Figure 1.
Analysis of matriptase mRNA and protein levels in a panel of 16 human breast cancer cell lines and a non-tumorigenic breast epithelial cell line MCF10A.
(A) Matriptase protein levels in indicated cell lines as determined using reverse phase protein arrays (RPPA). (B) Matriptase mRNA expression levels in indicated cell lines based on the data from Array Express (E-TABM-157). The “cut off” line in (A) was set at the value registered for the MDA-MB-436 cells. These cells displayed the highest RPPA read-out from all the cell lines that showed no detectable matriptase expression as validated by western blots presented in Fig. S1. Therefore the line represents the highest registered background reading. The “cut off” for Matriptase mRNA expression levels shown in (B) was determined using statistical information from the expression arrays. For some cell lines the level of expression was called as not significant above background (or “Absent”). The “cut off” was set to the expression level of the highest “Absent” cell line (HCC1569). Error bars represent standard deviations.
Figure 2.
Analysis of matriptase protein levels in a panel of 107 primary tumor biopsies from Edinburgh Breast Cancer Unit.
The samples were spotted on the same slides as cell lines in Fig. 1 and the “cut off” line is placed at identical value as in Fig. 1A. Error bars represent standard deviations.
Figure 3.
Matriptase (MT-SP1) protein levels in MDA-MB-231 and 4T1 cells as compared to human MDA-MB-468 cells (A), and primary mice keratinocytes (B).
Tubulin represents the loading control.
Figure 4.
Properties of the MDA-MB-231 cells engineered to overexpress matriptase
(MT-SP1 A, MT-SP1 B) and the respective control clones (Empty A, Empty B). (A) Bright field images of the selected clones and parental cells. (B) Western blots performed on total cell lysates, and (C) proteins precipitated from conditioned medium, using matriptase specific antibody. (D) In vitro growth curves for the selected clones and the parental cell line. There were no significant differences between the clones (p>0.05). (E) Representative fluorescence images of MT-SP1 B and Empty A cells stained with fluorescein-labelled phalloidin (actin cytoskeleton) and DAPI (nuclei). Similar results were obtained with MT-SP1 A and Empty B clones. Scale bar 10 µm.
Figure 5.
Properties of the 4T1 cells engineered to overexpress matriptase (MT-SP1 A, MT-SP1 B) and the respective control clones (Empty A, Empty B).
(A) Bright field images of the selected clones and parental cells. (B) Western blots performed on total cell lysates, and (C) proteins precipitated from conditioned medium, using matriptase specific antibody. (D) In vitro growth curves for the selected clones and the parental cell line. Although some “between-clone” variations were found (p<0.05) they were not associated with the presence or absence of MT-SP1 overexpression. (E) Representative fluorescence images of MT-SP1 B and Empty B cells stained with fluorescein-labelled phalloidin (actin cytoskeleton) and DAPI (nuclei). Similar results were obtained with MT-SP1 A and Empty A clones respectively. Scale bar 10 µm.
Figure 6.
In vitro adhesive and migratory properties of selected 4T1 clones stably
overexpressing matriptase (MT-SP1) and respective control cells. (A) Migratory properties of indicated cells as determined in the Transwell migration assay. No statistically significant differences between the clones were found (p>0.05). (B) Attachment strength of indicated clones as determined in the detachment assay. The results obtained for MT-SP1 overexpressing clones were significantly different (p<0.005) from those obtained for control clones and parental cell line. (C) Western blots illustrating E-cadherin expression in the indicated cell lines (left), and representative immunofluorescence pictures of E-cadherin staining (green) in MT-SP1 overexpressing cells and control cells (right). (D) Western blots illustrating β-catenin expression in the indicated cell lines (left), and representative immunofluorescence pictures of β-catenin staining (green) in MT-SP1 overexpressing cells and control cells (right). The immunofluorescence data in panels (C) and (D) are for clones 4T1 Empty B and 4T1 MT-SP1 B respectively, but analogous results were obtained in 4T1 Empty A and 4T1 MT-SP1 A clones. Blue color represents DAPI staining (nuclei). The individual (not overlayed) images for β-catenin and DAPI stainings presented in panel (D) are provided in Figure S4. Error bars in (A) and (B) represent standard errors. Scale bars 30 µm.
Figure 7.
In vivo growth characteristics of indicated MDA-MB-231 and 4T1 cells engineered to stably overexpress matriptase (MT-SP1) and the respective control
clones. The cells were injected into mammary fat pads of CD1 nude mice and grown as described in the materials and methods section. There were at least five animals in each group. (A) Growth curves of MDA-MB-231 MT-SP1 B (black circles) and MDA-MB-231 Empty A (white circles) clones. The MT-SP1 overexpressing cells grew significantly slower than the empty vector control cells (p = 0.003 on day 27). (B) Growth curves for 4T1 MT-SP1 B (black circles) and 4T1 Empty B (white circles) clones. The MT-SP1 overexpressing cells grew slower than the empty vector control cells although this was not statistically significant (p = 0.107 on day 16 and p = 0.187 on day 20). Error bars represent standard errors.