Genotypes and Antimicrobial Susceptibility Profiles of Hemolytic Escherichia coli from Diarrheic Piglets

Abstract

Hemolytic Escherichia coli are important pathogens in neonatal and weaned pigs. In this study, we analyzed 95 hemolytic E. coli isolated from intestinal contents or fecal samples of diarrheic piglets in 15 states of the United States between November 2013 and December 2014. Phenotypic antimicrobial susceptibility was determined through Sensititre BOFO6F plates for all the strains. They were all resistant to clindamycin, penicillin, tiamulin, tilmicosin, and highly resistant to oxytetracycline (91.6%), chlortetracycline (78.9%), ampicillin (75.8%), and sulfadimethoxine (68.4%). 86.2% of them were multidrug resistant. Whole genome sequencing (WGS) showed that 55 strains were enterotoxigenic E. coli (ETEC) and 40 strains were non-ETEC, and the strains belonged to 22 known and 2 novel sequence types (STs). ST100 and ST10 were the main and predominant STs in ETEC strains, whereas the non-ETEC strains were diverse with ST23 and ST761 as the main STs. Antibiotic resistance gene/mutation profiling of the genomes confirmed the results of antimicrobial susceptibility test. Notably, significant differences were found in the susceptibility to enrofloxacin between ETEC and non-ETEC (58.2% vs. 5.0%) and gentamicin (32.7% vs. 7.5%). ampH, ampC2, and ampC1 were the most common beta-lactamase genes in all E. coli strains, and extended-spectrum beta-lactamase (ESBL) genes were rare in these isolates. This study provides new insights into antibiotic resistance and genotypes of intestinal pathogenic E. coli associated with swine disease in the United States, and support the utility of WGS in accurate prediction of resistance to most antibiotics.

Introduction

Diarrhea induced by Escherichia coli in young piglets is a significant problem for swine production (Fairbrother and Gyles, 2012). Outbreaks may occur within the first week after birth, during the nursing period, or 1–2 weeks after weaning or after abrupt changes in environment or nutrition. Enteric colibacillosis may result in significant economic losses due to increased mortality, decreased weight gain, and associated cost for treatments, vaccinations, and feed supplements (Fairbrother and Gyles, 2012). There are six different types of intestinal pathogenic E. coli, each of which may possess several of the many virulence factors. Among them, enterotoxigenic E. coli (ETEC) and enteropathogenic E. coli (EPEC) are two major pathotypes involved in swine diarrhea (Nataro and Kaper, 1998; Kaper et al., 2004). ETEC produce enterotoxins (LT, STa, STb), which stimulate fluid secretion into the intestine and subsequently result in diarrhea (Levine, 1981; Evans and Evans, 1996). EPEC was initially associated with infant diarrhea in children; subsequently, it was found that EPEC also induce similar symptoms in pigs. EPEC do not secrete toxin, but attach to the intestinal mucosa and cause lesions of attaching and effacing (Janke et al., 1989). Although less common, some Shiga toxin-producing E. coli (STEC) strains cause edema disease in addition to diarrhea in young pigs (Francis, 1999).

Various approaches have been used to prevent and control pathogenic E. coli in pigs, including passive administration of antibodies, dietary supplementation such as prebiotics and dietary preventive measures, genetic breeding for ETEC-resistant herds, and oral use of live nontoxigenic E. coli vaccines (Zhang, 2014). Although several preventive approaches have shown some promise and efficacy for both neonatal and postweaning diarrhea, antibiotics are still frequently used to treat enteric colibacillosis (Luppi, 2017). The use of antibiotics in pig industry for therapeutic treatment or for prophylactic or metaphylactic purposes has resulted in the selection of antibiotic-resistant bacteria and favored their emergence and spread within the production system (Burow et al., 2014). The presence of resistant bacteria in animals is worrying not only from a veterinary clinical perspective but from a zoonotic aspect as well (Marshall and Levy, 2011). Resistant bacteria among food-producing animals may pose a health risk to humans through several routes, which include the consumption of animal products, exposure to resistant microorganisms from contact with animals, and the contamination of ground and surface waters by wastes containing antimicrobials and resistant microorganisms (Marshall and Levy, 2011; Verraes et al., 2013).

Antimicrobials commonly used to treat enteric colibacillosis in swine are chosen for their ability to achieve therapeutic outcomes. The most frequently used are enrofloxacin, apramycin, ceftiofur, neomycin, gentamicin, and amoxicillin/clavulanic acid (Fairbrother and Gyles, 2012). Antibiotic resistance to apramycin, neomycin, trimethoprim-sulfonimide, and colistin has been increasingly observed, in particular in ETEC strains (Zhang, 2014). Managing diarrhea in pigs requires an understanding of the pathotypes, genotypes, and antimicrobial resistance (AMR) profiles of E. coli to implement appropriate strategies for prevention and control. Toward this goal, we collected 95 hemolytic E. coli isolates associated with pig diarrhea from 15 different states in the United States. In this study, we characterized the sequence type (ST) of these isolates and profiled their AMRs by both phenotypic test and genome sequencing.

Materials and Methods

Collection of E. coli isolates

Ninety-five E. coli isolates were used and most of them were obtained from the intestinal contents or fecal samples of pigs from 15 states in the United States between November 2013 and December 2014 (Supplementary Table S1 to show the metadata of each isolate). The pigs were 10 days to 16 weeks old and clinically had severe diarrhea or acute death. The samples were shipped to Iowa State University (Ames, IA) diagnostic laboratory in ice-cooled containers. Once received, E. coli were isolated using blood agar plates (5% sheep blood) and identified by Matrix Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI-TOF MS). Pathotypes of E. coli were classified by whole genome sequencing (WGS) of toxins and adherence genes (toxins such as heat stable toxin, heat labile toxin, Shiga toxins 1, 2, and 2e, and adherence factors such as K88, K99, F41, 987P, and F18) as described previously (Nguyen et al., 2005).

