Reduction of Gli1 expression significantly decreased migration and invasion of 231 cells. a QRT analysis demonstrated decreased Gli1 mRNA expression in 231 cells 48 h after transfection of 2 different siRNAs targeting Gli1 (siGli1-1, siGli1-2) compared with a non-targeting siRNA control (NT). Decreased expression was confirmed prior to each experiment. b and c Transwell migration (b) and invasion (c) assays were performed for 24 h. Silencing Gli1 significantly decreased migration (P < 0.001, ANOVA) and invasion (P < 0.001, ANOVA) of 231 cells in comparison to the NT control cells. Data were normalized to the NT control and are the mean and standard error of three independent experiments performed in triplicate. d The repressor form of Gli3 (Gli3R) was expressed by lentiviral transduction (pLenti6-Gli3R) of 231 cells (231-Gli3R). Gli3 overexpression in comparison to the vector control (231-Vector) was confirmed by QRT analysis. Increased expression was confirmed prior to each experiment. e Expression of Gli3R down-regulated Gli1 mRNA as measured by QRT in 231 cells. Data are the mean and standard error of three independent experiments (P = 0.022, t test). f and g Expression of Gli3R decreased migration (f) (P = 0.039, t test) and invasion (g) (P < 0.001, t test) of 231 cells (after 24 h) in transwell migration and invasion assays as compared with the vector control. Data are normalized to the vector control and are the mean and standard error of three independent experiments performed in triplicate

Expression of MMP11 is up-regulated by Gli1, and MMP11 is important for Gli1-mediated migration and invasion of 231 cells. a Over-expression of Gli1 resulted in an up-regulation of MMP11 mRNA in 231-Gli1 cells as measured by QRT. Data are the mean and standard deviation (P < 0.001, t test). b Western blot analysis demonstrated elevated expression of the secreted, active form of MMP-11 protein in conditioned media harvested from 231-Gli1 cells in comparison to the conditioned medium from the same number of vector control cells (231-Vector). The immunoblot demonstrated two separate bands of ~65 kD (latent proenzyme form) and ~45 kD (active form) of secreted MMP-11 protein. A cell lysate with a high level of MMP11 (ThermoScientific) was included as a positive control (Control). This immunoblot is representative of at least four separate experiments. c QRT analysis demonstrated that silencing Gli1 using siRNA reduces expression of MMP-11 mRNA in 231 and SUM1315 cells in comparison to the non-targeting control (NT). The data are the mean and standard error of two (231) or three (SUM1315) experiments performed in triplicate (P < 0.01 for each, ANOVA). d QRT analysis indicated a reduced expression of MMP-11 after transfection of siRNAs targeting MMP-11 (si-MMP11) in comparison to a non-targeting siRNA control (NT) in 231-Gli1 cells. e and f Western blot analysis of conditioned media from 231-Vector/NT, 231-Gli1/NT and 231-Gli1 with siRNA targeting MMP-11 (231-Gli1/siMMP-11) demonstrated a reduction in secreted MMP-11 protein in 231-Gli1/siMMP-11 in comparison to 231-Gli1 to level that was similar to 231-Vector/NT. This experiment was repeated three times and densitometric representation of these experiments in f is normalized to 231-Vector/NT (P = 0.034, t test). g and h Transwell migration (g) and invasion (h) assays were performed after silencing MMP-11 in 231-Gli1 cells. Silencing MMP11 significantly decreased transwell migration (P = 0.006, t test) and invasion (P = 0.001, t test) of 231-Gli1 cells (after 24 h). Data are normalized to the vector control and are the mean and standard error of three independent experiments performed in triplicate

Reduction of Gli1 expression by shRNA inhibits migration, invasion and metastasis. a and b Reduced expression of Gli1 by different shRNA in 231 cells (shGli1-1, shGli1-2) markedly decreased migration (a) (P < 0.001, ANOVA) and invasion (b) (P < 0.001, ANOVA) after 24 h in comparison to NT control cells. Data are the mean and standard error of three independent experiments performed in triplicate. c Silencing Gli1 significantly inhibited the number of metastasis of 231 cells to the lungs of female, athymic nude mice (P < 0.001, ANOVA) in an experimental metastasis assay. Parental 231 cells (n = 10 animals) or 231 cells transduced with the NT control (n = 10), shGli1-1 (n = 6), shGli1-2 (n = 7) or a combination of the two shGli1 (n = 10) were injected into the tail-veins. The numbers of metastases in the lungs were determined approximately 6 weeks after injection. d The volumes of each pulmonary metastasis was also estimated and the mean volume of the metastases in the lungs of the mice injected with shGli1-1 and shGli2-1 was smaller than the metastases in the NT control group (P < 0.001 for both comparisons, t test), but this was not the case for the shGli1-1/2 group because of several larger metastases identified at the pulmonary surfaces in this group. e and f Histologic examination confirmed that the metastases in the shGli1 groups were smaller than those in the NT and parental controls. Representative metastases (outlined in black) from the NT control (e) were larger than those in shGli1 groups (shGli1-2 shown here) (f) in hematoxylin & eosin (H&E)-stained histologic sections (×200 original magnification, scale bar = 50 μm). g and h H&E-stained sections of the metastases (outlined in black) identified at the pulmonary surfaces revealed that the metastatic cells only coated the pleural surface and did not extend deeply into the parenchyma in shGli1-1/2 mice (h), in contrast to those found at the pulmonary surfaces of the NT (g) and parental control mice (×200 original magnification, scale bar = 50 μm)

