molecular characterization of resistance to extended-spectrum cephalosporins in clinical

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Dec. 2011, p. 5666–5675 Vol. 55, No. 120066-4804/11/$12.00 doi:10.1128/AAC.00656-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Molecular Characterization of Resistance to Extended-SpectrumCephalosporins in Clinical Escherichia coli Isolates from

Companion Animals in the United States�†Bashar W. Shaheen,1* Rajesh Nayak,1 Steven L. Foley,1 Ohgew Kweon,1 Joanna Deck,1

Miseon Park,1 Fatemeh Rafii,1 and Dawn M. Boothe2

Division of Microbiology, National Center for Toxicological Research, U.S. Food and Drug Administration, Jefferson, Arkansas,1

and Department of Anatomy, Physiology and Pharmacology, College of Veterinary Medicine, Auburn University, Auburn, Alabama2

Received 12 May 2011/Returned for modification 16 August 2011/Accepted 6 September 2011

Resistance to extended-spectrum cephalosporins (ESC) among members of the family Enterobacteriaceaeoccurs worldwide; however, little is known about ESC resistance in Escherichia coli strains from companionanimals. Clinical isolates of E. coli were collected from veterinary diagnostic laboratories throughout theUnited States from 2008 to 2009. E. coli isolates (n � 54) with reduced susceptibility to ceftazidime orcefotaxime (MIC > 16 �g/ml) and extended-spectrum-�-lactamase (ESBL) phenotypes were analyzed. PCRand sequencing were used to detect mutations in ESBL-encoding genes and the regulatory region of thechromosomal gene ampC. Conjugation experiments and plasmid identification were conducted to examine thetransferability of resistance to ESCs. All isolates carried the blaCTX-M-1-group �-lactamase genes in additionto one or more of the following �-lactamase genes: blaTEM, blaSHV-3, blaCMY-2, blaCTX-M-14-like, and blaOXA-1.Different blaTEM sequence variants were detected in some isolates (n � 40). Three isolates harbored ablaTEM-181 gene with a novel mutation resulting in an Ala184Val substitution. Approximately 78% of theisolates had mutations in promoter/attenuator regions of the chromosomal gene ampC, one of which was anovel insertion of adenine between bases �28 and �29. Plasmids ranging in size from 11 to 233 kbp weredetected in the isolates, with a common plasmid size of 93 kbp identified in 60% of isolates. Plasmid-mediatedtransfer of �-lactamase genes increased the MICs (>16-fold) of ESCs for transconjugants. Replicon typingamong isolates revealed the predominance of IncI and IncFIA plasmids, followed by IncFIB plasmids. Thisstudy shows the emergence of conjugative plasmid-borne ESBLs among E. coli strains from companion animalsin the United States, which may compromise the effective therapeutic use of ESCs in veterinary medicine.

Escherichia coli is one of the primary causes of urinary tractinfections and pyometra, particularly in dogs (11, 23). Drug-resistant E. coli strains have been associated with urinary tractand other nosocomial infections in veterinary hospitals (14).�-Lactams are widely used in veterinary medicine to treatinfections caused by E. coli in dogs and cats (26). This wide-spread use may have contributed to the substantial increase inthe emergence of E. coli resistance to �-lactams in companionanimals over the last several decades (26, 42, 43). The emer-gence of �-lactam-resistant bacteria in companion animals andthe transfer of resistant isolates to humans poses a potentialserious risk to public health (22). Of particular concern arebacteria from these animals with resistance to extended-spec-trum cephalosporins (ESCs), such as ceftiofur and ceftriaxone,which are critical in veterinary and human medicine for treat-ment of bacterial infections (2).

Resistance to ESCs in E. coli is mediated primarily by pro-duction of class A extended-spectrum �-lactamases (ESBLs),which hydrolyze oxyimino-cephalosporins but are not activeagainst cephamycins and carbapenems and can be inhibited by

�-lactamase inhibitors (e.g., clavulanic acid) (28). Most ESBLsand �-lactamase inhibitor-resistant �-lactamases are derivedfrom the classical TEM-1, TEM-2, and SHV-1 enzymes as aresult of amino acid substitutions in their sequences (2). Theplasmid-encoded CTX-M family members, which confer highlevels of resistance to ESCs, have been reported to be foundworldwide in E. coli strains isolated from human and otheranimal sources (1, 3, 4, 9, 26). CTX-M �-lactamases exhibitgreater activity against cefotaxime than ceftazidime (1). ClassC �-lactamase and plasmid-encoded AmpC-like CMY �-lac-tamases are also widely distributed in ESC-resistant E. colistrains of animal origin (27, 48, 50). Unlike CTX-M enzymes,CMY enzymes exhibit activity against cephamycins, such ascefoxitin, in addition to oxyimino-cephalosporins and cannotbe inactivated by clavulanate (2, 26). Resistance to ESCs alsois associated with hyperproduction of chromosomal AmpC�-lactamases by either gene amplification or mutations in thepromoter or the attenuator region of the chromosomal ampCgene (3, 4, 35). Other possible mechanisms are the class DOXA ESBLs (2, 28) and changes in membrane permeability(31).

ESBLs in clinical strains of E. coli isolated from humans andfood-producing animals have been characterized in variousstudies (3, 4, 20, 30, 33, 35, 47), but few studies have beenperformed on E. coli isolates from dogs or cats (26). E. colistrains from companion animals producing CTX-M-1, CTX-M-15, CMY-2, and SHV-12 have been found in Europe (7,

* Corresponding author. Mailing address: Division of Microbiology,National Center for Toxicological Research, U.S. Food and DrugAdministration, Jefferson, AR 72079. Phone: (870) 543-7599. Fax:(870) 543-7307. E-mail: [email protected].

† Supplemental material for this article may be found at http://aac.asm.org/.

� Published ahead of print on 26 September 2011.

5666

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

15–17), whereas strains with CTX-M-14 were reported in Chile(34). Recently, CTX-M-14, CTX-M-15, and SHV-12 ESBLshave been identified in canine and feline E. coli isolates asso-ciated with urinary tract infections (UTIs) in the United States(37).

Knowledge about the diversity and prevalence of resistanceto broad-spectrum �-lactams among Enterobacteriaceae fromcompanion animals in the United States remains limited. Inthis study, we evaluated the prevalence of resistance to thesedrugs among canine and feline isolates of E. coli from veteri-nary clinics in the United States and characterized the mech-anisms of resistance to ESCs by class A and C �-lactamases.We have also determined the resistance phenotypes, the ge-netic relationships among isolates, and the role of plasmids inresistance development and horizontal gene transfer.

