

**IgG13**

**IgG14**

**IgG15**

**IgG16**

**IgG19**

molecular mass markers.

(Polosukhina et al., 2004, 2005, 2006).

efficient than by intact Abs (Dubrovskaya et al., 2003; Galvita et al., 2007; Andrievskaya et al., 2000, 2002; Baranovskii et al., 2001; Nevinsky et al., 2005, 2010a, 2010b). A similar situation was observed for human milk IgGs and sIgAs with DNase and RNase activities (Kanyshkova et al., 1997; Nevinsky et al., 2000a). In addition, both H and L-chains of sIgAs had affinity to DNA-cellulose but only L-subunits hydrolyzed DNA and RNA (Nevinsky et al., 2000c). At the same time, it was demonstrated that the catalytic center of recombinant variable fragment (scFv) of DNase IgGs from AI-prone MRL-lpr/lpr mice may be located at the interface between the light and heavy chains and that after separation both of these chains are able to hydrolyze DNA (Kim et al., 2006).

It was shown that only separated light chains of AIDS IgGs hydrolyze DNA, while heavy chains is catalytically inactive (Odintsova et al., 2006a). On the contrary, both light and heavy chains of mouse IgGs after separation were active in the ATP hydrolysis (Andryushkova et al., 2009). Intact rat pIgGs and their separated H- and L-chains possess both peroxidase and oxidoreductase catalytic activity (Nevinsky et al., 2010a). The observed IN-hydrolyzing activity of AIDS pIgM L- and H-chains separated by SDS-PAGE (Fig. 3) may have different underlying causes. First, it is possible that these pIgMs contain a mixture of Abzs with only light or only heavy chains being catalytically active. However, similarly to mouse monoclonal DNase (Kim et al., 2006), catalytic centers of AIDS IgMs hydrolyzing IN may be located at the interface between the light and heavy chains, with both separated chains capable of hydrolysis of IN.

From the crystal structure of a catalytic Ab with esterase-like activity, it was concluded that the ligand *p*-nitrophenyl acetate interacts with amino acid residues of both light and heavy chains of Abzs and that both types of subunits are required for catalysis (Golinelli-Pimpaneau et al., 1994). Taken together, it is obvious that light and heavy chains of different Abzs including AIDS Abs can contribute to the active sites of Abzs in different ways.

The next question concerning the structural diversity of AIDS Abzs relates to the type of the proteolytic activity of their catalytic sites. Proteolytic IgGs from the sera of patients with asthma hydrolyzing VIP (Paul et al., 1989), Hashimoto thyroiditis and rheumatoid arthritis hydrolyzing thyroglobulin (Li eta l., 1995; Kalaga et al., 1995) are serine-like proteases, and their activity is most strongly reduced after incubation with specific serine protease inhibitors PMSF or AEBSF. It was shown that casein-hydrolyzing sIgAs from human milk (Odintsova et al., 2006; 2011) and hMBP-hydrolyzing IgGs and IgMs from the sera of patients with MS (Polosukhina et al., 2004, 2005, 2006) contain not only Ab subfractions with serine-like, but also specific subfractions with metal-dependent activity.

We have analyzed the type of proteolytic activity of AIDS pIgGs and pIgMs in the hydrolysis of IN and -casein. It was shown, that in contrast to milk sIgAs (Odintsova et al., 2006; 2011) and similarly to several other proteolytic Abzs, only specific inhibitor of serine protease AEBSF significantly suppress AIDS pIgG-dependent hydrolysis of -casein, while EDTA has no significant effect.

We have analyzed the type of IN-hydrolyzing activity of ten AIDS pIgGs; several typical examples are given in Fig. 6 (Baranova et al., 2009). Leupeptin, an inhibitor of many different proteases, demonstrated significant inhibition of proteolytic activity of only 2 of 10 individual AIDS IgGs (Table 2). A similar situation was observed for a specific inhibitor of acidic proteases, pepstatin A, which significantly inhibited IN-hydrolyzing activity of only two IgGs. Surprisingly, a significant inhibition of serine protease-like activity was also found only for 2 of 10 AIDS IgGs (Table 2). Proteolytic activity of five of ten AIDS IgGs was inhibited by 40-96 % after incubating the IgGs with EDTA. An incubation of IgGs with

efficient than by intact Abs (Dubrovskaya et al., 2003; Galvita et al., 2007; Andrievskaya et al., 2000, 2002; Baranovskii et al., 2001; Nevinsky et al., 2005, 2010a, 2010b). A similar situation was observed for human milk IgGs and sIgAs with DNase and RNase activities (Kanyshkova et al., 1997; Nevinsky et al., 2000a). In addition, both H and L-chains of sIgAs had affinity to DNA-cellulose but only L-subunits hydrolyzed DNA and RNA (Nevinsky et al., 2000c). At the same time, it was demonstrated that the catalytic center of recombinant variable fragment (scFv) of DNase IgGs from AI-prone MRL-lpr/lpr mice may be located at the interface between the light and heavy chains and that after separation both of these

It was shown that only separated light chains of AIDS IgGs hydrolyze DNA, while heavy chains is catalytically inactive (Odintsova et al., 2006a). On the contrary, both light and heavy chains of mouse IgGs after separation were active in the ATP hydrolysis (Andryushkova et al., 2009). Intact rat pIgGs and their separated H- and L-chains possess both peroxidase and oxidoreductase catalytic activity (Nevinsky et al., 2010a). The observed IN-hydrolyzing activity of AIDS pIgM L- and H-chains separated by SDS-PAGE (Fig. 3) may have different underlying causes. First, it is possible that these pIgMs contain a mixture of Abzs with only light or only heavy chains being catalytically active. However, similarly to mouse monoclonal DNase (Kim et al., 2006), catalytic centers of AIDS IgMs hydrolyzing IN may be located at the interface between the light and heavy chains, with both separated

From the crystal structure of a catalytic Ab with esterase-like activity, it was concluded that the ligand *p*-nitrophenyl acetate interacts with amino acid residues of both light and heavy chains of Abzs and that both types of subunits are required for catalysis (Golinelli-Pimpaneau et al., 1994). Taken together, it is obvious that light and heavy chains of different

The next question concerning the structural diversity of AIDS Abzs relates to the type of the proteolytic activity of their catalytic sites. Proteolytic IgGs from the sera of patients with asthma hydrolyzing VIP (Paul et al., 1989), Hashimoto thyroiditis and rheumatoid arthritis hydrolyzing thyroglobulin (Li eta l., 1995; Kalaga et al., 1995) are serine-like proteases, and their activity is most strongly reduced after incubation with specific serine protease inhibitors PMSF or AEBSF. It was shown that casein-hydrolyzing sIgAs from human milk (Odintsova et al., 2006; 2011) and hMBP-hydrolyzing IgGs and IgMs from the sera of patients with MS (Polosukhina et al., 2004, 2005, 2006) contain not only Ab subfractions with

We have analyzed the type of proteolytic activity of AIDS pIgGs and pIgMs in the hydrolysis of IN and -casein. It was shown, that in contrast to milk sIgAs (Odintsova et al., 2006; 2011) and similarly to several other proteolytic Abzs, only specific inhibitor of serine protease AEBSF significantly suppress AIDS pIgG-dependent hydrolysis of -casein, while

We have analyzed the type of IN-hydrolyzing activity of ten AIDS pIgGs; several typical examples are given in Fig. 6 (Baranova et al., 2009). Leupeptin, an inhibitor of many different proteases, demonstrated significant inhibition of proteolytic activity of only 2 of 10 individual AIDS IgGs (Table 2). A similar situation was observed for a specific inhibitor of acidic proteases, pepstatin A, which significantly inhibited IN-hydrolyzing activity of only two IgGs. Surprisingly, a significant inhibition of serine protease-like activity was also found only for 2 of 10 AIDS IgGs (Table 2). Proteolytic activity of five of ten AIDS IgGs was inhibited by 40-96 % after incubating the IgGs with EDTA. An incubation of IgGs with

Abzs including AIDS Abs can contribute to the active sites of Abzs in different ways.

serine-like, but also specific subfractions with metal-dependent activity.

chains are able to hydrolyze DNA (Kim et al., 2006).

chains capable of hydrolysis of IN.

EDTA has no significant effect.

iodoacetamide (a specific inhibitor of thiol proteases) usually has no remarkable effect on their proteolytic activity (Paul et al., 1989; Li et al., 1995; Kalaga et al., 1995; Odintsova et al., 2006; 2011; Polosukhina et al., 2004, 2005, 2006). Surprisingly, however, IN-hydrolyzing activity of all 10 IgGs was inhibited by iodoacetamide by 12-98 % (Baranova et al., 2009).

Fig. 6. SDS-PAGE analysis of a decrease in the intensity of protein band after its incubation with pIgGs from AIDS patients in different conditions. IN was incubated alone (1); in the presence of AIDS pIgGs from the sera of six different AIDS patients (IgG-12 – IgG19) and in the absence of other components (2) or in the presence of AEBSF (3), leupeptin (4), iodoacetamide (5), EDTA (6), and pepstatin A (lane 7). Arrows indicate the positions of molecular mass markers.

Similar surprising situation was observed for ten AIDS pIgMs with IN-hydrolyzing activity (Table 2) (Baranova et al., 2010). Leupeptin, significantly inhibited the proteolytic activity of only two of ten individual AIDS IgMs, and very weak inhibition was observed for one more preparation (Table 2). A specific inhibitor of acidic proteases, pepstatin A, significantly inhibited the IN-hydrolyzing activity of only three pIgMs and demonstrated weak inhibition of two preparations (Table 2). Surprisingly, a significant inhibition of serine protease-like activity by AEBSF was found only for two and weak suppression of the activity for another of ten AIDS IgMs. Proteolytic activity of five of ten AIDS IgMs was inhibited by 33–91% after incubating the IgMs with 0.01 M EDTA, while this chelating reagent at 0.1 M concentration decreased the relative activity of six preparations by 64–98% and inhibited three more preparations for ~8-10% (Table 2). As for AIDS IgGs, iodoacetamide inhibited the IN-hydrolyzing activity of all ten of ten Abs by 30–99% (Table 2). The inhibition of AIDS IgGs and IgMs with EDTA was comparable with that for IgGs from patients with MS (Polosukhina et al., 2004, 2005, 2006).

Iodoacetamide, a specific inhibitor of thiol proteases, usually does not significantly affect the activity of proteolytic Abzs (≤ 3–7% inhibition) (see above). Therefore, it was surprising that the IN-hydrolyzing activity of AIDS IgGs was suppressed by iodoacetamide in all 100% preparations by 12–98% (average value 65.7 ± 20.6%) in a stark contrast with other known proteolytic Abzs. A similar result was observed for AIDS IgMs; iodoacetamide suppressed the IN-hydrolyzing activity by 30–99% (average value 75.6 ± 21.2%) in all ten Ab preparations. Interestingly, there was no statistically significant difference in the inhibition of AIDS pIgGs and pIgMs by iodoacetamide (*p* = 0.2). Our findings support the idea that the pools of pIgGs and IgMs of AIDS patients can contain IN-hydrolyzing Abzs of four types

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 169

leupeptin and AEBSF for IgM7, and pepstatin A and AEBSF for IgM13 (Table 2). Very

In principle it is possible that the pools of IgGs and IgMs from AIDS patients may be "cocktails" of Abz molecules, with each molecule possessing only one of four alternative types of proteolytic activity: serine-, acidic-, thiol-, or metal-dependent. Yet the effects of two, three, and four inhibitors of different protease types did not always add to 100% inhibition. Only in three of ten IgM preparations (IgM8, IgM9, and IgM11) this sum was less or comparable with 100% (76–114%), while for other seven IgMs it varied from 125% to 289% (Table 2). Three IgG preparations (IgG9, IgG10, and IgG13) demonstrated this sum from 249 to 255 % (Table 2). Since IgM9 had only thiol protease-like activity, and IgM8 and IgM11 could be significantly suppressed (94–99%) only with iodoacetamide but lost their activity only marginally in the presence of EDTA or pepstatin A (by 7–8 %), it is most likely that in these patients most of the Abz molecules possess only the thiol protease type of proteolytic activity (Table 2). However, since the proteolytic activity in seven of ten IgMs and five of ten IgGs was summarily suppressed by specific inhibitors of serine, acidic, metaldependent, and thiol proteases by more than 100% (110–289%, Table 2), it is possible that the immune system of HIV-infected patients produces anti-IN Abzs with a combined structure of the active center, carrying amino acid residues typical of different proteases. For example, we suggest that the pools of IgM4, IgM7, and IgM13 (209–289% of the summarized inhibition) contain IgM molecules with extremely complicated active centers containing structural elements of thiol and metal-dependent proteases, which may be additionally combined with structural elements of the active centers of acidic proteases (IgM4), serine proteases (IgM7), or both (IgM13). Similar suggestion is reason in the case of several AIDS

Theoretically, a mammalian immune system can produce up to 106 variants of Abs against one antigen. An extreme diversity of RNase and DNase IgG and/or IgM Abzs from the sera of patients with MS and SLE and autoimmune prone MRL-lpr/lpr mice was observed (Baranovskii *et al.,* 1998, 2001; Andrievskaya *et al*., 2000, 2002; Kuznetsova *et al.,* 2007; Nevinsky *et al.*, 2003, 2005, 2010a, 2010b). It was shown that different patients (and animals) may have a relatively small or an extremely large pool of polyclonal nuclease Abzs containing different relative amounts of light chains of - and -types, demonstrating maximal activity at various optimal pHs, having a different net charge, activated or not by different metal ions, characterized by different substrate specificities. MS IgGs of all four subclasses (IgG1-IgG4) were catalytically active in the hydrolysis of hMBP (Legostaeva *et al*.,

We have analyzed the pH dependencies of the initial rates of DNA hydrolysis by AIDS catalytic IgGs (Odintsova et al., 2006a). Fig. 7A demonstrates three pH dependencies of different types which were revealed for catalytic pIgGs from the sera of 3 different patients. In contrast to all human DNases having one pronounced pH optimum in hydrolysis of DNA (Baranovskii et al., 2004), catalytic Abs usually show high DNase activity at a wide range of pH values between 5.5-9.0. Nevertheless, as one can see from Fig. 7A, one of the IgG preparations has a pronounced optimum at pH 8.0; second at pH 7.5, while third one

comparable situations were observed for AIDS IgGs and IgMs (Table 2).

IgGs (Table 2) (Baranova et al., 2009, 2010).

**5.2 pH optima diversity of AIDS abzymes** 

2010) and DNA (Parkhomenko *et al*., 2010).

demonstrates two marked pH optima at pH 7.5 and 8.5.


\*\*The decrease in the intensity of initial IN band estimated from SDS-PAGE electrophoresis data in the absence of inhibitor was taken for 100 %, for each preparation, a mean of 3 repeats are used. \*\*Sum of the effects of different compounds on the proteolytic activity

(leupeptin+Pepstatin+iodoacetamide + EDTA).

Table 2. Inhibition of proteolytic activity of individual IgGs and IgMs (from 10 AIDS patients) in the hydrolysis of HIV integrase by specific inhibitors of proteases of different types

resembling thiol, serine, acidic, and metal-dependent proteases, the ratio of which may be individual for every HIV-infected patient.

Interestingly, only IgM9 and three of IgGs (IgG11, IgG14, and IgG19) demonstrated significant inhibition by one inhibitor (iodoacetamide). At the same time other IgG and IgM preparations were sensitive to two or three inhibitors (Table 2). For example, IgM10 and IgM11 was strongly suppressed by iodoacetamide and EDTA; IgM23 was sensitive to AEBSF and iodoacetamide; IgM4 and IgM8 showed strong or at least some inhibition of the activity by three inhibitors (iodoacetamide, EDTA, and pepstatin A), while IgM12 was sensitive to leupeptin instead of pepstatin A (Table 2). Surprisingly, three of ten preparations (IgM5, IgM7, and IgM13) could be significantly inhibited by four different inhibitors. Of these, iodoacetamide and EDTA were common inhibitors for all three IgM preparations, while two other inhibitors were different: pepstatin A and leupeptin for IgM5,

AEBSF Leupeptin Pepstatin A Iodoacetamide 0.1 M

IgG9 42±5 74±8 51±5 85±9 0 252 IgG10 0 70±7 0 83±9 96±9 249 IgG11 0 0 0 66±7 0 66 IgG12 0 0 0 66±5 44±5 110 IgG13 0 0 59±6 98±8 98±8 255 IgG14 0 0 0 87±7 3±1 90 IgG15 0 11±3 0 33±4 45±4 89 IgG16 0 0 0 78±8 40±3 118 IgG17 49±5 0 0 12±1.5 0 61 IgG19 0 0 0 49±5 2±1 51 IgM4 0 0 48 ± 5 97 ± 8 98 ± 2 243 IgM5 0 47 ± 5 36 ± 5 36 ± 5 80 ± 7 199 IgM7 8 ± 1 68 ± 7 0 95 ± 9 94 ± 8 265 IgM8 0 0 7 ± 2 99 ± 8 8 ± 2 114 IgM9 0 0 0 76 ± 6 0 76 IgM10 0 0 0 83 ± 7 93 ± 8 176 IgM11 0 0 0 94 ± 7 8 ± 3 102 IgM12 0 7 ± 1 0 55 ± 5 98 ± 8 153 IgM13 46 ± 5 0 88 ± 9 91 ± 10 64 ± 4 289 IgM23 80 ± 8 0 5 ± 2 30± 5 10 ± 2 125 \*\*The decrease in the intensity of initial IN band estimated from SDS-PAGE electrophoresis data in the

absence of inhibitor was taken for 100 %, for each preparation, a mean of 3 repeats are used.

in the hydrolysis of HIV integrase by specific inhibitors of proteases of different types

Table 2. Inhibition of proteolytic activity of individual IgGs and IgMs (from 10 AIDS patients)

resembling thiol, serine, acidic, and metal-dependent proteases, the ratio of which may be

Interestingly, only IgM9 and three of IgGs (IgG11, IgG14, and IgG19) demonstrated significant inhibition by one inhibitor (iodoacetamide). At the same time other IgG and IgM preparations were sensitive to two or three inhibitors (Table 2). For example, IgM10 and IgM11 was strongly suppressed by iodoacetamide and EDTA; IgM23 was sensitive to AEBSF and iodoacetamide; IgM4 and IgM8 showed strong or at least some inhibition of the activity by three inhibitors (iodoacetamide, EDTA, and pepstatin A), while IgM12 was sensitive to leupeptin instead of pepstatin A (Table 2). Surprisingly, three of ten preparations (IgM5, IgM7, and IgM13) could be significantly inhibited by four different inhibitors. Of these, iodoacetamide and EDTA were common inhibitors for all three IgM preparations, while two other inhibitors were different: pepstatin A and leupeptin for IgM5,

\*\*Sum of the effects of different compounds on the proteolytic activity

(leupeptin+Pepstatin+iodoacetamide + EDTA).

individual for every HIV-infected patient.

Inhibition, %\*

EDTA

Sum of effects\*\*

Number of prep.

leupeptin and AEBSF for IgM7, and pepstatin A and AEBSF for IgM13 (Table 2). Very comparable situations were observed for AIDS IgGs and IgMs (Table 2).

In principle it is possible that the pools of IgGs and IgMs from AIDS patients may be "cocktails" of Abz molecules, with each molecule possessing only one of four alternative types of proteolytic activity: serine-, acidic-, thiol-, or metal-dependent. Yet the effects of two, three, and four inhibitors of different protease types did not always add to 100% inhibition. Only in three of ten IgM preparations (IgM8, IgM9, and IgM11) this sum was less or comparable with 100% (76–114%), while for other seven IgMs it varied from 125% to 289% (Table 2). Three IgG preparations (IgG9, IgG10, and IgG13) demonstrated this sum from 249 to 255 % (Table 2). Since IgM9 had only thiol protease-like activity, and IgM8 and IgM11 could be significantly suppressed (94–99%) only with iodoacetamide but lost their activity only marginally in the presence of EDTA or pepstatin A (by 7–8 %), it is most likely that in these patients most of the Abz molecules possess only the thiol protease type of proteolytic activity (Table 2). However, since the proteolytic activity in seven of ten IgMs and five of ten IgGs was summarily suppressed by specific inhibitors of serine, acidic, metaldependent, and thiol proteases by more than 100% (110–289%, Table 2), it is possible that the immune system of HIV-infected patients produces anti-IN Abzs with a combined structure of the active center, carrying amino acid residues typical of different proteases. For example, we suggest that the pools of IgM4, IgM7, and IgM13 (209–289% of the summarized inhibition) contain IgM molecules with extremely complicated active centers containing structural elements of thiol and metal-dependent proteases, which may be additionally combined with structural elements of the active centers of acidic proteases (IgM4), serine proteases (IgM7), or both (IgM13). Similar suggestion is reason in the case of several AIDS IgGs (Table 2) (Baranova et al., 2009, 2010).

#### **5.2 pH optima diversity of AIDS abzymes**

Theoretically, a mammalian immune system can produce up to 106 variants of Abs against one antigen. An extreme diversity of RNase and DNase IgG and/or IgM Abzs from the sera of patients with MS and SLE and autoimmune prone MRL-lpr/lpr mice was observed (Baranovskii *et al.,* 1998, 2001; Andrievskaya *et al*., 2000, 2002; Kuznetsova *et al.,* 2007; Nevinsky *et al.*, 2003, 2005, 2010a, 2010b). It was shown that different patients (and animals) may have a relatively small or an extremely large pool of polyclonal nuclease Abzs containing different relative amounts of light chains of - and -types, demonstrating maximal activity at various optimal pHs, having a different net charge, activated or not by different metal ions, characterized by different substrate specificities. MS IgGs of all four subclasses (IgG1-IgG4) were catalytically active in the hydrolysis of hMBP (Legostaeva *et al*., 2010) and DNA (Parkhomenko *et al*., 2010).

We have analyzed the pH dependencies of the initial rates of DNA hydrolysis by AIDS catalytic IgGs (Odintsova et al., 2006a). Fig. 7A demonstrates three pH dependencies of different types which were revealed for catalytic pIgGs from the sera of 3 different patients. In contrast to all human DNases having one pronounced pH optimum in hydrolysis of DNA (Baranovskii et al., 2004), catalytic Abs usually show high DNase activity at a wide range of pH values between 5.5-9.0. Nevertheless, as one can see from Fig. 7A, one of the IgG preparations has a pronounced optimum at pH 8.0; second at pH 7.5, while third one demonstrates two marked pH optima at pH 7.5 and 8.5.

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 171

different individuals analyzed, but also other subfractions of Abs, the activity of which may be comparable with or less than that of the major subfraction. In addition, the number of *K*<sup>m</sup> and *V*max values, characterizing interaction of different monoclonal of polyclonal Abzs with their specific substrates, can significantly increase when they measured at several pH values

It was shown previously, that nuclease and protease Abzs from the sera of AI patients and animals are very heterogeneous in their affinity for cognate substrate and can be separated into many fractions by chromatography on affinity resins bearing immobilized substrate (Baranovskii et al., 2001; Andrievskaya et al., 2002; Semenov et al., 2004; Kuznetsova et al., 2007; Nevinsky et al., 1998, 2005, 2010a, 2010b). We have analyzed the affinity of AIDS pIgGs for human -casein by chromatography on casein-Sepharose (Odintsova et al., 2006b). Interestingly, when IgGs were eluted from casein-Sepharose by a KCl gradient (0–3 M), the protein (and casein-hydrolyzing activity) was distributed all over the chromatography profile. A similar result was obtained at AIDS IgGs chromatography on RT- and HSA-Sepharoses (Odintsova et al., 2006b). The data indicate for extreme affinity heterogeneity of casein-, RT-, and HSA-hydrolyzing abzymes to cognate protein substrates (Odintsova et al.,

We have subjected an equimolar mixture of pIgGs (and IgMs) from five AIDS patients to affinity chromatography on IN-Sepharose. Only 15±3 % of the total IgGs (Fig. 8A) and 173 % of the total IgMs (Fig. 8B) were bound to IN-Sepharose (Baranova et al., 2009, 2010). As we have shown previously, the fraction of Abzs with different catalytic activities including in the serum of AI patients usually does not exceed 0.1-5 % of total Igs (Nevinsky et al., 2005, 2010a, 2010b). Therefore, it was surprising that IN-Sepharose can bind up to 15-17 % of the total pIgGs and pIgMs. At the same time, IN is known as a very hydrophobic protein which can interact nonspecifically with different hydrophobic compounds including other proteins. Taking this into account we could suppose that immobilized IN binds anti-IN pIgGs and pIgMs in a specific manner, and interacts with some other IgGs and IgMs non-specifically.

A B

**RA, %**

**80**

**0**

(5 l of dialyzed eluate, total volume of the reaction mixture 10 l) corresponding to complete hydrolysis of 0.3 mg/ml IN after 16 h of incubation was taken for 100%.

Fig. 8. Affinity chromatography of the mixture of five individual AIDS pIgGs (A) and pIgMs (B) on IN-Sepharose: (), absorbance at 280 nm; the relative activity (RA) of IgGs and IgMs

**20**

**V**

**40**

**60**

**0 5 10 15 20 25**

**Fraction number**

**III**

**Gradient**

**<sup>4</sup> <sup>3</sup> <sup>2</sup>**

**3 M MgCl2**

**Gly-HCl, pH 2.6**

**IV**

**2 M MgCl2**

**3 M NaCl**

**0.0 0.2 0.4 0.6 0.8 1.0 1.2**

**Loading**

**I**

**1**

**0 5 10 15 20 25**

**2**

**II**

**Fraction number**

**Gradient**

**III**

**3 M NaCl 0.2 M NaCl Gly-HCl, pH 2.6 A280 RA, %**

**3 M MgCl2 2 M MgCl2**

**IV**

**3**

**4**

**V**

**5.3 Affinity and relative catalytic activity diversity of AIDS abzymes** 

(Nevinsky, 2010b).

2006).

**0.2**

**0.4 0.6**

**I**

**1**

**II**

**0.2 M NaCl Loading**

**0.8**

**1.0**

**A280**

We have analyzed the pH dependencies of the initial rates of human casein hydrolysis by four individual AIDS IgGs. In contrast to all human proteases having one pronounced pH optimum (Horl et al., 1987; Rao et al., 1998), catalytic AIDS IgGs demonstrated high specific casein-hydrolyzing activity within a wide range of pH values (5.0–9.0) and the pH profile for each IgG was unique (Fig. 7B) (Odintsova et al., 2006).

It is well known that canonical mammalian, bacterial, and plant proteases, depending on their biological function, can have optimal pH values ranging from acidic (2.0) to neutral and alkaline (8-10) (Horl et al., 1987; Rao et al., 1998). We have measured the relative activity of AIDS IgGs and IgMs at pH from 3.0 to 10.5. In contrast to all human proteases, catalytic IgGs demonstrated high specific IN-hydrolyzing activity within a wide range of pH values (3.0–10).

Fig. 7. pH dependences of the relative DNase (A), casein-hydrolyzing (B) activity of IgGs, and IN-hydrolyzing activity (RA) of individual pIgGs (C) and pIgMs (D) from the sera of several different AIDS patients (number of Ab correspond to number of patient). The average error in the initial rate determination at each pH from two experiments did not exceed 7-10 %.

The pH profile for each IgG and IgM was unique; each preparation demonstrated from three to seven brightly expressed optima at different pH from 3 to 11 (Fig. 7). Taking into account the effective hydrolysis of IN at pH 3.0, one cannot exclude that human immune system of AIDS patients could in principle produce IgGs and IgMs with a proteolytic activity similar to that of stomach acidic proteases. The above results clearly demonstrate that pIgGs and pIgMs from individual AIDS patients can consist of different sets of catalytic Ab subfractions demonstrating quite distinct enzymic properties in the hydrolysis of DNA, human casein, and integrase.

Overall, a pool of many auto-Abs may contain very different monoclonal Abzs with various pH optima. It should be mentioned that the RAs of Abzs from patients with different AI and viral diseases are usually compared at one fixed pH, in which all samples are more or less active. Changing the reaction pH, one can reveal not only the major fraction of Abzs in

We have analyzed the pH dependencies of the initial rates of human casein hydrolysis by four individual AIDS IgGs. In contrast to all human proteases having one pronounced pH optimum (Horl et al., 1987; Rao et al., 1998), catalytic AIDS IgGs demonstrated high specific casein-hydrolyzing activity within a wide range of pH values (5.0–9.0) and the pH profile

It is well known that canonical mammalian, bacterial, and plant proteases, depending on their biological function, can have optimal pH values ranging from acidic (2.0) to neutral and alkaline (8-10) (Horl et al., 1987; Rao et al., 1998). We have measured the relative activity of AIDS IgGs and IgMs at pH from 3.0 to 10.5. In contrast to all human proteases, catalytic IgGs demonstrated high specific IN-hydrolyzing activity within a wide range of pH values

for each IgG was unique (Fig. 7B) (Odintsova et al., 2006).

**3 4 5 6 7 8 9 10 11 12**

**pH**

 **IgG3**

Fig. 7. pH dependences of the relative DNase (A), casein-hydrolyzing (B) activity of IgGs, and IN-hydrolyzing activity (RA) of individual pIgGs (C) and pIgMs (D) from the sera of several different AIDS patients (number of Ab correspond to number of patient). The average error in the initial rate determination at each pH from two experiments did not

The pH profile for each IgG and IgM was unique; each preparation demonstrated from three to seven brightly expressed optima at different pH from 3 to 11 (Fig. 7). Taking into account the effective hydrolysis of IN at pH 3.0, one cannot exclude that human immune system of AIDS patients could in principle produce IgGs and IgMs with a proteolytic activity similar to that of stomach acidic proteases. The above results clearly demonstrate that pIgGs and pIgMs from individual AIDS patients can consist of different sets of catalytic Ab subfractions demonstrating quite distinct enzymic properties in the hydrolysis of DNA,

Overall, a pool of many auto-Abs may contain very different monoclonal Abzs with various pH optima. It should be mentioned that the RAs of Abzs from patients with different AI and viral diseases are usually compared at one fixed pH, in which all samples are more or less active. Changing the reaction pH, one can reveal not only the major fraction of Abzs in

**RA, %**

**40**

**60**

**80**

**100**

**RA, %**

**IgM1**

**4.8 6.0 7.2 8.4 9.6**

**56789**

**IgG16**

**IgG8 IgG14 IgG4**

 **IgM2**

 **IgM4**

**D**

**B**

**pH**

 **IgM3**

**pH**

 **IgG1**

**IgG7 IgG6**

**pH**

**IgG5**

exceed 7-10 %.

human casein, and integrase.

**RA, %**

 **IgG2**

**C**

**A**

**56789**

(3.0–10).

different individuals analyzed, but also other subfractions of Abs, the activity of which may be comparable with or less than that of the major subfraction. In addition, the number of *K*<sup>m</sup> and *V*max values, characterizing interaction of different monoclonal of polyclonal Abzs with their specific substrates, can significantly increase when they measured at several pH values (Nevinsky, 2010b).

#### **5.3 Affinity and relative catalytic activity diversity of AIDS abzymes**

It was shown previously, that nuclease and protease Abzs from the sera of AI patients and animals are very heterogeneous in their affinity for cognate substrate and can be separated into many fractions by chromatography on affinity resins bearing immobilized substrate (Baranovskii et al., 2001; Andrievskaya et al., 2002; Semenov et al., 2004; Kuznetsova et al., 2007; Nevinsky et al., 1998, 2005, 2010a, 2010b). We have analyzed the affinity of AIDS pIgGs for human -casein by chromatography on casein-Sepharose (Odintsova et al., 2006b). Interestingly, when IgGs were eluted from casein-Sepharose by a KCl gradient (0–3 M), the protein (and casein-hydrolyzing activity) was distributed all over the chromatography profile. A similar result was obtained at AIDS IgGs chromatography on RT- and HSA-Sepharoses (Odintsova et al., 2006b). The data indicate for extreme affinity heterogeneity of casein-, RT-, and HSA-hydrolyzing abzymes to cognate protein substrates (Odintsova et al., 2006).

We have subjected an equimolar mixture of pIgGs (and IgMs) from five AIDS patients to affinity chromatography on IN-Sepharose. Only 15±3 % of the total IgGs (Fig. 8A) and 173 % of the total IgMs (Fig. 8B) were bound to IN-Sepharose (Baranova et al., 2009, 2010).

As we have shown previously, the fraction of Abzs with different catalytic activities including in the serum of AI patients usually does not exceed 0.1-5 % of total Igs (Nevinsky et al., 2005, 2010a, 2010b). Therefore, it was surprising that IN-Sepharose can bind up to 15-17 % of the total pIgGs and pIgMs. At the same time, IN is known as a very hydrophobic protein which can interact nonspecifically with different hydrophobic compounds including other proteins. Taking this into account we could suppose that immobilized IN binds anti-IN pIgGs and pIgMs in a specific manner, and interacts with some other IgGs and IgMs non-specifically.

Fig. 8. Affinity chromatography of the mixture of five individual AIDS pIgGs (A) and pIgMs (B) on IN-Sepharose: (), absorbance at 280 nm; the relative activity (RA) of IgGs and IgMs (5 l of dialyzed eluate, total volume of the reaction mixture 10 l) corresponding to complete hydrolysis of 0.3 mg/ml IN after 16 h of incubation was taken for 100%.

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 173

**12-18 M**

**V, M IN / 17 h**

Fig. 9. The dependencies of the initial rates of IN hydrolysis upon the IN concentration in the reaction catalyzed by non-separated by affinity chromatography two individual pIgMs (A) and two pIgGs (B) from different patients in coordinates V *vs* [S]. IgM12, IgM32, IgG15, and IgG32 were used in different concentrations. Arrows show different hyperbolic

**0 10 20 30 40 50 60 70**

**[IN], M**

**30-50 M**

**B**

**-0,02 -0,01 0,00 0,01 0,02 0,03 0,04 0,05**

**[IN], M**

**IgG**

**IgM**

**1 / V, min / M IN**

Fig. 10. The dependencies of the initial rates of IN hydrolysis upon the IN concentration in the reaction catalyzed by fractions of pIgGmix and pIgMmix purified on IN-Sepharose (Fig. 8; fractions number 2; IgGload and IgMload) in coordinates V vs [S] (A). Determination of the *Km*  for IN and *Vmax* values using the Lineweaver-Burk plot in the case of the same IgGload and

fraction 24 eluted with an acidic buffer) demonstrated 3– and 10-fold respectively higher affinity than that for pIgMload fraction (130± 30 M). The *Km* and *kcat* values corresponding to fractionated pIgGs and pIgMs agree with the relative contents and specific activities of the major Ab subfractions isolated by affinity chromatography (Fig. 8). On overall, the above data demonstrate extreme diversity of HIV IgGs and IgMs in their affinity to IN and in the

Affinity chromatography of DNase and RNase Abs from AI patients and animals (Baranovskii et al., 2001; Andrievskaya et al., 2002; Kuznetsova et al., 2007), healthy rabbits immunized with DNA, RNA, DNase I, DNase II, pancreatic RNase (Krasnorutskii et al.

**IgG32**

**80-100 M**

**IgG15**

**0 5 10 15 20 25 30 35**

**15-20 M**

**A**

**[IN], M**

**0 20 40 60 80 100 120 140 0,00**

**[IN], M**

**IgM32**

**30-50 M**

**IgM12**

fragments of complicated curves corresponding to the total dependencies.

**IgG**

**A IgM B**

**0**

**0,02 0,04 0,06 0,08 0,10 0,12 0,14**

**V, M IN / min**

IgMload (B).

relative *kcat* values (Figs 8-10).

**2**

**5-10 M**

**V, M IN / 8 h**

**4**

**6**

The pIgG and IgM fractions of the first peak (83-85% of protein loaded on IN-Sepharose) with a very low affinity for IN possess high IN-hydrolyzing activity (peak 1, Fig. 8). As it was shown previously non-separated on affinity resins pIgGs and IgMs from AIDS patients contain small subfractions hydrolyzing specifically not only HIV IN (Baranova et al., 2009, 2010), but also HIV RT, HSA and human casein (Odintsova et al., 2006). However, the fractions of pIgGs and pIgMs having high affinity to IN-Sepharose and eluted from this sorbent with different concentrations of NaCl, MgCl2 and acidic buffer (protein peaks 2-4, activity peaks II-V; Fig. 8) hydrolyzed only IN. Thus, IgGs and IgMs with IN-independent activities do not have affinity for IN-Sepharose, but some other Abs can be bound with IN non-specifically.

The total IN-hydrolyzing activities of pIgGs and pIgMs were distributed all over the chromatography profiles and in the case of both Abs five peaks of IN-hydrolyzing activity (I – V, Fig. 8) were brightly expressed. The data obtained are indicative of extreme heterogeneity of IN-hydrolyzing pIgGs and pIgMs in their affinity to IN.

When Abzs are highly heterogeneous, the dependence of *V* on the substrate concentration for non-fractionated Abs may appear inconsistent with simple Michaelis-Menten kinetics and may be described by a sum of several hyperbolic curves corresponding to different Ab subfractions. However, the contribution of some subfractions to the total curve may be small, or they may have comparable *Km* and *V*max (*kcat*) values. As a rule, only when significant differences (≥ 5-fold) exist between the *Km* and *V*max values for different Ab subfractions it is possible to determine these parameters characterizing individual subfractions of polyclonal Abz from the aggregated initial rate curves.

First, we have measured the *Km* and *V*max values in the reaction of IN hydrolysis using two individual preparations of pIgGs and pIgMs not fractionated on IN-Sepharose. The initial rate data obtained for these Abs at the increasing IN concentration were inconsistent with the Michaelis-Menten kinetics and the dependences corresponded to at least three or four hyperbolic curves with several segments reflecting different *Km* values, which were approximately in the ranges of 5-10, 15-20, 30-50, and higher than 70-100 M (Fig. 9). Similar situation was observed for the mixtures of equal amounts of electrophoretically homogeneous IgGs (pIgGmix) and IgMs (pIgMmix) from the sera of five AIDs patients.

For more detailed analysis of *Km* and *kcat* values characterizing different Ab fractions within total pool of Abzs (pIgGmix and pIgMmix) we have analyzed several individual pIgGmix and pIgMmix fractions eluted from IN-Sepharose (Fig. 8). First, we have measured the *Km* and *V*max values in the reaction catalyzed by IgGmix (IgGload) and IgMmix (IgMload) corresponding to the second fraction eluted under loading of Abs on IN-Sepharose (Fig. 8). The dependencies of V/[S] (hyperbolic curves; Fig. 10A) and 1/V *vs* 1/[S] (Fig. 10B) demonstrated virtually normal Michaelis-Menten character for second fractions of pIgGs and pIgMs. The *Km* and *kcat* for IgGload (156± 40 M; 0.3±0.1 min-1) and IgMload (130± 30 M; 2.0±0.4 min-1) were determined. The V/[S] and 1/V *vs* 1/[S] dependences for individual fractions of pIgGs and pIgMs eluted from IN-Sepharose in gradient of NaCl concentration (IgGsalt and IgMsalt) and by acidic buffer (IgGacid and IgMacid) had also typical Michaelis-Menten character.

The affinity of pIgGs for IN (in terms of *Km* values) increased with the increase of their affinity to IN-Sepharose; for IgGsalt (*Km* = 44±4.0 M) corresponding to fraction 8 (eluted with the salt) and IgGacid (*Km* = 14±1.0 M) corresponding to fraction 24 (eluted with an acidic buffer) the affinity was 3.5– and 11-fold respectively higher than that for IgGload (156±40 M). Similar situation was observed for the separated individual fractions of IgMmix (Fig. 8B); pIgMsolid (*Km* =43±4.0 M; fraction 10 eluted with NaCl) and pIgMacid (*Km* = 12.8±1.0 M;

The pIgG and IgM fractions of the first peak (83-85% of protein loaded on IN-Sepharose) with a very low affinity for IN possess high IN-hydrolyzing activity (peak 1, Fig. 8). As it was shown previously non-separated on affinity resins pIgGs and IgMs from AIDS patients contain small subfractions hydrolyzing specifically not only HIV IN (Baranova et al., 2009, 2010), but also HIV RT, HSA and human casein (Odintsova et al., 2006). However, the fractions of pIgGs and pIgMs having high affinity to IN-Sepharose and eluted from this sorbent with different concentrations of NaCl, MgCl2 and acidic buffer (protein peaks 2-4, activity peaks II-V; Fig. 8) hydrolyzed only IN. Thus, IgGs and IgMs with IN-independent activities do not have affinity for IN-Sepharose, but some other Abs can be bound with IN non-specifically. The total IN-hydrolyzing activities of pIgGs and pIgMs were distributed all over the chromatography profiles and in the case of both Abs five peaks of IN-hydrolyzing activity (I – V, Fig. 8) were brightly expressed. The data obtained are indicative of extreme

When Abzs are highly heterogeneous, the dependence of *V* on the substrate concentration for non-fractionated Abs may appear inconsistent with simple Michaelis-Menten kinetics and may be described by a sum of several hyperbolic curves corresponding to different Ab subfractions. However, the contribution of some subfractions to the total curve may be small, or they may have comparable *Km* and *V*max (*kcat*) values. As a rule, only when significant differences (≥ 5-fold) exist between the *Km* and *V*max values for different Ab subfractions it is possible to determine these parameters characterizing individual

First, we have measured the *Km* and *V*max values in the reaction of IN hydrolysis using two individual preparations of pIgGs and pIgMs not fractionated on IN-Sepharose. The initial rate data obtained for these Abs at the increasing IN concentration were inconsistent with the Michaelis-Menten kinetics and the dependences corresponded to at least three or four hyperbolic curves with several segments reflecting different *Km* values, which were approximately in the ranges of 5-10, 15-20, 30-50, and higher than 70-100 M (Fig. 9). Similar situation was observed for the mixtures of equal amounts of electrophoretically

homogeneous IgGs (pIgGmix) and IgMs (pIgMmix) from the sera of five AIDs patients.

For more detailed analysis of *Km* and *kcat* values characterizing different Ab fractions within total pool of Abzs (pIgGmix and pIgMmix) we have analyzed several individual pIgGmix and pIgMmix fractions eluted from IN-Sepharose (Fig. 8). First, we have measured the *Km* and *V*max values in the reaction catalyzed by IgGmix (IgGload) and IgMmix (IgMload) corresponding to the second fraction eluted under loading of Abs on IN-Sepharose (Fig. 8). The dependencies of V/[S] (hyperbolic curves; Fig. 10A) and 1/V *vs* 1/[S] (Fig. 10B) demonstrated virtually normal Michaelis-Menten character for second fractions of pIgGs and pIgMs. The *Km* and *kcat* for IgGload (156± 40 M; 0.3±0.1 min-1) and IgMload (130± 30 M; 2.0±0.4 min-1) were determined. The V/[S] and 1/V *vs* 1/[S] dependences for individual fractions of pIgGs and pIgMs eluted from IN-Sepharose in gradient of NaCl concentration (IgGsalt and IgMsalt) and by acidic buffer

The affinity of pIgGs for IN (in terms of *Km* values) increased with the increase of their affinity to IN-Sepharose; for IgGsalt (*Km* = 44±4.0 M) corresponding to fraction 8 (eluted with the salt) and IgGacid (*Km* = 14±1.0 M) corresponding to fraction 24 (eluted with an acidic buffer) the affinity was 3.5– and 11-fold respectively higher than that for IgGload (156±40 M). Similar situation was observed for the separated individual fractions of IgMmix (Fig. 8B); pIgMsolid (*Km* =43±4.0 M; fraction 10 eluted with NaCl) and pIgMacid (*Km* = 12.8±1.0 M;

heterogeneity of IN-hydrolyzing pIgGs and pIgMs in their affinity to IN.

subfractions of polyclonal Abz from the aggregated initial rate curves.

(IgGacid and IgMacid) had also typical Michaelis-Menten character.

Fig. 9. The dependencies of the initial rates of IN hydrolysis upon the IN concentration in the reaction catalyzed by non-separated by affinity chromatography two individual pIgMs (A) and two pIgGs (B) from different patients in coordinates V *vs* [S]. IgM12, IgM32, IgG15, and IgG32 were used in different concentrations. Arrows show different hyperbolic fragments of complicated curves corresponding to the total dependencies.

Fig. 10. The dependencies of the initial rates of IN hydrolysis upon the IN concentration in the reaction catalyzed by fractions of pIgGmix and pIgMmix purified on IN-Sepharose (Fig. 8; fractions number 2; IgGload and IgMload) in coordinates V vs [S] (A). Determination of the *Km*  for IN and *Vmax* values using the Lineweaver-Burk plot in the case of the same IgGload and IgMload (B).

fraction 24 eluted with an acidic buffer) demonstrated 3– and 10-fold respectively higher affinity than that for pIgMload fraction (130± 30 M). The *Km* and *kcat* values corresponding to fractionated pIgGs and pIgMs agree with the relative contents and specific activities of the major Ab subfractions isolated by affinity chromatography (Fig. 8). On overall, the above data demonstrate extreme diversity of HIV IgGs and IgMs in their affinity to IN and in the relative *kcat* values (Figs 8-10).

Affinity chromatography of DNase and RNase Abs from AI patients and animals (Baranovskii et al., 2001; Andrievskaya et al., 2002; Kuznetsova et al., 2007), healthy rabbits immunized with DNA, RNA, DNase I, DNase II, pancreatic RNase (Krasnorutskii et al.

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 175

vary in the range of 0.001–40 min–1 (Gololobov et al., 1995; Kalaga et al., 1995; Nevinsky et al., 2005, 2010a, 20010b). There are only several exceptions. For example, the specific activity of RNase IgGs from AI patients was about 1-20% of that for RNase A and of six known human sera RNases, while poly(A) was hydrolyzed by Abzs 2-10-fold faster than by RNase A, one of the most active RNases known (Buneva et al., 1994; Baranovskii et al., 1997, 1998; Vlasov et al., 1998). At the same time, the specific activity of homogeneous Abzs of several SLE and MS patients was about 40-400% of that for RNase A (Baranovskii et al., 1998). In addition, the specific nucleotide-hydrolyzing activities of mouse polyclonal IgGs is 1–4 orders of magnitude higher that those of known natural Abzs (Andryushkova et al., 2009). The *kcat* values for AIDS pIgGs hydrolyzing scDNA ((2.1-29.6)×10-2 min-1), human casein (2.0×10-2 min-1), HSA (3.6×10-5 min-1), and HIV RT (2.2×10-3 min-1), as well IgGs (0.3-2.9 min-1) and IgMs (2.0-6.4 min-1) purified on IN-Sepharose (Table 2) in the IN hydrolysis were comparable with those for known Abzs (Odintsova et al., 2006a, 2006b; Baranova et al., 2009,

Currently there are no methods that can efficiently separate Abzs from catalytically inactive Abs against the same antigen. In addition, as it was shown above, IN-Sepharose interact not only with anti-IN IgMs and IgGs but bind non-specifically some other Abs. Even partial purification of IgGs and IgMs on IN-Sepharose (or other specific affinity resins) leads to significant increase in the *kcat* value for IN and other substrates hydrolysis. Since the specific activities in all cases were calculated using the total concentration of purified pIgGs and pIgMs and affinity chromatography on IN-Sepharose (and other affinity sorbents) cannot separate catalytically active and inactive anti-IN Abs, the specific IN-hydrolyzing activities of the individual monoclonal subfractions in the pIgG and pIgM pools may be higher than those of non-fractionated or partially fractionated Abs. It should be mentioned that specific activities of some Abzs are often significantly lower than those for canonical enzymes with the same catalytic activities. However, it is impossible to dismiss the RAs of Abs from patients with AI and viral infection as biologically inessential since they are comparable with those for many canonical and very important enzymes, for example, restriction endonucleases and repair enzymes (Gololobov et al., 1995; Nevinsky et al., 2005, 2010, 2010). Thus, IN-hydrolyzing IgGs and IgMs from HIV-infected patients are very heterogeneous in their affinity to IN-Sepharose, demonstrate different *Km* and *Vmax* values and different subfractions of Abzs can hydrolyze various substrates at pH from 3 to 10. In addition, in contrast to other Abzs with proteolytic activity they can possess for different types of

proteolytic activities: thiol-, metal-dependent, serine- and acidic-like.

**6.1 Casein hydrolysis by AIDS abzymes** 

**6. Peculiarities of protein hydrolysis by AIDS abzymes and canonical** 

syndrome, polymyositis, AI myocarditis, etc. (Amino *et al.*, 1982; Freeman *et al.*, 1986).

Casein hydrolyzing Abzs was found not only in the sera of HIV-infected patients (Odintsova et al., 2006b) but also in the milk of lactating women (Odintsova et al., 2005; 2011). At the first glance, no obvious immunizing factors could be found in clinically healthy pregnant and lactating women. However, pregnancy could "activate" or "trigger" autoimmune-like manifestations in clinically healthy women, and a sharp exacerbation of AI reactions can occur in some cases soon after childbirth (Amino *et al.*, 1982; Freeman *et al.*, 1986). Postnatal AI pathologies arise sometimes, including SLE, Hashimoto's thyroiditis, phospholipids

2010).

**proteases** 

2008a, 2008b, 2008c, 2008d, 2009) as well as AI Abs with proteolytic and other activities (Nevinsky et al, 1998; Semenov et al., 2004; Legostaeva et al., 2010) on resins bearing immobilized specific substrates using elution of Abs with different concentration of NaCl and an acidic buffer always leads to separation of Abs into many Abz subfractions with different affinity to immobilized substrate. In addition, the affinity of separated fractions for immobilized substrate increased gradually with the increase in eluting NaCl concentration, but the *Km* (and *Vmax*) values corresponding to each fraction eluted from affinity sorbent are individual for every patient analyzed. It means, that the apparent number of monoclonal Abzs with different catalytic properties within the polyclonal Abs pool may be significantly underestimated since it is impossible to separate Abzs with comparable affinities for a specific substrate or to distinguish monoclonal Abzs with similar kinetic parameters. Some minor monoclonal Abzs with a relatively high activity and even major Abzs with low activity may be hidden by major Abzs with high activity. As a consequence, the *Km* (and *Vmax*) values determined using Abzs not fractionated on affinity sorbents more often characterize the interaction of substrates with a major fraction of Abzs with the maximal content and highest relative enzymatic activity at condition used. At the same time, these characteristic are very useful for comparisons of Abs with different substrate specificities and from patients with different diseases. For example, it was shown that affinity of specific major fractions of AIDS IgGs non-fractionated on affinity resins to HSA (*Km* = (1.8 ± 0.6)×10-8 M, *kcat* = (3.6± 1.1)×10-5 min-1) is 270-290-fold higher than to HIV RT (*Km* = (4.9 ± 0.5)×10-6 M, *kcat* = (2.2± 0.2)×10-3 min-1) and to -casein (*Km* = (5.3 ± 0.5)·10-6 M, *kcat* = (2.0± 0.2)×10-2 min-1) (Odintsova et al., 2006). At the same time, the relative rate of -casein hydrolysis was 9- and 550-fold higher than that for HIV RT and HSA, respectively. The *Km* for casein ((7.3±1.2)×10-6 M; *k*cat = 0.75±0.05 min-1) estimated in the reaction catalyzed by polyclonal sIgAs from human milk (Odintsova et al., 2005) is comparable with that for IgGs from AIDS patients, while the *kcat* is ~38-fold higher. This difference is most likely due to a higher content of anticasein proteolytic Abzs in human milk in comparison with blood of AIDS patients.

The affinity of AIDS pIgGs hydrolyzing HIV RT, human casein, and HSA (*Km* = 0.018 – 5.3 M; see above) as well as IN-Sepharose-purified AIDS pIgGs and pIgMs for IN (*Km* = 12.8- 156 M) in terms of *Km* values is comparable with typical affinities (*Km* = 0.038–7.3 M) (Paul et al., 1989; Kalaga et al., 1995; Legostaeva et al., 2010; Nevinsky et al., 2005, 2010a, 2010b) of Abzs hydrolyzing different proteins.

The *Km* (and *kcat*) values for plasmid scDNA in the reaction catalyzed by two individual nonfractionated AIDS IgG preparations were determined (Odintsova et al., 2006a). In the case of one of two preparations analyzed, the initial rate of DNA hydrolysis increases with increase in DNA concentration according to the Michaelis–Menten kinetics and only one pair of *K*<sup>m</sup> (53 ± 9 nM) and *kcat* ((2.1 ± 0.1)×10-2 min-1) was observed. For the second pIgG preparation two pairs of *K*<sup>m</sup> (2.6 ± 0.1 and 4.4 ± 0.7 nM) and *kcat* values ((6.7 ± 0.1)×10-2 and (29.6 ± 5.0)×10- 2 min-1) were revealed*.* Thus, the affinity the scDNA substrate for AIDS IgGs varied (in terms of *Km* values) in the range 2.6–53 nM, which correspond to typical *Kd* values for Ab– antigen interactions and is about 3–4 orders of magnitude higher than affinity of scDNA for DNase I (*KM* = 46–58 M) (Gololobov et al., 1995). These *Km* values for scDNA are comparable with the *Km* for plasmid DNA (43 nM) reported previously for IgG from SLE patients (Gololobov et al., 1995).

The catalysis mediated by artificial Abzs against reaction transition states is usually characterized by relatively low reaction rates: *kcat* values are 102–106-fold lower than for canonical enzymes (Keinan, 2005). The known *kcat* values for natural Abzs from AI patients

2008a, 2008b, 2008c, 2008d, 2009) as well as AI Abs with proteolytic and other activities (Nevinsky et al, 1998; Semenov et al., 2004; Legostaeva et al., 2010) on resins bearing immobilized specific substrates using elution of Abs with different concentration of NaCl and an acidic buffer always leads to separation of Abs into many Abz subfractions with different affinity to immobilized substrate. In addition, the affinity of separated fractions for immobilized substrate increased gradually with the increase in eluting NaCl concentration, but the *Km* (and *Vmax*) values corresponding to each fraction eluted from affinity sorbent are individual for every patient analyzed. It means, that the apparent number of monoclonal Abzs with different catalytic properties within the polyclonal Abs pool may be significantly underestimated since it is impossible to separate Abzs with comparable affinities for a specific substrate or to distinguish monoclonal Abzs with similar kinetic parameters. Some minor monoclonal Abzs with a relatively high activity and even major Abzs with low activity may be hidden by major Abzs with high activity. As a consequence, the *Km* (and *Vmax*) values determined using Abzs not fractionated on affinity sorbents more often characterize the interaction of substrates with a major fraction of Abzs with the maximal content and highest relative enzymatic activity at condition used. At the same time, these characteristic are very useful for comparisons of Abs with different substrate specificities and from patients with different diseases. For example, it was shown that affinity of specific major fractions of AIDS IgGs non-fractionated on affinity resins to HSA (*Km* = (1.8 ± 0.6)×10-8 M, *kcat* = (3.6± 1.1)×10-5 min-1) is 270-290-fold higher than to HIV RT (*Km* = (4.9 ± 0.5)×10-6 M, *kcat* = (2.2± 0.2)×10-3 min-1) and to -casein (*Km* = (5.3 ± 0.5)·10-6 M, *kcat* = (2.0± 0.2)×10-2 min-1) (Odintsova et al., 2006). At the same time, the relative rate of -casein hydrolysis was 9- and 550-fold higher than that for HIV RT and HSA, respectively. The *Km* for casein ((7.3±1.2)×10-6 M; *k*cat = 0.75±0.05 min-1) estimated in the reaction catalyzed by polyclonal sIgAs from human milk (Odintsova et al., 2005) is comparable with that for IgGs from AIDS patients, while the *kcat* is ~38-fold higher. This difference is most likely due to a higher content of anti-

casein proteolytic Abzs in human milk in comparison with blood of AIDS patients.

Abzs hydrolyzing different proteins.

patients (Gololobov et al., 1995).

The affinity of AIDS pIgGs hydrolyzing HIV RT, human casein, and HSA (*Km* = 0.018 – 5.3 M; see above) as well as IN-Sepharose-purified AIDS pIgGs and pIgMs for IN (*Km* = 12.8- 156 M) in terms of *Km* values is comparable with typical affinities (*Km* = 0.038–7.3 M) (Paul et al., 1989; Kalaga et al., 1995; Legostaeva et al., 2010; Nevinsky et al., 2005, 2010a, 2010b) of

The *Km* (and *kcat*) values for plasmid scDNA in the reaction catalyzed by two individual nonfractionated AIDS IgG preparations were determined (Odintsova et al., 2006a). In the case of one of two preparations analyzed, the initial rate of DNA hydrolysis increases with increase in DNA concentration according to the Michaelis–Menten kinetics and only one pair of *K*<sup>m</sup> (53 ± 9 nM) and *kcat* ((2.1 ± 0.1)×10-2 min-1) was observed. For the second pIgG preparation two pairs of *K*<sup>m</sup> (2.6 ± 0.1 and 4.4 ± 0.7 nM) and *kcat* values ((6.7 ± 0.1)×10-2 and (29.6 ± 5.0)×10- 2 min-1) were revealed*.* Thus, the affinity the scDNA substrate for AIDS IgGs varied (in terms of *Km* values) in the range 2.6–53 nM, which correspond to typical *Kd* values for Ab– antigen interactions and is about 3–4 orders of magnitude higher than affinity of scDNA for DNase I (*KM* = 46–58 M) (Gololobov et al., 1995). These *Km* values for scDNA are comparable with the *Km* for plasmid DNA (43 nM) reported previously for IgG from SLE

The catalysis mediated by artificial Abzs against reaction transition states is usually characterized by relatively low reaction rates: *kcat* values are 102–106-fold lower than for canonical enzymes (Keinan, 2005). The known *kcat* values for natural Abzs from AI patients vary in the range of 0.001–40 min–1 (Gololobov et al., 1995; Kalaga et al., 1995; Nevinsky et al., 2005, 2010a, 20010b). There are only several exceptions. For example, the specific activity of RNase IgGs from AI patients was about 1-20% of that for RNase A and of six known human sera RNases, while poly(A) was hydrolyzed by Abzs 2-10-fold faster than by RNase A, one of the most active RNases known (Buneva et al., 1994; Baranovskii et al., 1997, 1998; Vlasov et al., 1998). At the same time, the specific activity of homogeneous Abzs of several SLE and MS patients was about 40-400% of that for RNase A (Baranovskii et al., 1998). In addition, the specific nucleotide-hydrolyzing activities of mouse polyclonal IgGs is 1–4 orders of magnitude higher that those of known natural Abzs (Andryushkova et al., 2009). The *kcat* values for AIDS pIgGs hydrolyzing scDNA ((2.1-29.6)×10-2 min-1), human casein (2.0×10-2 min-1), HSA (3.6×10-5 min-1), and HIV RT (2.2×10-3 min-1), as well IgGs (0.3-2.9 min-1) and IgMs (2.0-6.4 min-1) purified on IN-Sepharose (Table 2) in the IN hydrolysis were comparable with those for known Abzs (Odintsova et al., 2006a, 2006b; Baranova et al., 2009, 2010).

Currently there are no methods that can efficiently separate Abzs from catalytically inactive Abs against the same antigen. In addition, as it was shown above, IN-Sepharose interact not only with anti-IN IgMs and IgGs but bind non-specifically some other Abs. Even partial purification of IgGs and IgMs on IN-Sepharose (or other specific affinity resins) leads to significant increase in the *kcat* value for IN and other substrates hydrolysis. Since the specific activities in all cases were calculated using the total concentration of purified pIgGs and pIgMs and affinity chromatography on IN-Sepharose (and other affinity sorbents) cannot separate catalytically active and inactive anti-IN Abs, the specific IN-hydrolyzing activities of the individual monoclonal subfractions in the pIgG and pIgM pools may be higher than those of non-fractionated or partially fractionated Abs. It should be mentioned that specific activities of some Abzs are often significantly lower than those for canonical enzymes with the same catalytic activities. However, it is impossible to dismiss the RAs of Abs from patients with AI and viral infection as biologically inessential since they are comparable with those for many canonical and very important enzymes, for example, restriction endonucleases and repair enzymes (Gololobov et al., 1995; Nevinsky et al., 2005, 2010, 2010). Thus, IN-hydrolyzing IgGs and IgMs from HIV-infected patients are very heterogeneous in their affinity to IN-Sepharose, demonstrate different *Km* and *Vmax* values and different subfractions of Abzs can hydrolyze various substrates at pH from 3 to 10. In addition, in contrast to other Abzs with proteolytic activity they can possess for different types of proteolytic activities: thiol-, metal-dependent, serine- and acidic-like.

#### **6. Peculiarities of protein hydrolysis by AIDS abzymes and canonical proteases**

#### **6.1 Casein hydrolysis by AIDS abzymes**

Casein hydrolyzing Abzs was found not only in the sera of HIV-infected patients (Odintsova et al., 2006b) but also in the milk of lactating women (Odintsova et al., 2005; 2011). At the first glance, no obvious immunizing factors could be found in clinically healthy pregnant and lactating women. However, pregnancy could "activate" or "trigger" autoimmune-like manifestations in clinically healthy women, and a sharp exacerbation of AI reactions can occur in some cases soon after childbirth (Amino *et al.*, 1982; Freeman *et al.*, 1986). Postnatal AI pathologies arise sometimes, including SLE, Hashimoto's thyroiditis, phospholipids syndrome, polymyositis, AI myocarditis, etc. (Amino *et al.*, 1982; Freeman *et al.*, 1986).

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 177

HIV-1 integrase catalyzes insertion of a DNA copy of the viral genome into the host genome (Asante-Appiah & Skalka, 1999)*.* Therefore IN, together with RT and protease, is the main

Specific interactions between HIV IN and long terminal repeats are required for insertion of viral DNA into the host genome. The use of a method based on stepwise increase in ligand complexity allowed an estimation of the relative contributions of each nucleotide from oligonucleotides (ODNs) to the total affinity for IN (Bugreev et al., 2003). It was shown that IN interacts with ODNs through superposition of weak contacts with their bases and, more importantly, with the internucleotide phosphate groups. Formation of the IN complex with specific DNA cannot by itself account for the major contribution of enzyme specificity, which lies in the *kcat* term; the rate of 3'-processin reaction is increased by more than 5 orders of magnitude upon transition from non-specific to specific oligonucleotides (Bugreev et al.,

In solution, HIV-1 IN exists in a dynamic equilibrium of monomers, dimers, tetramers and high-order oligomers (Deprez et al., 2000). We have recently analyzed the activity of different purified oligomeric forms of recombinant IN obtained after stabilization by platinum crosslinking and shown that these forms do not share the same *in vitro* catalytic properties (Faure et al., 2005). While monomers were inactive for all specific IN activities, dimers were able to catalyze the 3'-processing and the insertion of only one LTR into a short target DNA. In contrast, a tetramer of IN catalyzed the full-site integration of the two viral

To characterize the influence of the determinants of DNA substrate specificity on the oligomerization status of IN, the small-angle X-ray scattering technique was used (Baranova et al., 2007). Under special conditions in the absence of ODNs IN existed only as monomers. IN preincubation with specific ODNs led mainly to formation of dimers, the relative amount of which correlated well with the increase in the enzyme activity. Under these conditions, tetramers were scarce. Nonspecific ODNs stimulated formation of catalytically inactive dimers and tetramers. Complexes of monomeric, dimeric and tetrameric forms of IN with specific and nonspecific ODNs had varying radii of gyration (Rg), suggesting that the specific sequence-dependent formation of IN tetramers occurs by dimerization of two dimers of different structure. From the data it was concluded that the DNA-induced oligomerization of HIV-1 IN is of extreme importance to provide substrate specificity and to

It is known that a formation of multiple contacts between the same or various subunits of oligomeric enzymes is usually provided by multiple hydrophobic and electrostatic contacts and hydrogen bonds. A similar situation was observed for the dimeric forms of HIV-1 IN by X-ray crystallography (for review see Wlodawer 1999; Chiu & Davies, 2004 ). Analysis of effects of specific and nonspecific ODNs on the rate of IN proteolysis by chymotrypsin, trypsin, and proteinase K can provide useful information concerning a possible decrease in the accessibility of aromatic and positively charged amino acid residues after an enzyme binds its substrates, changes its conformation, or forms contacts between its subunits. It was interesting to compare the effect of different ODNs on the cleavage of IN by Abzs and

**6.2 Integrase hydrolysis by AIDS abzymes** 

important target of anti-HIV drugs.

LTR ends into a target DNA.

increase the enzyme activity (Baranova et al., 2007).

**6.2.2 Effect of DNA on the integrase hydrolysis by different proteases** 

2003).

**6.2.1 Specific regularities of integrase interaction with DNA** 

Fig. 11. SDS-PAGE analysis of products of [32P]casein hydrolysis by human milk sIgA and different proteases (A) or AIDS IgGs (B) (autoradiograph). A, Lanes 1, 2, and 3 correspond to [32P]-casein incubated with sIgAs for 20, 40, and 60 min, respectively; lanes 4 and 5, incubation with trypsin for 10 and 15 min; lanes 6 and 7, incubation with proteinase K for 10 and 15 min; lane 8, incubation with chymotrypsin for 10 min; 9, casein incubated alone for 60 min. The reaction mixture (10 l) for analysis of proteolytic activity of different proteases contained 6.2 g/ml casein and sIgAs or enzymes at the following concentrations: 0.32 g/ml trypsin, 0.064 g/ml chymotrypsin, 0.1 g/ml proteinase K, and 20 g/ml sIgA. B, [32P]casein was incubated for 14 h in the absence of Abs (lane 1) and in the presence IgGs from the sera of different AIDS patients: lane 2, IgG-1 (14 h), lanes 3 and 4, IgG2 (7 and 14 h, respectively), lane 5, IgG-3 (14 h).

Parenteral or oral administration of various proteins to animals late in pregnancy leads to the production of the corresponding Abs at high levels (Fey *et al.*, 1973; Mestecky *et al.*, 1987). Thus, pregnant women may be effectively immunized by contacts with compounds of viruses and bacteria that are not immunogenic in other healthy humans. There may be also some degree of autoimmunization during pregnancy similar to that in AI patients (Nevinsky et al., 2005, 2010a, 2010b and refs therein).

It was shown that lactation is associated with the appearance of sIgA and IgG abzymes with DNase, RNase (Kanyshkova *et al.*, 1997; Nevinsky *et al.*, 2000a, 2000b), ATPase (Semenov *et al.*, 2004), amylolytic (Savel'ev *et al.*, 2001), protein- (Nevinsky *et al.*, 1998), lipid- (Gorbunov *et al.*, 2005) and polysaccharide (Karataeva *et al.*, 2006a, 2006b) kinase activities in human milk. The specific activities of milk Abzs are significantly higher than those of Abzs from the blood of healthy lactating women and patients with different AI pathologies (Nevinsky *et al.,* 2003, 2005, 2010a, 2010b and refs therein).

We have compared the hydrolysis of -casein by canonical proteases, human milk sIgA (Fig. 11A) and three different individual AIDS IgGs (Fig. 11B) (Odintsova et al., 2006b; 2011). The patterns of -casein hydrolysis by milk sIgA, AIDS IgG, trypsin, chymotrypsin, and proteinase K were quite different. In addition, there was observed remarkable difference in the hydrolysis of -casein by three individual AIDS IgG-1, IgG-2, and IgG-3 (Fig 11B).

It should be mentioned that AIDS IgGs demonstrated only serine-like protease activity (Odintsova et al., 2006b), while milk IgAs additionally possess Me-dependent activity (Odintsova et al., 2011). Thus, possible ways of the production of Abzs with caseinhydrolyzing activity in healthy human mothers and in AIDS patients may be different.

## **6.2 Integrase hydrolysis by AIDS abzymes**

176 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

**94 67 43**

**B**

**1 2 3 4 5 kDa**

**67 43**

**30**

**20**

**14 2.**

**30**

**20**

**14 2.**

Fig. 11. SDS-PAGE analysis of products of [32P]casein hydrolysis by human milk sIgA and different proteases (A) or AIDS IgGs (B) (autoradiograph). A, Lanes 1, 2, and 3 correspond to [32P]-casein incubated with sIgAs for 20, 40, and 60 min, respectively; lanes 4 and 5, incubation with trypsin for 10 and 15 min; lanes 6 and 7, incubation with proteinase K for 10 and 15 min; lane 8, incubation with chymotrypsin for 10 min; 9, casein incubated alone for 60 min. The reaction mixture (10 l) for analysis of proteolytic activity of different proteases contained 6.2 g/ml casein and sIgAs or enzymes at the following concentrations: 0.32 g/ml trypsin, 0.064 g/ml chymotrypsin, 0.1 g/ml proteinase K, and 20 g/ml sIgA. B, [32P]casein was incubated for 14 h in the absence of Abs (lane 1) and in the presence IgGs from the sera of different AIDS patients: lane 2, IgG-1 (14 h), lanes 3 and 4, IgG2 (7 and 14 h,

Parenteral or oral administration of various proteins to animals late in pregnancy leads to the production of the corresponding Abs at high levels (Fey *et al.*, 1973; Mestecky *et al.*, 1987). Thus, pregnant women may be effectively immunized by contacts with compounds of viruses and bacteria that are not immunogenic in other healthy humans. There may be also some degree of autoimmunization during pregnancy similar to that in AI patients

It was shown that lactation is associated with the appearance of sIgA and IgG abzymes with DNase, RNase (Kanyshkova *et al.*, 1997; Nevinsky *et al.*, 2000a, 2000b), ATPase (Semenov *et al.*, 2004), amylolytic (Savel'ev *et al.*, 2001), protein- (Nevinsky *et al.*, 1998), lipid- (Gorbunov *et al.*, 2005) and polysaccharide (Karataeva *et al.*, 2006a, 2006b) kinase activities in human milk. The specific activities of milk Abzs are significantly higher than those of Abzs from the blood of healthy lactating women and patients with different AI pathologies (Nevinsky *et* 

We have compared the hydrolysis of -casein by canonical proteases, human milk sIgA (Fig. 11A) and three different individual AIDS IgGs (Fig. 11B) (Odintsova et al., 2006b; 2011). The patterns of -casein hydrolysis by milk sIgA, AIDS IgG, trypsin, chymotrypsin, and proteinase K were quite different. In addition, there was observed remarkable difference in the hydrolysis of -casein by three individual AIDS IgG-1, IgG-2, and IgG-3 (Fig 11B). It should be mentioned that AIDS IgGs demonstrated only serine-like protease activity (Odintsova et al., 2006b), while milk IgAs additionally possess Me-dependent activity (Odintsova et al., 2011). Thus, possible ways of the production of Abzs with caseinhydrolyzing activity in healthy human mothers and in AIDS patients may be different.

**1 2 3 4 5 6 7 8 9 kDa**

**A**

respectively), lane 5, IgG-3 (14 h).

(Nevinsky et al., 2005, 2010a, 2010b and refs therein).

*al.,* 2003, 2005, 2010a, 2010b and refs therein).

## **6.2.1 Specific regularities of integrase interaction with DNA**

HIV-1 integrase catalyzes insertion of a DNA copy of the viral genome into the host genome (Asante-Appiah & Skalka, 1999)*.* Therefore IN, together with RT and protease, is the main important target of anti-HIV drugs.

Specific interactions between HIV IN and long terminal repeats are required for insertion of viral DNA into the host genome. The use of a method based on stepwise increase in ligand complexity allowed an estimation of the relative contributions of each nucleotide from oligonucleotides (ODNs) to the total affinity for IN (Bugreev et al., 2003). It was shown that IN interacts with ODNs through superposition of weak contacts with their bases and, more importantly, with the internucleotide phosphate groups. Formation of the IN complex with specific DNA cannot by itself account for the major contribution of enzyme specificity, which lies in the *kcat* term; the rate of 3'-processin reaction is increased by more than 5 orders of magnitude upon transition from non-specific to specific oligonucleotides (Bugreev et al., 2003).

In solution, HIV-1 IN exists in a dynamic equilibrium of monomers, dimers, tetramers and high-order oligomers (Deprez et al., 2000). We have recently analyzed the activity of different purified oligomeric forms of recombinant IN obtained after stabilization by platinum crosslinking and shown that these forms do not share the same *in vitro* catalytic properties (Faure et al., 2005). While monomers were inactive for all specific IN activities, dimers were able to catalyze the 3'-processing and the insertion of only one LTR into a short target DNA. In contrast, a tetramer of IN catalyzed the full-site integration of the two viral LTR ends into a target DNA.

To characterize the influence of the determinants of DNA substrate specificity on the oligomerization status of IN, the small-angle X-ray scattering technique was used (Baranova et al., 2007). Under special conditions in the absence of ODNs IN existed only as monomers. IN preincubation with specific ODNs led mainly to formation of dimers, the relative amount of which correlated well with the increase in the enzyme activity. Under these conditions, tetramers were scarce. Nonspecific ODNs stimulated formation of catalytically inactive dimers and tetramers. Complexes of monomeric, dimeric and tetrameric forms of IN with specific and nonspecific ODNs had varying radii of gyration (Rg), suggesting that the specific sequence-dependent formation of IN tetramers occurs by dimerization of two dimers of different structure. From the data it was concluded that the DNA-induced oligomerization of HIV-1 IN is of extreme importance to provide substrate specificity and to increase the enzyme activity (Baranova et al., 2007).

#### **6.2.2 Effect of DNA on the integrase hydrolysis by different proteases**

It is known that a formation of multiple contacts between the same or various subunits of oligomeric enzymes is usually provided by multiple hydrophobic and electrostatic contacts and hydrogen bonds. A similar situation was observed for the dimeric forms of HIV-1 IN by X-ray crystallography (for review see Wlodawer 1999; Chiu & Davies, 2004 ). Analysis of effects of specific and nonspecific ODNs on the rate of IN proteolysis by chymotrypsin, trypsin, and proteinase K can provide useful information concerning a possible decrease in the accessibility of aromatic and positively charged amino acid residues after an enzyme binds its substrates, changes its conformation, or forms contacts between its subunits. It was interesting to compare the effect of different ODNs on the cleavage of IN by Abzs and

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 179

at the carboxyl side of lysine and arginine residues (Fig. 12). Thus, these ODNs most probably stimulate formation of dimeric forms of IN with more Lys- and Arg-dependent electrostatic contacts between the monomers. A weak protective effect of specific and nonspecific ODNs was observed in the case of proteinase K, which is mostly sequenceindependent. Thus, specific and nonspecific DNAs stimulate the formation of different IN oligomeric forms, in which aromatic and charged amino acid residues in different extent accessible for Abzs, chymotrypsin, and trypsin. The findings correlate with the results obtained by small-angle X-ray scattering, which show that all nonspecific and specific ODNs stimulate different changes in the structure of IN monomers and dimers free of DNA

MALDI-TOF analysis of the fragments formed after IN incubation with pIgG and pIgM purified on IN-Sepharose was carried out (Odintsova E., Baranova S., and Nevinsky G.A., personal communication). The cross-sections of longitudinal slices of the gel corresponding to the products with approximate mol. masses 29 ± 2 (P1), 22 ± 2 (P2), 16 ± 2 (P3), and 12 ± 2 (P4), as well as 30± 2 (P0) kDa were cut out as shown on Fig. 13 and the proteins were eluted from the gel. MALDI-TOF analysis has shown that P1–P4 fractions contain from four to eight major fragments with different molecular masses. The peptides found in the P1–P4 fractions were digested with trypsin under standard conditions for MALDI analysis, and the

> P1 P2 P3 P4

Fig. 13. SDS-PAGE analysis of IN hydrolysis by pIgG-1 (lane 3, A), IgG-6 (lane 2, B), pIgM-2 (lane 4, A) and pIgM-6 (lane 3, B) purified on IN-Sepharose after 9 h (A) and 2 h (B) of incubation in a nonreducing 12% gel followed by silver staining. Lanes 1 (B) and 2 (A), IN incubated in the absence of Abs. Lane 1 (A) protein molecular mass markers. Gel zones P1, P2, P3, and P4 (A) as well as P0 (B) were used for MALDI-TOF analysis (see the text).

Seven antigenic determinants (AGDs) have been reported for HIV IN corresponding to amino acid residues 5–22 (AGD1), 14–35 (AGD2) (Yi et al., 2000), 58–141 (AGD3), 141–172 (AGD4), 248–264 (AGD5) (Bizub-Bender et al., 1994), 208–228 (AGD6), and 251–271(AGD7) (Nilsen et al., 1996) (underlined in Fig. 14). Interestingly, 6–7 cleavage sites found by MALDI corresponded to the N-terminal stretch of residues 11–35, belonging to two overlapping antigenic determinants AGD1 and AGD2 (Fig. 14). Three clusters of cleavage sites were located within the long AGD3. A block of 12 closely spaced cleavage sites corresponded to the N-terminal part of AGD4. Only one cleavage site was located within AGD5 and four sites corresponded to AGD6 and AGD7. At the same time, some sites of IN cleavage, most notably a cluster of 16 sites between residues 175 and 202, did not corresponds to any IN

P0

**1 2 3 4** 

**A**

P0

**1 2 3 B**

P1

P0

P2 P3 P4

(Baranova et al., 2007).

hydrolyzates were studied.

AGD known at this moment.

**14.4 18.4 25.0**

**35.0 kDa**

canonical proteases. The specific single-stranded (ss) 5'-GTGTGGAAAATCTCTAGCA (19- CA), ss 5'-GTGTGGAAAATCTCTAGCAGT (21-GT), ss 5'-ACTGCTAGAGATTTTCCACAC (21-COM, complementary to 21-GT and to 19-CA), double-stranded (ds) 21-GT (21-GT•21- COM) and ds 19-CA (19-CA•21-COM) corresponding to terminal repeats of viral DNA were used.

While nonspecific d(pT)n markedly decreased the IgG-dependent hydrolysis of IN, d(pA)<sup>n</sup> and d(pA)n•d(pT)n demonstrated no detectable protective effect (Fig. 12) (Odintsova E., Baranova S., and Nevinsky G.A., personal communication).

Fig. 12. SDS-PAGE analysis of HIV IN hydrolysis by IgGs and canonical proteases. Polyclonal IgGs from HIV-infected patients (A), chymotrypsin (B), trypsin (C), and proteinase K (D) were used for hydrolysis of IN after its preincubation in the absence and in the presence of various ODNs leading to the formation of different IN oligomeric forms. IN was pre-incubated for 1 h at 30°C in the absence (lane 7) or in the presence of saturating concentrations (0.2–1 mM) of ss d(pA)24 (lane 1), ss d(pT)24 (lane 2), ds d(pA)24•d(pT)<sup>24</sup> (lane 3), ss 21-COM (lane 4), ss 21-GT (lane 5), ds 21-GT•21-COM (lane 6). Then, preincubated mixture was diluted 3-fold and one of three canonical proteases or pIgGs from HIV-infected patients was added. After 5–10 min incubation in the presence of trypsin (19 M), chymotrypsin (1.6 M), proteinase K (19 M) and 6 h in the presence of pIgGs (0.17 M) the reaction was stopped and the efficiency of IN hydrolysis was analysed by SDS-PAGE.

The best protection from the hydrolysis by IgGs was observed for ss and especially ds specific ODNs (Fig. 12). Overall, the protective effects of all specific and nonspecific ss and ds ODNs from hydrolysis of IN by chymotrypsin were comparable. Therefore, one can suggest that the formation of IN complex with specific and nonspecific ODNs led to a similar decrease in the accessibility of aromatic amino acid residues as a result of their shielding by ODNs and/or involvement of these residues to the formation of multiple contacts at the interfaces of IN oligomer subunits. In contrast to chymotrypsin, nonspecific ODNs strongly protects IN from hydrolysis by trypsin, which cleaves peptide chains mainly

canonical proteases. The specific single-stranded (ss) 5'-GTGTGGAAAATCTCTAGCA (19- CA), ss 5'-GTGTGGAAAATCTCTAGCAGT (21-GT), ss 5'-ACTGCTAGAGATTTTCCACAC (21-COM, complementary to 21-GT and to 19-CA), double-stranded (ds) 21-GT (21-GT•21- COM) and ds 19-CA (19-CA•21-COM) corresponding to terminal repeats of viral DNA

While nonspecific d(pT)n markedly decreased the IgG-dependent hydrolysis of IN, d(pA)<sup>n</sup> and d(pA)n•d(pT)n demonstrated no detectable protective effect (Fig. 12) (Odintsova E.,

**kDa**

**14.4 18.4 25.0 35.0**

**kDa**

**14.4 18.4 25.0 35.0**

**1 2 3 4 5 6 7 C M 1 2 3 4 5 6 7 C M**

 **1 2 3 4 5 6 7 C M 1 2 3 4 5 6 7 C M** 

Fig. 12. SDS-PAGE analysis of HIV IN hydrolysis by IgGs and canonical proteases. Polyclonal IgGs from HIV-infected patients (A), chymotrypsin (B), trypsin (C), and

proteinase K (D) were used for hydrolysis of IN after its preincubation in the absence and in the presence of various ODNs leading to the formation of different IN oligomeric forms. IN was pre-incubated for 1 h at 30°C in the absence (lane 7) or in the presence of saturating concentrations (0.2–1 mM) of ss d(pA)24 (lane 1), ss d(pT)24 (lane 2), ds d(pA)24•d(pT)<sup>24</sup> (lane 3), ss 21-COM (lane 4), ss 21-GT (lane 5), ds 21-GT•21-COM (lane 6). Then, preincubated mixture was diluted 3-fold and one of three canonical proteases or pIgGs from HIV-infected patients was added. After 5–10 min incubation in the presence of trypsin (19 M), chymotrypsin (1.6 M), proteinase K (19 M) and 6 h in the presence of pIgGs (0.17 M) the reaction was stopped and the efficiency of IN hydrolysis was analysed by

The best protection from the hydrolysis by IgGs was observed for ss and especially ds specific ODNs (Fig. 12). Overall, the protective effects of all specific and nonspecific ss and ds ODNs from hydrolysis of IN by chymotrypsin were comparable. Therefore, one can suggest that the formation of IN complex with specific and nonspecific ODNs led to a similar decrease in the accessibility of aromatic amino acid residues as a result of their shielding by ODNs and/or involvement of these residues to the formation of multiple contacts at the interfaces of IN oligomer subunits. In contrast to chymotrypsin, nonspecific ODNs strongly protects IN from hydrolysis by trypsin, which cleaves peptide chains mainly

Baranova S., and Nevinsky G.A., personal communication).

**A B**

**C D**

were used.

SDS-PAGE.

at the carboxyl side of lysine and arginine residues (Fig. 12). Thus, these ODNs most probably stimulate formation of dimeric forms of IN with more Lys- and Arg-dependent electrostatic contacts between the monomers. A weak protective effect of specific and nonspecific ODNs was observed in the case of proteinase K, which is mostly sequenceindependent. Thus, specific and nonspecific DNAs stimulate the formation of different IN oligomeric forms, in which aromatic and charged amino acid residues in different extent accessible for Abzs, chymotrypsin, and trypsin. The findings correlate with the results obtained by small-angle X-ray scattering, which show that all nonspecific and specific ODNs stimulate different changes in the structure of IN monomers and dimers free of DNA (Baranova et al., 2007).

MALDI-TOF analysis of the fragments formed after IN incubation with pIgG and pIgM purified on IN-Sepharose was carried out (Odintsova E., Baranova S., and Nevinsky G.A., personal communication). The cross-sections of longitudinal slices of the gel corresponding to the products with approximate mol. masses 29 ± 2 (P1), 22 ± 2 (P2), 16 ± 2 (P3), and 12 ± 2 (P4), as well as 30± 2 (P0) kDa were cut out as shown on Fig. 13 and the proteins were eluted from the gel. MALDI-TOF analysis has shown that P1–P4 fractions contain from four to eight major fragments with different molecular masses. The peptides found in the P1–P4 fractions were digested with trypsin under standard conditions for MALDI analysis, and the hydrolyzates were studied.

Fig. 13. SDS-PAGE analysis of IN hydrolysis by pIgG-1 (lane 3, A), IgG-6 (lane 2, B), pIgM-2 (lane 4, A) and pIgM-6 (lane 3, B) purified on IN-Sepharose after 9 h (A) and 2 h (B) of incubation in a nonreducing 12% gel followed by silver staining. Lanes 1 (B) and 2 (A), IN incubated in the absence of Abs. Lane 1 (A) protein molecular mass markers. Gel zones P1, P2, P3, and P4 (A) as well as P0 (B) were used for MALDI-TOF analysis (see the text).

Seven antigenic determinants (AGDs) have been reported for HIV IN corresponding to amino acid residues 5–22 (AGD1), 14–35 (AGD2) (Yi et al., 2000), 58–141 (AGD3), 141–172 (AGD4), 248–264 (AGD5) (Bizub-Bender et al., 1994), 208–228 (AGD6), and 251–271(AGD7) (Nilsen et al., 1996) (underlined in Fig. 14). Interestingly, 6–7 cleavage sites found by MALDI corresponded to the N-terminal stretch of residues 11–35, belonging to two overlapping antigenic determinants AGD1 and AGD2 (Fig. 14). Three clusters of cleavage sites were located within the long AGD3. A block of 12 closely spaced cleavage sites corresponded to the N-terminal part of AGD4. Only one cleavage site was located within AGD5 and four sites corresponded to AGD6 and AGD7. At the same time, some sites of IN cleavage, most notably a cluster of 16 sites between residues 175 and 202, did not corresponds to any IN AGD known at this moment.

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 181

substrates with both light and heavy chains of Abzs. The separation of the light chains can lead to a decrease in the lifetime of the existence of the complex and, as a consequence, to an increase in the turnover number and *Vmax* (*kca*t) of the reaction catalyzed by L-chains*.* Taken together, the absence or very weak interaction of short substrates with heavy chains of AIDS Abzs in contrast with globular molecules proteins (and higher rate of the reaction) may be a main reason of a decrease of specificity of Abzs action in the case of short oligopeptides; one cannot exclude that light (or heavy) chains of some Abzs can effectively hydrolyzed short

We have shown that *in vitro* IgGs and IgMs hydrolyzing IN significantly decrease the 3' processing and integration reaction catalyzed by IN (for example, Fig. 15) (Odintsova E.,

**[IgG], M**

**01234567**

**2.8 M**

**0.05 0.10 0.15 0.20**

**[IgM], M**

According to modern point of view, the immune response to the viral components is the most important factor providing slow transition of HIV infection to the stage of AIDS (Fauci et al., 2008). Since AIDS anti-IN anti-RT Abs can efficiently hydrolyze IN and RT (Odintsova et al., 2006b; Baranova et al., 2009, 2010), a positive role of abzymes in counteracting the infection cannot be excluded and these polyclonal and corresponding monoclonal Abzs with proteolytic activities are potentially interesting for designing new anti-HIV agents. In addition, detection of IN-hydrolyzing activity can be useful for diagnostic purposes and for

The field of monoclonal Abzs with immunotherapeutic potential has recently been reviewed (see "Introduction"). Abzs that cleave HIV envelope gp120 protein may find their use in the treatment of AIDS (Tellier, 2002; Stockwin & Holmes, 2003). pIgG degrading gp120 was also obtained taking advantage of the susceptibility of SJL mice to a peptide-induced AI disorder, experimental AI encephalomyelitis (Ponomarenko et al., 2006). Immunization of specific pathogen-free SJL mice with structural fragments of gp120 fused in-frame with the encephalitogenic MBP(85-101) peptide resulted in a pronounced disease-associated immune response against these antigens. This strategy can be generalized to create catalytic vaccines against viral pathogens (Ponomarenko et al., 2006). In addition, Abzs with different catalytic

Fig. 15. Effect of IN-hydrolyzing IgGs and IgMs from AIDS patients on the rate of the 3'-

oligopeptides of any sequences.

processing reaction.

**7. Conclusion** 

Baranova S., and Nevinsky G.A., personal communication).

**Relative activity, %**

assessment of the immune status in AIDS patients.

activities can be used for different purposes (see "Introduction").

**0.05 M**

Fig. 14. All trypsin-independent points of IN cleavage determined from the MALDI-TOF analysis directly mol. masses of the P0–P4 fractions and after their cleavage by trypsin. Numbers next to the arrows show the position of the residue within the IN sequence; numbers under short arrows and in parentheses correspond to the frequency of these cleavage sites found using different approaches. Seven known antigenic determinants of IN are underlined in the figure.

Interestingly, a similar situation was observed for MS IgGs specifically hydrolyzing MBP; in addition to the sites of cleavage within four known AGDs of MBP, several sites were outside these determinants (Ponomarenko et al., 2006). Thus, the number of structurally different antigenic determinants in the case of IN may be great.

Interestingly, in contrast to the absence of hydrolysis of non-specific globular proteins by anti-IN Abs, they first cleave IN with the accumulation of long fragments corresponding mainly to known AGDs and then are capable of further degradation of these long intermediates, and the formation of very short products was observed after 72–100 h of IN incubation In with Abzs (Odintsova E., Baranova S., and Nevinsky G.A., personal communication). AIDS anti-IN pIgGs and IgMs hydrolyze specific 20-25-mer oligopeptides corresponding to the IN AGDs ~30–70-fold faster than nonspecific long 20-25-mer oligopeptides corresponding to AGDs of human myelin basic protein and HIV RT. In addition, AIDS anti-IN Abzs can hydrolyze very short 3-4-mer nonspecific oligopeptides 100–300-fold more slowly than specific ones. Therefore, the recognition and digestion of globular proteins and relatively short oligopeptides by the Abzs proceeds in different ways. Since catalytic centers of Abzs specifically hydrolyzing different proteins including IN are usually located on the light chains of Abs (Nevinsky et al., 2005, 2010a, 2010b, and refs therein), the observed hydrolysis of short oligopeptides can be a consequence of their interaction with light or heavy chains without significant contacts with alternative chains.

Interestingly, separated light chains of pIgGs, pIgMs, and pIgAs from the sera of patients with different AI and viral diseases usually significantly more active than intact Abs in the hydrolysis of DNA, RNA, oligosaccharides, and proteins (Nevinsky et al., 2005, 2010a, 2010b, and refs therein). This phenomenon may be a consequence of a higher affinity of intact Abs, as compared with separated light chains, for different substrates due to interaction of the substrates with both light and heavy chains of Abzs. The separation of the light chains can lead to a decrease in the lifetime of the existence of the complex and, as a consequence, to an increase in the turnover number and *Vmax* (*kca*t) of the reaction catalyzed by L-chains*.* Taken together, the absence or very weak interaction of short substrates with heavy chains of AIDS Abzs in contrast with globular molecules proteins (and higher rate of the reaction) may be a main reason of a decrease of specificity of Abzs action in the case of short oligopeptides; one cannot exclude that light (or heavy) chains of some Abzs can effectively hydrolyzed short oligopeptides of any sequences.

We have shown that *in vitro* IgGs and IgMs hydrolyzing IN significantly decrease the 3' processing and integration reaction catalyzed by IN (for example, Fig. 15) (Odintsova E., Baranova S., and Nevinsky G.A., personal communication).

Fig. 15. Effect of IN-hydrolyzing IgGs and IgMs from AIDS patients on the rate of the 3' processing reaction.

## **7. Conclusion**

180 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

MALNGIDKA E DHEKYHSN R AMVSDFNL PVVA K EIVA CDKCQL KGE MHGQVDC -57 **Q W PS A**

**90-91 (1) 92-93 (2)**

**E V YPS**

**145-156 (1)**

**157-158 (1)**

<sup>2</sup> <sup>1</sup> <sup>2</sup> <sup>1</sup> <sup>2</sup> <sup>4</sup> <sup>1</sup> <sup>1</sup> <sup>4</sup> <sup>4</sup> <sup>1</sup>


**179-180 (1)**

**A F G KA A**


**158-159 (1)**

<sup>4</sup> <sup>4</sup> <sup>5</sup> <sup>1</sup> <sup>5</sup> <sup>1</sup> <sup>4</sup> <sup>3</sup> <sup>3</sup> <sup>2</sup> <sup>3</sup> <sup>2</sup> <sup>13</sup> <sup>1</sup> <sup>3</sup> <sup>1</sup>

KGE **PS**

**159-160 (1)**


**113-114 (1)**

**G RQ K D A**

**AGD5**


SP**G**IWQLDCTHL GKIILVAVH ASGYIE AEV I PA ETGQETA FLLKLAGRW VKT IHTDNG

**147-148 (1)**

**86-87 (1) 89-90 (1)**

**AGD1 AGD2**

<sup>2</sup> <sup>2</sup> <sup>2</sup> <sup>1</sup> <sup>3</sup> 3 <sup>1</sup>

**191-192 (1)**

**247-248 (2) 250-251 (2) 259-260 (2)**

<sup>3</sup> <sup>1</sup> <sup>1</sup> <sup>3</sup> <sup>3</sup> <sup>1</sup> <sup>1</sup> <sup>2</sup> <sup>2</sup>

**AGD6 AGD7**

**AGD3**

**120-121 (1) 127-128 (1)**

**184-185 (1) 189-190 (1)**

<sup>2</sup> <sup>2</sup>

are underlined in the figure.

**11-12 (1) 21-22 (4) 34-35 (1) 35-36 (2) 46-47 (1)**

NFTSTTV KA CWWAGIKQE GIPYNPQ S Q VVESM NN E L KIIGQVRDQ EHLKTAVQM

**148-149 (1)**

Fig. 14. All trypsin-independent points of IN cleavage determined from the MALDI-TOF analysis directly mol. masses of the P0–P4 fractions and after their cleavage by trypsin. Numbers next to the arrows show the position of the residue within the IN sequence; numbers under short arrows and in parentheses correspond to the frequency of these cleavage sites found using different approaches. Seven known antigenic determinants of IN

Interestingly, a similar situation was observed for MS IgGs specifically hydrolyzing MBP; in addition to the sites of cleavage within four known AGDs of MBP, several sites were outside these determinants (Ponomarenko et al., 2006). Thus, the number of structurally different

Interestingly, in contrast to the absence of hydrolysis of non-specific globular proteins by anti-IN Abs, they first cleave IN with the accumulation of long fragments corresponding mainly to known AGDs and then are capable of further degradation of these long intermediates, and the formation of very short products was observed after 72–100 h of IN incubation In with Abzs (Odintsova E., Baranova S., and Nevinsky G.A., personal communication). AIDS anti-IN pIgGs and IgMs hydrolyze specific 20-25-mer oligopeptides corresponding to the IN AGDs ~30–70-fold faster than nonspecific long 20-25-mer oligopeptides corresponding to AGDs of human myelin basic protein and HIV RT. In addition, AIDS anti-IN Abzs can hydrolyze very short 3-4-mer nonspecific oligopeptides 100–300-fold more slowly than specific ones. Therefore, the recognition and digestion of globular proteins and relatively short oligopeptides by the Abzs proceeds in different ways. Since catalytic centers of Abzs specifically hydrolyzing different proteins including IN are usually located on the light chains of Abs (Nevinsky et al., 2005, 2010a, 2010b, and refs therein), the observed hydrolysis of short oligopeptides can be a consequence of their interaction with light or heavy chains without significant contacts with alternative chains. Interestingly, separated light chains of pIgGs, pIgMs, and pIgAs from the sera of patients with different AI and viral diseases usually significantly more active than intact Abs in the hydrolysis of DNA, RNA, oligosaccharides, and proteins (Nevinsky et al., 2005, 2010a, 2010b, and refs therein). This phenomenon may be a consequence of a higher affinity of intact Abs, as compared with separated light chains, for different substrates due to interaction of the

VFIH NFKRK G IGGYSAGE IVDIIATDI T KELQKQIT IQNFRVYYR NKDPLWKGP

**211-212 (1)**

KLLWKGE GA VIQDNSDIK VPRRKAKII DYGKQMAGD CVASRQDED

**AGD4**

2 1

**V VR D**

**139-140 (2)**

1

antigenic determinants in the case of IN may be great.

According to modern point of view, the immune response to the viral components is the most important factor providing slow transition of HIV infection to the stage of AIDS (Fauci et al., 2008). Since AIDS anti-IN anti-RT Abs can efficiently hydrolyze IN and RT (Odintsova et al., 2006b; Baranova et al., 2009, 2010), a positive role of abzymes in counteracting the infection cannot be excluded and these polyclonal and corresponding monoclonal Abzs with proteolytic activities are potentially interesting for designing new anti-HIV agents. In addition, detection of IN-hydrolyzing activity can be useful for diagnostic purposes and for assessment of the immune status in AIDS patients.

The field of monoclonal Abzs with immunotherapeutic potential has recently been reviewed (see "Introduction"). Abzs that cleave HIV envelope gp120 protein may find their use in the treatment of AIDS (Tellier, 2002; Stockwin & Holmes, 2003). pIgG degrading gp120 was also obtained taking advantage of the susceptibility of SJL mice to a peptide-induced AI disorder, experimental AI encephalomyelitis (Ponomarenko et al., 2006). Immunization of specific pathogen-free SJL mice with structural fragments of gp120 fused in-frame with the encephalitogenic MBP(85-101) peptide resulted in a pronounced disease-associated immune response against these antigens. This strategy can be generalized to create catalytic vaccines against viral pathogens (Ponomarenko et al., 2006). In addition, Abzs with different catalytic activities can be used for different purposes (see "Introduction").

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 183

Amino, N.; Mori, H.; Iwatani, Y.; Tanizawa, O.; Kawashima, M.; Tsuge, I.; Ibaragi K. &

Asante-Appiah, E. & Skalka, A. M. (1999). HIV-1 integrase: structural organization,

Andrievskaya, O. A.; Buneva, V. N.; Naumov, V. A. & Nevinsky, G. A. (2000). Catalytic

Andrievskaya, O. A.; Buneva, V. N.; Baranovskii, A. G.; Gal'vita, A. V.; Benzo, E. S.;

Andryushkova, A. A.; Kuznetsova, I. A.; Orlovskaya, I. A.; Buneva, V. N. & Nevinsky, G. A.

Andryushkova, A. S.; Kuznetsova, I. A.; Buneva, V. N.; Toporkova, L. B.; Sakhno, L. V.;

Andryushkova, A. S.; Kuznetsova, I. A.; Orlovskaya, I. A.; Buneva, V. N. & Nevinsky, G. A.

Baranova S.; Tuzikov F. V.; Zakharova O. D.; Tuzikova N. A.; Calmels C.; Litvak S.; Tarrago-

Baranova, S. V.; Buneva, V. N.; Kharitonova, M. A.; Sizyakina, L. P.; Calmels, C.; Andreola,

Baranovskii, A. G.; Matushin, V. G.; Vlassov, A. V.; Zabara, V. G.; Naumov, V. A.; Buneva,

Baranovskii, A. G.; Kanyshkova, T. G.; Mogelnitskii, A. S.; Naumov, V. A.; Buneva, V. N.;

erythematosus. *Med. Sci. Monit*., 6(3), 460 –470, ISSN 1234-1010

erythematosus. *Immunol. Lett.,* 81(3), 191-198, ISSN 0165-2478

*Mol. Med*., 11(3), 531-551, ISSN 1582-1838

MRL/MpJ-lpr mice. *FEBS Lett.,* 580(21), 5089-5095, ISSN 0014-5793

MRL-lpr/lpr mice. *Int. immunol.,* 21(8), 935-945, ISSN 0953-8178

1 integrase. *Nucleic Acids Res.,* 35(3), 975-987, ISSN 0305-1048

Kumahara, Y.; Miyai, K. (1982). High prevalence of transient post-partum thyrotoxicosis and hypothyroidism. *N. Engl. J. Med.,* 306(14), 849-852, ISSN 0028-

conformational changes, and catalysis. *Adv. Virus Res.,* 52, 351-369, ISSN 0065-3527

heterogeneity of polyclonal RNA-hydrolyzing IgM from sera of patients with lupus

Naumov, V. A. & Nevinsky, G. A. (2002). Catalytic diversity of polyclonal RNAhydrolyzing IgG antibodies from the sera of patients with systemic lupus

(2006). Antibodies with amylase activity from the sera of autoimmune-prone

Tichonova, M. A.; Chernykh, E. R.; Orlovskaya, I. A. & Nevinsky, G. A. (2007) Formation of different abzymes in autoimmune-prone MRL-lpr/lpr mice is associated with changes in colony formation of haematopoetic progenitors. *J. Cell.* 

(2009) Nucleotide-hydrolyzing antibodies from the sera of autoimmune-prone

Litvak L.; Parissi V. & Nevinsky G. A. (2007). Small-angle X-ray characterization of the nucleoprotein complexes resulting from DNA-induced oligomerization of HIV-

M. L.; Parissi, V. & Nevinsky, G. A. (2009). HIV-1 integrase-hydrolyzing antibodies from sera of HIV-infected patients. *Biochimie,* 91(9), 1081-1086, ISSN 0300-9084 Baranova, S. V.; Buneva, V. N.; Kharitonova, M. A.; Sizyakina, L. P.; Calmels, C.; Andreola,

M. L.; Parissi, V. & Nevinsky, G. A. (2010). HIV-1 integrase-hydrolyzing IgM antibodies from sera of HIV-infected patients. *Int. Immunol.,* 22(8), 671-680, ISSN

V. N. & Nevinskii, G. A. (1997). DNA- and RNA-hydrolyzing antibodies from the blood of patients with various forms of viral hepatitis. *Biochemistry (Moscow),*

Gusev, E. I.; Boiko, A. N.; Zargarova, T. A.; Favorova, O. O. & Nevinsky, G. A. (1998). Polyclonal antibodies from blood and cerebrospinal fluid of patients with multiple sclerosis effectively hydrolyze DNA and RNA. *Biochemistry (Moscow),* 

**10. References** 

4793

0953-8178

62(12), ISSN 0006-2979

63(11), 1239-1248, ISSN 0006-2979

In conclusion, a number of studies of Abzs show the extremely wide potential of the immune system in producing Abzs possessing very different enzymatic activities, which very often are not comparable with those of known enzymes, and natural Abs with specified and novel functions may have wide potential for biotechnology and medicine.

## **8. Acknowledgments**

This research was made possible in part by grants from Analytical Departmental Special program "Development of Scientific Potential of High School (No. 2.1.1/5580)" from the Presidium of the Russian Academy of Sciences ("Molecular and Cellular Biology Program" No. 22.7, "Fundamental Sciences to Medicine" No. 21.16), Russian Foundation for Basic Research (08-04-00315, 09-04-00804), and funds from the Siberian Division of the Russian Academy of Sciences.

## **9. Abbreviations**

Abs - antibodies Abzs - abzymes - or catalytically active antibodies AG - antigen AI - autoimmune AD - autoimmune disease AIDS - human autoimmune deficit syndrome BSA - bovine serum albumin CC - correlation coefficient HSCs - hematopoietic stem cells sc - supercoiled ss and ds - single- and double-stranded - respectively CBA - (CBAxC57BL)F1 mice HT - Hashimoto's thyroiditis hMBP - human myelin basic protein MFT - microsomal fraction of thyrocytes MBP - myelin basic protein MS - multiple sclerosis nat-DNA and den-DNA - native and denatured DNA - respectively MHO - maltoheptaose pAbs and pIgGs - polyclonal Abs and IgGs - respectively RF - rheumatoid factor SLE- systemic lupus erythematosus - SDS-PAGE - SDS-polyacrylamide gel electrophoresis TBE - tick-borne encephalitis VIP - vasoactive intestinal peptide RA - relative activity CFU-GM - granulocytic-macrophagic colony-forming unit BFU-E - erythroid burst-forming unit CFU-GEMM - granulocytic-erythroid-megakaryocytic- macrophagic colony-forming unit

#### **10. References**

182 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

In conclusion, a number of studies of Abzs show the extremely wide potential of the immune system in producing Abzs possessing very different enzymatic activities, which very often are not comparable with those of known enzymes, and natural Abs with specified

This research was made possible in part by grants from Analytical Departmental Special program "Development of Scientific Potential of High School (No. 2.1.1/5580)" from the Presidium of the Russian Academy of Sciences ("Molecular and Cellular Biology Program" No. 22.7, "Fundamental Sciences to Medicine" No. 21.16), Russian Foundation for Basic Research (08-04-00315, 09-04-00804), and funds from the Siberian Division of the Russian

and novel functions may have wide potential for biotechnology and medicine.

**8. Acknowledgments** 

Academy of Sciences.

**9. Abbreviations**  Abs - antibodies

AD - autoimmune disease

BSA - bovine serum albumin CC - correlation coefficient HSCs - hematopoietic stem cells

CBA - (CBAxC57BL)F1 mice HT - Hashimoto's thyroiditis

MBP - myelin basic protein MS - multiple sclerosis

MHO - maltoheptaose

RF - rheumatoid factor

RA - relative activity

TBE - tick-borne encephalitis VIP - vasoactive intestinal peptide

BFU-E - erythroid burst-forming unit

hMBP - human myelin basic protein MFT - microsomal fraction of thyrocytes

AG - antigen AI - autoimmune

sc - supercoiled

Abzs - abzymes - or catalytically active antibodies

AIDS - human autoimmune deficit syndrome

ss and ds - single- and double-stranded - respectively

nat-DNA and den-DNA - native and denatured DNA - respectively

SLE- systemic lupus erythematosus - SDS-PAGE - SDS-polyacrylamide gel electrophoresis

CFU-GEMM - granulocytic-erythroid-megakaryocytic- macrophagic colony-forming unit

pAbs and pIgGs - polyclonal Abs and IgGs - respectively

CFU-GM - granulocytic-macrophagic colony-forming unit


Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 185

Faure, A.; Calmels, C.; Desjobert, C.; Castroviejo, M.; Caumont-Sarcos, A.; Tarrago-Litvak,

Founel, S. & Muller, S. (2002) Antinucleosome antibodies and T-cell response in systemic

Freeman, R.; Rosen, H. & Thysen, B. (1986). Incidence of thyroid dysfunction in an

Galvita, A. V.; Baranovskii, A. G.; Kuznetsova, I. A.; Vinshu, N. V.; Galenok, V. A.; Buneva,

Gao, Q. S.; Sun, M.; Tyutyulkova, S.; Webster, D.; Rees, A.; Tramontano, A.; Massey, R. J. &

Garmashova, N. V.; Kazanskii, V. E.; Tyshkevich, O. B.; Doronin, B. M.; Buneva, V. N. &

Goldfarb, M. F. (2001). Analysis of casein using two-dimensional electrophoresis, western blot, and computer imaging. *Adv. Exp. Med. Biol*., 501, 535-539, ISSN 0065-2598 Golinelli-Pimpaneau, B.; Gigant, B.; Bizebard, T.; Nazava, J.; Saludjian, P.; Zemel, R.; Tawfik,

Gololobov, G. V.; Mikhalap, S. V.; Starov, A. V.; Kolesnikov, A. F. & Gabibov, A. G. (1994).

Gololobov, G. V.; Chernova, E. A.; Schourov, D. V.; Smirnov, I. V.; Kudelina, I. A. &

Gottlieb, A. A. & Shwartz, R. H. (1972) Antigen-RNA interactions. *Cell Immunol*., 5(2), 341-

Grusby, M. J.; Mitchell, S. C.; Nabavi, N.; & Glimcher, L. H. (1990) Casein expression in

Fauci, A. S., Braunwald, E., Kasper, D. L., Hauser, S. L., Longo, D. L., Jameson, J. L. &

Fey, H.; Butler, R. & Marti, F. (1973). The production in the pregnant cow of anti-human

Hanson, C. V.; Nishiyama, Y. & Paul, S. (2005). Catalytic antibodies and their applications.

Hentati, B.; Sato, M. N.; Payelle, B.; Avrameas, S. & Ternynck, T. (1994). Beneficial effect of

Hill Professional, ISBN 0071466339 / 9780071466332, USA

*Curr. Opin. Biotechnol.,* 16(6), 631-636, ISSN 0958-1669

vitro. *Nucleic Acids Res*., 33(3), 977-986, ISSN 0305-1048

from patients with diabetes. *Russ. J. Immunol.,* 1, 116-131.

encephalitis. *Mol. Biol. (Moscow),* 38(4), 723-730, ISSN 0026-8933

*Biochem. Biotechnol.,* 47(2-3), 305-314, ISSN 0273-2289

*USA,* 92(1), 254-257, ISSN 0027-8424

*Chem.,* 269(51), 32389-32393, ISSN 0021-9258

9926

ISSN 0960-9822

362, ISSN 0008-8749

ISSN 0042-9007

8424

lupus erythematosus. *Ann. Med. Interne (Paris),* 153(8), 513-519.

L.; Litvak, S. & Parissi, V. (2005) HIV-1 integrase crosslinked oligomers are active in

unselected postpartum population. *Arch. Intern. Med.,* 146(7), 1361-1364, ISSN 0003-

V. N. & Nevinsky, G. A. (2007). A peculiarity of DNA hydrolysis by antibodies

Paul, S. (1994). Molecular cloning of a proteolytic antibody light chain. *J. Biol.* 

Nevinskii, G. A. (2004). Antibodies to DNA in the blood of patients with tick-borne

D. S.; Eshhar, Z.; Green, B. S. & Knossow, M. (1994). Crystal structure of catalytic antibody Fab with esterase-like activity. *Current Biology*, *Structure*, 2(3), 175-183,

DNA-protein complexes. Natural targets for DNA-hydrolyzing antibodies. *Appl.* 

Gabibov, A. G. (1995) Cleavage of supercoiled plasmid DNA by autoantibody Fab fragment: application of the flow linear dichroism technique. *Proc. Natl. Acad. Sci* 

cytotoxic T lymphocytes. *Proc. Natl. Acad. Sci. USA*, 87(17), 6897-6901, ISSN 0027-

Loscalzo, J. (2008). *Harrison's Principles of Internal Medicine*, 17th Edition. McGraw-

immunoglobulin to be used for the antiglobulin test. *Vox. Sang.,* 25(3), 245-253,

polyclonal immunoglobulins from malaria-infected BALB/c mice on the lupus-like syndrome of (NZB x NZW)F1 mice. *Eur. J. Immunol.,* 24(1), 8-15, ISSN 0014-2980


Baranovskii, A. G.; Ershova, N. A.; Buneva, V. N.; Kanyshkova, T. G.; Mogelnitskii, A. S.;

Barbas, C. F. 3rd; Heine, A.; Zhong, G.; Hoffmann, T.; Gramatikova, S.; Bjornestedt, R.; List,

Barzilai, O.; Ram, M. & Shoenfeld, Y. (2007). Viral infection can induce the production of autoantibodies. *Curr. Opin. Rheumatol*., 19(8), 636-643, ISSN 1040-8711 Bigazzi, P. E. (1983). In: *Mechanisms of Immunology*, Cohen, S.; Wozd, P. E. & McKlaskie, R. T.

Bizub-Bender, D.; Kulkosky, J. & Skalka, A. M. (1994). Monoclonal antibodies against HIV

Blanco, F.; Kalsi, J. & Isenberg, D. A. (1991) Analysis of antibodies to RNA in patients with

Bugreev, D. V.; Baranova, S.; Zakharova,O. D.; Parissi, V.; Desjobert, C.; Sottofattori, E.;

Buneva, V. N.; Andrievskaia, O. A.; Romannikova, I. V.; Gololobov, G. V.; Iadav, R. P.;

Chiu, T. K. & Davies D. R. (2004) Structure and function of HIV-1 integrase. *Curr. Top. Med.* 

Deprez, E.; Tauc, P.; Leh, H., Mouscadet, J. F.; Auclair, C. & Brochon, J. C. (2000). Oligomeric

Dias, S.; Jovic, F.; Renard, P. Y.; Taran, F.; Créminon, C.; Mioskowski, C. & Grassi, J. (2002)

Gabibov, A. G.; Ponomarenko, N. A.: Tretyak, E. B.: Paltsev, M. A. & Suchkov, S. V. (2006)

Dubrovskaya, V. V.; Andryushkova, A. S.; Kuznetsova, I. A.; Toporkova, L. B.; Buneva, V.

Earnshaw, W. C. & Rothfield, N. (1985). Identification of a family of human centromere

HIV-1 integrase. *Biochemistry,* 42(30), 9235-9247. ISSN 0006-2960

anisotropy. *Biochemistry,* 39(31), 9275-9284, ISSN 0006-2960

*Immunol. Methods,* 269(1-2), 81-98, ISSN 0022-1759

*Autoimmun. Rev.,* 5(5), 324-330, ISSN 1568-9972

*Biochemistry (Mosc),* 69(6), 587-601, ISSN 0006-2979

2085-2092, ISSN 0036-8075

1105-1115, ISSN 0889-2229

0026-8933

1088, ISSN 0006-2979

4), 313-321, ISSN 0009-5915

(Ed.), pp. 181-206, Moscow: Meditsina.

*Exp. Immunol*., 86(1), 66-70, ISSN 0009-9104

*Chem.,* 4(9) 965-977, ISSN 1568-0266

Doronin, B. M.; Boiko, A. N.; Gusev, E. I.; Favorova, O. O. & Nevinsky, G. A. (2001). Catalytic heterogeneity of polyclonal DNA-hydrolyzing antibodies from the sera of patients with multiple sclerosis. *Immunol. Lett*., 76(3), 163-167, ISSN 0165-2478 Baranovskii, A. G.; Buneva, V. N. & Nevinsky, G. A. (2004) Human deoxyribonucleases.

B.; Anderson, J.; Stura, E. A.; Wilson, I. A. & Lerner, R. A. (1997) *Science,* 278(5346),

type 1 integrase: clues to molecular structure. *AIDS Res. Hum. Retroviruses*, 10(9),

systemic lupus erythematosus and other autoimmune rheumatic diseases*. Clin.* 

Balbi, A.; Litvak, S.; Tarrago-Litvak, L. & Nevinsky, G. A. (2003). Dynamic, thermodynamic, and kinetic basis for recognition and transformation of DNA by

Iamkovoi, V. I. & Nevinskii, G. A. (1994) Interaction of catalytically active antibodies with oligoribonucleotides. *Mol. Biol. (Moscow),* 28(4), 738 – 743, ISSN

states of the HIV-1 integrase as measured by time-resolved fluorescence

Immunologically driven chemical engineering of antibodies for catalytic activity. *J.* 

Catalytic autoantibodies in clinical autoimmunity and modern medicine.

N.; Orlovskaya, I. A. & Nevinsky, G. A. (2003). DNA-hydrolyzing antibodies from sera of autoimmune-prone MRL/MpJ-lpr mice, *Biochemistry (Mosc.)*, 68(10), 1081-

proteins using autoimmune sera from patients with scleroderma. *Chromosoma,* 91(3-


Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 187

Krasnorutskii, M. A.; Buneva, V. N. & Nevinsky, G. A. (2008a). Antibodies against DNA

Krasnorutskii, M. A.; Buneva, V. N. & Nevinsky, G. A. (2008b). Antibodies against RNA hydrolyze RNA and DNA. *J. Mol. Recognit*., 21(5), 337–346, ISSN 0952-3499 Krasnorutskii, M. A.; Buneva, V. N. & Nevinsky, G. A. (2008c). Immunization of rabbits with

Krasnorutskii, M. A.; Buneva, V. N. & Nevinsky, G. A. (2008d) Anti-RNase Antibodies

Krasnorutskii, M. A.; Buneva, V. N. & Nevinsky, G. A. (2009). Immunization of rabbits with

Kuznetsova, I. A.; Orlovskaya, I. A.; Buneva, V. N. & Nevinsky, G. A. (2007). Activation of

Lacroix-Desmazes, S.; Moreau, A.; Sooryanarayana; Bonnemain, C.; Stieltjes, N.; Pashov, A.;

Lacroix-Desmazes, S.; Bayry, J.; Kaveri, S.V.; Hayon-Sonsino, D.; Thorenoor, N.;

Li, L.; Paul, S.; Tyutyulkova, S.; Kazatchkine, M. D. & Kaveri, S. (1995). Catalytic activity of anti-thyroglobulin antibodies. *J. Immunol.,* 154(7), 3328-3332, ISSN 0022-1767 Martin, A. B. & Schultz, P. G. (1999). Opportunities at the interface of chemistry and biology.

Matsiota-Bernard, P.; Mahana, W.; Avrameas, S. & Nauciel, C. (1993). Specific and natural

susceptible and resistant mice. *Immunology,* 79(3), 375-380, ISSN 0019-2805 Matsiota-Bernard, P.; Hentati, B.; Pie, S.; Legakis, N.; Nauciel, C. & Avrameas, S. (1996)

Mestecky, J. & McGhee, J. R. (1987) Immunoglobulin A (IgA): molecular and cellular

61, ISSN 0273-2289

9639

*Recognit*., 21(4), 233–242, ISSN 0952-3499

hydrolyze RNA and DNA. *Int. Immunol.*, 20(8), 1031-1040, ISSN 0953-8178

activities. *Int. Immunol.,* 21(4), 349-360, ISSN 0953-8178

against pancreatic ribonuclease A

1044-1047, ISSN 1078-8956

153-245, ISSN 0065-2776

*Mol. Med*., 14(3), 699–709, ISSN 1582-1838

*Trends Cell Biol.,* 9(12), 24-28, ISSN 0962-8924

*Clin. Exp. Immunol.,* 104(2), 228-235, ISSN 0009-9104

activities of anti-DNA autoantibodies from sera of patients with lymphoproliferative and autoimmune diseases*. Appl. Biochem. Biotechnol*., 75(1), 45-

hydrolyze DNA and RNA*. Biochemistry (Moscow),* 73(11), 1547-1560, ISSN 0006-2979

DNase I produces polyclonal antibodies with DNase and RNase activities. *J. Mol.* 

DNase II leads to formation of polyclonal antibodies with DNase and RNase

DNA-hydrolyzing antibodies from the sera of autoimmune-prone MRL-lpr/lpr mice by different metal ions. *Biochim. Biophys. Acta,* 1774(7), 884-896, ISSN 1570-

Sultan, Y.; Hoebeke, J.; Kazatchkine, M. D. & Kaveri, S. V. (1999). Catalytic activity of antibodies against factor VIII in patients with hemophilia A. *Nat. Med.,* 5(9),

Charpentier, J.; Luyt, C.E.; Mira, J. P.; Nagaraja, V.; Kazatchkine, M. D.; Dhainaut, J. F. & Mallet, V. O. (2005). High levels of catalytic antibodies correlate with favorable outcome in sepsis. *Proc. Natl. Acad. Sci. USA,* 102(11) 4109-4113, ISSN 0027-8424 Legostaeva, G. A.; Polosukhina, D. I.; Bezuglova, A. M.; Doronin, B. M.; Buneva, V. N. &

Nevinsky, G. A*.* (2010) Affinity and catalytic heterogeneity of polyclonal myelin basic protein-hydrolyzing IgGs from sera of patients with multiple sclerosis. *J. Cell* 

antibody production during *Salmonella typhimurium* infection in genetically

Beneficial effect of *Salmonella typhimurium* infection and of immunoglobulins from *S. typhimurium*-infected mice on the autoimmune disease of (NZB x NZW) F1 mice.

interactions involved in IgA biosynthesis and immune response. *Adv. Immunol.,* 40,


Hhachn B. Ch. (1996). Systemic lipus erythematosus. In: *Internal diseases,* E.E. Braunvald,

Horl, W.H.; Wanner, C. & Schollmer, P. (1987). Proteinases in catabolism and malnutrition. *JPEN J. Parenter. Enteral. Nutr.* 11(5 Suppl), 98S-103S, ISSN 0148-6071 Ikeda, R.; Takasaki, Y.; Hirokawa, K.; Takeuchi, K. & Hashimoto, H. (2003). Clinical

Ikehara, S.; Kawamura, M. & Takao, F. (1990). Organ-specific and systemic autoimmune

Ivanen, D. R.; Kulminskaya, A. A.; Eneyskaya, E. V.; Ershova N. A.; Shabalin, K. A.;

Ivanen, D. R.; Kulminskaya, A. A.; Shabalin, K. A.; Isaeva-Ivanova, L. V.; Ershova N. A.;

Janda, K. D.; Lo, L. C.; Lo, C. H.; Sim, M. M.; Wang, R.; Wong, C. H. & Lerner, R. A. (1997).

Jerne, N. K. (1974) Towards a network theory of the immune system. *Ann. Immunol. (Paris),*

Li, L.; Paul, S.; Tyutyulkova, S.; Kazatchkine, M. D. & Kaveri, S. (1995). Catalytic activity of anti-thyroglobulin antibodies. *J. Immunol.,* 154(7), 3328-3332, ISSN 0022-1767 Kalaga, R., Li, L., O'Dell, J. R., & Paul, S. (1995). Unexpected presence of polyreactive

Kanyshkova, T. G.; Semenov, D. V.; Khlimankov, D.; Buneva, V. N. & Nevinsky, G. A. (1997)

Karataeva, N. A.; Gorbunov, D.; Prokudin, I. V.; Buneva, V. N.; Kulminskaya, A. A.;

polysaccharide kinase activity. *Immunol. Lett.,* 103(1), 58-67, ISSN 0165-2478 Karataeva, N. A.; Buneva, V. N. & Nevinsky, G. A. (2006b) Polysaccharide kinase activity of

Keinan E.E. (Ed.), (2005). *Catalytic antibodies,* Wiley-VCH Verlag GmbH and Co. KgaA, ISBN

Kim, J. S.; Lee, S. Y.; Lee, W. R.; Sohn, J. N.; Chung, Y. C.; Shim, H. K.; Lee, S. C.; Kwon, M.

sequence specificity. *J. Biol. Chem.,* 281(22), 15287-15295, ISSN 0021-9258 Kozyr, A. V.; Kolesnikov, A. V.; Aleksandrova, E. S.; Sashchenko, L. P.; Gnuchev, N. V.;

rheumatoid arthritis. *J. Immunol.,* 155(5), 2695-2702, ISSN 0022-1767

human mothers. *FEBS Lett..* 416(1), 23-26, ISSN 0014-5793

disease. *J. Rheumatol*., 30(5), 998-1005, ISSN 0315-162X

*USA,* 87(21), 8341-8344, ISSN 0027-8424

*Sci. Monit*., 10(8), BR273-BR280, ISSN 1234-1010

275(5302), 945-948, ISSN 0036-8075

3-527-30688-9, Morlenbach, Germany

407-419, Moscow: Medicine.

*57*(suppl. 11), 253-260.

125C(1-2), 373-389.

K.D., Isselbakher, R.G. Petersdorf, D. D. Wilson, D.B. Martin & A.S. Fauchi (Eds.),

significance of antibodies to TS1-RNA in patients with mixed connective tissue

diseases originate from defects in hematopoietic stem cells. *Proc. Natl. Acad. Sci.* 

Savel'ev, A. N.; Kanyshkova, T. G.; Buneva V. N.; Nevinsky G. A. & Neustroev, K. N. (2002) Human autoantibodies with amylolytic activity. *Biologia (Bratislava),* 

Saveliev, A. N.; Nevinsky, G. A.; & Neustroev, K. N. (2004). Catalytic properties of IgMs with amylolytic activity isolated from patients with multiple sclerosis. *Med.* 

Chemical selection for catalysis in combinatorial antibody libraries. *Science,*

catalytic antibodies in IgG from unimmunized donors and decreased levels in

DNA-hydrolyzing activity of the light chain of IgG antibodies from milk of healthy

Neustroev, K. N. & Nevinsky, G. A. (2006a). Human milk antibodies with

human milk IgG antibodies. *Biochemistry (Mosc.),* 71(11), 1207-1221, ISSN 0006-2979

H. & Kim, Y. S. (2006). Heavy and light chain variable single domains of an anti-DNA binding antibody hydrolyze both double- and single-stranded DNAs without

Favorov, P. V.; Kotelnikov, M. A.; Iakhnina, E. I.; Astsaturov, I. A.; Prokaeva, T. B.; Alekberova, Z. S.; Suchkov, S. V. & Gabibov, A. G. (1998). Novel functional activities of anti-DNA autoantibodies from sera of patients with lymphoproliferative and autoimmune diseases*. Appl. Biochem. Biotechnol*., 75(1), 45- 61, ISSN 0273-2289


Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 189

Odintsova, E .S.; Buneva, V. N. & Nevinsky, G. A. (2005). Casein-hydrolyzing activity of sIgA antibodies from human milk. *J. Mol. Recognit.,* 18(5) 413-421, ISSN 0952-3499 Odintsova, E. S.; Kharitonova, M. A.; Baranovskii, A. G.; Siziakina, L. P.; Buneva, V. N. &

Odintsova, E. S.; Kharitonova, M. A.; Baranovskii, A. G.; Siziakina, L. P.; Buneva, V. N. &

Odintsova, E. S.; Zaksas, N. P.; Buneva, V. N. & Nevinsky, G. A. (2011). Metal dependent

Parkhomenko, T. A.; Odintsova, E. S.; Buneva, V. N.; Kunder, E. V.; Zhyltsov, I. V.;

857-864, ISSN 0026-8933

ISSN 0006-2979

45-59, ISSN 0952-3499

1536, ISSN 0022-1767

*Ass. J.,* 39*,* 850-853.

7(6), 473-479, ISSN 1568-9972

Nevinskii, G. A. (2006a). DNA-hydrolyzing IgG antibodies from the blood of patients with acquired immune deficiency syndrome. *Mol. Biol. (Moscow),* 40(5),

Nevinskii, G. A. (2006b). Proteolytic activity of IgG antibodies from blood of acquired immunodeficiency syndrome patients. *Biochemistry (Mosc),* 71(3), 251-261,

hydrolysis of β-casein by sIgA antibodies from human milk. *J. Mol. Recognit.,* 24(1),

Senkovich, S. A.; Generalov, I. I. & Nevinsky, G. A. (2009). DNA-hydrolyzing activity of IgG antibodies from the sera of patients with diseases caused by different bacterial infections. *J. Cell. Mol. Med.,* 13(9A), 2875-2887, ISSN 1582-1838 Parkhomenko, T. A.; Buneva, V. N.; Tyshkevich, O. B.; Generalov, I. I.; Doronin, B. M. &

Nevinsky, G. A. (2010). DNA-hydrolyzing activity of IgG antibodies from the sera of patients with tick-borne encephalitis. *Biochimie,* 92(5), 545-554, ISSN 0300-9084 Parkhomenko, T.A.; Legostaeva, G.A.; Doronin, B.M.; Buneva, V.N. & Nevinsky, G.A.

(2010). IgGs containing light chains of the lambda and kappa type and of all subclasses (IgG1-IgG4) from sera of patients with multiple sclerosis hydrolyze

Catalytic hydrolysis of vasoactive intestinal peptide by human autoantibody.

catalytic antibodies: peptide-hydrolyzing activities of Bence Jones proteins and VL

Caturegli, P. & Rose, N. R. (1997). Chacterization of thyroglobulin-directed and polyreactive catalytic antibodies in autoimmune disease. *J. Immunol.,* 159(3), 1530-

Paul, S.; Volle, D. J.; Beach, C. M.; Johnson, D. R.; Powell, M. J. & Massey, R. J. (1989).

Paul, S.; Li, L.; Kalaga, R.; Wilkins-Stevens, P.; Stevens, F. J. & Solomon, A. (1995). Natural

Paul, S.; Li, L.; Kalaga, R.; O'Dell, J.; Dannenbring, R. E. Jr.; Swindells, S.; Hinrichs, S.;

Paul, S.; Nishiyama, Y.; Planque, S. & Tagushi, H. (2006). Theory of proteolytic antibody

Pauling, L. (1946). Molecular basis of biological specificity. *Chem. Eng. News,* 24, 13751-13777. Pollack, S. J.; Jacobs, J. W.; & Schultz, P. G. (1986). Selective chemical catalysis by an

Pisetsky, D. (2001). Immune response to DNA in systemic lupus erythematosus. *Isr. Med.* 

Planque, S.; Nishiyama, Y.; Taguchi, H.; Salas, M.; Hanson, C. & Paul, S. (2008). Catalytic

Polosukhina, D. I.; Kanyshkova, T. G.; Doronin, B. M.; Tyshkevich, O. B.; Buneva, V. N.;

antibodies to HIV: physiological role and potential clinical utility. *Autoimmun. Rev.,* 

Boiko, A. N.; Gusev, E. I.; Favorova, O. O. & Nevinsky, G. A. (2004). Hydrolysis of

DNA. *J. Mol. Recognit*., 23(5), 486-494, ISSN 0952-3499

fragment. *J. Biol. Chem.,* 270(25), 15257-15261, ISSN 0021-9258

occurrence. *Immunol Lett.,* 103(1), 8-16, ISSN 0165-2478

antibody. *Science,* 234(4783), 1570-1573, ISSN 0036-8075

*Science,* 244(4909), 1158-1162, ISSN 0036-8075


Mitsuhashi, S.; Saito, R.; Kurashige, S. & Yamashugi, N. (1978) Ribonucleic acid in the immune response. *Moll. Cell. Biochem.* 20(3), 131-147, ISSN 0300-8177 Nagata, S. & Suda, T. (1995). Fas and Fas ligand: *lpr* and *gld* mutations. *Immunol. Today*,

Nevinsky, G. A.; Kit, Y. Y.; Semenov, D. V.; Khlimankov, D. Y. & Buneva, V. N. (1998).

Nevinsky, G. A.; Kanyshkova, T. G. & Buneva, V. N. (2000b) Natural catalytic antibodies

Nevinsky, G. A.; Kanyshkova, T. G.; Semenov, D. V.; Vlassov, A. V.; Gal'vita, A. V. &

Nevinsky, G. A.; Breusov, A. A.; Baranovskii, A. G.; Prints, A. V.; Kanyshkova, T. G.;

Nevinsky, G. A. & Buneva, V. N. (2002a). Human catalytic RNA- and DNA-hydrolyzing

Nevinsky, G.A.; Favorova, O.O. & Buneva, V.N. (2002b). *Catalytic* antibodies: New characters

Nevinsky, G.A. & Buneva, V.N. (2003). Catalytic antibodies in healthy humans and patients

Nevinsky, G.A. & Buneva, V.N. (2005). Natural catalytic antibodies-abzymes. In: *Catalytic* 

Nevinsky, G. A. & Buneva, V. N. (2010a). Natural catalytic antibodies in norm, autoimmune, viral, and bacterial diseases. *ScientificWorldJournal*, 10, 1203-1233, ISSN 1537-744X Nevinsky GA. (2010b). Natural catalytic antibodies in norm and in autoimmune diseases. In:

Nilsen, B. M.; Haugan, I. R.; Berg K.; Olsen L.; Brown P. O. & Helland D. E. (1996).

Nishi, Y. (2002). Evolution of catalytic antibody repertoire in autoimmune mice. *J. Immunol.* 

Nishi, Y. (2003). Enzyme/abzyme prodrug activation systems: potential use in clinical

antibodies. *J. Immunol. Methods* 269(1-2), 235-249, ISSN 0022-1759

Golemis, (Ed.), 523-534, Cold Spring Harbor, NY

Nova Science Publishers, Inc., ISBN 9781616680077, USA

oncology. *Curr. Pharm. Des*., 9(26), 2113-2130, ISSN 1381-6128

phosphorylation. *Appl. Biochem. Biotechnol.,* 75(1), 77-91, ISSN 0273-2289 Nevinsky, G. A.; Semenov, D. V. & Buneva, V. N. (2000a) Catalytic antibodies (abzymes)

Secretory immunoglobulin A from human milk catalyzes milk protein

induced by stable transition-state analogs. *Biochemistry (Moscow)*, *65*(11), 1233-1244,

(abzymes) in normalcy and pathology. *Biochemistry (Mosc),* 65(11), 1245-1255, ISSN

Buneva, V. N. (2000c). Secretory immunoglobulin A from healthy human mothers' milk catalyzes nucleic acid hydrolysis. *Appl. Biochem. Biotechnol.,* 83(1-3), 115-129,

Galvita, A. V.; Naumov, V. A. & Buneva, V. N. (2001). Effect of different drugs on the level of DNA-hydrolyzing polyclonal IgG antibodies in sera of patients with Hashimoto's thyroiditis and nontoxic nodal goiter. *Med. Sci. Monit.,* 7(2), 201-211,

in the protein repertoire. In*:Protein-protein interactions. A molecular cloning manual.* E.

with autoimmune and viral pathologies. *J. Cell. Mol. Med.,* 7(3), 265-276, ISSN 1582-

*antibodies.* E. Keinan, (Ed.), 503-567, VCH-Wiley press, ISBN 3-527-30688-9,

*Autoimmune Diseases: Symptoms, Diagnosis and Treatment,* K. J. Brenner, (Ed), 1-107,

Monoclonal antibodies against human immunodeficiency virus type 1 integrase: epitope mapping and differential effects on integrase activities in vitro. *J. Virology,*

16(1), 39–43.

ISSN 0006-2979

ISSN 0273-2289

ISSN 1234-1010

Morlenbach, Germany

70(3), 1580-1587, ISSN 0022-538X

*Methods,* 269(1-2), 213-233, ISSN 0022-1759

1838

0006-2979


myelin basic protein by polyclonal catalytic IgGs from the sera of patients with multiple sclerosis. *J. Cell Mol. Med.,* 8(3), 359-368, ISSN 1582-1838

Natural Catalytic Antibodies in Norm and in HIV-Infected Patients 191

Shuster, A. M.; Gololobov, G. V.; Kvashuk, O. A.; Bogomolova, A. E.; Smirnov, I. V. &

Steinman, L. (2001). Multiple sclerosis: a two-stage disease. *Nat. Immunol*., 2(9), 2762-2764,

Stockwin, L. H. & Holmes, S. (2003). Antibodies as therapeutic agents: vive la renaissance!

Sugiama, Y. & Yamamota, T. (1996). Characterization of serum anti-phospholipid antibodies

Tanaka, F. & Barbas, C. F., 3rd. (2002). Reactive immunization: a unique approach to catalytic antibodies. *J. Immunol. Methods,* 269(1-2), 67-79, ISSN 0022-1759 Tawfik, D. S.; Chap, R.; Green, B. S.; Sela, M. & Eshhar, Z. (2002). Unexpectedly high

Tellier, C. (2002). Exploiting antibodies as catalysts: potential therapeutic applications.

Ternynck, P., Falanga, B., Unterkircher, C., Gregoire. J., Da Silva, L. P. & Avrameas S. (1991).

Thayer, M. M.; Olender, E. H.; Arvai, A. S.; Koike, C. K.; Canestrelli, I. L.; Stewart, J. D.;

Thiagarajan, P.; Dannenbring, R.; Matsuura, K.; Tramontano, A.; Gololobov, G. & Paul, S.

