Serum visfatin and vaspin levels in normoglycemic
first-degree relatives of Iranian patients with type 2 diabetes mellitus
Samad Akbarzadeh, Iraj Nabipour *, Seyed Mojtaba Jafari, Ali Movahed, Niloofar Motamed, Majid Assadi, Najmeh Hajian
DepartmentofEndocrineandMetabolicDiseases,ThePersianGulfTropicalMedicineResearchCenter,BushehrUniversityofMedicalSciences, Boostan19Alley,ImamKhomeiniSt,Bushehr7514763448,Iran
1. Introduction
Since the mid-1990s, several adipocytokines have been identified in relation to obesity and comorbidities, type 2 diabetesmellitusanditscomplications,metabolicsyndrome, immunity, neuroendocrine, and cardiovascular functions [1–3]. The growing number of adipocyte-derived, cytokine- likehormonesincludesnovelinsulinresistanceinducersand insulin-mimetic/insulin-sensitizermoleculessuchasadipo- nectin,leptin,visfatin,andvaspin.
Visfatin, which is identical to the ‘‘pre-B cell-colony enhancing factor’’ and ‘‘nicotinamide phosphoribosyltrans- ferase,’’issecretedabundantlybythevisceralfatofhumans andmiceandmimicstheactionofinsulin[4].Further,invitro studiesshowedthatglucosesignalingforvisfatinreleasein adipocytes involves the phosphatidylinositol 3-kinase (PI3- kinase)andproteinkinaseBpathway[5]andtheregulationof glucose uptake by visfatin in human osteoblasts involves insulinreceptorphosphorylation[6]. Although currentdata maysuggestthatinsulinandvisfatinactinconcerttolower article info
Articlehistory:
Received5December2010 Receivedinrevisedform 11September2011 Accepted3October2011
Publishedonline22October2011
Keywords:
Adipocytokine Betacell
Insulinresistance Insulinsensitivity Diabetesmellitus Obesity
abstract
Aim: Toinvestigatecirculatingvisfatinandvaspinlevelsinfirst-degreerelativesofsubjects withtype2diabetesmellitus(FDRs)whofrequentlyhavehighervalueofHOMA-IRandbeta celldysfunction.
Methods: Serumvisfatinandvaspinconcentrationsweremeasuredin179Iraniansubjects (90normoglycemicFDRsand89age-andsex-matchedhealthycontrols)usingenzyme- linkedimmunosorbentassay(ELISA)methods.
Result: SerumvisfatinlevelsweresignificantlylowerintheFDRswhencomparedtothe controls (1.710.93ng/mlversus 2.692.02ng/ml, p=0.0001). However, no significant differencewasfoundinserumvaspinconcentrationsbetweentheFDRsandthecontrols (0.4520.254ng/mlversus 0.4090.275ng/ml,p>0.05). In multiple logistic regression analysis,theFDRsshowedasignificantassociationwithlowervisfatinlevelsafteradjust- mentsforage,sex,BodyMassIndex,systolicanddiastolicbloodpressures,lipidprofile, blood glucose levels and HOMA-IR [odds ratios (OR)=1.71, 95% confidence interval (1.30–2.25);p<0.0001].
Conclusion: The FDRs showed a significant association with lower visfatin levels. The observedlowercirculatingvisfatinlevelsinFDRsmaysuggestapathophysiologicalrole forvisfatininbetacelldysfunctioninthisgroup.
#2011ElsevierIrelandLtd.Allrightsreserved.
*Correspondingauthor.Tel.:+987712541827;fax:+987712541828.
E-mailaddress:[email protected](I.Nabipour).
ContentsavailableatSciverseScienceDirect
Diabetes Research and Clinical Practice
j o u r n a lh o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / d i a b r e s
0168-8227/$–seefrontmatter#2011ElsevierIrelandLtd.Allrightsreserved.
doi:10.1016/j.diabres.2011.10.004
bloodglucose[7],humanstudieshaveshowninconsistentand conflictingresultsregardingassociationsbetweenvisfatinand insulin-mimeticeffect,insulinresistance,betacellfunction impairment, adiposity, subcutaneous versus visceral fat distribution,anddiabetes[1,3,8].
Theplasmalevel ofvisfatinwas associated withtype2 diabetes mellitus, independent of knownrisk factors, in a Chinese population [9]. Visfatin levels increased in Asian Indianswithtype2diabetesmellitus,buttheassociationwas lostafteradjustingforobesity[10].Inanotherstudy,visfatin levelswerereportedtobehigherinthediabeticgroupthanthe controls,but therewas nosignificant differenceinvisfatin levelsbetweentheimpairedglucose-tolerantgroupandthe healthy controls [11]. Visfatin levels were correlated with HbA1c in patients with type 2 diabetes and lowered by intensiveglycemiccontrol[12].Higherserumlevelsofvisfatin inwomenwithgestationaldiabeteswerealsoreported[13,14].
Plasmavisfatinconcentrationwasnegativelyassociatedwith vascularendothelialfunctionintype2diabetesmellitus[15].
Visfatinwasshowntobesynthesizedinculturedmesangial cellsandtostimulateglucoseuptakeinglomerularmesangial cells.Thus,itwashypothesizedthatvisfatinmaycontributeto accelerationofdiabeticnephropathythroughtheaggravation ofmetabolicalterations[16].
