Gut microbiota signatures across BMI categories in adults and combined diagnostic value with clinical indicators for early obesity risk stratification

Background: The global prevalence of obesity continues to rise, with gut microbiota implicated as a key environmental factor in its development. However, microbial heterogeneity across BMI categories and its potential as an early diagnostic marker remain poorly understood. Here, we systematically compared the gut microbiota of adults with normal weight, overweight, and obesity, and evaluated the diagnostic utility of integrated microbial and clinical indicators.

Methods: In this study, a cohort of 214 participants was recruited and categorized into three groups based on BMI: normal weight (Nor, n = 85), overweight (Ow, n = 91), and obesity (Ob, n = 38). The gut microbiota was analyzed using 16S rRNA sequencing, which facilitated the assessment of α- and β-diversity, the identification of differentially abundant bacterial genera, and the evaluation of Spearman correlations with clinical indicators. Additionally, a diagnostic model was developed utilizing random forest, Least Absolute Shrinkage and Selection Operator (LASSO) regression and receiver operating characteristic (ROC) curve analysis.

Results: Clinical data showed no significant age differences among groups (p > 0.05), but the Ob group had markedly higher weight, BMI, total cholesterol, triglycerides, and uric acid levels (p < 0.001). Microbial analysis revealed reduced alpha diversity (Shannon index) in the Ob group and distinct microbial community structures among groups (PCoA, p < 0.001). LEfSe analysis showed enrichment of Escherichia-Shigella, Lactobacillus, and Parabacteroides in the Ob group, while Faecalibacterium, Enterobacter, and Agathobacter were predominant in the Nor group. The Ow group exhibited intermediate or specific enrichment, particularly in Ruminococcus gnavus group, Morganella, and Clostridium innocuum group. Correlation analysis indicated that bacteria abundant in the Nor group negatively correlated with BMI, TG, LDL-C, and positively with HDL-C, whereas those enriched in the Ow and Ob groups showed opposite trends. Four genera differentiated Ob from Ow individuals (AUC = 0.684), eight distinguished Ob from Nor (AUC = 0.787), and five separated Ow from Nor (AUC = 0.721). Integrating microbial and clinical data notably improved classification, with AUCs increasing to 0.908, 0.969, and 0.995, respectively.

Conclusion: Intestinal microecological profiles vary significantly across BMI categories, and a model combining microbiota with metabolic markers shows strong potential for early obesity detection.

1 Introduction

Obesity is a widespread metabolic disorder characterized by chronic and complex traits. Its etiology is multifactorial, encompassing genetic, environmental, and behavioral influences, and its prevalence is on the rise globally (Janić et al., 2025). This condition serves as a significant risk factor for various health issues, including cardiovascular disease, type 2 diabetes, non-alcoholic fatty liver disease, and certain cancers (Alexander et al., 2019; Wang et al., 2024; Yu et al., 2024; Leitzmann et al., 2025; Mo et al., 2025). The World Obesity Federation (2025) report projects that the number of adults with elevated body mass index (BMI) worldwide will increase to 2.9 billion by 2030. Similarly, China is experiencing a parallel trend, with the rates of obesity and overweight individuals expected to reach 70.5% by 2030 (World Obesity Federation, 2025). Furthermore, there is a concerning annual increase in the number of overweight adolescents and children, with a noticeable trend towards younger populations becoming obese [NCD Risk Factor Collaboration (NCD-RisC), 2024]. This situation poses a substantial burden on global public health systems. Consequently, it is of paramount clinical and social importance to gain a comprehensive understanding of the pathogenesis of obesity and to identify new targets for early intervention.

In recent years, advancements in microbiome research and high-throughput sequencing technologies have positioned gut microbiota as a significant factor influencing host metabolism and immune function, thereby emerging as a prominent subject of investigation in the field of metabolic disease research. The gut microbiota is essential for modulating energy intake and expenditure, regulating metabolic processes and inflammatory responses, and preserving the integrity of the intestinal barrier (Carmody and Bisanz, 2023; Wu et al., 2024a). Research indicates that germ-free mice subjected to fecal microbiota transplantation for the purpose of addressing obesity exhibited a marked increase in body weight and associated obesity symptoms, thereby implying a potential causal relationship between gut microbiota and obesity (Borrello et al., 2022; Yarahmadi et al., 2024). Furthermore, investigations have revealed that individuals with obesity often present with dysbiosis in their gut microbiota, accompanied by alterations in the abundance of critical bacterial genera. Recent extensive metagenomic analyses have identified Megamonas as a potential “obesogenic bacterium” among obese individuals in China, and its enrichment is significantly positively correlated with BMI and waist circumference (Wu et al., 2024b). Similarly, Enterobacter cloacae has been identified as a significant contributor to obesity. Research indicates that this bacterium constitutes approximately one-third of the total microbial population in the intestines of individuals with obesity. Furthermore, studies have demonstrated that Enterobacter cloacae is capable of inducing obesity and exacerbating inflammation in murine models (Fei and Zhao, 2013). Individuals with obesity frequently exhibit a reduction in the presence of anti-obesity bacteria, including species such as Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii (Liu et al., 2017; Song et al., 2023). And, research indicates that the abundance of Faecalibacterium prausnitzii is significantly and negatively correlated with both weight loss and the reduction of fat mass (Kahleova et al., 2020); the prevalence of Bacteroides thetaiotaomicron demonstrates a negative correlation with serum glutamate levels, which are associated with fat accumulation and weight gain (Liu et al., 2017). Furthermore, an increasing body of research has demonstrated that the intestinal microbiota engage in interactions with the host through the production of metabolites (Nicholson et al., 2012). Microorganisms are capable of generating essential metabolites, including short-chain fatty acids (SCFAs) and bile acid derivatives, through the fermentation of dietary fiber (Delzenne et al., 2025). SCFAs, particularly propionic acid and butyric acid, play a significant role in lipid metabolism by modulating hypothalamic activity via the blood–brain barrier, thereby decreasing food intake and inhibiting lipogenesis (Mukhopadhya and Louis, 2025). Likewise, the synthesis of secondary bile acids has been associated with enhanced energy expenditure and a reduction in fat accumulation (Ticho et al., 2019). In general, current research has established that individuals with obesity frequently exhibit specific traits, including dysbiosis in microbial diversity, a diminished presence of butyrate-producing bacteria (notably Faecalibacterium) and bacteria responsible for producing secondary bile acids, alongside an elevated abundance of pro-inflammatory bacteria. This phenomenon is referred to as the “obesity-type microecological imbalance” (Breton et al., 2022)^.

Existing research has established a significant relationship between obesity and intestinal microecology; however, there is considerable variability in the microecological profiles among individuals with obesity. Initial investigations highlighted an enrichment of Firmicutes in obese populations, accompanied by a reduction in Bacteroidetes, leading to the conclusion that an elevated Firmicutes/Bacteroidetes ratio is a characteristic feature of obesity (Breton et al., 2022). Recent research has introduced alternative viewpoints, with certain findings indicating an elevation in Bacteroidetes among individuals with obesity, or proposing that there is no significant variation in the Firmicutes/Bacteroidetes ratio (Magne et al., 2020). Correspondingly, a prevailing agreement among various studies indicates that individuals with obesity tend to demonstrate a reduced diversity of intestinal microbiota. However, research conducted by Japanese scholars has demonstrated that the diversity of intestinal microbiota in individuals with obesity is significantly higher than that found in individuals without obesity (Uema et al., 2022). These divergent findings highlight the variability in research results concerning intestinal microorganisms among various populations of individuals with obesity.

The alterations and composition of intestinal microbiota are influenced by a variety of factors, including demographic variations (such as ethnicity and geographic location), dietary patterns, lifestyle choices, sample size, and the methodologies employed in analysis (Brooks et al., 2018; Dwiyanto et al., 2021; Borrello et al., 2022; Andreu-Sánchez et al., 2025). More importantly, a study has demonstrated notable variations in the diversity and composition of gut microbiota among individuals with differing BMIs (Wang et al., 2025). Nonetheless, there remain significant gaps in the existing systematic research concerning the relationship between BMI and intestinal microbiota in China. Notably, studies that incorporate normal, overweight, and obese populations within a cohesive framework for comparative analysis are particularly limited. Furthermore, there is a deficiency in the investigation of microbial markers that could facilitate early intervention during the overweight phase. It is also important to acknowledge that BMI does not provide a direct measure of adipose tissue. Individuals with comparable BMI values may exhibit divergent health outcomes, indicating the necessity for additional metrics to assess the health implications associated with overweight and obesity (Sweatt et al., 2024). Consequently, to address these gaps, it is essential not only to undertake systematic research across BMI strata but also to explore the potential of gut microbiota, combined with routine metabolic markers, for stratifying metabolic health independent of BMI itself.

