Effects of Exercise on AMPK Signaling and Downstream Components to PI3K in Rat with Type 2 Diabetes

Exercise can increase skeletal muscle sensitivity to insulin, improve insulin resistance and regulate glucose homeostasis in rat models of type 2 diabetes. However, the potential mechanism remains poorly understood. In this study, we established a male Sprague–Dawley rat model of type 2 diabetes, with insulin resistance and β cell dysfunction, which was induced by a high-fat diet and low-dose streptozotocin to replicate the pathogenesis and metabolic characteristics of type 2 diabetes in humans. We also investigated the possible mechanism by which chronic and acute exercise improves metabolism, and the phosphorylation and expression of components of AMP-activated protein kinase (AMPK) and downstream components of phosphatidylinositol 3-kinase (PI3K) signaling pathways in the soleus. As a result, blood glucose, triglyceride, total cholesterol, and free fatty acid were significantly increased, whereas insulin level progressively declined in diabetic rats. Interestingly, chronic and acute exercise reduced blood glucose, increased phosphorylation and expression of AMPKα1/2 and the isoforms AMPKα1 and AMPKα2, and decreased phosphorylation and expression of AMPK substrate, acetyl CoA carboxylase (ACC). Chronic exercise upregulated phosphorylation and expression of AMPK upstream kinase, LKB1. But acute exercise only increased LKB1 expression. In particular, exercise reversed the changes in protein kinase C (PKC)ζ/λ phosphorylation, and PKCζ phosphorylation and expression. Additionally, exercise also increased protein kinase B (PKB)/Akt1, Akt2 and GLUT4 expression, but AS160 protein expression was unchanged. Chronic exercise elevated Akt (Thr308) and (Ser473) and AS160 phosphorylation. Finally, we found that exercise increased peroxisome proliferator-activated receptor-γ coactivator 1 (PGC1) mRNA expression in the soleus of diabetic rats. These results indicate that both chronic and acute exercise influence the phosphorylation and expression of components of the AMPK and downstream to PIK3 (aPKC, Akt), and improve GLUT4 trafficking in skeletal muscle. These data help explain the mechanism how exercise regulates glucose homeostasis in diabetic rats.


Introduction
Obesity and insulin resistance are markers of type 2 diabetes [1]. Consumption of a high-fat diet (HFD) can induce obesity, insulin resistance and hyperinsulinemia, and is a widely used animal model to study insulin resistance [2,3,4]. However, this model does not develop hyperglycemia, another characteristic of type 2 diabetes in humans, limiting experimental studies focusing on the effects of exercise and antidiabetic drugs on blood glucose levels [5]. Therefore, extragenetic, outbred Sprague-Dawley rats are commonly fed a HFD in combination with low-dose streptozotocin (STZ) to establish type 2 diabetes, characterized by insulin resistance and b cell dysfunction [5,6]. This model replicates the pathogenesis and metabolic characteristics of type 2 diabetes in humans [5,6,7]. However, the potential mechanism involved in these changes is still poorly understood. In addition, few studies have examined whether exercise influences skeletal muscle AMPK activity, downstream components of the insulin signaling pathway, or metabolism in HFD-and low-dose STZinduced diabetic rats.
AMPK is a fuel-sensing enzyme that senses an increase in the intracellular AMP/ATP ratio during exercise or hypoxia [8,9]. Interestingly, animal models of insulin resistance show decreased AMPK activity in skeletal muscle [10,11]. Changes in AMPK activity may accelerate the progression of insulin resistance and metabolic abnormalities [12]. Exercise can activate skeletal muscle AMPK, effectively resisting diet-or lipid-induced insulin resistance [13,14]. However, in mouse models with decreased AMPK activity, inhibiting AMPK activity only weakly or did not influence muscle contraction-induced glucose absorption [15,16]. Additionally, LKB1, the main upstream kinase of AMPK, can influence glucose transport in skeletal muscle, and skeletal muscle-specific deprivation of LKB1 was reported to partly inhibit muscle contraction-induced glucose absorption [17]. Considering the potential discrepancies in these earlier results, it is necessary to further examine the mechanism by which exercise influences the AMPK pathway for regulation of metabolism.
Insulin and exercise/muscle contraction promote intracellular glucose transporter 4 (GLUT4) translocation to membrane, increasing skeletal muscle glucose uptake [18,19]. Chronic exercise also promotes the increase in the number of GLUT4 transporters [18]. Early studies have shown that insulin and exercise both enhance GLUT4 translocation and glucose transport through divergent signaling mechanisms. The insulin signaling pathway involves rapid phosphorylation of insulin receptor and the tyrosine residues of insulin receptor substrate-1 and -2, and activation of PI3K [20,21]. By contrast, exercise and muscle contraction do not affect insulin receptor or insulin receptor substrate-1 phosphorylation, or PI3K activity [21,22]. However, further studies of the insulin signaling molecules downstream to PI3K have shown that atypical protein kinase C (aPKC)f/l is responsive to both insulin and exercise [23,24]. Insulin stimulation results in aPKCf/l translocation to the plasma membrane in cultured muscle cells and also increases its association with GLUT4 vesicles [25]. But, no studies have confirmed the effects of exercise on aPKC expression and phosphorylation in rats with HFD-and low-dose STZinduced diabetes.
As type 2 diabetes is common in the middle-aged and elderly population, we used 15-month-old Sprague-Dawley rats in this study, and diabetes was established by HFD feeding and low-dose STZ injection. Based on previous studies [13,14,26], we hypothesized that exercise would improve the phosphorylation and expression of AMPK and its signaling pathway components, as well as molecules downstream of PI3K, to ultimately enhance insulin sensitivity in skeletal muscle. Thus, the present study sought to investigate whether this model of diabetes was associated with impaired skeletal muscle AMPK and PI3K signaling pathways, and whether chronic and acute exercise could improve the phosphorylation and expression of signaling molecules involved in both pathways, particularly aPKC. The study also examined the effects of HFD on AMPK and aPKC.

