CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

425 CD4+ Tissue-Resident Memory T Cells Are an Important Reservoir for HIV Persistence Angela R. Wahl ; J.Victor Garcia University of North Carolina, Chapel Hill, NC, US Background: Tissue-resident memory T cells (T RM infection and the major cellular reservoir for HIV in peripheral blood, the contribution of CD4+ T RM

) are a recently described subset of memory T cells present in tissues. Although memory CD4+ T cells are a key target for HIV

to HIV infection and persistence is not known. Currently, studies of T RM

in humans

are limited to biopsied and cadaveric tissue; T RM the effect of HIV infection on the T RM do not recirculate and are virtually absent from peripheral blood. Therefore, despite their potential importance to HIV persistence, compartment has not been evaluated. To address this critical need, we utilized BLT humanized mice to investigate HIV infection of T RM in vivo and establish their contribution to HIV persistence. Methods: We performed a flow cytometric analysis of peripheral blood and cells isolated from the lymphoid (spleen, lymph nodes, and bone marrow) and non-lymphoid tissues (liver and lung) of uninfected BLT mice and BLT mice infected with a CCR5 tropic HIV-1 strain (JR-CSF, CH040 or THRO). Human memory T cells (CD3+CD45RO+) that expressed CD69 were classified as tissue-resident (T RM ). CD25 and HLA-DR negative T RM were classified as resting. T RM were isolated from the tissues of HIV-infected BLT mice by FACS and HIV-DNA and HIV-RNA quantitated by real-time PCR. Results: As observed in humans, T RM were present in lymphoid and non-lymphoid tissues of BLT mice but virtually absent from peripheral blood (p<0.0001). Resting CD4+ T RM were also present in lymphoid and non-lymphoid tissues of BLT mice (mean % of CD4+ T RM = spleen: 27%, lymph node: 61%, bone marrow: 28%, liver: 35%, lung: 33%). During HIV infection, CD4+memory T cells were significantly depleted from the tissues of BLT mice (p=0.0002). CD4+ T RM were also significantly decreased in HIV-infected BLT mice (p=0.0003) and preferentially depleted from non-lymphoid tissues (>85% decrease) in comparison to lymphoid tissues (27-55% decrease). Quantitative real-time PCR demonstrated the presence of HIV-DNA and HIV-RNA in CD4+ T RM of all tissues analyzed from HIV-infected BLT mice (mean: 3X10 4 HIV-DNA and 1X10 5 HIV-RNA copies per 10 6 cells). Conclusions: Human T RM are present in BLT mice and their tissue distribution recapitulates the human condition. Importantly, HIV infects and depletes CD4+ T RM in tissues. In addition, T RM contain significant levels of viral DNA and produce high levels of viral RNA. Currently, we are evaluating the contribution of CD4+ T RM to the latent HIV reservoir in vivo . 426 HIV-1 Reprograms Resting CD4 T Cells via Foxo1 and L-Selectin Suppression Background: Forkhead Box protein 01 (Foxo1, FKHR) is a transcription factor central to T cell biology. Foxo1 is necessary for the maintenance of T cell quiescence and for the expression of CD62L (L-selectin), an adhesion molecule that controls T cell migration into lymph nodes. TCR stimulation suppresses Foxo1 activity through PI3K/Akt-dependent phosphorylation of Foxo1 and its relocation to the cytoplasm. Foxo1 inactivation results in CD62L transcriptional repression and the activation or repression of a series of genes that regulate T cell activation, migration, survival, growth and differentiation. Methods: Purified peripheral blood and tonsil CD4 T cells were infected with HIV-1. Blood and tonsil cells were analyzed by flow cytometry. Blood cells were analyzed by ImageStream (Amnis) for Foxo1 and NF- κ B localization and TransAM ELISA (Active Motif) for Foxo1 sequence-specific DNA binding activity. Cellular and viral mRNA were analyzed by TaqMan qRT-PCR. Results: HIV-1 down-modulated CD62L on productively infected naïve and memory (TCM and TEM) resting CD4 T cells. Partial T cell activation was further revealed by upregulation of the mRNA and surface expression of CD69, a Foxo1-suppressed gene. HIV-1-induced