Maki Goto and Atsushi Goto have contributed equally to this work.
Low-molecular-weight adiponectin and high-molecular-weight adiponectin levels in relation to diabetes
Version of Record online: 10 SEP 2013
Copyright © 2013 The Obesity Society
Volume 22, Issue 2, pages 401–407, February 2014
How to Cite
Goto, M., Goto, A., Morita, A., Deura, K., Sasaki, S., Aiba, N., Shimbo, T., Terauchi, Y., Miyachi, M., Noda, M., Watanabe, S. and for the Saku Cohort Study Group (2014), Low-molecular-weight adiponectin and high-molecular-weight adiponectin levels in relation to diabetes. Obesity, 22: 401–407. doi: 10.1002/oby.20553
Funding agencies: This work was supported in part by a grant from the Japan Diabetes Foundation.
Disclosure: The authors declare that there are no conflicts of interest.
Author contributions: MG and AG designed the study, conceived experiments, analyzed data, and wrote the manuscript. AM, KD, SS, NA, MM, and SW conducted the study and contributed to the discussion. TS contributed to the analysis and the discussion. YT, and MN contributed to the discussion, and reviewed/edited the manuscript. The authors acknowledge all members for their participation in this research and also thank Masanobu Ikeda and Tomoko Yasuda for their assistance.
- Issue online: 3 FEB 2014
- Version of Record online: 10 SEP 2013
- Accepted manuscript online: 1 JUL 2013 08:17AM EST
- Manuscript Accepted: 17 JUN 2013
- Manuscript Received: 12 MAR 2013
To evaluate the association between adiponectin complexes (high-molecular-weight [HMW], middle-molecular-weight [MMW], and low-molecular-weight [LMW] adiponectin) and diabetes.
Design and Methods
We conducted a case-control study, based on a cohort in Saku, Japan. Among 2565 participants, 300 participants with diabetes and 300 matched controls (430 men and 170 women) were analyzed.
After adjusting for age, physical activity, hypertension, family history, alcohol use, smoking, and menopausal status, total, HMW, and LMW, but not MMW adiponectin levels were inversely associated with diabetes: total adiponectin, odds ratio comparing the highest with the lowest quartiles, 0.46 (95% confidence interval, 0.25–0.82; P for trend = 0.046); HMW, 0.40 (95%CI, 0.22–0.72; P = 0.046); MMW, 1.04 (95%CI, 0.60–1.77; P = 0.81); and LMW, 0.51 (95%CI, 0.29–0.89; P = 0.01). The associations between total and HMW adiponectin and diabetes attenuated after adjustment for BMI (P = 0.15 and 0.13, respectively), but LMW remained (P = 0.04). When stratified by sex, LMW adiponectin levels were associated with diabetes in men only. None of the associations were significant after adjustment for HOMA-IR.
Decreased LMW, total, and HMW adiponectin levels are associated with diabetes. These associations may be secondary to adiposity or insulin resistance.
Adiponectin, a protein secreted from adipocytes, has been reported to improve insulin sensitivity and reduce the risk of type 2 diabetes ([1-4]). Human adiponectin exists in three multimer forms, high-molecular-weight (HMW), middle-molecular-weight (MMW), and low-molecular-weight (LMW) (). HMW, total adiponectin, and the ratio of HMW to total adiponectin have been reported to be associated with type 2 diabetes risk ([6, 7]). In contrast to HMW adiponectin, the role of LMW adiponectin in the risk of diabetes is less defined. In a study among octogenarians, decreased LMW adiponectin was associated with type 2 diabetes (). Additionally, in a study among type 1 diabetic patients, HMW was upregulated, while MMW and LMW adiponectin was downregulated compared with controls (). Furthermore, studies reported that weight reduction was associated with a reduction in LMW adiponectin levels (), and high LMW adiponectin levels were associated with a decreased risk of developing Barrett's esophagus (). Although the mechanisms remain unclear, LMW adiponectin has been reported to have anti-inflammatory effects; a proposed mechanism is the suppression of IL-6 secretion and stimulation of IL-10 secretion ([12-15]). Furthermore, adiponectin levels differ according to ethnicity; Asian women had lower total and HMW adiponectin levels compared to Caucasian women ().