Vaspin(visceraladiposetissue-derivedserpin),amember oftheserineproteaseinhibitorfamily,isanoveladipocyto- kinewithinsulin-sensitizingeffects[17].Thecirculatinglevel and tissue expression of vaspin increased at the peak of obesityandinsulinresistanceinOtsukaLong–EvansTokush- imafatty(OLETF)rats,ananimalmodelofabdominalobesity andtype2diabetesmellitus[18].Ithasbeensuggestedthatthe compensatoryincreaseinvaspinlevelsduringinsulinresis- tance mayamelioratethe action of unknownup-regulated proteases, which impair the insulin action [18]. However, circulating vaspin levels in relation to insulin sensitivity/
insulin resistancestatus and glucose metabolism arecon- flictingandremaintobeelucidated.
First-degree relatives of subjects with type 2 diabetes mellitus(FDRs)frequentlyhave bothinsulinresistance[19]
andbetacelldysfunction[20]andbeara40%lifetimeriskfor developing type 2 diabetes mellitus [21]. With such a background of the pathophysiology of glucose metabolism inFDRs,investigatingthecirculatinglevelsofnoveladipocy- tokines, such as visfatin and vaspin, in this group is worthwhile.Thepresentstudy,forthefirsttime,soughtto exploreserumvisfatinandvaspinconcentrationsinFDRs.
2. Methods
2.1. Subjectsandphysicalmeasurements
Atotalof90 glucose-tolerantsubjects(41men, 49women, meanage40.559.36years)offirst-degreerelativesofIranian patientswithtype2diabetesmellitus(FDRs)whoconsecu- tivelyvisitedtheEndocrineClinicofFatemeh-ZahraUniversi- ty Hospital were enrolled in this study. We selected an age-andsex-matchedglucose-tolerantcontrolgroup(49men, 40women,meanage41.3759.47years)fromthePersianGulf Healthy Heart Study participants [22] or subjects with no
apparent disease history,who underwent a routinehealth checkup at General Clinic of Fatemeh-Zahra University Hospital. They had no family history of diabetes mellitus, type 2 diabetes or endocrine disorders, cardiovascular diseases,orhepaticorrenaldysfunctionandwerenottaking antirheumaticdrugsandmedicationsformetabolicdiseases.
The participants in the control group had anthropometric measurementscomparabletothoseoftheFDRs(Table1).All subjectswereaskedtofastandtopresenttothePersianGulf HealthResearchCenterbetween7:30and9:30a.m.,afteran overnightfast.Theyunderwenta75-goralglucosetolerance test(OGTT).Thedefinitionofglucosetolerantwasasubject whohadafastingplasmaglucoselevellowerthan110mg/dl and asecond-hourbloodsugar levelafteroralglucoseload (2h-OGTT)lowerthan140mg/dl[23].
Bloodpressurewasassessedtwiceattherightarmaftera 15-minrestinthesittingposition,usingastandardmercury sphygmomanometer.Heightandweightweremeasuredusing astadiometer.Heavyoutergarmentsandshoeswereremoved beforeheight andweightweremeasured.BodyMassIndex (BMI)wascalculated.Waistcircumferencewasdefinedatthe midwaylevelbetweenthecostalmarginsandtheiliaccrests.
Hipcircumferencewasmeasuredat thelevel ofthegreater trochanters.
Thestudywasapprovedbythemedical-ethicalcommittee ofBushehrUniversityofMedicalSciences.
2.2. Laboratorymeasurements
Afastingbloodsamplewastaken,allsampleswerepromptly centrifuged,andserawereseparatedandkeptfrozenat 708C
Table1–Thegeneralcharacteristics,includingblood pressureandanthropometricmeasurements,andthe biochemicalparametersoffirst-degreerelativesofsub- jectswithtype2diabetesmellitus(FDRs)andhealthy controls.
FDRs(n=90) Control(n=89)
Female/maleratio 49/41 40/49
Age(years) 40.559.63 41.379.47
SystolicBP(mmHg) 115.013.8 115.8212.84 DiastolicBP(mmHg) 74.729.95 74.4510.47
Height(cm) 165.889.31 166.898.87
Weight(kg) 75.9213.43 74.3214.75
Waistcircumference(cm) 94.4110.19 93.1012.63 Hipcircumference(cm) 104.769.975 102.9312.42
WHR 0.900.14 0.910.17
BMI(kg/m2) 27.574.25 26.524.24
Fastingglucose(mg/dl) 80.58.56 79.7710.46 2h-OGTT(mg/dl) 92.1715.68 92.0316.65 Totalcholesterol(mg/dl) 198.5436.85 196.9237.24 Triglyceride(mg/dl) 168.5182.3 155.7175.0 HDL-cholesterol(mg/dl) 4310.38 45.4411.65 LDL-cholesterol(mg/dl) 121.0131.55 120.0432.44 Insulin(mIU/ml) 8.75(5.25–16.68) 6.56(5.03–10.18)* HOMA-IR 1.82(1.05–3.06) 1.29(0.98–1.93)* BMI,BodyMassIndex;WHR,waisttohipratio;BP,BloodPressure;
2h-OGTT, second-hour oral glucose tolerance test; HOMA-IR, homeostasismodelofassessmentindex.
