HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 24, 21998-22009, June 14, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/24/21998    most recent M200715200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kainulainen, T. Articles by McCulloch, C. A. Search for Related Content PubMed PubMed Citation Articles by Kainulainen, T. Articles by McCulloch, C. A. Social Bookmarking                      What's this?

Cell Death and Mechanoprotection by Filamin A in Connective Tissues after Challenge by Applied Tensile Forces^*

Tiina Kainulainen^§¶, Alexandra Pender, Mario D'Addario, Yuanyi Feng^**, Predrag Lekic, and Christopher A. McCulloch^§§

From the  Canadian Institutes of Health Research Group in Matrix Dynamics, University of Toronto, Toronto, Ontario M5S 3E2, Canada, ^§ Department of Prosthetic Dentistry and Stomatognathic Physiology, University of Oulu, Oulu, Finland, ^** Department of Neurology, Beth Israel Deaconess Medical Center, Harvard University, Boston, Massachusetts 02115, and the  Department of Preventive Dental Sciences, Faculty of Dentistry, University of Manitoba, Winnipeg, Manitoba R3E 0W2, Canada

Received for publication, January 23, 2002, and in revised form, March 4, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cells in mechanically challenged environments must cope with high amplitude forces to maintain cell viability and tissue homeostasis. Currently, force-induced cell death and the identity of mechanoprotective factors are not defined. We examined death in cultured periodontal fibroblasts, connective tissue cells that are exposed to heavy applied forces in vivo. Static tensile forces (0.48 piconewtons/µm² cell area) were applied through magnetite beads coated with collagen or bovine serum albumin. There was a time-dependent increase of the percentage of propidium iodide-permeable cells in force-loaded cultures incubated with collagen but not bovine serum albumin beads, indicating a role for integrins. Cells exhibited reduced mitochondrial membrane potential, increased caspase-3 activation, nuclear condensation, terminal deoxynucleotidyl transferase nick end labeling staining, and detachment from the culture dish. The caspase-3 inhibitor acetyl-Asp-Glu-Val-Asp-aldehyde reduced detachment 3-fold. There was a rapid (<10-s) decrease in plasma membrane potential after force application, which, in filamin A-deficient melanoma cells, contributed to irreversible cell depolarization. In fibroblast cultures, cells with increased permeability to propidium iodide exhibited ~2-fold less filamin A content than impermeable cells. Fibroblasts transfected with antisense filamin A constructs or with filamin A constructs without an actin-binding domain exhibited 2-3-fold increased proportions of dead cells relative to controls. We conclude that high amplitude forces delivered through integrins can promote apoptosis in a proportion of cells and that filamin A confers mechanoprotection by preventing membrane depolarization.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In mechanically challenged tissues, the magnitude, direction, and frequency of physical forces can regulate extracellular matrix remodeling (e.g. bone trabeculae and periodontal ligament) (1, 2), indicating that the resident connective tissue cell populations can sense applied forces and appropriately generate signals to mediate homeostatic responses (3, 4). Applied physical forces can also induce cell death (5-7). For example, stretching forces promote cell death in cultures of cardiac myocytes and rat alveolar type II cells (5, 7) while cells of heavily loaded mineralized connective tissues exhibit increased death in vivo (8, 9). Accordingly the failure to adapt to applied mechanical stimuli may provide the pathobiological basis for the tissue destruction and loss of homeostasis observed in heavily loaded connective tissues (e.g. osteoarthritis) (10).

Both proliferation and death can regulate the size of a cell population. Appropriate regulation depends on specific paracrine and autocrine signals, ensuring that a cell divides only when more cells are required and that a cell survives only when and where needed (11-15). In addition to chemical signals, cells in mechanically active environments must cope with potentially injurious forces to avoid irreversible membrane damage (9). Notably, the cytoskeleton distributes and buffers applied forces at the plasma membrane, thereby reducing the conductance of stretch-activated ion channels and protecting against force-induced cell death (16). However, the force delivery systems and the nature of the mechanisms that prevent death of cells in mechanically active environment are not defined.

Recent data indicate that one of the earliest events in chemically induced cell death in vitro is plasma membrane depolarization (17). Accordingly, we considered that applied forces delivered through extracellular matrix protein receptors located on the plasma membrane of cells may initiate the death process by inducing depolarization. Conversely, membrane-associated proteins that dampen force-induced depolarization would be expected to protect against force-induced cell death. In this context the actin-binding protein filamin A (ABP-280) may protect cells from mechanically induced cell death by virtue of its ability to cross-link actin filaments (18, 19), increase the rigidity of cell membranes and the underlying cortical actin (20), and dampen force-induced calcium fluxes (21). Further, applied forces can induce transcriptional activation of the gene for filamin A (22), thereby suggesting a regulatory mechanism for mechanoprotection at the gene level.

In view of these findings, we examined the nature of the cell death that is induced by mechanical forces in fibroblasts from tissues constitutively exposed to high amplitude forces in vivo. We tested the hypotheses that cells exposed to static tensile forces applied through integrins die by apoptosis, that tensile forces cause membrane depolarization and that filamin A protects against force-induced death.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Antibodies-- Rabbit anti-mouse/rat filamin A antibody was a generous gift of Dr. Martin Stahlhut (23). Mouse anti-human filamin A antibody (clone PM6/317) was from Serotec (Cedarlane Laboratories, Hornby, Ontario, Canada). Mouse anti-human -actin (clone AC-15) was from Sigma. Fluorescein isothiocyanate-conjugated mouse monoclonal antibody to human ₁ integrin was from Beckman-Coulter (clone 4B4; Burlington, Ontario, Canada). Rabbit antibody to human active caspase-3 was from R & D Systems, Inc. (Minneapolis, MN). Magnetite microparticles were from Sigma. Acidified soluble collagen (Vitrogen 100) was from Cohesion Technologies (Palo Alto, CA).

