Applications Of Gene Guns In The Immune System

The Revised Authoritative Guide To Vaccine Legal Exemptions

Vaccines Have Serious Side Effects

Get Instant Access

Gene gun technologies are most potent when applied to easily exposed targets and when expressing gene products that have high-specific activity or whose biological effects engage systems that amplify the protein's effects. The most potent amplification system in the body is the immune system because this has evolved to detect a small number of infectious agents and amplify a response that may involve millions of cells and billions of effector molecules. Given this, some of the most potent application areas of the gene gun stimulate the immune system by delivering either antigens, cytokines, or both to the host organism. Current applications of the gene gun for immunological applications fall into the following categories:

1. Genetic immunization to provoke cellular and humoral immune responses by delivering antigen genes from pathogens or cancers to the immune system

2. Genetic immunization to divert problematic allergic or autoimmune responses by delivering antigens from the allergen or autoreactive immune cell or by delivery of cytokine genes to skew the T helper-type response

3. Cancer gene therapy to amplify weak responses to self- or mutant antigens present in tumors by delivering genes encoding immunostimulatory or immune presentation proteins

A. Genetic Immunization to Increase Immune Responses

The gene gun and naked DNA injection approaches for in vivo transfection with plasmids in mammals were developed essentially in parallel in the 1980s. With the advent of these approaches, it became theoretically possible to use them to deliver antigen genes into the cells of a host to provoke immune responses. Genetic immunization is performed by introducing the gene(s) for protein antigens into the host animal rather than introducing the antigen itself. Once the plasmid is delivered into the host cell, the gene is expressed and produces the antigen intracellularly. The process therefore essentially uses the host animal itself as a bioreactor to generate its own vaccine antigens to drive both antibody and cellular immune responses.

1. History

The first published demonstration of this novel approach to immunization was by Stephen Johnston's group (26). In this first paper published in 1992, the ''wand'' gene gun (Fig. 3B) was used to deliver plasmids encoding human growth hormone and human a1-antitrypsin into the skin of mice. This in situ gene delivery allowed the host cells to produce these secreted proteins such that they were recognized by the immune system and provoked potent antibody responses against these foreign antigens. The authors coined the term ''genetic immunization'' to describe this process of using genes to immunize a host. Subsequent publications have called the technology ''DNA-based immunization,'' ''DNA vaccines,'' and ''polynucleotide vaccines.'' Recent terminology seems to have settled toward calling the process ''genetic immunization'' and the application a ''genetic vaccine,'' a ''DNA vaccine,'' or a ''gene-based vaccine.''

This first publication by Johnston's group was interestingly delayed by over a year and half by a number of short-sighted and erroneous reviews that produced comments such as ''•••this is cute at best•••'' and "•••this will induce tolerance•••". Even later reviews were skeptical about the utility of this technology with titles such as: ''Genetic immunization—the biological equivalent of cold fusion?'' (27). These reviewer comments were obviously off the mark in predicting the impact of the approach because Johnston's paper has been referenced more than 600 times since 1992, and there are now more than 350 publications in the PubMed database using the approach.

2. Antibody Production

Genetic immunization can elicit antibody responses against encoded antigens to make polyclonal and monoclonal antibodies in the laboratory (26,28). This application is particularly useful for raising antibodies against proteins that cannot be purified directly or that cannot be readily expressed in Esche-richia coli, provided the gene for the protein is available. Antibody production is a practical nonvaccine application of the technology that is particularly useful in this genomic era of biology, when an investigator is more likely to possess the gene that encodes a protein before they will actually have the protein itself. As such, genetic immunization is a simple approach that allows the gene to be used to make antibody tools to analyze proteins and bridge the gap between genomics and proteomics. Given this, a number of companies now make their antibodies only by genetic immunization and a few offer this service commercially to investigators.

