The Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS) identification versus biochemical tests: a study with enterobacteria from a dairy cattle environment

h ttp : / / w w w . b j m i c r o b i o l . c o m . b r /

Veterinary

Microbiology

The

Matrix-Assisted

Laser

Desorption

Ionization–Time

of

Flight

Mass

Spectrometry

(MALDI-TOF

MS)

identification

versus

biochemical

tests:

a

study

with

enterobacteria

from

a

dairy

cattle

environment

Naiara

Miranda

Bento

Rodrigues

_,

_Greiciane

_Franc¸a

_Bronzato

_,

Gabrielli

Stefaninni

Santiago

_,

_Larissa

_Alvarenga

_Batista

_Botelho

_,

Beatriz

Meurer

Moreira

_,

_Irene

_da

_Silva

_Coelho

_,

_Miliane

_Moreira

_Soares

_de

_Souza

_,

Shana

de

Mattos

de

Oliveira

Coelho

a_{UniversidadeFederalRuraldoRiodeJaneiro,DepartamentodeMicrobiologiaeImunologiaVeterinária,Seropédica,RJ,Brazil} b_{LaboratóriodeInvestigac¸ãoemMicrobiologiaMédica–LIMM,InstitutodeMicrobiologiaPaulodeGoés,RiodeJaneiro,RJ,Brazil}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received29December2015 Accepted18July2016

Availableonline21September2016 AssociateEditor:JorgeLuizMello Sampaio

Keywords:

Bovinemastitis

Enterobacteriaidentification Matrix-AssistedLaserDesorption Ionization–TimeofFlightMass Spectrometry

a

b

s

t

r

a

c

t

Mastitisadverselyaffectsmilkproductionandingeneralcowsdonotregain their full productionlevelspostrecovery,leadingtoconsiderableeconomiclosses.Moreoverthe per-centagedecreaseinmilkproductiondependsonthespecificpathogenthatcausedthe infectionandenterobacteriaareresponsibleforthisgreaterreduction.Phenotypictestsare amongthecurrentlyavailablemethodsusedworldwidetoidentifyenterobacteria;however theytendtomisdiagnosethespeciesdespitethemultipletestscarriedout.Ontheother handTheMatrix-AssistedLaserDesorptionIonization–TimeofFlightMassSpectrometry (MALDI-TOFMS)techniquehasbeenattractingattentionforitspreciseidentificationof sev-eralmicroorganismsatspecieslevel.Inthecurrentstudy,183enterobacteriaweredetected inmilk(n=47)andfecalsamples(n=94)fromcows,andsamplesfromwater(n=23)and milklines(n=19).AllthesesampleswerecollectedfromafarminRiodeJaneirowiththe specificpurposeofpresentingtheMALDI–TOFMStechniqueasanefficientmethodology toidentifyEnterobacteriaceaefrombovineenvironments.TheMALDI–TOFMStechnique resultsmatchedthebiochemicaltestresultsin92.9%(170/183)oftheenterobacteriaspecies andthegyrBsequencingconfirmed100%oftheproteomictechniqueresults.Theamino

∗ _{Correspondingauthor.}

E-mail:[email protected](S.M.Coelho).

http://dx.doi.org/10.1016/j.bjm.2016.07.025

1517-8382/©2016SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

aciddecarboxylationtestmadethemostmisidentificationsandEnterobacterspp.wasthe mostmisidentifiedgenus(76.9%,10/13).Theseresultsaimtoclarifythecurrentbiochemical errorsinenterobacteriaidentification,consideringisolatesfromabovineenvironment,and showtheimportanceformorecarefulreadingsofphenotypictestswhichareoftenusedin veterinarymicrobiologylaboratories.

©2016SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.Thisis anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/

licenses/by-nc-nd/4.0/).

Introduction

Brazil is the fifth largest milk producer in the world and hadaproductionof6128billionlitersinthefirsthalf2015.1 DairylivestockarepresentthroughoutBrazilcontributingto thegenerationofemploymentandregionaldevelopmentand consequently dairy farming is important for the country’s economy.2_{However,environmentalbovinemastitisisamajor} challengeintheprimarysectorasitcausesadropinmilk pro-duction,prematurediscardofanimals,treatmentcostsand hasanegativeeffecton thequality ofmilk, whichinturn interferesintheindustrialprocessingofdairyproducts.3

EnvironmentalmastitisisassociatedmainlytoGram neg-ative bacteria,especially Enterobacteriaceae.Thisfamily of bacteriacolonizestheintestineofmammalsand birdsand alsoinfectsthe mammary glandsafter milkingdue tothe contactoftheudderwith infectedwater, bedding,milking environmentorcattleshed.4_{Thefamily Enterobacteriaceae} hasmorethan 50 genera and over200 species. Escherichia, Klebsiella,Enterobacter,SerratiaandProteusaregenusfrequently isolatedfromdairyenvironments.Escherichiacoliarenaturally presentinfecesfromwarm-bloodedanimalspecies,Klebsiella, EnterobacterandSerratiamarcescensinhabitsoils,grains,and water.Proteusspp.usuallycontaminantshosewaterusedto washuddersbeforemilking.5,6_{Thus,characterizationof} circu-latingEnterobacteriaceaeinadairyproductionenvironment isessentialtounderstandtheimportanceofthesebacteriaas agentsofbovinemastitis,aswellastheirroutesof contami-nation.

