Damian Rafal Plichta

Interactions between diet and the gut microbiota are fundamental to metabolic health, shaping energy balance and disease susceptibility^1-5. However, the underlying mechanisms by which dietary and microbial factors converge to regulate host physiology remain unclear. Here we show that protein availability profoundly modulates the functional landscape of the gut microbiota and promotes remodelling of white adipose tissue (WAT). Specifically, low-protein diets (LPDs) robustly induce signature genes of browning in WAT to a similar extent to that seen in response to classical stimuli, such as cold exposure or β-adrenergic receptor activation^6-8. LPD-mediated browning was markedly diminished in germ-free mice, and this defect was rescued by colonization with defined bacterial consortia made up of strains that were isolated and down-selected from the faeces of either LPD-fed mice or healthy human volunteers with ¹⁸F-fluorodeoxyglucose positron emission tomography (FDG-PET)-confirmed brown- or beige-fat activity^9-12. Microbiota-induced browning was mediated both by bile acids driving the activation of the farnesoid X receptor (FXR) in adipose progenitor cells, and by nrfA-encoding commensal-derived ammonia driving the expression of fibroblast growth factor 21 (FGF21) in hepatocytes. The bile acid-FXR and ammonia-FGF21 axes both have non-redundant, essential roles in promoting WAT browning. These findings highlight a mechanistic link between diet, gut microbial metabolism and adipose tissue remodelling, uncovering microbiota-dependent pathways by which the host responds to dietary cues.

High levels of circulating cholesterol are associated with human cardiovascular diseases and an altered gut microbiome. Still, major gaps exist in our understanding of the interactions of cholesterol with gut microbes. The reductive transformation of cholesterol to the poorly absorbed sterol coprostanol by human gut bacteria has long been known, but the genetic and biochemical basis for this activity is only partially elucidated. Here, we discover and characterize a gut bacterial enzyme that catalyzes the reduction of cholestenone to coprostanone, the second step in the intestinal sterol metabolism ( ism ) pathway for coprostanol production. We identify a gene encoding a previously unknown 5β-reductase, IsmB, a new member of the Fe–S cluster flavoenzyme superfamily, in the coprostanol producing organism Eubacterium coprostanoligenes . Biochemical characterization of IsmB confirms it is an anaerobic Fe–S cluster flavoenzyme and reveals specificity for reduction of an unanticipated intermediate, 5-cholesten-3-one, to coprostanone, revising the ism pathway. We also identify and characterize homologs of IsmB encoded in uncultured human gut bacteria that also encode the previously identified ism pathway enzyme IsmA, further supporting the role of IsmB in coprostanol formation. Finally, analysis of human stool metagenomics and metabolomics datasets further confirms the relevance of IsmB in the human gut microbiome, and analyses of human serum metabolomics from Framingham Heart Study participants reveal negative correlation between serum cholesterol levels and the presence of IsmA/IsmB encoders in the gut. Together, these results show the utility of combining biochemistry and stool metagenomic analysis for gut microbial enzyme discovery, and suggests IsmA/IsmB-encoding gut bacteria carry potential benefits for cholesterol homeostasis and cardiovascular health.

Background: The functional annotation of uncharacterized microbial enzymes from metagenomic data remains a significant challenge, limiting our understanding of microbial metabolic dynamics. Traditional annotation methods often rely on sequence homology, which can fail to identify remote homologs or enzymes with structural rather than sequence conservation. To address this gap, we developed CAZyLingua, the first annotation tool to use protein language models (pLMs) for the accurate classification of carbohydrate-active enzyme (CAZyme) families and subfamilies. Results: CAZyLingua demonstrated high performance, maintaining precision and recall comparable to state-of-the-art hidden Markov model-based methods while outperforming purely sequence-based approaches. When applied to a metagenomic gene catalog from mother/infant pairs, CAZyLingua identified over 27,000 putative CAZymes missed by other tools, including horizontally-transferred enzymes implicated in infant microbiome development. In datasets from patients with Crohn’s disease and IgG4-related disease, CAZyLinuga uncovered disease-associated CAZymes, highlighting an expansion of carbohydrate esterases (CEs) in IgG4-related disease. A CE17 enzyme predicted to be overabundant in Crohn’s disease was functionally validated, confirming its catalytic activity on acetylated manno-oligosaccharides. Conclusions: CAZyLingua is a powerful tool that effectively augments existing functional annotation pipelines for CAZymes. By leveraging the deep contextual information captured by pLMs, our method can uncover novel CAZyme diversity and reveal enzymatic functions relevant to health and disease, contributing to a further understanding of biological processes related to host health and nutrition.

Human immunodeficiency virus (HIV) infection alters gut microbiota composition and function, but the impact of geography and antiretroviral therapy remains unclear. Here we determined gut microbiome alterations linked to HIV infection and antiretroviral treatment in 327 individuals with HIV and 260 control participants in cohorts from Uganda, Botswana and the USA via faecal metagenomics. We found that while HIV-associated taxonomic differences were mostly site specific, changes in microbial functional pathways were broadly consistent across the cohorts and exacerbated in individuals with acquired immunodeficiency syndrome. Microbiome perturbations associated with antiretroviral medications were also geography dependent. In Botswana and Uganda, use of the non-nucleoside reverse transcriptase inhibitor efavirenz was linked to depletion of Prevotella, disruption of interspecies metabolic networks, exacerbation of systemic inflammation and atherosclerosis. Efavirenz-associated Prevotella depletion may occur through cross-inhibition of prokaryotic reverse transcriptases involved in antiphage defences, as shown by computational and in vitro experiments. These observations could inform future geography-specific and microbiome-guided therapy.

