OPN gene polymorphisms influence the risk of knee OA and OPN levels in synovial fluid in a Chinese population.

0.31 MB·English

by  Jiang, Yongqing

Jiangetal.ArthritisResearch&Therapy2013,15:R3
http://arthritis-research.com/content/15/1/R3
RESEARCH ARTICLE Open Access
OPN gene polymorphisms influence the risk of
knee OA and OPN levels in synovial fluid in a
Chinese population
Yongqing Jiang*, Meng Yao, Qingpeng Liu and Changwei Zhou
Abstract
Introduction: A body of studies suggests the role of osteopontin (OPN) in onset and development of
osteoarthritis (OA), however, the association between OPN polymorphisms and OA susceptibility as well as its
clinical features has not been reported.
Methods: A total of 750 patients with primary knee OA and 794 healthy volunteer were enrolled as controls. Both
OA and control groups were interviewed to obtain demographic and clinical data. Three polymorphisms of OPN
gene, namely, -156GG/G, -443C/T and -66T/G were determined. The levels of the full length and the thrombin-
cleaved OPN in synovial fluid (SF) from OA subjects were measured.
Results: We found the polymorphisms of the -443C/T and the -66/T/G were significantly associated with the OA
risk and the radiographic severity. The -443TT and -66GG showed protective effect against developing OA and
were associated with lower Kellgren-Lawrence grade. Besides, the polymorphisms of -443C/T and -66T/G
significantly affected the thrombin-cleaved OPN levels in SF from OA subjects. Subjects with -443TT and -66GG
genotypes had lower thrombin-cleaved OPN levels in SF. The thrombin-cleaved OPN levels in SF were positively
correlated to the radiographic severity of OA.
Conclusions: Our findings suggest that certain OPN gene polymorphisms may be used as molecular markers for
the susceptibility and severity of OA.
Introduction component[9,10].Inrecentyears,anumberofgenepoly-
Osteoarthritis(OA)isadegenerativejointdisorderresult- morphisms involved in development of knee OA have
ing in substantial morbidity and disability in the elderly beenidentified, such asthose localized in, or adjacentto
[1,2]. It is now accepted that the excessive, spontaneous the encoding sequences for the vitamin D receptor [11],
production of cytokines and mediators of inflammation estrogenreceptoralpha[12]orcalcitonin[13].
plays a significant role in the molecular pathogenesis of Osteopontin (OPN) is a matricellular protein that
OA,contributingtoahighlycatabolicstate, chondrocyte through interactions with its receptors, integrins a4b1,
apoptosis, and the resultant progressive degeneration of a9b1,av(b1,b3,b5),andCD44variants,participatesina
articular cartilage [3-5]. The etiology of OA is largely wide range of physiologic and pathologic processes,
unknown. Aging, trauma, hormonaland mechanical fac- including wound healing, bone turnover, tumor genesis,
torsarereported to contributetothe onsetandprogres- inflammation, andimmune responses [14-17]. The asso-
sion of OA [6-8]. Inheritance studies involving family ciation between OPN and joints had been studied. OPN
groupsandtwinpairshavedemonstratedthatalargepro- expression during chondrocyte maturation is one of the
portion of OA cases can be attributed to genetic factors, importanteventsinvolvedincartilage-to-bonetransitions,
suggesting that the etiology of OA clearly has a genetic and OPN is reportedly involved in the molecular patho-
genesisofOA,contributingtoprogressivedegenerationof
articular cartilage [18]. Clinically, OPN in plasma and
*Correspondence:[email protected] synovialfluid(SF)hasbeenreportedtoberelatedtopro-
DepartmentofOrthopedics,TheSecondAffiliatedHospitalofHarbin
gressive joint damage in knee OA, suggesting that OPN
MedicalUniversity,194NangangDistrict,Harbin,150081,China
©2013Jiangetal.;licenseeBioMedCentralLtd.ThisisanopenaccessarticledistributedunderthetermsoftheCreativeCommons
AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin
anymedium,providedtheoriginalworkisproperlycited.
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page2of8
http://arthritis-research.com/content/15/1/R3
mayserveasabiochemicalmarkerfordeterminingdisease determinedusingTaqMan5’allelicdiscriminationassay.
severity[18]. It was performed using a commercially available kit,
The expression of OPN is significantly influenced by Assays-on-DemandTM single nucleotide polymorphism
geneticpolymorphismsofitspromoter[19].Severalpoly- (SNP) genotyping products (Applied Biosystems, Foster
morphismsinthehumanOPNencodinggenehavebeen City, CA, USA). SNP amplification assays were used
identifiedindifferentpopulations,ofwhichthe-156GG/ accordingtothemanufacturer’sinstructions.Inshort,10
G,-443C/Tand-66T/Gpolymorphismsweremostlystu- ngofsampleDNAin25μLofreactionsolution,contain-
died. These OPN genetic polymorphisms have been ing 12.5 μL of the 2× TaqMan® Universal PCR Mix
reported to be associated with inflammatory disease, (AppliedBiosystems),and1.25μLofpre-developedassay
suchassystemiclupuserythematosus[20],chronichepa- reagentfromtheSNPgenotypingproductcontainingtwo
titisC[21],lupusnephritis[22],andlargearteryathero- primersandtwoMCB-Taqmanprobes.Thereactioncon-
sclerosis[23]. ditionconsisted of pre-incubation at 50°Cfor 2minutes,
AlthoughabodyofstudiessuggeststheroleofOPNin andat95°Cfor10minutes,followedby40cyclesof95°C
theonsetanddevelopmentofOA,theassociationbetween for15s,and60°Cfor1minute.Amplificationswereper-
®
OPNpolymorphismsandOAsusceptibility,aswellasits formed in an ABI Prism 7500 Sequence Detection
clinicalfeatures,hasneverbeenreported.Wecarriedout System(AppliedBiosystems).Toreconfirmtheresults of
a case-control study to clarify whether the abovemen- the genotypes, we randomly selected 25 patients and 25
tionedgeneticpolymorphismsintheOPNgeneareasso- controls(10%ofthestudy subjects) forre-genotypingby
ciated with the susceptibility and severity of OA in a directsequencing;theresultswere100%identical.
