Diabetologia

, Volume 50, Issue 8, pp 1615–1620

Impact of the peroxisome proliferator activated receptor-γ coactivator-1β (PGC-1β) Ala203Pro polymorphism on in vivo metabolism, PGC-1β expression and fibre type composition in human skeletal muscle

Authors

    • Department of Clinical Sciences/Diabetes and EndocrinologyLund University, CRC, University Hospital MAS
  • L. Wegner
    • Steno Diabetes Center
  • G. Andersen
    • Steno Diabetes Center
  • P. Almgren
    • Department of Clinical Sciences/Diabetes and EndocrinologyLund University, CRC, University Hospital MAS
  • T. Hansen
    • Steno Diabetes Center
  • O. Pedersen
    • Steno Diabetes Center
    • Faculty of Health ScienceUniversity of Aarhus
  • L. Groop
    • Department of Clinical Sciences/Diabetes and EndocrinologyLund University, CRC, University Hospital MAS
  • A. Vaag
    • Department of Clinical Sciences/Diabetes and EndocrinologyLund University, CRC, University Hospital MAS
    • Steno Diabetes Center
  • P. Poulsen
    • Steno Diabetes Center
Article

DOI: 10.1007/s00125-007-0729-6

Cite this article as:
Ling, C., Wegner, L., Andersen, G. et al. Diabetologia (2007) 50: 1615. doi:10.1007/s00125-007-0729-6

Abstract

Aims/hypothesis

Peroxisome proliferator activated receptor-γ coactivator-1β (PGC-1β, also known as PPARGC1B) expression is reduced in skeletal muscle from patients with type 2 diabetes mellitus and in elderly subjects. Ala203Pro, a common variant in the PGC-1β gene is associated with obesity. The aim of this study was to investigate whether the PGC-1β Ala203Pro polymorphism influences the age-related decline in skeletal muscle PGC-1β expression, in vivo metabolism and markers for muscle fibre type composition.

Materials and methods

The PGC-1β Ala203Pro polymorphism was genotyped in 110 young (age 28.0 ± 1.9 years) and 86 elderly (age 62.4 ± 2.0 years) twins and related to muscle PGC-1β expression, in vivo metabolism and markers for fibre type composition.

Results

Insulin-stimulated non-oxidative glucose metabolism (NOGM; p = 0.025) and glycolytic flux rate (GF; p = 0.026) were reduced in young Ala/Ala carriers compared with carriers of a 203Pro allele. In addition, a regression analysis, correcting for covariates, showed that the PGC-1β 203Pro allele was positively related to insulin-stimulated NOGM and GF in the young twins. While muscle expression of PGC-1β was reduced in elderly compared with young carriers of the Ala/Ala genotype (p ≤ 0.001), there was no significant age-related decline in PGC-1β expression in carriers of the 203Pro allele (p ≥ 0.4). However, a regression analysis, correcting for covariates, showed that only age was significantly related to muscle PGC-1β expression. Finally, PGC-1β expression correlated positively with markers for oxidative fibres in human muscle.

Conclusions/interpretation

This study suggests that young carriers of a PGC-1β 203Pro allele have enhanced insulin-stimulated glucose metabolism and may be protected against an age-related decline in PGC-1β expression in muscle.

Keywords

Fibre type Gene expression In vivo metabolism PGC-1β Polymorphism PPARGC1B Skeletal muscle

Abbreviations

DZ

dizygotic

ERR

oestrogen receptor-related receptor

GEE

generalised estimating equations

GF

glycolytic flux

LBM

lean body mass

MZ

monozygotic

NOGM

non-oxidative glucose metabolism

PGC-1β

peroxisome proliferator activated receptor-γ coactivator-1β

PPAR

peroxisome proliferator activated receptor

Introduction

Peroxisome proliferator activated receptor (PPAR)-γ coactivator-1β (PGC-1β) is part of a small family of transcriptional co-activators that also includes PPAR-γ coactivator-1α (PGC-1α) and PGC-1 related coactivator (PRC). These co-activators influence metabolic pathways in muscle, liver, adipose tissue, pancreas and heart by co-activating a number of transcription factors including PPARγ, PPARα, hepatocyte nuclear factor 4 alpha (HNF4α), nuclear respiratory factor 1 (NRF-1), myocyte enhancer factor 2 (MEF2), oestrogen receptor-related receptor (ERR)α and forkhead box O1 (FOXO1) [1].

