Potential conflict of interest: Nothing to report.
The Hedgehog signaling pathway plays a pivotal role during embryonic development, stem cell maintenance, and wound healing. Hedgehog signaling also is deregulated in many cancers. However, the role of this signaling pathway in the carcinogenesis of cholangiocarcinoma (CCC) is still unknown. In this study, we investigated the effects of Hedgehog inhibition by cyclopamine and 5E1 in cultured human CCC cell lines and in vivo using a xenograft mouse model. We also investigated the involvement of Hedgehog in epithelial to mesenchymal transition (EMT), migration, and CCC tumor growth. Sonic hedgehog (Shh) ligand was highly expressed in 89% of human CCC tissues and in CCC cell lines. Cyclopamine and 5E1 treatments effectively inhibited cell proliferation, migration, and invasion by down-regulating the Hedgehog target genes glioblastoma 1 and glioblastoma 2. In vitro and in vivo, we detected an increase in epithelial marker, E-cadherin, after Hedgehog inhibition. In addition, we saw an increase in necrotic areas and a decrease in mitotic figures in cyclopamine and 5E1-treated CCC xenograft tumors. Conclusion: This study supports the presence of autocrine Hedgehog signaling in human CCC, where CCC cells produce and respond to Shh ligand. Blocking the Hedgehog pathway inhibited EMT and decreased the viability of CCC cells. In addition, cyclopamine and 5E1 inhibited the growth of CCC xenograft tumors. (HEPATOLOGY 2013)
The incidence and mortality rate of cholangiocarcinoma (CCC) is increasing worldwide.1, 2 CCC represents the second most common primary hepatobiliary cancer.3 Most of the CCC tumors are adenocarcinomas arising from epithelial cells lining the intra- and extrahepatic biliary tract system.4, 5 The median survival of CCC patients is less than 24 months,6, 7 primarily because CCC is detected at an advanced stage, which impairs the possibility of curative surgery. Treatment with photodynamic therapy, systemic chemotherapy, and/or radiotherapy are the only options for patients with inoperable disease.8, 9 Previous studies have shown that CCC is characterized by a series of highly recurrent genetic abnormalities, including KRAS, BRAF, p53, SMAD, and p16INK4a mutations.10-15 Currently, the combination of gemcitabine and cisplatin is the standard chemotherapeutic regimen for patients undergoing first-line treatment.16, 17 Thus, strategies are still needed to overcome this deadly disease.
Hedgehog signaling regulates cell fate decisions, including proliferation, apoptosis, migration, and differentiation.18 Sonic hedgehog (Shh) was first discovered in a Drosophila screen for embryonic genes.19 Shh interacts with the cell surface receptor Patched (Ptc), which is expressed on Hedgehog target cells. In the absence of Hedgehog, Ptc represses the transmembrane protein Smoothened (Smo). When Shh is present, it binds to Ptc and releases the repression of Smo by Ptc. Smo can activate the Gliblastoma target genes glioblastoma 1 (Gli1), glioblastoma 2 (Gli2), and glioblastoma 3 (Gli3).18, 19 Aberrant activation of the Hedgehog pathway has been described for different cancer types.20-31 Recent studies have shown a correlation between the activation of epithelial to mesenchymal transition (EMT) as a result of Hedgehog signaling and cancer stem cells in gastrointestinal and nongastrointestinal malignancies.32-35 EMT promotes cancer invasion and metastasis.36-38 The hallmark of EMT is loss of the homotypic adhesion molecule epithelial cadherin (E-cadherin), which is expressed in most epithelial tissues, and gain of mesenchymal markers.39, 40 However, few studies have investigated EMT in CCC.41, 42
The cellular biological roles of Hedgehog in CCC are not completely known. It is unknown whether Hedgehog signaling in CCC is mediated by autocrine signaling or paracrine signaling or both. In the present study, we sought to further examine the potential benefits of targeting hedgehog in CCC by studying in the impact of cyclopamine and the monoclonal antibody (anti-Shh antibody) 5E1 treatment on human CCC cell lines. Our work revealed that cyclopamine, as well as 5E1, can block cell proliferation, EMT, migration, invasion, and apoptosis in human CCC cell lines in vitro. Additionally, both cyclopamine and 5E1 can suppress human CCC xenograft tumors in vivo. These findings support the development of therapeutic strategies targeting Hedgehog signaling and confirm the presence of Shh autocrine signaling in human CCC.
