Involvement of mTOR in Type 2 CRF Receptor Inhibition of Insulin Signaling in Muscle Cells

Abstract

Type 2 corticotropin-releasing factor receptor (CRFR2) is expressed in skeletal muscle and stimulation of the receptor has been shown to inhibit the effect of insulin on glucose uptake in muscle cells. Currently, little is known about the mechanisms underlying this process. In this study, we first showed that both in vivo and in vitro CRFR2 expression in muscle was closely correlated with insulin sensitivity, with elevated receptor levels observed in insulin resistant muscle cells. Stimulation of CRFR2 by urocortin 2 (Ucn 2), a CRFR2-selective ligand, in C2C12 myotubes greatly attenuated insulin-induced glucose uptake. The inhibitory effect of CRFR2 signaling required cAMP production and is involved the mammalian target of rapamycine pathway, as rapamycin reversed the inhibitory effect of CRFR2 stimulation on insulin-induced glucose uptake. Moreover, stimulation of CRFR2 failed to inhibit glucose uptake in muscle cells induced by platelet-derived growth factor, which, similar to insulin, signals through Akt-mediated pathway but is independently of insulin receptor substrate (IRS) proteins to promote glucose uptake. This result argues that CRFR2 signaling modulates insulin's action likely at the levels of IRS. Consistent with this notion, Ucn 2 reduced insulin-induced tyrosine phosphorylation of IRS-1, and treatment with rapamycin reversed the inhibitory effect of Ucn 2 on IRS-1 and Akt phosphorylation. In conclusion, the inhibitory effect of CRFR2 signaling on insulin action is mediated by cAMP in a mammalian target of rapamycine-dependent manner, and IRS-1 is a key nodal point where CRFR2 signaling modulates insulin-stimulated glucose uptake in muscle cells.

Skeletal muscle insulin resistance with impaired glucose disposal has been considered as a primary defect for type 2 diabetes and metabolic syndrome (1–3). With insulin resistance, skeletal muscle diverts ingested carbohydrate away from muscle glycogen storage into hepatic de novo lipogenesis, secondarily leading to liver steatosis and hypelipidemia (1). The exact underlying mechanisms of muscle insulin resistance have not been fully delineated and most likely involved a complex interaction between extrinsic and intrinsic factors. Stress has been proposed as a risk factor for insulin resistance and metabolic syndrome (4, 5). Surprisingly, little is known about the detailed mechanism by which stress may modulate insulin action and impact insulin sensitivity in muscle cells.

Corticotropin-releasing factor (CRF), a 41-amino acid polypeptide isolated originally from the ovine hypothalamus (6), plays a central role in coordinating the hypothalamic-pituitary-adrenal axis under basal and stress conditions and integrates endocrine, autonomic, and behavioral responses to stressors (7, 8). In addition to CRF, members of the CRF peptide family, including urocortin (Ucn) 1–3, have been identified in mammals (9–12). The function of CRF peptides is mediated by 2 G protein-coupled receptors: type 1 CRF receptor (CRFR1) and CRFR2 (7, 8). These 2 receptors share 69% amino acid homology but have different pharmacological properties with respect to ligands: CRF binds selectively to CRFR1, Ucn 1 binds both receptors with equal high affinity, and Ucn 2 and 3 bind CRFR2 with high affinity with only modest affinity to CRFR1 (7, 13). Stimulation of both CRFRs lead to activation of adenylyl cyclase, which subsequently increases intracellular cAMP levels (8). The resulting increase in cAMP levels mediates most CRF-induced physiological functions (8).

Accumulating evidence has suggested that the CRF family peptides and their receptors play an important role in modulating glucose homeostasis. Mice deficient in CRFR2 have enhanced glucose tolerance, increased insulin sensitivity, and are protected from high-fat diet-induced insulin resistance compared with wild-type (WT) littermates (14). Similar phenotypes have also been observed in Ucn 2-null mice (15). Consistent with mutant mouse studies, acute peripheral injection of Ucn 2 impairs glucose tolerance in a CRFR2-dependent manner (15). These studies suggest activation of CRFR2 negatively impacts glucose homeostasis.

