Low FLNa expression levels are correlated with breast cancer progression. (A) FLNa expression levels were examined by blinded pathology analysis of TMA (n = 222 samples) from breast cancer patients with progressive disease: benign, in situ carcinoma, and invasive carcinoma. Histogram represents normalized mean of epithelial expression of FLNa levels (confidence intervals [CI] of 95% ± SEM) using a monoclonal antibody that recognizes the FLNa N terminus. Images below the histogram correspond to representative immunostaining of breast cancer tissue cores from progressive breast cancers. Differences among the three groups were compared with one-way analysis of variance and the post-hoc Tukey honestly significant difference test for multiple comparisons (*, P < 0.005). (B) An independent TMA was assembled from a total of 258 lymph node (LN)–positive and –negative breast cancer patients’ samples. Histogram shown represents normalized mean of epithelial expression of FLNa (confidence intervals of 95% ± SEM) as determined by quantitative evaluation of immunostaining. A significant decrease of FLNa expression was observed in lymph node–positive (n = 127) versus lymph node–negative breast cancer (n = 155; *, P = 0.008).

Inhibition of FLNa promotes cancer progression to metastases in a preclinical orthotopic breast cancer model. MDA-231-ErbB2 cells expressing empty retroviral particles (PSR) or particles containing control scrambled shRNA (PSR-scramble) or FLNa-specific shRNA (PSR-FLNa) were implanted into the mammary fat pad of 10 SCID mice per cell type. Primary tumors were excised when they reached a size of 0.6–0.8 cm³, and animals were kept under observation for a period of 3 mo to allow metastasis formation. Mice were then subjected to autopsy, and lung metastases were examined. (A) Representative Western blot analysis of total cell extracts from MDA-231-ErbB2 tumor cells isolated from lung metastatic nodules confirming efficient stability of FLNa knockdown after in vivo implantation. GAPDH was used as an internal control. MM, molecular mass. (B) Graphs show the mean number of lung metastases ± SEM (*, P < 0.005), and images show representative lungs with metastatic nodules. Arrows indicate surface lung metastatic nodules. Bar, 8 mm.

Inhibition of FLNa stimulates breast cancer cell migration and invasion. (A) Western blot showing FLNa and FLNb expression in MDA-231, MDA-231-ErbB2, MCF-7-ErbB2, and BT-20 cells stably engineered to express FLNa-shRNA (PSR-FLNa) and their matched control cells expressing PSR puro vector alone (PSR). GAPDH was used as an internal control. MM, molecular mass. (B) Invasion of MDA-231, MDA-231-ErbB2, MCF-7-ErbB2, and BT-20 cells expressing PSR (control) and FLNa-shRNA using the Boyden chamber assay as described in Materials and methods. Bar graph represents the mean number of invaded cells (*, P < 0.01), and representative images of invasive cells on the membrane are shown. Results are expressed as the mean ± SD of five independent experiments and five fields per condition. (C) Cell migration analyzed by the wound healing assay. MDA-231-ErbB2 cells were wounded and monitored over 24 h to determine the rate of migration into the scratched area. Bar graph represents the mean ± SD of five independent experiments, and representative images of the wound captured at different time points are shown. (D) Cell motility analyzed using the phagokinetic track assay as described in Materials and methods. Bar graphs shows a quantification (mean ± SD) of the cleared areas from four independent experiments with >100 tracks assessed per experiment (*, P < 0.01). Images on the right represent brightfield microscopic images of the cleared phagokinetic areas. Bars: (B and C) 150 µm; (D) 120 µm.

FLNa regulates FA disassembly. (A) BT-20 cells expressing empty retroviral particles (PSR) or FLNa-specific shRNA (PSR-FLNa) were seeded on fibronectin- or collagen-coated dishes and grown at 70% confluence. Total protein extracts were probed using antibodies against the indicated proteins. Bar graph shows quantification of ratio of phosphorylated/total proteins (mean ± SD) from four independent experiments (*, P < 0.05). (B) Serum-starved control (PSR) and FLNa-silenced (PSR-FLNa) BT-20 cells were incubated with 10 µM nocodazole (NZ) for 4 h. Control untreated cells and treated cells at the indicated times after nocodazole washout were fixed and immunoassayed with anti–paxillin-pY118 or anti-FLNa. Three independent experiments were performed. Bar, 20 µm. (C) Quantification of FA areas using Volocity software. Data are expressed as the area of FAs per DAPI-stained nucleus and were obtained from at least 20 cells per condition. Quantifications show mean ± SD from three independent experiments as in B (*, P < 0.01). (D) Representative Western blots of FAK and paxillin during FA disassembly. BT-20 cell lysates were collected at the indicated times as in B and immunoblotted with anti–FAK-pY397 or anti–paxillin-pY118. Bar graphs represent the densitometry result as percentage of density of PSR control at 0 min (which is 100%) from four independent Western blots. MM, molecular mass.

