6-Bromoindirubin-3′-Oxime Inhibits JAK/STAT3 Signaling and Induces Apoptosis of Human Melanoma Cells

The Janus activated kinase (JAK) family includes JAK1, JAK2, JAK3, and TYK2. Whereas JAK1, JAK2 and TYK2 are ubiquitously expressed in mammalian cells, JAK3 expression is restricted to hematopoietic cells (1–2). The kinases of the JAK family activate STAT proteins in response to different cytokines and growth factors (2–5). In normal cells, JAK/STAT signaling is tightly regulated. However, in cancer cells, it is persistently activated due to the aberrant activation of JAK family kinases or other tyrosine kinases. As one of the more recently recognized oncogenic signaling pathways, JAK/STAT3 signaling has been shown to be important for tumorigenesis. Among the 7 STAT family members, STAT3 is most frequently aberrantly activated in human cancer cells. Persistent activation of JAK/STAT3 signaling contributes to the malignancy of tumors by promoting tumor cell proliferation and survival, angiogenesis, and immune evasion (6–10). Thus, JAK/STAT3 signaling is a promising molecular target in cancer therapy.

Recently, small-molecule inhibitors of JAK/STAT3 signaling have been identified for development of cancer therapeutics. Discovery of the JAK2 V617F mutation in myeloproliferative disease has prompted development of selective JAK2 inhibitors for treatment of hematologic disorders (11–13). Clinical studies of JAK2 inhibitors are currently ongoing (14–15). Furthermore, the role of constitutive JAK2 kinase activity in myeloproliferative neoplastic growth provides the rationale for investigating JAK inhibitors in solid tumors. Small-molecule inhibitors of JAK/STAT3 signaling have been reported to suppress cancer cell growth, both in vitro and in vivo (16–22).

Indirubin, a bis-indole alkaloid, is the active ingredient of Dang Gui Long Hui Wan, a traditional Chinese medicine for treatment of chronic myelocytic leukemia (CML) (23–24). Indirubins, namely indirubin and some indirubin derivatives (IRD), are known to be inhibitors of cyclin-dependent kinases (CDK), glycogen synthase kinase-3 (GSK-3), and glycogen phosphorylase b. In addition, some IRDs induce apoptosis in human cancer cells through inhibition of Src/STAT3 signaling (25–29), thus suggesting that STAT3 signaling might be a potential target for IRDs.

Bromo-IRDs are novel synthetic IRDs with improved potency (30–33). In this study, we have identified 6BIO as a pan-JAK inhibitor, selectively targeting JAK/STAT3 signaling in human melanoma cells. Among a series of synthetic IRDs, 6BIO shows the best anticancer activity, and induces apoptosis of melanoma cells. 6BIO suppressed tumor growth with low toxicity in an A2058 human melanoma xenograft mouse model. We investigated the effects of 6BIO particularly on human melanoma cells because of a need for more effective therapeutics for this tumor site. Dacarbazine, the only Food and Drug Administration (FDA)-approved drug for treatment of metastatic melanoma, has an approximately 15% response rate (34–35). Thus, 6BIO as a pan-JAK inhibitor represents a promising lead compound for development of new anticancer therapeutics, particularly for melanoma.

The preparation of IRDs has been described previously (30). Compounds were dissolved in dimethyl sulfoxide (DMSO) at a concentration of 20 mmol/L and stored at −20°C before use for in vitro experiments and treatments in cells. For in vivo experiments, 6BIO was freshly prepared in 30% Solutol (Basf) at a concentration of 10 mg/mL. Anti-survivin was obtained from Novus and Anti–β-actin was obtained from Sigma. Horseradish peroxidase (HRP)-labeled anti-mouse and anti-rabbit secondary antibodies were procured from GE Healthcare. All other antibodies were from purchased from Cell Signaling.

The A2058, G361, SK-MEL-5, and SK-MEL-28 human melanoma, DU145 and LNCaP prostate cancer, U266 myeloma, and MDA-MB-468 breast cancer cell lines were obtained from the American Type Culture Collection. MDA-MB-468 cells were maintained in Dulbecco's modified Eagle medium. Other types of cells were maintained in RPMI 1640 medium. The cell-culture medium was supplemented with 10% heat-inactivated FBS and 1% penicillin/streptomycin (P/S).

