Background:
Currently, ~1.5 million children are living with HIV, worldwide. ART cannot clear the HIV reservoir, resulting in the need for life-long adherence to therapy. Using SHIV-infected pediatric rhesus macaques (RMs) on long-term ART, we recently demonstrated transcriptionally active HIV in the gastrointestinal (GI) tract. Interestingly, the GI tract was also the first anatomic site of HIV reactivation after ART interruption. However, factors contributing to viral persistence in the GI tract remain unknown. Here, we tested whether pediatric GI microbiome can influence transcriptionally active HIV in this compartment.
Methods:
Feces were obtained from SHIV.CH505-infected virally-suppressed infant RMs. All RMs received the same triple-drug ART regimen (TFV+FTC+DTG) and were categorized into early ART (n=7, ART start: 4 days post-infection, pi), intermediate ART (n=9, ART start: 2 wks pi), late ART (n=14, ART start: 8 wks pi) and late ART+ therapeutic vaccine (n=8, ART start: 8 wks pi, vaccine on ART) groups. The fecal microbiome was profiled using 16s rRNA sequencing. Levels of 21 bile acids and 162 polar metabolites in feces were measured by LC-MS. To determine the impact of microbiome on cell-associated (CA)-RNA, we used Analysis of Compositions of Microbiomes with Bias Correction (bacterial taxa) and linear regression models (metabolites).
Results:
Irrespective of the regimen and duration of treatment, the ratio of SHIV CA-RNA/DNA was higher in the CD4+ T cells of rectal biopsies, compared to blood or lymph nodes. Treatment impacted both fecal-microbial and -metabolomic profiles. After correcting for therapy-based differences, we identified bacterial species Sarcina ventriculi, Lactococcus lactis and Treponema succinifaciens and metabolite adenosine diphosphate to be directly associated with increased SHIV CA-RNA. Additionally, the levels of Eubacterium hallii, Ruminococcus bromii, Terrisporobacter mayombei, D-glucose, D-glucuronic acid, citric acid, D-fructose and 4-hydroxyproline were inversely correlated with CA-RNA in the GI tract. Finally, a statistical mediation analysis indicated that the association of commensal Eubacterium hallii with reduced viral persistence was mediated via D-glucose.
Conclusions:
In the pediatric RM model, gut microbiome appeared to impact intestinal SHIV persistence. Building a GI microbiome from early life that can dampen HIV persistence and prevent reactivation may contribute to the goal of ART-free viral suppression in children living with HIV.