Chinesecohort.
OPN level detections
Materials and methods SFwasobtainedfrom175OApatients;nonewasobtained
Patients from controls due to ethical concerns. The levels of the
Atotalof750patientswithprimaryOAofthekneewere full length and the thrombin-cleaved OPN in SF were
recruited from February 2007 to December 2011. The determined by capture ELISA according to the protocol
diagnosisofkneeOAwasbasedontheAmericanCollege provided by the manufacturer (Calbiochem, San Diego,
of Rheumatology criteria [24]. The severity of OA was CA, USA). The sensitivity for OPNwas 3.33 ng/ml, with
evaluated byKellgren-Lawrence (KL)grade.We enrolled intra- and inter-assay coefficients of variation (CV) of <
794healthyvolunteersascontrols.BoththeOAandcon- 5%,and<10%,respectively[25].
trolgroupswereinterviewedtoobtaindemographicdata
and information on all of the established risk factors. Western blot analysis
Patientswithotheretiologiescausingkneediseases,such A 5-μL aliquotfromeachsample of SF wassubjected to
as inflammatory arthritis (rheumatoid, polyarthritic, or SDS-PAGE; the separated proteins were electro-trans-
autoimmunedisease),post-traumaticorpost-septicarthri- ferred onto polyvinylidene fluoride membranes (Milli-
tis, skeletal dysplasia or developmental dysplasia were pore, Bedford, MA, USA). Nonspecific proteins on the
excludedfromtheOAgroup. membraneswereblockedin5%skimmedmilkpowderin
Allthecontrolswereenrolledinthisstudybasedonthe PBS(thrombin-cleavedOPNovernightat4°C.Immuno-
following:1)nohistory,orsignsorsymptomsofarthritis blotting was performed with use of the rabbit against
orjointdiseases(pain,swelling,tenderness,orrestriction human Full length OPN Anti-Osteopontin antibody
ofmovement)and2)normalappearancesonradiography (ab8448,1:1,000dilution,Abcam,Cambridge,MA,USA)
oftheknee.Theclinicalcharacteristicsofallenrolledsub- and thrombin-cleaved OPN antibodies (Mouse Anti-
jects,includingage,sex,bodymassindex(BMI),smoking Human Osteopontin N-Half Monoclonal, 1:1,000 dilu-
status, knee activity and regular exercise were recorded. tion,CosmoBioCo.,Ltd.,Toyo,Japan).Themembranes
Obesity was defined as BMI > 30 kg/M2. The study was were then incubated with the appropriate horseradish
approvedbytheethicsreviewcommitteeofHarbinMedi- peroxidase (HRP)-conjugated secondary antibodies
calUniversityandwritteninformedconsentwasobtained (1:2,500 dilution). Immunoreactive proteins were visua-
fromallparticipants. lized with the western blotting luminol reagent (Santa
CruzBiotechnology,SantaCruz,CA,USA).
OPN gene polymorphisms
DNAwasextractedfrom peripheral whole bloodusing a Statistical analyses
Qiagen DNA Isolation Kit (Qiagen, Valencia, CA, USA). Chisquare(c2)testswereusedtocomparegenotypefre-
Three single nucleotide polymorphismson the promoter quencyanddemographicdistributionsbetweencasesand
region of OPN gene, including -66T/G (rs28357094), controls.Multiplelogisticregressionanalysiswasusedto
-156G/GG(rs17524488),and-443C/T(rs11730582),were evaluate if each SNP was independently associated with
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page3of8
http://arthritis-research.com/content/15/1/R3
OA when adjusted for the potential confounding effects adjusted OR was 0.652 (P = 0.017) compared with the
ofimportantclinicalvariables.Theoddsratios(OR)and -66TTgenotype.TheadjustedORforthe-66Gallelecar-
95% CIs were calculated. The associations between the riagewas0.815(P=0.011).
OPN haplotypes and OA risk were analyzed. The D’ The associations between the OPN haplotypes and
valueandr2forthethreeSNPsthatwestudiedwerecal- knee OA risk were analyzed in this study. The D’ value
culated using SHEsis software [26]. All other analyses for the three SNPs were calculated with the SHEsis soft-
wereperformedusing SPSSsoftware(StatisticalPackage ware [26]. All three SNPS were in strong linkage dise-
for the Social Sciences, version 16.0, SPSS Inc, Chicago, quilibrium (LD) (all D’ > 0.8). The estimated haplotype
IL,USA). frequencies of the OPN SNPs are shown in Table 3. The
haplotype G T G represented a protective effect
-156 -443 -66
Results for developing knee OA (adjusted OR 0.754, P = 0.005).
