By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Please be advised that we experienced an unexpected issue that occurred on Saturday and Sunday January 20th and 21st that caused the site to be down for an extended period of time and affected the ability of users to access content on Wiley Online Library. This issue has now been fully resolved. We apologize for any inconvenience this may have caused and are working to ensure that we can alert you immediately of any unplanned periods of downtime or disruption in the future.
Previously, we have shown that sorafenib sensitizes hepatocellular carcinoma (HCC) to apoptosis induced by TNF-related apoptosis-inducing ligand (TNFSF10; TRAIL). Here, we report that sorafenib and SC-49 sensitize HCC cells to CS-1008, a novel anti-human death receptor 5 (TNFRSF10B) antibody.
HCC cell lines (PLC5, Huh-7, and Hep3B) were treated with CS-1008 and/or sorafenib and analysed in terms of apoptosis and signal transductions.
SC-49 is a sorafenib derivative, which is devoid of kinase inhibitory activity. Both sorafenib and SC-49 down-regulated the phosphorylation of STAT3 at Tyr705 and subsequently reduced the levels of STAT3-regulated proteins, Mcl-1, survivin and cylcin D1, in CS-1008-treated HCC cells. Knockdown of STAT3 by RNA interference overcame apoptotic resistance to CS-1008 in HCC cells, and ectopic expression of STAT3 in HCC cells abolished the sensitizing effects of sorafenib and SC-49 on CS-1008-induced apoptosis, indicating that inhibition of STAT3 mediates the enhancing effects of these compounds when combined with CS-1008. Importantly, inhibition of SHP-1 by adding a specific SHP-1 inhibitor reduced the effects of SC-49 and CS-1008 on p-STAT3 and apoptosis, whereas co-treatment of CS-1008 with SC-49 increased the activity of SHP-1. These data indicate that the combined effects of CS-1008 and SC-49 on HCC are mediated by SHP-1. Moreover, the combination of CS-1008 and SC-49 inhibited HCC xenograft tumour growth in vivo.
Conclusions and Implications
Sorafenib and its derivative SC-49 sensitize HCC cells to the antitumour effects of CS-1008 through SHP-1-dependent inactivation of STAT3.
Hepatocellular carcinoma (HCC) is the fifth most common cancer in the world. It is a highly lethal malignancy that has a high recurrence rate despite surgical resection (Tanaka and Arii, 2009; Verslype et al., 2009). Advanced or recurrent HCC is frequently resistant to current chemotherapeutic agents and radiation; therefore, the development of targeted agents with tolerable toxicity is needed to advance anti-HCC therapy (Tanaka and Arii, 2009;). Sorafenib, a multi-kinase inhibitor, has been approved for clinical use in advance HCC after it was found to improve overall survival significantly in two prospective randomized phase III trials for patients with advanced HCC (Llovet et al., 2008; Cheng et al., 2009). Sorafenib inhibits multiple kinases, including the Ras/Raf/MAPK/ERK signalling pathway, the angiogenic pathways VEGFR2, VEGFR3, as well as PDGFRβ and other kinases such as FLT3 and FGFR1 (Adnane et al., 2006; Liu et al., 2006; Wilhelm et al., 2006; Auclair et al., 2007).
Among various targeted strategies for HCC treatment, TNF-related apoptosis inducing ligand (TNFSF10; TRAIL), which targets receptor-mediated apoptosis, represents an attractive option (Johnstone et al., 2008; Wang, 2008; Falschlehner et al., 2009). As a member of the TNF superfamily, TNFSF10 initiates apoptosis by binding to two important death domain-containing death receptors (DRs), TNFRSF10A (DR4) and TNFRSF10B (DR5) (Wang and El-Deiry, 2003; Rowinsky, 2005; Wiezorek et al., 2010). TNFSF10 or TNFSF10 agonists bind to TNFRSF10A or TNFRSF10B and form death-inducing signalling complex (DISC), which is a multi-protein complex consisting of an adaptor molecule, FADD and the initiator of extrinsic pathway caspase-8. (Wang and El-Deiry, 2003; Johnstone et al., 2008) Activated caspase-8 is capable of both initiating an extrinsic apoptotic pathway in type I cells (through activation of caspase-3, -6, and -7) and triggering the intrinsic pathway in type II cells (through activation of Bid) (Li et al., 1998; Johnstone et al., 2008).
