Open Collections

UBC Faculty Research and Publications

Total and high molecular weight adiponectin and ethnic-specific differences in adiposity and insulin… Sulistyoningrum, Dian C; Gasevic, Danijela; Lear, Scott A; Ho, Joe; Mente, Andrew; Devlin, Angela M Nov 13, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12933_2013_Article_755.pdf [ 228.88kB ]
JSON: 52383-1.0215999.json
JSON-LD: 52383-1.0215999-ld.json
RDF/XML (Pretty): 52383-1.0215999-rdf.xml
RDF/JSON: 52383-1.0215999-rdf.json
Turtle: 52383-1.0215999-turtle.txt
N-Triples: 52383-1.0215999-rdf-ntriples.txt
Original Record: 52383-1.0215999-source.json
Full Text

Full Text

ORIGINAL INVESTIGATION Open AccessTotal and high molecular weight adiponectin andethnic-specific differences in adiposity and insulinresistance: a cross-sectional studyDian C Sulistyoningrum1, Danijela Gasevic2, Scott A Lear2,3, Joe Ho1, Andrew Mente4 and Angela M Devlin1*AbstractBackground: Ethnic-specific differences in insulin resistance (IR) are well described but the underlying mechanismsare unknown. Adiponectin is an insulin sensitizing adipocytokine that circulates as multiple isoforms, with highmolecular weight (HMW) adiponectin associated with greatest insulin sensitivity. The objective of this study is todetermine if plasma total and HMW adiponectin concentrations underlie ethnic-specific differences in IR.Methods: Healthy Canadian Aboriginal, Chinese, European, and South Asian adults (N = 634) were assessed forsociodemographics; lifestyle; fasting plasma insulin, glucose, and total and HMW adiponectin; and adipositymeasures [BMI, waist circumference, waist-to-hip ratio, percent body fat, and subcutaneous and visceral adiposetissue (quantified by computed tomography)]. The homeostasis model assessment-insulin resistance (HOMA-IR)assessed IR.Results: South Asians had the greatest HOMA-IR, followed by Aboriginals, Chinese, and Europeans (P < 0.001).Plasma total and HMW adiponectin concentrations were lower in Chinese and South Asians than Aboriginal andEuropeans (P < 0.05). Total and HMW adiponectin were inversely associated with HOMA-IR (P < 0.001). Ethnicitymodified the relationship between HMW adiponectin and HOMA-IR with stronger effects observed in Aboriginals(P = 0.001), Chinese (P = 0.002), and South Asians (P = 0.040) compared to Europeans. This was not observed for totaladiponectin (P = 0.431). At mean total adiponectin concentrations South Asians had higher HOMA-IR than Europeans(P < 0.001).Conclusions: For each given decrease in HMW adiponectin concentrations a greater increase in HOMA-IR is observedin Aboriginals, Chinese, and South Asians than Europeans. Ethnic-specific differences in HMW adiponectin may accountfor differences in IR.Keywords: High molecular weight adiponectin, Ethnicity, Insulin resistance, Visceral adipose tissueIntroductionInsulin resistance (IR) plays an important role in theetiology of type 2 diabetes [1]. Rates of type 2 diabetesand IR are increasing in certain ethnic populations, suchas Aboriginal, Chinese, and South Asian compared toEuropean populations [2-4]. However the underlyingmechanisms accounting for these increased rates areunknown.Adiponectin is an insulin-sensitizing adipocytokinethat may contribute to ethnic-specific differences in IR.Circulating adiponectin concentrations are inversely pro-portional to adiposity [5], and adiponectin circulates asmultiple isoforms characterized as low, medium, andhigh molecular weight (HMW), with the HMW isoformbeing the most abundant and demonstrating the greatestinsulin sensitizing properties [6,7].Few studies have explored ethnic-specific differencesin the relationship between adiponectin and IR andthe findings have been inconsistent. One investigationreported no relationship in South Asian and EuropeanSouth Africans [8]. Whereas others observed a stronger* Correspondence: adevlin@cfri.ubc.ca1Department of Pediatrics, University of British Columbia, Child and FamilyResearch Institute, 272-950 West 28th Avenue, Vancouver V5Z 4H4, CanadaFull list of author information is available at the end of the articleCARDIOVASCULAR DIABETOLOGY© 2013 Sulistyoningrum et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of theCreative Commons Attribution License (, which permits unrestricted use,distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons PublicDomain Dedication waiver ( applies to the data made available in thisarticle, unless otherwise stated.Sulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 between total and HMW adiponectin con-centrations and IR in Europeans than in South Asians[9] and a stronger relationship in South Asian andAboriginals compared to Chinese and Europeans [10].However, these studies only considered surrogate mea-sures of adiposity, such as BMI or waist circumference,when assessing the relationship between adiponectinand IR, which may account for the discrepant findings.Further studies are required that consider body fatdistribution, especially visceral adipose tissue (VAT),given that VAT