cells Article Injured Achilles Tendons Treated with Adipose-Derived Stem Cells Transplantation and GDF-5 AndreaAparecidadeAro1,2,*,GianeDanielaCarneiro1,LuisFelipeR.Teodoro1, FernandaCristinadaVeiga3,DaniloLopesFerrucci1,GustavoFerreiraSimões1, PriscylaWaleskaSimões4 ID,LúciaElviraAlvares3,AlexandreLeiteR.deOliveira1, CristinaPontesVicente1,CaioPerezGomes5 ID,JoãoBoscoPesquero5 ID, MarceloAugustoM.Esquisatto2,BenedictodeCamposVidal1andEdsonRosaPimentel1 1 DepartmentofStructuralandFunctionalBiology,InstituteofBiology,StateUniversityof Campinas–UNICAMP,CharlesDarwin,s/n,CP6109,13083-970Campinas,SP,Brazil; [email protected](G.D.C.);[email protected](L.F.R.T.);[email protected](D.L.F.); [email protected](G.F.S.);[email protected](A.L.R.d.O.);[email protected](C.P.V.); [email protected](B.d.C.V.);[email protected](E.R.P.) 2 BiomedicalSciencesGraduateProgram,HerminioOmettoUniversityCenter–UNIARARAS, 13607-339Araras,SP,Brazil;[email protected] 3 DepartmentofBiochemistryandTissueBiology,InstituteofBiology,StateUniversityof Campinas–UNICAMP,CharlesDarwin,s/n,CP6109,13083-970Campinas,SP,Brazil; [email protected](F.C.d.V.);[email protected](L.E.A.) 4 Engineering,ModelingandAppliedSocialSciencesCenter(CECS),BiomedicalEngineeringGraduate Program(PPGEBM),UniversidadeFederaldoABC(UFABC),AlamedadaUniversidades/n, 09606-045SãoBernardodoCampo,SP,Brazil;[email protected] 5 DepartmentofBiophysics,FederalUniversityofSaoPaulo–Unifesp,PedrodeToledo,699, 04039-032SaoPaulo,SP,Brazil;[email protected](C.P.G.);[email protected](J.B.P.) * Correspondence:[email protected];Tel.:+55-19-3543-1423 (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1) (cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7) Received:19July2018;Accepted:23August2018;Published:31August2018 Abstract: Tendon injuries represent a clinical challenge in regenerative medicine because their naturalrepairprocessiscomplexandinefficient. Thehighincidenceoftendoninjuriesisfrequently associatedwithsportspractice,aging,tendinopathies,hypertension,diabetesmellitus,andtheuse ofcorticosteroids. Thegrowinginterestofscientistsinusingadipose-derivedmesenchymalstem cells(ADMSC)inrepairprocessesseemstobemostlyduetotheirparacrineandimmunomodulatory effectsinstimulatingspecificcellularevents. ADMSCactivitycanbeinfluencedbyGDF-5,whichhas beensuccessfullyusedtodrivetenogenicdifferentiationofADMSCinvitro. Thus,wehypothesized thattheapplicationofADMSCinisolationorinassociationwithGDF-5couldimproveAchilles tendon repair through the regulation of important remodeling genes expression. Lewis rats had tendonsdistributedinfourgroups: Transected(T),transectedandtreatedwithADMSC(ASC)or GDF-5(GDF5),orwithboth(ASC+GDF5). Inthecharacterizationofcellsbeforeapplication,ADMSC expressed the positive surface markers, CD90 (90%) and CD105 (95%), and the negative marker, CD45(7%). ADMSCwerealsodifferentiatedinchondrocytes,osteoblast,andadipocytes. Onthe 14thdayafterthetendoninjury,GFP-ADMSCwereobservedinthetransectedregionoftendons intheASCandASC+GDF5groups, andexhibitedand/orstimulatedasimilargenesexpression profile when compared to the invitro assay. ADMSC up-regulated Lox, Dcn, and Tgfb1 genes expressionincomparisontoTandASC+GDF5groups,whichcontributedtoalowerproteoglycans arrangement, and to a higher collagen fiber organization and tendon biomechanics in the ASC group. TheapplicationofADMSCinassociationwithGDF-5down-regulatedDcn,Gdf5,Lox,Tgfb1, Mmp2, and Timp2 genes expression, which contributed to a lower hydroxyproline concentration, lower collagen fiber organization, and to an improvement of the rats’ gait 24 h after the injury. Cells2018,7,127;doi:10.3390/cells7090127 www.mdpi.com/journal/cells Cells2018,7,127 2of22 Inconclusion,althoughtheliteraturedescribesthebeneficeffectofGDF-5forthetendonhealing process,ourresultsshowthatitsapplication,isolatedorassociatedwithADMSC,cannotimprove therepairprocessofpartialtransectedtendons,indicatingthehighereffectivenessoftheapplication ofADMSCininjuredAchillestendons. OurresultsshowthattheapplicationofADMSCininjured AchillestendonswasmoreeffectiveinrelationtoitsassociationwithGDF-5. Keywords: repair;extracellularmatrix;collagen;gait;biomechanics;geneexpression 1. Introduction Tendoninjuriesareverycommonandtheirnaturalrepairprocessisextremelyslow,complex, and inefficient due to their intrinsic hypocellularity and hypovascularity, representing a clinical challenge to orthopedists mainly because these injuries often respond poorly to treatment [1]. As reviewed by Zabrzyn´ski et al. [2], the occurrence of tendon injuries is associated with