CS-1008 is a novel TNFRSF10B agonist that exerts TNFSF10-like activity. It is a humanized anti-human TNFRSF10B antibody manufactured from a murine anti-human TNFRSF10B monoclonal antibody, TRA-8 (Yada et al., 2008). So far, CS-1008 has shown selective cytotoxicity towards tumour cells expressing TNFRSF10B (Yada et al., 2008) and an excellent safety profile in humans (Saleh et al., 2008). CS-1008 monotherapy induces apoptosis in various cancer cells and CS-1008 in combination with some chemotherapeutic agents (such as gemcitabine or docetaxel) has also been found to have enhanced antitumour activity (Yada et al., 2008). In a phase I trial, no dose-limiting toxicity was reported for CS-1008 at doses up to 8 mg·kg−1 weekly (Saleh et al., 2008).
Although TNFSF10 may be applied as anti-HCC strategy, more and more studies have reported that the efficacy of TNFSF10-induced apoptosis in HCC cells is insufficient, often due to resistance to TNFSF10 or its agonists (Shin et al., 2002; Pathil et al., 2006; Chen et al., 2009). Resistance to TNFSF10 may be induced at any step in the apoptosis signalling cascade, from the receptor level (mutations or overexpression of TNFRSF10A or TNFRSF10B) (Zhang and Fang, 2005), or defects in DISC assembly, (Eggert et al., 2001; Okano et al., 2003) through to dysfunctions of the anti-apoptotic Bcl-2 family proteins (Bcl-2, Bcl-xl, Mcl-1, etc) (Fulda et al., 2002; Kim et al., 2008) and pro-apoptotic proteins (Bax or Bak) or defects in mitochondria-derived activator of caspases (Smac/Diablo) (Zhang et al., 2001; Zhang and Fang, 2005). Of particular note, Mcl-1, an anti-apoptotic Bcl-2 family protein, plays a critical role in conferring TNFSF10 resistance (Kim et al., 2008). Data have shown that overexpression of Mcl-1 can neutralize TNFSF10-induced signalling (Meng et al., 2007; Ricci et al., 2007). Moreover, directly or indirectly destabilizing or disabling Mcl-1 can restore TNFSF10 sensitivity (Taniai et al., 2004; Hall and Cleveland, 2007). Interestingly, Mcl-1 is a highly regulated cell death and survival controller that responds to various cytokines and growth factors (Yang-Yen, 2006), and can be regulated by a number of transcription factors, including NF-κB targeting the cAMP response element (CRE-2) motif, and STAT3 targeting the sis-inducible element (SIE) motif of mcl-1 promoter region (Wang et al., 2003; Yang-Yen, 2006; Kim et al., 2008).
STAT3 is considered a potential anti-cancer therapeutic target because of its crucial role in transcriptional regulation of genes involved in cell proliferation and survival and it is constitutively activated in common human cancers, including HCC (Li et al., 2006; Germain and Frank, 2007; Kusaba et al., 2007). In response to the stimulation of cytokines, growth factors and hormones, STAT3 is phosphorylated (activated) and homodimerizes or heterodimerizes with STAT1 in the cytoplasm; it then translocates to the nucleus to regulate a number of genes, including genes that encode apoptosis-related proteins and cell cycle regulators (i.e. Bcl-2, Bcl-xl, Mcl-1, survivin and cyclin D1). In cancer cells, constitutively activated STAT3 directly contributes to tumourigenesis, invasion and metastasis (Germain and Frank, 2007). Targeting STAT3 using antisense oligonucleotide reduces the growth and metastasis of HCC cells in vitro and in vivo (Li et al., 2006). Importantly, reducing constitutive STAT3 activity has been shown to sensitize human hepatoma cells to TNFSF10-mediated apoptosis (Kusaba et al., 2007). Moreover, a number of protein tyrosine phosphatases have been shown to negatively regulate STAT3 signalling through direct dephosphorylation of p-STAT3 (Tyr705); these include members of the SH2-domain containing tyrosine phosphatase family (SHP-1 and SHP-2) and protein tyrosine phosphatase 1B (PTP-1B) (Ke et al., 2007; Chen et al., 2008; Kunnumakkara et al., 2009; Pandey et al., 2009). Therefore, activity of protein tyrosine phosphatases may be critical for the regulation of STAT3 phosphorylation in cancer cells.
