Functional dissection of human protease μ-calpain in cell migration using RNAi

FEBS Letters 580 (2006) 3246–3256

Functional dissection of human protease l-calpain in cell migration using RNAi Meiqun Wua,d, Zhenbao Yua, Jinjiang Fana, Antoine Caronb, Malcolm Whitewayc,d, Shi-Hsiang Shena,e,* a

Mammalian Cell Genetics Group, Biotechnology Research Institute, National Research Council of Canada, Montreal, Que., Canada H4P 2R2 b Genomics and Gene Therapy Vectors Group, Biotechnology Research Institute, National Research Council of Canada, Montreal, Que., Canada H4P 2R2 c Genetics Group, Biotechnology Research Institute, National Research Council of Canada, Montreal, Que., Canada H4P 2R2 d Department of Biology, McGill University, Montreal, Que., Canada H3A 1B1 e Department of Medicine, McGill University, Montreal, Que., Canada H3G 1A4 Received 17 March 2006; revised 28 April 2006; accepted 2 May 2006 Available online 6 May 2006 Edited by Felix Wieland

Abstract Calpains are a family of calcium-dependent cysteine proteases involved in a variety of cellular functions. Two isoforms, m-calpain and l-calpain, have been implicated in cell migration. However, since conventional inhibitors used for the studies of the functions of these enzymes lack specificity, the individual physiological function and biochemical mechanism of these two isoforms, especially l-calpain, are not clear. In contrast, RNA interference has the potential to allow a sequencespecific destruction of target RNA for functional assay of gene of interest. In the present study, we found that small interfering RNAs-mediated knockdown of l-calpain expression in MCF-7 cells that do not express m-Calpain led to a reduction of cell migration. This isoform-specific function of l-calpain was further confirmed by the rescue experiment as overexpression of l-calpain but not m-calpain could restore the cell migration rate. Knockdown of l-calpain also altered cell morphology with increased filopodial projections and a highly elongated tail that seemed to prevent cell spreading and migration with reduced rear detachment ability. Furthermore, knockdown of l-calpain decreased the proteolytic products of filamin and talin, which were specifically rescued by overexpression of l-calpain but not m-calpain, suggesting that their proteolysis could be one of the key mechanisms by which l-calpain regulates cell migration.  2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Calpain; RNA interference; Cell migration; Cleavage; Cytoskeleton

1. Introduction In multicellular organisms, cell migration is critical to many normal and pathophysiological processes, such as embryonic development, wound healing and tumor cell invasion and metastasis. Cell migration consists of multiple of steps, including extension of protrusions, formation of new cell-matrix adhesions, contraction of cell body, and tail detachment [1]. Each of these biophysical processes is controlled coordinately

* Corresponding author. Fax: +1 514 496 6319. E-mail address: [email protected] (S.-H. Shen).

by multiple biochemical signaling cascades. Cleavage of signaling and cytoskeletal proteins by proteases plays an important role in the regulation of such cascades [2]. The calpains are a large family of calcium-dependent cysteine proteases, which consists of two well-conserved and characterized proteins, l-calpain (calpain 1, CAPN1) and m-calpain (calpain2, CAPN2), and at least 10 additional mammalian calpains [3]. Calpains cleave several components of focal adhesions, such as talin, paxillin and focal adhesion kinase, resulting in disassembly of these complexes, remolding of cytoskeleton, loss of substrate adhesion and increased cell migration [4]. Calpain family members have been also implicated in various other biological and pathological phenomena, such as apoptosis, myoblast fusion, tumor metastasis, ischemic and traumatic brain injury, Alzheimer’s disease, muscular dystrophy, and recently, type 2 diabetes [5–13]. Although specific calpain isoforms dictate the specificity of some of these responses, l-calpain and m-calpain, the two ubiquitous isoforms, are considered to play a vital role in most of these biological processes. Both l-calpain and m-calpain are heterodimers composed of a large 80 kDa catalytic subunit and a common small 30 kDa regulatory subunit called calpain 4. They share 66% sequence similarity but differ in the concentration of calcium required for activation in vitro. Millimolar level of calcium is required for half-maximal activation for m-calpain in vitro, whereas micromolar amounts of calcium is enough for l-calpain [3]. Up to now, no substrate specificity between these two isoforms has been found. One major hinder is the lack of isoform-specific inhibitors for calpains [14–16]. Inhibition of calpain activity by different pharmacological inhibitors, and calpastatin, the endogenous inhibitor of calpains argues for the isoform-specific involvement of calpains. Furthermore, fibroblasts from calpain 4 gene knockout (calpain 4/) mice displayed decreased migration rates and loss of cytoskeletal stress fibers, without dissection of particular contribution of two calpain isoforms to the observed phenotypes [17]. While many functional studies are based on combined effects of l-calpain and m-calpain, the specific role of calpain isoforms in focal adhesion dissociation and cell migration remains poorly understood. Because l-calpain and m-calpains seem to have different distribution patterns in multiple subcellular locations [18], they might contribute to cell migration in their distinct extents. Therefore, it is necessary to understand the role of each

0014-5793/$32.00  2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2006.05.003

