HIV genotyping

Abstract

The development of viral resistance to antiretroviral drugs used for treatment of human immunodeficiency virus 1 (HIV) infection is an important cause of antiretroviral treatment (ARV) failure and limits options for alternative antiretroviral regimens. Prevention, characterisation and clinical management of such resistance are receiving increasing attention. The primary objective of this project was to study the naturally occurring variants of HIV-1 present in a group of ARV-naiVe patients. The presence of pre-existing mutations may aid clinicians in designing optimal ARV combinations for the country. The secondary objective was to choose a suitable method for doing the analysis. Published primer sequences and inhouse methods for the different steps of the procedure were used, but this was changed to an established commercial system (viroseq) because of superior sensitivity and the fact that it is FDA approved. The study population consisted of 19 adult ARV-na·ive AIDS patients recruited from Tsepo House, and Medi Inn, Bloemfontein. The CD4 counts indicated that the immune systems of these patients were severely compromised, the highest count being 348 and the mean value 184 cells/mm3 whole blood. Therefore, according to the Department of Health's criteria, all of them qualified for ARV treatment. The viral loads were high, varying from 23 000 to >750 000 RNA copies/ml plasma. This demonstrates how people differ in their response to the viral infection. All patients were in the terminal stage of disease, yet displayed up to thirty-fold differences in viral load. After RT-PCR and sequencing, the sequences were trimmed to 99 codons of the full protease and the first 335 codons of the reverse transcriptase reading frames. These were translated to amino acids and used separately in phylogenetic analyses to study their relatedness to each other and to other isolates. A benefit of phylogenetic analysis is to indicate possible contamination of one sample by another as this will show tight clustering of some samples. The form and branch distances of the trees found clearly eliminated this possibility. The sequences were compared to 20 other South African isolates, randomly chosen from the Los Alamos (http://www.hiv.lanl.gov/content/index) HIV-1 sequence database. Both the protease and reverse transcriptase trees revealed that the Bloemfontein sequences do not differ appreciably from those found in the rest of the country, since they tend to diffuse through the tree rather than to cluster on their own. All the patients tested positive for subtype C, complying with the demographical data. Very few mutations were detected in the reverse transcriptase (RT) reading frame, although a mutation (K103N) which confers high level resistance against non-nucleoside RT inhibitors was found in one isolate. This may mean that a small percentage of patients may harbour a virus that is naturally resistant to these drugs. In the protease reading frame, mutations at nine amino acid positions have been designated primary or major resistance mutations. None of the primary mutations were found, although several secondary mutations (of lesser significance) contributing to reduced susceptibility (e.g., M361 and 193L) were found in 95% of our samples. This was not unexpected as these polymorphisms are extremely common in subtype C viruses. These genetic differences may be clinically relevant when considering long-term strategies for patients infected with nonS subtypes. As the public sector ARV rollouts gather momentum, the emergence of drugresistant isolates is sure to follow and the laboratory services must be geared to provide the backup needed by the clinicians to plan salvage therapy. This project has contributed to the baseline knowledge of the infected population enabling us to anticipate the emerging resistance mutations.