Introduction

Type 2 diabetes is a heterogeneous disease characterised by insulin resistance and impaired insulin secretion. Abdominal obesity is a major risk factor for insulin resistance and type 2 diabetes [1]. Genetic variation in the calpain 10 gene (CAPN10) is associated with type 2 diabetes [2]. Calpain-10 is a calcium-activated protein expressed in all human tissues [3], but the biochemical and physiological mechanism(s) by which it affects the risk of type 2 diabetes remain unclear. The G allele of the single nucleotide polymorphism (SNP)-43, has been associated with insulin resistance [4] and type 2 diabetes in both cross-sectional [2] and prospective studies [5]. However, other studies have not found an association of the G allele with type 2 diabetes or insulin resistance [6–8], and other SNPs and/or haplotypes in the same locus have been proposed to be more important for glucose metabolism [6, 9–11]. The mechanisms by which the G allele of SNP-43 leads to lower rates of glucose disposal and higher rates of lipid oxidation [4] are unknown, but have been proposed to be related to high levels of NEFA [12]. Alternatively, polymorphisms in this locus could regulate energy expenditure [4], thermogenesis [13] and insulin secretion [14].

In this study we investigated whether SNP-43, SNP-44, insertion/deletion (InDel)-19 and SNP-63 polymorphisms of CAPN10 regulate insulin action, insulin secretion, energy expenditure, substrate oxidation and abdominal obesity in 158 middle-aged Finnish non-diabetic offspring of patients with type 2 diabetes.

Subjects and methods

Subjects

Study I

All subjects participating in Study I were of Finnish ancestry. Genetically the Finnish population is relatively homogeneous, originating mainly from southern (European) and eastern (Asian) immigration about 2,000 years ago [15]. The study group consisted of 158 non-diabetic offspring of patients with type 2 diabetes from our ongoing study [16]. Patients with type 2 diabetes were randomly selected from patients living in the Kuopio University Hospital region (population of 250,000). One to four offspring from each family were included and altogether 102 families were studied (57 families with one, 35 families with two, nine families with three, and one family with four offspring). On day 1, all subjects underwent an OGTT to evaluate their glucose tolerance according to the World Health Organization criteria [17]. All subjects had normal liver, kidney and thyroid function, no history of excessive alcohol intake and no severe chronic diseases. Informed consent was obtained from all subjects after the purpose and potential risks of the study were explained to them. The protocol was approved by the Ethics Committee of the University of Kuopio and was in accordance with the Helsinki Declaration.

Study II

An independent study population was collected to confirm the findings of Study I. The primary aim of this study is to investigate the effects of SNPs in genes of interest with respect to the risk of type 2 diabetes and coronary heart disease in a random sample of Finnish men, aged from 50 to 70 years, living in Kuopio town (population of 90,000), eastern Finland. The first 234 men of this ongoing population-based study to fulfil the following criteria: a history of type 2 diabetes in first-degree relatives of the participants, BMI <30 kg/m2 to match BMI of Study I, and a non-diabetic OGTT according to the World Health Organization criteria [17], were included in Study II. The protocol includes a 1-day visit to the Clinical Research Unit of the University of Kuopio. Study II was approved by the Ethics Committee of the University of Kuopio and was in accordance with the Helsinki Declaration.

Methods

Clinical measurements

Waist circumference was measured in the middle-point between the lower ribs and iliac crest to the nearest 0.5 cm. Blood pressure was measured in the sitting position after a 10-min rest. The mean of two measurements was used in statistical analyses.

Metabolic studies

(Study I) On day 2, metabolic studies were performed after an overnight fast. First, an IVGTT was performed to determine the first-phase insulin secretion capacity [18] as previously described [16]. Immediately after an IVGTT, the degree of insulin resistance was evaluated by the euglycaemic clamp technique [19]. After the baseline blood drawing, a priming dose of insulin (Actrapid 100 IU/ml; Novo Nordisk, Gentofte, Denmark) was administered during the initial 10 min to raise plasma insulin concentration quickly to the desired level, where it was maintained by a continuous insulin infusion of 240 pmol·kg−1·min−1. Blood glucose was clamped at 5.0 mmol/l for the next 120 min by the infusion of 20% glucose at varying rates according to blood glucose measurements performed at 5-min intervals. The mean rates of glucose infusion during the last hour of the clamp were used to calculate the rates of insulin-stimulated whole-body glucose uptake (WBGU). Indirect calorimetry was performed with a computerised flow-through canopy gas analyser system (Deltatrac; Datex, Helsinki, Finland) as previously described [20]. Gas exchange and urinary nitrogen excretion were measured in the fasting state and during the last 30 min of the euglycaemic clamp. The first 10 min of each measurement was discarded and the mean value of the last 20 min was used in calculations. The rates of glucose and lipid oxidation and energy expenditure were calculated according to Ferrannini [21]. The rates of non-oxidative glucose disposal during the euglycaemic clamp were estimated by subtracting the rates of glucose oxidation from the rates of WBGU.

