*4.2.3.2. Commercial immunoglobulin formulations*

What factors could have led to the heterogeneity in study results? Recognizable differences in the methods used for serum preparations and their administration, as well as the lack of appropriate experimental controls probably accounted in part for the studies outcomes. Furthermore, it is important to recognize that immune serum is a polyclonal preparation that includes antibodies with multiple specificities and isotypes. Consequently, polyclonal sera may contain antibodies of different subclasses and functional categories that can affect theoutcome of infection. For example, IgG3 murine monoclonal antibodies protected against *M.tuberculosis* [27] but failed to protect against *Cryptococcus neoformans* [41]. An IgG3 nonprotectivemonoclonal antibody to *C. neoformans*, became protective upon subclass switching to IgG1 [41]. In addition to intrinsic factors associated with the antibody structure, other parameters such as the genetic background of the microbe and the immunocompetence of the

For some microorganisms, such as *Samonella typhimurium* and *C. neoformans,* passive antibody therapy efficacy depends on the mouse strain used [42, 43]. In the same way, some microbial

The animal model used is another important parameter that varies between different experi‐ ments cited in the literature [45]. Timing, the route of infection, the magnitude of the infecting inoculum are some additional variables that could affect antibody protection studies [46]. Despite their variability, the results obtained with serum therapy were valuable, demonstrat‐ ing some beneficial effect of serum on the course of TB in humans, mainly in cases of early or localized TB [45]. Moreover, it was demonstrated that long periods of treatment were necessary

A recent study re-examined the usefulness of immune serum in the context of a therapeutic vaccine against TB [32]. This vaccine, named RUTI, is generated from detoxified *M. tuberculo‐ sis* cell fragments that facilitate a balanced T helper response to a wide range of antigens along with intense antibody production [47]. Local accumulation of specific CD8+ T cells and a strong humoral response after immunization are characteristic features of RUTI that contribute to its protective properties. In that study, immune serum was generated by immunizing mice with RUTI [32]. Severe Combined Immunodeficiency (SCID) mice were inoculated with *M. tuberculosis* and treated with chemotherapy for 3–8 weeks. After chemotherapy they were treated for up to 10 weeks with intraperitoneal injections of the generated immune serum. Mice treated with immune serum from RUTI vaccinated animals showed significant decreases in lung CFU in addition to reduced extent of granulomatous response and abscess formation [47]. These results indicate that protective serum antibodies can be elicited by vaccination, and that

Evidence for the stimulatory role of specific polyclonal antibodies on cellular immunity in experimental mycobacterial infections was reported by de Valliere and colleagues in 2005 [15].

host could alter the outcome of antibody protection experiments.

antibodies may be usefully combined with chemotherapy [32, 47, 48].

strains are more susceptible to the effects of antibodies [44].

60 Tuberculosis - Current Issues in Diagnosis and Management

to achieve a sustained effect [45].

*4.2.2. Polyclonal mouse antibodies*

*4.2.3. Human gammaglobulins*

*4.2.3.1. Specific human polyclonal antibodies*

Human Intravenous Immunoglobulin (IVIG) has been used to treat individuals with immune deficiencies and patients with inflammatory, autoimmune and infectious conditions [49, 50, 51]. Several groups tested the effect of human immunoglobulin preparation on mycobacterial infection. Roy and colleagues showed that treatment of *M. tuberculosis* infected mice with one cycle of IVIG led to the substantially lower bacterial loads in the spleen and lungs following its administration either at early or at late stage of infection 52]. The effect of the administration of a commercial preparation of human immunoglobulin (hIg) in a mouse model of intranasal infection with BCG was evaluated by Olivares and colleagues [33]. This group demonstrated the passage of specific antibodies to saliva and lung lavage following the intranasal or intraperitoneal administration of human hIg to mice. This treatment inhibited BCG coloniza‐ tion of the lungs of treated mice. A similar inhibitory effect was observed after infection of mice with hIg -opsonized BCG [33].