Time-lapse quantification of FA dynamics after FLNa inhibition. (A) Control (PSR) and FLNa-silenced (PSR-FLNa) BT-20 cells transfected with EYFP-paxillin were plated on chamber slide, stimulated with 10 ng/ml EGF, and then analyzed by time-lapse spinning disk confocal microscopy. Representative frames are shown. (B) Enlargements of the boxed regions of A showing white arrows that indicate positions of paxillin-containing FAs. (C) Quantification of EYFP-paxillin persistence in FAs. Duration measurements were quantified by counting the amount of time lapsed between the first and last frames in which an individual adhesion was followed. (D) Rate constants of EYFP-paxillin disassembly at individual adhesions were calculated as described in Materials and methods. (C and D) Quantifications show the mean ± SD from three independent experiments (*, P < 0.01). Bars: (A) 15 µm; (B) 5 µm.

Regulation of FA turnover by FLNa is calpain dependent. (A) Control (PSR) or FLNa-silenced (PSR-FLNa) BT-20 and MDA-231-ErbB2 cells were grown to confluence, and total lysates were used for the luminescent Calpain-Glo protease assay. The relative luminescence units (RLU) were averaged and normalized to the amount of proteins in cell lysates. The experiments were repeated four times, and the results are expressed as mean ± SD. (B) Control (PSR) or FLNa-silenced (PSR-FLNa) MDA-231-ErbB2 cells were cultured on 10-cm dishes and stimulated or not with degraded type I collagen (De.Collagen) for the indicated times. Where indicated, cells were treated with 50 µM of calpain inhibitor ALLN (22 h). Lysates were prepared at the indicated time points and analyzed by Western blotting using the corresponding antibodies. Control indicates unstimulated. Bar graph shows quantification and comparison of FA protein cleavage (mean ± SD) at the 5-min time point after degraded collagen stimulation from Western blot results generated from five independent experiments (*, P < 0.001). CT, C terminus; MM, molecular mass; NT, N terminus. (C) Control (PSR) and FLNa-silenced (PSR-FLNa) BT-20 cells were incubated with 10 µM nocodazole (NZ) plus 100 µM ALLN for 4 h followed by a washout. Bar graph shows the area of FAs per DAPI-stained nucleus (mean ± SD) obtained from five fields per condition and from three independent nocodazole washout assays. CTL, untreated.

Time-lapse quantification of FA dynamics after calpain 2 inhibition. (A) Western blot showing calpain 2 siRNA (sequence 1) efficiency in control and FLNa-silenced BT-20 cells. GAPDH was used as an internal control. MM, molecular mass. (B) Control (PSR) or FLNa-silenced (PSR-FLNa) BT-20 cells stably expressing EYFP-paxillin were transfected with calpain 2 siRNA, stimulated with EGF, and then analyzed by time-lapse spinning disk confocal microscopy. White arrows show the positions of paxillin-containing FAs. Four independent experiments were performed, and representative frames are shown. Bar, 5 µm. (C, left) Quantification of the persistence of EYFP-paxillin in control and FLNa knockdown cells. (right) Quantification of EYFP-paxillin disassembly. Quantifications show the mean ± SD from three independent experiments.

