CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

365 DISTINCT VULNERABILITY OF VIRAL RESERVOIR CELLS TO NK CELL- MEDIATED IMMUNE EFFECTS Stephane Hua 1 , Hsiao-Hsuan Kuo 1 , Shivaali Maddali 1 , Kevin Einkauf 2 , Chenyang Jiang 2 , Xiaodong Lian 1 , Mathias Lichterfeld 2 , Xu G. Yu 1 1 Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA, 2 Brigham and Women’s Hospital, Boston, MA, USA Background: Potent antiretroviral therapy leads to suppression of HIV-1 replication but is associated with establishment of a stable reservoir of latently- infected cells that can fuel rebound viremia upon treatment discontinuation. Innate cellular immune responses mediated by NK cells have been inversely associated with reductions in viral reservoir size during antiretroviral therapy, suggesting that latently-infected CD4 T cells may be susceptible to NK cell- mediated immune effects. Methods: Primary CD4 T cells were activated and in vitro infected with a dual- reporter virus allowing us to distinguish cells with productive or latent HIV-1 infection, followed by flow cytometry-based analysis of NK cell receptor ligands; moreover, the same analysis was performed in primary CD4 T cells from ART- treated patients. Single-cell RNA-Seq analysis of in vitro, latently infected cells was performed and compared to the transcriptome of productively infected and uninfected cells. The susceptibility of latently infected CD4 T cells to NK cell- mediated killing was analyzed using functional cytotoxicity assays. Results: In vitro, latently infected CD4 T cells expressed significantly higher levels of the activating NK cell ligand ULBP1, and were significantly enriched for a population of cells simultaneously expressing a combination of the three NKG2D ligands ULBP1, MICA and MICB. Upregulation of these molecules was associated with transcriptional activation of the ATM DNA damage response pathway. Functional assays demonstrated an increased susceptibility of latently-infected CD4 T cells to NK cell-mediated killing; this vulnerability to cytotoxic immune effects of NK cells was most obvious in latently-infected CD4 T cells expressing ULBP1, MICA and MICB, and was abrogated by antibodies directed against NKG2D. An upregulation of NKG2D ligands was also observed in patient-derived CD4 T cells from ART-treated patients, and denoted a subset of CD4 T cells characterized by increased expression of immune checkpoint and activation markers. Conclusion: Latently-infected CD4 T cells seem to express a distinct signature of activating NK cell receptors, likely in response to activation of DNA damage response signals that may result from viral latency. Expression of activating NK cell receptors on latently-infected CD4 T cells can increase the susceptibility to NK cell killing, and may represent a distinct vulnerability of the viral reservoir that provides novel targets for therapeutic viral eradication studies. 366 WITHDRAWN / INTENTIONALLY UNASSIGNED 367 WITHDRAWN / INTENTIONALLY UNASSIGNED 368 INDUCIBILITY OF LATENT HIV-1 IN RESTING CD4+ MEMORY T-CELL SUBSETS Kyungyoon J. Kwon 1 , Andrew Timmons 1 , Srona Sengupta 1 , Francesco R. Simonetti 1 , Hao Zhang 1 , Rebecca Hoh 2 , Steven G. Deeks 2 , Rafick-Pierre Sekaly 3 , Janet Siliciano 1 , Robert Siliciano 1 1 Johns Hopkins University, Baltimore, MD, USA, 2 University of California San Francisco, San Francisco, CA, USA, 3 Case Western Reserve University, Cleveland, OH, USA Background: The latent reservoir for HIV-1 is comprised mainly of resting memory CD4+ T cells that harbor integrated replication-competent proviruses that are not actively transcribed. Recent work indicated that the proliferation of infected cells is a major factor in the generation and stability of the latent reservoir. Several groups have shown that latently infected cells that clonally expand in vivo can be activated in vitro without producing virus. One hypothesis to explain this observation is that certain subpopulations of reservoir cells, such as particular CD4+memory subsets, are predisposed to delay reactivation of the latent virus or promote a deeper state of latency. To evaluate this possibility, we studied resting naïve, central memory, transitional memory, and effector memory CD4+ T cells from 10 ART-suppressed HIV patients in a multiple stimulation viral outgrowth assay (MSVOA). Methods: Culture readouts were measured by p24 ELISA. Flow cytometry was used to track CCR7 and CD27 expression throughout the culture. A droplet- digital-PCR (ddPCR)-based assay developed in our lab was performed on each