In addition to the central nervous system, CRFR2 is expressed in the periphery, including in the skeletal muscle (10, 16, 17). Importantly, it has been shown that the levels of CRFR2 in skeletal muscle are elevated by high-fat feeding and chronic variable stress (18), conditions that are associated with muscle insulin resistance (19–24). Moreover, a functional study shows that in primary skeletal muscle cells and in C2C12 myotubes stimulation of CRFR2 by Ucn 2 attenuates phosphorylation of Akt and inhibits insulin-induced glucose uptake (15). These results strongly suggest that CRFR2 signaling is involved in attenuating insulin action in skeletal muscle induced by stress or high-fat feeding. Currently, little is known regarding signaling pathways of CRFR2 in muscle cells to modulate insulin signaling. A number of cellular signaling molecules, including protein kinase A (PKA), exchange proteins activated by cAMP (Epac), and mammalian target of rapamycine (mTOR) (25–29) have been shown to serve as downstream effectors of cAMP pathway. In the present study, we first confirmed both in vivo and in vitro that CRFR2 expression is elevated in insulin resistant muscle cells. Moreover, using C2C12 myotubes as an in vitro model, we demonstrate that cAMP is crucial in mediating the effect of CRFR2 in suppressing insulin-induced glucose uptake via a rapamycin-sensitive pathway, and tyrosine phosphorylation of insulin receptor substrate 1 (IRS-1) is a key nodal point where CRFR2 signaling modulates insulin action.

Materials and Methods

Animals

Male ob/ob mice and WT littermates (n = 8/group, 10–12 wk old) were purchased from The Jackson Laboratory. Mice were maintained on a 12-hour light, 12-hour dark cycle and had ad libitum access to rodent chow and water during the acclimation period. After 1 week of acclimation period, mice were killed, and the hindlimb skeletal muscle consisting of the gastrocnemius, the soleus, and the extensor digitorum longus was quickly harvested and stored at −80°C until analysis. Exercise experiment was conducted as previously described (30). Briefly, adult male mice (n = 6/group, C57BL/6J; The Jackson Laboratory) were subjected to voluntary running (4 wk) in cages equipped with a running wheel and wheel-running activity was monitored continuously with Dataquest Acquisition and Analysis System (Data Sciences International). Mice with locked wheel in the cage were served as control. After 4 weeks of voluntary running, mice were killed, and muscles were harvested from each mouse. All studies were approved by the University of Virginia Animal Care and Use Committee.

Cell culture

Mouse C2C12 myoblasts (American Type Culture Collection) were maintained in DMEM containing 10% fetal bovine serum (HyClone) and penicillin/streptomycin at 37°C in a 5% CO₂. When cells reached confluence, the medium was replaced with DMEM supplemented with 2% horse serum and replaced every 2 days. After 4 additional days, the differentiated C2C12 cells had fused into myotubes as previously described (31). C2C12 cells were used for experiments between 5 and 6 days after differentiation. To induce insulin resistance in C2C12 myotubes, after 5–6 days of differentiation, C2C12 myotubes were treated insulin (10nM) or palmitate (0.75mM) for 24 hours as described elsewhere (32) to induce insulin resistance. Mouse Ucn 2 and astressin-2B (Ast2B) were generously provided by Dr Jean Rivier at The Salk Institute (La Jolla, CA). Pharmacological agents were purchased from Sigma.

Glucose uptake assay

Differentiated C2C12 cells were incubated with serum-free medium for overnight and washed 2 times with prewarmed Kreb-Ringer phosphate buffer (100mM NaCl, 5mM KCl, 1mM MgCl₂, 1mM NaH₂PO₄, and 1mM CaCl₂). The cells were pretreated with testing agents for 20 minutes before insulin treatment for 10 minutes. Glucose uptake was initiated by adding [³H]2-deoxy-D-glucose (0.5 μCi/mL; MP Biomedicals) into the medium for 10 minutes. Nonspecific glucose uptake was determined by performing glucose uptake in the presence of cytochalasin B. Reaction was terminated by 3 washes of cells with ice-cold Kreb-Ringer phosphate buffer. Cells were then lysed with 0.1% sodium dodecyl sulfate, and aliquots were removed for determining radioactivity and protein concentration was determined by Bradford assay (Bio-Rad). Each experiment was repeated minimum 3 times performed in triplicates.

Lentiviral production

Lentiviral vector expressing small hairpin RNA (shRNA) to suppress the expression of CRFR2 was generated as previously described (33). Briefly, shRNA against CRFR2 expressing cassettes driven by a RNA polymerase III promoter U6 were constructed and cloned into a lentiviral packaging vector (p156RRLsinPPtCMV-GFP-PREU3Nhe, kindly provided by Inder Verma, The Salk Institute). The viral vector also expresses green fluorescent protein under the control of a cytomegalovirus promoter, thus allowing identification of cells infected by the viral vector. In addition, a control vector containing a nontarget scrambled sequence was also generated. CRFR2 shRNA vector or control vector with additional 3 lentiviral packaging plasmids were cotransfected into HEK239FT cells. After transfection, the supernatant containing the viral particles was harvested, clarified, and concentrated by centrifugation.