MATERIALS AND METHODS

Bacterial isolates and antimicrobial susceptibility testing. Canine and felineclinical isolates of E. coli were acquired from clinical veterinary microbiologylaboratories in California, Illinois, Massachusetts, North Carolina, and Ohio(IDEXX Reference Laboratories, Inc., Westbrook, ME) between May 2008 andMay 2009 (n � 944). Isolates were initially cultured from samples received fromveterinary practitioners after collection from dogs or cats with presumed infec-tions. These isolates were isolated from urine, wound, ear, genital tract, anal sac,nasal structure, and soft tissue samples. Bacterial isolates were shipped overnightto the Clinical Pharmacology Laboratory at Auburn University as pure cultureson Trypticase soy agar slants. Upon arrival of the isolates, their identity wasconfirmed by plating them on MacConkey and CHROMagar agar plates and byindole testing using Kovacs reagent (Remel/Thermo Fisher Scientific, Lenexa,KS). The isolates were grown on Trypticase soy agar (TSA; Difco, Detroit, MI)and stored in Trypticase soy broth (Difco) containing 25% glycerol at �80°C.Isolates were classified into four geographical regions based on origin: South(North Carolina), West (California), Midwest (Ohio and Illinois), and Northeast(Massachusetts).

Antimicrobial susceptibility testing was performed for all isolates (n � 944)using custom microdilution susceptibility plates according to the manufacturer’sprotocol (Trek Diagnostic Systems, Inc., Cleveland, OH). A panel of 15 antimi-crobial agents was tested (see Table S1 in the supplemental material). The MICswere recorded using the Sensititre Vizion system (Trek Diagnostic Systems). E.coli ATCC 25922 was used for quality control. The results were interpreted byCLSI guidelines (13). Multidrug resistance (MDR) was defined as resistance totwo or more drug classes (the class of �-lactams includes three subclasses:penicillins, cephalosporins [i.e., narrow, expanded, and broad spectrum], andcarbapenems). Isolates that exhibited at least an 8-fold increase in their MICs(per CLSI guidelines) of either ceftazidime or cefotaxime alone versus the MICof the same drug tested in combination with clavulanic acid were identified aspotential ESBL producers (13).

The susceptibilities of the isolates to ceftazidime or cefotaxime were deter-mined, and a total of 54 isolates, which expressed the ESBL phenotype (n � 29)per CLSI guidelines or for which the MIC of ceftazidime or cefotaxime was �16�g/ml (n � 25), were selected as presumptive ESBL producer strains for thisstudy.

PFGE. All 54 isolates were analyzed for genetic relatedness by pulsed-field gelelectrophoresis (PFGE) using the XbaI restriction enzyme according to the CDCPulseNet protocol (http://www.cdc.gov/pulsenet/protocols.htm). The DNA fin-gerprints were analyzed with BioNumerics software (version 6.0; Applied Maths,Austin, TX). Matching and dendrogram analysis of the PFGE patterns wereconducted using unweighted pair group method with arithmetic averages (UP-GMA) analysis and the Dice correlation coefficient. The Dice band-based sim-ilarity coefficient, with a band position tolerance of 1% and an optimization of1.5%, was used for clustering.

PCR analysis of genes conferring resistance to ESCs. The �-lactamase genesblaTEM, blaCMY, blaSHV, blaOXA, and blaCTX-M were detected by PCR, usingspecific primers and conditions previously described (see Table S2 in the sup-plemental material). Bacterial DNA was extracted by the boiling method (44)and used as the template for PCR. Appropriate positive (E. coli CVM 15103[blaTEM], CVM 1290 [blaCMY], CVM 15104 [blaSHV], CVM 40118 [blaOXA], andCVM 35778 [blaCTX-M]) and negative controls were included in each PCR.Amplicons from CTX-M genes (group 1 and group 14) were sequenced using the

appropriate primers (see Table S2) and dideoxy terminators in an automatedDNA sequencer (ABI 310; Applied Biosystems, Foster City, CA). The sequenceswere analyzed using a BLAST search (http://www.ncbi.nlm.nih.gov/guide/dna-rna/) and compared with previously reported sequences for �-lactamases (http://www.lahey.org/Studies). Sequence alignments were performed using VectorNTI Advance 11 software (Invitrogen, Carlsbad, CA).

In silico structural analysis. The crystal structures of wild-type TEM-1 (PDBID, 1ERO) and mutant (A184V; PDB ID, 1JTD) were used for computationalanalysis in this study. The databases RCSB PDB (http://www.rcsb.org/pdb/home/home.do) and PDBsum (http://www.ebi.ac.uk/pdbsum/) provided the experi-mentally determined PDB structure files and their at-a-glance structural over-view. The three-dimensional homology model for wild-type protein (Ala-184)was generated using SWISS-MODEL (http://swissmodel.expasy.org/workspace/index.php) with the PDB 1JTD (A184V) as the template. SuperPose (version1.0) was used for structural superposition and RMSD (root mean square devi-ation) calculation of the wild-type and mutant proteins. To predict changes inprotein stability upon the point mutation, CUPSAT (http://cupsat.tu-bs.de/) wasadopted. PyMOL (0.99RC6) (http://www.pymol.org) was used for visualizationof structure figures.

Sequencing of the regulatory region of the chromosomal ampC gene. PCRamplification and sequencing of a 191-bp fragment of the promoter-and-atten-uator region of the chromosomal ampC gene were performed using the primersampC-F (5�-AATGGGTTTTCTACGGTCTG-3�) and ampC-R (5�-GGGCAGCAAATGTGGAGCAA-3�) (5). Multisequence alignment was performed tocompare the sequences and detect mutations in the promoters and attenuatorregions with sequences of the same region in the E. coli K-12 ampC gene, usingVector NTI Advance 11 software (Invitrogen).

Plasmid characterization: S1 nuclease digestion and plasmid Inc/Rep typing.PFGE with S1 nuclease digestion of whole genomic DNA (Promega, Madison,WI) was performed for all 54 isolates (46) to detect and estimate the size of largeplasmids. The XbaI-digested DNA from a strain of Salmonella serotype Braen-derup H9812 was used as a molecular weight standard. A dendrogram of theplasmid profiles was generated by using BioNumerics software and Dice coeffi-cients for band matching with a 2% position tolerance and an optimization of 1%used for generating a dendrogram. Plasmid Inc/Rep typing, including FIA, FIB,FIC, HI1, HI2, I1-I�, L/M, N, P, W, T, A/C, K, B/O, X, Y, F, and FIIA, wasperformed as described previously (6).

Transfer of resistance genes by conjugation. Conjugation experiments wereperformed by plate mating to determine if plasmid coding for CTX-M, TEM,and CMY-2 enzymes was transferable from ESBL-resistant phenotypes. Ninedonors, representative of different �-lactamase enzymes, were used for matingwith a �-lactam-susceptible recipient strain of E. coli, J53 AZr, which is resistantto sodium azide. Cells from late exponential phase of the donors and recipientwere overcrossed on blood agar plate and incubated at 37°C for 18 h. Transcon-jugants were selected on TSA medium supplemented with 300 �g/ml sodiumazide and 2 �g/ml cefotaxime or 32 �g/ml ampicillin. Microdilution susceptibilitytesting was performed to determine the MICs for the transconjugants. PCRamplifications and plasmid replicon typing were performed on the DNA ex-tracted from the transconjugants to determine the presence of ESBL genes andincompatibility groups, respectively.