Table 1 shows demographic and clinical characteristics Meanwhile, the GG C T showed a higher risk
-156 -443 -66
of all subjects in the study. There were no significant for developing OA (OR 1.943, P < 0.001).
differences in sex, age, smoking status, history of work Weanalyzedthegenotypeandtheradiographicseverity
involving heavy labor, hypertension, or diabetes mellitus ofOA inthe patient group,whowere groupedinto sub-
between knee OA cases and controls. The OA patient jects with KL grade = < 3 and those with KL grade > 3
group had a markedly higher BMI than the controls (Table3).We foundthatthe-443C/Tand-66T/Ggeno-
(P = 0.001). types were significantly different between OA subjects
Table 2 describes the genotype distributions and allele with KL grade = < 3 and those with KL grade > 3 (both
frequencies of OPN polymorphisms in OA and control P < 0.001). Logistic regression analysis showed that the
subjects.Thegenotypefrequenciesforallpolymorphisms -66GG genotype and -443TT carriers were less likely to
did not differ significantly from those expected under havesevereOA(KLgrade>3)comparedwith-66TTand
Hardy-Weinbergequilibrium (both P >0.05). The geno- -443CCcarriers.TheadjustedORfor-443TTcarrierswas
typeandallelefrequenciesatOPN-156/G/GGweresimi- 0.529(OR0.664,P=0.011,with-443CCasthereference)
lar in OA and control subjects. In contrast, the -443C/T and the adjusted OR for -66GG carriers was 0.635 (OR
and-66/T/Ggenotypeweresignificantlydifferentbetween 0.757, P = 0.023, with -66TT as the reference). The car-
kneeOAsubjectsandcontrols.TheOApatientshadsig- riage of -443T and -66G represented a protective role
nificantlylowerratesof-443TTand-66GGthancontrols against developing severe OA (adjusted OR 0.659 and
(P < 0.001 and P = 0.013, respectively). Accordingly, the 0.732, P = 0.008 and 0.017, respectively) (Table 4). The
-443T and -66G allele frequencies were lower in OA genetic polymorphisms of -156G/GG did not influence
patientsthanincontrols(P=0.018andP=0.009,respec- theseverityofOA.
tively).Logisticregressionanalysisshowedasignificantly We further analyzed the correlationbetween the OPN
decreasedriskforkneeOAforthe-443TTgenotypecom- levelinSFandtheradiographicseverityofOA.Themean
pared with the -443CC genotype (OR 0.525, P < 0.001) fulllengthOPNlevelsweresimilaramongKLgrade2,KL
after adjustment for sex, age, smoking status, family his- grade 3 and KL grade 4 subgroups (data not shown). In
tory and history of heavy labor. The adjusted OR for the contrast, the mean thrombin-cleaved OPN levels in OA
-443T allele carriage was 0.689 (P <0.001). Similarly the subjectswerequitedifferentamongtheseKL-gradedsub-
-66GG genotype carriers had a lower riskfor OA, as the groups.Themeanthrombin-cleavedOPNlevelwas,669.2
Table 1Demographic andclinical characteristics ofall subjects inthe study
Variables Cases(n=750) Control(n=794) P
Age,years,mean±SD 65.2±7.6 65.1±5.4 ns
Female,n(%) 401(53.47%) 406(51.13%) ns
BMI,kg/m2,mean±SD 27.7±2.4 22.5±3.1 0.001
Smoker,n(%) 283(37.73%) 291(36.65%) ns
FamilyhistoryofOA,n(%) 60(8.00%) 67(8.43%) ns
Historyoflaborwork,n(%) 140(18.66%) 158(19.89%) ns
Hypertension 198(26.4%) 206(25.9%) ns
Diabetesmellitus 152(20.3%) 166(20.9%) ns
KLgrade
=<3 351 -
>3 399 -
BMI,bodymassindex;OA,osteoarthritis;KL,Kellgren-Lawrence;ns,notsignificant.
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page4of8
http://arthritis-research.com/content/15/1/R3
Table 2Genotype distributions andallele frequencies ofOPNpolymorphisms inOA andcontrol subjects
OPN Allele OApatients, % Controls, % GlobalP- Adjusted 95%CI AdjustedP-
polymorphism frequency n n value OR value
-156/G/GG GG 203 27.07 213 26.83 0.557 1
GGG 352 46.93 391 49.24 0.956 0.732 1.276 0.701
GGGG 195 26.00 190 23.93 1.787 0.887 1.552 0.674
G 758 50.53 817 51.45 0.318 1
GG 742 49.47 771 48.55 1.038 0.785 1.267 0.658
-443C/T CC 167 22.27 119 14.99 <0.001 1
CT 384 51.20 405 51.01 0.763 0.457 0.998 0.003
TT 199 26.53 270 34.01 0.525 0.392 0.872 <0.001
C 718 47.87 643 40.49 0.018 1
T 782 52.13 945 59.51 0.689 0.575 0.896 <0.001
-66T/G TT 188 25.07 153 19.27 0.013 1
TG 369 49.20 402 50.63 0.738 0.516 0.988 0.035
GG 193 25.73 239 30.10 0.657 0.494 0.874 0.017
T 745 49.67 708 44.58 0.009 1
G 755 50.33 880 55.42 0.876 0.765 0.876 0.011
OPN,osteopontin;OA,osteoarthritis;OR,oddsratio.
