Glucose inhibits the insulin-induced activation of the insulin-degrading enzyme in HepG2 cells
- 1.2k Downloads
Hepatic insulin degradation decreases in type 2 diabetes. Insulin-degrading enzyme (IDE) plays a key role in insulin degradation and its gene is located in a diabetes-associated chromosomal region. We hypothesised that IDE may be regulated by insulin and/or glucose in a liver cell model. To validate the observed regulation of IDE in vivo, we analysed biopsies of human adipose tissue during different clamp experiments in men.
Human hepatoma HepG2 cells were incubated in normal (1 g/l) or high (4.5 g/l) glucose medium and treated with insulin for 24 h. Catalytic activity, mRNA and protein levels of IDE were assessed. IDE mRNA levels were measured in biopsies of human subcutaneous adipose tissue before and at 240 min of hyperinsulinaemic, euglycaemic and hyperglycaemic clamps.
In HepG2 cells, insulin increased IDE activity under normal glucose conditions with no change in IDE mRNA or protein levels. Under conditions of high glucose, insulin increased mRNA levels of IDE without changes in IDE activity. Both in normal and high glucose medium, insulin increased levels of the catalytically more active 15a IDE isoform compared with the 15b isoform. In subcutaneous adipose tissue, IDE mRNA levels were not significantly upregulated after euglycaemic or hyperglycaemic clamps.
Insulin increases IDE activity in HepG2 cells in normal but not in high glucose conditions. This disturbance cannot be explained by corresponding alterations in IDE protein levels or IDE splicing. The loss of insulin-induced regulation of IDE activity under hyperglycaemia may contribute to the reduced insulin extraction and peripheral hyperinsulinaemia in type 2 diabetes.
KeywordsEuglycaemic–hyperinsulinaemic clamp HepG2 cells Hyperglycaemic–hyperinsulinaemic clamp Insulin-degrading enzyme Subcutaneous adipose tissue Type 2 diabetes mellitus
Hypoxanthine phosphoribosyltransferase 1
Ribosomal protein large protein 0
Quantitative real-time PCR
Type 2 diabetes is characterised by insulin resistance, pancreatic beta cell dysfunction, and probably alterations in insulin metabolism [1, 2]. Decreased hepatic insulin degradation is an early phenotypical marker of disturbances in insulin metabolism and it was observed in first-degree relatives of type 2 diabetes patients , in obese insulin-resistant persons and children with metabolic syndrome [3, 4, 5]. Decreased insulin degradation may intensify insulin resistance via chronically elevated circulating fasting and postprandial insulin levels. However, mechanisms leading to the alteration in insulin degradation remain unclear.
Insulin-degrading enzyme (IDE) is thought to be a major enzyme responsible for insulin degradation . IDE is a 110 kDa zinc-requiring metalloproteinase located in the cytoplasm, cell membranes and some cell organelles (endosomes, peroxisomes and mitochondria) and is secreted into the extracellular space [6, 7, 8]. Insulin is the preferred substrate for IDE, but a large body of other substrates, including glucagon, atrial natriuretic peptide and β-amyloid peptide, were reported . IDE has also regulatory functions for proteasome activity, steroid receptors, peroxisomal fatty acid oxidation, growth and development [9, 10, 11, 12]. IDE is ubiquitously expressed, both in insulin-sensitive and in non-insulin-sensitive cells, indicating a multifunctional role for this protein .
Linkage with the IDE chromosome region, 10q23–q25, was identified for type 2 diabetes and related quantitative traits [13, 14]. Evidence for a putative influence of IDE on the pathogenesis of type 2 diabetes was confirmed in association studies in several independent human populations [15, 16, 17] and in a recently published meta-analysis .