Human cancer cells were seeded onto 96-well plates at a density of 5,000 cells per well. After 24-hour incubation, cells were treated with IRDs or DMSO as the vehicle control for 48 hour. An MTS Reagent (CellTiter 96 AQueousOne Solution Cell proliferation Assay; Promega) was added to each well according to the manufacturer's instructions. Absorbance was measured at 490 nm within 4 hours of the reaction by a Microplate Reader (Bio-Rad). The values of cell viability were calculated as percentages of absorbance from treated samples to absorbance from the vehicle control.

Cells were seeded on 6-well plates with 50,000 cells per well in 3 mL of RPMI 1640 medium supplemented with 10% FBS and 1% P/S. After 24-hour incubation, cells were treated with 6BIO or DMSO for 24 or 48 hours. After treatment, all cells were collected, and apoptotic cells were detected by flow cytometry by the Annexin V–FITC Apoptosis Detection Kit (BD Biosciences) according to the manufacturer's instructions.

Cell lysates were prepared in radioimmunoprecipitation assay (RIPA) buffer supplemented with inhibitors of proteases (Roche Diagnostics GmbH) and sodium orthovanadate, an inhibitor of phosphotases (Aldrich). Protein concentrations were determined by BioMate Spectrometer (Thermo) and protein assay (Bio-Rad). A sample of 40 μg of each protein was resolved in 8% or between 8% and 16% gradient SDS-PAGE gels (Pierce). After gel electrophoresis, proteins were transferred to Hybond-C membranes (Amersham). The membranes were blocked in 5% nonfat milk in PBS containing 0.1% Tween 20 (Polysorbate 20; PBST) at room temperature for 1 to 3 hours followed by an overnight incubation at 4°C with primary antibodies in PBST containing 5% nonfat milk. The membranes were then washed with PBST and incubated with HRP-conjugated secondary antibody in 5% nonfat milk/PBST solution for 1 to 3 hours at room temperature, or overnight at 4°C. Specific proteins were detected by exposure to X-ray film by using SuperSignal West Pico Chemiluminescent Substrate or SuperSignal West Dura Extended Duration Substrate (Pierce).

Recombinant JAKs, substrates, and ³³P-labeled ATP were used for in vitro kinase assays. The recombinant JAK catalytic domains were tagged with glutathione S-transferase (GST) and purified from insect cells. The substrate was prepared in freshly constituted Base Reaction Buffer [comprising 20 mmol/L HEPES, pH 7.5; 10 mmol/L MgCl₂; 1 mmol/L EGTA; 0.02% BRIJ-35; 0.02 mg/mL BSA; 0.1 mmol/L Na₃VO₄; 2 mmol/L dithiothreitol (DTT); and 1% DMSO]. Then, required cofactors and kinase were added into the substrate solution. Compounds in DMSO were delivered into the kinase reaction mixture, and the ³³P-labeled ATP (specific activity: 0.01 μCi/μL final) was added into the reaction mixture to initiate the reaction. The kinase reaction mixture was incubated for 120 minutes at room temperature. Reactions were spotted onto P81 ion-exchange paper (Whatman) for measurement of radioactivity.

BALB/c mice (6–8 weeks old) were purchased from the National Cancer Institute for toxicity study. Immunodeficient NOD/SCID/IL2Rgamma null (NSG) mice (female; 6–8 weeks old) were purchased from The Jackson Laboratory for use as the xenograft model. The experimental protocol for animal experiments was approved by the Institutional Animal Care and Use Committee (IACUC) of the Beckman Research Institute at City of Hope Medical Center. A2058 human melanoma cells at a density of 5 × 10⁶ cells in serum-free medium were inoculated subcutaneously into the dorsal area of NSG mice to create the xenograft model. When tumors became palpable, 6BIO or vehicle control was administered via oral gavage once daily at a dose of 50 mg/kg body weight. Tumor growth was monitored every other day. Tumor volumes were measured every 3 to 4 days. Tumor volumes were calculated by the formula: 0.5 × (larger diameter) × (small diameter)².

