Abstract

Objectives

To investigate the occurrence and characteristics of extended-spectrum β-lactamase (ESBL)- and AmpC-producing Enterobacteriaceae isolates in clinical samples of companion animals and horses and compare the results with ESBL/AmpC-producing isolates described in humans.

Methods

Between October 2007 and August 2009, 2700 Enterobacteriaceae derived from clinical infections in companion animals and horses were collected. Isolates displaying inhibition zones of ≤25 mm for ceftiofur and/or cefquinome by disc diffusion were included. ESBL/AmpC production was confirmed by combination disc tests. The presence of resistance genes was identified by microarray, PCR and sequencing, Escherichia coli genotypes by multilocus sequence typing and antimicrobial susceptibility by broth microdilution.

Results

Sixty-five isolates from dogs (n = 38), cats (n = 14), horses (n = 12) and a turtle were included. Six Enterobacteriaceae species were observed, mostly derived from urinary tract infections (n = 32). All except 10 isolates tested resistant to cefotaxime and ceftazidime by broth microdilution using clinical breakpoints. ESBL/AmpC genes observed were blaCTX-M-1, -2, -9, -14, -15,blaTEM-52, blaCMY-2 and blaCMY-39. blaCTX-M-1 was predominant (n = 17). blaCTX-M-9 occurred in combination with qnrA1 in 3 of the 11 Enterobacter cloacae isolates. Twenty-eight different E. coli sequence types (STs) were found. E. coli carrying blaCTX-M-1 belonged to 13 STs of which 3 were previously described in Dutch poultry and patients.

Conclusions

This is the first study among a large collection of Dutch companion animals and horses characterizing ESBL/AmpC-producing isolates. A similarity in resistance genes and E. coli STs among these isolates and isolates from Dutch poultry and humans may suggest exchange of resistance between different reservoirs.

Introduction

Members of the family Enterobacteriaceae commonly express plasmid-encoded broad-spectrum β-lactamases (TEM-1, TEM-2 and SHV-1) that confer resistance to amino-penicillins and first-generation cephalosporins, but not to third- and fourth-generation cephalosporins. The introduction of third-generation cephalosporins in the 1980s was a milestone in antimicrobial chemotherapy and improved the treatment options in human and veterinary medicine.1 Unfortunately, resistance to extended-spectrum cephalosporins (ESCs) emerged a few years later. Resistance to ESCs in Enterobacteriaceae is most often related to the production of extended-spectrum β-lactamases (ESBLs) or AmpC β-lactamases. ESBLs confer resistance to amino-penicillins, cephalosporins and monobactams and are inhibited by clavulanic acid. AmpC β-lactamases have a broader spectrum of resistance, including the cephamycins, and are not inhibited by β-lactamase inhibitors. The production of ESBLs and AmpC β-lactamases is often plasmid mediated. Moreover, these plasmids frequently carry genes encoding resistance to other drug classes, such as fluoroquinolones, aminoglycosides, sulfa-derivatives and trimethoprim.2,3 Therefore treatment options for infections caused by ESBL- and/or AmpC-producing organisms are limited. Initially these organisms were associated with hospitals and institutional care in humans, but they are now increasingly found in the community and in food-producing animals, particularly poultry, suggesting an exchange of organisms or genes between the different reservoirs, or a different antibiotic use behaviour.4 In the Netherlands, ESCs are authorized for use in food-producing animals (ceftiofur, cefquinome) and companion animals (cefovecin). Resistance to ESCs has been studied in detail in Gram-negative bacteria isolated from humans and food-producing animals.5,6 However, data on ESBL- and AmpC β-lactamase-producing Enterobacteriaceae in companion animals and horses are limited.7 The objective of this study was to investigate the occurrence of ESBL/AmpC-producing organisms in clinical samples, to further characterize these isolates and to compare the results with ESBL- and AmpC-producing isolates described in humans.

Materials and methods

Bacterial isolates

The Veterinary Microbiological Diagnostic Center (VMDC) of Utrecht University investigates samples from all over the Netherlands. Per year approximately 10 000 samples are submitted for bacteriological analysis and about 48% of these samples originate from dogs, 18% from cats, 12% from horses, 10% from food-producing animals and 12% from other animals including reptiles, birds and mammals other than those mentioned. From October 2007 to August 2009, 10 755 isolates, derived from approximately 18 000 samples, of which 2700 were Enterobacteriaceae, were tested for susceptibility to a wide range of antimicrobial agents by disc diffusion using a semi-confluent inoculum of the bacteria on Iso-Sensitest agar (bioTRADING, Mijdrecht, The Netherlands) with Neo-Sensitabs (Rosco Diagnostica, Taastrup, Denmark). Species were identified by conventional biochemical methods. Salmonella serovar identification was performed using microtitre and slide agglutination methods (O- and H-group antigens) according to the latest version of the Kauffman–White scheme. All Enterobacteriaceae isolates that displayed an inhibition zone ≤25 mm for ceftiofur (dog/cat/turtle isolates) and/or cefquinome (horse isolates) were included in the study and stored at −80°C until further analysis.

Phenotypic ESBL/AmpC testing

The isolates were tested phenotypically for ESBL production by combination disc tests using cefotaxime and ceftazidime with and without clavulanic acid (Becton Dickinson) according to CLSI guidelines.8 In addition, a cefoxitin disc (30 μg, Becton Dickinson) was added to this test, to detect AmpC phenotypes. All isolates classified as intermediate or resistant using CLSI criteria (≤17 mm) to cefoxitin were suspected to be AmpC producers.

