| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
BRIEF REPORT |
Steno Diabetes Center (H.S., P.P., A.A.V.), 2820 Gentofte, Denmark; and Department of Clinical Sciences/Diabetes and Endocrinology (C.L., L.G.), Lund University, Clinical Research Centre, University Hospital Malmö, S-205 02 Malmö, Sweden
Address all correspondence and requests for reprints to: Pernille Poulsen, Steno Diabetes Center, Niels Steensens Vej 2, 2820 Gentofte, Denmark. E-mail: pepn{at}steno.dk.
| Abstract |
|---|
|
|
|---|
Objectives: The aim of the study was to investigate the genetic and nongenetic control of plasma adiponectin and muscle AdipoR1/R2 gene expression and the impact of these components on in vivo glucose and fat metabolism.
Design and Participants: Plasma adiponectin and muscle gene expression of AdipoR1/R2 were measured before and during insulin infusion in 89 young and 69 elderly monozygotic and dizygotic twins. Insulin action, and glucose and fat oxidation rates were determined using hyperinsulinemic euglycemic clamps and indirect calorimetry.
Results: We demonstrated a genetic component in the control of plasma adiponectin and AdipoR1/R2 gene expression. Furthermore, levels of adiponectin and AdipoR1/R2 were influenced by age, sex, abdominal obesity, and aerobic capacity. Intrapair correlations in monozygotic twins indicated a nongenetic influence of birth weight on plasma adiponectin and AdipoR2 expression. Nonoxidative glucose metabolism was associated with AdipoR1 and plasma adiponectin, in young and elderly twins, respectively. In addition, plasma adiponectin was related to glucose and fat oxidation in younger subjects.
Conclusions: Plasma adiponectin and muscle gene expression of its specific receptors are controlled by genetic and several specific nongenetic factors. The data suggest that the "adiponectin axis" plays a role in in vivo insulin action and nonoxidative glucose metabolism.
| Introduction |
|---|
|
|
|---|
Genetic factors and several nongenetic factors, including age, physical inactivity, obesity, and low birth weight, contribute to the risk of T2D. The influence of these factors on adiponectin and its receptor levels remains to be determined. Twin studies are instrumental in quantifying the extent to which a phenotype is determined by genetic and nongenetic factors. Furthermore, measuring correlations between within-twin pair differences in monozygotic (MZ) twins allows associations of genetic and nongenetic origin to be distinguished.
This study aimed to investigate the relative influence of genetic and nongenetic factors [age, sex, birth weight, obesity, and aerobic capacity (VO2max)] on plasma adiponectin and the expression of AdipoR1/R2 in skeletal muscle before and during insulin stimulation, as well as the impact of this "adiponectin axis" on glucose and fat metabolism, measured by "gold standard" techniques in a cohort of young and elderly MZ and dizygotic (DZ) twins (11, 12).
| Subjects and Methods |
|---|
|
|
|---|
Young (n = 89) and elderly (n = 69) same-sex MZ and DZ healthy twins underwent a 2-h (40 mU/m2/min) hyperinsulinemic euglycemic clamp, including indirect calorimetry, as previously described (11, 12, 13). Muscle biopsies were obtained from the vastus lateralis muscle before and during insulin infusion as previously described (11). The present study was approved by the regional Ethical Committees and conducted in accordance with the Helsinki Declaration.
Analytical methods
Fasting levels of plasma adiponectin were determined by ELISA (Linco Research, St. Charles, MO). The intraassay coefficient of variation was 4.3% (n = 12). AdipoR1 and R2 mRNA levels were quantified using TaqMan real-time PCR with an ABI 7900 system (Applied Biosystems, Foster City, CA), and gene-specific probes and primer pairs for AdipoR1 (Assays-on-demand, Hs00360422_m1; Applied Biosystems) and AdipoR2 (Assays-on-demand, Hs00226105_m1; Applied Biosystems). The transcript quantities of AdipoR1 and R2 were normalized to the mRNA level of cyclophilin A (4326316E), and expressed as arbitrary units (AUs).
Statistical methods
Comparisons between young/elderly and male/female subjects were performed using the "proc mixed" (ANOVA) procedure in SAS (version 8.2; SAS Institute Inc., Cary, NC), with adjustment for intratwin pair relationships (11). Data are presented as mean (SD). Absolute phenotypic correlations (Spearman) were calculated using SigmaStat 3.0 (Systat Software, Inc., San Jose, CA).
Intratwin-pair correlations between differences in phenotypic variables within MZ and DZ twin pairs (i.e. twin A - twin B) allow adjustment for the influence of common environmental and maternal factors, and, in particular, for genotype in MZ twins. Accordingly, a significant intratwin-pair correlation within MZ twins (regardless of the correlation in DZ twins) is of nongenetic origin, while a significant correlation only among DZ twins suggests a genetic origin. The correlation analyses were performed using SigmaStat 3.0.