3. Vaccine Applications

Johnston's paper demonstrated that genetic immunization could be used to produce antibodies with a gene gun (26). Wolff's observation that muscle injection of naked DNA could transfect host cells (15) prompted several other investigators to explore whether plasmids could be used for vaccination against infectious agents, including influenza and human immunodeficiency virus 1 (HIV-1) (29-31). This approach was theoretically attractive for vaccine applications because these technologies allowed proteins to be produced intracellu-larly in living animals, such that they could be presented by major histocompatibility complex (MHC) class I molecules to stimulate T cell responses. Therefore, a cell would look infected to the immune response without actually having to be infected with a biohazardous agent. The feasibility of generating cellular immune responses by genetic immunization was first published in 1993 by Margaret Liu's group at Merck, where they demonstrated the production of cytotoxic T lymphocyte (CTL) responses against the nucleoprotein of influenza virus (29). In this same publication and in one published the same year by Harriet Robinson's group, both demonstrated that genetic immunization could be used to vaccinate animals and protect them from infection (31).

These observations provided excellent proof of principle for the application of genetic immunization for vaccine approaches driving both humoral and cellular immune responses. Prior to this, combined antibody and cellular responses could only generally be achieved using attenuated forms of the pathogen itself or by use of infectious viral vaccines that carry considerable biosafety baggage. In contrast, genetic immunization generated qualitatively similar immune responses as these whole pathogen or whole virus vaccines, but in the form of a single plasmid without any vector antigens. This meant that a genetic vaccine could provoke immune responses against its encoded genes without generating immune responses against itself, so multiple immunizations with the same vector were feasible. Likewise, since a genetic vaccine typically carries only 1 or a few genes from a pathogen, this type of vaccine could not cause the infection it is intended to prevent (32). This contrasts with potent whole pathogen vaccines such as that for poliovirus that can infrequently escape attenuation and actually cause the disease they should repel (33).

These benefits and others (Fig. 7) have led to the application of genetic vaccines against a growing list of infectious agents and against cancer for prophylactic and therapeutic immunization. The gene gun has been applied for vaccination against HIV, simian immunodeficiency virus (SIV), and feline immuno deficiency virus (FIV) (34-53), hepatitis (54-58), schistosoma (59), rabies and pseudorabies (60,61), encephalitis viruses (62-67), foot and mouth disease virus (68), rubella virus (69), infectious hematopoietic virus (70), papilloma virus (71,72), herpesviruses (73), pseudomonas aeruginosa

Advantages of Genetic Vaccines

Antigens are intracellular —CTLs

DNA is a stable vaccine

Vaccine is not tied to pathogen biology

■ No risk of infection

- High-level antigen expression

■ Can easily remove problem pathogen antigens

■ Can easily add other genes

■ Can easily modity antigens using recombinant DNA technology Vaccine does not express vector antigens 'Can deliver same vaccine multiple times Most efficient vaccine method for priming

Disdvantages of Genetic Vaccines

Lower-level production of antigens than proliferating vaccines or viral vaccines

■ Weaker antibody responses

■ Less efficient as a booster vaccine

■ Less robust when scaled up to large animals and humans

Exogenous DNA can recombine with the genome

* Chance of oncogenesis

True for any vector that delivers DNA Less chance for plasmid than for integrating vectors (i.e., retrovirus, AAV) Exogenous DNA can provoke anti-DNA antibodies

■ Theoretical chance of provoking autoimmunity

Figure 7 Advantages and disadvantages of genetic vaccines.

(74), chlamydia (75), and cancer (51,76-81). Beyond these demonstrations of genetic immunization for prophylactic vaccination, notable examples of gene gun applications of genetic vaccines are for biodefense [e.g., ebola virus (82), anthrax (83)], for emerging pathogens [e.g., ebola virus (82) and hanta virus (32)], and for unique vaccine applications [e.g., control of gingivitis (84)]. The gene gun has been applied for vaccination predominantly in mice, but also in rabbits (39,85), rats (86), chickens (31), cats (53), cattle (73,87), sheep (87), fish (70), nonhuman primates (35), and humans (58).