The human cholangiocarcinoma cell lines TFK1, SZ1, and EGI1 were generously provided by Nisar Malek. SZ1 was established from a surgically resected tumor specimen, which was histologically diagnosed as adenocarcinoma of moderate differentiation with cholangiolar differentiation. All cell lines were cultured in Roswell Park Memorial Institute 1640 medium plus GlutaMAX (Invitrogen, Karlsruhe, Germany) supplemented with 10% fetal calf serum (Biochrom, Berlin, Germany) and 100 U/mL penicillin/streptomycin (Invitrogen, Karlsruhe, Germany) at 37°C in 5% CO2.
Drug Preparation and In Vitro Treatment.
Stock solutions of cyclopamine, Shh antagonist, and tomatidine, a nonfunctional cyclopamine analog (Calbiochem, EMD Chemicals Inc., Darmstadt, Germany), were prepared by dissolving them in dimethyl sulfoxide (DMSO; AppliChem, Darmstadt, Germany). Cells were treated with DMSO, cyclopamine, or tomatidin at different concentrations (5 μM, 10 μM, 15 μM) and were analyzed after 24, 48, and 96 hours. Additionally, cells were treated with 5E1 antibody (40 μg/mL), SHH antagonist, and NS1 condition medium for 24, 48, and 96 hours (Developmental Studies Hybridoma Bank, Iowa City, IA).
The protocols for cell viability, invasion, migration assay, and cell death detection by fluorescence-activated cell sorting analysis are described in the Supporting Information.
Histology, Immunoblotting, Tissue Microarray, and Real-Time Polymerase Chain Reaction.
For detailed information of antibodies, primers and protocols for histology, tissue microarray samples, real-time polymerase chain reaction (PCR), and protein analysis can be found in the Supporting Information.
Animals and Treatment.
NMRI-nu/nu female nude mice were obtained from Charles River Laboratories International (Wilmington, MA). For the cyclopamine experiment, four mice were used for both the vehicle group and the cyclopamine (10 mg/kg) group. For the 5E1-treated mice, three mice were used for the vehicle group and the 5E1 (40 μg/mL) group. The mice were maintained under standard conditions according to the institutional guidelines for animal care of the Hannover Medical School and Lower Saxony. The injection, treatment procedure and calculation of tumors are described in the Supporting Information.
All experiments were repeated two or three times. The results were analyzed using software GraphPad prism version 5.0 (GraphPad Software, San Diego, CA) and SPSS version 11.0 (SPSS, Chicago, IL). The tests included one-way analysis of variance and Student t tests along with Bonferroni posttest and paired and unpaired t tests. Differences were considered statistically significant at P < 0.05.
Shh Is Expressed in CCC Cell Lines and Human Cancer Tissues.
To determine whether Hedgehog signaling is involved in the carcinogenesis of CCC, expression of Shh was assessed in TFK1, EGI1, and SZ1 cell lines via western blotting. Shh was highly expressed in all analyzed CCC cell lines (Fig. 1A). We then verified the activation of Hedgehog signaling by checking Shh protein expression in 119 human CCC tissues from 67 patients with different tumor gradings via immunohistochemistry staining (Fig. 1C). We observed that 89% (106 of 119) of the samples showed Shh expression. In contrast, normal bile ducts showed no Shh expression (Fig 1B). Interestingly, Shh had a stronger expression in tissues taken from patients with poorly differentiated tumors (grade 2 and grade 3) compared with highly differentiated tumors (grade 1) (Fig. 1D). The high expression of Shh in human CCC cell lines and tumor tissues makes it a good target for the treatment of poorly differentiated tumors.
