Effects of Ca2’ on the Conformation and Enzymatic Activity of Smooth Muscle Myosin*

The influence of Ca2+ on the enzymatic and physical properties of smooth muscle myosin was studied. The actin-activated ATPase activity of phosphorylated gizzard myosin and heavy meromyosin is higher in the presence of Ca2+ than in its absence, but this effect is found only at lower MgC12 concentrations. As the MgCl2 concentration is increased, Ca2+ sensitivity is decreased. The concentration of Ca” necessary to activate ATPase activity is higher than that required to saturate calmodulin. The similarity of thepCa dependence of ATPase activity and of Ca2+ binding to myosin and the competition by M e + indicate that these effects involved the Ca2+-Mg2+ binding sites of gizzard myosin. For the actin dependence of ATPase activity of phosphorylated myosin at low concentrations of MgC12, both V, and K, are influenced by Ca2+. The formation of small polymers by phosphorylated myosin in the presence of Ca2+ could account for the alteration in the affinity for actin. For the actin dependence of phosphorylated heavy meromyosin at low MgClz concentrations, Ca2+ induces only an increase in Vmu. To detect alterations in physical properties, two techniques were used: viscosity and limited papain hydrolysis. For dephosphorylated myosin, 6 S or 10 s, Ca2+-dependent effects are not detected using either technique. However, for phosphorylated myosin the decrease in viscosity corresponding to the 6 S to 10 S transition is shifted to lower KC1 concentrations by the presence of Ca2+. In addition, a Ca2+ dependence of proteolysis rates is observed with phosphorylated myosin but only at low ionic strength, i.e. under conditions where myosin assumes the folded conformation.

The influence of Ca2+ on the enzymatic and physical properties of smooth muscle myosin was studied. The actin-activated ATPase activity of phosphorylated gizzard myosin and heavy meromyosin is higher in the presence of Ca2+ than in its absence, but this effect is found only at lower MgC12 concentrations. As the MgCl2 concentration is increased, Ca2+ sensitivity is decreased. The concentration of Ca" necessary to activate ATPase activity is higher than that required to saturate calmodulin. The similarity of thepCa dependence of ATPase activity and of Ca2+ binding to myosin and the competition by M e + indicate that these effects involved the Ca2+-Mg2+ binding sites of gizzard myosin. For the actin dependence of ATPase activity of phosphorylated myosin at low concentrations of MgC12, both V , , and K , are influenced by Ca2+. The formation of small polymers by phosphorylated myosin in the presence of Ca2+ could account for the alteration in the affinity for actin. For the actin dependence of phosphorylated heavy meromyosin at low MgClz concentrations, Ca2+ induces only an increase in Vmu. To detect alterations in physical properties, two techniques were used: viscosity and limited papain hydrolysis. For dephosphorylated myosin, 6 S or 10 s, Ca2+-dependent effects are not detected using either technique. However, for phosphorylated myosin the decrease in viscosity corresponding to the 6 S to 10 S transition is shifted to lower KC1 concentrations by the presence of Ca2+. In addition, a Ca2+ dependence of proteolysis rates is observed with phosphorylated myosin but only at low ionic strength, i.e. under conditions where myosin assumes the folded conformation.
It is generally accepted that the phosphorylation-dephosphorylation of the 20,000-dalton light chains of myosin form an important component of the regulatory mechanism in smooth muscle (1,2). In its simplest interpretation the phosphorylation theory would predict that myosin phosphorylation allows the activation of Mg2+-ATPase by actin and hence increases the cross-bridge cycling rate leading to tension development; dephosphorylation would reverse the process resulting in the formation of an "inactive" myosin and relaxation. It is apparent that this scenario is oversimplified and the physiological responses of smooth muscle require a more complex or versatile regulatory mechanism. For example, it was found that intact fibers from several types of smooth *This work was supported by Grants HL 23615 and HL 20984 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. muscle exhibited a Ca2+-dependent resistance to stretch (3,4) that was thought to reflect attached noncycling cross-bridges. Subsequently, the attached noncycling bridges, termed latchbridges, were shown to occur with dephosphorylated myosin (5) and possessed a Ca2+ sensitivity distinct from that of the calmodulin-dependent myosin light chain kinase (6). Thus, it was proposed that light chain phosphorylation is correlated with the cross-bridge cycling rate (5) but that a second Ca2+dependent mechanism regulates stress maintenance (6,7). The latter unidentified mechanism would presumably have a higher affinity for Ca2+ than calmodulin since it operates when the calmodulin-myosin light chain kinase complex is inactive. This idea is consistent with recorded Ca2+ transients in vascular smooth muscle cells that showed that tension can be maintained at a lower Ca2+ level than that required for force development (8).
