Chlamydia trachomatis homotypic inclusion fusion is promoted by host microtubule trafficking

MISSING IMAGE

Material Information

Title:
Chlamydia trachomatis homotypic inclusion fusion is promoted by host microtubule trafficking
Series Title:
Richards TS, AE Knowlton, and SS Grieshaber. “Chlamydia trachomatis homotypic inclusion fusion is promoted by host microtubule trafficking” BMC Microbiology 2013; 13: 185 http://www.biomedcentral.com/1471-2180/13/185/abstract doi:10.1186/1471-2180-13-185
Physical Description:
Journal Article
Creator:
Richards, Theresa S
Knowlton, Andrea E
Grieshaber, Scott S
Publisher:
BioMed Central (BMC Microbiology)
Publication Date:

Notes

Abstract:
Background: The developmental cycle of the obligate intracellular pathogen Chlamydia is dependant on the formation of a unique intracellular niche termed the chlamydial inclusion. The inclusion is a membrane bound vacuole derived from host cytoplasmic membrane and is modified significantly by the insertion of chlamydial proteins. A unique property of the inclusion is its propensity for homotypic fusion. The vast majority of cells infected with multiple chlamydial elementary bodies (EBs) contain only a single mature inclusion. The chlamydial protein IncA is required for fusion, however the host process involved are uncharacterized. Results: Here, through live imaging studies, we determined that the nascent inclusions clustered tightly at the cell microtubule organizing center (MTOC) where they eventually fused to form a single inclusion. We established that factors involved in trafficking were required for efficient fusion as both disruption of the microtubule network and inhibition of microtubule trafficking reduced the efficiency of fusion. Additionally, fusion occurred at multiple sites in the cell and was delayed when the microtubule minus ends were either no longer anchored at a single MTOC or when a cell possessed multiple MTOCs. Conclusions: The data presented demonstrates that efficient homotypic fusion requires the inclusions to be in close proximity and that this proximity is dependent on chlamydial microtubule trafficking to the minus ends of microtubules. Keywords: Chlamydia, Dynein, Microtubules, Vesicle fusion, Centrosomes
Acquisition:
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Haeden Roberson.
Publication Status:
Published

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the submitter.
System ID:
IR00004274:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Chlamydiatrachomatishomotypicinclusionfusion ispromotedbyhostmicrotubuletraffickingRichards etal. Richards etal.BMCMicrobiology 2013, 13 :185 http://www.biomedcentral.com/1471-2180/13/185

PAGE 2

RESEARCHARTICLEOpenAccessChlamydiatrachomatishomotypicinclusionfusion ispromotedbyhostmicrotubuletraffickingTheresaSRichards,AndreaEKnowltonandScottSGrieshaber*AbstractBackground: Thedevelopmentalcycleoftheobligateintracellularpathogen Chlamydia isdependantonthe formationofauniqueintracellularnichetermedthechlamydialinclusion.Theinclusionisamembranebound vacuolederivedfromhostcytoplasmicmembraneandismodifiedsignificantlybytheinsertionofchlamydial proteins.Auniquepropertyoftheinclusionisitspropensityforhomotypicfusion.Thevastmajorityofcells infectedwithmultiplechlamydialelementarybodies(EBs)containonlyasinglematureinclusion.Thechlamydial proteinIncAisrequiredforfusion,howeverthehostprocessinvolvedareuncharacterized. Results: Here,throughliveimagingstudies,wedeterminedthatthenascentinclusionsclusteredtightlyatthecell microtubuleorganizingcenter(MTOC)wheretheyeventuallyfusedtoformasingleinclusion.Weestablishedthat factorsinvolvedintraffickingwererequiredforefficientfusionasbothdisruptionofthemicrotubulenetworkand inhibitionofmicrotubuletraffickingreducedtheefficiencyoffusion.Additionally,fusionoccurredatmultiplesites inthecellandwasdelayedwhenthemicrotubuleminusendswereeithernolongeranchoredatasingleMTOC orwhenacellpossessedmultipleMTOCs. Conclusions: Thedatapresenteddemonstratesthatefficienthomotypicfusionrequirestheinclusionstobein closeproximityandthatthisproximityisdependentonchlamydialmicrotubuletraffickingtotheminusendsof microtubules. Keywords: Chlamydia,Dynein,Microtubules,Vesiclefusion,CentrosomesBackgroundChlamydiatrachomatis causessexuallytransmittedinfectionsandistheleadingcauseofpreventableblindness worldwide[1]. Chlamydia areGram-negative,obligate intracellularbacteriawithaunique,biphasicdevelopmentalcyclethattakesplaceinamembrane-boundvacuole termedtheinclusion.Theinfectiousbutmetabolicallyinactiveelementarybody(EB)attachestoepithelialcellsand initiatesitsuptakethroughparasitemediatedendocytosis [2].Onceinternalized,EBsdifferentiateintometabolically activebutnon-infectiousreticulatebodies(RBs)which replicatebybinaryfission.Astheinfectionprogresses, RBsdifferentiateintoEBsinanasynchronousmannerand theseinfectiousEBsareeventuallyreleasedintothehost toinitiateaadditionalroundsofinfection. Followinginfection,theinclusionmembraneismodified throughtheinsertionofmultiplebacterialtypethree secretedeffectorproteins[3].Theseinclusionsarenonfusogenicwiththeendosomalandlysosomalpathways[4]. Inclusionsaretraffickedalongmicrotubulesinadyneindependentmannertothemicrotubuleorganizingcenter (MTOC)wheretheyintercepthost-derivedlipidstomaintaintheintegrityoftheexpandinginclusion[5].Thus, despitebeingsequesteredwithinamembrane-bound vacuole,chlamydiaemanipulatethehostandsubverthost pathwaystoestablishanenvironmentthatisnotonlyconducivetoreplicationanddifferentiationbutalsosimultaneouslyprotectedfromhostimmuneresponses. Athighmultiplicitiesofinfection,multipleinclusions fuseintoasingleinclusion.Thisfusioneventiscriticalfor pathogenicity;rareisolateswithnon-fusogenicinclusions areclinicallyassociatedwithlessseveresignsofinfection andlowernumbersofrecoverablebacteriathanwild-type isolates[6].Inclusionfusionoccursevenbetweendifferent C.trachomatis serovarspotentiallyfacilitatinggeneticexchangebetweenserovars[7].Previousstudieshavedemonstratedthatthefusionofchlamydialinclusionsrequires *Correspondence: sgriesha@ufl.edu DepartmentofOralBiology,CollegeofDentistry,UniversityofFlorida,Box 100424,Gainesville,FL32610,USA 2013Richardsetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.Richards etal.BMCMicrobiology 2013, 13 :185 http://www.biomedcentral.com/1471-2180/13/185

