**2. DNA vaccination**

#### **2.1 DNA vaccines: An emerging field**

In 1990, Wolff and collaborators found that bacterial plasmid DNA encoding a reporter gene could result in *in vivo* expression of the encoded protein after simple intramuscular injection without the need for more complex vectors (Wolff et al. 1990). Following Wolff's findings a new era of vaccination started.

Naked-DNA vaccines are for definition vectors based on bacterial plasmids engineered to express the disease-specific antigen using promoter elements active in mammalian cells, without the addition of surrounding chemicals or a viral coat. The main advantages of naked DNA vaccines are safety and production in large amount, as well as stability at different temperatures and, more important, flexibility in design, since multiple antigenic targets or multiple independent cytokines or co-stimulatory sequences can be incorporated into a single DNA vector. They are also likely to be attractive from a health economics perspective: they are relatively easy to manufacture in large quantities in contrast to the complicated processes requested for attenuated virus vaccines, and do not require any special transportation or storage conditions that could hinder their widespread distribution as it happens for live pathogens-based vaccines which need to be distributed and stored in cold conditions. The genes encoding the antigens can be chemically constructed without deriving them from live virulent organisms so avoiding for operators and patients the risks of exposure to dangerous pathogens. They are commonly delivered by a simple intramuscular injection. In mammals the skeletal muscle represents approximately the 30% of the body mass, and muscle fibres are ideal targets for DNA transfection. These are stable and large syncytial cells containing several nuclei that can actively take part in immune reactions. For the easy accessibility of the skeletal muscle and good vasculature, the delivery of DNA vaccine into this organ is highly preferable (Wiendl et al. 2005). Immunisation with DNA induces all three arms of adaptive immunity (antibodies, helper T cells, CTLs), and even innate immune responses can be easily and rapidly made while maintaining fidelity to the immunological aspects necessary for a pathogen, yet excluding other undesirable proteins or immune responses (Liu 2011).

On the other hand, a major disadvantage of plasmid DNA vaccines is their poor immunogenicity when administered as an unformulated intramuscular injection. Large quantities of DNA are required to induce only modest immunogenicity and many efforts have focussed on the development of new technologies aimed at increasing the DNA vaccine potency (Chiarella et al. 2008a). That said, better strategies are needed in designing more effective vectors and combined protocols so as to induce a strong immune response to weakly immunogenic antigens. These strategies comprise new insights in studying the mechanism of action and induction of the immune response in a host injected with a DNAbased vaccine. Recently, most relevant patented strategies have been developed to enhance the plasmid DNA vaccine immunogenicity taking into account DNA plasmid construction,

target antigen and limited DNA carrying capacity (Harrington et al. 2002; Ramirez et al. 2000). The recent advances made on the knowledge of the immune system biology have led to consider non-viral systems as naked DNA vaccination an alternative, safer and promising approach for introducing foreign antigens into the host to induce an immune response. At the moment, non-viral systems, especially those based on plasmid DNA delivery, have become increasingly desirable in both basic research laboratories and clinical settings.

In 1990, Wolff and collaborators found that bacterial plasmid DNA encoding a reporter gene could result in *in vivo* expression of the encoded protein after simple intramuscular injection without the need for more complex vectors (Wolff et al. 1990). Following Wolff's findings a

Naked-DNA vaccines are for definition vectors based on bacterial plasmids engineered to express the disease-specific antigen using promoter elements active in mammalian cells, without the addition of surrounding chemicals or a viral coat. The main advantages of naked DNA vaccines are safety and production in large amount, as well as stability at different temperatures and, more important, flexibility in design, since multiple antigenic targets or multiple independent cytokines or co-stimulatory sequences can be incorporated into a single DNA vector. They are also likely to be attractive from a health economics perspective: they are relatively easy to manufacture in large quantities in contrast to the complicated processes requested for attenuated virus vaccines, and do not require any special transportation or storage conditions that could hinder their widespread distribution as it happens for live pathogens-based vaccines which need to be distributed and stored in cold conditions. The genes encoding the antigens can be chemically constructed without deriving them from live virulent organisms so avoiding for operators and patients the risks of exposure to dangerous pathogens. They are commonly delivered by a simple intramuscular injection. In mammals the skeletal muscle represents approximately the 30% of the body mass, and muscle fibres are ideal targets for DNA transfection. These are stable and large syncytial cells containing several nuclei that can actively take part in immune reactions. For the easy accessibility of the skeletal muscle and good vasculature, the delivery of DNA vaccine into this organ is highly preferable (Wiendl et al. 2005). Immunisation with DNA induces all three arms of adaptive immunity (antibodies, helper T cells, CTLs), and even innate immune responses can be easily and rapidly made while maintaining fidelity to the immunological aspects necessary for a pathogen, yet excluding other undesirable

