Genomic Features of High-Priority Salmonella enterica Serovars Circulating in the Food Production Chain, Brazil, 2000-2016.

Multidrug-resistant (MDR) Salmonella enterica has been deemed a high-priority pathogen by the World Health Organization. Two hundred and sixty-four Salmonella enterica isolates recovered over a 16-year period (2000 to 2016) from the poultry and swine production chains, in Brazil, were investigated by whole-genome sequencing (WGS). Most international lineages belonging to 28 serovars, including, S. enterica serovars S. Schwarzengrund ST96, S. Typhimurium ST19, S. Minnesota ST548, S. Infantis ST32, S. Heidelberg ST15, S. Newport ST45, S. Brandenburg ST65 and S. Kentucky ST198 displayed MDR and virulent genetic backgrounds. In this regard, resistome analysis revealed presence of qnrE1 (identified for the first time in S. Typhimurium from food chain), qnrB19, qnrS1, blaCTX-M-8, blaCTX-M-2 and blaCMY-2 genes, as well as gyrA mutations; whereas ColpVC, IncHI2A, IncHI2, IncFIA, Incl1, IncA/C2, IncR, IncX1 and po111 plasmids were detected. In addition, phylogenetic analysis revealed multiple independent lineages such as S. enterica serovars S. Infantis, S. Schwarzengrund, S. Minnesota, S. Kentucky and S. Brandenburg. In brief, ocurrence and persistence of international lineages of S. enterica serovars in food production chain is supported by conserved genomes and wide virulome and resistome.

Introduction

Salmonella enterica displaying resistance to fluoroquinolone and third-generation cephalosporin remains one of the most pressing global concerns, being deemed a high-priority pathogen by the World Health Organization (WHO)[1]. In this regard, quinolone efflux pumps (oqxA/oqxB) and plasmid-mediated quinolone resistance (PMQR) genes, such as qnrB and qnrS variants have contributed to the increase in fluoroquinolone resistance[2-4]. On the other hand, the coexistence of PMQR and extended-spectrum beta-lactamases (ESBLs), most CTX-M-type, in S. enterica isolates is being increasingly reported at the human-animal interface worldwide[4-6].

Since, there is a growing understanding that the food production chain plays an important role both in the transmission of antibiotic-resistant pathogens and in their evolution and dissemination[4]. Genomic investigation of high-priority S. enterica serovars is a fundamental component of epidemiological surveillance work, particularly in major food-producing countries. Therefore, we have conducted a genomic investigation of S. enterica isolates recovered over a 16-year period from the poultry and swine meat production chains, in Brazil, which currently is one of the largest exporters of chicken and swine meat. In this regard, our results highlight persistence and dissemination of international lineages of multidrug-resistant S. enterica serovars exhibiting a highly virulent genetic background, in the food production chain.

Materials and Methods

Bacterial isolates

During a national surveillance study, 264 nontyphoidal Salmonella enterica (NTS) isolates recovered over a 16-year period (2000 to 2016), from the poultry and swine production chains, in Brazil, have been subjected to antimicrobial resistance screening. Specifically, in this study, we have focused on high-priority S. enterica isolates (n = 43) harboring fluoroquinolones, extended-spectrum β-lactamase (ESBL), and/or plasmid-mediated AmpC (pAmpC) resistance genes. Non-clinical samples were obtained from different points of poultry and swine production chain (Supplementary-Table 1) representing four different geographic regions in Brazil: South (Parana and Santa Catarina), Southeast (Sao Paulo and Minas Gerais), Midwest (Distrito Federal), and Northeast (Bahia) (Supplementary-Table 1). Bacterial isolation and serotyping were performed as previously reported[7-9]. All isolates (n = 43) underwent phenotypic and molecular characterization by Kirby-Bauer disc diffusion (Cefar), as well as broth microdilution using Sensititre® Gram Negative Plates (Trek Diagnostic Systems, OH) and whole-genome sequencing, respectively. Mininmum inhibitory concentration (MIC) values were interpreted according to Clinical and Laboratory Standards Institute (CLSI)[10] and the National Antimicrobial Resistance Monitoring System (NARMS)[11].

Whole-Genome sequencing

DNA extraction was performed using a commercial kit (QiAmp tissue, Qiagen, Germany) according to manufacturer’s guidelines. Genomic DNA of S. enterica isolates (n = 43) harboring fluoroquinolones, ESBL and/or pAmpC resistance genes were sequenced at a 300-bp paired-end-read using the Nextera XT library preparation kit at the MiSeq platform (Illumina, San Diego, CA).

Genomic data analysis

FastQ data were transferred into CLC genomic workbench 10.1.1 (Qiagen) and reads underwent strict quality control as previously described[12]. Subsequently, the software has ensured the appropriate size of the read length (300-bp), as well GC% content around 50%. Sequences obtained were de novo assembled using default settings in CLC workbench 10.1.1 (Qiagen). Resulting contigs were used to determine resistome (ResFinder 3.1), plasmidome (PlasmidFinder 2.0), multilocus sequence typing (MLST 2.0), plasmid MLST (pMLST 2.0), Salmonella pathogenicity island (SPIfinder 1.0), using default settings for all databases such as select threshold for %ID (≥90%) available in Center for Genomic Epidemiology (www.genomicepidemiology.org).

