We have previously reported that chronic leptin administration (2 wk) increases fatty acid (FA) oxidation and triacylglycerol hydrolysis in rodent soleus muscle. Acute stimulation of AMP-activated protein kinase (AMPK) results in a repartitioning of FA toward oxidation and away from esterification in rodent soleus muscle and has recently been shown to be responsible, at least in part, for the acute stimulatory effect of leptin on FA oxidation. Therefore, we hypothesized that the effects of chronic leptin treatment on muscle FA metabolism are mediated in part through an increased expression and/or activation of AMPK and a subsequent phosphorylation of acetyl-CoA carboxylase and a decrease in malonyl-CoA content. Female Sprague-Dawley rats were infused for 2 wk with leptin (0.5 mg · kg−1 · day−1) using subcutaneously implanted mini-osmotic pumps. Control and pair-fed animals received saline-filled implants. Leptin levels were elevated approximately fourfold (P < 0.001) in treated animals, relative to controls. Chronic leptin treatment resulted in an ∼2- to 3-fold greater protein expression of AMPK catalytic (α2) and regulatory (β2) units as well as a 1.5- to 2-fold increase in Thr172 phosphorylation of AMPK in both soleus and white gastrocnemius muscles. The increased expression/phosphorylation of AMPK was not the result of an altered energy status of the muscle. Correspondingly, there was also a 1.5- to 2-fold increase in acetyl-CoA carboxylase (ACC) phosphorylation after leptin treatment in soleus and white gastrocnemius. In spite of the measured increase in ACC phosphorylation after leptin treatment, we were unable to detect a decrease in resting malonyl-CoA content in either muscle. However, taken as a whole, our data support recent evidence in rodent muscle that leptin stimulates FA oxidation through stimulation of AMPK and a subsequent downregulation of ACC activity.
- osmotic mini-pumps
- fat oxidation
- acetyl-coenzyme A carboxylase
- adenosine monophosphate-activated protein kinase
the chronic administration of leptin has been demonstrated to have pronounced metabolic effects in rodents, including improvement of insulin sensitivity and glucose uptake in skeletal muscle (32). In addition, leptin administration leads to a reduction of triacylglycerol (TG) stores in several tissues, including skeletal muscle (29, 30, 32), which is related to an increased capacity to hydrolyze intramuscular TG and oxidize fatty acids (FA), while the capacity to transport FA across the sarcolemma is decreased (30, 31). In contrast, increased intramuscular TG storage and impaired FA oxidation are strongly associated with the development of insulin resistance (15,23). Thus the improved insulin sensitivity observed in rodents after chronic leptin administration may be due, in part, to the improvement in FA metabolism, leading to decreased storage of tissue TG.
Although leptin's acute and chronic effects on muscle FA metabolism are well documented, the mechanism(s) underlying these effects are not well understood. In particular, we recently demonstrated that an increased ability to oxidize FA and hydrolyze intramuscular TG after chronic leptin administration in rodents was not the result of alterations in indicators of aerobic capacity (citrate synthase), mitochondrial capacity to oxidize FA (β-hydroxyacyl dehydrogenase), or increased expression of hormone-sensitive lipase (30). Recently, considerable research has focused on the role of AMP-activated protein kinase (AMPK) as an acute regulator of substrate metabolism in skeletal muscle. AMPK is activated by an increase in the AMP-to-ATP ratio, as well as a decrease in phosphocreatine (PCr; see Refs. 9 and 10), and has been shown to stimulate both glucose uptake and FA oxidation (1, 3, 16, 19) in this tissue. Acute activation of AMPK using the cell-permeable compound 5-amino-4-imidazolecarboxamide ribofuranoside (AICAR) has been shown to repartition FA toward oxidation and away from storage in oxidative murine muscle (19), an effect identical to that observed with leptin (18). Indeed, a role of AMPK in mediating leptin's acute stimulatory effect on FA oxidation was recently demonstrated in murine soleus (Sol) muscle (17).
However, there is no evidence to date demonstrating whether leptin's chronic stimulatory effects on muscle FA metabolism might be due, at least in part, to the increased expression and/or activation of AMPK. The antidiabetic drugs rosiglitazone and metformin, which have beneficial effects on blood lipids, have been shown to stimulate AMPK in muscle (8, 33). Leptin treatment, which improves insulin sensitivity in rodents (32), has also been demonstrated to improve insulin sensitivity and decrease muscle TG in leptin-deficient humans suffering from severe lipodystrophy (25). Taken together, these data strongly suggest that leptin may be exerting its chronic effects on muscle FA metabolism and, ultimately, on insulin sensitivity by increasing the activity of muscle AMPK.