Thequalityofmicrobialidentificationcanimpactthe vet-erinaryclinicalmanagementbecoming anessentialtaskto controlmastitis.Theidentificationofthebacteriainroutine clinical microbiology laboratories is still based on pheno-typictests,however,somestrainswithinaspeciesmayhave smallbiochemicaldifferencesthatcanresultinfalseresults

in vitro.7,8 _{Phenotypicproperties are unstable attimes and} theirexpressionisdependentuponchangesinthe environ-mentalconditions, e.g., growthsubstrate,temperature, and pHlevels.9_{Inaddition,commerciallyavailableidentification} systemssometimesareunabletoidentifysomeorganisms.10 TheMatrix-AssistedLaserDesorptionIonization–Timeof FlightMassSpectrometryhasbeen usedforbacteria detec-tion and identification using an easy to follow protocol.11 The method is based on the accurate determination of their protein mass that is compared to available profiles storedinasoftwaredatabaseidentifyingthespecieinafew minutes.8_{MALDI–TOFMSprovidesfastandaccurateresults} onspeciesidentificationthusimprovingtheoverall microbio-logicaldiagnosis.12

Moleculartechniques,suchas16SrRNAsequencing,are thegoldstandardforbacterialidentificationbutthecurrent bacteriacharacterizationbythesetechniquesisunusualand tooexpensiveforveterinarylaboratories.However,theycanbe usedasatooltohelpunderstandthediscrepanciesbetween thecommonlyusedprotocols.Forexample,thegyrBgene– encodinggyrasesubunitBortopoisomeraseIIhasbeen con-sideredmorediscriminatingthan 16SrRNA,sinceitallows thedifferentiationandidentificationofcloselyrelatedspecies withintheEnterobacteriaceaefamily.

This study compared the MALDI–TOF MS technique to traditionalbiochemical-basemethodstoidentify Enterobacte-riaceaefrombovinemilkandfeces,aswellasinsamplesfrom waterandmilklinesofadairyproductionsystemtodetermine the possibleerrorsinbiochemicaltests. Also,theincorrect identificationswereconfirmedthroughgyrBgenesequencing.

Materials

and

methods

Ethicsstatement

Thisstudywascarriedoutincooperationwithadairyfarmer inthemunicipalityofBarradoPiraí,RiodeJaneiro,Brazil.This reportisnotintendedtobeafieldstudy;insteaditdescribes the applicationand accuracyof MALDI–TOF MS technique and conventionalbiochemicaltestsusedinlaboratoriesfor the identification of coliform bacteria associatedto bovine mastitis.Therefore,detailsand influenceoftheseasonson samplesarenotincluded.Thisstudywasconductedaccording to ethical standards and approved by the Ethics Commit-teeandBiosafetyoftheinstitutionunderprotocolnumber: CEUA-3664040915.Thesamplesusedinthecurrentstudywere obtainedfromsamplessubmittedforroutineveterinary diag-nosis, which was unrelated to this research. Thereport is focusedonthemicrobiologicalanalysisfollowingthe isola-tionofbacteriafromthemilkandfecalsamples,anddidnot directlyinvolveanyoftheanimals.

Materialcollection

ThepresentstudywasperformedinthetownofBarradoPiraí, RiodeJaneiro,Brazil(Latitude:22◦ 2814 South,Longitude: 43◦4936West)in2014and2015.Apoolofmilksamplesfrom 94cowstestedpositivebytheCaliforniaMastitisTest(CMT) werecollected,overthreeconsecutiveweeks.Atotalof94 rec-talfecessamplesfromthesesamelactatingcowswerealso collected.Representingmilklinesamples(n=48):10samples fromworkers’hands,10nasalsamplesfromworkers,20 milk-ingmachinesamplesand8nasalsamplesofpets(dogsand

catspresentinthemilkingparlor).Finally,onesamplefrom eachfarmwatersupplywascollected–waterfromwell,weir, faucet,drinkingfountainandbrook,makingatotalof19water samples.

Phenotypicidentification

Thebovinemilkandmilklinesampleswerefirstinoculated onbloodagar(bloodagarbaseenrichedwith5%sheepblood), whilethefecalandwatersampleswereinoculatedonEMB (EosinMethylenBlue)agarandincubatedat35◦C(±2◦C)for 24h.Then,theisolatesweresubmittedtoroutine microbio-logicaldiagnostics,includinginoculationinselectivemedium foranalysisofculturalproperties.TheGramnegativebacteria identificationwasfollowedbyaprotocolcomprising:glucose andlactosefermentationwithgasproduction,H2S(hydrogen

sulfide),indole,motility,acetoinandmixedacidproduction, lysine and ornithine decarboxylation, arginine dehydroge-nase,ureaseproduction,citrateandmalonate.13_Alltestswere carriedoutintriplicate.ThestrainsEscherichiacoliATCC25922 and KlebsiellapneumoniaeATCC700603 were usedasquality control.

