Gut Microbiota Shifts After a Weight Loss Program in Adults with Obesity: The WLM3P Study

Abstract

Background: The gut microbiota is increasingly recognized as a key modulator in obesity management, influencing host energy balance, lipid metabolism, and inflammatory pathways. With obesity prevalence continuing to rise globally, dietary interventions that promote beneficial microbial shifts are essential for enhancing weight loss outcomes and long-term health.

Objective: This study investigated the effects of the multicomponent Weight Loss Maintenance 3 Phases Program (WLM3P), which integrates caloric restriction, a high-protein low-carbohydrate diet, time-restricted eating (10h TRE), dietary supplementation (prebiotics and phytochemicals), and digital app-based support on gut microbiota composition compared to a standard low-carbohydrate diet (LCD) in adults with obesity. The analysis focused exclusively on the 6-month weight loss period corresponding to Phases 1 and 2 of the WLM3P intervention.

Methods: In this sub-analysis of a randomized controlled trial (ClinicalTrials.gov Identifier: NCT04192357), 58 adults with obesity (BMI 30.0-39.9 kg/m²) were randomized to the WLM3P (n = 29) or LCD (n = 29) groups. Stool samples were collected at baseline and 6 months for 16S rRNA sequencing. Alpha and beta diversity were assessed, and genus-level differential abundance was determined using EdgeR and LEfSe. Associations between microbial taxa and clinical outcomes were evaluated using regression models.

Results: After 6-month, the WLM3P group showed a significant increase in alpha diversity (p = 0.03) and a significant change in beta diversity (p < 0.01), while no significant changes were observed in the LCD group. Differential abundance analysis revealed specific microbial signatures in WLM3P participants, including increased levels of Faecalibacterium. Notably, higher Faecalibacterium abundance was associated with greater reductions in fat mass (kg, %) and visceral adiposity (cm²) in the WLM3P group compared to LCD (p < 0.01).

Conclusions: These findings suggest a potential microbiota-mediated mechanism in weight loss, where Faecalibacterium may enhance fat reduction effectiveness in the context of the WLM3P intervention.

Keywords: Faecalibacterium; dietary interventions; fat mass loss; gut microbiota; visceral fat loss; weight loss.

Figure 1

(a) Alpha diversity analysis evaluated by Shannon index at baseline and after 6 months of intervention, by groups (WLM3P and LCD). Pink boxes represent Shannon index values at baseline and blue boxes represent Shannon index after 6 months, compared by Wilcoxon test. (b) Principal coordinates analysis (PCoA) of beta diversity changes in WLM3P group and LCD using Bray−Curtis distances and compared by PERMANOVA test. Pink dots represent values at baseline and blue dots represent values after 6 months.

Figure 2

LEfSe analysis of microbial communities for WLM3P group (a) and for LCD group (b), by time. The figure illustrates the results of the Linear discriminant analysis Effect Size (LDA) applied to microbial communities. The bars represent the logarithmic LDA scores, indicating the effect size of different taxa across the experimental groups. Taxa with an LDA score greater than 2.0 are highlighted, suggesting significant differences in abundance. Blue colour represents the microbiota at baseline and red corresponds to microbiota composition after 6 months, providing a clear visual representation of the taxa that are significantly enriched or depleted in each group, by time.

Figure 3

Heatmap of Spearman’s correlation coefficients between anthropometric and biochemical determinations and the abundance of microbial genera normalized by CLR after six months of intervention. The left panel (a) corresponds to the WLM3P group, and the right panel (b) to the LCD group. Each cell represents the correlation between a clinical parameter (x-axis) and a microbial genus (y-axis), with the color indicating the strength and direction of the correlation (green for positive, red for negative). Significant correlations are marked with an asterisk (*). The strength of the correlations (Spearman’s rho) is color-coded according to the scale on the right, ranging from −1 to 1.

Figure 4

Predictive margins of the intervention groups (WLM3P and LCD) on weight-related outcomes as a function of the relative abundance of Faecalibacterium at 6 months. The graphs represent the linear prediction of (a) fat mass loss (kg), (b) fat mass loss (%), and (c) visceral fat loss (cm²) across a range of Faecalibacterium abundances in the gut microbiota normalized by Centered log ratio. In all three graphs, the blue line represents WLM3P group (blue line), and red line represents the LCD group. Error bars represent the 95% confidence intervals (CIs) for the predicted values, and the non-overlapping CIs in several cases suggest statistically significant differences between the two groups.