Zoonotic potential of multidrug-resistant extraintestinal pathogenic Escherichia coli obtained from healthy poultry carcasses in Salvador, Brazil

The

Brazilian

Journal

of

INFECTIOUS

DISEASES

w w w .e l s e v i e r . c o m / l o c a t e / b j i d

Original

article

Zoonotic

potential

of

multidrug-resistant

extraintestinal

pathogenic

Escherichia

coli

obtained

from

healthy

poultry

carcasses

in

Salvador,

Brazil

José

Vitor

Lima-Filho

_,

_Liliane

_Vilela

_Martins

_,

Danielle

Cristina

de

Oliveira

Nascimento

_,

_Roberta

_Ferreira

_Ventura

_,

Jacqueline

Ellen

Camelo

Batista

_,

_Ayrles

_Fernanda

_Brandão

_Silva

_,

Maria

Taciana

Ralph

_,

_Renata

_Valenc¸a

_Vaz

_,

_Carlos

_Boa-Viagem

_Rabello

_,

Isabella

de

Matos

Mendes

da

Silva

_,

_Joaquim

_{Evêncio-Neto}

a_{DepartmentofBiology,UniversidadeFederalRuraldePernambuco,Recife,PE,Brazil}

b_{DepartmentofZootechny,UniversidadeFederalRuraldePernambuco,Recife,PE,Brazil}

c_{DepartmentofMorphologyandAnimalPhysiology,UniversidadeFederalRuraldePernambuco,Recife,PE,Brazil}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received4July2012 Accepted12September2012 Availableonline3January2013

Keywords:

ExPEC MDRstrains

Increasedsurvivalgene Experimentalinfections

a

b

s

t

r

a

c

t

Thezoonoticpotentialtocausehumanand/oranimalinfectionsamongmultidrug-resistant extraintestinalpathogenicEscherichiacolifromavianoriginwasinvestigated.Twenty-seven extraintestinalpathogenicE.coliisolatescontainingtheincreasedsurvivalgene(iss)were obtainedfromtheliversofhealthyanddiseasedpoultrycarcassesattwoslaughterhouses inSalvador,northeasternBrazil.Theantimicrobialresistance-susceptibilityprofileswere conductedwithantibioticsofavianand/orhumanusebythestandardizeddisc-diffusion method.Antimicrobialresistancewashigherforlevofloxacin(51.8%),amoxicillin/clavulanic acid(70.4%), ampicillin(81.5%), cefalotin (88.8%), tetracycline(100%) and streptomycin (100%).Theminimuminhibitoryconcentrationsabovetheresistancebreakpointsof doxy-cycline,neomycin,oxytetracyclineandenrofloxacinreached,respectively,88.0%,100%,75% and91.7%oftheisolates.Strainswithhighandlowantimicrobialresistancewerei.p. admin-isteredtoSwissmice,andhistopathologicalexaminationwascarriedoutsevendaysafter infection.Resistancetogoatandhumanserumcomplementwasalsoevaluated.Theresults showthatSwissmicechallengedwithstrain2B(resistantto11antimicrobials)provokeda severedegenerationofhepatocytesbesideslymphocyticinfiltrationintheliver,whereas thespleenshowedareasofdegenerationofthewhiteandredpulp.Conversely,thespleen andliverofmicechallengedwithstrain4A(resistanttotwoantimicrobials)were morpho-logicallypreserved.Inaddition,complementresistancetogoatandhumanserumwashigh forstrain2Bandlowforstrain4A.Ourdatashowthatmultidrugresistanceand pathogen-esiscanbecorrelatedinextraintestinalpathogenicE.colistrainsobtainedfromapparently healthypoultrycarcasses,increasingtheriskforhumanpublichealthy.

©2013ElsevierEditoraLtda.Allrightsreserved.

∗ _{Corresponding author at: Laboratório de Microbiologia e Imunologia, Departamento de Biologia, Universidade Federal Rural de}

Pernambuco,R.DomManoeldeMedeiross/n,CampusDoisIrmãos,Recife,PE,52171-900,Brazil.Tel.:+55318133206312. E-mailaddress:[email protected](J.V.Lima-Filho).