PAGE 3

bacterialproteinsynthesisandisinhibitedduringgrowth at32C[8].Specifically,theinclusionmembraneprotein IncAisrequiredforthehomotypicfusionofchlamydial inclusions[9]. Theimportanceofbothinclusiontraffickingandinclusionfusionhavebeenestablishedbuttherolethatinclusiontraffickingplaysinpromotingfusionhasnotbeen investigated.Inthisstudywedemonstratethatinclusion migrationalongmicrotubulespromotesinclusionfusion. Interestingly,althoughthisdyneindependentmigration wasrequiredforthenormaltimingofinclusionfusion,inhibitionofthistraffickingwaseventuallyovercomelater duringinfection.MethodsOrganismsandcellcultureAllcellswereobtainedfromtheAmericanTypeCultureCollection.Celllinesare:McCoy(McCoyB,CRL-1696),HeLa (HeLa229,CCL-2.1),Cos7(COS-7,CRL-1651)andneuroblastoma(N1E-115,CRL-2263). Chlamydiatrachomatis serovarsare:L2(LGV434),G (UW-524/CX)andJ(UW36/CX). C.trachomatis werepropagatedinMcCoyorHeLa cells.EBswerepurifiedbyReno grafin(Bristol-MyersSquibb, NewYork,NY,USA)densitygradientcentrifugationaspreviouslydescribed[10,11].HeLaandCos7cellsweregrown inRPMI-1640(Lonza,Basel,S witzerland)supplemented with10%FBS(Gibco/LifeTechnologies,GrandIsland,NY, USA)and10 g/mLgentamicin(Gibco).McCoyand neuroblastomacellsweregrowninDMEM(Lonza) supplementedwith10%FBS(Gibco)and10 g/mLgentamicin(Gibco).Allcellsweregrownin5%CO2at37C.InfectionsAllinfectionswerecarriedoutasfollowsunlessotherwise noted.Cellswereincubatedwith C.trachomatis EBsin Hank ’ sbalancedsaltsolution(HBSS)(Invitrogen/Life Technologies,GrandIsland,NY,USA)for30minat22C. Theinoculumwasreplacedwithprewarmed,37C,completemedia.Fornocodazoletreatedcells,theinoculum wasreplacedwithprewarmed,37C,completemedia containing5 g/mLnocodazole.Infectedcellswereincubatedin5%CO2at37C.SynchronizedinfectionsCellswereincubatedwith C.trachomatis EBsinHBSS (Invitrogen)atMOI=1000for5minat22C.Thecells werewashedthreetimeswithHBSSplus100 g/mLheparin(Pharmacia,Peapack,NJ,USA)andtwicewithHBSS withoutheparin.Prewarmed,37C,completemediawas addedandinfectedcellswereincubatedin5%CO2at37C.TransfectionsandplasmidsHeLacellsweregrownon12mmnumber1.5borosilicateglasscoverslipscoatedwithPoly-L-lysine(SigmaAldrich,St.Louis,MO,USA)toobtainamonolayerof approximately65%confluency.TransfectionswerecarriedoutusingLipofectamine2000(Invitrogen)according tothemanufacturer ’ sinstructions.Expressionfromthe transfectedvectorswasallowedtoproceedforatleast 24hpriortoexperimentation.Expressionvectorsused werepEGFP-C3(Clontech,MountainView,CA,USA), EB1-GFPandEB1.84-GFP.TheEB1-GFPplasmidwasa kindgiftfromDrJenniferS.Tirnauer,Universityof ConnecticutHealthCenter.TheEB1.84-GFPplasmidwas generatedbyPCRcloningoftheNterminalendofEB1 andcloningintopDest-NGFPasdescribedbyAskham etal.[12].Micro-injectionsCos7cellsweregrownon25mmnumber1.5borosilicate glasscoverslipscoatedwithPoly-L-lysine(Sigma-Aldrich) toobtainamonolayerofapproximately50%confluency. Micro-injectionwasperformedusinganautomatedsystemdescribedpreviously[5].Cellswereinjectedwith eithermousemonoclonalantibodytodic74.1(Covance, Princeton,NJ,USA)orantiCD80(Invitrogen).Following injection,cellswerewashedoncewithprewarmed,37C, completemedia,andfreshprewarmedmediawasadded. Approximately10 – 15minafterinjection,thecellswere infectedwith C.trachomatis andincubatedin5%CO2at 37C.Thecellswerefixedwith4%paraformaldehydeand permeabilizedwith0.5%TritonX100.TheinjectedantibodiesweredetectedusingAlexaFluor488-conjugated goatanti-mouseIgG(MolecularProbes/LifeTechnologies, GrandIsland,NY,USA).AntibodiesandmicroscopyForfluorescentantibodystaining,infectedcellswere fixedwithcoldmethanolfor10min.Antibodiesused intheseexperimentsweremousemonoclonalanti-tubulin(Sigma-Aldrich),anti-chlamydialinclusion membraneproteinIncAagiftfromDr.DanRockey, attheOregonStateUniversity,andanti-chlamydial MOMPagiftfromDr.HarlanCaldwell,RockyMountainLabsNIAID. C.trachomatis wasstainedwithhumanserum(Sigma-Aldrich)unlessotherwisenoted.To visualizetheprimaryantibodies,cellswereincubated withtheappropriateAlexaFluorconjugatedsecondary antibody:488,567or647againstmouse,rabbitor humanIgG(MolecularProbes).TovisualizeDNA,cells werestainedwiththefar-redfluorescentdyeDRAQ5 (BiostatusLimited,Leicestershire,UK).Imageswere acquiredusingaspinningdiskconfocalsystemconnected toaLeicaDMIRBmicroscopewitha63oil-immersion objective,equippedwithaPhotometricscascade-cooled EMCCDcamera,underthecontroloftheOpenSource softwarepackage Manager(http://www.micro-manager. org/).ImageswereprocessedusingtheimageanalysisRichards etal.BMCMicrobiology 2013, 13 :185 Page2of7 http://www.biomedcentral.com/1471-2180/13/185

