Acute activation of the serine-threonine kinase Akt is cardioprotective and increases glucose uptake, at least in part, through enhanced expression of GLUT4 on the sarcolemma. The effects of chronic Akt activation on glucose uptake in the heart remain unclear. To address this issue, we examined the effects of chronic Akt activation on glucose uptake, glycogen storage, and relevant glucose transporters in the hearts of transgenic mice. We found that chronic cardiac activation of Akt led to a substantial increase in the rate of basal glucose uptake (P < 0.05) but blunted the response to insulin (1.9 vs. 18.1-fold increase compared with baseline) using NMR in ex vivo perfused heart. Basal glucose uptake was also increased in Akt transgenic mice in vivo (P < 0.005). These changes were associated with an increase on glycogen deposition, examined with histochemical staining, biochemical (>6-fold, P < 0.001) and in vivo radioactive (5-fold, P < 0.01) assays. Studies in chimeric hearts of female X-linked transgenic Akt mice suggested that increased glycogen deposition occurred as a cell autonomous effect of transgene expression. Interestingly, although sarcolemmal GLUT1 was not significantly altered, chronic Akt activation actually decreased plasma membrane GLUT4. Moreover, intracellular pools of GLUT1 were modestly reduced, whereas intracellular GLUT4 was substantially reduced. It seems likely that neither GLUT1 nor GLUT4 explains the increase in basal glucose uptake but that these reductions contribute to the loss of insulin responsiveness that we observed. These data demonstrate that chronic Akt activation increases basal glucose uptake and glycogen deposition while inhibiting the response to insulin.
- glucose transporters
- transgenic mice
in the normal adult heart, fatty acids are the preferential substrate for metabolism and ATP generation. However, glucose is an important alternative energy source in specific pathological conditions, such as ischemia (33), where the more favorable bioenergetics of glucose metabolism are thought to confer important benefits (reviewed in Ref. 9). Glucose utilization is initiated by glucose uptake and in some setting this appears to be the rate-limiting step in glucose utilization (26). Thus understanding control of glucose uptake in the heart may have implications for our approach to a variety of cardiac conditions.
In most tissues, insulin provides the dominant stimulus to enhance glucose transport, largely through activation of the downstream signaling molecules insulin receptor substrate-1 (IRS-1), phosphatidylinositol-3 kinase (PI3K), and the serine/threonine kinase Akt (or protein kinase B). We (30) previously found that acute activation of Akt significantly increased glucose uptake in vitro in cardiomyocytes to a level comparable to that seen with pharmacological insulin stimulation, accompanied by increased sarcolemmal glucose transporter-4 (GLUT4) in vivo. This observation is consistent with prior studies in L6 skeletal myoblasts (15, 42) and adipocytes (5, 17, 22). However, less is known about the effects of chronic Akt activation in the heart.
Glucose uptake is controlled by GLUTs on the plasma membrane. GLUT1 and GLUT4 appear to be the major glucose transporters in cardiac and skeletal muscle as well as in adipose tissue. Although GLUT4 is more abundant than GLUT1 in both the heart and skeletal muscle, the amount of GLUT1 relative to GLUT4 is much higher in the heart than in skeletal muscle (12, 27). In addition, the distribution of these glucose transporters differs between cardiac and skeletal muscle. In skeletal muscle, GLUT4 is more abundant in the membranes of intracellular vesicles and is recruited to the plasma membrane after insulin stimulation, whereas GLUT1 is localized mainly to the plasma membrane at baseline and is not significantly affected by insulin stimulation (43). In contrast, in the heart, GLUT1 is also located in intracellular pool and is recruited to the plasma membrane in response to insulin (2). Thus both GLUT1 and GLUT4 may contribute to enhanced glucose uptake after insulin stimulation in the heart (35).
In addition to GLUTs, hexokinase II plays an important role in regulating insulin-stimulated glucose uptake (3, 14) by catalyzing the first committed step in glycolysis, phosphorylation of glucose to glucose 6-phosphate. Hexokinase II enhances glucose uptake and glycogen synthesis in the heart (25).