Another aspect that is not understood is the effect of Ca2+ on shortening velocity. Although this is a controversial topic (see references in Ref. 9) there are several reports that claim that Ca2+ influences shortening velocity of intact (10,ll) and skinned (9,12,13) smooth muscle fibers. This can occur at high levels of phosphorylation without any marked change in the extent of phosphorylation (9,13) and is therefore different from the formation of latch-bridges. The requirement for Ca2+ is not obligatory since contraction of skinned smooth muscle fibers can be induced in the absence of Ca2+ by the addition of a Ca2+-independent myosin light chain kinase (14). However, it is likely that under suitable conditions (the free Mg2+ concentration is particularly important), the cycling rate of a phosphorylated cross-bridge can be modified by varying the Ca2+ concentration. The molecular basis for such an effect is unknown.
The role of the Ca2+-myosin interaction in the regulation of smooth muscle activity is controversial. In several reports, it is claimed that the actin-activated ATPase of phosphorylated, or, thiophosphorylated, myosin is not Ca2+-dependent (48-53). Other investigators have observed that in the absence of Ca2+ the actin-activated ATPase of phosphorylated myosin (39, 54-57) or heavy meromyosin (HMM') (58) is reduced. There is also a report (59) that the actin-activated ATPase of fully phosphorylated gizzard myosin is not Ca2+-dependent, whereas the M$+-ATPase of partially phosphorylated myosin is Ca2+-dependent. Many of these apparently contradictory reports c m be resolved if the ionic conditions, especially the M$+ concentration, of the various ATPase assays are taken into consideration. In order to increase ATPase activity, many investigators use relatively high concentrations of Mg2+ in their assays (for references see Ref. 60) and, as pointed out (39,56), these conditions would reduce, or eliminate, a Ca2+sensitive response. If the binding of Ca2+ to the Ca2+-Mg2+ sites of phosphorylated myosin is responsible for Ca2+ sensitivity, as proposed by Chacko and Rosenfeld (39), such an effect would be expected.
From the evidence mentioned above, it is likely that Ca2+ can influence the myosin-actin interaction in vitro and it is possible that the binding of Ca2+ to myosin may play an important regulatory role in the intact muscle. To obtain more information on the effects of Ca2+ on smooth muscle myosin we initiated this study. Of particular interest were the effects of Ca2+ on actin-activated ATPase activity and on the properties and conformation of myosin. Previously it was suggested (61) that the folded (10 S ) and extended (6 S) forms of monomeric myosin had distinct enzymatic properties and that some characteristic of the 6 S or 10 S conformation is a determinant of ATPase activity. With this in mind it was pertinent to determine if Ca2+-dependent conformational changes of myosin could be detected that in turn might reflect the Caz+ dependence of ATPase activity. The results are presented below.