4. Genetic Immunization Allows Combinatorial Delivery of Antigen Genes for Multivalent Vaccines and Vaccine Discovery

In most applications against infectious agents, genetic immunization is performed to deliver a specific antigen gene from the pathogen to provoke protective cellular or humoral responses. Early genetic vaccines generally delivered single antigen genes to the host for immunization (29,30). Subsequent applications have included multiple genes to provoke multiva-lent immune responses (88). Another approach is to deliver whole genomes of pathogen genes by Expression Library Immunization (ELI) (50,89,90). The ELI approach can be applied to screen whole pathogen genomes to identify vaccine genes where none existed previously or can be used as an approach to delivery all or many antigens from a pathogen to provoke many immune responses simultaneously (50,89,90).

5. Reengineering Antigens

Unlike other vaccines, genetic vaccines are simple plasmids. Therefore, any recombinant technique can be applied to these vaccines to fundamentally manipulate how the vaccine is expressed and how it interacts with the immune system. Recombinant engineering of pathogen and nonpathogen plasmids has utility to increase antigen expression and boost the level of immune responses against these antigens. For example, antigens can be codon-optimized by rebuilding whole genes from oligonucleotides to replace poorly translated codons of pathogen genes with well-translated ones optimized for mammalian expression (91). Other approaches involve fusing poorly translated antigen genes to well-expressed mammalian proteins to increase translation in cis(92).

For pathogen vaccines, reengineering the antigen may be a necessity because many pathogens have evolved potent methods to hide not only themselves, but also their antigens from the immune system. As such, reengineering antigens in genetic vaccines provides an approach to break down immune evasion mechanisms that are built into the structure of pathogen proteins. Antigens can be reengineered to increase antibody responses by fusing them to secretory leaders or whole secreted proteins (89,93). CD8+ CTL responses can be increased in genetic vaccines by fusing antigens to proteasome targeting proteins such as ubiquitin (89) or by fusing them to cytokine domains such as Flt-3 ligand (94). CD4+ T helper responses can be increased by fusing antigens to lysosomal targeting proteins like LAMP-1 for intracellular targeting (95) and to proteins like the Fc domain of immunoglobulin for extracellular targeting to antigen-presenting cells (APCs) (96). Likewise, proteins can be delivered as a cocktail of plasmids, each expressing an overlapping fragment of the protein to break down secondary structures in the antigen that prevent processing by the proteasome and that hide subdominant epi-topes (97). Fragmenting pathogen antigens also has the benefit of inactivating the function of proteins that may be frankly toxic or that inhibit immune presentation of pathogen antigens. By these methods, one can engineer vaccines that are markedly more potent than the pathogen or cancer itself.

6. Amplifying Genetic Vaccine Potency

Genetic vaccines have demonstrated robust protection against a wide variety of pathogens. In most cases, proof of principle has been demonstrated in mouse models where the amounts of DNA delivered could mediate high-level immune responses. Although many vaccines mediate protective levels of immune responses in large animals or humans, other genetic vaccines have failed this scale-up when translated into larger animals (36,37). This has been a particular problem for human immunodeficiency virus (HIV-1), simian immunodeficiency virus (SIV), and shiv chimeric (SHIV) vaccines aimed at testing in nonhuman primates, where initial genetic vaccines failed to mediate protection (36) or mediated only partial protection (37). These weaker responses in larger animals are likely due to the fact that genetic vaccines generally produce nanogram to microgram amounts of antigen, levels that may be inadequate to effectively stimulate potent immune responses in a large mammal. A number of approaches are being tested to amplify genetic vaccine immune response for human applications, including (1) codon optimization, (2) coadministration of cytokine genes, and (3) use of heterologous vaccines to boost genetic vaccine priming.

7. Codon Optimization

Early work using genetic immunization with HIV-1 envelope plasmids generally raised weak antibody responses (98-100). This problem appears to be due in part to the poor codonbias of HIV genes that makes their expression in mammalian drastically reduced compared with other genes (91). The level of expression of envelope can be increased more that 100-fold by codonoptimization in which codons that are poorly represented by mammalian tRNAs can be replaced by codons from highly expressed mammalian genes. This increased expression by codonoptimization produces recombinant HIV genes that mediate substantially better immune responses after genetic immunization (101). Given this, most current genetic vaccine trials in primates or humans now use antigen genes that have been completely reengineered by codonoptimization strategies (102).

