sign in sign up
<<
Home , IFNA16, IFNB1, Osteopontin

5 181

U Razny and others turnover expression 181:5 525–537 Clinical Study in obesity

Effect of resistance on whole blood mRNA and microRNA expression affecting bone turnover

Urszula Razny, Anna Polus, Joanna Goralska, Anna Zdzienicka, Anna Gruca, Maria Kapusta, Maria Biela, Aldona Dembinska-Kiec, Bogdan Solnica and Correspondence should be addressed Malgorzata Malczewska-Malec to U Razny Department of Clinical Biochemistry, Jagiellonian University Medical College, Krakow, Poland Email [email protected]

Abstract

Objective: To evaluate the effect of insulin resistance in obesity on the expression in whole blood of mRNA and miRNA affecting bone homeostasis as well as to estimate the influence of oral glucose load (OGTT) on serum concentration in obese individuals with and without insulin resistance. Design: Cross-sectional study. Methods: Carboxylated (cOC), undercarboxylated (ucOC) and total osteocalcin were measured by ELISA in the serum of obese subjects with insulin resistance (n = 41) and obese subjects without insulin resistance (n = 41) (control group) during OGTT. Analysis of (microarray) and miRNAs (real-time PCR) was performed in venous blood (representating samples) collected before OGTT from obese with insulin resistance and controls. Results: Obese subjects with insulin resistance (higher HOMA-IR and lower oral glucose insulin sensitivity index) presented significantly increased expression of WNT signalling inhibitors DKK1( , DKK2, SOST, SFRP1) and downregulation of the key factor in WNT signalling – β catenin participating in differentiation. Expression of miRNA involved in osteoblastogenesis was also inhibited (miR-29b, miR-181a, miR-210, miR-324-3p). During OGTT, contrary to the

European Journal of Endocrinology control group, subjects with insulin resistance presented suppression of cOC and total OC decrease after 1 and 2 h of oral glucose load. Conclusions: Obese subjects with insulin resistance may have defects in osteoblastogenesis that was demonstrated via key signalling molecules mRNA downregulation, and increased expression of WNT antagonists as well as inhibition of expression of miRNA participating in the regulation of differentiation. Disturbed osteoblastogenesis in insulin-resistant subjects results in the suppression of blood carboxylated and total osteocalcin decrease during OGTT.

European Journal of Endocrinology (2019) 181, 525–537

Introduction

Inappropriate secretion of insulin and insulin resistance matrix (2, 3). During bone resorption, undercarboxylated as well as chronic low-grade inflammation in obesity leads osteocalcin (ucOC) is released and participates in to disturbances in bone turnover. Ferron et al. reported glucose and lipid metabolism (4, 5). Studies in mice that knockout of the on osteoblasts led showed that knockout of the osteocalcin gene (Ocn−/−) to low bone mass in mice (1). By secreting osteocalcin led to obesity, increased serum triglycerides as well as (OC) osteoblasts participate in bone formation as well suppression of insulin secretion and glucose intolerance as energy metabolism. Osteocalcin in carboxylated form (6). Human studies also showed that obese subjects with (cOC) interacts with hydroxyapatite crystals of the bone pre-diabetes had decreased ucOC blood levels, which

https://eje.bioscientifica.com © 2019 European Society of Endocrinology Published by Bioscientifica Ltd. https://doi.org/10.1530/EJE-19-0542 Printed in Great Britain Downloaded from Bioscientifica.com at 09/23/2021 09:31:55PM via free access

-19-0542 European Journal of Endocrinology https://eje.bioscientifica.com full explanationofthepurpose andnatureofprocedures written consentwasobtained fromeachsubjectafter 2009 tillJuly2013.Priorto participationinthestudy Medical CollegeinCracow, PolandfromSeptember the DepartmentofBiochemistry, JagiellonianUniversity patient ClinicofObesityandLipidDisorderTreatment at Study participants were recruited from patients at theOut- Cracow, Poland(opinionnumberKBET/82/B/2009). Bioethics CommitteeoftheJagiellonianUniversity in (Declaration ofHelsinki)andwasapprovedbythe with The Code of Ethics of the World Medical Association The cross-sectional study was performed in accordance Subjects Subjects andmethods insulin resistance. OGTT coulddifferincomparisontosubjectswithout and consequently, blood osteocalcin concentration after and miRNA affecting bone turnover could be disturbed that duetoinsulinresistance,theexpressionofmRNA osteocalcin concentrationinobesity. We hypothesise as welltheinfluenceoforalglucoseloadonserum homeostasis in whole bloodfromobese individuals on theexpressionofmRNAandmiRNAaffectingbone resistance and disturbed oralglucose insulin sensitivity processes inhealthyaswelldiseasestates. in wholebloodcouldpartiallyreflectboneturnover bone formationandresorption,geneexpressionpatterns consideration thecirculatory componentoftheprocess osteoclastogenesis ( bloodstream, andthenmigratetobonesforfinal circulatein haematopoietictissues,afterwards inthe precursors,whichstarttheirdifferentiation in turncouldbeconsideredcellcycle-arrestedquiescent markers ofboneformation( of adultsubjectsandwereassociated with biochemical osteoblast-lineage cells werepresent in thecirculation from peripheralblood( could beobtainednotonlyfrombonemarrowbutalso cells (MSC)fromwhichosteoblastsaredifferentiated bone formation,studiesreportthatmesenchymalstem cells, andtheendocrinesystem.Namely, asregards tightly controlled by osteoblasts, , immune of bone remodelling ( insulin ( was inverselyassociatedwithHOMA-IRandserum Clinical Study In ourstudyweconcentratedontheeffectofinsulin Recent studiesfocusedonthecirculatory component 7 , 8 ). 11 9 , , 9 10 12 ). Ithasbeendocumentedthat , , 11 10 13 , ). Circulating monocytes 12 , U Raznyandothers 14 ), which appears to be , 15 ). Taking into index (OGIS)( groups depending on oralglucose insulin sensitivity Subjects includedinthestudyweredividedintotwo criteria. Thus,thestudyincluded82participants. participate and 63 subjects did not meet theinclusion 195 participantsrecruitedforthestudy, 50declinedto or liver function were excluded from the study. Of the mellitus andothermetabolicdisorders,impairedkidney diseases, , andchronicinflammation),diabetes Subjects sufferingfromchronicdiseases(cardiovascular and men, aged 25–65 years were included into the study. Segmental BodyCompositionAnalyserTANITA (Tanita, with the bioelectrical impedance method using the waist-to-hip ratio(WHR)andbodycompositionestimated anthropometrical measurements: body mass index (BMI), values n (control group with OGIS values andobese(25 used. Overweight were measuredbyenzymatic colorimetricmethodson ucOC were measured. Serum glucose and triglycerides free fattyacids(NEFAs), triglycerides(TGs),cOCand collected duringtheOGTTglucose, insulin,non-esterified Diagnostics nv, Sint-Niklaas, Belgium).Inbloodsamples was measured by immunoturbidimetric method (APTEC measured usingELISA(R&DSystemsEurope,Ltd).CRP was calculatedusingtheFriedewaldformula.IL-6 (Maxmat SA,Montpellier, France).LDLcholesterol colorimetric methodsusingtheMaxMatAnalyzer Total and HDL cholesterol were measured by enzymatic markers (CRP,cholesterol and inflammatory IL-6). was usedforbiochemicaltests:totalcholesterol,HDL Serum frombloodsamplescollectedbeforetheOGTT Biochemical tests the test. alcohol ordrinkscontainingcaffeinefor3daysbefore in 250 and 120minaftertheingestionof75 Blood sampleswerecollectedatbaseline(fasting),60, morning hours(8:00–10:00 Federation (IDF)guidelines.Tests wereperformedinthe Organization (WHO)andInternationalDiabetes test (OGTT)conductedaccordingtotheWorld Health a2-horalglucosetolerance Participants underwent Oral glucosetolerancetest Tokyo, Japan). in obesity Bone turnovergeneexpression

= 41) andthegroupwithinsulinresistance(OGIS ≤ mL ofwater. Allpatientswereaskednottodrink 392mL*min 16 ): subjectswithoutinsulinresistance − 1 *m Downloaded fromBioscientifica.com at09/23/202109:31:55PM − 2 , h) after10-hovernight fasting. n

= 41). All subjects underwent 41). All subjects underwent ≤ BMI > 181 392mL*min g glucosedissolved < 40 kg/m :5 2 ) women − 1 526 *m − via freeaccess 2 ,