Compared with healthy control subjects, patients with type 2 diabetes mellitus have reduced expression of PGC-1α (also known as PPARGC1A) and PGC-1β (also known as PPARGC1B), as well as of a set of genes involved in oxidative phosphorylation in skeletal muscle [2, 3]. We originally showed that a common variant in the PGC-1α gene, Gly482Ser, is associated with increased risk of type 2 diabetes mellitus as well as with altered lipid and glucose metabolism [4]. Furthermore, we demonstrated that combinations of genetic and environmental factors influence the PGC-1α and PGC-1β expression in human muscle [5]. Interestingly, the PGC-1α Gly482Ser polymorphism was found to accelerate an age-related decline in the expression of both PGC-1α and PGC-1β in muscle [5]. In a recent study, we identified a polymorphism in the PGC-1β gene, Ala203Pro, where the common allele Ala203 conferred an increased risk of obesity [6]. In addition, recent studies in rodents suggest that PGC-1β influences the composition of fibre types in skeletal muscle [79].

The present study aimed to investigate whether this common polymorphism in PGC-1β, Ala203Pro, influences in vivo glucose and fat metabolism as well as PGC-1β expression in muscle of young (n = 110, age 28.0 ± 1.9 years) and elderly (n = 86, age 62.4 ± 2.0 years) twins without known diabetes. We also investigated whether PGC-1β relates to markers for fibre type composition in skeletal muscle of twins.

Materials and methods

Participants

Participants were identified through the Danish Twin Register. Ninety-eight twin pairs (33 younger monozygotic [MZ]; 22 younger dizygotic [DZ]; 21 elderly MZ; 22 elderly DZ) without known diabetes were enrolled in the clinical examination. The clinical examination and characteristics of these twins have been described previously [5]. Informed consent was obtained from all subjects. The present study was approved by the regional ethical committees and conducted according to the Helsinki Declaration.

Analysis of PGC-1β, myosin heavy chain (MHC)7, MHCIIa and MHCIIx/d mRNA levels in muscle

Gene expression was analysed in 156 muscle biopsies of the vastus lateralis muscle taken in both the fasting state and after a 2 h hyperinsulinaemic–euglycaemic clamp (40 mU m−2 min−1) from 78 twin pairs (26 younger MZ; 18 younger DZ; 15 elderly MZ; 19 elderly DZ) using TaqMan Real-Time PCR with an ABI 7900 system (Applied Biosystems, Foster City, CA, USA) and assays-on-demand (all Applied Biosystems) for PGC-1β (Hs00370186_m1), myosin heavy chain (MHC, also known as MYH)7 (Hs00165276_m1), MHCIIa (Hs00430042_m1) and MHCIIx/d (Hs00428600_m1). The transcript quantity was normalised to that of cyclophilin A mRNA (4326316E; Applied Biosystems).

Genotyping

The PGC-1β Ala203Pro variant was genotyped using PCR restriction fragment length polymorphism analysis with the primers 5′-GTG GGG CTT TGT CAG TGA AT-3′ and 5′-GGA CTC CTG GAG GCA TGG TG-3′ and digestion with NlaIV. The restriction enzyme digests were separated on 4% agarose gels and visualised by ethidium bromide staining.

Statistical methods

Because of the strong intrapair correlation of twin data, conventional tests of differences between variable (y) means are not valid. To correct for this dependence we used generalised estimating equations (GEE) methodology (y = α + βx) to provide valid standard errors for the β coefficients [10, 11]. The β coefficient was estimated from all observations, whereas for calculation of the variance of β, each family was considered as one cluster.

To identify factors which were independently associated with the response variable, we used backwards regression analysis with a p > 0.05 as the defining criteria for exclusion. In this analysis, too, we used generalised estimation equations (GEE) methodology to obtain valid tests.

Results

Clinical characteristics

As presented previously, elderly twins were more obese and had reduced glucose disposal rate, storage and oxidation than younger twins [5].