±476(mean±SD)pg/mlinpatientswithKLgrade2OA, patients. The thrombin-cleaved OPN expressions in the
5,203.6±385pg/mlinthosewithKLgrade3,and5,642.7 SF from -66 GG patients were lower than those from
±350pg/mlinthosewithKLgrade4(Figure1).Positive -66GT and -66TT carriers (Figure 3). The protein
correlationwasfoundbetweenthethrombin-cleavedOPN expression of full-length OPN levels in SF were similar
levels (but not the full length OPN levels) in SF and OA among -443TT carriers compared with the -443TC and
severity(Pearsoncorrelationcoefficient0.692,P<0.001). -443CC, as well as the -66GG, -66GT and -66TT geno-
We next analyzed the thrombin-cleaved OPN levels in type carriers (figure not shown).
SF according to the genotypes of the OPN gene poly-
morphisms. Our results showed that the thrombin- Discussion
cleaved OPN levels in SF were significantly lower in Inthisstudy,weinvestigatedtheroleofOPNgenepoly-
-443TT carriers compared with the -443TC and -443CC morphisms in determining the susceptibility to, and
carriers (4,728 ± 354 vs. 5,398 ± 354 and 5334 ± 323, radiographicseverityofOAinaChinesecohort.Ofthree
pg/ml, both P < 0.05). Similarly, the thrombin-cleaved gene polymorphisms in this study, we found the poly-
OPN levels in SF from -66GG were lower than from the morphisms of two loci, namely, the -443C/T and the
-66GT and -66TT carriers (4,989 ± 329 vs. 5,413 ± 357 -66T/G, were significantly associated with the risk and
and 5,426 ± 274, pg/ml, both P < 0.05). The genetic radiographic severity of OA. The -443TT and -66GG
polymorphisms of -156G/GG did not influence the polymorphismsshowedaprotectiveeffectagainstoccur-
thrombin-cleaved OPN levels in SF (Figure 2). Western renceofOAandwereassociatedwithlowerradiographic
blot results showed that the thrombin-cleaved OPN severity. Inaddition,thepolymorphismsof-443C/Tand
expressions in the SF from -443TT patients were mark- -66T/Gsignificantlyaffectedthethrombin-cleaved OPN
edly lower than that from the -443TC and -443CC levelsin SF from OA patients.Patientswiththe -443TT
Table 3Associations between OPNhaplotypes andriskofkneeOA
Haplotypeanalysis*
-156G/GG -443C/T -66T/G Cases(frequency) Controls(frequency) Chi2 Pearson’sP Oddsratio(95%CI)
G C T 166 194. 4.451 0.065 0.789(0.633,1.092)
G T G 218 262 8.044 0.005 0.754(0.620,0.917)
G C T 209 206 0.206 0.650 0.953(0.775,1.172)
G T G 154 177 3.422 0.064 0.806(0.642,1.013)
GG C T 273 145 37.47 0.000 1.943(1.566,2.409)
GG C G 222 165 6.324 0.012 1.317(1.062,1.633)
GG T T 231 210 0.181 0.670 1.045(0.854,1.278)
*Allhaplotypeswithfrequency<0.03areignoredinthisanalysis.
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page5of8
http://arthritis-research.com/content/15/1/R3
Table 4Genotype andthe radiographic severity in patients withOA
Genotype KL=<3,numberof % KL>3,numberof % GlobalP- Adjusted 95%CI AdjustedP-
patients patients value OR value
-156/G/ GG 92 26.21 91 22.81 0.462 1
GG
GGG 166 47.29 206 51.63 0.787 0.564 1.178 0.223
GGGG 93 26.50 102 25.56 0.916 0.638 1.376 0.653
G 350 49.86 388 48.62 1
GG 352 50.14 410 51.38 0.934 0.786 1.186 0.648
-443C/T CC 88 25.07 79 19.80 <0.001 1
CT 176 50.14 188 47.12 0.816 0.537 1.264 0.352
TT 87 24.79 132 33.08 0.664 0.365 0.879 0.011
C 352 50.14 346 43.36 1
T 350 49.86 452 56.64 0.659 0.647 0.948 0.008
-66T/G TT 97 27.64 91 22.81 0.462 1
TG 158 45.01 191 47.87 0.865 0.587 1.234 0.174
GG 96 27.35 117 29.32 0.757 0.438 0.944 0.023
T 332 47.29 373 46.74 1
G 370 52.71 425 53.26 0.7326 0.668 0.973 0.017
KL,Kellgren-Lawrencegrade
and -66GG genotypes had markedly lower thrombin- OPNisanextracellularmatrixproteinwithpleiotropic
cleaved OPN levels in SF than those with -443CT, properties,andhasbeenrecently recognizedasapoten-
-443CC and -66 GT and -66TT. Moreover, the throm- tial inflammatory cytokine [27]. Expression of OPN has
bin-cleaved OPN levels in SF were positively correlated been observed in the joints of patients with rheumatoid
withthe radiographic severity ofOA.To the bestofour arthritis(RA)aswellasOA[28].ThefunctionofOPNis
knowledge, this is the first study of the significance of modulatedbyproteasedigestion,andathrombin-cleaved
OPN gene polymorphisms in the susceptibility to, and form of OPN is involved in the pathogenesis of various
severity of OA. Our findings suggest that certain OPN inflammatorydisorders,includingOAandRA[29,25,30].
