Oral Microbiota in Adolescents With Perinatal HIV-Infection

Objectives: The Pediatric HIV/AIDS Cohort AMP Study is a longitudinal study of perinatally HIV-infected (PHIV) and HIV-exposed, uninfected (PHEU) adolescents. PHIV were shown to have more caries than PHEU. Our objectives are 1) to compare levels of oral microbiota; 2) to assess whether caries-associated or periodontitis-associated microorganisms differed in PHIV versus PHEU.
Methods: V3-V4 regions of 16S rDNA from subgingival plaque (n=289) and throat wash expectorants (n=296) were sequenced using MiSeq. Human Oral Microbe Identification using NGS (HOMINGS) was used to assign bacterial identity mostly at species level. Zero-inflated negative binomial regression analyses were conducted separately for each taxon count and logistic regression analysis for caries or, separately, periodontitis, including interaction term between HIV group and taxon counts. Both regression analyses adjusted for samples' total reads and subjects' age. False discovery was controlled at 5% (Benjamini-Hochberg).
Results: Number of taxa detected and Simpson diversity differed little in PHIV and PHEU. However, group differences in average counts met false discovery threshold for 39 taxa in plaque and 22 taxa in throat washes, with more taxa exhibiting higher levels in PHEU than in PHIV. In plaque, Desulfolobus, Porphyromonas, and Filifactor alocis had higher levels in PHIV. Taxa exhibiting higher levels in PHEU included health-associated taxa Leptotrichia, Actinomyces, Rothia, and Corynebacterium. In plaque, PHIV and PHEU with periodontitis had similar associations with known periodontitis-associated taxa, Parvimonas micra, Porphyromonas gingivalis, and Filifactor alocis. In throat washes, PHIV and PHEU with caries had similar associations with caries-associated taxa, Streptococcus mutans, Scardovia and Lactobacillus.
Conclusions: Counts of many bacterial taxa differed between PHIV and PHEU. The PHIV oral microbiome seemed to have fewer “health”-associated taxa. The groups exhibited similar associations of caries or periodontitis with specific taxa, though detection of such differences was underpowered. Results are consistent with the hypothesis that HIV-infection, or its treatment, contributes to oral dysbiosis.