On the other hand, a major disadvantage of plasmid DNA vaccines is their poor immunogenicity when administered as an unformulated intramuscular injection. Large quantities of DNA are required to induce only modest immunogenicity and many efforts have focussed on the development of new technologies aimed at increasing the DNA vaccine potency (Chiarella et al. 2008a). That said, better strategies are needed in designing more effective vectors and combined protocols so as to induce a strong immune response to weakly immunogenic antigens. These strategies comprise new insights in studying the mechanism of action and induction of the immune response in a host injected with a DNAbased vaccine. Recently, most relevant patented strategies have been developed to enhance the plasmid DNA vaccine immunogenicity taking into account DNA plasmid construction,

**2. DNA vaccination** 

**2.1 DNA vaccines: An emerging field** 

proteins or immune responses (Liu 2011).

new era of vaccination started.

epitope and antigen choice, selection and use of new adjuvants and different delivery methods.


In table 1 the advantages and disadvantages of DNA vaccines are listed.

Table 1. Advantages and disadvantages of plasmid DNA vaccines.

#### **2.2 Mechanism of action and induction of the immune response**

The crucial event responsible for the initiation of an immune response against a foreign antigen is recognition by specialized cells namely the antigen presenting cells (APCs), uptake and presentation of the antigen to naïve lymphocytes and induction of effector T helper (Th), cytotoxic (CTL) and B lymphocytes.

In this context the mechanism of action of DNA vaccines looks very simple. Once the DNA vaccine is delivered into the skeletal muscle, the plasmid DNA is taken up by the resident DCs and by the muscle fibres. While transfected muscle cells behave as permanent antigen reservoir as well as target of immune effector cells (Payette et al. 2001), resident DCs have the property to leave the muscle tissue and move to the closest draining lymph nodes in order to process and present the antigen to T lymphocytes (**Fig. 1**). DCs are specialized in capturing extracellular antigens by receptor-mediated endocytosis and pinocytosis mechanisms and following antigen uptake they undergo a complex multi-step maturation process. DC maturation depends also on the microbial and pathogens-derived signals which increase their capacity to migrate towards the draining lymph node. While DCs move to the lymphoid organs, they interact with various chemokines which contribute further to their maturation process (Palucka et al. 2010). Once in the lymph nodes, DCs shift from an antigen-capturing cell to a T sensitizing cell, being capable to present antigen in association with the class I and class II MHC molecules to CTLs and Th lymphocytes. Interaction between the DC and the T lymphocyte induces formation of the immunological synapse (IS) *via* complex MHCantigen- T cell receptor (TCR) resulting in the clonal expansion of the T lymphocyte and differentiation in T memory cell. Professional DCs can also capture antigens released in the interstitial space by skeletal muscle fibres or in form of apoptotic bodies activating the cross-presentation pathway (**Fig. 1**) (Russo et al. 2000). This route allows presentation of extracellular/exogenous antigens through the MHC-I restriction pathways (Kurts et al. 2010). Therefore, extracellular antigens which normally induce a humoural immune response can also access to the MHC-I compartment through endoplasmic reticulum, leading to simultaneous stimulation of the CTL immune response. Antigen synthesized

DNA Vaccination by Electrogene Transfer 175

DNA plasmid vaccines offer several advantages when compared to viral vectors or live attenuated vaccines. First of all they are easy to construct and to manipulate which is an important characteristic required to vaccines against pandemic diseases. They are also very stable at room temperature, do not require particular storage conditions which makes them ideal candidates for long-term delivery in other countries. The antigen can be chemically synthesized and cloned directly into the plasmid vaccine simplifying the operations of amplification by molecular techniques and avoiding to work with potentially dangerous live antigen source. The DNA constructs can be easily made to encode for modified proteins deprived of regions that might be dangerous or toxic to the recipient cell or that might suppress the immune response. DNA is a highly flexible molecule, the basic construct can be manipulated in several ways by genetic engineering in order to increase antigen expression, immunogenicity and uptake by recipient cells. All the modifications can affect both the vector backbone and the gene sequence incorporated into the plasmid, which can include adjuvant-like sequences with stimulating activity on the immune system (Abdulhaqq and Weiner 2008). By using this approach, enhanced antigen specific immune responses were observed, suggesting that this could be a general method for targeting antigen to selected cell types. Different strategies can be used for enhancing the plasmid DNA vaccine potency. A first strategy allows to improve the vector construct *i.e.* by working on the plasmid backbone design and construction; a second strategy allows to improve the codon usage in