However, to the best of our knowledge, no previous studies have investigated the associations between adiponectin complexes and diabetes in Asians, with measurement of the HMW, MMW, and LMW multimers. To investigate the associations between adiponectin complexes and diabetes, we conducted a case-control study among Japanese participants.
This study was a cross-sectional analysis of the Saku cohort, which was launched in 2009 at Saku General Hospital Human Dock Center in Saku City, Nagano Prefecture, Japan. Participants who visited for a health checkup between May 5, 2009 to September 30, 2010 and who agreed to participate in the cohort were included in the study. From the study population at baseline (n = 2565), we excluded subjects with missing data (n = 30), age <50 years old (n = 350), and age ≥80 (n = 16). Of the remaining 2169 participants, 301 participants were defined as having diabetes. According to WHO criteria (), diabetes was defined by fasting plasma glucose levels ≥126 mg/dL, 2-h post-load glucose levels ≥200 mg/dL after a 75-g oral glucose tolerance test, or diabetes diagnosed by physicians. Of the remaining 1868 participants, 542 participants who were defined to have impaired glucose tolerance or impaired fasting glucose according to the WHO criteria were excluded. Control participants were randomly selected from the remaining 1326 participants and individually matched to cases based on age and sex (n = 301). Of these 602 participants, 1 case-control pair was excluded from the analysis because of the absence of a remaining serum sample for 1 male case, leaving 300 diabetes cases and 300 matched controls in the analysis. This study was reviewed and approved by the Ethical Committee of the National Institute of Health and Nutrition and Saku Central Hospital. Participants received a precise explanation of the study and provided their written informed consent.
Subjects' height (cm) and weight (kg) were measured with an automatic scale (Tanita, BF-220, Tokyo, Japan). Body mass index (BMI) was calculated as the weight (kg) divided by the squared height (m2). Waist circumference was measured at the umbilicus level while the subject was in a standing position. Blood pressure was measured using a validated automated blood pressure monitor (ES-H55; Terumo, Tokyo, Japan) with subjects in a sitting position. Physical activity levels were obtained by asking the participants about their average frequency of physical activity: rarely/never, one to three times per month, one to two times per week, and more than three times per week.
Following an overnight fast, blood samples were collected when the participants attended a health checkup at the Saku Health Dock Center. Blood samples were collected in tubes containing EDTA and heparin for the measurement of fasting plasma glucose, insulin, and HbA1c levels, and the remaining frozen serum samples were sent to the laboratory at the National Institute for Health and Nutrition and stored in deep freezers. Serum gel separator tubes were used for the measurement of total cholesterol, HDL cholesterol, and triglyceride (TG) levels. Routine laboratory blood analyses were performed at the Saku Central Hospital. HbA1c levels were measured using a high-performance liquid chromatography method (TOSOH HLC-723 G8; Tosoh Corporation, Tokyo, Japan), with intra- and interassay coefficients of variation (CVs) of 0.5–1.4% and 0.6–1.3%, respectively. The values for HbA1c were collected as Japan Diabetes Society (JDS) values and then converted to National Glycohemoglobin Standardization Program (NGSP) values using the following conversion formula: HbA1c (NGSP) = 1.02 × HbA1c (JDS) + 0.25% (). Plasma glucose levels were analyzed using an enzymatic method (ECO glucose buffer; A&T Corporation, Kanagawa, Japan), with intra- and interassay CVs of 0.3–0.5% and 0.6–0.8%, respectively. Plasma insulin levels were analyzed using an electrochemiluminescence immunoassay (Modular E170; Roche Diagnostics, Mannheim, Germany), with intra- and interassay CVs of 0.5–2.0% and 3.2–3.6%, respectively. To evaluate insulin sensitivity, we used the homeostasis model assessment for insulin resistance (HOMA-IR), which was calculated as follows: fasting insulin (µIU/mL) × fasting glucose (mg/dL)/405 (). Serum total cholesterol, HDL cholesterol, and TG concentrations were determined using enzymatic methods (serum total cholesterol: Detaminar L TC II, Kyowa Medex, Tokyo, Japan; HDL cholesterol: Cholestest N HDL, Sekisui Medical Co. Ltd., Tokyo, Japan; and TG concentration: Mizuho TG-FR Type II, Mizuho Medi, Saga, Japan) and an autoanalyzer BM-2250 (Nihon Denshi, Tokyo, Japan), with intra- and interassay CVs of ≤1.7% and ≤2.3%, respectively.