Escherichia coli genotyping

In order to compare the genotypes of the E. coli isolates found in this study with ESBL-producing E. coli in former human studies, all E. coli strains were genotyped by multilocus sequence typing (MLST) as described previously.9 Sequences were uploaded on the MLST web site (http://mlst.ucc.ie/).

Susceptibility testing

All isolates were tested for antimicrobial susceptibility by determining MICs using broth microdilution, according to the international standard ISO 20776-1:2006. MICs were determined for ampicillin (concentration range 0.5–32 mg/L), cefotaxime (0.06–4 mg/L), ceftazidime (0.25–16 mg/L), ciprofloxacin (0.008–4 mg/L), nalidixic acid (4–64 mg/L), colistin (2–4 mg/L), gentamicin (0.25–32 mg/L), kanamycin (4–128 mg/L), sulfamethoxazole (8–1024 mg/L), trimethoprim (0.5–32 mg/L), streptomycin (2–128 mg/L), tetracycline (1–64 mg/L), chloramphenicol (2–64 mg/L) and florfenicol (2–64 mg/L) using E. coli ATCC 25922 and Enterococcus faecalis ATCC 29212 as control isolates. Multidrug resistance was defined as resistance to three or more antimicrobial agents included in the following list of eight antibiotics: cefotaxime (R >2 mg/L), ciprofloxacin (R >1 mg/L), colistin (R >2 mg/L), gentamicin (R >4 mg/L), sulfamethoxazole (R >256 mg/L), trimethoprim (R >4 mg/L), tetracycline (R >8 mg/L) and chloramphenicol (R >8 mg/L). Isolates were classified as resistant based on EUCAST clinical breakpoints (www.eucast.org) and, if not available (for sulfamethoxazole and tetracycline), CLSI clinical breakpoints.8Serratia marcescens is intrinsically resistant to colistin and Proteus mirabilis to both colistin and tetracycline.10 These bacterial species were not included to calculate multiresistance.

β-Lactamase identification

All isolates were screened for blaSHV, blaLEN, blaTEM, blaOXA, blaCTX-M, blaDHA, blaACC, blaMOX, blaFOX or blaCMY gene families by miniaturized microarray (Identibac, AMR-ve 05 genotyping, Alere International, Tilburg, The Netherlands).11 The gene families that responded positively in the array were further typed by PCR and sequencing using primers displayed in Table 1. PCR consisted of 30 cycles (30 s of denaturation at 94°C, 30 s of annealing at temperatures mentioned in Table 1 and 60 s of extension at 72°C) after one step of 5 min at 94°C. Amplicons were purified and sequenced as described previously.5 All E. coli isolates with AmpC phenotypes that could not be attributed to the presence of a plasmid-mediated AmpC gene were tested for mutations in the ampC promoter/attenuator region as described previously.12 PCRs to detect blaSHV, blaCTX-M or blaTEM genes were performed in isolates with an ESBL phenotype in which the array was negative for probes encoding ESBL gene families (Table 1).

Table 1.

Primers and PCR conditions used in this study to detect ESBL, plasmid-mediated AmpC, chromosomal ampC and qnr genes