Establishment of Type 2 Diabetes
The rats were randomly assigned to the following groups: control(CON), high-fat diet (HFD), high-fat diet+chronic exercise (HFD +CE), high-fat diet+streptozotocin (HFD+STZ), high-fat diet+streptozotocin+chronic exercise (HFD+STZ+CE), and highfat diet+streptozotocin+acute exercise( HFD+STZ+AE) (8 rats per group). The CON group of rats received the routine control diet for 20 weeks, with free access to water. The HFD groups were fed with the HFD for 20 weeks, with free access to water. After 8 weeks of diet intervention, the HFD+STZ, HFD+STZ+CE, and HFD+STZ+AE groups received one intraperitoneal injection of low-dose STZ (30 mg/kg, Sigma), while other groups were injected once with citrate buffer solution (pH 4.4, 1 mL/kg). Four weeks after STZ injection, the animals were deprived of food for 12 h. Diabetes was regarded as successfully induced in rats with fasting blood glucose levels $7.8 mmol/L and postprandial blood glucose $11.1 mmol/L [5,7]. Body mass, water intake, and food intake were determined every week.

Chronic Exercise
The rats were trained to swim over 2 d, for 10-20 min per session, to reduce water-induced stress. Two or three rats, as one group, were placed in a plastic pool, 45 cm in diameter, 60 cm deep, and water temperature of 34-35uC. After the initial training, the rats underwent chronic exercise for 1 h/d, 5 d/wk, for 8 wk. The exercise program was conducted essentially as described by Luciano et al [28] with some modifications.

Acute Exercise
The HFD+STZ+AE rats received acute exercise after 20 weeks of HFD. Acute exercise was conducted in the same conditions as for chronic exercise. Acute exercise consisted of two sessions lasting 1.5 h each, with a 45 min interval between each session. The exercise program was conducted as previously described [29] with some modifications.

Blood Sample Collection and Blood Biochemistry
At 24-36 h after the final session of chronic exercise, or 8-16 h after acute exercise, all rats were anesthetized by an intraperitoneal injection of sodium thiopental (40 mg (kg body weight) 21 ). Blood samples were collected from tail veins and centrifuged at 1 1006g for 10 min to separate serum. Serum glucose, triglyceride, total cholesterol, and free fatty acid concentrations were determined using an autoanalyzer (RT-1904C; Rayto, Shenzhen, China). Serum insulin concentrations were determined using a radioimmunoassay, as previously described [30].

