CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

395 Latent HIV-1 Reactivation and Lysosomal Destabilization Synergize to Host Cell Death Metodi Stankov 1 ; Christina Suhr 1 ; Hazel Lin 1 ; Diana Panayotova-Dimitrova 2 ; Christine Goffinet 1 ; Georg Behrens 1 1 Hannover Medical School, Hannover, Germany; 2 Mannheim Clinic of University of Heidelberg, Mannheim, Germany

Background: Modern strategies for purging the latent HIV-1 reservoir use pharmacological reactivation followed by elimination of cells that reactivated viral latency either by the virus or/and patient’s immune system ‘shock and kill’. Unfortunately, most of the current approaches demonstrated only partial reactivation and limited reduction of the reservoir. We aimed to develop a pharmacological strategy targeting cellular pro-survival mechanisms in order sensitize and eliminate host cells with even low or incomplete levels of HIV-1 reactivation. Methods: We used in vitro models of HIV latency (J-lat full length clones, ACH2 and U1) and the latency-reversing agent panobinostat (Pan). We targeted the pro-survival pathway of autophagy and lysosomal integrity by the lysosome-destabilizing agents (LDA) chloroquine (CQ) or mefloquin (Mef). Autophagy and lysosomal pathways were monitored using flow cytometry, fluorescent microscopy and western blotting. Latent HIV-1 reactivation was confirmed through HIV-1-GFP expression, intracellular p24 abundance, and secretion Results: Our results confirmed that CQ and Mef suppress autophagic flux and destabilize lysosomal membrane. Pan mediated reactivation led to specific cell death, which closely correlated to the levels of reactivation (R 2 =0.8821; p=0.0178). At Pan concentrations associated with HIV-1 reactivation, but only little host cell death, co-treatment with CQ increased the death rate about four fold (Control 1.71%; Pano 17.4%; CQ 10% and Pano+CQ 63.2% p<0.0001). These synergistic and selective effects were further enhanced and were more rapid when we combined lysosomal disruption by Mef with boosted lysosomal activity by partial nutrient reduction (Control 9.2%; Pano 23.6%; Mef 8.8%, RPMI 13.3 and Pano+Mef+RPMI 74.7% p<0.0001). Conclusions: We conclude that interference with the pro-survival pathway of autophagy at the level of lysosomal stability efficiently supports host cell death upon incomplete HIV-1 reactivation. These in vitro results suggest that combined pharmacological intervention may assist to eradicate latently HIV-1 infected cells. 396 Noninduced Proviral Genome Characterization in Perinatal HIV Infection Kaitlin Rainwater-Lovett 1 ; Carrie Ziemniak 1 ; DouglasWatson 2 ; Katherine Luzuriaga 3 ; Priyanka Uprety 4 ;Yahui Chen 1 ;Ya-Chi Ho 1 ; Deborah Persaud 1 1 Johns Hopkins University School of Medicine, Baltimore, MD, US; 2 University of Maryland, Baltimore, MD, US; 3 University of Massachusetts Medical School, Worcester, MA, US; 4 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, US; 5 Johns Hopkins University School of Medicine, Baltimore, MD, US Background: The latent HIV reservoir was recently estimated as 60-times larger than previously thought due to replication-competent, non-induced proviral genomes (NIPG), creating a major impediment for cure. The frequency of circulating cells harboring proviral DNA approaches the limit of detection of ultrasensitive droplet digital PCR (ddPCR) in perinatally-infected individuals who initiated antiretroviral therapy (ART) before 6 months of age, suggesting early ART affects reservoir establishment. The extent to which NIPG contribute to HIV persistence in early treated perinatal infection is unknown. Methods: P24-negative culture wells each containing one million resting CD4+ T cells from 21 perinatally-infected individuals aged 4-21 years who were virologically-suppressed by 6 months of age and for a median of 13 years (IQR: 10.5, 16) were screened for HIV rt or gag by nested PCR. The gag gene was sequenced from near full-length HIV amplicons at a clonal level as described by Ho et al. Total HIV DNA concentrations were quantified with ddPCR. Values below the limit of detection were set to the limit of detection. Results: The median proviral burden was 15.2 copies HIV DNA/million resting CD4+ T cells (IQR: 4.8, 33.2). Of 426 million cultured resting CD4+ T cells, 9 wells yielded inducible provirus from 5 individuals (24%), resulting in a median of 0.067 infectious units per million cells. Of the remaining 389 p24-negative culture wells, 81 wells (19%) from 16 individuals yielded 281 NIPG gag sequences; of which, 188 (67%) were intact, 62 (22%) were hypermutated, and 31 (11%) contained major deletions. Individuals with induced provirus had 25% (95% CI: 13%, 38%) more intact and 8% (95% CI: 6%, 27%) fewer hypermutated gag sequences among their NIPG than those without induced provirus (p=0.001 and 0.021). The proportion of gag sequences with deletions did not differ between those with (4%) and without (12%) induced provirus (p=0.160). The proportion of 41 near-full length sequences obtained from the 188 NIPG with intact gag sequences with major deletions did not differ between those with (70%) and without (65%) induced provirus (p=1.00). Conclusions: Early treated, perinatally-infected individuals readily harbor NIPG. Individuals with non-inducible provirus had few intact and many hypermutated NIPG, suggesting NIPG analysis in combination with quantitative co-culture are important measures for identifying perinatally-infected children who may benefit from immunotherapeutic approaches. 397 Multiple Rounds of T-Cell Activation Induce Additional HIV-1 From the Latent Reservoir Background: HIV-1 has the ability to establish latency within resting memory CD4+ T cells, generating a major barrier to eradication. The current gold standard assay to measure the size of the reservoir is the viral outgrowth assay (VOA), which uses a lectin, PHA, to induce global activation of resting CD4+ T cells from patients on antiretroviral therapy (ART). Previous studies have shown that the size of the latent reservoir may be 60-fold larger than originally estimated due to the presence of proviruses that are genetically intact, but are not detected in the VOA following a single round of T cell activation. We hypothesize that multiple rounds of T cells activation will induce additional outgrowth of viruses from the latent reservoir, thus providing a better indication of the true size of the reservoir. Methods: Resting CD4+ T cells are isolated from patients on ART and plated at 200,000 cells per well in a VOA. After 8 days of culture with PHA, half of the volume from each initial well is split into new wells for a second round of PHA stimulation, performed in the same manner as the first round. After another 8 days, half the volume from these wells is split again into replicate wells, and so forth for a total of 4 stimulations. Viral outgrowth is detected 21 days following each respective round of PHA stimulation by both a p24 ELISA and a qPCR that detects full-length HIV-1 mRNA. Results: Our results demonstrate that additional viral outgrowth is observed after additional rounds of activation. In some patients, most of the virus is induced in the first and second rounds of activation, while in other patients, each round induces additional latently infected cells to produce virus. The data demonstrates that one round of maximum T cell activation is not sufficient to induce all proviruses that are capable of producing replication-competent virus. Conclusions: We conclude that more than one round of T cell activation was able to induce additional replication-competent viruses, indicating the possibility that viral reactivation is governed by stochastic processes and that the latent viral reservoir may be larger than initially thought. Nina N. Hosmane ; Adam A. Capoferri; Robert F. Siliciano Johns Hopkins University School of Medicine, Baltimore, MD, US

Poster Abstracts

288

CROI 2015