© 2003 by the American Diabetes Association, Inc.
Reduced Skeletal Muscle Oxygen Uptake and Reduced ß-Cell FunctionTwo early abnormalities in normal glucose-tolerant offspring of patients with type 2 diabetes
1 Department of Endocrinology and Metabolism, Eberhard-Karls-University, Tübingen, Germany Address correspondence and reprint requests to PD Dr. med. Andreas Fritsche, Medizinische Universitätsklinik, Otfried-Müller-Str. 10, D-72076 Tübingen, Germany. E-mail: andreas.fritsche{at}med.uni-tuebingen.de.
OBJECTIVEStudies on insulin sensitivity and insulin secretion in subjects with a familial predisposition for type 2 diabetes mellitus (T2DM) traditionally produce inconsistent results. This may be due to small sample size, subject selection, matching procedures, and perhaps lack of a measure of physical fitness. RESEARCH DESIGN AND METHODSIn the present study, we specifically tested the hypothesis that a family history of T2DM is associated with reduced VO2max, measured by incremental bicycle ergometry, independent of insulin sensitivity estimated from an oral glucose tolerance test (OGTT; n = 424) and measured by a euglycemic-hyperinsulinemic clamp (n = 185). Subjects included in the study were young (34 ± 10 years), healthy, and normal glucose tolerant with either a first-degree relative (FDR) with T2DM (n = 183), a second-degree relative with T2DM (n = 94), or no family history of T2DM (control subjects, n = 147). BMI, percent body fat, waist-to-hip ratio (WHR), and habitual physical activity (HPA; standard questionnaire) were comparable among groups. FDRs had significantly lower VO2max than control subjects: 40.5 ± 0.6 vs. 45.2 ± 0.9 ml O2/kg lean body mass, P = 0.01 after adjusting for sex, age, BMI, HPA, and insulin sensitivity (euglycemic-hyperinsulinemic clamp). RESULTSBMI, percent body fat, waist-to-hip ratio (WHR), and habitual physical activity (HPA; standard questionnaire) were comparable among groups. FDRs had significantly lower VO2max than control subjects: 40.5 ± 0.6 vs. 45.2 ± 0.9 ml O2/kg lean body mass, P = 0.01 after adjusting for sex, age, BMI, HPA, and insulin sensitivity (euglycemic-hyperinsulinemic clamp). Insulin sensitivity per se was not affected by family history of T2DM after adjusting for age, sex, BMI, and percent body fat (P = 0.76). The appropriateness of ß-cell function for the individual insulin sensitivity (disposition index: product of a validated secretion parameter [OGTT] and sensitivity [clamp]) was significantly lower in FDRs (87 ± 4 units) versus control subjects (104 ± 6 units, P = 0.02 after adjusting for sex, age, and BMI). Analyses of the larger OGTT group produced essentially the same results. CONCLUSIONSIn conclusion, these data are compatible with the hypothesis that familial predisposition for T2DM impairs maximal oxygen consumption in skeletal muscle. Because habitual physical activity was not different, genetic factors may be involved. Conceivably, reduced VO2max precedes skeletal muscle insulin resistance, providing a partial explanation for discrepancies in the literature.