A 2-sided t test was used to evaluate statistical significance of differences between treated and control groups. P < 0.05 was considered statistically significant.

We screened synthetic IRDs by using MTS cell-viability assays in A2058 cells, which are derived from metastatic lymph node melanoma. Two bromo-IRDs—MLS-2407 (6BIO) and MLS-2408—showed the most potent anticancer activity (Table 1). 6BIO is an IRD with double modifications of oxime- and bromo-groups, whereas MLS-2412 is an IRD with double modifications of oxime- and fluoro-groups at 3′- and 6-positions, respectively. MLS-2446 and -2406 are IRDs with single modification of oxime-group or bromo-group at 3′- or 6-position. Among 6BIO, MLS-2446, -2406, and -2412, 6BIO decreased the viability of melanoma cells to 31% at a concentration of 10 μmol/L, whereas MLS-2446, -2406, and -2412 showed marginal anticancer activities (Table 1), indicating that modifications with both 3′-oxime- and 6-bromo-groups are important for anticancer potency. Furthermore, 6BIO inhibits cell viability in other human cancer cells (Supplementary Fig. S1). Thus, 6BIO was selected for further study.

In primary screening assays, A2058 human melanoma cells were sensitive to 6BIO. We further compared the effects of 6BIO on A2058, G361, SK-MEL-5, and SK-MEL-28 human melanoma cell lines (Fig. 1). Cell viability was inhibited in all 4 melanoma cell lines. IC₅₀ values are approximately 5 μmol/L for A2058 and G361 cells, and 12 μmol/L for SK-MEL-5 and SK-MEL-28 cells (Fig. 1A). Previous studies showed that human melanoma cells harbor persistently activated STAT3, which contributes to tumorigenesis and metastasis (6). Phosphorylation of JAK2, Src, and STAT3 was inhibited at 10 μmol/L of 6BIO 4 hours after treatment in melanoma cells (Fig. 1B). Phosphorylation of Src was inhibited in G361, SK-MEL-5, and SK-MEL-28 melanoma cell lines, whereas phosphorylation of JAK2 was inhibited in A2058 and G361 cell lines and only partially inhibited in SK-MEL-5 and SK-MEL-28, both of which are more resistant cell lines. Phosphorylation of Src was not inhibited in A2058 human melanoma cells although phosphorylation of JAK2 was substantially inhibited (Fig. 1B). Consistent inhibition of phosphorylation of JAK2 in all melanoma cell lines suggests that JAK family kinases may be relevant targets in melanoma cells.

Human melanoma cells express JAK1, JAK2, and TYK2. We further investigated effects of 6BIO on JAK1, JAK2, and TYK2 in A2058, SK-MEL-5, SK-MEL-28, and G361 cell lines. In A2058 cells, 6BIO inhibited phosphorylation of JAK1, JAK2, and TYK2 in a dose-dependent manner at both 4 hours (Fig. 2A) and 24 hours (Fig. 2B) after treatment as well as in a time-dependent manner (Fig. 2C). Inhibition of phosphorylation of JAK1, JAK2, and TYK2 was observed in A2058 cells as early as within 5 minutes after treatment with 10 μmol/L 6BIO (Supplementary Fig. S2A). Furthermore, reduction of levels of phosphorylated JAK1, JAK2, and TYK2 was detected in SK-MEL-5, SK-MEL-28, and G361 cells at 4 hours after treatment with 10 μmol/L 6BIO (Supplementary Fig. S2B). To determine whether 6BIO is a pan-JAK inhibitor, we further investigated the effect of 6BIO on JAK enzymatic activity in an in vitro kinase assay. Activities of JAK family kinases were inhibited in a dose-dependent manner. The IC₅₀ values are 0.03 μmol/L for TYK2, 1.5 μmol/L for JAK1, 8.0 μmol/L for JAK2 (Fig. 2D), and 0.5 μmol/L for JAK3 (Supplementary Fig. S3). These findings suggest that 6BIO is a novel pan-JAK small-molecule inhibitor.

