The Influence of HIV-1 Subtype C LTR Genotype on Latency Potential

The persistence of latent viral reservoirs, that are insensitive to antiretroviral therapy (ART), remains the greatest barrier to HIV-1 eradication. The role that viral factors play in HIV-1 latency establishment and maintenance is poorly understood, and characterisation of these factors is imperative for the development of curative strategies or interventions that could lead to HIV-1 remission in infected individuals. Subtype level genotypic variation within regulatory elements of the HIV-1 promoter, the long terminal repeat (LTR), has been shown to influence latency establishment in in vitro models. We investigated the influence of inter-participant subtype C LTR genotypic variation on the establishment of latency in a dual reporter HIV-1 plasmid model and evaluated potential correlates of this latency potential. Long terminal repeats from 11 ART-naïve, acutely subtype-C infected women in the CAPRISA 004 cohort from Durban, South Africa were cloned into an HIV-1-expressing vector (pRGH) used to generate pseudovirions following HEK293T transfection. Pseudoviruses harboured a gag-eGFP gene under the control of the participant LTR, allowing measurement of active replication, and an mCherry gene under the control of a constitutive CMV promoter allowing measurement of viral integration. Latency potential was expressed as the ratio of mCherry only (latent) to eGFP and mCherry (active replication), as measured by flow cytometry after infection of Jurkat E6-1 and CEM.NKR CCR5+ cell lines before and after T cell activation with PMA/Ionomycin. A panel of LTRs cloned into a pGL4.10 luciferase expression vector were used to measure basal LTR expression and Tat-induced LTR expression. All LTR sequences were classified as subtype C, with an average inter-participant pairwise DNA distance of 7.6%. The median basal LTR activity was approximately two times higher than that of the BaL isolate (interquartile range: 1.38-2.14), and Tatinduced activity approximately nine times higher than that of BaL (interquartile range: 6.16-10.33). We observed consistently greater proportions of latently infected cells than actively infected cells. In Jurkat E6-1 cells, the median latent:active infection ratio was 1.97 (range 0.86-2.83; three experiments). Latency was reversible in a proportion of cells as the median latent:active infection ratio decreased to 0.55 (range 0.46-0.78). The latent:active ratio was unchanged, post-stimulation, in CEM.NKR CCR5+ cells and was therefore found not to be a suitable cell-line for the model. Latency potential did not correlate with basal or Tat-induced activity (Spearman correlation tests, basal p=0.25, r=-0.38, Tat-induced p=0.42, r=-0.27). The DNA distance in characterised functional sites from consensus did not correlate with latency potential (Spearman correlation test p=0.67, r=0.14). Our data suggest that HIV-1 LTRs have intrinsic properties which influence latency potential and the proportion of latently infected cells early post-infection. However, since differences were independent of basal and Tat-induced LTR activity, other factors such as regulatory element interaction and the efficiency of recruitment of molecules responsible for establishing latency, such as histone modifiers, may play a role.

Grantor dc:publisher.institution

Department of Pathology

Year dc:date.issued

2018

Author dc:creator

• Doolabh, Deelan Sudhir

Advisors dc:contributor.advisor

• Abrahams, Melissa-Rose
• Williamson, Carolyn

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/29620

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/29620