Insulin Tolerance Test and Serum Insulin Quantification
Insulin tolerance tests were performed after blood sample collection. Briefly, the rats were fasted for 12 h, anesthetized, and administered with 1.5 IU/kg artificial insulin (Sigma). Blood samples were collected at 0, 5, 10, 15, 20, 25, and 30 min after injection, centrifuged at 11006g, at 4uC for 15 min, and stored at -20uC to determine glucose concentrations. The rate constant for plasma glucose disappearance (K itt ) was calculated using the formula 0.693/biological half life. The plasma glucose half-life (t 1/

Skeletal Muscle Sampling
After the insulin tolerance test, the animals were intraperitoneally injected with sodium thiopental (200 mg/kg body weight; recommended by the US National Institutes of Health) and sacrificed. Bisected soleus was used from rat to perform glucose uptake and placed in 6 mL of Krebs-Ringer bicarbonate buffer (KRBB: 117 mmol/L NaCl, 4.7 mmol/L KCl, 2.5 mmol/L CaCl 2 , 1.2 mmol/L KH 2 PO 4 , 1.2 mmol/L MgSO 4 , 24.6 mmol/L NaHCO 3 ) supplemented with 8 mmol/L D-glucose for 20 min. For insulin treatment, muscles were preincubated and then incubated in the presence of 1 mM insulin for 40 min. The buffer was continuously infused with 95% O 2 /5% CO 2 at 37uC [32,33]. The uptake of 3-O-methyl-D-glucose (3MG) was determined as described below. The remainders of the soleus were placed in liquid nitrogen and stored at -80uC for western blotting and real time PCR. Meanwhile, gastrocnemius muscles were isolated and stored at -80uC to measure glycogen content.

3MG Uptake
To detect glucose transport, the muscles were cultured in KRBB containing 1 mmol/L 3MG (1.5 mCi/mL) and 7 mmol/L D-[14C] mannitol (0.3 mCi/mL; New England Nuclear, Boston, MA, USA) at 30uC for 10 min [32,33]. The tissue was then frozen in liquid nitrogen. The muscles were weighed, thawed in 1 mol/L NaOH (450 mL) at 80uC for 10 min, neutralized with 1 mol/L HCl, and centrifuged at 20 0006g for 2 min. Liquid scintillation counting was used to determine the radioactivity of double-labeled digested protein solution and to calculate 3MG uptake.

Skeletal Muscle Glycogen Content
The frozen muscles were weighed, digested with 1 mol/L NaOH (1:9 wt/vol) at 80uC for 10 min, neutralized with 1 mol/L HCl, and mixed with 6 mol/L HCl to a final concentration of 2 mol/L HCl. The resulting solution was incubated at 85uC for 2 h and neutralized with 5 mol/L NaOH [35]. A glucose hexokinase assay kit (Sigma) was used to determine the concentration of hydrolyzed glucose, and glucose content was determined as mmol per 1 g tissue (wet weight).

Real-time PCR
Total RNA was extracted from the frozen muscles using RNAiso Plus (Takara, Dalian, China), and the RNA concentration was measured at 260/280 nm. RNA samples were reversetranscribed using a PrimeScript RT reagent kit (Takara). Realtime PCR and subsequent data were analyzed using a 7900HT

Statistical Analysis
Results are expressed as mean 6 SE. Differences between the control and HFD, control and HFD+STZ, HFD and HFD+exercise, and HFD+STZ and HFD+STZ+exercise groups were compared by one-way analysis of variance. Values of P,0.05 were considered statistically significant. Data were analyzed using JMP software (SAS Institute, Cary, NC, USA).

Characteristics of the Rat Model of Type 2 Diabetes and Effects of Chronic Exercise
As shown in Table 1, compared with the control group, no differences were found in body mass or epididymal fat mass in the HFD+STZ group. However, food intake, blood glucose, triglyceride, total cholesterol, and free fatty acid were increased, while K itt , 3MG uptake and glycogen content were lower in the HFD+STZ group. These results indicate that HFD+STZ induced disorders in glucose and lipid metabolism, common metabolic characteristics of diabetes, and could be used to establish a rat model of type 2 diabetes, mimicking diabetes in humans. Compared with HFD rats, food intake and blood glucose were increased, while body mass, epididymal fat, insulin, and K itt were reduced in the HFD+STZ rats (Table 1). Meanwhile, compared with control rats, the HFD increased body mass, epididymal fat mass, blood insulin concentration, blood total cholesterol and free fatty acid concentrations, but reduced K itt , 3MG uptake and glycogen content, confirming the HFD used induced insulin resistance.
As shown in Table 1, chronic exercise reduced blood glucose, triglyceride, total cholesterol, and free fatty acid concentrations, and increased the glucose disappearance rate, 3MG absorption and glycogen content in HFD+STZ rats. However, blood insulin concentration, body mass and epididymal fat mass remained unchanged. In addition, chronic exercise decreased body mass, epididymal fat mass, blood insulin concentration, and total cholesterol, but increased K itt , 3MG absorption and glycogen content in the HFD rats (Table 1).

