Mutagenesis analysis of functionally important domains within the C-terminal end of smooth muscle caldesmon.

The ability of chicken gizzard smooth muscle caldesmon (CaD) to inhibit actomyosin ATPase activity is due mainly to an inhibitory domain that resides within the C-terminal 67 amino acid residues of the CaD molecule. In the present study, a series of C-terminal truncation and internal deletion mutants of chicken gizzard smooth muscle CaD were systematically designed using a site-directed mutagenesis approach, and these mutant proteins were overexpressed in a baculovirus expression system. Analysis of actin binding and inhibition of actomyosin ATPase activity using these mutants identified a strong actin-binding motif of 6 amino acid residues (from Lys718 to Glu723), which also form the core sequence for CaD-induced inhibition of actomyosin ATPase. However, maximal inhibition by CaD requires the presence of residues 728-731, which are not associated with actin binding. Our data provide direct evidence for the requirement of actin binding to a specific region in CaD for CaD-induced inhibition of actin activation of smooth muscle myosin ATPase. Furthermore, our findings also show that the region between residues 690 and 717 is responsible for the weak inhibition of actomyosin ATPase and reveal that the inhibitory determinants located in the regions between residues 690 and 717 and residues 718 and 756 can function independently.

Smooth muscle caldesmon (CaD) 1 is a multifunctional elongated molecule that is thought to play a role in modulating the interaction between myosin and actin (1-5; for review, see Refs. 6 and 7). Amino acid sequence and functional analyses have shown that the structure of smooth muscle CaD consists of an N-terminal domain, a helical middle region, and an C-terminal domain (6 -8). The N-terminal domain contains the major myosin binding core sequence that interacts with the S2 region of myosin (9 -11), whereas the C-terminal domain is responsible for interaction with several proteins including actin, tropomyosin, tubulin, calmodulin, and several other Ca 2ϩ -binding proteins (3,6,(12)(13)(14)(15). The middle region has been considered as a spacer, serving only a structural role (8).
The key functions of smooth muscle CaD in cell motility and contraction are its ability to down-regulate actomyosin ATPase activity, presumably by competing with myosin for actin binding (16,17) and, as determined in an in vitro motility assay, to inhibit the movement of actin filaments over myosin heads, probably through a tethering pattern produced by the binding of the CaD N-terminus to the S2 region of myosin and the C-terminal region to actin (17,18). The mechanism by which CaD inhibits actomyosin ATPase activity is not well understood, and it is rather complex, involving an enhancement of the inhibition by tropomyosin and reversal by Ca 2ϩ -calmodulin and/or phosphorylation of mitogen-activated protein kinase (1-5, 17, 18 -22).
Domain mapping studies using CaD fragments obtained from limited proteolysis and chemical cleavage or using recombinant CaD fragments produced in bacterial expression systems have localized both the inhibitory domains and major actin/calmodulin binding sites to the C-terminal fragments with molecular masses ranging from ϳ20 to 40 kDa (8,12,19,(23)(24)(25). These functional domains have been further narrowed to a smaller peptide consisting of the 98 C-terminal amino acids from Trp 659 to Pro 756 (26,27). Using a series of C-terminal truncated proteins of chicken gizzard smooth muscle CaD expressed in a baculovirus system, we showed that both the actin binding site(s) and the inhibitory domain(s) of CaD are composed of continuous multiple determinants located in the last 99 amino acids, whereas the main inhibitory determinants lie in the region between residues 690 and 756 (28). However, the precise localization of these functional determinants in the C-terminal end of CaD has remained unknown.
In this study, we generated a series of CaD C-terminal truncation mutants and internal deletion mutants using site-directed mutagenesis and polymerase chain reaction-based cloning strategies. Utilizing these mutants, we addressed the following questions. 1) Do the inhibitory or actin-binding motifs located within the last 67 amino acids of CaD function independently? 2) Do the actin binding and actomyosin ATPase inhibitory determinants overlap or do they juxtapose closely with each other? 3) What is the minimal sequence within the last 39 amino acids of the C terminus of CaD that is responsible for actin binding and inhibition of actomyosin ATPase?