Target Primer Annealing temperature (°C) Sequence (5′-3′) Product size (bp) Reference 
CTX-M families CTX-M-F 55 ATG TGC AGY ACC AGT AAR GTK ATG GC 592 5 
 CTX-M-R  TGG GTR AAR TAR GTS ACC AGA AYS AGC GG   
CTX-M-1 CTX-1-SEQ-F 60 CCC ATG GTT AAA AAA TCA CTG C >1000 28 
 CTX-1-SEQ-R  CAG CGC TTT TGC CGT CTA AG   
CTX-M-2 CTX-M-2F 55 ATG ATG ACT CAG AGC ATT CG 886 5 
 CTX-M-2R  TTA TTG CAT CAG AAA CCG TG   
CTX-M-9 CTX-M-9-1F 55 TGG TGA CAA AGA GAG TGC AAC G 875 20 
 CTX-M-4R  TCA CAG GCC TTC GGC GAT   
CTX-M-14/17 CTX-M-979255 CTA TTT TAC CCA GCC GCA AC 238 this study 
 CTX-M-91029 GTT ATG GAG CCA CGG TTG AT   
SHV SHV-F 55 TTA TCT CCC TGT TAG CCA CC 796 5 
 SHV-R  GAT TTG CTG ATT TCG CTC GG   
TEM TEM-F 55 GCG GAA CCC CTA TTT G 964 5 
 TEM-R  ACC ATT GCT TAA TCA GTG AG   
CMY CMY-F 58 ATG ATG AAA AAA TCG TTA TGC TGC 1138 5 
 CMY-R  GCT TTT CAA GAA TGC GCC AGG   
Chromosomal ampC AmpC1-71 55 AAT GGG TTT TCT ACG GTC TG 191 12 
 AmpC2120  GGG CAG CAA ATG TGG AGC AA   
OXA-1 OXA-1-F 55 ATG AAA AAC ACA ATA CAT ATC AAC TTC GC 820 29 
 OXA-1-R  GTG TGT TTA GAA TGG TGA TCG CAT T   
OXA-2 OXA-2-F 55 ACG ATA GTT GTG GCA GAC GAA C 601 29 
 OXA-2-R  ATY CTG TTT GGC GTA TCR ATA TTC   
qnrA qnrA-F 53 ATT TCT CAC GCC AGG ATT TG 516 13 
 qnrA-F  GAT CGG CAA AGG TTA GGT CA   
qnrB qnrB-F 53 GAT CGT GAA AGC CAG AAA GG 469 13 
 qnrB-R  ACG ATG CCT GGT AGT TGT CC   
qnrS qnrS-F 53 ACG ACA TTC GTC AAC TGC AA 417 13 
 qnrS-R  TAA ATT GGC ACC CTG TAG GC   
Target Primer Annealing temperature (°C) Sequence (5′-3′) Product size (bp) Reference 
CTX-M families CTX-M-F 55 ATG TGC AGY ACC AGT AAR GTK ATG GC 592 5 
 CTX-M-R  TGG GTR AAR TAR GTS ACC AGA AYS AGC GG   
CTX-M-1 CTX-1-SEQ-F 60 CCC ATG GTT AAA AAA TCA CTG C >1000 28 
 CTX-1-SEQ-R  CAG CGC TTT TGC CGT CTA AG   
CTX-M-2 CTX-M-2F 55 ATG ATG ACT CAG AGC ATT CG 886 5 
 CTX-M-2R  TTA TTG CAT CAG AAA CCG TG   
CTX-M-9 CTX-M-9-1F 55 TGG TGA CAA AGA GAG TGC AAC G 875 20 
 CTX-M-4R  TCA CAG GCC TTC GGC GAT   
CTX-M-14/17 CTX-M-979255 CTA TTT TAC CCA GCC GCA AC 238 this study 
 CTX-M-91029 GTT ATG GAG CCA CGG TTG AT   
SHV SHV-F 55 TTA TCT CCC TGT TAG CCA CC 796 5 
 SHV-R  GAT TTG CTG ATT TCG CTC GG   
TEM TEM-F 55 GCG GAA CCC CTA TTT G 964 5 
 TEM-R  ACC ATT GCT TAA TCA GTG AG   
CMY CMY-F 58 ATG ATG AAA AAA TCG TTA TGC TGC 1138 5 
 CMY-R  GCT TTT CAA GAA TGC GCC AGG   
Chromosomal ampC AmpC1-71 55 AAT GGG TTT TCT ACG GTC TG 191 12 
 AmpC2120  GGG CAG CAA ATG TGG AGC AA   
OXA-1 OXA-1-F 55 ATG AAA AAC ACA ATA CAT ATC AAC TTC GC 820 29 
 OXA-1-R  GTG TGT TTA GAA TGG TGA TCG CAT T   
OXA-2 OXA-2-F 55 ACG ATA GTT GTG GCA GAC GAA C 601 29 
 OXA-2-R  ATY CTG TTT GGC GTA TCR ATA TTC   
qnrA qnrA-F 53 ATT TCT CAC GCC AGG ATT TG 516 13 
 qnrA-F  GAT CGG CAA AGG TTA GGT CA   
qnrB qnrB-F 53 GAT CGT GAA AGC CAG AAA GG 469 13 
 qnrB-R  ACG ATG CCT GGT AGT TGT CC   
qnrS qnrS-F 53 ACG ACA TTC GTC AAC TGC AA 417 13 
 qnrS-R  TAA ATT GGC ACC CTG TAG GC   

Additional resistance gene identification

Antibiotic resistance genes were detected using the same microarray that was used for the screening of β-lactamase genes. This array included probes for 40 non-β-lactam resistance gene families known to occur in Gram-negative bacteria.11 Isolates with positive signals for qnr genes, which encode for plasmid-mediated quinolone resistance genes, were further analysed by PCR and sequencing using primers for qnrA, qnrB and qnrS as displayed in Table 1 and as described previously.13

Results

During the study period (October 2007–August 2009), 2% of all Enterobacteriaceae (n = 65) displayed inhibition zones ≤25 mm for ceftiofur or cefquinome. Between January 2008 and August 2009, this was found for 3% of the Enterobacteriaceae derived from dogs, 4% from cats and 8% from horses. For 2007 these data were not available. In 2007 a total of eight isolates matched the inclusion criteria. In 2008, 28 isolates were included, and in 2009, 29 isolates. All 65 isolates originated from clinical infections from dogs (n = 38), cats (n = 14), horses (n = 12) and a turtle. Except for two horses, all animals belonged to different owners and were submitted by 37 veterinary clinics located in 11 provinces in the Netherlands. One isolate was cultured from a dog that lived in an animal shelter in Germany. Information about antibiotic treatment was not available for all animals, but given the fact that most veterinary practices send in their samples only when initial antibiotic treatment fails, the isolates obtained are considered to be derived from animals that have received one or more antibiotic treatments. The isolates were cultured from urine (n = 32), wound (n = 15), peritoneal fluid (n = 5), uterus (n = 5), trachea (n = 3), blood (n = 2), faeces (n = 1), bile (n = 1) and a throat sample (n = 1). Most isolates were E. coli (n = 48). Other isolates were identified as Enterobacter cloacae (n = 11), P. mirabilis (n = 3), Salmonella enterica subspecies enterica 4,[5],12:b:− (n = 1), S. marcescens (n = 1) and Citrobacter freundii (n = 1).

Combination disc tests resulted in 29 isolates displaying an ESBL phenotype, 22 an AmpC phenotype, 6 a mixed ESBL/AmpC phenotype (displaying synergy with clavulanic acid and resistance to cefoxitin) and 2 displayed inconclusive results in the confirmation test (resistant to cefotaxime and/or ceftazidime, but no synergy with clavulanic acid and susceptible to cefoxitin). This was designated to be an inhibitor-resistant ESBL (IRE) type. In the remaining six isolates (four from dogs and two from cats), although with inhibition zones of ≤25 mm for ceftiofur, the ESBL phenotypic test was negative. These six isolates had MICs between ≤0.06 and 0.5 mg/L for cefotaxime and between ≤0.25 and 0.5 mg/L for ceftazidime, and molecular analysis of these isolates did not result in the detection of an ESBL and/or AmpC gene. The isolates that showed either an ESBL, ESBL/AmpC or IRE type were all resistant to cefotaxime (MIC >2 mg/L) and/or ceftazidime (MIC >4 mg/L). Four isolates displaying an AmpC phenotype had cefotaxime and ceftazidime MIC values below the clinical breakpoints. In sum, of 2700 isolates screened, 55 isolates (2.0%) were resistant to either cefotaxime or ceftazidime.

