AICAR was obtained from Wako Pure Chemical Industries. Antibodies for phospho-AMPKα (Thr172) (40H9), total-AMPKα, phosphor-ACC (Ser79), total-ACC and LKB1 were obtained from Cell Signaling Technology. GAPDH antibody was obtained from Merck Millipore. Secondary antibodies for mouse IgG and rabbit IgG were obtained from Santa Cruz Biotechnoligy.

siRNA for DARP was obtained from Dharmacon, and the negative control siRNA (scramble) was obtained from Ambion.

C2C12 mouse skeletal myoblast cell line was cultured in DMEM supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin. Myogenesis was induced as described previously [10]. Briefly, when cells reached 70–80% confluency, the growth medium was switched to differentiation medium (DMEM supplemented with 2% horse serum and 1% penicillin and streptomycin) to induce myogenesis. Differentiation medium was changed every other day until cells well differentiate into myotubes (regularly for 5–9 days).

Differentiated C2C12 myotubes (3–5 days after giving differentiation medium) were transfected with 50 nM siRNA for either DARP or scramble by using Lipofectamine RNAiMAX (Invitrogen), according to the manufacturer's instructions. The fresh differentiation medium was given after 24 h, and cells were incubated for another 48 h before use for experiments.

Preparation of retrovirus was performed as reported previously [11]. Briefly, cDNA for targeting gene (DARP-FLAG) was subcloned into pMSCVneo vector (Clontech). GP2-293 packaging cells were transfected with DARP-pMSCVneo plasmid and pVSV-G plasmid (Clontech) using Lipofectamine 3000. In a parallel way, GP2-293 cells were transfected with GFP-pMSCVneo and pVSV-G plasmids to prepare viruses for negative control. Fresh growth medium was replaced 24 h after transfection, and cells were incubated for another 24h, followed by collection of the virus-containing culture medium. For infection, myoblasts of 50–60% confluency were incubated in the mixture of growth medium and the virus-containing culture medium at 1:1 ratio in the presence of 8 μg/ml polybrene for 24 h. Thereafter, cells were given fresh medium and incubated for 24–48 h followed by protein extraction.

All protocols for animal experiments were approved by the Ethics Review Committee for Animal Experimentation in Kyoto Prefectural University of Medicine. The total number of mice used in this study is 72. Detailed method for the generation of DARP-deficient mice was described previously [12]. Exon 1–3 of DARP gene were replaced by cDNA encoding LacZ and a pGK neo cassette in the targeted allele.

DARP-/- and WT mice (C57BL6) were fed with normal chow (containing 23.1% protein and 5.1% fat). For all the experiments using mice, mice were regularly anesthetized with isoflurane prior to every procedure to minimize suffering of the mice. When sacrifice the mice, mice were anesthetized with isoflurane and then sacrified by cervical dislocation.

Treatment with AICAR in vivo was performed as described previously [13]. C57BL6 mice and DARP-/- mice were given an intraperitoneal injection of AICAR (0.25 mg/g) for 5 days prior to the ipGTT analysis. The condition of mice treated with AICAR was regularly monitored every day.

Treatment with AICAR in vitro was performed as described previously [14]. Differentiated C2C12 myotubes transfected with scramble or DARP siRNA were treated with AICAR at a final concentration of 1 mM for 1 h, followed by protein extraction in RIPA buffer.

Cell or tissue lysates were prepared in RIPA buffer containing protease and phosphatase inhibitors, and then immunoblotting was performed as described previously [15]. Lysates containing the same amount of proteins (~60 μg for culture cells and ~120 μg for quadriceps muscles) were subjected to SDS—PAGE, followed by transferring onto the nitrocellulose membrane. The membranes were blocked in 5% non-fat milk in TBS containing 0.05% tween20 at room temperature for 1 h. Membranes were then incubated with specific antibodies for target molecules. The dilution of primary antibodies was; phospho-AMPKα (1:1000), total-AMPKα (1:1000), phospho-ACC (1:1000), total-ACC (1:1000), LKB1 (1:1000), GAPDH (1:2000) and secondary antibodies for mouse IgG and rabbit IgG (1:4000).

Quantification of mRNA expression of target genes was performed as described previously [16]. Total RNAs of skeletal muscle and C2C12 cells were isolated by using Trizol (Invitrogen), followed by purification with NucleoSpin RNA Clean-up (MACHEREY-NAGEL). Complementary DNA was synthesized from 0.5–1 μg total RNA using PrimeScript RT Reagent kit with gDNA Eraser (TaKaRa). PCR reactions were prepared by using KAPA SYBR FAST Master Mix Universal (KAPA BIOSYSTEMS) followed by the real-time PCR using Thermal Cycler Dice (TaKaRa). Nucleotide sequence of each primer is shown in Table 1. mRNA levels for target genes relative to 18S rRNA or actin was shown for all the experiments.

The intraperitoneal glucose tolerance test (ipGTT) and insulin tolerance test (ITT) were performed as reported previously [17]. Briefly, for the ipGTT, mice were fasted for 8 h by depleting chows in the morning, and subsequently D-glucose was intraperitoneally administrated at the dose of 1.5 g/kg. Blood was collected at indicated time points, and blood glucose was measured by the glucose oxidase method (Sanwa Kagaku). For the ITT, mice in the absence of fasting treatment were subcutaneously given human insulin at the dose of 1 U/kg. Blood was collected at indicated time points, and blood glucose was measured. For quantitative analysis, the are-under-curve for ipGTT and ITT was also calculated.

Myotubes were serum-starved for 6 h and subsequently incubated with DMSO (control) or 2 mM AICAR for 15 min. Glucose uptake was determined by 2-deoxyglucose uptake with an enzymatic photometric assay by using 2-deoxyglucose uptake measurement kit (COSMO BIO Co. Ltd.) according to the manufacturer’s instructions.

Oxidation of glucose was assessed by measuring the production of ¹⁴CO₂ from (¹⁴C)-D-glucose as described previously [18]. Briefly, myotubes transfected with either scramble or DARP siRNA were incubated with serum- and glucose-depleted medium containing fatty acid-free BSA and 5550 Bq/mL (¹⁴C)-D-glucose at a ratio of 1:2.5 in the presence or absence of AICAR (2 mM). To measure conversion of (¹⁴C)-D-glucose into ¹⁴CO₂, a piece of Whatman papers wet with phenylethylamin/methanol (1:1) were taped onto the top of culture dish (35mm) to trap the CO₂ produced during the incubation period. After incubation for 2 h, 200 μl of H₂SO₄ (4 mol/l) was added to release all the ¹⁴CO₂, and cells were further incubated for 1 h. Subsequently, the seal was removed, and the pieces of Whatman papers were carefully collected and their radioactivities were measured by liquid scintillation counting.

All data are presented as mean±S.E. Differences between groups were analyzed by the Student’s t-test or one-way ANOVA. P<0.05 was considered statistically significant.

We previously reported that DARP is highly expressed in the skeletal muscle, and modest expression was detected in the heart and brown adipose tissue, assessed by northern blot analysis [7]. Therefore, we quantitatively analyzed the mRNA expression of DARP in mouse tissues, and found that DARP is expressed predominantly in skeletal muscle (Fig 1A). We then examined whether DARP expression is differentially regulated during myogenesis. C2C12 myoblasts expressed DARP at considerably low level, and DARP expression was substantially increased during myotube differentiation, indicating that DARP is predominantly expressed in mature myotube (Fig 1B). We also investigated ANKRD2 expression during myogenesis. ANKRD2 showed the lesser degree of upregulation during myogenesis, and undifferentiated myoblasts expressed ANKRD2 at a comparable level to mature myotubes (Fig 1B).

To investigate a potential role of DARP in energy metabolism in vivo, we analyzed metabolic phenotypes of DARP-/- mice. DARP-/- mice showed body weight similar to wildtype (WT) mice, with a little tendency to increase (Fig 2A). We then performed glucose and insulin tolerance test using these mice. DARP-/- mice exhibited significantly reduced blood glucose levels after intraperitoneal administration of glucose comparing to those in WT mice, despite similar body weight (Fig 2B and 2C). However, insulin sensitivity assessed by insulin tolerance test was not different between DARP-/- and WT mice (Fig 2D). These results suggest that genetic loss of DARP enhances glucose tolerance without affecting insulin sensitivity.

We then analyzed the expression of GLUT1 and GLUT4 in skeletal muscle. Expression of GLUT4 in skeletal muscle was not different between WT and DARP-/- mice, while GLUT1 expression was reduced in DARP-/- mice comparing to that in WT mice (Fig 3A). Because GLUT-1 expression is regulated by HIF-1α, we examined the expression levels of other HIF-1α target genes in skeletal muscle of these mice. The expression of vascular endothelial growth factor (VEGF) and heme oxygenase-1 (Hmox1) in skeletal muscle was similar in WT and DARP-/- mice (Fig 3B). We also analyzed expression of GLUT1 and GLUT4 in C2C12 myotubes transfected with either scramble or DARP siRNA. Gene-silencing of DARP did not affect the expression of both GLUT1 and GLUT-4 in C2C12 myotubes (Fig 3C). These data suggest that there is no direct link between DARP and HIF-1α/GLUT1. Of note, DARP-/- mice demonstrated significantly accentuated phosphorylation of AMPK in skeletal muscle comparing to that in WT mice (Fig 3D). In consistent with the enhanced AMPK activity, phosphorylation of acetyl-CoA carboxylase (ACC) was also accentuated in skeletal muscle of DARP-/- mice (Fig 3E).

Because AMPK mediates insulin-independent glucose uptake in muscle [19–21], these results suggest that loss of DARP might enhance glucose tolerance by activating AMPK. Therefore, we further examined the role of DARP in AMPK activation. Gene-silencing of DARP significantly enhanced phosphorylation of AMPK and ACC in C2C12 myotube (Fig 4A and 4B). Furthermore, we found that DARP-silencing significantly increased glucose uptake in C2C12 myotubes (Fig 4C). Treatment with AICAR, an AMPK activator [20], enhanced phosphorylation of AMPK and ACC in C2C12 myotubes, and the difference of their phosphorylation between control and DARP-knocked down cells disappeared after treatment with AICAR (Fig 4B). Accordingly, the increase of glucose uptake by DARP-silencing was abrogated by the treatment with AICAR (Fig 4C). In addition, glucose oxidation was enhanced by DARP-silencing in C2C12 myotubes, and this increase of glucose oxidation was also abolished by AICAR-treatment (Fig 4D and 4E). In contrast, overexpression of DARP significantly reduced the phosphorylation of AMPK in C2C12 myoblasts comparing to the control GFP-transfected cells (Fig 4F). These data strongly suggest that DARP negatively regulates AMPK activity in the skeletal muscle.

We finally investigated whether AMPK activation is attributable to the enhanced glucose tolerance observed in DARP/- mice. Administration of AICAR abolished the better glucose tolerance in DARP-/- mice, indicating that enhanced AMPK activity largely accounts for the enhanced glucose tolerance induced by DARP-deletion (Fig 5A). We confirmed that phosphorylation of AMPK and ACC in skeletal muscle was not different between WT and DARP-/- mice after the treatment with AICAR (Fig 5B). To investigate the mechanisms responsible for the modification of AMPK activity by DARP, we examined the expression of LKB1, a major upstream kinase that phosphorylates AMPK [4,22]. Gene-silencing of DARP significantly increased protein expression of LKB1 in skeletal muscle in vivo and C2C12 myotubes in vitro, suggesting that DARP modifies AMPK activity at least partially by modulating LKB1 expression in skeletal muscle (Fig 5C and 5D). In contrast, mRNA expression of LKB1 was not affected by DARP-silencing, indicating that DARP modifies the LKB1 expression at protein levels but not mRNA levels (Fig 5E and 5F).

Taken together, we unveiled a previously unknown function of DARP in the regulation of glucose homeostasis by modulating AMPK activity, shedding light on DARP as a new therapeutic target for the treatment of blood glucose dysregulation in diabetic patients.