CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

194 STRUCTURAL AND RNA BINDING MODEL OF APOBEC3G N-TERMINAL DOMAIN FOR NEW DRUG DESIGNS Hirofumi Fukuda 1 , Anamaria D. Sarca 1 , Kazuo Yamashita 2 , Song Lin Li 2 , Luca Sardo 3 , Jessica Smith 4 , Hirotaka Ebina 1 , Kotaro Shirakawa 1 , Kei Sato 1 , Daron Standley 1 , Yoshio Koyanagi 1 , Taisuke Izumi 1 , Akifumi Takaori-Kondo 1 1 Kyoto University, Kyoto, Japan, 2 Osaka University, Sukta, Osaka, Japan, 3 University of the Sciences, Philadelphia, PA, USA, 4 United States Army Medical Research Institute of Infectious Diseases, Frederick, MD, USA Background: APOBEC3G (A3G) is a cellular protein that inhibits HIV infection. The interaction of the A3G N-terminal domain (NTD) with RNA is essential for its virion incorporation, but this interaction is not completely understood. A3G-NTD is also recognized by HIV-1 Viral infectivity factor (Vif) and A3G-Vif binding leads to A3G degradation. Designing inhibitors for A3G-Vif interactions is a novel drug development strategy; however, targeting A3G-Vif interaction could negatively affect A3G-RNA interaction that is required for A3G’s antiviral activity. To develop a novel anti-HIV drug, it is necessary to understand A3G-RNA structural binding profile, we generated an in silico docking model to simulate the RNA- binding of A3G-NTD. Methods: The solubilized A3G-NTD structure has been recently revealed, and its amino-acid homology is almost 80% of the wild-type A3G-NTD. RNA association ability may be diminished resulting in deficient virion incorporation. We constructed a model of wild-type A3G-NTD based on the solubilized A3G-NTD, then we simulated A3G-RNA docking patterns with single-stranded RNA. With this model, for each amino acid we calculated the RNA binding propensity (BP) by geometrical information and measured the contact frequency (CF) with RNA. We evaluated the accuracy of this structural model by introducing Alanine substitutions for amino acids, which were predicted to be directly involved in RNA binding based on their BP and CF. Results: We confirmed the accuracy of our RNA docking model with several alanine-substituted mutants, which have been reported to associate with RNA or do not relate to RNA interaction. We have additionally determined three novel residues as RNA associated amino-acids by BP and CF calculated by the model, involved in RNA binding (N20, R55, and S95). Based on our model with residues, which have already been identified as binding sites for Vif (Three different RNA molecules with homology model in Figure), a wide area of the Vif interaction (colored dark grey on the model in Figure) is overlapped with the RNA binding surface. Conclusion: Designing new drugs that can inhibit A3G-Vif interaction requires high specificity as not to affect RNA binding to the same N-terminal Domain. Our three-dimensional structural and dynamic RNA-binding model will provide new insights for this purpose. Based on this model, the area surrounding DPD motif enclosed with the dotted line in Figure might be a sole target for in silico drug designing to prohibit only A3G-Vif interaction.

195 SEMINAL CYTOKINE/CHEMOKINE NETWORK AND HIV TRANSMISSION IN MEN WHO HAVE SEX WITH MEN Christophe Vanpouille 1 , Andrea Lisco 2 , Leonid Margolis 1 , Laura Layman 3 , Martin Hoenigl 3 , Sara Gianella 3 1 National Institute of Child Health and Human Development, Bethesda, MD, USA, 2 National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA, 3 University of California San Diego, La Jolla, CA, USA Background: The cytokine/chemokine network in genital secretion reflects the functional status of immune cells in the genital tract, where the key events related to sexual transmission occur. Here, we describe differences in cytokine/ chemokine levels in semen between HIV-infected men who did transmit and did not transmit HIV to their male sexual partner Methods: Participants were men with primary HIV infection and their recent male sexual partners (HIV-positive or negative). HIV transmission among sero- concordant partnerships was defined as phylogenetic linkage (≤1.5% genetic distance in pol), and the HIV-positive partner with the earlier estimated date of infection (EDI) was considered the source. Among sero-discordant couples, the HIV-positive partner was considered the (potential) source. All analyses were restricted to source partners. Sources with sero-discordant partners were classified as non-transmitters (n=23), and those with sero-concordant phylogenetically-linked partners were considered transmitters (n=21). For each source partner (transmitter or non-transmitter), semen was collected and a panel of 34 cytokine/chemokines were measured by Luminex. Principal- components and clustering methods were used to select cytokine/chemokines with the strongest association with transmission, which were included in a multiple logistic regression model in the presence of the following covariates from the source partner: ART status, EDI, age, race/ethnicity, CD4+ and CD8+ cells, HIV RNA levels in blood and semen, presence of seminal HSV-2, EBV or CMV DNA. Results: At the univariate level, participants classified as transmitters had significantly higher levels of IL-13, lower levels of M-CSF, INF-, IL-17, TGF- and Eotaxin compared to non-transmitters (Table 1). Individuals classified as transmitters were also more likely to be ART naïve, HIV-infected for >1 year, have lower CD4+ cells, detectable seminal HIV RNA and EBV DNA. In multivariable model, higher IL-13, lower Eotaxin and detectable HIV RNA in semen remained significantly associated with presence of HIV transmission within the observed partnership. Conclusion: While detectable HIV RNA in semen was the strongest predictor of HIV transmission, altered cytokine/chemokine network might play a role in HIV transmission. For example, the observed association between higher IL-13 and HIV transmission is consistent with previous vaccine studies reporting stronger anti-HIV response when inhibitors of IL-13 were used and should be further investigated.

Poster Abstracts

196 BENEFICIAL IMPACT OF EARLY TREATMENT ON RESTRICTION FACTOR EXPRESSION PROFILE Clarissa Van Hecke 1 , Magdalena Sips 1 , Eva Malatinkova 1 , Ward De Spiegelaere 1 , Karen Vervisch 1 , Chris Verhofstede 1 , Margaret Johnson 2 , Sabine Kinloch-de Loes 2 , Wim Trypsteen 1 , Linos Vandekerckhove 1 1 Ghent University, Ghent, Belgium, 2 Royal Free Hospital, London, UK

69

CROI 2018