Influence of SHH/GLI1 axis on EMT mediated migration and invasion of breast cancer cells

Abstract

Sonic Hedgehog signaling is critical for breast morphogenesis and cancer. The present study was conducted to explore the influence of SHH/GLI1 axis on epithelial mesenchymal transition and invasion in breast cancer cells. SHH/GLI1 positive samples demonstrated high expression of Snail and Vimentin with relatively low expression of E-cadherin. Overexpression of Vimentin and Snail in SHH/GLI1 positive patients was also associated with poor overall survival. Interestingly, GANT61 (GLI1 inhibitor) exposure significantly reduced cell viability and induced apoptosis at 10 µM. Suppression of Hedgehog pathway either by CRISPR mediated SHH knock out or GANT61 altered regulation of EMT markers in breast cancer cells. Moreover, in-activation of SHH/GLI1 axis also significantly restricted cell migration and invasiveness. These findings suggest that targeting SHH/GLI1 axis alters expression of EMT markers and abrogates neoplastic invasion in breast cancer cells.

Figure 1

Transcriptional profiling of SHH, GLI1 and EMT markers in breast cancer cohort. (a) Scatter plots showing mean relative mRNA expression of SHH, GLI1, E-cadherin, Vimentin and Snail in tumor samples as compared to adjacent normal tissues (Wilcoxon signed rank test, ***p < 0.0001). (b) Strong correlation was observed between expression of SHH and GLI1 in tumor samples (Spearman Correlation). (c) Immunohistochemical analysis of SHH, GLI1, E-Cadherin, Vimentin and Snail in tumor samples. (d) Box plot of SHH, GLI1 and EMT markers showing expressional variation between metastatic and non-metastatic patients (Mann whitney U test, ***p < 0.0001). Expression of GLI1 was significantly higher in luminal B and TNBC subtypes (e) while tGLI1 was exclusively found in triple negative patients (f). Second quartile of box whiskers plots are median values of data and ends of the whiskers are representative of minimum and maximum data points (**p < 0.01, ***p < 0.0001). Graphical data points and images are representative of at least three independent experiments.

Figure 2

Association of EMT markers with overall survival in SHH/GLI1 positive patients. (a) Kaplain Meier plots of SHH and GLI1 showing association with poor overall survival in Pakistani breast cancer cohort. (b) Expression profile of E-cadherin, Vimentin and Snail in SHH/GLI1 high and low patients. (c) Kaplain Meier plots of E-cadherin, Vimentin and Snail in SHH/GLI1 positive patients showing association with unfavorable outcome. Data is representative of 3 independent experiments (***p < 0.0001).

Figure 3

Effect of GANT61 on cell proliferation and apoptosis of breast cancer cells. Cell viability assays were conducted using CCK-8, to observe the effect of GANT61 on cell proliferation in MDA-MB-231 and MCF-7 in both (a). Dose dependent (5, 10, 15, 20 µM) and (b). Time dependent manner (24, 48, 72, 96hrs). IC50 of GANT61 was determined to be 10 µM. MDA-MB-231 and MCF-7 cells were treated with 10uM GANT61 and the control medium containing DMSO (Unpaired t test, ***, ^###p < 0.0001). Cell apoptosis was assessed using Annexin V-Cy3 kit and readings were taken after 48hrs of exposure at different concentrations (5, 10, 15, 20 µM) for (c). MDA-MB-231 and (d). MCF-7. Histograms showing quantitative assessment of apoptosis in c and d. Scatter plots and peaks in first panel of c and d represent untreated, unlabeled control, second panel untreated labeled control, and other panels illustrate different doses of GANT61. All results are representative of three independent experiments.

Figure 4

Inhibition of Hedgehog pathway using CRISPR/CAS9 mediated SHH knock-out and GANT61. (a). Schematic diagram of CRISPR/CAS9 mediated SHH knockout showing sequence of exon 1 having sgRNA (green), PAM sequence (red) and representative sequence showing modification at cutting site in knockout cells as compared to wild type. Western blot (b) and qRT-PCR (c) showing effective silencing of SHH, PTCH1 and GLI1 in KO1 and KO2 cells. Expression of SHH, PTCH1 and GLI1 was restored in KO1 rescue (KOR) cells in both lines (Anova with Dunnette post hoc test, *p < 0.05, **p < 0.001, ***p < 0.0001). Box plot demonstrating dose dependent decrease in transcription of SHH and GLI1 after treatment with 10 µM GANT61 in (d). MDA-MB-231 and (e). MCF-7 cells. Box plots (f) and western blots (g) showing down regulation of SHH, PTCH1 and GLI1 after 10 µM GANT61 treatment (Unpaired t test, ***p < 0.0001). All results are representative of three independent experiments.

Figure 5

Suppression of Hedgehog pathway activation using CRISPR/CAS9 mediated SHH knock-out and GANT61 in breast cancer cells. Immunofluorescence staining indicating down regulation of SHH and GLI1 in MDA-MB-231 (a–c) and MCF-7 (d–f) in CRISPR/CAS9 mediated SHH knockout (SHHKO) and rescued (SHHKOR) cells. Similar decrease in pathway activation was observed after GANT61 (10 μM) treatment in MDA-MB-231 (g–i) and MCF-7 (j–l) cells (Scale bar 100 µm). Box plots (c,f,i,l) indicate amount of SHH and GLI1 in both cell lines after immunofluorescence quantification in relative units (r.u.). Horizontal lines represent median values and whiskers indicate minimum and maximum values (Anova with Dunnette post hoc test in knockout and Unpaired t test in GANT61 treated cells, ***p < 0.0001). All results are representative of three independent experiments.

Figure 6

Inhibition of SHH/GLI1 axis alters expression of Snail, E-cadherin and Vimentin in breast cancer cells. Transcriptional variation of EMT markers (Snail, E-cadherin and Vimentin) after CRISPR/CAS9 mediated SHH knockout and GANT61 (10 µM) treatment in (a). MDA-MB-231 and MCF-7 cells (b). Western blot indicating altered expression of EMT markers after SHH knockout and GANT61 treatment in MDA-MB-231 and MCF-7 cells. Immunostaining showing expression of EMT markers in c). MDA-MB-231 and d). MCF-7 cells in SHHKO and GANT61 (10 µM) treated cells in comparison to wild type cells exposed to recombinant SHH (Scale bar 100 µm). Box plots (e,f) indicate amount of protein in both cell lines after immunofluorescence quantification in relative units (r.u.). Horizontal lines represent median values and whiskers indicate minimum and maximum values (Anova with Dunnette post hoc test, ***p < 0.0001). All results are representative of three independent experiments.

Figure 7

Inhibition of Hedgehog pathway using in-vitro models decreases migratory and invasive abilities of breast cancer cells. Wound healing assay was used to assess migration of breast cancer cells following GANT61 treatment, SHH knockout (SHHKO1) and knockout rescue (SHHKOR) in MDA-MB-231 (a) and MCF-7 (b) recorded after every 12 hours. (c) Box plots showing overall difference in invasion of cells after 48hrs measured using transwell assay in both cell lines. Invasion decreased in SHH knockout and GANT61 treated cells while rescued cells showed similar pattern as control cells. Horizontal lines represent median values and whiskers indicate minimum and maximum values (Anova with Dunnette post hoc test, ***p < 0.0001). (d) Representative cell invasion picture (Scale bar 50 µm). All results are representative of three independent experiments.