Caldesmon inhibits skeletal actomyosin subfragment-1 ATPase activity and the binding of myosin subfragment-1 to actin.

Smooth muscle contraction is controlled in part by the state of phosphorylation of myosin. A recently discovered actin and calmodulin-binding protein, named caldesmon, may also be involved in regulation of smooth muscle contraction. Caldesmon cross-links actin filaments and also inhibits actin-activated ATP hydrolysis by myosin, particularly in the presence of tropomyosin. We have studied the effect of caldesmon on the rate of hydrolysis of ATP by skeletal muscle myosin subfragment-1, a system in which phosphorylation of the myosin is not important in regulation. Caldesmon is a very effective inhibitor of ATP hydrolysis giving up to 95% inhibition. At low ionic strength (approximately 20 mM) this effect does not require smooth muscle tropomyosin, whereas at high ionic strength (approximately 120 mM) tropomyosin enhances the inhibitory activity of caldesmon at low caldesmon concentrations. Cross-linking of actin is not essential for inhibition of ATP hydrolysis to occur since at high ionic strength there is very little cross-linking as determined by a low speed sedimentation assay. Under all conditions examined, the decrease in the rate of ATP hydrolysis is accompanied by a decrease in the binding of myosin subfragment-1 to actin. Furthermore, caldesmon weakens the equilibrium binding of myosin subfragment-1 to actin in the presence of pyrophosphate. We conclude that caldesmon has a general weakening effect on the binding of skeletal muscle myosin subfragment-1 to actin and that this weakening in binding may be responsible for inhibition of ATP hydrolysis.

Smooth muscle contraction is controlled in part by the state of phosphorylation of myosin. A recently discovered actin and calmodulin-binding protein, named caldesmon, may also be involved in regulation of smooth muscle contraction. Caldesmon cross-links actin filaments and also inhibits actin-activated ATP hydrolysis by myosin, particularly in the presence of tropomyosin. We have studied the effect of caldesmon on the rate of hydrolysis of ATP by skeletal muscle myosin subfragment-1, a system in which phosphorylation of the myosin is not important in regulation. Caldesmon is a very effective inhibitor of ATP hydrolysis giving up to 95% inhibition. At low ionic strength (-20 mM) this effect does not require smooth muscle tropomyosin, whereas at high ionic strength (-120 mM) tropomyosin enhances the inhibitory activity of caldesmon at low caldesmon concentrations. Crosslinking of actin is not essential for inhibition of ATP hydrolysis to occur since at high ionic strength there is very little cross-linking as determined by a low speed sedimentation assay. Under all conditions examined, the decrease in the rate of ATP hydrolysis is accompanied by a decrease in the binding of myosin subfragment-1 to actin. Furthermore, caldesmon weakens the equilibrium binding of myosin subfragment-1 to actin in the presence of pyrophosphate. We conclude that caldesmon has a general weakening effect on the binding of skeletal muscle myosin subfragment-1 to actin and that this weakening in binding may be responsible for inhibition of ATP hydrolysis.
The force producing interaction between myosin and actin can be inhibited by modification of either the myosin filaments or the actin filaments. Activation of vertebrate striated muscle is mediated through the actin filaments. Binding of calcium to the troponin-tropomyosin complex causes a change in position of tropomyosin molecules on the actin filaments (2)(3)(4) which has the effect of increasing the rate of a process in the ATPase cycle, probably Pi release, that occurs after the binding of myosin to actin (5-9). Regulation of vertebrate smooth muscle contraction is primarily controlled by the state of the myosin filament. Many studies have shown that phosphorylation of myosin by the Ca2+-calmodulin-dependent enzyme, myosin light chain kinase, is essential for smooth muscle contraction (see, for example, Refs. [10][11][12][13]. The effect * This work was supported by Grant AM 35216 from the National Institutes of Health. A preliminary report of this work was presented at the Biophysical Society Meeting, San Francisco, CA, February 9, 1986 (1). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$ To whom correspondence should be sent.
of myosin phosphorylation in smooth muscle is similar to the effect of Ca2+ binding to troponin-tropomyosin in striated muscle in that a process which occurs after actin-myosin binding is accelerated (14). Similar mechanisms have also been demonstrated in uitro for other myosin-linked regulatory systems (15)(16)(17). Although myosin phosphorylation is required for smooth muscle contraction, it appears from physiological studies that another regulatory system may also be operative (18). The possibility of a second regulatory system in smooth muscle is also supported by the recent characterization by Sobue et al. (19,20) of an actin-binding protein that inhibits actomyosin ATPase activity. This protein, caldesmon, appears to be the same inhibitory factor identified independently by Marston and Smith (21). The possibility that caldesmon is a regulatory protein has been increased by the demonstration of its specific location on thin filaments of smooth muscle (22). Caldesmon has since been identified in skeletal and cardiac muscle raising the possibility that it serves a regulatory role in these muscles also (23,24).