Hybridization using Southern blotting. The plasmids from donors andtransconjugants were separated by PFGE with S1 nuclease digestion (S1-PFGE)(46). The resolved plasmids were transferred to nylon membranes (GE Health-care, Little Chalfont, England) by capillary transfer and hybridized to specificprobes generated from the purified PCR products, using the oligonucleotideprimers shown in Table S2 in the supplemental material. A BrightStar psoralen-biotin nonisotopic labeling kit (Ambion, Inc., Austin, TX) was used to label thePCR products, which were then used to detect the genes for CTX-M-15, CMY-2,or TEM in the plasmid DNA of the bacterial isolates. The hybridization stepswere performed according to the manufacturer’s protocol (Ambion, Inc.).

RESULTS

Susceptibility of ESBL-producing E. coli strains. Of the 944E. coli strains isolated from companion animals with UTIs andother infections, 54 (�6%) exhibited reduced susceptibility toceftazidime or cefotaxime (MIC � 16 �g/ml). Of these, 29(�54%) were confirmed as ESBL-producing E. coli strainsaccording to CLSI guidelines (13). Thus, the prevalence ofESBL-producing E. coli causing clinical infections in compan-ion animals was 3%. The majority of ESBL-producing E. coli

VOL. 55, 2011 �-LACTAMASE-PRODUCING E. COLI FROM COMPANION ANIMALS 5667

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

5668

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

isolates were recovered from urine specimens (35%, 19/54)(Fig. 1). The MIC50s and MIC90s (�g/ml) of �-lactams forESBL-producing E. coli strains were as follows: ampicillin,�256 and �256; amoxicillin-clavulanic acid, 64 and 64; ceph-alothin, �1,024 and �1,024; cefoxitin, 128 and 256; cefpo-doxime, 256 and �256; ceftazidime, 32 and 256; cefotaxime, 32and �1,024; and meropenem, 0.5 and 4. All isolates had highMIC90s toward all �-lactams, including narrow-, expanded-,and broad-spectrum cephalosporins, but only four isolates(NCTR-929, -926, -938, and -941) exhibited resistance to car-bapenems (i.e., meropenem) (Fig. 1). Among isolates produc-ing ESBL, the frequencies of resistance to other drugs were asfollows: enrofloxacin, 92%; gentamicin, 44%; chlorampheni-col, 44%; doxycycline, 82%; and trimethoprim-sulfamethoxa-zole, 44% (Fig. 1). All but one strain, NCTR-936, of ESBL-producing E. coli exhibited the MDR phenotype (Fig. 1).

Molecular characterization of ESBL genes. The ESC genesblaTEM, blaSHV, blaCTX-M-1 group, blaCMY, and blaOXA-10 weredetected in 74%, 17%, 100%, 89%, and 30% of ESBL-produc-ing E. coli, respectively. The blaOXA-2 group was not detectedin any isolates. Two variants of blaCTX-M genes were identifiedin 63% of the isolates. Sequencing of blaCTX-M subgroupsrevealed blaCTX-M encoding CTX-M-15 as being most preva-lent (78%), whereas blaCTX-M encoding CTX-M-14 and CTX-M-24 was found in 20% (n � 11) and 42% (n � 23) of theisolates, respectively (Fig. 1). Sequencing of the blaTEM generevealed blaTEM-1 (n � 33) and two inhibitor-resistant TEM(IRT) �-lactamase-encoding genes, blaTEM-33 (IRT-5) (n � 3)and blaTEM-30 (IRT-2) (n � 1) (Fig. 1). Additionally, a novelblaTEM gene, named blaTEM-181, resulting from a Ala184Valmutation, was detected in three clinical isolates, NCTR-920,-926, and -927.

All ESBL-producing E. coli isolates in which blaSHV,blaOXA-10, and blaCMY were amplified by PCR producedSHV-3, OXA-1, and CMY-2 variants, respectively (Fig. 1).

Mutations in the ampC promoter-attenuator region in theESBL-producing E. coli isolates. Mutations in the promoterand attenuator regions of the chromosomal ampC gene werefound in �78% (42/54) of isolates (Fig. 1). These mutationswere detected in association with ESBL genes, either alone atpositions 58 and �28 or in the following combinations: �18,�1, 20, and 58; �18, �1, and 58; �28 and 58; �28,34 and 58; �32 and 58; �28, 22, 26, 27, and 32;and 22, 26, 27, and 32. The distance between the �35and �10 boxes of the ampC gene, which also plays an impor-tant role in strengthening the ampC promoter (8), was 16 bp inall isolates. However, one novel insertion of adenine betweenbases �28 and �29, which increased the distance between theconserved regions to 17 bp, was detected in the ampC pro-moter region of isolate NCTR-929.

PFGE typing. PFGE subtyping of all 54 ESBL-producing E.coli isolates generated 51 XbaI-PFGE patterns (Fig. 1). No

genetic relatedness was found among the isolates carrying dif-ferent ESBLs by PFGE at the cutoff value of �90% (Fig. 1).However, three distinct PFGE clusters, each containing twoclosely related isolates (�93%), were identified (Fig. 1). Iso-lates in two clusters carried the same resistance determinants.In one cluster, isolates NCTR-920 and -922 displayed the sameESBL resistance genes and profiles, with the exception of fos-fomycin resistance in NCTR-922. Isolates NCTR-926 and -927,in the second cluster, also exhibited the same ESBL resistancegenes and phenotypes, with the exception of meropenem re-sistance in NCTR-926. However, the third cluster, comprisedof NCTR-911 and -935, had different antimicrobial suscepti-bility profiles and resistance gene determinants (Fig. 1).

In silico structural analysis of the wild-type (TEM-1 [A184])and mutant (TEM-181 [A184V]) enzymes. As shown in Fig. 2,amino acid residue 184 is located in the -helix H10 (position182 to 196 in 1JTD), surrounded by the -helices H8 (144 to154) and H2 (71 to 86) and the �-sheet of five antiparallelstrands. In the mutant strains NCTR-920, -926, and -927, theside chain of Val-184 of the helix is totally exposed on theenzyme surface, with a solvent accessibility of 24.5%. Onthe other hand, in the corresponding helix of the enzyme ofthe wild type, the Ala-184 shows 20.5% solvent accessibility.The distance between the residue Val-184 and the reactiveSer-70 of the catalytic site at the N terminus of the -helix H2is �17 Å, measured from C to C. To verify the proteinstability upon the point mutation in both the wild type (Ala-184) and the mutant (Val-184), ��G values were analyzed.Interestingly, the overall stability change, calculated using theatom and torsion angle potentials together, was increased as a

FIG. 1. Dendrographic analysis of XbaI PFGE data for canine and feline clinical isolates of Escherichia coli isolates from the United States.Black, gray, and white squares represent resistant, intermediate, and susceptible isolates, respectively. Abbreviations: CA, California; NC, NorthCarolina; IL, Illinois; MA, Massachusetts; OH, Ohio; ND, not determined. The boxes identify three distinct PFGE clusters, each with two closelyrelated (�93%) isolates that display either the same (NCTR-920 and -922; NCTR-926 and -927) or different (NCTR-911 and -935) ESBL-resistantgenes and different susceptibility profiles.

