CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

333 CYTOSINE METHYLATION OF THE HIV LTR WHEN STARTING ART IN ACUTE VS CHRONIC INFECTION Sarah LaMere 1 , Sara Gianella 1 , Antoine Chaillon 1 , Christina Huynh 1 , Susan J. Little 2 , Davey M. Smith 2 1 University of California San Diego, La Jolla, CA, USA, 2 University of California San Diego, San Diego, CA, USA Background: Targeting epigenetic mechanisms of HIV latency could be important for eradication. Cytosine methylation of mammalian promoters is one mechanism that represses transcription, and previous studies showed that it accumulates in a CpG island of the 5’ LTR in HIV latency models. Studies from clinical samples are rare and have produced conflicting results. Methods: We compared the % cytosine methylation in 13 HIV-infected persons: 6 who started ART during early HIV infection within 5 months from the estimated date of infection and were sampled within 2 years, and 7 from a cohort who began ART during chronic infection. Both cohorts were composed of caucasian MSMmen with ages ranging from 21-50Y, all with undetectable viral loads. DNA from PBMCs was extracted and bisulfite converted, and the proximal LTR was amplified and sequenced. The entire LTR from unconverted DNA was also sequenced and subjected to base composition analysis (BCA). BCA was also performed on sequences from the LANL database from acute (Fieberg stage 2) and chronic (Fieberg stage 5-6) infection. One-tailed Mann-Whitney p-values were used for all statistical tests. Results: The %methylated non-CpG cytosines exceeded the %methylated CpG for most (10/13) individuals. Most non-CpG cytosines existed in CTG motifs. There was a significantly higher % CpG methylation in individuals treated early compared to those treated during chronic infection (p=0.036). There was no difference in non-CpG cytosine methylation between the two groups (p=0.38). Base composition analysis showed that the chronic group had a higher observed to expected ratio (O/E) of CpG residues in the LTR than the acute group (p=0.067), while the O/E ratio of CTG trinucleotides was not different (p=0.42). Sequence analysis of the LTR from acute vs chronically infected untreated individuals from the LANL database yielded similar results, with the chronic group having a higher O/E of CpG than the acute group (p=0.0073). Conclusion: CpG %methylation was lower in the chronic group, and this could be due to an increased CpG frequency in the LTR of chronically infected individuals. Non-CpG methylation was recently reported in mammalian genes, but this is the first report of its presence in HIV provirus. In this context it is seen most often in CTG trinucleotides, which exhibit no difference between acute and chronic groups. Non-CpG methylation is a novel and potentially important mechanism in the brain and should be explored further in latent reservoir tissues. 334 IMPACT OF TREATMENT INTERRUPTION ON HIV RESERVOIRS AND IMMUNOLOGIC PARAMETERS Katherine Clarridge 1 , Jana Blazkova 1 , Kevin Einkauf 2 , Marissa Hand 1 , Mary Petrone 1 , Eric Refsland 1 , Jesse Justement 1 , Victoria Shi 1 , Erin Huiting 1 , Catherine Seamon 1 , Susan Moir 1 , Michael Sneller 1 , Mathias Lichterfeld 2 , Tae-Wook Chun 1 1 NIAID, Bethesda, MD, USA, 2 Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA Background: Suppression of human immunodeficiency virus (HIV) and improvements in health outcomes have been achieved in infected individuals receiving antiretroviral therapy (ART). Nonetheless, the vast majority will experience plasma viral rebound upon cessation of therapy, underscoring the need for developing additional therapeutic strategies that allow durable virologic remission following the interruption of ART. Analytical treatment interruption (ATI) is an essential component of future clinical trial design to determine the efficacy of immune-based therapies in suppressing and/or eradicating HIV. Here, we investigated the effect of short-term ATI on the HIV reservoir and immunologic parameters in HIV-infected individuals. Methods: In depth immunologic and virologic analyses were conducted using clinical specimens obtained from ten HIV-infected individuals prior to ART discontinuation, during ATI, and following reinitiation of ART. The effect of ATI on the HIV reservoir was determined by measuring the level of HIV proviral DNA, cell-associated HIV RNA, and replication-competent HIV in CD4+ T cells. Characterization of intact and defective near full-length HIV proviral DNA was performed using single-genome, next-generation sequencing. Examination of immunologic parameters included longitudinal analyses of CD4+ and CD8+ T cells as well as cytokine and inflammation markers in plasma. Expression of activation and exhaustion markers was analyzed using flow cytometry.