Tramontano, A.; Janda, K. D. & Lerner, R. A. (1986a). Catalytic antibodies*. Science,* 234(4783),

Tramontano, A.; Janda, K. D. & Lerner, R. A. (1986b). Chemical reactivity at an antibody

Vlasov, A. V.; Baranovskii, A. G.; Kanyshkova, T. G.; Prints, A. V.; Zabara, V. G.; Naumov,

Suzuki, H. (1994). Recent advances in abzyme studies. *J. Biochem. (Tokyo),* 115(4), 623-628. Tanaka, A.; Lindor, K.; Ansari, A. & Gershwin, M. E. (2000). Fetal microchimerisms in the mother: immunologic implications. *Liver Transpl.,* 6(2) 138-143, ISSN 1527-6465 Tanaka, F. (2002). Catalytic antibodies as designer proteases and esterases. *Chem. Rev*.,

*Expert. Opin. Biol. Ther.,* 3(7), 1133-1152, ISSN 1471-2598

667, ISSN 0036-8075

102(12), 4885-4906, ISSN 0009-2665

*USA*, 92*,* 2145-2149, ISSN 0027-8424

39(21), 6459-6465, ISSN 0006-2960

1316(3), 217-223, ISSN 1570-9639

1566-1570, ISSN 0036-8075

*Transfus. Clin. Biol.,* 9(1), 1-8, ISSN 1246-7820

*chabaudi*. *Int. Immunol*., 3, 29-37, ISSN 0953-8178

ISSN 1529-2908

8727

2836

0026-8933

Gabibov, A. G. (1992). DNA hydrolyzing autoantibodies. *Science,* 256(5057), 665-

in patients with multiple sclerosis. *Tohoku J. Exp. Med.,* 178, 203-215, ISSN 10040-

occurrence of catalytic antibodies in MRL/lpr and SJL mice immunized with a transition-state analog: is there a linkage to autoimmunity? *Proc. Natl. Acad. Sci.* 

Induction of high levels of IgG autoantibodies in mice infected with *Plasmodium* 

Benkovic, S. J.; Getzoff, E. D. & Roberts, V. A. (1999). Structural basis for amide hydrolysis catalyzed by the 43C9 antibody. *J. Mol. Biol.,* 291(2), 329-345, ISSN 0022-

(2000). Monoclonal antibody light chain with prothrombinase activity. *Biochemistry,*

binding site elicited by mechanistic design of a synthetic antigen. *Proc. Natl. Acad. Sci. USA,* 83(18), 6736-6740, ISSN 0027-84Tyutyulkova, S.; Gao, Q. S.; Thompson, A.; Rennard, S. & Paul, S. (1996). Efficient vasoactive intestinal polypeptide hydrolyzing autoantibody light selected by phage display. *Biochim. Biophys. Acta,*

V. A.; Breusov, A. A.; Giege R.; Buneva, V. N. & Nevinskii, G. A. (1998). Substrate specificity of serum DNA- and RNA-hydrolyzing antibodies of patients with polyarthritis and autoimmune thyroiditis. *Mol. Biol. (Mosk),* 32(3), 559-569, ISSN


Polosukhina, D. I.; Buneva, V. N.; Doronin, B. M.; Tyshkevich, O. B.; Boiko, A. N.; Gusev, E.

Polosukhina, D. I.; Buneva, V. N.; Doronin, B. M.; Tyshkevich, O. B.; Boiko, A. N.; Gusev, E.

Rao, M. B.; Tanksale, A. M.; Ghatge, M. S. & Deshhange, V. V. (1998). Molecular and

Raptis, L. & Menard, H. A. (1980). Quantitation and characterization of plasma DNA in

Sun, M.; Li, L.; Gao, Q. S. & Paul, S. (1994). Antigen recognition by an antibody light chain. *J.* 

Sun, M.; Gao, Q. S.; Kirnarskiy, L.; Rees, A. & Paul, S. (1997). Cleavage specificity of a

Sato, T.; Uvhiumi, T.; Arakawa, M. & Kominami, R. (1994) Serological association of lupus

Savel'ev, A. N.; Kanyshkova, T. G.; Kulminskaya, A. A.; Buneva, V. N.; Eneyskaya, E. V.;

Savel'iev, A. N.; Ivanen, D. R.; Kulminskaya, A. A.; Ershova, N. A.; Kanyshkova, T. G.;

Semenov, D. V.; Kanyshkova, T. G.; Karotaeva, N. A.; Krasnorutskii, M. A.; Kuznetsova, I.

Shoenfeld, Y.; Ben-Yehuda, O.; Messinger, Y.; Bentwitch, Z.; Rauch, J.; Isenberg, D. I. &

Shoenfeld, Y.; Teplizki, H. A.; Mendlovic, S.; Blank, M.; Mozes, E. & Isenberg, D. A. (1989).

Schochetman, G. & Massey, R. (1985). International Patent no. W085/02414.

idiotypes. *Immunol. Lett.,* 17(3), 285-291, ISSN 0165-2478

multiple sclerosis. *J. Cell Mol. Med.,* 8(3), 359-368, ISSN 1582-1838

*Monit.,* 11(8), BR266- BR272, ISSN 1234-1010

*Biol. Chem.,* 269(1), 734-738, ISSN 0021-9258

*Mol. Biol.,* 271(3), 374-385, ISSN 0022-2836

10(2), BR23-BR33, ISSN 1234-1010

*Immunopathol.,* 51(3), 313-325.

*Lett.,* 103(1), 75-81, ISSN 0165-2478

635.

1391-1399.

0009-9104

152.

myelin basic protein by polyclonal catalytic IgGs from the sera of patients with

I.; Favorova, O. O. & Nevinsky, G. A. (2005). Hydrolysis of myelin basic protein by IgM and IgA antibodies from the sera of patients with multiple sclerosis. *Med. Sci.* 

I.; Favorova, O. O. & Nevinsky, G. A. (2006). Metal-dependent hydrolysis of myelin basic protein by IgGs from the sera of patients with multiple sclerosis. *Immunol.* 

biotechnological aspects of microbial proteases. *Microbiol. Mol. Biol Rev*., 62(3), 597-

normals and patients with systemic lupus erythematosus. *J. Clin. Invest*., 66(6),

proteolytic antibody light chain and effects of the heavy chain variable domain. *J.* 

autoantibodies to a limited functional domain of 28S ribosomal RNA and to the ribosomal proteins bound to the domain. *Clin. Exp. Immunol*., 98(1), 35-39, ISSN

Filatov, M. V.; Nevinsky, G. A. & Neustroev, K. N. (2001). Amylolytic activity of IgG and sIgA immunoglobulins from human milk. *Clin. Chim. Acta,* 314(1-2), 141-

Buneva, V. N.; Mogelnitskii, A. S.; Doronin, B. M.; Favorova, O. O.; Nevinsky, G. A. & Neustroev, K. (2003). Amylolytic activity of IgM and IgG antibodies from patients with multiple sclerosis. *Immunol. Lett.,* 86(3), 291-297, ISSN 0165-2478

A.; Buneva, V. N. & Nevinsky, G. A. (2004). Catalytic nucleotide-hydrolyzing antibodies in milk and serum of clinically healthy human mothers. *Med. Sci. Monit.,*

Gadoth, N. (1988) Autoimmune diseases other than lupus share common anti-DNA

The role of the human anti-DNA idiotype 16/6 in autoimmunity. *Clin. Immunol.* 


Vlasov, A. V.; Helm, M.; Naumov, V. A.; Breusov, A. A.; Buneva, V. N.; Florentz, C.; Giege, R. & Nevinskii, G. A. (1999). Features of tRNA hydrolysis by autoantibodies from blood serum of patients with certain autoimmune and virus diseases. *Mol. Biol. (Moscow),* 33(5), 866-872, ISSN 0026-8933

**10** 

*South Africa* 

**RNAi-Based Gene Expression** 

Fiona T. van den Berg and Marc S. Weinberg

*Antiviral Gene Therapy Research Unit (AGTRU), University of the Witwatersrand* 

Antiretroviral drugs have made a significant impact on HIV/AIDS disease progression and have significantly extended the life expectancy of HIV-infected individuals, particularly when used in combination therapies such as HAART (highly active antiretroviral therapy). However, despite this success, recent reports indicate that HIV morbidity and mortality remain at epidemic proportions. It is estimated that over 33 million people are currently infected with the virus worldwide, while sub-Saharan Africa accounted for approximately 70% of all infected individuals and AIDS-related deaths in 2009 (UNAIDS, 2010). Issues of drug resistance, drug toxicity, correct patient compliance and the inability to remove latent reservoirs of infection remain significant problems to overcome. The need to develop novel and improved therapeutic strategies against HIV therefore remains an important medical objective. Gene-based therapies hold much promise as alternative treatment strategies for HIV/AIDS. Anti-viral gene therapies aim to provide a once-off, long-lasting treatment against the virus and thereby address some of the shortcomings associated with current

A gene therapy against HIV offers several unique advantages, including the sustained inhibition of viral replication and the removal of virus from cellular reservoirs. Moreover, by improving specificity, the common toxic side effects associated with current antiretrovial regimens can be diminished. A number of different RNA-based and protein-based gene therapy strategies have been explored and some have reached phase 1 and 2 clinical trials. Our research focuses on the development of RNA-based antiviral strategies and in particular, those that utilise gene expression strategies based on RNA interference (RNAi). In this chapter, we examine basic concepts and review recent advances in the development of expressed RNAibased systems against HIV, with a focus on progress in construct and target design. We also discuss topics related to the use of RNAi-based strategies, including appropriate construct expression, target specificity, viral escape mutations and effective construct delivery. We aim to identify desirable properties of an RNAi-based anti-HIV therapy and highlight the future

RNA interference (RNAi) is a gene silencing phenomenon in which RNA molecules act to silence the expression of particular genes at a post-transcriptional level in the cell. RNAi has

developments that are required to make this approach a reality.

**1. Introduction**

antiretroviral therapies.

**2. RNA interference**

**Strategies to Combat HIV** 


## **RNAi-Based Gene Expression Strategies to Combat HIV**

Fiona T. van den Berg and Marc S. Weinberg *Antiviral Gene Therapy Research Unit (AGTRU), University of the Witwatersrand South Africa* 

## **1. Introduction**

192 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Vlasov, A. V.; Helm, M.; Naumov, V. A.; Breusov, A. A.; Buneva, V. N.; Florentz, C.; Giege,

Vlassov, A.; Florentz, C.; Helm, M.; Naumov, V.; Buneva, V.; Nevinsky, G. & Giege, R.

Wentworth, P. Jr.; Liu, Y.; Wentworth, A. D.; Fan, P.; Foley, M. J. & Janda, K. D. (1998). A

Wójcik, T. & Kieć-Kononowicz, K. (2008). Catalytic activity of certain antibodies as a

Yi, J.; Arthur, J. W.; Dunbrack, R. L.; Jr. & Skalka A. M. (2000). An inhibitory monoclonal

Zhou, Y. X.; Karle, S.; Taguchi, H.; Planque, S.; Nishiyama, Y. & Paul, S. (2002). Prospects for

Zouali, M. (2001). B cell tolerance to self in systemic autoimmunity. *Arch. Immunol. Ther. Exp.* 

HIV-1 integrase. *J. Biol. Chem.,* 275(49), 38739-38748, ISSN 0021-9258 Zandman-Goddard, G.; Shoenfeld, Y. (2002). HIV and autoimmunity. *Autoimmun. Rev.,* 1(6),

hydrolysis. *Proc. Natl. Acad. Sci. USA,* 95(11), 5971-5975, ISSN 0027-8424 Wlodawer, A. (1999). Crystal structures of catalytic core domains of retroviral integrases and

with RNAse activity, *Nucleic Acid Res.,* 26(3), 5243-5250, ISSN 0305-1048 Watanabe-Fukunaga, R.; Brannan, C. I. & Copeland, N. G. (1992). Lymphoproliferation

*(Moscow),* 33(5), 866-872, ISSN 0026-8933

*Nature,* 356(6367), 314-317, ISSN 0968-0004

*Chem.,* 15(16), 1606-1615, ISSN 0929-8673

*(Warsz),* 49(5), 361-365, ISSN 0004-069X

329-337, ISSN 1568-9972

ISSN 0022-1759

0065-3527

R. & Nevinskii, G. A. (1999). Features of tRNA hydrolysis by autoantibodies from blood serum of patients with certain autoimmune and virus diseases. *Mol. Biol.* 

(1998). Characterization and selectivity of catalytic antibodies from human serum

disorder in mice explained by defects in Fas antigen that mediates apoptosis.

bait and switch hapten strategy generates catalytic antibodies for phosphodiester

role of divalent cations in enzymatic activity, *Adv.Virus. Res.,* 52, 335-350, ISSN

potential tool for drug synthesis and for directed prodrug therapies. *Curr. Med.* 

antibody binds at the turn of the helix-turn-helix motif in the N-terminal domain of

immunotherapeutic proteolytic antibodies. *J. Immunol. Methods,* 269(1-2), 257-268,

Antiretroviral drugs have made a significant impact on HIV/AIDS disease progression and have significantly extended the life expectancy of HIV-infected individuals, particularly when used in combination therapies such as HAART (highly active antiretroviral therapy). However, despite this success, recent reports indicate that HIV morbidity and mortality remain at epidemic proportions. It is estimated that over 33 million people are currently infected with the virus worldwide, while sub-Saharan Africa accounted for approximately 70% of all infected individuals and AIDS-related deaths in 2009 (UNAIDS, 2010). Issues of drug resistance, drug toxicity, correct patient compliance and the inability to remove latent reservoirs of infection remain significant problems to overcome. The need to develop novel and improved therapeutic strategies against HIV therefore remains an important medical objective. Gene-based therapies hold much promise as alternative treatment strategies for HIV/AIDS. Anti-viral gene therapies aim to provide a once-off, long-lasting treatment against the virus and thereby address some of the shortcomings associated with current antiretroviral therapies.

A gene therapy against HIV offers several unique advantages, including the sustained inhibition of viral replication and the removal of virus from cellular reservoirs. Moreover, by improving specificity, the common toxic side effects associated with current antiretrovial regimens can be diminished. A number of different RNA-based and protein-based gene therapy strategies have been explored and some have reached phase 1 and 2 clinical trials. Our research focuses on the development of RNA-based antiviral strategies and in particular, those that utilise gene expression strategies based on RNA interference (RNAi). In this chapter, we examine basic concepts and review recent advances in the development of expressed RNAibased systems against HIV, with a focus on progress in construct and target design. We also discuss topics related to the use of RNAi-based strategies, including appropriate construct expression, target specificity, viral escape mutations and effective construct delivery. We aim to identify desirable properties of an RNAi-based anti-HIV therapy and highlight the future developments that are required to make this approach a reality.

## **2. RNA interference**

RNA interference (RNAi) is a gene silencing phenomenon in which RNA molecules act to silence the expression of particular genes at a post-transcriptional level in the cell. RNAi has

RNAi-Based Gene Expression Strategies to Combat HIV 195

Fig. 1. The mammalian microRNA biogenesis pathway which mediates RNAi. Artificial antiviral constructs can be introduced at points 1, 2 and 3 of the pathway and processed to

inhibit viral replication.

become a popular tool in the development of antiviral therapeutics as the potent silencing mechanism can be redirected against viral genes to inhibit viral replication. RNAi was first described by Fire *et. al*. (Fire *et al.*, 1998) and has since been demonstrated in a number of different organisms including yeast, plants and animals. The mediators of RNAi are short 21- and 22- nucleotide (nt) RNAs known as small interfering RNAs (siRNAs) or microRNAs (miRNAs) and are derived from longer double stranded (ds) RNAs. SiRNAs/miRNAs direct the silencing of complementary gene transcripts in a sequence-specific manner. RNAi was first described in mammalian cells in 2001 (Elbashir *et al.*, 2001a) and subsequent research has moved swiftly to reveal a number of pathway components and mechanisms. Mammalian RNAi is now emerging as a complex network with several alternative RNA forms and levels of regulatory interactions (Breving and Esquela-Kerscher, 2009); (Ding *et al.*, 2009). RNAi–based therapeutics, however, still make use of the central RNAi pathway involved in miRNA biogenesis (Figure 1). Our discussion of RNAi-based strategies begins here with a description of this major RNAi pathway and how it can be redirected to inhibit viral replication.

The mammalian RNAi pathway mediates gene silencing through the generation of miRNAs. MiRNA regulation of genes is both essential and ubiquitous and has been implicated in the regulation of developmental timing, cellular differentiation, apoptosis, cell proliferation and organ development (Bartel, 2004). MiRNAs are expressed from non-protein-coding genes in intergenic or intronic regions as single or polycistronic transcripts by RNA polymerase II (Lee *et al.*, 2004); (Cullen, 2004). MiRNA transcripts are usually several kilobases long and fold back upon themselves to form characteristic hairpin structures known as primarymicroRNAs (pri-miRNAs) with flanking sequences, a partially duplexed stem and a terminal loop. Pri-miRNAs are processed in the RNAi pathway in two successive enzymatic steps (Lee *et al.*, 2002b) to produce mature miRNAs from the double-stranded stem region.

The first processing step occurs in the nucleus where the pri-miRNA is cleaved asymmetrically by the "microprocessor" complex to produce a shorter ~ 70 nt hairpin known as a precursor-microRNA (pre-miRNA) with a 2 nt overhang at the 3' hydroxyl end. The microprocessor complex includes two essential proteins, namely, the RNAse III enzyme Drosha and the DGCR8 (DiGeorge critical region 8) protein (Han *et al.*, 2004). The premiRNA is exported from the nucleus to the cytoplasm by the nuclear karyopherin Exportin-5 (Exp-5) in a Ran-GTP-dependant manner (Yi *et al.*, 2003); (Lund *et al.*, 2004). In the second processing step, the pre-miRNA is cleaved asymmetrically by another RNase III enzyme, Dicer (Paddison *et al.*, 2002) to produce staggered ~22 base pair (bp) miRNA/miRNA\* duplex with 2 nt overhangs at each 3' hydroxyl end. Dicer is thought to form a complex with TRBP (TAR RNA-binding protein) (Chendrimada *et al.*, 2005) and PACT (protein activator of protein kinase PKR) (Lee *et al.*, 2006).

One strand of the miRNA/miRNA\* duplex is selected as the mature miRNA or guide strand and loaded into the RNA-induced silencing complex (RISC) (Martinez *et al.*, 2002a). The guide strand (miRNA) is typically selected from the duplex as a result of weaker 5' end pairing, while the remaining passenger strand (miRNA\*) is degraded (Khvorova *et al.*, 2003); (Schwarz *et al.*, 2003). In certain cases, both strands of the duplex may be capable of RISC incorporation. RISC facilitates sequence-specific gene silencing and is directed by the guide sequence to complementary regions in the 3' untranslated regions of target messenger RNAs. RISC-targeting results in the cleavage, degradation or translational suppression of a gene transcript, depending on the level of total complementation between the miRNA and

become a popular tool in the development of antiviral therapeutics as the potent silencing mechanism can be redirected against viral genes to inhibit viral replication. RNAi was first described by Fire *et. al*. (Fire *et al.*, 1998) and has since been demonstrated in a number of different organisms including yeast, plants and animals. The mediators of RNAi are short 21- and 22- nucleotide (nt) RNAs known as small interfering RNAs (siRNAs) or microRNAs (miRNAs) and are derived from longer double stranded (ds) RNAs. SiRNAs/miRNAs direct the silencing of complementary gene transcripts in a sequence-specific manner. RNAi was first described in mammalian cells in 2001 (Elbashir *et al.*, 2001a) and subsequent research has moved swiftly to reveal a number of pathway components and mechanisms. Mammalian RNAi is now emerging as a complex network with several alternative RNA forms and levels of regulatory interactions (Breving and Esquela-Kerscher, 2009); (Ding *et al.*, 2009). RNAi–based therapeutics, however, still make use of the central RNAi pathway involved in miRNA biogenesis (Figure 1). Our discussion of RNAi-based strategies begins here with a description of this major RNAi pathway and how it can be redirected to inhibit

The mammalian RNAi pathway mediates gene silencing through the generation of miRNAs. MiRNA regulation of genes is both essential and ubiquitous and has been implicated in the regulation of developmental timing, cellular differentiation, apoptosis, cell proliferation and organ development (Bartel, 2004). MiRNAs are expressed from non-protein-coding genes in intergenic or intronic regions as single or polycistronic transcripts by RNA polymerase II (Lee *et al.*, 2004); (Cullen, 2004). MiRNA transcripts are usually several kilobases long and fold back upon themselves to form characteristic hairpin structures known as primarymicroRNAs (pri-miRNAs) with flanking sequences, a partially duplexed stem and a terminal loop. Pri-miRNAs are processed in the RNAi pathway in two successive enzymatic steps (Lee *et al.*, 2002b) to produce mature miRNAs from the double-stranded stem region. The first processing step occurs in the nucleus where the pri-miRNA is cleaved asymmetrically by the "microprocessor" complex to produce a shorter ~ 70 nt hairpin known as a precursor-microRNA (pre-miRNA) with a 2 nt overhang at the 3' hydroxyl end. The microprocessor complex includes two essential proteins, namely, the RNAse III enzyme Drosha and the DGCR8 (DiGeorge critical region 8) protein (Han *et al.*, 2004). The premiRNA is exported from the nucleus to the cytoplasm by the nuclear karyopherin Exportin-5 (Exp-5) in a Ran-GTP-dependant manner (Yi *et al.*, 2003); (Lund *et al.*, 2004). In the second processing step, the pre-miRNA is cleaved asymmetrically by another RNase III enzyme, Dicer (Paddison *et al.*, 2002) to produce staggered ~22 base pair (bp) miRNA/miRNA\* duplex with 2 nt overhangs at each 3' hydroxyl end. Dicer is thought to form a complex with TRBP (TAR RNA-binding protein) (Chendrimada *et al.*, 2005) and PACT (protein activator

One strand of the miRNA/miRNA\* duplex is selected as the mature miRNA or guide strand and loaded into the RNA-induced silencing complex (RISC) (Martinez *et al.*, 2002a). The guide strand (miRNA) is typically selected from the duplex as a result of weaker 5' end pairing, while the remaining passenger strand (miRNA\*) is degraded (Khvorova *et al.*, 2003); (Schwarz *et al.*, 2003). In certain cases, both strands of the duplex may be capable of RISC incorporation. RISC facilitates sequence-specific gene silencing and is directed by the guide sequence to complementary regions in the 3' untranslated regions of target messenger RNAs. RISC-targeting results in the cleavage, degradation or translational suppression of a gene transcript, depending on the level of total complementation between the miRNA and

viral replication.

of protein kinase PKR) (Lee *et al.*, 2006).

Fig. 1. The mammalian microRNA biogenesis pathway which mediates RNAi. Artificial antiviral constructs can be introduced at points 1, 2 and 3 of the pathway and processed to inhibit viral replication.

RNAi-Based Gene Expression Strategies to Combat HIV 197

Fig. 2. Various forms of therapeutic constructs used to trigger antiviral RNAi. Each construct generates antiviral siRNA or miRNA guide sequences (green, yellow, red and blue) which initiate silencing of complementary viral targets. Constructs may be non-expressed or

expressed in a singular or multiple format (combinatorial-RNAi).

target sequence (Hutvagner and Zamore, 2002); (Zeng *et al.*, 2003). Guide strands with nearperfect complementation tend to mediate cleavage of the target by activating the core catalytic Argonaute-2 (Ago-2) protein of RISC. Target cleavage is also a typical characteristic of siRNAmediated gene silencing. Most miRNAs exhibit incomplete complementation, resulting in translational suppression. Complete complementation, however, is still required for the seed region of the miRNA (5' position 2-7) for effective target knockdown (Brennecke *et al.*, 2005) and a single mismatch in this region can prevent silencing (Elbashir *et al.*, 2001b).

To redirect the RNAi silencing pathway to silence viral genes, artificial anti-viral siRNAs and miRNAs can be introduced into the cell to enter different points of the RNAi pathway (Figure 1: 1, 2, 3). Antiviral guide sequences are designed to be complementary to viral transcripts and can be incorporated into various forms of artificial RNAi intermediates including pri-miRNAs, pre-miRNAs, and miRNAs. Once in the pathway, these anti-viral intermediates are processed to give therapeutic guide sequences which act to suppress viral gene expression and inhibit viral replication. RNAi strategies have been used against incoming viral RNA to prevent integration, but it appears as if the strength of RNAi remains in its role in post-transcriptionl gene silencing (PTGS). Incoming viral RNA appears to be unsuitable for targeting as it is bound by several proteins and transcribed within a short time frame, which may limit its susceptibility to RNAi. A number of different antiviral constructs and targeting and delivery strategies have been investigated against HIV with varying success and are discussed in the following sections.

## **3. Antiviral RNAi constructs**

The potential application of RNAi for the treatment of HIV was recognised shortly after the first application of RNAi modalities in mammalian cells (Capodici *et al.*, 2002; Coburn and Cullen, 2002; Jacque *et al.*, 2002; Lee *et al.*, 2002a; Martinez *et al.*, 2002b). However, it soon became apparent that RNAi therapies fall into two broad groups: those that are expressed in the cell and those that are not. The basic forms of antiviral RNAi constructs are shown in Figure 2. There are specific advantages and drawbacks associated with both synthetic and expressed constructs with regard to delivery, duration of inhibition and dose control (Table 1), but which type of anti-viral construct is best suited for the treatment of HIV?

## **3.1 Non-expressed, synthetic RNAi constructs**

Small interfering RNAs (siRNAs) are the most common form of non-expressed RNAi constructs. Initial studies of RNAi induction in mammalian cells showed that siRNAs can be used as powerful tools for artificial gene silencing. Despite the relative potency of siRNAs, their use in a permanent therapeutic application is limited by the lack of continued expression (Tuschl and Borkhardt, 2002). Although, this feature can be useful for particular applications where topical administration is possible and doses can be more easily controlled. An siRNA (ALN-RSV01) against nucleocapsid expression of Respiratory Syncytial Virus (RSV) was successfully delivered to healthy individuals in the form of a nasal spray in a randomized, double-blind, placebo-controlled clinical trial (Devincenzo *et al.*, 2010). The treatment was shown to decrease the number of infected subjects by 38%, independently of other factors like pre-existing RSV antibody and intranasal proinflammatory cytokines. While siRNAs may not be suitable for once-off gene therapies, this example demonstrates how novel delivery methods can enable successful siRNA use in a therapeutic setting.

target sequence (Hutvagner and Zamore, 2002); (Zeng *et al.*, 2003). Guide strands with nearperfect complementation tend to mediate cleavage of the target by activating the core catalytic Argonaute-2 (Ago-2) protein of RISC. Target cleavage is also a typical characteristic of siRNAmediated gene silencing. Most miRNAs exhibit incomplete complementation, resulting in translational suppression. Complete complementation, however, is still required for the seed region of the miRNA (5' position 2-7) for effective target knockdown (Brennecke *et al.*, 2005)

To redirect the RNAi silencing pathway to silence viral genes, artificial anti-viral siRNAs and miRNAs can be introduced into the cell to enter different points of the RNAi pathway (Figure 1: 1, 2, 3). Antiviral guide sequences are designed to be complementary to viral transcripts and can be incorporated into various forms of artificial RNAi intermediates including pri-miRNAs, pre-miRNAs, and miRNAs. Once in the pathway, these anti-viral intermediates are processed to give therapeutic guide sequences which act to suppress viral gene expression and inhibit viral replication. RNAi strategies have been used against incoming viral RNA to prevent integration, but it appears as if the strength of RNAi remains in its role in post-transcriptionl gene silencing (PTGS). Incoming viral RNA appears to be unsuitable for targeting as it is bound by several proteins and transcribed within a short time frame, which may limit its susceptibility to RNAi. A number of different antiviral constructs and targeting and delivery strategies have been investigated against HIV with

The potential application of RNAi for the treatment of HIV was recognised shortly after the first application of RNAi modalities in mammalian cells (Capodici *et al.*, 2002; Coburn and Cullen, 2002; Jacque *et al.*, 2002; Lee *et al.*, 2002a; Martinez *et al.*, 2002b). However, it soon became apparent that RNAi therapies fall into two broad groups: those that are expressed in the cell and those that are not. The basic forms of antiviral RNAi constructs are shown in Figure 2. There are specific advantages and drawbacks associated with both synthetic and expressed constructs with regard to delivery, duration of inhibition and dose control (Table

Small interfering RNAs (siRNAs) are the most common form of non-expressed RNAi constructs. Initial studies of RNAi induction in mammalian cells showed that siRNAs can be used as powerful tools for artificial gene silencing. Despite the relative potency of siRNAs, their use in a permanent therapeutic application is limited by the lack of continued expression (Tuschl and Borkhardt, 2002). Although, this feature can be useful for particular applications where topical administration is possible and doses can be more easily controlled. An siRNA (ALN-RSV01) against nucleocapsid expression of Respiratory Syncytial Virus (RSV) was successfully delivered to healthy individuals in the form of a nasal spray in a randomized, double-blind, placebo-controlled clinical trial (Devincenzo *et al.*, 2010). The treatment was shown to decrease the number of infected subjects by 38%, independently of other factors like pre-existing RSV antibody and intranasal proinflammatory cytokines. While siRNAs may not be suitable for once-off gene therapies, this example demonstrates how novel delivery methods can enable successful siRNA use in a

1), but which type of anti-viral construct is best suited for the treatment of HIV?

and a single mismatch in this region can prevent silencing (Elbashir *et al.*, 2001b).

varying success and are discussed in the following sections.

**3.1 Non-expressed, synthetic RNAi constructs** 

**3. Antiviral RNAi constructs** 

therapeutic setting.

Fig. 2. Various forms of therapeutic constructs used to trigger antiviral RNAi. Each construct generates antiviral siRNA or miRNA guide sequences (green, yellow, red and blue) which initiate silencing of complementary viral targets. Constructs may be non-expressed or expressed in a singular or multiple format (combinatorial-RNAi).

RNAi-Based Gene Expression Strategies to Combat HIV 199

Exogenous siRNAs in the form of duplexes were found to be the most effective at redirecting the silencing mechanism against both endogenous and transfected genes (Elbashir *et al.*, 2001a), while the characteristic 2 nt overhangs at both the 5' and 3' ends are an essential feature (Caplen *et al.*, 2001). Artificial siRNAs are selected for RISC incorporation more directly and there is no need for Drosha or Dicer processing (Figure 1, point 1). This can prevent saturation of the RNAi pathway components and interference with the essential miRNA biogenesis. The use of short siRNA duplexes is preferable over longer dsRNAs which were previously used, as interferon responses in the cell can be avoided. The most effective synthetic siRNA duplexes are designed to be about 19 – 21 nts in length with 3' overhangs, but it has also been shown that shorter siRNAs of only 16 nts may trigger more potent RNAi as a result of higher RISC-loading capacity(Chu and Rana,

The chemical synthesis of siRNA duplexes is conceptually simple, as are the methods of electroporation, microinjection and liposome-mediated transfections for siRNA introduction into cells. The continuous need for these methods in large-scale siRNA use, however, can become labour-intensive. Coupled with the transient nature of siRNA-induced gene silencing (maximum ~ 1 week), the advantages of an expressed siRNA construct in therapeutic applications are clear. Nevertheless, novel delivery mechanisms for synthetic siRNAs have been developed for HIV. Specifically, studies in humanized mouse models of HIV using anti-gp120 aptamer-siRNA chimeras have demonstrated their potential as specific antiviral agents (Neff *et al.*, 2011); (MacRae *et al.*, 2008); (Zhou and Rossi, 2010). Similarly, siRNAs conjugated to single-chained monoclonal antibody fragments targeted to T-cell or lymphocyte-specific receptors have shown promise *in vivo* (Kim *et al.*, 2010; Kumar *et al.*, 2008). However, these methods are in the earlier stages of development and there are still several delivery and potency hurdles which must be overcome. The focus of this review will be on expressed RNAi modalities, where siRNA duplexes are expressed from plasmid

SiRNAs that are generated from expression cassettes have the advantage of sustained production which makes them suitable for long-term, once-off therapeutic applications. Recent developments in construct design and delivery methods have shown much promise for the advancement of siRNA expression systems against HIV. In earlier siRNA expression studies, linear cassettes were constructed to code for both sense and antisense sequences of the siRNA duplex under the control of separate promoters and termination signals (Lee *et al.*, 2002a); (Miyagishi and Taira, 2002). Expressed siRNAs then associate posttranscriptionally to form a duplex with 2 – 4 nt uridine overhangs. This system was found to be effective against HIV-1 sequences and siRNAs targeting a highly-accessible region of the *rev* transcript were found to inhibit viral transcript expression (Lee *et al.*, 2002a). However, the reliability of correct siRNA duplex association *in vivo* is questionable. Most siRNA expression constructs are now designed to generate mimics of RNAi intermediates in the form of siRNA or miRNA hairpin precursors which are processed by Drosha and/or Dicer enzyme complexes (Figure 2). This is preferable for more reliable siRNA processing in a manner that is regulated by the RNAi pathway, but inappropriate expression levels can lead

The choice of promoter is therefore critical in achieving suitable levels of construct expression. Polymerase III (pol III) promoter sequences, like human U6 snRNA (small

DNA vectors with lasting effects on gene silencing.

**3.2 Expressed RNAi constructs**

to saturation of critical RNAi components.

2008).


Table 1. A comparison of non-expressed and expressed antiviral RNAi constructs.

#### **3.1.1 siRNAs**

SiRNA duplexes have been shown to effectively silence a number of HIV target genes. SiRNAs against mRNAs of the Gag protein and CD4 cellular receptor have been shown to inhibit post-integrative expression events with a four-fold reduction in viral entry and a 47 % decrease in p24 expression in HIV cell culture challenge assays (Novina *et al.*, 2002). SiRNAs against *tat* and *rev* transcripts specifically inhibited protein function and viral replication in human T cell lines and primary lymphocytes (Coburn and Cullen, 2002). SiRNA duplexes against the long terminal repeat (LTR) and accessory genes *vif* and *nef* were shown to be effective at inhibiting viral production from infectious molecular clones by 30 to 50 fold in 24 hours (Jacque *et al.*, 2002).

Small interfering RNAs

No. Therapeutic effects are transient. Multiple or continuous treatments are

No. SiRNAs are chemically

No. SiRNAs are suitable substrates for direct association

Less likely as processing and export enzymes are not utilised.

predicted and controlled as construct expression is not a

modifications can be included for enhanced absorption or targeting to a specific tissue.

Table 1. A comparison of non-expressed and expressed antiviral RNAi constructs.

SiRNA duplexes have been shown to effectively silence a number of HIV target genes. SiRNAs against mRNAs of the Gag protein and CD4 cellular receptor have been shown to inhibit post-integrative expression events with a four-fold reduction in viral entry and a 47 % decrease in p24 expression in HIV cell culture challenge assays (Novina *et al.*, 2002). SiRNAs against *tat* and *rev* transcripts specifically inhibited protein function and viral replication in human T cell lines and primary lymphocytes (Coburn and Cullen, 2002). SiRNA duplexes against the long terminal repeat (LTR) and accessory genes *vif* and *nef* were shown to be effective at inhibiting viral production from infectious molecular clones by 30 to

(siRNAs)

required.

synthesised.

with RISC.

Dose Doses can be more easily

factor.

50 fold in 24 hours (Jacque *et al.*, 2002).

Delivery SiRNAs can be administered directly. Chemical

Singular forms

forms

Combinatorial

Potential as a once-off treatment?

Regulated cellular production?

Dicer/ Drosha processing?

Saturation of the RNAi pathway?

**3.1.1 siRNAs** 

**Non-Expressed Constructs Expressed Constructs** 

Multiple siRNAs Multiple shRNA cassettes

Short hairpin RNAs (shRNAs) Primary microRNA mimics

Extended short hairpins (e-shRNAs) Long hairpin RNAs (lhRNAs,

Polycistronic microRNA mimics

Yes. Auto-expression sustains the supply of anti-viral effectors.

Yes, but this is dependent on the type of promoter selected. Cell-specific or inducible promoters are favourable.

Yes. Anti-viral siRNA/miRNA guides must be processed from RNA precursors by Drosha and/or Dicer.

A serious concern related to overexpression of artificial constructs. Choice of expression system is

Exact expression levels under specific promoters can only be determined empirically and may vary with time

More complex. Viral vectors are often used for transduction, but issues with safety and efficacy persist. Cell-based delivery systems appear promising.

and genetic background.

(shRNA-miRs)

dlhRNAs)

critical.

Exogenous siRNAs in the form of duplexes were found to be the most effective at redirecting the silencing mechanism against both endogenous and transfected genes (Elbashir *et al.*, 2001a), while the characteristic 2 nt overhangs at both the 5' and 3' ends are an essential feature (Caplen *et al.*, 2001). Artificial siRNAs are selected for RISC incorporation more directly and there is no need for Drosha or Dicer processing (Figure 1, point 1). This can prevent saturation of the RNAi pathway components and interference with the essential miRNA biogenesis. The use of short siRNA duplexes is preferable over longer dsRNAs which were previously used, as interferon responses in the cell can be avoided. The most effective synthetic siRNA duplexes are designed to be about 19 – 21 nts in length with 3' overhangs, but it has also been shown that shorter siRNAs of only 16 nts may trigger more potent RNAi as a result of higher RISC-loading capacity(Chu and Rana, 2008).

The chemical synthesis of siRNA duplexes is conceptually simple, as are the methods of electroporation, microinjection and liposome-mediated transfections for siRNA introduction into cells. The continuous need for these methods in large-scale siRNA use, however, can become labour-intensive. Coupled with the transient nature of siRNA-induced gene silencing (maximum ~ 1 week), the advantages of an expressed siRNA construct in therapeutic applications are clear. Nevertheless, novel delivery mechanisms for synthetic siRNAs have been developed for HIV. Specifically, studies in humanized mouse models of HIV using anti-gp120 aptamer-siRNA chimeras have demonstrated their potential as specific antiviral agents (Neff *et al.*, 2011); (MacRae *et al.*, 2008); (Zhou and Rossi, 2010). Similarly, siRNAs conjugated to single-chained monoclonal antibody fragments targeted to T-cell or lymphocyte-specific receptors have shown promise *in vivo* (Kim *et al.*, 2010; Kumar *et al.*, 2008). However, these methods are in the earlier stages of development and there are still several delivery and potency hurdles which must be overcome. The focus of this review will be on expressed RNAi modalities, where siRNA duplexes are expressed from plasmid DNA vectors with lasting effects on gene silencing.

#### **3.2 Expressed RNAi constructs**

SiRNAs that are generated from expression cassettes have the advantage of sustained production which makes them suitable for long-term, once-off therapeutic applications. Recent developments in construct design and delivery methods have shown much promise for the advancement of siRNA expression systems against HIV. In earlier siRNA expression studies, linear cassettes were constructed to code for both sense and antisense sequences of the siRNA duplex under the control of separate promoters and termination signals (Lee *et al.*, 2002a); (Miyagishi and Taira, 2002). Expressed siRNAs then associate posttranscriptionally to form a duplex with 2 – 4 nt uridine overhangs. This system was found to be effective against HIV-1 sequences and siRNAs targeting a highly-accessible region of the *rev* transcript were found to inhibit viral transcript expression (Lee *et al.*, 2002a). However, the reliability of correct siRNA duplex association *in vivo* is questionable. Most siRNA expression constructs are now designed to generate mimics of RNAi intermediates in the form of siRNA or miRNA hairpin precursors which are processed by Drosha and/or Dicer enzyme complexes (Figure 2). This is preferable for more reliable siRNA processing in a manner that is regulated by the RNAi pathway, but inappropriate expression levels can lead to saturation of critical RNAi components.

The choice of promoter is therefore critical in achieving suitable levels of construct expression. Polymerase III (pol III) promoter sequences, like human U6 snRNA (small

RNAi-Based Gene Expression Strategies to Combat HIV 201

The choice of stem and loop structures must also be carefully considered in shRNA design. A completely duplexed stem assists in preserving shRNA structure and can be useful for the prevention of 3' – 5' exonuclease attack (Paul *et al.*, 2002), but is not a necessity and the high level of duplex stability may also interfere with strand selection. While typical stem lengths of about 19 bp are effective, longer 29 bp stems can be more potent triggers of RNAi with more effective processing, suggesting that Dicer requires a minimum stem length for efficient cleavage (Siolas *et al.*, 2005). Loop sizes are more variable and can be anywhere between 3 and 9 nts in size. A recent investigation has confirmed that loop sequences are indeed critical in determining shRNA function against HIV-1 sequences (Schopman *et al.*, 2010). ShRNAs with sub-optimal loop sequences (Brummelkamp *et al.*, 2002) can be slightly altered to increase RNA activity by up to 7 fold. The size of optimal loops appears to be between 7 and 10 nts, while decreasing loops to 5 nts or less appears to be detrimental to RNAi activity. Particular loop structures, especially those derived from pri-miRNAs, can enhance processing of weak shRNAs. The importance of loop structure may be attributed to Dicer co-factors, like the KH-type splicing regulatory protein (KSRP), which binds to the

terminal loop and affects processing (Vermeulen *et al.*, 2005); (Trabucchi *et al.*, 2009) .

appropriate promoter and target selection.

**3.2.2 Mimics of microRNA precursors** 

Pol III promoters like U6 or H1 are well suited for the constitutive expression of shRNAs in a range of cell types. The pol III termination signal consists of a short stretch of uridine residues which are cleaved at the termination site after two residues. This is ideal for the generation of a 3' UU overhang in the hairpin, which is important for the efficiency and the specificity of siRNA processing by Dicer. Robust shRNA expression from pol III promoters can be detrimental for a therapeutic application. High levels of sustained expression can lead to cytotoxicity and even to a lethal saturation of the RNAi pathway. The long-term effects of robust shRNA expression were investigated in the livers of adult mice and found to cause liver injury, organ failure and death within one month (Grimm *et al.*, 2006). Morbidity was associated with the downregulation of natural liver miRNAs, which suggested that competition exists for components of the RNAi pathway such as Exportin-5. In a more recent publication, Ago-2 (Slicer) was identified as the primary rate-limiting determinant of both *in vitro* and *in vivo* RNAi efficacy, toxicity, and persistence (Grimm *et al.*, 2010). Ago/shRNA coexpression studies have shown that increased Ago-2 and Exp-5 expression can rescue long-term U6-driven shRNA expression in adult mice with enhanced silencing of exogenous and endogenous hepatic targets, reduced hepatotoxicity, and extended RNAi stability of more than 3 months. The benefits of using a weaker promoter were demonstrated in this study where *in vivo* toxicity was alleviated, allowing for sustained target silencing of over a year. Overall, shRNAs are very potent gene-silencing moieties, but their safe and effective use in anti-HIV gene therapies is dependent on

There seem to be several advantages in creating antiviral constructs with properties that are similar to endogenous miRNA precursors. This includes the incorporation of mismatches into the stem region, the use of longer stems and different terminal loops. Enhanced silencing has been observed for siRNAs derived from hairpins based on precursor–miRNAs (pre-miRNAs). SiRNAs against the HIV-1 *tat* gene were placed into the natural pre-miR-30 backbone and found to be 80% more effective at reducing HIV replication than the same guide expressed from a conventional shRNA (Boden *et al.*, 2004). Hairpins based on primary-microRNAs (pri-miRNAs) with pol II promoters have also been shown to induce

nuclear RNA U6) or H1 (human RNAse P H1), are commonly used to drive efficient expression of short downstream sequences and often feature in short hairpin expression cassettes. They have been well characterised in earlier ribozyme expression studies (Good *et al.*, 1997) and are suitable to drive nuclear expression in a wide range of human cell types (Paul *et al.*, 2002). However, the high level of constitutive expression from pol III promoters can be undesired in a long-term therapeutic treatment. Polymerase II (pol II) promoters, like the human cytomegalovirus (CMV) promoter, are now being favoured for lower and potentially regulatable expression of pri- and pre-mRNA mimics. They allow for safer tissue-specific expression of constructs with tighter *in vivo* regulation (Cullen, 2005); (Giering *et al.*, 2008). Other types of inducible promoters have also been investigated for regulated expression in the presence of an activator molecule (Jacque *et al.*, 2002), which would satisfy the need for greater control of construct expression in therapeutic applications.

#### **3.2.1 Short hairpin RNAs**

Short hairpin RNAs (shRNAs) were developed for the expression of siRNA duplexes (Paul *et al.*, 2002). ShRNAs are essentially mimics of precursor-miRNAs that are processed by Dicer to produce staggered siRNA duplexes. ShRNAs typically have short, completely complementary stem regions of about 19 -29 base pairs (bp), a 2 nt 3' overhang and one of several commonly used terminal loops. Synthetic shRNAs have been shown to trigger more effective gene silencing than siRNA duplexes with the same guide sequences (Siolas *et al.*, 2005). The association of shRNAs with Dicer may result in more effective loading of guide sequences onto RISC, as Dicer forms part of the RISC-loading complex (RLC) (MacRae *et al.*, 2008).

A variety of different guide sequences can be expressed from shRNAs. Sequence composition of individual guides can affect the processing efficiency, but common shRNA formats generally give high levels of expression. An shRNA against the HIV-1 transactivator (Tat) protein gene was incorporated into an H1-driven expression cassette and delivered to cells through the use of a recombinant AAV (adeno-associated virus) DNA vector (Boden *et al.*, 2003). In a cell culture challenge assay with the infectious molecular clone HIV-1NL4.3, HIV-1 p24 antigen levels were decreased by 97% 48 hours post-transfection in cells expressing the sh*tat* compared to control cells. The high mutability of HIV, however, severely hinders the potency of silencing by a single shRNA in a long-term application. In cells stably expressing sh*tat*, HIV-1 replication was reduced by 95% in the first three weeks, but had again risen by day 25 as a result of a nonsynonymous mutation in the targeted region.

The silencing efficacy of an shRNA mostly depends on the level of conservation of the HIV target sequence and shRNAs with equal processing do not necessarily result in the same level of HIV inhibition. ShRNAs against the viral integrase sequence (shIN) and the U3 region of the viral genome required for integration (shU3) showed a more potent inhibitory effect on HIV-1 replication than sh*tat* in shRNA-transduced MT-4 or primary CD4\_ T cells (Nishitsuji *et al.*, 2006). In p24 viral replication assays, sh*int* produced a ~ 4 fold reduction in p24 production, while sh*tat* resulted in a ~2.5 fold reduction four days post-infection. In contrast, a similar hairpin against the U5 region of the viral genome resulted in weak inhibition, possibly due to high GC content. At 10 days postinfection, viral replication was again detected in the shTattransformed MT-4 cells, while HIV-1 replication was undetectable for up to 1 month postinfection, in cells that received shIN or shU3. While this is an improvement, the use of single shRNAs is still unsuitable for long-term HIV suppression.

nuclear RNA U6) or H1 (human RNAse P H1), are commonly used to drive efficient expression of short downstream sequences and often feature in short hairpin expression cassettes. They have been well characterised in earlier ribozyme expression studies (Good *et al.*, 1997) and are suitable to drive nuclear expression in a wide range of human cell types (Paul *et al.*, 2002). However, the high level of constitutive expression from pol III promoters can be undesired in a long-term therapeutic treatment. Polymerase II (pol II) promoters, like the human cytomegalovirus (CMV) promoter, are now being favoured for lower and potentially regulatable expression of pri- and pre-mRNA mimics. They allow for safer tissue-specific expression of constructs with tighter *in vivo* regulation (Cullen, 2005); (Giering *et al.*, 2008). Other types of inducible promoters have also been investigated for regulated expression in the presence of an activator molecule (Jacque *et al.*, 2002), which would satisfy the need for greater

Short hairpin RNAs (shRNAs) were developed for the expression of siRNA duplexes (Paul *et al.*, 2002). ShRNAs are essentially mimics of precursor-miRNAs that are processed by Dicer to produce staggered siRNA duplexes. ShRNAs typically have short, completely complementary stem regions of about 19 -29 base pairs (bp), a 2 nt 3' overhang and one of several commonly used terminal loops. Synthetic shRNAs have been shown to trigger more effective gene silencing than siRNA duplexes with the same guide sequences (Siolas *et al.*, 2005). The association of shRNAs with Dicer may result in more effective loading of guide sequences onto RISC, as Dicer forms part of the RISC-loading complex (RLC) (MacRae *et al.*, 2008). A variety of different guide sequences can be expressed from shRNAs. Sequence composition of individual guides can affect the processing efficiency, but common shRNA formats generally give high levels of expression. An shRNA against the HIV-1 transactivator (Tat) protein gene was incorporated into an H1-driven expression cassette and delivered to cells through the use of a recombinant AAV (adeno-associated virus) DNA vector (Boden *et al.*, 2003). In a cell culture challenge assay with the infectious molecular clone HIV-1NL4.3, HIV-1 p24 antigen levels were decreased by 97% 48 hours post-transfection in cells expressing the sh*tat* compared to control cells. The high mutability of HIV, however, severely hinders the potency of silencing by a single shRNA in a long-term application. In cells stably expressing sh*tat*, HIV-1 replication was reduced by 95% in the first three weeks, but had again risen by day 25 as a result of a nonsynonymous mutation in the targeted

The silencing efficacy of an shRNA mostly depends on the level of conservation of the HIV target sequence and shRNAs with equal processing do not necessarily result in the same level of HIV inhibition. ShRNAs against the viral integrase sequence (shIN) and the U3 region of the viral genome required for integration (shU3) showed a more potent inhibitory effect on HIV-1 replication than sh*tat* in shRNA-transduced MT-4 or primary CD4\_ T cells (Nishitsuji *et al.*, 2006). In p24 viral replication assays, sh*int* produced a ~ 4 fold reduction in p24 production, while sh*tat* resulted in a ~2.5 fold reduction four days post-infection. In contrast, a similar hairpin against the U5 region of the viral genome resulted in weak inhibition, possibly due to high GC content. At 10 days postinfection, viral replication was again detected in the shTattransformed MT-4 cells, while HIV-1 replication was undetectable for up to 1 month postinfection, in cells that received shIN or shU3. While this is an improvement, the use of

single shRNAs is still unsuitable for long-term HIV suppression.

control of construct expression in therapeutic applications.

**3.2.1 Short hairpin RNAs**

region.

The choice of stem and loop structures must also be carefully considered in shRNA design. A completely duplexed stem assists in preserving shRNA structure and can be useful for the prevention of 3' – 5' exonuclease attack (Paul *et al.*, 2002), but is not a necessity and the high level of duplex stability may also interfere with strand selection. While typical stem lengths of about 19 bp are effective, longer 29 bp stems can be more potent triggers of RNAi with more effective processing, suggesting that Dicer requires a minimum stem length for efficient cleavage (Siolas *et al.*, 2005). Loop sizes are more variable and can be anywhere between 3 and 9 nts in size. A recent investigation has confirmed that loop sequences are indeed critical in determining shRNA function against HIV-1 sequences (Schopman *et al.*, 2010). ShRNAs with sub-optimal loop sequences (Brummelkamp *et al.*, 2002) can be slightly altered to increase RNA activity by up to 7 fold. The size of optimal loops appears to be between 7 and 10 nts, while decreasing loops to 5 nts or less appears to be detrimental to RNAi activity. Particular loop structures, especially those derived from pri-miRNAs, can enhance processing of weak shRNAs. The importance of loop structure may be attributed to Dicer co-factors, like the KH-type splicing regulatory protein (KSRP), which binds to the terminal loop and affects processing (Vermeulen *et al.*, 2005); (Trabucchi *et al.*, 2009) .

Pol III promoters like U6 or H1 are well suited for the constitutive expression of shRNAs in a range of cell types. The pol III termination signal consists of a short stretch of uridine residues which are cleaved at the termination site after two residues. This is ideal for the generation of a 3' UU overhang in the hairpin, which is important for the efficiency and the specificity of siRNA processing by Dicer. Robust shRNA expression from pol III promoters can be detrimental for a therapeutic application. High levels of sustained expression can lead to cytotoxicity and even to a lethal saturation of the RNAi pathway. The long-term effects of robust shRNA expression were investigated in the livers of adult mice and found to cause liver injury, organ failure and death within one month (Grimm *et al.*, 2006). Morbidity was associated with the downregulation of natural liver miRNAs, which suggested that competition exists for components of the RNAi pathway such as Exportin-5. In a more recent publication, Ago-2 (Slicer) was identified as the primary rate-limiting determinant of both *in vitro* and *in vivo* RNAi efficacy, toxicity, and persistence (Grimm *et al.*, 2010). Ago/shRNA coexpression studies have shown that increased Ago-2 and Exp-5 expression can rescue long-term U6-driven shRNA expression in adult mice with enhanced silencing of exogenous and endogenous hepatic targets, reduced hepatotoxicity, and extended RNAi stability of more than 3 months. The benefits of using a weaker promoter were demonstrated in this study where *in vivo* toxicity was alleviated, allowing for sustained target silencing of over a year. Overall, shRNAs are very potent gene-silencing moieties, but their safe and effective use in anti-HIV gene therapies is dependent on appropriate promoter and target selection.

#### **3.2.2 Mimics of microRNA precursors**

There seem to be several advantages in creating antiviral constructs with properties that are similar to endogenous miRNA precursors. This includes the incorporation of mismatches into the stem region, the use of longer stems and different terminal loops. Enhanced silencing has been observed for siRNAs derived from hairpins based on precursor–miRNAs (pre-miRNAs). SiRNAs against the HIV-1 *tat* gene were placed into the natural pre-miR-30 backbone and found to be 80% more effective at reducing HIV replication than the same guide expressed from a conventional shRNA (Boden *et al.*, 2004). Hairpins based on primary-microRNAs (pri-miRNAs) with pol II promoters have also been shown to induce

RNAi-Based Gene Expression Strategies to Combat HIV 203

coding sequences can also occur. Viral escape mutations are, however, not limited to point mutations in the siRNA target sequences. Mutations can occur in other regions of the genome that alter the local RNA secondary structure of the target site and diminish siRNA

The key to successful HIV inhibition lies in the targeting of several highly conserved regions simultaneously in a combinatorial approach (co-RNAi). This strategy has been used in a number of conventional drug regimens and aims to reduce the emergence of viral escape mutants by inhibiting multiple HIV targets. Expressed shRNA and shRNA-miR constructs can be adapted to produce multiple siRNAs and combined in single plasmid vectors. RNAi constructs can also be more easily adapted than small molecules in response to viral evolution. Several studies have investigated viral mutation pathways in response to particular therapeutic stimuli in order to identify and block anticipated escape paths. Interestingly, viral escape paths against shRNA therapy differ to those triggered by drug

Multiple short hairpins can be used against HIV simply through the use of multiple vectors or through the design of consecutive shRNA constructs. The simultaneous use of two separate hairpins against the CCR5 and CXCR4 cellular receptors has been shown to protect transduced primary macrophages against HIV infection (Lee *et al.*, 2003). The consistent delivery and expression of two separate shRNAs in an equal ratio is not precise using a simultaneous approach and multiple shRNA constructs are preferable for more controlled expression levels. A bi-specific construct containing a U6-driven shRNA against CXCR4 and an H1-driven shRNA against the CCR5 has been shown to effectively downregulate both targets simultaneously (Anderson and Akkina, 2005). When the siRNA expressing transduced cells were challenged with X4 and R5 tropic HIV-1, they demonstrated marked viral resistance. Targeting of three different HIV regions is even more favourable for effective coRNAi and has been demonstrated using a multi-shRNA construct. Three H1 driven shRNAs against two *pol* and one *gag* sequence were successfully used to create an

additive inhibition of viral production and delay viral escape (ter Brake *et al.*, 2006).

Combining multiple shRNAs with the same construct structure can be problematic. The use of repeated promoter sequences can lead to rearrangements and deletions of whole transcriptional units as a result of recombination in lentiviral delivery vectors. To prevent this, non-identical pol III promoters U6, H1, and 7SK and the polymerase II U1 promoter can be used to drive simultaneous expression in a multi-shRNA cassette which can inhibit HIV without viral escape (ter Brake *et al.*, 2008). However, equivalent expression of each siRNA is not guaranteed and high expression levels of several anti-viral guides still occurs which can lead to even more serious saturation toxicity (McIntyre *et al.*, 2009). ShRNAs, even in a multiple format, are therefore not necessarily the most preferable expression

To avoid issues associated with the toxicity of multiple promoter-driven constructs, several adjacent siRNA sequences can be incorporated into single long hairpin constructs (lhRNAs) under the control of one promoter. Consecutive Dicer cleavage is required

binding (Westerhout *et al.*, 2005).

therapy (Applegate *et al.*, 2010).

**3.3.1 Multiple short hairpin RNAs** 

systems for therapeutic applications.

**3.3.2 Long hairpin RNAs** 

potent, stable and regulatable gene silencing *in vivo*, even when present as a single copy in the genome (Dickins *et al.*, 2005). These pri-miR mimics have been described as secondgeneration shRNAs and termed shRNA-miRs (Silva *et al.*, 2005).

Artificial miRNAs not only show a greater inhibitory efficacy against HIV targets when compared to conventional shRNAs (Liu *et al.*, 2008), but may also be better at suppressing imperfect HIV-1 targets (Liu *et al.*, 2009a). This enhanced silencing ability has been attributed to more efficient processing in the RNAi pathway by both Drosha and Dicer enzyme complexes. Pri-miRNA mimics may also be subjected to regulatory mechanisms and other important components of the RNAi pathway, unlike substrate mimics introduced further on in the pathway (Obernosterer *et al.*, 2006). In addition, there may be functional differences between RISC-siRNA and RISC-miRNA with respect to Ago protein association. As might be implied from nature, it is necessary to maintain several key elements of natural pri-miRNA structures for effective processing of artificial miRNAs. It has been suggested that a large terminal loop (≥ 10 nts), a stem between 26 and 40 bp and at least 40 nts of nonstructured flanking RNA sequences are required for efficient processing by Drosha (Zeng *et al.*, 2005). Single-stranded flanking sequences may form part of the Drosha-RNA interface (Zeng and Cullen, 2005) and it seems logical to preserve the natural flanking sequences in the use of miRNA precursors as scaffolds. Preservation of natural loop sequences also appears to be desirable and has been shown to rescue the inhibitory potential of weakly functioning shRNAs (Schopman *et al.*, 2010).

In comparisons of the silencing ability of shRNA and artificial microRNA constructs with similar guide strands, shRNAs were generally found to produce a more potent silencing effect (Boudreau *et al.*, 2008). This has, however, been attributed to a higher level of expression both *in vivo* and *in vitro.* As already discussed, higher expression of shRNAs is undesirable in a therapeutic setting. Artificial miRNAs with a lower expression are processed more efficiently in the RNAi pathway and cause less of a bottleneck which can lead to saturation toxicity. SiRNAs expressed from a microRNA backbone do not appear to show the same level of inhibitory competition for nuclear export by Exportin-5 and incorporation into RISC (Castanotto *et al.*, 2007). Pri-miRNA mimics therefore appear to be safer option for therapeutic use and show less disruption of natural microRNA biogenesis (Boudreau *et al.*, 2009). Pri-miRNAs, however, do not show consistent processing over a range of different guide sequences, as can be observed for shRNAs. Pri-miRNAs therefore appear to be a more favourable expression format for siRNAs, but sufficient processing of guide sequences must be assessed empirically.

#### **3.3 Combinatorial RNAi constructs**

Despite the potency of RNAi against HIV targets in short-term studies, the sustained inhibition of viral replication is not possible with a single siRNA construct. Viral escape mutations arise readily in response to the strong selective pressure of effective RNAi constructs. SiRNAs directed against the viral *nef* gene and introduced into human T cells by retroviral transduction successfully inhibit viral replication at first, but after several weeks of culture RNAi-resistant viruses developed (Gregory *et al.*, 2004). Viral mutations included nucleotide substitutions or deletions in the Nef gene that modified or deleted the siRNA-Nef target sequence. Similarly, expressed shRNAs targeting the HIV-1 *tat* gene soon give rise to a viral quasispecies harbouring a point mutation in the shRNA target region which abolishes antiviral activity of tat shRNA (Boden *et al.*, 2003). Silent mutations in protein-

potent, stable and regulatable gene silencing *in vivo*, even when present as a single copy in the genome (Dickins *et al.*, 2005). These pri-miR mimics have been described as second-

Artificial miRNAs not only show a greater inhibitory efficacy against HIV targets when compared to conventional shRNAs (Liu *et al.*, 2008), but may also be better at suppressing imperfect HIV-1 targets (Liu *et al.*, 2009a). This enhanced silencing ability has been attributed to more efficient processing in the RNAi pathway by both Drosha and Dicer enzyme complexes. Pri-miRNA mimics may also be subjected to regulatory mechanisms and other important components of the RNAi pathway, unlike substrate mimics introduced further on in the pathway (Obernosterer *et al.*, 2006). In addition, there may be functional differences between RISC-siRNA and RISC-miRNA with respect to Ago protein association. As might be implied from nature, it is necessary to maintain several key elements of natural pri-miRNA structures for effective processing of artificial miRNAs. It has been suggested that a large terminal loop (≥ 10 nts), a stem between 26 and 40 bp and at least 40 nts of nonstructured flanking RNA sequences are required for efficient processing by Drosha (Zeng *et al.*, 2005). Single-stranded flanking sequences may form part of the Drosha-RNA interface (Zeng and Cullen, 2005) and it seems logical to preserve the natural flanking sequences in the use of miRNA precursors as scaffolds. Preservation of natural loop sequences also appears to be desirable and has been shown to rescue the inhibitory potential of weakly

In comparisons of the silencing ability of shRNA and artificial microRNA constructs with similar guide strands, shRNAs were generally found to produce a more potent silencing effect (Boudreau *et al.*, 2008). This has, however, been attributed to a higher level of expression both *in vivo* and *in vitro.* As already discussed, higher expression of shRNAs is undesirable in a therapeutic setting. Artificial miRNAs with a lower expression are processed more efficiently in the RNAi pathway and cause less of a bottleneck which can lead to saturation toxicity. SiRNAs expressed from a microRNA backbone do not appear to show the same level of inhibitory competition for nuclear export by Exportin-5 and incorporation into RISC (Castanotto *et al.*, 2007). Pri-miRNA mimics therefore appear to be safer option for therapeutic use and show less disruption of natural microRNA biogenesis (Boudreau *et al.*, 2009). Pri-miRNAs, however, do not show consistent processing over a range of different guide sequences, as can be observed for shRNAs. Pri-miRNAs therefore appear to be a more favourable expression format for siRNAs, but sufficient processing of

Despite the potency of RNAi against HIV targets in short-term studies, the sustained inhibition of viral replication is not possible with a single siRNA construct. Viral escape mutations arise readily in response to the strong selective pressure of effective RNAi constructs. SiRNAs directed against the viral *nef* gene and introduced into human T cells by retroviral transduction successfully inhibit viral replication at first, but after several weeks of culture RNAi-resistant viruses developed (Gregory *et al.*, 2004). Viral mutations included nucleotide substitutions or deletions in the Nef gene that modified or deleted the siRNA-Nef target sequence. Similarly, expressed shRNAs targeting the HIV-1 *tat* gene soon give rise to a viral quasispecies harbouring a point mutation in the shRNA target region which abolishes antiviral activity of tat shRNA (Boden *et al.*, 2003). Silent mutations in protein-

generation shRNAs and termed shRNA-miRs (Silva *et al.*, 2005).

functioning shRNAs (Schopman *et al.*, 2010).

guide sequences must be assessed empirically.

**3.3 Combinatorial RNAi constructs** 

coding sequences can also occur. Viral escape mutations are, however, not limited to point mutations in the siRNA target sequences. Mutations can occur in other regions of the genome that alter the local RNA secondary structure of the target site and diminish siRNA binding (Westerhout *et al.*, 2005).

The key to successful HIV inhibition lies in the targeting of several highly conserved regions simultaneously in a combinatorial approach (co-RNAi). This strategy has been used in a number of conventional drug regimens and aims to reduce the emergence of viral escape mutants by inhibiting multiple HIV targets. Expressed shRNA and shRNA-miR constructs can be adapted to produce multiple siRNAs and combined in single plasmid vectors. RNAi constructs can also be more easily adapted than small molecules in response to viral evolution. Several studies have investigated viral mutation pathways in response to particular therapeutic stimuli in order to identify and block anticipated escape paths. Interestingly, viral escape paths against shRNA therapy differ to those triggered by drug therapy (Applegate *et al.*, 2010).

#### **3.3.1 Multiple short hairpin RNAs**

Multiple short hairpins can be used against HIV simply through the use of multiple vectors or through the design of consecutive shRNA constructs. The simultaneous use of two separate hairpins against the CCR5 and CXCR4 cellular receptors has been shown to protect transduced primary macrophages against HIV infection (Lee *et al.*, 2003). The consistent delivery and expression of two separate shRNAs in an equal ratio is not precise using a simultaneous approach and multiple shRNA constructs are preferable for more controlled expression levels. A bi-specific construct containing a U6-driven shRNA against CXCR4 and an H1-driven shRNA against the CCR5 has been shown to effectively downregulate both targets simultaneously (Anderson and Akkina, 2005). When the siRNA expressing transduced cells were challenged with X4 and R5 tropic HIV-1, they demonstrated marked viral resistance. Targeting of three different HIV regions is even more favourable for effective coRNAi and has been demonstrated using a multi-shRNA construct. Three H1 driven shRNAs against two *pol* and one *gag* sequence were successfully used to create an additive inhibition of viral production and delay viral escape (ter Brake *et al.*, 2006).

Combining multiple shRNAs with the same construct structure can be problematic. The use of repeated promoter sequences can lead to rearrangements and deletions of whole transcriptional units as a result of recombination in lentiviral delivery vectors. To prevent this, non-identical pol III promoters U6, H1, and 7SK and the polymerase II U1 promoter can be used to drive simultaneous expression in a multi-shRNA cassette which can inhibit HIV without viral escape (ter Brake *et al.*, 2008). However, equivalent expression of each siRNA is not guaranteed and high expression levels of several anti-viral guides still occurs which can lead to even more serious saturation toxicity (McIntyre *et al.*, 2009). ShRNAs, even in a multiple format, are therefore not necessarily the most preferable expression systems for therapeutic applications.

#### **3.3.2 Long hairpin RNAs**

To avoid issues associated with the toxicity of multiple promoter-driven constructs, several adjacent siRNA sequences can be incorporated into single long hairpin constructs (lhRNAs) under the control of one promoter. Consecutive Dicer cleavage is required

RNAi-Based Gene Expression Strategies to Combat HIV 205

(Ely *et al.*, 2009). Different pri-miRNA backbones have different expression aptitudes for individual sequences and the ordering of pri-miRNA expression units can affect both expression and silencing abilities. The preservation of natural pri-miRNA structural elements is still required in multiple constructs with a minimum of 22 nt of natural flanking sequence required at the 5′ arm and at least 15 nt at the 3′ arm (Zeng and Cullen, 2005). In addition to this, extra restriction sequences must often be included to create the tandem format. This can be very useful for creating modular pri-miRNA units that can be exchanged as required. On the other hand, extra artificial and repetitive natural flanking sequences in the expressed transcript can interact in an unexpected fashion to form undesirable secondary structures which prevent or alter processing of the intended guide sequences. Multiple guide sequences can be incorporated into other natural miRNA precursor forms. In an earlier example, the BIC non-coding RNA with its embedded miR-155 miRNA precursor was used as a scaffold for construction of the SIBR vector (Chung *et al.*, 2006). Synthetic miRNA sequences were incorporated into a modified miR-155 stem-loop, along with flanking sequences from the third exon of the BIC transcript, which proved to be sufficient for the expression of miR-155. It was found that two artificial miRNAs could be expressed from a single polycistronic transcript to give effective inhibition of targets without a decrease in the efficacy of individual target suppression. Alternatively, up to 8 tandem copies of the same artificial miRNA can be expressed from the SIBR vector in tandem for enhanced expression, but this is not a favourable option for HIV inhibition where strong

A simpler approach for polycistronic design is to mimic entire naturally occurring polycistronic pri-miRNA units. Multiple effective siRNAs can be inserted into a naturally occurring polycistronic scaffold and expressed from a single promoter sequence. The mir-17-92 polycistron has been successfully used as a scaffold for four siRNAs against *rev/tat*, *gag, pol* and leader HIV sequences (Liu *et al.*, 2008). In this example, each siRNA sequence was initially incorporated into an individual pri-miRNA structure with about 40 nts of flanking sequences and assessed. In doing so, the passenger strand was altered with the use of predictive secondary structure software to maintain all mismatches, bulges and thermodynamic stability as far as possible. Positioning of guide sequences in each primiRNA hairpin was found to be crucial for optimal processing. Individual hairpins showed moderate anti-HIV activity, but co-expression of two or more hairpins in a polycistronic format gave greatly enhanced silencing from each individual pri-miRNA component. Antiviral siRNAs have also been engineered into the tri-cistronic miR-106b cluster (Aagaard *et al.*, 2008) to produce 3 siRNAs against *tat/rev, tat* and *rev.* In both of these examples, polycistronic expression systems appear to have an intrinsic inhibitory activity greater than

In all examples of mimic design, it appears that the preservation of key structural elements is crucial for effective processing and inhibitory function. Although the predictive software for this purpose is of a very high standard, folded structures and sequence interactions *in vivo* can never be guaranteed. This is of particular concern when modular pri-miRNA units are being combined in a novel way. Guide sequence expression from pri-miRNA mimics is also variable and can depend on both the anti-viral sequence and backbone used. Transposition of 19 nt siRNAs from shRNA expression systems into pri-miRNA units can be tricky as miRNA sequences can be up to 24 nts in length. Only one guide can be used per pri-miRNA hairpin, which means that combinatorial constructs will inevitably contain quite a lot of extra, non-guide sequence. Construct size can be a limiting factor for insertion into

silencing of a single target should be avoided.

that of conventional shRNA constructs or individual hairpins.

along the length of the hairpin to release individual siRNA duplexes (Paddison *et al.*, 2002). A modified long hairpin against a 50 nt region of the integrase gene effectively suppressed both wild-type and sh*int*-resistant viral strains (Nishitsuji *et al.*, 2006). A U6 driven long hairpin RNA spanning a possible 60 bp of a 5'LTR target region has shown silencing of respective target sequences and inhibition of HIV replication (Barichievy *et al.*, 2007). The greatest silencing in this format was observed for the target corresponding to the base of the hairpin stem.

In a more pre-meditated approach, several well-characterised shRNA sequences can be concatenated into a single long or extended shRNA (e-shRNA). E-shRNAs were designed with two siRNAs against *nef* and *pol* HIV-1 sequences which were efficiently processed and showed viral inhibition (Liu *et al.*, 2007). The position of the two siRNAs was found to be critical for the generation of functional siRNAs. In a further step, the generation of three siRNAs from a single U6-driven hairpin was investigated against *tat*, *rev* and *vif* (Saayman *et al.*, 2008). All sequences were capable of target silencing depending on their position within the hairpin and processing efficiency decreased from the stem of the hairpin towards the terminal loop. Spacing between the siRNA sequences within the duplex stem region can also affect processing efficiency. E-shRNAs can be extended to include a maximum of 3 siRNAs with an optimal length of 66 bp. Further stem extension results in a loss of RNAi activity (Liu *et al.*, 2009a). A size limit of 80 bp has also been suggested and the incorporation of G:U wobbles may have several advantages related to hairpin expression (Sano *et al.*, 2008).

A further advancement which circumvents the length limitation of lhRNAs is the use of a long hairpin concatenation. A recent study has shown that four functional anti-HIV siRNAs can be derived from a single Pol III-generated transcript comprising two adjacent long hairpin RNA precursors (Saayman *et al.*, 2010). Two active anti-HIV siRNAs were engineered into each of two lhRNAs, which were arranged in tandem to create a double long hairpin (dlhRNA) expression cassette. Each hairpin component was found to generate two of four unique siRNA sequences (*tat, nef*, LTR and *int*) and thereby mediate dualtargeting. Processing of the individual siRNAs was found to be affected by both internal ordering and spacing between siRNAs. An inverse correlation between siRNA silencing potency and increased spacing was observed, while processing at the 3' position of each lhRNA was more variable. Optimal siRNA processing was found to occur when only one mismatched base pair was placed between each siRNA in accordance with predicted Dicer cleavage intervals. Effective multiple processing was achieved by manipulating the order of the siRNA-encoding sequences to create an optimized combinatorial dlhRNA expression cassette. Despite the use of a pol III promoter, expression potency of the individual guides is diluted and therefore less likely to cause toxic saturation. This work has highlighted the versatility of dlhRNAs and shown that they are a promising construct form for effective silencing of multiple HIV targets.

#### **3.3.3 Polycistronic primary microRNA mimics**

Safe and controlled expression of siRNAs is a particular concern in coRNAi. Once again, a logical way of doing this is to mimic mammalian microRNA expression systems. MiRNAs are often expressed as pol II-driven polycistronic units in the cell and multiple siRNAs can be expressed in a similar fashion. In some systems, effective singular miRNA mimics, like those based on the pri-miR30 backbone, were simply incorporated in tandem under the control of a single pol II promoter to express two or three artificial guides (Han *et al.*, 2006)

along the length of the hairpin to release individual siRNA duplexes (Paddison *et al.*, 2002). A modified long hairpin against a 50 nt region of the integrase gene effectively suppressed both wild-type and sh*int*-resistant viral strains (Nishitsuji *et al.*, 2006). A U6 driven long hairpin RNA spanning a possible 60 bp of a 5'LTR target region has shown silencing of respective target sequences and inhibition of HIV replication (Barichievy *et al.*, 2007). The greatest silencing in this format was observed for the target corresponding to

In a more pre-meditated approach, several well-characterised shRNA sequences can be concatenated into a single long or extended shRNA (e-shRNA). E-shRNAs were designed with two siRNAs against *nef* and *pol* HIV-1 sequences which were efficiently processed and showed viral inhibition (Liu *et al.*, 2007). The position of the two siRNAs was found to be critical for the generation of functional siRNAs. In a further step, the generation of three siRNAs from a single U6-driven hairpin was investigated against *tat*, *rev* and *vif* (Saayman *et al.*, 2008). All sequences were capable of target silencing depending on their position within the hairpin and processing efficiency decreased from the stem of the hairpin towards the terminal loop. Spacing between the siRNA sequences within the duplex stem region can also affect processing efficiency. E-shRNAs can be extended to include a maximum of 3 siRNAs with an optimal length of 66 bp. Further stem extension results in a loss of RNAi activity (Liu *et al.*, 2009a). A size limit of 80 bp has also been suggested and the incorporation of G:U wobbles may have several advantages related to hairpin expression (Sano *et al.*,

A further advancement which circumvents the length limitation of lhRNAs is the use of a long hairpin concatenation. A recent study has shown that four functional anti-HIV siRNAs can be derived from a single Pol III-generated transcript comprising two adjacent long hairpin RNA precursors (Saayman *et al.*, 2010). Two active anti-HIV siRNAs were engineered into each of two lhRNAs, which were arranged in tandem to create a double long hairpin (dlhRNA) expression cassette. Each hairpin component was found to generate two of four unique siRNA sequences (*tat, nef*, LTR and *int*) and thereby mediate dualtargeting. Processing of the individual siRNAs was found to be affected by both internal ordering and spacing between siRNAs. An inverse correlation between siRNA silencing potency and increased spacing was observed, while processing at the 3' position of each lhRNA was more variable. Optimal siRNA processing was found to occur when only one mismatched base pair was placed between each siRNA in accordance with predicted Dicer cleavage intervals. Effective multiple processing was achieved by manipulating the order of the siRNA-encoding sequences to create an optimized combinatorial dlhRNA expression cassette. Despite the use of a pol III promoter, expression potency of the individual guides is diluted and therefore less likely to cause toxic saturation. This work has highlighted the versatility of dlhRNAs and shown that they are a promising construct form for effective

Safe and controlled expression of siRNAs is a particular concern in coRNAi. Once again, a logical way of doing this is to mimic mammalian microRNA expression systems. MiRNAs are often expressed as pol II-driven polycistronic units in the cell and multiple siRNAs can be expressed in a similar fashion. In some systems, effective singular miRNA mimics, like those based on the pri-miR30 backbone, were simply incorporated in tandem under the control of a single pol II promoter to express two or three artificial guides (Han *et al.*, 2006)

the base of the hairpin stem.

silencing of multiple HIV targets.

**3.3.3 Polycistronic primary microRNA mimics** 

2008).

(Ely *et al.*, 2009). Different pri-miRNA backbones have different expression aptitudes for individual sequences and the ordering of pri-miRNA expression units can affect both expression and silencing abilities. The preservation of natural pri-miRNA structural elements is still required in multiple constructs with a minimum of 22 nt of natural flanking sequence required at the 5′ arm and at least 15 nt at the 3′ arm (Zeng and Cullen, 2005). In addition to this, extra restriction sequences must often be included to create the tandem format. This can be very useful for creating modular pri-miRNA units that can be exchanged as required. On the other hand, extra artificial and repetitive natural flanking sequences in the expressed transcript can interact in an unexpected fashion to form undesirable secondary structures which prevent or alter processing of the intended guide sequences.

Multiple guide sequences can be incorporated into other natural miRNA precursor forms. In an earlier example, the BIC non-coding RNA with its embedded miR-155 miRNA precursor was used as a scaffold for construction of the SIBR vector (Chung *et al.*, 2006). Synthetic miRNA sequences were incorporated into a modified miR-155 stem-loop, along with flanking sequences from the third exon of the BIC transcript, which proved to be sufficient for the expression of miR-155. It was found that two artificial miRNAs could be expressed from a single polycistronic transcript to give effective inhibition of targets without a decrease in the efficacy of individual target suppression. Alternatively, up to 8 tandem copies of the same artificial miRNA can be expressed from the SIBR vector in tandem for enhanced expression, but this is not a favourable option for HIV inhibition where strong silencing of a single target should be avoided.

A simpler approach for polycistronic design is to mimic entire naturally occurring polycistronic pri-miRNA units. Multiple effective siRNAs can be inserted into a naturally occurring polycistronic scaffold and expressed from a single promoter sequence. The mir-17-92 polycistron has been successfully used as a scaffold for four siRNAs against *rev/tat*, *gag, pol* and leader HIV sequences (Liu *et al.*, 2008). In this example, each siRNA sequence was initially incorporated into an individual pri-miRNA structure with about 40 nts of flanking sequences and assessed. In doing so, the passenger strand was altered with the use of predictive secondary structure software to maintain all mismatches, bulges and thermodynamic stability as far as possible. Positioning of guide sequences in each primiRNA hairpin was found to be crucial for optimal processing. Individual hairpins showed moderate anti-HIV activity, but co-expression of two or more hairpins in a polycistronic format gave greatly enhanced silencing from each individual pri-miRNA component. Antiviral siRNAs have also been engineered into the tri-cistronic miR-106b cluster (Aagaard *et al.*, 2008) to produce 3 siRNAs against *tat/rev, tat* and *rev.* In both of these examples, polycistronic expression systems appear to have an intrinsic inhibitory activity greater than that of conventional shRNA constructs or individual hairpins.

In all examples of mimic design, it appears that the preservation of key structural elements is crucial for effective processing and inhibitory function. Although the predictive software for this purpose is of a very high standard, folded structures and sequence interactions *in vivo* can never be guaranteed. This is of particular concern when modular pri-miRNA units are being combined in a novel way. Guide sequence expression from pri-miRNA mimics is also variable and can depend on both the anti-viral sequence and backbone used. Transposition of 19 nt siRNAs from shRNA expression systems into pri-miRNA units can be tricky as miRNA sequences can be up to 24 nts in length. Only one guide can be used per pri-miRNA hairpin, which means that combinatorial constructs will inevitably contain quite a lot of extra, non-guide sequence. Construct size can be a limiting factor for insertion into

RNAi-Based Gene Expression Strategies to Combat HIV 207

only about 14 % of these rare potential sites correspond to functional siRNA predictions. This finding has called for serious reconsideration of the clinical potential of numerous previous

Highly conserved target sites are essential in a therapeutic application to successfully inhibit mixed and fluctuating viral populations. Ideal target regions of the HIV genome include essential regulatory regions of viral gene expression. Among these are the primer activation signal (PAS), primer binding site (PBS), packaging signal (), central polypurine tract (cPPT), central termination sequence (CTS) and 3' polypurine tract (3' PPT). These regions are conserved at the nucleotide sequence level, presumably to conserve secondary RNA structures which are important for viral fitness. Other highly conserved potential siRNA target sites include the packaging signal, TAR/polyA and regions in protease and integrase protein codes. It does not seem possible for a single siRNA to target all known HIV-strains, but this work by Naito *et. al.* suggests that it is theoretically possible to target >99% of circulating subtype M strains with escape resistance by combining only two siRNAs against

In other previous studies, it has been suggested that four conserved sequences will be sufficient to inhibit several hundred circulating viral stains (Leonard and Schaffer, 2005). It is, however, unlikely that sequences across several viral strains will harbour sufficient sequence identity to be effectively targeted by the same set of four shRNAs. It has since been proposed that more than four shRNAs should be utilised in therapeutic design such that each viral strain will be effectively targeted by at least four shRNAs (McIntyre *et al.*, 2011). ShRNA combinations of seven H1-driven expression units were found to provide up to 87% coverage for all known HI V strains and 100% coverage of clade B subtypes. Position within a specific (1 – 7) multi-shRNA cassette generally had little effect on the suppressive activity of individual shRNAs when expressed in isolation, but when shRNA expression was simultaneous, expression decreased for shRNAs in position 3 – 7. The effective and equal expression of 6 or 7 tandem shRNAs is a challenge and the use of so many pol III promoter units poses a substantially higher risk of toxic saturation of the endogenous RNAi pathway. However, the possibility of inhibiting multiple viral strains simultaneously is a tempting motivation for further development of both multi-shRNA and larger polycistronic mimic expression systems. As an alternative to highly mutable viral sequences, host dependency factors (HDFs) encoded by the cell can also be targeted to further inhibit viral replication. The CD4 receptor required for viral entry is an obvious choice, but is also present on other host cells in which silencing of CD4 may result in undesirable side-effects. The CCR5 and CXCR4 co-receptors are more attractive targets for silencing and have been investigated (Novina *et al.*, 2002, Song *et al.*, 2003). There are however many other host factors involved in HIV replication, such as those required for Tat binding to TAR (cyclin T1 and CDK9) and those that bind to the LTR to control gene expression (NF-κB, SP1, LBP, and LEF). SiRNAs against the NF-κB p65 subunit resulted in decreased viral replication (Surabhi and Gaynor, 2002). Large screen studies have also revealed numerous other potential targets (Brass *et al.*, 2008). Cellular targeting is promising, but must be used with caution as the inhibition of cellular proteins

can have widespread effects on cellular function with undesirable side effects.

A primary concern with the use of RNAi-based strategies is that of safety and the specificity of the inhibitory effect *in vivo*. The presence of double stranded RNA (dsRNA) can activate

**5. Safety & toxicity of RNAi activators** 

studies which generally target regions outside of this highly conserved category.

highly conserved viral sequences.

viral vectors, but polycistronic pri-miRNA units are still generally within an acceptable size range. In general, the use of pri-miRNA mimics requires more planning and testing of individual components, while the final construct behaviour can only really be observed experimentally. This makes polycistronic miRNA expression systems more labour intensive, but thorough testing should be part of any therapeutic strategy. The extra input may be well worthwhile if the potential advantages of combined HIV targeting at an appropriate expression level with regulated and efficient processing can be realised.

## **3.3.4 Therapeutic constructs**

Overall, it appears as if dlhRNAs or polycistronic mimics appear to possess the best combinations of desirable properties for a therapeutic RNAi application. Developments are, however, still required before these constructs can be implemented in a clinical setting. Expression systems can be further optimised to give restricted expression in target cell populations and therefore reduce the risk of unwanted off target effects (OTEs). More specific expression can be achieved through the use of a haematopoietic or T-cell-specific promoter (Liu *et al.*, 2008). The WAS promoter, for example, is active in human hematopoietic precursor cells (CD34+), T lymphocytes, B cells and dendritic cells, but not in non-haematopoietic cells and may be an excellent candidate (Charrier *et al.*, 2007). Expression could ideally be further restricted to HIV infected cells by using the HIV-1 LTR promoter to express the miRNA polycistron only in the presence of the viral Tat protein (Unwalla *et al.*, 2004). Furthermore, RNAi activators are probably best used in combination with other types of RNA- or protein-based anti-HIV constructs in a therapeutic application to mediate an even more potent viral inhibition that does not rely on a single genetic mechanism. A polycistronic miRNA mimic, for example, can be combined with a TAR decoy for enhanced viral inhibition. ShRNAs can also be applied therapeutically in combination with other RNA-based constructs, for example, an anti-CCR5 ribozyme and a TAR decoy for greater protection against from HIV-1 challenge (Wilson *et al.*, 2003). It therefore seems that the best therapeutic approaches involve the use of combinations of both RNAi triggers and different types of inhibitory mechanisms, while maintaining natural RNAi processing and overall cellular function as far as possible.

## **4. RNAi target selection**

A critical factor in the success of any RNAi-inducing therapeutic strategy is the choice of target sequence. Highly effective therapeutic effectors can be rendered ineffective in a clinical setting if careful consideration is not given to the long-term targeting strategy. SiRNAs have been designed against most regions of HIV-encoded RNAs, including *tat*, *rev*, *gag*, *pol*, *nef*, *vif*, *env*, *vpr*, and the long terminal repeat (Figure 2). However, there is no single Achilles heel in the HIV genome and targeting of several highly conserved regions across multiple viral strains is a requirement for a clinically relevant RNAi-based therapy. *In silico* approaches for target identification are therefore crucial, although targeting strategies must still be experimentally validated.

Highly conserved HIV sequences are rare. In an extensive study of siRNA target prediction for optimal design of siRNAs, highly conserved sequences were analysed from the Los Alamos HIV Sequence Database covering 495 divergent strains of subtype M (Naito *et al.*, 2007). Of the 4 million potential 21-mer siRNA target sites, only 5.2 % showed a level of conservation greater than 50%. Highly conserved (> 80%) siRNA target sequences are very rare (< 1 %) and

viral vectors, but polycistronic pri-miRNA units are still generally within an acceptable size range. In general, the use of pri-miRNA mimics requires more planning and testing of individual components, while the final construct behaviour can only really be observed experimentally. This makes polycistronic miRNA expression systems more labour intensive, but thorough testing should be part of any therapeutic strategy. The extra input may be well worthwhile if the potential advantages of combined HIV targeting at an appropriate

Overall, it appears as if dlhRNAs or polycistronic mimics appear to possess the best combinations of desirable properties for a therapeutic RNAi application. Developments are, however, still required before these constructs can be implemented in a clinical setting. Expression systems can be further optimised to give restricted expression in target cell populations and therefore reduce the risk of unwanted off target effects (OTEs). More specific expression can be achieved through the use of a haematopoietic or T-cell-specific promoter (Liu *et al.*, 2008). The WAS promoter, for example, is active in human hematopoietic precursor cells (CD34+), T lymphocytes, B cells and dendritic cells, but not in non-haematopoietic cells and may be an excellent candidate (Charrier *et al.*, 2007). Expression could ideally be further restricted to HIV infected cells by using the HIV-1 LTR promoter to express the miRNA polycistron only in the presence of the viral Tat protein (Unwalla *et al.*, 2004). Furthermore, RNAi activators are probably best used in combination with other types of RNA- or protein-based anti-HIV constructs in a therapeutic application to mediate an even more potent viral inhibition that does not rely on a single genetic mechanism. A polycistronic miRNA mimic, for example, can be combined with a TAR decoy for enhanced viral inhibition. ShRNAs can also be applied therapeutically in combination with other RNA-based constructs, for example, an anti-CCR5 ribozyme and a TAR decoy for greater protection against from HIV-1 challenge (Wilson *et al.*, 2003). It therefore seems that the best therapeutic approaches involve the use of combinations of both RNAi triggers and different types of inhibitory mechanisms, while maintaining natural

A critical factor in the success of any RNAi-inducing therapeutic strategy is the choice of target sequence. Highly effective therapeutic effectors can be rendered ineffective in a clinical setting if careful consideration is not given to the long-term targeting strategy. SiRNAs have been designed against most regions of HIV-encoded RNAs, including *tat*, *rev*, *gag*, *pol*, *nef*, *vif*, *env*, *vpr*, and the long terminal repeat (Figure 2). However, there is no single Achilles heel in the HIV genome and targeting of several highly conserved regions across multiple viral strains is a requirement for a clinically relevant RNAi-based therapy. *In silico* approaches for target identification are therefore crucial, although targeting strategies must

Highly conserved HIV sequences are rare. In an extensive study of siRNA target prediction for optimal design of siRNAs, highly conserved sequences were analysed from the Los Alamos HIV Sequence Database covering 495 divergent strains of subtype M (Naito *et al.*, 2007). Of the 4 million potential 21-mer siRNA target sites, only 5.2 % showed a level of conservation greater than 50%. Highly conserved (> 80%) siRNA target sequences are very rare (< 1 %) and

expression level with regulated and efficient processing can be realised.

RNAi processing and overall cellular function as far as possible.

**3.3.4 Therapeutic constructs** 

**4. RNAi target selection** 

still be experimentally validated.

only about 14 % of these rare potential sites correspond to functional siRNA predictions. This finding has called for serious reconsideration of the clinical potential of numerous previous studies which generally target regions outside of this highly conserved category.

Highly conserved target sites are essential in a therapeutic application to successfully inhibit mixed and fluctuating viral populations. Ideal target regions of the HIV genome include essential regulatory regions of viral gene expression. Among these are the primer activation signal (PAS), primer binding site (PBS), packaging signal (), central polypurine tract (cPPT), central termination sequence (CTS) and 3' polypurine tract (3' PPT). These regions are conserved at the nucleotide sequence level, presumably to conserve secondary RNA structures which are important for viral fitness. Other highly conserved potential siRNA target sites include the packaging signal, TAR/polyA and regions in protease and integrase protein codes. It does not seem possible for a single siRNA to target all known HIV-strains, but this work by Naito *et. al.* suggests that it is theoretically possible to target >99% of circulating subtype M strains with escape resistance by combining only two siRNAs against highly conserved viral sequences.

In other previous studies, it has been suggested that four conserved sequences will be sufficient to inhibit several hundred circulating viral stains (Leonard and Schaffer, 2005). It is, however, unlikely that sequences across several viral strains will harbour sufficient sequence identity to be effectively targeted by the same set of four shRNAs. It has since been proposed that more than four shRNAs should be utilised in therapeutic design such that each viral strain will be effectively targeted by at least four shRNAs (McIntyre *et al.*, 2011). ShRNA combinations of seven H1-driven expression units were found to provide up to 87% coverage for all known HI V strains and 100% coverage of clade B subtypes. Position within a specific (1 – 7) multi-shRNA cassette generally had little effect on the suppressive activity of individual shRNAs when expressed in isolation, but when shRNA expression was simultaneous, expression decreased for shRNAs in position 3 – 7. The effective and equal expression of 6 or 7 tandem shRNAs is a challenge and the use of so many pol III promoter units poses a substantially higher risk of toxic saturation of the endogenous RNAi pathway. However, the possibility of inhibiting multiple viral strains simultaneously is a tempting motivation for further development of both multi-shRNA and larger polycistronic mimic expression systems. As an alternative to highly mutable viral sequences, host dependency factors (HDFs) encoded by the cell can also be targeted to further inhibit viral replication. The CD4 receptor required for viral entry is an obvious choice, but is also present on other host cells in which silencing of CD4 may result in undesirable side-effects. The CCR5 and CXCR4 co-receptors are more attractive targets for silencing and have been investigated (Novina *et al.*, 2002, Song *et al.*, 2003). There are however many other host factors involved in HIV replication, such as those required for Tat binding to TAR (cyclin T1 and CDK9) and those that bind to the LTR to control gene expression (NF-κB, SP1, LBP, and LEF). SiRNAs against the NF-κB p65 subunit resulted in decreased viral replication (Surabhi and Gaynor, 2002). Large screen studies have also revealed numerous other potential targets (Brass *et al.*, 2008). Cellular targeting is promising, but must be used with caution as the inhibition of cellular proteins can have widespread effects on cellular function with undesirable side effects.

#### **5. Safety & toxicity of RNAi activators**

A primary concern with the use of RNAi-based strategies is that of safety and the specificity of the inhibitory effect *in vivo*. The presence of double stranded RNA (dsRNA) can activate

RNAi-Based Gene Expression Strategies to Combat HIV 209

Cell-based delivery appears to be the most promising approach for the development of a realistic therapeutic strategy (Figure 3). Essentially, haematopoietic stem cells (HSCs) are collected from suitable donor candidates and transduced *ex vivo* with anti-HIV expression constructs. Lentiviral vectors are preferable for this purpose as they can mediate integration of therapeutic constructs into the cellular genome even in non-dividing cells for long-term construct expression (Naldini *et al.*, 1996). Transduced HSCs are then infused into an HIVinfected patient where they can give rise to HIV-resistant cell populations. This method allows for controlled transduction of target cell populations where aberrant integrative

Cell-based delivery may involve allogeneic or autologous cell transplantation. In a pivotal allogeneic study, replicating HIV remained undetected in a recipient patient 20 months after transplantation and termination of HAART (Hutter *et al.*, 2009). In this case, HSCs were obtained from a donor homozygous for a naturally occurring HIV-resistant phenotype and successfully transplanted into an HIV-infected patient with acute myeloid leukemia following myeloablative therapy. Cells with the Δ32CCR5 mutation harbour a 32 bp deletion in the gene for chemokine receptor 5 and are protected from infection by R5-tropic viral subtypes. This strategy proved to be effective for treating leukaemia and preventing viral replication. Notorious HIV-reservoirs, like the intestinal lamina propria, remained HIV free 159 days after transplantation. While this approach seems ideal for effective HIV inhibition *in vivo*, homozygous donors for the CCR5 mutation are rare, occurring in only ~ 1% of the white population. Nonetheless, this promising strategy can be adapted for the

Autologous cell–based approaches do not require matching of HLA genotypes and avoid the host-versus-graft complications. In a recent RNA-based example, a ribozyme (OZ1) against the *tat-vpr* region of the HIV genome was delivered to patients through transduced autologous CD34+ hematopoietic progenitor cells (Mitsuyasu, 2009). Progenitor cells were transduced with a murine retroviral vector encoding the OZ1 ribozyme. The cell-delivery method was assessed through a randomised, double-blind, placebo-controlled phase 2 gene transfer clinical trial with 74 HIV-1 infected individuals. The OZ1 group showed significantly lower viral loads after 40 weeks and significantly higher CD4+ lymphocyte counts through 100 weeks. This study demonstrated that cell-delivered gene transfer can be

In a more recent clinical trial, three RNA-based anti-HIV constructs were introduced into patients undergoing transplantation for AIDs-related lymphoma (DiGiusto *et al.*, 2010). HIVinfected individuals represent a unique and ethically-sound research group where marrow ablation can be performed prior to transplantation. A *tat/rev* short hairpin, TAR decoy and CCR5 ribozyme combination construct was used to modify the patient's own CD34+ cells through lentiviral transduction. Transduced cells showed no difference in haematopoietic potential compared to non-transduced cells in *in vitro* analysis and were successfully engrafted in four patients. Expression of the anti-HIV moieties was initially as high as 22 %, but declined to ~1% over four weeks of cell culture. Persistent siRNA expression was observed at low levels for up to 24 months in multiple lineages. No short-term toxicity was associated with the infusion of the genetically modified cells, and observed toxicities were instead related to the standard autologous hematopoietic cell transplantation (HCT)

events may be detected and eliminated *ex vivo*.

delivery of HSCs with an artificial HIV-resistant genotype.

both a safe and effective therapeutic strategy.

procedure.

cellular defence mechanisms which lead to a non-specific halt in translation and cell death. DsRNA induces an interferon type 1 (IFN-1) response in the cell which in turn activates the transcription of other immune effector molecules, IFN stimulated genes (ISGs) and Dicerrelated pathways (de Veer *et al.*, 2005); (Karpala *et al.*, 2005). DsRNA can also activate the retinoic-acid inducible gene-I (RIG-I) and members of the oligoadenylate synthetase (OAS) receptor family which catalyze the synthesis of 2'-5' oligoadenylates to activate a latent cellular endoribonuclease (RNASEL), which in turn cleaves cellular and viral mRNAs. A key effector molecule is the dsRNA-responsive protein kinase receptor (PKR) which functions to block translation of both viral and cellular proteins. PKR activation is typically induced by long dsRNA molecules (>30 nts), but can also be induced by exogenously introduced short 19–29 nt dsRNAs. SiRNAs and shRNAs can induce an IFN response in cells through toll-like receptors (TLRs), particularly TLR3 (Kariko *et al.*, 2004). Ironically, these immune responses play a role in viral defense systems of the cell, but can create issues for the introduction of artificial anti-viral constructs. However, siRNA sequences tend to be only weak inducers of the IFN response and the use of siRNA expression systems can be effective in avoiding an immune response (Robbins *et al.*, 2006).

Another major concern with RNAi activators is the unintentional suppression of cellular transcripts with partial sequence complementation described as off-target effects (OTEs). As the seed region (position 2-7 nt) is the most crucial determinant of target specificity, it seems probable that several cellular transcripts will be susceptible to such a short region of sequence complementation. Some microarray studies have shown that even targets with one or two base pair mismatches with an siRNA can be affected (Jackson *et al.*, 2003). The use of multiple guide sequences is required for effective long-term viral inhibition, but this increases the number of potential OTEs. In a number of studies, cellular toxicity has not been observed, suggesting that OTEs may not necessarily have a significant impact on cellular function (Liu *et al.*, 2008). Nonetheless, extensive attempts should be made to predict potential OTEs before clinical application.

Saturation of the endogenous RNAi pathway with highly expressed RNAi-activators, like pol III-driven shRNAs, can have potentially lethal toxic side effects (Grimm *et al.*, 2006). As already mentioned, polycistronic pri-miRNA mimics can be useful in avoiding competition with the components of the pathway through reduced expression levels and more regulated processing of guide sequences. The natural properties of these mimics may also be useful in avoiding immune stimulation and careful target selection may reduce unwanted OTEs. Although, the complete reduction of OTEs is unlikely and extensive screening of *in vivo* cell expression patterns may be the only real way to assess the extent of undesirable effects.

## **6. RNAi towards the clinic**

A number of RNAi-based effectors have reached clinical trials, but safe and effective expression and delivery of RNAi constructs remains an obstacle for most therapeutic approaches. Recent developments have shown much promise in addressing common delivery issues. Novel nanotechnologies have been used for the delivery of exogenous siRNAs (Davis *et al.*, 2010), while lentiviral vectors that are stably transduced with an extended hairpin expression cassette have been shown to durably inhibit HIV-1 in T-cells (Liu *et al.*, 2009b). Finer details of lentiviral optimisation are now being elucidated and it seems that unique strategies are required for shRNA and miRNA expressing vectors (Schopman *et al.*, 2010).

cellular defence mechanisms which lead to a non-specific halt in translation and cell death. DsRNA induces an interferon type 1 (IFN-1) response in the cell which in turn activates the transcription of other immune effector molecules, IFN stimulated genes (ISGs) and Dicerrelated pathways (de Veer *et al.*, 2005); (Karpala *et al.*, 2005). DsRNA can also activate the retinoic-acid inducible gene-I (RIG-I) and members of the oligoadenylate synthetase (OAS) receptor family which catalyze the synthesis of 2'-5' oligoadenylates to activate a latent cellular endoribonuclease (RNASEL), which in turn cleaves cellular and viral mRNAs. A key effector molecule is the dsRNA-responsive protein kinase receptor (PKR) which functions to block translation of both viral and cellular proteins. PKR activation is typically induced by long dsRNA molecules (>30 nts), but can also be induced by exogenously introduced short 19–29 nt dsRNAs. SiRNAs and shRNAs can induce an IFN response in cells through toll-like receptors (TLRs), particularly TLR3 (Kariko *et al.*, 2004). Ironically, these immune responses play a role in viral defense systems of the cell, but can create issues for the introduction of artificial anti-viral constructs. However, siRNA sequences tend to be only weak inducers of the IFN response and the use of siRNA expression systems can be

Another major concern with RNAi activators is the unintentional suppression of cellular transcripts with partial sequence complementation described as off-target effects (OTEs). As the seed region (position 2-7 nt) is the most crucial determinant of target specificity, it seems probable that several cellular transcripts will be susceptible to such a short region of sequence complementation. Some microarray studies have shown that even targets with one or two base pair mismatches with an siRNA can be affected (Jackson *et al.*, 2003). The use of multiple guide sequences is required for effective long-term viral inhibition, but this increases the number of potential OTEs. In a number of studies, cellular toxicity has not been observed, suggesting that OTEs may not necessarily have a significant impact on cellular function (Liu *et al.*, 2008). Nonetheless, extensive attempts should be made to predict

Saturation of the endogenous RNAi pathway with highly expressed RNAi-activators, like pol III-driven shRNAs, can have potentially lethal toxic side effects (Grimm *et al.*, 2006). As already mentioned, polycistronic pri-miRNA mimics can be useful in avoiding competition with the components of the pathway through reduced expression levels and more regulated processing of guide sequences. The natural properties of these mimics may also be useful in avoiding immune stimulation and careful target selection may reduce unwanted OTEs. Although, the complete reduction of OTEs is unlikely and extensive screening of *in vivo* cell expression patterns may be the only real way to assess the extent of undesirable effects.

A number of RNAi-based effectors have reached clinical trials, but safe and effective expression and delivery of RNAi constructs remains an obstacle for most therapeutic approaches. Recent developments have shown much promise in addressing common delivery issues. Novel nanotechnologies have been used for the delivery of exogenous siRNAs (Davis *et al.*, 2010), while lentiviral vectors that are stably transduced with an extended hairpin expression cassette have been shown to durably inhibit HIV-1 in T-cells (Liu *et al.*, 2009b). Finer details of lentiviral optimisation are now being elucidated and it seems that unique strategies are required for shRNA and miRNA expressing vectors

effective in avoiding an immune response (Robbins *et al.*, 2006).

potential OTEs before clinical application.

**6. RNAi towards the clinic** 

(Schopman *et al.*, 2010).

Cell-based delivery appears to be the most promising approach for the development of a realistic therapeutic strategy (Figure 3). Essentially, haematopoietic stem cells (HSCs) are collected from suitable donor candidates and transduced *ex vivo* with anti-HIV expression constructs. Lentiviral vectors are preferable for this purpose as they can mediate integration of therapeutic constructs into the cellular genome even in non-dividing cells for long-term construct expression (Naldini *et al.*, 1996). Transduced HSCs are then infused into an HIVinfected patient where they can give rise to HIV-resistant cell populations. This method allows for controlled transduction of target cell populations where aberrant integrative events may be detected and eliminated *ex vivo*.

Cell-based delivery may involve allogeneic or autologous cell transplantation. In a pivotal allogeneic study, replicating HIV remained undetected in a recipient patient 20 months after transplantation and termination of HAART (Hutter *et al.*, 2009). In this case, HSCs were obtained from a donor homozygous for a naturally occurring HIV-resistant phenotype and successfully transplanted into an HIV-infected patient with acute myeloid leukemia following myeloablative therapy. Cells with the Δ32CCR5 mutation harbour a 32 bp deletion in the gene for chemokine receptor 5 and are protected from infection by R5-tropic viral subtypes. This strategy proved to be effective for treating leukaemia and preventing viral replication. Notorious HIV-reservoirs, like the intestinal lamina propria, remained HIV free 159 days after transplantation. While this approach seems ideal for effective HIV inhibition *in vivo*, homozygous donors for the CCR5 mutation are rare, occurring in only ~ 1% of the white population. Nonetheless, this promising strategy can be adapted for the delivery of HSCs with an artificial HIV-resistant genotype.

Autologous cell–based approaches do not require matching of HLA genotypes and avoid the host-versus-graft complications. In a recent RNA-based example, a ribozyme (OZ1) against the *tat-vpr* region of the HIV genome was delivered to patients through transduced autologous CD34+ hematopoietic progenitor cells (Mitsuyasu, 2009). Progenitor cells were transduced with a murine retroviral vector encoding the OZ1 ribozyme. The cell-delivery method was assessed through a randomised, double-blind, placebo-controlled phase 2 gene transfer clinical trial with 74 HIV-1 infected individuals. The OZ1 group showed significantly lower viral loads after 40 weeks and significantly higher CD4+ lymphocyte counts through 100 weeks. This study demonstrated that cell-delivered gene transfer can be both a safe and effective therapeutic strategy.

In a more recent clinical trial, three RNA-based anti-HIV constructs were introduced into patients undergoing transplantation for AIDs-related lymphoma (DiGiusto *et al.*, 2010). HIVinfected individuals represent a unique and ethically-sound research group where marrow ablation can be performed prior to transplantation. A *tat/rev* short hairpin, TAR decoy and CCR5 ribozyme combination construct was used to modify the patient's own CD34+ cells through lentiviral transduction. Transduced cells showed no difference in haematopoietic potential compared to non-transduced cells in *in vitro* analysis and were successfully engrafted in four patients. Expression of the anti-HIV moieties was initially as high as 22 %, but declined to ~1% over four weeks of cell culture. Persistent siRNA expression was observed at low levels for up to 24 months in multiple lineages. No short-term toxicity was associated with the infusion of the genetically modified cells, and observed toxicities were instead related to the standard autologous hematopoietic cell transplantation (HCT) procedure.

RNAi-Based Gene Expression Strategies to Combat HIV 211

A genetic approach using expressed RNAi modalities offers the possibility of a once-off treatment against HIV with permanent and sustained viral inhibition and without common toxic side effects associated with current drug regimens. In this chapter, we have discussed the pros and cons of several RNAi-inducers that can be used to inhibit HIV replication. To summarise, an ideal RNAi-based gene therapy against HIV will make use of a combination of effective siRNA sequences in a single expression vector against at least four, but preferably six or seven, highly conserved viral target sequences or host dependency factors. This will provide potent silencing of target sequences across all known viral strains and prevent the emergence of viral escape mutants. Each siRNA sequence needs to be expressed at an equivalent and appropriate level under the control of a regulatable or HIV-inducible pol II promoter to avoid biased targeting and prevent the toxic and potentially lethal competition-based saturation of the natural RNAi pathway. The use of polycistronic primiRNA mimic expression systems appears most favourable and the preservation of natural structural motifs appears to enhance processing and silencing capabilities, as well as avoid activation of the innate immune system, which may otherwise occur with the introduction of exogenous constructs. Off-target effects should be modelled as far as possible before therapeutic testing and should be limited to non-significant effects. The most suitable delivery method to date appears to be through autologous cell-based gene transfer transplantation in a myeloablated recipient background with *ex vivo* lentiviral transduction of the patient's own haematopoietic progenitor cells. This is quite a comprehensive list of desirable properties for an ideal RNAi-based therapy against HIV and is the result of intensive research over the past decade. Further developments are necessary for the realisation of a safe and effective genetic therapy against HIV, but in light of the research

This work was supported by the South African National Research Foundation, Medical Research Council and the Poliomyelitis Research Foundation. FvdB is a recipient of a

Aagaard, L. A., Zhang, J., von Eije, K. J., Li, H., Sætrom, P., Amarzguioui, M., and Rossi, J. J.

Anderson, J., and Akkina, R. (2005). HIV-1 resistance conferred by siRNA cosuppression of

Applegate, T. L., Birkett, D. J., McIntyre, G. J., Jaramillo, A. B., Symonds, G., and Murray, J.

(2008). Engineering and optimization of the miR-106b cluster for ectopic expression

CXCR4 and CCR5 coreceptors by a bispecific lentiviral vector. *AIDS Res Ther 2*, 1-

M. (2010). In silico modeling indicates the development of HIV-1 resistance to multiple shRNA gene therapy differs to standard antiretroviral therapy.

scholarship from the Mellon Postgraduate Mentoring Programme.

of multiplexed anti-HIV RNAs. *Gene Therapy*, 1 - 14.

**7. Conclusion** 

presented here, we are moving closer.

**8. Acknowledgments** 

**9. References**

12.

*Retrovirology 7*, 83.

Fig. 3. An overall scheme for an RNAi-based therapy against HIV. SiRNAs or expressed RNAi constructs are delivered to infected cells by direct or vector-mediated methods

This study has demonstrated a viable approach for effective therapeutic expression of RNAi-based constructs against HIV-1. Persistent, constitutive shRNA expression over 24 months was not found to be toxic to peripheral blood cells and there was no evidence for lineage-specific toxicity. An artificial, anti-HIV expression vector can therefore be stably expressed in human blood cells without significant toxic side effects. The method of *ex vivo* lentiviral transduction and autologous cell-based gene transfer has been demonstrated as a safe and effective means of construct expression. The long term inhibitory effect on viral replication and evolution in the absence of anti-retroviral drugs remains to be seen. The demonstration of sustained anti-viral siRNA expression, however, has moved us one step closer to the realisation of a clinically applicable once-off treatment against HIV-1 infection. Further improvements in transduction and transplantation procedures are likely to yield even more favourable results for the therapeutic application of RNAi.

#### **7. Conclusion**

210 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Fig. 3. An overall scheme for an RNAi-based therapy against HIV. SiRNAs or expressed RNAi constructs are delivered to infected cells by direct or vector-mediated methods

even more favourable results for the therapeutic application of RNAi.

This study has demonstrated a viable approach for effective therapeutic expression of RNAi-based constructs against HIV-1. Persistent, constitutive shRNA expression over 24 months was not found to be toxic to peripheral blood cells and there was no evidence for lineage-specific toxicity. An artificial, anti-HIV expression vector can therefore be stably expressed in human blood cells without significant toxic side effects. The method of *ex vivo* lentiviral transduction and autologous cell-based gene transfer has been demonstrated as a safe and effective means of construct expression. The long term inhibitory effect on viral replication and evolution in the absence of anti-retroviral drugs remains to be seen. The demonstration of sustained anti-viral siRNA expression, however, has moved us one step closer to the realisation of a clinically applicable once-off treatment against HIV-1 infection. Further improvements in transduction and transplantation procedures are likely to yield A genetic approach using expressed RNAi modalities offers the possibility of a once-off treatment against HIV with permanent and sustained viral inhibition and without common toxic side effects associated with current drug regimens. In this chapter, we have discussed the pros and cons of several RNAi-inducers that can be used to inhibit HIV replication. To summarise, an ideal RNAi-based gene therapy against HIV will make use of a combination of effective siRNA sequences in a single expression vector against at least four, but preferably six or seven, highly conserved viral target sequences or host dependency factors. This will provide potent silencing of target sequences across all known viral strains and prevent the emergence of viral escape mutants. Each siRNA sequence needs to be expressed at an equivalent and appropriate level under the control of a regulatable or HIV-inducible pol II promoter to avoid biased targeting and prevent the toxic and potentially lethal competition-based saturation of the natural RNAi pathway. The use of polycistronic primiRNA mimic expression systems appears most favourable and the preservation of natural structural motifs appears to enhance processing and silencing capabilities, as well as avoid activation of the innate immune system, which may otherwise occur with the introduction of exogenous constructs. Off-target effects should be modelled as far as possible before therapeutic testing and should be limited to non-significant effects. The most suitable delivery method to date appears to be through autologous cell-based gene transfer transplantation in a myeloablated recipient background with *ex vivo* lentiviral transduction of the patient's own haematopoietic progenitor cells. This is quite a comprehensive list of desirable properties for an ideal RNAi-based therapy against HIV and is the result of intensive research over the past decade. Further developments are necessary for the realisation of a safe and effective genetic therapy against HIV, but in light of the research presented here, we are moving closer.

#### **8. Acknowledgments**

This work was supported by the South African National Research Foundation, Medical Research Council and the Poliomyelitis Research Foundation. FvdB is a recipient of a scholarship from the Mellon Postgraduate Mentoring Programme.

### **9. References**


RNAi-Based Gene Expression Strategies to Combat HIV 213

Chu, C.-Y., and Rana, T. M. (2008). Potent RNAi by short RNA triggers. *RNA 14*, 1714-

Chung, K. H., Hart, C. C., Al-Bassam, S., Avery, A., Taylor, J., Patel, P. D., Vojtek, A. B., and

Coburn, G. A., and Cullen, B. R. (2002). Potent and specific inhibition of human

Cullen, B. R. (2004). Transcription and Processing of Human microRNA Precursors.

Cullen, B. R. (2005). Human immunodeficiency virus: nuclear RNA export unwound. *Nature* 

Davis, M. E., Zuckerman, J. E., Choi, C. H., Seligson, D., Tolcher, A., Alabi, C. A., Yen, Y.,

Devincenzo, J., Lambkin-Williams, R., Wilkinson, T., Cehelsky, J., Nochur, S., Walsh, E.,

Dickins, R. A., Hemann, M. T., Zilfou, J. T., Simpson, D. R., Ibarra, I., Hannon, G. J., and

DiGiusto, D. L., Krishnan, A., Li, L., Li, H., Li, S., Rao, A., Mi, S., Yam, P., Stinson, S., Kalos,

Ding, X. C., Weiler, J., and Grosshans, H. (2009). Regulating the regulators: mechanisms controlling the maturation of microRNAs. *Trends Biotechnol 27*, 27-36. Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., and Tuschl, T. (2001a).

Elbashir, S. M., Lendeckel, W., and Tuschl, T. (2001b). RNA interference is mediated by 21-

Ely, A., Naidoo, T., and Arbuthnot, P. (2009). Efficient silencing of gene expression with

Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E., and Mello, C. C. (1998).

Giering, J. C., Grimm, D., Storm, T. A., and Kay, M. A. (2008). Expression of shRNA from a

and 22-nucleotide RNAs. *Genes & Development 15*, 188-200.

administered siRNA via targeted nanoparticles. *Nature 464*, 1067-1070. de Veer, M. J., Sledz, C. A., and Williams, B. R. (2005). Detection of foreign RNA:

implications for RNAi. *Immunol Cell Biol 83*, 224-228.

syncytial virus. *Proc Natl Acad Sci U S A*.

mammalian cells. *Nature 411*, 494-498.

microRNA precursors. *Nat Genet 37*, 1289-1295.

interference based on BIC/miR-155. *Nucleic Acids Res 34*, e53.

Turner, D. L. (2006). Polycistronic RNA polymerase II expression vectors for RNA

immunodeficiency virus type 1 replication by RNA interference. *J Virol 76*, 9225-

Heidel, J. D., and Ribas, A. (2010). Evidence of RNAi in humans from systemically

Meyers, R., Gollob, J., and Vaishnaw, A. (2010). A randomized, double-blind, placebo-controlled study of an RNAi-based therapy directed against respiratory

Lowe, S. W. (2005). Probing tumor phenotypes using stable and regulated synthetic

M.*, et al.* (2010). RNA-based gene therapy for HIV with lentiviral vector-modified CD34(+) cells in patients undergoing transplantation for AIDS-related lymphoma.

Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured

modular trimeric Pol II expression cassettes comprising microRNA shuttles. *Nucleic* 

Potent and specific genetic interference by double-strandedRNAin Caenorhabditis

tissue-specific pol II promoter is an effective and safe RNAi therapeutic. *Mol Ther* 

1719.

9231.

*433*, 26-27.

*Molecular Cell 16*, 861-865.

*Sci Transl Med 2*, 36ra43.

*Acids Res 37*, e91.

*16*, 1630-1636.

elegans. *Nature 391*, 806-811.


Barichievy, S., Saayman, S., Von Eije, K. J., Morris, K. V., Arbuthnot, P., and Weinberg, M. S.

Bartel, D., P. (2004). MicroRNAs: Genomics, Biogenesis, Mechanism, and Function. *Cell 116*,

Boden, D., Pusch, O., Lee, F., Tucker, L., and Ramratnam, B. (2003). Human

Boden, D., Pusch, O., Silbermann, R., Lee, F., Tucker, L., and Ramratnam, B. (2004).

Boudreau, R. L., Martins, I., and Davidson, B. L. (2009). Artificial MicroRNAs as siRNA

Boudreau, R. L., Monteys, A. M., and Davidson, B. L. (2008). Minimizing variables

Brass, A. L., Dykxhoorn, D. M., Benita, Y., Yan, N., Engelman, A., Xavier, R. J., Lieberman, J.,

Brennecke, J., Stark, A., Russell, R. B., and Cohen, S. M. (2005). Principles of microRNA-

Breving, K., and Esquela-Kerscher, A. (2009). The complexities of microRNA regulation:

Brummelkamp, T. R., Bernards, R., and Agami, R. (2002). A system for stable expression of

Caplen, N. J., Parrish, S., Imani, F., Fire, A., and Morgan, R. A. (2001). Specific inhibition of

Capodici, J., Kariko, K., and Weissman, D. (2002). Inhibition of HIV-1 infection by small interfering RNA-mediated RNA interference*. J Immunol 169*, 5196-5201. Castanotto, D., Sakurai, K., Lingeman, R., Li, H., Shively, L., Aagaard, L., Soifer, H.,

Charrier, S., Dupre, L., Scaramuzza, S., Jeanson-Leh, L., Blundell, M. P., Danos, O., Cattaneo,

Chendrimada, T. P., Gregory, R. I., Kumaraswamy, E., N., J., Cooch, N., Nishikura, K., and

short interfering RNAs in mammalian cells. *Science 296*, 550-553.

systems. *Proc Natl Acad Sci U S A 98*, 9742-9747.

through a functional genomic screen. *Science 319*, 921-926.

*Oligonucleotides 17*, 419-431.

hairpins. *Nucleic Acids Research 32*, 1154-1158.

target recognition. *PLoS Biol 3*, e85.

*Nucleic Acids Research 35*, 5154-5164.

from WAS patients. *Gene Ther 14*, 415-428.

processing and gene silencing. *Nature 436*, 740-744.

281-297.

11535.

*Therapy 17*, 169-175.

1834-1844.

*Biology.*

(2007). The Inhibitory Efficacy of RNA POL III-Expressed Long Hairpin RNAs Targeted to Untranslated Regions of the HIV-1 5\_ Long Terminal Repeat.

immunodeficiency virus type 1 escape from RNA interference. *J Virol 77*, 11531-

Enhanced gene silencing of HIV-1 specific siRNA using microRNA designed

Shuttles: Improved Safety as Compared to shRNAs In vitro and In vivo. *Molecular* 

among hairpin-based RNAi vectors reveals the potency of shRNAs. *RNA 14*,

and Elledge, S. J. (2008). Identification of host proteins required for HIV infection

mirandering around the rules. *The International Journal of Biochemistry & Cell* 

gene expression by small double-stranded RNAs in invertebrate and vertebrate

Gatignol, A., Riggs, A., and Rossi, J. J. (2007). Combinatorial delivery of small interfering RNAs reduces RNAi efficacy by selective incorporation into RISC.

F., Aiuti, A., Eckenberg, R., Thrasher, A. J.*, et al.* (2007). Lentiviral vectors targeting WASp expression to hematopoietic cells, efficiently transduce and correct cells

Shiekhatter, R. (2005). TRBP recruits the Dicer complex to AGO2 for microRNA


RNAi-Based Gene Expression Strategies to Combat HIV 215

Lee, N. S., Dohjima, T., Bauer, G., Li, H., Li, M.-J., Ehsani, A., Salvaterra, P., and Rossi, J.

Lee, Y., Hur, I., Park, S. Y., Kim, Y. K., Suh, M. R., and Kim, V. N. (2006). The role of PACT in

Lee, Y., Jeon, K., Lee, J.-T., Kim, S., and Kim, V. N. (2002b). MicroRNA maturation: stepwise processing and subcellular localization. *The EMBO Journal 21*, 4663-4670. Lee, Y., Kim, M., Han, J., Yeom, K. H., Lee, S., Baek, S. H., and Kim, V. N. (2004).

Leonard, J. N., and Schaffer, D. V. (2005). Computational design of antiviral RNA

Liu, Y. P., Gruber, J., Haasnoot, J., Konstantinova, P., and Berkhout, B. (2009a). RNAi-

Liu, Y. P., Haasnoot, J., and Berkhout, B. (2007). Design of extended short hairpin RNAs for

Liu, Y. P., Haasnoot, J., ter Brake, O., Berkhout, B., and Konstantinova, P. (2008). Inhibition

Liu, Y. P., von Eije, K. J., Schopman, N. C., Westerink, J. T., ter Brake, O., Haasnoot, J., and

Lund, E., Guttinger, S., Calado, A., Dahlberg, J. E., and Kutay, U. (2004). Nuclear export of

MacRae, I. J., Ma, E., Zhou, M., Robinson, C. V., and Doudna, J. A. (2008). In vitro reconstitution of the human RISC-loading complex. *PNAS 105*, 512-517. Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., and Tuschl, T. (2002a). Sinlge-

Martinez, M. A., Clotet, B., and Este, J. A. (2002b). RNA interference of HIV replication.

McIntyre, G. J., Groneman, J. L., Yu, Y. H., Jaramillo, A., Shen, S., and Applegate, T. L.

McIntyre, G. J., Groneman, J. L., Yu, Y. H., Tran, A., and Applegate, T. L. (2011). Multiple

Mitsuyasu, e. a. (2009). Phase 2 gene therapy trials of an anti-HIV ribozyme in autologous

transcripts in human cells. *Nature Biotechnology 19*, 500-505.

the RNA silencing pathway. *The EMBO Journal 25*, 522-532.

*77*, 11964-11972.

4051-4060.

1645-1654.

574.

*6*, 55.

*Ther 8*, 1.

*Nucleic Acids Res 37*, 6194-6204.

*Nulceic Acids Research 36*, 2811-2824.

hairpin RNAs. *Mol Ther 17*, 1712-1723.

*Trends Immunol 23*, 559-561.

CD34+ cells. *Nature medicine 15*.

microRNA precursors. *Science 303*, 95-98.

HIV-1 inhibition. *Nucleic Acids Res 35*, 5683-5693.

or CCR5-specific small interfering RNAs expressed from a lentivirus vector*. J Virol* 

(2002a). Expression of small interfering RNAs targeted against HIV-1 rev

MicroRNA genes are transcribed by RNA polymerase II. *The EMBO Journal 23*,

interference strategies that resist human immunodeficiency virus escape. *J Virol 79*,

mediated inhibition of HIV-1 by targeting partially complementary viral sequences.

of HIV-1 by multiple siRNAs expressed from a single microRNA polycistron.

Berkhout, B. (2009b). Combinatorial RNAi against HIV-1 using extended short

stranded antisense siRNAs guide target RNA cleavage in RNAi. Cell *110*, 563-

(2009). 96 shRNAs designed for maximal coverage of HIV-1 variants. *Retrovirology* 

shRNA combinations for near-complete coverage of all HIV-1 strains. *AIDS Res* 


Good, P. D., Krikos, A. J., Li, S. X., Bertrand, E., Lee, N. S., Giver, L., Ellington, A., Zaia, J. A.,

Gregory, R. I., Yan, K.-p., Amuthan, G., Chendrimada, T., Doratotaj, B., Cooch, N., and

Grimm, D., Streetz, K. L., Jopling, C. L., Storm, T. A., Pandey, K., Davis, C. R., Marion, P.,

Grimm, D., Wang, L., Lee, J. S., Schurmann, N., Gu, S., Borner, K., Storm, T. A., and Kay, M.

Han, J., Lee, Y., Yeom, K.-H., Kim, Y. K., Jin, H., and Kim, V. N. (2004). The Drosha-

Han, J., Lee, Y., Yeom, K.-H., Nam, J.-W., Heo, I., Rhee, J.-K., Sohn, S. Y., Cho, Y., Zhang, B.-

Hutter, G., Nowak, D., Mossner, M., Ganepola, S., Mussig, A., Allers, K., Schneider, T.,

Jackson, A. L., Bartz, S. R., Schelter, J., Kobayashi, S. V., Burchard, J., Mao, M., Li, B., Cavet,

Jacque, J. M., Triques, K., and Stevenson, M. (2002). Modulation of HIV-1 replication by

Kariko, K., Bhuyan, P., Capodici, J., and Weissman, D. (2004). Small interfering RNAs

Karpala, A. J., Doran, T. J., and Bean, A. G. (2005). Immune responses to dsRNA: implications for gene silencing technologies. *Immunol Cell Biol 83*, 211-216. Khvorova, A., Reynolds, A., and Jayasena, S. D. (2003). Functional siRNAs and miRNAs

Kim, S. S., Peer, D., Kumar, P., Subramanya, S., Wu, H., Asthana, D., Habiro, K., Yang, Y. G.,

Kumar, P., Ban, H. S., Kim, S. S., Wu, H., Pearson, T., Greiner, D. L., Laouar, A., Yao, J.,

Lee, M. T., Coburn, G. A., McClure, M. O., and Cullen, B. R. (2003). Inhibition of human

signaling through toll-like receptor 3. *J Immunol 172*, 6545-6549.

Delta32/Delta32 stem-cell transplantation. *N Engl J Med 360*, 692-698. Hutvagner, G., and Zamore, P. D. (2002). A microRNA in a multiple-turnover RNAi enzyme

microRNA/short hairpin RNA pathways. *Nature 441*, 537-541.

persistence in the adult mouse liver. *J Clin Invest 120*, 3106-3119.

microRNAs by the Drosha-DGCR8 Complex. *Cell 125*, 887-901.

human cell nuclei. *Gene Ther 4*, 45-54.

microRNAs. *Nature 432*.

complex. *Science 297*, 2056-2060.

by RNAi. *Nat Biotechnol 21*, 635-637.

RNA interference. *Nature 418*, 435-438.

exhibit strand bias. *Cell 115*, 209-216.

1 infection in humanized mice*. Cell 134*, 577-586.

*Ther 18*, 370-376.

3027.

Rossi, J. J., and Engelke, D. R. (1997). Expression of small, therapeutic RNAs in

Shiekhattar, R. (2004). The Microprocessor complex mediates the genesis of

Salazar, F., and Kay, M. A. (2006). Fatality in mice due to oversaturation of cellular

A. (2010). Argonaute proteins are key determinants of RNAi efficacy, toxicity, and

DGCR8 complex in primary miRNA processing. *Genes & Development 18*, 3016-

T., and Kim, V. N. (2006). Molecular Basis for the Recognition of Primary

Hofmann, J., Kucherer, C., Blau, O.*, et al.* (2009). Long-term control of HIV by CCR5

G., and Linsley, P. S. (2003). Expression profiling reveals off-target gene regulation

mediate sequence-independent gene suppression and induce immune activation by

Manjunath, N., Shimaoka, M., and Shankar, P. (2010). RNAi-mediated CCR5 silencing by LFA-1-targeted nanoparticles prevents HIV infection in BLT mice. *Mol* 

Haridas, V., Habiro, K.*, et al.* (2008). T cell-specific siRNA delivery suppresses HIV-

immunodeficiency virus type 1 replication in primary macrophages by using Tat-

or CCR5-specific small interfering RNAs expressed from a lentivirus vector*. J Virol 77*, 11964-11972.


RNAi-Based Gene Expression Strategies to Combat HIV 217

Silva, J. M., Li, M. Z., Chang, K., Ge, W., C., G. M., Rickles, R. J., Siolas, D., Hu, G., Paddison,

Siolas, D., Lerner, C., Burchard, J., Ge, W., Linsley, P. S., Paddison, P. J., Hannon, G. J., and

Song, E., Lee, S. K., Dykxhoorn, D. M., Novina, C., Zhang, D., Crawford, K., Cerny, J., Sharp,

Surabhi, R. M., and Gaynor, R. B. (2002). RNA interference directed against viral and cellular

ter Brake, O., Konstantinova, P., Ceylan, M., and Berkhout, B. (2006). Silencing of HIV-1

ter Brake, O., t Hooft, K., Liu, Y. P., Centlivre, M., von Eije, K. J., and Berkhout, B. (2008).

Trabucchi, M., Briata, P., Garcia-Mayoral, M., Haase, A. D., Filipowicz, W., Ramos, A.,

Tuschl, T., and Borkhardt, A. (2002). Small interfering RNAs: a revolutionary tool for the analysis of gene function and gene therapy. *Molecular Interventions 2*, 158- 167.

Unwalla, H. J., Li, M. J., Kim, J. D., Li, H. T., Ehsani, A., Alluin, J., and Rossi, J. J. (2004).

Vermeulen, A., Behlen, L., Reynolds, A., Wolfson, A., Marshall, W. S., Karpilow, J., and

Westerhout, E. M., Ooms, M., Vink, M., Das, A. T., and Berkhout, B. (2005). HIV-1 can

Wilson, J. A., Jayasena, S., Khvorova, A., Sabatinos, S., Rodrigue-Gervais, I. G., Arya, S.,

Yi, R., Qin, Y., Macara, I. G., and Cullen, B. R. (2003). Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes & *Development 17*, 3011-3016. Zeng, Y., and Cullen, B. R. (2005). Efficient Processing of Primary microRNA Hairpins by

Zeng, Y., Yi, R., and Cullen, B. R. (2003). MicroRNAs and small interfering RNAs can inhibit

mRNA expression by similar mechanisms. *PNAS 100*, 9779-9784.

the biogenesis of a subset of microRNAs. *Nature 459*, 1010-1014.

UNAIDS (2010). UN Report on the Global AIDS Epidemic, pp. 1 - 364.

propagated in human liver cells. *PNAS 100*, 2783-2788.

mouse and human genomes. *Nature Genetics 37*, 1281-1288.

primary macrophages*. J Virol 77*, 7174-7181.

*Biotechnology 23*, 227-231.

inhibition. *Mol Ther 16*, 557-564.

*Biotechnol 22*, 1573-1578.

efficiency. *RNA 11*, 674-682.

*Chemistry 280*, 27595-27603.

genome. *Nucleic Acids Res 33*, 796-804.

12963-12973.

892.

P. J., Schlabach, M. R.*, et al.* (2005). Second-generation shRNA libraries covering the

Cleary, M. A. (2005). Synthetic shRNAs as potent RNAi triggers. *Nature* 

P. A., Lieberman, J., Manjunath, N., and Shankar, P. (2003). Sustained small interfering RNA-mediated human immunodeficiency virus type 1 inhibition in

targets inhibits human immunodeficiency Virus Type 1 replication*. J Virol 76*,

with RNA Interference: A Multiple shRNA Approach*. Molecular Therapy 14*, 883-

Lentiviral vector design for multiple shRNA expression and durable HIV-1

Gherzi, R., and Rosenfeld, M. G. (2009). The RNA-binding protein KSRP promotes

Negative feedback inhibition of HIV-1 by TAT-inducible expression of siRNA. *Nat* 

Khvorova, A. (2005). The contributions of dsRNA structure to Dicer specificity and

escape from RNA interference by evolving an alternative structure in its RNA

Sarangi, F., Harris-Brandts, M., Beaulieu, S., and Richardson, C. D. (2003). RNA interference blocks gene expression and RNA synthesis from hepatitis C replicons

Drosha Requires Flanking Nonstructured RNA Sequences. *The Journal of Biological* 


Miyagishi, M., and Taira, K. (2002). U6 promoter-driven siRNAs with four uridine 3'

Naito, Y., Nohtomi, K., Onogi, T., Uenishi, R., Ui-Tei, K., Saigo, K., and Takebe, Y. (2007).

Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M., and

Neff, C. P., Zhou, J., Remling, L., Kuruvilla, J., Zhang, J., Li, H., Smith, D. D., Swiderski, P.,

Nishitsuji, H., Kohara, M., Kannagi, M., and Masuda, T. (2006). Effective Suppression of

Novina, C. D., Murray, M. F., Dykxhoorn, D. M., Beresford, P. J., Riess, J., Lee, S. K.,

Obernosterer, G., Leuschner, P. J., Alenius, M., and Martinez, J. (2006). Post-transcriptional

Paddison, P. J., Caudy, A. A., and Hannon, G. J. (2002). Stable suppression of gene expression by RNAi in mammalian cells. *Proc Natl Acad Sci U S A 99*, 1443-1448. Paul, C. P., Good, P. D., Winer, I., and Engelke, D. R. (2002). Effective expression of small

Robbins, M. A., Li, M., Leung, I., Li, H., Boyer, D. V., Song, Y., Behlke, M. A., and Rossi, J.

Saayman, S., Arbuthnot, P., and Weinberg, M. S. (2010). Deriving four functional anti-HIV

Saayman, S., Barichievy, S., Capovilla, A., Morris, K. V., Arbuthnot, P., and Weinberg, M. S.

Single U6 RNA Pol III-Expressed Long Hairpin RNA. *PLoS ONE 3*, e2602. Sano, M., Li, H., Nakanishi, M., and Rossi, J. J. (2008). Expression of long anti-HIV-1 hairpin

Schopman, N. C., Liu, Y. P., Konstantinova, P., ter Brake, O., and Berkhout, B. (2010).

Schwarz, D. S., Hutvagner, G., Du, T., Xu, Z., A., N., and Zamore, P. D. (2003). Asymmetry

in the assembly of the RNAi enzyme complex. *Cell 115*, 199-208.

J. (2006). Stable expression of shRNAs in human CD34+ progenitor cells can avoid induction of interferon responses to siRNAs in vitro. *Nat Biotechnol 24*, 566-

siRNAs from a single Pol III-generated transcript comprising two adjacent long

(2008). The Efficacy of Generating Three Independent Anti-HIV-1 siRNAs from a

RNAs for the generation of multiple siRNAs: advantages and limitations. *Mol Ther* 

Optimization of shRNA inhibitors by variation of the terminal loop sequence.

*Biotechnol 20*, 497-500.

*Transl Med 3*, 66ra66.

*Virology 80*, 7658-7666.

by a lentiviral vector. *Science 272*, 263-267.

inhibition of HIV-1 infection. *Nat Med 8*, 681-686.

regulation of microRNA expression. *Rna 12*, 1161-1167.

interfering RNA in human cells. *Nat Biotechnol 20*, 505-508.

hairpin RNA precursors. *Nucleic Acids Res 38*, 6652-6663.

*4*, 80.

571.

*16*, 170-177.

*Antiviral Res 86*, 204-211.

overhangs efficiently suppress targeted gene expression in mammalian cells*. Nat* 

Optimal design and validation of antiviral siRNA for targeting HIV-1. *Retrovirology* 

Trono, D. (1996). In vivo gene delivery and stable transduction of nondividing cells

Rossi, J. J., and Akkina, R. (2011). An aptamer-siRNA chimera suppresses HIV-1 viral loads and protects from helper CD4(+) T cell decline in humanized mice. *Sci* 

Human Immunodeficiency Virus Type 1 through a Combination of Short- or Long-Hairpin RNAs Targeting Essential Sequences for Retroviral Integration. *Journal of* 

Collman, R. G., Lieberman, J., Shankar, P., and Sharp, P. A. (2002). siRNA-directed


Zeng, Y., Yi, R., and Cullen, B. R. (2005). Recognition and cleavage of primary microRNA precursors by the nuclear processing enzyme Drosha. *The EMBO Journal 24*, 138- 148.