FIG. 2. Tertiary structure of the class A TEM-1 (1JTD; Val-184).The locations and side chains of the mutation are highlighted. Thediagram was drawn with PyMOL (0.99RC6).


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

result of the conversion of Ala to Val, with a predicted ��G of0.6 kcal/mol. On the other hand, a Val-to-Ala conversion de-creased stability by �2.7 kcal/mol. The RMSD ( carbons) ofthe wild-type and mutant proteins was 0.06 Å, indicating nosignificant structural change.

Plasmid characterization of the ESBL-producing isolates.S1-PFGE analysis revealed plasmids ranging from 11 kbp to233 kbp (Fig. 3). Common plasmids with approximate sizes of

93 and 55 kbp were identified in 60% (n � 32) and 43% (n �23) of isolates, respectively. The S1-PFGE dendrogram sepa-rated the plasmids into 32 distinct clusters, based on a �90%cutoff (Fig. 3). There were 12 distinct S1-PFGE clusters with�2 indistinguishable isolates. Of these, one large cluster con-tained six indistinguishable isolates from different animals withtwo large plasmids, 93 kbp and 55 kbp.

The plasmid replicons IncI1 and IncFIA were most prev-

FIG. 3. Dendrographic analysis of plasmid replicons from clinical E. coli isolates originating from companion animals and subjected to PFGEwith S1 nuclease digestion. Plasmid sizes are in kbp. Abbreviations; A, ampicillin; X, amoxicillin-clavulanic acid; C, cephalothin; F, fosfomycin; O,cefoxitin; P, cefpodoxime; T, cefotaxime; Z, ceftazidime; M, meropenem; E, enrofloxacin; D, doxycycline; H, chloramphenicol; G, gentamicin; S,trimethoprim-sulfamethoxazole.

5670 SHAHEEN ET AL. ANTIMICROB. AGENTS CHEMOTHER.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

alent (96%, 52/54), followed by IncFIB (87%) among E. coliisolates (Fig. 3). Replicon typing identified IncB/O (n � 2),FIC (n � 10), N (n � 1), Y (n � 10), K/B (n � 1), and HI1(n � 1) replicons.

Conjugation experiments. A subset of 27 randomly selecteddonor E. coli strains were conjugated with a recipient strain, E.coli J53 AZr. After mating, a total of nine (33%) of transcon-jugants were selected for further analysis based on resistanceto ampicillin (32 �g/ml) or cefotaxime (2 �g/ml). The MICs fordifferent �-lactams and ESBL genes detected by PCR in thedonors and transconjugants are shown in Table 1. All blaCMY

and blaCTX-M-15 and two blaTEM genes were transferred anddetected by PCR in the transconjugants.

In addition to ampicillin or cefotaxime, all donors and theirtransconjugants were resistant to amoxicillin-clavulanic acid,cephalothin, cefazolin, cefoxitin, cefpodoxime, and ceftiofur.Resistance to ceftazidime (MIC � 32 �g/ml) was observed insix isolates and their respective transconjugants (Table 1). Alltransconjugants exhibited an increase of at least 16-fold inMICs of cefotaxime and ceftazidime compared to the recipi-ent, E. coli J53 AZr, while clavulanic acid did not alter sensi-tivity of the transconjugants to these drugs. Resistance tocefepime was not transferred to the transconjugant (Trans-918) (Table 1). All transconjugants remained susceptible toenrofloxacin, gentamicin, chloramphenicol, doxycycline andtrimethoprim-sulfamethoxazole (data not shown).

Hybridization experiments. Following electrophoresis, plas-mids detected by PFGE with S1 nuclease digestion were usedfor Southern blotting. Probes were used to localize theblaCMY-2 and blaCTX-M-15 genes in a subset of donors (NCTR-890, -916, -917, -918, -920, and -927) and their transconjugants(Fig. 4). Plasmid DNA from these isolates yielded strong hy-bridization signals with the blaCMY-2 gene probe (Fig. 4). Eachof these six donors had a unique plasmid profile and harboredtwo or three large plasmids (size range, 50 to 150 kbp),whereas only one large plasmid (size range, 50 to 93 kbp) wasdetected in transconjugants (Fig. 4B). The blaCMY-2 probehybridized with plasmids of the same size in donors andtransconjugants. Similarly, the blaCTX-M-15 probe hybridizedwith the same-size blaCMY-2-positive plasmids, suggesting apossible cointegration event with the blaCMY-2 gene into thesame large plasmid.

The blaTEM probe from five of the six donors (NCTR-890,-916, -920, -926, and -927) hybridized to plasmids of 150, 93, 55,55, and 55 kbp, respectively (Fig. 5). No hybridization signal wasdetected in the transconjugants (Fig. 5B). Plasmids that hybrid-ized to the blaTEM probe were different from those that hybrid-ized to blaCMY-2 and blaCTX-M-15 probes (Fig. 4B and 5B).

DISCUSSION

Our study shows that ESBL-producing E. coli strains can beisolated from companion animals in the United States. MostESBL-producing E. coli strains exhibited the MDR phenotype.Diverse �-lactamase families associated with high-spectrumactivity against �-lactams were found in isolates and CTX-M,mainly blaCTX-M-1-group, was endemic in the United States. Insome isolates we also found a novel blaTEM gene, blaTEM-181,with a mutation resulting in the conversion of Ala184 to Val.We also detected novel mutations in the promoter-attenuator

region of ampC. Resistance genes encoding �-lactamases werelocated on transferable-conjugative plasmids, with different in-compatibility groups conferring high MICs to ESCs.

Data on the prevalence and characterization of ESBL-pro-ducing bacteria causing infections in companion animals in theUnited States are limited (26). Our results showed that theincidence of resistance to ESCs in 944 clinical isolates wassimilar to that reported from two studies in Europe (17). In ourstudy, nearly 6% of isolates exhibited reduced susceptibility toESCs. Of these, 3% (29 of 944) of the isolates produced ES-BLs. The incidence of ESBL-producing strains reported inChina and in one study from the United States were 68% (69of 101) and 40% (60 of 150), respectively (29, 37). The differ-ences may reflect the greater number of isolates screened forESBLs in our study (29, 37).

The PFGE results revealed a high degree of variation inantimicrobial resistance patterns among the 54 isolates. Thisresult demonstrates the huge genetic heterogeneity among E.coli isolates containing transferable ESC-resistance genes, sug-gesting that the dissemination of ESBL genes is not due to thepresence of a single clone (1, 3). Among the 29 ESBL-produc-ing isolates, most harbored several ESBL genes known to in-crease the spectrum of activity of ESBL enzymes and conferhigh-level resistance to �-lactams (Fig. 1; Table 1). Interest-ingly, 78% of isolates selected for this study carried blaCTX-M

(i.e., blaCTX-M-15). blaCTX-M-15 was first reported in 6% of E.coli isolates collected from canines and felines with UTIs in theUnited States (37). The CTX-M-1 group has been documentedin companion animals worldwide (7, 15–17, 34). Of particularpublic health concern is the emergence of the CTX-M-15-positive B2-O25:H4-ST131 clone in E. coli strains from com-panion animals (17, 36, 41).