Abbreviations: FDR, first-degree relative FHD, family history of diabetes GIR, glucose infusion rate HPA, habitual physical activity ISI, insulin sensitivity index OGTT, oral glucose tolerance test T2DM, type 2 diabetes mellitus WHR, waist-to-hip ratio
Familial predisposition is an important risk factor for type 2 diabetes mellitus (T2DM). Studies performed to identify the underlying mechanisms have produced conflicting results. Whereas some primarily demonstrated impaired insulin secretion in subjects with a family history of T2DM, others reported reduced insulin sensitivity as the main finding. To date, no consensus has been reached (1,2). Issues such as differences in experimental techniques, recruitment mechanisms (inclusion and exclusion criteria), matching procedures, adjustments for covariates, and personal bias of investigators are probably involved. In addition, few of the available studies have controlled for physical fitness, which is a strong independent determinant of insulin sensitivity (3). Conceivably, familial clustering of insulin resistance may be the corollary of a sedentary lifestyle prevailing in a particular family and have nothing to do with a primary (i.e., genetic) defect in insulin-stimulated glucose disposal. The existence of familial aggregation of behavioral traits such as television viewing, depression, or reduced physical activity is well documented (411). On the other hand, it is possible that genetic factors primarily affect the ability of skeletal muscle to maximize its oxidative capacity in response to endurance training. In this scenario, reduced glucose disposal would be a consequence of impaired skeletal muscle substrate oxidation rather than impaired insulin signaling. Consistent with this hypothesis, in insulin resistant offspring of patients with T2DM, maximal oxygen consumption was reduced compared with insulin sensitive control subjects without family history of diabetes (FHD) (12). Moreover, muscle of offspring of patients with T2DM was found to be enriched with type II fibers (13), whereas maximal oxygen consumption is predominantly a function of the type 1 fiber content (14,15). Therefore, the aim of the present study was to test the hypothesis that family history of T2DM is associated with reduced maximal oxidative capacity independent of alterations in insulin sensitivity. For this purpose, we studied a large group of healthy, normal glucose tolerant subjects with family history of T2DM (n = 279) and without family history of T2DM (n = 147). To provide a comprehensive picture of metabolic traits controlling glycemia, we included validated estimates for insulin secretion and calculated a disposition index. The size of the study group permitted mathematical adjustments for relevant covariates using general estimating equation procedures.
Subjects were recruited from the ongoing Tübingen Family Study for Type 2 Diabetes (TÜF). The Tübingen Family Study recruits family members of subjects with type 2 diabetes. Over the years, subjects without any family history were asked to undergo metabolic checkup. This group was used as a control group in the present study. A total of 424 healthy subjects with normal glucose tolerance according to World Health Organization criteria were selected for analyses. In a total of 185 subjects (77 control subjects and 75 first-degree relatives [FDRs]), insulin sensitivity data from euglycemic-hyperinsulinemic clamp were available for analyses. After the nature of the study was explained, all subjects gave informed written consent and underwent the standard procedures of the protocol, including medical history, physical examination, routine blood test, and oral glucose tolerance test (OGTT). The local ethics committee approved all protocols.
Analytical procedures and measurements
Body composition
Family history status
Habitual physical activity
Maximal oxygen consumption
Euglycemic-hyperinsulinemic clamp
Calculations
First phase insulin secretion.
Disposition index.
Statistical analyses
Subject characteristics (all subjects) FDRs of type 2 diabetic patients were significantly older than the control group without FHD or second-degree relatives. The groups were comparable with respect to BMI, percent body fat, and WHR. Fasting and 2-h blood glucose levels during OGTT were significantly higher in the FDR group compared with the control group (Table 1). This effect was still significant after adjusting for age and percent body fat (P = 0.0045 for fasting and P = 0.0047 for 2-h glucose). Characteristics of the subjects who also underwent a euglycemic-hyperinsulinemic clamp were similar to those of the OGTT group.
Relationships among maximal oxygen consumption, physical activity, and insulin sensitivity Maximal oxygen consumption during incremental exercise on a bicycle ergometer was higher in men than in women (48.5 ± 0.9 vs. 39.3 ± 0.4 ml · kg lean body mass-1 · min-1, P < 0.001). Habitual physical activity was positively correlated to maximal oxygen consumption (r = 0.24, P < 0.001) and insulin sensitivity (clamp) (r = 0.23, P < 0.001). Furthermore, maximal oxygen consumption was correlated to insulin sensitivity (clamp) (r = 0.42, P < 0.001). Age (r = -0.29, P < 0.001), body fat content (r = -0.41, P < 0.001) and 2-h glucose (r = -0.18, P < 0.0001) were negatively correlated with maximal oxygen consumption (Fig. 1).