Construction of Recombinant Baculovirus Transfer Vectors-DNA
oligonucleotide primers were synthesized using an Applied Biosystems DNA/RNA synthesizer. The C-terminal truncation mutants of chicken gizzard smooth muscle CaD cDNA were constructed by a polymerase chain reaction cloning approach as described (29) using the same upstream oligonucleotide primer, 5ЈACTGGGATCCTATAAATATGGAT-GACTTTGAACGC-3Ј. The sequences of the downstream oligonucleotide primers used to generate these truncation mutants were as follows: CaD 1-740 , 5ЈGTCAGGATCCCTACGTAGCTGTTACCTT-3Ј; CaD 1-735 , 5ЈGTCAGGATCCCTATGAAGAAGAAGCAGC-3Ј; CaD 1-727 , 5ЈGTCAG-GATCCCTAAACTGACTGCTTCTC-3Ј; and CaD 1-717 , 5ЈGTCAGGATC-CCTAGCCGGATACATCTCCTGG-3Ј. The underlined sequences in each primer represent the complementary portion, and all the primers contain a BamHI restriction site. Polymerase chain reaction amplifica-tions were performed as described (29). Each CaD truncation mutant was amplified in two independent polymerase chain reactions to check for nucleotide misincorporation during the amplification. The amplified products were introduced into the Pvl941 vector at the BamHI cloning site following BamHI digestion. The single-stranded pBluescript CaD containing full-length CaD cDNA was used as a template for sitedirected mutagenesis reactions. The sequences of these mutagenic primers were as follows: CaD⌬732-735, 5Ј-TGAAAAGCCAGCTAAGG-TAACAGCTACG-3Ј; CaD⌬728 -735, 5Ј-GAGAAGCAGTCAGTTAAGG-TAACAGCTACG-3Ј; CaD⌬724 -735, 5Ј-CGTAATCTCTGGGAGAAGG-TAACAGCTACG-3Ј; CaD⌬721-735, 5ЈTCCGGCAAACGTAATAAGGT-AACAGCTACG-3Ј; and CaD⌬718 -735, 5ЈGGAGATGTATCCGGCAAG-GTAACAGCTACG-3Ј. The internal deletion mutants of smooth muscle CaD were generated using an oligonucleotide-directed mutagenesis system (Amersham Corp.) according to the manufacturer's instructions. Briefly, using these oligonucleotides as primers, the complementary DNA strand was synthesized with T 7 DNA polymerase and then ligated with T 4 ligase. The non-mutant strand was nicked with NciI, digested with exonuclease III, polymerized by DNA polymerase I, and circularized by T 4 ligase. The resulting double-stranded plasmid DNAs were transformed into TG1 bacteria. The CaD internal deletion mutants described above were sequenced using a commercial kit (U. S. Biochemical Corp.) to confirm their structures. The mutated CaD inserts were released by BamHI digestion and introduced into the Pvl941 vector at a BamHI cloning site. The correct orientation of each of the C-terminal truncation and internal deletion mutants was confirmed by restriction mapping and by DNA sequencing (U. S. Biochemical Corp.).
Transfection and Isolation of Recombinant Baculovirus-Transfection of the CaD mutant cDNAs and isolation of the recombinant baculoviruses were carried out as described (28 -30).