is strongly associated with IR [11] andVAT deposition differs according to ethnicity [12].Furthermore, despite the evidence of ethnic-specificdifferences in body fat distribution [12] little is knownregarding ethnic-specific differences in the relationshipbetween adiposity and adiponectin. Most prior studieshave only considered BMI and waist circumference[10,13,14]. Direct measures of adiposity have rarely beenutilized and were predominantly explored in African,Caucasian, and Hispanic American populations [15,16].Less is known about the relationship between adiposityand adiponectin in other ethnic groups like Aboriginals,Chinese, and South Asians and whether it differs fromthat of Europeans.The objectives of this study are: i) to explore ethnic-specificdifferences in the relationship between adiposity (bodycomposition and body fat distribution) and plasma totaland HMW adiponectin concentrations; and ii) to investi-gate whether ethnicity modifies the relationship of plasmatotal and HMW adiponectin concentrations with IR inde-pendently of visceral adiposity in a multi-ethnic populationof Aboriginals, Chinese, Europeans, and South Asians.MethodsStudy participants and assessmentsStudy participants were recruited as part of the Multicul-tural-Community Health Assessment Trial (M-CHAT)[17]. As previously described [12], the cohort includes ap-parently healthy men and women (aged 30 to 65 years old)who were of exclusive European (continental Europe,Ireland, and the United Kingdom), Chinese (China, HongKong, and Taiwan), Aboriginal (reserve and non-reserveresidents), or South Asian (Bangladesh, India, Nepal,Pakistan, and Sri Lanka) descent living in and aroundVancouver, Canada. Only those study participants of Ab-original descent with at least 3 grandparents of exclusiveAboriginal origin were included in the study due to highprevalence of mixed ethnic origins in Aboriginal popula-tions. At the time of recruitment and enrolment in thestudy, participants had no known cardiovascular disease ortype 1 or 2 diabetes diagnoses and must have had a stablebody weight in the past 3 months (no more than 2 kgof body weight lost or gained). For non-Aboriginals, im-migrants (who have lived in Canada for more than 3 yearsto allow for acculturation) or first or second generationCanadians were included. The recruitment for study partic-ipants was done in such a way that there is an equal re-presentation of participants with healthy, overweight,and obese BMI in each ethnic group. The study was ap-proved by the University of British Columbia, Children andWomen’s Health Centre of British Columbia, ProvidenceHealth Care, and Simon Fraser University Research EthicsBoards. All study participants provided informed consent.Participants were assessed for age, sex (self-report),sociodemographics (education and income levels), familyhistory of type 2 diabetes, and lifestyle factors (smokingstatus). Total daily energy intakes were quantified bycollecting a 3-d food record and analyzed by a registereddietitian using commercially available software (FoodProcessor SQL software, ESHA, Salem, OR). Leisure-time physical activity was assessed as the average minutesper week of activity over the previous year (self-report)as previously described [12].Assessment of insulin resistanceInsulin resistance was calculated using the previouslyvalidated homeostasis model assessment of insulin re-sistance (HOMA-IR) [18]. Fasting plasma glucose con-centrations were determined by a glucose hexokinase IImethod using an ADVIA 1650 analyzer (Bayer HealthCare, Morristown, NJ). Fasting plasma insulin concen-trations were quantified by a commercial ELISA assayusing Immulite 2500 analyzer (Diagnostic Products, LosAngeles, CA). Coefficients of variations for intra- andinter-assay precision for were less than 7.4%.Adiposity and body fat distributionBody mass index was calculated by body weight (kg)/height (m)2. Waist circumference (WC) was measuredby taking the average of two readings at the point ofmaximal narrowing. Hip circumference was recorded asthe mean of two measures taken at the point of maximalgluteal protuberance from the lateral view over under-garments. Waist/hip ratio (WHR) was calculated bydividing WC by hip circumference. Total body fat wasmeasured by dual-energy X-ray absorptiometry with aNorland XR-36 scanner (Norland Medical Systems,White Plains, NY). The percentage of total body fat wascalculated by dividing total body fat by total body mass.A computed tomography scan was used to take a single10 mm-thick cross-sectional slice at the lumbar 4/5(L4/L5) of the inter-vertebral disc to quantify totalabdominal adipose tissue and VAT. Surface area of thescan was read using the CTi Advantage Scanner (GE,Milwaukee, WI) and the computation of surface areaswere performed using SliceOmatic v.4.2. (Tomovision,Montreal). Total abdominal adipose tissue is defined asthe total area in cm2 within an attenuation range ofSulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 2 of 8−190) to (−30) Hounsfield