Figure 1.
Breast and prostate cancer cell lines express RANK (Receptor Activator of NFκB).
RANK expression was analyzed by RT-qPCR (a). Protein lysates from MDA-MB-231, MDA-231BO2 and PC-3 cells were analyzed for RANK protein expression by Western blot. β-Actin was used as loading control (b). All experiments were run in triplicate. Error bars represent variation between technical replicates (n = 3).
Figure 2.
RANKL-RANK pathway mediates migration and invasion of breast and prostate cancer cells.
RANKL stimulus does not affect the proliferation of MDA-231BO2 breast cancer cells (a). Migration (b–e) and invasion (f–h) assays were performed. Migration assays were performed with Oris Cell Migration Assay (b–d) or using 96-well chemotaxis chambers with polycarbonate filters (8 µm pore size) (e). Breast and prostate cancer cell lines have different basal migration levels (b). RANKL (1 µg/ml) increases migration of MDA-MB-231, MDA-231BO2 human breast cancer cells and PC-3 human prostate cancer cells, while migration of PC-3KDRANK cells in response to RANKL is significantly decreased. PC-3shNT prostate cancer cells were used as control (c). RANKL increases migration of MDA-231BO2 cells in a dose-dependent manner (d). Increased migration of MDA-231BO2 cells in response to RANKL (2 µg/ml) was abrogated by neutralizing RANKL (with 2.5 µg/ml anti-RANKL antibody), and is similar to the response to the cytokine SDF-1α (100 ng/ml) (e). Invasion assays using 96-well chemotaxis chamber with polycarbonate filters (8 µm pore size) coated with human type I collagen showed that RANKL increases invasion of MDA-231BO2 human breast cancer cells in a dose-dependent manner (f, g). RANKL (1 µg/ml) had a similar effect to the cytokine SDF1α (100 ng/ml). Neutralized RANKL or siRNA mediated knockdown of RANK significantly decreased RANKL stimulation. siRNA mediated knockdown of MMP-1 impaired invasion in a similar level of cells treatment with the PI3K inhibitor wortmannin (100 nM) (h). All experiments were run in triplicate. Error bars represent variation between technical replicates, except for siRNA mediated knockdown of MMP-1 and RANK were it represents the average of three independent clones. n = 3, *p<0.05, **p<0.01, ***p<0.005, using one-way ANOVA with a Newman-Keuls multiple comparison test.
Figure 3.
Activation of RANKL-RANK pathway up-regulates MMP-1 expression in breast cancer cells.
MMP-1 expression upon RANKL stimulus was analyzed by RT-qPCR. MDA-231BO2 breast cancer cells or PC-3 prostate cancer cells were cultured with 1 µg/ml RANKL and total RNA was extracted at different time points. MMP-1 mRNA expression (mean ± SEM) was measured by RT-qPCR (n = 3) (a). MMP-1 expression at the protein level upon RANKL stimulus was analyzed at different time points by Western blot. MDA-231BO2 cells were cultured with 1 µg/ml RANKL for 60 min. Protein lysates from the treated cells were analyzed for MMP-1 protein expression. β-Actin was used as loading control (b). PC-3 prostate cancer cells were transfected with pGL4.15[luc2P/hygro] plasmid containing the MMP-1 gene promoter sequence (−592/−31). Cells were serum-starved for 24 h, and then treated with 1 µg/ml RANKL for 60 min before measuring luciferase activity. Results are expressed as the mean ± SEM (n = 3) of the relative luciferase activity (c). RANKL induces ERK1/ERK2 and JNK phosphorylation on MDA-231BO2 and PC-3 cells. RANK knockdown abrogated JNK phosphorylation. Cells were serum-starved for 24 h and stimulated with 1 µg/ml RANKL for the indicated time periods. ERK1/ERK2 activation (Thr202/Tyr204 phosphorylation; p-ERK), and JNK activation (Thr183/Tyr185; p-JNK) were detected by Western blot. Total ERK1/2, JNK, and control β-actin protein levels are shown. Phosphorylated protein levels, p-ERK1/2 and p-JNK, were normalized by densitometry to total protein levels, ERK1/2 and JNK (prior normalized to β-actin levels) (d,e). All experiments were run in triplicate. Error bars represent variation between technical replicates. n = 3, *p<0.05, **p<0.01, ***p<0.005, using one-way ANOVA with a Newman-Keuls multiple comparison test.
Figure 4.
Knockdown of MMP-1 decreases osteolytic lesions and osteoclast recruitment to tumor-bone interface in vivo.
Representative MMP-1 staining in demineralized bone sections from mice with bone metastases inoculated with MDA-231BO2 parental, shNT or shMMP14.4 cells. MMP-1 expression was quantified according to stain intensity (0–3). Results are expressed as the mean ± SEM. ***p<0,005 with a one-way ANOVA with a Newman-Keuls multiple comparison test (n = 6–13 per group) (a). Representative x-ray images from hind limbs of mice 4 weeks post inoculation with MDA-231BO2 parental, shNT and shMMP-1 cells. Arrows indicate osteolytic lesions (b). Osteolytic lesion area measured on radiographs of hind limbs and forelimbs of mice with bone metastases. Results are expressed as the mean area ± SEM per mouse (n = 6–13 per group). ** p<0.01 compared to parental or shNT clones using a two-way ANOVA with a Bonferroni post-test at 4 weeks (c). Representative histology of femurs with tumor indicated by arrows (d). Tumor burden in hind limbs and forelimbs was measured by quantitative histomorphometry. Results are expressed as the mean ± SEM area per mouse. A one-way ANOVA with a Newman-Keuls multiple comparison test showed a significant difference between parental and shMMP14.4 groups (p<0.05) but no significant differences between shNT and shMMP14.4 groups (n = 6–13 per group) (e). Representative bone histology of the femurs. Osteoclasts are indicated by arrows (f). Osteoclast number was measured in the femur at 200× magnification. Results are expressed as the number of osteoclasts (OC) per mm2 bone-tumor interface (BTI). Results are expressed as the mean ± SEM OC/BTI (g). ns – no significant **p<0,01, ***p<0,005 with a one-way ANOVA with a Newman-Keuls multiple comparison test (n = 6–13 per group).