From the stored frozen samples, adiponectin was measured in the laboratory of Saku Central Hospital. Serum levels of HMW (≥octadecamer), MMW (hexamer), LMW (trimer and albumin-binding trimer), and total adiponectin were determined using an enzyme-linked immunosorbent assay kit (Daiichi Pure Chemicals, Ibaraki, Japan) () with intra- and interassay CVs of <7.9% and <9.6%, respectively.
Baseline characteristics were compared between case and control subjects using the paired t-test for continuous variables and McNemar's test for categorical variables. Spearman correlation coefficients (r) were calculated to evaluate associations among HbA1c, HOMA-IR, total, HMW, MMW, LMW adiponectin, BMI, and HDL in controls. We used the student t-test to examine the associations between adiponectin complexes and exercise, alcohol intake, or smoking status.
Odds ratios (ORs) and 95% confidence intervals were calculated according to quartiles based on the joint distribution of cases and controls; the lowest quartile was used as the reference. We fitted conditional logistic regression models to estimate the association between adiponectin levels and diabetes. In Model 1, we stratified matched pairs using conditional logistic regression models. We further adjusted for smoking status (never, past, or current), physical activity (almost none, 1–3 times/week, 4–6 times/week, or every day), history of hypertension, family history of diabetes, alcohol use (almost none, occasional, or regular), and menopausal status (Model 2). We additionally included BMI and HOMA-IR in the models (Models 3 and 4, respectively). P-values for linear trend were computed based on median levels in categories. In addition, we performed stratified analyses according to sex. To assess non-multiplicative interactions of the adiponectin levels with sex, we fitted logistic models with product terms for each of these interactions, treating the biomarkers as continuous variables. Statistical analyses were conducted using STATA (version 12.0; StataCorp, College Station, TX).
Compared to control participants, diabetic participants tended to have lower LMW adiponectin and higher HOMA-IR, BMI, waist circumference, and TG levels (Table 1). The correlation coefficients between HbA1c; HOMA-IR; total, HMW, MMW, LMW adiponectin levels; BMI; and HDL level are shown in Table 2. Among control participants, total, HMW, MMW, and LMW adiponectin levels were positively associated with HDL cholesterol levels and inversely associated with HOMA-IR and BMI (Table 2). Correlation coefficients were similar when stratified by sex (data not shown).