To investigate the effects of chronic Akt activation in the heart, we generated transgenic (TG) mice with cardiac-specific expression of a constitutively active form of Akt (myr-Akt) driven by the myosin heavy chain-α promoter and measured glucose uptake at baseline and in response to insulin with the ex vivo Langendorff perfusion system using 31P-NMR spectroscopy. We found an increase in basal glucose uptake but a blunted response to insulin stimulation. In addition to ex vivo perfused heart, we confirmed increased basal glucose uptake using in vivo awake mice. Although the level of total GLUT1 expression was similar in both TG and littermate nontransgenic (NTG) mice, GLUT4 in TG mice was substantially lower than in NTG mice. Sucrose gradient protein fractionation demonstrated that, in intracellular vesicular pools, GLUT4 in hearts from TG mice was dramatically decreased in addition to a slight reduction on GLUT1. These reductions may contribute to impaired insulin sensitivity in hearts of TG mice. These studies delineate the effects of chronic Akt activation on glucose uptake in the heart and may have relevance for heart failure or other conditions in which Akt activity is enhanced (16).
Generation of TG mice.
The Subcommittee on Research Animal Care in Massachusetts General Hospital, Harvard Medical School, approved all aspects of animal care and experimentation performed in this study. TG mice expressing the hemagglutinin (HA)-tagged Akt with src myristoylation (myr) signal under the direction of the murine α-myosin heavy-chain promoter have previously been described in detail (29). Two viable lines (TG20 and TG564) have been backcrossed >10 generations to C57BL/6 wild-type mice. These lines exhibit stable Mendelian transgene inheritance consistent with autosomal (TG564) and X-linked transmission (TG20). 10- to 18-wk-old age-matched mice were used in all experiments, and either NTG or C57BL/6 wild-type mice were used as control. Because of X-chromosome inactivation, hearts from female TG20 mice exhibit chimeric transgene expression (29). TG564 mice were used for the described experiments unless specified in the figure legend.
Ex vivo glucose uptake using 31P-NMR spectroscopy.
After animals were deeply anesthetized with pentobarbital sodium (60 mg/kg ip) following 30 min of heparin (1,000 IU/kg ip) anticoagulation, hearts from overnight-fasted mice were subjected to the ex vivo Langendorff model perfused with phosphate-free Krebs-Henseleit buffer (in mM: 118 NaCl, 25 NaHCO3, 5.3 KCl, 2.5 CaCl2, 1.2 MgSO4, 0.5 EDTA, 5 pyruvate, and 5 glucose). 31P-NMR spectra were collected as previously described (40). Briefly, after a 20-min control perfusion, the phosphate-free Krebs-Henseleit buffer was switched to a buffer containing 3 mM of the glucose analog 2-deoxy-d-glucose (2-DG), instead of glucose, and 1.2 mM KH2PO4 to replenish the phosphate pool. All data were normalized to γ-ATP peak area for quantification. The transport rate (mmol·l−1·min−1) of 2-DG was assessed by the time-dependent accumulation of 2-DG-phosphate (2-DG-P). The rates were measured both before and after the addition of insulin (2 mU/ml) to the perfusate (40).
Sections were prepared as previously described (29). Periodic Acid-Schiff (PAS) staining was performed with the PAS Staining System (Sigma), according to the manufacturer's instructions. Some samples were pretreated with α-amylase before PAS staining to confirm the specificity of positive staining. Biochemical assay demonstrated myocardial glycogen content with the amount of glucose released from glycogen after alkaline extraction as previously described (40).
Basal glucose uptake and glycogen synthesis in vivo.