Proteomicidentification(MALDI–TOFMS)

Furthermore,allisolateswere evaluatedbyMALDI–TOFMS. Initially,thesampleswereinoculatedinBHIagarat37◦Cfor 24h. Eachculturewastransferredtoa microplate(96 MSP, Bruker® _{–Billerica, USA). Thebacterial sediment was}

cov-eredwithalysissolution(70%formicacid;Sigma–Aldrich®_).

Then a 1-␮L aliquot of matrix solution (alpha-ciano-4-hidroxi-cinamic acid diluted in 50% acetonitrile and 2.5% trifluoroacetic acid, Sigma–Aldrich®_{)was added. The}

spec-traofeach samplewere generatedinamassspectrometer (MALDI–TOFLTMicroflex,Bruker®_{)equippedwitha337nm}

nitrogenlaser ina linear path, controlled bythe FlexCon-trol 3.3(Bruker®_{)program.Thespectra were collectedin a}

massrangebetween2000and20,000m/s,andthenwere ana-lyzedbytheMALDIBiotyper2.0(Bruker®_{)program,usingthe}

standardconfigurationforbacteriaidentification,which com-paredthespectrumofthesampleswiththereferencesinthe database.Theresultsvaryona0–3scale,wherethehighest valuemeansamoreprecisematchandreliableidentification. Inthisstudy,weacceptedvaluesformatchinggreaterthanor equalto2.

gyrBsequencing

The bacterial DNA was extracted after thermal lysis and the700bpgyrBfragmentamplificationwasobtainedby Poly-meraseChainReactiontechnique(PCR).ThePCRwasperformed usingtheprimersUP1and181r.14,15_{Theconcentrationsused} forthereactions were:buffer10×(10mMTris–HCl,pH 9.0; 50mMKCl and0.1%TritonX-100), 2.0mMofMgCl2, 0.5␮M

ofeachprimer,0.2mMofdNTP(Fermentas)and0.5UofTaq polymerase(Fermentas)inatotalreaction volumeof20␮L with20ngofthe extractedDNA.Thefragmentswere eval-uated by electrophoresis in agarose gel (1.5%) and reveled withthe SYBRgreendye(Invitrogen),making visualization inultraviolet light possibleas well asbeing able torecord

theampliconswiththeimagecapturesystemL-PIXEX (Loc-cus Biotecnologia). The PCR products were purified using an Exo-Sap (USB Corporation, Cleveland, Ohio), as recom-mendedbythemanufacturerandsubmittedtothesequencer ABI3130xlofAppliedBiosystems(BiotechnologyLaboratory of the Genomic Sciences Post-Graduation Program of the Catholic University ofBrasilia). Thesequences were edited usingtheBioeditprogram16_{andMegaversion4.0}17_and com-paredwithothersequencesintheNCBUdatabase(GenBank:

http://www.ncbi.nlm.nih.gov/)usingtheBLASTalgorithm.18

Statistics

ThebiochemicaltestsandMALDI–TOFMSassociationwere evaluatedbyChi-Squaretest(2_{)andFisher’stestwith95%}

confidenceinterval,consideringwrongidentificationin gen-eraorataspecielevel.

Results

and

discussion

Atotalof183enterobacteriaweredetectedfrom443samples: 51.3% (94/183) were from bovinefeces, 25.6% (47/183) from bovinemilk,12.6%(23/183)fromwaterand10.3%(19/183)from themilkline.Allisolatesweresubmittedtobiochemicaltests andproteomicsandbothproceduresidentifiedisolatesata specielevel.

TheMALDI–TOFMSidentifiedE.coliastheprevalentspecie in all samplesevaluated (83%, 152/183) (Table 1).Although

E.coliwasisolatedmainlyfromthefecessamplesitwasalso detectedinothersamples.Thisspecieisoneofthemost thor-oughlystudiedfree-livingorganisms.Itisalsoaremarkably diversespeciesbecausesomeE.colistrainsliveasharmless commensals in animal intestines.19 _{Its presence isa} defi-niteindicationoffecalcontaminationandsomestrainscan cause awide variety ofintestinal and extra-intestinal dis-eases, suchasdiarrhea,urinarytractinfections,septicemia andmastitis.20

The milksample showed the highest variablyof enter-obacteriawithsevenoftheelevenbacteriaisolated.Thisis probablyduetothemethodologyofthemilkcollectionused inthisstudy–forthreeconsecutiveweeks,pointingto possi-blechangesinmilkmicrobiome.However,becauseofthelow numberofsamplesmoreinvestigationsofeachcowevaluated wouldbenecessarytoconfirmthishypothesis.Inspiteofthat, thepresenceofalargevariabilityofenterobacteriais worry-ingsincethesebacteriafrequentlyareassociatedtoclinical bovinemastitis.21

TheMALDI–TOFMStechniquematchedwith92.9%(n=170) of the biochemical identifications results (Table 2). Other authorshavealsofoundahighcorrelationbetweenthesetwo techniques. Eigneretal.(2009)22 _{described95.2% of} consis-tencyin1116bacterialstrainspreviouslyidentifiedinclinical routineandBizzinietal.(2010)11_{obtainedcorrect} identifica-tionof95.1%of1278strainsatspeciesleveland3%atgenus level.Oneofthemajoradvantagesofusingthistechnologyfor bacterialidentificationisthetime-to-result,whichisreduced from1to6daystolessthananhour.Inaddition,MALDI–TOF MS allowedaccurate bacterialidentificationofalarge vari-ety of bacteria. Genotypic identification using coa, nuc and

Table1–EnterobacteriaidentifiedbyMALDI–TOFMSisolatedfromdifferentsamplesinabovineenvironment.

