The intricate ecosystem of the human gut microbiome plays a pivotal role in health and disease, with its functional capacity largely shaped by carbohydrate-active enzymes (CAZymes) that break down complex carbohydrates from diet and host sources. Recent advances have enabled scientists to dive deeper into how these enzymatic repertoires differ across global human populations, revealing metabolic signatures tied closely to dietary habits and lifestyle. A groundbreaking study conducted by Ducarmon et al., published in Nature Microbiology, introduces a large-scale analytical framework named Cayman that systematically profiles gut microbial CAZyme compositions from metagenomic data, unveiling striking differences between hunter-gatherer and industrialized societies.
In this study, the authors harnessed metagenomic datasets from 3,960 individuals spanning two distinct cohorts: hunter-gatherer and subsistence-based traditional societies (HIS; n=3,166) and lower-middle-income societies (LMIS; n=794). By deploying principal coordinate analysis on CAZyme profiles derived from these populations, the research team observed a statistically robust separation between these groups. This finding, corroborated by permutational multivariate analysis of variance (PERMANOVA), highlights distinct functional capacities aligned with differing environmental and dietary exposures.
Contrary to prior smaller studies suggesting greater CAZyme diversity in LMIS populations, often linked to their fiber-rich diets, the comprehensive analysis spearheaded by Ducarmon and colleagues uncovered a surprisingly greater number of unique CAZymes in HIS populations. This increased richness persisted even when focusing exclusively on enzymes targeting dietary fibers, challenging earlier assumptions and suggesting a more nuanced relationship between diet, lifestyle, and microbiome functionality.
The team addressed potential confounders such as differences in read-mapping rates and sequencing depth, confirming that these technical factors did not explain the observed disparities in CAZyme diversity. This bolsters the conclusion that HIS microbiomes fundamentally harbor a broader enzymatic repertoire, which may reflect adaptations to diverse polysaccharide substrates encountered in their environments and diets.
Expanding on substrate specificity, the researchers leveraged a detailed annotation scheme to disentangle the functional landscape of microbiome CAZyme repertoires between HIS and LMIS individuals. HIS microbiomes showed significantly higher ratios of mucin-degrading to dietary fiber-degrading enzymes, as well as elevated glycosaminoglycan (GAG)-targeting CAZymes relative to fibers. This pattern aligns with dietary tendencies among hunter-gatherers, who consume more animal-derived glycans and have lower fiber intake compared to populations with agriculturally based diets rich in plant fibers. These metabolic signatures offer a compelling glimpse into how evolutionary and ecological pressures sculpt microbial capabilities in different human niches.
Furthermore, substrate enrichment analyses revealed that HIS microbiomes are not only biased towards host- and animal-derived glycans metabolism but also enriched for enzymes targeting non-starch polysaccharides such as pectins and gums. Given that these polysaccharides function as food additives commonly found in processed foods, this observation potentially marks an intriguing intersection of traditional diets with the influence of modern food consumption patterns, suggesting microbial adaptation to anthropogenic dietary components.
Delving deeper into the taxonomic underpinnings of these functional differences, the authors applied linear modeling to identify CAZyme families significantly enriched either in HIS or LMIS cohorts. They discovered notable enrichment of four resistant starch-targeting GH13 subfamilies in LMIS metagenomes, a finding consistent with higher consumption of whole grains and legumes. Taxonomic association analyses revealed that these LMIS-enriched CAZymes predominantly derived from Proteobacteria and Spirochaetes—commonly found in non-industrialized populations but reduced or lost in industrialized gut microbiomes, emphasizing the dynamic nature of microbial community composition.
Conversely, CAZymes enriched in HIS populations, including the highly prevalent CBM58 domain associated with the SusG starch utilization protein, predominantly mapped to Bacteroides. This genus’s enrichment in HIS metagenomes underscores its central ecological role and expanded enzymatic versatility in these communities. However, the team made it clear that Bacteroides alone does not fully explain the observed differences in CAZyme richness.
