Survival in the environment is a possible key factor for

the expansion of Escherichia coli strains producing

extended-spectrum b-lactamases

GUSTAF STARLANDER, HONG YIN, PETRA EDQUIST and �ASA MELHUS

Department of Medical Sciences/Section of Clinical Bacteriology, Uppsala University, Uppsala, Sweden

Gustaf Starlander, Hong Yin, Petra Edquist, �Asa Melhus. Survival in the environment is a possiblekey factor for the expansion of Escherichia coli strains producing extended-spectrum b-lactamases.APMIS 2014; 122: 59–67.

Acquired resistance to cephalosporins in Enterobacteriaceae is a global problem. After an outbreak atUppsala University Hospital of extended-spectrum b-lactamase (ESBL)-positive Klebsiella pneumoniaeproducing CTX-M-15, there was a shift from AmpC to ESBL production among Escherichia coli isolates.To explore the basis for this epidemiological shift, 46 E. coli isolates (ESBLs, n = 23; AmpC, n = 23) werecharacterized with regard to genetic relatedness, b-lactamase, replicon and integron types, antibiotic resis-tance profiles, and genes encoding virulence factors. In addition, the survival in the environment and onhospital-associated materials was analysed. CTX-M-15 was the most frequent ESBL (78%). Only three(13%) of the AmpC enzymes were harboured on plasmids (CMY-2, DHA-1). Independent of plasmid-mediated beta-lactamase, IncF plasmids predominated and only class I integrons were detected. The ESBLproducers carried more virulence genes (p = 0.04), exhibited a broader resistance phenotype (p = 0.01) andsurvived significantly longer (p = 0.03) on different materials than the AmpC-producing isolates. In conclu-sion, ESBL-producing isolates had properties which are likely to augment their competitiveness. Apartfrom antibiotic resistance and virulence factors, extended survival in the environment could be a selectivetrait for successful ESBL-producing E. coli strains.

Key words: E. coli; AmpC; ESBL; environment; virulence genes; plasmids.

Gustaf Starlander, Department of Clinical Microbiology, Uppsala University Hospital, SE-751 85 Uppsala,Sweden. e-mail: gustaf_starlander@hotmail.com

Escherichia coli is a Gram-negative rod belong-ing to the Enterobacteriaceae family. Althoughpart of the human gastrointestinal flora, it isone of the clinically most important patho-gens. E. coli is the leading cause of urinarytract infections (1) and among the mostcommon causes of septicaemia (2). It can alsocause neonatal meningitis, pneumonia andgastroenteritis.

Extraintestinal pathogenic E. coli isolatespossess certain virulence factors, which arescarce among commensal isolates (3). Thesevirulence factors facilitate colonization andinvasion of the host and cause cell injury. Apartfrom virulence factors, multiple antibiotic resis-tance is a suggested selective trait in epidemio-logical studies (4). High-risk clones, such asO25b-ST131, play a major role in the dissemina-tion of b-lactamases with extended spectra (4, 5).The tenacity of these clones to disseminate andsurvive is, however, poorly understood.Received 3 December 2012. Accepted 7 March 2013

59

APMIS 122: 59–67 © 2013 APMIS. Published by John Wiley & Sons Ltd.

DOI 10.1111/apm.12102

Escherichia coli with acquired resistance tobroad-spectrum cephalosporins is an increasingclinical problem, which is augmented by asso-ciated resistance to other classes of antibiotics,thereby leaving few treatment options (6).AmpC is a b-lactamase that is localized on theE. coli chromosome (cAmpC); it is naturallyexpressed in limited amounts, but mutationscan cause overexpression of AmpC and leadto cephalosporin-resistance (7). There are alsoplasmid-mediated versions of AmpC (pAm-pCs). These are derived from the chromosomesof other bacterial species, but can sometimesbe transferred to E. coli (8).The most widely reported b-lactamases are,

however, extended-spectrum b-lactamases (ES-BLs) of the CTX-M types. Their rapid dissemi-nation is taking pandemic proportions (9, 10),although there are geographic differences.So far, the Scandinavian countries have beenrelatively spared (11). In 2005, the first majoroutbreak of a CTX-M-15-producing Klebsiellapneumoniae in Sweden was identified at UppsalaUniversity Hospital (12). During a 3-year period,it involved approximately 300 patients. Beforethe outbreak, AmpC was the predominatingenzyme among E. coli isolates with resistance tocephalosporins. After the outbreak, there was asteadily increasing isolation frequency of ESBL-producing E. coli at the hospital, a scenario simi-lar to that described in Spain and UK (10, 13).The aim of this study was to explore the

basis for the observed change in the epide-miology of multiresistant E. coli at UppsalaUniversity Hospital by characterizing isolatescarrying either ESBL or AmpC enzymes.

