Abstract
Obesity and smoking are major global health threats with synergistic impacts on metabolic disease. The combined effects of these factors on the oral-gut (Orointestinal) microbiome axis, a crucial interface for systemic immunity and metabolism, remain poorly characterized. In this cross-sectional study, 108 age- and sex-matched participants were stratified into four groups (n = 27/group): Non-Obese Non-Smokers (Control), Obese Non-Smokers, Non-Obese Smokers, and Obese-Smokers. Orointestinal microbiomes were profiled via 16 S rRNA sequencing of stool and oral rinse samples, followed by comprehensive bioinformatics analysis. Compartment-specific patterns existed in relation to smoking and obesity. The oral microbiome of Obese-Smokers exhibited a significant, dose-dependent reduction in diversity (34% loss at > 20 cigarettes/day; padj = 0.008), synergistic depletion of oral commensals such as Elizabethkingia (log(2)FC = − 6.33, padj = 0.0032) and enrichment of pathobionts such as Escherichia_Shigella (log(2)FC = 3.2, padj = 0.00068). In contrast, the gut microbiome maintained stable bacterial diversity across groups, despite profound compositional shifts (PERMANOVA, padj = 0.003), with smokers showing a significantly elevated Firmicutes/Bacteroidetes ratio (1.78 vs. 0.91 in controls; padj = 0.006). Critically, Obese-Smokers revealed a significant bidirectional pattern of the Orointestinal axis, specifically in the oral cavity, as evidenced by the significant Orointestinal co-occurrence of the oral pathobionts Neisseria and Leptotrichia (r = 0.67, p = 0.0001), which potentially highlights microbial covariation across habitats. A random forest model identified the oral taxon F0058 (Saccharimonadaceae) as a potential biomarker for Obese-Smoker status (AUC = 0.975, padj = 0.017). Obesity and smoking are associated with synergistic orointestinal dysbiosis, pathobiont expansion, commensal loss, and potential microbial covariation across habitats. The vulnerable oral microbiome, specifically the F0058 biomarker, offers a promising target for noninvasive risk stratification and intervention. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13568-026-02048-y.