Cells-- Studies of functional adaptations to applied force in periodontal ligament and gingiva indicate that the periodontium is an excellent model system to study mechanoprotective phenomena (16). Cultured human gingival fibroblasts and Rat-2 cells, which are phenotypically very similar to periodontal cells (24), were used primarily in this study since, with a collagen-magnetic bead system, forces can be applied through integrins to the actin cytoskeleton (25). Primary cultures of human gingival fibroblast were derived as described (26). Cells from passages 6-14 were grown as monolayer cultures as described (26). We also used Rat-2 fibroblasts (ATCC CRL 1764; American Type Culture Collection, Manassas, VA), which were cultured as described earlier (27). For experiments in which filamin A was present or absent, melanoma cell lines were grown in -MEM¹ with 8% fetal bovine serum and 0.5 mg/ml G418 (Invitrogen) (19). The filamin A+ melanoma cells (A7) were originally derived by stable transfection of a parental filamin A cell line (M2T) with a mammalian expression vector (LK444) that either did (A7) or did not (M2T) contain the cDNA for full-length ABP-280 (19).

Plasmids and Transient Transfections-- Rat-2 fibroblasts were transfected with Effectine (Qiagen, Mississauga, Ontario, Canada) according to the manufacturer's instructions. Cells were grown to 80% confluence in 35-mm diameter dishes and transfected with 5 µg of plasmid DNA. Fresh culture medium was changed after 16 h. Cells were washed with PBS and used for subsequent experiments 48 h later. All cells were co-transfected with a green fluorescent protein--actin fusion construct to estimate transfection efficiency.

Sense and antisense PCR-amplified constructs corresponding to the first 333 bp of the coding region of the filamin A gene were generated and ligated into the SmaI site of the pCI mammalian expression vector (Promega, Madison, WI). Briefly, PCR-amplified products were generated from oligonucleotides corresponding to positions +1 to +21 (1A; 5'-ATGAGTAGCTCCCACTCTCGG-3') and +313 to +333 (1B; 5'-AGGTGCTCAGCCAGAAGAAGA-3'). PCR amplification was performed on 100 ng of DNA obtained from human gingival fibroblasts or Rat-2 cells using Pfu Turbo DNA polymerase (Stratagene, La Jolla, CA) according to the manufacturer's instructions. The blunt-ended product was purified from a 2.0% agarose gel (QIAEX II Gel Extraction Kit; Qiagen) and ligated into the SmaI site of pCI (Promega) using T4 DNA Ligase (Invitrogen). The correct orientation of the insert was established using diagnostic restriction enzyme cleavage and confirmed through sequencing performed at the DNA Sequencing Facility (Center for Applied Genomics, Hospital for Sick Children, Toronto, Ontario, Canada).

Human filamin A expression vectors with and without the actin binding domains were produced as follows. First, to tag the human c-Myc monoclonal antibody 9E10 epitope to the N terminus of filamin A, a PCR fragment was synthesized using human filamin A cDNA as template and the oligonucleotides AGCTTGCCATGGAACAAAAGTTGATTTCTGAAGAAGATTTGAGTAGCTCCCACTCTCGG and TCTTCTTCCACGGCGCG as primers. The PCR product was cloned into pCR-topo^TM (Invitrogen), sequenced, and digested with HindIII and SalI. This product was used to replace the N-terminal sequence of full-length human filamin A cDNA in pCDNA3 (Invitrogen), which digested completely at the HindIII site and partially at the SalI site. To make the actin binding domain deletion of the human filamin A, a SalI site was introduced into the Myc-tagged full-length filamin A cDNA at position 697 by the Stratagene QuikChange^TM site-directed mutagenesis system using the oligonucleotides GCCATGCGCGGCGGTCGACTGGCTGGGCATCCC and GGGATGCCCAGCCAGTCGACCGCCTGCTGCATGGC. The resulting cDNA was digested and religated between the newly created SalI site and the original unique SalI site at position 61 of the human filamin A cDNA, which resulted in the deletion of cDNA encoding amino acids Thr²³ to Asp²³³ of human filamin A.

Force Generation-- A force generation model was used as previously described (25). Briefly, magnetite microparticles (Fe₃O₄) were coated with collagen (1 mg/ml) or BSA (1 mg/ml), neutralized to pH 7.4, rinsed with PBS, and incubated with cells. Following a 15-min incubation, excess unbound particles were removed by washing with ice-cold PBS. Cells were supplemented with fresh Dulbecco's modified Eagle's medium or -MEM. A ceramic permanent magnet (Jobmaster, Mississauga, Ontario, Canada) was placed on top of the dish to generate a tensile force of ~0.48 pN/µm² cell area, a force level that is comparable with that which may be applied to cells in vivo during normal function (21).

Identification and Quantification of Force-induced Cell Death-- After force application, cells were vitally stained with propidium iodide (PI; 20 µg/ml; Molecular Probes, Inc., Eugene, OR) for 5 min, washed with PBS, fixed with formaldehyde (2% in PBS) for 10 min, and stained with DAPI (in 0.2% Triton X-100) for total cell counts. Samples were examined with a fluorescence microscope at ×250. The number of PI-permeable cells and the total number of cells (DAPI-positive cells) were counted in three different sampling grids for each sample to yield the percentage of permeable cells. Values from three separate areas were averaged for each sample.

