Changes in Gut Microbiota–Related Metabolites and Long-term Successful Weight Loss in Response to Weight-Loss Diets: The POUNDS Lost Trial

OBJECTIVE: Adiposity and the gut microbiota are both related to the risk of type 2 diabetes. We aimed to comprehensively examine how changes induced by a weight-loss diet intervention in gut microbiota–related metabolites, such as trimethylamine N-oxide (TMAO) and its precursors (choline and l-carnitine), were associated with improvements in adiposity and regional fat deposition. RESEARCH DESIGN AND METHODS: This study included 510 overweight and obese individuals who were randomly assigned one of four diets varying in macronutrient intake. We examined associations of 6-month changes in blood metabolites (TMAO, choline, and l-carnitine) with improvements in body weight (BW), waist circumference (WC), body fat composition, fat distribution, and resting energy expenditure (REE). RESULTS: Individuals with a greater reduction of choline (P < 0.0001) and l-carnitine (P < 0.01) rather than TMAO showed significant losses of BW and WC at 6 months. The reduction of choline was significantly predictive of decreases in body fat composition, fat distribution, and REE. Results of sensitivity analysis showed that the baseline diabetes risk status, such as the presence of hyperglycemia (31% of the total participants) and fasting glucose levels, did not modify the associations. Early changes in choline and l-carnitine were significantly predictive of weight loss over 2 years (P < 0.05 for all). Individuals with increases in choline or l-carnitine were 2.35-times (95% CI 1.38, 4.00) or 1.77-times (1.06, 2.95) more likely to fail to lose weight (–5% or more loss) at 2 years. CONCLUSIONS: Overweight and obese individuals who showed decreases in circulating choline or l-carnitine levels achieved greater improvements of adiposity and energy metabolism by eating a low-calorie weight-loss diet, suggesting that such metabolites are predictive of individuals’ response to the treatment. Further investigations are necessary to confirm our findings, particularly in a population with prediabetes that is more representative of the U.S. population with obesity.