The other enzyme conferring �-lactam resistance in 74% ofour isolates was a TEM-type �-lactamase (74%) (Fig. 1); theblaTEM-1 gene was the most prominent (82.5%). The blaTEM-1

variant was previously identified in clinical E. coli isolates fromcompanion animals in Europe but not in the United States (16,18, 40). Although the TEM-1 variant is not considered anESBL, its clinical importance relies on its potential to undergomutations to increase its activity against other �-lactams (10,25). We have identified two IRT �-lactamase-encoding genes,blaTEM-33 (IRT-5) and blaTEM-30 (IRT-2), from four E. coliisolates (NCTR-890, -914, -917, and -918). The fact that bla-TEM gene variants are more commonly detected in human E.coli isolates may reflect the fact that more studies have beenperformed with E. coli strains from humans than from com-panion animals (5, 12, 33). The new blaTEM-181 gene found inthis study carried one mutation, which led to the Ala184Valsubstitution. Isolates NCTR-920, -926, and -927, harboringblaTEM-181, not only were resistant to aminopenicillins but alsoexhibited high-level (MIC, 64 to 128 �g/ml) resistance to cef-tazidime, a broad-spectrum cephalosporin (Fig. 1). The effectof the amino acid substitution in this new variant is unclear interms of the enzyme function. Ala and Val are nonpolar aminoacids with alkyl side chains. Hence, the switch of Ala to Val atposition 184 does not significantly change the polarity of -he-lix H10. Ala is one of the best helix-forming residues in com-bination with lysine, glutamine, and methionine, whose sidechains have straight-chain aliphatic regions, whereas the�-branched Val is invariably a poor helix former (38). How-


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

TA

BL

E1.

Ant

imic

robi

alsu

scep

tibili

typr

ofile

sof

E.c

olii

sola

tes

used

inco

njug

atio

nex

peri

men

tsa

Gro

upan

dis

olat

e

MIC

(�g/

ml)

ofan

timic

robi

alag

ent

Pres

ence

orab

senc

eof

:Pl

asm

ids

AM

IA

UG

AM

PF

AZ

FE

PF

OT

FO

XPO

DT

AZ

TIO

AX

OC

EP

F/C

T/C

TE

MC

MY

CT

X-M

-Ilik

egr

oup

CT

X-M

-I4

like

grou

pO

XA

Plas

mid

size

(s)

(kbp

)R

eplic

onty

pe(s

)

Don

ors

NC

TR

-927

4>

32>

32>

16�

18

64>

3232

816

>16

816

��

��

�55

,93

I1,F

IC,F

IAN

CT

R-9

29>

64>

32>

32>

16�

1>

64>

64>

3212

8>

8>

64>

1664

64�

��

��

120,

55I1

,FIC

,FIA

NC

TR

-917

4>

32>

32>

16�

116

>64

>32

32>

816

>16

816

��

��

�55

,93

I1,F

IA,F

IBN

CT

R-9

188

32>

32>

16>

16>

64>

64>

3232

>8

>64

>16

1616

��

��

�40

,55,

93I1

,FIA

NC

TR

-890

2>

32>

32>

164

32>

64>

3212

8>

864

>16

1632

��

��

�50

,140

,150

I1,F

IA,F

IBN

CT

R-9

002

>32

>32

>16

�1

8>

64>

3232

>8

16>

1616

32�

��

��

100

I1,F

IC,F

IA,

FIB

NC

TR

-926

2>

32>

32>

16�

18

64>

3232

>8

16>

1616

32�

��

��

55,9

3I1

,FIA

,FIB

NC

TR

-920

2>

32>

32>

16�

116

>64

>32

32>

832

>16

816

��

��

�55

,93

I1N

CT

R-9

162

>32

>32

>16

�1

32>

64>

3264

>8

32>

1616

16�

��

��

37,6

4,93

I1,F

IA

Rec

ipie

ntJ5

3Az

18

4�

8�

1�

0.25

�4

1�

0.25

0.5

�0.

25�

8�

0.12

�0.

12�

��

��

ND

ND

Tra

nsco

njug

ants

Tra

ns-9

271

>32

>32

>16

�1

864

>32

328

16>

164

8�

��

��

93I1

Tra

ns-9

29�

0.5

>32

>32

>16

�1

864

>32

32>

816

>16

816

��

��

�55

I1T

rans

-917

2>

32>

32>

16�

116

>64

>32

32>

816

>16

816

��

��

�55

I1T

rans

-918

132

>32

>16

�1

464

>32

88

8>

162

8�

��

��

93I1

Tra

ns-8

902

>32

>32

>16

�1

1664

>32

32>

832

>16

816

��

��

�55

I1T

rans

-900

232

>32

>16

�1

432

>32

8>

88

>16

44

��

��

�10

0I1

Tra

ns-9

26�

0.5

>32

>32

>16

�1

864

>32

32>

816

>16

1616

��

��

�93

I1T

rans

-920

�0.

5>

32>

32>

16�

18

64>

3216

>8

16>

164

8�

��

��

93I1

Tra

ns-9

161

>32

>32

>16

�1

1664

>32

64>

816

>16

168

��

��

�64

I1

aA

UG

,am

oxic

illin

-cla

vula

nic

acid

;A

MP,

ampi

cilli

n;T

IO,c

eftio

fur;

AX

O,c

eftr

iaxo

ne;

CE

P,ce

phal

othi

n;F

OX

,cef

oxiti

n;T

AZ

,cef

tazi

dim

e;F

AZ

,cef

azol

in;

FE

P,ce

fepi

me;

T/C

,cef

tazi

dim

e/cl

avul

anic

acid

;F

/C,

cefo

taxi

me/

clav

ulan

icac

id;F

OX

,cef

oxiti

n;F

OT

,cef

otax

ime;

POD

,cef

podo

xim

e.N

D,n

otde

term

ined

.Num

bers

inbo

ldar

ere

sist

ance

brea

kpoi

ntso

rMIC

swhi

chin

crea

sed

atle

ast1

6-fo

ldre

lativ

eto

thos

efo

rthe

reci

pien

t(E

.col

iJ53

AZ

r ).


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

ever, an Ala-to-Val mutation on -helix H10 increased ratherthan decreased the overall stability of the mutant TEM-181.The stability effect caused by a difference in intrinsic sec-ondary structure propensity is probably superseded by otherenergetic effects related to tertiary contacts. Moreover, Val-184 is located 17.44 Å from reactive Ser-70 of the catalyticsite (Fig. 2), and no significant structural changes werecaused by the substitution, suggesting that substitution ofthis residue (Val-184) may not contribute directly to thesubstrate profile and catalytic parameters of the �-lacta-mase. Therefore, as suggested in the case of the Met182Thrmutation (45), the Ala184Val substitution might function asa suppressor of folding of TEM or stability defects resultingfrom mutations directly associated with �-lactam specificityand catalysis. Alternatively, this mutation could facilitaterequired structural changes at other positions. Other en-zymes which display Ala-to-Val substitutions at 184 alongwith a mutation(s) at a different position(s) have beenreported elsewhere (http://www.lahey.org/Studies/temtable.asp).