MATERIALS AND METHODS
Smooth muscle myosin was prepared from frozen turkey gizzards by a method modified from that of Persechini and Hartshorne (62) as follows. Turkey gizzards are trimmed of fat and connective tissue and minced. The mince (usually 250-500 g) is suspended in 3 volumes of buffer A, i.e. 10 mM Tris-HC1 (pH 7.5), 50 mM KCl, 2 mM EGTA, 25 mM MgCL, 0.2 mM dithiothreitol, 3% (v/v) Triton X-100, homogenized for 20 s in a Waring blender, and centrifuged at 1500 X g for 5 min. The supernatant is discarded and the pellet is subjected to 3 cycles of homogenization and centrifugation with buffer A, using 3 volumes based on the original mince weight for each cycle. The pellet is suspended in 3 volumes of buffer B, i.e. 10 mM Tris-HC1 (pH 7.5), 100 mM KC1, 2 mM EGTA, 0.2 mM dithiothreitol, homogenized (Waring blender, 20 s), and centrifuged at 8,000 X g for 10 min. The supernatant is discarded and the pellet is subjected to two additional cycles of homogenization and centrifugation using buffer B. Extraction of myosin is achieved by suspending the pellet in 1.5 volumes (based on original mince weight) of 40 mM imidazole (pH 6.8), 5 mM ATP, 4 mM EDTA, 2 mM EGTA, 0.5 mM dithiothreitol and homogenizing in a Waring blender for 20 s. The pH is adjusted to 6.9, the mixture is left on ice for 15 min, with occasional stirring, and centrifuged at 14,000 X g for 20 min. The supernatant is filtered through glass wool and the pellet is discarded. The pH of the supernatant is adjusted to 7.6 and 1 M MgClz is added slowly, using a peristaltic pump, to a final concentration of 150 mM. Additional ATP is added to increase the ATP concentration by 2. is left, with occasional stirring, for 10 min, and centrifuged at 14,000 X g for 10 min. The supernatant is centrifuged overnight at 75,000 X g. The resulting supernatant is diluted with 10 volumes of cold Hz0 and the precipitated myosin is collected by centrifugation at 10,000 X g for 10 min. The myosin pellet is suspended in approximately 10 volumes (based on pellet volume) of 10 mM sodium phosphate (pH 7.6), 1.5 mM EGTA, 5 mM ATP and homogenized by hand in a glass homogenizer. After complete suspension, the pH is adjusted to 7.6. 1 M MgClz is added slowly to a final concentration of 150 mM, and the mixture is centrifuged at 160,000 X g for 3 h. The supernatant is diluted with 10 volumes of cold Hz0 and the precipitated myosin is collected by centrifugation at 10,000 X g for 10 min. The pelleted myosin is dissolved in approximately 10 volumes (based on pellet volume) of 0.5 M KCl, 10 mM Tris-HC1 (pH 7.5), 1 mM dithiothreitol using a glass homogenizer. When the myosin is dissolved, cold HzO is added to give a final KC1 concentration of 0.1 M. After 15 min on ice the mixture is centrifuged at 14,000 X g for 15 min. The supernatant is myosin and the pellet contains myosin plus contaminant actin. (The amount of myosin pelleted by this procedure is variable, but an excessive precipitation, i.e. a loss of over 30% total myosin, often reflects a failure to maintain pH during the earlier MgClz precipitation steps.) The supernatant is diluted with an equal volume of cold HzO and 1 M MgClZ added to 10 mM. After 1 h on ice, the precipitated myosin is collected by centrifugation at 10,000 X g for 15 min, and the pellet is dissolved in 0.5 M KC1, 1 mM NaHC03, 1 mM dithiothreitol and dialyzed uersus this buffer. The average yield of myosin is 2.5 mg/g, wet weight, gizzard mince. The myosin prepared by this procedure is usually dephosphorylated as assessed by urea gel electrophoresis (63).
Other proteins were prepared by the following procedures: HMM by a-chymotryptic hydrolysis of gizzard myosin (64); myosin light chain kinase from frozen turkey gizzards (65); calmodulin from frozen bull and goat testes (65); actin from rabbit skeletal muscle (66); and tropomyosin from frozen turkey gizzards (67).