In the 29 isolates with an ESBL phenotype the following ESBL genes were found: blaCTX-M-1 (n = 17), blaCTX-M-15 (n = 4), blaCTX-M-14 (n = 1), blaCTX-M-2 (n = 2), blaTEM-52 (n = 3) and a combination of blaCTX-M-1 and blaCTX-M-14 (n = 1). In one isolate no ESBL gene could be detected (Table 2). No blaSHV or blaOXA genes encoding for ESBL production were detected.

Table 2.

Characteristics of Enterobacteriaceae isolates displaying inhibition zones of ≤25 mm for ceftiofur or cefquinome from clinical infections in companion animals and horses

Phenotype bla and qnr genes Number of isolates Species E. coli MLST Year of isolation Source Material 
ESBL (n = 29) CTX-M-1 E. coli 117, 162, 141, 770, 2030, 2226, 461 2008 (n = 5), 2009 (n = 2) dog (n = 2), cat (n = 2), horse (n = 3) peritoneal fluid (n = 2), urine (n = 2), wound (n = 1), uterus (n = 2) 
 CTX-M-1, TEM-1 E. coli 88, 162 (n = 2), 457, 362, 34 (n = 3), 58 2008 (n = 3), 2009 (n = 6) dog (n = 4), horse (n = 5) urine (n = 4), blood (n = 1), uterus (n = 3), trachea (n = 1) 
 CTX-M-1, CTX-M-14, TEM-1 E. coli 1287 2008 horse wound 
 CTX-M-1, TEM-80 E. coli 461 2009 dog wound 
 CTX-M-14 E. coli 1287 2008 horse wound 
 CTX-M-15 E. coli 648 2008 dog throat 
 CTX-M-15, TEM-1 E. coli 88, 156 2008 (n = 2) dog (n = 2) urine (n = 2) 
 CTX-M-15, TEM-1, OXA-1 E. coli 131 2008 dog urine 
 CTX-M-2, TEM-1 E. coli 156 (n = 2) 2007, 2008 cat, horse urine, wound 
 TEM-52 E. coli (n = 2), S. enterica (n = 1) 58, 93 2007 (n = 1), 2009 (n = 2) dog (n = 2), cat (n = 1) urine (n = 2), faeces (n = 1) 
 TEM-1, OXA-1 E. coli 117 2007 dog peritoneal fluid 
AmpC (n = 22) CMY-2 E. coli (n = 3), P. mirabilis (n = 3) 297, 372, 2227 2008 (n = 3), 2009 (n = 3) dog (n = 5), cat (n = 1) urine (n = 6) 
 CMY-2, OXA-1 E. coli 88 2007 dog urine 
 CMY-2, SHV-1 E. coli 12 2008 cat wound 
 CMY-2, TEM-1 E. coli 68 (n = 2) 2008, 2009 dog (n = 2) urine (n = 1), wound (n = 1) 
 CMY-39, qnrB17vara C. freundii NA 2007 turtle trachea 
 chromosomal ampC mutationsb E. coli 372, 58, 88 2008 (n = 1), 2009 (n = 2) dog (n = 2), cat (n = 1) peritoneal fluid (n = 1), bile (n = 1), urine (n = 1) 
 chromosomal ampC type 3c, TEM-1 E. coli 539 2009 dog urine 
 TEM-1 E. cloacae NA 2009 horse wound 
 TEM-1, qnrS1/S3 E. cloacae NA 2008 cat (n = 2) wound (n = 2) 
 none E. cloacae NA 2008 (n = 1), 2009 (n = 3) dog (n = 2), cat (n = 2) urine (n = 4) 
ESBL/AmpC (n = 6) CTX-M-9 E. cloacae NA 2008 cat wound 
 CTX-M-9, qnrA1 E. cloacae NA 2008 (n = 1), 2009 (n = 2) dog (n = 3) blood, trachea, urine 
 CTX-M-15, CMY-2, TEM-1 E. coli 648 2009 dog wound 
 CTX-M-14, TEM-1, chromosomal ampC type 4c E. coli 405 2007 dog peritoneal fluid 
IRE (n = 2) TEM-30 E. coli 88 2007 dog wound 
 none S. marcescens NA 2007 dog wound 
Susceptible (n = 6) OXA-1 E. coli 88 2009 dog urine 
 TEM-1 E. coli 448, 950, 1642 2008 (n = 1), 2009 (n = 2) dog (n = 3) urine (n = 3) 
 TEM-1, OXA-1 E. coli 1642 2009 cat wound 
 none E. coli 2225 2008 cat urine 
Phenotype bla and qnr genes Number of isolates Species E. coli MLST Year of isolation Source Material 
ESBL (n = 29) CTX-M-1 E. coli 117, 162, 141, 770, 2030, 2226, 461 2008 (n = 5), 2009 (n = 2) dog (n = 2), cat (n = 2), horse (n = 3) peritoneal fluid (n = 2), urine (n = 2), wound (n = 1), uterus (n = 2) 
 CTX-M-1, TEM-1 E. coli 88, 162 (n = 2), 457, 362, 34 (n = 3), 58 2008 (n = 3), 2009 (n = 6) dog (n = 4), horse (n = 5) urine (n = 4), blood (n = 1), uterus (n = 3), trachea (n = 1) 
 CTX-M-1, CTX-M-14, TEM-1 E. coli 1287 2008 horse wound 
 CTX-M-1, TEM-80 E. coli 461 2009 dog wound 
 CTX-M-14 E. coli 1287 2008 horse wound 
 CTX-M-15 E. coli 648 2008 dog throat 
 CTX-M-15, TEM-1 E. coli 88, 156 2008 (n = 2) dog (n = 2) urine (n = 2) 
 CTX-M-15, TEM-1, OXA-1 E. coli 131 2008 dog urine 
 CTX-M-2, TEM-1 E. coli 156 (n = 2) 2007, 2008 cat, horse urine, wound 
 TEM-52 E. coli (n = 2), S. enterica (n = 1) 58, 93 2007 (n = 1), 2009 (n = 2) dog (n = 2), cat (n = 1) urine (n = 2), faeces (n = 1) 
 TEM-1, OXA-1 E. coli 117 2007 dog peritoneal fluid 
AmpC (n = 22) CMY-2 E. coli (n = 3), P. mirabilis (n = 3) 297, 372, 2227 2008 (n = 3), 2009 (n = 3) dog (n = 5), cat (n = 1) urine (n = 6) 
 CMY-2, OXA-1 E. coli 88 2007 dog urine 
 CMY-2, SHV-1 E. coli 12 2008 cat wound 
 CMY-2, TEM-1 E. coli 68 (n = 2) 2008, 2009 dog (n = 2) urine (n = 1), wound (n = 1) 
 CMY-39, qnrB17vara C. freundii NA 2007 turtle trachea 
 chromosomal ampC mutationsb E. coli 372, 58, 88 2008 (n = 1), 2009 (n = 2) dog (n = 2), cat (n = 1) peritoneal fluid (n = 1), bile (n = 1), urine (n = 1) 
 chromosomal ampC type 3c, TEM-1 E. coli 539 2009 dog urine 
 TEM-1 E. cloacae NA 2009 horse wound 
 TEM-1, qnrS1/S3 E. cloacae NA 2008 cat (n = 2) wound (n = 2) 
 none E. cloacae NA 2008 (n = 1), 2009 (n = 3) dog (n = 2), cat (n = 2) urine (n = 4) 
ESBL/AmpC (n = 6) CTX-M-9 E. cloacae NA 2008 cat wound 
 CTX-M-9, qnrA1 E. cloacae NA 2008 (n = 1), 2009 (n = 2) dog (n = 3) blood, trachea, urine 
 CTX-M-15, CMY-2, TEM-1 E. coli 648 2009 dog wound 
 CTX-M-14, TEM-1, chromosomal ampC type 4c E. coli 405 2007 dog peritoneal fluid 
IRE (n = 2) TEM-30 E. coli 88 2007 dog wound 
 none S. marcescens NA 2007 dog wound 
Susceptible (n = 6) OXA-1 E. coli 88 2009 dog urine 
 TEM-1 E. coli 448, 950, 1642 2008 (n = 1), 2009 (n = 2) dog (n = 3) urine (n = 3) 
 TEM-1, OXA-1 E. coli 1642 2009 cat wound 
 none E. coli 2225 2008 cat urine 

