Diabetic db/db mice exhibit profound insulin resistance in vivo, but the specific degree of cardiac insensitivity to insulin has not been assessed. Therefore, the effect of insulin on cardiomyocytes from db/db hearts was assessed by measuring two metabolic responses (deoxyglucose uptake and fatty acid oxidation) and the phosphorylation of two enzymes in the insulin-signaling cascade [Akt and AMP-activated protein kinase (AMPK)]. Maximal insulin-stimulated deoxyglucose transport was reduced to 58 and 40% of control in cardiomyocytes from db/db mice at two ages (6 and 12 wk). Insulin-stimulated deoxyglucose uptake was also reduced in myocytes from transgenic db/db mice overexpressing the insulin-sensitive glucose transporter (db/db-hGLUT4). Treatment of db/db mice for 1 wk with an insulin-sensitizing peroxisome proliferator-activated receptor-γ agonist (COOH) completely normalized insulin-stimulated deoxyglucose uptake. Insulin had no direct effect on palmitate oxidation by either control or db/db cardiomyocytes, but the combination of insulin and glucose reduced palmitate oxidation, likely an indirect effect secondary to increased glucose uptake. Insulin had no effect on AMPK phosphorylation from either control or db/db cardiomyocytes. Insulin increased the phosphorylation of Akt in all cardiomyocyte preparations (control, db/db, COOH-treated db/db) to the same extent. Thus insulin has selective metabolic actions in mouse cardiomyocytes; deoxyglucose uptake and Akt phosphorylation are increased, but fatty acid oxidation and AMPK phosphorylation are unchanged. Insulin resistance in db/db cardiomyocytes is manifested by reduced insulin-stimulated deoxyglucose uptake.
- cardiac metabolism
- glucose uptake
- fatty acid oxidation
diabetic db/db mice provide a monogenic model of obesity and type 2 diabetes (13, 24). Insulin resistance is the earliest phenotypic change in db/db mice, evident at 10–12 days of age (14). At 8–12 wk, diabetic db/db mice exhibit glucose intolerance in response to an oral glucose challenge (18) and a reduced hypoglycemic response to a bolus injection of insulin (21) compared with nondiabetic control db/+ mice. Severe insulin resistance is also observed with hyperglycemic hyperinsulinemic clamps (9). Recently, Carley et al. (10) reported that chronic (6 wk) oral administration of an insulin-sensitizing peroxisome proliferator-activated receptor-γ (PPARγ) agonist (COOH) to db/db mice improved diabetic status by normalizing hyperglycemia.
On the basis of glucose homeostasis, skeletal muscle is usually the dominant organ responsible for in vivo insulin resistance (34). However, mechanisms of insulin resistance can be tissue specific. In humans with type 2 diabetes, heart glucose uptake is insulin sensitive despite insulin resistance in skeletal muscle (20, 35).
Insulin has a number of acute metabolic actions on the heart (8): stimulation of glucose uptake, glycogen synthesis, glycolysis and glucose oxidation, and inhibition of fatty acid (FA) oxidation. Insulin stimulated glycolytic rates in perfused hearts from control mice (6, 23, 27) and increased glucose uptake into control cardiomyocytes (4, 10, 27). The metabolic phenotype of db/db hearts has been assessed with ex vivo perfusions containing glucose and palmitate as exogenous substrates; glucose utilization was reduced, and FA utilization was enhanced (1, 5, 10, 29). However, the effect of insulin on the metabolism of db/db hearts has not been examined.
Akt is a serine/threonine protein kinase that is activated by insulin via a phosphatidylinositol 3-kinase-dependent pathway (7). Akt activation has been implicated in regulation of insulin-stimulated glucose uptake, and Akt2-null mice are insulin resistant (17). AMP-activated protein kinase (AMPK) stimulates both glucose utilization and FA oxidation (11). Recently, Kovacic et al. (23) have reported that Akt can inactivate AMPK in mouse hearts. Thus inhibition of AMPK by insulin may be a contributory mechanism to the observation that cardiac FA oxidation is inhibited by insulin (16, 23).