**0**

**11**

*San Luis Potosí, S.L.P.*

<sup>1</sup>*Mexico* <sup>2</sup>*Brasil*

**Exponential Equilibria and Uniform**

*Programa de Mestrado em Engenharia Industrial, Minas Gerais*

E. Castellanos-Velasco1 and J. Santos-Ferreira2

**Boundedness of HIV Infection Model**

<sup>1</sup>*Facultad de Ciencias, Electrónica, Universidad Autónoma de San Luis Potosí,*

<sup>2</sup>*Centro Universitário do Leste de Minas Gerais, Campus Coronel Fabriciano-M.G.,*

According to data and statistics in a global summary of the Acquired immune deficiency syndrome (AIDS) epidemic from The World Health Organization (WHO), by the end of 2007 an estimated 33 million people worldwide were living with human immunodeficiency virus, HIV. That same year, some 2 million died of AIDS, and the number of people receiving antiretroviral therapy (ART) was reported in 2.990.000, while an estimated of about 9.700.000 the people needing ART. In other words, globally, less than one person in five at risk of HIV has access to basic HIV prevention services. The same study indicates a total 31% as the ART

Highly Active Antiretroviral Therapy (HAART) has demonstrated to be effective at slowing the progression of (HIV) infection to Acquired immune deficiency syndrome (AIDS) and, subsequently, to improve quality of life for infected people. However, if on the one hand the cocktail of drugs has been making possible to extend patient's lives, on the other hand the many problems associated with it and its high cost, particularly to poor people are a clear indication that new approaches to address the situation are needed. Most efforts to control HIV replication has been focused on developing and optimizing antiretroviral therapies. The immune system of human beings contains different types of cells that help protect the body from infections. One of these types of specialized cells are called Cluster of Differentiation Antigen 4 (CD4) or T-cells, by the fact that CD4 is a glycoprotein predominantly found on the surface of helper T cells. The Human Immunodeficiency Virus (HIV) is a retrovirus and therefore it needs cells from a host so that it can make copies of itself. The CD4 cells are receptors for HIV and they aid the virus to initiate its replication process by enabling it to enter into its host. HIV is essentially considered as an infection of the immune in the sense that this virus infects and damages CD4 during the virus replication process. The more virus is produced by infected cells, the higher is the viral load and consequently, lower will be the number of functioning CD4 cells. When this number of uninfected cells declines

below a critical value, the immune system is seriously deteriorated by HIV.

In section 2, it is shown the theoretical reference used to analyze the asymptotic behavior of the solution to the nonlinear perturbed system. The analysis concerning the origin, x=0 as

**1. Introduction**

coverage at that same period (WHO, 2007).

Zhou, J., and Rossi, J. J. (2010). Aptamer-targeted cell-specific RNA interference. *Silence 1*, 4.

## **Exponential Equilibria and Uniform Boundedness of HIV Infection Model**

E. Castellanos-Velasco1 and J. Santos-Ferreira2

<sup>1</sup>*Facultad de Ciencias, Electrónica, Universidad Autónoma de San Luis Potosí, San Luis Potosí, S.L.P.* <sup>2</sup>*Centro Universitário do Leste de Minas Gerais, Campus Coronel Fabriciano-M.G., Programa de Mestrado em Engenharia Industrial, Minas Gerais* <sup>1</sup>*Mexico*

<sup>2</sup>*Brasil*

#### **1. Introduction**

218 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Zeng, Y., Yi, R., and Cullen, B. R. (2005). Recognition and cleavage of primary microRNA

Zhou, J., and Rossi, J. J. (2010). Aptamer-targeted cell-specific RNA interference. *Silence 1*,

148.

4.

precursors by the nuclear processing enzyme Drosha. *The EMBO Journal 24*, 138-

According to data and statistics in a global summary of the Acquired immune deficiency syndrome (AIDS) epidemic from The World Health Organization (WHO), by the end of 2007 an estimated 33 million people worldwide were living with human immunodeficiency virus, HIV. That same year, some 2 million died of AIDS, and the number of people receiving antiretroviral therapy (ART) was reported in 2.990.000, while an estimated of about 9.700.000 the people needing ART. In other words, globally, less than one person in five at risk of HIV has access to basic HIV prevention services. The same study indicates a total 31% as the ART coverage at that same period (WHO, 2007).

Highly Active Antiretroviral Therapy (HAART) has demonstrated to be effective at slowing the progression of (HIV) infection to Acquired immune deficiency syndrome (AIDS) and, subsequently, to improve quality of life for infected people. However, if on the one hand the cocktail of drugs has been making possible to extend patient's lives, on the other hand the many problems associated with it and its high cost, particularly to poor people are a clear indication that new approaches to address the situation are needed. Most efforts to control HIV replication has been focused on developing and optimizing antiretroviral therapies.

The immune system of human beings contains different types of cells that help protect the body from infections. One of these types of specialized cells are called Cluster of Differentiation Antigen 4 (CD4) or T-cells, by the fact that CD4 is a glycoprotein predominantly found on the surface of helper T cells. The Human Immunodeficiency Virus (HIV) is a retrovirus and therefore it needs cells from a host so that it can make copies of itself. The CD4 cells are receptors for HIV and they aid the virus to initiate its replication process by enabling it to enter into its host. HIV is essentially considered as an infection of the immune in the sense that this virus infects and damages CD4 during the virus replication process. The more virus is produced by infected cells, the higher is the viral load and consequently, lower will be the number of functioning CD4 cells. When this number of uninfected cells declines below a critical value, the immune system is seriously deteriorated by HIV.

In section 2, it is shown the theoretical reference used to analyze the asymptotic behavior of the solution to the nonlinear perturbed system. The analysis concerning the origin, x=0 as a stable equilibrium point. On the other hand, the functions that represent the perturbation have the nonlinear dynamic and the function that force the localization of equilibrium point. That function allows to characterize the behavior of the trajectory around origin, x=0. In section 3 are analyzed the properties of the equilibrium in the origin that corresponds to the infected state and the asymptotic behavior of the solution to the model of 3 EDO presented by (Barao & Lemos, 2007; Perelson & Nelson, 1999; Santos & Middleton, 2008). The model is used to characterize the dynamic infection of the disease. In the last section are made some conclusions about exponential equilibrium and uniform boundedness of the model solution.

#### **2. Perturbed system**

Consider the following perturbed system

$$
\dot{\mathbf{x}} = \mathbf{h}(t, \mathbf{x}),
\tag{1}
$$

**Definition 1.** *The equilibrium point x* = 0 *of (3) is*

*function on a neighbourhood D of x* = 0*, such that*

*Proof.* Given *ε >* 0, choose *r* ∈ (0,*ε*] such that

Let *<sup>α</sup>* <sup>=</sup> *min*�*x*�<sup>=</sup>*rV*(*x*). Then, *<sup>α</sup> <sup>&</sup>gt;* 0 by (5). Take *<sup>β</sup>* <sup>∈</sup> (0, *<sup>α</sup>*), and let

*Then, x* = 0 *is stable. Moreover, if*

*Then x* = 0 *is asymptotically stable.*

This follows from (5) since

and *V*(0) = 0, there is *δ >* 0 such that

*• Unstable, if not stable.*

stability sense.

*• Stable, if for each ε >* 0*, there is δ* = *δ*(*ε*) ≥ 0 *such that*

*• Asymptotically stable if it is stable and δ >* 0 *can be chosen such that*

*V*˙ (*x*) =

*n* ∑ *i*=1

�*x*(0)� *< δ* ⇒ �*x*(*t*)� *< ε*, ∀*t* ≥ 0.

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 221

*<sup>t</sup>*→<sup>∞</sup> *<sup>x</sup>*(*t*) = 0.

*∂V ∂xi*

*hi*(*x*). (4)

*<sup>V</sup>*˙ (*x*) <sup>≤</sup> <sup>0</sup> *in D*. (5)

*<sup>V</sup>*˙ (*x*) *<sup>&</sup>lt;* <sup>0</sup> *in D* − {0}. (6)

*n* ∑ *i*=1

�*x*(0)� *< δ* ⇒ lim

*∂V ∂xi x*˙*<sup>i</sup>* =

Let *<sup>V</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup>* be a differentiable function defined in a domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup><sup>n</sup>* that contains the origin. The derivative of *V*(*x*) along the trajectories of (3), denoted by *V*˙ (*x*), is given by

The function *V*˙ (*x*) is dependent on the system's equation. Hence, if *V*˙ (*x*) is negative, *V*(*x*) will decrease along the solution of (3). The following lemma (Khalil, 2002) states Lyapunov's

**Lemma 1.** *Let x* = 0 *be an equilibrium point for (3). Let V* : *D* → *R be a continuously differentiable*

*V*(0) = 0 *and V*(*x*) *>* 0 *in D* − {0},

*Br* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*|�*x*� ≤ *<sup>r</sup>*} <sup>⊂</sup> *<sup>D</sup>*.

Ω*<sup>β</sup>* = {*x* ∈ *Br*|*V*(*x*) ≤ *β*} .

The set Ω*<sup>β</sup>* has the property that any trajectory starting in Ω*<sup>β</sup>* at *t* = 0, stays in Ω*<sup>β</sup>* for all *t* ≥ 0.

*<sup>V</sup>*˙ (*x*(*t*)) <sup>≤</sup> <sup>0</sup> <sup>⇒</sup> *<sup>V</sup>*(*x*(*t*)) <sup>≤</sup> *<sup>V</sup>*(*x*(0)) <sup>≤</sup> *<sup>β</sup>*, <sup>∀</sup>*<sup>t</sup>* <sup>≥</sup> 0.

Since Ω*<sup>β</sup>* is a compact set (is closed and bounded since it is contained in *Br*), the system in (3) has a unique solution defined for all *t* ≥ 0, whenever *x*(0) ∈ Ω*β*. Since *V*(*x*) is continuous

�*x*� ≤ *δ* ⇒ *V*(*x*) *< β*.

where *<sup>h</sup>* : [0, <sup>∞</sup>) <sup>×</sup> *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* is continuous and *<sup>t</sup>* is locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, and *D* is an open connected set that contains the origin *x* = 0. Now, consider the right-hand side of (1), then by adding and subtracting *f*(*x*) known as the nominal system around the origin, we can rewrite the right-hand side as

$$h(t, \mathbf{x}) = f(\mathbf{x}) + [h(t, \mathbf{x}) - f(\mathbf{x})]\_{\mathbf{x}}$$

and define

$$g(\mathbf{x}) + d(t) = h(t, \mathbf{x}) - f(\mathbf{x})\_\prime$$

Hence, the perturbed system (1) can be written as

$$\begin{aligned} \dot{\mathbf{x}} &= h(t, \mathbf{x}) = f(\mathbf{x}) + \mathbf{g}(\mathbf{x}) + d(t) \\ \dot{\mathbf{x}} &= e(\mathbf{x}) + d(t), \end{aligned} \tag{2}$$

where *<sup>f</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* and *<sup>g</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* are locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, *<sup>d</sup>*(*t*) is a uniformly bounded disturbance that satisfies |*d*(*t*)| ≤ *δ* for all *t* ≥ 0 and *e*(*x*) = *f*(*x*) + *g*(*x*). The nominal system in *f*(*x*) could have a stable or asymptotically stable equilibrium point at the origin. The approach of the Lyapunov method will allow us to draw conclusions about the system when the nominal system is perturbed, whether such perturbation is an autonomous or a non autonomous perturbation respectively.

#### **2.1 Nonlinear autonomous perturbation**

Consider the autonomous system

$$
\dot{\mathfrak{x}} = h(\mathfrak{x}),
\tag{3}
$$

where *<sup>h</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* is locally Lipschitz map from a domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*.

Suppose *x*˜ ∈ *D* is an equilibrium point of (3), that is *h*(*x*˜) = 0. The aim is to characterize the stability for the case when the equilibrium point is at the origin, that is *x*˜ = 0. For autonomous system, there is a convergence of the trajectory to a set, the same as the asymptotic stability of the origin. A major concern in analysing the stability of dynamical system is the robustness of various stability properties to uncertainties in the system's model. In the following, it is introduced the stability definition.

**Definition 1.** *The equilibrium point x* = 0 *of (3) is*

*• Stable, if for each ε >* 0*, there is δ* = *δ*(*ε*) ≥ 0 *such that*

$$\|\|\mathbf{x}(\mathbf{0})\|\| < \delta \Rightarrow \|\|\mathbf{x}(t)\|\| < \varepsilon, \ \forall t \ge \mathbf{0}.$$

*• Unstable, if not stable.*

2 Will-be-set-by-IN-TECH

a stable equilibrium point. On the other hand, the functions that represent the perturbation have the nonlinear dynamic and the function that force the localization of equilibrium point. That function allows to characterize the behavior of the trajectory around origin, x=0. In section 3 are analyzed the properties of the equilibrium in the origin that corresponds to the infected state and the asymptotic behavior of the solution to the model of 3 EDO presented by (Barao & Lemos, 2007; Perelson & Nelson, 1999; Santos & Middleton, 2008). The model is used to characterize the dynamic infection of the disease. In the last section are made some conclusions about exponential equilibrium and uniform boundedness of the model solution.

where *<sup>h</sup>* : [0, <sup>∞</sup>) <sup>×</sup> *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* is continuous and *<sup>t</sup>* is locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, and *D* is an open connected set that contains the origin *x* = 0. Now, consider the right-hand side of (1), then by adding and subtracting *f*(*x*) known as the nominal system around the

*h*(*t*, *x*) = *f*(*x*)+[*h*(*t*, *x*) − *f*(*x*)],

*g*(*x*) + *d*(*t*) = *h*(*t*, *x*) − *f*(*x*),

*x*˙ = *h*(*t*, *x*) = *f*(*x*) + *g*(*x*) + *d*(*t*)

where *<sup>f</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* and *<sup>g</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* are locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, *<sup>d</sup>*(*t*) is a uniformly bounded disturbance that satisfies |*d*(*t*)| ≤ *δ* for all *t* ≥ 0 and *e*(*x*) = *f*(*x*) + *g*(*x*). The nominal system in *f*(*x*) could have a stable or asymptotically stable equilibrium point at the origin. The approach of the Lyapunov method will allow us to draw conclusions about the system when the nominal system is perturbed, whether such perturbation is an autonomous

Suppose *x*˜ ∈ *D* is an equilibrium point of (3), that is *h*(*x*˜) = 0. The aim is to characterize the stability for the case when the equilibrium point is at the origin, that is *x*˜ = 0. For autonomous system, there is a convergence of the trajectory to a set, the same as the asymptotic stability of the origin. A major concern in analysing the stability of dynamical system is the robustness of various stability properties to uncertainties in the system's model. In the following, it is

*x*˙ = *h*(*t*, *x*), (1)

*<sup>x</sup>*˙ <sup>=</sup> *<sup>e</sup>*(*x*) + *<sup>d</sup>*(*t*), (2)

*x*˙ = *h*(*x*), (3)

**2. Perturbed system**

and define

Consider the following perturbed system

origin, we can rewrite the right-hand side as

Hence, the perturbed system (1) can be written as

or a non autonomous perturbation respectively.

where *<sup>h</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* is locally Lipschitz map from a domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*.

**2.1 Nonlinear autonomous perturbation** Consider the autonomous system

introduced the stability definition.

*• Asymptotically stable if it is stable and δ >* 0 *can be chosen such that*

$$\|\mathfrak{x}(0)\| < \delta \Rightarrow \lim\_{t \to \infty} \mathfrak{x}(t) = 0.$$

Let *<sup>V</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup>* be a differentiable function defined in a domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup><sup>n</sup>* that contains the origin. The derivative of *V*(*x*) along the trajectories of (3), denoted by *V*˙ (*x*), is given by

$$\dot{V}(\mathbf{x}) = \sum\_{i=1}^{n} \frac{\partial V}{\partial \mathbf{x}\_{i}} \dot{\mathbf{x}}\_{i} = \sum\_{i=1}^{n} \frac{\partial V}{\partial \mathbf{x}\_{i}} h\_{i}(\mathbf{x}). \tag{4}$$

The function *V*˙ (*x*) is dependent on the system's equation. Hence, if *V*˙ (*x*) is negative, *V*(*x*) will decrease along the solution of (3). The following lemma (Khalil, 2002) states Lyapunov's stability sense.

**Lemma 1.** *Let x* = 0 *be an equilibrium point for (3). Let V* : *D* → *R be a continuously differentiable function on a neighbourhood D of x* = 0*, such that*

$$V(0) = 0 \text{ and } \; V(\mathbf{x}) > 0 \text{ in } D - \{0\},$$

$$\dot{V}(\mathbf{x}) \le 0 \text{ in } D. \tag{5}$$

*Then, x* = 0 *is stable. Moreover, if*

$$
\dot{V}(\mathbf{x}) < 0 \text{ in } D - \{0\}. \tag{6}
$$

*Then x* = 0 *is asymptotically stable.*

*Proof.* Given *ε >* 0, choose *r* ∈ (0,*ε*] such that

$$B\_r = \{ \mathbf{x} \in \mathbb{R}^n | \|\mathbf{x}\| \le r \} \subset D.$$

Let *<sup>α</sup>* <sup>=</sup> *min*�*x*�<sup>=</sup>*rV*(*x*). Then, *<sup>α</sup> <sup>&</sup>gt;* 0 by (5). Take *<sup>β</sup>* <sup>∈</sup> (0, *<sup>α</sup>*), and let

$$\Omega\_{\beta} = \{ \mathbf{x} \in B\_r | V(\mathbf{x}) \le \beta \}.$$

The set Ω*<sup>β</sup>* has the property that any trajectory starting in Ω*<sup>β</sup>* at *t* = 0, stays in Ω*<sup>β</sup>* for all *t* ≥ 0. This follows from (5) since

$$\dot{V}(\mathbf{x}(t)) \le 0 \Rightarrow V(\mathbf{x}(t)) \le V(\mathbf{x}(0)) \le \beta, \quad \forall t \ge 0.$$

Since Ω*<sup>β</sup>* is a compact set (is closed and bounded since it is contained in *Br*), the system in (3) has a unique solution defined for all *t* ≥ 0, whenever *x*(0) ∈ Ω*β*. Since *V*(*x*) is continuous and *V*(0) = 0, there is *δ >* 0 such that

$$||\mathfrak{x}|| \le \delta \Rightarrow V(\mathfrak{x}) < \beta.$$

**Corollary 1.** *Let* Ω*<sup>c</sup> be a compact (closed and bounded) set with the property that every solution of (3) which starts in* Ω*<sup>c</sup> remains for all future time in* Ω*c. Let V* : Ω*<sup>c</sup>* → *R be a continuously differentiable function such that <sup>V</sup>*˙ (*x*) <sup>≤</sup> <sup>0</sup>*,* <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c. Let E be the set of all points in* <sup>Ω</sup>*<sup>c</sup> where <sup>V</sup>*˙ (*x*) = <sup>0</sup>*. Let M be*

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 223

**Corollary 2.** *Let x* = 0 *be an equilibrium point for (1). Let V* : Ω*<sup>c</sup>* → *R be a continuously differentiable positive definite function on a neighbourhood* <sup>Ω</sup>*<sup>c</sup> of x* <sup>=</sup> <sup>0</sup>*, such that <sup>V</sup>*˙ (*x*) <sup>≤</sup> <sup>0</sup>*,* <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c.*

With the previous stability criteria for equilibria point about the origin, it is necessary to

Consider the autonomous case in (2), when *d*(*t*) = 0 for all *t* ≥ 0. Suppose that the origin *x* = 0 is inside of *D* and is an equilibrium point for the nominal system *f*(*x*), that is, *f*(0) = 0.

*<sup>f</sup>*(*x*) = *<sup>f</sup>*(0) + *<sup>∂</sup> <sup>f</sup>*(*z*)

where *z* is a point on the line segment connecting *x* to the origin. The above equality is valid for any point *x* ∈ *D* such that the line segment connecting *x* to the origin lies entirely in *D*.

> *x* + *∂ f*(*z*)

*<sup>x</sup>* <sup>=</sup> *<sup>∂</sup> <sup>f</sup>*(0) *∂x*

*<sup>∂</sup><sup>x</sup>* , *and g*(*x*) =

*∂ f*(*z*)

*<sup>∂</sup><sup>x</sup>* <sup>−</sup> *<sup>∂</sup> <sup>f</sup>*(0) *∂x*

for any *k >* 0, there exists *r >* 0, such that ∀�*x*�<sup>2</sup> *< r*. It is possible to approximate in a small neighbourhood of the origin the nonlinear system *f*(*x*) by its linearization about the origin

(*x*) = *Ax*, *where A* <sup>=</sup> *<sup>∂</sup> <sup>f</sup>*(0)

**Corollary 3.** *The equilibrium point x* = 0 *of (8) is stable if and only if all eigenvalues of A satisfy Re*(*λi*) ≤ 0*. The equilibrium point x* = 0 *is asymptotically stable if and only if all eigenvalues of A satisfy Re*(*λi*) *<* 0*. When all eigenvalues of A satisfy Re*(*λi*) *<* 0*, A is called a stability matrix or a Hurwitz matrix. The origin of (8) is asymptotically stable if and only if A is a stability matrix.*

 

The following corollary characterizes the stability properties of the origin.

*∂x x*,

*∂ f*(*z*)

 2

*, and suppose that no solution can stay forever in S, other than the trivial*

*<sup>∂</sup><sup>x</sup>* <sup>−</sup> *<sup>∂</sup> <sup>f</sup>*(0) *∂x*

*<sup>∂</sup><sup>x</sup>* <sup>−</sup> *<sup>∂</sup> <sup>f</sup>*(0) *∂x*

�*x*�<sup>2</sup> ≤ *k*�*x*�2,

*f*(*x*) = *Ax* + *g*(*x*),

 *x*,

 *x*.

*<sup>∂</sup><sup>x</sup>* . (8)

*the largest invariant set in E. Then every solution starting in* Ω*<sup>c</sup> approaches M as t* → ∞*.*

**Remark 3.** *When V*˙ (*x*) *is negative definite, S* = 0*. Then, corollary 2 coincide with lemma 1.*

*Let S* =

**2.1.1 Mean value**

By the mean value

where

*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c*|*V*˙ (*x*) = <sup>0</sup>

Since *f*(0) = 0, we can write *f*(*x*) as

The function *g*(*x*) satisfies

introduce the specific analysis for autonomous perturbed system.

*<sup>f</sup>*(*x*) = *<sup>∂</sup> <sup>f</sup>*(*z*) *∂x*

*<sup>A</sup>* <sup>=</sup> *<sup>∂</sup> <sup>f</sup>*(0)

�*g*(*x*)�<sup>2</sup> ≤

˙

*solution. Then, the origin x* = 0 *is asymptotically stable.*

$$\text{Then}$$

and

$$
\mathcal{B}\_{\delta} \subset \Omega\_{\beta} \subset \mathcal{B}\_{r}.
$$

$$
\mathfrak{x}(0) \in \mathcal{B}\_{\delta} \Rightarrow \mathfrak{x}(0) \in \Omega\_{\beta} \Rightarrow \mathfrak{x}(t) \in \Omega\_{\beta} \Rightarrow \mathfrak{x}(t) \in \mathcal{B}\_{r}.
$$

Therefore,

$$\|\|x(0)\|\| < \delta \Rightarrow \|\|x(t)\|\| < r \le \varepsilon, \ \forall t \ge 0.$$

which shows that the equilibrium point *x* = 0 is stable. Now, to show asymptotic stability it is necessary to show that *x*(*t*) → 0 as *t* → ∞, that is, for every *a >* 0, there is *T >* 0 such that �*x*(*t*)� *< a*, for all *t > T*. For every *a <* 0, we can choose *b >* 0 such that Ω*<sup>b</sup>* ⊂ *Ba*. Therefore, it is sufficient to show that *V*(*x*(*t*)) → 0 as *t* → ∞. Since *V*(*x*) is monotonically decreasing and bounded from below by zero,

$$V(\mathfrak{x}(t)) \to \mathfrak{c} \ge 0 \text{ as } t \to \infty.$$

To show that *c* = 0, suppose by contradiction *c >* 0. By continuity of *V*(*x*), there is *d >* 0 such that *Bd* ⊂ Ω*c*. The limit *V*(*x*(*t*)) → *c >* 0 implies that the trajectory *x*(*t*) lies outside the ball *Bd* <sup>⊂</sup> <sup>Ω</sup>*<sup>c</sup>* for all *<sup>t</sup>* <sup>≥</sup> 0. When *<sup>V</sup>*˙ (*x*) is integrated on *<sup>t</sup>*, it follows by (6) that

$$V(\mathbf{x}(t)) = V(\mathbf{x}(0)) + \int\_0^t \dot{V}(\mathbf{x}(\tau))d\tau \le V(\mathbf{x}(0)) + kt\_\prime$$

where *<sup>k</sup>* <sup>=</sup> <sup>−</sup>*maxa*≤�*x*�≤*rV*˙ (*x*) *<sup>&</sup>lt;* 0. Since the right-hand side will eventually become negative, the inequality contradicts the assumption that *c >* 0.

**Remark 1.** *The origin is stable if there is a continuously differentiable positive definite function V*(*x*) *so that V*˙ (*x*) *is negative semi-definite, and it is asymptotically stable if V*˙ (*x*) *is negative definite.*

**Remark 2.** *The theorem's conditions are only sufficient. Failure of a Lyapunov function candidate to satisfy the conditions for stability or asymptotic stability does not mean that the equilibrium is not stable or asymptotically stable. It only means that such a stability property cannot be established by using this Lyapunov function candidate.*

For the case when the origin *x* = 0 is asymptotically stable, it is often interesting to determine how far from the origin can the trajectory be and still converges to the origin as *t* → ∞. This gives rise to the definition of the region of attraction or basin.

**Definition 2.** *Let x*(*t*, *x*(0)) *be the solution of (3) that starts at initial state x*<sup>0</sup> *at time t* = 0*. Then, the region of attraction is defined as the set of all points x such that* lim*t*→<sup>∞</sup> *x*(*t*, *x*(0)) = 0*.*

To find the exact region of attraction analytically might be difficult or even impossible. However, Lyapunov functions can be used to estimate the region of attraction, that is, to find sets contained in the region of attraction. From the proof of Lemma 1, we say that if there is a Lyapunov function that satisfies the conditions of asymptotic stability over a domain *D*, and if

$$\Omega\_{\mathbb{C}} = \{ \mathfrak{x} \in \mathbb{R}^{\mathbb{N}} | V(\mathfrak{x}) \le c \}, \tag{7}$$

is bounded and contained in *D*, then every trajectory starting in Ω*<sup>c</sup>* remains in Ω*c*, and approaches the origin as *<sup>t</sup>* <sup>→</sup> <sup>∞</sup>. The set in (7) with *<sup>V</sup>*˙ (*x*) <sup>≤</sup> 0, <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*<sup>c</sup>* is a positively invariant set, since, as we showed in the proof of Lemma 1, a solution starting in Ω*<sup>c</sup>* remains in Ω*<sup>c</sup>* for all *t* ≥ 0. Now, it is introduced the following corollaries known as the LaSalle invariance principle and the Barbashin-Krasovskii theorem.

**Corollary 1.** *Let* Ω*<sup>c</sup> be a compact (closed and bounded) set with the property that every solution of (3) which starts in* Ω*<sup>c</sup> remains for all future time in* Ω*c. Let V* : Ω*<sup>c</sup>* → *R be a continuously differentiable function such that <sup>V</sup>*˙ (*x*) <sup>≤</sup> <sup>0</sup>*,* <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c. Let E be the set of all points in* <sup>Ω</sup>*<sup>c</sup> where <sup>V</sup>*˙ (*x*) = <sup>0</sup>*. Let M be the largest invariant set in E. Then every solution starting in* Ω*<sup>c</sup> approaches M as t* → ∞*.*

**Corollary 2.** *Let x* = 0 *be an equilibrium point for (1). Let V* : Ω*<sup>c</sup>* → *R be a continuously differentiable positive definite function on a neighbourhood* <sup>Ω</sup>*<sup>c</sup> of x* <sup>=</sup> <sup>0</sup>*, such that <sup>V</sup>*˙ (*x*) <sup>≤</sup> <sup>0</sup>*,* <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c. Let S* = *<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*c*|*V*˙ (*x*) = <sup>0</sup> *, and suppose that no solution can stay forever in S, other than the trivial solution. Then, the origin x* = 0 *is asymptotically stable.*

**Remark 3.** *When V*˙ (*x*) *is negative definite, S* = 0*. Then, corollary 2 coincide with lemma 1.*

With the previous stability criteria for equilibria point about the origin, it is necessary to introduce the specific analysis for autonomous perturbed system.

#### **2.1.1 Mean value**

4 Will-be-set-by-IN-TECH

*B<sup>δ</sup>* ⊂ Ω*<sup>β</sup>* ⊂ *Br*,

*x*(0) ∈ *βδ* ⇒ *x*(0) ∈ Ω*<sup>β</sup>* ⇒ *x*(*t*) ∈ Ω*<sup>β</sup>* ⇒ *x*(*t*) ∈ *Br*.

�*x*(0)� *< δ* ⇒ �*x*(*t*)� *< r* ≤ *ε*, ∀*t* ≥ 0. which shows that the equilibrium point *x* = 0 is stable. Now, to show asymptotic stability it is necessary to show that *x*(*t*) → 0 as *t* → ∞, that is, for every *a >* 0, there is *T >* 0 such that �*x*(*t*)� *< a*, for all *t > T*. For every *a <* 0, we can choose *b >* 0 such that Ω*<sup>b</sup>* ⊂ *Ba*. Therefore, it is sufficient to show that *V*(*x*(*t*)) → 0 as *t* → ∞. Since *V*(*x*) is monotonically decreasing

*V*(*x*(*t*)) → *c* ≥ 0 *as t* → ∞. To show that *c* = 0, suppose by contradiction *c >* 0. By continuity of *V*(*x*), there is *d >* 0 such that *Bd* ⊂ Ω*c*. The limit *V*(*x*(*t*)) → *c >* 0 implies that the trajectory *x*(*t*) lies outside the ball

where *<sup>k</sup>* <sup>=</sup> <sup>−</sup>*maxa*≤�*x*�≤*rV*˙ (*x*) *<sup>&</sup>lt;* 0. Since the right-hand side will eventually become negative,

**Remark 1.** *The origin is stable if there is a continuously differentiable positive definite function V*(*x*) *so that V*˙ (*x*) *is negative semi-definite, and it is asymptotically stable if V*˙ (*x*) *is negative definite.*

**Remark 2.** *The theorem's conditions are only sufficient. Failure of a Lyapunov function candidate to satisfy the conditions for stability or asymptotic stability does not mean that the equilibrium is not stable or asymptotically stable. It only means that such a stability property cannot be established by*

For the case when the origin *x* = 0 is asymptotically stable, it is often interesting to determine how far from the origin can the trajectory be and still converges to the origin as *t* → ∞. This

**Definition 2.** *Let x*(*t*, *x*(0)) *be the solution of (3) that starts at initial state x*<sup>0</sup> *at time t* = 0*. Then,*

To find the exact region of attraction analytically might be difficult or even impossible. However, Lyapunov functions can be used to estimate the region of attraction, that is, to find sets contained in the region of attraction. From the proof of Lemma 1, we say that if there is a Lyapunov function that satisfies the conditions of asymptotic stability over a domain *D*, and

is bounded and contained in *D*, then every trajectory starting in Ω*<sup>c</sup>* remains in Ω*c*, and approaches the origin as *<sup>t</sup>* <sup>→</sup> <sup>∞</sup>. The set in (7) with *<sup>V</sup>*˙ (*x*) <sup>≤</sup> 0, <sup>∀</sup>*<sup>x</sup>* <sup>∈</sup> <sup>Ω</sup>*<sup>c</sup>* is a positively invariant set, since, as we showed in the proof of Lemma 1, a solution starting in Ω*<sup>c</sup>* remains in Ω*<sup>c</sup>* for all *t* ≥ 0. Now, it is introduced the following corollaries known as the LaSalle invariance

*the region of attraction is defined as the set of all points x such that* lim*t*→<sup>∞</sup> *x*(*t*, *x*(0)) = 0*.*

*<sup>V</sup>*˙ (*x*(*τ*))*d<sup>τ</sup>* <sup>≤</sup> *<sup>V</sup>*(*x*(0)) + *kt*,

<sup>Ω</sup>*<sup>c</sup>* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*|*V*(*x*) <sup>≤</sup> *<sup>c</sup>*} , (7)

 *t* 0

*Bd* <sup>⊂</sup> <sup>Ω</sup>*<sup>c</sup>* for all *<sup>t</sup>* <sup>≥</sup> 0. When *<sup>V</sup>*˙ (*x*) is integrated on *<sup>t</sup>*, it follows by (6) that

*V*(*x*(*t*)) = *V*(*x*(0)) +

gives rise to the definition of the region of attraction or basin.

principle and the Barbashin-Krasovskii theorem.

the inequality contradicts the assumption that *c >* 0.

Then

and

if

Therefore,

and bounded from below by zero,

*using this Lyapunov function candidate.*

Consider the autonomous case in (2), when *d*(*t*) = 0 for all *t* ≥ 0. Suppose that the origin *x* = 0 is inside of *D* and is an equilibrium point for the nominal system *f*(*x*), that is, *f*(0) = 0. By the mean value

$$f(x) = f(0) + \frac{\partial f(z)}{\partial x} x\_{\prime}$$

where *z* is a point on the line segment connecting *x* to the origin. The above equality is valid for any point *x* ∈ *D* such that the line segment connecting *x* to the origin lies entirely in *D*. Since *f*(0) = 0, we can write *f*(*x*) as

$$f(\mathbf{x}) = \frac{\partial f(z)}{\partial \mathbf{x}} \mathbf{x} = \frac{\partial f(0)}{\partial \mathbf{x}} \mathbf{x} + \left[\frac{\partial f(z)}{\partial \mathbf{x}} - \frac{\partial f(0)}{\partial \mathbf{x}}\right] \mathbf{x},$$

$$f(\mathbf{x}) = A\mathbf{x} + g(\mathbf{x}),$$

where

$$A = \frac{\partial f(0)}{\partial \mathbf{x}}, \text{ and } \operatorname{g}(\mathbf{x}) = \left[\frac{\partial f(z)}{\partial \mathbf{x}} - \frac{\partial f(0)}{\partial \mathbf{x}}\right] \mathbf{x}.$$

The function *g*(*x*) satisfies

$$\|g(\mathbf{x})\|\_{2} \le \left|\frac{\partial f(z)}{\partial \mathbf{x}} - \frac{\partial f(0)}{\partial \mathbf{x}}\right|\_{2} \|\mathbf{x}\|\_{2} \le k \|\mathbf{x}\|\_{2'}$$

for any *k >* 0, there exists *r >* 0, such that ∀�*x*�<sup>2</sup> *< r*. It is possible to approximate in a small neighbourhood of the origin the nonlinear system *f*(*x*) by its linearization about the origin

$$
\dot{f}(\mathbf{x}) = A\mathbf{x}, \text{ where } A = \frac{\partial f(0)}{\partial \mathbf{x}}.\tag{8}
$$

The following corollary characterizes the stability properties of the origin.

**Corollary 3.** *The equilibrium point x* = 0 *of (8) is stable if and only if all eigenvalues of A satisfy Re*(*λi*) ≤ 0*. The equilibrium point x* = 0 *is asymptotically stable if and only if all eigenvalues of A satisfy Re*(*λi*) *<* 0*. When all eigenvalues of A satisfy Re*(*λi*) *<* 0*, A is called a stability matrix or a Hurwitz matrix. The origin of (8) is asymptotically stable if and only if A is a stability matrix.*

Consider a quadratic Lyapunov function candidate

$$V(\mathbf{x}) = \mathbf{x}^T P \mathbf{x}\_{\prime\prime}$$

where *P* is a real symmetric positive definite matrix. The derivative of *V* along the trajectories of the linear system (8) is given by

$$\dot{V}(\mathbf{x}) = \dot{\mathbf{x}}^T P \mathbf{x} + \mathbf{x}^T P \dot{\mathbf{x}} = \mathbf{x}^T \left( P \mathbf{A} + A^T P \right) \mathbf{x} = -\mathbf{x}^T Q \mathbf{x}.$$

Asymptotic stability of the origin can be also investigated using Lyapunov's equation, as it is shown on corollary 4.

**Corollary 4.** *A matrix A is a stability matrix, that is, Re*(*λi*) *<* 0 *for all eigenvalues of A, if and only if for any given positive definite symmetric matrix Q there exists a positive definite symmetric matrix P that satisfies the Lyapunov equation*

$$PA + A^T P = -Q.\tag{9}$$

The first term on the right-hand side is negative definite, while the second term is indefinite.

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 225

�*g*(*x*)�<sup>2</sup> ≤ *k*�*x*�2, ∀�*x*�<sup>2</sup> *< r*.

*<sup>V</sup>*˙ (*x*) ≤ −*xTQx* <sup>+</sup> <sup>2</sup>*k*�*P*�2�*x*�<sup>2</sup>

*<sup>x</sup>TQx* <sup>≥</sup> *<sup>λ</sup>min*(*Q*)�*x*�<sup>2</sup>

where *λmin*(·) denotes the minimum eigenvalue of a matrix. Note that *λmin*(*Q*) is real and

*<sup>V</sup>*˙ (*x*) ≤ − (*λmin*(*Q*) <sup>−</sup> <sup>2</sup>*k*�*P*�2�) �*x*�<sup>2</sup>

ensures that *V*˙ (*x*) is negative definite. By lemma 1, we conclude that the origin is

*x*˙ = *h*(*t*, *x*) = *f*(*x*) + *g*(*x*) + *d*(*t*),

where *<sup>f</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* and *<sup>g</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* are locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, *<sup>d</sup>*(*t*) is a uniformly bounded disturbance that satisfies |*d*(*t*)| ≤ *δ* for all *t* ≥ 0 and *e*(*x*) = *f*(*x*) + *g*(*x*). The notions of stability and asymptotic stability of the equilibrium of a non autonomous system are basically the same as Definition 1 for autonomous system, see (Khalil, 2002). The difference here is that, while the solution of an autonomous system depends only on (*t* − *t*0), the solution of a non autonomous system may depend on both *t* and *t*0. Here, in that case the

�*x*(*t*0)� *< δ* ⇒ �*x*(*t*)� *< ε*, ∀*t* ≥ 0.

*<sup>t</sup>*→<sup>∞</sup> *<sup>x</sup>*(*t*) = 0.

�*x*(*t*0)� *< δ* ⇒ lim

It is necessary to introduce special scalar functions that will help to characterize and study the

*λmin*(*Q*) �*P*�<sup>2</sup>

,

*k <* 1 2

*x*˙ = *e*(*x*) + *d*(*t*),

Therefore, the stability behavior of the equilibrium point will be dependent on *t*0.

*• Stable, if for each ε >* 0*, and any t*<sup>0</sup> ≥ 0 *there is δ* = *δ*(*ε*, *t*0) ≥ 0 *such that*

*• Asymptotically stable if it is stable and δ >* 0 *can be chosen such that*

behavior of a solution for the non autonomous system.

2,

2.

2,

For any *k >* 0, there exists *r >* 0. Hence, after majorize the right-hand side

positive since *Q* is symmetric and positive definite. Thus

**2.2 Non linear non autonomous perturbation**

function *d*(*t*) �= 0 for all *t* ≥ 0 about the origin.

**Definition 3.** *The equilibrium point x* = 0 *of (2) is*

But the function *g*(*x*) satisfies

but

By choosing

asymptotically stable.

*• Unstable, if not stable.*

Consider the system given in (2)

If *Q* is positive definite, then the origin is asymptotically stable, that is, *Re*(*λi*) *<* 0, for all eigenvalues of *A*. Here it follows the usual procedure of Lyapunov's method, where it choose *V*(*x*) to be positive definite and then check negative definiteness of *V*˙ (*x*).

**Remark 4.** *If matrix A is a stability matrix, then P is a unique solution of (9).*

The Lyapunov equation can be used to test whether or not a matrix *A* is a stability matrix, as an alternative to calculating the eigenvalues of *A*. The existence of a Lyapunov function will allow us to draw conslusions about the system when the linear term *Ax* is perturbed, whether such perturbation is a linear perturbation in the coefficients of *A* or a nonlinear autonomous perturbation. The following lemma is known as Lyapunov's indirect method or Lyapunov first method.

**Lemma 2.** *Let x* = 0 *be an equilibrium point for the nonlinear system*

$$
\dot{\mathfrak{x}} = f(\mathfrak{x}),
\tag{10}
$$

*where f* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup> is continuously differentiable and D is a neighbourhood of the origin. Let*

$$A = \left. \frac{\delta f(\mathbf{x})}{\delta \mathbf{x}} \right|\_{\mathbf{x}=\mathbf{0}}$$

.

*Then, the origin is asymptotically stable if Re*(*λi*) *<* 0 *for all eigenvalues of A.*

*Proof.* Let *A* be a stability matrix. Then, by corollary 4 it is known that for any positive definite symmetric matrix *Q*, the solution *P* of the Lyapunov equation (9) is positive definite. Consider

$$V(\mathbf{x}) = \mathbf{x}^T P \mathbf{x}\_r$$

as a Lyapunov function candidate for the nonlinear system. The derivative of *V*(*x*) along the trajectories of the system is given by

$$\dot{V}(\mathbf{x}) = \mathbf{x}^T P \mathbf{e}(\mathbf{x}) + \mathbf{e}^T(\mathbf{x}) P \mathbf{x} = \mathbf{x}^T P \left( A \mathbf{x} + \mathbf{g}(\mathbf{x}) \right) + \left( \mathbf{x}^T A^T + \mathbf{g}^T(\mathbf{x}) \right) P \mathbf{x}.$$

$$\dot{V}(\mathbf{x}) = \mathbf{x}^T \left( P \mathbf{A} + A^T P \right) \mathbf{x} + 2 \mathbf{x}^T P \mathbf{g}(\mathbf{x}) = -\mathbf{x}^T Q \mathbf{x} + 2 \mathbf{x}^T P \mathbf{g}(\mathbf{x}).$$

The first term on the right-hand side is negative definite, while the second term is indefinite. But the function *g*(*x*) satisfies

$$\|\|g(\mathfrak{x})\|\|\_{2} \le k \|\|\mathfrak{x}\|\|\_{2\prime} \,\,\,\forall \|\|\mathfrak{x}\|\|\_{2} < r.$$

For any *k >* 0, there exists *r >* 0. Hence, after majorize the right-hand side

$$\dot{V}(\mathbf{x}) \le -\mathbf{x}^T \mathbf{Q} \mathbf{x} + 2k \|P\|\_2 \|\mathbf{x}\|\_{2'}^2$$

but

6 Will-be-set-by-IN-TECH

*V*(*x*) = *xTPx*,

where *P* is a real symmetric positive definite matrix. The derivative of *V* along the trajectories

Asymptotic stability of the origin can be also investigated using Lyapunov's equation, as it is

**Corollary 4.** *A matrix A is a stability matrix, that is, Re*(*λi*) *<* 0 *for all eigenvalues of A, if and only if for any given positive definite symmetric matrix Q there exists a positive definite symmetric matrix*

If *Q* is positive definite, then the origin is asymptotically stable, that is, *Re*(*λi*) *<* 0, for all eigenvalues of *A*. Here it follows the usual procedure of Lyapunov's method, where it choose

The Lyapunov equation can be used to test whether or not a matrix *A* is a stability matrix, as an alternative to calculating the eigenvalues of *A*. The existence of a Lyapunov function will allow us to draw conslusions about the system when the linear term *Ax* is perturbed, whether such perturbation is a linear perturbation in the coefficients of *A* or a nonlinear autonomous perturbation. The following lemma is known as Lyapunov's indirect method or Lyapunov

*where f* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup> is continuously differentiable and D is a neighbourhood of the origin. Let*

*<sup>A</sup>* <sup>=</sup> *<sup>δ</sup> <sup>f</sup>*(*x*) *δx*

*Proof.* Let *A* be a stability matrix. Then, by corollary 4 it is known that for any positive definite symmetric matrix *Q*, the solution *P* of the Lyapunov equation (9) is positive definite. Consider

*V*(*x*) = *xTPx*,

as a Lyapunov function candidate for the nonlinear system. The derivative of *V*(*x*) along the

 *x*=0 .

*PA* + *ATP*

*<sup>x</sup>* <sup>=</sup> <sup>−</sup>*xTQx*.

*PA* <sup>+</sup> *<sup>A</sup>TP* <sup>=</sup> <sup>−</sup>*Q*. (9)

*x*˙ = *f*(*x*), (10)

*<sup>x</sup>* <sup>+</sup> <sup>2</sup>*xTPg*(*x*) = <sup>−</sup>*xTQx* <sup>+</sup> <sup>2</sup>*xTPg*(*x*).

*x<sup>T</sup> A<sup>T</sup>* + *gT*(*x*)

 *Px*,

*TPx* <sup>+</sup> *<sup>x</sup>TPx*˙ <sup>=</sup> *<sup>x</sup><sup>T</sup>*

*V*(*x*) to be positive definite and then check negative definiteness of *V*˙ (*x*). **Remark 4.** *If matrix A is a stability matrix, then P is a unique solution of (9).*

**Lemma 2.** *Let x* = 0 *be an equilibrium point for the nonlinear system*

*Then, the origin is asymptotically stable if Re*(*λi*) *<* 0 *for all eigenvalues of A.*

*V*˙ (*x*) = *xTPe*(*x*) + *eT*(*x*)*Px* = *xTP* (*Ax* + *g*(*x*)) +

*PA* + *ATP*

Consider a quadratic Lyapunov function candidate

*V*˙ (*x*) = *x*˙

of the linear system (8) is given by

*P that satisfies the Lyapunov equation*

trajectories of the system is given by

*<sup>V</sup>*˙ (*x*) = *<sup>x</sup><sup>T</sup>*

shown on corollary 4.

first method.

$$\mathbf{x}^{\mathsf{T}} \mathbf{Q} \mathbf{x} \ge \lambda\_{\min}(\mathbf{Q}) \|\mathbf{x}\|\_{2\mathsf{T}}^2$$

where *λmin*(·) denotes the minimum eigenvalue of a matrix. Note that *λmin*(*Q*) is real and positive since *Q* is symmetric and positive definite. Thus

$$\dot{V}(\mathfrak{x}) \le -\left(\lambda\_{\min}(Q) - 2k||P||\_2||\right)||\mathfrak{x}||\_2^2.$$

By choosing

$$k < \frac{1}{2} \frac{\lambda\_{\min}(\mathbf{Q})}{||P||\_2}.$$

ensures that *V*˙ (*x*) is negative definite. By lemma 1, we conclude that the origin is asymptotically stable.

#### **2.2 Non linear non autonomous perturbation**

Consider the system given in (2)

$$\begin{aligned} \dot{\mathfrak{x}} &= h(t, \mathfrak{x}) = f(\mathfrak{x}) + \mathfrak{g}(\mathfrak{x}) + d(t)\mathfrak{y}, \\ \dot{\mathfrak{x}} &= e(\mathfrak{x}) + d(t)\mathfrak{y} \end{aligned}$$

where *<sup>f</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* and *<sup>g</sup>* : *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup>* are locally Lipschitz in *<sup>x</sup>* on domain *<sup>D</sup>* <sup>⊂</sup> *<sup>R</sup>n*, *<sup>d</sup>*(*t*) is a uniformly bounded disturbance that satisfies |*d*(*t*)| ≤ *δ* for all *t* ≥ 0 and *e*(*x*) = *f*(*x*) + *g*(*x*).

The notions of stability and asymptotic stability of the equilibrium of a non autonomous system are basically the same as Definition 1 for autonomous system, see (Khalil, 2002). The difference here is that, while the solution of an autonomous system depends only on (*t* − *t*0), the solution of a non autonomous system may depend on both *t* and *t*0. Here, in that case the function *d*(*t*) �= 0 for all *t* ≥ 0 about the origin.

Therefore, the stability behavior of the equilibrium point will be dependent on *t*0.

**Definition 3.** *The equilibrium point x* = 0 *of (2) is*

*• Stable, if for each ε >* 0*, and any t*<sup>0</sup> ≥ 0 *there is δ* = *δ*(*ε*, *t*0) ≥ 0 *such that*

$$\|\|\mathbf{x}(t\_0)\|\| < \delta \Rightarrow \|\|\mathbf{x}(t)\|\| < \varepsilon, \ \forall t \ge 0.$$


$$\|\mathfrak{x}(t\_0)\| < \delta \Rightarrow \lim\_{t \to \infty} \mathfrak{x}(t) = 0.$$

It is necessary to introduce special scalar functions that will help to characterize and study the behavior of a solution for the non autonomous system.

Lyapunov stability theorems give sufficient conditions for stability, asymptotic stability, and so on. They do not say whether the given conditions are also necessary. There are converse theorems which establish, that for many Lyapunov stability theorems the given conditions are indeed necessary (Hahn, 1967; Krasovskii, 1967). The converse theorems are proved by actually constructing auxiliary functions that satisfy the conditions of the respective theorems. Almost always this construction assumes the knowledge of the solution of the differential equation. The origin *x* = 0 of the perturbed non-autonomous system (2), may not be an equilibrium point. We can no longer study stability of the origin as an equilibrium point, nor should we expect the solution of the perturbed system to approach the origin as *t* → ∞. If the perturbation terms *g*(*x*) and *d*(*t*) are small in some sense, then the solution *x*(*t*) will be

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 227

**Definition 7.** *The solution of x*˙ = *h*(*t*, *x*) *is said to be uniformly bounded if there exist constants a*

**Theorem 1.** *Let D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup><sup>n</sup>* | �*x*� *<sup>&</sup>lt; <sup>r</sup>*} *and h* : [0, <sup>∞</sup>) <sup>×</sup> *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup> be continuous in t and locally*

*for all* �*x*� ≥ *μ >* 0*, and for all x* ∈ *D, where w*1(·)*, w*2(·) *and w*3(·) *are positive definite functions*

*Proof.* By definition 6, it is necessary to prove that the origin is uniformly asymptotically stable in order to have an uniformly bounded solution. First, consider the derivative of *V*(*t*, *x*) along

Ω*t*,*<sup>ρ</sup>* = {*x* ∈ *D*|*V*(*t*, *x*) ≤ *w*1(*ρ*)} .

*w*2(�*x*�) ≤ *w*1(*ρ*) *and V*(*t*, *x*) ≤ *w*1(*ρ*).

<sup>2</sup> (*w*1(*ρ*))

*∂V ∂x*

�*x*(*t*0)�, *t* − *t*<sup>0</sup>

1 *w*2(*μ*) 

*h*(*t*, *x*) ≤ *w*<sup>3</sup>

<sup>2</sup> (*w*1(*r*))*. Then, there exists a positive definite and decreasing function u*(·, ·) *and a finite*

 *x* 

�*x*(*t*0)� *< μ* ⇒ �*x*(*t*)� *< a*, ∀*t > t*<sup>0</sup> + *T*. (14)

*w*1(�*x*�) ≤ *V*(*t*, *x*) ≤ *w*2(�*x*�), (15)

*. Furthermore, if wi*(*r*) = *kirc, for some positive constants ki and c,*

*h*(*t*, *x*) ≤ −*w*<sup>3</sup> (�*x*�).

since

, (16)

, ∀*t*<sup>0</sup> ≤ *t < t*1, (17)

, ∀*t* ≥ *t*1, (18)

bounded by a small bound, that is �*x*(*t*)� will be small for sufficiently large *t*.

The following Lyapunov like theorem is useful to show uniform boundedness.

*Lipschitz in x. Let V* : [0, ∞) × *D* → *R be a continuously differentiable function such that*

*∂V ∂x*

*and b, and for every μ* ∈ (0, *b*) *there is a constant T such that*

*and μ < w*−<sup>1</sup>

*for all* �*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup>

the trajectories of (2)

The set <sup>Ω</sup>*t*,*<sup>ρ</sup>* contains the ball

*time t*<sup>1</sup> *(dependent on x*(*t*0) *and μ) such that*

2 *w*1(*r*) 

*It is said globally uniformly bounded if (14) holds for arbitrarily large μ.*

*∂V <sup>∂</sup><sup>t</sup>* <sup>+</sup>

�*x*(*t*)� ≤ *<sup>w</sup>*−<sup>1</sup>

*<sup>∂</sup><sup>t</sup>* <sup>+</sup>

�*x*� ≤ *<sup>w</sup>*−<sup>1</sup>

�*x*(*t*)� ≤ *u*

*then u*(*r*,*s*) = *kre*−*<sup>α</sup>s, with k* = (*k*2/*k*1)1/*c, and α* = (*k*3/*k*2*c*)*.*

Let *ρ < r*, and define a time-dependent set Ω*t*,*<sup>ρ</sup>* by

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) = *<sup>∂</sup><sup>V</sup>*

**Definition 4.** *A scalar function w*(*r*) ∈ *R is said to be positive definite, if it is continuous and w*(*r*) *>* 0 *for* |*r*| *>* 0 *and w*(0) = 0*. The scalar function is radially unbounded if w*(*r*) → ∞ *as* |*r*| → ∞*.*

**Definition 5.** *A scalar function u*(*r*,*s*) ∈ *R is said to be positive definite and decreasing, if for each fixed s, the function u*(*r*,*s*) *>* 0 *and u*(*r*, 0) = 0 *is continuous with respect to r, and for each fixed r the function u*(*r*,*s*) *is continuous and decreasing with respect to s and u*(*r*,*s*) → 0 *as* |*s*| → ∞*.*

The following corollary states some properties of positive definite functions.

**Corollary 5.** *Let w*1(·) *and w*2(·) *be positive definite functions on domain D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*, �*x*� *<sup>&</sup>lt; <sup>r</sup>*}*. Consider the following difference between scalar functions*

$$k\_1 w\_1(r) - k\_2 w\_2(||\mathbf{x}||) \ge 0, \text{ where } ||\mathbf{x}|| < r.$$

*If the leading term can be factorized, then the bound is given by*

$$\|\mathbf{x}\| \le \gamma \quad \text{where} \quad \gamma = w\_2^{-1}[w\_1(r)].$$

The following stability properties of the origin are given.

**Definition 6.** *The equilibrium point x* = 0 *of (2) is*

*1. Uniformly stable, if there exists a positive definite function w*(·) *and a positive constant r, independent of t*<sup>0</sup> *such that*

$$\|\|\mathbf{x}(t)\|\| \le w(\|\|\mathbf{x}(t\_0)\|\|), \forall t \ge t\_0 > \mathbf{0}, \forall \|\|\mathbf{x}(t\_0)\|\| < r. \tag{11}$$

*2. Uniformly asymptotically stable, if there exist a positive definite and decreasing function u*(·, ·) *and a positive constant r, independent of t*<sup>0</sup> *such that*

$$\|\mathbf{x}(t)\| \le \mathfrak{u}(\|\mathbf{x}(t\_0)\|, t - t\_0), \forall t \ge t\_0 > 0, \forall \|\mathbf{x}(t\_0)\| < r. \tag{12}$$


$$
\mu(r,s) = kre^{-as}, \ \ k > 0, \ \mathfrak{a} > 0. \tag{13}
$$

To establish uniform asymptotic stability of the origin, it is necessary to verify inequality (12) with the aid of an auxiliary scalar differential equation. The following corollary defines a scalar solution of a special equation.

**Corollary 6.** *Consider the scalar differential equation*

$$
\dot{y} = -w(y)\_\prime \ y(t\_0) = y\_0.
$$

*where w*(·) *is a locally Lipschitz positive definite function. Then, this equation has a unique solution y*(*t*) *defined for all t* ≥ *t*<sup>0</sup>

$$y(t) = \sigma(y(t\_0), t - t\_0)$$

*where σ*(*r*,*s*) *is a positive definite and decreasing function, see Definition 5.*

8 Will-be-set-by-IN-TECH

**Definition 4.** *A scalar function w*(*r*) ∈ *R is said to be positive definite, if it is continuous and w*(*r*) *>* 0 *for* |*r*| *>* 0 *and w*(0) = 0*. The scalar function is radially unbounded if w*(*r*) → ∞ *as*

**Definition 5.** *A scalar function u*(*r*,*s*) ∈ *R is said to be positive definite and decreasing, if for each fixed s, the function u*(*r*,*s*) *>* 0 *and u*(*r*, 0) = 0 *is continuous with respect to r, and for each fixed r the function u*(*r*,*s*) *is continuous and decreasing with respect to s and u*(*r*,*s*) → 0 *as* |*s*| → ∞*.*

**Corollary 5.** *Let w*1(·) *and w*2(·) *be positive definite functions on domain D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*, �*x*� *<sup>&</sup>lt; <sup>r</sup>*}*.*

*k*1*w*1(*r*) − *k*2*w*2(�*x*�) ≥ 0, *where* �*x*� *< r*.

*1. Uniformly stable, if there exists a positive definite function w*(·) *and a positive constant r,*

*2. Uniformly asymptotically stable, if there exist a positive definite and decreasing function u*(·, ·) *and*

*3. Globally uniformly asymptotically stable, if inequality (12) is satisfied for any initial state x*(*t*0)*.*

To establish uniform asymptotic stability of the origin, it is necessary to verify inequality (12) with the aid of an auxiliary scalar differential equation. The following corollary defines a

*y*˙ = −*w*(*y*), *y*(*t*0) = *y*0.

*where w*(·) *is a locally Lipschitz positive definite function. Then, this equation has a unique solution*

*y*(*t*0), *t* − *t*<sup>0</sup>

*u*(*r*,*s*) = *kre*−*α<sup>s</sup>*

*y*(*t*) = *σ*

*where σ*(*r*,*s*) *is a positive definite and decreasing function, see Definition 5.*

<sup>2</sup> [*w*1(*r*)].

�*x*(*t*)� ≤ *w*(�*x*(*t*0)�), ∀*t* ≥ *t*<sup>0</sup> *>* 0, ∀�*x*(*t*0)� *< r*. (11)

�*x*(*t*)� ≤ *u*(�*x*(*t*0)�, *t* − *t*0), ∀*t* ≥ *t*<sup>0</sup> *>* 0, ∀�*x*(*t*0)� *< r*. (12)

, *k >* 0, *α >* 0. (13)

�*x*� ≤ *<sup>γ</sup>*, *where <sup>γ</sup>* <sup>=</sup> *<sup>w</sup>*−<sup>1</sup>

The following corollary states some properties of positive definite functions.

*Consider the following difference between scalar functions*

*If the leading term can be factorized, then the bound is given by*

The following stability properties of the origin are given.

**Definition 6.** *The equilibrium point x* = 0 *of (2) is*

*a positive constant r, independent of t*<sup>0</sup> *such that*

*4. Exponentially stable if inequality (12) is satisfied with*

**Corollary 6.** *Consider the scalar differential equation*

*independent of t*<sup>0</sup> *such that*

scalar solution of a special equation.

*y*(*t*) *defined for all t* ≥ *t*<sup>0</sup>


Lyapunov stability theorems give sufficient conditions for stability, asymptotic stability, and so on. They do not say whether the given conditions are also necessary. There are converse theorems which establish, that for many Lyapunov stability theorems the given conditions are indeed necessary (Hahn, 1967; Krasovskii, 1967). The converse theorems are proved by actually constructing auxiliary functions that satisfy the conditions of the respective theorems. Almost always this construction assumes the knowledge of the solution of the differential equation. The origin *x* = 0 of the perturbed non-autonomous system (2), may not be an equilibrium point. We can no longer study stability of the origin as an equilibrium point, nor should we expect the solution of the perturbed system to approach the origin as *t* → ∞. If the perturbation terms *g*(*x*) and *d*(*t*) are small in some sense, then the solution *x*(*t*) will be bounded by a small bound, that is �*x*(*t*)� will be small for sufficiently large *t*.

**Definition 7.** *The solution of x*˙ = *h*(*t*, *x*) *is said to be uniformly bounded if there exist constants a and b, and for every μ* ∈ (0, *b*) *there is a constant T such that*

$$\|\|\mathbf{x}(t\_0)\|\| < \mu \Rightarrow \|\|\mathbf{x}(t)\|\| < a, \ \forall t > t\_0 + T. \tag{14}$$

*It is said globally uniformly bounded if (14) holds for arbitrarily large μ.*

The following Lyapunov like theorem is useful to show uniform boundedness.

**Theorem 1.** *Let D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup><sup>n</sup>* | �*x*� *<sup>&</sup>lt; <sup>r</sup>*} *and h* : [0, <sup>∞</sup>) <sup>×</sup> *<sup>D</sup>* <sup>→</sup> *<sup>R</sup><sup>n</sup> be continuous in t and locally Lipschitz in x. Let V* : [0, ∞) × *D* → *R be a continuously differentiable function such that*

$$w\_1(\|\mathbf{x}\|) \le V(t, \mathbf{x}) \le w\_2(\|\mathbf{x}\|), \tag{15}$$

$$\frac{\partial V}{\partial t} + \frac{\partial V}{\partial x}h(t, x) \le w\_3(x),\tag{16}$$

*for all* �*x*� ≥ *μ >* 0*, and for all x* ∈ *D, where w*1(·)*, w*2(·) *and w*3(·) *are positive definite functions and μ < w*−<sup>1</sup> <sup>2</sup> (*w*1(*r*))*. Then, there exists a positive definite and decreasing function u*(·, ·) *and a finite time t*<sup>1</sup> *(dependent on x*(*t*0) *and μ) such that*

$$\|\|\mathbf{x}(t)\|\| \le \mu \left( \|\|\mathbf{x}(t\_0)\|\|, t - t\_0 \right), \ \forall t\_0 \le t < t\_1. \tag{17}$$

$$\|\|x(t)\|\| \le w\_1^{-1} \left(w\_2(\mu)\right), \quad \forall t \ge t\_1. \tag{18}$$

*for all* �*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup> 2 *w*1(*r*) *. Furthermore, if wi*(*r*) = *kirc, for some positive constants ki and c, then u*(*r*,*s*) = *kre*−*<sup>α</sup>s, with k* = (*k*2/*k*1)1/*c, and α* = (*k*3/*k*2*c*)*.*

*Proof.* By definition 6, it is necessary to prove that the origin is uniformly asymptotically stable in order to have an uniformly bounded solution. First, consider the derivative of *V*(*t*, *x*) along the trajectories of (2)

$$
\dot{V}(t, \mathbf{x}) = \frac{\partial V}{\partial t} + \frac{\partial V}{\partial \mathbf{x}} h(t, \mathbf{x}) \le -w\_3 \left( ||\mathbf{x}|| \right).
$$

Let *ρ < r*, and define a time-dependent set Ω*t*,*<sup>ρ</sup>* by

$$\Omega\_{t,\rho} = \{ \mathfrak{x} \in D | V(t, \mathfrak{x}) \le w\_1(\rho) \} \dots$$

The set <sup>Ω</sup>*t*,*<sup>ρ</sup>* contains the ball �*x*� ≤ *<sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*)) since

$$w\_2(\|\mathbf{x}\|) \le w\_1(\rho) \quad \text{and} \quad V(t, \mathbf{x}) \le w\_1(\rho).$$

enter the set Ω*t*,*<sup>η</sup>* in finite time. Once inside the set, the solution remains inside for all *t* ≥ *t*1.

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 229

*V*(*t*, *x*(*t*)) ≤ *w*1(*η*), ∀*t* ≥ *t*1,

�*x*(*t*)� ≤ *η*, ∀*t* ≥ *t*1.

uniformly asymptotically stable. The exponentially decaying for *w*1(·), *w*2(·) and *w*3(·) is

 *<sup>r</sup> k*2

*σ*(*r*,*s*) = *re*−(*k*3/*k*2)*<sup>s</sup>*

Inequalities (17) and (18) show that the solution *x*(*t*) is uniformly bounded for all *t* ≥ *t*0.

*bound approaches zero. Sometimes, it is possible to combine inequalities (17) and (18) in one inequality*

 + *w*−<sup>1</sup>

Now, let us illustrate how Theorem 1 is used in the analysis of the perturbed system (2), when the origin of the nominal system is exponentially stable and the system has a uniform bounded

**Lemma 3.** *Let x* = 0 *be an exponentially stable equilibrium point of the nominal system (8). Let V* : [0, ∞) × *D* → *R be a Lyapunov function of the nominal system that satisfies inequalities (15) and*

*for all t* ≥ 0*, x*(*t*) ∈ *D, and some positive constants* 0 *< θ <* 1*,* 0 *< ζ <* 1*, c*<sup>2</sup> *>* 0*, c*<sup>4</sup> *>* 0*, c*<sup>5</sup> *>* 0

�*x*(*t*)� ≤ *k*�*x*(*t*0)�*exp*(−*α*(*t* − *t*0)), ∀*t*<sup>0</sup> ≤ *t < t*1,

�*x*(*t*)� ≤ *b*, ∀*t* ≥ *t*1,

*k*2*rce*−(*k*3/*k*2)*<sup>s</sup> k*1

, *ki >* 0, *c >* 0, *i* = 1, 2, 3.

.

<sup>1</sup> (*w*2(*μ*)) *be a positive definite function called the bound of μ. As μ* → 0*, the*

*c*5 *rθ c*<sup>1</sup> *c*2 ,

<sup>√</sup>*c*1/*c*2*, the solution of the perturbed system x*(*t*) *satisfies*

<sup>1</sup> (*w*2(*μ*)), ∀*t* ≥ *t*0. (19)

1/*<sup>c</sup>* .

1/*<sup>c</sup> c* <sup>=</sup> *<sup>k</sup>*<sup>3</sup> *k*2 *r*. <sup>2</sup> (*w*1(*ρ*))}. Thus, inequality

<sup>2</sup> (*w*1(*ρ*))}, which implies that the origin *x* = 0 is

Hence, any initial state *<sup>x</sup>*(*t*0) can be included in the set {�*x*� ≤ *<sup>w</sup>*−<sup>1</sup>

*wi*(*r*) = *kir<sup>c</sup>*

*w*(*r*) = *k*<sup>3</sup>

*<sup>u</sup>*(*r*,*s*) =

Hence, the origin is exponentially stable. The property completes the proof.

�*x*(*t*0)�, *t* − *t*<sup>0</sup>

*(16), where D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*, �*x*� *<sup>&</sup>lt; <sup>r</sup>*}*. Suppose the perturbation term g*(*x*) + *<sup>d</sup>*(*t*) *satisfies*

�*g*(*x*)� ≤ *<sup>c</sup>*4�*x*�, �*d*(*t*)� ≤ *<sup>δ</sup> <sup>&</sup>lt; <sup>ζ</sup>*

Hence, the positive definite and decreasing function *σ*(·, ·) is given by

(12) is satisfied for all {�*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup>

Subsequently, the function *u*(·, ·) is given by

�*x*(*t*)� ≤ *u*

*respectively. Then, for all* �*x*(*t*0)� *< r*

Further, for scalar function

**Remark 5.** *Let w*−<sup>1</sup>

solution.

*and*

Therefore,

given by

and

Also, the set Ω*t*,*<sup>ρ</sup>* is a subset of the ball {�*x*� ≤ *ρ*} since *w*1(�*x*�) ≤ *w*1(*ρ*). Thus

$$\begin{aligned} \left\{ \mathfrak{x} \in \mathbb{R}^n ||\mathfrak{x}|| \le w\_2^{-1}(w\_1(\rho)) \right\} &\subset \Omega\_{t,\rho\_{\mathcal{V}}}\\ \Omega\_{t,\rho} &\subset \{ \mathfrak{x} \in \mathbb{R}^n ||\mathfrak{x}|| \le \rho \} \subset D\_{\mathcal{V}} \end{aligned}$$

for all *t* ≥ 0. For any *t*<sup>0</sup> ≥ 0 and any *x*(*t*0) ∈ Ω*t*,*ρ*, the solution starting at (*t*0, *x*(*t*0)) stays in <sup>Ω</sup>*t*,*<sup>ρ</sup>* for all *<sup>t</sup>*<sup>0</sup> <sup>≥</sup> 0. This follows from the fact that *<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) is negative on *<sup>D</sup>* − {0}; hence *V*(*t*, *x*) is decreasing. Therefore, the solution starting at (*t*0, *x*(*t*0)) is defined for all *t* ≥ *t*<sup>0</sup> and *<sup>x</sup>*(*t*) <sup>∈</sup> *<sup>D</sup>*. Now, it will assume that �*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*)) . Then

$$\dot{V}(t, x) \le -w\_3\left(||\mathfrak{x}(t\_0)||\right) \le -w\_3\left(w\_2^{-1}(w\_1(\rho))\right).$$

Let *y*(*t*) satisfy the auxiliary autonomous first order differential equation

$$
\dot{y} = w(y)\_\prime \ y(t\_0) = V(t\_0, x(t\_0)) \ge 0.
$$

It is clear that

$$V(t, \mathfrak{x}(t)) \le y(t), \ \forall t \ge t\_0.$$

By corollary 6, there exists a positive definite and decreasing function *σ*(*r*,*s*) such that

$$V(t, \mathbf{x}(t)) \le \sigma(V(t\_0, \mathbf{x}(t\_0)), t - t\_0), \quad \forall t \ge t\_0.$$

Therefore, any solution starting in Ω*t*,*<sup>ρ</sup>* satisfies the inequality

$$\begin{aligned} w\_1(||\mathfrak{x}(t)||) &\le V(t, \mathfrak{x}(t)), \\ ||\mathfrak{x}(t)|| &\le w\_1^{-1}(V(t, \mathfrak{x}(t))), \\ ||\mathfrak{x}(t)|| &\le w\_1^{-1}(\sigma(V(t\_0, \mathfrak{x}(t\_0)), t - t\_0)), \\ ||\mathfrak{x}(t)|| &\le w\_1^{-1}(\sigma(w\_2(||\mathfrak{x}(t\_0)||), t - t\_0)), \\ ||\mathfrak{x}(t)|| &\le u(||\mathfrak{x}(t\_0)||, t - t\_0). \end{aligned}$$

Since *μ < w*−<sup>1</sup> <sup>2</sup> (*w*1(*r*)), we can choose *<sup>ρ</sup> <sup>&</sup>lt; <sup>r</sup>* such that *<sup>μ</sup> <sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*)). Furthermore, for any �*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*r*)), we can choose *ρ* close enough to *r* such that �*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*)). Let *<sup>η</sup>* <sup>=</sup> *<sup>w</sup>*−<sup>1</sup> <sup>1</sup> (*w*2(*μ*)). Then

$$B\_{\mu} \subset \Omega\_{t,\eta} \subset B\_{\eta} \subset B\_{\rho} \subset D\_{\prime}$$

and

$$
\Omega\_{\mathfrak{t},\eta} \subset \Omega\_{\mathfrak{t},\rho} \subset B\_{\rho} \subset D.
$$

The sets Ω*t*,*<sup>ρ</sup>* and Ω*t*,*<sup>η</sup>* have the property that a solution starting inside either set cannot leave it because *<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) is negative on the boundary. Therefore, if �*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*)), the solution *x*(*t*) will belong to Ω*t*,*<sup>ρ</sup>* for all *t* ≥ *t*0. For a solution starting inside Ω*t*,*η*, the inequality (18) is satisfied for all *t* ≥ *t*0. For a solution starting inside Ω*t*,*ρ*, but outside Ω*t*,*η*, let *t*<sup>1</sup> be the first time it enters Ω*t*,*η*. This time *t*<sup>1</sup> could be *t*<sup>0</sup> (if the solution starts inside Ω*t*,*η*) or infinite (if it never enters Ω*t*,*η*). Since *u*(�*x*(*t*0)�, *t* − *t*0) → 0 as *t* → ∞, there is a finite time after which *u*(�*x*(*t*0)�, *t* − *t*0) *< μ* for all *t*. Therefore, the time *t*<sup>1</sup> must be finite; that is, the solution must enter the set Ω*t*,*<sup>η</sup>* in finite time. Once inside the set, the solution remains inside for all *t* ≥ *t*1. Therefore,

$$V(t, \mathfrak{x}(t)) \le w\_1(\eta)\_{\prime} \,\,\,\forall t \ge t\_{1\prime}$$

and

10 Will-be-set-by-IN-TECH

<sup>2</sup> (*w*1(*ρ*))

<sup>Ω</sup>*t*,*<sup>ρ</sup>* <sup>⊂</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*|�*x*� ≤ *<sup>ρ</sup>*} <sup>⊂</sup> *<sup>D</sup>*,

for all *t* ≥ 0. For any *t*<sup>0</sup> ≥ 0 and any *x*(*t*0) ∈ Ω*t*,*ρ*, the solution starting at (*t*0, *x*(*t*0)) stays in <sup>Ω</sup>*t*,*<sup>ρ</sup>* for all *<sup>t</sup>*<sup>0</sup> <sup>≥</sup> 0. This follows from the fact that *<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) is negative on *<sup>D</sup>* − {0}; hence *V*(*t*, *x*) is decreasing. Therefore, the solution starting at (*t*0, *x*(*t*0)) is defined for all *t* ≥ *t*<sup>0</sup> and

*y*˙ = *w*(*y*), *y*(*t*0) = *V*(*t*0, *x*(*t*0)) ≥ 0.

*V*(*t*, *x*(*t*)) ≤ *y*(*t*), ∀*t* ≥ *t*0.

*V*(*t*, *x*(*t*)) ≤ *σ*(*V*(*t*0, *x*(*t*0)), *t* − *t*0), ∀*t* ≥ *t*0.

�*x*(*t*)� ≤ *<sup>w</sup>*−<sup>1</sup>

*w*1(�*x*(*t*)�) ≤ *V*(*t*, *x*(*t*)),

<sup>1</sup> (*σ*(*V*(*t*0, *x*(*t*0)), *t* − *t*0)),

<sup>1</sup> (*σ*(*w*2(�*x*(*t*0)�), *t* − *t*0)), �*x*(*t*)� ≤ *u*(�*x*(*t*0)�, *t* − *t*0).

<sup>1</sup> (*V*(*t*, *x*(*t*))),

<sup>2</sup> (*w*1(*ρ*)).

<sup>2</sup> (*w*1(*ρ*)), the solution

<sup>2</sup> (*w*1(*r*)), we can choose *ρ* close enough to *r* such that

By corollary 6, there exists a positive definite and decreasing function *σ*(*r*,*s*) such that

�*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup>

<sup>2</sup> (*w*1(*ρ*))

 *w*−<sup>1</sup>

⊂ Ω*t*,*ρ*,

 . Then

<sup>2</sup> (*w*1(*ρ*))

 .

Also, the set Ω*t*,*<sup>ρ</sup>* is a subset of the ball {�*x*� ≤ *ρ*} since *w*1(�*x*�) ≤ *w*1(*ρ*). Thus

*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*|�*x*� ≤ *<sup>w</sup>*−<sup>1</sup>

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −*w*<sup>3</sup> (�*x*(*t*0)�) ≤ −*w*<sup>3</sup>

Let *y*(*t*) satisfy the auxiliary autonomous first order differential equation

Therefore, any solution starting in Ω*t*,*<sup>ρ</sup>* satisfies the inequality

�*x*(*t*)� ≤ *<sup>w</sup>*−<sup>1</sup>

�*x*(*t*)� ≤ *<sup>w</sup>*−<sup>1</sup>

<sup>2</sup> (*w*1(*r*)), we can choose *<sup>ρ</sup> <sup>&</sup>lt; <sup>r</sup>* such that *<sup>μ</sup> <sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup>

because *<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) is negative on the boundary. Therefore, if �*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup>

<sup>1</sup> (*w*2(*μ*)). Then

*B<sup>μ</sup>* ⊂ Ω*t*,*<sup>η</sup>* ⊂ *B<sup>η</sup>* ⊂ *B<sup>ρ</sup>* ⊂ *D*,

Ω*t*,*<sup>η</sup>* ⊂ Ω*t*,*<sup>ρ</sup>* ⊂ *B<sup>ρ</sup>* ⊂ *D*. The sets Ω*t*,*<sup>ρ</sup>* and Ω*t*,*<sup>η</sup>* have the property that a solution starting inside either set cannot leave it

*x*(*t*) will belong to Ω*t*,*<sup>ρ</sup>* for all *t* ≥ *t*0. For a solution starting inside Ω*t*,*η*, the inequality (18) is satisfied for all *t* ≥ *t*0. For a solution starting inside Ω*t*,*ρ*, but outside Ω*t*,*η*, let *t*<sup>1</sup> be the first time it enters Ω*t*,*η*. This time *t*<sup>1</sup> could be *t*<sup>0</sup> (if the solution starts inside Ω*t*,*η*) or infinite (if it never enters Ω*t*,*η*). Since *u*(�*x*(*t*0)�, *t* − *t*0) → 0 as *t* → ∞, there is a finite time after which *u*(�*x*(*t*0)�, *t* − *t*0) *< μ* for all *t*. Therefore, the time *t*<sup>1</sup> must be finite; that is, the solution must

*x*(*t*) ∈ *D*. Now, it will assume that

It is clear that

Since *μ < w*−<sup>1</sup>

�*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup>

and

Furthermore, for any �*x*(*t*0)� *<sup>&</sup>lt; <sup>w</sup>*−<sup>1</sup>

<sup>2</sup> (*w*1(*ρ*)). Let *<sup>η</sup>* <sup>=</sup> *<sup>w</sup>*−<sup>1</sup>

$$\|\mathfrak{x}(t)\| \le \eta\_{\prime} \quad \forall t \ge t\_1.$$

Hence, any initial state *<sup>x</sup>*(*t*0) can be included in the set {�*x*� ≤ *<sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*))}. Thus, inequality (12) is satisfied for all {�*x*(*t*0)� ≤ *<sup>w</sup>*−<sup>1</sup> <sup>2</sup> (*w*1(*ρ*))}, which implies that the origin *x* = 0 is uniformly asymptotically stable. The exponentially decaying for *w*1(·), *w*2(·) and *w*3(·) is given by

$$w\_i(r) = k\_i r^c, \ k\_i > 0, \ c > 0, \ i = 1, 2, 3.$$

Further, for scalar function

$$w(r) = k\_3 \left[ \left(\frac{r}{k\_2}\right)^{1/\varepsilon} \right]^c = \frac{k\_3}{k\_2} r.$$

Hence, the positive definite and decreasing function *σ*(·, ·) is given by

$$
\sigma(r, s) = r e^{-(k\_3/k\_2)s}.
$$

Subsequently, the function *u*(·, ·) is given by

$$u(r,s) = \left[\frac{k\_2 r^c e^{-(k\_3/k\_2)s}}{k\_1}\right]^{1/c}$$

Hence, the origin is exponentially stable. The property completes the proof.

Inequalities (17) and (18) show that the solution *x*(*t*) is uniformly bounded for all *t* ≥ *t*0.

**Remark 5.** *Let w*−<sup>1</sup> <sup>1</sup> (*w*2(*μ*)) *be a positive definite function called the bound of μ. As μ* → 0*, the bound approaches zero. Sometimes, it is possible to combine inequalities (17) and (18) in one inequality*

$$\|\mathbf{x}(t)\| \le \mathfrak{u}\left(\|\mathbf{x}(t\_0)\|, t - t\_0\right) + \mathfrak{w}\_1^{-1}(\mathfrak{w}\_2(\mu)), \ \forall t \ge t\_0. \tag{19}$$

.

Now, let us illustrate how Theorem 1 is used in the analysis of the perturbed system (2), when the origin of the nominal system is exponentially stable and the system has a uniform bounded solution.

**Lemma 3.** *Let x* = 0 *be an exponentially stable equilibrium point of the nominal system (8). Let V* : [0, ∞) × *D* → *R be a Lyapunov function of the nominal system that satisfies inequalities (15) and (16), where D* <sup>=</sup> {*<sup>x</sup>* <sup>∈</sup> *<sup>R</sup>n*, �*x*� *<sup>&</sup>lt; <sup>r</sup>*}*. Suppose the perturbation term g*(*x*) + *<sup>d</sup>*(*t*) *satisfies*

$$\|\|g(\mathbf{x})\|\| \le c\_4 \|\mathbf{x}\|\| \nu \quad \|d(t)\|\| \le \delta < \frac{\zeta}{c\_5} r\theta\sqrt{\frac{c\_1}{c\_2}}\nu$$

*for all t* ≥ 0*, x*(*t*) ∈ *D, and some positive constants* 0 *< θ <* 1*,* 0 *< ζ <* 1*, c*<sup>2</sup> *>* 0*, c*<sup>4</sup> *>* 0*, c*<sup>5</sup> *>* 0 *respectively. Then, for all* �*x*(*t*0)� *< r* <sup>√</sup>*c*1/*c*2*, the solution of the perturbed system x*(*t*) *satisfies*

$$\|\mathbf{x}(t)\| \le k \|\mathbf{x}(t\_0)\| \|\exp(-\alpha(t - t\_0)), \ \forall t\_0 \le t < t\_1.$$

*and*

$$\|\|\mathbf{x}(t)\|\| \le b\_{\prime} \quad \forall t \ge t\_{1\prime}$$

*for some finite time t*1*, where*

$$\begin{aligned} k &= \sqrt{\frac{c\_2}{c\_1}}, & a &= \frac{(1-\theta)\overline{\zeta}}{2c\_2}, & b &= \frac{c\_5}{\overline{\zeta}}\frac{\delta}{\theta}k, \\ c\_1 &= \lambda\_{\min}(P), & c\_2 &= \lambda\_{\max}(P), \\ c\_3 &= \lambda\_{\min}(Q), & c\_4 &\le c\_3 - \overline{\zeta}. \end{aligned}$$

*Proof.* Consider *V*(*t*, *x*) as a Lyapunov function candidate. The derivative of *V*(*t*, *x*) along the trajectories of (2) satisfies

$$\dot{V}(t,x) \le -c\_3 \|\mathbf{x}\|\_2^2 + \left\|\frac{\partial V}{\partial \mathbf{x}}\right\|\_2 \|g(\mathbf{x})\|\_2 + \left\|\frac{\partial V}{\partial \mathbf{x}}\right\|\_2 \|d(t)\|\_2,$$

$$\dot{V}(t,\mathbf{x}) \le -c\_3 \|\mathbf{x}\|\_2^2 + c\_4 \|\mathbf{x}\|\_2^2 + c\_5 \delta \|\mathbf{x}\|\_2,$$

$$\dot{V}(t,\mathbf{x}) \le -(c\_3 - c\_4 - \zeta) \|\mathbf{x}\|\_2^2 - \zeta \|\mathbf{x}\|\_2^2 + c\_5 \delta \|\mathbf{x}\|\_2,$$

$$\dot{V}(t,\mathbf{x}) \le -\zeta \|\mathbf{x}\|\_2^2 + c\_5 \delta \|\mathbf{x}\|\_2, \ 0 < \zeta < 1,$$

$$\dot{V}(t,\mathbf{x}) \le -(1 - \theta)\zeta \|\mathbf{x}\|\_2^2 - \theta\zeta \|\mathbf{x}\|\_2^2 + c\_5 \delta \|\mathbf{x}\|\_2^2,$$

$$\dot{V}(t,\mathbf{x}) \le -(1 - \theta)\zeta \|\mathbf{x}\|\_{2'}^2 \ 0 < \theta < 1, \ \forall \|\mathbf{x}\|\_2 \ge \delta c\_5 / \delta\zeta.$$

By following application of theorem 1 completes the proof.

The bound *b* is proportional to the upper bound on the perturbation *δ*. Once again, this result can be viewed as a robustness property of nominal system having exponentially uniform equilibria at the origin, because it shows that arbitrarily small (uniformly bounded) perturbations, will not result in large steady-state derivations from the origin.

#### **3. HIV infection model approximation: Third order ODE**

Consider the following model for HIV infection that involves a 3rd order ODE (Barao & Lemos, 2007), (Perelson & Nelson, 1999), and (Santos & Middleton, 2008)

$$\begin{aligned} \frac{dT}{dt} &= \mathbf{s} - d\_{\mathsf{T}}T - \beta TV, \\ \frac{d\tilde{T}}{dt} &= \beta TV - \delta \tilde{T}, \\ \frac{dV}{dt} &= p\tilde{T} - cV, \end{aligned} \tag{20}$$

where *T* denotes the concentration of uninfected target cells (specially, CD4+helper T cells), *T*˜ is the concentration of infected target cells and *V* denotes the concentration of virions. There are two equilibrium points for the system given in (20). One of these is termed the uninfected state and is given by

$$T = \frac{s}{d\_T}, \quad \tilde{T} = 0, \ V = 0. \tag{21}$$

The other equilibrium is termed the infected state and is given by

$$T = \frac{\delta c}{\beta p'}, \ \tilde{T} = \frac{s}{\delta} - \frac{c d\_{\mathsf{T}}}{\beta p'}, \ V = \frac{ps}{\delta c} - \frac{d\_{\mathsf{T}}}{\beta}. \tag{22}$$

**Parameter Description Value/units**

*s* uninfected cells 10mm−<sup>3</sup> per day

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 231

*<sup>β</sup>* free virus particles 2.4×10−<sup>5</sup> mm−<sup>3</sup> per day

*dT* uninfected cells 0.02 per day

*δ* infected cell 0.24 per day

*p* produced per 100 per day

*c* free particle 2.4 per day

On reference (Santos & Middleton, 2008), both equilibrium points (21) and (22) were studied. The uninfected state (21), see parameter values on Table 1, is an unstable equilibrium, where even a small perturbation (e.g introduction of HIV virus to system's dynamic) leads to divergence. For infected state in (22), it is concluded that the infected equilibrium is locally stable for the parameter values given on Table 1. The qualitative behavior of a non linear system near an equilibrium point can be determined via linearisation (Khalil, 2002). The

> −(*dT* + *βV*) 0 −*βT βV* −*δ βT* 0 *p* −*c*

<sup>|</sup>*λ<sup>I</sup>* <sup>−</sup> *<sup>A</sup>*<sup>|</sup> <sup>=</sup> *<sup>λ</sup>*<sup>3</sup> <sup>+</sup> *<sup>a</sup>*1*λ*<sup>2</sup> <sup>+</sup> *<sup>a</sup>*2*<sup>λ</sup>* <sup>+</sup> *<sup>a</sup>*<sup>3</sup> (24)

⎤

⎦ , (23)

*t* Time days

Source term for

Death rate of

Infection rate of

Death rate of

Rate of virions

infected cells Death rate of

virions

system can be linearised by computing the Jacobian which for (20) is given by

= ⎡ ⎣

uniformly bounded estimation for the equilibria point behavior at perturbation.

where *A* is a stability matrix for the evaluation of infected state given in (22) that leads to the

Now, it is necessary to study the dynamics for the perturbation and to determine the extent of stability region to know how large a perturbation from the equilibrium can be allowed and it

*<sup>δ</sup><sup>c</sup>* (*c* + *δ*), and *a*<sup>3</sup> = (*βsp* − *δcdT*). From (24), the Hurwitz stability conditions are *ak >* 0, for *k* = 1, 2, 3 and *a*1*a*<sup>2</sup> − *a*<sup>3</sup> *>* 0. This stability domain is very conservative, because of the local behavior about the equilibrium point in (22). In next subsection, for equilibrium point in (22), we need to probe the exponentially uniform stable property and describe boundedness of the region of attraction. We propose to work with the converse Lyapunov stability analysis in order to obtain a

*∂x x*=[*T*,*T*˜,*V*]

, *<sup>a</sup>*<sup>2</sup> <sup>=</sup> *<sup>β</sup>sp*

**3.1 Linearisation on infected and uninfected equilibrium point**

*<sup>A</sup>* <sup>=</sup> *<sup>∂</sup> <sup>f</sup>*

Table 1. Parameters for HIV model

characteristic polynomial

�

*c* + *δ* + *<sup>β</sup>sp δc* �

**3.2 Non autonomous perturbation analysis**

where *a*<sup>1</sup> =


Table 1. Parameters for HIV model

12 Will-be-set-by-IN-TECH

, *<sup>α</sup>* <sup>=</sup> (<sup>1</sup> <sup>−</sup> *<sup>θ</sup>*)*<sup>ζ</sup>* 2*c*<sup>2</sup>

*Proof.* Consider *V*(*t*, *x*) as a Lyapunov function candidate. The derivative of *V*(*t*, *x*) along the

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −*c*3�*x*�<sup>2</sup>

The bound *b* is proportional to the upper bound on the perturbation *δ*. Once again, this result can be viewed as a robustness property of nominal system having exponentially uniform equilibria at the origin, because it shows that arbitrarily small (uniformly bounded)

Consider the following model for HIV infection that involves a 3rd order ODE (Barao &

*dt* <sup>=</sup> *<sup>s</sup>* <sup>−</sup> *<sup>d</sup>τ<sup>T</sup>* <sup>−</sup> *<sup>β</sup>TV*,

where *T* denotes the concentration of uninfected target cells (specially, CD4+helper T cells), *T*˜ is the concentration of infected target cells and *V* denotes the concentration of virions. There are two equilibrium points for the system given in (20). One of these is termed the uninfected

*<sup>δ</sup>* <sup>−</sup> *cd<sup>τ</sup>*

*<sup>β</sup><sup>p</sup>* , *<sup>V</sup>* <sup>=</sup> *ps*

*<sup>δ</sup><sup>c</sup>* <sup>−</sup> *<sup>d</sup><sup>τ</sup>*

*dt* <sup>=</sup> *<sup>β</sup>TV* <sup>−</sup> *<sup>δ</sup>T*˜,

*dt* <sup>=</sup> *pT*˜ <sup>−</sup> *cV*,

*dT*˜

*dV*

�*g*(*x*)�<sup>2</sup> +

*c*<sup>1</sup> = *λmin*(*P*), *c*<sup>2</sup> = *λmax*(*P*), *c*<sup>3</sup> = *λmin*(*Q*), *c*<sup>4</sup> ≤ *c*<sup>3</sup> − *ζ*.

> <sup>2</sup> + *∂V ∂x* 2

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −(*c*<sup>3</sup> <sup>−</sup> *<sup>c</sup>*<sup>4</sup> <sup>−</sup> *<sup>ζ</sup>*)�*x*�<sup>2</sup>

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −*ζ*�*x*�<sup>2</sup>

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −(<sup>1</sup> <sup>−</sup> *<sup>θ</sup>*)*ζ*�*x*�<sup>2</sup>

perturbations, will not result in large steady-state derivations from the origin.

Lemos, 2007), (Perelson & Nelson, 1999), and (Santos & Middleton, 2008) *dT*

> *<sup>T</sup>* <sup>=</sup> *<sup>s</sup> dT*

, *<sup>T</sup>*˜ <sup>=</sup> *<sup>s</sup>*

The other equilibrium is termed the infected state and is given by

*<sup>T</sup>* <sup>=</sup> *<sup>δ</sup><sup>c</sup> βp* , *<sup>b</sup>* <sup>=</sup> *<sup>c</sup>*<sup>5</sup> *ζ δ θ k*,

> *∂V ∂x* 2

<sup>2</sup> <sup>+</sup> *<sup>c</sup>*4�*x*�<sup>2</sup>

<sup>2</sup> <sup>−</sup> *<sup>ζ</sup>*�*x*�<sup>2</sup>

<sup>2</sup> <sup>−</sup> *θζ*�*x*�<sup>2</sup>

<sup>2</sup>, 0 *< θ <* 1, ∀�*x*�<sup>2</sup> ≥ *δc*5/*θζ*.

<sup>2</sup> + *c*5*δ*�*x*�2, 0 *< ζ <* 1,

�*d*(*t*)�2,

<sup>2</sup> + *c*5*δ*�*x*�2,

<sup>2</sup> + *c*5*δ*�*x*�2,

<sup>2</sup> + *c*5*δ*�*x*�2,

, *T*˜ = 0, *V* = 0. (21)

*<sup>β</sup>* . (22)

(20)

*for some finite time t*1*, where*

trajectories of (2) satisfies

state and is given by

*k* =

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −*c*3�*x*�<sup>2</sup>

*<sup>V</sup>*˙ (*t*, *<sup>x</sup>*) ≤ −(<sup>1</sup> <sup>−</sup> *<sup>θ</sup>*)*ζ*�*x*�<sup>2</sup>

**3. HIV infection model approximation: Third order ODE**

By following application of theorem 1 completes the proof.

*c*<sup>2</sup> *c*1

#### **3.1 Linearisation on infected and uninfected equilibrium point**

On reference (Santos & Middleton, 2008), both equilibrium points (21) and (22) were studied. The uninfected state (21), see parameter values on Table 1, is an unstable equilibrium, where even a small perturbation (e.g introduction of HIV virus to system's dynamic) leads to divergence. For infected state in (22), it is concluded that the infected equilibrium is locally stable for the parameter values given on Table 1. The qualitative behavior of a non linear system near an equilibrium point can be determined via linearisation (Khalil, 2002). The system can be linearised by computing the Jacobian which for (20) is given by

$$A = \frac{\partial f}{\partial \mathbf{x}}\bigg|\_{\mathbf{x} = [T, \mathbf{\mathcal{T}}, V]} = \begin{bmatrix} -(d\_T + \beta V) & 0 & -\beta T \\ \beta V & -\delta & \beta T \\ 0 & p & -c \end{bmatrix} \tag{23}$$

where *A* is a stability matrix for the evaluation of infected state given in (22) that leads to the characteristic polynomial

$$|\lambda I - A| = \lambda^3 + a\_1 \lambda^2 + a\_2 \lambda + a\_3 \tag{24}$$

where *a*<sup>1</sup> = � *c* + *δ* + *<sup>β</sup>sp δc* � , *<sup>a</sup>*<sup>2</sup> <sup>=</sup> *<sup>β</sup>sp <sup>δ</sup><sup>c</sup>* (*c* + *δ*), and *a*<sup>3</sup> = (*βsp* − *δcdT*).

From (24), the Hurwitz stability conditions are *ak >* 0, for *k* = 1, 2, 3 and *a*1*a*<sup>2</sup> − *a*<sup>3</sup> *>* 0. This stability domain is very conservative, because of the local behavior about the equilibrium point in (22). In next subsection, for equilibrium point in (22), we need to probe the exponentially uniform stable property and describe boundedness of the region of attraction. We propose to work with the converse Lyapunov stability analysis in order to obtain a uniformly bounded estimation for the equilibria point behavior at perturbation.

#### **3.2 Non autonomous perturbation analysis**

Now, it is necessary to study the dynamics for the perturbation and to determine the extent of stability region to know how large a perturbation from the equilibrium can be allowed and it can still be sure that the solution remains toward the equilibrium (Hahn, 1967), (Khalil, 2002). Consider the following change of variables in (20)

$$\mathbf{x} = [T, \vec{T}, V]^T = [\mathbf{x}\_1, \mathbf{x}\_2, \mathbf{x}\_3]^T.$$

The HIV model equations (20) can be rewritten as a perturbed model

$$\begin{aligned} \dot{\mathbf{x}}\_1 &= \mathbf{s} - d\_T \mathbf{x}\_1 - \beta \mathbf{x}\_1 \mathbf{x}\_3 = -d\_T \mathbf{x}\_1 + (\mathbf{s} - \beta \mathbf{x}\_1 \mathbf{x}\_3), \\ \dot{\mathbf{x}}\_2 &= \beta \mathbf{x}\_1 \mathbf{x}\_3 - \delta \mathbf{x}\_2 = -\delta \mathbf{x}\_2 + (\beta \mathbf{x}\_1 \mathbf{x}\_3), \\ \dot{\mathbf{x}}\_3 &= p \mathbf{x}\_2 - c \mathbf{x}\_3. \end{aligned} \tag{25}$$

The compact form of (25) is

$$
\dot{\mathbf{x}} = f(\mathbf{x}) + \mathbf{g}(\mathbf{x}) + d(t), \tag{26}
$$

It is possible to find the unique solution with matrix *P* = *P<sup>T</sup> >* 0 be positive definite. By

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 233

25 0 0 0 3.2902 <sup>×</sup> <sup>10</sup><sup>3</sup> 7.8914 0 7.8914 0.2083

The candidate Lyapunov function *V*(*x*) = *xTPx* needs to satisfy the following four conditions,

**Remark 6.** *A candidate Lyapunov function V*(*x*) *is used to investigate for the nominal system and its stable or asymptotically stable equilibrium point at the origin, and determine if perturbed system (26)*

*<sup>∂</sup><sup>x</sup> <sup>f</sup>*(*x*) + *<sup>∂</sup><sup>V</sup>*

where *f*(*x*) is a function which describes the nominal system in (26), *g*(*x*) is a function which describes the perturbation about the origin in (26) and satisfy the growing bound given in (28) and *d*(*t*) is a function for a bounded perturbation in (26). Hence, by using the bound (28)

�*g*(*x*)�<sup>2</sup> +

<sup>2</sup> *<sup>β</sup>*�*x*�<sup>3</sup>

*θζ* , *<sup>M</sup>*

�

6.5804 <sup>×</sup> <sup>10</sup><sup>3</sup>

*<sup>V</sup>*˙ (*x*) <sup>≤</sup> (<sup>1</sup> <sup>−</sup> *<sup>θ</sup>*)*ζ*�*x*�<sup>2</sup>

<sup>2</sup> <sup>+</sup> <sup>50</sup>*s*�*x*�<sup>2</sup>

Then, the function *V*˙ (*x*) will be negative definite if the following conditions are satisfied

∀�*x*�<sup>2</sup> <sup>≥</sup> min �50*<sup>s</sup>*

*<sup>∂</sup><sup>x</sup> <sup>g</sup>*(*x*) + *<sup>∂</sup><sup>V</sup>*

� � � � *∂V ∂x* � � � � 2

*∂x*

�*d*(*t*)�2,

<sup>2</sup> + 50*s*�*x*�2,

<sup>2</sup>; 0 *< θ <* 1.

, (32)

<sup>2</sup>; 0 *< ζ <* 1, *M >* 0,

*d*(*t*), (31)

By evaluation the derivative of *V*(*x*) along the trajectories of perturbed system (26)

*<sup>x</sup>*˙ <sup>=</sup> *<sup>∂</sup><sup>V</sup>*

given for *g*(*x*), their corresponding results for the function *V*(*x*) are given by

<sup>2</sup> +

<sup>2</sup> + � � � � *∂V ∂x* � � � � 2 ⎤

⎦ . (30)

taking the parameter values in Table 1 the matrix *P* is

<sup>2</sup> <sup>≤</sup> *<sup>V</sup>*(*x*) <sup>≤</sup> *<sup>λ</sup>max*(*P*)�*x*�<sup>2</sup>

*<sup>∂</sup><sup>x</sup>* �<sup>2</sup> <sup>=</sup> �2*xTP*�<sup>2</sup> <sup>≤</sup> <sup>2</sup>�*P*�2�*x*�<sup>2</sup> <sup>≤</sup> <sup>2</sup>*λmax*(*P*)�*x*�2,

*<sup>V</sup>*˙ (*x*) = *<sup>∂</sup><sup>V</sup>*

*<sup>V</sup>*˙ (*x*) ≤ −�*x*�<sup>2</sup>

*<sup>V</sup>*˙ (*x*) ≤ −�*x*�<sup>2</sup>

*<sup>V</sup>*˙ (*x*) ≤ −*ζ*�*x*�<sup>2</sup>

*∂x*

*<sup>λ</sup>min*(*P*) = 0.1894, *<sup>λ</sup>max*(*P*) = 3.2902 <sup>×</sup> <sup>10</sup>3.

*<sup>∂</sup><sup>x</sup> Ax* <sup>=</sup> <sup>−</sup>*xTQx*,

*<sup>∂</sup><sup>x</sup>* �2�*Ax*�<sup>2</sup> ≤ −*λmin*(*Q*)�*x*�<sup>2</sup>

*λmin*(*Q*) = 1, *Q* = *I* = *Q<sup>T</sup> >* 0.

<sup>2</sup>*λmax*(*P*) = 6.5804 <sup>×</sup> 103.

*<sup>∂</sup><sup>x</sup>* �2|*d*(*t*)<sup>|</sup> <sup>=</sup> <sup>|</sup>2*xTP*||*d*(*t*)|,

*<sup>∂</sup><sup>x</sup>* �2|*d*(*t*)| ≤ 2|*P*||*x*||*d*(*t*)|,

*can obtain a uniform bounded value.*

*<sup>i</sup>*=<sup>1</sup> *λi*(*P*)|*xi*||*di*(*t*)| = 0.5|*x*1|.

for being a positive definite scalar function

1. *<sup>λ</sup>min*(*P*)�*x*�<sup>2</sup>

*<sup>∂</sup><sup>x</sup> <sup>f</sup>*(*x*) = *<sup>∂</sup><sup>V</sup>*

2. *<sup>∂</sup><sup>V</sup>*

3. � *<sup>∂</sup><sup>V</sup>*

4. � *<sup>∂</sup><sup>V</sup>*

� *<sup>∂</sup><sup>V</sup>*

<sup>≤</sup> <sup>2</sup> <sup>∑</sup><sup>3</sup>

� *<sup>∂</sup><sup>V</sup>*

*P* =

2,

⎡ ⎣

2,

where *β* ≥ 0 is unknown, *d* is bounded disturbance that satisfies |*d*(*t*)| ≤ *δ*, for *t* ≥ 0, and

$$f(\mathbf{x}) = A\mathbf{x} = \frac{\partial f}{\partial \mathbf{x}}\bigg|\_{\mathbf{x}=0} \mathbf{x} = \begin{bmatrix} -d\_T & 0 & 0\\ 0 & -\delta & 0\\ 0 & p & -c \end{bmatrix} \begin{bmatrix} \mathbf{x}\_1\\ \mathbf{x}\_2\\ \mathbf{x}\_3 \end{bmatrix} \prime$$

$$g(\mathbf{x}) = \begin{bmatrix} -\beta \mathbf{x}\_1 \mathbf{x}\_3\\ \beta \mathbf{x}\_1 \mathbf{x}\_3\\ 0 \end{bmatrix}, \quad d(t) = \begin{bmatrix} s\\ 0\\ 0 \end{bmatrix}. \tag{27}$$

Suppose the perturbation *g*(*x*) satisfies the uniform bound

$$\|\|g(\mathbf{x})\|\|\_{2} \le \beta \begin{bmatrix} |\mathbf{x}\_{1}| |\mathbf{x}\_{3}| \\ |\mathbf{x}\_{1}| |\mathbf{x}\_{3}| \\ \mathbf{0} \end{bmatrix} \le \frac{\beta}{2} \begin{bmatrix} \|\mathbf{x}\|\_{2} \\ \|\mathbf{x}\|\_{2} \\ \mathbf{0} \end{bmatrix}.\tag{28}$$

The linearisation about the origin *x* = 0 for the perturbed system in (20) is described by matrix *A* in (27). The stability analysis of matrix *A* is given by the eigenvalues

$$\begin{aligned} |\lambda I - A| &= \begin{vmatrix} \lambda + d\_T & 0 & 0 \\ 0 & \lambda + \delta & 0 \\ 0 & -p & \lambda + c \end{vmatrix}, \\ |\lambda I - A| &= (\lambda + d\_T)(\lambda + \delta)(\lambda + c). \end{aligned}$$

Matrix *A* is Hurwitz when *dT >* 0, *δ >* 0, *c >* 0.

The converse theorem of Lyapunov is based on linearisation about the origin, *x* = 0. The theorem supposes that matrix *A* is Hurwitz, in other words the nominal system. Then, there exists a candidate Lyapunov function *V*(*x*) = *xTPx*, which permit to analyse the stability by evaluating its derivative along the trajectories of the nominal system (27) such that

$$\dot{V}(\mathbf{x}) = \dot{\mathbf{x}}^T P \mathbf{x} + \mathbf{x}^T P \dot{\mathbf{x}}\_{\prime}$$

$$\dot{V}(\mathbf{x}) = \mathbf{x}^T A^T P \mathbf{x} + \mathbf{x}^T P A \mathbf{x}\_{\prime}$$

$$\dot{V}(\mathbf{x}) = \mathbf{x}^T [A^T P + P A] \mathbf{x}\_{\prime}$$

By solving the Lyapunov equation

$$A^T P + PA = -Q, \ Q = Q^T > 0. \tag{29}$$

It is possible to find the unique solution with matrix *P* = *P<sup>T</sup> >* 0 be positive definite. By taking the parameter values in Table 1 the matrix *P* is

$$P = \begin{bmatrix} 25 & 0 & 0 \\ 0 & 3.2902 \times 10^3 & 7.8914 \\ 0 & 7.8914 & 0.2083 \end{bmatrix}. \tag{30}$$

The candidate Lyapunov function *V*(*x*) = *xTPx* needs to satisfy the following four conditions, for being a positive definite scalar function


14 Will-be-set-by-IN-TECH

can still be sure that the solution remains toward the equilibrium (Hahn, 1967), (Khalil, 2002).

*x*˙1 = *s* − *dTx*<sup>1</sup> − *βx*1*x*<sup>3</sup> = −*dTx*<sup>1</sup> + (*s* − *βx*1*x*3),

where *β* ≥ 0 is unknown, *d* is bounded disturbance that satisfies |*d*(*t*)| ≤ *δ*, for *t* ≥ 0, and

*x* = ⎡ ⎣

⎤


The linearisation about the origin *x* = 0 for the perturbed system in (20) is described by matrix


The converse theorem of Lyapunov is based on linearisation about the origin, *x* = 0. The theorem supposes that matrix *A* is Hurwitz, in other words the nominal system. Then, there exists a candidate Lyapunov function *V*(*x*) = *xTPx*, which permit to analyse the stability by

> *V*˙ (*x*) = *x<sup>T</sup> ATPx* + *xTPAx*, *V*˙ (*x*) = *xT*[*ATP* + *PA*]*x*.

� � � � � �

evaluating its derivative along the trajectories of the nominal system (27) such that

*V*˙ (*x*) = *x*˙

*x*˙2 = *βx*1*x*<sup>3</sup> − *δx*<sup>2</sup> = −*δx*<sup>2</sup> + (*βx*1*x*3),

*∂x x*=0

−*βx*1*x*<sup>3</sup> *βx*1*x*<sup>3</sup> 0

> ⎡ ⎣

⎡ ⎣ *<sup>T</sup>* = [*x*1, *x*2, *x*3]

−*dT* 0 0 0 −*δ* 0 0 *p* −*c*

⎦ , *d*(*t*) =

⎤ <sup>⎦</sup> <sup>≤</sup> *<sup>β</sup>* 2

*λ* + *dT* 0 0 0 *λ* + *δ* 0 0 −*p λ* + *c*

*TPx* + *xTPx*˙,

*T*.

*x*˙ = *f*(*x*) + *g*(*x*) + *d*(*t*), (26)

⎤ ⎦ ,

⎦ . (27)

⎦ . (28)

⎤ ⎦ ⎡ ⎣ *x*1 *x*2 *x*3

⎤

⎤

� � � � � � ,

*<sup>A</sup>TP* <sup>+</sup> *PA* <sup>=</sup> <sup>−</sup>*Q*, *<sup>Q</sup>* <sup>=</sup> *<sup>Q</sup><sup>T</sup> <sup>&</sup>gt;* 0. (29)

⎡ ⎣ *s* 0 0

⎡ ⎣ �*x*�<sup>2</sup> �*x*�<sup>2</sup> 0

(25)

*x* = [*T*, *T*˜, *V*]

The HIV model equations (20) can be rewritten as a perturbed model

*x*˙3 = *px*<sup>2</sup> − *cx*3.

*<sup>f</sup>*(*x*) = *Ax* <sup>=</sup> *<sup>∂</sup> <sup>f</sup>*

*g*(*x*) =

�*g*(*x*)�<sup>2</sup> ≤ *β*

*A* in (27). The stability analysis of matrix *A* is given by the eigenvalues


Suppose the perturbation *g*(*x*) satisfies the uniform bound

Matrix *A* is Hurwitz when *dT >* 0, *δ >* 0, *c >* 0.

By solving the Lyapunov equation

Consider the following change of variables in (20)

The compact form of (25) is


**Remark 6.** *A candidate Lyapunov function V*(*x*) *is used to investigate for the nominal system and its stable or asymptotically stable equilibrium point at the origin, and determine if perturbed system (26) can obtain a uniform bounded value.*

By evaluation the derivative of *V*(*x*) along the trajectories of perturbed system (26)

$$\dot{V}(\mathbf{x}) = \frac{\partial V}{\partial \mathbf{x}}\dot{\mathbf{x}} = \frac{\partial V}{\partial \mathbf{x}}f(\mathbf{x}) + \frac{\partial V}{\partial \mathbf{x}}g(\mathbf{x}) + \frac{\partial V}{\partial \mathbf{x}}d(\mathbf{t}),\tag{31}$$

where *f*(*x*) is a function which describes the nominal system in (26), *g*(*x*) is a function which describes the perturbation about the origin in (26) and satisfy the growing bound given in (28) and *d*(*t*) is a function for a bounded perturbation in (26). Hence, by using the bound (28) given for *g*(*x*), their corresponding results for the function *V*(*x*) are given by

$$\begin{aligned} \dot{V}(\mathbf{x}) \le -||\mathbf{x}||\_2^2 + \left\|\frac{\partial V}{\partial \mathbf{x}}\right\|\_2 ||\mathbf{g}(\mathbf{x})||\_2 + \left\|\frac{\partial V}{\partial \mathbf{x}}\right\|\_2 ||\mathbf{d}(t)||\_{2'} \\ \dot{V}(\mathbf{x}) \le -||\mathbf{x}||\_2^2 + \frac{6.5804 \times 10^3}{2} \beta ||\mathbf{x}||\_2^3 + 50s ||\mathbf{x}||\_{2'} \\ \dot{V}(\mathbf{x}) \le -\xi ||\mathbf{x}||\_2^2 + 50s ||\mathbf{x}||\_2^2; \ 0 < \xi < 1, \ M > 0, \\ \dot{V}(\mathbf{x}) \le (1 - \theta)\xi ||\mathbf{x}||\_2^2; \ 0 < \theta < 1. \end{aligned}$$

Then, the function *V*˙ (*x*) will be negative definite if the following conditions are satisfied

$$\forall \|\|x\|\|\_{2} \ge \min\left\{\frac{50s}{\theta \zeta}, M\right\},\tag{32}$$

where

$$M = \frac{\beta 6.5804 \times 10^3}{2(1 - \zeta)} > 0, \ 0 < \theta < 1, \ 0 < \zeta < 1.$$

Therefore, the function *<sup>V</sup>*˙ (*x*) is negative definite inside the ball �*x*� *<sup>&</sup>lt; <sup>r</sup>* <sup>√</sup>*c*1/*c*2. The ball defines the region of attraction for the solution, when condition (32) is satisfied. It is concluded that the origin *x* = 0 is exponentially uniform stable and the solution for system (20) is uniformly bounded in the large for disturbances that satisfy |*d*(*t*)| ≤ *δ*, for all *t* ≥ 0.

#### **3.3 Simulation of trajectories and the region of attraction**

For the bound given in (32), the following simulations are shown, with initial condition *T*(0) = 520, *T*˜(0) = 0 and *V*(0) = 1. The phase space for system (20) is depicted in Figure 1 for parameter values *<sup>s</sup>* <sup>=</sup> 10, and *<sup>β</sup>* <sup>=</sup> 2.4 <sup>×</sup> <sup>10</sup>−5.

Fig. 2. Region of attraction for: *T*(0) = 520, *T*˜(0) = 0, *V*(0) = 1.

Fig. 3. Time response for three initial conditions: a) *T*(0) = 180.2, *T*˜(0) = 77.6, *V*(0) = 3305;

Exponential Equilibria and Uniform Boundedness of HIV Infection Model 235

point is the same, regardless of the the given initial condition. The proposed change in the values causes a shift in the response for the viral load for variable *V*, without modifying the time response for *T* and *T*˜, see Fig. 3. Also, the possibility of the initial condition (180.2, 0, 500) is worth considering in Fig. 1,b) and Fig. 3. That represents the beginning of an infection with a high viral load. This generates a more benign transient concerning the viral load dynamics. The aim of plotting the trajectories generated from different initial conditions is to depict the dynamics generated by the three state HIV model. Remember that the perturbation term *d*(*t*)

In the paper of (Barao & Lemos, 2007), the study is made for 3rd order ODE, which focus on the analysis of eigenvalues resulting from linearisation around the equilibrium points (Santos

b) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 500; c) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 1.

is constant.

**4. Conclusions**

Fig. 1. Phase space for: *<sup>s</sup>* <sup>=</sup> 10, *<sup>β</sup>* <sup>=</sup> 2.4 <sup>×</sup> <sup>10</sup>−<sup>5</sup> with three initial conditions (order from above to bottom): a) *T*(0) = 180.2, *T*˜(0) = 77.6, *V*(0) = 3305; b) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 500; c) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 1.

The result describing the region of attraction is useful for the clinical personal studying HIV behavior, since it allows to predict the infection development and then choose treatment options. This model does not describe the infection behavior when AIDS has already developed. The region of attraction describes the zone for which, given any initial state condition within it, its future dynamics will be particularly slow, i.e. exponentially uniform. In the positive sector (*xi >* 0, *i* = 1, 2, 3) of the trajectory space, the solutions will be exponentially uniform, but two different types of conditions are analysed: those starting outside and those starting inside of the domain. The latter are from an invariant set. In Fig. 2, the trajectory corresponds to initial condition given in the paper (Barao & Lemos, 2007). That trajectory belongs to solution with fast dynamics that becomes slow as soon as it traverses the invariant set.

One point which is closer to conditions found in reality is the point (180.2, 0, 1), which is depicted in Fig. 1, c). Here, the viral load is small, but the number of uninfected cells is zero. This makes us think of an HIV-infected patient who is not having a large viral load. This information may be useful to configure control law that locate the starting condition at a point such that no large viral load is generated. It is important to keep in mind that the attraction 16 Will-be-set-by-IN-TECH

defines the region of attraction for the solution, when condition (32) is satisfied. It is concluded that the origin *x* = 0 is exponentially uniform stable and the solution for system (20) is

For the bound given in (32), the following simulations are shown, with initial condition *T*(0) = 520, *T*˜(0) = 0 and *V*(0) = 1. The phase space for system (20) is depicted in Figure 1 for

Fig. 1. Phase space for: *<sup>s</sup>* <sup>=</sup> 10, *<sup>β</sup>* <sup>=</sup> 2.4 <sup>×</sup> <sup>10</sup>−<sup>5</sup> with three initial conditions (order from above to bottom): a) *T*(0) = 180.2, *T*˜(0) = 77.6, *V*(0) = 3305; b) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 500;

The result describing the region of attraction is useful for the clinical personal studying HIV behavior, since it allows to predict the infection development and then choose treatment options. This model does not describe the infection behavior when AIDS has already developed. The region of attraction describes the zone for which, given any initial state condition within it, its future dynamics will be particularly slow, i.e. exponentially uniform. In the positive sector (*xi >* 0, *i* = 1, 2, 3) of the trajectory space, the solutions will be exponentially uniform, but two different types of conditions are analysed: those starting outside and those starting inside of the domain. The latter are from an invariant set. In Fig. 2, the trajectory corresponds to initial condition given in the paper (Barao & Lemos, 2007). That trajectory belongs to solution with fast dynamics that becomes slow as soon as it traverses the invariant

One point which is closer to conditions found in reality is the point (180.2, 0, 1), which is depicted in Fig. 1, c). Here, the viral load is small, but the number of uninfected cells is zero. This makes us think of an HIV-infected patient who is not having a large viral load. This information may be useful to configure control law that locate the starting condition at a point such that no large viral load is generated. It is important to keep in mind that the attraction

<sup>2</sup>(<sup>1</sup> <sup>−</sup> *<sup>ζ</sup>*) *<sup>&</sup>gt;* 0, 0 *<sup>&</sup>lt; <sup>θ</sup> <sup>&</sup>lt;* 1, 0 *<sup>&</sup>lt; <sup>ζ</sup> <sup>&</sup>lt;* 1.

<sup>√</sup>*c*1/*c*2. The ball

*<sup>M</sup>* <sup>=</sup> *<sup>β</sup>*6.5804 <sup>×</sup> <sup>103</sup>

**3.3 Simulation of trajectories and the region of attraction**

parameter values *<sup>s</sup>* <sup>=</sup> 10, and *<sup>β</sup>* <sup>=</sup> 2.4 <sup>×</sup> <sup>10</sup>−5.

c) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 1.

set.

Therefore, the function *<sup>V</sup>*˙ (*x*) is negative definite inside the ball �*x*� *<sup>&</sup>lt; <sup>r</sup>*

uniformly bounded in the large for disturbances that satisfy |*d*(*t*)| ≤ *δ*, for all *t* ≥ 0.

where

Fig. 2. Region of attraction for: *T*(0) = 520, *T*˜(0) = 0, *V*(0) = 1.

Fig. 3. Time response for three initial conditions: a) *T*(0) = 180.2, *T*˜(0) = 77.6, *V*(0) = 3305; b) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 500; c) *T*(0) = 180.2, *T*˜(0) = 0, *V*(0) = 1.

point is the same, regardless of the the given initial condition. The proposed change in the values causes a shift in the response for the viral load for variable *V*, without modifying the time response for *T* and *T*˜, see Fig. 3. Also, the possibility of the initial condition (180.2, 0, 500) is worth considering in Fig. 1,b) and Fig. 3. That represents the beginning of an infection with a high viral load. This generates a more benign transient concerning the viral load dynamics. The aim of plotting the trajectories generated from different initial conditions is to depict the dynamics generated by the three state HIV model. Remember that the perturbation term *d*(*t*) is constant.

#### **4. Conclusions**

In the paper of (Barao & Lemos, 2007), the study is made for 3rd order ODE, which focus on the analysis of eigenvalues resulting from linearisation around the equilibrium points (Santos & Middleton, 2008). The disadvantage of linearizing about the equilibrium point when it is not the origin, is that the non linearities of the system are not taken into account.

**Part 4** 

**Treatment, Care and Support** 

**of HIV/AIDS Patients** 

Lyapunov converse analysis allows to obtain bounds on the phase space so that the exponential stability of the equilibrium point at the origin is guaranteed.

This means that the system trajectory describes an exponentially uniform trajectory as it approaches to stable equilibrium point. It can be seen that, there are initial conditions which are not within the given sets but their respective trajectories eventually reach the stable equilibrium point.

This dynamic characteristic is studied for the kind of nonlinear system which is studied in this chapter. It must be emphasized that the region of attraction will always determined by the initial conditions and the parameter values. It is also interesting in the future, to study the repulsion region, that means, the region which corresponds to the unstable equilibrium region. Both regions, attraction and repulsion are located in a manifold. The closer the initial condition is to the manifold in which the equilibrium point is located, the less stressful will the patient suffer from the dynamics. That is the main reason to justify the search for manifolds where the uniformly exponentially stable trajectories are found.

## **5. References**


## **Part 4**

## **Treatment, Care and Support of HIV/AIDS Patients**

18 Will-be-set-by-IN-TECH

236 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

& Middleton, 2008). The disadvantage of linearizing about the equilibrium point when it is

Lyapunov converse analysis allows to obtain bounds on the phase space so that the

This means that the system trajectory describes an exponentially uniform trajectory as it approaches to stable equilibrium point. It can be seen that, there are initial conditions which are not within the given sets but their respective trajectories eventually reach the stable

This dynamic characteristic is studied for the kind of nonlinear system which is studied in this chapter. It must be emphasized that the region of attraction will always determined by the initial conditions and the parameter values. It is also interesting in the future, to study the repulsion region, that means, the region which corresponds to the unstable equilibrium region. Both regions, attraction and repulsion are located in a manifold. The closer the initial condition is to the manifold in which the equilibrium point is located, the less stressful will the patient suffer from the dynamics. That is the main reason to justify the search for manifolds

Barao, M. & Lemos, J.M. (2007). Nonlinear control of HIV-1 infection with a singular

Perelson, A.S. & Nelson, P.W. (1999). Mathematical analysis of HIV-1 dynamics in vivo. *SIAM*

Santos-Ferreira, J. & Middleton, R. (2008). A preliminary analysis of HIV infection dynamics, *Proceedings of ISSC-2008*, pp. 1-6, Galway, Rep. of Ireland, (June 2008). Hahn, W. (1967). *Stability of motion*, Springer-Verlag, ISBN 67-23956, New York, USA. Khalil, H.K. (2002). *Nonlinear systems*, Prentice-Hall, ISBN 013-0673-89-7, New Jersey, USA. Krasovskii, N.N. (1967). *Stability of motion*, Stanford University Press, ISBN engrus, Stanford,

perturbation model. *Biomedical signal precessing and control*, Vol. 2, September 2007,

not the origin, is that the non linearities of the system are not taken into account.

exponential stability of the equilibrium point at the origin is guaranteed.

where the uniformly exponentially stable trajectories are found.

*Review*, Vol.41, No.1, 3-44, ISSN 0036-1445.

248–257, ISSN 1746-8094.

equilibrium point.

**5. References**

USA.

**12** 

*Canada* 

**Glycosphingolipids in HIV/AIDS:** 

Clifford A. Lingwood2,3,6 and Donald R. Branch1,2,4,5

*6Research Institute, Hospital for Sick Children, Toronto, Ontario* 

*1Department of Medicine, University of Toronto,* 

*3Department of Biochemistry, University of Toronto,* 

**The Potential Therapeutic Application** 

*2Department of Laboratory Medicine and Pathobiology, University of Toronto,* 

*4Division of Cell and Molecular Biology, Toronto General Research Institute, 5Research and Development, Canadian Blood Services, Toronto, Ontario,* 

It has been 30 years since the original description of acquired immunodeficiency syndrome (AIDS) was first reported (Gottlieb et al., 1981). Since this initial discovery, human immunodeficiency virus (HIV) and HIV pathogenesis ranks near the top as one of the most studied human diseases in the history of medical science. Unfortunately, we are no closer now than we were back in the early 1980s at finding a cure. Although there has been significant progress in treatment, there continues to be an increase in the numbers of

Current dogma says that HIV type 1 (HIV-1) is the most common HIV virus and that it infects immune cells called helper T-cells. HIV can also infect other cells involved in the immune response such as monocytes, macrophages and dendritic cells. The virus has an envelope that mediates its tropism for immune cells. The viral envelope protein gp120 first recognizes and binds to the CD4 molecule located on the cell-surface of CD4+ helper T-cells (Dalgleish et al., 1984). Although researchers insisted for over a decade that CD4 was the only cellular receptor required for HIV to infect cells, some in the research community were sceptical and aware from clinical and laboratory findings that something else must be required for HIV infection. Indeed, in 1996 it was discovered that CD4 does not act alone. Another family of cell-surface receptors, the chemokine receptors, were shown to be required, in conjunction with CD4, for successful infection with HIV-1 (Feng et al., 1996, Alkhatib et al., 1996). These receptors are important for the tropism of the virus. Thus, the CXCR4 chemokine receptor, directs infection of T cells by T cell-tropic HIV-1, whereas the CCR5 and CCR3 chemokine receptors are responsible for infection of monocytes with monocyte-tropic HIV-1 virions. However, as more studies were undertaken, it became clear that many other chemokine receptors could support HIV-1 infection and that the tropism was directed more to the chemokine co-receptor than to the cell type. Thus, HIV viruses that require CXCR4 are now known as X4 HIV-1 while virus that recognizes CCR5, or other members of this family of chemokine receptor, are now known as R5 HIV-1. (Dragic et al.,

infected people and those dying from AIDS throughout the world.

**1. Introduction** 

## **Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application**

Clifford A. Lingwood2,3,6 and Donald R. Branch1,2,4,5 *1Department of Medicine, University of Toronto,* 

*2Department of Laboratory Medicine and Pathobiology, University of Toronto, 3Department of Biochemistry, University of Toronto, 4Division of Cell and Molecular Biology, Toronto General Research Institute, 5Research and Development, Canadian Blood Services, Toronto, Ontario, 6Research Institute, Hospital for Sick Children, Toronto, Ontario Canada* 

## **1. Introduction**

It has been 30 years since the original description of acquired immunodeficiency syndrome (AIDS) was first reported (Gottlieb et al., 1981). Since this initial discovery, human immunodeficiency virus (HIV) and HIV pathogenesis ranks near the top as one of the most studied human diseases in the history of medical science. Unfortunately, we are no closer now than we were back in the early 1980s at finding a cure. Although there has been significant progress in treatment, there continues to be an increase in the numbers of infected people and those dying from AIDS throughout the world.

Current dogma says that HIV type 1 (HIV-1) is the most common HIV virus and that it infects immune cells called helper T-cells. HIV can also infect other cells involved in the immune response such as monocytes, macrophages and dendritic cells. The virus has an envelope that mediates its tropism for immune cells. The viral envelope protein gp120 first recognizes and binds to the CD4 molecule located on the cell-surface of CD4+ helper T-cells (Dalgleish et al., 1984). Although researchers insisted for over a decade that CD4 was the only cellular receptor required for HIV to infect cells, some in the research community were sceptical and aware from clinical and laboratory findings that something else must be required for HIV infection. Indeed, in 1996 it was discovered that CD4 does not act alone. Another family of cell-surface receptors, the chemokine receptors, were shown to be required, in conjunction with CD4, for successful infection with HIV-1 (Feng et al., 1996, Alkhatib et al., 1996). These receptors are important for the tropism of the virus. Thus, the CXCR4 chemokine receptor, directs infection of T cells by T cell-tropic HIV-1, whereas the CCR5 and CCR3 chemokine receptors are responsible for infection of monocytes with monocyte-tropic HIV-1 virions. However, as more studies were undertaken, it became clear that many other chemokine receptors could support HIV-1 infection and that the tropism was directed more to the chemokine co-receptor than to the cell type. Thus, HIV viruses that require CXCR4 are now known as X4 HIV-1 while virus that recognizes CCR5, or other members of this family of chemokine receptor, are now known as R5 HIV-1. (Dragic et al.,

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 241

The hydrophilic core sugar sequence defines the carbohydrate moieties, and these protrude into the extracellular space (Stults et al., 1989). The different moieties comprising GSLs have different roles in these multifunctional membrane lipids. GSLs are classified as neutral, acidic (anionic) and basic (cationic) glycolipids (Hakomori, 1986). Acidic GSLs usually contain either a sialic acid, which largely encompass the gangliosides, or a sulphate group. Basic glycolipids are very rare, but include plasmalopsychosine or glyceroplasmalopsychosine (Hikita et al., 2002). For the most part, GSLs are comprised of four groups, characterised by their basic core structure: globo- (defined by Gal1-4 Gal), lacto- (Gal1-3GlcNAc), neolacto- (Gal1-4GlcNAc)

The ceramide backbone of GSLs is synthesized on the cytosolic leaflet of the rough ER, through condensation of L-serine and fatty-acyl co-enzymeA and subsequent enzymatic modification (Huwiler et al., 1998). Ceramide may be converted in the lumen of the ER into gala-series glycolipids by addition of galactose, via -glycosidic linkage, producing the first in the series, galactosylceramide (GalCer) (Sprong et al., 1998). The addition of a sulphate group to the 3-position of the sugar residue on GalCer will give rise to sulphatide (SGC). Ceramide may alternatively be transported to the Golgi apparatus where the first sugar added is glucose, via -glycosidic link, producing glucosylceramide (GlcCer). The precursor for most GSL structures is GlcCer, which is synthesised by glucosyltransferase located in the cytosol (Futerman & Pagano, 1991). GlcCer can be then translocated by the flippase function of the drug resistance pump, P-glycoprotein (P-gp), to the Golgi lumen (De Rosa et al., 2004, Lala et al., 2000). Here subsequent synthesis of all other GSLs takes place through highly specific glycosyltransferases (Lannert et al., 1998). The first product that is formed from GlcCer is lactosyl ceramide (Gal1-4Glc cer, LC), which can then be sialylated, to give monosialoganglioside (sialic acid 2-3 Gal1-4Glc cer, GM3). Alternatively, LC is galactosylated to form globotriaosyl ceramide (Gal1-4 Gal1-4Glc cer, Gb3), which can be further converted to globotetraosyl ceramide (GalNAc1-3 Gal1-4 Gal1-4Glc cer, Gb4) (Figure 2). The major GSLs contain ~5 sugars or less although GSLs containing over 60 sugar residues have been described (Miller-Podraza et al., 1993), and more than 400 GSL species have been reported (Hakomori, 2008). Newly synthesised GSLs follow anterograde vesicular traffic through the Golgi compartments and are directed to the plasma membrane, where they are integrated into the outer leaflet. GSLs follow a process of recycling between intracellular compartments and the plasma membrane, before final endocytosis and transportation through endosomal compartments to the lysosomes (Huwiler et al., 1998). Here, highly specific glycosylhydrolases remove the terminal sugar sequentially from the GSLs, to release the ceramide backbone, which is subsequently catabolised or recycled. It is important to note here, that deficiencies in specific glycosylhydrolases manifest specifically as lysosomal storage diseases, where there is an accumulation of GSL in the lysosome (Kolter & Sandhoff, 1998). These include: Tay-Sachs disease, which accumulates GM2; Gaucher's disease, which accumulates glucocerebroside; and Fabry's disease, which

The biological functions of GSLs are many and varied, and may particularly relate to the distribution pattern within the membrane. One such functional role is attributed to the maintenance of membrane structural rigidity, and the ordering of the membrane structure

and ganglio- (Gal1-3GalNAc) series.

accumulates Gb3 (Kanfer & Hakomori, 1983).

**2.2 Cellular functions** 

in lipid rafts (discussed below).

1996, Littman, 1998).Thus, the current paradigm for HIV infection is that X4 or R5 HIV-1 first binds via its envelope gp120 to CD4 on T-cells expressing either CXCR4 or CCR5. The binding to CD4 results in a conformational change in the structure of a part of the virus envelope gp120 known as the variable V3 loop. This change in conformation of the gp120 exposes a binding site for either chemokine co-receptor. Following binding of the virus to the chemokine co-receptor, another conformational change occurs in the gp120 that exposes another viral membrane protein called gp41. It is the gp41 that then is able to cause the fusion of the virus envelope to the host cell membrane so that the virus can release its contents into the host target cell and begin the infectious process.

Perhaps not that surprising given the history of HIV/AIDS, the HIV-1 paradigm for productive infection continues to change as more studies are undertaken. A role for a family of cell-surface-expressed neuropeptide receptors has been proposed to be important for productive HIV infection (Branch et al., 2002) and cell-surface-expressed glycosphingolipids (GSLs) have been proposed to act as HIV-1 fusion receptors (Fantini et al., 1997; Nehete et al., 2002). Thus, despite 30 years of intense research, we continue to find new and surprising aspects of HIV pathogenesis that have eluded us over the years. One of these more recent findings is the possible therapeutic potential of GSLs in HIV/AIDS.

## **2. Glycosphingolipids**

#### **2.1 Biochemistry, biosynthesis and degradation**

GSLs are carbohydrate-lipid conjugates almost exclusively restricted to the outer leaflet of the plasma membrane bilayer of mammalian cells. The hydrophobic backbone, ceramide, consists of a fatty acid chain linked to a sphingosine base, and is common to all GSLs. The alkyl chains of the lipid moiety (ceramide) are embedded in the bilayer and vary in chain length, saturation and hydroxylation (Huwiler et al., 2000; Hakomori, 1993) (Figure 1).

Fig. 1. Different functional domains of GSLs.

1996, Littman, 1998).Thus, the current paradigm for HIV infection is that X4 or R5 HIV-1 first binds via its envelope gp120 to CD4 on T-cells expressing either CXCR4 or CCR5. The binding to CD4 results in a conformational change in the structure of a part of the virus envelope gp120 known as the variable V3 loop. This change in conformation of the gp120 exposes a binding site for either chemokine co-receptor. Following binding of the virus to the chemokine co-receptor, another conformational change occurs in the gp120 that exposes another viral membrane protein called gp41. It is the gp41 that then is able to cause the fusion of the virus envelope to the host cell membrane so that the virus can release its

Perhaps not that surprising given the history of HIV/AIDS, the HIV-1 paradigm for productive infection continues to change as more studies are undertaken. A role for a family of cell-surface-expressed neuropeptide receptors has been proposed to be important for productive HIV infection (Branch et al., 2002) and cell-surface-expressed glycosphingolipids (GSLs) have been proposed to act as HIV-1 fusion receptors (Fantini et al., 1997; Nehete et al., 2002). Thus, despite 30 years of intense research, we continue to find new and surprising aspects of HIV pathogenesis that have eluded us over the years. One of these more recent

GSLs are carbohydrate-lipid conjugates almost exclusively restricted to the outer leaflet of the plasma membrane bilayer of mammalian cells. The hydrophobic backbone, ceramide, consists of a fatty acid chain linked to a sphingosine base, and is common to all GSLs. The alkyl chains of the lipid moiety (ceramide) are embedded in the bilayer and vary in chain length, saturation and hydroxylation (Huwiler et al., 2000; Hakomori, 1993) (Figure 1).

contents into the host target cell and begin the infectious process.

findings is the possible therapeutic potential of GSLs in HIV/AIDS.

**2.1 Biochemistry, biosynthesis and degradation** 

Fig. 1. Different functional domains of GSLs.

**2. Glycosphingolipids** 

The hydrophilic core sugar sequence defines the carbohydrate moieties, and these protrude into the extracellular space (Stults et al., 1989). The different moieties comprising GSLs have different roles in these multifunctional membrane lipids. GSLs are classified as neutral, acidic (anionic) and basic (cationic) glycolipids (Hakomori, 1986). Acidic GSLs usually contain either a sialic acid, which largely encompass the gangliosides, or a sulphate group. Basic glycolipids are very rare, but include plasmalopsychosine or glyceroplasmalopsychosine (Hikita et al., 2002). For the most part, GSLs are comprised of four groups, characterised by their basic core structure: globo- (defined by Gal1-4 Gal), lacto- (Gal1-3GlcNAc), neolacto- (Gal1-4GlcNAc) and ganglio- (Gal1-3GalNAc) series.