Chronic Exercise Enhanced the Phosphorylation and Expression of Protein Kinase Cascades in the AMPK Pathway
To verify that the increase in exercise-stimulated glucose uptake was related to changes of the AMPK pathway signalling in this setting, we measured phosphorylation and expression of AMPKa1/2, AMPKa1, AMPKa2, LKB1 and ACC in muscle lysates. Phosphorylation of AMPKa1/2, AMPKa1, AMPKa2 and LKB1 in the muscle was reduced (P = 0.025, P = 0.013, P = 0.001 and P = 0.016; Fig. 1A, B, C and E) in the HFD+STZ group compared with the control group. However, the changes in phosphorylation were increased by chronic exercise. These changes were accompanied by a similar trend in protein expression of AMPKa1/2, AMPKa1, AMPKa2 and LKB1 (Fig. 1F). By contrast, the phosphorylation and expression of AMPK downstream target, ACC, were significantly greater in the HFD+STZ group compared with the control group ( Fig. 1D and F); these changes were suppressed by chronic exercise. Moreover, the results showed an increased ACC phosphorylation and a decreased AMPKa2 protein expression in the HFD+STZ group versus the HFD group (P = 0.001 and P = 0.009; Fig. 1D and F). There was no significant effect of exercise on ACC phosphorylation and AMPKa2 protein expression in HFD rats, whereas exercise improved these changes in HFD+STZ rats as described above. Chronic exercise also increased the phosphorylation and expression of AMPKa1/2( Fig. 1A and F), AMPKa1( Fig. 1B and F), as well as LKB1 phosphorylation (Fig. 1E) in the HFD rats.

Chronic Exercise Increased Phosphorylation and Expression of aPKC
It is unknown whether the effects of chronic exercise on aPKC expression and phosphorylation in HFD+STZ rats. Our results showed phosphorylation of PKCf/l, and phosphorylation and protein expression of PKCf in the muscle were significantly lower in the HFD+STZ group than in the control group (P = 0.011, P = 0.013, and P = 0.002; Fig. 2A, B and C). Chronic exercise improved the changes. Besides, the HFD significantly reduced the phosphorylation of PKCf (P = 0.039), but exercise increased PKCf phosphorylation (P = 0.047; Fig. 2B). Meanwhile, the phosphorylation of PKCf/l was marked decreased (P = 0.019; Fig. 2A) in the HFD+STZ rats compared with the HFD rats. However, the effects of exercise on phosphorylation of PKCf/l were only observed in the HFD+STZ rats.

Effects of Chronic Exercise on Components of the Insulin Signaling Pathway
To establish the impairments of the insulin signaling pathway and the effects of chronic exercise, we examined downstream components of PI3K, including phosphorylation and protein expression of Akt1, Akt2 and AS160, phosphorylation of Akt (Thr 308 ) and Akt (Ser 473 ), protein expression of Akt and GLUT4 both in HFD+STZ and HFD rats. Compared with the control group, phosphorylation of Akt1, Akt2, Akt (Thr 308 ), Akt (Ser 473 ) and AS160 in the muscle was reduced (P = 0.019, P = 0.003, P = 0.04, P = 0.029, P = 0.016; Fig. 3A, B, C, D and E) in the HFD+STZ group. Additionally, we observed the muscle protein content of Akt1 (P = 0.026; Fig.3G), Akt2(P = 0.025; Fig.3G) and Akt (P = 0.002; Fig.3G) was lower in the same group than in the control group. Chronic exercise improved the changes. Meanwhile, we also analyzed the protein expression of AS160 and GLUT4. In the line with the effects on Akt protein expression, exercise increased protein content of GLUT4 (P = 0.032; Fig. 3F), but AS160 protein expression was unchanged (Fig. 3G) in the HFD+STZ rats. In addition, chronic exercise increased the phosphorylation of Akt2 and Akt (Thr 308 ) and AS160 (Fig. 3B, C and E), as well as protein content of Akt1, Akt and GLUT4 in the HFD rats.