Antimicrobial susceptibility test

A standard antimicrobial susceptibility testing was performed on all of the isolates using commercially available Sensititre BOFO6F plates (Thermo Fisher Scientific, Waltham, MA) in the form of broth microdilution at the Veterinary Diagnostic Laboratory of Iowa State University (ISU VDL) following the Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI, 2013). Eighteen antimicrobials of veterinary and human health importance were tested, including ampicillin (AMP), ceftiofur (CEF), chlortetracycline (CTC), oxytetracycline (OTC), enrofloxacin (ENR), florfenicol (FFL), gentamicin (GEN), neomycin (NEO), trimethoprim/sulfamethoxazole (STX), sulfadimethoxine (SDM), spectinomycin (SPT), clindamycin (CLI), danofloxacin (DFX), penicillin (P), tiamulin (TIA), tilmicosin (TIL), tulathromycin (TUL), and tylosin (TYL). Minimum inhibitory concentrations were determined and the isolates were categorized as susceptible, intermediate, or resistant based on the clinical breakpoints set by CLSI (2015). Isolates resistant to three or more than three antimicrobial classes were considered as multidrug-resistant (MDR) isolates.

DNA extraction and library preparation

A single colony of E. coli was inoculated into Luria Bertani broth (Becton Dickinson, Franklin Lakes, NJ) and incubated at 37°C with shaking overnight. E. coli cells were collected and genomic DNA (gDNA) was extracted using ChargeSwitch gDNA mini bacteria kit (Life Technologies, Carlsbad, CA). DNA quality was measured by Nanodrop (Thermo Fisher Scientific) and a Qubit fluorometer double-stranded DNA high sensitivity (dsDNA HS) kit (Life Technologies). Indexed genomic libraries were prepared using the Nextera XT kit (Illumina, San Diego, CA). Genomic library DNA was quantified by Qubit fluorometer dsDNA HS kit (Life Technologies).

Genome sequencing and assembly

Twenty-four multiplexed strains were put in one lane and sequenced on an Illumina Miseq Desktop Sequencer (Illumina) at Iowa State University with 250 bp paired end strategy. Resultant reads were demultiplexed. Adapters and low-quality nucleotides were trimmed and/or removed from raw reads using Trimmomatic-0.36 (Bolger et al., 2014). Postprocessed reads were assembled using SPAdes Genome Assembler version 3.8.1-Linux (Bankevich et al., 2012). Assembly quality was evaluated using both custom scripts and abyss-fac command in ABySS (www.bcgsc.ca/platform/bioinfo/software/abyss) to determine N50, longest contigs, total length of contigs, GC% content, and other parameters as appropriate (Supplementary Table S4 to show the summary of assembly statistics). Next-generation sequencing (NGS) data were deposited on National Center for Biotechnology Information (NCBI) database under the Sequence Read Archive (SRA) accession SRP148876.

Antibiotic resistance genes detection and molecular typing

Multilocus sequence typing (MLST) was performed by using the classic scheme with seven housekeeping genes with an online service, Center for Genomic Epidemiology (https://cge.cbs.dtu.dk/services/MLST). Detection of antibiotic resistance genes were performed using the SRST2 toolkit version 0.2.0 (Inouye et al., 2014) with default parameter. The up-to-date database of Antibiotic Resistance Gene-ANNOTation (ARG-ANNOT) was used as the reference (Gupta et al., 2014) resistance gene database. Point mutations of gyrA, gyrB, parC, and parE related to quinolone resistance were detected by BLASTn. Core-genome phylogenetic tree was generated by Parsnp (Treangen et al., 2014) in the Harvest package using assembled genomes. The parameters for the maximum likelihood (ML) tree construction were defaulted with the evolutionary model of General Time Reversible (GTR) and 1000 resamples for bootstrap. The tree and the molecular features of each isolate were visualized by the online tool iTOL (Letunic and Bork, 2016).

Statistical analysis

Statistical analysis was performed using descriptive statistics, and Pearson's chi-squared test in SPSS (version 20) was employed for testing group differences, with p < 0.05 set as the level of significant differences and p < 0.01 set as the level of extremely significant differences. Genotype–phenotype correlations were statistically analyzed by using Cohen's kappa coefficient as described previously (Feuerman and Miller, 2008).

Results

Antimicrobial susceptibility

Among the 95 hemolytic E. coli isolates, the majority of them (78 of 95) were from Iowa, Minnesota, Indiana, Illinois, and South Dakota (Supplementary Table S1). Antimicrobial susceptibility test by Sensititre BOFO6F plates set showed that all isolates were resistant to clindamycin, penicillin, tiamulin, and tilmicosin (no interpretation available to danofloxacin, tulathromycin, and tylosin) (Supplementary Table S2). The other antibiogram of the 95 isolates was characterized by high incidence of resistance to oxytetracycline and chlortetracycline (91.6% and 78.9.1%, respectively), intermediate incidence of resistance to the ampicillin and sulfadimethoxine (75.8% and 68.4%, respectively), and low prevalence of resistance to gentamicin (21.1%) (Table 1). Notably, 28.4% isolates were resistant to ceftiofur, and 35.8% isolates were resistant to enrofloxacin. Overall, 88.4% of the isolates were resistant to more than one antibiotic and 38.9% (37/95) were resistant to more than seven antimicrobials, only 5.3% were susceptible to all drugs tested, and 86.2% were MDR.

Table 1.

Minimum Inhibitory Concentrations of 11 Antimicrobial Agents for 55 Enterotoxigenic Escherichia coli and 40 Nonenterotoxigenic Escherichia coli Isolated from Pigs with Clinical Diarrhea

