CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

459

Role of HIV Integration Site on Clonal Expansion of Infected Cells and Maintenance of Latency Virender K Pal 1 , Frauke Muecksch 2 , Ali Danesh 3 , Marie Canis 1 , Tan T. Huynh 1 , Theodora Hatziioannou 1 , R. Brad Jones 3 , Guinevere Q. Lee 3 , Paul D. Bieniasz 1 1 The Rockefeller University, New York, NY, USA, 2 Heidelberg University, Heidelberg, Germany, 3 Weill Cornell Medicine, New York, NY, USA Background: Latent reservoirs of HIV-1 are the major barrier to achieving a cure. Because latently infected cells are rare in humans there is a paucity of understanding of how latent reservoirs survive and expand, escape immune clearance, and reactivate upon ART cessation. The site of HIV-1 proviral DNA integration into the host genome could (i) determine the transcription status of the provirus and (ii) influence the expansion of infected clones during ART. In this study we are interested in the characterizing proviral integration landscape associated with clonal expansion and latency of HIV-1 infected cells in an in vivo model. Methods: Primary memory CD4+ T cells isolated from healthy human donors were infected with dual reporter tagged, defective HIV-1 to generate large populations (millions) of human memory CD4+ T cells each presumptively carrying a single reporter provirus at a distinct integration site. These cells were grafted into NSG mice and analyzed for proviral transcriptional dynamics over a period of time. HIV-1 integration sites were characterized by PCR based amplification of host-viral junction and next-generation sequencing. Results: Our preliminary analysis of HIV-1 integration sites in these long term persistent HIV-1 infected cells led to the identification of 1777 unique integration sites (UIS). Consistent with previous reports, HIV-1 integration was favored in genic regions (60%) of human chromosomes as compared to non-genic regions (40%). Genic HIV-1 integrations were primarily in introns (94%). We found 84 genes that harbored HIV-1 integrations in multiple different clones and in most cases these integrations were clustered within a single intron, suggesting possible hotspots within a gene for integration or for clone survival. Two different mice had clones with HIV-1 integrants in the same 25 genes suggesting a growth or survival advantage for clones harboring HIV-1 integration in these genes. Overall, 641 UIS were associated with expanded clones persisting in vivo and clonally expanded cells were more likely to have HIV-1 integration inside genes than cells where clonal expansion was not detected. The genes in which HIV-1 integration was associated with clonal expansion belonged to pathways with roles in cell component biogenesis, mitosis, regulation of cell cycle, and chromatin remodeling. Conclusion: Clonal expansion of HIV-1 infected cells in this model system is associated with proviral integration in a set of genes which provide selective growth advantage in vivo. HIV Integration Site Features Associated With Persistence of Infected Cells Under CD8 Pressure Noemi L Linden 1 , Alexander McFarland 2 , Ali Danesh 1 , John Everett 2 , Scott Sherrill-Mix 3 , Carole Lee 2 , Chanson J. Brumme 4 , Dennis Copertino 1 , Zabrina Brumme 5 , Itzayana Miller 1 , Tan T. Huynh 1 , aoife Roche 2 , Jared Weiler 1 , Frederic D. Bushman 2 , R. Brad Jones 1 1 Weill Cornell Medicine, New York, NY, USA, 2 University of Pennsylvania, Philadelphia, PA, USA, 3 Michigan State University, East Lansing, MI, USA, 4 British Columbia Centre for Excellence in HIV/AIDS, Vancouver, Canada, 5 Simon Fraser University, Vancouver, Canada Background: HIV integration sites (IS) in elite controllers and after long-term ART are highly clonal and skewed towards transcriptionally silent genomic loci. The mechanisms contributing to this IS landscape remain unclear, hampering the design of cure therapeutics. Using a participant-derived xenograft mouse model, we observed that CD8+ T-cell pressure was sufficient to recapitulate key features of this IS profile, including enrichment for large clones and integrations distal from genes. Here, we expand our dataset and assess whether prolonged elite controller CD8+ T-cell pressure is associated with integrations into specific gene pathways. Methods: NSG mice were engrafted with memory CD4+ T-cells from two HLA-B27+ HIV male elite controllers and infected with HIVJRCSF to test the effect of autologous memory CD8+ T-cell engraftment on the proviral landscape. T cell responses and HIV viral loads were monitored weekly by flow cytometry with a Gag-KK10 tetramer and by qPCR. At week 8, splenic DNA was subjected to linker-mediated PCR for IS sequencing. IS were identified with the AAvengeR software pipeline. We performed Reactome pathway analysis on 6,402 and 12,406 unique IS from 18 CD8+ and 9 CD8- mice respectively. A

