Figure 4. Change in the expression of epithelial and mesenchymal cell markers after GSI IX treatment in human pancreatic cancer. (A) KP3 and (B) BxPC3 cells were treated with control (DMSO) and GSI (2.5 mM, 5 mM and 10 mM) for 96 h. The expression of EMT markers: E-cadherin, N-cadherin, Slug and Vimentin were analyzed by Western blot. b-actin was used as a loading control. Both (A) Kp-3 and (B) BxPC3 showed no change in expression of epithelial marker E-cadherin but resulted in a GSI dose-independent down regulation of mesenchymal markers N-cadherin and Vimentin. We also detected a down regulation of the EMT transcriptional factor Slug after GSI treatment for both pancreatic cancer cell lines.
Figure 5. GSI treatment in sorted pancreatic tumor initiating CD44+/EpCAM+ cells reduces cell proliferation and selectively inhibits EMT. (A) Table showing the expression of CD44+, EpCAM+ cells and the combination of CD44+/EpCAM+ cells in pancreatic cancer cell line KP3. (B) CD44+/EpCAM+ cells were treated with GSI (2.5 mM, 5 mM and 10 mM) and control (DMSO) for 48 h to determine the role of Notch in regulating the cell proliferation. Cell proliferation was inhibited in a dose- and time-dependent manner. Note that these results reveal the anti-proliferative effects of GSI on human pancreatic tumor CD44+/EpCAM+ initiating cells. (C) Light microscopic pictures (106 magnification) were taken at 48 h to show the effect of GSI on cell proliferation. (D) The down regulation of the Notch pathway was confirmed by Western Blot for Notch downstream target Hes1. Compared to unsorted and sorted DMSO treated cells Hes1 showed a dose-dependent down regulation after GSI treatment. (D) CD44 and EpCAM were down regulated in a dose dependent manner. The black arrow is marking the protein lane of CD44. (E) Epithelial marker E-cadherin was unaltered, but mesenchymal marker N-cadherin, Vimentin and Slug showed dose-dependent down regulation. Consistent to its role in other solid tumors, CSCs are also responsible for tumor recurrence as well as tumour metastasis in pancreatic cancer. Several studies have shown the significant role of human pancreatic CD44+, CD133+ and ESA+ (EpCAM+) CSCs [40]. Hermann et al. has reported that human pancreatic CSCs are highly tumorigenic and others have shown that CSCs contain high levels of Notch1 and Notch2 [16,37]. We hypothesized that deregulation of the Notch signalling pathway by GSI will be a useful approach to treat pancreatic cancer stem cells. Li et al. also showed successfully that injection of human PDAC CD44+/ESA+ (EpCAM+) cells developed xenograft tumors, whereas CD442/ESA2 (EpCAM2) cells had a significant decrease in tumorigenic potential [23]. Besides that it was demonstrated that CD44+ cells are responsible for gemcitabine resistance in PDAC cells [49]. Based on this prior work, we have chosen to analyze pancreatic tumor initiating CD44+/EpCAM+ cells. In contrast, we did not use xenografts from human PDAC, but rather analyzed the common used human metastatic pancreatic cancer cell line KP3.
Our in vitro results clearly showed that the inhibition of Notch signalling by GSI resulted in a dose-dependent growth attenuation of human pancreatic tumor initiating CD44+/EpCAM+ cells, down regulated the Notch target Hes1 and decreased EMTrelated molecules. This might confirm the role of Notch in the maintenance of the CD44+/EpCAM+ subpopulation. We further proceeded to test the efficacy of GSI on pancreatic tumor initiating CD44+/EpCAM+ cells. In pancreatic cancer xenografts, we found a significant inhibition of tumor growth in all of the GSI treated xenografts compared to the vehicle group. To attenuate possible drug side effects, we followed an established 3-day on and 4-day off dosing schedule [29]. GSI and vehicle treated animals showed no significant changes of body weight or other abnormalities. Our current study clearly demonstrated that in vivo inhibition of pancreatic tumor initiating CD44+/EpCAM+ cells by GSI resulted in a down regulation of mesenchymal cell markers and up-regulation of epithelial cell markers and can suppress tumorigenesis.
Figure 6. GSI IX inhibits the growth of subcutaneously injected pancreatic CD44+/EpCAM+ cells. (A) Diagram shows time course of tumor growth in GSI and vehicle treated animals. The tumor growth volume was significantly decreased in the treated compared to the control group. P values were calculated with student’s t-test; (`*’ signifies P,0.05) error bars represent SD. (B) Pictures of mice taken at 5 weeks. The left mouse is treated with vehicle, the right mouse with GSI. The arrows are marking the growing xenografts. (C) Histology images of representative xenograft tumors from vehicle and GSI treated mice (206 magnification). (E) Ki67 staining of vehicle and GSI treated xenografts. (F) Quantification of Ki67 staining in the xenograft tumors, showing significant reduction in Ki67+ cells following GSI treatment (206 magnification). Differences were considered as statistically significant when the P-value was less ,0.05 and non significant “n.s.” when the P-value was higer .0.05. Error bars show standard deviation. (D) Western Blot analysis showed a down regulation of the Notch downstream target Hes1 in xenograft tumors of GSI treated animals compared to control mice. (D) Protein expression of CD44, EpCAM, E-cadherin, N-cadherin and Slug were analyzed by the Western blot analysis and b-actin was used as a loading control. Note that in vivo treatment with GSI is sufficient to reverse the EMT-associated “cadherin-switch” in xenograft tumors. Taking into account other signaling pathways like Hedgehog and mTOR which are known to affect pancreatic CSCs as well, we cannot rule out by our analysis that GSI IX also has some other indirect effects. Our findings support the previously identified CD44+ and EpCAM+ CSC population in human PDAC. In conclusion, our data confirmed and expanded on the CSC hypothesis and also showed an evidence for the first time that CD44+ and EpCAM+ xenografts underwent EMT partly upon the treatment with GSI. Taken together, this study demonstrates the central role of Notch signalling pathway in pancreatic cancer pathogenesis independent of their tumorigenicity and that pancreatic tumor initiating CD44+/EpCAM+ cells are down regulated by GSI. Future clinical investigations are needed to confirm these results and to establish GSI as a part of the treatment regime for pancreatic cancer patients.
Abstract
Notch signaling is essential to the regulation of cell differentiation, and aberrant activation of this pathway is implicated in human fibrotic diseases, such as pulmonary, renal, and peritoneal fibrosis. However, the role of Notch signaling in hepatic fibrosis has not been fully investigated. In the present study, we show Notch signaling to be highly activated in a rat model of liver fibrosis induced by carbon tetrachloride (CCl4), as indicated by increased expression of Jagged1, Notch3, and Hes1. Blocking Notch signaling activation by a c-secretase inhibitor, DAPT, significantly attenuated liver fibrosis and decreased the expression of snail, vimentin, and TGF-b1 in association with the enhanced expression of E-cadherin. The study in vitro revealed that DAPT treatment could suppress the EMT process of rat hepatic stellate cell line (HSC-T6). Interestingly, DAPT treatment was found not to affect hepatocyte proliferation in vivo.