European Journal of Endocrinology Gene Springversion13(Agilent Technologies). To identify resistance). Microarray data analysis was performed using group, fivesamplesfrom obesesubjectswithinsulin nine sampleswereused(four samplesfromthecontrol (Agilent Technologies). Formicroarrayhybridisation, SureScan MicroarrayScannerusingextractionsoftware of thefluorescencesignalwasperformedbyAgilent hybridisation toahumangenemicroarray. Detection Signal intensity of the labelled cRNA was measured by incorporation weremeasuredusingNanodropND-1000. the manufacturer’s protocol.RNAconcentration and dye was doneusingtheQuickAmplabellingkitaccording to hybridised to a single array. Labelling of 100 from AgilentTechnologies. Eachseparate RNAsamplewas Color 39494humanoligonucleotidearraysandreagents Gene expressionprofilesweredeterminedusingSingle Quantification ofmRNAtranscripts with RIN spectrophotometer (ThermoFisherScientific).Samples usingaNanoDropND-1000UV-VISby spectrophotometry RNA 6000 Nano (Agilent Technologies) and quantified were assessedinanAgilentBioanalyzer2100usingthe RNA/miRNA PurificationKit.RNAandmiRNAquality PAXgene 96BloodRNAKitandGeneMATRIX Universal − room temperatureincubation,sampleswerefrozenat Blood RNATubes (BectonDickinson),followinga2-h before OGTT(fastingsample)wascollectedintoPAXgene For analysisofgeneexpression and miRNAs, venous blood Sample collection,RNAandmiRNAisolation Diagnostics GmbH). enzymatic quantitativecolorimetricmethod(Roche was measuredimmediatelyinnon-frozenplasmaby as thesumofcOCanducOC.NEFAs concentration Shiba, Japan). Total osteocalcin level was calculated Serum cOCanducOCweredeterminedbyELISA(Takara, available onthewebsite was computed using a calculator for Excel spread sheet index (OGIS)proposedbyMari load insulinsensitivity, anoralglucoseinsulinsensitivity of assessment(HOMA-IR)( insulin resistancewasestimatedusinghomeostasismodel (LKB Instruments,MountWaverley, Australia).Basal Louvain-la-Neuve, Belgium)usingthegammacounter immunoradiometric method(DIAsource, ImmunoAssays, the MaxMat Analyzer. Serum insulin was measured by 80°C. Total RNA,miRNAenriched,wasisolatedusingthe Clinical Study ≻ 7.0wereselectedforfurtheranalysis. 17 http://webmet.pd.cnr.it/ogis ). To assesspostoralglucose et al U Raznyandothers . ( 16 ) wasused.OGIS ng of total RNA .

for theaccuratequantitationof754humanmicroRNA Human MicroRNAA+BCardsSetv3.0wereemployed by Megaplex™PreAmpKitandPrimers.TaqMan® Array Human PoolARTprimers.Preamplificationwasenabled TaqMan MicroRNA Reverse Transcription Kit andMegaPlex Scientific). Reversetranscriptionwasperformedwith TaqMan Low-DensityArray(TLDA)wasused(Thermo For relativequantification ofmiRNAbyReal-timePCR, transcripts byTLDA Quantification ofmicroRNA(miRNA,miR) without insulinresistanceandusedasexpressionlevel. in relationtoobesewithinsulinresistanceversus for foldofchange(FC)calculation.FCwascalculated the medianofallsamples.Normalisedsignalwasused (shifted to75percentile) withbackgroundcorrectionto signals wasnormalisedusingmedianshiftalgorithm differentially expressedgeneseachdatasetoffluorescence with Benjamini–Hochberg multipletestcorrection). generated usingstatistical filtering(Moderated obese withoutinsulinresistance with blood ofobesewithimpaired insulinsensitivityversus The listofsignificantly regulatedtranscriptsinwhole Statistical software(StatSoftPolska,Krakow, Poland). significant. Statisticalanalyseswereperformedwiththe Tukey ANOVA. Values atsingletimepointswerecomparedby during OGTTwerecalculatedwithrepeated-measures glucose, insulin, NEFAs, triglycerides and osteocalcins normally distributeddata).Potentialdifferencesinserum by unpaired versus obese without insulin resistance were calculated 75%). Differencesbetweenobesewithinsulinresistance asmedianandlower-upperquartilerange(25– otherwise Normally distributeddataarepresentedasmean Continuous variables were log transformed if required. of homogeneityvariance Levene’s testwasused. analyse dataforaGaussiandistribution.Forverification square ( Analyses ofnominaldatawereperformedusingchi- Statistical analysis were usedasendogenouscontrols. 2 tool andexpressedasRQ miRNA levelswerecalculatedusingDataAssistv3.01 Real-Time PCRsystem(ThermoScientific).Relative by real-timePCR.Thearrayswererunin7900HTFast in obesity Bone turnovergeneexpression ^( − ΔCt )obese without insulin resistance post hoc χ 2 ) test.TheShapiro–Wilk testwasusedto t test. -test orMannWhitney P valueslessthan0.05wereconsidered Downloaded fromBioscientifica.com at09/23/202109:31:55PM . RNU44, RNU48 and U6 snRNA . RNU44, RNU48 and U6 snRNA = 2 ^ ( https://eje.bioscientifica.com − ΔCt ) obesewithinsulinresistance 181 :5 U test(fornon- P

< .5 was 0.05 ± T 527

s -test . e . via freeaccess m / .

European Journal of Endocrinology https://eje.bioscientifica.com WC, waistcircumference;WHR,waist-to-hip ratio. C-reactive ;IL-6,interleukin6; LDL,low-densitylipoprotein;NEFA,non-esterifiedfattyacids;OC,osteocalcin; ucOC,undercarboxylatedosteocalcin; BMI, bodymassindex;BP,bloodpressure; cOC,carboxylatedosteocalcin;HDL,high-densitylipoprotein;HOMA-IR, homeostaticmodelassessment;CRP, non-normally distributedvariables; nominal datawereanalysedb *Significant differencebetweenobese withinsulinresistanceandobesewithoutresistance,(unpaired Total OC(ng/mL) cOC (ng/mL) ucOC (ng/mL) Vitamin D(ng/mL) CRP (mg/L) IL-6 (pg/mL) HOMA-IR Insulin (uIU/mL) Glucose (mmol/L) Triglycerides (mmol/L) NEFA (mmol/L) LDL cholesterol(mmol/L) HDL cholesterol(mmol/L) Total cholesterol(mmol/L) Diastolic BP(mmHg) Systolic BP(mmHg) Adipose tissuemass(%) Males Females WHR Males Females WC (cm) BMI (kg/m Sex, female(%) Age (years) Table 1 in The basalcharacteristicsofparticipantsissummarised Characteristics ofthesubjects Results adipose tissuemass. regarding possibleconfounderssuchassex,age,BMIand without insulinresistancewerenotstatisticallydifferent not adjustedforconfoundersbecauseobesewithand well asregulatedtranscriptsandmiRNAexpressionwere groups. Thecomparisonofbiochemicalmeasurementsas comparison ofthenormalised2 without insulinresistancewascalculatedusing blood betweenobesewithinsulinresistanceversus PathVisio 3software( Analysis ofregulatedpathwayswasperformedusing Clinical Study Table 1 P value for differences in miRNA expression in whole Basal characteristicsofsubjectsincludedinthestudy. 2 . Bothgroupsofsubjects(subjectswithinsulin ) https://www.pathvisio.org U Raznyandothers Obese withinsulinresistance ( − ΔCt ) valuesofthetwo 33.10 (31.12–35.54) 13.24 (10.48–18.64) 38.60 (33.35–41.55) 115 (109.5–121) 2.18 (1.00–3.81) 4.04 (2.41–5.46) 1.02 (0.99–1.04) 0.84 (0.82–0.91) 99 (92.8–106.5) 130 (120–140) 48.32 17.38 0.72 3.50 1.27 5.49 9.03 5.47 3.57 1.62 5.31 1.69 84 (80–90) ). 68 ± ± ± ± ± ± ± ± ± ± ± ± t 0.52 0.29 0.37 0.19 0.08 0.12 0.03 0.14 0.04 0.16 1.63 1.40 -test as y χ

2 test), † ‡ P ( n <

= 41)

to controlgroup( insulin resistancehadhigherserumincomparison glucose between7.8and10.5 in9individualswith2-hserum tolerance wasobserved serum glucose(5.6–6.4 respectively. However, 12 subjects had elevated fasting was abovetheupperreferencelimit:5.55and7.8 these subjectsneitherfastingnor2-hmeanserumglucose (9.08 group of subjects with insulin resistance than controls During OGTT1-hserumglucosewashigherinthe value (4.04(2.41–5.46)vs2.60(2.15–3.67)( vs 5.06 subjects hadhigherfastingserumglucose(5.31 without insulinresistance),theresistance However, incomparisontocontrolgroup(subjects BMI, waistcircumference, WHR andadiposetissuemass. n resistance after glucoseloadbyabout500%andthengraduallyfell 0.05 in obesity Bone turnovergeneexpression

= 41) didnotdiffersignificantlyregardingage,sex, .

