Hypermethylation of hepatic Gck promoter in ageing rats contributes to diabetogenic potential
- First Online:
- Cite this article as:
- Jiang, M.H., Fei, J., Lan, M.S. et al. Diabetologia (2008) 51: 1525. doi:10.1007/s00125-008-1034-8
Hepatic glucokinase (GCK) is a key enzyme in glucose utilisation. Downregulation of its activity is associated with insulin resistance and type 2 diabetes mellitus. However, it is unknown whether hepatic Gck expression is influenced by age and is involved in ageing-mediated diabetes, and whether the degree of methylation of the hepatic Gck promoter is correlated with the transcription of Gck. To address the question, we evaluated hepatic Gck transcription and promoter methylation in young (14 weeks), adult (40 weeks) and aged (80 weeks) rats.
Hepatic glycogen, Gck expression and the kinase activity of GCK were measured in three age groups. The CpG methylation status was determined by both bisulphite direct sequencing and clone sequencing of the PCR amplificates of Gck promoter. The causal relationship between Gck methylation and mRNA expression was confirmed by treating rat primary hepatocytes with 5-aza-2′-deoxycytidine (5-Aza-CdR).
We have shown an age-associated decline in hepatic glycogen, Gck expression levels and the kinase activity of hepatic GCK. The eleven CpG sites studied displayed age-related progressive methylation changes in hepatic Gck promoter, which were confirmed by two methods: direct and clone sequencing. After 5-Aza-CdR treatment of rat primary hepatocytes, there was a fourfold increase in Gck expression.
Our results demonstrate that an age-related increase in methylation is negatively associated with hepatic Gck expression, suggesting that DNA methylation could be involved in increasing age-dependent susceptibility to hepatic insulin resistance and diabetes. Thus, the epigenetic modification of the hepatic Gck promoter may represent an important marker for diabetogenic potential during the ageing process.
KeywordsAgeing Diabetes Glucokinase Insulin resistance Methylation
Insulin resistance, glucose intolerance and type 2 diabetes mellitus are becoming progressively more common in older populations [1, 2, 3]. These age-related changes may be caused in part by increased adiposity and the decreased physical activity in older people [4, 5]. Although ageing plays a key role in the pathogenesis of insulin resistance, the underlying molecular mechanism is still poorly understood.
The pathophysiology of ageing and ageing-related diabetes is complex. Insulin resistance is a major contributor to development of the disease and to its complications [6, 7, 8]. Insulin resistance occurs because of defective insulin-mediated glucose utilisation by the liver and peripheral tissues. The progression of insulin resistance to diabetes is caused by numerous metabolic events spanning a period of many years [9, 10, 11]. Glucose phosphorylation by glucokinase (GCK) is the initial event in glucose metabolism in the liver. Thus, glucose phosphorylation by GCK is important in the regulation of glucose utilisation in hepatocytes. The main role of insulin in the activation of glycolytic genes appears to be the stimulation of GCK synthesis, which leads to increased glucose utilisation [12, 13, 14, 15, 16]. Regulation of Gck activity is mainly due to changes in the transcription of this gene. In fact, decreased GCK enzymatic activity has been reported in patients with type 2 diabetes mellitus , and mutations in the Gck gene can cause MODY-2 [18, 19]. Accordingly, Gck is considered to be as a diabetes-susceptibility gene.
Since type 2 diabetes mellitus incidence increases in older populations and their metabolic situations deteriorate over time, we hypothesised that the effect of age on glucose metabolism is associated with the level of hepatic Gck transcription and attempted to investigate mechanisms of the age regulation of Gck expression in liver. Ling et al demonstrated that DNA methylation is associated with age-dependent decline in NDUFB6 expression in human skeletal muscle , opening up the possibility that involvement of epigenetic factors like DNA methylation could predispose an individual to insulin resistance and type 2 diabetes mellitus. Accordingly, we specifically examined whether epigenetic modification was involved in the regulation of hepatic Gck expression in rats during postnatal life and whether it played a role in the pathogenesis of insulin resistance.
The aim of this study was to test whether the level of Gck expression and the activity of hepatic GCK were age-related and whether the degree of DNA methylation in Gck promoter changes with age, from young (14 weeks) to adult (40 weeks) and aged (80 weeks) animals. In addition, to reveal the functional significance of DNA methylation in rat livers, we also investigated whether Gck expression can be reactivated after treating rat primary hepatocytes with the DNA methyltransferase inhibitor, 5-aza-2′-deoxycytidine (5-Aza-CdR).