The ceramide backbone of GSLs is synthesized on the cytosolic leaflet of the rough ER, through condensation of L-serine and fatty-acyl co-enzymeA and subsequent enzymatic modification (Huwiler et al., 1998). Ceramide may be converted in the lumen of the ER into gala-series glycolipids by addition of galactose, via -glycosidic linkage, producing the first in the series, galactosylceramide (GalCer) (Sprong et al., 1998). The addition of a sulphate group to the 3-position of the sugar residue on GalCer will give rise to sulphatide (SGC). Ceramide may alternatively be transported to the Golgi apparatus where the first sugar added is glucose, via -glycosidic link, producing glucosylceramide (GlcCer). The precursor for most GSL structures is GlcCer, which is synthesised by glucosyltransferase located in the cytosol (Futerman & Pagano, 1991). GlcCer can be then translocated by the flippase function of the drug resistance pump, P-glycoprotein (P-gp), to the Golgi lumen (De Rosa et al., 2004, Lala et al., 2000). Here subsequent synthesis of all other GSLs takes place through highly specific glycosyltransferases (Lannert et al., 1998). The first product that is formed from GlcCer is lactosyl ceramide (Gal1-4Glc cer, LC), which can then be sialylated, to give monosialoganglioside (sialic acid 2-3 Gal1-4Glc cer, GM3). Alternatively, LC is galactosylated to form globotriaosyl ceramide (Gal1-4 Gal1-4Glc cer, Gb3), which can be further converted to globotetraosyl ceramide (GalNAc1-3 Gal1-4 Gal1-4Glc cer, Gb4) (Figure 2). The major GSLs contain ~5 sugars or less although GSLs containing over 60 sugar residues have been described (Miller-Podraza et al., 1993), and more than 400 GSL species have been reported (Hakomori, 2008). Newly synthesised GSLs follow anterograde vesicular traffic through the Golgi compartments and are directed to the plasma membrane, where they are integrated into the outer leaflet. GSLs follow a process of recycling between intracellular compartments and the plasma membrane, before final endocytosis and transportation through endosomal compartments to the lysosomes (Huwiler et al., 1998). Here, highly specific glycosylhydrolases remove the terminal sugar sequentially from the GSLs, to release the ceramide backbone, which is subsequently catabolised or recycled. It is important to note here, that deficiencies in specific glycosylhydrolases manifest specifically as lysosomal storage diseases, where there is an accumulation of GSL in the lysosome (Kolter & Sandhoff, 1998). These include: Tay-Sachs disease, which accumulates GM2; Gaucher's disease, which accumulates glucocerebroside; and Fabry's disease, which accumulates Gb3 (Kanfer & Hakomori, 1983).

#### **2.2 Cellular functions**

The biological functions of GSLs are many and varied, and may particularly relate to the distribution pattern within the membrane. One such functional role is attributed to the maintenance of membrane structural rigidity, and the ordering of the membrane structure in lipid rafts (discussed below).

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 243

stream during an immune response (Sperandio, 2006) Furthermore, cell-cell interactions can take place between GSLs, and these interactions have been documented between GM3-Gg3, GM3-LacCer, and SGC-GalCer (Kojima & Hakomori, 1991, Koshy et al., 1999) and more recently, between Gb3-GalCer and Gb3-GlcCer (Mahfoud et al., 2010) and between Gg4– GM1, Gg4-Gg3 and Gg4-LacCer(Emam et al., 2010). GSLs acting as functional receptors include sulphated galactolipids (SGC, SGG), which are receptors for hsp70's (Boulanger et al., 1995, Mamelak et al., 2001b); GM1, a co-receptor for FGF2 (Rusnati et al., 2002); and nerve cell gangliosides GD1a and GT1b that bind myelin-associated glycoprotein (MAG),

GSLs, particularly gangliosides, may also have an impact on cell growth and motility (Hakomori & Igarashi, 1993). In terms of cell growth, several GSLs have been shown to interact with growth factor receptors, such as FGFR and EGFR, and modulate growth (Bremer, 1994, Weis & Davis, 1990). GM3 interaction with the insulin receptor is important in type 2 diabetes (Tagami et al., 2002, van Eijk et al., 2009). While the mechanisms of inhibitory or stimulatory effects are not well understood, in many cases receptor-associated tyrosine kinases are inhibited. Cell motility is controlled by integrin function, and is specifically affected by gangliosides. The GSL, GM3 is able to inhibit motility by interaction within a complex of N-glycosylated alpha3 integrin and tetraspanin CD9 (Ono et al., 2001).

Biological membrane lipids are not homogeneously distributed but can be organized into heterogeneous microdomains or lipid rafts of increased membrane order. Lipid rafts within the plasma membrane of eukaryotic cells present different physical assemblies of proteins and lipids. Specifically, rafts are comprised of increased concentrations of GSLs, certain phospholipids, and cholesterol, as well as scaffold and/or functional membrane proteins (Hooper, 1999, Simons & Ehehalt, 2002, Simons & Ikonen, 1997). Several membrane proteins preferentially associate with lipid rafts, and these include glycosylphosphatidylinositol (GPI)-anchored cell surface proteins within the outer leaflet, and cytosolic palmitoylated and myristoylated proteins, and cholesterol- or phospholipid-binding proteins (Rajendran & Simons, 2005). The proteins and lipids cooperate to form dynamic membrane assemblies to facilitate transmembrane information flow (Lingwood & Simons, 2010) One morphologically identifiable raft structure is caveolae, which are flask-shaped invaginations of the membrane associated with caveolin scaffolding protein (Kurzchalia & Parton, 1999). Lipid rafts are small, highly dynamic and detergent-insoluble, and while these assemblies are fluid, they represent a more ordered region within the membrane. This "liquid-ordered" domain is more tightly packed than the surrounding bilayer, and this is largely due to the saturated hydrocarbon chains of raft-associated GSL and phospholipids (Simons & Vaz, 2004). Thus, the degree of saturation and hydroxylation of GSLs may greatly affect the "liquid-ordered" state of the membrane, as well as the degree of clustering or association with membrane proteins (Brown & London, 1997, Hakomori et al., 1998b). Because raft formation is dependant on lipid structure, lipids of the appropriate structure are capable of forming microdomains in model membranes (Dietrich et al., 2001, Radhakrishnan et al., 2000). However such model membrane systems do not fully reflect plasma membrane

Two fundamental properties of lipid rafts associated with their physical attributes, are their capacity to selectively incorporate proteins, and their ability to coalesce to form larger

and inhibit nerve regeneration (Vyas et al., 2002).

**2.3 Lipid rafts** 

microdomains (Kaiser et al., 2009).

Fig. 2. Biosynthesis of Globotriaosylceramide (Gb3).

Several GSLs act as cellular antigens or cell-type specific markers, although the functional significance of this is not well understood. Perhaps the most well-known cell-type specific antigens are the GSLs that comprise the histo-blood group antigens, which include members of the Lewis (Le), ABH, I/i and P/P1/Pk blood groups. Differential expression of GSLs is also particularly depicted during development. The GSLs Lex (stage-specific embryonic antigen 1, SSEA-1), which is also a Lewis blood-group antigen, Gb5 (SSEA-3) and monosialyl-Gb5 (SSEA-4) are variably expressed at specific stages of embryonic development (Kannagi et al., 1983a, Kannagi et al., 1982, Kannagi et al., 1983b, Solter & Knowles, 1978). This differential expression profile has been shown to be important for cell adhesion and cell-cell contact, and such (carbohydrate-carbohydrate) interactions may be essential in developmental processes (Eggens et al., 1989). Indeed, SSEA-4 is a human multipotent stem cell marker.

Interestingly, cell-specific expression of GSLs is also common during differentiation and this is well demonstrated in the haematopoietic cell system. Myeloid cells are characterised by Lex, Neutrophils specifically express the GSL marker LC (CDW17) and the major T-cell (and monocyte) GSL is ganglioside GM3 (Schwartz-Albeiz et al., 1991, Sorice et al., 2004). Furthermore Gb3, which has been defined as CD77, is a marker of germinal centre B cells in humans, and thus a marker of differentiation (Mangeney et al., 1991, Wiels et al., 1991).

Given the role of GSLs in development and differentiation, it is not surprising that GSL expression may be aberrant in tumour development, and several have been identified as tumour-associated antigens (Hakomori, 1985). Developmentally regulated GSLs may be reexpressed, or the GSL profile modified, to specifically aid in tumour progression through adhesion functions or tumour growth modulation (Hakomori, 1996, Hakomori, 2002).

It is interesting to note that certain GSLs have been documented to act as cell adhesion molecules, and even as functional receptors, on the cell surface. The GSL sialyl 6-sulfo Lex acts as an adhesion ligand for selectins on leukocytes and activated endothelial cells, facilitating the process termed "rolling", a critical step in migration of cells from the blood stream during an immune response (Sperandio, 2006) Furthermore, cell-cell interactions can take place between GSLs, and these interactions have been documented between GM3-Gg3, GM3-LacCer, and SGC-GalCer (Kojima & Hakomori, 1991, Koshy et al., 1999) and more recently, between Gb3-GalCer and Gb3-GlcCer (Mahfoud et al., 2010) and between Gg4– GM1, Gg4-Gg3 and Gg4-LacCer(Emam et al., 2010). GSLs acting as functional receptors include sulphated galactolipids (SGC, SGG), which are receptors for hsp70's (Boulanger et al., 1995, Mamelak et al., 2001b); GM1, a co-receptor for FGF2 (Rusnati et al., 2002); and nerve cell gangliosides GD1a and GT1b that bind myelin-associated glycoprotein (MAG), and inhibit nerve regeneration (Vyas et al., 2002).

GSLs, particularly gangliosides, may also have an impact on cell growth and motility (Hakomori & Igarashi, 1993). In terms of cell growth, several GSLs have been shown to interact with growth factor receptors, such as FGFR and EGFR, and modulate growth (Bremer, 1994, Weis & Davis, 1990). GM3 interaction with the insulin receptor is important in type 2 diabetes (Tagami et al., 2002, van Eijk et al., 2009). While the mechanisms of inhibitory or stimulatory effects are not well understood, in many cases receptor-associated tyrosine kinases are inhibited. Cell motility is controlled by integrin function, and is specifically affected by gangliosides. The GSL, GM3 is able to inhibit motility by interaction within a complex of N-glycosylated alpha3 integrin and tetraspanin CD9 (Ono et al., 2001).

#### **2.3 Lipid rafts**

242 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Several GSLs act as cellular antigens or cell-type specific markers, although the functional significance of this is not well understood. Perhaps the most well-known cell-type specific antigens are the GSLs that comprise the histo-blood group antigens, which include members of the Lewis (Le), ABH, I/i and P/P1/Pk blood groups. Differential expression of GSLs is also particularly depicted during development. The GSLs Lex (stage-specific embryonic antigen 1, SSEA-1), which is also a Lewis blood-group antigen, Gb5 (SSEA-3) and monosialyl-Gb5 (SSEA-4) are variably expressed at specific stages of embryonic development (Kannagi et al., 1983a, Kannagi et al., 1982, Kannagi et al., 1983b, Solter & Knowles, 1978). This differential expression profile has been shown to be important for cell adhesion and cell-cell contact, and such (carbohydrate-carbohydrate) interactions may be essential in developmental processes (Eggens et al., 1989). Indeed, SSEA-4 is a human

Interestingly, cell-specific expression of GSLs is also common during differentiation and this is well demonstrated in the haematopoietic cell system. Myeloid cells are characterised by Lex, Neutrophils specifically express the GSL marker LC (CDW17) and the major T-cell (and monocyte) GSL is ganglioside GM3 (Schwartz-Albeiz et al., 1991, Sorice et al., 2004). Furthermore Gb3, which has been defined as CD77, is a marker of germinal centre B cells in humans, and thus a marker of differentiation (Mangeney et al., 1991, Wiels et al., 1991). Given the role of GSLs in development and differentiation, it is not surprising that GSL expression may be aberrant in tumour development, and several have been identified as tumour-associated antigens (Hakomori, 1985). Developmentally regulated GSLs may be reexpressed, or the GSL profile modified, to specifically aid in tumour progression through adhesion functions or tumour growth modulation (Hakomori, 1996, Hakomori, 2002). It is interesting to note that certain GSLs have been documented to act as cell adhesion molecules, and even as functional receptors, on the cell surface. The GSL sialyl 6-sulfo Lex acts as an adhesion ligand for selectins on leukocytes and activated endothelial cells, facilitating the process termed "rolling", a critical step in migration of cells from the blood

Fig. 2. Biosynthesis of Globotriaosylceramide (Gb3).

multipotent stem cell marker.

Biological membrane lipids are not homogeneously distributed but can be organized into heterogeneous microdomains or lipid rafts of increased membrane order. Lipid rafts within the plasma membrane of eukaryotic cells present different physical assemblies of proteins and lipids. Specifically, rafts are comprised of increased concentrations of GSLs, certain phospholipids, and cholesterol, as well as scaffold and/or functional membrane proteins (Hooper, 1999, Simons & Ehehalt, 2002, Simons & Ikonen, 1997). Several membrane proteins preferentially associate with lipid rafts, and these include glycosylphosphatidylinositol (GPI)-anchored cell surface proteins within the outer leaflet, and cytosolic palmitoylated and myristoylated proteins, and cholesterol- or phospholipid-binding proteins (Rajendran & Simons, 2005). The proteins and lipids cooperate to form dynamic membrane assemblies to facilitate transmembrane information flow (Lingwood & Simons, 2010) One morphologically identifiable raft structure is caveolae, which are flask-shaped invaginations of the membrane associated with caveolin scaffolding protein (Kurzchalia & Parton, 1999).

Lipid rafts are small, highly dynamic and detergent-insoluble, and while these assemblies are fluid, they represent a more ordered region within the membrane. This "liquid-ordered" domain is more tightly packed than the surrounding bilayer, and this is largely due to the saturated hydrocarbon chains of raft-associated GSL and phospholipids (Simons & Vaz, 2004). Thus, the degree of saturation and hydroxylation of GSLs may greatly affect the "liquid-ordered" state of the membrane, as well as the degree of clustering or association with membrane proteins (Brown & London, 1997, Hakomori et al., 1998b). Because raft formation is dependant on lipid structure, lipids of the appropriate structure are capable of forming microdomains in model membranes (Dietrich et al., 2001, Radhakrishnan et al., 2000). However such model membrane systems do not fully reflect plasma membrane microdomains (Kaiser et al., 2009).

Two fundamental properties of lipid rafts associated with their physical attributes, are their capacity to selectively incorporate proteins, and their ability to coalesce to form larger

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 245

labeling of the adaGb3 'raft" band. The raft band is still formed –seen under visible light- in the presence of SPC3. Thus both the SPC3 peptide (and presumably, gp120 and the intact HIV virus) and VT1 B selectively bind the same Gb3 containing raft structures. This would correlate with the raft requirement for HIV infectivity and VT cytotoxicity (Falguieres et al., 2001).

GSLs have been shown to play a role in many pathogen interactions with host cells. As previously described, several GSLs represent histo-blood group molecules, and there is a longstanding association between pathogens and these particular blood groups, which are not necessarily limited to expression on erythrocytes. Such interactions have been defined both in protective qualities conferred by a specific blood type, and in pathogen interactions

Several GSLs, including those categorised as blood group antigens, have been identified as adherence receptors for bacteria, or bacterial toxins (Lingwood, 1998). The globo-series of GSLs expressed on urogenital epithelia, particularly monosialyl-Gb5, are infection sites for *Escherichia coli* (Stapleton et al., 1992). The Leb antigen is required for surface adherence of *Helicobacter pylori*, known to cause gastritis and peptic ulcers, to gastric mucosa, and Group O Le (b+) secretors are thus likely most susceptible to this pathogen (Borén et al., 1993). The minimal structure of sialyl-lactosylceramide (GM3) is crucial for colonization and adherence to epithelium via fimbria-dependant binding of *Haemophilus influenzae*, which causes a variety of diseases from meningitis to upper respiratory infection (van Alphen et al., 1991).Ganglioseries GSLs such as asialoGM1(Gg4), are binding targets for the pili of certain *Pseudomonas aeruginosa* strains, which are opportunistic pathogens that target and colonize epithelial cells of the lung (Comolli et al., 1999). These GSLs are not receptors for these organisms (Emam et al., 2006) but can assist host cell invasion (Emam et al., 2010). Finally, bacterial toxins, which are soluble proteins, often bind to GSLs to elicit their effects. Glycolipid receptors include ganglioside GM1, bound by cholera toxin (De Haan & Hirst, 2004) from *Vibrio cholerae*, and Gb3, which is utilized by *Escherichia coli* elicited verotoxins (VT) (Petruzziello et al., 2009), susceptibility to

The involvement of GSLs in the host cell attachment of viruses, and also fusion in terms of enveloped viruses, has long been recognised (Haywood, 1994). The initial step of viral attachment to the susceptible cell is crucial in the process of establishing an infection. The sialic acid motif, which is widely presented on acidic GSLs, is perhaps the most broadly recognised adhesion component utilised by viruses, from small non-enveloped DNA polyomaviruses to larger enveloped RNA influenza viruses (Gilbert & Benjamin, 2004, Miller-Podraza et al., 2000, Tsai et al., 2003). The GSL neolactotetraosylceramide (nLc4Cer) is a key receptor for the enveloped Dengue virus, an infectious agent transmitted by mosquitoes (Aoki et al., 2006). The ganglioside GD1a has been identified as a critical component for viral binding of Sendai virus, and fusion of this enveloped virus with its

GSLs and lipid rafts themselves are important for many microbial pathogens and often form preferential sites for pathogen interactions (Lafont et al., 2002, Samuel et al., 2001, van der Goot & Harder, 2001, Vieira et al., 2010). Pathogenic interactions may be vast and varied. For example, lipid rafts serve as key platforms for entry of parasitic agents, such as *Plasmodium* 

target is abolished if GD1a is not present for initial contact (Epand et al., 1995).

with blood group antigens (Moulds & Moulds, 2000, Rios & Bianco, 2000 )

cholera toxin (and *E. Coli* LT) is blood group O related.

**2.5 Pathogens and lipid rafts** 

**2.4 Pathogens and GSL receptors** 

domains. It is not surprising therefore, that lipid rafts play a role in protein sorting, membrane trafficking and signal transduction (Brown & London, 2000, Lajoie & Nabi, 2010). Because of the thermodynamic formation of lipid rafts, the cell utilizes them as centres or "hot spots" for transmembrane signal transduction for a variety of membrane receptors (Hakomori & Igarashi, 1995, Simons & Toomre, 2000). Ligand-induced receptor dimerization and successive cytosolic phosphorylation cascades occur in microdomains, and as such membrane receptors often partition into such domains upon ligation, and may subsequently be internalised and traffic through said domains (Dykstra et al., 2001). This can result in direct ligand interaction with GSLs (Hakomori et al., 1998a, Iwabuchi et al., 2000) without necessarily, the involvement of a transmembrane protein (Katagiri et al., 1999, Mori et al., 2000). This implies a mechanism of communication between the cell surface and the cytosolic lipid bilayer leaflets. Cytosolic signal transduction proteins, such as *src*-family tyrosine kinases and small G-proteins, are often associated with the cytosolic surface of such domains in a transient and surface ligand-regulated manner (Dykstra et al., 2001, Hakomori, 2000, Katagiri et al., 1999).

Lipid rafts are also involved in internalisation and intracellular trafficking of proteins and lipids(Lajoie & Nabi, 2010, Mukherjee & Maxfield, 2000) and likely, their attendant signalling. An endocytic role has been established for cavolae and lipid rafts, which may translocate and endocytose GPI-anchored proteins in particular (Parton & Richards, 2003). Other raft-mediated routes of internalisation have been identified where the GPI-anchor acts as a targeting signal in the traffic to an endosomal organelle called the GPI-anchored protein enriched early endosomal compartment (Sabharanjak et al., 2002). It is clear therefore that several distinct raft-mediated trafficking pathways exist. It is important to note however, that lipid rafts are not distributed randomly in the endosomal pathway, but are excluded from the degradative compartments, although this is not well understood (Nichols et al., 2001, Simons & Gruenberg, 2000).

### **2.3.1 Defining a new assay for lipid raft formation**

Lipid rafts are isolated from cells from the Triton insoluble fraction separated on a discontinuous sucrose ultracentrifuge gradient. Due to their atypical density, the rafts separate as a band above the 30% sucrose layer. The majority of proteins sediment to the bottom, while components found in this fraction are deemed lipid raft associated. This has not been studied for purified (glyco)lipids. We have developed this procedure as a new method for examining the 'raft' forming capability of glycolipids. Soluble adamantylGb3 (adaGb3), natural Gb3 or Gb3+cholesterol were mixed with Triton and placed at the bottom of the sucrose gradient, below the 30% layer, the lower half of which now contains FITClabeled VT1 B subunit. The gradients are centrifuged at 66K rpm for 3 days. Any rafts structures formed will float up through the FITC-VT1 B layer and the raft band should thus be fluorescently labeled. When this was performed with Gb3 alone, no fluorescent band was formed. In contrast, a distinct fluorescent band was formed for Gb3+cholesterol. However adaGb3 formed the strongest labeled 'raft' band (Mahfoud et al., 2002b). While the characteristics of the structures formed by adaGb3 in this band remain to be fully characterized, this supports the "raft-like" character of adaGb3. Moreover, this is an excellent method for determining the properties and components required for optimal raft formation (Nutikka & Lingwood, 2004). We have shown that cholesterol is one requirement. A fifty fold molar excess of the SPC3 peptide from the glycolipid binding V3 loop of gp120 of HIV, which strongly binds adaGb3 (Mahfoud et al., 2002b) is able to eliminate FITC-VT1 B labeling of the adaGb3 'raft" band. The raft band is still formed –seen under visible light- in the presence of SPC3. Thus both the SPC3 peptide (and presumably, gp120 and the intact HIV virus) and VT1 B selectively bind the same Gb3 containing raft structures. This would correlate with the raft requirement for HIV infectivity and VT cytotoxicity (Falguieres et al., 2001).

#### **2.4 Pathogens and GSL receptors**

244 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

domains. It is not surprising therefore, that lipid rafts play a role in protein sorting, membrane trafficking and signal transduction (Brown & London, 2000, Lajoie & Nabi, 2010). Because of the thermodynamic formation of lipid rafts, the cell utilizes them as centres or "hot spots" for transmembrane signal transduction for a variety of membrane receptors (Hakomori & Igarashi, 1995, Simons & Toomre, 2000). Ligand-induced receptor dimerization and successive cytosolic phosphorylation cascades occur in microdomains, and as such membrane receptors often partition into such domains upon ligation, and may subsequently be internalised and traffic through said domains (Dykstra et al., 2001). This can result in direct ligand interaction with GSLs (Hakomori et al., 1998a, Iwabuchi et al., 2000) without necessarily, the involvement of a transmembrane protein (Katagiri et al., 1999, Mori et al., 2000). This implies a mechanism of communication between the cell surface and the cytosolic lipid bilayer leaflets. Cytosolic signal transduction proteins, such as *src*-family tyrosine kinases and small G-proteins, are often associated with the cytosolic surface of such domains in a transient and surface ligand-regulated manner (Dykstra et al., 2001, Hakomori,

Lipid rafts are also involved in internalisation and intracellular trafficking of proteins and lipids(Lajoie & Nabi, 2010, Mukherjee & Maxfield, 2000) and likely, their attendant signalling. An endocytic role has been established for cavolae and lipid rafts, which may translocate and endocytose GPI-anchored proteins in particular (Parton & Richards, 2003). Other raft-mediated routes of internalisation have been identified where the GPI-anchor acts as a targeting signal in the traffic to an endosomal organelle called the GPI-anchored protein enriched early endosomal compartment (Sabharanjak et al., 2002). It is clear therefore that several distinct raft-mediated trafficking pathways exist. It is important to note however, that lipid rafts are not distributed randomly in the endosomal pathway, but are excluded from the degradative compartments, although this is not well understood (Nichols et al.,

Lipid rafts are isolated from cells from the Triton insoluble fraction separated on a discontinuous sucrose ultracentrifuge gradient. Due to their atypical density, the rafts separate as a band above the 30% sucrose layer. The majority of proteins sediment to the bottom, while components found in this fraction are deemed lipid raft associated. This has not been studied for purified (glyco)lipids. We have developed this procedure as a new method for examining the 'raft' forming capability of glycolipids. Soluble adamantylGb3 (adaGb3), natural Gb3 or Gb3+cholesterol were mixed with Triton and placed at the bottom of the sucrose gradient, below the 30% layer, the lower half of which now contains FITClabeled VT1 B subunit. The gradients are centrifuged at 66K rpm for 3 days. Any rafts structures formed will float up through the FITC-VT1 B layer and the raft band should thus be fluorescently labeled. When this was performed with Gb3 alone, no fluorescent band was formed. In contrast, a distinct fluorescent band was formed for Gb3+cholesterol. However adaGb3 formed the strongest labeled 'raft' band (Mahfoud et al., 2002b). While the characteristics of the structures formed by adaGb3 in this band remain to be fully characterized, this supports the "raft-like" character of adaGb3. Moreover, this is an excellent method for determining the properties and components required for optimal raft formation (Nutikka & Lingwood, 2004). We have shown that cholesterol is one requirement. A fifty fold molar excess of the SPC3 peptide from the glycolipid binding V3 loop of gp120 of HIV, which strongly binds adaGb3 (Mahfoud et al., 2002b) is able to eliminate FITC-VT1 B

2000, Katagiri et al., 1999).

2001, Simons & Gruenberg, 2000).

**2.3.1 Defining a new assay for lipid raft formation** 

GSLs have been shown to play a role in many pathogen interactions with host cells. As previously described, several GSLs represent histo-blood group molecules, and there is a longstanding association between pathogens and these particular blood groups, which are not necessarily limited to expression on erythrocytes. Such interactions have been defined both in protective qualities conferred by a specific blood type, and in pathogen interactions with blood group antigens (Moulds & Moulds, 2000, Rios & Bianco, 2000 )

Several GSLs, including those categorised as blood group antigens, have been identified as adherence receptors for bacteria, or bacterial toxins (Lingwood, 1998). The globo-series of GSLs expressed on urogenital epithelia, particularly monosialyl-Gb5, are infection sites for *Escherichia coli* (Stapleton et al., 1992). The Leb antigen is required for surface adherence of *Helicobacter pylori*, known to cause gastritis and peptic ulcers, to gastric mucosa, and Group O Le (b+) secretors are thus likely most susceptible to this pathogen (Borén et al., 1993). The minimal structure of sialyl-lactosylceramide (GM3) is crucial for colonization and adherence to epithelium via fimbria-dependant binding of *Haemophilus influenzae*, which causes a variety of diseases from meningitis to upper respiratory infection (van Alphen et al., 1991).Ganglioseries GSLs such as asialoGM1(Gg4), are binding targets for the pili of certain *Pseudomonas aeruginosa* strains, which are opportunistic pathogens that target and colonize epithelial cells of the lung (Comolli et al., 1999). These GSLs are not receptors for these organisms (Emam et al., 2006) but can assist host cell invasion (Emam et al., 2010). Finally, bacterial toxins, which are soluble proteins, often bind to GSLs to elicit their effects. Glycolipid receptors include ganglioside GM1, bound by cholera toxin (De Haan & Hirst, 2004) from *Vibrio cholerae*, and Gb3, which is utilized by *Escherichia coli* elicited verotoxins (VT) (Petruzziello et al., 2009), susceptibility to cholera toxin (and *E. Coli* LT) is blood group O related.

The involvement of GSLs in the host cell attachment of viruses, and also fusion in terms of enveloped viruses, has long been recognised (Haywood, 1994). The initial step of viral attachment to the susceptible cell is crucial in the process of establishing an infection. The sialic acid motif, which is widely presented on acidic GSLs, is perhaps the most broadly recognised adhesion component utilised by viruses, from small non-enveloped DNA polyomaviruses to larger enveloped RNA influenza viruses (Gilbert & Benjamin, 2004, Miller-Podraza et al., 2000, Tsai et al., 2003). The GSL neolactotetraosylceramide (nLc4Cer) is a key receptor for the enveloped Dengue virus, an infectious agent transmitted by mosquitoes (Aoki et al., 2006). The ganglioside GD1a has been identified as a critical component for viral binding of Sendai virus, and fusion of this enveloped virus with its target is abolished if GD1a is not present for initial contact (Epand et al., 1995).

#### **2.5 Pathogens and lipid rafts**

GSLs and lipid rafts themselves are important for many microbial pathogens and often form preferential sites for pathogen interactions (Lafont et al., 2002, Samuel et al., 2001, van der Goot & Harder, 2001, Vieira et al., 2010). Pathogenic interactions may be vast and varied. For example, lipid rafts serve as key platforms for entry of parasitic agents, such as *Plasmodium* 

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 247

of GSLs. Gb3 shares the amphipathic characteristics of all GSLs, and their fatty acid chain length, saturation and hydroxylation, may vary yielding various Gb3 isoforms. Gb3 may also partition into lipid-rafts and interact with raft-associated proteins. Gb3 is widely expressed in a variety of tissues, but is a major GSL of human renal cortex(Boyd & Lingwood, 1989), heart, spleen and placenta; (Kojima et al., 2000). Moreover, it has been described in a number of epithelial and endothelial cell lines. Gb3, or CD77, is expressed as a differentiation antigen on a subset of tonsillar B lymphocytes in the germinal center, where expression is very specific, and only occurs at a restricted stage (Mangeney et al., 1991, Wiels et al., 1991). It is interesting to note that human Burkitt lymphoma cells, which are characteristically derived from B cells, also express Gb3 (Wiels et al., 1981; Kim et al., 2011). Finally, human erythrocytes are characterized by two major globo-series GSLs, Gb3 or Pk, being one, the other being Gb4, or P. Gb3 is also upregulated in many human tumours

Gb3 expressed on B-lymphocytes, has specifically been implicated in signal transduction resulting from CD19 engagement (Maloney & Lingwood, 1994). Indeed, the extracellular domain of CD19 presents a Gb3 binding site, (with sequence similarity to the Gb3-binding VT1B subunit of *Escherichia coli*). Gb3 is crucial for CD19 induced homotypic adhesion of B cells and this suggests a potential role for Gb3 in adhesion during B cell development (Maloney & Lingwood, 1994). Gb3 has further been shown to mediate CD19 directed apoptosis of B cells, which may be important during B cell selection (Khine et al., 1998). This occurs following CD19 ligation, where Gb3 mediates targeting and intracellular traffic of

Gb3 expression has also been shown to affect the binding capacity of IFN-α for its receptor, α2 interferon receptor IFNAR1, on B lymphoid cells (Ghislain et al., 1992). The amino

(Devenica et al., 2010)

**3.2 Function** 

Fig. 3. Chemical structure of globotriaosylceramide.

CD19 to the ER and nuclear envelope (Khine et al., 1998).

*falciparum*, which causes malaria. Following attachment of the *P. falciparum* merozoite to erythrocytes, the membrane invaginates taking up the parasite within a parasitophorous vacuolar membrane (PVM) (Haldar et al., 2002). Lipid rafts are critical for the formation of the PVM, as are the raft-associated proteins internalized with the vacuole (Lauer et al., 2000). Indeed, even 'non-classic' infectious agents require lipid rafts, as demonstrated by the requirement for the prion proteins which partition into rafts during the conversion of PrPc to infectious PrPsc (Simons & Ehehalt, 2002).

Bacteria often favour lipid rafts during host-cell interactions (Heung et al., 2006),. Raft association may provide a platform for colonisation, through signalling, cytoskeleton rearrangements and membrane ruffling (Manes & Martinez, 2004). Intracellular bacteria rely on lipid rafts to enter host-cells, which provides protection from degradation and immune detection. This is demonstrated by *Mycobacterium* spp., which exploits rafts to generate phagosomes in which to survive within the cell, allowing the bacterium to evade antigen processing (Gatfield & Pieters, 2000). Toxins produced by non-intracellular bacteria, are particularly dependant on rafts for host-cell interaction, which mediates oligomerization, internalization and intracellular trafficking (Fivaz et al., 1999). Cholera toxin binding ganglioside GM1 is the current "gold standard' for identification of such rafts (Lencer, 2001) Lipid rafts are integral in the retrograde transport of cholera toxin to the Golgi (Lencer & Saslowsky, 2005).VT is also dependant on the organization of its receptor, Gb3, into rafts for intracellular routing (Falguieres et al., 2001) and cytopathology (Khan et al., 2009).

Lipid rafts are also fundamental in viral infection, predominantly in the process of viral entry, both for enveloped and non-enveloped viruses (Manes et al., 2003). In the case of nonenveloped viruses, lipid rafts are important in the process of viral attachment and subsequent internalization and trafficking to the appropriate sub-cellular niche. Caveolae lipid rafts are required for non-enveloped simian virus 40 (SV40) interactions with MHC-I and viral entry, a process that can be inhibited with cholesterol chelators (Norkin, 1999). Lipid rafts facilitate the traffic of SV40 to the ER through the Golgi (Parton, 1994). Interestingly, rafts appear to be involved in a sorting process in viral trafficking, as other viruses, such as the echovirus, enter by caveolae but do not traffic to the ER(Marjomaki et al., 2002). Enveloped viruses are particularly dependant on lipid raft domains for the process of viral/cell membrane fusion. Cholesterol and sphingolipids, which define these domains, have been identified as critical in the process of fusion for alphaviruses, such as the Semliki-forest virus (Ahn et al., 2002). Interestingly, disrupting raft formation by replacing cholesterol with androstenol did not hinder envelope glycoprotein insertion, but replacing sphingolipids with dipalmitoylphosphatidylcholine was inhibitory, emphasizing the importance of GSLs in the process of fusion (Waarts et al., 2002). Finally, lipid rafts are critical in the process of enveloped virus assembly and budding for many viruses, including influenza, measles, filoviruses and HIV (Bavari et al., 2002 , Luo et al., 2008, Manie et al., 2000, Scheiffele et al., 1999 ). Membrane rafts are an efficient system of concentrating viral proteins in a specific region, may provide a specific lipid composition for the virus, and also exclude/include host-proteins from the viral envelope (Manes et al., 2003).

#### **3. Globotriaosylceramide**

#### **3.1 Characteristics and expression**

The neutral glycolipid, globotriaosylceramide (Gb3), is defined by the trisaccharide core unit (Gal1-4Gal1-4Glc) linked to a ceramide backbone, and as such belongs to the globo-series of GSLs. Gb3 shares the amphipathic characteristics of all GSLs, and their fatty acid chain length, saturation and hydroxylation, may vary yielding various Gb3 isoforms. Gb3 may also partition into lipid-rafts and interact with raft-associated proteins. Gb3 is widely expressed in a variety of tissues, but is a major GSL of human renal cortex(Boyd & Lingwood, 1989), heart, spleen and placenta; (Kojima et al., 2000). Moreover, it has been described in a number of epithelial and endothelial cell lines. Gb3, or CD77, is expressed as a differentiation antigen on a subset of tonsillar B lymphocytes in the germinal center, where expression is very specific, and only occurs at a restricted stage (Mangeney et al., 1991, Wiels et al., 1991). It is interesting to note that human Burkitt lymphoma cells, which are characteristically derived from B cells, also express Gb3 (Wiels et al., 1981; Kim et al., 2011). Finally, human erythrocytes are characterized by two major globo-series GSLs, Gb3 or Pk, being one, the other being Gb4, or P. Gb3 is also upregulated in many human tumours (Devenica et al., 2010)

Fig. 3. Chemical structure of globotriaosylceramide.

### **3.2 Function**

246 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

*falciparum*, which causes malaria. Following attachment of the *P. falciparum* merozoite to erythrocytes, the membrane invaginates taking up the parasite within a parasitophorous vacuolar membrane (PVM) (Haldar et al., 2002). Lipid rafts are critical for the formation of the PVM, as are the raft-associated proteins internalized with the vacuole (Lauer et al., 2000). Indeed, even 'non-classic' infectious agents require lipid rafts, as demonstrated by the requirement for the prion proteins which partition into rafts during the conversion of PrPc to

Bacteria often favour lipid rafts during host-cell interactions (Heung et al., 2006),. Raft association may provide a platform for colonisation, through signalling, cytoskeleton rearrangements and membrane ruffling (Manes & Martinez, 2004). Intracellular bacteria rely on lipid rafts to enter host-cells, which provides protection from degradation and immune detection. This is demonstrated by *Mycobacterium* spp., which exploits rafts to generate phagosomes in which to survive within the cell, allowing the bacterium to evade antigen processing (Gatfield & Pieters, 2000). Toxins produced by non-intracellular bacteria, are particularly dependant on rafts for host-cell interaction, which mediates oligomerization, internalization and intracellular trafficking (Fivaz et al., 1999). Cholera toxin binding ganglioside GM1 is the current "gold standard' for identification of such rafts (Lencer, 2001) Lipid rafts are integral in the retrograde transport of cholera toxin to the Golgi (Lencer & Saslowsky, 2005).VT is also dependant on the organization of its receptor, Gb3, into rafts for

intracellular routing (Falguieres et al., 2001) and cytopathology (Khan et al., 2009).

exclude/include host-proteins from the viral envelope (Manes et al., 2003).

The neutral glycolipid, globotriaosylceramide (Gb3), is defined by the trisaccharide core unit (Gal1-4Gal1-4Glc) linked to a ceramide backbone, and as such belongs to the globo-series

**3. Globotriaosylceramide** 

**3.1 Characteristics and expression** 

Lipid rafts are also fundamental in viral infection, predominantly in the process of viral entry, both for enveloped and non-enveloped viruses (Manes et al., 2003). In the case of nonenveloped viruses, lipid rafts are important in the process of viral attachment and subsequent internalization and trafficking to the appropriate sub-cellular niche. Caveolae lipid rafts are required for non-enveloped simian virus 40 (SV40) interactions with MHC-I and viral entry, a process that can be inhibited with cholesterol chelators (Norkin, 1999). Lipid rafts facilitate the traffic of SV40 to the ER through the Golgi (Parton, 1994). Interestingly, rafts appear to be involved in a sorting process in viral trafficking, as other viruses, such as the echovirus, enter by caveolae but do not traffic to the ER(Marjomaki et al., 2002). Enveloped viruses are particularly dependant on lipid raft domains for the process of viral/cell membrane fusion. Cholesterol and sphingolipids, which define these domains, have been identified as critical in the process of fusion for alphaviruses, such as the Semliki-forest virus (Ahn et al., 2002). Interestingly, disrupting raft formation by replacing cholesterol with androstenol did not hinder envelope glycoprotein insertion, but replacing sphingolipids with dipalmitoylphosphatidylcholine was inhibitory, emphasizing the importance of GSLs in the process of fusion (Waarts et al., 2002). Finally, lipid rafts are critical in the process of enveloped virus assembly and budding for many viruses, including influenza, measles, filoviruses and HIV (Bavari et al., 2002 , Luo et al., 2008, Manie et al., 2000, Scheiffele et al., 1999 ). Membrane rafts are an efficient system of concentrating viral proteins in a specific region, may provide a specific lipid composition for the virus, and also

infectious PrPsc (Simons & Ehehalt, 2002).

Gb3 expressed on B-lymphocytes, has specifically been implicated in signal transduction resulting from CD19 engagement (Maloney & Lingwood, 1994). Indeed, the extracellular domain of CD19 presents a Gb3 binding site, (with sequence similarity to the Gb3-binding VT1B subunit of *Escherichia coli*). Gb3 is crucial for CD19 induced homotypic adhesion of B cells and this suggests a potential role for Gb3 in adhesion during B cell development (Maloney & Lingwood, 1994). Gb3 has further been shown to mediate CD19 directed apoptosis of B cells, which may be important during B cell selection (Khine et al., 1998). This occurs following CD19 ligation, where Gb3 mediates targeting and intracellular traffic of CD19 to the ER and nuclear envelope (Khine et al., 1998).

Gb3 expression has also been shown to affect the binding capacity of IFN-α for its receptor, α2 interferon receptor IFNAR1, on B lymphoid cells (Ghislain et al., 1992). The amino

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 249

The P1PK and GLOB blood group system antigens are of particular interest, with many defined pathogen interactions. Both Pk (or Gb3) and P (or Gb4) are receptors for P pili of uropathogenic *E. coli* (Leffler & Svanborg-Eden, 1981). Pk has been shown to act as a receptor for the porcine bacteria *Streptococcus suis* (Haataja et al., 1994). Indeed, Pk is also known to act as a receptor for bacterial toxins, Shigella or Verotoxins specifically, produced by *Shigella dysenteriae* and Enterohemorrhagic *E. coli* (Bitzan et al., 1994, Pellizzari et al., 1992) but no association with P blood group status has been found (Jelacic et al., 2002). Viruses have also been shown to have interactions with blood group antigens. In terms of the P blood group system, Parvovirus B19 utilizes the P antigen as its receptor to infect cells, and individuals with the p phenotype lacking P are resistant to the virus (Brown et al., 1994). More recently, the P1PK and GLOB blood group system antigens, specifically, Pk, has been implicated as

Enterohemorrhagic *Escherichia coli* induce disease, characteristically haemolytic uremic syndrome (HUS), by the production of verotoxins (VT). VTs are capable of binding to Gb3, thus Gb3 contributes to the pathology of VT-induced disease (Lingwood, 2000, Lingwood et al., 1987 ) VT is comprised of a single toxic 'A' subunit and non-covalently associated pentameric 'B' subunits responsible for receptor (Gb3) binding. Only cells with Gb3 surface expression are sensitive to VT toxicity (Okuda et al., 2006, Waddell et al., 1990). VT interaction with the sugar moiety of Gb3 is dependant on the lipid moiety in its membrane environment (Arab & Lingwood, 1996, Kiarash et al., 1994, Pellizzari et al., 1992), which is crucial in internalization and subcellular targeting of VT (Arab & Lingwood, 1998 , Smith et al., 2006). The intracellular routing of VT thus is also dependant on the organization of Gb3 into lipid rafts (Falguieres et al., 2001). Indeed, VT binding to cell surface Gb3 within lipid microdomains has been shown to activate cytosolic raft-associated src kinase ((Katagiri et al., 1999, Mori et al., 2000)) indicating Gb3 can mediate transmembrane signals. Furthermore, Gb3 containing cells where Gb3 is not present in rafts are insensitive to VT cytotoxicity (Falguieres et al., 2001, Ramegowda & Tesh, 1996). In cells sensitive to VT cytotoxicity, the toxin is internalized via both clathrin independant/dependant pathways(Lauvrak et al., 2004) and undergoes retrograde transport via the reverse of the secretory system to the Golgi and ER/nucleus (Arab & Lingwood, 1998, Khine & Lingwood, 1994, Sandvig et al., 1994) Highly VT sensitive cells contain higher levels of short fatty acid containing Gb3 isoforms and retrograde transport the VT/Gb3 complex to the ER/nucleus. Less VT sensitive cells have longer fatty acid containing Gb3 isoforms and retrograde transport VT to the Golgi only (Arab & Lingwood, 1998). Interestingly, Gb3 is maintained in lipid rafts

Fabry disease is an X-linked lysosomal storage disorder, as a result of a genetic defect in the lysosomal enzyme α-galactosidase A, which results in reduced enzyme activity (Brady, 1967). This enzyme is normally responsible for the removal of the terminal Gb3 galactose residue, through hydrolysis of the 1-4 glycosidic linkage. Thus, Gb3, and potentially other -galactose terminal lipids accumulate in the lysosomes to abrogate their normal function and the function of these organelles. Clinical manifestations of the disorder are related to the

**4. Relationship to disease** 

**4.1 Verotoxin-induced disease**

having a role in HIV infection (Branch, 2010)(see below).

during retrograde transport (Smith et al., 2006).

**4.2 Fabry disease** 

terminus of IFNAR1 is able to bind to Gb3, sharing sequence similarity with the VT1B subunit (Lingwood & Yiu, 1992) and binding to Gb3 likely affects the subsequent signaling. Gb3 has thus been shown to be critical for IFNAR1-dependant α2IFN induced growth inhibition, mediated by short chain fatty acid Gb3 isoforms (Khine & Lingwood, 2000). Furthermore, Gb3 is important in α2IFN IFNAR1 signaling to induce antiviral activity, which is mediated by long fatty acid isoforms of Gb3 (Khine & Lingwood, 2000). Gb3 has been implicated in angiogenesis and is found in tumour neovasculature (Heath-Engel & Lingwood, 2003), and can promote tumour metastases (Kovbasnjuk et al., 2005)

### **3.3 Blood group antigens**

Gb3 belongs to the P1PK and GLOB blood group system, that have red cell phenotypes termed P/P1/Pk (Table 1) and whose expression profile is not limited to erythrocytes. The structure galabiose (Gal1-4Gal) is the terminal structure of Pk, also known as Gb3, and P1 blood group antigens, whilst it is the precursor for P antigen, also known as globoside or Gb4, which terminates with 1-3GalNAc(Spitalnik & Spitalnik, 1995). P1 and P2 are the two common P/P1/Pk-related blood group phenotypes. P1 individuals (~80% of Caucasians but only ~20% of Asians)(Daniels, 2002) express P and P1 but only expose low amounts of Pk antigens on their cell surfaces. P2 individuals (~20% of Caucasians and ~80% of Asians)(Daniels, 2002) express only P and low amounts of Pk antigens. There are also rare phenotypes defined by a deficiency in one or more of the P/P1/Pk blood group antigens. Individuals deficient in P antigen have mutations in the *B3GALNT1* gene causing lack of functional Gb4 synthase (β3GalNAc transferase) (Hellberg et al., 2002, Hellberg et al., 2004), and consequently express high amounts of unmodified precursor, Pk. These individuals may express P1 antigen (P1k phenotype) or not (P2 k). Although uncertain for many years (Hellberg et al., 2005; Iwamura et al., 2003), the molecular basis for P1/P2 has recently been elucidated (Thuresson et al., 2011). Individuals who do not express any P/P1/Pk antigens have mutations in the *A4GALT* gene, causing lack of functional Gb3 synthase (α4Gal transferase), and have the rare p blood group phenotype (Furukawa et al., 2000, Hellberg et al., 2002, Hellberg et al., 2003, Steffensen et al., 2000). Similar to the ABO blood group system, naturally occurring antibodies are formed against the P/P1/Pk antigens when missing (Spitalnik & Spitalnik, 1995). Recent studies show anti-Pk is present in all normal sera (Pochechueva et al., 2010)


Table 1. Red Blood Cell Phenotypes in the P1Pk and GLOB Blood Group System (Spitalnik & Spitalnik, 1995; Branch, 2010)

## **4. Relationship to disease**

248 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

terminus of IFNAR1 is able to bind to Gb3, sharing sequence similarity with the VT1B subunit (Lingwood & Yiu, 1992) and binding to Gb3 likely affects the subsequent signaling. Gb3 has thus been shown to be critical for IFNAR1-dependant α2IFN induced growth inhibition, mediated by short chain fatty acid Gb3 isoforms (Khine & Lingwood, 2000). Furthermore, Gb3 is important in α2IFN IFNAR1 signaling to induce antiviral activity, which is mediated by long fatty acid isoforms of Gb3 (Khine & Lingwood, 2000). Gb3 has been implicated in angiogenesis and is found in tumour neovasculature (Heath-Engel &

Gb3 belongs to the P1PK and GLOB blood group system, that have red cell phenotypes termed P/P1/Pk (Table 1) and whose expression profile is not limited to erythrocytes. The structure galabiose (Gal1-4Gal) is the terminal structure of Pk, also known as Gb3, and P1 blood group antigens, whilst it is the precursor for P antigen, also known as globoside or Gb4, which terminates with 1-3GalNAc(Spitalnik & Spitalnik, 1995). P1 and P2 are the two common P/P1/Pk-related blood group phenotypes. P1 individuals (~80% of Caucasians but only ~20% of Asians)(Daniels, 2002) express P and P1 but only expose low amounts of Pk antigens on their cell surfaces. P2 individuals (~20% of Caucasians and ~80% of Asians)(Daniels, 2002) express only P and low amounts of Pk antigens. There are also rare phenotypes defined by a deficiency in one or more of the P/P1/Pk blood group antigens. Individuals deficient in P antigen have mutations in the *B3GALNT1* gene causing lack of functional Gb4 synthase (β3GalNAc transferase) (Hellberg et al., 2002, Hellberg et al., 2004), and consequently express high amounts of unmodified precursor, Pk. These individuals may express P1 antigen (P1k phenotype) or not (P2k). Although uncertain for many years (Hellberg et al., 2005; Iwamura et al., 2003), the molecular basis for P1/P2 has recently been elucidated (Thuresson et al., 2011). Individuals who do not express any P/P1/Pk antigens have mutations in the *A4GALT* gene, causing lack of functional Gb3 synthase (α4Gal transferase), and have the rare p blood group phenotype (Furukawa et al., 2000, Hellberg et al., 2002, Hellberg et al., 2003, Steffensen et al., 2000). Similar to the ABO blood group system, naturally occurring antibodies are formed against the P/P1/Pk antigens when missing (Spitalnik & Spitalnik, 1995). Recent studies show anti-Pk is present in all normal

**Phenotype Frequency Red Blood Cell Antigens Serum Antibodies** 

P2k Rare Pk Anti-P1, anti-P

P Rare None Anti-P1, anti-P, anti-Pk

Table 1. Red Blood Cell Phenotypes in the P1Pk and GLOB Blood Group System (Spitalnik

P1 75% P1, P, Pk None P2 25% P, Pk Anti-P1 P1k Rare P1, Pk Anti-P

Lingwood, 2003), and can promote tumour metastases (Kovbasnjuk et al., 2005)

**3.3 Blood group antigens** 

sera (Pochechueva et al., 2010)

& Spitalnik, 1995; Branch, 2010)

The P1PK and GLOB blood group system antigens are of particular interest, with many defined pathogen interactions. Both Pk (or Gb3) and P (or Gb4) are receptors for P pili of uropathogenic *E. coli* (Leffler & Svanborg-Eden, 1981). Pk has been shown to act as a receptor for the porcine bacteria *Streptococcus suis* (Haataja et al., 1994). Indeed, Pk is also known to act as a receptor for bacterial toxins, Shigella or Verotoxins specifically, produced by *Shigella dysenteriae* and Enterohemorrhagic *E. coli* (Bitzan et al., 1994, Pellizzari et al., 1992) but no association with P blood group status has been found (Jelacic et al., 2002). Viruses have also been shown to have interactions with blood group antigens. In terms of the P blood group system, Parvovirus B19 utilizes the P antigen as its receptor to infect cells, and individuals with the p phenotype lacking P are resistant to the virus (Brown et al., 1994). More recently, the P1PK and GLOB blood group system antigens, specifically, Pk, has been implicated as having a role in HIV infection (Branch, 2010)(see below).

### **4.1 Verotoxin-induced disease**

Enterohemorrhagic *Escherichia coli* induce disease, characteristically haemolytic uremic syndrome (HUS), by the production of verotoxins (VT). VTs are capable of binding to Gb3, thus Gb3 contributes to the pathology of VT-induced disease (Lingwood, 2000, Lingwood et al., 1987 ) VT is comprised of a single toxic 'A' subunit and non-covalently associated pentameric 'B' subunits responsible for receptor (Gb3) binding. Only cells with Gb3 surface expression are sensitive to VT toxicity (Okuda et al., 2006, Waddell et al., 1990). VT interaction with the sugar moiety of Gb3 is dependant on the lipid moiety in its membrane environment (Arab & Lingwood, 1996, Kiarash et al., 1994, Pellizzari et al., 1992), which is crucial in internalization and subcellular targeting of VT (Arab & Lingwood, 1998 , Smith et al., 2006). The intracellular routing of VT thus is also dependant on the organization of Gb3 into lipid rafts (Falguieres et al., 2001). Indeed, VT binding to cell surface Gb3 within lipid microdomains has been shown to activate cytosolic raft-associated src kinase ((Katagiri et al., 1999, Mori et al., 2000)) indicating Gb3 can mediate transmembrane signals. Furthermore, Gb3 containing cells where Gb3 is not present in rafts are insensitive to VT cytotoxicity (Falguieres et al., 2001, Ramegowda & Tesh, 1996). In cells sensitive to VT cytotoxicity, the toxin is internalized via both clathrin independant/dependant pathways(Lauvrak et al., 2004) and undergoes retrograde transport via the reverse of the secretory system to the Golgi and ER/nucleus (Arab & Lingwood, 1998, Khine & Lingwood, 1994, Sandvig et al., 1994) Highly VT sensitive cells contain higher levels of short fatty acid containing Gb3 isoforms and retrograde transport the VT/Gb3 complex to the ER/nucleus. Less VT sensitive cells have longer fatty acid containing Gb3 isoforms and retrograde transport VT to the Golgi only (Arab & Lingwood, 1998). Interestingly, Gb3 is maintained in lipid rafts during retrograde transport (Smith et al., 2006).

#### **4.2 Fabry disease**

Fabry disease is an X-linked lysosomal storage disorder, as a result of a genetic defect in the lysosomal enzyme α-galactosidase A, which results in reduced enzyme activity (Brady, 1967). This enzyme is normally responsible for the removal of the terminal Gb3 galactose residue, through hydrolysis of the 1-4 glycosidic linkage. Thus, Gb3, and potentially other -galactose terminal lipids accumulate in the lysosomes to abrogate their normal function and the function of these organelles. Clinical manifestations of the disorder are related to the

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 251

cholesterol can induce a fatty acid-dependent GSL conformational change (Lingwood et al., 2011). The identification of a family of fatty acyl co-A selective ceramide synthases (Stiban et al., 2010, Teufel et al., 2009) provides the metabolic means to regulate the differential

As indicated, cell membrane GSLs can be organized into cholesterol enriched microdomains. Such microdomains are typically correlated with resistance to detergent extraction *in vitro* (Lingwood & Simons, 2007). While this procedure can induce domain pooling and the relationship between detergent resistance and natural cell membrane GSL domains has yet to be established (Westerlund & Slotte, 2009), detergent resistance indicates stronger lateral membrane interactions. Detergent resistant, cholesterol enriched plasma membrane domains have been shown to be important for HIV infection by most studies (Del Real et al., 2002 , Gummuluru et al., 2003, Liao et al., 2001 , Manes et al., 2000, Nguyen & Hildreth, 2000 , Popik et al., 2002 , Raulin, 2002) but not all (Percherancier et al., 2003).

The role of GSLs in HIV infection must be considered both in terms of GSL species generally distributed in the membrane or restricted to lipid microdomains. Several studies have shown that the binding of the gp120 HIV adhesin to GSL is dependent not only on the carbohydrate, but also the lipid moiety of the GSL (Mahfoud et al., 2009, Mahfoud et al.,

 The interaction of GSLs with cholesterol is modulated by the fatty acid chain length and the binding of HIV gp120 to Gb3/cholesterol vesicles has been shown to be a function of the fatty acid composition in that C16 fatty acid Gb3 was bound but C17, C18 and C20 Gb3 were not. C22 and C24 fatty acid containing Gb3s were bound(Mahfoud et al., 2009). The Gb3 fatty acid isoforms not recognized by gp120 in this context, have fatty acid chain lengths which are of the order of the dimensions of the cholesterol molecule, suggesting that these fatty acid isoforms have the minimum 'hydrophobic mismatch' (Niemela et al., 2009) and therefore interact more effectively with cholesterol. The interaction of GSLs with cholesterol has been shown in modeling studies to induce a conformation change in the headgroup to become parallel to the plane of the cholesterol containing membrane rather than perpendicular, as seen in the absence of cholesterol (Hall et al., 2010, Lingwood et al., 2011, Yahi et al., 2010 ). In such a membrane parallel carbohydrate format, the accessibility of the carbohydrate to carbohydrate binding ligands, such as gp120, will be restricted (but lateral interaction with the membrane may be enhanced). In GSL/cholesterol model detergent resistant membranes separated by sucrose gradient centrifugation, the major GSL fraction was not recognized by gp120 (GSLs –sulfatide, galactosyl ceramide and Gb3) (Mahfoud et al., 2010). Only a minor fraction of smaller vesicles were bound. Such smaller vesicles may display the GSL in a more disperse format, even in the presence of cholesterol, and thereby defray the effect of this potential cholesterol-induced conformational change. Moreover, the fatty acid isoforms of Gb3 negative for gp120 binding were dominant negative in mixtures of saturated Gb3 fatty acid isoforms, whereas addition of the unsaturated C24:1 Gb3 was dominant positive, suggesting that membrane fluidity in these vesicles could be a key factor in determining availability of the GSL carbohydrate for gp120 binding(Mahfoud et al., 2009). Thus the interaction of gp120 with membrane GSLs is extremely complex depending also, on the membrane bilayer organization and perhaps curvature and fluidity. In addition, host cell GSLs taken up into the viral membrane at the time of plasma membrane budding may also play a direct role in HIV dendritic cell targeting (Hatch et al., 2009) and T cell infection.

synthesis of such GSL fatty acid isoforms.

**5.1.2 GSLs and HIV infection** 

2002b, Villard et al., 2002 ).

cell-type-specific expression of Gb3 (Huwiler et al., 2000). Thus patients with Fabry disease typically experience renal dysfunction, myocardial and skin lesions, and joint pain, which relate to the major tissue distribution of Gb3 (Hakomori, 1986).

## **5. HIV and GSLs**

## **5.1 GSL receptors**

Initial contact of HIV with the host cell surface must occur before the virus can initiate infection. HIV envelope glycoprotein gp120 targets CD4 and CCR5 or CXCR4 chemokine coreceptors on monocytes and T-cells respectively, as the major HIV-host cell interaction (Alkhatib et al., 1996, Dalgleish et al., 1984, Feng et al., 1996). GSLs have been implicated in HIV infection since the original description of the binding of GalCer and sulfatide (3' sulfogalactosyl ceramide, SGC) by the HIV adhesin gp120 (Bhat et al., 1993, Bhat et al., 1991) and indeed gp120 binding to these species is considered the primary mechanism by which non-CD4 expressing cells are 'infected' by HIV(Dorosko & Connor, 2010, Harouse et al., 1995, Magerus-Chatinet et al., 2007, Ullrich et al., 1998). GSLs bound by gp120 include GalCer, SGC, Gb3 and the ganglioside GM3 (Delezay et al., 1996, Hammache et al., 1998a). It has been suggested that GM3 is bound only by gp120 from R5 strains whereas Gb3 is bound by both X4 and R5 strains (Hammache et al., 1999). GSL analogues have been shown to inhibit HIV infection(Fantini et al., 1997, Faroux-Corlay et al., 2001, Garg et al., 2008, Lund et al., 2006, Weber et al., 2000)and the efficacy of such analogues depends on the nature of both the carbohydrate and lipid moieties. In addition, GalCer binds to gp120 associated gp41 (Alfsen & Bomsel, 2002), the fusion heptad repeat C-terminal peptide of which, mediates viral/host membrane fusion (Shnaper et al., 2004). Nevertheless, the exact role of the GSLs in HIV infection remains unclear. Early suggestions were that GSL binding within lipid rafts facilitated a simultaneous recognition of CD4 and chemokine receptor by gp120(Fantini, 2003). However, the fact that the GSL-binding site (Delezay et al., 1996), defined as 2 alpha helixes with a central aromatic amino acid sequence(Mahfoud et al., 2002a), responsible for gp120-GSL binding, is contained within the same V3 loop as amino acids crucial for chemokine receptor binding (Xiao et al., 1998), suggest that the binding of GSLs within the V3 loop would more likely provide an inhibitory, rather than stimulatory effect on chemokine receptor binding. To address this potential dichotomy the unusual membrane properties of GSLs must be considered.

### **5.1.1 GSL conformation and lipid heterogeneity**

A single glycosphingolipid (i.e. a single carbohydrate species with a heterogeneous ceramide moiety) can differentially recognize two or more ligands, specific for the carbohydrate sequence. This can be based on differential recognition of the hydroxyl groups within the sugar sequence as has been shown for Verotoxin variants and monoclonal anti-Gb3 (Chark et al., 2004). Differential binding of anti-GM1 and cholera toxin to GM1 lipid isoforms has also been reported (Iglesias-Bartolome et al., 2009). This is consistent with differential ligand recognition of GSL lipid isoforms by ligands which bind the same carbohydrate sequence. The oligosaccharide moiety of glycolipids shows considerable flexibility in conformation and nine potential energy minima have been defined by molecular modeling (Nyholm & Pascher, 1993). This potential for differential carbohydrate conformation which can be regulated by the relative plane of the plasma membrane may therefore reflect the lipid composition and its membrane microenvironment. Indeed, cholesterol can induce a fatty acid-dependent GSL conformational change (Lingwood et al., 2011). The identification of a family of fatty acyl co-A selective ceramide synthases (Stiban et al., 2010, Teufel et al., 2009) provides the metabolic means to regulate the differential synthesis of such GSL fatty acid isoforms.

As indicated, cell membrane GSLs can be organized into cholesterol enriched microdomains. Such microdomains are typically correlated with resistance to detergent extraction *in vitro* (Lingwood & Simons, 2007). While this procedure can induce domain pooling and the relationship between detergent resistance and natural cell membrane GSL domains has yet to be established (Westerlund & Slotte, 2009), detergent resistance indicates stronger lateral membrane interactions. Detergent resistant, cholesterol enriched plasma membrane domains have been shown to be important for HIV infection by most studies (Del Real et al., 2002 , Gummuluru et al., 2003, Liao et al., 2001 , Manes et al., 2000, Nguyen & Hildreth, 2000 , Popik et al., 2002 , Raulin, 2002) but not all (Percherancier et al., 2003).

### **5.1.2 GSLs and HIV infection**

250 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

cell-type-specific expression of Gb3 (Huwiler et al., 2000). Thus patients with Fabry disease typically experience renal dysfunction, myocardial and skin lesions, and joint pain, which

Initial contact of HIV with the host cell surface must occur before the virus can initiate infection. HIV envelope glycoprotein gp120 targets CD4 and CCR5 or CXCR4 chemokine coreceptors on monocytes and T-cells respectively, as the major HIV-host cell interaction (Alkhatib et al., 1996, Dalgleish et al., 1984, Feng et al., 1996). GSLs have been implicated in HIV infection since the original description of the binding of GalCer and sulfatide (3' sulfogalactosyl ceramide, SGC) by the HIV adhesin gp120 (Bhat et al., 1993, Bhat et al., 1991) and indeed gp120 binding to these species is considered the primary mechanism by which non-CD4 expressing cells are 'infected' by HIV(Dorosko & Connor, 2010, Harouse et al., 1995, Magerus-Chatinet et al., 2007, Ullrich et al., 1998). GSLs bound by gp120 include GalCer, SGC, Gb3 and the ganglioside GM3 (Delezay et al., 1996, Hammache et al., 1998a). It has been suggested that GM3 is bound only by gp120 from R5 strains whereas Gb3 is bound by both X4 and R5 strains (Hammache et al., 1999). GSL analogues have been shown to inhibit HIV infection(Fantini et al., 1997, Faroux-Corlay et al., 2001, Garg et al., 2008, Lund et al., 2006, Weber et al., 2000)and the efficacy of such analogues depends on the nature of both the carbohydrate and lipid moieties. In addition, GalCer binds to gp120 associated gp41 (Alfsen & Bomsel, 2002), the fusion heptad repeat C-terminal peptide of which, mediates viral/host membrane fusion (Shnaper et al., 2004). Nevertheless, the exact role of the GSLs in HIV infection remains unclear. Early suggestions were that GSL binding within lipid rafts facilitated a simultaneous recognition of CD4 and chemokine receptor by gp120(Fantini, 2003). However, the fact that the GSL-binding site (Delezay et al., 1996), defined as 2 alpha helixes with a central aromatic amino acid sequence(Mahfoud et al., 2002a), responsible for gp120-GSL binding, is contained within the same V3 loop as amino acids crucial for chemokine receptor binding (Xiao et al., 1998), suggest that the binding of GSLs within the V3 loop would more likely provide an inhibitory, rather than stimulatory effect on chemokine receptor binding. To address this potential dichotomy the unusual membrane

A single glycosphingolipid (i.e. a single carbohydrate species with a heterogeneous ceramide moiety) can differentially recognize two or more ligands, specific for the carbohydrate sequence. This can be based on differential recognition of the hydroxyl groups within the sugar sequence as has been shown for Verotoxin variants and monoclonal anti-Gb3 (Chark et al., 2004). Differential binding of anti-GM1 and cholera toxin to GM1 lipid isoforms has also been reported (Iglesias-Bartolome et al., 2009). This is consistent with differential ligand recognition of GSL lipid isoforms by ligands which bind the same carbohydrate sequence. The oligosaccharide moiety of glycolipids shows considerable flexibility in conformation and nine potential energy minima have been defined by molecular modeling (Nyholm & Pascher, 1993). This potential for differential carbohydrate conformation which can be regulated by the relative plane of the plasma membrane may therefore reflect the lipid composition and its membrane microenvironment. Indeed,

relate to the major tissue distribution of Gb3 (Hakomori, 1986).

**5. HIV and GSLs 5.1 GSL receptors** 

properties of GSLs must be considered.

**5.1.1 GSL conformation and lipid heterogeneity** 

The role of GSLs in HIV infection must be considered both in terms of GSL species generally distributed in the membrane or restricted to lipid microdomains. Several studies have shown that the binding of the gp120 HIV adhesin to GSL is dependent not only on the carbohydrate, but also the lipid moiety of the GSL (Mahfoud et al., 2009, Mahfoud et al., 2002b, Villard et al., 2002 ).

 The interaction of GSLs with cholesterol is modulated by the fatty acid chain length and the binding of HIV gp120 to Gb3/cholesterol vesicles has been shown to be a function of the fatty acid composition in that C16 fatty acid Gb3 was bound but C17, C18 and C20 Gb3 were not. C22 and C24 fatty acid containing Gb3s were bound(Mahfoud et al., 2009). The Gb3 fatty acid isoforms not recognized by gp120 in this context, have fatty acid chain lengths which are of the order of the dimensions of the cholesterol molecule, suggesting that these fatty acid isoforms have the minimum 'hydrophobic mismatch' (Niemela et al., 2009) and therefore interact more effectively with cholesterol. The interaction of GSLs with cholesterol has been shown in modeling studies to induce a conformation change in the headgroup to become parallel to the plane of the cholesterol containing membrane rather than perpendicular, as seen in the absence of cholesterol (Hall et al., 2010, Lingwood et al., 2011, Yahi et al., 2010 ). In such a membrane parallel carbohydrate format, the accessibility of the carbohydrate to carbohydrate binding ligands, such as gp120, will be restricted (but lateral interaction with the membrane may be enhanced). In GSL/cholesterol model detergent resistant membranes separated by sucrose gradient centrifugation, the major GSL fraction was not recognized by gp120 (GSLs –sulfatide, galactosyl ceramide and Gb3) (Mahfoud et al., 2010). Only a minor fraction of smaller vesicles were bound. Such smaller vesicles may display the GSL in a more disperse format, even in the presence of cholesterol, and thereby defray the effect of this potential cholesterol-induced conformational change. Moreover, the fatty acid isoforms of Gb3 negative for gp120 binding were dominant negative in mixtures of saturated Gb3 fatty acid isoforms, whereas addition of the unsaturated C24:1 Gb3 was dominant positive, suggesting that membrane fluidity in these vesicles could be a key factor in determining availability of the GSL carbohydrate for gp120 binding(Mahfoud et al., 2009). Thus the interaction of gp120 with membrane GSLs is extremely complex depending also, on the membrane bilayer organization and perhaps curvature and fluidity. In addition, host cell GSLs taken up into the viral membrane at the time of plasma membrane budding may also play a direct role in HIV dendritic cell targeting (Hatch et al., 2009) and T cell infection.

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 253

Fig. 4. **Conformational change in gp120 induced by CCR5 binding and relation to GSL binding.** The NMR structure of the V3 loop alone (A) or with bound CCR5 N terminus (B) is shown according to Huang et al., 2007. The amino acids of the GSL binding site at the loop apex are boxed in (A). The V3 loop initially disorganized, becomes more rigid on binding of the CCR5 N terminus to the V3 loop base. Gb3 has been arbitrarily placed and oriented with

its glucose moiety stacked over the phenylalanine of the CCR5 unbound loop (C) to illustrate the potential of GSL binding to affect this V3 loop conformational change.

and limited data has been shown for vaginal mucosal transmission

ceramide in gastroepithelial/neuronal/renal cells targeted by HIV (Harouse et al., 1995). Infectious HIV utilizes GSLs and lipid rafts to traverse the host mucosa and access underlying susceptible target cells during transmission. This process is called transcytosis, whereby "receptors" mediate the transcellular traffic of the virus across the tight epithelial cell barrier, rather than productive infection (Bomsel and Alfsen 2003). Thus uptake of the virus occurs at one pole of the cells and infectious virus is released at the opposite pole, gaining access to the submucosa. GalCer has been shown to bind to both HIV-gp120 and gp41 (Alfsen and Bomsel 2001). However, HIV binding to GalCer in epithelial cells does not result in HIV/host cell fusion necessary for productive infection but rather mediates HIV transcytosis (Bomsel, 1997), Bomsel and David 2002). This has been demonstrated in epithelial cell lines, where HIV 'hijacks' the vesicular pathway in order to cross the cell. The process of transcytosis via GalCer has also been shown to occur in primary intestinal epithelial cells, specifically for R5 HIV-1 strains, and is particularly dependant on lipid rafts, as disruption of rafts substantially reduces uptake (Meng Wei 2002). It has further been suggested that transcytosis may occur across specialized M cells, which provide an epithelial barrier, to lymphoid Peyer's patches in the gastrointestinal tract (Fotopoulos 2002). Mucosal dendritic cells express GalCer which can mediate HIV uptake and transfer to T cells (Magerus-Chatinet et al., 2007). To date, the vast majority of evidence suggests transcytosis is a process that facilitates HIV transmission across the gastrointestinal mucosa,

This differential availability of cell membrane Gb3 for example, is dramatically highlighted by the differential binding of various monoclonal antibodies to Gb3 and verotoxin B subunit to lymphoid cells which synthesize Gb3 (Kim et al., 2011), despite the fact that these antibodies show similar efficacy to detect the Gb3 once extracted from the cells and separated by TLC. Gb3 positive cells which do not bind any Gb3 ligands have been reported(Sekino et al., 2004).

The masking of membrane GSL is also dependent on the relative cholesterol concentration. In model GSL membranes, the thickness of the carbohydrate layer was an inverse function of the cholesterol concentration, suggesting that the sugar can adopt intermediate conformations between the membrane perpendicular (thickest) and parallel (thinnest) conformation according to membrane cholesterol content (Lingwood et al., 2011), extending the potential for conformational regulation of GSL receptor function.

The differential expression of membrane GSLs within and without cholesterol enriched lipid rafts may provide the explanation for the differential function ascribed to GSLs in HIV infection; first as promotors of fusion/infection (Puri et al., 1998, Puri et al., 2004) and then as inhibitors of these functions (Lund et al., 2009, Ramkumar et al., 2009). It is conceivable that different GSL carbohydrate conformers can play different roles at different (or the same) times in infection.

Amino acids within the gp120 V3 loop defined by mutational analysis as crucial for chemokine receptor binding(Xiao et al., 1998) coincide with 3 amino acids of the concensus GSL binding site at the V3 loop apex (Delezay et al., 1996), together with 2 distinct amino acids, one in the base of each alpha helix comprising the V3 loop(Xiao et al., 1998). The GSL hexapeptide binding domain has been synthesized as a separate peptide and shown to bind the same GSLs *in vitro* as observed for the intact gp120 (Delezay et al., 1996). The V3 loop must open following CD4 -gp120 binding to allow the chemokine receptor to bind (Wang et al., 1999) whereas gp120 -GSL binding is observed in the absence of CD4 (Mahfoud et al., 2009). Thus GSL binding to the apex of the V3 loop could well alter the alpha helix conformation at the base of the V3 loop to modulate chemokine receptor binding. It is possible that binding of different GSLs, or different lipid isoforms of the same GSL, could differentially alter the conformation of the V3 loop to enhance or inhibit CCR5 binding.

This concept is consistent with NMR studies which indicate that the N terminus of CCR5 binds within the base of the V3 loop (Huang et al., 2007). In combination, soluble CD4 and CCR5 reduced proteolytic susceptibility of the V3 loop of gp120, consistent with binding. A model was proposed (Huang et al., 2007) by which the CCR5 N terminus bound to the base of the V3 loop (via tyrosine sulfate) and then the second extracellular loop of CCR5 associated with the V3 loop apex. Alternatively, the extracellular CCR5 loop associated with the V3 loop apex first, followed by CCR5 N terminus binding to the V3 loop base. The binding of the CCR5 N terminus to the V3 loop base was found to cause a conformational change to rigidify the V3 loop. Such a conformation change might be impeded or promoted by GSL binding within the GSL binding site at the V3 loop apex (Figure 4).

If the first GSL sugar is primarily responsible for binding, it is possible that the effect on V3 conformation could be dependent on the number and character of additional sugar residues. In the absence of the gp120 conformational change induced by CD4 binding, V3 loop binding to GSLs via this apical binding site could mediate a less effective mechanism for HIV internalization. Cell membrane GSLs undergo a natural process of internalization and recycling and GSLs function in receptor mediated endocytosis of appropriate GSL binding ligands. This could thereby provide a basis for the observed association of galactosyl

This differential availability of cell membrane Gb3 for example, is dramatically highlighted by the differential binding of various monoclonal antibodies to Gb3 and verotoxin B subunit to lymphoid cells which synthesize Gb3 (Kim et al., 2011), despite the fact that these antibodies show similar efficacy to detect the Gb3 once extracted from the cells and separated by TLC. Gb3 positive cells which do not bind any Gb3 ligands have been

The masking of membrane GSL is also dependent on the relative cholesterol concentration. In model GSL membranes, the thickness of the carbohydrate layer was an inverse function of the cholesterol concentration, suggesting that the sugar can adopt intermediate conformations between the membrane perpendicular (thickest) and parallel (thinnest) conformation according to membrane cholesterol content (Lingwood et al., 2011), extending

The differential expression of membrane GSLs within and without cholesterol enriched lipid rafts may provide the explanation for the differential function ascribed to GSLs in HIV infection; first as promotors of fusion/infection (Puri et al., 1998, Puri et al., 2004) and then as inhibitors of these functions (Lund et al., 2009, Ramkumar et al., 2009). It is conceivable that different GSL carbohydrate conformers can play different roles at different (or the

Amino acids within the gp120 V3 loop defined by mutational analysis as crucial for chemokine receptor binding(Xiao et al., 1998) coincide with 3 amino acids of the concensus GSL binding site at the V3 loop apex (Delezay et al., 1996), together with 2 distinct amino acids, one in the base of each alpha helix comprising the V3 loop(Xiao et al., 1998). The GSL hexapeptide binding domain has been synthesized as a separate peptide and shown to bind the same GSLs *in vitro* as observed for the intact gp120 (Delezay et al., 1996). The V3 loop must open following CD4 -gp120 binding to allow the chemokine receptor to bind (Wang et al., 1999) whereas gp120 -GSL binding is observed in the absence of CD4 (Mahfoud et al., 2009). Thus GSL binding to the apex of the V3 loop could well alter the alpha helix conformation at the base of the V3 loop to modulate chemokine receptor binding. It is possible that binding of different GSLs, or different lipid isoforms of the same GSL, could differentially alter the conformation of the V3 loop to enhance or inhibit CCR5 binding. This concept is consistent with NMR studies which indicate that the N terminus of CCR5 binds within the base of the V3 loop (Huang et al., 2007). In combination, soluble CD4 and CCR5 reduced proteolytic susceptibility of the V3 loop of gp120, consistent with binding. A model was proposed (Huang et al., 2007) by which the CCR5 N terminus bound to the base of the V3 loop (via tyrosine sulfate) and then the second extracellular loop of CCR5 associated with the V3 loop apex. Alternatively, the extracellular CCR5 loop associated with the V3 loop apex first, followed by CCR5 N terminus binding to the V3 loop base. The binding of the CCR5 N terminus to the V3 loop base was found to cause a conformational change to rigidify the V3 loop. Such a conformation change might be impeded or promoted

the potential for conformational regulation of GSL receptor function.

by GSL binding within the GSL binding site at the V3 loop apex (Figure 4).

If the first GSL sugar is primarily responsible for binding, it is possible that the effect on V3 conformation could be dependent on the number and character of additional sugar residues. In the absence of the gp120 conformational change induced by CD4 binding, V3 loop binding to GSLs via this apical binding site could mediate a less effective mechanism for HIV internalization. Cell membrane GSLs undergo a natural process of internalization and recycling and GSLs function in receptor mediated endocytosis of appropriate GSL binding ligands. This could thereby provide a basis for the observed association of galactosyl

reported(Sekino et al., 2004).

same) times in infection.

Fig. 4. **Conformational change in gp120 induced by CCR5 binding and relation to GSL binding.** The NMR structure of the V3 loop alone (A) or with bound CCR5 N terminus (B) is shown according to Huang et al., 2007. The amino acids of the GSL binding site at the loop apex are boxed in (A). The V3 loop initially disorganized, becomes more rigid on binding of the CCR5 N terminus to the V3 loop base. Gb3 has been arbitrarily placed and oriented with its glucose moiety stacked over the phenylalanine of the CCR5 unbound loop (C) to illustrate the potential of GSL binding to affect this V3 loop conformational change.

ceramide in gastroepithelial/neuronal/renal cells targeted by HIV (Harouse et al., 1995). Infectious HIV utilizes GSLs and lipid rafts to traverse the host mucosa and access underlying susceptible target cells during transmission. This process is called transcytosis, whereby "receptors" mediate the transcellular traffic of the virus across the tight epithelial cell barrier, rather than productive infection (Bomsel and Alfsen 2003). Thus uptake of the virus occurs at one pole of the cells and infectious virus is released at the opposite pole, gaining access to the submucosa. GalCer has been shown to bind to both HIV-gp120 and gp41 (Alfsen and Bomsel 2001). However, HIV binding to GalCer in epithelial cells does not result in HIV/host cell fusion necessary for productive infection but rather mediates HIV transcytosis (Bomsel, 1997), Bomsel and David 2002). This has been demonstrated in epithelial cell lines, where HIV 'hijacks' the vesicular pathway in order to cross the cell. The process of transcytosis via GalCer has also been shown to occur in primary intestinal epithelial cells, specifically for R5 HIV-1 strains, and is particularly dependant on lipid rafts, as disruption of rafts substantially reduces uptake (Meng Wei 2002). It has further been suggested that transcytosis may occur across specialized M cells, which provide an epithelial barrier, to lymphoid Peyer's patches in the gastrointestinal tract (Fotopoulos 2002). Mucosal dendritic cells express GalCer which can mediate HIV uptake and transfer to T cells (Magerus-Chatinet et al., 2007). To date, the vast majority of evidence suggests transcytosis is a process that facilitates HIV transmission across the gastrointestinal mucosa, and limited data has been shown for vaginal mucosal transmission

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 255

requires lipid rafts in the process of entry, but in immune evasion, suppressing host-cell signalling during replication, and egress of the virus from the host (Manes et al., 2000,

HIV Nef is a myristoylated protein that is associated with rafts, and this association is necessary for its function. Nef is involved in down-regulation of CD4 and MHC-I molecules, crucial for viral infectivity and immune evasion. These functions are dependant on the Nef targeting and trafficking function, through clathrin coated pits and early endosome associations, and are thus dependant on lipid raft (Bresnahan et al., 1998 , Piguet & Trono, 1999). Interestingly, Nef has been shown to inhibit Gb3 retrograde transport (Johannes et al.,

GSL-enriched lipid rafts are required for viral egress in addition to entry. Assembly and incorporation of envelope glycoproteins in the virion envelope is regulated by an interaction between the gp41 cytoplasmic tail and the MA domain of the Gag precursor peptide (Hourioux, Brand 2000). During post-transctiptional modification, the MA domain of Gag is myristoylated, and the envelope precursor gp160 is palmitoylated. These modifications target these proteins to lipid rafts, which promote the assembly of budding virions (Ono & Freed, 2001) . Furthermore, because HIV selectively buds from lipid rafts, the viral envelope is enriched in lipid raft components, including cholesterol and GlcCer (Brugger et al., 2006, Nguyen & Hildreth, 2000). The viral membrane GSL content can affect infectious potential

Increased Gb3 and GM3 synthesis can be detected at an early stage in HIV-1 infected individuals. In addition, antibodies to these GSLs have been detected in HIV patients (Fantini et al., 1998b). These GSLs have important functions within the immune system, with regards to cell growth, signalling and motility. They are of particular importance as markers in lymphocyte differentiation, where Gb3 is a marker of B cell development (Wiels et al., 1991)and GM3 of monocytes and T-cells. In addition, GM3 containing microdomains are functional in T cell motility (Gomez-Mouton et al., 2001) and signalling (Sorice et al., 2000). Thus, perturbations in GSL expression, and antibodies produced to GSL in HIV-1 infection

It is of interest to note that HIV infected patients are more prone to haemolytic uremic syndrome (HUS) (Turner et al., 1997). HUS is characterised by thrombotic microangiopathy of the renal glomeruli mediated by verotoxin/Gb3 binding(Muthing et al., 2009). It is thus interesting that transgenic mice, in which the HIV genome has been incorporated into the germ line, show renal Gb3 synthesis is selectively upregulated to induce renal disease (Liu et

In the quest for new drug targets, such as the entry inhibitors, and subsequent potential microbicide candidates, attention has been turned to HIV interactions with lipid rafts and GSLs. Several studies have investigated cholesterol-depletion as a means of disrupting lipid rafts to prevent HIV-1 fusion and entry (Liao et al., 2001; Liao et al., 2003) It has also been proposed that increasing ceramide levels in CD4+ lymphocytes and monocyte-derived macrophages may block HIV infection, perhaps inhibiting HIV fusion by disrupting normal lipid raft organization and function (Finnegan & Blumenthal, 2006).These studies have used several mechanisms to increase ceramide, including pharmacological agents, such as *N*-(4-

Nguyen & Hildreth, 2000, Peterlin & Trono, 2003)

2003).

(Hatch et al., 2009)

**5.4 Clinical links to GSL** 

may be immunosuppressive.

**5.5 Inhibiting HIV at the membrane level** 

al., 1999)

## **5.2 Membrane fusion**

For productive infection, HIV enters cells directly via plasma membrane penetration, which requires fusion of the viral envelope with the host cell membrane (Marsh & Helenius, 1989). Membrane fusion is particularly dependant on lipid rafts, which have a central role in HIV infection. Depletion of cholesterol, a key component of lipid rafts, renders cells resistant to infection and membrane fusion, a phenotype rescued upon re-introduction of cholesterol (Manes, del Real 2000, Liao, Cimakasky et al 2001). GSLs, important components of lipid rafts, have also been shown to play a role in membrane fusion. Complete lack of GSLs protects CD4 positive cells from HIV infection (Hug et al., 2000, Puri et al., 2004, Rawat et al., 2003).Interestingly, reconstitution of GSL deficient cells with Gb3, and to a lesser extent GM3, was able to restore membrane fusion in these model systems (Hug et al., 2000, Puri et al., 1998, Puri et al., 1999). However, no other GSLs were able to rescue this phenotype. This impediment could also be overcome by the over-expression of CD4 and CXCR4, suggesting the role for GSLs is facilitative (Hammache et al., 1999, Puri et al., 1998, Puri et al., 1999, Rawat et al., 2003). These findings have been supported by reports that both Gb3 and GM3, when introduced into the cell membrane of CD4+ T lymphocytes, have the potential to enhance HIV-1 fusion and entry of a broad range of isolates (Hug et al., 2000). It has also been shown that non-human cells expressing CD4, ordinarily not permissive to HIV-1 infection, become permissive to membrane fusion upon introduction of Gb3 (Puri et al., 1998). Our studies using a different glucosyl ceramide synthase inhibitor are conssitent with an inhibitory role for GSLs(Gb3)(Ramkumar et al., 2009) and HIV resistance is also conferred by high GM3 levels (Rawat et al., 2004).

For membrane fusion to proceed, HIV-gp120 binding to CD4 and chemokine co-receptor, must initiate conformational change in gp120 and the associated transmembrane gp41 (Freed et al., 1992, Jones et al., 1998). At physiological levels, CD4 and the co-receptors are not physically associated in the membrane in the absence of HIV-1 (Jones, korte et al 1998). However, CD4 and CCR5 are both present in lipid rafts, albeit separate rafts, and their associations with rafts have been shown to be required for infection. Indeed, both CD4 and CCR5 may interact with lipid rafts containing GM3 and Gb3 (Hammache et al., 1999, Hammache et al., 1998b, Manes et al., 2001, Millan et al., 1999, Sorice et al., 1997). Interestingly, CXCR4 is not normally associated with rafts, and is separated from CD4, which is ordinarily associated with GM3 rafts. Upon HIV-gp120 interactions with CD4 however, CXCR4 is physically recruited into these rafts for membrane fusion (Sorice et al., 2000, Sorice et al., 2001).

CD4 has been shown to insert into Gb3 or GM3 monolayers, as has HIV-gp120 (Hammache et al., 1999). Since both CD4 and chemokine receptors are found in, or are recruited to, lipid rafts for HIV infection, it was proposed that CD4 binds GSLs in rafts to promote gp120/GSL interactions (Fantini et al., 2000). GSL within rafts may then function to promote the migration of the CD4-gp120 complex to an appropriate, initially distal, coreceptor (Hammache et al., 1999). This would in turn promote clustering and thus co-operative interactions between the CD4-GSL-chemokine coreceptors (Rawat et al., 2006). HIV-gp120 binding interactions within the GSL-containing domain could then induce the conformational changes necessary to effect membrane fusion. Indeed, the fusion complex has specifically been shown to assemble in lipid rafts (Manes, et al., 2000 ).

#### **5.3 Infectivity and viral egress**

It is interesting to note that there appears to be an overall role for GSL containing lipid rafts in HIV infection (Manes et al., 2000, Popik et al., 2002). More specifically, HIV not only requires lipid rafts in the process of entry, but in immune evasion, suppressing host-cell signalling during replication, and egress of the virus from the host (Manes et al., 2000, Nguyen & Hildreth, 2000, Peterlin & Trono, 2003)

HIV Nef is a myristoylated protein that is associated with rafts, and this association is necessary for its function. Nef is involved in down-regulation of CD4 and MHC-I molecules, crucial for viral infectivity and immune evasion. These functions are dependant on the Nef targeting and trafficking function, through clathrin coated pits and early endosome associations, and are thus dependant on lipid raft (Bresnahan et al., 1998 , Piguet & Trono, 1999). Interestingly, Nef has been shown to inhibit Gb3 retrograde transport (Johannes et al., 2003).

GSL-enriched lipid rafts are required for viral egress in addition to entry. Assembly and incorporation of envelope glycoproteins in the virion envelope is regulated by an interaction between the gp41 cytoplasmic tail and the MA domain of the Gag precursor peptide (Hourioux, Brand 2000). During post-transctiptional modification, the MA domain of Gag is myristoylated, and the envelope precursor gp160 is palmitoylated. These modifications target these proteins to lipid rafts, which promote the assembly of budding virions (Ono & Freed, 2001) . Furthermore, because HIV selectively buds from lipid rafts, the viral envelope is enriched in lipid raft components, including cholesterol and GlcCer (Brugger et al., 2006, Nguyen & Hildreth, 2000). The viral membrane GSL content can affect infectious potential (Hatch et al., 2009)

### **5.4 Clinical links to GSL**

254 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

For productive infection, HIV enters cells directly via plasma membrane penetration, which requires fusion of the viral envelope with the host cell membrane (Marsh & Helenius, 1989). Membrane fusion is particularly dependant on lipid rafts, which have a central role in HIV infection. Depletion of cholesterol, a key component of lipid rafts, renders cells resistant to infection and membrane fusion, a phenotype rescued upon re-introduction of cholesterol (Manes, del Real 2000, Liao, Cimakasky et al 2001). GSLs, important components of lipid rafts, have also been shown to play a role in membrane fusion. Complete lack of GSLs protects CD4 positive cells from HIV infection (Hug et al., 2000, Puri et al., 2004, Rawat et al., 2003).Interestingly, reconstitution of GSL deficient cells with Gb3, and to a lesser extent GM3, was able to restore membrane fusion in these model systems (Hug et al., 2000, Puri et al., 1998, Puri et al., 1999). However, no other GSLs were able to rescue this phenotype. This impediment could also be overcome by the over-expression of CD4 and CXCR4, suggesting the role for GSLs is facilitative (Hammache et al., 1999, Puri et al., 1998, Puri et al., 1999, Rawat et al., 2003). These findings have been supported by reports that both Gb3 and GM3, when introduced into the cell membrane of CD4+ T lymphocytes, have the potential to enhance HIV-1 fusion and entry of a broad range of isolates (Hug et al., 2000). It has also been shown that non-human cells expressing CD4, ordinarily not permissive to HIV-1 infection, become permissive to membrane fusion upon introduction of Gb3 (Puri et al., 1998). Our studies using a different glucosyl ceramide synthase inhibitor are conssitent with an inhibitory role for GSLs(Gb3)(Ramkumar et al., 2009) and HIV resistance is also

For membrane fusion to proceed, HIV-gp120 binding to CD4 and chemokine co-receptor, must initiate conformational change in gp120 and the associated transmembrane gp41 (Freed et al., 1992, Jones et al., 1998). At physiological levels, CD4 and the co-receptors are not physically associated in the membrane in the absence of HIV-1 (Jones, korte et al 1998). However, CD4 and CCR5 are both present in lipid rafts, albeit separate rafts, and their associations with rafts have been shown to be required for infection. Indeed, both CD4 and CCR5 may interact with lipid rafts containing GM3 and Gb3 (Hammache et al., 1999, Hammache et al., 1998b, Manes et al., 2001, Millan et al., 1999, Sorice et al., 1997). Interestingly, CXCR4 is not normally associated with rafts, and is separated from CD4, which is ordinarily associated with GM3 rafts. Upon HIV-gp120 interactions with CD4 however, CXCR4 is physically recruited into these rafts for

CD4 has been shown to insert into Gb3 or GM3 monolayers, as has HIV-gp120 (Hammache et al., 1999). Since both CD4 and chemokine receptors are found in, or are recruited to, lipid rafts for HIV infection, it was proposed that CD4 binds GSLs in rafts to promote gp120/GSL interactions (Fantini et al., 2000). GSL within rafts may then function to promote the migration of the CD4-gp120 complex to an appropriate, initially distal, coreceptor (Hammache et al., 1999). This would in turn promote clustering and thus co-operative interactions between the CD4-GSL-chemokine coreceptors (Rawat et al., 2006). HIV-gp120 binding interactions within the GSL-containing domain could then induce the conformational changes necessary to effect membrane fusion. Indeed, the fusion complex

It is interesting to note that there appears to be an overall role for GSL containing lipid rafts in HIV infection (Manes et al., 2000, Popik et al., 2002). More specifically, HIV not only

has specifically been shown to assemble in lipid rafts (Manes, et al., 2000 ).

**5.2 Membrane fusion** 

conferred by high GM3 levels (Rawat et al., 2004).

membrane fusion (Sorice et al., 2000, Sorice et al., 2001).

**5.3 Infectivity and viral egress** 

Increased Gb3 and GM3 synthesis can be detected at an early stage in HIV-1 infected individuals. In addition, antibodies to these GSLs have been detected in HIV patients (Fantini et al., 1998b). These GSLs have important functions within the immune system, with regards to cell growth, signalling and motility. They are of particular importance as markers in lymphocyte differentiation, where Gb3 is a marker of B cell development (Wiels et al., 1991)and GM3 of monocytes and T-cells. In addition, GM3 containing microdomains are functional in T cell motility (Gomez-Mouton et al., 2001) and signalling (Sorice et al., 2000). Thus, perturbations in GSL expression, and antibodies produced to GSL in HIV-1 infection may be immunosuppressive.

It is of interest to note that HIV infected patients are more prone to haemolytic uremic syndrome (HUS) (Turner et al., 1997). HUS is characterised by thrombotic microangiopathy of the renal glomeruli mediated by verotoxin/Gb3 binding(Muthing et al., 2009). It is thus interesting that transgenic mice, in which the HIV genome has been incorporated into the germ line, show renal Gb3 synthesis is selectively upregulated to induce renal disease (Liu et al., 1999)

#### **5.5 Inhibiting HIV at the membrane level**

In the quest for new drug targets, such as the entry inhibitors, and subsequent potential microbicide candidates, attention has been turned to HIV interactions with lipid rafts and GSLs. Several studies have investigated cholesterol-depletion as a means of disrupting lipid rafts to prevent HIV-1 fusion and entry (Liao et al., 2001; Liao et al., 2003) It has also been proposed that increasing ceramide levels in CD4+ lymphocytes and monocyte-derived macrophages may block HIV infection, perhaps inhibiting HIV fusion by disrupting normal lipid raft organization and function (Finnegan & Blumenthal, 2006).These studies have used several mechanisms to increase ceramide, including pharmacological agents, such as *N*-(4-

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 257

modulating host/microbial interactions. In collaboration with Fantini's group , we showed that gp120 can insert into a Gb3 monolayer at the water/air interphase in a Langmuir trough (Mahfoud et al., 2002b). However, there was a 2 hour lag-phase prior to binding/insertion which then proceeded at a sigmoidal rate. AdaGb3, although water-soluble, also can form a monolayer at a water/air surface interface. Binding and insertion of gp120 into such a monolayer was exponential and immediate. Thus adaGb3 is by far (>1000x), a superior ligand for gp120 than the native Gb3. These results were duplicated using the SPC3 peptide from the V3 loop of gp120 which contains the GSL binding domain (Delezay et al., 1996). In the studies with the peptide, the lag-phase prior to Gb3 binding/insertion was even more exaggerated, being three hours as compared to immediate insertion into adaGb3. The lag phase observed for gp120 insertion into Gb3 monolayers was removed if the monolayer is formed with 20% cholesterol. This suggests that the gp120 may be interacting with Gb3 containing lipid rafts or microdomains (of which cholesterol is a key component). The lag phase seen in the absence of cholesterol, may be a function of the ability of Gb3 to organize into suitable microdomains for gp120 binding and the sigmoidal curve suggests a cooperative effect, once a few domains have been formed. The immediate binding and insertion into adaGb3 monolayers suggests that this organization required for gp120 insertion is already present in the adaGb3 monolayer. Interestingly, although gp120 binds SGC, we found no evidence for gp120 insertion into SGC monolayers, even in the presence of cholesterol. Similarly, no gp120 insertion into adamantylSGC monolayers was seen(Mahfoud et al., 2002b). Our recent work showing that 50% cholesterol can mask membrane Gb3 from gp120(Mahfoud et al., 2010)

SGC was shown to inhibit HIV infection of CD4 negative HT29 cells (Fantini et al., 1998a) without inhibition of HIV cell binding. In these studies, SGC was incorporated into the host cell membrane thereby increasing HIV binding, since gp120 binds SGC(Bhat et al., 1993), but fusion with the host cell membrane was inhibited(Fantini et al., 1998a). Although these studies also implicated GalCer in these cells as mediating HIV infection, HT29 cells are Gb3 positive. Thus, this is consistent with a role for Gb3 rather than SGC, in HIV–cell fusion. It is possible that in addition to forming microdomains poorly itself, SGC could interfere with rafts containing other GSLs. AdamantylSGC (Whetstone & Lingwood, 2003) is a soluble inhibitor of gp120-SGC binding and may prove more effective than the poorly soluble SGC.

Comparison of the "compressibility" of Gb3 and adaGb3 monolayers shows that the adaGb3 structure is more rigid and able to withstand greater increases in surface pressure without collapsing (Mahfoud et al., 2002b). This is consistent with a microdomain format for the adaGb3 monolayer. If adaGb3 is a "superligand" for gp120 as our studies indicate, adaGb3 might be able to interfere with the process of HIV infection even for (T) cells which do not express Gb3 (Akashi et al., 1988). We therefore tested whether adaGb3 was able to modify HIV infectivity *in vitro*. 200µM adaGb3 was able to reduce HIV infectivity in Jurkat T cells using a multiplicity of infectivity ratio of 0.6 (60x higher than standard practice) by ~70% over a 4 day infection period (figure 9) as monitored by ELISA of host cell production of viral nucleoprotein p24gag. Amino adamantane itself showed no inhibition. *Thus, this approach does represent a novel basis for the control of HIV infectivity*. Moreover, in our studies to use adaGb3 to protect mice against VT, we have shown that adaGb3 itself (4mg/kg) shows

indicates a bimodal concentration dependent cholesterol effect.

**6.2 AdaGb3 inhibits HIV infection** 

no side effects *in vivo*.

hydroxyphenyl) retinamide and fenretinide, treatment with sphingomyelinase or addition of long-chain ceramide. Lipid-raft altering compounds may have dual efficacy in treatment of HIV/AIDS. Microorganisms causing opportunistic infections in AIDS patients often rely on lipid-raft mediated mechanisms to elicit their effect, thus HIV treatments altering lipidrafts may be protective.

As GSLs are critical in the process of HIV infection and pathogenesis, targeting of these molecules may give rise to the development of novel therapeutics. Not only are GSLs key components of lipid rafts, but they also play several roles during HIV binding and host cell fusion. The efficacy of inhibiting HIV infection by targeting GSLs has already been demonstrated *in vitro*. Peptide analogues of the V3 loop of gp120, including those that define the GSL binding site, are effective as inhibitors of HIV-membrane fusion (Delezay et al., 1996, Savarino et al., 2003). Furthermore, analogues of galactosyl ceramide have been found to be protective against T cell infection *in vitro*, where the hydrophobic aglycone moiety of GalCer played an important role (Fantini, 2000, Fantini et al., 1997, Faroux-Corlay et al., 2001)

## **6. Generation of GSL mimics**

Given the importance of GSLs in HIV infection, and the demonstrated anti-HIV potential of GalCer analogues, it is particularly advantageous to develop soluble GSL analogues. GSL binding and receptor function is significantly regulated by lipid modulation (Lingwood, 1996). Despite the fact that the carbohydrate moiety of the GSL defines the specificity of binding interactions, the lipid-free sugar shows minimal binding activity (Boyd et al., 1994, Mamelak et al., 2001a). Thus, gp120 binding to the GSL receptors is abrogated if the lipid moiety, that is the anchor to the cell membrane, is removed (Faroux-Corlay et al., 2001, Mylvaganam & Lingwood, 1999b, Villard et al., 2002).In the membrane bilayer, GSLs comprise three domains the external aqueous sugar domain, the internal liquid crystalline domain and the "interface" between them. The "interface" region modulates the receptor function of the carbohydrate in response to the liquid crystalline domain, and likely plays a role in lipid raft organization.

In order to generate GSL analogues and maintain the interface character, an adamantane frame was used to replace the fatty acid (Mylvaganam & Lingwood, 2003). This rigid, globular, cagelike hydrophobic structure close to the interface region perturbs the lateral packing of the glycolipid, and thus bilayer structure formation, thereby promoting solubility. The Gb3 analogue, adamantylGb3 (adaGb3), was shown to preferentially partition into water in an organic/aqueous solvent system (Mylvaganum & Lingwood, 1999a). This compound, unlike the lipid-free Gb3 sugar, maintained its receptor function and was able to inhibit VT/ Gb3 binding, protecting cells against this toxin (Mylvaganam & Lingwood, 1999a).

We utilized the same strategy to develop a soluble analogue of SGC and GalCer. We substituted the fatty acid of SGC with an adamantane or with a norbornane (smaller) frame and, as with the Gb3 case, the conjugates partitioned into water (although adamantylGalCer was significantly less soluble), rather than the organic phase (Whetstone & Lingwood, 2003). AdamantylSGC retained its receptor activity (Mamelak et al., 2001a, Whetstone & Lingwood, 2003)

#### **6.1 AdaGb3 as a mimic of lipid rafts: a 'superligand' for HIV gp120**

We have found that adaGb3 has a variety of additional unusual physical properties which indicate that adamantyl-glycolipids may have unusual biological effects, particularly in

hydroxyphenyl) retinamide and fenretinide, treatment with sphingomyelinase or addition of long-chain ceramide. Lipid-raft altering compounds may have dual efficacy in treatment of HIV/AIDS. Microorganisms causing opportunistic infections in AIDS patients often rely on lipid-raft mediated mechanisms to elicit their effect, thus HIV treatments altering lipid-

As GSLs are critical in the process of HIV infection and pathogenesis, targeting of these molecules may give rise to the development of novel therapeutics. Not only are GSLs key components of lipid rafts, but they also play several roles during HIV binding and host cell fusion. The efficacy of inhibiting HIV infection by targeting GSLs has already been demonstrated *in vitro*. Peptide analogues of the V3 loop of gp120, including those that define the GSL binding site, are effective as inhibitors of HIV-membrane fusion (Delezay et al., 1996, Savarino et al., 2003). Furthermore, analogues of galactosyl ceramide have been found to be protective against T cell infection *in vitro*, where the hydrophobic aglycone moiety of GalCer

played an important role (Fantini, 2000, Fantini et al., 1997, Faroux-Corlay et al., 2001)

binding, protecting cells against this toxin (Mylvaganam & Lingwood, 1999a).

**6.1 AdaGb3 as a mimic of lipid rafts: a 'superligand' for HIV gp120** 

We utilized the same strategy to develop a soluble analogue of SGC and GalCer. We substituted the fatty acid of SGC with an adamantane or with a norbornane (smaller) frame and, as with the Gb3 case, the conjugates partitioned into water (although adamantylGalCer was significantly less soluble), rather than the organic phase (Whetstone & Lingwood, 2003). AdamantylSGC retained its receptor activity (Mamelak et al., 2001a, Whetstone &

We have found that adaGb3 has a variety of additional unusual physical properties which indicate that adamantyl-glycolipids may have unusual biological effects, particularly in

Given the importance of GSLs in HIV infection, and the demonstrated anti-HIV potential of GalCer analogues, it is particularly advantageous to develop soluble GSL analogues. GSL binding and receptor function is significantly regulated by lipid modulation (Lingwood, 1996). Despite the fact that the carbohydrate moiety of the GSL defines the specificity of binding interactions, the lipid-free sugar shows minimal binding activity (Boyd et al., 1994, Mamelak et al., 2001a). Thus, gp120 binding to the GSL receptors is abrogated if the lipid moiety, that is the anchor to the cell membrane, is removed (Faroux-Corlay et al., 2001, Mylvaganam & Lingwood, 1999b, Villard et al., 2002).In the membrane bilayer, GSLs comprise three domains the external aqueous sugar domain, the internal liquid crystalline domain and the "interface" between them. The "interface" region modulates the receptor function of the carbohydrate in response to the liquid crystalline domain, and likely plays a role in lipid raft organization. In order to generate GSL analogues and maintain the interface character, an adamantane frame was used to replace the fatty acid (Mylvaganam & Lingwood, 2003). This rigid, globular, cagelike hydrophobic structure close to the interface region perturbs the lateral packing of the glycolipid, and thus bilayer structure formation, thereby promoting solubility. The Gb3 analogue, adamantylGb3 (adaGb3), was shown to preferentially partition into water in an organic/aqueous solvent system (Mylvaganum & Lingwood, 1999a). This compound, unlike the lipid-free Gb3 sugar, maintained its receptor function and was able to inhibit VT/ Gb3

rafts may be protective.

**6. Generation of GSL mimics** 

Lingwood, 2003)

modulating host/microbial interactions. In collaboration with Fantini's group , we showed that gp120 can insert into a Gb3 monolayer at the water/air interphase in a Langmuir trough (Mahfoud et al., 2002b). However, there was a 2 hour lag-phase prior to binding/insertion which then proceeded at a sigmoidal rate. AdaGb3, although water-soluble, also can form a monolayer at a water/air surface interface. Binding and insertion of gp120 into such a monolayer was exponential and immediate. Thus adaGb3 is by far (>1000x), a superior ligand for gp120 than the native Gb3. These results were duplicated using the SPC3 peptide from the V3 loop of gp120 which contains the GSL binding domain (Delezay et al., 1996). In the studies with the peptide, the lag-phase prior to Gb3 binding/insertion was even more exaggerated, being three hours as compared to immediate insertion into adaGb3. The lag phase observed for gp120 insertion into Gb3 monolayers was removed if the monolayer is formed with 20% cholesterol. This suggests that the gp120 may be interacting with Gb3 containing lipid rafts or microdomains (of which cholesterol is a key component). The lag phase seen in the absence of cholesterol, may be a function of the ability of Gb3 to organize into suitable microdomains for gp120 binding and the sigmoidal curve suggests a cooperative effect, once a few domains have been formed. The immediate binding and insertion into adaGb3 monolayers suggests that this organization required for gp120 insertion is already present in the adaGb3 monolayer. Interestingly, although gp120 binds SGC, we found no evidence for gp120 insertion into SGC monolayers, even in the presence of cholesterol. Similarly, no gp120 insertion into adamantylSGC monolayers was seen(Mahfoud et al., 2002b). Our recent work showing that 50% cholesterol can mask membrane Gb3 from gp120(Mahfoud et al., 2010) indicates a bimodal concentration dependent cholesterol effect.

SGC was shown to inhibit HIV infection of CD4 negative HT29 cells (Fantini et al., 1998a) without inhibition of HIV cell binding. In these studies, SGC was incorporated into the host cell membrane thereby increasing HIV binding, since gp120 binds SGC(Bhat et al., 1993), but fusion with the host cell membrane was inhibited(Fantini et al., 1998a). Although these studies also implicated GalCer in these cells as mediating HIV infection, HT29 cells are Gb3 positive. Thus, this is consistent with a role for Gb3 rather than SGC, in HIV–cell fusion. It is possible that in addition to forming microdomains poorly itself, SGC could interfere with rafts containing other GSLs. AdamantylSGC (Whetstone & Lingwood, 2003) is a soluble inhibitor of gp120-SGC binding and may prove more effective than the poorly soluble SGC.

#### **6.2 AdaGb3 inhibits HIV infection**

Comparison of the "compressibility" of Gb3 and adaGb3 monolayers shows that the adaGb3 structure is more rigid and able to withstand greater increases in surface pressure without collapsing (Mahfoud et al., 2002b). This is consistent with a microdomain format for the adaGb3 monolayer. If adaGb3 is a "superligand" for gp120 as our studies indicate, adaGb3 might be able to interfere with the process of HIV infection even for (T) cells which do not express Gb3 (Akashi et al., 1988). We therefore tested whether adaGb3 was able to modify HIV infectivity *in vitro*. 200µM adaGb3 was able to reduce HIV infectivity in Jurkat T cells using a multiplicity of infectivity ratio of 0.6 (60x higher than standard practice) by ~70% over a 4 day infection period (figure 9) as monitored by ELISA of host cell production of viral nucleoprotein p24gag. Amino adamantane itself showed no inhibition. *Thus, this approach does represent a novel basis for the control of HIV infectivity*. Moreover, in our studies to use adaGb3 to protect mice against VT, we have shown that adaGb3 itself (4mg/kg) shows no side effects *in vivo*.

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 259

Rosa Borges et al. (2010) found that if the synthesized analogues utilized multimeric Gb3 sites, increased efficacy for inhibition of HIV-1 could be achieved. These investigators covalently attached multiple head groups of either Gb3 or GM3 to a dendrimer core to produce multivalent dendrimeric compounds that were water soluble and showed a much lower IC50 for inhibition of HIV-1 infection, compared to either FSL-Gb3 or adaGb3 (Rosa Borges et al., 2010). Thus, soluble Gb3 analogues have important therapeutic potential to

1-Deoxygalactonojirimycin (DGJ) is an alkylated imino sugar with a galactose head and a onecarbon side chain. It was found to be highly tolerable *in vivo* and established as a potent competitive inhibitor of *α-galactosidase A* (α-Gal A)(Hamanaka et al., 2008). DGJ has been proposed as a specific chemical chaperone for treatment of diseases including Fabry (Fan et al., 1999). Studies have indicated that oral administration of DGJ to transgenic mice expressing a human mutant α-Gal A substantially elevated enzyme activity in major organs (Fan et al., 1999). Because Gb3 has now been suggested by the studies of Lund et al. (Lund et al., 2006). as an important component for prevention of the HIV entry mechanism, the use of DGJ to pharmacologically increase Gb3 expression may be useful for HIV prevention strategies. Recent studies (Ramkumar et al., 2009) used DGJ to increase the cell-surface expression of Gb3 on the monocyte cell line, THP-1, which is infectable with X4 HIV-1. DGJ used at 100 μM was able to increase the cell-surface expression of Gb3 of THP-1 cells by approximately

To inhibit Gb3 expression in these cells, these investigators used the compound, D-threo-1 phenyl-2-palmitoylamino-3-pyrrolidino-1-propanol (P4) which inhibits GlcCer synthase, the enzyme responsible for producing GlcCer (Inokuchi & Radin, 1987). This is a key enzyme in the biosynthesis of GSLs because most are glucosylceramide-based. P4 is the most potent inhibitor of this glycosyltransferase (Lee et al., 1999). Using 2 μM P4, Ramkumar et al (Ramkumar et al., 2009) were able to completely inhibit Gb3 expression with subsequent X4 HIV-1 infection increased up to 20-fold. DGJ had little effect on the infection of a Gb3 negative subclone of THP-1 cells and FACS analysis indicated that after DGJ treatment, CD4 and HIV co-receptor levels were similar in the Gb3 expressing and non-expressing THP-1 cell lines by these two compounds. Also, as DGJ was found to be non-toxic in the THP-1 cells up to concentrations of 300 µM, the inhibitory effect was not a result of cytotoxicity of DGJ treatment. The authors concluded that their results indicated that the difference in HIV

infection was due solely to the modulation of the expression of the levels of Gb3.

To examine the effects of DGJ and P4 treatment on subsequent infection with R5 HIV-1 virus, the glioblastoma cell line, U87, that had been transfected to express CD4 and the chemokine co-receptor, CCR5, was used. Ramkumar et al. (Ramkumar et al., 2009) again found that treatment of these cells with DGJ resulted in a significant inhibition of R5 HIV-1 infection while treatment with P4 caused a doubling in the infection. They concluded that pharmacologically increasing Gb3 expression using DGJ treatment or inhibition of Gb3 expression using P4 demonstrates a linear relationship between Gb3 expression and infection with either X4 or R5 HIV-1. In addition, their studies suggest that pharmacologically increasing Gb3 is an effective and novel means to prevent HIV-1 infection *in vitro* and that this approach should be explored for *in vivo* treatment of HIV infection.

block HIV from interacting with CD4+ target cells (Figure 5D).

20-fold. Subsequent X4 HIV-1 infection was decreased significantly.

**7. Pharmacologic modulation of Gb3 expression** 

In collaboration with Blumenthal's group at NIH we have found that adaGb3 is also able to inhibit gp120/CD4/chemokine coreceptor dependent host cell fusion irrespective of gp120 type (R5 or X4 tropic, HIV-1 or HIV-2) as monitored in an indicator system in which gp120 is transfected into one indicator cell and the chemokine receptor into another (Lund et al., 2006).

#### **6.3 FSL-Gb3**

Recently, additional GSL analogues have been shown to act to inhibit HIV-1 infection *in vitro*. A completely synthetic water soluble analogue of Gb3 termed **F**unctional head **S**pacer **L**ipid tail-Gb3 (FSL-Gb3) was shown to inhibit X4 and R5 HIV-1 infection with a similar 50% inhibitory activity (IC50) as adaGb3 (Harrison et al., 2010). This Gb3 analogue was unique in that the lipid tails were replaced with phosphatidylethanolamine and a spacer region containing multiple ionic residues allowed for complete solubility in aqueous media. A novel synthetic process maintains the carbohydrate moiety of Gb3 coupled to phosphatidylethanolamine through a phosphate linker. This molecule gains its solubility through the insertion of charged nitrogen and phosphate containing groups that are located between the glycone and aglycone moietites. The molecule is completely synthetic, completely soluble in aqueous solutions, and available in large quantities for testing. In addition. animal studies have shown no toxicity at millimolar quantities systemically. The unusual tail also allows for this analogue to insert itself into cell membranes and convert an HIV-permissive Gb3-negative T-cell into a Gb3-positive T-cell that resists HIV infection (Harrison et al., 2010).

Harrison et al. (2010) have used the FSL-Gb3 to show that it can inhibit HIV infection by two different mechanisms. First, as with adaGb3, mixing the FSL-Gb3 with either X4 or R5 HIV-1 results in inhibition of HIV infection with approximately the same IC50 as with adaGb3 (Harrison et al., 2010). This was shown both for laboratory strains of HIV-1 as well as for clinical isolates of R5 HIV-1 viruses. In addition, FSL-Gb3, apparently due to its particular hydrophobic tail structure, was shown to insert itself into cell membranes, retaining proper cell-surface conformation of the carbohydrate moieties. Cellular insertion of FSL-Gb3 was able to result in conversion of a human CD4+ T-cell that completely lacked Gb3 expression into a T-cell that highly expresses Gb3. This property of FSL-Gb3 to convert a permissive HIV target cell into a less permissive cell for HIV infection is a major finding; thus, providing acquired resistance to HIV-1 infection as a possible therapeutic approach. The following: "In addition, preliminary work has shown potential for soluble Gb3 analogues, including FSL-Gb3, to act as microbicides to inhibit mucosal HIV transmission (Harrison et al., 2011). FSL-Gb3 shows great promise as a possible therapeutic, *in vivo*, as it would be potentially capable of inhibiting HIV infection both systemically to reduce viremia but also by its insertion into CD4+ T cells resulting in inhibition of HIV infection by blocking viral entry into its normal primary host targets.

#### **6.4 Multimeric GSLs**

The finding that a soluble GSL analogue was capable of inhibiting HIV infection soon led to other studies where the investigators used a series of C-glycoside synthetic analogues of GalCer (Garg et al., 2008). These investigators showed that two of six analogues were able to bind gp120 and inhibit X4 and R5 strains and dual-tropic HIV-mediated fusion and entry in the absence of any significant cytotoxicity. In addition to HIV, these investigators also showed that soluble GSLs may inhibit additional enveloped viruses such as vesticular stomatitis virus (VSV) (Garg et al., 2008).

In collaboration with Blumenthal's group at NIH we have found that adaGb3 is also able to inhibit gp120/CD4/chemokine coreceptor dependent host cell fusion irrespective of gp120 type (R5 or X4 tropic, HIV-1 or HIV-2) as monitored in an indicator system in which gp120 is transfected into one indicator cell and the chemokine receptor into another (Lund et al., 2006).

Recently, additional GSL analogues have been shown to act to inhibit HIV-1 infection *in vitro*. A completely synthetic water soluble analogue of Gb3 termed **F**unctional head **S**pacer **L**ipid tail-Gb3 (FSL-Gb3) was shown to inhibit X4 and R5 HIV-1 infection with a similar 50% inhibitory activity (IC50) as adaGb3 (Harrison et al., 2010). This Gb3 analogue was unique in that the lipid tails were replaced with phosphatidylethanolamine and a spacer region containing multiple ionic residues allowed for complete solubility in aqueous media. A novel synthetic process maintains the carbohydrate moiety of Gb3 coupled to phosphatidylethanolamine through a phosphate linker. This molecule gains its solubility through the insertion of charged nitrogen and phosphate containing groups that are located between the glycone and aglycone moietites. The molecule is completely synthetic, completely soluble in aqueous solutions, and available in large quantities for testing. In addition. animal studies have shown no toxicity at millimolar quantities systemically. The unusual tail also allows for this analogue to insert itself into cell membranes and convert an HIV-permissive Gb3-negative T-cell into a Gb3-positive T-cell that resists HIV infection (Harrison et al., 2010). Harrison et al. (2010) have used the FSL-Gb3 to show that it can inhibit HIV infection by two different mechanisms. First, as with adaGb3, mixing the FSL-Gb3 with either X4 or R5 HIV-1 results in inhibition of HIV infection with approximately the same IC50 as with adaGb3 (Harrison et al., 2010). This was shown both for laboratory strains of HIV-1 as well as for clinical isolates of R5 HIV-1 viruses. In addition, FSL-Gb3, apparently due to its particular hydrophobic tail structure, was shown to insert itself into cell membranes, retaining proper cell-surface conformation of the carbohydrate moieties. Cellular insertion of FSL-Gb3 was able to result in conversion of a human CD4+ T-cell that completely lacked Gb3 expression into a T-cell that highly expresses Gb3. This property of FSL-Gb3 to convert a permissive HIV target cell into a less permissive cell for HIV infection is a major finding; thus, providing acquired resistance to HIV-1 infection as a possible therapeutic approach. The following: "In addition, preliminary work has shown potential for soluble Gb3 analogues, including FSL-Gb3, to act as microbicides to inhibit mucosal HIV transmission (Harrison et al., 2011). FSL-Gb3 shows great promise as a possible therapeutic, *in vivo*, as it would be potentially capable of inhibiting HIV infection both systemically to reduce viremia but also by its insertion into CD4+ T cells resulting in inhibition of HIV infection by blocking viral

The finding that a soluble GSL analogue was capable of inhibiting HIV infection soon led to other studies where the investigators used a series of C-glycoside synthetic analogues of GalCer (Garg et al., 2008). These investigators showed that two of six analogues were able to bind gp120 and inhibit X4 and R5 strains and dual-tropic HIV-mediated fusion and entry in the absence of any significant cytotoxicity. In addition to HIV, these investigators also showed that soluble GSLs may inhibit additional enveloped viruses such as vesticular

**6.3 FSL-Gb3**

entry into its normal primary host targets.

stomatitis virus (VSV) (Garg et al., 2008).

**6.4 Multimeric GSLs** 

Rosa Borges et al. (2010) found that if the synthesized analogues utilized multimeric Gb3 sites, increased efficacy for inhibition of HIV-1 could be achieved. These investigators covalently attached multiple head groups of either Gb3 or GM3 to a dendrimer core to produce multivalent dendrimeric compounds that were water soluble and showed a much lower IC50 for inhibition of HIV-1 infection, compared to either FSL-Gb3 or adaGb3 (Rosa Borges et al., 2010). Thus, soluble Gb3 analogues have important therapeutic potential to block HIV from interacting with CD4+ target cells (Figure 5D).

## **7. Pharmacologic modulation of Gb3 expression**

1-Deoxygalactonojirimycin (DGJ) is an alkylated imino sugar with a galactose head and a onecarbon side chain. It was found to be highly tolerable *in vivo* and established as a potent competitive inhibitor of *α-galactosidase A* (α-Gal A)(Hamanaka et al., 2008). DGJ has been proposed as a specific chemical chaperone for treatment of diseases including Fabry (Fan et al., 1999). Studies have indicated that oral administration of DGJ to transgenic mice expressing a human mutant α-Gal A substantially elevated enzyme activity in major organs (Fan et al., 1999). Because Gb3 has now been suggested by the studies of Lund et al. (Lund et al., 2006). as an important component for prevention of the HIV entry mechanism, the use of DGJ to pharmacologically increase Gb3 expression may be useful for HIV prevention strategies.

Recent studies (Ramkumar et al., 2009) used DGJ to increase the cell-surface expression of Gb3 on the monocyte cell line, THP-1, which is infectable with X4 HIV-1. DGJ used at 100 μM was able to increase the cell-surface expression of Gb3 of THP-1 cells by approximately 20-fold. Subsequent X4 HIV-1 infection was decreased significantly.

To inhibit Gb3 expression in these cells, these investigators used the compound, D-threo-1 phenyl-2-palmitoylamino-3-pyrrolidino-1-propanol (P4) which inhibits GlcCer synthase, the enzyme responsible for producing GlcCer (Inokuchi & Radin, 1987). This is a key enzyme in the biosynthesis of GSLs because most are glucosylceramide-based. P4 is the most potent inhibitor of this glycosyltransferase (Lee et al., 1999). Using 2 μM P4, Ramkumar et al (Ramkumar et al., 2009) were able to completely inhibit Gb3 expression with subsequent X4 HIV-1 infection increased up to 20-fold. DGJ had little effect on the infection of a Gb3 negative subclone of THP-1 cells and FACS analysis indicated that after DGJ treatment, CD4 and HIV co-receptor levels were similar in the Gb3 expressing and non-expressing THP-1 cell lines by these two compounds. Also, as DGJ was found to be non-toxic in the THP-1 cells up to concentrations of 300 µM, the inhibitory effect was not a result of cytotoxicity of DGJ treatment. The authors concluded that their results indicated that the difference in HIV infection was due solely to the modulation of the expression of the levels of Gb3.

To examine the effects of DGJ and P4 treatment on subsequent infection with R5 HIV-1 virus, the glioblastoma cell line, U87, that had been transfected to express CD4 and the chemokine co-receptor, CCR5, was used. Ramkumar et al. (Ramkumar et al., 2009) again found that treatment of these cells with DGJ resulted in a significant inhibition of R5 HIV-1 infection while treatment with P4 caused a doubling in the infection. They concluded that pharmacologically increasing Gb3 expression using DGJ treatment or inhibition of Gb3 expression using P4 demonstrates a linear relationship between Gb3 expression and infection with either X4 or R5 HIV-1. In addition, their studies suggest that pharmacologically increasing Gb3 is an effective and novel means to prevent HIV-1 infection *in vitro* and that this approach should be explored for *in vivo* treatment of HIV infection.

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 261

and appears transient, these infected cells could serve as reservoirs of latent HIV provirus and may become activated under certain conditions to produce a round of progeny virions which would have the potential to infect other cells such as CD4+ T-cells and maintain or re-

Several hypotheses have arisen to explain the infection of CD4-negative cells within the current paradigm of only CD4 and chemokine co-receptors playing a role. A popular theory is that the availability of CXCR4 in CD4-negative cells is sufficient for viral fusion and entry. However, the evidence addressing this idea is contradictory. In support, human CD4-

CXCR4+ pre-T cell lines can be infected with HIV-1 (Borsetti et al., 2000). Furthermore,

transfected to over-express CXCR4 (Baiocchi et al., 1997). In contradiction, the CD4-negative human B-cell line Raji is not permissive to HIV-1 infection, even though it expresses functional CXCR4 (Speck et al., 1999). Therefore, the absolute dependence of HIV-1 on CXCR4, even in

Fig. 6. **Inhibition of HIV infection of epithelial cells by soluble Gb3**. Caco-2 CD4-negative epithelial cells are infected by HIV but the infection can be inhibited using soluble adaGb3.

**3**

Taken together, the current paradigm that requires the availability of both a primary CD4 receptor plus a co-receptor, either CXCR4 or CCR5, in order for HIV to infect a target cell is not sufficient to explain other cell infections were either the primary receptor and/or the coreceptor are not present, or where there is a lack of infection when both receptors are present. Indeed, there is ample evidence that these receptors are not always sufficient for viral infection. Further examples include human CD4-negative astrocytes that express functional CCR5 and CXCR4 and are resistant to infection by HIV-1 strains (Boutet et al., 2001) and CD4+CXCR4+ cells, also resistant to infection with HIV-1 (Moriuchi et al., 1997). This was shown by infecting U937 monocyte-derived cell lines that were shown to be either permissive or nonpermissive for infection by HIV-1. All but one of these cell lines expressed

adaNH2 is a control soluble GSL for adaGb3

the absence of CD4, does not completely account for the ability of HIV-1 to infect CD4-

human megakaryocytic cells are fully resistant to HIV-1 infection until they are

CCR5-

cells.

establish an active infection.

CD4+CXCR4-

## **8. HIV infection of CD4 negative cells**

The current paradigm indicates that infection with HIV-1 depends entirely on the recognition of its primary and co-receptors for viral fusion and entry into a target cell. Unfortunately, this paradigm is insufficient to completely explain the pathogenesis of HIV-1. This is because there are many instances of HIV-1 infection where either the primary and/or co-receptors are missing from the infected cell. Indeed, HIV-1-infected CD4 negative cells have been identified *in vivo*, including various brain cells (Pumarola-Sune et al., 1987, Ward et al., 1987, Wiley et al., 1986)epithelial cells (Nelson et al., 1988), cardiomyocytes (Barbaro et al., 1998), CD4 negative lymphocytes (Livingstone et al., 1996, Saha et al., 2001a), renal tubular epithelial cells(Marras et al., 2002, Wyatt & Klotman, 2007), hepatocytes (Fromentin et al., 2011) and thymocytes (Kitchen et al., 1997). HIV-1 has also been shown to infect CD4-negative neural and epithelial cells *in vitro*, although not productively (Clapham et al., 1989, Tateno et al., 1989). However, it has been shown that HIV-1 can productively infect CD4-CD8+ T lymphocytes *in vitro* (Saha et al., 2001b).

Our own work supports the idea of HIV infection of CD4-negative cells. Using kidneyderived cell lines such as ACHN and 293 as well as a colon-derived cancer cell line called Caco-2, we have been able to show transient infection with an X4 virus (Figure 5). We have also shown that soluble Gb3 can inhibit the infection of CD4-negative Caco-2 epithelial cell lines (Figure 6) as well as human CD4-negative cell lines derived from the cervix or endometrium (Harrison et al., 2011). Although, the infection of these cell lines is not robust

Fig. 5. **HIV can infect CD4-negative epithelial cells.** HIV infection of ACHN kidneyderived cell line and Caco-2 colon cancer derived epithelial cell line. Trypsin is used to insure that the p24 antigen being used as a measure of productive HIV infection is not derived from external virions sticking to the cell membranes but from budding virions indicating a round of replication of the virus.

The current paradigm indicates that infection with HIV-1 depends entirely on the recognition of its primary and co-receptors for viral fusion and entry into a target cell. Unfortunately, this paradigm is insufficient to completely explain the pathogenesis of HIV-1. This is because there are many instances of HIV-1 infection where either the primary and/or co-receptors are missing from the infected cell. Indeed, HIV-1-infected CD4

al., 1987, Ward et al., 1987, Wiley et al., 1986)epithelial cells (Nelson et al., 1988),

Saha et al., 2001a), renal tubular epithelial cells(Marras et al., 2002, Wyatt & Klotman, 2007), hepatocytes (Fromentin et al., 2011) and thymocytes (Kitchen et al., 1997). HIV-1 has also been shown to infect CD4-negative neural and epithelial cells *in vitro*, although not productively (Clapham et al., 1989, Tateno et al., 1989). However, it has been shown that

Our own work supports the idea of HIV infection of CD4-negative cells. Using kidneyderived cell lines such as ACHN and 293 as well as a colon-derived cancer cell line called Caco-2, we have been able to show transient infection with an X4 virus (Figure 5). We have also shown that soluble Gb3 can inhibit the infection of CD4-negative Caco-2 epithelial cell lines (Figure 6) as well as human CD4-negative cell lines derived from the cervix or endometrium (Harrison et al., 2011). Although, the infection of these cell lines is not robust

HIV-1 can productively infect CD4-CD8+ T lymphocytes *in vitro* (Saha et al., 2001b).

Fig. 5. **HIV can infect CD4-negative epithelial cells.** HIV infection of ACHN kidneyderived cell line and Caco-2 colon cancer derived epithelial cell line. Trypsin is used to insure that the p24 antigen being used as a measure of productive HIV infection is not derived from external virions sticking to the cell membranes but from budding virions

indicating a round of replication of the virus.

cells have been identified *in vivo*, including various brain cells (Pumarola-Sune et

lymphocytes (Livingstone et al., 1996,

**8. HIV infection of CD4 negative cells** 

cardiomyocytes (Barbaro et al., 1998), CD4 negative-

negative-

and appears transient, these infected cells could serve as reservoirs of latent HIV provirus and may become activated under certain conditions to produce a round of progeny virions which would have the potential to infect other cells such as CD4+ T-cells and maintain or reestablish an active infection.

Several hypotheses have arisen to explain the infection of CD4-negative cells within the current paradigm of only CD4 and chemokine co-receptors playing a role. A popular theory is that the availability of CXCR4 in CD4-negative cells is sufficient for viral fusion and entry. However, the evidence addressing this idea is contradictory. In support, human CD4- CCR5- CXCR4+ pre-T cell lines can be infected with HIV-1 (Borsetti et al., 2000). Furthermore, CD4+CXCR4 human megakaryocytic cells are fully resistant to HIV-1 infection until they are transfected to over-express CXCR4 (Baiocchi et al., 1997). In contradiction, the CD4-negative human B-cell line Raji is not permissive to HIV-1 infection, even though it expresses functional CXCR4 (Speck et al., 1999). Therefore, the absolute dependence of HIV-1 on CXCR4, even in the absence of CD4, does not completely account for the ability of HIV-1 to infect CD4 cells.

Fig. 6. **Inhibition of HIV infection of epithelial cells by soluble Gb3**. Caco-2 CD4-negative epithelial cells are infected by HIV but the infection can be inhibited using soluble adaGb3. adaNH2 is a control soluble GSL for adaGb3

Taken together, the current paradigm that requires the availability of both a primary CD4 receptor plus a co-receptor, either CXCR4 or CCR5, in order for HIV to infect a target cell is not sufficient to explain other cell infections were either the primary receptor and/or the coreceptor are not present, or where there is a lack of infection when both receptors are present. Indeed, there is ample evidence that these receptors are not always sufficient for viral infection. Further examples include human CD4-negative astrocytes that express functional CCR5 and CXCR4 and are resistant to infection by HIV-1 strains (Boutet et al., 2001) and CD4+CXCR4+ cells, also resistant to infection with HIV-1 (Moriuchi et al., 1997). This was shown by infecting U937 monocyte-derived cell lines that were shown to be either permissive or nonpermissive for infection by HIV-1. All but one of these cell lines expressed

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 263

Ahn, A., D. L. Gibbons & M. Kielian (2002) The fusion peptide of Semliki Forest virus

Akashi, M., F. Takaku, H. Nojiri, Y. Miura, Y. Nagai & M. Saito (1988) Neutral and sialosyl

Alfsen, A. & M. Bomsel (2002) HIV-1 gp41 envelope residues 650-685 exposed on native

Alkhatib, G., C. Combadiere, C. C. Broder, Y. Feng, P. E. Kennedy, P. M. Murphy & E. A.

Aoki, C., K. I. Hidari, S. Itonori, A. Yamada, N. Takahashi, T. Kasama, F. Hasebe, M. A.

recognized by dengue virus type 2. *Journal of Biochemistry.* 139, (pp. 607-14). Arab, S. & C. Lingwood (1998) Intracellular targeting of the endoplasmic reticulum/nuclear

Baiocchi, M., E. Olivetta, C. Chelucci, A. C. Santarcangelo, R. Bona, P. d'Aloja, U. Testa, N.

Barbaro, G., G. Di Lorenzo, B. Grisorio & G. Barbarini (1998) Cardiac involvement in the

Bavari, S., C. M. Bosio, E. Wiegand, G. Ruthel, A. B. Will, T. W. Geisbert, M. Hevey, C.

Bhat, S., R. V. Mettus, E. P. Reddy, K. E. Ugen, V. Srikanthan, W. V. Williams & D. B. Weiner

maps to amino acids 206-275. *AIDS Res Hum Retroviruses.* 9, (pp. 175-81). Bhat, S., S. L. Spitalinik, F. Gonzalez-Scarano & D. H. Silberberg (1991) Galactosyl ceramide

cofactor for macrophage-tropic HIV-1. *Science.* 272, (pp. 1955-8).

surface bilayer film and liposomes. *Glycoconj. J.* 13, (pp. 159-166).

Investigators. *AIDS Res Hum Retroviruses.* 14, (pp. 1071-7).

*Journal of Experimental Medicine.* 195, (pp. 593-602).

HIV-1 infection. *Blood.* 89, (pp. 2670-8).

88, (pp. 7131-7134).

associates with sterol-rich membrane domains. *Journal of Virology.* 76, (pp. 3267-75).

glycosphingolipid composition and metabolism of human T-lymphoblastic cell line MOLT-3 cells: distinctive changes as markers specific for their differentiation.

virus act as a lectin to bind epithelial cell galactosyl ceramide. *J Biol Chem.* 277, (pp.

Berger (1996) CC CKR5: a RANTES, MIP-1alpha, MIP-1beta receptor as a fusion

Islam, K. Hatano, K. Matsuoka, T. Taki, C. T. Guo, T. Takahashi, Y. Sakano, T. Suzuki, D. Miyamoto, M. Sugita, D. Terunuma, K. Morita & Y. Suzuki (2006) Identification and characterization of carbohydrate molecules in mammalian cells

envelope by retrograde transport may determine cell hypersensitivity to Verotoxin: sodium butyrate or selection of drug resistance may induce nuclear toxin targeting via globotriaosyl ceramide fatty acid isoform traffic. *J Cell Physiol.* 177, (pp. 646-660). Arab, S. & C. A. Lingwood (1996) Influence of phospholipid chain length on

verotoxin/globotriaosyl ceramide binding in model membranes: comparison of a

Komatsu, P. Verani & M. Federico (1997) Human immunodeficiency virus (HIV) resistant CD4+ UT-7 megakaryocytic human cell line becomes highly HIV-1 and HIV-2 susceptible upon CXCR4 transfection: induction of cell differentiation by

acquired immunodeficiency syndrome: a multicenter clinical-pathological study. Gruppo Italiano per lo Studio Cardiologico dei pazienti affetti da AIDS

Schmaljohn, A. Schmaljohn & M. J. Aman (2002) Lipid raft microdomains: a gateway for compartmentalized trafficking of Ebola and Marburg viruses. *The* 

(1993) The galactosyl ceramide/sulfatide receptor binding region of HIV-1 gp120

or a derivative is an essential component of the neural receptor for human immunodeficiency virus type 1 envolope glycoprotein gp 120. *Proc. Natl. Acad. Sci.* 

**10. References** 

*Blood.* 72, (pp. 469-79).

25649-59.).

both functional CXCR4 and CD4. One of these cell lines that was nonpermissive lacked CXCR4, but when this receptor was transfected back into this cell line, it remained nonpermissive to infection 1 (Moriuchi et al., 1997).

## **9. Summary**

Studies have indicated that human PBMCs with an intracellular or cell-surface accumulation of Gb3 were less susceptible to HIV infection (Lund et al., 2005, Lund et al., 2009).These PBMCs were derived from patients with Fabry disease and from healthy P1k blood group phenotype individuals having a pathologic or natural, respectively, elevation of Gb3. AdaGb3, FSL-Gb3, and multivalent dendrimeric-Gb3, all soluble Gb3 analogues, have been shown to be effective inhibitors of HIV regardless of strain or tropism, and also to inhibit drug resistant HIV strains and prevent HIV infection of CD4-negative epithelial cells. Therefore, Gb3 may be a natural host resistance factor and increasing its expression *in vitro* using soluble analogues, such as FSL-Gb3, that can insert into T-cells that do not naturally express Gb3, and/or the use of a pharmacologic agent, such as DGJ, to increase Gb3 expression, may decrease HIV-1 susceptibility. Importantly, the further development of soluble Gb3 analogues, especially multivalent analogues expressing multiple Gb3 sugar moieties having increased affinity and avidity for the V3 loop of HIV gp120, may provide for novel and highly effective HIV therapeutics to prevent or treat HIV/AIDS (Figure 7).

Fig. 7. **Potential novel HIV therapeutic.** Soluble Gb3 analogues may be able to bind to HIV gp120 protruding from the HIV envelop and prevent HIV from interacting with the primary and/or co-receptors for HIV; thus, preventing HIV infection.