For detection of individual apoptotic cells, DNA strand breaks (28) were detected with terminal deoxynucleotidyl transferase (Roche Diagnostics), which catalyzes polymerization of nucleotides on free 3'-ends of DNA (fluorescence TUNEL method). Paraformaldehyde-fixed cells were permeabilized with Triton X-100, incubated in TUNEL reaction mixture, washed, and stained with DAPI. As a positive control, cells were treated for 24 h with staurosporine (1 µM), a protein kinase inhibitor that induces apoptosis in a wide variety of cell types (29). Transmission electron microscopy was performed after a 24-h force application. Cells remaining attached to the dish were fixed with 4% paraformaldehyde and 2.5% glutaraldehyde. Detached cells in the culture medium were sedimented, fixed, and included in the samples for analysis. Thin sections were stained with uranyl acetate and lead citrate and examined in a Hitachi H7000 transmission electron microscope.

For assessment of force-induced cell death in vivo, latex rubber separators were placed between the first and second mandibular molar teeth of male Wistar rats (250 g). This maneuver applies tensile forces to rat periodontal ligament cells in vivo (30). Controls consisted of animals without separators. Rats were killed at 3 days after stimulation, and jaws were prepared for paraffin sections. The presence of apoptotic cells was determined with TUNEL assays (Intergen, Purchase, NY) followed by counterstaining with methyl green.

Immunoblotting and Immunohistochemistry-- Immunoblotting was performed as described earlier (22). For immunohistochemistry, Rat-2 cells were cultured on 18-mm diameter coverslips, and force generation was performed as described above. Cells were fixed with paraformaldehyde, permeabilized in 0.1% Triton X-100, blocked with 2% BSA, and stained with rabbit anti-human/mouse caspase-3 (0.3 µg/ml), biotinylated second antibody, and fluorescein isothiocyanate-streptavidin. As a positive control, cells were treated with staurosporine (1 µM for 24 h) and stained for TUNEL as described above.

Mitochondrial Membrane Potential-- JC-1 is a mitochondrial membrane potential-sensitive dye that exists largely as a green fluorescent monomer at low membrane potential in dead or dying cells. At higher (physiological) mitochondrial membrane potentials, JC-1 forms red fluorescent "J-aggregates." Analysis of JC-1-stained cell suspensions was performed with a FACSTAR Plus flow cytometer (Becton Dickinson). Rat-2 cells were cultured on 60-mm dishes and subjected to 1 h or overnight force application. To retrieve floating cells that had detached previously from the bottom of the dish, the growth medium was sedimented, and floating cells were suspended in buffer containing 5% fetal bovine serum. For cells remaining attached to the bottom of the dish after force application, cells were prepared as single cell suspensions by tryspinization. Cells were sedimented and resuspended in buffer before adding JC-1 (0.5 µM; Molecular Probes) for 10 min. For positive controls, the mitochondrial membrane potential was dissipated with carbonyl cyanide m-chlorophenylhydrazone (CCCP; 1 µmol; Sigma). A minimum of 10,000 cells were analyzed for each sample. In some experiments, cell sorting was used to divide the cell population into cells with low or high red fluorescence.

Plasma Membrane Potential-- Gingival fibroblasts, Rat-2, or melanoma cells were plated onto coverslips at ~5 × 10⁵/coverslip. Magnetite bead loading was done as described above. DiBAC4 solution (Ref. 32; 5 µM in ethanol) was added to each coverslip for 10 min as described (17). Force was applied for 30 s, and the fluorescence due to DiBaC4 was measured with a photomultiplier tube optically interfaced to an inverted fluorescence microscope (Photon Technology International, London, Ontario, Canada). Excitation monochromators were set to 470 nm, and emission was collected through a bandpass filter (520/10 nm). In control experiments, cells were loaded with BSA-coated beads.

Caspase-3-- Rat-2 cells were treated with staurosporine and with or without acetyl-Asp-Glu-Val-Asp-aldehyde (DEVD-CHO; 1 µM), a caspase-3 inhibitor (Calbiochem) (38). After a 4-h incubation, cells were stained with PI and DAPI as described above. Rat-2 cells were also incubated with the caspase-3 inhibitor during force treatment, and the PI-permeable floaters were counted with a hemocytometer.

Statistical Analysis-- For all studies, experiments were repeated at least three times. For quantitative data, means and S.E. were computed. When appropriate, two sample comparisons were analyzed by Student's unpaired t test and statistical significance set at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Effect of Force on Cultured Cells-- After application of static tensile forces (0.48 pN/µm² cell area) through collagen-coated beads, attached Rat-2 cells and human gingival fibroblasts were stained with PI to estimate the percentage of membrane-permeable cells. Fibroblasts and Rat-2 cells exhibited similar kinetics in the increased percentage of PI-permeable cells over time (Fig. 1, A and B). After 4 h of force application, ~3% of human gingival fibroblasts and 5% of Rat-2 cells were PI-permeable; after 24 h of force application, ~30% of both human gingival fibroblasts and Rat-2 cells were PI-permeable. For both Rat-2 cells and gingival fibroblasts, cells incubated with BSA-coated beads and subjected to force exhibited ~5% PI-permeable cells by 24 h, and cells incubated with collagen beads but without force exhibited ~4% PI-permeable cells after 24 h. These results were not due to loss of BSA-coated beads from the cells, since the area of cells coated with collagen or BSA-coated beads was similar, as has been shown earlier (23). In separate experiments in which the levels of applied forces were adjusted by altering the distance between the magnet pole face and the bead-loaded cells, lower force levels (~0.2 pN/µm² cell area) produced no substantial increase of PI-stained cells above no-force controls. Higher force levels (~1.0 pN/µm² cell area) caused rapid increases of permeability to PI in >50% of the cells and in some instances caused detachment of the collagen-coated beads from the cells. Accordingly, for all subsequent experiments, force levels of 0.48 pN/µm² cell area were applied. This force level is thought to be comparable with levels applied in vivo (21) and did not cause artifactual removal of the cells from the dish.