Hedgehog Signaling Is Essential for Maintenance of Growth of Human CCC.
The aberrant activation of Hedgehog signaling has been reported for many solid tumors. To assess the role of the Hedgehog signaling pathway in CCC, we examined the effects of cyclopamine, a known Hedgehog inhibitor on the proliferation of human CCC cell lines. Proliferation assays showed that cyclopamine treatment reduced the number of viable TFK1 and SZ1 cells in a dose-dependent (5, 10, and 15 μM) and time-dependent (0, 24, 48, and 96 hours) manner compared with vehicle (DMSO and tomatidine) (Supporting Fig. 1A-D, Fig. 2A-D). To confirm that the pharmacological inhibition of cyclopamine is specific, we silenced the expression of Smo via small interfering RNA (siRNA) transfection. These results confirmed the previous cyclopamine results (Supporting Fig. 3A,B). To validate that the Shh ligand is essential to the carcinogenesis of CCC, we used 5E1, an Shh antagonist. Treatment with 5E1 led to a decrease in the proliferation of CCC cells (Fig. 2A-D). This decrease in proliferation by cyclopamine, Smo siRNA, and 5E1 was paralleled by a decrease in the Hedgehog targets genes Gli1 and Gli2 (Fig. 2E and Supporting Figs. 1E, 2A-C, and 3C). Taken together, Hedgehog signaling in general and the Shh signaling pathway in particular is necessary for the growth of CCC cells.
Cyclopamine Treatment Induces Apoptotic Cell Death in Human CCC Cell Lines.
To gain more insight into the cyclopamine-induced decrease in CCC cell proliferation, we analyzed the cell cycle profile of the CCC cell lines. Interestingly, we found that 15 μM cyclopamine led to a significant increase in the subG1 population in both cell lines at different time points, with the highest percentage being at 96 hours (33% of SZ1 cells and 27% of TFK1 cells) (Fig. 3A). Additionally, to analyze apoptosis, we performed an annexin V assay. Cyclopamine-treated cells showed a significant increase of apoptotic cells (Fig. 3B). To further confirm our results, we investigated the presence of biochemical markers of apoptosis. We noticed a time-dependent increase of the protein poly(adenosine diphosphate ribose) polymerase (PARP) cleavage after 15 μM cyclopamine treatment for SZ1 and TFK1 (Fig. 3C). We conclude that cyclopamine treatment induced significant levels of apoptosis in human CCC cell lines, which correlates with the increase in the population of cells in the SubG1 fraction of the cell cycle.
Targeting of Hedgehog Results in Inhibition of Migration and Invasion in CCC.
Tumor metastasis is an intricate process that is regulated by many pathways, including the Hedgehog signaling pathway. This process depends on the ability of cancerous cells to migrate and invade. Thus, we sought to investigate the role of Hedgehog on the migratory and invasive abilities of human CCC cell lines. First, we examined the migration ability of SZ1 and TFK1 cells by employing wound healing migration assays (Fig. 4A-D). For both cell lines, significant inhibition of wound healing was observed with cyclopamine treatment (P < 0.05). In contrast, 70%-80% wound healing was seen after 24 hours in all untreated cells. Next, we measured the invasive potential of human CCC cell lines after cyclopamine treatment. In correlation with migration inhibition, blocking the Hedgehog pathway via cyclopamine treatment reduced the invasive potential of SZ1 by 90% and TFK1 by 70% (Fig. 5A,B). Thus, treatment with cyclopamine can successfully inhibit migration of human CCC cells and has an anti-invasive effect on human CCC cell lines.
Hedgehog Signaling Pathway Regulates EMT in Human CCC Cell Lines by Modulating E-cadherin Expression.
