Actin mediated regulation of muscle contraction

Pharmac.Ther.Vol. 55, pp. 95-148, 1992 Printed in Great Britain. All rights reserved 0163-7258/92 $15.00 © 1993PergamonPress Ltd Associate Editor: J. G. CORY ACTIN MEDIATED REGULATION OF MUSCLE CONTRACTION JOSEPH M. CHALOVICH* Department of Biochemistry, East Carolina University, School of Medicine, Greenville, NC 27858-4354, U.S.A. Abstract--Striated and smooth muscles have different mechanisms of regulation of contraction which can be the basis for selective pharmacological alteration of the contractility of these muscle types. The progression in our understanding of the tropomyosin-troponin regulatory system of striated muscle from the early 1970s through the early 1990s is described along with key concepts required for understanding this complex system. This review also examines the recent history of the putative contractile regulatory proteins of smooth muscle, caldesmon and calponin. A contrast is made between the actin linked regulatory systems of striated and smooth muscle. CONTENTS 1. Introduction 2. Hydrolysis of ATP by Myosin and Actomyosin 3. Regulation by Tropomyosin-Troponin 3.1. The tropomyosin-troponin complex 3.2. Conformationai changes of tropomyosin-troponin 3.3. Result of tropomyosin movement 3.4. The steric blocking model 3.5. The allosteric model of regulation 3.6. The nature of the regulated transition 3.6.1. Evidence from fiber studies 3.7. Challenges to the allosteric model of regulation 3.8. Weak and strong binding myosin crossbridges 3.9. Stabilization of weak and strong binding states 4. Complications of Actin Mediated Regulation 4.1. Multiple step binding of myosin to actin 4.2. Actin-tropomyosin conformationai states 5. Summary: Regulation by Tropomyosin-Troponin 6. Smooth Muscle Actin Binding Proteins 6.1. Caldesmon 6.2. Calponin 7. Conclusion Acknowledgements References 95 97 103 103 105 107 108 110 111 115 117 119 121 122 122 127 131 132 132 134 135 135 135 1. I N T R O D U C T I O N The ability to sustain directed movement is a fundamental criterion of life. We have, in our bodies, several types of 'molecular motors' which can convert chemical energy into the mechanical *Supported by Grants AR35216 and AR40540-01AI from the National Institutes of Health. Abbreviations--S-1, subfragment 1 of myosin--the globular catalytic region; HMM, heavy meromyosin-the large fragment of myosin with 2 catalytic sites; LC1, light chain 1; LC2, light chain 2; LC3, light chain 3; EDC, 1-(ethyl-3°(3-dimethylamino)propyl)carbodiimide; pPDM, N,N'-p-phenylenedimaleimide; ATP~S, adenosine 5'-(~,,-thio)triphosphate; ~ATP, 1, N6-ethenoadenosine triphosphate; ~ADP, 1, N6-ethenoadenosine diphosphate. 95 96 J.M. CHALOVICH Rod "I= chymotrypsin+ EDTAT J~4t Myosin ~Lchymotryps~4- MgCl2 HMM ~ Lldld FIG. 1. Basic structure of skeletal muscle myosin and its subfragments. Myosin (center) consists of a coiled coil tail (the coiling is not shown) which terminates in two globular head regions shown as solid ellipsoids. These globular heads or S-1 regions bind to actin and hydrolyze ATP. Each head contains two noncovalently bound polypeptide light chains: one LC2 (indicated by an asterisk), which can be phosphorylated and either LC1 or LC3. Myosin is insoluble as a result of aggregation of the tail region into thick filaments. Soluble fragments suitable for kinetic and binding studies can be produced by digestion of myosin with chymotrypsin. Digestion in the presence of MgCI 2 produces insoluble light meromyosin (LMM) and the soluble catalytically active fragment heavy meromyosin (HMM). Digestion in the absence of Mg2÷ produces an insoluble rod fragment and two S-1 fragments. Note that one light chain, LC2, is lost during the preparation of S-1 with chymotrypsin; the remaining light chain is either LC1 or LC3. energy of movement. The most obvious of these molecular motors is myosin which works in conjunction with the protein actin. Actomyosin, the complex of actin and myosin, is responsible for the movement of skeletal, cardiac and smooth muscles. The basic mechanism by which force or movement is produced by these three types of muscles is thought to be similar. However, important differences exist in the regulatory apparatus of these three muscle types. Thus a detailed knowledge of these regulatory systems provides the possibility of selectively intervening in the function of a single muscle type. Because muscle contraction requires the participation of both actin and myosin it is reasonable that regulatory systems might be directed toward both of these proteins. Both actin-linked and myosin-linked regulatory systems do exist. Actin-linked regulation involves muscle specific actin binding proteins which can control either the binding of actin to myosin or affect the ability of actin to participate with myosin as a cofactor in ATP hydrolysis. This type of regulation is common in both skeletal and cardiac muscle. The possibility also exists that a different type of an actin-linked regulatory system is operative in smooth muscle. Myosin-linked regulation in human muscle occurs by phosphorylation of the light chain components of myosin. This regulatory system is of greatest importance in smooth muscle and in actomyosin directed movement in nonmuscle cells. This review is limited to actin-linked regulation. The tropomyosin-troponin system of cardiac and skeletal muscle is heavily emphasized and caldesmon and calponin, of smooth muscle, are described briefly. The reason for this limitation is so that proper attention can be given to key concepts, such as ATP hydrolysis, weak and strong binding states of myosin and 'active' and 'inactive' forms of the actin filament. These concepts are important in all types of regulation but are most easily developed through a discussion of the tropomyosin-troponin system. The discussion of tropomyosin-troponin begins with the concept that Ca 2÷ binding to troponin causes changes in the troponin complex which ultimately change the interaction of tropomyosin with actin. Possible events by which this change in the tropomyosin-actin interaction can lead to Actin mediated regulation of muscle contraction 97 FIG. 2. Model of actin decorated with tropomyosin (top) and tropomyosin alone (bottom). Each actin monomer is shown as consisting of two globular domains. Actin monomers are arranged in a helical array and tropomyosin can bind along each groove of the helix (tropomyosin is shown binding along one of the two grooves in the figure). A single tropomyosin molecule spans 7 actin monomers. The actin structure can be thought of as two actin strands twisted about each other with crossover points every 350-380/~ with about 13 actin monomers between these crossovers. Tropomyosin is a coiled coil of two helical subunits. Tropomyosin tends to associate in a head to tail fashion as shown in the bottom panel. Courtesy of Dr George Phillips. activation of contraction are then considered in the 'steric' and 'allosteric' models of regulation. Possible step(s) of ATP hydrolysis by actomyosin which are finally activated by C a 2 + are considered in Sections 3.9, 4.1 and 4.2. An understanding of regulation of contraction requires a physical picture of the proteins involved. The structure of the regulatory proteins will be introduced as needed. The key players in all of the following sections are myosin and actin, so their structures will be introduced here. The protein myosin is schematically illustrated in Fig. 1. The helical tail region of myosin is responsible for the formation of the thick filaments in muscles and causes the formation of insoluble myosin aggregates in low ionic strength solution. The catalytically active regions of myosin are the two globular heads. Each head contains two noncovalently associated polypeptides known as light chains. One of these, light chain 2 (LC2), can be phosphorylated. The other light chain present on each head can be of two isoforms, light chain 1 (LC1) or LC2. The two myosin heads of a single myosin molecule are called a crossbridge since this unit bridges the gap between the thick myosin filaments and thin actin filaments in muscle. To avoid the problems associated with aggregation of myosin, solution studies are often done using proteolytic fragments of myosin known as heavy meromyosin (HMM) or subfragment 1 (S-l) (Cooke, 1989). The structures of these fragments are illustrated in Fig. 1. The detailed 3-dimensional structure of S-1 is not completely known but the work is in progress (Winkelmann et al., 1991). The protein actin exists in two forms, monomeric or globular and polymeric or fibrous. As shown in Fig. 2, fibrous actin has a helical structure (Milligan et al., 1990; Holmes et al., 1990; De Rosier, 1990; Egelman and DeRosier, 1991). Recently the atomic structure of a complex of actin and deoxyribonuclease I has been determined to 3 A resolution (Kabsch et al., 1990). Actin structure and properties are reviewed elsewhere (Carlier, 1991; Pollard, 1990; Vandekerckhove, 1990; Korn, 1982). 2. H Y D R O L Y S I S O F ATP BY MYOSIN A N D A C T O M Y O S I N An important concept in understanding movement and the regulation of movement is the mechanism by which myosin and actomyosin hydrolyze ATP. Myosin can, by itself, hydrolyze ATP at a low rate. It is only when myosin binds to actin that ATP is hydrolyzed rapidly and that movement is possible. 98 J.M. CHALOVICH Initial attempts to decipher the mechanism of ATP hydrolysis by myosin were done in the simple case where actin is absent. Numerous studies (Bagshaw et al., 1974; Koretz and Taylor, 1975; Taylor, 1977; Inoue and Tonomura, 1973; Chock and Eisenberg, 1979) on the kinetics of ATP hydrolysis by myosin (M) and its subfragments, have lead to the schemes similar to the following one: M + T ~ M - A T P ~ M * A T P ~ M * * A D P P i ~ M + ADP + Pi Myosin binds to ATP in at least two stages beginning with a weak collision complex and undergoing a conformational change to form a stable M*ATP complex. This complex, M*ATP, has an increase in myosin tryptophan fluorescence (indicated by the *). The binding to ATP is followed by a rapid cleavage of the phosphate anhydride bond which gives a further increase in fluorescence. This hydrolysis step is much faster than the subsequent steps in the reaction so that there is an initial rapid formation of a mixture of M-ATP and M-ADP-Pi (Lymn and Taylor, 1970). In assays using acid to quench the reaction, this myosin-bound Pi is released and gives rise to an apparent 'initial burst' of phosphate. The slow steady state rate of ATP hydrolysis by myosin was thought to be due to the slow release of products from the myosin (Taylor et al., 1970). The release of both Pi and ADP may be multiple step processes but these are not shown in the scheme. Rate constants for the steps in this scheme have been tabulated elsewhere (Woledge et al., 1985). In the presence of magnesium ions, actin causes a large acceleration in the rate of ATP hydrolysis by myosin (Maruyama and Watanabe, 1962). However, detailed analyses of the kinetics of actin activation are not possible in this system because of the complex intermolecular reactions known as superprecipitation occur during this process (Maruyama and Gergely, 1962). The kinetics of ATP hydrolysis by actomyosin are also difficult to study because they do not follow simple Michaelis-Menten kinetics. That is, the values of Vmaxand KATPaseare not unique but differ at high and low actin concentration ranges (Pope et al., 1981; Strzelecka-Golaszewska et al., 1979). The reasons for this complex behavior are not completely understood. Most studies of actomyosin ATPase activity have been done using a soluble proteolytic subfragment of myosin such as HMM or S-1 as described earlier (Eisenberg and Moos, 1967; Yagi et al., 1965; Sekiya et al., 1967). The ATPase activities of these subfragments follow Michaelis-Menten kinetics, with respect to actin concentration and the rates of ATP hydrolysis are 10 times faster than with myosin and are more representative of the rates in a working muscle (Eisenberg and Moos, 1968). Most of the studies described in this review deal with myosin subfragments. Another interesting experimental approach is to use synthetic myosin 'minifilaments' which have many of the advantages of myosin subfragrnents while still retaining the entire myosin molecule (Reisler, 1980). Early experiments with HMM showed that high degrees of activation of ATPase rates occurred under conditions where there was very little binding of actin to HMM (Leadbeater and Perry, 1963; Eisenberg and Moos, 1967). It was later suggested that the binding of ATP to HMM greatly weakened the binding of HMM to actin (Eisenberg and Moos, 1968). Although the binding of HMM-ATP to actin was weak, the rate of ATP hydrolysis increased in a hyperbolic manner as the free actin concentration was increased. Analysis of the actin concentration dependence using the Michaelis-Menten approach allowed the assignment of values of Vmaxand KAxPasefor this reaction. The large stimulatory effect of actin on rate was assumed to be due to an increase in the rate of product release. Lymn and Taylor subsequently observed that the addition of ATP to an actin-HMM complex caused a very rapid dissociation of HMM from actin (Lymn and Taylor, 1971). This rate of dissociation, at low protein concentrations, was about 10-fold faster than the phosphate burst or hydrolysis step. In fact, the rate of dissociation may be on the order of 5000 sec ~at 20 °C (Millar and Geeves, 1988). Lymn and Taylor suggested that hydrolysis of ATP to ADP + Pi occurred on the free HMM after dissociation from actin was completed. The M**DPi complex could then reassociate with actin, at a rate proportional to the actin concentration (Fig. 3A). Lymn and Taylor also showed directly that the rate of product dissociation from the acto-HMM-ADP-Pi complex was much faster than from the HMM-ADP-Pi complex. The Lymn-Taylor model was later modified as a result of the observation that a lower free actin concentration was required to reach 50% of the maximum ATPase (KAxw~)activity than required Actin mediated regulation of muscle contraction ® M ~--~ If AM M-ATP¢ I, t fsst ~ 99 .ADP . . . . • M, ADP...~,_ M.pi • . . . . It If ADP.. .AOP... A.M'p i ~ A ' M A'M-ATP slow ® M ._~ 1t AM M.pi I "" t fast _.~ ~ "'Pi, " .... It It A-M-ATP M-ATP M AM M-ATP._~ slow ADP ._~ . ,ADP . . . . • M,ADP , ADP~_ ,ADP A.M.Pil I . - ~ A'M slow K M ADP ~ M.AD P_.._91~ M.AD P K11 "Pil ~- "Pl," .... .ADP ADP KIO ADP K12 A'M-ATP _ . ~ A'M. Pil _ . ~ A'"'., Pill ~ A'M" "1-"..~ FIG. 3. Three models of ATP hydrolysis by myosin in the presence of actin from the 1970s through the 1980s. M is myosin or a myosin subfragment and A is actin in these schemes. Broken arrows are used to indicate a transition with a very low probability of occurrence. (A) The Lymn-Taylor model (Lymn and Taylor, 1971). (B) The refractory state model (Eisenberg et al., 1972; Mulhern and Eisenberg, 1976). (C) The 6 state model (Stein et al., 1979; Stein, 1988). Subscripts to Pi are used to indicate proposed conformational changes. to reach 50% binding of S-1 to actin in the presence of ATP (gbinding) (Eisenberg et al., 1972).* However, if product release were rate-limiting as in the Lymn-Taylor model, the apparent KATPase would equal the observed dissociation constant. To explain this and subsequent similar observations (Mulhern and Eisenberg, 1976; Marston, 1978; Wagner and Weeds, 1979) it was necessary to postulate that either the hydrolysis step (that is, the burst) was rate limiting or that an additional step was present in the cycle. Because the burst was observed to be a rather fast process Eisenberg and coworkers proposed that a rate limiting conformational change occurred just prior to product release (Stein et al., 1979, 1984, 1981; Stein, 1988). The model proposed by Eisenberg and coworkers contained an additional step shown in Fig. 3B as M-ADP-Pi(I) to M-ADP-Pi(II) and referred in older literature as M - A D P - P i ( R ) to M - A D P - P i ( R ) . t This 'refractory state' model allowed the essential features of the ATPase cycle to be maintained while still allowing for a high rate of ATP hydrolysis with little binding of S-1 to actin. That is, because the transition from state I to state II is slow, state I, which was thought to be refractory toward binding to actin, is the most heavily populated state. The presence of this extra state allows for successful modeling of both the difference between KATPase and Kbinding and the large phosphate burst when myosin is bound to actin. An alternate *All binding constants and Michaelis constants are expressed here as association constants. tThe suffixes (R) and (N) refer to the refractory and nonrefractory states, respectively. The term refractory was used to indicate that this state M'°DPi was refractory toward binding actin. It was later shown that this state did, in fact, bind to actin and the nomenclature was changed to (I) and (II). 100 J.M. CHALOVICH model, in which state II is omitted would be possible if the difference between KnTPasoand Kbi,ai,~ were small (Rosenfeld and Taylor, 1984) and if the size of the phosphate burst decreased substantially when myosin bound to actin (Tesi et al., 1990; Belknap et al., 1992). There continues to be a discussion over whether the hydrolysis step is rate-limiting or whether an additional step, following hydrolysis is rate limiting. One significant change in the original Lymn-Taylor model for which there is agreement regards the irreversible nature of the dissociation of acto-S-1 by ATP. Several studies suggested that myosin could bind to actin even in the presence of ATP. Marston postulated reversible binding of S- I - A T P to actin as a way of explaining biphasic curves of ATPase rate as a function of actin concentration (Marston, 1978). Another indication of binding of myosin to actin, in the presence of ATP, comes from the actin concentration dependence on the inhibition of resynthesis of ATP from ADP and Pi (Sleep and Hutton, 1980). Actin inhibits the resynthesis of ATP since actin binds more tightly to M - A D P than to M - A T P . To fit their experimental data, Sleep and Hutton assumed binding constants of 2 x 104 M- 1 for the binding of M - A T P to actin and 4 × 106 M- I for the binding of S-1-ADW ° to actin. Although not direct measures of binding, these experiments did suggest that S-1-ATP could bind to actin, although less tightly than does S-I-ADP. The first direct measurement of the binding of S-1-ATP to actin was done by Stein et al. (1979). They observed that upon mixing S-1-ATP with actin in a stopped-flow spectrophotometer that there was a very rapid formation of an acto-S-1 complex. The rate of this process was faster than the phosphate burst completed within 5 msec of mixing at 15 °C. At the 1 m u ATP used in this experiment the binding of ATP to S-1 or acto-S-1 and the subsequent dissociation of S-I from actin both have rate constants near 1000 sec-~ (Lymn and Taylor, 1971; Johnson and Taylor, 1978). Therefore, the binding observed was due to the rapid equilibrium binding of S-1-ATP to actin to form A-S-1-ATP. The binding observed by Stein et al. was due not only to S - I - A T P but also to S-1-ADP-Pi. This is because S-1-ATP and S-1-ADP-Pi have an equilibrium constant near 1 at low temperatures (Sleep and Taylor, 1976) and this equilibrium is rapidly established (Wagner and Weeds, 1979; Sleep and Boyer, 1978). In support of this, Stein et al. observed that the level of binding remained constant before and after the initial burst of ATP hydrolysis. Furthermore, the same initial level of binding was observed if S-1 was mixed with actin + ATP or if S-t was preincubated with ATP (giving the S-1-ADP-Pi state) prior to mixing with actin. This showed that both S-1-ATP and S-1-ADP-Pi bound rapidly to actin with similar binding constants. The amount of this binding increased with the actin concentration and extrapolated to 100% binding at infinite actin concentration.* Following complete hydrolysis of the ATP there was a further increase in binding since the S-1-ADP complex bound to actin more tightly than the S - i - A T P complexes. This binding of S-1-ATP and S-1-ADP-Pi to actin is much weaker than the interaction that occurs in the presence of ADP. In addition to showing that ATP did not lead to complete and irreversible dissociation of actomyosin, Stein et al. (1979) showed that ATP hydrolysis could occur at high actin concentrations where the S-1 remained bound to actin throughout the cycle (Fig. 3C). If cleavage of the terminal phosphate anhydride bond (the burst) could occur only when myosin was not bound to actin then the rate of ATP hydrolysis should be inhibited at very high actin concentrations when the binding of S - I - A T P to actin was favored. However, such inhibition was not observed (Chalovich et al., 1984b). The hydrolysis of ATP without mandatory dissociation of S-I from actin was confirmed by Mornet and coworkers (Mornet et al., 1981). They observed that S-1 could be covalently crosslinked to actin using a carbodiimide reagent, 1-(ethyl-3-(3-dimethylamino)propyl)carbodiimide (EDC). The covalent acto-S-1 complex hydrolyzed ATP at a rate similar to the Vmaxobtained at very high aetin concentrations. Thus ATP hydrolysis does not require that myosin detach from actin (see also Stein et al., 1985; Biosca et al., 1985; Rosenfeld and Taylor, 1984). It will be seen *Stein et al. (1979) were able to observe binding by using high protein concentrations and low ionic strength. In a muscle the actin and myosin are arranged in a lattice and binding reactions do not require diffusion of two proteins together but more nearly resemble an isomerization reaction. As a result, the interaction of myosin with actin in muscle occurs as if the actin concentration were very high; that is around 1 mM or higher (Brenner et al., 1986b). This very high 'effective actin concentration' cannot be duplicated in solution and so, low ionic strength conditions are used to favor association reactions. Actin mediated regulation of muscle contraction ~ ~ ATP ADP Pl .., ~ tADP 101 very slow .... ~ADP .,. J-Pi ,~ K9 "~' ~. f .~Fi ~K10 I weak ~ I , I , FIG. 4. A crossbridge model of ATP hydrolysis and force production. Actin is represented as a string of circles and a myosin S-1 group is shown projecting from the backbone of a myosin thick filament. Transition Kl0 is thought to represent the force producing event shown as a change in the binding of myosin to actin as Pi is released. later that this weak binding of S-1 to actin during ATP hydrolysis is a key point in the regulation of contraction. The model shown in Fig. 3C shows the nondissociating pathway of ATP hydrolysis. Note that this discussion of ATP hydrolysis is, by no means, complete. Rather those concepts essential for the discussion of regulation have been emphasized. For a more complete discussion of the mechanism of actomyosin ATPase activity, the reader is directed to several excellent reviews (Hibberd and Trentham, 1986; Stein, 1988; Homsher and Millar, 1990). A plausible description of the contractile cycle, based on the description of Eisenberg and Greene (1980), is given below (see Fig. 4). Myosin or actomyosin binds rapidly to ATP and the terminal phosphoanhydride bond is cleaved in an equilibrium reaction. This results in formation of an equilibrium among a number of 'weak binding' crossbridge states including M-ATP, AM-ATP, M - A D P - P i and AM-ADP-Pi. Evidence has been presented that M-ADP-Pi and A M - A D P - P i exist in two conformational states and that the transitions from M-ADP-Pil to M A D P - P i . and from AM-ADP-Pi] to A M - A D P - P i , are slow steps. Little is known about the properties of the states M - A D P - P i . and A M - A D P - P i . . The release of products occurs next and, since this step is very slow for myosin alone, appreciable flux through the cycle occurs only through the transition AM-ADP-Pi to AM-ADP. The phosphate release step is very important since this step is implicated in force production (Hibberd et al., 1985; Pate and Cooke, 1989; Metzger and Moss, 1991). The binding of myosin to actin is thought to change dramatically with this phosphate release. The relaxed state is often depicted as a '90 °' configuration of the crossbridge (Reedy et al., 1965) and following activation and Pi release the binding is depicted as a '45 °' attached state (Pringle, 1967; Huxley, 1969). The force-producing change may not actually be a change from a 90 ° to a 45 ° attached state; the force-producing transition has not been identified. However it is convenient to express force production as such a change as shown in Fig. 4. A central concept to the rotating crossbridge hypothesis of contraction is that each chemical state of the myosin crossbridge (i.e. M-ATP, M-ADP-Pi, M-ADP, etc.) have a preferred type of interaction with actin. Thus it follows that the rate and equilibrium constants for all reactions involving binding of myosin to actin and all transitions between two actomyosin species are thought to vary with the type of attachment to actin, or strain on the crossbridge (Hill, 1974; Eisenberg and Hill, 1978). In a muscle fiber, a single crossbridge may be constrained by the other crossbridges in the fiber or by an outside force. Thus a single crossbridge may not always be at its preferred configuration of attachment. Thus myosin states containing bound ATP or ADP + Pi are most stable (least strained) at the '90 °' configuration while states containing ADP or no nucleotide are most stable (least strained) at the '45 °' configuration. The change in configuration from the 90 ° to the 45 ° configuration is thought to produce force or movement. Although this is the favored working hypothesis, proving that such a conformational change is the force producing event is one of the great challenges in muscle research. Following the release of Pi, the muscle is in a force producing state. If allowed to shorten, the crossbridges will move to their new stable state of 45 ° type attachment. To complete the cycle, ADP is released to form the AM complex which then binds to ATP forming, once again, A M - A T P but 102 J.M. CHALOVICH now in a highly strained conformation. That is, while the A M - A T P state is most stable at a 90 ° attachment, it is now constrained in a 45 ° type attachment. This