Family history of T2DM and maximal oxygen consumption Maximal oxygen consumption was highest in the control group without family history of T2DM (45.2 ± 0.9 ml · kg lean body mass-1 · min-1) and significantly lower in second- and first-degree relatives of type 2 diabetic patients (42.7 ± 0.9 and 40.5 ± 0.6, P < 0.001, ANOVA). In contrast, habitual physical activity was comparable in all groups (8.2 ± 0.1 in control subjects versus 8.3 ± 0.1 in second-degree relatives versus 8.1 ± 0.1 in first-degree relatives, P = 0.27, ANOVA) (Fig. 2). In a multivariate regression analysis with maximal oxygen consumption as a dependent variable, the effect of family history status remained significant (P = 0.04) after adjusting for the effects of sex (P < 0.001), age (P = 0.11), BMI (P = 0.06), HPA (P < 0.001), and insulin sensitivity (clamp) (P = 0.01) (see Table 2). The results were similar using an insulin sensitivity estimate from the OGTT (P = 0.05 for the effect of family history status).
Family history status and its relationship with insulin sensitivity (euglycemic-hyperinsulinemic clamp) and insulin secretion Insulin sensitivity (clamp) tended to be lower in FDRs compared with the control group (0.119 ± 0.107 vs. 0.100 ± 0.087 µmol · kg-1 · min-1 · pmol-1, P = 0.051). This marginal effect of family history on insulin sensitivity disappeared completely after adjusting for sex, age, BMI, and HPA in a multivariate analysis (P = 0.87) (see Table 2). Similarly, family history status had no significant effect on insulin sensitivity estimated in the larger OGTT group after adjusting for the same covariates (P = 0.14). Moreover, no differences in first-phase insulin secretion between the control group and FDRs were observed (1,051 ± 43 vs. 992 ± 35, P = 0.28 and P = 0.30 after adjusting for sex, age, and BMI). The calculated disposition index (ISIclamp x SecretionOGTT), however, was significantly higher in control subjects compared with FDRs (104 ± 6 vs. 87 ± 4 units, P = 0.02), indicating reduced insulin secretion relative to insulin sensitivity in FDR subjects. This relationship remained significant (P = 0.046) after adjusting for the effects of sex, age, and BMI (see Table 2). The disposition index purely derived from OGTT was also significantly higher in control subjects than in FDRs (22,468 ± 883 vs. 18,291 ± 874, P < 0.001). This relationship remained significant (P = 0.02) after adjusting for sex, age, and BMI (Fig. 3).
The aim of the present study was to identify early metabolically relevant abnormalities predisposing to type 2 diabetes in FDRs of patients with the disease. There was a trend for lower insulin sensitivity in FDRs who also tended to be somewhat more obese and older. After adjusting, however, insulin sensitivity became comparable between the groups, indicating that most of the original trend was secondary to differences in fatness. Insulin secretion assessed by a validated index based on fasting and 30-min insulin and glucose concentrations during an OGTT was also comparable among the patients with FHD. However, insulin secretion relative to insulin sensitivity (expressed as the so-called disposition index) was significantly lower in FDR subjects. This is in line with results of previous studies identifying reduced ß-cell compensation to insulin resistance in individuals with FHD (2124).