EnterobacteriaidentifiedbyMALDI–TOFMS Samples Total

Milk Milkline Water Feces

Escherichiacoli 35 8 15 94 152 Enterobactercloacae 2 5 – – 7 Klebsiellapneumoniae 3 2 – – 5 Klebsiellaoxytoca 3 – 2 – 5 Enterobacterasburiae 1 3 1 – 5 Raoultellaornithinolytica – – 2 – 2 Serratiamarcescens 2 – – – 2 Enterobacteraerogenes – – 2 – 2 Citrobacterfreundii 1 – – – 1 Morganellamorganii – 1 – – 1 Providenciastuartii – – 1 – 1 Total 47 19 23 94 183

Table2–ThephenotypicmethodsincomparisontoMALDI–TOFMSinidentificationofenterobacteriaisolatedfroma bovineenvironment.

SpeciesidentifiedbyMALDI–TOF(averagescore) Isolates(%)identifiedbyconventionalphenotypicmethod

N◦ofisolates Correctidentification Misidentification

Specielevel Specielevel Genuslevel

Escherichiacoli(2.391) 152 152(100) – – Enterobactercloacae(2.161) 7 4(57.1) 2(28.5)a _1(14.2)b Enterobacterasburiae(2.243) 5 – 5(100)a,c _– Enterobacteraerogenes(2.450) 2 – 2(100)c _– Citrobacterfreundii(2.273) 1 1(100) – – Klebsiellaoxytoca(2.314) 5 5(100) – – Klebsiellapneumoniae(2.276) 5 4(80) 1(20)d _– Morganellamorganii(2.422) 1 1(100) – – Providenciastuartii(2.228) 1 1(100) – – Raoultellaornithinolytica(2.598) 2 – – 2(100)b Serratiamarcescens(2.311) 2 2(100) – – Total 183 170(92.9) 10(5.5) 3(1.6) a _{Pantoeaagglomerans.} b _{Klebsiellapneumoniae.} c _{Enterobactercloacae.} d _{Klebsiellaoxytoca.}

ADNr23S genescompared toMALDI–TOF had 100% sensi-tivityand specificity, accordingtoMotta etal.(2014)23 _ina

Staphylococcus spp.study. Duboiset al.(2010)24 _{and Carpaij} etal.(2011)25_{demonstratedthatproteomictechniquesarean} accuratemethodfordeterminingcoagulase-negative Staphy-lococcus(CNS) speciescomparedtomolecular identification ofclinicalisolates.Afterproteomicidentification,13(7.10%, 13/183)isolatesmisidentifiedweresubmittedtogyrB sequenc-ing.Therewerenoamplificationresultsforfiveisolatesasit wasnotpossibletoextracttheirDNA(1 K.pneumoniae,1E. cloacae,1E.asburieand2Raoultellaornithinolytica).Therefore, onlyeightenterobacteriawithnon-concordantidentification wereanalyzedwiththistechnique.Additionally, 15isolates withconcordantidentificationweresequencedtobethe

con-trol(Table3).ThegyrBsequencingresultsdemonstratedsix

specieshadbeenidentifiedcorrectlybyMALDI–TOFMS:two

E.aerogenes(A12andA17),twoE.asburiae(L62andL25)and twoE.cloacae(L62andL4).Oneisolate(L52)identifiedasP. agglomerans by biochemical tests and as E. asburieby pro-teomic testwas characterized asgenus Enterobacter. These

resultsshow thatgyrB sequencingand theMALDI–TOF MS techniqueproducesimilarresults.Anotherisolateidentified asE. clocaeandasE. asburiebybiochemicaland proteomic techniques,respectively,wasalsocharacterizedonlyatgenus levelbysequencing.Thus,inthepresentstudytheproteomic techniquewasusedasthe“goldstandard”forevaluatingthe sensitivityandspecificityofphenotypictestsusedin enter-obacteriaidentification.

Thespeciescorrectlyidentifiedbybiochemicaltestswere

E. coli(n=152), Klebsiellaoxytoca (n=5), Klebsiellapneumoniae

(n=4),Enterobactercloacae(n=4),Serratiamarcescens(n=2), Cit-robacter freundii (n=1), Morganellamorganii (n=1), Providencia stuartii(n=1).Thebiochemicalteststhatmisreadspecies iden-tification,inthiswork,arepresentedinTable4.