Recently, we have reported that sorafenib sensitizes HCC cells to TNFSF10 (Chen et al., 2010), and STAT3 is a major kinase-independent target of sorafenib in HCC (Tai et al., 2011). We have discovered that several sorafenib derivatives are novel STAT3 inhibitors (Chen et al., 2011). In this study, we demonstrated that SC-49, a novel sorafenib analogue, is able to sensitize HCC cells to the antitumour effects of CS-1008.
The receptor nomenclature used in this paper conforms to Br J Pharmacol's Guide to Receptors and Channels (Alexander et al., 2011).
Reagents and antibodies
CS-1008 and sorafenib (Nexavar®) were kindly provided by Daiichi Sankyo Co., Ltd. (Tokyo, Japan) and Bayer Pharmaceuticals (West Haven, CT, USA) respectively. For in vitro studies, sorafenib at various concentrations was dissolved in DMSO and then added to the cells in 5% FBS-containing DMEM. Antibodies for immunoblotting such as Akt1, Mcl-1 and PARP were purchased from Santa Cruz Biotechnology (San Diego, CA, USA). Other antibodies such as anti-pERK (1/2), ERK2, survivin, cylcin D1, Bcl-xL, Bid, caspase-3, caspase-8, caspase-9, phospho-STAT3 (Tyr705), STAT3 and phosphor-Akt (Ser473) were from Cell Signaling (Danvers, MA, USA).
Cell culture and Western blot analysis
The Huh-7 HCC cell line was obtained from the Health Science Research Resources Bank (HSRRB; Osaka, Japan; JCRB0403). The PLC/PRF/5 (PLC5), Sk-Hep-1, Hep3B and U937 cell lines were obtained from American Type Culture Collection (ATCC; Manassas, VA, USA). All cells obtained from HSRRB or ATCC were immediately expanded and frozen such that all cell lines could be restarted every 3 months from a frozen vial of the same batch of cells. No further authentication was done in our lab. Cells were maintained in DMEM supplemented with 10% FBS, 100 U·mL−1 penicillin G, 100 μg·mL−1 streptomycin sulfate and 25 μg·mL−1 amphotericin B in a humidified incubator at 37°C in an atmosphere of 5% CO2 in air. Lysates of HCC cells treated with drugs at the indicated concentrations for various periods of time were prepared for immunoblotting of caspase-3, PARP, p-STAT3, STAT3, etc. Western blot analysis was performed as previously reported (Chen et al., 2008).
The following three methods were used to assess drug-induced apoptotic cell death: detection of DNA fragmentation with the Cell Death Detection ELISA kit (Roche Diagnostics, Indianapolis, IN, USA), Western blot analysis of caspase activation and PARP cleavage, and measurement of apoptotic cells by flow cytometry (sub-G1). The elisa was conducted according to the manufacturer's instructions.
Gene knockdown using siRNA
Smart pool siRNA reagents, including a control (D-001810-10), and mcl-1, STAT3, SHP-1, SHP-2, and PTP-1B were all purchased from Dharmacon Inc. (Chicago, IL, USA). The procedure has been described previously (Chen et al., 2008).
PLC5 with ectopic expression of STAT3
STAT3 cDNA (KIAA1524) was purchased from Addgene plasmid repository (http://www.addgene.org/). PLC5 cells with stable expression of STAT3 were then treated with drugs, harvested, and processed for Western blot analysis as described previously (Chen et al., 2008).
Activity of Raf-1 and SHP-1
A tyrosine phosphatase assay kit (R-22067) was used for assessing SHP-1 activity (Molecular Probes, Invitrogen, Grand Island, NY, USA). The Raf-1 kinase cascade assay kit (Upstate-Millipore, Billerica, MA) was used to examine Raf-1 kinase activity. The VEGFR1 kinase activity kit was purchased from Reaction Biology Corp. (Malvern, PA, USA).