8. Coadministration of Cytokine Genes

In some cases, cytokine-expressing plasmids have been added to the plasmid mixture as genetic adjuvants to augment immune responses against pathogens (103-105). However, most genetic vaccine approaches against pathogens are directed at delivering antigen genes rather than cytokine genes. This contrasts with most gene gun applications for cancer, where it is more typical to deliver cytokines into tumors or tumor cells (106) because, in many cases, cancer antigens and their genes are unknown. Combining genetic vaccines with cytokine plasmids as genetic adjuvants is a robust method to increase immune responses (104,107,108). This approach has recently been quite effective at amplifying not only CTL responses, but also increasing control of pathogenic viruses in rhesus macaques (102). In this case, a relatively simple genetic vaccine expressing codon-optimized SIV gag and HIV env was used to immunize rhesus macaques by intramuscular (i.m.) injection. This work demonstrated control of viremia in the macaques only when IL-2-Ig plasmid (a more stable IL-2 protein) was inoculated with the genetic vaccines (102). This highlights the application of cytokine genetic adjuvant to rescue genetic vaccine responses and also highlights the need to provide robust T helper cell support for vaccines aimed at driving CD8 T cell responses.

9. Use of Heterologous Vaccines for Boosting

An alternate approach to amplify immune responses has been to boost genetic vaccines with heterologous vaccines such as recombinant protein vaccines or recombinant viruses (109,110). The improved responses observed with protein or viral vector boosting are likely related to the larger amounts of antigen that are delivered by these methods. Although these protein or viral vaccines are useful for boosting, in most cases, they are not as potent as genetic vaccines for immune priming, perhaps by the ability of DNA vaccines to directly transfect dendritic cells (111) or to cross-prime antigens to dendritic cells (112). Genetic vaccines are also useful for priming, because they do not encode vector-specific antigens. In contrast, viral vectors such as vaccinia or adenovirus generate immune responses against the intended vaccine antigen and also against proteins of the viral carrier. Therefore, when viral vectors are used, they generate immune responses against themselves that preclude their effective readminstration. In contrast, genetic vaccines express only the vaccine antigen. Thus, they impart no immunological memory against the vector itself and can be readministered multiple times or used to prime antigen-specific responses for later amplification by viral vectors with more robust antigen production. Given these observations, a number of protocols intended for humans involve DNA priming with a heterologous vaccine for boosting. One good example of this is a combined DNA and modified vaccinia ankara (MVA) poxvirus vaccine that is now being developed for a phase III clinical trial in Africa as a candidate acquired immune deficiency syndrome (AIDS) vaccine (113). Similarly, Merck is pursuing a DNA prime and adenoviral vector boost strategy to amplify responses (reviewed in (61)).

10. Genetic Immunization of Humans

Although the gene gun has been applied to many types of animals, only 1 pathogen-directed application has been reported in the literature in humans. In this trial, Powderject Vaccines is testing a hepatitis B virus genetic vaccine using the gene gun to immunize healthy volunteers (58). Three groups of 4 people were immunized epidermally by gene gun with 1, 2, or 4 ^g of plasmid-encoding hepatitis B virus surface antigen. The immunization was well tolerated and provoked CD8 T cell responses as well as protective levels of antibodies in all volunteers. The amount of DNA used in this gene gun trial contrasts with other trials using naked DNA injection that use 0.1 to 1.8 milligrams of DNA (114-119).

It should be noted that China is well ahead of the United States in transferring genetic vaccines into humans as demonstrated by the fact that, since 1997, the primary clinical vaccine against hepatitis B in China is a genetic vaccine (120). Although not a gene gun approach, this large scale application of genetic vaccines in China is a fundamental milestone in this vaccine technology.

