The H9C2 cell line, which derives from embryonic rat heart cells, was purchased from the Cell Bank of Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). SDG was purchased from Sigma-Aldrich (Merck Millipore; Darmstadt, Germany; purity >97% as determined by HPLC). JAK2/STAT3 inhibitor WP1066 was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA; purity >97% as determined by HPLC). Antibodies against poly-(ADP-ribose) polymerase (PARP, cat. no. 9532), p-JAK2 (Tyr-1007/Tyr-1008, cat. no. 3771), JAK2 (cat. no. 3230), p-STAT3 (Tyr-705, cat. no. 9145), p-STAT3 (Ser-727, cat. no. 9134), STAT3 (cat. no. 12640), Src (cat. no. 2191), B-cell lymphoma-extra-large (Bcl-xL, cat. no. 2762) and induced myeloid leukemia cell differentiation protein (Mcl-1, cat. no. 94296) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). P-Src (Tyr-418, cat. no. ab40660) was purchased from Abcam (Cambridge, MA, USA). Annexin V/propidium iodide (PI) Apoptosis Detection kit was purchased from BD Biosciences (San Jose, CA, USA). Remaining reagents used in the study were obtained from commercial sources.

H9C2 cells were plated at a density of 5.×10³ cells per well in 96-well plates. Cells were treated with 50, 100 and 150 μM SDG with or without 5 μM JAK2/STAT3 inhibitor for 24 h at 37°C and were treated with 400 μM H₂O₂ for 2 h at 37°C. Cell viability was detected using an MTT assay as previously described (17). Briefly, following treatment, the culture medium was removed from each well and replaced with 0.5 mg/ml MTT solution for 4 h at 37°C. Following incubation, culture medium was removed and 150 μl dimethyl sulfoxide (Sigma-Aldrich; Merck Millipore) per well was added for formazan solubilization. The formazan was quantified by determining the absorbance at 490 nm, using an microplate reader (Thermo Fisher Scientific, Inc., Walham, MA, USA). The viability of the cardiomyocytes (%) was calculated as follows: Viability (% of control)=optical density (OD) mean test group/OD mean control group ×100.

H9C2 cells were plated at a density of 5.0×10³ cells/well in 96-well plates. H9C2 cells were pretreated with 50, 100 and 150 μM SDG for 24 h at 37°C, and were stimulated with 400 μM H₂O₂ for 2 h at 37°C. Subsequently, 96-well plates were centrifuged at 400 × g for 5 min at 4°C to obtain the cellular supernatant and the LDH release was quantified using LDH Release assay kit (cat. no. C0016; Beyotime Institute of Biotechnology, Haimen, China) following the manufacturer's protocol.

Flow cytometry (BD Biosciences) was performed to quanitfy cell apoptosis. H9C2 cells were treated with indicated doses of SDG with/without the JAK2/STAT3 inhibitor for 24 h and then treated with H₂O₂ for 2 h. The apoptosis index of H9C2 cells was detected using the Annexin V/PI Apoptosis Detection kit following the manufacturer's protocol. The data was analyzed using FlowJo version 6.0 (Tree Star, Inc., Ashland, OR, USA).

Western blot assay was performed as previously described (18). Briefly, protein extracts of cultured cells were prepared in RIPA lysis buffer (Beyotime Institute of Biotechnology, Haimen, China). Protein concentration of each sample was detected using BCA protein assay kit (Thermo Fisher Scientific, Inc.). Protein samples (15 μg) were separated by 10% sodium dodecyl sulfate gel electrophoresis, and were transferred onto polyvinylidene fluoride membranes (Merck Millipore), subsequently blocked in Tris-buffered saline including 5% non-fat milk. Membranes were incubated at 4°C overnight with the corresponding primary antibodies [PARP 1:1,000, p-STAT3 (Tyr-705) 1:1,000, p-STAT3 (Ser-727), STAT3 1:1,000, p-JAK2 (Tyr-1007/ Tyr-1008) 1:1,000, JAK2 1:1,000, p-Src (Tyr-418) 1:1,000, Src 1:1,000, Mcl-1 1:1,000, Bcl-xL 1:1,000, GAPDH 1:5,000]. Then membranes were incubated with horseradish peroxidase-conjugated anti-rabbit secondary antibody (cat no. 7074; 1:1,000; Cell Signaling Technology, Inc.) for 1 h at 37°C. Each experiment was repeated three times. Band intensities were quantified using ImageJ version 1.47 (National Institutes of Helath, Bethesda, MD, USA) and normalized to the intensities of GAPDH loading control bands.