gene polymorphisms may be used as molecular markers More than fifty single nucleotide polymorphisms have
forsusceptibilityto,andseverityofOA. beenidentifiedinthe human OPNencoding gene indif-
ferentpopulations[31].WeselectedthreeSNPSbasedon
previousassociationwithautoimmuneand/orinflamma-
torydiseasesandabilitytodeterminehaplotypespredict-
ingserumOPNconcentrations[32,21,33,34].TheSNPat
-66 was predicted to bind the specificity protein 1 (SP1)
transcription factor, as it is known that the sequence
between-68and-59isanSP1bindingsiteinthehuman
OPN promoter [35]. A previous study showed that SP1
andSP3factorsrecognizethissiteandthattheallelewith
the -66Tnucleotide inside the SP1recognitionsequence
has a higher binding affinity [31]. Among patients with
cancer, those with G/G at -156 have been found to have
higher concentrations of OPN than those with G/GG or
GG/GG.PatientswithG/Gat-156havemorefrequently
beendiagnosedwithadvancedstagethanwithearlystage
non-small cell lungcancer (NSCLC) [36].Another study
revealedthatthefunctional-443C/Tpolymorphisminflu-
encesOPNgeneexpressioninmelanomacellsviabinding
ofc-Mybtranscriptionfactor[37].Todate,thesegenetic
Figure1Meanthrombin-cleavedosteopontin(OPN)levelsin variations of the OPN gene have been described to be
synovialfluid(SF)INpatientswithosteoarthritis(OA)
associated with inflammatory diseases, including lupus
accordingtoKellgren-Lawrence(KL)grade.
erythematosus, multiple sclerosis, urolithiasis, primary
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page6of8
http://arthritis-research.com/content/15/1/R3
Figure 2 Mean thrombin-cleaved osteopontin (OPN) levels in synovial fluid (SF) according to the genotypes of OPN gene
polymorphisms.
biliary cirrhosis, and autoimmune lymphoproliferative A study reported that the -66 polymorphism modifies
syndrome. thebindingaffinityfortheSP1/SP3transcriptionfactors.
Our results in this study showed that the thrombin- TheauthorsfoundthatSP1andSP3factorsrecognizethis
cleaved OPN levels in SF were significantly lower in site and that the allele with the T nucleotide inside the
-443TT compared with -443TC and -443CC carriers. A SP1 recognition sequence has a higher binding affinity.
previousstudy examinedtheeffectofthe-443T/Cpoly- OverexpressionofSP1inducedhigherpromoteractivation
morphismontranscriptionoftheOPNgene.Theauthors on a plasmid carrying the T allele compared with the
measuredpromoteractivitywithadualluciferaserepor- equivalentplasmid carryingtheGallele. Consistentwith
terassaysystemandcomparedtheactivitiesofthe-443C thisfinding,wefoundthethrombin-cleavedOPNlevelsin
and -443T alleles with the MKN28 and SGC-7901 cell SFfrom-66GGwaslowerthan-66GTand-66TT.
lines.They foundsignificantlyhigherluciferaseactivities Two previous studies investigated the association
were generatedwiththe pGL3-Cconstructcomparedto between the OPN gene polymorphisms and RA. Unlike
thepGL3-Tconstruct[38].Arecentstudyofmetastases thereportedeffectoftheOPNSNPconferringpredispo-
in melanoma found that those homozygous for the sition to common diseases such as multiple sclerosis or
-443Calleleexpressedsignificantlyhigher levelsofOPN systemic lupus erythematosus, four SNPs of the OPN
mRNAcomparedtothosethatwereeitherheterozygous gene(327T/C,795C/T,1128A/G,and1284A/C)didnot
(CT) or homozygous for the -443T allele. Transcription contribute to RA susceptibility in a Spanish population
factorc-MybbindstotheregionoftheOPNpromoterin [39].AnotherstudyinaChinesepopulationshowedthat
an allele-specific manner and induces enhanced activity OPNgenepolymorphismsdonotcorrelatewithsuscept-
ofthe-443Ccomparedtothe-443TOPNpromoter[37]. ibilitytoRA[40].Inthepresentstudy,wefoundthatthe
These studies provide information on the mechanism -443C/T and -66T/G were associated not only with risk
under which the -443T/C polymorphism regulates the of OA, but also its radiographic severity, suggesting the
OPN gene transcript activity, thus influencing OPN role of OPN genetic polymorphisms inOA and RA may
expression. bequitedifferent.
Figure3Thrombin-cleavedosteopontin(OPN)levelsinsynovialfluid(SF)fromdifferentgenotypecarriers.GAPDH,glyceraldehyde-3-
phosphatedehydrogenase.