sports, aging,tendinophaties,hypothyroidism,hypertension,diabetesmellitus,arthropathies,corticosteroids, vitaminCdeficiency,andothers. Collagenfibersarethemaincomponentoftheabundantextracellular matrix(ECM)oftendons,andwiththeproteoglycans(PG),thecollagenfibersformahighlyorganized supramolecularstructure,abletoattendtothebiomechanicaldemandsonthetissue[3,4]. Particularly, smallleucine-richproteoglycans(SLRPs),suchasdecorin,fibromodulim,andbiglycan,arethemost abundant PG in tendons, with decorin representing 80% of the total proteoglycan content of the tissue[5]. SLRPsinteractwiththecollagenfibrils,actinginthefibrillogenesisofcollagen,andprobably regulatingthegrowthindiameterofthesefibrils[6]. Besidestheparallelbundlesofpredominantly type I collagen fibers and PG, non-collagenous glycoproteins, matrix metalloproteinases (MMP), growthfactors,andcellsalsocompriseelementsofthetendons[5,7–9]. Thespecificmechanicaltendonpropertiesaredirectlyrelatedtothehighorganizationofcollagen bundles[10]. Afterinjuries,thestructuralorganizationandcompositionoftendonsarenotcompletely restored, withafibrousscarformationthatcancausesignificantdysfunctionandjointmovement inability[11,12]. Aftertherepairprocess,tendonsbecomebiomechanicallyweakened,makingitmore pronetore-rupture[9,13]. Clearly,newtreatmentsareneededwiththeobjectiveofimprovingtendon repair,consideringthattheseinjuriesreachthepopulationthatiseconomicallyactive,andthattheuse ofcelltherapyusingmesenchymalstemcells(MSC)couldbeagoodstrategy. Adult tissues are attractive MSC sources, which are characterized as undifferentiated cells, with mesodermal differentiation potential and self-renew and high proliferative capacities [14]. AdiposetissueisanalternativesourceofMSCthatcanbeobtainedbyalessinvasivemethodunder localanesthesia,withlittleassociatedpatientdiscomfortandinlargerquantitiesascomparedwithbone marrow[15]. Theeffectivenessofadipose-derivedmesenchymalstemcells(ADMSC)inregenerative medicine seems to be due to their paracrine effects that stimulate specific cellular events, like the growthfactorsandcytokinesdelivery[16],aswellastheirimmunomodulatoryeffectstopromote tendonregeneration[17]. Besidesrecruitingprogenitorcellsfromsometissues,theADMSCcouldalso differentiatespecificcellsduringtissuerepair[18–20]. However,theliteratureiscontroversialabout theeffectsofADMSCtransplantationduringtendonrepair. Insomeanimalmodels,therewasgreater structuralorganization,biomechanicalproperties,densityofcollagenfibers,collagentypesIandIII genesexpression,andgrowthfactorssynthesisasfibroblastgrowthfactor(FGF),vascularendothelial growthfactor(VEGF),andtransforminggrowthfactor(TGF-β)afterADMSCapplicationinrelationto untreatedtendons[7,21–23]. Conversely,nodifferenceswereobservedafterthiscellulartherapyinthe biomechanicalparametersorcollagenproduction[7,17,23]. It is widely accepted that the control of stem cell activity is influenced by several different environmental factors [24], including growth factors [1]. To drive tenogenic ADMSC in invitro differentiation, insulinlikegrowthfactor(IGF)-1orTGF-βinaco-culturewithprimarytenocytes Cells2018,7,127 3of22 and growth differentiation factor-5 (GDF-5) have been used successfully [19,20]. GDF-5 belongs totheTGF-βsuperfamilyanditisknownascartilage-derivedmorphogeneticprotein-1(CDMP-1) and bone-derived morphogenetic protein-14 (BMP-14). Storm et al. [25] documented two other membersofthissuperfamily,suchasGDF-6(BMP-13)andGDF-7(BMP-12). GDFfamilymembers arealsocalledbone-derivedmorphogeneticproteins(BMP)becausetheyarefoundpredominantly during the development of endochondral bones [26] and joint formation [27]. Wolfman et al. [28] demonstratedthatthethreemembersofBMPinducetheformationofconnectivetissuerichincollagen I,withanimportantroleintheprocessoftendonhealing[29,30]. Fromthisstudy,theBMPceased tobeconsideredasonlychondrogenicandosteogenicfactors,assuggestedbythename,andwere consideredtendinogenicfactors. Forslundandcolleagues[30],whoshowedincreasedAchillestendon resistanceafterrupture,reportedtheinitialsuccessofBMP-12,-13,and-14intendonhealing.Inastudy ofChhabraetal.