Purification of Recombinant CaD and Other Proteins-Recombinant CaDs were prepared according to Wang et al. (29). Smooth muscle actin, CaD, myosin (fully phosphorylated), and tropomyosin were purified from chicken gizzard (31)(32)(33). Bovine brain calmodulin was prepared according to Dedman and Kaetzel (34 (35) using bovine serum albumin as a standard. Binding of CaD mutants to calmodulin was determined using calmodulin-coupled agarose (Sigma) as described (28). Nonspecific binding was estimated by . The CaD mutants produced in some baculovirus expression systems show slight degradation (ranging from 3 to 7%, depending on the time between transfection and cell harvesting). We reported that the function (actin binding) of the baculovirus-generated full-length CaD, having slight degradation (5%), is indistinguishable from that of native CaD, suggesting that the presence of degradation in each of the purified baculovirus-generated recombinant CaD or CaD mutants has no significant effect on the binding data. It is possible that the cleaved site in the recombinant CaD is in its N terminus.
incubation of agarose (without coupled calmodulin) with increasing concentrations of 14 C-labeled full-length CaD or CaD mutants, and these values were subtracted from each point on the binding curve.
The binding of full-length CaD and CaD mutants to actin was determined in the presence or absence of tropomyosin using 14 C-labeled and unlabeled proteins as described (5,28). All determinations of the binding assays were done in triplicate. The apparent dissociation constants for calmodulin, actin, and actin-tropomyosin bindings were determined by the method of Scatchard (36) and by weighted nonlinear leastsquares curve fitting as described by Munson and Rodbard (37).
ATPase Assays-ATP assays were carried out at 25°C as described (38). Specific assay conditions are described in the figure legends.

Two or More Functionally Independent Actomyosin ATPase Inhibitory Motifs Reside in the Last 67 Amino Acid Residues of
CaD-We recently localized the major actin binding determinant(s) and inhibitory motifs for actomyosin ATPase in the chicken gizzard smooth muscle CaD molecule to its C-terminal 67 amino acid residues, presumably involving the regions between residues 690 -717 and 718 -756 (28). However, it is not known whether these two regions contain a functionally independent inhibitory motif(s) and where the functional domains in the C-terminal end of CaD precisely localize. Internal deletions of selected regions in the CaD molecule were made to investigate the dependence of the ATPase inhibitory determinants identified in the last 67 amino acids of the C terminus of CaD. Fig. 1A shows the schematic structure of the internal deletion mutant (CaD-(⌬690 -717)) of chicken gizzard smooth muscle CaD lacking residues Asn 690 to Gly 717 . This CaD mutant was overexpressed in a baculovirus expression system. The purity of CaD-(⌬690 -717) used for our experiments was assessed by SDS-PAGE (Fig. 1B). As shown in Fig. 2, the binding of CaD-(⌬690 -717) to actin began to level off at a stoichiometry of 0.08 mol of CaD per mol of actin, and the binding was saturated at a CaD concentration of 5 m (molar ratio of CaD to actin, 1:4). CaD-(⌬690 -717) binding to actin was higher in the presence of tropomyosin than in its absence (mol of CaD/mol of actin, 0.095 versus 0.08). However, the binding of CaD-(⌬690 -717) to both actin and tropomyosinactin was lower than that for the full-length CaD (CaD 1-756 ). The apparent K d values for the actin binding and tropomyosinactin binding of CaD-(⌬690 -717) obtained from Scatchard analysis (36, 37) were 0.63 Ϯ 0.033 ϫ 10 Ϫ6 M and 0.34 Ϯ 0.02 ϫ 10 Ϫ6 M, respectively, slightly higher than those for full-length CaD (0.52 Ϯ 0.03 ϫ 10 Ϫ6 M for actin-binding; 0.28 Ϯ 0.015 ϫ 10 Ϫ6 M for tropomyosin-actin binding).
Analysis of the inhibitory effect of CaD-(⌬690 -717) on actomyosin ATPase activity revealed strong inhibition of actin or tropomyosin-actin-activated myosin ATPase (Fig. 3), although the inhibition was slightly lower than that produced by CaD 1-756 . These results, together with our previous data (28), suggest that the sequence from Asn 690 to Gly 717 contains a weak inhibitory motif and the stretch from Lys-718 to Pro-756 embraces a strong inhibitory motif, respectively. The two inhibitory motifs are functionally independent.