Xenograft tumour growth
Male NCr athymic nude mice (5–7 weeks of age) were obtained from the National Laboratory Animal Center (Taipei, Taiwan). All studies involving animals are reported in accordance with the ARRIVE guidelines for reporting experiments involving animals (Kilkenny et al., 2010; McGrath et al., 2010). We used a total of 30 mice for the therapeutic evaluations and each separate experimental group consisted of 6-8 mice. In addition, all nude mice were housed in a light-controlled room with a 12 h day/night cycle and were given free access to water and food. The temperature of the animal room was kept at 25 °C.
All experimental procedures using these mice were performed in accordance with protocols approved by the Institutional Laboratory Animal Care and Use Committee of National Taiwan University. Each mouse was inoculated s.c. in the dorsal flank with 1 × 106 PLC5 cells suspended in 0.1 mL of serum-free medium containing 50% Matrigel (BD Biosciences, Bedford, MA, USA). When tumours reached 200–300 mm3, mice received an i.v. injection of CS-1008 (200 μg) three times a week, SC-49 (5 mg·kg−1) p.o. once daily or in combination. Controls received vehicle. Tumours were measured weekly using calipers, and their volumes calculated using the following standard formula: width2 × length × 0.52.
Immunohistochemical studies were performed on formalin-fixed, paraffin-embedded tissue. The primary monoclonal antibodies used were p-STAT3 (1:50 dilution; Cell Signaling), STAT3 (1:200; Cell Signaling), CD31 (1:40; BD Pharmingen, San Jose CA, USA). Antigen retrieval was performed at pH 9.0 using Epitope Retrieval 2 solution (Leica Microsystems, Wetzlar, Germany) for 30 min at 100°C. Slides were then stained using the Leica Microsystems BONDMAX autostainer according to the manufacturer's protocol.
Comparisons of mean values were performed using the independent samples t-test in SPSS for Windows 11.5 software (SPSS, Inc., Chicago, IL, USA).
Sorafenib sensitizes resistant HCC cells to CS-1008-induced apoptosis
To investigate the effects of sorafenib and CS-1008 on HCC cells, we first examined the apoptotic effects of both drugs on a panel of three human HCC cell lines Hep3B, PLC5 and Huh-7 at clinically relevant concentrations (≤1000 ng·mL−1); however, the combination of sorafenib and CS-1008 overcame the resistance and induced apoptosis in all cell lines tested, in a dose-dependent manner (Figure 1A). Next, we examined the effect of sorafenib on CS-1008-induced apoptosis, as assessed by DNA fragmentation in all HCC cell lines. DNA fragmentation was determined by cell death elisa after 48 h of treatment. As shown in Figure 1B, combining sorafenib at 7.5 μM with CS-1008 reversed the resistance in all three cell lines and induced significant apoptosis. Moreover, we further examined the apoptotic pathway by Western blot analysis. Our data indicated that co-treatment with CS-1008 and sorafenib activated caspase-8 then induced cleavage of Bid and subsequently activated caspase-9 and caspase-3 and PARP cleavage (Figure 1C). These data suggest that this intrinsic pathway played a role in mediating the combined effect of sorafenib and CS-1008 on apoptosis in HCC cells.
Sorafenib and CS-1008 co-treatment down-regulates p-STAT3 in HCC cells
Previous studies have suggested that Mcl-1, an anti-apoptotic Bcl-2 family protein, may play a role in mediating the sensitizing effect of sorafenib to TNFSF10 in cancer cells (Meng et al., 2007; Ricci et al., 2007; Kim et al., 2008). As STAT3 regulates the expression of Mcl-1, we next examined its related proteins including phospho-STAT3 (p-STAT3), STAT3 and STAT3-regulated proteins, which include Mcl-1, survivin and cyclin D1. As shown in Figure 2A, co-treatment with sorafenib and CS-1008 down-regulated p-STAT3 (Tyr705) and related proteins, Mcl-1, survivin and cyclin D1, in all the cells tested without altering of total protein levels of STAT3. In addition, down-regulation of p-STAT3 was associated with the cleavage of PARP as shown by the analysis of apoptosis induced in cells exposed to sorafenib and CS-1008 for 48 h (Figure 2A). Furthermore, we found that the combination of sorafenib and CS-1008 down-regulated p-STAT3 in PLC5 cells in both a dose- and time-dependent manner (Figure 2B). Interestingly, co-treatment with sorafenib and CS-1008 did not affect phospho-Erk, suggesting that the sensitizing effect of sorafenib on CS-1008 is not associated with its Raf-1 activity (Figure 2C). Notably, the combination of sorafenib and CS-1008 did not alter the expression of Bax and Bcl-xl (Figure 2C).