PAGE 4

softwareImageJ(http://rsb.info.nih.gov/ij/).Projections wereconstructedusingtheImageJimagesoftware (WayneRasband,U.S.NationalInstitutesofHealth, http://rsb.info.nih.gov/ij).ResultsInclusionfusionoccursattheMTOCThelocationanddynamicsofinclusionfusionarecurrentlypoorlyunderstood.T odeterminethesubcellular locationoffusioninmultiplyinfectedcells,HeLacells weretransfectedwithEB1-GFP.EB1isamicrotubule endplusendtrackingproteinandservestoidentify thesiteofthemicrotubuleorganizingcenter(MTOC). Eighteenhourspost-transfection,cellswereinfected with C.trachomatis atMOI~20.Infectedcellswere imagedevery10minutesforatotalof24hours.Representativetimepoints(Figure1)revealedthatearlyduringinfection,multipleinclusionswerepresentadjacent tocellcentrosomes(Figure1,8:50 – 11:30hpi).Asthe infectionproceeded,fusionoccurredbetweenclosely groupedinclusions(Figure1,11:30 – 12:30hpi).Fusion continueduntiltherewasasingleinclusion(Figure1, 12:30hpi)whichcontinuedtoexpandasthedevelopmentalcycleprogressed(Figure1,compare12:30and 15:50hpi).Intheseexperiments,fusionwasonlyobservedbetweeninclusionstightlyclusteredaroundthe MTOC/centrosomeofthehostcell.(AlsoseeAdditional file1:Movie1).Intactmicrotubulesarerequiredforefficientinclusion fusionWedemonstratedthatfusionoccursatthecentrosomes andwehavepreviouslyreportedthattraffickingonmicrotubulesisrequiredforthelocalizationofchlamydial inclusionsatthecentrosomes.Weaskedwhetherthe microtubulenetworkinfluencedinclusionfusion.HeLa cellswereinfectedwith C.trachomatis .Followinginfection,cellswereincubatedinthepresenceorabsenceof nocodazoleandthenfixedeverytwohoursbetween10 and24hpi.Inclusionfusion occurredatapproximately 14hpiforuntreatedcells(Figure2A).Incellsthathad beentreatedwithnocodazole,fusionwassignificantly delayed.Nocodazole-treatedcellshadanaverageofeight inclusionspercellat24hpi(Figure2A).Fusionwasnot completelyabolishedbynocodazoletreatmentsuggesting thatthefusionmachinerydoesnotrequiremicrotubules butinsteadthatthemicrotubulesacceleratefusion.Representativepicturesofnocodazoletreatedanduntreated cellsareshowninFigure2BandC,respectively.InhibitingdyneinfunctioninHeLacellsinhibitsinclusion fusionChlamydialmicrotubuletraffickingisdependentonthe hostmicrotubulemotorproteindynein.Toinvestigate Figure1 Inclusionfusionoccursatthecentrosomes. HeLacells weretransfectedwithEB1-GFPtovisualizecentrosomes(arrowinA). Eighteenhourspost-transfection,cellswereinfectedwithC. trachomatisatMOI=20.Duringinfection,cellswerephotographed every10minutesuntil24hpi.Timespostinfectionareindicatedin eachcorrespondingimage. Figure2 InclusionfusionisdelayedinHeLacellstreatedwith nocodazole. HeLacellswereinfectedwith C.trachomatis atMOI~ 9inthepresenceandabsenceofnocodazole(Noc)andfixedat 10,12,14,16,20,22and24hpi.Cellswerestainedwithhuman seraandanti-g-tubulinantibodiesandinclusionswereenumerated (A) .RepresentativetreatedanduntreatedHeLacells (B and C, respectively ) Richards etal.BMCMicrobiology 2013, 13 :185 Page3of7 http://www.biomedcentral.com/1471-2180/13/185