Presence of a Core Inhibitory Sequence in the Region between Residues 718 and 735-To delineate more precisely the inhibitory sequence in the region between residues 718 and 756, FIG. 4. The CaD C-terminal truncation mutants. A, the structure of these truncation mutants is the same as in Fig. 1A except for different C-terminal ends. B, full-length CaD and CaD C-terminal truncation mutants were purified as described (29) and analyzed by 4 -12% gradient SDS-PAGE followed by Coomassie Blue staining. Lane 1, molecular mass standards (see CaD truncation mutants that have identical N-terminal ends but differ in their C-termini due to the deletion of different portions from the C-terminus were constructed (Fig. 4A). All C-terminal truncation mutants were purified to Ͼ95% homogeneity (Fig. 4B). Actin binding of CaD 1-740 , CaD 1-735 , and CaD 1-727 , i.e. the CaD truncation mutants that progressively removed 16, 21, and 29 amino acids from the C-terminal end of CaD, was virtually indistinguishable from that of full-length CaD (Fig. 5A). In the presence of tropomyosin, the binding of CaD 1-740 , CaD 1-735 , and CaD 1-727 to actin increased in a manner similar to that for full-length CaD (Fig. 5B). In contrast, the binding of CaD 1-717 , missing 39 amino acids from the C-terminus of CaD, to both actin and tropomyosin-actin was 66 and 70%, respectively, as compared with full-length CaD (Fig. 5, A  and B). This suggests that a 10-amino acid stretch from Lys 718 to Val 727 covers the strong actin-binding motif in the C-terminal end of CaD. The apparent K d values of both actin binding and tropomyosin-actin binding for CaD 1-740 , CaD 1-735 , and CaD 1-727 were identical to those of full-length CaD as described above. Fig. 6 illustrates the ability of these truncation mutants to inhibit actomyosin ATPase activity. The inhibition of actin-or tropomyosin-actin-activated ATPase activity was unchanged with deletion of up to 21 amino acid residues (CaD 1-740 and CaD 1-735 ); however, further deletion of 8 (CaD 1-727 ) amino acid residues resulted in an 8 and a 10% reduction in the inhibition of the actin-activated myosin ATPase in the absence and the presence of tropomyosin, respectively, as compared with fulllength CaD. The inhibitions of actomyosin ATPase activity by CaD 1-717 in the absence or the presence of tropomyosin were, respectively, 37 and 44% of those produced by full-length CaD. These results suggest that the 18-amino acid stretch from Lys 718 to Ser 735 is responsible for the strong inhibition of actomyosin ATPase.
In another series of experiments, a site-directed mutagenesis approach was used to generate internal deletion mutants of chicken gizzard smooth muscle CaD, in which 4 -18 amino acid residues in the region from Lys 718 to Ser 735 were progressively nest-deleted (Fig. 7A). These mutants were used to identify the core sequence necessary for inhibition of actomyosin ATPase activity. All CaD internal deletion mutants were overexpressed in a baculovirus expression system and purified to Ͼ95% homogeneity (Fig. 7B), although their yields varied slightly. As expected, CaD⌬732-735 and CaD⌬728 -735, lacking 4 and 8 amino acid residues, respectively, bound to actin or tropomyosin-actin as efficiently as full-length CaD (Fig. 8A and 8B). The absence of residues 724 -727 also did not impair the binding of CaD to actin (Fig. 8A). However, further deletion of 6 amino acid residues between Lys 718 and Glu 723 caused a considerable decrease in the binding of CaD to actin (Fig. 8A). Apparently, the 6-amino acid stretch, KRNLWE, contributes to the strong actin-binding motif located within the last 39 amino acid residues of CaD. The apparent dissociation constants of CaD-(⌬721-735) and CaD-(⌬718 -735) were 0.72 Ϯ 0.038 and 1.03 Ϯ 0.047 ϫ 10 Ϫ6 M, respectively. The binding affinity of CaD-(⌬732-735), CaD-(⌬728 -735), and CaD-(⌬724 -735) for actin was very similar to that of full-length CaD.