Effects of Acute Exercise on Metabolic Parameters
Blood glucose concentrations measured at 8-16 h after acute exercise in the HFD+STZ+AE group were significantly lower than those in the HFD+STZ group. However, acute exercise did not improve body mass, or blood triglyceride, total cholesterol, free fatty acid, or insulin concentrations. Acute exercise increased the glucose disappearance rate in the HFD+STZ+AE group, although this was still lower than that in the control group (Table 1 and 2). 3MG uptake was reduced by 55% in the HFD+STZ group compared with the control group, but was increased by 30% at 16 h after acute exercise in the HFD+STZ+AE group (P,0.05), indicating that acute exercise does not completely compensate for muscle insulin resistance in HFD+STZ rats. Although acute exercise increased glycogen content in the HFD+STZ+AE group compared with the HFD+STZ group, this difference was not statistically significant (Table 2).

Acute Exercise Improves AMPK Pathway in HFD+STZ Rats
We also examined the effect of acute exercise on the AMPK pathway at different signaling levels. Compared with the HFD+STZ group, acute exercise increased AMPKa1/2 phosphorylation and protein content (P = 0.049 and P = 0.030, respectively; Fig. 4A and F). Acute exercise also reversed the effects of HFD+STZ on AMPKa1 phosphorylation, and AMPKa2 phosphorylation and protein expression (P = 0.004, P = 0.024, and P = 0.009, respectively; Fig. 4B, C and F), but not AMPKa1 protein expression (Fig. 4F). Acute exercise also decreased ACC protein expression (P = 0.008; Fig. 4F), but not ACC phosphorylation (Fig. 4D). Moreover, acute exercise increased LKB1 protein expression (P = 0.018; Fig. 4F).

Acute Exercise Reversed the Effects of HFD+STZ on the Phosphorylation and Expression of aPKC
It has not been observed whether acute exercise influences the phosphorylation and expression of aPKC in HFD+STZ rats. Our data showed acute exercise increased the phosphorylation of PKCf/l after the acute exercise, as compared with the HFD+STZ group (P = 0.043; Fig. 5A). Moreover, PKCf phosphorylation and protein expression were significantly upregulated  in the HFD+STZ+AE group compared with the HFD+STZ group (P = 0.002 and P = 0.003, respectively; Fig. 5B and C).

Effects of Acute Exercise on Components of the Insulin Signaling Pathway in Muscle in HFD+STZ rats
To investigate skeletal muscle responses to acute exercise, we analyzed downstream components of PI3K insulin signaling. Acute exercise enhanced Akt1 phosphorylation and protein expression (P = 0.038 and P = 0.041, respectively; Fig. 6A and E). Acute exercise also increased Akt2 phosphorylation (P = 0.049; Fig. 6B), but not Akt2 protein expression (Fig. 6E). AS160 phosphorylation only tended to increase after the acute exercise in the HFD+STZ group (Fig. 6C). Besides, acute exercise also significantly increased GLUT4 protein expression in the HFD+STZ group (P = 0.022; Fig. 6D).

Effects of Chronic and Acute Exercise on PGC1a and NRF1 mRNA Expression
We evaluated the effects of exercise on PGC1a and NRF1 gene expression in the HFD and HFD+STZ groups. PGC1a mRNA expression in the muscle was significantly lower in the HFD and HFD+STZ groups compared with the control group (P,0.001 and P,0.001, respectively; Fig. 7A). Chronic exercise increased PGC1a mRNA expression in the HFD+CE and HFD+STZ+CE groups (P = 0.048 and P,0.001, respectively). NRF1 mRNA expression was also significantly reduced in the HFD and HFD+STZ groups compared with the control group (P = 0.001 and P = 0.001, respectively; Fig. 7B). Although chronic exercise increased NRF1 mRNA expression in the HFD+CE and HFD+STZ +CE groups, the difference had no statistical significance. Acute exercise in the HFD+STZ+AE group significantly increased the mRNA expression of both PGC1a and NRF1 compared with the HFD+STZ group (P = 0.001 and P = 0.045; Fig. 7C and D).

Summary of Findings
The results of this study showed that the pathogenesis and metabolic characteristics of rats with HFD+STZ-induced type 2 diabetes were similar to those in humans. Chronic and acute exercise improved phosphorylation and protein expression levels of components of the skeletal muscle AMPK pathway, as well as the phosphorylation and expression of aPKC, Akt and Akt isoforms. Together, these changes conferred enhanced GLUT4 translocation to membrane and promoted 3MG uptake in these rats. Chronic and acute exercise also reversed the downregulation of PGC1a mRNA caused by HFD+STZ.

Mechanisms of Chronic Exercise-stimulated Glucose Uptake
Changes of AMPK signaling. AMPK plays a key role in fat and carbohydrate metabolism. Presumably, defects in AMPK signaling could somehow result in the metabolic abnormalities of type 2 diabetes [35]. It has become apparent that AMPK was important in mediating the acute and chronic effects of exercise on glucose and lipid metabolism in skeletal muscle [36,37]. However, under insulin-resistant conditions, the results of studies of food-induced or genetic obesity differed. Skeletal muscle AMPK phosphorylation was reduced in HFD-induced obese Wistar rats and in obese Zucker rats [35,38]. Acute exercise was reported to increase AMPK phosphorylation in Wistar rats [38], but not in obese Zucker rats [35]. AMPK phosphorylation and protein expression were not adversely affected in the skeletal muscle of ob/ ob mice and obese Wistar rats with insulin resistance induced by cafeteria diet [39,40]. However, exercise enhanced AMPKa2 expression in Wistar rats above [39]. These differing results may be due to differences in the species of the experimental animals. Interestingly, we found marked differences between the HFD and HFD+STZ groups in this study. In particular, the HFD+STZ group showed altered AMPKa1 and a2 expression, AMPKa1 and a2Thr 172 phosphorylation and ACC expression and phosphorylation, all of which were improved by exercise. A previous study demonstrated that exercise increased AMPKa2 expression and AMPK Thr 172 phosphorylation in diabetic patients [41]. In the present study, AMPKa1 expression and Thr 172 phosphorylation were reduced in HFD-induced obese and insulin-resistant rats, which were reversed by chronic exercise, consistent with previous results [4]. Although the mechanisms responsible for the different effects of obesity and insulin resistance, especially in type 2 diabetes, on AMPK signaling components are unclear, it is apparent that chronic exercise increase AMPK expression and downregulate expression and phosphorylation of its downstream target, ACC. These effects of exercise are partly associated with enhanced glucose transport [36] and lipid oxidation [37] at least.
Findings in aPKC. The characteristics of aPKC, which lie downstream of PI3K, have been described in recent years. In particular, it has been reported that aPKC plays a key role in normal insulin-stimulated glucose uptake [42]. Several studies have shown that insulin stimulation and exercise can increase skeletal muscle aPKC activity [23,42,43,44]. However, glucose absorption capacity is more strongly induced by insulin-mediated aPKC activity than by exercise-mediated signaling [23,44]. HFDinduced insulin resistance in rats reduced insulin's ability to stimulate PKCf/l activity, even though the phosphorylation and protein expression levels of PKCf and PKCl remained unchanged [26,45,46]. In the present study, HFD+STZ reduced PKCf/l phosphorylation, as well as PKCf phosphorylation and expression in skeletal muscle, effects that were reversed by chronic exercise. Kanzaki et al [47] reported that PKCf/l interacts with vesicles containing GLUT4. PKCf regulates munc18, a protein that facilitates the transport of vesicles containing GLUT4 [48]. Therefore, exercise may enhance the activity of PKCf/l to aid GLUT4 translocation. However, downstream signaling of aPKC that promote translocation of GLUT4 vesicles are still unknown, and the precise mechanism by which aPKC regulate GLUT4 translocation is now being explored [42,43].
Downstream components of PI3K. Akt, AS160, and GLUT4, are all components of the classical insulin pathway. Chronic exercise increased the phosphorylation and protein expression of the Akt1 and 2 isoforms in the HFD+STZ group, consistent with the responses of Akt (Ser 473 ) and Akt (Thr 308 ) phosphorylation to exercise. However, a previous study showed that exercise only restored Akt1 expression in rats with HFDinduced insulin resistance, and Akt2 content was unchanged [4]. Akt is the key component of the insulin signaling pathway, and promotes glucose transport via AS160 [49]. AS160 was the first component of the insulin signaling pathway found to directly promote GLUT4 membrane transport [42]. A recent study has confirmed the effects of AS160 in inducing glucose absorption in skeletal muscle [50]. In the present study, we found the phosphorylation of AS160 and GLUT4 expression were increased in the HFD+CE and HFD+STZ+CE groups. Therefore, the results from the present study indicate that exercise-induced activation of the Akt-AS160-GLUT4 pathway is important for improving skeletal muscle glucose handling in HFD rats and HFD+STZ-induced diabetic rats.
AMPK upstream kinase, LKB1. Experiments using mice with muscle-specific LKB1 knockout showed that LKB1 is a major kinase upstream of AMPK involved in the contractile response, and is an important regulator of skeletal muscle glucose transport [17,51]. However, Sriwijitkamol et al [52] reported that exercise did not affect LKB1 content or activity in obese and type 2 diabetic patients. In the present study, chronic exercise improved LKB1 content and phosphorylation, which were otherwise reduced in the HFD and HFD+STZ groups. An earlier study showed that skeletal muscle LKB1 expression was reduced in obese and insulin-resistant Zucker rats, which was reversed by chronic exercise [35]. The differences in results among these studies may be due to differences in exercise pattern and/or intensity.