11. Gene Guns Versus Naked DNA Injection for Genetic Vaccines

Investigators in genetic immunization tend to fall into two camps: those that use gene guns and those that use naked DNA injection. Prior to commercialization of the Helios, most investigators could not obtain a gene gun unless they built one themselves or entered into a research collaboration with companies making guns. Even now when one can buy a gene gun commercially, the cost of the instrument runs in excess of $15,000 for academics, making it prohibitive to many investigators. Given these parameters, the vast majority of ge netic vaccine research has been performed using a syringe and needle to inject large amounts of plasmids as naked DNA in saline. Based on Wolff's results (15), most work has involved naked DNA injection into skeletal muscle. Raz's group subsequently demonstrated that intradermal injection could also be used for genetic immunization (121). At least in mice, the site of intradermal injection has marked effects on the efficiency of transfection, where injection into the rigid structure of the tail can generate 10- to 100-fold higher transfection than i.m. injection or injection at other dermal sites (2). This may be due to the ability to produce higher pressures upon injection into the tough tail versus generating lower pressures in other sites that can expand upon injection. Indeed, the fact that injections in small animals work substantially better than in large primates may be due to the fact that injections into mouse muscle deliver 25 to 50 ^L of volume into a tissue that may itself only normally be 50 to 500 ^L in volume. In contrast, injections into primates and humans are never delivered with same relative volumes as in mice. Given that high-pressure injection into a number of tissues including muscle increases transfection efficiency (122,123), it is likely that the poor scale-up of naked DNA injection large animals and humans is in part due to the use of too small of volumes into the muscle and dermis.

In many applications, the gene gun and naked DNA injection yield comparable immune responses when both are applied under optimal conditions. However, the fundamental mechanisms by which cells are transfected by the 2 approaches are quite different and can yield qualitatively different immune responses.

12. Gene Guns Transfect Dendritic Cells and Non-APCs, Naked DNA Injection Transfects Non-APCs

The gene gun is typically applied for genetic immunzation by directly transfecting epidermal cells of the skin. As such, the gene gun can not only transfect nonantigen-presenting kera-tinocytes, but the gun also transfects Langerhans dendritic cells directly in vivo (111). In contrast, naked DNA injection predominantly transfects non-APCs when applied in the skeletal muscle. Therefore, in i.m. injection, the vast majority of antigen presentation to T cells occurs by transfer of antigens from the transfected muscle cells to host APCs for cross-presentation (124). In contrast, the gene gun may drive immune presentation either by cross-presentation or by direct transfection of dendritic cells, allowing these transfected APCs to express and present antigens themselves. Although some work delivering antigen plasmids under the control of tissue-specific promoters has suggested that gene gun-mediated immune responses occur only by cross-presentation (125), the promoters used were somewhat mismatched for the biology of epidermal cells and were not directly validated by in vivo reporter activity. Other work has compared the efficiency of gene gun modified Langerhans cells that were transfected either directly in vivo or in vitro and then reintro-duced into the animals (126). In this case, the investigators found that transfection of as few as 500 dendritic cells in vitro in the absence of non-APC transfection-generated immune responses comparable to that by direct gene gun delivery. Although direct transfection of dendritic cells in vivo by the gene gun appears promising, it is conceivable that this trans-fection may actually render the cells less efficient at presentation, or they may only inefficiently present their antigens expressed in an autocrine fashion. Further work is required to determine which transfected cells is driving immune responses after gene gun transfection.

13. Gene Gun Delivery Is More Efficient and Consistent Than Naked DNA Injection

The gene gun is inherently more consistent shot to shot than i.m. injection (108). Increasing the amount of DNA in an i.m. injection increases the ''hit'' rate, but does not obviate animals that fail to respond (Fig. 8). To reduce the likelihood that an animal will be ''missed'' by i.m. injection in a given immunization round, it is advisable to inject each animal in four separate sites. To achieve the same level of antigen production, one typically needs to deliver 50 to 100 times as much plasmid by naked DNA injection than by gene gun (31,108). Typical delivery amounts for the gene gun are 1 to 2.5 ^g per shot. For i.m. naked DNA injection, they are typically 50 to 100 ^g per shot in a mouse. This difference in efficiency is likely related to the ability of the gun to deliver plasmids directly into the cytoplasm of cells, whereas syringe injection delivers DNA extracellularly where more than 99% of the injected DNA is rapidly degraded into nonexpressible DNA fragments by the action of extracellular nucleases (2).