IDE knockout mice are characterised by the classic features of type 2 diabetes: decreased insulin degradation, hyperinsulinaemia and glucose intolerance . Loss-of-function mutations or pharmacological inhibition of IDE increases amyloid accumulation in pancreatic beta cells and in the central nervous system [19, 20, 21]. On the other hand, IDE overproduction increases insulin degradation and decreases the efficiency of insulin stimulation in the insulin signalling pathway . These data demonstrate that the regulation of IDE expression and/or the activity of the protein may contribute to the pathogenesis of type 2 diabetes.
All insulin-sensitive cells contain IDE and degrade insulin. However, the liver is the main site of insulin clearance, removing approximately 75% of the insulin during the first portal passage . Few reports have been published concerning the functional regulation of IDE in the liver or in liver cells [24, 25, 26].
In the present study, we performed a detailed analysis of the regulation of IDE functions by different concentrations of insulin and glucose in HepG2 cells. In this liver cell model, IDE regulation was analysed at three levels: IDE transcription, IDE mRNA translation and IDE protein activity. Moreover, in humans we assessed the regulation of IDE by different clamped glucose and insulin concentrations in vivo in subcutaneous adipose tissue.
Human hepatoma HepG2 cells were cultured in DMEM (Biochrom, Berlin, Germany) containing 10% (vol./vol.) FBS (Gibco/Invitrogen, Eggenstein, Germany) and 1 g/l d-glucose. Cells were incubated in a humidified atmosphere of 5% CO2 at 37°C. For experiments, HepG2 cells were maintained in complete medium with 10% (vol./vol.) FBS and either normal (1.0 g/l) or high (4.5 g/l) concentrations of d-glucose for 48–72 h to 80% confluence. Incubation of cells in high-glucose medium was used as a model of insulin resistance . After 24 h of serum starvation cells were treated with 0.1, 1, 10, 100 or 200 nmol/l insulin (Sigma Aldrich, Taufkirchen, Germany) for 24 h. As an osmotic control, 3.64 g/l mannitol (Sigma Aldrich) was added to normal glucose medium in simultaneous wells.
Total RNA from subcutaneous adipose tissue biopsy samples was extracted using the RNeasy Lipid Tissue Mini Kit (Qiagen, Hilden, Germany). Total RNA was extracted from HepG2 cells with Trizol reagent (Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions. The RNA concentration was measured and quality controlled using an ND-1000 spectrophotometer (Nanodrop, PeqLab, Erlangen, Germany).
Reverse transcription PCR (RT-PCR)
Primers used in the present study
QuantiTect Primer Assay (Qiagen)
Quantitative real-time PCR
Quantitative real-time PCR (qRT-PCR) was performed in a 384-well plate in an ABI Prism 7700 sequence detection system (Applied Biosystems). Quantitative real-time PCR was carried out on equal amounts of cDNA in triplicate for each sample from three independent experiments using Power SYBR Green PCR Master Mix (Applied Biosystems). The thermal cycling conditions for qRT-PCR were as follows: initial denaturation at 95°C for 10 min, followed by 47 cycles of denaturing at 95°C for 15 s, and annealing/extension for 1 min at 60°C. Quantification of mRNA levels was performed by the standard curve method. IDE expression in the cell experiments was measured using IDE_real_F1/IDE_real_R1 primers and normalised to the hypoxanthine phosphoribosyltransferase 1 (HPRT1) gene as internal control. IDE expression in adipose tissue biopsy samples was measured using IDE_real_F2/IDE_real_R2 primers and normalised to the ribosomal protein large protein 0 (RPLP0) gene. The RPLP0 QuantiTect Primer Assay was purchased from Qiagen.
For the determination of exons 15a- and 15b-specific IDE mRNA levels, previously described primers were used . All other oligonucleotides for qRT-PCR were designed with Primer Express software (PE Applied Biosystems, Darmstadt, Germany). PCR primer pairs for qRT-PCR were located in different exons to prevent possible amplification of genomic DNA. Sequences of all primers used are listed in Table 1.