View larger version (54K): [in this window] [in a new window]   Fig. 1.   Effect of mechanical force on cell death in cultured Rat-2 cells and human gingival fibroblasts. A, percentage of propidium iodide-permeable Rat-2 cells (left panel) and human gingival fibroblasts (right panel). Cells were incubated with collagen-coated (1 mg/ml) magnetite beads at a ratio of 10 beads per cell and subjected to vertically directed static forces (0.48 pN/µm2 of cell area) for the indicated time points. After force application, cells were stained with PI and DAPI, and the percentage of PI-permeable cells as a function of total (DAPI-stained) cells was determined by fluorescence microscopy. Numbers below bars indicate hours of applied magnetic force; *, collagen bead loading but without force application. B, DAPI staining of Rat-2 cells after 24 h of force application shows that cells are shrunken, and many of the nuclei are condensed and brightly stained (left panel); PI-stained nuclei show marked condensation and intense, uniformly bright staining in dead cells (middle panel); bright field micrograph to show cells with collagen-coated magnetite beads (right panel). Magnification is ×100. C, all detached cells (floaters) in culture medium are PI-permeable (upper panel) prior to fixation and permeabilization. DAPI staining after fixation and permeabilization shows nuclei of the same cells (lower panel). Magnification is ×250. D, electron photomicrographs showing morphological changes in Rat-2 cells after 24-h force treatment. Some cells have diffusely distributed chromatin and large lysosome-like structures (panel 1), whereas other cells exhibit loss of cytoplasmic structure, and only remnants of organelles can be recognized (panels 2 and 3).

Exposure of cells to force applied through collagen-coated beads caused a time-dependent increase in the number of floating cells. After force application, the medium was removed from cultures, and sedimented cells were stained with PI, fixed, permeabilized, and stained with DAPI. Enumeration of cytospin preparations showed that after 24 h of force, there were ~3-5-fold increased numbers of detached cells compared with control cultures without force. All detached cells (floaters) were permeable to PI (Fig. 1C) and did not attach to culture dishes after replating. Electron microscopy of force-treated Rat-2 cells (Fig. 1D, 1-3) showed degenerative changes including condensation of chromatin and the formation of large lysosome-like structures (Fig. 1D, 1). In some cells, there was also loss of normal cytoplasmic structure, and in many other cells only remnants of organelles could be recognized (Fig. 1D, 2 and 3). The morphological aspects of apoptotic cell death in the floating and attached cells were similar, including chromatin condensation and large lysosome-like structures, although the floating cells were much smaller (see below).

We considered that the induction of cell death by force may promote the release of factors from dying or dead cells that promote cell death in other, still living cells. To test this concept, we incubated Rat-2 cells with collagen beads and then applied (or did not apply) force for 14 h as described above. The medium from the force-treated and control cells was collected, and the floating cells were pelleted. This "conditioned medium" was added full strength to fresh cultures of Rat-2 cells for 4 h, and cells were vitally stained with PI. The percentage of PI-stained cells was close to 0 in both force-treated and control media (one ×25 microscopic field counted in five replicate cultures; percentage of PI-positive cells as follows: force-treated medium = 1.0 ± 0.4%; control-treated medium = 0.7 ± 0.3%; p > 0.2). Thus, we were unable to detect any soluble factors released from mechanically stimulated cells that may promote cell death.

Characterization of Cell Death-- After exposure to tensile forces for 4-24 h, morphological examination of cells remaining attached to the culture dish and floating cells showed that after staining with PI and DAPI, the floating cells were shrunken, and many nuclei were condensed and brightly stained. There was no difference in the morphological appearance of nuclei from floating and PI-permeable attached cells. To detect the presence of presumptive apoptotic cells after force application, in situ DNA nick labeling was used to stain both attached and floating cells. After overnight force application, Rat-2 cells showed high proportions of positively stained cells, indicating the presence of fragmented DNA (~20%; Fig. 2A). The nuclei of TUNEL-positive cells were also stained with PI (Fig. 2A, 2). DAPI staining showed more condensed brightly stained nuclei in presumptive apoptotic cells (Fig. 2A, 3). As a positive control for apoptosis, separate cultures were treated with staurosporine, and these cultures showed large numbers of TUNEL-positive cells (Fig. 2A, 4). These in vitro results with cultured cells were consistent with results obtained from experimental force application for 3 days in vivo in which we counted >5 TUNEL-positive fibroblasts per high power microscopic field at sites of increased mechanical stress in the periodontium (Fig. 2B). In samples without force application, there was <1 TUNEL-positive cell per microscopic field (p < 0.01; n = 3 animals per group; 3 fields per animal).