† Mean ±

0.38 vs7.90 ± mlL ( 0.08mmol/L Obese withoutinsulinresistance ± n

s

= . e . 41 vsindividualswithoutinsulinresistance m . ; allsuchvalues. 38.80 (33.85–41.70) 32.08 (28.98–34.82) 16.71 (9.36–24.57) P 1.53 (0.63–4.07) 2.60 (2.15–3.67) 0.98 (0.95–0.99) 0.82 (0.80–0.85) ±

= 122 (116–130) 114 (102–116) 96 (90–101.5) 14.50 45.68 0.39mmol/L, 9.25 6.21 3.04 1.33 5.06 1.39 0.78 3.59 1.29 5.50 0.016). Seruminsulinincreased1 81 (76–90) t- P Downloaded fromBioscientifica.com at09/23/202109:31:55PM test orMann–Whitney

80 = ± ± ± ± ± ± ± ± ± ± ± ± mmol/L). Impairedglucose 0.018)) andhigherHOMA-IR 0.51 0.42 0.28 0.12 0.08 0.11 0.04 0.16 0.03 0.17 1.20 1.88 ‡ Median (25–75%);allsuchvalues. mmol/L. Subjectswith P .0) ( =0.001) 181 ( n

= 41) :5 U testfor Fig. 1A P =0.027)). > mmol/L, 0.291 0.137 0.446 0.207 0.919 0.224 0.027 0.088 0.018 0.049 0.432 0.670 0.646 0.945 0.438 0.298 0.954 0.024 0.113 0.237 0.281 0.124 0.05 0.292 P ± 528 * 0.08 ). In via freeaccess h European Journal of Endocrinology mass inbothgroupsofpatients ( connected withsimilarBMI, WHRandadiposetissue 1.53 (0.63–4.07)mg/L, 1.33 markers:IL-6(1.62 in seruminflammatory significant differencesbetweenthestudiedgroups ( of subjectswithinsulinresistancethancontrols triglycerides concentration was higher in the group ( of thetestincomparisontobaselinevalues( groups plasmaNEFAs graduallydecreasedafter1and2 between thestudiedgroups( load, wefoundnostatisticallysignificantdifferences P except for serum triglycerides (1.69 LDL, HDL and total cholesterol between studied groups There werenostatisticallysignificantdifferencesinserum and approximately by 50% in the control group ( after 2 follows: *** Tukey and time(ab).Valuesatsinglepointswerecomparedby time (b)anddifferencesowingtointeractionbetweengroup without insulinresistanceparticipants(a),differencesover differences betweenobesewithinsulinresistanceand were performedwithrepeated-measuresANOVAanddenote test. Dataarepresentedasmeans (NEFAs) (C)andtriglycerides(D)duringoralglucosetolerance Blood levelsofglucose(A),insulin(B),non-esterifiedfattyacids Figure 1 BD AC P Fig. 1C =0.049). Clinical Study 0.034) ( =0.034) As regardsplasmaNEFAs inresponsetoglucose ± post hoc 0.19pg/mL, h by about 20% in the group with insulin resistance ). However, wefoundthatduringOGTTserum P Fig. 1D

< tests.Significantdifferencesareindicatedas 0.001. ). We anystatistically did notobserve P

= 0.224), CRP(2.18(1.00–3.81) vs P

= 0.919), whichprobablywas P

± = U Raznyandothers

0.309). Inbothstudied s . Table 1 e . m ± . Statisticalanalyses

0.12 vs 1.39 ). P ±

< .9 vs 0.19 Fig. 1B 0.001) ± 0.11, h ). Figure 2 * tests. Significantdifferencesareindicatedasfollows: Values atsingletimepointswerecomparedbyTukey differences owingtointeractionbetweengroupandtime(ab). resistance participants(a),differencesovertime(b)and obese withinsulinresistanceandwithout repeated-measures ANOVAanddenotedifferencesbetween means (C) duringoralglucosetolerancetest.Dataarepresentedas undercarboxylated osteocalcin(ucOC)(B)andtotal Blood concentrationsofcarboxylatedosteocalcin(cOC)(A), differences were observed inserumucOCduringOGTT differences wereobserved glucose load( and 2 reduced at1 subjects withinsulinresistance ( osteocalcin concentrationincontrolgroupbutnot in with insulinresistance.SimilartoserumcOC,total baseline values.Thiseffectwassuppressedinsubjects vs 6.21 vs 6.21 serum cOCgraduallydecreasedafter1 2A was differentinrespecttoserumcOC( P in both studied groups (5.47 Althoughbaselineserumosteocalcin(cOC)wassimilar osteocalcin duringOGTT Effects ofinsulinresistanceinobesityonserum in obesity Bone turnovergeneexpression AB P .3) ( =0.139)

< ). Surprisingly, in subjects without insulinresistance 0.05, ** h (7.53 ±

± s ± . e 0.19ng/mL, 0.19ng/mL, . m Table 1 P . Statisticalanalyseswereperformedwith

< h (8.17 Fig. 2C ± 0.01.

0.43 vs9.25 C ), theresponsetooralglucoseload ). However, nostatisticallysignificant ±

0.44 vs9.25 P Downloaded fromBioscientifica.com at09/23/202109:31:55PM

P <

<

0.01) ofOGTTcomparedto

0.05) and 2 ± ± 0.44ng/mL,

0.19 vs 6.21 https://eje.bioscientifica.com P .4) a gradually was =0.041) ± 0.44ng/mL, 181 :5 h (5.06 P h (5.41 P ± .2) ( =0.028)

< 0.19ng/mL, .1 after 0.01)

post hoc P

< ± ± 529 0.01) 0.32 0.32 Fig. via freeaccess

European Journal of Endocrinology https://eje.bioscientifica.com alpha collagen( ( of alkalinephosphatase( regulated inobesesubjects. We founddownregulation of osteoblastsdifferentiation wasstatisticallysignificantly expression inwholebloodofgenesbeingearlymarkers in comparisontosubjectswithoutinsulinresistance.The differentially regulatedinobesewithinsulinresistance signalling differentiation and maturation were also molecules: co-stimulatory CRP resistance alsorevealedupregulation of genescodingfor Analysis ofgeneexpressioninsubjectswithinsulin IFNA21 cytokineswasalsoupregulated: inflammatory CXCL10 individuals withoutinsulinresistance: increased insubjectswithinsulinresistanceversus of genesinvolvedintheregulationchemotaxiswere was upregulated: response /antigens takingpartininflammatory response, wefoundthatexpressionofgenescodingfor factor ( participatinginPDK/AKTsignallingpathway Isoforms ( downregulation of receptors for IGF1 ( to subjectswithoutinsulinresistance,theyhad pathway genesexpressioninwholeblood.Incomparison to controlgrouppresentedinhibitionofinsulinsignalling response. insulin signalling,cytokinesandinflammatory vitamin Dreceptorpathway, TGFbsignallingpathway, signalling, osteoblast signalling, Wnt Signalling pathway, subjects differregardingfollowingpathways:osteoclast PathVisio 3softwarerevealed thattwogroupsofstudied Table 2 obese subjectswithinsulinresistancearepresentedin Results ofgeneexpressionanalysisinwholeblood differentiation the regulationofosteoblastsandosteoclasts expression ofmRNAandmiRNAparticipatingin Effects ofinsulinresistanceinobesityonthe ( reduced serumucOClevelsduring1 values, bothgroupsofsubjectspresentedgradually between thestudiedgroups.Incomparisontobaseline P PDPK1

