INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

GLUCOSE IS A major contributor of myocardial ATP production, with ATP being generated by glycolysis and glucose oxidation. Glycolysis is the first stage of glucose metabolism and converts glucose to pyruvate. The second stage of the pathway is glucose oxidation where glycolytically derived pyruvate is converted to acetyl CoA, which is further metabolized in the mitochondria to produce ATP. Although this is a well-regulated process, the effects of elevated levels of fatty acids on glucose metabolism can result in fatty acid-derived acetyl CoA decreasing the production of glucose-derived acetyl CoA via inhibition of the pyruvate dehydrogenase complex (28). This ability of fatty acid oxidation to inhibit glucose oxidation has been termed the Randle glucose-fatty acid cycle (27).

The inhibition of glucose metabolism by fatty acids is particularly relevant in a variety of pathological conditions where circulating levels of fatty acids are markedly increased (see Ref. 22 for review). The shift in substrate utilization away from glucose has been shown both clinically and experimentally to be detrimental to the myocardium in a variety of ischemic and nonischemic cardiovascular diseases. Pharmacological agents designed to stimulate myocardial glucose oxidation via mechanisms involving the Randle glucose-fatty acid cycle have proven to be efficacious in reducing ischemic injury (see Ref. 21 for review).

Recent evidence in skeletal muscle suggests that alternative pathways involving specific lipid signaling events may regulate glucose metabolism independent of the Randle cycle (32). This leaves open the possibility that lipids can control glucose utilization at no less than two levels, namely, the inhibition of glucose oxidation via fatty acid-derived acetyl CoA as well as intracellular lipid-mediated signaling mechanisms.

The effects of lipid signaling have been investigated as the mechanisms underlying insulin resistance in skeletal muscle. Although many factors may contribute to insulin resistance, one hypothesis is that elevated levels of plasma fatty acids are a key factor in rendering muscle tissue insensitive to insulin signaling (31, 32). Much of this information has originated from correlative studies that simply link insulin resistance in skeletal muscle to elevations in plasma fatty acid levels (see Ref. 26 for review). More recently, insulin resistance in a cultured skeletal muscle cell line has been suggested to result from a lipid-dependent inactivation of protein kinase B (PKB) (32). This dephosphorylation and inactivation of PKB is reported to be due to an excess accumulation of de novo synthesized ceramide (originating from excess palmitate) that may activate a ceramide-inducible phosphatase (32). Recent data in skeletal muscle have also suggested that fatty acid control of PKB phosphorylation may be a consequence of activation of a mitogen-activated protein (MAP) kinase pathway (32). The activation of extracellular signal-regulated kinase (ERK) and p38 MAP kinase have been reported to occur in response to palmitate but not other long-chain fatty acids (32). However, inhibition of ERK or p38 MAP kinase by PD98059 or SB203580, respectively, does not prevent palmitate- or ceramide-induced PKB inactivation (32). This suggests that other factors or pathways are involved in mediating palmitate-induced inactivation of PKB. However, it is not known whether de novo-synthesized ceramide can alter PKB activity in the heart, whether elevated fatty acid levels elicit signaling events in heart as in skeletal muscle, or whether alterations in PKB activation can significantly modify insulin-stimulated glucose uptake and metabolism in the heart.

PKB is a serine/threonine protein kinase that is activated by insulin via a multistep pathway involving a phosphatidylinositol 3-kinase (PI-3K)-dependent mechanism (see Ref. 5 for review). Insulin binds to the insulin receptor at the cell surface and activates receptor tyrosine kinase activity. Insulin receptor substrates are subsequently activated by the insulin receptor, promoting the activation of PI-3K and the subsequent formation of phosphatidylinositol 3,4,5-trisphosphate [PI(3,4,5)P₃]. PI(3,4,5)P₃ is responsible for the activation of 3-phosphoinositide-dependent kinase (PDK1), which phosphorylates PKB at threonine residue 308. Phosphorylation at this site leads to autophosphorylation at serine residue 473 (37), and this latter phosphorylation of PKB is indicative of its activation state (1). Of particular importance to glucose metabolism is the activation of PKB and increased glucose uptake and glycogen synthesis (34, 39). The control of glucose uptake via translocation of GLUT4 to the plasma membrane may be one mechanism by which PKB can activate glucose transport and subsequent metabolism (15). In addition, the phosphorylation and inhibition of glycogen synthase kinase (GSK)-3 by PKB activates glycogen synthase and thereby promotes glycogen synthesis (19). The precise consequences of these events on myocardial glucose metabolism have not been defined. In addition, how lipids can modulate glucose uptake and utilization in the heart has not been explored in detail.