effects or immune clearance. Immune recognition requires the presentation of HIV peptides by MHC at the cell surface following LRA-induced HIV protein expression and processing. Some of the LRA that showed promising in vitro activity but limited efficacy in clinical trials such as HDAC inhibitors and PKC agonists have not been evaluated for their effects on HIV antigen processing and presentation by MHC. Methods: We assessed the hydrolytic activities of cytosolic peptidases in live primary CD4 T cells treated with HDACi or PKCa using a fluorometric assay. The effects on antigen processing were studied by an in vitro degradation experiment of synthetic HIV peptides in cytosolic extracts of mock- or LRA- treated CD4 T cells. Finally, we analyzed the effects of LRA treatment of PBMCs on the MHC-peptidome by direct acid elution from the cell surface and mass spectrometry analysis. Results: Treatment with HDACi significantly decreased the hydrolytic activities of the proteasome and cytosolic aminopeptidases in primary CD4 T cells by up to 0.8-fold, whereas treatment with PKCa increased these activities by up to 4.9-fold. Consequently, the degradation of HIV synthetic peptides in cytosolic extracts from LRA-treated primary CD4 T cells yielded altered degradation patterns with 4-8 sites significantly changed in a 35-mer and 2-4 in a 24-mer depending on the LRA. Additionally, there were changes in the generation of known epitopes as well as other potential HLA-binding peptides. Treatment with LRA partly changed the MHC-peptidome of PBMCs, showing common peptides as well as different numbers and location of peptides derived from some common source proteins. Conclusion: The observed changes in the hydrolytic activities of cytosolic peptidases in primary CD4 T cells upon LRA-treatment result in altered peptide degradation and a modified MHC-peptidome. The effects are both drug- and sequence-dependent. These results suggest that productively infected and LRA-reactivated CD4 T cells might present different HIV epitopes to the immune system, thus requiring different CD8 T cells responses for effective clearance of reservoirs. 332 SPONTANEOUS REACTIVATION OF LATENT HIV-1 PROMOTERS IS LINKED TO THE CELL CYCLE Yik Lim Kok 1 , Stefan Schmutz 2 , Anne Inderbitzin 1 , Kathrin Neumann 1 , Audrey Kelley 1 , Lisa Jörimann 1 , Mohaned Shilaih 1 , Valentina Vongrad 1 , Roger Kouyos 1 , Huldrych F. Günthard 1 , Christian Berens 3 , Karin Metzner 1 1 University Hospital Zurich, Zurich, Switzerland, 2 University of Zurich, Zurich, Switzerland, 3 Friedrich Loeffler Institute, Jena, Germany Background: Long-lived latently HIV-1-infected cells represent a major barrier to eradication of the virus. Studies examining these cells are hindered by their low frequencies in HIV-1-infected individuals. We, therefore, developed a dual- fluorescence HIV-1-based vector containing a pair of genetic insulators flanking a constitutive fluorescent reporter gene cassette to study HIV-1 latency. Methods: SUP-T1 cells were transduced with our VSV-G-pseudotyped genetic insulators-containing vector. The protective effects of the genetic insulators were demonstrated through long-term (up to 394 days) stable fluorescence profiles in cell populations representing both active and latent HIV-1 infections. The capability of our genetic insulators-containing vector to reproduce HIV-1 integration site patterns was confirmed by the analysis of various features in 1,941 vector integration sites derived from two independent transductions. With our verified HIV-1 infection model, monoclonal cells representing latent infections were sorted and the reactivation potentials of latent HIV-1 promoters at various integration sites were examined. Spontaneous reactivation of HIV-1 promoters in these cells were observed and factors contributing to this phenomenon were investigated by means of transcriptomic and cell cycle analyses. Results: The reactivation potentials of latent HIV-1 promoters are influenced by both vector integration sites and integrity of the vector genomes. In latent cell clones exhibiting a small subpopulation of cells with spontaneously reactivated HIV-1 promoters, higher expression levels of genes involved in cell cycle progression are observed in these cell subpopulations compared to the majority of cells with HIV-1 promoters that remained latent. Consistently, larger fractions of spontaneously reactivated cells are in the S and G2 phases of the cell cycle. Furthermore, genistein and nocodazole treatment of these cell clones, which halt cell cycling in the G2 phase, resulted in a 1.4–2.9-fold increase in spontaneous reactivation. Conclusion: Our stable HIV-1 infection and latency model reveals that the spontaneous reactivation of latent HIV-1 promoters is linked to the cell cycle.

Poster Abstracts

CROI 2018 116