is an unstable situation. The rate of dissociation of M - A T P from A M - A T P is very fast and this negatively strained crossbridge can detach and reattach to another actin monomer in a more favorable orientation. It is interesting to note that M - A D P has been recently shown to attach and detach with actin, rather rapidly, in isometric muscle (Brenner, 1991). The implications of this observation are beyond the scope of the present review but are discussed in the report by Brenner (1991). Some care is necessary when attempting to compare kinetic measurements made in solution to kinetic measurements within a muscle fiber. Among other differences is that, in muscle, various transitions are dependent on the strain or 'angle' of attachment of myosin to actin; in solution no strain is possible and the myosin attaches to actin at its most favorable position (Hill, 1974; Eisenberg and Hill, 1978). Thus, the rate-limiting step in solution is always the same reaction step, whereas the rate-limiting step in muscle can change to a different reaction step depending on whether the muscle is contracting isometrically (no motion) or isotonically (with shortening). According to the 1957 crossbridge model of Huxley, crossbridges bind at a moderate rate ( f ) to actin, the force-producing conformational change occurs and the crossbridges detach rapidly (g2) at the end of this conformational change or power stroke (Huxley, 1957). During the conformational change, when the crossbridge is doing work, the rate of detachment (g~) is the slowest step. We now know that the crossbridge cycle is somewhat more complex than supposed at the time that this model was proposed. The rates of crossbridge attachment and detachment are now known to be too fast to be the transitions f and g. Rather, it is more proper to think o f f as the transition in the forward direction from the nonforce producing (i.e, M-ATP, AM-ATP, M-ADP-Pi, AM-ADP-Pi, etc.) to the force producing states (i.e. AM-ADP, AM); the reverse of these transitions,f_ is assumed to be very slow and is usually ignored. Similarly, g represents those processes which allow a return to the nonforce producing states, continuing in the forward direction and is not to be confused with f _ . Again, the reverse process, g is usually assumed to be very slow and is neglected. In terms of Fig. 4, attachment and detachment reactions occur in the vertical direction; the processes defined by f and g represent a cycle from (M-ATP + AM-ATP) to A M - A D P and continuing with release of ADP and finally rebinding of ATP to complete the cycle. Since the process described by ' f ' involves several sequential reaction steps it is common to refer to an apparent value o f f or fapp" The value Offapp in the fiber is similar to the rate constant for the rate limiting step of ATP hydrolysis in solution. This was shown by comparing the rate of force redevelopment of a lightly loaded shortening muscle following a quick restretch (Brenner, 1985) to the rate of ATP hydrolysis of myosin S-1 crosslinked to actin to maximize the ATPase rate (Stein et al., 1985; Mornet et al., 1981). The rate of these transitions was found to be similar in the fiber and in solution (Brenner and Eisenberg, 1986). When a fiber is contracting at maximum velocity the value of gapp is thought to be large. That is, the force-producing conformational change of myosin on the actin is allowed to occur without resistance and the gapp is thought to be very fast for negatively strained crossbridges (i.e. crossbridges at the end of the power stroke). In mechanisms such as that proposed by A. F. Huxley, the value ofgapp is dictated by the rate constant of detachment of myosin from actin (Huxley, 1957). This in turn is probably limited by the rate at which ADP is released from the myosin (ATP rebinding to actomyosin is very fast as is the subsequent detachment of myosin from actin). Siemankowski and White observed that at physiological temperature and ATP concentration, ADP release is slow enough such that the rate of dissociation of S-1 from actin could limit the rate of unloaded shortening in cardiac muscle (Siemankowski and White, 1984; Siemankowski et aL, 1985). A more extensive discussion of the dependence of rate constants on the displacement of myosin can be found elsewhere (Eisenberg et al., 1980; Eisenberg and Hill, 1978). The force, stiffness and ATPase activity of a fiber can be shown to be related to the values of fapp and gapp: Force = F = n *~'*fapp/(f~pp + gapp) Stiffness = S = n * ~*fapp/( lapp + gapp) ATPase = n * b *f app*gapp / ( f app "1- gapp) Actin mediated regulation of muscle contraction 103 where n is the number of active crossbridges per half sarcomere (the natural contractile unit), and ~ are the mean force and stiffness, respectively, for a crossbridge in a force producing state and b is the number of half sarcomeres in a given fiber (Brenner, 1988). 3. REGULATION BY TROPOMYOSIN-TROPONIN 3.1. THE TROPOMYOSIN--TROPONINCOMPLEX It is clear from the previous section that ATP hydrolysis and movement require the active participation of both actin and myosin. That is, while myosin is capable of cleaving the terminal phosphate anhydride bond of ATP, it remains the function of actin to facilitate the release of the products of hydrolysis. Because of the important role played by both actin and myosin, modification of either protein could modulate the cycle of ATP hydrolysis. The first actin-linked regulatory system discovered is the relaxing factor from vertebrate striated muscle (Ebashi and Ebashi, 1964). This relaxing factor was found to be composed of tropomyosin, discovered earlier by Bailey (1948) and an additional complex of proteins called troponin (Ebashi and Kodama, 1965). Tropomyosin exists as a dimeric ~ helical coiled coil (Caspar et al., 1969; Cohen and SzentGyorgyi, 1957). Tropomyosin binds to F-actin with a stoichiometry of 1 tropomyosin: 7 actin monomers (Bremel et al., 1972; Spudich and Watt, 1971) and lies along the groove of the actin helix (Hanson and Lowy, 1963; O'Brien et al., 1971). Figure 2 shows a model of tropomyosin and the actin-tropomyosin complex from the crystal structure of tropomyosin at 15 /~ resolution (Phillips et al., 1986). The crystal structure of tropomyosin has recently been determined to 9/~ resolution by X-ray diffraction (Whitby et al., 1992). The primary structure of tropomyosin has 14 groups of clustered acidic amino acid residues which are thought to be involved in binding to actin (McLachlan and Stewart, 1976; Parry, 1975). These acidic residues can be grouped into 2 sets of quasiequivalent sites, ~ and fl (Hitchcock-DeGregori and Varnell, 1990; McLachlan and Stewart, 1976). Only the ~ sites are thought to be regular enough to be important in actin binding (Phillips et al., 1986). However, genetic deletion analysis has been used to argue for 14 quasiequivalent actin binding sites (Hitchcock-DeGregori, 1992). The other component of the regulatory complex, troponin, was shown to consist of at least two polypeptides (Hartshorne and Mueller, 1968; Schaub and Perry, 1969). Three component polypeptides isolated from the troponin complex were later shown to be required for full Ca 2÷ regulation (Greaser and Gergely, 1971). These three components were named troponin I, troponin T and troponin C. In the presence of tropomyosin, troponin I inhibits ATP hydrolysis by actomyosin and the addition of troponin C neutralizes the effect of troponin I. Upon the further addition of troponin T, the system becomes fully Ca2÷-sensitive. It was later shown that the complex of troponin I and troponin T inhibits ATP hydrolysis even in the absence of tropomyosin (Eisenberg and Kielley, 1974). Troponin C prevents the inhibition of ATPase activity in either the absence of troponin T or tropomyosin but restores full regulation in the presence of the other components. Troponin T was thought to prevent Troponin C from neutralizing the inhibitory effect of TNT in the absence of Ca 2÷ Figure 5 illustrates the probable locations of the troponin components, tropomyosin and actin as viewed down the actin axis. The protein-protein contacts at high and low free [Ca 2÷] are shown. Another view of the actin-tropomyosin-troponin complex, which shows the location of tropomyosin and troponin within the actin helix is shown in Fig. 6. Readers are referred to other excellent reviews for greater details about these protein-protein interactions (Zot and Potter, 1987; Leavis and Gergely, 1984). Only an outline of the properties of the troponin subunits is given below. Troponin I binds to troponin C, troponin T and to actin (see Leavis and Gergely, 1984). Troponin I, alone can bind to actin in a 1:1 complex and inhibit ATP hydrolysis (Perry et al., 1972; Eisenberg and Kielley, 1974; Eaton et al., 1975). This inhibition is enhanced in the presence of tropomyosin and the stoichiometry of binding required for inhibition is reduced to 1 troponin I per 7 actin monomers, the same stoichiometry as the binding of tropomyosin to actin. A 21-residue cyanogen bromide fragment of troponin I, comprising residues 96 to 116 also inhibits actin 104 J. M. CHALOVICH °° ° .:Tm; ; Tm'," °°, "Tin" Fio. 5. Hypothetical view down the axis of an actin filament (A) showing the location of tropomyosin (Tm), troponin T (TnT), troponin I (TnI) and troponin C (TnC). The upper figure represents the state with two molecules of Ca 2÷ (c) bound to each troponin C at the Ca 2÷-specific regulatory sites. The other metal binding sites of troponin C probably contain bound Mg 2÷ (m). The broken circles, in the upper figure, indicate the proposed position of tropomyosin in the absence of Ca 2÷ . The bottom figure shows the proposed change in the troponin interactions when Ca 2÷ is removed from troponin C. Note particularly the proposed change in position of tropomyosin from its position at higher Ca 2÷ concentrations (shown as a broken circle in the lower figure). Drawn after an illustration given by Dr James Potter. activated ATPase activity (Syska et al., 1976; Wilkinson and Grand, 1978). A synthetic peptide of troponin I has been prepared which inhibits actomyosin ATPase activity; the activity of this peptide is enhanced by tropomyosin (Talbot and Hodges, 1979). In a later paper, Talbot and Hodges (Talbot and Hodges, 1981) varied the synthetic amino acid composition and showed that residues 105-114 are of particular importance for inhibition. Troponin T is an elongated molecule (Flicker et al., 1982) which binds to tropomyosin at two sites (White et al., 1987). The amino terminus of troponin T binds tropomyosin at 1/3 of the length of tropomyosin from its C-terminal end and extends beyond the C-terminal end and interacts with the adjacent tropomyosin; this can be seen in Fig. 6. Residues 1-70 of troponin T are implicated in the binding to the C-terminus and the overlap region of tropomyosin while residues 71-158 at the C-terminus and about 1/3 of the length of tropomyosin away from the C-terminus. Troponin T also binds to troponin I. In the presence of tropomyosin, troponin T can inhibit actomyosin ATPase (Chong et al., 1983). In the presence of tropomyosin, troponin I and whole troponin are equally potent inhibitors whereas twice the molar concentration of troponin T is required to reach the same level of inhibition. Troponin T does not bind to actin; it exerts its effects through some change in the tropomyosin. Troponin C reverses 50% or more of the inhibition of ATPase activity of troponin I, even in the absence of Ca: +; addition of Ca 2÷ totally reverses the inhibition. In the case of troponin T, troponin C does not affect the degree of inhibition unless Ca 2÷ is present and then about 50% reversal is achieved. Troponin C is a Ca 2÷-binding subunit. Troponin C binds to both the T and I subunits. As stated earlier, the troponin C - C a 2÷ complex reverses the inhibition of ATPase activity caused by the other inhibitory subunits. Troponin C has 4 Ca 2÷ binding sites (Potter and Gergely, 1975; Ikemoto et Actin mediated regulation of muscle contraction 105 FIG. 6. Model of the regulated actin filament in the presence and absence of Ca 2+ . As in Fig. 2, tropomyosin is shown in only one of the actin grooves. In contrast to Fig. 2, the troponin complex can be seen; it is shown in the presence of bound Ca 2+ (top) and absence of bound Ca 2÷ (bottom). Notice the slight change in position of tropomyosin relative to the actin monomers as well as to the troponin. This is a good illustration of how subtle the conformational change is which triggers contraction. Courtesy of Dr George Phillips. al., 1974) and it is the binding to the weak binding sites that is responsible for regulation (see Zot and Potter, 1987). Troponin C resembles calmodulin somewhat in structure (Babu et al., 1985; Herzberg and James, 1985; Sundaraligam et al., 1985). Notably, troponin C has the EF hand Ca 2+ binding site characteristic of the family of calcium binding proteins including calbindin and parvalbumin (Kretsinger, 1980). Fragments of troponin C have been characterized which are capable of restoring Ca 2+ sensitivity to the inhibition of ATPAse activity by troponin I alone or in the presence of troponin T (Weeks and Perry, 1978; Grabarek et al., 1981). 3.2. CONFORMATIONAL CHANGES OF TROPOMYOSIN TROPONIN The details of the events that occur between the binding of Ca 2+ to troponin C and the production of force in striated muscle are not known. However, this is an area of intensive investigation and an outline of the molecular events is being formed. We consider below some changes that occur upon the binding of Ca :+ to troponin C. An advance in the understanding of the troponin complex came with the determination of the crystal structure of troponin C (Herzberg and James, 1985, 1988; Sundaralingam et al., 1985). Figure 7 shows a representation of the crystal structure obtained in the absence of Ca 2+ In general, binding of Ca 2+ to the low affinity, regulatory binding sites (I and I1) are thought to form a structure similar to that of the structural, high affinity Ca2+-binding sites (III and IV). This results in a movement of helices B and C away from helices A and D. This could cause exposure of more potential sites of contact with troponin I and alter the binding of troponin C to troponin. Recent evidence suggests that troponin C may be folded so that the N-terminal and C-terminal regions are in close contact unlike the structure shown in Fig. 7 (Swenson and Fredricksen, 1992). 106 J. M. CHALOVICH ÷ I /oo III - FIG. 7. Proposed 3D structure of troponin C (Herzberg and James, 1985; Sundaralingam et al., 1985). Helicies labeled A through G are shown by cylinders. The two high-affinity Ca 2+ binding sites (III and IV) are located at the COOH-terminus and the low affinity, regulatory sites (I and II) are at the N-terminal region. Only two molecules of Ca 2÷ were seen in the crystal (shown as circles in regions III and IV). The bold letters X and Y represent Gln48 and Gln 82. The bond between Gln 48 and Gln 82 was introduced experimentally to test models of structural change (see text). Two experimental tests of this model have been done. Grabarek et al. (1990) substituted cys residues for Gln 48 and Gln 82 (X and Y in Fig. 7, respectively). Formation of a disulfide bridge between these two cys residues created a bridge between the B and C helices which prevented the opening of a cavity for the binding of Ca 2+ to site II. This had the effects of reducing the Ca 2+ affinity, of locking the troponin C into its inhibitory configuration at both high and low Ca 2+ and reducing the binding to troponin I. Normal activity of the troponin C was restored by reducing the disulfide bridge. Fujimori et al. (1990) took a similar approach to this problem. They produced mutants where either Glu 57 (helix C) or Glu 88 (helix D) were replaced by lys residues. These mutations caused small decreases in the affinity for Ca z + particularly for the regulatory sites. The formation of maximum tension of skinned rabbit muscle fibers whose troponin C was exchanged with these mutant troponin C's was shifted to higher concentrations of Ca 2+. The decreased regulatory function of these mutants was proposed to be due to the formation of a salt bridge in the mutants (either LysSV-Glu 88 or Glu57-Lys88). Such a salt bridge would conceivably stabilize the troponin C having no Ca 2+ bound to the specific regulatory sites. The changes in the structure of troponin C, upon binding to Ca ~+, result in changes in a number of other protein interactions. These many changes have already been described in detail (Leavis and Gergely, 1984; Zot and Potter, 1987). In general, all of the interactions among the 3 troponin subunits become stronger, in the presence of Ca 2+, while interactions of the troponin subunits with actin and tropomyosin become weaker. As shown in Figs 5 and 6, the tighter binding of troponin C to troponin I, in the presence of Ca z +, occurs at the loss of direct interaction between troponin ! and both actin and tropomyosin. In addition, the binding of troponin T to tropomyosin is weakened. The idea that has emerged is that, at low Ca 2+, the troponin complex keeps tropomyosin away from its preferred location of binding to actin. The binding of Ca 2+ to troponin C allows tropomyosin to bind to actin in its preferred location. The precise nature of all of the changes that occur with Ca 2+ binding are not yet known but there is a great deal of work being done on this area. This last event, the movement of tropomyosin on actin, was historically the first change detected in the actin-tropomyosin-troponin complex. The observation of a change in the binding of tropomyosin to actin came from careful observations of X-ray diffraction and optical diffraction Actin mediated regulation of muscle contraction 107 patterns of muscles and muscle proteins. Upon activation of a muscle there is no obvious change in the gross structure of the actin itself seen by X-ray diffraction studies (Elliott et al., 1967; Huxley and Brown, 1967). However, one reflection, the 2rid layer line, was observed in active muscle but not in relaxed muscles (Huxley, 1972). This 2nd layer line reflection was shown to be due to the tropomyosin molecule (O'Brien et al., 1971). Huxley confirmed that the 2nd and 3rd layer lines were strengthened by tropomyosin and proposed that tropomyosin forms a continuous strand along each of the two grooves of actin. An increase in the 2nd layer line and decrease in the third layer line was observed upon activation of frog skeletal muscles (Huxley, 1972; Vibert et al., 1972) as well as smooth and molluscan muscles (Vibert et al., 1972). These data were explained by models by which the tropomyosin molecule moved into a different position on the actin filament in the presence of Ca 2÷ (Spudich et al., 1972; Parry and Squire, 1973; Haselgrove, 1972; Huxley, 1972). In these models, actin was assumed to be spherical and tropomyosin was approximated by a cylinder. Tropomyosin was assumed to always be closely bound to actin and the movement of tropomyosin was restricted to an arc of constant radius from the center of the actin filament. With this simple model, the observed changes in the diffraction pattern could be reproduced by a change in position of tropomyosin to one closer to the actin groove in active muscle. The range of positions of tropomyosin in both the presence and absence of Ca 2÷ depended on further assumptions which differed in each model. In Haselgrove's model, tropomyosin was thought to move from an angle of 50° to one of 70 ° upon activation but a reasonably good fit could also be obtained if in the active state the tropomyosin was between 60 ° and 90 °. A recent computer generated model of the actin-tropomyosin-troponin (Fig. 6) clearly shows the proposed structural change of actin-tropomyosin-troponin. The change in tropomyosin position was thought to be smaller in molluscan 'catch' or vertebrate smooth muscle than in vertebrate striated muscle (Parry and Squire, 1973). The binding of Ca 2÷ alone, was sufficient for the observed changes in X-ray reflections in vertebrate skeletal muscle since muscles pulled out of overlap, so that no contact between myosin and actin is possible, give the same changes in reflections (Haselgrove, 1972). The change in position of tropomyosin was confirmed by 3-dimensional reconstructions of electron micrographs using optical diffraction methods (Wakabayashi et al., 1975). In this study, whole muscle was not used. Rather, to simplify the system, the structure of actin-tropomyosin (used as a model of the active filament) was compared to the structure of actin-tropomyosintroponin I-troponin T (used as a model of the inactive filament). It will be shown later that full activity of the actin-tropomyosin-troponin complex requires the binding of S- 1 to actin. Therefore the changes observed may not reflect a change from total inhibition to total activation. This is also a consideration in fiber studies where the fiber is pulled out of overlap and in most solution studies. As stated earlier, the change in position of tropomyosin on actin probably results from the Ca2÷-induced changes in the troponin complex. Phillips et al. suggested that in the presence of Ca 2÷, tropomyosin interacts favorably with actin primarily through the ct sites. In the presence of Ca 2÷ the troponin subunits interact strongly with each other and weakly with actin and tropomyosin (Phillips et al., 1986). Upon removal of Ca 2÷ an increased strength of binding of troponin to actin and tropomyosin occurs at the expense of a displacement of tropomyosin from its stable ~ site mediated binding to actin. In relaxed muscle, tropomyosin may be weakly and dynamically attached to actin. The question that arises is how this change in interaction of tropomyosin to actin regulates striated muscle contraction. 3.3. RESULT OF TROPOMYOSIN MOVEMENT The previous discussion showed that the binding of Ca 2+ to troponin C ultimately leads to a change in binding of tropomyosin to actin. This change in actin-tropomyosin must affect the interaction between actin and myosin in some way as to regulate contraction. Before considering evidence for different types of alteration of the actin-myosin interaction, it is helpful to consider what the possibilities for regulation are. This can be done by examination of the kinetic model shown in Fig. 4. One point that is apparent is that since tropomyosin-troponin binds to actin, only those processes which involve (a) the binding of myosin to actin, (b) a transition between two actomyosin states or (c) a transition between two actin conformational states (to be discussed later) J ~ 55/2--B 108 J.M. CHALOVICH can be directly regulated. Thus, none of the transitions occurring along the top line of Fig. 4 may be regulated. In contrast, a regulatory system which functions through the myosin molecule may conceivably affect any step in the cycle. Inhibition of the binding of M-ATP and M-ADP-Pi to actin would result in inhibition of ATP hydrolysis and contraction since both processes require binding of myosin to actin. Inhibition of the binding of M-ADP to actin, however, would not directly inhibit contraction since this occurs after the power stroke and after elimination of Pi. Regulation of the rate of transition between two actomyosin intermediates (bottom line of Fig. 4) can only be effective if that same transition is slow in the absence of actin. Inhibition of the rate of the burst step (M-ATP to M-ADP-Pi) would not be a very efficient regulatory event since the burst is equally fast when myosin is not bound to actin. Hydrolysis could then proceed by 'side-stepping' the unfavorable reaction. In contrast, regulation could occur by control of Pi release from AM--ADP-Pi to AM-ADP since phosphate release is very slow for detached crossbridges. Relaxed striated muscles are characterized by low force and low stiffness (low rigidity). Therefore, the most heavily occupied states of relaxed muscle must be detached states or weakly attached states that have rapid rates of dissociation (see Section 3.8). For example, inhibition of the rate of ADP release would not be a good candidate for regulation since this would result in the accumulation of the tightly bound actin-myosin-ADP complex resulting in a rigid muscle. Therefore, the most reasonable candidates for regulation are the binding of 'weak' crossbridges to actin and the kinetic transitions associated with Pi release from attached crossbridges. The first possibility is an example of competitive inhibition whereas the second possibility would be a noncompetitive type of inhibition. Another point to be considered is whether regulation involves RECRUITMENT or a GRADED RESPONSE (kinetic modulation). That is, for a muscle that is operating at 50% maximum activity, are 50% of the crossbridges active (recruitment, see Podolsky and Teichholz, 1970) or are all of the crossbridges 50% active (graded response or kinetic modulation, see Julian, 1969). For tropomyosin-troponin regulation this is equivalent to asking whether each group of seven actin monomers acts rather independently and can be 'active' or 'inactive' irrespective of its neighboring actin groups or whether all groups of seven actin monomers are in one of many possible states of activation. Recent mechanical studies suggest that the latter possibility is correct, that is activation of striated muscle by Ca 2÷ is due to kinetic modulation (Brenner, 1988). 