Tyrosine phosphorylation of calponins Inhibition of the interaction with F-actin Julien Abouzaglou 1 , Christine Be ´ nistant 1 , Mario Gimona 2 , Claude Roustan 3 , Rhida Kassab 1 and Abdellatif Fattoum 1 1 Centre de Recherches de Biochimie Macromole ´ culaire du CNRS, Montpellier, France; 2 Institute of Molecular Biology, Salzburg, Austria; 3 UMR 5539, CNRS, UM2, EPHE, Montpellier, France The phosphorylation-dephosphorylation of serine and threonine residues of calponin is known to modulate in vitro its interaction with F-actin and is thought to regulate several biological processes in cells, involving either of the calponin isoforms. Here, we identify, for the first time, tyrosine- phosphorylated calponin h3 within COS 7 cells, before and after their transfection with the pSV vector containing cDNA encoding the cytoplasmic, Src-related, tyrosine kin- ase, Fyn. We then describe the specific tyrosine phosphory- lation in vitro of calponin h1 and calponin h3 by this kinase. 32 P-labeling of tyrosine residues was monitored by combined autoradiography, immunoblotting with a specific phospho- tyrosine monoclonal antibody and dephosphorylation with the phosphotyrosine-specific protein phosphatase, YOP. PhosphorImager analyses showed the incorporation of maximally 1.4 and 2.0 mol of 32 P per mol of calponin h3 and calponin h1, respectively. As a result, 75% and 68%, respectively, of binding to F-actin was lost by the phos- phorylated calponins. Furthermore, F-actin, added at a two- or 10-fold molar excess, did not protect, but rather increased, the extent of 32 P-labeling in both calponins. Structural analysis of the tryptic phosphopeptides from each 32 P-labe- led calponin revealed a single, major 32 P-peptide in calponin h3, with Tyr261 as the phosphorylation site. Tyr261 was also phosphorylated in calponin h1, together with Tyr182. Col- lectively, the data point to the potential involvement, at least in living nonmuscle cells, of tyrosine protein kinases and the conserved Tyr261, located in the third repeat motif of the calponin molecule, in a new level of regulation of the actin– calponin interaction. Keywords: actin; calponin h1; calponin h3; tyrosine phos- phorylation. Calponin belongs to a family of actin-binding proteins, which includes the basic and neutral calponin isoforms, designated h1 and h2, respectively, both of which are present in smooth muscle [1,2], and the acidic calponin, h3, expressed in smooth muscle as well as in nonmuscle cells [3]. They are thought to be involved in a variety of biological processes, such as the regulation of smooth muscle contraction, by inhibiting the actomyosin ATPase activity [4,5], the organization of the actin cytoskeleton in smooth muscle and nonmuscle cells by stabilizing actin networks [6–8], and cell signaling at the surface membrane of vascular smooth muscle [9,10]. All calponin isoforms are composed of a conserved N-terminal calponin homology domain, followed by a primary actin-binding consensus sequence and three 30-residue tandem repeats harboring a secondary actin-binding site and associated with a C-terminal tail of variable length and primary structure [11–13]. Their actin- binding properties are believed to be regulated by two main factors, revealed by in vitro studies, namely the interaction of calponin with calcium-binding proteins, such as calmod- ulin [5], and the phosphorylation of specific serine/threonine residues by protein kinase C and Ca 2+ calmodulin- dependent protein kinase II [14,15]. Both regulatory events lead to the dissociation of the calponin–F-actin complex in vitro. Given the relatively low affinity of calponin for calmodulin [5], the second process would be the most active effector in vivo. On the other hand, tyrosine phosphoryla- tion of calponin could represent an additional mechanism involved in the regulation of the calponin–actin interaction. In this regard, in recent years several actin-binding proteins, including villin [16,17], gelsolin [18,19] and cortactin [20], have been shown to be tyrosine phosphorylated with concomitant changes in their actin-binding properties, and most are, like calponin, proteins participating in the reorganization of the actin cytoskeleton. Less well charac- terized is the tyrosine phosphorylation of calponins and its effects on the actin-binding capacity of the different calponin isoforms. To gain further insight into the mech- anisms regulating the calponin–actin interaction, we assessed in this study, for the first time, the production of tyrosine-phosphorylated calponin h3 in COS 7 cells and investigated the tyrosine phosphorylation in vitro of both calponin h3 and calponin h1 by the cytoplasmic, Src-related protein tyrosine kinase, Fyn. Our quantitative data illustrate the specific incorporation of 32 P into each calponin, which takes place both in the absence and in the presence of F-actin, together with the resulting significant decrease of F-actin binding to either phosphorylated calponin. More- over, structural studies, using radio-Edman sequencing, and Correspondence to A. Fattoum, Centre de Recherches de Biochimie Macromole ´ culaire, CNRS UPR 1086, 1919, route de Mende, F-34293, Montpellier Cedex 5, France. Fax: + 33 467 521559, Tel.: + 33 467 613338, E-mail: fattoum@crbm.cnrs-mop.fr Note: a website is available at http://www.crbm.html (Received 4 February 2004, accepted 28 April 2004) Eur. J. Biochem. 