Assemblies were annotated with PROKKA version 1.14-dev[13]. The core genome was defined and aligned using Roary software version 3.11.2, with the BlastP threshold set to 95%[14]. The Roary software performs an all-against-all Blast of sequences produced by an initial CD-Hit clustering. The BlastP threshold was selected to include the mode of hits over a range of thresholds from 60 to 100%, while simultaneously providing a strict criterion to determine a set of vertically inherited core genes that can confidently be used to infer a phylogeny. A pan-genome gene presence-absence matrix was visualized with Phandango[15] using the gene presence/absence output from Roary. Single nucleotide polymorphisms were extracted from the alignment using SNP-sites version 2.3.3[16].

Phylogeny was reconstructed using RAxML version 8.2.12 with a General Time Reversible Model with Gamma distribution for rate heterogeneity[17]. The resulting phylogeny was tested against 600 bootstrap replications and the number of replications necessary was determined by implementing the majority rule, autoMR convergence criteria in the RAxML software[18]. The resulting phylogeny was visualized and annotated using iTol version 3[19].

Sequence data accession number

Sequence data were deposited as part of the GenomeTrakr Project: Thakur Molecular Epidemiology Laboratory, NC State University. We deposited all sequences (n = 43) in GenBank and their accession numbers are listed in Table 1.

Table 1

Features of S. enterica isolates (n = 43) harboring fluoroquinolones, extended-spectrum β-lactamase (ESBL), and/or plasmid-mediated AmpC (pAmpC) resistance genes with a wide virulome.