An important finding of the present study is that maximal oxygen consumption was reduced in FDRs despite essentially identical physical activity scores (derived from a questionnaire). The fact that this difference was observed in the absence of any detectable difference in insulin sensitivity is the most interesting aspect but represents somewhat of a pathophysiological quandary. That is, how would an abnormality in oxidative muscle metabolism during exercise ultimately affect glucose metabolism if not via peripheral (i.e., muscle in the insulin-stimulated state) glucose uptake? There are two possible conjectures: first, our measurement of peripheral insulin sensitivity is too unphysiologic (steady-state insulin concentration during the clamp It is important to emphasize that although the study population had normal glucose tolerance, the abnormal subphenotypes somehow impacted fasting and 2-h glucose levels, which were significantly higher in FDRs. The lower disposition is probably the most relevant factor responsible for the differences in glycemia. Nevertheless, mechanism secondary to maximal oxygen consumption but outside the diagnostic scope of the euglycemic-hyperinsulinemic clamp may also be involved. Previously, reduced maximal oxygen uptake was demonstrated in insulin-resistant relatives of patients with T2DM (12). In our study, maximal oxygen uptake was a strong determinant of insulin sensitivity but, in contrast to oxygen uptake, insulin sensitivity was not different between the patients with FHD. These findings are difficult to reconcile, but the observation that in patients with overt type 2 diabetes, resistance training improves insulin sensitivity without altering VO2max (25) in principle supports the possibility of a dissociation between effects of exercise and insulin sensitivity on skeletal muscle. Conceivably, reduced maximal oxygen consumption of skeletal muscle represents a primary defect predisposing skeletal muscle to insulin resistance when other confounding factors (e.g., physical inactivity) come into play. The relative content of type 1 fibers in skeletal muscle was identified as a main determinant of skeletal muscle oxygen uptake during exercise (14,15). Type 1 muscle fibers are slow-twitch muscle fibers with preferentially oxidative metabolism rich in myoglobin, mitochondria, oxidative enzymes, and lipids (26). Therefore, the mechanism underlying our observation is likely to involve altered fiber composition of skeletal muscle, specifically a reduced number of type 1 fibers, which was found to be associated with adiposity and insulin resistance (27). Additionally, a high proportion of type 1 muscle fibers has been shown to protect against weight gain in response to a long-term positive energy balance (28). Moreover, fiber type proportions of skeletal muscle have been demonstrated to be genetically determined (29). In accordance with all these findings, the number of type 2 muscle fibers were found to be reduced in FDRs of patients with T2DM (13). It is also necessary to point out that use of questionnaires in assessing habitual physical activity cannot be without problems. The same quantity or quality of physical exercise may seem more or less strenuous to different individuals. However, it is unlikely that the degree of underreporting is influenced by family history status. Therefore, and in view of the large number of subjects reported herein, our overall conclusions should remain valid. Finally, this study is not merely another report of prediabetic subphenotypes in subjects with and without FHD. Instead, this study presents novel aspects that complement previously published work. First, by including every subject in the database and performing ANCOVA rather than matching and excluding "un-matchable" individuals, we largely avoided any selection bias. Second, we present VO2max data in addition to insulin sensitivity and demonstrate that there may be dissociation between skeletal muscle glucose and exercise metabolism, which implies the possibility that only in subgroups does skeletal muscle have a primary role in the pathogenesis of T2DM. Third, we found abnormalities in both ß-cells and insulin sensitivity tissue but only conditionally, i.e., using the disposition index or measuring skeletal muscle oxidative capacity. This clearly demonstrates the subtlety of the influence of FHD on diabetes-relevant traits and may help to reconcile conflicting data and some ongoing debates. In summary, we have demonstrated two metabolic abnormalities in a young and healthy group of subjects with a familial predisposition of type 2 diabetes: reduced insulin secretion relative to insulin sensitivity and reduced maximal oxygen consumption during exercise, which was not the result of lack of exercise.
This study was supported by a research grant from the Deutsche Forschungsgemeinschaft (Fr 1561/11) and the European Community (QLRT-199900674). M.S. is currently supported by a Heisenberg grant of the Deutsche Forschungsgemeinschaft. We thank all the research volunteers for their participation and Elke Maerker, Anna Teigeler, and Heike Luz for their excellent technical assistance.
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances. Received for publication February 13, 2003. Accepted for publication March 21, 2003.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||