Antimicrobials	Int/MIC			ETEC				Non-ETEC
Antimicrobials	S	I	R	No. of susceptible strains (%)	No. of intermediate strains (%)	No. of resistant strains (%)	No. of NI strains (%)	No. of susceptible strains (%)	No. of intermediate strains (%)	No. of resistant strains (%)	No. of NI strains (%)
Ampicillin	4	16	16	5 (9.1)	1 (1.8)	49 (89.1)	0 (0)	16 (40)	1 (2.5)	23 (57.5)	0 (0)
Ceftiofur	0.5	4	8	39 (70.9)	2 (3.6)	14 (25.5)	0 (0)	27 (67.5)	0 (0)	13 (32.5)	0 (0)
Chlortetracycline	1–4	8	8	5 (9.1)	6 (10.9)	44 (80.0)	0 (0)	5 (12.2)	4 (10)	31 (75.6)	0 (0)
Oxytetracycline	1–2	—	8	3 (5.5)	0 (0)	52 (94.5)	0 (0)	5 (12.5)	0 (0)	35 (87.5)	0 (0)
Enrofloxacin	0.12–0.25	—	2	23 (41.8)	0 (0)	32 (58.2)	0 (0)	38 (95)	0 (0)	2 (5)	0 (0)
Florfenicol	2	4	8	17 (30.9)	16 (29.1)	22 (40.0)	0 (0)	5 (12.5)	13 (32.5)	22 (55)	0 (0)
Gentamicin	1	8	16	32 (58.2)	5 (9.1)	18 (32.7)	0 (0)	37 (92.5)	0 (0)	3 (7.5)	0 (0)
Neomycin	4	-	32	28 (50.9)	0 (0)	27 (49.1)	0 (0)	23 (57.5)	0 (0)	17 (42.5)	0 (0)
Spectinomycin	8	16	64	4 (7.3)	23 (41.8)	24 (43.6)	4 (7.3)	1 (2.5)	17 (42.5)	14 (35)	8 (20)
Sulfadimethoxine	≤256	—	>256	21 (38.2)	0 (0)	34 (61.8)	0 (0)	9 (22.5)	0 (0)	31 (77.5)	0 (0)
Trimethoprim/sulfamethoxazole	≤2	—	>2	38 (69.1)	0 (0)	17 (30.9)	0 (0)	25 (62.5)	0 (0)	15 (37.5)	0 (0)
Clindamycin	—	—	16	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)	0 (0)
Penicillin	—	—	8	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)	0 (0)
Tiamulin	—	—	32	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)	0 (0)
Tilmicosin	—	—	64	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)	0 (0)
Tulathromycin	—	—	—	0 (0)	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)
Tylosin	—	—	—	0 (0)	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)
Danofloxacin	—	—	—	0 (0)	0 (0)	0 (0)	55 (100)	0 (0)	0 (0)	0 (0)	40 (100)

MIC levels are given in μg/mL.

ETEC, enterotoxigenic E. coli; I, intermediate; Int/MIC, interpretation/minimum inhibitory concentration; NI, no interpretation available based on antimicrobial, organism, species, and tissue combination; R, resistant; S, susceptible.

The 95 isolates were further classified into two groups based on their toxin genes contents as determined by WGS analysis. Fifty-five isolates were ETEC, and the other 40 were non-ETEC (Supplementary Table S2). As shown in Supplementary Table S2 about the antibiogram of each isolate all ETEC isolates were resistant to at least one antimicrobial. About 89.1% ETEC isolates were resistant to at least three classes of antibiotics, and 20.0% ETEC isolates were resistant to all six antimicrobial classes. In non-ETEC, 77.5% strains were resistant to multiple antibiotics, but 12.5% isolates were susceptible to all the antibiotics tested.

The prevalence of oxytetracycline resistance was similar in the ETEC isolates (94.5%) and the non-ETEC isolates (87.5%). Seven (AMP, CTC, OTC, ENR, GEN, NEO, and SPT) antibiotics showed higher resistance rates in the ETEC isolates than in the non-ETEC isolates (Table 1). Interestingly, we found that enrofloxacin resistance was significantly higher in the ETEC strains (58.2%) than in the non-ETEC (5.0%) (p < 0.01). Likewise, a similar observation was also found for gentamicin resistance, which was 32.7% in ETEC versus 7.5% in non-ETEC (p < 0.01). In ETEC strains ceftiofur resistance was the lowest (25.5%), whereas in non-ETEC strains enrofloxacin resistance was the lowest (5.0%).

MLST analysis

The core genome phylogenetic tree indicated only a few main clusters for the isolates, but MLST analysis revealed that there were 22 known and 2 novel STs (Fig. 1). The classification of phylogenetic subgroups demonstrated that the 95 isolates were mainly in phylogroup A (n = 76), and only 14, 3, and 2 isolates belonged to phylogroups B1, B2, and D, respectively. The predominant ST was ST100 (n = 30) and the second most common was ST10 (n = 18). In the ETEC group, 29 of the 55 isolates (52.7%) were ST100 and 13 isolates were ST10 (23.6%), which represent the most prevalent STs overall. As shown in Figure 1, the non-ETEC isolates were contrastively scattered into multiple MLST types. Only one isolate belonged to ST100 and five isolates (12.5%) belonged to ST10. Interestingly, of the 30 ST100 isolates, all but one (IA723-018, ETEC) were MDR. Also, all the ST88 (n = 4) isolates in ETEC were resistant to all classes of the tested antibiotics.

FIG. 1.

Genomic analysis of 95 Escherichia coli genomes sequenced in this study. A maximum-likelihood phylogenetic tree was constructed using the core-genome Single Nucleotide Polymorphisms (SNPs) and midpoint rooted. Bootstrap values (1000 replicates) >70 are shown. The phylogenetic groups, A, B1, B2, and D are present in different color bars. Sequence type is shown along with each strain. The presence and absence of antibiotic resistance genes are denoted by filled circles and blanks, respectively. Different colors for the antibiotic resistance gene represent different antibiotic group. Details of the genes in each category are shown in Supplementary Table S3.

Prevalence of antibiotic resistance genes by WGS

To uncover the potential resistance mechanisms, the resistance genes in all the 95 isolates were analyzed using the NGS data. In total, 59 resistance genes were detected, which are involved in resistance to nine categories of antibiotics, including beta-lactams, aminoglycosides, rifampicin, chloramphenicol, methoxypyrimidine, macrolides, quinolones, sulfonamides, and tetracyclines (Fig. 1). The results showed that the beta-lactamase genes were most prevalent in the isolates. All the isolates encoded ampH and ampC2, whereas 93.7% isolates had ampC1. Tetracycline resistance genes ranked second and were present in 86.3% (82/95) isolates, including tet(A), tet(B), tet(C), tet(D), and tet(M). Among them, tet(A) accounted for 56.8%, but tet(C) and tet(M) were only detected in one isolate. Fifty-nine isolates (62.1%) had resistance genes to aminoglycosides and 14 genes known to confer resistance to aminoglycosides were detected; however, armA, the 16S ribosomal RNA (rRNA) methyltransferase conferring high-level resistance to aminoglycosides, was only present in one isolate. Multiple chloramphenicol-resistant genes catA1, catA2, catBx, cmlA, and floR were detected in the isolates, but cmlA was the most prevalent one, occurring in 29.5% isolates. floR, which emerged recently and had spread quickly all over the world, was the next prevalent chloramphenicol resistance (18.9%).