Previously, we reported (10) that insulin-stimulated glucose uptake was reduced in cardiomyocytes from db/db mice at 12 wk of age, using a single 10 nM insulin concentration. The objectives of this investigation were as follows: first, to extend our preliminary description of insulin resistance in db/db cardiomyocytes (10) by examining the ability of a range of insulin concentrations to increase glucose uptake into cardiomyocytes from db/db mice at two ages, corresponding to early (6 wk) and established (12 wk) stages in their diabetic phenotype (1, 13); second, to examine insulin-stimulated glucose uptake into cardiomyocytes from transgenic db/db-hGLUT4 mice overexpressing insulin-regulatable GLUT4 glucose transporters (5, 18); third, to further assess cardiac insulin resistance in db/db cardiomyocytes by determining the effect of insulin on a different metabolic response, FA oxidation; and fourth, to study the ability of insulin to regulate two enzymes (Akt and AMPK) in the insulin-signaling cascade (7, 11, 32) by use of cardiomyocytes from control db/+ and db/db mice as well as db/db mice treated with the insulin-sensitizing PPARγ agonist COOH (10).
Male C57BL/KsJ-leprdb/leprdb (db/db) and lean control heterozygote (db/+) mice were purchased from Jackson Laboratories (Bar Harbor, ME). Transgenic mice with overexpression of the human insulin-regulated glucose transporter (db/+-hGLUT4 and db/db-hGLUT4) were raised at the University of Calgary (5, 6). Animals were given ad libitum access to food and water and were housed under a 12:12-h light-dark cycle. All experiments were approved by the University of Calgary Animal Welfare Committee. Experiments were conducted with db/db mice at 6–8 and 12–15 wk of age, as noted, reflecting early and established stages in the age-dependent diabetic progression of this monogenic model of obesity and insulin resistance (1, 13).
In some experiments, db/db mice (11 wk of age) were treated for 1 wk with COOH, a nonthiazolidinedione PPARγ agonist (10). Preliminary experiments established that insulin sensitization (normalization of hyperglycemia in untreated db/db mice) could be achieved after only 1 wk of treatment compared with the chronic (6 wk) oral protocol utilized previously (10). COOH (from Merck Research Laboratories, Rahway, NJ) was administered as a food admixture by addition of COOH to powdered chow (ProLab RMH 2500/5P14; PMI International, Brentwood, MO) to give a daily dosage of 30 mg/kg body wt.
To assess diabetic status, plasma glucose concentration was measured as described by Carley et al. (10). Because mice were heparinized before heart perfusions to isolate cardiomyocytes or to assess cardiac function in ex vivo-perfused hearts, plasma lipids were not measured routinely, because heparin treatment produces artifactual changes in lipid concentrations due to the release of lipoprotein lipase into the circulation.
Isolation of cardiomyocytes.
Mouse ventricular cardiomyocytes were prepared essentially as described by Carley et al. (10). Mice (6–8 or 12–15 wk of age) were injected with 100 U of heparin intraperitoneally 15 min before administration of pentobarbital sodium (250 mg/kg ip). The heart was rapidly excised and arrested in ice-cold buffer A, consisting of (in mM) 120 NaCl, 5.4 KCl, 1.2 MgSO4, 1.2 NaH2PO4, 5.6 glucose, 20 NaHCO3, 0.6 CaCl2, 10 2,3-butanedione monoxime, and 5 taurine, pH 7.5. The aorta was then cannulated, and the heart was retrogradely perfused at 37°C, with buffer A gassed with 95% O2-5% CO2 for 4 min, followed by 10–14 min with buffer A containing 25 μM CaCl2 and 59 U/ml type II collagenase (Worthington) or 0.1 mg/ml Liberase Blendzyme-1 (Roche). The coronary flow rate was set at 2.5 ml/min. The ventricles were then removed and digested at 37°C for 5–10 min longer in the presence of collagenase, 50 μM CaCl2, and 1% (wt/vol) FA-free BSA. When Liberase Blendzyme-1 was used for the perfusion, ventricles were digested in perfusion buffer for 10 min. Dispersed myocytes were filtered through an 85-μm mesh, gently pelleted by centrifugation, and resuspended in buffer A containing 100 μM CaCl2 and 0.6% FA-free BSA. Calcium concentrations were increased gradually to 1.0 mM in subsequent washings. The final viability of cardiomyocytes (percentage of rod-shaped cells that excluded trypan blue) was 85–90% (db/+) and 70–85% (db/db) with an overall yield of 1–1.5 × 106 cells/heart. To measure glucose uptake or FA oxidation, the cells were washed once with MEM (Sigma) containing 2% fetal serum albumin, 100 U/ml penicillin, and 100 μg/ml streptomyocin and then plated in laminin-coated tissue culture dishes. Studies were conducted 60 min after the platedown to allow viable cells to stick to the laminin so that nonviable cells could be removed before measurement of glucose uptake or FA oxidation.