The presence of tropomyosin increased the binding of CaD and the internal deletion mutants to actin (Fig. 8B). In the presence of tropomyosin, the apparent K d values of actin binding for CaD-(⌬721-735) and CaD-(⌬718 -735) were 0.41 Ϯ 0.032 and 0.63 Ϯ 0.041 ϫ 10 Ϫ6 M, respectively.

FIG. 6. Inhibition of actin-activated myosin ATPase by the CaD C-terminal truncation mutants.
Actin or tropomyosin-actin-activated myosin ATPase activities were determined as described (38). Symbols are the same as in Fig. 5. A, effect of CaD C-terminal truncation mutants on actin-activated ATPase activity and B on tropomyosin-actin-activated ATPase activity. spectively, suggesting that deletion of residues 732-735 and 724 -727 has no effect on CaD-induced inhibition of actin-activated myosin ATPase. These results demonstrate that there are two inhibitory determinants between residues 718 -723 and 728 -731 that are separated by 4 amino acid residues from Lys 724 to Val 727 . These two determinants directly participated in producing the strong inhibitory activity by smooth muscle CaD because deletion of residues 721-735 resulted in a reduction by 24% in the inhibition of actomyosin ATPase as compared with the full-length CaD, and further removal of three residues from Lys 718 to Asn 720 caused a dramatic decrease by 61% (Fig. 9A). Apparently, the core sequence is composed of the 6 amino acid residues between Lys 718 and Glu 723 . Our data also suggest that residues 728 -731 that are irrelevant to the binding of CaD to actin (as shown in Fig. 8) are associated with a weak inhibition of actomyosin ATPase.
In the presence of tropomyosin, the inhibition by both fulllength CaD and these internal deletion mutants was enhanced (Fig. 9B).

Reversal of the Inhibition of Actomyosin ATPase by CaD Truncated and Internal Deletion Mutants in the Presence of
Ca 2ϩ -Calmodulin-We previously obtained some evidence to suggest that the region between residues 690 and 717 is responsible for the high affinity binding to calmodulin (28). To obtain more accurate information about the binding affinity of CaD to calmodulin, we examined the binding of CaD-(⌬690 -717) to calmodulin using calmodulin-coupled agarose and [ 14 C]iodoacetamide-labeled or unlabeled full-length CaD and CaD-(⌬690 -717). Tropomyosin, which does not bind to calmodulin, was used as a negative control. Nonspecific binding was considered as background. As shown in Fig. 10, A and B, the absence of CaD residues 690 -717 caused a 41% reduction in the binding to calmodulin, but under the same assay conditions, tropomyosin did not bind to calmodulin-coupled agarose. The binding of both CaD 1-756 and CaD-(⌬690 -717) increased upon raising their molar concentrations (Fig. 10C). Interestingly, the calmodulin binding for both CaD 1-756 and CaD-(⌬690 -717) leveled off at a stoichiometry of ϳ1 mol of CaD/mol of calmodulin, although the maximal calmodulin-binding of CaD-(⌬690 -717) was significantly lower than that of CaD 1-756 . The apparent dissociation constant of CaD-(⌬690 -717) was 1.96 Ϯ 0.058 ϫ 10 Ϫ6 M, much higher than that of full-length CaD (0.98 Ϯ 0.061 ϫ 10 Ϫ6 M). Our results provide unequivocal evidence that a second strong calmodulin-binding motif is located in the residues 690 -717, presumably involving residues Asn 690 to Lys 695 .