CD62L down modulation was inhibited by the PI3K inhibitors LY294002 and PI-103, indicating that activation of the PI3K/Akt pathway contributes to this effect. ImageStream analysis demonstrated relocation of Foxo1 to the cytoplasm in productively infected resting naïve and memory CD4 T cells, and TransAM ELISA demonstrated loss of Foxo1 transactivation activity. A series of Foxo1 regulated mRNA were decreased or increased in productively infected resting CD4 T cells, including CD62L, IL-7r α , Myc, KLF2, CCR5, Fam65b, S1P 1 (EDG1), CD52 and Cyclin D2, indicating a profound reprogramming of these cells by HIV-1. Application of the Foxo1 inhibitor AS1842856 prior to or just after infection accelerated de novo viral gene expression and enhanced infected cell viability. Conclusions: HIV-1 suppression of Foxo1 activity is apparently a strategy to promote its replication in resting CD4 T cells. Dysregulation of Foxo1 may contribute to immune activation, viral load and HIV-1 pathogenesis. As Foxo1 is an investigative cancer therapy target, the ongoing development of Foxo1 interventions may assist the quest to specifically suppress or activate HIV-1 replication in vivo. 427 Measurements of Viral Transcription in Elite Suppressor CD4+ T Cells Christopher W. Pohlmeyer ; C. Korin Bullen; Greg Laird; Alyssa R. Martin;VictoriaWalker-Sperling; Stanley U. Chioma; Robert F. Siliciano; Joel Blankson Johns Hopkins University School of Medicine, Baltimore, MD, US Background: Elite suppressors (ES) are patients who control HIV replication without antiretroviral therapy. Prior studies have shown that the frequency of latently infected cells in these patients is much lower than patients on suppressive antiretroviral regimens. However the frequency of CD4+ T cells that express HIV-1 mRNA at baseline and following T cell stimulation is unknown. In this study we compared HIV-1 transcription levels in CD4+ T cells from chronic progressors (CPs) on suppressive antiretroviral regimens and ES. Methods: To measure intracellular HIV-1 mRNA, we isolated CD4 T cells from PBMCs of ES and CPs. Replicates of 5x10 6 cells were stimulated with PMA/ionomycin or DMSO for 24 hours. The cells were collected and lysed in Trizol for RNA extraction and subsequent quantification by qPCR. RNA from supernatants were collected and measured for released HIV-1 mRNA. Results: A comparison of cell associated HIV-1 mRNA in CD4+ T cells of HAART-suppressed CPs and ES shows that ES have significantly less HIV-1 mRNA per 5x10 6 cells before stimulation. HIV1 mRNA was uniformly detected in CPs, but was present at very low levels in just 2 of 8 ES (p>0.05). When 5x10 6 CD4+ T cells were stimulated with PMA/ ionomycin, the levels of cell-associated HIV-1 mRNA increased in 4 of 7 ES. Additionally, when measuring HIV-1 mRNA levels in culture supernatant following stimulation of 5 x 10 6 CD4+ T cells with PMA/ionomycin, we detected release of virus from just 2 of 8 ES compared to 5 of 5 CPs. When more replicates were analyzed, viral release was seen in 4 ES. 2 of these patients showed positivity in 1 of 5 replicates (25x10 6 cells), releasing approximately 3,000 and 5,000 copies HIV-1 RNA each. Poisson statistics suggests an 89% chance that the signal observed reflects a single cell releasing virus, and our HIV-1 mRNA measurements fit with current estimates of the burst size of an infected CD4+ T cell. Conclusions: In the present study, we demonstrate that the baseline levels of cell associated HIV-1 mRNA in ES are significantly lower than those observed in CPs per 5x10 6 cells. However an increase in viral transcription following T cell stimulation was observed. These results further characterize the size of the latent reservoir in ES and confirm earlier studies that suggested that some of these patients are infected with replication-competent virus. BenjaminTrinité 1 ; Chi N. Chan 1 ; Caroline S. Lee 1 ; Joy Folkvord 2 ; Elizabeth Connick 2 ; David N Levy 1 1 New York University, New York, NY, US; 2 University of Colorado Anschutz Medical Campus, Aurora, CO, US

Poster Abstracts

300

CROI 2015