OUP user menu

Initial steps of insulin signaling and glucose transport are defective in the type 2 diabetic rat heart

Martine Desrois, Robert J Sidell, Dominique Gauguier, Linda M King, George K Radda, Kieran Clarke
DOI: http://dx.doi.org/10.1016/j.cardiores.2003.11.021 288-296 First published online: 1 February 2004

Abstract

Objective: Whole body insulin resistance and diabetes are risk factors for cardiovascular diseases, yet little is known about insulin resistance in the diabetic heart. The aim of this work was to define the insulin response in hearts of the Goto–Kakizaki (GK) rat, a polygenic model of spontaneous type 2 diabetes. Methods: We measured d[2-3H]glucose uptake before and after insulin stimulation, plus initial steps of the insulin signaling pathway after insulin infusion via the caudal vena cava in hearts from the male Wistar and spontaneously diabetic GK rats. Results: Despite normal basal d[2-3H]glucose uptake, insulin-stimulated glucose uptake was 50% (p<0.03) lower in GK rat hearts compared with their Wistar controls. Total GLUT4 protein was depleted by 28% (p<0.01) in GK rat hearts. We found 31% (p<0.0001) and 38% (p<0.001) decreased protein levels of insulin receptor β (IRβ)-subunit and insulin receptor substrate-1 (IRS-1), respectively, in GK rat hearts with 37% (p<0.02) and 45% (p<0.01) lower insulin-stimulated tyrosine phosphorylation of these proteins. Owing to the decreased IRS-1 protein levels, GK rat hearts had a 41% (p<0.0001) decrease in insulin-stimulated IRS-1 protein association with the p85 subunit of phosphatidylinositol 3-kinase, despite normal phosphatidylinositol 3-kinase protein expression. Insulin-stimulated serine phosphorylation of protein kinase B was the same in all hearts, as was protein kinase B expression. Conclusion: We conclude that decreased insulin receptor β, IRS-1 and GLUT4 proteins are associated with insulin resistance in type 2 diabetic rat hearts.

Keywords
  • Diabetes
  • Signal transduction
  • Protein phosphorylation
  • Receptors
  • Membrane transport

1. Introduction

Worldwide there are close to 150 million people with diabetes mellitus [1]. Despite cardiovascular disease being the leading cause of death in diabetic patients [1], most studies of insulin signaling in type 2 diabetes are of skeletal muscle or adipose tissue, with very few heart studies. Yet type 2 diabetic patients have 2-fold higher mortality after myocardial infarction and a 2.4-fold increased risk of congestive heart failure than non-diabetic patients [1], with the relation between whole body insulin resistance and cardiovascular risk well recognized [2].

Type 2 diabetes is a complex disorder, being influenced by a combination of environmental and genetic factors, yet identification of the genes responsible for the development of the disease has proven difficult, there being more than 60 potential genes for diabetes susceptibility in humans [3]. Consequently, polygenic animal models of type 2 diabetes have proven invaluable for candidate gene identification. For example, at least six independent genetic loci are responsible for defects in glucose and insulin metabolism in the Goto–Kakizaki (GK) rat [4], a highly inbred strain derived from outbred, glucose intolerant Wistar rats, that spontaneously develops type 2 diabetes within the first few weeks of age [5]. GK rats exhibit mild hyperglycaemia and hyperinsulinaemia, impaired glucose-induced insulin secretion, marked glucose intolerance, hepatic glucose overproduction and peripheral insulin resistance [4,5]. Glucose transport and insulin signaling are impaired in isolated GK rat adipocytes [6] and skeletal muscle [7], but effects on the heart have not been reported to date.

Positron emission tomographic (PET) studies have shown that type 2 diabetic patients have either lower [8] or normal [9] myocardial fluorine-18 fluorodeoxyglucose (FDG) uptake during insulin infusion, although it is unknown whether decreased insulin-stimulated glucose uptake is caused by defects in insulin signaling and/or glucose transport. Because alterations in the insulin response may contribute to myocardial dysfunction, particularly during ischaemia, we investigated, for the first time to our knowledge, myocardial insulin resistance and signaling in a polygenic model of type 2 diabetes, the GK rat. A preliminary report of this work has been presented in abstract form [10].