Genotypic characteristics of resistance to β-lactam antibiotics

Beta-lactam antimicrobials are the most common treatments for bacterial infections and continue to be the prominent antimicrobials resisted by Gram-negative bacteria worldwide. Ten genes encoding beta-lactamases were identified in the E. coli isolates tested in this study. The gene ampH was present in all the isolates, two genotypes of ampC (i.e., ampC2 and ampC1) were found in 100% and 93.7% of the isolates, respectively. bla _TEM was present in 53 (55.8%) of the isolates. Nine variants of bla _TEM were found, which were bla _{TEM-33, 34, 54, 70, 104, 105,} _{150, 186, 198}. Among them, TEM-198 was the most prevalent variant (64.2%, 34/53). Among the nine detected variants of bla _TEM in this study, bla _TEM-70 and bla _TEM-186 were extended-spectrum beta-lactamase (ESBL) genes (Pitout and Laupland, 2008), but ESBL genes were not very common in these isolates, except for two bla _TEM; only bla _CTX-M-32, bla _SHV-12, and bla _OXA-1 were detected in one, one, and three isolates, respectively.

bla _CTX-M and bla _SHV encode CTX-M and SHV beta-lactamases, respectively. The presence of bla _CTX-M or bla _SHV was detected frequently in bla _TEM-198 isolates; however, bla _OXA-1 did not coexist with other beta-lactamase genes. Bla _CMY was present in 27.4% of the isolates (26/95) and included three variants bla _CMY-44, bla _CMY-33, and bla _CMY-22. bla _CMY-44 was the most prevalent variant (53.8%, 14/26), followed by bla _CMY-33 (42.3%, 11/26).

Genotypic characteristics of resistance to quinolone

The resistance to quinolones in E. coli is mediated by the plasmid-mediated quinolone resistance (PMQR) genes (Hopkins et al., 2008), or more commonly by mutations in the quinolone resistance-determining region (QRDR) of genes encoding DNA gyrase genes (gyrA and gyrB) and topoisomerase IV genes (parC and parE) (Yoshida et al., 1991; Everett et al., 1996; Vila et al., 1996; Hooper, 1999; Sorlozano et al., 2007). Among 95 E. coli isolates tested in this study, only one strain (isolate IA724-016) carried two PMQR genes (oqxA and oqxB) and another strain (isolate IA724-023) carried the qnrB gene. However, 43.2% (41/95) of the isolates carried gyrA mutations and 29.5% (28/95) carried parC mutations, which were known to be responsible for quinolone resistance. Among them, 25 isolates carried both gyrA and parC mutations. For gyrA, 39 isolates had the Ser-83→Leu substitution, and 11 isolates had the Asp-87→Asn (n = 6), Asp-87→Tyr (n = 2), or Asp-87→Gly (n = 3) substitutions. Nine isolates had both Ser-83 and Asp-87 substitutions by Ser-83→Leu and Asp-87→Asn (n = 6), Asp-87→Tyr (n = 1) or Asp-87→His (n = 2). In regard to parC, all 28 isolates had a substitution of Ser-80→Ile. Also, 25 isolates carried mutations in both gyrA and parC, but none of these isolates had both the gyrA/parC mutations and PMQR genes concurrently. Of the ETEC isolates, 67.3% (37/55) were identified to be resistant to quinolone by the known gyrA or parC-resistant mutations, whereas four non-ETEC isolates carried known gyrA or parC-resistant mutations, which was consistent with the results of the phenotypic susceptibility testing (Fig. 1).

Genotypic characteristics of resistance to other antibiotics

Fifty-five isolates carried the sul family genes conferring resistance to sulfonamides. However, 11 additional isolates were phenotypically resistant to sulfonamides, but did not have known resistant genes, and their mechanisms remain to be identified. For trimethoprim resistance, 27 isolates carried gene drf, consistent with phenotypic results. The drf variants included dfrA19, dfrA1, dfrA15b, dfrA5, dfrA7, dfrA8, and dfrA12. Of the 95 isolates, 12 carried macrolides resistance gene ereA, ermB, mefB, mphA, mphB, mphE, or msrE, and 3 had both mphE and msrE. Three isolates had ADP-ribosyl transferase (arr) gene resistant to rifampicin, but their resistance was not tested phenotypically in this study.

Comparison between genotypes and phenotypes of AMRs

To determine the utility of WGS in predicting resistance to different antibiotics, we compared the antibiotic resistance phenotypes with their corresponding genotypes. As shown in Table 2, the resistances to aminoglycosides, chloramphenicol, methoxypyrimidine, quinolones, sulfonamides, and tetracyclines were highly correlated between genotypes and phenotypes. Because of the 100% presence of ampC2 gene, the genotype of beta-lactams was 100%, but for the two detected antibiotics ampicillin and ceftiofur, their genotype was 75.8%. There were another two special antibiotics: rifampicin and macrolides. On the Sensititre BOFO6F plate, there were no rifampicin antibiotics. Meanwhile, all the macrolides were special for Gram-positive bacteria; as a result, all the isolates were resistant to tiamulin and tilmicosin (no interpretation available to tulathromycin and tylosin). In general, WGS accurately predicted the vast majority of resistance phenotypes from MDR E. coli strains.

Table 2.

Comparison Between Genotypes and Phenotypes of Antimicrobial Resistance

Antibiotic	Phenotype: susceptible		Phenotype: resistant
Antibiotic	Genotype: resistant	Genotype: susceptible	Genotype: resistant	Genotype: susceptible	Kappa ^a
AMP	3	20	72	0	0.91
CEF	0	39	27	0	1
CTC	1	9	72	3	0.79
OTC	0	8	82	5	0.73
ENR	9	52	33	1	0.78
FFL	1	21	31	13	0.58
GEN	34	35	21	0	0.32
NEO	18	33	40	4	0.54
SPT	1	4	37	1	0.77
SDM	0	30	54	11	0.76
SXT	26	37	29	3	0.42
Overall	93	288	498	41	0.69

Genotype–phenotype agreement of antibiotic resistances to clindamycin, penicillin, tiamulin, tilmicosin, danofloxacin, tulathromycin, and tylosin were not analyzed.