An expression vector for genetic vaccination generally consists of the following elements: 1) a promoter/enhancer sequence; 2) the gene of interest encoding the target antigen; 3) a polyadenylation/transcriptional terminator sequence; 4) a resistance gene for plasmid selection and an origin of replication (Ori) in order to allow production of high numbers plasmid copies. The capacity of a plasmid DNA vector to drive gene expression is obtained by an optimal combination of all these requisites, and by the possibility of introducing various modifications into the plasmid backbone. The promoters mostly used for gene expression vectors are the cytomegalovirus immediately-early promoter (CMV) the simian virus SV40 early promoter (SV40) and the Rous sarcoma virus promoter (RSV). The CMV promoter is the most popular, as it drives gene expression in a wide range of cells and tissues (Lundquist et al. 1999). Plasmid DNA vectors can also contain tissue specific, synthetic and controllable promoters, whose sequences are designed for a specific use (Papadakis et al. 2004). When the expression of a gene is desired in certain tissues, promoters that control expression in a cell- tissue-specific manner are used. For instance the alpha skeletal muscle actin promoter is specifically used for selective expression in skeletal muscle cells whereas endothelial cell-specific promoters are used to drive expression in the tumour-associated endothelial cells (Dong and Nor 2009). The flexibility of DNA vector permits investigators to exploit the concept of gene expression optimisation by creating synthetic promoters, such as enhancer/promoters composed of numerous combinations of various regulatory sequences (Edelman et al. 2000; Li et al. 1999). One of these regulatory synthetic sequence, the hybrid CMV-Ub promoter, was found to have higher expression than the natural muscle promoters (Yew et al. 2001). In order to display gene expression kinetics, naturally regulated systems were developed by incorporating sequence elements that respond to the local environment of the given cell or tissue or that are regulated by small molecule drugs (Yew 2005). Thus transgene expression can be regulated by modulating expression of these transcriptional factors or by altering their activity through drug administration. However, regulation of *in vivo*

**2.3 DNA plasmid construction** 

order to maximize the antigen synthesis.

by DC or skeletal muscle cell can also be released in the extracellular environment and activate directly the B lymphocytes through antigen-antibody interaction (**Fig.1**).

Considering the mechanism described above, plasmid DNA vaccines are able to stimulate all the principal effector cells of the adaptive immune system but due to the presence of CpG islands intrinsic to the DNA structure they can also can mimic some aspects of live infection, activating important signals of the innate immune system (Matzinger 2007)**.** Theoretically genetic immunisation by plasmid DNA vaccines seems to confer the same broad immunological advantages as immunisation with live, attenuated vaccines does, without the accompanying safety concerns associated with live infection, such as reversion to the virulent form and/or incomplete inactivation of live vaccines. However there are some obstacles that make naked DNA vaccination less potent than traditional vaccines.

Fig. 1. Mechanism of action of plasmid DNA vaccination.

The antigen sequence is cloned into a bacterial plasmid vector specific for vaccination. The DNA vaccine is administered by intramuscular injection. After plasmid uptake by muscle cells, the gene coding for the antigen is translated into a protein. Transfected skeletal muscle cells can present the antigen to CTL through MHC-I molecules (**A**) as well as through DCs cross presentation (**B**). Plasmid is also uptaken by resident DCs which synthesize the antigen and present it in association with MHC-I molecules and MHC-II molecules to CD8+ T and CD4+ T-helper lymphocytes (**C**). Furthermore, antigen is released in the extracellular environment inducing the production of antibodies by B lymphocytes whose activation is mediated by the CD4+ T-helper cells (**D**). SMC, skeletal muscle cell; DC, dendritic cell; CTL, cytotoxic lymphocyte.CD8 T cell, CD8 T lymphocyte; CD4 Th cell, CD4 T helper lymphocyte; B cell, B lymphocyte.