1413-8670/$–seefrontmatter©2013ElsevierEditoraLtda.Allrightsreserved. http://dx.doi.org/10.1016/j.bjid.2012.09.004

Introduction

ThespreadofmultidrugresistanceamongavianEscherichiacoli

isusuallyattributedtotheselectivepressureexertedbythe antimicrobialsincludedinbroilerfeedforthepast60years.1 Johnson et al.2 _{reinforced this suspicion since} antibiotic-resistant and antibiotic-susceptible E. coli isolates from retailpoultryproductshadsimilarphylogeneticbackground, and otherwise emerged from the same source population. Regardingantimicrobialexposureinpoultryproduction, mul-tidrug resistance among extraintestinal pathogenic E. coli

(ExPEC)alliedtocommunityacquiredinfectionsisincreasing inprevalenceinmanypartsoftheglobe.3_{Theindiscriminate} useofantimicrobialsishighindevelopingcountries,where humanantibioticsare often accessible fornon-therapeutic usesinhealthyanimals.4_{Therefore,colonizationof} asymp-tomaticpoultrybymultidrugresistantExPECaugmentsthe probabilityofresistancegeneacquisitionbyhumanstrains throughthefood-chain.

Avianpathogenic E.coli(APEC)iscommonlyreported as anExPECpathotype,butitsgenotypeisnotclearlydefined. Accordingto Kwonet al.5 _{atleast fivevirulencegenes are} present inAPEC, many ofthemfound inplasmid pTJ100.6 However, the increased survival gene (iss) has been iden-tified as a virulence marker to distinguish between avian andhumanExPEC.7_{Thisgeneexertsanti-complement} resis-tance and has been found in conserved regions of ColV and ColBMplasmids,8,9 _{which are often identified in APEc} strains.6,10_{RecentfindingsalsosuggestthatAPECandhuman} neonatalmeningitis E. coli(NMEC) are subpathotypesfrom ExPEC involvedin pathogenesis of bothavian and human infections.11 _{Therefore, hybrid plasmids (ColV- and} ColBM-associatedplasmids)harboringanumberofdistinctvirulence genesandMDR-encodingislandscanalsobepresentinExPEC isolates.11 Consequently,it isnow clearthatselected high-virulentandmultidrugresistantExPECstrainsofavianorigin represent subpathotypes ofgreat danger to humanpublic health.

Brazilhasrankedfirstintheworldinexportsofpoultry meatsince2004.12 _{TypicalselectionsofpoultryinBrazilian} slaughterhousestake into account macroscopicalterations thatcanresultindiscardingcarcasses.13_{However,sub-clinical} symptomsarenotalwaysperceivedand oftendonotshow aclueofthepresenceofpathogenicmicroorganisms.Thus, theriskofvirulenceanddrugresistancegenetransmission between avian E. coli obtained trough the food-chain and intestinal commensal E. coliis increasing. The goal of the presentstudywastoinvestigatethe pathogenesisto mam-malian hostsofselected multidrug resistant ExPECstrains obtainedfromhealthypoultrycarcassesinSalvador,Brazil.

Material

and

methods

ExtraintestinalpathogenicE.colistrainswereobtainedfrom theliversofpoultrycarcasseswith(n=9)orwithout(n=18) macroscopicalterationsattwoslaughterhousesinSalvador,

Table1–ExPECstrainsobtainedfromtheliverofpoultry carcasses.

E.coli Macroscopicaspect Presenceof geneissa Presenceof genestxb 4A Noalteration + − 5A Noalteration + − 36A Noalteration + − 41A Noalteration + − 43A Noalteration + + 44A Noalteration + − 2B Noalteration + − 5B Noalteration + − 37B Noalteration + − 38B Noalteration + + 39B Noalteration + − 42B Noalteration + − 15C Noalteration + − 21C Noalteration + − 42C Noalteration + − 32 Noalteration + − 35 Noalteration + − 46 Noalteration + +

24A Salmonellasepticemia + −

30C Salmonellasepticemia + − 31C Salmonellasepticemia + − 48A Ascitis + − 48B Ascitis + − 52B Cachexy + − 54B Cachexy + − 60A Colibacillosis + − 55B Colibacillosis + −

a _{Increasedserumsurvivalgene.} b _{Shiga-liketoxingene.}

capital of the state of Bahia, northeast Brazil (Table 1). Carcasseswithoutmacroscopicalterationswere considered healthy andapprovedforhumanconsumption.Theorgans werecollected understerileconditions.Allbacterialstrains wereidentifiedthroughbiochemicaltests,andthepresenceof theissgenewasdeterminedaccordingtothemethodreported inapreviousstudy.14_{Theisolatesweremaintainedintubes} containingbrainheartinfusionagarat8◦Cuntiluse.