Strain ID	Serotype	Year	Location	Source	Resistance profile	PMQR	Resistance genes	Plasmid	Sequence type	Virulence genes
STy2 SAMN08606827	Typhimurium	2000	PR	Pig liver	CHL-TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, aadA1, aac(3)-IIa, floR, sul1, tet (A), drfA1	IncHI2A, IncHI2, IncFIA	ST3438	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy13 SAMN08874445	Typhimurium	2013	PR	Pork	CHL-TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, aac(3)-IIa, aadA1, floR, sul1, tet (A), dfrA1	IncHI2A, IncHI2, IncFIA	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy05 SAMN08386803	Typhimurium	2015	SP	Broiler chicken	TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, aadA1, aac(3)-lla, aph(6)-ld, strA, sul1, tet(A), dfrA1	IncHI2A, IncHI2, IncFIA	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy06 SAMN07279711	Typhimurium	2015	SP	Broiler chicken	CHL-TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, strA, strB, aac(3)-IIa, aadA1, floR, sul1, tet (A), drfA1	IncHI2A, IncHI2, IncFIA	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy07 SAMN08606818	Typhimurium	2015	SP	Broiler chicken	CHL-TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, aac(3)-lla, aph(6)-ld, strA, aadA2, floR, sul1, tet(A), dfrA1	IncHI2A, IncHI2, IncFIA	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy08 SAMN08874450	Typhimurium	2015	SP	Broiler chicken	CHL-TET-CIP-GEN-NAL-SXT-AMP-STR	qnrE1	blaTEM-1B, aadA1, aac(3)-lla, floR, sul1, tet(A), dfrA1	IncHI2A, IncHI2, IncFIA, ColpVC	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy011 SAMN08386758	Typhimurium	2015	SP	Broiler chicken	CHL-TET-CIP-GEN-NAL-SXT-AMP	qnrE1	blaTEM-1B, aadA1, aac(3)-lla, floR, tet(A)	IncHI2A IncHI2, IncFIA, po111	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy013 SAMN07283712	Typhimurium	2015	SP	Broiler chicken	CHL-TET-CIP-GEN-NAL-SXT-AMP	qnrE1	blaTEM-1A, aadA1, aac(3)-IIa, sul1, tet (A)	IncHI2A, IncHI2, IncFIA, Incl1	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
STy015 SAMN07279560	Typhimurium	2015	SP	Swine stomach	CHL-TET-CIP-NAL-AMP-STR	oqxA, oqxB,	blaTEM-1A strA, strB, floR, sul2, tet (B)	IncR	ST19	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SI017 SAMN08387036	Infantis	2015	SP	Swine muscle	CHL-TET-CIP-AMP-STR	qnrS1	blaTEM-1B, aadA2, inu(F),floR, tet(A)	IncR	ST32	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SI018 SAMN08387035	Infantis	2015	SP	Swine muscle	CHL-TET-CIP-NAL-AMP-STR	qnrS1	blaTEM-1B, aadA2, inu(F),floR, tet(A)	IncR	ST32	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SI690 SAMN08951109	Infantis	2016	SC	Chicken cage after cleaning	Pansusceptible	qnrB19	—	—	ST32	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc117 SAMN08874439	Schwarzengrund	2016	SP	Swab	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc119 SAMN08874447	Schwarzengrund	2016	SP	Mechanically recovered meat	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc123 SAMN08874435	Schwarzengrund	2016	SP	Mechanically recovered meat	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc126 SAMN08874434	Schwarzengrund	2016	SP	Chicken thigh	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc130 SAMN08874436	Schwarzengrund	2016	SP	Chicken cage after cleaning	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc140 SAMN08874410	Schwarzengrund	2016	DF	Chicken wing paddle	CIP-NAL	qnrB19	—	—	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc146 SAMN08874411	Schwarzengrund	2016	DF	Chicken carcass	CIP-NAL	qnrB19	—	ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc149 SAMN08874405	Schwarzengrund	2016	DF	Chicken carcass	TET-CIP-GEN-NAL-SXT-STR	qnrB19	aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc150 SAMN08874406	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc151 SAMN08951138	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc156 SAMN08951134	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc161 SAMN08951135	Schwarzengrund	2016	MG	Mechanically recovered meat	CIP-NAL	qnrB19	—	ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SN141 SAMN08951153	Schwarzengrund	2016	DF	Chicken carcass	CIP-NAL	qnrB19	—	ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SN143 SAMN08951160	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SN145 SAMN08951155	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aph(3’)-Ia, aac(3)-IVa, strA, strB, aph(4)-Ia, aadA1, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SMi132 SAMN08951176	Schwarzengrund	2016	SP	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, aph(4)-Ia, aadA1, strA, strB, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SMi152 SAMN08951180	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IVa, aph(4)-Ia, aadA1, strA, strB, sul1, tet(A), tet(B), dfrA1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH147 SAMN08951096	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IV, aph(6)-Id, aadA1, aph(3'')-Ib, aph(3')-Ia, aph(4)-Ia, sul1, tet(A), tet(B), dfrA1, qacEdelta1	IncHI2A, IncHI2, ColpVC	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH154 SAMN08951103	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-8, blaCTX-M-2, aadA15, aph(3')-Ic, aph(4)-Ia, strA, strB, sul1, tet(A), tet(B), lnu(A), qacEdelta1	IncHI2A, IncHI2, ColpVC, Incl1	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH157 SAMN08951105	Schwarzengrund	2016	DF	Chicken carcass	TET-AXO-CIP-GEN-NAL-XNL-SXT-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-IV, aph(6)-Id, aadA1, aph(3'')-Ib, aph(4)-Ia, sul1, tet(A), tet(B), dfrA1, qacEdelta1	IncHI2A, IncHI2, ColpVC, Incl1	ST96	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SMi124 SAMN08951178	Minnesota	2016	SP	Chicken carcass	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP-STR	qnrB19	blaCTX-M-8, blaCMY-2, aadA1, sul2, tet(A)	IncA/C2	ST548	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SMi160 SAMN08951185	Minnesota	2016	MG	Chicken feet	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP-STR	qnrB19	blaCMY-2, sul2, tet (A), aadA1, aph(3')Ia	IncA/C2	ST548	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SMi294 SAMN08951202	Minnesota	2016	SP	Mechanically recovered meat	Pansusceptible	qnrB5	aac(6')-Iaa	ColpVC	ST548	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc153 SAMN08951101	Minnesota	2016	DF	Chicken carcass	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP-STR	qnrB19	blaCMY-2, sul2, tet (A), aadA1, aph(3')Ia	IncA/C2	ST548	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SK497 SAMN08951172	Kentucky	2016	BA	Chicken liver	Pansusceptible	qnrB19	aac(6')-Iaa	ColpVC	ST198	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc142 SAMN08874409	Newport	2016	DF	Chicken carcass	AXO-CIP-GEN-NAL-XNL-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-VIa, aadA1, sul1	ColpVC	ST45	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SN144 SAMN08951147	Newport	2016	DF	Chicken carcass	AXO-CIP-GEN-NAL-XNL-AMP-STR	qnrB19	blaCTX-M-2, aac(3)-VIa, aadA1, sul1	—	ST45	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH291 SAMN09207939	Brandenburg	2016	SP	Chicken breast	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP	qnrB19	blaCMY-2, fosA7, sul2, tet (A)	ColpVC, Incl1	ST65	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH686 SAMN09207883	Brandenburg	2016	SC	Mechanically recovered meat	TET-AXO-AUG2-CIP-NAL-XNL-AMP	qnrB19	tet(A)	ColpVC	ST65	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SH159 SAMN08951099	Heidelberg	2016	MG	Chicken cage after cleaning	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP	qnrB5	blaCMY-2, fosA7, sul2, tet (A), aac(6')-Iaa	ColpVC, IncA/C2, Incl1, IncX1	ST15	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA
SSc139 SAMN08874407	Heidelberg	2016	SP	Chicken wing	FOX-TET-AXO-AUG2-CIP-NAL-XNL-AMP-STR	qnrB19	blaCMY-2, fosA7, sul2, tet(A)	ColpVC, IncA/C2, Incl1, IncX1	ST15	invA, slyA, fimH, phoP, phoQ, sipABC, mgtA

Results

Salmonella isolates and distribution of serotypes

Amongst 264 S. enterica isolates, we identified 28 serotypes distributed in the six regions studied (Supplementary-Table 1). Most isolates included S. Heidelberg (n = 81), S. Typhimurium (n = 43), S. Infantis (n = 35), S. Schwarzengrund (n = 21), and S. Enteritidis (n = 20). The remaining sixty-four isolates were classified in 23 different serotypes, including rarely reported S. Abony (n = 6), S. Isangi (n = 4), S. Rochdale (n = 3), S. Saphra (n = 2), S. Orion (n = 2) S. Ouakam (n = 1), S. Grumpensis (n = 1), S. Carrau (n = 1), S. Abaetetuba (n = 1), and S. Idikan (n = 1) (Supplementary-Table 1).