Because of the observation that palmitate can significantly modify insulin-stimulated PKB activation in skeletal muscle, the aim of this study was to determine if palmitate also affects myocardial PKB activation. In addition, we sought to define the relationship between insulin-stimulated PKB phosphorylation, palmitate, and the control of glucose uptake, glycolysis, and glucose oxidation in the heart. This was performed using isolated perfused rat hearts where perfusion conditions can be controlled, physiologically relevant levels of cardiac work can be performed, and rates of glucose uptake and metabolism can be directly measured. Cultured cardiac cells were also used to investigate the signaling pathways involved in the control of PKB activity by palmitate.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials. Fatty acid-free BSA was purchased from Sigma. Ex-Cell 320 medium was obtained from JRH Biosciences; norepinephrine, fibronectin, and retinoic acid were from Sigma; Bacto Gelatin was from Difco; and all other tissue-culture solutions were purchased from GIBCO. Primary antibodies used in this study were rabbit anti-Akt, rabbit anti-phospho-Akt (Ser-473 and Thr-308), rabbit anti-phospho-GSK-3/ (Ser-21/9), rabbit anti-phospho-p38 MAP kinase (Thr-180/Tyr-182), and rabbit anti-phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibodies, all from New England Biolabs. Goat anti-rabbit secondary antibodies were obtained from Santa Cruz Biotechnology.

Heart perfusions. Isolated rat hearts were perfused in working mode as we have previously described (11, 16). Three groups of hearts were perfused with Krebs-Henseleit solution (118 mM NaCl, 4.7 mM KCl, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 2.5 mM CaCl₂, and 25 mM NaHCO₃) containing 11 mM dual-labeled [5-³H]- and [U-¹⁴C]glucose as follows: glucose alone (G), glucose with the addition of 100 µU/ml insulin (GI), or glucose, 100 µU/ml insulin, and 1.2 mM palmitate prebound to 3% BSA (GIP). Rates of glycolysis and glucose oxidation were measured directly from the production of ³H₂O and ¹⁴CO₂ as described previously (16). Hearts were perfused aerobically for either 45 or 80 min. This allows for the determination of glycogen synthesis measured from the increase in [5-³H]- and [¹⁴C]glucosyl units in total myocardial glycogen (16).

Immunoblot analysis. Boiled samples of heart tissue homogenates or cell homogenates were subjected to SDS-PAGE and transferred to nitrocellulose as previously described (13). Membranes were blocked in 5% BSA/1× TBS and then immunoblotted with either rabbit anti-phospho-Akt (Ser-473 or Thr-308), rabbit anti-Akt, rabbit anti-phospho-GSK-3/ (Ser-21/9), rabbit anti-phospho-p38 MAP kinase (Thr-180/Tyr-182), and rabbit anti-phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibodies (1:1,000 dilution) in 5% BSA/1× TBS. After being washed extensively, the membranes were incubated with a peroxidase-conjugated goat anti-rabbit secondary antibody in 5% BSA/1× TBS. After further washing, the antibodies were visualized using the Pharmacia enhanced chemiluminescence Western blotting and detection system.