To understand the contributions of individual genera, the study quantified how many HIS-enriched CAZyme families each genus encoded, revealing that while Bacteroides harbors the largest repertoire of such families, genera like Eubacterium and Clostridium were even stronger predictors of overall community CAZyme richness in linear models. This insight suggests that a consortium of key bacterial players rather than a single genus shapes the functional enzyme landscape across human populations, reflecting complex microbiome interdependencies and functional redundancies.
Building upon taxonomic-function links, the authors constructed multivariable regression models connecting species abundances within genera to abundances of individual CAZyme families. These analyses confirmed strong, predictable relationships between microbial taxa and the metabolic enzymes they encode. For instance, Bifidobacterium species showed exceptionally high predictive power for CAZyme families GH13_30 and GH13_3, the latter lacking prior experimental validation but showing congruent abundance patterns, pointing to Bifidobacterium as the primary contributor in both HIS and LMIS individuals.
Intriguingly, certain associations differed markedly between HIS and LMIS individuals despite similar taxonomic abundances. The genus Collinsella, for example, predicted GH13_30 CAZyme abundance robustly in LMIS but not in HIS groups, indicating that other taxa fulfill this enzymatic function in hunter-gatherer microbiomes. Similarly, the α-fucosidase family GH95 displayed opposite genus-level associations: it correlated strongly with Bacteroides in HIS yet with Prevotella in LMIS individuals, suggesting functional replacement of enzyme carriers as gut microbial ecology shifts with lifestyle.
Notably, Prevotella has not been previously recognized as a mucin-degrading genus or known to encode GH95 enzymes experimentally, emphasizing the power of combining large-scale metagenomics with bioinformatics models to infer novel microbial functional roles. This phenomenon of taxa assuming different enzymatic functions in distinct populations highlights the adaptive plasticity of the gut microbiome.
Another compelling case is the glycosyltransferase family GT31, which associates dominantly with Akkermansia in HIS microbiomes but not in LMIS samples. Instead, in LMIS individuals, the single-celled parasite Blastocystis shows stronger linkage to GT31 abundance. Such findings underscore complex inter-species interactions and the role of non-bacterial microorganisms in shaping gut enzymatic profiles that vary by population context.
Together, these findings reflect a dynamic gut ecosystem where diet, environment, and microbial community structure collectively shape functional enzymatic landscapes. The Cayman framework offers a powerful toolset for unraveling this complexity, facilitating large-scale comparative studies that bridge the gap between taxonomy and function with unprecedented resolution.
This study not only clarifies how global human populations differ in their microbial carbohydrate metabolism but also illuminates the link between lifestyle transitions and gut microbial functionality. Such knowledge is critical for understanding the long-term health implications of dietary modernization and for devising microbiome-targeted therapies tailored to specific ecological and cultural settings.
Moreover, uncovering population-specific microbial enzyme repertoires can guide precision nutrition and the development of microbiota-accessible carbohydrates that balance the metabolic capabilities of distinct gut communities. This highlights the urgent need to preserve microbial diversity and function amid global dietary homogenization and industrialization.
In summary, through the integration of metagenomics, innovative bioinformatics, and ecological modeling, the work by Ducarmon et al. revolutionizes our understanding of human gut microbiome enzymatic ecology. It sets the stage for future efforts to decode the co-evolutionary dance between microbes, diet, and health across the human species.
Subject of Research: Gut microbiome carbohydrate-active enzyme (CAZyme) repertoires in distinct human populations.
Article Title: Cayman enables large-scale analysis of gut microbiome carbohydrate-active enzyme repertoires.
Article References:
Ducarmon, Q.R., Karcher, N., Giri, S. et al. Cayman enables large-scale analysis of gut microbiome carbohydrate-active enzyme repertoires. Nat Microbiol (2026). https://doi.org/10.1038/s41564-026-02318-2
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s41564-026-02318-2
Tags: carbohydrate-active enzymes in human gutCayman analytical framework for microbiomesCAZyme diversity in traditional societiesenvironmental impact on gut enzyme profilesfunctional capacity of gut microbiotaglobal human population microbiome comparisongut microbiome enzyme analysishunter-gatherer vs industrialized gut microbiomeslarge-scale metagenomic CAZyme profilingmetagenomic data in gut microbiome researchmicrobial metabolism and dietary habitssubsistence-based diet microbiome studies