To determine whether the strong calmodulin-binding site in the region between residues 658 and 689 (26), which is further CaD-(⌬718 -735) (Ⅺ) to actin or tropomyosin-actin was measured as described in Fig. 2. A, binding of the CaD internal deletion mutants to actin and B to tropomyosin-actin. narrowed down to residues 658 -666 (8), is functionally involved in down-regulating CaD-induced inhibition of actin-activated myosin ATPase, we analyzed the calmodulin-induced reversal of the inhibition by either full-length CaD or CaD-(⌬690 -717). Reversal of the inhibition of actomyosin ATPase activity by CaD-(⌬690 -717), shown in Fig. 3, by calmodulin is depicted in Fig. 11. The inhibition caused by both CaD 1-756 and CaD-(⌬690 -717) decreased upon increasing the concentration of calmodulin, and the percentage of reversal by calmodulin for the inhibition caused by CaD-(⌬690 -717) and full-length CaD was similar. This result indicates that the CaM-binding site between 658 and 689 is important for calmodulin to reverse the inhibition of actomyosin ATPase activity by CaD. Furthermore, tropomyosin had no effect on the calmodulin-induced reversal of inhibition by either full-length CaD or CaD-(⌬690 -717) (Fig.  11). At a stoichiometry of 0.075 mol of CaD/mol of calmodulin, the inhibitions caused by both CaD-(⌬690 -717) and CaD 1-756 were completely reversed.
Reversal of the inhibition of actomyosin ATPase by CaD truncated and internal deletion mutants (shown in Figs. 4 and  7) in the presence of Ca 2ϩ -calmodulin was also examined. The inhibition by these truncation mutants was reversed by calmodulin in the presence of Ca 2ϩ in a pattern similar to that seen with full-length CaD (data not shown). A stoichiometry of 0.075 mol of CaD/mol of calmodulin was required for the complete reversal of the inhibition by either full-length CaD or these truncation mutants (data not shown). Additionally, the inhibition by these mutants described in Fig. 9 was completely reversed at about 13:1 molar ratio of calmodulin to CaD (data not shown).

DISCUSSION
Our previous study indicated that the region necessary for inhibition of actomyosin ATPase was localized mainly in the C-terminal 67 amino acid residues (28). The 67-amino acid stretch also contains two or more functional motifs for actin binding and one high affinity calmodulin binding site (28). However, the precise localizations of these functional domains and the exact relationship between them remain unclear. In the present study, we generated a series of C-terminal truncation mutants and internal deletion mutants that, unlike bacterially expressed CaD peptides, have intact C-terminal domains and central helical regions but different C-terminal ends. Thus, the major regions of the CaD molecule and the structural relationships during protein-protein interaction remain intact, a feature that is impossible to obtain with CaD fragments derived from limited proteolysis or bacterial expression systems (8,12,19,(23)(24)(25)(26)(27). Using the CaD truncation mutants and internal deletion mutants expressed in the baculovirus expression system, we extend our previous study and have precisely determined the major ATPase inhibitory domain and one strong actin-binding motif located in the last 39 amino acid residues of chicken gizzard smooth muscle CaD.
We show here that the region between residues 690 and 717 is responsible for weak inhibitory activity and weak actin binding, whereas the region between residues 718 and 756 contributes to strong inhibitory activity and strong actin binding. Evidently, the regions between residues 690 -717 and 718 -756 each contain at least one inhibitory determinant for the CaDinduced inhibition of actomyosin ATPase, and these inhibitory determinants can function independently. However, the deletion of one binding site is likely to affect not only the affinity but also the stoichiometry as shown in Figs. 2, 5, and 8. There are two possible explanations for the decrease in the stoichiometry of binding. 1) The conformation of CaD is changed by the deletion of either one of the actin-binding sites. 2) The conformation of the actin upon binding of CaD may vary depending on the presence or absence of either one of the actinbinding sites on the CaD. Hence, the difference in the stoichiometry of binding observed upon deletion of one binding site without affecting the other may be attributed to these confor- FIG. 9. Inhibition of actin-activated myosin ATPase activity by the CaD internal deletion mutants. Assay conditions were as described under "Materials and Methods." Symbols are the same as in Fig. 8. A, effect of CaD internal deletion mutants on actin-activated ATPase activity and B on tropomyosin-actin-activated ATPase activity. mational changes, and it is possible that the same phenomenon may apply for the binding of CaD mutants deficient in CaMbinding site to CaM, as shown below.
