ContentslistsavailableatScienceDirect
ExperimentalCellResearch
journalhomepage:www.elsevier.com/locate/yexcr
Liraglutide,aglucagon-likepeptide-1receptoragonist,facilitatesosteogenic
proliferationanddi?erentiationinMC3T3-E1cellsthrough
phosphoinositide3-kinase(PI3K)/proteinkinaseB(AKT),extracellular
signal-relatedkinase(ERK)1/2,andcAMP/proteinkinaseA(PKA)signaling
pathwaysinvolvingβ-catenin
XuelunWu
a,b
,ShilunLi
b
,PengXue
a,b
,YukunLi
a,b,
?
a
DepartmentofEndocrinology,TheThirdHospitalofHebeiMedicalUniversity,139ZiqiangRoad,Shijiazhuang050051,HebeiProvince,PRChina
b
KeyOrthopaedicBiomechanicsLaboratoryofHebeiProvince,139ZiqiangRoad,Shijiazhuang050051,HebeiProvince,PRChina
ARTICLEINFO
Keywords:
GLP-1
Osteoblast
Proliferation
Di?erentiation
Signalingpathway
ABSTRACT
Previousstudieshaveproventhatglucagon-likepeptide-1(GLP-1)anditsreceptoragonistexertfavorable
anabolice?ectsonskeletalmetabolism.However,whetherGLP-1coulddirectlyimpactosteoblast-mediated
boneformationisstillcontroversial,andtheunderlyingmolecularmechanismremainstobeelucidated.Thusin
thispaper,weinvestigatedthee?ectsofliraglutide,aglucagon-likepeptide-1(GLP-1)receptoragonist,on
murineMC3T3-E1preosteoblastsproliferationanddi?erentiationandexploredthepotentialcellularbasis.Our
studycon?rmedthepresenceofGLP-1RinMC3T3-E1,anddemonstratedthatliraglutidepromotesosteoblasts
proliferationatanintermediateconcentration(100nM)andtime(48h),upregulatedtheexpressionofosteo-
blastogenicbiomarkersatvariousstages,andstimulatedosteoblasticmineralization.Liraglutidealsoelevated
theintracellularcAMPlevelandphosphorylationofAKT,ERKandβ-cateninsimultaneouslywithincreased
nuclearβ-catenincontentandtranscriptionalactivity.PretreatmentofcellswiththeinhibitorsLY294002,
PD98059,H89andGLP-1Randβ-cateninsiRNApartiallyblockedtheliraglutide-inducedsignalingactivation
andattenuatedthefacilitatinge?ectofliraglutideonMC3T3-E1cells.Collectively,liraglutidewascapableof
actinguponosteoblastsdirectlythroughGLP-1RbyactivatingPI3K/AKT,ERK1/2,cAMP/PKA/β-cat-Ser675
signalingtopromoteboneformationviaGLP-1R.Thus,GLP-1analoguesmaybepotentialtherapeuticstrategy
forthetreatmentofosteoporosisindiabetics.
1.Introduction
Osteoporosisandtype2diabetesmellitus(T2DM)aretwocommon
metabolicdiseases,andbotharewellknowntobeassociatedwithhigh
morbidityandeconomicburden[1–3].Althoughthesetwoprevalent
conditionshavepreviouslybeenconsideredseparatediseases,accu-
mulatingevidencehasshownthattheycouldbelinkedbysimilar
predispositionsandmolecularmechanisms[4].Thus,speci?cclassesof
anti-hyperglycemicmedicationscoulddirectlyin?uencebonemeta-
bolismunrelatedtoglucosehomeostasis.Glucagon-likepeptide-1(GLP-
1)anditsreceptoragonists(GLP-1RAs)havebeenpreviouslyreported
toplayapositiveroleinbonemetabolismgiventhatincretinshave
variouspleiotropice?ectsonliver,adiposetissue,bone,cardiovascular
andgastrointestinalsystem[5,6],drawingincreasingattention.
GLP-1,akeyincretinhormone,issecretedbyintestinalendocrineL
cellsandamelioratesglucosehomeostasisbystimulationofinsulin
secretion,suppressionofglucagonrelease,andinhibitionofgastric
emptyingandappetite,withitsbioactivitieslimitedbyrapidde-
gradationbydipeptidylpeptidase-4(DPP-4)[7].GLP-1exertsitsbio-
logicalroleprimarilybybindingtotheGLP-1receptor(GLP-1R),which
belongstotheG-proteincoupledreceptorfamily[8]highlyexpressed
onpancreaticβcellsbutalsowidelydistributedinextrapancreaticcells
includingbone[9].Inviewoftheabove-mentionedproperties,GLP-1
RAs,suchasexenatideandliraglutide,basedonincretine?ectshave
beenandarebeingdevelopedforT2DMtreatment.
Boththepreclinicalandclinicalresearchcurrentlyavailablesuggest
thatGLP-1andGLP-1RAsplayapositiveimportantroleinfacilitating
boneformationandsuppressingboneresorption,thusimprovingbone
strength[10–14].Thesee?ectshavebeendemonstratedinvivoinro-
dentsandinvitro,inGLP-1Rdeletionanimals[15–17],streptozotocin-
http://dx.doi.org/10.1016/j.yexcr.2017.09.018
Received15August2017;Receivedinrevisedform9September2017;Accepted12September2017
?
Correspondingauthorat:DepartmentofEndocrinology,TheThirdHospitalofHebeiMedicalUniversity,139ZiqiangRoad,Shijiazhuang050051,HebeiProvince,PRChina.
E-mailaddress:lyk19621225@163.com(Y.Li).
ExperimentalCellResearchxxx(xxxx)xxx–xxx
0014-4827/?2017ElsevierInc.Allrightsreserved.
Pleasecitethisarticleas:Wu,X.,ExperimentalCellResearch(2017),http://dx.doi.org/10.1016/j.yexcr.2017.09.018
inducedT2DMandfructose-inducedinsulinresistant(IR)rats[10,11],
agedovariectomized(OVX)rats[12],hyperlipidicrats[18]andos-
teoblasts[19].Theresultshavebeenverypositiveinmost[10–12,20]
butnotallstudies[21].Additionally,ameliorationofbonemetabolism
andtheanti-osteoporosise?ectsofGLP-1andGLP-1RAsinrodent
experimentshaveprimarilybeendemonstratedinosteoblastlineages
[22].
However,whetherGLP-1Rispresentinosteoblaststomediatea
directe?ectisstillunclear.Previousinvitroexperimentshavesug-
gestedthatthereisnodirecte?ectofGLP-1ineitherosteoblastsor
osteoclasts[15].Recently,Nuche-Berenguerandco-workersidenti?ed
thepresenceofaspeci?cfunctionalGLP-1RinMC3T3-E1cells[23].
Kimetal.foundthatpancreaticGLP-1RwaspresentonMLO-Y4cells
andosteocytesofratfemurs[20].AnotherstudybythePereirateam
con?rmedthattheGLP-1Rwasexpressedinbonemarrowcells,pri-
maryosteoclasts,osteoblastsandalateosteocyticcellline[24].In
summary,GLP-1Ractivationmaycontributetoosteogenesis.
Togainfurtherinsightintothepathophysiologyofosteoporosisand
thepharmacologicalpropertiesofGLP-1anditsanalogues,thedown-
streammolecularmechanismsunderlyingtheosteogenice?ectshave
beeninvestigated.GLP-1inducesmultiplesignalingpathwaysthrough
activationofthepleiotropicGLP-1R,includingclassicalcAMP/PKA
pathwayandothersecondmessengers[25–29].Amongthem,cAMP/
PKA[30,31],PI3K/Akt[32,33]andMEK/ERK1/2[34]signaling
pathwayshavebeenshowntobeinvolvedinosteoblastmediatedbone
formation.InMC3T3-EIcells,datasupportaGLP-1-inducedstimulation
uponPI3KandMAPKactivitiesinosteoblasticcells[23].Similarly,
anotherstudyfoundthatExendin-4canalsopromotetheproliferation
anddi?erentiationofMC3T3-E1osteoblastsviaMAPKs[19].Inlightof
recent?ndingsthatlinkGLP-1asadirectactivatorofWntsignaling,
GLP-1canstimulateβ-catenin(thee?ectorofWntsignaling)phos-
phorylationataC-terminalSer675position,thusleadingtothestabi-
lizationandincreasedtranslocationofβ-catenintothenucleus,derived
fromactivationofthecAMP/PKAsignalingcascade[35,36].Mengand
colleaguesrevealedthatactivationoftheGLP-1regulatedosteogenic
di?erentiationandinhibitedBMSCadipogenicdi?erentiationthrough
cAMP/PKA/β-cateninandPKA/PI3K/AKT/GSK3βsignaling[37].Thus,
wehypothesizedthatGLP-1mightinducecAMP/PKA,PI3K/AKTand
ERK1/2signalingactivationwithβ-catenininvolvementtoachieveits
promotinge?ectinosteoblast-mediatedboneformation.