Our study was designed with two primary innovative aims: first, to systematically identify the unique and transitional gut microbial signatures of the overweight state, a critical yet understudied pre-obesity phase; and second, to develop and evaluate a predictive model that integrates these microbial markers with clinical metabolic indicators for obesity stratification, deliberately excluding BMI to test its intrinsic utility. By identifying core intestinal microorganisms associated with different BMI categories and clinical metabolic indicators independent of BMI, we developed a predictive model that integrates microbial markers significantly correlated with both BMI and clinical indicators. This study offers theoretical insights and identifies potential biomarkers for the early detection and targeted intervention of obesity, highlighting the variations in intestinal microorganisms among individuals with differing BMI classifications.

2 Methods and materials

2.1 Volunteer recruitment

This research received approval from the Ethics Committee of Zhongnan Hospital (Code: 2025054) and was carried out in strict compliance with the principles outlined in the Declaration of Helsinki. Participants classified as overweight or obesity were recruited from the Weight Loss and Metabolism Center at Zhongnan Hospital, while the normal control group was sourced from the Physical Examination Department. It is noteworthy that all participants had not engaged in any specific dietary interventions in the 12 months preceding their recruitment and maintained a varied diet to accurately reflect the intestinal flora composition associated with their typical dietary practices. Inclusion criteria: (1) participants aged between 18 and 60 years, irrespective of gender; (2) individuals must be in good health, without significant organic diseases or chronic infectious conditions; (3) participants must not have received antibiotics, probiotics, or medications that regulate intestinal function within the 3 months prior to sampling; (4) participants were categorized by BMI in accordance with the most recent guidelines for the diagnosis and treatment of obesity (Department of Medical Administration, National Health Commission of the People′s Republic of China, 2024): normal group (Nor): 18.5 ≤ BMI < 24.0 kg/m²; overweight group (Ow): 24.0 ≤ BMI < 28.0 kg/m²; Obesity group (Ob): BMI ≥ 28.0 kg/m². Exclusion criteria: (1) individuals who are pregnant or currently breastfeeding; (2) participants who have received antibiotics, probiotics, or immunomodulators, or who have undergone significant surgical interventions within the last 3 months; (3) individuals with a documented history of gastrointestinal disorders, including inflammatory bowel disease and constipation, as well as metabolic disorders such as type 2 diabetes and hypertension, and major systemic diseases such as cancers, chronic liver disease, or renal disease; (4)individuals with a history of substance abuse, alcoholism, or psychiatric or neurological disorders; and (5) individuals exhibiting atypical dietary practices or imbalanced dietary patterns, such as strict vegetarians or those following extreme high-fat diets. The clinical characteristics of all participants are presented in Table 1.

Table 1

Table 1. Comparison of baseline characteristics among different BMI groups.

2.2 Sample collection

Fecal samples were obtained utilizing a sterile stool collection kit. Participants were instructed to collect stool samples in their homes, following the provided guidelines. They utilized a sampling spoon to gather 1–2 g of stool, which was subsequently placed into a stool sampling tube. The samples were then promptly stored in the freezer of a domestic refrigerator at approximately −20 °C. Within 24 h, the samples were transported to the laboratory on dry ice and subsequently stored at −80 °C until analysis.

2.3 DNA extraction, PCR amplification and sequencing

Genomic DNA was isolated from stool samples utilizing the HiPure Stool DNA Mini Kit (Magen, Guangzhou, China), adhering closely to the protocols outlined by the manufacturer. Following the extraction process, the concentration of the DNA was quantified using the Qubit 4 Fluorometer (Thermo Fisher Scientific, United States) in conjunction with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, United States), which is designed to provide high sensitivity for the detection of low-concentration DNA. Afterwards the integrity of genomic DNA was assessed through 1% agarose gel electrophoresis. Following this, the V3-V4 hypervariable region of the bacterial 16S rRNA gene was amplified via polymerase chain reaction (PCR) utilizing universal primers (341F, 5′-CCTACGGGNGGCWGCAG-3′ and 805R, 5′-GACTACHVGGGTATCTAATCC-3′). The PCR was conducted using KAPA HiFi HotStart ReadyMix (Roche, Switzerland) and was executed on a Veriti 96-well gradient PCR instrument (Applied Biosystems, United States). The resulting amplified PCR products underwent library construction, were purified using AMPure XP magnetic beads (Beckman Coulter, United States), and were subsequently subjected to paired-end sequencing (2 × 250 bp) on the Illumina MiSeq PE250 platform.

2.4 Data analysis

Raw sequencing data were processed in QIIME2 (version 2021.2.0). Quality control was performed using the DADA2 plugin, which removed low-quality reads and chimeras and generated high-quality, non-redundant amplicon sequence variants (ASVs). After denoising, the minimum paired-end sequencing depth across all samples was approximately 50,000 reads. Taxonomic assignment was conducted against the SILVA database (version 138).

To ensure comparability across samples, all α- and β-diversity analyses were performed after rarefying samples to a depth of 20,000 reads. Alpha diversity was quantified using the Shannon index to evaluate within-sample microbial diversity. Beta diversity was assessed using the Bray–Curtis distance matrix, visualized by principal coordinate analysis (PCoA). Differences in community structure among groups were tested using ANOSIM and Adonis. To identify differentially abundant taxa across BMI categories, Linear Discriminant Analysis Effect Size (LEfSe) was applied.

Spearman correlation analysis was conducted to examine the relationships between microbial genera and clinical parameters, including LDL-C, TC, TG, and BMI. To further identify microbial or metabolic markers capable of distinguishing Nor, Ow, and Ob individuals, a random forest classifier was constructed at the genus level, followed by feature selection using LASSO regression. Receiver Operating Characteristic (ROC) analyses were performed to determine the predictive performance of each model.

To investigate whether metabolic and microbial markers could stratify obesity independently of BMI, BMI was excluded from the diagnostic model and used solely as a grouping variable. A multivariate model integrating microbial markers with metabolic indicators (TG, HDL-C, UA) was then developed, and its discriminatory capacity was evaluated using the area under the curve (AUC).

2.5 Statistical analysis

Statistical analyses were performed using R software (version 4.3.2). Group differences in clinical variables and alpha diversity indices were assessed using the Kruskal–Wallis test. Beta diversity differences were evaluated using ANOSIM and Adonis based on the Bray–Curtis distance matrix. For multiple testing in differential abundance and correlation analyses, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) method. Model performance from random forest, LASSO, and ROC analyses was quantified using the area under the curve (AUC), and using logistic regression analyses employing Firth’s penalized likelihood estimation were used to obtain bias-reduced, finite parameter estimates. Statistical significance was defined as p < 0.05 or FDR-adjusted q < 0.05.