Basal rates of glucose uptake and glycogen synthesis in heart were assessed in awake TG and littermate NTG mice (n = 4–5 for both groups) as previously described (20). After an overnight fast, [3-3H]glucose (0.1 μCi/min; PerkinElmer Life and Analytical Sciences, Boston, MA) was continuously infused for 2 h, and 2-deoxy-d-[1-14C]glucose (2-[14C]DG) was administered as a bolus (10 μCi) at 30 min before the end of experiments. Plasma samples were taken at 5-min intervals during the last 30 min of experiments to measure the concentrations of 2-[14C]DG and [3H]glucose. At the end of study, mice were anesthetized with pentobarbital sodium injection, and heart samples were rapidly taken for biochemical assays. Because 2-DG is a glucose analog that is phosphorylated but not further metabolized, glucose uptake in heart can be estimated by determining the myocardial content of 2-[14C]DG-6-P and plasma 2-[14C]DG profile, which was fitted with a double exponential or linear curve using MLAB (Civilized Software, Bethesda, MD). Cardiac glycogen synthesis in vivo was estimated by determining the incorporation of 3H into myocardial glycogen as previously described (20).
Cardiac muscle fractionation.
Sucrose gradient purification was used to isolate cytosolic, intracellular vesicle pools and plasma membrane fractions from cardiac muscle, as previously described (43). In brief, hearts from 12- to 16-wk-old male mice were removed from deeply anesthetized animals by means of pentobarbital sodium (60 mg/kg ip), washed with cold PBS, and homogenized in cold lysis buffer (20 mM HEPES, 250 mM sucrose, 1 mM EDTA, 5 mM benzamidine, 1 μM aprotinin, 1 μM pepstatin, 1 μM leupeptin, and 1 mM phenylmethylsulfonyl fluoride, pH 7.4). The homogenate was then centrifuged at 2,000 g for 10 min, and supernatants removed and centrifuged at 9,000 g for 20 min. Pellets were resuspended in 500 μl of PBS with protease inhibitors, as in the lysis buffer, and kept as a part of plasma membrane fraction (P1). Supernatants were then centrifuged at 180,000 g for 90 min, and the supernatants from this centrifugation were removed and kept as the cytosolic fraction. The remaining pellets were resuspended in 500 μl of PBS with protease inhibitors. Protein concentration was determined, and the same amount of protein from each sample was loaded on a 10–30% (wt/wt) continuous sucrose gradient and centrifuged at 48,000 rpm for 55 min in an SW-50 rotor. Gradients were separated into 26 fractions, starting from the bottom of the tube (heavy fraction). The pellet was resuspended in PBS and named P2 (plasma membrane fraction 2). Each experiment was done with a pair of littermate TG and NTG mice. Fractions isolated from TG and NTG hearts were subjected in parallel to SDS-PAGE (12%) and transferred to a single nitrocellulose membrane for immunoblotting.
Frozen hearts were crushed in liquid nitrogen and homogenized in cold lysis buffer, as previously described (29). After the samples were lysed, SDS-PAGE was performed under reducing conditions on 10 or 12% separation gels with a 4% stacking gel. Proteins were transferred to nitrocellulose membranes. Membranes were incubated with primary antibodies to GLUT1 (FabGennix), GLUT4 (FabGennix), GSK-3β (Cell Signaling), phospho-Ser9 GSK-3β (Cell Signaling), β1-integrin (Santa Cruz Biotechnology), SCAMP2 (Abcam), HA (12CA5, Roche), or hexokinase type II (IREAGQR, Chemicon) and subsequently with horaseradish peroxidase-conjugated secondary antibodies. Signal was detected using enhanced chemiluminescence and band intensity quantified using NIH Image.
Total RNA was isolated from TG564 and NTG mice and quantitative RT-PCR (QRT-PCR) performed using the Brilliant One-Step QRT-PCR kit (Stratagene, La Jolla, Ca) containing SYBR Green I (Sigma), as described previously (6). Accumulation of PCR product was monitored in real time and the crossing threshold (CT) was determined with the Mx4000 system (Stratagene). Fold change in gene expression was determined using the ΔCT method with normalization to GAPDH (6, 32). QRT-PCRs were performed with the following sets of primers: for GLUT4, forward 5′-ATGGCTGTCGCTGGTTTCTC and reverse 5′-TAAGGACCCATAGCATCCGC; for GAPDH, forward 5′-TGGTGAAGCAGGCATCTGAG and reverse 5′-TGCTGTTGAAGTCGCAGGAG.