Caldesmon has been shown to inhibit the actin-activated hydrolysis of ATP in uitro by skeletal muscle myosin (25,26) as well as smooth muscle myosin (27, 28) and smooth muscle HMM' (29). In addition, the source of actin may be either smooth muscle (26,30) or skeletal muscle (25,27,28). In the presence of tropomyosin, the reported level of inhibition of ATP hydrolysis ranges from 50 (28) to 95% (26), whereas in the absence of tropomyosin reports range from virtually no inhibition (28,30) to around 40% (25, 27). These studies were carried out at low actin concentrations and ionic strengths greater than 50 mM. Caldesmon has been shown to cross-link actin filaments (31) and this cross-linking has been held responsible for the inhibition of ATP hydrolysis (25) although in another study it was shown that the caldesmon concentration required for cross-linking is greater than that required for inhibition of ATP hydrolysis (32).
We have studied the effect of caldesmon on the hydrolysis of ATP by skeletal myosin subfragment-1 (S-1) and skeletal actin. While caldesmon is present in low concentrations in mammalian skeletal and cardiac muscle (23,24), our primary reason for choosing skeletal muscle S-1 is to avoid any possibility of interference from the myosin-linked regulation common to smooth muscle myosin. We observed that at low ionic strength caldesmon is an effective inhibitor of ATPase activity even in the absence of tropomyosin. This decrease in ATPase activity occurs in parallel with actin cross-linking and an inhibition of the binding of S-1 .ATP with actin. At high ionic strength, caldesmon has a diminished cross-linking activity but inhibition of ATP hydrolysis still occurs. Tropo-'The abbreviations used are: HMM, heavy meromyosin; S-1, myosin subfragment-1; EGTA, [ethylenebis(oxyethylenenitrilo)tetraacetic acid AMP-PNP, adenyl-5'-yl imidodiphosphate. 5711 myosin enhances the activity of caldesmon a t high ionic strength. Under all conditions the decreased ATPase rate is associated with a decreased association between S-l.ATP and actin. Finally, caldesmon also inhibits the equilibrium binding of S-1 to actin in the presence of pyrophosphate. This weakening in the binding of S-1 to actin by caldesmon may be the cause of the decreased rate of ATP hydrolysis.

MATERIALS AND METHODS
Myosin was isolated from the back and leg muscles of rabbits by standard procedures (33). S-1 was prepared by chymotryptic digestion of myosin (34). Actin was isolated by the procedure of Spudich and Watt (35) as modified by Eisenberg and Kielley (36). Tropomyosin was isolated from turkey gizzards by the method of Bretscher (31). S-1 was radioactively labeled by reaction with ["C]iodoacetamide (37) and applied to a small affinity column of ATP attached to agarose through C-8 (Sigma) equilibrated with 10 mM imidazole, 2 mM MgC12, 1 mM dithiothreitol. The column was washed with the same buffer and the "C-labeled S-1 was eluted with 0.1 M KCI. Radioactively labeled S-1 was used only for binding studies performed in the absence of ATP. Calmodulin was isolated from porcine brain by the method of Yazawa et al. (38). Caldesmon was prepared by a method involving modification of several published procedures (19,21,39). Fresh turkey gizzards were cleaned, minced, and homogenized on the middle setting of a Polytron homogenizer in 4 volumes of buffer A (20 mM Tris-HCI, 40 mM NaC1, 1 mM MgC12, 1 mM dithiothreitol, 1 mM EGTA, 0.2 mM phenylmethylsulfonyl fluoride, 0.03 mg/ml soybean trypsin inhibitor, and 2 pg/ml leupeptin, p H 7.5) containing 0.05% Triton X-100. After centrifugation, the myofibrils were washed in buffer A without Triton X-100. Caldesmon was extracted by homogenization for 15 s in buffer B (0.3 M KCI, 5 mM MgCI,, 50 mM imidazole-HCI, 0.2 mM phenylmethylsulfonyl fluoride, 0.03 mg/ml soybean trypsin