CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

297LB TRANSCRIPTOMIC HOST RESPONSES TO RHCMV/SIV VACCINATION IN RHESUS MACAQUES Connor Driscoll 1 , Richard Green 1 , Monica Rojas-Peña 1 , Fredrik Barrenas 2 , Jean Chang 1 , Elise Smith 1 , Lynn Law 1 , Jan Komorowski 2 , Scott Hansen 3 , Louis J. Picker 3 , Michael Gale 1 1 University of Washington, Seattle, WA, USA, 2 Uppsala University, Uppsala, Sweden, 3 Oregon Health and Sciences University, Portland, OR, USA Background: The rhesus cytomegalovirus (RhCMV) strain 68-1 vaccine against simian immunodeficiency virus (SIV) induces a T cell response that protects over 50% of vaccinated rhesus macaques (RM) to clear infection against multiple SIV challenge including distinct challenge routes. To define the molecular features of the host response to vaccination and the underlying gene signature of vaccine protection we assessed the transcriptional responses of protected and non- protected RMs following vaccination prior to challenge. Methods: Two groups of 15 RMs were administered 68-1 RhCMV/SIV vaccine via oral or subcutaneous delivery. Following vaccination, RMs were subjected to repeated limiting dose intrarectal SIVmac239 challenge until infected as detected as plasma virus or the de novo development of SIV Vif-specific T cell response. During the vaccination phase, blood samples were collected at time points near prime, boost, and before challenge. mRNA-seq was performed followed by bioinformatics analysis including differential gene expression, co-expression clustering, and functional enrichment analyses. Results: In the subcutaneous and orally-vaccinated groups, 53% and 60% of RMs cleared SIV infection after virus challenge, respectively. Protected RMs showed both an acute and sustained increase in gene expression indicative of myeloid cell responses, including genes and gene networks involved in Toll-like receptor signaling, inflammasome induction, and monocyte activation. We identified a noncanonical T-cell signaling signature in protected animals that was characterized by a decrease in of Zap70 and Tbx21 with concomitant increase in Ido1 expression. Importantly, we identified an interleukin (IL)-15/ STAT5 signaling module that links with immune cell trafficking and protection from SIV infection. A rule-based machine learning analysis confirmed that gene expression signatures controlling TLR activation of macrophages and myeloid cell activation, could predict vaccine protection. Assessment of an independent cohort of naive RM treated with IL-15 revealed gene networks of the IL-15 response. Integration analyses were conducted to identify a subset of IL-15 response genes in vaccinated animals whose induced expression tracks with vaccine protection. This gene expression signature was maintained post-virus challenge. Conclusion: Our results define noncanonical T cell activation, inflammatory signaling, myeloid cell activation, and IL-15 response as features of RhCMV-SIV vaccine protection. 298 EARLY ART PARTLY REDUCES HIV-INDUCED MUCOSAL B-CELL HYPERACTIVATION Alexandra Schuetz 1 , Yuwadee Phuang-Ngern 1 , Carlo Sacdalan 2 , Suchada Sukhumvittaya 1 , Eugène Kroon 2 , Rungsun Rerknimitr 3 , Robin L. Dewar 4 , Nitiya Chomchey 2 , Rapee Trichaviroj 1 , Nittaya Phanuphak 2 , Merlin L. Robb 5 , Nelson L. Michael 5 , Mark de Souza 2 , Jintanat Ananworanich 5 , for the RV254/SEARCH 010 and RV304/SEARCH 013 Study Groups 1 Armed Forces Research Institute of Medical Sciences in Bangkok, Bangkok, Thailand, 2 Thai Red Cross AIDS Research Center, Bangkok, Thailand, 3 Chulalongkorn University, Bangkok, Thailand, 4 National Cancer Institute, Frederick, MD, USA, 5 US Military HIV Research Program, Silver Spring, MD, USA Background: HIV-induced immune activation is a hallmark of HIV pathogenesis and disease progression. This includes hyperactivation of B cells with increased differentiation of B cells to plasmablasts (PB) and a decrease in memory B cells (BMem). We evaluated the effects of antiretroviral therapy (ART) initiated during acute HIV infection (AHI) on mucosal B cells. Methods: Individuals at Fiebig stages I to III of AHI, underwent sigmoid biopsy at the time of HIV diagnosis (AHI; n=14) and 24 months post ART initiation (n=18). Matched HIV-uninfected (HIV-; n=33) and individuals who had initiated ART during chronic HIV infection (CHI - 12 to 36 months post ART initiation; n=6) served as controls. Mucosal mononuclear cells were isolated and multiparameter flow cytometry was performed to determine the frequency of B cell subsets (PB: Lin-CD20+/-CD19+CD27hiCD38hi; BMem: Lin- CD20+CD19+CD27+CD38-IgD-IgM-). Results: Median gut HIV RNA was 3.8 log10 [copies/mg tissue] at AHI diagnosis and undetectable after 24 months of ART in 17/18 participants (p≤0.001). There

