oa Editorial [Hot topic: The Significance of HIV-1 Genetic Diversity for Vaccine Development (Guest Editor: Michael M. Thomson)]

A recent phase III vaccine trial in Thailand has shown for the first time that a vaccine can prevent HIV-1 infection in humans [1]. Although the efficacy was modest, and the protection might have been short-lived, the results of this trial have raised hopes in the field of HIV-1 vaccine research after a number of failures. The vaccination protocol consisted in priming with a canarypox virus vector component expressing Gag, protease, and gp41 derived from an HIV-1 subtype B isolate and gp120 from a CRF01_AE virus, and boosting with envelope glycoproteins of isolates of subtype B and CRF01_AE. The genetic composition of the immunogens was guided by phylogenetic and molecular epidemiological studies of the viruses circulating in the targeted population, among which a CRF01_AE variant, introduced in Thailand the late 1980s, is the predominant clade, with a subtype B strain introduced around the same time, circulating as a minor genetic form, and by cross-neutralization studies indicating that the Thai CRF01_AE variant correlates with a neutralization serotype [2]. Whether the genetic relatedness between the vaccine immunogens and the locally circulating HIV-1 variants favored vaccine-mediated protection is the subject of ongoing studies. Whatever their results, the design of the immunogens of this vaccine and of others planned for future testing, in which the immunogens match the HIV-1 variants circulating among the targeted population, exemplify the growing and widely accepted idea that HIV-1 genetic diversity is an essential element to be taken into consideration in HIV-1 vaccine design, which is the central topic of the articles of this special issue of Current HIV Research. This idea has not always had a general support, with most earlier work failing to find correlations between antibody-mediated neutralization and HIV-1 subtypes, and most studies on T cell-mediated responses emphasizing the detection of cross-clade reactivities. Based on those studies, it was suggested that a HIV-1 vaccine design approach targeted to subtypes seemed “unnecessary and without scientific foundation” [3]. However, multiple subsequent studies have supported that there may be a correlation between HIV-1 clades and antibody-mediated neutralization (reviewed by van Gils and Schuitemaker) and have detected cell-mediated responses with preferential or specific intraclade reactivities (reviewed by McKinnon et al., who also underscore the importance of inducing cross-reactive CD8+ T cell responses for vaccine protection, and of using multiple immunological assays for their detection). The relevance of viral genetic diversity for vaccine efficacy is also supported by results obtained with other lentiviruses. A study with an attenuated equine infectious anemia virus (EIAV) vaccine in ponies found an inverse linear correlation between vaccine efficacy and increasing genetic divergence of the challenge virus gp90 (the EIAV envelope surface glycoprotein), with an amino acid distance of as little as 6% from the vaccine surface glycoprotein resulting in a 25% reduction in protection from disease, which increased to 50% when the divergence was 13% [4]. On the other hand, single-subtype feline immunodeficiency virus vaccines have been protective only against homologous or homologous-subtype in vitro-derived challenges [5]. Further knowledge on the significance of HIV-1 variants for immune responses relevant for vaccine protection in humans might come, apart from the vaccine trials themselves, from large population-based studies on the incidence of superinfection (reviewed by Chohan et al.) in areas where multiple clades circulate. The difficulty in detecting strong correlations between HIV-1 subtypes and immune responses may derive from multiple causes: a) large intrasubtype genetic distances, which may currently be similar to or greater than intersubtype divergence at the time of the origin of subtypes, with epitopes originally shared within clades being progressively lost through accumulation of mutations along subtype diversification; b) frequent intersubtype recombination in areas where multiple subtypes circulate, which may confound the results of cross-neutralization assays when envelope sequences of the isolates used in the assays or of those inducing the antibody responses are not fully characterized, as occurred in some of the earlier studies; c) cryptic dual infections with viruses of different clades, in which one of the variants remains undetected due to its low abundance, with immune responses induced by the minority variant being erroneously interpreted as cross-clade responses; d) the use of assays to detect cytotoxic T lymphocyte responses that may not reflect antiviral activity [6]. Therefore, it is important that these points be taken into consideration when conducting studies on HIV-1 cross-clade immune responses. Given the great degree of genetic divergence within HIV-1 subtypes, it may also be relevant to consider the intrasubtype phylogenetic structure in vaccine-related studies. Such structure is apparent in some geographic areas, where a large proportion of viruses branch within well supported intrasubtype clusters [7-9]. In other areas, extensive intrasubtype recombination may have blurred the distinction between clusters, which may not be apparent when employing the usual methods of phylogenetic analysis. In this regard, the detection of regional clustering of neutralization, as reported for subtype C isolates from South Africa [10], may reflect the existence of an underlying intrasubtype phylogenetic clustering (Thomson MM, unpublished results). The lower genetic diversity within intrasubtype clusters or variants, compared to subtypes, would be expected to be reflected in stronger correlations with immune responses (as observed with the Thai CRF01_AE variant in neutralization assays [2]), which may be of relevance for the genetic composition of the vaccine immunogens used in areas where such clusters or variants are circulating. Different strategies for immunogen design have been devised to broaden the immune responses effective across the wide diversity of HIV-1 variants. One of them is the use of polyvalent formulations consisting of cocktails of immunogens derived from different HIV-1 clades, similarly to other widely used vaccines of proven efficacy against other pathogens (reviewed by Lu et al.). Another is the use of immunogens with centralized (either ancestral, consensus, or center of tree) sequences, designed to minimize the genetic distances with a majority of circulating viruses within a clade (reviewed by Arenas and Posada, who underscore the importance of taking into account recombination and complex models of evolution for the design of such immunogens). Additional strategies include the use of computationally designed mosaic proteins assembled from natural sequences optimized to maximize coverage of the most common potential T-cell epitopes across HIV-1 clades [11], and of immunogens incorporating the most highly conserved regions of the HIV-1 proteome, in which escape mutations would be predicted to severely compromise virus viability [12-14]. These strategies are not mutually exclusive and could be used in combination. In rare HIV-1-infected individuals antibodies are produced possessing broad and potent neutralizing activity across HIV-1 clades (reviewed by Gonzalez et al.). Rational design of Env immunogens based on the epitopes targeted by such antibodies would be another strategy for broadening humoral responses elicited by HIV-1 vaccines. However, it is uncertain whether such a vaccine could induce these rare antibodies in the general population, if their elicitation depends on the particular primary B-cell repertoires present in a few individuals and on a lengthy affinity maturation process of the antibody response in the context of a chronic HIV-1 infection [15]. Given the lack of adequate animal models, definitive knowledge on the significance of HIV-1 genetic diversity for vaccine efficacy in humans will only come from large-scale vaccine trials in populations among which different HIV-1 variants (including subtypes, circulating recombinant forms, and intraclade variants) are circulating. However, cumulating evidence, as reviewed in this issue, supports the relevance of the correlations of HIV-1 genetic forms to immune responses for vaccine design. This constitutes a rapidly expanding area of research which will most likely be of key importance for the development of effective preventive vaccines against HIV-1.