HepG2 cell protein extracts were prepared in cell lysis buffer (Cell Signaling, Frankfurt, Germany), sonicated and centrifuged at 700 g for 10 min to remove cell debris. Protein concentrations in total cell lysates were measured by the bicinchoninic acid assay (Pierce/Thermo Scientific, Schwerte, Germany) using a Wallac Victor2 multilabel plate reader (PerkinElmer, Freiburg, Germany). Lysates were separated by 12% (vol./vol.) SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride membranes (Millipore, Schwalbach, Germany). The membranes were blocked with 5% (wt/vol.) non-fat dry milk in Tris-buffered saline with 0.1% (vol./vol.) Tween 20 and hybridised with the appropriate primary antibodies: rabbit IDE antibody (1:1,000; Chemicon, Hofheim, Germany) or rabbit β-actin antibody (1:2,000; Sigma Aldrich), followed by incubation with anti-rabbit horseradish peroxidase-conjugated secondary antibodies (1:2,000; Cell Signaling Technology). Immunoreactive bands were visualised using the Lumiglo chemiluminescence detection system (Cell Signaling Technology). The chemiluminescence signal was quantified using a CCD camera and Image Reader LAS-1000 Pro v 2.1 software (Fujifilm, Tokyo, Japan) and an AIDA Image Analyzer (Raytest, Straubenhardt, Germany). Relative IDE protein levels were calculated by normalisation of the IDE levels with β-actin signals, and mean values from three independent experiments were taken.
IDE activity assay
HepG2 cells were disrupted by scraping in CytoBuster Protein extraction reagent (Novagen, Schwalbach, Germany) for the preparation of total cell extracts or hypotonic buffer (50 mmol/l Tris–HCl, pH 7.4) for the preparation of membrane and cytosolic fractions, followed by 30 min incubation and extrusion through a 23-gauge hypodermic needle. Membrane and cytosolic fractions were prepared as described previously . In brief, the post-nuclear supernatant fraction was centrifuged at 100,000 g for 1 h to separate cytosolic (supernatant fraction) and membrane (pellet) fractions. The pellet was resuspended in 100 µl of 50 mmol/l Tris-HCl (pH 7.4) and stored at −80°C.
The traditional insulin degradation assay using 125I-labelled insulin and trichloroacetic acid [21, 28] requires radioactive material to be handled. Therefore, IDE activity was assessed with the non-radioactive InnoZyme Insulysin/IDE Immunocapture Activity Assay Kit (Calbiochem/Merck, Schwalbach, Germany) which has an effective working range of 50–1,000 ng/ml. Fluorometric quantification of insulin degradation by membrane samples was performed using the FRET substrate (Mca-GGFLRKHGQEDDnp), which is a component of the InnoZyme Insulysin/IDE Immunocapture Activity Assay Kit. For this assay, membrane suspensions were diluted 1:75 in 50 mmol/l Tris-HCl. Five microlitres of the diluted membrane fraction was incubated with 100 µl of 1× FRET substrate solution in assay buffer for 3.5 h at 37°C. Sample fluorescence was determined at an excitation wavelength of 320 nm and an emission wavelength of 405 nm on a SpectraMax M2e plate reader (Molecular Devices, Ismaning, Germany). Reactions were performed in duplicate for samples from three independent experiments, and the percentage of hydrolysis was calculated from standards containing rat recombinant IDE. Enzyme activity was assessed as relative fluorescent units per mg of total protein. Protein concentration was determined by the bicinchoninic acid assay (Pierce).
Glucose clamp studies
The study protocol was approved by the ethics committee of Brandenburg (registration number: AS 2(a)/2005), Germany. Before the study, informed written consent was obtained from all participants.
Seventeen healthy non-diabetic men with moderate obesity participated in two of the following procedures: (1) an infusion test with placebo (infusion of 0.9% [wt/vol.] NaCl solution) (n = 11); (2) a hyperinsulinaemic–euglycaemic clamp (EC) with continuous infusion of 240 pmol m−2 min−1 insulin at a steady-state capillary plasma glucose concentration of 4.4 mmol/l (n = 10); (3) a hyperinsulinaemic–hyperglycaemic clamp (HC) with continuous infusion of 240 pmol m−2 min−1 insulin at a steady-state capillary plasma glucose concentration of 7.8 mmol/l (n = 8).