The phenotypic test parameters (sensitivity and speci-ficity)showedhighvaluesduetotheirhighlevelperformance confirmed by MALDI–TOF MS. After the proteomic iden-tification the thirteen misidentified isolates were used to investigatethemistakesoccurred inthe phenotypic identi-fication.Asdescribedabovefourteentestswereusedinthe

Table3–ThegyrBsequencingresultsinenterobacteriaidentifiedbybiochemicaltestsandMALDI–TOFMS.a

Strainsa _{Biochemicalidentification MALDI–TOFMS BLASTn NCBI Maximumidentity}

L65 E.coli E.coli E.coli CP013253.1 99%

L22 E.coli E.coli E.coli KF914219.1 98%

L57 E.coli E.coli E.coli CP013253.1 92%

L64 E.coli E.coli E.coli CP013253.1 97%

F56 E.coli E.coli E.coli CP013253.1 98%

A21 E.coli E.coli E.coli DQ386875.1 96%

A24 E.coli E.coli E.coli CP011018.1 96%

A16 E.coli E.coli E.coli CP011331.1 99%

A16 E.coli E.coli E.coli CP011331.1 99%

L10 E.cloacae E.cloacae E.cloacae EF064834.1 99%

L19 E.cloacae E.cloacae E.cloacae CP012167.1 88%

L58 E.cloacae E.cloacae E.cloacae AB084013.1 99%

L34 S.marcescens S.marcescens S.marcescens JX103478.1 99%

L60 K.pneumoniae K.pneumoniae K.pneumoniae CP008700.1 98%

A4 K.oxytoca k.oxytoca K.oxytoca FJ617358.1 84%

A5 E.cloacae E.asburiae Enterobactersp. KF871115.1 99% A12 E.cloacae E.aerogenes E.aerogenes CP011574.1 98% A17 E.cloacae E.aerogenes E.aerogenes F0203355.1 95% L52 P.agglomerans E.asburiae Enterobactersp. FJ617364.1 90% L62 P.agglomerans E.asburiae E.asburiae CP011591.1 95% L68 P.agglomerans E.cloacae E.cloacae EF064834.1 99% L4 P.agglomerans E.cloacae E.cloacae AB084017.1 92% L25 P.agglomerans E.asburiae E.asburiae CP012162.1 82%

a _{Theboldvaluesshowthestrainsusedasacontrol.}

Table4–Biochemical-basedteststhatdifferfromtheidentificationbyMALDI–TOFMS.

Phenotypicidentification (no.ofisolates)

MALDI–TOFMSidentification(no.ofisolates)

Species Genus Biochemicaltestsresponsiblefor

misidentificationi

Enterobactercloacae(4) Enterobacterasburiae(2) – Acetoina_,mixedacidb_{productionand}

motilityc

Enterobacteraerogenes(2) – Ureaseproductiond_,lysine

decarboxylatione_{andargininehydrolysis}f

Klebsiellaoxytoca(1) Klebsiellapneumoniae(1) – Indoleproductiong

Klebsiellapneumoniae(3) – Enterobactercloacae(1) Motilityc_{,Argininehydrolysis}f_,Lisinee

andOrnithinedecarboxylationh

– Raoultellaornithinolytica(2) Ornithinedecarboxylationh_,indoleg_and

mixedacidproductionb

Pantoeaagglomerans(5) – Enterobactercloacae(2) Lisinee_{andornithinedecarboxylation}h

– Enterobacterasburiae(3) Ornithinedecarboxylationh

a _{SE:100%andSP:99%.} b _{SE:98%andSP:100%.} c _{SE:99%andSP:89%.} d_{SE:100%andSP:99%.} e _{SE:81%andSP:99%.} f _{SE:90%andSP:99%.} g _{SE:99%andSP:95%.} h _{SE:93%andSP:100%.}

i _{Sensitivity(SE)andspecificity(SP)ofthebiochemicaltests.}

biochemical-based identification and eight of them were detectedasbeingresponsibleforwrongresults;thesetests were:bacterialmotilityinsemi-solidmedium,acetoin produc-tion(VogesProskauertest),mixedacidsproduction(Methyl RedTest),ureaseandindoleproduction,argininehydrolysis, lysineandornithinedecarboxylation.

Amongthe discordant results, Enterobacter spp.was the mostcommongenusmisidentified(77%, 10/13).Enterobacter

asburiae(n=5)weremisidentifiedasPantoeaagglomerans(n=3) and Enterobactercloacae(n=2).SomeE.asburiewere consid-eredfalsepositivebytheVPtestbutacetoinproductionbythis speciecanbeexpressedinonly79%ofisolateswhichprobably leadstomisidentification.Pantoeaagglomeranswasbelonging toEnterobactergenus,classifiedasEnterobacteragglomerans.26_It isdifficulttodifferentiatePantoeaspp.fromtheothermembers ofthisfamily,suchastheEnterobacter,KlebsiellaandSerratia

species.Recently,sixspecieswereassignedtothe Enterobac-tercloacaecomplex,includingEnterobacterasburiae,Enterobacter cloacae,Enterobacterhormaechei,Enterobacterkobei,Enterobacter ludwigii,andEnterobacternimipressuralis.Insurveillance stud-ies,Enterobacter speciesare notoftenclassified beyondthe genuslevel,probablybecauseidentificationisdifficult.Only the Enterobacter spp. isolates that belong to the E. cloacae

complexareofclinicalsignificanceandareincreasingly iso-latedashumanpathogens.Thus,morepreciseidentification oftheE.cloacaecomplexisolatesmaypermitdifferentiation betweenpathogens,commensalandtransitionalspecies.Up tonowphenotypic identificationofspeciesand subspecies withintheE.cloacaecomplexhavebeenlargelyunreliableand irreproducible.27

