Core rumen microbes are functional generalists that sustain host metabolism and gut ecosystem function

nature ecology & evolution Article https://doi.org/10.1038/s41559-025-02904-3 Core rumen microbes are functional generalists that sustain host metabolism and gut ecosystem function Received: 5 November 2024 Accepted: 17 October 2025 Published online: xx xx xxxx Check for updates Omar E. Tovar-Herrera 1,2,3, Ido Grinshpan1,2,3, Gil Sorek Liron Levin4, Sarah Moraïs 1,2,3 & Itzhak Mizrahi 1,2,3 , Ido Lybovits 1,2,3 , 1,2,3 Some microbes persist across diverse gut microbiomes, raising the question of what features define these core taxa and allow them to persist across hosts. Using the rumen microbiome as a model system, we show that core microbes exhibit distinct attributes of ecological generalists, including greater strain variability and broader functional capacity, linked to larger genome sizes. By analysing ~3,000 genomes of core and non-core microbes and metabolically measuring their functional attributes with both biochemical assays and untargeted/targeted metabolomics, we find that these traits enable core microbes to be metabolically independent while also supporting non-core microbes and the host. Core taxa produce essential metabolites, such as amino acids and vitamins, and encode fibre-degrading enzymes crucial for host nutrition. Additionally, they engage in cross-feeding, providing non-core microbes with vital nutrients. This independence positions core microbes as foundational pillars of gut ecosystem stability, and influencing these microbes could modulate microbiome functionality and ruminant host metabolism, with possible downstream consequences for food security and environmental sustainability. Microbial communities, including those in the gut of animals, are exposed to constant environmental fluctuations that directly affect their assembly and dynamics. Despite these continual changes, certain community members are consistently found across several hosts or habitats. Such microbes can be described under the generalist–specialist framework, where those with high occupancy are defined as habitat-generalists1. These persistent and widespread microbes associated with specific gut ecosystems such as the human gut or the rumen environment, are described as members of the core microbiome2. In recent years, substantial efforts have been made to understand the contributions of core microbes to host fitness, health and disease3–7. However, the underlying factors contributing to their persistence across several hosts and habitats remain poorly understood. Positive interactions and lower intraspecific competition have been pointed out as features of microbial core gut species in fish7, while a cohesive phylogenetic structure and higher heritability index were properties of some core members in cows8. In this study, we aim to explore the connection between habitatgeneralist microbes, defined as the host-associated core microbiome, and their functional attributes, by systematically integrating ecological and evolutionary perspectives to generalize the roles of core taxa. We chose the rumen microbiome ecosystem for this purpose, as the intricate symbiotic relationship between ruminants and their microbiome has become a model system for studying the host–microbiome axis9. The rumen ecosystem provides the bovine host with up to 80% of its energy requirements, as well as essential metabolites such as amino acids and vitamins, through the activity of its resident microbial community10,11. This remarkable process involves diverse microbial 1 National Institute of Biotechnology in the Negev, Ben-Gurion University of the Negev, Beer-Sheva, Israel. 2Department of Life Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel. 3The Goldman Sonnenfeldt School of Sustainability and Climate Change, Ben-Gurion University of the Negev, Beer-Sheva, Israel. 4Bioinformatics Core Facility, llse Katz Institute for Nanoscale Science and Technology, Ben-Gurion University of the Negev, Beer-Sheva, Israel. e-mail: Nature Ecology & Evolution Article (i) (ii) Number of genera a https://doi.org/10.1038/s41559-025-02904-3 8 Ref. 16 Ref. Succiniclasticum Saccharofermentans* Selenomonas Ruminococcus Butyrivibrio Pseudobutyrivibrio Anaeroplasma Prevotella Succinivibrio Methanobrevibacter Fibrobacter Treponema Unclassified genus 1 Lachnospiraceae NK3A20 group >80% animals 200 Non-core (95.08%) 426 genera 100 Core (4.92%) 14 genera 0 0 Ref. (iii) 300 0.25 0.50 0.75 1.00 Prevalence 17 8 Ref. , n = 1,016 16 Ref. , n = 742 17 Ref. , n = 16 b (i) (iv) Number of genomes Dataset Ref. 18 4,941 Ref. 19 10,373 Ref. 20 410 Total genomes = 15,726 14 (ii) >90% completeness <10% contamination (dRep and GTDBTK) Genomes without taxonomic annotation were removed Total dereplicated genomes = 4,655 (iii) 2,553 non-core 1,260 core Metabolic pathway analysis (Anvi'o) CAZyme profiling (dbCAN2) Total annotated genomes = 3,813 Non-core Core genera Prevotella Ruminococcus Saccharofermentans Lachnospiraceae NK3A20 group Pseudobutyrivibrio Selenomonas Succiniclasticum Anaeroplasma Treponema Fibrobacter Succinivibrio Butyrivibrio Fig. 1 | A global core rumen microbiome is represented by 14 microbial genera. a, Identification of core rumen microbiome genera. (i) Worldwide locations of ruminant samples used in our analysis. Countries from which samples were collected in refs. 8,16,17 are depicted in shades of blue on the global map. (ii) Global core rumen genera are also identified in a 1,016-cow European study. The prevalence of genera across the 1,016-cow cohort of ref. 8 is shown: grey depicts genera with less than 80% prevalence (non-core) across the cohort while green indicates more than 80% (core). (iii) Correspondence of the 14 core genera found in at least two of the studies from refs. 8,16,17 cohorts. The circles in shades of blue indicate which microbes were identified as core in the study listed at the bottom of each column. The microbial names on the right represent the 14 core genera found in ref. 8. Although Saccharofermentans was only detected as a core microbe in ref. 8, we included it in our analysis because of the large number of available representative genomes for this genus. The number below the name of the dataset indicates the number of animals included in the study. b, Genome database of core and non-core rumen microbes. (i) Number and source of MAGs and genomes collected. Genomes derived from ref. 20 correspond to isolates from the Hungate Collection. (ii) Taxonomic annotation and quality control. Preprocessing of the database of the genome, including completeness, contamination, dereplication and taxonomic annotation using dRep 2.0 and GTDBTK, resulting in 4,655 curated MAGs and genomes. (iii) Functional annotation of dereplicated genomes. Metabolic pathways and glycosyl hydrolases in each genome were annotated using anvi’o and dBCan 2.0. Genomes lacking taxonomic annotation or with fewer than five representatives per gen (...truncated)