Clinical Study < .0) ( 0.001) , The mRNAlevelsofgenesresponsibleforosteoblasts Obese subjectswithinsulinresistanceincomparison CSF1 FOXO1 , , . Analysisofregulatedpathwaysperformedby , IFNA14 AKT2, GSK3A,GSK3B CXCL9 (MCP1),receptorforIL-1( PIK3CD, PIK3CG, PIK3C3, PIK3R1, PIK3R2 Fig. 2B . As regards cytokines and inflammatory . Asregardscytokinesandinflammatory SPP1 , COL1A1 , IFNB1 ). IL6 ), bonesialoprotein( CXCL11 , IL12B , ) inobesesubjectswithinsulin IFNA16 CD40 ALPL . Expression of otherpro- , IL12A ) aswelltranscription , U Raznyandothers ) andupregulationof CD80. , IFNA6 . ThemRNAlevels h and2 IGF1R IL1R1 , CCL2 IFNA7 IBSP ), P13-Kinase ) aswell h ofOGTT ), type1 , , IFNA10 CXCL8 IFNB1 ), . , , group, insulin-resistantsubjects haddecreasedexpression antagonist expression ( osteoblastic differentiation due to upregulation of BMP was alsodownregulated,which couldnegativelyaffect obese subjects with insulinresistance, miR-181a expression disturbances inbonemicrostructure( in subjectswithtype2diabetesandisassociated demonstrated thatmiR-324-3pexpressionisinhibited compared tovolunteerswithoutinsulinresistance. It was miR-324-3p inobesesubjectswithinsulinresistance Table 3 osteoblasts andosteoclastsdifferentiationisshown in (TRAP) genes, suchastartrate-resistantacidphosphatase( expression of inhibitionof we observed Although geneexpressionof TNFRSF11B osteoblast differentiation( resistance. Expression of key transcription factors in family weredownregulatedinobesesubjectswithinsulin and itsnuclearpartners and upregulation ofWntsignallingagonists ACVR2A DKK1 However, theexpressionofWntsignallingantagonists: TGFB ( and also revealedupregulationofBMPantagonist in insulin resistance subjects. Analysis of gene expression BMP7 but mRNAforbonemorphogenicproteins( downregulation of which modulateosteoblastdifferentiation,weobserved was unchanged(datanotshown).Amongfactors However, expressionratioof found upregulatedexpressionof resistance. AsregardsRANKL/RANK/OPGpathway, we significantly upregulatedinobesesubjectswithinsulin signallingdifferentiationwerestatistically without insulinresistance. of subjectswithinsulinresistanceascomparedto differentiation factor was upregulated.Expressionofanotherosteoblast expression ofRUNX2degradationproteingene( DLX5 resistance. The expression of osteocalcin gene ( in obesity Bone turnovergeneexpression Expression ofmiRNAsparticipatinginregulation The mRNAlevelsofgenesparticipatingin FZD TGFB3 ) andthereceptorforPTH( , , , TGFBR2 TWIST1 DKK2 ITGA5 and . We foundsignificantdownregulation of andtheirreceptor ()and aswellinhibitionofthereceptorfor , ), wasdownregulated. ) inobesesubjectswithinsulinresistance. NFATC1 and WIF1 DKK4 TGFB1 TWIST2 wasupregulated.We alsofound , sclerostin( ATF4 as well as regulated osteocytes’ Downloaded fromBioscientifica.com at09/23/202109:31:55PM GREM2 LEF-1/TCF4 , receptorforvitaminD( wasinhibitedinwholeblood RUNX2 SPI1 (PU.1). ) wasupregulated.However, CSF1 (MCP1) TNSF11/TNFRSF11B LRP5 PTH1R ). Contrary to control ). Contrary SOST , osterix . However, transcriptionfactor 181 TNFSF11A (RANK), 18 TNSF11 (RANKL) ) wereupregulated ), WNT3A :5 , Moreover, gene 19 SFRP1 wasincreased, ). Inbloodof BMP2 SP7 , , SMURF2 CTNNB1 , WNT5A BGLAP , was1. GREM2 DUSP2 MSX2 BMP4 VDR ACP5 530 via freeaccess ), ) ) . , , ,

European Journal of Endocrinology

Table 2 The differentially expressed genes of whole blood involved in pathways related to bone remodelling in obese subjects with insulin resistance. Significantly Clinical Study regulated transcripts (NCBI symbols) with P < 0.05 in whole blood of obese with insulin resistance versus obese without insulin resistance was generated using Moderated T-test with Benjamini–Hochberg multiple test correction.

Gene symbol Gene name FC P (corr) Ensembl ID Gene ID GenBank accession Ref Seq accession Insulin signalling pathway AKT2 v-Akt murine thymoma viral oncogene homolog 2 −2.75 0.021 ENST00000476266 208 NM_001626 NM_001626 FOXO1 Forkhead box O1 −2.94 0.019 ENST00000379561 2308 NM_002015 NM_002015 GSK3A Glycogen synthase kinase 3 alpha −2.03 0.021 ENST00000222330 2931 NM_019884 NM_019884 GSK3B Glycogen synthase kinase 3 beta −2.44 0.021 ENST00000473886 2932 NM_002093 NM_002093 IGF1R Insulin-like 1 receptor −2.72 0.021 ENST00000268035 3480 NM_000875 NM_000875 PDPK1 3-Phosphoinositide dependent protein kinase 1 −2.29 0.023 ENST00000342085 5170 NM_002613 NM_002613 PIK3C3 Phosphatidylinositol 3-kinase catalytic subunit type 3 2.10 0.036 ENST00000587328 5289 NM_002647 NM_002647 − U Raznyandothers PIK3CD Phosphatidylinositol-4.5-bisphosphate 3-kinase catalytic −2.10 0.022 ENST00000377346 5293 NM_005026 NM_005026 subunit delta PIK3CG Phosphatidylinositol-4.5-bisphosphate 3-kinase catalytic −2.33 0.045 ENST00000496166 5294 NM_002649 NM_002649 subunit gamma PIK3R1 Phosphoinositide-3-kinase regulatory subunit 1 (alpha) −2.60 0.021 ENST00000521381 5295 NM_181523 NM_181523 PIK3R2 Phosphoinositide-3-kinase regulatory subunit 2 (beta) −2.14 0.021 ENST00000222254 5296 NM_005027 NM_005027 Inflammatory molecules pathways CCL2 (C-C motif) ligand 2 2.31 0.021 ENST00000582017 6347 NM_002982 NM_002982 CRP C-reactive protein pentraxin-related 2.44 0.021 ENST00000368111 1401 NM_000567 NM_000567 CSF1 Colony-stimulating factor 1 () 2.38 0.021 ENST00000369801 1435 NM_172210 NM_172210 CD40 CD40 molecule. TNF receptor superfamily member 5 1.76 0.040 ENST00000477696 958 NM_001250 NM_001250 CD80 CD80 molecule 2.68 0.035 ENST00000264246 941 NM_005191 NM_005191 in obesity Bone turnovergeneexpression CXCL8 Chemokine (C-X-C motif) ligand 8 2.88 0.021 ENST00000307407 3576 NM_000584 NM_000584 CXCL9 Chemokine (C-X-C motif) ligand 9 2.24 0.030 ENST00000264888 4283 NM_002416 NM_002416 CXCL10 Chemokine (C-X-C motif) ligand 10 2.02 0.034 ENST00000306602 3627 NM_001565 NM_001565 CXCL11 Chemokine (C-X-C motif) ligand 11 2.23 0.021 ENST00000306621 6373 NM_005409 NM_005409 IFNA6 alpha 6 2.22 0.021 ENST00000259555 3443 NM_021002 NM_021002 IFNA7 Interferon alpha 7 2.50 0.048 ENST00000239347 3444 NM_021057 NM_021057 IFNA10 Interferon alpha 10 2.26 0.021 ENST00000357374 3446 NM_002171 NM_002171 IFNA14 Interferon alpha 14 2.08 0.031 ENST00000380222 3448 NM_002172 NM_002172 IFNA16 Interferon alpha 16 1.99 0.036 ENST00000380216 3449 NM_002173 NM_002173 IFNA21 Interferon alpha 21 2.19 0.021 ENST00000380225 3452 NM_002175 NM_002175 Downloaded fromBioscientifica.com at09/23/202109:31:55PM IFNB1 Interferon beta 1 2.16 0.021 ENST00000380232 3456 NM_002176 NM_002176 IFNB1 Interferon beta 1 fibroblast 2.16 0.021 ENST00000380232 3456 NM_002176 NM_002176 IL1R1 1 receptor type I 2.02 0.024 ENST00000413623 3554 NM_000877 NM_000877 https://eje.bioscientifica.com IL6 3.99 0.021 ENST00000406575 3569 XM_005249745 XM_005249745 IL12A Interleukin 12A 1.99 0.048 ENST00000496308 3592 NM_000882 NM_000882 IL12B Interleukin 12B 2.19 0.021 ENST00000231228 3593 NM_002187 NM_002187 181

Osteoblasts differentiation pathways :5 ALPL liver/bone/ −2.27 0.029 ENST00000374832 249 NM_000478 NM_000478 ATF4 Activating transcription factor 4 −2.38 0.021 ENST00000337304 468 NM_001675 NM_001675 ACVR2A Activin A receptor type IIA 2.18 0.021 ENST00000495775 92 NM_001616 NM_001616 BMP2 Bone morphogenetic protein 2 2.34 0.021 ENST00000378827 650 NM_001200 NM_001200

BMP4 Bone morphogenetic protein 4 2.25 0.021 ENST00000558984 652 NM_001202 NM_001202 531

via freeaccess BMP7 Bone morphogenetic protein 7 2.26 0.021 ENST00000395863 655 NM_001719 NM_001719

(Continued) European Journal of Endocrinology https://eje.bioscientifica.com

Table 2 Continued. Clinical Study

Gene symbol Gene name FC P (corr) Ensembl ID Entrez Gene ID GenBank accession Ref Seq accession COL1A1 Collagen type I alpha 1 1.87 0.048 ENST00000225964 1277 NM_000088 NM_000088 CTNNB1 Catenin (cadherin-associated protein) beta 1 88kDa −2.49 0.021 ENST00000396185 1499 NM_001904 NM_001904 DKK1 Dickkopf WNT signalling pathway inhibitor 1 2.10 0.022 ENST00000373970 22943 NM_012242 NM_012242 DKK2 Dickkopf WNT signalling pathway inhibitor 2 2.25 0.021 ENST00000285311 27123 NM_014421 NM_014421 DKK4 Dickkopf WNT signalling pathway inhibitor 4 2.48 0.025 ENST00000220812 27121 NM_014420 NM_014420 DLX5 Distal-less homeobox 5 2.18 0.021 ENST00000222598 1749 NM_005221 NM_005221 DUSP2 Dual specificity phosphatase 2 1.95 0.032 ENST00000488952 1844 NM_004418 NM_004418 FZD3 Frizzled class receptor 3 2.28 0.040 ENST00000537916 7976 NM_017412 NM_017412 FZD4 Frizzled class receptor 4 2.23 0.021 ENST00000531380 8322 NM_012193 NM_012193 FZD5 Frizzled class receptor 5 2.27 0.021 ENST00000295417 7855 NM_003468 NM_003468

