<%BANNER%>






Ureaplasma parvum infection alters filamin A dynamics in host cells
http://www.biomedcentral.com ( Publisher's URL )
CITATION PDF VIEWER
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/IR00000587/00001
 Material Information
Title: Ureaplasma parvum infection alters filamin A dynamics in host cells
Series Title: BMC Infect Dis. 2011 Apr 20;11:101
Physical Description: Journal Article
Creator: Reyes, Leticia
Publisher: BioMed Central Ltd
Place of Publication: 236 Gray's Inn Road
London WC1X 8HB
Publication Date: April 20, 2011
 Subjects
Genre:
Spatial Coverage:
 Notes
Abstract: Background: Ureaplasmas are among the most common bacteria isolated from the human urogenital tract. Ureaplasmas can produce asymptomatic infections or disease characterized by an exaggerated inflammatory response. Most investigations have focused on elucidating the pathogenic potential of Ureaplasma species, but little attention has been paid to understanding the mechanisms by which these organisms are capable of establishing asymptomatic infection. Methods: We employed differential proteome profiling of bladder tissues from rats experimentally infected with U. parvum in order to identify host cell processes perturbed by colonization with the microbe. Tissues were grouped into four categories: sham inoculated controls, animals that spontaneously cleared infection, asymptomatic urinary tract infection (UTI), and complicated UTI. One protein that was perturbed by infection (filamin A) was used to further elucidate the mechanism of U. parvum-induced disruption in human benign prostate cells (BPH-1). BPH-1 cells were evaluated by confocal microscopy, immunoblotting and ELISA. Results: Bladder tissue from animals actively colonized with U. parvum displayed significant alterations in actin binding proteins (profilin 1, vinculin, a actinin, and filamin A) that regulate both actin polymerization and cell cytoskeletal function pertaining to focal adhesion formation and signal transduction (Fisher’s exact test, P < 0.004; ANOVA, P < 0.02). This phenomenon was independent of clinical profile (asymptomatic vs. complicated UTI). We selected filamin A as a target for additional studies. In the BPH-1 model, we confirmed that U. parvum perturbed the regulation of filamin A. Specifically, infected BPH-1 cells exhibited a significant increase in filamin A phosphorylated at serine2152 (P ≤ 0.01), which correlated with impaired proteolysis of the protein and its normal intracellular distribution. Conclusion: Filamin A dynamics were perturbed in both models of infection. Phosphorylation of filamin A occurs in response to various cell signaling cascades that regulate cell motility, differentiation, apoptosis and inflammation. Thus, this phenomenon may be a useful molecular marker for identifying the specific host cell pathways that are perturbed during U. parvum infection.
Acquisition: Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Leticia Reyes.
Publication Status: Published
Funding: The project described was supported by Award Number K08DK075651 from the National Institute of Diabetes And Digestive And Kidney Diseases (NIDDK). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIDDK or the National Institutes of Health.” Experimental portions of this project were also supported by an intramural grant obtained through the University of Florida, College of Veterinary Medicine. Publication of this article was funded in part by the University of Florida Open-Access Publishing Fund.
 Record Information
Source Institution: University of Florida Institutional Repository
Holding Location: University of Florida
Rights Management: All rights reserved by the submitter.
System ID: IR00000587:00001

Downloads

This item is only available as the following downloads:

1471-2334-11-101 ( PDF )


Full Text

PAGE 1

RESEARCHARTICLE OpenAccessUreaplasmaparvum infectionaltersfilamina dynamicsinhostcellsAymanBAllam1,SophieAlvarez2,3,MaryBBrown1andLeticiaReyes1*AbstractBackground: Ureaplasmas areamongthemostcommonbacteriaisolatedfromthehumanurogenitaltract. Ureaplasmas canproduceasymptomaticinfectionsordiseasecharacterizedbyanexaggeratedinflammatory response.Mostinvestigationshavefocusedonelucidatingthepathogenicpotentialof Ureaplasma species,but littleattentionhasbeenpaidtounderstandingthemechanismsbywhichtheseorganismsarecapableof establishingasymptomaticinfection. Methods: Weemployeddifferentialproteomeprofilingofbladdertissuesfromratsexperimentallyinfectedwith U. parvum inordertoidentifyhostcellprocessesperturbedbycolonizationwiththemicrobe.Tissuesweregrouped intofourcategories:shaminoculatedcontrols,animalsthatspontaneouslyclearedinfection,asymptomaticurinary tractinfection(UTI),andcomplicatedUTI.Oneproteinthatwasperturbedbyinfection(filaminA)wasusedto furtherelucidatethemechanismof U.parvum -induceddisruptioninhumanbenignprostatecells(BPH-1).BPH-1 cellswereevaluatedbyconfocalmicroscopy,immunoblottingandELISA. Results: Bladdertissuefromanimalsactivelycolonizedwith U.parvum displayedsignificantalterationsinactin bindingproteins(profilin1,vinculin, a actinin,andfilaminA)thatregulatebothactinpolymerizationandcell cytoskeletalfunctionpertainingtofocaladhesionformationandsignaltransduction(Fisher sexacttest,P<0.004; ANOVA,P<0.02).Thisphenomenonwasindependentofclinicalprofile(asymptomaticvs.complicatedUTI).We selectedfilaminAasatargetforadditionalstudies.IntheBPH-1model,weconfirmedthat U.parvum perturbed theregulationoffilaminA.Specifically,infectedBPH-1cellsexhibitedasignificantincreaseinfilaminA phosphorylatedatserine2152(P 0.01),whichcorrelatedwithimpairedproteolysisoftheproteinanditsnormal intracellulardistribution. Conclusion: FilaminAdynamicswereperturbedinbothmodelsofinfection.PhosphorylationoffilaminAoccurs inresponsetovariouscellsignalingcascadesthatregulatecellmotility,differentiation,apoptosisandinflammation. Thus,thisphenomenonmaybeausefulmolecularmarkerforidentifyingthespecifichostcellpathwaysthatare perturbedduring U.parvum infection.BackgroundUreaplasmaparvum and U.urealyticum areamong themostcommonbacteriaisolatedfromthehuman urogenitaltract[1-3].Infectionratesashighas40to 80%inwomenandupto50%inmenhavebeen reported[3].Mostinfectionsofthelowerurogenital tractappeartobeasymptomatic[1].However,both speciesof Ureaplasma arealsoresponsibleforavarietyofdiseasessuchaschori oamnionitis,spontaneous abortion,prematurebirth, stillbirth,pos tpartumendometritis,neonatalneuropathies,andpneumoniawith bronchopulmonarydysplasia[1,2,4,5].Ureaplasmasare alsoimplicatedinawiderangeofurinarytractdiseasesincludingurinarytractinfection(UTI)[6], encrustedcystitis[7],urethritis[8],chronicprostatitis [9],andurolithiasis[10].Mostinvestigationshave focusedonelucidatingthepathogenicpotentialof Ureaplasma species,butlittleattentionhasbeenpaid tounderstandingthemechanismsbywhichthese organismsarecapableofest ablishingasymptomatic infection. *Correspondence:lreyes@ufl.edu1DepartmentofInfectiousDisease&Pathology,CollegeofVeterinary Medicine,UniversityofFlorida,Gainesville,FL,USA FulllistofauthorinformationisavailableattheendofthearticleAllam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 2011Allametal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.