### **10. References**

262 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

both functional CXCR4 and CD4. One of these cell lines that was nonpermissive lacked CXCR4, but when this receptor was transfected back into this cell line, it remained

Studies have indicated that human PBMCs with an intracellular or cell-surface accumulation of Gb3 were less susceptible to HIV infection (Lund et al., 2005, Lund et al., 2009).These PBMCs were derived from patients with Fabry disease and from healthy P1k blood group phenotype individuals having a pathologic or natural, respectively, elevation of Gb3. AdaGb3, FSL-Gb3, and multivalent dendrimeric-Gb3, all soluble Gb3 analogues, have been shown to be effective inhibitors of HIV regardless of strain or tropism, and also to inhibit drug resistant HIV strains and prevent HIV infection of CD4-negative epithelial cells. Therefore, Gb3 may be a natural host resistance factor and increasing its expression *in vitro* using soluble analogues, such as FSL-Gb3, that can insert into T-cells that do not naturally express Gb3, and/or the use of a pharmacologic agent, such as DGJ, to increase Gb3 expression, may decrease HIV-1 susceptibility. Importantly, the further development of soluble Gb3 analogues, especially multivalent analogues expressing multiple Gb3 sugar moieties having increased affinity and avidity for the V3 loop of HIV gp120, may provide for novel and highly effective HIV therapeutics to prevent or treat HIV/AIDS (Figure 7).

Fig. 7. **Potential novel HIV therapeutic.** Soluble Gb3 analogues may be able to bind to HIV gp120 protruding from the HIV envelop and prevent HIV from interacting with the primary

and/or co-receptors for HIV; thus, preventing HIV infection.

nonpermissive to infection 1 (Moriuchi et al., 1997).

**9. Summary**


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 265

Chark, D., A. Nutikka, N. Trusevych, J. Kuzmina & C. Lingwood (2004) Differential

Clapham, P. R., J. N. Weber, D. Whitby, K. McIntosh, A. G. Dalgleish, P. J. Maddon, K. C.

Comolli, J. C., A. R. Hauser, L. Waite, C. B. Whitchurch, J. S. Mattick & J. N. Engel (1999)

Dalgleish, A. G., P. C. Beverley, P. R. Clapham, D. H. Crawford, M. F. Greaves & R. A. Weiss

De Haan, L. & T. R. Hirst (2004) Cholera toxin: a paradigm for multi-functional engagement

De Rosa, M. F., D. Sillence, C. Ackerley & C. Lingwood (2004) Role of Multiple Drug

Del Real, G., S. Jimenez-Baranda, R. A. Lacalle, E. Mira, P. Lucas, C. Gomez-Mouton, A. C.

Delezay, O., D. Hammache, J. Fantini & N. Yahi (1996) SPC3, a V3 loop-derived synthetic

Devenica, D., V. Cikes Culic, A. Vuica & A. Markotic (2010) Biochemical, pathological and

Dietrich, C., L. A. Bagatolli, Z. N. Volovyk, N. L. Thompson, M. Levi, K. Jacobson & E.

Dorosko, S. M. & R. I. Connor (2010) Primary human mammary epithelial cells endocytose

Dykstra, M. L., A. Cherukuri & S. K. Pierce (2001) Floating the raft hypothesis for immune

Eggens, I., B. Fenderson, T. Toyokuni, B. Dean, M. Stroud & S.-I. Hamomori (1989) Specific interaction between Lex and Lex determinants. *J Biol Chem.* 264, (pp. 9476-9484). Emam, A., W. G. Carter & L. C.A (2010) Glycolipid-dependent, protease sensitive

of cellular mechanisms (Review). *Mol Membr Biol.* 21, (pp. 77-92).

CD4 to nonraft membrane domains. *J Exp Med.* 196, (pp. 293-301.).

oncological relevance of Gb3Cer receptor. *Med Oncol.* (pp.

(pp. 1-13).

3625-30).

1417-1428).

381, (pp. 667-73).

160-6.).

muscle cells. *Nature.* 337, (pp. 368-70.).

retrovirus. *Nature.* 312, (pp. 763-7). Daniels, G. (2002). Blackwell Science, Malden, MA

*Chem.* 279, (pp. 7867-7876).

*Biochemistry.* 35, (pp. 15663-71).

cells *Open Microbiol J.* 4, (pp. 106-15).

carbohydrate epitope recognition of globotriaosyl ceramide by verotoxins and monoclonal antibody: Role in human renal glomerular binding. *Eur. J. Biochem.* 271,

Deen, R. W. Sweet & R. A. Weiss (1989) Soluble CD4 blocks the infectivity of diverse strains of HIV and SIV for T cells and monocytes but not for brain and

Pseudomonas aeruginosa gene products PilT and PilU are required for cytotoxicity *in vitro* and virulence in a mouse model of acute pneumonia. *Infect Immun.* 67, (pp.

(1984) The CD4 (T4) antigen is an essential component of the receptor for the AIDS

Resistance Protein 1 in neutral but not acidic glycosphingolipid biosynthesis. *J. Biol.* 

Carrera, A. C. Martinez & S. Manes (2002) Blocking of HIV-1 infection by targeting

peptide inhibitor of HIV-1 infection, binds to cell surface glycosphingolipids.

Gratton (2001) Lipid rafts reconstituted in model membranes. *Biophys J.* 80, (pp.

HIV-1 and facilitate viral infection of CD4+ T lymphocytes. *J Virol.* 84, (pp. 10533- 42).Dragic, T., V. Litwin, G. P. Allaway, S. R. Martin, Y. Huang, K. A. Nagashima, C. Cayanan, P. J. Maddon, R. A. Koup, J. P. Moore & W. A. Paxton (1996) HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. *Nature.* 

receptors: access to rafts controls receptor signaling and trafficking. *Traffic.* 2, (pp.

internalization of *Pseudomonas aeruginosa* into cultured human respiratory epithelial


Bitzan, M., S. Richardson, C. Huang, B. Boyd, M. Petric & M. Karmali (1994) Evidence that

Bomsel, M. (1997) Transcytosis of infectious human immunodeficiency virus across a tight

Borén, T., P. Falk, K. A. Roth, G. Larson & S. Normark (1993) Attachment of *Helicobacter* 

Borsetti, A., C. Parolin, B. Ridolfi, L. Sernicola, A. Geraci, B. Ensoli & F. Titti (2000) CD4-

Boulanger, J., D. Faulds, E. M. Eddy & C. A. Lingwood (1995) Members of the 70kDa heat

Boyd, B. & C. A. Lingwood (1989) Verotoxin receptor glycolipid in human renal tissue.

Boyd, B., Z. Zhiuyan, G. Magnusson & C. A. Lingwood (1994) Lipid modulation of glycolipid

Branch, D. R. (2010) Blood groups and susceptibility to virus infection: new developments.

Branch, D. R., L. J. Valenta, S. Yousefi, D. Sakac, R. Singla, M. Bali, B. M. Sahai & X. Z. Ma

Bresnahan, P. A., W. Yonemoto, S. Ferrell, D. Williams-Herman, R. Geleziunas & W. C.

Brown, D. & E. London (2000) Structure and function of sphingolipid- and cholesterol-rich

Brown, D. A. & E. London (1997) Structure of detergent-resistant membrane domains: does

Brown, K. E., J. R. Hibbs, G. Gallinella, S. M. Anderson, E. D. Lehman, P. McCarthy & N. S.

actively facilitates productive HIV-1 infection. *AIDS.* 16, (pp. 309-19). Bremer, E. G. (1994). Glycosphingolipids as Effectors of Growth and Differentiation

membrane rafts. *J. Biol. Chem.* 275, (pp. 17221-17224).

of human erythrocytes *in vitro*. *Infect Immun.* 62, (pp. 3337-3347).

and simian immunodeficiency viruses. *J Virol.* 74, (pp. 6689-94).

human epithelial cell line barrier. *Nat Med.* 3, (pp. 42-7.).

(pp. 1892-1895).

CXCR4. *Glia.* 34, (pp. 165-77).

*Nephron.* 51, (pp. 207-210).

*Chem.* 13, (pp. 565-77).

(pp. 1-7).

(pp. 2641-6).

*Curr Opin Hematol.* 17, (pp. 558-64).

Verotoxins (Shiga-like toxins) from *Escherichia coli* bind to P blood group antigens

*pylori* to human gastric epithelium mediated by blood group antigens. *Science.* 262,

independent infection of two CD4(-)/CCR5(-)/CXCR4(+) pre-T-cell lines by human

shock protein family specifically recognize sulfoglycolipids: role in gamete recognition and mycoplasma related infertility. *J. Cell. Physiol.* 165, (pp. 7-17). Boutet, A., H. Salim, Y. Taoufik, P. M. Lledo, J. D. Vincent, J. F. Delfraissy & M. Tardieu

(2001) Isolated human astrocytes are not susceptible to infection by M- and T-tropic HIV-1 strains despite functional expression of the chemokine receptors CCR5 and

receptor function: Presentation of galactose a1-4 galactose disaccharide for Verotoxin binding in natural and synthetic glycolipids. *Eur. J. Biochem.* 223, (pp. 873-878). Brady, R. O. (1967) Enzymatic abnormalities in diseases of sphingolipid metabolism. *Clin* 

(2002) VPAC1 is a cellular neuroendocrine receptor expressed on T cells that

Greene (1998) A dileucine motif in HIV-1 Nef acts as an internalization signal for CD4 downregulation and binds the AP-1 clathrin adaptor. *Curr Biol.* 8, (pp. 1235-8).

phase separation occur in biological membranes? *Biochem Biophys Res Commun.* 240,

Young (1994) Resistance to parvovirus B19 infection due to lack of virus receptor (erythrocyte P antigen). *The New England Journal of Medicine.* 330, (pp. 1192-6). Brugger, B., B. Glass, P. Haberkant, I. Leibrecht, F. T. Wieland & H. G. Krausslich (2006) The

HIV lipidome: a raft with an unusual composition. *Proc Natl Acad Sci U S A.* 103,


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 267

Furukawa, K., K. Iwamura, M. Uchikawa, B. N. Sojka, J. Wiels, T. Okajima & T. Urano (2000)

Futerman, A. H. & R. E. Pagano (1991) Determination of the intracellular sites and topology of glucosylceramide synthesis in rat liver. *Biochem J.* 280 ( Pt 2), (pp. 295-302). Garg, H., N. Francella, K. A. Tony, L. A. Augustine, J. J. Barchi, Jr., J. Fantini, A. Puri, D. R.

Gatfield, J. & J. Pieters (2000) Essential role for cholesterol in entry of mycobacteria into

Ghislain, J., C. Lingwood, M. Maloney, L. Penn & E. Fish (1992) Association between the ifn

Gilbert, J. & T. Benjamin (2004) Uptake pathway of polyomavirus via ganglioside GD1a.

Gomez-Mouton, C., J. L. Abad, E. Mira, R. A. Lacalle, E. Gallardo, S. Jimenez-Baranda, I. Illa,

Gottlieb, M. S., R. Schroff, H. M. Schanker, J. D. Weisman, P. T. Fan, R. A. Wolf & A. Saxon

Gummuluru, S., M. Rogel, L. Stamatatos & M. Emerman (2003) Binding of Human

Haataja, S., K. Tikkanen, U. Nilsson, G. Magnusson, K. Karlsson & J. Finne (1994)

Hakomori, S. (1985) Aberrant glycosylation in cancer cell membranes as focused on glycolipids: overview and perspectives. *Cancer Res.* 45, (pp. 2405-14).

Hakomori, S. (1993) Structure and function of sphingoglycolipids in transmembrane signalling and cell-cell interactions. *Biochem Soc Trans.* 21 ( Pt 3), (pp. 583-95). Hakomori, S. (1996) Tumor malignancy defined by aberrant glycosylation and

Hakomori, S. (2002) Glycosylation defining cancer malignancy: new wine in an old bottle.

Hakomori, S. & Y. Igarashi (1993) Gangliosides and glycosphingolipids as modulators of cell growth, adhesion, and transmembrane signaling. *Adv Lipid Res.* 25, (pp. 147-62). Hakomori, S. & Y. Igarashi (1995) Functional role of glycosphingolipids in cell recognition

Cholesterol-Dependent Pathway. *J Virol.* 77, (pp. 12865-74).

adhesin specificities. *J Biol Chem.* 269, (pp. 27466-27472).

Hakomori, S. (1986) Glycosphingolipids. *Scientific American.* 254, (pp. 40-53).

*Proc Natl Acad Sci U S A.* 99, (pp. 10231-3).

and signaling. *J Biochem (Tokyo).* 118, (pp. 1091-1103).

sphingo(glyco)lipid metabolism. *Cancer Res.* 56, (pp. 5309-5318).

individuals. *J Biol Chem.* 275, (pp. 37752-6).

macrophages. *Science.* 288, (pp. 1647-50).

*Journal of Virology.* 78, (pp. 12259-67).

*Acad Sci U S A.* 98, (pp. 9642-7.).

*N Engl J Med.* 305, (pp. 1425-31).

80, (pp. 54-61).

12, (pp. s114).

Molecular basis for the p phenotype. Identification of distinct and multiple mutations in the alpha 1,4-galactosyltransferase gene in Swedish and Japanese

Mootoo & R. Blumenthal (2008) Glycoside analogs of beta-galactosylceramide, a novel class of small molecule antiviral agents that inhibit HIV-1 entry. *Antiviral Res.* 

alpha receptor and the membrane glycolipid globotriaosyl ceramide. *J Interfer Res.* 

A. Bernad, S. Manes & A. C. Martinez (2001) Segregation of leading-edge and uropod components into specific lipid rafts during T cell polarization. *Proc Natl* 

(1981) Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: evidence of a new acquired cellular immunodeficiency.

Immunodeficiency Virus Type 1 to Immature Dendritic Cells Can Occur Independently of DC-SIGN and Mannose Binding C-Type Lectin Receptors via a

Oligosaccharide-receptor interaction of the Gal alpha 1-4Gal binding adhesin of *Streptococcus suis*. Combining site architecture and characterization of two variant


Emam, A., A. Yu, H.-J. Park, R. Mahfoud, J. Kus, L. Burrows & C. Lingwood (2006)

Epand, R. M., S. Nir, M. Parolin & T. D. Flanagan (1995) The role of the ganglioside GD1a as

Falguieres, T., F. Mallard, C. Baron, D. Hanau, C. Lingwood, B. Goud, J. Salamero & L.

Fantini, J. (2000) Synthetic soluble analogs of glycolipids for studies of virus-glycolipid

Fantini, J. (2003) How sphingolipids bind and shape proteins: molecular basis of lipid-

Fantini, J., D. Hammache, O. Delezay, G. Pieroni, C. Tamalet & N. Yahi (1998a) Sulfatide

Fantini, J., D. Hammache, O. Delezay, N. Yahi, C. Andre-Barres, I. Rico-Lattes & A. Lattes

Fantini, J., D. Hammache, G. Pieroni & N. Yahi (2000) Role of glycosphingolipid microdomains in CD4-dependent HIV-1 fusion. *Glycoconj J.* 17, (pp. 199-204). Fantini, J., C. Tamalet, D. Hammache, C. Tourrés, N. Duclos & N. Yahi (1998b) HIV-1-

Faroux-Corlay, B., J. Greiner, R. Terreux, D. Cabrol-Bass, A. M. Aubertin, P. Vierling & J.

Finnegan, C. & R. Blumenthal (2006) Dissecting HIV fusion: identifying novel targets for

Fivaz, M., L. Abrami & F. G. van der Goot (1999) Landing on lipid rafts. *Trends in Cell* 

Freed, E. O., E. L. Delwart, G. L. Buchschacher, Jr. & A. T. Panganiban (1992) A mutation in the

interferes with fusion and infectivity. *Proc Natl Acad Sci U S A.* 89, (pp. 70-4). Fromentin, R., M. R. Tardif & M. J. Tremblay (2011) Inefficient fusion due to a lack of

anti- HIV-1 activity, and gp120 binding. *J Med Chem.* 44, (pp. 2188-203.). Feng, Y., C. C. Broder, P. E. Kennedy & E. A. Berger (1996) HIV-1 entry cofactor: functional

attachment. *Microbiol.* 152, (pp. 2789-2799).

*Nat Med.* 5, (pp. 112-5).

*Mol Life Sci.* 60, (pp. 1027-32).

inhibits HIV-1 entry into CD4-

272, (pp. 7245-7252).

(pp. 872-7).

*Biology.* 9, (pp. 212-3).

a receptor for Sendai virus. *Biochemistry.* 34, (pp. 1084-9).

interactions. *Methods in Enzymology.* 311, (pp. 626-638).

*Syndromes and Human Retrovirology.* 19, (pp. 221-229).

entry inhibitors. *Infect Disord Drug Targets.* 6, (pp. 355-67).

Laboratory and clinical *Pseudomonas aeruginosa* strains do not bind glycosphingolipids *in vitro* or during type IV pili-mediated initial host cell

Johannes (2001) Targeting of Shiga toxin b-subunit to retrograde transport route in association with detergent-resistant membranes. *Mol. Biol. Cell.* 12, (pp. 2453-68.). Fan, J. Q., S. Ishii, N. Asano & Y. Suzuki (1999) Accelerated transport and maturation of

lysosomal alpha-galactosidase A in Fabry lymphoblasts by an enzyme inhibitor.

protein interactions in lipid shells, rafts and related biomembrane domains. *Cell* 

(1997) Synthetic soluble analogs of galactosylceramide (GalCer) bind to the V3 domain of HIV-1 gp 120 and Inhibit HIV-1-induced fusion and entry. *J. Biol. Chem.* 

induced perturbations of glycosphingolipid metabolism are cell-specific and can be detected at early stages of HIV-1 infection. *Journal of Acquired Immune Deficiency* 

Fantini (2001) Amphiphilic anionic analogues of galactosylceramide: synthesis,

cDNA cloning of a seven-transmembrane, G protein-coupled receptor. *Science.* 272,

human immunodeficiency virus type 1 transmembrane glycoprotein gp41 dominantly

attachment receptor/co-receptor restricts productive human immunodeficiency virus type 1 infection in human hepatoma Huh7.5 cells. *J Gen Virol.* 92, (pp. 587-97).

/CXCR4+ cells. *Virology.* 246, (pp. 211-220).


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 269

Hatch, S. C., J. Archer & S. Gummuluru (2009) Glycosphingolipid composition of human

Heath-Engel, H. M. & C. A. Lingwood (2003) Verotoxin sensitivity of ECV304 cells *in vitro* 

Hellberg, A., M. A. Chester & M. L. Olsson (2005) Two previously proposed P1/P2-

Hellberg, A., J. Poole & M. L. Olsson (2002) Molecular basis of the globoside-deficient P(k)

Hellberg, A., A. Ringressi, V. Yahalom, J. Safwenberg, M. E. Reid & M. L. Olsson (2004)

group system and collection. *British Journal of Haematology.* 125, (pp. 528-36). Hellberg, A., R. Steffensen, V. Yahalom, B. N. Sojka, H. E. Heier, C. Levene, J. Poole & M. L.

Heung, L. J., C. Luberto & M. Del Poeta (2006) Role of sphingolipids in microbial

Hikita, T., K. Tadano-Aritomi, N. Iida-Tanaka, S. B. Levery, I. Ishizuka & S. Hakomori (2002)

Hooper, N. (1999) Detergent-insoluble glycosphingolipid/cholesterol-rich membrane domains, lipid rafts and caveolae. *Molr Membr Biol.* 16, (pp. 145-156). Huang, C. C., S. N. Lam, P. Acharya, M. Tang, S. H. Xiang, S. S. Hussan, R. L. Stanfield, J.

Hug, P., H. Lin, T. Korte, X. Xiao, D. Dimitrov, J. Wang, A. Puri & B. R. (2000)

Huwiler, A., D. Fabbro & J. Pfeilschifter (1998) Selective ceramide binding to protein kinase Calpha and -delta isoenzymes in renal mesangial cells. *Biochem.* 37, (pp. 14556-14562). Huwiler, A., T. Kolter, J. Pfeilschifter & K. Sandhoff (2000) Physiology and pathophysiology of sphingolipid metabolism and signaling. *Biochim Biophys Acta.* 1485, (pp. 63-99). Iglesias-Bartolome, R., A. Trenchi, R. Comin, A. L. Moyano, G. A. Nores & J. L. Daniotti

gene. *The Journal of Biological Chemistry.* 277, (pp. 29455-9).

dendritic cell-mediated HIV-1 trans-infection. *J Virol.* 83, (pp. 3496-506). Haywood, A. M. (1994) Virus receptors: binding, adhesion strengthening, and changes in

viral structure. *Journal of virology.* 68, (pp. 1-5).

phenotype. *Transfusion.* 43, (pp. 899-907).

pathogenesis. *Infect Immun.* 74, (pp. 28-39).

significance. *Neurochem Res.* 27, (pp. 575-81).

gp120 and CD4. *Science.* 317, (pp. 1930-4).

74, (pp. 6377-6385).

*Acta.* 1788, (pp. 2526-40).

*Angiogenesis.* 6, (pp. 129-41.).

immunodeficiency virus type 1 (HIV-1) particles is a crucial determinant for

and *in vivo* in a xenograft tumour model: VT1 as a tumour neovascular marker.

differentiating and nine novel polymorphisms at the A4GALT (Pk) locus do not correlate with the presence of the P1 blood group antigen. *BMC Genetics.* 6, (pp. 49).

blood group phenotype. Identification of four inactivating mutations in the UDP-Nacetylgalactosamine: globotriaosylceramide 3-beta-N-acetylgalactosaminyltransferase

Genetic heterogeneity at the glycosyltransferase loci underlying the GLOB blood

Olsson (2003) Additional molecular bases of the clinically important p blood group

Cationic glycosphingolipids in neuronal tissues and their possible biological

Robinson, J. Sodroski, I. A. Wilson, R. Wyatt, C. A. Bewley & P. D. Kwong (2007) Structures of the CCR5 N terminus and of a tyrosine-sulfated antibody with HIV-1

Glycosphingolipids promote entry of a broad range of human immunodeficiency virus type 1 isolates into cell lines expressing CD4, CXCR4,and/or CCR5. *J Virol.* 

(2009) Differential endocytic trafficking of neuropathy-associated antibodies to GM1 ganglioside and cholera toxin in epithelial and neural cells. *Biochim Biophys* 


Hakomori, S., S. Yamamura & A. K. Handa (1998a) Signal transduction through

Hakomori, S.-I. (2000) Cell adhesion/recognition and signal transduction through

Hakomori, S.-I., K. Handa, K. Iwabuchi, S. Yamamura & A. Prinetti (1998b) New insights in

Hakomori, S. I. (2008) Structure and function of glycosphingolipids and sphingolipids: recollections and future trends. *Biochim Biophys Acta.* 1780, (pp. 325-46). Haldar, K., N. Mohandas, B. U. Samuel, T. Harrison, N. L. Hiller, T. Akompong & P.

Hall, A., T. Rog, M. Karttunen & I. Vattulainen (2010) Role of glycolipids in lipid rafts: a

Hamanaka, R., T. Shinohara, S. Yano, M. Nakamura, A. Yasuda, S. Yokoyama, J. Q. Fan, K.

Hammache, D., G. Piéroni, N. Yahi, O. Delézay, N. Koch, H. Lafont, C. Tamalet & J. Fantini

Hammache, D., N. Yahi, M. Maresca, G. Pieroni & J. Fantini (1999) Human erythrocyte

Hammache, D., N. Yahi, G. Pieroni, F. Ariasi, C. Tamalet & J. Fantini (1998b) Sequential

Harouse, J. M., R. G. Collman & F. Gonzalez-Scarano (1995) Human immunodeficiency

Harrison, A., M.L. Olsson, R.B. Jones, S. Ramkumar, D. Sakac, B. Binnington, S. Henry, C.A.

cofactors. *Biochem. Biophys. Res. Commun.* 246, (pp. 117-122).

glycosphingolipid microdomain. *Glycoconjugate J.* 17, (pp. 143-151).

enriched microdomains. *Ann N Y Acad Sci.* 845, (pp. 1-10).

coupled with signaling. *Glycobiol.* 8, (pp. xi-xix).

*Phys Chem B.* 114, (pp. 7797-807).

7967-7971).

*J Virol.* 73, (pp. 5244-5248).

*Virol.* 69, (pp. 7383-90).

malaria parasites. *Cellular Microbiology.* 4, (pp. 383-95).

lysosomes. *Biochim Biophys Acta.* 1782, (pp. 408-13).

glyco(sphingo)lipids. Introduction and recent studies on glyco(sphingo)lipid-

glycosphingolipid function: "glycosignaling domain," a cell surface assembly of glycosphingolipids with signal transducer molecules, involved in cell adhesion

Cheresh (2002) Protein and lipid trafficking induced in erythrocytes infected by

view through atomistic molecular dynamics simulations with galactosylceramide. *J* 

Kawasaki, M. Watanabe & S. Ishii (2008) Rescue of mutant alpha-galactosidase A in the endoplasmic reticulum by 1-deoxygalactonojirimycin leads to trafficking to

(1998a) Specific interaction of HIV-1 and HIV-2 surface envelope glycoproteins with monolayers of galactosylceramide and ganglioside GM3. *J Biol Chem.* 273, (pp.

glycosphingolipids as alternative cofactors for human immunodeficiency virus type 1 (HIV-1) entry: Evidence for CD4-induced interactions between HIV-1 gp120 and reconstituted membrane microdomains of glycosphingolipids (Gb3 and GM3).

interaction of CD4 and HIV-1 gp120 with a reconstituted membrane patch of ganglioside GM3: Implications for the role of glycolipids as potential HIV-1 fusion

virus type 1 infection of SK-N-MC cells: domains of gp120 involved in entry into a CD4-negative, galactosyl ceramide/3' sulfo-galactosyl ceramide-positive cell line. *J* 

Lingwood & D.R. Branch (2010) A synthetic globotriaosylceramide analogue inhibits HIV-1 infection *in vitro* by two mechanisms. *Glycoconj. J.* 27, (pp. 515-524). Harrison, A., S. Henry, R. Mahfoud, A. Manis, A. Albertini, Y. Gaudin, C. A. Lingwood &

D.R. Branch (2011) A novel VSV/HIV pseudotype approach for the study of HIV microbicides without requirement for level 3 biocontainment. *Future Virol. In press*


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 271

Khine, A. A. & C. A. Lingwood (2000) Functional significance of globotriaosylceramide in a2

Kiarash, A., B. Boyd & C. A. Lingwood (1994) Glycosphingolipid receptor function is

Kim, M., D. Sakac, B. Binnington, K. Fernandes, S. Shi, LingwoodC. & D. Branch (2011)

Kitchen, S. G., C. H. Uittenbogaart & J. A. Zack (1997) Mechanism of human

Kojima, N. & S. Hakomori (1991) Cell adhesion, spreading, and motility of GM3-expressing cells based on glycolipid -glycolipid interaction. *J. Biol. Chem.* 266, (pp. 17552-17558). Kojima, Y., S. Fukumoto, K. Furukawa, T. Okajima, J. Wiels, K. Yokoyama, Y. Suzuki, T.

Kolter, T. & K. Sandhoff (1998) Glycosphingolipid degradation and animal models of GM2-

Koshy, K. M., J. Wang & J. M. Boggs (1999) Divalent cation-mediated interaction between

Kovbasnjuk, O., R. Mourtazina, B. Baibakov, T. Wang, C. Elowsky, M. A. Choti, A. Kane &

transformation of colon cancer. *Proc Natl Acad Sci U S A.* 102, (pp. 19087-92). Kurzchalia, T. & R. Parton (1999) Membrane microdomains and caveolae. *Curt Opin Cell* 

Lafont, F., G. Tran Van Nhieu, K. Hanada, P. Sansonetti & F. G. van der Goot (2002) Initial

Lajoie, P. & I. R. Nabi (2010) Lipid rafts, caveolae, and their endocytosis. *Int Rev Cell Mol* 

Lala, P., S. Ito & C. A. Lingwood (2000) Transfection of MDCK cells with the MDR1 gene

Lannert, H., K. Gorgas, I. Meißner, F. T. Wieland & D. Jeckel (1998) Functional organization

182, (pp. 97-108).

(pp. 11138-11146).

71, (pp. 5713-22).

306-18.).

6251).

*Biol.* 11, (pp. 424-431).

*Biol.* 282, (pp. 135-63).

*Chem.* 273, (pp. 2939-2946).

*Immunol. Methods* 371, (pp. 48-60).

glycosphingolipids. *J Biol Chem.* 275, (pp. 15152-6).

gangliosidoses. *J Inherit Metab Dis.* 21, (pp. 548-63).

CD44-IpaB interaction. *Embo J.* 21, (pp. 4449-57.).

interferon/ type I interferon receptor mediated anti viral activity'. *J. Cell. Physiol.* 

modified by fatty acid content: Verotoxin 1 and Verotoxin 2c preferentially recognize different globotriaosyl ceramide fatty acid homologues. *J. Biol. Chem.* 269,

Comparison of detection methods for cell surface globotriaosylceramide. *J.* 

immunodeficiency virus type 1 localization in CD4-negative thymocytes: differentiation from a CD4-positive precursor allows productive infection. *J Virol.* 

Urano & M. Ohta (2000) Molecular cloning of globotriaosylceramide/CD77 synthase, a glycosyltransferase that initiates the synthesis of globo series

cerebroside sulfate and cerebrosides: an investigation of the effect of structural variations of lipids by electrospray ionization mass spectrometry. *Biophys J.* 77, (pp.

M. Donowitz (2005) The glycosphingolipid globotriaosylceramide in the metastatic

steps of Shigella infection depend on the cholesterol/sphingolipid raft-mediated

results in a major increase in globotriaosyl ceramide and cell sensitivity to verocytotoxin: role of P-gp in glycolipid biosynthesis. *J Biol Chem.* 275, (pp. 6246-

of the Golgi apparatus in glycosphingolipid biosynthesis. Lactosylceramide and subsequent glycosphingolipids are formed in the lumen of the late Golgi. *J Biol* 


Inokuchi, J.-I. & N. S. Radin (1987) Preparation of the active isomer of 1-phenyl-2-

Iwabuchi, K., K. Handa & S. Hakomori (2000) Separation of glycosphingolipid-enriched

Iwamura, K., K. Furukawa, M. Uchikawa, B. N. Sojka, Y. Kojima, J. Wiels, H. Shiku & T. Urano

A clue to the solution of the P1/P2/p puzzle. *J Biol Chem.* 278, (pp. 44429-38). Jelacic, S., C. L. Wobbe, D. R. Boster, M. A. Ciol, S. L. Watkins, P. I. Tarr & A. E. Stapleton

Jones, P. L., T. Korte & R. Blumenthal (1998) Conformational changes in cell surface HIV-1

Kaiser, H. J., D. Lingwood, I. Levental, J. L. Sampalo, L. Kalvodova, L. Rajendran & K.

Kannagi, R., S. B. Levery & S. Hakomori (1983a) Sequential change of carbohydrate antigen

developmentally regulated antigen SSEA-1. *J Biol Chem.* 257, (pp. 14865-74). Kannagi, R., R. Stroup, N. A. Cochran, D. L. Urdal, W. W. Young Jr. & S.-I. Hakomori, .

Katagiri, Y., T. Mori, H. Nakajima, C. Katagiri, T. Taguchi, T. Takeda, N. Kiyokawa & J.

Khan, F., F. Proulx & C. A. Lingwood (2009) Detergent-resistant globotriaosyl ceramide may

Khine, A. A., M. Firtel & C. A. Lingwood (1998) CD77-dependent retrograde transport of

during germinal center B-cell apoptosis. *J. Cell. Physiol.* 176, (pp. 281-292). Khine, A. A. & C. A. Lingwood (1994) Capping and receptor mediated endocytosis of cell

microdomains. *J. Biol. Chem.* 274, (pp. 35278-35282).

internalization. *J. Cell. Physiol.* 161, (pp. 319-332).

Membrane to the Endoplasmic Reticulum. *Traffic.* 4, (pp. 323-32).

Kanfer, J. & S. Hakomori (1983). *Sphingolipid Biochemistry* Plenum Press, New York

synthetase. *J. Lipid Res.* 28, (pp. 565-571).

co-receptors. *J Biol Chem.* 273, (pp. 404-9).

*Acad Sci USA.* 106, (pp. 16645-50).

*Kidney Int.* 75, (pp. 1209-16).

*Methods Enzymol.* 312, (pp. 488-94).

decanoylamino-3-morpholino-1-propanol, inhibitor of murine glucocerebroside

microdomains from caveolar membrane characterized by presence of caveolin.

(2003) The blood group P1 synthase gene is identical to the Gb3/CD77 synthase gene.

(2002) ABO and P1 blood group antigen expression and stx genotype and outcome of childhood Escherichia coli O157:H7 infections. *J. Infect. Dis.* 185, (pp. 214-9). Johannes, L., V. Pezo, F. Mallard, D. Tenza, A. Wiltz, A. Saint-Pol, J. Helft, C. Antony & P.

Benaroch (2003) Effects of HIV-1 Nef on Retrograde Transport from the Plasma

envelope glycoproteins are triggered by cooperation between cell surface CD4 and

Simons (2009) Order of lipid phases in model and plasma membranes. *Proc Natl* 

associated with differentiation of murine leukemia cells: i-I antigenic conversion and shifting of glycolipid synthesis. *Proc Natl Acad Sci U S A.* 80, (pp. 2844-8). Kannagi, R., E. Nudelman, S. B. Levery & S. Hakomori (1982) A series of human erythrocyte

glycosphingolipids reacting to the monoclonal antibody directed to a

(1983b) Factors affecting expression of glycolipid tumor antigens: Influence of ceramide composition and coexisting glycolipid on the antigenicity of gangliotriaosylceramide in murine lymphoma cells. *Cancer Res.* 43, (pp. 4997-5005,).

Fujimoto (1999) Activation of Src family kinase induced by Shiga toxin binding to globotriaosyl ceramide (Gb3/CD77) in low density, detergent-insoluble

define verotoxin/glomeruli-restricted hemolytic uremic syndrome pathology.

CD19 to the nuclear membrane: Functional Relationship between CD77 and CD19

bound verotoxin(Shiga-like toxin) 1; Chemical identification of an amino acid in the B subunit necessary for efficient receptor glycolipid binding and cellular


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 273

Livingstone, W. J., M. Moore, D. Innes, J. E. Bell & P. Simmonds (1996) Frequent infection of

Lund, N., D. Branch, M. Mylvaganam, D. Chark, X. Ma, D. Sakac, B. Binnington, J. Fantini,

Luo, C., K. Wang, Q. Liu de, Y. Li & Q. S. Zhao (2008) The functional roles of lipid rafts in T

Magerus-Chatinet, A., H. Yu, S. Garcia, E. Ducloux, B. Terris & M. Bomsel (2007) Galactosyl

Mahfoud, R., A. Manis, B. Binnington, C. Ackerley & C. A. Lingwood (2010) A major

Mahfoud, R., A. Manis & C. Lingwood (2009) Fatty acid-dependent globotriaosyl ceramide

Mahfoud, R., M. Mylvaganam, C. A. Lingwood & J. Fantini (2002b) A novel soluble analog

Maloney, M. D. & C. A. Lingwood (1994) CD19 has a potential CD77 (globotriaosyl

Mamelak, D., M. Mylvaganam, E. Tanahashi, H. Ito, H. Ishida, M. Kiso & C. Lingwood

Manes, S., G. del Real, R. A. Lacalle, P. Lucas, Gomez-Mouton, C., , S. Sanchez-Palomino, R.

position required for hsc70 recognition. *Carbohydr Res.* 335, (pp. 91-100.). Mamelak, D., M. Mylvaganam, H. Whetstone, E. Hartmann, W. Lennarz, P. Wyrick, J.

*Escherichia coli* pathogenesis. *J. Exp. Med.* 180, (pp. 191-201).

*Biochemistry.* 40, (pp. 3572-3582).

derived dendritic cells to autologous T cells. *Virology.* 362, (pp. 67-74). Mahfoud, R., N. Garmy, M. Maresca, N. Yahi, A. Puigserver & J. Fantini (2002a)

HIV-1 proteins. *J Biol Chem.* 277, (pp. 11292-6).

Heterosexual Transmission Study Group. *Lancet.* 348, (pp. 649-54).

4980-91).

59).

55).

*Immunol.* 5, (pp. 1-7).

peripheral blood CD8-positive T-lymphocytes with HIV-1. Edinburgh

A. Puri, R. Blumenthal & C. Lingwood (2006) A novel soluble mimic of the glycolipid globotriaosylceramide inhibits HIV infection. *AIDS.* 20, (pp. 333-343). Lund, N., D. R. Branch, D. Sakac, C. Lingwood, C. Siatskas, C. Robinson, R. Brady & J.

Medin (2005) Lack of Susceptibility of Cells from Patients with Fabry Disease to Infection with R5 Human Immunodeficiency Virus. *AIDS.* 19, (pp. 1543-1546). Lund, N., S. Ramkumar, M. L. Olsson, D. Sakac, Y. Yahalom, C. Levene, A. Hellberg, X. Ma,

D. Jung, B. Binnington, C. A. Lingwood & D. R. Branch (2009) The human Pk histoblood group antigen provides protection against HIVinfection. *Blood.* 113, (pp.

cell activation, immune diseases and HIV infection and prevention. *Cell Mol* 

ceramide expressed on dendritic cells can mediate HIV-1 transfer from monocyte

Identification of a common sphingolipid-binding domain in Alzheimer, prion and

fraction of glycosphingolipids in model and cellular cholesterol containing membranes are undetectable by their binding proteins. *J Biol Chem.* 285, (pp. 36049-

receptor function in detergent resistant model membranes *J Lip Res.* 50, (pp. 1744-

of the HIV-1 fusion cofactor, globotriaosylceramide(Gb3), eliminates the cholesterol requirement for high affinity gp120/Gb3 interaction. *J. Lipid Res.* 43, (pp. 1670-1679).

ceramide)-binding site with sequence similarity to verotoxin B-subunits: Implications of molecular mimicry for B cell adhesion and enterohemorrhagic

(2001a) The aglycone of sulfogalactolipids can alter the sulfate ester substitution

Raulston, H. Han, P. Hoffman & C. Lingwood (2001b) Hsp70s contain a specific sulfogalactolipid binding site. Differential aglycone influence on sulfogalactosyl ceramide binding by prokaryotic and eukaryotic hsp70 family members.

Delgado, J. Alcami, E. Mira & C. Martinez-A (2000) Membrane raft microdomains


Lauer, S., J. VanWye, T. Harrison, H. McManus, B. U. Samuel, N. L. Hiller, N. Mohandas &

Leffler, H. & C. Svanborg-Eden (1981) Glycolipid receptors for uropathogenic Escherichia

Lencer, W. (2001) Microbes and microbial toxins: paradigms for microbial-mucosal toxins.

Lencer, W. I. & D. Saslowsky (2005) Raft trafficking of AB5 subunit bacterial toxins. *Biochim* 

Liao, Z., L. M. Cimakasky, R. Hampton, D. H. Nguyen & J. E. Hildreth (2001) Lipid rafts and

Lingwood, C. A. (1996) Aglycone Modulation of Glycolipid Receptor Function. *Glycoconj J.* 

Lingwood, C. A. (1998) Oligosaccharide receptors for bacteria: a view to a kill. *Current* 

Lingwood, C. A. (2000) Verotoxin/globotriaosyl ceramide recognition: angiopathy, angiogenesis and antineoplasia. *Bioscience Reports.* 19, (pp. 345-354). Lingwood, C. A., H. Law, S. Richardson, M. Petric, J. L. Brunton, S. DeGrandis & M. Karmali

Lingwood, C. A. & S. C. K. Yiu (1992) Glycolipid modification of a2-interferon binding:

Lingwood, D., B. Binnington, T. Róg, I. Vattulainen, M. Grzybek, U. Coskun, C. Lingwood &

Lingwood, D. & K. Simons (2007) Detergent resistance as a tool in membrane research. *Nat* 

Lingwood, D. & K. Simons (2010) Lipid rafts as a membrane-organizing principle. *Science.* 

Littman, D. R. (1998) Chemokine receptors: keys to AIDS pathogenesis? *Cell.* 93, (pp. 677-80). Liao, Z., D.R. Graham & J.E. Hildreth (2003) Lipid rafts and HIV pathogenesis: virion-

Liu, X.-H., C. A. Lingwood & P. Ray (1999) Recruitment of renal tubular epithelial cells

toxin. *Am J Physiol - Gastro Liver Physiol.* 280, (pp. G781-786).

1. *AIDS Res Hum Retroviruses.* 17, (pp. 1009-19.).

verotoxin *in vitro*. *J. Biol. Chem.* 262, (pp. 8834-8839).

toxin) B-subunit. *Biochem. J.* 283, (pp. 25-26).

activity. *Nature Chem. Biol.* 7, (pp. 260-2).

*Res Hum Retroviruses*. 19, (pp. 675-87).

disease. *Kidney Int.* 55, (pp. 554-561).

*Biol Chem.* 274, (pp. 14662-14669).

*Biophys Acta.* 1746, (pp. 314-21).

13, (pp. 495-503).

*Biology.* 2, (pp. 695-700).

*Protoc.* 2, (pp. 2159-65).

327, (pp. 46-50).

920-9).

K. Haldar (2000) Vacuolar uptake of host components, and a role for cholesterol and sphingomyelin in malarial infection. *The EMBO Journal.* 19, (pp. 3556-64). Lauvrak, S. U., M. L. Torgersen & K. Sandvig (2004) Efficient endosome-to-Golgi transport of Shiga toxin is dependent on dynamin and clathrin. *J Cell Sci.* 117, (pp. 2321-31). Lee, L., A. Abe & J. A. Shayman (1999) Improved inhibitors of glucosylceramide synthase. *J* 

coli on human erythrocytes and uroepithelial cells. *Infection and Immunity.* 34, (pp.

V. Cholera: invasion of the intestinal epithelial barrier by a stably folded protein

HIV pathogenesis: host membrane cholesterol is required for infection by HIV type

(1987) Glycolipid binding of purified and recombinant *Escherichia coli*-produced

Sequence similarity between a2-interferon receptor and the verotoxin (Shiga-like

K. Simons (2011) Cholesterol modulates glycolipid conformation and receptor

associated cholesterol is required for fusion and infection of susceptible cells. *AIDS* 

expressing Verotoxin-1 (Stx1) receptors in HIV-1 transgenic mice with renal


mediate lateral assemblies required for HIV-1 infection. *EMBO Reports.* 1, (pp. 190- 196).

Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 275

Muthing, J., C. H. Schweppe, H. Karch & A. W. Friedrich (2009) Shiga toxins,

Mylvaganam, M. & C. Lingwood (1999a) Adamantyl globotriaosyl ceramide- a monovalent

Mylvaganam, M. & C. A. Lingwood (1999b) A convenient oxidation of natural

Nelson, J. A., C. A. Wiley, C. Reynolds-Kohler, C. E. Reese, W. Margaretten & J. A. Levy

Nguyen, D. H. & J. E. Hildreth (2000) Evidence for budding of human immunodeficiency

Nichols, B., A. Kenworthy, R. Polishchuk, R. Lodge, T. Roberts, K. Hirschberg, R. Phair & L.-

Niemela, P. S., M. T. Hyvonen & I. Vattulainen (2009) Atom-scale molecular interactions in

Norkin, L. C. (1999) Simian virus 40 infection via MHC class I molecules and caveolae.

Nutikka, A. & C. Lingwood (2004) Generation of receptor active, globotriaosyl

Nyholm, P.G. & I. Pascher (1993) Orientation of the saccharide chains of glycolipids at the

Okuda, T., N. Tokuda, S. Numata, M. Ito, M. Ohta, K. Kawamura, J. Wiels, T. Urano, O.

Ono, A. & E. O. Freed (2001) Plasma membrane rafts play a critical role in HIV-1 assembly

Ono, M., K. Handa, S. Sonnino, D. A. Withers, H. Nagai & S. Hakomori (2001) GM3

receptor. *Biochem. Biophys. Res. Commun.* 257, (pp. 391-394).

HIV-1 strains. *Antiviral Res*. 56, (pp. 233-51).

complex. *J Cell Biol.* 153, (pp. 529-41).

*Immunological Reviews.* 168, (pp. 13-22).

*Biochemistry.* 40, (pp. 6414-21).

with gastrointestinal symptoms. *Lancet.* 1, (pp. 259-62.).

lipid raft mixtures. *Biochim Biophys Acta.* 1788, (pp. 122-35).

glycosphingolipid receptor activity. *Glycoconj. J.* 20, (pp. 33-38).

sensitivity to verotoxins. *J Biol Chem.* 281, (pp. 10230-5).

and release. *Proc Natl Acad Sci U S A.* 98, (pp. 13925-30.).

252-64).

(pp. 3264-72).

1225-34).

glycosphingolipid diversity, and endothelial cell injury. *Thromb Haemost.* 101, (pp.

soluble glycolipid mimic which inhibits verotoxin binding to its glycolipid

glycosphingolipids to their "ceramide acids"for neoglycoconjugation: Bovine serum albumin-glycoceramide acid conjugates as investigative probes for HIV coat protein gp120-glycosphingolipid interactions. *J Biol Chem.* 274, (pp. 20725-20732). Mylvaganam, M. & C. A. Lingwood (2003). A preamble to aglycone reconstruction for membrane-presented glycolipids Wiley-VCH Press, Weinheim, Germany Nehete, P.N., E.M. Vela, M.M. Hossain, A.K. Sarkar, N. Hahi, J. Fantini & K.J. Sastry (2002)

A post-CD4-binding step involving interaction of the V3 region of viral gp120 with host cell surface glycosphingolipids is common to entry and infection by diverse

(1988) Human immunodeficiency virus detected in bowel epithelium from patients

virus type 1 selectively from glycolipid-enriched membrane lipid rafts. *J Virol.* 74,

S. J. (2001) Rapid cycling of lipid raft markers between the cell surface and Golgi

ceramide/cholesterol lipid 'rafts' *in vitro*: a new assay to define factors affecting

membrane surface: conformational analysis of the glucose-ceramide and the glucose-glyceride linkages using molecular mechanics (MM3). *Biochem*. 32, (pp.

Tajima & K. Furukawa (2006) Targeted disruption of Gb3/CD77 synthase gene resulted in the complete deletion of globo-series glycosphingolipids and loss of

ganglioside inhibits CD9-facilitated haptotactic cell motility: coexpression of GM3 and CD9 is essential in the downregulation of tumor cell motility and malignancy.


Manes, S., G. del Real & A. C. Martinez (2003) Pathogens: raft hijackers. *Nature reviews.* 

Manes, S., R. Lacalle, C. Gomez-Mouton, G. del Real, E. Mira & A. C. Martinez (2001)

Manes, S. & A. C. Martinez (2004) Cholesterol domains regulate the actin cytoskeleton at the

Mangeney, M., Y. Richard, D. Coulaud, T. Tursz & J. Wiels (1991) CD77: an antigen of germinal center B cells entering apoptosis. *Eur. J. Immunol.* 21, (pp. 1131-1140). Manie, S. N., S. de Breyne, S. Vincent & D. Gerlier (2000) Measles virus structural

Marjomaki, V., V. Pietiainen, H. Matilainen, P. Upla, J. Ivaska, L. Nissinen, H. Reunanen, P.

Marras, D., L. Bruggeman, F. Gao, N. Tanji, M. Mansukhani, A. Cara, M. Ross, G. Gusella, G.

Marsh, M. & A. Helenius (1989) Virus entry into animal cells. *Adv Virus res.* 36, (pp. 107-151). Millan, J., J. Cerny, V. Horejsi & M. Alonso (1999) CD4 segregates into specific detergentresistant T-cell membrane microdomains. *Tissue Antigens.* 53, (pp. 33-40). Miller-Podraza, H., C. Andersson & K. A. Karlsson (1993) New method for the isolation of

Miller-Podraza, H., L. Johansson, P. Johansson, T. Larsson, M. Matrosovich & K. A. Karlsson

Mori, T., N. Kiyokawa, Y. U. Katagiri, T. Taguchi, T. Suzuki, T. Sekino, N. Sato, K. Ohmi, H.

Moriuchi, H., M. Moriuchi, J. Arthos, J. Hoxie & A. S. Fauci (1997) Promonocytic U937

Moulds, J. M. & J. J. Moulds (2000) Blood group associations with parasites, bacteria, and

Mukherjee, S. & F. Maxfield (2000) Role of membrane organization and membrane domains

viruses. *Transfusion Medicine Reviews.* 14, (pp. 302-11).

in endocytic lipid trafficking. *Traffic.* 1, (pp. 203-211).

leading edge of moving cells. *Trends Cell Biol.* 14, (pp. 275-8).

for virus assembly. *Journal of virology.* 74, (pp. 305-11).

associated nephropathy. *Nature Medicine.* 8, (pp. 522-6).

*Journal of Virology.* 76, (pp. 1856-65).

196).

(pp. 147-157).

1168, (pp. 330-9).

(pp. 975-82).

1260-1268).

71).

*Immunology.* 3, (pp. 557-68).

mediate lateral assemblies required for HIV-1 infection. *EMBO Reports.* 1, (pp. 190-

Membrane raft microdomains in chemokine receptor function. *Sem Immunol.* 13,

components are enriched into lipid raft microdomains: a potential cellular location

Huttunen, T. Hyypia & J. Heino (2002) Internalization of echovirus 1 in caveolae.

Benson, V. D'Agati, B. Hahn, M. Klotman & P. Klotman (2002) Replication and compartmentalization of HIV-1 in kidney epithelium of patients with HIV-

polyglycosylceramides from human erythrocyte membranes. *Biochim Biophys Acta.* 

(2000) A strain of human influenza A virus binds to extended but not short gangliosides as assayed by thin-layer chromatography overlay. *Glycobiology.* 10,

Nakajima, T. Takeda & J. Fukimoto (2000) Globotriaosyl ceramide (CD77/Gb3) in the glycolipid-enriched membrane domain participates in B-cell receptor-mediated apoptosis by regulating Lyn kinase activity in human B cells. *Exp. Hemat.* 28, (pp.

subclones expressing CD4 and CXCR4 are resistant to infection with and cell-to-cell fusion by T-cell-tropic human immunodeficiency virus type 1. *J Virol.* 71, (pp. 9664-


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 277

Ramkumar, S., D. Sakac, B. Binnington, D. R. Branch & C. A. Lingwood (2009) Induction of

Raulin, J. (2002) Human immunodeficiency virus and host cell lipids. Interesting pathways

Rawat, S. S., S. A. Gallo, J. Eaton, T. D. Martin, S. Ablan, V. N. KewalRamani, J. M. Wang, R.

Rawat, S. S., M. Viard, S. A. Gallo, R. Blumenthal & A. Puri (2006) Sphingolipids, cholesterol, and HIV-1: A paradigm in viral fusion. *Glycoconj J.* 23, (pp. 189-97). Rawat, S. S., M. Viard, S. A. Gallo, A. Rein, R. Blumenthal & A. Puri (2003) Modulation of

Rios, M. & C. Bianco (2000) The role of blood group antigens in infectious diseases. *Seminars* 

Rosa Borges, A., L. Wieczorek, B. Johnson, A. J. Benesi, B. K. Brown, R. D. Kensinger, F. C.

Rusnati, M., C. Urbinati, E. Tanghetti, P. Dell'Era, H. Lortat-Jacob & M. Presta (2002) Cell

Sabharanjak, S., P. Sharma, R. G. Parton & S. Mayor (2002) GPI-anchored proteins are

Saha, K., J. Zhang, A. Gupta, R. Dave, M. Yimen & B. Zerhouni (2001a) Isolation of primary HIV-1 that target CD8+ T lymphocytes using CD8 as a receptor. *Nat Med.* 7, (pp. 65-72). Saha, K., J. Zhang & B. Zerhouni (2001b) Evidence of productively infected CD8+ T cells in

Samuel, B. U., N. Mohandas, T. Harrison, H. McManus, W. Rosse, M. Reid & K. Haldar

Sandvig, K., M. Ryd, Ø. Garred, E. Schweda & P. K. Holm (1994) Retrograde transport from

Savarino, A., T. Bensi, A. Chiocchetti, F. Bottarel, R. Mesturini, E. Ferrero, L. Calosso, S. Deaglio,

Scheiffele, P., A. Rietveld, T. Wilk & K. Simons (1999) Influenza viruses select ordered lipid

is regulated by butyric acid and cAMP. *J. Cell. Biol.* 126, (pp. 53-64).

loop of the viral envelope glycoprotein gp120. *FASEB J.* 17, (pp. 461-3).

globotriaosyl ceramide levels. *Glycobiology.* 19, (pp. 76-82).

78, (pp. 7360-8).

*Biol.* 20, (pp. 243-54).

*in Hematology.* 37, (pp. 177-85).

HIV-1. *Virology.* 408, (pp. 80-8).

*Syndr.* 26, (pp. 199-207).

*Chem.* 276, (pp. 29319-29.).

*Chemistry.* 274, (pp. 2038-44).

2. *Proc Natl Acad Sci U S A.* 99, (pp. 4367-72).

independent pinocytic pathway. *Dev Cell.* 2, (pp. 411-23).

in research for a new HIV therapy. *Prog Lipid Res.* 41, (pp. 27-65).

HIV resistance: Cell susceptibility to infection is an inverse function of

Blumenthal & A. Puri (2004) Elevated expression of GM3 in receptor-bearing targets confers resistance to human immunodeficiency virus type 1 fusion. *J Virol.* 

entry of enveloped viruses by cholesterol and sphingolipids (Review). *Mol Membr* 

Krebs, B. Wigdahl, R. Blumenthal, A. Puri, F. E. McCutchan, D. L. Birx, V. R. Polonis & C. L. Schengrund (2010) Multivalent dendrimeric compounds containing carbohydrates expressed on immune cells inhibit infection by primary isolates of

membrane GM1 ganglioside is a functional coreceptor for fibroblast growth factor

delivered to recycling endosomes via a distinct cdc42-regulated, clathrin-

patients with AIDS: implications for HIV-1 pathogenesis. *J Acquir Immune Defic* 

(2001) The role of cholesterol and glycosylphosphatidylinositol-anchored proteins of erythrocyte rafts in regulating raft protein content and malarial infection. *J Biol* 

the Golgi complex to the ER of both Shiga toxin and the nontoxic Shiga B-fragment

E. Ortolan, S. Butto, A. Cafaro, T. Katada, B. Ensoli, F. Malavasi & U. Dianzani (2003) Human CD38 interferes with HIV-1 fusion through a sequence homologous to the V3

domains during budding from the plasma membrane. *The Journal of Biological* 


Parton, R. G. (1994) Ultrastructural localization of gangliosides; GM1 is concentrated in

Parton, R. G. & A. A. Richards (2003) Lipid rafts and caveolae as portals for endocytosis:

Pellizzari, A., H. Pang & C. A. Lingwood (1992) Binding of verocytotoxin 1 to its receptor is

Percherancier, Y., B. Lagane, T. Planchenault, I. Staropoli, R. Altmeyer, J. Virelizier, F.

Petruzziello, T. N., I. A. Mawji, M. Khan & P. A. Marsden (2009) Verotoxin biology: molecular events in vascular endothelial injury. *Kidney Int Suppl.* (pp. S17-9). Piguet, V. & D. Trono (1999) The Nef protein of primate lentiviruses. *Rev Med Virol.* 9, (pp.

Pochechueva, T., A. Chinarev, M. Spengler, E. Korchagina, V. Heinzelmann-Schwarz, N.

Popik, W., T. M. Alce & W. C. Au (2002) Human immunodeficiency virus type 1 uses lipid

Pumarola-Sune, T., B. A. Navia, C. Cordon-Cardo, E. S. Cho & R. W. Price (1987) HIV

Puri, A., P. Hug, K. Jernigan, J. Barchi, H.-Y. Kim, J. Hamilton, J. Wiels, G. J. Murray, R. O.

Puri, A., P. Hug, K. Jernigan, P. Rose & R. Blumenthal (1999) Role of glycosphingolipids in

Puri, A., S. S. Rawat, H. M. Lin, C. M. Finnegan, J. Mikovits, F. W. Ruscetti & R. Blumenthal

Radhakrishnan, A., T. G. Anderson & H. M. McConnell (2000) Condensed complexes, rafts,

Rajendran, L. & K. Simons (2005) Lipid rafts and membrane dynamics. *J Cell Sci.* 118, (pp.

Ramegowda, B. & V. L. Tesh (1996) Differentiation-associated toxin receptor modulation,

envelope glycoprotein. *Proc. Natl. Acad. Sci.* 95, (pp. 14435-14440).

Bovin & R. Rieben (2011) Multiplex suspension array for human anti-carbohydrate

raft-colocalized CD4 and chemokine receptors for productive entry into CD4(+) T

antigen in the brains of patients with the AIDS dementia complex. *Ann Neurol.* 21,

Brady & R. Blumenthal (1998) The neutral glycosphingolipid globotriaosylceramide promotes fusion mediated by a CD4-dependent CXCR4-utilizing HIV type 1

HIV-1 entry: requirement of globotriosylceramide (Gb3) in CD4/CXCR4-

(2004) An inhibitor of glycosphingolipid metabolism blocks HIV-1 infection of

and the chemical activity of cholesterol in membranes. *Proc Natl Acad Sci.* 97, (pp.

cytokine production, and sensitivity to Shiga-like toxins in human monocytes and

influenced by differences in receptor fatty acid content. *Biochemistry.* 31, (pp. 1363-

Arenzana-Seisdedos, D. Hoessli & F. Bachelerie (2003) HIV-1 Entry into T-cells Is Not Dependent on CD4 and CCR5 Localization to Sphingolipid-enriched, Detergent-resistant, Raft Membrane Domains. *J Biol Chem.* 278, (pp. 3153-61). Peterlin, B. M. & D. Trono (2003) Hide, shield and strike back: how HIV-infected cells avoid

new insights and common mechanisms. *Traffic.* 4, (pp. 724-38).

caveolae. *J Histochem Cytochem.* 42, (pp. 155-166).

immune eradication. *Nat Rev Immunol.* 3, (pp. 97-107).

antibody profiling. *Analyst.* 136, (pp. 560-9).

dependent fusion. *Biosci Rep.* 19, (pp. 317-325).

moncytic cell lines. *Infect. Immun.* 64, (pp. 1173-1180).

primary T-cells. *Aids.* 18, (pp. 849-58).

cells. *J Virol.* 76, (pp. 4709-22.).

70).

111-20).

(pp. 490-6).

12422-12427).

1099-102).


Glycosphingolipids in HIV/AIDS: The Potential Therapeutic Application 279

Stapleton, A., E. Nudelman, H. Clausen, S. Hakomori & W. E. Stamm (1992) Binding of

Steffensen, R., K. Carlier, J. Wiels, S. B. Levery, M. Stroud, B. Cederen, S. B. Nilsson, E. P.

Stults, C. L. M., C. H. Sweeley & B. A. Macher (1989) Glycosphingolipids; structure biological source and properties. *Methods Enzymol.* 179, (pp. 167-214). Tagami, S., J. Inokuchi Ji, K. Kabayama, H. Yoshimura, F. Kitamura, S. Uemura, C. Ogawa,

Tateno, M., F. Gonzalez-Scarano & J. A. Levy (1989) Human immunodeficiency virus can

Teufel, A., T. Maass, P. R. Galle & N. Malik (2009) The longevity assurance homologue of yeast lag1 (Lass) gene family (review). *Int J Mol Med.* 23, (pp. 135-40). Thuresson, B., J.S. Westman & M.L. Olsson (2011) Identification of a novel A4GALT exon reveals the genetic basis of the P1/P2 histo-blood groups. *Blood*. 117, (pp. 678-87). Tsai, B., J. M. Gilbert, T. Stehle, W. Lencer, T. L. Benjamin & T. A. Rapoport (2003)

Turner, M., K. Kher, T. Rakussan, L. D'Angelo, S. Kapur, D. Selby & P. Ray (1997) Atypical

Ullrich, R., W. Schmidt, T. Zippel, T. Schneider, M. Zeitz & E. O. Riecken (1998) Mucosal

van Alphen, L., L. Gellen-van den Broek, L. Blaas, M. van Ham & J. Dankert (1991) Blocking

van der Goot, F. G. & T. Harder (2001) Raft membrane domains: from a liquid-ordered membrane phase to a site of pathogen attack. *Semin Immunol.* 13, (pp. 89-97). van Eijk, M., J. Aten, N. Bijl, R. Ottenhoff, C. P. van Roomen, P. F. Dubbelhuis, I. Seeman, K.

Villard, R., D. Hammache, G. Delapierre, F. Fotiadu, G. Buono & J. Fantini (2002) Asymmetric

tools to study virus-glycolipid interactions. *Chembiochem.* 3, (pp. 517-25.).

safe door for micro-organisms? *Biol Cell.* 102, (pp. 391-407).

*Investigation.* 90, (pp. 965-72).

(pp. 3085-92).

22, (pp. 4346-55).

*Ped Nephrol.* 11, (pp. 161-163).

*Infect Immun.* 59, (pp. 4473-4477).

HIV infection. *Pathobiol.* 66, (pp. 145-150).

4287-90.).

and signaling. *Adv Exp Med Biol.* 688, (pp. 60-71).

uropathogenic Escherichia coli R45 to glycolipids extracted from vaginal epithelial cells is dependent on histo-blood group secretor status. *The Journal of Clinical* 

Bennett, C. Jersild & H. Clausen (2000) Cloning and expression of the histo-blood group Pk UDP-galactose: Galbeta 1-4Glcbeta 1-Cer alpha 1, 4-galactosyltransferase. Molecular genetic basis of the p phenotype. *J Biol Chem.* 275, (pp. 16723-16729). Stiban, J., R. Tidhar & A. H. Futerman (2010) Ceramide synthases: roles in cell physiology

A. Ishii, M. Saito, Y. Ohtsuka, S. Sakaue & Y. Igarashi (2002) Ganglioside GM3 participates in the pathological conditions of insulin resistance. *J Biol Chem.* 277,

infect CD4-negative human fibroblastoid cells. *Proc Natl Acad Sci U S A.* 86, (pp.

Gangliosides are receptors for murine polyoma virus and SV40. *The EMBO Journal.* 

hemolytic uremic syndrome in human immunodeficiency virus-1 infected children.

of fimbria-mediated adherence of *Haemophilus influenzae* by sialyl gangliosides.

Ghauharali-van der Vlugt, H. S. Overkleeft, C. Arbeeny, A. K. Groen & J. M. Aerts (2009) Reducing glycosphingolipid content in adipose tissue of obese mice restores insulin sensitivity, adipogenesis and reduces inflammation. *PLoS One.* 4, (pp. e4723). Vieira, F. S., G. Correa, M. Einicker-Lamas & R. Coutinho-Silva (2010) Host-cell lipid rafts: a

synthesis of water-soluble analogues of galactosylceramide, an HIV-1 receptor: new


Schwartz-Albeiz, R., B. Dorken, P. Moller, N. T. Brodin, D. A. Monner & B. Kniep (1991)

Shnaper, S., K. Sackett, S. A. Gallo, R. Blumenthal & Y. Shai (2004) The C- and the N-

Simons, K. & R. Ehehalt (2002) Cholesterol, lipid rafts, and disease. *J Clin Invest.* 110, (pp.

Simons, K. & J. Gruenberg (2000) Jamming the endosomal system: lipid rafts and lysosomal

Simons, K. & E. Ikonen (1997) Functional rafts in cell membranes. *Nature.* 387, (pp. 569-572). Simons, K. & D. Toomre (2000) Lipid rafts and signal transduction. *Nature Rev.* 1, (pp. 31-39). Simons, K. & W. L. Vaz (2004) Model systems, lipid rafts, and cell membranes. *Annu Rev* 

Smith, D. C., D. J. Sillence, T. Falguieres, R. M. Jarvis, L. Johannes, J. M. Lord, F. M. Platt & L.

Sorice, M., T. Garofalo, R. Misasi, A. Longo, V. Mattei, P. Sale, V. Dolo, R. Gradini & A. Pavan

Sorice, M., I. Parolini, T. Sansolini, T. Garofalo, V. Dolo, M. Sargiacomo, T. Tai, C. Peschle,

Speck, R. F., U. Esser, M. L. Penn, D. A. Eckstein, L. Pulliam, S. Y. Chan & M. A. Goldsmith

Sperandio, M. (2006) Selectins and glycosyltransferases in leukocyte rolling *in vivo*. *FEBS J.* 

Spitalnik, P. & S. Spitalnik (1995) The P blood group system: Biochemical, serological and

Sprong, H., B. Kruithof, R. Leijendekker, J. W. Slot, G. van Meer & P. van der Sluijs (1998)

protein of the endoplasmic reticulum. *J Biol Chem.* 273, (pp. 25880-25888).

immunodeficiency virus type 1. *Curr Biol.* 9, (pp. 547-50.).

clinical aspects. *Transfus Med Rev.* 9, (pp. 110-122).

M. Roberts (2006) The Association of Shiga-like Toxin with Detergent-resistant Membranes Is Modulated by Glucosylceramide and Is an Essential Requirement in the Endoplasmic Reticulum for a Cytotoxic Effect. *Mol Biol Cell.* 17, (pp. 1375-87). Solter, D. & B. B. Knowles (1978) Monoclonal antibody defining a stage-specific mouse embryonic antigen (SSEA-1). *Proc Natl Acad Sci U S A.* 75, (pp. 5565-9). Sorice, M., T. Garofalo, L. Misasi R, A, , J. Mikulak, V. Dolo, G. Pontieri & P. A. (2000)

Association between GM3 and CD4-Ick complex in human peripheral blood

(2001) Evidence for cell surface association between CXCR4 and ganglioside GM3 after gp120 binding in SupT1 lymphoblastoid cells. *FEBS Lett.* 506, (pp. 55-60.). Sorice, M., A. Longo, T. Garofalo, V. Mattei, R. Misasi & A. Pavan (2004) Role of GM3-

enriched microdomains in signal transduction regulation in T lymphocytes.

M. R. Torrisi & A. Pavan (1997) Evidence for the existence of ganglioside-enriched plasma membrane domains in human peripheral lymphocytes. *J Lipid Res.* 38, (pp.

(1999) A trans-receptor mechanism for infection of CD4-negative cells by human

UDP-galactose:ceramide galactosyltransferase is a class I integral membrane

corresponding to germinal center B cells. *Int Immunol.* 2, (pp. 929-936). Sekino, T., N. Kiyokawa, T. Taguchi, H. Takenouchi, J. Matsui, W. R. Tang, T. Suzuki, H.

*Microbiol Immunol.* 48, (pp. 377-87).

*Biophys Biomol Struct.* 33, (pp. 269-95).

lymphocytes. *Glycoconj J.* 17, (pp. 247-252).

*Glycoconj J.* 20, (pp. 63-70).

969-80).

273, (pp. 4377-89).

597-603).

with cholesterol, respectively. *J Biol Chem.* (pp.

storage diseases. *Trends Cell Biol.* 10, (pp. 459-62).

Neutral glycosphingolipids of the globo-series characterize activation stages

Nakajima, M. Saito, K. Ohmi, Y. U. Katagiri, H. Okita, H. Nakao, T. Takeda & J. Fujimoto (2004) Characterization of a shiga-toxin 1-resistant stock of vero cells.

terminal regions of gp41 ectodomain fuse membranes enriched and not enriched


Vyas, A., H. Patel, S. Fromholt, M. Heffer-Lauc, K. Vyas, J. Dang, M. Schachner & R. Schnaar (2002) Gangliosides are functional nerve cell ligands for myelin-associated glycoprotein (MAG), an inhibitor of nerve regeneration. *Proc Natl Acad Sci.* 99, (pp. 8412-7).

**13** 

*South Africa* 

**Drug-Drug Interactions as a Challenge** 

*Walter Sisulu University, Department of Pharmacology, P Bag X 1, Mthatha,* 

Human immunodeficiency virus (HIV) is a major challenge in the medical fraternity worldwide. According to the UNAIDS report on the global AIDS epidemic, in 2004 it had affected more than 42 million people, and of these 25 million resided in the sub-Sahara

This virus has no cure; the lives of the infected patients can only be prolonged using lifelong highly active antiretroviral (ARV) therapy (HAART). HAART has been proven to suppress HIV-1 viral replication continuously thus reducing mortality and morbidity in treated patients. It has further been proven that HAART is only highly effective if prescribed in combination of more than 2 drugs. However these drug combinations can be presented with potential drug-drug interactions (DDIs) an important cause of adverse drug reactions

DDIs are well-recognised causes of adverse drug effects (ADEs) (Bates *et al*., 1995). According to Juurlink *et al*. (2003), DDIs do cause particularly important type of adverse drug event because they are often predictable based on previous reports, clinical studies,

According to Johnson *et al*. (1999), DDIs are classified as an important category of ADEs. Drug interactions result in undesirable modification of the action of one or more concurrently administered agents. The interaction may cause treatment failure, an increased pharmacologic effect, or a toxic effect, which may be fatal. Because DDIs usually have a specific time course (i.e., onset and duration), they are more predictable (and preventable) then ADRs (adverse drug reactions). Bates *et al*. (1999) state that Preventable DDIs account

In HIV-infected patients, the introduction of HAART has led to reduced morbidity and mortality in treated patients (Egger *et al.*, 2002). However, in a substantial proportion of patients, the effectiveness of HAART has not been sufficient due to occurrence of virological failure and immunological decay (Bartlett *et al.*, 2001). All this has been due to failure to

The possible causes of DDIs include drug combinations, lack of communication between the prescribers and medical history, increase in the number of newly marketed drugs and polypharmacy. Specific patients who are risk for DDIs include the elderly, people living with HIV/AIDS. Patients with HIV are more at risk for the virus because they are treated

for about one third of ADEs but incur about one half of the total ADE costs.

determine drug interactions and prevention of toxic effects. (Boffito *et al.*, 2005).

**1. Introduction** 

Africa (UNAIDS, 2004 & Bhigjee, 2005).

and an understanding of pharmacological principles.

(ADRs) (Highleyman, 2007).

**in the Treatment of HIV/AIDS** 

Norah L. Katende-Kyenda


## **Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS**

Norah L. Katende-Kyenda

*Walter Sisulu University, Department of Pharmacology, P Bag X 1, Mthatha, South Africa* 

## **1. Introduction**

280 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Vyas, A., H. Patel, S. Fromholt, M. Heffer-Lauc, K. Vyas, J. Dang, M. Schachner & R. Schnaar

Waarts, B. L., R. Bittman & J. Wilschut (2002) Sphingolipid and cholesterol dependence of

Waddell, T., A. Cohen & C. A. Lingwood (1990) Induction of verotoxin sensitivity in

Wang, W. K., T. Dudek, M. Essex & T. H. Lee (1999) Hypervariable region 3 residues of HIV

Ward, J. M., T. J. O'Leary, G. B. Baskin, R. Benveniste, C. A. Harris, P. L. Nara & R. H. Rhodes

Weber, K. T., D. Hammache, J. Fantini & B. Ganem (2000) Synthesis of glycolipid analogues

Weis, F. M. B. & R. J. Davis (1990) Regulation of epidermal growth factor receptor signal

Westerlund, B. & J. P. Slotte (2009) How the molecular features of glycosphingolipids affect domain formation in fluid membranes. *Biochim Biophys Acta.* 1788, (pp. 194-201). Whetstone, D. & C. Lingwood (2003) 3'Sulfogalactolipid binding specifically inhibits Hsp70

Wiels, J., M. Fellous & T. Tursz (1981) Monoclonal antibody against a Burkitt-lymphoma-

Wiels, J., M. Mangeney, C. Tetaud & T. Tursz (1991) Sequential shifts in the three major

Wiley, C. A., R. D. Schrier, J. A. Nelson, P. W. Lampert & M. B. Oldstone (1986) Cellular

Xiao, L., S. M. Owen, I. Goldman, A. A. Lal, J. J. deJong, J. Goudsmit & R. B. Lal (1998) CCR5

the V3 domain that predicts CCR-5 usage. *Virology.* 240, (pp. 83-92). Yahi, N., A. Aulas & J. Fantini (2010) How cholesterol constrains glycolipid conformation for

glycosphingolipid series are associated with B cell differentiation. *Int Immunol.* 3,

localization of human immunodeficiency virus infection within the brains of acquired immune deficiency syndrome patients. *Proc Natl Acad Sci U S A.* 83, (pp. 7089-93). Wyatt, C. M. & P. E. Klotman (2007) HIV-associated nephropathy in the era of antiretroviral

coreceptor usage of non-syncytium-inducing primary HIV-1 is independent of phylogenetically distinct global HIV-1 isolates: delineation of consensus motif in

optimal recognition of Alzheimer's beta amyloid peptide (Abeta1-40). *PLoS One.* 5,

associated antigen. *Proc Nat Acad Sci (Wash.).* 78, (pp. 6485-6488).

viral antigens in fixed tissue sections. *Am J Pathol.* 127, (pp. 199-205).

liposomes. *The Journal of Biological Chemistry.* 277, (pp. 38141-7).

*Proc. Natl. Acad. Sci. USA.* 87, (pp. 7898-7901).

transduction. *J Biol Chem.* 265, (pp. 12059-12066).

ATPase activity *in vitro*. *Biochem.* 42, (pp. 1611-1617).

therapy. *Am J Med.* 120, (pp. 488-92).

implications. *Proc Natl Acad Sci U S A.* 96, (pp. 4558-62).

8412-7).

10, (pp. 1011-4).

(pp. 1289-300).

(pp. e9079).

(2002) Gangliosides are functional nerve cell ligands for myelin-associated glycoprotein (MAG), an inhibitor of nerve regeneration. *Proc Natl Acad Sci.* 99, (pp.

alphavirus membrane fusion. Lack of correlation with lipid raft formation in target

receptor deficient cell lines using the receptor glycolipid globotriosyl ceramide.

type 1 gp120 involved in CCR5 coreceptor utilization: therapeutic and prophylactic

(1987) Immunohistochemical localization of human and simian immunodeficiency

that disrupt binding of HIV-1 gp120 to galactosylceramide. *Bioorg Med Chem Lett.* 

Human immunodeficiency virus (HIV) is a major challenge in the medical fraternity worldwide. According to the UNAIDS report on the global AIDS epidemic, in 2004 it had affected more than 42 million people, and of these 25 million resided in the sub-Sahara Africa (UNAIDS, 2004 & Bhigjee, 2005).

This virus has no cure; the lives of the infected patients can only be prolonged using lifelong highly active antiretroviral (ARV) therapy (HAART). HAART has been proven to suppress HIV-1 viral replication continuously thus reducing mortality and morbidity in treated patients. It has further been proven that HAART is only highly effective if prescribed in combination of more than 2 drugs. However these drug combinations can be presented with potential drug-drug interactions (DDIs) an important cause of adverse drug reactions (ADRs) (Highleyman, 2007).

DDIs are well-recognised causes of adverse drug effects (ADEs) (Bates *et al*., 1995). According to Juurlink *et al*. (2003), DDIs do cause particularly important type of adverse drug event because they are often predictable based on previous reports, clinical studies, and an understanding of pharmacological principles.

According to Johnson *et al*. (1999), DDIs are classified as an important category of ADEs. Drug interactions result in undesirable modification of the action of one or more concurrently administered agents. The interaction may cause treatment failure, an increased pharmacologic effect, or a toxic effect, which may be fatal. Because DDIs usually have a specific time course (i.e., onset and duration), they are more predictable (and preventable) then ADRs (adverse drug reactions). Bates *et al*. (1999) state that Preventable DDIs account for about one third of ADEs but incur about one half of the total ADE costs.

In HIV-infected patients, the introduction of HAART has led to reduced morbidity and mortality in treated patients (Egger *et al.*, 2002). However, in a substantial proportion of patients, the effectiveness of HAART has not been sufficient due to occurrence of virological failure and immunological decay (Bartlett *et al.*, 2001). All this has been due to failure to determine drug interactions and prevention of toxic effects. (Boffito *et al.*, 2005).

The possible causes of DDIs include drug combinations, lack of communication between the prescribers and medical history, increase in the number of newly marketed drugs and polypharmacy. Specific patients who are risk for DDIs include the elderly, people living with HIV/AIDS. Patients with HIV are more at risk for the virus because they are treated

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 283

 **Drug binding**: This was illustrated by in vivo work which showed that methadone concentrations were decreased when administered with ritonavir, due to displacement

 **Drug metabolism:** An example of this kind of interaction is between PIs and NNRTIS that act as inhibitors or inducers of cytochrome P450 (CYP450). Ritonavir is the most potent CYP450 and therefore the most likely to interact with other drugs such as amiodarone, cisapride or pethidine. Likewise efavirenz induces the metabolism of indinavir and

 **Transport system:** One case report demonstrated a 48% decrease in valproic acid concentration after a patient had been started on lopinavir/ritonavir-based regimen. This interaction was likely to be due to the ability of ritonavir to induce valproic acid

The result of pharmacokinetic DDIs may be an increase or decrease in the concentration of

Drugs are metabolised by two types of reactions: phase 1 reactions that involve oxidation, reduction or hydrolysis in which drugs are turned into more polar compounds and phase 11 reactions that involve coupling drugs with some other substance (e.g. glucuronic acid) to make (usually) inactive compounds (Cohen *et al*., 2002). These reactions make drugs more easily excretable. Drug metabolism interactions can increase or decrease the amount of drug

Enzyme induction frequently affects phase 1 oxidation, which requires the presence of the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) and the haemcontaining protein cytochrome P450. Enzyme inducers like carbamazepine, phenytoin, phenobarbital, PIs and NNRTIs, increase the activity of the microsomal enzymes (cytochrome P450 isoenzyme), increasing the rate of metabolism and excretion. One study reported that there was a decreased metabolism and subsequent toxicity of carbamazepine when concomitantly administered with ritonavir (Bates & Herman, 2006; Young, 2005). A case study reported of a patient who was prescribed ritonavir with midazolam concomitantly, and developed extreme sedation and possibly respiratory depression due to

Enzyme inhibitors inhibit the microsomal enzymes (cytochrome P450 isoenzymes), decreasing the rate of metabolism and excretion of other drugs that are metabolised by these same enzymes (Cohen *et al*., 2002). Examples of these drugs are PIs and delavirdine presenting drug interactions with statins because they are metabolised by the same enzyme (Geletko & ZuWallack, 2001). These drugs begin to accumulate in the body and toxicity may

the drug at the site of action. The mechanism most common is drug metabolism.

saquinavir by reducing their plasma concentrations (Piscitelli & Gallicano, 2001). **Excretion:** In this case, the NRTIs may have additive or synergistic adverse effects, so if for example stavudine is administered with zalcitabine or didanosine, because these drugs are eliminated primarily by the kidney peripheral neuropathy caused by

of methadone from plasma binding sites (Piscitelli & Gallicano, 2001).

levels of ciprofloxacin (Sahai *et al.,* 1993).

stavudine (Lee & Henderson, 2001).

**3.2.1 Drug metabolism interactions** 

the inhibition of midazolam metabolism.

**3.2.1.1 Enzyme induction** 

**3.2.1.2 Enzyme inhibition** 

metabolism via glucuronidation (Sheehan *et al*., 2006).

available by inhibition or induction of metabolism (Cohen *et al*., 2002).

ions in the buffer may chelate with ciprofloxacin, resulting in subtherapeutic blood

using HAART which consist of at least three agents with the risk rising from 13% in patients taking two drugs to 82% in those taking seven drugs or more (Sanderson, 2005).

The main focus in this abstract will be on the pharmacological aspects of DDIs between ARVs. The topics to be covered in this chapter will include:


## **2. Concept of DDIs**

The term drug-drug interactions can be defined as *"the pharmacological or clinical response to the administration of a drug combination different from that anticipated from the known effects of the two agents when given alone"* (Tatro, 2009). As described by Tatro (2009), the effect of a DDI may be one of the following:


## **3. Different types of DDIs**

According to Seden *et al*. (2009), DDIs may arise due to the pharmacokinetics or pharmacodynamics of administered compounds. DDIs can be classified as pharmacokinetic or pharmacodynamics (Young, 2005; Cohen *et al*., 2002) or pharmaceutical (Hall, 1986).

## **3.1 Pharmaceutical interactions**

Pharmaceutical interactions occur when two drugs are given together, e.g., in an infusion, or when a drug reacts with the infusion solution. While it is necessary to be aware of this type of interaction, it is relatively uncommon (Hall, 1986).

#### **3.2 Pharmacokinetic interactions**

Pharmacokinetic interactions may be defined as those interactions in which the disposition of the first drug is altered by the second drug or precipitant drug. As a result, the effect of the first drug is either diminished or increased. Pharmacokinetic interactions are divided into those that affect (Swart & Harris, 2005; Young, 2005; Cohen *et al*., 2002):

 **Drug absorption**: An example of this interaction is when didanosine containing an aluminium-magnesium antacid buffer, is administered with ciprofloxacin, the metallic

using HAART which consist of at least three agents with the risk rising from 13% in patients

The main focus in this abstract will be on the pharmacological aspects of DDIs between

The term drug-drug interactions can be defined as *"the pharmacological or clinical response to the administration of a drug combination different from that anticipated from the known effects of the two agents when given alone"* (Tatro, 2009). As described by Tatro (2009), the effect of a DDI

Antagonism, such as a loss of blood pressure control by clonidine when tricyclic

Synergism, as an example of which is the increased anticoagulant effect resulting from

 Idiosyncratic, such as the possible though rare severe effects that have been associated with patients concurrently receiving pethidine and monoamine oxidase inhibitor

According to Seden *et al*. (2009), DDIs may arise due to the pharmacokinetics or pharmacodynamics of administered compounds. DDIs can be classified as pharmacokinetic or pharmacodynamics (Young, 2005; Cohen *et al*., 2002) or pharmaceutical (Hall, 1986).

Pharmaceutical interactions occur when two drugs are given together, e.g., in an infusion, or when a drug reacts with the infusion solution. While it is necessary to be aware of this type

Pharmacokinetic interactions may be defined as those interactions in which the disposition of the first drug is altered by the second drug or precipitant drug. As a result, the effect of the first drug is either diminished or increased. Pharmacokinetic interactions are divided

 **Drug absorption**: An example of this interaction is when didanosine containing an aluminium-magnesium antacid buffer, is administered with ciprofloxacin, the metallic

into those that affect (Swart & Harris, 2005; Young, 2005; Cohen *et al*., 2002):

taking two drugs to 82% in those taking seven drugs or more (Sanderson, 2005).

3. Drug-drug interactions rating systems and their significance levels.

7. The role of pharmacists in preventing DDIs in clinical practice and;

ARVs. The topics to be covered in this chapter will include:

6. The pharmacological aspects of DDIs between ARVs.

antidepressants are added to a regimen.

administering salicylates and warfarin.

of interaction, it is relatively uncommon (Hall, 1986).

8. Recommendations on the clinical management of DDIs.

1. The concept of DDIs. 2. The different types of DDIs.

9. Conclusion.

**2. Concept of DDIs** 

may be one of the following:

(Jankel & Fitterman, 1993).

**3. Different types of DDIs** 

**3.1 Pharmaceutical interactions** 

**3.2 Pharmacokinetic interactions** 

4. The possible causes of DDIs. 5. Patients that are at risk for DDIs. ions in the buffer may chelate with ciprofloxacin, resulting in subtherapeutic blood levels of ciprofloxacin (Sahai *et al.,* 1993).


The result of pharmacokinetic DDIs may be an increase or decrease in the concentration of the drug at the site of action. The mechanism most common is drug metabolism.

## **3.2.1 Drug metabolism interactions**

Drugs are metabolised by two types of reactions: phase 1 reactions that involve oxidation, reduction or hydrolysis in which drugs are turned into more polar compounds and phase 11 reactions that involve coupling drugs with some other substance (e.g. glucuronic acid) to make (usually) inactive compounds (Cohen *et al*., 2002). These reactions make drugs more easily excretable. Drug metabolism interactions can increase or decrease the amount of drug available by inhibition or induction of metabolism (Cohen *et al*., 2002).

#### **3.2.1.1 Enzyme induction**

Enzyme induction frequently affects phase 1 oxidation, which requires the presence of the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) and the haemcontaining protein cytochrome P450. Enzyme inducers like carbamazepine, phenytoin, phenobarbital, PIs and NNRTIs, increase the activity of the microsomal enzymes (cytochrome P450 isoenzyme), increasing the rate of metabolism and excretion. One study reported that there was a decreased metabolism and subsequent toxicity of carbamazepine when concomitantly administered with ritonavir (Bates & Herman, 2006; Young, 2005). A case study reported of a patient who was prescribed ritonavir with midazolam concomitantly, and developed extreme sedation and possibly respiratory depression due to the inhibition of midazolam metabolism.

#### **3.2.1.2 Enzyme inhibition**

Enzyme inhibitors inhibit the microsomal enzymes (cytochrome P450 isoenzymes), decreasing the rate of metabolism and excretion of other drugs that are metabolised by these same enzymes (Cohen *et al*., 2002). Examples of these drugs are PIs and delavirdine presenting drug interactions with statins because they are metabolised by the same enzyme (Geletko & ZuWallack, 2001). These drugs begin to accumulate in the body and toxicity may

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 285

(Sahai *et al*., 1993). It is therefore recommended that the two drugs be administered at different times. Absorption of many drugs, such as delavirdine, atazanavir, aspirin, ciprofloxacin, and digoxin, can be significantly impaired by concurrent administration of

A mechanism of DDIs that may go unrecognised is where one drug changes the gastrointestinal transit time. In doing so, the pharmacokinetics of not altering the transit time can be changed, leading to changes in the drug's pharmacological actions. Drugs with anticholinergic properties and opioids will slow gastrointestinal motility, while drugs, such as metoclopramide and laxatives will increase gastric emptying and gastric transit and

Drugs that change the normal pH of the stomach can affect absorption characteristics of other drugs. This is an essentially important point, considering the widespread use of proton pump inhibitors, although only a few clinically relevant interactions have been identified (e.g. ketoconazole) (Delafuente, 2003). The results of data in a study done by O'Connor-Semmes *et al*. (2001) suggested that the elderly may be more sensitive to the

According to Piscitell and Gallicano (2001), when didanosine is administered with indinavir, changes in pH may significantly alter drug absorption of indinivarir because of an increase in pH due to the didanosine buffer. It is therefore recommended that didanosine and

It has been reported by Fulco *et al*. (2006) and Tran *et al*. (2001) that acid-suppressive therapy with histamine-2 (H2) blockers, proton pump inhibitors or antacids can cause a decrease in the absorption of some PIs. This is due to changes in the pH of the gastrointestinal tract. PIs like atazanavir, fosamprenavir, tipranavir have been found to have significant interactions with acid-suppressive therapy that require intervention due to the potential for virological

The most common DDI affecting drug distribution is alteration in protein binding. This type of interaction occurs when there is competitive inhibition for protein binding sites. This allows for the unbound fraction of the drugs to be increased, and it is the free fraction that is responsible for pharmacological activity (Young, 2005). Most of the clinically significant interactions involve drugs that are highly protein bound and have a narrow therapeutic index. An example of this is when the cytidine analogue lamivudine inhibits phosphorylation of another cytidine analog, zalcitabine, resulting in high incidence of

An example of this DDI is when zidovudine and stavudine are co-administered, the two NRTIs do compete for cellular thymidine kinase, the enzyme that is responsible for the monophosphorylation of both drugs to nucleotides. The inhibitory effect impairs the

Most of the clinically important types of pharmacokinetic DDIs are those altering a drug's metabolism. Many elderly patients, but not all, have underlying impaired CYP450

antacids by a variety of mechanisms (Fulco *et al,* 2006).

generally increase the rate of absorption (Benet *et al*., 1990).

increase in gastric pH compared to younger adults.

indinavir be administered at least one hour apart.

failure from inadequate ARV concentration (Fulco *et al*., 2006).

toxicities. Therefore such combinations should be avoided (Young, 2005).

efficacy of stavudine when combined with zidovudine (Havril *et al*., 2000).

**3.3.1.2 Gastrointestinal motility** 

**3.3.1.3 Altered gastric pH** 

**3.3.1.4 Altered distribution** 

**3.3.1.5 Altered metabolism** 

*3.3.1.5.1 Cytochrome P450 isoenzyme* 

develop within 2 to 3 days. The clinical significance of the enzyme inhibition interaction depends on the extent to which serum levels rise. Some drugs may have additive or synergistic adverse effects. For example, zidovudine may cause anaemia and neutropnenia, so drugs causing bone marrow suppression should be prescribed with caution if used concomitantly (Matheny *et al*., 2001). Another example of this metabolism was the administration of simvastatin with saquinavir/ritonavir, the interaction leading to increased levels of simvastatin by more than 3000% (Fichenbaum *et al*., 2002b). This could put the patient at risk for adverse effects like myalgias, rhabdomyolysis, elevated creatinine phosphokinase and hepatic dysfunction (Dube *et al*., 2003).

## **3.2.2 Cytochrome P450 isoenzymes**

Cytochrome P450 is a large family of related isoenzymes of which about 30 have been identified. The most frequently involved in drug interactions are CYP3A4 and CYP2D6. There are many drugs that are metabolised by these cytochrome P450 isoenzymes including ARVs (Clarke *et al*., 2008). Drugs may be metabolised by more than one cytochrome isoenzyme. For example, the majority of PIs and NNRTIs and antidepressants are substrates for, and can inhibit or induce the CYP450 system and have the potential to cause clinical drug interactions including serotonin syndrome, a potential fatal complication. According to Swart and Harris (2005) it is of value to know which particular isoenzymes are responsible for the metabolism of a specific drug as this makes it possible to predict with which other drugs it may possibly interact.

#### **3.3 Pharmacodynamic interactions**

Pharmacodynamic interactions are those where the effects of one drug are changed by the presence of another drug at its site of action, without alterations in the concentrations of either drug (Young, 2005; Cohen *et al*., 2002). Sometimes one drug competes directly with another for particular receptors, but often the reaction is more indirect and involves the interference with physiological mechanisms, making pharmacodynamic interactions more difficult to classify than pharmacokinetic interactions (DeVane as quoted by Delafuente, 2003). There are four basic subdivisions as quoted by Swart and Harris (2005):


#### **3.3.1 Mechanisms of drug-drug interactions**

Drugs interact with one another through various mechanisms which include altered absorption, altered distribution, altered metabolism and altered elimination.

#### **3.3.1.1 Altered absorption**

Drug interactions can occur where one drug changes the absorption characteristics of another drug. The binding of one drug to another causes changes in gastric pH, and changes in gastrointestinal motility and can cause these drug interactions (Cohen *et al*., 2002). One example is that of didanosine which contains an aluminium-magnesium antacid buffer. When administering didanosine with ciprofloxacin, the metallic ions in the buffer may chelate concomitantly with ciprofloxacin resulting in subtherapeutic levels of the antibiotic (Sahai *et al*., 1993). It is therefore recommended that the two drugs be administered at different times. Absorption of many drugs, such as delavirdine, atazanavir, aspirin, ciprofloxacin, and digoxin, can be significantly impaired by concurrent administration of antacids by a variety of mechanisms (Fulco *et al,* 2006).

## **3.3.1.2 Gastrointestinal motility**

284 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

develop within 2 to 3 days. The clinical significance of the enzyme inhibition interaction depends on the extent to which serum levels rise. Some drugs may have additive or synergistic adverse effects. For example, zidovudine may cause anaemia and neutropnenia, so drugs causing bone marrow suppression should be prescribed with caution if used concomitantly (Matheny *et al*., 2001). Another example of this metabolism was the administration of simvastatin with saquinavir/ritonavir, the interaction leading to increased levels of simvastatin by more than 3000% (Fichenbaum *et al*., 2002b). This could put the patient at risk for adverse effects like myalgias, rhabdomyolysis, elevated creatinine

Cytochrome P450 is a large family of related isoenzymes of which about 30 have been identified. The most frequently involved in drug interactions are CYP3A4 and CYP2D6. There are many drugs that are metabolised by these cytochrome P450 isoenzymes including ARVs (Clarke *et al*., 2008). Drugs may be metabolised by more than one cytochrome isoenzyme. For example, the majority of PIs and NNRTIs and antidepressants are substrates for, and can inhibit or induce the CYP450 system and have the potential to cause clinical drug interactions including serotonin syndrome, a potential fatal complication. According to Swart and Harris (2005) it is of value to know which particular isoenzymes are responsible for the metabolism of a specific drug as this makes it possible to predict with which other

Pharmacodynamic interactions are those where the effects of one drug are changed by the presence of another drug at its site of action, without alterations in the concentrations of either drug (Young, 2005; Cohen *et al*., 2002). Sometimes one drug competes directly with another for particular receptors, but often the reaction is more indirect and involves the interference with physiological mechanisms, making pharmacodynamic interactions more difficult to classify than pharmacokinetic interactions (DeVane as quoted by Delafuente,

Drugs interact with one another through various mechanisms which include altered

Drug interactions can occur where one drug changes the absorption characteristics of another drug. The binding of one drug to another causes changes in gastric pH, and changes in gastrointestinal motility and can cause these drug interactions (Cohen *et al*., 2002). One example is that of didanosine which contains an aluminium-magnesium antacid buffer. When administering didanosine with ciprofloxacin, the metallic ions in the buffer may chelate concomitantly with ciprofloxacin resulting in subtherapeutic levels of the antibiotic

2003). There are four basic subdivisions as quoted by Swart and Harris (2005):

absorption, altered distribution, altered metabolism and altered elimination.

Additive or synergistic interactions and combined toxicity

 Interactions due to changes in drug transport mechanisms Interactions due to disturbances in fluid and electrolyte

phosphokinase and hepatic dysfunction (Dube *et al*., 2003).

**3.2.2 Cytochrome P450 isoenzymes** 

drugs it may possibly interact.

**3.3 Pharmacodynamic interactions** 

Antagonistic or opposing interactions

**3.3.1 Mechanisms of drug-drug interactions** 

**3.3.1.1 Altered absorption** 

A mechanism of DDIs that may go unrecognised is where one drug changes the gastrointestinal transit time. In doing so, the pharmacokinetics of not altering the transit time can be changed, leading to changes in the drug's pharmacological actions. Drugs with anticholinergic properties and opioids will slow gastrointestinal motility, while drugs, such as metoclopramide and laxatives will increase gastric emptying and gastric transit and generally increase the rate of absorption (Benet *et al*., 1990).

#### **3.3.1.3 Altered gastric pH**

Drugs that change the normal pH of the stomach can affect absorption characteristics of other drugs. This is an essentially important point, considering the widespread use of proton pump inhibitors, although only a few clinically relevant interactions have been identified (e.g. ketoconazole) (Delafuente, 2003). The results of data in a study done by O'Connor-Semmes *et al*. (2001) suggested that the elderly may be more sensitive to the increase in gastric pH compared to younger adults.

According to Piscitell and Gallicano (2001), when didanosine is administered with indinavir, changes in pH may significantly alter drug absorption of indinivarir because of an increase in pH due to the didanosine buffer. It is therefore recommended that didanosine and indinavir be administered at least one hour apart.

It has been reported by Fulco *et al*. (2006) and Tran *et al*. (2001) that acid-suppressive therapy with histamine-2 (H2) blockers, proton pump inhibitors or antacids can cause a decrease in the absorption of some PIs. This is due to changes in the pH of the gastrointestinal tract. PIs like atazanavir, fosamprenavir, tipranavir have been found to have significant interactions with acid-suppressive therapy that require intervention due to the potential for virological failure from inadequate ARV concentration (Fulco *et al*., 2006).

### **3.3.1.4 Altered distribution**

The most common DDI affecting drug distribution is alteration in protein binding. This type of interaction occurs when there is competitive inhibition for protein binding sites. This allows for the unbound fraction of the drugs to be increased, and it is the free fraction that is responsible for pharmacological activity (Young, 2005). Most of the clinically significant interactions involve drugs that are highly protein bound and have a narrow therapeutic index. An example of this is when the cytidine analogue lamivudine inhibits phosphorylation of another cytidine analog, zalcitabine, resulting in high incidence of toxicities. Therefore such combinations should be avoided (Young, 2005).

An example of this DDI is when zidovudine and stavudine are co-administered, the two NRTIs do compete for cellular thymidine kinase, the enzyme that is responsible for the monophosphorylation of both drugs to nucleotides. The inhibitory effect impairs the efficacy of stavudine when combined with zidovudine (Havril *et al*., 2000).

#### **3.3.1.5 Altered metabolism**

#### *3.3.1.5.1 Cytochrome P450 isoenzyme*

Most of the clinically important types of pharmacokinetic DDIs are those altering a drug's metabolism. Many elderly patients, but not all, have underlying impaired CYP450

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 287

Many drugs and drug metabolites are excreted in the urine via renal tubular secretion. Two drugs can compete for the same active secretion sites in the tubule allowing for decreased elimination and potentially toxic serum concentrations (Lesho & Gey, 2003). Alteration in urine pH can also affect drug elimination. Alkalinisation of the urine will decrease elimination of drugs that are weak bases and decreases in urine pH will increase their elimination. Acidification of the urine will decrease renal elimination of drugs that are weak

This mechanism happens in interactions that alter drug bioavailability by decreasing it and these are commonly found in PIs. The reason is that PIs induce CYP450 isoenzymes, so drugs like phenytoin, rifampin, carbamazepine, phenobarbitone, and dexamethasone can increase the hepatic clearance, thereby decreasing plasma concentrations of the PIs (Lesho & Gey, 2003). All this result in increase in toxicity of the drugs (Lesho & Gey, 2003). In the elderly as stated by Delafuente (2003), more common and potentially more significant are DDIs that affect renal function. Glomerular filtration rates decline with advanced aging. To compensate for this physiologic change, a compensatory production of vasodilatory renal prostaglandins occurs (Delafuente, 2001). However, according to Swedko *et al*. (2003) in frail elderly patients, serum creatinine concentrations may be very misleading, often in the

Most rating systems as employed by Tatro (2009) and De Maat *et al*. (2004) indicate:

Significance level 1 indicates a major contradiction or a drug interaction that requires very careful monitoring. According to Strain *et al*. (2002b) the effects are potentially lifethreatening or capable of causing permanent damage. The clinician needs to document why he or she is prescribing this combination, and the medical necessity to use both drugs concomitantly only if there is no alternative or the potential benefit outweighs the risk. Drug combinations producing an interaction with a significance level 1 are combinations that result in serious and potentially life-threatening adverse effects such as arrhythmia,

Obviously, if this combination is to be used the drug(s) in question must be prescribed with an explanation as to the need for their concomitant use and must be preceded by very cautious monitoring. Documentation of the clinician's awareness of the potential serious – level 1 – interaction should be accomplished at the time of prescribing this potentially dangerous combination. In addition, it is obligatory to alert the other health care providers' of the potential interactions and adverse outcomes which they could expect. Obviously, the optimum choice, if possible, is to use an alternative medication to avoid significance level 1

With significance level 2, the effects may cause deterioration in a patient's status. Additional treatment, hospitalisation or extension of hospital stay may be necessary (Strain *et al*.,

*3.3.1.5.4 Altered renal elimination* 

normal range despite poor renal function.

**4. DDIs rating system: significance levels** 

respiratory depression and/or death (Winston & Boffito, 2005).

acids (Hasten, 1995).

major significance,

minor significance.

moderate significance; and

**4.1 Major significance: level 1** 

interactions (Strain *et al*., 2002a).

**4.2 Moderate significance: level 2** 

metabolising capability. According to Flockhart and Tanus-Santos (2002), six CYP450 isoenzymes that have been identified to be involved in oxidative metabolism of most commonly used drugs are: CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4.

Interactions involving the CYP450 enzymes are often due to either inhibition of an isoenzyme, leading to increased blood or tissue concentrations of the substrate, or induction of an isoenzyme, causing enhanced metabolism and lower substrate concentrations (Delafuente, 2003:137). According to Johnson *et al.* (1999) enzyme inhibition is the mechanism most often responsible for life-threatening interactions.32 Such interactions have been observed when zalcitabine is combined with stavudine or didanosine producing severe peripheral neuropathy, pancreatitis, and lactic acidosis (Simpson & Tagliati, 1995).

Induction of certain CYP450 isoenzymes, for example CYP2C9/19 by lopinavir/ritonavir and nelfinavir was reported by Honda *et al.* (1999) that it could lead to an increase in the matabolism of antiepileptic drugs like phenytoin, a norrow therapeutic index drug. The reduction in the anticonvulsant serum concentration could lead to seizures.

### *3.3.1.5.2 Cytochrome P450 inhibition*

Competitive binding at the enzyme's binding site between two drugs is often responsible for inhibition of a drug's metabolism. The onset of CYP450 inhibition depends on the inhibiting drug's half-life. For drugs with short half-lives, enzyme inhibition occurs quickly and clinically significant interactions can be apparent within 1 or 2 days (Cheng *et al*., 2009). Inhibition of CYP450 is also dose-dependent. Higher doses of an inhibitory drug will cause greater amounts of competitive inhibition than lower doses. Although sufficient data are not available to help in clinical situations as stated by Delafuente (2003), knowing the CYP450 enzymes involved in a drug's metabolism can be used to predict and avoid clinical problems resulting from drug interactions.

All currently marketed PIs – atazanavir, amprenavir, fosamprenavir, indinavir, lopinavir, nelfinavir, ritonavir and saquinavir (Young, 2005) – and the NNRTI delavirdine inhibit CYP3A4 (Piscitelli & Gallicano, 2001) decrease the hepatic clearance of CYP3A4 substrates and increase their plasma levels.

#### *3.3.1.5.3 Cytochrome P450 induction*

The onset of enzyme induction is usually longer than that of enzyme inhibition (Chapron, 2001). Enzyme induction is dependent on the half-life of the synthesis of new CYP450 isoenzymes and is dependent on the half-life of the inducing drug. Like inhibition of CYP450 enzymes, shorter half-life drugs will have a shorter onset of induction. A drug with half-life, such as phenobarbital, may take one week before enzyme induction is seen.

Drugs often involved in induction of CYP450 isoenzymes are carbamazepine, phenytoin, phenobarbital, primidone, and rifampicin (Clarke *et al*., 2008). Aging may impair enzyme induction, but this is not a universal finding as stated by Chapron (2001).

Of the ARVs, the NNRTIs nevirapine and efavirenz induce CYP3A4, thus increasing the hepatic clearance of CYP3A4 substrates and decreasing their plasma levels (Piscitelli & Gallicano, 2001). Other ARVs like PIs induce CYP450 isoenzymes, and it has been reported that drugs like phenytoin, rifampin, carbamazepine, phenobarbital, and dexamethasone can increase the hepatic clearance and therefore decrease plasma concentrations of the PIs (Lesho & Gey, 2003).

### *3.3.1.5.4 Altered renal elimination*

286 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

metabolising capability. According to Flockhart and Tanus-Santos (2002), six CYP450 isoenzymes that have been identified to be involved in oxidative metabolism of most commonly used drugs are: CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4. Interactions involving the CYP450 enzymes are often due to either inhibition of an isoenzyme, leading to increased blood or tissue concentrations of the substrate, or induction of an isoenzyme, causing enhanced metabolism and lower substrate concentrations (Delafuente, 2003:137). According to Johnson *et al.* (1999) enzyme inhibition is the mechanism most often responsible for life-threatening interactions.32 Such interactions have been observed when zalcitabine is combined with stavudine or didanosine producing severe peripheral neuropathy, pancreatitis, and lactic acidosis

Induction of certain CYP450 isoenzymes, for example CYP2C9/19 by lopinavir/ritonavir and nelfinavir was reported by Honda *et al.* (1999) that it could lead to an increase in the matabolism of antiepileptic drugs like phenytoin, a norrow therapeutic index drug. The

Competitive binding at the enzyme's binding site between two drugs is often responsible for inhibition of a drug's metabolism. The onset of CYP450 inhibition depends on the inhibiting drug's half-life. For drugs with short half-lives, enzyme inhibition occurs quickly and clinically significant interactions can be apparent within 1 or 2 days (Cheng *et al*., 2009). Inhibition of CYP450 is also dose-dependent. Higher doses of an inhibitory drug will cause greater amounts of competitive inhibition than lower doses. Although sufficient data are not available to help in clinical situations as stated by Delafuente (2003), knowing the CYP450 enzymes involved in a drug's metabolism can be used to predict and avoid clinical

All currently marketed PIs – atazanavir, amprenavir, fosamprenavir, indinavir, lopinavir, nelfinavir, ritonavir and saquinavir (Young, 2005) – and the NNRTI delavirdine inhibit CYP3A4 (Piscitelli & Gallicano, 2001) decrease the hepatic clearance of CYP3A4 substrates

The onset of enzyme induction is usually longer than that of enzyme inhibition (Chapron, 2001). Enzyme induction is dependent on the half-life of the synthesis of new CYP450 isoenzymes and is dependent on the half-life of the inducing drug. Like inhibition of CYP450 enzymes, shorter half-life drugs will have a shorter onset of induction. A drug with

Drugs often involved in induction of CYP450 isoenzymes are carbamazepine, phenytoin, phenobarbital, primidone, and rifampicin (Clarke *et al*., 2008). Aging may impair enzyme

Of the ARVs, the NNRTIs nevirapine and efavirenz induce CYP3A4, thus increasing the hepatic clearance of CYP3A4 substrates and decreasing their plasma levels (Piscitelli & Gallicano, 2001). Other ARVs like PIs induce CYP450 isoenzymes, and it has been reported that drugs like phenytoin, rifampin, carbamazepine, phenobarbital, and dexamethasone can increase the hepatic clearance and therefore decrease plasma concentrations of the PIs

half-life, such as phenobarbital, may take one week before enzyme induction is seen.

induction, but this is not a universal finding as stated by Chapron (2001).

reduction in the anticonvulsant serum concentration could lead to seizures.

(Simpson & Tagliati, 1995).

*3.3.1.5.2 Cytochrome P450 inhibition* 

problems resulting from drug interactions.

and increase their plasma levels. *3.3.1.5.3 Cytochrome P450 induction* 

(Lesho & Gey, 2003).

Many drugs and drug metabolites are excreted in the urine via renal tubular secretion. Two drugs can compete for the same active secretion sites in the tubule allowing for decreased elimination and potentially toxic serum concentrations (Lesho & Gey, 2003). Alteration in urine pH can also affect drug elimination. Alkalinisation of the urine will decrease elimination of drugs that are weak bases and decreases in urine pH will increase their elimination. Acidification of the urine will decrease renal elimination of drugs that are weak acids (Hasten, 1995).

This mechanism happens in interactions that alter drug bioavailability by decreasing it and these are commonly found in PIs. The reason is that PIs induce CYP450 isoenzymes, so drugs like phenytoin, rifampin, carbamazepine, phenobarbitone, and dexamethasone can increase the hepatic clearance, thereby decreasing plasma concentrations of the PIs (Lesho & Gey, 2003). All this result in increase in toxicity of the drugs (Lesho & Gey, 2003). In the elderly as stated by Delafuente (2003), more common and potentially more significant are DDIs that affect renal function. Glomerular filtration rates decline with advanced aging. To compensate for this physiologic change, a compensatory production of vasodilatory renal prostaglandins occurs (Delafuente, 2001). However, according to Swedko *et al*. (2003) in frail elderly patients, serum creatinine concentrations may be very misleading, often in the normal range despite poor renal function.

## **4. DDIs rating system: significance levels**

Most rating systems as employed by Tatro (2009) and De Maat *et al*. (2004) indicate:


## **4.1 Major significance: level 1**

Significance level 1 indicates a major contradiction or a drug interaction that requires very careful monitoring. According to Strain *et al*. (2002b) the effects are potentially lifethreatening or capable of causing permanent damage. The clinician needs to document why he or she is prescribing this combination, and the medical necessity to use both drugs concomitantly only if there is no alternative or the potential benefit outweighs the risk. Drug combinations producing an interaction with a significance level 1 are combinations that result in serious and potentially life-threatening adverse effects such as arrhythmia, respiratory depression and/or death (Winston & Boffito, 2005).

Obviously, if this combination is to be used the drug(s) in question must be prescribed with an explanation as to the need for their concomitant use and must be preceded by very cautious monitoring. Documentation of the clinician's awareness of the potential serious – level 1 – interaction should be accomplished at the time of prescribing this potentially dangerous combination. In addition, it is obligatory to alert the other health care providers' of the potential interactions and adverse outcomes which they could expect. Obviously, the optimum choice, if possible, is to use an alternative medication to avoid significance level 1 interactions (Strain *et al*., 2002a).

### **4.2 Moderate significance: level 2**

With significance level 2, the effects may cause deterioration in a patient's status. Additional treatment, hospitalisation or extension of hospital stay may be necessary (Strain *et al*.,

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 289

Emergency department physicians do not routinely screen for potential drug interactions due to unavailability of a medication history. In a study by Beers *et al*. (1990) it was stated that groups of patients at higher risk of drug complications, the elderly and those taking multiple medications, did not appear to receive more cautious care. Neither the physician's record nor the instructions given to the patient indicated that prescribing physicians recognised the potential adverse reactions that were introduced. There is need for physians to screen for interactions. A patient's advanced age or a long list of medications should cause the physicians to be more reticent in prescribing. Fewer medications should be given

There is a considerable number of newly marketed drugs with a growing number of possible combinations. Complex disease states often require the concurrent use of these drug combination therapies so as to be highly effective (Bergk *et al*., 2004). Nevertheless, as supported by Merlo *et al*. (2001) multiple drug use is also associated with the occurrence of DDIs. Therefore the majority of these interactions can be compensated by dose adjustment or prevented by a well-considered sequence of administration (Bergk *et al*., 2004). The considerable number of newly marketed drugs with a number of possible combinations raises the need to support general practitioners with the pertinent information for careful

Polypharmacy, the use of two or more medications by one patient, has become prevalent especially in elderly patients (Gaeta *et al*., 2002). Beers *et al*. (1990) in their study showed that those 65 years of age and older used an average of two to six prescribed medications and one to four non-prescribed medications per day. The frequency of polypharmacy in the elderly increases the incidence of adverse drug reactions and interactions, and it is the most significant contributing factor for DDIs. Patient's past medical history and medication has to be evaluated by the physicians. According to Seden *et al*. (2009), polypharmacy is largely unavoidable for patients receiving ARVs in both the developed and developing world and resource-poor

Patients that are at risk for DDIs are discussed in this section with specific reference to the

The incidence of adverse drug reactions and interactions in the elderly has been reported to be two to three times the incidence in younger patients (Nolan & O'Malley, 1998). According to Sloan (1992), this increased risk for the elderly may be related to impaired organ reserve capacity, multiorgan dysfunction, and altered pharmacokinetics and

The HIV infection is treated by using HAART, which involves a regimen of at least three agents to be effective (Seden *et al*., 2009). In a study on DDIs in general medical patients,

setting, with life-long treatment and change of drug combinations along the way.

to the elderly and to high medication users.

approach to patients.

**5.4 Polypharmacy** 

**6. Patients at risk for DDIs** 

**6.1 The elderly** 

pharmacodynamics.

elderly and the HIV/AIDS patients.

**6.2 People living with HIV/AIDS** 

**5.3 Increase in number of newly marketed drugs** 

2002b). The potential interaction must also be documented and the clinical outcome(s) must be monitored carefully so that unacceptable, pernicious reactions are halted as soon as possible. According to Strain *et al*. (2002b) it is essential that the clinician document that the potential drug interactions were considered when using this combination. It is also essential to alert the patient's health care providers to the potential interactions so that they are observed early in their course.

## **4.3 Minor significance: level 3**

As stated by Strain *et al*. (2002b) the effects are usually mild. Consequences may be bothersome or unnoticeable, but should not significantly affect the therapeutic outcome. Additional treatment is usually not required (Tatro, 2005 & Sewester, 2001). According to Strain *et al*. (2002a) significance level 3 does not preclude the use of a specific drug, but clinical decision making requires acknowledging if the adverse reactions (e.g. nausea and rash) might be precluded by choosing an alternative drug. The potential interaction and its mechanism(s) needs documentation in the patient's medical chart and the patient's health care providers need to be informed.

Another rating system is employed by *Drug Interaction Facts* which utilises 5 point significance classification scheme (Tatro, 2009) and *Facts and Comparisons* (McEvoy, 2000) and they recommended the following:


## **5. Possible causes of DDIs**

## **5.1 Drug combinations**

Drug combinations of interacting drugs are among the major causes leading to DDIs (Seden *et al*., 2009). Drug combinations are more common in an elderly population using many drugs (Bjőrkman *et al,*. 2002). A large proportion of these combinations are likely to be part of a normal drug regimen. In a study done by Bjőrkman *et al*. (2002), in DDIs most of the drug combinations increased the risk of ADRs and lowered therapeutic effects as stated by Seymour and Routledge (1998). In all potential DDIs, 50% of the combinations could result in an adverse drug reaction and 50% in a suboptimal therapeutic effect. However, combination ARV treatment is a potent and effective therapy for HIV infection (Pontali, 2008). This is also a disadvantage because ARV drugs frequently interact amongst themselves and other drugs as was identified by KatendeKyenda et al (2007). Since some of these drug combinations have negative effects, more attention must be focused on detecting and monitoring patients using such combinations and could also be addressed by dose adjustment.

### **5.2 Lack of communication and medication history**

Communication between emergency departments and primary care physicians often does not occur (Beers *et al*., 1990), and primary care physicians do not take down medication histories optimally and therefore, the physicians responsible for follow-up may be unaware of the changes made in therapy.

Emergency department physicians do not routinely screen for potential drug interactions due to unavailability of a medication history. In a study by Beers *et al*. (1990) it was stated that groups of patients at higher risk of drug complications, the elderly and those taking multiple medications, did not appear to receive more cautious care. Neither the physician's record nor the instructions given to the patient indicated that prescribing physicians recognised the potential adverse reactions that were introduced. There is need for physians to screen for interactions. A patient's advanced age or a long list of medications should cause the physicians to be more reticent in prescribing. Fewer medications should be given to the elderly and to high medication users.

## **5.3 Increase in number of newly marketed drugs**

There is a considerable number of newly marketed drugs with a growing number of possible combinations. Complex disease states often require the concurrent use of these drug combination therapies so as to be highly effective (Bergk *et al*., 2004). Nevertheless, as supported by Merlo *et al*. (2001) multiple drug use is also associated with the occurrence of DDIs. Therefore the majority of these interactions can be compensated by dose adjustment or prevented by a well-considered sequence of administration (Bergk *et al*., 2004). The considerable number of newly marketed drugs with a number of possible combinations raises the need to support general practitioners with the pertinent information for careful approach to patients.

#### **5.4 Polypharmacy**

288 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

2002b). The potential interaction must also be documented and the clinical outcome(s) must be monitored carefully so that unacceptable, pernicious reactions are halted as soon as possible. According to Strain *et al*. (2002b) it is essential that the clinician document that the potential drug interactions were considered when using this combination. It is also essential to alert the patient's health care providers to the potential interactions so that they are

As stated by Strain *et al*. (2002b) the effects are usually mild. Consequences may be bothersome or unnoticeable, but should not significantly affect the therapeutic outcome. Additional treatment is usually not required (Tatro, 2005 & Sewester, 2001). According to Strain *et al*. (2002a) significance level 3 does not preclude the use of a specific drug, but clinical decision making requires acknowledging if the adverse reactions (e.g. nausea and rash) might be precluded by choosing an alternative drug. The potential interaction and its mechanism(s) needs documentation in the patient's medical chart and the patient's health

Another rating system is employed by *Drug Interaction Facts* which utilises 5 point significance classification scheme (Tatro, 2009) and *Facts and Comparisons* (McEvoy, 2000)

Drug combinations of interacting drugs are among the major causes leading to DDIs (Seden *et al*., 2009). Drug combinations are more common in an elderly population using many drugs (Bjőrkman *et al,*. 2002). A large proportion of these combinations are likely to be part of a normal drug regimen. In a study done by Bjőrkman *et al*. (2002), in DDIs most of the drug combinations increased the risk of ADRs and lowered therapeutic effects as stated by Seymour and Routledge (1998). In all potential DDIs, 50% of the combinations could result in an adverse drug reaction and 50% in a suboptimal therapeutic effect. However, combination ARV treatment is a potent and effective therapy for HIV infection (Pontali, 2008). This is also a disadvantage because ARV drugs frequently interact amongst themselves and other drugs as was identified by KatendeKyenda et al (2007). Since some of these drug combinations have negative effects, more attention must be focused on detecting and monitoring patients using such combinations and could also be addressed by dose

Communication between emergency departments and primary care physicians often does not occur (Beers *et al*., 1990), and primary care physicians do not take down medication histories optimally and therefore, the physicians responsible for follow-up may be unaware

Usually avoid combination: use combination only under special circumstances.

observed early in their course.

**4.3 Minor significance: level 3** 

care providers need to be informed.

and they recommended the following:

**5. Possible causes of DDIs** 

of the changes made in therapy.

**5.1 Drug combinations** 

adjustment.

Avoid combination: risk always outweighs benefit.

No interaction: evidence suggests no interaction.

**5.2 Lack of communication and medication history** 

 Minimise risk: take action as necessary to reduce risk. No action needed: risk of adverse outcomes appears small.

Polypharmacy, the use of two or more medications by one patient, has become prevalent especially in elderly patients (Gaeta *et al*., 2002). Beers *et al*. (1990) in their study showed that those 65 years of age and older used an average of two to six prescribed medications and one to four non-prescribed medications per day. The frequency of polypharmacy in the elderly increases the incidence of adverse drug reactions and interactions, and it is the most significant contributing factor for DDIs. Patient's past medical history and medication has to be evaluated by the physicians. According to Seden *et al*. (2009), polypharmacy is largely unavoidable for patients receiving ARVs in both the developed and developing world and resource-poor setting, with life-long treatment and change of drug combinations along the way.

## **6. Patients at risk for DDIs**

Patients that are at risk for DDIs are discussed in this section with specific reference to the elderly and the HIV/AIDS patients.

#### **6.1 The elderly**

The incidence of adverse drug reactions and interactions in the elderly has been reported to be two to three times the incidence in younger patients (Nolan & O'Malley, 1998). According to Sloan (1992), this increased risk for the elderly may be related to impaired organ reserve capacity, multiorgan dysfunction, and altered pharmacokinetics and pharmacodynamics.

#### **6.2 People living with HIV/AIDS**

The HIV infection is treated by using HAART, which involves a regimen of at least three agents to be effective (Seden *et al*., 2009). In a study on DDIs in general medical patients,

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 291

Drug interactions associated with NRTIs and NtRTIs are few because these drugs are not metabolised by the CYP450 system (Clarke *et al.,* 2008). However, drug interactions may still occur within these drugs as was demonstrated by Katende-Kyenda et al. (2008a). One of the few pharmacodynamic interactions encountered in HIV medicine occurs, for example with co-administered zidovudine and stavudine, since both drugs are thymidine analogues and they can compete for the same phosphorylation site in the growing chain of HIV DNA, resulting in an antagonistic, pharmacodynamic interaction (Piscitelli & Gallicano, 2001). It is

The use of didanosine (ddl) is complicated by drug interactions (Cohen *et al*, 2002). It is a buffered tablet form containing magnesium and calcium to improve systemic absorption. It, however, interacts with certain antibiotics like ciprofloxacin, tetracycline and therefore, to minimise the interaction, didanosine should be administered at least two hours after or six hours before the fluoroquinolone (Knupp & Barbhaiya, 1999). Concurrent use of didanosinebuffered tablets may also impair the absorption of the PI atazanavir, since atazanavir requires an acidic environment for absorption (Product Information Videx EC, 2003). To minimise the interaction, patients should take a didanosine-buffered tablet two hours after

The most significant didanosine drug interaction reported occurs when didanosine is used concurrently with the NrRTI tenofovir. The didanosine AUC increases by 60% and therefore it is recommended that in patients receiving these two drugs concurrently and weighing > 60 kg, the didanosine dosage should be reduced from 400 mg to 250 mg once daily or from 250 mg to 200 mg in patients who weigh less than 60 kg (Young, 2005). For severely underweight patients, the dose should be further reduced to 125mg once daily (Faragon & Piliero, 2004). All patients receiving concurrent tenofovir and didanosine should be closely monitored for didanosine -related toxicities such as pancreatitis, hyperlactatemia, and lactic

Drugs in this group are prone to drug interactions because they are extensively metabolised via CYP3A4 and can act as either inducers or inhibitors of CYP3A4. Nevirapine and efavirenz are inducers of CYP3A4, while delavirdine is an inhibitor of CYP3A4 (Pfister *et al*., 2003). Therefore, when one of these drugs is combined with a drug that is also metabolised

Nevirapine presents with numerous drug interactions, being a CYP3A4 inducer, and drug interactions associated with it lead to an increase in metabolism and reduced concentration of the co-administered drug. For example, when nevirapine is concurrently given with methadone, withdrawal symptoms may occur as a result of reduced methadone levels (Pinzanni *et al*., 2000). Efavirenz is a potent inducer of CYP3A4 in vivo. Like the PIs, EFV is

The induction properties of efavirenz can result in reduced concentrations of concurrently administered drugs that are metabolised by CYP3A4 and it is therefore contraindicated with midazolam, triazolam and ergotamine derivative since there is a potential for increased drug concentrations of these medications and associated toxicity (Product Infomation Sustiva, 2003). Efavirenz, as a potent inducer of CYP3A4 is suggested to have a potential interaction with lopinavir and ritonavir, both of which inhibit CYP3A4. This interaction was

therefore recommended that these two drugs never to be combined.

or one hour before taking atazanavir.

acidosis, regardless of didanosine dosage adjustments.

**7.4 Non-Nucleoside Reverse Transciptase Inhibitors (NNRTIs)** 

by CYP3A4, a drug interaction may occur (Clarke *et al*., 2008).

extensively metabolised primarily by the CYP3A4 (Pfister *et al*., 2003).

Sanderson (2005) found that the risk of DDIs rose from 13% in patients taking two drugs to 82% in patients taking seven drugs or more.

## **7. Pharmacological aspects of DDIs between ARVs**

DDIs are a serious complication of taking multiple medications and account for 3% to 5% of all hospital medication errors (Leape *et al.,* 1995). According to Clarke *et al.* (2008), the consequences of drug interactions vary ranging from drug toxicities to therapeutic failures, or loss of effectiveness and can significantly affect a patient's clinical outcome. Of particular concern are drug interactions in patients infected with HIV who are receiving HAART because it involves a regimen of a least three agents (Seden *et al*., 2009).

HAART has revolutionised the management of HIV-1 infection and the ARV therapy has improved steadily in terms of efficacy, tolerability, and dosing convenience since the advent of HAART in 1995 (Chandwani & Shuter, 2008). HAART consists of four classes that are available for ARV therapy: (Nucleoside/nucleotide reverse transcriptase inhibitors (N(t)RTIs; non-nucleoside reverse transcriptase inhibitors (NNRTIs); protease inhibitors (PIs); and a fusion inhibitor). The strongly recommended regimen based on the existing efficacy data, is either NNRTI-based or PI-based HAART (Yeni *at al*., 2004).

#### **7.1 Clinically significant drug interactions associated with Highly Active Antiretroviral Therapy**

One of the most challenging issues faced by health care providers treating patients with HIV-1 infection is the complex problem of DDIs associated with HAART (Seden *et al*., 2009; Clarke *et al*., 2008; Pontali, 2007; Cohen *et al*., 2002). The guidelines for the initial treatment of HIV infection recommend the use of at least three ARVs (Bartlett *et al.,* 2006a), each of which is associated with significant drug interactions (DHHS, 2003). Drug interactions associated with HIV medications can be classified into those that alter the pharmacokinetics and those that alter pharmacodynamics (Seden *et al*., 2009).

Pharmacokinetic drug interactions result in a change in pharmacokinetic parameters, such as the area under the curve (AUC), which measures drug exposure, peak concentration (Cmax), through concentration or half-life (Young, 2005; Cohen, 2002). Pharmacodynamic interactions result in alterations in the pharmacologic activity of the medication; not causing a change in the pharmacokinetic (Young, 2005; Cohen, 2002). The most common drug interactions in HIV medicine are pharmacokinetic interactions as a result of a change in the absorption, distribution and metabolism and the result of the concurrently administered medication (Piscitelli & Gallicano, 2001).

#### **7.2 Influence of cytochrome P450 (CYP450) on DDIs in HIV**

The cytochrome P450 enzyme system is responsible for the biotransformation of drugs from active to inactive metabolites that are readily excreted by the body. DDIs are more common in PIs and NNRTIs (Seden *et al*., 2009; Winston & Boffito, 2005; Young, 2005; Cohen *et al*., 2002). Of the numerous isoenzymes of CYP450 that have been identified, the ones responsible for elimination of drugs used in HAART are CYP3A, CYP1A2, and CYP2D2 (Clarke *et al*., 2008).

#### **7.3 Nucleoside/Nucleotide Reverse Transcriptase Inhibitors (NRTIs & NtRTIs)**

The NRTIs are valuable ARV agents in the treatment of HIV infection because they constitute the "backbone" of highly active ARV therapy regimens (Waters & Boffito, 2007).

Sanderson (2005) found that the risk of DDIs rose from 13% in patients taking two drugs to

DDIs are a serious complication of taking multiple medications and account for 3% to 5% of all hospital medication errors (Leape *et al.,* 1995). According to Clarke *et al.* (2008), the consequences of drug interactions vary ranging from drug toxicities to therapeutic failures, or loss of effectiveness and can significantly affect a patient's clinical outcome. Of particular concern are drug interactions in patients infected with HIV who are receiving HAART

HAART has revolutionised the management of HIV-1 infection and the ARV therapy has improved steadily in terms of efficacy, tolerability, and dosing convenience since the advent of HAART in 1995 (Chandwani & Shuter, 2008). HAART consists of four classes that are available for ARV therapy: (Nucleoside/nucleotide reverse transcriptase inhibitors (N(t)RTIs; non-nucleoside reverse transcriptase inhibitors (NNRTIs); protease inhibitors (PIs); and a fusion inhibitor). The strongly recommended regimen based on the existing

**7.1 Clinically significant drug interactions associated with Highly Active Antiretroviral** 

One of the most challenging issues faced by health care providers treating patients with HIV-1 infection is the complex problem of DDIs associated with HAART (Seden *et al*., 2009; Clarke *et al*., 2008; Pontali, 2007; Cohen *et al*., 2002). The guidelines for the initial treatment of HIV infection recommend the use of at least three ARVs (Bartlett *et al.,* 2006a), each of which is associated with significant drug interactions (DHHS, 2003). Drug interactions associated with HIV medications can be classified into those that alter the pharmacokinetics and those

Pharmacokinetic drug interactions result in a change in pharmacokinetic parameters, such as the area under the curve (AUC), which measures drug exposure, peak concentration (Cmax), through concentration or half-life (Young, 2005; Cohen, 2002). Pharmacodynamic interactions result in alterations in the pharmacologic activity of the medication; not causing a change in the pharmacokinetic (Young, 2005; Cohen, 2002). The most common drug interactions in HIV medicine are pharmacokinetic interactions as a result of a change in the absorption, distribution and metabolism and the result of the concurrently administered

The cytochrome P450 enzyme system is responsible for the biotransformation of drugs from active to inactive metabolites that are readily excreted by the body. DDIs are more common in PIs and NNRTIs (Seden *et al*., 2009; Winston & Boffito, 2005; Young, 2005; Cohen *et al*., 2002). Of the numerous isoenzymes of CYP450 that have been identified, the ones responsible for elimination of drugs used in HAART are CYP3A, CYP1A2, and CYP2D2 (Clarke *et al*., 2008).

The NRTIs are valuable ARV agents in the treatment of HIV infection because they constitute the "backbone" of highly active ARV therapy regimens (Waters & Boffito, 2007).

**7.3 Nucleoside/Nucleotide Reverse Transcriptase Inhibitors (NRTIs & NtRTIs)** 

82% in patients taking seven drugs or more.

that alter pharmacodynamics (Seden *et al*., 2009).

medication (Piscitelli & Gallicano, 2001).

**7.2 Influence of cytochrome P450 (CYP450) on DDIs in HIV** 

**Therapy** 

**7. Pharmacological aspects of DDIs between ARVs** 

because it involves a regimen of a least three agents (Seden *et al*., 2009).

efficacy data, is either NNRTI-based or PI-based HAART (Yeni *at al*., 2004).

Drug interactions associated with NRTIs and NtRTIs are few because these drugs are not metabolised by the CYP450 system (Clarke *et al.,* 2008). However, drug interactions may still occur within these drugs as was demonstrated by Katende-Kyenda et al. (2008a). One of the few pharmacodynamic interactions encountered in HIV medicine occurs, for example with co-administered zidovudine and stavudine, since both drugs are thymidine analogues and they can compete for the same phosphorylation site in the growing chain of HIV DNA, resulting in an antagonistic, pharmacodynamic interaction (Piscitelli & Gallicano, 2001). It is therefore recommended that these two drugs never to be combined.

The use of didanosine (ddl) is complicated by drug interactions (Cohen *et al*, 2002). It is a buffered tablet form containing magnesium and calcium to improve systemic absorption. It, however, interacts with certain antibiotics like ciprofloxacin, tetracycline and therefore, to minimise the interaction, didanosine should be administered at least two hours after or six hours before the fluoroquinolone (Knupp & Barbhaiya, 1999). Concurrent use of didanosinebuffered tablets may also impair the absorption of the PI atazanavir, since atazanavir requires an acidic environment for absorption (Product Information Videx EC, 2003). To minimise the interaction, patients should take a didanosine-buffered tablet two hours after or one hour before taking atazanavir.

The most significant didanosine drug interaction reported occurs when didanosine is used concurrently with the NrRTI tenofovir. The didanosine AUC increases by 60% and therefore it is recommended that in patients receiving these two drugs concurrently and weighing > 60 kg, the didanosine dosage should be reduced from 400 mg to 250 mg once daily or from 250 mg to 200 mg in patients who weigh less than 60 kg (Young, 2005). For severely underweight patients, the dose should be further reduced to 125mg once daily (Faragon & Piliero, 2004). All patients receiving concurrent tenofovir and didanosine should be closely monitored for didanosine -related toxicities such as pancreatitis, hyperlactatemia, and lactic acidosis, regardless of didanosine dosage adjustments.

#### **7.4 Non-Nucleoside Reverse Transciptase Inhibitors (NNRTIs)**

Drugs in this group are prone to drug interactions because they are extensively metabolised via CYP3A4 and can act as either inducers or inhibitors of CYP3A4. Nevirapine and efavirenz are inducers of CYP3A4, while delavirdine is an inhibitor of CYP3A4 (Pfister *et al*., 2003). Therefore, when one of these drugs is combined with a drug that is also metabolised by CYP3A4, a drug interaction may occur (Clarke *et al*., 2008).

Nevirapine presents with numerous drug interactions, being a CYP3A4 inducer, and drug interactions associated with it lead to an increase in metabolism and reduced concentration of the co-administered drug. For example, when nevirapine is concurrently given with methadone, withdrawal symptoms may occur as a result of reduced methadone levels (Pinzanni *et al*., 2000). Efavirenz is a potent inducer of CYP3A4 in vivo. Like the PIs, EFV is extensively metabolised primarily by the CYP3A4 (Pfister *et al*., 2003).

The induction properties of efavirenz can result in reduced concentrations of concurrently administered drugs that are metabolised by CYP3A4 and it is therefore contraindicated with midazolam, triazolam and ergotamine derivative since there is a potential for increased drug concentrations of these medications and associated toxicity (Product Infomation Sustiva, 2003). Efavirenz, as a potent inducer of CYP3A4 is suggested to have a potential interaction with lopinavir and ritonavir, both of which inhibit CYP3A4. This interaction was

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 293

The PIs are extensively metabolised by the cytochrome P450 (CYP) enzymes present in the liver and small intestine (Winston & Boffito, 2005). Therefore drug interactions involving PIs will occur largely as a result of enzyme induction or enzyme inhibition (Barry *et al*., 1999). Some PIs can alter metabolism and thus the plasma concentration of other PIs, creating complex drug interactions when a second PI is added to HAART. According to Van Heeswijk *et al.* (2001), additionally, favourably positive DDIs can increase the exposure to PIs, allowing the use of lower doses at reduced dosing frequencies with fewer dietary

Protease inhibitors have differing affinities for the CYP3A4 isoenzyme. The most potent inhibitor of CYP3A4 is ritonavir (Cooper *et al*., 2003), whereas the least potent is saquinavir. CYP3A4 inhibition associated with indinavir, nelfinavir, and amprenavir, and atazanavir tends to be intermediate. Ritonavir is often the most likely medication in the PI class to cause drug interactions because in addition to its CYP3A4 inhibition, it also inhibits CYP2D6 and induces CYP1A2 and CYP2C9 (Clarke *et al*., 2009). However, ritonavir is often used to enhance the pharmacokinetic parameters of co-administered PIs like indinavir (Kappelhoff *et al*., 2005), due to its potent inhibition of their metabolism by CYP3A4 (Zeldin &

The use of boosted double PI regimen is presented with complex unexpected pharmacokinetic interactions (Winston & Boffito, 2005). Therefore combinations like tipranavir/ritonavir with others must be avoided because such combinations have shown to significantly reduce plasma concentrations of saquinavir, amprenavir and lopinavir (Boffito *et al*., 2005). Another interesting interaction that was observed by Boffito *et al*. (2005) was with the boosted double combinations of atazanavir/saquinavir/ritonavir. Saquinavir levels are enhanced in this regimen further than when dosed with ritonavir alone, thus suggesting

Although the number of clinically relevant DDIs is probably low, DDIs may be responsible for a substantial number of hospital admissions. Therefore the pharmacist is responsible for preventing the use of unsafe or non-effective drug regimens. Specifically, pharmacists should avoid the dispensing of combinations of drugs that may cause serious DDIs (Becker

Many drug interactions can be avoided or managed safely if adequate time and precautions are taken by a patient's pharmacist. Having the pharmacist provide patient counselling on the use of prescription and non-prescription medication, disease state(s), and the safety of concurrent use of herbal products plays a major role in avoiding drug interactions (Brown,

According to Lien and Lien (1994), many patients visit more than one doctor for their different diseases and receive more than one drug at a time, and often doctors are unaware of all the medications their patients are taking and the risks to which their patients are exposed when treated with multiple drugs. Since pharmacists in the community setting or hospital, are the most accessible health care providers, they are able to intervene when faced with potential drug interactions that may occur during patients' multiple drug therapy. Adverse DDIs are the major cause of morbidity and mortality. Cancer patients, for example, are particularly at high risk of such interactions because they commonly receive multiple

**8. Role of pharmacists in preventing DDIs in clinical practice** 

**7.8 Protease inhibitor interactions** 

restrictions.

Petruschke, 2004).

*et al*., 2005).

2004).

a role for this as a once daily regimen.

assessed in a parallel group study in which PI-experienced, NNRTI-naive, HIV-infected patients received different doses of these agents (Young, 2005).

