Adhesion and invasion of Clostridium perfringens type A into epithelial cells

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

Veterinary

Microbiology

Adhesion

and

invasion

of

Clostridium

perfringens

type

A

into

epithelial

cells

Luis

A.

Llanco

_,

_Viviane

_Nakano

_,

_Claudia

_T.P.

_de

_Moraes

_,

_Roxane

_M.F.

_Piazza

_,

Mario

J.

Avila-Campos

a_{AnaerobeLaboratory,InstituteofBiomedicalScience,UniversityofSaoPaulo,SaoPaulo,SP,Brazil}

b_{BacteriologyLaboratory,ButantanInstitute,SaoPaulo,SP,Brazil}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received11July2016 Accepted8September2016 Availableonline7July2017 AssociateEditor:BeatrizErnestina CabilioGuth

Keywords:

Adhesion Invasion

ClostridiumperfringenstypeA

Necroticenteritis

tpeLgene

a

b

s

t

r

a

c

t

Clostridiumperfringensisthecausativeagentfornecroticenteritis.Itsecretesthemajor

vir-ulencefactors,and␣-andNetB-toxinsthatareresponsibleforintestinallesions.TheTpeL toxinaffectscellmorphologybyproducingmyonecrosis,butitsroleinthepathogenesisof necroticenteritisisunclear.Inthisstudy,thepresenceofnetBandtpeLgenesinC.perfringens

typeAstrainsisolatedfromchickenswithnecroticenteritis,theircytotoxiceffectsandrole inadhesionandinvasionofepithelialcellswereevaluated.Six(27.3%)ofthe22C. perfrin-genstypeAstrainswereharboringthetpeLgeneandproducedmorphologicalalterationsin Verocellsafter6hofincubation.StrainstpeL(−)inducedstrongcellroundingafter6hof incubationandproducedcellenlargement.Noneofthe22strainsharborednetBgene.All thesixtpeL(+)genestrainswereabletoadheretoHEp-2cells;however,onlyfourofthem (66.6%)wereinvasive.Thus,theseresultssuggestthatthepresenceoftpeLgeneorTpeL toxinmightberequiredfortheadherenceofbacteriatoHEp-2cells;however,itcouldnot haveanyroleintheinvasionprocess.

©2017SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.Thisis anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/

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

Clostridiumperfringensspore-forminggram-positivebacterium

isthecausativeagentofgaseousgangreneinhumans, entero-toxaemiaincattle,and necroticenteritis inchicken.Based onthe typeofpotentandlethaltoxins produced,theyare classifiedintofivetoxinotypes:fromAtoE.1–3

Largeclostridialcytotoxins(LCTs)compriseC.difficiletoxin A(TcdA)andtoxinB(TcdB),C.sordelliilethaltoxin(TcsL)and

∗ _{Correspondingauthorat:DepartmentofMicrobiology,UniversityofSaoPaulo,Av.Prof.LineuPrestes,1374,05508-900SaoPaulo,SP,}

Brazil.

E-mail:mariojac@usp.br(M.J.Avila-Campos).

1 _{Theseauthorscontributedequallytothework.}

hemorrhagic toxin (TcsH), C. novyi alphatoxin (TcnA), and toxinC.perfringenslargecytotoxin(TpeL),areimportant vir-ulencefactorsinpathogenesisofmyonecrosisandintestinal diseases.4–6

C.perfringensisabletoproducesotherpotenttoxinsand

enzymes, including NetB related tohuman and veterinary diseases.However,theroleoftheNetB-toxininthenecrotic

http://dx.doi.org/10.1016/j.bjm.2017.06.002

1517-8382/©2017SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

enteritisisstillcontroversialbecauseithasbeendetectedin bothhealthyandsickanimals.7

TpeLtoxinwasinitiallydetectedinC.perfringenstypeC iso-latedfromswineandinC.perfringenstypeAstrainsisolated fromchicken.ItislethaltomiceandcytotoxictoVerocells, causinganenlargement,rounded cells,and forming aggre-gates,eventuallydetachedcellsfromtheplate.4,8,9

Inearlystudy,chickensinoculatedwithC.perfringensTpeL (+)showedhighercapacitytocausesevereintestinallesions andanearlieronsetofsymptomsthanchickensnot inocu-latedwithTpeL.10_{ThissuggeststhatTpeLmaypotentiateor}

contributetothepathogenesisofnecroticenteritiscausedby

C.perfringenstypeAstrains.9

Adhesion process to epithelial cells is the first step to the bacterial colonization, and this is mediated by fim-briae or non-fimbriae appendices.11 _{C. prefringens is able}

to produce biofilm on cell surfaces and the presence of pili as well as the production of sialidase can collaborate with this process.12 _McClane13 _{showed that adherent C.}

perfringensstrainsincreasetheneuraminidaseandtoxin

pro-duction.