Two isolates of Raoultella ornithinolytica were misidenti-fiedas Klebsiellapneumoniae. Theindole, VMand ornithine decarboxylationtestsareresponsibletodifferentiateR. ornithi-nolytica and K. pneumoniae and probably were mistakenly interpreted. Historically, these species are closely related. Drancourtetal.(2001)28_{usingphylogeneticanalysisofrpoB} genusconfirmedtheheterogeneityofKlebsiellaspp.and pro-posedtheformationofthreegroups:GroupIcomprisethree subspecies,K. pneumoniae subsp. pneumoniae, K.pneumoniae

subsp.ozaenaeandK.pneumoniaesubsp.rhinoscleromatis;Group IIcompriseK.terrigena,K.ornithinolytica,K.planticolaandK. trevisanii;andGroupIIIisrepresentedbyK.oxytoca.Basedon thisevidence,theseauthorsproposedadivisionofKlebsiella

genus:Klebsiellaspp.andRaoultellaspp.AllGroupIIspecies weretransferredtothenewgenusRaoultellaspp.Inthepresent studyRaoultellaornithinolytica wasisolatedfromwater sam-ples.Thisspeciesisoftenrecoveredfromwater,soil,plants andoccasionallymammalmucosa,andsomaybeassociated tomastitis.23

OneisolateofKlebsiella pneumoniaewas misidentifiedat the genus-level. By proteomic technique it was identified as Enterobacter cloacae. Although belonging to a different genus it is included in the same tribe – Klebsielleae, along withHafnia,Serratia andPantoeagenera.In thisstudy, Kleb-siellapneumoniaepresentedfalsenegativeresultsforarginine, ornithineandmotilitytestsaswellasfalsepositivetolysine. Theaminoacidsdecarboxylationtestsareextremely impor-tant,especiallyfortheseparationofcertainmembersofthe

Klebsiella-Enterobacter-Serratia-Hafniagroup.13

AnotherK.pneumoniaewasmisidentifiedasK.oxytoca.The indoleproductioncoulddifferentiatethembutsomestrains donotproduceclassicreactionsleadingtodesignationof sev-eraladditionalspeciesandconsequentlyinducingtheerrorof identification.29_{AndthemisidentificationofE.aerogenesasE.}

cloacaewasprobablyduetotheureaseinterpretationreading forE.cloacae,whichisvariableinmostcases.13

Consideringall the tests usedin the present study,the lysinedecarboxylationpresentedthelowestsensitivity per-centage(81%).Interpretationofthetestsfordecarboxylation ofamino acids must beaccurate especiallybecause some isolatespresent aslowreaction resulting inawrong iden-tification.Additionally,thebacterialdensity,theincubation timeandthepHproperlyadjustedarerequiredstandardsto minimizetheriskoffalseresults.25_Ingeneral, microorgan-ismsfirstfermentglucosetolowerthepHsothattheoptimal hydrogen ion concentration for decarboxylase activity is

reached.Decarboxylationresultsintheformationofamines and consequently increases the pH. Positive results are usuallyobtainedafter18–24h ofincubation butwithsome strainsthetestsmustbeheldforaslongas4days.Oxygen isgenerally excludedbyoverlaying thebroth withparaffin ormineraloilorbytheadditionof0.3%agar.30,31_Amedium withagarcannotbeheldlongerthan24handthusafewfalse negativeresultsshouldbeexpected.Alargenumberof micro-biologists haveadoptedthe use ofthelysine Falkowbroth insteadoftheMölleragar,becausetheFalkowtesteliminates theanaerobicoracidenvironment.However,thistestcannot beusedtodetectthelysinedecarboxylaseactivityofcertain members of the Klebsiella-Enterobacter-Serratia-Hafnia group considering the acetyl methyl carbinol production which interferesattheendofalkalinepHchangeleadingtoafalse negativeinterpretation.13

Thebacterialmotilitytestinducedmisidentificationoftwo isolates(E.cloacaeandE.asburiae).Someauthorssuggestthe useofsemi-solidmediumwithtetrazoliumsaltstoevaluate themotility.Thissaltwasconvertedintoredandinsoluble complexesbythereducing propertiesofmost enterobacte-riathusimprovingthevisualization.32_{Inthesameway,the} indoleandmixedacidsproductiontestsalsodependon indi-vidualviewingmakingthistestsubjectiveinmostcases.The subjectivityofphenotypictestshasincreasedthesearchfor moredefinitivetechniquesinbacterialidentificationsuchas moleculartools.9_{But,equallyimportantwithregardtothese} molecularmethodsisthescientificaccuracyofthe biochemi-calidentificationofeachspeciessinceclinicalandveterinary laboratoriescontinuetorelyuponthebiochemicalproperties andnotupon16SrDNAgenesequencingtoidentifybacterial strains.