Genotype–phenotype agreement was statistically analyzed by using Cohen's kappa coefficient (Feuerman and Miller, 2008).

AMP, ampicillin; CEF, ceftiofur; CTC, chlortetracycline; ENR, enrofloxacin; FFL, florfenicol; GEN, gentamicin; NEO, neomycin; OTC, oxytetracycline; SDM, sulfadimethoxine; SPT, spectinomycin; STX, trimethoprim/sulfamethoxazole.

Discussion

Diarrhea caused by pathogenic E. coli, including ETEC and EPEC, is one of the most common diseases in young piglets. Vaccination is the main intervention, but antibiotics are still needed in some cases, such as urgent or severe diseases. Characterization of the antibiotic susceptibility profiles of E. coli would be helpful to guide the clinical use of antibiotics. In this study, we investigated the susceptibility to 18 clinically used antibiotics and the corresponding molecular mechanisms for resistance in hemolytic E. coli isolates from 15 states in the United States from 2013 to 2014. It was found that both ETEC and non-ETEC isolates were highly resistant to oxytetracycline, ampicillin, chlortetracycline, and sulfadimethoxine, which is similar to what was reported in other studies (Fairbrother, 1999; Hariharan et al., 2004). The resistance rate to ceftiofur was at 25.9% in ETEC in this study. However, the resistance rate of ceftiofur has been rising up continuously; Daniel et al. found cephalothin resistance significantly increased over time (0.43%/year) among animal E. coli isolates (Tadesse et al., 2012). Winokur et al. (2000) found 16% of clinical E. coli isolates from cattle and swine in Iowa to be resistant to extended-spectrum cephalosporins. This may be due to the common use of cephalosporins as first-line agents for ETEC infection in both animals and humans, therefore resulting in the rising of resistance to ceftiofur (Winokur et al., 2001).

To date, >350 different ESBL variants have been reported and classified into nine distinct families, including TEM, SHV, CTX-M, PER, VEB, GES, BES, TLA, and OXA (Bush and Jacoby, 2010). In this study, only CTX-M, TEM, SHV, and OXA were detected. bla _CTX-M genes are usually plasmid-borne and coexist with other resistance genes, making them easily transfer between bacteria (D'Andrea et al., 2013; Seiffert et al., 2013). CTX-M has many variants, which are being frequently detected (Doi et al., 2012; Rao et al., 2014); however, in this study only one strain was found to carry the variant CTX-M-32. Previous studies showed that CTX-M ESBL producing E. coli were often coresistant to more classes of antibiotics such as trimethoprim–sulfamethoxazole, tetracycline, gentamicin, and ciprofloxacin than bacteria producing other types of ESBLs (Pitout et al., 2004; Valverde et al., 2004). But in this study, the isolate carrying bla _CTX-M-32 encoded only one tetracycline genes, tet(A). These results may suggest that the CTX-M ESBL E. coli has not yet become prevalent in pigs in the United States.

In addition, eight variants of bla _TEM and two variants of CMY-2 β-lactamases were detected in this study. Interestingly, this is the first report of bla _CMY-22 of CMY-2 β-lactamases in E. coli, which was initially reported in Salmonella enterica (Zioga et al., 2009). Altogether there were only seven ETEC isolates (5.3%) that carried ESBL genes, indicating ESBL genes may not be the main resistance mechanism of resistance to beta lactams in E. coli from swine. We did find a high number strains with bla _CMY and most of them had ampC. Bla _CMY-2 encodes an ampC-type beta-lactamase that hydrolyzes third-generation cephalosporins (Zhao et al., 2001). Bla _CMY-2, which probably originated from Citrobacter freundii (Wu et al., 1999; Philippon et al., 2002) has been disseminated horizontally within and between several bacterial species in Enterobacteriaceae, including important enteric pathogens (Winokur et al., 2000; Carattoli et al., 2002). Isolates harboring both ampC and bla _CMY genes were resistant to ceftiofur, whereas those isolates only encoding ampC but not bla _CMY were susceptible to ceftiofur (Supplementary Tables S2 and S3). The phenotypic and genotypic AMR profiles of E. coli in diarrheic piglets will guide the clinical use of antibiotics.

Enrofloxacin is an important fluoroquinolone (FQ) antibiotic to treat respiratory and enteric infections in pigs (Huang et al., 2000; Shaw, 2012; Ziółkowski et al., 2014; Pomorska-Mól et al., 2015). In recent years, studies have shown an increasing resistance to FQ in E. coli in several countries. In Spain, quinolone resistance among E. coli isolates has been found to increase from 9% to 17% in an over 5 years study (Garau et al., 1999). In China, the average enrofloxacin resistance between 2008 and 2015 has climbed up to 51.11% in swine E. coli isolates (Zhang et al., 2017). In this study, the enrofloxacin resistance between 2013 and 2014 in the United States was high at 58.2% in ETEC isolates. Noticeably, there was significant difference between ETEC isolates (58.2%) and non-ETEC isolates (5%). ETEC can produce enterotoxin and may cause more severe diseases to pigs, so the high enrofloxacin resistance in ETEC would be a significant threat to swine production.

Significant differences were also observed in genetic diversity between ETEC and non-ETEC strains. Molecular typing showed that ETEC isolates were relatively homogenous, with ST100 (52.7%) and ST10 (23.6%) as the most common STs. In contrast, the non-ETEC isolates were more genetically diverse than the ETEC isolates, with no predominant STs (Fig. 1 and Supplementary Table S3). ST100 isolates are the predominant ETEC type and important pig pathogens in the United States, Canada, Germany, and Thailand (http://mlst.warwick.ac.uk/mlst/dbs/Ecoli). ST10 spreads globally and colonizes humans, poultry, wild birds, and pigs (Okeke et al., 2010; Bergeron et al., 2012; Manges and Johnson, 2012; Poirel et al., 2012). A previous study showed that ST10 and ST23 were also the main STs of porcine ETEC isolates in the United States (Shepard et al., 2012). ST10 and ST23 strains commonly carry certain antibiotic resistance genes, such as ampC-type beta-lactamases, NDM-type carbapenemases, and other ESBLs (Oteo et al., 2009; Crémet et al., 2010; Mushtaq et al., 2011). In this study, no ST23 strains were detected in the ETEC isolates, but six isolates belonged to ST23 clonal complex and were resistant to at least five categories of antibiotic. All the ST23 E. coli isolates from this study were found in the non-ETEC collection. Interestingly, 66.7% of these non-ETEC ST23 strains were sensitive to all the tested antibiotics, which is different from the previous reports. The core-genome Single Nucleotide Polymorphism (SNP) phylogenetic tree showed the 95 E. coli isolates clustered with the phylogroups. Isolates in the same group were clustered closely in the whole genome tree (Fig. 1). Most of ETEC strains (n = 50) were in the phylogroup A. Five isolates in phylogenetic groups B2 and D were all non-ETEC strains. Similar to the MLST analysis, WGS phylogenetic analysis also showed that non-ETEC isolates were more diverse than the ETEC isolates.