RNA analysis

Total RNA was extracted from C2C12 cells or mouse muscle samples with Tri Reagent according to the manufacturer's protocol (Molecular Research Center, Inc). Quality of RNA was confirmed by ethidium bromide staining in 2% agarose gel. Gene sequences were obtained from the GenBank database. Primers used for PCR analysis (Supplemental Table 1) were designed using Primer 3. Single-strand cDNA was synthesized using iScript cDNA Synthesis kits (Bio-Rad). Real-time PCR was performed using a MyIQ Single Color Real-Time PCR Detection System iCycler (Bio-Rad). Amplification products were verified by melting curves and agarose gel electrophoresis. All reactions were performed in duplicate. The results of relative expression were normalized to cyclophilin or β-actin mRNA levels in each sample.

Western blot analysis

Cells were rinsed with ice-cold PBS and solubilized in lysis buffer containing 50mM HEPES, 150mM NaCl, 10mM EDTA, 0.5% Triton X-100, 5% glycerol, 0.25% Nonidet P-40, 1mM phenylmethylsulphonyl fluoride, 1-μg/mL aprotinin, 1-μg/mL leupeptin, 100mM NaF, and 1mM Na₃VO₄ (pH 7.4) and incubated for 30 minutes at 4°C. Whole-cell lysates were centrifuged at 4°C for 10 minutes to remove insoluble compartment. The proteins in the supernatants were separated by SDS-PAGE gels and transferred to nitrocellulose membranes. After blocking with Tris-buffered saline Tween-20 (10mM Tris, 150mM NaCl, and 0.05% Tween 20; pH 7.6) containing 5% nonfat milk for 1 hour at room temperature, the membranes were incubated with the appropriate antibodies overnight at 4°C. Antibodies to phospho-S473 Akt, phospho-T389 S6 kinase (S6K), Akt, and S6K were from Cell Signaling Technology. Antiphospho-Y608/612 IRS-1 and anti-IRS-1 were from Millipore. Anti-α-tubulin was from Sigma. The proteins were visualized by enhanced chemiluminescence using horseradish peroxidase-conjugated antirabbit or mouse IgG. Densitometric analysis of the bands was performed using AlphaImager 2200 (Alpha Innotech, Inc) and analyzed by Multigauge software (version 3.0). Each experiment was repeated 2–3 times and performed in triplicates.

cAMP assay

Serum-starved C2C12 cells were preincubated for 30 minutes with 0.1mM 3-isobutyl-1-methylxanthine and then exposed to Ucn 2 or other peptides for 30 minutes at 37°C. Intracellular cAMP was measured in triplicate using a commercial ELISA kit (Biomedical Technologies).

Statistical analysis

Results are expressed as mean ± SEM. A 2-tailed Student's t test was used for comparisons between ob/ob and WT mice, between exercised and control mice, and between control C2C12 cells and cells treated with insulin or palmitate. For differences in glucose uptake or protein phosphorylation in C2C12 cells, two-way ANOVA was used followed by post hoc analyses for between group comparisons. P < .05 was considered statistically significant.

Results

Expression of CRFR2 in skeletal muscle of murine diabetic model and exercised mice

To probe the possible functional role of CRF signaling in skeletal muscle insulin resistance, we first examined the expression of CRFR2 in skeletal muscle of diabetic ob/ob obese mice and WT littermates. As shown in Figure 1A, CRFR2 mRNA levels were significantly increased in the muscle of obese diabetic ob/ob mice compared with that of WT mice. On the other hand, CRFR2 mRNA (Figure 1B) and protein levels (Supplemental Figure 1C) were decreased in skeletal muscle of chronically exercise mice compared with that of sedentary control mice.

Figure 1.

Expression of CRFR2 in skeletal muscle of mice. A, CRFR2 mRNA levels in total RNA samples isolated from skeletal muscle of ob/ob obese mice and WT control mice. B, Expression of CRFR2 in skeletal muscle samples from mice with 4 weeks of running exercise or sedentary controls. **, P < .01; *, P < .05 vs controls.