Althoughalargenumberofstudieshavebeencommittedtoex-
ploringthebene?cialanabolice?ectsofGLP-1onskeletalhealth,there
hasbeenlessinterestinliraglutide,asyntheticGLP-1RAthatshares
97%homologywiththestructureofhumanGLP-1andexhibitsre-
sistancetoDPP-IVinactivationandamuchlongercirculatinghalf-life
[38].Recently,severalstudiesindicatedthatthedrugliraglutidecan
exertosteogenicandanti-resorptivee?ectswhenadministeredtodia-
beticratsornon-diabeticosteoporoticOVXrats[24,39,40];however,
theunderlyingmechanismsresponsibleforthesefavorablechangesare
lessknown.
Therefore,theaimofourstudywastoinvestigatethee?ectiveness
ofliraglutideonthefunctionofmurineMC3T3-E1preosteoblastsandto
explorethepossibleunderlyingmolecularmechanisms.Wealsoaimed
todeterminewhetherGLP-1RAcanhaveadirecte?ectonbonecellsin
vitrothroughareceptorexpressedinbonecells.
2.Materialsandmethods
2.1.Reagents
Glycerol2-phosphatedisodiumsalthydrate(β-GP),ascorbicacid,
alizarinred,3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbro-
mide(MTT),cetylpyridiniumchlorideanddimethylsulfoxide(DMSO)
werepurchasedfromSigma-Aldrich(St.Louis,MO,USA).Antibodies
targetingextracellularsignal-regulatedkinase(ERK),p-ERK,AKT,p-
AKTandβ-cateninwerepurchasedfromCellSignalingTechnology
(Danvers,MA,USA).GLP-1RantibodywaspurchasedfromAbgent(San
Diego,CA,USA,CatalogNo.AP52040).β-catenin-Ser675antibodywas
purchasedfromAbcam(Cambridge,MA,USA,CatalogNo.ab58615).
Peroxidase-conjugatedA?niPuregoatanti-rabbitIgG(H+L),GAPDH
andβ-actin(asaninternalcontrol)antibodieswerepurchasedfrom
ZSGB-BIO(Beijing,China).TheERKinhibitorPD98059,AKTinhibitor
LY294002andPKAinhibitorH89werepurchasedfromSelleckchem
(Houston,TX,USA).GLP-1R-siRNA,β-catenin-siRNAandsiRNA
scramblewerepurchasedfromGenePharmaCompany(Shanghai,
China).Allotherchemicalswereofanalyticalgradeandavailable
commercially.
2.2.Cellcultureanddi?erentiationinduction
MC3T3-E1cellswereobtainedfromAmericanTypeCulture
Collection(ATCCCRL-2594)andculturedinalphaminimalessential
medium(α-MEM)supplementedwith10%fetalbovineserum(FBS)
and1%penicillin–streptomycininahumidi?edincubatorat37°Cwith
5%CO
2
.Theculturemediawasrefreshedevery2–3days.Whenthe
cellsreachedapproximately80%con?uence,theyweresubcultured
intoanewculture?askata1:2ratioorthecompletemediumwas
replacedwithosteogenicinductionmediumcontainingα-MEM,10%
FBS,1%penicillin–streptomycin,50mg/L-ascorbicacidand10mMβ-
GPinthepresenceorabsenceofliraglutide(100nM)for28days.
2.3.Cellviabilityassay
MC3T3-E1cellswereseededatadensityof5×10
4
cells/200μL/
wellin96-wellplatesinα-MEM.Afterattachmentfor6h,themedium
wasremoved,andthecellswereincubatedwithliraglutide(0,10,100,
500,or1000nM)basedonpreviousresearch[24]for24,48or72h.
Followingtreatments,20μLoffreshlypreparedMTT(5mg/ml)was
added,andtheplateswereincubatedat37°Cforanother4htoform
crystals.Then,150μLofDMSOwasaddedtofullydissolvethecrystals,
and?nally,theabsorbanceofeachwellwasmeasuredatawavelength
of490nmusingamicroplatespectrophotometer(BioTekInstruments,
SanJose,CA,USA).
2.4.ALPactivityassay
MC3T3-E1cellsweregrownin24-wellplatesfortheALPactivity
assayafter0,7,and14daysofosteogenicdi?erentiation.Then,the
cellswereharvestedandlysedwithproteinlysisbu?er(Beyotime
Biotechnology,Shanghai,China),and30μLsampleswerequanti?ed
withanALPDetectionKit(NanjingJianchengBiotechInstitute,China)
andaspectrophotometeratawavelengthof520nmaccordingtothe
protocol.Eachvaluewasnormalizedtotheproteinconcentration.
MC3T3-E1cellswerecollectedonday7forinhibitionandinterference
experiments.
2.5.Reversetranscription-polymerasechainreaction(RT-PCR)and
quantitativereal-timeRT-PCR
ToidentifytheexpressionofGLP-1RmRNAinmouseMC3T3-E1
cells,RT-PCRwasconducted.TotalRNAwasisolatedfromcultured
MC3T3-E1cellsandmousepancreasasapositivecontrolusingTRIzol
reagent(Invitrogen,Carlsbad,CA,USA)accordingtotherecommended
protocol,andthen,cDNAwaspreparedwithaTIANScriptRTKIT
(TIANGEN,Beijing,China).PCRwasperformedinavolumeof50μL
thatincluded25μLofPCRmix(TIANGEN,Beijing,China)asfollows:
pre-denaturationat95°Cfor5minfollowedby40cyclesofthethree-
stepPCR(denaturationat95°Cfor45s,annealingat53°Cfor45s,and
extensionat72°Cfor1min).Then,5μLoftheampli?edproductswere
visualizedwithethidiumbromide(EB)after2%agarosegelelectro-
phoresis.
Toinvestigateosteogenicdi?erentiationmarkers,including
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
2
collagen-1(COL1),osteocalcin(OC),runt-relatedtranscriptionfactor-2
(RUNX2)andosteoprotegerin(OPG),alongwithGLP-1mRNAexpres-
sion,totalRNAwasextractedfromMC3T3-E1cellsusingEastep
?
Super
TotalRNAExtractionKit(Promega,Madison,WI,USA)followingthe
manufacturer''sinstructionsduringdi?erentiationstages(days0,7,14,
21and28).Equalamounts(1.0μg)oftotalRNAwerereversetran-
scribedusingaGoScript?ReverseTranscriptionSystem(Promega,
Madison,WI,USA)toobtaincDNA.PCRwasperformedinastandard
20μLreactionvolumecontaining10μLGoTaq
?
qPCRMasterMix
(Promega,Madison,WI,USA)usingaBIO-RADCFX96touchq-PCR
system.Ampli?cationparametersconsistedofaninitialdenaturation
for10minat95°Cand40cyclesoftwosteps(95°Cfor15s,and60°C
for1min).ToevaluateWntsignalingactivation,theexpressionlevelof
TCF7L2mRNAwasalsoquantitatedasdescribedabove.
QuantitativePCRassayswereconductedintriplicateforeach
sample.Therelativequanti?cationexpressionofeachgenewasde-
terminedusingtheΔΔCtmethodandwasnormalizedtoGAPDHmRNA.