|With diabetes||No diabetes|
|(n = 300)||(n = 300)||P|
|Age (years)||64.1 ± 6.7||64.1 ± 6.7||0.72|
|HbA1c (%)||6.6 ± 1.0||5.6 ± 0.3||<0.001|
|Fasting glucose (mg/dL)||127.5 ± 31.1||97.0 ± 6.5||<0.001|
|HOMA-IR||2.1 ± 2.1||1.0 ± 0.8||<0.001|
|Total adiponectin (μg/mL)||5.2 ± 3.3||5.6 ± 2.9||0.13|
|HMW adiponectin (μg/mL)||2.7 ± 2.4||2.9 ± 2.2||0.17|
|MMW adiponectin (μg/mL)||1.1 ± 0.7||1.1 ± 0.7||0.91|
|LMW adiponectin (μg/mL)||1.4 ± 0.6||1.5 ± 0.6||0.03|
|BMI (kg/m2)||24.1 ± 3.2||22.9 ± 2.8||<0.001|
|Waist circumference (cm)||86.6 ± 8.4||82.9 ± 8.6||<0.001|
|Total cholesterol (mg/dL)||199 ± 29||199 ± 28||0.95|
|Triglycerides (mg/dL)||126 ± 78||106 ± 55||<0.001|
|LDL cholesterol (mg/dL)||118 ± 27||119 ± 24||0.75|
|HDL cholesterol (mg/dL)||57 ± 14||60 ± 16||0.04|
|Serum creatinine (mg/dL)||0.8 ± 0.2||0.8 ± 0.2||0.34|
|(P = 0.09)|
|(P = 0.25)||(P < 0.001)|
|(P = 0.12)||(P < 0.001)||(P < 0.001)|
|(P = 0.75)||(P = 0.004)||(P < 0.001)||(P < 0.001)|
|(P = 0.91)||(P < 0.001)||(P < 0.001)||(P < 0.001)||(P = 0.21)|
|(P = 0.26)||(P < 0.001)||(P < 0.001)||(P < 0.001)||(P = 0.002)||(P = 0.03)|
|(P = 0.84)||(P < 0.001)||(P < 0.001)||(P < 0.001)||(P < 0.001)||(P < 0.001)||(P < 0.001)|
Table 3 shows the relation between adiponectin complexes and exercise, alcohol, or smoking. Total, HMW, and MMW adiponectin levels were higher in non-drinkers and in non-smokers.
|<3 times a week||>3 times a week|
|Exercise||(n = 175)||(n = 125)||P|
|Total adiponectin (μg/mL)||5.7 ± 3.1||5.4 ± 2.6||0.35|
|HMW adiponectin (μg/mL)||3.1 ± 2.5||2.7 ± 1.8||0.20|
|MMW adiponectin (μg/mL)||1.1 ± 0.7||1.1 ± 0.7||0.85|
|LMW adiponectin (μg/mL)||1.5 ± 0.6||1.6 ± 0.6||0.67|
|Alcohol intake||None (n = 110)||>1 time a week (n = 190)||P|
|Total adiponectin (μg/mL)||6.4 ± 2.8||5.1 ± 2.8||<0.001|
|HMW adiponectin (μg/mL)||3.5 ± 2.1||2.6 ± 2.2||<0.001|
|MMW adiponectin (μg/mL)||1.3 ± 0.8||1.0 ± 0.6||0.002|
|LMW adiponectin (μg/mL)||1.6 ± 0.7||1.5 ± 0.6||0.13|
|Smoking status||None (n = 137)||Current or past smoking (n = 163)||P|
|Total adiponectin (μg/mL)||6.4 ± 3.3||4.9 ± 2.3||<0.001|
|HMW adiponectin (μg/mL)||3.6 ± 2.6||2.4 ± 1.7||<0.001|
|MMW adiponectin (μg/mL)||1.3 ± 0.8||1.0 ± 0.5||<0.001|
|LMW adiponectin (μg/mL)||1.5 ± 0.7||1.5 ± 0.6||0.92|
Total and HMW adiponectin were inversely associated with diabetes after multivariate adjustment for age, smoking status, physical activity, history of hypertension, family history of diabetes, alcohol use, and menopausal status (Table 4; Models 1 and 2). Odds ratios (ORs) comparing the highest with the lowest quartiles were 0.46 (95% confidence interval, 0.25–0.82; P for trend = 0.046) for total adiponectin, 0.40 (95% CI, 0.22–0.72; P for trend = 0.046) for HMW adiponectin, 1.04 (95% CI, 0.60–1.77; P for trend = 0.81) for MMW adiponectin, and 0.51 (95% CI, 0.29–0.89; P for trend = 0.01) for LMW adiponectin. After adjustment for BMI, the associations between diabetes and total and HMW adiponectin were attenuated (P for trend = 0.15 and 0.13, respectively), but the association between LMW adiponectin and diabetes remained (P for trend = 0.04; Model 3). This association between LMW adiponectin level and diabetes attenuated after adjustment for HOMA-IR (P for trend = 0.22; Model 4). MMW adiponectin levels showed no associations in all the Models (Models 1–4). The P-values for interaction by sex were 0.45 for total adiponectin, 0.50 for HMW, 0.27 for MMW, and 0.86 for LMW adiponectin levels, indicating insufficient evidence to support sex differences in the associations between adiponectin multimers and diabetes.
