FileName Folder Description AgingDeclineHostMicrobiomeInteractions.pdf . Full research article "Metabolic modeling reveals the aging-associated decline of host-microbiome metabolic interactions in mice" SupplFigures.pdf . Supplementary figures related to main Figures 1-6 supplTables1_2024-03.ods . Supplementary data tables related to main Figure 1 in manuscript supplTables2_2024-02.ods . Supplementary data tables related to main Figure 2 in manuscript supplTables3_2024-03.ods . Supplementary data tables related to main Figure 3 in manuscript supplTables4_2024-02.ods . Supplementary data tables related to main Figure 4 in manuscript supplTables5_2024-03.ods . Supplementary data tables related to main Figure 5 in manuscript supplTables6_2024-03.ods . Supplementary data tables related to main Figure 6 in manuscript obj_metamouse-2023-05-10.rds databases gapseq microbiome metabolic models for use with R df_rxn2subsys20230510MetaMouse.rds databases Reactions mapped to subsystems/pathways df_rxnAnnotation.rds databases RXN-IDs to Reaction Names gapseqPathwayIDsToNames.csv databases gapseq and MetaCyc Pathway IDs to names df_mouseGOannotaion_2023-05-16.rds databases mouse gene symbols relation to GeneOntology pathways and terms GRCm38.102.EnsemblToGeneSymbol.tsv.gz databases Mouse Ensembl gene ids converted to gene symbols for reference genome version GRCm38.102 metaboliteAnnotation.csv databases translation table from model metabolite/reaction ID to metabolite name df_transcriptMetaData.rds mouseTranscriptome Mouse metadata for RNA-Seq. data mtx_CountsColonAllAges.rds mouseTranscriptome RNA-Seq. gene expression read counts for Colon mtx_CountsLiverAllAges.rds mouseTranscriptome RNA-Seq. gene expression read counts for Liver mtx_CountsBrainAllAges.rds mouseTranscriptome RNA-Seq. gene expression read counts for Brain simple_phylogeny_fromGTDBTKmsa_2023-05-09.tree mouseMetagenome Phylogenetic tree of 181 mouse gut bacteria MAGs from multiple sequence alignment by GTDB-Tk df_MAGabundances.rds mouseMetagenome Read depth coverage of each MAG per Sample df_metadataMAG.rds mouseMetagenome Mouse metadata for MAG data FVAactiveReactions.rds mouseMetagenome Predicted active reactions for each microbial metabolic model based on FVA df_rxnAbundancesMM.rds mouseMetagenome FVA predicted active reaction abundances of mouse microbiome metabolic models df_MMRxnsByAge.rds mouseMetagenome FVA predicted active reaction abundances of mouse microbiome metabolic models, analyzed with linear models by Age microbiotaModelsSBML.zip mouseMetagenome Metabolic models reconstructed from MAGs of the mouse microbiota df_colonRNAnormFiltered.rds combineDataLayers Variance stabilized and near zero variance filtered gene expression data for Colon df_liverRNAnormFiltered.rds combineDataLayers Variance stabilized and near zero variance filtered gene expression data for Liver df_brainRNAnormFiltered.rds combineDataLayers Variance stabilized and near zero variance filtered gene expression data for Brain df_colonMetaData.rds combineDataLayers Mouse metadata for partial correlations of microbiome reactions with colon gene expression data df_liverMetaData.rds combineDataLayers Mouse metadata for partial correlations of microbiome reactions with liver gene expression data df_brainMetaData.rds combineDataLayers Mouse metadata for partial correlations of microbiome reactions with brain gene expression data df_MMrxnFiltered.rds combineDataLayers Abundance of microbiome active reactions per mouse, near-zero variance filtered df_pcorRxn2ColonRNATop.rds combineDataLayers Significance filtered top correlation pairs of microbiome reactions to colon gene expression df_pcorRxn2LiverRNATop.rds combineDataLayers Significance filtered top correlation pairs of microbiome reactions to liver gene expression df_pcorRxn2BrainRNATop.rds combineDataLayers Significance filtered top correlation pairs of microbiome reactions to brain gene expression df_pcorRxn2ColonRNA.rds . Full table of all correlation pairs of microbiome reactions to colon gene expression df_pcorRxn2LiverRNA.rds . Full table of all correlation pairs of microbiome reactions to liver gene expression df_pcorRxn2BrainRNA.rds . Full table of all correlation pairs of microbiome reactions to brain gene expression metamodel_analysis.zip metaModel Source code and data files for running the metamodel and related analysis

Summary Aging is the predominant cause of morbidity and mortality in industrialized countries, yet the molecular mechanisms driving aging and especially the contribution by the microbiome remain unclear. We combined multi-omics with metabolic modeling to comprehensively characterize host–microbiome interactions during aging in mice. Our findings reveal a complex dependency of host metabolism on known and novel microbial interactions. We observed a pronounced reduction in metabolic activity within the aging microbiome accompanied by reduced beneficial interactions between bacterial species. These microbial changes coincided with increased inflammaging as well as a corresponding downregulation of key host pathways, predicted by our model to be microbiome-dependent, that are crucial for maintaining intestinal barrier function, cellular replication, and homeostasis. Our results elucidate microbiome–host interactions that potentially influence host aging processes, focusing on microbial nucleotide metabolism as a pivotal factor in aging dynamics. These pathways could serve as future targets for the development of microbiome-based anti-aging therapies.

Data and code availability Metagenomic raw read and MAG assembly data was deposited in the European Nucleotide Archive (ENA) under BioProject PRJEB73981 (ebi.ac.uk/ena/browser/view/PRJEB73981). Individual accession numbers for each MAG were listed in Supplementary Table S1.2. Gene expression data was published in the GEO database under record GSE262290 (ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE262290) and record GSE278548 (ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE278548). Metabolomics data has been made available at the MassIVE database (massive.ucsd.edu) with identifiers MSV000094409, MSV000094410 and MSV000094436. The metamodel can be found under accession MODEL2310020001 in the BioModels database (ebi.ac.uk/biomodels/MODEL2310020001) 97. Detailed sample metadata, the microbial metabolic models and supplementary resources as well as source code used for data analysis (github.com/sciwitch/MouseMicrobiomeAging) were deposited in a Zenodo record (doi.org/10.5281/zenodo.10844503).