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

calpain in cell motility, especially l-calpain, whose requirement of Ca2+ concentration for activation by its autolyzed active form is much closer to the physiological condition than that of m-calpain on the basis of in vitro studies [3]. RNA interference (RNAi) has a great potential to distinguish distinct functions of each member in a closely related gene family or to selectively target a mutant gene, especially to study the functions of a particular isoform [19,20]. Generally, many eukaryotic genes are expressed as multiple isoforms through the differential utilization of transcription/translation initiation sites or alternative splicing. The conventional method to study an isoform-specific function is either to express the isoform in a clean background without the presence of other members, or to use an isoform-specific inhibitor. However, these methods are time-consuming and generally unavailable. Taking advantage of the RNAi technology, human adaptor protein ShcA has been investigated in an isoform-specific manner through suppression of other non-target isoforms using common small interfering RNAs (siRNA) [21]. In the very similar way, five isoforms of vascular endothelial growth factor have also been successfully studied [19]. RNAi strategy provides a novel tool for studying the functions of various gene isoforms and may contribute to exploiting human isoform-specific functions of calpains in cell migration. Recently, Franco et al. utilized siRNAs to specifically knock down the expression of l-calpain and m-calpain, and found that only m-calpain but not l-calpain is involved in the regulation of membrane protrusion and adhesion dynamics of fibroblasts [22,23]. Because of the different expression patterns and different subcellular locations of l-calpain and m-calpain, we reasoned that l-calpain might contribute to cell migration in other cell types although its function is not required in some fibroblasts as reported by Franco et al. [22,23]. In this study, we dissected the function of l-calpain in cell migration using RNAi. We found that knockdown of l-calpain in MCF-7 human breast cancer cells resulted in the reduction of cell mobility, change of cell morphology and increased stability of cytoskeletal proteins.

2. Materials and methods 2.1. Materials and cell lines Mouse anti-l-calpain monoclonal antibody was from Affinity BioReagents (Golden, CO). Rabbit anti-m-calpain, mouse anti-Vinculin (clone VIN-11-5), mouse anti-Talin (clone 8D4) and mouse anti-Actin (clone AC-74) antibodies were purchased from Sigma (St. Louis, MO). Mouse anti-Filamin A monoclonal antibodies (clones TI10 and PM6/ 317) were obtained from Chemicon (Temecula, CA). Mouse and rabbit anti-IgG-horseradish peroxidase conjugated polyclonal antibodies were obtained from Bio-Rad (Hercules, CA). Nitrocellulose membrane Hybond-ECL was purchased from Amersham Biosciences (Little Chalfont, Buckinghamshire). ECL chemiluminescence reagent was from Perkin–Elmer Life Science Inc (Boston, MA). Protease inhibitor cocktail without EDTA was purchased from Roche Diagnostics (Indianapolis, IN). Fibronectin of human plasma was purchased from Sigma. MCF-7 cells (Human breast cancer cells) and HeLa cells (Human cervical cancer cell line) were from American Type Culture Collection and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum at 37 C in a 5% CO2 atmosphere. 2.2. Plasmid construction for RNA interference The vector-based RNAi constructs were built in the plasmid pSilenceTM 3.1-H1 (Ambion, Austin, TX). Four siRNAs (Fig. 1A) were designed using the siRNA target finder program from the Ambion Inc. web page (www.ambion.com/techlib/misc/siRNA_finder.html) and compared to the appropriate human genome database using

3247 BLAST to eliminate from any target sequences with more than 16 contiguous base pairs of homology to other coding sequences. SiRNAs were then screened based on internal stability profile [24,25]. Plasmid vectors encoding a small hairpin RNA (shRNA) with two 19-mer-reverse and complementary stems separated by a 9-mer loop were constructed using pSilenceTM siRNA Construction Kit (Ambion, Austin, TX) according to the manufacturer’s instructions. 2.3. Plasmid construction for protein expression in mammalian cells Full-length l-calpain cDNA was amplified from a human brain cDNA library by PCR using pfu DNA polymerase. The PCR product was digested with NheI and XhoI and then cloned into the mammalian expression vector pcDNA3.1 (Invitrogen) at the same sites. A silent mutant of l-calpain carrying six mutant sites within the siRNA targeting sequence (Fig. 3A) was generated by PCR-based site-directed mutagenesis to create plasmid pcDNA3.1/l-calpain-re. 2.4. Transfection and immunoblot analysis HeLa cells and MCF-7 cells were transfected using Lipofectamine 2000 (Life Technologies, Inc.) as described in the manufacturer’s protocol. Stable cell lines were isolated by selection with 300 lg/ml of hygromycin for three weeks after transfection. For immunoblot analysis, monolayer cells were washed twice with cold phosphate-buffered saline (PBS), and lysed in lysis buffer (50 mM HEPES, pH 7.4, 150 mM sodium chloride and 1% Triton X-100) containing EDTA-free protease inhibitor cocktail. The samples were centrifuged at 13 000 rpm for 5 min at 4 C to remove insoluble materials. Total protein was quantified using the Bio-Rad Protein Assay kit with bovine serum albumin as standard protein. The proteins were blotted onto nitrocellulose membranes after separation by 7.5–10% SDS–PAGE. The membranes were blocked overnight with 5% milk in Tris-buffered saline (TBS) and incubated for 2 h at room temperature with primary antibody. After washing four times with TBS containing 0.05% Tween-20 (TBS-T), the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h and then washed four times with TBST. The blots were developed using the ECL chemiluminescence reagent kit according to the manufacturer’s instructions. 2.5. RNA extraction and Northern blot analysis Total cellular RNA was extracted using the RNeasy Midi Kit (Qiagen). Thirty micrograms of total RNA from each cell line was electrophoresed in a 1.2% agarose/formaldehyde gel, transferred to a nylon membrane by upward capillary transfer, and crosslinked by a UV Stratalinker 2400 (Stratagene). The membrane was hybridized overnight at 42 with a cDNA probe labeled with 32P by random priming using Ready-To-GoTM DNA Labeling Beads (Amersham Biosciences) in ULTRAhybTM Hybridization Buffer (Ambion) according to the manufacturer’s specifications. The membrane was washed twice in 2· SSC–0.1% SDS for 5 min and twice in 0.1· SSC–0.1% SDS for 15 min at 42 C and then autoradiographed. 2.6. Casein zymography l-Calpain activity in cell extracts was analyzed by casein zymography in a standard Tris-glycine system [26]. Briefly, cultured cells were washed and scraped into lysis buffer (50 mM HEPES, pH 7.6, 150 mM sodium chloride, 10% glycerol, 1% Triton X-100, 5 mM EDTA, 10 mM b-mercaptoethanol, 100 mM phenylmethylsulfonyl fluoride and 10 lg/ml of leupeptin). After centrifugation at 12 000 · g for 5 min at 4 C to remove any insoluble material, 30 lg of protein was run on a non-denaturing acrylamide gel containing casein (0.2%, w/v) at 125 V for 3 h. The gel was then removed and incubated overnight at room temperature with 5 mM Ca2+ and 10 mM b-mercaptoethanol in 25 mM Tris–HCl (pH 7.6) buffer, followed by fixation and staining with Coomassie brilliant blue R250. As a control, an identical gel was treated with incubation buffer containing 5 mM EDTA in place of Ca2+. Protease activity was detected as a clear band in the dark blue background after destaining the gel. 2.7. Wound-healing assay The ability of cells to migrate in monolayer cultures was assessed by a wound-healing assay. Prior to the assay, 6-well dishes were precoated with 1.3 lg/ml of fibronectin in PBS overnight at 4 C. Cells were seeded and cultured to 90% confluency. On the day of assay, a