3.4. THE STERICBLOCKINGMODEL Both competitive and noncompetitive mechanisms were recognized at the time of the discovery of the change in position of tropomyosin on actin (Wakabayashi et al., 1975). However, it was generally assumed, that the movement of tropomyosin competitively inhibited the binding of myosin to actin (Parry and Squire, 1973; Haselgrove, 1972; Huxley, 1972). This model is called the steric blocking model of regulation. Steric blocking is also a special case of recruitment since the inability of myosin crossbridges to bind to actin would result in a decrease in the number of crossbridges which are able to go through the contractile cycle. There were several reasons for assuming a steric blocking mechanism. First, as stated earlier, the type of conformational change in actin expected for an allosteric modulation was not observed upon activation, at least in early studies. Thus an allosteric change in the actin as a result of changes in tropomyosin position was considered unlikely (Wakabayashi et al., 1975). This is an important point in distinguishing the steric blocking model from other models of regulation. That is, in the steric blocking model, actin is considered to be static and tropomyosin is responsible for controlling access of myosin to the actin sites. A second indication of steric blocking came from studies done in the early 1970s which suggested that bound myosin crossbridges overlapped the tropomyosin binding sites on actin, in the absence of Ca 2+ (Moore et aL, 1970; DeRosier and Moore, 1970). Localizing the position of S-1 and tropomyosin on the actin filament in a muscle is no easy task as subsequent studies showed. Following the work of Moore et al. (1970), which supported a steric blocking mechanism, Seymour and O'Brien (1980) observed that the tropomyosin and myosin heads bind to opposite sides of the actin filament where steric blocking would be impossible. In the next year, Taylor and Amos (1981) provided a new interpretation which placed the myosin and Actin mediated regulation of muscle contraction 109 FIG. 8. View down the axis of an actin filament (A) showing the positions of tropomyosin in the 'relaxed' position (open broken circle) and in the 'active' position (circle T). Two proposed locations of the S-1 region of myosin are shown on the actin. Steric blocking of the binding of myosin would be possible in position S-1a (Taylor and Amos, 1981) but not in position S- 1b (Seymour and O'Brien, 1980). Allosteric regulation would be possible in either position. tropomyosin in close proximity once more where steric blocking could occur. The latter two orientations of tropomyosin and S- 1 are illustrated in Fig. 8. It is clear that steric blocking involves more than movement of tropomyosin; the movement must overlap the binding site of S-1 sufficiently to prevent myosin binding in relaxed muscle. The reader is referred to an interesting and brief summary of the changing position of tropomyosin and myosin (Squire, 1981). A further complication, for steric blocking, came from the observation that a single S-1 could be covalently crosslinked to two actin molecules with the crosslinking agent E D C (Mornet et al., 1981). This result was supported by subsequent image reconstructions (Amos et al., 1982; Wakabayashi and Toyoshim~, 1981). In this type of model, the tropomyosin could be located between the two myosin binding sites. This provides a tempting speculation that the tropomyosin movement could inhibit the transition from one type of bound state to another type of bound state rather than by inhibiting binding. At the time of postulation of the steric blocking model, the tropomyosin molecule was assumed to be rigid and statically bound to actin in either of two positions. However, detailed studies of the crystal structure of tropomyosin suggest a very different picture. The persistence length of tropomyosin (the longest distance over which molecular motion are linked) has been estimated to be 2-4 molecular lengths (Phillips and Chacko, 1992). Similarly, electric birefringence measurements of skeletal tropomyosin may be modeled as a semi-flexible rod with a persistence length of 150 nm or 3.6 molecular lengths (Swenson and Stellwagen, 1989). This persistence length places an upper theoretical limit on the cooperativity through tropomyosin, in the absence of bound S-I. Moreover, this persistence length is a manifestation of the high flexibility of tropomyosin (Phillips and Chacko, 1992). Tropomyosin cannot be treated as a rigid rod as assumed in early models of steric blocking. It is thought, for instance, that even if one actin molecule, of a group of 7 covered by tropomyosin, were blocked from binding to myosin, other actins within the group could bind to myosin. Thus, a strict steric blocking may not be possible. As the structures of S-I, actin and tropomyosin are becoming known to a higher level of resolution, it is becoming clear that there is not a complete overlap of S-1 and tropomyosin binding sites on actin. Milligan et al. (1990) recently used image analysis cryoelectron microscopy to localize the binding sites of S-1 and tropomyosin on the reconstituted protein complex. In activating conditions, S-1 was seen to form its major contact with the outer domain of actin and extend toward the inner domain of actin near the tropomyosin binding site. S-1 also binds to the top of the outer domain of an adjacent long pitch monomer in agreement with earlier cross linking studies (Marianne-Pepin et al., 1985; Mornet et al., 1981). A prediction of models of regulation which involve movement of tropomyosin on the actin filament is that this movement precedes force development in the muscle. To test this prediction, 110 J.M. CHALOVICH Huxley and coworkers measured the time course of structural changes in muscle during activation using synchrotron radiation (Kress et al., 1986). The second actin layer-line, thought to be due to reflect the position of tropomyosin, increases following stimulation and reaches half maximum intensity after about 17 msec. The equatorial 11 reflection, which monitors the attachment of myosin crossbridges to actin, reaches its half maximum intensity at 25-30 msec. Finally, tension reaches its half maximum intensity at 40--50 msec after stimulation. In agreement with earlier studies, the change in position of tropomyosin was not dependent on the binding of crossbridges to actin although a model in which Ca 2+ gives only partial activation with further activation occurring upon crossbridge binding could not be ruled out. These data strongly suggest that the change in orientation of the tropomyosin molecule is an important event in regulation since it precedes both an increase in crossbridge attachment and tension development. Huxley and Kress (1985) suggested that crossbridges first bind to actin in a 'preforce-generating' state whose binding is controlled sterically by the tropomyosin position on actin. Force was thought to be produced following transition into a 'force-generating' state. Brenner (1990) showed that the data of Huxley and Kress can be explained without postulating the existence of the 'preforce-generating' states but by assuming that force is produced from a transition from weak binding to strong binding states (Brenner, 1990). In this alternate model the affinity of weak binding crossbridges is only slightly altered by Ca 2÷ and the major regulatory event is the rate of transition from weak binding to strong binding states. The hypothesis that regulation of contraction does not occur by steric blocking is explored later in this chapter. Not all of the changes in the actin-tropomyosin-troponin complex precede myosin crossbridge attachment. Ishii and Lehrer have used the excimer fluorescence of rabbit skeletal tropomyosin labeled a t c y s 19° with N-(1-pyrenyl)-iodoacetamide to monitor tropomyosin-actin conformational changes (Ishii and Lehrer, 1990, 1991). No change in excimer fluorescence occurs until after myosin crossbridges bind to actin; Ca 2÷ itself has little effect (Ishii and Lehrer, 1991). It is also noteworthy that the X-ray diffraction changes attributed to movement of tropomyosin are seen in fibers pulled out of overlap so that there is no binding of myosin to actin. However, further changes in ATPase activity (Bremel et al., 1972; Murray et al., 1982; Nagashima and Asakura, 1982; Pemrick and Weber, 1976; Dancker, 1992; Cande, 1986; Meeusen and Cande, 1979) and in fiber contractility (Schnekenbuhl et al., 1992; Swartz et al., 1992) occur when crossbridges bind to the actin filament. Therefore, the observed X-ray changes may not reflect the total regulatory response. 3.5. THE ALLOSTERICMODEL OF REGULATION Although the steric blocking model is simple and intuitively pleasing, it does not readily explain all the experimental data. Weber and colleagues observed that under some conditions, tropomyosin-troponin can actually stimulate the rate of ATP hydrolysis above that of actin alone (Bremel et al., 1972; Murray et al., 1982, 1980). In the steric blocking model, stimulation of ATPase activity would be impossible since actin by itself was considered to be maximally active and tropomyosin could have either no effect (in Ca 2+) or inhibit ATP hydrolysis (at low free Ca 2+) by blocking the binding of myosin. Tropomyosin alone can also inhibit or potentiate (Williams et al., 1984; Lehrer and Morris, 1984; Eaton, 1976) the rate of actin-activated ATP hydrolysis. This potentiation is also evident in smooth muscle tropomyosin which is not normally associated with troponin (Williams et al., 1984; Lehrer and Morris, 1984; Chacko and Eisenberg, 1990). These observations are more readily explained by a model in which tropomyosin alters the conformation of actin in an allosteric fashion. It is interesting to note that maleimidobenzoyl-G-actin (Bettache et al., 1990) stabilized with phalloidin has a potentiated ATPase rate in the absence of tropomyosin or troponin (Miki and Hozumi, 1991). This raises the possibility that actin alone can be stabilized in a more active or less active state suggesting that conformational changes in actin are important. Other effects of tropomyosin on ATPase activity are most easily explained by postulating conformational changes in the actin caused by tropomyosin. Rabbit skeletal actin does not activate the ATPase activity of Limulus myosin whereas skeletal actin-tropomyosin does activate the ATPase (Lehman and Szent-Gyorgyi, 1972). Also, skeletal muscle tropomyosin produces partial inhibition of acto-HMM ATPase activity and this inhibition is increased by the addition of troponin I (Eaton et al., 1975; Wilkinson et al., 1972). Therefore, the effects of tropomyosin and Actin mediated regulation of muscle contraction 111 tropomyosin-troponin may be graded. It is possible to explain a graded inhibition by assuming that tropomyosin can adopt a continuum of positions which block the binding of myosin to varying degrees. However, there is no way to obtain enhancement of the rate over that with actomyosin alone by such a mechanism. Furthermore, the binding of S-1 to actin is actually enhanced by tropomyosin (Eaton et al., 1975; Eaton, 1976). Several experimental approaches have also given indication of a change in actin structure during activation. Yanagida and Oosawa (1980) measured the changes in polarization of fluorescence of the fluorescent ADP analog 1, N6-ethenoadenosine diphosphate (eADP) following its incorporation into F-actin. Single muscle fibers were extracted to remove myosin, tropomyosin and troponin (Yanagida and Oosawa, 1978) and eADP was then incorporated into the actin. Tropomyosin and troponin, but not myosin, were then added back to the fibers. The fluorescence polarization of the resulting fibers was Ca2+-sensitive. The angle of the base plane of eADP changed at pCa 2÷ of about 6 indicating a conformation change in the actin or a change in orientation of the actin monomers. The actin filaments were also more flexible in the presence of Ca 2+. The effect of tropomyosin-troponin on actin structure was also examined by polarized fluorescence of phalloidin-rhodamine-labeled F-actin in extracted fibers (Dobrowolski et al., 1988). The binding of myosin S-1 to such extracted fibers caused a change in orientation of the probe and an increase in the flexibility of actin thought to represent activation of the actin filament (Galazkiewicz et al., 1987). Yagi and Matsubara (1989) obtained evidence from X-ray diffraction of frog semitendinosis muscle fibers that an actin conformational change accompanied activation. The frog fibers were highly stretched so that interaction of myosin with actin was impossible. Upon activation, the equator, the second layer line at 1/18 nm-1 and the 5.9 nm layer line integrated intensity increased while the intensity of the first layer line at 1/36 nm- ~decreased by 48%. The first actin layer line partially overlaps the first myosin layer line at 1/43 nm-i. However, the changes were thought to be due to actin since the myosin layer line of highly stretched muscle does not change (Huxley et al., 1980; Yagi and Matsubara, 1980). The various changes in intensity could not be fitted by a model assuming the Egelman and DeRosier 2-domain structure of actin (Egelman and De Rosier, 1983) and treating tropomyosin as a cylinder with two positions on the actin filament even if the mass of troponin T (now known to be extended along the tropomyosin) were considered. Therefore, Yagi and Matsubara (1980) concluded that an additional conformational change in actin must occur. Kress et al. (1986) also observed that the 5.9 nm actin layer line changes simultaneously with changes in tropomyosin diffraction. The change in 5.9 nm layer line occurred even in overstretched muscles although to a lower extent. Tropomyosin movement, itself, should not contribute to a change in this layer line. Poppet al. (1991) have recently shown that by moving the small domain of actin, it is possible to duplicate the X-ray diffraction pattern of the actin-tropomyosin complex. These authors suggested that the thin filament conformational changes, seen during activation, are not due to tropomyosin alone but might also involve changes in actin structure. 3.6. THE NATURE OF THE REGULATED TRANSITION The key postulate of the steric blocking model is that the binding of myosin-ADP-Pi to actin is inhibited in the absence of Ca 2+. The first attempt at measuring the effect of Ca 2÷ on the binding of myosin-S-1 to actin-tropomyosin-troponin in solution was done using a stopped-flow device (Chalovich et al., 1981). It had been shown earlier that the weak binding of S-1-ADP-Pi to actin could be measured by observing the rapid increase in turbidity due to the binding of S-1 to actin in the presence of ATP (Stein et al., 1979). Rapid measurement of binding was required because the binding of S-1-ADP to actin is much tighter than the binding of S-1-ADP-Pi to actin and in the presence of Ca 2+, the rate of Pi release is very fast and could cause erroneously tight binding. The stopped-flow method was applied to the measurement of binding of S- 1 to actin-tropomyosintroponin in the presence of ATP in both the presence and absence of Ca 2+. Although C a 2+ increased the rate of ATP hydrolysis 25-fold, it resulted in only about a 2-fold increase in the binding constant Kbindi,g.This surprising result did not support the steric blocking model but rather indicated that the change in conformation of the actin-tropomyosin-troponin complex acted 112 J.M. CHALOVICH allosterically to regulate the rate of a process that occurred after myosin bound to actin. Shortly thereafter, another method of measuring the binding of S-1-ATP + S-1-ADP-Pi was developed which was based on the separation of free and bound S-1 species by rapid sedimentation in an ultracentrifuge (Chalovich and Eisenberg, 1982). This method gave excellent agreement with the stopped-flow method. Under conditions where 60% of the S-1 was bound to actin in either the presence or absence of Ca 2÷, the rate of ATP hydrolysis was more than 20-fold faster in the presence of Ca 2+. This showed that tropomyosin-troponin was capable of regulating the rate of some process that occurred after binding of S-1 to actin. These data alone, do not allow one to conclude that crossbridges are bound to actin in relaxed muscle. Such an extrapolation is difficult for several reasons: (1) Low ionic strength was used to promote binding so that the signal was large enough to measure accurately. In muscle the protein concentrations and the geometry favor binding so that it occurs even at much higher ionic strengths. These effects cause the 'effective' actin concentration, in muscle, to be quite high. In the rabbit psoas muscle, the effective actin concentration is 1.5~5.5 mM (Brenner et al., 1986b). (2) A single headed subfragment of myosin, S-I was used in the assays since myosin is insoluble at low ionic strength. (3) The geometry of binding is different in solution and in a fiber. Several efforts were taken to bridge the gap between these observations made in solution to a muscle fiber. Wagner and Giniger (1981) and Wagner and Stone (1983) confirmed the results of Chalovich et al. (198 l) on the binding of S-1 to actin-tropomyosin-troponin but observed different behavior with the two headed myosin fragment HMM. In the presence of Ca 2+, H M M bound with the same affinity as S-1. In the absence of Ca 2+ 33% of the H M M bound with the same affinity as in the presence of Ca 2+ but the remainder bound with a binding constant 1/20th of that in the presence of Ca 2+. Single headed H M M showed the same behavior. The greater sensitivity of binding of H M M and single-headed H M M was found to be due to the presence on intact LC2 which is lost during the preparation of S-1 (see Fig. 1). Phosphorylation of the LC2 did not further alter the binding. This should not be confused with phosphorylation of LC2 in smooth muscle myosin which appears to be required for actin activation of ATP hydrolysis. In this case phosphorylation is reported to have no effect on the binding of gizzard myosin (Sellers et al., 1982). In some types of smooth muscle, phosphorylation is reported to affect the KAvPaseand it has been argued that this is probably due to a strengthening of binding upon phosphorylation in these types of muscle (Wagner, 1986; Wagner and George, 1986). In a subsequent study, Wagner used urea gel electrophoresis to quantitate the amount of intact LC2 in the H M M and corrected the binding constant measured in the absence of Ca 2+ by assuming that those H M M molecules with a degraded LC2 (the degraded LC2 has an apparent MW of 17,000) will have the same binding constant as in the presence of Ca 2+ (Wagner, 1984). With these corrections, the Ca2+-dependence in binding became 10-fold. This difference was still too low to explain the difference in the rate of ATP hydrolysis meaning that tropomyosin-troponin affected a rate process. Furthermore, steady-state rate analyses indicated that the Vma× increased from 0.2-6.7 sec 1 (KATp,seincreased from 5 x 10 3 t o 5 × 10 4 M - I and Kbindingincreased from 4 x 10 3 t o 4 × 10 4 M - 1 ) , However, these data would suggest that a significant change in binding could occur in addition to changes in the rate of some process that occurs subsequent to the binding. Chalovich and Eisenberg (1986) also examined the binding of chymotryptic HMM, with intact LC2, to actin-tropomyosin-troponin. The binding constant, measured in the presence of Ca 2 +, was 3-fold higher than in the absence of Ca 2+ . Correction of these data for degraded LC2 brought the difference in affinity to 5-fold. A 'double binding' experiment was done to determine if there could be a larger difference in binding. In this experiment, the binding of H M M ATP to actintropomyosin-troponin was measured in the absence of Ca 2+. The supernatant, was presumably enriched in the fraction of H M M containing intact LC2 and having the weakest binding. This H M M was removed and the binding was repeated. Similar binding constants were obtained in both experiments indicating that the presence of LC2 does not have a dramatic effect on the Ca2+-sensitivity of binding. While there is a small difference in the binding of H M M to actin, the primary effect of tropomyosin-troponin was again to inhibit the rate of a process that occurred after binding was completed. An important consideration is whether a larger change in binding constant would occur at more physiological conditions, particularly at higher ionic strength. Inoue and Tonomura (1982) Actin mediated regulation of muscle contraction 113 measured the binding of S-1 and HMM to actin-tropomyosin-troponin in both the presence and absence of Ca 2÷ as a function of ionic strength. At low ionic strength, they observed a 25% increase in the fraction of myosin bound to actin in the presence of Ca 2+. With increasing ionic strength the fraction of S-1 bound, decreased in parallel, in the presence and absence of Ca 2÷ . Calculation of binding constants from the data indicated that at 100 mM ionic strength there was less than a 6-fold difference in the binding constant in the presence and absence of Ca 2+. This result is in reasonable agreement with other studies. At 50 mM ionic strength, Chalovich and Eisenberg (1982) observed an increase in association constant of 1.5-fold while the ATPase activity increased 55-fold. El-Saleh and Potter (1985) studied the ATPase activity and binding in ATP at 134 mr~ ionic strength, 25 °. Under these conditions the association constant increased from 2.5 × 1 0 3 tO 2.7 x 103 M-1 upon the addition of Ca 2+. Therefore, even at ionic strengths close to physiological, tropomyosin-troponin and Ca 2÷ have relatively little effect on the binding of myosin subfragments to actin. A question that is always raised regarding binding measurements is whether the binding is specific. The interaction between S-1 or HMM and actin-tropomyosin-troponin saturates at a 1:1 ratio of S-1 to actin (Chalovich et al., 1981, 1983; Chalovich and Eisenberg, 1982) and it is correlated with a biological function which is the increase in rate of ATP hydrolysis (Chalovich et al., 1981; Chalovich and Eisenberg, 1982). This is particularly evident from a study in the effect of light chain composition on the binding of S-1 to actin. Small changes in the binding of these isozymes of S- 1 to actin are reflected in changes in the rate of ATP hydrolysis; that is a weakening of binding is associated with a weakening of the KATPase(Chalovich et al., 1984b). The ionic strength dependence of the binding of S-1-ATP is similar to that of the tight binding of S-1-AMP-PNP to actin (Chalovich et al., 1983). Additional evidence for specificity comes from the competition of S-1-ATP and HMM-ATP with caldesmon binding to actin in solution (Chalovich, 1988; Chalovich et al., 1990) and in single fibers (Brenner et al., 1991; Chalovich et al., 1991a,b). It is important to realize that the allosteric model of regulation does not preclude a change in myosin crossbridge binding to active muscle. Upon activation there is a large increase in the population of strong binding crossbridges. These strong binding crossbridges will, in fact, dominate many of the properties of the muscle because of their higher affinity and slower binding kinetics. Steady-state kinetics of the inhibition by tropomyosin-troponin also indicate that regulation is not a simple competitive steric blocking (Chalovich and Eisenberg, 1982). At very low ionic strength, the Vmax for ATP hydrolysis is about 18-fold greater in the presence of Ca 2÷ but there is only about a 2-fold increase in the KATpas~(association constant). A large change in the Vm~xis also seen at 50 mM (Chalovich and Eisenberg, 1982) and 134 mM (E1-Saleh and Potter, 1985) ionic strength. This large change in the Vm~x shows that in the inhibited state actin is rather ineffective in accelerating a step associated with product even when bound to myosin. Together with the observations that Ca 2+ has little effect on KAvPase and Kbinding this indicates that actin can exist in two or more conformational states which differ in their ability to catalyze product release. In a test of the ability of tropomyosin-troponin to modulate a rate process directly, two laboratories studied the effect of Ca 2+ on the ATPase activity of the S- 1-actin complex which was covalently crosslinked with EDC (King and Greene, 1985; Rouayrenc et al., 1985). The rate of ATPase activity was 20-fold higher in the presence of Ca 2+ than in the absence of Ca 2÷ although there was no possibility of dissociation of S-1 from the actin. This result was obtained only on preparations with low ratios of S-1 crosslinked to actin (King and Greene, 1985). At ratios of S-i to actin of 2:10 the rate in the absence of Ca 2+ was 92% of that at high Ca 2+. This was thought to be due to 'turning on' of the actin filament (see Section 4.2). Earlier kinetic studies with HMM had competitive type inhibition patterns (Eisenberg and Kielley, 1970; Parker et al., 1970). However, these studies did not involve the use of very high actin concentrations thus making an estimation of kinetic parameters difficult. Kinetic studies with tropomyosin, in the absence of troponin, also are indicative of regulation by an allosteric type of mechanism. Neither smooth nor skeletal muscle tropomyosin is a steric blocker or competitive inhibitor of actin activated S-1 ATPase activity (Sobieszek, 1982). Skeletal muscle tropomyosin, reduces the ATPase activity to about half of its initial value while smooth muscle tropomyosin activates the ATPase activity by up to 100%. Potentiation of ATPase activity, above the rate obtained with S-1 and actin alone, cannot be explained by steric blocking. 