[31],micedeficientintheGDF-5genehadapoorhealingprocess,withlesserstructural organizationanddecreasedbiomechanicalpropertiesoftendons,evidencingtheimportanceofthis growthfactorduringtendonrepairprocesses. Currently,celltherapyusingtheADMSCassociated withtheexogenousapplicationofgrowthfactorsrepresentsagreatpotentialintheprocessoftendon repair. Despitepromisingstudiesinanimals,notreatmentassociatedwiththeapplicationofADMSC intendoninjurieshasbeenusedinclinicsduetothelackofknowledgeonmolecularaspectsinvolving thosetherapies. TheobjectiveofthepresentstudywastotestthehypothesisthattheapplicationofADMSCin isolationorassociatedwithGDF-5couldimproveAchillestendonrepair. TheuseofGDF-5wasbased ontheliteraturethatdemonstratesitsimportanceduringtendonhealingandtheroleofGDF-5in modulatingADMSCtenogenicdifferentiationinvitro.Thus,thedown-orup-regulationofremodeling genesexpressioninresponsetoADMSCandGDF-5applicationwereanalyzed,andtheinvolvement ofthosegenesintherestorationofthestructural,biomechanical,andfunctionalpropertiesofAchilles tendonsafterpartialtransection. 2. MaterialsandMethods 2.1. InVitroExperiments 2.1.1. IsolationofADMSCandCellCulture TheprocedurewasdoneaccordingtoYangetal.[32]withsomemodifications. Adiposetissue wasobtainedfromtheinguinalregionof10maleLewisratsbetween90–120days. Adiposetissuewas cutandwashedinDulbecco’smodifiedphosphatebufferedsalinesolution(DMPBSFlushwithout calciumandmagnesium)containing2%streptomycin/penicillintoremovecontaminatingbloodcells. Then,0.2%collagenase(Sigma-Aldrich® Inc.,SaintLouis,MO,USA)wasaddedtodegradationofthe ECMandthesolutionwasmaintainedat37◦Cundergentlestirringfor1htoseparatethestromal cellsfromprimaryadipocytes. Dissociatedtissuewasfilteredusingcellstrainers(40µm)andthe inactivationofcollagenasewasthendonebytheadditionofanequalvolumeofDulbecco’smodified Eagle’smedium(DMEM)supplementedwith15%fetalbovineserum(FBS),followedbycentrifugation at1800rpmfor10min. Thesuspendingportioncontaininglipiddropletswasdiscardedandthepellet wasresuspendedinDMEM(containing50mg/Lpenicillinand50mg/Lstreptomycin)with15%FBS, andtransferredto25cm2flasksfor48hours. Afterconfluence,cellsweretransferredto75cm2flasks (1stpassage). Themediumwasreplacedafter48handthenevery3days. Culturesweremaintained at37◦Cwith5%CO untilthe5thpassage(5P),alwaysatupto80%confluency. 2 2.1.2. FlowCytometry ADMSC at 5P (n = 4) were trypsinized and centrifuged at 1800 rpm for 10 min, and counted usingtheNeubauerchamber. 1×106ADMSCwereresuspendedin200uLofDMPBSFlushwith2% BSA(bovineserumalbumin). Fortheimmunophenotypicpanel,thefollowingantibodieswereused: Cells2018,7,127 4of22 CD90-APC,CD105-PE,andCD45-APCdoubleconjugated(eBioscience®Inc.,SanDiego,CA,USA) were diluted 1:200 and incubated for 40 min at room temperature. Subsequently, ADMSC were washedtwicewith500µLofDMPBSFlushandcentrifugedat2000rpmfor7min. TheADMSCwere resuspendedinDMPBSFlushwith2%BSA,followedbytheflowcytometryanalysis. 2.1.3. InVitroDifferentiationPotentialofADMSC Usingdifferentmediaforeachtypeofdifferentiation,ADMSC(5P)werecultured(2×104cells) according to Yang et al. [32] with some modifications. Osteogenic differentiation (n = 3): DMEM supplementedwith10%FBS,0.1µMdexamethasone,200µMascorbicacid,and10mMβ-glycerol phosphate. Adipogenic differentiation (n = 3): DMEM supplemented with 10% FBS, 1 µmol/L dexamethasone,50µmol/Lindomethacin,0.5mM3-isobutyl-1-methyl-xanthine,and10µMinsulin. Condrogenicdifferentiation(n=3): DMEMsupplementedwith10%FBS,acidfree15mMHEPES, 6.25µg/mLinsulin,10ng/mLTGF-β1,and50nMAsAP.Culturesweremaintainedat37◦Cwith5% CO ,andthecompletemediumswerereplacedtwiceaweek. Attheendoffourweeks,cellswere 2 fixedwith4%paraformaldehydefor20minandstainedwith2%AlizarinRedS(pH4.1)during5min forcalciumdetection,with0.025%ToluidineblueinMcIlvainebuffer(0.03Mcitricacid,0.04Msodium phosphatedibasic,pH4.0)during10minforproteoglycansdetection,andwith1%SudanIVduring 5mintoshowlipiddroplets. SampleswereimagedontheAxiovertS100(ZEISS)invertedmicroscope. 2.1.4. CellViability Flasksof75cm2 withADMSCin5P(n=3)weretrypsinizedandcentrifugedat1800rpmfor 10min.Thepelletobtainedfromeachflaskswasresuspendedin1mLofDMPBSFlush.Then,analiquot of10µLofeachculturewasstainedwith0.4%TrypanBlue,andtheADMSCwereplacedforanalysis intheCountessIIFL(LifeTechnologies®Inc.,Carlsbad,CA,USA)equipmentforcellconcentration andviabilitymeasurements. 2.1.5. ContrastbyDifferentialInterference(DIC) For the birefringence analyses of nuclei, ADMSC (5P) in culture (n = 3) were stained with 0.025%toluidinebluesolutionin0.1MMcIlvainephosphatebufferatpH4.3for30min,followedby analysesintheOlympusBX-51(OlympusAmerica,CenterVallery,PA,USA)polarizingmicroscope equippedwithaQ-color5camera(OlympusAmerica),andusingtheImagePro-plusv.6.3software forWindows™(MediaCybernetics,SilverSpring,MD,USA)[4,33,34]. Withthemicroscopeandthe software,itispossibletocarryoutanalysisofDICandtheanisotropicproperties,bothindividuallyand incombination[34]. DICobservationswereperformedafteradditionofthetwocondenser’sWollaston prisms in the light path. With DIC-PLM, the optical path differences in samples can be detected throughtheslidingofanotherWollastonprismthatispositionedundertheanalyzer. Theresulting colorsinthesamplearecomparedwiththecolorsinaMichel-Lévychart. 