AlloftheabovePCRprimers,includingthoseforGLP-1R(5′-
GGGCCAGTAGTGTGCTACAA-3′,5′-CTTCACACTCCGACAGGTCC-3′),
COL1(5′-GAGAGGTGAACAAGGTCCCG-3′,5′-AAACCTCTCTCGCCTC
TTGC-3′),OC(5′-GCACACCTAGCAGACACCAT-3′,5′-TATTGCCCTCCT
GCTTGGAC-3′),RUNX2(5′-TCGCCTCACAAACAACCACA-3′,5′-
AAAACAAAACGGAGTGAGCAAA-3′),OPG(5′-CCTTGCCCTGAC
CACTCTTAT-3′,5′-CACACACTCGGTTGTGGGT-3′),TCF7L2(5′-
CTAGGATGGCAAGGTCAGCC-3′,5′-CTTGACTGTCGGTGTGACGA-3′),
andGAPDH(5′-CCTTCCGTGTTCCTACCC-3′,5′-CCCAAGATGCCCTTG
AGT-3′),weredesignedusingtheNCBI-BLASTsystem.Forinhibition
andinterferenceexperiments,cellswerepretreatedwithinhibitoror
siRNAandharvestedonday7afterundergoingdi?erentiation.
2.6.Alizarinredstaining(AR-S)
Toexaminethematrixmineralizationofosteoblasts,MC3T3-E1
cellswereplatedin6-wellplatesataconcentrationof1×10
5
perwell
inthedi?erentiationmediumfor21days,andAlizarinredstainingwas
performed.Cellswere?xedwith4%paraformaldehydeatroomtem-
peraturefor10min.Then,thewellswererinsedthreetimeswithdis-
tilledwater,andcellsweredyedwith1%AR(Sigma,USA)at37°Cfor
30min.Finally,cellswerewashedthoroughlywithdistilledwaterand
air-driedforphotography.Afterthat,ARwasprecipitatedwith10%
cetylpyridiniumchloride(Sigma,USA)for30minatroomtemperature.
ThecalciumconcentrationsweredeterminedbydeterminingtheOD
valueat562nmwithamicroplatespectrophotometer.Alldatawere
normalizedtothetotalproteincontent.
2.7.Westernblotanalyses
TomeasurethephosphorylationlevelofthesignalingproteinsAKT,
ERK1/2,andβ-catenin,MC3T3-E1cellswereculturedin25cm
2
?asks
forproteinextraction.Aftercellsgrewtocon?uence,thecomplete
mediumwassubstitutedwithserum-freemedium,andcellswerecul-
turedforanother12h.Then,thecellswereexposedtoliraglutide
(100nM)for0–60mininfreshmedium.Cellswerewashedwithcold
PBSthreetimes,andtotalproteinwasextractedattheindicatedtime
pointusingRIPAlysisbu?erwithphosphataseinhibitors(Beyotime
Biotechnology,Shanghai,China).ABCAassay(Beyotime
Biotechnology,Shanghai,China)wasusedtodetermineproteincon-
centration,andequalamountsofproteinwereseparatedby5%SDS-
PAGEandtransferredtoPVDFmembranesusingasemidrytransfer
Fig.1.ThepresenceofGLP-1RinMC3T3-E1osteoblasts.RepresentativeRT-PCR(A),Westernblot(B),andconfocalmicroscopy(C)indicatingthepresenceoftheGLP-1receptorin
MC3T3-E1cellsandmousepancreas(aspositivecontrol).(C)Upperpanel:400×magni?cation;lowerpanel:200×magni?cation.GLP-1RmRNA(D)andproteinexpression(E)in
di?erentiatingMC3T3-E1cellsinducedbyosteogenicmediumfor0,7,14,21,and28days.TheresultswerenormalizedtoGAPDHasaninternalcontrol.Experimentswereperformedat
leastthreetimes,andthedataareexpressedasthemean±SD.P<0.05andP<0.01vs.controlcells.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
3
apparatus(ATTO).Themembranewasblockedwith5%non-fatmilkin
TBSTfor1hatroomtemperaturebeforeincubationwithspeci?cpri-
maryantibodiesovernightat4°Candsubsequentlywasincubatedwith
HRP-labeledsecondaryantibody(1:3000dilution)for90minatroom
temperature.Lastly,blotswereprocessedusinganECLkit(SantaCruz
Biotechnology),exposedto?lm,andthendetectedwithTanon1600
ImagingSystem(Shanghai,China).Tocontrolforimmunedetection
e?ciency,p-AKT,totalAKT,p-ERK1/2,totalERK1/2,p-β-catenin-
Ser675,totalβ-cateninwerediluted1:1000withGAPDHasaninternal
control.
NuclearandcytoplasmicfractionsofMC3T3-E1cellswereprepared
fordetectionoftheβ-cateninproteinlevelusingaNuclearProtein
ExtractionKitprovidedbySolarbio(Beijing,China).Thehousekeeping
proteinsβ-actinandHistone3wereusedasinternalcontrolsfornor-
malization.
Westernblotanalysisservedtocon?rmthepresenceofGLP-1R
proteininMC3T3-E1cells,withmousepancreasasapositivecontrol,
andtodetectalterationsinGLP-1Rexpressionduringtheosteogenic
di?erentiationperiod.Theanti-GLP-1Rantibodyusedwasrabbit
polyclonal(AP52040,Abgent,53kDa).
2.8.MeasurementofintracellularcyclicAMP
MC3T3-E1cellswereculturedin25cm
2
?asksformeasurementof
cAMP.Atharvesttime,thecellsupernatantswereobtainedafterul-
trasonicationandrepeatedfreezingandthawing.CytoplasmiccAMP
contentwasmeasuredinlysatesusingaCyclicAMPELISAAssaykit
(NanjingJianchengBiotechInstitute,China)accordingtothe
manufacturer''sinstructions.Todetectshort-termmodi?cationofthe
cAMPlevelfor0–60min,thecellswereplacedinserum-freemedium
for12h.
2.9.Confocalmicroscopy
MC3T3-E1cells(10
5
cells/well)wereculturedonaglasscoverslip
in6-wellplatesfor24h.Then,thecellswere?xedin4%paraf-
ormaldehydefor30min.Subsequently,the?xedcellswerewashed
twicewith0.1%BSAinPBSfor10min,permeabilizedwith0.3%
TritonX-100inPBS,washedagainwithPBSfor10min,blockedfor1h
inPBScontaining10%bovineserumalbumin(BSA),andincubated
overnightat4°Cwithanti-GLP-1Rantibodydiluted1:200inPBS
containing1%BSA.Thenextday,thecoverslipswerewashedtwicein
PBSfor10minandincubatedfor1hatroomtemperatureawayfrom
lightwithAlexaFluor
?
594-conjugatedgoatanti-rabbitIgGH&L
(MolecularProbes,Eugene,OR,USA)diluted1:200inPBScontaining
1%BSA.Thecoverslipsweremountedwithglycerolmediumcon-
taining4′,6-diamidino-2-phenylindole(DAPI,Sigma-Aldrich)for
10minatroomtemperature.Confocalimageswereobservedundera
FV1000FLUOVIEWconfocallaserscanningmicroscope(Olympus,
Tokyo,Japan).
2.10.Inhibitionstudy
MC3T3-E1cellswerepretreatedwiththepharmacologicalPI3K
inhibitorLY294002(20μM)theERKinhibitorPD98059(20μM),orthe
PKAinhibitorH89(20μM)2hbeforeliraglutidetreatment.Follow-up
Fig.2.E?ectsofliraglutideoncellviabilityanddi?erentiationinMC3T3-E1cells.(A)CellviabilityafterMC3T3-E1cellswereculturedfor24,48,or72hwithliraglutideattheindicated
concentrations.(B)ALPactivityinMC3T3-E1cellsafterliraglutidetreatmentatdi?erentdosesondays0,7,and14.(C–F)ThemRNAexpressionofboneformationmarkersCol-1,OC,
Runx2,andOPGinMC3T3-E1cellsculturedfor0,7,14,21and28daysinosteogenesisinductionmediumvia?uorogenicquantitativePCR.(G,H)MineralizedmatrixinMC3T3-E1cells
anditsquanti?cation.Alizarinredstainingwasperformedonday21;mineralizednodulesapparentbyvisualinspection(G)andquanti?cationbycetylpyridiniumchloridedissolution
(H)arepresented.(n=3).P<0.05andP<0.01comparedwiththecontrolgroup.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
4
experimentswereperformedbasedontheadditionofaboveinhibitors.
Forcalciumdepositionlastingfor21days,alowerconcentration
(10μM)ofvariousinhibitorswasappliedforAlizarinRedstaining.
2.11.GLP-1Randβ-cateninsilencingviaRNAinterference
TwosmallinterferingRNAstargetingGLP-1R(5′-
CCGGACCUUUGAUGACUAUTT-3′,5′-AUAGUCAUCAAAGGUCCGGTT-
3′)andβ-catenin(5′-GGGUUCCGAUGAUAUAAAUTT-3′,5′
-AUUUAUAUCAUCGGAACCCTT-3′)andascrambledsiRNA(sense5′-
UUCUCCGAACGUGUCACGUTT-3′,antisense5′-ACGUGACACG
UUCGGAGAATT-3′)weresynthesizedbyGenePharmaBiotechnology
(Shanghai,China).MC3T3-E1cellswereplatedinaT25?askand
culturedfor24hwithoutantibioticsandthentransfectedwithsiRNA
(50pmol/well)usingEntransterTM-R4000(EngreenBiosystemCo,
Ltd.)accordingtothemanufacturer''sinstructions.Cellswerecultured
for48h.ThesiRNAsilencinge?ciencywasdeterminedbyprotein
analysisforfurtherexperiments.Becausethematrixmineralizationof
MC3T3-E1cellswasevaluatedbyAlizarinredstainingonday21
duringdi?erentiation,cellsenduredthreeroundsoftransfectionprior
toperformingexperiments[41,42].
2.12.Statisticalanalysis
Dataareexpressedasthemean±standarddeviation(SD)andwere
analyzedusingSPSSsoftware,version18.0.Allexperimentswerere-
peatedthreetimes.StatisticalanalysiswasperformedusingStudent''st-
testandone-wayANOVAfollowedbyposthocmethods.P<0.05was
consideredstatisticallysigni?cant.
3.Results
3.1.ExpressionofGLP-1RinMC3T3-E1cells
TodeterminethedirectroleofGLP-1inMC3T3-E1cells,we?rst
con?rmedthepresenceofGLP-1RinMC3T3-E1cellsusingRT-PCR.
Theresultsrevealeda111-bpbandspeci?ctoGLP-1Rwithmouse
pancreastissueasapositivecontrol(Fig.1A).Inaddition,westernblot
analysis(Fig.1B)andconfocalmicroscopy(Fig.1C)furthervalidated
theexpressionofGLP-1RproteinonthesurfaceofMC3T3-E1cellsas
expected.Moreover,weinvestigatedthee?ectsofliraglutideon
MC3T3-E1cellsduringdi?erentiationstages,andwefoundthatGLP-
1RmRNAandproteinexpressionwerebothincreasedduringosteo-
genicinductionfor28days,withthehighestvalueatday21.The
observede?ectswerenottime-dependent(Fig.1DandE).
3.2.Liraglutidepromotedproliferationanddi?erentiationinMC3T3-E1
cells
MC3T3-E1cellsweretreatedwithliraglutide(0,10,100,500,or
Fig.3.E?ectsofliraglutideonAKT,ERK,andβ-cateninphosphorylationandontheintracellularcAMPlevelinMC3T3-E1cells.(A–C)AKT,ERK,andβ-cateninphosphorylationin
MC3T3-E1cellstreatedwithliraglutide.Relativeratiosofphosphorylated/totalproteinarelistedonthepanels.(D–F)E?ectsofspeci?cselectiveinhibitors(20μMLY294002,20μM
PD98059,or20μMH89)onliraglutide-inducedupregulationofAKT,ERK,andβ-cateninphosphorylation.(G,H)TheintracellularcAMPlevelinMC3T3-E1cellswhensubjectedto
di?erentconcentrationsofliraglutide(G)ortreatedwith100nMliraglutidefor60minatdi?erenttimepoints(H).ThedataarepresentedasthemeanandSD(n=3/group).P<0.05
andP<0.01comparedwiththecontrolgroup.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
5
1000nM)todetermineitse?ectsontheproliferationanddi?erentia-
tionofMC3T3-E1cells.Preliminarily,MTTassaysshowedthat100nM
liraglutideat48hhadthehigheste?ectoncellviabilitycomparedto
thecontrolgroup,peakingatanintermediateconcentrationandtime
(Fig.2A).Inparallelwiththisanalysis,ALPactivityexhibitedthelar-
gestincreaseat100nMafterexposuretoliraglutide,althoughthetrend
wastime-dependentunderinduction(Fig.2B).Basedontheabove
results,100nMliraglutidewasselectedastheoptimalconcentration
forfurtherstudies.Subsequently,quantitativereal-timeRT-PCRwas
usedtoexaminethemRNAexpressionofbonemarkers,includingCol-
1,OC,RUNX2andOPG(Fig.2C–F),thatarecorrelatedwithosteo-
blasticdi?erentiation.Inbrief,ourdatashowedthatCol-1mRNAex-
pressionreachedahighlevelintheearlyperiodofdi?erentiation,
RUNX2mRNAexpressionwasslightlyincreased,andtheOPGmRNA
levelroseduringtheearlyandmiddleperiod,allpeakingonday14
(P<0.05).However,OCdisplayedasustainedincreaseinmRNAex-
pressionalmostthroughouttheentiredi?erentiationprocesscontrolled
tobaseline.ThesedataindicatedthatMC3T3-E1cellswereundergoing
acharacteristicosteogenicdi?erentiationprocessformaturation.After
stimulationwithliraglutide,thesetendencieswerenoticeablyen-
hanced.ThemRNAlevelsofCol-1andOPGweresigni?cantlyelevated,
withthehighestincreaseonday14(P<0.05),andOCandRUNX2
mRNAexpressionwasmostvisiblyaugmentedonday21(P<0.05).In
addition,visualinspectionusingAlizarinredstaining(Fig.2G)and
quanti?cationofcalciumdepositionviacetylpyridiniumchloridedis-
solution(Fig.2H)showedanincreaseinmineralizationforcellstreated
withliraglutidecomparedwiththatofcellsculturedindi?erentiation
mediaalone(P<0.05).
3.3.PI3K/AKT,ERK1/2,andcAMP/PKAsignalingmediatedtheanabolic
e?ectofliraglutideinMC3T3-E1cellsviaGLP-1Randinvolvingβ-catenin
ToelucidatehowGLP-1interactedwiththedownstreamsignaling
activatedbyGLP-1R,westernblotting,ELISAandsiRNAassayswere
performed.Ourresultsshowedthatliraglutidesimultaneouslyacti-
vatedPI3K/AKT(Fig.3A),ERK1/2(Fig.3B),andcAMP/PKA(Fig.3H)
signalingpathwaysinsidetheshorttimeof0–60min.Westernblotting
indicatedthatliraglutideupregulatedAKTandERK1/2phosphoryla-
tionwiththehighestimpactat30minforERK1/2butat45minfor
AKT,andthephosphorylationofERKandAKTbyliraglutidewaso?set
bypretreatmentwiththespeci?cinhibitorsLY294002(AKTinhibitor,
20μM)(Fig.3D)andPD98059(ERK1/2inhibitor,20μM)(Fig.3E),
respectively.ForcAMP,theELISArevealedthattheoptimumcon-
centrationofliraglutidetoincreasecAMPcontentwas100nMwhen
MC3T3-E1cellsweresubjectedtodi?erentconcentrationsofliraglutide
(Fig.3G).Furthermore,similartoERK1/2,cAMPinductionbylir-
aglutidewasmaximizedat30min(Fig.3H).Inaddition,ourwestern
blotresultsalsoshowedthatliraglutidetreatmentaugmentedthe
phosphorylationofβ-cateninataC-terminalSer675positionatthe
sametimeascAMPactivation,peakingat30min(Fig.3C),andpre-
treatmentwith20μMH89(PKAinhibitor)attenuatedthephosphor-
ylationofβ-cateninatC-terminalSer675inducedbyliraglutide
(Fig.3F),indicatingthatcAMP/PKA/β-catenin-Ser675signalingcas-
cadewasinvolvedinthisstimulatorye?ect.
Then,toinvestigatewhethercAMP/PKA,PI3K/AKT,andERK1/2
activationwaslinkedwithGLP1-R,siRNAwasusedtosuppressGLP1-R
(Fig.4A),andthelevelofcAMP/PKA/β-catenin-Ser675,PI3K/AKT,
andERK1/2activationwassigni?cantlyblocked(Fig.4C–F),which
Fig.4.E?ectsofGLP-1Randβ-cateninsmallinterferingRNAtransfectiononliraglutide-inducedupregulationofAKT,ERK,andβ-cateninphosphorylationandoncAMPcontent.(A,B)
GLP-1Randβ-cateninproteinexpressionsuppressedbysiRNA.(C)ThecAMPcontentaftertreatmentwithscrambledGLP-1RsiRNAandGLP-1R-siRNA.(D–F)AKT,ERK,andβ-catenin
phosphorylationafterSiGLP-1Rinterference.Barsrepresentthemean±SD.P<0.05,P<0.01.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
6
suggeststhatliraglutidemediatedtheactivationofcAMP/PKA,PI3K/
AKTandERK1/2throughGLP-1R.