Over-expression of Gli1 enhanced migration and invasion of MDA-MB-231 breast cancer epithelial cells. a HA-tagged Gli1 was over-expressed by transduction (pLJD-HA-Gli1) of MDA-MB-231 (231-Gli1) cells followed by mass selection. Quantitative RT-PCR (QRT) confirmed an increase in Gli1 mRNA expression in 231-Gli1 in comparison to the empty vector control cells (231-Vector). Over-expression was confirmed prior to each experiment. b Western blot analysis with anti-HA confirmed that HA-tagged Gli1 was over-expressed in 231-Gli1 cells. Over-expression was also confirmed using a polyclonal antibody that recognizes Gli1 (anti-Gli1). F9 cells are a teratocarcinoma cell line with a high Gli1 expression and are a positive control. β-actin was also probed for as a loading control. c Transwell migration assays were performed for 24 h. The number of cells that migrated through the transwell filter was increased in 231-Gli1 cells. Data were normalized to the vector control and are the mean and standard error of three independent experiments performed in triplicate (P = 0.002, t test). d Transwell invasion assays through basement membrane material were performed for 24 h. The number of cells that traversed the transwell filter was increased in 231-Gli1 cells. Data were normalized to the vector control and are the mean and standard error of three independent experiments performed in triplicate (P = 0.027, t test)

Reduction of Gli1 expression significantly impaired migration and invasion of ERα negative SUM1315 breast cancer cells. a QRT analysis demonstrated that endogenous Gli1 expression was reduced in SUM1315 by transfection of two different siRNAs targeting Gli1 (siGli1-1, siGli1-2) in comparison to the non-targeting control (NT). Decreased expression was confirmed prior to each experiment. b and c Silencing Gli1 expression significantly inhibited migration (b) (P < 0.001, ANOVA) and invasion (c) (P < 0.001, ANOVA) of SUM1315 cells (after 24 h) in transwell migration and invasion assays. Data are normalized to the NT control and are the mean and standard error of three independent experiments performed in duplicate or triplicate

Sustained inhibition of Gli1 expression by shRNA or HPI-1 impaired growth of 231 cells. a Reduction in Gli1 expression 2 weeks after transduction of shRNA targeting Gli1 (shGli1-1, shGli1-2) in 231 cells was verified by western blot analysis using a rabbit polyclonal antibody (anti-Gli1). β-actin is a loading control. F9 cells are a teratocarcinoma cell line with a very high expression of Gli1 and are a positive control. b QRT analysis demonstrated that a reduction in Gli1 mRNA was maintained for 1 month after transduction of shGli1-1 and shGli1-2 in 231 cells in comparison to transduction of a non-targeting shRNA control (NT). c Cell growth was measured at 5 and 10 days after transduction of shRNA targeting Gli1 in 231 cells by MTT assay. Reduced expression of Gli1 for 5 days had little impact on cell growth, whereas, after 10 days of inhibition of Gli1 expression there was a significant decrease in cell growth (P < 0.001, ANOVA). Data are the mean and standard error of three independent experiments performed in triplicate. d To determine whether there was an increase in apoptosis in 231 cells with sustained reduction in Gli1 expression by shRNA (shGli1-1, shGli1-2), western blot analysis for cleaved PARP was performed on day 9 after transduction with shRNA targeting Gli1. The immunoblot was also probed for β-actin as a loading control. There was an increase in cleaved PARP, as indicated by an increase in the 89 kD cleavage product, in the shGli1-1 and shGli1-2 cells in comparison to the NT control. e The mean optical density of the bands corresponding to cleaved PARP on immunoblots from three replicate, independent experiments was greater in the shGli1 cells in comparison to the NT control cells (P < 0.001, ANOVA). The optical densities for PARP were normalized to the optical densities of the corresponding bands for β-actin. f 231 cells with sustained suppression of Gli1 expression (shGli1-1) and the corresponding NT control cells were immunostained for Ki-67 on day 9 after transduction. The percentages of Ki-67 labeled cells (a minimum of 500 cells was counted) was significantly lower in the shGli-1 cells in comparison to the NT control (P = 0.02, t test). Data are the mean and standard error of three independent experiments. g 231 cells were treated with HPI-1 (5 and 10 μM), an antagonist of Gli-mediated transcription, for 48 h. Inhibition of Gli-mediated transcription was indicated by a reduction in Gli1 mRNA, as quantified by QRT, in treated cells in comparison to the vehicle control. Data are the mean and standard deviation (P < 0.01, ANOVA). h MTT assay was performed on 231 cells treated with HPI-1 for 3 and 8 days. HPI-1 did not decrease growth after 3 days of treatment, but inhibited growth after 8 days of treatment (P < 0.001, ANOVA). Data are the mean and standard error of three independent experiments performed in triplicate