The procedures involved for the limited proteolysis of myosin by papain were as described previously (68). The extent of myosin degradation was estimated by integrating the area of intact myosin heavy chain observed at different times of digestion on SDS-polyacrylamide gels. Caz+-binding to gizzard myosin was estimated at pH 7.0 by equilibrium dialysis under the conditions given in the figure legend. The total concentration of EGTA ( i e . EGTA and Ca2+-EGTA) was 0.1 mM. The apparent dissociation constant at pH 7.0 for Caz+-EDTA was assumed to be 1 X 10-6"7 M. Myosin (4 mg/ml) was dialyzed overnight at 4 "C at different pCa values using a EMD 101 equilibrium dialysis system (Hoefer Scientific Instruments). %a (45CaClz, New England Nuclear) was measured by scintillation counting (Beckman LS 9000 Scintillation System) in 100-pl aliquots from each side of the membrane and the amount of bound &Ca was estimated. For practical reasons neither ATP nor actin were included in the samples for equilibrium dialysis. ATPase activities were measured at 25 "C, as described previously (69), under conditions given in the figure legends. Routine assays were carried out at a pH 7.5, with the exception that pCa dependence of ATPase activity ( Fig. 2) was determined at pH 7.0. Phosphorylation of myosin was as described in the figure legends and determined by the procedure of Walsh et al. (65). Electrophoresis was carried out on 7.5-20% polyacrylamide gradient slab gels in the presence of 0.1% SDS using the discontinuous buffer system of Laemmli (70). The gels were stained with Coomassie Brilliant Blue R 250 (Sigma) and scanned by a GS 300 Scanning Densitometer (Hoefer Scientific Instruments) attached to a LCI-100 Laboratory Computing Integrator (Perkin Elmer). The measurements of viscosity and sedimentation velocity were as outlined previously (67). Other procedures are given by Walsh et al. (71).

RESULTS
The actin-activated ATPase of phosphorylated gizzard myosin (approximately 101 molar ratio of actin:myosin, respectively) at varying levels of MgCl2 and in the presence and absence of Ca2+ is shown in Fig. 1A. (The indicated concentration of MgCL is the total concentration; in the presence of 1 mM ATP it is assumed that the free Mg2+ concentration is approximately 1 mM less than the total concentration.) This figure illustrates an important point, namely that the level of actin-activated ATPase activity is markedly dependent on the MgC1, concentration, and that the Ca2+ dependence of this activity is more pronounced at lower MgClz concentrations. It was suggested previously (60) that the low specific actomyosin ATPase activity, characteristic of lower MgC12 concentrations, reflects the presence of the 10 S conformation.
Increasing the MgClz concentration results in higher activities and the conversion of 10 S to 6 S myosin. As can be seen from Fig. 1A the increase in ATPase activity also is associated with a gradual decrease of Ca2+ sensitivity, expressed as a percentage of the two activities (+Ca"). At 6 mM MgC12 and above, there is little significant difference in the ATPase activities measured in the presence and in the absence of Ca2+.
A similar Ca2+-dependent response of ATPase activity was obtained for phosphorylated myosin in the absence of calmodulin and myosin light chain kinase (results not shown).
Myosin (40-50 mg) was prephosphorylated to 1.8 mol of P/ mol of myosin under the conditions given in the legend to In the presence of gizzard tropomyosin (Fig. 1B) the actinactivated ATPase is activated, both in the presence and absence of Ca".
There is a marked dependence of ATPase activity on MgClz concentration, as observed in the absence of tropomyosin. A Ca2+-dependent response of ATPase activity also is found and this is exaggerated at lower MgClz concentrations. As the MgCIz concentration is increased, the ATPase activity-in the absence of Ca2+ increases until it equals that measured in the presence of Caz+, i.e. Ca2+ sensitivity is lost. This occurs at 3 to 4 mM MgC1, (total), and this is considerably lower than the concentration of MgClz at which Ca2+ sensitivity is lost in the absence of tropomyosin. Miyata and Chacko (72) have shown recently that the Ca2+ activation of actin-activated ATPase activity in the presence of tropomyosin is not due to an effect of Ca2+ on binding of tropomyosin to actin.
Throughout the experiments illustrated in Fig. 1, the level of myosin phosphorylation remained constant, at 1.7 to 1.8 mol of P/mol of myosin. There was no detectable dephosphorylation for up to 30 min after the addition of EGTA. It is unlikely, therefore, that any of the observed differences in ATPase activity are due to altering levels of myosin phosphorylation.