All data are presented as means ± SE from at least three independent experiments and were compared using a two-tailed Student's t-test. The null hypothesis was rejected at P < 0.05.
Glucose uptake in ex vivo myr-Akt transgenic mice.
To minimize variability due to endogenous insulin release, we measured basal and insulin-stimulated glucose uptake in hearts from overnight-fasted TG and NTG mice. Cardiac glucose uptake was assessed in an ex vivo Langendorff perfusion system using 31P-NMR spectroscopy and glucose analog 2-DG. At baseline, the rate of 2-DG-P accumulation was significantly higher in TG than in NTG mice (Fig. 1A, top). Cumulative data demonstrated that the basal glucose uptake in TG mice was greater than that seen in NTG mice (0.87 ± 0.18 vs. 0.18 ± 0.10 mmol·l−1·min−1, P < 0.05, Fig. 1B). After baseline 2-DG-P accumulation was measured for 40 min, insulin (final concentration = 2 mU/ml) was added to the perfusate and glucose uptake assessed in the same way. As we expected, glucose uptake in NTG hearts was enhanced by insulin more than 18-fold. In contrast, insulin-stimulating glucose uptake in the heart of TG mice demonstrated much less response than that seen in NTG mice (1.9 vs. 18.1-fold, Fig. 1, A and B). Although glucose uptake after insulin perfusion in TG mice is almost twice bigger than that in baseline, there is no statistical difference (P = 0.06, Fig. 1B).
Glycogen content in myr-Akt TG mice.
To examine whether enhanced glucose uptake resulted in increased storage of glycogen, we performed histochemical staining and biochemical analyses. PAS staining of hearts from fasted NTG mice did not reveal significant glycogen stores (Fig. 2A). In contrast, PAS staining of hearts from both male and female TG20 mice revealed substantial glycogen accumulation. Treatment with α-amylase completely abrogated this staining, confirming its specificity (Fig. 2A). As noted above, the X-linked transgene in TG20 mice results in chimeric transgene expression in hearts from female mice (29). Consistent with this, fewer cardiomyocytes in female TG20 hearts stained positive with PAS compared with those in male TG20 mice, although staining in male TG20 mice was also somewhat heterogenous (Fig. 2A). These data suggest that the effect of myr-Akt on glycogen deposition is likely a cell autonomous effect of the transgene rather than a secondary effect of the hypertrophy seen in both male and female TG20 mice. PAS staining in TG564 mice appeared similar to that seen in TG20 male mice (data not shown). To quantitate glycogen content, we measured the amount of glucose released by alkaline extraction (40). Glycogen content in Akt TG mice was more than sixfold higher than that seen in NTG littermates (37.1 ± 1.2 vs. 5.5 ± 2.6 μmol glucose/g wet wt, P < 0.001; Fig. 2B). We next examined expression and phosphorylation (Ser9) of GSK-3β, which causes inhibition of GSK-3 and consequently increases the activity of glycogen synthase. To minimize the contribution of leukocytes and other circulating factors, we studied GSK-3β phosphorylation in the heart after a 20-min ex vivo perfusion in the Langendorff system. The level of phosphorylation in TG564 mice was substantially higher than that in NTG littermates (Fig. 2C).
Basal glucose uptake and glycogen synthesis in vivo.
Akt TG mice show cardiac hypertrophy (2.3-fold greater than NTG in TG564) (29), which could cause an overestimation of glucose uptake in the ex vivo perfused heart system of NMR. Therefore, we also evaluated cardiac glucose uptake in vivo using a combination of continuous [3-3H]glucose infusion and bolus 2-[14C]DG injection in awake mice. In vivo, the basal rate of cardiac glucose uptake in Akt TG mice was substantially higher than that in NTG mice (320.6 ± 56.3 vs. 49.1 ± 10.6 nmol·g−1·min−1, P < 0.005; Fig. 3A). This change in basal glucose uptake in vivo agrees well with that measured by NMR in ex vivo perfused hearts (4.8-fold increase ex vivo vs. 6.5-fold in vivo). By use of these samples, basal glycogen synthesis in vivo was estimated as the rate of 3H incorporation into myocardial glycogen. The results indicate that basal glycogen synthesis in hearts from Akt TG mice was markedly elevated compared with the NTG mice (20.7 ± 4.0 vs. 4.2 ± 2.4 nmol·g−1·min−1, P < 0.01; Fig. 3B). Increased glycogen synthesis in vivo was also consistent with the result of accumulated glycogen content in the hearts of Akt TG mice in histological analysis and biochemical assay.