Deoxyglucose uptake by isolated cardiomyocytes.
Glucose uptake assays were performed as described by Carley et al. (10), using radiolabeled 2-deoxyglucose as a glucose analog that is taken up by glucose transporters but then undergoes only one metabolic step (hexokinase reaction), resulting in the formation of 2-deoxyglucose phosphate. Thus metabolic trapping of 2-deoxyglucose phosphate within cells such as cardiomyocytes has been utilized extensively as a marker for glucose transport (4, 27). However, it must be acknowledged that, under certain circumstances such as altered nutritional state or the presence of an FA (oleate) in the reperfusion buffer, glucose metabolism by perfused rat hearts does not correlate perfectly with 2-deoxyglucose uptake (15). Therefore, results in this investigation will be presented as 2-deoxyglucose uptake.
Plated cardiomyocytes (35 mm) were washed twice with glucose-free DMEM (GIBCO) containing 0.2% FA-free BSA and 1.0 mM pyruvate (incubation buffer). Pyruvate is added as an energy source, since glucose is removed from the incubation medium to permit measurement of radiolabeled 2-deoxyglucose uptake (4). Cells were then incubated in the absence and in the presence of insulin (1–10 nM) in 2.0 ml of incubation buffer for 40 min at 37°C, with 95% O2-5% CO2 gassing. Twenty microliters of a 2-deoxyglucose solution containing 130 μl of glucose-free DMEM, 15 μl of a 200 mM 2-deoxyglucose solution, and 5 μCi of 2-deoxy-[3H]glucose (ICN Biomedicals) were added to the dishes, and the incubation was continued for 20 min. The buffer was then aspirated, and the cells were washed twice with cold PBS. Cells were lysed in 300 μl of 1 M NaOH at 37°C for 20 min and then washed with 200 μl of NaOH. Thirty microliters of 12 M HCl were added to 400 μl of the lysate to normalize the pH, and radioactivity was measured. Protein assay was performed with 10 μl of the lysate by use of a BCA Protein Assay Kit (Pierce Chemical). Deoxyglucose uptake is expressed as nanomoles per hour per milligram of protein.
FA oxidation by isolated cardiomyocytes.
FA oxidation experiments were performed with modifications to the method described by van der Lee et al. (36). Cardiomyocytes were allowed to attach to laminin-coated 60-mm center-well organ culture dishes. After being washed twice with glucose-free DMEM containing 0.2% FA-free BSA, cells were incubated in 1.0 ml of glucose-free DMEM containing 0.4 mM palmitic acid, 3% FA-free BSA, and 0.5 μCi [1-14C]palmitic acid in the absence or presence of glucose (5 mM), insulin (10 nM), or glucose and insulin combined. Incubations were carried out for 2 h at 37°C with 95% O2-5% CO2 gassing. Four hundred microliters of 5% KOH were then injected into the center wells of the dishes, and the oxidation was terminated by injecting 400 μl of 1 M H2SO4 into the incubation buffer. The dishes were sealed tight and stored at 4°C overnight. The trapping medium (KOH) was then assessed for 14CO2 by liquid scintillation counting. Oxidation rate was calculated by subtracting trapped 14CO2 at zero time from trapped 14CO2 after 2 h of incubation and is expressed as nanomoles CO2 per hour per milligram of protein. Total protein was determined using the BCA Protein Assay Kit. Preliminary experiments established that generation of 14CO2 was reasonably linear for incubation times up to 2 h; longer incubation times resulted in loss of viability (<50%).
Isolated heart perfusions.
