MOAA0104 - Oral Abstract Session
The effect of HIV-infection and highly-active anti-retroviral therapy on the human gut microbiome investigated by phylogenetic-targeted 454 pyrosequencing: a meta-genomic profile of duodenal biopsies, aspirates, and stool over time
Presented by Collin L. Ellis (United States).
C.L. Ellis1, C.-S. Li2, S. Mann3, Z.-M. Ma4, H. Overman2, A. Maniar1, T. Yotter1, E. Tsuchida5, T.H. Knight1, J.C. Rutledge1, C.J. Miller1,4, R.B. Pollard1, D.M. Asmuth1
1University of California Davis School of Medicine, Internal Medicine, Sacramento, United States, 2University of California Davis, Davis, United States, 3Mather Veteran's Administration Hospital, Internal Medicine, Mather, United States, 4California National Primate Research Center, Davis, United States, 5CARES Clinic, Sacramento, United States
Background: Intestinal bacterial translocation contributes to persistent immune activation and HIV-pathogenesis. We have previously shown that the distribution of bacteria at the order-level in the stools of HIV-infected patients (pts) influences peripheral and gut tissue immune activation and CD4+ depletion. Spatial meta-genomic profiling of the gut microbiome by extensive phylogenetic sequencing of multiple different intestinal specimen-types from HIV pts to establish the distribution of supra- and sub- order taxa has not previously been reported.
Methods: ART-naive pts underwent endoscopy before and 9 mo after initiating HAART for duodenal biopsy (DB) and lumen aspirate (LA). Stool was collected at the same time. After microbial DNA was extracted from each specimen, 454 pyro-sequencing of 16S rDNA amplicons recovered from DB, LA, and stool from pts and controls was performed to assess 1st order abundances of the stable class-level and dynamic genus-level.
Results: Samples from 14 subjects are available for analysis. 1st order class-level % abundance results (stool/biopsy are longitudinal) and 2 controls are available for comparison. Compared to controls, pre-ART stool had 0.32%> proinflammatory Gammaproteobacteria (GP) (0% in controls) and all other classes of Proteobacteria (P) appear to over-colonize during infection. POST-ART stool had 2.5%> GP (0% control). Over all, little change was observed after HAART. A striking preponderance of Cyanobacteria (>26%) was observed in pre-ART DB compared to controls which is a primitive blue-green algae. The 2 pre-ART LA specimens, a novel sample-type, produced an absolute quantity of: 5-13% GP, 11-21% Beta-P, 1.2-3.4% Alpha-P, and 0-0.6% Epsilon-P, 6-10% Clostridiales, 2-9% Bacteroidetes, 4-7% Actinobacteria, 0.5-3.2% Erysipelotrichi, 2-37% Bacilli, 0.12-11% Fusobacteria, and 0.17-1.2% Cyanobacteria.
Conclusions: Consistent with our previous reports of order-level gut bacteria in HIV-patients, we observed greater representation of proinflammatory/inflammation-thriving class-level bacteria. We also observed unique distributions of bacteria in samples from the LA and DB specimens which were not clearly impacted by HAART therapy.
Back to the Programme-at-a-Glance