Expression of CRFR2 in differentiated C2C12 myotubes

To evaluate the function of CRFR2 in skeletal muscle cells in vitro, we first determined the expression of CRFR2 in differentiated C2C12 myotubes. CRFR2 mRNA levels were examined in undifferentiated C2C12 myoblasts and in C2C12 myotubes at 2, 4, and 6 days after differentiation. As shown in Figure 2A, CRFR2 mRNA levels were low in undifferentiated myoblasts and CRFR2 expression was increased concomitantly with differentiation. The expression of CRFR2 in C2C12 differentiation was also confirmed at protein levels (Supplemental Figure 2). The function of the receptor in differentiated C2C12 cells was validated by Ucn 2 dose dependently stimulating cAMP production in differentiated myotubes (Supplemental Figure 3). Furthermore, the effects on Ucn 2-induced cAMP accumulation was mediated by CRFR2, as pretreating cells with Ast2B, a CRFR2-selective antagonist, completely blocked Ucn 2-induced cAMP accumulation (Supplemental Figure 3). To determine whether elevated CRFR2 expression in the muscle of diabetic mice is also observed in vitro, insulin resistance was induced in differentiated C2C12 myotubes by chronic high-insulin treatment or palmitic acid feeding. As expected, both treatments induced insulin resistance (see Figure 4 below and Supplemental Figure 4), and importantly, the levels of CRFR2 mRNA (Figure 2) and protein (Supplemental Figure 1) in insulin resistant C2C12 myotubes were greatly elevated. We also confirmed the CRFR2 overexpression by Ucn 2-induced cAMP production, as chronic insulin treated cells showing enhanced sensitivity to Ucn 2 in inducing cAMP production than in control cells (Supplemental Figure 5).

Figure 2.

Expression of CRFR2 in C2C12 myotubes. A, CRFR2 mRNA levels in total RNA samples isolated from C2C12 cells at various time points of differentiation. B and C, CRFR2 mRNA levels in C2C12 myotubes treated with insulin (10nM) (B) or palmitate (C) or respective vehicle for 24 hours. **, P < .01; *, P < .05 vs vehicle controls.

Figure 4.

Effect of pharmacological agents in Ucn 2 inhibition of insulin-induced glucose uptake. Serum-starved cells were treated with Ucn 2 and/or MDL-12330A (5μM) (A), forskolin (5μM) (B), or rapamycin (100nM) (C) for 20 minutes followed by insulin (10nM) treatment for 10 minutes. Bars with different letters were significantly different with P < .05.

Activation of CRFR2 suppresses insulin-induced glucose uptake in C2C12 myotubes

We then evaluated the effect of stimulation of CRFR2 in muscle cells on insulin-mediated glucose uptake. As expected, insulin treatment had an significant effect on glucose uptake (F_1,62 = 62.9, P < .001). Insulin (10nM) significantly stimulated glucose uptake in C2C12 cells (Figure 3A). Ucn 2 (10nM) (Figure 3) alone up to 100nM (data not shown) had no significant effect on glucose uptake. Importantly, pretreating cells with Ucn 2 (10nM) greatly attenuated insulin-induced glucose uptake, and the effect of Ucn 2 is mediated by CRFR2 as pretreating cells with Ast2B (100nM) abolished the inhibitory effects of Ucn 2 (Figure 3A).

Figure 3.

Effect of CRFR2 stimulation by Ucn 2 on insulin-induced 2-deoxyglucose (2-DG) uptake in C2C12 myotubes. A, Serum-starved cells were treated with Ucn 2 and/or Ast2B for 20 minutes followed by insulin (10nM) treatment for 10 minutes. B, Glucose uptake in C2C12 myotubes infect with lentiviral vector expressing shRNA against CRFR2 (CRFR2 KD) or scramble control (control) treated with insulin and/or Ucn 2. C, Glucose uptake in C2C12 cells infected with scramble (control) or CRFR2 shRNA viral vector. Cells were then first treated with chronic high insulin (10nM) for 24 hours to induce insulin resistance before the glucose uptake experiment. Note that insulin failed to induce glucose uptake in control viral vector infected cells. Bars in A with different letter were significant different with P < .05. *, P < .05 vs control cells treated with insulin; #, P < .05 vs vehicle controls; **, P < .01 vs control cells treated with insulin.

Effect of CRFR2 knockdown in glucose uptake in insulin resistant muscle cells

We further evaluated the effect of Ucn 2 on glucose uptake in CRFR2 knockdown C2C12 myotubes. A lentiviral vector expressing shRNA against mouse CRFR2 (33) was used to suppress CRFR2 expression in C2C12 myotubes (Supplemental Figure 6). We found that the effect of Ucn 2 in inhibiting insulin-induced glucose uptake was reduced in CRFR2 knockdown cells (Figure 3B). Interestingly, insulin-induced glucose uptake was further enhanced in CRFR2-deficient cells compared with that in control cells (Figure 3B). We then determined whether CRFR2 knockdown improves insulin sensitivity in insulin resistant muscle cells. As shown in Figure 3C, chronic high-insulin treatment abolished insulin-induced glucose uptake. Importantly, CRFR2 knockdown restored insulin sensitivity in glucose uptake in insulin resistant C2C12 cells (Figure 3C).