Validation of STAT3
Several approaches were used to validate the finding that inhibition of STAT3 signals is responsible for the sensitizing effect of sorafenib on CS-1008-induced apoptosis in HCC cells. Firstly, we knocked down the protein expression of Mcl-1 and STAT3 by use of small interference RNA (siRNA). PLC5 cells were transfected with either control, survivin siRNA or STAT3 siRNA for 48 h then exposed to DMSO or CS-1008 at the indicated doses for another 48 h. Silencing Mcl-1 and STAT3 significantly sensitized PLC5 cells to CS-1008-induced apoptosis (P < 0.05) (Figure 2D, left and middle), suggesting that inhibition of the STAT3 signalling pathway is important for the sensitivity of HCC cells towards CS-1008. Next, we examined the effects of sorafenib in combination with CS-1008 in both wild-type PLC5 cells and PLC5 cells with ectopic expression (overexpression) of STAT3. Over-expression of STAT3 significantly reduced the combined effects of sorafenib plus CS-1008 on p-STAT3 and apoptosis (P < 0.05) (Figure 2D, right). Together, these results confirm the importance of STAT3 inhibition in mediating the combined effect of CS-1008 and sorafenib.
SHP-1 plays a role in mediating the effects of apoptosis induced by sorafenib and CS-1008
To elucidate the mechanism by which sorafenib plus CS-1008 down-regulated p-STAT3 in HCC cells, we investigated the roles of several protein phosphatases on the effect of sorafenib plus CS-1008 on p-STAT3 and apoptosis. Firstly, we altered the expression of SHP-1, by using siRNA, in PLC5 cells and showed that silencing SHP-1 significantly reduced the effects of sorafenib plus CS-1008 on p-STAT3 and apoptosis (Figure 3A, left). This suggests that SHP-1 mediates the effects of these drugs on p-STAT3 and apoptosis. Notably, co-treatment with sorafenib and CS-1008 did not affect the expression level of SHP-1 in HCC cells. Therefore, we measured SHP-1 phosphatase activity in PLC5 cells that were treated with sorafenib plus CS-1008. As shown in Figure 3A (right), sorafenib plus CS-1008 significantly increased the activity of SHP-1 (P < 0.05). Moreover, as sorafenib is a kinase inhibitor, we examined whether sorafenib plus CS-1008 enhanced SHP-1 activity by affecting the phosphorylation of SHP-1. According to previous reports, phosphorylation of SHP-1 at Tyr536 may enhance its activity and phosphorylation at Ser591 may down-regulate its activity. However, our data showed that neither sorafenib alone nor co-treatment with CS-1008 altered phospho-SHP-1 at either site (Figure 3B, left). In addition, we examined whether the combination of drugs affected the protein-protein interactions between SHP-1 and STAT3. Our data show that the amount of STAT-SHP1 complex did not alter significantly after co-treatment with the two drugs, suggesting that this combination treatment did not affect the interaction between SHP-1 and STAT3 protein (Figure 3B right). Finally we also examined other protein tyrosine phosphatases such as SHP-2 and PTP-1B that could also regulate the STAT3 signalling pathway. However, neither knockdown of SHP-2 nor silencing of PTP-1B affected the effect of sorafenib plus CS-1008 on p-STAT3 signalling and apoptosis (Figure 3C and D). These data indicate that SHP-2 and PTP-1B are not involved in mediating the effects of sorafenib on p-STAT3 signalling and apoptosis induced by CS-1008.