Thus the primary objective of the present study was to determine whether chronic (2 wk) administration of leptin to rodents would increase the expression and/or activation (phosphorylation) of AMPK, as well as the phosphorylation of one of its target enzymes, acetyl-CoA carboxylase (ACC), in oxidative (Sol) and glycolytic [white gastrocnemius (WG)] muscles. Furthermore, we also measured indicators of the resting energy status in each fiber type [ATP, PCr, creatine (Cr), ADP, AMP, and IMP], as well as metabolites associated with altered muscle FA metabolism (acetyl-CoA and malonyl-CoA).
Female Sprague-Dawley rats (Charles River, Quebec, Canada) weighing between 180 and 225 g were randomly assigned to one of the following groups: ad libitum-fed saline-treated, pair-fed saline-treated (PF), or leptin-treated (n = 6–9/group) animals. Animals were anesthetized with halothane, and a small incision was made through the skin on the upper region of the back between the scapulae. Mini-osmotic pumps (2ML2; Durect, Cupertino, CA) were filled with either sterile PBS (control, PF) or murine leptin (donated by Amgen, Thousand Oaks, CA) and inserted through the incision. The incision was closed with a single surgical staple. A leptin dosage of 0.5 mg · kg−1 · day−1was used as in previous studies (2, 11, 30). Animals were maintained on a reverse 12:12-h light-dark cycle, and water was freely accessible to all groups. Food intake was ad libitum for both the control- and leptin-treated animals, whereas PF-treated animals were fed the same amount of chow as the leptin-treated animals. Body mass was monitored over the 2-wk treatment period. The Committees on Animal Care at the Universities of Waterloo and Guelph approved all procedures.
Blood was collected at the completion of treatments (2 wk) via cardiac puncture after the excision of the Sol and WG muscles. Blood samples were centrifuged at 8,000 g, and the serum was collected and stored at −80°C until further analysis. Muscles were immediately frozen upon being excised and were stored under liquid nitrogen. All tissue samples were taken while rats were in the fed state, between 0900 and 1100, to eliminate diurnal variability. Serum leptin and insulin were assayed in duplicate using RIA kits (Linco, St. Charles, MO). Serum glucose was assayed fluorometrically, and free fatty acids (FFA) were assayed using a colorometric kit from Wako (Wako Chemicals).
Protein expression of regulatory (AMPKβ2) and catalytic (AMPKα2) subunits of AMPK in Sol and WG muscles and the phosphorylation (i.e., activation) of AMPK were measured via Western blotting. Briefly, muscles were homogenized in solution containing 210 mM sucrose, 2 mM EGTA, 40 mM sodium chloride, 30 mM HEPES, 5 mM EDTA, 2 mM phenylmethylsulfonyl fluoride, and protease inhibitors (1 μM pepstatin A, 2 μM leupeptin, and 20 mM 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate). Proteins were precipitated with 0.5 M potassium chloride and 25 mM tetrasodium pyrophosphate and spun down in an ultracentrifuge (Beckman XL-90) at 48,000 g for 75 min at 4°C. Protein pellets were rehomogenized in 10 mM Tris-1.0 mM EDTA, pH 7.4, and SDS (Bio-Rad, Mississauga, ON) was added to bring the final concentration to 5%; centrifugation followed at 1,000 g for 10 min at room temperature. Protein concentration was measured using a bicinchoninic acid protein assay. Thirty micrograms of muscle protein were separated with the use of standard SDS-PAGE (4% stacking gel, 10% running gel) and transferred to membranes (Immobilon polyvinylidene difluoride; Bio-Rad) by semidry transfer (20 V, 20 min). Membranes were then incubated with the various antibodies (3 μg/ml) as per the manufacturer's directions [anti-AMPKβ2 (BD Biosciences, Pharmingen, Mississauga, ON); anti-AMPKα2 (Upstate Biotech); anti-phospho-AMPK-Thr172 (Cell Signaling Technology, Beverly, MA)]. Immunoblot signals were generated via enhanced chemiluminescence (ECL, Amersham, Mississauga, ON) and captured on X-ray film (Amersham).
Seventy micrograms of protein were separated by SDS-PAGE (4% stacking gel, 8% running gel) and transferred to the nitrocellulose membrane (100 V for 60 min). Membranes were then blocked in Tris-buffered saline-Tween 20 (TBS-T) containing 5% nonfat dried milk for 2 h at room temperature with constant agitation followed by three washes in TBS-T. Membranes were then incubated overnight in 5% TBS-T milk containing the anti-phospho-ACC (Ser79; Upstate Biotech) at a dilution of 1:1,000. The next day, membranes were washed three times with TBS-T and incubated with goat anti-rabbit horseradish peroxidase-conjugated IgG (Upstate Biotech) at a dilution of 1:2,000 for 1.5 h and then washed three times with TBS-T. Phosphorylated ACC was then visualized using procedures described previously.