units for adipose tissue inthe L4/L5 cross-sectional slice image. VAT is the totalarea within the aforementioned attenuation range thatfalls within the abdominal wall. Subcutaneous adiposetissue (SAT) was calculated as the difference betweentotal abdominal adipose tissue and VAT.Quantification of plasma adiponectin concentrationsPlasma total and HMW adiponectin concentrations werequantified using a commercially available ELISA (ALPCODiagnostics, Salem, NH). HMW adiponectin concentra-tions were quantified by pretreating plasma with proteinaseK, which selectively digests low and medium molecularweight adiponectin. The inter-assay (between plates) coeffi-cient of variability (CV) was 6.15% and the intra-assayCV was 2.4%.Statistical analysesOf the 822 potentially available M-CHAT study partici-pants, we excluded 188 because of lack of informationon HOMA-IR, total- or HMW adiponectin, which left634 individuals for our analyses. Distribution of continu-ous variables was explored for normality, and variablesthat were not normally distributed (HOMA-IR, total-and HMW-adiponectin, VAT, SAT, physical activity)were natural logarithm (ln) transformed. Continuousvariables are presented as means (95% confidence inter-val [CI]) or geometric means (95% CI) if not normallydistributed. Ethnic differences in categorical variableswere explored using Chi-square test. The difference incontinuous variables was investigated by analysis ofvariance (ANOVA) or general linear modeling where theanalyses were controlled for age and sex. Post hoc pair-wise comparisons were performed using Bonferroni teststo adjust for multiple comparisons.Linear regression models were used to determine therelationship between adiposity measures (BMI, WC,waist/hip ratio, percent total body fat, SAT and VAT; in-dependent variables) and total- and HMW- adiponectin(dependent variables). Separate models were used foreach adiposity variable and each of the adiponectin vari-ables, while adjusting for covariates (age, sex, ethnicity,education, smoking status, physical activity, and totaldietary energy intake). To explore whether ethnicitymodifies the relationship between adiposity and adipo-nectin, we investigated the interactions of ethnicity witheach of the adiposity measures. We found that ethnicitydid not modify the relationship of adiposity with eithertotal adiponectin or HMW adiponectin (p > 0.05 for allethnicity by adiposity interactions). Furthermore, giventhe sex differences in body fat accumulation, weadditionally explored whether sex modifies the relation-ship of adiposity with total and HMW adiponectin.Given that all interactions were not significant (p > 0.05for all), the relationship between adiposity and adiponec-tin was explored on a total sample.The relationship between total and HMW adiponectin(independent variables) and HOMA-IR (dependent vari-able) was examined using linear regression. We exploredinteractions of ethnicity with both total and HMW adi-ponectin in order to explore whether ethnicity modifiesthe relationship between adiponectin and IR. While eth-nicity was found not to modify the association betweentotal adiponectin and IR (ethnicity by total adiponectininteraction, p = 0.431), the ethnicity by HMW adiponec-tin interaction was found significant (p = 0.010). There-fore, the relationship between total adiponectin and IRwas explored on a total sample after adjusting for ethni-city and other covariates of interest (age, sex, familyhistory of diabetes mellitus, education, smoking status,physical activity, total dietary energy intake, BMI andWHR). Ethnic specific differences in the relationship be-tween HMW adiponectin and IR were also exploredusing linear regression including the interaction term andadjusting for the above-mentioned covariates. To evaluatewhether the association of total- and HMW- adiponectinconcentrations with HOMA-IR was independent of VAT,models were additionally adjusted for VAT.All regression models were checked for multicolinear-ity; variance inflation indices and tolerance statisticswere well below 10 and above 0.2, respectively, indicat-ing that there was no collinearity in our data. Statisticalanalyses were performed using Statistical Package forSocial Sciences (IBM SPSS version 19). Results areconsidered significant if p < 0.05 (two-sided).ResultsStudy participant characteristicsSignificant ethnic-specific differences in age, education,current smoking, physical activity, dietary energy intake,measures of adiposity, plasma glucose concentrations, andHOMA-IR were observed (Table 1). Specifically, Chineseand Europeans were significantly older, had a higher levelof formal education, and higher dietary total energy intakecompared to Aboriginals and South Asians. Furthermore,Aboriginals and Europeans were more likely to be currentsmokers, but engage in greater physical activity thanChinese and South Asians. Aboriginals had significantlyhigher BMI, WC, and WHR compared to the other ethnicgroups, whereas Chinese participants had the lowest BMIand