Antimicrobialresistance-susceptibilityprofileofExPEC strains

The resistance-susceptibility profiles to antimicrobials of avianorhumanuseweredeterminedbythestandard disc-diffusion method, following the recommendations of the ClinicalandLaboratoryStandardsInstitute(formally,National CommitteeforClinicalLaboratoryStandards).15_{Briefly,filter} paperdiscs5mmindiameter impregnatedwithantibiotics (Cecon) were added to cultures in Petridishes (0.5 on the McFarland scale, corresponding to 108_{cells/mL) containing} Mueller–Hintonagar(Oxoid).After24hincubationat35◦C,the diameteroftheinhibitionzonewasmeasuredwithacaliper. Alltestswerecarriedoutinduplicateagainst13 antimicro-bialsandtheresultswereinterpretedassensitive,moderately sensitive or resistant. Thebreakpoints for resistancewere thoserecommendedbytheCLSI.Theoverallresistancerate was calculated as the number of non-susceptible isolates

Table2–AntimicrobialsauthorizedbyBrazilian authoritiesusedasgrowthpromotersinbroilerfeed.

Antimicrobialclass Antimicrobials Dosage(g/ton) Oligosaccharide Avilamicin 2.5–10 Peptide Bacitracinmethylene

disalicylate

4–55 Peptide Zincbacitracin 4–55 Benzenederivative Chlorhexidine 10–20 Macrolide Spiramycin 5 Peptide Enramycin 3–10 Phosphoglycolipids Flavomycin 1–2 Lincosamide Lincomycin 2.2–4.4 Peptide Colistinsulfate 2–10 Streptogramin Virginiamycin 5.5–16.5 Quinolone Clorohidroxiquinolin 15–30 Macrolide Tylosin

tartrate/phosphate

4–55

Source:BrazilianMinisteryofAgriculture.

dividedbythetotalnumberofisolates.Multidrugresistance wasdeterminedwhenbacterialisolateswereresistanttofour antimicrobialsofatleastthreedifferentclasses.Theisolates obtainedfrom a singlesample withan identical antibiotic resistance/sensitivityprofileweretreatedasasinglestrain. Inthiscase,bacterialreplicateswerenotconducted.

Minimuminhibitoryconcentrationoftypical antimicrobialsusedonBrazilianfarms

The minimum inhibitory concentration (MIC) of antimi-crobials commonly used on Brazilian farms (doxycycline, neomycin,oxytetracyclineandenrofloxacin) wasmeasured throughthebrothdilutionmethodinconcentrationsranging from 1.9to1000␮g/mL.16 _{TheMIC wasthe lowest} concen-tration that caused visible inhibition ofgrowth, while the minimum bactericidal concentration(MBC) wasthe lowest concentration resulting in no growth after the incubation periodof24hat37◦C.Allassayswereperformedinduplicate with25ExPECstrainsresistantto4–11antimicrobialsthrough thedisc-diffusionmethod.Alistofantimicrobialsauthorized bytheBraziliangovernmenttobeincludedinbroilerfeedis showninTable2.

Complementresistanceassay

Thecomplementresistancetestwascarriedoutwithselected ExPECstrainsobtainedfromhealthycarcasses,andwith dis-tinctdrug resistanceprofile,followingthe methodadapted fromSamuelsenetal.17_{Samplesofbloodwereobtainedfrom} goats and a healthy humanvolunteer under sterile condi-tions and allowedtocoagulate. Theblood was centrifuged (7000rpm/5min)andbloodserumwasseparatedintoanew tube.Briefly,190␮Loftheserumplus10␮LofE.colistrains (107_{cells/mL)wereculturedtogetherinwellsofsterileElisa} plates,andincubatedat37◦Cfor180min.Then,aliquotsof 10␮Lweresampledattimes0,60,120and180min,andadded toPetridishescontainingMacConkeyagarforenumeration ofthecolonyformingunits.Theabsenceofspecific

antibod-iesfortheExPECstrainsusedinthisstudywasconfirmedby

invitroagglutinationtests.

Animals

AdultmaleSwissmiceweighingapproximately35g,obtained fromKeizo-AzamiImmunopathologyLaboratory(LIKA/UFPE), wereused.Theanimalswerekeptinananimalhousewithfree accesstowaterandcommercialsterilediet(Purina,Paulínia, SP,Brazil).Themicewere handledaccordingtoestablished experimentalprocedures.

ExperimentalinfectionwithselectedExPECstrains

Themicewereseparated intothreegroups(n=5)and chal-lenged by the intraperitoneal (i.p.) route with 0.2mL of bacterial suspensionscontaining106_{CFU/mL. Experimental} infectionswerecarriedoutwithstrains4A,41Aand2B. Clin-icalsymptomssuchasprostration,weightlossandmortality wereobserveddailyforsevendayspost-infection.Survivors weresacrificedunderanesthesiawithHalothane(Halocarbon Laboratories, USA).Sections ofthe liver were submittedto enumerationofcolonyformingunits(CFU/g)andhistological examination.