Involvement of adenylyl cyclase and mTOR in the inhibitory effect of CRFR2 signaling on insulin-induced glucose uptake in muscle cells

CRFR2 has been shown to predominately couple to G_αs to stimulate cAMP production (8). As shown in Figure 4A, pretreating C2C12 cells with (cis-N-(2-Phenylcyclopentyl)-azacyclotridec-1-en-2-amine hydrochloride) MDL-12330A, an adenylyl cyclase inhibitor, effectively blocked the effect of Ucn 2 on insulin-induced glucose uptake. (P < .01), indicating that cAMP is necessary in mediating the inhibitory effect of CRFR2 signaling on insulin-induced glucose uptake. Similar to Ucn 2, forskolin, which stimulates adenylyl cyclase to raise cellular cAMP levels, attenuated insulin-stimulated glucose uptake (Figure 4B). Pretreating cells with a number of PKA pathway inhibitors, including H-89 and Rp-cAMPS, had no effect on Ucn 2 inhibition of insulin action (Supplemental Figure 7). Similarly, antagonizing Epac pathway with brefedlin failed to restore insulin action (Supplemental Figure 7). These results thus suggested that the inhibitory effect of CRFR2 signaling on insulin action does not require either PKA or Epac-mediated pathway. On the other hand, pretreating cells with rapamycin, which blocks mTOR complex 1 (mTORC1), effectively reversed the inhibitory effect of Ucn 2 on insulin-induced glucose uptake (Figure 4C), indicating that mTOR is involved in Ucn 2 inhibitory effect on insulin action.

Effect of CRFR2 signaling on insulin and platelet-derived growth factor (PDGF)-induced glucose uptake

PDGF has been shown to stimulate glucose uptake by signaling through Akt pathway in an IRS-1-independent manner (32, 34, 35). Thus, we examined the effect of Ucn 2 on PDGF-induced glucose uptake in C2C12 myotubes. As expected, PDGF (10 ng/mL) greatly stimulated glucose uptake in muscle cells (Figure 5). Importantly, Ucn 2 completely blocked insulin-induced glucose uptake but had no effect on PDGF-induced glucose uptake (Figure 5). This result thus argues that CRFR2 signaling modulates insulin's action likely at the levels of IRS-1.

Figure 5.

Effect of Ucn 2 on glucose uptake induced by PDGF in C2C12 myotubes. Cells were treated with Ucn 2 (10nM) for 20 minutes followed by treatment of insulin (10nM) or PDGF (10 ng/mL) for 10 minutes. **, P < .01; *, P < .05 vs respective vehicle treatment.

Effect of Ucn2 and rapamycin on insulin-induced tyrosine phosphorylation of IRS-1

Tyrosine phosphorylation of IRS-1 has been shown to be critical in the activation of phosphoinositide 3-kinase and glucose uptake by insulin (36). We thus tested whether CRFR2-induced inhibition of insulin-stimulated glucose uptake was accompanied by decreased IRS-1 tyrosine phosphorylation. A significant interaction between insulin and Ucn 2 was observed on IRS-1 tyrosine phosphorylation (F_1,47 = 18.6, P < .01) with insulin at 10nM significantly increasing (P < .01) IRS-1 tyrosine phosphorylation (Figure 6, A and B). Although stimulation of CRFR2 with Ucn 2 alone did not alter basal IRS-1 tyrosine phosphorylation when compared with control cells (Figure 6A), Ucn 2 treatment significantly suppressed insulin stimulation of IRS-1 tyrosine phosphorylation (Figure 6, A and B). We also examined whether exposure of these cells to Ucn 2 modulates IRS-1 protein content. IRS-1 total protein levels (Figure 6A) were not altered after insulin or Ucn 2 treatment, suggesting that Ucn 2 impairs IRS-1 tyrosine phosphorylation without altering IRS-1 protein synthesis under acute conditions. Similar to IRS-1 phosphorylation, Ucn 2 also significantly attenuated serine phosphorylation of Akt by insulin (Figure 6A). In contrast to tyrosine phosphorylation, Ucn 2 treatment failed to modulate IRS-1 phosphorylation at Ser636/639 (Supplemental Figure 8).

Figure 6.