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page7of8
http://arthritis-research.com/content/15/1/R3
ElevatedlevelsofOPNhavebeenfoundinSFfromRA Abbreviations
patientsandincreasedlevelsofOPNhavebeencorrelated BMI:bodymassindex;CV:coefficientofvariation;ELISA:enzyme-linked
immunosorbentassay;GAPDH:glyceraldehyde-3-phosphatedehydrogenase;
with increased levels ofmultiple inflammatory cytokines
KL:Kellgren-Lawrence;NSCLC:non-smallcelllungcancer;OA:osteoarthritis;
[28].Arecentstudy showed that thrombin-cleavedOPN OPN:osteopontin;OR:oddsratio;PBS:phosphate-bufferedsaline;RA:
levelsfromSFsamplescorrelatewithradiographicseverity rheumatoidarthritis;SF:synovialfluid;SP:specificityprotein.
determinedbyKLgradingofkneeOAinaJapanesestudy Authors’contributions
[29]. In that study, immunohistochemistry of the syno- YJ:studydesign,manuscriptwriting;MY,CZandQL:OPNgenotypingand
viumshowedstrongerreactivityinsamplesfromsubjects clinicaldatacollection.Allauthorshavereadandapprovedthemanuscript
forpublication.
with advanced OA. Consistent with this finding, we
observedinthepresentstudythatthrombin-cleavedOPN Competinginterests
levelsfromSFweresignificantly higherinthose with KL Theauthorsdeclarethattheyhavenocompetinginterests.
grade>3thaninthosewithKLgrade=<3;bystratifying
Acknowledgements
the thrombin-cleaved OPN levels according to the OPN WethankDr.LiZhenandXuweiHoufortheirhelpinstatisticalanalyses.
genotypes,wefoundthatthe-443and-66polymorphisms ThisstudywassupportedbytheNationalNaturalScienceFoundationof
China(30770543).
had a substantial influence on thrombin-cleaved OPN
levels.The-443TTand-66GG,whichshowedaprotective Received:29July2012 Revised:18November2012
role against susceptibility to OA, had markedly lower Accepted:20December2012 Published:5January2013
thrombin-cleaved OPN levels. Our data confirmed the
References
positivecorrelationbetweenthrombin-cleavedOPNlevels
1. HunterDJ,WilsonDR:Imagingtheroleofbiomechanicsinosteoarthritis.
inSFandOAseverity.Wesuggestthatthrombin-cleaved RheumDisClinNorthAm2009,35:465-483.
OPNmayserveasabiochemicalmarkerfordetermining 2. CoulibalyY,DiakiteAA,KeitaM,DiakiteI,DembeleM,DialloG:
Epidemiologyandtherapyofosteoarthritisinchildren.MaliMed2009,
diseaseseverityandcouldbepredictiveofprognosiswith
24:7-10.
respecttotheprogressionofkneeOA.Sincepreviousstu- 3. AbramsonSB:Inflammationinosteoarthritis.JRheumatolSuppl2004,
dies showed that the alleles at polymorphic sites in the 70:70-76.
4. BonnetCS,WalshDA:Osteoarthritis,angiogenesisandinflammation.
OPNgeneinfluencetheOPNlevelbyinfluencingthelevel
Rheumatology(Oxford)2005,44:7-16.
of promoter activity and characterizing their ability to 5. BenitoMJ,VealeDJ,FitzGeraldO,vandenBergWB,BresnihanB:Synovial
bind transcription factors, we here postulated that the tissueinflammationinearlyandlateosteoarthritis.AnnRheumDis2005,
64:1263-1267.
influence of OPN polymorphisms on risk and severity of
6. EsserS,BaileyA:Effectsofexerciseandphysicalactivityonknee
OA is very likely due to its effect on the synthesis of the osteoarthritis.CurrPainHeadacheRep2011,15:423-430.
activeOPNprotein. 7. BaeJY,ParkKS,SeonJK,KwakDS,JeonI,SongEK:Biomechanicalanalysis
oftheeffectsofmedialmeniscectomyondegenerativeosteoarthritis.
Some limitations of this study need to be addressed.
MedBiolEngComput2012,50:53-60.
First, there was a large proportion of obese patients in 8. AndersonDD,MarshJL,BrownTD:Thepathomechanicaletiologyofpost-
this study. As obesity is an important risk factor for OA, traumaticosteoarthritisfollowingintraarticularfractures.IowaOrthopJ
2011,31:1-20.
not only as a mechanical factor but also due to the
9. ZhangW,DohertyM:Howimportantaregeneticfactorsin
related increase in cytokines, which contribute to OA osteoarthritis?Contributionsfromfamilystudies.JRheumatol2005,
development. Thus, there might have been enrollment 32:1139-1142.