Thewesternblottingresultsalsoshowedthatliraglutidetreatment
ledtoincreasedaccumulationofβ-catenininthecytoplasmanden-
hancedtranslocationofβ-cateninintothenucleus(Fig.5A–C).Nuclear
β-cateninassociateswithTCF7L2toformatranscriptionallyproductive
complexthattargetsWntsignalinge?ectors.Hence,ourPCRresults
showedupregulationofTCF7L2mRNAexpressioncomparedwiththe
levelinthecontrolundertheimpactofliraglutide(Fig.5G).Subse-
quently,pretreatmentwithH89(PKAinhibitor)andinterventionwith
siGLP-1Rdownregulatedthenuclearlocalizationofβ-cateninandthe
inductionofTCF7L2inducedbyliraglutide(Fig.5A–H),providingdi-
rectevidenceoftheinvolvementofcAMP/PKA/β-catenininthecom-
plexprocessinducedbyliraglutidethroughGLP-1R.Inaddition,the
accumulationofβ-catenininthecytoplasmandtranslocationofβ-ca-
tenintothenucleusaswellasthetranscriptionofTCF7L2werealso
partlyinhibitedbyLY294002andPD98059,indicatingthatthePI3K/
AKTandERK1/2pathwaysmaycross-talkwithβ-catenintoactivate
Wntsignaling.
3.4.Thee?ectsofliraglutideinpromotingtheproliferationand
di?erentiationofMC3T3-E1cellswereabolishedbyspeci?cinhibitorsand
siRNAs
Speci?cinhibitorsandsiRNAswereutilizedtoinvestigatetherole
ofcAMP/PKA,PI3K/AKTandERK1/2intheproliferation,di?erentia-
tionandfunctionofMC3T3-E1osteoblasticcells.InMTTassays,
treatmentwiththeinhibitorPD98059causedthemostobviousat-
tenuatione?ect;LY294002showedtheleastbutstillsigni?cantde-
crease,andH89exhibitedalevelofinhibitionsomewhereinbetween
them(Fig.6A).ThemRNAexpressionofosteoblasticdi?erentiation
markersafterinhibitortreatmentwasmuchlowerthanthatinthe
blanks(Fig.6E–H).PD98059andLY294002preferentiallyinhibitedthe
expressionofRunx2(Fig.6G),leavingH89withthelowestlevelofOC
expression(Fig.6F),buttheyallrestrainedmatrixcalci?cationbased
onamineralizationstainingassay(Fig.7AandC).Lastly,suppressing
GLP1-RwithsiRNAdistinctlyattenuatedthefacilitatione?ectoflir-
aglutideoncellviability(Fig.6B),theexpressionofdi?erentiation
genes(Fig.6I–L)andthemineralizationcapacity(Fig.7BandD)of
MC3T3-E1osteoblasticcells.Tofurthercon?rmthekeyroleofβ-ca-
tenininthemodulatinge?ectofGLP-1RonMC3T3-E1cellprolifera-
tionanddi?erentiation,wealsoexploredthee?ectofsilencingβ-ca-
tenin(Fig.4G).Ourresultsshowedthatthee?ectofliraglutideoncell
viability(Fig.6B),ALPactivity(Fig.6D),theexpressionofosteoblast
di?erentiationmaturationmarkers(Fig.6I–L)andmineralization
(Fig.7BandD)wereabolishedbysiβ-catenin.Insummary,liraglutide
inducedcAMP/PKA,PI3K/AKT,andERK1/2signalingpathwayacti-
vationviaGLP-1Rwithβ-catenininvolved,whichplaysapivotalrolein
theproliferation,di?erentiationandmineralizationofMC3T3-E1os-
teoblasticcells(Fig.8).
Fig.5.Involvementofβ-catenininliraglutide-stimulatedPI3K/AKT,ERK1/2andcAMP/PKAsignalinginMC3T3-E1osteoblasticcells.(A–C)Cytoplasmicβ-catenin(C-β-cat)andnuclear
β-catenin(N-β-cat)proteinlevelsaftertreatmentwithLY294002,PD98059,orH89.(D–F)C-β-catandN-β-catproteinlevelsaftertreatmentwithsiGLP1-Rorsiβ-catenin.Western
blottingwasperformed,andtheresultswerenormalizedtoβ-actinorhistoneH3.1.(G–H)Real-timePCRforTCF7L2mRNAinMC3T3-E1cellssubjectedtodi?erenttreatments.Thedata
arepresentedasthemean±SD(n=3/group).P<0.05andP<0.01vsvehicletreatment.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
7
4.Discussion
GLP-1isexpectedtoexertpotentialbene?ciale?ectsonbonehealth
beyondglycemia,possiblyduetodirectboneanabolicactivity[43].
However,thisputativeroleofGLP-1asamodulatorofboneturnover
andtheunderlyingmechanismsremainuncertain.Moreover,thereis
stillwidespreadconfusionregardingthepresenceofGLP-1Rinbone.In
ourstudy,werecon?rmedthepresenceofGLP-1RinMC3T3-E1os-
teoblasts,corroboratingmostofthepreviousresults[19,44].However,
theinabilityofNuche-Berenguerandcolleaguestodetectpancreatic
GLP-1Rinthesamecelllinemaybedueinparttodi?erencesinthesize
oftheampli?cationtemplate[45].Inaddition,wealsoobservedthat
GLP-1RexpressionwasincreasedinMC3T3-E1cellsduringinduced
osteogenicdi?erentiation,whichwasinlinewithobservationsinadi-
pose-derivedstemcells(ADSCs)[46],despitedisparitiesintheex-
pressiontendency,perhapsasaresultofthedi?erentcellsource,which
stilldidnotpreventusfromdrawingtheconclusionthatGLP-1Racti-
vationinMC3T3-E1osteoblastsmaycontributetoosteogenesis.
GLP-1exertsnumerouspleiotropice?ectsinvarioustissues,reg-
ulatingenergyabsorptionanddisposal,aswellascellproliferationand
survival[47,48].Forosteoblasts,GLP-1causedsigni?cantincreasesin
cellviabilityinthehumanMG-63cellline[45],andexendin-4in-
creasedcellproliferationatcertainconcentrations[19],asevidenced
byMTTassay.Inagreementwiththeaboveresults,wefoundthatlir-
aglutidepromotedMC3T3-E1osteoblastproliferationatdi?erent
concentrationsandtimes,bothwithapeake?ectatanintermediate
point,butnotinadose-andtime-dependentmanner.Theexplanation
forthismaybethatlowerconcentrationsandshortertimesarenot
su?cienttoinducemeaningfulalterationsinviability,whileGLP-1R
maybesubjectedtosomedesensitizationorrapidattenuationofre-
ceptorsensitivityafterlongorhighexposuretoagonists[45].
ALPactivityandCol1,OC,RUNX2,andOPGmRNAexpressionwere
usedasbonemarkerstoevaluatetheabilityofMC3T3-E1osteoblaststo
di?erentiateafterinduction.ALPandCol1havebeendemonstratedto
beearlymarkers,whileOCNappearslate,concomitantwithminer-
alization[49].Runx2isamastertranscriptionalregulatorofosteoblast
di?erentiation,asevidencedbyupregulatedosteoblast-relatedgenes
suchasCol1,ALP,BSPandOCN[50].OPGmRNAexpressionhasbeen
foundtobeincreasedattheonsetofmineralizationandtoremainhigh
throughoutprimaryosteoblasticdi?erentiation[51].Here,ourdata
showedthatliraglutidesigni?cantlyupregulatedtheexpressionofos-
teoblastogenicbiomarkersatvariousstageswhilestimulatingosteo-
blasticmineralization,despiteadivergenceinbasalexpressionandthe
peake?ectspointcomparedtothosereportedbyFengetal.[19]and
Nuche-Berengueretal.[23],possiblyonaccountofcomparablemed-
ication,dosesandexposuretimesalongwithGLP-1Rexpressionin
MC3T3-E1cellsduringinducedosteogenicdi?erentiation.Insummary,
liraglutideconferredadirectpositiveactiononMC3T3-E1pre-
osteoblastdi?erentiationandmineralizationtowardboneformation.
