The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Contents lists available at SciVerse ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

Insulin Receptor Substrate protein 53 kDa (IRSp53) is a negative regulator of myogenic differentiation Ashish Misra 1 , Bhawana George 1 , Rajamuthiah Rajmohan, Neeraj Jain, Ming Hwa Wong, Ravi Kambadur, Thirumaran Thanabalu ∗ School of Biological Sciences, Nanyang Technological University, Singapore 637551, Republic of Singapore

a r t i c l e

i n f o

Article history: Received 2 September 2011 Received in revised form 27 February 2012 Accepted 29 February 2012 Available online 24 March 2012 Keywords: I-BAR WASP Filopodia Vinculin SH3 domain

a b s t r a c t Fusion of mononucleated myoblasts to generate multinucleated myotubes is a critical step in skeletal muscle development. Filopodia, the actin cytoskeleton based membrane protrusions, have been observed early during myoblast fusion, indicating that they could play a direct role in myogenic differentiation. The control of filopodia formation in myoblasts remains poorly understood. Here we show that the expression of IRSp53 (Insulin Receptor Substrate protein 53 kDa), a known regulator of filopodia formation, is down-regulated during differentiation of both mouse primary myoblasts and a mouse myoblast cell line C2C12. Over-expression of IRSp53 in C2C12 cells led to induction of filopodia and decrease in cell adhesion, concomitantly with inhibition of myogenic differentiation. In contrast, knocking down the IRSp53 expression in C2C12 cells led to a small but significant increase in myotube development. The decreased cell adhesion of C2C12 cells over-expressing IRSp53 is correlated with a reduction in the number of vinculin patches in these cells. Mutations in the conserved IMD domain (IRSp53 and MIM (missing in metastasis) homology domain) or SH3 domain of IRSp53 abolished the ability of this protein to inhibit myogenic differentiation and reduce cell adhesion. Over-expression of the IMD domain alone was sufficient to decrease the cell–extracellular matrix adhesion and to inhibit myogenesis in a manner dependent on its function in membrane shaping. Based on our data, we propose that IRSp53 is a negative regulator of myogenic differentiation which correlates with the observed down regulation of IRSp53 expression during myoblast differentiation to myotubes. © 2012 Elsevier Ltd. All rights reserved.

1. Introduction Skeletal muscle formation and regeneration depend on the fusion of mono-nucleated cells to generate multinucleated myotubes that further differentiate into myofibers. Skeletal muscle formation can be divided into a series of steps, namely, formation of muscle precursor cells, myoblast proliferation, cell cycle arrest, fusion of myoblast and the formation of multinucleated myotubes (Andres and Walsh, 1996; Charge and Rudnicki, 2004; Rochlin et al., 2010; Nowak et al., 2009). The fusion of myoblasts requires a series of cellular events, including cell–extracellular matrix (ECM) adhesion, cell migration, cell–cell adhesion and membrane fusion (Knudsen and Horwitz, 1977). The actin cytoskeleton plays an essential role in cell adhesion, cell migration and muscle formation (Nowak et al., 2009; Kim et al., 2007; Le Clainche and Carlier, 2008).

∗ Corresponding author. Tel.: +65 6316 2832; fax: +65 6791 3856. E-mail address: [email protected] (T. Thanabalu). 1 These authors contributed equally to this work. 1357-2725/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2012.02.020

Myogenesis is regulated by a number of growth factors including insulin-like growth factors, IGF-I and IGF-II, which play a vital role in skeletal muscle differentiation and growth (Ren et al., 2008). IRSp53 (Insulin Receptor Substrate protein 53 kDa) was initially identified due to phosphorylation on its tyrosine residues upon stimulation by IGF-I (Yeh et al., 1996) and subsequently identified as a WAVE1 interacting protein in a yeast two-hybrid screen (Miki et al., 2000). The formation of myotubes from myoblasts requires morphological changes in cell shape and migration, which involve deformation of membranes (Swailes et al., 2006; Louis et al., 2008) and IRSp53 has been shown to produce membrane protrusion in animal cells (Suetsugu et al., 2006). It has been suggested that membrane protrusions such as filopodia play a critical role in myogenesis (Chen and Olson, 2001; Nowak et al., 2009). IRSp53 is an adaptor protein with an IMD (IRSp53 and missing in metastasis domain) at the N-terminus followed by a central Cdc42/Rac interactive binding (CRIB) domain and a SH3 domain at the Cterminus (Scita et al., 2008). IMD is also referred to as the I-BAR domain (Inverse-BIN-Amphiphysins-RVS) and it plays a role in Factin bundling and filopodia formation (Yamagishi et al., 2004). It has been suggested that IRSp53 adopts an auto-inhibited conformation, due to the interaction between the N terminus of IRSp53