inhibitor, 2 pg/ml leupeptin, 1 mM dithiothreitol, 10 p~ n-propyl gallate, p H 7.0). After centrifugation, caldesmon was harvested from the supernatant by ammonium sulfate precipitation between 30 and 50% saturation. The ammonium sulfate precipitate was dissolved in buffer C (0.5 M KCI, 20 mM Tris-HCI, 1 mM EDTA, 1 mM dithiothreitol, 1 mM EGTA, pH 8.0) and dialyzed against the same buffer. The solution was clarified by centrifugation a t 160,000 X g for 1 h and loaded onto a Affi-Gel Blue (Bio-Rad) column (30 ml of resin/100 g of tissue used) equilibrated with buffer C. After washing the column thoroughly, the caldesmon was eluted with buffer C containing a total of 1 M KCI. Fractions containing caldesmon were dialyzed against 2 changes of buffer D (100 mM NaCI, 10 mM imidazole-HCI, 1 mM dithiothreitol, pH 6.7). The caldesmon was loaded onto a column of cellulose phosphate (Whatman P11; 10 ml of hydrated resin/100 g of tissue) equilibrated with buffer D and washed thoroughly. Caldesmon was eluted by increasing the concentration of NaCl to 200 mM. Fig. 1 shows a sodium dodecyl sulfate gel of the caldesmon and other proteins used in this study. The high molecular weight contaminants located between caldesmon and S-1 in h n e A arise from myosin digestion. The heavily overloaded sample of pure caldesmon shown in lane B is virtually free of visible contaminants. Caldesmon isolated by this procedure does not become phosphorylated upon incubation with [y-'*P]ATP and calmodulin.
The concentration of caldesmon was determined by the Lowry method (40) using bovine serum albumin as a standard. Other protein concentrations were determined by their absorption a t 280 nm. The molecular mass of caldesmon was assumed to be 140,000 daltons.
ATPase Assays-ATPase rates were measured at 25 "C by the liberation of "' Pi from [y-"P]ATP (6). Each assay consisted of three time points. The conditions for the ATPase assays are given in the figure legends.
Binding Assays-The binding of S-1 to actin in the presence of ATP was measured by sedimenting the acto-S-1 in an ultracentrifuge and determining the free S-1 concentration by an NH:/ EDTA/ATPase assay as described earlier (6). Binding assays were performed in 1.0 ml of solution having a composition that is described in the figure legends. Caldesmon was added to actin or actin-tropomyosin and allowed to stir for several minutes prior to the addition of S-I. After a further mixing for 1 min at 4 'C, the ATP was added.
The solution was stirred for 1 min and placed into centrifuge tubes, brought to 25 "C, and centrifuged in a Ti-50 rotor a t 40,000 rpm for 20 min. All of the supernatant was removed and an aliquot was s-1" """7-Tropomysin FIG. 1. 1% Sodium dodecyl sulfate/8.5% polyacrylamide mini slab gel electrophoresis of the proteins used in this study. A, mixture of 4 pg of each of caldesmon, myosin S-1, actin, and tropomyosin. E, 16 pg of caldesmon following elution from the cellulose phosphate column.
used for S-1 determination. Caldesmon had no effect on the NH:/EDTA/ATPase rate of S-I. Binding of "C-labeled S-1 was done by the method of Greene and Eisenberg (41).
Actin Cross-linking Studies-The cross-linking of actin by caldesmon was estimated by a low speed sedimentation assay under the same conditions used for the binding studies (42). Fig. 2 shows that the normal actin-activated ATP hydrolysis is greatly inhibited by the addition of caldesmon at low ionic strength in the absence of tropomyosin. At 2.5 p~ caldesmon monomer, where the caldesmon to actin ratio is 0.1, the ATPase activity is about 70% inhibited.* Even greater inhibition is observed at higher concentrations of caldesmon. In experiments not shown here, we have observed that the addition of smooth muscle tropomyosin does not enhance the inhibitory activity of caldesmon under these conditions.

