The oral cavity of chronically homeless adults is home for unusual extremophile environmental bacteria.

Chronically homeless adults (CHA) face limited healthcare access, frequent substance use, contact with animals that significantly alter their oral microenvironment. This study aimed to characterize the oral microbiota in this understudied population. We analyzed saliva from CHA and asymptomatic men with no history of homelessness in Mexico City. V1-V3 regions of the 16S rRNA gene were sequenced in extracted DNA. Taxonomic classification was performed using human and non-human databases. Diversity metrics were calculated, and informative species identified through machine learning. Microbial correlation networks were inferred using NetCoMi. Bacterial diversity was significantly higher in the CHA group. Although groups shared 369 species, only eight were exclusive to asymptomatic and 149 to CHA. Several taxa had not been reported in humans but found in extreme environments, like Megasphaera cerevisiae adapted to high ethanol and Syntrophocurvum alkaliphilum from a hypersaline lake. Marked differences in microbial network structure were observed between groups. These findings highlight the adaptability of extremophile bacteria to extremely altered human environments, as in marginalized population.