Fat distribution

(Study I) Abdominal fat distribution was evaluated by computed tomography (Somatom Plus S; Siemens, Erlangen, Germany) at the level of the fourth lumbar vertebra. One hundred and twenty-nine subjects participated in the computed tomography study. Subcutaneous and intra-abdominal fat areas were calculated as previously described [22].

Analytical methods

Plasma glucose levels in the fasting state and in the OGTT (Studies I and II) and IVGTT and blood glucose levels during the euglycaemic clamp (Study I) were measured by the glucose oxidase method (2300 Stat Plus; Yellow Springs Instruments, Yellow Springs, OH, USA). For the determination of plasma insulin, blood was collected in EDTA-containing tubes, and after centrifugation the plasma was stored at −20°C until the analysis. Plasma insulin concentration was determined by a commercial double-antibody solid-phase RIA (Phadeseph Insulin RIA 100; Pharmacia Diagnostics AB, Uppsala, Sweden). An insulinogenic index was calculated as a ratio between changes in insulin and glucose during the first 30 min of the OGTT. Lipoprotein fractionation was performed by ultracentrifugation and selective precipitation, as previously described [23]. Cholesterol and triglyceride levels from whole serum and lipoprotein fractions were assayed by automated enzymatic methods (Boehringer-Mannheim, Mannheim, Germany). Serum NEFA were determined from fresh frozen samples by an enzymatic method (Wako Chemicals GmbH, Neuss, Germany). Non-protein urinary nitrogen was measured by an automated Kjeldahl method.

Genotyping of the SNPs in

CAPN10 Genotyping of SNP-43, SNP-44 and SNP-63 of CAPN10 was carried out using TaqMan Allelic Discrimination Assays (Applied Biosystems, Foster City, CA, USA). Primers are available upon request (J. Pihlajamäki). The TaqMan genotyping reaction was amplified on a GeneAmp PCR system 2700 and fluorescence was detected using an ABI Prism 7000 sequence detector (Applied Biosystems). The insertion/deletion polymorphism InDel-19 of CAPN10 was amplified using published PCR primers [6], and PCR products were separated on a 3% NuSieve agarose gel. We used the following coding for the alleles of CAPN10; SNP-43 (rs3792267): G=allele 1, A=allele 2; SNP-44 (rs2975760): T=allele 1, C=allele 2; InDel-19 (rs3842570): two copies of a 32 bp repeat=allele 1, three copies of a 32 bp repeat=allele 2; and SNP-63 (rs5030952): C=allele 1, T=allele 2. Only four SNP-43, InDel-19 and SNP-63 haplotypes were found (111, 121, 112 and 221). In this study the C allele of the SNP-44 was observed only with haplotype 111 (Table 1). We re-genotyped 10% of samples. So far, reproducibility of the results in this and other studies has been 100%.

Table 1 The effect of the CAPN10 haplotypes on glucose tolerance and hyperinsulinaemia in an OGTT, fasting serum triglycerides, the rates of WBGU and intra-abdominal fat area (IAF) in 158 offspring of patients with type 2 diabetes (QTDT-adjusted for age, sex and BMI)
Full size table

Statistical analyses

Analysis of the data was performed with the SPSS/Win programs (version 10.0; SPSS, Chicago, IL, USA). Insulin, NEFA and triglyceride concentrations and insulinogenic index were log transformed before statistical analyses to achieve a normal distribution. The effect of genotypes in families was studied with a mixed linear model including age, sex, BMI and family relationship as covariates when appropriate (Study I). In Study II, analysis of covariance (ANCOVA) was applied to adjust for confounding variables. For the haplotype analysis of CAPN10, haplotype frequencies were estimated and likely haplotypes reconstructed for each individual using the MERLIN program [24]. The effect of each haplotype on quantitative parameters was analysed with the family-based test of linkage disequilibrium (LD) using the QTDT program with age, sex and BMI as covariates, when appropriate [25]. A value of p<0.05 was considered as statistically significant. Data are presented as means±SD.