2. Methods

2.1. Materials and antibodies

Insulin receptor β (IRβ), insulin receptor substrate-1 (IRS-1), horseradish peroxidase-labeled anti-rabbit, anti-mouse, anti-sheep IgG antibodies and Protein A-agarose were obtained from Santa Cruz Biotechnology (Autogen Bioclear, UK). Anti-phosphotyrosine antibody (clone 4G10) was from Upstate Biotechnology (Lake Placid, NY). PKB expression and phosphorylation were determined using anti-PKB and anti-phospho-PKB (Ser473) rabbit antibodies, respectively (New England Biolabs, UK). Anti-GLUT4 rabbit antibody was a kind gift from Prof. G.D. Holman (Bath University, UK). Anti-GLUT1 antibody was from Biogenesis (Biogenesis, UK). The enhanced chemiluminescence (ECL) detection system was from Amersham (Pharmacia Biotech, UK). Plasma glucose was measured using an assay kit (Sigma).

2.2. Animals

Age-matched, male control Wistar (n = 20) or GK rats (n = 20) (10 months, 400–550 g), bred at the Wellcome Trust Centre for Human Genetics, University of Oxford, were used. The rats had free access to chow and water. Food was withdrawn 12 h before the experiments. The University of Oxford Animal Ethics Review Committees and the Home Office (London, UK) approved all procedures performed in this study. The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.3. Heart perfusions

Control (n = 6) and GK (n = 6) rats were anaesthetised by intraperitoneal injection of 60 mg/ml sodium pentobarbitone (Sagatal, Rhône Mérieux, Dublin, Ireland). Hearts were quickly excised and arrested in ice-cold heparin-containing Krebs–Henseleit buffer (KHB). Hearts were weighed before cannulation via the ascending aorta for retrograde Langendorff-perfusion at a constant temperature of 37 °C and a constant pressure of 100 mm Hg. Hearts were perfused with a modified, phosphate-free KHB (pH 7.4) containing (in mM) 118, NaCl, 4.7 KCl, 1.2 MgSO4, 1.75 CaCl2, 25 NaHCO3, 0.5 EDTA and 11 glucose, and continuously gassed with a mixture of 95% O2 and 5% CO2. A drainage cannula was inserted into the left ventricle through a left atrial incision to vent the Thebesian flow via the apex.

2.4. Glucose uptake in response to insulin

Glucose uptake was measured as the rate of cleavage of H+ from glucose [11,12]. Hearts were perfused with 250 ml recirculating KHB containing 11 mM glucose with 14.5 μCi/mmol d[2-3H]glucose (Amersham, Amersham, Bucks). Insulin was added to the buffer reservoir after 30 min to a final concentration of 3 U/l to ensure maximal stimulation of glucose transport. Perfusion was continued for another 30 min and buffer samples were taken from the reservoir every 4 min throughout the protocol. The glucose used (μmol/g ww) was plotted against time, and the rates of glucose uptake (μmol/g ww/min), with or without insulin, were calculated.

2.5. GLUT4 and GLUT1 determinations

Frozen cardiac tissue was homogenised in lysis buffer containing 75 mM Tris (pH 6.8), 3.8% SDS, 4 M urea and 20% glycerol. The homogenates were boiled at 98–100 °C (5 min) and centrifuged at 13,000 rpm (5 min). Protein concentrations of the supernatants were determined using an assay kit (Pierce, USA) and bovine serum albumin (BSA) as a standard. Samples were boiled with 5% β-mercaptoethanol (5 min) and stored. For direct immunoblotting, equal amounts of solubilised protein (20–50 μg/lane) were resuspended in Laemmli sample buffer containing 20 mM dithiothreitol (DTT) and separated by 10% SDS-PAGE. The resolved proteins were electrophoretically transferred to nitrocellulose membranes using a transfer buffer (48 mM Tris, 39 mM glycine, 0.0375% SDS and 20% methanol (pH 8.8)) and a semidry transfer apparatus (BioRad, Hercules, CA). We performed Ponceau staining to verify the quality of the transfer and equal amounts of protein in each lane. The nitrocellulose membranes were incubated in TBS-T (0.9% NaCl, 10 mM Tris and 0.1% Tween-20) supplemented with 5% milk to reduce non-specific binding and incubated for 2 h at room temperature with anti-GLUT4 antibody (1/4000) or overnight at 4 °C with anti-GLUT1 antibody (1/1000). After the addition of the secondary antibody, signal was detected by ECL and quantified by densitometry using Qscan 32 image analysis software (Biosoft, USA). A Wistar rat heart homogenate was run as an internal standard for quantification in each assay.