View larger version (52K): [in this window] [in a new window]   Fig. 2.   Identification of apoptotic cells in mechanically active environments. Detection of DNA strand breaks in nuclei of force-induced Rat-2 cells. TUNEL assay using terminal deoxynucleotidyl transferase (panel 1) shows that 24 h after force application, high proportions of cells stain with TUNEL, indicating the presence of fragmented DNA. TUNEL-positive cells show bright, co-localized staining with PI (panel 2). DAPI staining shows nuclei of some cells that exhibit nuclear condensation. *, magnetite beads superimposed on cell. As positive control, cells treated with staurosporine (1 µM, for 24 h) show TUNEL-stained nuclei. Magnifications are ×400 (panels 1-3) and ×250 (panel 4). B, left panel, drawing of in vivo force model to illustrate application of tensile forces on fibroblasts in periodontal ligament and location of photomicrographs. Middle panel, low magnification image showing periodontal ligament (P) and surrounding tissues: tooth (T), alveolar bone (AB), and marrow elements (M). The higher magnification image shows apoptotic periodontal ligament cells (arrows) after 3-day force stimulation with latex rubber separator. The presence of apoptotic cells was determined by TUNEL assay. Magnifications are ×250 and ×400. C, immunohistochemical localization of active caspase-3 in force-treated Rat-2 cells. Panels 1 and 2, after a 2-h force application, a small proportion (~5%) of cells stained positively (green) for anti-active caspase-3. The inset (*) shows cells without force application in which no caspase-3 stained cells were detected. Panels 3 and 4, cells treated with staurosporine (1 µM for 24 h) show positive staining for active caspase-3 and nuclei. Magnification is ×250.

We also used staining for active caspase-3 to evaluate late stage effectors of apoptosis in cells following force application in vitro. After 2-h or overnight force application, Rat-2 cells showed positive staining for active caspase-3 (Fig. 2C, 1). Control cells without force application were not stained (Fig. 2, inset). Control cells pretreated with staurosporine for 4 h showed staining for active caspase-3 (Fig. 2C, 3 and 4). After 4 h of treatment with staurosporine and co-incubation with DEVD-CHO, a specific caspase-3 inhibitor, there were fewer PI-permeable cells compared with control cultures without the caspase-3 inhibitor (Fig. 3A). Accordingly, we considered that applied force may induce cell death through a caspase-3-dependent pathway. Rat-2 cells were incubated with the caspase-3 inhibitor DEVD-CHO during force application, and the detached, PI-permeable cells were counted (Fig. 3B). We found that force increased the numbers of floating cells by 7-fold. When cells were treated with the caspase-3 inhibitor simultaneously with force application, the numbers of PI-positive floating cells were reduced nearly 3-fold (p < 0.01). However, there was still a 3-fold higher number of floating cells in the cultures treated with force and inhibitor compared with inhibitor without force. Evidently, the force-induced increase of cell death is not completely blocked by DEVD-CHO, indicating either incomplete blockage of caspase-3 activation or the existence of other, non-caspase-dependent mechanisms of cell death.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Caspase-3 and force-induced cell death. A, cells were treated with staurosporine for 4 h and either co-incubated with the caspase-3 inhibitor DEVD-CHO (panels 1 and 2) or with vehicle (panels 3 and 4). Cells were vitally stained with PI, fixed, and counterstained with DAPI. Note that the caspase-3 inhibitor reduced the numbers of PI-permeable cells. B, cells were incubated with collagen-coated beads, and force was applied (+) or not (). Cells were also treated (+) or not () with the caspase inhibitor DEVD-CHO. After 4 h, the number of PI-permeable floating cells in the medium was counted. Force increased the numbers of floating cells by 7-fold. The caspase-3 inhibitor reduced the numbers of floating cells by ~3-fold (p < 0.01), although despite this, force-treated cells still showed 3-fold higher numbers of floating cells than DEVD-treated cultures without force.

Mechanical Force Induces Loss of Mitochondrial Membrane Potential-- Evaluation of cell death by TUNEL, caspase-3, and morphology gives insight into late stages of apoptosis (34) but does not indicate whether force influences critical events that occur at earlier stages. Indeed, at 2 h after force application, there was only a small difference between control and force-loaded cells when stained with PI (Fig. 1, A and B). Accordingly, we assessed the effect of force on mitochondrial membrane potential as an earlier stage event in the overall apoptotic process. We recognize that caspase activation and dissipation of mitochondrial membrane potential are not necessarily functionally related (34) and used mitochondrial membrane potential simply as a marker of an earlier event in cell death. Flow cytometric analysis of cells subjected to static tensile force applied through collagen-coated beads for 1 h showed that in cells remaining attached to the culture dish after force application, there was only a modest reduction of forward scatter (cell volume), side scatter (cytoplasmic granularity), or the green or red fluorescence attributable to JC-1 staining (Fig. 4). In cultures loaded with collagen beads that were not subjected to force, we found no change whatsoever. In contrast, cells that were detached from the culture dish after force application (floaters) showed reduced side and forward scatter and a very large reduction of red fluorescence after JC-1 staining, an estimate of mitochondrial membrane potential. These results were similar to cells treated with CCCP, which showed a decrease of forward and side scatter and greatly reduced red fluorescence. In cells incubated with BSA beads and loaded with force, we found very few floating cells and no change in JC-1 staining. Following overnight force application with collagen but not BSA beads, flow cytometric measurements of floating cells also showed reduced light scatter and mitochondrial membrane potential (data not shown).

View larger version (124K): [in this window] [in a new window]   Fig. 4.   A, flow cytometry assay to estimate mitochondrial potential in Rat-2 cells. Positive control for cell death used cells incubated in the presence or absence of 1 µmol/µl CCCP for 15 min and stained with JC-1. CCCP caused a reduction of forward and side scatter (top panel) and red fluorescence (middle panels) compared with untreated cells. Three separate experiments were conducted with similar results. B, cells subjected to constant static force (0.48 pN/µm2 cell area) for 1 h show that for cells remaining attached to the culture dish after force application, there was no change in forward scatter, side scatter, and green or red fluorescence. Cells without force application show similar results to untreated control cells in A. Cells detached from the culture dish after force application (floaters) show large reductions of side and forward scatter and 10-fold reduced red fluorescence after JC-1 staining. Three separate experiments were conducted, which produced results similar to that shown here.

Plasma Membrane Depolarization after Mechanical Force Application-- Since irreversible plasma membrane depolarization is an early stage indicator of apoptosis in a wide variety of cell types (17), we used the membrane potential dye DiBAC₄ to quantify effects of mechanical stretching on plasma membrane potential in single, attached cells. DiBAC₄ is an anionic oxonal dye that exhibits increased fluorescence intensity at 520 nm after membrane depolarization. In control experiments that were designed to test the validity of the measurement system, we depolarized cells with solutions containing various concentrations of K⁺. Graded KCl solutions were prepared by adjusting KCl and NaCl concentrations in Dulbecco's modified Eagle's medium to maintain osmolarity at 290 mosmol. The normal KCl and NaCl concentrations in Dulbecco's modified Eagle's medium are 5.4 and 155.5 mM. For graded potassium media, the KCl concentration was set at 5.4 mM (normal), 25, 50, 75, or 100 mM, whereas the NaCl concentration was adjusted to maintain normal osmolarity. Gingival fibroblasts were depolarized with increasing concentrations of extracellular K⁺. In the presence of DiBAC_4, cells without treatment showed minimal photobleaching over time (Fig. 5A). KCl-treated cells showed increases in DiBAC₄ fluorescence (Fig. 5B; 25 mM shown; 50, 75, and 100 mM KCl not shown), indicating that the measurement system did indeed report plasma membrane potential (17). Due to variations of dye loading and cell size between individual cells, the base-line fluorescence of single cells varied considerably. Consequently, we analyzed the data on the basis of time-dependent variations of fluorescence as a function of the base line values. To examine the effect of mechanical stretch on plasma membrane potential, magnetic forces were applied to collagen or BSA-coated beads preincubated with Rat-2 cells as described above. When cells were subjected to a constant force applied through collagen beads (Fig. 5C), there was an initial rapid depolarization. This was followed by a slow reduction of fluorescence, which represents a slow repolarization to control levels and perhaps some photobleaching, although our control experiments without force showed only minimal photobleaching (Fig. 5A). If the magnetic force was removed, the cells repolarized much more quickly and would exhibit a second depolarization when the magnetic force was applied for a second time (Fig. 5D). If BSA-coated beads were incubated with cells, the size of the depolarization was modest (Fig. 5E). In some cells, we noticed that if force was applied and removed and then reapplied, the fluorescence dropped precipitously because of dye loss (Fig. 5F), an indication of a catastrophic increase of membrane permeability. These cells subsequently rounded up and detached from the coverslip.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Plasma membrane potential after force application. Single attached cells were incubated with collagen- or BSA-coated beads, vitally stained with DiBAC4, and fluorescence (photon counts) was measured by microscopic fluorimetry (excitation = 470 nm; emission = 520 nm). A, collagen bead-loaded cell without force application shows very little photobleaching over 300 s. B, cells were depolarized with 25 mM KCl at ~30 s, and fluorescence sharply increases thereafter. C, cell with collagen bead shows a sharp increase of fluorescence after force application (In). Note the decay of fluorescence over time, which is due to fluorescence photobleaching and repolarization. D, cell with collagen bead shows sharp increase (In) followed by decrease after force removal (Out) and rapid return to base-line levels. If force is repeated (In), fluorescence increases again, and the cell repolarizes and photobleaches. E, cell with BSA bead shows only a modest change of fluorescence after force application. F, cell with collagen bead shows sharp increase of fluorescence after force (In) followed by repolarization and photobleaching. After the second force application, there is no response, and dye leaks from the cell as fluorescence approaches 0. This cell subsequently detached from the culture dish.

In view of these findings and earlier data showing that conductance of stretch-activated channels is dependent on the abundance of cortical actin filaments (35), we considered that stabilization of the cell membrane by cortical actin filaments may protect against force-induced increases of membrane permeability and membrane damage. Notably, the actin-binding protein filamin A cross-links actin filaments (18) and reduces the amplitude of calcium fluxes following tensile force application (25). Accordingly, we examined force-induced depolarization in A7 cells (n = 15 cells) and M2T cells (n = 14 cells). In 13 out of 15 A7 cells (expressing filamin A), following a 1-s force application, fluorescence increased initially and returned to base-line levels within 50 s, similar to the results shown in Fig. 5C. In contrast, in 10 out of 14 M2T cells, the fluorescence initially increased and then did not return to base-line levels within 200 s. In the other four M2T cells, the fluorescence increased and then precipitously dropped to near zero levels, similar to what is shown in Fig. 5F, indicating a sudden loss of dye and increase of membrane permeability. These data indicated that filamin A expression in melanoma cells regulates force-induced membrane depolarization and possibly membrane integrity.

Filamin A Expression-- The apparent protection against membrane depolarization provided by filamin A suggested that we should first examine filamin A expression in force-loaded cells and relate filamin A expression levels to PI permeability. We applied force for 4 h through collagen-coated beads to cultures of human gingival fibroblasts, prepared cell suspensions (including floaters), vitally stained the suspensions with PI, washed, fixed, permeabilized, immunostained for filamin A, and analyzed by two-parameter flow cytometry (red emission, PI; green emission, filamin A). Cells were analyzed initially for PI staining, and then the filamin A staining was analyzed in the PI-permeable or the PI-impermeable cell populations. For cultures incubated with collagen-coated beads, there were ~2-fold higher levels of filamin A content in the PI-impermeable cells than in the PI-permeable cells (PI-permeable, 147 ± 24 fluorescence units; PI-impermeable, 321 ± 35 fluorescence units; p < 0.01). We also considered that since filamin A binds to ₁ integrins (20), there was the possibility that the force effect on cells may be due to force-induced reduction of ₁ integrins on the cell surface. However, flow cytometry analysis of surface ₁ integrin expression showed that cells incubated with beads and treated with or without force for 18 h showed no difference of staining intensity (n = 3 samples; force = 37 ± 3 fluorescence units; no force = 36 ± 3 fluorescence units; p > 0.2).

Collectively, these data indicated that filamin A may indeed provide protection against mechanical force-induced death. However, these data did not rule out the possibility that if force selectively activated caspases in the dying cells, the reduced amount of immunostained filamin that we observed in the PI-permeable cells was simply the result of an apoptotic cleavage process and not an indication that filamin A is a protective factor in mechanically induced death. Notably, filamins are caspase substrates in dying human lymphocytes (36). Accordingly, we examined filamin A and -actin content by immunoblotting in Rat-2 cells and human gingival fibroblasts after force treatment (Fig. 6, A and B). Equal amounts of total protein were loaded in each well and separated on 6% polyacrylamide gels. The amount of -actin did not change detectably after all treatments, and consequently, when quantifying the amount of filamin A by densitometry, we adjusted filamin A content as a proportion of -actin. There was no change of filamin A protein content in Rat-2 cells, but there was a 4-fold, time-dependent increase of filamin A in human gingival fibroblasts between 0 and 24 h after force application (Fig. 6, A and B). There was no evidence of the caspase-dependent cleavage of filamin A into 95- and 110-kDa fragments that has been previously reported in T-lymphocytes and Jurkat cells undergoing Fas-dependent death (36).

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Detection of filamin A in force-treated Rat-2 cells and human gingival fibroblasts. A and B, immunoblot analyses of filamin A and -actin protein content in force-treated Rat-2 cells (A) and human gingival fibroblasts (B). Equal amounts of total protein were loaded in each well and separated on 6% polyacrylamide gels. Quantitative analyses of filamin A were performed using densitometry of scanned blots (mean ± S.E.). Numbers below the bars and lanes indicate hours of applied magnetic force; *, bead loading without force incubation. Note that force caused no substantial increase of filamin A in Rat-2 cells but caused a substantial increase over time in the human gingival fibroblasts. For both Rat-2 cells and gingival fibroblasts, there was no appearance of filamin A fragments after force application. C, Rat-2 cells were transfected with 5 µg of filamin A constructs corresponding to the first 333 bp of the coding region of the filamin A gene (antisense and sense) or an expression plasmid for human filamin A but lacking the actin binding domain (No actin binding domain) or an expression plasmid with the human filamin A gene but containing the actin binding domain (see "Experimental Procedures"). After a 2-h force application, cells were vitally stained with PI, fixed, permeabilized, and stained with DAPI, and the percentage of PI-permeable cells was counted. Data are mean percentage of PI-permeable cells ± S.E. (n = 3).

If filamin A is mechanoprotective, we considered that intentional alteration of the expression levels of filamin A should be associated with increased protection against force-induced cell death. Since transfection efficiency of human gingival fibroblasts is much lower than Rat-2 cells (5% compared with 60%), we used the Rat-2 cells for the transfection experiments. The Rat-2 cells also exhibit stable levels of filamin A after force application (Fig. 6, A and B), which simplified the interpretation of these experiments. Rat-2 cells were transiently transfected with various human filamin A constructs, and force experiments were conducted 48 h later. Transfection efficiencies for all constructs as evaluated by green fluorescent protein -actin co-transfections were equivalent (~60%). Since constitutive filamin A expression did not change after force treatment in the Rat-2 cells and force caused only a small increase of PI staining in Rat-2 cells after 2 h (Fig. 1A), we reasoned that if filamin A is mechanoprotective, we should be able to detect changes in the percentage of PI-stained cells at early time periods if we altered the levels of filamin A that participate in mechanoprotective processes. Accordingly, following transfections, after 2 h of force application, cells were vitally stained with PI, fixed, and counterstained with DAPI. Cells with antisense constructs exhibited >3-fold higher percentages of PI-permeable cells than cells transfected with the sense constructs (p < 0.001), which were in turn not significantly different from mock-transfected control cells (Fig. 6C; p > 0.2). The mock-transfected control cells exhibited percentages of dead cells that were very similar to the nontransfected cells shown in Fig. 1A. The mechanoprotective effect was due in large part to the ability of filamin A to cross-link actin filaments, since there were 3-fold higher proportions of PI-permeable cells in cultures that were transfected with a construct in which the actin-binding domain of human filamin A was deleted (p < 0.001). Replacement of the actin-binding domain in the construct restored the proportion of PI-permeable cells to control levels.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The main findings of this study are that tensile forces applied through the collagen receptor can induce a time- and amplitude-dependent increase of apoptosis in fibroblasts and that the expression of filamin A protects against force-induced cell depolarization and death. Previous in vivo studies of high amplitude stretching of myocytes (5), chondrocytes (10), and osteoclasts (6) have shown increased proportions of apoptotic cells after force application, but the force levels, direction, and extracellular matrix attachment proteins required for increased cell death were not defined. Here we show that perpendicular tensile forces applied to collagen- but not BSA-coated beads can rapidly induce apoptosis in fibroblasts as demonstrated by such classical markers of programmed cell death as loss of cytoplasmic structure, increased PI staining, caspase-3 activation, nuclear shrinkage, reduced flow cytometric side and forward scatter, DNA strand breaks, and reduced mitochondrial membrane potential. Most of these features were seen in both attached and floating cells, but we noticed that, in particular, reduced mitochondrial membrane potential was most readily seen in the floating cells. These floating cells may have undergone anoikis, a type of apoptotic cell death involving deprivation of integrin receptor ligation (37) in which the applied physical force caused cell detachment followed by death. However, whether cells remained attached or were detached did not seem to affect the outcome of force-induced death, since all of the classical signs of apoptosis were seen in both floating and attached cells (e.g. DNA strand breaks by TUNEL staining, nuclear shrinkage, ultrastructural changes).