Influence of the PGC-1β Ala203Pro polymorphism on the age-related decline in muscle gene expression

We have previously demonstrated that elderly twins have significantly lower mRNA levels of PGC-1β in muscle than younger twins [5]. We demonstrate here that this age-related decline in muscle PGC-1β expression might be influenced by the PGC-1β Ala203Pro polymorphism (Fig. 1a,b). Both in the basal state (young 1.37 ± 0.09 vs elderly 0.93 ± 0.07; p = 0.0007) and in the insulin-stimulated state (young 2.51 ± 0.17 vs elderly 1.58 ± 0.12; p = 0.00004), the elderly twins carrying the PGC-1β Ala203Ala genotype had lower PGC-1β gene expression than young Ala203Ala carriers. There was, however, no significant influence of age on muscle PGC-1β expression in carriers of a PGC-1β 203Pro allele, either in the basal (young 1.24 ± 0.16 vs elderly 1.21 ± 0.21; p = 0.9) or in the insulin-stimulated state (young 2.41 ± 0.43 vs elderly 1.92 ± 0.24; p = 0.4; Fig. 1a,b). We performed a regression analysis to test whether any of the following factors influence skeletal muscle PGC-1β mRNA expression: age, sex, percentage body fat, birth-weight, zygosity, the PGC-1β Ala203Pro polymorphism and the interaction between age and the PGC-1β Ala203Pro polymorphism,. When correcting for these variables in a regression analysis, only age was significantly related to skeletal muscle PGC-1β expression (basal expression, regression coefficient =  −0.01; p = 0.003; insulin-stimulated expression, regression coefficient =  −0.02; p < 0.0005).
https://static-content.springer.com/image/art%3A10.1007%2Fs00125-007-0729-6/MediaObjects/125_2007_729_Fig1_HTML.gif
Fig. 1

The effect of the PGC-1β Ala203Pro polymorphism on the age-related decline in skeletal muscle gene expression in twins. The mRNA level of PGC-1β was measured in skeletal muscle of young (open bars) and elderly (closed bars) twins a before and b after a hyperinsulinaemic clamp and related to the PGC-1β Ala203Pro polymorphism. Results are expressed as the mean±SEM for Ala/Ala carriers (n = 80 and n = 50 for young and elderly, respectively) and carriers of a 203Pro allele, Ala/Pro + Pro/Pro (n = 8 and n = 18 for young and elderly, respectively). To correct for the lack of independence of twin data, all comparisons of mean differences between groups were performed using generalised estimation equations (GEE) methodology. **, p < 0.001; #, p = 0.9; †, p < 0.0001; ‡, p = 0.4

Impact of the PGC-1β Ala203Pro polymorphism on in vivo glucose and fat metabolism

Young carriers of a PGC-1β 203Pro allele had increased glucose storage or non-oxidative glucose metabolism (NOGM; 48.8 ± 5.4 vs 38.0 ± 1.6 μmol kg lean body mass [LBM]−1 min−1 [8.78 ± 0.97 vs 6.83 ± 0.28 mg kg LBM−1 min−1]; p = 0.025) and glycolytic flux (GF; 30.9 ± 3.2 vs 24.1 ± 1.0 μmol kg LBM−1 min−1 [5.56 ± 0.57 vs 4.34 ± 0.19 mg kg LBM−1 min−1]; p = 0.026) during insulin stimulation compared with the Ala/Ala group (Table 1). However, no significant influence of this polymorphism on glucose disposal, glucose and fat oxidation was detected in the young twins (Table 1). Furthermore, this polymorphism did not influence glucose or fat metabolism in the elderly twins (data not shown).
Table 1

The influence of the PGC-1β Ala203Pro polymorphism on insulin-stimulated metabolic variables in young twins

Variable

Ala/Ala

Ala/Pro+Pro/Pro

p value

n=94

n=12

Glucose disposal rate

μmol kg LBM−1 min−1

62.6 ± 2.0

73.6 ± 5.3

 

mg kg LBM−1 min−1

11.52 ± 0.33

13.26 ± 0.96

0.09

NOGM

μmol kg LBM−1 min−1

38.0 ± 1.6

48.8 ± 5.4

 

mg kg LBM−1 min−1

6.83 ± 0.28

8.78 ± 0.97

0.025

Glucose oxidation

μmol kg LBM−1 min−1

26 ± 0.81

24.8 ± 1.5

 

mg kg LBM−1 min−1

4.68 ± 0.15

4.47 ± 0.27

0.8

GF

μmol kg LBM−1 min−1

24.1 ± 1.0

30.9 ± 3.2

 

mg kg LBM−1 min−1

4.34 ± 0.19

5.56 ± 0.57

0.026

Fat oxidation (mg kg LBM−1 min−1)

0.33 ± 0.05

0.38 ± 0.09

0.76

Data are expressed as means±SEM

To adjust for the lack of independence between MZ and DZ twins, all comparisons of mean differences between groups were performed using generalised estimation equations (GEE) methodology