2.6. Insulin signaling

Control (n = 12) and GK (n = 12) rats were anaesthetised as described above. After laparotomy, a 500 μl bolus of human insulin (Velosulin, 10 U/kg body weight, dose producing maximal phosphorylation of myocardial proteins [13]) or vehicle (0.9% NaCl) was injected into the caudal vena cava. After 90 s or 180 s [14], the hearts were rapidly excised, washed in cold phosphate buffered-saline (PBS) and frozen. The tissue was powdered and homogenised [13] (Power Control Unit, Kinematica, Luzern, Switzerland). Protein concentrations were determined as described above.

For direct immunoblotting, equal amounts of solubilised protein (20–50 μg per lane) were resuspended in Laemmli sample buffer containing 100 mM DTT and separated using 6% SDS-PAGE. Blots were probed with antiphosphotyrosine (1 μg/ml), antiinsulin receptor (2 μg/ml) or IRS-1 (1/200) antibodies 2 h at room temperature or overnight at 4 °C with anti-p85 subunit of PI 3-kinase (1 μg/ml) antibody. Proteins were visualised and quantified as described above.

For immunoprecipitation studies, IRβ and IRS-1 antibodies were incubated with Protein-A-agarose beads overnight at 4 °C. The complexes were washed three times with PBS and once with buffer containing (in mM) 100 Tris, 150 NaCl, 10 EDTA, 5 sodium orthovanadate, 2 phenylmethylsulfonyl fluoride (PMSF), 10 sodium pyrophosphate, 10 p-nitrophenylphosphate disodium, 6 μg/ml leupeptin and 0.5% Triton X100. The solubilised proteins (1–5 mg) were added to the antibody/protein-A-agarose complexes for 2 h at 4 °C. The immunocomplexes were centrifuged and washed. The immunoprecipitates were resuspended in Laemmli sample buffer (100 mM DTT), resolved in SDS-PAGE and the tyrosine phosphoproteins were identified by phosphotyrosine antibody immunoblotting.

The association of the p85 subunit of PI 3-kinase with phosphorylated IRS-1 was demonstrated by blotting the nitrocellulose membranes of IRS-1 immunoprecipitates with antip85 antibody [14] with detection as described above.

2.7. PKB level and phosphorylation

To detect both protein levels and serine phosphorylation of PKB, total lysates (100 μg protein/lane) were separated using 8% SDS-PAGE and blots were probed with anti-phospho-PKB (Ser473) or anti-PKB rabbit polyclonal antibodies. PKB levels and phosphorylation were quantified as described above.

2.8. Data analyses

The bands in fluorographs were quantified by densitometry and the results expressed as means±S.E.M. Heart samples from GK and control rats were processed in parallel for each analysis. Statistical significance was evaluated by ANOVA, followed by a modified Student's t test when appropriate. Differences were considered significant at p<0.05.

3. Results

3.1. Physiological characteristics of diabetic rats

The GK rats had a 26% lower (p<0.0001) body weight than the control rats (409±8 vs. 550±15 g), although the heart wet weights were the same (1.8±0.20 and 1.8±0.2 g, respectively). After an overnight fast, serum glucose concentrations were 76% higher (p<0.001) in GK than in control rats (10.2±1.2 and 5.8±0.1 mM, respectively).

3.2. Basal glucose uptake and insulin response

Basal glucose uptake rates were the same in all hearts at 0.56±0.08 and 0.62±0.2 μmol/g ww/min for control and GK rat hearts, respectively (Fig. 1A). The response to insulin in the GK rat hearts was half that of the controls, in that insulin stimulated glucose uptake by 148% in control, but only by 73% (p<0.03) in GK rat hearts. Thus, GK rat hearts were significantly insulin resistant.

Fig. 1

Glucose uptake (A) and GLUT4 (B) and GLUT1 (C) protein contents in control and GK rat hearts. (A) Basal (shaded) and insulin-stimulated (open) glucose uptake rates (μmol/gww/min) were measured as the rate of cleavage of H+ from d[2-3H]glucose in isolated perfused control (n = 6) and GK (n = 6) rat hearts (*p<0.05 vs. -Insulin, vs. control). GLUT4 (B) and GLUT1 (C) protein contents were determined by immunoblotting (see Section 2). Blots show GLUT4 and GLUT1 proteins from a control and a GK rat heart. Equal protein loading is checked using Ponceau red staining. The bar graphs below the blots show results from six control and six GK rat hearts. Data are means ± SEM and are expressed after assigning a value of 100% to the control. *p<0.01 vs. control.