For all clamps, human insulin (Actrapid; Novo Nordisk, Bagsværd, Denmark) and glucose (Serag Wiessner, Naila, Germany) were used. Throughout the clamp, plasma glucose concentrations were monitored every 5 min and used to regulate plasma glucose by the adjustment of a variable infusion of glucose. A deviation of a single capillary glucose concentration of >10% during assumed steady-state conditions was defined as non-steady state. In the hyperinsulinaemic–hyperglycaemic clamp, 150 mg diazoxide (Proglycem; Schering, Berlin, Germany) was given twice (at −60 and 120 min of the experiment) in combination with constant, continued insulin infusion for the suppression of endogenous insulin secretion. Deviation of the plasma insulin concentration was <10% during the steady state. All tests were conducted for 240 min. Subcutaneous adipose tissue biopsy samples were taken 40 min before the start of insulin/placebo infusion and at the end (240 min) of clamp tests. After removal, tissue samples were frozen in liquid nitrogen and stored at −80°C until total RNA isolation.
All values are expressed as mean ± SD. Differences between means were analysed using two-tailed Student’s t test. A p value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS 14.0 (SPSS, Chicago, IL, USA).
IDE expression in HepG2 cells in the presence of different insulin concentrations
IDE expression in HepG2 cells under normal and high glucose concentrations
We compared IDE mRNA and protein levels in HepG2 cells in normal (1 g/l) or high (4.5 g/l) concentrations of d-glucose. Mannitol treatment, used as an osmotic control, induced no changes in IDE expression (data not shown). High glucose concentrations alone induced no effects on the mRNA level of IDE, while 10 nmol/l insulin induced a significant increase in IDE mRNA in a high glucose concentration, in comparison with untreated cells (Fig. 1b). Western blotting of total protein extracts revealed no significant effects of glucose on IDE protein level, with a trend to increased IDE protein levels in the high glucose condition (Fig. 1d). IDE protein levels in cytosolic and membrane fraction of cell samples also did not exhibit significant alterations after insulin or/and high glucose treatments (data not shown).
High glucose inhibits the effect of insulin on IDE activity in HepG2 cells
Absolute IDE activity detected in HepG2 cell samples
Absolute IDE activity
1 g/l glucose medium
4.5 g/l glucose medium
Insulin 10 nmol/l
Insulin 10 nmol/l
81.0 ± 70.9
136.2 ± 91.0
82.1 ± 56.3
95.3 ± 63.6
125.9 ± 66.2
216.1 ± 85.6
157.3 ± 16.7
161.1 ± 29.2
300.0 ± 82.2
247.6 ± 82.5
265.6 ± 24.5
391.3 ± 65.2
Search for alternative IDE transcripts of different sizes in insulin- and glucose-treated cells
Insulin treatment increases the 15a/15b IDE isoform ratio
IDE mRNA expression in human subcutaneous adipose tissue after the clamps
To study the regulation of IDE expression by insulin and glucose in vivo in humans, we performed qRT-PCR in cDNA from subcutaneous adipose tissue taken before (−40 min) and after (240 min) the clamp procedures. Seventeen healthy obese non-diabetic men (age 47.4 ± 8.4 years, BMI 32.5 ± 2.2 kg/m2, waist circumference 110.5 ± 7.1 cm) participated in clamp experiments.
IDE mRNA levels were highly variable among the individuals studied (22.5 ± 13.5%). No correlations between IDE mRNA levels in subcutaneous adipose tissue and anthropometric data (age, BMI, waist circumference, percentage of body fat) and basal and steady-state insulin and glucose concentrations were detected (data not shown).