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

(aa 1–178) and the central region (aa 180–317). Binding of Cdc42 to the CRIB domain of IRSp53 relieves the auto-inhibitory interaction and promotes the formation of filopodia (Krugmann et al., 2001). The SH3 domain of IRSp53 has been shown to bind to a number of proteins, including WAVE1, WAVE2, N-WASP and WIRE (Miki et al., 2000; Lim et al., 2008; Misra et al., 2010). Cell–ECM adhesion is vital for many cellular processes such as proliferation, differentiation, cell shape changes, actin organization and migration (Enomoto et al., 1993; Boettiger et al., 1995; Velleman and McFarland, 2004). Cell–ECM adhesion is mediated mainly by integrins, a family of transmembrane heterodimeric glycoprotein receptors that physically link extracellular matrix (ECM) to the intracellular cytoskeleton at sites known as focal adhesions (FA) (Liu et al., 2011). Vinculin links the actin cytoskeleton to integrins and is often used as a marker for FA (Adams et al., 1998). The purpose of the current study was to determine the role of IRSp53 in myogenic differentiation. We found that the expression of IRSp53 was down regulated during the differentiation of myoblast to myotubes. Knocking down the expression of IRSp53 in C2C12 cells enhanced the myogenic differentiation while overexpression of IRSp53 inhibited myogenic differentiation of C2C12 cells. The over-expression of IRSp53 caused a significant reduction in cell–fibronectin adhesion and inhibited the assembly of focal adhesions containing vinculin. The IMD domain of IRSp53 (IRSp53IMD ) is sufficient to induce filopodia, reduce cell–ECM adhesion and inhibit myogenic differentiation suggesting that the IMD domain of IRSp53 is responsible for negative regulation of myogenic differentiation.

929

2.4. Gel electrophoresis and immunoblotting Cells were lysed using RIPA lysis buffer, the resulting lysate was boiled in SDS-PAGE sample buffer for 5 min, and 30 ␮g of proteins were resolved on 10% SDS-PAGE gel and transferred onto nitrocellulose membrane. The membrane was probed with appropriate primary antibodies anti-MyHC (Bader et al., 1982) (Developmental Studies Hybridoma Bank, IA, USA), anti-MyoD (Becton, Dickinson and Company, USA), anti-IRSp53 (Santa Cruz Biotechnology, CA, USA) and developed using secondary antibody conjugated with horse radish peroxidase (Sigma–Aldrich, USA). All the experiments were carried out in triplicates and densitometry was performed with Image J analysis software (NIH). Band densities were measured as area under curve. Sample intensity was normalized to GAPDH intensity and Student’s t-test was performed to determine statistical significance.

2.5. Measurement of fusion index C2C12 cells undergoing differentiation were fixed and stained with DAPI and anti-MyHC (Bader et al., 1982). We captured six different images from random fields. The fusion index was calculated as the ratio of the nuclei in myotubes to the total number of nuclei in the field. The assays were done in triplicate and the error bars represent the standard deviation of 3 independent experiments.

2.6. Immunofluorescence 2. Materials and methods 2.1. Cell culture and transfection Mouse C2C12 cells were maintained in growth media, GM (DMEM/10% fetal bovine serum) at 37 ◦ C in a 5% CO2 humidified atmosphere. C2C12 cells were microporated using the Neon Transfection System (Invitrogen, CA, USA) according to manufacturer’s instructions. Briefly, 5 × 106 cells in 100 ␮l of resuspension buffer were mixed with 10 ␮g of plasmid DNA and subjected to three 10 ms pulses at 1650 V. We were able to achieve 80–90% transfection efficiency with this method. Transfected cells were incubated for 36 h in GM before analysis or induced to differentiate by switching to differentiation medium (DM) (DMEM/2% horse serum). 2.2. Mouse Primary myoblasts isolation Mouse Primary myoblasts were isolated from newborn normal mice (C57BL6) as described previously (Springer et al., 2002). The hind limb muscles were isolated, minced using scalpel and digested with 0.2% collagenase (Sigma, MO, USA) for 1 h. The cells were harvested by centrifugation and seeded in Ham F-10 media with 20% FBS and 2.5 ng/ml bFGF (basic Fibroblast Growth Factor). The cells were grown to confluence and switched to 2% horse serum (HS) containing DMEM for induction of differentiation.

C2C12 grown on coverslips were probed with appropriate primary antibodies and alexa-488 secondary antibody while the actin cytoskeleton was visualized using alexa-568 Phalloidin (Misra et al., 2010). Fluorescence images were acquired using Olympus IX51 fitted with Cool SNAPHQ camera and analyzed using Metamorph software (Molecular Devices).

2.7. Cell binding and spreading assay Human Fibronectin (Sigma) dissolved in sodium bicarbonate buffer was used to coat the wells of 96-well microtiter plates (Misra et al., 2007). C2C12 cells were labeled with Calcein AM (Invitrogen) and added to fibronectin coated wells (20,000 cells per well) and allowed to adhere for 30 min and the bound cells were quantified using a fluorescent plate reader after washing off the unbound cells. The assays were done in triplicate and the error bars represent the standard deviation of 3 independent experiments (Misra et al., 2007). Spreading assay was carried out by adding the cells to wells coated with fibronectin and imaging them at 10 min intervals. The mean surface area was calculated using MetaMorph software (Molecular Devices, CA, USA). The data is an average of 3 independent experiments with a total of 30 cells quantified for each experiment. Quantification of vinculin patches were carried out as described previously (Benoit et al., 2009). We counted the number of vinculin patches in 20 cells per transfection and from three independent experiments.

2.3. DNA constructs Plasmids expressing mouse specific shRNA (psh-IRSp53) and the corresponding scrambled shRNA (pscr-IRSp53) were constructed in pFIV (open Biosystems, CO, USA). Plasmids expressing IRSp53 (pIRSp53) and its mutants (pIRSp534A , pIRSp532A ) have been described previously (Misra et al., 2010). Plasmid expressing IRSp53IMD (1–250 aa) and IRSp53SH3 (367–521 aa) were generated in the same plasmid.