After the 2 wk of leptin administration, resting concentrations of acetyl-CoA, adenine nucleotides, and Cr were measured in neutralized HClO3 (0.5 N, 2 mM EDTA) extracts (pH = 7–7.4) in Sol and WG muscles, as previously described (4, 5, 27). Free contents of ADP and AMP were calculated from the near-equilibrium Cr kinase and adenylate kinase reactions, respectively, as outlined by Lawson and Veech (14). Malonyl-CoA concentrations were determined in neutralized HClO3 (0.5 N, 50 μM dithiothreitol) extracts (pH = 4–5) using reverse-phase HPLC (26).
Because of tissue requirements for Western blotting and metabolite analyses, we were unable to determine FA oxidation in isolated Sol muscles. However, we had previously determined that FA oxidation was increased in contracting Sol strips after chronic leptin treatment (30). In addition, because of technical difficulties in securing undamaged tendons in representative glycolytic muscles suitable for incubation (tibialis anterior, extensor digitorum longus, and epitrochlearis), FA oxidation was not determined in glycolytic muscle.
All data are reported as means ± SE. Results were analyzed using ANOVA procedures, and a Tukey's post hoc test was used to test significant differences revealed by the ANOVA. Significance was accepted at P ≤ 0.05.
Body Mass and Serum Leptin
Body mass was reduced (P = 0.01) in both leptin- and PF-treated animals (pooled mean, 236 ± 4 g), relative to control (272 ± 7 g) after 2 wk of treatment. Changes in serum leptin, insulin, FFA, and glucose as a consequence of the treatments are presented in Table 1.
Protein expression of catalytic (AMPKα2) and regulatory (AMPKβ2) subunits was significantly increased in Sol and WG muscles after 2 wk of chronic exposure to leptin (Figs. 1 and2). AMPKα2 expression was increased 2.1- and 1.9-fold relative to the control group in Sol (P < 0.001) and WG (P < 0.001), respectively. Similarly, AMPKβ2 expression was increased 2.8- and 2.4-fold relative to the control group in Sol (P < 0.001) and WG (P < 0.005). The expression of each AMPK subunit in both Sol and WG muscles was not statistically different between control and PF groups. Phosphorylation of AMPK-Thr172 was also increased significantly in Sol (1.5-fold, P ≤ 0.001) and WG (2-fold, P = 0.010) muscles with leptin treatment (Fig3), indicating an increase in the activity of AMPK at rest.
After leptin treatment, ACC phosphorylation (Fig4) was increased relative to control, ∼1.5-fold in WG (P = 0.06) and ∼2-fold in Sol (P = 0.027).
Energy status of the cell.
Indicators of cellular energy status were generally unaffected by chronic leptin treatment. PCr, Cr, total ADP, total AMP, and calculated free ADP and AMP concentrations were not statistically different after chronic leptin treatment in either Sol (Table2) or WG (Table3) muscles. ATP content was unaffected in Sol but was decreased in WG (P = 0.0097) after leptin treatment relative to PF. IMP content in Sol was significantly greater in the leptin group relative to both control (P = 0.0303) and PF (P = 0.0393) groups but was unaffected in WG.
Sol malonyl-CoA content (Table 4) was not affected by treatments, but WG malonyl-CoA was increased after leptin treatment relative to control (P = 0.0645) and PF (P = 0.0474) groups. Sol acetyl-CoA content (Table 4) was increased more than threefold with leptin relative to the control group but showed no significant change in WG with PF or leptin treatments. Muscle lactate was unaffected by the treatments in Sol and WG (data not shown).
This study is the first to demonstrate that chronic leptin administration increases the protein expression and phosphorylation of AMPK, as well as the phosphorylation of ACC, in rodent skeletal muscle. Increases in AMPK content and phosphorylation were observed in both oxidative (Sol) and glycolytic (WG) muscles and were not the result of an altered energy status of the cell. This corresponded to an increased phosphorylation of one of AMPK's primary targets, ACC, and presumably, a decrease in its activity. However, a corresponding decrease in the resting malonyl-CoA content was not observed in either muscle. Taken as a whole, our data support recent evidence in rodent muscle that leptin stimulates FA oxidation through stimulation of AMPK and a subsequent downregulation of ACC activity.