PAGE 2

Wepreviouslydevelopedanexperimentalmodelof UTIthathasprovidedsomeinsightsintothehostfactorsassociatedwithasymptomaticinfectionandcomplicateddisease[11-13].Specifically,geneticallyinbred Fisher(F344)ratsthatwereexperimentallyinoculated with U parvum developedthreeclinicaloutcomes. Approximatelyonethirdofi noculatedanimalsspontaneouslyclearedinfectionfromtheurinarytractby2 weekspostinoculation.Theanimalsthatremained infectedexhibitedtwodistin ctclinicalprofilesofUTI: asymptomaticinfectionori nfectioncomplicatedbyan exaggeratedinflammatoryre sponsewithbladderstone formationassequela[11-13]. U.parvum organismscan befoundcolonizingtheuroe pithelialsurfaceinboth clinicalprofilesofUTI.However,inanimalswithcomplicatedUTI, U.parvum canalsobefoundwithinthe submucosaofthebladder,whichmaybethedriving forcebehindthepersistentandexaggeratedinflammatoryresponse. Anotablefeatureinanima lswithasymptomaticUTI wasthepresenceofquiescenturoepitheliumdespitethe presenceof U.parvum ,whichisincontrasttowhat occurswithUTIcausedbyotherbacterialspecies [14,15].Basedonourobservations,wepostulatedthat ureaplasmasperturburoepithelialfunctioninamanner thatinterfereswithinnateimmunedefenseandsupports microbialcolonization.Inordertobegintoidentifythe hostcellprocessesthatareperturbedby Ureaplasma duringinfection,weuseddifferentialproteomicstoprofilebladdertissuesfromF344ratsexperimentallyinoculatedwith U.parvum .Tissuesfromeachclinicalprofile (shaminoculatedcontrol,culturenegativeanimals,animalswithasymptomaticUTI,andcomplicatedUTI group)wereanalyzedinthisstudy. Inthisreportweshowthatbladdertissuefromanimalsactivelycolonizedwith U.parvum displaysignificantalterationsinactinbindingproteinsthatregulate bothactinpolymerizationandcellcytoskeletalfunction pertainingtofocaladhesionformationandsignaltransduction.Thisphenomenonis independentofclinical profile(asymptomaticvs.complicatedUTI).Weselected theactin-bindingproteinfilaminAasatargetforadditionalstudiesbasedonproteomeprofilingresultsas wellasitsintegralroleincellsignalingeventsrelatedto innateimmunity[16,17].Weevaluatedtheimpactof U. parvum infectiononfilaminAusingthebenignprostate hyperplastic(BPH-1)celllineasamodelofinfection.In theBPH-1model,weconfirmedthat U.parvum perturbedtheregulationoffilam inA.Specifically,infected BPH-1cellsexhibitedasignificantincreaseinfilaminA phosphorylatedatserine2152,whichcorrelatedwith impairedproteolysisoffilaminAanditsintracellular distribution.MethodsSampleselectionandproteinextractionforratbladder proteomestudiesRatbladdertissuesfromanimalsexperimentallyinfected witharatadaptedstrainof U.parvum wereusedfor differentialproteomeprofiling[12].Allprocedureswere performedinaccordancewiththeUniversityofFlorida InstitutionalAnimalCareandUseCommittee.Briefly, animalswereinoculatedwit hsterile10Bbroth(control group)or109CFUof U.parvum .Tissueswerecollected 2weekspost-inoculation[12].Attimeoftissueharvest, bladderfromeachanimalwasdividedinto3sections. Onesectionwasusedforhist opathology,anothersectionwasculturedfor U.parvum ,andthethirdsection wasflashfrozeninliquidnitrogenandstoredat-80C forfutureanalysis. Forproteomestudies,tissuesweregroupedaccording toclinicalprofile.Thenegativegroupincludedanimals inoculatedwith109CFUof U.parvum thatwereculture negativeintheurinarytractat2weekspostinoculation (PI).Basedonacuteinfectionstudies[11],100%ofanimalsinoculatedwiththisdosewereinfectedwith U. parvum foratleast72hours,thereforethenegative groupwascomposedofanimalsthatspontaneously clearedinfection.AnimalswithintheUTIgroupwere culturepositiveattimeofnecropsy,hadminimalhistologicchangesinbladdertis sue[12],lowurinecytokine levels,andnoevidenceofstruviteuroliths.Animals withintheStruvitegroupwereculturepositiveattime ofnecropsy,hadextensivehistologicchangesinbladder tissue,werepositiveforstruvites,andhadmarkedelevationsinurinepro-inflammatorycytokines.Animals sham-inoculatedwithsterile10Bbrothservedasuninfectedcontrols.Inordertominimizevariabilityof U. parvum proteinloadbetweeninfectedgroups,onlytissuesthathadsimilarlogCFU(2.4to2.7)attimeof necropsywerechosenfortheseexperiments.Protein fromtissueswasextractedwithTrizol(Invitrogen Corp.,Carlsbad,CA)accordingtothemanufacturer s protocol.Pelletedprotein extractswereallowedtoair dryandstoredat-20Cbeforeanalysis.Quantitativeproteomicanalysisusingpeptide-labeling andoffline2D-LC-MS/MSThreeindependentiTRAQexperimentswereperformed. Eachexperimentincludedon ebiologicalreplicatefrom acontrol,negative,UTI,andstruvitegroup.Sample processing,offline2D-LC-MS/MS,proteinidentification andquantitationwereperformedaspreviouslydescribed [11].TandemmassspectrawereextractedbyAnalyst(v 1.1.;AppliedBiosystems/MDSSciex).Concatenationof theforwardandrandomsequencesfromtheIPIrat databasev3.32[18]wereusedforproteinidentification.Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page2of13

PAGE 3

Proteinidentificationsearcheswereperformedusing MS/MSdatainterpretationalgorithmsfromProtein Pilot (Paragon algorithm,v2.0,AppliedBiosystems/MDSSciex)[19]andMascot(v2.2,Matrix Science,London,UK).Theconfidencelevelforprotein identificationwassetto1. 3(95%).Thefalsediscovery rateforalliTRAQ experimentsrangedfrom0.0%to 0.93%.ProteinratiosweregeneratedwithProGroup algorithmandautomaticallyc orrectedforbias.Protein quantificationwasperformedwithaminimumofthree spectrathatwerepresentinallproteinsamplesthat wereanalyzedwithintheexperiment.Onlyprotein ratioswithanerrorfactor(EF)<2wereretainedfor furtheranalysis.EFisameasureofthevariationamong thedifferentiTRAQ ratios(thegreaterthevariation, thegreatertheuncertainty)andrepresentsthe95% uncertaintyrangeforareportedratio.ThecalculatedPvalueobtainedwiththeProGroup algorithmisbased on95%confidenceinterval.Enrichmentanalysisofproteinratiosthatcomparedthe Negativegrouptoanimalswithactiveinfection(UTIand Struvitegroups)Sinceourinterestwastoidentifyperturbationsthat werepresentinallanimalswithactiveUTI,regardless oftheclinicalprofile,weonlyincludedproteinratios thatexhibitedthesameresponsetoinfectioninall infectedgroups.Forexample,iftheproteinratiowas decreasedinbothUTIandSt ruvitegroups,thenthat proteinwasincluded.Iftheratiowasincreasedinthe UTIgroupbutdecreasedinth eStruvitegroup,then thatproteinwasconsideredtopotentiallybeaffectedby thehostinflammatoryresponseandwasexcludedfrom theanalysis.Proteinsweregroupedaccordingtogeneral biologicalfunctionsasassignedintheUniprot/Swissprot database.ProteinratioswereconsideredsignificantlydifferentiftheyhadPvalueslessthan0.05asdetermined bytheProGroup algorithm;theseproteinswere assignedabinaryvalueof1.Allothersratioswereconsideredinsignificantandwereassignedabinaryvalueof 0.FisherexacttestwithBonferronicorrectionformultiplecomparisonswasusedtoidentifyanybiological functioncategoriesthatweresignificantlyoveror underrepresentedinanimalswithactiveUTIcompared toanimalsthatspontaneouslyclearedinfection.EnrichmentanalysiswasperformedwithJMPGenomics3.0 software(SASInstituteInc.,Cary,NC).ANOVAandhierarchicalclusteringofratbladdertissue proteomeprofilesANOVAwasusedtoidentifytheproteomeprofilesthat werecommonamonganimalswithactiveUTIbutsignificantlydifferentfromanimalsthatclearedinfection (Negativegroup).ProteinratioscomparingNegativeto control,UTItocontrol,andStruvitetocontrolwere generatedwiththeProGroup algorithm.Proteome datasetsthatcontainedonlyproteinswithanEF<2 andthatalsowereidentifiedinallthreeindependent iTRAQexperimentswereanalyzedwithJMPGenomics 3.0software(SASInstituteInc.,Cary,NC).Thequality ofthedatawasassessedbydistributionanalysis,box plotsandkerneldensityestimatesandstandardized priortoANOVA(rowbyrowmodeling).ANOVAwas performedwithafalsediscoveryratesetat a =0.05. Theleastsquaresmeansofproteinsthatsignificantly differedamonggroups(P<0.03)werethenclusteredby themethodofWard(proteintoproteinwithingroup, aswellasgrouptogroup).Infectionstudieswithbenignprostatehyperplasticcells (BPH-1)BPH-1cellswereagiftfromDr.CharlesRosser(MD AndersonCancerCenter,Orlando,FL).ForallexperimentsBPH-1cellswereculturedat37Cin5%CO2in completeRPMIcontaining10%fetalcalfserum.For eachexperiment,thenumberofviablecellswasdeterminedbytrypanbluestaining.Cellnumberswere adjustedtoaconcentrationof107cellspermlandplatedin6wellplates(CorningInc.,Lowell,MA)orLabTekII8wellglassslidechambers(Nuncbrand,Rochester,NY).After24hours,th ecellculturemediumwas changedwithfreshcompleteRPMIandcellswere checkedfor50to60%confluencepriortoinfection with U.parvum .Eachexperimentcontainedthreebiologicalreplicateswithineachtreatmentgroup,andeach experimentwasrepeatedatleasttwice. Forsupernatantexperiments,10mlofRPMImedium washarvestedfromBPH-1cellsthatwereexposedto sterile10Bbrothor109CFUof U.parvum for24 hours.Harvestedcellsupernatantswerefirstclarifiedby centrifugationat15,000gat4Cfor10minutesto removeanycelldebris.Theclarifiedsupernatantwas thenfilteredthroughsterile0.1 msyringefiltersto removeanybacteria.BPH-1cellcultureswereinoculatedwithprocessedsupernatantsandmaintainedat37 Cand5%CO2for72hoursbeforeharvesting.Preparationandcultureof U.parvumForallBPH-1cellinfectionexperiments,ourrat adaptedstrainof U.parvum [13]wasgrowntomidlog phase(approximately14hours),whichwasconfirmed byopticaldensityreadingobtainedat550nm.Thelog CFUofeachinoculumwasalsoconfirmedbyculture onA8agaraspreviouslydescribed[13].ImmunocytologicassaysAntiU.parvum rabbitpolyclonalantibody(agiftfrom Dr.JanetRobertson,MedicalMicrobiologyandAllam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page3of13