#### **2.3 DNA plasmid construction**

174 Non-Viral Gene Therapy

by DC or skeletal muscle cell can also be released in the extracellular environment and

Considering the mechanism described above, plasmid DNA vaccines are able to stimulate all the principal effector cells of the adaptive immune system but due to the presence of CpG islands intrinsic to the DNA structure they can also can mimic some aspects of live infection, activating important signals of the innate immune system (Matzinger 2007)**.** Theoretically genetic immunisation by plasmid DNA vaccines seems to confer the same broad immunological advantages as immunisation with live, attenuated vaccines does, without the accompanying safety concerns associated with live infection, such as reversion to the virulent form and/or incomplete inactivation of live vaccines. However there are some obstacles that make naked DNA vaccination less potent than traditional

DC

MHC-II MHC-I

CD4 Th cell

**C** 

Antigen processing and presentation by DCs to naive Th and CTL

Plasmid DNA vaccine

DC maturation and migration to the lymph node

CD8 T-cell

activate directly the B lymphocytes through antigen-antibody interaction (**Fig.1**).

Fig. 1. Mechanism of action of plasmid DNA vaccination.

SMC

MHC-I

SMC killing by CTL

SMC Apoptotic bodies **B** 

**A** Cross-presentation

CTL migration

CTL

Antigen secretion

B cell

CD4 Th cell

lymphocyte; B cell, B lymphocyte.

The antigen sequence is cloned into a bacterial plasmid vector specific for vaccination. The DNA vaccine is administered by intramuscular injection. After plasmid uptake by muscle cells, the gene coding for the antigen is translated into a protein. Transfected skeletal muscle cells can present the antigen to CTL through MHC-I molecules (**A**) as well as through DCs cross presentation (**B**). Plasmid is also uptaken by resident DCs which synthesize the antigen and present it in association with MHC-I molecules and MHC-II molecules to CD8+ T and CD4+ T-helper lymphocytes (**C**). Furthermore, antigen is released in the extracellular environment inducing the production of antibodies by B lymphocytes whose activation is mediated by the CD4+ T-helper cells (**D**). SMC, skeletal muscle cell; DC, dendritic cell; CTL, cytotoxic lymphocyte.CD8 T cell, CD8 T lymphocyte; CD4 Th cell, CD4 T helper

vaccines.

B cell activation

**D** 

DNA plasmid vaccines offer several advantages when compared to viral vectors or live attenuated vaccines. First of all they are easy to construct and to manipulate which is an important characteristic required to vaccines against pandemic diseases. They are also very stable at room temperature, do not require particular storage conditions which makes them ideal candidates for long-term delivery in other countries. The antigen can be chemically synthesized and cloned directly into the plasmid vaccine simplifying the operations of amplification by molecular techniques and avoiding to work with potentially dangerous live antigen source. The DNA constructs can be easily made to encode for modified proteins deprived of regions that might be dangerous or toxic to the recipient cell or that might suppress the immune response. DNA is a highly flexible molecule, the basic construct can be manipulated in several ways by genetic engineering in order to increase antigen expression, immunogenicity and uptake by recipient cells. All the modifications can affect both the vector backbone and the gene sequence incorporated into the plasmid, which can include adjuvant-like sequences with stimulating activity on the immune system (Abdulhaqq and Weiner 2008). By using this approach, enhanced antigen specific immune responses were observed, suggesting that this could be a general method for targeting antigen to selected cell types. Different strategies can be used for enhancing the plasmid DNA vaccine potency. A first strategy allows to improve the vector construct *i.e.* by working on the plasmid backbone design and construction; a second strategy allows to improve the codon usage in order to maximize the antigen synthesis.

An expression vector for genetic vaccination generally consists of the following elements: 1) a promoter/enhancer sequence; 2) the gene of interest encoding the target antigen; 3) a polyadenylation/transcriptional terminator sequence; 4) a resistance gene for plasmid selection and an origin of replication (Ori) in order to allow production of high numbers plasmid copies. The capacity of a plasmid DNA vector to drive gene expression is obtained by an optimal combination of all these requisites, and by the possibility of introducing various modifications into the plasmid backbone. The promoters mostly used for gene expression vectors are the cytomegalovirus immediately-early promoter (CMV) the simian virus SV40 early promoter (SV40) and the Rous sarcoma virus promoter (RSV). The CMV promoter is the most popular, as it drives gene expression in a wide range of cells and tissues (Lundquist et al. 1999). Plasmid DNA vectors can also contain tissue specific, synthetic and controllable promoters, whose sequences are designed for a specific use (Papadakis et al. 2004). When the expression of a gene is desired in certain tissues, promoters that control expression in a cell- tissue-specific manner are used. For instance the alpha skeletal muscle actin promoter is specifically used for selective expression in skeletal muscle cells whereas endothelial cell-specific promoters are used to drive expression in the tumour-associated endothelial cells (Dong and Nor 2009). The flexibility of DNA vector permits investigators to exploit the concept of gene expression optimisation by creating synthetic promoters, such as enhancer/promoters composed of numerous combinations of various regulatory sequences (Edelman et al. 2000; Li et al. 1999). One of these regulatory synthetic sequence, the hybrid CMV-Ub promoter, was found to have higher expression than the natural muscle promoters (Yew et al. 2001). In order to display gene expression kinetics, naturally regulated systems were developed by incorporating sequence elements that respond to the local environment of the given cell or tissue or that are regulated by small molecule drugs (Yew 2005). Thus transgene expression can be regulated by modulating expression of these transcriptional factors or by altering their activity through drug administration. However, regulation of *in vivo*