271, 2615–2623 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04190.x direct amino acid sequencing of 32 P-labeled peptides from each calponin, allowed identification of the conserved Tyr261, located in the third repeat motif of the calponin molecule, as the major target of the kinase reaction. Together, the results suggest that the enzymatic tyrosine phosporylation of calponin represents a biologically rele- vant process which can occur both in vivo and in vitro,with Tyr261 of the protein forming a novel site involved in the regulation of interactions between calponins and F-actin. Materials and methods Chemicals Monoclonal anti-phosphotyrosine Ig (clone 4G10) was purchased from Upstate Biotechnology Inc., and [ 32 P]dATP[cP] (5000 CiÆmmol )1 ) was from Amersham Biosciences. The recombinant full-length human protein tyrosine kinase, Fyn, had a specific activity of 1035 UÆmg )1 and was obtained from Upstate Biotechnology Inc. The recombinant protein tyrosine phosphatase, YOP, from Yersinia enterocolitica was supplied by New England Bio- Labs Inc. at a specific activity of 100 000 UÆmg )1 .Trypsin was from Worthington and ATP was from Boehringer. Protein preparations F-actin from rabbit skeletal muscle was prepared as described previously [21]. Basic calponin h1 was isolated from fresh turkey gizzards, as described previously [21]. Recombinant rat acidic calponin h3 was expressed as described previously [13]. The protein concentrations were measured spectrophotometrically using A 1% 280 ¼ 11.7 for actin, 7.6 for calponin h1, and 8.9 for calponin h3 [22]. Tyrosine phosphorylation of calponins with Fyn Calponin h1 or calponin h3 (0.16 lg) were phosphorylated in vitro with increasing concentrations of Fyn (0–0.20 UÆlg )1 of calponin) in an assay mixture (30 lL) containing 25 m M Hepes, pH 7.5, 1 m M dithiothreitol, 12 m M MnCl 2 ,0.16lCi of [ 32 P]dATP[cP] and 5 l M ATP. The reactions were carried out for 60 min at 30 °Cand stopped by the addition of Laemmli sample buffer. Tyro- sine-phosphorylated calponins were separated by SDS– PAGE and the radioactivity incorporated into each protein was detected by autoradiography and quantified by Phos- phorImager analysis (Typhon 9200; Amersham Bioscienc- es) or by densitometric measurements of the autoradiograms using the Scan Analysis software of Biosoft (Cambridge, UK). The calponin phosphorylation in the presence of F-actin was conducted under the same experi- mental conditions using 0.20 U of Fyn per lg of calponin, a calponin/actin molar ratio of 1 : 2 or 1 : 10, and an incubation period of 10 min. A phosphorylation control of F-actin, without calponin, was carried out in parallel. After gel electrophoresis, the amounts of radioactivity associated with calponin and actin were measured. The dephosphorylation of calponins was achieved using the final 30 lL of Fyn kinase mixture, which was adjusted to 50 m M Tris/HCl, pH 7.0, 100 m M NaCl, 2 m M EDTA, 5m M dithiothreitol, 0.01% Brij 35 and supplemented with 1mgÆmL )1 BSA and 50 U of YOP protein tyrosine phosphatase. After 60 min at 30 °C, the reaction was terminated by the addition of Laemmli sample buffer. The extent of tyrosine dephosphorylation was assessed by SDS/ PAGE and autoradiography. Cell culture and transfections COS 7 cells were grown to 60–80% confluence in DMEM (Dulbecco’s modified Eagle’s medium) supplemented with 10% fetal bovine serum and penicillin/streptomycin (10 UÆmL )1 /10 lgÆmL )1 )at37°Cand5%CO 2 . Cells were transfected with the pSV vector containing the cDNA encoding Fyn kinase [23], or with the empty pSV vector. The transfection was conducted by using the lipofectamin system (Gibco) according to the protocol of the manufacturer. The transfected cells were incubated for 36 h and, after the addition of 1 m M Na vanadate, incubated for a further 30 min. They were then washed twiceincoldNaCl/P i (PBS), scraped into 400 lLof RIPA buffer [20 m M Tris/HCl, pH 7.5, 150 m M NaCl, 1% (v/v) Triton-X-100, 1% Na deoxycholate, 0.1% SDS containing 1 m M dithiothreitol, 1 m M Na vanadate, 1 m M NaF, 1 m M phenylmethanesulfonyl fluoride, 10 lgÆmL )1 aprotinin and 20 l M leupeptin] and then incubated at 4 °C for 10 min. The overall cell medium was subse- quently potterized 10· in a Dounce potter with a small clearance to achieve an efficient extraction of proteins, and then centrifuged at 20 000 g for 5 min at 4 °C. The supernatants were collected and subjected to immuno- precipitation with affinity-purified anti-(calponin h3) Ig (20 lg). The immunocomplexes were recovered over Protein A–Sepharose beads (Pharmacia), which were washed once in RIPA buffer and twice in WLB buffer (20 m M Tris/HCl, pH 7.5, 150 m M NaCl, 1% Nonidet P-40 containing 1 m M dithiothreitol and 1 m M Na vanadate). Beads were resuspended in Laemmli sample buffer and centrifuged. The supernatants were submitted to 10% acrylamide gel electrophoresis and Western blotting analyses using anti-phosphotyrosine, anti- (calponin h3) or anti-Fyn [23] Ig. Electrophoresis and immunoblots Unless stated otherwise, SDS–PAGE was carried out in 5–18% gradient acrylamide gels [24]. The running buffer was 50 m M Tris, 100 m M boric acid, pH 8.0. The gels were stained with Coomassie Blue R-250 and destained with 7% acetic acid. Densitometric analysis of protein bands on electrophoretic gels was carried out using the SCAN ANALYSIS software of Biosoft. Following electro- phoresis, phosphorylated calponins were then transferred to nitrocellulose membranes and subjected to immuno- blotting with phosphotyrosine-specific monoclonal anti- body (diluted 1 : 1000) and secondary anti-mouse immunoglobulin G (IgG) conjugated to horseradish peroxidase (diluted 1 : 5000). The reaction was revealed by the enhanced chemiluminescence technique, using the kit from Amersham Biosciences [25]. Immunoblots were prepared in a similar manner using affinity-purified anti- (calponin h3) Ig directed to the specific C-terminal tail of the protein (E 311-Q 326) [26]. 2616 J. Abouzaglou et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Co-sedimentation assays Calponin h1 or h3 (5 lg), in a final volume of 50 lL, was first phosphorylated with nonradioactive ATP (5 l M )and 3.4UofFynin25 m M Hepes, pH 7.5, under the conditions described above. F-actin was then added at a twofold molar excess over calponin. After incubation for 30 min at 30 °C, the protein mixtures were centrifuged, in a Beckman Airfuge, at 90 000 g for 30 min at 25 °C. The same experiments, carried out in the absence of Fyn, were used as controls. The supernatants and pellets that were homo- genized in the same buffer, were subjected to gel electro- phoresis. Structural analyses of tryptic phosphopeptides 32 P-labeled calponin h1 or calponin h3 (7 lg) were first purified by SDS–PAGE in 10% gels. After staining with Coomassie Blue R-250, the corresponding bands were excised and subjected to in-gel digestion with trypsin according to Rosenfeld et al. [27]. The resulting digests were extracted from the gel slices with 50% acetonitrile and concentrated in vacuo. Each peptide mixture was fraction- ated by reverse-phase HPLC on a 2· 100 mm, Aquapore C8 column (Applied Biosystems), eluted with a linear acetonitrile gradient (5–30%) in 0.1% trifluoroacetic acid for 60 min, at a flow rate of 0.18 mLÆmin )1 , and monitored by Cerenkov counting. The radioactive fractions were separated and radiosequenced with a Procise sequencer (Applied Biosystems) using a program which allows detec- tion of the position of 32 P-labeled amino acids by counting the radioactivity of the anilino-thiazolinone derivative released at each Edman degradation cycle. For direct amino acid sequencing, each radioactive peak was further purified by HPLC on a 1· 100 mm Aquapore C 18 column eluted with a linear gradient of 9–33% acetonitrile. Sequence determination of each pure phosphopeptide was performed with the same Procise sequencer using the standard pulsed- liquid program. Results Tyrosine phosphorylation of calponin h3 in COS 7 cells We initiated our work by experiments aiming to determine the ability of calponin h3 to be phosphorylated on tyrosine in cultured COS 7 cells by cytoplasmic, Src-related protein tyrosine kinases, such as Fyn. Our attention was focused towards this enzyme because it is an ubiquitous kinase expressed in most tissues and it exhibits the functional properties of all the other Src kinases [28]. Moreover, transfection of COS 7 with the pSV vector carrying Fyn cDNA takes place with a high yield. To assess tyrosine phosphorylation of the endogenous calponin h3 present in COS 7 cells, we employed protein extracts derived from cells transfected either with the pSV vector carrying the Fyn cDNA, or with the empty pSV vector. Na vanadate was added to all extraction buffers to inhibit protein phosphotyrosine phosphatases. Calponin h3 was then immunoprecipitated with the affinity-purified anti- (calponin h3) Ig. The immune complexes were separated by SDS gel electrophoresis and probed with primary antibodies directed to phosphotyrosine or to calponin h3. COS 7 cells transfected with the empty pSV vector gave rise to a detectable band, migrating at 36 kDa, which was identified as calponin h3 by immunoblotting with the anti-(calponin h3) Ig (Fig. 1A, lane 1, panel c). This band showed a faint intensity when probed with anti-phosphotyrosine Ig (Fig. 1A, lane 1, panel b). The cells transfected with the pSV vector containing Fyn cDNA yielded an identical calponin h3 species (Fig. 1A, lane 2, panel c) but with a significantly increased intensity when probed with the phosphotyrosine antibody (Fig. 1A, lane 2, panel b). The densitometric measurements of the blots, presented in Fig. 1B, show a threefold higher intensity of the phos- phorylated calponin in the Fyn cDNA-transfected cells Fig. 1. Tyrosine phosphorylation of calponin h3 in COS 7 cells. (A) The lysates prepared from COS 7 cells transfected with pSV vector con- taining Fyn cDNA (lane 2) or with the empty pSV vector (lane 1) were treated with anti-(calponin h3) Ig, as indicated in the Materials and methods; the resulting immunoprecipitates were subjected to acryla- mide gel electrophoresis (10% gel) and Western blot analyses with anti- (Fyn kinase) Ig (a), anti-(phosphotyrosine) Ig (b), or anti-(calponin h3) Ig (c). Lane 3: control immunoblots of purified rat calponin h3 (120 ng) using any of the three antibodies. (B) Densitometric meas- urements of the Western blots corresponding to COS 7 transfected with the empty pSV vector (a) or the pSV vector containing Fyn cDNA (b). Hatched bars, calponin h3 blots; dark bars, tyrosine- phosphorylated calponin h3 blots; white bars, Fyn kinase blots. Ó FEBS 2004 Tyrosine phosphorylated calponin–actin interaction (Eur. J. Biochem. 271) 2617 (Figs 1B, b) than in the control cells transfected with the empty pSV vector (Fig. 1B, a), whereas the level of total calponin was unchanged. Moreover, the immunoprecipitate from the former cells displayed an additional band reactive to Fyn kinase antibodies (Fig. 1A, line 2, panel a, and Fig. 1B, b), suggesting that the production in these cells of a tight complex between the expressed kinase and the endogenous calponin h3, was resistant to the immunopre- cipitation of the latter protein. A much smaller amount of this band was also obtained from the control cells (Fig. 1A, line 1, panel a; and Fig. 1B, a). These findings clearly indicated that in normal COS 7 cells, a fraction of the endogenous calponin h3 was tyrosine phosphorylated, and the amount of this fraction was noticeably enhanced upon transfection of the cells with Fyn cDNA. They suggest that, in vivo, under physiological conditions, calponin probably represents a good substrate for Src family protein tyrosine kinases. This proposal is supported by the data, reported below, describing the tyrosine phosphorylation in solution of the calponin isoforms h1 and h3 by Fyn. In vitro phosphorylation of calponin h3 and calponin h1 by Fyn tyrosine kinase Incubation of purified calponin h3 or calponin h1, in the presence of increasing concentrations of Fyn kinase and [ 32 P]dATP[cP], resulted in the progressive and efficient incorporation of radioactive 32 P in both calponins. How- ever, the maximum extent of phosphorylation was different for the two calponins. At saturation, the stoichiometry of the phosphorylation reaction reproducibly plateaued at 1.4 mol of 32 P per mol of calponin h3 (Fig. 2A) and 2.0 mol of 32 P per mol of calponin h1 (Fig. 2B). These results suggest that the two calponin isoforms in solution are adequate substrates for in vitro phosphorylation with Fyn. Upon SDS gel electrophoresis, each 32 P-labeled calponin displayed an electrophoretic band migrating identically to the unlabeled protein (Fig. 3A,B). To determine whether the calponins were phosphorylated on tyrosine residues, we subjected each radioactively phosphorylated protein to dephosphorylation by the phosphotyrosine-specific protein phosphatase YOP and to immunoblotting with monoclonal anti(phosphotyrosine) Ig. The treatment of calponin h1 with YOP for 30 min caused a complete loss of the incorporated 32 P (Fig. 3C, lanes a and b); similarly, incubation of YOP with calponin h3 resulted in a total dephosphorylation of the protein (Fig. 3C, lanes c and d). These results demon- strate that the Fyn-catalyzed phosphorylation of calponins was also a reversible reaction. On the other hand, the Western blot analyses presented in Fig. 3D clearly indicate that the phosphorylated calponins were reactive to the anti(phosphotyrosine) Ig (Fig. 3D, lanes b and d) but not the unphosphorylated proteins (Fig. 3D, lanes a and c). Together, the findings support the conclusion that the phosphorylation of calponin h1 and calponin h3 takes place on tyrosine residues. Inhibition of actin binding to tyrosine-phosphorylated calponins To investigate the effects of tyrosine phosphorylation on the actin-binding activity of the two calponins, we analyzed, by quantitative co-sedimentation experiments, the interaction of F-actin with each modified calponin isoform. Calponin h3 or calponin h1, both exhibiting the maximum level of tyrosine phosphorylation, were mixed with F-actin at a molar ratio of 1 : 2 and then subjected to high-speed centrifugation. The partitioning of each phosphorylated calponin, and of actin, between the supernatant and pellet fractions, was determined by densitometry of Coomassie Blue-stained electrophoretic gels (Fig. 4B), and the data were compared with those obtained in the control co-sedimenta- tion experiments carried out with calponins incubated in the phosphorylation medium devoid of Fyn (Fig. 4A). The densitometric analyses depicted in Fig. 4C show that  98% of the control unphosphorylated calponin h1 (Fig. 4C, lanes a) and calponin h3 (Fig. 4C, lanes c) are associated with the F-actin pellet. In contrast,  68% of phosphorylated calpo- nin h1 (Fig. 4C, lanes b) and 78% of phosphorylated calponin h3 (Fig. 4C, lanes d) failed to bind to F-actin and remained in the supernatant fractions. These data strongly indicate that tyrosine phosphorylation of calponin h1 or h3 decreases the amounts of calponin bound to F-actin and imply that the affinity of each calponin can be modulated along the actin filament by tyrosine phosphorylation. Fig. 2. In vitr o phosphorylation of calponin h3 (A) and calponin h1 (B) by Fyn tyrosine kinase. Calponins (0.16 lg) in 30 lL of a kinase assay consisting of 25 m M Hepes, pH 7.5, 1 m M dithiothreitol, 12 m M MnCl 2 ,0.16lCi of [ 32 P]dATP[cP] and 5 l M ATP were treated for 60 min at 30 °C with the indicated concentrations of Fyn kinase. The phosphorylation level of each calponin was monitored by autoradio- graphy followed by PhosphorImager analyses. Data representing at least five independent experiments are shown. 2618 J. Abouzaglou et al. (Eur. J. Biochem. 271) Ó FEBS 2004 We did not extend the present study to calponin h2 because this particular calponin isoform includes a mutated inactive primary actin-binding site [5], which precludes any comparison of the functional effects of its phosphorylation relative to those found with the two other calponins. Next, we examined the influence of F-actin on the extent of phosphorylation of calponin h1 and calponin h3 by Fyn. In order to ensure detection of any blocking effect of F-actin, the phosphorylation reaction with [ 32 P]dATP[cP] was performed for 15 min instead of 30 min, and with the use of two molar ratios of calponin/actin corresponding to 1 : 2 or 1 : 10. After gel electrophoresis, the radioactivity associated with each protein was revealed by autoradiog- raphy (Fig. 5) and quantified by PhosphorImager analysis (Table 1). The phosphorylation of the complexes between F-actin and calponin h1 (Fig. 5, lane c) or calponin h3 (Fig. 5, lane d) led to the incorporation of c 32 P, not only into each calponin but also into actin. The radiolabeling of actin by Fyn kinase was not unexpected because, previously, actin has been shown to be phosphorylated in vivo in Dictyostelium [29–31] and plant cells [32] by nonidentified tyrosine protein kinases with a unique target site, identified as the tyrosine residue at position 53 [31]. Our quantitative radioactivity measurements, depicted in Table 1, show that nearly 1 mol of 32 P was incorporated per mol of actin, both in the absence and presence of calponins. We conclude that F-actin in solution is also an excellent substrate for Fyn kinase, which probably phosphorylates the protein exclu- sively on Tyr53. Furthermore, this efficient actin phos- phorylation can be considered as an internal marker pinpointing the good experimental conditions of our Fyn kinase assays. On the other hand, the data of Table 1 further show that actin not only failed to hinder the phosphorylation of both calponins, but even caused an increase in the phosphorylation levels of each calponin isoform as compared to the level of the control calponins, especially when a calponin/actin molar ratio of 1 : 10 was employed. The observed shift, from 0.7 to 2.0 mol of 32 Pper mol of calponin h1, and from 1.4 to 3.0 mol of 32 Ppermol of calponin h3, probably results from an actin-induced conformational change in the calponin molecule, which enhances the exposure of phosphorylatable tyrosine resi- dues to Fyn kinase. This conclusion is supported by control cosedimentation experiments (data not shown) using phos- phorylated actin and native calponin h1 or calponin h3, which did not show a noticeable change in the binding of either calponin to the modified actin. Thus, the actin- stimulated phosphorylation of calponins we observed was not the result of a nonspecific alteration of the actin structure caused by its co-phosphorylation. Identification of the tyrosine phosphorylation sites of calponin h3 and calponin h1 To determine the major sites of calponin h3 and calponin h1 that are phosphorylated by Fyn, we first subjected the tryptic in-gel digest of each 32 P-labeled protein to fractionation by HPLC. By radioactivity counting of the eluted fractions, we identified a single major radioactive peak in the digest of calponin h3 (Fig. 6A) and two radioactive peaks, designated a and b, in the digest of calponin h1 (Fig. 6B). Each peptide peak was subse- quently processed for radio-Edman sequencing and, after further HPLC purification, each posphopeptide was submitted to N-terminal sequencing. In the peak from Fig. 3. Electrophoretic profiles on a 5–18% gradient polyacrylamide gel of 32 P-labeled calponin h3 and calponin h1 after autoradiography (B). These profiles were compared with the patterns of control calponins on the same gel stained with Coomassie Blue R-250 (A). (C) Autoradio- grams of phosphorylated calponin h1 and calponin h3 before (lanes a and c, respectively) and after treatment with YOP protein tyrosine phosphatase, as described in the Materials and methods (lanes b and d, respectively). (D) Anti-(phosphotyrosine) probed Western blots of phosphorylated (lanes b and d) and unphosphorylated (lanes a and c) calponins. The experimental conditions of the immunoblots were as reported in the Materials and methods. Ó FEBS 2004 Tyrosine phosphorylated calponin–actin interaction (Eur. J. Biochem. 271) 2619 calponin h3, a single phosphotyrosine was found in cycle 5 and the N-terminal segment of the corresponding pure peptide was identified as GMSV. The only amino acid sequence of rat calponin h3 [5] accounting for these results is G257-R265, with the phosphorylation site on Tyr261. In the peak a of calponin h1, one 32 P-labeled tyrosine was identified in cycle 10 and the N-terminal end of the purified peptide was determined as GASQQG. Conse- quently, the corresponding calponin h1 sequence should be G252-R265 and the phosphorylated tyrosine is also localized at Tyr261, a position which is conserved in all the calponin sequences. Finally, the same analyses carried out on peak b led to the identification of the calponin h1 sequence, F173-R185, with Tyr182 as the phosphorylated residue. The presence of one and two phosphopeptides in calponin h3 and calponin h1, respectively, is essentially in agreement with the observed stoichiometry of 32 P incor- poration in each protein. Fig. 4. Tyrosine phosphorylation of calponins inhibits their binding to F-actin. Samples (5 lg) of calponin h1 or calponin h3, either unphosphorylated (A) or phosphorylated with nonradioactive ATP (B), were incubated with F-actin (molar ratio 1 : 2) in 25 m M Hepes, pH 7.5, 1 m M dithiothreitol, 12 m M MnCl 2 for 30 min at 30 °C and then subjected to high-speed centrifugation. The supernatant and pellet fractions were separated by gel electrophoresis and the partioning of calponin h1 (lanes 1) and calponin h3 (lanes 2) between the supernatant and pellet fractions was determined after staining with Coomassie Blue R-250. (C) The gels were analyzed by densitometry and the percentage of the control calponin h1 (lanes a) or calponin h3 (lanes c), together with the percentage of phosphorylated calponin h1 (lanes b) or calponin h3 (lanes d) in either fraction were calculated. The data represent the mean of three experiments. Hatched bars, calponin h1; dark bars, calponin h3. 2620 J. Abouzaglou et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Discussion In this study, we first investigated the enzymatic tyrosine phosphorylation of the calponin h3 isoform in COS 7 cells before and after transfection with the pSV vector containing the cDNA encoding the tyrosine kinase, Fyn. Tyrosine phosphorylated calponin was detected in both cases, with a marked increase of the phosphotyrosine content of the endogenous calponin upon transfection of the cells. These results strongly suggest that tyrosine phosphorylation of calponin h3 by Src kinases can occur in vivo and its biological effects remain to be elucidated. To assess the functional consequences of this phosphorylation on the F-actin–calponin interaction, purified calponin h3 and calponin h1 in solution were phosphorylated by Fyn kinase. Our data revealed that both calponin isoforms are conveni- ent substrates for specific tyrosine phosphorylation by this kinase, with concomitant inhibition of their binding to F-actin. However, although a stoichiometric quantity of phosphotyrosine was formed in each calponin, the complete inhibition of actin binding to any phosphorylated calponin did not occur, as the maximal reduction of actin binding observed was 75%. It seems likely that tyrosine phosphory- lation modulates the amount of calponin bound to actin, probably by lowering the affinity of the phosphocalponin– F-actin complex, rather than by fully dissociating the Fig. 5. Tyrosine phosphorylation of calponins is enhanced by F-actin. Calponin h1 or calponin h3 were phosphorylated in the absence (lanes a and b, respectively) or presence (lanes c and d, respectively) of F-actin at a molar ratio of 1 : 2. The experimental conditions for the phos- phorylation reaction were as reported in the Materials and methods. After gel electrophoresis and autoradiography, the radioactivity was measured by PhosphorImager analysis. The data are presented in Table 1. Table 1. Compared phosphorylation of calponins by Fyn kinase per- formed in the absence or presence of varying concentrations of F-actin. The experimental conditions for the phosphorylation reactions and radioactivity measurements were as described in the Materials and methods. Proteins Moles of 32 P incorporated per mol of protein Calponin h1 0.7 Calponin h3 1.4 Calponin h1 (+ F-actin at a molar ratio of 1 : 2) 1.0 Calponin h1 (+ F-actin at a molar ratio of 1 : 10) 2.0 Calponin h3 (+ F-actin at a molar ratio of 1 : 2) 1.8 Calponin h3 (+ F-actin at a molar ratio of 1 : 10) 3.0 Actin 0.9 a a Value found both for the control actin alone and for actin complexed to either calponin. Fig. 6. Separation of 32 P-labeled tryptic peptides of calponin h3 (A) and calponin h1 (B) by reverse-phase HPLC. After in-gel digestion of each phosphorylated protein with trypsin, the corresponding tryptic peptide mixtures were extracted and applied to an Aquapore C8 reverse-phase column, which was then eluted with a linear gradient of 0–30% acetonitrile for 60 min. The radioactive peptides were separated and their amino acid sequences were determined by combining their radio- Edman degradation with their direct N-terminal sequencing. Phos- phorylated Tyr261 was identified in the phosphopeptide representing the major peak of calponin h3 and in the phosphopeptide corres- ponding to peak a of calponin h1. Ó FEBS 2004 Tyrosine phosphorylated calponin–actin interaction (Eur. J. Biochem. 271) 2621 complex. An essentially similar modulation of the interac- tion of tyrosine phosphorylated villin with F-actin has been reported previously [16]. The association of F-actin with either calponin h1 or h3 during the kinase reaction did not hinder the phosphorylation process of the calponins; however, it increased their phosphotyrosine content. The lack of F-actin protection against tyrosine phosphorylation of calponins contrasts with the F-actin-induced blocking effect on the enzymatic Ser/Thr phosphorylation of calpo- nin h1 in vitro, reported previously [33]. On the other hand, the observed extent of actin-induced phosphorylation was different for the two calponins as, for calponin h1 it reached the same level as found in the absence of actin, whereas for calponin h3 it significantly exceeded the level obtained without actin. Thus, the actin-bound conformations of the two calponins are probably different. The F-actin-depend- ent tyrosine phosphorylation of calponin h3 could be of biological significance as the additional phosphorylated sites may also contribute to modulate actin binding. Further investigations are currently underway for assessing its influence on the actin–calponin interaction and for identi- fying the sites specifically stimulated by actin binding. In the work presented here, we have localized the tyrosines phosphorylated without actin to Tyr261 in calponin h3 and to Tyr261 and Tyr182 in calponin h1. The latter residue is within the first calponin repeat and represents a chymotrypsin cleavage site residing at the C-terminus of the central, protease-ensitive segment of calponin h1 that starts at Tyr144. Although being conserved in calponin h3, Tyr182 did not undergo phosphorylation in this isoform. Probably, the protein structure around this particular residue is different in calponin h3 from that of calponin h1. This suggestion is supported by the differences observed in the chymotryptic cleavage patterns of the two calponin isoforms. It should be noted that its phosphory- lation did not have any effect on F-actin binding as the extent of inhibition of the calponin–actin interaction was quite similar for the two phosphorylated isoforms. Thus, Tyr261 in calponin h3 or calponin h1 represents the only site whose phosphorylation promotes the observed inhibition of the protein binding to F-actin. Tyr261 is positioned one amino acid apart from Thr259, which was shown to be phosphorylated in calponin h1 by Rho-kinase [34]. The amino acid sequence around these two residues is highly conserved, being GMTVY261GL in chicken gizzard calpo- nin h1 and GMSVY261GL in rat calponin h3, the latter containing a serine residue instead of a threonine at position 259 [5]. Furthermore, Tyr261 is located in a strategic region of the calponin molecule, namely the conserved third repeat motif in the C-terminal portion of the protein. It is also a specific residue of calponins, as in the thin filament associated Unc87 protein from Caenorhabditis elegans, which contains seven copies of the calponin repeat, phenyl- alanine replaces the tyrosine residue in all repeats [35]. In the reported 3D reconstruction of the F-actin–calponin com- plex, the position of the calponin repeats relative to the actin–calponin interface has not been determined [36]. However, the Unc87 repeats were shown to directly bind to F-actin, both in vitro and in vivo [37]. In addition, the direct involvement of the calponin repeats in the stabilization of the actin filaments was recently described [38]. Together, these findings support the conclusion that the third calponin repeat motif, comprising Tyr261, may directly contribute to the building of a stable calponin–actin interface. Conse- quently, phosphorylation of the critical Tyr261 would cause a conformational change in the repeat domain with a marked reduction of the binding strength at the calponin– actin interface, thereby promoting the observed partial dissociation of the complex between F-actin and phosphor- ylated calponins. Previously, the association of F-actin with calponin repeats was found to be regulated by the tail region of calponin in an isoform-specific manner [39]. The present investigation indicates that it could be further modulated by direct tyrosine phosphorylation of the repeats. In conclusion, our studies open the way for the characterization of tyrosine phosphorylated calponins in other cell types using the Fyn cDNA transfection approach and the determination of their roles in the regulation in vivo of the structural dynamics and the interactions of F-actin. Acknowledgements This research was supported by grants from the Centre National de la Recherche Scientifique, the Institut National de la Sante ´ et de la Recherche Me ´ dicale, and the Association Franc¸ aise contre les myopathies. We thank J. L. Derancourt for kindly performing the structural analyses of tryptic phosphopeptides and Dr S. Roche for providing us with the PSV–Fyn construct and the Fyn2 antibody. References 1.Gimona,M.,Herzog,M.,Vandekerckhove,J.&Small,J.V. (1990) Smooth muscle specific expression of calponin. FEBS Lett. 274, 159–162. 2. Strasser,P.,Gimona,M.,Moessler,H.,Herzog,M.&Small,J.V. (1993) Mammalian calponin. Identification and expression of genetic variants. FEBS Lett. 330, 13–18. 3. Applegate, D., Feng, W., Green, R.S. & Taubman, M.B. (1994) Cloning and expression of a novel acidic calponin isoform from rat aortic vascular smooth muscle. J. Biol. Chem. 269, 10683–10690. 4. Winder, S.J. & Walsh, M.P. (1990) Smooth muscle calponin. Inhibition of actomyosin MgATPase and regulation by phos- phorylation. J. Biol. Chem. 265, 10148–10155. 5. Gimona, M. & Small, J.V. (1996) Calponin. In Biochemistry of Smooth Muscle Contraction, Vol 7 (Barany, M., ed), pp. 91–103. Academic Press, New York. 6. North, A., Gimona, M., Cross, R.A. & Small, J.V. (1994) Cal- ponin is localised in both the contractile apparatus and the cytoskeleton of smooth muscle cells. J. Cell Sci. 107, 437–444. 7. Kake, T., Kimura, S., Takahashi, K. & Maruyama, K. (1995) Calponin induces actin polymerization at low ionic strength and inhibits depolymerization of actin filaments. Biochem. J. 312, 587–592. 8. Leinweber, B., Tang, J.X., Stafford, W.F. & Chalovich, J.M. (1999) Calponin interaction with alpha-actinin-actin: evidence for a structural role for calponin. Biophys. J. 77, 3208–3217. 9.Menice,C.B.,Hulvershorn,J.,Adam,L.P.,Wang,C L.A.& Morgan, K.G. (1997) Calponin and mitogen-activated protein kinase signaling in differentiated vascular smooth muscle. J. Biol. Chem. 272, 25157–25161. 10. Leinweber, B., Parissenti, A., Gallant, C., Gangopadhyay, S.S., Kirwan-Rhude, A., Leavis, P.C. & Morgan, K.G. (2000) Reg- ulation of protein kinase C by the cytoskeletal protein calponin. J. Biol. Chem. 257, 40329–40336. 2622 J. Abouzaglou et al. (Eur. J. Biochem. 271) Ó FEBS 2004 11. Gimona, M. & Mital, R. (1998) The single CH domain of calponin is neither sufficient nor necessary for F-actin binding. J. Cell Sci. 111, 1813–1821. 12. Mezgueldi,M.,Mendre,C.,Calas,B.,Kassab,R.&Fattoum,A. (1995) Characterization of the regulatory domain of gizzard calponin. Interaction of the 145–163 region with F-actin, calcium- binding protein, and tropomyosin. J. Biol. Chem. 270, 8867–8876. 13. Burgstaller, G., Kranewitter, W.J. & Gimona, M. (2002) The molecular basis for the autoregulation of calponin by isoform- specific C-terminal tail sequences. J. Cell Sci. 115, 2021–2029. 14. Nakamura,F.,Mino,T.,Yamamoto,J.,Naka,M.&Tanaka,T. (1993) Identification of the regulatory site in smooth muscle cal- ponin that is phosphorylated by protein kinase C. J. Biol. Chem. 268, 6194–6201. 15. Tang, D C., Kang, H M., Jin, J P., Fraser, E.D. & Walsh, M.P. (1996) Structure–function relations of smooth muscle calponin. The critical role of serine 175. J. Biol. Chem. 271, 8605–8611. 16. Zhai, L., Zhao, P., Panebra, A., Guerrerio, A.L. & Khurana, S. (2001) Tyrosine phosphorylation of villin regulates the organiza- tion of the actin-cytoskeleton. J. Biol. Chem. 276, 36163–36167. 17.Zhai,L.,Kumar,N.,Panebra,A.,Zhao,P.,Parryl,A.L.& Khurana, S. (2002) Regulation of actin dynamics by tyrosine- phosphorylation: identification of tyrosine phosphorylation sites within the actin-severing domain of villin. Biochemistry 41, 11750–11760. 18. DeCorte,V.,Gettemans,J.&Vandekerckhove,J.(1997)Phos- phatidylinositol 4,5-bisphosphate specifically stimulates PP60 (c-src) catalyzed phosphorylation of gelsolin and related actin- binding proteins. FEBS Lett. 401, 191–196. 19.Wang,Q.,Xie,Y.,Du,Q S.,Wu,X J.,Feng,X.,Mei,L., McDonald, J.M. & Xiong, W C. (2003) Regulation of the for- mation of osteoclastic actin rings by proline-rich tyrosine kinase 2 interacting with gelsolin. J. Cell Biol. 160, 565–575. 20. Huang,C.,Ni,Y.,Wang,T.,Gao,Y.,Haudenschild,C.C.& Zhan, X. (1997) Down-regulation of the filamentous actin cross- linking activity of cortactin by src-mediated tyrosine phosphor- ylation. J. Biol. Chem. 272, 13911–13915. 21. Mezgueldi, M., Fattoum, A., Derancourt, J. & Kassab, R. (1992) Mapping of the functional domains in the amino-terminal region of calponin. J. Biol. Chem. 267, 15943–15951. 22. Fattoum, A., Roustan, C., Smyczynski, C., Der Terrossian, E. & Kassab, R. (2003) Mapping the microtubule binding regions of calponin. Biochemistry 42, 1274–1282. 23. Roche, S., Koegl, M., Barone, M.V., Roussel, M.F. & Court- neidge, S.A. (1995) DNA synthesis induced by some but not all growth factors requires Src family protein tyrosine kinases. Mol. Cell. Biol. 15, 1102–1109. 24. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680– 685. 25. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. & Struhl, K. (1994) Current Proto- cols in Molecular Biology. Greene Publishing Associates/Wiley, New York. 26. Plantier, M., Fattoum, A., Menn, A., Ben-Ari, Y., Der Terrossian, E. & Represa, A. (1999) Acidic calponin immunoreactivity in postnatal rat brain and cultures: subcellular localization in growth cones, under the plasma membrane and along actin and glial filaments. Eur. J. Neurosci. 11, 2801–2812. 27. Rosenfeld, J., Capdevielle, J., Guillemot, J.C. & Ferrara, P. (1992) In-gel digestion of proteins for internal sequence analysis after one or two-dimentional gel electrophoresis. Anal. Biochem. 203, 173–179. 28. Thomas, S.M. & Brugge, J.S. (1997) Cellular functions regulated by Src family kinases. Annu. Rev. Cell Dev. Biol. 13, 513–609. 29. Schweiger, A., Mihalache, O., Ecke, M. & Gerisch, G. (1992) Stage-specific tyrosine phosphorylation of actin in Dictyostelium discoideum cells. J. Cell Sci. 102, 601–609. 30. Howard, P.K., Sefton, B.M. & Firtel, R.A. (1993) Tyrosine phoshorylation of actin in Dictyostelium associated with cell-shape changes. Science 259, 241–244. 31. Jungbluth, A., Eckerskorn, C., Gerish, G., Lottspeich, F., Stocker, S. & Schweiger, A. (1995) Stress-induced tyrosine phosphorylation of actin in Dictyostelium cells and localization of the phosphor- ylation site to tyrosine-53 adjacent to the DnaseI binding loop. FEBS Lett. 375, 87–90. 32. Kameyama, K., Kishi, Y., Yoshimura, M., Kanzawa, N., Same- shima, M. & Fsuchiya, T. (2000) Tyrosine phosphorylation in plant bending. Nature 407, 37–40. 33. Nagumo, H., Sakurada, K., Seto, M. & Sasaki, Y. (1994) Phos- phorylation of calponin by PKC is blocked by F-actin in vitro. Biochem. Biophys. Res. Commun. 203, 1502–1507. 34. Kaneko, T., Amano, M., Maeda, A., Goto, H., Takahashi, M.I. & Kaibuchi, K. (2000) Identification of calponin as a novel substrate of Rho-kinase. Biochem. Biophys. Res. Commun. 273, 110–116. 35. Goetinck, S. & Waterson, R.H. (1994) The Caenorhabditis elegans muscle-affecting gene, unc-87, encodes a novel thin filament- associated protein. J. Cell Biol. 127, 79–93. 36. Bramham, J., Hodgkinson, J.L., Smith, B.O., Uhrin, D., Barlow, P.N. & Winder, S.J. (2002) Solution structure of the Calponin CH domain and fitting to the 3D-helical reconstruction of F-actin calponin. Structure 10, 249–258. 37. Kranewitter, W.J., Ylanne, J. & Gimona, M. (2001) Unc-87 is an actin bundling protein. J. Biol. Chem. 276, 6306–6312. 38. Gimona, M., Kaverina, I., Resch, G.P., Vignal, E. & Burgstaller, G. (2003) Calponin repeats regulate actin filament stability and formation of Podosomes in smooth muscle cells. Mol. Biol. Cell 14, 2482–2491. 39. Burgstaller, G., Kranewitter, W.J. & Gimona, M. (2002) The molecular basis for the autoregulation of calponin by isoform- specific C-terminal tail sequences. J. Cell Sci. 115, 2021–2029. Ó FEBS 2004 Tyrosine phosphorylated calponin–actin interaction (Eur. J. Biochem. 271) 2623 . increased their phosphotyrosine content. The lack of F-actin protection against tyrosine phosphorylation of calponins contrasts with the F-actin- induced blocking effect on the enzymatic Ser/Thr phosphorylation. functional effects of its phosphorylation relative to those found with the two other calponins. Next, we examined the influence of F-actin on the extent of phosphorylation of calponin h1 and calponin. villin with F-actin has been reported previously [16]. The association of F-actin with either calponin h1 or h3 during the kinase reaction did not hinder the phosphorylation process of the calponins; however,