Enumerationofcolonyformingunits

Thespleensandliversweredissected,weighedand macera-tedinPBS(1:10or1:100,w/v)understerileconditions.Serial decimaldilutionsweremadeand0.1mLaliquotswereplated ontoMacConkeyagar(Oxoid).Thecolonieswerecountedafter incubationat37◦Cfor24handtheresultsexpressedasCFU/g oforgan.

Histologicalexamination

Tissues were fixedin10% formaldehyde andprocessedfor paraffin embedding.Thesections (5␮m) werestained with hematoxylin–eosinandtheslideswerecodedandexamined byasinglepathologist,whowasunawareoftheexperimental conditionsofeachgroup.

Statisticalanalyses

Thestatisticalsignificanceofdatawasassessedbyanalysis ofvariance(ANOVA),followedbyStudent’st-test.Thelevelof significancewasdeterminedasp<0.05.

Results

The antimicrobial resistance-susceptibility profiles of 27 extraintestinal pathogenic E. coli strains from carcassesof healthy and diseased poultry are shown in Table 3. The majority of ExPEC were resistant to at least four antibi-oticsfromdifferentclasses.Themostprevalentphenotypes wereresistanttolevofloxacin(51.8%),amoxicillin/clavulanic acid(70.4%),ampicillin(81.5%),cefalotin(88.8%),tetracycline (100%)andstreptomycin(100%).Theoverallmultidrug resis-tancevariedfrom4to11antimicrobialsandreached92.6%of

Table3–Resistance-susceptibilityprofilesofExPECstrainstoantimicrobialsofavianorhumanuseinBrazil.

Antimicrobialclass Antimicrobials Disc content (␮g) Use: Avian(A) Human(H) Resistance breakpoint (mm) Healthy carcasse Diseased carcasse Overall Resistance %b Resistant strains S%a _{I% R% S% I% R%}

Penicillin Ampicillin 10 A/H ≤13 11.2 5.5 83.3 11.3 11 77.7 81.5 44A,15C,48B, 36A,43A,5B, 21C,42C,35, 31C,52B,55B, 5A,24A,54B, 660A,38B,37B, 339B,46,41A, 22B

Aminoglycoside Amikacin 30 A/H ≤14 72.4 11 16.6100 0 0 1.1 36A,42B,2B Gentamicin 10 A/H ≤12 61.3 11 27.7 56.6 0 44.4 33.3 32,48B,52B, 55B,42B,54B, 37B,41A,2B

Streptomycin 10 H ≤11 0 0 100 0 0 100 100 4A,48A,32,

44A,15C,30C, 48B,36A,43A, 5B,21C,42C, 35,31C,52B, 55B,5A,42B, 24A,54B,60A, 38B,37B,39B, 46,41A,2B Fluoroquinolone Ciprofloxacin 5 A/H ≤15 22.3 27.750 67.7 0 33.3 44.4 30C,43A,35,

42B,24A,60A, 38B,37B,39B, 46,41A,2B Levofloxacin 5 H ≤13 27.9 11 61.1 55.6 11.1 33.3 51.8 15C,30C,43A, 42C,35,5A, 42B,24A,60A, 38B,37B,46, 41A,2B

Phenicol Chloramphenicol 30 A/H ≤12 66.8 5.5 27.7 88.9 11.1 0 18.5 37B,39B,46, 41A,2B Cephalosporin Ceftazidim 30 H ≤14 56.7 27.716.6 66.7 22.2 11.1 14.8 54B,39B,46, 41A Cefalotin 30 H ≤14 5.7 5.5 88.8 0 11.1 88.9 88.8 32,44A,15C, 30C,48B,36A, 43A,5B,21C, 42C,31C,52B, 55B,5A,42B, 24A,54B,60A, 38B,37B,39B, 46,41A,2B Carbapenem Imipenem 10 H ≤13 100 0 0 100 0 0 0

-Tetracycline Tetracycline 30 A/H ≤14 0 0 100 0 0 100 100 4A,48A,32, 44A,15C,30C, 48B,36A,43A, 5B,21C,42C, 35,31C,52B, 55B,5A,42B, 24A,54B,60A, 38B,37B,39B, 46,41A,2B Monobactam Aztreonam 30 H ≤15 33.8 22.244 44.5 44.4 11.1 33.3 5B,21C,31C, 5A,54B,38B, 9B,46,41A, 2B

Table3–(Continued)

Antimicrobialclass Antimicrobials Disc content (␮g) Use: Avian(A) Human(H) Resistance breakpoint (mm) Healthy carcasse Diseased carcasse Overall Resistance %b Resistant strains S%a _{I% R% S% I% R%} Beta-lactam/beta-lactamase inhibitor Amoxycillin/ clavulanicacid 20/10 H ≤13 22.3 5.5 72.233.4 0 66.6 70.4 44A,36A,5B, 21C,42C,35, 31C,52B,55B, 5A,24A,54B, 60A,38B,37B, 39B,46,41A, 2B

a _{PercentageofE.colistrainssensitive(S),intermediarysensitive(I)orresistant(R)outofthetotalofisolates.}

b _{Theoverallresistancerateisgivenbythenumberofnon-susceptibleisolatesdividedbythetotalnumberofisolatessubmittedtoantibiogram}

tests.