WC. Percent body fat, VAT, SAT, plasma insulinconcentrations, and HOMA-IR were significantly hig-her among Aboriginals and South Asians compared toChinese and Europeans (P < 0.05 for all).Adiponectin, ethnicity, and insulin resistanceTotal and HMW adiponectin concentrations were simi-lar in Aboriginals and Europeans and significantly higherSulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 3 of 8 to Chinese and South Asians (Table 1). Totaland HMW adiponectin concentrations were significantlyinversely associated with all measures of adiposity(BMI, WC, WHR, percent body fat, VAT, and SAT) afteradjustments for age, sex, ethnicity, education, smokingstatus, physical activity and total dietary energy intakes(Table 2). The strongest relationships were observed be-tween WHR and ln total adiponectin [B (95% CI): -2.118(−2.714, -1.523), P < 0.001], and between WHR and lnHMW adiponectin concentrations [B (95% CI): -2.663(−3.609, -1.717), P < 0.001].A significant inverse association between plasma totaladiponectin concentration and HOMA-IR was foundafter adjustment for age, sex, ethnicity, family history ofdiabetes mellitus, education, smoking status, physical ac-tivity, dietary total energy intake, BMI, and WHR (Model1) (Table 3). In this model, ethnicity had a significant effecton HOMA-IR. Specifically, HOMA-IR was significantlyhigher in South Asians compared to Europeans [B (95%CI): 0.203 (0.087, 0.318), P = 0.001]. When the model wasfurther adjusted for VAT (Model 2, Table 3), the relation-ship between plasma total adiponectin concentration andTable 1 Characteristics of study participants by ethnic groupAboriginal (n = 142) Chinese (n = 163) European (n = 164) South Asian (n = 165) Overall p valueAge (years) 45.7 (44.4-47.1) 48.1 (46.7-49.4) 51.1 (49.8-52.4) 45.1 (43.8-46.4) < 0.001 b, d, e, fFemales 75 (52.8%) 90 (55.2%) 78 (47.6%) 83 (50.3%) 0.550Education < 0.001High school or less 55 (38.7%) 53 (32.5%) 33 (20.1%) 68 (41.2%)More than high school 87 (61.3%) 110 (67.5%) 131 (79.9%) 97 (58.8%)Current smoker 41 (28.9%) 6 (3.7%) 13 (7.9%) 6 (3.6%) < 0.001Physical activity (min/week) †* 288.3 (240.6-345.5) 162.1 (136.6-192.5) 247.6 (208.9-293.5) 145.5 (122.6-172.4) < 0.001 a, c, d, fDiet (total energy, kcals) † 1821 (1725–1918) 2010 (1923–2097) 2100 (2012–2188) 1755 (1665–1845) < 0.001 a, b, e, fBMI (kg/m2)† 29.2 (28.4-29.9) 25.7 (25.0-26.4) 27.7 (27.0-28.4) 27.7 (27.1-28.4) < 0.001 a-eWC (cm) † 95.7 (94.0-97.4) 83.5 (81.9-90.8) 89.2 (87.6-90.8) 88.7 (87.1-90.3) < 0.001 a-eWHR† 0.94 (0.93-0.95) 0.88 (0.87-0.88) 0.87 (0.86-0.88) 0.89 (0.88-0.90) < 0.001 a-cPercent body fat (%)† 35.0 (33.9-36.0) 30.8 (29.8-31.7) 31.9 (31.0-32.9) 36.4 (35.4-37.4) < 0.001 a, b, e, fVAT (cm2)†* 117.6 (109.1-126.8) 92.4 (86.2-99.1) 95.8 (89.2-102.8) 118.5 (110.5-127.1) < 0.001 a, b, e, fSAT (cm2)†* 308.3 (287.4-330.6) 212.3 (199.1-226.6) 260.1 (243.5-277.8) 299.8 (280.9-319.9) < 0.001 a, b, d-fGlucose (mmol/L) †* 5.25 (5.15-5.35) 5.27 (5.18-5.37) 5.10 (5.01-5.20) 5.35 (5.25-5.45) < 0.001 fInsulin (μU/mL) †* 72.60 (66.35-79.51) 60.70 (55.76-66.02) 57.69 (52.93-62.87) 77.71 (71.38-84.61) < 0.001 a, b, e, fHOMA-IR†* 2.44 (2.21-2.70) 2.05 (1.86-2.25) 1.88 (1.71-2.07) 2.66 (2.42-2.92) < 0.001 b, e, fTotal adiponectin (μg/mL) †* 5.17 (4.80-5.56) 4.15 (3.88-4.45) 5.18 (4.83-5.56) 4.41 (4.12-4.73) < 0.001 a, c, d, fHMW adiponectin (μg/mL) †* 2.45 (2.18-2.75) 1.59 (1.43-1.77) 2.35 (2.11-2.62) 1.84 (1.65-2.05) < 0.001 a, c, d, fContinuous data presented as mean (95% CI). Categorical data presented as n (%).†Means are adjusted for age and sex. *Geometric means (95% CI).The overall p-value: ethnic differences in continuous and categorical variables were explored by ANOVA and Chi-square tests, respectively. Post hoc pairwisecomparisons: ap < 0.05, Aboriginal vs. Chinese, bp < 0.05, Aboriginal vs. European, cp < 0.05, Aboriginal vs. South Asian, dp < 0.05, Chinese vs. European, ep < 0.05,Chinese vs. South Asian, fp < 0.05, European vs. South Asian.Table 2 Association between adiposity and adiponectinln total adiponectin ln HMW adiponectinB (95% CI) Standardized B p value B (95% CI) Standardized B p valueBMI −0.021 (−0.030, -0.013) −0.191 < 0.001 −0.035 (−0.048, -0.022) −0.199 < 0.001WC −0.011 (−0.014, -0.007) −0.259 < 0.001 −0.017 (−0.023, -0.011) −0.255 < 0.001WHR −2.118 (−2.714, -1.523) −0.385 < 0.001 −2.663 (−3.609, -1.717) −0.305 < 0.001% BF −0.007 (−0.013, -0.001) −0.136 0.019 −0.011 (−0.021, -0.002) −0.134 0.019ln VAT −0.267 (−0.350, -0.184) −0.264 < 0.001 −0.339 (−0.471, -0.208) −0.210 < 0.001ln SAT −0.120 (−0.209, -0.031) −0.111 0.008 −0.225 (−0.363, -0.086) −0.131 0.002All linear regression models were adjusted for age, sex, ethnicity, formal level of education, smoking status, physical activity, and energy intake.B, beta-value; BMI, body mass index; % BF, percent body fat; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue; WC, waist circumference; WHR,waist-to-hip ratio.Sulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 4 of 8 was slightly attenuated but