C F7 M

A 54 9

H el a

C os -7

29 3A

A

H S57 8T B T54 9

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256 N C I/A D R -R M ES D A -M B 23 1 T47 D

3248

Anti - µ-Calpain Anti - m- Calpain Anti - Actin µ-Calpain siRNAs

B

siRNA1 : AAGACCTATGGCATCAAGTGG siRNA2 : AAGAAGTTGGTGAAGGGCCAT siRNA3 : AAGAGGAGATTGACGAGAACT siRNA4 : AAGAAGCTGTACGAGCTCATC

C

nM

siRNA1 10 40

siRNA2 10 40

siRNA4 Control 0 10 40

siRNA3 10 40

Anti-µ-Calpain

D

C

on

tr

ol

si

R

N

A

si R N A µC al pa in

no

si R N A

Anti-Actin

Anti - µ -Calpain

Anti - Actin E

µ-Calpain

GAPDH

F

µ - Calpain

Fig. 1. Screening of cell lines and functional siRNAs and knockdown of human l-calpain. (A) Western blot analysis of l-calpain and m-calpain expression in human cell lines. (B) Sequences of siRNAs used to knockdown human l-calpain. (C) The siRNAs synthesized in vitro by T7 polymerase (Ambion Inc.) were transiently transfected into HeLa cells. Forty-eight hours after transfection, the cells were lysed and whole cell lysates were subjected to Western blot analysis as described under Section 2. (D) MCF-7 stable cell lines were generated as described under Section 2. Total proteins extracted from wild-type, l-calpain siRNA-expressing or control siRNA-expressing MCF-7 stable cell lines were subjected to Western blot analysis. (E) Northern blot analysis of total RNAs extracted from wild-type, l-calpain siRNA-expressing or control siRNA-expressing MCF-7 stable cell lines. The DNA probes for l-calpain and GAPDH control were prepared as described under Section 2. (F) Casein zymogram. Thirty micrograms of cell lysates from wild-type, l-calpain siRNA-expressing or control siRNA-expressing MCF-7 stable cell lines were run on a native acrylamide gel containing casein, followed by incubation, fixation and staining as described under Section 2.

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256 line was drawn with a marker on the bottom of the dish. Two separate wounds were scratched with a sterile pipette tip through the cells moving perpendicular to the line drawn. The cells were then rinsed once with PBS to remove any cellular debris and replaced with new medium containing 10% of FBS. Photographs were taken just above and just below each line marker immediately after wounding and at intervals up to 36 h. The experiments were repeated in duplicate wells at least three times. 2.8. Cell growth assay Prior to seeding, 100-mm plates were pre-coated with 1.3 lg/ml of fibronectin in PBS overnight at 4 C. Cells were plated at a density of 8 · 104 cells per plate. At indicated times, cells were trypsinized, stained with trypan blue and counted by a hemocytometer. Three counts were taken for each sample and two independent experiments were preformed. The average was applied for calculating the cell number. 2.9. Time lapse videomicroscopy Cells were plated at 15% density onto a glass Delta T dish (0.15 mm; Bioptechs, Inc., Butler, PA) coated with 1.3 lg/ml of fibronectin and allowed to culture overnight at 37 C under 5% CO2. The dish was partitioned into three compartments for testing three different conditions. On the day of the assay, the cell dish was placed directly onto a heated stage after replacing the medium with CO2-independent L15 medium.

3249 A Leica DMIRB inverted microscope with motorized Ludl hardware was used to capture individual focused images for 10 h at 10-min intervals on a CCD camera (Qimaging) attached to a PowerMac G5 computer (Apple, Cuperttino, CA). Image capture and processing were done using Openlab Software Version 3 (Improvision). Cell motility was assessed by drawing a rectangular region of interest (ROI) encompassing each cell at time 0. The time-lapse movie for each stage position was played and each cell was scored for motility according to an arbitrary criterion: if most of a cell’s area exited the ROI at any time, the cell was scored as motile.