3 Results

3.1 Baseline characteristics of participants

In this study, a total of 214 volunteers were recruited, comprising 38 individuals in the Ob group, 91 in the Ow group, and 85 in the Nor group (Table 1). The BMI values among the three groups exhibited statistically significant differences (p < 0.0001), while no significant age differences were observed (p = 0.092). The fasting blood glucose (Glu-AC) levels in the Ob group (7.64 ± 1.81 mmol/L) were significantly higher than those observed in the Ow group (6.26 ± 1.37 mmol/L) and the Nor group (5.19 ± 0.47 mmol/L). Additionally, lipid profiling indicated persistent abnormalities, with the Ob group demonstrating the highest levels of triglycerides (TG, 2.56 ± 0.82 mmol/L) and low-density lipoprotein cholesterol (LDL-C, 3.03 ± 0.79 mmol/L), as well as the lowest levels of high-density lipoprotein cholesterol (HDL-C, 1.19 ± 0.21 mmol/L) among all groups (p < 0.0001). The Ob group exhibited significantly elevated serum albumin (Alb, 46.43 ± 2.89 g/L) and uric acid (Ua, 473.23 ± 124.96 μmol/L) levels compared to Ow (Alb, 44.05 ± 4.28 g/L, Ua, 365.57 ± 98.08 μmol/L) and Nor groups (Alb, 42.33 ± 3.35 g/L, Ua, 310.47 ± 55.34 μmol/L) (p < 0.0001), confirming metabolic dysregulation in Ob. Notably, the Ob group demonstrated concurrent increases in hematological parameters (WBC: 6.44 ± 1.79 × 10⁹/L; HGB: 147.16 ± 15.71 g/L; PLT: 267.58 ± 64.89 × 10⁹/L) and blood pressure (SBP: 126.79 ± 12.52 mmHg; DBP: 85.11 ± 10.06 mmHg) relative to Ow (121.11 ± 11.27/78.65 ± 8.40 mmHg) and Nor (117.36 ± 5.78/80.22 ± 5.68 mmHg) groups. These findings suggest a triad of pathological manifestations in obesity: chronic low-grade inflammation, prothrombotic tendency, and elevated cardiovascular risk. No significant intergroup differences were observed in serum creatinine (Cr, p = 0.858) or urea nitrogen (urea, p = 0.155) levels among the Ob, Ow, and Nor groups, suggesting preserved renal function in Ob individuals. However, the Ob group exhibited a characteristic metabolic triad: systemic dysregulation (evidenced by elevated Glu-AC and lipid profiles), chronic low-grade inflammation (reflected by increased WBC count), and elevated cardiovascular risk (demonstrated by higher blood pressure). Notably, while the Ow group showed intermediate abnormalities in certain parameters compared to the Nor group, the degree of abnormality was substantially smaller than in the Ob group, suggesting Ow status as a critical window for preventive intervention.

3.2 Gut microbiota diversity

High-throughput sequencing of 214 fecal samples generated 13,027,572 high-quality 16S rRNA sequences, which were clustered into 97,141 ASVs. Venn analysis demonstrated distinct ASV distribution patterns among Ob, Ow, and Nor groups (Figure 1A), with the Ob group exhibiting the highest proportion of unique ASVs (28.7% vs. Ow 15.4% and Nor 9.8%). The Ob, Ow, and Nor groups contained 20,103 (30.9%), 26,516 (40.8%), and 13,931 (21.4%) unique ASVs, respectively, with only 613 ASVs (0.9%) shared among all three groups. Notably, the Nor and Ow groups shared 3,205 ASVs (4.9%), significantly more than the 185 ASVs (0.3%) shared between the Nor and Ob groups, demonstrating closer microbial community similarity between the Nor and Ow groups than between the Nor and Ob groups. The box plot of Shannon index demonstrated significant differences in α-diversity among Ob, Ow, and Nor groups (Figure 1B), with the Ob group exhibiting the lowest median value. PCoA based on Bray-Curtis distance revealed distinct clustering patterns, where both Ob and Ow groups showed significant separation from the Nor group along PCoA1 axis (Figure 1C, p < 0.001). ANOSIM analysis further confirmed the inter-group differences (Figure 1D, R = 0.066, p < 0.001), indicating statistically valid grouping with microbial community variations exceeding intra-group heterogeneity.

Figure 1

Figure 1. Gut microbiota characteristics across BMI groups. (A) Venn diagram illustrating the distribution of amplicon sequence variants (ASVs) among obesity, overweight, and normal-weight groups, with unique and shared ASV counts annotated. (B) Box plots of the Shannon diversity index (α-diversity) across the three groups. (C) Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity depicting β-diversity differences among groups. (D) ANOSIM results showing microbial community segregation (R statistic and p-value indicated; Bray–Curtis distance used). Higher R values indicate stronger between-group separation. Ob, n = 38; Ow, n = 91; Nor, n = 85. Statistical significance was assessed using the Kruskal–Wallis test, followed by the Benjamini–Hochberg procedure. * p < 0.05; ** p < 0.01.

3.3 Taxonomic composition of the gut microbiota

Microbial composition analysis identified 330 genera spanning 16 phyla across all samples. The top 10 phyla by average relative abundance are presented in Figure 2A, with Bacillota demonstrating the highest dominant phyla (Ob: 53.7%; Ow: 52.2%; Nor: 58.5%) and no significant intergroup differences (p > 0.05). Pseudomonadota, Bacteroidota, and Actinomycetota were similarly dominant across cohorts. At the genus level, Bacteroides, Escherichia-Shigella, and Faecalibacterium exhibited the highest relative abundances. Comparative analysis revealed Bacteroides and Faecalibacterium were enriched in the Nor group, whereas Escherichia-Shigella showed an inverse distribution pattern (Figure 2B). LefSe analysis (LDA score > 2, p < 0.05) combined with relative abundance heatmap (Figure 2C) identified distinct microbial markers across groups. The Ob group exhibited significant enrichment of Escherichia-Shigella, Collinsella, Lactobacillus, and Parabacteroides, with Escherichia-Shigella and Lactobacillus showing higher abundance in Ob compared to both Ow and Nor groups. Parabacteroides abundance in Ob was significantly elevated versus Nor but did not differ from Ow. Conversely, the Ow group demonstrated enrichment of Ruminococcus gnavus group, Morganella, Clostridium innocuum group, and Hungatella. Thirteen genera including Faecalibacterium, Enterobacter, and Agathobacter were preferentially enriched in Nor, with most showing significantly higher abundance than both Ob and Ow groups (except Erysipelotrichaceae UCG-003). These findings indicate that genus-level disparities primarily distinguish Nor from both Ob and Ow groups.

Figure 2

Figure 2. Differences analysis of gut microbial community structure in Ob, Ow, and Nor people. (A) Relative abundance histogram of the top 10 phyla in each BMI group. (B) Relative abundance histogram of the top 10 genera in each BMI group. (C) LEfSe results showing linear discriminant analysis (LDA) scores (LDA > 2.0) and corresponding heatmap of significantly different genera across groups. Ob, n = 38; Ow, n = 91; Nor, n = 85. Group differences were evaluated using the Kruskal–Wallis test with Benjamini–Hochberg false discovery rate correction. Colored diamonds indicate intergroup statistical comparisons (adjusted p-values).

3.4 Correlation analysis between gut microbial and clinical indicators

Spearman correlation analysis between LEfSe-identified differential bacterial genera and clinical indicators revealed BMI-associated microbial signatures across groups (Figure 3). In the Nor cohort, Coprococcus, Faecalibacterium, Lachnospiraceae UCG-001, Lachnospiraceae NK4A136 group, and Lachnospiraceae ND3007 group exhibited significant inverse correlations with BMI (p < 0.001). Conversely, the Ob group demonstrated positive correlations for Lactobacillus and Escherichia-Shigella, while Ow individuals showed similar positive associations with Ruminococcus gnavus group and Morganella abundance. Significant microbiota-clinical correlations were observed across BMI groups (Figure 3). In the Nor cohort, four genera, including Faecalibacterium and Fusicatenibacter, exhibited negative associations with Glu-AC, while six genera, including Lachnospiraceae NK4A136 group and Lachnospiraceae UCG-001, showed inverse correlations with TG. Six genera including Coprococcus were negatively associated with Alb, and four genera, including Faecalibacterium and Lachnospiraceae UCG-001, demonstrated negative correlations with LDL-C. Ten genera, including Lachnospiraceae NK4A136 group and Coprococcus, were positively correlated with HDL-C. Conversely, the Ow and Ob cohorts displayed reversed patterns. In the Ow group, the Ruminococcus gnavus group showed positive correlations with TG, LDL-C, TC, and UA but negative with HDL-C, while Morganella was positively associated with LDL-C. Clostridium innocuum group tracked positively with Alb, and Hungatella inversely with WBC. In the Ob group, Lactobacillus was positively correlated with Glu-AC and TG yet negatively with HDL-C, paralleled by Parabacteroides’s positive association with Glu-AC.

Figure 3

Figure 3. Heatmap showing spearman correlation coefficients between significantly altered genera and clinical parameters (BMI, TG, LDL-C, HDL-C, Glu-AC, Alb, UA, etc.). Red indicates positive correlation and blue indicates negative correlation. Statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001 (FDR-adjusted).