remained signifi-cant [B (95% CI): -0.300 (−0.385, -0.214), P < 0.001].The effect of ethnicity on HOMA-IR was also attenu-ated with inclusion of VAT in the model; however,HOMA-IR remained significantly higher in SouthAsians compared to Europeans [B (95% CI): 0.155(0.044, 0.266), P = 0.006] (Model 2, Table 3). While theassociation between total adiponectin and HOMA-IRwas independent of ethnicity; ethnicity was foundto modify the relationship between plasma HMW adi-ponectin concentration and HOMA-IR (Figure 1).Compared to Europeans, the effect of HMW-adiponec-tin on HOMA-IR was significantly different in Abo-riginals (B (95% CI) = −0.238 (−0.391, -0.086), P = 0.002),Chinese (B (95% CI) = −0.199 (−0.328, -0.069), P = 0.003,and South Asians (B (95% CI) = −0.146 (−0.285, -0.008), P =0.038). The addition of VAT did not substantiallychange the modifying effect of ethnicity on the relation-ship between HMW adiponectin and HOMA-IR.Namely, after additional adjustment for VAT, the effectof HMW-adiponectin on HOMA-IR in Aboriginals(B (95% CI) = −0.236 (−0.381, -0.092), P = 0.001), Chinese(B (95% CI) = −0.195 (−0.318, -0.072), P = 0.002), andSouth Asians (B (95% CI) = −0.137 (−0.268, -0.006),P = 0.040) was still significantly different from that ofEuropeans.DiscussionThe current study was designed to explore ethnic-specific differences in the relationship between directmeasures of adiposity and plasma total and HMW adi-ponectin concentrations in a multi-ethnic population ofAboriginals, Chinese, Europeans, and South Asians. Wefurther determined whether ethnicity modifies the rela-tionship of plasma total and HMW adiponectin concen-trations with IR, independent of visceral adiposity. Wereport that all measures of adiposity, surrogate (BMI,WC, WHR) and direct (percent body fat, SAT, andVAT), are inversely associated with plasma total andHMW adiponectin concentrations but that ethnicity hasno effect on these relationships. We also found thatplasma total and HMW adiponectin concentrations areinversely associated with HOMA-IR and that inclusionof VAT in the model only slightly attenuated the rela-tionship. Most interestingly, we found that ethnicitymodified the relationship between HMW adiponectinand IR in that for every decrease in HMW adiponectinconcentration a greater increase in HOMA-IR wasobserved in Aboriginals, Chinese, and South Asians thanin Europeans. This ethnic-specific effect was not ob-served for the relationship between total adiponectinand IR.Adiponectin is an insulin-sensitizing adipocytokine thatmay be a mediating factor that underlies ethnic-specificdifferences in IR. Prior studies by others have been incon-clusive [8-10,19,20] and little is known regarding the roleof body composition and body fat distribution in this rela-tionship. We have previously reported that South Asiansand Aboriginals have significantly higher VAT depositioncompared to Chinese and Europeans [12] and this is ac-companied by ethnic-differences in IR [21]. Most import-antly, in the current study, we found that ethnicitymodifies the relationship between plasma HMW adipo-nectin concentrations and HOMA-IR, in that for eachgiven decrease in plasma HMW adiponectin concentra-tion, a greater increase in HOMA-IR was observed inAboriginals, Chinese, and South Asians compared toEuropeans. The same interaction was not observed forplasma total adiponectin concentrations. Taken togetherour findings suggest that HMW adiponectin may play arole in the ethnic-specific differences in IR, which was pre-viously attributed, in part, to ethnic-specific differences inVAT deposition. Furthermore, the differences in HMWadiponectin concentrations may also account for differ-ences in diabetes severity and cardiovascular complicationsobserved among different indigenous populations [22].As has been reported by others, our findings that eth-nicity had no effect on the relationship between plasmatotal and HMW adiponectin concentrations and surro-gate measures of adiposity (BMI, WC, WHR), percentbody fat, and body fat distribution (VAT and SAT) is inTable 3 Association between total adiponectin andHOMA-IROutcome: in HOMA-IR Total adiponectinB (95%CI) p valueModel 1ln Total adiponectin −0.331 (−0.421, -0.242) < 0.001EthnicityAboriginal vs. European 0.001 (−0.126, 0.125) 0.995Chinese vs. European 0.061 (−0.052, 0.174) 0.287South Asian vs. European 0.203 (0.087, 0.318) 0.001BMI 0.054 (0.043, 0.064) < 0.001WHR 2.280 (1.530, 3.031) < 0.001Model 2ln Total adiponectin −0.300 (−0.385, -0.214) < 0.001EthnicityAboriginal vs. European 0.022 (−0.098, 0.141) 0.722Chinese vs. European 0.061 (−0.046, 0.169) 0.264South Asian vs. European 0.155 (0.044, 0.266) 0.006BMI 0.030 (0.018, 0.041) < 0.001WHR 1.251 (0.488, 2.013) 0.001VAT 0.473 (0.351, 0.594) < 0.001Outcome variable is ln HOMA-IR. Model 1 is adjusted for ethnicity, age, sex,family history of diabetes, formal level of education, BMI, WHR, smoking status,physical activity and energy intake. Model 2 = Model 1 + VAT.BMI, body mass index; VAT, visceral adipose