PAGE 5

theroleofdyneinininclusionfusion,weinjectedCos7 cellswithanti-dyneinintermediatechainantibodies (DIC74.1).Followinginjection,cellswereinfectedwith C.trachomatis .Uninjectedcellswereinfectedinparallel. Cellswerefixedat6and24hpi.Incellsthathadbeen injectedwithanti-dyneinantibodies,inclusionclustering wasdecreasedearlyininfectionandinclusionfusion decreased(Figure3AandB,respectively).At24hpi, therewasasignificantdifferencebetweeninjectedand uninjectedcells(P<0.001);injectedcellsaveragedthree inclusionsperinfectedcellwhileuninjectedcellsaveragedoneinclusionperinfectedcell(Figure3C).FusionisdelayedinneuroblastomacellsWeestablishedthatinclusionfusionoccursatcellcentrosomesandbothdyneinandmicrotubulespromotefusion. Wenextaskedwhetherinfectionofcellswithmultiple centrosomeswouldleadtomultiplesitesoffusion.The mouseneuroblastomacelllineN115hassignificantcentrosomenumberdefectscontaininganaverageofeight centrosomespercell[13,14].Thisallowedustoaskwhetherdefectsincentrosomenumberswouldaffectinclusionfusion.HeLaandneuroblastomacellswereinfected with C.trachomatis atthreedifferentmultiplicitiesof infection.Infectionswerefixedat3hpiandeverytwo hoursbetween10and24hpi.Earlyinclusionswerepresent nearthetightlyclusteredcen trosomesinHeLacellsbutin neuroblastomacells,whichhavemultiplecentrosomesdistributedthroughoutthecell, earlyinclusionswerepresent throughoutthehostcytosol clusteredatthescattered centrosomes(Figure4A3hpiand4B3hpi,respectively). At24hpi,infectedHeLacellshadasingleinclusionadjacenttothecentrosomes(Figure424hpi).Whilesome infectedneuroblastomacellshadsingleinclusionsat 24hpi,infectedneuroblastomacellscouldstillbefound withmultipleunfusedinclusions(Figure4B24hpi).In infectedHeLacells,fusionofchlamydialinclusions occurredatapproximately12-14hpi(Figure4C).Fusion wasdelayedinneuroblastomacells,occurringatapproximately16-18hpi(Figure4D).Neuroblastomacellsarefusioncompetentandinclusion membraneproteinIncAispresentontheirinclusion membranesInordertodeterminewhetherneuroblastomaswerefusioncompetent,HeLaandneuroblastomacellswereseriallyinfectedwithdifferent C.trachomatis serovars.Cells wereinfectedwith C.trachomatis serovarGfor40hours andthensuperinfectedwith C.trachomatis serovarL2for fourhours.InbothHeLacellsandneuroblastomas,fusion occurredbetweeninclusionscontainingGandL2indicatingthattheinclusionsinneuroblastomacellsarefusion competent(Figure5Aand5B).Theinclusionmembrane proteinIncAisrequiredforinclusionfusionanddelaysin IncAmembranelocalizationleadtodelayedhomotypic fusion[8,9,15].Therefore,weassessedthelocationof IncAintheinfectedneuroblastomacells.HeLaand neuroblastomacellswereinfectedwith C.trachomatis serovarL2,fixedat24hpiandstainedwithantibodies toIncA.IncAwaspresentoninclusionmembranesin bothHeLaandneuroblastomacells(Figure5Cand5D, respectively).Takentogether,thesedatademonstratethat thedelayininclusionfusionobservedinneuroblastoma cellsisnotduetodifferencesinfusioncompetencyor todifferencesinthepresenceofIncA.Additionally,when infectedneuroblastomasweregrownonfibronectin micropatternstoforcecentro someclustering,inclusion fusionwasrestored(Additionalfile2:FigureS1). Figure3 Chlamydialinclusiontraffickingandfusionisdyneindependent. Cos7cellswereinfectedwith C.trachomatis serovarL2following micro-injectionwithanti-dyneinantibodies.Uninjectedcellswereinfectedinparallel.Twenty-fourhourspostinfection,cellswerefixedands tained withhumansera(red)andtheappropriatesecondaryfortheanti-dyneinantibody(green).Representativepictureofanti-dyneininjectedcellsat6 and 24hpi (A and B, respectively ) .Inclusionsperinfectedcellwereenumeratedforinjectedanduninjectedcellsat24hpi,P<0.0001 (C) Richards etal.BMCMicrobiology 2013, 13 :185 Page4of7 http://www.biomedcentral.com/1471-2180/13/185

PAGE 6

Fusionisdelayedincellswithunanchoredmicrotubule minusendsChlamydialinclusionfusionoccursathostcentrosomes andisdelayedwhenextracentrosomesarepresent.Inclusionmigrationisunidirectionalresultinginthechlamydial inclusionresidingatthecellcentrosomeforitsentire intracellulargrowthphase.Inthecell,thecentrosomeacts astheorganizingcenterthatanchorsthemajorityof microtubuleminusends.Wehypothesizethatinclusion fusionispromotedbyinclusioncrowdingattheanchored minusendsofmicrotubules.Todetermineiffusionis dependentonmicrotubuleminusendanchoring,we transfectedHeLacellswiththeGFPtaggedEB1mutant, EB1.84-GFP.CellsexpressingEB1.84-GFPhavedefectsin microtubuleorganizationandcentrosomalanchoring resultinginunanchoredfreemicrotubuleminusends [12].Whenwecomparedinclusionfusioninthecellsthat hadbeenmocktransfectedtocellstransfectedwith EB1.84-GFP,theEB1.84producingcellsweremarkedly delayedininclusionfusion.At24hpi,transfectedcells averaged1.7inclusionsperinfectedcellwhilemock transfectedcellsaveragedoneinclusionperinfectedcell (P<0.001).Wealsoquantitatedthedistributionofinclusionnumbersinthesecells,slightlyunderhalfofthe cellstransfectedwithEB1.84-GFPcontainedoneinclusion (46%)whilethemajorityofmocktransfectedcells(92%) hadasingleinclusion(Figure6AandB,respectively). Additionally,manyoftheEB1.85transfectedcellshad Figure4 Inclusionfusionisdelayedincellswithmultipleunclusteredcentrosomes. HeLacells (A) andneuroblastomas (B) wereinfected with C.trachomatis atMOI~27andfixedat3and24hpi.Cellswerestainedwithanti-g-tubulinantibodies(green)andhumansera(red).HeLa cells (C) andneuroblastomas (D) wereinfectedwith C.trachomatis atMOI~3,9and27andfixedat10,12,14,16,20,22and24hpi.Cellswere stainedwithhumanseraandinclusionswereenumerated. Figure5 NeuroblastomasarefusioncompetentandIncA localizestotheinclusionmembraneduringinfection. HeLacells (A) andneuroblastomas (B) wereinfectedwith C.trachomatis serovarG.At40hpi,cellsweresuperinfectedwith C.trachomatis serovarL2andfixedfourhoursaftersuperinfection.Cellswere stainedwithhumansera(red)andanti-L2MOMPantibodies(green). HeLacells (C) andneuroblastomas (D) wereinfectedwith C. trachomatis serovarL2atMOI~9andfixed24hpi.Cellswere stainedwithhumansera(blue)andanti-IncAantibodies(green). Richards etal.BMCMicrobiology 2013, 13 :185 Page5of7 http://www.biomedcentral.com/1471-2180/13/185

PAGE 7

fourormoreinclusionspercell,whilemocktransfected cellsneverhadmorethantwoinclusionpercell(Figure6A andB,respectively).Representativeimagesofinclusions intransfectedandmocktransfectedcellsareshownin Figure6CandD,respectively.DiscussionandconclusionTheabilityof C.trachomatis inclusionstofuseiscriticalto pathogenicity.Comparedtowildtypestrains,rareisolates withnon-fusogenicinclusionsareclinicallyassociatedwith lessseveresignsofinfectionandlowernumbersofrecoverablebacteria[6].Incellculturehowever,arolefor inclusionfusionhasyettobedetermined.Matchedpairs ofnon-fusingandfusingstrainsaswellasnocodazole treatedanduntreatedmatchedsetsgrowatsimilarrates andproducecomparablenumbersofprogeny[16,17]. Chlamydialinclusionfusionishowevercriticaltopathogenicitythoughtheexactreasonforthisremainselusive. Homotypicinclusionfusionin C.trachomatis isaphenotypesharedbyallserovars.Consideringthatthemetabolicallyactiveformofthisobligateintracellularorganism isspatiallysequestered,itisplausiblethatsharingasingle inclusoplasmfacilitatesgeneticand/ornutrientexchange betweenbetweenco-infectingtrachomatisserovarsthus promotingtheirfitnesswithinapopulation.Itiswell establishedthat C.trachomatis storessugarsintheform ofglycogenintheinclusion[18,19]andthisglycogen storageislinkedtovirulenceaslossofthechlamydial crypticplasmidresultsinbothlossofglycogenstorageas wellasreducedvirulence[20].Homotypicinclusion fusionwouldallowthisresourcetobesharedbybacteria andmayleadtoacompetitivegrowthadvantageinahostileenvironmentsuchasthereproductivetrackduring invivoinfection. Acompleteunderstandingofmechanismsandfactors requiredforhomotypicfusioniscurrentlyunknown.The chlamydialinclusionmembraneproteinIncAistheonly chlamydialfactorknowntoberequiredforhomotypic inclusionfusion[9,21].Additionally,nohostfactorshave beenidentifiedtoberequiredforhomotypicfusion. Here,wedescribeanovelroleforproperinclusiontraffickingininclusionfusion.Throughlivecellimaging studies,weshowedthatinclusionfusionoccurspredominantlyatasinglesitewithinhostcells.ThissitewasinvariablytheMTOCofthecell(Figure1andAdditional file1).Earlyininfection,multipleinclusionsclustertightly attheMTOCandremainassociatedastheseinclusions begintofuse.Afterfusioniscomplete,thesingleinclusion retainsitscloseassociationwiththeMTOCasitcontinuestoexpand. TheMTOCcontainsthecellscentrosomesandactsas anorganizingfociforthecell.Additionally,theMTOC actsasthenucleationpointforcellularmicrotubules. Hostmicrotubulesarepolymerizedinapolarfashion; theplusendsundergorapidpolymerizationwhilethe minusendsareanchoredattheMTOCwhichallowsfor directionaltransportalongthemicrotubules.Wepreviouslydemonstratedthatthethenascentchlamydialinclusiontrafficksalongmicrotubulesusingthemicrotubule motorproteindynein[5].Thisstudydemonstratesthat inclusionmigrationisacriticalcomponentforefficientfusionasboththedyneinmotorproteinandintactmicrotubulesareimportantforinclusionfusion.Therequirement forbothanintactmicrotubulenetworkandthedynein motorproteinalongwiththeobservationthatfusiontakes placebetweencloselyadjacentinclusionssuggeststhat migrationtoacentrallocationinthecellisamechanism tophysicallydrivetheinclusionstogether.Thisincreases thelikelihoodthatthefusogenicproteinIncAonneighboringinclusionswillinteract,therebyenhancingatimely fusion.Thishypothesisisfurthersupportedbytheobservationthatwhentheminusendsofthemicrotubulesare notanchored(EB1.84expressingcells)ornotanchored atasinglesiteinthecell(neuroblastomas),fusionwas severelydelayed.Interestingly,inneuroblastomacells,the nonfusedinclusionsappeartobeincloseproximityto eachotherhowevertheresolutionoffluorescencemicroscopycannotresolvemolecularlevelinteractions.This suggeststhatforthechlamydialfusionproteinIncAto interactwithanIncAproteinonasecondinclusion,the distancebetweenthemwouldlikelyneedtobeverysmall. Figure6 TransfectionwithEB1.84-GFPdisruptsinclusion fusion. HeLacellsweretransfectedwithEB1.84-GFPormock transfected.Theyweretheninfectedwith C.trachomatis .Twentyfourhourspostinfection,cellswerefixedandstainedwithhuman seraandinclusionsperinfectedcellwereenumerated.The distributioninthenumberofinclusionsperinfectedcellisshown fortheEB1.84-GFPtransfectedandmocktransfectedcellsin A and B ,respectively.Mocktransfectedcellswerealsostainedwithanti-gtubulinantibodies(green).Representativetransfectedandmock transfectedcellsshownin C and D ,respectively. Richards etal.BMCMicrobiology 2013, 13 :185 Page6of7 http://www.biomedcentral.com/1471-2180/13/185

PAGE 8

Interestingly,fusionisonlydelayedunderthesecircumstancessuggestingthateventuallymultipleinclusionsin thecellcomeincloseenoughcontactfortheIncAdriven fusionsystemtomediatefusion. Overallourdatasupportamodelwherenascent chlamydia-containinginclusionstrafficalongmicrotubulesusingthedyneinmotorproteintodirectionally traffictotheminusendsofmicrotubules.Iftheminus endsofthemicrotubulesareanchoredattheMTOC, thenthemultipleinclusionsmakeclosecontactandare spatiallyarrangedtoencouragefusion.Interestingly,this traffickingtakesplacepriortoIncAexpression.Inclusion migrationisrapidandoccurswithinthefirstfewhoursof infectionhoweverIncAisonlyexpressedduringthemid cycleofchlamydialinfection,about8hoursafterinfection [22].Theinclusionsaremaintainedinafusion-supporting organizationuntilfusionisinitiatedthroughIncAprotein expressionandinsertionintotheinclusionmembrane. Thissuggestsastepwisepathwayofestablishingthe mature,fusion-competentchlamydialinclusion. Wehaveshownthatinclusionfusionoccursathostcell centrosomesandthatinorderforfusiontoresultinasingleinclusion,nascentinclusionsmustbetransportedby dyneinalongintact,anchoredmicrotubulestoasingle site.Comprehendingtheroleofmicrotubuletraffickingin inclusionfusiondynamicsiscrucialtoacompleteunderstandingofthemechanismsbywhichthisobligateintracellularpathogenpromotesitsintracellularsurvivaland pathogenicity.AdditionalfilesAdditionalfile1: Inclusionfusionoccursatminusendsof microtubules. MovieofFigure1. Additionalfile2:Figure2. Centrosomepositioningaffectschlamydial inclusionlocalization.Uninfectedandinfectedneuroblastomaswere platedonCYTOOchips(glasscoverslipsimprintedwithfibronectin micropatterns).Eachmicropatternisindicatedinthelowerleftofthetop panel.Infectedcellswerefixedat12and24hpi(topandbottompanel foreachshape,respectively).Cellswerestainedwithantibodiestogtubulin(green)andChlamydia(red).Nucleicacidisvisualizedbystaining withDRAQ5(blue). Authors ’ contributions TRcarriedouttheinfectionsandimmunofluorescenceexperimentsand draftedthemanuscript.AKacquiredconfocalimagesandcontributedto dataanalysis.SGcontributedtodataanalysisandfinalizedthemanuscript. Allauthorsreadandapprovedthefinalmanuscript. Received:13May2013Accepted:1August2013 Published:7August2013 References1.WeinstockH,BermanS,CatesW: Sexuallytransmitteddiseasesamong Americanyouth:incidenceandprevalenceestimates,2000. PerspectSex ReprodHealth 2004, 36: 6 – 10. 2.CliftonDR,FieldsKA,GrieshaberSS,DooleyCA,FischerER,MeadDJ, CarabeoRA,HackstadtT: AchlamydialtypeIIItranslocatedproteinis tyrosine-phosphorylatedatthesiteofentryandassociatedwith recruitmentofactin. ProcNatlAcadSciUSA 2004, 101: 10166 – 10171. 3.DehouxP,FloresR,DaugaC,ZhongG,SubtilA: Multi-genome identificationandcharacterizationofchlamydiae-specifictypeIII secretionsubstrates:theIncproteins. BMCGenomics 2011, 12: 109. 4.HackstadtT,FischerER,ScidmoreMA,RockeyDD,HeinzenRA: Originsand functionsofthechlamydialinclusion. TrendsMicrobiol 1997, 5: 288 – 293. 5.GrieshaberSS,GrieshaberNA,HackstadtT: Chlamydiatrachomatisuses hostcelldyneintotraffictothemicrotubule-organizingcenterinap50 dynamitin-independentprocess. JCellSci 2003, 116: 3793 – 3802. 6.GeislerWM,SuchlandRJ,RockeyDD,StammWE: Epidemiologyand clinicalmanifestationsofuniqueChlamydiatrachomatisisolatesthat occupynonfusogenicinclusions. JInfectDis 2001, 184: 879 – 884. 7.RidderhofJC,BarnesRC: Fusionofinclusionsfollowingsuperinfectionof HeLacellsbytwoserovarsofChlamydiatrachomatis. InfectImmun 1989, 57: 3189 – 3193. 8.FieldsKA,FischerE,HackstadtT: InhibitionoffusionofChlamydia trachomatisinclusionsat32degreesCcorrelateswithrestrictedexport ofIncA. InfectImmun 2002, 70: 3816 – 3823. 9.HackstadtT,Scidmore-CarlsonMA,ShawEI,FischerER: TheChlamydia trachomatisIncAproteinisrequiredforhomotypicvesiclefusion. CellMicrobiol 1999, 1: 119 – 130. 10.HowardL,OrensteinNS,KingNW: Purificationonrenografindensity gradientsofChlamydiatrachomatisgrownintheyolksacofeggs. ApplMicrobiol 1974, 27: 102 – 106. 11.ScidmoreMA: CultivationandLaboratoryMaintenanceofChlamydia trachomatis. CurrProtocMicrobiol 2005, Chapter11: Unit11A – 1. 12.AskhamJM,VaughanKT,GoodsonHV,MorrisonEE:Evidencethatan interactionbetweenEB1andp150(Glued)isrequiredfortheformation andmaintenanceofaradialmicrotubulearrayanchoredatthe centrosome. MolBiolCell 2002, 13: 3627 – 3645. 13.SharpGA,OsbornM,WeberK: Ultrastructureofmultiplemicrotubule initiationsitesinmouseneuroblastomacells. JCellSci 1981, 47: 1 – 24. 14.KnowltonAE,BrownHM,RichardsTS,AndreolasLA,PatelRK,GrieshaberSS: Chlamydiatrachomatisinfectioncausesmitoticspindlepoledefects independentlyfromitseffectsoncentrosomeamplification. Traffic 2011, 12: 854 – 866. 15.SuchlandRJ,RockeyDD,BannantineJP,StammWE: IsolatesofChlamydia trachomatisthatoccupynonfusogenicinclusionslackIncA,aprotein localizedtotheinclusionmembrane. InfectImmun 2000, 68: 360 – 367. 16.SuchlandRJ,JeffreyBM,XiaM,BhatiaA,ChuHG,RockeyDD,StammWE: IdentificationofconcomitantinfectionwithChlamydiatrachomatis IncA-negativemutantandwild-typestrainsbygenomic,transcriptional, andbiologicalcharacterizations. InfectImmun 2008, 76: 5438 – 5446. 17.SchrammN,WyrickPB: CytoskeletalrequirementsinChlamydia trachomatisinfectionofhostcells. InfectImmun 1995, 63: 324 – 332. 18.GORDONFB,QUANAL: Occurenceofglycogenininclusionsofthe psittacosis-lymphogranulomavenereum-trachomaagents. JInfectDis 1965, 115: 186 – 196. 19.FanVS,JenkinHM: GlycogenmetabolisminChlamydia-infectedHeLa-cells. JBacteriol 1970, 104: 608 – 609. 20.RussellM,DarvilleT,Chandra-KuntalK,SmithB,AndrewsCW,O ’ ConnellCM: Infectivityactsasinvivoselectionformaintenanceofthechlamydial crypticplasmid. InfectImmun 2011, 79: 98 – 107. 21.RockeyDD,FischerER,HackstadtT: Temporalanalysisofthedeveloping Chlamydiapsittaciinclusionbyuseoffluorescenceandelectron microscopy. InfectImmun 1996, 64: 4269 – 4278. 22.Scidmore-CarlsonMA,ShawEI,DooleyCA,FischerER,HackstadtT: IdentificationandcharacterizationofaChlamydiatrachomatisearly operonencodingfournovelinclusionmembraneproteins. MolMicrobiol 1999, 33: 753 – 765.doi:10.1186/1471-2180-13-185 Citethisarticleas: Richards etal. : Chlamydiatrachomatishomotypic inclusionfusionispromotedbyhostmicrotubuletrafficking. BMC Microbiology 2013 13 :185.Richards etal.BMCMicrobiology 2013, 13 :185 Page7of7 http://www.biomedcentral.com/1471-2180/13/185