Conclusion

In thepresent study,the MALDI–TOFMS techniqueresults matchedin92.9%withthe phenotypictestresultsand the

gyrB sequencing confirmed the proteomic results in 100%.

E.coliwasprevalentinallsamplescollectedfromthebovine environment and the milk bovine sample presented the highest microbiota variation. The E. coli specie was easily identifiedbyMALDI–TOFMSwhileEnterobacterspp.wasthe most misidentifiedgenus. Thephenotypic test parameters showed high values (sensitivity>81% and specificity>89%) due totheir high level performanceconfirmedby the pro-teomic technique; however eight of them were associated withmisidentificationatspeciesorgenuslevels.Theamino acid decarboxylationtest wasthe cause ofthe majorityof themisidentifications.Thisshowstheimportanceformore carefulreadingsofphenotypictestswhichareoftenusedin veterinaryclinicalmicrobiologylaboratories.

Conflicts

of

interest

Theauthorsdeclarenoconflictsofinterest.

Acknowledgements

ThisstudywassupportedbytheNationalCouncilfor Scien-tific and TechnologicalDevelopment (CNPq, Riode Janeiro,

Brazil–process308528/2011-5)andFoundationforResearch SupportintheStateofRiodeJaneiro(FAPERJ).

r

e

f

e

r

e

n

c

e

s

1. InstitutoBrasileirodeGeografiaeEstatística(IBGE)Brasil MinistériodoDesenvolvimento,IndústriaeComércio Exterior.2015.

2. WillersCD,FerrazSP,CarvalhoLS,RodriguesLB.

Determinationofindirectwaterconsumptionand

suggestionsforcleanerproductioninitiativesforthe

milk-producingsectorinaBrazilianmiddle-sizeddairy

farming.JCleanProd.2014;72:146–152.

3. JiangQ,YangY,XinK,etal.Microbialdiversityanalysisof

subclinicalmastitisindairycattleinNortheastChina.AfrJ

MicrobiolRes.2015:687–694.

4. PerezNetoF,ZappaV.Mastiteemvacasleiteira:revisãode

literatura.RevCientEletronMedVet.2011;16:1–28.

5. HoganJS,GonzalezRN,HarmonRJ,etal.LaboratoryHandbook

onBovineMastitis.Madison,WI,USA:NationalMastitis

Council,Inc.;1999.

6. HoganJS,WeissWP,SmithKL,TodhunterDA,Schoenberger

PS,SordilloLM.EffectsofanEscherichiacoliJ5vaccineon

mildclinicalcoliformmastitis.JDairySci.1995;78:285–290.

7. CarrollKC,WeinsteinMP.Manualandautomatedsystems

fordetectionandidentificationofmicroorganisms.In:

MurrayPR,BaronEJ,JorgensenJH,LandryML,PfallerMA,

eds.ManualofClinicalMicrobiology.9thed.Washington,DC:

AmericanSocietyforMicrobiol;2007:192–217.

8. SengP,RolainJM,FournierPE,LaScolaB,DrancourtM,

RaoultD.MALDI–TOFmassspectrometryapplicationsin

clinicalmicrobiology.FutureMicrobiol.2010;5:1733–1754.

9. Rosselló-MoraR,AmannR.Thespeciesconceptfor

prokaryotes.FEMSMicrobiolLett.2001;25:39–67.

10.JandaJM,AbbottSL.Bacterialidentificationforpublication:

whenisenoughenough?JClinMicrobiol.

2002;40(6):1887–1891.

11.BizziniA,GreubG.Matrix-assistedlaserdesorption

ionizationtime-of-flightmassspectrometry,arevolutionin

clinicalmicrobialidentification.ClinMicrobiolInfect.

2010;16:1614–1619.

12.MurrayPR.Matrix-assistedlaserdesorptionionization

time-of-flightmassspectrometry:usefulnessfortaxonomy

andepidemiology.ClinMicrobiolInfect.2010;16:1626–1630.

13.KonemanWE,AllenSD,JandaWM,SchreckenbergerPC,

WinnWCJr.Asenterobacteriaceae.In:KonemanEW,ed.

Diagnósticomicrobiológico–textoeatlascolorido.6.ed.Riode

Janeiro:MédicaeCientífica;2012:263–329.

14.YamamotoS,HarayamaS.PCRamplificationanddirect

sequencingofgyrBgeneswithuniversalprimersandtheir

applicationtothedetectionandtaxonomicanalysisof

Pseudomonasputidastrains.ApplEnvironMicrobiol.

1995;61:1104–1109.

15.KasaiH,WatanabeK,GasteigerE,etal.Constructionofthe

gyrBdatabasefortheidentificationandclassificationof

bacteria.GenomeInformSerWorkshopGenomeInform.