3. Results 3.1. Knockdown of l-calpain reduces cell migration To study isoform-specific function of l-calpain, we screened cell lines which specifically express l-calpain. We found that MCF-7 cells express high level of l-calpain but not m-calpain (Fig. 1A). We next used siRNAs to disrupt l-calpain expression. Specifically designed siRNAs (Fig. 1B) were screened by thermodynamic profiling [24,25] and verified by transient transfection into HeLa cells followed by Western blot analysis (Fig. 1C). The results showed that two of these siRNAs, siR-

Fig. 2. Knockdown of l-calpain reduces cell migration. (A) l-calpain siRNA-expressing stable cell line and control siRNA-expressing stable cell line were seeded on six-well tissue culture dishes pre-coated with 1.3 lg/ml of fibronectin and cultured to confluent cell monolayers, which were then wounded as described under Section 2 and photographed at the indicated intervals. (B) The wound areas were measured and calculated and the results shown are mean ± S.E. from three independent experiments. Statistical analysis was performed by Student’s t test (P = 0.086 for 18 h, P < 0.005 for 36 h). (C) The movement of l-calpain siRNA-expressing stable MCF-7 cells and control siRNA-expressing stable MCF-7 cells was measured by quantitative time lapse videomicroscopy as described under Section 2. The results shown are mean ± S.E. of 60 cells from two independent experiments. Statistical analysis was performed by Student’s t test (P < 0.001).

3250

NA1 and siRNA2, could knock down approximately 60% of l-calpain expression. siRNA2 displayed a lower internal strand stability (approximately 6.2 kcal/mol) at the 5 0 -antisense (5 0 -AS) terminus, while siRNA1 was enriched with nucleotides that had lower internal strand stability (approximately 7.6 kcal/mol) at position 9–14 (counting from the 5 0 -AS end) [24,25]. siRNA3 and siRNA4 did not meet these two criteria and had no effect on silencing l-calpain. siRNA2 was selected for the subsequent study and named l-calpain siRNA, whereas non-functional siRNA4 was used as negative control siRNA. Previous research indicated that l-calpain is a long-lived protein with a metabolic half-life of approximately 5 days [27]. In order to maximize the knockdown efficiency for functional assays of l-calpain on cell migration, stable cell lines expressing siRNAs were made with MCF-7 cells. Western blot analysis revealed that l-calpain expression in the MCF-7 cells stably transfected with the l-calpain siRNA-expressing plas-

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

mid was inhibited up to 70–80% as compared both to cells transfected with the control siRNA-expressing plasmid, and to wild-type MCF-7 cells (Fig. 1D). Northern blot analysis further verified the disruption of l-calpain mRNA (Fig. 1E). Casein zymography demonstrated that l-calpain activity was significantly reduced in the l-calpain siRNA-expressing stable cell lines as compared to control siRNA-expressing stable cell lines and wild-type MCF-7 cells (Fig. 1F). Thus, siRNA-mediated silencing of l-calpain leads to substantial inhibition of its expression and activity in the MCF-7 stable cell line. To determine whether selective silencing of l-calpain is sufficient to affect normal cell migration, we carried out a woundhealing assay and time-lapse videomicroscopy of the MCF-7 cells plated on 1.3 lg/ml of fibronectin-coated surface that confers an intermediate cell-substratum adhesiveness maximizing cell migration [28,29]. When confluent monolayers of transfected cells were analyzed for their ability to move to the wounded area, control siRNA-expressing cells migrated into

Fig. 3. Rescue of l-calpain expression results in restoration of a normal cell motility and morphology. (A) Six-site silent mutation of nucleotide was made at the siRNA targeting sequence of l-calpain-wild-type (l-calpain-wt) to generate l-calpain-re. (B) l-calpain siRNA-expressing MCF-7 stable cell line was transiently transfected with pcDNA3.1 empty vector and pcDNA3.1/l-calpain-re respectively. Cells were lysed 72 h after transfection and 30 lg of total protein extract was then subjected to Western blot analysis. (C) Casein zymogram. The l-calpain siRNA-expressing stable cell line was transiently transfected with pcDNA3.1 empty vector and pcDNA3.1/l-calpain-re respectively. Seventy-two hours after transfection, 30 lg of cell lysates from each cell line were subjected to a native acrylamide gel containing casein, followed by incubation, fixation and staining as described under Section 2. (D) l-calpain siRNA-expressing MCF-7 stable cells were transiently cotransfected with a GFP-expressing vector (pFERD/GFP) and pcDNA3.1/l-calpain-re, pcDNA3.1/m-calpain or empty vector. Sixteen hours after transfection, the cells were seeded on six-well tissue culture dishes and cultured to confluent cell monolayers, which were then wounded as described under Section 2 and photographed at indicated intervals. The same transfected cells were lysed and subjected to Western bolt analysis 72 h after transfection. (E) The l-calpain siRNA-expressing stable MCF7 cells were cotransfected with a GFP-expressing vector (pFERD/GFP) and pCDNA3.1/l-calpain-re or empty vector. More than 105 cells in each sample were subjected to the motile and static assay using time-lapse videomicroscopy as described under Section 2. Statistical analysis was performed by McNemar’s test (P < 0.05).

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

3251

the open space within the first 18 h and almost closed the wound within 36 h (Fig. 2A). In contrast, most of the calpain siRNA-expressing cells remained at the initial wound margin after 18 h and the wound was only half-closed after 36 h. The average recovery percentage of the wounded portion was measured and calculated. Overall, knockdown of l-calpain resulted in 45–50% of reduced recovery rate of the wounded portion (Fig. 2B). Time-lapse videomicroscopy analysis showed that calpain siRNA-expressing cells moved at a

much lower velocity (12.5 (m/h)) than that of the control siRNA-expressing cells (17.6 (m/h)) (Fig. 2C), which is consistent with the wound-healing assay. 3.2. Rescue of l-calpain expression but not m-calpain expression restores normal cell motility of MCF-7 cells To determine if the phenotypes observed in the l-calpain siRNA-expressing stable MCF-7 cell line is the direct consequence of the knockdown of targeted l-calpain, we carried