2-4% using LTR as the denominator. Proviruses that were confirmed intact by full-length sequencing comprised 0.07-2% of the LTR+ MDA wells. Two of the 10 clones carrying sequence-intact proviruses increased in size, one decreased, and seven did not change significantly over 1-4 years. Conclusion: In 3 donors with non-suppressible viremia on ART, we identified 10 clones carrying sequence-intact proviruses comprising <2% of the total proviral population. All 10 intact proviruses were integrated in genes that are normally expressed in memory CD4+ T cells, including the 5 in KRAB-ZNF genes. Only 1/10 clones declined in size on long-term ART. These results indicate that a stable pool of intact proviruses integrated in expressed genes can persist on long-term ART.

Poster Abstracts

458

Host Gene Activation at the Integration Site Reduces HIV Expression by Transcriptional Interference Sam Weissman , Miriam Viazmenski, Jack Collora, Yang-Hui Jimmy Yeh, Ya-Chi Ho Yale University, New Haven, CT, USA Background: The integration site affects HIV persistence in vivo. A provirus shares and competes with local host genes for transcription machinery that regulates expression. How host promoter activity at the integration site affects HIV expression remains unsettled. We postulated that host promoter activation increases local chromatin accessibility to transcription factors and increases HIV expression. Methods: We isolated 9 Jurkat T cell clones, each with HIV provirus (NL4-3-d6 dE-dsGFP) stably integrated into (a) introns of an actively transcribed gene in the same orientation ( VAV1, RAP1B, SPECC1 ), (b) introns of an actively transcribed gene in the opposite orientation ( INPPL1, FNBP1, KLF12, EEF2K ), and (c) intergenic regions. Using CRISPR-mediated activation or inhibition (CRISPRa/CRISPRi), we activated or inhibited the local host promoter using host-targeting sgRNA. We used nontargeting sgRNA as a control. We examined how host promoter activity affects HIV gene expression using ATAC-seq, RNA-seq, and qRT-PCR. We assessed differential gene expression and chromatin accessibility using edgeR and chromatin footprints with TOBIAS. Results: Activation of the host promoter dominantly decreased HIV gene expression in 4 out of 7 cell line clones: SPECC1, KLF12, EEF2K (each FDR<10 -6 ), and VAV1 (FDR<10 -3 ), in both orientations of HIV integration. A smaller trend was seen with RAP1B (FDR=0.07). For the cell line clone in which HIV integrated 7kb from the host promoter, activation of the host promoter INPPL1 increased HIV expression. The two clones harboring a provirus in intergenic regions expressed substantial proviral RNA despite lower chromatin accessibility. We next examined whether host promoter activity changes chromatin accessibility and transcription factor binding to the HIV LTR. Activating the host promoter of VAV1 , RAP1B , SPECC1 , and EEF2K significantly reduced chromatin accessibility of the downstream HIV 5' LTR (FDR<10 -6 ; KLF12 n.s.) and displaced HIV transcription factors. In contrast, activating host promoter INPPL1 increased HIV LTR chromatin accessibility. Conclusion: We identified transcriptional interference at the HIV integration site. Activation of host genes harboring an HIV integration site displaces transcription factors bound to the HIV 5' LTR and inhibits HIV expression. In vivo studies to understand or modulate latency should account for transcriptional interference.

460

CROI 2024 117