Resveratrol attenuates constitutive STAT3 and STAT5 activation through induction of PTPε and SHP-2 tyrosine phosphatases and potentiates sorafenib-induced apoptosis in renal cell carcinoma

Resveratrol (RES), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), propidium iodide (PI), Tris base, glycine, NaCl, sodium dodecylsulfate (SDS), and bovine serum albumin (BSA) were purchased from Sigma-Aldrich (St. Louis, MO). RPMI 1640, fetal bovine serum (FBS), antibiotic-antimycotic mixture, and LightShift® Chemiluminescent EMSA kit were obtained from Thermo Fisher Scientific Inc. (Waltham, MA). 5’-biotinylated STAT3 and STAT5 was from Bioneer Corporation (Daejeon, Korea). Alexa Fluor® 488 donkey anti-rabbit IgG (H + L) antibody, and 0.4 % trypan blue vital stain, and TMRE (tetramethylrhodamine, ethyl ester) were obtained from Life Technologies (Grand Island, NY). Anti-phospho-STAT3(Tyr705), anti-phospho-STAT3(Ser727), anti-phospho-JAK1(Tyr1022/1023), anti-JAK1, anti-phospho-JAK2(Tyr1007/1008), anti-JAK2, and anti-phospho-Src(Tyr416) antibodies were purchased from Cell Signaling Technology (Beverly, MA). Anti-STAT3, anti-phospho-STAT5(Tyr 694/Tyr 699), anti-STAT5, anti-Src, anti-PTPε, anti-SHP-2, anti-bcl-2, anti-bcl-xL, anti-survivin, anti-IAP-1, anti-IAP-2, anti-COX-2, anti-VEGF, anti-MMP-9 (matrix metalloproteinase-9), anti-caspase-3, anti-cleaved caspase-3, anti-PARP, anti-cyclin D1, anti-cyclin E, anti-Bax, anti-p21, anti-p53, anti-β-actin, and horseradish peroxidase (HRP)-conjugated secondary antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Annexin V staining kits (ApoScan) were purchased from BioBud (Seoul, Korea). TUNEL (terminal transferase mediated dUTP-fluorescein nick end labeling) assay kit was from Roche Diagnostics GmbH (Mannheim, Germany).

Human Renal cell carcinoma Caki-1 and 786-O were obtained from the American Type Culture Collection (Manassas, VA). Caki-1 and 786-O cells were cultured in RPMI 1640 medium containing 10 % FBS. Media were also supplemented with 100 U/ml of penicillin and 100 μg/ml of streptomycin.

Western blot analysis was performed using a method described previously [36].

Electrophoretic mobility shift assay (EMSA) was performed as described previously [36]. The membrane was detected following manufacturer instructions using LightShift® Chemiluminescent EMSA kit (Waltham, MA).

Immunocytochemistry was performed as described previously [37].

Reverse transcription polymerase chain reaction was performed using a method described previously [38].

Caki-1 and 786-O cells were plated in 6-well plates and allowed to adhere for overnight incubation. On the day of transfection, 6 μl of Lipofectamin 2000 (Invitrogen, Carlsbad, CA) were added to 50 nM PTPε and SHP-2 siRNA in a final volume of 1 ml culture medium. After 5 h of transfection, cells were treated with RES for 8 h and whole-cell extracts were prepared for PTPε, SHP-2, phospho-STAT3, and STAT3 analysis by Western blot analysis.

Cell cycle analysis was performed as described previously [37].

Annexin V assay was performed using a method described previously [37]

Mitochondrial membrane potential assay was performed using a method described previously [37].

Clonogenic assay was performed as described previously [39].

Real-time cell proliferation analysis assay was performed using a method described previously [39].

Invasion assay was performed as described previously [37].

Cell viability was measured by an MTT assay to detect NADH-dependent dehydrogenase activity [40], as described previously [39].

Analysis of drug interactions were performed using a method described previously [41]

Statistical analysis was performed by Student’s t-test and one way analysis of variance, (ANOVA). A p value of less than 0.05 was considered statistically significant.

We first tested whether RES (structure of resveratrol, Fig. 1a) inactivated STAT3 in Caki-1 and 786-O renal cancer cells. Previous studies have shown that STAT3 is a key point of convergence of multiple oncogenic signaling pathways [42‐44]. The ability of RES to modulate constitutive STAT3 activation in RCC cells was investigated by Western blot analysis using antibodies which recognize STAT3 phosphorylation at tyrosine 705 and serine 727. As shown in Fig.1b, RES inhibited the constitutive activation of STAT3 in both Caki-1 and 786-O cells in a dose-dependent manner, with maximum inhibition occurring at around 50 μM, but had no effect on the expression of STAT3 protein (Fig. 1b, third panels). Whether RES affects the activation of another STAT family isoform, STAT5 in renal cancer cells was also investigated. We noticed that RES substantially reduced constitutive STAT5 activation in both Caki-1 and 786-O cells, without affecting total STAT5 levels as analyzed by Western blot analysis (Fig. 1c).