In this study, 264 isolates were screened for resistance to fluoroquinolones and/or broad-spectrum cephalosporins, of which 40 isolates were considered as high priority Salmonella strains due to broad-spectrum cephalosporin- or fluoroquinolone-resistance profiles[1]. Then, genomic investigation was performed on 43 isolates, including high priority Salmonella strains displaying an MDR (n = 36), defined as resistant to three or more classes of antimicrobial compounds[20]; or a quinolone-resistant profile (n = 4). Additionally, 3 representative pan-susceptible Salmonella serovars S. Infantis (SI690), S. Minnesota (SMi294) and S. Kentucky (SK497) were also investigated by WGS, for comparative analysis (Table 1).

Moreover, of the six Brazilian states studied, Sao Paulo (SP) harbored nineteen S. enterica isolates beloging to different serovars such as S. Typhimurium (n = 7), S. Schwarzengrund (n = 6), S. Infantis (n = 2), S. Minnesota (n = 2), S. Brandenburg (n = 1) and S. Heidelberg (n = 1). Salmonella Infantis (n = 1) and S. Brandenburg (n = 1) were isolated from the Santa Catarina (SC) state while the Distrito Federal (DF) harbored S. Schwarzengrund (n = 13), S. Newport (n = 2) and S. Minnesota (n = 1). Finally, the state of Minas Gerais (MG) harbored serotypes S. Schwarzengrund (n = 1), S. Minnesota (n = 1), and S. Heidelberg (n = 1), followed by Parana (PR) and Bahia (BA) with occurrence of S. Typhimurium (n = 2) and S. Kentucky (n = 1), respectively as summarized in Table 1 and Fig. 1.

Figure 1

Distribution of S. enterica isolates (n = 43) harboring fluoroquinolones, extended-spectrum β-lactamase (ESBL), and/or plasmid-mediated AmpC (pAmpC) resistance genes and antimicrobial resistance genes (ARGs) over a 16-year period in Brazil. The map showing the distribution of S. enterica (n = 43) was created using an online service (https://mapchart.net/). Footnotes: *DF, Distrito Federal; MG, Minas Gerais; SP, São Paulo; PR, Paraná; SC, Santa Catarina; BA, Bahia.

Resistome

While these 43 Salmonella isolates reimaned susceptible to colistin and carbapenems, additional genes encoding resistance to aminoglycoside [aadA1, aadA2, aac(6′), aac(3)-Iva, aph(4)-Ia, aac(3)-IIa], sulfonamide [sul1, sul2], tetracycline [tet(A), tet(B)], trimethoprim [drfA1], phenicol [floR], streptomycin [strA, strB], fosfomycin [fosA7], macrolide [inu(F)], and quaternary ammonium [qacEdelta1] were confirmed by WGS as shown in Table 1. All of these genes have been previously reported in Salmonella isolates from variety sources including food and human[5,21].

Among 43 selected isolates, which presented genes conferring resistance to fluoroquinolones, ESBLs and/or pAmpC, 36 carried both PMQR and β-lactams genes, whereas 7 isolates presented PMQR but not β-lactams encoding genes. These isolates belonged to eight serovars including S. Schwarzengrund (n = 20), S. Typhimurium (n = 9), S. Minnesota (n = 4), S. Infantis (n = 3), S. Heidelberg (n = 2), S. Newport (n = 2), S. Brandenburg (n = 2) and S. Kentucky (n = 1).

The PMQR qnrB19 gene, (n = 32) [8 isolates were only positive for qnrB19, and 24 co-produced CTX-M-2, CTX-M-8 or CMY-2 genes] was the most commom quinolone resistance gene observed followed by qnrE1 (n = 8) [1 was positive for TEM-1A and 7 co-produced TEM-1B], qnrS1 (n = 2) [both co-produced TEM-1B], and oqxA/oqxB (n = 1) [co-produced TEM-1A].

The highest PMQR gene diversity was observed in strains isolated from samples collected in Sao Paulo which harbored qnrE1 (n = 6), qnrB19 (n = 9), qnrS1 (n = 2) and oqxA/B (n = 1). Subsequently, Distrito Federal [n = 16; (qnrB19)], Minas Gerais [n = 3; (qnrB19)], Santa Catarina [n = 2; (qnrB19)], Parana [n = 2; (qnrE1) and Bahia [n = 1; (qnrB19)] harbored different genes. The ESBL/pAampC distribution is shown in Table 1. Among these 43 isolates the most frequent source of S. enterica was poultry (34/43; 79%), followed by swine (5/43; 11.6%) and different sources (4/43; 9.3%), including chicken cage after cleaning (n = 3) and swab (n = 1) (Table 1).