Mouse hearts were perfused in working mode, as described by Carley et al. (10). The perfusate contained 11 mM glucose and 0.7 mM [9,10-3H]palmitate. FA oxidation was calculated by measurement of 3H2O in perfusate samples removed at 20-min intervals during the 60-min working heart perfusion protocol (preload pressure of 15 mmHg; afterload pressure of 50 mmHg). Steady-state rates of palmitate oxidation were calculated from the average of 3H2O formation in the three perfusate samples and expressed as micromoles per minute per gram dry weight of hearts (10). Functional parameters of working heart performance (heart rate, aortic and coronary flows) were also monitored.
Cardiomyocytes from db/+, db/db, or COOH-treated db/db mice were incubated in glucose-free DMEM containing 0.2% FA-free BSA and 1 mM pyruvate for 40 min in the absence and presence of 10 nM insulin, as described in Deoxyglucose uptake by isolated cardiomyocytes. Cells were then washed twice with cold PBS, scraped, and lysed for 15 min on ice in 100 μl of lysis buffer consisting of 20 mM Tris·HCl (pH 7.4), 50 mM NaCl, 50 mM NaF, 5 mM Na pyrophosphate, 0.25 M sucrose, 1% Triton X-100, mammalian protease inhibitor mixture, phosphatase inhibitor mixture, and 1 mM dithiothreitol. After centrifugation at 800 g for 10 min, the supernatants were frozen in liquid nitrogen and stored at −85°C. Protein concentration of the supernatants was determined using the BCA Protein Assay kit.
The insulin-stimulated phosphatidylinositol 3-kinase pathway results in phosphorylation of Akt at Ser473 and Thr308 and enzyme activation (7). AMPK is activated by phosphorylation of Thr172 in the catalytic α-subunit (11). Therefore, immunoblotting for phosphoenzymes was conducted. Boiled samples of cell homogenates were subjected to SDS-PAGE in gels containing 8% acrylamide and transferred to nitrocellulose as previously described (23). Membranes were blocked in 5% milk/1× TBS/0.1% Tween 20 and then immunoblotted at 1:1,000 dilution with either rabbit anti-phospho-α-AMPK (Thr172), rabbit anti-α-AMPK, rabbit anti-phospho-Akt (Ser473), and rabbit anti-Atk (Cell Signaling Technology) overnight at 4°C. After being washed extensively, the membranes were incubated with peroxidase-conjugated goat anti-rabbit secondary antibody in 5% milk/1× TBS/0.1% Tween 20. After a further washing, the antibodies were visualized using the Pharmacia enhanced chemiluminescence Western blotting detection system.
Data are expressed as means ± SE. Differences in the effects of insulin on metabolism (deoxyglucose uptake and FA oxidation) and on insulin-signaling enzymes (Akt and AMPK) in cardiomyocytes were determined by ANOVA with a Student-Newman-Keuls test for pairwise comparisons. Differences between means were considered statistically significant when the P values were <0.05.
Effect of insulin on deoxyglucose uptake.
The effect of varying concentrations of insulin on deoxyglucose uptake by control and db/db cardiomyocytes is shown in Fig. 1, A and B. In this representative experiment, insulin produced a concentration-dependent increase in deoxyglucose uptake by control cardiomyocytes isolated from 6- to 7-wk db/+ mice, with half-maximal stimulation at 1–2 nM (Fig. 1A). In contrast, the maximal insulin stimulation of deoxyglucose uptake into db/db cardiomyocytes was reduced. Similar results were obtained when cardiomyocytes were isolated from control and db/db mice at 12–13 wk of age (Fig. 1B); the maximal stimulatory effect of insulin on deoxyglucose uptake into db/db cardiomyocytes was reduced markedly without a change in sensitivity. The concentration of insulin that produced half-maximal stimulation of deoxyglucose uptake was 1.6 ± 0.3 nM (n = 3) and 1.8 ± 0.07 nM (n = 5) in control and db/db cardiomyocytes, respectively. Therefore, additional experiments were performed with submaximal (1 nM) and maximal (10 nM) insulin concentrations (Fig. 1, C and D). Cardiomyocytes from 6- to 8-wk db/db mice showed no change in basal (no insulin) deoxyglucose uptake, but insulin-stimulated uptake was reduced significantly (to 58% of control) at both insulin concentrations (Fig. 1C). Basal deoxyglucose uptake was reduced significantly in 12- to 15-wk db/db cardiomyocytes, with a pronounced reduction in insulin-stimulated glucose uptake (to 40% of control) at both 1 and 10 nM insulin concentrations (Fig. 1D).
