CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Methods: This was a cross-sectional study where we first tested 129 HIV-1-infected subjects and 27 HIV-negative controls in Barcelona (BCN0). Findings were internally validated in 110 subjects from BCN0 providing a second fecal sample 1 month later (BCN1). External validation was obtained in 77 HIV-1-infected and 7 non-infected subjects from Stockholm (STK). In all study participants, we produced MiSeq TM 16S rRNA sequence data on fecal microbiomes and collected comprehensive metadata. Alpha and beta diversity analyses of the gut microbiota were performed. LASSO regression was used to quantify the strength of the association between sexual practice, HIV-1 status and global fecal microbiota composition. Results: Men who have sex with men (MSM) consistently had a significantly richer and more diverse fecal microbiota than non-MSM individuals. After stratifying for sexual practice, HIV-1 infection remained consistently associated with reduced bacterial richness. The lowest microbial richness was observed in HIV-1-infected individuals with immune- virological discordant phenotype. Fecal microbiomes strongly clustered by sexual practice rather than by HIV-1 serostatus, with high concordance between BCN0 and BCN1 (Procrustes m2=0.3475, PROTEST p=0.001). The fecal microbiota composition in BCN and STK significantly differed by sexual practice, with MSM and non-MSM subjects mostly belonging to the Prevotella and Bacteroides enterotypes, respectively. Cross-validation accuracy of the LASSO model was very high for sexual practice (mean AUC=95%), confirming a different fecal microbiota composition in MSM and non-MSM individuals after excluding multiple other potential confounders. In contrast, HIV-1 status was not associated with consistent changes in the global fecal microbiota composition at genus level. Conclusions: Gay men have a distinct gut microbiota composition, which is a potential confounder of all human fecal microbiome studies. Yet, HIV-1 infection remains independently associated with reduced bacterial richness, which offers new avenues for interventions to improve HIV immune dysfunction. 261 Background: During early HIV infection, the intestinal mucosa is a critical reservoir of viral replication and profound CD4+ T cell depletion, with subsequent loss of epithelial barrier function. This is believed to result in translocation of luminal bacterial products, which may contribute to systemic immune activation and eventual immune exhaustion. We investigated the role of the enteric microbiome in HIV disease progression by using high-throughput sequencing to conduct a comprehensive survey of changes in gut bacterial populations using stool paired with patient plasma and clinical metadata. Methods: A total of 124 stool samples with matching plasma and clinical data were analyzed; 90 of these subjects completed a comprehensive diet survey. Patient cohorts include HIV uninfected and infected (controllers, untreated, and ARV-treated) subjects. DNA extracted from fecal samples was used to perform 16S rRNA sequencing. QIIME was used to group viable sequences into Operational Taxon Units (OTUs) with >97% shared similarity, and PICTRUSt was used to infer gene expression of representative microbial communities. Soluble levels of sCD14, sCD163, CRP, and I-FABP/FABP2 in plasma were measured. Results: At all taxonomic levels the HIV-uninfected subjects demonstrated more intra-subject bacterial community variability with greater levels of Verrucomicrobia and lower dominant levels of Bacteroidetes at the phyla level than HIV-infected subjects. Weighted Unifrac analysis of bacterial communities demonstrated significant clustering on a principal coordinate plane (p<0.01 PERMANOVA testing via Adonis). Communities from HIV uninfected and HIV+ untreated patients clustered independently, but those from infected patients on ARVs were spread between the two groups. Relationships among bacterial communities, inferred microbial gene expression, and systemic markers were assessed. Conclusions: We identified specific HIV-associated alterations in the gut bacterial communities and their inferred gene expression, and associations with soluble clinical markers. These alterations suggest that the enteric microbiome is significantly altered by HIV infection and may directly contribute to disease progression. Impact of HIV-Associated Changes in the Gut Microbiome on Disease Progression Jesús Luévano ; David B. Gootenberg; Jeffrey M. Paer; Bruce D.Walker; Douglas S. Kwon Ragon Inst of MGH, MIT, and Harvard, Cambridge, MA, USA

Poster Abstracts

262 Altered Gut Microbes Enhance Mucosal CD4 T Cell Infection and Depletion Ex Vivo Stephanie M. Dillon 1 ; Eric L. Lee 1 ; Andrew M. Donovan 1 ; Kejun Guo 1 ; Michael S. Harper 1 ; Daniel N. Frank 1 ; Martin D. McCarter 1 ; Mario L. Santiago 1 ; CaraWilson 2 1 Univ of Colorado Anschutz Med Campus, Aurora, CO, USA; 2 Univ of Colorado Hosp, Aurora, CO, USA

Background: Early HIV-1 infection is characterized by high levels of HIV-1 replication and CD4 T cell depletion in the gut mucosa as well as epithelial barrier breakdown and translocation of microbial products into the lamina propria (LP) and systemic circulation. HIV-induced disruption of gut homeostasis is associated with changes in the gut microbiome (dysbiosis) that are linked to mucosal and systemic immune activation. Mechanisms by which dysbiotic bacteria contribute to HIV-1 pathogenesis are poorly characterized. Here we investigated the impact of exposure of representative bacterial species previously shown to be in either high or low abundance in the colonic mucosa of viremic HIV-1 infected individuals (HIV-altered mucosal bacteria; HAMB) on LP CD4 T cell function, infection by HIV-1, and survival in vitro. Methods: LP mononuclear cells isolated from normal human jejunums (n=11) were infected with CCR5-tropic HIV-1 BaL or mock-infected and cultured with high abundance HAMB (3 gram-negative, GN) or low abundance HAMB (2 GN, 2 gram-positive, GP) or with control GN Escherichia coli . Flow cytometry was used to measure levels of CD4 T cell infection

101

CROI 2016