All of the methods for characterizing force-induced apoptosis used here are markers for relatively late stages of the apoptotic process (34), including activation of caspase-3. Notably, inhibition of caspase-3 by DEVD-CHO (38) reduced cell death 3-fold compared with vehicle controls, but there was still a 3-fold higher rate of cell death than in those cells that were not subjected to force. Evidently, there are earlier, rate-determining steps that mediate force-induced apoptosis well in advance of caspase-3 activation. This observation and previous work showing that irreversible cell depolarization may provide insight into much earlier stages in the apoptotic process (17) suggested to us that tensile forces acting on matrix receptors on the cell membrane may cause irreversible cell depolarization, which in turn leads to detachment or death. Indeed, measurements of plasma membrane potential with DiBAC4 showed that in healthy cells after stimulation with either chronic or short term tensile force applications, there was an initially rapid depolarization followed by a return of membrane potential to base-line levels. In contrast, cells that failed to repolarize after force application or that showed dye loss, subsequently exhibited shrinkage and detachment from the culture dish. Consequently, in agreement with Bortner et al. (39), we considered that loss of K⁺ gradients, membrane depolarization, and failure to repolarize could be early events in the initiation of stretch-induced cell death.

In view of the ability of cortical actin filaments to regulate stretch-induced ion channel activity (35), we evaluated the impact of filamin A on stretch-induced membrane depolarization. Filamin A is an actin-binding protein that stiffens the cell membrane (19, 20) by virtue of its ability to cross-link adjacent actin filaments and facilitate the formation of orthogonal filament networks (18). As anticipated, filamin A expression in melanoma cells substantially reduced the deleterious effects of stretch-induced membrane depolarization and also was associated with reduced cell death in fibroblast populations with various levels of constitutive filamin A expression. Previous data have also shown that stretch-induced calcium entry and cell death are increased in melanoma cells without filamin A expression (25). This associative data was shown here to depend on the level of filamin A expression (by antisense experiments) and most notably on the actin binding domain of filamin A, since deletion of the actin binding domain of filamin A from the transfected expression plasmid caused increases of force-induced cell death. In summary, to our knowledge, this is the first report on stretch-induced loss of membrane potential linked to apoptosis and cell death. Moreover, in the context of constitutive mechanoprotective systems, we show how an actin binding protein that alters the rheological properties of the cell membrane can also prevent membrane depolarization and cell death. While the mechanism by which filamin A dampens membrane injury is apparently actin-dependent, it should be noted that filamin A also binds to several other membrane-associated proteins (20), and consequently its protective role may not be mediated solely through actin filaments. In conclusion, basic knowledge of the pathological and protective mechanisms of force-loaded biological systems has particular relevance to skeletal (9) and cardiovascular tissues (5) in which cell death can have irreversible consequences for tissue remodeling and homeostasis.

	ACKNOWLEDGEMENTS

We thank Wilson Lee for flow cytometry, Steven Doyle for electron microscopy, Caroline Chu for preparation of the manuscript, Andras Kapus for very helpful comments, and Honghong Chen for assistance with immunostaining.

	FOOTNOTES

^* This work was supported by Canadian Institutes of Health Research (CIHR) operating, maintenance, and group grants (to C. M.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ Supported by the Academy of Finland and by personnel support from the Canadian Arthritis Network.

Supported by a CIHR fellowship.

^§§ To whom correspondence should be addressed: Rm. 244, Fitzgerald Bldg., University of Toronto, 150 College St., Toronto, Ontario M5S 3E2, Canada. Tel.: 416-978-1258; Fax: 416-978-5956; E-mail: christopher.mcculloch@utoronto.ca.

Published, JBC Papers in Press, March 21, 2002, DOI 10.1074/jbc.M200715200

	ABBREVIATIONS

The abbreviations used are: MEM, minimal essential medium; PBS, phosphate-buffered saline; BSA, bovine serum albumin; pN, piconewton(s); PI, propidium iodide; DAPI, 4',6-diamidino-2-phenylindole; CCCP, carbonyl cyanide m-chlorophenylhydrazone; DEVD-CHO, acetyl-Asp-Glu-Val-Asp-aldehyde; TUNEL, terminal deoxynucleotidyl transferase nick end labeling.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Kwon, E. Hanna, D. Lorang, M. He, J. S. Quick, A. Adem, C. Stevenson, J.-Y. Chung, S. M. Hewitt, E. Zudaire, et al. Functional Characterization of Filamin A Interacting Protein 1-Like, a Novel Candidate for Antivascular Cancer Therapy Cancer Res., September 15, 2008; 68(18): 7332 - 7341. [Abstract] [Full Text] [PDF]

				B. C. Mak, Q. Wang, C. Laschinger, W. Lee, D. Ron, H. P. Harding, R. J. Kaufman, D. Scheuner, R. C. Austin, and C. A. McCulloch Novel Function of PERK as a Mediator of Force-induced Apoptosis J. Biol. Chem., August 22, 2008; 283(34): 23462 - 23472. [Abstract] [Full Text] [PDF]

</

				S. Na, O. Collin, F. Chowdhury, B. Tay, M. Ouyang, Y. Wang, and N. Wang Rapid signal transduction in living cells is a unique feature of mechanotransduction PNAS, May 6, 2008; 105(18): 6626 - 6631. [Abstract] [Full Text] [PDF]