The Ca2+ dependence of the actin-activated ATPase activity of phosphorylation gizzard myosin at two concentrations of MgCl2, 0.1 and 1 mM (total), is shown in Fig. 2. As in the previous experiment the myosin was prephosphorylated (-1.8 mol of myosin) and then assayed at pH 7.0 at varying concentrations of Ca2+, using a Ca2+-EGTA buffer system. The pCa of half-maximal activity (assuming the activity at pCa 4 to be the maximum) for the 0.1 mM MgClz assays is approximately 10 pM, and for the 1 mM MgClz assays is approximately 30 p~. Since the binding sites involved on the myosin are Ca2+-Mg2+ sites, a shift to lower affinity at higher M e concentrations is expected. However, on this limited experimental basis and with the relatively slight shift ofpCa values the significance of these results should not be overemphasized. Under identical conditions (Le. pH, calmodulin concentration, etc.) activation of myosin light chain kinase occurs at lower Ca2+ concentrations, and half-maximal phosphorylation was obtained at approximately 1 PM Ca2+ (Fig. 2). Thus, it appears that under these assay conditions the Ca2+ sites on myosin The binding of Ca2+ to dephosphorylated myosin at 3 levels of MgC12 is shown in Fig. 3. For practical reasons these determinations differed from the ATPase assays in that ATP and actin were not included. The results show clearly that the binding of Ca2+ to myosin is dependent on the MgC12 concentration and is almost eliminated at 6 mM M e . Half-saturation in 0.1 mM MgClz occurred at approximately 35 p M ca2+ and at approximately 50 p~ Ca2+ in 1 mM MgCI2. The binding of Ca2+ to phosphorylated myosin in 0.1 mM MgC12 also is shown (Fig. 3) and the binding curve is the same as that obtained with dephosphorylated myosin.
In general the above results confirm those of others (39,56) in that it is demonstrated that the Ca2+ sensitivity of actin-activated ATPase and binding of Ca2+ to myosin are markedly affected by the M$+ concentration. The following experiments explore in more detail the Ca2+ dependence of actin-activated ATPase activity using gizzard myosin and HMM.
In Fig. 4   omer and a small myosin polymer (in the order of 2 to 6 molecules). As shown in Fig. 4, the extent of polymer formation appears to be Ca2+-dependent. In the presence of Ca2+ (-1 X M) the polymer boundary is considerably larger than in the absence of Ca2+. The proportion of polymer-tomonomer could affect the ATPase properties, and it might be predicted that a myosin polymer would bind actin stronger than the myosin monomer. Therefore, this may be at least a partial explanation for the variations in K, observed in these experiments.
In 10 mM MgC12 the actin-activated ATPase activity is not To check if Ca2+-induced conformational changes in myosin could be detected, two approaches were taken. The first was to determine if the viscosity of myosin is altered in the presence and absence of Ca2+, and the second utilized limited proteolysis with papain as a conformational probe.
The viscosity data for phosphorylated gizzard myosin are shown in Fig. 7. As described previously (61) myosin at higher KC1 concentrations (in this experiment, above 0.25 M) exists as 6 S, and as the ionic strength is reduced the 10 S conformation is formed. This transition is reflected by a decrease in viscosity. As can be seen (Fig. 7) the presence of Ca2+ shifts the viscosity transition to lower KC1 concentrations. This could be due either to a relative resistance of the 6 S myosin plus Ca2+ to form the folded conformation, or to an increased formation of small polymers with 10 S myosin plus Ca". As shown above (Fig. 4), the sedimentation velocity profiles of phosphorylated myosin indicated an enhanced tendency for polymer formation in the presence of Ca". For dephospho-

ACTIN" (JJM)"
rylated myosin an effect of Ca2+ is not detected and the viscosity transitions in the presence and absence of Ca2+ are identical (Fig. 7).