Fig. 3 (continued)

3252

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

Fig. 3 (continued)

out the rescue experiment by overexpressing wild-type l-calpain and m-calpain. In order to avoid degradation of the transfected l-calpain by the existing siRNA in the l-calpain siRNA-expressing stable cell line, a six-site silent mutation was made at the siRNA targeting sequence of wild-type l-calpain (l-calpain-wt) to generate l-calpain-rescue (l-calpain-re) plasmid (Fig. 3A). Since the changed codons were selected based on the maximum frequency of codon usage, we anticipated a successful rescue of l-calpain expression. As expected, a highly elevated expression level of l-calpain was achieved by ectopic transfection of l-calpain-re plasmid into the l-calpain siRNA-expressing stable cell line (Fig. 3B). The restoration of l-calpain activity was also verified by casein zymography, showing that the activity was fully recovered and even higher than that in wild-type MCF-7 cells (Fig. 3C). Taken together, rescue of l-calpain creates a normal level of activity and can be further applied to analyze reverted phenotypes. To test whether the revertant could reinstate l-calpain-related migration, the l-calpain siRNA-expressing MCF-7 stable cells were transfected with l-calpain-re-expressing vector, m-calpain-expressing vector and empty vector respectively, and then replated on fibronectin at low density to facilitate cell spreading and migration. An increase in the migration rate as the direct consequence of l-calpain re-expression was demonstrated by the wound-healing assay when l-calpain-re was co-transfected with pFRED/GFP into confluent l-calpain siRNA-expressing stable cells (Fig. 3D). However, expression of m-calpain did not restore the l-calpain function for the cell migration (Fig. 3D). The expression of the transfected l-calpain and m-calpain was confirmed by Western blot (Fig. 3D). The motile/static assay by time lapse videomicroscopy also demonstrated 75% increase of migration rate in l-calpain-re-transfected cells compared to the control (Fig. 3E). Therefore, MCF-7 cells lacking l-calpain show a notable reduction in cell migration and translocation, whereas the restored expression of l-calpain but not m-calpain establishes its activity. 3.3. Knockdown of l-calpain alters cell morphology In order to determine if the observed phenotype might be caused by different cell growth rates, we therefore counted cell numbers at individual time points under the same culture condition as that in the wound-healing assay. The result showed that the growth rates of these two stable cells were similar

(Fig. 4A). We next determined if the knockdown of l-calpain affects cell morphology that may be linked to the reduction of cell mobility. MCF-7 cells expressing the control siRNA were mostly cuboidal in shape and very few of them had some background with irregular shapes including short protrusions (Fig. 4B). However, compared to the control, 6-fold more of the l-calpain siRNA-expressing MCF-7 stable cells displayed filopodial projections and a highly elongated tail (Fig. 4B and C) that seemed to prevent cell spreading and migration through a reduced rear detachment ability. More interestingly, recovery of expression of l-calpain by transfecting the l-calpain siRNA-expressing MCF-7 stable cell line with the l-calpain-re-expressing vector restored the cell morphology to that of wild-type MCF-7 cells, whereas the phenotype in the control vector transfection essentially maintained that of the l-calpain siRNA-expressing stable cell line (Fig. 4B and C). Together, these results reveal that decreased l-calpain activity is related to altered morphology of MCF-7 cells. 3.4. Knockdown of l-calpain increases the stability of intracellular filamin and talin It has been demonstrated that integrin-mediated signals target calpain to focal adhesion complexes, in which some components have been identified as possible calpain substrates. However, the relevant substrates in vivo for calpain-dependent adhesion turnover, cell rear detachment and subsequent cell movement remain unclear. In an attempt to clarify potential substrates involved in l-calpain-regulated migration, the expression level and proteolysis of several focal adhesion components including filamin A, talin and vinculin were studied. As shown in Fig. 5A, native filamin A (280 kDa) was cleaved giving rise to a 190-kDa fragment (N-terminus) and a 90-kDa fragment (C-terminus) in the control siRNA-expressing stable cell line. Knockdown of l-calpain by siRNA dramatically reduced the production of both the 190-kDa fragment and 90kDa fragment (Fig. 5A). Similarly, intact talin (230 kDa) also appeared to be proteolyzed producing a 190-kDa fragment that was observed in the control siRNA-expressing stable cell line and absent in the l-calpain siRNA-expressing stable cell line (Fig. 5A). Both 190-kDa and 90-kDa cleavage fragments of filamin A and the 190-kDa proteolytic product of talin are identical in size to the previously identified calpain-mediated cleavage products of filamin [30] and talin [31,32] in vitro. In accordance with the full recovery of l-calpain activity, even improved cleavage of filamin and talin were seen in cells rescued by over expression of l-calpain, but not m-calpain (Fig. 5B). This pattern suggested that l-calpain is specifically responsible for cleavage of filamin and talin in MCF-7 cells. In either cell type, the level of vinculin was unchanged (Fig. 5). Thus, these results suggest that knockdown of l-calpain increases the intracellular stability of filamin and talin. Cleavage of these two adhesion components in vivo by l-calpain may define the central mechanism by which l-calpain regulates cell motility.