2.1.6. RealTime-PCRArrayforAnalysisofADMSCGeneExpression For the total RNA extraction of the ADMSC (5P) using Trizol® reagent, Invitrogen, Carlsbad, CA, USA, three different cultures were analyzed. A spectrophotometer (NanoDrop® ND-1000, Thermo Fisher Scientific®, Waltham, MA, USA) was used to quantify the RNA in each sample by determining the absorbance ratio at 260 and 280 nm. 0.5 µg from the total extracted RNA of each sample was used for the synthesis of cDNA using the RT2 First Strand Kit (QIAGEN®, Hilden, Germany) and thermocycler Mastercycler Pro (Eppendorf®, Hamburg, Germany), also following the manufacturer’s instructions. The cDNA was frozen at −20 ◦C until tested. The RT-PCR array reaction was performed using the RT2 Profiler PCR Arrays (A format) kit in combination with the RT2 SYBR Green Mastermixes (QIAGEN®, Hilden, Germany) on the thermocycler apparatus 7300 (ABI Applied Biosystems®, Foster City, CA, USA), following the manufacturer’s instructions. For each culture sample, three types of reaction controls were used: Cells2018,7,127 5of22 1. Positive PCR control; 2. Reverse transcriptase control; and 3. Control for contamination of rat genomic DNA. The Glyceraldehyde-3-phosphate dehydrogenase (Gapdh, NM_017008) was used as an endogenous control for each sample. The following genes were analyzed (QIAGEN®): Scleraxis (Scx, NM_001130508); Tenomodulin (Tnmd, NM_022290); Tumor necrosis factor (TNF superfamily, member2)(Tnf, NM_012675); Interleukin1beta(II1b, NM_031512); Transforminggrowthfactor, beta1 (Tgfb1,NM_021578);Matrixmetallopeptidase2(Mmp2,NM_031054);Matrixmetallopeptidase9(Mmp9, NM_031055);Matrixmetallopeptidase8(Mmp8,NM_022221);TIMPmetallopeptidaseinhibitor2(Timp2, NM_021989);Decorin(Dcn,NM_024129);Lysyloxidase(Lox,NM_017061);andGrowthdifferentiation factor 5 (Gdf5, XM_001066344). Reactions were made in a single cDNA pipetting for each gene, including the endogenous control. ∆CT values were obtained by the difference between the CT valuesofthetargetgenesandtheGapdhgene. The2−∆CTmethod[35]wasusedtocalculatethegene expressionforeachtargetgene. 2.2. InVivoExperiments 2.2.1. ExperimentalGroups Atotalof110maleLewisrats(120-day-old),withfreeaccesstofoodandwater,weredividedinto 5experimentalgroups(22animalsforeachgroup):Normal(N):Ratswithtendonswithouttransection; Transected(T):RatswithpartiallytransectedtendonsandtreatedwithtopicalapplicationofDMPBS Flushinthetransectedregion;Mesenchymalstemcellsderivedfromadiposetissue(ASC):Ratswith transectedtendonswithsubsequenttransplantofADMSC(3.7×105cells)inthetransectedregion; GDF-5(GDF5): RatswithtransectedtendonsandtreatedwithtopicalapplicationofDMPBSFlushin thetransectedregion+GDF5application(500ng)24hafterpartialtransection;andwithADMSCand GDF-5(ASC+GDF5): RatswithtransectedtendonsandtreatedwithsubsequenttransplantofADMSC (3.7 × 105 cells)inthetransectedregion+GDF5application(500ng)24hafterpartialtransection. Animalswereeuthanizedonthe14thdayaftertransectionbyanoverdoseofanesthetic(Ketamine andXylazine). AnimalsofgroupNwereeuthanizedat134daysandthetendonswithouttransection werecollectedforanalysis. 2.2.2. PartialTransectionoftheCalcanealTendonandApplicationofADMSCandGDF-5 The animals were anesthetized with intraperitoneal injection of Ketamine (90 mg/Kg) and Xylazine (12 mg/Kg), and the right lower paws submitted to antisepsis and trichotomy. For the exposureofthecalcaneustendon,alongitudinalincisionwasmadeintheanimal’sskin,followed byatransversepartialtransectionperformedintheproximaltendonregionlocatedatadistanceof 4mmfromitsinsertioninthecalcaneusbone[36,37]. Approximately3.7×105 ADMSC(5P)were resuspendedin15µLofDMPBSFlushandtransplantedinthetransectedregionoftendonsinthe ASCgroupusingapipette. 15µLofDMPBSFlushwasappliedintendonsoftheTgroup,andthe GDF5grouptendonsreceivedanapplicationof15µLofDMPBSFlush+500ngofGDF-524hafter the tendon transection. All applications were made in the region of the tendon where the partial transectionwasperformed. Then,theskinwassuturedwithnylonthread(Shalon5-0)andaneedle (1.5cm). AllsurgicalandexperimentalprotocolswereapprovedbytheInstitutionalCommitteefor EthicsinAnimalResearchoftheStateUniversityofCampinas-UNICAMP-Brazil(Protocolnº2905-1). 