Effect of Ucn 2 on IRS-1 and Akt phosphorylation in insulin treated C2C12 myotubes. A, Representative Western blottings of tyrosine phosphorylated and total IRS-1 and serine phosphorylated and total Akt in cells pretreated with Ucn 2 (10nM) for 20 minutes followed by 10 minutes of insulin (10nM) stimulation. B, Quantification of tyrosine phosphorylation of IRS-1 in C2C12 myotubes. Bars with different letters were significantly different with P < .05.

We also determined phosphorylation of IRS-1 and Akt in C2C12 myotubes treated with PDGF with or without Ucn 2. Consistent with glucose uptake study shown in Figure 5, PDGF significantly stimulated Akt phosphorylation in C2C12 cells (Figure 7) regardless of Ucn 2 treatment. Importantly, PDGF failed to induce tyrosine phosphorylation of IRS-1 (Figure 7), further confirming that glucose uptake induced by PDGF is independent of IRS-1 (32).

Figure 7.

Effect of Ucn 2 on IRS-1 and Akt phosphorylation in PDGF-treated C2C12 myotubes. A, Representative Western blottings of tyrosine phosphorylated and total IRS-1 and Akt in cells pretreated with Ucn 2 (10nM) for 20 minutes followed by 10 minutes of PDGF (10 ng/mL) treatment. B, Quantification of phosphorylation of IRS-1 and Akt in C2C12 myotubes. Bars with different letters were significantly different with P < .05.

Because rapamycin treatment restored insulin-induced glucose uptake in the face of Ucn 2, we investigated the possible effect of rapamycin on tyrosine phosphorylation of IRS-1 in C2C12 myotubes. As expected, rapamycin treatment blocked phosphorylation of S6K (Figure 8), a downstream effector of mTORC1 (37, 38). Consistent with our functional study (Figure 4), pretreating cells with rapamycin neutralized the inhibitory effect of Ucn 2 on insulin-induced phosphorylation of IRS-1 and Akt (Figure 8).

Figure 8.

Effect of Rapamycin on Ucn 2 inhibition of IRS-1 and Akt phosphorylation in insulin treated C2C12 myotubes. A, Representative Western blottings of phosphorylated and total S6K, IRS-1, and Akt in cells pretreated with either Ucn 2 and/or rapamycin before insulin treatment. B–D, Quantitative bar graphs showing the detection of phosphorylation of S6K, IRS-1, and Akt in C2C12 myotubes. **, P < .01 and *, P < .05 vs vehicle treatment.

Discussion

There is much evidence suggesting that stress is a risk factor in the pathophysiology of muscle insulin resistance and diabetes (4, 5), yet the molecular mechanism by which stress modulates glucose homeostasis remains unclear. The CRF family of peptides plays a pivotal role in coordinating an array of physiological adaptations to stress (7). It has been demonstrated that CRF peptides and their receptors are expressed not only in the brain but also in a number of peripheral tissues, including skeletal muscle, digestive tract, and endocrine pancreas (17, 39–42). Importantly, the levels of CRFR2 in skeletal muscle are greatly elevated in chronically stressed animals or mice fed with high-fat diet (18), conditions that are associated with insulin resistance (19–24). In the present study, we confirmed an early observation (18) that CRFR2 is expressed in differentiated C2C12 myotubes with minimal expression in undifferentiated C2C12 myoblasts. The present study further showed that levels of CRFR2 were significantly elevated in insulin resistant muscle cells both in vivo and in vitro compared with control muscle cells. Moreover, CRFR2 levels in the skeletal muscle of chronic exercised mice were significantly lowered than that in nonexercise sedentary control mice. These results demonstrate that CRFR2 levels are closely correlated with the status of insulin sensitivity, with elevated receptor levels found in insulin resistant cells, whereas reduced levels found in muscle with improved insulin sensitivity. These observations thus support a potential functional role of CRFR2 in regulating insulin sensitivity and glucose homeostasis in muscle cells.