In vitro PKB assay. Powdered heart tissue (100 mg) was homogenized in 400 µl of buffer A [50 mM Tris · HCl, pH 7.5, 0.1% (wt/vol) Triton X-100, 1 mM EDTA, 1 mM EGTA, 50 mM sodium fluoride, l0 mM sodium B-glycerophosphate, 5 mM sodium pyrophosphate, 0.1% (vol/vol) 2-mercaptoethanol, 1:1,000 dilution of mammalian tissue protease inhibitor cocktail (Sigma P 8340), 1:100 dilution of phosphatase inhibitor cocktail (Sigma P 5726), and 1:100 dilution of phosphatase inhibitor cocktail (Sigma P 2850)]. The homogenized samples were centrifuged at 11,000 g for 10 min at 4°C, and the supernatant was removed. After determination of the protein concentrations of each sample supernatant, 30 µl of immobilized PKB 1G1 monoclonal antibody (Cell Signaling Technology) were added to 200 µg of supernatant, the volumes were adjusted to 100 µl with buffer A, and the samples were incubated on a rotator at 4°C for 2 h. After being washed two times with buffer A containing 0.5 M NaCl, two times with buffer B [50 mM Tris · HCl, pH 7.5, 0.03% (wt/vol) Brij-35, 0.1 mM EGTA, 0.1% 2-mercaptoethanol, 1:1,000 dilution of mammalian tissue protease inhibitor cocktail, and 1:100 dilution of both phosphatase inhibitor cocktails (see above)], and once with ADB buffer [20 mM MOPS, pH 7.2, 25 mM B-glycerol phosphate, 5 mM EGTA, 1 mM dithiothreitol, 1:1,000 dilution of mammalian tissue protease inhibitor cocktail, and 1:100 dilution of both phosphatase inhibitor cocktails (see above)], the antibody/agarose/enzyme complex was resuspended in 10 µl of ice-cold ADB buffer. The subsequent kinase assay was performed as described in the PKB kinase assay protocol provided by the suppliers of the crosstide peptide (Upstate Biotechnology).

Cell culture. HL-1 cells are cardiac muscle cells derived from a mouse atrial tumor and were established by Dr. W. C. Claycomb, as described previously (6). Before being cultured, tissue-culture dishes were coated with gelatin/fibronectin (2 µg/cm²). Cells were cultured in Ex-Cell 320 medium supplemented with fetal bovine serum (10%), insulin (15 µg/ml), penicillin/streptomycin (100 U/ml:100 µg/ml), nonessential amino acids (0.1 mM), endothelial cell growth supplement (50 µg/ml), retinoic acid (1 µM), and norepinephrine (0.1 mM) as described (6).

Cell treatments. After being cultured for 48 h, HL-1 cells were serum starved for 24 h, washed in serum-free medium, and then treated with either 3% BSA or 3% BSA prebound to 1.2 mM palmitate for 6 h. Pretreated cells were then incubated for an additional hour in the absence or presence of insulin (200 nM). For experiments involving treatment with C₂-ceramide, HL-1 cells were serum starved and washed as above and then subsequently subjected to either no treatment (0 nM insulin), insulin (200 nM) for 1 h, C₂-ceramide (32 µM) for 2 h, C₂-ceramide (32 µM) for 2 h with the inclusion of insulin (200 nM) for the second hour, C₂-ceramide (32 µM) for 3 h, or C₂-ceramide (32 µM) for 3 h with the inclusion of insulin (200 nM) for the third hour. For experiments involving the determination of the effects of palmitate on insulin-pretreated cells, HL-1 cells were serum starved as above and then treated with either glucose alone or glucose plus insulin (200 nM). After 1 h, the cells were washed, and the media were replaced with insulin-free media containing either 3% BSA or 1.2 mM palmitate/3% BSA for an additional 3 h. For all cell culture experiments, immediately after the final incubation stage cells were harvested in a lysis buffer [20 mM Tris · HCl (pH 7.4), 50 mM NaCl, 50 mM NaF, 5 mM Na pyrophosphate, 0.25 M sucrose, protease inhibitor cocktail (Sigma), phosphatase inhibitor cocktail (Sigma), and 1 mM dithiothreitol] and prepared for SDS-PAGE.