Although the OXA-1 type �-lactamase is not regarded as anESBL and does not hydrolyze ESCs, this enzyme hydrolyzes�-lactamase inhibitors (e.g., clavulanic acid), compromisingeffective treatment during combined use of these drugs (21). Inour study, �28% of E. coli strains carried this variant, com-pared to those found in canine uropathogenic E. coli strains inPortugal (0.8%) (18). The blaOXA-1 gene is widely dissemi-nated among human E. coli strains of phylogroup B2 and/orplasmids containing blaCTX-M-15, blaTEM-1, and aac(6�)-Ib-crresistance genes (30, 39). The increased incidence of the OXA-type lactamase could be linked to clinical use of �-lactams witha �-lactamase inhibitor for treatment of clinical infections incompanion animals (26).

Among plasmid-borne class C �-lactamase genes, blaCMY-2

was detected in 89% of the isolates. All isolates positive for theCMY-2 variant were resistant to broad-spectrum cephalospo-rins. The presence of a CMY-2 �-lactamase in E. coli strainsfrom companion animals has been described from the UnitedStates (42) and Italy (7, 28). The blaCMY-2 gene can be trans-

ferred between different bacterial species and between animalsand humans (49). The high prevalence of class C �-lactamases,particularly blaCMY-2, is of concern, since these �-lactamasesnot only confer resistance to a wide range of ESCs used fortreatment of serious infections in humans or animals but alsoare not affected by �-lactamase inhibitors (2, 26). In the pres-ence of CMY-2 in clinical E. coli strains, the MICs of ceftazi-dime and cefotaxime in the presence of clavulanic acid wereincreased �8-fold compared to that for E. coli J53 AZr, whilefor select transconjugants, increases in the MICs varied from�16-fold (e.g., NCTR-918) to �130-fold (e.g., NCTR-926)(Table 1). Such variation in the expression of CMY-2 �-lacta-mase in E. coli isolates, as displayed by diverse transconjugantMICs, possibly reflects different copy numbers of the cmygenes associated with their relationship to plasmid size andcopy number. This also might reflect differences in the putativepromoter sequence on blaCMY-2 transcription, as the promoteris found within an ISEcp1-like element located upstream ofblaCMY-2 (19). The presence of CMY in E. coli isolates wouldmask the detection of ESBLs, which requires an innovativeapproach.

In our study, we found that both IncI and IncFIA (96%),followed by IncFIB (87%), were the most prevalent plasmidreplicon types in E. coli. Similarly, a recent study has identifiedIncI, IncFIA, and IncFIB as common replicon types in CTX-M-15-type-ESBL-producing E. coli strains from companionanimals in Europe (17). Given the high incidence of CTX-M-15-type ESBLs in our isolates (Fig. 1), the data suggest acommon plasmid backbone on which the blaCTX-M-15 gene islikely to be located. In contrast to a previous study (17), weobserved a wide diversity of plasmids from different incompat-ible groups (Fig. 3). However, none of the isolates was positivefor the FII replicon, previously identified in E. coli from hu-mans (20). Multiple plasmid replicons (�4 replicon types)were observed in 20 E. coli strains (Fig. 3). The presence ofmultiple replicons in Salmonella and E. coli isolates from hu-mans has been described previously (24, 32); however, theirclinical significance is still unknown.

Conjugation experiments demonstrated that �-lactam resis-

FIG. 4. PFGE of plasmid DNA of donors and transconjugants(A) and Southern hybridization of plasmid DNA from donors andtransconjugants with a blaCMY-2 probe (B). Lanes A to F, donorsNCTR-890, -916, -917, -918, -920, and -927, respectively; lanes 1 to 6,transconjugants Trans-890, -916, -917, -918, -920, and -927, respec-tively.

FIG. 5. PFGE of plasmid DNA of donors and transconjugants(A) and Southern hybridization of plasmid DNA from donors andtransconjugants with a blaTEM probe (B). Lanes A to F, donors NCTR-890, -916, -918, -920, -926, and -927, respectively; lanes 1 to 6, transcon-jugants Trans-890, -916, -918, -920, -926, and -927, respectively.


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

tance was transferred from nine clinical E. coli isolates amongthe 27 tested (Table 1). The transconjugants expressed resis-tance to aminopenicillins and narrow-, expanded-, and broad-spectrum cephalosporins. PCR and Southern blotting revealedthe presence of blaCMY, blaCTX-M, and blaTEM genes in donorstrains and their respective transconjugants (Table 1; Fig. 4and 5). Due to the presence of a CMY-type �-lactamase, theMICs of ceftazidime or cefotaxime combined with clavulanicacid for transconjugants were increased �16-fold relative tothe recipient, E. coli J53 AZr. The blaCMY and blaCTX-M geneswere located on the same plasmids (e.g., 93-kbp plasmids ofNCTR-920 and -927) in the donors and were cotransferred tothe recipients (Fig. 4). The blaTEM gene was located on differ-ent size plasmids (e.g., 55-kbp plasmids of NCTR-920 and-927) (Fig. 5). PCR detected blaTEM in all donors and only twotransconjugants (NCTR-926 and -927) (Table 1). The blaTEM

hybridization signal was absent from two donors and alltransconjugants, which may indicate that blaTEM can be pres-ent in other, potentially transferrable genetic elements thatwere not readily detected by Southern hybridization of theplasmids. All transconjugants were susceptible to other drugclasses (data not shown). These results suggest that MDRgenes could be carried on integron-associated plasmids andthat selective pressure of certain antimicrobials may be neces-sary to mobilize these plasmids.

Mutation at position �32 of ampC, which is necessary forthe overexpression of this gene (5, 8), was found in only oneisolate (NCTR-940) (Fig. 1) along with a mutation at 58. InE. coli, the primary mechanism of hyperproduction of a chro-mosomal ampC gene is mediated through mutations in thepromoter and an attenuator of ampC, which confers resistanceto ampicillin, cefoxitin, and ESCs (47). A previous study hasshown that mutations at �32 overexpressed the ampC gene124-fold (47). Other important mutations, at positions �42and �1 (5, 8), were not found in this study. Mutations at �18,which play an important role in over expression of the ampCgene by creating a new �10 box (8), were found in 18% of theisolates. Mutation at �28 with an uncertain function related tothe expression of ampC (35) were observed in one-third of ourisolates, and there was one novel insertion of adenine between�28 and �29 in E. coli NCTR-929. This insertion created analternate displaced promoter with new �35 and �10 boxes,separated by 17 bp, and could be responsible for overexpres-sion of ampC (47). In our study, the effect of the importantmutations and insertion in the ampC promoter region on theresistance phenotype was not evaluated. Their clinical impacton the resistance phenotype could be masked by the presenceof plasmid-mediated �-lactam resistance found in our isolates.To the best of our knowledge, this is the first report to describemutations associated with chromosomal AmpC �-lactamasesin E. coli strains from companion animals.