3.3. GLUT4 and GLUT1 protein expression

The GK rat hearts had 28% lower (p<0.01) GLUT4 protein content compared with control rat hearts (Fig. 1B), but GLUT1 expression was the same for all hearts (Fig. 1C).

3.4. Insulin signaling

Finding that the GK rat hearts were insulin resistant, we measured the initial steps of insulin signaling. A time-course experiment was performed and after 90 s of insulin infusion into the caudal vena cava, tyrosine phosphorylation of myocardial proteins was maximal with the phosphorylation of proteins present in at least two bands (170 and 98 kDa) in control rat hearts (data not shown). The identities of the two insulin-stimulated tyrosine phosphoproteins were confirmed by immunoprecipitation with specific antibodies. The 98-kDa band represented the IRβ-subunit and the 170-kDa band corresponded to IRS-1 (Fig. 2). For subsequent studies, 90 s of insulin infusion was used to achieve maximal phosphorylation of myocardial proteins. We found a 31% lower IRβ protein content (Fig. 3A, p<0.0001) in GK compared with control rat hearts; therefore, insulin-stimulated tyrosine phosphorylation of IRβ was 37% lower (Fig. 3B, p<0.02). Insulin receptor phosphorylation was decreased to the same extent as the decreased receptor protein expression in the GK and the control rat hearts, suggesting that the decrease in tyrosine-phosphorylated insulin receptor may have been a consequence of decreased insulin receptor protein (Fig. 3C). We observed a 38% decrease in IRS-1 protein expression in GK compared with control rat hearts (Fig. 4A, *p<0.001), and insulin-stimulated tyrosine phosphorylation of IRS-1 was 45% lower (Fig. 4B, *p<0.01). The phosphorylation/IRS-1 protein ratio was not significantly different between the GK and control rat hearts, showing that the decrease in IRS-1 protein was responsible for the decrease in insulin-stimulated tyrosine phosphorylation of IRS-1 (Fig. 4C).

Fig. 4

IRS-1 protein content (A) and tyrosine phosphorylation (B) in rat hearts. Data are means±S.E.M. and expressed as arbitrary units. (A) IRS-1 protein content in control (left) and GK (right) rats with or without insulin. The bar graph is data from six control and six GK rat hearts. Insulin did not change IRS-1 level in any hearts but lower IRS-1 protein content was found in GK compared with control (*p<0.001). (B) Hearts from one control and one GK were processed in the same experiments and separated on the same gel. Results are expressed relative to phosphotyrosine content under basal condition (given an arbitrary value of 100). Basal IRS-1 tyrosine phosphorylation was similar in both groups but insulin-stimulated IRS-1 phosphorylation was lower in GK compared with control (p<0.01). (C) Ratio of IRS-1 tyrosine phosphorylation per IRS-1 protein was calculated from insulin-stimulated tyrosine phosphorylation and protein contents.

Fig. 3

IRβ protein content (A) and tyrosine phosphorylation (B) in rat hearts. Data are means±S.E.M. and expressed as arbitrary units. (A) IRβ protein content with or without insulin from control (left) and GK (right) rats. The bar graph is data from six control and six GK rats. Insulin did not change IRβ level in both groups. Lower IRβ protein content was found in GK compared with control (*p<0.0001). (B) Hearts from one control and one GK were processed in the same experiments and separated on the same gel. Results are expressed relative to phosphotyrosine content under basal condition (given an arbitrary value of 100). Basal IRβ tyrosine phosphorylation was similar in both groups but insulin-stimulated IRβ phosphorylation was lower in GK compared with control (p<0.02). (C) Ratio of Insulin receptor tyrosine phosphorylation and the insulin receptor protein. The ratio was the same for all rat hearts, suggesting that phosphorylation was limited by content.

Fig. 2

Identification of insulin-stimulated tyrosine phosphoproteins in rat hearts. After 90 s insulin, solubilized proteins from frozen tissue were immunoprecipited with anti-IRβ (left) or anti-IRS-1 (right) antibodies and then immunoblotted with antiphosphotyrosine antibody. Molecular mass standards were depicted in the margin. Positions of IRβ and IRS-1 were indicated by arrows.