It has been demonstrated that the rate of limited papain hydrolysis of myosin is sensitive to conformation (68,73) in that 10 S myosin is more resistant to proteolysis than 6 S myosin. This approach was used to screen for possible Ca2+dependent conformational changes in gizzard myosin. The extent of proteolysis was estimated from the area of the myosin heavy chain observed on SDS-polyacrylamide gels. Time courses of heavy chain disappearance for myosin exposed to papain under various conditions are shown in Fig. 8. in control experiments on casein digestion that variations in KC1 concentrations from 50 mM to 0.4 M had negligible effects on the rate of papain hydrolysis.) For 10 S dephosphorylated myosin the proteolysis rate is considerably slower than that for 6 S myosin and there is no effect of Ca2+ (Fig. 8). (The concentration of papain used with dephosphorylated 10 S myosin was 3-fold higher than that used in the other experiments.) The only conditions under which an effect of Ca2+ could be detected was for phosphorylated 10 S myosin. For phosphorylated 10 S myosin in the absence of Ca", the proteolysis rate was faster than for dephosphorylated 10 S myosin, and this indicates an alteration in the molecule induced by phosphorylation (see "Discussion") that is not evident from sedimentation velocity experiments. In the pres- ence of Ca2+, the proteolysis of 10 S phosphorylated myosin is more rapid than in the absence of Ca" and approaches the hydrolysis rates observed for 6 S myosin. Thus, using the technique of limited papain hydrolysis, the only detectable effects of Ca2+ on proteolysis rates are observed when 10 S phosphorylated myosin is used.

DISCUSSION
The data presented above indicates that the actin-activated ATPase of phosphorylated gizzard myosin can be modified by a Ca2+-dependent process. The Ca2+-dependent response is seen both in the presence and absence of tropomyosin and is seen also for the ATPase activity of acto-HMM. In all instances the Ca2+ effect is influenced by the prevalent M e concentration and in general is favored by lower concentrations of M$+. Although these results are qualitatively similar to previous studies (39, 56, 58) there are some distinctions. For example, in the presence of tropomyosin and using skeletal muscle actin we find both a M$+ and Ca2+ dependence of actin-activated ATPase activity. Nag and Seidel(56) found little Ca2+ dependence using skeletal actin plus tropomyosin, whereas with gizzard actin plus tropomyosin they observed a marked Ca2+ dependence at lower MgClz concentrations. Ca2+ dependence in our system is seen even at low M P levels and does not show an optimum at approximately 2 mM free M$+ as found with arterial myosin (39). Rees and Fredericksen (55) with porcine aorta myosin found a stimulation by Ca" of the actin-activated ATPase of dephosphorylated myosin and this is not found with our system. Previously it was reported that the ATPase activity of acto-arterial HMM showed a marked Ca2+ dependence only in the presence of tropomyosin (58), whereas in our system tropomyosin is not essential.
The most likely site for Ca2+ binding, and therefore the site responsible for these Ca2+ effects, is the myosin molecule. The evidence for this is as follows. 1) Myosin from each muscle type is known to bind Ca2+. With the exception of the molluscan-type myosin that possesses Ca2+-specific sites, other myosins bind Ca2+ at Ca2+-M$+ sites ( The effect of Ca2+ on the actin-activated myosin ATPase was to alter both the apparent dissociation constant for actin (KO) and Vmx. It is suggested that the alteration of KO could be associated with the polymer formation which was shown to occur in the presence of Ca" (see Fig. 4). To evaluate the relative contributions of K, and V, , , to the Ca" dependence of ATPase and to eliminate complications due to polymer formation, we studied the actin-activated ATPase of gizzard HMM. For the ATPase activity of acto-HMM a Ca2+ dependence is observed at lower MgClz concentrations but not at higher MgC12 concentrations (see Figs. 5 and 6). The Ca2+ effect is due to a shift in Vma. K, in the presence and absence of Ca2+ remains constant. In the intact muscle one would not expect to see effects due to. myosin polymerization since it has been reported that thick filaments exist both in resting and contracting muscle (74). Thus, it might be predicted that the increase in V, , observed in vitro with phosphorylated myosin would be reflected in vivo by an increase in shortening velocity. In support of this, Siegman et al. (13) found that an increase in external Ca2+ caused approximately a 2-fold increase in the unloaded shortening velocity of intact taenia coli fibers. With skinned gizzard fibers at constant levels of phosphorylation, increasing Ca2+ concentrations also increase shortening velocity. 2 The sedimentation velocity experiments show that at low ionic strength a fraction of the phosphorylated myosin forms a small polymer and that polymer formation is more noticeable in the presence of Ca2' . The presence of this polymer could be at least partly responsible for many of the observed effects of Ca2+. The viscosity measurements, for example, would be sensitive to the presence of polymers. Whether or not the enhanced formation of polymers in the presence of Ca2+ could explain the proteolysis results is not known. However, it is difficult to understand how a fractional polymer population could dominate the entire proteolysis time course and it should also be pointed out that the rate of proteolysis of monomeric 6 S myosin is very similar to that of the phosphorylated 10 S myosin plus Ca". In addition it should be emphasized that the enhanced polymer formation is in itself an indication of a Ca2+-induced change. A dramatic example of how conformation can alter aggregation properties of myosin is illustrated by the differences in aggregation for 10 S and 6 S myosins (75). The effect of Ca2+ on the ATPase activity of acto-HMM suggests that Ca2+-dependent changes other than polymer formation are implicated, although these remain to be defined.