We tested the effect of RES on STAT3 and STAT5 in nuclei. As shown by Western blot analysis in Fig. 1d, RES inhibits nuclear STAT3 and STAT5 in Caki-1 and 786-O cells.

We next determined whether RES can modulate the DNA-binding ability of STAT3 and STAT5 proteins in RCC cells. EMSA analysis of nuclear extracts prepared from Caki-1 and 786-O cells showed that RES reduced STAT3 and STAT5-DNA binding activities in a dose-dependent manner (Fig. 1e and f). These results show that RES abrogates the DNA binding ability of STAT3 and STAT5.

We next analyzed whether RES can suppress nuclear translocation of STAT3 and STAT5 in RCC cells. Figure 1g and h clearly demonstrates that RES reduced the translocation of STAT3 and STAT5 to the nucleus in both Caki-1 and 786-O cells.

In order to determine which upstream signaling molecules are involved in RES-mediated STAT3/5 inactivation, we examined the effects of RES on the phosphorylation of JAK1, JAK2, and Src in Caki-1 and 786-O cells. Cells were treated with the indicated concentrations of RES for 6 h. As shown in Fig. 2a, JAK1 and JAK2 were constitutively active in Caki-1 and 786-O cells and the treatment with RES clearly suppressed its phosphorylation in a concentration-dependent manner. The levels of total JAK1/2 remained unchanged under the same conditions (Fig. 2a, second and fourth panels).

We also determined the effect of RES on constitutive activation of Src kinase in both Caki-1 and 786-O cells. Interestingly, RES suppressed the constitutive phosphorylation of Src kinase in a concentration-dependent manner in these cells (Fig. 2b).

To analyze whether RES-induced inhibition of STAT3 phosphorylation could be due to activation of a protein tyrosine phosphatase (PTPase), Caki-1 and 786-O cells were treated with the broad-acting tyrosine phosphatase inhibitor sodium pervanadate. We noted that prevented the RES-induced inhibition of STAT3 activation (Fig. 2c). This data suggests that tyrosine phosphatases may be involved in RES-induced inhibition of STAT3 activation in RCC cells.

Protein tyrosine phosphatases have been also implicated in the STATs signaling pathways [45]. Thus, we examined whether RES regulates the expression of PTPε, and SHP-2, which are non-transmembrane PTPs expressed most abundantly in hematopoietic cells [36, 37, 46, 47]. As shown in Fig. 2d, RES led to increased expression of PTPε in Caki-1 cells, and SHP-2 in 786-O at the protein level. RES also enhanced mRNA level of PTPε C and SHP-2 in a dose dependent manner in Caki-1 and 786-O cells (Fig. 4e).

We determined whether the suppression of PTPε expression by siRNA would abrogate the inhibitory effect of RES on STAT3 activation in Caki-1 cells. Western blotting showed that RES-induced PTPε expression was effectively abolished in the cells treated with PTPε siRNA; whereas treatment with scrambled siRNA had no effect (Fig.2f, left, first panel). We also found that RES did not suppress STAT3 activation in cells treated with PTPε siRNA (Fig. 2f, left, third panel).

We also determined whether the deletion of SHP-2 expression by siRNA would abolish the suppressive effect of RES on STAT3 activation in 786-O cells. We found that RES-induced SHP-2 expression was effectively prevented in the cells treated with SHP-2 siRNA; treatment with scrambled siRNA had no effect (Fig. 2f, right, first panel). RES did not suppress STAT3 activation in cells treated with SHP-2 siRNA (Fig. 2f, right, third panel).

We also examined the effects of RES on cell cycle progression in Caki-1 and 786-O cells. As shown in Fig. 3a, RES-induced an increased accumulation of the cell population in S phase and a corresponding decrease of cells in the G1/G0 phases on Caki-1 and 786-O cells.

To evaluate the anti-cancer effects of RES, we also examined the apoptosis-inducing effects of RES by using the Annexin V assay. The Annexin V positive cells were increased as compared with the non-treated cells and Annexin V-FITC and PI positive cells were also increased in Caki-1 and 786-O cells (Fig. 3b).

We also examined whether RES induces changes in the mitochondrial membrane potential in Caki-1 and 786-Ocells. TMRE is a cell permeable, positively-charged, red-orange dye that readily accumulates in active mitochondria due to their relative negative charge. Depolarized or inactive mitochondria have decreased membrane potential and fail to sequester TMRE. As shown in Fig. 3c, the accumulation of TMRE labeled mitochondria was notably reduced on RES treatment for 24 h as compared with the non-treated cells. These results suggest that RES induces both early and late apoptosis accompanied with mitochondrial dysfunction in these RCC cells.

