KEGG-Based Functional Signatures Complement Taxonomic Profiles Associated with Spontaneous Decolonisation of...

Understanding the functional potential of the gut microbiota for carbapenem-resistant Enterobacterales (CRE) decolonisation is essential for developing novel non-antibiotic strategies to promote their clearance. In a previous study, we identified distinct taxonomic signatures associated with spontaneous CRE decolonisation (DeCol). Here, we aimed to determine whether these taxonomic differences were accompanied by differences in the predicted functional potential of the gut microbiota. Patients were identified from a database of individuals colonised with CRE. We performed Illumina shotgun metagenomic sequencing on 14 persistent CRE carriage (Col) and 23 DeCol patients with OXA-48-producing isolates. Bioinformatic analysis was performed using SqueezeMeta and differential abundance of functional and metabolic genes was assessed using DESeq2. Several antimicrobial resistance genes, including blaOXA-48, were underrepresented in DeCol patients. In contrast, DeCol patients showed an overrepresentation of genes associated with motility, regulated adhesion, short-chain fatty acid (SCFA)-related pathways and alternative carbohydrate metabolism. These orthologue enrichment patterns are consistent with functions previously linked to intestinal homeostasis in the literature. Conversely, Col patients exhibited an overrepresentation of genes associated with redox defence, biofilm formation and amino acid metabolism, suggesting distinct predicted functional profiles between persistent carriage and spontaneous decolonisation. Spontaneous CRE decolonisation was associated with distinct KEGG-based functional signatures and a lower abundance of antimicrobial resistance determinants. These functional profiles were consistent with the taxonomic differences previously identified in the same cohort and generate hypotheses regarding microbiome functions that may contribute to colonisation clearance. Because these findings are based on gene-content analysis, they reflect predicted functional potential rather than direct evidence of metabolic activity. Further multi-omics and experimental studies are required to validate these observations.