3.5 Clinical indicators improve diagnostic performance of bacterial markers

LEfSe analysis identified 21 differential microbial markers (Figure 2C), from which potential biomarkers for distinguishing Ow and Ob were selected using random forest analysis (Figure 4A). Subsequent LASSO regression refined the discriminatory microbial markers in the gut microbiome. Based on the top 10 species identified by random forest, we identified eight intestinal microorganisms that effectively differentiated the Nor and Ob groups, including Lachnospiraceae ND3007 group, Lachnospiraceae NK4A136 group, Fusicatenibacter, Coprococcus, Lactobacillus, Ruminococcus gnavus group, Escherichia-Shigella, and Faecalibacterium. The composite microbial marker demonstrated diagnostic potential with an AUC of 0.787 (Figures 4B–D). Furthermore, four markers including Collinsella, Fusicatenibacter, Lactobacillus, and Lachnospiraceae ND3007 group were effectively differentiated Ob from Ow, with an AUC value of 0.684 (Figures 4E–G). Similarly, five microorganisms including Ruminococcus gnavus group, Lachnospiraceae NK4A136 group, Coprococcus, Faecalibacterium, Escherichia−Shigella, effectively distinguished Nor from Ow individuals, achieving an AUC of 0.721 (Figures 4H–J).

Figure 4

Figure 4. Identification of gut microbial biomarkers distinguishing BMI groups. (A) Mean decrease Gini (MDG) values from random forest ranking the top 10 discriminatory microbial genera. (B) LASSO regression model for Nor vs. Ob classification. (C) Coefficients of selected composite microbial markers (blue: negatively associated with Ob; orange: positively associated). (D) ROC curve for Nor vs. Ob discrimination. (E) LASSO model for Ob vs. Ow classification. (F) Coefficients of microbial markers distinguishing Ob from Ow. (G) ROC curve for Ob vs. Ow classification. (H–J) LASSO model, coefficients, and ROC curve for Nor vs. Ow classification. AUC indicates area under the ROC curve; 95% confidence intervals are shown in brackets. All models were trained using stratified 10-fold cross-validation, and p-values were FDR-corrected, ROC curves and AUC values were derived from Firth-corrected models.

In addition, random forest analysis identified 10 key clinical indicators (excluding BMI) for distinguishing Ow and Ob (Figure 5A). LASSO regression revealed seven indicators (Glu-AC, TC, Alb, DBP, WBC, UA, SBP) that effectively differentiated Ob from Ow, achieving an AUC of 0.822 (Figures 5B–D). Similarly, six indicators (HDL-C, Glu-AC, LDL-C, TC, TG, DBP) distinguished Nor from Ob with perfect discrimination (AUC = 0.895; Figures 5E–G), while seven indicators (HDL-C, Glu-AC, LDL-C, TG, Alb, SBP, UA) differentiated Nor from Ow (AUC = 0.971; Figures 5H–J). The diagnostic performance was further enhanced through combined analysis. Integrating key clinical indicators with bacterial biomarkers improved discrimination between Ob and Ow (AUC increased from 0.684 to 0.908; Figure 5K). Similarly, combined models for Nor vs. Ow and Nor vs. Ob achieved AUC values of 0.969 and 0.995, respectively (Figures 5L,M). These findings demonstrate the synergistic potential of clinical indicators and microbial biomarkers for early obesity detection.

Figure 5

Figure 5. Enhancement of BMI-group discrimination through clinical indicators alone and combined with microbial markers. (A) Mean decrease Gini (MDG) values ranking the top 10 discriminative clinical indicators (excluding BMI) in random forest analysis. (B) LASSO regression model for Ob vs. Ow classification. (C) Coefficients of selected clinical predictors. (D) ROC curve for Ob vs. Ow classification. (E–G) LASSO model, coefficients, and ROC curve for Nor vs. Ob classification. (H–J) LASSO model, coefficients, and ROC curve for Nor vs. Ow classification. (K–M) ROC curves for combined microbial–clinical models for Ob vs. Ow (K), Nor vs. Ow (L), and Nor vs. Ob (M). All AUC values include 95% confidence intervals. All analyses were performed using stratified 10-fold cross-validation, and p-values were FDR-corrected, ROC curves and AUC values were derived from Firth-corrected models.

4 Discussion

Over the past decade, the gut microbiota has been increasingly recognized as a key contributor to obesity pathogenesis. Substantial evidence demonstrates that obese individuals exhibit significant alterations in microbial composition, functional capacity, and metabolic output, particularly characterized by reduced butyrate-producing bacteria, elevated pro-inflammatory taxa, and impaired SCFAs biosynthesis (Bouter et al., 2017; Sun et al., 2018; Aron-Wisnewsky et al., 2021). However, as research cohorts expand and analytical techniques advance, the conventional definition of an “obesity-associated microbiome” requires critical reevaluation (Magne et al., 2020). Critically, most prior studies have treated obesity as a single state (e.g., obesity vs. normal) or focused exclusively on extreme BMI groups, overlooking the transitional “overweight” phase. From a public health perspective, this intermediate stage represents both a precursor to obesity and the most reversible window for microbial ecological interventions. Our study provides novel insights into the gut microbiota’s role in obesity by focusing on the often-overlooked overweight population and by constructing a diagnostic framework that deliberately excludes BMI.

Our results demonstrate that Ob exhibits a significantly lower Shannon index compared to Nor, reflecting reduced α-diversity of gut microbiota—a finding consistent with prior research (Hu et al., 2022). It is noteworthy that the absolute values of the Shannon index in our cohort (primarily ranging between 2 and 3.7 for the Nor group) appear lower than some published benchmarks. This discrepancy may be attributed to methodological variations (e.g., DNA extraction kits, 16S rRNA gene region targeted [V3-V4], sequencing platform, and bioinformatic pipelines), population-specific factors (e.g., genetic background, geography, and diet of our Chinese cohort), or a combination thereof. While cross-study comparisons of absolute diversity values should be interpreted with caution, the significant relative difference observed between the Ob and Nor groups within our standardized analytical framework remains valid and informative. Notably, no significant α-diversity difference was observed between the Ow and Ob groups. β-diversity analysis revealed a gradient distribution of microbial communities across BMI categories, with Ob and Nor groups forming distinct clusters and Ow positioned intermediately, suggesting BMI as a key driver of microbiota restructuring (Wang et al., 2025).