Polymerase chain reaction and microarrays have been widely used to detect resistance gene and virulence factors (Boerlin et al., 2005; Nguyen et al., 2005). Nevertheless, these methods only detect particular genes and are unable to uncover new or rare resistance mechanisms. In recent years, considerable progress has been made in NGS technology. NGS has become a routine diagnostic and genotyping tool for disease surveillance, resistance prediction, and evolutionary analysis in infectious diseases (Javid and Török, 2016; Bossé et al., 2017; Ellington et al., 2017). In this study, NGS was applied to analyze 95 E. coli isolates of swine origin from 15 states in United States from 2013 to 2014. All the known resistance genes and mutations were identified directly from the genome sequences and we found the sequencing results were highly consistent with those of antimicrobial susceptibility test. A few cases were found to be inconsistent between their phenotypes and genotypes. The discrepancies may be due to the low level of expression, nonfunctional known resistance gene, or emergent novel resistance mechanisms in the strains compared with the resistance gene database. Phenotypic testing is still the current norm for determining antibiotic resistance, whereas WGS enables to investigate the prevalence of resistance genes and the genomic epidemiology of bacteria pathogens simultaneously in one analysis. Our study showed that NGS is a powerful tool for identification of genotypes and AMRs in E. coli.

In conclusion, we characterized the ST of 95 hemolytic E. coli from diarrheic piglets in the United States, and profiled their AMRs by both phenotypic test and genome sequencing. ETEC isolates were relatively homogenous, with ST100 and ST10 as the most common STs, but the non-ETEC isolates were more genetically diverse and did not have predominant STs. The isolates were 100% resistant to clindamycin, penicillin, tiamulin, and tilmicosin, and were also highly resistant to oxytetracycline, chlortetracycline, ampicillin, and sulfadimethoxine. 86.2% of them were MDR. Especially, ETEC isolates were highly resistant to enrofloxacin. This study not only provides new insights into antibiotic resistance and genotypes of intestinal pathogenic E. coli associated with swine disease in the United States, but also supports the utility of WGS in prediction of resistance to most antibiotics. The limitation of our study is the relatively small sample size across 15 states that may not really represent the pig population in the United States. Meanwhile, E. coli isolates may not be representative of E. coli found in the intestines of general swine population since these samples were from clinically ill and dead swine, and these swine may have been previously treated with antibiotics, which would greatly increase the chances of recovering resistant E. coli from the fecal samples. Therefore, further investigation is required to elucidate the correlations between the resistances and the demographic characteristics, such as the age, origin, and farm.

Footnotes

Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

References

Bankevich

, Nurk

, Antipov

, Gurevich

, Dvorkin

, Kulikov

, Lesin

, Nikolenko

, Pham

, Prjibelski

. SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol, 2012; 19:455–477.

Bergeron

, Prussing

, Boerlin

, Daignault

, Dutil

, Reid-Smith

, Zhanel

, Manges

. Chicken as reservoir for extraintestinal pathogenic Escherichia coli in humans, Canada. Emerg Infect Dis, 2012; 18:415.

Boerlin

, Travis

, Gyles

, Reid-Smith

, Lim

NJH

, Nicholson

, McEwen

, Friendship

, Archambault

. Antimicrobial resistance and virulence genes of Escherichia coli isolates from swine in Ontario. Appl Environ Microbiol, 2005; 71:6753–6761.

Bolger

, Lohse

, Usadel

. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics, 2014; 30:2114–2120.

Bossé

, Li

, Rogers

, Crespo

, Li

, Chaudhuri

, Holden

, Maskell

, Tucker

, Wren

. Whole genome sequencing for surveillance of antimicrobial resistance in Actinobacillus pleuropneumoniae . Front Microbiol, 2017; 8:311.

Burow

, Simoneit

, Tenhagen

, Kasbohrer

. Oral antimicrobials increase antimicrobial resistance in porcine E. coli—A systematic review. Prev Vet Med, 2014; 113:364–375.

Bush

, Jacoby

. Updated functional classification of β-lactamases. Antimicrob Agents Chemother, 2010; 54:969–976.

Carattoli

, Tosini

, Giles

, Rupp

, Hinrichs

, Angulo

, Barrett

, Fey

. Characterization of plasmids carrying CMY-2 from expanded-spectrum cephalosporin-resistant Salmonella strains isolated in the United States between 1996 and 1998. Antimicrob Agents Chemother, 2002; 46:1269–1272.

[CLSI] Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals: Approved Standard, 4th ed. CLSI document VET01-A4. Wayne, PA: CLSI, 2013.

10.

CLSI. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals: Approved Standard, 3rd ed. CLSI supplement VET01S. Wayne, PA: CLSI, 2015.

11.

Crémet

, Caroff

, Giraudeau

, Dauvergne

, Lepelletier

, Reynaud

, Corvec

. Occurrence of ST23 complex phylogroup A Escherichia coli isolates producing extended-spectrum AmpC β-lactamase in a French hospital. Antimicrob Agents Chemother, 2010; 54:2216–2218.

12.

D'Andrea

, Arena

, Pallecchi

, Rossolini

. CTX-M-type β-lactamases: A successful story of antibiotic resistance. Int J Med Microbiol, 2013; 303:305–317.

13.

Doi

, Park

, Rivera

, Adams-Haduch

, Hingwe

, Sordillo

, Lewis

2nd , Howard

, Johnson

, Polsky

. Community-associated extended-spectrum β-lactamase–producing Escherichia coli infection in the United States. Clin Infect Dis, 2012; 56:641–648.