MATERIALS AND METHODS

Setting

Uppsala University Hospital is a tertiary hospitalwith about 1100 beds and a catchment area of 1.4million people. Approximately 52 000 patients areadmitted and 650 000 visits are made annually.

The outbreak of the CTX-M-15-producingK. pneumoniae strain started in May 2005, and inOctober 2006 an intervention program was initiatedthat lasted until the end of 2007. The programfocused on the outbreak strain and no other ESBLproducers. It included isolation of patients, rein-forcement of hygienic measures and a change in theantibiotic policy (14, 15).

To follow the epidemiology, the isolation frequen-cies of E. coli and K. pneumoniae isolates resistantto broad-spectrum cephalosporins from urine sam-ples were obtained from January 2006 to June 2009from the database of the Department of ClinicalMicrobiology, Uppsala University Hospital. Thisdatabase keeps record of all analyses performed atthe laboratory from 2005 and forward. E. coli iso-lates resistant to broad-spectrum cephalosporins andwith no clavulanate inhibition were assumed to beAmpC producers. Infections caused by the epidemicE. coli clone O25b-ST131 were notifiable locallyduring the latter part of the study period, and dataobtained by screening for the pabB gene with PCR(16) were also available.

Bacteria and media

All E. coli isolates with resistance to cefotaximeand/or ceftazidime (inhibition zone <24 mm or MIC>1 mg/L) collected from January 2006 throughDecember 2008 at the Department of ClinicalMicrobiology, Uppsala University Hospital, wereincluded in the study. The bacteria were cultured onblood and cysteine lactose-electrolyte-deficient agar(Becton Dickinson and Company, Sparks, MD,USA) at 35 °C for 24–48 h in room atmosphere.They were identified by conventional methods or aVITEK 2 instrument (bioM�erieux, Lyon, France).

The ESBL phenotype was confirmed by disc dif-fusion synergy test (17), and the presence of AmpCwas identified by a method described by Coudronet al. (18) after a slight modification (300 lg insteadof 400 lg of boronic acid was used in the discs).From different years and sources, 23 ESBL-positiveand 23 AmpC-positive isolates were thereafter ran-domly chosen for further analyses.

Antibiotic susceptibility

Antibiotic susceptibility to aztreonam, cefotaxime,ceftazidime, ceftibuten, cefuroxime, piperacillin-tazobactam, imipenem, meropenem, nalidixic acid,tobramycin, gentamicin and trimethoprim-sulpha-methoxazole was tested by the disc diffusion methodon IsoSensitest agar (Oxoid Ltd., Basingstoke, Eng-land), according to the method recommended by theSwedish Reference Group for Antibiotics 2009,SRGA (www.srga.org). Multiresistance was definedas resistance to an extended-spectrum cephalosporinand to at least three other classes of non-b-lactamantibiotics.

DNA preparation and PCR

DNA was prepared by heating a bacterial suspen-sion to 95 °C for 5–10 min. The template was added

60 © 2013 APMIS. Published by John Wiley & Sons Ltd

STARLANDER et al.

to HotStarTaq master mix (Qiagen AB�; Solna,Sweden) with primers (Eurogentech S.A., Seraing,Belgium) in a final volume of 25 lL. The PCRreactions were processed in a GeneAmp PCRSystem 9700 cycler (PE Applied Biosystems, FosterCity, CA, USA), the program consisting of 94 °Cfor 3–12 min, followed by 25–30 cycles of 30 s at94 °C, 30–60 s at a primer-specific temperature,1–3 min at 68–72 °C and a final extension step at72 °C for 5–10 min. The PCR products wereseparated by electrophoresis on 1–1.5% agarose gels(GeneChoice Inc., Frederick, MD, USA). Thepositive controls were provided by the Departmaentof Clinical Microbiology, if not otherwise indicated.Before the use of any positive control, the amplifiedproduct was sequenced to verify its specificity.