114 J.M. CHALOVICH Furthermore, the inhibition of ATPase activity by pure skeletal tropomyosin is due to both a 6-10-fold increase in the Kmpase (i.e. an increase in the apparent affinity) and a 6-10-fold reduction in the Vm,x. The inhibition was not of the competitive type predicted by the steric blocking mechanism. While these data do not indicate the mechanism of inhibition by the intact tropomyosin-troponin complex they do show that tropomyosin alone acts as an allosteric uncompetitive inhibitor or activator. This supports the role of conformational changes in actin in the regulation of muscle contraction. The preceding discussion dealt specifically with the regulatory system of vertebrate skeletal muscle; similar conclusions have also been reached using cardiac muscle regulatory proteins (Tobacman and Adelstein, 1986). Evidence has been presented that regulation involves inhibition of a kinetic transition aside from any changes in binding of myosin to actin that may occur. Chalovich and Eisenberg suggested that Pi release could be the step regulated by Ca 2+ by considering the reactions within the dashed box of Fig. 4 (Chalovich et al., 1981; Chalovich and Eisenberg, 1982). In this simple scheme, the equilibrium between M - A D P - P i and A M - A D P is path independent. That is, K9 x Kt5 = K~4 ><K~0. The reaction given by/£9 cannot be Ca 2+ dependent and KI4 is only slightly Ca 2+ sensitive. The binding of myosin-ADP complexes to actin (Kts) is Ca 2+ dependent (Greene and Eisenberg, 1980). Therefore, to maintain detailed balance, the value of Kl0 must be Ca 2+ dependent. Several items mentioned above require comment. First, the conclusion that Pi release is regulated must be true only if the model shown in Fig. 4 is correct. As seen later this may not be the case. Second, the effect of Ca 2+ on the equilibrium binding reaction K15 does not imply a steric blocking mechanism. This is because this step occurs only after the force producing step and also because the change in K~5 of about 20-fold is not nearly great enough to produce the observed regulation. Note that reduction of the value of Vmax causes the same degree of inhibition at all actin concentrations. In contrast, a decrease in an association constant gives a degree of inhibition which is dictated by the free actin concentration. At any finite actin concentration, the degree of inhibition is less than the simple ratio of Kbmding(inhibited)/Kbindmg(active). Rather, the fraction of activity is given by the ratio (KbindJ,g(inhibited) + [S])/(Kb~,Omg(active)+ [S]). Third, a reduction in the association constant K~0 can occur either by a decrease in the rate of Pi release, by an increase in the rate of rebinding of Pi to M - A D P , or to a combination of the two. Fourth, while the rate of Pi release, or some related step might be regulated, it is also possible that additional kinetic transitions are regulated. There are some indications that the binding of nucleotides to actin-myosin, particularly in muscle fibers, is somewhat sensitive to C a 2+. Kraft et al. (1990) monitored the binding of nucleotides to crossbridges, by observing the characteristic changes in equatorial X-ray diffraction that accompany the detachment of crossbridges from actin. This detachment of crossbridges is directly the result of binding of nucleotides to the crossbridges. In the case of the ATP analog, ATP 7S, the equatorial intensity ratio of the two inner most equatorial reflections (I,/Lo) reached a plateau at 50-100 #M ATPTS in the absence of Ca z+ but at 10-20 mM in the presence of Ca 2+. The binding of ATP and ATPTS to actomyosin, in solution, also appears to be Ca 2+ dependent but to a lesser degree (A. M. Resetar and J. M. Chalovich, unpublished observation). The binding of nucleotide is probably not of direct regulatory significance since the ATP concentration in muscle is probably saturating in both the presence and absence of Ca 2+. Rather, the decrease in affinity of nucleotides, in the presence of Ca 2+, may reflect the more rapid rate of dissociation of nucleotides at high Ca 2+ concentrations. Rosenfeld and Taylor (1987) and later Stein and Chalovich (1991) measured the effect of Ca `-+ on the rapid burst or hydrolysis step ( A M - A T P to AM-ADP-Pi). Rosenfeld and Taylor measured this transition directly while Stein and Chalovich measured the rate of the fluorescence increase that occurs parallel to the burst. Both studies showed that the rate of the burst was insensitive to the concentration of Ca 2+. Rosenfeld and Taylor (1987) also measured the effect of C a 2+ on the rate of both substrate and product release from acto-S-l. In one experiment, actin-S-1 was mixed with a stoichiometric amount of ATP and allowed to 'age' for a brief period to produce a mixture of A M - A T P and A M - A D P - P i . To this mixture was then added an excess of 1, N6-ethenoadenosine triphosphate, (eATP), a fluorescent ATP analog. The change in fluorescence as eATP bound to acto-S-I was Actin mediated regulation of muscle contraction 115 limited by the rate of displacement of ATP and ADP + Pi. The rate of dissociation of A T P - and ADP + Pi Was reduced 10-20-fold in the absence of free Ca 2+ While these studies show that some step associated with Pi release is regulated, the Pi release step, itself, may not be the regulated step. This conclusion comes from the response of the isometric force produced by a single muscle fiber upon a rapid increase in the Pi concentration by laser flash photolysis of caged Pi (Homsher and Millar, 1990). The rate of the tension transient at 0.7 mu Pi was independent of [Ca 2+] suggesting that Ca 2+ does not directly affect the rate of Pi release. Moss' group subsequently observed a 3-fold Ca 2+ sensitivity of the rate of tension decline following a jump in Pi; the rate of this transition reached a plateau at high phosphate concentrations (Metzger and Moss, 1991; Walker et al., 1991). While the 3-fold regulation is not sufficient to explain the regulation of the system, the observed saturation indicates that Pi release occurs in more than one step. It is possible that the Pi release step, itself, is regulated to a small degree but this is not the major regulatory effect. The primary regulated step could be a conformational change which precedes Pi release. Possible regulated conformational changes will be discussed in Sections 4.1 and 4.2. 3.6.1. Evidence f r o m Fiber Studies If tropomyosin-troponin is able to inhibit ATP hydrolysis allosterically without a large change in the binding constant of myosin crossbridges to actin, then one should see attached crossbridges in relaxed muscle. Yet, it is the very fact that relaxed muscle has low stiffness that nurtured the concept of steric blocking. Biochemical measurements of the binding of myosin to actin, in the presence of ATP, have shown that the kinetics of binding are very fast (Stein et al., 1979; Chalovich et al., 1981). Thus, crossbridges in relaxed muscle may attach and detach very rapidly making the stiffness difficult to measure. To illustrate this point, it is helpful to consider how stiffness measurements are carried out (Brenner et al., 1982; Schoenberg et al., 1984). A single muscle fiber is mounted between a force transducer and a displacement generator (e.g. moving coil etc.). The muscle is stretched and the resulting force is measured. The distance of the stretch is determined from the optical diffraction pattern of the muscle. Stiffness is given as the ratio of dForce/dl. The stiffness is due both to connective tissue and to deformation of myosin bound to actin. The component due to crossbridges can be determined by measuring the stiffness at different extents of overlap of the thin and thick filaments (different sarcomere lengths). The value of the crossbridge stiffness is proportional to the number of attached crossbridges and is also dependent on the attachment and detachment rates of the crossbridges compared to the speed of the stretch. That is, during and subsequent to a stretch, there is detachment and reattachment of crossbridges to different actin monomers so as to return to the lowest energy equilibrium conformation (lowest crossbridge strain and zero force). If the time taken to make the measurement is long relative to the redistribution time then the stiffness will be underestimated. Actual measurements of stiffness of relaxed muscle have confirmed that at low speeds of stretch very little stiffness is observed. However, as the speed of stretch is increased, the stiffness also increases so that at the highest speed of stretch, the stiffness is about 33% of rigor stiffness (Brenner et al., 1982) where all of the crossbridges are attached (Cooke and Franks, 1980). Thus, under conditions of low temperature and low ionic strength a large number of crossbridges are attached to actin but the attached crossbridges produce no force. These crossbridges appear to be very similar to the weak binding crossbridges observed in solution during the steady-state hydrolysis of ATP. The number of crossbridges bound to actin does decrease at more physiological ionic strength. In relaxed muscle, at 20 mM ionic strength, about 65-95% of the crossbridges are attached. At 170 mM ionic strength roughly 2-10% of the crossbridges are attached in relaxing conditions while 10%, or more, of the crossbridges would be attached in active muscle (Brenner et al., 1986a). Several types of experiments confirm that the stiffness of relaxed muscle is truly due to bound crossbridges. The stiffness of relaxed fibers varies linearly with the degree of overlap of the thin and thick filaments and reaches a minimum value at zero overlap (Brenner et al., 1982). Equatorial X-ray diffraction patterns of relaxed muscles at low ionic strength confirmed the existence of extra 116 J.M. CHALOVICH mass associated with the actin filament as would occur with the binding of myosin crossbridges (Brenner et al., 1984). The number of weakly bound crossbridges decreases as the ionic strength increases from 20 to 100 mM although some crossbridges are attached even at 100 mM ionic strength. In a subsequent X-ray diffraction study, Yu and Brenner (1989) estimated that 30% of crossbridges are attached at 100 mM ionic strength while at 20 mM ionic strength 60% of the crossbridges are attached in relaxed muscle. Interestingly, the differences in X-ray diffraction pattern between relaxed and rigor muscle fibers could not be totally explained by a change in the fraction of attached crossbridges. Yu and Brenner suggested that in relaxed muscle there is a different conformation of the attached crossbridge from that in active muscle. Some of the characteristic properties of relaxed muscle are also seen in equilibrium states where crossbridges are clearly bound to actin such as in the presence of pyrophosphate. Like the stiffness of relaxed fibers, the stiffness of fibers in PPi decreases with increasing ionic strengths (Brenner et al., 1986a). Also, the observed stiffness of rabbit psoas fibers increases with the speed of stretch both with PPi and ATP (Brenner et al., 1986a). The difference between PPi and ATP is that much faster speeds of stretch (about 103-fold greater) are required to detect significant stiffness in the presence of ATP. Interestingly, in the presence of PPi, the stiffness is a multiexponential function of the speed of stretch indicating that crossbridges detach with a family of rate constants. In contrast to the stiffness of relaxed fibers, the stiffness is highly Ca 2+ dependent and is highly cooperative with the concentration of PPi. This cooperativity of binding is characteristic of strong binding states. Theoretical models of crossbridge binding suggest that a change in the attachment rate constant causes a vertical scaling of plots of stiffness versus the log of the speed of the stretch (i.e. the maximum observed stiffness decreases) while there is no change in the speed of stretch required to give the maximum change (Schoenberg, 1985; Tozeren and Schoenberg, 1986; Anderson and Schoenberg, 1987). In contrast, changes in the detachment rate constant cause a change in the speed of stretch required to give the maximum change. The maximum observed stiffness is unchanged as long as the detachment rate constant is less than the attachment rate constant but does decrease when the detachment rate constant becomes greater than the attachment rate constant (Brenner, 1990). Fitting relaxed stiffness/speed of stretch data at a series of ionic strengths showed that increasing the ionic strength causes a large decrease in the association rate constant and a small decrease in the detachment rate constant. The detachment rate constants for relaxed fibers are of the order 2 × 103 to 2 × 104 sec I (Schoenberg, 1988) while in the presence of ADP, the detachment rate constants are of the order of 0.01 sec ~. In rigor, the rate of detachment is too slow to measure. In the presence of A M P - P N P and PPi, the rate constants range from 0.01-10sec -~ and 0.1-100sec -l, respectively (Schoenberg and Eisenberg, 1985). In general, the effect of different nucleotides is to alter the rate of the detachment rate constant both in fibers (Schoenberg, 1985) and in solution (Marston, 1982). The detachment rate constants of relaxed crossbridges are equal to or greater than those measured in solution (Lymn and Taylor, 1971; Millar and Geeves, 1983). The range of rate constants is due to the different strain of different crossbridges resulting from the mismatch between the actin helix and the myosin repeat distance. Another indication that the stiffness and X-ray measurements are actually detecting bound crossbridges comes from the addition of an inhibitor of weak crossbridge binding to muscle fibers (Brenner et al., 1991; Chalovich et al., 1991a,b). The inhibitor used for such studies is caldesmon or the actin binding subunit of caldesmon (Sobue et al., 1981, 1982). Caldesmon and its 20 kDa actin binding fragment, are competitive inhibitors of the binding of myosin subfragments to actin (Chalovich et al., 1987). Fortuitously, the binding constant of caldesmon to actin is much stronger (about 1000-fold) than the binding constant of weak binding crossbridges to actin but is similar to the binding constant of strong binding crossbridges to actin, in solution (Velaz et al., 1989). Thus conditions can be chosen in which caldesmon totally inhibits attachment in the weak binding configurations with little effect on attachment in strong binding crossbridge configurations. Thus, caldesmon is not only a probe of attached crossbridges but is a probe of weakly bound crossbridges. The 20 kDa actin binding fragment of caldesmon behaves similarly and has the advantage that it does not bind to myosin (Velaz et al., 1990). The addition of caldesmon or the 20 kDa fragment of caldesmon to a relaxed fiber causes a decrease in the stiffness of a relaxed fiber to about 20% of its original value (Brenner et al., 1991). Actin mediated regulation of muscle contraction 117 The speed dependence of the observed relaxed fiber stiffness indicates that caldesmon causes a true decrease in binding of the crossbridges and not an increase in the rate of dissociation of weakly bound crossbridges. The displacement of mass from the actin filament was confirmed by X-ray diffraction studies showing a decrease in the IH/Ito ratio (Brenner et al., 1991). This result confirms the presence of attached crossbridges which are incapable of producing force. The weak crossbridges, that are bound in relaxed muscle, are intermediates in the crossbridge cycle. That is, inhibition of weak binding crossbridges, by the 20 kDa fragment of caldesmon, has the effect of inhibiting force production in parallel. Inhibition of weak binding crossbridges, without affecting strong binding crossbridges, is sufficient to inhibit force production. It is important to note that caldesmon is a rather specific inhibitor of the weak binding crossbridges. Thus, under conditions where relaxed stiffness is greatly inhibited, caldesmon has no effect on crossbridge stiffness in rigor or in the presence of magnesium pyrophosphate (MgPPi) (Brenner et al., 1991). The absence of an effect in the presence of MgPPi is important. In the presence of bound PPi, crossbridges detach at ionic strengths where force production in an active fiber remains high; force producing crossbridges must have an affinity greater than this (see Hill, 1974). While these experiments show the existence of weakly bound crossbridges in relaxed fibers, the conditions used are nonphysiological (5°C and 50 mM ionic strength). Recently experiments using caldesmon as a probe of weakly bound crossbridges have been done at ionic strengths as high as 170 mM and at temperatures as high as 20°C (Kraft et al., 1991). Weak binding crossbridges do remain bound to actin in relaxed fibers, under these conditions and these crossbridges are essential for the completion of the crossbridge cycle (Chalovich et al., 1991a). Mechanical measurements of single muscle fibers indicate that Ca 2÷ control is on the rate of transition between different bound crossbridge states (Brenner, 1988). The rate constant of force redevelopment (kr~dev) following a brief isotonic shortening with restretch to starting sarcomere length was shown to equalfapp + gapp, wherefapp is the rate constant form nonforce-generating-states to force-generating states, while gappis the rate constant for the return to nonforce-generating states. The value of kr~dev, which is independent of the number of attached crossbridges, was shown to increase as the level of activation of the muscle increased. This observation supports the concept that tropomyosin-troponin has a direct effect on the kinetics of crossbridge cycling. By measuring isometric force and ATPase activity, in parallel experiments, Brenner (1988) was able to show that the value Offapp was most sensitive to changes in Ca 2+ (see equations in Section 2). It is interesting to contrast these results with those obtained using fibers saturated with caldesmon. Altering the concentration of the actin binding fragment of caldesmon reduced isometric force but had no effect o n kredev. The contrast between the effects of caldesmon and tropomyosin-troponin in solution and in the fiber is striking. If caldesmon is functioning by a competitive binding mechanism then tropomyosin-troponin functions by a very different mechanism. The picture that has emerged, to this point, is that Ca 2+ binding to troponin C causes alterations in the interactions among the troponin components and tropomyosin. This results in a change in the position of tropomyosin on the actin filament. The position of tropomyosin in the relaxed state appears to partially overlap the strong myosin binding site but does not appear to have a major effect on the weak binding site. The movement of tropomyosin to the active site causes a small increase in the binding of weak states and a larger increase in binding of strong binding states. The most striking effect of the change in position of tropomyosin is an increase in the rate of transition from the weak binding states to the strong binding states. 3.7. CHALLENGES TO THE ALLOSTERIC MODEL OF REGULATION Studies with the ATP analog, ATP7 S, have been used as evidence for the steric blocking (Dantzig et al., 1988). The nucleotide analog ATPTS is hydrolyzed slowly by actomyosin but does not support tension development in skinned muscle fibers. ATPvS dissociates acto--S-1 at 10% of the rate of ATP (Goody and Hofmann, 1980; Millar and Geeves, 1988). The rate of hydrolysis of the phosphoanhydride bond of ATPy S is much slower than that of ATP (Bagshaw et al., 1972) leading to the hypothesis that ATPy S binds to actomyosin to produce a state analogous to AM-ATP. In contrast, the binding of ATP to actomyosin leads to an equilibrium distribution of AM-ATP and AM-ADP-Pi. 118 J.M. CHALOVICH While Ca 2+ had little effect on stiffness of single muscle fibers in ATP, the stiffness increased significantly at high Ca 2÷ concentrations in the presence of ATPy S (Dantzig et al., 1988). This led Dantzig et al. (1988) to suggest that the binding of the AM-ATP type state (as opposed to the AM-ADP-Pi type state) may be regulated by a steric blocking mechanism although the coexistence of an allosteric mechanism was not excluded. There are several reasons why different nucleotides could have different effects on fiber stiffness. First, although ATP~ S produces a weak interaction between S-1 and actin other criteria have not been used to confirm that this is a b o n a f i d e weak state. ATP7 S does behave quite differently from ATP in that the rate of hydrolysis of ATP~ S has little actin activation and is not regulated by tropomyosin-troponin (Dantzig et aL, 1988). Second, Ca 2+ can cause an increase in crossbridge stiffness by decreasing the affinity of the acto-S-1 complex for nucleotide, or by decreasing the on or off rates of crossbridge binding as well as by directly increasing crossbridge binding. Third, Ca 2÷ dependent changes in nucleotide binding to actomyosin may cause the appearance of regulation of crossbridge attachment because of the tight binding of rigor crossbridges. In fact, Ca 2÷ has been shown to weaken the binding of ATP 7 S to crossbridges in single muscle fibers (Kraft et aL, 1990). Thus, Kraft et al. (1990) observed that the fiber stiffness, in the presence of Ca 2÷, decreased as the concentration of ATP~S increased. At saturating concentrations of ATP~ S there is little difference in stiffness in the presence and absence of calcium. Therefore, the results with ATP 7 S are similar to those previously reported for ATP. Recent transient kinetic measurements indicate that tropomyosin-troponin has little effect on the binding of either M-ATP or M-ADP-Pi to actin (Stein and Chalovich, 1991). Measurements of both the burst magnitude (which gives the distribution between M - A T P + A M - A T P and M-ADP-Pi + AM-ADP-Pi) and the fraction of all S-I-ATP states bound to actin, showed that significant amounts of both M-ATP and M-ADP-Pi bound to actin in relaxing conditions, in solution. The binding of M-ATP to acti~tropomyosin-troponin is thought to change from 4.8 x 1 0 4 to 5.6 x 1 0 4 M - I upon the addition of C a 2+ while the binding of M-ADP-Pi increases from 1.5 × 104 to 1.9 × 104 M-'. Thus, none of the known weak binding states of ATP exhibit a large Ca 2+ dependence of binding. A potentially important point in the discussion of vertebrate striated muscle regulation is that variations in the contractile proteins and regulatory proteins could exist among species. All of the solution and mechanics studies which indicated an allosteric mechanism of regulation were done with rabbit skeletal muscle while frog muscle was typically used in the X-ray diffraction results and time resolved X-ray diffraction work used to support steric blocking. Different muscle types could differ in the Ca 2+ effect on the weak binding and strong binding crossbridge states and also on the kinetics of the transition from the weak to the strong states. It is important to determine how great the differences are between the rabbit and frog muscles. Schoenberg has shown that the number of weak crossbridges, in skinned single frog fibers, may be 1/3 to 1/5 the number in rabbit muscle (Schoenberg, 1988). In that experiment, the speed of stretch was not great enough to reach a plateau of stiffness. Therefore, part of the difference between rabbit and frog could be in the kinetics of crossbridge binding. Jung et al. (1989) also observed stiffness in relaxed frog fibers which was interpreted as weak binding crossbridges. They also observed that the population of weak binding crossbridges was less than in rabbit skeletal muscle, not exceeding 45% of rigor stiffness at low ionic strength. Jung et al. (1989) observed that rigor stiffness of the frog increased somewhat with the speed of stretch; correcting for this effect caused a reduction of the relaxed stiffness compared to the observed value. In contrast, Bagni et al. (1991) suggested that there may not be weak binding crossbridges in intact frog muscles. Rather, they interpreted the observed stiffness to be a viscous or viscoelastic response independent of crossbridge binding. Clearly, this problem requires more attention. In fish (turbot) muscle there is apparently both a decrease in the number of bound weak binding crossbridges in relaxed muscle and an increase in the rates of attachment and detachment of the crossbridges compared with rabbit muscle (Brenner, 1990). A weak binding type state also appears to exist in smooth muscle and nonmuscle cells. Turkey gizzard smooth HMM (Sellers et al., 1982) and Acanthamoeba myosin I (Albanesi et al., 1983) both bind to actin even when dephosphorylated and the rate of ATPase activity is very low. Acanthamoeba myosin II, which is inactivated by phosphorylation binds to actin even in the Actin mediated regulation of muscle contraction 119 phosphorylated state (Collins et al., 1982). Molluscan myosin, which is activated by binding directly to Ca 2÷ remains weakly bound to actin even in the absence of Ca 2+ (Chalovich et al., 1984a); the effect in molluscan muscle is, however, disputed (Sellers et al., 1991). The existence of weak binding to actin, in itself, is not proof that weak binding crossbridges exist in these muscles. Yet these observations hint that such states do exist in these muscle types. The observation that the 20 kDa fragment of caldesmon inhibits force production in chicken gizzard muscle is a further indication of weak binding crossbridges in smooth muscle (Pfitzer et al., 1992). 