Molecular docking and dynamics analyses (MD) were performed to investigate the mechanism of interaction between SDG and JAK2. X-ray crystallographic structures of JAK2 (PDB ID: 4C61) (19) were used as starting structures for prediction of the binding site on SDG (CID: 9917980). The crystallographic structures of the JAK2 protein were prepared for molecular docking by removing water and co-crystallized ligands. Energy minimization of ligands was performed using YASARA version 16.9.23 (www.yasara.org). The protein kinase domain of JAK2 was defined in the present study as the protein-ligand binding site of JAK2. Molecular docking simulation was performed using Autodock Vina version 1.1.2 (vina.scripps.edu; The Scripps Research Institute, La Jolla, CA, USA) (20). PyMol version 1.3 and Ligplot⁺ version 1.4 were used to visualize the JAK2-SDG complex (21,22). The conformation with the lowest binding energy (lower is better) was used as star conformation for MD simulation.

MD simulation was performed using YASARA (23). All simulations were run with the AMBER ff99sb force field option selected. The JAK2-SDG complex was solvated in a cube box of 0.9% NaCl, and the distance between solute and the box was 5 Å. Simulations were run at 298 K followed by simulated annealing minimizations which start at 298 K and velocities were scaled down with 0.9 every ten steps for 500 steps. Subsequently, 70 ns MD simulation was performed with a time step of 2 fsec and the simulation snapshots were saved every 100 psec.

Data are presented as the mean ± standard deviation. Calculations were performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL, USA). One-way analysis of variance was performed to determine the significance of multiple comparisons. Least Significant Difference was used for statistical comparisons between treatment groups. P<0.05 was considered to indicate a statistically significant difference.

To investigate whether SDG prevents H9C2 from oxidative stress-mediated apoptosis, cells were pretreated with various doses of SDG for 24 h and subsequently stimulated with H₂O₂ for 2 h. Cell viability of H9C2 was reduced by H₂O₂ treatment, but was increased in a dose-dependent manner in SDG treatment groups (Fig. 1A). LDH release assay indicated that treatment with H₂O₂ significantly increased LDH release, whereas SDG treatment reduced it, in a dose-dependent manner (Fig. 1B). Subsequently, the protective effect of SDG against H₂O₂-induced apoptosis was examined, by detecting PARP expression and performing Annexin/PI assay. It was determined that SDG treatment led to a dose-dependent reduction of cleaved PARP expression levels (Fig. 1C). Accordingly, SDG treatment significantly reduced the apoptosis index of H9C2 cells (Fig. 1D and E). This suggested that SDG protected cardiomyocytes against oxidative stress-mediated cell apoptosis.

It has been previously established that STAT3 activation prevents cardiomyocyte apoptosis (15,16). Therefore, whether SDG increased the activity of STAT3 in H₂O₂-treated H9C2 cells was investigated. It was demonstrated that SDG caused a dose-dependent increase in phosphorylated STAT3 (p-STAT3, at Tyr-705 and Ser-727) expression levels (Fig. 2A-C). It was further examined whether SDG affected STAT3-target genes in H9C2 cells. Mcl-1 and Bcl-xL are target genes of STAT3 involved in cell apoptosis (24). It was observed that Mcl-1 and Bcl-xL protein levels were significantly increased in H₂O₂-treated H9C2 cells following SDG treatment (Fig. 2D-F). These findings indicated that SDG rescued STAT3 signaling activation in H₂O₂-treated H9C2 cardiomyocytes.