At the genus level, Coprococcus, Faecalibacterium, Lachnospiraceae UCG-001, and Lachnospiraceae NK4A136 group were significantly enriched in the Nor group. Notably, Faecalibacterium-a genus consistently linked to metabolic health-exhibits reduced abundance in obese individuals, as demonstrated by prior studies (Maioli et al., 2021; Islam et al., 2022). The Lachnospiraceae NK4A136 group and Lachnospiraceae UCG-001 exhibited a positive correlation with SCFA production capacity, a metabolic pathway critically linked to enhanced energy metabolism and attenuated inflammatory responses (Ma et al., 2020). Furthermore, our findings corroborate prior evidence indicating an inverse association between the Lachnospiraceae NK4A136 group and obesity (Companys et al., 2021). Notably, while certain studies report increased Coprococcus abundance in obese individuals, significant functional heterogeneity exists within this genus. Multiple independent cohort and intervention studies demonstrate that specific species (e.g., C. eutactus) correlate with enhanced insulin sensitivity, reduced BMI, and possess butyrate-producing and anti-inflammatory capabilities (Yang et al., 2023; González et al., 2025). Thus, the Coprococcus enrichment observed in our normal-weight group likely represents a “beneficial subpopulation,” with its depletion potentially indicating loss of functionally protective microbiota in obesity. However, the Ruminococcus gnavus group, Morganella, and Clostridium innocuum group were significantly enriched in the Ow cohort. Notably, both the Ruminococcus gnavus group and Morganella are associated with pro-obesity effects (Grahnemo et al., 2022; Sun et al., 2023), whereas the Clostridium innocuum group-a recognized probiotic-exhibits intestinal regulatory and potential anti-obesity properties (Bian et al., 2022; Mei et al., 2022). These findings suggest that the gut microbiota in the Ow stage may represent a transitional equilibrium, characterized by the concurrent expansion of pro-inflammatory taxa and compensatory proliferation of protective flora. The Escherichia-Shigella, Collinsella, and Parabacteroides genera were significantly enriched in the Ob group. Escherichia-Shigella exhibited obesogenic and pro-inflammatory properties in the gut of obese individuals, showing significant positive correlations with obesity markers including BMI (Hong et al., 2024; Gao et al., 2025). Similarly, Collinsella demonstrated consistent positive associations with obesity-related parameters such as TG, TC, and BMI (Gallardo-Becerra et al., 2020; Wang et al., 2020). While Parabacteroides is generally regarded as a potentially beneficial genus, certain studies have reported its positive association with BMI (Chiu et al., 2014; Guan et al., 2021). Notably, P. merdae and P. distasonis demonstrate significant enrichment in individuals with hypertension and diabetes, with Parabacteroides exhibiting particularly high abundance in obese patients with type 2 diabetes. These findings suggest that specific Parabacteroides species may contribute to metabolic dysregulation (Yan et al., 2017; Hasain et al., 2020; Yang et al., 2021). The Parabacteroides genus demonstrates significantly higher abundance in individuals with inflammatory bowel disease or elevated inflammatory indices, suggesting a potential association between Parabacteroides proliferation and systemic inflammation levels (Aranaz et al., 2021; Iliodromiti et al., 2023). This observation aligns with the elevated WBC counts consistently reported in obese populations, which further corroborates systemic inflammatory responses. Notably, our findings reveal increased Parabacteroides abundance in obese individuals, potentially indicating the expansion of specific Parabacteroides subspecies linked to metabolic dysregulation and pro-inflammatory alterations in the gut microbiota. Of particular note is that the abundance of Lactobacillus was significantly increased in the obese group, a result that seems to contradict the “beneficial” image of this genus that is often widely used in probiotic preparations. However, in fact, the Lactobacillus genus is highly heterogeneous, and different species and even different strains of the same species have significant differences in their effects on host metabolism. For example, Lactobacillus acidophilus has been shown to be associated with weight gain, while Lactobacillus plantarum has been shown to improve obesity (Million et al., 2012). Furthermore, a study found that the number of Lactobacillus species in obese patients was significantly higher than lean controls, a finding consistent with our results (Armougom et al., 2009). However, the limitations of 16S rRNA sequencing preclude precise species-level differentiation. We therefore hypothesize that the enriched Lactobacillus population in obesity may represent an obesity-promoting subpopulation or strains opportunistically amplified under metabolic stress. These observations underscore the need to avoid oversimplified classifications of Lactobacillus as uniformly “beneficial” or “harmful,” instead advocating for functional subpopulation characterization through future strain-level analyses.

The Spearman correlation network we constructed revealed the multi-layer associations between differential bacterial genera and clinical metabolic indicators. In the Nor group, butyrate-producing bacteria including Faecalibacterium, Coprococcus, Lachnospiraceae ND3007 group, and Lachnospiraceae NK4A136 group were significantly enriched. These genera exhibited negative correlations with BMI, TG, and LDL-C, while demonstrating positive associations with HDL-C. These observations align with established evidence that elevated butyrate levels correlate with reduced adiposity and inverse associations with obesity markers and blood glucose levels (Amiri et al., 2023; Zhang et al., 2024). These findings are consistent with our results.

However, the Ruminococcus gnavus group in the Ow cohort exhibited inverse correlations with obesity-related metabolic markers (BMI, TC, and LDL-C) compared to the Nor group. This bacterial group has been widely recognized as a potential biomarker for multiple pathologies, including obesity, depression, and inflammatory bowel disease (Henke et al., 2019; Juge, 2023; He et al., 2025). Its elevated abundance in obese individuals aligns with existing evidence linking R. gnavus to adverse metabolic phenotypes, such as increased adiposity, elevated waist circumference, hypertriglyceridemia, and reduced HDL-C levels (Grahnemo et al., 2022). Notably, Hungatella exhibits a negative correlation with WBC counts, implying its potential protective role in modulating host inflammatory responses. This observation aligns with existing evidence demonstrating Hungatella’s capacity to enhance intestinal barrier function and immune regulation through SCFA production, which may suppress low-grade inflammation (Xing et al., 2025). While the exact mechanisms are not yet fully understood, these findings indicate that Hungatella may engage in compensatory upregulation of anti-inflammatory pathways in response to overweight conditions. This hypothesis merits additional research. Notably, the Escherichia−Shigella genus was significantly enriched in the Ob group and exhibited a positive correlation with BMI-a finding consistent with prior reports (Hong et al., 2024; Gao et al., 2025). While most studies associate Lactobacillus with metabolic benefits, our data revealed its positive correlation with Glu-AC and TG levels, aligning with limited but corroborative evidence (Li et al., 2022; Crudele et al., 2023; Feng et al., 2023). This discrepancy suggests strain- or host-specific metabolic effects of Lactobacillus. Conversely, Parabacteroides showed exclusive positive correlation with Glu-AC, potentially reflecting its dietary adaptation to high-fat/high-sugar conditions (de Wouters d'Oplinter et al., 2021).

Utilizing machine learning models, we successfully identified bacterial genera capable of distinguishing BMI stratification, with model AUC values ranging from 0.721 to 0.787. Notably, integrating these microbial markers with conventional clinical metabolic indicators (e.g., TG, HDL-C, UA) substantially enhanced diagnostic performance. It is important to highlight that, given that BMI has been utilized as the primary grouping variable in this research, this approach serves to mitigate the potential for pseudo-correlations arising from the interaction of cross-variables within the model, on the one hand, BMI alone often inadequately classifies obesity-related diseases at the individual level, as individuals with similar BMI values frequently exhibit divergent metabolic phenotypes (Coral et al., 2025). Therefore, we deliberately excluded BMI from the diagnostic model during the modeling process to test whether other metabolic indicators have independent stratification capabilities. The results demonstrated that key metabolic indicators (including TG, HDL-C, and UA) significantly enhanced the model’s discriminative performance independent of BMI, confirming their intrinsic predictive value for obesity risk stratification. Notably, when integrated with microbial markers, the combined model exhibited a substantial increase in AUC values, indicating that this “microbiome-metabolic” framework maintains robust diagnostic potential without BMI data. These findings highlight the clinical translatability of this integrated approach for non-invasive metabolic risk assessment.

While this study provides novel insights into structural variations of gut microbiota across different BMI categories, several limitations should be acknowledged. First, the cross-sectional design precludes causal inference between microbial alterations and metabolic indicators. While we propose that the microbial signatures identified in the overweight group hold potential for early risk stratification, this cross-sectional snapshot cannot demonstrate that these signatures actually predict future progression to obesity. Future studies incorporating longitudinal follow-up of overweight individuals are essential to validate the predictive power of these microbial markers for true early risk stratification. Such studies would track whether individuals with the ‘dysbiotic’ overweight profile we identified are indeed more likely to develop obesity and associated metabolic disorders over time. Second, the use of 16S rRNA sequencing limits taxonomic resolution and does not allow species- or strain-level differentiation. Thus, the observed associations—such as those involving Lactobacillus—may reflect underlying strain variability that cannot be resolved with 16S data alone. Future work using metagenomic approaches will be required to validate these findings and characterize functional differences more accurately. Third, key lifestyle factors (e.g., diet, physical activity, and sleep patterns) were not systematically incorporated into the analysis. These variables not only influence gut microbiota composition but may also confound the observed associations between BMI and microbial profiles. Future studies should employ comprehensive dietary frequency surveys and energy intake assessments, coupled with standardized exercise and sleep monitoring, to mitigate confounding effects and strengthen the model’s explanatory power. Additionally, the study’s single-center design and geographically restricted sampling may limit the generalizability of the findings. Given that gut microbiota composition is influenced by multifactorial determinants (e.g., genetics, diet, and environmental exposures), future multi-center studies involving diverse ethnic populations are warranted to validate and enhance the external validity of these conclusions.

5 Conclusion

This study revealed BMI-related alterations in gut microbiota composition and diversity, and demonstrated that integrating microbial features with clinical indicators enhances obesity stratification beyond BMI alone. These findings provide valuable reference evidence for the early stratification of obesity risk, suggesting that microbiota-informed assessment may complement traditional anthropometric measures. To advance toward practical application, future studies should include longitudinal cohorts, expanded sample sizes, and higher-resolution microbial profiling to clarify causal relationships and to determine whether these microbial signatures can reliably support early risk identification or guide individualized intervention strategies.

Data availability statement

The 16S rRNA gene sequencing data generated in this study have been deposited in the NCBI BioProject database under accession number PRJNA1394711.