2.2.3. PreparationofSectionsinFreezing TendonswereplacedinTissue-Tek®,frozen,andcutincryostat(seriallongitudinalcutsof7µm thickness). The sections were fixed using a 4% formaldehyde solution in Millonig buffer (0.13 M sodiumphosphateand0.1Msodiumhydroxide,pH7.4)for20min,followedbybirefringenceand lineardichroismanalysis. Cells2018,7,127 6of22 2.2.4. DAPIStaining Immediately after fixation, the sections (n = 4) were incubated with DAPI (4(cid:48),6-Diamidino-2-phenylindole dihydrochloride) (0.1 mg/mL in methanol) for 5 min at 37 ◦C. Thesectionswereanalyzedbyfluorescencemicroscope(OlympusBX60)andtheimagescapturedby theQCapture4.0program. 2.2.5. PolarizationMicroscopy: BirefringenceMeasurements After fixation, image analyses of the tendons (n = 4) were evaluated to detect differences in morphology based on the aggregation and organization of the collagen bundles, which reflect the variation of birefringence intensity. Birefringence properties were studied using an Olympus BX53 polarizing microscope and an image analyzer (Life Science Imaging Software, Version 510_UMA_cellSens16_Han_en_00). Because the birefringence appears visually as brilliance, thisphenomenonwasmeasuredwithanimageanalyzerandexpressedasgrayaverage(GA)values inpixels(8bits=1pixel). Thelargertendonaxiswaspositionedat45◦ tothecrossedanalyzerand polarizer. Ascollagenbundlesexhibittwotypesofbirefringence,intrinsicbirefringence(Bi)andform ortexturalbirefringence(Bf)[38],thetotalbirefringence(thesumofBiandBf)wasusedinthisstudy. Measurementsofthetendonsineachexperimentalgroupweremadeafterimmersingthesections in water, a condition in which total birefringence is highly detectable [4,33,34,38]. The number of measurementsofGAwasrepresentedasthemedianandtheywerechosenatrandomin16sections fromfourtendonsofeachgroup. 2.2.6. LinearDichroismMeasurements Linear dichroism measurements (n = 4) were obtained from toluidine blue-stained sections. LineardichroismmeasurementshaveshownthatGAGchainspresentincollagenfiberPGsarelinearly distributedandpredominantlyparalleltothelongestfiberaxis[3,39,40]. Inthiscase,lineardichroism is an extrinsic phenomenon, resulting from the arrangement of toluidine blue molecules that are electrostaticallyboundtotheanioniclinksitesoftheorientedsubstrate. Thedichroicratio(DR=d/d⊥) wasdeterminedbythetoluidineblueabsorbanceintheparallel(d)andperpendicular(d⊥)positions ofthetendon’slongestaxis,withregardtothepolarizedlightplane(PLP)[39,40]. Lineardichroism wasmeasuredusinganOlympusBX53polarizingmicroscope(Objective: OlympusUPlanFLN40×; Camera:OlympusQ-color5;Polarizer:OlympusU-POT)andanimageanalyzer(LifeScienceImaging Software,Version510_UMA_-cellSens16_Han_en_00). Thenumberofmeasurements(~100)ofGA wasrepresentedasthemedianandtheywerechosenatrandomin16sectionsfromfourtendonsof eachgroup. 2.2.7. RealTime-PCRArray Thecollectedtendons(n=4)wereplacedinstabilizingsolution(RNA-later,QIAGEN®Hilden, Germany) and maintained at −20 ◦C. For the total RNA extraction, the tendons were sprayed using liquid nitrogen and then homogenized in a tube containing 5 stainless steel balls (2.3 mm diameter,Biospec)bybeingshakeninaTissueLyserLTinstrument(QIAGEN® Hilden,Germany), with2repetitions(60s)intercaletedwithicecooling(2min)betweeneachshakingstep(5). TotalRNA was isolated from each sample using the RNeasy® Fibrous Tissue Mini Kit (QIAGEN® Hilden, Germany),followingthemanufacturer’sinstructions. Aspectrophotometer(NanoDrop®ND-1000, Thermo Fisher Scientific®, Waltham, Massachusetts, USA) was used to quantify the RNA in each samplebydeterminingtheabsorbanceratioat260and280nm. 0.5µgfromthetotalextractedRNAof eachsamplewasusedforthesynthesisofcDNAusingtheRT2 FirstStrandKit(QIAGEN® Hilden, Germany)andthermocyclerMastercyclerPro(Eppendorf®Hamburg,Germany),alsofollowingthe manufacturer’sinstructions. ThecDNAwasfrozenat−20◦Cuntiltested. TheRT-PCRarrayreaction wasperformedusingtheRT2ProfilerPCRArrays(Aformat)kitincombinationwiththeRT2SYBR Cells2018,7,127 7of22 GreenMastermixes(QIAGEN®Hilden,Germany)onthethermocyclerapparatus7300(ABIApplied Biosystems®,FosterCity,CA,USA),followingthemanufacturer’sinstructions.Foreachanimalsample, threetypesofreactioncontrolswereused: 1. PositivePCRcontrol;2. Reversetranscriptasecontrol; and3. ControlforcontaminationofratgenomicDNA.TheGlyceraldehyde-3-phosphatedehydrogenase (Gapdh, NM_017008) was used as an endogenous control for each sample. The following genes were analyzed (QIAGEN® Hilden, Germany): Scleraxis (Scx, NM_001130508); Tenomodulin (Tnmd, NM_022290); Tumor necrosis factor (TNF superfamily, member 2) (Tnf, NM_012675); Interleukin 1beta (II1b, NM_031512); Transforming growth factor, beta 1 (Tgfb1, NM_021578); Matrix metallopeptidase 2 (Mmp2,NM_031054);Matrixmetallopeptidase9(Mmp9,NM_031055);Matrixmetallopeptidase8(Mmp8, NM_022221); TIMP metallopeptidase inhibitor 2 (Timp2, NM_021989); Decorin (Dcn, NM_024129); Lysyloxidase(Lox,NM_017061);andGrowthdifferentiationfactor5(Gdf5,XM_001066344). Reactions weremadeinasinglecDNApipetteforeachgene,includingtheendogenouscontrol. ∆CTvalueswere obtainedbythedifferencebetweentheCTvaluesofthetargetgenesandtheGapdhgene. Thesevalues werenormalizedbysubtractingthe ∆CTvalueofthecalibratorsample(Tgroup)toobtain ∆∆CT values. A2−∆∆CT methodwasusedtocalculatetherelativeexpressionlevel(foldchange)foreach targetgene. Resultswererepresentedastherelativegeneexpressionincomparisontothecalibrator samplethatisequalto1. 