Statistical analysis. All data are presented as means ± SE. The unpaired t-test was used for the determination of statistical differences between two groups. For comparison of three groups, analysis of variance followed by the Newman-Keuls test was used. A value of P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of palmitate on glucose metabolism in working hearts. Hearts in each of the three perfusion groups (G, GI, and GIP) had similar values of left ventricular work and coronary flow and appeared to be functionally identical (data not shown). Hearts perfused with glucose only demonstrated a glycolytic rate of 5.26 ± 0.30 µmol · g dry wt¹ · min¹ (Fig. 1A, open bars), which was increased by 29% (Fig. 1A, gray bars) in the presence of insulin. In the presence of glucose, insulin, and palmitate, the rate of glycolysis decreased to below the rate observed in hearts perfused with glucose alone (Fig. 1A, solid bars). A similar profile, namely, stimulation by insulin and inhibition by palmitate, was observed in the three groups with respect to glucose oxidation rates (Fig. 1B). Glucose uptake, calculated from the sum of rates of glycogen synthesis and glycolysis, was significantly increased in hearts perfused with glucose and insulin, compared with glucose alone (Fig. 1C). However, in the presence of palmitate, insulin-stimulated glucose uptake was completely inhibited to a rate similar to that observed in hearts perfused with glucose alone (Fig. 1C).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Effects of perfusate composition on rates of glucose uptake, glycolysis, and glucose oxidation. Rat hearts were perfused with Krebs solution that contained either 11 mM glucose (G), 11 mM glucose and 100 µU/ml insulin (GI), or 11 mM glucose, 100 µU/ml insulin, and 1.2 mM palmitate (GIP) as described in METHODS. Rates of glycolysis (A) and glucose oxidation (B) were measured directly from the production of 3H2O and 14CO2 from 3H/14C dual-labeled glucose. Glucose uptake (C) was calculated as the sum of glycogen synthesis (determined from the incorporation of radiolabeled glucose into glycogen) and glycolysis. Values are means ± SE of measurements from 6-8 hearts. * Significant difference from the G group; tsignificant difference from the GIP group as assessed by analysis of variance followed by the Newman-Keuls test.

Effects of palmitate on insulin-stimulated PKB signaling in rat hearts. To understand better the mechanisms involved in the palmitate-induced alterations of myocardial glucose metabolism, extracts from hearts from the three perfusion groups (G, GI, and GIP) were subjected to immunoblot analysis to assess levels of total PKB or phosphorylated PKB (Ser-473 or Thr-308; Fig. 2, A and B). Hearts perfused with glucose alone had relatively little phosphorylated PKB protein. As expected, upon treatment with insulin, the level of phosphorylated PKB increased significantly. However, in the presence of palmitate, the stimulatory effect of insulin was blunted.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Interactions between palmitate and insulin on protein kinase B (PKB) and mitogen-activated protein (MAP) kinase signaling. Extracts of the three groups of perfused rat hearts (G, GI, and GIP as described in Fig. 1) were subjected to immunoblot analysis as described in METHODS. A: a representative immunoblot performed using anti-phospho-PKB (Ser-473 and Thr-308) antibodies, an anti-phospho-GSK-3/ (Ser-21/9) antibody, and an anti-PKB antibody that served as a protein loading control. B: a quantitative assessment of immunoblots subjected to densitometric analysis. C: PKB activity, assayed in vitro using the crosstide peptide as a substrate, is shown for extracts of hearts from the three perfusion conditions: G, GI, and GIP. D: immunoblots performed using anti-phospho-p38 MAP kinase (Thr-180/Tyr-182) and anti-phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibodies (P-p38 MAPK and P-ERK, respectively). Extracts from rat hearts subjected to 30 min of severe no-flow ischemia were used as a positive control for P-p38 MAPK (ISCH). * Significant difference from the glucose only group (G); tsignificant difference from the GIP group as assessed by analysis of variance followed by the Newman-Keuls test.