To investigate the selectivity of inhibition on JAK/STAT3 signaling, Src and other two important signaling proteins, Akt and mitogen-activated protein kinase (MAPK; Erk1/2) were examined. Western blots were done to detect the levels of total and phosphorylated Src, STAT3, Akt, and MAPK (Erk1/2; Fig. 3). In contrast to the inhibition of phosphorylation of JAK family kinases (Fig. 2A), phosphorylation of Src was not changed (Fig. 3A). Phosphorylation of STAT3 was decreased whereas phosphorylation of Akt and MAPK (Erk1/2) was not inhibited in all of the 4 human melanoma cell lines (Fig. 3). These observations suggest that, among signaling pathways of STAT3, Akt, and MAPK, 6BIO selectively inhibits JAK/STAT3 signaling in human melanoma cells.

STAT3 activation is involved in cell survival in human cancer cells (6). We examined STAT3 downstream antiapoptotic proteins including Mcl-1, Bcl-2, Bcl-xL, and survivin. Expression of Mcl-1 was decreased in a dose-dependent manner in A2058 cells (Fig. 4A) and other human melanoma cell lines (Fig. 4B). Cleaved forms of PARP and caspase-3 were detected at 5 μmol/L 24 hours after treatment. The levels of full-length PARP and caspase-3 were decreased corresponding to the increased levels of cleaved forms (Fig. 4C). 6BIO induced apoptosis of human melanoma cells in both dose- and time-dependent manners (Fig. 4D). These results show that 6BIO induces melanoma cell apoptosis at least in part by inhibition of Mcl-1.

The anticancer activity of 6BIO in vivo was studied with an A2058 human melanoma xenograft mouse model. We first tested the toxicity of 6BIO in BALB/c normal mice. 6BIO, at the dose of 100 mg/kg, was found to be safe for the mice (Fig. 5A). Then, we used a dose of 50 mg/kg for 6BIO therapy study in the NSG mouse xenograft model by oral administration once daily for 2 weeks. As shown in Figure 5B, the tumor growth was significantly suppressed. No side effects were observed in the 6BIO- treated mice. These findings show the antitumor activity of 6BIO in vivo against human melanoma cells.

Indirubin has been extensively derivatized by chemical modifications to improve its selectivity profile, kinase inhibitory effect, and solubility (26, 30). In particular, modifications have been carried out on positions 5 or 6 on one indole ring and 3′ of the other indole ring. The IRDs shown in Table 1 were synthesized with modifications at 3′-, 5-, and 6-positions. 6BIO and MLS-2408 showed the most potent anticancer activity. These active bromo-IRDs contain a bromo-group at the 6-position and a hydrophilic group at the 3′-position. Our data show that either 3′- or 6-substitution alone, as shown in MLS-2403 and -2446, or -2406, -2409, and -2411, is not sufficient to improve anticancer activity. Furthermore, 6BIO inhibited cell viability strongly whereas 6-fluoroindirubin-3′-oxime (6FIO; MLS-2412) is a very weak inhibitor of cell viability (Table 1), indicating bromo-substitution is more important than fluoro-substitution at the 6-position. Substitution at the 5-position with a nitro-group as in MLS-2418 and -2419 reduced anticancer activity, compared with 6BIO and MLS-2408. This finding suggests that modification at the 5-position with a nitro-group is unnecessary for anticancer activity of 6-bromoindirubins. Comparing the structures and activities of 6BIO, MLS-2446, and MLS-2406, we found that both substitution of an oxime-group at the 3′-position and substitution of a bromo-group at the 6-position are important to enhance anticancer activity.