In summary, �6% of clinical E. coli isolates from companionanimals had reduced susceptibility to ESCs. The resistant phe-notypes harbored different variants of ESBL genes. The genes,which are located on conjugative plasmids, conferred resis-tance to ESCs in recipient cells. The emergence and dissemi-nation of conjugative plasmid-mediated ESC resistance in ca-nine and feline E. coli isolates can serve as a potential reservoirof ESBL genes. Hence, prudent use of ESCs by veterinariansin treating companion animals is warranted.

ACKNOWLEDGMENTS

We thank Shaohua Zhao (Center for Veterinary Medicine, U.S.Food and Drug Administration) for kindly providing ESBL-positivecontrol strains, George A. Jacoby and Quynh T. Tran for providing E.coli J53 AZr, and Alessandra Carattoli for providing replicon typing-positive control strains for this study. We thank George A. Jacoby forhelpful discussions and John B. Sutherland, Mark Hart, and Carl E.Cerniglia for their critical review of the manuscript. Special thanks goto Carl E. Cerniglia for his encouragement and support of this work.

Bashar W. Shaheen is supported by the Oak Ridge Institute forScience and Education. Views presented in this article do not neces-sarily reflect those of the U.S. Food and Drug Administration.

REFERENCES

1. Bonnet, R. 2004. Growing group of extended-spectrum �-lactamases: theCTX-M enzymes. Antimicrob. Agents Chemother. 48:1–14.

2. Bradford, P. A. 2001. Extended-spectrum �-lactamases in the 21st century:characterization, epidemiology, and detection of this important resistancethreat. Clin. Microbiol. Rev. 14:933–951.

3. Brinas, L., et al. 2005. Mechanisms of resistance to expanded-spectrumcephalosporins in Escherichia coli isolates recovered in a Spanish hospital. J.Antimicrob. Chemother. 56:1107–1110.

4. Brinas, L., et al. 2005. Monitoring and characterization of extended-spec-trum �-lactamases in Escherichia coli strains from healthy and sick animals inSpain in 2003. Antimicrob. Agents Chemother. 49:1262–1264.

5. Brinas, L., M. Zarazaga, Y. Saenz, F. Ruiz-Larrea, and C. Torres. 2002.�-Lactamases in ampicillin-resistant Escherichia coli isolates from foods,humans, and healthy animals. Antimicrob. Agents Chemother. 46:3156–3163.

6. Carattoli, A., et al. 2005. Identification of plasmids by PCR-based replicontyping. J. Microbiol. Methods 63:219–228.

7. Carattoli, A., et al. 2005. Extended-spectrum �-lactamases in Escherichia coliisolated from dogs and cats in Rome, Italy, from 2001 to 2003. Antimicrob.Agents Chemother. 49:833–835.

8. Caroff, N., E. Espaze, D. Gautreau, H. Richet, and A. Reynaud. 2000. Anal-ysis of the effects of �42 and �32 ampC promoter mutations in clinicalisolates of Escherichia coli hyperproducing AmpC. J. Antimicrob. Che-mother. 45:783–788.

9. Cartelle, M., et al. 2004. High-level resistance to ceftazidime conferred by anovel enzyme, CTX-M-32, derived from CTX-M-1 through a single Asp240-Gly substitution. Antimicrob. Agents Chemother. 48:2308–2313.

10. Chaibi, E. B., D. Sirot, G. Paul, and R. Labia. 1999. Inhibitor-resistant TEM�-lactamases: phenotypic, genetic and biochemical characteristics. J. Anti-microb. Chemother. 43:447–458.

11. Chen, Y. M., P. J. Wright, C. S. Lee, and G. F. Browning. 2003. Uropatho-genic virulence factors in isolates of Escherichia coli from clinical cases ofcanine pyometra and feces of healthy bitches. Vet. Microbiol. 94:57–69.

12. Chmelnitsky, I., Y. Carmeli, A. Leavitt, M. J. Schwaber, and S. Navon-Venezia. 2005. CTX-M-2 and a new CTX-M-39 enzyme are the major ex-tended-spectrum �-lactamases in multiple Escherichia coli clones isolated inTel Aviv, Israel. Antimicrob. Agents Chemother. 49:4745–4750.

13. CLSI. 2008. Performance standards for antimicrobial disk and dilution sus-ceptibility testing for bacteria isolated from animals, 3rd ed. Approved stan-dard M31-A3. Clinical and Laboratory Standards Institute, Wayne, PA.

14. Cooke, C. L., R. S. Singer, S. S. Jang, and D. C. Hirsh. 2002. Enrofloxacinresistance in Escherichia coli isolated from dogs with urinary tract infections.J. Am. Vet. Med. Assoc. 220:190–192.

15. Costa, D., et al. 2008. Prevalence of antimicrobial resistance and resistancegenes in faecal Escherichia coli isolates recovered from healthy pets. Vet.Microbiol. 127:97–105.

16. Costa, D., et al. 2006. Detection of Escherichia coli harbouring extended-spectrum �-lactamases of the CTX-M, TEM and SHV classes in faecalsamples of wild animals in Portugal. J. Antimicrob. Chemother. 58:1311–1312.

17. Ewers, C., et al. 2010. Emergence of human pandemic O25:H4-ST131 CTX-M-15 extended-spectrum-�-lactamase-producing Escherichia coli amongcompanion animals. J. Antimicrob. Chemother. 65:651–660.

18. Feria, C., E. Ferreira, J. D. Correia, J. Goncalves, and M. Canica. 2002.Patterns and mechanisms of resistance to �-lactams and �-lactamase inhib-itors in uropathogenic Escherichia coli isolated from dogs in Portugal. J.Antimicrob. Chemother. 49:77–85.

19. Giles, W. P., et al. 2004. DNA sequence analysis of regions surroundingblaCMY-2 from multiple Salmonella plasmid backbones. Antimicrob. AgentsChemother. 48:2845–2852.

20. Gonullu, N., et al. 2008. Dissemination of CTX-M-15 �-lactamase genescarried on Inc FI and FII plasmids among clinical isolates of Escherichia coliin a university hospital in Istanbul, Turkey. J. Clin. Microbiol. 46:1110–1112.

21. Gonzalez-Sanz, R., S. Herrera-Leon, M. de la Fuente, M. Arroyo, and M. A.Echeita. 2009. Emergence of extended-spectrum �-lactamases and AmpC-


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

type �-lactamases in human Salmonella isolated in Spain from 2001 to 2005.J. Antimicrob. Chemother. 64:1181–1186.

22. Guardabassi, L., S. Schwarz, and D. H. Lloyd. 2004. Pet animals as reser-voirs of antimicrobial-resistant bacteria. J. Antimicrob. Chemother. 54:321–332.