3.5. Insulin-stimulated IRS-1-associated PI 3-kinase

After insulin stimulation, the p85 subunit of PI 3-kinase that immunoprecipitated with phosphorylated IRS-1 was 41% lower (p<0.0001) in GK compared with control rat hearts, correlating with the 45% lower insulin-stimulated phosphorylation of IRS-1 (see above). Total protein levels of the p85 subunit of PI 3-kinase were similar in all hearts (Fig. 5B).

Fig. 5

IRS-1-associated PI 3-kinase (A), p85 (PI 3-kinase) protein content (B), PKB protein content (C) and serine-phosphorylation (D) in rat hearts. Data are means±S.E.M. and expressed as arbitrary units. (A) IRS-1-associated PI 3-kinase with or without insulin from control (left) and GK (right) rat hearts. Hearts from one control and one GK were processed in the same experiments and separated on the same gel. Results are expressed relative to phosphotyrosine content under basal condition (given an arbitrary value of 100). Basal IRS-1-associated PI 3-kinase was similar in both groups but lower insulin-stimulated IRS-1-associated PI 3-kinase was found in GK compared with control (*p<0.0001 vs control). The bar graph is from six control and six GK rats. (B) p85 (PI 3-kinase) protein content with or without insulin from control (left) and GK (right) rats. The bar graph is from six control and six GK rats. There was no difference in p85 (PI 3-kinase) protein content in control and GK rat hearts. (C) PKB protein content with or without insulin from control (left) and GK (right) rats. The bar graph is from six control and six GK rats. There was no difference in PKB protein content in control and GK rat hearts. (D) PKB serine phosphorylation with or without insulin from control (left) and GK (right) rats. Hearts from one control and one GK were processed in the same experiments and separated on the same gel. Results are expressed relative to phosphoserine content under basal condition (given an arbitrary value of 100). Basal and insulin-stimulated serine phosphorylation of PKB were similar in control and GK rat hearts.

3.6. PKB level and phosphorylation

PKB protein levels were the same in all hearts (Fig. 5C). Insulin increased the serine phosphorylation of PKB by 99±12% and 102±16% in control and GK rat hearts without significant differences between groups (Fig. 5D).

4. Discussion

Although diabetic patients are two to four times more likely to have heart disease than the normal population [2], the cardiac response to insulin stimulation remains an understudied complication of diabetes. Here, we examined hearts from a non-obese, polygenic rat model of type 2 diabetes, the GK rat, and found a significant loss of GLUT4, insulin receptor β-subunit and IRS-1 proteins with decreased insulin-stimulated phosphorylation of both the insulin receptor β-subunit and IRS-1, which significantly decreased the association between phosphorylated IRS-1 and PI 3-kinase, despite normal PI 3-kinase protein expression. We suggest that the loss of insulin receptor β, IRS-1 and GLUT4 proteins underlies insulin resistance in the GK diabetic rat heart.

We found that GK rat hearts had normal basal glucose uptake rates, but 50% lower insulin-stimulated glucose uptake than control rat hearts, confirming reports of low insulin-stimulated glucose uptake in type 2 diabetic patient heart [8] and skeletal muscle [15]. We found the same total GLUT1 protein levels in all hearts, which explained their similar basal glucose uptake rates. However, total GLUT4 protein expression was decreased by 28% in GK rat heart, similar to the 20–25% decrease in the obese Zucker rat heart [12,16]. The lower GLUT4 protein content, as found in skeletal muscle of type 2 diabetic patients [17], could have contributed to the insulin resistance in GK rat heart, but was probably not the only cellular change involved.

There are few studies of insulin signaling in the type 2 diabetic heart. Eckel et al. [18] have reported reduced insulin sensitivity in diabetic BB rat cardiocytes and have suggested the existence of postreceptor defects. Abnormalities have been reported in the obese Zucker rat that differ considerably from those found here in the GK rat. Unlike the GK rat heart, isolated myocytes from 3 to 4 month old obese Zucker rats have normal tyrosine phosphorylation of the insulin receptor β-subunit and IRS-1, a 370% increased serine and/or threonine phosphorylation of IRS-1 with a 30–40% reduction in IRS-1 abundance, and blunted IRS-1-associated PI 3-kinase activity after insulin stimulation [19]. In contrast to isolated myocytes, intact obese Zucker rat hearts become insulin resistant only after 6 months of age, but have normal insulin receptor and IRS-1 protein levels [12]. In short, although Zucker and GK rats are considered to be models of type 2 diabetes, there are strain differences in myocardial insulin signaling that arise because the Zucker rat was bred for an obese phenotype that resulted from a spontaneous mutation of the leptin receptor gene [20], the insulin resistance being secondary to the morbid obesity [21], whereas the GK rat was bred for glucose intolerance [5].