RESULTS
Since an inhibitor of actomyosin ATPase activity could function either by weakening the binding of myosin to actin in the presence of ATP or by inhibiting the rate of some process in the hydrolysis reaction, we determined the effect of caldesmon on the association of S-1 with actin in the presence of ATP. The results of these binding studies are also shown in Fig. 2. In the absence of caldesmon, roughly 50% of the S-1.ATP was bound to actin. The fraction of S-l.ATP bound decreased with increasing caldesmon concentrations in a manner that paralleled the inhibition of ATPase activity. Once again, when the ratio of caldesmon to actin was 0.1, the binding was decreased to about 30% of the initial level. Therefore, the decreased ATPase activity is most likely the result of a decreased binding of S-1 to actin. To demonstrate that the effect of caldesmon on the binding of S-1. ATP to actin is specific, we attempted to reverse this inhibition of binding by the addition of calmodulin. The inset to Fig. 2 shows that the inhibition of binding by caldesmon is, in fact, reversed by the addition of a large molar excess of calmodulin to the reaction in the presence of 0.5 mM CaC12. It is important to note that the curve drawn through the points in the inset is arbitrary.
Since the protein assay of Lowry et al. (40) does not give an absolute protein concentration, the ratios given must be considered approximate. We considered the possibility that the inhibition of binding caused by caldesmon may have been the result of aggregation of actin filaments by caldesmon. We were alerted to this possibility by our observation that addition of caldesmon to actin solutions under the conditions of Fig. 2 caused a large increase in turbidity, suggesting cross-linking of the actin filaments. Also, during the course of this investigation, reports of cross-linking of actin by caldesmon have been published (25, 31, 32). It was important, therefore, to quantitate the degree of cross-linking caused by caldesmon at low ionic strength. Fig. 3 shows the results of low speed sedimentation assays used to estimate the level of actin cross-linking as a function of caldesmon concentration. This experiment, done at 4 p~ actin, shows that caldesmon does cause actin to aggregate. Caldesmon inhibits ATP hydrolysis although the extent of inhibition is greater than in Fig. 2. This difference reflects a combination of different protein preparations used and the errors associated with measuring rates close to the rate of ATP hydrolysis by S-1 alone. The amount of actin sedimented at low speed parallels the degree of inhibition of ATPase activity at low caldesmon/actin ratios. Thus the possibility exists that the effect of caldesmon on both the ATPase activity and the binding of S-1 to actin may be the result of cross-linking of actin filaments. The inset to Fig. 3  The ionic strength dependence on the amount of actin sedimented at low speed is shown in Fig. 4. In the experiments shown here the caldesmon was dialyzed overnight against 17 mM imidazole-HC1, pH 7.0,7 mM MgCl,, 1 mM EGTA, and 1 mM dithiothreitol, sodium chloride was then added to each reaction mixture to give the required ionic strength. Whereas caldesmon is a potent cross-linker at low ionic strength, this activity disappears at ionic strengths greater than 50 mM. It is important to note that caldesmon/actin mixtures must be handled gently since aggregation can occur, even at high ionic strength, following vigorous mixing. Studies on the effect of caldesmon on ATPase activity and binding were repeated at high ionic strength where cross-linking did not occur. Fig. 5 shows the results of ATPase assays and low speed sedimentation studies at 125 mM ionic strength and 25 PM actin. At these conditions caldesmon remains an effective inhibitor of ATP hydrolysis in both the presence (open circles) and absence (solid circles) of smooth muscle tropomyosin. However, in contrast to low ionic strength experiments, tropomyosin does enhance the inhibitory activity of caldesmon. The ratio of caldesmon to actin required to produce 50% inhibition of ATPase activity is reduced from about 0.08 to 0.02 in the presence of tropomyosin in this experiment. I n further contrast to low ionic strength experiments, the inhibition of ATP hydrolysis is not accompanied by aggregation of actin. Even at 5 PM caldesmon (a caldesmon to actin ratio of 0.2) where there is 95% inhibition of ATP hydrolysis, no Actomyosin ATPase and Binding observable cross-linking of actin has occurred. Inhibition of ATPase activity is therefore not dependent on the crosslinking of actin filaments.