10. BianQ,WangYJ,LiuSF,LiYP:Osteoarthritis:geneticfactors,animal
bias in the recruitment of patients with primary OA
models,mechanisms,andtherapies.FrontBiosci(EliteEd)2012,4:74-100.
into this study. Second, we enrolled a Chinese cohort of 11. KeenRW,HartDJ,LanchburyJS,SpectorTD:Associationofearly
patients with knee OA only, so, whether the positive osteoarthritisofthekneewithaTaqIpolymorphismofthevitaminD
receptorgene.ArthritisRheum1997,40:1444-1449.
roles of OPN gene polymorphisms we found in this
12. JinSY,HongSJ,YangHI,ParkSD,YooMC,LeeHJ,HongMS,ParkHJ,
study exist in other types of OA, or in other ethnic YoonSH,KimBS,YimSV,ParkHK,ChungJH:Estrogenreceptor-alpha
populations, remains unknown. Third, we did not per- genehaplotypeisassociatedwithprimarykneeosteoarthritisinKorean
population.ArthritisResTher2004,6:R415-421.
form functional analyses to further disclose the mechan-
13. MaganaJJ,Galvez-RosasA,Gonzalez-HuertaC,Duarte-SalazarC,Lara-
ism under which the OPN gene polymorphisms affect AlvaradoL,Soria-BastidaMA,Cortes-GonzalezS,Miranda-DuarteA:
the susceptibility to and severity of OA. Associationofthecalcitoningene(CA)polymorphismwithosteoarthritis
ofthekneeinaMexicanmestizopopulation.Knee2010,17:157-160.
14. GiachelliCM,SteitzS:Osteopontin:aversatileregulatorofinflammation
Conclusions andbiomineralization.MatrixBiol2000,19:615-622.
Inthisstudy we found the polymorphismsof two loci of 15. HeilmannK,HoffmannU,WitteE,LoddenkemperC,SinaC,SchreiberS,
HayfordC,HolzlohnerP,WolkK,TchatchouE,MoosV,ZeitzM,SabatR,
OPN gene, namely, the -443C/T and the -66T/G, were
GünthertU,WittigBM:Osteopontinastwo-sidedmediatorofintestinal
significantlyassociatedwithriskandradiographicseverity inflammation.JCellMolMed2009,13:1162-1174.
of OA, suggesting that these OPN gene polymorphisms 16. WolakT,KimH,RenY,KimJ,VaziriND,NicholasSB:Osteopontin
modulatesangiotensinII-inducedinflammation,oxidativestress,and
maybeusedasmolecularmarkersforthesusceptibilityto,
fibrosisofthekidney.KidneyInt2009,76:32-43.
andseverityofOA.
Jiangetal.ArthritisResearch&Therapy2013,15:R3 Page8of8
http://arthritis-research.com/content/15/1/R3
17. Zhivkova-GalunskaM,AdwanH,EyolE,KleeffJ,KolbA,BergmannF, 34. GolledgeJ,MullerJ,ShephardN,ClancyP,SmallwoodL,MoranC,DearAE,
BergerMR:Osteopontinbutnotosteonectinfavorsthemetastatic PalmerLJ,NormanPE:Associationbetweenosteopontinandhuman
growthofpancreaticcancercelllines.CancerBiolTher2010,10:54-64. abdominalaorticaneurysm.ArteriosclerThrombVascBiol2007,27:655-660.
18. HonsawekS,TanavaleeA,SakdinakiattikoonM,ChayanupatkulM, 35. WangD,YamamotoS,HijiyaN,BenvenisteEN,GladsonCL:Transcriptional
YuktanandanaP:Correlationofplasmaandsynovialfluidosteopontin regulationofthehumanosteopontinpromoter:functionalanalysisand
withdiseaseseverityinkneeosteoarthritis.ClinBiochem2009,42:808-812. DNA-proteininteractions.Oncogene2000,19:5801-5809.
19. ChiuYW,TuHF,WangIK,WuCH,ChangKW,LiuTY,KaoSY:The 36. ChangYS,KimHJ,ChangJ,AhnCM,KimSK:Elevatedcirculatinglevelof
implicationofosteopontin(OPN)expressionandgenetic osteopontinisassociatedwithadvanceddiseasestateofnon-smallcell
polymorphismsofOPNpromoterinoralcarcinogenesis.OralOncol2010, lungcancer.LungCancer2007,57:373-380.
46:302-306. 37. SchultzJ,LorenzP,IbrahimSM,KundtG,GrossG,KunzM:Thefunctional
20. TrivediT,FranekBS,GreenSL,KariukiSN,KumabeM,MikolaitisRA,JollyM, -443T/Costeopontinpromoterpolymorphisminfluencesosteopontin
UtsetTO,NiewoldTB:Osteopontinallelesareassociatedwithclinical geneexpressioninmelanomacellsviabindingofc-Mybtranscription
characteristicsinsystemiclupuserythematosus.JBiomedBiotechnol2011, factor.MolCarcinog2009,48:14-23.