Dataaremeans(SD),exceptforinsulinand HOMA-IR[medians (interquartileranges)].
*pvalues<0.05(theFDRsincomparisontothecontrols).
until used. Analyses for biochemical parameters (blood sugars,triglyceride,andcholesterollevels)werecarriedout atthePersianGulfHealthResearchCenteronthedayofblood collection using a Selectra 2 autoanalyzer (Vital Scientific, Spankeren,theNetherlands).Glucosewasassayedwiththe enzymatic (glucose oxidase) colorimetric method using a commercialkit(ParsAzmunInc.,Tehran,Iran).Serumtotal cholesterol and HDL (high-density lipoprotein) cholesterol weremeasuredusingcholesteroloxidasephenolaminoanti- pyrine and triglycerides using the glycerol-3 phosphate oxidasephenolaminoantipyrine enzymaticmethod.Serum LDL(low-densitylipoprotein)cholesterolwascalculatedusing the Friedwald formula; LDL cholesterol was not calculated whenthetriglyceridesconcentrationwas>400mg/dl.Insulin wasmeasuredusingacommerciallyavailableenzyme-linked immunosorbent assay kit (Insulin ELISA, DRG diagnostics, Marburg,Germany).Theassaysensitivitywas1.76mIU/ml;the intra-andinterassaycoefficientsofvariancewere1.79–2.6%
and2.88–5.99%,respectively.Insulinresistancewasassessed by calculating the homeostasis model of assessment index (HOMA-IR) using the equation: fasting insulin (mIU/
ml)fastingglucose(mg/dl)/405.
To detect visfatin and vaspin in the serum samples, commercially (Cat. No. V0523EK and Cat. No. V0712EK, respectively)availableenzyme-linkedimmunosorbentassay kits (AdipoGen, Seoul, Korea) were used according to the manufacturer’sinstructions.Theassaysensitivityforvisfatin was 0.10ng/ml; the intra- and interassay coefficients of variancewere3.8–5.5%and6.4–9.5%,respectively.Theassay sensitivity forvaspinwas 0.012ng/ml;theintra-and inter- assay coefficients of variance were 1.3–3.8% and 3.3–9.1%, respectively.
2.3. Statisticalanalysis
Normal distribution of the data was controlled with the Kolmogorov–Smirnov test. Probability values <5% were considered statistically significant. The significance of the difference in the results between the two groups was determinedwithchi-squareanalysisusing22contingency tables. Atwo-tailed t-testwas usedtocompare the values across groups in the presence of a normal distribution.
SignificantdifferenceswereassessedwiththeMann–Whitney Utestintheabsenceofanormaldistribution.Multiplelinear regression models were used to assess the association betweenadipocytokines(visfatinandvaspin)levels(indepen- dentvariables)andanthropometric,bloodpressuremeasure- ments,andbiochemicalvariables(eachfactorwasconsidered aseparatedependentvariableinaseriesofmodels).Because thedistributionsofseruminsulinlevelsandHOMA-IRwere skewed, logarithmically transformed values were used for statistical analysis. Insulin and HOMA-IR showed normal distributions after the logarithmic transformations. Binary logistic regression analysis was used to ascertain the associationsbetweenthevisfatinlevelsandthepresenceof FDRs.Sex, age, BMI,systolic and diastolic blood pressures, glucose levels, HOMA-IR and lipid status parameters were consideredcovariates,andthepresenceofFDRsthedepen- dentvariable.Standardorsimultaneous(EnterMethod)type ofregressionanalysiswasused.Allstatisticalanalyseswere
performedusingthePASWStatisticsGradPack18(SPSSInc., Chicago,IL).
3. Results
The general characteristics, including blood pressure and anthropometricmeasurements,andthebiochemicalparam- etersofthestudygroupsareshowninTable1.Therewasno significantdifferencebetweenthetwogroupsregardingblood pressure measurements, height, weight, waist and hip circumferences, waist to hip ratio, and BMI. There was also no significant difference in the fasting blood glucose, 2h-OGTT,andlipidprofilesbetweenthetwogroups.However, theFDRshadhigherHOMA-IRandseruminsulinlevelsthan thecontrolgroup(p<0.05,Table1).
SerumvisfatinlevelsweresignificantlylowerintheFDRs when compared to the controls (1.710.93ng/ml versus 2.692.02ng/ml,p=0.0001).However,nosignificantdiffer- encewasfoundintheserumvaspinconcentrationsbetween the FDRs and the controls (0.4520.254ng/ml versus 0.4090.275ng/ml, p=0.305; Fig. 1). No gender difference was found for the circulating visfatin and vaspin levels (p>0.05).
TherewasnosignificantdifferencebetweentheFDRsand control groupsregardingtheprevalence ofconsumptionof antihypertensive agents (7.8% versus 4.5%), hypocholester- olemicdrugs(6.7%versus5.6%)andaspirin(5.6%versus7.9%).
Atotalof10womenintheFDRsgroupand13womeninthe controlgroupwereonanticontraceptivepills.