work we assessed the in vivo antiviral effect of dasatinib using humanized mice infected with HIV-1. Methods: Human CD34+ hematopoietic stem cell-engrafted NSG mice (hu-CD34) were treated for 5 days with dasatinib 20mg/kg/day (n=5) or with placebo (citrate buffer solution) (n=5). Then, all mice were intraperitoneally injected with purified HIV-1NL4-3 (17,500 TCID50) and treated for 21 days with dasatinib or placebo, in the absence of antiretroviral treatment (ART). Results: 1) Viral load in hu-CD34 mice treated with dasatinib was 4.7-, 3.8- and 3.5-fold lower than the placebo group after 7, 15 and 21 days of infection, respectively. Two mice from dasatinib group persistently showed undetectable viral load. 2) Proviral load in blood of mice treated with dasatinib remained 1.6-, 4.6- and 2.2-fold lower than the placebo group after 7, 15 and 21 days post-infection, respectively. 3) Proviral load in GALT was 3.0-fold smaller at 21 days post-infection in the dasatinib-treated group. 4) Treatment with dasatinib affected the distribution of CD4 and CD8 subpopulations: CD4 and CD8 TCM cells were respectively 2.0- and 2.7-fold lower than the placebo group; CD4 and CD8 TEM cells were 4.0- and 6.3-fold lower; CD4 and CD8 TEMRA cells were 1.5- and 3.5-fold lower; whereas CD4 and CD8 naïve T cells were 1.5- and 1.4-fold higher. Conclusion: Daily oral treatment with dasatinib in the absence of ART interfered with HIV-1 acute infection in hu-CD34 mice. Dasatinib reduced viral load and proviral reservoir size in blood and GALT, and modified the distribution of CD4 and CD8 subpopulations. This study is the first proof of concept that dasatinib decreases HIV-1 reservoir in vivo, supporting the use of dasatinib in combination with ART to reduce the reservoir size, particularly in patients with acute infection. Himanshu Batra 1 , Jingen Zhu 1 , Swati Jain 1 , Neeti Ananthswamy 1 , Marthandan Mahalingam 1 , Pan Tao 1 , Mary F. Kearney 2 , Frank Maldarelli 2 , Venigalla Rao 1 1 The Catholic University of America, Washington, DC, USA, 2 National Cancer Institute, Frederick, MD, USA Background: Persistent latent reservoir of CD4+ T cells containing stably integrated and transcriptionally silent form of HIV-1 provirus is a major barrier to cure HIV. An effective curative approach would be to activate the latent reservoir and selectively disrupt the proviral genome as well as elicit immune responses to neutralize any remaining viruses. We hypothesize that the 120 x 86 nm bacteriophage T4 capsid (head) nanoparticle carrying a combination of HIV cure molecules; proteins, DNAs, and RNAs can be targeted to CD4+ T cells for delivery and cure. About 170 kb DNA can be packaged inside the capsid and ~1025 proteins fused to the outer capsid proteins Soc (small outer capsid protein, 870 copies) and Hoc (highly antigenic outer capsid protein, 155 copies) can be displayed on the capsid surface. Methods: Empty phage T4 capsids are purified from E.coli infected with mutant phage that lacks tail, neck, and outer capsid proteins. The DNA packaging motor (gp17) is attached to the portal vertex of the capsid. Packaging is carried out by adding plasmid DNAs containing Cas9 and/or gRNAs and ATP. The packaged heads are displayed with combinations of proteins; HIV envelope protein or the CD4 binding DARPin fused to Soc or Hoc and Soc-Cas9/gRNA complex. Results: HIV cure T4 nanoparticles were assembled by packaging 7-8 molecules of 6-8 kb LTR-gRNA-Cas9 plasmid DNA inside and displaying ~150 gRNA- Cas9-Soc complexes and 155 molecules of CD4DARPin-Hoc or 30 envelope trimers from a transmitted founder virus. These in vitro assembled HIV Cure T4 nanoparticles bound specifically to CD4+ T cells (A3.01) and disrupted HIV-1 provirus integrated into HEK 293T cells, as determined by T7 EI assay. Unexpectedly, the T4 nanoparticles also activated J-Lat 10.6 full length cells (P value =0.0024, t-test) and the activation was dependent on the displayed targeting ligand specific to CD4 receptor. Experiments are underway to determine if the T4 nanoparticles can deliver cargo into these activated J-Lat cells and disrupt the proviral genome. Conclusion: The phage T4 nanoparticle is a unique and powerful vehicle for delivery of complex cargos into mammalian cells. This study shows proof-of- principle data on targeting to CD4+ cells, provirus disruption, and activation of the latent J-Lat T cells. With its large capacity and engineering flexibility to deliver complex cargos, the T4 nanoparticle system has the potential to be an effective platform for targeted HIV-1 cure

364 A BACTERIOPHAGE T4 NANOPARTICLE PLATFORM FOR TARGETED CURATIVE THERAPY AGAINST HIV-1

Poster Abstracts

CROI 2019 132