2011:802581. 38. ZhaoF,ChenX,MengT,HaoB,ZhangZ,ZhangG:Genetic
21. MochidaS,HashimotoM,MatsuiA,NaitoM,InaoM,NagoshiS,NaganoM, polymorphismsintheosteopontinpromoterincreasestheriskof
EgashiraT,MishiroS,FujiwaraK:Geneticpolymorphimsinpromoter distancemetastasisanddeathinChinesepatientswithgastriccancer.
regionofosteopontingenemaybeamarkerreflectinghepatitisactivity BMCCancer2012,12:477.
inchronichepatitisCpatients.BiochemBiophysResCommun2004, 39. UrcelayE,MartinezA,Mas-FontaoA,Peris-PertusaA,Pascual-SalcedoD,
313:1079-1085. BalsaA,Fernandez-ArqueroM,delaConchaE:Osteopontingene
22. MiyazakiT,OnoM,QuWM,ZhangMC,MoriS,NakatsuruS,NakamuraY, polymorphismsinSpanishpatientswithrheumatoidarthritis.J
SawasakiT,EndoY,NoseM:Implicationofallelicpolymorphismof Rheumatol2005,32:405-409.
osteopontininthedevelopmentoflupusnephritisinMRL/lprmice.Eur 40. XuG,SunW,HeD,WangL,ZhengW,NieH,NiL,ZhangD,LiN,ZhangJ:
JImmunol2005,35:1510-1520. Overexpressionofosteopontininrheumatoidsynovialmononuclear
23. KimY,LeeC:Haplotypeanalysisrevealedageneticinfluenceof cellsisassociatedwithjointinflammation,notwithgenetic
osteopontinonlargearteryatherosclerosis.JBiomedSci2008,15:529-533. polymorphism.JRheumatol2005,32:410-416.
24. AltmanR,AschE,BlochD,BoleG,BorensteinD,BrandtK,ChristyW,
CookeTD,GreenwaldR,HochbergM:Developmentofcriteriaforthe doi:10.1186/ar4129
classificationandreportingofosteoarthritis.Classificationof Citethisarticleas:Jiangetal.:OPNgenepolymorphismsinfluencethe
osteoarthritisoftheknee.DiagnosticandTherapeuticCriteria riskofkneeOAandOPNlevelsinsynovialfluidinaChinese
CommitteeoftheAmericanRheumatismAssociation.ArthritisRheum population.ArthritisResearch&Therapy201315:R3.
1986,29:1039-1049.
25. HasegawaM,NakoshiY,IinoT,SudoA,SegawaT,MaedaM,YoshidaT,
UchidaA:Thrombin-cleavedosteopontininsynovialfluidofsubjects
withrheumatoidarthritis.JRheumatol2009,36:240-245.
26. SaitohE,MatsuiA,InaoM,MimuraS,KohH,OhnoA,NagoshiS,KayanoH,
YoshimotoT,MochidaS,FujiwaraK:Extrahepaticmanifestationsin
transgenicmiceofosteopontininhepatocytes-Acluetoadventof
pathologicalstateinvariousorgansofchronichepatitisCpatients.
HepatolRes2005,31:197-202.
27. AtturMG,DaveMN,StuchinS,KowalskiAJ,SteinerG,AbramsonSB,
DenhardtDT,AminAR:Osteopontin:anintrinsicinhibitorof
inflammationincartilage.ArthritisRheum2001,44:578-584.
28. SharifSA,DuX,MylesT,SongJJ,PriceE,LeeDM,GoodmanSB,
NagashimaM,MorserJ,RobinsonWH,LeungLL:Thrombin-activatable
carboxypeptidaseBcleavageofosteopontinregulatesneutrophil
survivalandsynoviocytebindinginrheumatoidarthritis.ArthritisRheum
2009,60:2902-2912.
29. HasegawaM,SegawaT,MaedaM,YoshidaT,SudoA:Thrombin-cleaved
osteopontinlevelsinsynovialfluidcorrelatewithdiseaseseverityof
kneeosteoarthritis.JRheumatol2011,38:129-134.
30. YamamotoN,SakaiF,KonS,MorimotoJ,KimuraC,YamazakiH,OkazakiI,
SekiN,FujiiT,UedeT:EssentialroleofthecrypticepitopeSLAYGLR
withinosteopontininamurinemodelofrheumatoidarthritis.JClin
Invest2003,112:181-188.
31. GiacopelliF,MarcianoR,PistorioA,CatarsiP,CaniniS,KarsentyG,
RavazzoloR:Polymorphismsintheosteopontinpromoteraffectits
transcriptionalactivity.PhysiolGenomics2004,20:87-96.
32. ChiocchettiA,IndelicatoM,BensiT,MesturiniR,GiordanoM,SamettiS,
CastelliL,BottarelF,MazzarinoMC,GarbariniL,GiacopelliF,ValesiniG, Submit your next manuscript to BioMed Central
SantoroC,DianzaniI,RamenghiU,DianzaniU:Highlevelsofosteopontin and take full advantage of:
associatedwithpolymorphismsinitsgeneareariskfactorfor
developmentofautoimmunity/lymphoproliferation.Blood2004,
• Convenient online submission
103:1376-1382.
33. ChiocchettiA,ComiC,IndelicatoM,CastelliL,MesturiniR,BensiT, • Thorough peer review
MazzarinoMC,GiordanoM,D’AlfonsoS,Momigliano-RichiardiP,LiguoriM,
• No space constraints or color figure charges
ZorzonM,AmorosoA,TrojanoM,MonacoF,LeoneM,MagnaniC,
DianzaniU:Osteopontingenehaplotypescorrelatewithmultiple • Immediate publication on acceptance
sclerosisdevelopmentandprogression.JNeuroimmunol2005, • Inclusion in PubMed, CAS, Scopus and Google Scholar
163:172-178.
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit