PROJECT SUMMARY/ABSTRACT Sepsis causes profound changes in host metabolism, including acute and persistent muscle loss. Among critically ill patients, body composition (including lean muscle mass) predicts mortality, length of stay, and functional status, and patients with sepsis experience dramatic changes in body composition. Survivors of sepsis commonly face profound functional deficits due to acute and persistent muscle catabolism. Our research group has recently discovered that in murine models of sepsis, the gut microbiome participates in the pathogenesis of acute muscle loss. Using murine models of sepsis, we have discovered that variation in gut microbiota (e.g., enrichment with butyrate-producing Lachnospiraceae spp.) explains variation in acute muscle loss. Modulation of gut microbiota can attenuate acute muscle loss. These novel findings suggest that the gut microbiome is an potential therapeutic target for sepsis-induced muscle loss. Yet no study has investigated whether gut microbiota contribute to acute muscle loss in a human sepsis cohort. Our central hypothesis is that specific gut bacteria (e.g., Lachnospiraceae spp.) calibrate the body’s metabolic response in sepsis, both propelling and impeding acute muscle loss. The rationale is that these discoveries will facilitate development of therapies for the prevention and treatment of muscle loss in sepsis. Specific Aim 1: To determine the interaction between gut microbiota and body composition in predicting clinical outcomes among septic patients. We hypothesize that core microbiome and morphometric predictors of 28-day outcomes are not independent, but instead have a significant interaction, supporting a causal relationship between the gut microbiome, body composition, and clinical trajectories. We will test this hypothesis using a retrospective cohort of 800 hospitalized patients with sepsis. We will characterize baseline gut microbiota (via 16S rRNA amplicon gene sequencing of admission rectal swabs), body composition (using morphomic analysis of CT scans performed within 48 hours of microbiome sampling), and 28-day clinical outcomes (using validated techniques of EMR interrogation). We will account for key confounding exposures (e.g., antibiotics and corticosteroids) using our rich clinical data capture. Specific Aim 2: To identify and interrogate the gut microbiota that predict acute muscle loss in patients with sepsis. We hypothesize that core features of baseline (admission) gut microbiota predict degree of muscle loss in septic patients. We will test this hypothesis using a subgroup of 300 patients in Aim 1 who had two CT scans performed more than 72 hours apart during the same hospitalization. We will use validated morphometric indices to test the hypothesis that baseline gut microbiota predict change in body composition (e.g., loss of lean muscle mass). We will perform metagenomic sequencing and in silico metabolic inference to identify candidate microbial and metabolic pa...