14.

Ellington

, Ekelund

, Aarestrup

, Canton

, Doumith

, Giske

, Grundman

, Hasman

, Holden

, Hopkins

. The role of whole genome sequencing in antimicrobial susceptibility testing of bacteria: Report from the EUCAST Subcommittee. Clin Microbiol Infect, 2017; 23:2–22.

15.

Evans

Jr , Evans

. Escherichia coli in diarrheal disease. In Baron

(ed.): Medical Microbiology. Galveston TX: University of Texas Medical Branch at Galveston, 1996. pp. 337–345.

16.

Everett

, Jin

, Ricci

, Piddock

. Contributions of individual mechanisms to fluoroquinolone resistance in 36 Escherichia coli strains isolated from humans and animals. Antimicrob Agents Chemother, 1996; 40:2380–2386.

17.

Fairbrother

. Neonatal Escherichia coli Diarrhea. Diseases of Swine, 8th ed. Ames, IA: Iowa State University Press, 1999:433–441.

18.

Fairbrother

, Gyles

. Colibacillosis. In Zimmerman

, Karriker

, Ramirez

, Schwartz

, Stevenson

(eds.): Diseases of Swine, 10th Edition. Oxford: Wiley-Blackwell, 2012. pp. 723–749.

19.

Feuerman

, Miller

. Relationships between statistical measures of agreement: Sensitivity, specificity and kappa. J Eval Clin Pract, 2008; 14:930–933.

20.

Francis

. Colibacillosis in pigs and its diagnosis. Swine Health Prod, 1999; 7:241–244.

21.

Garau

, Xercavins

, Rodríguez-Carballeira

, Gómez-Vera

, Coll

, Vidal

, Llovet

, Ruíz-Bremón

. Emergence and dissemination of quinolone-resistant Escherichia coli in the community. Antimicrob Agents Chemother, 1999; 43:2736–2741.

22.

Gupta

, Padmanabhan

, Diene

, Lopez-Rojas

, Kempf

, Landraud

, Rolain

. ARG-ANNOT, a new bioinformatic tool to discover antibiotic resistance genes in bacterial genomes. Antimicrob Agents Chemother, 2014; 58:212–220.

23.

Hariharan

, Coles

, Poole

, Page

. Antibiotic resistance among enterotoxigenic Escherichia coli from piglets and calves from piglets and calves with diarrhea. Can Vet J, 2004; 45:605.

24.

Hooper

. Mechanisms of fluoroquinolone resistance. Drug Resist Updat, 1999; 2:38–55.

25.

Hopkins

, Day

, Threlfall

. Plasmid-mediated quinolone resistance in Salmonella enterica, United Kingdom. Emerg Infect Dis, 2008; 14:340.

26.

Huang

, Yu

, You

, Zhang

. Enrofloxacin transdermal absorben liniment in prevention and treatment of diarrhea caused by Escherichia coli in piglets. J Fujian Agric Univ, 2000; 29:498–501.

27.

Inouye

, Dashnow

, Raven

, Schultz

, Pope

, Tomita

, Zobel

, Holt

. SRST2: Rapid genomic surveillance for public health and hospital microbiology labs. Genome Med, 2014; 6:90.

28.

Janke

, Francis

, Collins

, Libal

, Zeman

, Johnson

. Attaching and effacing Escherichia coli infections in calves, pigs, lambs, and dogs. J Vet Diagn Invest, 1989; 1:6–11.

29.

Javid

, Török

. Whole-genome sequencing for the diagnosis of drug-resistant tuberculosis. Lancet Infect Dis, 2016; 16:17.

30.

Kaper

, Nataro

, Mobley

. Pathogenic Escherichia coli . Nat Rev Microbiol, 2004; 2:123.

31.

Letunic

, Bork

. Interactive tree of life (iTOL) v3: An online tool for the display and annotation of phylogenetic and other trees. Nucleic Acids Res, 2016; 44:W242–W245.

32.

Levine

. Adhesion of enterotoxigenic Escherichia coli in humans and animals. Ciba Found Symp, 1981:80:142–160.

33.

Luppi

. Swine enteric colibacillosis: Diagnosis, therapy and antimicrobial resistance. Porcine Health Manag, 2017; 3:16.

34.

Manges

, Johnson

. Food-borne origins of Escherichia coli causing extraintestinal infections. Clin Infect Dis, 2012; 55:712–719.

35.

Marshall

, Levy

. Food animals and antimicrobials: Impacts on human health. Clin Microbiol Rev, 2011; 24:718–733.

36.

Mushtaq

, Irfan

, Sarma

, Doumith

, Pike

, Pitout

, Livermore

, Woodford

. Phylogenetic diversity of Escherichia coli strains producing NDM-type carbapenemases. J Antimicrob Chemother, 2011; 66:2002–2005.

37.

Nataro

, Kaper

. Diarrheagenic Escherichia coli . Clin Microbiol Rev, 1998; 11:142–201.

38.

Nguyen

, Le Van

, Le Huy

, Gia

, Weintraub

. Detection and characterization of diarrheagenic Escherichia coli from young children in Hanoi, Vietnam. J Clin Microbiol, 2005; 43:755–760.

39.

Okeke

, Wallace-Gadsden

, Simons

, Matthews

, Labar

, Hwang

, Wain

. Multi-locus sequence typing of enteroaggregative Escherichia coli isolates from Nigerian children uncovers multiple lineages. PLoS One, 2010; 5:e14093.

40.

Oteo

, Diestra

, Juan

, Bautista

, Novais Â, Pérez-Vázquez

, Moyá

, Miró

, Coque

, Oliver

. Extended-spectrum β-lactamase-producing Escherichia coli in Spain belong to a large variety of multilocus sequence typing types, including ST10 complex/A, ST23 complex/A and ST131/B2. Int J Antimicrob Agents, 2009; 34:173–176.

41.

Philippon

, Arlet

, Jacoby

. Plasmid-determined AmpC-type β-lactamases. Antimicrob Agents Chemother, 2002; 46:1–11.

42.

Pitout

, Hanson

, Church

, Laupland

. Population-based laboratory surveillance for Escherichia coli-producing extended-spectrum β-lactamases: Importance of community isolates with bla CTX-M genes. Clin Infect Dis, 2004; 38:1736–1741.