14. T Helper Cell Bias for Gene Guns and Naked DNA Injection

Depending on the pathogen, protection can be mediated antibody responses, cellular responses, or a combination of both. For some infectious agents, antibodies are able to neutralize incoming pathogen. For other pathogens, CD8+ T cell responses are advantageous because these can kill or inactivate infected or neoplastic cells. Fortunately for viral and cancer vaccines, the plasmid DNA used for genetic vaccines is inherently biased toward generating CD8+ T cell by virtue of the fact that it is quite potent at biasing T helper responses toward a TH1 phenotype. This bias occurs because the bacterial plasmid DNA bears unmethylated CpG motifs that stimulate IFN-7 and IL-12 production from natural killer (NK) cells and macrophages (127). This production of cytokines biases immune responses toward the TH1 phenotype against any antigen present after injection, whether it is encoded by the DNA itself or delivered as a protein (128).

Because naked DNA injection delivers 50 to 100 times as much DNA as the gene gun, i.m. and intradermal (i.d.) injection has a stronger TH1 bias than the gene gun [(24) and Fig. 8]. Both methods of gene delivery are able to drive TH1 responses at the early phases of responses as evidenced by interferon (IFN)-^ production by T helper cells, but the gene gun appears to have increased TH2 character (at least for some antigens) later in the T helper responses, as evidenced by increases in interleukin (IL)-4 production after IFN-7 re-

Figure 8 TH1 skewing by excess plasmid DNA after gene gun and i.m. injection. BALB/c mice were immunized a single time as indicated, and their antibody responses were evaluated by isotype-specific ELISA against a1-antitrypsin (AAT). (A) Comparison of the effects of large amounts of antigen-coding plasmid (CMV-AAT), noncoding plasmid (CMVB), and eukaryotic DNA (SS DNA) on antibody levels and isotype. (B) Comparison of the effects of noncoding excess DNA and GM-CSF expressing plasmid on antibody boosting and isotype when delivered into the same site as the antigen plasmid or the contralateral muscle.

Figure 8 TH1 skewing by excess plasmid DNA after gene gun and i.m. injection. BALB/c mice were immunized a single time as indicated, and their antibody responses were evaluated by isotype-specific ELISA against a1-antitrypsin (AAT). (A) Comparison of the effects of large amounts of antigen-coding plasmid (CMV-AAT), noncoding plasmid (CMVB), and eukaryotic DNA (SS DNA) on antibody levels and isotype. (B) Comparison of the effects of noncoding excess DNA and GM-CSF expressing plasmid on antibody boosting and isotype when delivered into the same site as the antigen plasmid or the contralateral muscle.

sponses wane (129). Because of this, the gene gun fairs worse in controlling some pathogens that are highly dependent on potent CD8+ T cell responses for protection (130). Conversely, the TH2 bias is the likely explanation for why the gene gun is generally more robust at generating antibody responses than i.m. injection in mice (131) and in primates (132). Therefore, the gun has better utility for protection requiring neutralizing antibodies [e.g., hepatitis B (58)]

Although many investigators call i.m. injection ''TH1'' and the gene gun ''TH2'', this is inaccurate. It is more accurate to say that both can produce mixed TH1/TH2 responses and that i.m. injection skews responses toward TH1, whereas the gene gun skews responses toward TH2. The TH2 bias of the gun is not absolute because the gene gun can provoke quite potent TH1-driven CD8 + T cell responses with nanogram amounts of plasmid (131) and even after single immunization (50). Further, the TH2 bias of the gene gun can be converted to more of a TH1 responses by coimmunization with cytokines like IFN-a or IL-12 (133).