Inthisstudy,thepresenceofnetBandtpeLgenesinC.

per-fringenstypeAstrainsisolatedfrom chickenswithnecrotic

enteritisandtheircytotoxiceffectsandroleintheadhesion andinvasionoftheepithelialcellswereevaluated.

Chickensbelonged to five Brazilian states, Ceará (CE, 1 chicken),SaoPaulo(SP,2chickens),Paraná(PR,3fchicken), SantaCatarina(SC,2chickens),andRioGrandedoSul(RS,1 chicken).Twenty-twoC.perfringensstrainspreviouslyisolated fromthe intestinalsamplesofnine chickens withnecrotic enteritis,identifiedandtoxinotypedasC.perfringenstypeA bybiochemicaltests. Thepresenceofthe cpa gene codify-ingthetoxin␣inallC.perfringensstrainswasalsoevaluated byPCR.ThisstudywasapprovedbytheEthicsCommitteeof theBiomedicalSciencesInstitute,UniversityofSaoPaulo(No. 104/CEEA).

Bacterial DNA was obtained in accordance with Sam-brooketal.14_{Briefly,thebacteriagrowninto5mLbrainheart}

infusion(BHI) were harvested by centrifugation(14,000×g,

10min), and the aliquots of supernatant were maintained at −80◦_{C until use. The pellets were washed twice with}

0.1Mphosphate-bufferedsaline(PBS,pH7.2),andincubated with10mg/mLlysozymeat37◦Cfor3h.Then,20%SDSand 20mg/mLproteinase K were added and incubated at 55◦C for2h. TheDNA wasextracted byusing equalvolumesof phenol–chloroformandelutedin100␮LofTE.PCRassayswere performedtodetect netBand tpeLgeneswithfinalreaction volumesof25␮Lcontaining10×PCRbuffer,1.5mMMgCl2,

0.2mMdNTPmix,0.5UPlatinumTaqDNApolymerase (Invitro-gen),0.4mMofeachprimer5,15_{and1ngofDNA.Thermocycler}

(PEAppliedBiosystemsGeneAmpPCRSystem9700)was pro-grammed to: initial denaturation at94◦C (5min), followed by30 cyclesof94◦C(1min),55◦C(1min)and72◦C(1min), andafinalextensionat72◦C(7min).ThePCRproductswere analyzedon1%agarosegelstainedwithethidiumbromide (0.5mg/mL)andphotographedbyusingKodakDigitalSystem DC-120.DNAfromC.perfringensJGS5369strainwasusedas control.DNAwassequencedusingMegaBACE1000system. Sequencingdatawere analyzedbyusing NucleotideBLAST (NCBI,Bethesda,Maryland,USA).

6h

24h

48h

72h

B

A

D

C

F

E

H

G

J

I

Fig.1–CytotoxicofC.perfringenstypeAstrainonVerocells indifferenttimes.(A,C,E,G)C.perfringenstpeL(+)gene;(B, D,F,H)C.perfringenstpeL(−)gene;(I)Verocells(control);(J) VerocellsandBHI.TimesofIncubation:AandB,6h;Cand D,24h;EandF,48h;GandH,72h.Magnification:1000×.

The cytotoxic effect at different times (6, 24, 48, and 72h)was evaluated on Vero African GreenMonkey kidney cell line (Vero, ATCC CRL-81) cultured inDulbecco’s modi-fiedEagle’smedium(DMEM; Vitrocell,Embriolife,SP,Brazil) supplemented with 2% (v/v) fetal bovine serum (FBS). In each well of a24-well microplate,1mLof cell suspension (1×105_{cells/mL)wasdispensed,andincubatedat37}◦_Cin5%

CO2for24h.TheVerocellsweretreatedwiththesupernatant

ofthebacteriaculturedinBHIfor24h.Morphological alter-ationsincellswereassessedbyusingalightmicroscope.All assayswereindependentlyperformedtwiceinduplicate.