Effects of altered glucose oxidation rates on PKB phosphorylation. As shown above, perfusing rat hearts with 1.2 mM palmitate led to significantly reduced glucose oxidation rates well below values obtained in hearts perfused with glucose alone (Fig. 1). Although we hypothesize that decreased PKB phosphorylation alters glucose utilization, it is possible that palmitate-induced decreases in glucose oxidation rates may be the cause, and not the consequence, of decreased PKB phosphorylation. To investigate this cause and effect relationship, we examined whether maintaining high glucose oxidation rates despite the presence of 1.2 mM palmitate could modulate PKB phosphorylation. To stimulate glucose oxidation in the presence of palmitate, hearts were aerobically perfused for 60 min with GIP in the absence or presence of 3 mM dichloroacetate (DCA). DCA is a known inhibitor of pyruvate dehydrogenase kinase and therefore activates the pyruvate dehydrogenase complex (PDH) and increases glucose oxidation rates (36). This activation of PDH results in a stimulation of glucose oxidation (3-fold; data not shown) even in the presence of 1.2 mM palmitate. Despite this effect, PKB phosphorylation was not elevated in DCA-perfused hearts compared with GIP hearts (Fig. 3, lanes 1-4 and 9-12). These data suggest that the observed palmitate-induced effects on PKB phosphorylation are not simply a consequence of depressed glucose oxidation rates, but rather suggest that altered PKB phosphorylation contributes to altered glucose metabolism.

View larger version (55K): [in this window] [in a new window]   Fig. 3.   Effects of glucose oxidation and fatty acid oxidation on PKB phosphorylation. Extracts of rat hearts that had been perfused with Krebs solution containing glucose, insulin, and palmitate (GIP) and in the absence (lanes 1, 2, 5, 6, 9, 10, 13, and 14) or presence of either 3 mM dichloroacetate (DCA; lanes 3, 4, 11, and 12) or 2 mM oxfenicine (lanes 7, 8, 15, and 16) were subjected to immunoblot analysis using anti-phospho-PKB (Ser-473 and Thr-308) antibodies. Blots obtained with an anti-PKB antibody served as a loading control. The extent of PKB phosphorylation was not affected by the presence of either an activator of glucose oxidation (DCA) or an inhibitor of palmitate oxidation (oxfenicine).

Effects of reducing fatty acid oxidation on PKB phosphorylation. We and others have previously shown that perfusing rat hearts with 1.2 mM palmitate (GIP) also results in an acceleration of fatty acid oxidation rates (12, 30). Because by-products of fatty acid oxidation could be responsible for signaling altered PKB phosphorylation, we also investigated the effects of reducing fatty acid oxidation on PKB phosphorylation state using the fatty acid oxidation inhibitor oxfenicine. Oxfenicine is a potent carnitine palmitoyltransferase (CPT)-1 inhibitor that prevents the transport of long-chain fatty acids into the mitochondria and greatly reduces fatty acid oxidation rates (33, 40). The phosphorylation state of PKB was not different in GIP hearts perfused with 2 mM oxfenicine compared with untreated GIP hearts (Fig. 3, lanes 5-8 and 13-16), suggesting that fatty acid metabolites are not involved in palmitate-induced inhibition of PKB phosphorylation. In addition to reported decreases in fatty acid oxidation rates (33, 40), GIP perfused hearts also demonstrate a sixfold increase in glucose oxidation rates with the addition of 2 mM oxfenicine (data not shown). Thus these data further suggest that alterations in glucose oxidation rates are not responsible for controlling PKB phosphorylation, confirming the data obtained with DCA-perfused hearts. Finally, because the inhibition of CPT-1 also reduces the concentration of palmitoylcarnitine (23), our studies also suggest that palmitoylcarnitine does not play a role in controlling PKB phosphorylation.

