Silibinin suppresses growth of human colorectal carcinoma SW480 cells in culture and xenograft through down-regulation of beta-catenin-dependent signaling

Abstract

Mutations in APC/beta-catenin resulting in an aberrant activation of Wnt/beta-catenin pathway are common in colorectal cancer (CRC), suggesting that targeting the beta-catenin pathway with chemopreventive/anticancer agents could be a potential translational approach to control CRC. Using human CRC cell lines harboring mutant (SW480) versus wildtype (HCT116) APC gene and alteration in beta-catenin pathway, herein we performed both in vitro and in vivo studies to examine for the first time whether silibinin targets beta-catenin pathway in its efficacy against CRC. Silibinin treatment inhibited cell growth, induced cell death, and decreased nuclear and cytoplasmic levels of beta-catenin in SW480 but not in HCT116 cells, suggesting its selective effect on the beta-catenin pathway and associated biologic responses. Other studies, therefore, were performed only in SW480 cells where silibinin significantly decreased beta-catenin-dependent T-cell factor-4 (TCF-4) transcriptional activity and protein expression of beta-catenin target genes such as c-Myc and cyclin D1. Silibinin also decreased cyclin-dependent kinase 8 (CDK8), a CRC oncoprotein that positively regulates beta-catenin activity, and cyclin C expression. In a SW480 tumor xenograft study, 100- and 200-mg/kg doses of silibinin feeding for 6 weeks inhibited tumor growth by 26% to 46% (P < .001). Analyses of xenografts showed that similar to cell culture findings, silibinin decreases proliferation and expression of beta-catenin, cyclin D1, c-Myc, and CDK8 but induces apoptosis in vivo. Together, these findings suggest that silibinin inhibits the growth of SW480 tumors carrying the mutant APC gene by down-regulating CDK8 and beta-catenin signaling and, therefore, could be an effective agent against CRC.

Figure 1

Silibinin inhibits the growth of human CRC SW480 cells. Cells were plated overnight and treated with 50 to 200 εM silibinin for 24 to 72 hours. At the end of treatment, cells were collected and counted on a hemocytometer after staining with Trypan blue dye under the microscope for (A) total cell number (live and dead cells together) and (B) dead cells represented as percent dead cells. Data shown are mean ±SD. *P <.001 compared with the control. SB indicates silibinin.

Figure 2

Silibinin selectively decreases the total, cytoplasmic, and nuclear pools of β-catenin in human CRC SW480 cells. Cells were plated overnight and treated with 50 to 100 εM silibinin for 24 to 72 hours. At the end of treatment, SW480 (A) and HCT116 (B) cells were analyzed for total β-catenin levels by Western immunoblot analysis. HCT116 cells were also analyzed for the total cell number after these treatments (B), and the data shown are mean ± SD. ^$P < .05 compared with the control. (C) In separate studies, after similar treatments of SW480 cells, cytoplasmic and nuclear fractions were prepared and analyzed for β-catenin levels by Western immunoblot analysis. Membranes were reprobed with histone H1 (N, nuclear fraction) or GST-π (C, cytoplasmic fraction) as loading controls and as markers for purity of the fractions. Densitometry data shown below each band represent fold change compared with the control after normalization with respective loading controls. (D) Cellular localization of β-catenin was also observed by immunofluorescent staining followed by confocal imaging. For this, SW480 cells grown on coverslips were treated with either DMSO alone (control) or 100 εM silibinin for 48 hours, fixed, and stained for β-catenin (red) or nuclei with DAPI (blue). Original magnification, x1000.

Figure 3

Silibinin inhibits β-catenin-mediated transcriptional activity and the expression of its target genes in human CRC SW480 cells. (A) Cells were plated to 40% to 50% confluence overnight and transfected with 1 εg of TOP/FOP FLASH plasmid constructs along with 300 ng of pRL-CMV plasmid for 24 hours and then treated with DMSO alone (control) or 100 εM silibinin for another 24 hours. Luciferase activity was measured using Dual Luciferase Assay kit (Promega) following the manufacturer's instruction. Data shown represent mean ± SD of three independent observations. (B) Cells were plated overnight and treated with 50–100 εM silibinin for 24 to 72 hours. Total cell lysates were then analyzed by Western immunoblot analysis for CDK8, cyclin C, cyclin D1, and c-Myc levels. Membranes were reprobed with β-actin as loading control. Densitometry data shown below each band represent fold change compared with control after normalization with respective loading controls (β-actin). ^$P < .05 compared with control.

Figure 4

Silibinin treatment inhibits human CRC SW480 xenograft growth in athymic nude mice. Mice were subcutaneously injected with SW480 cells mixed with Matrigel and, after 24 hours, gavaged with CMC (control group) or 100 (SB-100) and 200 mg/kg body weight/day doses (SB-200) of silibinin for 5 days/week for 6 weeks. (A) Tumor volume/mouse as a function of time, (B) tumor weight/mouse at the end of study, (C) average body weight/mouse, and (D) average diet consumption/mouse/day were analyzed as detailed in Materials and Methods. Data shown in panels A and B are mean ± SE from eight mice in each group.

Figure 5

Silibinin treatment inhibits cell proliferation and induces apoptosis in human CRC SW480 xenograft. Xenograft tumor tissues were analyzed for PCNA- and TUNEL-positive cells. Representative photographs for IHC staining of PCNA-positive (A) and TUNEL-positive (C) cells in tumor tissue from the vehicle control and silibinin-fed groups, respectively, are shown at 400x magnification. (B) Percent PCNA labeling index and (D) percent TUNEL-positive apoptotic cells in tissues were analyzed as detailed in Materials and Methods. Data shown represent mean ± SE from eight mice in each group. SB-100 and SB-200 indicates 100 and 200 mg/kg body weight silibinin, respectively.

Figure 6

Silibinin treatment suppresses β-catenin, cyclin D1, c-Myc, and CDK8 expression in human CRC SW480 xenograft. Xenograft tumor tissues were analyzed for (A) β-catenin-, (B) cyclin D1-, (C) c-Myc-, and (D) CDK8-positive cells as detailed in Materials and Methods. Data shown represent mean ± SE from eight mice in each group.