2.2.8. DosageofHydroxyproline Hydroxyproline was used as an indicator of the amount of total collagen in the tendons of thedifferentgroups(n=5)usedpreviouslyinthebiomechanicalassay. Thetendonswerecutand immersedinacetonefor48h,followedbyasolutioncontainingchloroform:ethanol(2:1)alsofor48h. Afterdehydration,thesampleswereplacedfordryingintheovenat37◦C.Sampleswereweighed and hydrolyzed in 6N HCl (1 mL/10 mg of tissue) for 4 h at 130 ◦C according to Stegemann and Stalder[41]withsomemodifications,andneutralizedwith6NNaOH.Theabsorbanceofthesamples wasmeasuredat550nmusingamicroplatereader(ExpertPlus,Asys®,Holliston,MA,USA). 2.2.9. EvaluationoftheMaxContactIntensityoftheRatPawafterPartialTransection The CatWalk system (Noldus Inc., Wageningen, The Netherlands) was used to analyze the gait recovery of the animals (n = 5). In this protocol, the rats crossed a walkway (100 cm length, 5cmwidth,and0.6cmthickness)withaglassfloorilluminatedfromthelongedgeinadarkroom. Dataacquisitionwasperformedwithahigh-speedcamera(PulnixTM-765ECCD),andthepawprints wereautomaticallyclassifiedbythesoftware. Thepawprintswereobtainedduringthe3daysbefore thepartialtransectionofthetendonstoassessthenormalstandardgaitoftheanimals,andtheywere collectedagainafterthelesions. Post-operativedatawereassessedonthe1st,3rd,5th,7th,9th,11th, and13thdaysfollowingsurgicallesion. Theparametersusedhereinwere“MaxContactIntensity”, corresponding to the pressure exerted by the paw on the glass floor during gait. The intensity of magnificationcanvaryfrom0to255pixels. 2.2.10. BiomechanicalParameters Tendons from experimental groups (n = 5) were collected and stored at −20 ◦C until tested. Beforethebiomechanicaltest,thetendonswerethawedandmeasuredwithpachymeter,considering their length, width, and thickness. For the biomechanical assay, the tendons were maintained in PBS to prevent their fibers from drying out. Then, the tendons were fixed to metal claws by the myotendinousjunctionandbytheosteotendinousjunctionforcorrectalignmentintheequipment (Texturometer, MTSmodelTESTSTARII).Ineachbiomechanicalassay, tendonsweresubjectedto a gradual increase of load at a displacement velocity of 1 mm/s by using a load 0.05 N until the tendonruptured. BiomechanicalparameterswereanalyzedaccordingtoBiancalanaetal.[42]and Tomiossoetal.[9],suchasmaximumforce(N)andmaximumdisplacement(mm),whichwereused tocalculatethemaximumstress(Mpa)andmaximumstrain(L)oftendonsfromtheexperimental Cells2018,7,127 8of22 groups. The cross-sectional area of the calcaneus tendon was calculated by assuming an elliptical approximation(A=πWd/4)usingmeasurementsofwidth(W)andthickness(d)valuesfromthe sameSparrowetal.[43]. Themaximumstressvalue(MPa)wasestimatedbytheratiobetweenthe maximumload(N)andthecross-sectionalarea(mm2). Themaximumdeformation(L)wascalculated through (L = L − L/L), where (L ) is the value of the final length before rupture, and (L) is the f i i f i initialtendonlengthvalue. Stress-straincurveswereconstructedusingthemeanofeachmechanical propertyobtainedforeachgroup. 2.2.11. StatisticalAnalysis All results were presented as the mean and standard deviation for the values with a normal distribution(orinterquartilerangeandmedianforthevaluesthatdidnotadheretotheGaussian distribution). For the data with normal distribution, the analysis of variance (ANOVA) was used, Cells 2018, 7, x 8 of 23 followedbytheTukeypost-hoctestforintra-groupanalysis(inthecaseofstatisticalsignificance), or the Student’s t(LTi)e iss tthep irneiticael tdeneddon bleyngtthh vealuLe.e Svtreesns-estrTaien scut.rveFs owrered caontastruthcteadt udsinidg thne omteaand ofh eeacrhe to the Gaussian mechanical property obtained for each group. distribution,thenon-parametrictestoftheKruskal-Wallistestfollowedbythepost-hoctestofDunn 2.2.11. Statistical Analysis forintra-groupanalysis(inthecaseofstatisticalsignificance)wasusedortheUTestofMann-Whitney. All results were presented as the mean and standard deviation for the values with a normal StatisticalanalysiswasperformedinthesoftwareStatisticalPackageforSocialSciences(SPSS)version