23. Hagman, R., and I. Kuhn. 2002. Escherichia coli strains isolated from theuterus and urinary bladder of bitches suffering from pyometra: comparisonby restriction enzyme digestion and pulsed-field gel electrophoresis. Vet.Microbiol. 84:143–153.

24. Hopkins, K. L., et al. 2006. Replicon typing of plasmids carrying CTX-M orCMY �-lactamases circulating among Salmonella and Escherichia coli iso-lates. Antimicrob. Agents Chemother. 50:3203–3206.

25. Knox, J. R. 1995. Extended-spectrum and inhibitor-resistant TEM-type�-lactamases: mutations, specificity, and three-dimensional structure. Anti-microb. Agents Chemother. 39:2593–2601.

26. Li, X. Z., M. Mehrotra, S. Ghimire, and L. Adewoye. 2007. �-lactam resis-tance and �-lactamases in bacteria of animal origin. Vet. Microbiol. 121:197–214.

27. Liebana, E., et al. 2004. Characterization of �-lactamases responsible forresistance to extended-spectrum cephalosporins in Escherichia coli and Sal-monella enterica strains from food-producing animals in the United King-dom. Microb. Drug Resist. 10:1–9.

28. Livermore, D. M. 1995. �-Lactamases in laboratory and clinical resistance.Clin. Microbiol. Rev. 8:557–584.

29. Ma, J., et al. 2009. High prevalence of plasmid-mediated quinolone resis-tance determinants qnr, aac(6�)-Ib-cr, and qepA among ceftiofur-resistantEnterobacteriaceae isolates from companion and food-producing animals.Antimicrob. Agents Chemother. 53:519–524.

30. Machado, E., et al. 2006. Dissemination in Portugal of CTX-M-15-, OXA-1-,and TEM-1-producing Enterobacteriaceae strains containing the aac(6�)-Ib-cr gene, which encodes an aminoglycoside- and fluoroquinolone-modify-ing enzyme. Antimicrob. Agents Chemother. 50:3220–3221.

31. Martinez-Martinez, L., et al. 2000. Activities of imipenem and cephalospo-rins against clonally related strains of Escherichia coli hyperproducing chro-mosomal �-lactamase and showing altered porin profiles. Antimicrob.Agents Chemother. 44:2534–2536.

32. Mataseje, L. F., et al. 2009. Characterization of Canadian cefoxitin-resistantnon-typhoidal Salmonella isolates, 2005–06. J. Antimicrob. Chemother. 64:723–730.

33. Miro, E., et al. 2002. Prevalence of clinical isolates of Escherichia coli pro-ducing inhibitor-resistant �-lactamases at a University Hospital in Barce-lona, Spain, over a 3-year period. Antimicrob. Agents Chemother. 46:3991–3994.

34. Moreno, A., H. Bello, D. Guggiana, M. Dominguez, and G. Gonzalez. 2008.Extended-spectrum �-lactamases belonging to CTX-M group produced byEscherichia coli strains isolated from companion animals treated with enro-floxacin. Vet. Microbiol. 129:203–208.

35. Mulvey, M. R., et al. 2005. Molecular characterization of cefoxitin-resistant

Escherichia coli from Canadian hospitals. Antimicrob. Agents Chemother.49:358–365.

36. Nicolas-Chanoine, M. H., et al. 2008. Intercontinental emergence of Esch-erichia coli clone O25:H4-ST131 producing CTX-M-15. J. Antimicrob. Che-mother. 61:273–281.

37. O’Keefe, A., T. A. Hutton, D. M. Schifferli, and S. C. Rankin. 2010. Firstdetection of CTX-M and SHV extended-spectrum �-lactamases in Esche-richia coli urinary tract isolates from dogs and cats in the United States.Antimicrob. Agents Chemother. 54:3489–3492.

38. Pace, C. N., and J. M. Scholtz. 1998. A helix propensity scale based onexperimental studies of peptides and proteins. Biophys. J. 75:422–427.

39. Pai, H., M. R. Kim, M. R. Seo, T. Y. Choi, and S. H. Oh. 2006. A nosocomialoutbreak of Escherichia coli producing CTX-M-15 and OXA-30 �-lactamase.Infect. Control Hosp. Epidemiol. 27:312–314.

40. Pomba-Feria, C., and M. Canica. 2003. A novel sequence framework (bla-TEM-1G) encoding the parental TEM-1 �-lactamase. FEMS Microbiol. Lett.220:177–180.

41. Pomba, C., J. D. da Fonseca, B. C. Baptista, J. D. Correia, and L. Martinez-Martinez. 2009. Detection of the pandemic O25-ST131 human virulentEscherichia coli CTX-M-15-producing clone harboring the qnrB2 andaac(6�)-Ib-cr genes in a dog. Antimicrob. Agents Chemother. 53:327–328.

42. Sanchez, S., et al. 2002. Characterization of multidrug-resistant Escherichiacoli isolates associated with nosocomial infections in dogs. J. Clin. Microbiol.40:3586–3595.

43. Shaheen, B. W., D. M. Boothe, O. A. Oyarzabal, and T. Smaha. 2010.Antimicrobial resistance profiles and clonal relatedness of canine and felineEscherichia coli pathogens expressing multidrug resistance in the UnitedStates. J. Vet. Intern. Med. 24:323–330.

44. Shaheen, B. W., O. A. Oyarzabal, and D. M. Boothe. 2010. The role of class1 and 2 integrons in mediating antimicrobial resistance among canine andfeline clinical E. coli isolates from the US. Vet. Microbiol. 144:363–370.

45. Sideraki, V., W. Huang, T. Palzkill, and H. F. Gilbert. 2001. A secondarydrug resistance mutation of TEM-1 �-lactamase that suppresses misfoldingand aggregation. Proc. Natl. Acad. Sci. U. S. A. 98:283–288.

46. Sirichote, P., et al. 2010. Molecular characterization of extended-spectrumcephalosporinase-producing Salmonella enterica serovar Choleraesuis iso-lates from patients in Thailand and Denmark. J. Clin. Microbiol. 48:883–888.

47. Tracz, D. M., et al. 2007. ampC gene expression in promoter mutants ofcefoxitin-resistant Escherichia coli clinical isolates. FEMS Microbiol. Lett.270:265–271.

48. Whichard, J. M., et al. 2005. �-lactam resistance and Enterobacteriaceae,United States. Emerg. Infect. Dis. 11:1464–1466.

49. Winokur, P. L., D. L. Vonstein, L. J. Hoffman, E. K. Uhlenhopp, and G. V.Doern. 2001. Evidence for transfer of CMY-2 AmpC �-lactamase plasmidsbetween Escherichia coli and Salmonella isolates from food animals andhumans. Antimicrob. Agents Chemother. 45:2716–2722.

50. Zhao, S., et al. 2001. Identification and expression of cephamycinase blaCMYgenes in Escherichia coli and Salmonella isolates from food animals andground meat. Antimicrob. Agents Chemother. 45:3647–3650.


Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

ac o

n 04

Dec

embe

r 20

21 b

y 17

7.15

4.22

7.23

.

molecular characterization of resistance to extended-spectrum cephalosporins in clinical

Documents