SC-49, a sorafenib derivative, sensitizes HCC cells to CS-1008
Previously, we showed that STAT3 mediates the anti-tumour effect of sorafenib on HCC, and this effect was not related to inhibition of kinase activity (Tai et al., 2011). We then modified sorafenib and synthesized several new sorafenib analogues, which, by enhancing the activity of SHP-1, are potent inhibitors of STAT3. Here, we investigated the ability of SC-49 to sensitize HCC cells to CS-1008. As shown in Figure 4A, SC-49 is similar to sorafenib structurally but without Raf-1 inhibitory activity. Like sorafenib, the combination of SC-49 and CS-1008 also induced significant apoptosis in TNFSF10-resistant HCC cells. Combining sorafenib at 5 μM with CS-1008 reversed the resistance in all three cell lines and induced significant apoptosis within 24 h (Figure 4B and C). As shown in Figure 4D, co-treatment with SC-49 and CS-1008 activated caspase-8 then induced cleavage of Bid and subsequently activated caspase-9 and caspase-3 and PARP cleavage. It is noteworthy that in comparison with sorafenib, SC-49 sensitized HCC cells to CS-1008 at lower concentrations (5 μM vs. 7.5 μM) and in a shorter period of time (24 h vs. 48 h). These data suggest that SC-49 is more potent than sorafenib in sensitizing HCC cells to CS-1008.
SC-49 showed better apoptotic effects than sorafenib in HCC
To further investigate the effect of SC-49 on angiogenesis, we tested the effect of SC-49 on the activity of VEGFR1 in HUVEC cells. As shown in Figure 5A (left), sorafenib as a kinase inhibitor significantly inhibited the activity of VEGFR1 in HUVEC cells. However, unlike sorafenib, SC-49 did not affect the activity of VEGFR1 in HUVEC cells. Next, we examined the effect of SC-49 on p-STAT1 and p-STAT5; SC-49 down-regulated both p-STAT1 and p-STAT5 in a dose-dependent manner (Figure 5A, right). Furthermore, in HCC cell lines, Huh-7 and Hep3B, we found that SC-49 induced more apoptotic cell death than sorafenib (Figure 5B). In addition, SC-49 was more effective at down-regulating p-STAT3 than sorafenib in HCC cells (Figure 5C). These data suggest that SC-49, a sorafenib derivative without kinase inhibitory activity, is a more potent anti-tumour agent than sorafenib and that its effect is induced by targeting the STAT3 signalling pathway. Notably, a recent study has shown that depletion of tumour-associated macrophages may be associated with the effect of sorafenib on a liver metastasis model (Zhang et al., 2010). In this regard, we tested the effect of SC-49 on U937, a human macrophage cell line (Passmore et al., 2001) and again found that SC-49 down-regulated p-STAT3 in a dose-dependent manner in these cells (Figure 5D).
SHP-1-dependent inhibition of STAT3 mediates the effect of SC-49
We next examined the mechanism by which SC-49 sensitized HCC cells to CS-1008. Our data showed that the combination of SC-49 and CS-1008 decreased p-STAT3 in three TNFSF10-resistant cell lines. Co-treatment with SC-49 plus CS-1008 also down-regulated STAT3-driven proteins, Mcl-1, survivin and cylcin D1, in all three HCC cell lines (PLC5, Huh-7 and Hep3B) (Figure 6A). As shown in Figure 6B, overexpression of STAT3 reversed the sensitizing effects of SC-49 on CS-1008, suggesting that STAT3 plays a role in mediating the effect of SC-49. We next employed a specific SHP-1 inhibitor to test whether SHP-1 mediates the effect of SC-49 on p-STAT3, and found that the SHP-1 inhibitor abolished the sensitizing effect of SC-49 on the action of CS-1008 significantly, indicating that SHP-1 is a mediator of this effect of SC-49 (Figure 6C). As shown in Figure 6D (left), SC-49 at 5 μM enhanced the activity of SHP-1. Notably, SC-49 did not affect the phosphorylation of SHP-1 (Figure 6D, right). To further explore the mechanism by which SC-49 affected the activity of SHP-1, we tested the effect of SC-49 on SHP-1-containing cell lysates to determine whether SC-49 enhances the activity of SHP-1 by affecting its interactions with other proteins. Briefly, PLC5 cells were immunoprecipitated with anti-SHP-1 antibody. Protein extract which included SHP-1 complex was further incubated with SC-49 at 7.5 μM and/or CS-1008 at 1000 ng·mL−1 for 30 min and then SHP-1 phosphatase activity assay was performed. Our data showed that SC-49 alone or in combination with CS-1008 increased the phosphatase activity of SHP-1-containg lysates (Figure 6E). These data suggest that SC-49 enhances the activity of SHP-1 by directly interacting with it. However, further work is needed to elucidate details of the interactions between SC-49 and SHP-1.