4. Discussion The biological function of calpains has been extensively studied using pharmacological inhibitors (for review, see Refs. [3–13]). However, these functional studies are based on com-

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

3253

Fig. 4. Knockdown of l-calpain does not affect growth but causes morphological change of MCF-7 cells. (A) l-calpain siRNA-expressing stable cell line and control siRNA-expressing stable cell line were seeded on 100-mm dishes pre-coated with 1.3 lg/ml of fibronectin at a density of 8 · 104 cells/ dish. Viable cells were then counted at the times indicated. (B) l-calpain siRNA-expressing and control siRNA-expressing stable cell lines were plated on 1.3 lg/ml fibronectin at 10% density and incubated at 37 C under 5% CO2. Phase-contrast images were acquired 16 h later using a 32· objective. The arrows highlight extended filopodia and/or elongated tails. (C) The number of the cells showing extended filopodia and/or an elongated tail was counted and the percentage of the total number of cells was calculated. The results shown are mean ± S.E. from at least three independent experiments. Statistical analysis was performed by Student’s t test.

3254

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

Fig. 4 (continued)

bined effects of l-calpain and m-calpain since the calpain inhibitors used inactivate both enzymes [14–16]. RNAi has a great potential to distinguish distinct functions of each member in a closely related gene family. Recently, Franco et al. have knocked-down the expression of l-calpain and m-calpain in mouse embryonic fibroblasts, and found that knock-down lines of m-calpain altered cell membrane protrusion kinetics and adhesion dynamics, but knock-down lines of l-calpain showed normal morphology [23]. Similarly, Honda et al have found that knockdown of m-calpain, but not l-calpain, induced mitotic delay with chromosome misalignment [33]. In this study, using a similar gene knockdown method, we demonstrate the function of l-calpain in cell migration. To avoid the effect of m-calpain on the observation, we selected a cell line, human breast cancer MCF-7 cells, which do not express m-calpain. We found that knockdown of l-calpain expression led to a reduction in cell migration. The isoform-specific function of l-calpain was further confirmed by a rescue experiment as overexpression of l-calpain but not that of m-calpain could restore the cell migration rate. Recently Satish et al. reported that both l-calpain and m-calpain are involved in the regulation of cell migration of keratinocytes, but through a different signaling pathway. M-calpain is activated by EGF receptor signaling whereas l-calpain is activated by interferon-inducible protein 9 [34]. Taken together, these findings suggest that lcalpain and m-calpain have different biological functions. Cell migration can be considered as a cycle of multiple steps, including the forward extension of a protrusion, the stable attachment at the leading edge of the protrusion, the contraction and forward movement of cell body, and the release of adhesion and retraction at the cell rear. We found that l-calpain siRNA-expressing MCF-7 stable cells displayed increased filopodial projections and a highly protrusive morphology. This morphological change can cause either an increase or decrease in cell spreading and migration depending on which migration step is affected. Although the function of calpain as a positive regulator in cell migration has been well demonstrated, the mechanism of cell movement regulated by calpains remains unclear. Huttenlocher et al. reported that inhibition of calpain activity resulted in a change in CHO cell morphology

with many cells showing a highly elongated tail, suggesting that calpain disrupts focal adhesions at the rear of migrating cells [35]. However, Potter et al. found that filopodial and lamellipodial projection formation were severely impaired in NIH 3T3 cells treated with calpain inhibitors, suggesting that calpain activity is required for forward protrusion and spreading during cell migration [36]. It was also reported that bovine aortic-endothelial cell spreading was remarkably disrupted by calpain inhibitors [37]. Taken together, calpains may have multiple actions that control cell migration at different steps. The present study suggests that l-calpain may regulate the migration of MCF-7 cells by disassembling focal complexes at the rear of the cell and releasing tail attachment. Calpains have many substrates in vivo, including a number of focal adhesion complex proteins and cytoskeletal proteins [3]. In contrast to the results observed in mouse embryonic fibroblasts where talin was cleaved by m-calpain but not l-calpain [22,23], we found that talin and filamin cleavage decreased in l-calpain knocked-down MCF-7 cells, and increased again in cells rescued by overexpression of l-calpain, but not m-calpain, suggesting that calpains have distinct substrate specificities in different cells. To our knowledge, this is the first report demonstrating the proteolysis of filamin during calpain-regulated cell locomotion, although reports indicated previously that cleavage of filamin by calpain in vivo is associated with blood coagulation and platelet physiology [38–40]. Based on previous in vitro findings it is suspected that calpains are activated by intracellular Ca2+ fluxes of 5–50 lM for l-calpain and 0.2–1 mM for m-calpain [3]. Moreover, spatial patterns of l-calpain and m-calpain are probably different in multiple subcellular locations [18,41]. It is possible that the activation process and favorable substrates of l-calpain and m-calpain in vivo are different although they cleave most substrates in a similar manner. As a result, both l-calpain and mcalpain play important roles in cell migration, but each of them may function in different cell types and/or under different circumstances, as revealed in the present and previous studies [23,33,34]. Since calpain-regulated cell adhesion and migration have been historically linked to pathophysiological issues in tumor metastasis, wound healing and the immune and inflammatory responses, functional dissection of individual isoforms and specific therapeutic treatments are even more crucial. In addition, many human disorders are caused by overactivation of single isoforms of calpains. For example, lens m-calpain is thought to be over-activated in cataracts leading to crystalline precipitation and, ultimately, lens opacity [15,42]. m-calpain can be targeted for limiting prostate cancer invasion [43]. An excessive amount of m-calpain is also found in arthritic knee joints of mice [44]. l-calpain overexpression is associated with increased malignancy in human renal cell carcinoma [45]. However, the current lack of potent, selective and cellpermeable calpain inhibitors makes it difficult to control these situations. RNAi, which has been proven to be a powerful tool for dissecting the functions of each calpain isoform, may provide us with an attractive therapeutical approach in the treatment of various pathological disorders associated with over-activation of l-calpain and/or m-calpain, or other specific calpain family members. However, the isoform-specific function of l-calpain in cell adhesion was not experimentally demonstrated.