A clonogenic assay evaluates the potential of a single cell to resist treatments and grow into a colony [48]. The influence of RES on the clonogenic capacity of Caki-1 and 786-O cells was also evaluated. RES significantly inhibited colony formation and resulted in a remarkable decrease at colony formation ratio (Fig. 3d).

Because bcl-2, bcl-xL, survivin, IAP-1, and IAP-2 have been implicated in apoptosis and mitochondrial dysfunction, we next examined the effects of RES on the constitutive expression of these proteins by Western blot analysis. Cells were treated with 10, 30, or 50 μM of RES for 24 h. We observed that the constitutive expression of bcl-xl, bcl-2, survivin, IAP-1, and IAP-2 was dose-dependently reduced in RES-treated cells. Also, RES suppressed proteins linked with cell proliferation (COX-2), and angiogenesis (VEGF, and MMP-9) (Fig. 4a).

Whether suppression of constitutively active STAT3 in Caki-1 and 786-O cells by RES leads to apoptosis was also investigated. In these RCC cells treated with RES there was a dose-dependent activation of pro-caspase-3 and increased expression of cleaved caspase-3. Caki-1 and 786-O cells were treated with indicated concentration of RES for the 24 h, and then examined for caspase activation by Western blot using specific antibodies. We found a dose-dependent activation of caspase-3 by RES in these RCC cell lines (Fig. 4b, first panel). Activation of downstream caspases led to the cleavage of a 116 kDa PARP protein into 87 kDa fragments (Fig. 4b, Second panel). Taken together, these results suggest that RES induces caspase-3-dependent apoptosis in Caki-1 and 786-Ocells.

We also investigated whether RES blocks expression of various cell cycle associated proteins. As shown in Fig. 4c, RES induced a dose-dependent decrease in cyclin D1 and cyclin E protein levels, while an increased expression of p21 was observed after RES treatment (Fig. 4c, third panel). Taken together, these results indicate that RES decreased growth by blocking cell cycle progression via the upregulation of p21 and/or the downregulation of cyclin D1 and cyclin E expression.

To determine whether RES induces expression of Bax and p53, we examined the expression of these proteins in Caki-1 and 786-O cells by Western blot analysis. Figure 4d shows that RES induced the expression of Bax and p53 gene products in a dose-dependent manner.

To specifically examine the anti-tumor activity of RES on Caki-1 and 786-O cells, the cells were treated with 30 or 50 μM concentrations of RES, and then cell viability was analyzed every 15 min time intervals using the xCELLigence RTCA DP Instrument (Roche Diagnostics GmbH, Germany). As shown in Fig. 4e, RES significantly suppressed cell proliferation in RCC cells in a time-dependent manner.

Cyclin D1 and VEGF, all regulated by STAT3, are major players in tumor metastasis and has been shown to mediate tumor invasion [49‐51]. Whether RES can modulate tumor cell invasion activity was investigated in vitro. To determine this, Caki-1 and 786-O cells were analyzed by Boyden chamber assay. We found that these cells exhibited a very high invasive potential through a thick layer of Matrigel, and this invasive ability was significantly attenuated by RES (Fig. 4f).

Sorafenib (a kinase inhibitor) has been used for treating renal cell carcinoma patients. To determine whether RES potentiates the apoptotic effect of sorafenib, we treated 786-O cells with RES combined with sorafenib, and then examined the cell viability with a MTT assay. We found that RES indeed enhanced the cytotoxic effects of sorafenib using CalcuSynsoftware (BIOSOFT, Ferguson, MO) (Fig. 5a). Also, we further examined whether RES can potentiate the inhibitory effect of sorafenib on constitutive STAT3 and STAT5 phosphorylation in 786-O cells by Western blot analysis. As shown in Fig. 5b, RES substantially enhanced the inhibitory effects of sorafenib on phosphorylated STAT3 and STAT5 in 786-O cells. Interestingly, RES or sorafenib alone at sub-optimal concentrations had little effect on levels of bcl-2, bcl-xl, and surviving proteins in 786-O cells. However, treatment of cells with the combination of RES and sorafenib resulted in a marked attenuation in the expression levels of all of these proteins (Fig. 4c). Caspase-3 and PARP cleavage were further increased by the co-treatment of RES along with sorafenib rather than individual agents alone in 786-O cells (Fig. 5d). Overall, these results indicate that RES can significantly enhance the apoptotic effect of sorafenib in 786-O cells through the downregulation of various cell survival proteins.