Sequencing

The PCR products were purified with the ExoSAP-IT purification kit (Amersham Biosciences, Piscat-away, NJ, USA) and sequenced on both strands,using the BigDye Terminator v3.1 Cycle SequencingKit (Applied Biosystems). The sequencing reactionswere analyzed with the ABI3130 instrument(Applied Biosystems) and the sequences edited withthe Vector NTI Advance 10 software (Invitrogen,Carlsbad, CA, USA). The Basic Local AlignmentSearch Tool, BLAST, at the National Center forBiotechnology Information (http://www.ncbi.nlm.nih.gov) was employed to identify the b-lactamasesand other products. Multiple sequence alignmentswere generated with the BioEdit 7.0 sequence align-ment editor (Ibis Therapeutics, Carlsbad, CA USA).

Molecular detection of b-lactamases

Isolates with ESBL phenotype were screened byPCR for genes encoding blaTEMs, blaSHVs andblaCTX-M phylogenetic lineage groups I–IV (12,19). Genes encoding pAmpC enzymes were detectedin isolates with a positive aminophenylboronic acidtest, using the method described by P�erez-P�erezet al. (8). CTX-M- and pAmpC-positive isolateswere sequenced as described. E. coli isolates with anAmpC phenotype, but no pAmpC genes were con-sidered to harbour AmpC chromosomally.

Plasmid characterization and virulence factors

To characterize the plasmids carried by isolates withESBL or pAmpC, replicon typing was performed asdescribed by Carattoli et al. (20). In addition, integronsbelonging to classes 1 and 2 were detected by PCR,and the variable gene cassette region was amplifiedand sequenced (21). Genes encoding 17 different vir-ulence factors of E. coli were detected by PCR aspreviously described by Johnson et al. (3).

Rep-PCR

Rep-PCR was performed to examine the geneticrelatedness. Included primers were ERIC1R 5′-TGTAAGCTCCTGGGGATTCAC-3′, ERIC2 5′-AGTAAGTGACTGGGGT-GAGCG-3′ and A70-10 5′-CAGACACGCC-3′. The number of cycles was 45 andthe annealing temperature 36 °C. Isolates differingby one DNA band or more were assigned to differ-ent strains (22).

Survival test

Six AmpC-producing and six ESBL-producing E.coli isolates were randomly chosen for the survivaltest. To avoid a genetic bias, only isolates belongingto different strains were included. The inoculum wasprepared by inoculating one colony from an over-night culture into 2 mL of Luria–Bertani (LB) broth(Becton Dickinson) for incubation for 5 h to a via-ble count of 108 colony-forming units. From eachinoculum, 10 lL was spread onto an area measuring1 cm2 on 12 different hard and soft materials com-monly encountered in the hospital environment(Table 1). All hard materials included in the studywere disinfected before use, whereas all soft materi-als were cut into squares of 1 cm2 and sterilized inan autoclave. Sterile surgical gloves were cut understerile conditions. After 1, 3, 7, 14, 21 and 28 daysof inoculation, surviving bacteria were collected fromhard materials with pre-moistened cotton swabs,which were placed into tubes with 2 mL of LBbroth. Soft materials were placed directly into thebroth with sterilized forceps. The cultures were incu-bated overnight, and 10 lL of the broth was there-after pipetted onto blood agar for a new overnightincubation.

Statistical analyses

The Statistics Online Computational Resource wasused. Fisher’s exact test was applied when analyzingantibiotic resistance, whereas Wilcoxon signed ranktest was used when analyzing virulence factors andsurvival. As a substitute for measurement of thearea under the curve, the total number of days ofsurvival on each material was calculated for everybacterial isolate (23). A difference was consideredstatistically significant if p was <0.05.

