In recent years, the intricate relationship between human genetics and the gut microbiome has emerged as a focal point in biomedical research. The gut microbiota, a vast and diverse community of microorganisms residing within our digestive tract, profoundly influences health, metabolic functions, and disease susceptibility. Yet, despite the accumulation of microbial genome-wide association studies (GWAS) identifying host genetic variants linked to gut bacterial populations, the precise cellular contexts in which these genetic influences manifest have remained largely enigmatic. A groundbreaking study now bridges this critical knowledge gap by integrating microbial GWAS with single-cell RNA sequencing (scRNA-seq) across multiple human tissues, unveiling a detailed map of host-microbe interactions at unprecedented cellular resolution.
The research pioneers a computational framework termed single-cell Bacteria Polygenic Score, or scBPS, designed to harness the power of existing large-scale microbial GWAS data while leveraging the cellular granularity afforded by scRNA-seq. By analyzing an extensive repertoire of 207 microbial taxa alongside 254 distinct host cell types extracted from 24 human organs—including metabolically and immunologically pivotal tissues such as the liver, pancreas, lung, and intestine—this framework uniquely correlates genetic associations with specific cell populations. This approach transcends traditional bulk analyses, enabling the dissection of subtler, cell type-specific host-microbiome crosstalk that eluded earlier studies.
The validation of scBPS’s biological fidelity comes in part through its recapitulation of well-established physiological interactions. Notably, enrichments inferred by the model confirmed dominant communication channels between gut microbes and the digestive tract, as well as the liver’s epithelial compartment—a critical hub in metabolism and detoxification. This affirmation of scBPS’s predictive power bolsters confidence in the framework’s novel discoveries, especially those involving previously uncharted microbial-host cellular interactions.
Among the most striking findings is the robust association identified between the gut bacterium Collinsella and a specialized subpopulation of liver cells known as central-veinal hepatocytes. These hepatocytes play crucial roles in metabolic processes, particularly lipid and cholesterol metabolism, within the liver’s microarchitecture. By pinpointing this association, the study illuminates a potential mechanistic link through which gut microbiota may modulate crucial host metabolic pathways, with implications for understanding metabolic diseases and cardiovascular risk factors.
Translating computational insight into experimental validation, the researchers carried out in vivo murine experiments to probe the causal impact of Collinsella on hepatic functions. Through oral gavage of Collinsella in mice, followed by single-nuclei RNA sequencing of liver tissue, they uncovered transcriptional modulation of cholesterol metabolism pathways specifically within the central-veinal hepatocyte subpopulation. This experimental confirmation not only substantiates the computational predictions but also highlights the dynamic responsiveness of liver cells to microbial signals emanating from the gut microbiome.
Intriguingly, these findings underscore how distinct microbial taxa can selectively influence targeted cell populations, shifting the paradigm from broad microbiota-host associations to precise cellular dialogue. The ability to map these direct interactions at single-cell resolution provides a powerful lens for interpreting how microbial communities may contribute to human health pathways, metabolic homeostasis, and disease etiologies. This could pave the way for more tailored therapeutic interventions aiming to manipulate microbial ecosystems to beneficially reprogram specific host cell functions.
The robustness and reproducibility of the scBPS framework were further tested by applying it to independent microbial GWAS datasets alongside both single-cell and bulk transcriptome data. This cross-validation effort bolstered the model’s generalizability, confirming that its predictive outputs were not data-set specific but rather reflective of underlying biological truths. Such reproducibility is fundamental for establishing computational tools as reliable platforms for future investigations into host-microbial interplay.
Critically, scBPS’s methodology represents an innovative fusion of two rapidly advancing fields: microbial genomics and single-cell transcriptomics. While microbial GWAS provide associative links between host genotypes and microbiota compositions, scRNA-seq adds unparalleled resolution by delineating gene expression patterns at the level of individual cells within heterogeneous tissues. Integrating these dimensions allows researchers to assign functional significance to host genetic variants by revealing their effects within discrete cellular milieus interacting with gut microbes.
The emphasis on 24 distinct human organs, spanning metabolic, immune, and barrier tissues, also emphasizes the systemic nature of gut microbiome influence. Beyond the digestive tract, organs such as the pancreas and lung were included, shedding light on potential extraintestinal microbiome-host interactions. This broad multi-organ perspective is essential, as microbial metabolites and components circulate systemically, and immune training by gut microbes impacts distant sites, modulating inflammation and organ function.
Furthermore, the study sheds light on the cellular heterogeneity within tissues, particularly the liver, where different hepatocyte subpopulations exhibit spatial zonation and functional specialization. The central-veinal hepatocytes, identified as key responders to Collinsella presence, demonstrate how microenvironmental variation within an organ can define responsiveness to microbial cues. Such spatially resolved analyses will be instrumental in deciphering the layered complexity of host-microbe communications.
Mechanistic insights gleaned from this approach could revolutionize our understanding of metabolic diseases, especially those linked to dysregulation of cholesterol and lipid metabolism such as nonalcoholic fatty liver disease (NAFLD) and atherosclerosis. By elucidating how gut bacteria like Collinsella influence gene expression in liver cells, the study opens avenues for microbiome-targeted therapeutics that may complement or enhance existing pharmaceutical strategies tackling metabolic disorders.
Beyond metabolism, the scBPS framework offers potential for unraveling how host genetics and cell-type specific expression patterns affect immune responses, susceptibility to infections, and inflammatory conditions shaped by the gut microbiome. The precision and scalability of this computational approach make it adaptable for exploring diverse microbiome-related phenotypes in various population cohorts, ultimately aiding personalized medicine efforts.
In conclusion, the integration of microbial GWAS and single-cell transcriptomics through the innovative scBPS framework unveils a new frontier in microbial-host interaction research. By linking microbial taxa to specific host cell populations across multiple tissues, this study establishes a comprehensive map of host-microbe crosstalk that advances both fundamental biology and translational potential. The convergence of computational predictions with experimental validation marks a pivotal step toward understanding how the microbiome shapes human physiology at the cellular level.
As large-scale multi-omic datasets continue to expand, approaches like scBPS will be instrumental in harnessing this data deluge to extract meaningful biological insights. The study’s demonstration of linking genetic predispositions, microbial composition, and cell-type specific gene expression lays a promising foundation for future investigations into microbe-driven modulation of host function, offering hope for novel interventions benefiting human health worldwide.
Subject of Research: Host genetics, gut microbiome, and cell type-specific interactions in human tissues
Article Title: Integrating microbial GWAS and single-cell transcriptomics reveals associations between host cell populations and the gut microbiome
Article References:
Li, J., Ma, Y., Cao, Y. et al. Integrating microbial GWAS and single-cell transcriptomics reveals associations between host cell populations and the gut microbiome. Nat Microbiol (2025). https://doi.org/10.1038/s41564-025-01978-w
Image Credits: AI Generated