Male Wistar rats were obtained from the Animal Developmental Center, Chinese Academy of Sciences, Shanghai. We used three different age groups (n = 7 rats per group): 14 weeks (young), 40 weeks (adult) and 80 weeks (aged) in this study. Rats were housed under standard light–dark conditions (12/12 h light/dark cycle) and were given free access to commercial laboratory animal chow diet. Experiments were carried out after 16 h fasting. All experimental procedures involving the use of animals were conducted in accordance with NIH Guidelines and were reviewed and approved by the Animal Use and Care Committee of Fudan University.
The content of hepatic glycogen was measured as described [21, 22]. Briefly, livers were weighed and incubated in four volumes (vol./wt) 1 mol/l KOH for 20 min at 100°C. After cooling, the lysates were diluted tenfold with water then added slowly to the anthrone reagent, which was placed in cold water to prevent excessive heating. After mixing by lateral shaking, the tube was placed in a boiling water bath for exactly 5 min for colour development and subsequently cooled in tap water. Absorbance at 620 nm was measured against a blank control using glucose as a standard. The glycogen content was calculated using the formula: mg of glycogen per g of tissue = Du/Ds × 0.1 × 40 (fold total dilution) × 0.9, where DU is the OD value of the unknown, DS is the OD value of 0.1 mg of glucose in standard solution and 0.9 is the factor for converting the glucose value into glycogen value. Triplicate samples per liver were analysed and data are presented as mg of glucose units per g of liver.
RNA extraction and real-time PCR
Primer sequences for real-time PCR and bisulphite sequencing PCR of hepatic Gck promoter
mRNA expression by real-time PCR
Bisulphite sequencing PCR of Gck promoter
Determination of GCK activities
GCK activity was determined as described [25, 26]. Liver tissue (50 mg) was homogenised and diluted 20 times by buffer containing 50 mmol/l HEPES, 5 mmol/l MgCl2, 1 mmol/l EDTA, 100 mmol/l KCl and 2.5 mmol/l dithioerythritol at pH 7.4. The homogenate was then passed through a 23-gauge needle syringe (five times) to ensure the proper tissue suspension and later centrifuged at 10,000×g and 4°C. The reaction was started with 0.5 mmol/l NAD+, 4 U glucose-6-phosphate dehydrogenase and 5 mmol/l ATP at 30°C. The continuous spectrophotometric assay of glucose-phosphorylating activity was performed at two glucose concentrations: 100 mmol/l (measures all hexokinases, including GCK) and 0.5 mmol/l (measures only the low-Km hexokinases). The difference between the two assays gives GCK activity. One unit is the amount of enzyme that catalyses the formation of 1 μmol of substrate per min under the conditions of assay. The Bio-Rad DC (Bio-Rad Laboratories, Hercules, CA, USA) protein assay method was used to measure protein concentration.
Bisulphite-sequencing of Gck promoter
All reagents were purchased from Sigma (St Louis, MO, USA). Genomic DNA was prepared from the livers of three groups using the SDS and proteinase K methods, and then subjected to sodium bisulphite treatment . The Gck hepatic promoter sequence (−518 to 83) was amplified with AmpliTaq Gold polymerase (Applied Biosystems, Warrington, UK) by two rounds of PCR. Bisulphite-sequencing primers for rat Gck promoter were synthesised according to the reported sequences  (Table 1). PCR of the rat Gck promoter was performed under the following conditions: (1st round) 1 min 95°C, 2 min 50°C, 2 min 72°C for five cycles, then 1 min 95°C, 1 min 50°C, 1 min 72°C for 30 cycles. The first round products (2 μl) were then subjected to a second round PCR under the following conditions: (second round) 1 min 95°C, 2 min 58°C, 1 min 30 s 72°C for 35 cycles. We then sequenced 1 μl of the second round PCR product using a DNA sequencer (ABI PRISM 3730; Applied Biosystems, Foster City, CA, USA) with dye terminators (Perkin-Elmer, Foster City, CA, USA). The rest of the PCR products were kept and used in the following experiments.
Quantification of methylation levels
We used two methods to analyse levels of DNA methylation. First, we directly calculated the methylation level by measuring the heights of cytosine (shown as ‘C’) and thymine (shown as ‘T’) peaks at each CpG [29, 30, 31, 32, 33]. A single ‘C’ at the corresponding CpG site was considered to be 100% methylation and a single ‘T’ was considered to be no methylation; overlap of ‘C’ and ‘T’ was ranked as partial methylation. The percentage methylation was calculated as peak height of ‘C’ versus peak height of ‘C’ plus peak height of ‘T’ for each position.