It has been shown that activation of the Hedgehog pathway regulates EMT during bile duct development and chronic biliary injury.43 We hypothesized that blocking the Hedgehog pathway can partially reverse EMT in CCC. To address this question, we analyzed the protein expression of epithelial and mesenchymal markers after cyclopamine treatment. We observed a time-dependent increase in the protein expression of E-cadherin after cyclopamine treatment compared with vehicle. However, there was no change in the expression of N-cadherin (Fig. 5C). To further confirm the importance of E-cadherin in EMT modulation via the Hedgehog pathway, we knocked down E-cadherin using siRNA. This repression in E-cadherin (Supporting Fig. 4A) led to an abrogation in the effects of cyclopamine on migration and invasion of CCC cells (Supporting Figs. 4B,C and 5A,B). More importantly, we observed that the migration potential increased in cells with down-regulated E-cadherin expression (Fig. 5A,B). Our data show that cyclopamine treatment could selectively reverse the EMT phenotype in human CCC cell lines by relieving the inhibition on E-cadherin exerted by the Shh pathway.
Blocking the Hedgehog Pathway Effectively Inhibits the Growth of Xenograft Tumors.
We further investigated the in vivo role of the Hedgehog pathway in a xenograft mouse model. TFK1 cells were subcutaneously injected into both flanks, and treatment was initiated with either cyclopamine (10 mg/kg), 5E1 (40 μg/mL), or vehicle. We observed a significant suppression of tumor growth in cyclopamine and 5E1-treated animals compared with the vehicle group (Figs. 6A,B and 7A,B). In general, cyclopamine and 5E1 application were safe and showed no side effects. Mouse body weight was not significantly different between the analyzed groups (Supporting Figs. 6A and 7A). Interestingly, histological analysis of explanted tumors from cyclopamine and 5E1-treated mice showed a remarkable decrease in weight (Supporting Figs. 6B and 7B). Histologically, treatment increased necrotic areas in the cyclopamine and 5E1-treated groups (60%-65%) compared with the vehicle group (10%) (Figs. 6C and 7C). In line with these findings, we also found a lower number of cells undergoing mitosis in the cyclopamine and 5E1 group (Figs. 6C and 7C). Cytokeratin 7, 17, and 19 staining did not show any differences between the analyzed groups (data not shown). To analyze the effect of cyclopamine and 5E1 treatment on the proliferation of TFK1, we performed Ki67 staining. Cell proliferation was found to be similar in both mouse groups (Supporting Figs. 6C,D and 7C,D). Next, we confirmed that Hedgehog target genes Gli1 and Gli2 were effectively inhibited in cyclopamine- and 5E1-treated tumors (Figs. 6E and 7E). To further confirm the role of Hedgehog signaling in the regulation of the EMT process in xenografts, we analyzed the expression of E-cadherin and N-cadherin. Interestingly, E-cadherin was up-regulated in the cyclopamine-treated tumors compared with the vehicle group (Fig. 6F). However, in line with our in vitro data, we did not see any change of N-cadherin expression (Fig. 6F). On the other hand, 5E1 treatment showed no change in E-cadherin expression (Fig. 7F). Our data suggest that treatment with cyclopamine and 5E1 causes a significant inhibition of human CCC cells in vivo by increasing necrosis and influencing mitosis rate. In addition, it confirms a leading role of the Hedgehog signaling pathway in the carcinogenesis of CCC.
Previous studies have highlighted the role of Hedgehog in the pathogenesis of different cancers. On the other hand, its role in the carcinogenesis of CCC is still to be elucidated. Recent studies suggest a role for the activation of the paracrine Hedgehog signaling by platelet-derived growth factor (PDGF) in CCC.44 However, comparative experiments to analyze the autocrine mechanisms in CCC have not been performed yet.
To analyze the importance of the Hedgehog pathway activation in CCC, we have blocked the pathway using cyclopamine and 5E1. Here, we show for the first time that Shh is expressed in 89% of analyzed human CCC tissues and correlates with tumor grading. Tissue microarray analysis showed strong Shh staining in grade 2 and grade 3 tumors. Our novel finding of Shh expression in the majority of human CCC may be of clinical significance, because a reliable immunohistochemical marker for CCC and new molecular target is lacking.