PAGE 4

Immunology,UniversityofAlberta)wasusedtodetect bacteriainBPH-1cellcultures.RabbitmonoclonalantibodycloneEP2405Y(Epitomics,Burlingame,CA)that recognizestheCterminalregionoffilaminAwasused todetectintact,andcleavedformsofthemolecule. Mousemonoclonalanti-filamin1(cloneSPM182from SantaCruzBiotechnology,Inc,SantaCruz,CA)was usedtodetecttheintactformofthewholemolecule. RabbitmonoclonalantibodycloneEP2310AYwasused todetectFilaminAphosphorylatedatserine2152(AbcamInc.,Cambridge,MA).RabbitandmouseIgG isotypecontrolswereusedtoassessnon-specificbinding ofprimaryantibodies(ThermoScientific,Fremont,CA). Fordetectionpurposes,secondaryantibodieswereAlexa Fluor-488goatanti-mouseIgGandALEXAFluor-594 goatanti-rabbitIgG(Invitrogen,Corp.,Carlsbad,CA). NucleiwerestainedwithDAPIandpolymerizedactin wasstainedwithPhalloidinlabeledwithAlexaFluor488(Invitrogen,Corp.,Carlsbad,CA). Cellsgrownonsterileglassslideswerefixedfor3 minuteswith3.7%formaldehydeinphosphatebuffered saline(PBS),andprocesse dforimmunofluorescent stainingaspreviouslydescribed[11].ImageswerecapturedwithOlympusIX81-DSUSpinningDiskconfocal MicroscopeusingSlidebooksoftware(Olympus,Center Valley,PA). Fordeterminationof U.parvum infectionrates,foci thatdemonstratedcoloca lizationofDAPIwithantiU parvum antibodywerecountedaspositive.ForquantificationofcellswithnormalintracellularfilaminAdistribution,cellsthatexhibiteds trongintranuclearstaining offilaminAwerecountedaspositive.Forallstudies,a minimumoffivebiologicalreplicateswereevaluated andatleast200cellspersamplewerecounted.PreparationofwholecelllysatesforELISAAdherentcellsweregentlywashedtwicewithsterile PBS.Cellswerethenlysedwith1mloficecoldlysis buffer[50mMTris,pH7.5,0.15MNaCl,2mM EDTA,1mMEGTA,1%Triton-X]supplementedwith HALTproteaseinhibitorandHALTphosphataseinhibitor(PIERCEChemicals,Rockford,IL.).Cellsuspensions weretransferredtosteriletubesandsonicatedwith three10secondpulses(SonicDismembratormodel500, FisherScientific,Pittsburgh,PA).Disruptedcellsuspensionsweredividedinto200 laliquotsandstoredat -80C.DetectionoftotalandphosphorylatedfilaminAbyELISAOptEIAELISAreagentkitB(BDBiosciences,San Diego,CA)wasusedtoperformtheassayaspreviously described[11].Reagentsusedinthisstudyweremouse monoclonalanti-filamin1(cloneSPM182fromSanta CruzBiotechnology,Inc,SantaCruz,CA)forcapture,c terminalfilaminArabbitmonoclonal(Epitomics,Burlingame,CA)fordetectionoftotalfilaminA,andphosphoS2152filaminArabbitpolyclonalantibody(ab75978 fromAbcamInc.,Cambridge,MA)todetectthephosphorylatedprotein.Fornormalizationpurposes,the totalproteinconcentrationofeachsamplewasdeterminedbymicroBCAproteinassay(PierceChemicals, Rockwood,MD).Absorbancevalues(ABS)obtainedby ELISAweredividedbytheirtotalproteinconcentration sothatvaluesarereportedasABS/mgoftotalprotein.WesternblotanalysisFilaminAwasdetectedwithrabbitmonoclonalanti-C terminalantibody(Epitomics,Burlingame,CA).Calpastatin,calpain,andGAPDH weredetectedwithrabbit polyclonalantibodies(Abcam,Cambridge,MA). GAPDHwasusedasaloadingcontrol.Nuclearand cytoplasmicfractionswerepreparedwithNE-PER nuclearandcytoplasmicex tractionkitsupplemented withHALT proteaseinhibitorandHALT phosphataseinhibitor(ThermoScientific,Rockford,IL). CellextractswereloadedontoaNuPAGE10%BisTrisgel(Invitrogen,Carlsbad,CA)andsubjectedto105 Vfor1.5hoursinMOPS-SDSrunningbuffer.Protein transfertonitrocellulose membranefilterpaper,0.45 mporesize(Invitrogen,Carlsbad,CA),wasperformed at30Vfor1.5hoursintransferbuffer[Bicine25mM, Bis-Tris25mM,EDTA1mM,and10%methanol]. DetectionwasperformedwithSuperSignalWestPico CompleteRabbitIgGDetectio nKit(ThermoScientific, Rockford,IL)accordingtomanufacturer sinstructions. ChemiluminescencewasdetectedwithChemiDoc Imagingsystemanddensitometryofimagedbandswas performedwithQuantityOnev4.6.9software(Biorad LaboratoriesInc.HerculesCA).StatisticaldataanalysisofBPH-1cellexperimentsDatafrommultipleexperime ntsweregroupedtogether inordertomakestatisticalanalysispossible.Datawere analyzedbyone-wayANOVAwhenmorethantwo groupswereincludedintheanalysis.Fisher sProtected LeastSignificantDifference(PLSD)testwasusedwhen ANOVAindicatedasignificantdifferenceamonggroup means.Unpairedstudent sttestwasusedforcomparisonsthatwerelimitedtotwogroups.Forallanalyses,a probabilityofP 0.05wasconsideredsignificant.ResultsProteomeprofilingofF344ratbladdertissuesTwoapproaches,enrichmentanalysisandANOVA, wereusedtoidentifyproteinsthatdisplayedthesame responseto U.parvum infectionregardlessoftheclinicalprofile(UTIorStruvite).Inthefirstapproach, enrichmentanalysiswasperformedonproteinratiosAllam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page4of13