GLUT1 and GLUT4 expression in the myr-Akt TG mice.
To identify potential mechanisms contributing to glucose uptake in the heart, we examined the expression of the two main cardiac glucose transporters, GLUT1 and GLUT4. Total expression of GLUT4 was significantly reduced in whole cell lysates from TG mice compared with NTG mice (∼32% reduced in TG, P < 0.05; Fig. 4A). In contrast, overall GLUT1 expression appeared comparable between TG and NTG (Fig. 4A). Because GLUTs enhance glucose uptake functionally at the plasma membrane, we analyzed the distribution of both GLUT1 and -4 in hearts by sucrose gradient fractionation. This method distinguishes proteins associated with intracellular membrane structures such as intracellular vesicles from those associated with the plasma membrane (23, 43). Immunoblotting for the sarcolemmal protein β1-integrin confirmed localization of plasma membrane proteins to the P2 (and to a lesser extent P1) fractions without cross-contamination in the intracellular or cytosolic fractions (Fig. 4B). Secretory carrier-associated membrane proteins (SCAMPs) are located in cell surface recycling vesicles that contain GLUT4 (43). Immunoblotting for SCAMP2 confirmed the distribution of these small vesicles containing GLUT1 and GLUT4 in the heart (Fig. 4B). Immunoblotting for the HA epitope encoded by the Akt transgene revealed the presence of myr-Akt in both the plasma membrane and the cytosolic fractions (Fig. 4B), consistent with prior reports (4). In NTG mice, immunoreactive GLUT4 was evident in both the plasma membrane and intracellular vesicular fractions but not in the cytosol (Fig. 4B). Cumulative densitometric quantitation demonstrated that, at baseline, GLUT4 was predominantly in the intracellular vesicular fractions rather than in the plasma membrane fractions (3.8-fold higher in intracellular vesicular fractions; Fig. 4C), whereas GLUT1 protein was comparable in both sites (Fig. 4C). Sucrose gradient showed that in Akt TG hearts GLUT4 was substantially reduced in both the plasma membrane (21.6 ± 5.8 vs. 100, P < 0.01) and the intracellular vesicular fractions (137.6 ± 122.8 vs. 379.5 ± 156.5, P < 0.05; Fig. 4, B and C). The level of GLUT1 protein in Akt TG hearts was slightly lower in both the plasma membrane and the intracellular vesicular fractions than that in NTG hearts, although these differences did not achieve statistical significance (Fig. 4, B and C). To address the mechanism of increased basal glucose uptake in Akt TG hearts, we evaluated the distribution of GLUT1 protein in TG hearts between the plasma membrane (PM) and intracellular vesicular (IV) compartments. No statistically significant difference was seen in the PM/IV ratio of immunoreactive GLUT1 [1.21 ± 0.25 vs. 1.59 ± 0.47, NTG vs. TG, P = not significant (ns)]. To study a regulation of GLUT4 expression in the heart of myr-Akt TG mice, we examined the mRNA level of GLUT4 by QRT-PCR (Fig. 4D). By use of ΔCT, quantitative analysis was done with normalization to GAPDH. Cumulative data demonstrated that the mRNA level of GLUT4 in TG564 mice was lower than that in NTG mice (0.73 ± 0.05 vs. 1.02 ± 0.11, P < 0.05; Fig. 4D).