The same formulation was evaluated also in a mouse model of intratracheal infection with *M. tuberculosis*. Animals receiving human hIg intranasally 2h prior to intratracheal challenge demonstrated a significant decrease in lung bacillary load as compared with non-treated animals [29]. When *M. tuberculosis* was pre-incubated with hIg prior to challenge the same effect was observed [29].

The protective effect of the hIg formulation was abolished following pre-incubation with *M. tuberculosis* [29]. These results are suggestive of a potential role for specific human antibodies in the defense against mycobacterial infections.

Taken together, these studies provide support for the potential use of immunoglobulins against *M. tuberculosis*.

#### *4.2.3.3. Human secretory IgA*

Human secretory IgA (hsIgA) is the major class of antibody associated with immune protection of the mucosal surfaces [53]. Colostrum volume is above 102 mL in humans during the first three days after delivery [54]. The high percentage of (hsIgA) in human colostrum [55] strongly suggests its important role in passive immune protection against gastrointestinal and respi‐ ratory infections [56]. In one study performed by Alvarez and colleagues, hsIgA from human colostrum was obtained by anion exchange and gel filtration chromatographic methods, using DEAE Sepharose FF and Superose 6 preparative grade, respectively [57].

hsIgA was administered intranasally to BALB/c mice, and the level of this immunoglobulin in several biological fluids was determined by ELISA. The results showed the presence of this antibody in the saliva of animals that received the hsIgA, at all time intervals studied. In tracheobronchial lavage, hsIgA was detected at 2 and 3 hours after inoculation in animals that received the hsIgA [58]. Similar studies were performed by Falero and colleagues with monoclonal antibodies of IgA and IgG class [59]. Following demostration that hsIgA could be detected in several biological secretions after intranasal administration, the protective effect of this formulation against *M. tuberculosis* challenge was evaluated. Mice challenged with *M. tuberculosis* preincubated with hsIgA showed a statistically significant decrease in the mean number of viable bacteria recovered from the lungs compared to control mice and to the group that received the hsIgA before challenge with *M. tuberculosis*. Moreover, an increased level of iNOS production was also reported (Alvarez et al., mannuscript in preparation). Consistently with this result, a better organization of granulomatous areas with foci of lymphocytes and abundant activated macrophages were observed in the lungs of mice that received *M. tuberculosis* pre-incubated with hsIgA and sacrificed at 2 months postchallenge. Untreated animals, however, showed an increased area of bronchiectasis and atelectasis as well as fibrin deposits, accumulation of activated macrophages and lymphocytes.

to coat mycobacteria before administration to mice. In this study, spleen CFUs was reduced while lung CFUs did not [63]. These results suggest that binding of these antibodies to HBHA

The Role of Antibodies in the Defense Against Tuberculosis

http://dx.doi.org/10.5772/53950

63

The protective efficacy of a monoclonal antibody, TBA63 and IgA anti-Acr administered intranasally before and after the intranasal or aerosol challenge with *M. tuberculosis* was demonstrated in a study by Williams and colleagues [64]. In another series of experiments carried out by López and colleagues, the protective effect of this Mab administered intratra‐ cheally before an intratracheal challenge with virulent mycobacteria was evaluated. At 21 days post-infection, pre-treatment of mice with TBA63 caused a significant decrease in viable bacteria in the lungs compared to control mice or those treated with the Mab against the 38 kDa protein (TBA86) [65]. Consistent with the reduction of viable bacteria following treatment with TBA63, the area of peribronchial inflammation was also statistically smaller in this group

When the lungs of mice were histologically examined, granulomas were better organized in the infected animals that had received TBA63 than in controls or mice treated with TBA86. The reduction of CFU in lungs of the treated group was associated with milder histopathological changes, as indicated by the organization of the granulomas and less pneumonic area. The fact that this Mab promotes granuloma formation in mice infected intratracheally with *M. tuber‐ culosis* strongly suggests the close interaction between antibody mediated immunity and cell-