In addition to overall resistance, the MIC results reveals 13 resistance patterns among these S. enterica isolates (n = 43) (Table 1). The MIC values ranged from 0.5 µg/mL to 512 µg/mL among 14 antibiotics tested. The majority of quinolone-resistant phenotypes (QRP) isolates presented high-level resistance (MIC ≥32 µg/mL) to nalidixic acid and range between 0.5 to 8 µg/mL for ciprofloxacin. Additionally, most QRP isolates were resistant to levofloxacin (third-generation quinolone), moxifloxacin (fourth generation quinolone) and enrofloxacin (veterinary use only) by using Kirby-Bauer disc diffusion method. Regarding broad-spectrum cephalosporin-resistant, all isolates harbouring blaCTX-M-type presented high-level resistance against ceftriaxone (MIC >64 µg/ml) and ceftiofur (>8 µg/ml). Besides ceftriaxone and ceftiofur (veterinary use only), the isolates that carried blaCMY-2 had high-level resistance against cefoxitin (>32 µg/ml). Further, fourty isolates were resistant to ciprofloxacin and interestingly, one isolate (S. Infantis) was ciprofloxacin-resistant but nalidixic acid susceptible. Lastly, a total of 33 genes in this collection encoded resistance to β-lactams, aminoglicosydes, sulphonamides, tetracycline, phenicols, trimethoprim, microlides, fosfomycin and amonium quaternary (Fig. 2).

Figure 2

(A) Reconstructed phylogeny based on the core genome (3533 genes) of the 42 Salmonella strains. Percent of bootstrap samples in which nodes appeared are shown. The serotype of each isolate is labeled on its respective branch. Color strips depict the year, source, and geographic location of isolation, respectively. Poultry sources include broiler chicken, chicken wing, chicken wing paddle, chicken thigh, chicken feet, chicken breast, chicken cage after cleaning, chicken carcass, and Mechanically recovered chicken meat. Swine sources include: swine stomach, liver, muscle, and pork. Environmental sources include slaughterhouse and swab. (B) Presence and absence of selected antimicrobial resistance genes are shown, black indicating presence. (C) The gene presence/absence matrix depicts pan-genome variation.

Quinolone resistance-determining region (QRDR) among Salmonella enterica isolates

Thirteen isolates (30.2%) exhibited a single mutation in gyrA at codons Ser83 and Asp87, which has been most frequently reported in S. enterica[22]. Of these, eight S. Typhimurium presented mutation at codon Ser83Tyr, one S. Typhimurium isolate at codon Asp87Asn, two S. Brandenburg isolates at codon Asp87Gly, and two S. Heidelberg isolates at codon Ser83Phe. While gyrB, parC and parE were not identified, this single mutation in gyrA was sufficient to promote resistance at >32 µg/mL and >4 µg/mL for nalidixic acid and ciprofloxacin, respectively; it was particularly observed among S. Typhimurium isolates carrying qnrE1 and gyrA mutation, as shown in Table 2. Notably, the isolates, which harbored the combination between gyrA mutations and qnrE1, played a markedly greater role than gyrA and qnrB19 in mediating quinolone resistance (Table 2).

Table 2

Co-occurrence of QRDR and PMQR among S. enterica serovars.

S. enterica strain	Serotype	Source	Date (year)	Statesa	MIC (µg/mL)b		Quinolone resistance-determining region (QRDR)			PMQR
					NAL	CIP	Nucleotide change	Amino acid change	Gene
STy2	Typhimurium	Pig liver	2000	PR	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy13	Typhimurium	Pork	2013	PR	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy05	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy06	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy07	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy08	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy011	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy013	Typhimurium	Broiler chicken	2015	SP	>32	>4	TCC → TAC	Ser83Tyr	gyrA	qnrE1
STy015	Typhimurium	Swine stomach	2015	SP	>32	>0.5	GAC → AAC	Asp87Asn	gyrA	oqxA/oqxB
SH291	Brandenburg	Chicken breast	2016	SP	>32	>1	GAC → GGC	Asp87Gly	gyrA	qnrB19
SH686	Brandenburg	Mechanically recovered chicken meat	2016	SC	>32	>1	GAC → GGC	Asp87Gly	gyrA	qnrB19
SH159	Heidelberg	Chicken cage after cleaning	2016	MG	>32	>0.25	TCC → TTC	Ser83Phe	gyrA	qnrB19
SSc139	Heidelberg	Chicken wing	2016	SP	>32	>1	TCC → TTC	Ser83Phe	gyrA	qnrB19

aPR: Parana; SP: Sao Paulo; SC: Santa Catarina; MG: Minas Gerais;

bNAL: nalidixic acid; CIP: ciprofloxacin; MIC: minimum inhibitory concentrationa.

Identification of international lineages among fluoroquinolone- and cephalosporin-resistant Salmonella enterica serovars

We obtained among 43 S. enterica isolates a total of 9 different sequence types (STs) including the most frequently observed ST96 [S. Schwarzengrund (n = 20)], ST19 [S. Typhimurium (n = 8)] and ST548 [S. Minnesota (n = 4)]. As described above for S. Schwarzengrund, S. Typhimurium, and S. Minnesota, all STs were consistently associated with their respective serotypes: S. Infantis [ST32 (n = 3; 100%)], S. Heidelberg [ST15 (n = 2; 100%)], S. Newport [ST45 (n = 2; 100%)], S. Brandenburg [ST65 (n = 2; 100%)], and also the clinically important ST198 [S. Kentucky (n = 1; 100%)]. Only the S. Typhimurium contained two sequence types, being ST19 (n = 8) and ST3438 (n = 1), highlighting their genetic diversity.