Each antigen has its own TH1 or TH2 bias and the different delivery methods retain the inherent bias or skew it in one direction or the other. For example, protein immunization with human a1-antitrypsin (AAT) in mice produces IgG1 antibodies. If a 2.5 ^g of a plasmid-encoding AAT is delivered by either gene gun or i.m. injection, antibody responses are still predominantly IgG1 (Fig. 8 and data not shown). TH1 skewing (as evidenced by increased IgG2a antibodies) only occurs when more bacterial DNA is added in the i.m. injection (either more AAT plasmid or a nonexpressing plasmid) (Fig. 8A). A similar mixed response involving combined IgG1/IgG2a antibodies is observed if 1 gene guns 50 ^g of DNA. If 2.5 ^g AAT plasmid plus 47.5 ^g more AAT plasmid is introduced by i.m. injection, the response is still predominantly IgG1. In contrast, if one adds 2.5 ^g AAT plasmid plus 47.5 ^g of a nonexpressing plasmid (supplies cytosine-guanosine dinucle-otides (CpGs), but not more antigen), then stronger skewing toward IgG2a responses is observed after i.m. injection (Fig. 8). Therefore, for AAT, the default antibody response is IgG1. Adding more CpGs by gun or i.m. injection skews this toward a TH1 response as evidenced by increased (but not only) IgG2a antibodies.

The CpG effects by plasmid DNA appears to be a systemic one given that increased responses can be observed, whether the large amounts of plasmid is injected into the same site as the antigen-expression plasmid or into a contralateral muscle (Fig. 8B). Increased levels of response as well as increased numbers of animals that respond are observed, whether large amounts of plasmid DNA is injected in the same site or an alternate site to where the antigen plasmid is injected. This site effect is quite different than that observed when using cytokine genes as genetic adjuvants. For example, low-level antibody responses after i.m. injection of suboptimal amounts of a1-antitrypsin (AAT) plasmid can be amplified if a granu-locyte-macrophage colony stimulating factor (GM-CSF) plasmid is coinjected in the same site, but not if the cytokine plasmid is injected in another muscle (Fig. 8B). Note also, that unlike the CpG effect, the GM-CSF antibody boost retains the normal TH-bias of AAT by producing predominantly IgG1

antibodies. Interestingly, if GM-CSF plasmid is delivered into a nonmuscle site (e.g., the skin), this can actually antagonize the immune response (unpublished observations), suggesting the cytokine may attract APCs away from the site of antigen production to the wrong location.

In summary, the gene gun tends to bias responses toward a TH2 phenotype. Naked DNA injection biases responses toward the TH1 phenotype due to delivery of large amounts of unmethylated CpG motifs. Both approaches have the capacity to drive mixed TH1 and TH2 responses, and both can be skewed toward one response or the other by delivery of cytokine plasmids. The TH bias of the antigen is a primary factor in the character of responses. One should be aware that differences in responses are usually highly antigen and model specific, so the approaches with generally need to be compared for each investigator's own system.

B. Genetic Immunization to Reduce or Skew Immune Responses

Allergy and autoimmune diseases represent immunological diseases in which aberrant immune responses occur against exogenous allergens or against self-antigens, respectively. Genetic immunization has been applied in these situations to inactivate or skew the existing immune response to one that does not support cause the disease. Genetic immunization prevents allergic responses when the allergen antigen gene is delivered with CpG-rich DNA to skew the TH2-driven allergic response to a nonproblematic TH1 response. This approach has been applied successfully to prevent allergic responses against peanut (134), pollen (135), and milk antigens (136). For diseases like autoimmune diabetes, there are autoimmune T cell responses against antigens associated with pancreatic p cells. In NOD mice, the diabetes is mediated by responses against glutamic acid decarboxylase. Genetic immunization with a plasmid-encoding glutamic acid decarboxylase was able to prevent spontaneous diabetes in the NOD mice, demonstrating that the skewed responses could disrupt disease progression (137). An alternate approach is to generate immune responses against the antigen receptors of T cells that target self-antigens. Proof of principle for this approach was demonstrated for autoimmune encephalitis by expressing the genes encoding T cell receptor specific to the self-antigen myelin basic protein (138). This report was notable because the vaccine approach not only ablated the autoimmune disease prophylactically, but was also able to attenuate established autoimmunity. These reports on both allergy and autoimmune applications hold great promise for the application of genetic immunization for not only inducing new immune responses, but also ablating problematic immune responses.