Age-andsex-adjustedcorrelationsforserumvisfatinand vaspininrelationtoanthropometric,biochemical,andblood pressure measures are shown in Table 2. Age- and sex- adjustedserumvisfatinconcentrationlevelsweresignificant- lycorrelatedwithBMIintheFDRs(b=0.22,p=0.04).However, there were no significant correlations between visfatin concentrations and HOMA-IR, the fasting blood glucose, 2h-OGTT, triglyceride, total cholesterol, HDL-cholesterol, LDL-cholesterol, insulin and vaspin levels in the FDRs.
However,age-andsex-adjustedserumvisfatinconcentration levels were significantly correlated with total cholesterol
Fig.1–Serumvisfatinandvaspinconcentrationsinfirst- degreerelativesofsubjectswithtype2diabetesmellitus (FDRs,n=90)andhealthycontrols(n=89).Dataare meansWstandarddeviation.
(b=0.29,p=0.01)andLDL-cholesterol(b=0.28,p=0.01)inthe controlgroup.
Age-andsex-adjustedvaspinwasalsonotcorrelatedwith anthropometric, biochemical, or blood pressure measure- ments. However, age-and sex-adjustedserum vaspincon- centrationlevelsweresignificantlycorrelatedwithHOMA-IR intheFDRs(b= 0.25,p=0.03).
Table3showsunadjustedandadjustedoddsratios(95%CI) betweenthepresenceofFDRsandcirculatingadipocytokines levels. In multiple logistic regression analysis, the FDRs showedasignificant associationwith lowervisfatin levels, even after adjustments were made for age, sex, BMI, and biochemicalparameters[OR=1.71,CI(1.30–2.25);p<0.0001].
However,aftersuccessiveadjustmentforpotentialconfoun- ders,the FDRs didnot show asignificant association with vaspinlevels(Table3).
4. Discussion
Inthe currentstudy,we foundthatFDRs hada significant associationwithlowervisfatinlevels,evenafteradjustingfor age, sex, BMI, and biochemical parameters inmultivariate
analyses. However, no significant difference was found in serum vaspin concentrations between the FDRs and the controls.
Visfatinshowssomeinsulinmimeticproperties[24],and its molecular mechanism of actions is under vigorous challenge[4,24,25].Recently,Brownetal.[26]suggestedthat visfatin can significantly regulateinsulin secretion, insulin receptor phosphorylation, and intracellular signaling and expressionofanumberofbetacellfunction-associatedgenes inmousebetacells.
However,therearehugediscrepanciesinreportsofvisfatin inrelationtoinsulinsensitivityandinsulinresistance[1,8,13].
Similartoourstudy,alackofcorrelationbetweencirculating visfatin and insulin/HOMA [27,28] was reported, but in contrast, a significant correlation between visfatin levels and insulin/HOMA was observed by other authors [9,13].
Paganoetal.[28]reportedthatvisfatindidnotplayaroleinthe developmentofinsulinresistance,inhumanobesity,eitheras assessed byhomeostasismodel assessmentor duringlipid infusion.Thefindingsofthelatter studyareinaccordance with the results of other studies that did not find an association of serum visfatin with insulin resistance [10,27,29]. Taken together, the current knowledge doesnot Table2–Age-andsex-adjustedcorrelationsforserumvisfatinandvaspininrelationtoanthropometric,biochemical,and bloodpressuremeasuresinfirst-degreerelativesofsubjectswithtype2diabetesmellitus(FDRs)andhealthycontrols.
Visfatin Vaspin
FDRs Control FDRs Control
b pvalue b pvalue b pvalue b pvalue
BMI 0.22 0.04 0.13 0.23 0.01 0.93 0.01 0.96
WHR 0.05 0.61 0.15 0.18 0.01 0.94 0.05 0.68
SystolicBP 0.12 0.29 0.04 0.71 0.01 0.91 0.27 0.04
DiastolicBP 0.13 0.25 0.01 0.92 0.04 0.68 0.11 0.37
Fastingglucose 0.04 0.70 0.13 0.23 0.19 0.07 0.05 0.64
2h-OGTT 0.07 0.48 0.18 0.10 0.05 0.64 0.01 0.93
Triglyceride 0.10 0.38 0.21 0.06 0.10 0.34 0.11 0.36
Totalcholesterol 0.07 0.51 0.29 0.01 0.13 0.24 0.01 0.88
LDL-cholesterol 0.05 0.60 0.28 0.01 0.10 0.34 0.03 0.77
HDL-cholesterol 0.08 0.49 0.10 0.39 0.02 0.83 0.08 0.50
Insulin 0.12 0.27 0.02 0.81 0.18 0.12 0.21 0.07
HOMA-IR 0.13 0.24 0.01 0.92 0.25 0.03 0.21 0.07
Visfatin – – – – 0.11 0.30 0.13 0.27
Vaspin 0.11 0.30 0.13 0.27 – – – –
BMI,BodyMassIndex;WHR,waisttohipratio;BP,bloodpressure;2h-OGTT,second-houroralglucosetolerancetest;HOMA-IR,homeostasis modelofassessmentindex.
Table3–Unadjustedandadjustedoddsratios[OR;95%confidenceinterval(CI)]betweenthepresenceoffirst-degree relativesofsubjectswithtype2diabetesmellitus(FDRs)andcirculatingadipocytokines(visfatinandvaspin)levels.