E.colistrains.Inaddition,40.7%weresimultaneouslyresistant tostreptomycin,levofloxacin,ciprofloxacinandtetracycline. The proportion of highly multidrug resistant strains (8–11 antimicrobials) reached22.2%.Conversely, the aminoglyco-sideamikacin ofavianand humanusewere veryeffective against89.9%ofExPEC.TheMICandMBCoftypical antimi-crobialsusedinBrazilianfarmswere determinedforthose ExPECstrainsresistanttoatleast4antimicrobials.Thelevel ofresistanceagainstdoxycycline,neomycin,enrofloxacinand oxytetracyclinereached,respectively,88.0%,100%, 75%and 91.7%ofthestrains(Table4).

Thecomplement resistancetogoat serum was highfor strains 2Band46 andlowforstrains4Aand41A, whereas resistancetohumanserumwashighforstrains32,46 and 2Bandlowforstrains44and4A(Fig.1).Experimentscarried outwithlaboratoryanimalshaveshownthatSwissmice chal-lengedwithstrains4A,41Aand2Bcansurvivesevendaysafter infection.Afterthistime,thebacterialloadinthespleenand theliverreachedrespectively:3.74×103_and8.54×103_CFU/g for strain 4A; 4.8×102 _{and 1.56×10}3_{CFU/g forstrain 41A,} and 4.0×101 _{and 1.2×10}4_{CFU/g for strain 2B. However,} strain2Bprovokedseveredegenerationofhepatocytesbesides

Table4–ResistanceofmultidrugresistantExPECstrainsfrompoultrycarcassestoantimicrobialscommonlyusedin Brazilianfarms.

Strain Antimicrobials(␮g/mL)

Doxycycline Neomycin Enrofloxacin Oxytetracycline

MIC MBC MIC MBC MIC MBC MIC MBC

4A 125 – >1000 – 1.9 31.2 >1000 – 48A 31.2 – >1000– – 1.9 250 >1000 – 32 >1000 – 500 500 62.5 1000 250 250 44A 62.5 500 >1000 – 15.6 250 >1000 – 15C 62.5 1000 >1000 – 62.5 – 500 – 30C 1.9 250 500 500 1.9 1.9 1.9 500 48B 125 125 125 125 1.9 62.5 >1000 – 36A 1.9 500 250 500 1.9 500 >1000 – 43A 62.5 500 62.5 250 31.2 250 250 – 5B 62.5 500 >1000 – 1.9 1000 250 – 21C 125 500 500 1000 15.6 – 500 500 42C 31.2 – 125 250 3.9 500 1000 – 35 125 250 500 500 31.2 – 62.5 500 31C 125 500 250 250 15.6 1000 >1000 – 52B 31.2 – 500 – 7.8 500 500 – 55B 62.5 1000 >1000 – 31.2 1000 250 – 24A 62.5 250 >1000 – 125 – >1000 – 54B 62.5 250 >1000 – >1000 – 7.8 7.8 60A 62.5 500 >1000 – 15.6 1000 125 – 38B 125 125 >1000 – 62.5 125 125 125 37B 62.5 125 >1000 – 31.2 62.5 62.5 – 39B 15.6 62.5 500 500 7.8 7.8 >1000 – 46 62.5 125 500 500 31.2 250 >1000 – 41A 62.5 62.5 >1000 – NT NT NT NT 2B 125 125 125 125 31.2 31.2 1000 – Resistancebreakpoint ≥16 ≥32 ≥2 ≥16

7 6 5 4 3 2 1 0 60 120 Time (min) Human serum 180 41A 46 4A 44A 32 2B Log CFU/mL Log CFU/mL 7 6 5 4 3 2 1 0 60 120 Time (min) 180 46 41A 4A 44A 32 2B Goat serum

Fig.1–Resistancetoserumcomplementofmultidrug resistantavianExPECstrainsharboringtheissgene.

lymphocyticandlipidinfiltrationintheliver,whereasspleen showed areas of degeneration of the white and red pulp (Fig. 2Band D).Likewise,the liverofmicechallenged with strain41Ashowedmarkedvacuolizationofhepatocytesand lymphocyticinfiltration,whereas thespleenshowed leuko-cyte infiltration and white pulp without boundaries (data notshown).Conversely,strain4Ashowedhepatocytes, por-talspaceandlobularveinspreservedwithmildlymphocytic infiltration in the liver. Also, the spleen was morphologi-cally preserved with capsule, and white pulp withcentral arteriolebesidesred pulp (Fig. 2A and C). These histologi-calfindingsweregenerallyperceptibleinallmicefromeach animal group independently of the bacterial load in the organs.