Thebacterialadherenceassayswereperformedbyusing HEp-2 cells (1×105_{cells/mL), as previously described by}

A

C

B

D

Fig.2–AdhesionassayofC.perfringenstypeAstrainonHEp-2cells.(A)HEp-2control;(B)E.coliO42(EAEC)strain;(C)C. perfringensATCC13124strain,tpeL(−)gene;(D)C.perfringens47dstrain,tpeL(+)gene.Magnification:1000×.

FBSwasaddedtoHEp-2cells,whichwasfurtherinoculated with100␮Lof24-hbacterialculture(c.a.1.5×108_cells/mL).

Plateswereincubated (in5%CO2,6h),washed threetimes

with PBS,fixed with absolute methanol, and stained with May–Grünwald–Giemsa stain. The enteroaggregative strain

EscherichiacoliO42displayinganaggregativeadhesionpattern

wasusedasacontrol.All assayswereperformedin dupli-cate.

The invasion assay was performed in accordance with Nakanoetal.,8_{asfollows:100␮Lofbacterialsuspensionwas}

addedtoHEp-2cells,andplateswereincubated(in5%CO2,

2h). The non-adherent bacteria were killed by incubating themin1mLofDMEMsupplementedwith200mg/Lof

cefox-itin(forC.perfringens)or300mg/Lofgentamicin(forE.coli)

at37◦C,in5%CO2for1h;theantibioticconcentrationsdid

notaffectthemorphologyoftheepithelialcells.Then,cells werewashed inPBS (threetimes)andlysedwith400␮Lof Triton X-100 (1%,v/v), thoroughlyand transferredtotubes containing1.6mLBHIbroth.Aliquotsof100␮Lofthiswere platedonBrucellabloodagar(DifcoLaboratories,UK), incu-batedinanaerobiosis(90%N2+10%CO2)at37◦Cfor48h,and

thecolonyformingunit(CFU)wasdetermined.An enteroin-vasivestrain E. coli serotype O124:NM, and a non-invasive strainE.coliHB101wereused,aspositiveandnegativecontrol,

respectively.Allassayswereperformedinduplicate.An inva-sionwasexpressedasthepercentageofbacteriarecovered fromtheinitialinoculumaftertheantibiotictreatmentand lysisoftheepithelialcellsasreportedbyTangetal.(1993).18

The statistical analysis was performed by using GraphPad Prismversion6.0(GraphPadSoftware,LaJollaCalifornia,USA)

and a p-value of 0.05 was considered as statistically

sig-nificant. TheCFUwere comparedbyusing unpairedT-test,

andtheotherdatawerestatisticallyevaluatedbyChi-square test.

Inthisstudy,noneofthestrainsharboredthenetBgene inaccordance withLlancoetal.5_{; although,itisimportant}

tonotethattheproductionofthistoxindoesnot necessar-ilygene-dependent.16_{ThepartialsequencingofthetpeLgene}

fromsix(27.3%)outof22strainsshowed99%sequence sim-ilaritywithC.perfringens(AB262081.1)tpeL(+).Thesestrains alteredmorphologyofVerocellswhichdisplayed character-isticcellenlargement(Fig.1A)androundedcells(Fig.1E)as early as 6h ofincubation, and this effect waspronounced after48hofincubation.Ontheotherhand,thesupernatant fromstrainstpeL(−)geneinducedastrongcellroundingafter 6hofincubation;and after72hthecells werefoundtobe detachedcompletelywithoutforminganyaggregates(Fig.1H). ThiseffectoftpeL(−)strainsmightbeduetothetoxin-␣.The

initialstretchingorelongationofVerocells(Fig.1A)noticed withtpeL(+)strainswasduetothecharacteristicactionofthe toxinTpeLorbyanysynergisticorcompetitiveeffect,since thesupernatantwasused,andinaccordancewithprevious studies.3,17