Togainabetterunderstandingofthemechanismsunderlyingthe
Fig.6.E?ectsofGLP-1R-mediatedPI3K/AKT,ERK1/2andcAMP/PKAsignalingonMC3T3-E1cellproliferationanddi?erentiationstimulatedbyliraglutide.(A,B)E?ectofLY294002,
PD98059,orH89andsiGLP1-Rorsiβ-cateninonliraglutide-inducedincreaseofMC3T3-E1cellproliferationevaluatedbyMTTassays.(C-L)E?ectofLY294002,PD98059,orH89and
siGLP1-Rorsiβ-cateninontheliraglutide-inducedincreaseinMC3T3-E1celldi?erentiation.Cellswereinducedtodi?erentiateinosteogenicmediumwithspeci?cinhibitorsandsiRNAs
for7days,andthen,ALPactivitywasdetected(C,D),andquantitativePCRwasperformed(E-L).Thedataareexpressedasthemean±SDofthreeindependentexperiments.P<0.05
andP<0.01vsthecontrolgroup.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
8
functionofliraglutideinregulatingtheproliferationanddi?erentiation
ofMC3T3-E1preosteoblasts,weconcentratedonseveralintracellular
signalingpathwayssurroundingGLP-1.Basedonaseriesofstudies
[19,20,37]conductedforinsightintothemechanismsofGLP-1-induced
boneformationthroughGLP-1R,thepresentstudyshowedthatlir-
aglutideactivatedcAMP/PKA,PI3K/AKT,andERK1/2andphos-
phorylatedβ-cateninattheSer675positioninanextremelybrief
period,allwithamaximumenhancemente?ectat30min,withthe
Fig.7.Matrixmineralizationandquanti?cationin
MC3T3-E1cells.Cellswereculturedindi?erentia-
tionmediumandstainedwithalizarinredonday21
subjectedtoAKT,ERK,andcAMPselectivein-
hibitors(A,C)orGLP-1Randβ-cateninsiRNA(B,D).
Upper?guresshowimagesofAlizarinredstaining,
andthelower?gurespresentthequanti?cation
basedoncetylpyridiniumchloridedissolution(n=
3/group).P<0.05andP<0.01vscontrolcells.
Fig.8.Anillustrationofthesignalingpathwaysinvolvedinthe
roleofGLP-1inpromotingMC3T3-E1osteoblastproliferationand
di?erentiation.GLP-1ortheGLP-1agonist,liraglutideactivated
PI3K/AKT,ERK1/2andcAMP/PKAsignalingviaGLP-1R.Then,
activatedPKAinducedphosphorylationattheSer675site,thus
leadingtoβ-cateninnucleartranslocationandbindingtotran-
scriptionfactorTCF7L2totargetosteogenicWnte?ectors.
ActivationofPI3K/AKTandERK1/2alsopotentiatednuclear
accumulationofβ-catenin,with?nalboneanabolice?ects
achieved.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
9
exceptionofAKT,whichpeakedat45minaftertheonsetofstimula-
tion.GLP-1hasbeenreportedtoenhancebetacellproliferationvia
transactivationoftheEGFreceptor(EGFR),whichregulatedthePI3K
andAKTactivationrequiredforbasalWntsignaling[52].However,
whetherthissignalingoranothercontributedtothenon-synchronous
activationinsignalingcascadesevokedbyliraglutidewithadelayfor
AKTisnotyetclear.Subsequently,PI3K/AKT,ERK1/2,andcAMP/
PKA/Ser675pathwayswerecon?rmedtobeactivatedthroughGLP-1R
byliraglutidestimulationandtohavesomecross-talkwithWntsig-
nalingactivation,enablingustoestimatethattheWntsignaling
pathwayisthe?nalpathwayforliraglutide-mediatedosteoblastfunc-
tionbyGLP-1R,whichboreoutourearliersuspicion.However,data
fromNuche-Berengueretal.[23]failedtoverifythatGLP-1a?ected
thecanonicalWntpathway.Weinferthatthedi?erenceinourresultsis
probablytraceableinparttodi?erentpharmacologicalproperties,
medicationtiminganddosingregimen,whichrequiresfurtherin-
vestigation.Inaddition,inhibitionofcAMP/PKAdidnota?ectAKTor
ERKactivationunexpectedly(datanotshown),butNuche-Berenguer
[23]foundthatGLP-1inducedanincreaseinPI3KandMAPKactivities
byGPI/IPGs,andassuch,weassumedthatwhetherGLP-1mightin-
teractwiththedownstreampathwaythroughbothGLP-1RandGPI/
IPG-coupledreceptorsroutestoengenderdi?erentiale?ectsdependsto
greatextentonsignalingbiasandhighselectivity[53],whichneedsto
becon?rmedbyfurtherstudy.
Theuseofspeci?cinhibitorsandsiRNAssupportedtheindis-
pensablerolethatliraglutideplaysintheproliferation,di?erentiation
andfunctionofMC3T3-E1osteoblasticcells.Inproliferationexperi-
ments,theinhibitorPD98059displayedthemostobviousinhibitory
e?ect,treatmentwithLY294002showedamildbutsigni?cantde-
crease,andtheH89e?ectwasintermediatebetweentheothertwo,
suggestingthatthemitogenice?ectofliraglutiderequiredthecAMP/
PKAandERK1/2pathwaysmorethanthePI3K/AKTpathway.
Meanwhile,allthreeinhibitorsblockedtheexpressionofdi?erentiation
markers,asexpected.LY294002,PD98059,andH89exhibitedsimilar
inhibitorye?ectsonALPandCOL1,indicatingthatPI3K/AKT,ERK1/2,
andcAMP/PKA/Ser675signalingworktogetherintheearlystagesof
celldi?erentiation.LY294002andPD98059imposedtheirmaximum
eliminationimpactonRunx2,consistentwithpreviousstudiesthat
haveshownthatAKTandERKpotentiatedboneformationbymod-
ulatingtheactivityofRunx2[54],whileH89presentedasharpin-
hibitorye?ectonOC,suggestingapreferentiale?ectofcAMP/PKAon
thelaterstagesofosteoblasticdi?erentiation[23].Theirunanimous
in?uenceonmatrixcalci?cationclearlydemonstratedtheircommon
contributionstoosteoblastfunctionaftermaturation.Anyhow,these
downstreamsignalscomeintoplayataspeci?cstageofdi?erentiation
partiallyunresponsivetoGLP-1Rexpression(asindicatedbyFig.1D
andE),onceagainindicatedthepossibleimportanceofsignalingbiasin
thiscomplexsignalnetwork.
Additionally,astheoriginandtermination,GLP-1Randβ-catenin
interferenceviasiRNAtheoreticallyshouldcauseagreaterlossforthe
enhancementthatliraglutidegivesrisetoonMC3T3-E1osteoblastic
cells.However,theydisplaysimilare?ectsasthegroupusingeach
inhibitoronly,raisingthepossibilitythattherearestillothersignaling
pathwaysinvolvedinthiscomplexprocess.Inbrief,theimpactoflir-
aglutidedescribedabovecanbepartlyorcompletelypreventedbyin-
hibitorsandsiRNAsinMC3T3-E1cells,echoingtheconclusionswe
previouslyreached.
Inconclusion,ourresultsclearlyshowthattheGLP-1RAliraglutide
engendereddirectanabolice?ectsonmurineMC3T3-E1preosteoblasts
throughmultiplepathwaysinducedviaGLP-1Rinvitro.However,
whethertheresultsofthesepreclinicalstudiescanbeextrapolatedfrom
rodentstohumansinvivoisnotclear.Thus,furtherexperimentaland
long-termclinicalstudieswillberequiredtoascertainthee?cacyof
GLP-RAtherapiesinbonediseasesandtoidentifytheunderlyingpa-
thophysiologicalmechanisms,whichisparticularlyimportantnotonly
forabetterunderstandingoftheinterplaybetweenincretinsandbone
butalsoforthepotentialtherapeuticimplicationsofincretinsinske-
letalfragilityassociatedwithdiabetes.