Visfatin Vaspin
OR CI pvalue OR CI pvalue
Unadjusted 1.53 1.22–1.93 <0.0001 0.53 0.16–1.76 0.304
Adjustedforage+sex 1.56 1.23–1.98 <0.0001 0.51 0.15–1.73 0.282
Adjustedforage+sex+BMI 1.60 1.25–2.06 <0.0001 0.50 0.14–1.73 0.280
Adjustedforage+sex+WHR 1.55 1.22–1.96 <0.0001 0.51 0.15–1.75 0.287
Fullmodel 1.71 1.30–2.25 <0.0001 0.25 0.06–1.06 0.061
BMI,BodyMassIndex;WHR,waisttohipratio.
Fullmodelincludedage, sex,BMI,systolicanddiastolic bloodpressures,HOMA-IR,fastingglucose, second-hourbloodglucose,insulin, triglyceride,HDL-cholesterolandLDL-cholesterollevelsandFDRs(yesversusno).
suggest that visfatin plays an unequivocal role in the developmentofinsulinresistance.
Likewise, visfatin’s relationship with type 2 diabetes mellitus and its link with pathophysiological pathways in pre-diabeticstatesandmetabolicsyndromeareunclear,and thereisinconsistencyinthecorrelationsbetweencirculating visfatin and diabetes mellitus. In some studies, higher circulating visfatin levels in patients with type 2 diabetes mellitus[9,11]andgestationaldiabetesmellitus[13,14]were observedwhilelowerlevelsofvisfatininpatientswithtype1 diabetesmellitus[30]and gestationaldiabetesmellitus[31]
werereported.Nocorrelationbetweenvisfatinanddiabetes [15]independentofobesity[10]wasfoundinotherstudies.
Chen et al. [9] suggested that the compensatory elevated visfatin,whichhaspotentialinsulin mimeticproperties,in patientswithdiabetesmellitus,mayreflectanimpairmentof itsactionintargettissues,dysregulationinitsbiosynthesis,or aresponsetohyperglycemia/hyperinsulinemiaorthealtered stateofadipocytokines.
According to the above studies, it appears that frank impairment conditions of glucose metabolism (e.g., the diabetes state) are not good models for assessing the pathophysiologicallinkofvisfatinwithdiabetesmellitus.In the current study, we evaluated the circulating levels of visfatininglucose-tolerantFDRs.
Impaired glucose metabolism iscommon in FDRs [19].
Theyalsohavedefectsinbetacellfunctionatatimewhen they are not hyperglycemic, and this reduction affects insulin andamylin responses to glucose stimulation [20].
Since both insulin sensitivity and beta cell dysfunction predictconversiontodiabetes[32,33],FDRsareatincreased riskoftype2diabetes[21].Arecentlargepopulationstudy [33]showedthatbetacellimpairmentexistsintheoffspring oftype2diabetespatients, eveninthe absenceofinsulin resistance, suggesting that beta cell dysfunction is the outstanding determinant forthe development of diabetes mellitusinthisgroup.
Revolloetal.[34]providedseverallinesofevidencethat visfatin (Nampt)-mediated systemic nicotinamide adenine dinucleotide (NAD) biosynthesis is critical for beta cell function.Visfatin/Namptisanimportantrate-limitingmoiety in the NAD biosynthetic pathway from nicotinamide [35].
AlterationinNADlevelscouldaltertheenzymaticactivitiesof NDA-dependentdeacetylaseSirt1andaffectimportantmeta- bolic pathways that are essential for regulating glucose- stimulatedinsulinsecretioninpancreaticbetacells[34].
Revolloetal.[34]showedthatvisfatin/Nampt+/ miceand islets had defects in nicotinamide mononucleotide (NMN)/
NADbiosynthesis and glucose-stimulatedinsulinsecretion, and these defectswere corrected by the administration of NMN,aproductofthevisfatinreaction.
Asamajorlimitation,betacellfunctionwasnotassessedin thecurrentstudy.However,basedontheresultsoftheabove study [34], we can hypothesize that the observed lower circulatingvisfatinlevelsinourFDRsmayleadtoderange- ment in visfatin-mediated systemic NAD biosynthesis and earlydefectsinglucose-stimulatedinsulinsecretionintheir pancreatic betacells.Undoubtedly, assessment ofbetacell functionand measurement ofplasma NMN levels inFDRs wouldbehelpful.
It should be noted that the linking mechanism of low visfatinlevelsandearlybetacelldysfunctioninFDRsappears not to be operative in the advanced phases of beta cell deterioration, similar to type 2 diabetes mellitus. In fact, increasingcirculatingvisfatinlevelswasdemonstratedwith progressive beta cell deterioration [36]. To identify the pathophysiological stimuli that cause increasedproduction ofvisfatinintype2diabetes,asreportedbysomeauthors, furtherinvestigationsareneeded.
Wefoundasignificantcorrelationbetweenvisfatinlevels and BMIin the FDRs. Circulating visfatin concentrationin relation toobesityiscontroversial[27,28,37,38].In arecent study [39], visfatin/PBEF/Nampt mRNA levels were not correlated with measures of obesity, and the researchers suggested thatvisfatin isnotpredominantly secreted from visceralfat.
We found that age- and sex-adjusted serum visfatin concentrationlevels weresignificantlycorrelated withtotal cholesterolandLDL-cholesterolinthecontrolgroupbutthis positive correlation was lost in the FDRs group. It was suggested thatvisfatin-induced freefattyacid fluxtoliver, drives very low density lipoprotein production, leading to elevated plasma triglyceride and total cholesterol [39].