Experiments were also conducted with cardiomyocytes from transgenic db/+ mice overexpressing hGLUT4 transporters (6). Basal deoxyglucose uptake was increased from 10.7 ± 1.4 (n = 8) in db/+ cardiomyocytes to 185 ± 1 nmol·h−1·mg protein−1 in db/+-hGLUT4 cardiomyocytes, as expected (Fig. 1E). Notwithstanding this large increase in basal uptake, insulin (1 and 10 nM) still produced a statistically significant increase (1.28- and 1.46-fold, respectively) in deoxyglucose uptake into db/+-hGLUT4 cardiomyocytes. With db/db-hGLUT4 cardiomyocytes, basal deoxyglucose uptake (189 nmol·h−1·mg protein−1) was unchanged compared with control db/+-hGLUT4 uptake, but insulin no longer produced any stimulatory effect on deoxyglucose uptake (Fig. 1E).
One-week treatment with COOH had no effect on the body weight of db/db mice, but both elevated serum glucose concentrations (46 ± 1 mM) and enhanced FA oxidation rates (0.32 ± 0.05 μmol·min−1·g−1) by perfused working hearts, which characterize untreated db/db mice (n = 5), were completely normalized by COOH treatment (glucose, 13.1 ± 2 mM; FA oxidation, 0.20 ± 0.02 μmol·min−1·g−1; n = 6–7), as observed previously after a 6-wk treatment protocol (10). Acute COOH treatment significantly increased basal deoxyglucose uptake and completely normalized insulin-stimulated deoxyglucose uptake (Fig. 2). However, the direct addition of 1 μM COOH to the cardiomyocyte preincubation had no effect on insulin-stimulated deoxyglucose uptake (results not shown).
Effect of insulin on FA oxidation.
The direct effect of insulin on rates of FA oxidation by control and db/db cardiomyocytes was examined (Fig. 3). Addition of insulin or glucose separately had no effect on palmitate oxidation by control myocytes; however, the combination of insulin plus glucose reduced palmitate oxidation significantly (Fig. 3A). The inhibitory effect of insulin on palmitate oxidation by db/db cardiomyocytes was also glucose dependent (Fig. 3B). Surprisingly, the basal rate of palmitate oxidation by db/db cardiomyocytes was not different from control basal rate.
Effect of insulin on Akt and AMPK.
A 21-fold increase in Akt phosphorylation was observed when control cardiomyocytes were incubated with insulin, as shown by increased phospho-Akt (Ser473) content measured by immunoblotting with no change in total Akt levels (Fig. 4). Insulin activation of Akt was not reduced significantly in db/db cardiomyocytes; furthermore, insulin-stimulated Akt activity in cardiomyocytes from COOH-treated db/db cardiomyocytes was unchanged relative to untreated db/db cardiomyocytes.
Despite activation of Akt by insulin in control cardiomyocytes (Fig. 4), incubation with 10 nM insulin had no effect on levels of phospho-AMPK (Thr172) or on total AMPK (Fig. 5). Insulin also had no effect on AMPK activation state in either db/db or COOH-treated db/db cardiomyocytes.
Diabetic db/db mice exhibit profound insulin resistance in vivo (9). Therefore, the objective of this investigation was to assess cardiac insulin sensitivity by using isolated cardiomyocytes from control and db/db mice.
Glucose uptake and Akt activation.