AllthetpeL(+)strainsandtwotpeL(−)strainswereable toadheretoHEp-2 cells;however,no significantdifference wasobserved(p=0.1336).Theadhesionprofiledisplayed non-localized clusters (Fig. 2C and D). The invasive capacities observedinfourofthesixadherenttpeLpositivestrainsand twotpeLnegativestrainswerecomparable(p=1.0000).All non-adherent strains also did notinvade HEp-2 cells.The tpeL

negativeC.perfringens ATCC13124andenteroinvasiveE. coli

O143wereabletoadheretoandinvadeHEp-2cells.Although the higher level of invasion was recorded with tpeL (+) (log103.26CFU/mL)thanintpeL(−)strains(log102.89CFU/mL),

therewasnostatisticallysignificantdifference(p=0.9080). There are few reports from Brazil in which the pres-ence ofTpeLtoxin-producingC. perfringens typeAisolated fromchickenswithnecroticenteritisarestudied.TpeLtoxin belongstoLCTsthatexhibitslethalityinmiceand cytotox-icityinvariouscelllinesincludingVerocells.18_Someofthe

strainsstudiedherewereabletoadheretoandinvade HEp-2cellswhichmighthavecontributedinthepathogenicityof thesestrains.StudieshaveshownbyusingRAPD-PCR,MLST, and PFGE methods that some clones ofC. perfringens iso-latedfromdifferentoriginsmightbeassociatedtoenteritis necroticand theycanpersistforlongtimeatthediseased sites.19,20

Theability ofbacteria to adhereepithelial cells is con-sideredasimportantvirulencefactorinvarious pathogens,

suchasC.difficile,SalmonellaTyphimurium,Shigellaspp.,and

Helicobacterpylori.21_{Sincemoststudieshavebeenexclusively}

focusedonthetoxin production,the adhesioninC. perfrin-genshasnotbeenadequatelyevaluated.22_{Inhistopathological}

studies of the intestinal tissue from necrotic enteritis, C.

perfringens wasfound tobeadhered to intestinalcells and

necroticareas,suggestingthatthisorganismproducestoxins afterintestinalcolonization.21

Once the six adherent strains evaluated in this study belongedtochickensfromdifferentBrazilianstates,itis possi-bletosuggestthatthesestrainstpeL-positivearedistributedin mostregionofBrazil;however,morestudiesareneeded evalu-atingahighnumberofchickensandstrains.Inaddition,since sixstrainswereadherentandfourofthemwereinvasive,it issuggestedthattheTpeLtoxinhasanyroleintheadhesion, butnottoinvasion. Studieshaveshown thatthe adhesion processofC.perfringenstoHEp-2cellscouldbemediatedby thepresenceofpilitypeIV,whichhasanimportantroleinthe biofilmformationontheepithelialcells.23_{Thepresenceofpili}

orotherappendiceswasnotevaluated.

Inthisstudy,wehaveidentifiedtpeLharboringC.perfringens

typeAstrainsisolatedfromchickenswithnecroticenteritis, andourresultssuggeststhepossiblecontributionofthistoxin inaggravationofthisdisease.Inconclusion,thepresenceof

tpeLgeneorTpeLtoxinwould beassociatedwithadhesion butnotwithinvasionprocess.Furtherstudiesarewarranted tocharacterizethemechanismsinvolvedintheadhesionand invasionofepithelialcellsbytpeLpositiveandnegativeC. per-fringens.

Conflicts

of

interest

The authors declareno potential conflicts ofinterest with respecttotheresearch,authorship,and/orpublicationofthis article.

Acknowledgements

TheauthorsthanktoMissMarciaH.Fukugaitiforhertechnical support.ClostridiumperfringensJGS5369waskindlyprovided byDr.J.G.SongeratUniversity ofArizona,USA. Thisstudy wassupportedbygrantofCNPqNo.158799/2012-7andFAPESP 2013/17739-9.

r

e

f

e

r

e

n

c

e

s

1.BallDW,VanTassellRL,RobertsMD,HahnPE,LyerlyDM,

WilkinsTD.Purificationandcharacterizationofalpha-toxin

producedbyClostridiumnovyitypeA.InfectImmun.

1993;61:2912–2918.

2.SongerJG.Clostridialentericdiseasesofdomesticanimals.

ClinMicrobiolRev.1996;9:216–234.

3.AmimotoK,NoroT,OishiE,ShimizuM.Anoveltoxin

homologoustolargeclostridialcytotoxinsfoundinculture

supernatantofClostridiumperfringenstypeC.Microbiology.