3.8. WEAK AND STRONG BINDING MYOSIN CROSSBRIDGES It is now known that myosin binds to actin both in relaxed and active muscle and evidence has been provided that the weak binding species that exists in relaxed muscle is an intermediate in the ATPase cycle. The weakly and strongly bound crossbridge states have quite different properties, many of which have already been discussed. The most obvious difference between these two states is that the weak binding state does not produce force whereas the strong binding state does. It is this distinction which makes the study of the weak and strong binding states exciting. The expectation is that by characterizing the properties and structure of these two states the nature of the force producing event will be revealed. In addition to the difference in ability to produce force, the weak state has very rapid binding kinetics (Lymn and Taylor, 1971; Stein et al., 1979; Chalovich et al., 1981; White and Taylor, 1976; Goldman et al., 1984) whereas the kinetics of binding of the strong states are somewhat slower (Lymn and Taylor, 1971; White and Taylor, 1976; Marston, 1982; Konrad and Goody, 1982). This difference in binding kinetics is primarily due to the increased rate of detachment of the weak state. Similar observations have been made in muscle fibers and have been described in detail by a series of papers by Schoenberg (Schoenberg, 1985; Anderson and Schoenberg, 1987; Tozeren and Schoenberg, 1986). Part of the difference in detachment rates may be due to the binding of both myosin heads to actin in the absence of ATP and the possibility that the stiffness could be the same for single and two headed attachment (Schoenberg and Eisenberg, 1985; Schoenberg, 1988); that is, decay of rigor stiffness may require the improbable simultaneous detachment of both heads. Two additional points should be made regarding the rate constants of the weak and strong binding species. The first is that changes in the rate of dissociation of acto-S-1, by the binding of different nucleotides is associated with similar changes in the dissociation rates of the nucleotides upon binding of S-I to actin (Taylor, 1989; Goody and Holmes, 1983). Second, while binding of nucleotides to S-1 and increases in ionic strength both weaken the interaction with actin these effects occur by different mechanisms. Indications from both fiber studies (Schoenberg, 1988) and solution studies (Marston, 1985; Konrad and Goody, 1982) indicate that ionic strength effects attachment rate constant whereas the nucleotide affects the detachment rate constant. Another property mentioned earlier is that the association of the weak binding states to actin is weaker than that of the strong binding states under identical conditions o f measurement. Both the weak and strong interactions are ionic strength dependent so that increasing the ionic strength from 20-170 mM decreases the association constant of both by about 100-fold (Greene et al., 1983). It is, therefore, possible to make the weak states bind relatively tightly and the strong states bind weakly. In fact, weak binding states can be distinguished from strong binding states even when they are compared under conditions of equal affinity to actin (Chalovich et al., 1983). A comparison between the binding of pPDM-S-1, at low ionic strength and S-1-AMP-PNP, at high ionic strength, indicates that while the strength of binding is similar the S-1-AMP-PNP complex binds cooperatively whereas the pPDM-S-1 complex does not. The binding of weak states of S-1 and HMM has little Ca 2÷ sensitivity (Hill et aL, 1981; Chalovich and Eisenberg, 1982) whereas binding of strong states is Ca 2+ sensitive (Greene and Eisenberg, 1988; Brenner et al., 1986b; Greene, 1986; Greene and Eisenberg, 1980; Greene, 1982). A corollary is that weak binding is noncooperative (Chalovich et al., 1983; Greene et aL, 1986) whereas strong binding is cooperative (Greene, 1986; Greene and Eisenberg, 1980). This cooperativity is observed as a sigmoidal curve of either 0 (S-1 bound/actin total) or of the ATPase activity as the concentration of free S-1 is increased in the presence of tropomyosintroponin and absence of Ca 2+. This is a highly diagnostic criteria for weak and strong states and 120 J.M. CHALOVICH is observed independently of the strength of binding of S-1 to actin. These observations indicate that the actin-tropomyosin filament, in addition to the myosin, has different states (active and inactive). The significance of this actin-tropomyosin transition will be discussed later. Since tropomyosin interferes with the binding of strong binding states to actin but has little effect on weak binding states, it is likely that the areas of interaction between myosin and actin are different in these two types of states. However, this difference need not be great. In fact, while the terms '45 ° and 90 °' are used to represent the different actin-myosin states, most researchers believe that the actual transition is quite subtle. Only a limited region of myosin and actin may be involved in the force producing conformational change. Similarly, Marston's group reported that the addition of AMP-PNP (Yount et al., 1971) to a rigor (nucleotide free) fiber caused a small increase in length of the fiber (Marston et al., 1976, 1978). This was interpreted as a partial reversal of the force producing conformational change that results in force production; this result is disputed, however (Schoenberg, 1989). Negative stained acto-S-1, crosslinked with EDC, has a very different appearance in the presence and absence of ATP indicating two distinct types of binding of S-1 to actin in the weak and strong binding states (Craig et al., 1985). This was also shown by electron microscopy of EDC crosslinked pPDM modified S-1 which is a stable weak binding state (Applegate and Flicker, 1987). These results were confirmed by fluorescence energy transfer between SH1 thiol of S-1 and c y s 374 of actin in the crosslinked actin-S-1 complex (Arata, 1986). The 50/20K junction of S-l, in the crosslinked actin-S-1 complex, is resistant toward cleavage by trypsin in the presence of ADP but not in the presence of ATP or ATPy S (Duong and Reisler, 1989). Electron paramagnetic resonance studies of spin labeled myosin indicate that in the presence of ATP the attached myosin heads have considerable disorder on the # sec time scale. This has been observed both at low ionic strength (Berger et al., 1989), where weak binding is favored and with crosslinked acto-S-1 where dissociation is impossible (Svensson and Thomas, 1986). This rotational disorder was also shown to exist in fibers in which crossbridge attachment was verified by rapid stiffness measurements (Fajer et al., 1991). In contrast, rigor crossbridges have a high degree of order and are relatively immobile; the addition of ADP causes very little change in the electron paramagnetic resonance spectrum (Fajer et al., 1990). Further evidence for a structural change between the weak and strong states comes from fluorescence resonance energy transfer measurements which show that the distance between c y s 374 on actin and c y s 177 o n light chain 1 of myosin S-1 changes from 6 nm in rigor to < 3 nm in the weakly attached state in the presence of ATP (Bhandari, 1985; Trayer and Trayer, 1988). Also, there is evidence for a rather large volume change during the transition from the weak to the strong state (Geeves, 1991b). The volume change upon this isomerization is of a similar magnitude to that of the denaturation of myoglobin and indicates a rather large structural change. Evidence for crossbridge rotation does come from X-ray diffraction studies of muscle (Huxley et al., 1980). Upon activation, the myosin heads lose the helical symmetry characteristic of the thick filament indicating that there is a change in attachment which occurs upon activation. Fluorescence probes on myosin are also consistent with rotational transitions between actin-attached states (Borejdo et al., 1979; Burghardt et al., 1983). Similarly, a fluorescent probe on the SH2 thiol of myosin is ordered both in ATP and ADP while the orientation of the probes differ (Ajtai et al., 1989). However, other studies have failed to detect a rotational change of a probe on SH 1 of myosin upon stretching a rigor muscle (Cooke, 1986; Thomas, 1987). These negative results are as important as the positive results in determining which regions of actin and myosin are changing. Considerable effort is being expended toward mapping the areas of contact of actin and myosin in the weak and strong binding crossbridge states. The area of actin near residues 1-7 of the sequence are important for the binding of myosin in the presence of ATP (DasGupta and Reisler, 1989; Bertrand et al., 1989) and also for the binding of troponin I (Grabarek and Gergely, 1987; Levine et al., 1988). This region does not appear to be a major area of interaction with myosin in the absence of ATP (Mejean et al., 1987; Miller et al., 1987) and there is an intermediate effect with AMP-PNP, PPi and ADP (DasGupta and Reisler, 1991). We can expect to know much more about the interaction between actin and myosin in various states in the near future as a result of detailed stractural studies of actin (Kabsch et al., 1990; Holmes et al., 1990), actin-tropomyosin (Milligan et al., 1990; Flicker et al., 1991) and myosin S-1 (Winkelmann et al., 1991). Additional sites are becoming candidates for interactions in the weak Actin mediated regulation of muscle contraction 121 and strong states. It is possible that surfaces rather than point sites are responsible for each type of interaction. In this regard, it is interesting that site-directed mutagenesis of the 1-7 region of actin, thought to be important in the weak binding, is not as devastating as one would expect on the ATPase activity (Cook et al., 1992, 1993). On the other hand, substitution of lysine for asp 3 and asp 4 of fl actin reduced rigor binding (Aspenstrom and Karlsson, 1991). The technique of site directed mutagenesis in combination with nonmuscle motility assays and ATPase assays is also being done with Dictyostelium actin (K. Sutoh, personal communication). These types of studies should be of great help in mapping the areas of interaction of actin and myosin. 3.9. STABILIZATION OF WEAK AND STRONG BINDING STATES Several models of weakly and strongly binding states have been used by a number of researchers. Because they have many useful applications in the study of regulation and force production several of these paradigms are listed below. A very useful weak binding analog is that formed by introducing a chemical bridge between two SH groups of myosin S-1. Reaction of S-1 with N, N'-p-phenylenedimaleimide (pPDM) crosslinks the SH1 and SH2 groups of myosin (Reisler et al., 1974) and traps a molecule of ADP (Wells and Yount, 1982). Such pPDM-S-1 has a very low rate of ATPase activity. The pPDM-S-1 complex resembles a weak binding state in terms of the circular dichroisrn spectrum and in its lack of binding to actin at high ionic strength (Burke et al., 1976). In contrast, the intrinsic fluorescence of pPDM-S-I resembles a strong binding state (Burke et al., 1976; Perkins et al., 1981). More recent studies support the view that pPDM-S-1 resembles a weak binding state. Thus, pPDM-S-1 binds to actin or actin-tropomyosin-troponin with about the same affinity as S-1-ATP and exhibits little Ca 2+ effect of binding (Chalovich et al., 1983). Also, the 50/20K junction of pPDMS-1 is susceptible to tryptic cleavage when crosslinked to actin with EDC just as in the case of crosslinked actin-S-1 in the presence of ATP (Duong and Reisler, 1989). The 50/20K junction is resistant to cleavage in the presence of ADP, AMP-PNP or PPi. Thus the conformation of pPDM-S-1 is similar to S-1-ATP. While pPDM-S-1 is very similar to a weak binding state it is not in a 100% weak binding conformation in the absence of ATP. A slight cooperativity of pPDM-S-1 binding to actin-tropomyosin-troponin is observed in the presence of ADP or in the absence of nucleotide (Greene et aL, 1986). Furthermore, pPDM-S-1 has a slight tendency to cooperatively activate ATPase activity (King and Greene, 1987). Other methods ofcrosslinking the SH-1 and SH-2 groups of myosin exist (Wells and Yount, 1982) and these may also prove to be useful as weak binding models. A semi-stable weak binding state can be produced by the addition of vanadate to the myosinADP complex (Goodno and Taylor, 1982; Goodno, 1979, 1982). The M-ADP-Vi complex is thought to be an analog of the M-D-Pi complex (see also Wells and Bagshaw, 1984; Smith and Eisenberg, 1990). The use of this complex is limited by the release of Vi upon binding to actin. More recently, aluminum and beryllium complexes with fluoride have been shown to act as phosphate analogs which inhibit phosphotransfer reactions because of their inability to adopt a pentavalent conformation (Chabre, 1990). Both beryllium fluoride and aluminum fluoride have been shown to bind to myosin to form a stable M-ADP-Pi type state which could be useful in studying the weak binding state (Maruta et al., 1991; Muhlrad et al., 1992; Phan and Reisler, 1992). Different nucleotides may stabilize myosin into a state which resembles the weak or strong binding states to some extent. As discussed earlier, the ATPy S-S-1 complex is often used as an analog of the weak binding state, S-1-ATP. Other nucleoside and nonnucleoside triphosphates may produce weak type states, which although not stable, may be useful for characterizing the properties of the weak state (see for example Pate et al., 1991). Ethylene glycol reduces the affinity of S-1 for actin and may be useful in generating other states (Marston and Tregear, 1984). Both rigor binding and binding of S-1-AMP-PNP to actin are reduced 100-fold at 50% ethylene glycol. The rate of association of actin with S-1 is reduced 6-fold and the rate of dissociation of S-1 from actin is increased 30-fold in 50% ethylene glycol (Marston, 1982). The affinity for myosin is unaffected but the rate constants for both formation and dissociation are reduced 200-fold. Actin is effective in dissociating AMP-PNP and the rate of release of AMP-PNP is sensitive to Ca 2+ and tropomyosin-troponin. Mushtaq and Greene (1989) 122 J.M. CHALOVICH later observed that 40% ethylene glycol weakens the binding of S - I - A M P - P N P to actin but has little effect of the binding of S-1-ATP. More importantly, the binding of S - I - A M P - P N P to actin, in the presence of ethylene glycol is not cooperative. However, this reduction of cooperativity in binding may be due to a direct effect on the actin-tropomyosin-troponin complex (Mushtaq and Greene, 1989). So it is not entirely clear to what extent ethylene glycol induces the weak state of myosin and to what extent it stabilizes the inactive state of the actin-tropomyosin complex. As will be shown later these are entirely different things. Interestingly, Reisler and coworkers observed that subtilisin cleavage of actin weakens the binding of S-1 to actin both in the presence and absence of ATP (Schwyter et al., 1989). This has, to date, not been fully characterized but it could be another way of stabilizing the inactive state of actin. A third method of stabilizing either the active or inactive forms of the actin filament is by crosslinking the tropomyosin to actin by glutaraldehyde (Mikawa, 1979). This treatment produces hybrid states which are 'mostly activated' or 'mostly inactivated' (J. M. Chalovich, unpublished observation) and so must be used with caution. The agent 2,3-butanedione monoxime retards contraction in cardiac muscle and is thought to produce a state similar to the weak binding state. This agent inhibits the release of Ca 2+ from the sarcoplasmic reticulum in addition to reducing the response to Ca 2+ (Gwathmey et al., 1991). Thus the use of this agent to study weak binding crossbridges is limited to systems where Ca 2+ levels can be controlled artificially. Also, the crossbridge state produced in the presence of 2,3-butanedione has not been tested rigorously to determine if it really forms a state analogous to the weak state. 4. C O M P L I C A T I O N S OF A C T I N M E D I A T E D R E G U L A T I O N To this point, regulation has been discussed in terms of models such as that shown in Fig. 4. From that type of model, one would reach the conclusion that Pi release is the regulated step. Yet, two laboratories (Homsher and Millar, 1990; Metzger and Moss, 1991) have suggested that Pi release is not greatly sensitive to Ca 2+. Therefore, more detailed models are required to accurately define the transitions which are sensitive to Ca :+ . Two important concepts, must now be considered. The first concept to be considered involves multiple conformations of myosin and the second involves multiple conformations of actin. Evidence for the existence of these conformational changes will be presented below. It is possible, but not proven, that the transition between these states is regulated. It is also possible that these changes in myosin and actin occur in a concerted fashion although the states can exist independently. 4.1. MULTIPLE STEP BINDING OF MYOSIN TO ACTIN In Fig. 4, the assumption was made that the weak state is the species of myosin containing bound ATP or ADP and Pi while the strong state is myosin containing bound ADP or nucleotide free myosin. However, recent studies have indicated the possibility that all myosin-nucleotide complexes may be able to exist in multiple conformational states and it has been further suggested that some of these states may correspond to the weakly bound and strongly bound species of myosin. That is, each myosin-nucleotide complex is thought to have the potential to bind to actin in a weak or strong binding manner; the nucleotide bound to myosin can alter the equilibrium to favor either the weak binding conformation (as in the presence of ATP) or the strong binding conformation (as with ADP). An early indication that there may be multiple myosin nucleotide complexes came from the work of Sleep and Hutton (1980). The rate of incorporation of Pi into ATP (the reverse of ATP hydrolysis) was independent of the ADP concentration and was 50-fold faster than predicted assuming that ATP was resynthesized from Pi and ADP in solution. Sleep and Hutton concluded that ATP synthesis was the result of the addition of Pi, from solution, to an S - I - A D P complex that formed during ATP hydrolysis but which was not heavily populated. AM-ADP-Pi=A-M' - ADP~AMADP~AM + ADP Addition of ADP and Pi to actomyosin, in solution, would lead to the formation of AMD which could not readily form A - M ' - D since the equilibrium favors the A M D state. However, during the Actin mediated regulation of muscle contraction 123 hydrolysis of ATP in solution, the A-M'-ADP state becomes populated and it is this state to which solution Pi binds to form ATP. Taylor has suggested that there also may be a need to postulate a two step release of Pi (Taylor, 1991) since little inhibition of ATP hydrolysis occurs in myofibrils (the contractile apparatus isolated from homogenized muscle free of other organelles and cytoplasmic proteins) at high, 100 mM, concentrations of Pi whereas the tension produced by muscle is inhibited by low, 5 mM Pi (Hibberd et al., 1985). Trybus and Taylor (1980) observed biphasic increases in light scattering as S-1 was mixed with actin, actin-tropomyosin or actin-tropomyosin-troponin indicating at least two steps in the binding in each case. They interpreted this observation as a two step association process: A + M~-AMI~-AM 2 with k~ = 1.5e6 M-~sec -1, k_l = k 2 = 2 0 sec -~. The initial binding yielded a relatively stable complex (strong binding). This is quite different from the later studies where the initial complex was weak. Such a two step reaction, if it occurred with ADP, would be consistent with the data of Sleep and Hutton. Another indication of two step binding came from pressure relaxation studies (Criddle et al., 1985). Because of an increase in volume that occurs during the binding of myosin to actin, the rapid release of pressure results in an increase in the association of S-1 with actin. The rate of the reestablishment of equilibrium can be monitored by light scattering (which increases) or by the quenching of the fluorescence of a pyrene probe placed o n cys TM of actin (Geeves and Gutfreund, 1982). Two relaxation processes were observed for the quenching of pyrene fluorescence in the absence of nucleotide but only a single relaxation was observed by light scattering (Coates et al., 1985). A two step binding scheme was proposed to account for these data; the formation of A-M gives rise to an increase in light scattering while the pressure sensitive isomerization to AM is associated with a decrease in pyrene fluorescence. A+M, kl k_l 'A-M~ k2 ,AM k_2 The complex A-M was not thought to be a collision complex since its rate of formation, kj, is apparently not diffusion controlled (Coates et al., 1985). Geeves (1989) has pointed out that k~ may be nearly fast enough to be diffusion controlled but its activation energy is too high to be a simple collision complex (Coates et al., 1985). Furthermore, for the reaction of actin with S-I, the rate kl is too slow and the temperature dependence is too great to be diffusion limited (Geeves and Gutfreund, 1982; Marston, 1982; Siemankowski et al., 1985). An analysis of the values of the individual rate constants, in the presence of different nucleotides, led Geeves and his colleagues to propose that A-M corresponded to the weak binding state while AM corresponded to the strong binding state (reviewed by Geeves, 1991a,b). The association constant for the initial complex formed, the weak state, was observed to be about 103-104 M-1. The value of the equilibrium constant for the subsequent isomerization was, in contrast, highly sensitive to the nucleotide. In the presence of ATP, ATPyS or ADP + vanidate (all weak states) the value of/(2 was ~<0.01 even at 10 mM ionic strength where the interaction between actin and S-1 is maximal (Geeves and Jeffries, 1988). The value of K2 increases to 2.3 in PPi and to about 20 for ADP (Geeves and Jeffries, 1988) and is presumably greater than 280 in rigor at low ionic strength. The value in rigor was not determined at 0.01 M ionic strength but was 280 at 0.1 M, 74 at 0.3 u and 40 at 0.5 ra ionic strength (Coates et aL, 1985; Geeves and Halsall, 1986). Since the value of K2 is very small in the presence of ATP, only the weak binding state 'AM-ATP' would be populated under normal conditions. In support of this, the addition of ATP to a mixture of actin and S-1, under conditions where only partial dissociation occurred, resulted in a large increase in fluorescence. This result is expected if only state A-M was populated (Geeves et al., 1986). A scheme incorporating this new concept is shown in Fig. 9. The transition from any weak binding state to any strong binding state is thought to occur with a structural change in actin-myosin sufficient to produce force. Because of the dependence of the value of K2 with nucleotide, the transition to a force-producing state is favored only following the release of phosphate. In relaxed striated muscle, where the weak binding state but not the strong binding state is populated, one would expect that the four states in the upper left corner of Fig. 9 are in rapid JPT 55/2~C 124 J.M.CHALOVICH ~ | ATP a i ~ p t2AT" ~ADPI , ~"PW , I ~ ob ! w ADP C i d. It 2AO,P .~P FIG. 9. Crossbridge model of ATP hydrolysis and contraction incorporating the two step binding of myosin to actin as recently proposed (Geeves and Halsall, 1987). Vertical transitions indicate the two step binding of myosin to actin, the collision complex is not shown. Step 1 produces a low affinity complex regardless of the nucleotide bound to myosin. Step 2 is an isomerization to a tight binding state which is thought to correspond to crossbridge rotation. This isomerization is favorable following Pi release but probably does not occur to a significant extent prior to Pi release. Horizontal transitions represent changes in the chemical state of myosin; only part of the scheme for total ATP hydrolysis is shown for simplicity. The states contained within the box are thought to be in rapid equilibrium with each other in relaxed muscle. Activation of muscle leads to a State 1 to State 2 transition which can produce force (see text). equilibrium with one another. Following activation, Pi must be released for force to be produced. AM-ADP-Piweak can decay by two pathways. In one case, AM-ADP-Piweak can decay, by the direct dissociation of Pi, to state AM-ADPwe~k. For AM-ADPweak to produce force it can isomerize to AM-ADP,tro~g with a forward rate constant of 4 sec--1. This pathway is too slow to explain the rate of ATP hydrolysis during unloaded shortening. The alternative is for ADP to first dissociate from AM-ADPw~ak with force production occurring between A-Mw~k to AMst~ong (not shown on Fig. 9). This is also considered unlikely since the rate of dissociation of ADP from AM-ADPwo~k is slow at 2 sec-1. The second possible pathway for the decay of AM-ADP-Piw¢~k is for isomerization to AM-ADP-Pi,trong followed by Pi release to form AM-ADPst~o,g. It is important to determine whether the flux through this pathway is sufficiently fast to account for the rate of unloaded ATPase activity. Not much information is available on the properties of the A M - A D P - P i complex. The product of the rate of product release and the equilibrium constant of the rate of Pi release, K2ADP_Pi , should be greater than 20 sec- 1 The rate of Pi release has been suggested to be on the order of 1000 sec-~ (Stein et al., 1979) and if the value of the equilibrium constant is greater than 0.01 then this could be a possible pathway. While A M - A D P - P i is thought to be a weak state, one may consider the possibility of a second A M - A D P - P i state as proposed by Eisenberg and colleagues (Stein et al., 1979). The equilibrium constant between this type of AM-ADP-Piw,ak and AM-ADP-Pistro,g could be closer to 1 so that there is a reasonable flux through this pathway. There is data to support such a pathway. For example, Fortune et al. (1991) observed that at 12 °C the transients in response to the release of caged Pi are consistent with values of k÷ + k_ of 51 sec-1. Millar and Homsher (1992) measured a value of k + of 25 sec- 1 and k_ of 86 sec- t. This makes the value of K2 for this state equal to 0.3 so that the strong binding state would be partially populated and the flux through this pathway would be large. There is disagreement over the main pathway of ATP hydrolysis during maximum shortening. Geeves argued that the M - A D P state which is loosely bound to actin cannot be heavily populated K2ADP_Pi Actin mediated regulation of muscle contraction 125 in rapidly contracting muscle while Sleep and Hutton (1980) argued that this is the most heavily populated state. The measurements of Taylor (1991) are in agreement with the data of Sleep and Hutton since his observed value for the transition to the S-I-ADP state, which binds tightly to actin (k2ADPin Fig. 9), is much faster than the value of 4 sec-1 observed by Geeves (1991b). As Taylor has suggested, it is possible that the stopped flow studies and the pressure relaxation studies may not be measuring the same step. Geeves and Conibear (1992) have presented evidence that this is actually the case. At ionic strengths below 30 mM the observed rate constant of pyrene fluorescence change, observed with a stopped-flow apparatus, increases in a hyperbolic manner with the concentration of both S-1-ADP and S-1. As the ionic strength increases, the maximum rate, at saturating concentrations of S- 1-ADP or S- 1, increases for S- 1 and decreases for S- 1-ADP. At ionic strengths > 30 mM, deviations from a hyperbola occur in both cases. While the values of kobs are similar for S-1 and S-1-ADP, they are not monitoring the same transition. It is possible that the rate constant for the transition from the collision complex to AMw~akincreases with ionic strength while the subsequent transition to AMstrongdecreases with increasing ionic strength. At low ionic strength the rate constants for the two processes are identical but at higher ionic strength, the first process is considerably faster than the second process. The reported value of k2 is also too slow to explain the rate of detachment of crossbridges in skinned fibers in the presence of ADP (Brenner, 