FLNa down-regulation stimulates calpain activity via MAPK activation. (A) Lysates of parental BT-20 cells (CTL), cells expressing empty viral particles used for FLNa-shRNA expression (PSR), FLNa shRNA–expressing cells (PSR-FLNa), and their matched cells rescued with shRNA-resistant FLNa (PSR-FLNa⁺FLNa) or empty plasmid used for FLNa rescue (PSR-FLNa⁺pcDNA) were immunoblotted with FLNa-specific antibody. GAPDH was used as an internal control. (B) Control BT-20 cells (CTL), BT-20 cells expressing empty viral particles (PSR), BT-20 cells expressing FLNa-shRNA alone (PSR-FLNa, CTL), FLNa-shRNA cells pretreated for 2 h with 10 µM UO126 (PSR-FLNa, UO126), FLNa-rescued FLNa shRNA cells (PSR-FLNa, FLNa), and PSR-FLNa cells expressing empty plasmid (PSR-FLNa, pcDNA) were serum starved and then stimulated or not with 10 ng/ml EGF for 10 min. Total cell extracts were assayed for kinase activity (left) by immunocomplex kinase assay using anti-ERK2 for immunoprecipitation and MBP as kinase substrate in presence of γ-[³²P]ATP as described in Materials and methods or by Western blotting using anti–phospho-ERK1/2 (right). The radioactivity incorporated into MBP was quantified from the excised MBP bands by Cerenkov counting. (C) Samples as in B were used to determine calpain activity using the luminescent Calpain-Glo protease assay as described in Materials and methods. (B and C) Results are expressed as the mean ± SD of three independent experiments (*, P < 0.005). MM, molecular mass.

FLNa and talin are competitive substrates for cleavage by calpain and differentially regulate breast cancer cell invasion. (A) Control (PSR) and talin-silenced (PSR-Talin) BT-20 cells were cultured on 10-cm dishes and stimulated with degraded type I collagen (De.Collagen) to promote calpain-mediated cleavage of FA proteins. Control indicates unstimulated. Cell lysates were immunoblotted with the indicated antibodies. CT, C terminus; MM, molecular mass; NT, N terminus. (B) Quantification and comparison of FA protein cleavage (mean ± SD) at 5-min time point after degraded collagen stimulation from Western blot results of three independent experiments as in A and Fig. S5 B (*, P < 0.005). (C) Invasion of control (PSR), FLNa-silenced (PSR-FLNa), and talin-silenced (PSR-Talin) BT-20 cells using the Boyden chamber assay as described in Materials and methods. The bar graph shows the quantification of cell invasion results as the mean ± SD of four independent experiments and five fields per condition (*, P < 0.005; **, P < 0.01). Representative images of invasive cells on the membrane are shown. (D) Bar graph shows the quantification of cell migration analyzed by the wound healing assay. Control (PSR), FLNa-silenced (PSR-FLNa), and talin-silenced (PSR-Talin) BT-20 cells were wounded and monitored over 48 h to determine the rate of migration into the scratched area. The experiment was repeated four times, and results are expressed as mean ± SD. Representative images of the wound captured at different time point are shown. The vertical black lines delimit the wounded regions at different time points. Bars: (C) 150 µm; (D) 200 µm.

Cleavage of FLNa regulates FA turnover and cell invasion. (A) BT-20 siFLNa cells were transfected with construct either expressing GFP-tagged FLNa WT (GFP–FLNa-WT) or GFP-tagged FLNa mutant resistant to calpain cleavage (GFP–FLNa-Δ) and assayed for nocodazole (NZ)-based FA disassembly. Cells were untreated (control) or incubated with 10 µM nocodazole for 4 h. Cell lysates were collected at the indicated times and immunoblotted with anti–FAK-pY397 or anti–paxillin-pY118. (B) BT-20 siFLNa cells transfected with either GFP–FLNa-WT or GFP–FLNa-Δ were cultured on 10-cm dishes and stimulated or not with degraded type I collagen (De.Collagen) for the indicated times. Where indicated, cells were treated with 50 µM of calpain inhibitor ALLN (22 h). Lysates were prepared at the indicated time points and analyzed by Western blotting using the corresponding antibodies. Control indicates unstimulated. NT, N terminus. (C) Control BT-20 cells, BT-20 siFLNa cells, and BT-20 siFLNa cells expressing GFP–FLNa-WT or GFP–FLNa-Δ were placed in the top layer of Boyden migration chamber, and after 48 h, cells migrated through the chamber to the bottom membrane were measured. The graph represents the mean ± SD from four independent experiments (*, P < 0.005). (D) FLNa-silenced BT-20 cells were transfected with construct either expressing EGFP-tagged FLNa 14–16 repeats (EGFP-FLNa (14–16)) or expressing EGFP alone, stimulated or not stimulated with degraded type I collagen. Cell lysates were immunoblotted with the indicated antibodies. Bar graph shows quantification of cleavage of FAK (mean ± SD) from three independent experiments. MM, molecular mass.