Multi-drug resistant (MDR) non-typhoidal Salmonella (NTS) is a public health concern globally. This study reports the phenotypic and genotypic antimicrobial resistance (AMR) profiles of NTS isolates from bovine lymph nodes (n = 48) and ground beef (n = 29). Furthermore, we compared genotypic AMR data of our isolates with those of publicly available NTS genomes from Mexico (n = 2400). The probability of finding MDR isolates was higher in ground beef than in lymph nodes:χ2 = 12.0, P = 0.0005. The most common resistant phenotypes involved tetracycline (40.3%), carbenicillin (26.0%), amoxicillin-clavulanic acid (20.8%), chloramphenicol (19.5%) and trimethoprim-sulfamethoxazole (16.9%), while more than 55% of the isolates showed decreased susceptibility to ciprofloxacin and 26% were MDR. Conversely, resistance to cephalosporins and carbapenems was infrequent (0–9%). MDR phenotypes were strongly associated with NTS serovar (χ2 = 24.5, P<0.0001), with Typhimurium accounting for 40% of MDR strains. Most of these (9/10), carried Salmonella genomic island 1, which harbors a class-1 integron with multiple AMR genes (aadA2, blaCARB-2, floR, sul1, tetG) that confer a penta-resistant phenotype. MDR phenotypes were also associated with mutations in the ramR gene (χ2 = 17.7, P<0.0001). Among public NTS isolates from Mexico, those from cattle and poultry had the highest proportion of MDR genotypes. Our results suggest that attaining significant improvements in AMR meat safety requires the identification and removal (or treatment) of product harboring MDR NTS, instead of screening for Salmonella spp. or for isolates showing resistance to individual antibiotics. In that sense, massive integration of whole genome sequencing (WGS) technologies in AMR surveillance provides the shortest path to accomplish these goals.

Salmonella enterica (SE) can survive in surface waters (SuWa) and the role of non-host environments in its transmission has acquired increasing relevance. In this study, we conducted comparative genomic analyses of 172 SE isolates collected from SuWa across three months in six states of central Mexico during 2019. SE transmission dynamics were assessed using 87 experimental and 112 public isolates from Mexico collected during 2002-2019. We also studied genetic relatedness between SuWa isolates and human clinical strains collected in North America during 2005-2020. Among experimental isolates, we identified 41 SE serovars and 56 multi-locus sequence types (ST). Predominant serovars were Senftenberg (n=13), Meleagridis, Agona, and Newport (n=12 each), Give (n=10), Anatum (n=8), Adelaide (n=7), and Infantis, Mbandaka, Ohio and Typhimurium (n=6 each). We observed a high genetic diversity in the sample under study, as well as clonal dissemination of strains across distant regions. Some of these strains are epidemiologically important (ST14, ST45, ST118, ST132, ST198, and ST213), and were genotypically close to those involved in clinical cases in North America. Transmission network analysis suggests that SuWa are a relevant source of SE (0.7 source/hub ratio) and contributes to its dissemination as isolates from varied sources and clinical cases have SuWa isolates as common ancestors. Overall, the study shows SuWa act as reservoir of various SE serovars of public health significance. Further research is needed to better understand the mechanisms involved in SuWa contamination by SE, as well as develop interventions to contain its dissemination to food production settings. Study importance Surface waters are heavily used in food production worldwide. Several human pathogens can survive in these waters for long periods and disseminate to food production environments, contaminating our food supply. One of these pathogens is Salmonella enterica , a leading cause of foodborne infections, hospitalizations and deaths in many countries. This research demonstrates the role of surface waters as a vehicle for the transmission of Salmonella along food production chains. It also shows some strains circulating in surface waters are very similar to those implicated in human infections and harbor genes that confer resistance to multiple antibiotics, posing a risk to public health. The study contributes to expand our current knowledge on the ecology and epidemiology of Salmonella in surface waters.

Multi-drug resistant (MDR) non-typhoidal Salmonella (NTS) is a major public health concern globally. This study reports antibiotic susceptibility testing and genotypic antimicrobial resistance (AMR) profiles of NTS isolates from bovine lymph nodes (n=48) and ground beef (n=29). Furthermore, we compared genotypic AMR data of our isolates with those of publicly available NTS genomes from Mexico (n=1637). The probability of finding MDR isolates was higher in ground beef as compared to lymph nodes: χ2=12.0, P=0.0005. The most common resistant phenotypes involved tetracycline (40.3%), carbenicillin (26.0%), amoxicillin-clavulanic acid (20.8%), chloramphenicol (19.5%) and trimethoprim-sulfamethoxazole (16.9%), while over 55% of the isolates showed decreased susceptibility to ciprofloxacin and 26% were MDR. Occurrence of MDR isolates was strongly associated with NTS serovar (χ2=24.5, P<0.0001), with Typhimurium accounting for 40% of MDR strains. Most of these (9/10), carried Salmonella genomic island 1, which harbors multiple AMR genes (aadA2, blaCARB-2, floR, sul1, tetG) that confer a penta-resistant phenotype. Moreover, 100% of MDR isolates had mutations in the ramR gene. Among public NTS isolates from foods and clinical cases in Mexico to date, those from cattle had the highest proportion of MDR genotypes. Our results suggest attaining significant improvements in AMR meat safety may require the identification and removal (or treatment) of product harboring MDR NTS, instead of screening for Salmonella spp. or for isolates showing resistance to individual antibiotics. In that sense, massive integration of whole genome sequencing (WGS) technologies in AMR surveillance provides the shortest path to accomplish these goals.