However, there are controversies in reports of visfatin in relation to lipid metabolism. Positive correlations between visfatin concentrations and total cholesterol [39], HDL- cholesterol[40,41]andtriglyceridelevels[39]wereobserved byotherauthors.Incontrast,negativecorrelationsbetween visfatin concentrations and LDL-cholesterol [40], HDL- cholesterol and triglyceride [41] levels were reported.
Zahorska-Markiewiczdidnotobserveacorrelationbetween visfatinconcentrationsandlipidsinboththeobeseandlean groups[37].Thelackofcorrelationbetweenvisfatinlevelsand parametersoflipidmetabolismintheFDRsgroupwarrants furtherstudy.
Inourstudy,therewasnodifferenceincirculatingvaspin betweentheFDRsandthecontrols.Humanvaspinexpression was associated with obesity, insulin resistance, glucose metabolism in human [42] and insulin sensitivity in rats [43].Vaspinlevelswerereportedtobesignificantlyelevatedin nonobeseandobeseT2DMpatientscomparedwithcontrols [44].Inanotherstudy,therewasnosignificantdifferencein circulatingvaspinbetweennormalglucose-tolerantsubjects andpatientswithtype2diabetesmellitus[45].
Although the association of vaspin levels with insulin resistance,obesity,andglycemicstatusisstillcontroversial inhumans[46],itseemsthattheproductionofvaspinfrom adiposetissueisacompensatorymechanismtoameliorate severe insulin resistance in obesity and type 2 diabetes [42,45].
We did not measure insulin sensitivity with the gold- standard isoglycaemic-hyperinsulinemic clamp test in our subjects, which isa limitation; however, the FDRs didnot show a significant association with vaspin levels after adjustmentsforHOMA-IRinmultivariateanalysis(Table3).
SincegoodcorrelationsbetweenHOMA-IRandinsulinresis- tanceassessedbysome‘‘goldstandard’’methodshavebeen reported[47],wepostulatethatthelevelofinsulinresistance inFDRswasnotatathresholdtostimulatevaspinsecretion,as acompensatorymechanism.
Weacknowledgeseveralstudylimitations.Asourdataare cross-sectional, limited inferences can be made regarding temporality andcausation. Anotherlimitation ofourstudy includesthelackofassessmentofbetacellfunction.Sincewe assessed the investigated adipocytokines with singlemea- surements,the changes inthese adipocytokinesover time could not be reflected in the current study. To clarify circulatingadipocytokinesinrelationtoworseningofglucose tolerance overtime in FDRs, longitudinal studies arewar- ranted.Themeasurementofadditional adipocytokinesand inflammatorymarkersandcytokinesmeritsconsiderationto elucidatethecomplexsystemthatregulatesbetacellfunction andinsulinsensitivityinpre-diabetesstates.
In conclusion, there was no significant difference in circulatingvaspinlevelsbetweentheFDRsandthecontrols.
Therefore,circulatingvaspinmightnothaveadetermining valueasabiologicalmarkerinFDRs.Incontrast,theobserved lower circulating visfatin levels in FDRs may suggest a pathophysiologicalroleforvisfatininbeta cell dysfunction inthisgroup.
Because visfatin-mediated systemic NAD biosynthesis is vitalforbetacellfunction[34],furtherstudiesareneededto elucidatetheprecisemechanismsbehindunderproductionof visfatininFDRstopreventfurtherpathologicalconsequences includingdevelopmentoftype2diabetesmellitusinthisgroup.
Conflict of interest
Therearenoconflictsofinterest.
Acknowledgments
Thisstudywassupportedinpartbya grantfromBushehr ProvinceTechnologyandResearchCommitteeandResearch DeputyofBushehrUniversityofMedicalSciences.
references
[1] LagoF,Go´mezR,Go´mez-ReinoJJ,DieguezC,GualilloO.
Adipokinesasnovelmodulatorsoflipidmetabolism.
TrendsBiochemSci2009;34:500–10.
[2] KralischS,BluherM,PaschkeR,StumvollM,FasshauerM.
Adipokinesandadipocytetargetsinthefuture
managementofobesityandthemetabolicsyndrome.Mini RevMedChem2007;7:39–45.
[3] GulcelikNE,UsmanA,Gu¨rlekA.Roleofadipocytokinesin predictingthedevelopmentofdiabetesanditslate complications.Endocrine2009;36:397–403.
[4] FukuharaA,MatsudaM,NishizawaM,SegawaK, TanakaM,KishimotoK,etal.Visfatin:aproteinsecretedby visceralfatthatmimicstheeffectsofinsulin.Science 2005;307:426–30.
[5] HaiderDG,SchallerG,KapiotisS,MaierC,LugerA,WolztM.
Thereleaseoftheadipocytokinevisfatinisregulatedby glucoseandinsulin.Diabetologia2006;49:1909–14.
[6] XieH,TangSY,LuoXH,HuangJ,CuiRR,YuanLQ,etal.
Insulin-likeeffectsofvisfatinonhumanosteoblasts.Calcif TissueInt2007;80:201–10.
[7] KatwaLC,SeidelER.Visfatininpregnancy:proposed mechanismofpeptidedelivery.AminoAcids2009;37:555–8.