IRE, inhibitor-resistant ESBL; E. coli, Escherichia coli; S. enteric, Salmonella enterica subspecies enterica 4,[5],12:b:−; P. mirabilis, Proteus mirabilis; E. cloacae, Enterobacter cloacae; C. freundii, Citrobacter freundii; S. marcescens, Serratia marcescens; NA, not applicable.

aSequence is similar to qnrB17, but has one silent mutation at position 169 GCG → GCA.

bChromosomal ampC mutations type 2, 3 and 31 according to Mulvey et al.14

cChromosomal ampC types refer to mutations in the chromosomal ampC gene as described by Mulvey et al.14

Among the 22 isolates with an AmpC phenotype, 11 carried an ESBL or AmpC gene [blaCMY-2 (n = 10) and blaCMY-39 (n = 1)] and one of these isolates (a P. mirabilis with blaCMY-2) was clinically susceptible to cefotaxime and ceftazidime (Table S1, available as Supplementary data at JAC Online). Four isolates with an AmpC phenotype had mutations in the promoter region of the chromosomal ampC gene. These four isolates had mutations in the ampC gene promoter belonging to sequence types (STs) 2, 3 (n = 2) and 31 as described by Mulvey et al.14 Two of those isolates (one with type 2 and one with type 3) were clinically susceptible to cefotaxime and ceftazidime (Table S1). The remaining seven isolates displaying an AmpC phenotype all belonged to E. cloacae. In these isolates no plasmid-mediated AmpC gene was detected with the methods used (Tables 2 and S1). One isolate of this group had MICs of 1 mg/L and 0.5 mg/L for cefotaxime and ceftazidime, respectively.

Among the isolates showing a combined ESBL/AmpC phenotype, blaCTX-M-9 (n = 4, all E. cloacae), blaCTX-M-14 in combination with type 4 mutations in the ampC attenuator region14 (n = 1) and a combination of blaCTX-M-15 and blaCMY-2 were detected (n = 1) (Tables 2 and S1).

Among the two isolates with inconclusive results in the combination disc test (IRE), one carried blaTEM-30 and in the other isolate (S. marcescens) no ESBL or AmpC genes were observed (Table 2).