Palmitate signaling pathways in cultured cardiac myocytes. With the use of cultured cardiac cells to mimic the effects of palmitate observed in the intact heart, HL-1 cardiac cells were incubated in the absence or presence of 1.2 mM palmitate and then exposed to 200 nM insulin. Although this concentration of insulin is considerably higher than what is used in the perfused heart, direct comparisons between cultured cells and the intact heart perfusions can be misleading, because cultured cell lines are normally grown in high concentrations of insulin and can become partially resistant to the effects of insulin. Therefore, higher concentrations of insulin are needed in HL-1 cells to exert the same effects as in the perfused heart. As seen in Fig. 4A, lanes 3 and 4, insulin promotes PKB phosphorylation at Ser-473 (and Thr-308; data not shown) in HL-1 cells. In addition, HL-1 cells also possess the cell transduction mechanisms required for the inhibitory effects of palmitate on insulin-mediated phosphorylation of PKB (Fig. 4A, lanes 5 and 6). However, in a 3-h insulin washout experiment, the presence of 1.2 mM palmitate (Fig. 4B, lane 4) does not accelerate the rate at which PKB is dephosphorylated beyond that observed with the removal of insulin alone (Fig. 4B, lane 3). This was also the case after much shorter time points of 10, 30, 60, and 120 min (data not shown). Similar results were also obtained using an anti-phospho PKB (Thr-308) specific antibody (data not shown).

View larger version (53K): [in this window] [in a new window]   Fig. 4.   Alteration of PKB phosphorylation by palmitate in cultured cardiac myocytes. HL-1 cardiac myocytes were incubated in media containing either glucose (G), glucose and insulin (GI), or glucose, insulin, and palmitate (GIP) as described in METHODS. A: immunoblots obtained from extracts of these three groups of cells using an anti-phospho-PKB (Ser-473) antibody in addition to an anti-PKB antibody that served as a loading control. B: immunoblot analysis performed using anti-phospho-PKB (Ser-473) and anti-PKB antibodies on cellular extracts from HL-1 cardiac myocytes treated with either glucose alone (lane 1) or glucose and insulin (lanes 2, 3, and 4). After 1 h, the cells were washed and the media were replaced with insulin-free media containing either 3% BSA (lane 3) or 1.2 mM palmitate/3% BSA (lane 4) for an additional 3 h.

Ceramide signaling pathways in cultured cardiac myocytes. The mechanism by which palmitate is able to antagonize insulin-mediated stimulation of PKB phosphorylation in cardiac cells is unknown. However, in cultured skeletal muscle cells it has been demonstrated that elevated levels of de novo-synthesized ceramide originating from palmitate are responsible for promoting the dephosphorylation of PKB (32). Because the cultured cardiac cells behaved similarly to the intact working heart, we tested whether ceramide could be responsible for the decrease in PKB (S473) phosphorylation demonstrated in the perfused hearts. HL-1 cells treated with insulin in the presence of glucose demonstrated significant increases in PKB phosphorylation (Fig. 5A, lanes 1 and 2). Although these HL-1 cells are responsive to insulin, treatment with C₂-ceramide for 2 or 3 h, in the presence of insulin for the final hour (Fig. 5A, lanes 4 and 6, respectively), had no effect on insulin-stimulated PKB phosphorylation. However, C₂-ceramide did increase phosphorylation of p38 MAP kinase (data not shown), confirming the effectiveness of ceramide treatment on other pathways. We also investigated whether C₂-ceramide could accelerate PKB dephosphorylation beyond that demonstrated with the removal of insulin alone. The washout experiment shown in Fig. 5B shows that C₂-ceramide does not accelerate PKB dephosphorylation. Similar results were also obtained using an anti-phospho PKB (Thr-308) specific antibody (data not shown). Finally, ceramides contained in lipids extracted from homogenized heart tissue were deacylated by strong alkaline digestion to sphingosine, reacted with o-phthalaldehyde to form a fluorescent compound that was separated by HPLC, and quantified as described (3). These data show that there are no increases in the ceramide levels in hearts perfused with palmitate compared with the other two groups of perfused hearts (G, 204 ± 9.9 pmol C₁₆-sphigosine/g dry wt vs. GI, 193 ± 14.6 pmol C₁₆-sphigosine/g dry wt vs. GIP, 160 ± 9.2 pmol C₁₆-sphigosine/g dry wt). Taken together, these data suggest that, in contrast to findings in skeletal muscle (32), the effects of palmitate on PKB are not due to elevated ceramide levels and do not act in conjunction with p38 MAP kinase in cardiac cells.