[8] Saddi-RosaP,OliveiraCS,GiuffridaFM,ReisAF.Visfatin, glucosemetabolismandvasculardisease:areviewof evidence.DiabetolMetabSyndr2010;2:21.
[9] ChenMP,ChungFM,ChangDM,TsaiJC,HuangHF,ShinSJ, etal.Elevatedplasmalevelofvisfatin/pre-Bcellcolony- enhancingfactorinpatientswithtype2diabetesmellitus.
JClinEndocrinolMetab2006;91:295–9.
[10] SandeepS,VelmuruganK,DeepaR,MohanV.Serum visfatininrelationtovisceralfat,obesity,andtype2 diabetesmellitusinAsianIndians.Metabolism 2007;56:565–70.
[11] DogruT,SonmezA,TasciI,BozogluE,YilmazMI,GencH, etal.Plasmavisfatinlevelsinpatientswithnewly diagnosedanduntreatedtype2diabetesmellitusand impairedglucosetolerance.DiabetesResClinPract 2007;76:24–9.
[12] ZhuJ,SchottM,LiuR,LiuC,ShenB,WangQ,etal.
Intensiveglycemiccontrollowersplasmavisfatinlevels inpatientswithtype2diabetes.HormMetabRes2008;40:
801–5.
[13] LewandowskiKC,StojanovicN,PressM,TuckSM, SzoslandK,BienkiewiczM,etal.Elevatedserumlevelsof visfatiningestationaldiabetes:acomparativestudyacross variousdegreesofglucosetolerance.Diabetologia 2007;50:1033–7.
[14] KrzyzanowskaK,KruglugerW,MittermayerF,RahmanR, HaiderD,ShnawaN,etal.Increasedvisfatin
concentrationsinwomenwithgestationaldiabetes mellitus.ClinSci(Lond)2006;110:605–9.
[15] TakebayashiK,SuetsuguM,WakabayashiS,AsoY,Inukai T.Associationbetweenplasmavisfatinandvascular endothelialfunctioninpatientswithtype2diabetes mellitus.Metabolism2007;56:451–8.
[16] SongHK,LeeMH,KimBK,ParkYG,KoGJ,KangYS.
Visfatin:anewplayerinmesangialcellphysiologyand diabeticnephropathy.AmJPhysiolRenalPhysiol 2008;295:F1485–94.
[17] WadaJ.Vaspin:anovelserpinwithinsulin-sensitizing effects.ExpertOpinInvestigDrugs2008;17:327–33.
[18] HidaK,WadaJ,EguchiJ,ZhangH,BabaM,SeidaA,etal.
Visceraladiposetissue-derivedserineproteaseinhibitor:a uniqueinsulin-sensitizingadipocytokineinobesity.Proc NatlAcadSciUSA2005;102:10610–5.
[19] ErikssonJ,Franssila-KallunkiA,EkstrandA,SalorantaC, Widen,SchalinC,etal.Earlymetabolicdefectsinpersons atincreasedriskfornon-insulin-dependentdiabetes mellitus.NEnglJMed1989;321:337–43.
[20] KnowlesNG,LandchildMA,FujimotoWY,KahnSESO.
Insulinandamylinreleasearebothdiminishedinfirst- degreerelativesofsubjectswithtype2diabetes.Diabetes Care2002;25:292–7.
[21] PerseghinG,GhoshS,GerowK,ShulmanGI.Metabolic defectsinleannondiabeticoffspringofNIDDMparents:a cross-sectionalstudy.Diabetes1997;46:1001–9.
[22] NabipourI,AmiriM,ImamiSR,JahfariSM,ShafeiaeE, NosratiA,etal.Themetabolicsyndromeandnonfatal ischemicheartdisease;apopulation-basedstudy.IntJ Cardiol2007;118:48–53.
[23] CarnevaleSchiancaGP,RossiA,SainaghiPP,MaduliE, BartoliE.Thesignificanceofimpairedfastingglucose versusimpairedglucosetolerance:importanceof insulinsecretionandresistance.DiabetesCare2003;26:
1333–7.
[24] StephensJM,Vidal-PuigAJ.Anupdateonvisfatin/pre-Bcell colony-enhancingfactor,anubiquitouslyexpressed, illusivecytokinethatisregulatedinobesity.CurrOpin Lipidol2006;17:128–31.
[25] FukuharaA,MatsudaM,NishizawaM,SegawaK, TanakaM,KishimotoK,etal.Retraction.Science 2007;318:565.
[26] BrownJE,OnyangoDJ,RamanjaneyaM,ConnerAC,PatelST, DunmoreSJ,etal.Visfatinregulatesinsulinsecretion, insulinreceptorsignallingandmRNAexpressionof diabetes-relatedgenesinmousepancreaticbeta-cells.JMol Endocrinol2010;44:171–8.
[27] BerndtJ,Klo¨tingN,KralischS,KovacsP,FasshauerM, Scho¨nMR,etal.Plasmavisfatinconcentrationsandfat depot-specificmRNAexpressioninhumans.Diabetes 2005;54:2911–6.
[28] PaganoC,PilonC,OlivieriM,MasonP,FabrisR,SerraR, etal.Reducedplasmavisfatin/pre-Bcellcolony-enhancing factorinobesityisnotrelatedtoinsulinresistancein humans.JClinEndocrinolMetab2006;91:3165–70.