|Variable||Quartile of adiponectin levels||P for trend|
|1 (lowest)||2||3||4 (highest)|
The results of stratified analysis according to sex are shown in Tables 5 and 6. In men, LMW adiponectin, but not total, HMW, or MMW adiponectin, showed an inverse association with diabetes in Models 1–3. After multivariate adjustment for age, smoking status, physical activity, history of hypertension, family history of diabetes, and alcohol use, OR comparing the highest with the lowest quartiles was 0.52 (95% CI, 0.27–1.004; P for trend = 0.02) (Table 5; Model 2). The inverse association remained after adjustment for BMI (P for trend = 0.04; Model 3), and attenuated after adjustment for HOMA-IR (P for trend = 0.33; Model 4). In women, although the inverse association of HMW adiponectin with diabetes was suggested (OR = 0.41; 95% CI, 0.13–1.33; P for trend = 0.08) (Table 6; Model 2), there were no statistically significant associations between any forms of adiponectin levels and diabetes.
|Variable||Quartile of adiponectin levels||P for trend|
|1 (lowest)||2||3||4 (highest)|
|Variable||Quartile of adiponectin levels||P for trend|
|1 (lowest)||2||3||4 (highest)|
In this case-control study, we investigated the associations between adiponectin complexes and diabetes. Consistent with previous studies, total and HMW adiponectin levels were inversely associated with diabetes ([1, 7, 21]). Further, in this study, LMW adiponectin levels were found to be inversely associated with diabetes. The association between adiponectin complexes and diabetes was attenuated after adjustment for HOMA-IR, suggesting that insulin resistance may account for the associations.
HMW is thought to be the major active form of adiponectin in peripheral tissues ([22, 23]). LMW adiponectin may also play a role in glucose metabolism, and LMW adiponectin is predominant in the cerebrospinal fluid (CSF) ([5, 24]). A previous study among 154 octogenarians reported that lower LMW adiponectin levels were associated with diabetes (). Our results, coupled with earlier evidence, suggest that LMW adiponectin may also be inversely associated with diabetes.
In this study, the association between adiponectin levels and diabetes was attenuated after adjustment for BMI in the case of total and HMW adiponectin, but not LMW adiponectin. This observation is consistent with the findings of a previous study among 760 Japanese children, in which HMW but not LMW adiponectin levels were inversely associated with BMI (). This association between LMW adiponectin levels and diabetes attenuated after adjustment for HOMA-IR. This finding may indicate that the association between LMW adiponectin and diabetes observed in this study was secondary to insulin resistance or mediated by insulin resistance. Because adiponectin levels are reported to show sexual dimorphism, with women showing higher adiponectin levels ([26, 27]), we conducted a sensitivity analysis stratified by sex. The associations between adiponectin complexes and diabetes were all attenuated, with the exception for LMW adiponectin in men. Although we did not observe apparent sex differences in the adiponectin-diabetes association in the present study, we may have lacked statistical power (sample size) to detect sexual dimorphism.
Some limitations of the present study need to be addressed. First, participants in this study were Japanese individuals from one district, and the sample size was limited. Thus, generalizability may be a problem. Second, residual confounding by unknown or unmeasured confounding factors may have occurred. Finally, because a cross-sectional analysis was used, we were unable to establish temporal relationships. The major strengths of this study were the well-characterized population, matched case controls, and the fact that we measured the levels of each adiponectin complex.
In conclusion, our findings suggest that decreased LMW adiponectin levels, in addition to total or HMW adiponectin levels, may be associated with diabetes. To better understand the role of LMW adiponectin in diabetes or the sexual dimorphism of adiponectin in Asians, prospective studies with larger sample sizes and experimental studies are warranted.
The authors acknowledge all members for their participation in this research and also thank Masanobu Ikeda and Tomoko Yasuda for their assistance.