(P27-006-24) Fecal Microbiota and Metabolites Predict Metabolic Health Features Across Various Dietary Patterns

Objectives: Adhering to a healthy dietary pattern reduces the risk of developing metabolic diseases and nourishes the gut microbiota. Thus, understanding how the gut microbiota may underpin connections between diet and metabolic health is of great clinical interest. Herein, we examined the contributions of bacterial taxa and metabolites in predicting metabolic health features across dietary patterns in a cohort of 118 adults.

Methods: Metabolic health features included waist circumference, blood pressure (BP), and fasting blood glucose, triglyceride (TG), and HDL concentrations. Four dietary pattern scores were derived from the Diet History Questionnaire (DHQ): Dietary Approaches to Stop Hypertension (DASH), Healthy Eating Index-2020, Mediterranean Diet, and Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND). ANCOM-BC was used to identify differentially abundant bacteria across dietary patterns. Fecal volatile fatty acid and bile acid concentrations were measured via GC-LC and LC-MS/MS, respectively. Hierarchical linear regression (HLR) was used to assess the contributions of individual fecal bacteria (Step 2) and metabolites (Step 3) in predicting metabolic features across each dietary pattern (Step 1) while adjusting for sex, age, BMI, self-reported physical activity, and household income.

Results: DASH diet was the greatest predictor of systolic and diastolic BP in Step 1 models. Further examination of connections between DASH and BP revealed that models were improved by adding Eubacterium eligens (systolic R² = 0.27; diastolic R² = 0.22) and E.xylanophilum (systolic R² = 0.30; diastolic R² = 0.25) independently in Step 2. Adding deoxycholic acid (Step 3) with E. xylanophilum resulted in the best overall model fit (systolic R² = 0.32; diastolic R² = 0.26). MIND diet tended to predict TG concentrations in Step 1 (p = 0.06). Adding E. eligens in Step 2 improved the model (R² = 0.16), with the greatest overall model fits observed when isobutyrate (R² = 0.22) and isovalerate (R² = 0.20) were independently incorporated in Step 3.

Conclusions: This exploratory analysis identifies specific fecal bacteria and microbial metabolites that may contribute to the effect different dietary patterns have on metabolic health parameters.

Funding Sources: This work was supported in part by the USDA NIFA AFRI Predoctoral Fellowship (2022-11290).