Consistent with this notion, we showed that treating muscle cells with Ucn 2, a selective ligand for CRFR2, significantly inhibited insulin-induced glucose uptake, and this inhibitory effect was mediated by CRFR2, because either pretreating cells with Ast2B, a CRFR2-selective antagonist (43), or suppressing CRFR2 expression by lentiviral-mediated shRNA knockdown abrogated the effect of Ucn 2. In the present study, we observed an interesting discrepancy in insulin-induced glucose uptake between the 2 different approaches we used to abrogate CRFR2 function. Knocking down CRFR2 either further enhanced or restored insulin-induced glucose uptake in normal or insulin resistant muscle cells, respectively. Conversely, acute pharmacological blockade of CRFR2 alone failed to potentiate insulin-induced glucose uptake. The source of the apparent discrepancy is unclear. With CRFR2 shRNA knockdown, muscle cells were infected with lentiviral vector for 72 hours before the experiments. On the other hand, in the pharmacologic approach, cells were treated with Ast2B acutely for 20 minutes before insulin stimulation. Thus, it is reasonable to assume that, unlike acute pharmacological blockade, CRFR2 tone in the receptor knockdown cells is likely reduced before insulin treatment, which may inevitably modulate the sensitivity of cells to insulin. It will be of great interest to determine the mechanism by which chronic CRFR2 signaling modulates insulin signaling machinery. It is also noteworthy that Ucn 2 has been shown to be prominently expressed in skeletal muscle cells (17), thus raising the possibility that Ucn 2 is released from muscle cells and subsequently exerts an autocrine negative feedback mechanism via CRFR2 to modulate insulin action. Currently the regulation of Ucn 2 secretion from muscle cells is not known. A lack of potentiating effect of Ast2B on insulin-induced glucose uptake suggests that endogenous ligands of CRFR2 such as Ucn 2 are not released under basal conditions. Identification of possible factors (eg, insulin) that regulate Ucn 2 secretion in muscle cells will provide significant insight into the physiological role of endogenous CRFR2 signaling in modulating insulin action in muscle cells.

CRFRs have been shown to couple predominately to G_αs to stimulate adenylyl cyclase to promote cAMP production (8). The effects of cAMP are mediated by a number of downstream effectors, including PKA and Epac (25–27). In the present study, we found that MDL-12330A, an adenylyl cyclase inhibitor, blocked the effects of CRFR2 on the insulin-induced glucose uptake, indicating cAMP is important in mediating the inhibitory effect of CRFR2 signaling. On the other hand, both Epac and PKA inhibitors failed to reverse the inhibitory effects of CRFR2 on insulin-induced glucose, suggesting that Epac and PKA are not required in CRFR2-cAMP pathway to inhibit insulin's action. Unexpectedly, we found that rapamycin, a selective antagonist of mTORC1, effectively reversed the inhibition of CRFR2 on insulin-stimulated glucose uptake, suggesting mTOR signaling is involved in the inhibition by CRFR2 signaling on insulin-stimulated glucose uptake.

Extensive studies have shown that insulin induces glucose uptake in muscle cells by stimulating Akt signaling pathway in an IRS protein-dependent manner (1, 44, 45). It has been shown that stimulation of CRFR2 attenuates insulin-induced Akt phosphorylation in primary skeletal muscle cells and C2C12 myotubes (15), suggesting the inhibitory effect of CRFR2 signaling on insulin action probably lies between insulin receptor signaling and Akt phosphorylation. Similar to insulin, PDGF has been shown to stimulate glucose uptake in muscle cells by signaling through Akt pathway in an IRS-1-independent manner (34, 35). Consistent with earlier reports, in the present study, we found that PDGF significantly stimulated glucose uptake in muscle cells. Importantly, stimulation of CRFR2 by Ucn 2 failed to inhibit PDGF-induced glucose uptake and Akt phosphorylation. This result argues that CRFR2 signaling modulates insulin's action likely at the levels of IRS-1. Consistent with this notion, stimulation of CRFR2 greatly attenuated insulin-induced tyrosine phosphorylation of IRS-1. Moreover, similar to functional glucose uptake study, pretreating cells with rapamycin restored insulin stimulated phosphorylation of IRS-1 and Akt. These results together strongly argue that CRFR2 signaling modulates insulin signaling at the levels of IRS-1.

Although it has been shown that mTOR plays an important role as a downstream effector of insulin signaling (46, 47), accumulating evidence has also suggested that it can negatively impact insulin signaling (30, 48–50). Specifically, it has been shown stimulation of the mTOR pathway acutely inhibits insulin signaling to Akt and glucose transport in 3T3-L1 and human adipocytes (49). Currently the mechanism by which mTOR negatively modulates insulin signaling is not completely known. It has been shown that phosphorylation of Ser636/639 of IRS-1 serves as negative feedback mechanism in regulating insulin action (51, 52) and mTOR-mediated activation of S6kinase can increase phosphorylation of the serine residues of IRS-1 (53). In current study activation of CRFR2 failed to modulate phosphorylation of IRS-1 at Ser636/639 with or without insulin treatment, thus suggesting that the inhibitory effect of CRFR2 signaling on insulin action does not require the negative feedback through serine phosphorylation of IRS-1. Also of importance is that in the present study, rapamycin alone had no effect in insulin action. This suggests, at least under present experimental setting, that mTOR signaling is recruited with CRFR2 stimulation.