2.8. Filopodia measurement and analysis Filopodia induction was analyzed as described (Misra et al., 2010). We analyzed 30 fluorescent cells per transfection (membrane projections between 8 and 20 ␮m) (Lim et al., 2008) and three independent transfections were performed for each construct. Values presented in bar charts represent mean ± S.D.

930

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 1. Expression of IRSp53 is down regulated during differentiation. (A) C2C12 cells were grown in growth media to 80% confluence and then switched to differentiation media. Images were acquired at 24 h (day 1), 48 h (day 2) and 72 h (day 3) after changing to DM. (B) (I) Total lysate from C2C12 cells undergoing differentiation was prepared on the days indicated and analyzed by immunoblotting with anti-IRSp53 and anti-GAPDH. (II) Quantification of Western blots using densitometry (*P < 0.05).

2.9. Statistical analysis All experiments were carried out a minimum of three times with similar results. Statistical significance analysis was performed using Student’s t-test and *P < 0.05, **P < 0.01, ***P < 0.001. 3. Results 3.1. IRSp53 expression is down regulated during myogenic differentiation The expression of IRSp53 in differentiating C2C12 myoblast cells was investigated by subjecting confluent C2C12 cells to differentiating media (DM) for three days and isolating protein extracts at the indicated time points (Fig. 1). Two days after switching to DM, the myoblasts lost fibroblast morphology, aligned and fused to form myotubes. The immunoblotting of cell lysates showed that IRSp53 was expressed in C2C12 cells maintained in GM and that the expression of IRSp53 was down-regulated as the C2C12 myoblasts underwent myogenic differentiation (Fig. 1A and B). We also performed similar experiments with freshly isolated mouse primary myoblasts and found that the expression of IRSp53 was also downregulated as the mouse primary myoblast differentiated to form myotubes (Fig. S1). 3.2. Knocking down the expression of IRSp53 leads to the formation of thick myotubes The expression of IRSp53 is down regulated during myogenic differentiation suggesting that IRSp53 might play an inhibitory role in myogenesis. Thus we sought to determine whether artificial depletion of IRSp53 would lead to acceleration in differentiation. In order to determine the contribution of endogenous IRSp53 in myogenesis, we generated psh-IRSp53 (plasmid expressing IRSp53-shRNA) and confirmed the specificity of the shRNA by using reconstitution experiments. We used the scrambled pscrIRSp53 (scrambled sh-IRSp53) as a control (Misra et al., 2010). We microporated C2C12 cells with either psh-IRSp53 or pscr-IRSp53 and transferred the cells to the differentiation medium 36 h after microporation. C2C12 cells were fixed (on the days indicated),

permeabilized and probed with anti-MyHC (myosin heavy chain) antibodies followed by immunofluorescence microscopy to determine the differentiation status of the cells (Fig. 2A). Untransfected C2C12 cells (data not shown) or C2C12 cells transfected with the empty vector (Fig. 3A) differentiated in a similar manner to C2C12 cells transfected with pscr-IRSp53 (Fig. 2A). We calculated the fusion index at day 2 and day 3 by quantifying the number of nuclei in myotubes as a percentage of total number of nuclei. The C2C12 cells transfected with psh-IRSp53 were found to have a higher fusion index than cells transfected with pscr-IRSp53 (Fig. 2B). The ability of psh-IRSp53 to knock down the expression of IRSp53 in C2C12 was confirmed by immunoblot analysis (Fig. 2C). MyoD activates muscle specific genes and is an early marker of myogenic differentiation while MyHC is a late marker for myogenic differentiation. There was a significant increase in MyoD expression (DM2) in IRSp53 knockdown cells compared to cells transfected with scrIRSp53 and the cells transfected with psh-IRSp53 expressed higher levels of MyHC (DM3) compared to cells transfected with pscrIRSp53 (Fig. 2D). 3.3. Over-expression of IRSp53 inhibits myogenic differentiation Knocking down the expression of IRSp53 leads to acceleration in myogenic differentiation (Fig. 2) thus, we wondered if overexpression of IRSp53 could inhibit myogenic differentiation. We microporated C2C12 cells with pVect (control) or the IRSp53expressing plasmid (pIRSp53) and transferred the cells from GM to DM 36 h after microporation to initiate differentiation. C2C12 cells microporated with pVect underwent differentiation and formed myotubes (visible by DM2) whereas the C2C12 cells microporated with IRSp53 expressing plasmid differentiated poorly (Fig. 3A). We monitored myogenic differentiation by MyHC immunofluorescence and by determining the fusion index (Fig. 3B) which showed statistically significant reduction in differentiation of IRSp53 overexpressing cells. Analysis of the expression of MyoD and MyHC by Western blotting showed a reduced expression of both proteins upon over-expression of IRSp53 (Fig. 3D). To compare the expression levels of endogenous IRSp53 and the IRSp53 expressed from the plasmid, we transfected the plasmid expressing IRSp53-EGFP into C2C12 cells and probed the lysates with anti-IRSp53 antibody

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

931

Fig. 2. Knocking down IRSp53 expression leads to the formation of thicker myotubes. (A) C2C12 cells microporated with plasmid expressing either scr-IRSp53 or sh-IRSp53 were fixed after 3 days in DM, stained with DAPI and anti-MyHC. (B) Fusion index was calculated as the percentage of total nuclei incorporated in myotubes. (C) (I) Cell lysate from C2C12 cells microporated with plasmid expressing either sh-IRSp53 or the scr-IRSp53 were analyzed by immunoblotting with anti-IRSp53 and anti-GAPDH. (II) Quantification of Western blots using densitometry. (D) (I) Protein extract from C2C12 cells undergoing differentiation were isolated on the days as indicated and probed with anti-MyHC, anti-MyoD and anti-GAPDH. (II) Quantification of Western blots using densitometry (*P < 0.05, ***P < 0.001).