C. Gene Gun Applications for Cancer

The use of gene guns and genetic immunization for cancer is usually quite different than for vaccination against pathogens. First, for most pathogens, candidate vaccine antigens are already known and can be applied in the genetic vaccine. For those pathogens that lack candidate genes, ELI can be applied to screen the whole genome to find these candidates (89). In contrast, the proteins that could act as protective antigens against cancer cells are (1) usually unknown, (2) are usually specific to 1 type of cancer, and (3) may in some cases be specific to the individual patient. For those few cancers associated with or caused by infectious agents, one can vaccinate with genes from the pathogen that are associated with the tumor cell. Good examples of this are the use of the gene gun to vaccinate against papillomavirus antigens (139,140). Whole genome searches for antigens by ELI and other techniques is a staggering task for cancer as compared with viral or bacterial pathogens because the mammalian genome is 1000- to 10,000fold larger than the genomes of most pathogens. To combat this size issue, most investigators use cDNA libraries, rather than unbiased genome libraries, to reduce the complexity of their search. Although feasible, screening cDNA libraries for cancer antigens introduces a potent bias in the antigens that are searched because representation in the library is directly proportional to the expression level of the message in the cancer cells. Despite these considerable difficulties, investigators have identified several promising cancer antigens. Given this, the gene gun has been applied to deliver melanoma antigens gp100 (80) and tyrosinase-related protein 2 (81), prepro-calcitonin has been applied for medullary thyroid carcinoma (77), and alpha-fetoprotein has been applied for genetic vaccination against hepatocellular carcinoma (76).

The vast majority of gene gun applications for cancer are directed at generating cancer vaccines. In this application, the gun is used to transfect cancer cells either in vitro or in vivo, with plasmids encoding immunostimulatory proteins to amplify weak anticancer responses. By this approach, the cancer cell provides the antigens and the gene gun provides gene products to increase immune reactions against these antigens. One of the earliest approaches using gene guns for this application was by Kam Hui's group in Singapore, where they engineered their own gene gun and used this to transfect cancer cells with the genes for allogeneic MHC I proteins (78,141). Other applications have delivered genes encoding HLA-A, HLA-DR, and B7 costimulatory molecules (51). In most applications, the gene gun has been used to deliver cyto-kine genes to explanted tumors or tumor cell lines ex vivo (51,142-144). The advantage of the gene gun in these applications is that it mediated gene delivery without biohazardous viruses and without the need for cancer cells to be cycling or cultured for long periods of time. Alternate applications have delivered cytokines to the skin immediately above subcutaneous tumors (145-150), taking advantage of the ease in addressing the epidermis with the gun. Cytokines, including IL-2, IL-12, pro-IL-18, IFN-a, GM-CSF, and tumor necrosis factor (TNF)-a, have been delivered either alone or in combination to modulate cellular immune responses against the cancer cells. In the case of IL-12, direct transfection of the skin above intradermal sarcomas in mice mediated comparable protection to treament with recombinant IL-12 protein (151). However, an advantage associated with the localized delivery and production of IL-12 by the gene gun is that this avoided many of the toxic side effects associated with systemic IL-12 therapy (151). Further, transfection with cytokine genes by gene gun is substantially less expensive than systemic delivery of recombinant cytokine proteins.

Although the vast majority of gene gun testing for cancer has occurred in mouse models, 3 trials have been performed or are underway in humans. In 1 trial, melanoma and sarcoma explants from patients are to be transfected by the gene gun with a plasmid-encoding human GM-CSF and the cells were injected intradermally back into the patients as an autologous cancer vaccine (79). In 2 other trials, IL-7 or IL-12 plasmids were delivered by gene gun into melanoma cells and injected subcutaneously back into patients (152,153). Patients immunized with autologous IL-7-modified cells demonstrated little toxicity, increased CTL levels against the tumor, and 2 of the 6 patient demonstrated minor clinical responses (152). In the IL-12 trial, mild fever was the only side effect of the therapy and increases were observed in cytotoxic T cells (153). One of the 6 patients had a minor clinical response that was associated with infiltration of CD4 and CD8 T cells into metastatic sites distant from the vaccination site. From these phase I trials, the gene gun holds some promise for human tumor applications, particularly when it is applied to deliver known tumor antigens or when it is used to deliver cytokine genes for local production to augment immune responses against unknown tumor antigens.

Was this article helpful?

0 0
How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book

Post a comment