DNA Vaccination by Electrogene Transfer 177

increasing the affinity of the binding to the MHC molecules or by augumenting the binding ability of the peptide–MHC complex for the TCR. The first approach is the most widely used (Dudek NL et al. 2010). This strategy is commonly exploited in the design of cancer vaccines, allowing to convert a subdominant into a dominant epitope by making it more competitive in the binding to specific MHC alleles, and thereby enhancing the potency of the vaccine. The primary and/or secondary anchor residues of an epitope can be replaced with specific aminoacids that provide much of the specificity of binding to the MHC molecule**.** Epitope "enhancement" is possible for both classes of human MHC resulting in priming of the CD8+ cytotoxic T-lymphocytes (CTL) which can recognise the target epitope on tumour or infected cells and in activation of CD4+ T helper cells whose role is crucial for promoting humoural and cytolytic responses, regardless the CTL epitope enhancement. The selection of biologically relevant epitopes within an antigen sequence is performed with different bioinformatic softwares. Several databases of MHC binding peptides now exist and a number of programs performing such predictions are available on the web. We experienced the use of SYFPEITHI, BIMAS and PROPRED I-II programs (www.syfpeithi.de; wwwbimas.cit.nih.gov; www.imtech.res.in/raghava/propred)(Parker et al. 1994; Rammensee et al. 1999; Singh and Raghava 2001). They are based on different algorithms that provide estimation of the binding of a certain peptide sequence to a wide spectrum of human MHC molecules. Unfortunately, a major drawback of these programs is their intrinsic feature of being 'predictive' in estimating the binding affinity between the MHC molecule and the antigenic epitope, with approximately 70% reliability. Another limit is they cannot calculate the binding of the MHC-peptide complex to the T cell receptor, which is a crucial point to

Once immunogenic peptides are predicted *in silico* they need to be verified experimentally. This goal is achieved by performing *in vitro* assays to confirm their stability and binding capacity. After that, human CTL isolated from patients are assayed *in vitro* to verify their capability to recognize specifically the selected epitope on the tumour cells. This result represents the validation of the tumour antigens. Once the antigenic epitope has been validated, it can be taken in consideration for DNA vaccines manufacture and for initiating

It is widely accepted immunogenicity of DNA vaccines is generally weak in comparison to that of traditional vaccines. Although plasmid vaccines are capable to induce a complete immune response involving activation of CTLs, Th and B lymphocytes and a certain activation of the innate compartment of the immune system, these vaccines show low potency and efficacy when administered as unformulated injection. DNA vaccines like all the subunit vaccines, which are made of purified or recombinant antigens, require additional components to help stimulating a "comprehensive" immune reaction. Such

Adjuvants of current use, either in man or in animals, have for the most part been developed empirically, without a clear understanding of their cellular and molecular mechanisms of action. Indeed adjuvants were historically defined as "the dirty little secret of the

The main function of an adjuvant is to create a depot of antigen at the injection site, resulting in a gradual release of small quantities of antigen over a long period of time. The adjuvant also serves as a vehicle for delivering the antigen to the lymph organs, where antigenic

verify the immunogenicity of the tumour antigenic determinant.

"help" is provided by substances and components termed adjuvants.

vaccination trials.

immunologist" (Janeway 1989).

**2.5 Adjuvants** 

transgene expression by such approaches is unreliable, mainly due to the low levels of control associated with the complexity of these systems. Appropriate choice of regulatory elements and vector backbone can lead the gene expression kinetics from a few days to several months. Nuclear localization of plasmid DNA is another prerequisite for the effective antigen expression. To this purpose our laboratory developed a series of plasmids with a functional nuclear translocation sequence (NTS) (Ciafre et al. 1998). All the characteristics of plasmid DNA vaccines mentioned above are summarized in **Table 1**.