The total phenotypic variance is the sum of the variance attributable to effects of both genetic and environmental factors. The heritability (h2) expresses the proportion of variance attributable to genetic variance, and was assessed by biometric modeling as previously described (13).
| Results |
|---|
|
|
|---|
The clinical characteristics of the subjects have been described previously (11, 12, 13). In brief, the elderly [age 62.4 yr (0.2)] twins had greater total fat mass, higher trunk-leg fat ratios, and a higher level of fat oxidation (FOX), but lower VO2max, insulin-stimulated whole-body glucose uptake (glucose disposal rate), glucose oxidation (GOX), and nonoxidative glucose metabolism (NOGM) than the younger [age 28.0 yr (0.2)] twins.
The relationship between plasma adiponectin and muscle AdipoR1/R2 gene expression levels, and effect of age and insulin
AdipoR1 gene expression was similar in young and elderly twins. The expression of basal (P < 0.001) and clamp (P < 0.001) AdipoR2 was lower in elderly than in young twins. Insulin increased the expression of AdipoR2 significantly in both elderly and young twins, whereas AdipoR1 expression was not affected by insulin. Fasting plasma adiponectin values were significantly higher in elderly than in young twins (P < 0.001) (Table 1
).
|
|
Plasma adiponectin levels were significantly higher in females than males [mean (SD), mg/liter; males: 6.9 (3.8); females: 10.0 (4.1); P < 0.001], regardless of age. In contrast, men had higher AdipoR1 and AdipoR2 mRNA levels than women during both steady-state periods [mean (SD), AU; AdipoR1basal: 0.64 (0.23) vs. 0.50 (0.25), P < 0.001; AdipoR1insulin: 0.67 (0.25) vs. 0.46 (0.13), P < 0.001; AdipoR2basal: 0.44 (0.14) vs. 0.37 (0.09), P < 0.001; and AdipoR2insulin: 0.48 (0.15) vs. 0.39 (0.10), P < 0.001].
AdipoR1 mRNA level was positively associated with VO2max (basal: r = 0.41, P < 0.001; insulin: r = 0.30, P = 0.005) and trunk-leg fat ratio (basal: r = 0.40, P < 0.001; insulin: r = 0.34, P = 0.001) and negatively associated with total fat percentage (basal: r = 0.52, P < 0.001; insulin: r = 0.39, P < 0.001) in young subjects. In the elderly, AdipoR1 mRNA level was weakly associated with trunk-leg fat ratio (insulin: r = 0.27, P = 0.02).
AdipoR2 mRNA level was negatively associated with total fat percentage (basal: r = 0.31, P = 0.003; insulin: r = 0.31, P = 0.004) and positively associated with VO2max (basal: r = 0.21, P = 0.05) and trunk-leg fat ratio (r = 0.27, P = 0.01). In the elderly, AdipoR2 mRNA level was not associated with measures of VO2max or body composition.
Plasma adiponectin was negatively correlated with trunk-leg fat ratio in both age groups (young: r = 0.54, P < 0.0.001; elderly: r = 0.48, P < 0.0001) and positively associated with total fat percentage in the young (r = 0.38; P < 0.001).
h2 of plasma adiponectin and muscle AdipoR1/R2 expression levels
The h2 estimates suggest that there is a major genetic component of plasma adiponectin and AdipoR1/R2 gene expression in both young (adiponectin: h2 = 76%, AdipoR1; basal: h2 = 66%, insulin: h2 = 75%, AdipoR2; and basal: h2 = 73%, insulin: h2 = 64%) and elderly (adiponectin: h2 = 58%, AdipoR1; basal: h2 = 78%, insulin: h2 = 74%, AdipoR2; and basal: h2 = 77%, insulin: h2 = 39%) twins.
Impact of birth weight on plasma adiponectin and muscle AdipoR1/R2 gene expression levels
When performing intrapair correlations, eliminating a genetic influence in MZ twins, we demonstrated a positive correlation between plasma adiponectin and birth weight in young MZ twins (r = 0.37; P = 0.007), but not in DZ or elderly twins. We found a positive intrapair correlation between clamp AdipoR1 gene expression and birth weight in DZ (r = 0.41; P = 0.012), but not in MZ (r = 0.17; nonsignificant) elderly twins. Furthermore, in elderly twins there was a significant intrapair correlation between clamp AdipoR2 gene expression and birth weight (MZ: r = 0.36, P = 0.037; DZ: r = 0.35, P = 0.031).
Impact of adiponectin and its receptors on glucose and fat metabolism
Plasma adiponectin was associated with FOX (r = 0.23; P = 0.03) and GOX (r = 0.37; P < 0.001) in young twins, and positively associated with glucose disposal rate (r = 0.67; P < 0.001) and NOGM (r = 0.72; P < 0.001) in elderly twins. AdipoR1 gene expression was positively associated with NOGM (r = 0.24; P = 0.02) only in young twins. No associations were demonstrated between AdipoR2 gene expression and metabolic variables in either age group.
| Discussion |
|---|
|
|
|---|
Interestingly, we found an influence of age on the levels and relationship of adiponectin with its receptors. In younger subjects, adiponectin was inversely correlated with levels of AdipoR1/R2, consistent with findings from middle-aged subjects (7). However, no such association was seen in the elderly twins. Higher plasma adiponectin levels and lower AdipoR2 mRNA expression were observed in elderly than in younger twins. The higher adiponectin level, which is a key determinant of glucose metabolism in elderly subjects, may represent an effect that compensates for the reduction of receptor levels and/or "adiponectin resistance" with increasing age.