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

3255

Fig. 5. Knockdown of l-calpain increases the stability of intracellular filamin and talin. (A) l-calpain siRNA-expressing MCF-7 stable cell line and control siRNA-expressing stable cell line were seeded on dishes coated with 1.3 lg/ml of fibronectin. After 24 h incubation, cells were processed for Western blot analysis with the antibodies as indicated. (B) l-calpain siRNA-expressing cell line was transiently transfected with pCDNA3.1 empty vector, pcDNA3.1/l-calpain-re and pcDNA3.1/m-calpain respectively. Twenty-four hours after transfection, cells were detached and replated on dishes coated with 1.3 lg/ml of fibronectin. After additional 24 h incubation, cells were processed for Western blot analysis with the antibodies as indicated. Molecular mass (kDa) is shown to the left of the gels.

Acknowledgements: We thank Denis L’Abbe´, Normand Jolicoeur and Yves Fortin for technical supports. This work was supported by grants to S.-H.S. from the PENCE (The Protein Engineering Network of Centers of Excellence).

References [1] Webb, D.J., Parsons, J.T. and Horwitz, A.F. (2002) Adhesion assembly, disassembly and turnover in migrating cells – over and over and over again. Nat. Cell Biol. 4, E97–E100. [2] Glading, A., Lauffenburger, D.A. and Wells, A. (2002) Cutting to the chase: calpain proteases in cell motility. Trends Cell Biol. 12, 46–54. [3] Goll, D.E., Thompson, V.F., Li, H., Wei, W. and Cong, J. (2003) The calpain system. Physiol. Rev. 83, 731–801. [4] Carragher, N.O. and Frame, M.C. (2002) Calpain: a role in cell transformation and migration. Int. J. Biochem. Cell Biol. 34, 1539–1543. [5] Croall, D.E. and DeMartino, G.N. (1991) Calcium-activated neutral protease (calpain) system: structure, function, and regulation. Physiol. Rev. 71, 813–847. [6] Saido, T.C., Sorimachi, H. and Suzuki, K. (1994) Calpain: new perspectives in molecular diversity and physiological–pathological involvement. FASEB J. 8, 814–822. [7] Sorimachi, H., Ishiura, S. and Suzuki, K. (1997) Structure and physiological function of calpains. Biochem. J. 328 (Pt 3), 721– 732. [8] Carafoli, E. and Molinari, M. (1998) Calpain: a protease in search of a function? Biochem. Biophys. Res. Commun. 247, 193–203. [9] Huang, Y. and Wang, K.K. (2001) The calpain family and human disease. Trends Mol. Med. 7, 355–362. [10] Perrin, B.J. and Huttenlocher, A. (2002) Calpain. Int. J. Biochem. Cell Biol. 34, 722–725.

[11] Suzuki, K., Hata, S., Kawabata, Y. and Sorimachi, H. (2004) Structure, activation, and biology of calpain. Diabetes 53 (Suppl 1), S12–S18. [12] Branca, D. (2004) Calpain-related diseases. Biochem. Biophys. Res. Commun. 322, 1098–1104. [13] Bano, D., Young, K.W., Guerin, C.J., Lefeuvre, R., Rothwell, N.J., Naldini, L., Rizzuto, R., Carafoli, E. and Nicotera, P. (2005) Cleavage of the plasma membrane Na+/Ca2+ exchanger in excitotoxicity. Cell 120, 275–285. [14] Wang, K.K. and Yuen, P.W. (1997) Development and therapeutic potential of calpain inhibitors. Adv. Pharmacol. 37, 117–152. [15] Wang, K.K. and Yuen, P.W. (1994) Calpain inhibition: an overview of its therapeutic potential. Trends Pharmacol. Sci. 15, 412–419. [16] Donkor, I.O. (2000) A survey of calpain inhibitors. Curr. Med. Chem. 7, 1171–1188. [17] Dourdin, N., Bhatt, A.K., Dutt, P., Greer, P.A., Arthur, J.S., Elce, J.S. and Huttenlocher, A. (2001) Reduced cell migration and disruption of the actin cytoskeleton in calpain-deficient embryonic fibroblasts. J. Biol. Chem. 276, 48382–48388. [18] Lane, R.D., Allan, D.M. and Mellgren, R.L. (1992) A comparison of the intracellular distribution of mu-calpain, m-calpain, and calpastatin in proliferating human A431 cells. Exp. Cell Res. 203, 5–16. [19] Zhang, L., Yang, N., Mohamed-Hadley, A., Rubin, S.C. and Coukos, G. (2003) Vector-based RNAi, a novel tool for isoformspecific knock-down of VEGF and anti-angiogenesis gene therapy of cancer. Biochem. Biophys. Res. Commun. 303, 1169–1178. [20] Miller, V.M., Gouvion, C.M., Davidson, B.L. and Paulson, H.L. (2004) Targeting Alzheimer’s disease genes with RNA interference: an efficient strategy for silencing mutant alleles. Nucleic Acids Res. 32, 661–668. [21] Kisielow, M., Kleiner, S., Nagasawa, M., Faisal, A. and Nagamine, Y. (2002) Isoform-specific knockdown and expression of adaptor protein ShcA using small interfering RNA. Biochem. J. 363, 1–5.