PAGE 5

thatweregeneratedbycomparingtheNegativegroupto animalsthatwerestillactivelyinfectedwith U parvum (bothUTIandStruvitegroups).Onlyproteinsthatwere identifiedinallthreeinde pendentiTRAQexperiments andthatalsoexhibitedacommoneffectamonganimals activelyinfectedwith U.parvum wereusedforthisanalysis.Usingthesecriteria,28of84proteinsexhibited bothasignificant(P<0.05)andacommoneffectwith U.parvum colonization(listedinAdditionalfile1, TableS1).Thedistributionoftheseproteinsaccording totheirbiologicalfunctionissummarizedinFigure1a and1b.Enrichmentanalysisrevealedthatanimalscolonizedwith U.parvum exhibitedasignificantchangein proteinsthatregulateactinpolymerization(P<0.004, withBonferronicorrection).Theseactin-regulatingproteinswereprofilin1,filaminA, a actinin,vinculin,spectrinandtalin.Withtheexceptionofprofilin1,allactin bindingproteinsweresignificantlylowerinanimals colonizedwith U.parvum (AdditionalFile1,TableS1). Inthesecondanalysis,wecomparedtheproteomeprofileofanimalsinoculatedwith U parvum totheproteomeprofilefromshaminoculatedcontrols.All proteinratios(Negative/control,UTI/control,andStruvite/control)thathadanEF<2andthatwereidentified inallthreeiTRAQexperimentswerethenanalyzedby ANOVA.Twenty-eightof315proteinsdisplayeda 7% 7% 7% 11% 11% 11% 11% 4% 7% 3% Endocrine 0% Protein processing 0% lipid 0% Actin dynamics**22% C A 11% 9% 2% 11% 4% 7% 14% Protein processing 22% Actin dynamics 0% Transport 0% 7% 2% 7% 4%B Inflammation Endocrine Apoptosis /Immunity Extracellular matrix/StructuralSignal transduction Transport Oxidative metabolismLipid metabolism Actin dynamics Protein processing Translation Carbohydrate metabolism Unclassified Genetic process Figure1 ProteomeprofilingofF344rattissuesinoculatedwithsterilebrothor U.parvum .PanelsAandBrepresentthepercentof proteinsassignedtoeachbiologicalfunctiongroup.GeneontologydesignationswereobtainedfromtheUniprot/SwissprotDatabase.Protein ratiosofeachspecificproteinarefromUTIandStruvitegroupsdividedbytheNegativegroup(n=3).GraphAshowsthedistributionof proteinratiosthatexhibitedasignificantdifferencebyProGroup algorithm(P<0.05).**Biologicalfunctioncategoriesthatweredetermined tobesignificantlydifferentbyenrichmentanalysis.GraphBshowsthedistributionofproteinratiosthatwerenotsignificantlydifferentbetween Negativegroupandgroupsthatwereculturepositivefor U.parvum (UTIandStruvite).PanelCisahierarchicalclusterofstandardizedleast squaresmeansthatweresignificantlydifferent(P<0.02)amongNegative,Struvite,andUTIgroups.RatiosgeneratedwiththeProGroup algorithmwereanalyzedbyANOVA(n=3)withafalsediscoveryrate a =0.05. Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page5of13

PAGE 6

significantlydifferentpatternamongNegative,UTI,and Struvitegroups(P<0.03).Theseproteinsarelistedin Additionalfile1,TableS2.Theleastsquaredmeansof the28proteinswereclusteredandarepresentedinFigure1C.Proteinclusters5,6,and7showedasimilar patterninUTIandStruvitegroupswhichwasmarkedly differentfromtheNegativegroup.Cluster5contained apolipoproteinA-Iprecursor,peptidyl-prolylcis-trans isomeraseBprecursor,andcalmodulin.Cluster6containedisocitratedehydrog enase[NADP],EH-domain containing2protein, a -enolase,peroxiredoxin-2,creatinekinaseB-type,andcomplementC3precursor.Cluster7containedelongationfactor1a a -actinin,vinculin, andfilaminA. Table1liststhe8proteinsfoundtobesignificantly affectedby U.parvum asdetectedbybothenrichment analysisandANOVA.Fouroftheseproteins(profilin 1, a actinin,vinculin,andfilaminA)areinvolvedin theregulationofactinpolymerization.Withthe exceptionofprofilin1,theseactinbindingproteins weresignificantlydecreasedinanimalswithactive infection. BecausefilaminAisoneofthebettercharacterized proteinsandisinvolvedinregulatingbothsignal transductionandgeneexpression[16,17],wechoseto focusourattentiononthiskeyprotein.Weevaluated thedistributionoffilaminAinratbladdertissuesby immunofluorescentstainingwiththemonoclonalantibodyspecificforintactfila minA(seeAdditionalfile 2).Therewasnoappreciabledifferenceintheoverall intensityordistributionoffilaminAstaininginthe tissuesofuninfectedandinfectedanimals.Therefore, inordertofurthercharacterizetheeffectof U.parvum colonizationonfilaminAdynamicsofhostcells,itwas necessarytoexpandourstudiestoacellculture system.InfectionofhumanBPH-1cellswith U.parvumWeestablishedamodelofinfectionusingtheBPH-1 immortalized,differentiatede pithelialcellline[20,21]. Inadditiontobeingamenabletochronicinfectionwith mycoplasmas[22],prostatecellsexhibitadistinctintracellulardistributionoffilaminAthatfacilitatesevaluatingitsdynamics[22]. Inpreliminarystudies,thefrequencyof U.parvum colonizationofBPH-1cellswasevaluatedbyconfocal microscopyat24,48,and72hours.Inallexperiments, U.parvum organismswereconsistentlyfoundonthe hostcellmembrane(seeAdditionalfile3). U parvum colonizationofBPH-1cellswasdetectedbybothDAPI stainingand U.parvum specificantibodylabeling.At24 hours, U.parvum couldbedetectedon973.5% (meanSD,n=4)ofcellsthatwereevaluatedintwo separateexperiments.At48hours,colonizationrates werereducedto60.515%.By72hours,517.8%of cellswerecolonized.Weselectedthe72hourpost inoculationtimepointforalladditionalexperiments sincecolonizationratesappearedtostabilizebythat timepoint.IntracellulardistributionoffilaminAinBPH-1cellsWeevaluatedtheintracellulardistributionoffilaminA inuninfectedandinfectedBPH-1cellsbyconfocal microscopyandimmunoblotting.Uninfectedcells showedapunctatepatternoffilaminAwithinthe nucleus(seeFigure2a)thatwasseenonlywiththe antibodythatrecognizedthecleavedformoffilamin A.Whenamonoclonalantibodythatrecognizedonly intactfilaminAwasused,nuclearfilaminAappeared tobepresentinastriatedformthatresembledactin stressfibers.Thiswasconf irmedwithco-localization studiesoffilaminAwithpolymerizedactinasshown inAdditionalfile3.Thisin tracellulardistributionof Table1Listofproteinsfoundtobeperturbedinanimalswithactive U.parvum infectiona).Accessionb)Proteinname GeneOntologyBiologicalProcessc)Effectd)IPI00194097.5GcVitaminD-bindingprotein transportsvitaminDanditsmetabolites IPI00193485.2Isocitratedehydrogenase [NADP] carbohydratemetabolism IPI00767147.1SimilartoAlpha-enolase carbohydratemetabolism IPI00195372.1Elongationfactor1-alpha1 proteinbiosynthesis/translationalelongation IPI00231358.6Profilin-1 actinbinding IPI00454431.1Brain-specificalphaactinin1 actinbinding IPI00365286.3SimilartoVinculin cellmotility/celladhesion/lamellipodiumbiogenesis IPI00409539.3SimilartoFilamin-Aactinfilamentbinding/actincytoskeletonreorganization/glycoproteinbinding/cytoplasmic sequesteringofprotein/regulationoftranscriptionfactor/regulationofI-kappa b kinase/NFb cascade a)Proteinsfoundtobesignificantlyalteredbybothenrichmentanalysis(Figure1AandB)andANOVA(Figure1b). b)AccessionnumbersrefertotheInternationalProteinDatabase,http://www.ebi.ac.uk/IPI/IPIhelp.html. c)GeneontologydatawasobtainedfromthePantherdatabasehttp://www.pantherdb.organdtheUniprotKB/Swiss-ProtDatabasehttp://www.ebi.ac.uk/uniprot. d)EffectreferstotheproteinratiothatwasgeneratedbytheProGroup algorithm,whichcomparedactivelyinfectedanimals(UTIandStruvitegroups)to NegativegrouporControlgroup.ActualproteinratiosarelistedinAdditionalfile1,tablesS1andS2.Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page6of13