FZD6 Frizzled class receptor 6 2.50 0.022 ENST00000522566 8323 NM_003506 NM_003506 U Raznyandothers FZD7 Frizzled class receptor 7 2.21 0.021 ENST00000286201 8324 NM_003507 NM_003507 FZD8 Frizzled class receptor 8 2.17 0.021 ENST00000374694 8325 NM_031866 NM_031866 FZD9 Frizzled class receptor 9 2.60 0.021 ENST00000344575 8326 NM_003508 NM_003508 FZD10 Frizzled class receptor 10 2.22 0.021 ENST00000229030 11211 NM_007197 NM_007197 GREM2 Gremlin 2: DAN family BMP antagonist 5.08 0.021 ENST00000318160 64388 NM_022469 NM_022469 IBSP -binding sialoprotein 2.20 0.021 ENST00000226284 3381 NM_004967 NM_004967 LEF1 Lymphoid enhancer-binding factor 1 −2.64 0.021 ENST00000503879 51176 NM_016269 NM_016269 LRP5 Low-density lipoprotein receptor-related protein 5 3.02 0.021 ENST00000529993 4041 NM_002335 NM_002335 MSX2 msh homeobox 2 2.30 0.021 ENST00000239243 4488 NM_002449 NM_002449 PTH1R 1 receptor 2.18 0.021 ENST00000427125 5745 NM_000316 NM_000316 RUNX2 Runt-related transcription factor 2 2.11 0.031 ENST00000371436 860 NM_001015051 NM_001015051

SFRP1 Secreted frizzled-related protein 1 2.29 0.021 ENST00000220772 6422 NM_003012 NM_003012 in obesity Bone turnovergeneexpression SMAD4 SMAD family member 4 −4.07 0.021 ENST00000342988 4089 NM_005359 NM_005359 SMURF2 SMAD specific E3 ubiquitin protein ligase 2 2.26 0.021 ENST00000582081 64750 NM_022739 NM_022739 SOST Sclerostin 2.19 0.021 ENST00000301691 50964 NM_025237 NM_025237 SP7 2.13 0.022 ENST00000303846 121340 NM_152860 NM_152860 SPP1 Secreted phosphoprotein 1 2.08 0.022 ENST00000360804 6696 NM_001040058 NM_001040058 TCF4 Transcription factor 4 −2.28 0.021 ENST00000570177 6925 NM_003199 NM_003199 TGFB1 Transforming growth factor beta 1 −2.61 0.021 ENST00000221930 7040 NM_000660 NM_000660 TGFB3 Transforming growth factor beta 3 2.21 0.021 ENST00000238682 7043 NM_003239 NM_003239 TGFBR2 Transforming growth factor beta receptor II (70/80kDa) −1.84 0.048 ENST00000295754 7048 NM_001024847 NM_001024847

Downloaded fromBioscientifica.com at09/23/202109:31:55PM TWIST1 Twist family bHLH transcription factor 1 3.05 0.021 ENST00000354571 7291 XR_428085 XR_428085 TWIST2 Twist family bHLH transcription factor 2 2.24 0.021 ENST00000448943 117581 NM_057179 NM_057179 VDR Vitamin D (1.25- dihydroxyvitamin D3) receptor −2.04 0.026 ENST00000229022 7421 NM_001017535 NM_001017535 WIF1 WNT inhibitory factor 1 2.27 0.023 ENST00000286574 11197 NM_007191 NM_007191 WNT3A Wingless-type MMTV integration site family. member 3A 2.26 0.021 ENST00000284523 89780 NM_033131 NM_033131 WNT5A Wingless-type MMTV integration site family. member 5A 2.26 0.021 ENST00000264634 7474 NM_003392 NM_003392 Osteoclasts differentiation pathways 181 ACP5 Acid phosphatase 5 tartrate resistant (TRAP) −1.80 0.040 ENST00000592828 54 NM_001611 NM_001611 :5 ITGA5 Integrin alpha 5 (fibronectin receptor alpha polypeptide) −2.41 0.021 ENST00000552564 3678 NM_002205 NM_002205 NFATC1 Nuclear factor of activated T-cells cytoplasmic calcineurin- −2.30 0.021 ENST00000329101 4772 NM_172387 NM_172387 dependent 1 SPI1 Spi-1 proto-oncogene −2.40 0.021 ENST00000227163 6688 NM_001080547 NM_001080547 TNFRSF11A Tumour necrosis factor receptor superfamily member 11a. 2.27 0.021 ENST00000616710 8792 NM_003839 NM_003839 532

via freeaccess NFKB activator TNFRSF11B Tumour necrosis factor receptor superfamily member 11b 2.13 0.021 ENST00000521597 4982 NM_002546 NM_002546 TNFSF11 Tumour necrosis factor (ligand) superfamily member 11 2.10 0.027 ENST00000358545 8600 NM_033012 NM_033012 European Journal of Endocrinology normalised 2 subjects wascalculatedusingStudent’s obese withinsulinresistanceandwithout P hsa-miR-324-3p hsa-miR-210 hsa-miR-181a hsa-miR-29b miRNA without insulinresistanceindividuals. profile inobesesubjectswithinsulinresistanceversus Table 3 pathway defects.We downregulationof observed with insulinresistanceconfirmed insulinsignalling of geneexpressioninwhole bloodfromobesesubjects triglycerides incomparison to thecontrolgroup.Analysis basal andpost-glucoseload seruminsulin,glucoseand pre-diabetes. Obeseinsulin-resistantsubjectshadhigher glucose 7.8–10.5 subjects (22%)hadimpairedglucosetolerance(2-hserum had impairedfastingglucose(5.6–6.4 resistance hadnodiabetesbut12of41subjects(29%) higher HOMA-IRvalues.Studiedsubjectswithinsulin of insulinresistancedeterminedbylowerOGISand insulin resistance),thestudygrouphadahigherdegree peripheral circulation ( ( numbers whichcorrelatewithmarkersofboneformation circulate inperipheralblood inphysiologicallysignificant taken intoconsideration.Namely, osteoblast-lineagecells circulatory component ofboneremodellingshouldbe with boneturnoverisperformedinwholeblood, However, whenanalysisofgeneexpressionconnected from wholebloodresidesincirculating bloodleukocytes. in wholeblood.ItiswellknownthatRNAinformation resistance. Analysisofgeneexpressionwasperformed remodelling inobesesubjectswithandwithoutinsulin and microRNAinwholeblood,associatedwithbone The studyisthefirstonefocusingonbothmRNA Discussion inhibitors ( due toincreasedexpressionofosteoblastdifferentiation which couldcontributetoosteogenesisdisturbances Moreover, wefounddecreasedmiR-29bexpression, osteoblast proliferationanddifferentiationwereinhibited. of miR-210 in whole blood, which can indicate that 10 valuefordifferencesinmiRNAexpressionwholebloodbetween Clinical Study ). Osteoclastprecursorsarealsodetectableinthe In comparisontothecontrolgroup(subjectswithout Relative quantification(RQ)ofmiRNAexpression (–ΔCt TGFB3 ) valuesofthetwogroups. mmol/L), together 17 subjects(41%)had , ACVR2A 11 ). , DUSP2 t -test ascomparisonofthe U Raznyandothers 0.467 0.458 0.297 0.474 RQ ) ( 20 mmol/L) andnine ). P 0.012 0.035 0.009 0.046 value IGF1R , signalling pathway agonists ( upregulationofWNT precursors. Althoughweobserved in the commitment of early progenitors to osteoblast via mediatingWNTsignal transduction, plays a keyrole regulated by1 inhibit itsfunction. which interactswiththeRUNX2 DNA-bindingdomainto antiosteogenic functionis mediated byaTWISTbox, disappears inosteoblastprecursors( expression occursonlywhenof al et SMURF2 ( concentration couldbedecreasedbyupregulation of in obese subjects with insulin resistance, its protein in osteoblastdifferentiation– osteoblasts ( decreased boneformationrate,andareducednumber of to lowerbonemineraldensity, decreasedbonevolume, that conditionaldeletionof with insulinresistance.Ithasbeenreportedpreviously SMAD4 such as of geneexpressionforBMPspromotingboneformation, theupregulation gene expression.Althoughweobserved FZD found downregulation of response.Indeed,we participating ininflammatory BMP signallingpathwayandupregulationofgenes inhibiting expression ofgenesconnected with Wnt, in insulinsignallingcouldaffectosteoblastogenesisby ( propeptide tobonealkalinephosphatase(PINP/BAP) defined asquotientofprocollagentype1amino-terminal less collagenaswellhadlowerboneformationindex insulin resistantsubjectswerelessactiveandproduced markers. serum inflammatory did notdifferregardingBMI,adiposetissuemassaswell signalling pathway. Itshouldbestressedthatbothgroups P13-kinase isoforms,genesparticipatinginPDK/AKT in consequenceleadstothedownregulationof (DKK) families ( by SOST(sclerosteosisgeneproduct)( pathway. In turn LRP5/6 co-receptor activity is inhibited factor1(WIF1),preventingactivationofthe inhibitory with antagonists,including(SFRP)familyandWNT WIF-1 this pathwayantagonistswasalsoupregulated( in obesity Bone turnovergeneexpression 21 ). Inourstudieswealsoreportedthatdisturbances Transcriptional activityofRUNX2mightalsobe Although gene expression of key transcription factor Some studiesreportedthatosteoblastsfromobese ) aswelltheirreceptor . showedthat RUNX2-induced osteoblast gene , inBMPsignallingwasdownregulatedsubjects SOST BMP2 27 26 ) aswellTWIST1andTWIST2( , , DKK1 ). α BMP4 ,25-dihydroxy vitaminD3through 23 , , 24 , DKK2 BMP6 ), all of which bind LRP5/6, that Downloaded fromBioscientifica.com at09/23/202109:31:55PM ). WNTligandscaninteract ( β SMAD4 -catenin ( 25 https://eje.bioscientifica.com ), thegeneexpressionof RUNX2 LRP5 WNT3A inosteoblasts,leads 29 181 , theexpressionof ). TWISTproteins’ 22 wasupregulated CTNNBI :5 , ) andDickkopf WNT5A TWIST 28 β ), which catenin ). Bialek genes and 533 SFRP VDR via freeaccess . ,