To determine whether SDG activated the STAT3-associated signaling pathway directly or through inhibition of oxidative stress, phosphorylation levels of STAT3 following pretreatment with indicated concentrations of SDG for 24 h were investigated. As illustrated in Fig. 3A-C, pretreatment with SDG led to a dose-dependent increase of expression levels of p-STAT3 at the Tyr-705 and Ser-727 sites. It has been previously established that STAT3 may be activated by JAK2 and non-receptor tyrosine kinase Src (25,26). Therefore, the present study determined if activation of JAK2 and/or Src occurred in SDG-treated H9C2 cells. As presented in Fig. 3D-F, phosphorylation of Src did not change following SDG treatment, and expression of p-JAK2 was increased in SDG-treated groups. These findings suggested that SDG may directly activate the JAK2/STAT3 signaling pathway.

In order to determine if the anti-apoptotic effect of SDG in H9C2 cells was due to the activation of the JAK2/STAT3 signaling pathway, WP1066 which is a JAK2/STAT3 inhibitor (27) was used. As illustrated in Fig. 4A-D, phosphorylation of JAK2 and STAT3 was increased by SDG treatment; however, it was reduced in H9C2 cells treated with with WP1066 and with/without SDG. These results indicated that co-treatment with WP1066 inhibited the activation effect of SDG on the JAK2/STAT3 signaling pathway. Following treatment with SDG and/or WP1066 for 24 h, H9C2 cells were treated with H₂O₂ for 2 h, and phosphorylation of JAK2 and STAT3 was measured by western blot analysis. As demonstrated in Fig. 4E-J, co-treatment of WP1066 significantly reduced the phosphorylation levels of JAK2 and STAT3, as well as the expression levels of Mcl-1 and Bcl-xL, were increased by SDG in H₂O₂-treated H9C2 cells. These results confirmed that WP1066 abolished the activation effect of SDG on the JAK2/STAT3 signaling pathway. Finally, MTT and Annexin V/PI staining was performed to determine the role of JAK2/STAT3 signaling in the anti-apoptotic effect of SDG. As demonstrated in Fig. 4K, viablity of H9C2 cells was significantly reduced by treatment with H₂O₂ or WP1066 alone. SDG treatment rescued the viability of H₂O₂-stimulated H9C2 cells, whereas inhibition of JAK2/STAT3 impeded the protective effects of SDG on cell viability, reducing it from 82.5 to 50.3%. Accordingly, inhibition of JAK2/STAT3 partly abolished the anti-apoptotic effects of SDG on apoptotic rate from 10.8 to 25.6% (Fig. 4L). These results implied that the anti-apoptotic effect of SDG was, partially at least, due to the activation of the JAK2/STAT3 signaling pathway.

The present findings demonstrated that SDG significantly enhanced the phosphorylation levels of JAK2. Therefore, it is possible that SDG directly activates JAK2. Subsequently, the interaction between the protein kinase domain of JAK2 and SDG was investigated, by molecular docking and MD simulation. SDG molecule was docked into the protein kinase domain of the JAK2 protein (PDB: 4C61). The binding energy of the JAK2-SDG complex was −8.258 kcal/mol. Three- and two-dimensional structures of the JAK2-SDG complex are presented in Fig. 5A and B. Two hydrogen bonds were formed between SDG and JAK2 Leu-932 and Gln-853. The distance of hydrogen bonds between SDG and JAK2 Leu-932 and Gln-853 were Å, 2.85 and 2.80 Å, respectively. Two hydrogen bonds were formed between SDG and Arg-938 of JAK2, and the distance of hydrogen bonds between them was 2.99 and 2.95 Å, respectively. The best conformation of JAK2-SDG was used as the star conformation for the MD simulation by YASARA. Fig. 5C demonstrates the evolution of backbone atoms and the root-mean-square deviation (RMSD) of the complex with respect to the minimized structure. The RMSD track in the last 50 nsec fluctuated from 2.52 to 4.14 Å in the unbound state of the JAK2 protein, and from 2.32 to 3.26 Å in the JAK2-SDG complex. These results suggested strong binding between the kinase domain of JAK2 and SDG, which indicated that SDG may directly target JAK2.