PAGE 7

Figure2 IntracellulardistributionintactandcleavedfilaminAinBPH-1cells .Cellswereexposedtosterile10Bbroth,109CFUof U. parvum (UP),orcellculturesupernatant(super)for72hoursbeforeexaminationbyconfocalmicroscopy(A),Westernblot(B)anddensitometry (C).Confocalimagesweretakenat600magnificationandthescalebarisequalto10 m.CleavedandintactfilaminA(FilA)werestained withrabbit-antiCterminalfilaminA(red).IntactFilAwasstainedwithmouseanti-filamin1(green).BPH-1nucleiand U.parvum (whitearrow) wereidentifiedwithDAPIstain(blue).WesternblotanalysisforthedetectionofcleavedfilaminAwasperformedoncytosolic(cyt)andnuclear (nuc)fractionsfromuninfected(BPH)andinfected(UP)cells.TheblackarrowisdelineatingGAPDH,whichwasusedasaloadingcontrolanda confirmationthatthenuclearfractionwasnotcontaminatedwithcytosolicproteins.QuantitationofintactfilaminAwasperformedby densitometryofthecytosolicfractionsofuninfectedand U.parvum infectedcells.Theaveragequantitywithineachblotwasnormalizedby dividingtheaveragequantityoffilaminAproteinbandbytheaveragequantityoftheGAPDHband.ValuesrepresentthemeanSDof3 replicatesfrom3independentexperiments. Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page7of13

PAGE 8

filaminAwasconsistentwit hpreviousreportsinnormalprostatecells[23,24].Interestingly,asignificant proportionof U.parvum infectedBPH-1cellsexhibitedamarkedreductioninthepunctatenuclearstainingoffilaminAcoupledwithaconcurrentincreased offilaminAinthecytosol(P<0.0001).Specifically,68 10%of U.parvum infectedcellsascomparedwith 14.44%ofuninfectedcellsshowedthisabnormal phenotype. InordertodetermineiftheabnormalfilaminAphenotypecouldbetheresultofaparacrinemediatedhost responsetoinfection,weexposedBPH-1cellstoculture supernatantsobtainedfromshaminoculatedor U.parvum infectedcultures.Supernatantswereharvested fromBPH-1cellculturesat24hourspostinoculation with U.parvum .The24hourtimepointwaschosen becausecellculturesupernatantsobtainedatthispost inoculationtimepointcontainthehighestcytokine/chemokineconcentrations(u npublishedstudiesinour laboratory).Moreover,wewantedtominimizeany potentialconfoundingfactorsthatcouldresultfrom incubatingcellswithnutrientdepletedmedium,which couldoccurifalongerpostinoculationtimepointwas selected.AsshowninthetwobottompanelsinFigure 2a,supernatanttreatedculturesdidnotdisplayan abnormalfilaminAphenotypeaswasdetectedin U. parvum infectedcultures. Weevaluatedbothnuclearandcytosolicfractionsof celllysatesbyWesternblotwiththeantibodythat recognizestheC-terminalportionoffilaminA.As showninFigure2b,therewasamarkedreductionin thedetectionoffilaminAfragmentsinlysatesfrom U. parvum infectedcultures.Densitometryofintactfilamin Ainthecytosolicfractionofcelllysateswasalsoperformed(Figure2c).Forthisanalysis,ratiosweregeneratedbydividingtheaveragequantityofintactfilaminA bytheaveragequantityofGAPDHthatwasusedasa loadingcontrol.AsshowninFigure2c,therelative amountofintactfilaminAwasgreaterinthecytosolic fractionof U.parvum infectedcells.Thetotalamount offilaminAwasalsomeasuredbyELISA,andnosignificantdifferenceinthetotalamountoffilaminAamong infectedanduninfectedBPH-1cellswasdetected(data notshown).DetectionofphosphorylatedfilaminAinBPH-1cellsInprostatecells,adominantpathwayoffilaminAregulationinvolvescleavageoftheproteinbycalpain[23,24]. CalpainmediatedcleavageoffilaminAcanbereduced byphosphorylationoftheproteinatserine2152.Therefore,weassessedthedegreeoffilaminAphosphorylationatserine2152byimmunofluorescentmicroscopy (Figure3A)andELISA(Figure3B).Bothdetection methodsconfirmedthatonlycellsinfectedwith U. parvum displayedasignificantincreaseinphosphorylatedfilaminA.DetectionofcalpastatinandcalpaininBPH-1cellsRecentstudieshaveshownth atinfectionofhostcells with Mycoplasmahyorhinis causedinhibitionofcalpain activitythroughupregulationofitsinhibitor,calpastatin [25].Therefore,wealsoevaluatedtheeffectof U.parvum infectionontheintracellulardistributionandthe relativeconcentrationsofcalpainandcalpastatinin BPH-1cells.WedidnotdetectadifferenceintheintracellulardistributionofcalpainamonguninfectedBPH-1 cells, U.parvum infectedcells,andcellsincubatedwith supernatantsbyconfocalmicroscopy(datanotshown). Wealsodidnotobserveanyappreciabledifferencesin theamountofcalpainpresentwithinthecytosolicand nuclearfractionsofthesecellsbyWesternblot(data notshown).However,wedidobservedifferencesin boththeintracellulardistributionofcalpastatinandits relativeconcentrationamongthegroups.Specifically, U. parvum infectedcellsexhibitedlargeaggregatesofcalpastatinwithinthenucleus,andtheseaggregateswere moreprominentthanwhatwasobservedintheother groups(Figure4A).Moreover,Westernblotshowed thatcalpastatinwasreducedinthecytosolicfractionof U.parvum infectedcells(Figure4B),whichwasconfirmedbydensitometry(Figure4C.)DiscussionInourpreviousstudy,wefoundthatcolonizationofthe mucosalsurfaceofthebladderwasacommonfeature ininbredF344ratswitheitherasymptomaticorcomplicatedUTI[12].Basedonthesefindings,wepostulated that U.parvum maybecausingadisruptioninepithelial hostcellfunctioninamannerthatsupportsmicrobial colonization.Inordertoidentifyhostcellproteinsthat maybeperturbedby U.parvum colonization,weused differentialproteomeprofilingtocompareanimalswith activeUTItoanimalsthatspontaneouslyclearedinfectionorwithshaminoculatedcontrols.Weusedboth enrichmentanalysisandANOVAtoidentifycorrelationsbetweencolonizationwith U.parvum andperturbationsinfunctionalproteincategoriesorprotein networks.Bothmethodsrevealedthatanimalscolonized with U.parvum exhibitedsignificantperturbationsin proteinsthatnormallyregu lateactinpolymerization duringcellattachment,motilityandsignaltransduction events[26-31].Specifically,theconcentrationofprofilin 1, a actinin,vinculin,andfilaminAwerefoundtobe significantlyalteredbyinfection. WechosetofocusouradditionalexperimentsonfilaminAbecauseitsregulationhasbeenstudiedmore extensivelythanthatofprofilin1, a actinin,andvinculin[16,17,23,24,26].Thispr oteinisalsoinvolvedintheAllam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page8of13

PAGE 9

Figure3 IntracellulardistributionandquantificationofphosphorylatedfilaminAinuninfected, U.parvum infected,andsupernatant treatedBPH-1cells .Cellswereexposedtosterile10Bbroth,109CFUof U.parvum,orcellculturesupernatant(super)for72hoursbefore examinationbyconfocalmicroscopy(A),orELISA(B).PhosphorylatedfilaminA(red)wasdetectedwitharabbitmonoclonalantibody (EP2310AY).IntactfilaminA(green)wasdetectedwithmouseanti-filamin1.BPH-1nucleiand U.parvum (whitearrow)wereidentifiedwithDAPI stain(blue).Allimagesweretakenat600magnificationandthescalebarisequalto10 m.ELISAforphosphorylatedfilaminAwasperformed onwholecelllysates.AbsorbancevaluesweredividedbythetotalmgproteindeterminedbyBCAassay.ValuesrepresentthemeanSD(n= 5)ofphosphorylatedfilaminAinuninfected(BPH-1)andinfected(UP),uninfectedsupernatanttreated(BPHsuper),andinfectedsupernatant treated(UPsuper)cells.**PValueWasobtainedbyFishersPLSD. Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page9of13