View larger version (61K): [in this window] [in a new window]   Fig. 5.   Palmitate and ceramide signaling in cultured cardiac myocytes. Immunoblot analysis was performed using anti-phospho-PKB (Ser-473) and anti-PKB antibodies on cellular extracts from HL-1 cardiac myocytes. A: HL-1 cells were subjected to either no treatment (lane 1), insulin (200 nM) for 1 h (lane 2), C2-ceramide (32 µM) for 2 h (lane 3), C2-ceramide (32 µM) for 2 h with the inclusion of insulin (200 nM) for the second hour (lane 4), C2-ceramide (32 µM) for 3 h (lane 5), or C2-ceramide (32 µM) for 3 h with the inclusion of insulin (200 nM) for the third hour (lane 6). B: HL-1 cells were subjected to either no treatment (lane 1); insulin (200 nM) for 2 h (lane 2); insulin (200 nM) for 1 h, washed with insulin-free media and then treated with C2-ceramide for 1 h (lane 3); or insulin (200 nM) for 1 h, washed with insulin-free media and then treated with DMSO (control) for 1 h (lane 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this report we have used both isolated perfused working hearts and cultured cardiac myocytes to assess mechanisms underlying palmitate-induced inhibition of insulin-stimulated glucose uptake and utilization. Hearts perfused with insulin in the presence of glucose demonstrated increased rates of glucose uptake, glycolysis, and glucose oxidation compared with hearts perfused with glucose alone. Palmitate blunted the stimulatory effects of insulin on glucose uptake and glycolysis, resulting in a restoration of these values to those observed in hearts perfused with glucose alone. In addition, the rate of glucose oxidation was significantly impaired in hearts perfused with palmitate, as would be predicted by the Randle glucose-fatty acid cycle.

We also investigated whether PKB could respond to an elevated palmitate level and contribute to the regulation of glucose metabolism independent from the direct metabolic effects of fatty acids. Our data suggest that palmitate can inhibit insulin-stimulated glucose utilization, at least in part, via a palmitate-induced inhibition of PKB phosphorylation. The observation that insulin-stimulated PKB phosphorylation can be blunted by the presence of palmitate suggests that palmitate has effects on the PKB signaling pathway and may regulate glucose utilization in the heart. In addition, downstream targets of PKB, such as GSK-3, are also altered by the presence of palmitate, demonstrating a decrease in PKB activity and downstream signaling in the presence of palmitate. Finally, although phosphorylation of Thr-308 appears to be more sensitive to palmitate in the perfused heart (Fig. 2B), our data support the notion that phosphatases are not involved in palmitate-mediated reduction of PKB phosphorylation in the heart.

In addition to decreasing PKB phosphorylation, perfusing hearts with 1.2 mM palmitate significantly reduces glucose oxidation rates and enhances fatty acid oxidation rates (12, 30). Thus we investigated whether, in the continued presence of 1.2 mM palmitate, pharmacological restoration of the rate of glucose oxidation or inhibition of the rate of fatty acid oxidation could modify the palmitate-mediated inhibition of PKB phosphorylation. DCA and oxfenicine each significantly accelerated the rate of glucose oxidation to values similar to rates reported for hearts perfused with physiologically "normal" 0.4 mM palmitate (18, 25). Although DCA acts directly on glucose oxidative pathways, oxfenicine acts indirectly and accelerates glucose oxidation by inhibiting fatty acid oxidation. Therefore, oxfenicine not only reduced the production of fatty acid oxidative metabolites but also accelerated glucose oxidation. Despite these changes in glucose oxidation and fatty acid oxidation, we did not observe any associated increase in PKB phosphorylation. This suggests that palmitate-induced PKB dephosphorylation does not occur simply as a consequence of decreased glucose oxidation or increased fatty acid oxidation, but rather suggests that altered PKB phosphorylation is one of the causes of altered glucose uptake and metabolism.