[29] HammarstedtA,Pihlajama¨kiJ,RotterSopasakisV,GoggS, JanssonPA,LaaksoM,etal.Visfatinisanadipokine,butit isnotregulatedbythiazolidinediones.JClinEndocrinol Metab2006;91:1181–4.
[30] TorunerF,AltinovaAE,BukanN,ArslanE,AkbayE,ErsoyR, etal.Plasmavisfatinconcentrationsinsubjectswithtype1 diabetesmellitus.HormRes2009;72:33–7.
[31] ChanTF,ChenYL,LeeCH,ChouFH,WuLC,JongSB,etal.
Decreasedplasmavisfatinconcentrationsinwomenwith gestationaldiabetesmellitus.JSocGynecolInvestig 2006;13:364–7.
[32] WarramJH,MartinBC,KrolewskiAS,SoeldnerJS,KahnCR.
Slowglucoseremovalrateandhyperinsulinemiaprecede thedevelopmentoftypeIIdiabetesintheoffspringof diabeticparents.AnnInternMed1990;113:909–15.
[33] StadlerM,PaciniG,PetrieJ,LugerA,AnderwaldC, RISCInvestigators.Betacell(dys)functionin
non-diabeticoffspringofdiabeticpatients.Diabetologia 2009;52:2435–44.
[34] RevolloJR,Ko¨rnerA,MillsKF,SatohA,WangT,GartenA, etal.Nampt/PBEF/Visfatinregulatesinsulinsecretionin betacellsasasystemicNADbiosyntheticenzyme.Cell Metab2007;6:363–75.
[35] RevolloJR,GrimmAA,ImaiS.TheNADbiosynthesis pathwaymediatedbynicotinamide
phosphoribosyltransferaseregulatesSir2activity inmammaliancells.JBiolChem2004;279(3):50754–63.
[36] Lo´pez-BermejoA,Chico-Julia` B,Ferna`ndez-BalsellsM, RecasensM,EsteveE,CasamitjanaR,etal.Serumvisfatin increaseswithprogressivebeta-celldeterioration.Diabetes 2006;55:2871–5.
[37] Zahorska-MarkiewiczB,Olszanecka-GlinianowiczM, JanowskaJ,KocełakP,Semik-GrabarczykE,HoleckiM,etal.
Serumconcentrationofvisfatininobesewomen.
Metabolism2007;56:1131–4.
[38] HaiderDG,SchindlerK,SchallerG,PragerG,WolztM, LudvikB.Increasedplasmavisfatinconcentrationsin morbidlyobesesubjectsarereducedaftergastricbanding.
JClinEndocrinolMetab2006;91:1578–81.
[39] ChangYC,ChangTJ,LeeWJ,ChuangLM.Therelationship ofvisfatin/pre-B-cellcolony-enhancingfactor/
nicotinamidephosphoribosyltransferaseinadiposetissue withinflammation,insulinresistance,andplasmalipids.
Metabolism2010;59:93–9.
[40] ChenCC,LiTC,LiCI,LiuCS,LinWY,WuMT,etal.The relationshipbetweenvisfatinlevelsandanthropometric andmetabolicparameters:associationwithcholesterol levelsinwomen.Metabolism2007;56:1216–20.
[41] WangP,vanGreevenbroekMM,BouwmanFG,BrouwersMC, vanderKallenCJ,SmitE,etal.ThecirculatingPBEF/NAMPT/
visfatinlevelisassociatedwithabeneficialbloodlipid profile.PflugersArch2007;454:971–6.
[42] Klo¨tingN,BerndtJ,KralischS,KovacsP,FasshauerM, Scho¨nMR,etal.Vaspingeneexpressioninhumanadipose tissue:associationwithobesityandtype2diabetes.
BiochemBiophysResCommun2006;339:430–6.
[43] WangYM,WangWP,WangLP,Lu¨ QH,ZhouXH.Calorie controlincreasedvaspinlevelsofserumand
periepididymaladiposetissueindiet-inducedobeseratsin associationwithserumfreefattyacidandtumornecrosis factoralpha.ChinMedJ(Engl)2010;123:936–41.
[44] El-MesallamyHO,KassemDH,El-DemerdashE,AminAI.
Vaspinandvisfatin/Namptareinterestinginterrelated adipokinesplayingaroleinthepathogenesisoftype2 diabetesmellitus.Metabolism2010.June1[Epubahead ofprint].
[45] YounBS,Klo¨tingN,KratzschJ,LeeN,ParkJW,SongES, etal.Serumvaspinconcentrationsinhumanobesityand type2diabetes.Diabetes2008;57:372–7.
[46] ChangHM,ParkHS,ParkCY,SongYS,JangYJ.Association betweenserumvaspinconcentrationsandvisceraladipose tissueinKoreansubjects.Metabolism2010;59:1276–81.
[47] SongY,MansonJE,TinkerL,HowardBV,KullerLH,NathanL, etal.Insulinsensitivityandinsulinsecretiondeterminedby homeostasismodelassessmentandriskofdiabetesina multiethniccohortofwomen:theWomen’sHealthInitiative ObservationalStudy.DiabetesCare2007;
30:1747–52.