Mobilome

The most common plasmid incompatibility group in our collection was ColpVC (n = 27; 62%). Furthermore, a range of plasmids previously associated with multidrug-resistant foodborne bacteria which have been associated with clinical settings harbored: IncHI2A (n = 24; 55.8%), IncHI2 (n = 24; 55.8%), IncFIA (n = 8; 18.6%), Incl1 (n = 6; 14%), IncA/C2 (n = 5; 11.6%), IncR (n = 3; 7%), IncX1 (n = 2; 4.6%), po111 (n = 1; 2.3%) (Table 1).

The eight S. Typhimurium qnrE1-positive isolates exhibited identical genetic content, regardless of the source (swine or chicken) or year of isolation (2000, 2013, 2015) (Fig. 3). These plasmids were composed by IRL (inverted repeat left)-ISEcp1-IRR (inverted repeat right)-qnrE1-araJ-ahp in a total of 4,659-bp and were different from previous reports[23,24] (Fig. 3).

Figure 3

(A–E) Comparison of the genetic environments of qnrE1 gene. (F) Genetic environment of qnrS1. (G) Representative qnrB19 plasmid for 29 S. enterica isolates. Genes, different plasmids and shotgun sequences were extracted from GenBank database. Arrows indicate the positions and directions of the genes; Δ indicates the truncated gene. Regions with >99% identity are indicated in blue shadow.

Twenty-nine S. enterica containing qnrB19 genes had the plasmid sequences closed at 2,989-bp and each plasmid shared the same incompatibility group, ColpVC (Fig. 3). These small plasmids were identical to qnrB19 that were identified in E. coli in Brazil in 2016 (KX452393.1). They also shared 99% of identity with qnrB19-plasmids from Salmonella Muenchen in the United States in 2017 (KY991368.1) and from S. enterica serovars in Canada in 2018 (CP030230.1). The remaining three isolates had large plasmids ranging from 3,159-bp (S. Schwarzengrund) to 37,696-bp (S. Heidelberg). Notably, the qnrB19-backbone (37,696-bp) of S. Heidelberg (SH159) isolated from a chicken cage after cleaning in this study, showed 100% identity with S. Heidelberg previously reported from human, animal and food sources in Canada in 2016 (CP016580.1) denoting intercontinental spread of bacteria harboring these genes.

The genetic context of qnrS1 on S. Infantis had approximately 4,755-bp and this gene was surrounded upstream by truncated tnpR and downstream by truncated ISKpn19 that were carried on a core quinolone resistance genetic enviroment (Fig. 3). Additionally, the genetic context of blaCTX-M-8 showed that this gene was flanked by two copies of IS26; one which was located 1,922-bp upstream followed by the transposase (tnpA) and IS26 located 869-bp downstream for a total of 3,668-bp (IS26-tnpA-blaCTX-M-8-IS26). These data indicated that this genetic platform has the same mobilization apparatus of CTX-M-8-producing E. coli, which has been primarily isolated from retail chicken meat imported from Brazil[25]. Lastly, IncA/C2 was associated with sul2/tetA in S. Heidelberg, while IncHI2 carried blaCTX-M-2 and Incl1 was responsible by dissemination of blaCMY-2.

Virulome and Salmonella pathogenicity Island

All 43 Salmonella genomes were analyzed for virulence factors using Prokka anotation and SPIfinder 1.0 (https://cge.cbs.dtu.dk/services/SPIFinder/). Several virulence factors were shared between S. enterica serotypes. All isolates were positive for invA and slyA genes, which are responsible for host invasion and cytolysin production, respectively. Most S. enterica displayed important virulence factors involved in pathogenicity process. For example, fimH (n = 43) is an adhesin responsible for host cell specific recognition[26]. Further, were detected in all isolates components of Salmonella Pathogenicity Island (SPI) composed of virulence genes such as phoP, phoQ, pagP, sipA, sipB, sipC, and mgtA. The genes phoP and phoQ are responsible for the control of HilE expression which in turn regulates the expression of SPI-1[27]. Additionally, the others virulence genes (pagP, sipABC, mgtA) are involved in modification of lipid A (intracelular survival and ions transport), mechanisms not only involved in pathogenicity, but also in AMR[28].

Salmonella Pathogenicity Islands harbor a variety of virulence genes, of which most are chromosomally located. These genes are required for interaction among Salmonella spp. and hosts[29]. Most of them are integrated with SPIs and the majority of the Salmonella isolates possess SPI-1[29]. However, the absence or partial deletions of these genetic components in certain circumstances does not interfere in the ability to cause disease, remaining potentially pathogenic presenting infection process such as invasion, intracellular survival and replication[29]. In fact, our findings revealed that not all serovars harbor SPI-1 as shown in Table 3. In this regard, a limitation of this study was the lack of an in vivo analysis for confirmation of the virulence behavior. However, our results confirm previous reports of association between SPI and serotype[30] and highlight risk factors associated with Salmonella host infection[21].

Table 3

Representative distribution of Salmonella Pathogenicity Island among Salmonella serovars.