2007;153:1198–1206.

4.NagahamaM,OhkuboA,OdaM,etal.Clostridiumperfringens

TpeLglycosylatestheRacandRassubfamilyproteins.Infect

Immun.2011;79:905–910.

5.LlancoLA,NakanoV,FerreiraAJP,Avila-CamposMJ.

ToxinotypingandantimicrobialsusceptibilityofClostridium

perfringensisolatedfrombroilerchickenswithnecrotic

enteritis.IntJMicrobiolRes.2012;4:290–294.

6.LlancoLA,NakanoV,Avila-CamposMJ.Sialidaseproduction

andgeneticdiversityinClostridiumperfringenstypeAisolated

fromchickenwithnecroticenteritisinBrazil.CurrMicrobiol.

2015;70:330–337.

7.MartinTG,SmythJA.PrevalenceofnetBamongsomeclinical

isolatesofClostridiumperfringensfromanimalsintheUnited

States.VetMicrobiol.2009;136:202–205.

8.NakanoV,PiazzaRM,CianciarulloAM,etal.Adherenceand

invasionofBacteroidalesisolatedfromthehumanintestinal

tract.ClinMicrobiolInfect.2008;14:955–963.

9.NavaneethanU,GiannellaRA.Mechanismsofinfectious

diarrhea.NatClinPractGastrHepat.2008;5:637–647.

10.OlkowskiAA,WojnarowiczC,Chirno-TrejoM,DrewMD.

Responsesofbroilerchickensorallychallengedwith

Clostridiumperfringensisolatedfromfieldcasesofnecrotic

enteritis.ResVetSci.2006;81:99–108.

11.ParkerCT,SperandioV.Cell-to-cellsignalingduring

pathogenesis.CellMicrobiol.2009;11:363–369.

12.BorastonAB,Ficko-BleanE,HealeyM.Carbohydrate

recognitionbyalargesialidasetoxinfromClostridium

perfringens.Biochemistry.2007;46:11352–11360.

13.McClaneBA.ClostridiumperfringenstypeCisolatesrapidly

upregulatetheirtoxinproductionuponcontactwithhost

cells.Newinsightsintovirulence?Virulence.2010;1-2:97–100.

14.SambrookJ,FritischEF,ManiatisT.MolecularCloning:A

LaboratoryManual.2nded.USA:ColdSpringHarbor

LaboratoryPress;1989.

15.PetitL,GibertM,PopoffMR.Clostridiumperfringens:

toxinotypeandgenotype.TrendsMicrobiol.1999;7:104–110.

16.BradyJ,Hernandez-DoriaJD,BennettC,GuenterW,House

producingandcommensalisolatesofClostridiumperfringens

fromchickensfedorganicdiets.AvianPathol.

2010;39:475–481.

17.RichardJF,PetitL,GilbertM,MarvaudJC,BouchaudC,Popoff

MR.Bacterialtoxinsmodifyingtheactincytoskeleton.Int

Microbiol.1999;2:185–194.

18.TangP,FoubisterV,PucciarelliMG,FinlayBB.Methodsto

studybacterialinvasion.JMicrobiolMethods.1993;18:227–240.

19.StackebrandtE,KramerI,SwiderskiJ,HippeH.Phylogenetic

basisforataxonomicdissectionofthegenusClostridium.

FEMSImmunolMedMicrobiol.1999;24:253–258.

20.BakerA,DavisE,RehbergerT,RosenerD.Prevalenceand

diversityoftoxigenicClostridiumperfringensandClostridium

difficileamongswineherdsinthemidwest.ApplEnviron

Microbiol.2010;76:2961–2967.

21.VargaJJ,TheritB,MelvilleSB.TypeIVpiliandtheCcpA

proteinareneededformaximalbiofilmformationbythe

Gram-positiveanaerobicpathogenClostridiumperfringens.

InfectImmun.2008;76:4944–4951.

22.RoodJI,ColeST.Moleculargeneticsandpathogenesisof

Clostridiumperfringens.MicrobiolRev.1991;55:621–648.

23.CoursodonCF,GlockRD,MooreKL,CooperKK,SongerJG.

TpeL-producingstrainsofClostridiumperfringenstypeAare

highlyvirulentforbroilerchicks.Anaerobe.2012;18: