Generation of multivalent genome-wide T cell responses in HLA-A*0201 transgenic mice by an HIV-1 expression library immunization (ELI) vaccine.

Abstract

HIV-1 is a fundamentally difficult target for vaccines because of its high mutation rate and its repertoire of immune evasion strategies. To address these difficulties, a multivalent genetic vaccine or "live genetic vaccine" was recently developed against HIV-1 using the expression library immunization (ELI) approach. In this HIV-1 vaccine, all open reading flames of HTLV-IIIb are expressed as protein fragments to retain all viral T cell epitopes, but destroy protein toxicity, inactivate immune escape functions, and reveal subdominant epitopes. In addition, each antigen fragment is fused to the ubiquitin protein to increase antigen expression and target these antigens to the proteasome to enhance cytotoxic T lymphocyte (CTL) responses. This multivalent vaccine also has the advantage of being incapable of generating infectious HIV-1 virus because of the segregation of the HIV genome into 32 separate plasmids. In this work, we demonstrate the ability of this genetic vaccine to provoke robust HLA-A*0201-restricted T cell responses in MHC class I humanized mice against gag, pol, env, and nef after a single round of immunization. In addition, this HTLV-IIIb-derived vaccine demonstrated cross-clade, envelope-specific, HLA-restrieted CD8 responses against clades A, D, and E. HLA-restricted CD8 responses were generated against all 32 open reading frames encoded by the multi-plasmid genetic vaccine demonstrating that a broad repertoire of human relevant CD8 responses are provoked by this vaccine. This work supports this approach to generate multivalent T cell responses to control the highly mutable and immuno-evasive HIV-1 virus.

Presentation Summary

Michael Barry, PhD, opened his talk by acknowledging that his work focuses on developing new technologies and is not primarily immunologic or virologic in nature. Barry's lab works in gene therapy, including the use of gene therapy in vaccines, a strategy known as genetic immunization. The process basically involves putting antigens into plasmids and injecting them into animals or humans (with a syringe or a gene gun) to elicit immune responses (see Figure 1). This strategy appears capable of eliciting both cellular and humoral immune responses. An advantage of genetic vaccines is simplicity of the system overall to deliver antigens intracellularly, causing CTL responses. Also, DNA functions as a stable vaccine and is not tied to the biology of the pathogen. In other words, there is no risk of infection, yet there is high-level antigen expression; problem pathogen antigens (such as those involved in immune evasion) can be removed, or other genes added; and antigens can be easily manipulated using recombinant DNA technologies that are now well established.

[FIGURE 1 OMITTED]

One challenge to using live-attenuated vaccine for HIV is that the vaccine uses viral proteins to elicit immune responses. However, the virus has evolved to evade the immune system, which vaccines themselves are intended to activate. Therefore, Barry believes that any potential vaccine would become biased against the immune response. So, how can an effective vaccine be created to target a virus that evades the very immune system that the vaccine activates? To overcome this dilemma, the best possible antigens would be needed to elicit the best possible immune response. A useful approach may involve the use of an expression library, whereby the genome of a pathogen is fragmented and the pieces of DNA captured to create a pathogen library that represents most of the T cell epitopes that exist for that pathogen. The library can be broken into sublibraries and used in animal immunization studies to select for ideal antigens to use in vaccines. Such libraries can be created through a random sheering process or through a directed process with fragments created deliberately at specific points.

The strategy of using expression libraries as vaccines has worked with several types of bacterial pathogens. Considering HIV's relatively small genome, Barry's group has been looking at using an entire expression library as a "genomic vaccine" representing all or many of HIV's antigens. The question is whether such an expression library immunization (ELI) vaccine could deliver multivalent epitopes to drive immune responses to HIV. Barry and his colleagues have been studying such a vaccine, which uses 32 plasmids and is derived from HTLV-IIIb (see Figure 2). The vaccine antigens have been fused to ubiquitin, which has been shown to enhance MHC presentation and CTL responses. The group's work in mice has demonstrated that the ELI vaccine was able to elicit CTL responses against subdominant epitopes of gag, whereas a protein vaccine of gag plus adjuvant was able to include only one such response against the main gag epitope. Such broadened responses against one viral protein target may reveal new ways of effectively targeting CTL responses against "hidden" or subdominant epitopes of HIV.

[FIGURE 2 OMITTED]

In a Center for AIDS Research (CFAR) collaboration at Baylor College of Medicine, the vaccine was studied in rhesus macaques and, in at least some animals, seemed to generate multivalent T cell responses that apparently correlated with better control of viral load after challenge with a hybrid simian-human immunodeficiency virus (SHIV). Macroaggregated albumin (shown to target lung tissue) was used as an adjuvant to help elicit systemic as well as mucosal responses. Since the ELI vaccine that was used generates random epitope fragments, codon optimization may produce more robust CTL responses.

Studies in mice indicate that the presence of multiple antigens (32) does not appear to interfere with immune responses. Also, the immune responses appear to be HLA-restricted. In a study evaluating the relative contribution of each of the 32 plasmids to the vaccine's immunogenicity, 32 cell lines were created expressing HLA-A*0201 and one of the library members. When interferon-[gamma] production was measured in these cell lines after 6 hours of in vitro stimulation, every library member appeared to generate a response, thus suggesting a multivalent response. In another study, HLA transgenic mice were immunized with a gag-pol plasmid and an env plasmid. Next, T cell responses were measured against the panel of 32 cell lines. Fairly strong responses were seen in the genomic areas coding for the immunodominant epitopes gag, pol, and env. In contrast, when the ELI (whole library) vaccine was used, strong responses were still seen in the gag, pol, and env regions, but also in adjacent regions, further suggesting a multivalent response (see Figure 3). Ongoing research is looking at whether such responses can he elicited across clades. Preliminary research thus far suggests some cross-clade CD8 T cell responses.

[FIGURE 3 OMITTED]

Next steps include codon optimization of the libraries, investigation of other potential vectors, and the use of gene delivery vectors (like engineered adenoviruses with specific ligands attached) that can target immunologically relevant cells. The use of targeted vectors would decrease the amount of particles needed and the immunization of cells that do not play immunological roles.

Further Reading

Barry MA, Lai WC, Johnston SA. Protection against mycoplasma infection using expression-library immunization. Nature. 1995;377(6550):632-635.

Orson FM, Kinsey BM, Hua PJ, Bhogal BS, Densmore CL, Barry MA. Genetic immunization with lung-targeting macroaggregated polyethyleneimine-albumin conjugates elicits combined systemic and mucosal immune responses. J Immunol. 2000; 164(12):6313-6321.

Santra S, Barouch DH, Kuroda MJ, et al. Prior vaccination increases the epitopic breadth of the cytotoxic T-lymphocyte response that evolves in rhesus monkeys following a simian-human immunodeficiency virus infection. J Virol. 2002;76(12):6376-81.

Singh RA, Wu L, Barry MA. Generation of genome-wide CD8 T cell responses in HLA-A*0201 transgenic mice by an HIV-1 ubiquitin expression library immunization vaccine. J Immunol. 2002; 168(1):379-391.

Sykes KF, Lewis MG, Squires B, Johnston SA. Evaluation of SIV library vaccines with genetic cytokines in a macaque challenge. Vaccine. 2002;20(17-18):2382-95.