Results

Subjects in Study I were middle-aged (age 34.9±6.3 years [mean±SD]), their BMI was slightly above normal range (26.2±4.9 kg/m2) and they were non-diabetic on the basis of the OGTT (130 had normoglycaemia, three had isolated impaired fasting glucose and 25 had impaired glucose tolerance). The genotypes of SNP-44, SNP-43, InDel-19 and SNP-63 of CAPN10 were in Hardy–Weinberg equilibrium. SNP-43, InDel-19 and SNP-63 were in LD with each other (all pairwise D′>0.92 and p<0.006). In addition, SNP-44 was in LD with SNP-43 and InDel-19 (D′>0.80, p<0.05). Five haplotypes could be formed from the four polymorphisms (Table 1). No association of the genotypes with abnormal glucose tolerance was observed (p=0.712, impaired fasting glucose and impaired glucose tolerance combined).

In QTDT analysis only the 1221 haplotype was significantly associated with measures of glucose metabolism and intra-abdominal fat area (Table 1). Subjects with this haplotype had higher levels of 2-h insulin in the OGTT (p=0.023, adjusted for age, sex and BMI), higher levels of fasting triglycerides (p=0.016), lower rates of WBGU (p=0.010) and higher amount of intra-abdominal fat (p=0.004) than subjects without the haplotype. No effect of this haplotype on BMI, total or HDL-cholesterol, serum NEFA, insulin secretion (measured as insulin AUC during the first 10 min of the IVGTT or as an insulinogenic index) or on the rates of energy expenditure and substrate oxidation was observed (data not shown). Combined haplogenotypes of SNP-43, InDel-19 and SNP-63 had no statistically significant effect on insulin sensitivity or intra-abdominal fat area (data not shown).

The effect of the 1221 haplotype was due to the effect of SNP-43 only. SNP-43 was associated with larger area of intra-abdominal fat (p=0.009, adjusted for age, sex, BMI and family relationship) and tended to be associated with higher 2-h insulin in the OGTT (p=0.059), higher fasting triglycerides (p=0.071) and lower rates of WBGU per lean body mass (p=0.062, adjusted for age, sex and family relationship, Table 2). The association of the A allele of SNP-43 with lower rates of WBGU was not significant if adjusted for intra-abdominal fat area (p=0.449). However, the association with high intra-abdominal fat area (p=0.015), and high ratio of intra-abdominal fat to total abdominal fat area (p=0.005) remained significant after the adjustment for the rates of WBGU.

Table 2 Characteristics of the study group according to the genotypes of SNP-43
Full size table

We found a statistically significant interaction between sex and SNP-43 (p=0.037, adjusted for age, BMI and the rates of WBGU) on their effects on intra-abdominal fat area. The A allele of SNP-43 was associated with a large intra-abdominal fat area in men (p=0.014), but not in women (p=0.396) (Fig. 1). No interaction of SNP-43 with sex or their effects on the rates of WBGU or other parameters was observed. SNP-44 tended to be associated with the rates of WBGU (p=0.053) but this was probably because of LD with SNP-43. InDel-19 and SNP-63 were not associated with any of the parameters measured. We also found a statistically significant interaction between BMI (formed by dividing study population using median BMI of 25.3 kg/m2 as a cut-off point) and SNP-43 on their effects on intra-abdominal fat area (p=0.034). Figure 2 demonstrates that BMI correlated more strongly with the amount of intra-abdominal fat in subjects with the A allele than in subjects with the GG genotype.