Most isolates were multidrug resistant: 74% of all 61 isolates included in the calculation (all isolates, except P. mirabilis and S. marcescens) were resistant to three or more antimicrobial classes, and 32% were resistant to more than five classes of antibiotics. Among the isolates included in this calculation, nine were clinically susceptible to cefotaxime and/or ceftazidime, six of these were resistant to three or more antibiotics and one of them was resistant to more than five antibiotics. Besides β-lactamase genes, the following gene families were observed, encoding resistance to trimethoprim [dfrA1 (23%), dfrA12 (5%), dfrA14 (3%), dfrA17 (25%), dfrA19 (2%), dfrV (6%)], sulphonamides [sul1(42%), sul2 (12%), sul3 (2%)], fluoroquinolones [qnr (detects qnrA genes, 6%), qnrB (2%), qnrS (3%)], tetracyclines [tetA (6%), tetB (25%), tetC (2%)], aminoglycosides [aadA1 (29%), aadA2 (17%), aadA4 (14%), aac(6′)-Ib (8%), strA (3%), strB (25%), ant2a (8%)], chloramphenicol [cmlA1 (3%), catA1 (25%), catB3 (2%)], florfenicol [floR (5%)] and macrolides [ermB (3%), ereA (3%), ereB (20%)]. In addition, integrase genes intI1 (29%) and intI2 (3%) were found (Table S1). PCR and sequencing of the seven isolates positive for a qnr gene in the array demonstrated the presence of three qnrA1 genes; all three were detected in E. cloacae isolates from dogs also carrying blaCTX-M-9. Two qnrS1/S3 genes (no distinction could be made between qnrS1 and qnrS3 with the primers used) were observed in E. cloacae carrying blaTEM-1 isolated from cats and one qnrB17 variant (qnrB17 with one silent mutation at position 169 GCG → GCA) was found in C. freundii carrying blaCMY-39 isolated from a turtle. In one isolate (ID 32, Table S1), although positive for the qnr probe in the array, no qnr PCR product could be amplified with the primers used in this study (Table 1). This strain was susceptible to ciprofloxacin (MIC 0.015 mg/L) as well as to nalidixic acid (MIC ≤4 mg/L) (Table S1), confirming the absence of a qnr gene.

MLST of the E. coli isolates resulted in 28 different STs (Table 2). The predominant STs were ST88 (n = 6), ST162 (n = 3), ST58 (n = 3), ST34 (n = 3) and ST156 (n = 3). Three new STs were identified: ST2225, ST2226 and ST2227. Except for the two isolates from the horses owned by the same person, which both belonged to ST34, no relation was found between STs and the area where the animals lived, or where the isolates were collected (data not shown).

Discussion

A prevalence of 2% ESC resistance in clinical isolates derived from companion animals is comparable to what is found in other countries.15,16 In this study we used ceftiofur or cefquinome to detect resistance to ESCs. Veterinary diagnostic laboratories often use cephalosporins that are commonly used in veterinary practice to perform susceptibility tests. However, these antibiotics have been shown to be less suitable as indicator cephalosporins for isolates with ESBL or plasmid-mediated AmpC β-lactamases.17 In our study, screening with these cephalosporins using non-standardized interpretive criteria resulted in 19 (29%) apparently false-positive isolates, in which no ESBL and/or plasmid-mediated AmpC genes could be detected by array or PCR and sequencing. Six of these (all E. coli) were negative in the phenotypic confirmation test and had MICs below clinical breakpoints for cefotaxime and ceftazidime. Four other isolates (two E. coli, one P. mirabilis and one E. cloacae), although displaying an AmpC phenotype, also had MICs below the clinical breakpoints for both ESCs. This resulted in a misclassification of 15% (10/65) of the isolates being resistant to ESCs using ≤25 mm as cut-off value for ceftiofur or cefquinome in the disc diffusion test with Neo-Sensitabs (Rosco). To have fewer false-positive results in the future both the method and the interpretative criteria should be critically analysed and possibly reconsidered.

Overall a variety of ESBL genes was found within six bacterial species isolated from four animal species. blaCTX-M-1 was the predominant ESC resistance gene found (26%). This gene was also the predominant ESBL gene among Enterobacteriaceae studied in Dutch horses in 2003–05.18 At that time, only blaCTX-M-1 was isolated. Our study shows that in later years blaCTX-M-2 and blaCTX-M-14 also occurred in Enterobacteriaceae from Dutch horses.

blaCTX-M-1 was recently described as the predominant ESBL gene in cefotaxime-resistant E. coli and S. enterica isolates (49%) of Dutch poultry5 and in poultry meat isolates (49%) derived from Dutch supermarkets.6 In Dutch patients, blaCTX-M-1 was not predominant in clinical infections (21%–24%),6,19 but it was the predominant gene (46%) found in ESC-resistant E. coli derived from rectal swabs of human patients in Dutch hospitals.19 Our results show that blaCTX-M-1 was present among others in E. coli ST117 (n = 2), ST58 (n = 3) and ST162 (n = 3). These E. coli genotypes are described in Dutch poultry (ST117 and ST58) and patients (ST117, ST58 and ST162) carrying blaCTX-M-1.6 This suggests a clonal spread of these E. coli genotypes carrying blaCTX-M-1 between different hosts.