Basal deoxyglucose uptake in 6- to 8-wk db/db cardiomyocytes was unchanged relative to cardiomyocytes from control db/+ mice (Fig. 1C), consistent with the observation that basal glucose utilization in perfused hearts from 6-wk db/db mice was not different from control (1). In cardiomyocytes from 12- to 15-wk db/db mice, changes in basal deoxyglucose uptake were inconsistent; a significant reduction in basal uptake was noted in Fig. 1D but not in Fig. 2. By comparison, basal rates of glycolysis in perfused hearts from 12-wk db/db mice were reduced to 50% of control (5). It must be acknowledged that the considerable energy demand in beating perfused hearts will influence glucose transporter translocation; so it is perhaps not surprising that basal glucose uptake into quiescent cardiomyocytes does not correlate perfectly with glucose utilization by working perfused hearts. In addition, the rather harsh technique used to isolate cardiomyocytes could influence translocation of glucose transporters to the cell surface. Nevertheless, isolated cardiomyocytes provide a useful experimental model system to examine insulin sensitivity. For example, cardiomyocytes have been used to assess insulin signaling in cardiac-selective insulin receptor knockout mice (4) and in insulin-resistant ob/ob mice (27).
Insulin (10 nM) produced a marked (6- to 7-fold) stimulation of deoxyglucose uptake by control cardiomyocytes (Fig. 1, C and D), with half-maximal stimulation at 1.6 nM. This insulin responsiveness is very similar to results reported with cardiomyocytes from control C57BL/6J mice (27). For comparison, the serum concentrations of insulin in control mice under fed conditions range from 0.07 to 0.15 nM (1, 5), so insulin concentrations used with cardiomyocyte preparations are supraphysiological, likely reflecting loss of cell surface insulin receptors during isolation.
At 6 wk of age, db/db mice exhibit significant hyperglycemia (24 mM compared with 14 mM in control mice) and pronounced hyperinsulinemia (1), consistent with insulin resistance being the earliest phenotypic change in this monogenic model of obesity and type 2 diabetes (13). By comparison, 12-wk-old db/db mice have profound hyperglycemia (>40 mM glucose) and continuing hyperinsulinemia (1). Therefore, use of db/db mice at ∼6 and 12 wk allows age-dependent changes in cardiac phenotype to be monitored. Aasum et al. (1) have reported that perfused hearts from 6-wk-old db/db mice have an altered metabolic phenotype (increased FA oxidation but unchanged glucose oxidation) but normal contractile function, whereas 12-wk db/db hearts exhibit a more pronounced alteration in metabolism (decreased glucose utilization and increased FA oxidation) plus contractile dysfunction. Insulin-stimulated deoxyglucose uptake was reduced markedly in cardiomyocytes from db/db mice at both ages (Fig. 1); maximal stimulation of deoxyglucose uptake was diminished to 58 and 40% of control with cardiomyocytes from ∼6- and 12-wk diabetic db/db mice, respectively, with no change in insulin sensitivity. The 6-wk results show that cardiac insulin resistance precedes alterations in basal glucose utilization and contractile function in perfused db/db hearts (1).
Interestingly, the reduction in insulin-stimulated deoxyglucose uptake in db/db cardiomyocytes is much less than that observed with cardiomyocytes from obese ob/ob mice, where the ability of insulin to stimulate glucose uptake was completely abrogated (27). Obese ob/ob mice are glucose intolerant but exhibit only mild hyperglycemia (12, 27), in contrast to marked hyperglycemia in db/db mice. Thus the degree of cardiac insulin resistance can differ substantially from whole animal insulin resistance. It must be acknowledged that cardiac insulin resistance will make only a small contribution to changes in whole animal glucose homeostasis; skeletal muscle is the dominant organ responsible for in vivo glucoregulation (34). Nevertheless, alterations in substrate utilization due to cardiac insulin resistance may influence contractile function (5). Interestingly, cardiomyocytes from insulin-resistant obese Zucker rats showed no change in maximal insulin-stimulated glucose uptake but reduced insulin sensitivity (22). Of course, variations in isolation procedures can be a contributing factor to the differences observed with db/db and ob/ob mice and obese Zucker rats.
Basal glucose uptake was increased markedly in cardiomyocytes from db/+-hGLUT4 and db/db-hGLUT4 mice. Total hGLUT4 expression was elevated four- to sixfold in cardiac muscle from db/db-hGLUT4 mice (18); the 17-fold increase in basal deoxyglucose uptake (Fig. 1E) indicates preferential localization of functional hGLUT4 transporters in cell membrane of db/db-hGLUT4 cardiomyocytes. Despite markedly elevated basal deoxyglucose uptake, insulin was still able to produce a small but significant increase in deoxyglucose uptake (Fig. 1E). In contrast, insulin had no stimulatory effect on deoxyglucose uptake into db/db-hGLUT4 cardiomyocytes, indicating insulin resistance.