distribution (or interquartile range and median for the values that did not adhere to the Gaussian 22.0andforallthedisatrfibourtieomn). eFnort tihoe ndeatda wtieths tnsortmhael dsisitgribnuitfiionc,a thnec aenalleysvise olf αvar=ian0ce. 0(A5NaOnVdA) pwoasw useerd, ofthetestof95% followed by the Tukey post-hoc test for intra-group analysis (in the case of statistical significance), or wereconsidered. the Student’s t Test preceded by the Levene Test. For data that did not adhere to the Gaussian distribution, the non-parametric test of the Kruskal-Wallis test followed by the post-hoc test of Dunn 3. Results for intra-group analysis (in the case of statistical significance) was used or the U Test of Mann-Whitney. Statistical analysis was performed in the software Statistical Package for Social Sciences (SPSS) version 22.0 and for all the aforementioned tests the significance level α = 0.05 and 3.1. InVitro power of the test of 95% were considered. 3. Results 3.1.1. InVitroDifferentiationPotentialofADMSCandFlowCytometry 3.1. In Vitro Invitro ADMSC staining with Toluidine Blue, Alizarin Red S, and Sudan IV showed 3.1.1. In Vitro Differentiation Potential of ADMSC and Flow Cytometry differentiation of ADMSC at the 5P in chondrocytes, osteoblasts, and adipocytes, respectively In vitro ADMSC staining with Toluidine Blue, Alizarin Red S, and Sudan IV showed (Figure 1A–C). Fldoifwferecntyiattioonm oef tArDicMSaCn aatl ythse is5P (iFn icghuonrderoc1yDtes,,E o)stesohbloaswts, eadnd tahdeipopcyrteess, erenspceectivoefly CD90 (90%) and (Figure 1A–C). Flow cytometric analysis (Figure 1D,E) showed the presence of CD90 (90%) and CD105(95%)positivesurfacemarkers,andthenegativemarker,CD45(75%). CD105 (95%) positive surface markers, and the negative marker, CD45 (75%). Figure 1. In vitro adipogenic (A), condrogenic (B), and osteogenic (C) differentiation of Figure1.Invitroadipogenic(A),condrogenic(B),andosteogenic(C)differentiationofadipose-derived adipose-derived mesenchymal stem cells (ADMSC) in 5P: Observe intracellular lipid droplets mesenchymalstemstaicneedl lwsit(hA SuDdaMn IVS C(→)),i pnro5tePog:lyOcabnss setarivneed winitthr taolcueidlilnue lbalure (lip), iadndd exrtorapcellelutlasr scatlaciiunme dwithSudanIV (→),proteoglycansstasinteadi nweithd alwizairtihn rteod lSu (id). iBnaers =b Alu aend( C(cid:73): 2)0,0a μnmd; Be: 1x0t0r μamc.e (lDlu) Hlaisrtogcraamlcsi udemmonssttaraitne tehde withalizarinred x-axis fluorescence scale considered positive when the cell peak is above 101 (CD45) or 102 (CD105 S((cid:73)).Bars=AanadndC C:D2900)0. (µE)m C;onBtr:o1l 0fo0r µ–PmE .an(dD –)AHPCi s(twoitgh rvaemry slodwe mfluoorenscsetnrcae)t, ecotrhreespxon-daixngi stofl uorescencescale consideredpositivneonw-mharekend tcehlles. (cFe) lFllopwe caytkomiestray bofo AvDeM1SC0 f1or( CCDD1045,5 C)Do90r, a1n0d2 CD(C45D su1rf0ac5e maanrkderCs. D90).(E)Controlfor –PEand–APC (withverylowfluorescence),correspondingtonon-markedcells.(F)Flowcytometryof ADMSCforCD105,CD90,andCD45surfacemarkers. Cells 2018, 7, x 9 of 23 Cells2018,7,127 9of22 3.1.2. Fluorescence, Birefringence, and Contrast by Differential Interference (DIC) 3.1.2A. Fnlaulyosreiss ceunscien,gB irfelufroinregsecnecnec,ea nmdCicoronstcroasptyb ysDhoifwfeeredn tiAalDIMnteSrCfe-GreFnPc e(oDnI C5)P with fusiform fibroblast-like morphology (Figure 2A). When DIC was used, nucleoli were observed (Figure 2B). AnalysisusingfluorescencemicroscopyshowedADMSC-GFPon5Pwithfusiformfibroblast-like On AT (pH 4.3) staining, most of the nuclei were stained in blue, with the presence of granules of morphology(Figure2A).WhenDICwasused,nucleoliwereobserved(Figure2B).OnAT(pH4.3) various sizes in some regions being highly stained in blue (Figure 2C). Under polarizing microscopy, staining,mostofthenucleiwerestainedinblue,withthepresenceofgranulesofvarioussizesinsome the granules exhibited abnormal interference colors due to differences in the high packing of DNA regionsbeinghighlystainedinblue(Figure2C).Underpolarizingmicroscopy,thegranulesexhibited (Figure 2D). abnormalinterferencecolorsduetodifferencesinthehighpackingofDNA(Figure2D). FFigiguurree 22.. MMoorrpphhoollooggyyo offA DAMDMSCSCon o5nP :5(AP:) A(AD) MASDCM-GSFCP-:GFFibPr:o bFliabsrto-lbiklaesmt-loikrpe hmoloorgpyh,woliothgya, fuwsiitfho rma fushsaifpoerm.