GK rat skeletal muscle, liver and adipose tissues are insulin resistant [5–7,22], and here we have demonstrated that the GK rat heart also is insulin resistant. Although decreased IRS-1 tyrosine phosphorylation and diminished activation of IRS-1-dependent PI 3-kinase occurred in soleus, but not extensor digitorum longus (EDL) muscle from GK rats, levels of insulin receptor, IRS-1 and GLUT4 proteins were normal in soleus muscle [7]. In GK rat heart, we found 31% and 38% decreased protein levels of insulin receptor β-subunit and IRS-1, respectively, with 37% and 45% lower insulin-stimulated tyrosine phosphorylation of these proteins. Thus the decrease in protein content could have accounted for the decrease in insulin-stimulated tyrosine phosphorylation of insulin receptor β-subunit and IRS-1, as reported for insulin resistant human skeletal muscle [23]. Similarly decreased IRS-1 protein expression and impaired insulin receptor and IRS-1 tyrosine phosphorylation with lower IRS protein-associated PI 3-kinase activity have been reported for muscle, liver and adipose tissues of obese Zucker rats [24] and for insulin resistant human skeletal muscle [15,23,25]. Thus, the 38% loss of IRS-1 protein led to 41% lower association between phosphorylated IRS-1 and the p85 subunit of PI 3-kinase in GK rat hearts, which correlated with the 50% lower insulin-stimulated glucose uptake. Although we did not measure PI 3-kinase activity, many studies have shown that phosphorylated IRS-1 associated with the p85 subunit of PI 3-kinase accurately reflects PI 3-kinase activity in normal and insulin resistant tissues [6,19,25–27].

The decrease in the insulin receptor and IRS-1 proteins in the GK rat heart may be related to the high circulating insulin level [5], which caused internalization and degradation of the receptor in adipose tissue and liver [28,29]. Prolonged insulin treatment of adipocytes or hepatoma cells stimulated the disengagement of the insulin receptor from IRS proteins, making them accessible to proteases [28,30]. Thus, it is possible that the insulin resistance found in the GK rat heart was due to an insulin-induced degradation of insulin receptor and IRS proteins.

As reported for adipocytes [6] and human type 2 diabetic skeletal muscle [25], we found that neither PKB protein expression nor insulin-stimulated PKB serine phosphorylation were abnormal in the insulin-stimulated GK rat heart, despite decreased association between phosphorylated IRS-1 and PI 3-kinase and decreased glucose uptake. Although we did not measure PKB activity, others have shown myocardial PKB to be maximally activated within 2 min of insulin stimulation, with PKB serine phosphorylation accurately reflecting PKB activity [27]. In contrast, GK rat soleus and EDL muscles have decreased insulin-stimulated PKB activity [7,22], although PKB activity was decreased in EDL muscle only during maximal insulin stimulation when glucose transport, IRS-1 tyrosine phosphorylation and IRS-1-associated PI 3-kinase activity were normal [7]. Other studies have shown PKB activation by insulin to be impaired in adipocytes [31] and skeletal muscle [32] from type 2 diabetic patients, obese Zucker rats [24] and hyperglycaemic rats [33], although not necessarily correlating with diminished PI 3-kinase activity [33] or decreased glucose uptake [32]. Thus, the role of PKB in insulin resistance is unclear, but may involve tissue-specific changes in expression and activation [24] with only relatively small activation of PI 3-kinase perhaps necessary for the full expression of downstream signaling [34,35]. Alternatively, another signaling molecule or pathway, independent of PKB, may have decreased insulin-stimulated glucose uptake in the GK rat hearts, for example decreased PI 3-kinase p85α subunit recruitment to GLUT4 vesicles has been reported in obese Zucker rat cardiomyocytes [26]. In addition, the role of atypical protein kinase C (PKC) isoforms λ/ζdelta has been recently reported in the pathogenesis of insulin resistance in muscle of type 2 diabetic patients [36]. Interestingly, atypical PKCs (λ/ζ) are activated by insulin in a PI 3-kinase dependent manner and have been suggested to be involved in insulin-mediated glucose uptake [36,37]. Impaired insulin-stimulated PKC λ/ζ activity associated with defective insulin-induced activation of IRS-1-dependent PI 3-kinase and normal PKB activity have been shown in skeletal muscle and adipose tissues of GK rats [38]. In skeletal muscle of type 2 diabetic subjects, Kim et al. [36] reported decreased glucose disposal, impaired insulin-stimulated IRS-1 phosphorylation, PI 3-kinase and PKC λ/ζ activity but normal PKB activation. Decreased PKC activation most likely reflected impaired activation of IRS-1-dependent PI 3 kinase and decreased ability of lipid product, PI-3,4,5-(PO4)3 to directly activate PKC [39].