RESULTS

Epidemiological data

During the study period, a total of 134 ESBL-producing K. pneumoniae isolates fromurine samples were registered. The isolation

© 2013 APMIS. Published by John Wiley & Sons Ltd 61

b-LACTAMASES IN E. COLI

frequency of ESBL-producing K. pneumoniaedecreased markedly after initiation of theintervention program, and the outbreak strainwas mainly replaced by ESBL-producingE. coli. (See Fig. 1). The total number ofE. coli isolates with reduced susceptibility (I-R)to extended-spectrum cephalosporins in urinesamples during the same time period was 225,of which 202 isolates exhibited the ESBL phe-notype and 23 isolates the AmpC phenotype.About one of five of the ESBL-producingisolates belonged to the O25b-ST131 clone.Although AmpC-producing E. coli recovered

slowly, they never reached high numbers. Aftera sharp decline during the second half of 2008,their number was close to zero at the end ofthe observation period.

Studied subpopulation

The bacterial subpopulation randomly chosenfor further exploration was distributed asfollows: urine (n = 19), faeces (n = 19) andwound secretion (n = 8). Nine of the ESBL-positive and ten of the AmpC-positive isolateswere derived from faecal samples.

0

10

20

30

40

50

60

2006-01-01--2006-06-30

2006-07-01--2006-12-31

2007-01-01--2007-06-30

2007-07-01--2007-12-31

2008-01-01--2008-06-30

2008-07-01--2008-12-31

2009-01-01--2009-06-30

Time period

No.

of i

sola

tes

E. coli, ESBL-positive

E. coli, AmpC-positive

K. pneumoniae, ESBL-positive

O25b-ST131-clone positive

Intervention 2006

Fig. 1. Number of Klebsiella pneumoniae and Escherichia coli isolates from urine samples with reduced suscep-tibility (I-R) to broad-spectrum cephalosporins during the study period according to laboratory records atUppsala University Hospital.

Table 1. Survival of ESBL- and AmpC-producing E. coli isolates in days on different hospital-associatedmaterials

AmpC-producing E. coli isolates ESBL-producing E. coli isolates

A B C D E2 F2 G H1 I J K L

Hard materialsMarble 0 0 3 1 1 0 28 3 3 7 7 14Stainless steel 0 3 3 3 1 1 28 28 21 28 21 21Laminate 0 3 3 3 1 3 28 21 14 7 21 14Glass 1 0 7 1 0 0 14 3 1 21 3 3Wood 1 1 3 1 1 1 7 3 0 1 1 1

Soft materialsVinyl glove 7 3 14 14 14 7 28 28 28 28 21 28Synthetic surgical scrub 14 7 14 3 14 7 28 28 14 28 28 14Cotton–polyester coat 3 3 14 3 3 3 28 7 7 7 14 14Polyethene glove 28 28 28 21 28 28 28 28 28 21 28 28Latex surgical glove 3 0 3 3 1 1 21 1 0 14 3 3Nitrile glove 14 14 14 14 21 14 28 28 28 28 28 21Synthetic protection coat 28 28 28 28 28 28 28 28 28 28 28 28

1Belongs to the O25b-ST131 clone.2Plasmid-mediated AmpC.

62 © 2013 APMIS. Published by John Wiley & Sons Ltd

STARLANDER et al.

Among the 23 ESBL-producing isolates,CTX-M-15 predominated (n = 18), followedby CTX-M-14 (n = 5). Four isolates had bothCTX-M-14 and CTX-M-15. The most com-mon combination of b-lactamases was CTX-Mand TEM (57%). In one isolate, the CTX-M-type enzyme was combined with SHV. Twotypes of pAmpC were detected: CMY-2(n = 2) and DHA-1 (n = 1). None of the threeisolates carried TEM- or SHV-type enzymes.Replicon types were identified in all but two

isolates with plasmid-mediated b-lactamases.The IncF plasmids dominated totally amongboth the ESBL and pAmpC producers, andonly class I integrons were detected in the twogroups. Their gene cassettes contained genesencoding resistance to aminoglycosides (ami-noglycoside adenyl transferase) and/or trime-throprim (dihydrofolate reductase). One of theintegrons in a pAmpC isolate had no content.In general, the ESBL-producing E. coli

isolates were more broadly resistant to non-b-lactam antibiotics than AmpC-producingisolates (p = 0.01) (Fig. 2). Multiresistance wasidentified in 11 (48%) of the ESBL-producingisolates. The corresponding figure for AmpCproducers was 2 (9%). Reduced susceptibility(I-R) to piperacillin-tazobactam was observedin 70% of the ESBL producers and in 74% ofthe AmpC producers. None of the isolates wasresistant to carbapenems.

