HIV genetic variation: Clinical importance

Abstract

One of the most striking (and exhaustively documented) characteristics of HIV is its variability at nucleic acid and amino acid level. 1-3 Intuitively this would appear to be a vital component of the virus's ability to survive both the onslaught of the immune system and of anti-viral chemotherapy. Following intervention with antiviral drugs, there is now good evidence that resistant mutants become dominant in the quasispecies. 4 The effect of immunological pressure on the appearance of mutants is less clearly defined. The immune system certainly would seem to exert influence in that the decline in plasma viraemia after seroconversion is associated with evidence of a cytotoxic T-cell response 5 and as the disease progresses the viral RNA load correlates inversely with the immunocompetence of the individual. In addition, long-term survival correlates with a vigorous immune response. 6 Whether the variation in viral quasispecies can be said to be driven by immunological pressure is less certain. It is certainly not the only cause of viral variation and the evidence that ' immune selection' is taking place is still largely circumstantial. Immunological escape appears to be well documented in other micro-organisms such as the gonococcus 7-9 and trypanosomes 1° in which antigenic variation occurs due to differential splicing or recombination or acquisition of new DNA and subsequent alteration of surface proteins. The major difference between HIV infection and most others is that in HIV infection alone an antigen specific CD4 cell recruited to a site of viral replication, where antigen presenting cells are exhibiting peptides for which that T-cell is specific, has a very high chance of being infected and killed by the very virus it is there to eliminate. This powerful extra influence on the dynamics of infection makes it difficult to draw conclusions regarding cause and effect in the interplay between viral variation and the immune response. In the related animal model of SIV, and to some extent in HIV, there is circumstantial evidence of difference in selective pressure on viruses in vitro and in vivo. The nef gene is one of the best examples of an immunogenic protein ~1 essential for high level replication in vivo 12 and in which stop codons appear in the gene in vitro correlating with increased replication. 13 In the SIV model in vitro mutations in nef repair back to wild type when the virus is passaged in vivo, suggesting that Nef has an important function in vivo yet is inessential/inhibitory to replication in vitro. 14 The antibody mapping studies of the gp120 envelope protein by Moore and Sodroski 1~ and others also tell a somewhat similar story with the (admittedly somewhat indirect) evidence suggesting the gp120 adopts an open, highly fusogenic structure on in vitro passage compared to wild type isolates in which immunoreactive epitopes are concealed within a more closed structure leading to reduced fusogenicity. The data with regard to cell mediated immunity dependent epitopes is less complete and, therefore, slightly less compelling. Although in vitro display of aberrant peptides can reduce the immune response to the native peptide, 16 this has not been proven as being influential in vivo. Certainly the scenario requiring co-infection of a cell by two viruses, one containing an inhibitory peptide sequence, one with a stimulatory peptide sequence has not been established in vivo. So HIV variation exists. It is an inevitable consequence of the remarkable ability of the virus to mutate, although the majority of mutations are probably neutral or deleterious to the virus. The significance of different viral variants found in different organs 17 has yet to be fully clarified but may have important therapeutic implications. Specific tropism for brain capillary endothelial cells has been demonstrated in some variants and viral sequences responsible for this have been identified. 18 This is one example where variants might profoundly influence disease manifestations. One particularly important piece of the jigsaw which has recently fallen into place is the variation in the envelope glycoprotein and its correlation with the use of cell specific receptors used by the different variants. Fusin, which has homology to the ~ and [3 chemokine receptor families, was identified as a co-receptor for T-cell tropic H1V strains. ~9 Subsequently a study of individuals exposed to HIV but uninfected led to the discovery that another chemokine receptor, CCR-5, can