In vivo effects of SC-49 in Huh-7 xenograft tumours
To further examine the effect of SC-49, we next tested the effect of SC-49 on Huh-7 xenograft tumours in vivo. As shown in Figure 7A (left), treatment of mice with SC-49 at a dose of 10 mg·kg−1·day−1 p.o. significantly reduced the growth of the Huh-7 tumour and this anti-tumour effect was better than that of sorafenib in vivo. As shown in Figure 7A (right), animals had stable body weights throughout the course of study. In addition, SC-49 down-regulated p-STAT3 in Huh-7 tumours (Figure 7B). SC-49 and sorafenib enhanced the activity of SHP-1 in Huh-7 tumours (Figure 7C). Immunohistochemical staining for STAT3 showed no obvious significantly different cytoplasmic expression in all groups (Figure 7D). The treatment of both sorafenib and SC-49 decreased the nuclear expression of P-STAT3 (Figure 7D). From the immunohistochemical stain for CD-31, all the groups showed a similar vascular density in the tumour areas (Figure 7D).
These data indicate that SC-49 exhibited better in vivo effects than sorafenib through an SHP-1-dependent inhibitory effect on STAT3.
The effect of the combination of SC-49 and CS-1008 in vivo
To confirm whether the sensitizing effect of SC-49 in resistant cell lines has potentially relevant clinical implications, we assessed the effect of the combination of CS-1008 plus SC-49 on the growth of PLC5 tumours in vivo. Tumour-bearing mice were treated with vehicle or CS-1008 i.v. at a dose of 200 μg three times a week or SC-49 p.o. at a dose of 5 mg·kg−1·day−1, or a combination of the two, for the duration of the study. All animals tolerated the treatments well without an observable signs of toxicity and had stable body weights throughout the course of study. No gross pathological abnormalities were noted at necropsy.
Tumour growth was significantly inhibited by co-treatment with CS-1008 and SC-49 for 2 weeks (vs. control, P < 0.05), and tumour size in the co-treatment group was only one third of that of the control group at the end of the study (Figure 8A). Treatment with CS-1008 had no significant effect on PLC5 tumour growth. SC-49 alone showed modest effects on tumour growth. In addition, co-treatment of SC-49 and CS1008 significantly down-regulated p-STAT3 in PLC5 tumours (Figure 8B). Moreover, as shown in Figure 8C, co-treatment with SC-49 and CS-1008 enhanced SHP-1 activity significantly, indicating that SHP-1 plays a role in mediating the effects of the combination of drugs on PLC5 tumours. Together, these data indicate that a combination of CS-1008 and SC-49 exhibits good anti-tumour activity in vivo. Further clinical investigations are warranted.
Previous literature has consistently shown that sorafenib is capable of sensitizing various cancer cells, including HCC, to TNFSF10-induced apoptosis (Hall and Cleveland, 2007; Ricci et al., 2007; Koehler et al., 2009; Huang and Sinicrope, 2010; Llobet et al., 2010). For that reason, CS-1008 is currently undergoing phase II trials for the treatment of advanced HCC (NCI clinical trial: NCT01033240). The results should provide informative in vivo evidence of the activity of this combination strategy.