It has been suggested from studies in C₂C₁₂ skeletal muscle cells that the palmitate-mediated inhibition of PKB phosphorylation is due to a ceramide-mediated acceleration of PKB dephosphorylation (32). Thus we investigated the possibility that ceramide regulates PKB phosphorylation in cardiac myocytes. Cultured cardiac myocytes treated with glucose, glucose/insulin, or glucose/insulin/palmitate demonstrated alterations in PKB phosphorylation similar to those observed in the intact hearts (Figs. 4A and 2A, respectively). However, the effects of palmitate on PKB were ceramide independent in that exogenous ceramide could not prevent PKB phosphorylation in response to insulin or accelerate PKB dephosphorylation upon insulin withdrawal (Fig. 5, A and B). Recent work by Cazzolli et al. (4) demonstrates that palmitate disrupts insulin signaling by activating a protein phosphatase 2A-like phosphatase in skeletal muscle cells. However, our data do not implicate phosphatases in the control of palmitate-regulated PKB phosphorylation in the heart and may indicate tissue-specific regulation of PKB. This concept of tissue-specific regulation of enzymes involved in glucose metabolism is not without precedent. For instance, the larger phosphofructose-2-kinase (pfk-2) isoform present in the heart appears to be under phosphorylation control (2, 8, 9), whereas the smaller skeletal muscle isoform lacks the amino acids containing certain phosphorylation sites and is not regulated by phosphorylation (9, 29). Whether PKB activating kinases are differentially regulated in the heart and skeletal muscle is currently unknown.

Although the mechanism by which palmitate exerts its inhibitory effects on PKB phosphorylation is as yet unknown, one possibility is that palmitate may be inhibiting upstream PKB kinases such as integrin-linked kinase or PDK. Alternatively, palmitate may prevent the translocation of PKB to the plasma membrane (and subsequent activation) similar to what has been reported to occur as a result of the presence of ceramide (35). Regardless, it appears that cardiac myocytes act differently than skeletal muscle cells in the manner by which PKB phosphorylation is regulated by palmitate (Fig. 6).

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Proposed model for palmitate-dependent inhibition of insulin-induced PKB phosphorylation. Insulin signaling may be inhibited at many stages including 1) the site of insulin interaction with the insulin receptor (IR); 2) the activation of insulin receptor substrates (IRS) or phosphatidylinositol 3-kinase (PI-3K); 3) the activation of PDK1, -2, and/or ILK; and 4) the activation of PKB. Although the reduction of insulin binding to its receptor may be due directly to palmitate, intracellular mechanisms may be more complex. Intracellular palmitate has many fates (indicated by the boxed area), beginning with the conversion to palmitoyl CoA. Palmitoyl CoA is subsequently used for triglyceride synthesis (TG), ceramide synthesis, or can undergo -oxidation within the mitochondria to produce acetyl CoA. In the heart, palmitate or its metabolites may inhibit insulin signaling at any of these four stages. However, unlike skeletal muscle, ceramide does not appear to be responsible for alterations in PKB signaling in heart muscle, as indicated by the X.

In summary, this report describes a novel pathway by which myocardial glucose metabolism may be controlled by lipid signaling events, independent the metabolic actions of fatty acids described in the Randle glucose-fatty acid cycle. In addition, we have identified PKB as a central enzyme involved in this signaling pathway and have described how palmitate alters glucose uptake, glycolysis, and glucose oxidation in the heart. However, this study did not examine the effects of all long-chain fatty acids, and we cannot conclude that similar results would be obtained with other fatty acids such as oleate or linoleate. Finally, our results also suggest that palmitate may alter PI-3K activity. Indeed, PI-3K inhibitors have been shown to cause changes in glucose uptake in the heart (14), and it is possible that these changes may be attributed to alterations in PKB activation similar to those demonstrated with palmitate. These possibilities are currently being explored.

Although this study examined the potential metabolic effects of altered PKB activity, PKB has multiple nonmetabolic cellular functions (see Ref. 7 for review). Because palmitate can regulate PKB, the effects of palmitate may extend to include a wide variety of these cellular processes. Indeed, palmitate-induced apoptosis has been shown to occur in cardiac myocytes (10, 20) and may be controlled, at least in part, via decreased PKB phosphorylation (17). Future work will investigate these possibilities.