PAGE 10

Figure4 Intracellulardistributionandquantificationofcalpastatininuninfected, U.parvum infected,andsupernatanttreatedBPH-1 cells .Cellswereexposedtosterile10Bbroth,109CFUof U.parvum ,orcellculturesupernatantfor72hoursbeforeexaminationbyconfocal microscopy(A),Westernblot(B)anddensitometry(C).Confocalimagesweretakenat600magnificationandthescalebarisequalto10 m. Calpastatinwasdetectedwithrabbitpolyclonalantibody(red).BPH-1nucleiand U.parvum (whitearrow)wereidentifiedwithDAPIstain(blue). Westernblotanalysisforthedetectionofcalpastatinwasperformedoncytosolic(cyt)andnuclear(nuc)fractionsfromuninfectedcells(BPH), infected(UP),uninfectedsupernatanttreated(BPHS)andinfectedsupernatanttreated(UPS)cells.Mequalsmolecularweightmarker.GAPDH wasusedasaloadingcontrolandaconfirmationthatthenuclearfractionwasnotcontaminatedwithcytosolicproteins.Quantitationof calpastatinincytosolicfractionswasperformedbydensitometry.Theaveragequantitywithineachblotwasnormalizedbydividingtheaverage quantityofcalpastatinproteinbandbytheaveragequantityoftheGAPDHbandwithineachblot.ValuesrepresentthemeanSDof2 biologicalreplicatesfrom2independentexperiments. Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page10of13

PAGE 11

regulationofvariouscellsignalingeventsincluding pathwaysthatareimportantinbothinnateandadaptive immunity[32-35].AccordingtotheProGroup analysis,filaminAwassignificantlyreducedinanimals activelycolonizedwith U.parvum .However,immunofluorescentdetectionoffilaminAinratbladdertissues didnotsupporttheproteomedata.Thus,thechanges thatweobservedininfectedtissuesmaybeareflection ofpost-translationalevents suchasproteolysisorchemicalmodificationsthataffectproteinsolubilityand recoveryduringproteinextraction[27,28].Therefore,it wasnecessarytodevelopan invitro modelofinfection thatcouldbeusedtoelucidatethesemechanisms.BPH1cellswereusedasourmodelsystembecausetheyare adifferentiate d,immortalizedepithelialcelllineofthe urogenitaltract[20,21]thatisamenabletolong-term colonizationwith Mollicutes [22].Moreover,prostate cellsdisplayadistinctintracellulardistributionoffilaminAthatwewereabletoexploitinourstudies [23,24].Specifically,filaminAcleavagemediatedbycalpainisadominantpathwayinprostatecells.OncefilaminAiscleaved,theC-terminalfragmentsofthe proteintranslocatetothenucleus[23,24].Weobserved C-terminalfragmentsoffilaminAwithinthenucleus ofuninfectedcellsbyconfocalmicroscopy.Wealso observedthesefragmentswithinthecytosolicfractionof BPH-1cellsbyWesternblot.However,wedidnot detectthesefragmentswithinthenuclearfractionof thesecellsbythismethod,whichismostlikelyareflectionofthedetectionlimitsoftheassay.Despitethis limitation,bothconfocalmicroscopyandWesternblot identifiedadisruptioninthenormalprocessoffilamin Acleavagein U.parvum infectedcells.Thiseffect appearstobeduetotheactualpresenceofthemicrobe sinceitwasnotobservedincellsexposedtocellculture supernatantsfrominfectedcultures. Twoprinciplemechanismscanaccountforthe decreasedproteolysisoffil aminA.Theproteinitself becomesresistanttocalpainmediatedcleavagewhenit isphosphorylatedatserine2152[36,37].Thesecond mechanisminvolvesdirecti nhibitionofcalpainbyits naturalinhibitorcalpastatin[38,39].Changesinintracellularconcentrationsofcalpainandcalpastatinhave beenshowntocorrelatewithchangesinfilaminAcleavage[40].TheresultsofourstudiessuggestthatphosphorylationoffilaminAisthemostlikelymechanism for U.parvum mediateddisruptioninthissystem. U. parvum infectedcellshadsignificantlygreateramounts offilaminAphosphorylatedatserine2152thandidthe othergroups(uninfectedcellsorcellsincubatedwith supernatants). U.parvum infectiondidnotappearto affecttheintracellulardistributionofcalpainoritsoverallconcentrationasdetectedbyWesternblot.Furthermore,thechangesweobservedincalpastatindynamics of U.parvum infectedcellscouldactuallyreducethe abilityofcalapstatintoinhibitcalpain[41,42]. PhosphorylationoffilaminAmaybeadownstream effectofsignaltransductionthatisinitiatedatthehost cellmembrane/microbeinterface.BothcAMP-dependentkinase(PKA)[37]andribosomalS6kinase(RSK) [43]havebeenreportedtoendogenouslyphosphorylate filaminAatserine2152.Therefore,bothkinasesare potentialupstreameffectorsof U.parvum mediated effectsonfilaminA.However,ourresultssuggestthe cAMPpathwaymaybethemorelikelytargetof U.parvum infection.Forexample,theperturbationofcalpastatindynamicsobservedininfectedBPH-1cellscan occurwithincreasedintracellularcAMPandactivation ofPKA[41].Thechangesinvinculin, a actinin, a enolase,andelongationfactor1 a thatweredetectedbydifferentialproteomeprofilingofbladdertissuescanalso betheresultofincreasedintracellularcAMP[44-47]. DownstreameffectorsofcAMPsuchasEPAC (exchangeproteindirectlyactivatedbycAMP,also knownascAMP-GEF)andPKAhavebeenshownto modulateinflammationandt issueproliferation[48,49]. Thus,elucidatingtheupstreamcomponentsoffilamin Aphosphorylationmayprovidenewmechanistic insightsintothemechanismsofureaplasmalasymptomaticinfectionanddisease. Theeffectsweobservedin U.parvum infectedBPH-1 cellscanprovideanexplanationforthereductionof filaminAthatwasdetectedintheiTRAQexperiments. ThephosphorylationoffilaminAthatwasnotedin U. parvum infectedcellscorrelatedwithitsredistribution intolesssolublecompartmentsofBPH-1cells(cell membraneandthecellcytoskeleton).Sincelesssoluble proteinscanbelostbyTrizolextractionmethods,they canbeunderrepresentedwithinthefractionandmeasuredasadecreasebyiTRAQ analysis.Unfortunately, noneoftheavailableantibodiesspecificforphosphorylatedfilaminAatserine2152workedinratbladdertissuessowecouldnotdetermineifthisalsooccurredin infectedanimals.Despitetheselimitations,bothmodels ofinfectiondisplayedaperturbationoffilaminA dynamics,whichmayserveasaviablemolecularmarker fordelineatingthehostcel lsignaltransductionpathwaysthatareaffectedby U.parvum infection.ConclusionsProteomeprofilingofratbladdertissuesidentifiedasignificantperturbationinhostcellfilaminAduringcolonizationwith U.parvum Invitro infectionstudieswith BPH-1cellsconfirmedthat U.parvum colonization interferedwiththenormaldis tributionofintracellular filaminAbyinducingphosphorylationoftheproteinat serine2152.PhosphorylationoffilaminAoccursin responsetovariouscellsignalingcascadesthatregulateAllam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page11of13