A remarkable finding in our study was that all blaCTX-M-9 genes (n = 4) occurred in E. cloacae isolates, three derived from dogs and one from a cat. The canine isolates also harboured the plasmid-mediated quinolone resistance gene qnrA1. The combined presence of blaCTX-M-9 and qnrA1 has been described in outbreak strains of Enterobacter hormaechei (initially also classified as E. cloacae) in a Dutch hospital at Utrecht University Medical Centre carrying a conjugative plasmid pQC of the IncHI2 family with several complex integrons containing aadB, blaCTX-M-9 and qnrA1.20,21 The three canine isolates were not screened for the aadB gene, but had non-wild-type MICs of gentamicin, and two isolates contained a class 1 integron. To our knowledge this has not been reported previously in clinical isolates from dogs. The dogs and their owners originated from three different areas in the Netherlands and none of them originated from the area around Utrecht, which confirms the observed nationwide outbreak of qnrA1-positive multidrug-resistant E. cloacae in human patients.22 Moreover, in 2008 an incHI2 plasmid containing qnrA1 and blaCTX-M-9 was found in Salmonella Paratyphi B var. Java isolated from broilers.23,24 Together with our data this suggests that a plasmid with this combination of resistance genes is not only present in different areas within the Netherlands, but has also disseminated in different hosts, including humans, food-producing animals and companion animals.

Another interesting finding in our study was the presence of blaCTX-M-15 in different E. coli genotypes. Only one isolate derived from a dog carrying blaCTX-M-15 belonged to uropathogenic E. coli ST131. This ST has emerged globally in hospital and community settings and has been described frequently in humans25 and incidentally in isolates derived from dogs and horses in several European countries,25–27 emphasizing the intraspecies and pan-European spread of this resistant clone. Our data, however, confirm that blaCTX-M-15 in E. coli of dogs is not limited to E. coli ST131, and that other clones like ST88, ST156 and ST648 play a role in the dissemination of this resistance gene as well.

This is the first study providing information about the genes and genotypes involved in ESBL- or AmpC-producing isolates from Dutch companion animals and horses in a large group of isolates. The most prevalent gene found, blaCTX-M-1, was previously found in Dutch human and poultry isolates. This, together with the presence of the combination of blaCTX-M-9 and qnrA1 in E. cloacae isolates, which was previously described in human isolates, suggests transmission between the different reservoirs or the existence of a common source. The fact that companion animals often live in close contact with their owners makes the occurrence of transmission between them even more likely. This study shows that prudent usage of antibiotics in companion animals and horses should be emphasized and continued susceptibility surveillance to ESCs is recommended.

Funding

This work was supported by the Dutch Ministry of Economic Affairs, Agriculture and Innovation (grant number WOT-01-002-003).

Transparency declarations

None to declare.

Supplementary data

Table S1 is available as Supplementary data at JAC Online (http://www.jac.oxfordjournals.org/).

Acknowledgements

This work was partly presented in an oral presentation at the Scientific Spring Meeting of the Dutch Society for Medical Microbiology and the Dutch Society for Microbiology in Arnhem, The Netherlands, 2011 (abstract number 0194) and in a poster at the Fourth Symposium on Antimicrobial Resistance in Animals and the Environment (ARAE), Tours, France, 2011 (abstract number P55).

We would like to thank the EU Reference Laboratory for Antimicrobial Resistance at the Danish Technical University in Lyngby for providing the control strains that were used to perform the PCRs and the technicians working at the VMDC of Utrecht University for collecting the isolates that met the inclusion criteria.

References

1
Pfeifer
Y
Cullik
A
Witte
W
Resistance to cephalosporins and carbapenems in Gram-negative bacterial pathogens
Int J Med Microbiol
 , 
2010
, vol. 
300
 (pg. 
371
-
9
)
2
Hawkey
PM
Jones
AM
The changing epidemiology of resistance
J Antimicrob Chemother
 , 
2009
, vol. 
64
 