3256 [22] Franco, S.J., Rodgers, M.A., Perrin, B.J., Han, J., Bennin, D.A., Critchley, D.R. and Huttenlocher, A. (2004) Calpain-mediated proteolysis of talin regulates adhesion dynamics. Nat. Cell Biol. 6, 977–983. [23] Franco, S., Perrin, B. and Huttenlocher, A. (2004) Isoform specific function of calpain 2 in regulating membrane protrusion. Exp. Cell Res. 299, 179–187. [24] Reynolds, A., Leake, D., Boese, Q., Scaringe, S., Marshall, W.S. and Khvorova, A. (2004) Rational siRNA design for RNA interference. Nat. Biotechnol. 22, 326–330. [25] Khvorova, A., Reynolds, A. and Jayasena, S.D. (2003) Functional siRNAs and miRNAs exhibit strand bias. Cell 115, 209– 216. [26] Arthur, J.S. and Mykles, D.L. (2000) Calpain zymography with casein or fluorescein isothiocyanate casein. Methods Mol. Biol. 144, 109–116. [27] Zhang, W., Lane, R.D. and Mellgren, R.L. (1996) The major calpain isozymes are long-lived proteins. Design of an antisense strategy for calpain depletion in cultured cells. J. Biol. Chem. 271, 18825–18830. [28] Palecek, S.P., Loftus, J.C., Ginsberg, M.H., Lauffenburger, D.A. and Horwitz, A.F. (1997) Integrin-ligand binding properties govern cell migration speed through cell-substratum adhesiveness. Nature 385, 537–540. [29] Huttenlocher, A., Ginsberg, M.H. and Horwitz, A.F. (1996) Modulation of cell migration by integrin-mediated cytoskeletal linkages and ligand-binding affinity. J. Cell Biol. 134, 1551– 1562. [30] Fox, J.E., Goll, D.E., Reynolds, C.C. and Phillips, D.R. (1985) Identification of two proteins (actin-binding protein and P235) that are hydrolyzed by endogenous Ca2+-dependent protease during platelet aggregation. J. Biol. Chem. 260, 1060–1066. [31] Schoenwaelder, S.M., Yuan, Y., Cooray, P., Salem, H.H. and Jackson, S.P. (1997) Calpain cleavage of focal adhesion proteins regulates the cytoskeletal attachment of integrin alphaIIbbeta3 (platelet glycoprotein IIb/IIIa) and the cellular retraction of fibrin clots. J. Biol. Chem. 272, 1694–1702. [32] Tranqui, L. and Block, M.R. (1995) Intracellular processing of talin occurs within focal adhesions. Exp. Cell Res. 217, 149–156. [33] Honda, S., Marumoto, T., Hirota, T., Nitta, M., Arima, Y., Ogawa, M. and Saya, H. (2004) Activation of m-calpain is required for chromosome alignment on the metaphase plate during mitosis. J. Biol. Chem. 279, 10615–10623. [34] Satish, L., Blair, H.C., Glading, A. and Wells, A. (2005) Interferon-inducible protein 9 (CXCL11)-induced cell motility in

M. Wu et al. / FEBS Letters 580 (2006) 3246–3256

[35]

[36]

[37] [38]

[39] [40]

[41]

[42]

[43] [44]

[45]

keratinocytes requires calcium flux-dependent activation of mucalpain. Mol. Cell. Biol. 25, 1922–1941. Huttenlocher, A., Palecek, S.P., Lu, Q., Zhang, W., Mellgren, R.L., Lauffenburger, D.A., Ginsberg, M.H. and Horwitz, A.F. (1997) Regulation of cell migration by the calcium-dependent protease calpain. J. Biol. Chem. 272, 32719–32722. Potter, D.A., Tirnauer, J.S., Janssen, R., Croall, D.E., Hughes, C.N., Fiacco, K.A., Mier, J.W., Maki, M. and Herman, I.M. (1998) Calpain regulates actin remodeling during cell spreading. J. Cell Biol. 141, 647–662. Kulkarni, S., Saido, T.C., Suzuki, K. and Fox, J.E. (1999) Calpain mediates integrin-induced signaling at a point upstream of Rho family members. J. Biol. Chem. 274, 21265–21275. Azam, M., Andrabi, S.S., Sahr, K.E., Kamath, L., Kuliopulos, A. and Chishti, A.H. (2001) Disruption of the mouse mu-calpain gene reveals an essential role in platelet function. Mol. Cell. Biol. 21, 2213–2220. Onji, T., Takagi, M. and Shibata, N. (1987) Calpain abolishes the effect of filamin on the actomyosin system in platelets. Biochim. Biophys. Acta 912, 283–286. Verhallen, P.F., Bevers, E.M., Comfurius, P. and Zwaal, R.F. (1987) Correlation between calpain-mediated cytoskeletal degradation and expression of platelet procoagulant activity. A role for the platelet membrane-skeleton in the regulation of membrane lipid asymmetry? Biochim. Biophys. Acta 903, 206–217. Moraczewski, J., Piekarska, E., Bonavaud, S., Wosinska, K., Chazaud, B. and Barlovatz-Meimon, G. (1996) Differential intracellular distribution and activities of mu- and m-calpains during the differentiation of human myogenic cells in culture. C. R. Acad. Sci. III 319, 681–686. Yoshida, H., Murachi, T. and Tsukahara, I. (1984) Limited proteolysis of bovine lens alpha-crystallin by calpain, a Ca2+dependent cysteine proteinase, isolated from the same tissue. Biochim. Biophys. Acta 798, 252–259. Mamoune, A., Luo, J.H., Lauffenburger, D.A. and Wells, A. (2003) Calpain-2 as a target for limiting prostate cancer invasion. Cancer Res. 63, 4632–4640. Szomor, Z., Shimizu, K., Fujimori, Y., Yamamoto, S. and Yamamuro, T. (1995) Appearance of calpain correlates with arthritis and cartilage destruction in collagen induced arthritic knee joints of mice. Ann. Rheum. Dis. 54, 477–483. Braun, C., Engel, M., Seifert, M., Theisinger, B., Seitz, G., Zang, K.D. and Welter, C. (1999) Expression of calpain I messenger RNA in human renal cell carcinoma: correlation with lymph node metastasis and histological type. Int. J. Cancer 84, 6–9.