At baseline, GLUT4 expression in total lysate and sucrose gradient fractionation from the hearts of TG mice was significantly lower than that in NTG mice (Fig. 4, A–C). The reduction in intracellular or vesicular pools of GLUT4 likely contributes to the blunted insulin response seen in these mice. To confirm the response to insulin on the distribution of GLUT4 expression, we performed sucrose gradient fractionation in the ex vivo heart stimulated with a 20-min insulin perfusion, which was done with the same condition to 31P-NMR spectroscopy (Fig. 1). In NTG mice, the GLUT4 expression in the intracellular vesicular fraction was significantly reduced by insulin exposure (387.6 ± 153.4 vs. 172.2 ± 110.4, P < 0.05; Fig. 5). In addition, the PM/IV ratio was dramatically increased after 20-min insulin exposure (0.25 to 0.79), suggesting the translocation of GLUT4 from the intracellular vesicular fraction to the plasma membrane, although an increase of GLUT4 expression in the plasma membrane was not statistically significant (100 vs. 135.1 ± 11.2, unstimulated vs. insulin-stimulated hearts). In contrast, the PM/IV ratio in TG mice by insulin stimulation was comparable to that in unstimulated hearts (0.62 vs. 0.51, unstimulated vs. insulin-stimulated hearts) in addition to there being no change in the GLUT4 expression in the intracellular vesicular fraction (151.7 ± 71.0 vs. 120.1 ± 55.6, unstimulated vs. insulin-stimulated hearts) and in the plasma membrane (94.8 ± 14.2 vs. 68.0 ± 13.7, unstimulated vs. insulin-stimulated hearts). These data demonstrated less capacity of GLUT4 recruitment to the plasma membrane by acute insulin stimulation in TG mice.
Hexokinase II, relevant to glucose uptake in the myr-Akt TG mice.
To identify potential contributors to the increased basal glucose uptake seen in Akt TG mice, we examined hexokinase, which controls the first committed step in glycolysis (26). We observed no difference in hexokinae II expression between fasted TG and NTG mice (Fig. 6). Similarly, hexokinase was not different in fed TG and NTG mice (data not shown).
In the present study, we examined the effects of chronic Akt activation on glucose uptake, glycogen storage, and relevant glucose transporters in the heart. We found that chronic cardiac activation of Akt led to a substantial increase in the rate of basal glucose uptake but a blunted response to insulin. These changes were associated with an increase in glycogen deposition (>6-fold) and synthesis (5-fold). Studies in chimeric hearts of female X-linked transgenic Akt mice suggest that increased glycogen deposition occurs as a cell autonomous effect of transgene expression. Although acute Akt activation appears to increase glucose uptake predominantly through enhanced sarcolemmal GLUT4 (30), chronic Akt activation actually led to a significant decrease in total expression of GLUT4. Sucrose gradient showed that intracellular GLUT4 was substantially reduced in addition to a modest reduction of intracellular pools of GLUT1 in chronic Akt activation. Because translocation of these pools to the plasma membrane mediates the acute increase in glucose uptake seen in the heart after insulin stimulation (2, 12), it seems likely that this reduction contributes to the loss of insulin responsiveness that we observed. In fact, the maximal insulin-stimulated 2-DG uptake in hearts of TG mice was only one-half of that in NTG mice, reflecting a decrease in capacity of GLUT4 (32% in whole lysates and 64% in the intracellular vesicular fractions).
The observed increase in glycogen deposition in Akt TG hearts likely reflects the persistently increased glucose uptake in these hearts, which in other settings appears sufficient to increase glycogen content in the heart (40). However, other signaling effects could also contribute. For example, the activation of Akt by insulin inactivates GSK-3β through Ser9 phosphorylation, thereby enhancing activation of glycogen synthase (7, 8). In fact, after ex vivo perfusion, we did observe an increase in GSK-3β phosphorylation (Fig. 2C), suggesting that this mechanism could contribute to the increased glycogen in the heart of TG mice. Interestingly, previous studies (29, 37) with nonperfused, freshly isolated hearts did not detect this difference, perhaps reflecting a higher background contribution of noncardiomyocyte population, e.g., leukocytes, to GSK-3β phosphorylation. Recently, it has been reported that muscle-specific deletion of GLU4 in mice increases glycogen content despite reduced glucose transport (21). In that case, they demonstrated an increase of hexokinase II expression and protein phosphatase-1 (PP1) activity with dominating over GSK-3 effects. In the present study, GSK-3 effects seem more dominant on regulation of glycogen content, because we did see significant phosphorylation of GSK-3β with enhanced baseline glucose uptake but no increase of hexokinase II. We examined glycogen content and synthesis in Akt TG mice to see the effect of enhanced basal glucose uptake in chronic Akt activation. Although a series of ex vivo and in vivo experiments demonstrating an increase of basal glucose uptake and glycogen deposition supports the hypothesis that glycogen is a dominant fate of the increased glucose uptake in this setting, future studies will be directed at investigating other potential metabolic fates of glucose.