Suppl 1
(pg. 
i3
-
10
)
3
Pitout
JD
Laupland
KB
Extended-spectrum β-lactamase-producing Enterobacteriaceae: an emerging public-health concern
Lancet Infect Dis
 , 
2008
, vol. 
8
 (pg. 
159
-
66
)
4
Scientific Advisory Group on Antimicrobials of the Committee for Medicinal Products for Veterinary Use
Reflection paper on the use of third and fourth generation cephalosporins in food producing animals in the European Union: development of resistance and impact on human and animal health
J Vet Pharmacol Ther
 , 
2009
, vol. 
32
 (pg. 
515
-
33
)
5
Dierikx
C
van Essen-Zandbergen
A
Veldman
K
, et al.  . 
Increased detection of extended spectrum β-lactamase producing Salmonella enterica and Escherichia coli isolates from poultry
Vet Microbiol
 , 
2010
, vol. 
145
 (pg. 
273
-
8
)
6
Leverstein-van Hall
MA
Dierikx
CM
Cohen Stuart
J
, et al.  . 
Dutch patients, retail chicken meat and poultry share the same ESBL genes, plasmids and strains
Clin Microbiol Infect
 , 
2011
, vol. 
17
 (pg. 
873
-
80
)
7
Wieler
LH
Ewers
C
Guenther
S
, et al.  . 
Methicillin-resistant staphylococci (MRS) and extended-spectrum β-lactamases (ESBL)-producing Enterobacteriaceae in companion animals: nosocomial infections as one reason for the rising prevalence of these potential zoonotic pathogens in clinical samples
Int J Med Microbiol
 , 
2011
, vol. 
301
 (pg. 
635
-
41
)
8
Clinical and Laboratory Standards Institute
Performance Standards for Antimicrobial Susceptibility Testing: Twenty-first Informational Supplement M100-21
 , 
2011
Wayne, PA, USA
CLSI
9
Wirth
T
Falush
D
Lan
R
, et al.  . 
Sex and virulence in Escherichia coli: an evolutionary perspective
Mol Microbiol
 , 
2006
, vol. 
60
 (pg. 
1136
-
51
)
10
Magiorakos
AP
Srinivasan
A
Carey
RB
, et al.  . 
Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance
Clin Microbiol Infect
 , 
2011
11
Batchelor
M
Hopkins
KL
Liebana
E
, et al.  . 
Development of a miniaturised microarray-based assay for the rapid identification of antimicrobial resistance genes in Gram-negative bacteria
Int J Antimicrob Agents
 , 
2008
, vol. 
31
 (pg. 
440
-
51
)
12
Caroff
N
Espaze
E
Berard
I
, et al.  . 
Mutations in the ampC promoter of Escherichia coli isolates resistant to oxyiminocephalosporins without extended spectrum β-lactamase production
FEMS Microbiol Lett
 , 
1999
, vol. 
173
 (pg. 
459
-
65
)
13
Gay
K
Robicsek
A
Strahilevitz
J
, et al.  . 
Plasmid-mediated quinolone resistance in non-Typhi serotypes of Salmonella enterica
Clin Infect Dis
 , 
2006
, vol. 
43
 (pg. 
297
-
304
)
14
Mulvey
MR
Bryce
E
Boyd
DA
, et al.  . 
Molecular characterization of cefoxitin-resistant Escherichia coli from Canadian hospitals
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
358
-
65
)
15
SVARM 2009: Swedish Veterinary Antimicrobial Resistance Monitoring
 , 
2010
Uppsala, Sweden
National Veterinary Institute (SVA)
16
Costa
D
Poeta
P
Saenz
Y
, et al.  . 
Prevalence of antimicrobial resistance and resistance genes in faecal Escherichia coli isolates recovered from healthy pets
Vet Microbiol
 , 
2008
, vol. 
127
 (pg. 
97
-
105
)
17
Aarestrup
FM
Hasman
H
Veldman
K
, et al.  . 
Evaluation of eight different cephalosporins for detection of cephalosporin resistance in Salmonella enterica and Escherichia coli
Microb Drug Resist
 , 
2010
, vol. 
16
 (pg. 
253
-
61
)
18
Vo
AT
van Duijkeren
E
Fluit
AC
, et al.  . 
Characteristics of extended-spectrum cephalosporin-resistant Escherichia coli and Klebsiella pneumoniae isolates from horses
Vet Microbiol
 , 
2007
, vol. 
124
 (pg. 
248
-
55
)
19
Overdevest
I
Willemsen
I
Rijnsburger
M
, et al.  . 
Extended-spectrum β-lactamase genes of Escherichia coli in chicken meat and humans, The Netherlands
Emerg Infect Dis
 , 
2011
, vol. 
17
 (pg. 
1216
-
22
)
20
Paauw
A
Fluit
AC
Verhoef
J
, et al.  . 
Enterobacter cloacae outbreak and emergence of quinolone resistance gene in Dutch hospital
Emerg Infect Dis
 , 
2006
, vol. 
12
 (pg. 
807
-
12
)
21
Paauw
A
Caspers
MP
Leverstein-van Hall
MA
, et al.  . 
Identification of resistance and virulence factors in an epidemic Enterobacter hormaechei outbreak strain
Microbiology
 , 
2009
, vol. 
155
 (pg. 
1478
-
88
)
22
Leverstein-van Hall
MA
Blok
HE
Paauw
A
, et al.  . 
Extensive hospital-wide spread of a multidrug-resistant Enterobacter cloacae clone, with late detection due to a variable antibiogram and frequent patient transfer
J Clin Microbiol
 , 
2006
, vol. 
44
 (pg. 
518
-
24
)
23
MARAN 2008: Monitoring of Antimicrobial Resistance and Antibiotic Usage in Animals in the Netherlands in 2008
 , 
2010
Lelystad, The Netherlands
Central Veterinary Institute (CVI) of Wageningen UR
 
www.cvi.wur.nl/NL/publicaties/rapporten/maranrapportage (17 February 2012, date last accessed)
24
Veldman
K
Cavaco
LM
Mevius
D
, et al.  . 
International collaborative study on the occurrence of plasmid-mediated quinolone resistance in Salmonella enterica and Escherichia coli isolated from animals, humans, food and the environment in 13 European countries
J Antimicrob Chemother
 , 
2011
, vol. 
66
 (pg. 
1278
-
86
)
25
Pomba
C
da Fonseca
JD
Baptista
BC
, et al.  . 
Detection of the pandemic O25-ST131 human virulent Escherichia coli CTX-M-15-producing clone harboring the qnrB2 and aac(6′)-Ib-cr genes in a dog
Antimicrob Agents Chemother
 , 
2009
, vol. 
53
 (pg. 
327
-
8
)
26
Rogers
BA
Sidjabat
HE
Paterson
DL
Escherichia coli O25b-ST131: a pandemic, multiresistant, community-associated strain
J Antimicrob Chemother
 , 
2011
, vol. 
66
 (pg. 
1
-
14
)
27
Ewers
C
Grobbel
M
Stamm
I
, et al.  . 
Emergence of human pandemic O25:H4-ST131 CTX-M-15 extended-spectrum-β-lactamase-producing Escherichia coli among companion animals
J Antimicrob Chemother
 , 
2010
, vol. 
65
 (pg. 
651
-
60
)
28
Carattoli
A
Garcia-Fernandez
A
Varesi
P
, et al.  . 
Molecular epidemiology of Escherichia coli producing extended-spectrum β-lactamases isolated in Rome, Italy
J Clin Microbiol
 , 
2008
, vol. 
46
 (pg. 
103
-
8
)
29
Hasman
H
Mevius
D
Veldman
K
, et al.  . 
β-Lactamases among extended-spectrum β-lactamase (ESBL)-resistant Salmonella from poultry, poultry products and human patients in The Netherlands
J Antimicrob Chemother
 , 
2005
, vol. 
56
 (pg. 
115
-
21
)

Supplementary data