The 16 kDa protein (Acr antigen) has been defined as a major membrane protein peripherally associated with the membrane [67] carrying epitopes restricted to tubercle bacilli on the basis of B-cell recognition [68, 69]. The Acr antigen is present on the surface of tubercle bacilli and is highly expressed in organisms growing within infected macrophages, allowing it to be potentially targeted by specific antibodies either inside infected cells as well as extracellulary. A novel immunotherapy, combining treatment with anti-IL-4 antibodies, IgA antibody against 16 kDa protein and IFN-γ, showed the potential for passive immunoprophylaxis against TB. In genetically deficient IL-4-/- BALB/c mice, infection in both lungs and spleen was substan‐ tially reduced for up to 8 weeks. Administration of rIL-4 to IL-4-/- mice with increased bacterial

counts to wild-type levels and make mice refractory to protection by IgA/IFN-γ [70].

More recently, Balu and colleagues reported that intranasal administration of a human IgA1 Mab, obtained using a single-chain variable fragment derived from an Ab phage library with high affinity for hspX and the human FcαRI (CD91) IgA receptor together with recombinant mouse IFN-γ significantly inhibited pulmonary infection with *M. tuberculosis* H37Rv in mice transgenic for human CD91 but not in the CD91-negative controls. These results suggested that binding to CD91 was necessary for the IgA-imparted passive protection [71]. When the Mab was incubated with human whole-blood or monocyte cultures it inhibited H37Rv

Inhibition of the infection by the antibody was synergistic with human rIFN-γ in purified human monocytes cultures but not in whole blood cultures [71]. The demonstration of the role of FcαRI (CD91) in human IgA mediated protection contributes to understanding the mecha‐

mediated immunity to induce protection against intracellular pathogens (66).

can impede mycobacterial dissemination.

compared to the control group [65].

infection.

The pneumonic areas were more prominent in the untreated animals than in the groups treated with hsIgA and *M. tuberculosis* pre-incubated with hsIgA (Alvarez et al., manuscript in preparation)

#### *4.2.4. Monoclonal antibodies*

Since the first report on the use of the monoclonal antibody Mab 9d8 against *M. tuberculosis* [27], many similar studies have been reported [40]. This IgG3 monoclonal antibody (Mab) generated against arabinomannan (AM) capsular polysaccharide, increased the survival of intratracheally infected mice when the *M. tuberculosis* Erdman strain was pre-coated with it [27]. In this study, a longer survival associated with an enhanced granulomatous response in the lungs was found as compared to controls receiving an isotype-specific non-related Mab [27]. Another Mab, SMITB14, directed against the AM portion of LAM prolonged the survival of intravenously infected mice associated with reduced lung CFU and prevention of weight loss [60]. In this study, the authors demonstrated that protection was independent of the antibody Fc portion, because the F(ab')2 fragment also conferred a similar protective effect [60]. In another study, mice receiving the Mab 5c11 (an IgM antibody that recognizes other mycobacterial arabinose-containing carbohydrates in addition to AM) intravenously prior to Mannosylated lipoarabinomannan (ManLAM) administration, showed a significant clearance of Man-LAM and redirection of this product to the hepatobiliary system [26]. This study strongly supports an indirect effect of certain antibodies on the course of mycobacterial infection, altering problably the pharmacokinetics of mycobacterial components and contri‐ buting to protection against TB [26].