1991). However, it is likely that the k2 increases with strain (Hill, 1974; Eisenberg et al., 1980) so there may not be a real discrepancy on this point. In a study of the binding of S- I to actin, using a stopped-flow device to monitor changes in light scattering and pyrene-actin fluorescence, Taylor (1991) observed that the value of KI increased by about 50-fold in going from ATP to rigor conditions. This result indicates that there is no unique weak state formed during the binding of S-1 to actin. This could mean that there is a family of weak states with similar but not identical properties. This raises the question as to whether all weak states have identical properties. Another interesting point about the experiments utilizing pyrenyl actin, made by Taylor (1991) is that the penultimate C-terminal residue of actin is responsible for reporting what is assumed to be a change in the interaction between S-1 and actin. However, probes placed on the SH1 group of S-1 also sense this change in interaction with a similar rate constant (Taylor, 1991; Geeves, 1991 b). Thus, the second step of the binding process may be a concerted change in both actin and myosin. Multiple states of myosin S-1 have also been detected in the absence of actin and it is largely presumed that these S-1 states are responsible for the two step binding of myosin to actin. Shriver and Sykes observed two 31p NMR resonances for the fl phosphates of both ADP and AMP-PNP (Shriver and Sykes, 1981a). The multiple resonances were found to be due to at least two S-l-nucleotide conformations. In the case of ADP at 25 °C only the upfield resonance signal was observed whereas at 0 °C an additional resonance was seen 0.7 ppm downfield. The S-1-AMP-PNP complex produced two resonances at 25° but only the downfield resonance at low temperature. Similar results were obtained using 19F NMR with S-1 containing a fluorine probe on the SH1 residue to show directly that the conformation of the S-1 itself is different in the two binding conformations (Shriver and Sykes, 1981b). Also, u.v. difference spectroscopy, which is sensitive to the 3D structure of S-I, was shown to be both nucleotide and temperature dependent (Kamath and Shriver, 1989). Trybus and Taylor (1980, 1982) and Garland and Cheung (1979) observed multistep binding of nucleotides to S-1. Thus the binding of nucleotides as well as actin to S-1 is a multistep process. Trybus and Taylor proposed the following model for this binding: M+ADP~ Fluorescence 3~p NMR state State designation Nucleotide binding strength Actin binding strength r~ 'M'ADP~ r2 'M#ADP~ low high temperature (up field) t lower higher K3 'M*ADP high low temperature (downfield) r higher lower 126 J.M. CHALOVICH In this scheme, Kj represents the formation of a collision complex which has a forward rate constant of 1000 sec- J. This is followed by the formation of fluorescent species M # ADP which occurs at a rate of k2 of about 200 sec- ~. Lastly, there is an isomerization to a more fluorescent species, M * A D P which occurs at a rate k 3 + k 3 of 15 sec - l . Shriver and Sykes (1981a) argued that at 25 °C, where only the upfield 3~p N M R resonance of ADP-S-I is observed, Trybus and Taylor (1980) observed only a single exponential for binding to actin. Therefore, the low fluorescence state (M # ADP) was thought to be the high temperature (upfield) nmr state. Shriver and Sykes (1981 a) also point out that the observation of two distinct 3~p N M R resonances for ADP at low temperature is consistent with the relatively slow exchange (15 sec-~) between the two forms. Because there is a reciprocal relationship between the binding of nucleotides and actin to myosin and because ADP is more tightly bound in the M * A D P state than in the M # ADP state, the M * A D P state was suggested to be the strong actin binding species of myosin S-1. Shriver proposed a model of contraction in which the different myosin species observed are assumed to be the weak and strong states of myosin (Shriver, 1986). Shriver also considers it likely that the weak binding of myosin to actin is somewhat nucleotide dependent; that is, there is no universal weak actin-S-I complex. The idea that the two conformations of S-l, for which there is strong evidence, corresponds to the forms of S-1 which bind weakly and strongly to actin is attractive. However, there is only circumstantial evidence, at present, that this is the case. It should be possible to observe that the weak type of interactions between S-1 and actin are more prevalent at low temperatures. Additional evidence for a temperature dependent change of myosin conformation comes from the temperature dependence of the amplitude of intrinsic protein fluorescence with the binding of ATP or ATPyS to S-1 (Millar and Geeves, 1988). Similarly, multiple conformations of S-1 are indicated from the temperature dependence of the fluorescence decay of bound eADP (Aguirre et al., 1989) and of fluorescently labeled S-1 in the presence of ADP (Lin and Cheung, 1991). In the latter experiments it was noted that fluorescence emission of 5-(iodoacetamido)fluorescein covalently attached t o cys 7°7 (SH1) of S-I decayed in a biexponential manner in both the absence of nucleotide as well as in the presence of ADP. The fluorescence decay became monoexponential in 6 M guanidine hydrochloride indicating that the two decay processes were due to different conformations of S-1. The relative intensities f~ and f2 were used to estimate the relative amount of the two conformational species with fluorescence lifetimes tl (about 4 nsec) and t~ (about 1 nsec). Just as in the studies of Shriver (1986), the ratio of the species f t/f2 is temperature sensitive with the state f~ favored at high temperature. At room temperature, the high temperature form of the S1 and S1-ADP states predominate. The biphasic decay was also observed when S-1 or S-1-ADP were complexed with actin providing evidence that the change in S-1 would occur when complexed to actin. From the temperature dependence of the transition, the change in enthalpy and entropy were determined. It is interesting that these values were similar to those obtained by the same group using unmodified S-1 and e ADP as a reporter group. In the case of S-1-e ADP, the equilibrium constant shifted, by a factor of 7, toward the low temperature state in the presence of orthovanadate. This was interpreted as evidence for the transition from the low temperature state to the high temperature state being responsible for force production (Aguirre et al., 1989). The vanadate complex of ADP is known to produce a weak binding type state while ADP forms a strong binding type of state when bound to S-1. Thus the possibility exists that there is a relationship between these states and the weak and strong states. While the presence of multiple stable conformations of myosin and the binding of myosin to actin and nucleotides occurs in multiple steps, the schemes shown in Figs 9 and 4 are not greatly different. Force is generated between weak and strong binding states and states containing bound ATP or ADP + Pi are weak binding states while states containing bound ADP or no nucleotide exist primarily as strong binding states. However, in the new scheme the transition between weak and strong does not require a change in the chemical state, that is the state of the bound nucleotide, of the S-1. These two states could correspond to the beginning and end of the power stroke. In the same regard, it was mentioned, from observations with the nucleotide dependence of binding of pPDM-modified S-1 to actin, that there may be a continuum of states between weak and strong binding. This idea is consistent with the ideas presented in the multistep binding of S-I to actin. Actin mediated regulation of muscle contraction 127 The observed intermediate states of pPDM-S-1 (Greene et al., 1986) could be due to differences in the distribution of the weak and strong forms of S-1 and not to different unique states. The critical question for the present discussion is how these concepts alter the thinking about mechanisms of regulation of contraction. In the case of tropomyosin-troponin there is inhibition on the rate of transition from a weak binding state to a strong binding state. Exactly which step is regulated depends on the pathway of entry into a force-producing state as discussed earlier in this section. It is unlikely that the rate of Pi release from AM-ADP-Piweak is directly regulated as originally predicted (Chalovich et al., 1981; Chalovich and Eisenberg, 1982). If the major pathway in active muscle is to go from AM-ADP-Piw,ak to AM-ADP-Pistro,~ to AM-ADP then either the transition from AM-ADP-Piw~k to AM-ADP-Pistrong or the subsequent release of Pi could be inhibited. Pressure relaxation studies are consistent with regulatory control of the transition from the weak binding states to the strong binding states (McKillop and Geeves, 1991). That is, steps K2ATp, K2ADPPand K2ADPin Fig. 9. Such a possibility might also prove to be in better agreement with the Pi concentration jump experiments. Alternatively, the transition between two actintropomyosin states could be regulated (Section 4.2). 4.2. ACTIN-TROPOMYOSIN CONFORMATIONALSTATES In preceding sections, evidence was presented for two types of myosin crossbridges, the weak and strong crossbridge states. One way of distinguishing between these two classes of states is the way in which they interact with actin-tropomyosin or actin-tropomyosin-troponin. Binding of the strong binding myosin subfragment species to such regulated actin filaments is cooperative whereas binding of the weak binding states is not cooperative. The cooperativity seen in the interaction of strong binding S-1 with actin-tropomyosin-troponin implies that there are also two or more conformational states of the regulated actin filament. In the following section, this theme will be explored. Weber and colleagues published a series of papers detailing the cooperative increase in the rate of ATP hydrolysis with increasing myosin subfragment concentrations (Bremel et al., 1972; Bremel and Weber, 1972; Murray et al., 1980). The rate of ATP hydrolysis by S-l, in the presence of actin-tropomyosin-troponin and absence of C a 2 +, can become faster than the rate in the absence of the inhibitory proteins under conditions of low ATP or at high ratios of S-1 to actin. This was interpreted as rigor crossbridges being able to move the tropomyosin molecule into a noninhibitory position. Extensive modification of skeletal muscle myosin S-1 with N-ethylmaleimide (Pemrick and Weber, 1976; Nagashima and Asakura, 1982; Cande, 1986; Meeusen and Cande, 1979; Dancker, 1992) results in S-I with very low rate of ATPase activity and tight binding to actin even in the presence of ATP. The addition of this modified S-1 to ATPase assays, in the presence of actin-TM-TP, caused a marked enhancement of ATPase activity. The potentiation above the rate of unregulated actin could not be readily explained by steric blocking. Greene and Eisenberg (1988) followed this work with a series of detailed studies of the binding of myosin subfragment nucleotide complexes to actin-tropomyosin-troponin. As shown in Fig. 10 they observed that the strong type S-1 complexes (i.e. in the absence of nucleotide or presence of PPi, ADP, or AMP-PNP) bound cooperatively to regulated actin. The cooperativity is most marked in the absence of C a 2 + but is observed even in the presence of C a 2+. At low levels of saturation of the actin filament with S-l, the binding is weak. However, as the degree of saturation of the actin filament increases there is a cooperative transition to a more tightly associated acto-S-1 complex. This was interpreted as two forms of the actin filament which can be referred to as the active or 'on' form of actin and inactive 'off' form of the filament. In the model of Hill et al. (1980), these two forms of regulated actin were suggested to be in equilibrium with each other: kinactivation A-Tm-Tpactive ~ ' A-Tm-Tpinactiw kactivation The equilibrium constant for a group of 7 actin monomers covered by a single tropomyosin going from all actin monomers in the 'on' state to the 'off" state is given by L' = [A-Tm-TpinactivdA-TmTpactive] (Hill et al., 1980). The value of L' is decreased in the presence of Ca 2÷ thus favoring the active form to a greater extent. Typical values of L' are 150, at very low free Ca 2÷ and 1.6 at high 128 J.M. CHALOVICH 0.8 0.8 cd" 0.4 Q~ 0.4 o~ 2.0 4.0 6.0 8.0 L I I 0.4 [NEM-S-1] s I 0.8 [S-1]F(pM) FIG. 10. Cooperative binding of myosin S-1 to actin-tropomyosin-troponin (A) and effect of strong crossbridge binding on the ATPase activity (B). In the absence of Ca 2÷ the binding of S-1-PPi, a strong binding type state, to actin-tropomyosin-troponin is clearly cooperative (A) while the cooperativity is not obvious in the presence of Ca 2+. In (B) the percent of maximal ATPase activity is shown as a function of the amount of NEM-S-1 bound to actin. This NEM-S-1 is a strong type state which binds tightly to actin even in the presence of ATP but has no ATPase activity. In the absence of added NEM-S-1, the rate of ATP hydrolysis is more than 20 times faster at high rather than at very low Ca 2÷ concentration. As the amount of NEM-S-1 bound to actin increases, the rate of ATP hydrolysis increases BOTH in the presence and absence of Ca 2+. At high degrees of saturation the rates of ATP hydrolysis in the presence and absence of Ca 2÷ are similar. These curves were made from the data of Greene and Eisenberg (1988). free Ca 2+ (Greene and Eisenberg, 1988). A similar equilibrium was also thought to exist in the absence of regulatory proteins with actin being only partially in the active state. In the absence of tropomyosin, however, there is not the same cooperative change with Ca 2÷ or S-1 binding. Assuming that the actin is not maximally active provides a natural explanation for the potentiation phenomenon. To explain the cooperative binding of S-l, it was assumed that strong crossbridges bind to the two forms of actin differently. K1 A-Tm-Tpinactiv~ + S-1 ~ ' S-1-A-Tm-Tpina~tivc K2 A-Tm-Tp,ctive + S- 1 , ' S- 1-A-Tm-Tpactiv¢ Because/(2 is greater than/(1, strong type myosin states bind preferentially to S-1-A-Tm-Tpac,ve and shift the block of 7 actin monomers more toward the active state. Note that K~ and/(2 are equivalent to Ktumednet and Kturn~don in the notation of Greene and Eisenberg (1988). The greater the ratio of K d K I , the greater will be the driving force of transition to the active state of actin as myosin binds to actin and the greater the cooperativity in binding. This ratio would be near 1 for weak binding crossbridges and will be about 7 in the presence of PPi and 20 for rigor crossbridges (Greene and Eisenberg, 1988). In addition to the cooperativity inherent in changing 7 actin monomers as a group, Hill et al. (1980) also proposed nearest neighbor interactions between adjacent units of 7 actin monomers. The energy of interaction, Y, could change depending on the free [Ca 2+ ] and on the conformational state of the interacting groups of actin monomers (i.e. active-active, active-inactive, inactiveinactive). However, binding data have been fitted well using the same value of Y (about 20) both in the absence and presence of Ca 2÷ (Greene and Eisenberg, 1988). The importance of the energy of interaction of nearest neighbor tropomyosin molecules is seen in the cooperativity of binding of tropomyosin to actin (Wegner, 1979; Wegner and Walsh, 1981; Weigt et al., 1991; T o b a c m a n et al., 1992). Wegner (1979) observed a larger cooperative effect with skeletal tropomyosin than T o b a c m a n et al. (1992) observed with cardiac tropomyosin. However, in both cases the interaction parameter is substantial. The effect of this term Y can be studied by removal of the overlap region among adjacent tropomyosin molecules (Pan et aL, 1989). Actin mediated regulation of muscle contraction 129 In the Hill formalism, actin is considered to be a dynamic structure that is influenced by tropomyosin-troponin and by agents which alter the binding of tropomyosin-troponin to actin such as myosin and Ca 2÷. The model of Hill et al. (1980) successfully predicts the cooperative binding of the strong binding species of myosin to actin-tropomyosin-troponin to actin in both the presence and absence of Ca 2+. An effect ignored by the Hill model, which is nonetheless important, is the cooperativity inherent in the thin filament in the absence of myosin binding. The binding of Ca 2÷ to cardiac actin-tropomyosin-troponin (which has only a single CaZ+-specific binding site) is cooperative even in the absence of myosin (Mehegan and Tobacman, 1991). This is a smaller effect than the cooperativity observed as a result of myosin binding to the actin filament. However, it could be quite important in regulation by Ca 2+. The equilibrium constant for an entire actin filament is given by L' x Y. It is interesting to note that the value of L ' Y is reduced from 3000 to 32 in response to Ca 2÷. However, even in Ca 2+, the value of L ' Y indicates that the actin filament is still mostly in the inactive form. This accounts for the ability of S-1 to bind cooperatively even in the presence of Ca 2÷ and to further activate the ATPase rate. In the absence of Ca 2+, tropomyosin-troponin stabilizes the 'inactive' state to a large extent causing a reduction in ATPase activity. Hill et al. (1981) have extended the model of cooperative binding to include the resulting ATPase activity. In this model, it is assumed that actin in the 'inactive' state is ineffective in accelerating Pi release or the transition from the weak state to the strong state, while actin in the 'active' state is capable of accelerating the rate of ATP hydrolysis. Although this model successfully predicts much experimental data it may have to be revised to incorporate a different regulated step since Pi release may not be the regulated transition. Lehrer and Morris (1982) applied the cooperative model of inhibition of Hill et al. (1980) to the effect of tropomyosin and tropomyosin-troponin on the inhibition of actin activated ATPase activity. In their use of the model, they choose values of L', Y, K1 and Ks to fit their observed ATPase rate data. They also concluded that there was appreciable inhibition of the ATPase activity even in the presence of Ca 2+. Cooperativity in the interaction of strong binding S-1 with actin is also evident from the kinetics of S- 1 binding to actin-tropomyosin-troponin (Trybus and Taylor, 1980). In the presence of Ca 2+ the binding of S-1 (no nucleotide) to actin, actin-tropomyosin and actin-tropomyosin-troponin were similar in rate and all were biphasic. In the absence of Ca 2+ the binding of S-1 to actin-tropomyosin was distinctly different and was dependent on the relative concentrations of S- 1 and actin. If S-1 was in excess there was a lag in binding followed by a single exponential phase of binding which was similar to the slower of the two phases in the presence of Ca 2+. This lag could be eliminated by premixing 3 tool of S-1 per mol of actin. If actin was in excess, the binding was biphasic with the individual rates having about 20% of the magnitude of those in the presence of Ca 2+. The data, at low Ca z+ concentrations, were explained by two types of actin binding sites with different affinities for S-1. The lag was due to the transition from the 'inactive' to the 'active' state of actin (or 'blocked' to 'open' in the notation of Trybus and Taylor, 1980); preloading the actin with S-1 stabilized the active state so that the lag was eliminated. Trybus and Taylor (1980) also observed a change in the fluorescence of troponin I labeled with 4-(N-iodoacetoxyethyl-N-methyl)-7-nitrobenz-2-oxa-1, 3-diazole upon binding to S-l; this change was thought to monitor the change in the state of actin filament. When ATP was rapidly added to the labeled S-l-actin-tropomyosin-troponin complex there was a very rapid decrease in light scattering caused by the rapid dissociation of S-1. The fluorescence of troponin I increased but at a much slower rate. The maximum rate of the fluorescence change (obtained at 2 mM ATP) was thought to represent the rate constant for the transition of the actin filament from the active state to the inactive state. This rate was 250 sec-~ at 4-20 °C and 30 mM KC1 and 430 sec-~ in 0.1 M KC1. The reverse transition was also thought be very fast since a maximum rate of binding of S-1 to actin-tropomyosin-troponin was not observed in the absence of Ca 2+. Thus the active and inactive states of actin were thought to be in rapid equilibrium (Trybus and Taylor, 1980). The use of a fluorescent probe to monitor the state of the actin filament is a powerful tool to study regulation. However, probes placed directly on tropomyosin provide more direct measures of this transition than probes placed on troponin I. Thus, the value of L' required to fit the data 130 J.M. CHALOVICH of Trybus and Taylor (1980) was much greater than was used in direct binding studies (Greene and Eisenberg, 1980). This could mean that either the probe was not giving an accurate measure of the degree of activation of the actin filament or that the model of Hill (1980) used to fit the data was not adequate. In a subsequent study, Greene (1986) argued that the problem was in the probes bound to troponin I. Using the same probe as Trybus and Taylor (1980) and additionally 5'-iodoacetamidofluorescein, the change in fluorescence was found to parallel the fraction of actin units in the active states in the presence of Ca 2+. However, in the absence of Ca 2+ the data could be fit by the Hill model only by assuming that two tropomyosin units were required to move into the active state in order to give a fluorescence signal change (Greene, 1986). More recently, Ishii and Lehrer showed that fluorescent probes placed directly on tropomyosin provide an accurate measure of the fraction of actin-tropomyosin-troponin units in the activated state (Ishii and Lehrer, 1987, 1990; Lehrer and Ishii, 1988). Pyrenyliodoacetamide-labeled tropomyosin appears to be a particularly convenient way to monitor the state of the actin filament particularly at high ionic strength (Ishii and Lehrer, 1990). The excimer fluorescence of the tropomyosin increased with the fraction of actin units in the activated state caused by S-1 binding to actin-tropomyosin-troponin. Little change in fluorescence was observed with Ca 2* alone suggesting that Ca 2+ facilitates a change in the tropomyosin caused by S-I binding. Interestingly, the fluorescence was independent of nucleotide which favors the first cooperative model of Hill et al. (1980). From the solution results already discussed the activation of ATPase activity by Ca 2+ alone is characterized by an increase in the Vmax of 28-fold, an increase in KATPase of 3-fold and an increase in the binding of S-1-ATP to actin of about 2-fold (Chalovich and Eisenberg, 1982). The binding of strong crossbridges (i.e. NEM-S-1) to the Ca2+-activated actin filament results in a further increase in ATPase rate of 8-fold. This activation or potentiation is characterized by an additional 2-fold increase in V~ax, a 4-fold increase in KATPaseand no change in the Kbinding (Williams et al., 1988). Therefore, at low ionic strength, in solution, total activation by Ca 2+ and strong crossbridge binding results from a 56-fold increase in Vmaxand 12-fold increase in KATPase.In the absence of Ca 2+ greater than 60% saturation of the actin filament with strong binding crossbridges is required to achieve full activation of ATP hydrolysis while in the presence of Ca 2+ only about 15% saturation is required (Greene and Eisenberg, 1988). It is interesting at this point to note that this potentiation effect is observed even in smooth muscle which lacks the troponin complex. Thus, at 20 mM ionic strength, smooth muscle tropomyosin enhances the ATPase activity 3-fold by an increase in the Vmax(Chacko and Eisenberg, 1990). Further activation of ATPase activity occurs upon the binding of NEM treated S-1. This further activation is due to a 1.3-fold increase in the Vmax and a 7-fold increase in KATPase; Kbinding also increased, unlike the case with skeletal proteins. This change in binding was thought to be due to a change from single headed binding of HMM to double headed binding. These effects of smooth muscle proteins were somewhat dependent on the ionic strength (see Table 1 of Chacko and Eisenberg, 1990). Lehrer and Morris (1984) studied the cooperative activation of ATPase activity in the presence of both smooth and skeletal tropomyosin in the absence of troponin. In both cases, the ATPase activity increased cooperatively as the degree of saturation of the thin filament increased. While smooth and skeletal tropomyosin had the same qualitative effects, there were quantitative differences in the observed cooperativity. Fitting the model of Hill et al. (1980), to the data required larger Y and smaller L' values for smooth tropomyosin. Thus, in the presence of smooth tropomyosin, more units of actin-tropomyosin are initially in the active state and the transition from the inactive to the active state is more cooperative. The steeper cooperativity was thought to result from the reported greater stiffness of the smooth tropomyosin compared to skeletal muscle tropomyosin (Betteridge et al., 1983). This may not be the case since modeling of electric birefringence data suggest that smooth tropomyosin is actually more flexible than skeletal tropomyosin (Swenson and Stellwagen, 1989). Smooth and skeletal tropomyosins were also compared by Williams et al. (1984). Whereas skeletal tropomyosin tended to inhibit actin activated S-1 ATPase activity over all ionic strengths, smooth tropomyosin inhibits the ATPase activity by 60% at low ionic strength (20 mM) but stimulates the activity by 3-fold at high ionic strength (120 mM). The fully potentiated rates (formed by adding NEM-S-1 to the system) were the same for both smooth and skeletal tropomyosins. Actin mediated regulation of muscle contraction 131 Williams et al. (1984) explained these observations by postulating that, in the absence of bound S-I, smooth muscle tropomyosin induces a larger fraction of activated actin units than skeletal muscle tropomyosin. This result was confirmed by