Characterisation of HIV-1 Envelope features of breakthrough infections from the CAPRISA 004 Microbicide Trial

The CAPRISA 004 trial demonstrated the safety and a 39% efficacy of a 1% tenofovir (TFV) gel for the prevention of HIV-1 acquisition in young African women. It was subsequently shown that women assigned to the TFV arm who became infected had higher viral loads, slower anti-HIV-1 antibody avidity maturation, and higher Gag-specific IFN-γ+ CD4+ T cell responses; although replication capacity, as measured by Gag-Pro recombinant viruses, did not differ between arms. We thus aimed to investigate if there were differences in Envelope function, or TFV susceptibility, which may be selected for during transmission in those who became infected despite being assigned to the TFV arm. Viruses from 39 out of 48 recently HIV-1 infected individuals from the trial (matched on time post-infection and the presence of protective HLAs) were isolated. Isolate env genes were sequenced using a single genome amplification approach and were compared to plasma sequences from the same time-point. To evaluate phenotypic characteristics of env, inhibition assays were performed using the following inhibitors: tenofovir, maraviroc, T20, PSC-RANTES and anti-CD4 antibody clone SK3. In addition, envs for 19 participants were cloned and used to generate pseudoviruses which were evaluated for entry efficiency. Viral isolates were identical or very similar to viruses in circulation in vivo; however had a lower diversity, indicating that they were representative of in vivo virus but did not reflect the entire quasispecies in plasma. The TFV arm viruses were not more resistant to TFV than those in the placebo arm. A comparison of variable loop characteristics, distance to a consensus representative of viruses circulating in the region, and sensitivity to inhibitors or entry efficiencies of the viruses, also found no difference in genotypic nor phenotypic properties between study arms. When assessing the impact of viral phenotype on markers of disease progression, it was found that sensitivity to inhibitors did not contribute to VL or CD4+ count in this cohort. To evaluate envelope in isolation of the rest of the genome, pseudoviruses were generated from 11 participants. We found that PSV entry efficiency did not correlate with VL at isolation, 3 months post-infection and set-point, or with CD4+ counts at set-point. However, pseudovirus inhibitor sensitivities were significantly different to those of isolates for the inhibitors T20, anti-CD4 antibody SK3 and PSC-RANTES. Overall, the isolate env genotypic and phenotypic characteristics investigated in this study did not differ between trial arms. Interestingly, pseudoviruses showed significant differences in their sensitivity to entry inhibitors when compared to their corresponding isolate, highlighting the importance of caution when interpreting data from in vitro studies, and motivates for further evaluation of in vitro models.

Grantor dc:publisher.institution

Division of Virology

Year dc:date.issued

2016

Author dc:creator

• Ismail, Sherazaan Dineo

Advisors dc:contributor.advisor

• Williamson, Carolyn
• Selhorst, Philippe

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/22870

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/22870