Thus, insulin resistance in the GK rat heart may be due, at least in part, to the decreased association between phosphorylated IRS-1 and the p85 subunit of PI 3-kinase caused by decreased insulin receptor and IRS-1 protein levels.

As has been shown previously in the rat heart, whether in vivo [14], isolated and perfused in vitro or in isolated cardiomyocytes [27], maximal phosphorylation of myocardial proteins occurred within 90 s after stimulation with a bolus of insulin, the two proteins phosphorylated being the insulin receptor β-subunit and IRS-1. In this and other studies, changes in the insulin signaling pathway were investigated in the in vivo heart after insulin administration [14], whereas insulin-stimulated glucose uptake rates were measured in the isolated, perfused rat heart using well-established methodology [11,12,27]. As with other tissues, myocardial glucose uptake rates are difficult to measure in vivo and there is no reason to believe that in vivo insulin signaling would be different to that in vitro. The timing differences of 90 s for measuring signaling after an insulin bolus, and 30 min for glucose uptake arose from the different techniques required for the measurements and are similar to those in other studies in which both insulin signaling and glucose uptake have been measured [6,24,27].

Studies using gene knockout mice have shown that the insulin signaling abnormalities observed here could account for the principal metabolic changes in type 2 diabetic heart. Selective ablation of insulin receptors in the β cell resulted in a blunted response to glucose [40] and ablation of IRS-1 retarded growth and caused mild insulin resistance [41]. Independent 50% reductions in insulin receptor or IRS-1 in transgenic heterozygous mice did not cause insulin resistance, yet mice with heterozygous mutations in the insulin receptor and IRS-1 developed severe insulin resistance in skeletal muscle and liver with compensatory β cell hyperplasia [42]. Although defects in IRS-2 [25] and/or the p85α and p85β regulatory subunit isoforms of PI 3-kinase [26] may contribute to insulin resistance in the type 2 diabetic heart, we have not measured IRS-2 and did not distinguish between the p85 isoforms. Atypical protein kinase C isoforms may be also involved in the pathogenesis of insulin resistance [36,39] but it is not possible to gauge the importance of alternative signals downstream of PI3K from our work. Yet the changes we have observed in the GK type 2 diabetic rat heart have also been found in skeletal muscle from type 2 diabetic patients [15,25].

In conclusion, we have shown impaired insulin-stimulated glucose transport and decreased insulin receptor β-subunit, IRS-1 and GLUT4 protein levels in the type 2 diabetic GK rat heart. Lower insulin receptor β-subunit and IRS-1 proteins were associated with the decreased activation of insulin signaling and, with the loss of GLUT4 protein, may have impaired insulin-stimulated glucose uptake in the GK rat heart. Protein levels and insulin-stimulated PKB phosphorylation were normal in the GK rat heart, suggesting that PKB may not be a controlling factor in this model of cardiac insulin resistance. Defects in myocardial insulin response may have deleterious effects on glucose utilization and may underlie the metabolic abnormalities observed in type 2 diabetic heart.

Acknowledgements

We thank Yvonne Green for her expert technical assistance, Prof G.D. Holman (Bath University) for the kind gift of anti-GLUT4 antibody and Dr F. Ouali (Paris 7 University) for the serum glucose assays. This study was supported by grants from the British Heart Foundation (Martine Desrois, Kieran Clarke) and the Wellcome Trust (Robert Sidell, Linda King). Dominique Gauguier holds a Wellcome Trust Senior Fellowship in basic biomedical science.

Footnotes

  • Time for primary review 00 days

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
View Abstract