The codon usage is an important issue to consider in the DNA vaccine manufacture. Plasmid DNA are totally dependent on the host cell machinery for protein transcription and translation. Since codon usage of bacterial organisms is different from the codon usage of mammalian, it is mandatory to optimize codon usage in DNA vaccine to allow maximum antigen expression (Bojak et al. 2002). Various codon usage approaches are now commonly exploited in both non-human primate studies and clinical trials. This strategy has been successfully used to optimize the sequence of mycobacterial antigens (e.g., Ag85B) improving protein expression and thereby enhancing the immunogenicity of DNA vaccines against M. tuberculosis (Ko et al. 2005).

#### **2.4 Epitope and antigen selection**

The first requisite for a DNA vaccine to induce effective immune response is related to the choice of the target antigen. Sometimes the immunogenic determinant of a certain pathogen is unknown hence selection of antigen sequences has to be included as first step in the design of an epitope-based vaccine. The goal is to identify relevant T cell epitopes, able to bind to MHC class I and II molecules that are both effective and sufficient in vaccine protection against pathogen challenge or, in the case of cancer vaccines, T cell epitopes of malignant antigens that are not ignored by the immune system. In particular for tumours, highly immunogenic antigen determinants remain to be identified for most cancers types. The "direct immunological approach" which consists of deriving tumour cell lines from malignant biopsies, isolating the cancer antigens and expanding the human CTLs specific for that given antigen is now substituted by the "reverse immunological approach" (Sette and Rappuoli 2010). With this new method a candidate cancer antigen expressed on a tumour is selected by *in silico* studies. First tumour antigens are identified by exploiting immune assays based on the availability of specific polyclonal and monoclonal antibodies. Otherwise the serological analysis of recombinantly expressed clones technology (SEREX) is also supportive to identify novel tumour antigens where the blood serum of patients affected by neoplastic diseases is screened against tumour antigen cDNA expression libraries (Jager et al. 2004). Techniques employed in the molecular biology field are also helpful. The analysis of the human transcriptome based on DNA and RNA microarrays allows identification of cancer antigens in a high-throughput system. As following step, putative antigenic determinants are predicted with the aid of bioinformatics. This makes it possible to identify a variety of epitopes within an antigen sequence and to choose the best candidates for the binding to specific MHC molecules. This system works well for epitope discovery, and predictions of the MHC class I pathway is being further improved by integration with prediction tools for proteasomal cleavage and Transporter associated with Antigen Processing (TAP) binding (Larsen et al. 2007) . Furthermore, native epitopes that do not fit perfectly into the MHC groove can be modified at specific sites to increase their affinity to the MHC molecule of interest, leading to the generation of what are called "heteroclitic" epitopes (Dyall et al. 1998). The antigen determinant can be modified either by

transgene expression by such approaches is unreliable, mainly due to the low levels of control associated with the complexity of these systems. Appropriate choice of regulatory elements and vector backbone can lead the gene expression kinetics from a few days to several months. Nuclear localization of plasmid DNA is another prerequisite for the effective antigen expression. To this purpose our laboratory developed a series of plasmids with a functional nuclear translocation sequence (NTS) (Ciafre et al. 1998). All the characteristics of plasmid

The codon usage is an important issue to consider in the DNA vaccine manufacture. Plasmid DNA are totally dependent on the host cell machinery for protein transcription and translation. Since codon usage of bacterial organisms is different from the codon usage of mammalian, it is mandatory to optimize codon usage in DNA vaccine to allow maximum antigen expression (Bojak et al. 2002). Various codon usage approaches are now commonly exploited in both non-human primate studies and clinical trials. This strategy has been successfully used to optimize the sequence of mycobacterial antigens (e.g., Ag85B) improving protein expression and thereby enhancing the immunogenicity of DNA vaccines