tissue; WHR, waist-to-hip ratio.Sulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 5 of 8 to the findings of another Canadian study thatreported an inverse relationship between surrogate mea-sures of adiposity and plasma total adiponectin concen-trations in Aboriginals, Chinese, and Europeans but notin South Asians [10]. However, similar to our findings,an inverse relationship between serum total adiponectinconcentrations and VAT was reported in a population ofSouth Asians living in California [19].Adipose tissue is an endocrine organ that releases anumber of adipocytokines, such as leptin and adiponec-tin, which play important roles in appetite regulationand cellular energy metabolism [11]. In addition, adipo-nectin has insulin-sensitizing functions and may contrib-ute to ethnic-specific differences in IR. Similar to whatwe report here, another group reported that plasma totaland HMW adiponectin concentrations were also lowerin South Asians compared to Europeans but an inverserelationship with IR was only observed in Europeansand not in South Asians [9]. Lower total adiponectinconcentrations were also reported in a population ofSouth Asians compared to African or Chinese subjectsfrom Trinidad [20]. In our current study we report theopposite effect, with the strongest relationship betweenplasma HMW adiponectin concentrations and IR foundin Aboriginals, Chinese, and South Asians compared toEuropeans. Interestingly, we did not find an ethnic-specific effect on the relationship between plasma totaladiponectin concentrations and IR. This is in contrast tothe findings of another group that did report an effect ofethnicity on the inverse relationship between plasmatotal adiponectin concentrations and IR with SouthAsians and Aboriginals showing the strongest relation-ship compared to Europeans and Chinese [10].The role of adiponectin in the maintenance of glucose-insulin homeostasis has been investigated in the past twodecades. A study in isolated epididymal rat adipocytesincubated with globular recombinant adiponectin reportedthat adiponectin facilitated insulin-stimulated glucoseuptake via the activation of AMP-kinase pathway [23]. Astudy in male C57BL/6 J, ob/ob, and female non-obesediabetic mice showed that an injection with 84 μg/g bodyweight of recombinant full-length adiponectin reducedserum glucose concentrations and subsequently reducedhyperglycemia in the male ob/ob and female NOD mice[24]. In the same study, it was also reported that the pres-ence of 50 μg/mL of full-length adiponectin on isolated pri-mary hepatocytes increased the ability of insulin tosuppress glucose production. Furthermore, variants in thegene encoding one of the adiponectin receptors (ADIPOR2)were reported to be associated with increased risk of devel-oping type 2 diabetes and greater cardiovascular diseaserisk factors in individuals with glucose intolerance [25].In the current study we also considered body fat distri-bution and specifically we were interested in investigatingthe interplay between adiponectin and VAT as predictorsof HOMA-IR because of the evidence supporting a stronginverse association between adiponectin and VAT [26-28].A few studies have investigated whether adiponectinsynthesis and secretion is influenced by body fat location.A study in adipose tissue explants obtained from subjectsundergoing abdominal surgery reported that total adipo-nectin secretion by adipocytes from SAT explants waslower in obese than in non-obese subjects but this was notobserved for adipocytes from VAT explants [29]. Thesefindings suggest that adiponectin secretion by adiposetissue may depend on body fat location. In line with this,we report that VAT has a stronger inverse association withplasma total and HMW adiponectin concentrations thanSAT. In agreement with our findings, a study in culturedadipocytes from human omental (VAT) adipose tissue andSAT reported that adiponectin secretion by the omentaladipocytes was lower than that by SAT adipocytes [30].Figure 1 Interaction between plasma HMW adiponectin concentrations and ethnicity on HOMA-IR. The relationship betweenHMW-adiponectin on HOMA-IR in Aboriginals (dashed line), Chinese (solid dark grey line) and South Asians (solid light grey line) is different fromthat of Europeans (solid black line).Sulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 6 of 8 study also reported that adiponectin secretion by theomental adipocytes was increased in the presence insulin, aneffect that was but not observed in adipocytes from SAT,further supporting a role for adiponectin and VAT in IR.Study limitationsOur study has some limitations. We used HOMA-IRto assess IR, which is inferior to the euglycemic-hyperinsulinemic clamp method to assess IR. However,the use of HOMA-IR as a surrogate marker of IR is lessinvasive and more feasible in large population studies[18]. Furthermore, given our study is cross-sectional indesign we cannot determine the causal-effect of adipo-nectin on HOMA-IR. Despite these limitations, ourstudy has