Although the mechanisms employed by various tumours to evade TNFSF10-induced apoptosis are heterogeneous (Zhang and Fang, 2005), it has been suggested that Mcl-1 is the gateway to the sensitizing effect of sorafenib in cells (including HCC cells) that harbour defects in apoptosis mediated by the intrinsic pathway (Meng et al., 2007; Kim et al., 2008). In addition, aberrant activation of anti-apoptotic pathways, such as PI3K/Akt signalling, MAPK pathway and the NF-κB pathway, may also contribute to the development of TNFSF10 resistance in HCC cells (Bortul et al., 2003; Ehrhardt et al., 2003; Zhang and Fang, 2005). In particular, TNFSF10 treatment in resistant cells has been shown to induce Mcl-1 expression through the Raf and NF-κB-dependent pathway. Sorafenib, as a Raf kinase inhibitor, could, therefore, potentially block this TNFSF10-induced NF-κB–mediated transcriptional activation of Mcl-1, and NF-κB binding to the Mcl-1 promoter region (Kim et al., 2008). In the present study, we added to previous data by further showing that sorafenib suppresses other STAT3-regulated proteins (i.e. survivin and cyclin D1) in HCC. Our data also confirmed that the inhibition of STAT3 is the major mechanism by which sorafenib sensitizes TNFSF10 in HCC. It has previously been demonstrated that sorafenib inhibits STAT3 activity and enhances TNFSF10-mediated apoptosis in other cancer cells, including pancreatic cancer cells (Huang and Sinicrope, 2010), medulloblastomas cells (Yang et al., 2008) and cholangiocarcinoma cells (Blechacz et al., 2009). Taken together, these data indicate that STAT3 represents a novel anti-cancer target of sorafenib.
Another important finding in the current study is that sorafenib inhibits STAT3 by increasing SHP-1 activity (Figure 3). Our results showed that sorafenib increased SHP-1 activity but did not alter SHP-1 protein expression level and, despite being a kinase inhibitor, did not alter the phosphorylation of SHP-1 at either the Y-536 or S-591 sites, both known to change SHP-1 activity upon phosphorylation. Moreover, sorafenib did not influence the SHP-1 and STAT3 protein–protein interactions. In contrast, several chemical compounds such as acety-11-keto-β-boswellic acid and butein (3,4,2′,4′-tetrahydroxychalcone), are thought to inhibit STAT3 by the induction of SHP-1 expression (Kunnumakkara et al., 2009; Pandey et al., 2009). Nevertheless, the mechanism by which sorafenib influences SHP-1 activity remains to be elucidated and further studies are needed to address this issue. Interestingly, Blechacz et al. (2009) suggested that sorafenib inhibits STAT3 in cholangiocarcinoma cells by influencing SHP-2 activity through the down-regulation of phospho-SHP-2. However,Blechacz et al. (2009) did not show whether sorafenib also affects SHP-1 in cholangiocarcinoma cells. In contrast, in the present work we have shown that knockdown of SHP-2 did not alter the sensitizing effect of sorafenib on apoptosis and STAT3 phosphorylation in HCC cells (Figure 3C). However, it is possible that sorafenib inhibits STAT3 by affecting different protein tyrosine phosphatases in various cancer cells. More effort is needed to fully understand why sorafenib affects different protein tyrosine phosphatases in HCC and cholangiocarcinoma, and perhaps other cancer cells.
In conclusion, our results revealed that sorafenib as well as SC-49 have a synergistic effect with CS-1008 on HCC through SHP-1-dependent inhibition of STAT3 and indicate that the STAT3 signalling pathway may be a suitable target for the development of anti-HCC targeted agents. Sorafenib may serve as a lead compound for the development of more potent STAT3 inhibitors.
This study is supported by grants, NTUH 100P04 from National Taiwan University Hospital (K-F Chen), NSC99-2314-B-002-017-MY2 (K-F Chen), NSC 100-3112-B-002 -013 (A-L. Cheng), NSC 100-2325-B-002 -036 (K-F Chen), and NSC 100-2325-B-010 -007 (C-W Shiau) from National Science Council, Taiwan, Taiwan Clinical Oncology Research Foundation (C-Y Liu).
Disclosure of potential conflict of interest
A-LC is a consultant for and a member of the speaker's bureau of Bayer-Schering and a consultant of Daiichi Sankyo. K I is an employee of Daiichi Sankyo. Other authors have nothing relevant to this manuscript to disclose.