Heparin Binding Hemagglutinin Adhesin (HBHA) is a surface-exposed glycoprotein involved in the mycobacterial binding to epithelial cells and in mycobacterial dissemination [62]. Monoclonal antibodies 3941E4 (IgG2a) and 4058D2 (IgG3) directed against HBHA were used to coat mycobacteria before administration to mice. In this study, spleen CFUs was reduced while lung CFUs did not [63]. These results suggest that binding of these antibodies to HBHA can impede mycobacterial dissemination.

hsIgA was administered intranasally to BALB/c mice, and the level of this immunoglobulin in several biological fluids was determined by ELISA. The results showed the presence of this antibody in the saliva of animals that received the hsIgA, at all time intervals studied. In tracheobronchial lavage, hsIgA was detected at 2 and 3 hours after inoculation in animals that received the hsIgA [58]. Similar studies were performed by Falero and colleagues with monoclonal antibodies of IgA and IgG class [59]. Following demostration that hsIgA could be detected in several biological secretions after intranasal administration, the protective effect of this formulation against *M. tuberculosis* challenge was evaluated. Mice challenged with *M. tuberculosis* preincubated with hsIgA showed a statistically significant decrease in the mean number of viable bacteria recovered from the lungs compared to control mice and to the group that received the hsIgA before challenge with *M. tuberculosis*. Moreover, an increased level of iNOS production was also reported (Alvarez et al., mannuscript in preparation). Consistently with this result, a better organization of granulomatous areas with foci of lymphocytes and abundant activated macrophages were observed in the lungs of mice that received *M. tuberculosis* pre-incubated with hsIgA and sacrificed at 2 months postchallenge. Untreated animals, however, showed an increased area of bronchiectasis and atelectasis as well as fibrin

The pneumonic areas were more prominent in the untreated animals than in the groups treated with hsIgA and *M. tuberculosis* pre-incubated with hsIgA (Alvarez et al., manuscript in

Since the first report on the use of the monoclonal antibody Mab 9d8 against *M. tuberculosis* [27], many similar studies have been reported [40]. This IgG3 monoclonal antibody (Mab) generated against arabinomannan (AM) capsular polysaccharide, increased the survival of intratracheally infected mice when the *M. tuberculosis* Erdman strain was pre-coated with it [27]. In this study, a longer survival associated with an enhanced granulomatous response in the lungs was found as compared to controls receiving an isotype-specific non-related Mab [27]. Another Mab, SMITB14, directed against the AM portion of LAM prolonged the survival of intravenously infected mice associated with reduced lung CFU and prevention of weight loss [60]. In this study, the authors demonstrated that protection was independent of the antibody Fc portion, because the F(ab')2 fragment also conferred a similar protective effect [60]. In another study, mice receiving the Mab 5c11 (an IgM antibody that recognizes other mycobacterial arabinose-containing carbohydrates in addition to AM) intravenously prior to Mannosylated lipoarabinomannan (ManLAM) administration, showed a significant clearance of Man-LAM and redirection of this product to the hepatobiliary system [26]. This study strongly supports an indirect effect of certain antibodies on the course of mycobacterial infection, altering problably the pharmacokinetics of mycobacterial components and contri‐

Heparin Binding Hemagglutinin Adhesin (HBHA) is a surface-exposed glycoprotein involved in the mycobacterial binding to epithelial cells and in mycobacterial dissemination [62]. Monoclonal antibodies 3941E4 (IgG2a) and 4058D2 (IgG3) directed against HBHA were used

deposits, accumulation of activated macrophages and lymphocytes.

preparation)

*4.2.4. Monoclonal antibodies*

62 Tuberculosis - Current Issues in Diagnosis and Management

buting to protection against TB [26].

The protective efficacy of a monoclonal antibody, TBA63 and IgA anti-Acr administered intranasally before and after the intranasal or aerosol challenge with *M. tuberculosis* was demonstrated in a study by Williams and colleagues [64]. In another series of experiments carried out by López and colleagues, the protective effect of this Mab administered intratra‐ cheally before an intratracheal challenge with virulent mycobacteria was evaluated. At 21 days post-infection, pre-treatment of mice with TBA63 caused a significant decrease in viable bacteria in the lungs compared to control mice or those treated with the Mab against the 38 kDa protein (TBA86) [65]. Consistent with the reduction of viable bacteria following treatment with TBA63, the area of peribronchial inflammation was also statistically smaller in this group compared to the control group [65].