Ethics statement

The studies involving humans were approved by the Ethics Committee of Zhongnan Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

XC: Methodology, Investigation, Writing – original draft, Data curation. YW: Writing – original draft, Methodology, Investigation, Data curation. YL: Formal analysis, Writing – original draft, Validation. HW: Data curation, Investigation, Writing – review & editing. YY: Data curation, Investigation, Writing – review & editing. JT: Writing – review & editing, Data curation, Validation. HS: Methodology, Conceptualization, Writing – review & editing, Formal analysis. ZD: Conceptualization, Methodology, Funding acquisition, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Key R&D Program of China (grant no. 2024YFC3505100), the Medical Science and Technology Innovation Platform of Zhongnan Hospital, Wuhan University (grant no. PTXM2021016); the General Project of the Health Commission of Hubei Province (grant no. WJ2019M204).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Alexander, M., Loomis, A. K., van der Lei, J., Duarte-Salles, T., Prieto-Alhambra, D., Ansell, D., et al. (2019). Non-alcoholic fatty liver disease and risk of incident acute myocardial infarction and stroke: findings from matched cohort study of 18 million European adults. BMJ 367:l5367. doi: 10.1136/bmj.l5367,

PubMed Abstract | Crossref Full Text | Google Scholar

Amiri, P., Hosseini, S. A., Roshanravan, N., Saghafi-Asl, M., and Tootoonchian, M. (2023). The effects of sodium butyrate supplementation on the expression levels of PGC-1α, PPARα, and UCP-1 genes, serum level of GLP-1, metabolic parameters, and anthropometric indices in obese individuals on weight loss diet: a study protocol for a triple-blind, randomized, placebo-controlled clinical trial. Trials 24:489. doi: 10.1186/s13063-022-06891-9,

PubMed Abstract | Crossref Full Text | Google Scholar

Andreu-Sánchez, S., Blanco-Míguez, A., Wang, D., Golzato, D., Manghi, P., Heidrich, V., et al. (2025). Global genetic diversity of human gut microbiome species is related to geographic location and host health. Cell 188, 3942–3959.e3949. doi: 10.1016/j.cell.2025.04.014

Crossref Full Text | Google Scholar

Aranaz, P., Ramos-Lopez, O., Cuevas-Sierra, A., Martinez, J. A., Milagro, F. I., and Riezu-Boj, J. I. (2021). A predictive regression model of the obesity-related inflammatory status based on gut microbiota composition. Int. J. Obes. 45, 2261–2268. doi: 10.1038/s41366-021-00904-4,

PubMed Abstract | Crossref Full Text | Google Scholar

Armougom, F., Henry, M., Vialettes, B., Raccah, D., and Raoult, D. (2009). Monitoring bacterial community of human gut microbiota reveals an increase in Lactobacillus in obese patients and methanogens in anorexic patients. PLoS One 4:e7125. doi: 10.1371/journal.pone.0007125,

PubMed Abstract | Crossref Full Text | Google Scholar

Aron-Wisnewsky, J., Warmbrunn, M. V., Nieuwdorp, M., and Clément, K. (2021). Metabolism and metabolic disorders and the microbiome: the intestinal microbiota associated with obesity, lipid metabolism, and metabolic health-pathophysiology and therapeutic strategies. Gastroenterology 160, 573–599. doi: 10.1053/j.gastro.2020.10.057,

PubMed Abstract | Crossref Full Text | Google Scholar

Bian, X., Shi, T., Wang, Y., Ma, Y., Yu, Y., Gao, W., et al. (2022). Gut dysbiosis induced by antibiotics is improved by tangerine pith extract in mice. Nutr. Res. 101, 1–13. doi: 10.1016/j.nutres.2022.01.005,

PubMed Abstract | Crossref Full Text | Google Scholar

Borrello, K., Lim, U., Park, S. Y., Monroe, K. R., Maskarinec, G., Boushey, C. J., et al. (2022). Dietary intake mediates ethnic differences in gut microbial composition. Nutrients 14:660. doi: 10.3390/nu14030660,

PubMed Abstract | Crossref Full Text | Google Scholar

Bouter, K. E., van Raalte, D. H., Groen, A. K., and Nieuwdorp, M. (2017). Role of the gut microbiome in the pathogenesis of obesity and obesity-related metabolic dysfunction. Gastroenterology 152, 1671–1678. doi: 10.1053/j.gastro.2016.12.048

Crossref Full Text | Google Scholar

Breton, J., Galmiche, M., and Déchelotte, P. (2022). Dysbiotic gut Bacteria in obesity: An overview of the metabolic mechanisms and therapeutic perspectives of next-generation probiotics. Microorganisms 10:452. doi: 10.3390/microorganisms10020452,

PubMed Abstract | Crossref Full Text | Google Scholar

Brooks, A. W., Priya, S., Blekhman, R., and Bordenstein, S. R. (2018). Gut microbiota diversity across ethnicities in the United States. PLoS Biol. 16:e2006842. doi: 10.1371/journal.pbio.2006842,

PubMed Abstract | Crossref Full Text | Google Scholar

Carmody, R. N., and Bisanz, J. E. (2023). Roles of the gut microbiome in weight management. Nat. Rev. Microbiol. 21, 535–550. doi: 10.1038/s41579-023-00888-0,

PubMed Abstract | Crossref Full Text | Google Scholar

Chiu, C. M., Huang, W. C., Weng, S. L., Tseng, H. C., Liang, C., Wang, W. C., et al. (2014). Systematic analysis of the association between gut flora and obesity through high-throughput sequencing and bioinformatics approaches. Biomed. Res. Int. 2014:906168. doi: 10.1155/2014/906168,

PubMed Abstract | Crossref Full Text | Google Scholar

Companys, J., Gosalbes, M. J., Pla-Pagà, L., Calderón-Pérez, L., Llauradó, E., Pedret, A., et al. (2021). Gut microbiota profile and its association with clinical variables and dietary intake in overweight/obese and lean subjects: a cross-sectional study. Nutrients 13:2032. doi: 10.3390/nu13062032,

PubMed Abstract | Crossref Full Text | Google Scholar

Coral, D. E., Smit, F., Farzaneh, A., Gieswinkel, A., Tajes, J. F., Sparsø, T., et al. (2025). Subclassification of obesity for precision prediction of cardiometabolic diseases. Nat. Med. 31, 534–543. doi: 10.1038/s41591-024-03299-7,

PubMed Abstract | Crossref Full Text | Google Scholar

Crudele, L., Gadaleta, R. M., Cariello, M., and Moschetta, A. (2023). Gut microbiota in the pathogenesis and therapeutic approaches of diabetes. EBioMedicine 97:104821. doi: 10.1016/j.ebiom.2023.104821,

PubMed Abstract | Crossref Full Text | Google Scholar

de Wouters d'Oplinter, A., Rastelli, M., Van Hul, M., Delzenne, N. M., Cani, P. D., and Everard, A. (2021). Gut microbes participate in food preference alterations during obesity. Gut Microbes 13:1959242. doi: 10.1080/19490976.2021.1959242,

PubMed Abstract | Crossref Full Text | Google Scholar

Delzenne, N. M., Bindels, L. B., Neyrinck, A. M., and Walter, J. (2025). The gut microbiome and dietary fibres: implications in obesity, cardiometabolic diseases and cancer. Nat. Rev. Microbiol. 23, 225–238. doi: 10.1038/s41579-024-01108-z,

PubMed Abstract | Crossref Full Text | Google Scholar

Department of Medical Administration, National Health Commission of the People′s Republic of China (2024). Guideline for diagnosis and treatment of obesity (2024 edition). Chin. J. Dig. Surg. 23, 1237–1260. doi: 10.3760/cma.j.cn115610-20241017-00455

Crossref Full Text | Google Scholar

Dwiyanto, J., Hussain, M. H., Reidpath, D., Ong, K. S., Qasim, A., Lee, S. W. H., et al. (2021). Ethnicity influences the gut microbiota of individuals sharing a geographical location: a cross-sectional study from a middle-income country. Sci. Rep. 11:2618. doi: 10.1038/s41598-021-82311-3,

PubMed Abstract | Crossref Full Text | Google Scholar

Fei, N., and Zhao, L. (2013). An opportunistic pathogen isolated from the gut of an obese human causes obesity in germfree mice. ISME J. 7, 880–884. doi: 10.1038/ismej.2012.153,