European Journal of Endocrinology https://eje.bioscientifica.com influence osteoblastsdifferentiation. insulin resistance with insulinresistanceincomparison tosubjectswithout CXCL10 responsethatis taking partininflammatory on proteinlevelsinserumatfasting,expressionofgenes cytokines increasesinpro-inflammatory did notobserve contribute toboneturnoveralterations.Althoughwe responseandinsulin resistancecould in inflammatory pre-diabetes anddiabetes( β influence insulinsensitivity, whereasucOC couldaffect hand, Pollock was reportedinthestudybySchwetz( insulin resistanceincomparisontothecontrolgroupas suppression ofucOCdecreaseinobesesubjectswith increased boneformation.However, wedidnotobserve blood of subjects without insulin resistance could suggest released intocirculation) itsdecreaseduringOGTTin fact that cOC accumulates mainly in bone (only 15% is oral glucoseload( suppressed in comparison to non-obese volunteers after that inobesesubjectsthedecreaseserumcOCwas in agreementwithrecentlypublishedresultsindicating markers: totalandcarboxylatedosteocalcin.Suchdataare load withasuppresseddecreaseinserumboneformation subjects withoutinsulinresistance,respondedtoglucose of osteoprogenitorsinthecirculation. with insulinresistance,whichmayconfirmthepresence sialoprotein andcollagenwereupregulatedinsubjects differentiation towardsosteoblastsasosteopontin,bone ( osteocalcin isalatemarkerofosteoblastsdifferentiation regulated, which could be explained by the fact that data showedthatosteocalcingeneexpressionwasnot could inconsequence affect bone remodelling. Our resistance incomparisontothecontrolgroup,which inthegroupwithinsulin gene expressionwasobserved between studiedgroups.However, adecreasein mL), andtherewerenostatisticallysignificantdifferences vitamin Dconcentration(belowtargetrange:30–80 osteocalcin ( induces osteoblast-specificgenesexpressionsuchas transcriptional complexesatspecificpromotersand and interactionbetweenVDRRUNX2stabilises are co-expressed in developing skeletal elements, It wasreportedpreviouslythatRUNX2andVDR 31 - function and seems to be important in subjects with Clinical Study ). However, theexpressionofsomemarkerscell We postulatethatafternutrientsingestionan increase to Obese subjectswithinsulinresistancecontrary In ourstudy, bothgroupsofsubjectshadlowserum , CXCL9 Ocn t al et ) ( , . Suchgeneexpressionpatterns could 30 32 CXCL11 . reportedthatinhumans,cOCmay ). , 33 ). Taking intoconsiderationthe 35 wasupregulatedinsubjects ). U Raznyandothers 34 In vitro ). Ontheother studies CXCL8 VDR ng/ ,

mass. Indeed,wefoundincreasedexpressionof The RANKL/OPGratioisthereforeadeterminantofbone to RANKLandpreventingitfrominteractingwithRANK. the skeletonfromexcessiveboneresorptionbybinding osteoblasts andosteogenicstromalstemcellsprotects activation. Osteoprotegerin(OPG,TNFSF11)issecretedby of precursorsintomultinuclearosteoclastsandosteoclast osteoclasts andtheirprecursorsregulatesthedifferentiation Binding RANKL to its receptor, RANK, on the surface of Wnt/ by strikingchangesintheexpressionofregulators pathway. Resolutionofinflammationwasaccompanied cytokinesdisrupttheWNTsignalling Pro-inflammatory TNFRSF11B by modulatingtheRANK/RANKL/OPGsystem( cytokinescanstimulateosteoclastogenesis inflammatory maturation byinhibitingtheexpressionof tumour necrosisfactor cytokinessuchas reported thatpro-inflammatory of unrestrictedsomaticstem cellsfromhumancord ( and DUSP2(dualspecificity proteinphosphatase2) for exampleTGFB3(transforming growthfactorbeta) signal cascadetranscription programmes,including protein expression of negative regulators of the cellular miR-29b stimulatesosteoblastogenesisbyinhibiting osteoblast differentiation. expression ofmiR-29b,whichhasbeenshowntoinduce either translationalrepressionordegradation. mRNA transcriptspost-translationally, they provoke via the RNA interference pathway ( negatively regulate gene expression and cellular processes endogenous small,non-codingRNAsthatpositively or gene expression patterns. MicroRNAs (miRs/miRNAs) are obese subjectswithinsulinresistanceappearstoconfirm stimulate osteoclastogenesisandboneresorption. leads toinhibitionofosteoblastogenesisbutdoesnot These resultssuggestthatinsulinresistanceinobesity nuclear factorofactivatedT-cells cytoplasmic1( and themasterregulatorofosteoclastdifferentiation – is notactivatedinthesesubjects.Moreover, was equalto1,whichsuggeststhatosteoclastogenesis resistance. However, expressionratio genes including ( downregulated in obesesubjects with insulinresistance which regulateanumberofosteoclast-specificgenes,were in obesity Bone turnovergeneexpression 20 38 ). Mir-29balsostimulates osteogenic differentiation ). We thatexpressionofosteoclast-specific alsoobserved Data frompreviousstudiesrevealedthatpro- Analysis of miR expression revealed decreased Analysis ofmiRNAexpressioninwholeblood β -catenin pathway( and ACP5 (TRAP) RANK in obese subjects with insulin Downloaded fromBioscientifica.com at09/23/202109:31:55PM α 37 (TNFA) suppressedosteoblast ). In vitro , ITGA5 studiesshowedthat 39 was downregulated. 181 TNFSF11/TNRSF11B , :5 40 ). By targeting RUNX2 PU.1 (SPI1) TNFSF11 NFATC1 ( 534 27 36 via freeaccess ). ). ), ,

