Rational design based synthetic polyepitope DNA vaccine for eliciting HIV-specific CD8+ T cell responses

Abstract

Advances in defining HIV-1 CD8+ T cell epitopes and understanding endogenous MHC class I antigen processing enable the rational design of polyepitope vaccines for eliciting broadly targeted CD8+ T cell responses to HIV-1. Here we describe the construction and comparison of experimental DNA vaccines consisting of ten selected HLA-A2 epitopes from the major HIV-1 antigens Env, Gag, Pol, Nef, and Vpr. The immunogenicity of designed gene constructs was assessed after double DNA prime, single vaccinia virus boost immunization of HLA-A2 transgenic mice. We compared a number of parameters including different strategies for fusing ubiquitin to the polyepitope and including spacer sequences between epitopes to optimize proteasome liberation and TAP transport. It was demonstrated that the vaccine construct that induced in vitro the largest number of [peptide–MHC class I] complexes was also the most immunogenic in the animal experiments. This most immunogenic vaccine construct contained the N-terminal ubiquitin for targeting the polyepitope to the proteasome and included both proteasome liberation and TAP transport optimized spacer sequences that flanked the epitopes within the polyepitope construct. The immunogenicity of determinants was strictly related to their affinities for HLA-A2. Our finding supports the concept of rational vaccine design based on detailed knowledge of antigen processing.

Introduction

Development of an effective anti-HIV/AIDS vaccine is hindered by numerous factors, including high genetic variation of HIV-1 and its myriad strategies for avoiding immune detection. Attempts to design a vaccine using traditional technologies have largely failed. It appears that novel technologies, concepts, and approaches taking into account the specific features of host–HIV relationship will be required to develop an effective vaccine. One necessary feature is that the vaccine induced immune responses to a wide range of HIV-1 epitopes and variants.

One promising approach is to construct synthetic polyepitope vaccines encoding protective T cell epitopes conserved among various HIV-1 subtypes (Hanke et al., 1998a, Woodberry et al., 1999, Firat et al., 2001, Bazhan et al., 2004, Karpenko et al., 2004, Lorin et al., 2005). This approach avoids many limitations associated with vaccines based on attenuated viruses, whole inactivated viruses, or recombinant viral proteins (Levy, 1993). Ideally, polyepitope vaccines focus the responses on protective determinants excluding responses to non-protective or even detrimental determinants, for example, those inducing an immunopathology or suppressing the immune response.

It is likely that vaccines that provide maximal protection against HIV-1 will induce both broadly neutralizing antibodies and virus-specific CD8+ T cells. A large body of data supports a role for CD8+ T cells in controlling HIV replication and establishing low set points of viral titers in infected individuals (Ogg et al., 1998, Price et al., 1998, Yang et al., 1997, Wagner et al., 1998). It is now evident that effector memory CD8+ T cells persist for prolonged periods in peripheral tissues, where they can provide immediate immune surveillance and antiviral effector functions (Masopust et al., 2001), providing a rational basis for a vaccine strategy.

Based on these considerations, a number of groups are pursuing the development of polyepitope anti-HIV vaccines (Hanke et al., 1998a, Hanke et al., 1998b, Hanke et al., 1998c, Bazhan et al., 2004, Thomson et al., 1995, Thomson et al., 1996). We previously described a synthetic poly-CD8+ T cell epitope immunogen, TCI, a candidate DNA vaccine (Bazhan et al., 2004). The TCI gene product comprises 392 amino acid residues and contains over 80 selected overlapping optimal epitopes (both CD8+ T cell and CD4+ Th) from the major HIV immunogens Env, Gag, Pol, and Nef. Assessment of the applicability of the resulting DNA vaccine for eliciting CD8+ T cell responses has demonstrated promising immunogenicity in mice (Bazhan et al., 2004, Karpenko et al., 2004).

Buoyed by the efficacy of this approach, we sought to improve it by exploiting recent advances in the knowledge regarding endogenous antigen processing to rationally enhance the presentation of individual determinants present in polyepitope gene products. More specifically, we designed a second generation polyepitope vaccine based on (i) genetic attachment of ubiquitin sequence to the N-terminus of polyepitope constructs to target them to the proteasome (Varshavsky et al., 2000) and use of the amino acid residues flanking the determinants to provide (ii) a proteasomal processing of the polyepitope construct (Ishioka et al., 1999, Kuttler et al., 2000, Livingston et al., 2001) or (iii) the motifs for TAP (transporter associated with antigen processing) proteins, necessary for transporting the proteasome generated peptides into the endoplasmic reticulum (ER) (Uebel et al., 1999, Cardinaud et al., 2009). Here we describe the antigenicity and immunogenicity of several rational design based polyepitope DNA vaccines.