PAGE 12

cellmotility,differentiation,apoptosisandinflammation, whichmayberelevanttoureaplasmaldiseasepathogenesis.Thus,thisphenomenonmaybeausefulmolecular markerforidentifyingthespecifichostcellsignaling pathwaysperturbedduring U.parvum infection.AdditionalmaterialAdditionalfile1:TableS1-Ratbladderproteinsthatwere significantlyalteredbyactiveinfectionwith U.parvum as determinedwiththeProGroup algorithm .TableS1containsthe listofproteinratiosthatsignificantlydifferedamongratsinoculatedwith U.parvum (activeinfectionversusclearedinfection).Resultsare presentedasthemeanSDofthreebiologicalreplicatesobtainedfrom threeindependentiTRAQ experiments.Ratiosweregeneratedby dividingthespectralintensityvalueinUTIandStruvitegroupsbythe spectralintensityforaspecificpeptideintheNegativegroup.Only proteinratiosthatshowedasimilarresponseinbothUTIandStruvite groupsandweresignificantlydifferentfromtheNegativegroup(P< 0.05)arelisted.Proteinsweregroupedaccordingtotheirassigned biologicalfunction. TableS2-Ratbladdertissueproteomeprofiles thatweresignificantlydifferentamongratsinoculatedwith U. parvum asdeterminedbyANOVA .TableS2containsthelistofprotein ratiosobtainedwiththeProGroup algorithm(meanSD)thatwere significantlydifferentamongratsinoculatedwith U.parvum as determinedbyANOVA.Datawasobtainedfromthreebiological replicatesfromthreeindependentiTRAQ experiments.Proteinratios weregeneratedbydividingthespectralintensityoftheproteinina U. parvum inoculatedgroup(Negative,UTI,orStruvite)withthespectral intensityintheshaminoculatedcontrolusingtheProGroup algorithm.Proteinsthatshowedasignificantlydifferentprofile(P<0.03) arelisted.Proteinsaregroupedaccordingtotheclusterpatternshown inFigure1CthatwasobtainedbythemethodofWard.Thebiological functionofeachproteinwasobtainedfromthePantherdatabasehttp:// www.pantherdb.orgortheRatGenomeDatabasehttp://rgd.mcw.edu/ wg. Additionalfile2:ImmunohistochemicaldetectionofintactfilaminA inthebladdertissueofF344ratsinoculatedwithsterile10Bbroth or109CFUof U.parvum .Representativebladdertissuesectionsfrom isotypeprimaryantibodycontrol(A),shaminoculatedcontrol(B), asymptomaticUTI(C),andstruvite(D)groupsdemonstratingthe distributionofintactfilaminA(green).NucleiwerestainedwithDAPI (blue).Imagesare600magnification,L=bladderlumen,SM= submucosa,andarrowsarepointingtouroepithelium. Additionalfile3:ColocalizationoffilaminAwithpolymerizedactin inuninfectedand U.parvum infectedBPH-1cells .Representative imagesofcellsexamined72hoursafterinoculationwithsterile10B brothor109CFUof U.parvum .NucleiwerestainedwithDAPI(blue), whitearrowsarepointingto U.parvum coloniesthatweredetectedwith DAPIstaining.FilaminA(FilA)wasstainedwithrabbit-antiCterminal filaminA(red).PolymerizedactinwasstainedwithphalloidinAlexa-448 (green).Allimagesweretakenat400magnificationandthescalebaris equalto10 m. TheprojectdescribedwassupportedbyAwardNumberK08DK075651from theNationalInstituteofDiabetesAndDigestiveAndKidneyDiseases (NIDDK).Thecontentissolelytheresponsibilityoftheauthorsanddoesnot necessarilyrepresenttheofficialviewsoftheNIDDKortheNational InstitutesofHealth. Experimentalportionsofthisprojectwerealso supportedbyanintramuralgrantobtainedthroughtheUniversityofFlorida, CollegeofVeterinaryMedicine.Publicationofthisarticlewasfundedinpart bytheUniversityofFloridaOpen-AccessPublishingFund. Authordetails1DepartmentofInfectiousDisease&Pathology,CollegeofVeterinary Medicine,UniversityofFlorida,Gainesville,FL,USA.2InterdisciplinaryCenter forBiotechnologyResearch,UniversityofFlorida,Gainesville,FL32611,USA.3DonaldDanforth,PlantScienceCenter,975N.WarsonRdSt.Louis,MO 63132,USA. Authors contributions AAexecutedcellcultureexperimentsandcontributedtomanuscript preparation.SAdesignedandexecutedproteomestudies,andassistedin manuscriptpreparation.MBBparticipatedinthedesignandcoordinationof thestudyandassistedinmanuscriptpreparation.LRdesignedexperiments, executedanimalinfectionstudies,assistedincellcultureexperiments,data analysisandmanuscriptpreparation.Allauthorsconcurwiththefinal versionofthemanuscript. Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Received:29July2010Accepted:20April2011Published:20April2011 References1.VolgmannT,OhlingerR,PanzigB: Ureaplasmaurealyticum -harmless commensalorunderestimatedenemyofhumanreproduction?A review. Archivesofgynecologyandobstetrics 2005, 273(3) :133-139. 2.WaitesKB,SchelonkaRL,XiaoL,GrigsbyPL,NovyMJ: Congenitaland opportunisticinfections: Ureaplasmaspecies and Mycoplasmahominis Seminarsinfetal&neonatalmedicine 2009, 14(4) :190-199. 3.YiJ,YoonBH,KimEC: Detectionandbiovardiscriminationof Ureaplasma urealyticum byreal-timePCR. Molecularandcellularprobes 2005, 19(4) :255-260. 4.SchelonkaRL,WaitesKB: Ureaplasmainfectionandneonatallungdisease. SeminPerinatol 2007, 31(1) :2-9. 5.PinnaGS,SkevakiCL,Kafetzis: Thesignificanceof Ureaplasmaurealyticum asapathogenicagentinthepaediatricpopulation. CurrOpinInfectDis 2006, 19(3) :283-289. 6.McDonaldMI,LamMH,BirchDF,D ArcyAF,FairleyKF,PavillardER: Ureaplasmaurealyticum inpatientswithacutesymptomsofurinarytract infection. TheJournalofurology 1982, 128(3) :517-519. 7.GiannakopoulosS,AlivizatosG,DeliveliotisC,SkolarikosA,KastriotisJ, SofrasF: Encrustedcystitisandpyelitis. EurUrol 2001, 39(4) :446-448. 8.Taylor-RobinsonD,CsonkaGW,PrenticeMJ: Humanintra-urethral inoculationofureplasmas. QJMed 1977, 46(183) :309-326. 9.SkerkV,MarekovicI,MarkovinovicL,BegovacJ,SkerkV,BarsicN,MajdakGluhinicV: Comparativerandomizedpilotstudyofazithromycinand doxycyclineefficacyandtolerabilityinthetreatmentofprostate infectioncausedby Ureaplasmaurealyticum Chemotherapy 2006, 52(1) :9-11. 10.HedelinH: Uropathogensandurinarytractconcretionformationand catheterencrustations. Internationaljournalofantimicrobialagents 2002, 19(6) :484-487. 11.ReyesL,AlvarezS,AllamA,ReinhardM,BrownMB: Complicatedurinary tractinfectionisassociatedwithuroepithelialexpressionof proinflammatoryproteinS100A8. Infectionandimmunity 2009, 77(10) :4265-4274. 12.ReyesL,ReinhardM,BrownMB: Differentinflammatoryresponsesare associatedwith Ureaplasmaparvum -inducedUTIandurolithformation. BMCinfectiousdiseases 2009, 9 :9. 13.ReyesL,ReinhardM,O Donell LJ,StevensJ,BrownMB: Rat strainsdiffer insusceptibilityto Ureaplasmaparvum -inducedurinarytractinfection andstruvitestoneformation. Infectionandimmunity 2006, 74(12) :6656-6664. 14.MulveyMA,SchillingJD,MartinezJJ,HultgrenSJ: Badbugsand beleagueredbladders:interplaybetweenuropathogenic Escherichiacoli andinnatehostdefenses. ProcNatlAcadSciUSA 2000, 97(16) :8829-8835. 15.SivickKE,MobleyHL: Wagingwaragainsturopathogenic Escherichiacoli : winningbacktheurinarytract. Infectionandimmunity 2009, 78(2) :568-585. 16.PopowiczGM,SchleicherM,NoegelAA,HolakTA: Filamins:promiscuous organizersofthecytoskeleton. Trendsinbiochemicalsciences 2006, 31(7) :411-419. 17.StosselTP,CondeelisJ,CooleyL,HartwigJH,NoegelA,SchleicherM, ShapiroSS: Filaminsasintegratorsofcellmechanicsandsignalling. Nat RevMolCellBiol 2001, 2(2) :138-145.Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page12of13