Discussion

Poultry-associateddiseasescausedbyExPECcause massive economiclossesinthefoodindustry.18–20_{However,thebroad} use ofantimicrobialsbypoultryfarmersinrecent decades hasled tothe emergence ofmultidrug resistant strains in manypartsoftheglobe.21–24 _{Additionally,amongmultidrug} resistantE.colifromchickensandpigletsinChina,97% har-boredtheissgene,suggestingthisgenecouldalsobeusedas amultidrugresistancemarker.22 InBrazil,apreviousstudy

showedthattheratesofmultidrugresistanceamongavian

E. coli reached 77.5%.25 _{Regarding ExPEC strains harboring} the iss gene, we confirmed that the majority were resis-tant to several antimicrobials beyondthose authorizedfor inclusion in broiler feed. Although the number of strains investigatedwassmall,thelevelsofantimicrobialresistance toampicillin,gentamicin,streptomycin,ciprofloxacin, chlor-ramphenicolandtetracyclinewereveryhighincomparison tothelevelsfoundinarecentstudycarriedoutwith101E.coli

strainsfrombroilersandlayerhenswithcolibacillosis infec-tionsinBangladesh.26

Additionally,wehaveshownthatagreatnumberofavian ExPECareprobably␤-lactamaseproducingstrains,since resis-tance to amoxycillin/clavulanic acid was present in 70.4% of the isolates. Oteo et al.27 _{reported that resistance to} amoxicillin-clavulanic acidisincreasingamongE. colifrom human origin in Spanish hospitals, affecting 5.1% of9090 bloodisolates.Consideringthefood-chain,itisworryingthat theonlyinhibitorydrugagainstallavianExPECiscarbapenem imipinem,restricted tohumanuse. Alsoofparticular con-cernaretheresistanceofavianExPECtofluoroquinolessuch asenrofloxacin,whicharesimilartoantibioticsbeingused inhumanmedicine.28Forinstance,theincreasingresistance ofE.colistrainstociprofloxacinhasbeendetectedinseveral internationalstudies.29,30 _{Enrofloxacinhasbeenwithdrawal} fromnon-therapeuticuseinthepoultryindustryintheUnited States.31 _{InEurope,thedrugcannotbeusedintheanimals} from which eggs are produced for human consumption.32 However,inthis study thelevel ofantimicrobialresistance todoxycycline, neomycin,enrofloxacinand oxytetracycline reached alarminglevels. Althoughwe cannotconfirm that Brazilianfarmersrestricttheiruseofsuchantimicrobialsto treatmentregimens,itwasclearthatthesedrugsshouldno longerbeusedonBrazilianpoultryfarms.

AvianExPECstrainshavebeengeneticallyassociatedwith humanE.colicausingurinaryinfections.6_{Additionally,E.coli} strains ofthe EcoRgroupB2 relatedtohumanand animal extraintestinal infections have been detected amongAPEC strains obtainedfrom diseased and healthy chickens.33 In spiteoftheimplicationsofavianExPECincarcassesofhealthy poultryisunclear,aprevious studyinthe UKshowed that samplesofimportedchickenbreastswereoftenpositivefor

E. coliwithCTX-M-2 genotype relatedtohumaninfections inSouthAmerica.34_{Ourdatashowsthatresistanceofavian}

E.colitoserumcomplementlysesinmammalianhostswas notdirectlyrelatedtothepresenceoftheissgene.Thisfinding suggeststhatothergeneticdeterminantsamongE.coliofavian originmustberelatedtohumanserumresistance. Addition-ally,multidrug-resistantExPECstrains41Aand2Bwereoften moreresistanttogoatandhumanserumcomplement,and alsomorevirulenttotheexperimentallyinfectedmicethan strain4A,whichissensitivetothemajorityofantimicrobials. Previous studies have shown the correlation between serum resistance and virulence of E. coli causing diseases in turkeys and chickens.35 _{Moreover, the characterization} a transferable hybrid plasmid pAPEC-O103-ColBM encod-ing multidrug resistance and pathogenicity was recently described.11_{Eventhoughvirulenceandmultidrugresistance} genesarenotuniformlydistributedamongconjugative plas-mids,thediversityofavianExPECwithzoonoticpotentialto

Fig.2–HistologicaldamagetolaboratoryanimalsafterchallengewithavianExPEC.Theslidesoftheliver(AandB)and spleen(CandD)ofSwissmicechallengedwithstrain4A(AandC)and2B(BandD)areshown.Thefullarrow(AandB) indicatesleukocyteinfiltrationwhereasemptyarrowshowsdegenerationofhepatocytes(B).Thefullstarshowsportal space(A)andemptystarshowsdegenerationofredbesideswhitepulp(D).Theasteriskshowsthewhitepulpofthespleen innormalmorphology(C).

causehumandiseasescanbemorefrequentthanperceived. Forexample,inthepresentstudytheriskforhumanhealth wasparticularlyobservedforstrain2B,obtainedfrompoultry carcassesapprovedforhumanconsumption.Thus,anearly diagnosisprocedureshouldbefollowedonpoultryfarmsand atslaughterhousestoidentifyhazardousmicroorganismsin asymptomaticpoultry,andthereforeeliminatecarcassesthat arenotproperforhumanconsumption.Thedatareinforcethe generalconcernaboutthespreadofmultidrug-resistanceand virulencegenesbetweenavianandhumanE.colithroughthe food-chain.

Conflict

of

interest

Theauthorshavenoconflictofinteresttodeclare.

Acknowledgements

TheauthorsthanktheBrazilianCouncilofResearch(CNPq) for research funding. Prof. Lima-Filho was supported by a ScholarshipfundingfromProgramadeEducac¸ãoTutorial (PET-MEC/SESu).TheauthorsalsothankMariaHelena(LIKA/UFPE) forprovidingthelaboratoryanimalsusedinthisstudy.

r

e

f

e

r

e

n

c

e

s

1.JonesFT,RickeSC.Observationsonthehistoryofthe developmentofantimicrobialsandtheiruseinpoultryfeeds. PoultSci.2003;82:613–7.

2.JohnsonJR,McCabeJS,WhiteDG,etal.Molecularanalysisof

Escherichiacolifromretailmeats(2002–2004)fromtheUnited

Statesnationalantimicrobialresistancemonitoringsystem. ClinInfectDis.2009;49:195–201.

3.CollignonP.ResistantEscherichiacoli—wearewhatweeat. ClinInfectDis.2009;49:202–4.

4.CollignonP,HenrikCW,PeterB,etal.Theroutineuseof antibioticstopromoteanimalgrowthdoeslittletobenefit proteinundernutritioninthedevelopingworld.ClinInfect Dis.2005;41:1007–13.

5.KwonSG,ChaSY,ChoiEJ,etal.Epidemiologicalprevalenceof avianpathogenicEscherichiacolidifferentiatedbymultiplex PCRfromcommercialchickensandhatcheryinKorea.J BacteriolVirol.2008;38:179–88.

6.Rodriguez-SiekKE,GiddingsCW,DoetkottC,etal.

CharacterizingtheAPECpathotype.VetRes.2005;36:241–56. 7.JohnsonTJ,WannemuehlerY,JohnsonSJ,etal.Comparison

ofextraintestinalpathogenicEscherichiacolistrainsfrom humanandaviansourcesrevealsamixedsubset representingpotentialzoonoticpathogens.ApplEnviron Microbiol.2008;74:7043–50.

8.TivendaleKA,AllenJL,GinnsCA,etal.Associationofissand

iucA,butnottsh,withplasmid-mediatedvirulenceofavian

9. JohnsonTJ,JohnsonSJ,NolanLK.CompleteDNAsequenceof aColBMplasmidfromavianpathogenicEscherichiacoli

suggeststhatitevolvedfromcloselyrelatedColVvirulence plasmids.JBacteriol.2006;188:5975–83.

10.JohnsonTJ,GiddingsCW,HorneSM,etal.Locationof increasedserumsurvivalgeneandselectedvirulencetraits onaconjugativeRplasmidinanavianEscherichiacoliisolate. AvianDis.2002;46:342–52.

11.JohnsonTJ,DiannaJ,SubhashinieK,etal.Sequenceanalysis andcharacterizationofatransferablehybridplasmid encodingmultidrugresistanceandenablingzoonotic potentialforextraintestinalEscherichiacoli.InfectImmun. 2010;78:1931–42.

12.BrazilianChickenProducersandExportersAssociation;2012. http://www.abef.com.br/English/default.php[accessed 28.06.12].

13.Brasil.Portaria2010;1998.http://extranet.agricultura.gov.br/ sislegis-consulta/servlet/VisualizarAnexo?id=3162[accessed 28.06.12].

14.SilvaIMM,Evêncio-NetoJ,SilvaRM,etal.Caracterizac¸ão genotípicadosisoladosdeEscherichiacoliprovenientesde frangosdecorte.ArqBrasMedVetZootec.2011;63:333–9. 15.NationalCommitteeforClinicalLaboratoryStandards.