Fig. 1
figure 1

The effect of SNP-43 on the ratio of intra-abdominal to total abdominal fat area measured with computed tomography by sex (adjusted for age, BMI and the rates of WBGU). Data are means±SD. p=0.014

Full size image
Fig. 2
figure 2

The correlation between BMI and intra-abdominal fat in subjects with the GA/AA genotype (triangles, solid lines) or the GG genotype (circles, dashed lines) of SNP-43. CT computed tomography. r 2=035, p<0.001 (GG/AA genotype); r 2=021, p<0.001(GG genotype)

Full size image

Study II gave further evidence of the association of SNP-43 with abdominal obesity in men (Table 3). Waist circumference tended to be larger in carriers of the A allele than in carriers of the GG genotype in Study I, and it was larger in carriers of the A allele than in carriers of the GG genotype in Study II (p=0.022). In both studies, diastolic blood pressure and insulin AUC in an OGTT were higher among the carriers of the A allele of SNP-43 than among carriers of the GG genotype.

Table 3 Clinical characteristics of men in Studies I and II according to the genotypes of SNP-43
Full size table

Discussion

Genetic variation in CAPN10 increases the risk of type 2 diabetes [2, 26, 27] with variants affecting risk alone (e.g. SNP-44 and SNP-43) or in combination (e.g. SNP-43, InDel-19 and SNP-63 haplotype). The variants in CAPN10 can result in the synthesis of a mutant protein (e.g. Thr504Ala) or may affect the levels of expression of calpain-10, e.g. SNP-43 in fat and skeletal muscle [2, 4, 28, 29]. The novel finding of the present study was that SNP-43 affects intra-abdominal obesity and insulin sensitivity in offspring of patients with type 2 diabetes (Study I). Furthermore, in an independent sample of men who had a history of type 2 diabetes in first-degree relatives, the A allele of SNP-43 was associated with surrogate markers of abdominal obesity (waist) and insulin sensitivity (insulin AUC in an OGTT).

Insulin resistance is an inherited trait in offspring of patients with type 2 diabetes [30], and precedes this disease. The central question is how CAPN10 regulates insulin action in patients at risk of developing diabetes. In our study the A allele of SNP-43 was associated with intra-abdominal fat area even after the adjustment for insulin sensitivity. The GG genotype has been associated with improved insulin action in isolated subcutaneous adipocytes [31], which could explain the preferential storage of fat in subcutaneous depots, in contrast to intra-abdominal fat that is less sensitive to insulin action [32]. Lower accumulation of intra-abdominal fat could then explain better whole-body insulin sensitivity in subjects with the GG genotype in this study. The same genotype GG has also been associated with lower CAPN10 expression in fat, and thus could have a role in adipocyte metabolism [28]. Interestingly, interaction between SNP-43 and obesity resembles the findings of another identified type 2 diabetes risk polymorphism, Pro12Ala of the peroxisome proliferator-activated receptor γ2 gene, which is known to regulate adipocyte metabolism. The effect of this polymorphism is also different in lean and obese subjects, because the insulin sensitive Ala12 allele [33] is associated with lower prevalence of type 2 diabetes [34] but increased weight gain [35] and diabetes risk in obese subjects [36].

The effect of the A allele of SNP-43 on intra-abdominal obesity (Study I) was observed in men but not in women (Fig. 1, Table 3). This finding is supported by another sample of men (Study II) when waist circumference was used as a surrogate marker of abdominal obesity. This sex-related difference could be due to the higher prevalence of central obesity in men. No association of SNP-43 and waist-to-hip ratio was observed in our study, in agreement with a study of 70-year-old Swedish men [37]. An interaction of SNP-43 with sex (Fig. 1) or BMI (Fig. 2) on their effects on abdominal obesity has not been reported in previous studies, which could explain discrepant findings. However, we cannot exclude the possibility that our findings are population specific.

The GG genotype of SNP-43 has been associated with type 2 diabetes and insulin resistance [4, 27], whereas in this study the A allele, and a haplotype carrying this allele, was associated with intra-abdominal obesity. These apparently discrepant findings may be due to the fact that the effects of calpain-10 on clinical measures may be complex and difficult to investigate in cross-sectional studies, similar to the Pro12Ala polymorphism of the peroxisome proliferator-activated receptor γ2 gene [33–36]. Subjects with the GG genotype of SNP-43 may originally have better adipocyte and whole-body insulin sensitivity and this may promote the accumulation of fat, e.g. in skeletal muscle and liver.

In conclusion, our study suggests that SNP-43 is associated with abdominal obesity in subjects at high risk of developing type 2 diabetes, indicating that calpain-10 may have a role in adipocyte metabolism.