The first requisite for a DNA vaccine to induce effective immune response is related to the choice of the target antigen. Sometimes the immunogenic determinant of a certain pathogen is unknown hence selection of antigen sequences has to be included as first step in the design of an epitope-based vaccine. The goal is to identify relevant T cell epitopes, able to bind to MHC class I and II molecules that are both effective and sufficient in vaccine protection against pathogen challenge or, in the case of cancer vaccines, T cell epitopes of malignant antigens that are not ignored by the immune system. In particular for tumours, highly immunogenic antigen determinants remain to be identified for most cancers types. The "direct immunological approach" which consists of deriving tumour cell lines from malignant biopsies, isolating the cancer antigens and expanding the human CTLs specific for that given antigen is now substituted by the "reverse immunological approach" (Sette and Rappuoli 2010). With this new method a candidate cancer antigen expressed on a tumour is selected by *in silico* studies. First tumour antigens are identified by exploiting immune assays based on the availability of specific polyclonal and monoclonal antibodies. Otherwise the serological analysis of recombinantly expressed clones technology (SEREX) is also supportive to identify novel tumour antigens where the blood serum of patients affected by neoplastic diseases is screened against tumour antigen cDNA expression libraries (Jager et al. 2004). Techniques employed in the molecular biology field are also helpful. The analysis of the human transcriptome based on DNA and RNA microarrays allows identification of cancer antigens in a high-throughput system. As following step, putative antigenic determinants are predicted with the aid of bioinformatics. This makes it possible to identify a variety of epitopes within an antigen sequence and to choose the best candidates for the binding to specific MHC molecules. This system works well for epitope discovery, and predictions of the MHC class I pathway is being further improved by integration with prediction tools for proteasomal cleavage and Transporter associated with Antigen Processing (TAP) binding (Larsen et al. 2007) . Furthermore, native epitopes that do not fit perfectly into the MHC groove can be modified at specific sites to increase their affinity to the MHC molecule of interest, leading to the generation of what are called "heteroclitic" epitopes (Dyall et al. 1998). The antigen determinant can be modified either by

DNA vaccines mentioned above are summarized in **Table 1**.

against M. tuberculosis (Ko et al. 2005).

**2.4 Epitope and antigen selection** 

increasing the affinity of the binding to the MHC molecules or by augumenting the binding ability of the peptide–MHC complex for the TCR. The first approach is the most widely used (Dudek NL et al. 2010). This strategy is commonly exploited in the design of cancer vaccines, allowing to convert a subdominant into a dominant epitope by making it more competitive in the binding to specific MHC alleles, and thereby enhancing the potency of the vaccine. The primary and/or secondary anchor residues of an epitope can be replaced with specific aminoacids that provide much of the specificity of binding to the MHC molecule**.** Epitope "enhancement" is possible for both classes of human MHC resulting in priming of the CD8+ cytotoxic T-lymphocytes (CTL) which can recognise the target epitope on tumour or infected cells and in activation of CD4+ T helper cells whose role is crucial for promoting humoural and cytolytic responses, regardless the CTL epitope enhancement. The selection of biologically relevant epitopes within an antigen sequence is performed with different bioinformatic softwares. Several databases of MHC binding peptides now exist and a number of programs performing such predictions are available on the web. We experienced the use of SYFPEITHI, BIMAS and PROPRED I-II programs (www.syfpeithi.de; wwwbimas.cit.nih.gov; www.imtech.res.in/raghava/propred)(Parker et al. 1994; Rammensee et al. 1999; Singh and Raghava 2001). They are based on different algorithms that provide estimation of the binding of a certain peptide sequence to a wide spectrum of human MHC molecules. Unfortunately, a major drawback of these programs is their intrinsic feature of being 'predictive' in estimating the binding affinity between the MHC molecule and the antigenic epitope, with approximately 70% reliability. Another limit is they cannot calculate the binding of the MHC-peptide complex to the T cell receptor, which is a crucial point to verify the immunogenicity of the tumour antigenic determinant.

Once immunogenic peptides are predicted *in silico* they need to be verified experimentally. This goal is achieved by performing *in vitro* assays to confirm their stability and binding capacity. After that, human CTL isolated from patients are assayed *in vitro* to verify their capability to recognize specifically the selected epitope on the tumour cells. This result represents the validation of the tumour antigens. Once the antigenic epitope has been validated, it can be taken in consideration for DNA vaccines manufacture and for initiating vaccination trials.

#### **2.5 Adjuvants**

It is widely accepted immunogenicity of DNA vaccines is generally weak in comparison to that of traditional vaccines. Although plasmid vaccines are capable to induce a complete immune response involving activation of CTLs, Th and B lymphocytes and a certain activation of the innate compartment of the immune system, these vaccines show low potency and efficacy when administered as unformulated injection. DNA vaccines like all the subunit vaccines, which are made of purified or recombinant antigens, require additional components to help stimulating a "comprehensive" immune reaction. Such "help" is provided by substances and components termed adjuvants.

Adjuvants of current use, either in man or in animals, have for the most part been developed empirically, without a clear understanding of their cellular and molecular mechanisms of action. Indeed adjuvants were historically defined as "the dirty little secret of the immunologist" (Janeway 1989).

The main function of an adjuvant is to create a depot of antigen at the injection site, resulting in a gradual release of small quantities of antigen over a long period of time. The adjuvant also serves as a vehicle for delivering the antigen to the lymph organs, where antigenic

DNA Vaccination by Electrogene Transfer 179

contributed to increase the efficacy of a DNA vaccine against a simian immunodeficiency virus when it was fused with the immunoglobulin Fc fragment, resulting in augmentation of the cytokine half-life (Barouch et al. 2000). However, the use of IL-2 is now being limited by the emerging evidence that this cytokine can play a major role in maintaining self-tolerance and in supporting survival of CD25+ CD4+ regulatory T cells (T-regs) (Bayer et al. 2005;

IL-12 is another cytokine used in DNA vaccination. It acts on T and NK cells by inducing the generation of CTLs through T-helper 1 cell activation and IFN-γ production. The beneficial effect of IL-12 in pre-clinical experimental tumour models suggested the possibility of using IL-12 as an anti-tumour agent in clinical trials. Despite some toxicity associated with certain doses of IL-12 when administered as a drug in patients affected by melanoma and colon cancer, some clinical responses were observed; this indicates that IL-12 can be used in clinical protocols of cancer therapy where a toxic effect of the cytokine could be acceptable (Atkins et al. 1997; Gollob et al. 2003). Granulocyte/macrophage colony-stimulatory factor (GM-CSF) is probably the most attractive adjuvant for DNA vaccines for its ability to recruit antigen-presenting cells to the site where antigen synthesis occurs as well as for its capacity to stimulate DC maturation. Plasmid DNA vaccines were constructed fusing GM-CSF to the S antigen of Hepatitis B Virus (HBV) to vaccinate HBV-transgenic mice. This fusion construct worked well in conferring protection from the HBV to both normal and transgenic mice (Qing et al. 2010). In another study the utility of GM-CSF as a DNA vaccine adjuvant for glycoprotein B (gB) of pseudorabies virus (PrV) was evaluated in the vaccination of a murine model. Mouse co-inoculation with a vector expressing GM-CSF enhanced the protective immunity against PrV infection. This immunity was caused by the induction of increased humoural and cellular immunity in response to PrV antigen (Yoon et al. 2006). A DNA vaccine encoding the GM-CSF gene and a DNA vaccine encoding the H1N1 influenza (A/New Caledonia/20/99) HA antigen were co-administered by particle-mediated epidermal delivery in Rhesus Macaques. After three immunizations the DNA vaccines were shown to significantly enhance both the systemic and mucosal immunogenicity of the HA

Among genetic adjuvants the pathogen-derived immune-enhancing proteins are noteworthy for their ability to stimulate the immune system when they are fused with target antigens. Modified bacterial toxins, such as anthrax, diphteria and pertussis toxins, are being used in vaccination as effective carriers to deliver foreign epitopes which stimulate protective CTL responses in mammalian cells (Ballard et al. 1996; Carbonetti et al. 1999). However, the ability of modified toxins to activate the host immune system does not reside

The tetanus toxin Fragment C (FrC) is one of the widely used genetic adjuvant as a fusion partner for foreign antigens. This protein was found to increase the immunogenicity of the Schistosoma mansoni glutathione S-transferase antigen when administered as genetic fusion in a live Aro-attenuated vaccine strain of Salmonella (Khan et al. 1994b) and similar results were obtained when a vaccine construct consisting of a portion of P28 glutathione Stransferase was administering intravenously as C-terminal fusion to tetanus toxin FrC in a live Aro-attenuated vaccine strain of Salmonella (Khan et al. 1994a). In cancer vaccination, a domain of the tetanus toxin FrC fused to a single antigenic determinant was demonstrated able to induce an anti-tumoural CTL mediated response in vaccinated mice (Rice et al. 2002). Likewise, in vaccination against B cell lymphoma, DNA vaccines containing the idiotypic determinants of the Ig variable region provided protective immunity against the tumour

only in the delivery effect exerted on the fused antigen (Stevenson et al. 2004).

Setoguchi et al. 2005).

influenza vaccine (Loudon et al. 2010).

epitopes can be presented to T cells by professional APCs. However, recent advances in the immunological research suggest that most, if not all, adjuvants enhance T and B cell responses by engaging components of the innate immune system, rather than by exerting direct effects on the lymphocytes. In DNA vaccinations, adjuvants are used to achieve qualitative alteration of the immune response. Adjuvants confer to DNA or to subunit vaccines the ability to promote an immune response which might not occur in their absence. Here we describe certain classes of adjuvants most widely used in DNA vaccinations. A summary of the adjuvants described bellow is presented in **Table 2.**


Table 2. Adjuvants used in vaccination with naked DNA.