Acknowledgements
ThisworkwassupportedbyNaturalSciencefoundationofHebei
Province(grantNo:H2016206243),HebeiProvincial2016govern-
ment-fundedclinicaltalentstrainingandbasicresearchprojects(grant
No:361005),andHebeiCollegeNaturalSciencetoptalentproject
(grantNo:BJ2016037)fromChina.
Authorcontributions
Studyconceptanddesign,acquisitionofdata,analysisandinter-
pretationofdata,draftingofthemanuscript:XuelunWu,ShilunLi;
criticalrevisionofthemanuscriptforimportantintellectualcontent:
PengXue,YukunLi;statisticalanalysis:PengXue;obtainedfunding:
PengXue,YukunLi;administrative,technical,ormaterialsupport:
XuelunWu,ShilunLi;studysupervision:PengXue,YukunLi.
Con?ictofinterest
Theauthorshavenocon?ictsofinterests.
References
[1]G.Roglic,N.Unwin,P.H.Bennett,C.Mathers,J.Tuomilehto,S.Nag,V.Connolly,
H.King,Theburdenofmortalityattributabletodiabetes:realisticestimatesforthe
year2000,DiabetesCare28(2005)2130–2135.
[2]D.W.Dempster,Osteoporosisandtheburdenofosteoporosis-relatedfractures,Am.
J.Manag.Care17(Suppl6)(2011)S164–S169.
[3]F.Ponti,S.Guerri,C.Sassi,G.Battista,G.Guglielmi,A.Bazzocchi,Imagingof
diabeticbone,Endocrine(2017).
[4]B.Lecka-Czernik,Safetyofanti-diabetictherapiesonbone,Clin.Rev.BoneMiner.
Metab.11(2013)49–58.
[5]C.H.McIntosh,S.Widenmaier,S.J.Kim,Pleiotropicactionsoftheincretinhor-
mones,Vitam.Horm.84(2010)21–79.
[6]E.Ceccarelli,E.G.Guarino,D.Merlotti,A.Patti,L.Gennari,R.Nuti,F.Dotta,
Beyondglycemiccontrolindiabetesmellitus:e?ectsofincretin-basedtherapieson
bonemetabolism,Front.Endocrinol.4(2013)73.
[7]L.L.Baggio,D.J.Drucker,Biologyofincretins:GLP-1andGIP,Gastroenterology
132(2007)2131–2157.
[8]J.S.Dillon,Y.Tanizawa,M.B.Wheeler,X.H.Leng,B.B.Ligon,D.U.Rabin,H.Yoo-
Warren,M.A.Permutt,A.E.Boyd3rd,Cloningandfunctionalexpressionofthe
humanglucagon-likepeptide-1(GLP-1)receptor,Endocrinology133(1993)
1907–1910.
[9]J.E.Campbell,D.J.Drucker,Pharmacology,physiology,andmechanismsofincretin
hormoneaction,CellMetab.17(2013)819–837.
[10]B.Nuche-Berenguer,P.Moreno,P.Esbrit,S.Dapia,J.R.Caeiro,J.Cancelas,
J.J.Haro-Mora,M.L.Villanueva-Penacarrillo,E?ectofGLP-1treatmentonbone
turnoverinnormal,type2diabetic,andinsulin-resistantstates,Calcif.TissueInt.
84(2009)453–461.
[11]B.Nuche-Berenguer,P.Moreno,S.Portal-Nunez,S.Dapia,P.Esbrit,
M.L.Villanueva-Penacarrillo,Exendin-4exertsosteogenicactionsininsulin-re-
sistantandtype2diabeticstates,Regul.Pept.159(2010)61–66.
[12]X.Ma,J.Meng,M.Jia,L.Bi,Y.Zhou,Y.Wang,J.Hu,G.He,X.Luo,Exendin-4,a
glucagon-likepeptide-1receptoragonist,preventsosteopeniabypromotingbone
formationandsuppressingboneresorptioninagedovariectomizedrats,J.Bone
Miner.Res.:O?.J.Am.Soc.BoneMiner.Res.28(2013)1641–1652.
[13]B.Su,H.Sheng,M.Zhang,L.Bu,P.Yang,L.Li,F.Li,C.Sheng,Y.Han,S.Qu,
J.Wang,Riskofbonefracturesassociatedwithglucagon-likepeptide-1receptor
agonists''treatment:ameta-analysisofrandomizedcontrolledtrials,Endocrine48
(2015)107–115.
[14]G.Mabilleau,UseofGLP-1mimeticintype2diabetesmellitus:isittheendof
fragilityfractures?Endocrine48(2015)1–2.
[15]C.Yamada,Y.Yamada,K.Tsukiyama,K.Yamada,N.Udagawa,N.Takahashi,
K.Tanaka,D.J.Drucker,Y.Seino,N.Inagaki,Themurineglucagon-likepeptide-1
receptorisessentialforcontrolofboneresorption,Endocrinology149(2008)
574–579.
[16]G.Mabilleau,A.Mieczkowska,N.Irwin,P.R.Flatt,D.Chappard,Optimalbone
mechanicalandmaterialpropertiesrequireafunctionalglucagon-likepeptide-1
receptor,J.Endocrinol.219(2013)59–68.
[17]A.Mieczkowska,S.Mansur,B.Bouvard,P.R.Flatt,B.Thorens,N.Irwin,
D.Chappard,G.Mabilleau,Doubleincretinreceptorknock-out(DIRKO)mice
presentwithalterationsoftrabecularandcorticalmicromorphologyandbone
strength,Osteoporos.Int.:J.Establ.ResultCoop.Eur.Found.Osteoporos.Natl.
Osteoporos.Found.USA26(2015)209–218.
[18]B.Nuche-Berenguer,D.Lozano,I.Gutierrez-Rojas,P.Moreno,M.L.Marinoso,
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
10
P.Esbrit,M.L.Villanueva-Penacarrillo,GLP-1andexendin-4canreversehyperli-
pidic-relatedosteopenia,J.Endocrinol.209(2011)203–210.
[19]Y.Feng,L.Su,X.Zhong,W.Guohong,H.Xiao,Y.Li,L.Xiu,Exendin-4promotes
proliferationanddi?erentiationofMC3T3-E1osteoblastsbyMAPKsactivation,J.
Mol.Endocrinol.56(2016)189–199.
[20]J.Y.Kim,S.K.Lee,K.J.Jo,D.Y.Song,D.M.Lim,K.Y.Park,L.F.Bonewald,B.J.Kim,
Exendin-4increasesbonemineraldensityintype2diabeticOLETFratspotentially
throughthedown-regulationofSOST/sclerostininosteocytes,LifeSci.92(2013)
533–540.
[21]J.L.Perez-Castrillon,J.A.Riancho,D.deLuis,J.R.Caeiro,D.Guede,M.Gonzalez-
Sagrado,M.Ruiz-Mambrilla,M.Domingo-Andres,R.Conde,D.Primo-Martin,The
deleteriouse?ectofbariatricsurgeryoncorticalandtrabecularbonedensityinthe
femursofnon-obese,type2diabeticGoto-Kakizakirats,Obes.Surg.22(2012)
1755–1760.
[22]G.Luo,H.Liu,H.Lu,Glucagon-likepeptide-1(GLP-1)receptoragonists:potentialto
reducefractureriskindiabeticpatients?Br.J.Clin.Pharmacol.81(2016)78–88.
[23]B.Nuche-Berenguer,S.Portal-Nunez,P.Moreno,N.Gonzalez,A.Acitores,
A.Lopez-Herradon,P.Esbrit,I.Valverde,M.L.Villanueva-Penacarrillo,Presenceof
afunctionalreceptorforGLP-1inosteoblasticcells,independentofthecAMP-
linkedGLP-1receptor,J.Cell.Physiol.225(2010)585–592.
[24]M.Pereira,J.Jeyabalan,C.S.Jorgensen,M.Hopkinson,A.Al-Jazzar,J.P.Roux,
P.Chavassieux,I.R.Orriss,M.E.Cleasby,C.Chenu,Chronicadministrationof
Glucagon-likepeptide-1receptoragonistsimprovestrabecularbonemassandar-
chitectureinovariectomisedmice,Bone81(2015)459–467.
[25]E.Gomez,C.Pritchard,T.P.Herbert,cAMP-dependentproteinkinaseandCa2+
in?uxthroughL-typevoltage-gatedcalciumchannelsmediateRaf-independent
activationofextracellularregulatedkinaseinresponsetoglucagon-likepeptide-1in
pancreaticbeta-cells,J.Biol.Chem.277(2002)48146–48151.