There was a statistically significant difference(p = 0.04) in the number of genes encodingvirulence factors carried by isolates producingESBL (median 4, range 1–10) and AmpC(median 3, range 1–4). The highest number ofgenes was found among isolates derived fromfaeces. The two most frequent virulence geneswere fimH (type 1 fimbrial adhesion) and traT(serum survival gene), and this was indepen-dent of the ß-lactamase produced by theisolate. After these two genes, bmaE (M fimbrialadhesion) followed in AmpC producers, whileiutA and fyuA (both siderophores) were morecommon in isolates with ESBL production(Fig. 3). There was no difference in the num-ber of virulence genes among the isolatesincluded in the survival test and the rest of thestudied subpopulation (p = 0.6).Clonality was found among both ESBL and

AmpC producers. In the subpopulation stud-ied, five (22%) of the ESBL- and one (4%) ofthe cAmpC-producing isolates belonged to theO25b-ST131 clone. The ESBL-producing iso-lates belonging to the O25b-ST131 clone didnot carry more virulence factors than the restof the ESBL-producing isolates (p = 0.19), butall were multiresistant (p = 0.01). Among thecAmpC-producing isolates, two small clusterswere identified with Rep-PCR. The largestconsisted of five isolates and the other oftwo isolates. No other clusters than the

30%

17%

39%

52%

65%

87%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Nalidixic acid Aminoglycosides Trimethoprim-sulphamethoxazole

% o

f iso

late

s

AmpCESBL

Fig. 2. Percentage of isolates in the studied subgroup (n = 46) with reduced susceptibility to non-beta-lactamsin relation to the type of b-lactamase. In the aminoglycoside group, the isolate exhibited resistance to genta-micin, tobramycin or both.

© 2013 APMIS. Published by John Wiley & Sons Ltd 63

b-LACTAMASES IN E. COLI

O25b-ST131 clone was found in the ESBL-producing group.

Survival in the environment

The survival of the ESBL- (n = 6) and theAmpC-producing (n = 6) E. coli strains on dif-ferent materials is shown in Table 1. Marblewas the material on which the isolates survivedfor the shortest time period. The longestsurvival period was observed for synthetic pro-tection coats and polyethene exam gloves. TheESBL-producing isolates had a significantlylonger survival (p = 0.03) on all materialscompared with the AmpC-producing isolates,but a higher number of virulence factorsamong the ESBL-producing isolates tended toreduce the ability to survive in the environ-ment (p = 0.05). The median survival time ofESBL-producing isolates was 10.5 days (lower/upper quartiles = 3/21 days) on hard materialsand at least 28 days (14/28 days) on soft mate-rials. The corresponding figures for AmpC-producing isolates were 1 day (0.25/3 days)and 14 days (3/26 days), respectively. Nomajor differences were noticed when pAmpC

and cAmpC isolates were compared. Thefittest strain of them all was derived from aurine sample. It stayed alive for 28 days onnine of the twelve materials tested. It carriedCTX-M-14, TEM, dfrA14, fimH and fuyA.

DISCUSSION

In this study, 46 E. coli isolates with a reducedsusceptibility to broad-spectrum cephalospo-rins were characterized to better understandthe local epidemiology at Uppsala UniversityHospital. We found ESBL-producing isolatesto carry a higher number of virulence genes, tobe multiresistant more often and to persist fornotably longer time periods on materials fromthe hospital environment than AmpC produc-ers. Virulence factors and antibiotic resistanceare well-known selective traits, but persistencein the environment, especially on textiles andplastic gloves, seem to not have attracted theattention it probably deserves.The epidemiology of ESBL-producing