S. enterica Serovars	Salmonella Pathogenicity Islanda
S. Typhimurium	SPI-1, SPI-2, SPI-3, SPI-5, SPI-9, SPI-13, SPI-14, C63PI
S. Infantis	SPI-2, SPI-3, SPI-4, SPI-5, SPI-13, SPI-14, C63PI
S. Heidelberg	SPI-2, SPI-3, SPI-5, SPI-13, SPI-14, C63PI
S. Newport	SPI-3, SPI-4, SPI-5, SPI-13, SPI-14, C63PI
S. Schwarzengrund	SPI-3, SPI-4, SPI-13, SPI-14, C63PI
S. Brandenburg	SPI-2, SPI-3, SPI-4, SPI-13, C63PI
S. Minnesota	SPI-2, SPI-3, SPI-13, SPI-14, C63PI
S. Kentucky	SPI-2, SPI-3, SPI-4,
aC63PI: Centisome 63 pathogenicity island

Interestingly, regarding virulome, we identified IS200 only among S. Brandenburg isolates. This isertion sequence was recently described in S. Typhimurium to be involved in host gene expression[31,32].

Phylogenetic and evolutionary dinamics of S. enterica isolates

The core genome used for phylogeny reconstruction represented a sizeable portion of the pan-genome, 3533 out of 8286 total genes in the pan-genome (Fig. 2). Each serotype was represented by a monophyletic clade on the reconstructed phylogeny, and bootstrap values of these clades were greater than 99, representing a high confidence in the phylogenetic topology. Genomic variation was observed between serovars, and specifically, antimicrobial resistance genes varied by serovar. The quinolone resistance gene allele was largely similar within serovar and varied between serovars (Fig. 2). Isolates did not cluster by year, source, or geographic location across the phylogeny.

The investigation of genomic diversity between Salmonella isolates is useful from an epidemiological perspective. We observed that specifically in S. enterica serotype and sequence type are the main drivers for cluster analysis, as most of the time isolates were clustered together by serotype and not by resistance profile, year, source or geographic location. Additionally, our results reveal high similarity among serotype regardless of the year of isolation suggesting the widespread distribution and persistence of Salmonella strains in Brazil.

Although the evolutionary relatedness in S. enterica has been improved in the last decade due broad molecular approach studies[33], most often, remains difficult to determine when contamination begins[34] or which isolate is considered a common evolutionary ancestor. In this concern, SNP trees were reconstructed using Salmonella isolates (n = 508) retrieved from GenomeTrakr. Ten different clades were identified as shown in Supplementary Fig. S1 and Supplementary Fig. S2. Of these, five were monophyletic [clades-A, B, E, H, J], and five appear to be from novel clades, since our analysis revealed multiple independent lineages of S. enterica serovars S. Infantis [clade-C], S. Schwarzengrund [clade-D], S. Minnesota [clade-F], S. Kentucky [clade-G] and S. Brandenburg [clade-I] (Supplementary Fig. S1).

Clade A was composed of 10 MDR S. Typhimurium isolates recovered from Brazil. Besides our 7 isolates, 1 isolate was recovered in 2009 from an industrialized product (CFSAN033917), as well as 2 isolates were recovered from swine collected in 2012 (CFSAN068037) and 2015 (UFRGS-SA052). Non-association with international strains was observed. In this regard, most likely this clade is endemic in the Brazilian food sector. Most isolates (n = 13) clustered in clade B were clinical or host-associated recovered in USA (PRJNA230403). These isolates grouped together with S. Infantis isolates (SI017 and SI018) from this study. Interestingly, S. Infantis SI690 clustered with 2 pan-susceptible S. Infantis identified in 2016 by this study, constituting a new clade, named clade C.

All isolates from clade D were MDR S. Schwarzengrund recovered from different sources in Brazil. These 21 isolates did not cluster with isolates from other countries. Clade E was mostly constituted by S. Minnesota isolated in the United Kingdom (Supplementary Fig. S1). S. Minnesota isolates SMi124, SMi160 and SSc153 were clustered with 3 S. Minnesota isolated from Gallus gallus in Brazil and 38 S. Minnesota isolates recovered in UK, all being predominantly found in humans and food (Supplementary Fig. S1). Among S. Minnesota, most showed the same resistance profile and carried qnrB19 and blaCMY-2 supporting clonal spread of this lineage. On the other hand, S. Minnesota SMi294 did not cluster together with the previous clade being nested with another S. Minnesota isolate within clade F (Supplementary Fig. S1). Although this isolate carried qnrB19, the resistance profile was pan-susceptible. Therefore, the resistance genotype appear to be a determining factor to clustering this isolate outside of the clade E.

Regarding clade G, two pan-susceptible S. Kentucky ST198 were nested between each other, which appear not to be related to the highly-drug resistant clade disseminated in Africa[35].

Clade H, besides our 2 S. Newport isolates recovered from chicken carcass, included an isolate from pilgrims (SAMEA2673767) carrying blaCTX-M-2.

A novel clade (designated I), clustered only two S. Brandenburg isolates (Supplementary Fig. S1), both from poultry samples collected from two different geographic locations in Brazil (Table 1). These isolates were multidrug-resistant harboring important resistance genes, including qnrB19, blaCMY-2, fosA7, aac(6′)-Iaa, sul2 and tetA (Table 1). Lastly, clade J was composed of 403 S. Heidelberg isolates including strains from Brazil, UK and Germany. Of these isolates, 77 were from our Salmonella collection (Supplementary Fig. S2). S. Heidelberg, currently circulating in Europe, are most likely part of a Brazilian clade that has been the most frequent in Brazil. In this serovar, the presence of the plasmid-mediated blaCMY-2, readily mobilized, seems to be a major public health issue.

Discussion

S. enterica harboring qnrB19 has become the most common PMQR gene observed in Brazil and has been increasing in the US[36]. In addition to this study describing qnrB19 identified among 8 serotypes, there are only three additional reports worldwide on qnrB19-producing S. Infantis[37] S. Heidelberg[38] and S. Newport[39] isolated from Colombia, Venezuela and Poland, respectively. In this context, this is the first know report of qnrB19 in Salmonella serovars S. Schwarzengrund (poultry and environmental), S. Infantis (environmental), S. Minnesota (poultry), S. Brandenburg (poultry), and S. Kentucky (poultry), given that qnrB19 have been reported in Brazil only in S. Corvallis[40], in E. coli[41,42] or K. pneumoniae[43,44].

Isolates harboring qnrE1 had not previously been reported for Salmonella enterica isolated from poultry and swine. Interestingly, qnrE1 was reported in K. pneumoniae from human in Argentina[23] and in K. pneumoniae isolated from a parrot in Brazil[24]. In this regard, to our knowledge, this is the first report of qnrE1 (poultry and swine) in S. enterica serovar Typhimurium worldwide. These results emphasize the plasticity of this new plasmid and the presence of ISEcp1 might be the key vector to silent spread of this resistance gene. These results will aid in the development of mitigation strategies to limit the global distribution of bacteria harboring these genes, since genomic surveillance study allow us to predict, prevent and manage antibiotic resistance and virulence markers in One Health interface, providing substantial evidences that can be implemented in Brazilian food sector.

Two fluoroquinolone-resistant S. Infantis isolated from swine were found to be carrying qnrS1. Conversely, qnrS1 in Brazil was associated only with E. coli[41], K. pneumoniae[43,45] or Pseudomonas aeruginosa[46]. In this regard, the presence of oqxA/oqxB in swine highlights their importance as emergence of qnrS1 and oqxA/oqxB quinolone resistance genes being a public health concern in swine production chain.

Lastly, the high prevalence of PMQR shown in this study is consistent with those other studies that found qnrB19 in 2011[40], 2012[41], 2013[43], 2015[44], 2017[42], qnrS1 in 2012[41], 2013[43], 2014[45], 2016[46] and qnrE1 in 2017[24], in Brazil highlighting an urgent need to strengthen surveillance. This finding emphasizes the importance of persistence of quinolone resistance genes emerging in the poultry and swine production chain and calls for action to arrest further transmission and dissemination.

Although ESBL resistance genes are well described within the environment-food-human interface[6,47], the wide distribution of CTX-M-8, CTX-M-2 and CMY-2 ESBL/pAmpC genes are regarded as a threat to public health, as they are easily transferred horizontally to other foodborne pathogens and commensal bacteria in gut environment[48]. In fact, the high prevalence of ESBLs/pAmpC strains shown in this study is consistent and highlights the endemic occurrence of broad-spectrum cephalosporin strains in South America[49]. In addition, this is the first report in Brazil, identifying S. Heidelberg harboring a new gene, fosA7, which encodes resistance to fosfomycin [as previously described in S. Heidelberg, also isolated from broiler chicken in Canada][50].

Out of the nine sequence types found, ST19 was associated with strains isolated from swine and poultry. This ST19 appears to be closely related to the sub-Saharan Africa ST313 clade, which is gastroenteritis-associated and globally distributed[51,52]. However, based on previous investigations[53], as well as showed in clade A (Supplementary Fig. S1), we suggest that S. Typhimurium ST19 strains from Brazil are genetically distinct from those ST19 strains associated with gastroenteritis worldwide, given they did not cluster with international isolates.

Sequence type 32 was associated with strains isolated from swine and chicken cage after cleaning. This ST is highly conserved in S. Infantis and has been associated with a clonal dissemination in food sources and human[21,54,55]. Conversely, isolates from poultry and environmental sources were associated with ST96 and ST15. Interestingly, S. Schwarzengrund (ST96) was previously associated with ESBL from poultry in Brazil [2013][56], carbapenemase resistant KPC-2-producing S. Schwarzengrund from human in Argentina [2014][57], and most recently with mcr-1-producing S. Schwarzengrund isolated from poultry in Brazil [2018][58]. In addition, ST15 which had not been previously reported in Brazil has become the most prevalent and relevant serotype in Brazil (manuscript in preparation). Also of interest, STs 548, 198, 45 and 65 were associated with strains isolated from one source (poultry).

In order to support the current knowledge regard the epidemiological distribution of MDR strains between the food-animal-environmental interface, our results provide valuable information related to distribution of multidrug-resistant S. enterica serovars in food-producing animal settings. In addition to the range of mobile genetic elements identified in our isolate collection, these data provide additional insights into global mobility and genomic plasticity, which contribute to persistence of strains along food chain.

The widespread of multi-drug resistant S. enterica in poultry and swine production chain is concerning due the potential transmission to human in the end of food chain. Given these isolates are resistant to fluoroquinolones and third-generation cephalosporin raises a particular concern, since these antibiotics are the first choice for the treatment of salmonellosis. While, our results provide additional evidences of the global mobilization of international clones of S. enterica, over a 16 year-period, continuous surveillance and additional studies in MDR S. enterica isolated from human, needs to be established as mitigation strategies to limit their global spread.

Supplementary Dataset