The regulation of IDE activity in insulin-sensitive tissue in health and type 2 diabetes is very poorly understood. We performed a detailed analysis of the regulation of IDE function by insulin and glucose in a human hepatoma cell line because data for liver cells are controversial. Our data provide the first evidence of increased IDE activity in hepatoma cells after insulin treatment in normal glucose conditions. Neither insulin nor glucose treatment altered IDE mRNA or protein levels in HepG2 cells at normal glucose concentration. At high glucose concentration, an insulin-induced change in IDE activity was not detectable, but we observed a trend to upregulation of IDE protein levels. Moreover, in normal and high glucose medium, insulin increased levels of the catalytically more active 15a IDE isoform compared with the 15b IDE isoform. To validate mechanisms of IDE regulation in vivo, we analysed IDE mRNA levels in biopsies of subcutaneous adipose tissue under conditions of controlled hyperinsulinaemia combined with euglycaemia and hyperglycaemia. We detected a trend to increased IDE expression in subcutaneous adipose tissue during clamp experiments, which did not reach statistical significance.
The observed increase in IDE activity after insulin treatment in human hepatoma cells under normal glucose concentration obviously switches off the action of insulin: insulin induces an increase in IDE activity which leads to increased insulin degradation and decreased insulin signalling. However, we detected no regulation of IDE mRNA and protein levels by insulin at normal glucose concentration. The binding of some peptide substrates to one of the IDE subunits is known to lead to allosteric regulation of IDE activity, with the induction of a shift of the IDE dimer/tetramer equilibrium to the more active dimer and activation of the adjacent subunit . However, insulin binding was shown to induce no activation of IDE, probably because insulin, which is a dimer of A and B chains, can simultaneously bind to both subunits of the IDE dimer. Thus, the effect observed in our experiments cannot be explained by this mechanism.
Under conditions of high glucose, we observed a loss of insulin-induced changes in IDE activity accompanied by an increase in IDE gene expression. We were able to detect an insulin-induced increase in IDE mRNA level and a trend to increased IDE protein level. This observation is supported by data from a hyperglycaemic–hyperinsulinaemic clamp study, in which there were similar changes in IDE mRNA levels in subcutaneous fat tissue in vivo. It is possible that the deficit in IDE activity at a high glucose concentration may lead to a compensatory increase in IDE expression. A study carried out in primary hippocampal neuron culture suggested that insulin may induce the increase in IDE protein level through the phosphatidylinositol-3 kinase pathway; however, the glucose concentration in the culture medium was not stated .
Our findings suggest that hyperglycaemia itself provokes the known disturbance of IDE activity in type 2 diabetes. Results of analyses of IDE activity in biological fluids (blood cells, plasma, wound fluid, cerebrospinal fluid) of diabetic patients reported in the literature are controversial and apparently depend on a range of factors, including study design, the type of diabetes mellitus and the type of treatment for diabetes. One of the studies showed that IDE activity of erythrocytes was increased in patients with type 2 diabetes taking sulfonylureas, in a subgroup with well-controlled type 2 diabetes, and in patients with secondary failure of response to oral therapy, but it was unmodified in well-controlled type 1 diabetic patients . In another study an increase in IDE activity in plasma and erythrocytes was demonstrated both in insulin-dependent and in non-insulin-dependent diabetic patients . In rodents, a decrease in hepatic insulin degradation was observed in diabetic rats, which was restored to near normal levels following insulin treatment [25, 33].
Results of the present study suggest that the increase in IDE activity after insulin treatment and the disturbance of this regulation at a high glucose concentration cannot be explained by observed changes in IDE gene expression. To study other possible mechanisms of alteration of IDE activity under high glucose, we investigated the expression of alternative IDE isoforms in cell samples. No alternative splicing forms of IDE of different sizes were detected after insulin and/or glucose treatment. Furthermore, insulin significantly increased the level of the more catalytically active 15a IDE isoform in comparison with the 15b IDE isoform in both normal and high glucose conditions. Thus, the regulation of the 15a/15b isoform ratio may be one of the mechanisms of the increase in IDE activity after insulin treatment in normal glucose conditions. However, under high glucose conditions, loss of the effect of insulin on IDE activity was observed in the presence of an increased 15a/15b isoform ratio. Hence, the cause of this effect remains unidentified. Incubation of HepG2 cells in a high glucose concentration was shown to induce insulin resistance , which may result in the observed disturbance of regulation of IDE activity by insulin.