Mechanisms of Acute Exercise-stimulated Glucose Uptake
Acute exercise significantly enhanced insulin's ability to stimulate glucose absorption in skeletal muscle and counteracted insulin resistance, a typical feature of type 2 diabetes. However, the potential molecular mechanism remains poorly understood [42]. Therefore, in this study we subjected HFD+STZ-induced diabetic rats to a session of acute exercise. We found that acute exercise improved the phosphorylation and protein expression of the protein kinase cascade in the AMPK signaling pathway. Interestingly, acute exercise reversed the effects of HFD+STZ on aPKC phosphorylation and expression, increased Akt subunit phosphorylation and expression, elevated GLUT4 content, and enhanced glucose absorption in skeletal muscle. However, there were some notable differences in the characteristics of rats subjected to chronic and acute exercise. For example, blood triglyceride, total cholesterol, and free fatty acid concentrations were not affected by acute exercise. The main difference in signaling pathways between chronic and acute exercise was related to phosphorylation of ACC and LKB1, and AKT2 protein content, which did not change after an acute bout of exercise. It has been postulated that the discrepancy may reflect the influence of different modes of chronic and acute exercise, as well as exercise intensity. Given this contention, it is also plausible that differences in the signaling response between models could be related to the time points chosen for evaluation. However, the net effects of acute and chronic exercise on the AMPK and downstream to PIK3 signaling pathways were broadly comparable.

Gene Expression of PGC1a and NRF1 Response to Exercise
Exercise improves skeletal muscle AMPK activity, and enhances mitochondrial biogenesis and function [35,53]. An increase in mitochondrial quantity and function is thought to involve activation of the PGC1a gene by the AMPK-mediated peroxisome proliferator [54]. Therefore, AMPK may directly contribute to the exercise-induced increase in PGC1a gene expression. The results of this study suggest that by enhancing AMPK activity, chronic and acute exercise reversed the downregulation of gene expression in the HFD+STZ rats. As a transcriptional coactivator, PGC-1 is a major coordinator of mitochondrial biogenesis, and acts by interacting with NRF-1 [55]. Thus, we also assessed NRF1 gene expression in this study. We found that acute exercise restored the reduction in NRF1 gene expression induced by HFD+STZ in muscle.

Conclusion
In summary, chronic and acute exercise improved skeletal muscle AMPK signaling and downstream to PIK3 in rats with HFD-plus STZ-induced diabetes. Although the HFD adversely affected the phosphorylation and expression of several components of the AMPK signaling and downstream to PIK3, chronic exercise improved these changes. The present study provides important insights into the mechanisms by which exercise may correct the molecular disorders associated with obesity and type 2 diabetes.