Caldesmon's Effect on
While the high ionic strength experiment of Fig. 5 shows that cross-linking is not a prerequisite for inhibition of ATPase activity, this condition is not suitable for binding studies since the strength of binding of S-1 to actin decreases with increasing ionic strength (43). To determine the relationship between effects on binding and effects on ATPase activity, it is necessary to measure both binding and ATPase rates under conditions where cross-linking does not occur. Fig. 6 shows the effect of caldesmon on the binding of S-1 to actin and the ATPase rate, at 67 mM ionic strength in the presence of tropomyosin. Fig. 6A shows experiments done at 50 p~ actin. Under these conditions where there is very little cross-linking of the actin filaments, increasing concentrations of caldesmon cause inhibition of both ATP hydrolysis and the binding of S-1-ATP to actin. At 6 p~ caldesmon, where the ratio of caldesmon to actin is 0.12, 87% of the actin is soluble (not shown) but the ATPase activity and the fraction of S-1. ATP bound are reduced to about 10% of their initial values. Although the binding data in Fig. 6A are quite scattered due to the difficulty in measuring changes in binding where the maximum fraction of S-1 bound is of the order of 20%, there is clearly a trend toward decreased S-1 binding with increasing caldesmon concentration. The decrease in binding caused by caldesmon is better shown in Fig. 6 B where the actin concentration was increased to 75 p~ and the fraction of S-1 bound at 0 caldesmon was about 0.3. Again, under this condition caldesmon decreases the fraction of S-1. ATP bound to actin and this decrease parallels the decrease in ATPase activity. At a ratio of caldesmon to actin of 0.04, both the ATPase activity and the binding are reduced to roughly 50% of their initial value. Therefore, caldesmon weakens the binding of S-l.ATP to actin and this decreased affinity may be responsible for the decreased ATPase activity.
Since caldesmon weakens the binding of S-l.ATP to actin one might expect that caldesmon would have a similar effect on the equilibrium binding of other forms of S-1. One convenient equilibrium system to study is the S-1 .pyrophosphate complex which has a binding constant between that for S-1. ATP and the very tight binding S-1-ADP complex (44). Fig.  7 is a plot of the fraction of S-1 .pyrophosphate bound to actin as a function of caldesmon concentration under conditions where actin cross-linking does not occur. In the absence of caldesmon the binding constant of S-1.pyrophosphate to actin is enhanced about 4-fold by smooth muscle tropomyosin, in agreement with a 3.6-fold increase reported by Williams and Greene (45) for skeletal muscle tropomyosin. As the caldesmon concentration is increased, the fraction of S-1. pyrophosphate bound decreases to a low value. It is difficult to determine the absolute degree of inhibition in either case because of the errors associated with measuring very low levels of binding. However, it is clear that marked inhibition of binding occurred in both the presence and absence of tropomyosin. At 1 ~L M caldesmon, where the caldesmon to actin monomer ratio is 0.1, the binding in both the presence and absence of tropomyosin is reduced to about 58% of the respective original values.

DISCUSSION
Two lines of evidence support the conclusion that the inhibitory effect of caldesmon on skeletal acto-S-1 .ATPase activity is primarily the result of weakened binding between S-1 .ATP and actin. First, under several different conditions, and whether or not actin aggregates are present, the decreased rate of ATP hydrolysis is accompanied by a decreased association between S-1. ATP and actin. In those cases where the binding can be accurately measured, the correlation between inhibition of ATPase activity and the binding of S -l . A T P is good. In all cases the degree of binding in the absence of caldesmon is consistent with our earlier results (6) and it is reduced to a very low value at high caldesmon concentrations. The effect of caldesmon on the binding is reversible as demonstrated by the addition of excess calmodulin. The effect of caldesmon on binding does not appear to result from gross aggregation of actin filaments since the aggregation would be seen in low speed sedimentation experiments. In fact, our results strongly indicate that inhibition of ATP hydrolysis and binding are not caused by actin aggregation. While this present work was in progress, Moody et al. (32) also showed inhibition of ATP hydrolysis in the absence of actin crosslinking at low actin concentrations using electron microscopy to estimate actin cross-linking. Our demonstration of the ionic strength dependence of the cross-linking may help to explain why ATPase inhibition may sometimes seem to vary together with the degree of cross-linking.
The second indication that caldesmon functions in this system by weakening the binding of S-1 to actin comes from equilibrium binding studies in the presence of pyrophosphate. It was shown in these studies that caldesmon greatly weakens the binding of S-1.pyrophosphate to actin. There are several advantages to studying the binding in the presence of pyrophosphate. The binding of S-1-substrate complexes is complicated by the presence of intermediate states. A change in the distribution of these intermediate states could cause a change in the observed binding constant. Since pyrophosphate is not a substrate for S-1, no intermediate states are formed and the binding is straightforward.