When the lungs of mice were histologically examined, granulomas were better organized in the infected animals that had received TBA63 than in controls or mice treated with TBA86. The reduction of CFU in lungs of the treated group was associated with milder histopathological changes, as indicated by the organization of the granulomas and less pneumonic area. The fact that this Mab promotes granuloma formation in mice infected intratracheally with *M. tuber‐ culosis* strongly suggests the close interaction between antibody mediated immunity and cellmediated immunity to induce protection against intracellular pathogens (66).

The 16 kDa protein (Acr antigen) has been defined as a major membrane protein peripherally associated with the membrane [67] carrying epitopes restricted to tubercle bacilli on the basis of B-cell recognition [68, 69]. The Acr antigen is present on the surface of tubercle bacilli and is highly expressed in organisms growing within infected macrophages, allowing it to be potentially targeted by specific antibodies either inside infected cells as well as extracellulary. A novel immunotherapy, combining treatment with anti-IL-4 antibodies, IgA antibody against 16 kDa protein and IFN-γ, showed the potential for passive immunoprophylaxis against TB. In genetically deficient IL-4-/- BALB/c mice, infection in both lungs and spleen was substan‐ tially reduced for up to 8 weeks. Administration of rIL-4 to IL-4-/- mice with increased bacterial counts to wild-type levels and make mice refractory to protection by IgA/IFN-γ [70].

More recently, Balu and colleagues reported that intranasal administration of a human IgA1 Mab, obtained using a single-chain variable fragment derived from an Ab phage library with high affinity for hspX and the human FcαRI (CD91) IgA receptor together with recombinant mouse IFN-γ significantly inhibited pulmonary infection with *M. tuberculosis* H37Rv in mice transgenic for human CD91 but not in the CD91-negative controls. These results suggested that binding to CD91 was necessary for the IgA-imparted passive protection [71]. When the Mab was incubated with human whole-blood or monocyte cultures it inhibited H37Rv infection.

Inhibition of the infection by the antibody was synergistic with human rIFN-γ in purified human monocytes cultures but not in whole blood cultures [71]. The demonstration of the role of FcαRI (CD91) in human IgA mediated protection contributes to understanding the mecha‐ nisms involved as well as for using this knowledge for the future development of new immunotherapies for TB [71].

Together these studies suggest that B cells may have an important role in host defense against

The Role of Antibodies in the Defense Against Tuberculosis

http://dx.doi.org/10.5772/53950

65

The various effects of antibodies demonstrated in the studies analyzed, suggest that different mechanisms of action are involved in the effect of monoclonal and polyclonal antibodies on *M. tuberculosis*. Secretions found on mucosal surfaces contain significant levels of Igs, partic‐ ularly, IgA. IgA has both direct and indirect functional roles for combating infectious agents such as viruses and bacteria that cross the mucosal barrier [72]. Moreover, experimental evidence suggests that IgA associated with the pIgR may neutralize pathogens and antigens intracellularly during their transport from the basolateral to the apical zone of epithelial cells [73,74]. In addition, IgA may interact with Gal-3 (an intracellular binding β-galactosidase lectin), and interfere with the interaction of mycobacteria with the phagosomal membrane,

resulting in the decrease of bacterial survival and replication in the phagosome [75].