In order to confirm whether the method above was reliable for estimation of methylation levels, we sequenced subclones of PCR products [34, 35, 36]. In brief, the residual PCR amplified products after direct sequencing were purified by gel electrophoresis and subcloned into a pGEM-T vector (Promega, Madison, WI, USA). A total of 30 clones per rat liver were isolated using a kit (QiaPrep Spin Plasmid Miniprep; Qiagen, Hilden, Germany) and sequenced. The methylation level for each CpG site was calculated by dividing the number of methylated CpGs of each site by the total number of clones sequenced.
Treatment of 5-Aza-CdR in rat primary hepatocytes
Hepatocytes were isolated from a 40-week adult male Wistar rat fasted for 48 h by two-stage collagenase perfusion of the liver as described . The isolated hepatocytes were seeded at a density of 2 × 106 cells per 35 mm plate. After 6 h incubation at 37°C and 5% CO2, attached cells were treated with freshly prepared 5 μmol/l 5-Aza-CdR (Sigma) and the medium was changed every 24 h. RNA was isolated from both untreated and treated confluent hepatocytes after 72 h as described above. Then the transcription level of Gck was analysed with real-time PCR.
Results are presented as mean ± SD. Statistical analysis was conducted using the software package Minitab (version 184.108.40.206; Minitab, State College, PA, USA). Significance was assessed by Student’s t test and one-way ANOVA followed by Bonferroni t test for the comparisons of multiple means within an experiment, or by Kruskal–Wallis test and subsequent pair-wise comparisons. The statistical method used for each group of data is indicated and a p value of less than 0.05 was considered to be statistically significant.
Physical and biochemical characteristics of the experimental rats
General and biochemical characteristics
Young, 14 weeks
Adult, 40 weeks
Aged, 80 weeks
248.7 ± 25.0
402.0 ± 34.8**
629.9 ± 41.8**
5.04 ± 0.36
5.02 ± 0.22
5.09 ± 0.44
174.5 ± 17.9
215.2 ± 15.7**
233.3 ± 14.1**,†
Hepatic glycogen contents in three groups of rats
Gck expression and activity in different age groups
To determine the age-related changes in the Gck gene products, we measured hepatic Gck mRNA levels and GCK enzymatic activities from three age groups. The Gck mRNA levels were comparable in real-time PCR with either rat Hprt or beta-actin as an internal control. As shown in Fig. 1b, hepatic Gck expression, with rat Hprt as an internal control, was 60% (p < 0.05) lower in adult and 86% (p < 0.01) lower in aged groups than in the young animal group, respectively. Furthermore, the decrease trend of GCK activities in the three groups was very similar to that of Gck mRNA, i.e. about 50 and 80% decreases in adult and aged rats, respectively, compared with young rats (Fig. 1c). The results indicate that the age-related decline of hepatic GCK activity is mainly controlled at the transcription level and is consistent with the tendency of changes seen in hepatic glycogen contents.
Methylation levels of the Gck promoter in different age groups
5-Aza-2′-deoxycytidine treatment of the primary rat hepatocytes upregulated Gck expression
Insulin resistance in liver is a major contributor to the increased prevalence of type 2 diabetes mellitus with age. Hepatic glucose metabolism is a good predictor of insulin sensitivity. The primary function of insulin in liver is to stimulate the key enzyme of glycolysis, GCK, which enhances hepatic glucose disposal. To determine why insulin sensitivity decreases with age, we analysed the Gck expression and the activity of GCK in livers from young (14 weeks), adult (40 weeks) and aged (80 weeks) rats. The adult and the aged rats showed a sharp decrease in the transcription of Gck compared with the young group. These declines were of the same magnitude as that observed in GCK activities of adult and aged rats. We deduce from this that the downregulation of hepatic Gck with ageing would severely attenuate the effect of insulin in liver and diminish glucose utilisation (Fig. 5).
In our study, Gck expression levels in liver were positively associated with hepatic glycogen. Consistent with the trend of Gck expression, glycogen levels progressively decline with age. GCK plays a crucial role in glycogen synthesis, and small changes in its total activity resulted in much larger changes in glycogen synthesis [41, 42]. The potential mechanism of GCK-mediated glycogen synthesis might be through catalysation of glucose 6-phosphate synthesis and promotion of glycogen synthase activation in hepatocytes [43, 44, 45, 46] (Fig. 5). Furthermore, we determined the glycogen content in rat muscles, a tissue that does not express Gck, and found no difference among the three age groups (Electronic supplementary material [ESM] Fig. 1).
