CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

290 SKEWED DISTRIBUTION OF HIV-2 RESERVOIR WITH LIMITED INPUT OF CENTRAL MEMORY T CELLS Assia Samri 1 , Charlotte Charpentier 2 , Mélanie Bertine 2 , Mariama Diallo 1 , Sophie Even 1 , Sophie Matheron 3 , Rodolphe Thiébaut 4 , Brigitte Autran 5 , Francoise Brun-Vezinet 3 , for the ANRS CO5 IMMUNOVIR-2 Study Group 1 Pierre and Marie Curie Univ, Paris, France, 2 INSERM, Paris, France, 2 Bichat-Claude Bernard Hosp, Paris, France, 4 INSERM, Bordeaux, France, 5 Pitié-Salpêtrière Hosp, Paris, France Background: HIV-2 infection is characterized by a low pathogenicity and a low virus production. We tested the hypothesis of a limited distribution of the HIV-2 reservoir among central-memory CD4 T cells (TCM), similarly to models of HIV-1 functional cure or of non pathogenic SIV infection. Methods: 14 ARV-naïve patients with non-progressive infection included in the ANRS CO5 HIV-2 Cohort were assessed; their median CD4 cells/mm3 was 966 [IQR: 820-1216]. Subpopulations were sorted into CD3-CD4+monocytes, CD25+CD69+HLADR+ activated, CD25-CD69-HLADR- resting CD4+ T cells and among those into naïve (TN), TCM, transitional-memory (TTM) and effector-memory (TEM) cells. Cell-associated HIV-2-DNA was isolated from sorted subsets with QIAamp DNA Mini or Micro kit® (Qiagen) according to cell numbers. HIV-2 DNA was quantified using a real-time PCR assay with a limit of detection (LOD) 95% of 3 c/PCR and a limit of quantification (LOQ) of 6 c/PCR. HIV-2 reactivation assays were performed by CD8- T cells culture with anti-CD3+CD28+IL-2+IL-7 for 30 days. Results: Plasma viral load (pVL) was <40 c/mL in 12 patients, among them 4 had a positive ultra-sensitive pVL above 1 c/mL (IQR=1-12). Median total HIV-2 DNA in PBMC was above the LOQ in 13 patients with a median of 1.94 log10 c/106 PBMC (IQR=1.53-2.13). After sorting, HIV-2 DNA was undetectable among monocytes, and above the LOQ only in TTM from 4 patients (median=2.25 [IQR: 1.99-2.94] log10 c/106 cells) and in TCM from 1 patient (1.75 log10 c/106 cells). HIV-2 DNA levels were above the LOD in 3, 12, 9 and 10 patients in TN, TCM, TTM and TEM, respectively. When integrating subsets proportions, the median contribution of TN, TCM, TTM and TEM to the HIV-2 reservoirs was 0%, 33%, 46% and 8%, respectively. The HIV-2 DNA levels in TTM were positively correlated to those in PBMC (p=0.008; r=0.67). After CD8- T cells reactivation, HIV-2 RNA was detected in 3 of the 11 tested samples with the highest HIV-2 DNA values that were quantified in TTM (n=2) or TCM (n=1). Conclusion: Overall, these HIV-2-infected patients had low circulating HIV-2 reservoirs that were quantifiable in only 5 of the 14 patients tested, mainly distributed in TTM and reactivable in vitro in only 3 of these 5 patients. These results confirm the hypothesis of a limited reservoir in TCM, thus supporting the concept of the relative protection of central- memory T cells as an attribute of low pathogenicity models of HIV/SIV infection. 291 ALTERED STABILITY OF HIV-INFECTED MEMORY CELLS FOLLOWING VERY EARLY ART Louise Leyre 1 , Jintanat Ananworanich 2 , Claire Vandergeeten 3 , Eugene Kroon 4 , Nitiya Chomchey 5 , Siriwat Akapirat 6 , Merlin L. Robb 2 , Nittaya Phanuphak 7 , Nicolas Chomont 8 , for the RV254/SEARCH010 and RV304 Study Groups 1 Cntr de Rsr du Cntr Hosp de l’Univ de Montréal, Montréal, Québec, Canada 2 US Military HIV Rsr Prog, Silver Spring, MD, USA, 3 Vaccine and Gene Therapy Inst, Port St Lucie, FL, USA, 4 SEARCH, The Thai Red Cross AIDS Rsr Cntr, Bangkok, Thailand, 5 Armed Forces Rsr Inst of Med Scis, Bangkok, Thailand, 6 Thai Red Cross AIDS Rsr Cntr, Bangkok, Thailand, 7 Univ de Montréal, Montreal, QC, Canada Background: Initiation of ART during primary HIV infection restricts the size of the HIV reservoir. However, less is known about the cell subsets in which HIV integrates and persists during the earliest phase of HIV infection. Methods: In participants who underwent leukapheresis, integrated HIV DNA was quantified by Alu-PCR in sorted naïve (TNA), central (TCM), transitional (TTM) and effector (TEM) memory cells from infected Thai individuals who started ART within the first weeks of infection (Fiebig stages I-V based on HIV clinical assays: HIV RNA, p24, HIV antibodies). Participants who initiated ART during chronic infection were used as controls. Results: In acutely infected individuals, there was an increase in the frequency of infected TCM, TTM and TEMwhen transitioning from Fiebig I, II and III stages. All 3 memory subsets were equally infected