We then performed a regression analysis to test whether any of the following variables influenced insulin-stimulated glucose disposal and storage, glucose and fat oxidation, and GF: sex, percentage body fat, birth-weight, zygosity, the PGC-1β Ala203Pro polymorphism and both basal and insulin-stimulated muscle PGC-1β mRNA expression (Table 2). The regression analysis was performed separately in young and elderly twins, since the analysis above (Table 1) only demonstrated effects of the PGC-1β Ala203Pro polymorphism in young twins. After correction for covariates, the PGC-1β 203Pro allele was consistently positively related to insulin-stimulated NOGM (p = 0.03) and GF (p = 0.001) in the young twins (Table 2). Furthermore, in agreement with our previous study [5], basal PGC-1β mRNA expression was positively related to NOGM (p = 0.005) and fat oxidation (p = 0.006) and negatively related to glucose oxidation (p = 0.02; Table 2). Finally, when performing regression analysis in the elderly twins, the PGC-1β Ala203Pro polymorphism was found to be not significantly related to any of the metabolic variables studied (data not shown).
Table 2

Factors influencing insulin-stimulated glucose disposal rate, glucose storage (NOGM), glucose oxidation, glycolytic flux and fat oxidation in skeletal muscle of 110 young twins

Variable

Regression coefficient

p value

Glucose disposal rate

Sex (male vs female)

2.4

0.02

Percentage body fat

−0.17

0.01

Zygosity (MZ vs DZ)

−2.1

0.007

NOGM

Percentage body fat

−0.09

0.025

Zygosity (MZ vs DZ)

−1.8

0.003

PGC-1β Ala203Pro polymorphism

2.2

0.03

PGC-1β gene expression, basal

0.83

0.005

Glucose oxidation

Sex (male vs female)

1.1

0.0001

PGC-1β gene expression, basal

−0.35

0.02

GF

PGC-1β Ala203Pro polymorphism

2.1

0.001

Fat oxidation

PGC-1β gene expression, basal

0.15

0.006

By regression analysis we tested whether the following factors influence insulin-stimulated glucose disposal, glucose storage (NOGM), glucose oxidation, glycolytic flux and fat oxidation: sex (men coded 0, women coded 1), percentage body fat, birth-weight, zygosity, the PGC-1β Ala203Pro polymorphism and both basal and insulin-stimulated skeletal muscle PGC-1β gene expression

p < 0.05 for all final models

PGC-1β mRNA expression correlates with markers for oxidative fibre types in human skeletal muscle

We investigated whether basal PGC-1β expression correlates with basal mRNA expression of markers for three different fibre types, namely MHC7 (type I fibres, slow-twitch oxidative), MHCIIa (type IIA fibres, fast-twitch oxidative) and MHCIIx/d (type IIB fibres, fast-twitch glycolytic), in human skeletal muscle (n = 156, young and elderly twins). While both MHC7 and MHCIIa correlated positively with PGC-1β expression (r = 0.28; p = 0.0005 and r = 0.31; p = 0.00008, respectively), there was no correlation between MHCIIx/d and PGC-1β (r = −0.03; p = 0.6). Moreover, elderly twins had 25% lower mRNA expression of MHCIIa than young twins (p = 0.0005). However, the mRNA expression of MHC7 and MHCIIx/d did not change with age (p = 0.5 and p = 0.9, respectively). Finally, because PGC-1β correlated with MHC7 and MHCIIa expression, we tested whether the PGC-1β Ala203Pro polymorphism influenced the basal expression of these two fibre type markers in young and elderly twins. Indeed, elderly carriers of a PGC-1β 203Pro allele (n = 18) had increased MHC7 expression (2.1 ± 0.3 vs 1.6 ± 0.1; p < 0.05) compared with the Ala/Ala group (n = 62). However, there was no significant influence of this polymorphism on basal MHCIIa expression in elderly twins or on MHC7 and MHCIIa expression in young twins (data not shown).

Discussion

PGC-1β expression is reduced in muscle of healthy elderly individuals and in patients with type 2 diabetes mellitus [3, 5]. In addition, PGC-1β was previously found to increase the expression of mitochondrial genes and the formation of oxidative fibres in rodent muscle [1, 79]. We have also shown that this co-activator is positively related to NOGM and lipid oxidation in human muscle in vivo [5]. In the present study we found that in vivo measures of NOGM and GF were significantly increased in young carriers of the PGC-1β 203Pro allele. Our data also suggest that PGC-1β influences the formation of oxidative fibres in human skeletal muscle.