43.

Pitout

, Laupland

. Extended-spectrum β-lactamase-producing Enterobacteriaceae: An emerging public-health concern. Lancet Infect Dis, 2008; 8:159–166.

44.

Poirel

, Potron

, De La Cuesta

, Cleary

, Nordmann

, Munoz-Price

. Wild coastline birds as reservoirs of broad-spectrum-β-lactamase-producing Enterobacteriaceae in Miami Beach, Florida. Antimicrob Agents Chemother, 2012; 56:2756–2758.

45.

Pomorska-Mól

, Czyżewska-Dors

, Kwit

, Rachubik

, Lipowski

, Pejsak

. Immune response in pigs treated with therapeutic doses of enrofloxacin at the time of vaccination against Aujeszky's disease. Res Vet Sci, 2015; 100:68–74.

46.

Rao

, Lv

, Zeng

, Chen

, He

, Chen

, Wu

, Wang

, Yang

, Wu

. Increasing prevalence of extended-spectrum cephalosporin-resistant Escherichia coli in food animals and the diversity of CTX-M genotypes during 2003–2012. Vet Microbiol, 2014; 172:534–541.

47.

Seiffert

, Hilty

, Perreten

, Endimiani

. Extended-spectrum cephalosporin-resistant Gram-negative organisms in livestock: An emerging problem for human health?. Drug Resist Updat, 2013; 16:22–45.

48.

Shaw

. The use of antibiotics to treat bacterial infections in dogs and cats. Companion Animal, 2012; 17:37–41.

49.

Shepard

, Danzeisen

, Isaacson

, Seemann

, Achtman

, Johnson

. Genome sequences and phylogenetic analysis of K88-and F18-positive porcine enterotoxigenic Escherichia coli . J Bacteriol, 2012; 194:395–405.

50.

Sorlozano

, Gutierrez

, Jimenez

, de Dios Luna

, Martínez

. Contribution of a new mutation in parE to quinolone resistance in extended-spectrum-β-lactamase-producing Escherichia coli isolates. J Clin Microbiol, 2007; 45:2740–2742.

51.

Tadesse

, Zhao

, Tong

, Ayers

, Singh

, Bartholomew

, McDermott

. Antimicrobial drug resistance in Escherichia coli from humans and food animals, United States, 1950–2002. Emerg Infect Dis, 2012; 18:741.

52.

Treangen

, Ondov

, Koren

, Phillippy

. The Harvest suite for rapid core-genome alignment and visualization of thousands of intraspecific microbial genomes. Genome Biol, 2014; 15:524.

53.

Valverde

, Coque

, Sánchez-Moreno

, Rollán

, Baquero

, Cantón

. Dramatic increase in prevalence of fecal carriage of extended-spectrum β-lactamase-producing Enterobacteriaceae during nonoutbreak situations in Spain. J Clin Microbiol, 2004; 42:4769–4775.

54.

Verraes

, Van Boxstael

, Van Meervenne

, Van Coillie

, Butaye

, Catry

, de Schaetzen

, Van Huffel

, Imberechts

, Dierick

, Daube

, Saegerman

, De Block

, Dewulf

, Herman

. Antimicrobial resistance in the food chain: A review. Int J Environ Res Public Health, 2013; 10:2643–2669.

55.

Vila

, Ruiz

, Goni

, De Anta

. Detection of mutations in parC in quinolone-resistant clinical isolates of Escherichia coli . Antimicrob Agents Chemother, 1996; 40:491–493.

56.

Winokur

, Brueggemann

, DeSalvo

, Hoffmann

, Apley

, Uhlenhopp

, Pfaller

, Doern

. Animal and human multidrug-resistant, cephalosporin-resistant Salmonella isolates expressing a plasmid-mediated CMY-2 AmpC β-lactamase. Antimicrob Agents Chemother, 2000; 44:2777–2783.

57.

Winokur

, Vonstein

, Hoffman

, Uhlenhopp

, Doern

. Evidence for transfer of CMY-2 AmpC β-lactamase plasmids between Escherichia coli and Salmonella isolates from food animals and humans. Antimicrob Agents Chemother, 2001; 45:2716–2722.

58.

, Dornbusch

, Kronvall

, Norgren

. Characterization and nucleotide sequence of a Klebsiella oxytoca cryptic plasmid encoding a CMY-type β-lactamase: Confirmation that the plasmid-mediated cephamycinase originated from the Citrobacter freundii AmpC β-lactamase. Antimicrob Agents and chemotherapy, 1999; 43:1350–1357.

59.

Yoshida

, Bogaki

, Nakamura

, Yamanaka

, Nakamura

. Quinolone resistance-determining region in the DNA gyrase gyrB gene of Escherichia coli . Antimicrob Agents Chemother, 1991; 35:1647–1650.

60.

Zhang

, Shen

, Zhang

, Song

, Wang

, Shang

, Yue

, Qu

, Li

, Wu

. Surveillance of antimicrobial resistance among Escherichia coli from chicken and swine, China, 2008–2015. Vet Microbiol, 2017; 203:49–55.

61.

Zhang

Progress and challenges in vaccine development against enterotoxigenic Escherichia coli (ETEC)-associated porcine post-weaning diarrhea (PWD). J Vet Med Res, 2014; 1(2), 1006.

62.

Zhao

, White

, McDermott

, Friedman

, English

, Ayers

, Meng

, Maurer

, Holland

, Walker

. Identification and expression of cephamycinasebla CMY genes in Escherichia coli and Salmonella isolates from food animals and ground meat. Antimicrob Agents Chemother, 2001; 45:3647–3650.

63.

Zioga

, Whichard

, Kotsakis

, Tzouvelekis

, Tzelepi

, Miriagou

. CMY-31 and CMY-36 cephalosporinases encoded by ColE1-like plasmids. Antimicrob Agents Chemother, 2009; 53:1256–1259.

64.

Ziółkowski

, Jaroszewski

, Maślanka

, Grabowski

, Katolik

, Pawęska

, Siemianowska

, Jasiecka

, Markiewicz

, Spodniewska

. Influence of oral co-administration of a preparation containing calcium and magnesium and food on enrofloxacin pharmacokinetics. Res Vet Sci, 2014; 97:99–104.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.04 MB

0.08 MB