## **7.5 Non-Nucleoside Reverse Transcriptase and Protease Inhibitors Interactions**

When predicting potential drug interactions, it is important to know which P450 isoenzyme is responsible for the metabolism of a drug. Drug interactions between NNRTIs and PIs are common as was observed in a study by Katende-Kyenda (2008b), as all currently available agents in these two classes are metabolised mainly by the 3A4 isoenzyme of the CYP450 system (Fichtenbaum & Gerber, 2002). NNRTIs and PIs also inhibit or induce CYP3A4, decreasing or increasing hepatic clearance and, thereby, increasing or decreasing plasma levels, respectively, of drugs metabolised by CYP3A4. Therefore, depending primarily on the potency of each NNRTI or PI as an inhibitor or inducer of CYP3A4 and on the substrate affected, each one has a different drug interaction profile.

All currently marketed PIs – atazanavir, amprenavir, fosamprenavir, indinavir, lopinavir, nelfinavir, ritonavir, and saquinavir – and the NNRTI delavirdine inhibit CYP3A4 (Piscitelli & Gallicano, 2001). According to Von Moltke *et al*. (1998), ritonavir is the most potent CYP3A4 inhibitor and, consequently, has the most drug interactions, while amprenavir, indinavir, lopinavir, and nelfinavir appear to inhibit CYP3A4 equally, and saquinavir with the lowest inhibitory effect.

## **7.6 Effect of Protease Inhibitors on Nucleoside Analogues**

The nucleoside analogue reverse transcriptase inhibitor, tenofovir, does not appear to inhibit CYP3A4 isoenzyme significantly and, like most inhibitors, is excreted by the kidneys. Tenofovir, unlike other NRTIs, is associated with several drug interactions, particularly a bidirectional effect (i.e. agent can alter plasma levels of the other) with atazanavir, while atazanavir raises plasma levels of tenofovir (Holder, 2003).

Indinavir does not alter the pharmacokinetics of zidovudine, stavudine or lamivudine (Perry & Belfour, 1996:928). This is because the optimal absorption of indinavir requires a normal (acidic) gastric pH whereas an acid medium rapidly degrades didanosine, which is formulated with buffering agents to increase the pH. Therefore the administration of indinavir and didanosine should be separated by at least 1 hour to avoid an interaction mediated by altered drug absorption (Perry & Balfour, 1996).

#### **7.7 Effect of Non-Nucleoside Reverse Transciptase Inhibitors on Nucleoside Analogues**

Nevirapine is a potent and selective non-competitive inhibitor of reverse transcriptase (De Maat *et al*., 2003). It does not compete with template or nucleoside triphosphates, and therefore a significant interaction would not be expected. Nevirapine may reduce plasma zidovudine concentrations by 25% but does not influence plasma concentrations of didanosine or zalcitabine (Murphy & Montaner, 1996).

NRTIs, unlike NNRTIs and PIs, are not metabolised by the hepatic CYP3A4 ezyme system and – the exception of zidovudine and abacavir - undergo renal rather than biliary excretion. Zidovudine undergoes hepatic glucuronidation and abacavir is metabolised in the liver by alcohol dehydrogenase (Barry *et al.,*1999). Therefore, there is little potential for interaction between NRTIs and NNRTIs or between NRTIs and PIs. In addition, the NRTI class as a whole has fewer drug interactions than the NNRTI and PI classes have.

## **7.8 Protease inhibitor interactions**

292 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

assessed in a parallel group study in which PI-experienced, NNRTI-naive, HIV-infected

All currently marketed PIs – atazanavir, amprenavir, fosamprenavir, indinavir, lopinavir, nelfinavir, ritonavir, and saquinavir – and the NNRTI delavirdine inhibit CYP3A4 (Piscitelli & Gallicano, 2001). According to Von Moltke *et al*. (1998), ritonavir is the most potent CYP3A4 inhibitor and, consequently, has the most drug interactions, while amprenavir, indinavir, lopinavir, and nelfinavir appear to inhibit CYP3A4 equally, and saquinavir with

The nucleoside analogue reverse transcriptase inhibitor, tenofovir, does not appear to inhibit CYP3A4 isoenzyme significantly and, like most inhibitors, is excreted by the kidneys. Tenofovir, unlike other NRTIs, is associated with several drug interactions, particularly a bidirectional effect (i.e. agent can alter plasma levels of the other) with atazanavir, while

Indinavir does not alter the pharmacokinetics of zidovudine, stavudine or lamivudine (Perry & Belfour, 1996:928). This is because the optimal absorption of indinavir requires a normal (acidic) gastric pH whereas an acid medium rapidly degrades didanosine, which is formulated with buffering agents to increase the pH. Therefore the administration of indinavir and didanosine should be separated by at least 1 hour to avoid an interaction

Nevirapine is a potent and selective non-competitive inhibitor of reverse transcriptase (De Maat *et al*., 2003). It does not compete with template or nucleoside triphosphates, and therefore a significant interaction would not be expected. Nevirapine may reduce plasma zidovudine concentrations by 25% but does not influence plasma concentrations of

NRTIs, unlike NNRTIs and PIs, are not metabolised by the hepatic CYP3A4 ezyme system and – the exception of zidovudine and abacavir - undergo renal rather than biliary excretion. Zidovudine undergoes hepatic glucuronidation and abacavir is metabolised in the liver by alcohol dehydrogenase (Barry *et al.,*1999). Therefore, there is little potential for interaction between NRTIs and NNRTIs or between NRTIs and PIs. In addition, the NRTI class as a

**7.7 Effect of Non-Nucleoside Reverse Transciptase Inhibitors on Nucleoside** 

**7.5 Non-Nucleoside Reverse Transcriptase and Protease Inhibitors Interactions**  When predicting potential drug interactions, it is important to know which P450 isoenzyme is responsible for the metabolism of a drug. Drug interactions between NNRTIs and PIs are common as was observed in a study by Katende-Kyenda (2008b), as all currently available agents in these two classes are metabolised mainly by the 3A4 isoenzyme of the CYP450 system (Fichtenbaum & Gerber, 2002). NNRTIs and PIs also inhibit or induce CYP3A4, decreasing or increasing hepatic clearance and, thereby, increasing or decreasing plasma levels, respectively, of drugs metabolised by CYP3A4. Therefore, depending primarily on the potency of each NNRTI or PI as an inhibitor or inducer of CYP3A4 and on the substrate

patients received different doses of these agents (Young, 2005).

affected, each one has a different drug interaction profile.

**7.6 Effect of Protease Inhibitors on Nucleoside Analogues** 

atazanavir raises plasma levels of tenofovir (Holder, 2003).

mediated by altered drug absorption (Perry & Balfour, 1996).

didanosine or zalcitabine (Murphy & Montaner, 1996).

whole has fewer drug interactions than the NNRTI and PI classes have.

the lowest inhibitory effect.

**Analogues** 

The PIs are extensively metabolised by the cytochrome P450 (CYP) enzymes present in the liver and small intestine (Winston & Boffito, 2005). Therefore drug interactions involving PIs will occur largely as a result of enzyme induction or enzyme inhibition (Barry *et al*., 1999). Some PIs can alter metabolism and thus the plasma concentration of other PIs, creating complex drug interactions when a second PI is added to HAART. According to Van Heeswijk *et al.* (2001), additionally, favourably positive DDIs can increase the exposure to PIs, allowing the use of lower doses at reduced dosing frequencies with fewer dietary restrictions.

Protease inhibitors have differing affinities for the CYP3A4 isoenzyme. The most potent inhibitor of CYP3A4 is ritonavir (Cooper *et al*., 2003), whereas the least potent is saquinavir. CYP3A4 inhibition associated with indinavir, nelfinavir, and amprenavir, and atazanavir tends to be intermediate. Ritonavir is often the most likely medication in the PI class to cause drug interactions because in addition to its CYP3A4 inhibition, it also inhibits CYP2D6 and induces CYP1A2 and CYP2C9 (Clarke *et al*., 2009). However, ritonavir is often used to enhance the pharmacokinetic parameters of co-administered PIs like indinavir (Kappelhoff *et al*., 2005), due to its potent inhibition of their metabolism by CYP3A4 (Zeldin & Petruschke, 2004).

The use of boosted double PI regimen is presented with complex unexpected pharmacokinetic interactions (Winston & Boffito, 2005). Therefore combinations like tipranavir/ritonavir with others must be avoided because such combinations have shown to significantly reduce plasma concentrations of saquinavir, amprenavir and lopinavir (Boffito *et al*., 2005). Another interesting interaction that was observed by Boffito *et al*. (2005) was with the boosted double combinations of atazanavir/saquinavir/ritonavir. Saquinavir levels are enhanced in this regimen further than when dosed with ritonavir alone, thus suggesting a role for this as a once daily regimen.

## **8. Role of pharmacists in preventing DDIs in clinical practice**

Although the number of clinically relevant DDIs is probably low, DDIs may be responsible for a substantial number of hospital admissions. Therefore the pharmacist is responsible for preventing the use of unsafe or non-effective drug regimens. Specifically, pharmacists should avoid the dispensing of combinations of drugs that may cause serious DDIs (Becker *et al*., 2005).

Many drug interactions can be avoided or managed safely if adequate time and precautions are taken by a patient's pharmacist. Having the pharmacist provide patient counselling on the use of prescription and non-prescription medication, disease state(s), and the safety of concurrent use of herbal products plays a major role in avoiding drug interactions (Brown, 2004).

According to Lien and Lien (1994), many patients visit more than one doctor for their different diseases and receive more than one drug at a time, and often doctors are unaware of all the medications their patients are taking and the risks to which their patients are exposed when treated with multiple drugs. Since pharmacists in the community setting or hospital, are the most accessible health care providers, they are able to intervene when faced with potential drug interactions that may occur during patients' multiple drug therapy.

Adverse DDIs are the major cause of morbidity and mortality. Cancer patients, for example, are particularly at high risk of such interactions because they commonly receive multiple

Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 295

Bates, DW.; Collen, DJ.; Laird, N. et al. (1995). Incidence of adverse drug events and

Beers, M.H.; Storrie, M. & Lee, G. (1990). Potential adverse drug interactions in the emergency room. *Annals of internal medicine*, Vol.112, No.1, pp.61-64. Benet, L.Z.; Mitchell, J.R. & SHEINER, L.B. (1990). Pharmacokinetics: the dynamics of drug

Bergk, V.; Gasse, C.; Schnell, R. & Haefeli, W.E. (2004). Requirements for a successful

Bhigjee AL. (2005). Seizure in HIV/AIDS: a southern African perspective. Acta Neurology

Blower, P.; De Wit, R.; Goodin, S. & Apro, M. (2005). DDIs in oncology: why are they

Boffito, M,; Acosta, E,; Burger, D. et al. (2005). Current status and future prospects of

Chandwan, A. & Shuter, J. (2008). Lopinavir/ritonavir in the treatment of HIV-1 infection: a review. *Therapeutics and Clinical Risk Management,* Vol.4, No.5, pp: 1023-1033. Chapron, D.J. (2001). Drug disposition and response. (*In*: Delafuente, J.C., Stewart, R.B,

Cheng, J. W. M.; Frishman, W. H.; Aronow, W. S. (2009). Updates on Cytochrome P450-

Clarke S,; Stein, C. & Townsed ML. (2008). Drug-drug interactions with HIV antiretroviral

Cohen, B.H. & Lea, R.B. (2004). Essentials of statistics for the social and behavioral sciences.

Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale: Erlbaum,

Cohen, K.; Andrews, S. & Maartens, G. (2002). Drug interactions with antiretroviral therapy.

editors, *Therapeutics in the elderly*. 3rd ed. Cincinnati: Harvey Whitney Books Co.;

Mediated Cardiovascular Drug Interactions. *American Journal of Therapeutics,* Vol.16,

monitoring and applied clinical pharmacology in ART. *AIDS*, Vol.10, No.3, pp:375-92. Bjorkman, I.; Fastbom, J.; Schmidt, I.K. & Bernsten, C.B. (2002). DDIs interactions in the

elderly. *The Annals of pharmacotherapy*, Vol.36, No.11, pp: 1675-1681. Brown, CH. (2004). Overview of drug interactions: role of pharmacist in managing drug-

http://www.uspharmacist.com. Date of access: 3 August 2005.

therapy. *US pharmacist,* Vol. 33, No.4, pp:HS-3-HS-21.

*The South African Journal of HIV medicine, pp*: 42-37.

group. *Journal of the American Medical Association*, Vol.274 No.1, pp.29-34. Becker, E.E.; Ceerecker, M.L.; Kallewwrd, M.; et al. (2005). Potential determinants of DDI

pp.371-378.

60, No.8, pp: 595-602.

Vol.55, No.2, pp:117-142.

therapeutic drug

drug interactions.

2001. Pp: 257-88).

No. 2. pp:155-163.

Hooken: Wiley, 2004.

1988.

1811p Pergamons Press, Ney York NY).

Scandinavia Vol.112. Suppl.131, pp:4-7.

potential adverse drug events: implications for prevention: ADE prevention study

associated with dispensing in community pharmacies. *Drug safety,* Vol.28, No.5,

absorption, distribution, and elimination. (*In*: Goodman, G.A. Rall, T.W. Nies, A. & Taylor. P.. *Goodman & Gilman's: the pharmacological basis of therapeutics.* 1990: XVI-

implementation of drug interaction information systems in general practice: results of a questionnaire survey in Germany. *European journal of clinical pharmacology,* Vol.

important and can they be minimised? *Critical Reviews in Oncology/Hematology,* 

medications, including cytotoxic chemotherapy, hormonal agents and supportive care drugs (Blower *et al*., 2005). Increased awareness by pharmacists of the potential for drug interactions will allow health care providers to minimise the risk by selecting appropriate drugs and also by monitoring for signs of interaction.

According to Pezella (2005), in 2000, the number of patient deaths attributable to ADRs in the United States of America, was estimated to be 218 000 annually. More than 51% of approved drugs in the market in 2009 may have serious side-effects not detected before marketing approval. Therefore health plans and pharmacy benefit managers must work together to take effective steps to increase ADR monitoring and reporting and to proactively avoid ADRs through pharmacy management tools.

## **9. Recommendations regarding management of level 2 ARV DDIs in clinical practice**

The overall review revealed that most DDIs are identified between ARVs interacting at level 2 as identified by Tatro guidelines. Therefore the following recommendations can be formulated to manage these DDIs, based on the standard treatment guidelines for ARVs.


## **10. Conclusion**

This chapter dealt with drug-drug interactions as a challenge presented to healthcare providers in the management of HIV/AIDS. This worldwide epidemic can be managed using HAART which according to the recommended treatment guidelines, three or more drugs have to be prescribed. However some these combinations present with DDIs, the major cause of adverse drug events. DDIs can nevertheless be managed accordingly either by switching the drug combinations or by dosage adjustments. It is therefore the role both prescribers and specifically the pharmacists to identify the DDIs and working in collaboration manage them.

## **11. References**


medications, including cytotoxic chemotherapy, hormonal agents and supportive care drugs (Blower *et al*., 2005). Increased awareness by pharmacists of the potential for drug interactions will allow health care providers to minimise the risk by selecting appropriate

According to Pezella (2005), in 2000, the number of patient deaths attributable to ADRs in the United States of America, was estimated to be 218 000 annually. More than 51% of approved drugs in the market in 2009 may have serious side-effects not detected before marketing approval. Therefore health plans and pharmacy benefit managers must work together to take effective steps to increase ADR monitoring and reporting and to proactively

**9. Recommendations regarding management of level 2 ARV DDIs in clinical** 

The overall review revealed that most DDIs are identified between ARVs interacting at level 2 as identified by Tatro guidelines. Therefore the following recommendations can be formulated to manage these DDIs, based on the standard treatment guidelines for ARVs. Patients must be told the importance of consulting their doctors before using over- the

The prescriber should always check for potential DDIs then prescribing any

 While DDIs involving HIV drugs are essentially unavoidable, many can be managed through dosage adjustments as recommended by McNicholl & Coffey, (2007 & 2009)

This chapter dealt with drug-drug interactions as a challenge presented to healthcare providers in the management of HIV/AIDS. This worldwide epidemic can be managed using HAART which according to the recommended treatment guidelines, three or more drugs have to be prescribed. However some these combinations present with DDIs, the major cause of adverse drug events. DDIs can nevertheless be managed accordingly either by switching the drug combinations or by dosage adjustments. It is therefore the role both prescribers and specifically the pharmacists to identify the DDIs and working in

Barry, M.; Mulcahy, F.; Merry, C. et al. (1999). Pharmacokinetics and potential interactions

Bartlett, J.A.; Faith, M.J.; DeMasi, R. et al. (2006a). An updated systematic overview of triple

amongst ARV agents used to treat patients with HIV infection. *Clinical* 

combination therapy in ARV-naive HIV-infected adults. *AIDS,* Vol.20, No.16,

counter drugs that might interact with their prescribed ARVs;

Drug level monitoring of concurrent patients' medications should be done;

concomitant drug for a patient who is on ARV therapy;

*Pharmacokinetics,* vol.36, No.4 pp. 289-304.

drugs and also by monitoring for signs of interaction.

avoid ADRs through pharmacy management tools.

**practice** 

and McNicholl, (2009).

collaboration manage them.

pp:2051-2064.

**10. Conclusion** 

**11. References** 


http://www.uspharmacist.com. Date of access: 3 August 2005.


Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 297

Hasten, P.D. (2002). Appendix 11: important drug interactions. (*In*: Katzung BG, editor.

Havril, D.V.; Tierney, C. Fiedland, G.H. et al. (2000). In vivo antagonism with zidovudine

Highleyman L. (2007). Drug interactions and anti-HIV therapy. The body: the complete

Honda, M.; Yasuoka, A.; AokiI, M. & Oka. S. (1999). A generalized seizure following

Johnson, M.D.; Newkirk, G. & White, J.R. (1999). Clinically significant drug interactions.

Juurlink DN, Mamdani M, Kopp A et al. (2003). DDIs among eldely patients hospitalized for

Kappelhoff, B.S.; Huutelma, D.R. & Sankatsing, N. (2005). Population pharmacokinetics of

Katende-Kyenda N L, Lubbe MS. Serfontein JHP, Truter I. (2008a). Prevalence of drug-drug

Katende-Kyenda NL, Lubbe MS, Serfontein JHP and Truter I. (2008b). Prevalence of drug-

Katende-Kyenda NL, Lubbe MS, Serfontein JHP, Truter I. (2007). Effect of prescribed

Knupp, C.A & Barbhaiya, F.H. (1999). A multiple-dose pharmacokinetic interaction study

but asymptomatic. *Biopharmaceutical drug disposition,* Vol.18, No.1, pp:65-77. Leape, L.L. Bates, D. W. Cullen, D.J. (1995). Systems analysis of adverse drug events. ADE

Lee, D. & Bergman, U. (1994). Studies of drug utilization. (*In* STROM, B.L., ed. *Pharmacoepidemiology.* 2nd ed. Chichester: John Wiley & Sons. p. 380).

*British journal of clinical pharmacology,* Vol.60, No.3, pp:276-286.

Africa. *South African Medical Journal*, Vol.98, No.2, pp: 109-113.

*and Therapeutics,* Vol. 33, pp: 393-400.

 http://www.thebody.com/content/treat/art2579. (Accessed on 2 August 2007). Holder, S. (2003). Important new pharmacokinetic data for ReyatazTM (atazanavir sulphate)

providers]. Bristol-Meyers Squibb Company. August 8.

hospital admissions. *Drug safety,* Vol.9, No.1, pp:51-59.

1995. p. 986-95.

No.2. pp:193-219.

1658.

408.

pp:35-43.

HIV/AIDS resource. Available at:

pp:321-325.

Basic and clinical harmacology. 6th ed. Norwalk Connecticut: Appleton & Lange;

plus stavudine combination therapy. *Journal of infectious diseases,* Vol.182, No.1,

in combination with Vired® (tenofovir disoproxil fumarate) [Letter to health care

initiation of nelfinavir in a patient with HIV type 1, suspected due to interaction between nelfinavir and phenytoin. *Internal Medicine,* Vol.38, No.3, pp:302-303. Jankel, C.A. & Fitterman L.K. (1993). Epidemiology of drug-drug interactions as a cause of

What you need to know before writing prescriptions. *Drug interactions,* Vol.105,

drug toxicity. *Journal of the American Medical Association,* Vol.289, No.13, pp: 1652-

indinavir alone and in combination with ritonavir in HIV-1-infected patients.

interactions of antiretroviral agents in the private health care sector in South

drug interactions between antiretroviral agents in different age groups in a section of the private health care sector setting in South Africa.*Journal of Clinical Pharmacy* 

minimum benefits on the prevalence of possible drug-drug interactions of antiretroviral agents in a section of the private health care sector in South Africa: a 2 year comparative study. *International Journal of Pharmacy Practice* Vol.16, pp: 403-

between didanosine (Videx) and ciprofloxacin in male subjects seropositive for HIV

Prevention Study Group. *Journal of American medical association,* Vol.274, No.1,


 http://www.aidsinfo.nih.gov/contentsfiles/AdultandAdolescentsGL.pdf. Date of access: 6 June 2009.


http://www.thebody.com/contents/art/3000.html. Date accessed: 21 Nov. 2007.


Cooper, C.L.; Van Heeswijik, R.P.G.; Gallicano, K. & Cameron, D.W. (2003). A. Review of

De Maat, M,; Ekhart,G,; Huitema, A. et al. (2003). Drug interactions between ARV drugs and comedicated agents. *Clinical pharmacokinetics*, Vol.42, No.3, pp: 223-282. Delafuente, J.C. (2003). Understanding and preventing drug interactions in elderly patients.

Devane, C.L. & Nemeroff, C.B. (2000). Guide to psychotropic drug interactions. New York:

DHHS (Department of Health and Human Services). 2004. Department of Health and

http://www.aidsinfo.nih.gov/contentsfiles/AdultandAdolescentsGL.pdf. Date of

Dube, M.P.; Stein, J.H.; Aberg, J.A. et al. (2003). Guidelines for the evaluation of and

Egger, M. May, M. Chêne, G. et al. (2002). Prognosis of HIV-1 infected patients starting

F aragon , J.J. & Piliero, P.J. (2003). Drug interactions associated with HAART: focus on

Fargon, J.J. & Piliero, P.J. (2004). Clinically significant drug interactions associated with

 http://www.thebody.com/contents/art/3000.html. Date accessed: 21 Nov. 2007. Fichtenbaum , C.J. & Gerber. J.G. (2002). Interactions between Antiretroviral drugs and

Flockhart, D.A. & Tanus-Santos, J.E. (2002). Implications of cytochrome P450 interactions

Fulco, P.P.; Vora, U.B. & Bearman, G.M. (2006). Acid suppressive therapy and the effects on protease inhibitors. *Annals of Pharmacotherapy,* Vol.40, No.11, pp:1974-1983. Gaeta, T. J.; Fiorini, M. & Ender, (2002). Potential drug-drug interactions in elderly patients presenting with syncope. *The Journal of emergency medicine,* 22(2): 159-162. Geletko, S.M. & Zuwallack, A.R. (2001). Treatment of hyperlipidemia in HIV-infected patients. *American Society of Health- System Pharmacists,* Vol.58, No.7, pp:607-614. Hall, M.R.P. (1986). Drug interactions in geriatric patients. *Drugs and Aging* ed. By D.Platt.

infection. *Clinical Pharmacotherapy,*Vol.41, No.14, pp:1196-1211.

Human Services and Henry J. Kaiser Family Foundation. 2003. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. November 10,

management of dyslipidemia in HIV-infected adults receiving antiretroviral therapy: recommendations of the HIV Medical Association of the IDSA and the adult AIDS Clinical Trials Group. *Clinical Infectious Diseases,* Vol.37, No.5, pp:613-

highly active antiretroviral therapy: a collaborative analysis of prospective studies.

treatments for addiction and recreational drugs. *AIDS read,* Vol.13, No.9, pp:433-4,

highly active antiretroviral therapy. Infectious Diseases in Corrections. HEPP

drugs used for therapy of the metabolic complications encountered during HIV

when prescribing medication for hypertension. *Archives of internal medicine,* 

*Clinical reviews in oncology/haematology,* Vol.48, No.2, pp: 133-143.

*Diseases,* Vol.36, No.12, pp:1585-92.

MedWorks Media Inc.

2003. [Web:]

627.

access: 6 June 2009.

*Lancet*, 360(9327):119-29.

Report. January 2004.

Vol.162, No.4, pp:405-12.

Springer – Verlag Berlin Heidelberg. 1986.

437-41, 446-50.

low-dose ritonavir in protease inhibitor combination therapy. *Clinical Infectious* 


http://www.thebody.com/content/treat/art2579. (Accessed on 2 August 2007).


Drug-Drug Interactions as a Challenge in the Treatment of HIV/AIDS 299

Pinzanni,V.; Faucherre, V.; Peyiere, H. & Blayac, J.P. (2000). Methadone withdrawal

Piscitelli, S.C. & Gallicano, K.D. (2001). Interactions among drugs for HIV and opportunistic infections. *The New England journal of medicine,* Vol.344, No.13, pp:984-996. Pontali, E. (2007). Interactions of antiretroviral drugs with anti-infectives and other

SahaiI, J. Gallicano, K. Oliveras, L. (1993). Cations in the didanosine tablet reduce

Saanderson. N. (2005). DDIs: the silent epidemic. *Psychiatric service ,* Vol.56, No.1, pp: 22-24. Seden, K.; Back D & Khoo S. (2009). ARV drug interactions:often recognised, frequently

Sewester, C.S.; Olin, B. & Hebel, S.K. (editors). (2001). Drug facts and comparisons. Facts and

Seymor, R.M. & Routledga, P.A. (1998). Important drug-drug interactions in the elderly.

Sheehah, N.L.; Brouillette, M.J. Delisle, M.S. & Allan, J. 2006. Possible interaction between

Simpon, D.M. & Tagliati, M. (1995). Nucleoside analogue-associated peripheral neuropathy

Sloan, R. (1992). Principles of drug therapy in geriatric patients. *American family physician,* 

Stannic, A,; Granarana, JC. (2009). Review of antiretroviral agents for the treatment of HIV

Strain, J.J.; Karim, A.& Caliendo, G. (2002b). Cardiac drug-psychotropic drug update.

Strain J.J. Karim, A. Caliendo, G. Brodsky, M. et al. (2002a). Neurologic drug-psychotropic

Swart, A. & Harris V. (2005). Drug interactions with TB therapy. *Community Medical* 

Swedko, P.J. Clark, H.D. Paramsothy, K. & Akbari, A. (2003). Serum creatinine an

Tatro DS., ed. (2009). Drug interaction facts, and comparisons: drug interaction facts.TM St.

Tran, J.Q.; Gerber, J.G. & Kerr, B.M. (2001). Delavirdine: clinical pharmacokinetics and drug

UNAIDS Report on the global AIDS epidemic. Bangkok: XV International AIDS Conference,

http://www.unaids.org/bangkok2004/report.html (accessed March 25, 2007).

inadequate screening test for renal failure in elderly patients. *Archives of internal* 

drug update. *General hospital psychiatry,* Vol.24, No.5, pp: 290-310.

interactions. C*linical Pharmacokinetics,* Vol.40, No.3, pp:207-226.

antiretrovirals. *Chemotherapy,* Vol.53, No.1, pp:26-29.

comparisons. St. Louis, MO 2000. *Monthly update July.* 

*Syndrome & Human Retrovirology,*Vol.9, No.2, pp:153-161.

*Drugs Aging,* Vol.12, No.6, pp: 485-94.

Vol.45, No.6, pp:2709-18.

*pharmacotherapy,* Vol.40, No.1, pp:147-150.

infection *Formulary,* Vol. 44, No.1, pp: 47-54.

*Education,* Vol.23, No.2, pp: 56-60.

*medicine,* Vol.163, No.3, pp:356-60.

Louis, MO: WolterKluwer Health.

2004.

*General hospital psychiatry,* Vol.24, No.5, pp: 283-289.

pp:405-407.

pp:292-297.

No.1, pp: 5-8.

symptoms with nevirapine and efavirenz. *Annals of Pharmacotherapy,* Vol.34, No.3,

ciprofloxacin bioavailability. *Clinical pharmacology and therapeutics,*Vol.53, No.3,

unavoidable, sometimes unmanageable. *Journal of Antimicrobial therapy*. Vol.64,

lopinavir/ritonavir and valproic acid exacerbates bipolar disorder. *The annals of* 

in human immunodeficiency virus infection. *Journal of Acquired Immune Deficiency* 


http://www.aidsetc.org/aidsetc?page=et-03-00-04. Date of access: 26 Jun 2008.


 http://www.aids-etc.org/aidsetc?page=etres-display&resource=etres-320. Date of access: 25 June 2008.


Lee, L.M. & Henderson, D.K. (2001). Torelability of postexposure ARV prophylaxis for occupational exposures to HIV. *Drug Safety,* Vol.24, No.8, pp:587-597. Lesho, E.P. & Gey, D.C. (2003). Managing issues related to ARV therapy. *American family* 

Lien, L.L. & Lien, E.J. (1994). Preventing potential drug interactions in community pharmacy. *Journal of clinical pharmacology and therapeutics,* Vol.19, No.6, pp: 371-9. Lim, M.L. Min, S.S. Eron, J.J. et al. (2004). Coadministration of lopinavir/ritonavir and

Matheny, C.J.; Lamb, M.W.; Brower, K.L.R. & Pollack, G.M. 2001. Pharmacokinetic and

McEvoy, G.K. 2000. (ed). American hospital formulary service drug information. *American* 

McNicholl, I.A. & Coffey, S. (2007). Dosage adjustments for ARV-ARV drug interactions

The AIDS Education & Training Centers (AETC) National Resource Center and UCSF

 http://www.aidsetc.org/aidsetc?page=et-03-00-04. Date of access: 26 Jun 2008. McNicoll, I.R. & Coffey, S. (2009). Dosage Adjustments for ARV-ARV Drug Interactions. AETC National Resource Center and UCSF Center for HIV Information. July 2009.

 http://www.aidsetc.org/aidsetc?page=et-03-00-04. Date of access: 6 Oct 2009. McNicholl, I.R. (2009). Dosage adjustments for ARV-ARV Drug Interactions (Adult Dosing).

http://www.aids-etc.org/aidsetc?page=etres-display&resource=etres-320. Date of

Merlo, J.; Liedholm, H.; Lindbland, U. et al. (2001). Prescriptions with potential drug

Nolan, L. & O'Malley, K. (1998). Prescribing for the elderly: Part 1 – sensitivity of the elderly

O'Connor-Semmes, R.L.; Kersey, K.; Williams, D.H. (2001). Effect of ranitidine on the

Pezella, E. (2005). Preventing adverse drug reactions in the general population. *Managed care* 

Pfister M,; Labbe. L.. & Hammer S.C. (2003). Population pharmacokinetics and

study. *British medical journal,* Vol.323, No.3710, pp: 427-428.

Perry , CM. & Balfour, JA. (1996). Fomiviresn. *Drugs*, Vol.52, No.6, pp:928-62.

*society of health system pharmacists*. Bethesda, MD. 2000.

phenytoin results in two –way drug interaction through cytochrome P-450 induction. *Journal of acquired immune deficiency syndrome,* Vol.36, No.5, pp:1034-1040.

pharmacodynamic implications of P-glycoprotein modulation. *Pharmacotherapy,* 

AETC (The AIDS Education &Training Centers) National Resource Center and

interactions dispensed at Swedish pharmacies in January 1999: cross sectional

to adverse drug reactions. *Journal of American geriatric society,* Vol.36, No.2, pp: 142-

pharmacokinetics of triazolam and a-hydroxytriazolam in both young (19-60 years) and older (61-78 years) people *Clinical pharmacology and therapeutics,* Vol.70, No.2,

pharmacodynamics of efavirenz, nelfinavir and indinavir: adult AIDS clinical trial group study. *Antimicrobial agents and chemotherapy,*Vol.47, No.1, pp: 130-137.

*physician,* Vol.68, No.44, pp:675-686.

Vol.21, No.7, pp:778-798.

ACSF Center for HIV Information.

*interface,* Vol.18, No.1, pp: 49-52.

access: 25 June 2008.

149.

pp: 126-31.

(Adult Dosing).

Center for HIV Information, October 2007.


http://www.unaids.org/bangkok2004/report.html (accessed March 25, 2007).

Van Heeswijk, R.P. Veldkamp, A. Mulder, J. W. et al. (2001). Combinations of protease inhibitors for the treatment of HIV-1-infected patients. A review of pharmacokinetics and clinical experience. *Antiviral therapy,* Vol.6, No.4, pp:201-229.

**14** 

*1Colombia 2Spain* 

**Clinical Relevance of Drug Interactions** 

Pedro Amariles1, Newar Giraldo Alzate1 and Maria Jose Faus2

The identification, prevention, and clinical solution of drug interactions (DIs) are a critical aspect to achieve desired pharmacotherapy goals in patients infected with human immunodeficiency virus and/or affected by acquired immunodeficiency syndrome (HIV/AIDS) receiving antiretroviral (ARV) therapy, mainly because DIs may lead that ARV therapy will be unsafe and/or ineffective and thus, DIs may be clinical relevant. Additionally, in this group of patients the DIs are more frequent among other aspects by: The use of Highly Active Antiretroviral Therapy (HAART) or combined Antiretroviral Therapy (cART) includes three or more ARV drugs; (DHHS, 2011) therefore it is

 The pharmacokinetic properties of ARV drugs, for instance several of them are metabolized through complimentary cytochrome P450 isoenzymes, thus their

 ARV drugs are concurrently used with other class of medications for other common conditions, mainly infections and cardiovascular disease; and many of these medications used to treat these conditions are metabolized through complimentary

Accordingly, some studies illustrate that the 96% of patients receiving HAART or cART has at least a clinical condition or use a concomitant drug that could cause that ARV therapy may be unsafe (adverse drug reactions) or ineffectiveness (therapeutic failure). (Grimes et al., 30 2002) Therefore, identifying, preventing, and solving clinically relevant DIs is recognized both as a topic of great importance in achieving therapeutic goals for drug therapy (Kashuba, 2005) as a constant challenge to health care providers to HIV-infected patients receiving HAAR or cART. In addition, the clinical significance of a DI depends on the disposition and toxicity profile of the drug being administered. Thus, in HIV-infected patients assessing the clinical relevance of a DI is complex due to the large interpatient variability in pharmacokinetics exhibited by most ARV drugs, and then the evaluation and prediction of clinical effect of a DI is critical in the pharmacotherapy of patients with

therapeutic use could be accompanied by frequent DIs. (Miller et al., 2007)

cytochrome P450 isoenzymes, so several pharmacokinetic DIs may occur.

**1. Introduction** 

HIV/AIDS.

associated with a greater likelihood of DIs.

**in HIV-Infected Patients Receiving** 

**Antiretroviral Therapy** 

*1University of Antioquia, Medellin, 2University of Granada, Granada,* 


## **Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy**

Pedro Amariles1, Newar Giraldo Alzate1 and Maria Jose Faus2 *1University of Antioquia, Medellin, 2University of Granada, Granada, 1Colombia 2Spain* 

## **1. Introduction**

300 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Van Heeswijk, R.P. Veldkamp, A. Mulder, J. W. et al. (2001). Combinations of protease

Von Moltke, L.L. Greenblatt, D.J. Grassi, J.M. et al. (1998). Protease inhibitors as inhibitors of

Waters, L & Boffito, M. (2007). Pharmacology of current and investigational human

Winston, A. & Boffito, M. 2005. The management of HIV-1 protease inhibitor

Yeni, P.G.; Hammer, S.M.; Carpenter, C.C. et al. (2002). Antiretroviral treatment for adult

Zeldin, R.K. & Peetruschke, R.A. (2004). Pharmacological and therapeutic properties of

USA Panel. *Journal of American Association,* Vol.288, No.2, pp:222-235. Young, B. 2005. Mixing New Cocktails: Drug Interactions in ARV Regimens. *Aids patient care* 

*pharmacotherapy,* Vol.38, No.2, pp:106-111.

*and STDs* Vol.19, No.5, pp:286-297.

*antimicrobial chemotherapy* Vol.53, No.1, pp:4-9.

pp:213-221.

pp:1-5.

inhibitors for the treatment of HIV-1-infected patients. A review of pharmacokinetics and clinical experience. *Antiviral therapy,* Vol.6, No.4, pp:201-229.

human cytochromes P450: High risk associated with ritonavir. *The journal of clinical* 

immunodeficiency virus (HIV) nucleoside/nucleotide reverse transcriptase inhibitors in adults. *Anti-Infective Agents in Medicinal Chemistry,* Vol. 6, No.3,

pharmacokinetic interactions. *Journal of antimicrobial chemotherapy*, Vol.56, No.1,

HIV infection in 2002: updated recommendations of the International AIDS Society-

ritonavir-boosted protease inhibitor therapy in HIV-infected patients. *The Journal of* 

The identification, prevention, and clinical solution of drug interactions (DIs) are a critical aspect to achieve desired pharmacotherapy goals in patients infected with human immunodeficiency virus and/or affected by acquired immunodeficiency syndrome (HIV/AIDS) receiving antiretroviral (ARV) therapy, mainly because DIs may lead that ARV therapy will be unsafe and/or ineffective and thus, DIs may be clinical relevant. Additionally, in this group of patients the DIs are more frequent among other aspects by:


Accordingly, some studies illustrate that the 96% of patients receiving HAART or cART has at least a clinical condition or use a concomitant drug that could cause that ARV therapy may be unsafe (adverse drug reactions) or ineffectiveness (therapeutic failure). (Grimes et al., 30 2002) Therefore, identifying, preventing, and solving clinically relevant DIs is recognized both as a topic of great importance in achieving therapeutic goals for drug therapy (Kashuba, 2005) as a constant challenge to health care providers to HIV-infected patients receiving HAAR or cART. In addition, the clinical significance of a DI depends on the disposition and toxicity profile of the drug being administered. Thus, in HIV-infected patients assessing the clinical relevance of a DI is complex due to the large interpatient variability in pharmacokinetics exhibited by most ARV drugs, and then the evaluation and prediction of clinical effect of a DI is critical in the pharmacotherapy of patients with HIV/AIDS.

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 303

efficacy or increase the toxicity) **that may lead to therapeutic failure or adverse drug reactions associated to a previous or a concomitant use** of another drug (**drug-drug interactions**), including herbal drug products (**herbal-drug interactions**), certain type of food (**drug-food interactions**) or due to a patient´s physio-pathological condition (**drugdisease interaction**). (Amariles et al., 2007a) Additionally, changes on the results of certain laboratory tests that may produce some drugs (**drug - laboratory tests interactions**) (Maddox et al., 1980) or on the bioavailability of several nutrients (**drug - nutrient interactions**) (Chan, 2002; Santos & Boullata, 2005) may be considered as a DI. (Amariles,

Elsewhere, the increasing use of herbal products worldwide and the growth of the herbal product industry have led to rising the identification and characterization of clinically relevant DIs among several drugs with some of these products, for instance St. John's Wort (Hypericum Perforatum L.), which has been the characterization of another type of DI:

Although from a pharmacological perspective, some DIs may lead to a required therapeutic effect, for instance "pharmacokinetic enhancement or ritonavir boosting, strategy in which low doses of ritonavir -100 to 200 mg- (a cytochrome P4503A inhibitor) are used in combination with other protease inhibitors to increase antiretroviral drug exposure (Rathbun & Rossi, 2002), from a risk perspective, the efforts should focus on evaluating, predicting, and solving DIs with high probability to produce effects that are undesirable and to arise **toxicities or therapeutic failures**, which are termed as **clinically relevant drug** 

**Pharmacological mechanism of drug interactions.** The **previous or concomitant** use of a drug, herbal drug product, or food (like a patient´s physio-pathological condition) may cause a **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect** because such substance or situation causes an alteration that involve one or more of the three pharmacologic processes namely biopharmaceutics, pharmacokinetics, or

**The biopharmaceutics** is relating both to factors that influence the drug release from a drug product and the drug dissolution rate in the absorption site. Whereas, **the pharmacokinetics** is relating to drug absorption, distribution, metabolism, and excretion (ADME) (**what the body does to the drug**). As most drugs proceed through first order kinetics, and the process of ADME usually follows first order kinetics as well, the relationship between dose/time and drug plasma concentration for most drugs is linear. Thus, the **biopharmaceutics and pharmacokinetics phases determine the drug plasma concentration**. Similarly, since the relationship between plasma concentration and drug available on the site of action or biophase is linear, and the concentration on biophase is directly related to the intensity and duration of therapeutic response for most drugs as well, the relationship between drug plasma concentration and the magnitude of drug

**Pharmacodynamics phase (what the drug does to the body)**: the pharmacodynamics studies the **mechanism of action** (specific molecular drug-target interaction, usually as a result of binding to a receptor or an enzyme, through which a drug causes its **pharmacological response or effect** (pharmacological surrogate or clinical effect as result of

**herbal–drug drug interactions**. (Markowitz & DeVane, 2001)

**interactions.** (Amariles, 2002; Amariles et al., 2007a)

effect for most drugs is linear (figure 1).

drug- target interaction).

2002)

pharmacodynamics.

Since most of ARV drug DIs are clinically relevant, it is considered appropriate both to outline **the concept, types, mechanisms, and effects of ARV DIs on drug therapy, and to present a comprehensive summary of those drugs that are affected and the clinical relevance of ARV DIs.** In this way, the aim of this chapter is provide evidence and systematize information about DIs in HIV-infected receiving ART therapy, which allow define, evaluate, and predict the clinical relevance of the DIs, highlight those associated to pharmacokinetic mechanism. In this way, a proposal to identify, evaluate, and predict DIs considered as clinically relevant is presented, in which clinical relevance of a DI is defined according to the probability of their occurrence and to the severity of clinical effect in patient health (adverse event or therapeutic failure). (Amariles et al., 2007a)

Previous review about DIs with ARV, (Amariles et al., 2007b; Giraldo et al., 2010) achieved as a result of searched in Pubmed/Medline database, have showed that, in the case of clinically relevant pharmacokinetic interactions, nearly 80% are related to changes in systemic clearance, mainly associated to the systemic inhibition or induction of the metabolic activity of the cytochrome P-450 (CYP-450), mostly CYP3A4 isoform, whereas approximately 15% are related to changes in bioavailability (changes in gastrointestinal pH, presystemic clearance [mediated by CYP3A4 hepatic or intestinal]) or in P-glycoprotein activity). (Amariles et al., 2007b; Giraldo et al., 2010)

For this chapter, the earlier published information (Amariles et al., 2007a, 2007b; Giraldo et al., 2010; Amariles, 2002.) have been complement with information achieved from both a structured and systematic review of publications on Pubmed/Medline and references cited in relevant articles, and in other electronic databases (SIETES, MEDSCAPE, and TRIPDATABASE), and supplemented by other primary and secondary information sources to identify DIs in HIV-infected patients. Thus, searched MeSH terms were drug interactions, antiretroviral agents (or drugs), drug food interactions, drug nutrient interactions, drug laboratory test interference, drug in special situations (age, diseases), drug herbal plant interactions, computerized drug interactions, decision clinical computer based, and clinical relevance, clinically relevant or significantly relevant.

Finally, according with clinical relevance of the DIs, pairs of the identified DIs have been classified in four levels, according to rate probability and severity, (Amariles et al., 2007a, 2007 b; Giraldo et al., 2010) and then, the different drug pairs have been structured in a software designed to facilitate the identification, evaluation, and prediction of clinical relevant DIs. Current, 1,082 drug pairs of potential DIs have been identified, near to 80% of them due to pharmacokinetic mechanism (changes in plasma concentration), mainly associated to systemic enzyme inhibition. The scaling of these 1,082 drug pairs of recognized DIs, according to different dosage forms and strengths of identified drugs, generates a total of 6,087 pairs of DIs, in which, according their clinical relevance, 4,158 (68.3%) are clinical relevant (Levels 1 and 2) in HIV-infected patients receiving ARV therapy. Thus, the designed software meets the requirements defined for this type of program (Gaikwad et al., 2007; Rodríguez et al., 2009) and most important it facilitates the assessment, prediction, and decision on clinical relevance of 4,158 ARV DIs, which are considered of clinical interest in patients with HIV/AIDS (levels 1 and 2).

### **2. Drug interactions in HIV-infected patients receiving antiretroviral therapy**

#### **2.1 Concept, type, and mechanism of drug interactions**

**Concept of DI.** In patients with HIV/AIDS a DI could be assumed as **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect** (decrease the

Since most of ARV drug DIs are clinically relevant, it is considered appropriate both to outline **the concept, types, mechanisms, and effects of ARV DIs on drug therapy, and to present a comprehensive summary of those drugs that are affected and the clinical relevance of ARV DIs.** In this way, the aim of this chapter is provide evidence and systematize information about DIs in HIV-infected receiving ART therapy, which allow define, evaluate, and predict the clinical relevance of the DIs, highlight those associated to pharmacokinetic mechanism. In this way, a proposal to identify, evaluate, and predict DIs considered as clinically relevant is presented, in which clinical relevance of a DI is defined according to the probability of their occurrence and to the severity of clinical effect in patient

Previous review about DIs with ARV, (Amariles et al., 2007b; Giraldo et al., 2010) achieved as a result of searched in Pubmed/Medline database, have showed that, in the case of clinically relevant pharmacokinetic interactions, nearly 80% are related to changes in systemic clearance, mainly associated to the systemic inhibition or induction of the metabolic activity of the cytochrome P-450 (CYP-450), mostly CYP3A4 isoform, whereas approximately 15% are related to changes in bioavailability (changes in gastrointestinal pH, presystemic clearance [mediated by CYP3A4 hepatic or intestinal]) or in P-glycoprotein

For this chapter, the earlier published information (Amariles et al., 2007a, 2007b; Giraldo et al., 2010; Amariles, 2002.) have been complement with information achieved from both a structured and systematic review of publications on Pubmed/Medline and references cited in relevant articles, and in other electronic databases (SIETES, MEDSCAPE, and TRIPDATABASE), and supplemented by other primary and secondary information sources to identify DIs in HIV-infected patients. Thus, searched MeSH terms were drug interactions, antiretroviral agents (or drugs), drug food interactions, drug nutrient interactions, drug laboratory test interference, drug in special situations (age, diseases), drug herbal plant interactions, computerized drug interactions, decision clinical computer based, and clinical

Finally, according with clinical relevance of the DIs, pairs of the identified DIs have been classified in four levels, according to rate probability and severity, (Amariles et al., 2007a, 2007 b; Giraldo et al., 2010) and then, the different drug pairs have been structured in a software designed to facilitate the identification, evaluation, and prediction of clinical relevant DIs. Current, 1,082 drug pairs of potential DIs have been identified, near to 80% of them due to pharmacokinetic mechanism (changes in plasma concentration), mainly associated to systemic enzyme inhibition. The scaling of these 1,082 drug pairs of recognized DIs, according to different dosage forms and strengths of identified drugs, generates a total of 6,087 pairs of DIs, in which, according their clinical relevance, 4,158 (68.3%) are clinical relevant (Levels 1 and 2) in HIV-infected patients receiving ARV therapy. Thus, the designed software meets the requirements defined for this type of program (Gaikwad et al., 2007; Rodríguez et al., 2009) and most important it facilitates the assessment, prediction, and decision on clinical relevance of 4,158 ARV DIs, which are considered of clinical interest in

**2. Drug interactions in HIV-infected patients receiving antiretroviral therapy** 

**Concept of DI.** In patients with HIV/AIDS a DI could be assumed as **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect** (decrease the

health (adverse event or therapeutic failure). (Amariles et al., 2007a)

activity). (Amariles et al., 2007b; Giraldo et al., 2010)

relevance, clinically relevant or significantly relevant.

patients with HIV/AIDS (levels 1 and 2).

**2.1 Concept, type, and mechanism of drug interactions** 

efficacy or increase the toxicity) **that may lead to therapeutic failure or adverse drug reactions associated to a previous or a concomitant use** of another drug (**drug-drug interactions**), including herbal drug products (**herbal-drug interactions**), certain type of food (**drug-food interactions**) or due to a patient´s physio-pathological condition (**drugdisease interaction**). (Amariles et al., 2007a) Additionally, changes on the results of certain laboratory tests that may produce some drugs (**drug - laboratory tests interactions**) (Maddox et al., 1980) or on the bioavailability of several nutrients (**drug - nutrient interactions**) (Chan, 2002; Santos & Boullata, 2005) may be considered as a DI. (Amariles, 2002)

Elsewhere, the increasing use of herbal products worldwide and the growth of the herbal product industry have led to rising the identification and characterization of clinically relevant DIs among several drugs with some of these products, for instance St. John's Wort (Hypericum Perforatum L.), which has been the characterization of another type of DI: **herbal–drug drug interactions**. (Markowitz & DeVane, 2001)

Although from a pharmacological perspective, some DIs may lead to a required therapeutic effect, for instance "pharmacokinetic enhancement or ritonavir boosting, strategy in which low doses of ritonavir -100 to 200 mg- (a cytochrome P4503A inhibitor) are used in combination with other protease inhibitors to increase antiretroviral drug exposure (Rathbun & Rossi, 2002), from a risk perspective, the efforts should focus on evaluating, predicting, and solving DIs with high probability to produce effects that are undesirable and to arise **toxicities or therapeutic failures**, which are termed as **clinically relevant drug interactions.** (Amariles, 2002; Amariles et al., 2007a)

**Pharmacological mechanism of drug interactions.** The **previous or concomitant** use of a drug, herbal drug product, or food (like a patient´s physio-pathological condition) may cause a **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect** because such substance or situation causes an alteration that involve one or more of the three pharmacologic processes namely biopharmaceutics, pharmacokinetics, or pharmacodynamics.

**The biopharmaceutics** is relating both to factors that influence the drug release from a drug product and the drug dissolution rate in the absorption site. Whereas, **the pharmacokinetics** is relating to drug absorption, distribution, metabolism, and excretion (ADME) (**what the body does to the drug**). As most drugs proceed through first order kinetics, and the process of ADME usually follows first order kinetics as well, the relationship between dose/time and drug plasma concentration for most drugs is linear. Thus, the **biopharmaceutics and pharmacokinetics phases determine the drug plasma concentration**. Similarly, since the relationship between plasma concentration and drug available on the site of action or biophase is linear, and the concentration on biophase is directly related to the intensity and duration of therapeutic response for most drugs as well, the relationship between drug plasma concentration and the magnitude of drug effect for most drugs is linear (figure 1).

**Pharmacodynamics phase (what the drug does to the body)**: the pharmacodynamics studies the **mechanism of action** (specific molecular drug-target interaction, usually as a result of binding to a receptor or an enzyme, through which a drug causes its **pharmacological response or effect** (pharmacological surrogate or clinical effect as result of drug- target interaction).

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 305

with others important co-morbidities, such as dyslipidaemia, hypertension, tuberculosis, and opiate dependence, in which is needing to use several drugs that may be both substrate and

In general, the clinical relevance of a pharmacodynamic DI can often be evaluated, predicted, and monitored easily, because the process is supported by knowledge of the drugs mechanism of action and pharmacological effects (therapeutic and adverse), complemented with the definition and monitoring of parameters related to the drug clinical effects (clinical effects and toxicity profile), if possible in a quantitative way. Whereas, the pharmacokinetic DIs (**alterations in drug plasma concentration associated to changes in the release, dissolution, absorption, distribution, metabolism, or excretion of drug**) are more complex and may not be as easily evaluated, predicted, and monitored as the pharmacodynamic DIs; thus, the process requires both knowledge of pharmacology, pharmacotherapy, and clinical expertise. Thus, it is important to present a development proposal, which have been adjusted with goal to evaluate and predict the clinical relevance of pharmacokinetic DIs in HIV-infected patients receiving

**2.2.1 Identifying and assessing if one of the medication that the patient is using (or that the patient will be use) is considered as a drug with narrow therapeutic indices**  The therapeutic index of a drug is the ratio of the dose that produces toxicity (**drug plasma concentration that elicits the toxic effect in 50 percent of treated individuals -TD50-**), and the dose that produces a clinically desired or effective response in a population of individuals (**drug plasma concentration that elicits the therapeutic effect in 50 percent of** 

Both TD50 and ED50 are calculated from dose response curves, which represent the frequency with which each drug plasma concentration elicits the therapeutic effect or the

**Equation 1. Index therapeutic = TD50/ED50**

selected inducers or inhibitors of CYP3A4. (Josephson, 2010)

ARV therapy. (Amariles et al., 2007b; Giraldo et al., 2010)

toxic effect in the population (figure 2).

Fig. 2. Dose-responsive curves and Therapeutic Index

**the treated individuals -ED50-**) as shown in equation 1 (Katzung, 2009)

Fig. 1. Pharmacokinetics, plasmatic concentration (Cp), and pharmacologic effect.

Based on the pharmacological mechanism that explains the **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect**, the DI may be classified as:


#### **2.2 Proposal to evaluate and predict the clinical relevance of pharmacokinetic drug interactions in HIV-infected patients receiving antiretroviral therapy (Amariles et al., 2007b)**

HAART or cART has improved survival of HIV-infected patients, but they currently have chronic co-morbidities which require pharmacologic interventions with several medications, increasing the risk of DIs. In clinical practice, it is known that DIs may lead important pharmacotherapy problems especially for illnesses which require using various medications. Therefore, clinically relevant DIs are frequent among HIV-infected patients who are receiving ARV therapy. (Miller et al., 2007) In this context, ARV drugs may lower the efficacy or enhance side effects or toxicity of several of these drugs, and similarly some of these drugs may cause therapeutic failure or increase the toxicity of ARV drugs. (Fletcher et al., 2000)

DIs are especially important for drugs with narrow therapeutic indices and may either be pharmacodynamic or pharmacokinetic in nature. However, pharmacokinetic DIs may be more frequent complex to evaluate and to predict the effect among HIV-infected patients receiving ARV therapy. For instance, protease inhibitors (PIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs), and the CCR5 antagonist maraviroc are metabolized through the CYP450 system, mostly by CYP3A4. (DHHS, 2011) In addition, each of the NNRTIs and PIs induce and/or inhibit specific CYP450 enzymes and consequently are prone to cause pharmacokinetic DIs, (Pau & Boyd, 2010) mainly when they are concomitant used in patients with HIV and

**Pharmacodynamics Biophase Drug Effect** 

Based on the pharmacological mechanism that explains the **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect**, the DI may be classified as: Pharmacokinetic DI: The **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect is explained mainly by a change in the drug plasma concentration** associate with substance or situation that causes the DI. The change in the drug plasma concentration may be attributed to a biopharmaceutic (drug release and dissolution) and/or pharmacokinetic (ADME) alteration. Thus, biopharmaceutic DIs are included in pharmacokinetic DIs (**both biopharmaceutic and pharmacokinetic DIs are attributed to a change in the drug plasma concentration**). Pharmacodynamic DI: The **non-therapeutic and quantitative modification in the magnitude or duration of the drug effect** occurs without **change in the drug plasma concentration**. This type of DIs are mainly due to drugs or substances that have either similar (synergism) pharmacological effect or opposing (antagonistic) pharmacological effect or physio-pathological condition (drug-disease interactions) that contributes or facilities (synergism) the therapeutic or toxic effect of the drug, or that diminish or counteract its therapeutic effect (antagonism). In general, in one patient, the use of drugs those have a similar unsafe profile increases the likelihood and severity of adverse drug effects, for instance the use of drug-induced hepatic or renal toxicity.

**2.2 Proposal to evaluate and predict the clinical relevance of pharmacokinetic drug interactions in HIV-infected patients receiving antiretroviral therapy (Amariles et al.,** 

HAART or cART has improved survival of HIV-infected patients, but they currently have chronic co-morbidities which require pharmacologic interventions with several medications, increasing the risk of DIs. In clinical practice, it is known that DIs may lead important pharmacotherapy problems especially for illnesses which require using various medications. Therefore, clinically relevant DIs are frequent among HIV-infected patients who are receiving ARV therapy. (Miller et al., 2007) In this context, ARV drugs may lower the efficacy or enhance side effects or toxicity of several of these drugs, and similarly some of these drugs may cause therapeutic failure or increase the toxicity of ARV drugs. (Fletcher

DIs are especially important for drugs with narrow therapeutic indices and may either be pharmacodynamic or pharmacokinetic in nature. However, pharmacokinetic DIs may be more frequent complex to evaluate and to predict the effect among HIV-infected patients receiving ARV therapy. For instance, protease inhibitors (PIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs), and the CCR5 antagonist maraviroc are metabolized through the CYP450 system, mostly by CYP3A4. (DHHS, 2011) In addition, each of the NNRTIs and PIs induce and/or inhibit specific CYP450 enzymes and consequently are prone to cause pharmacokinetic DIs, (Pau & Boyd, 2010) mainly when they are concomitant used in patients with HIV and

**2007b)** 

et al., 2000)

Fig. 1. Pharmacokinetics, plasmatic concentration (Cp), and pharmacologic effect.

**Drug plasma concentration (Cp)**

**Pharmacokinetics**

**Dose/time**

with others important co-morbidities, such as dyslipidaemia, hypertension, tuberculosis, and opiate dependence, in which is needing to use several drugs that may be both substrate and selected inducers or inhibitors of CYP3A4. (Josephson, 2010)

In general, the clinical relevance of a pharmacodynamic DI can often be evaluated, predicted, and monitored easily, because the process is supported by knowledge of the drugs mechanism of action and pharmacological effects (therapeutic and adverse), complemented with the definition and monitoring of parameters related to the drug clinical effects (clinical effects and toxicity profile), if possible in a quantitative way. Whereas, the pharmacokinetic DIs (**alterations in drug plasma concentration associated to changes in the release, dissolution, absorption, distribution, metabolism, or excretion of drug**) are more complex and may not be as easily evaluated, predicted, and monitored as the pharmacodynamic DIs; thus, the process requires both knowledge of pharmacology, pharmacotherapy, and clinical expertise. Thus, it is important to present a development proposal, which have been adjusted with goal to evaluate and predict the clinical relevance of pharmacokinetic DIs in HIV-infected patients receiving ARV therapy. (Amariles et al., 2007b; Giraldo et al., 2010)

**2.2.1 Identifying and assessing if one of the medication that the patient is using (or that the patient will be use) is considered as a drug with narrow therapeutic indices**  The therapeutic index of a drug is the ratio of the dose that produces toxicity (**drug plasma concentration that elicits the toxic effect in 50 percent of treated individuals -TD50-**), and the dose that produces a clinically desired or effective response in a population of individuals (**drug plasma concentration that elicits the therapeutic effect in 50 percent of the treated individuals -ED50-**) as shown in equation 1 (Katzung, 2009)

Both TD50 and ED50 are calculated from dose response curves, which represent the frequency with which each drug plasma concentration elicits the therapeutic effect or the toxic effect in the population (figure 2).

Fig. 2. Dose-responsive curves and Therapeutic Index

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 307

which include drug as digoxin, lithium, vancomycin, aminoglycosides, and anticonvulsants. Although it is knowing that some drugs has narrow therapeutic index, for instance oral anticoagulants and insulin, they may not be included in this kind of lists, mainly due to: (1) Absence of population data of minimum toxic concentrations and minimum effective concentrations; (2) presence of a large interindividual pharmacodynamic variability (in some patient a similar drug plasma concentration may cause different magnitude or duration of the clinical effects); and (3) there are not a reasonable relationship between drug plasma concentrations and clinical effects. For that reason, although data for dose-response curves is obtained from several individuals, the therapeutic index is assumed more as an individual perspective whereas therapeutic range is assumed more as a population

In drug therapy process using drugs with a narrow therapeutic range or index, pharmacokinetic DIs may cause an **increase in blood/serum concentrations and thus could lead to drug toxic effects**, or may cause a **decrease in blood/serum concentrations and thus could lead to drug failure or ineffectiveness**. Therefore, in the process of evaluate and predict the clinical relevance of a DI, **the first stage is to establish whether any drugs that patient is using (or that will use) is considered as a narrow therapeutic range or index drug**. In general, if the drugs used have a broad therapeutic index or margin, the probability of a pharmacokinetic DI makes that the drug be unsafe or ineffective is low, because the probability of a change leads to achieved drug plasma concentrations are excluded at the therapeutic range or index (below of the minimum effective concentrations or above of the minimum toxic concentrations) is minimum. In general, if **pharmacokinetic DIs (as well as pharmacodynamic DIs) involves drug products that are not considered as narrow therapeutic range or index drugs they will be clinical irrelevant**; therefore in these cases

**ARV drugs as Narrow Therapeutic or Range Index Drugs.** Due to their pharmacological and clinical features, ARV drugs may be considered as narrow therapeutic or range index drugs. (FDA, 2005) For instance, although there are some controversies due to broad intraand inter-individual pharmacokinetic variability of ARV drugs, (Nettles et al., 2006) PIs and NNRTIs have defined drug plasma concentrations related to maximum efficacy and safety (therapeutic range) and, therefore ARV drugs are susceptible to therapeutic drug monitoring (assessing and monitoring drug concentration) (Justesen, 2006; Wertheimer et al., 2006), as shown in table 1. As consequence, it is possible to establish, that **patients that are receiving PI/NNRT have a high susceptibility to present clinically relevant DIs**. Additionally, some medications used in HIV-infected patients for treatment or prevention of some chronic co-morbidities and opportunistic infections may be considered as narrow therapeutic range or index drugs, for instance rifampin, rifabutin and other antibiotics, anticonvulsants, statins, antidepressants, antihypertensives, and opioids (DHHS, 2011).

the evaluating and predicting of DI may not be needed.

**2.2.2 Identifying and predicting consequences of drug interaction on the** 

**pharmacokinetics, plasma concentrations, and affected drug clinical effects (step** 

The consequences of a pharmacokinetic DI on the main pharmacokinetic process, on the drug plasma concentrations, and thus on the affected drug clinical effects, depends among other aspects, of: (1) the magnitude that the respective pharmacokinetic process affects

perspective.

**more complex)** 

From a clinical perspective, drug therapeutic range corresponds at drug plasma concentrations associate to likelihood of achieving, in the most patients, the maximum therapeutic effect with the minimum toxic effect. The probability that a pharmacotherapy process will be effective and safe increases if both the maximum drug plasma concentration and the minimum drug plasma concentration of the steady state **associate to a specific dosing schedule in a patient**  are included within the **therapeutic range (population)** or **therapeutic index (individual)** of the drug (**minimum effective level and minimum toxic level, which are theoretical**). If the concentrations achieved at steady state are outside of the therapeutic range or index (excluded), it is increases the probability of drug failure or ineffective (if the concentration is lesser than the minimum effective level) or of adverse or toxic effects (if the concentration is higher than the minimum toxic level). Therefore, drug therapy should obtain that the **achieved drug plasma concentrations** at steady state are included within the therapeutic range or index (**theoretical**) is a specific patient, as shown in figure 3.

Fig. 3. Relationship between drug plasma concentrations achieved in steady state and the theoretical therapeutic window (therapeutic index or therapeutic margin)

The probability that a pharmacokinetic DI causes that the drug **achieved drug plasma concentrations** at steady state are outside of the therapeutic range or index is inversely proportional to the difference between **minimum effective level and minimum toxic level** that are defined in the therapeutic index or range of the drug. The probability is higher (increased) for drugs with small difference between the minimum effective concentrations and the minimum toxic concentrations (drugs with a narrow therapeutic range or index).

It could be practice to consider a narrow therapeutic range or index drug if the drug: (FDA, 2005) (1) Require pharmacokinetics (therapeutic drug concentration) or pharmacodynamic (measuring clinical effects) monitoring, thus the effective and safe use of the drug require careful dosage titration and patient monitoring; (2) have less than a 2-fold difference in median lethal dose (LD50) and median effective dose (ED50) values; or (3) have less than a 2 fold difference in the minimum toxic plasma concentrations and the minimum effective plasma concentrations

From a clinical practical perspective, there are lists of narrow therapeutic index drugs, including their minimum effective concentrations and the minimum toxic concentrations,

From a clinical perspective, drug therapeutic range corresponds at drug plasma concentrations associate to likelihood of achieving, in the most patients, the maximum therapeutic effect with the minimum toxic effect. The probability that a pharmacotherapy process will be effective and safe increases if both the maximum drug plasma concentration and the minimum drug plasma concentration of the steady state **associate to a specific dosing schedule in a patient**  are included within the **therapeutic range (population)** or **therapeutic index (individual)** of the drug (**minimum effective level and minimum toxic level, which are theoretical**). If the concentrations achieved at steady state are outside of the therapeutic range or index (excluded), it is increases the probability of drug failure or ineffective (if the concentration is lesser than the minimum effective level) or of adverse or toxic effects (if the concentration is higher than the minimum toxic level). Therefore, drug therapy should obtain that the **achieved drug plasma concentrations** at steady state are included within the therapeutic range or index

Fig. 3. Relationship between drug plasma concentrations achieved in steady state and the

The probability that a pharmacokinetic DI causes that the drug **achieved drug plasma concentrations** at steady state are outside of the therapeutic range or index is inversely proportional to the difference between **minimum effective level and minimum toxic level** that are defined in the therapeutic index or range of the drug. The probability is higher (increased) for drugs with small difference between the minimum effective concentrations and the minimum toxic concentrations (drugs with a narrow therapeutic range or index). It could be practice to consider a narrow therapeutic range or index drug if the drug: (FDA, 2005) (1) Require pharmacokinetics (therapeutic drug concentration) or pharmacodynamic (measuring clinical effects) monitoring, thus the effective and safe use of the drug require careful dosage titration and patient monitoring; (2) have less than a 2-fold difference in median lethal dose (LD50) and median effective dose (ED50) values; or (3) have less than a 2 fold difference in the minimum toxic plasma concentrations and the minimum effective

Css minimum

Css maximum

Steady state

Minimum effective level

Minimum toxic level

Therapeutic window: Therapeutic index or therapeutic range

Maximum effective level

Minimum effective level

From a clinical practical perspective, there are lists of narrow therapeutic index drugs, including their minimum effective concentrations and the minimum toxic concentrations,

theoretical therapeutic window (therapeutic index or therapeutic margin)

plasma concentrations

**Drug plasma concentration**

(**theoretical**) is a specific patient, as shown in figure 3.

**Time** 

which include drug as digoxin, lithium, vancomycin, aminoglycosides, and anticonvulsants. Although it is knowing that some drugs has narrow therapeutic index, for instance oral anticoagulants and insulin, they may not be included in this kind of lists, mainly due to: (1) Absence of population data of minimum toxic concentrations and minimum effective concentrations; (2) presence of a large interindividual pharmacodynamic variability (in some patient a similar drug plasma concentration may cause different magnitude or duration of the clinical effects); and (3) there are not a reasonable relationship between drug plasma concentrations and clinical effects. For that reason, although data for dose-response curves is obtained from several individuals, the therapeutic index is assumed more as an individual perspective whereas therapeutic range is assumed more as a population perspective.

In drug therapy process using drugs with a narrow therapeutic range or index, pharmacokinetic DIs may cause an **increase in blood/serum concentrations and thus could lead to drug toxic effects**, or may cause a **decrease in blood/serum concentrations and thus could lead to drug failure or ineffectiveness**. Therefore, in the process of evaluate and predict the clinical relevance of a DI, **the first stage is to establish whether any drugs that patient is using (or that will use) is considered as a narrow therapeutic range or index drug**. In general, if the drugs used have a broad therapeutic index or margin, the probability of a pharmacokinetic DI makes that the drug be unsafe or ineffective is low, because the probability of a change leads to achieved drug plasma concentrations are excluded at the therapeutic range or index (below of the minimum effective concentrations or above of the minimum toxic concentrations) is minimum. In general, if **pharmacokinetic DIs (as well as pharmacodynamic DIs) involves drug products that are not considered as narrow therapeutic range or index drugs they will be clinical irrelevant**; therefore in these cases the evaluating and predicting of DI may not be needed.

**ARV drugs as Narrow Therapeutic or Range Index Drugs.** Due to their pharmacological and clinical features, ARV drugs may be considered as narrow therapeutic or range index drugs. (FDA, 2005) For instance, although there are some controversies due to broad intraand inter-individual pharmacokinetic variability of ARV drugs, (Nettles et al., 2006) PIs and NNRTIs have defined drug plasma concentrations related to maximum efficacy and safety (therapeutic range) and, therefore ARV drugs are susceptible to therapeutic drug monitoring (assessing and monitoring drug concentration) (Justesen, 2006; Wertheimer et al., 2006), as shown in table 1. As consequence, it is possible to establish, that **patients that are receiving PI/NNRT have a high susceptibility to present clinically relevant DIs**. Additionally, some medications used in HIV-infected patients for treatment or prevention of some chronic co-morbidities and opportunistic infections may be considered as narrow therapeutic range or index drugs, for instance rifampin, rifabutin and other antibiotics, anticonvulsants, statins, antidepressants, antihypertensives, and opioids (DHHS, 2011).

#### **2.2.2 Identifying and predicting consequences of drug interaction on the pharmacokinetics, plasma concentrations, and affected drug clinical effects (step more complex)**

The consequences of a pharmacokinetic DI on the main pharmacokinetic process, on the drug plasma concentrations, and thus on the affected drug clinical effects, depends among other aspects, of: (1) the magnitude that the respective pharmacokinetic process affects

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 309

(hepatic clearance) or by renal excretion (renal clearance) but also by other ways, such as biliary excretion. CL is a measure of the efficiency of human organism to remove irreversibly a drug from the systemic circulation or bloodstream by all routes of elimination,

Drugs are metabolized (changed) usually by enzymes found mainly in the liver but also in small intestine, lung, kidney, and skin to a metabolites by process known as drug metabolism or biotransformation. Biotransformation often changes no-polar or lipophilic drugs into metabolites more polar or hydrophilic, which tend to be excreted in the urine (renal excretion) or in the stool (biliary excretion) as glucuronate, sulphate or acetate conjugates. While, polar or hydrophilic drugs may be excreted without drug metabolism

A renal pharmacokinetic DI may be clinically relevant if: (1) there is a competitive inhibition of tubular secretion of the drug and (2) the **renal clearance contributes more than 30% and drug systemic clearance**. (Bonate et al., 1998; Launay et al., 2006) Additionally, certain drugs, for instance ritonavir may inhibit the renal secretion of certain drugs, which may be critical for drugs that are mainly eliminated by this via. For example, when digoxin is used concomitant with ritonavir, the PI may lead to an increase in the levels and pharmacological

Concepts provide above, particularly the equation 2, lead to understand why clinically relevant pharmacokinetic DIs are mainly explained by changes in **systemic clearance and bioavailability.** Thus, almost 80% of pharmacokinetic DIs are related to changes in systemic clearance, mainly associated to the systemic inhibition or induction of the metabolic activity of the CYP450 isoenzymes, and approximately 15% related to changes in bioavailability (changes in gastrointestinal pH, presystemic clearance [by hepatic or intestinal CYP3A4

The process of evaluating and predicting of the clinical relevance of a pharmacokinetic DI continues with the **identification if the main route of drug elimination is hepatic o renal, and thus if the drug is** remove irreversibly from the systemic circulation or bloodstream by excretion renal or by hepatic metabolism. **Generally, if the drug is eliminating by hepatic metabolism, the probability that a pharmacokinetic DI will be clinical relevant is elevated.** In these cases, subsequent to identify whether the systemic elimination of any drugs that patient is using (or drugs that patient will use) occurs primarily by hepatic metabolism, **the process must continue with the evaluation of the effect that may cause one possible hepatic metabolism inhibition or induction on** 

Because a near to 80% of clinically relevant pharmacokinetic DIs are related to hepatic metabolism, both systemic and presystemic, during the process of developing new drugs it is important to characterize and to predict DIs related to hepatic metabolism. With this goal, both cell cultures are used to establishing the ability of the new drug to modify the activity of major CYP450 isoenzymes, and the assessment of the susceptibility of the drug metabolism to be affected by drugs recognized as enzyme inhibitors and inducers. (Tucker,

The major route of elimination of PIs and NNRTIs is by hepatic metabolism and thus the pharmacokinetic DIs may be clinical relevant. Not at all, for nucleoside or nucleotide

isoenzyme]) or by in P-gp activity). (Amariles et al., 2007b; Giraldo et al., 2010)

mainly by biotransformation (or drug metabolism) and excretion.

through renal excretion.

effects of digoxin. (Ding et al., 2004)

2001; Obach et al., 2005, 2006)

**drug plasma concentrations.**


aAssessment as its active metabolite (m8). Cmin: minimum concentration

Table 1. Drug plasma concentrations related to effective minimum level and toxic minimum level that defined the therapeutic index or range of ARV drugs. (Justesen, 2006; Nettles et al., 2006; Wertheimer et al., 2006)

achieved drug plasma concentrations with the dosing-schedule used in the patient; and (2) the magnitude of change that the DI causes on altered pharmacokinetic process. These two aspects determine the influence of a DI both on the magnitude of change on drug plasma concentrations and on the probability of achieved drug plasma concentrations at steady state are excluded from the therapeutic range or index. Relate to this issue, the average steady-state concentration (Cpss) is an excellent estimator of achieved drug plasma concentrations, which is determined by the ratio between the drug delivery rate (input rate, which depends directly on the dose –D-, and the bioavailability –F-, and inversely on the dosing interval -**τ**-), and the clearance rate –CL- (output rate), as shown in equation 2.

$$\text{Equation 2. Cpss} = \frac{\mathbf{D} \ge \mathbf{F}}{\mathbf{\color{red}{x \ge CL}}}$$

**Bioavailability (F).** Although, bioavailability of drug is classically defined as the rate and extent to which the active ingredient (drug substance) is absorbed from the dosage form (upon oral administration), from a clinical perspective, this pharmacokinetic parameter may be assumed as the **amount of active ingredient** (drug substance) from the dosage form (drug product, upon oral administration) **that reaches the systemic circulation unchanged**. Therefore, the bioavailability of drug is influenced by: (1) the disintegration of the dosage form, the release of drug from a drug product, and the dissolution of drug in the absorption site; (2) the presystemic metabolism both in the gut lumen (extrahepatic) and in the liver by CYP450, especially CYP3A4; and (3) the contribution of intestinal transporters, which may decrease the bioavailability, by efflux effect of intestinal P-glycoprotein (P-gp) in drug absorbed from apical to basolateral), or may increase the bioavailability, by effect of the anionic organic polypeptide (TAOP), especially the type B. (Ho & Kim, 2005)

**Systemic or total clearance (CL)** is defined as the volume of plasma in the vascular compartment cleared of drug per unit of time [volume/time], mainly by hepatic metabolism

Table 1. Drug plasma concentrations related to effective minimum level and toxic minimum level that defined the therapeutic index or range of ARV drugs. (Justesen, 2006; Nettles et al.,

achieved drug plasma concentrations with the dosing-schedule used in the patient; and (2) the magnitude of change that the DI causes on altered pharmacokinetic process. These two aspects determine the influence of a DI both on the magnitude of change on drug plasma concentrations and on the probability of achieved drug plasma concentrations at steady state are excluded from the therapeutic range or index. Relate to this issue, the average steady-state concentration (Cpss) is an excellent estimator of achieved drug plasma concentrations, which is determined by the ratio between the drug delivery rate (input rate, which depends directly on the dose –D-, and the bioavailability –F-, and inversely on the

dosing interval -**τ**-), and the clearance rate –CL- (output rate), as shown in equation 2.

**Bioavailability (F).** Although, bioavailability of drug is classically defined as the rate and extent to which the active ingredient (drug substance) is absorbed from the dosage form (upon oral administration), from a clinical perspective, this pharmacokinetic parameter may be assumed as the **amount of active ingredient** (drug substance) from the dosage form (drug product, upon oral administration) **that reaches the systemic circulation unchanged**. Therefore, the bioavailability of drug is influenced by: (1) the disintegration of the dosage form, the release of drug from a drug product, and the dissolution of drug in the absorption site; (2) the presystemic metabolism both in the gut lumen (extrahepatic) and in the liver by CYP450, especially CYP3A4; and (3) the contribution of intestinal transporters, which may decrease the bioavailability, by efflux effect of intestinal P-glycoprotein (P-gp) in drug absorbed from apical to basolateral), or may increase the bioavailability, by effect of the anionic organic polypeptide (TAOP),

� � ��

Equation 2. Cpss = ���

**Systemic or total clearance (CL)** is defined as the volume of plasma in the vascular compartment cleared of drug per unit of time [volume/time], mainly by hepatic metabolism

**Drug Therapeutic range or index (ng/mL)** 

Saquinavir 250 – 600 Ritonavir (solo) 150 - 2.100 Indinavir 100 - 1.000 Nelfinavir 800 - 3.000 Amprenavir 400 - 2.200 Lopinavir/ritonavir 1.000 - 9.000 Atazanavir 150 - 1.000 Tipranavir 6.500 - 50.000

Nevirapine Cmin 3.500 Efavirenz 1.000 - 4.000

2006; Wertheimer et al., 2006)

especially the type B. (Ho & Kim, 2005)

Delavirdine Limited information aAssessment as its active metabolite (m8). Cmin: minimum concentration (hepatic clearance) or by renal excretion (renal clearance) but also by other ways, such as biliary excretion. CL is a measure of the efficiency of human organism to remove irreversibly a drug from the systemic circulation or bloodstream by all routes of elimination, mainly by biotransformation (or drug metabolism) and excretion.

Drugs are metabolized (changed) usually by enzymes found mainly in the liver but also in small intestine, lung, kidney, and skin to a metabolites by process known as drug metabolism or biotransformation. Biotransformation often changes no-polar or lipophilic drugs into metabolites more polar or hydrophilic, which tend to be excreted in the urine (renal excretion) or in the stool (biliary excretion) as glucuronate, sulphate or acetate conjugates. While, polar or hydrophilic drugs may be excreted without drug metabolism through renal excretion.

A renal pharmacokinetic DI may be clinically relevant if: (1) there is a competitive inhibition of tubular secretion of the drug and (2) the **renal clearance contributes more than 30% and drug systemic clearance**. (Bonate et al., 1998; Launay et al., 2006) Additionally, certain drugs, for instance ritonavir may inhibit the renal secretion of certain drugs, which may be critical for drugs that are mainly eliminated by this via. For example, when digoxin is used concomitant with ritonavir, the PI may lead to an increase in the levels and pharmacological effects of digoxin. (Ding et al., 2004)

Concepts provide above, particularly the equation 2, lead to understand why clinically relevant pharmacokinetic DIs are mainly explained by changes in **systemic clearance and bioavailability.** Thus, almost 80% of pharmacokinetic DIs are related to changes in systemic clearance, mainly associated to the systemic inhibition or induction of the metabolic activity of the CYP450 isoenzymes, and approximately 15% related to changes in bioavailability (changes in gastrointestinal pH, presystemic clearance [by hepatic or intestinal CYP3A4 isoenzyme]) or by in P-gp activity). (Amariles et al., 2007b; Giraldo et al., 2010)

The process of evaluating and predicting of the clinical relevance of a pharmacokinetic DI continues with the **identification if the main route of drug elimination is hepatic o renal, and thus if the drug is** remove irreversibly from the systemic circulation or bloodstream by excretion renal or by hepatic metabolism. **Generally, if the drug is eliminating by hepatic metabolism, the probability that a pharmacokinetic DI will be clinical relevant is elevated.** In these cases, subsequent to identify whether the systemic elimination of any drugs that patient is using (or drugs that patient will use) occurs primarily by hepatic metabolism, **the process must continue with the evaluation of the effect that may cause one possible hepatic metabolism inhibition or induction on drug plasma concentrations.**

Because a near to 80% of clinically relevant pharmacokinetic DIs are related to hepatic metabolism, both systemic and presystemic, during the process of developing new drugs it is important to characterize and to predict DIs related to hepatic metabolism. With this goal, both cell cultures are used to establishing the ability of the new drug to modify the activity of major CYP450 isoenzymes, and the assessment of the susceptibility of the drug metabolism to be affected by drugs recognized as enzyme inhibitors and inducers. (Tucker, 2001; Obach et al., 2005, 2006)

The major route of elimination of PIs and NNRTIs is by hepatic metabolism and thus the pharmacokinetic DIs may be clinical relevant. Not at all, for nucleoside or nucleotide

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 311

In clinical practice, inhibition of hepatic CYP enzymes or metabolic inhibition is one of the most common DI mechanisms and it is usually reversible and competitive. Most pharmacokinetic DIs occur when a drug increases or decreases metabolism of other drugs in the liver (CYP enzymes or glucuronidation). When a drug or substance causes a decrease in the systemic hepatic metabolism and clearance of a drug, this substance generates an increase in plasma concentrations and may lead to emerge of adverse drug events or toxicities. Thus, the process of evaluating and predicting the effects of this kind of DI needs to identify the drugs that are strong or moderate inhibitors of activity of

c. Determining if the pharmacokinetic DI related to hepatic metabolism might be either one-way or two-way direction (bidirectional) and if it might affect the levels and effects of the two drugs involved; situation that is likely to occur, because the need to use simultaneously either inhibitor drugs (e.g., macrolides and antifungal azoles) or inducer drugs (for instance rifampicin and carbamazepine) together with PIs and/or NNRTIs is

Changes in presystemic metabolism (due to inhibition or induction of extra-hepatic or hepatic CYP3A subfamily), in the activity of P-gp, or in the gastrointestinal pH may affect both the amount absorbed and the effects of some ARV drugs. In turn, antiretroviral drugs, particularly PIs, may modify the bioavailability and effects of some drugs, mainly through inhibition of presystemic metabolism and, to a lesser extent, of the activity of P-gp. Didanosine, especially in its dosage form as buffer solution, can change gastrointestinal pH

For drugs with kinetics elimination of first order or linear (most drugs used at therapeutic doses, phenytoin is one important exception), as it may be deduced from the equation 2, the increase or decreases of average steady-state concentration (Cpss) is inversely related to the decrease or increase of CL (Cpss1/CL). Generally, for narrow range or index drugs, a pharmacokinetic DI may be clinically relevant, if the DI causes a change in the achieved Cpss of +/-20% (FDA, 1999). As Cpss 1/CL, a decrease (associate to a reduction of hepatic metabolism) in the CL of 10%, 15%, 20%, 30%, 50%, and 75% may produce an increase in the Cpss of 11%, 18%, 25%, 43%, 100%, and 400%, respectively. The expected increasing of Cpss is calculated from the ratio of 100 divided by 100 less the percentage that CL is decreasing [100/ (100 less % of decrease of CL)]. Thus, if CL decreases in 10%, the increase of 11% is obtained from the ratio of 100 divided by 100 less 10 (100/90), which result is 111% and it is showing that Cpss increases in 11%. Similarly, 43% is obtained from the ratio of 100 divided by 100 less 30 (100/70) which result is 143% and it is showing that Cpss increases in 43%; and so similarly for other values. (Amariles, 2002) In the situations of increases of hepatic metabolism (induction enzymatic), it is requiring a minimal increase of 25% in CL, which

The drug plasma concentration before of the pharmacokinetic DI is another factor that contributes significantly to the magnitude of change in the clinical effect. For example, if diltiazem may decreased CL of quinidine (therapeutic range: 1-4 g/mL) in 35% (Laganiere et al., 1996), thus it may cause an increase of 54% in the achieved Cpss of quinidine. Thus, if the quinidine Cpss on time interaction was of 2.5 g/mL, it would increase to 3.85 g/mL, whereas if the quinidine Cpss was of 3.5 g mL, it would increase to 5.39 g/mL. As

**2.2.2.2 Evaluation of changes in the bioavailability of drugs orally administered** 

different CYP450 isozymes known also as "enzyme inhibitors for excellence".

common.

and the amount absorbed by some drugs.

may be caused a decrease of 20% in the achieved Cpss.

analogue reverse transcriptase inhibitors (NRTIs), due to they are eliminated primarily by renal excretion, clinically relevant pharmacokinetic DIs are less frequent. However, among NRTIs, abacavir is metabolized by alcohol dehydrogenase and zidovudine by gucoroniltransferase, which may cause that these two drugs to have interactions associated to changes in the activity of the respective enzymes. Additionally, there is evidence that tenofovir may modify the atazanavir metabolism and thus drug plasma concentrations and clinical effects of this PI; similarly, atazanavir and lopinavir/ritonavir may alter the drug plasma concentrations and clinical effects of tenofovir (see below). In addition, the systemic elimination of some drugs, considered as narrow therapeutic index or margin drugs, and commonly used in patients with HIV/AIDS (such as rifamycins, anticonvulsants, statins, and antidepressants) occurs by hepatic metabolism and, therefore, clinically relevant pharmacokinetic DIs are likely to arise.

#### **2.2.2.1 Evaluation of the effect that may cause one possible hepatic metabolism inhibition or induction on the plasma concentrations of the potentially affected drug**

The process of evaluation of a DI associated to hepatic metabolism needs: (1) identify the CYP450 enzyme which is responsible for the biotransformation of the drug whose metabolism can be altered, and (2) identifying agents that alter (induce or inhibit) the metabolic capacity of the CYP450 enzyme. The proper observance of this stage implies following three steps.


In the case of PIs and NNRTIs, their systemic metabolism occurs primarily by CYP3A4. Additionally, there are other drugs commonly administered to HIV/AIDS patients which are metabolized by CYP3 or CYP2 families.

b. Identifying drugs that may modify (inducing or inhibiting) the metabolic capacity of CYP2 and CYP3 families. For instance, rifampin, rifabutin, carbamazepine, phenobarbital, phenytoin, and Saint-John's-wort may induce the activity of CYP2 and CYP3 families, whereas azole antifungals, macrolides, calcium antagonists, immunosuppressants, and grapefruit juice may inhibit CYP3 family. Most the drugs ARV used in the treatment of HIV are metabolized by the CYP450 enzymes and they have the capacity to modify the activity of CYP3 family and, to a lesser extent, the activity of CYP2 family. Among the NNRTIs, efavirenz and nevirapine mainly induce CYP3A4 and CYP2B6, while delavirdine is primarily an enzymatic inhibitor of the CYP3A4, and etravirine inhibits CYP2C9 and CYP2C19 while inducing CYP3A4. However, efavirenz may inhibit the activity of CYP3A4, CYP2C9, and CYP2C19 isoenzymes. In vitro studies show most of the PIs are inhibitors of activity of CYP3A4 isoenzyme (atazanavir, darunavir/ritonavir, lopinavir/ritonavir, saquinavir, tipranavir/ritonavir) or strong CYP3A4 inhibitors (ritonavir, indinavir, nelfinavir). Some PIs both inhibit and induce CYP3A4 (amprenavir, fosamprenavir). Ritonavir also inhibits CYP2D6. **The NRTIs, the fusion inhibitor enfuvirtide, and the integrase inhibitor raltegravir are not metabolized by the CYP450 system**.

analogue reverse transcriptase inhibitors (NRTIs), due to they are eliminated primarily by renal excretion, clinically relevant pharmacokinetic DIs are less frequent. However, among NRTIs, abacavir is metabolized by alcohol dehydrogenase and zidovudine by gucoroniltransferase, which may cause that these two drugs to have interactions associated to changes in the activity of the respective enzymes. Additionally, there is evidence that tenofovir may modify the atazanavir metabolism and thus drug plasma concentrations and clinical effects of this PI; similarly, atazanavir and lopinavir/ritonavir may alter the drug plasma concentrations and clinical effects of tenofovir (see below). In addition, the systemic elimination of some drugs, considered as narrow therapeutic index or margin drugs, and commonly used in patients with HIV/AIDS (such as rifamycins, anticonvulsants, statins, and antidepressants) occurs by hepatic metabolism and, therefore, clinically relevant

**2.2.2.1 Evaluation of the effect that may cause one possible hepatic metabolism inhibition** 

The process of evaluation of a DI associated to hepatic metabolism needs: (1) identify the CYP450 enzyme which is responsible for the biotransformation of the drug whose metabolism can be altered, and (2) identifying agents that alter (induce or inhibit) the metabolic capacity of the CYP450 enzyme. The proper observance of this stage implies

a. Identification of the CYP450 enzyme which is responsible for the biotransformation of the drug that may be altered. A detailed and updated list of major CYP450 isoenzymes, together with their most common substrates, inhibitors, and inducers can be found on

In the case of PIs and NNRTIs, their systemic metabolism occurs primarily by CYP3A4. Additionally, there are other drugs commonly administered to HIV/AIDS patients which

b. Identifying drugs that may modify (inducing or inhibiting) the metabolic capacity of CYP2 and CYP3 families. For instance, rifampin, rifabutin, carbamazepine, phenobarbital, phenytoin, and Saint-John's-wort may induce the activity of CYP2 and CYP3 families, whereas azole antifungals, macrolides, calcium antagonists, immunosuppressants, and grapefruit juice may inhibit CYP3 family. Most the drugs ARV used in the treatment of HIV are metabolized by the CYP450 enzymes and they have the capacity to modify the activity of CYP3 family and, to a lesser extent, the activity of CYP2 family. Among the NNRTIs, efavirenz and nevirapine mainly induce CYP3A4 and CYP2B6, while delavirdine is primarily an enzymatic inhibitor of the CYP3A4, and etravirine inhibits CYP2C9 and CYP2C19 while inducing CYP3A4. However, efavirenz may inhibit the activity of CYP3A4, CYP2C9, and CYP2C19 isoenzymes. In vitro studies show most of the PIs are inhibitors of activity of CYP3A4 isoenzyme (atazanavir, darunavir/ritonavir, lopinavir/ritonavir, saquinavir, tipranavir/ritonavir) or strong CYP3A4 inhibitors (ritonavir, indinavir, nelfinavir). Some PIs both inhibit and induce CYP3A4 (amprenavir, fosamprenavir). Ritonavir also inhibits CYP2D6. **The NRTIs, the fusion inhibitor enfuvirtide, and the integrase** 

**or induction on the plasma concentrations of the potentially affected drug** 

pharmacokinetic DIs are likely to arise.

following three steps.

the following web sites:

are metabolized by CYP3 or CYP2 families.

http://medicine.iupui.edu/clinpharm/DDIs/ and

http://medicine.iupui.edu/clinpharm/DDIs/ClinicalTable.asp

**inhibitor raltegravir are not metabolized by the CYP450 system**.

In clinical practice, inhibition of hepatic CYP enzymes or metabolic inhibition is one of the most common DI mechanisms and it is usually reversible and competitive. Most pharmacokinetic DIs occur when a drug increases or decreases metabolism of other drugs in the liver (CYP enzymes or glucuronidation). When a drug or substance causes a decrease in the systemic hepatic metabolism and clearance of a drug, this substance generates an increase in plasma concentrations and may lead to emerge of adverse drug events or toxicities. Thus, the process of evaluating and predicting the effects of this kind of DI needs to identify the drugs that are strong or moderate inhibitors of activity of different CYP450 isozymes known also as "enzyme inhibitors for excellence".

c. Determining if the pharmacokinetic DI related to hepatic metabolism might be either one-way or two-way direction (bidirectional) and if it might affect the levels and effects of the two drugs involved; situation that is likely to occur, because the need to use simultaneously either inhibitor drugs (e.g., macrolides and antifungal azoles) or inducer drugs (for instance rifampicin and carbamazepine) together with PIs and/or NNRTIs is common.

### **2.2.2.2 Evaluation of changes in the bioavailability of drugs orally administered**

Changes in presystemic metabolism (due to inhibition or induction of extra-hepatic or hepatic CYP3A subfamily), in the activity of P-gp, or in the gastrointestinal pH may affect both the amount absorbed and the effects of some ARV drugs. In turn, antiretroviral drugs, particularly PIs, may modify the bioavailability and effects of some drugs, mainly through inhibition of presystemic metabolism and, to a lesser extent, of the activity of P-gp. Didanosine, especially in its dosage form as buffer solution, can change gastrointestinal pH and the amount absorbed by some drugs.

For drugs with kinetics elimination of first order or linear (most drugs used at therapeutic doses, phenytoin is one important exception), as it may be deduced from the equation 2, the increase or decreases of average steady-state concentration (Cpss) is inversely related to the decrease or increase of CL (Cpss1/CL). Generally, for narrow range or index drugs, a pharmacokinetic DI may be clinically relevant, if the DI causes a change in the achieved Cpss of +/-20% (FDA, 1999). As Cpss 1/CL, a decrease (associate to a reduction of hepatic metabolism) in the CL of 10%, 15%, 20%, 30%, 50%, and 75% may produce an increase in the Cpss of 11%, 18%, 25%, 43%, 100%, and 400%, respectively. The expected increasing of Cpss is calculated from the ratio of 100 divided by 100 less the percentage that CL is decreasing [100/ (100 less % of decrease of CL)]. Thus, if CL decreases in 10%, the increase of 11% is obtained from the ratio of 100 divided by 100 less 10 (100/90), which result is 111% and it is showing that Cpss increases in 11%. Similarly, 43% is obtained from the ratio of 100 divided by 100 less 30 (100/70) which result is 143% and it is showing that Cpss increases in 43%; and so similarly for other values. (Amariles, 2002) In the situations of increases of hepatic metabolism (induction enzymatic), it is requiring a minimal increase of 25% in CL, which may be caused a decrease of 20% in the achieved Cpss.

The drug plasma concentration before of the pharmacokinetic DI is another factor that contributes significantly to the magnitude of change in the clinical effect. For example, if diltiazem may decreased CL of quinidine (therapeutic range: 1-4 g/mL) in 35% (Laganiere et al., 1996), thus it may cause an increase of 54% in the achieved Cpss of quinidine. Thus, if the quinidine Cpss on time interaction was of 2.5 g/mL, it would increase to 3.85 g/mL, whereas if the quinidine Cpss was of 3.5 g mL, it would increase to 5.39 g/mL. As

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 313

**Comments and suggestions** 

especially gastrointestinal, muscular and

Recommendation: dose adjustment and

More likelihood of increasing QTc interval on the electrocardiogram and occurrence of cardiac arrhythmias, as well as dizziness Recommendation: avoid co-administration

hypertension, nervousness, hallucinations, seizures, gastrointestinal, and muscle

sedation and muscle weakness. Oxazepam, lorazepam or temazepam are an alternative, because they are eliminated by conjugation with glucuronic acid and are hardly affected by the simultaneous use of PIs. Recommendation:

Increased risk of myopathy, rhabdomyolysis, and even death. Avoid the use of lovastatin or simvastatin in patients using ritonavir, atazanavir and saquinavir. Recommendation:

interaction, such as pravastatin, fluvastatin, or

cardiac conduction. It is recommended to reduce the dose of these two drugs by half

headache, and visual disturbances. Doses should be adjusted: sildenafil to 25 mg/48 hours, tadalafil to 10 mg/72 hours, and

interval on the electrocardiogram and cardiac

disorders. Recommendation: dose adjustment

2: high risk Increased likelihood of security problems,

monitoring plasma levels

2: high risk Increased likelihood of ergotism:

and monitoring

rosuvastatin

Cisapride **a** 2: high risk Increased likelihood of increasing the QTc

2: high risk Increased risk of hypotension and reduced

2: high risk Increased risk of hypotension, priapism,

vardenafil to 2.5 mg/72 hours

2: high risk Increased likelihood of respiratory depression,

dose adjustment and monitoring

to use the lowest possible dose (for atorvastatin) and to monitor signs and symptoms of muscle toxicity, or use statins involving lower risk for this type of

cardiac conduction problems.

**Clinical relevance: level**

2: high risk

3: medium risk

**Drug group or drugs affected**

*Antiarrhythmic drugs <sup>a</sup>*

*Antihistamines anti-H1 <sup>a</sup>*

Dihydroergotamine

*Statins <sup>a</sup>* (Aberg et al., 2006; Bays, 2006; Benesic et al., 2004; Cooper et al., 2003; Fichtenbaum et al., 2002; Fichtenbaum & Gerber, 2002; Hare et al., 2002; Jacobson, 2004; Sax, 2006; Sudano et al., 2006)*.*

 Lovastatin Simvastatin Atorvastatin

 Rosuvastatin Fluvastatin Pravastatin

*Calcium antagonists nodihydro-pyridines <sup>a</sup>* Verapamil Diltiazem

*Phosphodiesterase type V <sup>a</sup>*

 Sildenafil Tadalafil Vardenafil

Terfenadine 2: high risk

Astemizole 1: very high risk

 Flecainide Disopyramide Amiodarone

*Ergot alkaloids <sup>a</sup>* Ergotamine

 Ergonovine Methylergonovine

*Benzodiazepines <sup>a</sup>* Midazolam Triazolam Alprazolam

consequence, the probability of toxicity at 5.39 g/mL (outside the therapeutic range) is higher than at 3.85 g/mL (within the therapeutic range). (Amariles, 2002)

## **2.3 Determining and predicting of the clinical relevance level (Amariles et al., 2007a)**

A relevance analysis of a DI should result in determining and predicting the clinical relevance level based on the gravity and probability of occurrence of the DI. The **probability of the DI is set to 3 categories: defined, probable, and possible**, whereas the **gravity of the DI is grouped into 3 categories: grave, moderate, and mild**. Based on the possible combinations of gravity and probability of occurrence, interactions can be grouped into 4 categories:


From a clinical perspective, predicting of the clinical relevance of a DI should be improvement by clinical experience and knowledge obtained from situations similar; thus, the clinical interpretation of the information, including drug history and the patient's clinical condition is critical. Therefore, the presence of concomitant diseases and the need to use other drugs, the condition of renal and hepatic function, as the age and nutritional condition are factors that influence the clinical relevance of a determine DI.

## **3. Clinical relevance of pharmacokinetic drug interactions in HIV-infected patients receiving antiretroviral therapy**

In this apart, using the previously described proposal, the evaluating and predicting the clinical relevance of pharmacokinetic DIs in HIV-infected patients receiving antiretroviral therapy is presented. (Amariles et al., 2007b; Giraldo et al., 2010)

#### **3.1 Drug interactions due to enzyme inhibition mediated by PIs or NNRTIs**

The process of evaluating and predicting the effects of this kind of DI needs to identify the drugs that are strong or moderate inhibitors of activity of different CYP450 isozymes known also as "enzyme inhibitors for excellence".

In general, PIs (ritonavir > indinavir ≈ nelfinavir ≈ lopinavir ≈ atazanavir ≈ amprenavir – fosamprenavir- ≈ darunavir ≈ tipranavir > saquinavir) (Boffito et al., 2006), delavirdine (Tran et al., 2001) and, in some cases, efavirenz, (DeSilva et al., 2001) can inhibit the systemic metabolism

consequence, the probability of toxicity at 5.39 g/mL (outside the therapeutic range) is

**2.3 Determining and predicting of the clinical relevance level (Amariles et al., 2007a)**  A relevance analysis of a DI should result in determining and predicting the clinical relevance level based on the gravity and probability of occurrence of the DI. The **probability of the DI is set to 3 categories: defined, probable, and possible**, whereas the **gravity of the DI is grouped into 3 categories: grave, moderate, and mild**. Based on the possible combinations of gravity and probability of occurrence, interactions can be grouped into 4

 **Level 1 (Very high risk)** resulting from the combination of: defined and grave or probable and grave. The simultaneous use of drugs is considered absolutely

 **Level 2 (High Risk)** resulting from the combinations of: possible and severe, defined and moderate, or probable and moderate. The simultaneous use of drugs is considered contraindicated: combined administration should be a*v*oided or, if it is need, the dosage regimen of affected drug may be adjusted and to assess signs and symptoms associated

 **Level 3 (Medium risk)** resulting from the combination of: possible and moderate, defined and mild, or probable and mild. The simultaneous use of drugs requires assessing signs and symptoms associated to treatment effectiveness and safety, ideally

**Level 4 (Low risk)** resulting from the combination can be mild. The interaction is of

From a clinical perspective, predicting of the clinical relevance of a DI should be improvement by clinical experience and knowledge obtained from situations similar; thus, the clinical interpretation of the information, including drug history and the patient's clinical condition is critical. Therefore, the presence of concomitant diseases and the need to use other drugs, the condition of renal and hepatic function, as the age and nutritional condition are factors that influence the clinical relevance of a determine DI.

**3. Clinical relevance of pharmacokinetic drug interactions in HIV-infected** 

**3.1 Drug interactions due to enzyme inhibition mediated by PIs or NNRTIs** 

In this apart, using the previously described proposal, the evaluating and predicting the clinical relevance of pharmacokinetic DIs in HIV-infected patients receiving antiretroviral

The process of evaluating and predicting the effects of this kind of DI needs to identify the drugs that are strong or moderate inhibitors of activity of different CYP450 isozymes known

In general, PIs (ritonavir > indinavir ≈ nelfinavir ≈ lopinavir ≈ atazanavir ≈ amprenavir – fosamprenavir- ≈ darunavir ≈ tipranavir > saquinavir) (Boffito et al., 2006), delavirdine (Tran et al., 2001) and, in some cases, efavirenz, (DeSilva et al., 2001) can inhibit the systemic metabolism

higher than at 3.85 g/mL (within the therapeutic range). (Amariles, 2002)

to treatment effectiveness and safety, ideally in a quantitative form.

categories:

contraindicated.

in a quantitative form.

little clinical relevance.

**patients receiving antiretroviral therapy** 

also as "enzyme inhibitors for excellence".

therapy is presented. (Amariles et al., 2007b; Giraldo et al., 2010)


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 315

**PIs Clinical** 

*Ritonavir, nelfinavir* 

*ATV, PV, DRV, SQV, FPV* 

*LPV/r* 3: medium

risk

**relevance: level** 

3: medium risk

drugs, which can lead to increased levels and toxicity of SSRIs (serotonin syndrome). It is recommended using half or the lowest dose possible of either drug, setting the dose in terms of their effectiveness and safety. Citalopram and possibly escitalopram and sertraline, due to their low metabolism and effect on CYP activity, are considered the most suitable option in HIV-infected patients

receiving antiretroviral therapy

Recommendation: reduce the dose of maraviroc up to 50% and monitoring

**Comments and suggestions** 

increase in the levels and toxicity simvastatin increases the risk of rhabdomyolysis and acute renal failure. Recommendation: dose adjustment and monitoring

Increased metabolism of drugs metabolized by CYP2B6, CYP2C8, CYP2C9, CYP2C19. Recommendation:

2: high risk PIs increase plasma levels of maraviroc.

*Atazanavir* 2: high risk Inhibition of CYP3A4 by ATV, the

monitoring

2: high risk The coadministration of TDF with

ATV, LPV, DRV and SQV increased Cp of tenofovir, while decreases Cp of

LPV/r may increase the metabolism of

stimulation of CYP2C9, as well as the R enantiomer by stimulation of

FPV. Recommendation: dose adjustment and monitoring

S enantiomer of warfarin by

2008; Acosta, 2002; Caballero & Nahata 2005; Currier et al., 2004; de Maat et al., 2003; Isbister & Buckley, 2005; Tseng &

Integrase inhibitors *Maraviroc* (Aberg, 2008)

**With specific PIs Drug group or drugs affected** 

Simvastatin (Schmidt et al.,

 Efavirenz Paclitaxel Losartan Diclofenac Phenytoin Amitriptyline Omeprazole Fluoxetine Warfarin Iibuprofen Glibenclamide (Dixit et al, 2007)

Tenofovir (Tong et al., 2007)

Warfarin (Hughes et al., 2007)

2007)

Foisy, 1999) Fluoxetine Fluvoxamine Venlafaxine Paroxetine Sertraline


monitoring

monitoring

Pimozide **a** 2: high risk Increased likelihood of involuntary

2: high risk

2: high risk

Warfarina (acenocumarol)

Fluticasone **a** (Arrington-Sanders et al., 2006)

*Opioid analgesics (*Armstrong & Cozza, 2003a, *2003b)* Oxycodone

 Oxycodone Buprenorphine

 Dihydrocodeine Hydrocodone

*Tricyclic and tetracyclic antidepressants* (Cvetkovic et al., 2003; de Maat et al., 2003; Oldfield & Plosker, 2006; Von et al., 1998) Amitriptyline Desipramine Nortriptyline Mirtazapine Trazodone Nefazodone

*Antidepressants reuptake inhibitors (SSRI)* (Aberg,

Codeine 3: medium risk

arrhythmias, as well as gastrointestinal disturbances and dizziness. Recommendation:

movements (tics), agitation, confusion, behavioral disturbances, and tachycardia. Recommendation: Dose adjustment and

combination is used. Recommendation: dose adjustment and monitoring international

PIs, tipranavir, and indinavir may increase fluticasone levels, and even lead to the

with the use of ritonavir in children. Recommendation: Dose adjustment and

CYP3A4 and thus the metabolism of oxycodone and buprenorphine, increasing their plasma concentrations as well as the likelihood of toxic effects (sedation and

development of Cushing syndrome, especially

Protease inhibitors may inhibit CYP2D6 and

Dihydrocodeine, hydrocodone, and codeine are pro-drugs and require activation by CYP2D6 or glucuronyltransferase. Therefore, protease inhibitors can inhibit the metabolism and the formation of the active compound and, therefore, the pharmacological effect of these drugs. However, codeine is considered a suitable option for pain control in HIV-infected patients receiving antiretroviral therapy

CYP2D6, and thus, the metabolism of these drugs, which can generate toxicity problems, particularly decreased conduction and cardiac arrest, as well as increased anticholinergic effects (constipation, dry mouth, urinary retention) and cardiac abnormalities. It is recommended to reduce the dose by half (or use the lowest dose possible) of these two drugs, with adjustments based on their

CYP2D6, and thus, the metabolism of these

Dose adjustment and monitoring

2: high risk The risk of bleeding may increase, when this

normalized ratio (INR)

respiratory depression)

2: high risk Ritonavir alone or with another PI can inhibit

effectiveness and safety

2: high risk Ritonavir alone or with another PI can inhibit

2: high risk PIs, especially ritonavir alone or with other


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 317

2: high risk

3: medium risk

3: medium risk

risk

risk

a Protease inhibitors (ritonavir → indinavir ≈ nelfinavir ≈ atazanavir≈ amprenavir- fosamprenavir- ≈ tipranavir → saquinavir), mainly through inhibition of CYP3A4, may decrease the metabolism of these

Table 2. General interactions due to enzyme inhibition by protease inhibitors (Boffito et al., 2006; Busti et al., 2004; DeSilva et al., 2001; DHHS, 2011; Krikorian & Rudorf, 2005; Piscitelli & Gallicano, 2001; Kashuba, 2005a; Robertson et al., 2005a; Tran et al., 2001; Winston &

of several drugs, increase their plasma levels, and may cause adverse drug reactions, which could cause grave health problems in patients. Thus, according to their clinical relevance, most

due to the inhibitory effect of ritonavir. Recommendation: dose adjustment and monitoring

may be increased, due to the inhibitory effect of PIs.

cannabinoids and marihuana derivates

Recommendation: dose adjustment and monitoring, or avoid use of

methadone may be increased, due to inhibitory effect of PIs on CYP3A4. Recommendation: dose adjustment of methadone and/or monitoring adverse effects of methadone

quetiapine are increased, due to inhibitory effect of atazanavir. Recommendation: dose adjustment of

Plasma concentrations and effects of

quetiapine and monitoring

these antineoplastics may be increased, due to inhibitory effect of

Recommendation: monitoring

LPV/r on CYP3A5.

2: high risk Plasma concentrations and effects of

2: high risk Plasma concentrations and effects of

cannabinoids

*ATV* 2: high risk Plasma concentrations and effects of

 Docetaxel Fluticasone Oxycodone Alprazolam Sirolimus Quetiapine

Cannabinoids (Abrams et al.,

Quetiapine (Hantson et al.,

Antineoplasics (Levêque et al., 2009; Makinson et

Boffito, 2005; Wire et al., 2006)

al., 2010) Irinotecan Vinblastine Vincristine

2010)

Methadone *LPV/r* 

2003)

2009; Gray et al., 2010; Gruber & McCance-Katz,

(McCance-Katz et al., 2003)

(McCance-Katz et al., 2004)

*ATV* (Friedland et al, 2005)

*FPV* (Cao et al.,

*IDV* 3: medium

*LPV/r* 3: medium

drugs, which can cause increased plasma concentrations and toxicity.

2010)

*IDV NFV RTV* 

*NFV* 

2008)


*ATV* 3: medium

*ATV/r* 1: very high

*Tipranavir* 1: very high

*FPV/r* 1: very high

*DRV/r* 3: medium

risk

risk

risk

risk

risk

2: high risk

risk

risk

Minociline (DiCenzo et al.,

Etravirine (Aberg,

2008)

2008)

Raltegravir (Aberg, 2008)

 Pravastatin Sildenafil Vardenafil Tadalafil

 Budesonide Quinine

Darunavir *SQV*(Sekar et

al., 2006a)

*LPV* (Sekar et al., 2006b)

*DRV* (Sekar et al., 2008)

*RTV and LPV/r* (Daveluy et al,

*IDV* 3: medium

*NVP* 3: medium

CYP1A2. Recommendation: INR

the intestinal bacterial flora. Recommendation: monitoring

co-administration.

co-administration

*LPV/r* 2: high risk Etravirine Cp may be increased by

*NFV* 2: high risk Concomitant use increase plasma

*Tipranavir* 2: high risk Plasma concentrations of raltegravir

*ATV* 2: high risk Plasma concentrations of raltegravir

Avoid co-administration

85%. Recommendation: dose adjustment and monitoring

Decreased levels and effect of ATV, possibly due to enterohepatic cycle interference associated to alterations of

ATV Cp is decreased by 38%. Avoid

Etravirine Cp was reduced by 75%.

FPV Cp is decreased by 77%. Avoid

Etravirine Cp may be decreased by 50%. Recommendation: monitoring

are reduced when it is used with tipranavir. Recommendation: dose

are increased when it is used with ATV. Recommendation: dose adjustment and monitoring

these PIs may be increased, due to the

Recommendation: dose adjustment and

these medications may be increased, due to the inhibitory effect of darunavir. Recommendation: dose adjustment and/or monitoring

these medications may be increased,

inhibitory effect of darunavir.

adjustment and monitoring

2: high risk Plasma concentrations and effects of

2: high risk Plasma concentrations and effects of

2: high risk Plasma concentrations and effects of

monitoring

concentrations of nelfinavir, due to the inhibitory effect of etravirine on CYP3A4. Recommendation: dose adjustment and monitoring (Schöller et al., 2006a; Sekar et al., 2006a)

monitoring


a Protease inhibitors (ritonavir → indinavir ≈ nelfinavir ≈ atazanavir≈ amprenavir- fosamprenavir- ≈ tipranavir → saquinavir), mainly through inhibition of CYP3A4, may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity.

Table 2. General interactions due to enzyme inhibition by protease inhibitors (Boffito et al., 2006; Busti et al., 2004; DeSilva et al., 2001; DHHS, 2011; Krikorian & Rudorf, 2005; Piscitelli & Gallicano, 2001; Kashuba, 2005a; Robertson et al., 2005a; Tran et al., 2001; Winston & Boffito, 2005; Wire et al., 2006)

of several drugs, increase their plasma levels, and may cause adverse drug reactions, which could cause grave health problems in patients. Thus, according to their clinical relevance, most

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 319

information available on clinically relevant interactions

induces warfarin metabolism

Cisapride *<sup>a</sup>* 2: high risk 2: high risk More likelihood of increasing QTc

a Delavirdine, primarily through inhibition of CYP3A4, and efavirenz through inhibition of CYP3A4 and

CYP2D6, may decrease the metabolism of these drugs, which can cause increased plasma

Table 3. General interactions due to enzyme inhibition by inhibitors of non-nucleoside transcriptase (DeMaat et al., 2003; DHHS, 2011; Krikorian & Rudorf, 2005; Piscitelli &

INR

dizziness

efavirenz

2: high risk The Cp of raltegravir is reduced

Inhibition of CYP3A4 and metabolism of oxycodone and buprenorphine, increasing plasma concentrations and the likelihood of toxic effects (sedation and respiratory depression)

activation by CYP2D6 or

effect of these drugs

2: high risk Efavirenz, in cases of deficiency of

Dihydrocodeine, hydrocodone, and codeine are pro-drugs and require

glucuronyl-transferase. Therefore, delavirdine may inhibit the metabolism and the formation of the active compound and,

consequently, the pharmacological

CYP2D6, may inhibit CYP2C9 and CYP2C19, which can lead to increased levels and toxicity of SSRIs (serotonin syndrome). We recommend using half or the lowest dose possible of these two drugs, adjusting the dose in terms of their effectiveness and safety

With delavirdine it increases the risk of bleeding. It is recommended dose adjustment and monitoring of

interval on the electrocardiogram and cardiac arrhythmias, as well as gastrointestinal disturbances and

when is co-administered with

No

2: high risk

3: medium risk

No

Warfarin*a (acenocumarol)* 2: high risk Efavirenz

information available on clinically relevant interactions with delavirdine

information available

*Opioid analgesics*  (Armstrong & Cozza,

2003a, 2003b)

 Oxycodone Buprenorphine

 Dihydrocodeine Hydrocodone Codeine

*Antidepressants reuptake inhibitors (SSRI)*  (Caballero & Nahata, 2005; Currier et al., 2004; DeSilva et al., 2001; Isbister & Buckley, 2005; Tseng & Foisy, 1999) Fluoxetine Fluvoxamine Venlafaxine Paroxetine Sertraline

Raltegravir (Aberg, 2008) No

Gallicano, 2001; Tran et al., 2001)

concentrations and toxicity.

could be classified in level 1 or 2. Table 2 lists the interactions mediated by enzyme inhibition caused by PIs. Table 3 contains the interactions mediated by delavirdine and efavirenz.


could be classified in level 1 or 2. Table 2 lists the interactions mediated by enzyme inhibition

1: very high

Efavirenz acts as an inducer of statins metabolism

interactions induction)

Information

2: high risk 2: high risk It increases the risk of hypotension,

(see

not available

schedule of these two drugs Diltiazem 3: medium

risk

**Clinical relevance: level Comments and suggestions** 

2: high risk 2: high risk Increased likelihood of ergotism:

2: high risk 2: high risk Increased likelihood of respiratory

co-administration

hypertension, nervousness, hallucinations, seizures, gastrointestinal and muscle disorders. Recommendation: dose adjustment and monitoring

depression, sedation and muscle weakness. Oxazepam, lorazepam or temazepam are an alternative, because they are eliminated by conjugation with glucuronic acid and are hardly affected by the

simultaneous use of PI

with PIs or delavirdine)

Especially with delavirdine, it increases the risk of hypotension and reduced cardiac conduction. Consider to use the half dosing

priapism, headache, and visual disturbances. Doses should be adjusted: sildenafil to 25 mg/48 hours, tadalafil to 10 mg/72 hours, and vardenafil to 2.5 mg/72 hours

Increased risk of myopathy, rhabdomyolysis, and even death. It should be avoided in patients using delavirdine (see text: 3.1.1 statins

More likely to increase QTc interval on the electrocardiogram and cardiac arrhythmias, as well as dizziness. Recommendation: avoid

caused by PIs. Table 3 contains the interactions mediated by delavirdine and efavirenz.

Delavirdine Efavirenz

Terfenadine 2: high risk 2: high risk

risk

2: high risk

3: medium risk

risk

Astemizole 1: very high

**Drug group or drugs affected** 

*Anti-H1 antihistamines<sup>a</sup>*

Dihydroergotamine

*Ergot alkaloids <sup>a</sup>* Ergotamine

 Ergonovine Methylergonovine

*Benzodiazepines <sup>a</sup>* Midazolam Triazolam Alprazolam

*Statins* (Bays, 2006; Cooper et al., 2003; Fichtenbaum & Gerber, 2002; Jacobson, 2004; Sax, 2006; Sudano et al., 2006.)

 Lovastatin Simvastatin Atorvastatin

 Rosuvastatin Fluvastatin Pravastatin

*Calcium antagonists not dihydro-pyridine <sup>a</sup>*

*Phosphodiesterase Type V* 

*inhibitors<sup>a</sup>* Sildenafil Tadalafil Vardenafil

Verapamil 2: high risk


a Delavirdine, primarily through inhibition of CYP3A4, and efavirenz through inhibition of CYP3A4 and CYP2D6, may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity.

Table 3. General interactions due to enzyme inhibition by inhibitors of non-nucleoside transcriptase (DeMaat et al., 2003; DHHS, 2011; Krikorian & Rudorf, 2005; Piscitelli & Gallicano, 2001; Tran et al., 2001)

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 321

On the other hand, it seems that the combination atazanavir with lopinavir/ritonavir does not generate this problem (although decreased levels of lopinavir by the action of atazanavir may occur (Colombo et al., 2006), which is considered a double pharmacokinetic extension (boosted) with PI, useful in a certain group of patients. (Ribera et al., 2006) Contrary to this effect (of atazanavir on lopinavir levels), when using a saquinavir/ritonavir and atazanavir (1.600/100 and 200 mg) an increase in saquinavir plasma and cellular levels is observed (without effect on ritonavir). Therefore, it is believed that adding atazanavir 200 mg/day to saquinavir/ritonavir 1.600/100 mg/day may be a good strategy for patients in who the

**Due to it is common used of this group of drugs in HIV infected patients receiving** 

**Statins.** The simultaneous use of PI or delavirdine with statins (lovastatin≈ simvastatin > atorvastatin > rosuvastatin ≈ fluvastatin ≈ pravastatin) increases the risk of myopathy, rhabdomyolysis, and even death. (Hare et al., 2002) Thus, for atorvastatin the lowest possible doses are recommended. Also, it is suggested to monitor muscle toxicity signs and symptoms or to use statins less likely to have this type of interaction, such as pravastatin, fluvastatin or rosuvastatin from. (Benesic et al., 2004; DHHS, 2011; Fichtenbaum et al., 2002; Fichtenbaum & Gerber, 2002; Jacobson, 2004; Sax, 2006) In this regard, statins (simvastatin, lovastatin, and atorvastatin, except pravastatin, fluvastatin, and rosuvastatin) are metabolized by CYP3A4 and, therefore, their use should be avoided in patients using PIs, especially ritonavir, atazanavir and saquinavir. Both metabolism and levels of pravastatin, as well as, most likely, fluvastatin and rosuvastatin are slightly affected by the combined use of ritonavir, indinavir, atazanavir, saquinavir, and nelfinavir and therefore they could be combined in patients receiving cART or HAART. (Aberg et al., 2006; Bays, 2006; Benesic et al., 2004; Cooper et al., 2003; Jacobson, 2004; Sudano et al., 2006). However, in some cases it may be necessary to increase the dose of pravastatin, as appears likely in the case of

**Selective serotonin reuptake inhibitors (SSRI).** The use of ritonavir (alone or in combination with another PI) by inhibiting CYP2D6 and efavirenz by inhibiting CYP2C9 and CYP2C19, when a deficit of CYP2D6 exists, may lead to increased levels and toxic effects of SSRIs (fluoxetine, fluvoxamine, venlafaxine, paroxetine and, sertraline). Particularly, it may lead to the development of serotonin syndrome (DeSilva et al., 2001). Therefore, it is advisable to use half the dose or the lowest dose possible of these drugs, setting the dose in terms of their effectiveness and safety. (DHHS, 2011) Additionally, citalopram, and possibly escitalopram and sertraline, due to their pharmacological properties (low metabolism and minimal effect on CYP activity), are considered the most suitable option in patients receiving cART or HAART.

Generally, serotonin syndrome is a disorder caused by an increase in serotonin levels, resulting in **cognitive disorders** (lethargy, confusion, coma, agitation, hallucinations, and seizures); **neuromuscular activity disorders** (myoclonus, tremor, hyperreflexia, rigidity, hyperactivity); **autonomic nervous system disorders** (hypotension or hypertension, tachycardia, chills, hyperthermia, diaphoresis, diarrhea, salivation, abdominal pain, tachypnea) (Isbister & Buckley, 2005). In a patient under treatment with a SSRI (or other drugs with serotonergic activity in the central nervous system), such change can occur due to pharmacodynamic interactions (simultaneous treatment with other drugs that increase serotonin activity or agonist) or to pharmacokinetic interactions (treatment with an CYP2D6

(Caballero & Nahata, 2005; Currier et al., 2004; DHHS, 2011; Kashuba, 2005)

**antiretroviral therapy; it is important focus following drugs therapeutic class:** 

Cpss are under the minimum effective one. (Ford et al., 2006)

nelfinavir. (Aberg et al., 2006)

From a practical perspective and specifying the type of PI, it is generally recommended to avoid the following combinations (level 1 or level 2: very high risk or high risk): (DHHS, 2011)


From a practical perspective and specifying the type of NNRTIs, it is generally recommended to avoid the following combinations (level 1 or level 2: very high risk or high risk): (DHHS, 2011)


Related to PIs and NNRTIs Di, it is important to note that for delavirdine, its combination with amprenavir or fosamprenavir can cause a decrease in the levels and effects of delavirdine. (Tran et al., 2001; Wire et al., 2006) Additionally, some studies have shown a possible interaction between fosamprenavir with lopinavir/ritonavir, in fact mediated by a two-way increase in CYP3A4 and/or P-glycoprotein (P-gp) metabolic activity by fosamprenavir and lopinavir. (Huang et al., 2001; Taburet et al., 2004; Tranet al., 2002) This interaction is associated to a 50-60% decrease in lopinavir plasma concentration, and a 60- 70% decrease in amprenavir levels (ritonavir levels do not change significantly). Consequently, combining these 3 PIs may be inappropriate, due to a high probability of generating a virologic failure (Kashuba et al., 2005) (level 2: high risk). The negative effect of this interaction is not neutralized by increasing ritonavir dose from 100 to 200 mg/12 hours; while this strategy is associated with increased gastric problems associated to the use of ritonavir. (Mauss et al., 2004)

From a practical perspective and specifying the type of PI, it is generally recommended to avoid the following combinations (level 1 or level 2: very high risk or high risk): (DHHS, 2011) Indinavir with: atazanavir, simvastatin or lovastatin, amiodarone, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine, or methyl

 Ritonavir (alone or with another PI) with: voriconazole (with a ritonavir dose higher than 400 mg/12 hours), fluticasone, simvastatin or lovastatin, amiodarone, flecainide, propafenone, or quinidine, cisapride, pimozide or clozapine, trazodone or nefazodone , astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine or

 Saquinavir with: simvastatin or lovastatin, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine, or methyl ergonovine. Lopinavir/ritonavir with: fluticasone, simvastatin or lovastatin, flecainide or propafenone, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam,

 Nelfinavir with indinavir, irinotecan, simvastatin or lovastatin, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine or

 Atazanavir with: simvastatin or lovastatin, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine, or methyl ergonovine. Amprenavir (same for the fosamprenavir) with: simvastatin or lovastatin, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine,

 Tipranavir with: fluticasone, simvastatin or lovastatin, amiodarone, flecainide, propafenone, or quinidine, cisapride, pimozide, astemizole or terfenadine, midazolam or triazolam, ergotamine, ergonovine, or methyl-ergonovina.( Boffito et al., 2006) From a practical perspective and specifying the type of NNRTIs, it is generally recommended to avoid the following combinations (level 1 or level 2: very high risk or high

 Delavirdine with: amprenavir or fosamprenavir, simvastatin or lovastatin, cisapride, astemizole or terfenadine, alprazolam, midazolam or triazolam, ergotamine, ergonovine

Efavirenz with: voriconazole, cisapride, astemizole or terfenadine, midazolam or

Related to PIs and NNRTIs Di, it is important to note that for delavirdine, its combination with amprenavir or fosamprenavir can cause a decrease in the levels and effects of delavirdine. (Tran et al., 2001; Wire et al., 2006) Additionally, some studies have shown a possible interaction between fosamprenavir with lopinavir/ritonavir, in fact mediated by a two-way increase in CYP3A4 and/or P-glycoprotein (P-gp) metabolic activity by fosamprenavir and lopinavir. (Huang et al., 2001; Taburet et al., 2004; Tranet al., 2002) This interaction is associated to a 50-60% decrease in lopinavir plasma concentration, and a 60- 70% decrease in amprenavir levels (ritonavir levels do not change significantly). Consequently, combining these 3 PIs may be inappropriate, due to a high probability of generating a virologic failure (Kashuba et al., 2005) (level 2: high risk). The negative effect of this interaction is not neutralized by increasing ritonavir dose from 100 to 200 mg/12 hours; while this strategy is associated with increased gastric problems associated to the use of

triazolam, ergotamine, ergonovine or methyl ergonovine.

ergonovine.

methylergonovine.

methylergonovine.

risk): (DHHS, 2011)

or methyl ergonovine.

ritonavir. (Mauss et al., 2004)

ergotamine, ergonovine, or methyl ergonovine.

or methyl-ergonovina. ( Wire et al., 2006)

On the other hand, it seems that the combination atazanavir with lopinavir/ritonavir does not generate this problem (although decreased levels of lopinavir by the action of atazanavir may occur (Colombo et al., 2006), which is considered a double pharmacokinetic extension (boosted) with PI, useful in a certain group of patients. (Ribera et al., 2006) Contrary to this effect (of atazanavir on lopinavir levels), when using a saquinavir/ritonavir and atazanavir (1.600/100 and 200 mg) an increase in saquinavir plasma and cellular levels is observed (without effect on ritonavir). Therefore, it is believed that adding atazanavir 200 mg/day to saquinavir/ritonavir 1.600/100 mg/day may be a good strategy for patients in who the Cpss are under the minimum effective one. (Ford et al., 2006)

#### **Due to it is common used of this group of drugs in HIV infected patients receiving antiretroviral therapy; it is important focus following drugs therapeutic class:**

**Statins.** The simultaneous use of PI or delavirdine with statins (lovastatin≈ simvastatin > atorvastatin > rosuvastatin ≈ fluvastatin ≈ pravastatin) increases the risk of myopathy, rhabdomyolysis, and even death. (Hare et al., 2002) Thus, for atorvastatin the lowest possible doses are recommended. Also, it is suggested to monitor muscle toxicity signs and symptoms or to use statins less likely to have this type of interaction, such as pravastatin, fluvastatin or rosuvastatin from. (Benesic et al., 2004; DHHS, 2011; Fichtenbaum et al., 2002; Fichtenbaum & Gerber, 2002; Jacobson, 2004; Sax, 2006) In this regard, statins (simvastatin, lovastatin, and atorvastatin, except pravastatin, fluvastatin, and rosuvastatin) are metabolized by CYP3A4 and, therefore, their use should be avoided in patients using PIs, especially ritonavir, atazanavir and saquinavir. Both metabolism and levels of pravastatin, as well as, most likely, fluvastatin and rosuvastatin are slightly affected by the combined use of ritonavir, indinavir, atazanavir, saquinavir, and nelfinavir and therefore they could be combined in patients receiving cART or HAART. (Aberg et al., 2006; Bays, 2006; Benesic et al., 2004; Cooper et al., 2003; Jacobson, 2004; Sudano et al., 2006). However, in some cases it may be necessary to increase the dose of pravastatin, as appears likely in the case of nelfinavir. (Aberg et al., 2006)

**Selective serotonin reuptake inhibitors (SSRI).** The use of ritonavir (alone or in combination with another PI) by inhibiting CYP2D6 and efavirenz by inhibiting CYP2C9 and CYP2C19, when a deficit of CYP2D6 exists, may lead to increased levels and toxic effects of SSRIs (fluoxetine, fluvoxamine, venlafaxine, paroxetine and, sertraline). Particularly, it may lead to the development of serotonin syndrome (DeSilva et al., 2001). Therefore, it is advisable to use half the dose or the lowest dose possible of these drugs, setting the dose in terms of their effectiveness and safety. (DHHS, 2011) Additionally, citalopram, and possibly escitalopram and sertraline, due to their pharmacological properties (low metabolism and minimal effect on CYP activity), are considered the most suitable option in patients receiving cART or HAART. (Caballero & Nahata, 2005; Currier et al., 2004; DHHS, 2011; Kashuba, 2005)

Generally, serotonin syndrome is a disorder caused by an increase in serotonin levels, resulting in **cognitive disorders** (lethargy, confusion, coma, agitation, hallucinations, and seizures); **neuromuscular activity disorders** (myoclonus, tremor, hyperreflexia, rigidity, hyperactivity); **autonomic nervous system disorders** (hypotension or hypertension, tachycardia, chills, hyperthermia, diaphoresis, diarrhea, salivation, abdominal pain, tachypnea) (Isbister & Buckley, 2005). In a patient under treatment with a SSRI (or other drugs with serotonergic activity in the central nervous system), such change can occur due to pharmacodynamic interactions (simultaneous treatment with other drugs that increase serotonin activity or agonist) or to pharmacokinetic interactions (treatment with an CYP2D6

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 323

**Comments and suggestions** 

bidirectional increase in the levels of antifungal and PIs, which can lead to toxicity problems. It is recommended using up to 200 mg/day of

ketoconazole or 200-400 mg/day of itraconazole. In the case of fluconazole combination with tipranavir / ritonavir an increase in levels of tipranavir, without significant changes in fluconazole is observed.

The coadministration of darunavir and ketoconazole increase the plasma concentrations of both drugs, creating the same toxic effects (Sekar et al., 2008)

bidirectional increase in the levels of macrolides and PIs is generated, which can lead to toxicity problems.

clarithromycin it is suggested to reduce 50-75% in the level, if the patient has a creatinine clearance <60 ml / minute). In general, it is considered that using erythromycin with a strong inhibitors of CYP3A4

Increased levels and toxicity of immunosuppressants can be generated, so it is recommended to monitor plasma levels and to adjust the treatment regime. In the case of nelfinavir/tacrolimus combination it is recommended to reduce by 50% the dose of the immunosuppressant (Jain et al., 2002). In the case of cyclosporine with lopinavir / ritonavir, a 5% initial reduction is recommended in cyclosporine dose (a reduction of up to 20%) may be necessary. (Vogel et al., 2004) Cyclosporine and tacrolimus may increase the levels and toxicity of PIs and therefore it is recommended to monitor signs and symptoms of toxicity, and if required, to adjust the dose

Except for amprenavir and fosamprenavir, a

Macrolides may increase QTc interval on the electrocardiogram. It is recommended to use maximum 1 g / day of these antibiotics (for

should be avoided (Ray et al., 2004)

a PIs can inhibit CYP3A4 and may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity. In turn, these drugs, also due to CYP3A4 inhibition, may decrease

Table 4. Clinical relevant bidirectional drug interactions mediated by PIs enzyme inhibition with other known drugs (Brophy et al., 2000; DHHS, 2011; Kashuba, 2005a, 2005b; Krikorian

2: high risk Except for amprenavir and fosamprenavir, there is a

(Vourvahis & Kashuba, 2007)

**Group drugs or drugs affected** 

*Azole antifungals <sup>a</sup>* (Polk et al., 1999; Vourvahis & Kashuba, 2007) Ketoconazole Itraconazole Fluconazole

Voriconazole 3: medium risk

2: high risk

2: high risk

the metabolism of PIs and increase plasma concentrations and toxicity.

& Rudorf, 2005; Robertson et al., 2005b; Young, 2005)

 

Macrolides **<sup>a</sup>** Erythromycin Clarithromycin

Immunosuppressants **<sup>a</sup>** (Jain et al., 2002; Vogel et al., 2004) Cyclosporine

Tacrolimus

**Clinical relevance of the bidirectional inhibition: level**

inhibitor or because of a situation leading to an increase in serotonin levels and effects). (DeSilva et al., 2001; Tseng & Foisy, 1999)

#### **3.2 Drug interactions mediated mainly by enzymatic inhibition of PIs and NNRTIs with other known enzyme inhibitors (for some drugs metabolism inhibition may occur while for others may occur metabolism induction)**

Since PIs and NNRTIs are metabolized by CYP3A, strong inhibitors of CYP3A isoenzyme may inhibit its metabolism, increase plasma levels and therefore the risk of developing adverse drug reactions or toxicities. In these cases, generally, inhibition is bidirectional (the strong inhibitors may increase the PIs and NNRTIs levels and toxicity as well as PIs and NNRTIs may increase the inhibitors plasma concentrations and toxicity). However, in some cases the interaction can be in one direction, while in others both NNRTI metabolism inhibition (by the known inhibitor) and known inhibitor metabolism induction (by the NNRTIs) may occur, especially with efavirenz and nevirapine. In table 4 (for PIs) and table 5 (for NNRTIs) the most relevant interactions of this type are included.

## Due to it is important some examples of the **one-way drug interactions mediated mainly by enzymatic inhibition of PIs and NNRTIs are detailed:**

#### **Azoles antifungals**


#### **Macrolides and immunosuppressants**


#### **3.3 Drug interactions due to enzyme induction of NNRTIS (Back et al., 2003; Bergshoeff et al., 2005; DHHS, 2011; Kashuba, 2005; Krikorian & Rudorf, 2005; Mildvan et al., 2002; (Pérez et al., 2009; Piscitelli & Gallicano, 2001; Saraga et al., 2006; Young, 2005)**

Efavirenz and nevirapine due to their ability to increase the hepatic metabolism may cause a decrease in PIs levels and consequently therapeutic failure. Therefore, in general, using a single PI with efavirenz (in particular) or nevirapine is considered contraindicated (level 2 interaction: high risk). For example, efavirenz can cause a 39% reduction in the minimum

inhibitor or because of a situation leading to an increase in serotonin levels and effects).

**3.2 Drug interactions mediated mainly by enzymatic inhibition of PIs and NNRTIs with other known enzyme inhibitors (for some drugs metabolism inhibition may occur** 

Since PIs and NNRTIs are metabolized by CYP3A, strong inhibitors of CYP3A isoenzyme may inhibit its metabolism, increase plasma levels and therefore the risk of developing adverse drug reactions or toxicities. In these cases, generally, inhibition is bidirectional (the strong inhibitors may increase the PIs and NNRTIs levels and toxicity as well as PIs and NNRTIs may increase the inhibitors plasma concentrations and toxicity). However, in some cases the interaction can be in one direction, while in others both NNRTI metabolism inhibition (by the known inhibitor) and known inhibitor metabolism induction (by the NNRTIs) may occur, especially with efavirenz and nevirapine. In table 4 (for PIs) and table 5

Due to it is important some examples of the **one-way drug interactions mediated mainly by** 

 Azoles antifungals (ketoconazole, itraconazole, fluconazole, voriconazole) with amprenavir. Antifungal may inhibit amprenavir hepatic metabolism (and probably fosamprenavir metabolism), which can lead to an increase in their levels, although such increase does not seem to be clinically relevant (level 3: medium risk). (Vourvahis &

 Fluconazole with tipranavir/ritonavir. Fluconazole causes an increase in the area under the curve. In any case, similar to what happens when PIs are used with azoles antifungals; it is advisable to monitor the hepatic function and not to exceed a dose of

 Ketoconazole with delavirdine. Ketoconazole may increase delavirdine levels (Level 3: medium risk), (DHHS, 2011) but ketoconazole levels do not change significantly.

 Clarithromycin with amprenavir (and probably fosamprenavir). Clarithromycin produces an 18% an increase in the area under the curve of amprenavir, while the effect

 Immunosuppressants with efavirenz or nevirapine. Cyclosporine and tacrolimus may increase the levels and toxicity of these two NNRTIs. It is thus recommended to monitor toxicity signs and symptoms of these drugs and, if required, to adjust its dose

Efavirenz and nevirapine due to their ability to increase the hepatic metabolism may cause a decrease in PIs levels and consequently therapeutic failure. Therefore, in general, using a single PI with efavirenz (in particular) or nevirapine is considered contraindicated (level 2 interaction: high risk). For example, efavirenz can cause a 39% reduction in the minimum

(DeSilva et al., 2001; Tseng & Foisy, 1999)

**while for others may occur metabolism induction)** 

**enzymatic inhibition of PIs and NNRTIs are detailed:** 

200 mg/daily fluconazole. (DHHS, 2011)

of PI on macrolide levels is lowest. (Brophy et al., 2000)

(Level 3: medium risk). (Jain et al., 2002; Vogel et al., 2004)

**3.3 Drug interactions due to enzyme induction of NNRTIS (Back et al., 2003; Bergshoeff et al., 2005; DHHS, 2011; Kashuba, 2005; Krikorian & Rudorf, 2005; Mildvan et al., 2002; (Pérez et al., 2009; Piscitelli & Gallicano, 2001; Saraga et al.,** 

**Macrolides and immunosuppressants** 

**Azoles antifungals** 

Kashuba, 2007)

**2006; Young, 2005)**

(for NNRTIs) the most relevant interactions of this type are included.


a PIs can inhibit CYP3A4 and may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity. In turn, these drugs, also due to CYP3A4 inhibition, may decrease the metabolism of PIs and increase plasma concentrations and toxicity.

Table 4. Clinical relevant bidirectional drug interactions mediated by PIs enzyme inhibition with other known drugs (Brophy et al., 2000; DHHS, 2011; Kashuba, 2005a, 2005b; Krikorian & Rudorf, 2005; Robertson et al., 2005b; Young, 2005)

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 325

2: high risk No

a NNRTIs may inhibit CYP3A4 and may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity. In turn, these drugs, also due to CYP3A4 inhibition, may

Table 5. Clinical relevant bidirectional drug interactions mediated by enzyme inhibition (induction in some cases) of NNRTIs with other known enzyme inhibitors (Brophy et al., 2000; Jain et al., 2002; Krikorian & Rudorf, 2005; Piscitelli & Gallicano, 2001; Polk et al., 1999; Ray et al., 2004; Robertson et al., 2005b; Sekar et al., 2008; Vogel et al., 2004; Vourvahis &

lopinavir steady state concentration. (Dailly et al., 2005; Solas et al., 2004) The use of another low-dose PI as a pharmacokinetic extension agent (boosted), for instance ritonavir, is a strategy to counter this problem. (Acosta, 2002; Rathbun & Rossi, 2002; Zeldin & Petruschke, 2004) When using atazanavir with efavirenz it is recommended to add 100 mg of ritonavir to the usual dose (300 mg/24 hours). For lopinavir/ritonavir and efavirenz, lopinavir/ritonavir should be increased from 400/100 mg/12 hours (3 capsules) to 533/133 mg/12 hours (4 capsules) without change in the efavirenz dose. (DHHS, 2011; Solas et al., 2004; Dailly et al., 2005). For pediatric patients, in order to compensate the induction of efavirenz hepatic metabolism, it is suggested to increase the lopinavir/ritonavir dose to 300/75 mg/m2 twice daily. (Dailly et al., 2006) Related to this topic, some studies have shown that cimetidine may be useful as a pharmacokinetic extension agent (effect similar to

Besides the effect of efavirenz and nevirapine on PIs levels, these ITINNs, (DeJesus et al., 2006) efavirenz (mainly) and nevirapine may increase metabolism, lower levels and cause therapeutic failure with various medications, such as statins (Gerber et al., 2005) and warfarin. In table 6 the most relevant interactions due to enzyme induction mediated by

decrease the metabolism of NNRTIs and increase plasma concentrations and toxicity.

ritonavir) when combined with saquinavir. (Boffito et al., 2002)

information available

Pimozide (DHHS,

Kashuba, 2007; Young, 2005)

these two NNRTIs are included.

No

information available

2011)

efavirenz, so it is recommended to monitor the levels of immunosuppressants (Tseng et al., 2002)

Efavirenz with drugs primarily metabolized by CYP3A4 may increase plasma

concentrations of drugs, due to a weak inhibitory effect on this enzyme. Recommendation: There may be potential for serious or lifethreatening reactions such as cardiac arrhythmias with pimozide, so it is not recommended for use

concomitantly


No

No

No

2: high risk 2: high risk No

3: medium risk

information available

information available

information available

**Clinical relevance: level** Comments and

With Delavirdine: Delavirdine levels increased with no change in ketoconazole

With nevirapine: increased levels of nevirapine and

decreased ketoconazole

In general, slight clinical

increased levels and potential liver toxicity of nevirapine and no changes on fluconazole

With delavirdine: increased levels of both drugs. The toxic effects of both drugs should be

A bidirectional increase

in the levels of macrolides and delavirdine and efavirenz, which may lead to problems of toxicity of macrolides

and NNRTIs

Immunosuppressants inhibit the metabolism of efavirenz and nevirapine. In addition, there are reports of decreased levels of cyclosporine by

monitored. Efavirenz and nevirapine: NNRTI levels increased and decreased levels voriconazole

levels.

levels

2: high risk With nevirapine:

relevance

 3: medium risk

Delavirdine Efavirenz Nevirapine suggestions

No

2: high risk 3: medium

risk

information available

3: medium risk

information available

**Drug group or drugs affected** 

Ketoconazole 3: medium

Itraconazole 3: medium

Voriconazole 3: medium

Fluconazole No

Macrolides **<sup>a</sup>** Erythromycin Clarithromycin

Immunosuppressant

Cyclosporine

Tacrolimus

s

risk

risk

risk

significant changes

3: medium risk

Azole antifungals **<sup>a</sup>**


a NNRTIs may inhibit CYP3A4 and may decrease the metabolism of these drugs, which can cause increased plasma concentrations and toxicity. In turn, these drugs, also due to CYP3A4 inhibition, may decrease the metabolism of NNRTIs and increase plasma concentrations and toxicity.

Table 5. Clinical relevant bidirectional drug interactions mediated by enzyme inhibition (induction in some cases) of NNRTIs with other known enzyme inhibitors (Brophy et al., 2000; Jain et al., 2002; Krikorian & Rudorf, 2005; Piscitelli & Gallicano, 2001; Polk et al., 1999; Ray et al., 2004; Robertson et al., 2005b; Sekar et al., 2008; Vogel et al., 2004; Vourvahis & Kashuba, 2007; Young, 2005)

lopinavir steady state concentration. (Dailly et al., 2005; Solas et al., 2004) The use of another low-dose PI as a pharmacokinetic extension agent (boosted), for instance ritonavir, is a strategy to counter this problem. (Acosta, 2002; Rathbun & Rossi, 2002; Zeldin & Petruschke, 2004) When using atazanavir with efavirenz it is recommended to add 100 mg of ritonavir to the usual dose (300 mg/24 hours). For lopinavir/ritonavir and efavirenz, lopinavir/ritonavir should be increased from 400/100 mg/12 hours (3 capsules) to 533/133 mg/12 hours (4 capsules) without change in the efavirenz dose. (DHHS, 2011; Solas et al., 2004; Dailly et al., 2005). For pediatric patients, in order to compensate the induction of efavirenz hepatic metabolism, it is suggested to increase the lopinavir/ritonavir dose to 300/75 mg/m2 twice daily. (Dailly et al., 2006) Related to this topic, some studies have shown that cimetidine may be useful as a pharmacokinetic extension agent (effect similar to ritonavir) when combined with saquinavir. (Boffito et al., 2002)

Besides the effect of efavirenz and nevirapine on PIs levels, these ITINNs, (DeJesus et al., 2006) efavirenz (mainly) and nevirapine may increase metabolism, lower levels and cause therapeutic failure with various medications, such as statins (Gerber et al., 2005) and warfarin. In table 6 the most relevant interactions due to enzyme induction mediated by these two NNRTIs are included.

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 327

information available

3: medium risk

information available

information available

information available

information available

No

No

2: high risk 3: medium risk

2: high risk No

3: medium risk

2: high risk 2: high risk No

2: high risk 2: high risk No

No

2: high risk No

3: medium risk

information available

information available

Darunavir No

Statins *<sup>a</sup>(*Gerber et al., 2005; Sekar et

Methadone (Altice et al., 1999; Bruce et al., 2006; Clarke et al., 2001; Stocker et

Ethinyl estradiol **<sup>a</sup>** (oral contraceptives) (Mildvan et al, 2002)

Valproic acid (Saraga et al., 2006)

Macrolides Clarithromycin (Schöller et al.,

2006b) Erythromycin

Maraviroc No

Warfarina **a** 3: medium

risk

No

information available

information available

al., 2007a) Simvastatin Atorvastatin Pravastatin Lovastatin Rosuvastatin

al., 2004)

information available

3: medium risk

etravirine with fosamprenavir.

Efavirenz may decrease plasma concentrations of darunavir (Schöller et al., 2007)

Possible reduction effect of these hypolipidemic. Pharmacological response should be traced and dose adjusted, where necessary is

recommended to monitor parameters the effectiveness of lovastatin (lipid profile) and if necessary make an adjustment in

See text: methadone and efavirenz (nevirapine)

Additional contraceptive method should be used (barrier

method, for example)

2: high risk Etravirine may decrease the

Efavirenz may decrease valproate concentrations by induction of glucoronosil diphosphate transferase

drugs plasma concentrations, because it exerts on enzyme induction of CYP3A4

Efavirenz is an inducer of CYP3A4 metabolism which can

lead to reduced plasma concentrations of maraviroc. Recommendation: In the presence of enzyme inducers such as efavirenz, maraviroc may be increased to 600 mg

medication dosage

Monitoring INR

(Holdich et al., 2007)


Delavirdine Efavirenz Etravirine

3: medium risk (Baker, 1998)

2: high risk 2: high risk 1: Very

2: high risk 2: high risk 1: Very

high risk

high risk

**Clinical relevance: level Comments and suggestions** 

hours

2: high risk Concomitant use of etravirine with saquinavir causes a decrease of the saquinavir plasma concentrations, due to etravirine induction effect on CYP3A4. (Etravirine, 2006; Harris et al., 2006b)

mg of ritonavir.

with etravirine.

2008)

25 and 35%

use of etravirine with atazanavir/ritonavir. (Aber,

Nevirapine may decrease plasma concentrations between

Fosamprenavir plasma concentration is decreased by 77% if it is administered concomitantly with etravirine. Recommendation: avoiding concomitant administration of

It is recommended to add 100

The plasma concentration of atazanavir is reduced by 38% if it is administered concomitantly

Recommendation: Concomitant

with efavirenz (specially) or nevirapine should be avoided. Efavirenz may decrease by 39% the minimum concentration of lopinavir. The dose should be increased to 533/133 mg/12

2: high risk 2: high risk 2: high risk In general, the use of a single PI

**Drug group or drugs affected** 

*Protease inhibitors* 

 Indinavir (IDV) (Harris et al., 2006b)

 Nelfinavir (NFV) Ritonavir (RTV) Amprenavir (APV)

 Tipranavir (TPV) Lopinavir/Riton avir (LPV/r) (Bergshoeff et al., 2005; Dailly et al., 2005; Solas et al.,

Saquinavir (SAQ) (Boffito et al., 2002)

Atazanavir (ATZ) (Mullin et al., 2004)

Fosamprenavir (f-APV) ( Back et

al, 2003)

4: low risk (Fletcher et al., 2000)

2004)

*(PI) <sup>a</sup>* 


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 329

**Methadone and efavirenz (nevirapine).** Efavirenz and nevirapine may reduce methadone area under curve (AUC) by 57% and 46%, respectively. (Altice et al., 1999) Therefore, when it is necessary to use methadone in patients that are using these NNRTIs, it is advisable to gradually increase the opiate dose from 10 to 10 mg, (Stocker et al., 2004) adjusting it based on effectiveness and safety parameters. Particular attention must be paid to withdrawal clinical manifestations. Withdrawal manifestations usually appear 7-10 days after initiating treatment with NNRTI and must not be confused with neurological toxicity symptoms associated with efavirenz (dizziness, headache, insomnia, concentration difficulty, nightmares, and agitation) that may occur within 1-2 days of starting treatment with efavirenz and which may be present during 14-28 days. (Bruce et al., 2006) Additionally, at sites with a suitable infrastructure, plasma levels of methadone can be monitored, aiming to reach a concentration of 400 and 250 micrograms/mL for (R, S)-methadone and (R) – methadone, respectively. (Bruce et al., 2006;

**Ritonavir as enzymatic inductor.** Ritonavir, alone or in combination with other PIs, due to its ability to induce mainly glucuronyl-transferase and, to a lesser extent CYP2B6, which

 Ethinyl estradiol (oral contraceptive component). The contraceptive effect may be diminished when used simultaneously with ritonavir. (Ouellet et al., 1998) It is therefore advisable to inform the patient of the need to use a barrier method as a

 Thyroid hormones. It may cause metabolic inactivation of thyroid hormones and treatment failure (level 2: high risk). It is therefore recommended to monitor and adjust the levels of

 Bupropion. It causes a decrease both in bupropion levels and its metabolite (hydroxybupropion), which could lead to ineffective treatment (Level 3: medium risk) and the need to double bupropión dose. (Hogeland et al., 2007) However, in theory, ritonavir can also act as an inhibitor of this isoenzyme and, therefore, increase concentrations of

**3.4 Drug interactions mediated by known enzyme inducers and consequences on PIs and NNRTIs efficacy (Benator et al., 2007; DHHS, 2011; Hamzeh et al., 2003; Kraft et al., 2004; Krikorian & Rudorf, 2005; Lim et al., 2004; Mullin et al., 2004; Romanelli &** 

The pharmacological effect of PIs and NNRTIs may be diminished and thus may appear therapeutic failure, associated with the emergence of resistance, when these ARV are combined or used simultaneously with known enzyme inducers (rifampicin, rifabutin, phenobarbital, primidone, carbamazepine, phenytoin), including St. John's wort. Rifabutin, a drug in the same family as rifampicin (rifamycins), is also characterized by inducing the metabolism of certain drugs, although to a lesser degree than rifampicin. (Finch et al., 2002)

**3.4.1 Interactions of rifamycins with protease inhibitors (PIs) or with non-nucleoside** 

**Rifampicin and PIs**. (Blumberg et al., 2003; Finch et al., 2002; Sekar et al., 2010; Spradling et al., 2002) In general, the use of rifampicin (rifabutin is preferred due to their lesser inductive

thyroid hormones in patients under treatment with ritonavir. (Touzot et al., 2006) Lamotrigine. It produces a decrease in anticonvulsant plasma levels, which can lead to ineffective treatment (level 2: high risk). It may be necessary to double lamotrigine dose

complementary birth control method (DHHS, 2011) (level 2: high risk).

in order to achieve therapeutic levels. (Van der Lee et al., 2006)

bupropion, so caution is advised until more data are available.

Clarke et al., 2001; DHHS, 2011; Stocker et al., 2004)

may produce ineffectiveness of some drugs:

**Pomeroy, 2003; Young, 2005)** 

**reverse transcriptase inhibitors (NNRTIs)** 


a Efavirenz, in particular, and nevirapine can primarily induce CYP3A4 and increase the metabolism of these drugs, which can cause a decrease plasma concentrations and therapeutic effects.

Table 6. Clinical relevant drug interactions due to enzyme induction mediated by NNRTIs (DeJesus et al., 2006; DHHS, 2011; Krikorian & Rudorf, 2005; Young, 2005)

NNRTI Efavirenz Nevirapine

Raltegravir No

Sildenafil No

Buprenorphine No

No

information available

information available

information available

information available

No

No

information available

information available

these drugs, which can cause a decrease plasma concentrations and therapeutic effects.

(DeJesus et al., 2006; DHHS, 2011; Krikorian & Rudorf, 2005; Young, 2005)

a Efavirenz, in particular, and nevirapine can primarily induce CYP3A4 and increase the metabolism of

Table 6. Clinical relevant drug interactions due to enzyme induction mediated by NNRTIs

3: medium risk

No

information available

twice daily (Abel et al., 2008) Simultaneous administration of etravirine and maraviroc causes

a decrease in plasma concentrations of maraviroc (Cmin decreased 29%, Cmax by

60% and AUC by 53%). Recommendation: dose adjustment (Davis et al., 2007)

with nevirapine and efavirenz can cause a decrease in plasma concentrations of etravirine and its therapeutic effect (Kakuda et

etravirine and raltegravir with efavirenz can cause a decrease in plasma concentration of raltegravir. Cmin is reduced by

(Anderson et al., 2008; Menard et al., 2009; Wittkop et al., 2009)

Coadministration of etravirine with sildenafil can cause a decrease in sildenafil concentrations by 57%. Recommendation:

dose adjustment of sildenafil

with buprenorphine may decrease plasma concentrations

of its active metabolite, norbuprenorphine, due to the inductive effect exerted by the

efavirenz on CYP3A4 (McCance-Katz et al., 2006)

Pérez et al., 2009)

2: high risk Coadministration of efavirenz

34% and AUC by 10%.

2: high risk Concomitant use of etravirine

al., 2006)

2: high risk 2: high risk Concomitant administration of

**Methadone and efavirenz (nevirapine).** Efavirenz and nevirapine may reduce methadone area under curve (AUC) by 57% and 46%, respectively. (Altice et al., 1999) Therefore, when it is necessary to use methadone in patients that are using these NNRTIs, it is advisable to gradually increase the opiate dose from 10 to 10 mg, (Stocker et al., 2004) adjusting it based on effectiveness and safety parameters. Particular attention must be paid to withdrawal clinical manifestations. Withdrawal manifestations usually appear 7-10 days after initiating treatment with NNRTI and must not be confused with neurological toxicity symptoms associated with efavirenz (dizziness, headache, insomnia, concentration difficulty, nightmares, and agitation) that may occur within 1-2 days of starting treatment with efavirenz and which may be present during 14-28 days. (Bruce et al., 2006) Additionally, at sites with a suitable infrastructure, plasma levels of methadone can be monitored, aiming to reach a concentration of 400 and 250 micrograms/mL for (R, S)-methadone and (R) – methadone, respectively. (Bruce et al., 2006; Clarke et al., 2001; DHHS, 2011; Stocker et al., 2004)

**Ritonavir as enzymatic inductor.** Ritonavir, alone or in combination with other PIs, due to its ability to induce mainly glucuronyl-transferase and, to a lesser extent CYP2B6, which may produce ineffectiveness of some drugs:


#### **3.4 Drug interactions mediated by known enzyme inducers and consequences on PIs and NNRTIs efficacy (Benator et al., 2007; DHHS, 2011; Hamzeh et al., 2003; Kraft et al., 2004; Krikorian & Rudorf, 2005; Lim et al., 2004; Mullin et al., 2004; Romanelli & Pomeroy, 2003; Young, 2005)**

The pharmacological effect of PIs and NNRTIs may be diminished and thus may appear therapeutic failure, associated with the emergence of resistance, when these ARV are combined or used simultaneously with known enzyme inducers (rifampicin, rifabutin, phenobarbital, primidone, carbamazepine, phenytoin), including St. John's wort. Rifabutin, a drug in the same family as rifampicin (rifamycins), is also characterized by inducing the metabolism of certain drugs, although to a lesser degree than rifampicin. (Finch et al., 2002)

#### **3.4.1 Interactions of rifamycins with protease inhibitors (PIs) or with non-nucleoside reverse transcriptase inhibitors (NNRTIs)**

**Rifampicin and PIs**. (Blumberg et al., 2003; Finch et al., 2002; Sekar et al., 2010; Spradling et al., 2002) In general, the use of rifampicin (rifabutin is preferred due to their lesser inductive

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 331

**Hypericum (St. John's wort) as enzyme inducer and ARV drugs.** In general, the combination of this herbal antidepressant with PIs or with NNRTIs, due to high probability of generating therapeutic failure, is considered contraindicated (level 1: very high risk). Thus, there are several reports of therapeutic failure of ritonavir and nevirapine, associated with the use of this herbal product. (Izzo, 2004; Winston & Boffito, 2005; Zhou et al., 2004) This substance may induce both the systemic metabolism by CYP3A4, and the presystemic metabolism by intestinal or hepatic CYP3A4; additionally it may induce P-gp activity

Table 7 contains details of interactions produced by rifamycins (rifampin and rifabutin), and table 8 contains detailed information on interactions mediated by traditional anticonvulsants.

In general, an increased gastrointestinal pH may affect the amount absorbed of delavirdine, indinavir, fosamprenavir, tipranavir, (Vourvahis & Kashuba, 2007) and atazanavir. (DeSilva et al., 2001; DHHS, 2011; Tomilo et al., 2006) However, it is basic to clarify some aspects: Concomitant administration of atazanavir with proton-pump inhibitors (omeprazole, lansoprazole, pantoprazole) (Tomilo et al., 2006) is not recommended (Level 2: high risk). The use of delavirdine with H2 antihistamines (cimetidine, ranitidine, famotidine, and nizatidine) and with proton-pump inhibitors (level 2: high risk) is also not recommended, due to a decrease near to 90% of the amount absorbed and, consequently, to a reduction in antiretroviral plasma concentrations, which can lead to treatment failure. (DHHS, 2011) In the case of atazanavir, famotidine can be used as an alternative to omeprazole, spacing its administration about 10 hours. A similar result can be achieved by adding low-dose ritonavir (100 mg) at 300 mg/day atazanavir, or by

Combined administration of antacids does not seem to significantly affect the amount absorbed of fosamprenavir, (Ford et al., 2005) or raltegravir; however it is recommended to separate the taking of the two drugs when they are used in pharmacological therapy. (Kiser

> 1: very high risk

**level** 

**Comments and suggestions (see text: rifampicin and PI** 

150 mg/48 hours, with no change in ritonavir

Rifampicin Rifabutin **and rifampicin and NNRTI)** 

2: high risk Adjust rifabutin dose

(bioavailability related interactions). (Lee et al., 2006)

**3.5 Drug interactions related to bioavailability** 

et al., 2010)

al., 2002)

**3.5.1 Drug interactions related to changes in gastrointestinal pH** 

increasing the dose to 400 mg/day of atazanavir. (Kashuba, 2005)

**Drug group or drugs affected Clinical relevance:** 

*PIs<sup>a</sup>*(Aaron et al., 2004; Benator et al., 2007; Blumberg et al., 2003; Burger et al., 2006; DeJong et al. 2004; Finch et al., 2002; Hamzeh et al., 2003; Kraft et al., 2004; La Porte et al., 2004; Polk et al., 2001; Ribera et al., 2007; Rolla et al., 2006; Spradling et al., 2002) Ritonavir (Aaron et al., 2004; Blumberg et al., 2003; DeJong et al. 2004; Finch et al., 2002; Spradling et

effect than rifabutin) with a single PI, including the most recent, such as amprenavir and atazanavir, is considered contraindicated (level 1: very high risk), because in most cases, rifampicin produces non-efficacy PIs concentrations. Besides, most treatment guidelines for HIV/AIDS patients extend this contraindication even to PIs combined with ritonavir used as a pharmacokinetic extension agent (enhancer). (DHHS, 2011) However, for saquinavir, (Rolla et al., 2006) atazanavir, (Burger et al., 2006) and lopinavir, (La Porte et al., 2004) some studies show that adding ritonavir may counteract such effect in to some extent. In these cases, both saquinavir/ritonavir dose must be adjusted to 400/400 mg twice daily, (Aaron et al., 2004; de Jong et al., 2004; Rolla et al., 2006;) and lopinavir/ritonavir to 400/400 mg/12 hours or to 800/200 mg/12 hours. (Aaron et al., 2004; de Jong et al., 2004; La Porte et al., 2004)

Regarding the concomitant administration of saquinavir/ritonavir (1000/100 mg once daily) with rifampicin, because of some reports of liver toxicity, it is recommended to avoid such a combination. (Kashuba, 2005) Additionally, some studies show that in order to achieve saquinavir adequate levels, when saquinavir is used in combination with ritonavir, it is advisable to use doses higher than 1.600/200 mg/day of this combination; (Ribera et al., 2007) however this situation could further increase the likelihood of liver toxicity problems. With respect to the advisable rifampicin dose (despite the possible inhibitory effect of PI), changes are not recommended, but liver functioning must be monitored. (La Porte et al., 2004; de Jong et al., 2004; Aaron et al., 2004) For amprenavir (Polk et al., 2001) even with low-dose ritonavir (100 or 200 mg/day), their use in combination with rifampicin is contraindicated (level 1: very high risk). (DHSS, 2011)

**Rifampicin and NNRTI**. (Finch et al., 2002; de Jong et al., 2004; Ribera et al., 2001; Ramachandran et al., 2006; Weiner et al., 2005) In general, the use of rifampicin together with delavirdine is considered contraindicated (Borin et al., 1997) (level 1: very high risk). When it needs to use simultaneously rifampin with efavirenz, it is recommended to increase the NNRTI dose from 600 to 800 mg/day. (DHHS, 2011; Matteelli et al., 2007) However, some studies conducted with patients weighing 50 kg in Thailand and infected with HIV and tuberculosis, treated with rifampicin and HAART, indicate that the use of the standard efavirenz dose (600 mg/day) may be enough to reach the desired levels and effects with NNRTI at 24 and 48 weeks. (Manosuthi et al., 2005, 2006) Similarly, results achieved in African patients support the use of efavirenz 600 mg/day. (Friedland et al., 2006) Additionally, there are some reports of toxicity with efavirenz at doses of 800 mg/day. (Brennan et al., 2005) Therefore, it is believed that with black patients weighing less than 55 kg, as well as with Hispanic or Asian patients who show evidence of susceptibility to efivarenz toxic effects, it is recommended to use a 600 mg/day dose of this NNRTI. (Matteelli et al., 2007) Thus, when it is indicate to use rifampicin with efavirenz no adjustments are recommended in the dose of rifampicin.

Consequently, in general: a) for patients receiving HAART, rifabutin should preferred to the use of for rifampicin, due to its significantly lower enzyme-inducing effect, b) the use of single PIs with rifampicin should be avoided; c) some studies show that rifampicin could be used with saquinavir, atazanavir, and lopinavir enhanced with ritonavir, but it is needing more evidence on the effectiveness and safety of such combinations, d) the use of delavirdine in conjunction with rifampicin or rifabutin is considered absolutely contraindicated, and e) in most cases where rifabutin or rifampicin is combined with PIs, dosage must be adjusted and potential toxicity must be monitored, particularly liver toxicity due to rifamycins.

**Hypericum (St. John's wort) as enzyme inducer and ARV drugs.** In general, the combination of this herbal antidepressant with PIs or with NNRTIs, due to high probability of generating therapeutic failure, is considered contraindicated (level 1: very high risk). Thus, there are several reports of therapeutic failure of ritonavir and nevirapine, associated with the use of this herbal product. (Izzo, 2004; Winston & Boffito, 2005; Zhou et al., 2004) This substance may induce both the systemic metabolism by CYP3A4, and the presystemic metabolism by intestinal or hepatic CYP3A4; additionally it may induce P-gp activity (bioavailability related interactions). (Lee et al., 2006)

Table 7 contains details of interactions produced by rifamycins (rifampin and rifabutin), and table 8 contains detailed information on interactions mediated by traditional anticonvulsants.

## **3.5 Drug interactions related to bioavailability**

330 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

effect than rifabutin) with a single PI, including the most recent, such as amprenavir and atazanavir, is considered contraindicated (level 1: very high risk), because in most cases, rifampicin produces non-efficacy PIs concentrations. Besides, most treatment guidelines for HIV/AIDS patients extend this contraindication even to PIs combined with ritonavir used as a pharmacokinetic extension agent (enhancer). (DHHS, 2011) However, for saquinavir, (Rolla et al., 2006) atazanavir, (Burger et al., 2006) and lopinavir, (La Porte et al., 2004) some studies show that adding ritonavir may counteract such effect in to some extent. In these cases, both saquinavir/ritonavir dose must be adjusted to 400/400 mg twice daily, (Aaron et al., 2004; de Jong et al., 2004; Rolla et al., 2006;) and lopinavir/ritonavir to 400/400 mg/12 hours or to 800/200 mg/12 hours. (Aaron et al., 2004; de Jong et al., 2004; La

Regarding the concomitant administration of saquinavir/ritonavir (1000/100 mg once daily) with rifampicin, because of some reports of liver toxicity, it is recommended to avoid such a combination. (Kashuba, 2005) Additionally, some studies show that in order to achieve saquinavir adequate levels, when saquinavir is used in combination with ritonavir, it is advisable to use doses higher than 1.600/200 mg/day of this combination; (Ribera et al., 2007) however this situation could further increase the likelihood of liver toxicity problems. With respect to the advisable rifampicin dose (despite the possible inhibitory effect of PI), changes are not recommended, but liver functioning must be monitored. (La Porte et al., 2004; de Jong et al., 2004; Aaron et al., 2004) For amprenavir (Polk et al., 2001) even with low-dose ritonavir (100 or 200 mg/day), their use in combination with rifampicin is

**Rifampicin and NNRTI**. (Finch et al., 2002; de Jong et al., 2004; Ribera et al., 2001; Ramachandran et al., 2006; Weiner et al., 2005) In general, the use of rifampicin together with delavirdine is considered contraindicated (Borin et al., 1997) (level 1: very high risk). When it needs to use simultaneously rifampin with efavirenz, it is recommended to increase the NNRTI dose from 600 to 800 mg/day. (DHHS, 2011; Matteelli et al., 2007) However, some studies conducted with patients weighing 50 kg in Thailand and infected with HIV and tuberculosis, treated with rifampicin and HAART, indicate that the use of the standard efavirenz dose (600 mg/day) may be enough to reach the desired levels and effects with NNRTI at 24 and 48 weeks. (Manosuthi et al., 2005, 2006) Similarly, results achieved in African patients support the use of efavirenz 600 mg/day. (Friedland et al., 2006) Additionally, there are some reports of toxicity with efavirenz at doses of 800 mg/day. (Brennan et al., 2005) Therefore, it is believed that with black patients weighing less than 55 kg, as well as with Hispanic or Asian patients who show evidence of susceptibility to efivarenz toxic effects, it is recommended to use a 600 mg/day dose of this NNRTI. (Matteelli et al., 2007) Thus, when it is indicate to use rifampicin with efavirenz no

Consequently, in general: a) for patients receiving HAART, rifabutin should preferred to the use of for rifampicin, due to its significantly lower enzyme-inducing effect, b) the use of single PIs with rifampicin should be avoided; c) some studies show that rifampicin could be used with saquinavir, atazanavir, and lopinavir enhanced with ritonavir, but it is needing more evidence on the effectiveness and safety of such combinations, d) the use of delavirdine in conjunction with rifampicin or rifabutin is considered absolutely contraindicated, and e) in most cases where rifabutin or rifampicin is combined with PIs, dosage must be adjusted and potential toxicity must be monitored, particularly liver

Porte et al., 2004)

contraindicated (level 1: very high risk). (DHSS, 2011)

adjustments are recommended in the dose of rifampicin.

toxicity due to rifamycins.

## **3.5.1 Drug interactions related to changes in gastrointestinal pH**

In general, an increased gastrointestinal pH may affect the amount absorbed of delavirdine, indinavir, fosamprenavir, tipranavir, (Vourvahis & Kashuba, 2007) and atazanavir. (DeSilva et al., 2001; DHHS, 2011; Tomilo et al., 2006) However, it is basic to clarify some aspects:

 Concomitant administration of atazanavir with proton-pump inhibitors (omeprazole, lansoprazole, pantoprazole) (Tomilo et al., 2006) is not recommended (Level 2: high risk). The use of delavirdine with H2 antihistamines (cimetidine, ranitidine, famotidine, and nizatidine) and with proton-pump inhibitors (level 2: high risk) is also not recommended, due to a decrease near to 90% of the amount absorbed and, consequently, to a reduction in antiretroviral plasma concentrations, which can lead to treatment failure. (DHHS, 2011) In the case of atazanavir, famotidine can be used as an alternative to omeprazole, spacing its administration about 10 hours. A similar result can be achieved by adding low-dose ritonavir (100 mg) at 300 mg/day atazanavir, or by increasing the dose to 400 mg/day of atazanavir. (Kashuba, 2005)

Combined administration of antacids does not seem to significantly affect the amount absorbed of fosamprenavir, (Ford et al., 2005) or raltegravir; however it is recommended to separate the taking of the two drugs when they are used in pharmacological therapy. (Kiser et al., 2010)


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 333

2: high risk 3: medium risk

risk

2: high risk 3: medium risk

**<sup>a</sup>** Rifampicin → rifabutin can induce CYP3A4 and increase the metabolism of these drugs, which can cause a decrease in plasma concentrations and virological response. In turn, some PIs or NNRTIs may modify the metabolism and concentrations or rifamycins, particularly by inhibition of CYP3A4, which can decrease their metabolism and increase plasma and concentrations and toxic effects of rifampicin

Table 7. Clinical relevant drug interactions mediated by rifampicin and rifabutin (DHHS,

In general terms, no dosage adjustment is recommended

increase rifabutin dose to 450-600 mg/day and to use the usual dose of efavirenz (600 mg/day) (Matteelli et

Co-administration of etravirine with rifampicin causes a significant decrease in plasma concentrations of etravirine, due to inductor effect on CYP3A4 of

Co-administration of etravirine with rifabutin causes a decrease by 37% of plasma concentrations of

maraviroc plasma concentrations by 66%. Recommendation: dose adjustment of maraviroc to

These drugs can reduce the

with rifabutin. With rifampicin it may be necessary to increase the dose from 200 mg/12 hours to 300 mg/12 hours, and to monitor liver function

2: high risk 2: high risk It is recommended to

al., 2007b)

rifampicin

etravirine

600 mg

McCance-Katz et al., 2006; Matteelli et al., 2007a; Ramachandran et al.,

2006; Ribera et al., 2001) Delavirdine (Borin et al., 1997; DeJong et al. 2004; McCance-Katz et al., 2006; Spradling et al., 2002) Nevirapine (Benator et al., 2007; Blumberg et al., 2003; Borin et al., 1997; Burger et al., 2006; Hamzeh et al., 2003; Kraft et al., 2004; La Porte et

al., 2004; Polk et al., 2001;

et al., 2005)

Entry inhibitor:

Maraviroc (Ogbuokiri, 2009)

2011; Tran et al., 2001)

and rifabutin, particularly liver and blood effects.

Ramachandran et al., 2006; Ribera et al., 2001, 2007; Rolla et al., 2006)

Etravirine (Abel et al., 2008) 2: high risk 3: medium

Efavirenz (Brennan et al., 2005; DeJong et al. 2004, Friedland et al., 2006; Manosuthi et al., 2005, 2006; Matteelli et al., 2007a, 2007b; Weiner


2: high risk Adjust the dose of rifabutin 150 mg/24 hours and indinavir at 1,000 mg/8

2: high risk Use usual dose of rifabutin

2: high risk Adjust rifabutin dose 150

2: high risk Adjust rifabutin dose 150

in atazanavir

2: high risk Adjust rifabutin dose 150

2: high risk Adjust rifabutin dose 150

2: high risk Adjust rifabutin dose 150

2: high risk Adjust rifabutin dose 150

changes in

dose

(300 mg/day), with no change in saquinavir

mg/24 hours and nelfinavir

mg/48hours, with no change

to 1,000 mg/8 hours

mg/24 hours, with no change in amprenavir

mg/24 hours, with no changes in fosamprenavir

mg/48hours, with no changes in tipranavir dose

mg/24 hours, with no

lopinavir/ritonavir dose

of rifabutin (RFB) is 25-Odesacetylrifabutin (desRFB), darunavir by inhibition of CYP3A4 may increase plasma concentrations of RFB, as well as its side effects

Rifampin may significantly decrease the plasma

concentrations of darunavir

(Sekar et al., 2010)

(Tibotec, 2008)

This combination is considered absolutely contraindicated

1: very high

risk

hours

1: very high risk

1: very high risk

1: very high risk

high risk

high risk

1: very high risk

1: very high risk

1: very high risk

1: very high risk

Darunavir 2: high risk 2: high risk The predominant metabolite

Indinavir (Hamzeh et al., 2003, Kraft

Saquinavir (Aaron et al., 2004; Blumberg et al., 2003; DeJong et al. 2004; Finch et al., 2002; Ribera et al., 2007; Rolla et al., 2006; Spradling et

Nelfinavir (Aaron et al., 2004; Benator et al., 2007; Blumberg et al., 2003; DeJong et al. 2004; Finch et al.,

2002; Spradling et al., 2002)

Atazanavir (Burger et al., 2006*)* 1: very

Amprenavir (Polk et al., 2001) 1: very

Fosamprenavir (Aaron et al., 2004;

Tipranavir (Vourvahis & Kashuba,

Lopinavir/ritonavir (Aaron et al., 2004; La Porte et al., 2004; Spradling

*Non-nucleosid transcriptase inhibitors (NNRTI)<sup>a</sup>* (DeJong et al. 2004; Finch et al., 2002; Friedland et al., 2006; Manosuthi et al., 2005, 2006;

DeJong et al. 2004)

2007)

et al., 2002)

et al., 2004)

al., 2002)


**<sup>a</sup>** Rifampicin → rifabutin can induce CYP3A4 and increase the metabolism of these drugs, which can cause a decrease in plasma concentrations and virological response. In turn, some PIs or NNRTIs may modify the metabolism and concentrations or rifamycins, particularly by inhibition of CYP3A4, which can decrease their metabolism and increase plasma and concentrations and toxic effects of rifampicin and rifabutin, particularly liver and blood effects.

Table 7. Clinical relevant drug interactions mediated by rifampicin and rifabutin (DHHS, 2011; Tran et al., 2001)

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 335

 Etravirine-Ranitidine. The concurrently use of etravirine and ranitidine for a period of eight days may cause a reduction in absorption of etravirine due to decrease of gastric

 Darunavir-Omeprazole/Ranitidine. Due to the absorption of some PIs is pH dependent gastric acid, antacids may inhibit absorption of PIs. However, some studies show that there are not clinically significant DIs between darunavir and omeprazole/ranitidine, thus no dose adjustment is required to concomitant

 Absorption of delavirdine is noticeably decreased at pH values below 3. Therefore, its administration with antacids may cause a decrease on levels and effects (level 3: medium risk), and it is recommended to space its administration, at least an hour. (Tran

Broadly, didanosine in buffered tablets may decrease absorption, levels and efficacy of delavirdine and indinavir (as well of other drugs that requiring an acidic environment for absorption), due to didanosine excipients maintain gastric pH above 3. However, the addition of didanosine in a new dosage form (enteric-coated granules within capsules) minimizes this effect, as evidenced by the results of studies on drugs whose bioavailability may be decreased by interactions with antacids such as indinavir, ketoconazole, and ciprofloxacin. (Damle et al., 2002b) Furthermore, the buffered tablet effect cannot be generalized for drugs requiring an acidic pH for absorption, such as itraconazole or fluconazole, whose bioavailability is not altered when administered simultaneously with

Related to this type of mechanism, in general, food (drug-food interactions) can decrease the amount absorbed and effects of several ARV drugs. For example, in the case of didanosine (buffered tablet and enteric-coated granules within capsules), it is recommended to take it on an empty stomach (2 hours before or two hours after meals), (la Porte et al., 2005) because a decrease (between 18 and 55%) in the amount absorbed can be produced (level 3: risk medium). (Damle et al., 2002c; Kearney et al., 2005) Such effect may be minimized if didanosine is administered in combination with indinavir/ritonavir or with tenofovir (La

Porte et al., 2005; Kearney et al., 2005) (see didanosine- tenofovir Interaction below).

**3.5.2 Interactions related to presystemic metabolism and/or P-glycoprotein activity**  P-gp (ABCB1) is one of the most extensively studied transporters regarding DIs. P-gp is expressed in multiple key organs in drug disposition such as small intestine, blood-brain barrier, kidney, and liver. Therefore, P-gp mediated DIs can occur at various organs and tissues. Thus, the induction or inhibition of P-gp can lead to drug-drug interactions. For instance, induction of the intestinal P-glycoprotein activity can cause reduced bioavailability of orally administered drugs and cause therapeutic failure. In contrast, the inhibition of the intestinal P-gp activity can lead to increased bioavailability, concentrations and adverse

Broadly, PIs, particularly ritonavir, are mechanism-based intestinal P-gp efflux activity inhibition, which partially explains the effect on the levels and clinical results of some drugs, (de Maat et al., 2003) such as cyclosporine (Izzedine et al. 2004) and digoxin. (Penzak et al., 2004) Ritonavir can also inhibit renal P-gp activity, which may lead to a decrease in renal clearance of digoxin, associated with an increase in the levels and effects of this drug. (Ding

(Schöller-Gyüre et al., 2008) acidity (level 3: medium risk).

didanosine, regardless of dosage form used. (Damle et al., 2002a)

administration. (Sekar et al., 2007b)

et al., 2001)

drug effects. (Josephson, 2010)

et al., 2004)


Table 8. Clinical relevant drug interactions mediated by traditional anticonvulsants (DHHS, 2011; Tran et al., 2001)

*Phenytoin Carbamazepine*

*Protease Inhibitors (PI) <sup>a</sup>* (Lim et al, 2004; Mullin et al., 2004; Romanelli & Pomeroy, 2003*)*  Ritonavir 2: high risk 2: high risk 2: high risk Traditional anticonvulsants

**Clinical relevance: level Comments and suggestions** 

(except valproic acid) can cause ineffectiveness of PI (including their combination with low dose ritonavir). (Ogbuokiri, 2009)

It is recommended the use of

anticonvulsants (gabapentine, lamotrigine, vigabatrin and

second-generation

Gabapentine, due to its pharmacological properties (it is not metabolized, it does not

affect CYP activity), is considered the best option (Mullin et al., 2004; Romanelli

& Pomeroy, 2003)

decrease the plasma

CYP2C9

(Acosta, 2002)

maraviroc

concentrations of phenytoin, due to the induction of

Delavirdine with any of the traditional anticonvulsants (except valproic acid) is considered contraindicated. Traditional anticonvulsants (except valproic acid) can cause NNRTI ineffectiveness

Traditional anticonvulsants (except valproic acid) can cause ineffectiveness of

2:high risk 2:high risk 2:high risk Lopinavir / ritonavir may

1:very high risk

2:high risk

2: high risk

3: medium risk

Table 8. Clinical relevant drug interactions mediated by traditional anticonvulsants (DHHS,

*Non-nucleoside transcriptase inhibitors (NNRTI) <sup>a</sup> (Mullin et al., 2004; Romanelli & Pomeroy,* 

No

information available

information available

information available

3: medium risk

topiramate).

**Drug group or drugs affected** 

 Lopinavir/ ritonavir

*Entry inhibitor*  Maraviroc (Ogbuokiri, 2009)

2011; Tran et al., 2001)

*2003; Tran et al., 2001; Tseng & Foisy, 1999)* 

risk

Nevirapine 2: high risk No

Efavirenz 2:high risk No

3: medium risk

Delavirdine 1: very high

*Phenobarbital and primidone*

 Indinavir 2: high risk 2: high risk 2: high risk Saquinavir 2: high risk 2: high risk 2: high risk Nelfinavir 2:high risk 2:high risk 2:high risk Atazanavir 2:high risk 2:high risk 2:high risk Amprenavir 2:high risk 2:high risk 2:high risk Fosamprenavir 2:high risk 2:high risk 2:high risk Tipranavir 2:high risk 2:high risk 2:high risk


Broadly, didanosine in buffered tablets may decrease absorption, levels and efficacy of delavirdine and indinavir (as well of other drugs that requiring an acidic environment for absorption), due to didanosine excipients maintain gastric pH above 3. However, the addition of didanosine in a new dosage form (enteric-coated granules within capsules) minimizes this effect, as evidenced by the results of studies on drugs whose bioavailability may be decreased by interactions with antacids such as indinavir, ketoconazole, and ciprofloxacin. (Damle et al., 2002b) Furthermore, the buffered tablet effect cannot be generalized for drugs requiring an acidic pH for absorption, such as itraconazole or fluconazole, whose bioavailability is not altered when administered simultaneously with didanosine, regardless of dosage form used. (Damle et al., 2002a)

Related to this type of mechanism, in general, food (drug-food interactions) can decrease the amount absorbed and effects of several ARV drugs. For example, in the case of didanosine (buffered tablet and enteric-coated granules within capsules), it is recommended to take it on an empty stomach (2 hours before or two hours after meals), (la Porte et al., 2005) because a decrease (between 18 and 55%) in the amount absorbed can be produced (level 3: risk medium). (Damle et al., 2002c; Kearney et al., 2005) Such effect may be minimized if didanosine is administered in combination with indinavir/ritonavir or with tenofovir (La Porte et al., 2005; Kearney et al., 2005) (see didanosine- tenofovir Interaction below).

## **3.5.2 Interactions related to presystemic metabolism and/or P-glycoprotein activity**

P-gp (ABCB1) is one of the most extensively studied transporters regarding DIs. P-gp is expressed in multiple key organs in drug disposition such as small intestine, blood-brain barrier, kidney, and liver. Therefore, P-gp mediated DIs can occur at various organs and tissues. Thus, the induction or inhibition of P-gp can lead to drug-drug interactions. For instance, induction of the intestinal P-glycoprotein activity can cause reduced bioavailability of orally administered drugs and cause therapeutic failure. In contrast, the inhibition of the intestinal P-gp activity can lead to increased bioavailability, concentrations and adverse drug effects. (Josephson, 2010)

Broadly, PIs, particularly ritonavir, are mechanism-based intestinal P-gp efflux activity inhibition, which partially explains the effect on the levels and clinical results of some drugs, (de Maat et al., 2003) such as cyclosporine (Izzedine et al. 2004) and digoxin. (Penzak et al., 2004) Ritonavir can also inhibit renal P-gp activity, which may lead to a decrease in renal clearance of digoxin, associated with an increase in the levels and effects of this drug. (Ding et al., 2004)

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 337

 Zidovudine, cisplatin or pemetrexed: The use of zidovudine in patients receiving cisplatin or pemetrexed may increase the risk of hematological toxicity (Makinson et al.,

Broadly, the use of drugs with a similar unsafe profile leads to an increased probability

increased risk of anemia (DHHS, 2011) (level 2: high risk).

hyperbilirubinemia (DHHS, 2011) (level 2: high risk).

(Shriner & Goetz, 1992) (Level 2 or 3: high or middle risk).

**5. Clinically relevant drug-disease interactions in HIV-infected patients** 

Due to it is important some examples related to clinically relevant drug-disease interactions

(level 2 or 3: high risk or middle).

**receiving antiretroviral treatment** 

in HIV-infected patients are detailed:

 Didanosine and ribavirin. This combination increases the likelihood of mitochondrial toxicity, which can lead to lactic acidosis and pancreatitis (Fleischer et al., 2004; Perronne et al. 2006) (level 2: high risk). Similarly, the didanosine-adefovir combination increases the risk of pancreatitis (Perronne et al. 2006) (level 2: high risk). Atazanavir and indinavir: This combination increases the likelihood of developing

 Drugs with additive hepatic toxicity: In general, concomitant use of hepatotoxic drugs enhances the probability of hepatic alterations, for instance, when rifampicin is used with isoniazid (Steele et al., 1991) or with pyrazinamide (Yee et al., 2003) for treatment of tuberculosis; or when acetaminophen is used with zidovudine

 Drugs with additive renal toxicity: Simultaneous use of renal toxic drugs increases the likelihood of problems in this organ, for instance the simultaneously use of adefovir, acyclovir (intravenously), cidofovir, foscarnet, indinavir, ritonavir, tenofovir, pentamidine, aminoglycosides, and amphotericin B (Fisher et al., 1989)

 Stavudine and didanosine: This combination is considered absolutely contraindicated (level 1: very high risk), due to increased likelihood of significant synergism of adverse drug problems associated with mitochondrial toxicity, which may manifest, especially as peripheral neuropathy, pancreatitis, and lactic acidosis. (Boubaker et al., 2001; Catanzaro et al., 2004; Coghlan et al., 2001; DHHS, 2011) Similarly, didanosine/zalcitabine and stavudine/zalcitabine combinations are considered contraindicated because of a greater likelihood and severity of peripheral neuropathy (Dalakas et al., 2001; Simpson & Tagliati, 1995) (level 1: very high risk). In general, NRTIs (lower risk with abacavir and lamivudine) may inhibit competitively the mitochondrial DNA polymerase gamma, an enzyme responsible for repairing mitochondrial DNA associated with oxidative alterations. (Dagan et al., 2002) Therefore, the use of NRTI may lead to depletion of DNA and uncoupling of mitochondrial respiratory chain and thus, to the accumulation of radicals and free fatty acids, as well as dicarboxylic acids, responsible for mitochondrial toxicity associated with this group of drugs. (Petit et al, 2005) Zidovudine and ganciclovir: This combination increases the likelihood of developing bone marrow suppression, a condition associated with the occurrence of severe hematologic toxicity and life threatening infections, including the progression of cytomegalovirus infection (Hochster et al., 1990) (level 1 : very high risk). Related to this type of interaction, another combination with additive hematologic toxicity combination is zidovudine with ribavirin, associated with an

2010) (level 3: medium risk).

and severity of adverse effects, for instance:

Simultaneous administration of saquinavir 500 mg (low dose ritonavir) with omeprazole is associated with an increase of 82% in AUC of saquinavir. Although the mechanism is unclear, it is believed that this effect is due to the fact that omeprazole can inhibit P-gp activity and other membrane transporters activity. (Winston et al., 2006) Similarly, saquinavir bioavailability increases significantly when administered with cimetidine, (Boffito et al., 2002) ranitidine or food, an effect independent of pH increase. (Kakuda & Falcon, 2006) These findings could lead to believe that food is likely to increase solubility and to decrease the hepatic first pass effect, while the effect of cimetidine, ranitidine, and omeprazole, similar to what has been observed with other known inhibitors of intestinal and hepatic CYP3A4, may be associated to decreased presystemic metabolism and, to a lesser extent, to decreased P-gp intestinal activity. (Amariles, 2007)

## **4. Clinically relevant pharmacodynamic drug interactions in HIV-infected patients receiving antiretroviral therapy**

Due to it is important some examples related to pharmacodynamic DIs in HIV-infected patients are detailed:


Simultaneous administration of saquinavir 500 mg (low dose ritonavir) with omeprazole is associated with an increase of 82% in AUC of saquinavir. Although the mechanism is unclear, it is believed that this effect is due to the fact that omeprazole can inhibit P-gp activity and other membrane transporters activity. (Winston et al., 2006) Similarly, saquinavir bioavailability increases significantly when administered with cimetidine, (Boffito et al., 2002) ranitidine or food, an effect independent of pH increase. (Kakuda & Falcon, 2006) These findings could lead to believe that food is likely to increase solubility and to decrease the hepatic first pass effect, while the effect of cimetidine, ranitidine, and omeprazole, similar to what has been observed with other known inhibitors of intestinal and hepatic CYP3A4, may be associated to decreased presystemic metabolism and, to a

**4. Clinically relevant pharmacodynamic drug interactions in HIV-infected** 

Due to it is important some examples related to pharmacodynamic DIs in HIV-infected

 Lamivudine and emtricitabine: this combination is considered inadequate, since these drugs show a minimum additive effect, as well as a similar profile of resistance (DHHS,

 Stavudine and zidovudine: this combination is considered inadequate, because their chemical similarity can lead to competition for intracellular phosphorylation zidovudine inhibits phosphorylation and effects of stavudine (Ray, 2005) which might cause therapeutic failure (Level 2: high risk). Concerning this interaction, it is important to note that prior exposure to zidovudine (2 to 45 months) does not influence the ability of HIV-infected patients to phosphorylate stavudine. (Hoggard et al., 2001) Due to similar mechanism, the lamivudine and zalcitabine combination is considered inadequate, since lamivudine may inhibit the phosphorylation of zalcitabine, which may lead to therapeutic failure (Becher et al., 2004; DHHS, 2011; Havlir et al., 2000)

 Abacavir, tenofovir, and lamivudine (or emtricitabine) as a triple NRTI therapy: This combination is considered contraindicated due to the rapid onset of virologic failure, when used as initial therapy (level 2: high risk). (DHHS, 2011) A similar consideration has been established for combination of tenofovir, didanosine, and lamivudine

 Amprenavir and fosamprenavir: This combination is not recommended, because fosamprenavir is a prodrug of amprenavir, thus their combined use increases the risk of adverse drug reactions without additional benefits (level 2: high risk).(DHHS, 2011) Stavudine and aminopterin: This combination is not recommended, because the combination causes an inhibition of hepatic mitochondrial DNA and hepatic toxicity

 Tenofovir and cisplatin or pemetrexed: The use of tenofovir in patients receiving cisplatin or pemetrexed may increase the risk of renal toxicity (3: medium risk).

 Tenofovir/emtricitabine, disulfiram and nifedipine: The use of nifedipine in HIVinfected patients receiving tenofovir/emtricitabine and disulfiram may cause lactic

lesser extent, to decreased P-gp intestinal activity. (Amariles, 2007)

**patients receiving antiretroviral therapy** 

patients are detailed:

2011) (level 2: high risk).

(level 2: high risk).

(emtricitabine). (DHHS, 2011)

(Makinson et al., 2010)

(Setzer et al., 2008) (level 1: very high risk).

acidosis (level 2: high risk). ( Moling et al., 2009)

	- Stavudine and didanosine: This combination is considered absolutely contraindicated (level 1: very high risk), due to increased likelihood of significant synergism of adverse drug problems associated with mitochondrial toxicity, which may manifest, especially as peripheral neuropathy, pancreatitis, and lactic acidosis. (Boubaker et al., 2001; Catanzaro et al., 2004; Coghlan et al., 2001; DHHS, 2011) Similarly, didanosine/zalcitabine and stavudine/zalcitabine combinations are considered contraindicated because of a greater likelihood and severity of peripheral neuropathy (Dalakas et al., 2001; Simpson & Tagliati, 1995) (level 1: very high risk). In general, NRTIs (lower risk with abacavir and lamivudine) may inhibit competitively the mitochondrial DNA polymerase gamma, an enzyme responsible for repairing mitochondrial DNA associated with oxidative alterations. (Dagan et al., 2002) Therefore, the use of NRTI may lead to depletion of DNA and uncoupling of mitochondrial respiratory chain and thus, to the accumulation of radicals and free fatty acids, as well as dicarboxylic acids, responsible for mitochondrial toxicity associated with this group of drugs. (Petit et al, 2005)
	- Zidovudine and ganciclovir: This combination increases the likelihood of developing bone marrow suppression, a condition associated with the occurrence of severe hematologic toxicity and life threatening infections, including the progression of cytomegalovirus infection (Hochster et al., 1990) (level 1 : very high risk). Related to this type of interaction, another combination with additive hematologic toxicity combination is zidovudine with ribavirin, associated with an increased risk of anemia (DHHS, 2011) (level 2: high risk).
	- Didanosine and ribavirin. This combination increases the likelihood of mitochondrial toxicity, which can lead to lactic acidosis and pancreatitis (Fleischer et al., 2004; Perronne et al. 2006) (level 2: high risk). Similarly, the didanosine-adefovir combination increases the risk of pancreatitis (Perronne et al. 2006) (level 2: high risk).
	- Atazanavir and indinavir: This combination increases the likelihood of developing hyperbilirubinemia (DHHS, 2011) (level 2: high risk).
	- Drugs with additive hepatic toxicity: In general, concomitant use of hepatotoxic drugs enhances the probability of hepatic alterations, for instance, when rifampicin is used with isoniazid (Steele et al., 1991) or with pyrazinamide (Yee et al., 2003) for treatment of tuberculosis; or when acetaminophen is used with zidovudine (Shriner & Goetz, 1992) (Level 2 or 3: high or middle risk).
	- Drugs with additive renal toxicity: Simultaneous use of renal toxic drugs increases the likelihood of problems in this organ, for instance the simultaneously use of adefovir, acyclovir (intravenously), cidofovir, foscarnet, indinavir, ritonavir, tenofovir, pentamidine, aminoglycosides, and amphotericin B (Fisher et al., 1989) (level 2 or 3: high risk or middle).

## **5. Clinically relevant drug-disease interactions in HIV-infected patients receiving antiretroviral treatment**

Due to it is important some examples related to clinically relevant drug-disease interactions in HIV-infected patients are detailed:

### **Pregnancy**

 Amprenavir oral solution, due to the high content of propylene glycol, as an excipient, can cause toxicity problems (DHHS, 2011) (level 2: high risk).

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 339

 **Amprenavir oral solution (propylene glycol) and oral ritonavir (ethanol excipient)**. Because propylene glycol and ethanol are metabolized by the same enzyme, accumulation and toxicity may occur. (DHHS, 2011) Also, related to the presence of excipients, the use of lopinavir/ritonavir oral solution (4.2% in ethanol) combined with metronidazole or disulfiram, substances that may inhibit the alcohol dehydrogenase, may lead to the development of disulfiram effect. (Cvetkovic & Goa, 2003; de Maat et

 **Atazanavir and tenofovir.** Tenofovir reduces atazanavir levels through a mechanism still to be established, so it is recommend using atazanavir (300 mg) together with low dose ritonavir (100 mg) as an enhancing agent. (Taburet et al., 2004) In addition, lopinavir/ritonavir and atazanavir may increase levels and toxic renal effects of tenofovir. (Perronne et al., 2006) It is important to illustrate that the inductive effects of tenofovir on the metabolism of atazanavir have not been documented to others PIs (Boffito et al., 2005) **Saquinavir and adefovir.** Adefovir (antiviral used in the treatment of hepatitis B)

 **Tenofovir and enzyme inducers.** NRTIs with inducing capacity (efavirenz and nevirapine) (Droste et al., 2006) and probably rifampicin, (Droste et al., 2005) do not cause variation in tenofovir levels and effects. Therefore, it is not recommended to

 **Strong or moderate inhibitors and PIs.** Some known enzyme inhibitors such as ketoconazole may decrease the extra-intestinal P-gp efflux activity and increasing PIs levels and effects in certain body areas, such as the cerebrospinal fluid, which have been evidenced for ritonavir and saquinavir. (Cvetkovic & Goa, 2003; Oldfield & Plosker,

 **Rosiglitazone and nevirapine.** Rosiglitazone, apparently due to increased CYP3A4 or P-gp activity, may decrease nevirapine levels. This effect does not occur when using

adjust the dose of either drug, when it is necessary to use such combinations. **Amprenavir capsules, vitamin E, and warfarin levels.** Amprenavir capsules contain an amount of vitamin E that exceeds recommended daily dose, so in patients treated with amprenavir in this dosage form, vitamin E supplementation should be avoided. In patients under treatment with warfarin, high levels of vitamin E (associated with intakes higher than 400 IU per day) increase the risk of bleeding. (Amsay et al, 2005,

lowers saquinavir levels and effects. (Perronne et al. 2006)

2006; Lin & Yamazaki, 2003; Lin, 2003; Khaliq et al., 2000)

risk).

al., 2003)

Heck et al., 2000)

less than 60 kg. (Kearney et al., 2005; Martinez et al., 2004; Antoniou et al., 2003) Additionally, even using reduced dose, didanosine toxicity signs and symptoms should be monitored (significant increase in serum amylase or lipase, neuropathy, paresthesia, nausea, vomiting, and abdominal pain). (DHHS, 2011) There are reports of didanosine toxicity (deadly lactic acidosis and acute hepatic failure), (Guo & Fung, 2004; Masía et al., 2005) at doses of 200 mg/day, especially in women weighing 60 or less kg. A similar effect could be observed when administering didanosine with other purine nucleoside phosphorylase inhibitory drugs, such as allopurinol, ganciclovir (Moling et al., 2009) and valganciclovir. (Tseng & Salit, 2007) Additionally, the use of didanosine-tenofovir combination may lead to decrease in the CD4 lymphocyte levels, thus this combination should be avoided (Anderson & Kakuda, 2006; Barreiro & Soriano, 2006) (level 1 or 2: Very high risk" or higher). Similarly, the **didanosine and ganciclovir** combination has also been associated with decreased CD4 cell levels (Tseng & Salit, 2007) (level 2: high

	- NNRTI and PI. In general this class of drugs should be administered with caution or avoided altogether in patients with moderate or severe liver failure (DHHS, 2011) (level 3: medium risk).
	- Amprenavir, fosamprenavir. Avoid their use in the presence of severe liver failure (Amariles et al., 2007c; DHHS, 2011) (level 3: medium risk).
	- Amprenavir/ritonavir, tipranavir / ritonavir. Their use is contraindicated in severe liver failure (Amariles et al., 2007c; DHHS, 2011) (level 2: high risk).

## **6. Other clinically relevant drug interactions in HIV-infected patients attributed to different or unclear mechanisms**

Other examples of clinically relevant DIs in HIV-infected patients attributed to different or unclear mechanisms are:


excipient, can cause toxicity problems (DHHS, 2011) (level 2: high risk).

teratogenic risk (AIDS Patient Care, 2005; DHHS, 2011) (level 2: high risk). Didanosine and stavudine, because there are reports of severe and even fatal lactic acidosis in pregnant women receiving stavudine and didanosine (DHHS, 2011)

**Moderate (Child Pugh score: 7-9 points) or severe (Child Pugh: over 9 points) liver** 

 NNRTI and PI. In general this class of drugs should be administered with caution or avoided altogether in patients with moderate or severe liver failure (DHHS,

Amprenavir, fosamprenavir. Avoid their use in the presence of severe liver failure

Amprenavir/ritonavir, tipranavir / ritonavir. Their use is contraindicated in severe

 **Women with CD4 lymphocyte counts> 250 cells/mm3 or men with CD4 lymphocyte counts > 400 cells/mm3.** For these groups of patients there are reports of serious, sometimes fatal, liver impairments attributed to nevirapine (DHHS, 2011) (level 2: high

 **Cirrhosis.** The combination of didanosine and ribavirin can lead to liver decompensation. Therefore, this combination is considered no-adequate for patients

 **HIV**. The use of rifapentine (rifamycin) is considered inadequate, due to fewer efficacies in preventing the onset of tuberculosis in this group of patients. For this reason this could be considered as a possible drug-disease interaction (DHHS, 2011)

Other examples of clinically relevant DIs in HIV-infected patients attributed to different or

 **Zidovudine and enzyme inducers.** Zidovudine is metabolized and inactivated by glucuronyl transferase. Thus, inducers or inhibitors of this enzyme can affect its levels and effects (Kiang et al, 2005). For example, rifampicin may increase zidovudine

 **Tenofovir and didanosine.** Tenofovir increases didanosine levels and toxicity (particularly, pancreatitis), (Martinez et al., 2004) apparently due to inhibition of its metabolism by purine nucleoside phosphorylase. (Ray et al., 2004) It is thus recommended to reduce the dose from 400 mg to 250 mg/24 hours in patients weighing

(Amariles et al., 2007c; DHHS, 2011) (level 3: medium risk).

with advanced liver fibrosis (Perronne, et al. 2006) (level 2: high risk).

**6. Other clinically relevant drug interactions in HIV-infected patients** 

metabolism and decrease its levels and effects. (Gallicano et al., 1999)

**attributed to different or unclear mechanisms** 

liver failure (Amariles et al., 2007c; DHHS, 2011) (level 2: high risk). **Children under 4 years.** Amprenavir oral solution. Due to the high content of propylene glycol, as an excipient, it can cause toxicity problems (DHHS, 2011) (level 2:

Amprenavir oral solution, due to the high content of propylene glycol, as an

 Efavirenz, especially during the first 3 months, because it represents a potential teratogenic risk, it is included in the D category of FDA classification of drugs for

**Pregnancy** 

**failure** 

high risk).

risk).

(level 2: high risk).

unclear mechanisms are:

(level 2: high risk).

2011) (level 3: medium risk).

less than 60 kg. (Kearney et al., 2005; Martinez et al., 2004; Antoniou et al., 2003) Additionally, even using reduced dose, didanosine toxicity signs and symptoms should be monitored (significant increase in serum amylase or lipase, neuropathy, paresthesia, nausea, vomiting, and abdominal pain). (DHHS, 2011) There are reports of didanosine toxicity (deadly lactic acidosis and acute hepatic failure), (Guo & Fung, 2004; Masía et al., 2005) at doses of 200 mg/day, especially in women weighing 60 or less kg. A similar effect could be observed when administering didanosine with other purine nucleoside phosphorylase inhibitory drugs, such as allopurinol, ganciclovir (Moling et al., 2009) and valganciclovir. (Tseng & Salit, 2007) Additionally, the use of didanosine-tenofovir combination may lead to decrease in the CD4 lymphocyte levels, thus this combination should be avoided (Anderson & Kakuda, 2006; Barreiro & Soriano, 2006) (level 1 or 2: Very high risk" or higher). Similarly, the **didanosine and ganciclovir** combination has also been associated with decreased CD4 cell levels (Tseng & Salit, 2007) (level 2: high risk).


lopinavir/ritonavir with nevirapine. (Oette et al., 2005) Thus, it is recommended to monitor nevirapine levels and effects, when this drug is used in combination with rosiglitazone (level 2: high risk).

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 341

d. **Software implementation.** The program, posterior to entering drug treatment of a specific patient, generate the list of possible interactions with clinical relevance, accompanied by the suggestion of the most appropriate process to be followed by the healthcare professional user of the program. In Colombia, the program have been implemented in 24 health institutions, in addition the access is free and it is available on the website http:www.udea.edu.co/pypfarmaceutica. However, further investigation to evaluate in more detail information regarding to positive predictive and negative

Due to ARV therapy is one of the most dynamic in terms of launching new products in the market, it is necessary to carry out structured and systematic review of publications on Pubmed/Medline and other electronic databases to identify new DIs in HIV-infected patients, receiving ART therapy, which must be complemented with evaluating the

In HIV-infected patients receiving antiretroviral therapy, closer to 80% of relevant clinical pharmacokinetic DIs are associated to induction or inhibition of the systemic hepatic metabolism, which is associated to systemic clearance for a significant number drugs. While closer to 20% of relevant clinical pharmacokinetic DIs are associated to changes in bioavailability related interactions, including changes in gastrointestinal pH, presystemic metabolism and/or P-gp transport activity. Therefore, in HIV-infected patients receiving HAART of cART, the assessing and predicting clinical relevance of a given pharmacokinetic DI need to understand and use of concepts related to induction and, particularly with enzyme inhibition (mainly in CYP3 and CYP2 subfamilies) and, to a lesser extent, to bioavailability (especially with changes in a drug presystemic metabolism and/or P-gp

It is becoming more evident that ARV drugs are metabolized via common pathways by CYP450 enzymatic complex, which leads to an increased probability of new clinical relevant pharmacokinetic DIs due to the inhibition or stimulation of CYP isoforms (mainly in CYP3

In general, for a patient using more than one drug with differential capacity to modify CYP3A4 and/or P-gp enzymatic activity (some induce while some others inhibit), it is difficult to predict the net effect on the levels and effects of a drug whose metabolism is affected. (Spradling et al., 2002) Therefore, it is necessary, from a theoretical perspective, to avoid the use of such schemes, which is difficult in patients with HIV/AIDS. In this context, some studies conducted to establish the influence of several antiretroviral drugs on CYP3A activity show that ritonavir/nelfinavir inhibitory effect is maintained and it counteracts efavirenz/nevirapine inducing effect, when they are administered in combination. Additionally, it has been found that chronic administration of ritonavir (200 mg/day) or nelfinavir (2.5 g/12 hours) does not increase CYP3A activity. (Fellay et al., 2005; Mouly et

Similar to other groups of patients, it is important to systematize, distribute, and use guidelines and recommendations based on the findings of studies indicating which

(Gaikwad et al., 2007; Rodríguez et al., 2009)

**8. Conclusions** 

transport activity).

al., 2006)

and CYP2 subfamilies).

predictive values of alert generates by program is required.

scientific evidence and classifying their clinical relevance.

(68.3%) are clinical relevant (Levels 1 and 2) in HIV-infected patients receiving ARV therapy. Thus, the software meets the requirements defined for this type of programs.

 **Maraviroc and raltegravir.** The concomitantly administration of these drugs can reduce peak concentrations of both drugs due to changes in pre-systemic elimination associated to changes in absorption and/or first pass metabolism; however, the exact mechanism of interaction has not been determined (Andrews et al., 2010) (3: medium risk).

## **7. Software for evaluating and predicting clinical relevance of drug interactions in HIV-infected patients receiving antiretroviral therapy**

The identification, systematization, evaluation, and prediction of DIs may be easier by using computer applications. Additionally, these kinds of informatics tools may contribute to reduce the risk to arise of clinical relevant DIs, and thus, the negative effect in goals of HAAR or cART in patients with HIV/SIDA. Thus, the design software that facilitates the identification and prediction the clinical relevance of drug interactions may be an important contribution to get the possible outcomes best in HIV-infected patients receiving HAART or cART. (Amariles et al., 2008) In this way, the following tasks have been done:


(68.3%) are clinical relevant (Levels 1 and 2) in HIV-infected patients receiving ARV therapy. Thus, the software meets the requirements defined for this type of programs. (Gaikwad et al., 2007; Rodríguez et al., 2009)

d. **Software implementation.** The program, posterior to entering drug treatment of a specific patient, generate the list of possible interactions with clinical relevance, accompanied by the suggestion of the most appropriate process to be followed by the healthcare professional user of the program. In Colombia, the program have been implemented in 24 health institutions, in addition the access is free and it is available on the website http:www.udea.edu.co/pypfarmaceutica. However, further investigation to evaluate in more detail information regarding to positive predictive and negative predictive values of alert generates by program is required.

## **8. Conclusions**

340 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

 **Maraviroc and raltegravir.** The concomitantly administration of these drugs can reduce peak concentrations of both drugs due to changes in pre-systemic elimination associated to changes in absorption and/or first pass metabolism; however, the exact mechanism of

The identification, systematization, evaluation, and prediction of DIs may be easier by using computer applications. Additionally, these kinds of informatics tools may contribute to reduce the risk to arise of clinical relevant DIs, and thus, the negative effect in goals of HAAR or cART in patients with HIV/SIDA. Thus, the design software that facilitates the identification and prediction the clinical relevance of drug interactions may be an important contribution to get the possible outcomes best in HIV-infected patients receiving HAART or

a. **Structured and systematic review of publications on Pubmed/Medline and other electronic databases, supplemented by other primary and secondary information sources to identify DIs in HIV-infected patients receiving ART therapy.** Thus, published articles of DIs in HIV-infected patients were identified by a comprehensive literature search using electronic databases of information sources (Medline/Pubmed, SIETES, Medscape, and Tripdatabase) to identify all full text or abstracts published in English and Spanish from January 1996 to February 2011. Additionally, a search was done in some specify journals: New England Journal of Medicine, British Medical Journal, and other recognized information sources: the electronic sheet drug interactions: www.drug-interactions.com; Philip D. Hansten, John R. Horn. Managing Clinically Important Drug Interactions, facts and comparison, 2003; Stockley IH. Drug Interactions. First edition. Pharma Editores. Spain, 2004, and Drug Interaction Facts,

b. **Classification of the clinical relevance of pairs of the identified DIs.** The accessing and predicting the clinical relevance of pairs of the identified DIs were based on the severity and probability of occurrence of the DIs. Based on the possible combinations of severity and probability of occurrence, DIs were grouped into 4 categories: Level 1 (very high risk), level 2 (high risk), level 3 (medium risk), an level 4 (low risk) as it was mentioned in numeral 2.3 (Determining and predicting of the clinical relevance level)

c. **Software design.** The results of the review and evaluation of clinical relevance of the DIs were used to design a platform with alert generator, report generator, constant evolution support systems data integrity, network management intranet and Internet Web and Windows platforms. The developed software facilitates the identification, evaluation, and prediction of clinical relevant of 1,082 drug pairs of potential DIs, near to 80% of them due to pharmacokinetic mechanism (changes in plasma concentration), mainly associated to enzyme inhibition. The scaling of these 1,082 drug pairs of the recognized DIs according to different dosage forms and strengths of identified drugs, generates a total of 6,087 pairs of DIs, which, according their clinical relevance, 4,158

interaction has not been determined (Andrews et al., 2010) (3: medium risk).

**7. Software for evaluating and predicting clinical relevance of drug interactions in HIV-infected patients receiving antiretroviral therapy** 

cART. (Amariles et al., 2008) In this way, the following tasks have been done:

Micromedex (Drug-REAX) computer program.

(Amariles et al., 2007a)

rosiglitazone (level 2: high risk).

lopinavir/ritonavir with nevirapine. (Oette et al., 2005) Thus, it is recommended to monitor nevirapine levels and effects, when this drug is used in combination with

> Due to ARV therapy is one of the most dynamic in terms of launching new products in the market, it is necessary to carry out structured and systematic review of publications on Pubmed/Medline and other electronic databases to identify new DIs in HIV-infected patients, receiving ART therapy, which must be complemented with evaluating the scientific evidence and classifying their clinical relevance.

> In HIV-infected patients receiving antiretroviral therapy, closer to 80% of relevant clinical pharmacokinetic DIs are associated to induction or inhibition of the systemic hepatic metabolism, which is associated to systemic clearance for a significant number drugs. While closer to 20% of relevant clinical pharmacokinetic DIs are associated to changes in bioavailability related interactions, including changes in gastrointestinal pH, presystemic metabolism and/or P-gp transport activity. Therefore, in HIV-infected patients receiving HAART of cART, the assessing and predicting clinical relevance of a given pharmacokinetic DI need to understand and use of concepts related to induction and, particularly with enzyme inhibition (mainly in CYP3 and CYP2 subfamilies) and, to a lesser extent, to bioavailability (especially with changes in a drug presystemic metabolism and/or P-gp transport activity).

> It is becoming more evident that ARV drugs are metabolized via common pathways by CYP450 enzymatic complex, which leads to an increased probability of new clinical relevant pharmacokinetic DIs due to the inhibition or stimulation of CYP isoforms (mainly in CYP3 and CYP2 subfamilies).

> In general, for a patient using more than one drug with differential capacity to modify CYP3A4 and/or P-gp enzymatic activity (some induce while some others inhibit), it is difficult to predict the net effect on the levels and effects of a drug whose metabolism is affected. (Spradling et al., 2002) Therefore, it is necessary, from a theoretical perspective, to avoid the use of such schemes, which is difficult in patients with HIV/AIDS. In this context, some studies conducted to establish the influence of several antiretroviral drugs on CYP3A activity show that ritonavir/nelfinavir inhibitory effect is maintained and it counteracts efavirenz/nevirapine inducing effect, when they are administered in combination. Additionally, it has been found that chronic administration of ritonavir (200 mg/day) or nelfinavir (2.5 g/12 hours) does not increase CYP3A activity. (Fellay et al., 2005; Mouly et al., 2006)

> Similar to other groups of patients, it is important to systematize, distribute, and use guidelines and recommendations based on the findings of studies indicating which

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 343

Amariles, P. (2007). Food consumption, cytochrome P450 3A4 (CYP3A4) presystemic

Amariles, P.; Giraldo, N. & Faus, M. (2007a) Clinical relevance of drug interactions. *Med Clin* 

Amariles, P.; Giraldo, N. & Faus, M. (2007b). Interacciones medicamentosas en pacientes

Amariles, P.; Giraldo, N.; Faus, M. & Gutiérrez, F. (2008). Proposal of a software for analysis

Amariles, P.; Lacampa, P. & Sáez-Benito, L. (2007c). Dosage adjustments for hepatic

Amsay, N.; Kenny, M.; Davies, G. & Patel, J. (2005). Complimentary and alternative

Anderson, P. & Kakuda, T. (2006). Comment on: Suboptimal CD4 gains in HIV-infected

Armstrong, S. & Cozza, K. (2003). Pharmacokinetic drug interactions of morphine, codeine,

Arrington-Sanders, R.; Hutton, N.; & Siberry, G. (2006). Ritonavir-fluticasone interaction

Andrews, E.; Glue, P.; Fang, J.; Crownover, P.; Tressler, R. & Damle, B. (2010). Assessment of

Antoniou, T.; Park-Wyllie, L. & Tseng, A. (2003). Tenofovir: A nucleotide analog for the

Back, D.; Gibbons, S. & Khoo, S. (2003). Pharmacokinetic drug interactions with nevirapine. *J* 

Barreiro, P. & Soriano, V. (2006). Suboptimal CD4 gains in HIV-infected patients receiving

Bays, H. (2006). Statin safety: An overview and assessment of the data-2005. *Am J Cardiol*,

Becher, F.; Landman, R.; Mboup, S.; Kane, C.; Canestri, A.; Liegeois, F.; et al. (2004).

Benator, D.; Weiner, M.; Burman, W.; Vernon, A.; Zhao, Z.; Khan, A. et al. (2007). Clinical

*Acquir Immune Defic Syndr*, Vol. 34, No. 1, pp. 8-14, ISSN 1077-9450.

*Pharmacol*, Vol. 69, No. 1, pp. 51-57, ISSN 1365-2125, January.

*(Barc)*, Vol. 129, pp. 27-35, ISSN 1578-8989.

*Farm Hosp*. Vol. 31, pp. 283-302, ISSN 1130-6343.

*Dis J*, Vol . 25, pp. 1044-1048, ISSN 0891-3668.

1130-6343.

December.

ISSN 1365-2141.

220-221, ISSN 1460-2091.

515-520, ISSN 1545-7206.

pp. 29-43, ISSN 0277-0008.

2091.

Baker, R. (1998). Efavirenz and Fortovase. *BETA*. Jul:9.

Vol. 97, No. 1, pp. 6-26, ISSN 0029149.

711, ISSN 2210-7711.

inhibitors, and bioavailability of saquinavir. *Farm Hosp*, Vol. 31, pp. 68-70, ISSN

infectados con el VIH: aproximación para establecer y evaluar su relevancia clínica.

of clinical relevance of antiretroviral drug interactions. *Pharm World Sci*. Vol. 30, pp.

dysfunction *Am J Health Syst Pharm*, Vol. 64, No. 24, pp. 2536-2538, ISSN 1535-2900,

medicine use among patients starting warfarin. *Br J Haematol*, Vol. 130, pp. 777-780,

patients receiving didanosine plus tenofovir. *J Antimicrob Chemother*, Vol. 58, pp.

and their derivatives: Theory and clinical reality, Part II. *Psychosomatics*, Vol .44, pp.

causing Cushing syndrome in HIV-infected children and adolescents. *Pediatr Infect* 

the pharmacokinetics of co-administered maraviroc and raltegravir. *Br J Clin* 

management of human immunodeficiency virus infection. *Pharmacotherapy*, Vol. 23,

didanosine plus tenofovir. *J Antimicrob Chemother*, Vol. 57, pp. 806-809, ISSN 1460-

Monitoring of didanosine and stavudine intracellular trisphosphorylated anabolite concentrations in HIV-infected patients. *AIDS*, Vol. 18, pp. 181-187, ISSN 1473-5571.

evaluation of the nelfinavir-rifabutin interaction in patients with tuberculosis and

combinations of a specified drug, of some specify therapeutic class, are the most appropriate for patients receiving cART or HAART. Such appropriate combinations should be chosen considering the lowest probability of the DIs, which should lead to a decrease in the use of less adequate combinations, as evidenced by studies in other countries. (Hulgan et al., 2005) The developing software based both on the severity of the effect, and on the probability of occurrence of a specify DI, including quality and quantity of literature that supports the interaction (evidence) in order to establish its clinical relevance, could be a notable contribution to the management of DIs in VIH-infected patients receiving ARV therapy.

Since proprietary databases and clinician assessment of severe DIs do not agree, developing a knowledge base for a DI alert system likely requires proprietary database information in conjunction with clinical opinion. (Smithburger et al., 2010) Thus, evaluation and prediction of relevant clinical DIs involves not only using DI alert system, but also the clinical interpretation of the alert and information, including drug history and the patient's clinical condition. In this task, a computer program that facilitates the evaluation, prediction, and decision on the clinical relevance of the DIs in HIV-infected patients receiving ART therapy have been designed, however its clinical utility requires be assessed in a study designed for this goal.

#### **9. References**


combinations of a specified drug, of some specify therapeutic class, are the most appropriate for patients receiving cART or HAART. Such appropriate combinations should be chosen considering the lowest probability of the DIs, which should lead to a decrease in the use of less adequate combinations, as evidenced by studies in other countries. (Hulgan et al., 2005) The developing software based both on the severity of the effect, and on the probability of occurrence of a specify DI, including quality and quantity of literature that supports the interaction (evidence) in order to establish its clinical relevance, could be a notable contribution to the management of DIs in VIH-infected patients receiving ARV therapy. Since proprietary databases and clinician assessment of severe DIs do not agree, developing a knowledge base for a DI alert system likely requires proprietary database information in conjunction with clinical opinion. (Smithburger et al., 2010) Thus, evaluation and prediction of relevant clinical DIs involves not only using DI alert system, but also the clinical interpretation of the alert and information, including drug history and the patient's clinical condition. In this task, a computer program that facilitates the evaluation, prediction, and decision on the clinical relevance of the DIs in HIV-infected patients receiving ART therapy have been designed, however its clinical utility requires be assessed in a study designed for

Aaron, L.; Saadoun, D.; Calatroni, I.; Launay, O.; Memain, N.; Vincent, V.; et al. (2004).

Abel, S.; Jenkins, T.; Whitlock, L.; Ridgway, C. & Muirhead, G. (2008). Effects of CYP3A4

Aberg, J. (2008). Drug-drug interactions with newer antiretroviral agents. *Top HIV Med*,

Aberg, J.; Rosenkranz, S.; Fichtenbaum, C.; Alston, B.; Brobst, S.; Segal, Y.; et al. (2006).

Abrams, D.; Hilton, J. & Leiser, R. (2003). Short-term effects of cannabinoids in patients with

Acosta, E. (2002). Pharmacokinetic enhancement of protease inhibitors. *J Acquir Immune* 

*AIDS* Patient Care. (2005). Antiviral briefs: Pregnancy category change for efavirenz. *AIDS* 

Altice, F.; Friedland, G. & Cooney, E. (1999). Nevirapine induced opiate withdrawal among

Amariles, P. (2002). Interacciones medicamentosas. En: Amariles P, editor. El medicamento.

*Infect*, Vol. 10, pp. 388-398, ISSN 1469-0691.

Vol.16, pp. 146–150, ISSN 15428826.

Vol. 139, pp. 258–266, ISSN 1539-3704.

957-962, ISSN 1473-5571.

*Defic Syndr*, Vol. 29, No. 1, pp. 11-18, ISSN 1525-4135.

*Patient Care STDS* Vol. 19, pp. 413, ISSN 1557-7449.

Tuberculosis in HIV-infected patients: A comprehensive review. *Clin Microbiol* 

inducers with and without CYP3A4 inhibitors on the pharmacokinetics of maraviroc in healthy volunteers. *Br. J. Clin. Pharmacol*, Vol. 65, No. 1, pp. 38–46,

Pharmacokinetic interaction between nelfinavir and pravastatin in HIVseronegative volunteers: ACTG Study A5108. *AIDS*, Vol. 20, pp. 725-729, ISSN

HIV-1 infection: A randomized, placebo-controlled clinical trial. *Ann Intern Med*,

injection drug users with HIV infection receiving methadone. *AIDS*, Vol. 13, pp.

Compendio básico para su utilización correcta. Medellín: *Impresos LTDA*, pp. 155-

this goal.

**9. References** 

1365-2125.

1473-5571.

246.


human immunodeficiency virus infection. *Pharmacotherapy*, Vol. 27, pp. 793-800, ISSN 0277-0008.

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 345

Caballero, J. & Nahata, M. (2005). Use of selective serotonin-reuptake inhibitors in the

Cao, Y.; Smith, P. & Wire, M. (2008). Pharmacokinetics and pharmacodynamics of

Chan, L. (2002). Drug-nutrient interaction in clinical nutrition. *Curr Opin Clin Nutr Metab* 

Clarke, S.; Mulcahy, F.; Tjia, J.; Reynolds, H.; Gibbons, S.; Barry, M.; et al. (2001). The

Coghlan, M.; Sommadossi, J.; Jhala, N.; Many, W.; Saag, M. & Johnson, V. (2001).

Colombo, S.; Buclin, T.; Franc, C.; Guignard, N.; Khonkarly, M.; Tarr, P.; et al. (2006).

Cooper, K.; Martin, P.; Dane, A.; Warwick, M.; Schneck, D. & Cantarini, M. (2003). Effect of

Currier, M.; Molina, G. & Kato, M. (2004). Citalopram treatment of major depressive

Cvetkovic, R. & Goa, K. (2003). Lopinavir/ritonavir: A review of its use in the management

Dagan, T.; Sable, C.; Bray, J. & Gerschenson, M. (2002). Mitochondrial dysfunction and

Dailly, E.; Allavena, C.; Raffi, F. & Jolliet, P. (2005). Pharmacokinetic evidence for the

Dailly, E.; Tribut, O.; Tattevin, P.; Arvieux, C.; Perré, P.; Raffi, F.; et al. (2006). Influence of

Dalakas, M.; Semino-Mora, C. & Leon-Monzon, M. (2001). Mitochondrial alterations with

Damle, B.; Hess, H.; Kaul, S. & Knupp, C. (2002a). Absence of clinically relevant drug

of HIV infection. *Drugs*, Vol. 63, pp. 769-802, ISSN 0012-6667.

antiretrovirals. *Curr Infect Dis Rep* Vol. 1, pp. 89-96.

*Care*. Vol. 5, pp. 327-332, ISSN 1473-6519.

213-217, ISSN 1365-2125.

33, pp. 1914-1921, ISSN 1537-6591.

11, pp. 53-62, ISSN 2040-2058.

pp. 322-329, ISSN 0009-9236.

Vol. 45, pp. 210-216, ISSN 1545-7206.

Vol. 1, pp. 397-412, ISSN 1567-7249.

34, ISSN 1365-2125.

526, ISSN 1432-1041.

1544, ISSN 1530-0307.

ISSN 0012-6667.

treatment of depression in adults with HIV. *Ann Pharmacother*, Vol. 39, pp. 141-145,

methadone enantiomers after coadministration with fosamprenavir-ritonavir in opioid-dependent subjects. *Pharmacotherapy*, Vol. 28, pp. 863–874, ISSN 0277-0008. Catanzaro, L.; Slish, J.; DiCenzo, R. & Morse, G. (2004). Drug interactions with

pharmacokinetics of methadone in HIV-positive patients receiving the nonnucleoside reverse transcriptase inhibitor efavirenz. *Br J Clin Pharmacol*, Vol. 51, pp.

Symptomatic lactic acidosis in hospitalized antiretroviral treated patients with human immunodeficiency virus infection: A report of 12 cases. *Clin Infect Dis*, Vol.

Ritonavir-boosted atazanavir-lopinavir combination: A pharmacokinetic interaction study of total, unbound plasma and cellular exposures. *Antivir Ther* Vol.

itraconazole on the pharmacokinetics of rosuvastatin. *Clin Pharmacol Ther*, Vol. 73,

disorder in Hispanic HIV and *AIDS* patients: A prospective study. Psychosomatics,

antiretroviral nucleoside analogue toxicities: What is the evidence? *Mitochondrion*

induction of lopinavir metabolism by efavirenz. *Br J Clin Pharmacol*, Vol. 60, pp. 32-

tenofovir, nevirapine and efavirenz on ritonavir-boosted atazanavir pharmacokinetics in HIV-infected patients. Eur *J Clin Pharmacol*, Vol. 62, pp. 523-

mitochondrial DNA depletion in the nerves of *AIDS* patients with peripheral neuropathy induced by 2'3'dideoxycytidine (ddC). *Lab Invest*, Vol. 81, pp. 1537-

interactions following simultaneous administration of didanosine-encapsulated,


Benesic, A.; Zilly, M.; Kluge, F.; Weissbrich, B.; Winzer, R.; Klinker, H.; et al. (2004). Lipid

Bergshoeff, A.; Fraaij, P. & Ndagijimana, J. (2005). Increased dose of lopinavir/ritonavir

children. *J Acquir Immune Defic Syndr*, Vol. 39, pp. 63-68, ISSN 1077-9450. Blumberg, H.; Burman, W.; Chaisson, R.; Daley, C.; Etkind, S.; Friedman, L.; et al. (2003).

Boffito, M.; Carriero, P. & Trentini, L. (2002). Pharmacokinetics of saquinavir co-

Boffito, M.; Maitland, D. & Pozniak, A. (2006). Practical perspectives on the use of tipranavir

Boffito, M.; Pozniak, A.; Kearney, B.; Higgs, C.; Mathias, A.; Zhong, L.; et al. (2005). Lack of

Bonate, P.; Reith, K. & Weir, S. (1998). Drug interactions at the renal level. Implications for drug development. *Clin Pharmacokinet*, Vol. 34, pp. 375-404, ISSN 1179-1926. Borin, M.; Chambers, J.; Carel, B.; Gagnon, S. & Freimuth, W. (1997). Pharmacokinetic study

Boubaker, K.; Flepp, M.; Sudre, P.; Furrer, H.; Haensel, A.; Hirschel, B.; et al. (2001).

Brennan, P.; Lyus, R.; Harrison, T.; Pakianathan, M. & Macallan, D. (2005). Pharmacokinetic

Brophy, D.; Israel, D.; Pastor, A.; Gillotin, C.; Chittick, G.; Symonds, W.; et al. (2000).

Bruce, R.; Altice, F.; Gourevitch, M. & Friedland, G. (2006). Pharmacokinetic drug

Burger, D.; Agarwala, S.; Child, M; Been-Tiktak, A.; Wang, Y. & Bertz, R. (2006). Effect of

studies. *J Clin Pharmacol*, Vol. 46, pp. 130-139, ISSN 0091-2700.

*Infection*, Vol . 32, pp. 229-233, ISSN 0153-4453.

Vol. 167, pp. 603-662, ISSN 0003-0805.

Vol. 61, pp. 544-553, ISSN 1532-6535.

Vol. 41, pp. 563-572, ISSN 1077-9450.

*Infect Dis* Vol. 33, pp. 1931-1937, ISSN 1537-6591.

ISSN 0277-0008.

1460-2091.

6569.

human immunodeficiency virus infection. *Pharmacotherapy*, Vol. 27, pp. 793-800,

lowering therapy with fluvastatin and pravastatin in patients with HIV infection and antiretroviral therapy: Comparison of efficacy and interaction with indinavir.

compensates for efavirenz-induced drug-drug interaction in HIV-1-infected

American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: Treatment of tuberculosis. *Am J Respir Crit CareMed*,

administered with cimetidine. *J Antimicrob Chemother*, Vol. 50, pp. 1081-1084, ISSN

in combination with other medications: Lessons learned from pharmacokinetic

pharmacokinetic drug interaction between tenofovir disoproxil fumarate and nelfinavir mesylate. *Antimicrob Agents Chemother*, Vol. 49, pp. 4386-4389, ISSN 1098-

of the interaction between rifampin and delavirdine mesylate. *Clin Pharmacol Ther*,

Hyperlactatemia and antiretroviral therapy: The Swiss HIV Cohort Study. *Clin* 

interactions between efavirenz and rifampicin in the treatment of HIV and tuberculosis: One size does not fit all. *AIDS*, Vol. 19, pp. 1541-1543. ISSN 1473-5571.

Pharmacokinetic interaction between amprenavir and clarithromycin in healthy male volunteers. *Antimicrob Agents Chemother*, Vol. 44, pp. 978-984, ISSN 1098-6596.

interactions between opioid agonist therapy and antiretroviral medications: Implications and management for clinical practice. *J Acquir Immune Defic Syndr*,

rifampicin on steady-state pharmacokinetics of atazanavir with ritonavir in healthy volunteers. *Antimicrob Agents Chemother*, Vol. 50, pp. 3336-3342, ISSN 1098-6596.


enteric-coated bead formulation with either itraconazole or fluconazole. *Biopharm Drug Dispos*, Vol. 23, pp. 59-66, ISSN 0142-2782.

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 347

interactions. Drug Metab Dispos, Vol. 35, pp.1853–1859, ISSN 0090-9556. Droste, J.; Kearney, B.; Hekster, Y. & Burger, D. (2006). Assessment of drug-drug

Droste, J.; Verweij-van, C.; Kearney, B.; Buffels, R.; Vanhorssen, P.; Hekster, Y.; et al. (2005).

FDA. (1999). US Food and Drug Administration. Guidance for Industry: In vivo Drug

FDA. (2005). Food and Drug Administration. Title 21, Volume 5: CFR 320.33. Code of

http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=32

Fellay, J.; Marzolini, C.; Decosterd, L.; Golay, K.; Baumann, P.; Buclin, T.; et al. (2005).

Fichtenbaum, C. & Gerber J. (2002). Interactions between antiretroviral *Drugs* and *Drugs*

infection. *Clin Pharmacokinet*, Vol . 41, pp. 1195-1211, ISSN 84-7666-135-5. Fichtenbaum, C.; Gerber, J.; Rosenkranz, S.; Segal, Y.; Aberg, J.; Blaschke, T.; et al. (2002).

Finch, C.; Chrisman, C.; Baciewicz, A. & Self, T. (2002). Rifampin and rifabutin drug interactions: An update. *Arch Intern Med*, Vol. 162, pp. 985-989, ISSN 1538-3679. Fisher, M.; Talbot, G.; Maislin, G.; McKeon, B.; Tynan, K. & Strom, B. (1989). Risk factors for amphotericin B-associated nephrotoxicity. *Am J Med*, Vol. 87, pp. 547-552. Fleischer, R.; Boxwell, D. & Sherman, K. (2004). Nucleoside analogues and mitochondrial

Fletcher, C.; Acosta, E.; Cheng, H.; Haubrich, R.; Fischl, M.; Raasch, R.; et al. (2000).

Ford, S.; Wire, M.; Lou, Y.; Baker, K. & Stein DS. (2005). Effect of antacids and ranitidine on

Ford, J.; Boffito, M.; Maitland, D.; Hill, A.; Back, D.; Khoo, S.; et al. (2006). Influence of

Competing drug-drug interactions among multidrug antiretroviral regimens used in the treatment of HIV- infected subjects: ACTG 884. *AIDS*, Vol. 14, No. 16, pp.

the single-dose pharmacokinetics of fosamprenavir. *Antimicrob Agents Chemother*,

atazanavir 200 mg on the intracellular and plasma pharmacokinetics of saquinavir

toxicity. *Clin Infect Dis*, Vol. 38, pp. 79-80, ISSN 1537-6591.

patients. Eur *J Clin Pharmacol*, Vol. 60, pp 865-873, ISSN 1432-1041.

Variations of CYP3A activity induced by antiretroviral treatment in HIV-1 infected

used for the therapy of the metabolic complications encountered during HIV

Pharmacokinetic interactions between protease inhibitors and statins in HIV seronegative volunteers: ACTG Study A5047. *AIDS*, Vol . 16, pp. 569-577, ISSN

*Immune Defic Syndr*, Vol. 41, pp. 37-43, ISSN 1077-9450.

Etravirine: R165335, TMC 125, TMC-125, TMC125. *Drugs* R & D 2006;7(6):367–73.

6569.

0.24.

1473-5571.

2495-2501, ISSN 1473-5571.

Vol. 49, pp. 467-469, ISSN 1098-6569.

MD, November 1999.

Federal Regulations. Available from:

protease inhibitors in human hepatocytes: Implications for predicting clinical drug

interactions between tenofovir disoproxil fumarate and the nonnucleoside reverse transcriptase inhibitors nevirapine and efavirenz in HIV-infected patients. *J Acquir* 

Pharmacokinetic study of tenofovir disoproxil fumarate combined with rifampin in healthy volunteers. *Antimicrob Agents Chemother*, Vol. 49, pp. 680-684, ISSN 1098-

Metabolism/Drug Interaction Studies - Study Design, Data Analysis, and Recommendations for Dosing and Labeling. Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER), Rockville,


Damle, B.; Mummaneni, V.; Kaul, S. & Knupp, C. (2002b). Lack of effect of simultaneously

Damle, B.; Yan, J.; Behr, D.; O'Mara, E.; Nichola, P.; Kaul, S.; et al. (2002c). Effect of food on

Daveluy, A.; Raignoux, C.; Miremont-Salamé, G.; Girodet, P.; Moore, N.; Haramburu, F.; et

European *AIDS* Conference. Madrid, Spain. Oct 24-27. Abstract P4.3/02. De Jong, B.; Israelski, D.; Corbett, E. & Small, P. (2004). Clinical management of

De Maat, M.; Ekhart, G.; Huitema, A.; Koks, C.; Mulder, J. & Beijnen J. (2003). Drug

DeJesus, E.; Piliero, P.; Summers, K.; Wire, M.; Stein, D.; Masterman, A.; et al. (2006).

DeSilva, K.; Le Flore, D.; Marston, B. & Rimland, D. (2001). Serotonin syndrome in HIV-

DHHS. (2005). Panel on Antiretroviral Guidelines for Adult and Adolescents. Guidelines for

of Health and Human Services. In: *AIDS* INFO, 10.01.2011; Available from: http://www.*AIDS*info.nih.gov/contentfiles/AdultandAdolescentGL.pdf. ISBN

DiCenzo, R.; Peterson, D.; Cruttenden, K.; Mariuz, P.; Rezk, N.; Hochreiter, J.; et al. (2008).

Ding, R.; Tayrouz, Y.; Riedel, K.; Burhenne, J.; Weiss, J.; Mikus, G.; et al. (2004). Substantial

Dixit, V.; Hariparsad, N.; Li, F.; Desai, P.; Thummel, K. & Unadkat J. (2007). Cytochrome

*Clin Pharmacol Ther*, Vol. 76, pp. 73-84, ISSN 1532-6535.

*Drug Dispos*, Vol. 23, pp. 59-66, ISSN 0142-2782.

*Chemother*, Vol. 46, pp. 385-391, ISSN 1098-6569.

*Pharmacol*, Vol. 42, pp. 419-427, ISSN 0091-2700.

*Pharmacokinet*, Vol. 42; pp. 223-282, ISSN 0312-5963.

Vol. 50, pp. 3157-3159, ISSN 1098-6569.

pp. 1281-1285, ISSN 1473-5571.

ISSN 0066-4219.

1588-080048

0066-4804.

enteric-coated bead formulation with either itraconazole or fluconazole. *Biopharm* 

administered didanosine encapsulated enteric bead formulation (Videx EC) on oral absorption of indinavir, ketoconazole, or ciprofloxacin. *Antimicrob Agents* 

the oral bioavailability of didanosine from encapsulated enteric-coated beads. *J Clin* 

al. (2009). Drug interactions between inhaled corticosteroids and enzymatic inhibitors. Eur *J Clin Pharmacol*, Vol.65, No. 7, pp. 743-745, ISSN 1432-1041. July Davis, J.; Schöller-Gyre, M.; Kakuda, T.; Ridgway, C.; Tweedy, S.; Ndongo, N.; et al. (2007).

An open, randomized, two period, crossover study in 2 cohorts to investigate the effect of steady state TMC125 and the combination of TMC/darunavir/ritonavir on the steady state pharmacokinetics of oral maraviroc in healthy subjects. 11th

tuberculosis in the context of HIV infection. *Annu Rev Med*, Vol. 55, pp. 283-301,

interactions between antiretroviral *Drugs* and comedicated agents. *Clin* 

Interaction between fosamprenavir, with and without ritonavir, and nevirapine in human immunodeficiency virus-infected subjects. *Antimicrob Agents Chemother*,

infected individuals receiving antiretroviral therapy and fluoxetine. *AIDS*, Vol .15,

the use of antiretroviral agents in HIV-infected adults and adolescents. Department

Effects of minocycline and valproic acid coadministration on atazanavir plasma concentrations in human immunodeficiency virus-infected adults receiving atazanavir-ritonavir. *Antimicrob Agents Chemother*, Vol. 52, pp. 3035–3059, ISSN

pharmacokinetic interaction between digoxin and ritonavir in healthy volunteers.

P450 enzymes and transporters induced by antihuman immunodeficiency virus

protease inhibitors in human hepatocytes: Implications for predicting clinical drug interactions. Drug Metab Dispos, Vol. 35, pp.1853–1859, ISSN 0090-9556.


and ritonavir 1600/100 mg administered once daily in HIV-infected patients. *J Antimicrob Chemother*, Vol. 58, pp. 1009-1016, ISSN 1460-2091.

Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 349

Havlir, D.; Tierney, C.; Friedland, G.; Pollard, R.; Smeaton, L.; Sommadossi, J.; et al. (2000).

Heck, A.; DeWitt, B. & Lukes, A. (2000). Potential interactions between alternative therapies and warfarin. *Am J Health Syst Pharm*, Vol. 57, pp. 1221-1227, ISSN 1535-2900. Ho, R. & Kim, R. (2005). Transporters and drug therapy: implications for drug disposition and disease. *Clin Pharmacol Ther*, Vol. 78, pp. 260-277, ISSN 1532-6535. Hochster, H.; Dieterich, D.; Bozzette, S.; Reichman, R.; Connor, J.; Liebes, L.; et al. (1990).

Hogeland, G.; Swindells, S.; McNabb, J.; Kashuba, A.; Yee, G. & Lindley, C. (2007).

Hoggard, P.; Sales, S.; Phiboonbanakit, D.; Lloyd, J.; Maher, B.; Khoo, S.; et al. (2001).

Huang, L.; Wring, S.; Woolley, J.; Brouwer, K.; Serabjit-Singh, C. & Polli, J. (2001). Induction

Hulgan, T.; Sterling, T.; Daugherty, J.; Arbogast, P.; Raffanti, S. & Ray, W. (2005). Prescribing

persons. *J Acquir Immune Defic Syndr*, Vol. 38, pp. 277-282, ISSN 1077-9450. Hughes, C.; Freitas, A. & Miedzinski LJ. (2007). Interaction between lopinavir / ritonavir

Isbister, G. & Buckley, N. (2005). The pathophysiology of serotonin toxicity in animals and

Izzedine, H.; Launay-Vacher, V.; Baumelou, A. & Deray, G. (2004). Antiretroviral and

Izzo, A. (2004). Drug interactions with St. John's Wort (Hypericum perforatum): A review

Jacobson, T. (2004). Comparative pharmacokinetic interaction profiles of pravastatin,

Jain, A.; Venkataramanan, R.; Shapiro, R.; Scantlebury, V.; Potdar, S.; Bonham, C.; et al.

humans. *Clin Neuropharmacol*, Vol. 28, pp. 205-214, ISSN 0362-5664.

inhibitors. Am J Cardiol, Vol . 94, pp. 1140-1146, ISSN 0029149.

and warfarin. *CMAJ*, Vol. 177, pp. 357–359, ISSN 1488-2329.

Opportunistic Infections.Denver (CO). 5-8 Feb. Abstract 575.

*Clin Pharmacol Ther* Vol. 81, pp. 69-75, ISSN 1532-6535.

*Dis*, Vol. 182, pp. 321-325, ISSN 1537-6613.

113, pp. 111-117, ISSN 1539-3704.

No. 8, pp. 2943-2947, ISNN 1098-6569.

532-541, ISSN 1523-1755.

4879.

*Dispos* Vol. 29, pp. 754-760, ISSN 0090-9556.

lopinavir and ritonavir in HIV+adults. 13th Conference on Retroviruses and

In vivo antagonism with zidovudine plus stavudine combination therapy. *J Infect* 

Toxicity of combined ganciclovir and zidovudine for cytomegalovirus disease associated with *AIDS*. An *AIDS* Clinical Trials Group Study. *Ann Intern Med*, Vol.

Lopinavir/ritonavir reduces bupropion plasma concentrations in healthy subjects.

Influence of prior exposure to zidovudine on stavudine phosphorylation in vivo and ex vivo. *Antimicrob Agents Chemother*, Vol. 45, pp. 577-582, ISSN 1098-6569. Holdich, T.; Shiveley, L. & Sawyer, J. (2007). Effect of Lamivudine on the plasma and

intracellular pharmacokinetics of apricitabine, a novel nucleoside reverse transcriptase inhibitor, in healthy volunteers. *Antimicrob Agents Chemother*, Vol. 51.

of P-glycoprotein and cytochrome P450 3A by HIV protease inhibitors. *Drug Metab* 

of contraindicated protease inhibitor and statin combinations among HIV-infected

immunosuppressive drug-drug interactions: An update. *Kidney Int*, Vol. 66, pp.

of the clinical evidence. Int J *Clin Pharmacol Ther*, Vol. 42, pp. 139-148, ISSN 0174-

simvastatin, and atorvastatin when coadministered with cytochrome P450

(2002). The interaction between antiretroviral agents and tacrolimus in liver and kidney transplant patients. *Liver Transpl*. Vol. 8, pp. 841-845, ISSN 1527-6473.


Friedland, G.; Andews, L. & Schreibman, T. (2005). Lack of an effect of atazanavir on steady-

Friedland, G.; Khoo, S.; Jack, C. & Lalloo, U. (2006). Administration of efavirenz (600

Gaikwad, R.; Sketris, I.; Shepherd, M. & Duffy, J. (2007). Evaluation of accuracy of drug

infected patients. *Br J Clin Pharmacol*, Vol. 48, pp. 168-179, ISSN 1365-2125. Gerber, J.; Rosenkranz, S.; Fichtenbaum, C.; Vega, J.; Yang, A.; Alston, B.; et al. (2005). Effect

Giraldo, N.; Amariles, P.; Gutiérrez, F.; Monsalve, M. & Faus, M. (2010). Aproximación para

Gray, D.; Roux, P.; Carrihill, M. & Klein, M. (2010). Adrenal suppression and Cushing's

Grimes, R.; Lal, L. & Lewis, S. (2002). Frequency of medical history items, drug interactions,

Gruber, V. & McCance-Katz, E. (2010). Methadone, buprenorphine, and street drug

Guo, Y. & Fung, H. (2004). Fatal lactic acidosis associated with coadministration of

Hamzeh, F.; Benson, C.; Gerber, J.; Currier, J.; McCrea, J.; Deutsch, P.; et al. (2003). Steady-

Hantson, P.; Di Fazio, V. & Wallemacq, P. (2010). Toxicokinetic interaction between

Hare, C.; Vu, M.; Grunfeld, C. & Lampiris, H. (2002). Simvastatin-nelfinavir interaction

Harris, M.; Zala, C.; Ramírez, S.; Woodfall, B.; Peeters, M.; Scholler, M.; et al. (2006).

healthcare. *Health Informatics J*, Vol. 13, pp. 163-177, ISSN 1741-2811. Gallicano, K.; Sahai, J.; Shukla, V.; Seguin, I.; Pakuts, A.; Kwok, D.; et al. (1999). Induction of

*Antimicrob Chemother*, Vol. 58, pp. 1009-1016, ISSN 1460-2091.

addiction. *AIDS*, Vol. 19, pp. 1635–1641, ISSN 1473-5571.

58, pp. 1299-1302. ISSN 1460-2091.

39, pp. 307-312, ISSN 1077-9450.

296-297, ISSN 2078-5131.

152-160, ISSSN 1548-3576.

1094, ISSN 0277-0008.

1058-4838.

*Farm Hosp*, Vol. 34, pp. 90-93, ISSN 1130-6343.

*Clin Trials*, Vol. 3, pp. 161-167, ISSN 1528-4336.

*Pharmacol Ther* Vol. 73, pp. 159-69, ISSN 1532-6535.

*Lett*, Vol.4, No. 1, pp.7-8, ISSN 1872-3128.

and ritonavir 1600/100 mg administered once daily in HIV-infected patients. *J* 

state pharmacokinetics of methadone in patients chronically treated for opiate

mg/day) with rifampicin results in highly variable levels but excellent clinical outcomes in patients treated for tuberculosis and HIV. *J Antimicrob Chemother*, Vol.

interaction alerts triggered by two electronic medical record systems in primary

zidovudine glucuronidation and amination pathways by rifampicin in HIV-

of efavirenz on the pharmacokinetics of simvastatin, atorvastatin, and pravastatin: Results of AIDS Clinical Trials Group 5108 Study. *J Acquir Immune Defic Syndr*, Vol.

establecer y evaluar la relevancia clínica de las interacciones medicamentosas en pacientes infectados con virus de la inmunodeficiencia humana: actualización 2009.

syndrome secondary to ritonavir and budesonide. S Afr Med J, Vol. 100, No. 5, pp.

and lifestyle characteristics that may interfere with antiretroviral medications. *HIV* 

interactions with antiretroviral medications. *Curr HIV/AIDS Rep*, Vol. 7, No. 3, pp.

didanosine and tenofovir disoproxil fumarate. *Pharmacotherapy*, Vol. 24, pp. 1089-

state pharmacokinetic interaction of modified-dose indinavir and rifabutin. *Clin* 

quetiapine and antiretroviral therapy following quetiapine overdose. *Drug Metab* 

implicated in rhabdomyolysis and death. *Clin Infect Dis*, Vol . 35, pp. 111-112, ISSN

Pharmacokinetics and safety of adding TMC125 to stable regimens of saquinavir,

lopinavir and ritonavir in HIV+adults. 13th Conference on Retroviruses and Opportunistic Infections.Denver (CO). 5-8 Feb. Abstract 575.


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 351

Laganiere, S.; Davies, R.; Carignan, G.; Foris, K.; Goernert, L.; Carrier, K.; et al. (1996).

Levêque, D.; Santucci, R.; Pavillet, J.; Herbrecht, R. & Bergerat J. (2009). Paralytic ileus

Lim, M.; Min, S.; Eron, J.; Bertz, R.; Robinson, M.; Gaedigk, A.; et al. (2004).

Lin, J. & Yamazaki, M. (2003). Role of P-glycoprotein in pharmacokinetics: Clinical

Lin, J. (2003). Drug-drug interaction mediated by inhibition and induction of P-glycoprotein.

Maddox, R.; Groves, W. & Ensom, R. (1980). Automated pharmaceutical assistance systems.

Makinson, A.; Pujol, J.; Le Moing, V.; Peyriere, H. & Reynes J. (2010). Interactions between

Manosuthi, W.; Sungkanuparph, S.; Thakkinstian, A.; Vibhagool, A.; Kiertiburanakul, S.;

Manosuthi, W.; Kiertiburanakul, S.; Sungkanuparph, S.; Ruxrungtham, K.; Vibhagool, A.:

Markowitz, J. & DeVane, C. (2001). The emerging recognition of herb-drug interactions with

Martinez, E.; Milinkovic, A.; de Lazzari, E.; Ravasi, G.; Blanco, J.; Larrousse, M:; et al. (2004).

tenofovir in HIV-infected adults. *Lancet*, Vol. 364, pp. 65-67, ISSN 0140-6736. Masía, M.; Gutiérrez, F.; Padilla, S.; Ramos, J. & Pascual, J. (2005). Severe toxicity associated

implications. *Clin Pharmacokinet*, Vol. 42, pp. 59-98, ISSN 0312-5963.

quinidine. *Clin Pharmacol Ther*, Vol. 60, pp. 255-264, ISSN 1532-6535. Launay, V.; Izzedine, H.; Karie, S.; Hulot, J.; Baumelou, A. & Deray, G. (2006). Renal tubular drug transporters. *Nephron Physiol*, Vol. 103, pp. 97-106, ISSN 1660-2137. Lee, L.; Andrade, A. & Flexner, C. (2006). Interactions between natural health products and

*Pharm World Sci*, Vol. 31, No. 6, pp. 619-621, ISSN 2210-7711.

*Adv Drug Deliv Rev*, vol. 55, pp. 53-81, ISSN 0169-4092.

rifampicin. *AIDS*, Vol. 19, pp. 1481-1486, ISSN 1473-5571.

*Med Syst*, Vol. 4, pp. 151-168, ISSN 0148-5598.

*AIDS* , Vol. 20, pp. 131-132, ISSN 1473-5571.

*STD AIDS* Vol. 16, pp. 646-648, ISSN 0956-4624.

Vol. 45, pp. 211-218, ISSN 0091-2700.

Vol. 43, pp. 1052-1059, ISSN 1537-6591.

36, pp. 1034-40, ISSN 1077-9450.

ISSN 1556-1380.

pp. 53-64.

coated capsule formulation of didanosine in healthy volunteers. *J Clin Pharmacol*,

Pharmacokinetic and pharmacodynamic interactions between diltiazem and

antiretroviral *Drugs*: Pharmacokinetic and pharmacodynamic effects. *Clin Infect Dis*

possibly associated with interaction between ritonavir/lopinavir and vincristine.

Coadministration of lopinavir/ritonavir and phenytoin results in twoway drug interaction through cytochrome P-450 induction. *J Acquir Immune Defic Syndr*. Vol.

I. Clinical pharmacokinetics. II. Drug interference with laboratory tests results. *J* 

cytotoxic chemotherapy and antiretroviral treatment in human immunodeficiency virus-infected patients with lung cancer. *J Thorac Oncol*, Vol. 5, No. 4, pp. 562-571,

Rattanasiri, S.; et al. (2005). Efavirenz levels and 24- week efficacy in HIV-infected patients with tuberculosis receiving highly active antiretroviral therapy and

Rattanasiri, S.; et al. (2006). Efavirenz 600 mg/day versus efavirenz 800 mg/day in HIV-infected patients with tuberculosis receiving rifampicin: 48 weeks results.

a focus on St. John's wort (Hypericum perforatum). *Psychopharmacol Bull*. Vol. 35,

Pancreatic toxic effects associated with co-administration of didanosine and

with the combination of tenofovir and didanosine: Case report and review. *Int J* 


Josephson, F. (2010). Drug-drug interactions in the treatment of HIV infection: focus on

Justesen, U. (2006). Therapeutic drug monitoring and human immunodeficiency virus (HIV)

Kakuda, T. & Falcon, R. (2006). Effect of food and ranitidine on saquinavir pharmacokinetics

Kakuda, T.; Schöller-Gyüre, M.; Woodfall, B.; De Smedt, G.; Peeters, M.; Vandermeulen, K.;

Kashuba, A.; Tierney, C.; Downey, G.; Acosta, E.; Vergis, E.; Klingman, K.; et al. (2005).

Kashuba, A. (2005). Drug-drug interactions and the *Pharmacotherapy* of HIV infection. Top

Katzung, B. (2009). Basic and Clinical Pharmacology. 11th ed. McGraw Hill Medical, ISBN

Kearney, B.; Sayre, J.; Flaherty, J.; Chen, S.; Kaul, S. & Cheng, A. (2005). Drug-drug and

Khaliq, Y.; Gallicano, K., Venance, S.; Kravcik, S. & Cameron, D. (2000). Effect of

Kiang, T.; Ensom, M. & Chang, T. (2005). UDP-glucuronosyltransferases and clinical drugdrug interactions. *Pharmacol Ther,* Vol. 106, pp. 97-132, ISSN 0163-7258. Kiser, J.; Bumpass, J.; Meditz, A.; Anderson, P.; Bushman, L.; Ray, M.; et al. (2010). Effect of

Kraft, W.; McCrea, J.; Winchell, G.; Carides, A.; Lowry, R.; Woolf, E.; et al. (2004). Indinavir

Krikorian, S. & Rudorf, D. (2005). Drug-drug interactions and HIV therapy: What should pharmacists know? *J Pharm Pract*, Vol. 18, pp. 278-294, ISSN 0897-1900. La Porte, C.; Colbers, E.; Bertz, R.; Voncken, D.; Wikstrom, K.; Boeree, M.; et al. (2004).

La Porte, C.; Verweij-van, C.; van Ewijk, N.; Aarnoutse, R.; Koopmans, P.; Reiss, P.; et al.

Infection. Glasgow. 12-16 November. Oral presentation PL5.2.

No. 6, pp. 530-539, ISSN 1365-2796, December.

7843.

0277-0008.

0071604057.

5003, ISSN 1098-6569.

305-313, ISSN 1552-4604.

6596.

pp. 145-152, ISSN 1473-5571.

HIV Med, Vol. 13, pp. 64-69, ISSN 1542-8826.

*Clin Pharmacol*, Vol. 45, pp. 1360-1367, ISSN 0091-2700.

*Clin Pharmacol Ther*, Vol. 68, pp. 637-646, ISSN 1532-6535.

pharmacokinetic enhancement through CYP3A inhibition. *J Intern Med*, Vol. 268,

antiretroviral therapy. *Basic Clin Pharmacol Toxicol*, Vol. 98, pp. 20-31, ISSN 1742-

and gastric pH in healthy volunteers. *Pharmacotherapy* Vol. 26, pp. 1060-1068, ISSN

et al. (2006). TMC125 in combination with other medications: summary of drugdrug interaction studies. 8th International Congress on Drug Therapy in HIV

Combining fosamprenavir with lopinavir/ritonavir substantiallyreduces amprenavir and lopinavir exposure: ACTG protocol A5143 results. *AIDS*, Vol. 19,

drug-food interactions between tenofovir disoproxil fumarate and didanosine. *J* 

ketoconazole on ritonavir and saquinavir concentrations in plasma and cerebrospinal fluid from patients infected with human immunodeficiency virus.

antacids on the pharmacokinetics of raltegravir in human immunodeficiency virusseronegative volunteers. *Antimicrob Agents Chemother*, Vol. 54, No. 12, pp. 4999-

and rifabutin drug interactions in healthy volunteers. *J Clin Pharmacol*, Vol. 44, pp.

Pharmacokinetics of adjusted-dose lopinavir-ritonavir combined with rifampin in healthy volunteers. *Antimicrob Agents Chemother*, Vol. 48, pp. 1553-1560, ISSN 1098-

(2005). Pharmacokinetic interaction study of indinavir/ritonavir and the enteric-

coated capsule formulation of didanosine in healthy volunteers. *J Clin Pharmacol*, Vol. 45, pp. 211-218, ISSN 0091-2700.


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 353

Oette, M.; Kurowski, M.; Feldt, T.; Kroidl, A.; Sagir, A.; Vogt, C.; et al. (2005). Impact of

Ogbuokiri, J. (2009). Optimizing care for patients on maraviroc salvage therapy. *HIV Clin* 

Oldfield, V. & Plosker, G. (2006). Lopinavir/ritonavir: A review of its use in the management of HIV infection. *Drugs*, Vol. 66, pp. 1275-1299, ISSN 0012-6667. Ouellet, D.; Hsu, A.; Qian, J.; Locke, C.; Eason, C.; Cavanaugh, J.; et al. (1998). Effect of

Pau, A. & Boyd, S. (2010). Recognition and management of significant drug interactions in

Penzak, S.; Shen, J.; Alfaro, R.; Remaley, A.; Natarajan, V. & Falloon J. (2004). Ritonavir

Perronne, C. (2006). Antiviral hepatitis and antiretroviral drug interactions. *J Hepatol*, Vol.

Petit, F.; Fromenty, B.; Owen, A. & Estaquier, J. (2005). Mitochondria are sensors for HIV

Piscitelli, S & Gallicano, K. (2001). Interactions among *Drugs* for HIV and opportunistic

Polk, R.; Brophy, D.; Israel, D.; Patron, R.; Sadler, B.; Chittick; G.; et al. (2001).

Rathbun, R. & Rossi, D. (2002). Low-dose ritonavir for protease inhibitor pharmacokinetic enhancement. *Ann Pharmacother* Vol. 36, pp. 702-706, ISSN 0012-6667. Ray, A.; Olson, L. & Fridland, A. (2004). Role of purine nucleoside phosphorylase in

*Antimicrob Agents Chemother* Vol. 48, pp. 1089-1095, ISSN 1098-6569. Ray, A. (2005). Intracellular interactions between nucleos(t)ide inhibitors of HIV reverse

Pharmacokinetic interaction between amprenavir and rifabutin or rifampin in healthy males. *Antimicrob Agents Chemother*, Vol. 45, pp. 502-508, ISSN 1098-6596. Polk, R.; Crouch, M.; Israel, D.; Pastor, A.; Sadler, B.; Chittick, G.; et al. (1999).

Pharmacokinetic interaction between ketoconazole and amprenavir after single doses in healthy men. *Pharmacotherapy*, Vol.19, pp. 1378-1384, ISSN 0277-0008. Ramachandran, G.; Hemanthkumar, A.; Rajasekaran, S.; Padmapriyadarsini, C.; Narendran,

G.; Sukumar, B.; et al. (2006). Increasing nevirapine dose can overcome reduced bioavailability due to rifampicin coadministration. *J Acquir Immune Defic Syndr*,

interactions between 2',3'-dideoxyinosine and allopurinol, ganciclovir, or tenofovir.

*Drugs*. *Trends Pharmacol Sci*, Vol. 26, pp. 258-264, ISSN 0165-6147.

infections. *N Engl Med*, Vol . 344, pp. 984-996, ISSN 0028-4793.

MDR1 genotypes. *Ther Drug Monit*, Vol. 26, pp. 322-330, ISSN 1536-3694. Pérez, V.; Sánchez, P. & Villar, S. (2009). Etravirine drug interactions. *Enferm Infecc Microbiol* 

*Clin Pharmacol Ther*, Vol. 78, pp. 582-592, ISSN 1532-6535.

*Summer*, Vol. 21, No. 3, pp. 11-15, ISSN 1551-8850

*Ther*, Vol. 88, No. 5, pp. 712-719, ISSN 1532-6535.

*Clin*, Vol. 27, No. 2, pp. 27-31, ISSN 1578-1852.

44, No. 1, pp. 119-125.

Vol. 42, pp. 36-41. ISSN 1077-9450.

transcriptase. *AIDS* Rev, Vol. 7, pp. 113-125.

*Br J Clin Pharmacol*, Vol. 46, pp. 111-116, ISSN 1365-2125.

2091.

qualitative relationships, quantitative predictions, and the rank-order approach.

rosiglitazone treatment on the bioavailability of antiretroviral compounds in HIV-positive patients. *J Antimicrob Chemother*, Vol. 56, pp. 416-419, ISSN 1460-

ritonavir on the pharmacokinetics of ethinyl estradiol in healthy female volunteers.

HIV patients: challenges in using available data to guide therapy. *Clin Pharmacol* 

decreases the nonrenal clearance of digoxin in healthy volunteers with known


Matteelli, A.; Regazzi, M.; Villani, P.; De Iaco, G.; Cusato, M.; Carvalho, A.; et al. (2007).

Mauss, S.; Scholten, S.; Wolf, E.; Berger, F.; Schmutz, G.; Jaeger, H.; et al. (2004). A

McCance-Katz, E.; Moody, D. & Morse, G. (2006). Interactions between buprenorphine and

McCance-Katz, E.; Rainey, P. & Friedland G. (2003). The protease inhibitor

McCance-Katz, E.; Rainey, P. & Smith, P. (2004). Drug interactions between opioide and

Menard, A.; Solas, C.; Mokthari, S; Bregigeon, S.; Drogoul, M.; Tamalet, C.; et al. (2009).

Mildvan, D.; Yarrish, R.; Marshak, A.; Hutman, H.; McDonough, M.; Lamson, M.; et al.

Miller, C.; El-Kholi, R-; Faragon, J. & Lodise, T. (2007). Prevalence and risk factors for

Moling, O.; Rimenti, G.; Pristerà, R.; Vedovelli, C.; Pagani, L.; Spoladore, G.; et al. (2009).

Mullin, P.; Green, G. & Bakshi, R. (2004). Special populations: The management of seizures

Nettles, R.; Kieffer, T.; Parsons, T.; Johnson, J.; Cofrancesco, J.; Gallant, J.; et al. (2006).

Obach, R.; Walsky, R.; Venkatakrishnan, K.; Gaman, E.; Houston, J. & Tremaine, L. (2006).

*J Acquir Immune Defic Syndr*, Vol. 29, pp. 471-477, ISNN 1077-9450.

delavirdine. *Clin Infect Dis*, Vol. 43, pp. S224–S234, ISSN 1537-6591.

patients. *Clin Infect Dis*, Vol. 37, pp. 476–482, ISSN 1537-6591.

*Am J Addict*, Vol.13, pp.163–180, ISSN 1521-0391.

cases. *AIDS*, Vol. 23, pp. 869-871, ISSN 1473-5571.

Vol. 27, No. 10, pp. 1379-1386, ISSN 0277-0008.

ISSN 1365-2125.

6293.

1537-6591.

1621.

8826.

Multiple-dose pharmacokinetics of efavirenz with and without the use of rifampicin in HIV-positive patients. *Curr HIV Res*, Vol. 5, pp. 349-353, ISSN 1570-

prospective, controlled study assessing the effect of lopinavir on amprenavir concentrations boosted by ritonavir. *HIV Med*, Vol. 5, pp. 15-17, ISSN ISSN 1542-

antiretrovirals I: Non-nucleoside reverse transcriptase inhibitors: Efavirenz and

lopinavir/ritonavir may produce opiate withdrawal in methadone-maintained

antiretroviral medications: Interaction between methadone, LAAM and nelfinavir.

Etravirine- raltegravir, a marked interaction in HIV-1 infected patients: about four

(2002). Pharmacokinetic interaction between nevirapine and ethinyl estradiol/norethindrone when administered concurrently to HIV-infected women.

clinically significant drug interactions with antiretroviral therapy. *Pharmacotherapy*,

Fatal lactic acidosis precipitated by nifedipine in a patient treated with disulfiram and antiretrovirals. *Drug Metab Lett*. Vol. 3, No. 3, pp. 176-180, ISSN 1872-3128. Mouly, S.; Rizzo-Padoin, N.; Simoneau, G.; Verstuyft, C.; Aymard, G.; Salvat, C.; et al.

(2006). Effect of widely used combinations of antiretroviral therapy on liver CYP3A4 activity in HIV-infected patients. *Br J Clin Pharmacol*, Vol. 62, pp. 200-209,

in HIV-positive patients. *Curr Neurol Neurosci Rep*, Vol.4, pp.308-14, ISSN: 1534-

Marked intraindividual variability in antiretroviral concentrations may limit the utility of therapeutic drug monitoring. *Clin Infect Dis*, Vol. 42, pp. 1189-1196, ISSN

The utility of in vitro cytochrome P450 inhibition data in the prediction of drugdrug interactions. *Pharmacol Exp Ther*, Vol. 316, pp. 336-348, ISSN 1521-0103. Obach, R.; Walsky, R.; Venkatakrishnan, K.; Houston, J. & Tremaine, L. (2005) In vitro

cytochrome P450 inhibition data and the prediction of drug-drug interactions:

qualitative relationships, quantitative predictions, and the rank-order approach. *Clin Pharmacol Ther*, Vol. 78, pp. 582-592, ISSN 1532-6535.


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 355

Schöller-Gyüre, M.; Woodfall, B.; Debroye, C.; De Marez, T.; Peeters, M.; Vandermeulen, K.;

Sekar, V.; De Pauw, M. & Mariën, K. (2007a). Pharmacokinetic interaction between

Sekar, V.; Lefebvre, E. & Boogaerts, G. (2006a). Pharmacokinetic interaction between the

Sekar, V.; Lefebvre, E.; De Marez, T.; De Pauw, M.; De Paepe, E.; Vangeneugden, T.; et al.

Sekar, V.; Lefebvre, E. & De Paepe, E.; et al. (2007b). Pharmacokinetic interaction between

Sekar, V.; Lefebvre, E. & Mariën, K. (2006b). Pharmacokinetic interaction between the HIV

Sekar, V.; Lavreys, L; Van de Casteele, T; Berckmans, C.; Guzman, S.; Vangeneugden, T.; et

Setzer, B.; Lebrecht, D. & Walker, U. (2008). Pyrimidine nucleoside depletion sensitizes to

Shriner, K. & Goetz, M. (1992). Severe hepatotoxicity in a patient receiving both

Simpson, D. & Tagliati, M. (1995). Nucleoside analogue associated peripheral neuropathy in

Smithburger, P.; Kane-Gill, S.; Benedict, N.; Falcione, B. & Seybert, A. (2010). Grading the

Solas, C.; Poizot-Martin, I.; Drogoul, M.; Ravaux, I.; Dhiver, C.; Lafeuillade, A.; et al. (2004).

acetaminophen and zidovudine. *Am J Med*, Vol. 93, pp. 94-96.

Prevention. Sydney. 22-25 Jul. Abstract WEPEA 106.

Vol. 51, No. 3, pp. 958–961, ISSN 1098-6569, March.

Toronto (ON). abstract no. 959.

4440-4445, ISSN 1098-6569.

*Retrovirol*, Vol. 9, pp. 153-161.

ISSN 1365-2125.

*Am J Pathol*. Vol. 172, No. 3, pp. 681-690.

Vol. 44, No. 11, pp. 1718-1724, ISSN 0012-6667.

Abstract 963

509–514, ISSN 2040-2058.

abstract no. A-367.

1179-1918.

International *AIDS* Society Conference on HIV Pathogenesis, Treatment and

et al. (2006). Pharmacokinetic interaction between TMC125 and rifabutin. 44th Annual Meeting of the Infectious Diseases Society of America. Toronto. 12-15 Oct.

TMC114/r and efavirenz in healthy volunteers. *Antivir Ther*, Vol. 12, No. 4, pp.

protease inhibitors TMC114 and lopinavir/ritonavir. 46th Interscience Conference on Antimicrobial Agents and Chemotherapy, Sep 27-30, San Francisco (CA),

(2008). Effect of repeated doses of darunavir plus low-dose ritonavir on the pharmacokinetics of sildenafil in healthy male subjects: phase I randomized, openlabel, two-way crossover study. *Clin Drug Investig*, Vol. 28, No. 84, pp. 79-85, ISSN

darunavir boosted with ritonavir and omeprazole or ranitidine in human immunodeficiency virus-negative healthy volunteers. *Antimicrob Agents Chemother,*

protease inhibitors TMC114 and saquinavir, in the presence of low-dose ritonavir. 44th Annual Meeting of the Infectious Disease Society of America, Oct 12-15;

al. (2010). Pharmacokinetics of darunavir/ritonavir and rifabutin coadministered in HIV-negative healthy volunteers. *Antimicrob Agents Chemother,* Vol. 54, No. 10, pp.

the mitochondrial hepatotoxicity of the reverse transcriptase inhibitor stavudine.

human immunodeficiency virus infection. *J Acquir Immune Defic Syndr Hum* 

severity of drug-drug interactions in the intensive care unit: a comparison between clinician assessment and proprietary database severity rankings. *Ann Pharmacother*.

Therapeutic drug monitoring of lopinavir/ritonavir given alone or with a nonnucleoside reverse transcriptase inhibitor. *Br J Clin Pharmacol*, Vol. 57, pp. 436-440,


Ray, W.; Murray, K.; Meredith, S.; Narasimhulu, S.; Hall, K. & Stein, C. (2004). Oral

Ribera, E.; Azuaje, C.; López, R.; Díaz, M.; Feijoo, M.; Pou, L.; et al. (2006). Atazanavir and

Ribera, E.; Pou, L.; López, R.; Crespo, M.; Falco, V.; Ocaña, I.; et al. (2001). Pharmacokinetic

Rolla, V.; da Silva Vieira, M.; Pereira, P.; Lourenço, M.; de Jesus, C.; Gonçalves, M.; et al.

Santos, C. & Boullata, J. (2005). An approach to evaluating drug-nutrient interactions.

Saraga, M.; Preisig, M. & Zullino, D. (2006). Reduced valproate plasma levels possible after

Sax, P. (2006). Strategies for management and treatment of dyslipidemia in HIV/AIDS.

Schmidt, G.; Hoehns, J.; Purcell, J.; Friedman, R. & Elhawi, Y. (2007). Severe rhabdomyolysis

and atazanavir. *J Am Board Fam Med*, Vol. 20, pp. 411–416, ISSN 1558-7118. Schöller, M.; Woodffall, B.; De Marez, T.; De Smedt, G.; Peeters, M.; Vandermeulen, K.; et al.

Schöller-Gyüre, M.; Kakuda, T.; De Smedt, G.; Vanaken, H.; Bouche, M.; Peeters, M.; et al.

Schöller-Gyüre, M.; Kakuda, T.; De Smedt, G.; Woodfall, B.; Bollen, S.; Peeters, M.; et al.

tuberculosis. *Clin Drug Investig*, Vol. 26, pp. 469-479, ISSN 1179-1918. Romanelli, F. & Pomeroy, C. (2003). Concurrent use of antiretrovirals and anticonvulsants in

*Pharmacotherapy*. Vol. 25, pp. 1789-1800, ISSN 0277-0008.

*AIDS* Care, Vol . 18, pp. 149-157, ISSN 0954-0121.

*Antimicrob Chemother*, Vol. 59, pp. 690-697, ISSN 1460-2091.

351, pp. 1089-1096, ISSN 1533-4406.

146, ISSN 1130-6343.

ISSN 1399-5618.

Abstract P278.

508-516, ISSN 1476-5381.

pp. 1433-1439, ISSN 1381-6128.

erythromycin and the risk of sudden death from cardiac causes. *N Engl J Med*, Vol.

lopinavir/ritonavir: Pharmacokinetics, safety and efficacy of a promising doubleboosted protease inhibitor regimen. *AIDS*, Vol. 20, pp. 1131-1139, ISSN 1473-5571. Ribera, E.; Azuaje, C.; López, R.; Domingo, P.; Curran, A.; Feijoo, M.; et al. (2007).

Pharmacokinetic interaction between rifampicin and the once-daily combination of saquinavir and low-dose ritonavir in HIV-infected patients with tuberculosis. *J* 

interaction between nevirapine and rifampicin in HIVinfected patients with tuberculosis. *J Acquir Immune Defic Syndr*, Vol. 28, pp. 450-453, ISSN 1077-9450. Robertson, S.; Penzak, S. & Pau, A. (2005). Drug interactions in the management of HIV infection. *Expert Opin Pharmacother*, Vol. 6, pp. 233-253, ISSN 1744-7666. Rodríguez, A.; Caraballo, M.; Palma, D.; Santos, B.; Molina, T.; Desongles, T.; et al. (2009).

Quality of interaction database management systems. *Farm Hosp*, Vol. 33, pp. 134-

(2006). Safety, efficacy and pharmacokinetics of ritonavir 400 mg/saquinavir 400 mg twice daily plus rifampicin combined therapy in HIV patients with

human immunodeficiency virus (HIV) seropositive patients. *Curr Pharm Des*, Vol.9,

introduction of efavirenz in a bipolar patient. Bipolar Disord, Vol. 8, pp. 415-417,

and acute renal failure secondary to concomitant use of simvastatin, amiodarone,

(2006). Pharmacokinetics of TMC125 with atazanavir and atazanavir/ritonavir. 8th Internacional Congress on Drug Therapy in HIV Infection. Glasgow. Nov 12-16,

(2008). A pharmacokinetic study of etravirine (TMC125) co-administered with ranitidine and omeprazole in HIV–negative volunteers. *Br J Pharmacol*, Vol. 66, pp.

(2007). Pharmacokinetic interaction between the non-nucleoside reverse transcriptase inhibitor TMC125 and atorvastatin in HIV-negative volunteers. 4th International *AIDS* Society Conference on HIV Pathogenesis, Treatment and Prevention. Sydney. 22-25 Jul. Abstract WEPEA 106.


Clinical Relevance of Drug Interactions in HIV-Infected Patients Receiving Antiretroviral Therapy 357

Tucker, G.; Houston, J. & Huang, S. (2001). Optimizing drug development: strategies to

Van der Lee, M.; Dawood, L.; Hofstede, H.; Graaff-Teulen, M.; Ewijk-Beneken, Caliskan, N.;

healthy subjects. *Clin Pharmacol Ther*, Vol. 80, pp. 159-168, ISSN 1532-6535. Vogel, M.; Voigt, E.; Michaelis, H.; Sudhop, T.; Wolff, M.; Turler, A.; et al. (2004).

Von, L.; Greenblatt, D.; Duan, S.; Daily, J.; Harmatz, J. & Shader, R. (1998). Inhibition of

Vourvahis, M. & Kashuba, A. (2007). Mechanisms of pharmacokinetic and

Weiner, M.; Benator, D.; Peloquin, C.; Burman, W.; Vernon, A.; Engle, M.; et al. (2005).

Wertheimer, B.; Freedberg, K.; Walensky, R.; Yazdanapah, Y. & Losina, E. (2006).

Winston, A.; Back, D.; Fletcher, C.; Robinson, L.; Unsworth, J.; Tolowinska, I.; et al. (2006).

Winston, A. & Boffito, M. (2005). The management of HIV-1 protease inhibitor

Wire, M.; Shelton, M. & Studenberg S. (2006). Fosamprenavir: Clinical pharmacokinetics and

Wittkop, L.; Breilh, D.; DaSilva, D.; Duffau, P.; Mercié, P.; Raymond, I.; et al. (2009).

Yee, D.; Valiquette, C.; Pelletier, M.; Parisien, I.; Rocher, I. & Menzies, D. (2003). Incidence of

active tuberculosis. *Am J Respir Crit Care Med*, Vol. 167, pp. 1472-1477. Young, B. (2005). Review: Mixing new cocktails: Drug interactions in antiretroviral regimens. *AIDS Patient Care STDS*, Vol. 19, pp. 286-297, ISSN 1557-7449.

*Pharmacol Ther*, Vol. 70, pp. 103-114, ISSN 1532-6535.

*Liver Transpl*, Vol 10, pp. 939-944, ISSN 1524-6473.

*Pharmacotherapy*, Vol. 27, pp.888-909, ISSN 0277-0008.

pp. 1184-1189, ISSN 1520-6017.

Vol. 7, pp. 59-69, ISSN 1528-4336.

6591.

7453.

1460-2091.

ISSN 1473-5571.

168, ISSN 0312-5963.

assess drug metabolism/transporter interaction potential-toward a consensus. *Clin* 

et al. (2006). Lopinavir/ritonavir reduces lamotrigine plasma concentrations in

Management of drug-to-drug interactions between cyclosporine A and the protease-inhibitor lopinavir/ritonavir in liver-transplanted HIV-infected patients.

desipramine hydroxylation (Cytochrome P450-2D6) in vitro by quinidine and by viral protease inhibitors: Relation to drug interactions in vivo. *J Pharm Sci*, Vol. 87,

pharmacodynamic drug interactions associated with ritonavirenhanced tipranavir.

Evaluation of the drug interaction between rifabutin and efavirenz in patients with HIV infection and tuberculosis. *Clin Infect Dis*,Vol.41, pp. 1343-1349. ISSN 1537-

Therapeutic drug monitoring in HIV treatment: A literature review. *HIV Clin Trials*,

Effect of omeprazole on the pharmacokinetics of saquinavir-500 mg formulation with ritonavir in healthy male and female volunteers. *AIDS*, Vol. 20, pp. 1401-1406,

pharmacokinetic interactions. *J Antimicrob Chemother*, Vol . 56, pp. 1- 5, ISSN 0305-

drug interactions of the amprenavir prodrug. *Clin Pharmacokinet*, Vol . 45, pp. 137-

Virological and immunological response in HIV-1-infected patients with multiple treatment failures receiving raltegravir and optimized background therapy, ANRS CO3 Aquitaine Cohort. *J Antimicrob Chemother*, Vol. 63, 1251-1255, ISSN

serious side effects from first-line antituberculosis *Drugs* among patients treated for


http://www.prezista.com/prezista/documents/us\_package\_insert.pdf


Spradling, P.; Drociuk, D.; McLaughlin, S.; Lee, L.; Peloquin, C.; Gallicano, K.; et al. (2002).

Steele, M.; Burk, R. & DesPrez, R. (1991). Toxic hepatitis with isoniazide and rifampin: A

Stocker, H.; Kruse, G. & Kreckel, P. (2004). Nevirapine significantly reduces the levels of

Taburet, A.; Raguin, G.; Le Tiec, C.; Droz, C.; Barrail, A.; Vincent, I.; et al. (2004). Interactions

Tibotec (2008). Prezista® (darunavir) tablet: US prescribing information. Raritan (NJ),

Tomilo, D.; Smith, P.; Ogundele, A.; Difrancesco, R.; Berenson, C.; Eberhardt, E.; et al. (2006).

healthy volunteers. *Pharmacotherapy*, vol. 26, pp. 341-436, ISSN 0277-0008. Tong, L.; Phan, T.; Robinson, K.; Babusis, D.; Strab, R.; Bhoopathy, S.; et al. (2007). Effects of

Touzot, M.; Beller, C.; Touzot, F.; Louet, A. & Piketty, C. (2006). Dramatic interaction

Tran, J.; Gerber, J. & Kerr, B. (2001). Delavirdine: Clinical pharmacokinetics and drug interactions. *Clin Pharmacokinet*, Vol. 40, pp. 207-226, ISSN 0312-5963. Tran, J.; Petersen, C.; Garrett, M.; Hee, B. & Kerr, B. (2002). Pharmacokinetic interaction

amprenavir. *Clin Pharmacol Ther*, Vol. 72, pp. 615-626, ISSN 1532-6535. Tseng, A.; Nguyen, M.; Cardella, C.; Humar, A & Conly, J. (2002). Probable interaction between efavirenz and cyclosporine. *AIDS*, Vol. 16, pp. 505-506, ISSN 1473-5571. Tseng, A. & Foisy M. (1999). Significant interactions with new antiretrovirals and psychotropic *Drugs*. *Ann Pharmacother*, Vol. 33, pp. 461-473, ISSN 0012-6667. Tseng, A. & Salit, I. (2007). CD4+ cell count decline despite HIV suppression: A probable

http://www.prezista.com/prezista/documents/us\_package\_insert.pdf

outbreak. *Clin Infect Dis*, Vol. 35, pp. 1106-1112, ISSN 1537-6591.

meta-analysis. *Chest*, Vol. 99, pp. 465-71, ISSN 1931-3543.

Vol. 48, pp. 2091-2096, ISSN 1098-6569.

pp. 310-323, ISSN 1532-6535.

25.03.2009 Available from:

3498–3504, ISSN 1098-6596.

0012-6667.

Vol. 20, pp. 1210-1212, ISSN 1473-5571.

Drug-drug interactions in inmates treated for human immunodeficiency virus and Mycobacterium tuberculosis infection or disease: An institutional tuberculosis

racemic methadone and (R)-methadone in human immunodeficiency virus-infected patients. *Antimicrob Agents Chemother*, Vol. 48, pp. 4148-4153, ISSN 1098-6569. Sudano, I.; Spieker, L.; Noll, G.; Corti, R.; Weber, R. & Luscher, T. (2006). Cardiovascular disease in HIV infection. *Am Heart J*, Vol . 151, pp. 1147-1155, ISSN 0002-8703. Taburet, A.; Piketty, C.; Chazallon, C.; Vincent, I.; Gerard, L.; Calvez, V:; et al. (2004).

Interactions between atazanavir-ritonavir and tenofovir in heavily pretreated human immunodeficiency virus-infected patients. *Antimicrob Agents Chemother*,

between amprenavir and the lopinavir-ritonavir combination in heavily pretreated patients infected with human immunodeficiency virus. *Clin Pharmacol Ther*, Vol 75,

Inhibition of atazanavir oral absorption by lansoprazole gastric acid suppression in

human immunodeficiency virus protease inhibitors on the intestinal absorption of tenofovir disoproxil fumarate in vitro. *Antimicrob Agents Chemother*, Vol.51, pp.

between levothyroxine and lopinavir/ritonavir in a HIV-infected patient. *AIDS*,

between amprenavir and delavirdine: Evidence of induced clearance by

didanosine-valganciclovir interaction. *Ann Pharmacother*, Vol. 41, pp. 512-517, ISSN


Zeldin, R. & Petruschke, R. (2004). Pharmacological and therapeutic properties of ritonavirboosted protease inhibitor therapy in HIV-infected patients. *J Antimicrob Chemother*, Vol. 53, pp. 4-9, ISSN 1460-2091.

**15** 

*Lisboa, Portugal* 

**Acute Kidney Injury** 

**in Hospitalized HIV-Infected** 

**Patients in the HAART Era:** 

*Department of Nephrology and Renal Transplantation,* 

Actually, more than 30 million people are affected with human immunodeficiency virus (HIV) infection worldwide [1]. Since the introduction of the highly active antiretroviral therapy (HAART) at the end of 1995, overall mortality of patients with HIV infection decreased dramatically as well as mortality caused by HIV infection or by an Acquired Immunodeficiency Syndrome (AIDS)-defining disease. Conversely, mortality due to kidney disease, liver disease, heart disease, and non-AIDS-defining cancers has proportionally

Renal disorders in HIV-infected patients can present as an acute or chronic condition and they are associated with increased morbidity and mortality in this population [5,6,7,8,9]. Acute kidney injury is a common complication in ambulatory HIV-infected patients treated with HAART and has been associated with prior renal impairment, lower CD4 levels, AIDS, hepatitis C virus (HCV) co-infection, and liver disease [10,11]. HIV-infected patients are also at increased risk for AKI development within hospitalization, related to volume depletion, sepsis, and the acute administration of nephrotoxic medications or radiocontrast. Before the advent of HAART, studies addressing AKI on HIV-infected patients typically included only severe cases of AKI which were identified through hospital records or biopsy databases [12,13,14]. The epidemiology of AKI in hospitalized HIV-infected patients in the HAART era has also not been extensively analyzed. In fact, few studies have focused on the clinical characteristics of AKI in hospitalized HIVinfected patients in the HAART era [15,16,17,18]. In this chapter, we provide a critical and contemporary review of AKI in hospitalized HIV-infected patients in the HAART era,

**1. Introduction** 

increased [2,3,4].

focusing on the incidence, risk factors, and outcome.

**An Epidemiological View** 

José António Lopes and Sofia Jorge

*Hospital de Santa Maria,* 

*Centro Hospitalar Lisboa Norte,* 

Zhou, S.; Chan, E.; Pan, S.; Huang, M. & Lee, E. (2004). Pharmacokinetic interactions of *Drugs* with St John's wort. *J Psychopharmacol*, Vol. 18, pp. 262-276, ISSN 1461-7285.

## **Acute Kidney Injury in Hospitalized HIV-Infected Patients in the HAART Era: An Epidemiological View**

José António Lopes and Sofia Jorge

*Department of Nephrology and Renal Transplantation, Hospital de Santa Maria, Centro Hospitalar Lisboa Norte, Lisboa, Portugal* 

#### **1. Introduction**

358 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Zeldin, R. & Petruschke, R. (2004). Pharmacological and therapeutic properties of ritonavir-

Zhou, S.; Chan, E.; Pan, S.; Huang, M. & Lee, E. (2004). Pharmacokinetic interactions of *Drugs* with St John's wort. *J Psychopharmacol*, Vol. 18, pp. 262-276, ISSN 1461-7285.

Vol. 53, pp. 4-9, ISSN 1460-2091.

boosted protease inhibitor therapy in HIV-infected patients. *J Antimicrob Chemother*,

Actually, more than 30 million people are affected with human immunodeficiency virus (HIV) infection worldwide [1]. Since the introduction of the highly active antiretroviral therapy (HAART) at the end of 1995, overall mortality of patients with HIV infection decreased dramatically as well as mortality caused by HIV infection or by an Acquired Immunodeficiency Syndrome (AIDS)-defining disease. Conversely, mortality due to kidney disease, liver disease, heart disease, and non-AIDS-defining cancers has proportionally increased [2,3,4].

Renal disorders in HIV-infected patients can present as an acute or chronic condition and they are associated with increased morbidity and mortality in this population [5,6,7,8,9]. Acute kidney injury is a common complication in ambulatory HIV-infected patients treated with HAART and has been associated with prior renal impairment, lower CD4 levels, AIDS, hepatitis C virus (HCV) co-infection, and liver disease [10,11]. HIV-infected patients are also at increased risk for AKI development within hospitalization, related to volume depletion, sepsis, and the acute administration of nephrotoxic medications or radiocontrast. Before the advent of HAART, studies addressing AKI on HIV-infected patients typically included only severe cases of AKI which were identified through hospital records or biopsy databases [12,13,14]. The epidemiology of AKI in hospitalized HIV-infected patients in the HAART era has also not been extensively analyzed. In fact, few studies have focused on the clinical characteristics of AKI in hospitalized HIVinfected patients in the HAART era [15,16,17,18]. In this chapter, we provide a critical and contemporary review of AKI in hospitalized HIV-infected patients in the HAART era, focusing on the incidence, risk factors, and outcome.

Acute Kidney Injury

Wyatt et

Lopes et

Lopes et

al 15 25.114 Retrospective,

al 16 <sup>489</sup>Retrospective,

al 17 <sup>97</sup>Retrospective,

single-center

N Design Setting Year of

multicenter Hospitalized 2003

single-center Hospitalized 2005-2007

Intensive

SAPS II- Simplified Acute Pathophysiology Score version II.

care unit 2002-2006

Table 1. Studies reporting the incidence, risk factors and mortality of acute kidney injury in the highly active antiretroviral therapy era. AKI- acute kidney injury. ICD-9- International Classification of Diseases, 9th Revision. OR- odds ratio. CI- confidence interval. HCVhepatitis C virus. SCr- serum creatinine. AIDS- Acquired Immunodeficiency Syndrome.

in Hospitalized HIV-Infected Patients in the HAART Era: An Epidemiological View 361

Definition of AKI

Code 584 of the ICD-9

1.5X SCr or SCr ≥0.5mg (if baseline SCr 4mg/dl)

1.5X SCr or SCr ≥0.5mg (if baseline SCr 4mg/dl)

Incidence of AKI

6%

18%

47%

Risk factors of AKI

Age (per year above mean) (adjusted OR 1.03, 95% CI, 1.02-1.04), Male (adjusted OR 1.16, 95% CI 1.04-1.30), diabetes mellitus (adjusted OR 1.27, 95% CI 1.08-1.49), chronic kidney disease (adjusted OR 5.48, 95% CI 4.58-6.56), liver disease (adjusted OR 1.59, 95% CI 1.40-1.79)

Pre-existing hypertension (adjusted OR 2.4, 95% CI 1.04-5.6, P=0.04), AIDS (adjusted OR 2.7, 95% CI 1.2-6, P=0.02), sepsis (adjusted OR 23, 95% CI 11-45.3, P<0.001), and nephrotoxic drugs administration (adjusted OR 2.8, 95% CI 1.4-5.8, P=0.004)

Age >60 years (adjusted OR 5.32, 95% CI 1.23-23, P=0.025), HCV co-infection (adjusted OR 3.42, 95% CI 1.08-10.85, P=0.037), SAPS II >50 (adjusted OR 2.35, 95% CI 1.2- 5.9, P=0.008)

Mortality (AKI versus non-AKI)

In-hospital mortality 27% versus 4.5% (adjusted OR 5.83; 95% CI, 5.11-6.65, P0.0001)

In-hospital mortality 27.3% versus 8% (adjusted OR 2.7, 95% CI 1.3-5.6, P=0.008)

60-day mortality (65% versus 24%, P0.0001)

hospitalization

#### **2. Incidence of AKI in hospitalized HIV-infected patients**

Three recent studies 15,16,17 have addressed specifically the incidence of AKI in hospitalized HIV-infected patients (Table 1). Wyatt et al 15 compared the incidence of AKI in HIV-infected patients before and after the introduction of HAART. For this purpose, all adult patients who were discharged from acute care hospitals in New York State during 1995 (pre-HAART era) and during 2003 (post-HAART era) were evaluated. The presence of AKI was determined by a diagnosis code 584 of the International Classification of Diseases, 9th Revision (ICD-9), which identified AKI based on the clinical judgement of the treating physician. There were 52.580 patients with documented HIV infection discharged from hospital in 1995, and 25.114 in 2003. Acute kidney injury was reported significantly more often during hospitalizations for HIV-infected patients than for uninfected patients in both 2003 (6% versus 2.7%) and 1995 (2.9% versus 1.0%). After adjusting for other covariates, HIV infection remained associated with an increased risk of AKI both in 2003 [adjusted odds ratio (OR) 2.82, 95% confidence interval (CI) 2.66-2.99) and in 1995 (adjusted OR 4.62, 95% CI 4.3-4.95). Lopes et al 16 conducted a cohort study including 489 HIV-infected patients hospitalized in a tertiary and teaching Portuguese Hospital between 2005 and 2007 to characterize AKI in this population. Acute kidney injury was defined and categorized according to "Risk Injury Failure Loss of kidney function End-stage kidney disease" (RIFLE) classification [19], and it was considered if there was an increase of baseline serum creatinine × 1.5 or in patients with baseline serum creatinine > 4 mg/dl if there was an acute rise in serum creatinine of at least 0.5 mg/dl. They found that 18% of patients had AKI within the hospitalization which was much higher than the incidence previously reported (6%) in hospitalized HIV-infected patients in HAART era 15. It should be remembered that in the study of Wyatt et al 15 the diagnosis of AKI was determined by a diagnosis code 584 of the ICD-9 based on the clinical judgement and documentation of the treating physician, and laboratory values were not reported. Administrative databases may be a powerful tool for the study of AKI, although the low sensitivity of the AKI codes still remains an important caveat 20. Therefore, in the study of Wyatt et al 15 the utilization of diagnostic code to identify AKI could not have captured an important number of cases.

In a previous report, Lopes et al 17 have also studied AKI in a small cohort of critically ill HIV-infected patients hospitalized in a tertiary and teaching Portuguese Hospital between 2002 and 2006. In this retrospective study, 47% of patients had AKI (defined by the RIFLE criteria) during the intensive care unit (ICU) stay.

#### **3. Risk factors of AKI in hospitalized HIV-infected patients**

In the HAART era, clinical conditions commonly associated with increased risk of AKI in the general population such as older age, Male, Black race, diabetes, prior hypertension, liver disease and pre-existing chronic kidney disease have also been reported as independent risk factors of AKI in hospitalized HIV-infected patients 15,16,17 (Table 1). Accordingly, renal function should be closely monitored during the hospitalization, and an adequate control of glycemia and blood pressure as well as the appropriate management of patients with acute or chronic liver insufficiency and/or chronic kidney disease could prevent the occurrence of AKI.

Three recent studies 15,16,17 have addressed specifically the incidence of AKI in hospitalized HIV-infected patients (Table 1). Wyatt et al 15 compared the incidence of AKI in HIV-infected patients before and after the introduction of HAART. For this purpose, all adult patients who were discharged from acute care hospitals in New York State during 1995 (pre-HAART era) and during 2003 (post-HAART era) were evaluated. The presence of AKI was determined by a diagnosis code 584 of the International Classification of Diseases, 9th Revision (ICD-9), which identified AKI based on the clinical judgement of the treating physician. There were 52.580 patients with documented HIV infection discharged from hospital in 1995, and 25.114 in 2003. Acute kidney injury was reported significantly more often during hospitalizations for HIV-infected patients than for uninfected patients in both 2003 (6% versus 2.7%) and 1995 (2.9% versus 1.0%). After adjusting for other covariates, HIV infection remained associated with an increased risk of AKI both in 2003 [adjusted odds ratio (OR) 2.82, 95% confidence interval (CI) 2.66-2.99) and in 1995 (adjusted OR 4.62, 95% CI 4.3-4.95). Lopes et al 16 conducted a cohort study including 489 HIV-infected patients hospitalized in a tertiary and teaching Portuguese Hospital between 2005 and 2007 to characterize AKI in this population. Acute kidney injury was defined and categorized according to "Risk Injury Failure Loss of kidney function End-stage kidney disease" (RIFLE) classification [19], and it was considered if there was an increase of baseline serum creatinine × 1.5 or in patients with baseline serum creatinine > 4 mg/dl if there was an acute rise in serum creatinine of at least 0.5 mg/dl. They found that 18% of patients had AKI within the hospitalization which was much higher than the incidence previously reported (6%) in hospitalized HIV-infected patients in HAART era 15. It should be remembered that in the study of Wyatt et al 15 the diagnosis of AKI was determined by a diagnosis code 584 of the ICD-9 based on the clinical judgement and documentation of the treating physician, and laboratory values were not reported. Administrative databases may be a powerful tool for the study of AKI, although the low sensitivity of the AKI codes still remains an important caveat 20. Therefore, in the study of Wyatt et al 15 the utilization of diagnostic

**2. Incidence of AKI in hospitalized HIV-infected patients** 

code to identify AKI could not have captured an important number of cases.

**3. Risk factors of AKI in hospitalized HIV-infected patients** 

criteria) during the intensive care unit (ICU) stay.

prevent the occurrence of AKI.

In a previous report, Lopes et al 17 have also studied AKI in a small cohort of critically ill HIV-infected patients hospitalized in a tertiary and teaching Portuguese Hospital between 2002 and 2006. In this retrospective study, 47% of patients had AKI (defined by the RIFLE

In the HAART era, clinical conditions commonly associated with increased risk of AKI in the general population such as older age, Male, Black race, diabetes, prior hypertension, liver disease and pre-existing chronic kidney disease have also been reported as independent risk factors of AKI in hospitalized HIV-infected patients 15,16,17 (Table 1). Accordingly, renal function should be closely monitored during the hospitalization, and an adequate control of glycemia and blood pressure as well as the appropriate management of patients with acute or chronic liver insufficiency and/or chronic kidney disease could


Table 1. Studies reporting the incidence, risk factors and mortality of acute kidney injury in the highly active antiretroviral therapy era. AKI- acute kidney injury. ICD-9- International Classification of Diseases, 9th Revision. OR- odds ratio. CI- confidence interval. HCVhepatitis C virus. SCr- serum creatinine. AIDS- Acquired Immunodeficiency Syndrome. SAPS II- Simplified Acute Pathophysiology Score version II.

Acute Kidney Injury

outcome among AKI patients [36].

AKI severity and mortality.

**5. Conclusions** 

in this specific population.

**6. References** 

in Hospitalized HIV-Infected Patients in the HAART Era: An Epidemiological View 363

acute episode, could account for decreased long-term survival of patients who developed AKI [33,34,35]. Moreover, CKD disease with subsequent hypertension, proteinuria and increased cardiovascular disease has been appointed as a possible cause of poor long-term

The development of AKI during the hospitalization also portends an ominous outcome among HIV-infected patients (Table 1). In the study of Wyatt et al 15, hospitalizations of HIV-infected patients that were complicated by AKI were also complicated by much higher in-hospital mortality that seen in admissions of HIV-infected patients without AKI and, furthermore, AKI independently increased in-hospital mortality of those patients. In the study of Lopes et al 16, the development of AKI was associated with lengthened time of hospitalization and increased in-hospital mortality. In fact, patients who developed AKI within the hospitalization had higher in-hospital mortality than those patients who did not develop AKI. After adjusting for other covariates, AKI still remained associated with increased in-hospital mortality. Furthermore, there was a relationship between more severe AKI and increased in-hospital mortality. In critically ill HIV-infected patients 17, AKI has also been associated with increased mortality, and there was a graded relationship between

The detrimental impact of AKI on patient outcome seems to persist after hospital discharge even in those patients who exhibit renal function recovery. Recently, Choi et al 17 conducted an observational cohort study in a national sample of 17.325 HIV-infected persons receiving care in the Veterans Health Administration who survived at least 90 days after discharge from their first hospitalization to examine the association between AKI experienced during their first hospitalization with the development of heart failure, atherosclerotic cardiovascular events, end-stage renal disease (ESRD), and death over a period of 2 decades. They found a graded and independent association between severity of

Acute kidney injury is a common complication in hospitalized HIV-infected patients in the HAART era. Older patients, Male, Black race patients, diabetic and /or hypertensive patients and patients with pre-existing chronic kidney disease, HCV co-infection and/or liver disease are at increased risk for AKI within the hospitalization and, therefore, renal function should be closely monitored in those patients. Sepsis is the most common etiology of AKI in this setting and should be promptly diagnosed and treated. The occurrence of AKI is associated with both increased short- and long-term mortality. Therefore, prevention of AKI should be an important task to accomplish in order to improve morbidity and mortality

AKI with heart failure, cardiovascular disease, ESRD, and death.

[1] UNAIDS. UNAIDS/WHO AIDS Epidemic Update: December 2009, 2009

Hepatitis C virus co-infection has also been associated with increased risk for AKI (Table 1). Hepatitis C virus co-infection is an increasingly important cause of morbidity and mortality in patients with HIV [2, and affects approximately 30% of HIV-infected individuals [21. Studies have demonstrated that co-infection with HIV and HCV translates into higher morbidity and mortality related to end-stage liver disease [22. A recent meta-analysis of 27 studies including data of more than 18.000 HIV-infected patients has also demonstrated that HCV co-infection was associated with an increased risk of AKI by 64% [23. Therefore, the association between HCV co-infection and risk for acute and chronic kidney disease supports existing guidelines for the diagnosis and management of kidney disease in patients with HIV [5.

Only two studies have specifically analyzed the etiology of AKI in hospitalized HIVinfected patients 16,17. In the study of Lopes et al 16, the etiology of AKI was multifactorial in 48.9% of patients. The most common etiologies of AKI in this cohort were sepsis (59%), nephrotoxic drugs administration (i.e. aminoglycosides, amphotericine B, vancomycin, acyclovir, gancyclovir and foscarnet) (37.5%), volume depletion (21.6%), and use of radiocontrast (20.5%). Other less common causes of AKI were tumour lysis syndrome, hemorrhage, acute urinary tract obstruction and thrombotic microangiopathy. In the ICU setting, Lopes et al 17 have also identified sepsis as the most common cause of AKI (86%) in HIV-infected patients. Therefore, prompt recognition and aggressive treatment of sepsis, adequate hydration, avoidance and serum monitoring of nephrotoxic drugs, and prophylaxis of contrast induced nephropathy could be important in diminishing the occurrence of AKI in this population [24,25].

The influence of HIV-related variables namely type of HIV, HAART, CD4 lymphocyte count, viral load and AIDS diagnosis in the development of AKI in hospitalized HIVinfected patients still remains to be established. In fact, only one study has attempted to study the impact of those variables on renal function in hospitalized HIV-infected patients 16. In this study, only AIDS-defining conditions were independently associated with AKI and none association was found with type of HIV, HAART, CD4 lymphocyte count and viral load. However, the limited number of studied patients did not allow the authors to conclude definitively about the influence of those variables in the development of AKI. Therefore, prospective and randomized studies with a large number of patients are still warranted to better determine the precise impact of those HIV-related variables on renal function among HIV-infected patients who are hospitalized.

## **4. Impact on outcome of AKI in hospitalized HIV-infected patients**

Acute kidney injury is a risk factor for short- and long-term mortality, and there is a graded relationship between severity of AKI and increased mortality [26,27,28,29,30,31,32]. The mechanism by which AKI contributes to increased mortality is not completely understood. Volume overload, coagulation abnormalities, an increased incidence of sepsis with multi-organ failure, and cytokine or immunemediated major organ dysfunction are other possible explanations for poor survival among AKI patients. The permanent injury to other vital organs caused by AKI, although the potential reversible nature of clinical AKI, in which serum creatinine can return to baseline after the acute episode, could account for decreased long-term survival of patients who developed AKI [33,34,35]. Moreover, CKD disease with subsequent hypertension, proteinuria and increased cardiovascular disease has been appointed as a possible cause of poor long-term outcome among AKI patients [36].

The development of AKI during the hospitalization also portends an ominous outcome among HIV-infected patients (Table 1). In the study of Wyatt et al 15, hospitalizations of HIV-infected patients that were complicated by AKI were also complicated by much higher in-hospital mortality that seen in admissions of HIV-infected patients without AKI and, furthermore, AKI independently increased in-hospital mortality of those patients. In the study of Lopes et al 16, the development of AKI was associated with lengthened time of hospitalization and increased in-hospital mortality. In fact, patients who developed AKI within the hospitalization had higher in-hospital mortality than those patients who did not develop AKI. After adjusting for other covariates, AKI still remained associated with increased in-hospital mortality. Furthermore, there was a relationship between more severe AKI and increased in-hospital mortality. In critically ill HIV-infected patients 17, AKI has also been associated with increased mortality, and there was a graded relationship between AKI severity and mortality.

The detrimental impact of AKI on patient outcome seems to persist after hospital discharge even in those patients who exhibit renal function recovery. Recently, Choi et al 17 conducted an observational cohort study in a national sample of 17.325 HIV-infected persons receiving care in the Veterans Health Administration who survived at least 90 days after discharge from their first hospitalization to examine the association between AKI experienced during their first hospitalization with the development of heart failure, atherosclerotic cardiovascular events, end-stage renal disease (ESRD), and death over a period of 2 decades. They found a graded and independent association between severity of AKI with heart failure, cardiovascular disease, ESRD, and death.

## **5. Conclusions**

362 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Hepatitis C virus co-infection has also been associated with increased risk for AKI (Table 1). Hepatitis C virus co-infection is an increasingly important cause of morbidity and mortality in patients with HIV [2, and affects approximately 30% of HIV-infected individuals [21. Studies have demonstrated that co-infection with HIV and HCV translates into higher morbidity and mortality related to end-stage liver disease [22. A recent meta-analysis of 27 studies including data of more than 18.000 HIV-infected patients has also demonstrated that HCV co-infection was associated with an increased risk of AKI by 64% [23. Therefore, the association between HCV co-infection and risk for acute and chronic kidney disease supports existing guidelines for the diagnosis and management of kidney disease in patients

Only two studies have specifically analyzed the etiology of AKI in hospitalized HIVinfected patients 16,17. In the study of Lopes et al 16, the etiology of AKI was multifactorial in 48.9% of patients. The most common etiologies of AKI in this cohort were sepsis (59%), nephrotoxic drugs administration (i.e. aminoglycosides, amphotericine B, vancomycin, acyclovir, gancyclovir and foscarnet) (37.5%), volume depletion (21.6%), and use of radiocontrast (20.5%). Other less common causes of AKI were tumour lysis syndrome, hemorrhage, acute urinary tract obstruction and thrombotic microangiopathy. In the ICU setting, Lopes et al 17 have also identified sepsis as the most common cause of AKI (86%) in HIV-infected patients. Therefore, prompt recognition and aggressive treatment of sepsis, adequate hydration, avoidance and serum monitoring of nephrotoxic drugs, and prophylaxis of contrast induced nephropathy could be important in diminishing the

The influence of HIV-related variables namely type of HIV, HAART, CD4 lymphocyte count, viral load and AIDS diagnosis in the development of AKI in hospitalized HIVinfected patients still remains to be established. In fact, only one study has attempted to study the impact of those variables on renal function in hospitalized HIV-infected patients 16. In this study, only AIDS-defining conditions were independently associated with AKI and none association was found with type of HIV, HAART, CD4 lymphocyte count and viral load. However, the limited number of studied patients did not allow the authors to conclude definitively about the influence of those variables in the development of AKI. Therefore, prospective and randomized studies with a large number of patients are still warranted to better determine the precise impact of those HIV-related variables on renal

with HIV [5.

occurrence of AKI in this population [24,25].

function among HIV-infected patients who are hospitalized.

**4. Impact on outcome of AKI in hospitalized HIV-infected patients** 

Acute kidney injury is a risk factor for short- and long-term mortality, and there is a graded relationship between severity of AKI and increased mortality [26,27,28,29,30,31,32]. The mechanism by which AKI contributes to increased mortality is not completely understood. Volume overload, coagulation abnormalities, an increased incidence of sepsis with multi-organ failure, and cytokine or immunemediated major organ dysfunction are other possible explanations for poor survival among AKI patients. The permanent injury to other vital organs caused by AKI, although the potential reversible nature of clinical AKI, in which serum creatinine can return to baseline after the Acute kidney injury is a common complication in hospitalized HIV-infected patients in the HAART era. Older patients, Male, Black race patients, diabetic and /or hypertensive patients and patients with pre-existing chronic kidney disease, HCV co-infection and/or liver disease are at increased risk for AKI within the hospitalization and, therefore, renal function should be closely monitored in those patients. Sepsis is the most common etiology of AKI in this setting and should be promptly diagnosed and treated. The occurrence of AKI is associated with both increased short- and long-term mortality. Therefore, prevention of AKI should be an important task to accomplish in order to improve morbidity and mortality in this specific population.

## **6. References**

[1] UNAIDS. UNAIDS/WHO AIDS Epidemic Update: December 2009, 2009

Acute Kidney Injury

in Hospitalized HIV-Infected Patients in the HAART Era: An Epidemiological View 365

[19] Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure – definition, outcome

[20] Waikar SS, Wald R, Chertow GM, et al. Validity of International Classification of

[21] Alter MJ. Epidemiology of viral hepatitis and HIV co-infection. *J Hepatol*. 2006;44:S6-S9 [22] Graham CS, Baden LR, Yu E, et al. Influence of human immunodeficiency virus

[23] Wyatt CM, Malvestutto C, Coca SG, et al. The impact of hepatitis C virus coinfection on

[24] Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign international

[25] Venkataraman R. Can we prevent acute kidney injury? *Crit Care Med*. 2008;36 (4

[26] Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury are

[27] Bagshaw SM, Laupland KB, Doig CJ, et al. Prognosis for long-term survival and renal

[28] Lopes JA, Jorge S, Resina C, et al. Acute kidney injury in patients with sepsis: a

[29] Lopes JA, Fernandes P, Jorge S, et al. Long-term risk of mortality after acute kidney injury in patients with sepsis: a contemporary analysis. *BMC Nephrol*. 2010;11:9 [30] Parikh CR, Coca SG, Wang Y, et al. Long-term prognosis of acute kidney injury after

[31] Covic A, Schiller A, Mardare NG, et al. The impact of acute kidney injury on short-term

[32] Tsagalis G, Akrivos T, Alevizaki M, et al. Long-term prognosis of acute kidney injury

[33] Kelly KJ. Distant effects of experimental renal ischemia/reperfusion injury. *J Am Soc* 

[34] Hassoun HT, Lie ML, Grigoryev DN, et al. Kidney ischemia-reperfusion injury

[35] Li X, Hassoun HT, Santora R, et al. Organ crosstalk: the role of the kidney. *Curr Opin* 

survival in an Eastern European population with stroke. *Nephrol Dial Transplant*.

induces caspase-dependent pulmonary fibrosis. *Am J Physiol Renal Physiol*.

contemporary analysis. *Int J Infect Dis*. 2009;13:176-181

acute myocardial infarction. *Arch Intern Med*. 2008;168:987-995

after first acute stroke. *Clin J Am Soc Nephrol*. 2009;4:616-622

(ADQI) Group. *Crit Care*. 2004;8:R204

*Soc Nephrol*. 2006;17:1688-1694

2001;33:562-569

2008;22:1799-807

2008;36:296-327

Suppl):S166-S171

*Care*. 2006;10:R73

2008;23:2228-2234

2009;1:F125-137

*Nephrol*. 2003;14:1549-1558

*Crit Care*. 2009;15:481-487

study. *Crit Care*. 2005;9:R700-709

measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative

Diseases, Ninth Revision, Clinical Modification Codes for Acute Renal Failure. *J Am* 

infection on the course of hepatitis C virus infection: a meta-analysis. *Clin Infect Dis*.

HIV-related kidney disease: a systematic review and meta-analysis. *AIDS*.

guidelines for management of severe sepsis and septic shock: 2008. *Crit Care Med*.

associated with hospital mortality in critically ill patients: a cohort analysis. *Crit* 

recovery in critically ill patients with severe acute renal failure: a population-based


[2] Selik RM, Byers RH Jr, Dworkin MS. Trends in diseases reported on U.S. death

[3] Krentz HB, Kliewer G, Gill MJ. Changing mortality rates and causes of death for HIV-

[4] Palella FJ Jr, Baker RK, Moorman AC, et al. Mortality in the highly active antiretroviral

[5] Gupta SK, Eustace JA, Winston JA, et al. Guidelines for the management of chronic

[6] Szczech LA, Hoover DR, Feldman JG, et al. Association between renal disease and

[7] Fine DM, Atta MG. Kidney disease in the HIV-infected patient. *AIDS Patient Care STDS*.

[8] Wyatt CM, Winston JA, Malvestutto CD, et al. Chronic kidney disease in HIV infection:

[9] Fine DM, Perazella MA, Lucas GM, et al. Renal disease in patients with HIV infection: epidemiology, pathogenesis and management. *Drugs*. 2008;68:963-980 [10] Franceschini N, Napravnik S, Eron JJ Jr, et al. Incidence and etiology of acute renal failure among ambulatory HIV-infected patients. *Kidney Int*. 2005;67:1526-1531 [11] Ibrahim F, Claire Naftalin C, Cheserem E, et al. Immunodeficiency and renal

[12] Valeri A, Neusy AJ. Acute and chronic renal disease in hospitalized AIDS patients. *Clin* 

[13] Rao TK, Friedman EA. Outcome of severe acute renal failure in patients with acquired

[14] Peraldi MN, Maslo C, Akposso K, et al. Acute renal failure in the course of HIV

[15] Wyatt CM, Arons RR, Klotman PE, et al. Acute renal failure in hospitalized patients

[16] Lopes JA, Fernandes J, Jorge S, et al. Acute renal failure in critically ill HIV-infected

[17] Choi AI, Li Y, Parikh C, et al. Long-term clinical consequences of acute kidney injury in

[18] Lopes JA, Melo MJ, Viegas A, et al. Acute kidney injury in hospitalized HIV-infected

immunodeficiency syndrome. *Am J Kidney Dis*. 1995;25:390-398

renal biopsies. *Nephrol Dial Transplant*. 1999;14:1578-158

2002;29:378-387

*Med*. 2005;6:99-106

2005;40:1559-1585

2007;21:813-824

2010,24:2239-2244

565

print]

*Nephrol*. 1991;35:110-118

patients. *Crit Care*. 2007;11:404

the HIV-infected. *Kidney Int*. 2010;78:478-485

*Acquir Immune Defic Syndr*. 2006;43:27-34

therapy. *Clin Infect Dis*. 2004;39:1199-1206

an urban epidemic. *AIDS*. 2007;21:2101-2103

certificates that mentioned HIV infection, 1987-1999. *J Acquir Immune Defic Syndr*.

infected individuals living in Southern Alberta, Canada from 1984 to 2003. *HIV* 

therapy era: changing causes of death and disease in the HIV outpatient study. *J* 

kidney disease in HIV-infected patients: recommendations of the HIV Medicine Association of the Infectious Diseases Society of America. *Clin Infect Dis*.

outcomes among HIV-infected women receiving or not receiving antiretroviral

impairment are risk factors for HIV-associated acute renal failure. *AIDS*.

infection: a single institution retrospective study of ninety-two patients and sixty

with HIV: risk factors and impact on in-hospital mortality. *AIDS*. 2006;20:561-

patients: a cohort analysis. *Nephrol Dial Transplant.* 2011 May 4. [Epub ahead of


[36] Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: a systematic review. *J Am Soc Nephrol*. 2006;17:2034-2047 **16** 

**Exploring the Nanotechnology-Based** 

Flávia Chiva Carvalho1, Rubiana Mara Mainardes2

*2State University of West-Center, UNICENTRO, Guarapuava,* 

and Maria Palmira Daflon Gremião1

*Brazil* 

 **Drug Delivery Systems for AIDS Treatment** 

*1School of Pharmaceutical Sciences, State University of São Paulo, UNESP, Araraquara,* 

The Acquired Immunodeficiency Syndrome (AIDS) and Human Immunodeficiency Virus (HIV) infection are a worldwide public health challenge. The emergence of antiretroviral therapy agents has significantly increased the life expectancy and the patient´s quality of life. In the 1990's, there was a great improvement in the knowledge of the disease, enlargement of therapeutic resources, a rise in life expectancy and the epidemiologic profile. Since the mid-1990's, the advancement of pharmacology studies and the arrival of protease inhibitor antiretroviral have given rise to a new era of anti-HIV agents, known as Highly Active Antiretroviral Therapy (HAART) (Geocze et al., 2010, Richman et al., 2009). The HAART's adhesion improved the clinical results, the control of the advancement of the disease and decreased the mortality rate, which resulted in an improvement of the patient's life quality. Despite the successful administration of HAART, latently infected cells can escape the viral immune response and persist for long periods of time (Alexaki et al., 2008). In addition, the HAART presents several collateral effects, such as fatigue, nausea, sickness, diarrhea and lipodystrophy. These symptoms contribute to a lack of treatment adhesion in the patient, resulting in a rise in the blood viral load and a decline in CD4+ T cells count, as well as an increased tolerance of anti-HIV drugs, treatment failure, increased opportunistic infections and in wasted investments (Geocze et al., 2010). Moreover, many antiretroviral drugs undergo extensive pre-systemic metabolism and instability in the gastrointestinal environment, resulting in inadequate and erratic oral absorption as well as low bioavailability. The half-life for most anti-HIV drugs is short, and thus, it requires frequent dosage administrations, leading to a decrease in patient compliance. Also, some antiretroviral classes present poor solubility, low absorption and limited bioavailability. Another limitation of the current HAART is the inefficiency of the regimens to eradicate HIV from various anatomical reservoirs (e.g., central nervous system (CNS) and gastrointestinal tract) and intracellular sites (e.g., macrophages, hepatocytes, dendritic cells and Langerhans cells) (Ojewole et al., 2008, Saksena & Haddad, 2003). Large concentrations are essential for eliminating HIV from these reservoirs to achieve the desired therapeutic effect, but these large doses contribute to severe side effects associated with anti-HIV therapy (Ojewole et al., 2008). Because drug development in the HIV field has slowed

**1. Introduction** 

## **Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment**

Flávia Chiva Carvalho1, Rubiana Mara Mainardes2 and Maria Palmira Daflon Gremião1 *1School of Pharmaceutical Sciences, State University of São Paulo, UNESP, Araraquara, 2State University of West-Center, UNICENTRO, Guarapuava, Brazil* 

## **1. Introduction**

366 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

[36] Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: a

The Acquired Immunodeficiency Syndrome (AIDS) and Human Immunodeficiency Virus (HIV) infection are a worldwide public health challenge. The emergence of antiretroviral therapy agents has significantly increased the life expectancy and the patient´s quality of life. In the 1990's, there was a great improvement in the knowledge of the disease, enlargement of therapeutic resources, a rise in life expectancy and the epidemiologic profile. Since the mid-1990's, the advancement of pharmacology studies and the arrival of protease inhibitor antiretroviral have given rise to a new era of anti-HIV agents, known as Highly Active Antiretroviral Therapy (HAART) (Geocze et al., 2010, Richman et al., 2009). The HAART's adhesion improved the clinical results, the control of the advancement of the disease and decreased the mortality rate, which resulted in an improvement of the patient's life quality. Despite the successful administration of HAART, latently infected cells can escape the viral immune response and persist for long periods of time (Alexaki et al., 2008). In addition, the HAART presents several collateral effects, such as fatigue, nausea, sickness, diarrhea and lipodystrophy. These symptoms contribute to a lack of treatment adhesion in the patient, resulting in a rise in the blood viral load and a decline in CD4+ T cells count, as well as an increased tolerance of anti-HIV drugs, treatment failure, increased opportunistic infections and in wasted investments (Geocze et al., 2010). Moreover, many antiretroviral drugs undergo extensive pre-systemic metabolism and instability in the gastrointestinal environment, resulting in inadequate and erratic oral absorption as well as low bioavailability. The half-life for most anti-HIV drugs is short, and thus, it requires frequent dosage administrations, leading to a decrease in patient compliance. Also, some antiretroviral classes present poor solubility, low absorption and limited bioavailability. Another limitation of the current HAART is the inefficiency of the regimens to eradicate HIV from various anatomical reservoirs (e.g., central nervous system (CNS) and gastrointestinal tract) and intracellular sites (e.g., macrophages, hepatocytes, dendritic cells and Langerhans cells) (Ojewole et al., 2008, Saksena & Haddad, 2003). Large concentrations are essential for eliminating HIV from these reservoirs to achieve the desired therapeutic effect, but these large doses contribute to severe side effects associated with anti-HIV therapy (Ojewole et al., 2008). Because drug development in the HIV field has slowed

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 369

Surfactants are extensively used as excipients in drug delivery and the understanding of the physicochemical properties and behavior of these amphiphilic compounds has undergone significant development. One reason for this development is that surface chemistry is a relatively young scientific discipline, and many studies have recognized its importance for the design and controlled use of drug delivery formulations (Malmsten, 2002). The surfactant systems comprise several types of arrangements, and they self-organize with different physicochemical properties that can be used to design new drug delivery systems that are able to solubilize both water-soluble and oil-soluble compounds. Depending on the composition and molecules of the component, surfactant systems can infinitely dilute or form highly stiff matrices, both in physiological conditions. Due the capacity of surfactant systems, which can form viscous and gel-like structures with different rheological characteristics, these systems are extremely versatile and can be delineated to be applied for different routes of administration, e.g., transdermal or mucosal administration, such as vaginal, nasal, rectal and sublingual. However, almost all studies aim oral administration, where anti-HIV drugs are associated in microemulsions, polymeric micelles, self-assembled drug delivery systems, liquid crystals and liposomes. Liposomes, contrary to the other surfactant organizations, have been widely explored for orally administered anti-HIV drugs. Although the oral route is still the primary mode of delivery for antiretroviral surfactant systems, they were found to be a suitable vehicle for anti-HIV intravenous and mucosal administration (described below).

Microemulsions are systems consisting of water, oil, and surfactant(s), which constitute a single optically isotropic and thermodynamically stable liquid dispersion. Such systems are useful for drug delivery due to their excellent stability, ease of preparation, optical clarity, as well as their capacity to dissolve hydrophilic and lipophilic drugs, frequently in high amounts. Microemulsions differ from emulsions and nanoemulsions because of their thermodynamic stability; these systems form spontaneously, exhibit reduced droplets sizes (typically 10–100 nm) (Malmsten, 2002), higher surface areas and free energy without the inherent creaming, flocculation, coalescence and sedimentation associated with emulsions (Gupta & Jain, 2010). Thus, microemulsions are considered to be an interesting possibility for anti-HIV drug delivery systems (Gupta & Jain, 2010). Carvalho et al. (2009) developed and characterized PPG-5-CETETH-20/oleic acid/water zidovudine-loaded microemulsions; the in vitro drug release assay showed that the drug release followed the Fickian diffusion through a disordered matrix, and the mechanism was identified by the use of Weibull mathematical model. Vyas et al. (2008), investigated the oral bioavailability of saquinavir incorporated in oil-in-water microemulsions in the study of enhanced brain disposition, a potential sanctuary site for HIV. Pharmacokinetics parameters were found to be higher in the brain, suggesting an enhanced rate and extended saquinavir absorption following oral administration of microemulsions. Thus, microemulsions may be very promising for HIV/AIDS therapy, in particular, for reducing the viral load in important anatomical

Polymeric micelles are nanostructures that have been utilized for improving aqueous solubility, mucosal permeability and disease-site targeting of several drug molecules.

**2. Surfactant systems** 

**2.1 Microemulsions** 

reservoir sites (Vyas et al., 2008).

**2.2 Polymeric micelles** 

(Hawkins, 2010), strategies currently being investigated to overcome these limitations include the design and development of novel drug delivery systems that can improve the efficacy of both existing and novel antiretroviral drugs (Ojewole et al., 2008). With the aim to reduce dosing frequency and to improve the compliance of the existing pharmacotherapy with patients, drug delivery system design is becoming complementary to new drug discovery (Sosnik et al., 2009). In the past decade, there has been an explosion of interest in the development of anti-HIV delivery systems (Ojewole et al., 2008). Evidence of this new interest is the emergence of several revision papers that have focused attention to the development of anti-HIV delivery systems, which have been published in the last two years (Geocze et al., 2010; Gupta & Jain, 2010; Hawkins, 2010; Neves et al., 2010; Sosnik et al., 2009, Wong et al., 2010, Khalil et al., 2011).

There is a special trend in research concerning the development of anti-HIV drug delivery systems, which apply nanotechnology to improve AIDS treatment. The basic concept behind the use of nanotechnology-based systems for antiretroviral drug delivery is the ability of these systems to compartmentalize as well as modify the properties and behavior of drugs in the biological medium. Through drug association with nanostructured systems, the properties that govern drug release are determined by the physicochemical properties of the nanosystems and not by the drug properties (Neves et al., 2010). These properties can include the protection of incorporated drugs from the metabolism, an increase of drug residence time in the human body and the possibility of targeting drugs to specific cells or organs. In addition, these properties can allow a dosage reduction, more appropriate dosage regimens, fewer adverse effects and increased patient compliance. Moreover, there is the possibility of incorporating different anti-HIV drugs in the same delivery system, which can also contribute to a simplification of drug administration schedules (Neves et al., 2010).

Despite the improvement of the nanotechnology-based studies, many of them are still in the pre-formulation or pre-clinical phases. However, the potential of nanotechnology-based drug delivery systems to improve AIDS treatment is evident. Thus, the goal of the current chapter is to organize a systematic review about this area of study. Particular emphasis was placed on surfactant and nanoparticulated systems. The surfactant systems are formed by amphiphilic compounds and include several types of arrangements; these systems selforganize with different physicochemical properties, which can be used to design new drug delivery systems. The most popular surfactant system that incorporates antiretroviral drugs are the liposomes (Zidan et al., 2010; Clayton et al., 2009; Kaur et al., 2008); however, anti-HIV drugs were also found to be associated with microemulsions (Vyas et al., 2008), polymeric micelles (Chiappetta et al. 2010; Kaparissides et al., 2006), self-assembled drug delivery systems (Jin et al., 2010) and liquid crystals (Carvalho et al., 2010a). In addition to surfactant systems, polymeric nanoparticles have been extensively studied as nanometric carriers, and these carriers presenting different morphologies, including nanospheres or nanocapsules. In this document, these are referred to as polymeric nanoparticles (Mainardes et al., 2010; Sharma & Garg, 2010, Schäfer et al., 1992, Shah & Amiji, 2006, Destache et al., 2009, Mahajan et al., 2009) and solid lipid nanoparticles (Aji Alex et al., 2010; Kuo & Chen, 2009; Shegokara et al., 2010). The advantages and limitations of each system are discussed, thus, this work can be used as a start point for researchers focusing on nanotechnologybased drug delivery systems for the treatment of AIDS.

## **2. Surfactant systems**

368 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

(Hawkins, 2010), strategies currently being investigated to overcome these limitations include the design and development of novel drug delivery systems that can improve the efficacy of both existing and novel antiretroviral drugs (Ojewole et al., 2008). With the aim to reduce dosing frequency and to improve the compliance of the existing pharmacotherapy with patients, drug delivery system design is becoming complementary to new drug discovery (Sosnik et al., 2009). In the past decade, there has been an explosion of interest in the development of anti-HIV delivery systems (Ojewole et al., 2008). Evidence of this new interest is the emergence of several revision papers that have focused attention to the development of anti-HIV delivery systems, which have been published in the last two years (Geocze et al., 2010; Gupta & Jain, 2010; Hawkins, 2010; Neves et al., 2010; Sosnik et al., 2009,

There is a special trend in research concerning the development of anti-HIV drug delivery systems, which apply nanotechnology to improve AIDS treatment. The basic concept behind the use of nanotechnology-based systems for antiretroviral drug delivery is the ability of these systems to compartmentalize as well as modify the properties and behavior of drugs in the biological medium. Through drug association with nanostructured systems, the properties that govern drug release are determined by the physicochemical properties of the nanosystems and not by the drug properties (Neves et al., 2010). These properties can include the protection of incorporated drugs from the metabolism, an increase of drug residence time in the human body and the possibility of targeting drugs to specific cells or organs. In addition, these properties can allow a dosage reduction, more appropriate dosage regimens, fewer adverse effects and increased patient compliance. Moreover, there is the possibility of incorporating different anti-HIV drugs in the same delivery system, which can also contribute to a simplification of drug

Despite the improvement of the nanotechnology-based studies, many of them are still in the pre-formulation or pre-clinical phases. However, the potential of nanotechnology-based drug delivery systems to improve AIDS treatment is evident. Thus, the goal of the current chapter is to organize a systematic review about this area of study. Particular emphasis was placed on surfactant and nanoparticulated systems. The surfactant systems are formed by amphiphilic compounds and include several types of arrangements; these systems selforganize with different physicochemical properties, which can be used to design new drug delivery systems. The most popular surfactant system that incorporates antiretroviral drugs are the liposomes (Zidan et al., 2010; Clayton et al., 2009; Kaur et al., 2008); however, anti-HIV drugs were also found to be associated with microemulsions (Vyas et al., 2008), polymeric micelles (Chiappetta et al. 2010; Kaparissides et al., 2006), self-assembled drug delivery systems (Jin et al., 2010) and liquid crystals (Carvalho et al., 2010a). In addition to surfactant systems, polymeric nanoparticles have been extensively studied as nanometric carriers, and these carriers presenting different morphologies, including nanospheres or nanocapsules. In this document, these are referred to as polymeric nanoparticles (Mainardes et al., 2010; Sharma & Garg, 2010, Schäfer et al., 1992, Shah & Amiji, 2006, Destache et al., 2009, Mahajan et al., 2009) and solid lipid nanoparticles (Aji Alex et al., 2010; Kuo & Chen, 2009; Shegokara et al., 2010). The advantages and limitations of each system are discussed, thus, this work can be used as a start point for researchers focusing on nanotechnology-

Wong et al., 2010, Khalil et al., 2011).

administration schedules (Neves et al., 2010).

based drug delivery systems for the treatment of AIDS.

Surfactants are extensively used as excipients in drug delivery and the understanding of the physicochemical properties and behavior of these amphiphilic compounds has undergone significant development. One reason for this development is that surface chemistry is a relatively young scientific discipline, and many studies have recognized its importance for the design and controlled use of drug delivery formulations (Malmsten, 2002). The surfactant systems comprise several types of arrangements, and they self-organize with different physicochemical properties that can be used to design new drug delivery systems that are able to solubilize both water-soluble and oil-soluble compounds. Depending on the composition and molecules of the component, surfactant systems can infinitely dilute or form highly stiff matrices, both in physiological conditions. Due the capacity of surfactant systems, which can form viscous and gel-like structures with different rheological characteristics, these systems are extremely versatile and can be delineated to be applied for different routes of administration, e.g., transdermal or mucosal administration, such as vaginal, nasal, rectal and sublingual. However, almost all studies aim oral administration, where anti-HIV drugs are associated in microemulsions, polymeric micelles, self-assembled drug delivery systems, liquid crystals and liposomes. Liposomes, contrary to the other surfactant organizations, have been widely explored for orally administered anti-HIV drugs. Although the oral route is still the primary mode of delivery for antiretroviral surfactant systems, they were found to be a suitable vehicle for anti-HIV intravenous and mucosal administration (described below).

#### **2.1 Microemulsions**

Microemulsions are systems consisting of water, oil, and surfactant(s), which constitute a single optically isotropic and thermodynamically stable liquid dispersion. Such systems are useful for drug delivery due to their excellent stability, ease of preparation, optical clarity, as well as their capacity to dissolve hydrophilic and lipophilic drugs, frequently in high amounts. Microemulsions differ from emulsions and nanoemulsions because of their thermodynamic stability; these systems form spontaneously, exhibit reduced droplets sizes (typically 10–100 nm) (Malmsten, 2002), higher surface areas and free energy without the inherent creaming, flocculation, coalescence and sedimentation associated with emulsions (Gupta & Jain, 2010). Thus, microemulsions are considered to be an interesting possibility for anti-HIV drug delivery systems (Gupta & Jain, 2010). Carvalho et al. (2009) developed and characterized PPG-5-CETETH-20/oleic acid/water zidovudine-loaded microemulsions; the in vitro drug release assay showed that the drug release followed the Fickian diffusion through a disordered matrix, and the mechanism was identified by the use of Weibull mathematical model. Vyas et al. (2008), investigated the oral bioavailability of saquinavir incorporated in oil-in-water microemulsions in the study of enhanced brain disposition, a potential sanctuary site for HIV. Pharmacokinetics parameters were found to be higher in the brain, suggesting an enhanced rate and extended saquinavir absorption following oral administration of microemulsions. Thus, microemulsions may be very promising for HIV/AIDS therapy, in particular, for reducing the viral load in important anatomical reservoir sites (Vyas et al., 2008).

#### **2.2 Polymeric micelles**

Polymeric micelles are nanostructures that have been utilized for improving aqueous solubility, mucosal permeability and disease-site targeting of several drug molecules.

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 371

Liquid crystals combine the properties of both liquid and solid states. They can be made to form a range of different nanostructures, including rods, lamellae, and bicontinuously interconnected structures, with alternative polar and non-polar layers, where aqueous drug solutions can be included. (Kaparissides et al., 2006; Malmsten, 2001). The spontaneous selfassembly of some lipids used to form liquid crystalline structures can offer a potentially new class of sustained release matrices. Depending to the liquid crystalline materials, they can be highly stable to dilution, which means they can persist as a reservoir for slow drug release in excess fluids, such as the gastrointestinal tract or subcutaneous regions. Drug release rates are directly related to the nanostructure of the matrix. The particular geometry into which the lipids assemble can be manipulated through either the use of additives to modify the assembly process or through modifying conditions, such as temperature (Boyd, 2010). The structure-forming lipids can absorb a certain amount of water and then spontaneously form gel-like phases with unique internal structures into which drugs can be incorporated. Moreover, non-toxic, biodegradable and bioadhesive properties also contribute to their applications towards drug delivery (Guo et al., 2010). Liquid crystal phases have been found to be mucoadhesive, with a range of mucosal surfaces; and the mechanism of mucoadhesion probably involves the rheological properties of the system, which are similar to the in situ gelling vehicles. These liquid crystal systems can be arranged in a very strong and viscous matrix that favors the mucosal retention, impeding the immediate removal of the formulation by the mucociliary clearance (Carvalho et al., 2010b). This property was used by Carvalho et al. (2010a) to develop a mucoadhesive surfactant system for the nasal administration of zidovudine. The nasal route has been explored to avoid the extensive, first-pass metabolism and poor oral bioavailability of drugs that suffer hepatic metabolism or gastric degradation when administered by the oral route. Thus, the nasal route is an option for enhancing the therapeutic efficacy of drugs and to reduce the extent of their firstpass effect because this route is highly vascular and has a great superficial area of absorption. However, there are some mechanisms that limit the intranasal absorption, such as the mucociliary clearance, which rapidly removes the formulation from the nasal cavity. Systems composed of PPG-5-CETETH-20 as surfactant, oleic acid and water have shown to display phase transition to the lamellar phase when put in contact with the aqueous nasalsimulated mucus (SM). The phase transition was accompanied by an increase in the system's elasticity, in addition to the presentation of suitable mucoadhesive force. Thus, a viscous and mucoadhesive liquid crystalline matrix can be formed when the formulations are in contact with the SM, which may prolong the residence time of zidovudine in the nasal cavity. These findings indicate a potentially useful system for the nasal administration of

Liposomes can be defined as associations of colloidal amphipathic lipids that spontaneously arrange themselves in closed structures, such as spherical shells containing aqueous cores. The unique aspect of the liposomes is that the hydrophilic drugs can be encapsulated in the aqueous layer, while the hydrophobic drugs can be incorporated into the phospholipid bilayer. They can range in size from 25 nm up to several microns, and liposomes are prepared from natural or synthetic phospholipids and cholesterol; in addition, they may include other substances, such as lipids and proteins (Sharma & Sharma, 1997). Conventional liposomes (without surface modification) are naturally taken up by cells of the

**2.4 Liquid cristal** 

zidovudine (Carvalho et al., 2010a).

**2.5 Liposomes** 

Compared to the conventional surfactant based micelles, polymeric micelles are composed of block copolymers. Although the structural "core–shell" arrangement is similar to surfactant-based micelles, polymer micelles self associate at much lower concentrations. Consequently, the thermodynamic and in vivo stability of polymeric micelles is relatively high (Sharma & Garg, 2010). A study by Chiappeta et al. (2010) showed that the solubility of efavirenz, a lipophilic first-line antiretroviral drug, could be improved. Micellar systems composed of N-methylated and N-alkylated poloxamines (X-shaped poly(ethylene oxide)– poly(propylene oxide, PEO–PPO) diblocks connected to a central ethylenediamine group), were investigated to optimize the oral pharmacotherapy effects of efavirenz. The in vitro release was sustained for at least 24 h. The authors suggested that the polymeric micelles could be promising nanocarriers for oral or parenteral drug delivery. The aqueous solubility of the drug was increased from 0.004 mg/mL to approximately 30 mg/mL, representing the best solubilization performance in an aqueous medium of any nanocarrier described thus far (Chiappetta et al. 2010). Other characteristics of polymeric micelles present the possibility for substitution of the block copolymer micelles with specific ligands, which is a very promising strategy for a broader range of sites of activity with a considerably higher selectivity (Kaparissides et al., 2006). Micelles can be tailored by attaching hydrophilic blocks to antibodies or other ligands specific for the type of receptors present within the disease site. Lectin receptors are present on HIV reservoirs, such as T lymphocytes, dendritic cells and macrophages; therefore, this can be a promising approach for viral reservoir targeting (Sharma & Garg, 2010).

#### **2.3 Self-assembled drug delivery systems of antiretroviral prodrugs**

A novel technology involving antiretroviral prodrugs with amphiphilic properties have been developed by Jin and co-workers (Jin et al., 2008, 2009, 2010). The molecular selfassembly properties of those prodrugs in aqueous media permit the formation of nanostructures with amphiphilic characteristics, allowing them to cross biomembranes and deliver themselves in vivo without carriers (Jin et al., 2008). Recently, a series of cholesteryl derivatives of antiviral nucleoside analogues were synthesized by this group, which involved acyclovir, zidovudine and didanosine. The morphologies and the morphological transformation of cholesterylsuccinyl didanosine was investigated as a prodrug with representative self-assembly behavior in aqueous media. Results showed that the resulted nanoparticulate system had a narrow size distribution, which allowed heat sterilization and showed a site-specific distribution for the anti-HIV therapy after IV administration (Jin et al., 2008). Another example of this technology was the synthesis of the amphiphilic prodrug anti-HIV zidovudine, cholesteryl-phosphonyl zidovudine. This system degraded quickly in biological environments and showed high anti-HIV activity; in addition, the system targeted the mononuclear phagocyte system (MPS) and was followed by degradation at the targeted organs (Jin et al., 2009). Furthermore, a study was recently published on the synthesis of an amphiphilic prodrug containing dual zidovudine (Jin et al., 2010). The stable and concentrated vesicular self-assemblies were prepared through injecting the prodrug solution into water followed by adding stabilizers and removing solvents. Properties, such as their nanoscale size, stability, anti-HIV activity and macrophage targeting effects, have demonstrated that the prodrug is a promising self assembled drug delivery system. Moreover, this kind of system containing different drugs would benefit a combination therapy for AIDS treatment (Jin et al., 2010).

#### **2.4 Liquid cristal**

370 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Compared to the conventional surfactant based micelles, polymeric micelles are composed of block copolymers. Although the structural "core–shell" arrangement is similar to surfactant-based micelles, polymer micelles self associate at much lower concentrations. Consequently, the thermodynamic and in vivo stability of polymeric micelles is relatively high (Sharma & Garg, 2010). A study by Chiappeta et al. (2010) showed that the solubility of efavirenz, a lipophilic first-line antiretroviral drug, could be improved. Micellar systems composed of N-methylated and N-alkylated poloxamines (X-shaped poly(ethylene oxide)– poly(propylene oxide, PEO–PPO) diblocks connected to a central ethylenediamine group), were investigated to optimize the oral pharmacotherapy effects of efavirenz. The in vitro release was sustained for at least 24 h. The authors suggested that the polymeric micelles could be promising nanocarriers for oral or parenteral drug delivery. The aqueous solubility of the drug was increased from 0.004 mg/mL to approximately 30 mg/mL, representing the best solubilization performance in an aqueous medium of any nanocarrier described thus far (Chiappetta et al. 2010). Other characteristics of polymeric micelles present the possibility for substitution of the block copolymer micelles with specific ligands, which is a very promising strategy for a broader range of sites of activity with a considerably higher selectivity (Kaparissides et al., 2006). Micelles can be tailored by attaching hydrophilic blocks to antibodies or other ligands specific for the type of receptors present within the disease site. Lectin receptors are present on HIV reservoirs, such as T lymphocytes, dendritic cells and macrophages; therefore, this can be a promising approach for viral reservoir

targeting (Sharma & Garg, 2010).

therapy for AIDS treatment (Jin et al., 2010).

**2.3 Self-assembled drug delivery systems of antiretroviral prodrugs** 

A novel technology involving antiretroviral prodrugs with amphiphilic properties have been developed by Jin and co-workers (Jin et al., 2008, 2009, 2010). The molecular selfassembly properties of those prodrugs in aqueous media permit the formation of nanostructures with amphiphilic characteristics, allowing them to cross biomembranes and deliver themselves in vivo without carriers (Jin et al., 2008). Recently, a series of cholesteryl derivatives of antiviral nucleoside analogues were synthesized by this group, which involved acyclovir, zidovudine and didanosine. The morphologies and the morphological transformation of cholesterylsuccinyl didanosine was investigated as a prodrug with representative self-assembly behavior in aqueous media. Results showed that the resulted nanoparticulate system had a narrow size distribution, which allowed heat sterilization and showed a site-specific distribution for the anti-HIV therapy after IV administration (Jin et al., 2008). Another example of this technology was the synthesis of the amphiphilic prodrug anti-HIV zidovudine, cholesteryl-phosphonyl zidovudine. This system degraded quickly in biological environments and showed high anti-HIV activity; in addition, the system targeted the mononuclear phagocyte system (MPS) and was followed by degradation at the targeted organs (Jin et al., 2009). Furthermore, a study was recently published on the synthesis of an amphiphilic prodrug containing dual zidovudine (Jin et al., 2010). The stable and concentrated vesicular self-assemblies were prepared through injecting the prodrug solution into water followed by adding stabilizers and removing solvents. Properties, such as their nanoscale size, stability, anti-HIV activity and macrophage targeting effects, have demonstrated that the prodrug is a promising self assembled drug delivery system. Moreover, this kind of system containing different drugs would benefit a combination Liquid crystals combine the properties of both liquid and solid states. They can be made to form a range of different nanostructures, including rods, lamellae, and bicontinuously interconnected structures, with alternative polar and non-polar layers, where aqueous drug solutions can be included. (Kaparissides et al., 2006; Malmsten, 2001). The spontaneous selfassembly of some lipids used to form liquid crystalline structures can offer a potentially new class of sustained release matrices. Depending to the liquid crystalline materials, they can be highly stable to dilution, which means they can persist as a reservoir for slow drug release in excess fluids, such as the gastrointestinal tract or subcutaneous regions. Drug release rates are directly related to the nanostructure of the matrix. The particular geometry into which the lipids assemble can be manipulated through either the use of additives to modify the assembly process or through modifying conditions, such as temperature (Boyd, 2010). The structure-forming lipids can absorb a certain amount of water and then spontaneously form gel-like phases with unique internal structures into which drugs can be incorporated. Moreover, non-toxic, biodegradable and bioadhesive properties also contribute to their applications towards drug delivery (Guo et al., 2010). Liquid crystal phases have been found to be mucoadhesive, with a range of mucosal surfaces; and the mechanism of mucoadhesion probably involves the rheological properties of the system, which are similar to the in situ gelling vehicles. These liquid crystal systems can be arranged in a very strong and viscous matrix that favors the mucosal retention, impeding the immediate removal of the formulation by the mucociliary clearance (Carvalho et al., 2010b). This property was used by Carvalho et al. (2010a) to develop a mucoadhesive surfactant system for the nasal administration of zidovudine. The nasal route has been explored to avoid the extensive, first-pass metabolism and poor oral bioavailability of drugs that suffer hepatic metabolism or gastric degradation when administered by the oral route. Thus, the nasal route is an option for enhancing the therapeutic efficacy of drugs and to reduce the extent of their firstpass effect because this route is highly vascular and has a great superficial area of absorption. However, there are some mechanisms that limit the intranasal absorption, such as the mucociliary clearance, which rapidly removes the formulation from the nasal cavity. Systems composed of PPG-5-CETETH-20 as surfactant, oleic acid and water have shown to display phase transition to the lamellar phase when put in contact with the aqueous nasalsimulated mucus (SM). The phase transition was accompanied by an increase in the system's elasticity, in addition to the presentation of suitable mucoadhesive force. Thus, a viscous and mucoadhesive liquid crystalline matrix can be formed when the formulations are in contact with the SM, which may prolong the residence time of zidovudine in the nasal cavity. These findings indicate a potentially useful system for the nasal administration of zidovudine (Carvalho et al., 2010a).

#### **2.5 Liposomes**

Liposomes can be defined as associations of colloidal amphipathic lipids that spontaneously arrange themselves in closed structures, such as spherical shells containing aqueous cores. The unique aspect of the liposomes is that the hydrophilic drugs can be encapsulated in the aqueous layer, while the hydrophobic drugs can be incorporated into the phospholipid bilayer. They can range in size from 25 nm up to several microns, and liposomes are prepared from natural or synthetic phospholipids and cholesterol; in addition, they may include other substances, such as lipids and proteins (Sharma & Sharma, 1997). Conventional liposomes (without surface modification) are naturally taken up by cells of the

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 373

the sterically stabilized liposomes remain concentrated in the spleen (Harvie et al., 1995, 1996). In a recent study, a prodrug of didanosine in a liposomal formulation displayed antiviral activity and showed a promising enhancement of the drug activity against HIV-1

The effect of the liposome composition and cholesterol on the cellular uptake of stavudine by human M/Ms was verified by Katragadda and co-workers (2000). The cells were uptaken more expressively by the negatively charged liposomes (containing phosphatidylserine and dicetyl phosphate) compared to either the neutral or positive liposomes. The authors suggested that the difference in stavudine liposome uptake in the presence of charge might be due in part to the extent of the interaction between the charged bilayer and the cells. Other studies involving stavudine and liposomes were studied by Garg et al. (2006, 2007). Primarily, they observed that the elimination half-life and mean residence time of stavudine were increased when they were encapsulated in the mannosylated and galactosylated liposomes. Stavudine-loaded mannosylated liposomes presented in vitro antiretroviral activity. In addition, the two liposomal formulations resulted in reduced hematological toxicity and enhanced the hepatic cellular uptake of the stavudine. Furthermore, the group demonstrated that the antiretroviral activity of stavudine in galactosylated liposomes is dose-dependent, in a study with infected cell culture (Garg et

Immunoliposomes (liposomes with antibody attached) have also been used to deliver antiretroviral drugs to HIV targets. The Betageri group (1993a, 1993b) attached a mouse antibody in liposomes containing stavudine-triphosphate or zalcitabine-triphosphate and observed a significant increase in uptake by human macrophages compared to the free drug and unmodified liposomes. Gagné and co-workers (2002) showed that immunoliposomes were very efficient in delivering high concentrations of indinavir to lymphoid tissues (126 times higher than the free drug) for at least 15 days, post a single subcutaneous injection in mice. The HLA-DR determinant of major histocompatibility complex class II is highly expressed on macrophages and activated CD4+ T cells. Also, the authors showed that the immunoliposomal indinavir was as efficient as the free drug to inhibit HIV-1 replication in cultured cells. A recent study (Clayton et al., 2009) demonstrated specific targeting and delivery of a novel protease inhibitor encapsulated in PEGylated immunoliposomes (coated with a F105 Fab' fragment). The immunoliposome was shown to enable selective delivery of the drug to HIV-1-infected cells and also demonstrated that the effect of the targeted drug on viral replication was greater than the effect of a comparable concentration of the free drug or non-targeted drug. Therefore, the potential of liposomes and various ligands for the active targeting of antiretroviral drugs loaded on liposomes has on development. These studies have shown potential benefits of liposomes for improving antiretroviral drug

Nanoparticles are solid, colloidal particles consisting of macromolecular substances varying in size from 10 to 1,000 nm. The drug can be dissolved, entrapped, adsorbed, attached or encapsulated into a nanoparticle. Depending on the method of preparation, nanospheres or nanocapsules can be developed with different properties; in addition, different release characteristics for the encapsulated therapeutic agent can also be developed. For nearly three decades, polymeric nanoparticles have been extensively studied due to their unique

in in vitro infected cell cultures (Lalanne et al., 2007).

al., 2008).

therapy.

**3. Polymeric nanoparticles** 

MPS, an important HIV reservoir. Additionally, the liposome surface can be modified to improve its properties. Ligands that promote active targeting of liposomes to HIV-infected cells and organs are interesting alternatives. Liposomes represent a convenient approach to improve the delivery of anti-HIV agents into infected cells, thereby improving the efficacy of drugs and reducing their adverse side effects (Desormeaux et al., 1998).

Monocytes/macrophages (M/Ms) are widely recognized as the secondary cellular target of HIV-1 and a crucial virus reservoir. HIV-1-infected M/Ms cells are widely distributed in all tissues and organs, including the CNS, and the HIV-1 replication in these cells is a crucial pathogenic event during the progression of viral infection. Also, M/Ms are resistant to the cytopathic effect of HIV-1 and produce viruses over a prolonged period, consisting of a long-term viral reservoir (Gartner et al., 1986; Garaci et al., 1999).

The primary research studies involving the application of liposomes in AIDS treatment are based in in vitro and in vivo (animals) experiments that consider the ability of liposomes to increase the intracellular delivery of antiretroviral drugs. The most popular drugs studied are zalcitabine, zidovudine, didanosine, stavudine and indinavir. One of the first studies that introduced the application of liposomes as carriers for anti-HIV drugs was realized by Szebeni and co-workers (1990). The group suggested that the capability of liposomes for targeting drugs in vivo to macrophages could potentially be exploited to improve the therapeutic index of dideoxynucleoside drugs. They also demonstrated the antiviral effects of 2',3'-dideoxycytidine-5'-triphosphate-loaded liposomes in cultured human M/Ms infected with HIV-1 and the higher drug stability in presence of liposome. Another study involving 2',3'-dideoxycytidine (zalcitabine) showed that the anionic character of the liposome seemed to be an important factor to obtain a high intracellular uptake. The lipid component can interfere in interactions between the cell and the liposome (Makabi-Panzu et al., 1998).

After the discovery of the zidovudine associated hematotoxicity (Ganser et al., 1989), the effect of liposome encapsulation on the bone marrow toxicity and antiviral activity of zidovudine in mice was determined by the Phillips group (1991, 1992). The results showed that zidovudine encapsulated in liposomes exhibited no bone marrow toxicity at doses that were cytotoxic with zidovudine solution; in addition, erythrocyte and leukocyte levels remained normal. Also, zidovudine loaded liposomes presented a better and prolonged antiretroviral response compared to the zidovudine solution. A more recent study showed that galactosylated liposomes reduced hematopoietic toxicity, enhanced cellular uptake and altered pharmacokinetics of zidovudine (Gard & Jain, 2006). Studies performed by the Jain group (2006, 2008) reported on the application of zidovudine loaded liposomes via transdermal route. The results showed that zidovudine permeation was higher from liposomal formulations, and it was able to target the drug to MPS organs more effectively than the free drug. In 2008, Kaur and co-workers demonstrated that mannosylatedliposomes were able to target zidovudine to the spleen and lymph nodes after subcutaneous administration. The mannose receptors in the spleen explain the role of mannose on the liposome surface and the highest drug localization in this organ.

The Désormeaux group (1994) was one of the first to study liposomal formulations for didanosine. They found that the liposomes modified the drug tissue distribution and plasma pharmacokinetics, resulting in a marked improvement of drug biodistribution, especially into the MPS. Furthermore, they reported that didanosine was efficiently targeted to lymph nodes and macrophage-rich tissue when it was loaded in liposomes. The group showed that the liposomes were able to increase the plasma half-life of the drug, and also

MPS, an important HIV reservoir. Additionally, the liposome surface can be modified to improve its properties. Ligands that promote active targeting of liposomes to HIV-infected cells and organs are interesting alternatives. Liposomes represent a convenient approach to improve the delivery of anti-HIV agents into infected cells, thereby improving the efficacy of

Monocytes/macrophages (M/Ms) are widely recognized as the secondary cellular target of HIV-1 and a crucial virus reservoir. HIV-1-infected M/Ms cells are widely distributed in all tissues and organs, including the CNS, and the HIV-1 replication in these cells is a crucial pathogenic event during the progression of viral infection. Also, M/Ms are resistant to the cytopathic effect of HIV-1 and produce viruses over a prolonged period, consisting of a

The primary research studies involving the application of liposomes in AIDS treatment are based in in vitro and in vivo (animals) experiments that consider the ability of liposomes to increase the intracellular delivery of antiretroviral drugs. The most popular drugs studied are zalcitabine, zidovudine, didanosine, stavudine and indinavir. One of the first studies that introduced the application of liposomes as carriers for anti-HIV drugs was realized by Szebeni and co-workers (1990). The group suggested that the capability of liposomes for targeting drugs in vivo to macrophages could potentially be exploited to improve the therapeutic index of dideoxynucleoside drugs. They also demonstrated the antiviral effects of 2',3'-dideoxycytidine-5'-triphosphate-loaded liposomes in cultured human M/Ms infected with HIV-1 and the higher drug stability in presence of liposome. Another study involving 2',3'-dideoxycytidine (zalcitabine) showed that the anionic character of the liposome seemed to be an important factor to obtain a high intracellular uptake. The lipid component can interfere in interactions between the

After the discovery of the zidovudine associated hematotoxicity (Ganser et al., 1989), the effect of liposome encapsulation on the bone marrow toxicity and antiviral activity of zidovudine in mice was determined by the Phillips group (1991, 1992). The results showed that zidovudine encapsulated in liposomes exhibited no bone marrow toxicity at doses that were cytotoxic with zidovudine solution; in addition, erythrocyte and leukocyte levels remained normal. Also, zidovudine loaded liposomes presented a better and prolonged antiretroviral response compared to the zidovudine solution. A more recent study showed that galactosylated liposomes reduced hematopoietic toxicity, enhanced cellular uptake and altered pharmacokinetics of zidovudine (Gard & Jain, 2006). Studies performed by the Jain group (2006, 2008) reported on the application of zidovudine loaded liposomes via transdermal route. The results showed that zidovudine permeation was higher from liposomal formulations, and it was able to target the drug to MPS organs more effectively than the free drug. In 2008, Kaur and co-workers demonstrated that mannosylatedliposomes were able to target zidovudine to the spleen and lymph nodes after subcutaneous administration. The mannose receptors in the spleen explain the role of mannose on the

The Désormeaux group (1994) was one of the first to study liposomal formulations for didanosine. They found that the liposomes modified the drug tissue distribution and plasma pharmacokinetics, resulting in a marked improvement of drug biodistribution, especially into the MPS. Furthermore, they reported that didanosine was efficiently targeted to lymph nodes and macrophage-rich tissue when it was loaded in liposomes. The group showed that the liposomes were able to increase the plasma half-life of the drug, and also

drugs and reducing their adverse side effects (Desormeaux et al., 1998).

long-term viral reservoir (Gartner et al., 1986; Garaci et al., 1999).

cell and the liposome (Makabi-Panzu et al., 1998).

liposome surface and the highest drug localization in this organ.

the sterically stabilized liposomes remain concentrated in the spleen (Harvie et al., 1995, 1996). In a recent study, a prodrug of didanosine in a liposomal formulation displayed antiviral activity and showed a promising enhancement of the drug activity against HIV-1 in in vitro infected cell cultures (Lalanne et al., 2007).

The effect of the liposome composition and cholesterol on the cellular uptake of stavudine by human M/Ms was verified by Katragadda and co-workers (2000). The cells were uptaken more expressively by the negatively charged liposomes (containing phosphatidylserine and dicetyl phosphate) compared to either the neutral or positive liposomes. The authors suggested that the difference in stavudine liposome uptake in the presence of charge might be due in part to the extent of the interaction between the charged bilayer and the cells. Other studies involving stavudine and liposomes were studied by Garg et al. (2006, 2007). Primarily, they observed that the elimination half-life and mean residence time of stavudine were increased when they were encapsulated in the mannosylated and galactosylated liposomes. Stavudine-loaded mannosylated liposomes presented in vitro antiretroviral activity. In addition, the two liposomal formulations resulted in reduced hematological toxicity and enhanced the hepatic cellular uptake of the stavudine. Furthermore, the group demonstrated that the antiretroviral activity of stavudine in galactosylated liposomes is dose-dependent, in a study with infected cell culture (Garg et al., 2008).

Immunoliposomes (liposomes with antibody attached) have also been used to deliver antiretroviral drugs to HIV targets. The Betageri group (1993a, 1993b) attached a mouse antibody in liposomes containing stavudine-triphosphate or zalcitabine-triphosphate and observed a significant increase in uptake by human macrophages compared to the free drug and unmodified liposomes. Gagné and co-workers (2002) showed that immunoliposomes were very efficient in delivering high concentrations of indinavir to lymphoid tissues (126 times higher than the free drug) for at least 15 days, post a single subcutaneous injection in mice. The HLA-DR determinant of major histocompatibility complex class II is highly expressed on macrophages and activated CD4+ T cells. Also, the authors showed that the immunoliposomal indinavir was as efficient as the free drug to inhibit HIV-1 replication in cultured cells. A recent study (Clayton et al., 2009) demonstrated specific targeting and delivery of a novel protease inhibitor encapsulated in PEGylated immunoliposomes (coated with a F105 Fab' fragment). The immunoliposome was shown to enable selective delivery of the drug to HIV-1-infected cells and also demonstrated that the effect of the targeted drug on viral replication was greater than the effect of a comparable concentration of the free drug or non-targeted drug. Therefore, the potential of liposomes and various ligands for the active targeting of antiretroviral drugs loaded on liposomes has on development. These studies have shown potential benefits of liposomes for improving antiretroviral drug therapy.

## **3. Polymeric nanoparticles**

Nanoparticles are solid, colloidal particles consisting of macromolecular substances varying in size from 10 to 1,000 nm. The drug can be dissolved, entrapped, adsorbed, attached or encapsulated into a nanoparticle. Depending on the method of preparation, nanospheres or nanocapsules can be developed with different properties; in addition, different release characteristics for the encapsulated therapeutic agent can also be developed. For nearly three decades, polymeric nanoparticles have been extensively studied due to their unique

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 375

delivery and because the use of a single antiretroviral in the treatment of HIV-1 only resulted in the development of resistant strains and treatment failures. In another study, the same group showed that these nanoparticles were able to maintain the plasmatic drug concentrations for a prolonged period, after intraperitoneal administration in mice. Also, the drug concentration in the brain was significantly higher with drug-loaded nanoparticles than with the free drug. Additionally, the antiretroviral drug-loaded nanoparticles were able to interact with the M/Ms infected with HIV-1 and inhibit virus replication up to 1000 fold for 10 days compared to the free drugs (Destache et al., 2010). Poly(lactic) acid (PLA) and PLA-polyethylene glycol (PLA-PEG) blended nanoparticles containing zidovudine were developed, and their uptake by polymorphonuclear leucocytes from rats was studied in vitro. The results showed that the PLA nanoparticles were more efficiently phagocytosed than PLA-PEG blends and were able to activate a larger number of cells than the blended PLA-PEG nanoparticles (Mainardes et al., 2009). Furthermore, the group evaluated the pharmacokinetic profile of these nanoparticles in rats after a single intranasal administration. Blended PLA-PEG nanoparticles exhibited a sustained release of the drug over 24 h, while PLA nanoparticles were sustained up to 10 h. The half-life of zidovudine also varied among the formulations. The slow elimination rate (Ke) resulted in significantly prolonged t1/2 values for zidovudine from the PLA and blended PLA-PEG nanoparticles compared to the zidovudine solution. Because of the slow release of zidovudine from the nanoparticles, its metabolic breakdown was also slower, increasing the mean half-life. The significant increase (p< 0.05) in the value of the area under curve (AUC) for the zidovudineloaded PLA-PEG nanoparticles, compared to the PLA nanoparticles and zidovudine aqueous solution, distinctly indicated the improved intranasal bioavailability of the blended system (Mainardes et al., 2010). Thus, the results of this study corroborated those of the first study, indicating that the physicochemical characteristics of nanoparticles intended for controlled drug release is very important because these characteristics can govern the application of the formulation and can be used to predict its behavior in the biological medium. The size and surface charge are also important parameters in a nanostructured system because these characteristics interfere directly in biological processes, such as the transport across biological membranes and the recognition by M/Ms and biodistribution. Another important factor that must be taken into account in the design strategies used to improve AIDS treatment is the brain delivery system of antiretroviral drugs. Because of the restricted entry of anti-HIV drugs, the brain is thought to form a viral sanctuary, and the treatment and control of HIV within this reservoir must be primordial. Nanoparticles can enhance the brain-drug delivery by three major pathways, which include the following: i) increasing the local drug gradient at the Blood Brain Barrier (BBB) by passive targeting, ii) allowing drug-trafficking by non-specific or receptor-mediated endocytosis and iii) blocking drug efflux transporters at the BBB (Wong et al., 2010). Consequently, the use of nanocarriers should help to achieve higher concentrations of encapsulated drugs and also

One of the most used polymers for the development of nanoparticles intended for brain delivery is poly-(butylcyanoacrylate) (PBCA) (Koziara et al., 2006). Studies have shown that the surface modification of PBCA nanoparticles using other polymers or surfactant agents, such as polysorbate 80, could increase the transport of particles through the BBB. Polysorbate 80 has been found to increase the translocation of nanoparticles by increasing the particle interaction with the low density lipoprotein (LDL) receptor-mediated endocytic pathway in brain endothelial cells and by inhibition the efflux function of P-gp (Goppert & Muller, 2005).

allow their prolonged residence in the CNS.

and valuable physicochemical and biological properties. Nanoparticles can improve drug actuation by the following characteristics: protecting it from degradation (higher physical stability during storage and in biological fluids), enhancing its transport and distribution (possibility through drug targeting by modification of surface charge with inserted ligands, such as antibodies, surfactants, and polymers) and prolonging its release (ability to sustain the drug release over a period of days to weeks). Therefore, nanoparticles may improve the plasma half-life of the entrapped drug (Allémann et al., 1993; Oppenheim, 1981). The drug pharmacokinetics parameters are altered when the drug is loaded in nanoparticles, and the particle surface composition plays an important role in drug bioavailability, which can be greater or lower than the drug solution/powder ratio, depending on the polymer used (Ubrich et al., 2005). Some characteristics of nanoparticles, such as particle size and surface charge, can be modulated by modifying some process parameters of formulation; they can be used in various applications. The research involving the applications of polymeric nanoparticles in AIDS treatment is primarily directed to increasing the intracellular and brain delivery of antiretroviral drugs. Thus, it is clear that M/Ms represent an important target for antiretroviral drugs and for carriers loaded with these drugs. The nanoparticles represent an attractive alternative in AIDS treatment because they consist of a carrier system intended for targeting M/Ms. When administered intravenously, conventional nanoparticles are rapidly cleared from the bloodstream by the MPS, represented by M/Ms. The particle uptake by cells is affected by the particle's physicochemical properties, such as particle size, surface charge, hydrophobicity and presence of a coating (varying in density/conformation) (Stolnik et al., 2005; Owens & Peppas, 2006).

Schäfer and co-workers (1992) were pioneers in studies involving antiretroviral drugs and macrophage targeting using nanoparticles. The authors found that the physicochemical properties, including the composition, surface characteristics and size, of poly(alkylcyanoacrylate) (PACA), poly(metylmethacrylate) (PMMA) and human serum albumin (HSA) nanoparticles containing zidovudine influenced the rate of uptake by macrophages, particularly when these cells were infected by HIV (up to 60% more than for uninfected macrophages). Also, the group demonstrated the effectiveness of poly(hexylcyanoacrylate) (PHCA) and HSA nanoparticles containing zidovudine and didanosine in preventing HIV infection in M/Ms cultures in vitro (Bender et al., 1994). Furthermore, the group prepared PHCA nanoparticles as carriers for saquinavir or zalcitabine and demonstrated that the both nanoparticles formulations led to a dosedependent reduction of HIV-1 antigen production in vitro in primary human M/Ms cultures (Bender et al., 1996). In a similar study, saquinavir carried in poly(ethyleneoxide) modified poly(epsilon-caprolactone) (PEO-PCL) nanoparticles was significantly internalized by the THP-1 human M/Ms cell line at a 10-fold higher rate than an aqueous solution of saquinavir (Shah & Amiji, 2006). In another study, Hillaireau et al. (2006) demonstrated that nanocapsules composed of PIBCA and poly(ethyleneimine) increased the intracellular uptake of azidothymidine-triphosphate 10- to 30-fold higher that the free drug, in a mouse macrophages culture. Destache and co-workers (2009) developed poly(lactic-co-glycolic) acid (PLGA) nanoparticles containing ritonavir, lopinavir and efavirenz and the results of the in vitro release of the drugs from the nanoparticles in human peripheral blood mononuclear cells showed an intracellular peak of each drug over a 28-day period, while the free drugs were eliminated in 2 days. The authors also demonstrated that nanoparticles were not significantly cytotoxic over macrophages. These results are important because they demonstrate that the three drugs can be incorporated into a single nanoparticle for drug

and valuable physicochemical and biological properties. Nanoparticles can improve drug actuation by the following characteristics: protecting it from degradation (higher physical stability during storage and in biological fluids), enhancing its transport and distribution (possibility through drug targeting by modification of surface charge with inserted ligands, such as antibodies, surfactants, and polymers) and prolonging its release (ability to sustain the drug release over a period of days to weeks). Therefore, nanoparticles may improve the plasma half-life of the entrapped drug (Allémann et al., 1993; Oppenheim, 1981). The drug pharmacokinetics parameters are altered when the drug is loaded in nanoparticles, and the particle surface composition plays an important role in drug bioavailability, which can be greater or lower than the drug solution/powder ratio, depending on the polymer used (Ubrich et al., 2005). Some characteristics of nanoparticles, such as particle size and surface charge, can be modulated by modifying some process parameters of formulation; they can be used in various applications. The research involving the applications of polymeric nanoparticles in AIDS treatment is primarily directed to increasing the intracellular and brain delivery of antiretroviral drugs. Thus, it is clear that M/Ms represent an important target for antiretroviral drugs and for carriers loaded with these drugs. The nanoparticles represent an attractive alternative in AIDS treatment because they consist of a carrier system intended for targeting M/Ms. When administered intravenously, conventional nanoparticles are rapidly cleared from the bloodstream by the MPS, represented by M/Ms. The particle uptake by cells is affected by the particle's physicochemical properties, such as particle size, surface charge, hydrophobicity and presence of a coating (varying in

Schäfer and co-workers (1992) were pioneers in studies involving antiretroviral drugs and macrophage targeting using nanoparticles. The authors found that the physicochemical properties, including the composition, surface characteristics and size, of poly(alkylcyanoacrylate) (PACA), poly(metylmethacrylate) (PMMA) and human serum albumin (HSA) nanoparticles containing zidovudine influenced the rate of uptake by macrophages, particularly when these cells were infected by HIV (up to 60% more than for uninfected macrophages). Also, the group demonstrated the effectiveness of poly(hexylcyanoacrylate) (PHCA) and HSA nanoparticles containing zidovudine and didanosine in preventing HIV infection in M/Ms cultures in vitro (Bender et al., 1994). Furthermore, the group prepared PHCA nanoparticles as carriers for saquinavir or zalcitabine and demonstrated that the both nanoparticles formulations led to a dosedependent reduction of HIV-1 antigen production in vitro in primary human M/Ms cultures (Bender et al., 1996). In a similar study, saquinavir carried in poly(ethyleneoxide) modified poly(epsilon-caprolactone) (PEO-PCL) nanoparticles was significantly internalized by the THP-1 human M/Ms cell line at a 10-fold higher rate than an aqueous solution of saquinavir (Shah & Amiji, 2006). In another study, Hillaireau et al. (2006) demonstrated that nanocapsules composed of PIBCA and poly(ethyleneimine) increased the intracellular uptake of azidothymidine-triphosphate 10- to 30-fold higher that the free drug, in a mouse macrophages culture. Destache and co-workers (2009) developed poly(lactic-co-glycolic) acid (PLGA) nanoparticles containing ritonavir, lopinavir and efavirenz and the results of the in vitro release of the drugs from the nanoparticles in human peripheral blood mononuclear cells showed an intracellular peak of each drug over a 28-day period, while the free drugs were eliminated in 2 days. The authors also demonstrated that nanoparticles were not significantly cytotoxic over macrophages. These results are important because they demonstrate that the three drugs can be incorporated into a single nanoparticle for drug

density/conformation) (Stolnik et al., 2005; Owens & Peppas, 2006).

delivery and because the use of a single antiretroviral in the treatment of HIV-1 only resulted in the development of resistant strains and treatment failures. In another study, the same group showed that these nanoparticles were able to maintain the plasmatic drug concentrations for a prolonged period, after intraperitoneal administration in mice. Also, the drug concentration in the brain was significantly higher with drug-loaded nanoparticles than with the free drug. Additionally, the antiretroviral drug-loaded nanoparticles were able to interact with the M/Ms infected with HIV-1 and inhibit virus replication up to 1000 fold for 10 days compared to the free drugs (Destache et al., 2010). Poly(lactic) acid (PLA) and PLA-polyethylene glycol (PLA-PEG) blended nanoparticles containing zidovudine were developed, and their uptake by polymorphonuclear leucocytes from rats was studied in vitro. The results showed that the PLA nanoparticles were more efficiently phagocytosed than PLA-PEG blends and were able to activate a larger number of cells than the blended PLA-PEG nanoparticles (Mainardes et al., 2009). Furthermore, the group evaluated the pharmacokinetic profile of these nanoparticles in rats after a single intranasal administration. Blended PLA-PEG nanoparticles exhibited a sustained release of the drug over 24 h, while PLA nanoparticles were sustained up to 10 h. The half-life of zidovudine also varied among the formulations. The slow elimination rate (Ke) resulted in significantly prolonged t1/2 values for zidovudine from the PLA and blended PLA-PEG nanoparticles compared to the zidovudine solution. Because of the slow release of zidovudine from the nanoparticles, its metabolic breakdown was also slower, increasing the mean half-life. The significant increase (p< 0.05) in the value of the area under curve (AUC) for the zidovudineloaded PLA-PEG nanoparticles, compared to the PLA nanoparticles and zidovudine aqueous solution, distinctly indicated the improved intranasal bioavailability of the blended system (Mainardes et al., 2010). Thus, the results of this study corroborated those of the first study, indicating that the physicochemical characteristics of nanoparticles intended for controlled drug release is very important because these characteristics can govern the application of the formulation and can be used to predict its behavior in the biological medium. The size and surface charge are also important parameters in a nanostructured system because these characteristics interfere directly in biological processes, such as the transport across biological membranes and the recognition by M/Ms and biodistribution.

Another important factor that must be taken into account in the design strategies used to improve AIDS treatment is the brain delivery system of antiretroviral drugs. Because of the restricted entry of anti-HIV drugs, the brain is thought to form a viral sanctuary, and the treatment and control of HIV within this reservoir must be primordial. Nanoparticles can enhance the brain-drug delivery by three major pathways, which include the following: i) increasing the local drug gradient at the Blood Brain Barrier (BBB) by passive targeting, ii) allowing drug-trafficking by non-specific or receptor-mediated endocytosis and iii) blocking drug efflux transporters at the BBB (Wong et al., 2010). Consequently, the use of nanocarriers should help to achieve higher concentrations of encapsulated drugs and also allow their prolonged residence in the CNS.

One of the most used polymers for the development of nanoparticles intended for brain delivery is poly-(butylcyanoacrylate) (PBCA) (Koziara et al., 2006). Studies have shown that the surface modification of PBCA nanoparticles using other polymers or surfactant agents, such as polysorbate 80, could increase the transport of particles through the BBB. Polysorbate 80 has been found to increase the translocation of nanoparticles by increasing the particle interaction with the low density lipoprotein (LDL) receptor-mediated endocytic pathway in brain endothelial cells and by inhibition the efflux function of P-gp (Goppert & Muller, 2005).

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 377

proved the suitability of this system to dissolve lipophilic anti-HIV drugs and sustain the drug release; in addition, these studies have shown the feasibility of scaling up SLN production. Cationic SLN were found to be beneficial to the entrapment efficiency of saquinavir. SLN were fabricated via a microemulsion method and stabilized by polysorbate 80; in addition, the lipid phase contained cationic stearylamine, dioctadecyldimethyl ammonium bromide, non-ionic Compritol 888 ATO and cacao butter. The in vitro drug release assay suggested that the carriers could sustain drug delivery without an apparent initial burst (Kuo & Chen, 2009). Aji Alex et al. (2010) investigated the use of SLNs to target intestinal lymphatic vessels. Lopinavir, a poor orally available anti-HIV, was successfully encapsulated in glyceryl behenate-based SLNs produced via a hot homogenization process followed by ultrasonication. In vitro release studies showed that SLNs presented a low release profile; the intestinal lymphatic transport study showed an increase in the cumulative percentage dose of lopinavir secreted into the lymph. These results significantly enhanced the percentage of lopinavir bioavailability. SLNs have been obtained using large-scale production methods, and the study of Shegokara et al. 2010 showed promising results, in which the scaling up of the stavudine production for intravenous injection was possible. The SLNs were produced by the high-pressure homogenization of the stavudine lipid melt, dispersed in a hot surfactant solution (preemulsion). For the investigated formulation, the homogenization system seemed to be rather

**5. Challenges involving clinical trials of antiretroviral drug delivery systems**  Despite the current increase of published original studies on nanotechnology-based antiretroviral drug delivery systems with promising strategies and pre-clinical results, these studies generally have not extended to the clinical studies and, consequently, patients have not received the benefits. Clinical trials are the best way to confirm the efficacy of new medicines; however, this type of study also utilizes placebos, which present serious ethical challenges. The placebo-group has been disapproved in cases of AIDS research because the patient that does not receive the effective regimen can suffer serious consequences in the absence of AIDS therapy. Studies using placebos have been considered unethical in the case where an efficient treatment is known (Scheffer, 2000). Since the Declaration of Helsinki (World Medical Association, 2008), a document with guidelines of ethical principles for the medical community about human experimentation, researchers worldwide must protect the life, health, privacy, and dignity of the human subject, although those principles may contradict many economic and political interests. Thus, a discussion about human experimentation and the investigation of new in vitro models in cells and animals are also extremely important to circumvent the problems with clinical trials of new antiretroviral

The development of systems for drug delivery will not only benefit the therapy of AIDS and other viral diseases but also accelerate the development of systems for bacterial diseases, fungi and mycobacteria. For this, new challenges for the future of drug delivery systems are the feasibility of scaling-up processes to bring to the market quickly innovative therapeutic and the possibility of obtaining multifunctional systems that will be able to fulfill the

robust, producing very similar SLN sizes.

drug delivery systems.

different biological and therapeutic requirements.

**6. Conclusion** 

Kuo and Chen (2006) showed that methylmethacrylate-sulfopropylmethacrylate (MMSPM) nanoparticles were able to significantly increase the BBB permeability of zidovudine and lamivudine by 100%, using blood-brain-microvascular endothelial cells model. In the same study, PBCA nanoparticles increased the BBB permeability of zidovudine 8- to 20-fold and lamivudine 10- to 18-fold. The authors also demonstrated that the drug permeability increased with the decrease in particle size of the two polymeric carriers. Furthermore, these authors observed an increase in the BBB permeability (in vitro) of stavudine-, delaviridineand saquinavir-loaded PBCA and MMSPM nanoparticles coated with polysorbate 80 and solid lipid nanoparticles; in addition, a higher drug permeability was obtained with smaller particles (Kuo & Fu, 2007).

The transferrin receptors present in the luminal membrane of brain endothelial cells have been used as preferential targets for enhanced antiretroviral drug delivery to the CNS by means of nanoparticulate systems (Kreuter, 2001). PEGylated albumin nanoparticles encapsulating zidovudine were prepared, and its surface was modified by anchoring transferrin as a ligand for brain targeting. A significant enhancement of brain localization of zidovudine was observed when it was delivered by transferrin-anchored PEGylated albumin nanoparticles compared to unmodified nanoparticles (Mishra et al., 2006).

Recently, the properties of cell-penetrating peptides have been explored to further enhance the cellular permeability of drug carrier systems. In this approach, certain proteins or peptides can be tethered to the hydrophilic drug of interest, and together, the construct possesses the ability to translocate across the plasma membrane and to deliver the payload intracellularly (Jeang et al., 1999). The Tat peptide, the most frequently used cell-penetrating peptide, is derived from the transcriptional activator protein encoded by HIV-1 (Torchilin, 2008). Thus, nanoparticles containing Tat are promising systems for transport across the BBB and entry into the brain. Therefore, Rao and co-workers (2008) hypothesized that anti-HIV drugs loaded in nanoparticles could bypass the efflux action of P-gp and that Tat conjugation would enhance their transport across the BBB, thereby enhancing the CNS bioavailability of anti-HIV drugs. In their study, ritonavir-loaded PLA nanoparticles conjugated with the Tat peptide were developed, and it was demonstrated to enhance and sustained brain delivery of the system without influencing the integrity of the BBB; these data suggested that the transport occurred through transcytosis across the endothelium of the brain vasculature. At two weeks post administration, the brain ritonavir level after administration of the conjugated nanoparticles was 800-fold higher than that with the drug delivered in solution. It was concluded that Tat-conjugated nanoparticles enhanced the ritonavir CNS bioavailability and maintained therapeutic drug levels in the brain for an effectively sustained period for reducing the viral load in the CNS, which acts as a reservoir for the replicating HIV-1 virus.

## **4. Solid Lipid Nanoparticles (SLN)**

In the last decade of the last century, SLN have gained considerable interest as novel particulate drug delivery systems. SLN are solid, particulate carriers that are nano-sized and composed of biodegradable/biocompatible lipids, suitable for the incorporation of lipophilic and hydrophilic drugs in the lipid matrix in high concentrations. SLNs can be prepared from fatty acids and the stabilization of dispersions with emulsifiers and co-emulsifiers, such as polysorbates, poloxamers, fatty acid co-esters, lecithin and bile salts (Gupta & Jain, 2010). Although few reports about the anti-HIV drug SLN have been published, some studies have proved the suitability of this system to dissolve lipophilic anti-HIV drugs and sustain the drug release; in addition, these studies have shown the feasibility of scaling up SLN production. Cationic SLN were found to be beneficial to the entrapment efficiency of saquinavir. SLN were fabricated via a microemulsion method and stabilized by polysorbate 80; in addition, the lipid phase contained cationic stearylamine, dioctadecyldimethyl ammonium bromide, non-ionic Compritol 888 ATO and cacao butter. The in vitro drug release assay suggested that the carriers could sustain drug delivery without an apparent initial burst (Kuo & Chen, 2009). Aji Alex et al. (2010) investigated the use of SLNs to target intestinal lymphatic vessels. Lopinavir, a poor orally available anti-HIV, was successfully encapsulated in glyceryl behenate-based SLNs produced via a hot homogenization process followed by ultrasonication. In vitro release studies showed that SLNs presented a low release profile; the intestinal lymphatic transport study showed an increase in the cumulative percentage dose of lopinavir secreted into the lymph. These results significantly enhanced the percentage of lopinavir bioavailability. SLNs have been obtained using large-scale production methods, and the study of Shegokara et al. 2010 showed promising results, in which the scaling up of the stavudine production for intravenous injection was possible. The SLNs were produced by the high-pressure homogenization of the stavudine lipid melt, dispersed in a hot surfactant solution (preemulsion). For the investigated formulation, the homogenization system seemed to be rather robust, producing very similar SLN sizes.

## **5. Challenges involving clinical trials of antiretroviral drug delivery systems**

Despite the current increase of published original studies on nanotechnology-based antiretroviral drug delivery systems with promising strategies and pre-clinical results, these studies generally have not extended to the clinical studies and, consequently, patients have not received the benefits. Clinical trials are the best way to confirm the efficacy of new medicines; however, this type of study also utilizes placebos, which present serious ethical challenges. The placebo-group has been disapproved in cases of AIDS research because the patient that does not receive the effective regimen can suffer serious consequences in the absence of AIDS therapy. Studies using placebos have been considered unethical in the case where an efficient treatment is known (Scheffer, 2000). Since the Declaration of Helsinki (World Medical Association, 2008), a document with guidelines of ethical principles for the medical community about human experimentation, researchers worldwide must protect the life, health, privacy, and dignity of the human subject, although those principles may contradict many economic and political interests. Thus, a discussion about human experimentation and the investigation of new in vitro models in cells and animals are also extremely important to circumvent the problems with clinical trials of new antiretroviral drug delivery systems.

## **6. Conclusion**

376 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Kuo and Chen (2006) showed that methylmethacrylate-sulfopropylmethacrylate (MMSPM) nanoparticles were able to significantly increase the BBB permeability of zidovudine and lamivudine by 100%, using blood-brain-microvascular endothelial cells model. In the same study, PBCA nanoparticles increased the BBB permeability of zidovudine 8- to 20-fold and lamivudine 10- to 18-fold. The authors also demonstrated that the drug permeability increased with the decrease in particle size of the two polymeric carriers. Furthermore, these authors observed an increase in the BBB permeability (in vitro) of stavudine-, delaviridineand saquinavir-loaded PBCA and MMSPM nanoparticles coated with polysorbate 80 and solid lipid nanoparticles; in addition, a higher drug permeability was obtained with smaller

The transferrin receptors present in the luminal membrane of brain endothelial cells have been used as preferential targets for enhanced antiretroviral drug delivery to the CNS by means of nanoparticulate systems (Kreuter, 2001). PEGylated albumin nanoparticles encapsulating zidovudine were prepared, and its surface was modified by anchoring transferrin as a ligand for brain targeting. A significant enhancement of brain localization of zidovudine was observed when it was delivered by transferrin-anchored PEGylated

Recently, the properties of cell-penetrating peptides have been explored to further enhance the cellular permeability of drug carrier systems. In this approach, certain proteins or peptides can be tethered to the hydrophilic drug of interest, and together, the construct possesses the ability to translocate across the plasma membrane and to deliver the payload intracellularly (Jeang et al., 1999). The Tat peptide, the most frequently used cell-penetrating peptide, is derived from the transcriptional activator protein encoded by HIV-1 (Torchilin, 2008). Thus, nanoparticles containing Tat are promising systems for transport across the BBB and entry into the brain. Therefore, Rao and co-workers (2008) hypothesized that anti-HIV drugs loaded in nanoparticles could bypass the efflux action of P-gp and that Tat conjugation would enhance their transport across the BBB, thereby enhancing the CNS bioavailability of anti-HIV drugs. In their study, ritonavir-loaded PLA nanoparticles conjugated with the Tat peptide were developed, and it was demonstrated to enhance and sustained brain delivery of the system without influencing the integrity of the BBB; these data suggested that the transport occurred through transcytosis across the endothelium of the brain vasculature. At two weeks post administration, the brain ritonavir level after administration of the conjugated nanoparticles was 800-fold higher than that with the drug delivered in solution. It was concluded that Tat-conjugated nanoparticles enhanced the ritonavir CNS bioavailability and maintained therapeutic drug levels in the brain for an effectively sustained period for reducing the viral load in the CNS, which acts as a reservoir

In the last decade of the last century, SLN have gained considerable interest as novel particulate drug delivery systems. SLN are solid, particulate carriers that are nano-sized and composed of biodegradable/biocompatible lipids, suitable for the incorporation of lipophilic and hydrophilic drugs in the lipid matrix in high concentrations. SLNs can be prepared from fatty acids and the stabilization of dispersions with emulsifiers and co-emulsifiers, such as polysorbates, poloxamers, fatty acid co-esters, lecithin and bile salts (Gupta & Jain, 2010). Although few reports about the anti-HIV drug SLN have been published, some studies have

albumin nanoparticles compared to unmodified nanoparticles (Mishra et al., 2006).

particles (Kuo & Fu, 2007).

for the replicating HIV-1 virus.

**4. Solid Lipid Nanoparticles (SLN)** 

The development of systems for drug delivery will not only benefit the therapy of AIDS and other viral diseases but also accelerate the development of systems for bacterial diseases, fungi and mycobacteria. For this, new challenges for the future of drug delivery systems are the feasibility of scaling-up processes to bring to the market quickly innovative therapeutic and the possibility of obtaining multifunctional systems that will be able to fulfill the different biological and therapeutic requirements.

Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 379

Carvalho F. C., Sarmento V. H. V., Chiavacci L. A., Barbi M. S., Gremião M. P. D. (2009).

Chiappetta D.A., Alvarez-Lorenzo C., Rey-Rico A., Taboada P, Concheiro A., Sosnik A.

Clayton R., Ohagen A., Nicol F., Del Vecchio A.M., Jonckers T.H., Goethals O., Van Loock

Desormeaux A., Bergeron M.G. (1998). Liposomes as drug delivery system: a strategic

Désormeaux A., Harvie P., Perron S., Makabi-Panzu B., Beauchamp D., Tremblay M., Poulin

Destache C.J., Belgum T, Goede M, Shibata A., Belshan M.A. (2010). Antiretroviral release

Destache C.J., Belgum T., Christensen K., Shibata A., Sharma A., Dash A. (2009).

Gagné J.F., Désormeaux A., Perron S., Tremblay W.J., Bergeron M.G. (2002). Targeted

Ganser A., Greher. J, Völkers B., Staszewski S., Hoelzer D. (1989). Inhibitory effect of

Garg M., Asthana A., Agashe H.B., Agrawal G.P., Jain N.K. (2006). Stavudine-loaded

Biochimica et Biophysica Acta – Biomembranes, Vol. 1558, pp. 198-210.

with AIDS. Expert Hematology, Vol. 17, pp. 321-5.

zidovudine. Journal of Pharmaceutical Sciences, Vol. 99, pp. 2367-74.

Pharmaceutics and Biopharmaceutics, Vol. 76, pp. 24–37,

liposomes. Antiviral Research, Vol. 84, pp. 142-149.

DOI:10.1002/jps

1–15.

DOI:10.1016/j.ejpb.2010.05.007

DOI:10.1016/j.antiviral.2009.08.003

DOI:10.3109/10611869808997877

Vol. 8, pp. 1545-1153.

DOI: 10.1093/jac/dkq318

Chemotherapy, Vol. 65, pp. 2183-7.

Diseases, Vol. 9, pp.198-206. DOI: 10.1186/1471-2334-9-198

DOI: 10.1016/S0005-2736(01)00432-1

DOI: 10.1211/jpp.58.5.0005

Development and in vitro evaluation of surfactant systems for controlled release of

(2010) N-alkylation of poloxamines modulates micellar assembly and encapsulation and release of the antiretroviral efavirenz. European Journal of

M., Michiels L., Grigsby J., Xu Z., Zhang Y.P., Gutshall L.L., Cunningham M., Jiang H., Bola S., Sarisky R.T., Hertogs K. (2009). Sustained and specific in vitro inhibition of HIV-1 replication by a protease inhibitor encapsulated in gp120-targeted

approach for the treatment of HIV infection. Journal of Drug Targeting, Vol. 6, pp.

L., Bergeron M.G. (1994). Antiviral efficacy, intracellular uptake and pharmacokinetics of free and liposome-encapsulated 2', 3'-dideoxyinosine. AIDS,

from poly(DL-lactide-co-glycolide) nanoparticles in mice. Journal of Antimicrobial

Combination antiretroviral drugs in PLGA nanoparticle for HIV-1. BMC Infectious

delivery of indinavir to HIV-1 primary reservoirs with immunoliposomes.

azidothymidine, 2'-3'-dideoxyadenosine, and 2'-3'-dideoxycytidine on in vitro growth of hematopoietic progenitor cells from normal persons and from patients

mannosylated liposomes: in-vitro anti-HIV-I activity, tissue distribution and pharmacokinetics. Journal of Pharmacy and Pharmacology, Vol. 58, pp. 605–616.

## **7. Acknowledgment**

The authors would like to thank the FAPESP (Fundo de Apoio a Pesquisa do Estado de São Paulo) and the CNPQ (Conselho Nacional de Desenvolvimento Científico e Tecnológico) from Brazil for financial support.

## **8. References**


DOI: 10.1211/jpp/62.04.0004

Carvalho F. C., Bruschi M. L., Evangelista R. C., Gremião M. P. D. (2010b) Mucoadhesive drug delivery systems. Brazilian Journal of Pharmaceutical Sciences, Vol. 46, pp. 1- 17.

DOI: 10.1590/S1984-82502010000100002

The authors would like to thank the FAPESP (Fundo de Apoio a Pesquisa do Estado de São Paulo) and the CNPQ (Conselho Nacional de Desenvolvimento Científico e Tecnológico)

Aji Alex M.R., Chacko A. J., Jose S., Souto E. B. (2011). Lopinavir loaded solid lipid

Alexaki A., Liu Y, Wigdahl B. (2008). Cellular reservoirs of HIV-1 and their role in viral

Allémann E., Gurny R., Doelker E. (1993). Drug loaded nanoparticles preparation methods

Bender A.R., Schäfer V., Steffan A.M., Royer C., Kreuter J., Rübsamen-Waigmann H., Von

Bender A.R., von Briesen H., Kreuter J., Duncan B., Rübsamen-Waigmann H. (1996).

Betageri G.V., Black S.D., Szebeni J., Wahl L.M., Weinstein J.N. (1993a). Fc-receptor-

Betageri G.V., Jenkins S.A., Ravis W.R. (1993b). Drug delivery using antibody-liposome conjugates. Drug Development and Industrial Pharmacy, Vol. 19, pp. 2109–2116.

Carvalho F. C., Bruschi M. L., Evangelista R. C., Gremião M. P. D. (2010b) Mucoadhesive

 http://www.pharm.monash.edu.au/research/mips/d4/liquidcrystals.html Carvalho F. C., Barbi M. S., Sarmento V. H. V., Chiavacci L. A., Netto F. M., Gremião M. P.

Antimicrobial Agents and Chemotherapy, Vol. 40, pp. 1467–1471.

Pharmaceutical Sciences, Vol. 18, n. 42, 11-18.

Biopharmaceutics, Vol. 39, pp. 173-91.

persistence. Current HIV Research, Vol. 6, pp. 388-400.

nanoparticles. Research Virology, Vol. 145, pp. 215–20.

nanoparticles (SLN) for intestinal lymphatic targeting. European Journal of

and drug targeting issues. European Journal of Pharmaceutics and

Briesen H. (1994). Inhibition of HIV in vitro by antiviral drug-targeting using

Efficiency of nanoparticles as a carrier system for antiviral agents in human immunodeficiency virus-infected human monocytes/macrophages in vitro.

mediated targeting of antibody-bearing liposomes containing dideoxycytidine triphosphate to human monocyte/macrophages. Journal of Pharmacy and

D. (2010a) Surfactant systems for nasal zidovudine delivery: structural, rheological and mucoadhesive properties. Journal of Pharmacy and Pharmacology, Vol. 62, pp.

drug delivery systems. Brazilian Journal of Pharmaceutical Sciences, Vol. 46, pp. 1-

**7. Acknowledgment** 

**8. References** 

from Brazil for financial support.

DOI:10.1016/j.ejps.2010.10.002

DOI:10.2174/157016208785861195

DOI: 10.1016/S0923-2516(07)80025-2

Pharmacology, Vol. 45, pp. 48–53.

Boyd B., (2010) Liquid crystals as drug delivery systems,

DOI:10.3109/03639049309069345

DOI: 10.1211/jpp/62.04.0004

DOI: 10.1590/S1984-82502010000100002

430–439.

17.


DOI:10.3109/10611869808997877


Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 381

Jin Y., Ai P., Xina R., Chen D. (2008) Morphological transformation of self-assembled

Jin Y., Xinga L., Tian Y., Li M., Gao C., Du L., Dong J., Chen H. (2009) Self-assembled drug

Kaparissides C., Alexandridou S., Kotti K., Chaitidou S. (2006). Recent advances in novel

Available from <http://www.azonano.com/Details.asp?ArticleID=1538>

Katragadda A., Bridgman R., Betageri G. (2000). Effect of liposome composition and

Kaur C.D., Nahar M., Jain N.K. (2008). Lymphatic targeting of zidovudine using surfaceengineered liposomes. Journal of Drug Targeting, Vol. 16, pp. 798–805.

Khalil N.M., Carraro E., Cótica L.F., Mainardes R.M. (2011). Potential of polymeric

Koziara J.M., Lockman P.R., Allen D.D., Mumper R.J. (2006). The blood-brain barrier and

Kreuter J. (2001). Nanoparticulate systems for brain delivery of drugs. Advanced Drug

Kuo Y. C. Chen H. H. (2009). Entrapment and release of saquinavir using novel cationic

Kuo Y.C. Chen H.H. (2006). Effect of nanoparticulate polybutylcyanoacrylate and

Kuo Y.C. Su FL. (2007). Transport of stavudine, delavirdine, and saquinavir across the blood–

Cellular Molecular Biology Letters, Vol. 5, pp. 483–493.

DOI:10.1016/j.jcis.2008.07.014

DOI:10.1016/j.ijpharm.2009.07.024

DOI: 10.1080/10611860802475688

DOI: 10.1517/17425247.2011.543673

DOI: 10.1016/S0169-409X(00)00122-8

DOI: 10.1016/j.ijpharm.2008.08.050

DOI: 10.1016/j.ijpharm.2006.07.044

Delivery Reviews, Vol. 47, pp. 65–81.

Pharmaceutics, Vol. 327, pp. 160–69.

Vol. 8, pp. 95-112.

DOI: 10.1166/jnn.2006.441

35.

213,

152.

381, pp. 40–48.

DOI: 10.2240/azojono0111

nanostructures prepared from cholesteryl acyl didanosine and the optimal formulation of nanoparticulate systems: Effects of solvents, acyl chain length and poloxamer 188. Journal of Colloid and Interface Science, Vol. 326, pp. 275–282.

delivery systems. Part 4. In vitro/in vivo studies of the self-assemblies of cholesteryl-phosphonyl zidovudine. International Journal of Pharmaceutics, Vol.

drug delivery systems, In: Journal of nanotechnology online, November 25th, 2010,

cholesterol on the cellular uptake of stavudine by human monocyte/macrophages.

nanoparticles in AIDS treatment and prevention. Expert Opinion on Drug Delivery,

brain drug delivery. Journal of Nanoscience and Nanotechnology, Vol. 6, p. 2712-

solid lipid nanoparticles. International Journal of Pharmaceutics, Vol. 365, pp. 206–

methylmethacrylate-sulfopropylmethacrylate on the permeability of zidovudine and lamivudine across the in vitro blood-brain barrier. International Journal of

brain barrier by polybutylcyanoacrylate, methylmethacrylatesulfopropylmethacrylate, and solid lipid nanoparticles. International Journal of Pharmaceutics, Vol. 340, pp. 143–


DOI: 10.1016/j.antiviral.2009.08.003


DOI:10.1016/j.ijpharm.2009.07.024

380 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Garg M., Dutta T., Jain N.K. (2007). Reduced hepatic toxicity, enhanced cellular uptake and

Geocze L., Mucci S., Marco M. A. D., Nogueira-Martins L. A., Citero V. A. (2010). Qualidade

Goppert T.M., Muller R.H. (2005). Polysorbate-stabilized solid lipid nanoparticles as

Gupta U., Jain N. K. (2010) Non-polymeric nano-carriers in HIV/AIDS drug delivery and

Harvie P., Desormeaux A., Bergeron M.C., Tremblay M., Beauchamp D., Poulin L., Bergeron

Harvie P., Desormeaux A., Gagne N., Tremblay M., Poulin L., Beauchamp D., Bergeron

Hawkins T. (2010). Understanding and managing the adverse effects of antiretroviral

Hillaireau H., Le Doan T., Appel M., Couvreur P. (2006). Hybrid polymer nanocapsules

Jain S., Tiwary A.K, Jain N.K. (2006). Sustained and targeted delivery of an anti-HIV agent

Jain S., Tiwary A.K., Jain N.K. (2008) PEGylated elastic liposomal formulation for lymphatic targeting of zidovudine. Current Drug Delivery, Vol. 5, pp. 275–281.

Jeang K.T., Xiao H., Rich E.A. (1999). Multifaceted activities of the HIV-1 transactivator of transcription, Tat. The Journal of Biological Chemistry, Vol. 274, pp. 28837–40.

Jin Y., Qia N., Tong L., Chen D. (2010). Self-assembled drug delivery systems. Part 5: Self-

assemblies of a bolaamphiphilic prodrug containing dual zidovudine. International

targeting. Advanced Drug Delivery Reviews, Vol. 62, pp. 478–490,

Journal of Pharmaceutics and Biopharmaceutics, Vol. 67, pp. 76–85.

DOI:10.1016/j.ejpb.2006.12.019

DOI: 10.1080/10611860500071292

DOI: 10.1016/j.addr.2009.11.018

DOI:10.1016/j.antiviral.2009.10.016

DOI:10.1016/j.jconrel.2006.09.016

DOI: 10.1016/j.antiviral.2009.08.003

DOI:10.2174/156720108785915078

DOI:10.1016/j.ijpharm.2009.11.018

DOI: 10.1074/jbc.274.41.28837

Vol.3, pp. 157–166.

Saúde Pública, Vol. 44, pp. 743-9, DOI: 10.1590/S0034-89102010000400019

dideoxyinosine. AIDS, Vol.9, pp. 701–707.

of Controlled Release, Vol. 116, pp. 346–52.

Journal of Pharmaceutics, Vol. 386, pp. 268–274,

therapy. Antiviral Research, Vol. 85, pp. 201–209,

altered pharmacokinetics of stavudine loaded galactosylated liposomes. European

de vida e adesão ao tratamento anti-retroviral de pacientes portadores de HIV. Rev.

colloidal carriers for intravenous targeting of drugs to the brain: comparison of plasma protein adsorption patterns. Journal of Drug Targeting, Vol.13, pp. 179–87.

M.G. (1996). Comparative pharmacokinetics, distributions in tissue and interactions with blood proteins of conventional and sterically stabilized liposomes containing 2, 3-dideoxyinosine. Antimicrobial Agents and Chemotherapy, Vol.40, pp. 225–229.

M.G. (1995). Lymphoid tissues targeting of liposome-encapsulated 2, 3-

enhance in vitro delivery of azidothymidine-triphosphate to macrophages. Journal

using elastic liposomal formulation: mechanism of action. Current Drug Delivery,


DOI: 10.1517/17425247.2011.543673

Koziara J.M., Lockman P.R., Allen D.D., Mumper R.J. (2006). The blood-brain barrier and brain drug delivery. Journal of Nanoscience and Nanotechnology, Vol. 6, p. 2712- 35.

DOI: 10.1166/jnn.2006.441


DOI: 10.1016/j.ijpharm.2008.08.050


Exploring the Nanotechnology-Based Drug Delivery Systems for AIDS Treatment 383

Phillips N.C., Tsoukas C. (1992). Liposomal encapsulation of azidothymidine results in

Rao K.S., Ghorpade A., Labhasetwar V. (2009). Targeting anti-HIV drugs to the CNS.

Richman D.D., Margolis D.M., Delaney M., Greene W.C., Hazuda D, Pomerantz R.J. (2009). The challenge of finding a cure for HIV infection. Science, Vol. 323, pp. 1304–07.

Saksena N.K., Haddad D.N. (2003). Viral reservoirs an impediment to HAART: new

Schäfer V., von Briesen H., Andreesen R., Steffan, A.M., Royer, C., Tröster, S., Kreuter, J.,

Scheffer M. (2000) Tecnologia, AIDS e ética em pesquisa. Interface - Comunicação, Saúde,

Shah L.K., Amiji M.M. (2006). Intracellular delivery of saquinavir in biodegradable

Sharma A., Sharma U.S. (1997). Liposomes in drug delivery: progress and limitations.

Sharma P., Garg S. (2010). Pure drug and polymer based nanotechnologies for the

Shegokara R., Singha K. K., Müller R. H. (2010). Production & stability of stavudine solid

Sosnik A., Chiappetta D. A., Carcaboso A. M. (2009). Drug delivery systems in HIV

Stolnik S., Illum L., Davis S.S. (1995). Long circulating microparticulate drug carriers.

Szebeni J., Wahl S.M., Betageri G.V., Wahl L.M., Gartner S, Popovic, M, Parker R.J., Black

strategies to eliminate HIV-1. Current Drug Targets Infection Disorders, Vol. 3, pp.

Rübsamen-Waigmann, H. (1992). Phagocytosis of nanoparticles by human immunodeficiency virus (HIV)-infected macrophages: a possibility for antiviral

polymeric nanoparticles for HIV/AIDS. Pharmaceutical Research, Vol. 23, p. 2638-

improved solubility, stability, bioavailability and targeting of anti-HIV drugs.

lipid nanoparticles — From lab to industrial scale, International Journal of

pharmacotherapy: What has been done and the challenges standing ahead. Journal

C.D., Weinstein J.N. (1990). Inhibition of HIV-1 in monocyte/ macrophage cultures

immunodeficiency syndrome. Blood, Vol. 79, pp. 1137–1143.

drug targeting. Pharmaceutical Research, Vol. 9, pp. 541–46.

International Journal Pharmaceutics, Vol. 154, pp. 123–140.

Advanced Drug Delivery Reviews, Vol. 62, pp. 491–502.

Advanced Drug Delivery Reviews, Vol. 16, pp. 195–214.

Expert Opinion on Drug Delivery, Vol. 6, pp. 771-84.

DOI:10.1517/17425240903081705

DOI: 10.1126/science.1165706

DOI: 10.2174/1568005033481187

DOI: 10.1023/A:1015852732512

DOI: 10.1007/s11095-006-9101-7

DOI:10.1016/S0378-5173(97)00135-X

DOI: 10.1016/j.addr.2009.11.019

Pharmaceutics, In press, DOI: 10.1016/j.ijpharm.2010.08.014

DOI: 10.1016/j.jconrel.2009.05.007

DOI: 10.1016/0169-409X(95)00025-3

of Controlled Release, Vol. 138, pp. 2–15,

Educação, Vol. 6, pp. 173-176.

179–206.

45.

decreased hematopoietic toxicity and enhanced activity against murine acquired

DOI: 10.1517/17425247.2011.543673


DOI: 10.1016/j.ijpharm.2010.05.020.


DOI: 10.1016/0169-409X(95)00025-3

Phillips N.C., Skamene E., Tsoukas C. (1991). Liposomal encapsulation of 3`-azido-3` deoxythymidine (AZT) results in decreased bone marrow toxicity and enhanced activity against murine AIDS-induced immunosuppression. Journal of Acquired Immune Deficiency Syndrome, Vol. 4, pp. 959–966.


DOI: 10.2174/1568005033481187

382 Understanding HIV/AIDS Management and Care – Pandemic Approaches in the 21st Century

Lalanne M., Paci A., Andrieux K., Dereuddre-Bosquet N., Clayette P., Deroussent A., Ré M.,

Mainardes, R.M.; Khalil, N.M.; Gremião, M.P.D. (2010) Intranasal delivery of zidovudine by

Makabi-Panzu B., Gourde P., Desormeaux A., Bergeron M.G. (1998). Intracellular and serum

Malmsten, M. (2002). Surfactants and polymers in drug delivery. Marcel Dekker, ISBN 978-

Mishra V., Mahor S., Rawat A., Gupta P.N., Dubey P., Khatri K., Vyas S.P. (2006). Targeted

Neves J., Amiji M. M., Bahia M. F., Sarmento B. (2010). Nanotechnology-based systems for

Ojewole E., Mackraj I., Naidooa P., Govender, T. (2008) Exploring the use of novel drug

Oppenheim R.C. (1981). Solid colloidal drug delivery systems: nanoparticles. International

Owens D.E., Peppas N.A. (2006). Opsonization, biodistribution, and pharmacokinetics of

Phillips N.C., Skamene E., Tsoukas C. (1991). Liposomal encapsulation of 3`-azido-3`-

HIV. Bioorganic Medicinal Chemistry Letters, Vol. 17, pp. 2237–2240. Mainardes R.M., Gremião M.P.D., Brunetti I.L., da Fonseca, A.M., Khalil, N.M. (2009).

Vassal G., Couvreur P., Desmaële D. (2007). Synthesis and biological evaluation of two glycerolipidic prodrugs of didanosine for direct lymphatic delivery against

Zidovudine-loaded PLA and PLA-PEG blend nanoparticles: influence of polymer type on phagocytic uptake by polymorphonuclear cells. Journal of Pharmaceutical

PLA and PLA-PEG blend nanoparticles. International Journal of Pharmaceutics,

stability of liposomal 2′, 3′-dideoxycytidine. Effect of lipid composition. Cellular

brain delivery of AZT via transferrin anchored pegylated albumin nanoparticles.

the treatment and prevention of HIV/AIDS, Advanced Drug Delivery Reviews,

delivery systems for antiretroviral drugs. European Journal of Pharmaceutics

polymeric nanoparticles. International Journal of Pharmaceutics, Vol. 307, pp. 93-

deoxythymidine (AZT) results in decreased bone marrow toxicity and enhanced activity against murine AIDS-induced immunosuppression. Journal of Acquired

DOI: 10.1517/17425247.2011.543673

Sciences, Vol. 98, pp. 257-67.

Molecular Biology, Vol. 44, pp. 277–284.

Journal of Drug Targeting, Vol. 14, pp. 45–53.

and Biopharmaceutics, Vol. 70, pp. 697-710.

Journal of Pharmaceutics, Vol.8, pp. 217-34.

Immune Deficiency Syndrome, Vol. 4, pp. 959–966.

DOI: 10.1002/jps.21406

Vol. 16, pp. 266-271, DOI: 10.1016/j.ijpharm.2010.05.020.

0-8247-0804-7, New York.

DOI: 10.1080/10611860600612953

Vol. 62, pp. 458–477, DOI: 10.1016/j.addr.2009.11.017

DOI 10.1016/j.ejpb.2008.06.020

DOI: 10.1016/0378-5173(81)90100-9

DOI: 10.1016/0169-409X(95)00025-3

102.


DOI: 10.1007/s11095-006-9101-7


by 2', 3'-dideoxycytidine-5'-triphosphate, free and in liposomes. AIDS Research and Humum Retroviruses, Vol. 6, pp. 691–702. DOI: 10.1089/aid.1990.6.691


http://www.wma.net/en/30publications/10policies/b3/index.html.

Zidan A.S., Spink C., Fortunak J., Habib M., Khan M.A. (2010). Near-infrared investigations of novel anti-HIV tenofovir liposmes. The AAPS Journal, Vol. 12, pp. 202-214. DOI: 10.1208/s12248-010-9177-1

Torchilin V.P. (2008). Tat peptide-mediated intracellular delivery of pharmaceutical nanocarriers. Advanced Drug Delivery Reviews, Vol. 60, pp. 548-58.

Ubrich N., Schmidt C., Bodmeier R., Hoffman M., Maincent, P. (2005). Oral evaluation in

Vyas T. K., Shahiwala A., Amiji M. M. (2008). Improved oral bioavailability and brain

Wong H. L., Chattopadhyay N., Wu X. Y., Bendayan R. (2010). Nanotechnology

World Medical Association (2008) Declaration of Helsinki - Ethical Principles for Medical

Zidan A.S., Spink C., Fortunak J., Habib M., Khan M.A. (2010). Near-infrared investigations of novel anti-HIV tenofovir liposmes. The AAPS Journal, Vol. 12, pp. 202-214.

International Journal of Pharmaceutics, Vol. 347, pp. 93–101,

http://www.wma.net/en/30publications/10policies/b3/index.html.

and Humum Retroviruses, Vol. 6, pp. 691–702.

Journal of Pharmaceutics, Vol. 288, p. 169-75.

Drug Delivery Reviews, Vol. 62, pp. 503–517,

Research Involving Human Subjects.

DOI: 10.1089/aid.1990.6.691

DOI: 10.1016/j.addr.2007.10.008

DOI:10.1016/j.ijpharm.2004.09.019

DOIi:10.1016/j.ijpharm.2007.06.016

DOI:10.1016/j.addr.2009.11.020

DOI: 10.1208/s12248-010-9177-1

by 2', 3'-dideoxycytidine-5'-triphosphate, free and in liposomes. AIDS Research

rabbits of cyclosporine-loaded Eudragit RS or RL nanoparticles. International

transport of saquinavir upon administration in novel nanoemulsion formulations.

applications for improved delivery of antiretroviral drugs to the brain. Advanced

## *Edited by Fyson Hanania Kasenga*

Like any other book on the subject of HIV/AIDS, this book is not a substitute or exhausting the subject in question. It aims at complementing what is already in circulation and adds value to clarification of certain concepts to create more room for reasoning and being part of the solution to this global pandemic. It is further expected to complement a wide range of studies done on this subject, and provide a platform for the more updated information on this subject. It is the hope of the authors that the book will provide the readers with more knowledge and skills to do more to reduce HIV transmission and improve the quality of life of those that are infected or affected by HIV/AIDS.

Understanding HIV/AIDS Management and Care - Pandemic Approaches in the 21st Century

Understanding HIV/AIDS

Management and Care

Pandemic Approaches in the 21st Century

*Edited by Fyson Hanania Kasenga*

Photo by satori13 / iStock