Role of AMPK as a Mediator of Leptin's Effects
Previous studies have demonstrated that activation of AMPK, by means of the cell-permeable compound AICAR, results in stimulation of FA oxidation (1, 12, 19). Stimulation of AMPK inhibits ACC while simultaneously activating malonyl-CoA decarboxylase (12,28) in oxidative muscle, resulting in a decrease in malonyl-CoA content and relief of carnitine palmitoyltransferase (CPT) I inhibition at the mitochondrial membrane. Recently, AMPK was identified as a critical part of the signaling cascade through which leptin acutely stimulates FA oxidation in skeletal muscle (17). We have preliminary evidence that iodotubercidin, an inhibitor of AMPK, prevents the acute, leptin-induced stimulation of FA oxidation in rat Sol (unpublished observations). Chronic leptin administration also results in increased FA oxidation in skeletal muscle and a significant reduction in intramuscular TG stores (30). However, before this study, the effects of chronic leptin administration on AMPK expression/activity had not been examined. Although we did not directly measure the activity of the catalytic units of AMPK (i.e., α1 and α2) in the present study, it has been demonstrated that Thr172 phosphorylation of AMPK is required for, and corresponds to, the actual enzyme activity (22,24). Furthermore, the activity of AMPKα2, and not AMPKα1, increases in response to AICAR, leptin, and contraction (6, 17, 21) and is directly related to Thr172 phosphorylation (6). Thus, given the significant increase in both α2 protein expression and Thr172 phosphorylation measured in the present study, it is likely that AMPK activity was increased. Furthermore, we were able to demonstrate that ACC, a primary target of AMPK, was phosphorylated in response to leptin treatment, further arguing that the actual activity of AMPK was increased. It should also be noted that the absolute increase in AMPK phosphorylation (1.5- to 2-fold) was actually less than the observed increase in AMPK protein content (2- to 3-fold). Thus, although the absolute amount of phosphorylated or activated AMPK increased as a function of leptin treatment, the fraction of total AMPKα2 in its phosphorylated state did not.
The present results extend the recent findings of Minokoshi et al. (17), who were the first to directly demonstrate that leptin acutely stimulates FA oxidation through the activation of AMPKα2 in oxidative Sol but not glycolytic muscle. Interestingly, in the present study, chronic leptin treatment resulted in altered AMPK and ACC phosphorylation in glycolytic and oxidative muscle, suggesting that leptin's chronic stimulatory effects on lipid metabolism might differ from its acute effects. However, the absence of FA oxidation measurements in WG in this study prevents such a conclusion. Although it would have been interesting to compare our observed lack of changes in malonyl-CoA content with the study from Kahn's group, malonyl-CoA was not determined in the former study. Furthermore, in the study by Minokoshi et al. (17), it was determined that leptin's effects on muscle FA metabolism were mediated both directly (early response) and indirectly through stimulation of the sympathetic nervous system (later response). It cannot be determined from the present study whether leptin's chronic effects on muscle FA metabolism are mediated through direct or indirect effects, which might also include secondary changes in other hormones, i.e., decrease in insulin. Nevertheless, our data are in agreement that leptin stimulates muscle FA oxidation, at least in part, through stimulation of the AMPK-ACC-malonyl-CoA axis.
Leptin treatment, which improves insulin sensitivity in rodents (32), has recently been demonstrated to improve insulin sensitivity and decrease muscle TG in leptin-deficient patients suffering from severe lipodystrophy (25). Similarly, the antidiabetic drugs rosiglitazone and metformin have also been shown to stimulate AMPK in both myotubes and human muscle (8, 33). Thus it is tempting to speculate that leptin's ability to improve insulin sensitivity in skeletal muscle is due in part to the stimulation of AMPK and consequently a repartitioning of FA toward oxidation and away from esterification, leading to a decrease in stored lipid. It is also possible that chronic stimulation of AMPK affects other processes, including FA transport across the sarcolemma and activation of hormone-sensitive lipase, which may contribute to the lowering of tissue TG.
Effects of Leptin Administration on Metabolism
Because of tissue requirements for Western blotting and metabolite analyses, we were unable to directly confirm our previous observations that 2 wk of leptin administration increases the capacity of rat Sol muscle to oxidize FA and hydrolyze intramuscular TG stores (30). Furthermore, the effect of chronic leptin administration on FA oxidation in glycolytic muscle was not determined in the present study. We recognize that this measurement would add further insight to the role of AMPK as a mediator of leptin's effects. However, we have previously shown that a chronic alteration in FA metabolism, whether induced by aerobic training (7) or leptin implants (30), is not necessarily evident in resting (quiescent) Sol strips because of the low metabolic rate. It is only when the metabolic rate is increased, via electrical stimulation, that the increased capacity for FA oxidation becomes evident. However, several of the glycolytic muscles are more difficult to remove with intact tendons at both ends, which is absolutely essential to retain viability during contraction. We have previously attempted to contract isolated glycolytic muscles and have encountered a rapid rate of fatigue that clearly compromises the interpretation of the metabolic findings.