The precise localization of the strong actin binding motif between residues 718 and 756 has been determined in our present study. Deletion of 29 amino acid residues from the C-terminal end of CaD had no effect on the binding of CaD to actin, whereas further deletion of residues spanning from Lys 718 to Val 727 markedly reduced the ability of CaD to bind to actin, suggesting that this 10-amino acid stretch harbors a strong actin-binding motif. The data in Fig. 8 further define the actin-binding motif as a 6-amino acid sequence from Lys 718 to Glu 723 . Thus, our findings rule out the possibility that the activity of the C-terminal actin-binding motif depends on a single amino acid residue. The actin-binding motif, KRNLWE, contains amino acids with positively and negatively charged side chains. An uncharged nonpolar amino acid, tryptophan, is also present in the composition of the actin-binding motif. In fact, tryptophan has been found to be the key residue associated with calmodulin binding in other proteins (39). Thus, it seems likely that tryptophan and the three charged amino acid residues act in concert to form the actin-binding pocket.
The last 21 amino acid residues from the C-terminal end of CaD do not participate in the CaD-induced inhibition of actomyosin ATPase. The inhibitory motif actually consists of two discontinuous epitopes represented by a 6-amino acid sequence from Lys 718 and Glu 723 and a 4-amino acid stretch from Lys 728 to Ala 731 . The 6-amino acid sequence, KRNLWE, contains the main inhibitory activity since the deletion of this sequence caused a more significant decrease in the CaD-induced inhibitory activity. On the other hand, the 6-amino acid sequence also encompasses the overall sequence responsible for the strong actin binding. This finding suggests that the inhibitory domain largely overlaps with the actin-binding motif, consistulin was determined as described under "Materials and Methods." A, lane 1, molecular mass standards (see ent with the requirement for binding of CaD to actin for CaDinduced inhibition of actomyosin ATPase. Note, however, that the maximal inhibition by CaD requires the presence of the sequence between residues 728 and 731 (Fig. 9), a region not associated with actin binding. Most likely, the sequence between residues 728 and 731 is required in maintaining the conformation of the inhibitory domain. Future studies with smooth muscle myosin subfragment-1 and these mutants or new CaD C-terminal point mutants will help us to define the precise nature of these interactions and shed light on the mechanism by which CaD down-regulates actin-activated myosin ATPase.
The C-terminal region of smooth muscle CaD contains at least two calmodulin-binding sites, as demonstrated by several investigators using synthetic peptides and bacterially expressed fragments (27,28,40,41). One calmodulin-binding site, designated CaM-binding site A, has been localized in a 7-residue segment from Trp 659 to Phe 665 (40). The other site, CaM-binding site B, was mapped to the region between residues 675 and 695 (41) and has recently been narrowed to a 6-residue stretch from Asn 690 to Lys 695 (28). Using indirect peptide-competitive binding assays, Marston et al. (41) showed that site B (residues 675-695), but not site A (residues 658 -666), is closely associated with the reversal of CaD-induced inhibition of actomyosin ATPase. By contrast, A. Wang and colleagues (42) recently reported that the functionally related calmodulin-binding site is site A. We recently showed that the CaM-binding motif in site B lies between residues 690 -695 (28). Our present data indicating that calmodulin reverses the inhibition induced by CaD-(⌬690 -717), which lacks the site B sequence, are in accord with the report of A. Wang and colleagues (42). However, we find that the binding of this mutant to calmodulin occurs at a much lower level than that for fulllength CaD. Thus, the importance of site B in reversing CaDinduced inhibition by calmodulin cannot be ruled out.