Genotypic and phenotypic variation in the human immunodeficiency viruses
Despite the involvement of multiple genetic variants of HIV in the causation of disease worldwide, most research has focussed on subtype B, the prevalent subtype in the western world. As a consequence, it is currently not clear whether genetically distinct HIV strains have different biological properties that cause differences in in vivo transmission, disease progression, replication capacity or sensitivity to antiretroviral drugs. In order to increase the ease of classification of viral diversity and thus aid studies into its importance, a novel genotyping tool for classifying HIV-1 subtype based on pol sequence, produced routinely during drug resistance monitoring, has been developed in this thesis. A dataset of 187 full-length HIV-1 sequences was used to generate Gag, Pol, Protease-Reverse Transcriptase (PR-RT) and Env protein sequence alignments. Phylogenetic analyses enabled generation of subtype specific alignments and, whilst sequence variation in the PR-RT dataset was low, this variation was adequate for PR-RT subtype assignment. The subtyping tool, named STAR, utilises position specific scoring matrices (PSSMs) derived from these subtype specific multiple sequence alignments and results in highly accurate reclassification of the subtype alignment sequences, with 98.6% of sequences being accurately assigned a subtype. Subsequent to the development of STAR the importance of HIV genetic variation classified as subtype, was addressed. A comparison of the relative growth capacity of HIV-1 primary isolates of subtypes A, B, C, D, F, group O and HIV-2 was performed in two T-cell environments. A novel reporter cell line was developed specifically to facilitate this work. Clear and consistent differences in in vitro growth phenotype in terms of rate and cytopathogenicity were detected, indicative of intrinsic differences between the HIV-1 types and subtypes. This work was extended by the utilisation of microarray technology which offers the possibility to analyse, at any given time point, the transcriptome of a virus-infected cell. A comparison of the transcriptional responses within T-cells to infection with HIV-1 subtype B, Group O and HIV-2 enabled identification of both core and diverging transcriptional response programs. Whilst the core response program provides insight into the most essential interactions between virus and host during HIV infection of T-cells, analysis of the diverging responses provide evidence that genetically divergent strains of HIV may interact differently with the host. It is proposed that these differences may have the potential to influence disease outcome.