The dramatic reduction in GLUT4 expression that we observed is consistent with prior studies documenting decreased GLUT4 after prolonged insulin exposure in other tissues (36, 39). Interestingly, a similar phenomenon is also seen in pressure overload hypertrophy in the heart (41), a condition in which Akt is activated (11). Using QRT-PCR, we showed a decrease of the mRNA level of GLUT4 in a model of chronic Akt activation, which is consistent with a prior report demonstrating that chronic insulin treatment decreases GLUT4 mRNA in an in vitro cell line (13). In addition to transcriptional regulation, other mechanisms, such as modification at translation, might be involved in this setting, as reported previously (39). To examine insulin-induced GLUT4 translocation, hearts stimulated with insulin were subjected to sucrose gradient protein purification. In NTG mice, acute insulin stimulation dramatically reduced the expression of GLUT4 in the intracellular vesicular fractions and increased the PM/IV ratio, although an increase of GLUT4 expression in the plasma membrane did not achieve statistical significance. On the other hand, the expression level of GLUT4 in the intracellular vesicular fractions in myr-Akt TG mice was not changed by insulin. These data suggest that chronic Akt activation reduces the expression level of GLUT4, especially in insulin-responsive compartments.
We did not observe an increase of GLUT1 expression previously noted in these settings (24, 36, 39, 41). Moreover, sucrose gradient fractionation demonstrated that GLUT1 in Akt TG hearts was slightly lower in both the plasma membrane and the intracellular vesicular fractions than that in NTG hearts, although these differences did not achieve statistical significance. Because GLUT1 in the heart, as with GLUT4, is recruited to the plasma membrane in response to insulin (2, 35), a reduction of both GLUT4 and GLUT1 in intracellular pools likely accounts for the blunted insulin response seen with chronic Akt activation.
Although GLUT4 is more abundant than GLUT1 in both the heart and skeletal muscle, the ratio of GLUT1 to GLUT4 is much higher in the heart than in skeletal muscle (12, 27). Furthermore, the localization of GLUT1 at baseline, seen predominantly in the plasma membrane, suggests that GLUT1 plays an important role in basal glucose uptake in the heart (35). Hearts from NTG showed that GLUT1, but not GLUT4, was predominantly localized in the plasma membrane, consistent with previous reports (12, 35). To address the mechanism of increased glucose uptake in Akt TG mice, we compared the distribution of GLUT1 in TG mice with that in NTG with the ratio of PM/IV immunoreactive protein. No difference was seen in this ratio, suggesting that GLUT1 did not account for the increased basal glucose uptake in TG mice. The present studies cannot exclude the possibility that the intrinsic activity of GLUT1 and GLUT4 might be altered and thereby contribute to the enhanced basal glucose uptake (18).
Although acute Akt activation shows beneficial effects on the ischemic heart (28), our findings raise the possibility that chronic activation of these pathways may cause adverse effects on cardiac function in some settings such as ischemia-reperfusion injury. Indeed, glycogen deposition and cardiac insulin resistance can be associated with cardiomyopathy in different settings (1, 31). Tissue insulin resistance has previously been associated with a variety of signaling abnormalities, including IRS-1/2 (see review in Ref. 38). In fact, we (32) have recently reported that chronic Akt activation impairs cardiac functional recovery and increases injury in an ex vivo model of ischemia-reperfusion injury through negative feedback inhibition of IRS-1/PI3K signaling. Reductions in the levels of both IRS-1 and IRS-2 culminate in impaired activation of PI3K in response to ligands such as insulin or IGF-I and reperfusion (32). Thus we infer that the combination of impaired upstream insulin signaling together with the alterations in GLUT1 and GLUT4 levels is likely to account for the diminished ability of Akt TG hearts to increase glucose uptake after insulin stimulation. This “cardiac insulin resistance” may have relevance for heart failure (16) or other conditions associated with chronic activation of Akt signaling. In these settings, Akt activation may reflect increased expression of growth factors such as IGF-I induced as an initially compensatory response (19, 34).