(Fig. 3C). Transfection of plasmid expressing IRSp53-EGFP also inhibited the differentiation of C2C12 cells to myotubes (data not shown). 3.4. IRSp53 inhibits cell adhesion and spreading on fibronectin Fibronectin is a major constituent of endomysium (Hantai et al., 1983) and myogenesis is inhibited in the presence of antibody that reduced adhesion of myoblasts to fibronectin (Neff et al., 1982;

Boettiger et al., 1995). Thus, we assessed the adhesion properties of C2C12 cells microporated with pVect or one of the following plasmids; pscr-IRSp53, psh-IRSp53 or pIRSp53. C2C12 cells microporated with pVect (control), pscr-IRSp53 (data not shown) or psh-IRSp53 had similar adhesion properties on fibronectin coated surface (Fig. 4A). C2C12 cells over-expressing IRSp53 showed reduced cell adhesion on fibronectin coated surface compared to control C2C12 cells (Fig. 4A). Since the IRSp53 over-expressing cells displayed a difference in cell–ECM adhesion, we checked for

932

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 3. Over-expression of IRSp53 inhibits myotubes formation. (A) C2C12 cells were microporated with either pVect or pIRSp53 and transferred to DM. Three days after transferring to DM, the cells were fixed, stained with DAPI and anti-MyHC. (B) Fusion index of C2C12 cells microporated with plasmids as indicated in panel A was calculated as the percentage of total nuclei incorporated in myotubes. (C) (I) Cell lysate from C2C12 cells microporated with either vector or plasmid expressing IRSp53-EGFP were analyzed by immunoblotting with anti-IRSp53 and anti-GAPDH. (II) Quantification of Western blots using densitometry. (D) (I) Protein extract from C2C12 cells undergoing differentiation were isolated on days as indicated and probed with anti-MyHC, anti-MyoD and anti-GAPDH. (II) Quantification of Western blots using densitometry (*P < 0.05, **P < 0.01, ***P < 0.001).

the spreading of these transfected C2C12 cells on a fibronectin coated surface. Consistent with the cell adhesion data, control C2C12 cells and IRSp53 knockdown C2C12 cells adopted a flattened morphology within 30 min of addition to the fibronectin coated surface, while C2C12 cells over-expressing IRSp53 showed significant reduction in spreading on fibronectin (Fig. 4B-I). This was also reflected in the surface area of the cells: the cells microporated with

pIRSp53 had surface area significantly smaller than the control cells (Fig. 4B-II). Vinculin rich focal adhesions play a critical role in cell adhesion (Harburger and Calderwood, 2009) and we have previously found that cells with impaired cell adhesion exhibited reduced vinculin patches (Misra et al., 2007). Cells were microporated with the above-mentioned plasmids (pscr-IRSp53, psh-IRSp53,

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

933

Fig. 4. IRSp53 inhibits cell adhesion and cell spreading on fibronectin. (A) C2C12 cells microporated with plasmids as indicated were used to carry out cell adhesion assay as described in materials and methods. (B) (I) C2C12 were microporated with plasmids as indicated and 36 h after transfection, the cell spreading assay was carried out as described in Section 2. (II) The surface area of the cells in random fields was quantified using Metamorph software and plotted (**P < 0.01).

pVect, pIRSp53) and probed with the anti-vinculin antibody, while the actin cytoskeleton was visualized by fluorescently labeled phalloidin. C2C12 cells transfected with pVect, pscr-IRSp53 and psh-IRSp53 displayed prominent vinculin patches as discrete

streaks in the cell periphery (Fig. 5A). There were more vinculin patches per cell in the IRSp53 knockdown cells as compared to pscrIRSp53 transfected C2C12 cells (Fig. 5A). In contrast, C2C12 cells transfected with pIRSp53 had smaller and fewer vinculin patches

934

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 5. IRSp53 inhibits the formation of vinculin patches. (A) (I) C2C12 cells were microporated with plasmids as indicated and probed with anti-vinculin primary antibody and subsequently probed with alexa-488 secondary antibody. The actin cytoskeleton was visualized using alexa-568 Phalloidin. (II) Quantification of number of vinculin patches per cell; averaged values from three replicates are shown. (B) C2C12 cells microporated with plasmids expressing IRSp53, scr-IRSp53 or sh-IRSp53 were lysed and analyzed by immunoblotting with anti-vinculin and anti-GAPDH (*P < 0.05, **P < 0.01).

compared cells transfected with pVect (Fig. 5A) and this is not due to reduced vinculin expression in IRSp53 over-expressing cells (Fig. 5B). 3.5. The IMD and SH3 domains of IRSp53 are essential for inhibition of myogenic differentiation IRSp53 is an adaptor protein that consists of IMD, CRIB and SH3 domains (Fig. 6A). The IMD of IRSp53 has been shown to have actin bundling activities and the four lysine residues (amino acid 142, 143, 146 and 147) in this domain have been shown to be essential for filopodia induction (Millard et al., 2005; Yamagishi et al., 2004). Thus, we mutated these residues to alanine and tested the function of the mutant IRSp534A in inhibiting myogenic differentiation. C2C12 cells microporated with pIRSp534A formed myotubes, unlike the cells microporated with pIRSp53, suggesting that the mutations in the IMD domain abolished the ability of IRSp53 to inhibit myoblast differentiation (Fig. 6B and C). The expression of the mutant IRSp534A was comparable to the expression of wild type IRSp53 (Fig. 6E). The expression of MyoD and MyHC were higher in C2C12 cells over-expressing IRSp534A as compared to C2C12 cells over-expressing IRSp53 (Fig. 6D) consistent with the higher fusion index observed in C2C12 cells expressing IRSp534A . The SH3 domain of IRSp53 mediates interaction with a number of proline-rich proteins including WAVE1, N-WASP and WIRE

(Lim et al., 2008; Miki et al., 2000; Misra et al., 2010). In order to determine the role of the SH3 domain in blocking C2C12 myogenic differentiation, we compared differentiation of C2C12 cells over-expressing IRSp53 and IRSp532A (F428A, P429A, mutation in the SH3 domain) (Lim et al., 2008). C2C12 cells over-expressing IRSp532A differentiated and formed myotubes indicating that SH3 domain is required for inhibition of myogenic differentiation by IRSp53 in this experimental system (Fig. 6B–D). The expression of MyoD and MyHC were higher in C2C12 cells over-expressing IRSp532A compared to C2C12 cells over-expressing IRSp53 (Fig. 6D) consistent with the higher fusion index observed in C2C12 cells over-expressing IRSp532A . The inability of the mutant IRSp532A to inhibit myogenic differentiation is not due to its poor expression (Fig. 6E). 3.6. The IMD and SH3 domains of IRSp53 regulate cell adhesion and filopodia induction We examined the roles of IMD and SH3 domains in regulating cell–matrix adhesion by comparing the cell adhesion properties of C2C12 cells expressing either wild type IRSp53 or its mutants (IRSp534A and IRSp532A ). Cells over-expressing IRSp534A or IRSp532A displayed higher cell adhesion to fibronectin compared to the cells over-expressing IRSp53. C2C12 cells expressing the IMD mutant, IRSp534A , displayed the higher cell adhesion

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

935

Fig. 6. The IMD and SH3 domains of IRSp53 are critical for the inhibition of myotubes formation. (A) Domain organization of IRSp53 and the mutants used in the present study. (B) C2C12 cells were microporated with plasmid expressing IRSp53 (WT or its mutants; IRSp532A , IRSp534A ) and transferred to DM 36 h after microporation. Three days after transferring to DM, the cells were fixed, stained with DAPI and anti-MyHC. (C) Fusion index of C2C12 cells microporated with plasmids as described in panel A was calculated as described in Fig. 2B. (D) (I) Protein extract from C2C12 cells undergoing differentiation were isolated on days as indicated and probed with anti-MyHC, anti-MyoD and anti-GAPDH. (II) Quantification of Western blots using densitometry. (E) Protein extract from C2C12 cells microporated with plasmid expressing IRSp53-EGFP or its mutants (IRSp532A , IRSp534A ) were analyzed by immunoblotting with anti-IRSp53 and anti-GAPDH (*P < 0.05, ***P < 0.001).

936

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 6. (Continued ).

to fibronectin than the control cells (Fig. 7A). We also carried out cell spreading assays on fibronectin coated surfaces and the results correlated with cell adhesion assay data: C2C12 cells overexpressing IRSp534A and IRSp532A displayed a faster spreading rate as compared to C2C12 cells over-expressing IRSp53 (Fig. 7B). Consistent with the cell adhesion assay, vinculin patches were more prominent in C2C12 cells expressing the mutants (IRSp534A and IRSp532A ) as compared to the cells expressing IRSp53 (Fig. 8) and cells over-expressing IRSp534A had higher number of vinculin patches than the control cells. This is not due to changes in the expression of vinculin (data not shown). Filopodia induction by IRSp53 was characterized by transfecting plasmids expressing, EGFP, IRSp53-EGFP or IRSp53 mutants (IRSp532A -EGFP, IRSp534A -EGFP). Almost 60% of the IRSp53-EGFP transfected cells exhibited filopodia while both IRSp534A and IRSp532A mutants provoked a reduced number of cells with filopodia (Fig. S2A and B). IRSp53-EGFP localized efficiently to the plasma membrane while IRSp532A -EGFP had diminished membrane localization as compared to IRSp53-EGFP. We did not detect any plasma membrane localization in the case of IRSp534A -EGFP (Fig. S2A and C). This implies that IMD and the SH3 domains of IRSp53 are critical for its localization to the plasma membrane as well as for generating membrane protrusions. 3.7. The IMD domain of IRSp53 is sufficient to inhibit myogenic differentiation The IMD and SH3 domains of IRSp53 play a critical role in inhibition of myogenic differentiation (Fig. 6). To substantiate the

