Global in silico genomic epidemiology of tetracycline resistance in Staphylococcus aureus: one health dynamics and genotype–phenotype associations
摘要
Tetracyclines remain widely used in human and veterinary medicine for the treatment of Staphylococcus aureus, including methicillin-resistant strains, and are key agents at the human–animal–environment interface. However, the global genomic epidemiology of tetracycline resistance mechanisms, their temporal stability, host structuring, and genotype–phenotype relationships across the tetracycline class remain incompletely resolved. We performed a large-scale in silico analysis to characterize tetracycline resistance determinants, minimum inhibitory concentration (MIC) patterns, and One Health dynamics in S. aureus. A total of 110,309 publicly available S. aureus genomes collected between 2000 and 2025 from 128 countries were retrieved from NCBI and subjected to rigorous quality control. Tetracycline resistance determinants including intrinsic efflux pumps (tet, mepA), acquired efflux genes (tet(K), tet(L)), ribosomal protection proteins (tet(M), tet(O), tet(S), tet(W)), rare tet alleles, and the ribosomal S10 mutation rpsJ Y58D were identified using AMRFinderPlus. Host source, geography, and multilocus sequence type (MLST) were integrated. Phenotypic MIC data for tetracycline, oxytetracycline, doxycycline, minocycline, and tigecycline were harmonized using Clinical and Laboratory Standards Institute/ European Committee on Antimicrobial Susceptibility Testing logic. Temporal trends, host enrichment, clonal structure, and gene–MIC associations were assessed using regression models and descriptive genomics. Intrinsic efflux determinants were nearly universal, with tet and mepA detected in > 99.9% of genomes, defining a fixed genomic background. Acquired resistance genes were less frequent and strongly host-structured: tet(K) (12.0%), tet(M) (11.0%), and tet(L) (1.9%) predominated, while other tet family members and rpsJ Y58D remained rare (< 1.1%). Environmental and animal isolates carried significantly higher frequencies of acquired determinants than human isolates (p < 0.001), consistent with One Health reservoirs of resistance. Temporal analyses revealed largely stable prevalence of tetracycline resistance mechanisms over two decades, with only modest declines in human-associated tet(M). MIC distributions showed preserved activity of tigecycline and minocycline (100% susceptible), high susceptibility to tetracycline (97%), and moderate non-susceptibility for doxycycline. Gene–phenotype alignment demonstrated that tet(K) and tet(M) were associated with elevated MIC tails, whereas isolates harboring only intrinsic efflux clustered at lower MICs. Global tetracycline resistance in S. aureus is characterized by a stable intrinsic efflux backbone with superimposed, host-structured acquisition of mobile tet genes that shape agent-specific MIC distributions. The persistence of higher resistance gene burdens in animal and environmental reservoirs underscores the importance of One Health surveillance. Despite widespread gene carriage, tetracyclines particularly tigecycline and minocycline retain strong phenotypic activity, highlighting opportunities for informed stewardship guided by genome-based monitoring.
Graphical Abstract