CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

However, because of the rarity of latently infected cells, the HIV-1 genome is poorly covered by bulk RNAseq. To address this limitation, we developed an RNAseq method with probe-based enrichment of HIV-1 reads. Methods: Resting, non-naïve CD4 T cells were isolated from leukapheresis samples from four HIV-1 Eradication and Latency Study (HEAL) participants. 15 million cells were treated for 24 hours with: 1) unstimulated 2) PMA-ionomycin (iono) 3) romidepsin (rmd) 4) bryostatin (bryo) 5) IL-15 6) rmd/bryo. Total RNA was extracted using Trizol. RNA was poly-A selected and libraries were generated following an adapted TruSeq library generation protocol. A custom set of tiling probes was used to enrich HIV-1 and control gene PCR constructs. The unenriched and enriched libraries were sequenced on NextSeq, 40x40 paired-end reads. The reads were aligned to the human transcriptome and HIV-1 genome. A custom script was used to count reads per gene and per region of the HIV-1 genome. Results: For both host control genes and HIV-1, we observed an average ~50-fold enrichment after probe capture (see Figure 1). HIV-1 reads aligned across all regions of the genome. PMA-iono and combination rmd/bryo had the greatest increase in HIV-1 transcription. To test the reproducibility of the probe-enrichment, we performed linear regression on normalized RNAseq reads from cells treated with PMA-iono. We observed a Pearson’s R2 of 0.95 for total RNAseq between two participants and 0.97 for enriched RNAseq between the same participants. We also found evidence of hypermutated HIV-1 RNAseq reads in the enriched samples. Conclusion: Our approach to analyzing host and HIV transcriptomes leverages next-generation sequencing to investigate latency reactivation. Probe-based enrichment allowed RNAseq quantification of HIV-1 reads from resting memory CD4+ T cells without the need for sorting of HIV-infected cell populations. We were able to measure HIV-1 transcription after reactivation from latency using a variety of latency reversal agents and compare HIV-1 gene expression across conditions. Analysis of differential host gene expression will yield insight into host factors necessary for HIV-1 reactivation in latently infected resting memory CD4+ T cells in persons with HIV.

355 NOD2 AND TLR8 AGONISTS ENHANCE IL-15-MEDIATED ACTIVATION OF HIV EXPRESSION Jasmine Kaur 1 , Jay Lalezari 2 , Rebecca Hoh 3 , Steven G. Deeks 3 , Wade Blair 1 , Tomas Cihlar 1 , Jeffrey Murry 1 1 Gilead Sciences, Inc, Foster City, CA, USA, 2 Quest Clinical Research, San Francisco, CA, USA, 3 University of California San Francisco, San Francisco, CA, USA Background: The latent HIV reservoir is a barrier to achieving an HIV cure. Individual reservoir-targeting agents have shown potential activity in exploratory clinical trials, but it is likely that activation of HIV expression would enhance and/or accelerate the depletion of the latent reservoir. We previously identified clinically advanced agents that modestly activate HIV expression in cells isolated from ART-suppressed people living with HIV (PLWHIV), including IL-15 and agonists of multiple pattern recognition receptors (PRRs), such as toll- like receptor (TLR) and Nucleotide-binding oligomerization domain-containing protein 2 (NOD2) agonists. Here we identify combinations of agents that have greater activity than either agent alone. Methods: Peripheral blood mononuclear cells (PBMCs) were isolated from ART-suppressed PLWHIV then treated with various PRR agonists individually or in combination with IL-15. Cytokine production and surface markers of T cell activation were assessed 24 hours after treatment initiation. HIV RNA in culture supernatants and T cell proliferation were quantified following a 4-day treatment with PRR agonists. Wilcoxon matched pair signed rank test and Bliss independence model were used for statistical and synergy analysis, respectively. Results: In PBMCs from 7 ART-suppressed PLWHIV, IL-15 alone induced a 3.9-fold increase in HIV expression relative to control, while NOD2 and TLR8 agonists induced 3.0- and 3.2-fold increases, respectively (geometric means, p < 0.05 for each). In combination with IL-15, NOD2 and TLR8 agonists had the greatest effect, increasing HIV expression 14- and 11-fold, respectively (p < 0.05 for both compared to IL-15 alone). This was not significantly different from that induced by PMA and ionomycin (18-fold). The combination of NOD2 and IL-15 showed the clearest synergy. Correspondingly, both NOD2 and TLR8 agonists increased the levels of cytokines and activation markers produced in response to IL-15 stimulation, but had minimal additional effect on CD4 T cell proliferation. Conclusion: Combining either NOD2 or TLR8 agonist with IL-15 significantly increased HIV expression and, in cells from several donors, approached that observed with the mitogenic activation control. This identifies clinically tested agents capable of robustly inducing HIV. It is important to consider that these combinations can also activate broader immunity and potentially augment immune-mediated reservoir clearance. 356 HIV-1 GENE EXPRESSION DURING REVERSAL OF LATENCY USING RNA-Seq WITH PROBE ENRICHMENT Phillip Tomezsko 1 , Silvi Rouskin 2 , Daniel R. Kuritzkes 3 , Athe Tsibris 3 1 Harvard Medical School, Boston, MA, USA, 2 MIT, Cambridge, MA, USA, 3 Brigham and Women's Hospital, Boston, MA, USA Background: Transcriptomic analysis of the human and HIV-1 expression profile that is essential for successful reactivation of latently infected cells promises to help inform the next generation of latency reversal agents.

Poster Abstracts

357 SURFACE ENGINEERING OF EXTRACELLULAR VESICLES TO TARGET HIV PERSISTENCE Pooja Bhardwaj 1, Rafal Kaminski 2 , Shivani Desai 1 , Ali Danesh 1 , Amir Afshari 1 , Nicholas Yam 1 , Kamel Khalili 2 , Archana Gupta 1 , Satish K. Pillai 1 1 Vitalant Research Institute, San Francisco, CA, USA, 2 Temple University, Philadelphia, PA, USA Background: Current limitations of antiretroviral therapy are driving interest in novel HIV eradication strategies. Extracellular vesicles (EVs) are nano-sized membrane vesicles involved in cell signaling which have shown promise as engineerable therapeutic agents. We used surface display technology to engineer HIV-targeting EVs (HTEVs) that block HIV infection and target infected cells. Methods: EVs were isolated from healthy donor plasma using polymer-based precipitation and column purification. EVs were decorated with single-chain variable fragment (scFv)-C1C2 fusion proteins targeting the HIV envelope protein. Surface-engineered EVs and HIV particles were fluorescently labeled and incubation reactions were visualized in dual-color channel and single- particle tracking analysis using a Nanoimager (ONI). Decorated EVs were incubated for two hours with a GFP-reporter HIV strain at 1:1, 2:1, and 4:1 ratios. Jurkat E6.1 cells and primary human CD4+ T cells were infected via spinoculation. Reporter virus was incubated with no EVs, undecorated EVs, or anti-PD-1 scFv-decorated EVs as negative controls. Jurkat 2D10 cells were induced with PMA/TSA to reactivate latent HIVNL4-3-Dgag/pol-GFP reporter

CROI 2020 123