Considering the significant role of Gck expression in hepatic insulin resistance and glycogen synthesis, we explored the potential mechanisms of Gck expression with ageing. Here, we have presented data to support the notion that differential DNA methylation results in the age-dependent reduction of Gck expression. The Gck promoter structure [13, 28, 38] is shown in Fig. 2a. The same region of this gene was hypomethylated in the young livers, with less than 10% methylation, while in adult livers the percentage methylation increased significantly (about 50%) and was highest (above 70%) in aged liver in the region investigated (Fig. 2b, c). These results demonstrate that hepatic Gck promoter methylation increased gradually with ageing (Fig. 5).
We quantified the methylation levels of Gck promoter in different age groups using bisulphite direct sequencing and clone sequencing of the PCR products derived from Gck promoter. In the present study, ‘T’ and ‘C’ overlapped at almost all of the CpG sites in the Gck promoter region, but there were significant differences in the three age groups (Fig. 3). From the sequencing chromatograms we directly calculated the degree of methylation of 11 CpG sites in the promoter region (Fig. 2b). Furthermore, we also sequenced the subclones of bisulphite-treated Gck promoter to confirm the methylation levels (Fig. 2c). The methylation levels calculated by two methods were quite similar, with fewer than 10% differences at some of the CpG sites. The minor differences can probably be attributed to the background noise affecting the exact peak heights of ‘C’ or ‘T’ in the sequencing chromatogram, as well as to the number of clones selected in this study. In spite of the minor differences, these two methods generated consistent results for calculating DNA methylation levels. Especially the direct sequencing of the PCR product is relatively straightforward.
To confirm the links between methylation and gene expression, a demethylation agent, 5-Aza-CdR, was used to treat primary adult rat hepatocytes. 5-Aza-CdR is a cytosine analogue that acts as a suicide substrate for DNA methyltransferase and hence is an inhibitor of DNA methyltransferase when incorporated into DNA at CpG dinucleotides of the target site for DNA methylation . As expected, 5-Aza-CdR significantly increased Gck mRNA transcription.
Gck gene has two promoters, one active in pancreas and the other active in liver . We determined pancreatic Gck expression using real-time PCR. There were no significant differences in the three age groups, which was consistent with earlier studies . Thus, we did not further study the methylation status of the Gck promoter specific to islets. We also quantified methylation of the liver-specific Gck promoter in rat skeletal muscles, a tissue that does not express Gck. Almost 100% methylation was found at the same region of the Gck promoter in all age groups (ESM Fig. 2). We were not sure whether the hypermethylation of Gck promoter was the major factor in silencing Gck mRNA expression in muscle or other Gck non-expressing tissues. However, according to our findings, the degree of CpG methylation in the Gck promoter increased in an age-dependent manner as a specific biological feature in rat livers, which should be the critical factor for suppression of Gck mRNA expression.
Ageing is associated with marked basal hyperinsulinaemia despite normal blood glucose. As shown in Table 2, the basal insulin levels were higher in the adult and aged rats than in young animals. Basal hyperinsulinaemia is suggestive of an insulin-resistant state, so both adult and aged rats exhibited marked whole-body insulin resistance. Furthermore, body weight increase (Table 2) and fat accumulation of healthy old rats might aggravate insulin resistance in liver and other insulin-targeting organs, so the combined impact of ageing and obesity on insulin sensitivity contributes greatly to insulin resistance in elderly people [1, 4, 50]. Accordingly, we cannot rule out the potential role of age-related obesity in hypermethylation of the Gck promoter, since, under normal conditions, rats increase weight as they age. We have now begun a new study of Gck methylation in either ageing or obese rats. Our preliminary findings indicate that obesity could increase susceptibility to age-induced insulin resistance and the level of hepatic Gck methylation in aged rats (M.H. Jiang, J. Fei, M.S. Lan, Z.P. Lu, M. Liu, W.W. Fan, X. Gao and D.R. Lu, unpublished results).
In summary, we have demonstrated that hypermethylation of hepatic Gck promoter is associated with the downregulation of hepatic Gck transcription in aged rats and contributes to diabetogenic potential (Fig. 5). However, the detailed mechanisms that link ageing and hypermethylation of Gck promoter still merit further investigation. Also, a systematic profiling of DNA methylation changes in the ageing process would yield valuable insights into the epigenetic mechanism of ageing, insulin resistance and other ageing-related diseases.
We would like to thank J. Gu (Shanghai Experimental Animal Center, China Academy of Sciences, Shanghai, China) for his help with animal studies. This work was supported in part by the National Key Basic Research Program of China (2006CB503910), Science and Technology Commission of Shanghai Municipality (07JC14011) and Shanghai Leading Academic Discipline Project (B111).
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.