Whatever the effects of Ca2+ prove to be, it is interesting that these are found only with 10 S phosphorylated myosin. Myosin in the 6 S conformation or dephosphorylated myosin in the 10 S conformation are not sensitive to Ca2+-induced changes and this argues against nonspecific Ca2+ effects. The definition of myosin as 10 S or 6 S is based on sedimentation velocity experiments. At low ionic strength ( i e . 85 mM KC11 and low concentrations of MgC12, the sedimentation profiles of phosphorylated and dephosphorylated myosin are similar (60,61) and it was assumed in both cases that the 10 S conformation is formed. On the other hand the proteolysis experiments suggest that the "10 S" phosphorylated myosin, plus or minus Ca2+, is distinct from the 10 S dephosphorylated myosin. Thus, the use of the term 10 S to describe both the phosphorylated and dephosphorylated forms implies an identity in conformation that may not be justified. It should be emphasized, however, that although the gross hydrodynamic properties of the two myosins are similar, the sedimentation coefficient of the phosphorylated myosin (at low ionic strength, etc.) was not calculated, and more exacting measurements might reveal a difference in sedimentation rates. Additional evidence to indicate a difference between 10 S phosphorylated myosin and 10 S dephosphorylated myosin is apparent from the actin-activated ATPase activities. Although the level of actin activation for phosphorylated myosin at low MgC1, concentrations (ix. operationally defined as 10 S) is low, it is considerably higher than the actin-activated ATPase activity of dephosphorylated myosin. Therefore, it is suggested that phosphorylated 10 S myosin in the absence of Ca2+ is distinct in some way from dephosphorylated myosin in the presence or absence of Ca2+. This difference becomes more evident when Ca2+ is bound to the phosphorylated 10 S myosin.
We suggested earlier (68, 69), based on limited proteolysis studies, that the interaction of subfragment 1 and subfragment 2 may be affected by the 6 S-10 S transition. In our opinion this interaction is probably more important in determining enzymatic activity than the folding and/or interactions involving the tail portion of the molecule. This could be the region that is altered by phosphorylation at low ionic strength and, as such, may not be easily detected by sedimentation velocity measurements.
It is important to establish if Caz+ influences the binding affinity of myosin to actin, as this may be pertinent to the attached or slowly cycling cross-bridges observed in fiber measurements (4, 5). Although we cannot be conclusive at this time, we feel it is unlikely that Ca2' binding to myosin is important in this process. Our results indicate that the range of Ca" concentrations over which the Ca2+ effects are observed are higher than that necessary to "activate" calmodulin. Murphy and colleagues (6,7 ) report that latch-bridges possess a greater sensitivity to Ca2+ than calmodulin. In other words the latch process would be subject to regulation at Ca" levels lower than that necessary to bind to calmodulin. The Ca2+ sensitivities of the two systems relative to the calmodulin threshold therefore are different. In addition, one would predict that the formation of attached cross-bridges is associated with an increased affinity of myosin and actin. We think that the changes observed for I C , can be explained by polymer formation and we do not find an increased affinity for actin in the absence of Ca2+, but rather a reduced affinity. It is possible that the effect of Ca2' on myosin filaments is different from that shown in our in vitro experiments. With this as a reservation, it can be proposed that our current data is not consistent with the involvement of Ca2+ binding to myosin in the formation of slowly cycling cross-bridges.