PubMed Abstract | Crossref Full Text | Google Scholar

Feng, X., Deng, M., Zhang, L., and Pan, Q. (2023). Impact of gut microbiota and associated mechanisms on postprandial glucose levels in patients with diabetes. J Transl Int Med 11, 363–371. doi: 10.2478/jtim-2023-0116,

PubMed Abstract | Crossref Full Text | Google Scholar

Gallardo-Becerra, L., Cornejo-Granados, F., García-López, R., Valdez-Lara, A., Bikel, S., Canizales-Quinteros, S., et al. (2020). Metatranscriptomic analysis to define the Secrebiome, and 16S rRNA profiling of the gut microbiome in obesity and metabolic syndrome of Mexican children. Microb. Cell Factories 19:61. doi: 10.1186/s12934-020-01319-y,

PubMed Abstract | Crossref Full Text | Google Scholar

Gao, Y., Liu, W., Ma, X., Xue, M., Li, Y., Bi, S., et al. (2025). The role of intestinal microbiota and its metabolites in the occurrence and intervention of obesity. Front. Microbiol. 16:1559178. doi: 10.3389/fmicb.2025.1559178,

PubMed Abstract | Crossref Full Text | Google Scholar

González, A., Fullaondo, A., and Odriozola, A. (2025). In search of healthy ageing: a microbiome-based precision nutrition approach for type 2 diabetes prevention. Nutrients 17:1877. doi: 10.3390/nu17111877,

PubMed Abstract | Crossref Full Text | Google Scholar

Grahnemo, L., Nethander, M., Coward, E., Gabrielsen, M. E., Sree, S., Billod, J. M., et al. (2022). Cross-sectional associations between the gut microbe Ruminococcus gnavus and features of the metabolic syndrome. Lancet Diabetes Endocrinol. 10, 481–483. doi: 10.1016/s2213-8587(22)00113-9

Crossref Full Text | Google Scholar

Guan, H.-h., Liu, C.-l., Pu, Y.-n., Dai, Y.-x., Zong, G., and Yan, Z. (2021). The effect of diet mediated by gut microbiota on adiposity phenotypes. Fudan Univ. J. Med. Sci. 48, 721–729. doi: 10.3969/j.issn.1672-8467.2021.06.002

Crossref Full Text | Google Scholar

Hasain, Z., Mokhtar, N. M., Kamaruddin, N. A., Mohamed Ismail, N. A., Razalli, N. H., Gnanou, J. V., et al. (2020). Gut microbiota and gestational diabetes mellitus: a review of host-gut microbiota interactions and their therapeutic potential. Front. Cell. Infect. Microbiol. 10:188. doi: 10.3389/fcimb.2020.00188,

PubMed Abstract | Crossref Full Text | Google Scholar

He, X., Hu, M., Xu, Y., Xia, F., Tan, Y., Wang, Y., et al. (2025). The gut-brain axis underlying hepatic encephalopathy in liver cirrhosis. Nat. Med. 31, 627–638. doi: 10.1038/s41591-024-03405-9,

PubMed Abstract | Crossref Full Text | Google Scholar

Henke, M. T., Kenny, D. J., Cassilly, C. D., Vlamakis, H., Xavier, R. J., and Clardy, J. (2019). Ruminococcus gnavus, a member of the human gut microbiome associated with Crohn's disease, produces an inflammatory polysaccharide. Proc. Natl. Acad. Sci. USA 116, 12672–12677. doi: 10.1073/pnas.1904099116,

PubMed Abstract | Crossref Full Text | Google Scholar

Hong, J., Fu, T., Liu, W., Du, Y., Bu, J., Wei, G., et al. (2024). An update on the role and potential molecules in relation to Ruminococcus gnavus in inflammatory bowel disease, obesity and diabetes mellitus. Diabetes Metab. Syndr. Obes. 17, 1235–1248. doi: 10.2147/dmso.S456173

Crossref Full Text | Google Scholar

Hu, J., Guo, P., Mao, R., Ren, Z., Wen, J., Yang, Q., et al. (2022). Gut microbiota signature of obese adults across different classifications. Diabetes Metab. Syndr. Obes. 15, 3933–3947. doi: 10.2147/dmso.S387523,

PubMed Abstract | Crossref Full Text | Google Scholar

Iliodromiti, Z., Triantafyllou, A. R., Tsaousi, M., Pouliakis, A., Petropoulou, C., Sokou, R., et al. (2023). Gut microbiome and neurodevelopmental disorders: a link yet to be disclosed. Microorganisms 11:487. doi: 10.3390/microorganisms11020487,

PubMed Abstract | Crossref Full Text | Google Scholar

Islam, M. R., Arthur, S., Haynes, J., Butts, M. R., Nepal, N., and Sundaram, U. (2022). The role of gut microbiota and metabolites in obesity-associated chronic gastrointestinal disorders. Nutrients 14:624. doi: 10.3390/nu14030624,

PubMed Abstract | Crossref Full Text | Google Scholar

Janić, M., Janež, A., El-Tanani, M., and Rizzo, M. (2025). Obesity: recent advances and future perspectives. Biomedicine 13:368. doi: 10.3390/biomedicines13020368,

PubMed Abstract | Crossref Full Text | Google Scholar

Juge, N. (2023). Microbe profile: Ruminococcus gnavus: the yin and yang of human gut symbionts. Microbiology (Reading) 169:001383. doi: 10.1099/mic.0.001383,

PubMed Abstract | Crossref Full Text | Google Scholar

Kahleova, H., Rembert, E., Alwarith, J., Yonas, W. N., Tura, A., Holubkov, R., et al. (2020). Effects of a low-fat vegan diet on gut microbiota in overweight individuals and relationships with body weight, body composition, and insulin sensitivity. A randomized clinical trial. Nutrients 12:2917. doi: 10.3390/nu12102917,

PubMed Abstract | Crossref Full Text | Google Scholar

Leitzmann, M. F., Stein, M. J., Baurecht, H., and Freisling, H. (2025). Excess adiposity and cancer: evaluating a preclinical-clinical obesity framework for risk stratification. EClinicalMedicine 83:103247. doi: 10.1016/j.eclinm.2025.103247,

PubMed Abstract | Crossref Full Text | Google Scholar

Li, X., Wang, Y., Yu, C., Yao, Y., Chen, X., Deng, Z. Y., et al. (2022). The signatures of liver metabolomics and gut microbiota in high-fat diet fed mice supplemented with rhododendrol. Food Funct. 13, 13052–13063. doi: 10.1039/d2fo01214f,

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, R., Hong, J., Xu, X., Feng, Q., Zhang, D., Gu, Y., et al. (2017). Gut microbiome and serum metabolome alterations in obesity and after weight-loss intervention. Nat. Med. 23, 859–868. doi: 10.1038/nm.4358,

PubMed Abstract | Crossref Full Text | Google Scholar

Ma, L., Ni, Y., Wang, Z., Tu, W., Ni, L., Zhuge, F., et al. (2020). Spermidine improves gut barrier integrity and gut microbiota function in diet-induced obese mice. Gut Microbes 12, 1832857–1832819. doi: 10.1080/19490976.2020.1832857,

PubMed Abstract | Crossref Full Text | Google Scholar

Magne, F., Gotteland, M., Gauthier, L., Zazueta, A., Pesoa, S., Navarrete, P., et al. (2020). The Firmicutes/Bacteroidetes ratio: a relevant marker of gut Dysbiosis in obese patients? Nutrients 12:1474. doi: 10.3390/nu12051474,

PubMed Abstract | Crossref Full Text | Google Scholar

Maioli, T. U., Borras-Nogues, E., Torres, L., Barbosa, S. C., Martins, V. D., Langella, P., et al. (2021). Possible benefits of Faecalibacterium prausnitzii for obesity-associated gut disorders. Front. Pharmacol. 12:740636. doi: 10.3389/fphar.2021.740636,

PubMed Abstract | Crossref Full Text | Google Scholar

Mei, H., Yang, S., Peng, A., Li, R., Xiang, F., Zheng, H., et al. (2022). Development of the gut microbiota in healthy twins during the first 2 years of life and associations with body mass index z-score: results from the Wuhan twin birth cohort study. Front. Microbiol. 13:891679. doi: 10.3389/fmicb.2022.891679,