In the young twins, AdipoR1 expression was associated with NOGM, and adiponectin was associated negatively with FOX and positively with GOX. In elderly twins, glucose disposal and NOGM were associated with circulating adiponectin, which implies a role for adiponectin in the control of muscle glycogen synthesis in elderly subjects (19). However, receptor expression levels were not associated with any metabolic variables. Thus, the relative influence of adiponectin and its specific receptors on in vivo insulin action seems to change with age. The lack of association between plasma adiponectin and glucose metabolism in the young is somewhat surprising. One explanation for this may be that young subjects are lean and insulin sensitive, and the association between adiponectin and insulin resistance may very well develop later in the course of becoming increasingly obese and insulin resistant (20). It could also be explained by an age-related change in the ratio of high-molecular weight adiponectin to total adiponectin, which has recently been suggested to be a better indicator of metabolic activity (21).
The role of AdipoR1/R2 in the regulation of glucose metabolism is controversial, with reports of positive and negative associations (6, 7, 9). These apparently paradoxical results may be a consequence of using different methods to measure insulin action, the size of the study populations, glucose tolerance, and, perhaps, as the results of this study suggest in particular, age. Importantly, in this large and thoroughly characterized study population, where in vivo insulin action was measured by the "gold standard" hyperinsulinemic clamp technique, we found indication of a specific role for the muscle AdipoR1 in the control of insulin action and NOGM in young but not elderly subjects. The age dependency of the influence of AdipoR1 expression level and, in particular, the absence of any distinct effect of muscle AdipoR2 expression level on in vivo insulin action requires further investigation. Regardless, the present results suggest that these two receptors could mediate differential cellular responses to adiponectin within and between tissues.
Plasma adiponectin was negatively associated with abdominal obesity, which is consistent with previous findings (7). Expression levels of AdipoR1 and AdipoR2 were, on the other hand, positively associated with trunk-leg ratio. A positive effect of acute and chronic training on AdipoR1 and AdipoR2 expression in skeletal muscle (7) has recently been reported, and is consistent with the positive association between VO2max and AdipoR1 and AdipoR2 in the present study. The higher plasma adiponectin level in females than in males is consistent with previous findings in young (22) and elderly (23) subjects, whereas a concomitant reduction of the receptor expression level has not been demonstrated before, to our knowledge.
Recent studies suggest that reduced levels of adiponectin may play an important role in insulin resistance associated with reduced fetal growth (24). Performing intrapair correlations, and eliminating any genetic influence in MZ twins, we demonstrated positive associations between birth weight and plasma adiponectin in young, and between birth weight and AdipoR2 expression in elderly MZ twins, indicating nongenetic associations. We recently showed that low birth weight is nongenetically associated with insulin resistance in elderly but not in young MZ twins (12). The association between plasma adiponectin and birth weight in young, but not in elderly twins, suggests that reduced plasma adiponectin levels may, to some extent, precede overt insulin resistance in elderly twins, but the "adiponectin axis" per se does not fully explain the nongenetic impact of birth weight on insulin action in elderly twins.
Although many statistical analyses were performed in this study, all were based upon a priori hypotheses, and correction for multiple testing was, therefore, not performed in the primary analysis. Regardless, it may be important to realize that the major findings in this study, including the interrelationship between circulating adiponectin and the expression of its receptors, the effect of age and gender on AdipoR1/R2 expression and adiponectin, and the association between adiponectin and glucose and FOX in the young, and adiponectin and nonoxidative glucose metabolism in the elderly, all remained statistically significant even after applying Bonferroni correction for multiple testing.
In conclusion, we have found evidence for a strong genetic component in the control of plasma adiponectin and muscle AdipoR1/R2 gene expression levels in twins. In addition, nongenetic factors, including age, sex, birth weight, abdominal obesity, and VO2max, influence the adiponectin-AdipoR1/R2 axis. Levels of AdipoR1, but not AdipoR2, gene expression in skeletal muscle and plasma adiponectin are related to nonoxidative glucose metabolism in young and elderly subjects, respectively.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: H.S., P.P., and C.L. have nothing to declare. L.G. has been a consultant for and served on advisory boards for Aventis-Sanofi, Bristol Myers Squibb, GlaxoSmithKline and Roche. A.A.V. has been a consultant for and served on advisory boards for Novo Nordisk.
First Published Online April 10, 2007
1 H.S. and P.P. contributed equally to this work. ![]()
Abbreviations: AdipoR, Adiponectin receptor; AU, arbitrary unit; DZ, dizygotic; FOX, fat oxidation; GOX, glucose oxidation; h2, heritability; MZ, monozygotic; NOGM, nonoxidative glucose metabolism; T2D, type 2 diabetes; VO2max, aerobic capacity.
Received August 17, 2006.
Accepted April 3, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |