L6 myoblasts, Rat skeletal muscle cell lines were maintained in DMEM supplemented with10% FBS, 10% antibiotic-antimycotic mix at 37°C under 5% CO₂ atmosphere. Cells were grown at a density of 1 × 10⁵cells/well. For differentiation, the cells were maintained in differentiation media containing 2% horse serum for 5–7 days. The concentration of quercetin was fixed at 10 and 100 μM and the period of incubation at 24 h based on our previous study (Dhanya et al., 2014).

For the evaluation of the effects of kinase inhibitors, fully differentiated L6 myotubes were starved in serum free (SF) medium for 1.5 h, and then the inhibitors of AMPK (20 μM dorsomorphin) or PI3K (100 nM wortmannin) were added, and cells were incubated for 30 min. Cells were then treated with quercetin (10 and 100 μm) for 24 h in SF-medium with/without corresponding inhibitors. The glucose uptake assay was initiated by the addition of 2-NBDG uptake. The cells were then washed twice with cold phosphate-buffered saline (PBS) and the fluorescence in cells were acquired by confocal microscope (Pathway 855, BD Bioscience, San Jose, CA, United States) equipped with filters having excitation (490 nm) and emission (525 nm) in the FITC range. The changes in glucose uptake in cells were analyzed using BD Image Data Explorer software.

Cells were washed with PBS and digested with trypsin. Total cell number were counted before cells are centrifuged at 8000 rpm for 3 min. Cell pellets were suspended in 3% perchloric acid and incubated on ice for 30 min. Within 1 h the pH of the lysate was adjusted between 6 and 8 with 2M KOH/0.3M MOPS. Precipitated salt was separated by centrifugation at 13000 × g for 10 min at 4°C. Aliquoted samples were stored at -80°C. Adenine nucleotide measurements were conducted by HPLC with a Phenomenex Gemini column (5mm, 0.46 cm × 15 cm, C18 110A) as previously described (Hahn-Windgassen et al., 2005). The nucleotides were detected spectrophotometrically at 259 nm and eluted at a flow rate of 1.0 ml/min. Internal standards (7.5 μM ATP, ADP, and AMP in ddH₂O) were used to quantify the samples. The HPLC buffer contained 20 mM KH₂PO₄ and 3.5 mM K₂HPO₄ at pH 6.1.

Mitochondrial membrane potential was measured using mitochondrial staining kit, JC-1 following manufactures instructions. The kit uses the cationic, lipophilic dye, 5,5′,6,6′tetrachloro-1,1′,3,3′tetraethylbenzimidazolocarbocyanine iodide (JC-1). In normal cells, due to the electrochemical gradient, the dye concentrates in the mitochondrial matrix, where it forms red fluorescent aggregates (JC-1 aggregates). Change in mitochondrial membrane potential prevents the accumulation of the JC-1 and thus, the dye is dispersed throughout the entire cell leading to shift from red (JC-1 aggregates) to green fluorescence (JC-1 monomers). The cells after treatments were incubated with a JC-1 staining solution for 20 min at 37°C. The stain was washed off with PBS and examined under spinning disk microscope, and images were collected, and fluorescence intensity was also measured. For JC-1 monomers and aggregates the fluorescence were measured at 490/530 nm and 525/590 nm, respectively. Valinomycin (1 μg/mL) was used as positive control for the measurement of dissipation of mitochondrial membrane potential.

Differentiated L6 myoblast (5–7 days) cultured in 96 black well plates were treated with compounds of standardized concentrations for 24 h. Intracellular calcium levels were detected by staining the various groups with Fura-2AM for 20 min at 37°C. The stain was washed off with PBS and visualized under a spinning disk confocal microscope (Pathway 855, BD Bioscience, San Jose, CA, United States) at an excitation-emission wavelength of 350 and 510 nm, respectively.

Total RNA from pretreated L6 myotubes were isolated using trizol (Invitrogen Corp., Grand Island, NY, United States) according to the manufacturer’s protocol. One microgram RNA was reverse transcribed by Superscript VILO cDNA synthesis kit. The primer sequences for tested genes were; PPIA: Forward- 5′CAAAGTTCCAAAGACAGCAGAAA3′, Reverse- 5′CTGTGAAAGGAGGAACCCTTATAG3′, GLUT 4: Forward- 5′TCGTGTGGCAAGATGTGTAT3′, Reverse- 5′GTGCCTATGTATGTGGGAGAAA3′, Akt: Forward- 5′GAGCTGTGAACTCCTCATCAA3′, Reverse- 5′TCTCCATAGTCCTCTGGGTAAG3′, PI3K: Forward- 5′GTGGACAAAGCAGAAGCATTAC3′, Reverse- 5′ACCCTGTGTTCTTTGTCTAGTG3′; IRS: Forward- 5′GAGTTGAGTTGGGCAGAGTAG3′, Reverse- 5′CATGTAATCACCACGGCTATTTG3′, AMPK: Forward- 5′CCTATGAAGAGGGCCACAATAA3′, Reverse- 5′AGGTCACGGATGAGGTAAGA3′, CaMKK: Forward- 5′CGCTGGTTCCCACTCTTATC3′, Reverse- 5′GCTCCCTGACTCTTTGCTATT3′, MAPK: Forward- 5′CCCAAGGCCCAGAAATATGA3′, Reverse- 5′AAGAACTGGCTTGGAGATGG3′. Peptidylprolyl isomerase A (PPIA) was used as reference gene. Quantification was performed using a real-time PCR system (Bio-Rad, Hercules, CA, United States) with SYBR green. The cycling parameters were as follows: initial denaturation at 95°C for 1 min, followed by 40 cycles of denaturation at 95°C for 20 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s. Results were presented as levels of expression relative to those of controls after normalization to PPIA using the 2^-ΔΔC_T method. The analysis was carried out in triplicates.

Differentiated L6 myoblast (5–7 day) cultured in 6-well plates were treated with compounds of standardized concentrations for 24 h. L6 cells were homogenized in 1 ml of RIPA lysis buffer (25 mM Tris-HCl pH 7.4, 25 mM NaCl, 0.5 mM EDTA, 1% Triton-X-100, 0.1% SDS) for 30 min on ice and were centrifuged at 12000 rpm for 10 min. The supernatants were collected, and protease inhibitor cocktail was added (Roche, Mannheim, Germany). Supernatants were then stored at -80°C until analysis. Upon thawing, protein content was assayed by the bicinchoninic acid method standardized to bovine serum albumin (Roche, Laval, QC, Canada). Each sample were loaded at approximately 40 μg on 10% polyacrylamide mini gels and transferred to nitrocellulose membrane (Millipore, Bedford, MA, United States). Membranes were blocked for 1 h at room temperature and then incubated overnight at 4°C in blocking buffer with appropriate phospho-specific or pan-specific antibodies against AMPK, P38MAPK, GLUT4, AKT, IRS1, IRS2 (each at 1: 500 to 1:1000). Followed by incubation with anti-rabbit HRP-conjugated secondary antibodies at 1:1000 to 1:4000 (Santa Cruz, CA, United States). The revelation was performed using diaminobenzidine method (Sigma, United States). Gel band intensities were evaluated by densitometry analysis using Image densitometry software (Quantity one-4.6.7, 1D analysis software, Bio-Rad, United States).

Following pretreatment with quercetin (10 and 100 μM) cells were washed with PBS and fixed for 5 min with 4% formaldehyde in PBS and quenched with 50 mM glycine in PBS for 10 min. Cells were blocked with 5% BSA in PBS for 1 h and incubated with monoclonal GLUT4 antibody solution (1:200 dilution in 1.5% BSA in PBS) at 4°C overnight followed by washing with PBS and 1 h incubation at room temperature with FITC-conjugated goat anti-mouse IgG secondary antibody (1:500 dilution, 1.5% BSA in PBS). Following PBS wash, images were acquired using laser scanning confocal microscope (Nikon A1R, Nikon Instruments, Melville, NY, United States) equipped with filters in the FITC range (i.e., excitation, 490 nm; and emission, 525 nm). Images were analyzed by NIS ELEMENTS software.

Results were expressed as means and standard deviations of the control and treated cells from triplicate measurements (n = 3) of three different experiments. Data were subjected to one-way ANOVA, and the significance of differences between means was calculated by Duncan’s multiple range test using SPSS, standard version 16 (SPSS, Inc.), and significance was accepted at P ≤ 0.05.

From our previous study, we got evidence for the possible antidiabetic potential of quercetin, as it stimulated 2-NBDG uptake in L6 myotubes via GLUT 4 translocation (Dhanya et al., 2014). As the effect of quercetin on glucose uptake was greater in magnitude than insulin, we assumed that it may employ other routes to attain its effect. To investigate whether the enhanced glucose uptake in L6 myotubes was mediated through PI3K activation, we examined the effects of selective inhibitors of PI3K and AMPK on 2-NBDG uptake. The effect of quercetin on 2-NBDG uptake was wortmannin insensitive as shown in Figure 1A, indicating that the insulin signaling pathway upstream of PI3K is not involved. Dorsomorphin treatment reduced the glucose uptake up to 80% as indicated in Figure 1B pointing to the conclusion that AMPK pathway has a significant role in 2-NBDG uptake induced by quercetin.

The cellular AMP, ADP and ATP concentrations of L6 myotubes on pretreatment of quercetin (100 μM) was measured by HPLC as indicated in Figure 2A. Quercetin pretreatment caused twofold increase in both AMP to ATP as well as ADP to ATP ratio as indicated in Figure 2B. Interestingly, besides the AMP: ATP ratio, AMP, ADP and ATP concentrations significantly increased in treated myotubes. This may be the result of increased ATP utilization in these cells. The AMP: ATP ratio is an essential factor for cellular AMPK activation. As the cellular AMP: ATP ratio varies as the square of the ADP: ATP ratio (Supplementary Table S1), we used the ADP: ATP ratio as a surrogate measure.

A transient depolarization was observed in L6 myotubes pretreated with quercetin (Supplementary Figure S1) and its relative fluorescence intensity was analyzed by BDI Explorer software as shown in Figure 3A. As we had earlier proved quercetin to be non-cytotoxic (≤10% toxicity at 100 μM for 24 h) (Dhanya et al., 2014), we ruled out the possibility of apoptosis or the opening of mitochondrial transition pore. Pretreatment of quercetin increased the AMP to ATP ratio which tightly correlated with its effect on mitochondrial membrane depolarization.

To assess the possible change in intracellular calcium levels in treated cells, a fluorescent indicator, Fura 2 AM was used. Our results showed that the flavonoid impaired calcium homeostasis by significantly increasing cytosolic calcium levels in L6 myotubes. Pretreatment of L6 myotubes with quercetin triggered an increase in cytosolic calcium levels as compared to control (Figure 3C), and its intensity was analyzed and represented in Figure 3B.

To evaluate the gene expression of the key signaling molecules involved in insulin and AMPK pathway, the mRNA levels were quantified by real-time PCR as shown in Figure 4A. Pretreatment of quercetin, at 100 μM showed a 14-fold increase in GLUT4 mRNA levels relative to control. There was a significant upregulation in Akt expression in L6 myotubes on quercetin pretreatment. CaMKK, AMPK, and MAPK, the key signaling molecules involved in AMPK were upregulated more than fivefold compared to control. Irs and PI3K showed onefold increase in mRNA levels on pretreatment compared to control.

To provide insight into the possible mechanism of glucose uptake stimulated by quercetin, the protein expression of IRS1, IRS2, AKT, P38MAPK and AMPK in L6 myotubes were investigated on pretreatment of quercetin by western blotting. Since the activation depends on the phosphorylation of the key signaling molecules, the expression of phosphorylated molecules were also examined. P38MAPK, AMPK, and AKT showed onefold increase in expression and phosphorylation of these proteins were also enhanced on pretreatment. GLUT4 expression increased to twofold in L6 myotubes, as expected there was a feeble expression of both IRS1 and IRS2 in L6 myotubes as shown in Figure 4B.

GLUT4 upregulation in L6 myotubes was monitored by immunoassay with fluorescent labeled secondary antibody at 24 h pretreatment. Effect of quercetin (100 μM) on translocation was much higher than that of positive control as shown in supplementary data, suggested the end molecular mechanism behind the induction of AMPK pathway in L6 myotubes. The results obtained by quantifying immunologically labeled GLUT4 receptors at the surface of intact cells were depicted in Supplementary Figure S2.