Also, since Lhe binding in pyrophosphate is stronger than in the presence of ATP, it is possible to accurately measure binding a t high ionic strengths and low actin concentrations where actin aggregation is not a problem.
Two related, but unanswered questions concern the possibility that a caldesmon-actin-S-1 complex can form and whether such a complex would be catalytically active. Preliminary results indicate that caldesmon can inhibit the ATPase activity of chemically cross-linked acto-S-1 by about 30%.3 While this is much smaller than the inhibition in binding that we observe, it must be considered a possibility that caldesmon has multiple inhibitory activities.
In a preliminary report Lash et al. (29) found that the inhibition of ATPase activity of smooth muscle HMM was associated with an increase in the binding of HMM.ATP to actin. While smooth and skeletal muscle myosins differ quantitatively in their strength of association with actin (46, 47) (smooth muscle myosin binds tighter than skeletal muscle myosin a t high ionic strength), qualitatively different effects of caldesmon are unexpected. Both myosins have similar mechanisms of actin-activated ATP hydrolysis (48) and the binding of both myosins to actin is increased as the nucleotide is changed from ATP to AMP-PNP to ADP (47). Based on this knowledge we would expect qualitative similarities in the effect of caldesmon on the binding of smooth and skeletal muscle myosin subfragments. The large difference observed between skeletal S-1 and smooth HMM may be due to: 1) differences between a single-headed subfragment and one with two heads, 2) to two different modes of attachment of the smooth and skeletal myosins to actin, or 3) to a kinetically controlled difference in the population of states between the two myosins. Further exploration of the difference between skeletal and smooth muscle myosins is likely to reveal much M. E. Hemric, C. E. Benson, and J. M. Chalovich, unpublished results. about the interaction of these contractile proteins with actin as well as clarify the function of the regulatory protein caldesmon.
A disputed topic is the requirement for tropomyosin in the inhibitory effect of caldesmon. Sobue et al. (28) using a smooth muscle myosin and skeletal actin, have reported that caldesmon functions primarily by reversing the potentiation of ATPase activity that occurs in the presence of tropomyosin. Caldesmon was also reported by Marston and Smith (30) to be relatively ineffective in the absence of tropomyosin when assayed with smooth muscle myosin and actin. In contrast, Ngai and Walsh (27), like Sobue et al. (28), used smooth muscle myosin and skeletal actin but reported significant inhibition by caldesmon in the absence of tropomyosin. Similarly, Dabrowska et al. (25), and Lim and Walsh (24) using skeletal myosin and actin, found that tropomyosin only enhanced the inhibitory activity of caldesmon. Using skeletal muscle actin and myosin S-1 we also find that caldesmon, by itself, is sufficient to inhibit ATP hydrolysis. Caldesmon is an inhibitor of ATPase activity and S-1 binding, whether or not tropomyosin is present. At very low ionic strength we have never observed a difference in the effect of caldesmon upon addition of tropomyosin. At 25 p~ actin and the highest ionic strength used, tropomyosin by itself increased the rate of ATP hydrolysis from 0.24 to 0.56 s-'. Caldesmon did not merely reverse this 2-fold potentiation of tropomyosin but reduced the rate to about 0.02 s-'. It therefore appears that while tropomyosin enhances caldesmon's inhibitory activity, caldesmon by itself is an inhibitor. One reason for our small tropomyosin dependence may be that the actin concentrations we use are much greater than those used by other researchers (23-26). If the saturation of actin with caldesmon is low due to a combination of high ionic strength and low protein concentration, tropomyosin may sufficiently strengthen the binding so that a large difference in the rate of ATP hydrolysis is observed. Smith and Marston (26) have, in fact, demonstrated an increase in the strength of binding of caldesmon to actin in the presence of tropomyosin. This is not to say that tropomyosin necessarily functions solely by strengthening the binding of caldesmon to actin. Moody et al. (32) have shown that when samples containing the same amount of bound caldesmon are compared, the one containing tropomyosin shows the greater degree of inhibition.
In summary, the inhibition of actin-activated ATP hydrolysis of skeletal S-1 by caldesmon shows the following properties. 1) Tropomyosin facilitates the activity of caldesmon possibly by enhancing the binding of caldesmon to actin but it is not essential. 2) Cross-linking of actin filaments is not necessary for inhibition of ATP hydrolysis. 3) Inhibition of ATP hydrolysis is associated primarily, and possibly completely, with a decrease in the affinity of S-1 .ATP and other S-1 species to actin.