PubMed Abstract | Crossref Full Text | Google Scholar

Million, M., Angelakis, E., Paul, M., Armougom, F., Leibovici, L., and Raoult, D. (2012). Comparative meta-analysis of the effect of Lactobacillus species on weight gain in humans and animals. Microb. Pathog. 53, 100–108. doi: 10.1016/j.micpath.2012.05.007,

PubMed Abstract | Crossref Full Text | Google Scholar

Mo, R., Zhang, Z., Zhou, Y., Wang, Y., Yin, P., Zhang, C., et al. (2025). A new prognostic model based on serum apolipoprotein AI in patients with HBV-ACLF and acutely decompensated liver cirrhosis. Lipids Health Dis. 24:35. doi: 10.1186/s12944-025-02434-8,

PubMed Abstract | Crossref Full Text | Google Scholar

Mukhopadhya, I., and Louis, P. (2025). Gut microbiota-derived short-chain fatty acids and their role in human health and disease. Nat. Rev. Microbiol. 23, 635–651. doi: 10.1038/s41579-025-01183-w,

PubMed Abstract | Crossref Full Text | Google Scholar

NCD Risk Factor Collaboration (NCD-RisC) (2024). Worldwide trends in underweight and obesity from 1990 to 2022: a pooled analysis of 3663 population-representative studies with 222 million children, adolescents, and adults. Lancet 403, 1027–1050. doi: 10.1016/s0140-6736(23)02750-2,

PubMed Abstract | Crossref Full Text | Google Scholar

Nicholson, J. K., Holmes, E., Kinross, J., Burcelin, R., Gibson, G., Jia, W., et al. (2012). Host-gut microbiota metabolic interactions. Science 336, 1262–1267. doi: 10.1126/science.1223813,

PubMed Abstract | Crossref Full Text | Google Scholar

Song, B., Zhao, K., Zhou, S., Xue, Y., Lu, H., Jia, X., et al. (2023). Association of the gut microbiome with fecal short-chain fatty acids, lipopolysaccharides, and obesity in young Chinese college students. Front. Nutr. 10:1057759. doi: 10.3389/fnut.2023.1057759,

PubMed Abstract | Crossref Full Text | Google Scholar

Sun, J., Germain, A., Kaglan, G., Servant, F., Lelouvier, B., Federici, M., et al. (2023). The visceral adipose tissue bacterial microbiota provides a signature of obesity based on inferred metagenomic functions. Int. J. Obes. 47, 1008–1022. doi: 10.1038/s41366-023-01341-1,

PubMed Abstract | Crossref Full Text | Google Scholar

Sun, L., Ma, L., Ma, Y., Zhang, F., Zhao, C., and Nie, Y. (2018). Insights into the role of gut microbiota in obesity: pathogenesis, mechanisms, and therapeutic perspectives. Protein Cell 9, 397–403. doi: 10.1007/s13238-018-0546-3,

PubMed Abstract | Crossref Full Text | Google Scholar

Sweatt, K., Garvey, W. T., and Martins, C. (2024). Strengths and limitations of BMI in the diagnosis of obesity: what is the path forward? Curr. Obes. Rep. 13, 584–595. doi: 10.1007/s13679-024-00580-1,

PubMed Abstract | Crossref Full Text | Google Scholar

Ticho, A. L., Malhotra, P., Dudeja, P. K., Gill, R. K., and Alrefai, W. A. (2019). Intestinal absorption of bile acids in health and disease. Compr. Physiol. 10, 21–56. doi: 10.1002/cphy.c190007,

PubMed Abstract | Crossref Full Text | Google Scholar

Uema, T., Millman, J. F., Okamoto, S., Nakamura, T., Yamashiro, K., Uehara, M., et al. (2022). Profile of gut microbiota and serum metabolites associated with metabolic syndrome in a remote island most afflicted by obesity in Japan. Sci. Rep. 12:17292. doi: 10.1038/s41598-022-21708-0,

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, Z., Usyk, M., Sollecito, C. C., Qiu, Y., Williams-Nguyen, J., Hua, S., et al. (2020). Altered gut microbiota and host metabolite profiles in women with human immunodeficiency virus. Clin. Infect. Dis. 71, 2345–2353. doi: 10.1093/cid/ciz1117,

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, W., Wang, F., Li, Y., Shi, Y., Wang, X., Chen, X., et al. (2025). Distinct gut microbiota profiles in Normal weight obesity and their association with Cardiometabolic diseases: results from two independent cohort studies. J. Cachexia. Sarcopenia Muscle 16:e13644. doi: 10.1002/jcsm.13644,

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, X., You, J., Tang, J., Li, X., Wang, R., Li, Y., et al. (2024). Interaction between non-alcoholic fatty liver disease and obesity on the risk of developing cardiovascular diseases. Sci. Rep. 14:24024. doi: 10.1038/s41598-024-74841-3,

PubMed Abstract | Crossref Full Text | Google Scholar

World Obesity Federation (2025). World Obesity Atlas 2025. Available online at: https://www.worldobesity.org/resources/resource-library/world-obesity-atlas-2025 (Accessed June 11, 2025).

Google Scholar

Wu, G., Xu, T., Zhao, N., Lam, Y. Y., Ding, X., Wei, D., et al. (2024a). A core microbiome signature as an indicator of health. Cell 187, 6550–6565.e11. doi: 10.1016/j.cell.2024.09.019,

PubMed Abstract | Crossref Full Text | Google Scholar

Wu, C., Yang, F., Zhong, H., Hong, J., Lin, H., Zong, M., et al. (2024b). Obesity-enriched gut microbe degrades myo-inositol and promotes lipid absorption. Cell Host Microbe 32, 1301–1314.e9. doi: 10.1016/j.chom.2024.06.012,

PubMed Abstract | Crossref Full Text | Google Scholar

Xing, R., Du, P., Wang, Z., Fan, Z., Wang, L., Huang, Y., et al. (2025). Porcine bile acids improve antioxidant status and immune function by increasing Hungatella abundance with different protein level diets in late-laying hens. Animals 15:400. doi: 10.3390/ani15040500,

PubMed Abstract | Crossref Full Text | Google Scholar

Yan, Q., Gu, Y., Li, X., Yang, W., Jia, L., Chen, C., et al. (2017). Alterations of the gut microbiome in hypertension. Front. Cell. Infect. Microbiol. 7:381. doi: 10.3389/fcimb.2017.00381,

PubMed Abstract | Crossref Full Text | Google Scholar

Yang, K., Niu, J., Zuo, T., Sun, Y., Xu, Z., Tang, W., et al. (2021). Alterations in the gut Virome in obesity and type 2 diabetes mellitus. Gastroenterology 161, 1257–1269.e13. doi: 10.1053/j.gastro.2021.06.056,

PubMed Abstract | Crossref Full Text | Google Scholar

Yang, R., Shan, S., Shi, J., Li, H., An, N., Li, S., et al. (2023). Coprococcus eutactus, a potent probiotic, alleviates colitis via acetate-mediated IgA response and microbiota restoration. J. Agric. Food Chem. 71:3273–3284. doi: 10.1021/acs.jafc.2c06697,

PubMed Abstract | Crossref Full Text | Google Scholar

Yarahmadi, A., Afkhami, H., Javadi, A., and Kashfi, M. (2024). Understanding the complex function of gut microbiota: its impact on the pathogenesis of obesity and beyond: a comprehensive review. Diabetol. Metab. Syndr. 16:308. doi: 10.1186/s13098-024-01561-z,

PubMed Abstract | Crossref Full Text | Google Scholar

Yu, J., Huang, J., Xia, T., Li, M., Li, R., Liu, T., et al. (2024). Molecular mechanisms of hepatic lipid metabolism disorders: focus on mitochondrial quality control. Portal Hypertens. Cirrhosis 3, 217–233. doi: 10.1002/poh2.95

Crossref Full Text | Google Scholar

Zhang, S., Zhang, Y., Li, J., Wang, X., Zhang, M., Du, M., et al. (2024). Butyrate and propionate are negatively correlated with obesity and glucose levels in patients with type 2 diabetes and obesity. Diabetes Metab. Syndr. Obes. 17, 1533–1541. doi: 10.2147/dmso.S434499,

PubMed Abstract | Crossref Full Text | Google Scholar