Baseline function in hearts from Akt TG mice manifested moderate increases in both positive and negative dP/dt measured ex vivo, whereas other major cardiac functions, such as the mean left ventricular systolic pressure (LVSP) and end-diastolic pressure (LVEDP) were comparable to those in NTG (32). However, functional recovery and injury after ischemia-reperfusion were dramatically worse in hearts from Akt TG mice compared with NTG (32). Because glucose is utilized as a critical energy source during cardiac ischemia, the inability of Akt TG hearts to enhance glucose uptake may well contribute to the observed deleterious effects of chronic Akt activation in this model. The current study demonstrating less expression of GLUT4 in Akt TG mice suggests an additional mechanism contributing to this cardiac deterioration after ischemia-reperfusion injury in addition to Akt-driven feedback inhibition of IRS-1(32).
Although the lower levels of GLUT1 and GLUT4 in the intracellular vesicular fractions likely contribute to the loss of insulin responsiveness, the molecular explanation for enhanced basal glucose uptake in Akt TG hearts has not been identified in the current study. Hexokinase II is known to increase glucose uptake in the heart (3, 14), but it was not significantly different between Akt TG and NTG hearts. A variety of other glucose transporters and enzymes contribute to glucose uptake in skeletal and heart muscle. For example, although GLUT8 expression in the heart is less than in other tissue such as testis, the high efficiency of this glucose transporter suggests that it may play a role in glucose uptake in the heart (10). However, GLUT8 was not increased in hearts from Akt TG mice (data not shown). It is possible that enhanced expression of novel cardiac glucose transporters contributes to the observed increase in basal glucose uptake. In this context, the Akt TG mice could provide a valuable substrate for investigating the possibility of novel glucose transportors and/or modulatory pathways by facilitating subtractive screens to identify such mechanisms.
In summary, the present data demonstrate that chronic Akt activation in the heart led to a substantial increase in the basal glucose uptake and glycogen deposition but blunted the response to insulin. Decreases in the insulin-responsive intracellular pools of GLUT1 and GLUT4 could represent an additional mechanism of insulin resistance on chronic Akt activation in the heart, in addition to IRS-1/PI3K inhibition (32). Additional studies will be necessary to define the contribution of these mechanisms to pathophysiological conditions in the heart such as cardiomyopathy or heart failure.
This work was supported in part by grants from the National Institutes of Health (HL-04250 to T. Matsui; HL-58119 and DK-47425 to M. J. Charron; U24 DK-59635 to J. K. Kim; HL-59246 and HL-67970 to R. Tian; and HL-59521 and HL-61557 to A. Rosenzweig), American Diabetes Association (1-04-RA-47 to J. K. Kim), and The Robert Leet and Clara Guthrie Patterson Trust Award to J. K. Kim. T. Matsui is a recipient of a Grant-in-Aid from the American Heart Association Northeast Research Consortium. I. Luptak is a recipient of the American Heart Association Northeast Consortium Fellowship. A. Rosenzweig and R. Tian are Established Investigators of the American Heart Association.
We thank Drs. Konstantin V. Kandror and Tatyana A. Kupriyanova for their helpful advice.
Present address of E.-G. Hong and J. K. Kim: Department of Cellular and Molecular Physiology, Pennsylvania State University College of Medicine, Hershey, PA.
Present address of T. Matsui, L. Li, and A. Rosenzweig: Cardiovascular Division, Beth Israel Deaconess Medical Center, Boston, MA.
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 © 2006 by American Physiological Society