Project Summary/Abstract The vaginal microbiome is associated with important health outcomes, including bacterial vaginosis (BV), preterm birth, HIV risk, and cervical cancer, but mechanisms for these associations are unclear and current microbiome-directed therapies have low efficacy. Metabolome alterations including high-level production of amines like putrescine are a defining feature of BV and a major cause of symptoms, but underlying mechanisms and host responses to BV-associated metabolites including amines are poorly characterized. In preliminary data from a South African cohort with high BV prevalence, 60% of detectable metabolites in vaginal fluid significantly differ between optimal Lactobacillus-dominant bacterial communities and non-optimal non-Lactobacillusdominant (NLD) communities. Strongly BV- and NLD-associated metabolites included putrescine and imidazole propionate (ImP), both of which are known to have adverse host effects in non-vaginal diseases but whose production and effects remain largely uncharacterized in the genital tract. These results suggest bacteria comprising NLD communities express enzymes that produce these metabolites. The study will (1) characterize metabolome and metagenome composition in additional cohorts to validate findings and identify candidate enzymes for putrescine and ImP production, (2) characterize the enzymes in cultured vaginal bacteria and assess effects of known pharmacologic inhibitors, and (3) assess effects of putrescine and ImP on metabolic and inflammatory host responses and test whether they can be blocked by known host pathway inhibitors. The results will elucidate vaginal metabolome-related pathophysiology and may directly identify candidate therapies. The PI is an Instructor in Medicine at Harvard Medical School and the Division of Infectious Diseases at Massachusetts General Hospital (MGH); in addition to the research plan this proposal includes a five-year career development plan. His career goal is to