AbstractObjectivesThe epidemiology of Enterococcus resistant to priority antibiotics including linezolid has mainly been investigated in developed countries and especially in hospitals. We aimed to evaluate the contribution of different non-human reservoirs for the burden of MDR enterococci in Tunisia, where scarce data are available.MethodsSamples (n = 287) were collected from urban wastewater (n = 57), retail meat (n = 29; poultry/bovine/ovine), milk (n = 89; bovine/ovine), farm animal faeces (n = 80; poultry/bovine/ovine) and pets (n = 32; rabbit/dogs/cats/birds) in different Tunisian regions (2014–17). They were plated onto Slanetz–Bartley agar after pre-enrichment without antibiotics. Standard methods were used for bacterial identification and characterization of antibiotic resistance and virulence genes (PCR), antibiotic susceptibility testing (disc diffusion/broth microdilution; EUCAST/CLSI) and clonality (SmaI-PFGE/MLST).ResultsAll samples carried Enterococcus (n = 377 isolates) resistant to antibiotics considered to be critical or highly important by WHO. Even without antibiotic selection, 38% of Enterococcus faecalis (Efs) and 22% of Enterococcus faecium (Efm) were identified as MDR. Linezolid-resistant isolates (5%; MIC = 8 mg/L) comprised six poxtA-carrying Efm (cow milk), seven optrA-carrying Efs (chicken faeces/meat) and five Efm lacking cfr/optrA/poxtA (poultry/bovine/ovine/wastewater). Clinically relevant Efm clones (clade A1) were identified in animal/meat sources. Ampicillin resistance (1%) was confined to ST18/ST78-like MDR Efm clones from bovine meat/milk samples carrying relevant virulence markers (e.g. ptsD/IS16).ConclusionsThis study provides evidence of the contribution of livestock and foodstuffs to the dispersal of acquired linezolid resistance genes including poxtA and optrA. We report the first poxtA-carrying Efm in Tunisia, and for the first time in bovine samples, stressing the urgent need for alternative measures to counteract the spread of linezolid-resistant enterococci globally.