Enterobacteriaceae depicted in this study wassimilar to that described by Cant�on (10)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

fimH

traT

fyuA

iutA

bmaE PAI

HPIibe

Acv

aC

afa/dr

aBC

sfa/fo

cDE

hlyA

focG

cnf1

sfaS

Genes encoding virulence factors

% o

f iso

late

s

AmpCESBL

Fig. 3. Number and type of genes encoding virulence factors among E. coli isolates with ESBL or AmpC pro-duction. The following genes were investigated: fimH – type 1 fimbriae (adhesin); traT – serum survival gene(surface lipoprotein); fuyA – yersiniabactin (siderophore); iutA – aerobactin (siderophore); bmaE – M fimb-riae (adhesin); PAI – pathogenicity island; HPI – high-pathogenicity island, ibeA – protein associated withendothelial cell invasion; cvaC – colicin V (bacteriocin); afa/draBC – M fimbriae (adhesin); sfa/focDE – F1Cfimbrial subunit (adhesin); hlyA – haemolysin (toxin); focG – F1C fimbriae (adhesin); cnf1 – cytotoxic necro-ticizing factor (toxin); sfaS – S fimbriae (adhesin).

64 © 2013 APMIS. Published by John Wiley & Sons Ltd

STARLANDER et al.

and Livermore (13); after an initial increase inthe isolation frequency of ESBL-producingK. pneumoniae, ESBL-producing E. coli tookover. Without doubt, the O25b-ST131 clonehas been very successful. The increases inESBL-producing E. coli could to some extentbe explained by the entry of this pandemicclone on the scene. However, most ESBL-producing E. coli isolates included in thisstudy carried more selective traits than theirAmpC-producing counterparts. Furthermore,the O25b-ST131 clone was represented amongthe AmpC producers, but that was not enoughto balance the situation. AmpC-producing iso-lates belonging to the O25b-ST131 clone havebeen reported by other research groups, butmost seem to be sporadic findings (5, 24–26).Only three of the isolates exhibiting the

AmpC phenotype carried the b-lactamaseencoding gene on the plasmid, which is in linewith other Scandinavian studies (26, 27). Oneof the pAmpCs was typed to DHA-1. This isthe first time a DHA-type enzyme is reportedfrom Sweden and is a rare finding in Scandina-via (26, 27). The plasmids of the ESBL andpAmpC producers had identical or similar rep-licon types. In addition, the integrons were allof class I and had similar contents. These find-ings indicate that similar IncF plasmids are cir-culating in different E. coli populations (28–31),but those harbouring certain CTX-M-typeenzymes appear to be better equipped withtraits necessary for dissemination on a globalscale.The quantity and quality of genes encoding

virulence factors may affect colonization andinfection rates and therefore the epidemiology.ESBL producers had significantly more viru-lence genes, but the clinical importance ofquantity was not quite clear as the highestnumber of virulence genes was carried by iso-lates from faecal samples and not by isolatesfrom infectious sites. Furthermore, the carriageof very large numbers of virulence genes, asobserved in some ESBL-producing isolates,appeared to be associated with a reduced fit-ness to survive in the environment. Not onlythe quantity but also the quality of virulencegenes differed between ESBL and AmpC pro-ducers. The ESBL-producing isolates favouredsiderophores, whereas the AmpC-producing

isolates had a relatively large number of Mfimbriae adhesion genes.A major difference among the two E. coli

groups was found when the survival in the envi-ronment was explored. Although only a limitednumber of isolates were compared, clear ten-dencies were recognized. In accordance withprevious studies (32, 33), there was a differencebetween E. coli isolates in their ability to surviveon different materials, but more importantly, theESBL-producing E. coli isolates tended to sur-vive much longer than the AmpC-producingisolates. This is likely a part of the CTX-M suc-cess: by surviving outside its host for anextended time, the bacterium can to be trans-mitted from patient to patient for monthsthrough contaminated objects, including clothesand other soft materials. Although the survivalof AmpC producers was relatively poor, theystill managed to spread among a small numberof patients. Without prompt and continuoususe of disinfectants, it can be difficult to inter-rupt transmission chains among bacteria evenof relatively competitiveness.In conclusion, although the two groups of

E. coli isolates had several features in com-mon, those producing ESBLs surpassed theAmpC-producing isolates in antibiotic resis-tance, virulence and survival in the environ-ment. Extended survival in the environment isprobably a selective trait for epidemicallysuccessful strains. Further studies in this areaare needed, but it is possible that the dissemi-nating potential of a multiresistant strain canbe identified without advanced molecular tech-niques at any microbiology laboratory byinvestigating the survival with a simple test.

FUNDING

This study was financially supported by theUppsala-€Orebro Regional Research Counciland the Collaborative Research Council inUppsala-V€aster�as.

TRANSPARENCY DECLARATION

There are no arrangements that could beconsidered to pose a conflict of interest.

© 2013 APMIS. Published by John Wiley & Sons Ltd 65

b-LACTAMASES IN E. COLI

REFERENCES

1. Ronald A. The etiology of urinary tract infec-tion: traditional and emerging pathogens. Am JMed 2002;113(Suppl 1A):14S–9S.

2. Bearman GM, Wenzel RP. Bacteremias: a lead-ing cause of death. Arch Med Res 2005;36:646–59.

3. Johnson JR, Stell AL. Extended virulencegenotypes of Escherichia coli strains frompatients with urosepsis in relation to phylogenyand host compromise. J Infect Dis 2000;181:261–72.

4. Peirano G, Pitout JD. Molecular epidemiologyof Escherichia coli producing CTX-M beta-lactamases: the worldwide emergence of cloneST131 O25:H4. Int J Antimicrob Agents2010;35:316–21.

5. Rogers BA, Sidjabat HE, Paterson DL. Escheri-chia coli O25b-ST131: a pandemic, multiresis-tant, community-associated strain. J AntimicrobChemother 2011;66:1–14.

6. Paterson DL. Recommendation for treatment ofsevere infections caused by Enterobacteriaceaeproducing extended-spectrum beta-lactamases(ESBLs). Clin Microbiol Infect 2000;6:460–3.

7. Jaurin B, Grundstrom T, Edlund T, NormarkS. The E. coli beta-lactamase attenuator medi-ates growth rate-dependent regulation. Nature1981;290:221–5.

8. Perez-Perez FJ, Hanson ND. Detection of plas-mid-mediated AmpC beta-lactamase genes inclinical isolates by using multiplex PCR. J ClinMicrobiol 2002;40:2153–62.

9. Pitout JD, Laupland KB. Extended-spectrumbeta-lactamase-producing Enterobacteriaceae:an emerging public-health concern. LancetInfect Dis 2008;8:159–66.

10. Canton R, Coque TM. The CTX-M beta-lactamase pandemic. Curr Opin Microbiol2006;9:466–75.

11. European Centre for Disease Prevention andControl. Antimicrobial resistance surveillance inEurope 2010. Annual Report of the EuropeanAntimicrobial Resistance Surveillance Network(EARS-Net). Stockholm, 2011.

12. Lytsy B, Sandegren L, Tano E, Torell E,Andersson DI, Melhus A. The first majorextended-spectrum beta-lactamase outbreak inScandinavia was caused by clonal spread of amultiresistant Klebsiella pneumoniae producingCTX-M-15. APMIS 2008;116:302–8.

13. Livermore DM, Hawkey PM. CTX-M: chang-ing the face of ESBLs in the UK. J AntimicrobChemother 2005;56:451–4.

14. T€angden T, Eriksson BM, Melhus �A, SvennbladB, Cars O. Radical reduction of cephalosporinuse at a tertiary hospital after educational

antibiotic intervention during an outbreak ofextended-spectrum beta-lactamase-producingKlebsiella pneumoniae. J Antimicrob Chemo-ther 2011;66:1161–7.

15. Ransj€o U, Lytsy B, Melhus �A, Aspevall O,Artinger C, Eriksson BM, et al. Hospitaloutbreak control requires joint efforts from hos-pital management, microbiology and infectioncontrol. J Hosp Infect 2010;76:26–31.

16. Clermont O, Dhanji H, Upton M, Gibreel T,Fox A, Boyd D, et al. Rapid detection of theO25b-ST131 clone of Escherichia coli encom-passing the CTX-M-15-producing strains. JAntimicrob Chemother 2009;64:274–7.

17. Jarlier V, Nicolas MH, Fournier G, PhilipponA. Extended broad-spectrum beta-lactamasesconferring transferable resistance to newer beta-lactam agents in Enterobacteriaceae: hospitalprevalence and susceptibility patterns. RevInfect Dis 1988;10:867–78.

18. Coudron PE. Inhibitor-based methods for detec-tion of plasmid-mediated AmpC beta-lactamasesin Klebsiella spp., Escherichia coli, and Proteusmirabilis. J Clin Microbiol 2005;43:4163–7.