within each Fiebig group, while TNA cells were rarely infected. Integrated HIV DNA was detected in at least one memory subset in 21/22 Fiebig II-V individuals (95%) and in all subsets from chronically infected controls (n=3). In contrast, 4/7 Fiebig I individuals (57%) were devoid of HIV DNA in all subsets (limit of detection: 10 copies/10^6 cells). The frequency of infected cells in each subset was strongly correlated to plasma viral load (p<0.0001 for TCM, TTM and TEM cells). After 24-96 weeks of ART, the frequency of cells with integrated HIV DNA decreased in TCM, TTM and TEM cells from all acutely treated individuals, whereas it remained stable in individuals treated during chronic infection. With the exception of TTM cells from 1 participant, all Fiebig I individuals showed undetectable levels of HIV DNA in all subsets. Importantly, the earlier ART was initiated, the steeper was the decay in integrated HIV DNA in TCM (505, 25, 8, and 2-fold decrease (FD) in Fiebig II, III, IV-V and chronic), TTM (279, 25, 5 and 2-FD) and TEM cells (50, 35, 3 and 1-FD). Conclusion: Memory cells harbouring integrated HIV DNA rapidly accumulated as plasma viral load increased. Whereas ART initiated in chronic infection had no impact on the amount of integrated HIV DNA in memory subsets, initiation of ART in acute infection decreased the frequency of infected TCM, TTM and TEM cells. This decrease was more pronounced in those who started very early, leading to undetectable levels of infected cells in Fiebig I individuals. These results suggest that the majority of memory cells infected during acute infection are short-lived. 292 CLONAL EXPANSION OF GENOME-INTACT HIV-1 IN FUNCTIONALLY POLARIZED T-CELL SUBSETS Guinevere Q. Lee 1 , Nina Orlova 1 , Erik Serrao 2 , Xiaoming Sun 1 , Navin F. Chowdhury 1 , Eric Rosenberg 3 , Alan N. Engelman 2 , Xu G. Yu 1 , Mathias Lichterfeld 4 1 Ragon Inst of MGH, MIT and Harvard, Cambridge, MA, USA, 2 Dana-Farber Cancer Inst, Boston, MA, USA, 3 Massachusetts General Hosp, Boston, MA, USA, 4 Brigham and Women’s Hosp, Boston, MA, USA Background: HIV-1 remains an incurable disease due to long-lasting reservoirs of infected cells that are unaffected by suppressive antiretroviral therapy. However, the mechanisms that maintain long-term stability of such viral reservoirs are not well understood. Here, we used a near-full-genome deep sequencing approach to characterize HIV-1 DNA content in highly purified CD4 T cell subsets with distinct functional polarization. Methods: Highly purified CD4 T cells secreting IFN-g (Th1), IL-4 (Th2), IL-9 (Th9), IL-17 (Th17), cytokine-negative control cells and unstimulated autologous CD4 T cells were sorted or enriched from large volumes of PBMC obtained from chronically-infected ART-suppressed HIV-infected patients. Total DNA extracted from individual cell populations was subjected to HIV-1 DNA quantification by ddPCR amplification of 5’LTR-gag (127bp) or to single-genome nested PCR with primers spanning near-full-genome HIV-1 (HXB2 coordinates 638-9632), followed by sequencing of individual products with Illumina MiSeq. Results: Little variation was found between total HIV-1 gag DNA levels among the different polarized T cell populations in seven study subjects. In subsequent full-genome sequencing assays from six additional patients, 894 HIV-1 DNA sequences were obtained; only 7% (66/894) were genome-intact. Defective genomes had: large deletions over 1000bp 70% (623/894), APOBEC3G/3F-associated hypermutations 21% (186/894), deletions over 6bp in packaging signal 1% (9/894), and premature stop codons in gag/pol/ env 0.3% (3/894). 352 HIV-1 DNA sequences were obtained from polarized CD4 T cell populations, of which 28 were sequence-intact. The proportion of genome-intact sequences relative to the pool of all viral species was highest in Th1 cells 15% (18/117), followed by cytokine-negative-Th 6% (6/99), Th9 4% (2/52), Th2 3% (1/34), Th17 2% (1/50), (Chi-square p=0.0003 for Th1 vs non-Th1 cells). In six distinct cases, groups of two to seven genome-intact viruses that shared 100% sequence identity over all 8995bp were detected from autologous polarized Th1 cells and/or unfractionated CD4 T cells, suggestive of clonal expansion of cells harboring intact provirus. Conclusion: Th1-polarized cells seem to represent an important reservoir of cells harboring genome-intact HIV-1 sequences during antiretroviral therapy. Proliferation of such cells may contribute to maintaining and expanding the HIV-1 infected cell pool.

Poster and Themed Discussion Abstracts

CROI 2017 115