readily generates mucosal and systemic immune responses in animal models. We developed a series of PIV5 vectors expressing SIVgag and HIV gp140 and administered them intranasally to rhesus macaques, in order to evaluate the potential of this mucosal vaccine vector as an HIV vaccine. Methods: HIV-1 JR-FL Env and SIVmac239 Gag were inserted individually into the PIV5 genome, and replication and expression of gene products validated in cell lines. Virus-like particles (VLPs) consisting of SIVmac239 Gag core and HIV-1 JR-FL Env were produced in stable, inducible manner frommammalian cells and purified. Macaques received four doses of PIV5-SIVgag + PIV5-HIVgp140 by intranasal administration, followed by boosting with SHIV virus-like particles (VLPs). Humoral and cellular immune responses to SIV and HIV antigens and to the PIV5 vector were measured over time. Results: Monkeys immunized with PIV5 constructs developed HIV Env and SIV Gag-specific binding Ab titers in a dose-dependent manner. VLP boosting further enhanced SIV/HIV-specific responses, including when the VLP boost followed priming with the lowest intranasal dose of PIV5. Gp120 binding titers correlated to the magnitude of antigen specific plasmablasts measured in blood on days 5 post boost, except for a peak plasmablast response after the initial VLP boost. Presensitization with a heterologous PIV5 vector decreased the frequencies of gp120-specific plasmablasts detected following PIV5-delivered immunizations. CD4+ and CD8+ T cell responses were elicited by PIV5 vectors and boosted following VLP administration. HIV and SIV-specific IgG and weak IgA responses were detected at mucosal sites. Neutralizing antibodies and ADCC responses remain under evaluation at the time of submission of this abstract. Conclusion: We demonstrate here for the first time that intranasal administration of PIV5-SIV/HIV vectors are well-tolerated and immunogenic. VLP boosting enhanced HIV- and SIV-specific humoral and cellular immune responses. This prime-boost approach may provide a novel approach to the development of systemic and mucosal protective responses against HIV. Jinhua Xiang , James H. McLinden, Qing Chang, Thomas M. Kaufman, Jack T. Stapleton University of Iowa, Iowa City, IA, USA Background: Although broadly neutralizing HIV antibodies are detected in infected (and sometimes uninfected) humans, immunization with HIV-1 antigens generally do not elicit broadly neutralizing antibodies. Clinical studies found that GBV-C E2 antibody is associated with improved survival in HIV and GBV-C co-infected individuals, and E2 antibody neutralizes HIV-1 infectivity by reducing virus entry in vitro. To determine if GBV-C E2 elicits broadly neutralizing HIV monoclonal antibodies (MAbs), we immunized mice with either E2 (lacking C-terminal transmembrane domain) or a synthetic peptide we previously demonstrated is involved in HIV entry and generated anti-E2 MAbs against both antigens. Here, we examined the interactions of both MAbs for their ability to inhibit HIV-1 infectivity, precipitate HIV-1 particles, and bind HIV-specific structural protein antigens. Methods: GBV-C E2 protein was expressed in CHO cells, and a synthetic peptide generated of the 17 amino acid E2 region involved in HIV entry. Mice were immunized with 25 ug E2 protein or peptide in IFA four times prior to sacrifice, and hybridomas generated. One anti-E2 and one anti-peptide hybridoma cell line (8H2 and 1C4 respectively) were identified frommore than 2,000 independent cultures using a capture E2 ELISA. HIV-1 envelope proteins (gp120, gp140 and gp41), gp41 peptides (MPER and T-20), and X4- and R5-tropic HIV-1 isolates representing clades A, B, and D were studied. Results: Both GBV-C anti-E2 MAbs precipitated HIV-1 particles and neutralized X4 and R5-tropic HIV isolates from diverse geographic regions representing three clades (IC50 ranging from 2.5 to 7.5 ug/mL). These antibodies did not neutralize mumps or yellow fever viruses, demonstrating specificity. 8H2 reacted with HIV-1 gp140 and gp41 proteins, but not gp120 using two types of ELISA methods. Both 8H2 and 1C4 recognized the HIV-1 gp41 fusion (MPER) peptide in immunoblot assays. Conclusion: Although HIV-1 antigens do not elicit broadly neutralizing HIV-1 antibodies, GBV-C E2 and an E2 peptide elicited HIV-1 neutralizing MAbs in mice. Since GBV-C E2 antibodies are associated with prolonged survival in three clinical studies, these data illustrate the potential for a novel antigen to incorporate into HIV-1 vaccine strategies.

Poster Abstracts

296 NOVEL ANTIGEN ELICITS BROADLY NEUTRALIZING ANTI-HIV MONOCLONAL ANTIBODIES IN MICE

CROI 2019 108