19. Pitout JD, Hossain A, Hanson ND. Phenotypicand molecular detection of CTX-M-beta-lactamasesproduced by Escherichia coli and Klebsiella spp.J Clin Microbiol 2004;42:5715–21.

20. Carattoli A, Bertini A, Villa L, Falbo V, Hop-kins KL, Threlfall EJ. Identification of plasmidsby PCR-based replicon typing. J MicrobiolMethods 2005;63:219–28.

21. Machado E, Canton R, Baquero F, Galan JC,Rollan A, Peixe L, et al. Integron content ofextended-spectrum-beta-lactamase-producingEscherichia coli strains over 12 years in a singlehospital in Madrid. Spain. Antimicrob AgentsChemother 2005;49:1823–9.

22. van Belkum A, Tassios PT, Dijkshoorn L, Ha-eggman S, Cookson B, Fry NK, et al. Guidelinesfor the validation and application of typingmethods for use in bacterial epidemiology. ClinMicrobiol Infect 2007;13(Suppl 3):1–46.

23. Johansson S, Melhus H. Vitamin A antagonizescalcium response to vitamin D in man. J BoneMiner Res 2001;16:1899–905.

24. Simner PJ, Zhanel GG, Pitout J, Tailor F,McCracken M, Mulvey MR, et al. Prevalenceand characterization of extended-spectrum beta-lactamase- and AmpC beta-lactamase-producingEscherichia coli: results of the CANWARD2007–2009 study. Diagn Microbiol Infect Dis2011;69:326–34.

25. Oteo J, Cercenado E, Cuevas O, Bautista V,Delgado-Iribarren A, Orden B, et al. AmpCbeta-lactamases in Escherichia coli: emergence ofCMY-2-producing virulent phylogroup D iso-lates belonging mainly to STs 57, 115, 354, 393,and 420, and phylogroup B2 isolates belonging to

66 © 2013 APMIS. Published by John Wiley & Sons Ltd

STARLANDER et al.

the international clone O25b-ST131. DiagnMicrobiol Infect Dis 2010;67:270–6.

26. Naseer U, Haldorsen B, Simonsen GS, SundsfjordA. Sporadic occurrence of CMY-2-producingmultidrug-resistant Escherichia coli of ST-com-plexes 38 and 448, and ST131 in Norway. ClinMicrobiol Infect 2010;16:171–8.

27. Jørgensen RL, Nielsen JB, Friis-Møller A, Fjeld-søe-Nielsen H, Schønning K. Prevalence andmolecular characterization of clinical isolates ofEscherichia coli expressing an AmpC phenotype.J Antimicrob Chemother 2010;65:460–4.

28. Carattoli A, Miriagou V, Bertini A, Loli A,Colinon C, Villa L, et al. Replicon typing ofplasmids encoding resistance to newer beta-lactams. Emerg Infect Dis 2006;12:1145–8.

29. Hopkins KL, Liebana E, Villa L, Batchelor M,Threlfall EJ, Carattoli A. Replicon typing of plas-mids carrying CTX-M or CMY beta-lactamasescirculating among Salmonella and Escherichia

coli isolates. Antimicrob Agents Chemother 2006;50:3203–6.

30. Marcade G, Deschamps C, Boyd A, Gautier V,Picard B, Branger C, et al. Replicon typing ofplasmids in Escherichia coli producing extended-spectrum beta-lactamases. J Antimicrob Chemo-ther 2009;63:67–71.

31. Naseer U, Haldorsen B, Tofteland S, HegstadK, Scheutz F, Simonsen GS, et al. Molecularcharacterization of CTX-M-15-producing clini-cal isolates of Escherichia coli reveals the spreadof multidrug-resistant ST131 (O25:H4) andST964 (O102:H6) strains in Norway. APMIS2009;117:526–36.

32. Neely AN. A survey of gram-negative bacteriasurvival on hospital fabrics and plastics. J BurnCare Rehabil 2000;21:523–7.

33. Scott E, Bloomfield SF. The survival and transferof microbial contamination via cloths, hands andutensils. J Appl Bacteriol 1990;68:271–8.

© 2013 APMIS. Published by John Wiley & Sons Ltd 67

b-LACTAMASES IN E. COLI