strengths, which include our elaborate assess-ment of adiponectin by quantifying plasma total andHMW adiponectin concentrations. This is importantbecause HMW adiponectin is suggested to be the mostabundant and have the greatest insulin-sensitizing prop-erties [6,7]. Furthermore, we included direct measures ofbody fat distribution (VAT) in our models, which ismore strongly associated with IR compared to othermeasures of adiposity [11]. Other limitations of ourstudy are the differences in age range and the relativelysmall sample size of the different ethnic groups.In summary, the relationship between plasma HMWadiponectin and HOMA-IR is influenced by ethnicitywith the greater effect observed for Aboriginals, Chinese,and South Asians compared to Europeans. Even afteradjustment for VAT, the association between adiponectinand HOMA-IR remained significant, suggesting that adi-ponectin plays a role in IR across different ethnic groupsindependent of VAT.AbbreviationsANOVA: Analysis of variance; CI: Confidence interval; HOMA-IR: Homeostasismodel assessment of insulin resistance; HMW: High molecular weight;IR: Insulin resistance; M-CHAT: Multicultural-community health assessmentTrial; SAT: Subcutaneous adipose tissue; VAT: Visceral adipose tissue;WC: Waist circumference; WHR: Waist-hip ratio.Competing interestsThe authors declare no conflicts of interest to disclose from any of theauthors.Authors’ contributionsSAL and AMD conceived and designed the research. DCS, JH, and SALperformed the research. DCS, DG, and AMD analyzed the data. DCS, DG, andAMD wrote the manuscript. SAL and AMD had primary responsibility for finalcontent. All authors read and approved the final manuscript.AcknowledgementsSupported by a grant from the Canadian Institutes of Health Research (toAMD). DCS was supported by a graduate studentship from the Child andFamily Research Institute. SAL is a Pfizer/Heart and Stroke Foundation ofCanada Chair in Cardiovascular Disease Prevention Research. AMD issupported by an Investigator Award from the Child and Family ResearchInstitute.Author details1Department of Pediatrics, University of British Columbia, Child and FamilyResearch Institute, 272-950 West 28th Avenue, Vancouver V5Z 4H4, Canada.2Department of Biomedical Physiology and Kinesiology, Simon FraserUniversity, 8888 University Dr, Burnaby BC V5A 1S6, Canada. 3Division ofCardiology, Providence Health Care, Vancouver, Canada. 4Population HealthResearch Institute, Hamilton Health Sciences, McMaster University, Hamilton,Canada.Received: 30 September 2013 Accepted: 8 November 2013Published: 13 November 2013References1. Warram JH, Martin BC, Krolewski AS, Soeldner JS, Kahn CR: Slow glucoseremoval rate and hyperinsulinemia precede the development of type IIdiabetes in the offspring of diabetic parents. Ann Intern Med 1990,113(12):909–915.2. Young TK, Reading J, Elias B, O’Neil JD: Type 2 diabetes mellitus inCanada’s first nations: status of an epidemic in progress. CMAJ 2000,163(5):561–566.3. Razak F, Anand S, Vuksan V, Davis B, Jacobs R, Teo KK, Yusuf S: Ethnicdifferences in the relationships between obesity and glucose-metabolicabnormalities: a cross-sectional population-based study. Int J Obes (Lond)2005, 29(6):656–667.4. Weber MB, Oza-Frank R, Staimez LR, Ali MK, Venkat Narayan KM: Type 2diabetes in asians: prevalence, risk factors, and effectiveness ofbehavioral intervention at individual and population levels. Ann Rev Nutr2012, 32(1):417–439.5. Cook JR, Semple RK: Hypoadiponectinemia cause or consequence ofhuman “insulin resistance”? J Clin Endocrinol Metab 2010, 95(4):1544–1554.6. Lara-Castro C, Luo N, Wallace P, Klein RL, Garvey WT: Adiponectinmultimeric complexes and the metabolic syndrome trait cluster. Diabetes2006, 55(1):249–259.7. Hara K, Horikoshi M, Yamauchi T, Yago H, Miyazaki O, Ebinuma H, Imai Y,Nagai R, Kadowaki T: Measurement of the high–molecular weight form ofadiponectin in plasma is useful for the prediction of insulin resistanceand metabolic syndrome. Diabetes Care 2006, 29(6):1357–1362.8. Ferris WF, Naran NH, Crowther NJ, Rheeder P, Van der Merwe L, Chetty N:The relationship between insulin sensitivity and serum adiponectinlevels in three population groups. Horm Metab Res 2005, 37(11):695–701.9. Martin M, Palaniappan LP, Kwan AC, Reaven GM, Reaven PD: Ethnicdifferences in the relationship between adiponectin and insulinsensitivity in south asian and caucasian women. Diabetes Care 2008,31(4):798–801.10. Mente A, Razak F, Blankenberg S, Vuksan V, Davis AD, Miller R, Teo K,Gerstein H, Sharma AM, Yusuf S, et al: Ethnic variation in adiponectin andleptin levels and their association with adiposity and insulin resistance.Diabetes Care 2010, 33(7):1629–1634.11. Tchernof A, Després J-P: Pathophysiology of human visceral obesity: anupdate. Physiol Rev 2013, 93(1):359–404.12. Lear SA, Humphries KH, Kohli S, Chockalingam A, Frohlich JJ, BirminghamCL: Visceral adipose tissue accumulation differs according to ethnicbackground: results of the multicultural community health