European Journal of Endocrinology (C-terminal telopeptideoftype Icollagen(CTX))markers alkaline phosphatase (BSAP)) as well as bone resorption propeptide oftypeIprocollagen, PINP)andbone-specific bone formation(productsof collagen synthesis(N-terminal mineralisation. To reflectboneturnoverothermarkersof marker ofboneformationandincreasesatlatestages of produced byosteoblasts.Thisproteinisconsideredalate – osteocalcin, which isanon-collagenous protein OGTT we measured one marker of bone remodelling the limitationsofstudywasalsofactthatduring groups ofsubjectsinthestudy(datanotshown).One of results ofbiochemicalmeasurements,thewholetwo subjected tomoleculartestsreflected,inrelationthe than biochemical parameters. However, the participants limitations wereanalysedinasmallergroupofsubjects and miRNAgeneexpressionpatternsduetofinancial the number of both sexes was unequal. Secondly, mRNA number ofsubjectsincludedinthestudywassmalland However, ourstudyhadsomelimitations.Firstly, the in comparisontoindividualswithoutinsulinresistance. during OGTTinobesesubjectswithinsulinresistance on proteinlevelsbysuppressionofserumcOCdecrease differentiation. Defectsinboneremodellingarereflected expression participatinginregulationofosteoblasts osteoblastogenesis antagonistsandinhibitionofmiRNA in osteoblastogenesis,increasedmRNAexpressionof via mRNAdownregulationforkeysignallingmolecules in osteoblastogenesissubjectswithinsulinresistance ( microstructure andhistomorphometry revealed positiveassociationofmiR-324-3pwithbone mineral densityatthelumbarspine( circulating miR-324-3p levels correlated with areal bone with insulinresistance.Recentstudiespresentedthat which couldsuggestboneformationdefectsinsubjects blood fromobesesubjectswithinsulinresistance. ( (AcvR1b), whichinhibitsTGF- positive regulationoftheactivinAreceptortype-1B differentiation of murine stromal BM-MSC ST2 cells by of metabolicsyndromecomponents( miR-181a expressionwasassociatedwithahighernumber monocytes ofleanpersons.Moreover, downregulationof Recent studies presented overexpression of miR-181ain insulin resistanceinobesityalsoindicatesbonedefects. 29b butalsoupregulationof blood ( 44 Clinical Study ). However, miR-210expressionwasinhibitedinwhole Summing up,resultsofourstudiessuggestdefects We downregulation of miR-324-3p, also observed Transfection ofsensemiR-210-stimulatedosteoblastic Inhibition ofmiR-181aexpressioninsubjectswith 41 ). We foundnotonlydownregulationofmiR TGFB3 β /activin signallingpathway U Raznyandothers and 42 18 , 19 DUSP2 ). Anotherstudy 43 ). ). mRNA. BOLIS (rn areet ubr 495; h Ntoa Science National the Metabolic 244995); number of agreement (grant Robustness (BIOCLAIMS) Food” of on Made CLAIMS Health Nutrigenomics—derived for “BIOmarkers Homeostasis Seventh its Programme through Commission Framework European supported was work This Funding be could that interest of conflict perceived asprejudicingtheimpartialityofresearchreported. no is there that declare authors The Declaration ofinterest defects indifferentdiseasestates. be a basis for further studies evaluating bone remodelling about circulating componentsofboneturnoverandcould that ourresultscouldcontributetoexpandknowledge reflect bonemicrostructure defects. However, webelieve blood ofobesesubjectswithinsulinresistanceindeed osteoblastogenesis geneandmiRexpressioninwhole It couldbeinterestingtoconfirmthatinhibitionof in thestudydidnothavebonemicrostructureanalysed. should bemeasured.Lastbutnotleast,subjectsincluded References as wellassistanceinlanguageandtextediting. The authors thank Monika Chojnacka for technical support in methodology Acknowledgements Polish and -2011/02/A/NZ2/00022) Ministry ofScienceandHigherEducationgrantnumber(K/ZDS/006287) DEC number grant (Poland, Centre in obesity Bone turnovergeneexpression 7 6 5 4 3 2 1 Razny U, Fedak D,Kiec-Wilk B, Goralska J,Gruca A,Zdzienicka A, Ferron M, Hinoi E,Karsenty G&Ducy P. Osteocalcindifferentially Ferron M &Lacombe J.Regulationofenergymetabolismby Lee NK, Sowa H,Hinoi E,Ferron M,Ahn JD,Confavreux C, Price PA, Rice JS&Williamson MK. of Conserved Hauschka PV, Lian JB, Cole DE&Gundberg CM.Osteocalcin Ferron M, Wei J, Yoshizawa T, DelFattore A,DePinho RA, Teti A, Malec M. Carboxylatedandundercarboxylated osteocalcinin Kiec-Klimczak M, Solnica B,Hubalewska-Dydejczyk A &Malczewska- 105 the NationalAcademyofSciences UnitedStatesofAmerica development ofmetabolicdiseasesin wild-typemice. regulates betacellandadipocytegeneexpressionaffectsthe abb.2014.05.022) and Biophysics the skeleton:osteocalcinandbeyond. (https://doi.org/10.1016/j.cell.2007.05.047) of energymetabolismbytheskeleton. Dacquin R, Mee PJ,McKee MD,Jung DY, Science matrix glaproteinfromshark,lamb,rat,cow, andhuman. serines intheSer-X-Glu/Ser(P)sequencesofvitaminK-dependent physrev.1989.69.3.990) Physiological Reviews and matrixglaprotein:vitaminK-dependentproteinsinbone. doi.org/10.1016/j.cell.2010.06.003) remodeling andenergymetabolism. Ducy P &Karsenty G.Insulinsignalinginosteoblastsintegratesbone 5266–5270. 1994 3 2014 822–830. (https://doi.org/10.1073/pnas.0711119105) 1989 561 137–146. (https://doi.org/10.1002/pro.5560030511) 69 Downloaded fromBioscientifica.com at09/23/202109:31:55PM 990–1047. (https://doi.org/10.1016/j. https://eje.bioscientifica.com Cell Archives ofBiochemistry Cell (https://doi.org/10.1152/ 2010 et al 2007 181 . Endocrineregulation 142 :5 130 296–308. 456–469. Proceedings of Protein 2008 (https:// 535 via freeaccess European Journal of Endocrinology https://eje.bioscientifica.com

20 19 18 17 16 15 14 13 12 11 10 9 8 Clinical Study Li Z, Hassan MQ,Jafferji M,Aqeilan RI, Garzon R,Croce CM,van Kocijan R, Muschitz C,Geiger E,Skalicky S,Baierl A,Dormann R, Feichtinger X, Muschitz C,Heimel P, Baierl A,Fahrleitner-Pammer A, Matsuda M &DeFronzo RA.Insulinsensitivityindicesobtained Mari A, Pacini G,Murphy E,Ludvik B&Nolan JJ.Amodelbased Mizoguchi T, Muto A,Udagawa N,Arai A,Yamashita T, Hosoya A, Shalhoub V, Elliott G, Chiu L,Manoukian R,Kelley M,Hawkins N, Matsuzaki K, Udagawa N,Takahashi N, Yamaguchi K, Yasuda H, Zhou Y, Deng HW&Shen H.Circulating monocytes:anappropriate Muto A, Mizoguchi T, Udagawa N,Ito S,Kawahara I,Abiko Y, Arai A, Eghbali-Fatourechi GZ, Mödder UI,Charatcharoenwitthaya N, Marupanthorn K, Tantrawatpan Ch, Kheolamai P, Lin X, Brennan-Speranza TC,Levinger I&Yeap BB. 29b contributetopositiveregulation ofosteoblastdifferentiation. Wijnen AJ, Stein JL,Stein GS&Lian JB.BiologicalfunctionsofmiR- doi.org/10.1210/jc.2016-2365) Clinical EndocrinologyandMetabolism postmenopausal osteoporosisandfragility fractures. Circulating microRNAsignaturesinpatientswithidiopathicand Plachel F, Feichtinger X,Heimel P, Fahrleitner-Pammer A, 22844-2) Scientific Reports and theirassociationtobonemicrostructurehistomorphometry. circulating microRNAsmiR-29b-3p,miR-550a-3p,andmiR-324-3p Redl H, Resch H,Geiger E,Skalicky S,Dormann R, org/10.2337/diacare.22.9.1462) insulin clamp. from oralglucosetolerancetesting:comparisonwiththeeuglycemic org/10.2337/diacare.24.3.539) tolerance test. method forassessinginsulinsensitivityfromtheoralglucose org/10.1083/jcb.200806139) in vivo. Identification ofcellcycle-arrestedquiescentosteoclastprecursors Ninomiya T, Nakamura H,Yamamoto Y, Kinugawa S, 2141.2000.02379.x) Haematology of osteoclastprecursorsinhumanblood. Davy E, Shimamoto G,Beck J,Kaufman SA, org/10.1006/bbrc.1998.8586) Biophysical Research Communications human peripheralbloodmononuclearcellcultures. differentiation factor(ODF)inducesosteoclast-likecellformationin Shima N, Morinaga T, Toyama Y, Yabe Y, Higashio K, 2561–2572. model forbone-relatedstudy. org/10.1002/jbmr.490) ofBoneandMineralResearchJournal osteoclast precursorscirculate inbloodandsettledownintobone. Harada S, Kobayashi Y, Nakamichi Y, bone.2006.12.064) humans. Characterization ofcirculating osteoblastlineagecellsin Sanyal A, Undale AH,Clowes JA,Tarara JE &Khosla S. doi.org/10.3892/ijmm.2017.2872) ofMolecularMedicine Journal International stromal cellsderivedfromhumanbonemarrowandumbilicalcord. enhances theosteogenicdifferentiationcapacityofmesenchymal Tantikanlayaporn D &Manochantr S.Bonemorphogeneticprotein-2 nu10070847) sensitivity. for aroleinglucosehomeostasisandmuscleregulationofinsulin Undercarboxylated osteocalcin:experimentalandhumanevidence org/10.1002/dmrr.2862) Metabolism Research andReviews metabolic complicationsofhumanobesityandprediabetes. The Journal ofCellBiology The Journal Bone Nutrients (https://doi.org/10.1007/s00198-015-3250-7) 2000 2007 Diabetes Care Diabetes Care 2018 111 2018 40 8 501–512. 1370–1377. 4867. 10 2001 1999 847. (https://doi.org/10.1038/s41598-018- International 2017 (https://doi.org/10.1046/j.1365- 2009 24 22 (https://doi.org/10.3390/ 2011 (https://doi.org/10.1016/j. 1998 539–548. 2016 1462–1470. et al 33 U Raznyandothers 184 e2862. 2017 26 . Lineage-committed British Journal of British Journal 246 101 541–554. 2978–2990. et al 199–204. (https://doi. 39 4125–4134. . Characterization (https://doi. 654–662. (https://doi. et al Biochemical and Journal of Journal et al et al (https://doi. 2015 . Bone-related (https://doi. (https://doi. . Osteoclast . et al (https:// (https:// Diabetes/ 26 .