It is interesting that 6BIO displays differential inhibition of JAK family kinase activity in an in vitro system by using recombinant JAKs (Fig. 2D). 6BIO inhibits TYK2 most potently (IC₅₀ value: 0.03 μmol/L), whereas the IC₅₀ value is 8 μmol/L for JAK2 kinase activity. The IC₅₀ value of JAK1 kinase activity is similar in vitro as it is in intact A2058 cells. The differential inhibition of phosphorylated JAKs in human melanoma cells is not as dramatic as the differential inhibition of JAK activity in an in vitro system (Fig. 2A and D). It is striking that 6BIO inhibits phosphorylation of JAK2 strongly in A2058 cells, but it is a weak inhibitor of JAK2 kinase activity in vitro. We have found that 6BIO inhibits phosphorylation of JAK1, JAK2, and TYK2 in cells very effectively although it is not a strong inhibitor for JAK2 in vitro. This finding is consistent with JAK transphosphorylation and reciprocal regulation of JAKs in cells. Previous studies showed that interaction between 2 members among JAK1, JAK2, and TYK2 regulates JAK activity. In particular, it has been found that TYK2 is required for kinase activity of JAK2 and/or JAK1 (4, 36–40). The long-term effect of 6BIO on A2058 human melanoma cells showed substantial differences between phosphorylated levels of JAK1, JAK2, and TYK2. Phosphorylation of JAK1 and TYK2 was most potently inhibited, whereas phosphorylation of JAK2 was least inhibited in a dose-dependent manner (Fig. 2B). This inhibition pattern is very similar to the inhibition of in vitro JAK activity (Fig. 2D). Our findings suggest that TYK2 may regulate kinase activity of JAK2 and/or JAK1 in human melanoma cells. In the in vitro system, 6BIO is a strong inhibitor of TYK2, a moderate inhibitor of JAK1, but a weak inhibitor of JAK2. However, it is a pan-JAK inhibitor in intact A2058, SK-MEL-5, SK-MEL-28, and G361 human melanoma cells (Fig. 2; Supplementary Fig. S2B), suggesting reciprocal regulation among JAK1, JAK2, and TYK2 in cells.

Furthermore, as a pan-JAK inhibitor, 6BIO inhibits phosphorylation of Src in human melanoma cells. As shown in Figure 1B, phosphorylation of both JAK2 and Src was inhibited in G361, SK-MEL-5, and SK-MEL-28 cells by 6BIO at 10 μmol/L at 4 hours after treatment, suggesting cooperation between JAK2 and SRC in G361, SK-MEL-5, and SK-MEL-28 cells to inhibit STAT3 signaling (41). In addition, in A2058 cells, phosphorylation of Src was inhibited in a dose-dependent manner 24 hours after treatment (Supplementary Fig. S2C), but it was not inhibited at 4 hours after treatment (Fig. 3A). This observation may indicate a secondary inhibition of Src by 6BIO in A2058 cells. As shown in Figure 3, comparing with phosphorylation of Src, Akt, and MAPK, the phosphorylation of STAT3 was inhibited in a dose-dependent manner by 6BIO 4 hours after treatment. This finding suggests the selective inhibition of JAK/STAT3 signaling mediated by 6BIO in human melanoma cells. However, we cannot conclude that 6BIO is an absolutely selective inhibitor of JAK/STAT3 signaling. Indirubin, an ATP-mimic, was originally identified as an inhibitor of CDKs. Later, indirubin and its derivatives have been found to inhibit other kinases, including GSK3-β and Src. Similarly, we cannot rule out other possible targets of 6BIO such as CDKs, GSK3-β, FLT3, platelet-derived growth factor receptor (PDGF-R), Src, and VEGF-R (31). In summary, we show that 6BIO is a pan-JAK inhibitor and inhibits the JAK/STAT3 signaling pathway, but we do not exclude inhibition of other pathways.

We have discovered that 6BIO induces apoptosis of human melanoma cells associated with inhibition of JAK/STAT3 signaling as a pan-JAK inhibitor. Two other important signaling proteins, Akt and MAPK, are not involved in the induction of apoptosis. Src may be an indirect target of 6BIO. Our findings suggest TYK2 may regulate the kinase activity of other JAK family members such as JAK2 and/or JAK1. In addition, 6BIO is a strong inhibitor of JAK3 (Supplementary Fig. S3). As JAK3 is exclusively expressed in hematopoietic cells, there are no further data available regarding JAK3 in melanoma cells, although 6BIO may be a promising drug candidate for treatment of blood malignancies. Importantly, 6BIO suppresses tumor growth in an A2058 human melanoma xenograft mouse model. In summary, 6BIO is a promising candidate for development as a novel therapeutic agent targeting JAK/STAT3 signaling in melanoma.

No potential conflicts of interest were disclosed.

This study was supported in part by Grant R01-CA115674 from NIH (R. Jove).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.