assessmenttrial (M-CHAT). Am J Clin Nutr 2007, 86(2):353–359.13. Khoo CM, Sairazi S, Taslim S, Gardner D, Wu Y, Lee J, Van Dam RM, ShyongTai E: Ethnicity modifies the relationships of insulin resistance,inflammation, and adiponectin with obesity in a multiethnic asianpopulation. Diabetes Care 2011, 34(5):1120–1126.14. Hulver MW, Saleh O, MacDonald KG, Pories WJ, Barakat HA: Ethnicdifferences in adiponectin levels. Metabolism 2004, 53(1):1–3.15. Hanley AJG, Bowden D, Wagenknecht LE, Balasubramanyam A, Langfeld C,Saad MF, Rotter JI, Guo X, Chen Y-DI, Bryer-Ash M, et al: Associations ofadiponectin with body Fat distribution and insulin sensitivity in nondiabetichispanics and african-americans. J Clin Endocrinol Metab 2007,92(7):2665–2671.16. Bidulescu A, Liu J, Hickson DA, Hairston KG, Fox ER, Arnett DK, Sumner AE,Taylor HA, Gibbons GH: Gender differences in the association of visceraland subcutaneous adiposity with adiponectin in african americans: thejackson heart study. BMC Cardiovasc Disord 2013, 13(9).doi:10.1186/1471-2261-13-9.Sulistyoningrum et al. Cardiovascular Diabetology 2013, 12:170 Page 7 of 8 Lear SA, Birmingham CL, Chockalingam A, Humphries KH: Study design ofthe multicultural community health assessment trial (M-CHAT): acomparison of body fat distribution in four distinct populations. Ethn Dis2006, 16(1):96–100.18. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC:Homeostasis model assessment: insulin resistance and β-cell functionfrom fasting plasma glucose and insulin concentrations in man.Diabetologia 1985, 28(7):412–419.19. Shah A, Hernandez A, Mathur D, Budoff MJ, Kanaya AM: Adipokines andbody fat composition in south asians: results of the metabolic syndromeand atherosclerosis in south asians living in america (MASALA) study.Int J Obes 2012, 36(6):810–816.20. Nayak S, Maharaj N, Fatt L-A: Association between altered lipid profile,body mass index, low plasma adiponectin and varied blood pressure intrinidadian type 2 diabetic and non-diabetic subjects. Ind J Med Sci 2012,66(9–10):214–221.21. Lear SA, Kohli S, Bondy GP, Tchernof A, Sniderman AD: Ethnic variation inFat and lean body mass and the association with insulin resistance.J Clin Endocrinol Metab 2009, 94(12):4696–4702.22. Maple-Brown LJ, Cunningham J, Zinman B, Mamakeesick M, Harris SB,Connelly PW, Shaw J, O’Dea K, Hanley AJ: Cardiovascular disease riskprofile and microvascular complications of diabetes: comparison ofindigenous cohorts with diabetes in australia and canada. CardiovascDiabetol 2012, 11(30). doi:10.1186/1475-2840-1111-1130.23. Wu X, Motoshima H, Mahadev K, Stalker TJ, Scalia R, Goldstein BJ:Involvement of AMP-activated protein kinase in glucose uptake stimu-lated by the globular domain of adiponectin in primary Rat adipocytes.Diabetes 2003, 52(6):1355–1363.24. Berg AH, Combs TP, Du X, Brownlee M, Scherer PE: The adipocyte-secretedprotein Acrp30 enhances hepatic insulin action. Nat Med 2001,7(8):947–953.25. Siitonen N, Pulkkinen L, Lindström J, Kolehmainen M, Schwab U, Eriksson JG,Ilanne-Parikka P, Keinänen-Kiukaanniemi S, Tuomilehto J, Uusitupa M:Association of ADIPOR2 gene variants with cardiovascular disease andtype 2 diabetes risk in individuals with impaired glucose tolerance: theFinnish Diabetes Prevention Study. Cardiovasc Diabetol 2011, 10(83).doi:10.1186/1475-2840-1110-1183.26. Cnop M, Havel PJ, Utzschneider KM, Carr DB, Sinha MK, Boyko EJ, RetzlaffBM, Knopp RH, Brunzell JD, Kahn SE: Relationship of adiponectin to bodyfat distribution, insulin sensitivity and plasma lipoproteins: evidence forindependent roles of age and sex. Diabetologia 2003, 46(4):459–469.27. Cote M, Mauriege P, Bergeron J, Almeras N, Tremblay A, Lemieux I, DespresJP: Adiponectinemia in visceral obesity: impact on glucose tolerance andplasma lipoprotein and lipid levels in men. J Clin Endocrinol Metab 2005,90(3):1434–1439.28. Park K-G, Park KS, Kim M-J, Kim H-S, Suh Y-S, Ahn JD, Park K-K, Chang Y-C,Lee I-K: Relationship between serum adiponectin and leptinconcentrations and body fat distribution. Diabetes Res Clin Pract 2004,63(2):135–142.29. Kovacova Z, Tencerova M, Roussel B, Wedellova Z, Rossmeislova L, Langin D,Polak J, Stich V: The impact of obesity on secretion of adiponectinmultimeric isoforms differs in visceral and subcutaneous adipose tissue.Int J Obes 2012, 36(10):1360–1365.30. Motoshima H, Wu X, Sinha MK, Hardy VE, Rosato EL, Barbot DJ, Rosato FE,Goldstein BJ: Differential regulation of adiponectin secretion fromcultured human omental and subcutaneous adipocytes: effects of insulinand rosiglitazone. J Clin Endocrinol Metab 2002, 87(12):5662–5667.doi:10.1186/1475-2840-12-170Cite this article as: Sulistyoningrum et al.: Total and high molecularweight adiponectin and ethnic-specific differences in adiposity andinsulin resistance: a cross-sectional study. Cardiovascular Diabetology2013 12:170.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at et al. Cardiovascular Diabetology 2013, 12:170 Page 8 of 8


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items