in obesity Bone turnovergeneexpression 21 26 25 24 23 22 35 34 33 32 31 30 29 28 27 Tan X, Weng T, Zhang J,Wang J, Li W, Wan H, Lan Y, Cheng X, Peng Y, Kang Q,Cheng H,Li X,Sun MH,Jiang W, Luu HH,Park JY, Mao B, Wu W, Li Y, Hoppe D,Stannek P, Glinka A&Niehrs C.LDL- Mao B, Wu W, Davidson G,Marhold J,Li M,Mechler BM,Delius H, Semenov M, Tamai K &He X.SOSTisaligandforLRP5/LRP6and Viljakainen H, Ivaska KK,Paldánius P, Lipsanen-Nyman M, Pollock NK, Bernard PJ,Gower BA,Gundberg CM, Wenger K, Schwetz V, Lerchbaum E, Schweighofer N,Hacker N,Trummer O, Paldanius PM, Ivaska KK,Hovi P, Andersson S,Eriksson JG, Bjarnason NH, Henriksen EE,Alexandersen P, Christgau S, Huang W, Yang S, Shao J&Li YP. Signalingandtranscriptional Paredes R, Arriagada G,Cruzat F, Villagra A, Olate J,Zaidi K,van Bialek P, Kern B,Yang X, Schrock M,Sosic D,Hong N,Wu H, Yu K, Kronenberg HM. Twist genesregulateRunx2andboneformation. Asagiri M &Takayanagi H. Themolecularunderstandingof murine postnatalbonehomeostasis. Hou N, Liu H, org/10.1002/jcb.10744) ofCellularBiochemistry Journal morphogenetic (BMPs)-mediatedosteogenicsignaling. Haydon RC &He TC.Transcriptional characterizationofbone 2001 receptor-related protein6isareceptorforDickkopfproteins. 664–667. receptors thatregulateWnt/beta-cateninsignaling. Hoppe D, Stannek P,Walter C, 26770–26775. a Wntsignalinginhibitor. jc.2013-3097) and Metabolism metabolism? Acase-controlstudy. Suppressed boneturnoverinobesity:alinktoenergy Saukkonen T, Pietiläinen KH,Andersson S,Laitinen K&Mäkitie O. doi.org/10.1074/jbc.M809787200) ofBiologicalChemistry The Journal cell function. concentrations inchildrenwithprediabetes isassociatedwithbeta- Misra S, Bassali RW &Davis CL.Loweruncarboxylated osteocalcin doi.org/10.4158/EP13110.OR) syndrome. polycystic ovary levels onoralglucoseloadinwomenbeinginvestigatedfor Borel O, Pieber TR,Chapurlat R&Obermayer-Pietsch B.Osteocalcin doi.org/10.1371/journal.pone.0063036) lowbirthweight. normal orvery osteocalcin associatewithinsulinlevelsinyoungadultsborn Väänänen K, Kajantie E&Mäkitie O.Total andcarboxylated s8756-3282(01)00662-7) in boneresorption. Henriksen DB &Christiansen C.Mechanismofcircadian variation doi.org/10.2741/2296) andVirtualBioscience: AJournal Library regulation inosteoblastcommitmentanddifferentiation. org/10.1128/MCB.24.20.8847-8861.2004) cells. to up-regulateratosteocalcingeneexpressioninosteoblastic factor Runx2interactswiththe1 Wijnen A, Lian JB, Stein GS, Stein JL, (https://doi.org/10.1016/S1534-5807(04)00058-9) of osteoblastdifferentiation. Ornitz DM, Olson EN, 5807(04)00069-3) Developmental Cell org/10.1016/j.bone.2006.09.023) osteoclast differentiation. 2162–2170. E1092–E1099. Molecular &CellularBiology 411 321–325. (https://doi.org/10.1038/nature756) (https://doi.org/10.1242/jcs.03466) Journal ofClinicalEndocrinologyandMetabolism Journal et al (https://doi.org/10.1074/jbc.M504308200) (https://doi.org/10.1210/jc.2010-2731) 2014 2004 . Smad4isrequiredformaintainingnormal Bone (https://doi.org/10.1038/35077108) 99 et al 6 2002 2155–2163. 317–318. Bone Journal ofBiologicalChemistry Journal . Atwistcodedeterminestheonset Downloaded fromBioscientifica.com at09/23/202109:31:55PM Endocrine Practice Developmental Cell 2003 et al 30 2007 2004 α 2009 307–313. PLOS ONE ,25-dihydroxyvitamin D3receptor . KremenproteinsareDickkopf Journal ofClinicalEndocrinologyJournal 90 (https://doi.org/10.1016/S1534- 40 Journal ofCellScience Journal (https://doi.org/10.1210/ et al. 2007 24 1149–1165. 284 251–264. 8847–8861. Bone-specific transcription 181 15676–15684. 2013 (https://doi.org/10.1016/ 12 2014 3068–3092. 2004 :5 8 (https://doi. e63036. 20 (https://doi. Nature 6 5–14. (https://doi. 423–435.

2005 2007 2002 Frontiers in (https:// (https:// (https:// 2011 (https:// Nature 280 536 120 417 via freeaccess 96

European Journal of Endocrinology

40 39 38 37 36 Clinical Study Hassan MQ, Tye CE, Stein GS&Lian JB.Non-codingRNAs: Lian JB, Stein GS,vanWijnen AJ, Stein JL,Hassan MQ,Gaur T& Takayanagi H. Osteoimmunology:sharedmechanismsandcrosstalk Matzelle MM, Gallant MA,Condon KW, Walsh NC, Manning CA, Gilbert L, He X,Farmer P, Rubin J,Drissi H,vanWijnen AJ, Lian JB, 2015 epigenetic regulatorsofbonedevelopmentandhomeostasis. org/10.1038/nrendo.2011.234) Nature Reviews.Endocrinology Zhang Y. MicroRNAcontrolofboneformationandhomeostasis. 2007 between theimmuneandbonesystems. (https://doi.org/10.1002/art.33504) signaling pathway. inflammation inducesosteoblastfunctionandregulatestheWnt Stein GS, Lian JB,Burr DB&Gravallese EM.Resolutionof (https://doi.org/10.1074/jbc.M106339200) necrosis factor- factor RUNX2(Cbfa1/AML3/Pebp2 Stein GS &Nanes MS.Expressionoftheosteoblastdifferentiation 81 7 292–304. 746–756. α . (https://doi.org/10.1038/nri2062) Journal ofBiologicalChemistry Journal (https://doi.org/10.1016/j.bone.2015.05.026) Arthritis andRheumatism Arthritis 2012 8 α 212–227. A)isinhibitedbytumor U Raznyandothers Nature Reviews.Immunology 2012 (https://doi. 2002 64 1540–1550. 277 2695–2701. Bone

Accepted 17September2019 Revised versionreceived10September2019 Received 15July2019

in obesity Bone turnovergeneexpression 44 43 42 41 Mizuno Y, Tokuzawa Y, Ninomiya Y, Yagi K, Yatsuka-Kanesaki Y, Zheng H, Liu J,Tycksen E, Nunley R&McAlinden A.MicroRNA- Hulsmans M, Sinnaeve P, Van derSchueren B,Mathieu C,Janssens S Trompeter HI, Dreesen J,Hermann E,Iwaniuk KM,Hafner M, febslet.2009.06.006) FEBS Letters promotes osteoblasticdifferentiationthroughinhibitionofAcvR1b. Suda T, Fukuda T, Katagiri T, Kondoh Y, Amemiya T, 123 PTEN/PI3K/AKT signalingandmitochondrialmetabolism. 181a/b-1 over-expressionenhancesosteogenesisbymodulating 1008) Metabolism disease. and coronary patients isassociatedwiththeoccurrenceofmetabolicsyndrome & Holvoet P. DecreasedmiR-181aexpressioninmonocytesofobese (https://doi.org/10.1186/1471-2164-14-111) stem cellsfromhumancordblood. miR-29b accelerateosteogenicdifferentiationofunrestrictedsomatic Renwick N, Tuschl T &Wernet P. MicroRNAsmiR-26a,miR-26b,and 92–102. 2012 2009 (https://doi.org/10.1016/j.bone.2019.03.020) 97 583 E1213–E1218. 2263–2268. Downloaded fromBioscientifica.com at09/23/202109:31:55PM Journal ofClinicalEndocrinologyand Journal (https://doi.org/10.1016/j. (https://doi.org/10.1210/jc.2012- BMC Genomics https://eje.bioscientifica.com 181 :5 2013 et al. 14 miR-210 Bone 111. 537 2019 via freeaccess