Antibodies may be critical, during the extracellular phases of intracellular facultative patho‐ gens. They may act by interfering with adhesion, by neutralizing toxins and by activating complement. Moreover, antibodies may be able to penetrate recently infected cells, bind internalised pathogens, and enhance antigen processing (76). Antibodies may also play a crucial role in modulating the immune response by activating faster secretion of selected cytokines that in turn, contribute to more efficient and rapid Th1 response [76,77], increasing the efficacy of co-stimulatory signals, enhancing Antibody Dependent Cellular Cytotoxicity (ADCC) and the homing of immune cells to the lungs after the respiratory infection [10,78, 79,

Examples of relevant potential action mechanisms of antibodies against *M. tuberculosis* were

Future applications of antibody formulations for the control of TB may include several

Antibody based therapy could potentially be useful in several scenarios. They could be used to shorten the standard treatment period of patients with uncomplicated TB when coupled with standard chemotherapy. However, they would be particularly important in the treatment of patients infected with Multidrug Resistant (MDR) and Extensively Drug Resistant (XDR)

*M. tuberculosis*.

80, 81, 82, 83].

**6.1. Treatment**

discussed by Glatman-Freedman [40].

**6. Potential uses of antibodies against TB**

strains, in combination with the standard treatment.

possibilities including treatment, prevention and diagnosis.

**5. Mechanisms of action**

#### *4.2.5. Transgenic mice*

Mouse models with deficiency in antibody production can provide useful information for understanding certain roles of antibodies in protection against mycobacterial infections. Rodríguez and colleagues reported that after immunization of IgA deficient (IgA-/-) and wild type mice by the intranasal route with the mycobacterial surface antigen PstS-1, IgA-/- mice were more susceptible to BCG infection than IgA+/+ mice [34]. Cytokine response analysis demonstrated reduction in the IFN-γ and TNF-α production in the lungs of IgA-/- as compared with IgA+/+ mice, suggesting that IgA may play a role in protection against mycobacterial infections in the respiratory tract. Furthermore, these authors demonstrated that immunized pIgR-/- mice were more susceptible to BCG infection than immunized wildtype mice [34].

In an attempt to elucidate whether antibody-mediated immunity has a special role in the defense against TB, different experiments with B cell knockout mice were performed by several authors. In 11016, Vordermeier and colleagues developed an infection model of TB in μ chain knockout Ig- mice. Organs from *M. tuberculosis* infected IgG- mice had three to eight fold elevated counts of viable bacilli compared with those from normal mice. This result suggested that B cells play a role in the containment of murine tuberculous infection [35]. In another study B cell KO mice and controls were infected by aerosol with *M. tuberculosis*. They were subse‐ quently given chemotherapy to destroy remaining bacilli and then re-challenged by aerosol exposure. There were not differences in the ability of animals to control this second infection, indicating that, in this low dose pulmonary infection model, any local production of antibodies neither impeded nor enhanced the expression of specific acquired resistance [36].

In another series of experiments the role of B cells was evaluated during early immune responses to infection with a clinical strain of *M. tuberculosis* (CDC 1561). In this study, despite comparable bacillary loads in the lungs, B cell KO mice had a less severe pulmonary granuloma formation and delayed dissemination of bacteria from lungs to peripheral organs. Additional analysis of lung cells demonstrated higher numbers of lymphocytes, particular‐ ly CD8+ T cells, macrophages, and neutrophils in wild-type and reconstituted mice as compared with B cell KO mice. These results demonstrate that less severe granuloma formation and delayed dissemination of mycobacteria found in B cell KO mice were dependent on B cells, (not antibodies, at least in this study) and were associated with modification of cellular infiltrate in the lungs [37]. This latter result differs from a study carried out by Maglione and colleagues in which B cell-/- mice demonstrated exacerbated immunopathology corresponding with enhanced pulmonary recruitment of neutrophils following aerosol challenge with *M. tuberculosis* Erdman strain [38]. Infected B cell-/- mice demonstrated increased production of IL-10 in the lungs, while IFN- γ, TNF-α, and IL-10R were not significantly different from those of wild type mice [38].. B cell-/- mice demonstrat‐ ed enhanced susceptibility to aerosol infection of 300 CFU of *M. tuberculosis* with elevated bacterial burden in the lungs but not in the spleen or liver [38].

Together these studies suggest that B cells may have an important role in host defense against *M. tuberculosis*.