This study suggests that Shh is responsible for the initiation of Hedgehog signaling in CCC mainly via autocrine mechanism. We have shown that cyclopamine and 5E1 treatments induced a dose- and time-dependent growth inhibition in human CCC cell lines and that the Hedgehog target genes Gli1 and Gli2 were down-regulated. Jinawath et al.45 showed that cyclopamine application increased the number of cells arresting at the G0/G1 phase of the cell cycle. In this study, Hedgehog inhibition led to an increase in the number of cells in the subG1 phase of the cell cycle, which correlates with the increase in the number of apoptotic cells. This increase in apoptosis was accompanied with an increase in the expression of apoptosis-related proteins such as cleaved PARP, which ultimately leads to DNA fragmentation.45 Thus, our data proves that Hedgehog signaling pathway is essential for the proliferation, growth, and maintenance of CCC. The results presented here provide insight into the autocrine mechanism by which Shh acts in human CCC cell lines. It has been suggested that tumors overexpressing Hedgehog primarily respond to autocrine mechanisms.46 Hence, cyclopamine treatment and 5E1 treatment are able to effectively influence CCC tumor growth. However, cyclopamine seems to be more efficient in inhibiting proliferation of CCC. This could be due to that fact that cholangiocytes might secrete other Hedgehog ligands. In addition, Fingas et al.44 showed that malignant cholangiocytes can activate Hedgehog signaling independent of the ligand via PDGF-mediated activation of Smo. Thus, cyclopamine is a general inhibitor of the Hedgehog pathway, which makes it more effective.
E-cadherin, a protein modulated during EMT, is known to be expressed by liver progenitors and biliary epithelial cells.43 Araki et al.41 demonstrated for extrahepatic CCC that the cadherin switch promotes tumor progression via transforming growth factor β (TGF-β) signaling. It is also known that EMT induced by TGF-β1/Snail activation is closely associated with an aggressive growth of CCC and that Slug plays a role in the regulation of E-cadherin expression and acquisition of invasive potential of extrahepatic CCC.24, 47 The high metastatic potential of CCC and the known role of Hedgehog in the regulation of bile duct development highlights the necessity to further investigate the mechanisms regulating the migration and invasion of this deadly disease.43 Indeed, we found that treatment with cyclopamine resulted in an in vitro inhibition of migration and invasion. Our findings show that this inhibition correlates with a time-dependent increase of E-cadherin. In addition, our results suggest that E-cadherin is essential for the cyclopamine-mediated inhibition of invasion and migration. Surprisingly, we could not detect any down-regulation of the mesenchymal marker N-cadherin, arguing for a selective control of EMT by cyclopamine.
In CCC xenografts, we found a significant inhibition of tumor growth under cyclopamine and 5E1 treatment compared with vehicle. Our current study clearly demonstrates that in vivo inhibition of xenografts via blocking of the Hedgehog pathway resulted in a down-regulation of Gli1 and Gli2. Notably, we found significant necrotic areas in cyclopamine- and 5E1-treated tumors. Moreover, we showed a decrease in mitotic figures after Hedgehog inhibition, whereas cell proliferation was not affected. These phenomena could be explained by the fact that cyclopamine directly triggers cell death in the tumors without affecting the cell cycle, which correlates with an increase in the subG1 fraction and in cells undergoing apoptosis in our in vitro experiments. Fingas et al.44 showed that myofibroblast-derived PDGF ligand promotes apoptosis resistance and postulated that PDGF ligand is secreted by myofibroblasts. However, in our xenograft tumors, myofibroblasts were not detected; therefore, their role in CCC carcinogenesis remains undeciphered in our model. Interestingly, 5E1 treatment did not alter the expression of E-cadherin. This finding suggests that cyclopamine and 5E1 have different mechanisms in blocking migration and invasion. Thus, further studies are needed to understand the variation in these mechanisms.
Our study demonstrates the underlying role of Shh in CCC pathogenesis and suggests that human CCC is induced by an autocrine mechanism. These findings highlight the importance of the role of Shh in tumor progression of CCC and argue for targeting of the Hedgehog pathway as a therapeutic approach.
We thank the Developmental Studies Hybridoma Bank for providing the 5E1 antibody.