[26]Y.Suzuki,H.Zhang,N.Saito,I.Kojima,T.Urano,H.Mogami,Glucagon-like
peptide1activatesproteinkinaseCthroughCa2+-dependentactivationofphos-
pholipaseCininsulin-secretingcells,J.Biol.Chem.281(2006)28499–28507.
[27]J.Buteau,S.Foisy,C.J.Rhodes,L.Carpenter,T.J.Biden,M.Prentki,Proteinkinase
Czetaactivationmediatesglucagon-likepeptide-1-inducedpancreaticbeta-cell
proliferation,Diabetes50(2001)2237–2243.
[28]H.Hui,X.Zhao,R.Perfetti,Structureandfunctionstudiesofglucagon-likepeptide-
1(GLP-1):thedesigningofanovelpharmacologicalagentforthetreatmentof
diabetes,Diabetes/Metab.Res.Rev.21(2005)313–331.
[29]C.Sanz,P.Vazquez,C.Blazquez,P.A.Barrio,M.AlvarezMdel,E.Blazquez,
Signalingandbiologicale?ectsofglucagon-likepeptide1onthedi?erentiationof
mesenchymalstemcellsfromhumanbonemarrow,Am.J.Physiol.Endocrinol.
Metab.298(2010)E634–E643.
[30]R.Siddappa,W.Mulder,I.Steeghs,C.vandeKlundert,H.Fernandes,J.Liu,
R.Arends,C.vanBlitterswijk,J.deBoer,cAMP/PKAsignalinginhibitsosteogenic
di?erentiationandboneformationinrodentmodels,TissueEng.PartA15(2009)
2135–2143.
[31]K.W.Lo,H.M.Kan,K.M.Ashe,C.T.Laurencin,ThesmallmoleculePKA-speci?c
cyclicAMPanalogueasaninducerofosteoblast-likecellsdi?erentiationandmi-
neralization,J.TissueEng.Regen.Med.6(2012)40–48.
[32]A.R.Guntur,C.J.Rosen,Theskeleton:amulti-functionalcomplexorgan:newin-
sightsintoosteoblastsandtheirroleinboneformation:thecentralroleof
PI3Kinase,J.Endocrinol.211(2011)123–130.
[33]I.M.McGonnell,A.E.Grigoriadis,E.W.Lam,J.S.Price,A.Sunters,Aspeci?crolefor
phosphoinositide3-kinaseandAKTinosteoblasts?Front.Endocrinol.3(2012)88.
[34]M.B.Greenblatt,J.H.Shim,L.H.Glimcher,Mitogen-activatedproteinkinase
pathwaysinosteoblasts,Annu.Rev.CellDev.Biol.29(2013)63–79.
[35]T.Jin,I.GeorgeFantus,J.Sun,WntandbeyondWnt:multiplemechanismscontrol
thetranscriptionalpropertyofbeta-catenin,Cell.Signal.20(2008)1697–1704.
[36]B.Gustafson,U.Smith,WNTsignallingisbothaninducerande?ectorofglucagon-
likepeptide-1,Diabetologia51(2008)1768–1770.
[37]J.Meng,X.Ma,N.Wang,M.Jia,L.Bi,Y.Wang,M.Li,H.Zhang,X.Xue,Z.Hou,
Y.Zhou,Z.Yu,G.He,X.Luo,ActivationofGLP?1receptorpromotesbonemarrow
stromalcellosteogenicdi?erentiationthroughbeta-catenin,StemCellRep.6
(2016)579–591.
[38]S.Madsbad,Exenatideandliraglutide:di?erentapproachestodevelopGLP-1re-
ceptoragonists(incretinmimetics)–preclinicalandclinicalresults,bestpractice&
research,Clin.Endocrinol.Metab.23(2009)463–477.
[39]H.X.Sun,N.Lu,X.Luo,L.Zhao,J.M.Liu,Liraglutide,theglucagon-likepeptide-1
receptoragonist,hasanabolicbonee?ectsindiabeticGoto-Kakizakirats,J.
Diabetes7(2015)584–588.
[40]N.Lu,H.Sun,J.Yu,X.Wang,D.Liu,L.Zhao,L.Sun,H.Zhao,B.Tao,J.Liu,
Glucagon-likepeptide-1receptoragonistLiraglutidehasanabolicbonee?ectsin
ovariectomizedratswithoutdiabetes,PLoSOne10(2015)e0132744.
[41]R.Rubin,A.Arzumanyan,A.R.Soliera,B.Ross,F.Peruzzi,M.Prisco,Insulinre-
ceptorsubstrate(IRS)-1regulatesmurineembryonicstem(mES)cellsself-renewal,
J.Cell.Physiol.213(2007)445–453.
[42]P.Xue,X.Wu,L.Zhou,H.Ma,Y.Wang,Y.Liu,J.Ma,Y.Li,IGF1promotesos-
teogenicdi?erentiationofmesenchymalstemcellsderivedfromratbonemarrow
byincreasingTAZexpression,Biochem.Biophys.Res.Commun.433(2013)
226–231.
[43]C.Hamann,S.Kirschner,K.P.Gunther,L.C.Hofbauer,Bone,sweetbone–osteo-
poroticfracturesindiabetesmellitus,Nat.Rev.Endocrinol.8(2012)297–305.
[44]E.Aoyama,I.Watari,K.A.Podyma-Inoue,M.Yanagishita,T.Ono,Expressionof
glucagon-likepeptide-1receptorandglucosedependentinsulinotropicpolypeptide
receptorisregulatedbytheglucoseconcentrationinmouseosteoblasticMC3T3-E1
cells,Int.J.Mol.Med.34(2014)475–482.
[45]E.L.Pacheco-Pantoja,L.R.Ranganath,J.A.Gallagher,P.J.Wilson,W.D.Fraser,
Receptorsande?ectsofguthormonesinthreeosteoblasticcelllines,BMCPhysiol.
11(2011)12.
[46]Y.K.Jeon,M.J.Bae,J.I.Kim,J.H.Kim,S.J.Choi,S.K.Kwon,J.H.An,S.S.Kim,
B.H.Kim,Y.K.Kim,I.J.Kim,Expressionofglucagon-likepeptide1receptorduring
osteogenicdi?erentiationofadipose-derivedstemcells,Endocrinol.Metab.29
(2014)567–573.
[47]D.J.Drucker,Glucagon-likepeptides:regulatorsofcellproliferation,di?erentia-
tion,andapoptosis,Mol.Endocrinol.17(2003)161–171.
[48]V.Gupta,Pleiotropice?ectsofincretins,IndianJ.Endocrinol.Metab.16(Suppl1)
(2012)S47–S56.
[49]C.S.Soltano?,S.Yang,W.Chen,Y.P.Li,Signalingnetworksthatcontrolthelineage
commitmentanddi?erentiationofbonecells,Crit.Rev.Eukaryot.GeneExpr.19
(2009)1–46.
[50]M.Fakhry,E.Hamade,B.Badran,R.Buchet,D.Magne,Molecularmechanismsof
mesenchymalstemcelldi?erentiationtowardsosteoblasts,WorldJ.StemCells5
(2013)136–148.
[51]G.P.Thomas,S.U.Baker,J.A.Eisman,E.M.Gardiner,ChangingRANKL/OPGmRNA
expressionindi?erentiatingmurineprimaryosteoblasts,J.Endocrinol.170(2001)
451–460.
[52]J.Buteau,S.Foisy,E.Joly,M.Prentki,Glucagon-likepeptide1inducespancreatic
beta-cellproliferationviatransactivationoftheepidermalgrowthfactorreceptor,
Diabetes52(2003)124–132.
[53]C.Koole,K.Pabreja,E.E.Savage,D.Wootten,S.G.Furness,L.J.Miller,
A.Christopoulos,P.M.Sexton,RecentadvancesinunderstandingGLP-1R(glu-
cagon-likepeptide-1receptor)function,Biochem.Soc.Trans.41(2013)172–179.
[54]J.F.Chau,W.F.Leong,B.Li,Signalingpathwaysgoverningosteoblastproliferation,
di?erentiationandfunction,Histol.Histopathol.24(2009)1593–1606.
X.Wuetal.ExperimentalCellResearchxxx(xxxx)xxx–xxx
11
|
|