( Bs)hCapoen.t r(aBs)t CbyonDtrifafsetr ebnyt iDalifIfnetreernfetiraeln Icnete(DrfIeCre):nTceh e(DreIdC)t:o Tbhlue erbedan tdo sbhlouwe ibnagntdh eshinotwerinfegr etnhcee ineftfeercfetroefntchee enfufecclte uofs tdhuee ntuochleiguhs edruceo tnoc ehnigtrhaetrio cnosnocfenmtraatteiroinals. oNf umclaetoelriiaplr. eNseunccleeoslei epnreinsetnhcee isneiteina lino f ththee ibnliutiealb oafn tdh(e(cid:73) b)l.u(eC b)aAnDdM (SC). s(tCa)i nAeDdMwSitCh AstTai(npeHd 4w.3it)h: NAoTt e(pthHe 4p.r3e)s: eNnoceteo tfhneu pclreeissetnacien eodf innubclleuie s(t→ain),eadn idn tbhleupe r(e→se)n, caenodf tghrea npurleessenofcev aorfi ogurasnsuizleess hoifg vhalyrisotuasin seidzeisn hbilguhelyin sstoaimneedr eigni obnluse( (cid:73)in) dsoumeteo rtehgeiodniss p(on)i bdiluitey otof pthheo sdpihsaptoengibroiluitpys .o(fD p)hPooslpahriaztaet iognromupicsr. o(sDco) pPyo:lIanritzhaetidoent amileicdroimscaogpey,: nIont etthhee abnormalinterferencecolorsduetoabnormaldispersionofbirefringencebecauseofthedifferences detailed image, note the abnormal interference colors due to abnormal dispersion of birefringence inthehighpackingofDNA.Bars=50µm(A),70µm(B,C),and35µm(D).(E)NumberofADMSC because of the differences in the high packing of DNA. Bars = 50 μm (A), 70 μm (B,C), and 35 μm (D). usedforapplicationintendons,presentingabout80%viabilityaftertripsinization.(F)RT-PCRarrayof (E) Number of ADMSC used for application in tendons, presenting about 80% viability after ADMSCon5PinvitroshowingtheexpressionprofileofthegenesLox,Dcn,Timp2,Mmp2,andTgfb1. tripsinization. (F) RT-PCR array of ADMSC on 5P in vitro showing the expression profile of the gNenoeesx Lporxe,s Dsiconn, Twiamspo2b, sMermvpe2d, faonrdS Tcxg,fbT1n. mNdo, eMxpmrpe9s,sGiodnf 5w,Tasn of,basnedrvIelbd1 fgoer nSecsx., Tnmd, Mmp9, Gdf5, Tnf, and Ilb1 genes. 3.1.3. CellViability 3.1.3. FCieglul rVei1aEbislihtyo wedameanof3.7×105ADMSCusedfortendons’applicationoftheADMSCand ASCF+iGguDrFe5 1gEr sohuopws,ewd iath maemane aonf 3o.f78 ×0 1%05o AfvDiaMbSleCA uDseMd SfCor( tFeingduorens2’E a)p. plication of the ADMSC and ASC+GDF5 groups, with a mean of 80% of viable ADMSC (Figure 2E). 3.1.4. Real-TimePCRArray 3.1.4. IRneathl-eTiRmTe-P PCCRR aArrrarayya nalysis of ADMSC on 5P, the expression profile of the genes, Lox, Dcn, Timp2,Mmp2,andTgfb1wasobserved. NoexpressionwasobservedfortheScx,Tnmd,Mmp9,Gdf5, In the RT-PCR array analysis of ADMSC on 5P, the expression profile of the genes, Lox, Dcn, Tnf,andIlb1genes. TheexpressionoftheMmp8genewasnotrepresentedbecauseonlyoneculture Timp2, Mmp2, and Tgfb1 was observed. No expression was observed for the Scx, Tnmd, Mmp9, Gdf5, expressedit(Figure2F). Tnf, and Ilb1 genes. The expression of the Mmp8 gene was not represented because only one culture expressed it (Figure 2F). Cells2018,7,127 10of22 Cells 2018, 7, x 10 of 23 3.23..I2n. IVni vVoivo 3.23.1.2..I1m. Immumnuonflouflouroersecsecnecnece InItnh ethpe rperseesnetnst tsutduyd,ys, isgingnifiifciacanntta altlteerraattiioonnss iinn tthhee EECCMM ooff tthhee AAcchhilillelse stetnednodno nwwereer oebosbersveervde d aftaefrte1r4 1d4a dyasyssi nsicneceth tehep apratritailalt rtarannsseeccttiioonn,, bbootthh ffrroomm tthhee aapppplliiccaatitoionn oof fththe eAADDMMSCS Cisoislaotleadte idn itnhet he injuinrjeudrerdeg rieognioann danfrdo mfrotmhe tahpe palpicpaltiicoantioonf tohfe tAheD AMDSCMaSsCs oacsisaotceidatwedit hwGithD FG-D5.FT-5h. eTcheell cmelilg mraitgiorantiaosns ay demasosanys trdaetmedonthsteraptreeds etnhcee porfeAseDncMe SoCf -GADFPMiSnCt-eGnFdPo nins etcetniodnosn osfebctoitohnsg roofu bpos,thA SgCrouanpds, AASSCC+ GanDdF 5, onAthSeC3+rGdDaFn5d, o1n4 tthhed 3aryds aanftde r14inthju drayy(sF aigftuerre in3j)u.ry (Figure 3). Figure3.ADMSC-GFPMigrationtoTRonthe3rdand14thdaysafterinjury:Observethepresence Figure 3. ADMSC-GFP Migration to TR on the 3rd and 14th days after injury: Observe the presence ofADMSC-GFP(A,D,G,J)intheASCandASC+GDF5groups. Visiblehighernumbersofcellscan of ADMSC-GFP (A,D,G,J) in the ASC and ASC+GDF5 groups. Visible higher numbers of cells can be beobservedonthe14thdayinbothgroups. DAPI:Nucleimarking(B,E,H,K).Mergedimagesof observed on the 14th day in both groups. DAPI: Nuclei marking (B,E,H,K). Merged images of ADMSC-GFPwithnucleimarkedwithDAPI(C,F,I,L).Bar=50µm. ADMSC-GFP with nuclei marked with DAPI (C,F,I,L). Bar = 50 μm. 3.2.2. Real-TimePCRArray 3.2.2. Real-Time PCR Array ThTeh1e2 1g2e ngeenseesx pexrepsrseisosnioann aanlyaslyiss,iso, botbatinaiendedth trhoruoguhghth teheR TR-TP-CPRCRa rarraryay(F (iFgiugruere4 )4,)s, hsohwowededt htahtatt he ADthMe SACDaMpSpCli caaptipolnicautpio-nre ugpu-lraetgeudlattheed etxhpe reexspsiroesnsioofnt hoef tLhoex L,oDxc, nD,cann, danTdg fTbg1fbg1e ngeensecso cmompapraerdedto tot he Tgtrhoeu Tp ,garonudpo, fanthde oMf tmhep 2M,Tmipm2p, 2T,imanpd2, Ganddf5 Ggdefn5e gseinnerse ilna trieolnatitoont htoe tAheS CA+SGC+DGFD5Fg5r oguropu.p.
Description: