Structure-function relations of smooth muscle calponin. The critical role of serine 175.

Calponin has been implicated in the regulation of smooth muscle contraction through its interaction with F-actin and inhibition of the actin-activated MgATPase activity of phosphorylated myosin. Both properties are lost following phosphorylation (primarily at serine 175) by protein kinase C or calmodulin-dependent protein kinase II. To evaluate further the functional importance of serine 175, wild-type calponin and three site-specific mutants (S175A, S175D, and S175T) were expressed in Escherichia coli and compared with calponin purified from chicken gizzard smooth muscle in terms of actin binding, actomyosin MgATPase inhibition, and phosphorylation by protein kinase C and calmodulin-dependent protein kinase II. The affinities of skeletal muscle F-actin for wild-type and S175T calponins were similar to that for the tissue-purified protein (K = 0.8, 1.3, and 1.0 μM, respectively), whereas the affinities for S175A and S175D calponins were much lower (K = 26.8 and 44.2 μM, respectively). Tissue-purified, wild-type, and S175T calponins displayed comparable inhibition of the smooth muscle actin-activated myosin MgATPase, whereas S175A and S175D calponins were much less effective. Phosphorylation confirmed serine 175 as the principal site of phosphorylation by both kinases. These results indicate that the hydroxyl side chain at position 175 of calponin plays a critical role in the binding of calponin to actin and inhibition of the cross-bridge cycling rate.

Calponin is a 32-kDa, thin filament-associated protein that has been implicated in the regulation of smooth muscle contraction (1)(2)(3)(4). Through its interaction with actin, calponin inhibits the actin-activated MgATPase activity of phosphorylated smooth muscle myosin (1) and inhibits the relative movement of actin and myosin in in vitro motility assays (5,6). Exogenous calponin attenuated Ca 2ϩ -induced contractions of permeabilized rabbit mesenteric arterial smooth muscle strips in a concentration-dependent manner (7). Inhibition of the actomyosin ATPase occurs predominantly by a reduction of V max rather than an effect on the affinity of phosphorylated myosin heads for actin (8 -10). The inhibitory effect of calponin can be alleviated by phosphorylation catalyzed by protein kinase C (PKC) 1 or Ca 2ϩ /calmodulin-dependent protein kinase II (CaM kinase II) (1,11) and restored following dephosphorylation by type 2A (12) or 2B protein phosphatases (13). Phosphorylation-induced loss of inhibition results from a marked reduction in the affinity of phosphorylated calponin for actin (1). Several examples of phosphorylation of calponin in intact muscle in response to various contractile stimuli have been reported (11, 14 -16), and in response to phenylephrine (an ␣ 1 -adrenergic agonist known to trigger activation of PKC) calponin translocates from cytosolic filamentous structures to a region underlying the sarcolemma (17). Other investigators, however, have reported that calponin is not phosphorylated in intact muscle (18 -20). The principal site of phosphorylation by either kinase is serine 175 (11), although Nakamura et al. (21) suggested that threonine 184 was the principal residue phosphorylated by PKC. Mezgueldi et al. (22) have defined the actin-binding domain of calponin as residues 145-182, which includes serine 175.
To investigate further the importance of serine 175 in the interaction of calponin with actin and inhibition of the actinactivated myosin MgATPase, we have expressed in bacteria wild-type calponin and three site-specific mutants in which serine 175 was replaced by alanine, aspartic acid, or threonine. The following properties of the expressed proteins were compared with calponin purified from chicken gizzard smooth muscle: (i) protein conformation, (ii) actin binding, (iii) inhibition of actin-activated myosin MgATPase activity, and (iv) phosphorylation by PKC and CaM kinase II. The results indicate that serine 175 plays a critical role in the interaction of calponin with actin and ATPase inhibition; in particular, a hydroxyl group in this position is essential for these functions. The results also confirm serine 175 as the principal site of phosphorylation by PKC and CaM kinase II.

EXPERIMENTAL PROCEDURES
Materials-[␥-32 P]ATP (Ͼ5000 Ci/mmol) was purchased from Amersham Corp. (Oakville, Ontario, Canada). L-␣-Phosphatidyl-L-serine and 1,2-diolein were purchased from Serdary Research Laboratories (London, Ontario, Canada). Urea was purchased from Serva Feinbiochemica Hanstoff (Heidelberg, Germany). PKC (a mixture of ␣, ␤, and ␥ isoenzymes) was purified from rat brain as described previously (23). Myosin light chain kinase (24), actin (25), myosin (26), tropomyosin (27), and calponin (1) were purified from chicken gizzard smooth muscle. CaM kinase II was co-purified from chicken gizzard with caldesmon because this preparation of the kinase was found to be much more stable than the isolated kinase (28). Actin was also purified from rabbit skeletal muscle as described by Zot and Potter (29). Calmodulin was purified from bovine brain (30). Monoclonal antibody CP-93 to turkey gizzard calponin was purchased from Sigma. Oligonucleotide primers were synthesized at the University Core DNA Services, University of Calgary. Electrophoresis reagents were purchased from Bio-Rad. All other chemicals were reagent grade or better and were purchased from Can Lab (Edmonton, Alberta, Canada).
Reverse Transcription-PCR-Chicken gizzard mRNA (5 g) was reverse transcribed at 52°C for 30 min with Superscript RNase H Ϫ Reverse Transcriptase (Life Technologies, Inc.) using an oligo(dT) primer. PCR was performed in a Perkin-Elmer System 2400 DNA Thermal Cycler using the following primers: 5Ј-ATGTCGAACGCGA-ACTTCAACCGC-3Ј (5Ј primer) and 5Ј-GGAATTCCTTATTGTGAGTT-GTAAAAGCTGTGGTTGAGGCC-3Ј (3Ј primer). These sequences are derived from the nucleotide sequence of calponin reported by Takahashi and Nadal-Ginard (31) with an EcoRI site incorporated at the 5Ј end of the 3Ј primer. The 50-l reaction mixture contained 50 mM Tris-HCl (pH 8.4), 50 mM KCl, 2.5 mM MgCl 2 , 0.6 mM of each dNTP, 1 M of each primer, AmpTaq polymerase (Perkin-Elmer), and cDNA templates (the reverse transcription products). Denaturation of the sample was affected at 94°C for 1 min followed by 35 cycles of 1 min of denaturation at 94°C, 1 min annealing at 55°C, and 2 min of extension at 72°C. A single PCR product of the expected size for ␣ calponin (879 base pairs) was produced, as shown by agarose gel electrophoresis. This cDNA hybridized with a unique 1.3 kb mRNA upon Northern blot analysis with chicken gizzard mRNA as expected (31). The PCR product was subcloned into pBSII(SKϩ) for sequencing using the Perkin-Elmer/ Applied Biosystems, Inc. (Mississauga, Ontario, Canada) automated fluorescent sequencing methodology and Genetics Computer Group, Inc. (Madison, WI) sequence analysis software. The nucleotide sequence was identical to that reported by Takahashi and Nadal-Ginard (31) except for four nucleotide substitutions: nucleotides 78, 95, 96, and 197 were C, G, T and A, respectively, rather than T, T, G, and C. We have confirmed our sequence in several independent experiments, and Gong et al. (32) have independently confirmed nucleotides C and A at positions 78 and 197, respectively. Only one of the four nucleotide changes alters the encoded amino acid (nucleotide 197): threonine 66 becomes lysine, in agreement with the peptide sequences reported by Mezgueldi et al. (22) and Vancompernolle et al. (33) and our own peptide sequencing data (3).
PCR Mutagenesis-S175A, S175D, and S175T mutants were produced by PCR mutagenesis using the overlap extension method (34). The complete nucleotide sequences of all mutants were confirmed to ensure that only the desired mutations were obtained.
Bacterial Expression and Protein Purification-Wild-type and mutant calponins were expressed in Escherichia coli and purified from bacterial lysates essentially as described by Gong et al. (32) by sequential chromatography on columns of CM-Sephadex, hydroxylapatite, and Superose 12. Calponin-containing fractions were identified by SDS-PAGE in 12.5% acrylamide mini-slab gels (35). The purified proteins, shown to be electrophoretically homogeneous, were concentrated by dialysis versus 30% polyethylene glycol 20,000 in 20 mM Tris-HCl (pH 7.5), 0.1 M KCl, 1 mM dithiothreitol and stored at Ϫ80°C. Wild-type, S175A, S175D, and S175T calponins each exhibited a molecular mass of 32 kDa by 0.1% SDS/7.5-20% polyacrylamide gradient slab gel electrophoresis and comigrated with calponin isolated from chicken gizzard. All proteins were recognized by monoclonal antibody CP-93 to turkey gizzard calponin by Western blotting carried out as described previously (36). Yields of all recombinant proteins were ϳ10 mg/l of bacterial culture.
Actin Binding-Various concentrations of calponin (1.75-4.5 M) were incubated at 20°C for 60 min with 11 M skeletal or smooth muscle actin with or without 2 M smooth muscle tropomyosin in Buffer A (20 mM Tris-HCl, pH 7.5, 0.1 M KCl, 2 mM MgCl 2 , 1 mM ATP, 1 mM dithiothreitol, 0.1 mM CaCl 2 ). The mixture (0.1 ml) was then centrifuged at 100,000 ϫ g for 60 min at 4°C to separate supernatant and pellet. SDS gel sample buffer (0.1 ml) was added to the supernatant. The pellet was resuspended in 0.1 ml of Buffer A, and SDS gel sample buffer (0.1 ml) was added. Samples were boiled for 2 min prior to SDS-PAGE in full-sized (12 ϫ 13.5 cm) 7.5-20% polyacrylamide gradient slab gels (35). Calponin and actin in the supernatant and pellet fractions were quantified by densitometric scanning of the Coomassie Blue-stained gels using a Pharmacia Image Master Desktop Scanning System. This protocol separates F-actin and bound calponin and tropomyosin (recovered in the pellet) from G-actin and free calponin and tropomyosin (recovered in the supernatant). Calponin alone does not sediment under these conditions (11). K d values for calponin binding to actin were determined from Scatchard plots of the binding data (for tissue-purified, wild-type, or S175T calponin) or were estimated by binding measurements carried out at high calponin concentrations (for S175A and S175D calponins).
Actin-activated Myosin MgATPase Assay-Smooth muscle actin (6 M), myosin (1 M), tropomyosin (1.15 M), calmodulin (0.6 M), and myosin light chain kinase (0.1 M) were incubated at 30°C in the absence and the presence of tissue-purified, wild-type, and mutant calponins at concentrations indicated in the text in 25 mM Tris-HCl (pH 7.5), 60 mM KCl, 10 mM MgCl 2 , 0.2 mM CaCl 2 , 1 mM dithiothreitol, and 1 mM [␥-32 P]ATP (3,500 -5,300 cpm/pmol) in a reaction volume of 0.6 ml. Reactions were started by the addition of ATP, and samples (50 l) were withdrawn at 1-min intervals up to 9 min for quantification of [ 32 P]phosphate released as described previously (37). ATPase rates in the absence of calponin were 149.8 Ϯ 38.9 nmol P i min Ϫ1 ⅐mg myosin Ϫ1 (mean Ϯ S.D., n ϭ 18) in the presence of Ca 2ϩ and 8.4 Ϯ 5.5 nmol P i min Ϫ1 ⅐mg myosin Ϫ1 (n ϭ 10) in the absence of Ca 2ϩ (1 mM EGTA replaced 0.2 mM CaCl 2 in the ATPase reaction mixture).
Monoclonal Antibody Epitope Competition Assays-Four monoclonal antibodies against different epitopes on chicken gizzard calponin 2 were used in an enzyme-linked immunosorbant assay-mediated competition experiment to evaluate the structural conservation of the mutant calponins under native conditions. Recombinant wild-type calponin was coated on 96-well microtiter plates (Falcon 3915) at 5 g/ml in 50 mM carbonate buffer (pH 9.6), 100 l/well, and incubated overnight as described previously (39) at 4°C. After blocking with 1% bovine serum albumin in phosphate-buffered saline (137 mM NaCl, 2.7 mM KH 2 PO 4 , 8 mM Na 2 HPO 4 , pH 7.4) containing 0.05% Tween-20 (PBS-T) and washing three times with PBS-T, a constant concentration (predetermined in the indirect enzyme-linked immunosorbant assay to give a binding in the upper one-third of the linear range against the coated wild-type calponin) of anti-calponin monoclonal antibodies CP1, 3, 4, or 8 mixed with serial dilutions of the wild-type, S175A, or S175D proteins (0.001-10 g/ml) were added and incubated at 4°C for 6 h. Following three washes with PBS-T, horseradish peroxidase-labeled rabbit antimouse polyvalent immunoglobulin (Sigma) second antibody was added and incubated at 37°C for 1 h. Following three further washes with PBS-T, H 2 O 2 /2,2Ј-azinobis(3-ethylbenzthiazolinesulfonic acid) substrate was added and color developed at room temperature. The A 450 nm was recorded using a Bio-Rad model 3550 UV automated microplate reader at a series of time points, and the values in the linear range were plotted against the concentration of the blocking protein.
Protein Concentrations-Concentrations of calponin and its mutants were determined by amino acid analysis using endogenous phenylalanine and exogenous norleucine for quantification. PKC and myosin light chain kinase concentrations were determined by the Coomassie Blue dye binding assay with dye reagent and ␥-globulin standard purchased from Pierce. Other proteins were quantified using the following values for the absorbance of a 1% solution with a path length of 1 cm: myosin, 4.5 at 280 nm (40); tropomyosin, 2.9 at 278 nm (41), calmodulin, 1.95 at 277 nm (42); skeletal muscle actin, 6.3 at 290 nm (43); and smooth muscle actin, 6.38 at 290 nm (44).

RESULTS
Two experimental approaches were used to determine whether or not the conformations of the mutant recombinant calponins were similar to that of wild-type calponin or calponin purified from chicken gizzard smooth muscle: (i) limited proteolysis and (ii) binding of a panel of anti-calponin monoclonal antibodies with distinct epitopes under nondenaturing conditions. Fig. 1 compares the time courses of proteolysis of tissuepurified (A), wild-type (B), S175A (C), S175T (D), and S175D (E) calponins by chymotrypsin. No significant differences are seen between the patterns of proteolysis of the five calponin species. Similar results were obtained with protease K, trypsin, and S. aureus V8 protease (data not shown). As shown in Fig.  2, the time courses of digestion of phosphorylated and unphosphorylated calponins by protease K were not significantly different. Similar results were obtained with trypsin, chymotrypsin, and S. aureus V8 protease (data not shown). We conclude, therefore, that the conformations around the protease-sensitive sites in calponin are conserved in the bacterially expressed wild-type protein, the three site-specific mutants, and phosphorylated calponin. Any alterations in functional properties of the mutants can therefore be assigned directly to the mutation rather than to a gross conformational difference in the mutant protein. Fig. 3 shows competition binding curves illustrating the binding of four monoclonal antibodies to recombinant wild-type calponin and the two mutants, S175A and S175D, which exhibited functional differences compared with wild-type calponin (see below). The titration curves showed similar blocking patterns for all four monoclonal antibodies by the wild-type and S175A calponins, indicating no detectable structural differences. The S175D mutant showed slightly decreased affinity for monoclonal antibodies CP1, CP3, and CP4 as compared with that of the wild-type or S175A calponins, indicating a minor structural change induced by introduction of a negatively charged residue at position 175. In additional experiments, it was observed that these four and an additional three monoclonal antibodies gave similar binding patterns to solid phase-immobilized wild-type, S175A, and S175T calponins. However, the S175D mutant exhibited lower maximal binding to CP1, 3, 4, 5, and 7 (data not shown). The binding of bacterially expressed calponins to smooth muscle actin or actin/tropomyosin was evaluated by a co-sedimentation assay. Each calponin was mixed with chicken gizzard actin with or without tropomyosin as described under "Experimental Procedures" and centrifuged at high speed to separate F-actin and bound proteins from G-actin and unbound proteins. Calponin purified from chicken gizzard (4 M) bound to smooth muscle actin-tropomyosin as expected (Table I); 74% of the calponin was recovered in the pellet fraction under these conditions. Wild-type calponin and the S175T mutant showed similar binding to actin/tropomyosin; 71 and 69% of the respective calponins was recovered in the pellet. On the other hand, S175A and S175D showed much reduced binding to actin/ tropomyosin; 14% of S175A and 30% of S175D was recovered in the pellet. Similar results were obtained in the absence of tropomyosin; the proportions of wild-type calponin, S175A, S175D, and S175T mutant calponins recovered in the pellet associated with smooth muscle F-actin were 86, 5.6, 33, and 69%, respectively (means of two experiments). Similar results were also obtained with skeletal muscle actin (Table I); the proportions of tissue-purified calponin, wild-type calponin, S175A, S175D, and S175T mutant calponins recovered in the pellet were 71, 72, 28, 20, and 66%, respectively.
We noticed that the proportion of sedimented smooth muscle actin was reduced in the presence of the three calponin mutants compared with tissue-purified or wild-type calponin, whereas the sedimentation of skeletal muscle actin was unaffected by the different calponins (Table I), suggesting that the three mutant calponins may have a minor effect on smooth muscle actin polymerization or cause a small degree of actin filament severing producing short filaments that do not sediment upon high speed centrifugation. Consequently, skeletal muscle actin/smooth muscle tropomyosin was chosen as the system to quantify the affinities of actin for the various calponin species to facilitate accurate quantification of the amount of actin-bound calponin as the calponin concentration is varied over the range 1.75-4.5 M as described under "Experimental Procedures." The following K d values for actin/tropomyosin binding were determined from the Scatchard plots shown in Fig. 4; 1.0 M for calponin purified from chicken gizzard, 0.8 M for wild-type calponin, and 1.3 M for S175T. These values are not significantly different, indicating that the bacterially expressed wild-type protein and the S175T mutant bind to actin with the same affinity as the native protein. The S175A and S175D mutants, on the other hand, bound to actin with much lower affinity; K d ϭ 26.8 M for S175A and 44.2 M for S175D.
The effects of calponin and its mutants on the actin-activated MgATPase activity of smooth muscle myosin are compared in Fig. 5. As expected, calponin purified from chicken gizzard markedly inhibited the ATPase. A similar degree of inhibition FIG. 1. Partial proteolysis of tissue-purified, wild-type, and mutant calponins. Calponins were subjected to limited proteolysis by chymotrypsin as described under "Experimental Procedures." The time courses of digestion were analyzed by 0.1% SDS/7.5-20% polyacrylamide gradient slab gel electrophoresis. A, calponin purified from chicken gizzard; B, bacterially expressed wild-type calponin; C, S175A; D, S175T; E, S175D calponins. The numbers indicate times of digestion (min), and the arrowheads denote the four major fragments generated by proteolysis.
was observed with the bacterially expressed wild-type calponin (20.8% of control ATPase activity was retained at 4 M and 12.4% at 5 M calponin) and the S175T mutant (Fig. 5). The S175A and S175D mutants, however, were relatively ineffective in inhibiting the actomyosin ATPase, consistent with their much reduced affinities for actin. SDS-PAGE and autoradiography of samples at the end of the ATPase reactions verified that phosphorylation of the 20-kDa light chain of smooth mus-cle myosin was unaffected by any of the calponin species (data not shown). Fig. 6 compares the time courses of phosphorylation of the various calponin species by PKC. Tissue-purified calponin and bacterially expressed calponin exhibit similar phosphorylation time courses with maximal stoichiometry of ϳ1.1 mol P i /mol calponin under these conditions. No phosphorylation was observed in the absence of Ca 2ϩ , phospholipid, and diacylglycerol,  3. Epitope competition assays between wild-type recombinant, S175A, and S175D calponins and four specific monoclonal antibodies. Wild-type calponin was coated on microtiter plates and analyzed by enzyme-linked immunosorbant assay for its interaction with monoclonal antibodies CP1, 3, 4, and 8 in the presence of 0.001-10 g/ml serial dilutions of wild-type, S175A, or S175D calponins. The competition curves were plotted as the binding of monoclonal antibody to the immobilized wild-type calponin versus the concentration of competing protein in solution.
which are required by PKC (a mixture of ␣, ␤, and ␥ isoenzymes) for activity. Much less phosphate incorporation into S175A (ϳ0.3 mol P i /mol) or S175D (ϳ0.4 mol P i /mol) was observed, consistent with our earlier conclusion that serine 175 represents the principal site of phosphorylation by PKC. The S175T mutant is clearly a substrate of PKC, although the kinetics of its phosphorylation are much slower than for the native or wild-type protein.
In the case of phosphorylation by CaM kinase II (Fig. 7), again the native and wild-type proteins are very similar, and phosphorylation required the presence of Ca 2ϩ and calmodulin. The S175A and S175D mutants were very poor substrates of this enzyme, even poorer than for PKC. The S175T mutant, however, proved to be as good a substrate for CaM kinase II as the native or wild-type protein. DISCUSSION Serine 175 has previously been implicated in the binding of calponin to actin and inhibition of the actin-activated myosin MgATPase (11). Phosphorylation by PKC or CaM kinase II occurs predominantly at this site with substoichiometric phosphorylation occurring at other sites: Thr 180 , Thr 184 , Ser 215 , Ser 254 , and Thr 259 by PKC and Ser 71 , Ser 254 , and Ser 215 by  4. Scatchard analysis of the binding of tissue-purified, wild-type, and S175T calponins to F-actin. Actin binding of tissuepurified (A), wild-type (B), and S175T calponin (C) was quantified as described under "Experimental Procedures," and the results are presented as Scatchard plots.
FIG. 5. Concentration dependence of the effects of tissue-purified and mutant calponins on the actin-activated MgATPase activity of smooth muscle myosin. Actomyosin ATPase activities were measured as described under "Experimental Procedures" at the indicated concentrations of chicken gizzard calponin (E), S175A (Ç), S175D (f), and S175T (Ⅺ) calponins.
CaM kinase II. Ser 175 is located within the actin-binding domain (residues 145-182) defined by Mezgueldi et al. (22). It should be noted, however, that Nakamura et al. (21) concluded that Thr 184 is the major site of phosphorylation by PKC. To address the functional importance of Ser 175 directly, we have expressed and characterized three site-specific mutants of calponin and compared their properties with those of bacterially expressed wild-type calponin and calponin purified from chicken gizzard smooth muscle. Ser 175 was mutated (i) to alanine to assess the importance of the hydroxyl side chain and to determine whether or not this residue indeed represents the principal phosphorylation site, (ii) to aspartic acid to assess the functional consequences of introduction of a negative charge at this position (i.e. in an attempt to mimic, at least in part, the effects of phosphorylation), and (iii) to threonine to assess the importance of a hydroxylated side chain at this position and to determine whether or not PKC or CaM kinase II is capable of phosphorylating threonine as well as serine in this position.
Prior to functional characterization of the mutants, it was necessary to verify that bacterial expression (in particular, lack of post-translational modification) and site-specific mutagenesis did not affect the conformation of the protein. Two experimental approaches that are very sensitive to differences in protein conformation were used to address this issue: (i) analysis of the time courses of proteolytic digestion at 0°C by four proteases with distinct cleavage site specificities and (ii) binding to calponin under nondenaturing conditions of a panel of monoclonal antibodies with distinct epitopes. Both methods indicated that the conformations of recombinant wild-type, S175A, and S175T calponins are indistinguishable from that of the native tissue-purified protein. The S175D mutant appears to have a slightly different conformation, reflected by decreased affinity for specific monoclonal antibodies, although its overall conformation is indistinguishable from that of native calponin, as revealed by protease sensitivity.
Wild-type calponin expressed in E. coli bound to smooth and skeletal muscle actins similarly to the protein purified from chicken gizzard. The K d of skeletal muscle actin for wild-type calponin (0.8 M) was found to be similar to that for native calponin (1.0 M). Consistent with these observations, wildtype calponin inhibited the actin-activated MgATPase activity of smooth muscle myosin to a similar extent as did the native protein. Furthermore, phosphorylation of wild-type and tissuepurified calponins by PKC and CaM kinase II occurred with similar time courses and maximal stoichiometries. These results indicate that bacterially expressed wild-type calponin is structurally and functionally indistinguishable from the protein purified from smooth muscle, thus confirming and extending the work of Gong et al. (32).
The properties of the S175A mutant are particularly intriguing. This mutant calponin binds very weakly to smooth or skeletal muscle actins (the K d of skeletal muscle actin for S175A was estimated to be 26.8 M) and is correspondingly a very poor inhibitor of the actomyosin ATPase. These results imply that the hydroxyl side chain in position 175 plays a crucial role in the interaction of calponin with actin, this interaction mediating the inhibition of actomyosin ATPase activity (1). The C-terminal region of actin (residues 326 -355) has been implicated as the binding site for calponin (22), and the three most C-terminal residues appear to be particularly important for binding to calponin (45). Clearer understanding of the involvement of Ser 175 in the interaction with actin will come from determination of the three-dimensional structure of calponin and, ultimately, of the calponin-actin complex.
Phosphorylation of S175A calponin by PKC and CaM kinase II revealed incorporation of substoichiometric amounts of phosphate: 0.3 mol P i /mol with PKC and 0.1 mol P i /mol with CaM kinase II. This compares with stoichiometries of phosphorylation of native and wild-type calponins of 1.1 and 1.0 mol P i /mol with PKC and CaM kinase II, respectively. These results strongly support our earlier conclusion that Ser 175 represents the principal site of phosphorylation by these two kinases. The low stoichiometry of S175A phosphorylation presumably represents phosphate incorporation into minor, nonfunctional FIG. 6. Phosphorylation of tissue-purified, wild-type, and mutant calponins by PKC. Chicken gizzard (E), wild-type (q), S175A (Ç), S175D (f), and S175T (Ⅺ) calponins were incubated with PKC as described under "Experimental Procedures," and phosphate incorporation was quantified at the indicated times. Phosphorylation of the tissue-purified protein was also measured in the absence of Ca 2ϩ , phospholipid, and diacylglycerol (å). The results are representative of eight independent experiments. FIG. 7. Phosphorylation of tissue-purified, wild-type, and mutant calponins by CaM kinase II. A, chicken gizzard (E), wild-type (q), S175A (Ç), S175D (f), and S175T (Ⅺ) calponins were incubated with CaM kinase II as described under "Experimental Procedures," and phosphate incorporation was quantified at the indicated times. Phosphorylation of the tissue-purified protein was also measured in the absence of Ca 2ϩ and calmodulin (å). B, autoradiogram of duplicate samples subjected to SDS-PAGE showing the time courses of phosphorylation of caldesmon (CaD) and the indicated calponin species (CaP). Caldesmon served as a useful internal control. The results are representative of six independent experiments. CG, chicken gizzard. sites identified earlier (11).
The properties of the S175D mutant are very similar to those of S175A and substantiate the conclusions that Ser 175 plays a crucial role in binding to actin and that it represents the principal site of phosphorylation. As shown earlier, phosphorylation of calponin dramatically reduces its affinity for actin and alleviates inhibition of the actomyosin MgATPase (1). The similar properties of the S175D mutant might suggest that the introduction of negative charge at position 175 disrupts the calponin-actin interaction. However, the fact that the S175A mutant behaves very similarly to S175D and phosphorylated calponin indicates that the hydroxyl group itself is required for this interaction. The slightly decreased affinity of S175D calponin for monoclonal antibodies CP1, 3, and 4 suggests that introduction of negative charge at this position affects the conformation that together with loss of the hydroxyl group contributes to the marked reduction in affinity for actin. Similar effects probably account for the dramatic consequences of phosphorylation of calponin at Ser 175 .
The properties of the S175T mutant support the importance of a hydroxyl-containing side chain at position 175. This mutant binds to smooth and skeletal muscle actins similarly to the native and wild-type proteins and has a comparable K d for skeletal muscle actin of 1.3 M. Consequently, it inhibits the actin-activated myosin MgATPase to a similar extent as does the native or wild-type protein. Finally, the S175T mutant is as good a substrate for CaM kinase II as is tissue-purified or wild-type calponin but not as good a substrate for PKC, reflecting a preference of PKC for serine rather than threonine at the site of phosphorylation.
In conclusion, our results indicate that serine 175 of calponin plays a crucial role in the interaction of calponin with actin and inhibition of the actin-activated MgATPase activity of smooth muscle myosin and that this residue is the principal site of phosphorylation by PKC and CaM kinase II. In particular, the integrity of the hydroxyl group in the side chain at position 175 is critically important for high affinity binding to actin.