equilibrium binding studies of the S-1-P-PNP complex. Several variations of the Hill model have been produced. One model allows for a continuum of tropomyosin positions on the actin filament (Hill et al., 1983). Under any set of conditions all of the tropomyosin molecules will exist at a particular state as opposed to there being a change in the distribution between two states in the earlier model. This latter model is not consistent with the observation that pPDM-S-1 can turn on the actin filament in the presence of Ca 2+ but not in the absence of Ca 2÷ (Greene et al., 1987). This pPDM experiment can be readily explained by the original Hill model assuming a ratio of K J K I of 2 for the pPDM-S-1. In another variation of this model, the transition from the inactive to the active state of regulated actin is assumed to be coupled with the transition from the weak binding state to the strong binding state of S-1 (Geeves and Halsall, 1987). This variation gives the same results as the original model of Hill et al. (1980). Other models of the cooperative binding of S-1 to regulated actin have also been proposed (Balazs and Epstein, 1983) but will not be discussed here. An important question in regulation is what the contribution of the potentiation of ATPase activity by the binding of 'strong binding' crossbridges is in muscle contraction. Preliminary evidence from three laboratories suggests that diffusion of S-1 modified with N-ethylmaleimide (a strong state of S-l) into single skinned striated muscle fibers activates the fibers even in relaxing conditions (Schnekenbuhl et al., 1991, 1992; Swartz and Moss, 1991; DasGupta and Reisler, 1991). The results appear to be similar to those in solution in that the maximum activity obtained by binding of strong binding crossbridges is similar in both the presence and absence of Ca 2÷. However, the mechanisms of activation by Ca 2÷ and strong crossbridge binding may be different. The diffusion of NEM-S-1 into a muscle is an artificial situation. In a living muscle where do these additional strong binding crossbridges come from which activate contraction? If tropomyosin-troponin controls the transition from 'weak' to 'strong' myosin crossbridges then activation by Ca 2÷ will cause an increase in the number of attached strong binding crossbridges. The increased number of strong or force producing crossbridges may further activate the muscle by the mechanisms described above in solution. The parameters of the Hill model could be quite different in a muscle fiber from what has been measured in solution. In particular the cooperativity is reported to be much greater in intact fibers than in solution (Brandt et al., 1987, 1990; Moss et al., 1985, 1986). This observation is primarily based on the change in Ca 2÷ sensitivity of the muscle upon partial extraction of troponin C from the muscle so as to reduce the number of tropomyosin units which could act as a cooperative unit. DasGupta and Reisler (1991) made an interesting observation that the binding of S-1-AMPPNP, S-1-ADP or S-1-PPi to actin is cooperative in the absence of tropomyosin but in the presence of antibodies directed against the 1-7 region of actin. This cooperativity occurred without the displacement of antibody from the actin. Thus it may be possible to stabilize the inactive and active forms of actin even in the absence of tropomyosin. They later found that these antibodies inhibited ATPase activity not only by displacing S-1-ATP from actin but also by inducing an inhibitory conformational change in actin which did not activate ATP hydrolysis (DasGupta and Reisler, 1992). Interestingly, Sutoh et al. (1991) observed that mutations in the N-terminal region of Dictyostelium actin has diminished Vmaxvalues without changes in KATpase-It has been shown that antibodies directed against tropomyosin can inhibit motility (Hegmann et al., 1989). One can imagine this occuring if the antibodies stabilized the inactive form of the actin filament much as troponin does in the absence of Ca 2÷. Both of these observations could lead to interesting probes for studying the regulatory apparatus. 5. SUMMARY: REGULATION BY TROPOMYOSIN-TROPONIN The tropomyosin-troponin regulatory system grows more complex by the year. Ca 2+ binding to troponin C causes a conformational change in troponin C that is not entirely characterized. This change makes the troponin subunits bind more tightly with each other and less tightly to actin and 132 J.M. CHALOVICH tropomyosin. The result of these changes is an alteration in the binding of tropomyosin to actin. This change in tropomyosin binding may have some effect on the affinity of myosin for actin. However this change in tropomyosin has other, more pronounced effects. When tropomyosin-actin units are in the active configuration the kinetics of ATP hydrolysis are greatly increased at a given level of binding of myosin to actin. Thus tropomyosin-troponin was postulated to control the rate of transition from weak binding myosin states to strong binding myosin states and this transition was thought to be coupled to the Pi release step. Subsequent studies showed that the transition from a weak to a strong conformation is inhibited but that this need not be directly coupled to Pi release; thus the weak to strong transition is thought to exist, in principle, for all actin-myosin interactions. The concept that tropomyosin-troponin controls a transition between two bound actomyosin complexes holds true. The change in binding of tropomyosin to actin was one of the early observations made and it remains of central importance. Thus, regulation of ATPase activity can occur independently of troponin. The binding of 'strong' binding myosin crossbridges can induce a change in the actin-tropomyosin complex which leads to greater activity. In the presence of tropomyosin and troponin, both Ca 2÷ and binding of strong myosin crossbridges tend to shift the actin distribution cooperatively toward the 'active' configuration. There is evidence that the change from 'weak' to 'strong' myosin states occurs in a concerted manner with the change from 'inactive' (or 'turned off' or 'closed') actin units to 'active' (or 'turned on' or 'open') actin units. It is not clear at present if activation of the actin filament by Ca 2÷ and strong crossbridge binding stabilize the same conformational state of the actin filament and produce the same type of activation. It is also not clear what the function of these different modes of activation are in striated or cardiac muscle. 6. SMOOTH MUSCLE ACTIN BINDING PROTEINS The primary switch for contraction of smooth muscle is phosphorylation of the regulatory light chains of smooth muscle myosin (Itoh et al., 1989). However, there are indications that additional regulatory systems might also be active in this muscle (Suematsu et al., 1991; Tansey et al., 1990; Gerthoffer, 1987; Fischer and Pfitzer, 1989). One reason for suspecting regulation by actin-binding proteins is the presence of the proteins tropomyosin, caldesmon (Sobue et al., 1981) and calponin (Takahashi et al., 1988) in smooth muscle and nonmuscle cells where phosphorylation of myosin is thought to be the key regulatory event. Both caldesmon (Marston and Lehman, 1985; Furst et al., 1986) and calponin (Gimona et al., 1990; Nishida et al., 1990) are localized in the actomyosin domain. Smooth muscle actin filaments have been shown to stimulate the ATPase activity of skeletal muscle myosin in a Ca2+-dependent manner (Marston and Smith, 1984; Marston and Lehman, 1985) although the skeletal myosin is not regulated by phosphorylation. The properties of caldesmon have been discussed in several reviews (Marston and Redwood, 1991; Sobue and Sellers, 1991; Chalovich et aL, 1990; Chalovich, 1988) but relatively little is known about calponin at this time. The discussion to follow is focused on issues related to those discussed earlier with the tropomyosin-troponin system of regulation particularly on how these proteins affect the weak and strong interactions between actin and myosin. 6.1. CALDESMON Several studies suggest that caldesmon plays a role in actin and myosin mediated motility. Caldesmon is phosphorylated and dissociates from actin microfilaments in transformed rat embryo-derived fibroblasts during mitosis (Yamashiro et aL, 1990; Yamashiro and Matsumura, 1991). Large structural changes occur in these cells during mitosis and caldesmon is thought to play a part in these changes. Receptor capping in mouse T-lymphoma cells was inhibited following the extracting of caldesmon with 25 mM MgC12 (Walker et al., 1989). Capping was restored upon the addition of purified smooth muscle caldesmon. Microinjection of an antibody which inhibits caldesmon binding to actin, inhibits granule movement in cultured chicken embryo fibroblasts (Hegmann et aL, 1991). High concentrations of the antibody caused the caldesmon to dissociate from stress fibers in the cells (Lin et al., 1991). Caldesmon and fragments of caldesmon inhibit force Actin mediated regulation of muscle contraction 133 production when allowed to diffuse into single skeletal muscle fibers (Brenner et al., 1991) or smooth muscle fiber bundles (Pfitzer et al., 1992). The inhibition of force was reversible, in both cases and occurred without dephosphorylation of myosin light chains in the smooth muscle. Antibodies against caldesmon have been used to reverse the Ca 2÷-regulation of skeletal myosin by vascular smooth muscle thin filaments (Marston et al., 1988). In short, removal of caldesmon impedes some motile events while addition of caldesmon to other systems inhibits motility. Because of the unique structure of caldesmon, both stimulation and inhibition are possible. Caldesmon binds to actin with an affinity near 107 M-1 (Velaz et al., 1989) and to myosin with an affinity near 106 M-t (Hemric and Chalovich, 1990). The myosin-binding activity has been localized to an N-terminal chymotryptic fragment of caldesmon (Velaz et al., 1990) while the actin binding region(s) are largely confined to a C-terminal 35 kDa fragment (Szpacenko and Dabrowska, 1986; Fujii et al., 1987). The center region of caldesmon is an extended helix of no known function (Wang et al., 1991) and is missing in nonmuscle caldesmon's (Ball and Kovala, 1988). Studies of the mechanism by which caldesmon inhibits actin activation of myosin ATPase activity have been complicated by the simultaneous binding of caldesmon to actin and myosin. However, inhibition of ATP hydrolysis by the C-terminal region of caldesmon, which does not bind to myosin, is strongly correlated with an inhibition of binding of myosin-subfragrnent-ATP complexes to actin (Velaz et al., 1989; Horiuchi et al., 1991) supporting the hypothesis that caldesmon is a competitive inhibitor of the binding of myosin to actin (Chalovich et al., 1987). Caldesmon also competes with the binding of other myosin-nucleotide complexes with actin (Chalovich et al., 1987; Hemric and Chalovich, 1988) but due to the relative binding constants (Velaz et al., 1989), caldesmon is a good inhibitor only of the weak crossbridges. Furthermore, the diffusion of caldesmon or the actin-binding fragments of caldesmon into single skeletal muscle fibers (which normally lack caldesmon) inhibits the binding of myosin to actin in the presence of ATP whereas no inhibition is seen in the presence of PPi, ADP, or in rigor (Brenner et al., 1991). Further evidence that caldesmon functions as a competitive inhibitor of the binding of myosin to actin is that caldesmon and myosin share some of the same binding regions on actin as seen by N M R (Levine et al., 1990), crosslinking studies (Bartegi et al., 1990) or by competition with antibodies directed against discrete regions of actin (Adams et al., 1990). Furthermore, caldesmon does not inhibit the ATPase activity of myosin S-1 chemically crosslinked to actin (Bartegi et al., 1990). In skeletal muscle fibers caldesmon clearly inhibits crossbridge binding in ATP as seen by stiffness and X-ray measurements but there is no evidence for a reduction in the rate of force redevelopment (Brenner et al., 1991; Chalovich et al., 1991a,b). Caldesmon was also found to weaken the binding of fluorescently labeled myosin S-1 to actin filaments in ghost fibers (Nowak et al., 1989). If caldesmon functions only by inhibiting the binding of myosin to actin then caldesmon should cause a decrease in the KAxPase without a change in the Vmax. Horiuchi et al. (1991) observed a 6-fold decrease in the KATPaso of smooth muscle H M M (association constant) while the Vmax was unchanged, in the absence of tropomyosin. However, in the presence of tropomyosin a 2-fold reduction in both the Vmaxand KATPascwas observed. Marston (1988), however, reported a decrease in the Vma× with no change in the KATPase with thioFhosphorylated smooth muscle HMM and suggested that caldesmon inhibited the rate of product release. Freedman et al. (1992) attempted to obtain an accurate estimation of the steady-state kinetic parameters by working under conditions where the binding of myosin to actin was maximized. The A-1 isozyme of skeletal S-1, which binds most tightly to actin, was used at very low ionic strength to maximize the binding of S-1 to actin. To avoid aggregation of caldesmon at low ionic strength, two SH groups of caldesmon were modified with iodoacetamide. Under these conditions they observed a decrease in the KATPaseto less than 8% of the original value and almost a 2-fold decrease in the Vma~both in the presence and absence of tropomyosin. The inhibition of ATPase for a given amount of caldesmon bound to actin is greater in the presence of smooth muscle tropomyosin (Velaz et al., 1989; Smith et al., 1987). Thus, in the presence of smooth tropomyosin, the ATPase activity decreases to a greater extent than does the binding of smooth H M M to actin (Horiuchi and Chacko, 1989). These data also suggest that, in the presence of tropomyosin, caldesmon might also inhibit the rate of a kinetic transition. The activation of ATPase activity by smooth tropomyosin alone is primarily due to an increase in the 134 J.M. CHALOVlCrt Vm~xbut the further increase with rigor S-1 binding is primarily due to an increase in binding of HMM to actin (Chacko and Eisenberg, 1990). On the other hand, the reversal of the potentiation by tropomyosin, by the actin binding fragment of caldesmon, both in the absence and presence of NEM-S-1, is associated with a large decrease in binding of HMM to actin (see Fig. 4; Horiuchi and Chacko, 1989). Thus while caldesmon clearly displaces weak binding myosin crossbridges from actin there is also the possibility of a reduction in some kinetic transition. Another unknown is the function of myosin binding by caldesmon. After the description of this myosin binding it was suggested that caldesmon might crosslink actin and myosin together to passively maintain force in smooth muscle. However, several recent experiments argue against this idea. The addition of the myosin-binding fragment of caldesmon to gizzard fibers in the latch state, where high tension is maintained with low ATPase activity, does not cause a reduction of force (Pfitzer et al., 1992). Furthermore, the addition of intact caldesmon has never been observed to cause an increase in force as might occur with such crosslinking (Pfitzer et al., 1992). Caldesmon has been shown in the in vitro actin motility assay to actually increase the movement of actin filaments under conditions at which actin binds weakly to myosin (Haeberle et al., 1991). This occurs only when the caldesmon concentration is very low so that only a small number of myosin binding sites on actin are blocked by caldesmon. The fact that actin motility can occur when it is tethered to myosin by caldesmon implies that the caldesmon is able to attach and detach with myosin or actin very rapidly on the time scale of motion. Such a tethering could not maintain force. If the binding of caldesmon to both myosin and actin are involved in the regulation of contraction then it is reasonable to expect that these interactions, in turn, are regulated. The binding of caldesmon to both actin (Sobue et al., 1982) and to myosin (Ikebe and Reardon, 1988; Hemric and Chalovich, 1988; Hemric et al., 1991) are reversed by Ca2÷-calmodulin. However, rather large concentrations of calmodulin are required for this reversal and it is uncertain what role this plays in vivo. Phosphorylation of caldesmon is also reported to inhibit the interaction of caldesmon with both actin (Ngai and Walsh, 1984; Yamashiro et al., 1991) and myosin (Sutherland and Walsh, 1989; Hemric et al., 1991). Phosphorylation of caldesmon does occur in smooth muscle (Adam et al., 1989; Barany et al., 1991). Caldesmon can be phosphorylated by several different protein kinases. Protein kinase C and cdc kinase phosphorylate sites located in the C-terminal region (Sutherland and Walsh, 1989; Ikebe and Reardon, 1990; Mak et al., 1991; Tanaka et al., 1990; Adam et al., 1990, 1992; Ikebe and Hornick, 1991) whereas protein kinase II phosphorylates sites on both the C- and N-terminal regions (Ikebe and Reardon, 1990). The order of phosphorylation by Ca2÷-calmodulin protein kinase II is: ser 73, ser 26, ser 726 and ser 587; the N-terminus is preferentially phosphorylated by the calmodulin dependent protein kinase II. Ca 2+-calmodulin dependent protein kinase II is thought to be the kinase which copurifies with caldesmon (Ikebe et al., 1990; Scott-Woo et al., 1990). Casein kinase II phosphorylates one site at the N-terminal region of caldesmon (Vorotnikov et al., 1988). It is interesting that kinases exist which phosphorylate both the N- and C-terminal regions of caldesmon which result in the inhibition of binding to both myosin and to actin. Thus, both interactions may be regulated. However, the kinase(s) which functions in intact muscle has not been identified. A recent report suggests that one kinase which has regulatory importance in muscle is a proline directed kinase such as cdc kinase or one of the microtubule associated protein kinases (Adam et al., 1992). 6.2. CALPONIN Much work is currently being done on calponin but relatively little is published at the present time. Calponin does inhibit the actin activated ATPase activity of myosin and its subfragments but there is not yet consensus on the mechanism of inhibition or on the reversal of inhibition. Abe et al. (1990) reported 25% and 40% inhibition of the actin activated ATPase activity, of phosphorylated myosin, in the absence and presence of tropomyosin, respectively. Maximum inhibition occurred with 1 calponin bound per 7-10 actin monomers. The normalized inhibition of ATPase activity by caldesmon was unchanged by the presence of calponin; that is while the ATPase activity was initially depressed by 25% by calponin, the final level of ATPase activity was Actin mediated regulation of muscle contraction 135 25% lower for caldesmon-calponin than for caldesmon alone. In another report, 78% inhibition of ATPase activity was obtained in the presence of calponin and smooth muscle tropomyosin and slightly less inhibition in the absence of tropomyosin (Winder and Walsh, 1990). Calponin also inhibits the ATPase activity of phosphorylated smooth muscle HMM (Horiuchi and Chacko, 1991). Calponin inhibits actin activated ATPase activity of phosphorylated smooth actin-HMM ATPase by 90% in the absence of tropomyosin and 80% in the presence of tropomyosin. The fraction of HMM bound to actin is decreased by about 70% by calponin in the presence or absence of tropomyosin. The Vmax decreases from 3 to 0.53 in the absence of tropomyosin and from 6.3 to 2.4 with tropomyosin present. Calponin caused a decrease in KATPase values from 1.6 × 104 to 0.96 × 104 M-~ in the absence of tropomyosin and from 1.9 x 104 to 1 × 104M-1 in the presence of tropomyosin. The reversal of the inhibitory effect of calponin is, at present, unclear. Makuch et al. (1991) have reported that Ca 2+-calmodulin reverses the inhibition of ATPase activity caused by calponin but no such effect was seen by two other laboratories (Winder and Walsh, 1990; Marston, 1991). Calponin can be phosphorylated in vitro by protein kinase C with a resulting reduction in the affinity of calponin for actin (Winder and Walsh, 1990). However, two laboratories have failed to observe calponin phosphorylation in contracting or resting arterial smooth muscle (Barany et al., 1991; Gimona et al., 1992). 7. CONCLUSION Although the tropomyosin-troponin complex was, for many years, the only known actin-based regulatory system of contraction we now have the possibility of additional systems involving tropomyosin, caldesmon and calponin. The current data indicate quite different mechanisms of inhibition by tropomyosin-troponin and tropomyosin-caldesmon. This is an exciting prospect since different mechanisms may allow independent and specific pharmacological intervention of muscle contraction. The differences in the mechanisms of these systems also provides a powerful tool for studying motility by artificially adding a second regulatory system. However, it must be emphasized that physiological function of caldesmon and calponin, unlike troponin, has not been well established. It is not clear if these proteins act independently or if caldesmon, calponin and perhaps other proteins, function in a concerted way such as the components of troponin in skeletal muscle. In addition, the means by which the inhibition by caldesmon and calponin are reversed is not known with certainty. The further characterization of actin-based regulation of smooth muscle contraction and nonmuscle cell motility is eagerly awaited. Acknowledgements--The author wishes to thank Drs Herbert C. Cheung, George N. Phillips, Jr, James D. Potter, Emil Reisler, Peter A. Rubenstein, Mark Schoenberg, Kazuo Sutoh and Leepo C. Yu for sharing their illustrations or providing information prior to publication and Bernhard Brenner, Michael A. Geeves and George Phillips for commenting on portions of this manuscript. He also thanks Drs Michael B~rfinyand Evan Eisenberg who helped to shape his thinking on muscle contraction and whose contributions to the field will be long remembered. REFERENCES ABE, M., TAKAHASHI,K. and HIWADA,K. (1990) Effect of calponin on actin-activated myosin ATPase activity. J. Biochem. (Tokyo) 108: 835-838. ADAM, L. P., HAEBERLE,J. R. and HATHAWAY,D. R. (1989) Phosphorylation of caldesmon in arterial smooth muscle. J. biol. Chem. 264: 7698-7703. ADAM, L. P., MILIO, L., BRENGLE, B. and HATHAWAY,D. R. (1990) Myosin light chain and caldesmon phosphorylation in arterial muscle stimulated with endothelin-1. J. Molec. cell. Cardiol. 22: 1017-1023. ADAM, L. P., GAPINSKI,C. J. and HATHAWAY,D. R. 0992) Phosphorylation sequences in h-caldesmon from phorbol ester-stimulated canine aortas. FEBS Lett. 302: 223-226. ADAMS,S., DAsGUPTA,G., CHALOVICH,J. M. and REISLER,E. (1990) Immunochemicalevidencefor the binding of caldesmon to the NH2-terminal segment of actin. J. biol. Chem. 265: 19652-19657. AGUIRRE,R., LIN, S.-H., GONSOULIN,F., WANG, C.-K. and CHEUNG,H. C. (1989) Characterization of the ethenoadenosine diphosphate binding site of myosin subfragment 1. Energetics of the equilibrium between two states of nucleotide-S-1 and vanadate induced global conformational changes. Biochemistry 28: 799-809. 136 J.M. CHALOVICH AJTAI, K., FRENCH,A. R. and BURGHARDT,T. P. (1989) Myosin crossbridge orientation in rigor and in the presence of nucleotide studied by electron spin resonance. Biophys. J. 56: 535-541. ALBANESI,J. P., HAMMER,J. A., III and KORN,E. n. (1983) The interaction of F-actin with phosphorylated and unphosphorylated myosins IA and IB from acanthamoeba castellanii. J. biol. Chem. 258: 10176-10181. AMOS,L. A., HUXLEY,H. E., HOLMES,K. C., GOODY,R. S. and TAYLOR,K. A. (1982) Structural evidence that myosin heads may interact with 2 sites on F-actin. Nature 299: 467-469. ANDERSON,M. L. and SCHOENBERG,i , (1987) Possible cooperativity in crossbridge detachment in muscle fibers having magnesium pyrophosphate at the active site. Biophys. J. 52: 1077-1082. APPLEGATE,D. and FUCKER,P. (1987) New states of actomyosin. J. biol. Chem. 262: 6856-6863. ARATA,T. (1986) Structure of the actin-myosin complex produced by crosslinking in the presence of ATP. J. Molec. Biol. 191: 107-116. ASPENSTROM,P. and KARLSSON,R. (1991) Interference with myosin S-1 by site-directed mutagenesis of actin. Eur. J. Biochem. 200: 35-41. BABU,A., SACK,J. S., GREENHOUGH,T. J., BUGY,C. E., MEANS,A. R. and COOK,W. J. (1985) Three-dimensional structure of calmodulin. Nature 315: 37-40. BAGNI,M. A., CECCI,G., COLOMO,F. and GARZELLA,P. (1991) Passive force response of isolated muscle fiber to fast stretches. J. Musc. Res. Cell Motility 13: 222. BAGSHAW,C. R., ECCLESTON,J. F., TRENTHAM,D. R., MATES,D. W. and GOODY,R. S. (1972) Transient kinetic studies of the Mg2+-dependent ATPase of myosin and its proteolytic subfragments. CoM Spring Harbor Syrup. Quant. Biol. 37: 127-135. BAGSHAW,C. R., ECCLESTON,J. F., ECKSTEIN,F., GOODY,R. S., GUTFREUND,H. and TRENTHAM,D. R. (1974) The magnesium ion-dependant adenosine triphosphatase of myosin. Biochem. J. 141: 351-364. BAILEY,K. (1948) Tropomyosin: A new asymmetric protein component of the muscle fibril. Biochem. J. 43: 271-279. BALL,E. H. and KOVALA,T. (1988) Mapping of caldesmon: Relationship between the high and low molecular weight forms. Biochemistry 27: 6093-6098. BALAZS,A. C. and EPSTEIN,I. R. (1983) Kinetic model for the interaction of myosin subfragment 1. Biophys. J. 44: 144-151. B~RANY, M., ROKOLYA,A. and B~RANY,K. (1991) Absence of calponin phosphorylation in contracting or resting arterial smooth muscle. FEBS Lett. 279: 65-68. BARTEGI,A., FATTOUM,A. and KASSAB,R. (1990) Cross-linking of smooth muscle caldesmon to the NH2terminal region of skeletal F-actin. J. biol. Chem. 265: 2231-2237. BELKNAP, B., WANG, X.-O. and WHITE, H. D. (1992) The bond splitting step is rate limiting for skeletal actomyosin-S1 ATP hydrolysis. Biophys. J. 61: A440. BURGER,C. L., SVENSSON,E. C. and THOMAS,D. D. (1989) Photolysis of a photolabile precursor of ATP (caged ATP) induces microsecond rotational motions of myosin heads bound to actin. Proc. hath. A cad. Sci. U.S.A. 86: 8753-8757. BERTRAND,R., CHAUSSEPIED,P., AUDEMARD,E. and KASSAB,R. (1989) Functional characterization of skeletal F-actin labeled on the NH2-terminal segment of residues 1-28. Fur. J. Biochem. 