PAGE 13

18.KerseyPJ,DuarteJ,WilliamsA,KaravidopoulouY,BirneyE,ApweilerR: The InternationalProteinIndex:anintegrateddatabaseforproteomics experiments. Proteomics 2004, 4(7) :1985-1988. 19.ShilovIV,SeymourSL,PatelAA,LobodaA,TangWH,KeatingSP,HunterCL, NuwaysirLM,SchaefferDA: TheParagonAlgorithm,anextgeneration searchenginethatusessequencetemperaturevaluesandfeature probabilitiestoidentifypeptidesfromtandemmassspectra. MolCell Proteomics 2007, 6(9) :1638-1655. 20.HaywardSW,DahiyaR,CunhaGR,BartekJ,DeshpandeN,NarayanP: Establishmentandcharacterizationofanimmortalizedbutnontransformedhumanprostateepithelialcellline:BPH-1. Invitrocellular& developmentalbiology 1995, 31(1) :14-24. 21.MitchellS,AbelP,WareM,StampG,LalaniE: Phenotypicandgenotypic characterizationofcommonlyusedhumanprostaticcelllines. BJU international 2000, 85(7) :932-944. 22.NamikiK,GoodisonS,PorvasnikS,AllanRW,IczkowskiKA,UrbanekC, ReyesL,SakamotoN,RosserCJ,TysonDR, etal : Persistentexposureto Mycoplasmainducesmalignanttransformationofhumanprostatecells. PLoSOne 2009, 4(9) :e6872. 23.BedollaRG,WangY,AsuncionA,ChamieK,SiddiquiS,MudryjMM, PrihodaTJ,SiddiquiJ,ChinnaiyanAM,MehraR, etal : Nuclearversus cytoplasmiclocalizationoffilaminAinprostatecancer: immunohistochemicalcorrelationwithmetastases. ClinCancerRes 2009, 15(3) :788-796. 24.LoyCJ,SimKS,YongEL,YoshidaN,OgataT,TanabeK,LiS,NakazatoM, KohuK,TakafutaT, etal : Filamin-Afragmentlocalizestothenucleusto regulateandrogenreceptorandcoactivatorfunctions. ProcNatlAcadSci USA 2003, 100(8) :4562-4567. 25.ElkindE,RechnitzerH,VaisidT,KornspanJD,BarnoyS,RottemS, KosowerNS: Mycoplasmahyorhinis upregulatescalpastatinandinhibits calpain-dependentproteolysisinSH-SY5Yneuroblastomacells. FEMS microbiologyletters 2010, 304(1) :62-68. 26.GarciaE,StracherA,JayD: CalcineurindephosphorylatestheC-terminal regionoffilamininanimportantregulatorysite:apossiblemechanism forfilaminmobilizationandcellsignaling. Archivesofbiochemistryand biophysics 2006, 446(2) :140-150. 27.BlaseC,BeckerD,KappelS,Bereiter-HahnJ: Microfilamentdynamics duringHaCaTcellvolumeregulation. Europeanjournalofcellbiology 2009, 88(3) :131-139. 28.SchnittlerHJ,SchneiderSW,RaiferH,LuoF,DieterichP,JustI,AktoriesK: Roleofactinfilamentsinendothelialcell-celladhesionandmembrane stabilityunderfluidshearstress. PflugersArch 2001, 442(5) :675-687. 29.JockuschBM,MurkK,RothkegelM,YarmolaEG,BubbMR: Theprofileof profilinsHowdepolymerizationcanpromotepolymerization:thecaseof actinandprofilin. RevPhysiolBiochemPharmacol 2007, 159(11) :131-149. 30.MierkeCT: Theroleofvinculinintheregulationofthemechanical propertiesofcells. CellBiochemBiophys 2009, 53(3) :115-126. 31.CampbellID: Studiesoffocaladhesionassembly. BiochemSocTrans 2008, 36(Pt2):263-266. 32.RuddCE,TaylorA,SchneiderH: CD28andCTLA-4coreceptorexpression andsignaltransduction. ImmunolRev 2009, 229(1) :12-26. 33.LeonardiA,Ellinger-ZiegelbauerH,FranzosoG,BrownK,SiebenlistU: Physicalandfunctionalinteractionoffilamin(actin-bindingprotein-280) andtumornecrosisfactorreceptor-associatedfactor2. JBiolChem 2000, 275(1) :271-278. 34.MartiA,LuoZ,CunninghamC,OhtaY,HartwigJ,StosselTP,KyriakisJM, AvruchJ: Actin-bindingprotein-280bindsthestress-activatedprotein kinase(SAPK)activatorSEK-1andisrequiredfortumornecrosisfactoralphaactivationofSAPKinmelanomacells. JBiolChem 1997, 272(5) :2620-2628. 35.WangQ,PattonWF,ChiangET,HechtmanHB,SheproD: Filamin translocationisanearlyendothelialcellinflammatoryresponseto bradykinin:regulationbycalcium,proteinkinases,andprotein phosphatases. JCellBiochem 1996, 62(3) :383-396. 36.vanderFlierA,SonnenbergA: Structuralandfunctionalaspectsof filamins. BiochimBiophysActa 2001, 1538(2-3) :99-117. 37.JayD,GarciaEJ,delaLuzIbarraM: InsitudeterminationofaPKA phosphorylationsiteintheC-terminalregionoffilamin. MolCellBiochem 2004, 260 :49-53. 38.MelloniE,MichettiM,SalaminoF,MinafraR,SparatoreB,PontremoliS: Isolationandcharacterizationofcalpainactivatorproteinfrombovine brain. Methodsinmolecularbiology 2000, 144 :99-105. 39.StifaneseR,AvernaM,SalaminoF,CantoniC,MingariMC,PratoC, PontremoliS,MelloniE: Characterizationofthecalpain/calpastatin systeminhumanhemopoieticcelllines. Archivesofbiochemistryand biophysics 2006, 456 :48-57. 40.PilopC,AreggerF,GormanRC,BrunisholzR,GerritsB,SchaffnerT, GormanJH,MatyasG,CarrelT,FreyBM: Proteomicanalysisinaortic mediaofpatientswithMarfansyndromerevealsincreasedactivityof calpain2inaorticaneurysms. Circulation 2009, 120(11) :983-991. 41.AvernaM,deTullioR,PassalacquaM,SalaminoF,PontremoliS,MelloniE: Changesinintracellularcalpastatinlocalizationaremediatedby reversiblephosphorylation. TheBiochemicaljournal 2001, 354(Pt1):25-30. 42.AvernaM,StifaneseR,DeTullioR,DefranchiE,SalaminoF,MelloniE, PontremoliS: Interactionbetweencatalyticallyinactivecalpainand calpastatin.Evidenceforitsoccurrenceinstimulatedcells. TheFEBS journal 2006, 273(8) :1660-1668. 43.WooMS,OhtaY,RabinovitzI,StosselTP,BlenisJ: RibosomalS6kinase (RSK)regulatesphosphorylationoffilaminAonanimportantregulatory site. Molecularandcellularbiology 2004, 24 :3025-3035. 44.MarjanovicS,WollbergP,SkogS,HeidenT,NelsonBD: Theeffectsof cAMPontheexpressionofglycolyticisozymesinactivatedperipheral humanTlymphocytes. Archivesofbiochemistryandbiophysics 1993, 302(2) :398-401. 45.PorterSE,Dwyer-NieldLD,MalkinsonAM: Regulationoflungepithelial cellmorphologybycAMP-dependentproteinkinasetypeIisozyme. Am JPhysiolLungCellMolPhysiol 2001, 280(6) :L1282-1289. 46.DellaFaziaMA,CastelliM,BartoliD,PieroniS,PettirossiV,PiobbicoD,ViolaMagniM,ServilloG: HOPS:anovelcAMP-dependentshuttlingprotein involvedinproteinsynthesisregulation. JCellSci 2005, 118(Pt 14) :3185-3194. 47.SwaneyJS,PatelHH,YokoyamaU,HeadBP,RothDM,InselPA: Focal adhesionsin(myo)fibroblastsscaffoldadenylylcyclasewith phosphorylatedcaveolin. JBiolChem 2006, 281(25) :17173-17179. 48.GloerichM,PonsioenB,VliemMJ,ZhangZ,ZhaoJ,KooistraMR,PriceLS, RitsmaL,ZwartkruisFJ,RehmannH, etal : SpatialregulationofcyclicAMPEpac1signalingincelladhesionbyERMproteins. Molecularandcellular biology 30(22) :5421-5431. 49.WallEA,ZavzavadjianJR,ChangMS,RandhawaB,ZhuX,HsuehRC,LiuJ, DriverA,BaoXR,SternweisPC, etal : SuppressionofLPS-inducedTNFalphaproductioninmacrophagesbycAMPismediatedbyPKAAKAP95-p105. SciSignal 2009, 2(75) :ra28.Pre-publicationhistory Thepre-publicationhistoryforthispapercanbeaccessedhere: http://www.biomedcentral.com/1471-2334/11/101/prepubdoi:10.1186/1471-2334-11-101 Citethisarticleas: Allam etal .: Ureaplasmaparvum infectionalters filaminadynamicsinhostcells. BMCInfectiousDiseases 2011 11 :101. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Allam etal BMCInfectiousDiseases 2011, 11 :101 http://www.biomedcentral.com/1471-2334/11/101 Page13of13


University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Powered by SobekCM