involvement of IMD and SH3 domains in IRSp53 function, we constructed two plasmids pIRSp53IMD (IRSp531–250 ; IMD domain) and pIRSp53SH3 (IRSp53367–521 ; SH3 domain) (Fig. 6A), microporated the plasmids into C2C12 cells and analyzed the differentiation of the cells into myotubes. C2C12 cells over-expressing IRSp53IMD differentiated inefficiently (Fig. 9A and B) while the cells overexpressing the IRSp53SH3 differentiated just like control cells (data not shown). Over-expression of IRSp53IMD affected the expression of myogenic markers similar to the over-expression by IRSp53 (Fig. 9C). Mutating the four lysine residues of the IMD domain abolished the ability of the IRSp53IMD4A mutant to inhibit C2C12 differentiation into myotubes (Fig. 9A and B) and the cells overexpressing IRSp53IMD4A had a higher fusion index than the control cells. The inability of IRSp53IMD4A mutant to inhibit differentiation was not due to poor expression of the mutant protein as compared to IRSp53IMD (Fig. 9D). C2C12 cells over-expressing IRSp53IMD had reduced cell adhesion similar to cells over-expressing IRSp53 (Fig. S3A) while the C2C12 cells over-expressing IRSp53IMD4A were found to bind efficiently to fibronectin coated surface and had a binding efficiency higher than the control cells (Fig. S3A). Binding deficiency of C2C12 cells over-expressing IRSp53IMD was reflected in the spreading assay as well, where even after 30 min C2C12 cells over-expressing IRSp53 or IRSp53IMD had not adopted a flattened morphology in contrast to C2C12 cells over-expressing IRSp53IMD4A (Fig. S3B). This correlated with the finding that cells over-expressing IRSp53IMD had a reduced number of vinculin patches similar to cells overexpressing IRSp53, while cells over-expressing IRSp53IMD4A had prominent vinculin patches, more than the control cells (Fig. S4).

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

937

Fig. 7. The IMD and SH3 domains of IRSp53 are critical for the inhibition of cell adhesion and spreading on fibronectin. (A) C2C12 cells microporated with plasmids as indicated were trypsinized 36 h after microporation and cell adhesion assay was carried out as described in Fig. 4A. (B) (I) C2C12 cells microporated with plasmids as indicated were used to carry out cell spreading assay as described in Section 2. (II) The surface area of the cells in random fields was quantified and plotted (*P < 0.05, **P < 0.01).

This was not due to changes in the expression of vinculin (data not shown). We also found that IRSp53IMD -EGFP alone is sufficient to localize to the membrane and induce filopodia in C2C12 cells, whereas mutation in the IMD domain abolished this activity (Fig. S5A–C). Thus, IRSp53IMD (IRSp531–250 ) had all the activities of the full length IRSp53 in regulating myogenesis, filopodia formation and cell–ECM adhesion suggesting that the IMD is the functional domain of IRSp53.

4. Discussion The fusion of myoblasts to form multinucleate myotubes is one of the critical events in the formation of skeletal muscle and in muscle regeneration (Charrasse et al., 2006). The actin cytoskeleton is crucial for many of the cellular processes and remodeling of the actin cytoskeleton may play a vital role in myogenesis (Kim et al., 2007). In this study, we show that the expression of IRSp53 is

938

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 8. The IMD and SH3 domains of IRSp53 play a critical role in regulating vinculin patch assembly. (A) C2C12 cells microporated with plasmids as indicated were fixed, permeabilized and probed with anti-vinculin primary antibody and alexa-488 secondary antibody. The actin cytoskeleton was visualized using alexa-568 Phalloidin. (B) Quantification of number of vinculin patches per cell; averaged values from three replicates are shown (**P < 0.01).

down-regulated during the formation of myotubes from myoblasts (Fig. 1 and Fig. S1), suggesting that IRSp53 might play a negative role in myogenic differentiation. The inhibitory role of IRSp53 in myogenic differentiation was confirmed by the finding that knocking down the expression of IRSp53 led to the acceleration of myotubes formation (Fig. 2), while over-expression of IRSp53 inhibited the formation of myotubes (Fig. 3). The N-terminal IMD domain of IRSp53 interacts with Rac (Miki and Takenawa, 2002), has the ability to negatively bend the membranes (Suetsugu et al., 2006) and binds to actin (Millard et al., 2005), whereas the C-terminal SH3 domain interacts with a number of proline-rich proteins such as, WAVE1, WAVE2, N-WASP, WIRE (Lim et al., 2008; Miki et al., 2000; Misra et al., 2010). Unlike the wild type (WT) IRSp53, neither the IMD domain mutant (IRSp534A ) nor the SH3 domain mutant (IRSp532A ) inhibited myogenic differentiation (Fig. 6). This implies that the IMD and/or SH3 domains may play a direct role in IRSp53-mediated inhibition of differentiation. Over-expression of IMD domain (IRSp531–250 ) severely compromised myogenic differentiation while over-expression of SH3 domain (IRSp53367–521 ) did not affect the myogenic differentiation of C2C12 cells. These results suggest that the IRSp53IMD is responsible for inhibition of C2C12 myoblast differentiation and that the SH3 domain could mask the activity of the IMD domain of IRSp53. We have previously shown that IRSp53 localizes efficiently to the plasma membrane in the presence of WIRE which binds to the SH3 domain of IRSp53 (Misra et al., 2010). The F-BAR domain of syndapin 1 is auto-inhibited by

its SH3 domain and this auto-inhibition is relieved by the binding of the SH3 domain to the proline-rich domain of dynamin 1 (Rao et al., 2010). Thus, it is possible that binding of the SH3 domain of IRSp53 to proline-rich proteins may lead to a change in conformation that exposes the IMD domain thereby allowing its membrane localization leading to its function in myogenic differentiation. Cell–ECM adhesion is critical for myogenic differentiation as monoclonal antibody specific to ␤1 integrin hampers cell adhesion to fibronectin and inhibits myogenesis (Neff et al., 1982; Boettiger et al., 1995). Knocking down the IRSp53 expression in C2C12 cells had no effect on cell adhesion and spreading (Fig. 4), in contrast to an earlier report which showed that knocking down IRSp53 expression in NIH3T3 fibroblast cells reduced cell spreading on fibronectin (Roy et al., 2009). This difference could be due to the different cell types used, fibroblasts versus myoblasts. Though there was no significant change in cell adhesion properties of IRSp53 knock down C2C12 cells, we observed an increased number of vinculin patches in the IRSp53 knock down cells (Fig. 5). In contrast to the IRSp53 knock down, over-expression of IRSp53 in C2C12 caused a significant decrease in cell adhesion and spreading on fibronectin (Fig. 4) which correlated with the reduced number of vinculin patches in C2C12 cells expressing IRSp53 (Fig. 5). Similar results were also observed in C2C12 cells over-expressing IRSp53IMD : inhibition of myogenic differentiation, reduced cell–ECM adhesion and vinculin patches. In contrast, the expression of the mutants IRSp534A , IRSp532A and IRSp53IMD4A

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

939

Fig. 9. Over-expression of IRSp53IMD is sufficient to inhibit myotube formation. (A) C2C12 cells were microporated with plasmids as indicated and transferred to DM. Three days after transferring to DM, the cells were fixed, stained with DAPI and anti-MyHC. (B) Fusion index of C2C12 cells microporated with plasmids as indicated in panel A was calculated as described in Fig. 2B. (C) Protein extract from C2C12 cells undergoing differentiation were isolated on days as indicated and probed with anti-MyHC, anti-MyoD and anti-GAPDH. (II) Quantification of Western blots using densitometry. (D) Cell lysate from C2C12 cells microporated with plasmid expressing IRSp53-EGFP, IRSp53IMD -EGFP or IRSp53IMD4A -EGFP were analyzed by immunoblotting with anti-IRSp53 and anti-GAPDH (*P < 0.05, **P < 0.01, ***P < 0.001).

940

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941

Fig. 9. (Continued ).

in C2C12 did not affect myogenic differentiation, cell–ECM adhesion or vinculin patch assembly. Over-expression of IRSp53IMD4A in C2C12 cells enhanced myogenic differentiation and the number of vinculin patches per cell, similar to the IRSp53 knock down cells. This suggests that IRSp53IMD4A interferes with the function of endogenous IRSp53 in regulating myogenic differentiation and vinculin patch assembly. It is possible that IRSp53IMD4A may be titrating out proteins or factors interacting with the endogenous protein. In agreement with previous findings (Lim et al., 2008; Krugmann et al., 2001), we observed membrane localization and filopodia induction by IRSp53 and IRSp53IMD while IRSp534A , IRSp532A and IRSp53IMD4A were inefficient in membrane localization and filopodia induction. Filopodial extension from the tip of the long axis of the myoblast is vital for myogenic differentiation (Nowak et al., 2009); however, IRSp53 induces filopodia all around the cell periphery. This non-polarized filopodia formation by IRSp53 may be detrimental to myogenic differentiation. Our results suggest that IRSp53 is a negative regulator of myogenesis and that the IMD domain of IRSp53 is responsible for this function. The IMD domain of IRSp53 induces filopodia, and causes reduction in cell adhesion. It is possible that either one or both of these activities is responsible for the inhibition of myogenic differentiation.

Acknowledgements This work was supported by the Tier-2 research grant (MOE2008-T2-1-026), Ministry of Education, Singapore. We thank S. Oliferenko and M. Featherstone for critical reading of the manuscript.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.biocel.2012.02.020. References Adams JC, Seed B, Lawler J. Muskelin, a novel intracellular mediator of cell adhesive and cytoskeletal responses to thrombospondin-1. EMBO J 1998;17:4964–74. Andres V, Walsh K. Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J Cell Biol 1996;132:657–66. Bader D, Masaki T, Fischman DA. Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J Cell Biol 1982;95:763–70. Benoit YD, Lussier C, Ducharme PA, Sivret S, Schnapp LM, Basora N, et al. Integrin alpha8beta1 regulates adhesion, migration and proliferation of human intestinal crypt cells via a predominant RhoA/ROCK-dependent mechanism. Biol Cell 2009;101:695–708. Boettiger D, Enomoto-Iwamoto M, Yoon HY, et al. Regulation of integrin alpha 5 beta 1 affinity during myogenic differentiation. Dev Biol 1995;169:261–72. Charge SB, Rudnicki MA. Cellular and molecular regulation of muscle regeneration. Physiol Rev 2004;84:209–38. Charrasse S, Comunale F, Grumbach Y, et al. RhoA GTPase regulates M-cadherin activity and myoblast fusion. Mol Biol Cell 2006;17:749–59. Chen EH, Olson EN. Antisocial, an intracellular adaptor protein, is required for myoblast fusion in drosophila. Dev Cell 2001;1:705–15. Enomoto MI, Boettiger D, Menko AS. [alpha]5 Integrin is a critical component of adhesion plaques in myogenesis. Dev Biol 1993;155:180–97. Hantai D, Gautron J, Labat-Robert J. Immunolocalization of fibronectin and other macromolecules of the intercellular matrix in the striated muscle fiber of the adult rat. Coll Relat Res 1983;3:381–91. Harburger DS, Calderwood DA. Integrin signalling at a glance. J Cell Sci 2009;122:159–63. Kim S, Shilagardi K, Zhang S, et al. A critical function for the actin cytoskeleton in targeted exocytosis of prefusion vesicles during myoblast fusion. Dev Cell 2007;12:571–86. Knudsen KA, Horwitz AF. Tandem events in myoblast fusion. Dev Biol 1977;58:328–38. Krugmann S, Jordens I, Gevaert K, et al. Cdc42 induces filopodia by promoting the formation of an IRSp53:mena complex. Curr Biol 2001;11:1645–55.