Muscle malonyl-CoA content was not decreased in either Sol or WG muscles after leptin treatment; in fact, malonyl-CoA was increased in WG. However, the facts that 1) AMPK expression/phosphorylation and ACC phosphorylation were increased in both muscle types and 2) that our previous findings do demonstrate an increase in FA oxidation in Sol after chronic leptin treatment (30) indicate that our lack of observed change in malonyl-CoA should be viewed cautiously. Our finding of unaltered malonyl-CoA content, despite the increased expression and phosphorylation of AMPK, and subsequent phosphorylation of ACC, is unexpected, given that acute activation of AMPK in murine Sol by AICAR has been shown previously to decrease malonyl-CoA content (1,12). However, we have also been unable to detect a decrease in malonyl-CoA content in Sol muscle acutely exposed to leptin, despite the fact that we can demonstrate a significant increase in Sol malonyl-CoA during exposure to insulin (unpublished findings in collaboration with S.J. Peters, N. Ruderman and A. Saha). There is recent evidence in rodent muscle for a malonyl-CoA-insensitive pool of CPT I (13), supporting the contention that a decrease in malonyl-CoA may not be essential to enhance FA flux across the mitochondrial membrane. Alternatively, and most feasibly, it is possible that our determinations of total cellular malonyl-CoA content may not reveal changes in malonyl-CoA in localized compartments near CPT I, as has been suggested previously (15).
Acutely, AMPK is thought to be largely regulated by alterations in the energy status of the cell, including a decrease in PCr and an increase in the ratio of AMP to ATP. Acute activation of AMPKα2 by leptin also corresponded to a significant increase in muscle AMP content (17) in mouse Sol. In the present study, we determined whether the chronic upregulation of AMPK expression and its phosphorylation state also corresponded to a change in the resting energy charge of the muscle. Although there was a significant decrease in ATP in WG, and an increase in IMP in Sol after leptin treatment, all other indicators of energy charge were unaffected, including PCr content and both measured total and calculated free content of ADP and AMP. However, this appears to be an area of some controversy. Metformin, which stimulates AMPK in muscle and myotubes, has been reported to increase the cellular ADP-to-ATP ratio, decrease PCr content in human muscle (20), and have no effect on the cellular energy charge in myotubes (8). Thus a decrease in the cellular energy charge may not be essential for the chronic upregulation of AMPK activity.
This is the first report that chronic leptin treatment in rodents results in both increased expression and phosphorylation of AMPK. On the strength of previous evidence establishing a link between acute AMPK stimulation and an increase in FA oxidation, as well as between chronic leptin administration and increased muscle FA oxidation, it is our belief that leptin's chronic effects on FA metabolism are also mediated, in part, through an increase in AMPK activity. Interestingly, the activation of AMPK through chronic leptin treatment does not appear to require a change in the chronic energy status of the muscle. Despite our inability to detect decreases in muscle malonyl-CoA content, the fact that we were able to detect an increase in the phosphorylation of ACC suggests that chronic leptin treatment stimulates the AMPK-ACC-malonyl-CoA axis. Finally, given that antidiabetic drugs, such as rosiglitazone and metformin, also stimulate AMPK in skeletal muscle, our results imply that leptin's ability to improve insulin sensitivity in both rodents and humans may be a result of increased AMPK activity.
We gratefully acknowledge the technical assistance and advice of Drs. J. Rosenfeld and Jeff Richards (McMaster University) regarding the HPLC determinations of malonyl-CoA. We also acknowledge the generous donation of murine leptin by Amgen (Thousand Oaks, CA) for these studies.
This study was funded by the Natural Sciences and Engineering Research Council of Canada (D. J. Dyck), the Heart and Stroke Foundation of Ontario (NA-4604; J. W. E. Rush), and the Canadian Institutes of Health Research (MOP-48184; J. W. E. Rush). G. Steinberg was supported by a Natural Science and Engineering Research Council of Canada Postgraduate Scholarship.
Address for reprint requests and other correspondence: D. J. Dyck, Dept. of Human Biology and Nutritional Sciences, Univ. of Guelph, Guelph, Ontario, Canada N1G 2W1 (E-mail:).
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.
First published November 19, 2002;10.1152/ajpendo.00318.2002
- Copyright © 2003 the American Physiological Society