Padronizac¸ãodosTestesdeSensibilidadeaAntimicrobianos porDiscodifusão:NormaAprovadaM2-A8,Oitava

Edic¸ão-Traduc¸ãoANVISA.Villanova,PA,USA:Clinicaland LaboratoryStandardsInstitute,NCCLS;2002.

16.KonemanEW,AllenSD,JandaWM,etal.Diagnóstico Microbiológico–TextoeAtlasColorido.2002(5).SãoPaulo, Brasil:MEDSI;2001.

17.SamuelsenØ,HauklandHH,UlvatneH,etal.

Anti-complementeffectsoflactoferrin-derivedpeptides. FEMSImmunolMedMicrobiol.2004;41:141–8.

18.Dho-MoulinM,FairbrotherJM.Avianpathogenic

Escherichiacoli(APEC).VetRes.1999;30:299–316.

19.EwersCT,Janssen,WielerLH.AvianpathogenicEscherichiacoli

(APEC).BerlMunchTierarztlWochenschr.2003;116:381–95. 20.KabirSML.AvianColibacillosisandSalmonellosis:acloser lookatepidemiology,pathogenesis,diagnosis,controland publichealthconcerns.IntJEnvironResPublicHealth. 2010;7:89–114.

21.BronzwaerSL,CarsO,BuchholzU,etal.AEuropeanstudyon therelationshipbetweenantimicrobialuseandantimicrobial resistance.EmergInfectDis.2002;8:278–82.

22.YangH,ShengC,DavidG,etal.Characterizationof multiple-antimicrobial-resistantEscherichiacoliisolatesfrom diseasedchickensandswineinChina.JClinMicrobiol. 2004;42:3483–9.

23.SalehiTZ,BonabSF.Antibioticssusceptibilitypatternof

Escherichiacolistrainsisolatedfromchickenswith

colisepticemiainTabrizProvince,Iran.IntJPoultrySci. 2006;5:677–84.

24.HaradaK,TetsuoA.Roleofantimicrobialselectivepressure andsecondaryfactorsonantimicrobialresistanceprevalence

inEscherichiacolifromfood-producinganimalsinJapan.J

BiomedBiotechnol.2012,

http://dx.doi.org/10.1155/2010/180682.

25. ZanattaGF,KanashiroAMI,CastroAGM,etal.Suscetibilidade deamostrasdeEscherichiacolideorigemaviáriaa

antimicrobianos.ArqInstBiol.2004;71:283–6.

26. HasanB,FaruqueR,DrobniM,etal.Highprevalenceof antibioticresistanceinpathogenicEscherichiacolifrom large-andsmall-scalepoultryfarmsinBangladesh.AvianDis. 2011;55:689–92.

27.OteoJ,CamposJ,LázaroE,etal.Increased

amoxicillin–clavulanicacidresistanceinEscherichiacoliblood isolates.SpainEmergInfectDis.2008;14:1259–62.

28.CoxJrLA,PopkenDA.Quantifyingpotentialhumanhealth impactsofanimalantibioticuse:enrofloxacinandmacrolides inchickens.RiskAnal.2006;26:135–46.

29.VellingaA,TanseyS,HanahoeB,etal.Trimethoprimand ciprofloxacinresistanceandprescribinginurinarytract infectionassociatedwithEscherichiacoli:amultilevelmodel.J AntimicrobChemother.2012;67(10):2523–30.

30. Morfin-OteroR,Tinoco-FavilaJ,SaderH,etal.Resistance trendsingram-negativebacteria:surveillanceresultsfrom twoMexicanhospitals,2005–2010.BMCResNotes.2012;5:277 [PMID:22676813].

31.FoodandDrugAdministration;2005.http://www.fda.gov/ AnimalVeterinary/SafetyHealth/RecallsWithdrawals/ ucm042004.htm[accessed28.06.12].

32.TheEuropeanAgencyfortheEvaluationofMedicinal Products,1998.http://www.emea.europa.eu/docs/enGB/ documentlibrary/MaximumResidueLimits-Report/2009/11/ WC500014142.pdf[accessed28.06.12].

33.EwersC,Esther-MariaA,DiehlI,etal.Intestineand environmentofthechickenasreservoirsforextraintestinal pathogenicEscherichia.ApplEnvironMicrobiol.

2009;75:184–92.

34.WarrenRE,EnsorVM,O’NeillP,etal.Importedchickenmeat asapotentialsourceofquinolone-resistantEscherichiacoli

producingextended.JAntimicrobChemother.2008;61: 504–8.

35.AllanBJ,JvandenHurkJV,PotterAA.Characterizationof

Escherichiacoliisolatedfromcasesofaviancolibacillosis.CanJ