cAMP has been suggested to regulate mTOR signaling (28, 29). A study by Kim et al (54) has identified a signaling pathway involving phosphodiesterase 4D (PDE4D) and its binding partner, Ras homolog enriched in brain (Rheb), in regulating the activity of mTORC1. Under basal conditions, PDE4D binds Rheb, a critical activator of mTOR, to prevent the molecule from binding and stimulating mTORC1. However, elevated intracellular cAMP levels recruits PDE4D to metabolize cAMP and leading to dissociation of PDE4D from Rheb, which consequently interacts and stimulates mTOR signaling. Both PDE4D and Rheb have been identified in muscle cells, including C2C12 cells (55). Thus, it is conceivable that this pathway may mediate the effect of CRFR2 signaling in attenuating insulin action. More studies are needed to resolve this issue.

Although often considered to be a negative outcome, insulin resistance may alternatively be viewed as an adaptation to an environmental stressor (56). It has been shown acute eccentric exercise actually results in transient insulin resistance (57, 58), and the resistance could not be attributed to circulating hormone and nutrient levels. Although the physiological relevance of this phenomenon is unclear, it is tempting to speculate that CRFR2 signaling in muscle cells is involved in the transient insulin resistance due to acute exercise to tone down insulin action in favor of glucose utilization. On the other hand, long-term exercise has been shown to enhance glucose uptake and improve insulin sensitivity (59–61). The present study showed that endurance exercise reduced CRFR2 expression in muscle, suggesting that lowering CRFR2 expression and signaling is a possible mechanism by which long-term exercise improves insulin sensitivity.

In conclusion, the present study shows that levels of CRFR2 in muscle cells are closely associated with insulin sensitivity. Stimulation of CRFR2 in muscle cells attenuates insulin-stimulated glucose uptake in muscle cells in a cAMP-dependent manner. Rapamycin reversed the inhibitory effect of CRFR2 on insulin induced tyrosine phosphorylation of IRS-1 and glucose uptake indicating the involvement of mTOR pathway in the inhibitory effect of CRFR2 signaling of insulin action (Figure 9). Together, these results indicate that CRFR2 is a possible molecular mechanism that contributes to muscle insulin resistance induced by stress or high-fat feeding.

Figure 9.

Schematic diagram showing the modulation of insulin-mediated glucose uptake in muscle cells by CRFR2 signaling pathway. AC, adenylyl cyclase; Glut4, glucose transporter-4; IR, insulin receptor; PDK1, phosphoinositide-dependent kinase-1; PI3K, phosphoinositide 3-kinase; PIP3, phosphatidylinositol (3,4,5)-triphosphate.

Acknowledgments

We thank Jessica Geisler, Giulia Dula, and Mercedeh Movassagh for providing technical assistance in RNA extraction and real-time PCR analysis. We also thank Dr Peilin Chen for providing comments on the manuscript.

This work was supported by National Institutes of Health Grants R01 DK078049 (to C.L.) and R01 AR050429 (to Z.Y.), the American Diabetes Association Junior Faculty Award 7-11-JF-21 (to T.H.), and the National Institutes of Health Grant R01 DK101946 (to T.H.).

Disclosure Summary: The authors have nothing to disclose.

Abbreviations

 
• Ast2B

  astressin-2B

 
• CRF

  corticotropin-releasing factor

 
• CRFR1

  type 1 CRF receptor

 
• Epac

  exchange proteins activated by cAMP

 
• IRS-1

  insulin receptor substrate 1

 
• MDL-12330A

  cis-N-(2-Phenylcyclopentyl)-azacyclotridec-1-en-2-amine hydrochloride

 
• mTOR

  mammalian target of rapamycin

 
• mTORC1

  mTOR complex 1

 
• PDE4D

  phosphodiesterase 4D

 
• PDGF

  platelet-derived growth factor

 
• PKA

  protein kinase A

 
• RheB

  Ras homolog enriched in brain

 
• S6K

  S6 kinase

 
• shRNA

  small hairpin RNA

 
• Ucn

  urocortin

 
• WT

  wild-type.

Author notes

^*

H.C. and H.L. contributed equally to this work.

Present address for V.L.: Department of Health and Human Physiology, Obesity Research Education Initiative, Fraternal Order of Eagles Diabetes Research Center, University of Iowa, Iowa City, IA 52242.