A. Misra et al. / The International Journal of Biochemistry & Cell Biology 44 (2012) 928–941 Le Clainche C, Carlier MF. Regulation of actin assembly associated with protrusion and adhesion in cell migration. Physiol Rev 2008;88:489–513. Lim KB, Bu W, Goh WI, et al. The Cdc42 effector IRSp53 generates filopodia by coupling membrane protrusion with actin dynamics. J Biol Chem 2008;283:20454–72. Liu H, Niu A, Chen SE, Li YP. {beta}3-Integrin mediates satellite cell differentiation in regenerating mouse muscle. FASEB J 2011;25:1914–21. Louis M, Zanou N, Van Schoor M, Gailly P. TRPC1 regulates skeletal myoblast migration and differentiation. J Cell Sci 2008;121:3951–9. Miki H, Takenawa T. WAVE2 serves a functional partner of IRSp53 by regulating its interaction with Rac. Biochem Biophys Res Commun 2002;293:93–9. Miki H, Yamaguchi H, Suetsugu S, Takenawa T. IRSp53 is an essential intermediate between Rac and WAVE in the regulation of membrane ruffling. Nature 2000;408:732–5. Millard TH, Bompard G, Heung MY, et al. Structural basis of filopodia formation induced by the IRSp53/MIM homology domain of human IRSp53. EMBO J 2005;24:240–50. Misra A, Lim RP, Wu Z, Thanabalu T. N-WASP plays a critical role in fibroblast adhesion and spreading. Biochem Biophys Res Commun 2007;364:908–12. Misra A, Rajmohan R, Lim RP, et al. The mammalian verprolin, WIRE induces filopodia independent of N-WASP through IRSp53. Exp Cell Res 2010;316:2810–24. Neff NT, Lowrey C, Decker C, et al. A monoclonal antibody detaches embryonic skeletal muscle from extracellular matrices. J Cell Biol 1982;95:654–66. Nowak SJ, Nahirney PC, Hadjantonakis AK, Baylies MK. Nap1-mediated actin remodeling is essential for mammalian myoblast fusion. J Cell Sci 2009;122:3282–93. Rao Y, Ma Q, Vahedi-Faridi A, et al. Molecular basis for SH3 domain regulation of F-BAR-mediated membrane deformation. Proc Natl Acad Sci USA 2010;107:8213–8.

941

Ren H, Yin P, Duan C. IGFBP-5 regulates muscle cell differentiation by binding to IGF-II and switching on the IGF-II auto-regulation loop. J Cell Biol 2008;182: 979–91. Rochlin K, Yu S, Roy S, Baylies MK. Myoblast fusion: when it takes more to make one. Dev Biol 2010;341:66–83. Roy BC, Kakinuma N, Kiyama R. Kank attenuates actin remodeling by preventing interaction between IRSp53 and Rac1. J Cell Biol 2009;184: 253–67. Scita G, Confalonieri S, Lappalainen P, Suetsugu S. IRSp53: crossing the road of membrane and actin dynamics in the formation of membrane protrusions. Trends Cell Biol 2008;18:52–60. Springer ML, Rando TA, Blau HM.Haines JL, et al., editors. Gene delivery to muscle. Current protocols in human genetics; 2002 [Chapter 13: Unit 13.4]. Suetsugu S, Murayama K, Sakamoto A, et al. The RAC binding domain/IRSp53-MIM homology domain of IRSp53 induces RAC-dependent membrane deformation. J Biol Chem 2006;281:35347–58. Swailes NT, Colegrave M, Knight PJ, Peckham M. Non-muscle myosins 2A and 2B drive changes in cell morphology that occur as myoblasts align and fuse. J Cell Sci 2006;119:3561–70. Velleman SG, McFarland DC. Beta1 integrin mediation of myogenic differentiation: implications for satellite cell differentiation. Poult Sci 2004;83: 245–52. Yamagishi A, Masuda M, Ohki T, et al. A novel actin bundling/filopodium-forming domain conserved in insulin receptor tyrosine kinase substrate p53 and missing in metastasis protein. J Biol Chem 2004;279:14929–36. Yeh TC, Ogawa W, Danielsen AG, Roth RA. Characterization and cloning of a 58/53-kDa substrate of the insulin receptor tyrosine kinase. J Biol Chem 1996;271:2921–8.