The risk of developing obesity and metabolic disease including insulin resistance, dyslipidaemia and type 2 diabetes mellitus increases with high-fat/high-energy diets and reduced physical activity. However, there is a large inter-individual variation in the metabolic response to diet and exercise and not all individuals exposed to a sedentary lifestyle and poor diets actually develop obesity or metabolic disease. A likely reason is that inherited factors influence the susceptibility to obesity; indeed, we recently identified such a genetic factor, the common PGC-1β Ala203 allele, which we found to be associated with increased risk of obesity [6]. The present study demonstrates further how this polymorphism influences glucose metabolism in an age-dependent manner. PGC-1β was previously suggested to influence metabolism in rodents by co-activating the ERR, resulting in elevated energy expenditure and resistance to obesity [12]. The PGC-1β Ala203Pro polymorphism changes an amino acid from alanine to proline, which might influence the ability of PGC-1β to co-activate transcription factors, e.g. ERR. This may explain, or at least contribute to, the positive effects of the PGC-1β 203Pro allele on peripheral (i.e. muscle) glucose metabolism in the young. The fact that elderly 203Pro carriers did not show an age-related reduction in muscle PGC-1β expression could be interpreted as an attempt to protect these subjects against reductions in glucose metabolism as they age. Despite this, we found no positive influence of the PGC-1β 203Pro allele on glucose metabolism in the elderly, suggesting that these subjects cannot compensate for the impact of other degenerative factors influencing the decline in metabolism during ageing.

In the present study we were unable to identify an influence of the PGC-1β Ala203Pro polymorphism on measures of obesity such as percentage body fat, BMI and WHR (data not shown). The most plausible reason for this is lack of statistical power, as this study included 196 twins compared with 7,790 middle-aged subjects in our previous case–control study [6].

Our detailed metabolic studies allowed us to gain insight into distinct effects of the PGC-1β Ala203Pro polymorphism on in vivo glucose metabolism and cellular glucose partitioning. Indeed, this polymorphism influenced NOGM (glycogen synthesis) as well as ‘exogenous’ (i.e. extracellular-derived glucose uptake) GF rates, as determined by the appearance rate of 3H-labelled water. This is in contrast to the glucose oxidation rate as determined from indirect calorimetry, which measures the oxidation of glucose from both intra- and extracellular components. We have previously reported apparently divergent results regarding glucose oxidation and GF rates [13], which are likely to be explained in particular by the unknown contribution of glucose oxidation from endogenous glycogen pools. In this study, the PGC-1β Ala203Pro polymorphism influenced insulin-stimulated GF but not glucose oxidation in the young twins. The explanation for this may be that the influence is exerted at the level of membrane glucose transport and that the intracellular response to that may be a compensatory elevated rate of glycogenolysis, resulting in near-normal total (endogeous and exogenous) glucose oxidation.

Recent studies of over-production of PGC-1β in rodent muscle suggest that PGC-1β plays a positive role in the formation of oxidative fibres in skeletal muscle [7, 8]. Moreover, soleus muscle from mice lacking PGC-1β showed impaired ATP-production and reduced expression of genes from the respiratory chain [9]. The present study suggests that PGC-1β also influences the formation of oxidative fibres in human muscle, since the expression of PGC-1β correlated positively with markers for both slow- and fast-twitch oxidative fibres. In contrast, Krämer et al. [14] did not find any correlation between the expression level of PGC-1β and fibre type composition in human muscle, which could be explained by the small number of subjects (n = 20).

In summary, our data provide evidence that carriers of a PGC-1β 203Pro allele have enhanced rates of insulin-stimulated glucose metabolism when young and may be protected against an age-related decline in skeletal muscle gene expression of this key co-activator of energy metabolism. This study also proposes a role for PGC-1β in determining fibre type composition in human skeletal muscle.

Acknowledgements

This investigation was supported by grants from Region Skåne, the Swedish Research Council, Novo Nordisk, the Danish Medical Researh Council, the Danish Diabetes Association, the European Union (EUGENE2, LSHM-CT-2004-512013), Malmö University Hospital, Swegene, the Diabetes Programme at Lund University, Hedlund, Bergvall, Wiberg, Påhlsson, Lundberg and Exgenesis (grant 005272). We thank M. Svensson, E. Nilsson and M. Modest for excellent technical assistance.

Duality of interest

The authors declare that there is no duality of interest associated with this manuscript.

Copyright information

© Springer-Verlag 2007