A range of chemical substances are known to inhibit IDE activity (chelators, divalent cations, insulin-binding inhibitors and thiol-blocking agents) . The IDE activity in cells may also be regulated by non-esterified fatty acids , nucleotide triphosphates  and post-translational modification . In particular, ATP induces conformational and aggregation changes in IDE and inhibits insulin degradation in vitro . Therefore, the increase in intracellular ATP in high glucose conditions might result in inhibition of the insulin-degrading activity of IDE. Moreover, IDE-interacting proteins might play a role in the disturbance of regulation of IDE activity in high glucose conditions. The recently described IDE interaction with the mitochondrial protein SIRT4 can alter IDE protease activity by ADP-ribosylation . IDE participates in a functional interaction with the 26 S proteasome, which is regarded as the principal site of ubiquitin-dependent intracellular protein degradation . Two IDE-interacting proteins (14 and 6 kDa) were described to inhibit its insulin-degrading activity in rodents, and one of them was identified as ubiquitin [40, 41].
Interestingly, functional changes in IDE were recently described in people with late-onset Alzheimer’s disease. Kim et al.  found reduced IDE activity, but unaltered IDE expression, in Alzheimer’s disease patients from chromosome 10-linked families, suggesting the possibility of systemic functional defects in IDE activity in these families. Type 2 diabetes and hyperinsulinaemia are known to increase the risk of Alzheimer’s disease developing in the elderly . Late-onset Alzheimer’s disease is also an IDE gene-associated disease, and IDE is being the possible link between the pathogenesis of type 2 diabetes and that of late-onset Alzheimer’s disease . Therefore, IDE is an attractive drug target for the treatment of type 2 diabetes and Alzheimer’s disease. Shen et al.  described high-resolution crystal structures of IDE, and these authors considered that this opened the door to the design of pharmacological modulators of IDE activity. The disturbance of insulin-induced regulation of IDE activity in high glucose conditions revealed in present study confirms the need to develop IDE activators, as advocated in a recent paper in Nature .
In summary, we demonstrated the regulation of IDE activity by insulin in a liver cell model and the loss of this regulation in high glucose conditions. This disturbance cannot be explained by corresponding alterations in IDE protein expression or IDE splicing. Moreover, we detected an insulin-induced increase in IDE mRNA levels and a trend to increased IDE protein level under high glucose conditions. Similar changes in IDE expression in subcutaneous fat tissue in vivo in hyperglycaemic–hyperinsulinaemic clamp experiments were observed. The loss of insulin-induced regulation of IDE activity under high glucose conditions may contribute to reduced insulin extraction and peripheral hyperinsulinaemia in type 2 diabetes.
We thank all study participants for their cooperation. We gratefully acknowledge the technical assistance of A. Wagner, S. Grosch, A. Ziegenhorn and K. Sprengel. We thank K. Wagner and A. Teichmann for help in the measurement of IDE activity. This study was supported by a grant from the German Research Foundation (DFG Grant No. Pf164/021002; N. Rudovich, Ö. Gögebakan and A. F. H. Pfeiffer) and by the German Academic Exchange Service (O. Pivovarova).
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
- 2.Rudovich NN, Rochlitz HJ, Pfeiffer AFHP (2007) Reduced hepatic insulin extraction in response to gastric inhibitory polypeptide compensates for reduced insulin secretion in normal-weight and normal glucose tolerant first-degree relatives of type 2 diabetic patients. Diabetes 53:2359–2365CrossRefGoogle Scholar
- 37.Del Carmen Camberos M, Cresto JC (2007) Insulin-degrading enzyme hydrolyzes ATP. Exp Biol Med (Maywood) 232:281–292Google Scholar