1998;9:13–21.

16.HallTA.BioEdit:auser-friendlybiologicalsequence

alignmenteditorandanalysisprogramforWindows

95/98/NT.NuclAcidsSympSer.1999;41:95–98.

17.KumarS,TamuraK,NeiM.MEGA3:integratedsoftwarefor

molecularevolutionarygeneticsanalysisandsequence

alignment.BriefBioinform.2004;5(2):150–163.

18.AltschulSF,MaddenTL,SchafferAA,etal.GappedBLAST

andPSI-BLAST:anewgenerationofproteindatabasesearch

programs.NucleicAcidsRes.1997;25:3389–3402.

19.FarasatT,BilalZ,YunusFUN.Isolationandbiochemical

identificationofEscherichiacolifromwastewatereffluentsof

foodandbeverageindustry.JCellMolBiol.2012;10:

13–18.

20.CiesielczukH[Doctoraldissertation]Extra-intestinal

pathogenicEscherichiacoliintheUK.Theimportancein bacteraemiaversusurinarytractinfection,colonizationof widespreadclonesandspecificvirulencefactors.QueenMary

UniversityofLondon;2015.

21.OikonomouG,BicalhoML,MeiraE,etal.Microbiotaofcow’s

milk;distinguishinghealthy,sub-clinicallyandclinically

diseasedquarters.PLOSONE.2014;9(1):85904.

22.EignerU,HolfelderM,OberdorferK,Betz-WildU,BertschD,

FahrA.Performanceofamatrix-assistedlaserdesorption

ionization-time-of-flightmassspectrometrysystemforthe

identificationofbacterialisolatesintheclinicalroutine

laboratory.ClinLab.2009;55:289.

23.MottaCC,RojasACCM,DubenczukFC,etal.Verificationof

molecularcharacterizationofcoagulasepositive

Staphylococcusfrombovinemastitiswithmatrix-assisted

laserdesorptionionization,time-of-flightmass

spectrometry(MALDI–TOFMS)massspectrometry.AfrJ

MicrobiolRes.2014;8:3861–3866.

24.DuboisD,LeysseneD,ChacornacJP,etal.Identificationofa

varietyofStaphylococcusspeciesbymatrix-assistedlaser

desorptionionization–timeofflightmassspectrometry.JClin

Microbiol.2010;48:941–945.

25.CarpaijN,WillemsRJ,BontenMJ,FluitAC.Comparisonof

theidentificationofcoagulase-negativestaphylococciby

matrix-assistedlaserdesorptionionizationtime-of-flight

massspectrometryandtufsequencing.EurJClinMicrobiol

InfectDis.2011;30:1169–1172.

26.BradyC,CleenwerckI,VenterS,CoutinhoT,DeVosP.

TaxonomicevaluationofthegenusEnterobacterbasedon

multilocussequenceanalysis(MLSA):proposaltoreclassify

E.nimipressuralisandE.amnigenusintoLelliottiagen.nov.as

Lelliottianimipressuraliscomb.nov.andLelliottiaamnigena

comb.nov.,respectively,E.gergoviaeandE.pyrinusinto

Pluralibactergen.nov.asPluralibactergergoviaecomb.nov.and

Pluralibacterpyrinuscomb.nov.,respectively,E.cowanii,E. radicincitans,E.oryzaeandE.arachidisintoKosakoniagen.nov.

SystApplMicrobiol.2013;36:309–319.

27.KämpferP,NienhüserA,PackroffG,WernickeF,MehlingA.

Molecularidentificationofcoliformbacteriaisolatedfrom

drinkingwaterreservoirswithtraditionalmethodsandthe

Colilert-18system.IntJHygEnvironHealth.2007;211:

374–384.

28.DrancourtM,BolletC,CartaA.Phylogeneticanalysesof

KlebsiellaspeciesdelineateKlebsiellaandRaoutellagen.nov.,

withdescriptionofRaoultellaornythinolyticacomb.Nov.,

Raoultellaterrigenacomb.nov.andRaoultellaplanticola.JSyst EvolMicrobiol.2001;51:925–932.

29.ChuW,ZereTR,WeberMM,etal.Indoleproduction

promotesEscherichiacolimixed-culturegrowthwith

Pseudomonasaeruginosabyinhibitingquorumsignaling.Appl EnvironMicrobiol.2012;78:411–419.

30.PedrazaJG,SanandresNP,VarelaZS,AguirreEH,CamachoJV.

AislamientomicrobiológicodeSalmonellaspp.y

herramientasmolecularesparasudetección.SaludUninorte.

2014;30:73–94.

31.GonzalezPJB,SotoVarelaZ,HernándezE,VillarealCJ.

AislamientodeSalmonellaspp.yherramientasmoleculares

parasudetección.RevistaCientíficaSaludUninorte.

2013;30(1):73–94.

32.MacFaddinJF.BiochemicalTestsforIdentificationofMedical

Bacteria.3rded.Philadelphia:LippincottWilliams&Wilkins; 2000.