Activation of brown and beige fat biogenesis promotes metabolic health in rodents and humans, but typically requires cold exposure or pharmacological activation of β-adrenergic receptors, which may pose cardiovascular risks. Dietary intervention represents a clinically viable alternative strategy to induce beige cells and thus enhance metabolic health, though the underlying mechanisms remain poorly understood. In this study, we identified specific microbiota members in both mice and humans that promote browning of white adipose tissue (WAT) and ameliorate metabolic disorders in the context of a low-protein diet (LPD). Diets with low protein, regardless of fat and carbohydrate content, induced robust WAT browning to a degree comparable to that achieved by cold exposure or β-adrenergic receptor agonist administration. LPD-mediated browning was markedly diminished in germ-free (GF) mice, highlighting the essential role of the gut microbiota. Microbiota-induced browning occurred independently of the immune system and primarily through mechanisms involving increased circulating deconjugated bile acids, activation of the farnesoid X receptor (FXR) in WAT progenitor cells, and enhanced hepatic production of FGF21. The browning defect in GF mice was rescued by transplanting microbiota from conventional mice or from brown/beige fat-positive healthy human volunteers, as determined by fluorodeoxyglucose positron-emission tomography (FDG-PET). Defined bacterial consortia, comprising strains isolated from mice or FDG-PET-positive human volunteers, were sufficient to elevate plasma bile acids and hepatic FGF21 levels by modulating nitrogen metabolism, ultimately restoring browning in response to a LPD. Our findings highlight the significant impact of diet-microbiota interactions on WAT browning and suggest their therapeutic potential for managing metabolic diseases.

Persistent colonization and outgrowth of potentially pathogenic organisms in the intestine can result from long-term antibiotic use or inflammatory conditions, and may perpetuate dysregulated immunity and tissue damage^1,2. Gram-negative Enterobacteriaceae gut pathobionts are particularly recalcitrant to conventional antibiotic treatment^3,4, although an emerging body of evidence suggests that manipulation of the commensal microbiota may be a practical alternative therapeutic strategy^5–7. Here we isolated and down-selected commensal bacterial consortia from stool samples from healthy humans that could strongly and specifically suppress intestinal Enterobacteriaceae. One of the elaborated consortia, comprising 18 commensal strains, effectively controlled ecological niches by regulating gluconate availability, thereby re-establishing colonization resistance and alleviating Klebsiella- and Escherichia-driven intestinal inflammation in mice. Harnessing these activities in the form of live bacterial therapies may represent a promising solution to combat the growing threat of proinflammatory, antimicrobial-resistant Enterobacteriaceae infection.

Determining the structure and phenotypic context of molecules detected in untargeted metabolomics experiments remains challenging. Here we present reverse metabolomics as a discovery strategy, whereby tandem mass spectrometry spectra acquired from newly synthesized compounds are searched for in public metabolomics datasets to uncover phenotypic associations. To demonstrate the concept, we broadly synthesized and explored multiple classes of metabolites in humans, including N-acyl amides, fatty acid esters of hydroxy fatty acids, bile acid esters and conjugated bile acids. Using repository-scale analysis^1,2, we discovered that some conjugated bile acids are associated with inflammatory bowel disease (IBD). Validation using four distinct human IBD cohorts showed that cholic acids conjugated to Glu, Ile/Leu, Phe, Thr, Trp or Tyr are increased in Crohn’s disease. Several of these compounds and related structures affected pathways associated with IBD, such as interferon-γ production in CD4⁺ T cells³ and agonism of the pregnane X receptor⁴. Culture of bacteria belonging to the Bifidobacterium, Clostridium and Enterococcus genera produced these bile amidates. Because searching repositories with tandem mass spectrometry spectra has only recently become possible, this reverse metabolomics approach can now be used as a general strategy to discover other molecules from human and animal ecosystems.

In metagenomics, the pool of uncharacterized microbial enzymes presents a challenge for functional annotation. Among these, carbohydrate-active enzymes (CAZymes) stand out due to their pivotal roles in various biological processes related to host health and nutrition. Here, we present CAZyLingua, the first tool that harnesses protein language model embeddings to build a deep learning framework that facilitates the annotation of CAZymes in metagenomic datasets. Our benchmarking results showed on average a higher F1 score (reflecting an average of precision and recall) on the annotated genomes of Bacteroides thetaiotaomicron , Eggerthella lenta and Ruminococcus gnavus compared to the traditional sequence homology-based method in dbCAN2. We applied our tool to a paired mother/infant longitudinal dataset and revealed unannotated CAZymes linked to microbial development during infancy. When applied to metagenomic datasets derived from patients affected by fibrosis-prone diseases such as Crohn’s disease and IgG4-related disease, CAZyLingua uncovered CAZymes associated with disease and healthy states. In each of these metagenomic catalogs, CAZyLingua discovered new annotations that were previously overlooked by traditional sequence homology tools. Overall, the deep learning model CAZyLingua can be applied in combination with existing tools to unravel intricate CAZyme evolutionary profiles and patterns, contributing to a more comprehensive understanding of microbial metabolic dynamics.