The decreased cardiac glucose utilization by db/db hearts has been ascribed to a GLUT4 translocation defect (18). Insulin-stimulated Akt phosphorylation was not reduced in cardiomyocytes from db/db hearts (Fig. 4); thus altered Akt activation must not be a contributory mechanism to the cardiac GLUT4 translocation defect (33). There are multiple other mechanisms downstream of Akt activation that could decrease insulin-stimulated glucose uptake into db/db cardiomyocytes (32); delineation of the specific mechanism(s) will be an important topic for future research. Reduced Akt activation by insulin has been observed in skeletal muscle and adipose tissue from db/db mice (19, 33), providing further evidence for tissue-specific features of insulin resistance in db/db mice. Interestingly, oral administration of a PPARγ agonist, COOH, to db/db mice for 1 wk completely normalized insulin-stimulated glucose uptake (Fig. 2), as observed previously after 6-wk treatment, but Akt activation by insulin was unchanged (Fig. 4).
Palmitate oxidation and AMPK activity.
The insulin inhibition of FA oxidation observed with control perfused mouse hearts (4, 27) could be a direct effect of insulin due to AMPK inhibition (16, 23). In control cardiomyocytes, insulin at much lower concentrations than in the previous studies (23) had no effect on either palmitate oxidation (Fig. 3A) or AMPK phosphorylation state (Fig. 5). However, the combination of insulin plus glucose did produce a significant reduction in palmitate oxidation (Fig. 3A). This suggests that the inhibitory effect of insulin on FA oxidation observed in whole perfused mouse hearts (4, 27) or isolated cardiomyocytes (Fig. 3) is indirect, perhaps secondary to increased glucose utilization (2). Similar results were observed by Awan and Saggerson (3) with rat cardiomyocytes. Recent experiments with perfused hearts from transgenic mice expressing a kinase-dead form of AMPK suggests that AMPK primarily regulates postischemic cardiac metabolism (31); there was no influence of AMPK on metabolism of normoxic hearts.
It should be noted that the glucose-dependent inhibition of FA oxidation by insulin does not rule out the possibility that insulin might promote FA uptake into mouse heart cardiomyocytes by stimulating the redistribution of fatty acid transporter FAT/CD36 to the cell membrane, as observed by Luiken et al. (26) with rat cardiomyocytes; increased intracellular FA levels as a consequence of increased uptake could be channeled by esterification into triacylglycerol stores rather than undergoing oxidation. Surprisingly, basal FA oxidation in db/db cardiomyocytes was not increased relative to control cardiomyocytes (Fig. 3); an elevated rate of FA oxidation is a characteristic of perfused db/db hearts, irrespective of whether FA is supplied as a palmitate-albumin complex (1, 5, 10) or lipoprotein lipase-derived FA from chylomicrons in the perfusate (29). This discrepancy between beating hearts and quiescent cardiomyocytes may reflect differences in the control of FA transporter translocation.
In summary, insulin had no direct (i.e., glucose-independent) effect on rates of FA oxidation, and AMPK phosphorylation was not changed by insulin. Maximal insulin-stimulated deoxyglucose uptake was reduced in cardiomyocytes from db/db mice at both early and established stages of their diabetic phenotype without any change in insulin sensitivity. Thus cardiac insulin resistance is a feature of db/db mice, at least with respect to deoxyglucose uptake. It must be acknowledged, however, that the degree of cardiac insulin resistance in db/db mice in vivo could be more pronounced because of the presence of circulating factors such as elevated plasma FA concentrations and adipokines (25, 28, 30). Therefore, an important objective for future investigations will be to investigate insulin sensitivity of the heart and other insulin target organs with in vivo methodologies applied to db/db mice.
This study was supported by operating grants from the Canadian Institutes of Health Research to D. L. Severson (MOP 13227) and J. R. B. Dyck (MOP 53088).
We acknowledge the expert technical assistance of Suzanne Kovacic.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by American Physiological Society