181: 747-754. BETTACHE,N., BERTRAND,R. and KASSAB,R. (1990) Maleimidobenzoyl-G-actin: Structural properties and interaction with skeletal myosin subfragment-1. Biochemistry 29: 9085-9091. BETTERIDGE,D., GRACEFFA,P., LEHRER,S. S. and SEIDEL,J. C. (1983) Evidence for the lack of localized melting and chain separation at cysteine in tropomyosin. Biophys. J. 41: 104a. BHANDARI,n. (1985) Resonance energy transfer evidence for two attached states of the actomyosin complex. FEBS Lett. 187: 160-166. BIOSCA,J. A., TRAVERS,F., BARMAN,T. E., BERTRAND,R., AUDEMARD,E. and KASSAB,R. (1985) Transient kinetics of adenosine 5'-triphosphate hydrolysis by covalently cross-linked actomyosin complex in water and 40% ethylene glycol by the rapid flow quench method. Biochemistry 24: 3814--3820. BOREJDO,J., PUTNAM,S. and MORALES,M. F. (1979) Fluctuations in polarized flourescence: Evidence that muscle crossbridges rotate repetitively during contraction. Proc. narD. Acad. Sci. U.S.A. 76: 6346-6350. BRANDT,P. W., DIAMOND,M. S., RUTCHIK,J. S. and SCHACHAT,F. H. (1987) Cooperative interactions between troponin-tropomyosin units extend the length of the thin filament in skeletal muscle. J. Molec. Biol. 195: 885-896. BRANDT, P. W., ROEMER,D. and SCHACHAT,F. H. (1990) Cooperative activation of skeletal muscle thin filaments by rigor crossbridges. The effect of troponin C extraction. J. Molec. Biol. 212: 473-480. BREMEL,R. D. and WEBER,A. (1972) Cooperation within actin filament in vertebrate skeletal muscle. Nature 238: 97-101. BREMEL,R. D., MURRAY,J. M. and WEBER,A. (1972) Manifestations of cooperative behavior in the regulated actin filament during actin-activated ATP hydrolysis in the presence of calcium. Cold Spring Harbor Syrup. Quant. Biol. 37: 267-275. BRENNER,B. (1985) Correlation between the crossbridge cycle in muscle and the actomyosin ATPase cycle in solution. J. Musc. Res. Cell Motil. 6: 659-664. BRENNER,B, (1988) Effect of Ca 2+ on cross-bridge turnover kinetics in skinned single rabbit psoas fibers: implications for regulation of muscle contraction. Proc. hath. Acad. Sci. U.S.A. 85: 3265-3269. BRENNER, B. (1990) Muscle mechanics and biochemical kinetics. In: Molecular Mechanisms in Muscular Contraction, pp. 77-149, SQUIRE,J. M. (ed.) Macmillan Press, London. Actin mediated regulation of muscle contraction 137 BRENNER, B. (1991) Dynamic actin interaction of crossbridges: A general principle and its implications for crossbridge action in muscle. Adv. Biophys. 27: 259-269. BRENNER,B. and EISENBERG,E. (1986) Rate of force generation in muscle: correlation with actomyosin ATPase in solution. Proc. natn. Acad. Sci. U.S.A. 83: 3542-3546. BRENNER, B., SCHOENBERG,M., CHALOVlCH,J. M., GREENE, L. E. and EtSENBERG, E. (1982) Evidence for crossbridge attachment in relaxed muscle at low ionic strength. Proc. HatH. Acad. Sci. U.S.A. 79:7288-7291. BRENNER,B., Yu, L. C. and PODOLSKY,R. J. (1984) X-ray evidence for crossbridge formation in relaxed muscle fibers at various ionic strengths. Biophys. J. 46: 299-306. BRENNER,B., CHALOVICH,J. M., GREENE,L. E., EISENBERG,E. and SCHOENBERG,M. (1986a) Stiffness of skinned rabbit psoas fibers in MgATP and MgPPi solution. Biophys. J. 50: 685-691. BRENNER,B., Yu, L. C., GREENE,L. E., EISENaERG,E. and SCHOENBERG,M. (1986b) Ca 2+-sensitive crossbridge dissociation in the presence of magnesium pyrophosphate in skinned rabbit psoas fibers. Biophys. J. 50: ll01-1108. BRENNER,B., Yu, L. C. and CHALOVICH,J. M. (1991) Parallel inhibition of active force and relaxed fiber stiffness in skeletal muscle by caldesmon: Implications for the pathway to force generation. Proc. natn. Acad. Sci. U.S.A. 88: 5739-5743. BURGHARDT, T. P., ANDO, T. and BOREJDO, J. (1983) Evidence for crossbridge order in contraction of glycerinated skeletal muscle. Proc. natn. Acad. Sci. U.S.A. 80: 7515-7519. BURKE, M., REISLER,E. and HARRINGTON,W. F. (1976) Effect of bridging the two essential thiols of myosin on its spectral and actin-binding properties. Biochemistry 15: 1923-1927. CANDE,Z. W. (1986) Preparation of N-ethylmaleimide-modified heavy meromyosin and its use as a functional probe of actomyosin-based motility. Meth. Enzymol. 134: 473-477. CARLIER,M.-F. (1991) Actin: Protein structure and filament dynamics. J. biol. Chem. 266: I-4. CASPAR,D. L. D., COHEN,C. and LONGLEY,W. (1969) Tropomyosin." Crystal structure, polymorphism and molecular interactions. J. Molec. Biol. 41: 87-107. CHABRE,M. (1990) Aluminofluoride and beryllofluoride complexes: new phosphate analogs in enzymoiogy. TIBS 15: 6-10. CHACKO,S. and EISENBERG,E. (1990) Cooperativity of actin-activated ATPase of gizzard heavy meromyosin in the presence of gizzard tropomyosin. J. biol. Chem. 265:2105-2110. CHALOVICH,J. M. (1988) Caldesmon and thin-filament regulation of muscle contraction. Cell Biophys. 12: 73-85. CHALOVICH, J, M. and EISENBERG, E. (1982) Inhibition of actomyosin ATPase activity by troponin-tropomyosin without blocking the binding of myosin to actin. J. biol. Chem. 257: 2432-2437. CHALOVlCH,J. M. and EISENBERG,E. (1986) The effect of troponin-tropomyosin on the binding of heavy meromyosin to actin in the presence of ATP. J. biol. Chem. 261: 5088-5093. CHALOVICH,J. M., CHOCK, P. B. and EISENBERG,E. (1981) Mechanism of action of troponin-tropomyosin: inhibition of actomyosin ATPase activity without inhibition of myosin binding to actin. J. biol. Chem. 256: 575-578. CHALOVICH,J. M., GREENE,L. E. and EISENBERG,E. (1983) Crosslinked myosin sub-fragment l: a stable analog of the subfragment-l-ATP complex. Proc. natn. Acad. Sci. U.S.A. 80: 4909-4913. CHALOVICHJ. M., CHANTLER,P. D., SZENT-GYORGYI,A. G. and EISENBERG,E. (1984a) Regulation of molluscan actomyosin ATPase activity. J. biol. Chem. 259: 2617-2621. CHALOVICH J. M., STEIN,L. A., GREENE,L. E. and EISENBERG,E. (1984b) Interaction of isozymes of myosin subfragment 1 with actin: effect of ionic strength and nucleotide. Biochemistry 23: 4885-4889. CHALOVICH J. M., CORNELIUS,P. and BENSON,C. E. (1987) Caldesmon inhibits skeletal actomyosin subfragment-1 ATPase activity and the binding of myosin subfragment-1 to actin. J. biol. Chem. 262: 5711-5716. CHALOVICH J. M., HEMRIC,M. E. and VELAZ,L. (1990) Regulation of ATP hydrolysis by caldesmon: A novel change in the interaction of myosin with actin. Ann. N Y Acad. Sci. 599: 85-99. CHALOVlCH,J. M., YU, L. C. and BRENNER,B. (1991a) Involvement of weak binding crossbridges in force production in muscle. J. Musc. Res. Cell Motil. 12: 503-506. CHALOVICHJ. M., Yu, L. C., VELAZ,L., KRAFT,TH. and BRENNER,B. (1991b) Caldesmon derived polypeptides as probes of force production in skeletal muscle. In: Peptides as Probes in Muscle Research, pp. 81-93, RUEGG, J. C. (ed.) Springer-Verlag, New York. CHOCK, S. P. and EISENBERG,E. (1979) The mechanism of the skeletal muscle myosin ATPase I. Identity of the myosin active sites. J. biol. Chem. 254: 3229-3235. CHONG, P. C. S., ASSELBERGS,P. J. and HODGES,R. S. 0983) Inhibition of rabbit skeletal muscle acto-S-1 ATPase by troponin. FEBS Lett. 153: 372-376. COATES,J. H., CRIDDLE,A. H. and GEEVES,M. A. (1985) Pressure-relaxation studies of pyrene-labeled actin and myosin subfragment 1 from rabbit skeletal muscle. Biochem. J. 232: 351-356. COHEN, C. and SZENT-GYORGYI,A. G. (1957) Optical rotation and helical polypeptide chain configuration in alpha proteins. J. Am. Chem. Soc. 79: 248. COLLINS,J. H., KUZNICKI,J., BOWERS,B. and KORN,E. D. (1982) Comparison of the actin binding and filament formation properties of phosphorylated and dephosphorylated acanthamoeba myosin II. Biochemistry 21: 6910-6915. COOK, R. K., BLACKE,W. T. and RUaENSTEIN,P. A. (1992) Removal of the amino-terminal acidic residues of yeast actin. Studies in vitro and in vivo. J. biol. Chem. 267: 9430-9436. 138 J.M. CHALOVICH COOK, R. K., ROOT,D., MILLER,C., RESILER,E. and RUBENSTEIN,P. A. (1993) Enhanced stimulation of myosin subfragment 1 ATPase activity by addition of negatively charged residues to the yeast actin NH 2 terminus. J. Biol. Chem., in press. COOKE, R. (1986) The mechanism of muscle contraction. CRC Crit. Rev. Biochem. 21:53-118. COOKE, R. (1989) Structure of the myosin head. Cell Motil. Cytoskeleton 14: 183-186. COOKE, R. and FRANKS,K. (1980) All myosin heads form bonds with actin in rigor rabbit skeletal muscle. Biochemistry 19: 2265-2269. CRAIG,R., GREENE,L. E. and EISENBERG,E. (1985) Structure of the actin-myosin complex in the presence of ATP. Proc. hath. Acad. Sci. U.S.A. 82: 3247-3251. CRIDDLE,A. H., GEEVES,M. A. and JEEFRIES,T. 0985) The use of actin labelled with N-(l-pyrenyl)iodoacetamide to study the interaction of actin with myosin subfragments and troponin/tropomyosin. Biochem. J. 232: 343-349. DANCKER,P. (1992) The modification of actomyosin ATPase activity by tropomyosin-troponin and its dependence on ionic strength, ATP-concentration and actin-myosin ratio. Z. NaturJbrsch. 29: 496-505. DANTZIG,J. A., WALKER,J. W., TRENTHAM,D. R. and GOLDMAN,Y. E. (1988) Relaxation of muscle fibers with adenosine 5'-[y-thio]triphosphate (ATP[yS]) and by laser photolysis of caged ATP[7S]: Evidence for Ca2+-dependent affinity of rapidly detaching zero-force crossbridges. Proc. hath. Acad. Sci. U.S.A. 85: 6716-6720. DASGUPTA,G. and REISLER,E. (1989) Antibody against the amino terminus of ~t-actin inhibits actomyosin interactions in the presence of ATP. J. Molec. Biol. 207: 833-836. DASGUPTA,G. and REISLER,E. (1991) Nucleotide-induced changes in the interaction of myosin subfragment 1 with actin: Detection by antibodies against the N-terminal segment of actin. Biochemistry 30: 9961-9966. DAsGUPTA, G. and REISLER,E. (1992) Actomyosin interactions in the presence of ATP and the N-terminal segment of actin. Biochemistry 31: 1836-1841. DE ROSIER,D. J. (1990) Protein structure: The changing shape of actin. Nature 347: 21-22. DEROSIER,D. J. and MOORE,P. B. (1970) Reconstruction of three-dimensional images from electron micrographs of structures with helical symmetry. J. Molec. Biol. 52: 355-369, DOBROWOLSKI,Z., BOROVIKOV,Y. S., NOWAK,E., GALAZKIEWICZ,B. and DABROWSKA,R. (1988) Comparison of Ca2+-dependent effects of caldesmon-tropomyosin-calmodulin and troponin-tropomyosin complexes on the structure of F-actin in ghost fibers and its interaction with myosin heads. Bioehim. biophys. Acta 956: 140-150. DUONG,A. M. and REISLER,E. (1989) Nucleotide-induced states of myosin subfragment 1 cross-linked to actin. Biochemistry 28: 3502-3509. EATON, B. L. (1976) Tropomyosin binding to F-actin induced by myosin heads. Science 192: 1337-1339. EATON, B. L., KOMINZ,D. R. and EISENBERG,E. (1975) Correlation between the inhibition of the acto-heavy meromyosin ATPase and the binding of tropomyosin to F-actin: effects of Mg 2÷, KC1, troponin I and troponin C. Biochemistry 14: 2718-2725. EBASHI,S. and EBASHI,F. (1964) A new protein component participating in the superprecipitation of myosin B. J. Biochem. (Tokyo)55: 604-613. EBASHI,S. and KODAMA,A. (1965) A new protein factor promoting aggregation of tropomyosin. J. Biochem. 58: 107-108. EGELMAN,E, H. and DEROSIER,D. J. (1983) A model for F-actin derived from image analysis of isolated filaments. J. Molec. Biol. 166: 623-629. EGELMAN,E. H. and DEROSIER,D. J. (1991) Angular disorder in actin: Is it consistent with general principles of protein structure. J. Molec. Biol. 217: 405-408. EISENBERG,E. and GREENE,L. E. (1980) The relation of muscle biochemistry to muscle physiology. A. Rer. Physiol. 42: 293-309. EISENBERG,E. and HILL,T. L. (1978) A cross-bridge model of muscle contraction. Prog. Biophys. Molec. Biol. 33: 55-82. EISENBERG,E. and KIELLEY,W. (1970) Native tropomyosin: effect on the interaction of actin with heavy meromyosin and subfragment-1. Biochem. biophys. Res. Commun. 40: 50-56. EISENBERG,E. and KIELLEY,W. W. (1974) Troponin-tropomyosin complex. Column chromatographic separation and activity of the three active troponin components with and without tropomyosin present. J, biol. Chem. 249: 4742-4748. EISENBERG,E. and Moos, C. (1967) The interaction of actin with myosin and heavy meromyosin in solution at low ionic strength. J. biol. Chem. 242: 2945-2951. EISENBERG,E. and Moos, C. 0968) The adenosine triphosphatase activity of acto-heavy meromyosin. A kinetic analysis of actin activation. Biochemistry 7: 1486-1489. EISENBERG,E., DOaKIN,L. and KIELLEY,W. (1972) Heavy meromyosin: evidence for a refractory state unable to bind to actin in the presence of ATP. Proc. hath. Acad. Sci. U.S.A. 69: 667-671. EISENBERG,E., HILL,T. L. and CHEN,Y. (1980) Crossbridge model of muscle contraction. Quantitative analysis. Biophys. J. 29: 195-227. ELLIOTT,G. F., LOWY,J. and MILLMAN,B. (1967) Low angle X-ray diffraction studies of living striated muscle during contraction. J. Molec. Biol. 25: 31-45. EL-SALEH,S. C. and POTTER,J. D. (1985) Calcium-insensitive binding of heavy meromyosin to regulated actin at physiological ionic strength. J. biol. Chem. 260: 14775-14779. Actin mediated regulation of muscle contraction 139 FAJER, P. G., FAJER, E. A., MATTA,J. J. and THOMAS,D. D. (1990) Effect of ADP on the orientation of spin-labeled myosin heads in muscle fibers: A high-resolution study with deuterated spin labels. Biochemistry 29: 5865-5871. FAJER, P. G., FAJER,E. A., SCHOENBERG,M. and THOMAS,D. D. (1991) Orientational disorder and motion of weakly attached crossbridges. Biophys. J. 60: 642--649. FISCHER,W. and PFITZER,G. (1989) Rapid myosin phosphorylation transients in phasic contractions in chicken gizzard smooth muscle. FEBS Lett. 258: 59-62. FLICKER,P. F., MILLIGAN,R. A. and APPLEGATE,D. (1991) Cryoelectron microscopy of Sl-decorated actin filaments. Adv. Biophys. 27: 185-196. FLICKER,P. F., PHILLIPS,G. N., JR and COHEN,C. (1982) Troponin and its interactions with tropomyosin. J. Molec. Biol. 162: 495-501. FORTUNE,N. S., GEEVES,M. A. and RANATUNGA,K. W. (1991) Tension responses to rapid pressure release in glycerinated rabbit muscle fibers. Proc. natn. Acad. Sci. U.S.A. 88: 7323-7327. FREEDMAN,M., HEMRIC,M. E. and CHALOVICH,J. M. (1992) Effect of caldesmon on steady state kinetics using skeletal muscle (A1) subfragment-1. Biophys. J. 61: A154. FuJii, T., IMAI,M., ROSENFELD,G. C. and BRYAN,J. (1987) Domain mapping of chicken gizzard caldesmon. J. biol. Chem. 262: 2757-2763. FUJIMORI,K., SORENSON,M., HERZBERG,O., MOULT,J. and REINACH,F. C. (1990) Probing the calcium-induced conformational transition of troponin C with site-directed mutants. Nature 345: 182-184. FURST, D. O., CROSS,R. A., DEMEY,J. and SMALL,J. V. (1986) Caldesmon is an elongated, flexible molecule localized in the actomyosin domains of smooth muscle. EMBO J. 5: 251-257. GALAZKIEWICZ,B., BOROVIKOV,Y. S. and DABROWSKA,R. (1987) The effect of caldesmon on actin-myosin interaction in skeletal muscle fibers. Biochim. biophys. Acta 916: 368-375. GARLAND,F. and Ch'EUNG,H. C. (1979) Flourescence stopped-flow study of the mechanism of nucleotide binding to myosin subfragment-1. Biochemistry 18: 5281-5289. GEEVES,M. A. (1989) Dynamic interaction between actin and myosin subfragment 1 in the presence of ADP. Biochemistry 28: 5864-5871. GEEVES,M. A. (1991a) The influence of pressure on actin and myosin interactions in solution and in single muscle fibers. J. Cell Sci. 98 (Suppl. 14): 31-35. GEEVES,M. A. (1991b) The dynamics of actin and myosin association and the crossbridge model of muscle contraction. Biochem. J. 274: 1-14. GEEVES,M. A. and CONIBEAR,P. B. (1992) The kinetics of binding actin to S1 and S1. ADP. Biophys. J. 61: A440. GEEVES,M. A. and GUTFREUND,H. (1982) The limiting rate of the ATP-mediated dissociation of actin from rabbit skeletal muscle myosin S-1. FEBS Lett. 160: 141-148. GEEVES, M. A. and HALSALL,D. J. (1986) The dynamics of the interaction between myosin subfragment 1 and pyrene-labelled thin filaments, from rabbit skeletal muscle. Proc. R. Soc. Lond. (Biol.) 229: 85-95. GEEVES,M. A. and HALSALL,D. J. (1987) Two-step ligand binding and cooperativity. A model to describe the cooperative binding of myosin subfragment 1 to regulated actin. Biophys. J. 52: 215-220. GEEVES, M. A. and JErVRIES,T. (1988) The effect of nucleotide upon a specific isomerization of actomyosin subfragment 1. Biochem. J. 256: 41--46. GEEVES,M. A., JEFFRIES,T. E. and MILLAR,N. C. (1986) ATP-induced dissociation of rabbit skeletal actomyosin subfragment 1. Characterizaton of an isomerization of the ternary acto-S1-ATP complex. Biochemistry 25: 8454-8458. GERTHOFF~R, W. T. (1987) Dissociation of myosin phosphorylation and active tension during muscarinic stimulation of tracheal smooth muscle. J. Pharmac. exp. Ther. 240: 8-15. GIMONA,M., HERZOG,M., VANDEKERCKHOVE,J. and SMALL,J. V. (1990) Smooth muscle specific expression of calponin. FEBS Lett. 274: 159-162. GIMONA,M., SPARROWS,M. P., STRASSER,P., HERZOG,M. and SMALL,J. V. (1992) Calponin and SM 22 isoforms in avian and mammalian smooth muscle. Absence of phosphorylation in vivo. Eur. J. Biochem. 205: 1067-1075. GOLDMAN, Y. E., HIBBERD, M. G. and TRENTHAM, D. R. (1984) Relaxation of rabbit psoas muscle fibers from rigor by photochemical generation of adenosine-5'triphosphate. J. Physiol. (Lond.) 354: 577-604. GOODNO, C. C. (1979) Inhibition of myosin ATPase by vanadate ion. Proc. natn. Acad. Sci. U.S.A. 76: 2620-2624. GOODNO,C. C. (1982) Myosin active-site trapping with vanadate ion. Meth. Enzymol. 85:116--123. GOODNO,C. C. and TAYLOR,E. W. (1982) Inhibition of actomyosin ATPase by vanadate. Proc. natn..4cad. Sci. U.S.A. 79: 21-25. GOODY, R. S. and HOFMANN,W. (1980) Stereochemical aspects of the interaction of myosin and actomyosin with nucleotides. J. Musc. Res. Cell Motil. 1:101-115. GOODY, R. S. and HOLMES,K. C. (1983) Crossbridges and the mechanism of muscle contraction. Biochim. biophys. Acta 726: 13-39. GRABAREK,Z. and GERGELY,J. (1987) Troponin I binds to the N-terminal twelve-residue segment of actin. Biophys. J. 51: 331a. JPT55/2--D 140 J.M. CHALOVICH GRABAREK,Z., DRABIKOWSKI,W., LEAVIS,P. C., ROSENFELD,S. S. and GERGELY,J. (1981) Proteolytic fragments of troponin C. Interactions with the other troponin subunits and biological activity. J. biol. Chem. 256: 13121-13127. GRABAREK,Z., TAN, R.-Y., WANG,J., TAO,T. and GERGELY,J. (1990) Inhibition of mutant troponin C activity by an intra-domain disulphide bond. Nature 345: 132-135. GREASER,M. L. and GERGELY,J. (1971) Reconstitution of troponin activity from three protein components. J. biol. Chem. 246: 4226-4233. GREENE L. E. (1982) The effect of nucleotide on the binding of myosin subfragment 1 to regulated actin. J. biol. Chem. 257: 13993-13999. GREENE L. E. (1986) Cooperative binding of myosin subfragment one to regulated actin as measured by fluorescence changes of troponin I modified with different fluorophores. J. biol. Chem. 261: 1279-1285. GREENE L. E. and EISENBERG,E. (1980) Cooperative binding of myosin subfragment-1 to the actin-troponintropomyosin complex. Proc. natn. Acad. Sci. U.S.A. 77: 2616-2620. GREENE L. E. and EISENBERG,E. (1988) Relationship between regulated actomyosin ATPase activity and cooperative binding of myosin to regulated actin. Cell Biophysics 12: 59-71. GREENE L. E., SELLERS,J. R., EISENBERG,E. and ADELSTEIN,R. S. (1983) Binding of gizzard smooth muscle myosin subfragment 1 to actin in the presence and absence of adenosine 5'-triphosphate. Biochemistry 22: 530-535. GREENE, L. E., CHALOVICH,J. M. and EISENBERG,E. (1986) Effect of nucleotide on the binding of N, N'-p-phenylenedimaleimide-modified S-1 to unregulated and regulated actin. Biochemistry 25: 704-709. GREENE,L. E., WILLIAMS,D. L., JR and EISENBERG,E. (1987) Regulation of actomyosin ATPase activity by troponin-tropomyosin: effect of the binding of the myosin subfragrnent 1 (S-I). ATP complex. Proc. natn. Acad. Sci. U.S.A. 84: 3102-3106. GWATHMEY,J. K., HAJJAR, R. J. and SOLARO,R. J. (1991) Contractile deactivation and uncoupling of crossbridges: Effects of 2, 3-butanedione monoxime on mammalian myocardium. Circ. Res. 69: 1280-1292. HAEBERLE,J. R., TRYBUS,K. M. and WARSHAW,D. M. (1991) Caldesmon-dependent regulation of thin filament motility. Biochem. J. 59: 58a. HANSON,J. and LowY, J. (1963) The structure of F-actin and of actin filament isolated from muscle. J. Molec. Biol. 6: 46--60. HARTSrtORNE,D. J. and MUELLER,H. (1968) Fractionization of troponin into two distinct proteins. Biochem. biophys. Res. Commun. 31: 647-653. HASELGROVE,J. C. (1972) X-ray evidence for a conformational change in the actin containing filaments of vertebrate striated muscle. CoM Spring Harbor Syrup. Quant. Biol. 37: 341-352. HEGMANN, T. E., LIN, J. L.-C. and LIN, J. J.-C. (1989) Probing the role of nonmuscle tropomyosin isoforms in intracellular granule movement by microinjection of monoclonal antibodies. J. Cell Biol. 109: 1141-1152. HEGMANN, T. E., SCHULTE, D. L., LIN, J. L.-C. and LIN, J. J.-C. (1991) Inhibition of intracellular granule movement by microinjection of monoclonal antibodies against caldesmon. Cell Motil. Cytoskel. 20: 109-120. HEMRIC,M. E. and CHALOVICH,J. M. (1988) Effect of caldesmon on the ATPase activity and the binding of smooth and skeletal myosin subfragments to actin. J. biol. Chem. 263: 1878-1885. HEMRIC,M. E. and CHALOVICH,J. M. (1990) Characterization of caldesmon binding to myosin. J. biol. Chem. 265: 19672-19678. HEMRIC,M. E., CAREY,J. O., Lu, F. and CHALOVICH,J. M. (1991) Binding of caldesmon to myosin: effect of phosphorylation and calcium-calmodulin. Biophys. J. 59: 57a. HERZBERG,O. and JAMES,M. N. G. (1985) Structure of the calcium regulatory muscle protein troponin-C at 2.8 /~ resolution. Nature 313: 653-659. HERZBERG,O. and JAMES,M. N. G. (1988) Refined crystal structure of troponin C from turkey skeletal muscle at 2.0 A resolution. J. Molec. Biol. 203: 761-779. HmBERD, M. G. and TRENTHAM,D. R. (1986) Relationships between chemical and mechanical events during muscular contracton. A. Rev. Biophys. biochem. Chem. 56: 119-161. HIBBERD,M. G., DANTZIG,J. A., TRENTHAM,D. R. and GOLDMAN,Y. E. (1985) Phosphate release and force generation in skeletal muscle fibers. Science 228: 1317-1319. HILL,T. L. (1974) Theoretical formalism for the sliding filament model of contraction of striated muscle. Prog. Biophys. Molec. Biol. 28: 267-340. HILL,T. L., EISENBERG,E. and GREENE,L. E. (1980) Theoretical model for the cooperative equilibrium binding of myosin subfragment 1 to the actin-troponin-tropomyosin complex. Proc. natn. Acad. Sci. U.S.A. 77: 3186-3190. HILL, T. L., EISENBERG,E. and CHALOVICH,J. M. (1981) Theoretical models for cooperative steady-state ATPase activity of myosin subfragment-1 on regulated actin. Biophys. J. 35: 99-112. HILL, T. L., EISENBERG,E. and GREENE,U E. (1983) Alternate model for the cooperative equilibrium binding of myosin subfragment- 1-nucleotide complex to actin-troponin-tropomyosin. Proe. natn. Acad. Sci. U.S.A. 80: 60-64. HITCHCOCK-DEGREGORI,S. E. (1992) Periodic actin binding sites in tropomyosin. Biophys. J. 61: A3, HITCHCOCK-DEGREGORI,S. E. and VARNELL,T. A. (1990) Tropomyosin has discrete actin-binding sites with sevenfold and fourteenfold periodicities. J. Molec. Biol. 214: 885-896. Actin mediated regulation of muscle contraction 141 HOLMES,K. C., PoPP, D., GEBHARD,W. and KABSCH,W. (1990) Atomic model of the actin filament. Nature 347: 44-49. HOMSHER,E. and MILLAR,N. C. (1990) Caged compounds and striated muscle contraction. A. Rev. Physiol. 52: 875-896. HORIUCHI,K. Y. and CHACKO,S. (1989) Caldesmon inhibits the cooperative turning-on of the smooth muscle heavy meromyosin by tropomyosin-actin. Biochemistry 28:9111-9116. HORIUCHI,K. Y. and CHACKO,S. (1991) The mechanism for the inhibition of actin-activated ATPase of smooth muscle heavy meromyosin by calponin. Biochem. biophys. Res. Commun. 176: 1487-1493. HORIUCHI,K. Y., SAMUEL,M. and CHACKO,S. (1991) Mechanism for the inhibition of acto-heavy meromyosin ATPase by the actin/calmodulin binding domain of caldesmon. Biochemistry 30: 712-717. HUXLEY,A. F. (1957) Muscle structure and theory of contraction. Prog. Biophys. biochem. Chem. 7: 255-318. HUXLEY,H. E. (1969) The mechanism of muscular contraction. Science 164: 1356-1366. HUXLEY,H. E. (1972) Structural changes in the actin and myosin containing filaments during contraction. Cold Spring Harbor Syrup. Quant. Biol. 37: 361-376. HUXLEY, H. E. and BROWN,W. (1967) The low-angle X-ray diagram of vertebrate striated muscle and its behavior during contraction and rigor. J. Molec. Biol. 30: 383-434. HUXLEY,H. E. and KRESS,M. (1985) Crossbridge behaviour during muscle contraction. J. Musc. Res. Cell Motil. 6: 153-161. HUXLEY, H. E., FARUQI,A. R., BORDAS,J., KOCH, M. H. J. and MILCH,J. R. (1980) The use of synchrotron radiation in time-resolved X-ray diffraction studies of myosin layer-line reflections during muscle contraction. Nature 284: 140-143. IKEBE,M. and HORNICK,T. (1991) Determination of the phosphorylation sites of smooth muscle caldesmon by protein kinase C. Arch. Biochem. Biophys. 288: 538-542. IKEBE, M. and REARDON,S. (1988) Binding of caldesmon to smooth muscle myosin. J. biol. Chem. 263: 3055-3058. IKEBE, M. and I~ARDON,S. (1990) Phosphorylation of smooth muscle caldesmon by calmodulin-dependent protein kinase II. Identification of the phosphorylation sites. J. biol. Chem. 265: 17607-17612. IKEBE,M., REARDON,S., SCOTT-WOo,G. C., ZHOU,Z. and KODA,Y. (1990) Purification and characterization of calmodulin-dependent multifunctional protein kinase from smooth muscle: Isolation of caldesmon kinase. Biochemistry 29:11242-11248. IKEMOTO,N., NAGY,B., BHATNAGAR,G. M. and GERGELY,J. (1974) Studies on a metal-binding protein of the sarcoplasmic reticulum. J. biol. Chem. 249: 2357-2365. INOUE,A. and TONOMURA,Y. (1973) Kinetic properties of the myosin-phosphatase-ADP complex. J. Biochem. (Tokyo) 73: 555-566. INOUE,A. and TONOMURA,Y. (1982) Regulation of binding of myosin subfragments with regulated actin by calcium ions in the presence of magnesium ATP. J. Biochem. 91: 1231-1239. IsmI, Y. and LEHRER,S. S. (1987) Fluorescence probe studies of the state of tropomyosin in reconstituted muscle thin filaments. Biochemistry 26: 4922-4924. ISnlI, Y. and LEHRER,S. S. (1990) Excimer fluorescence of pyrenyliodoacetamide-labeled tropomyosin: A probe of the state of tropomyosin in reconstituted muscle thin filaments. Biochemistry 29:1160-1166. lsnlI, Y. and LEHRER,S. S. (1991) Two-site attachment of troponin to pyrene-labeled tropomyosin. J. biol. Chem. 266: 6894-6903. ITOH, T., IKEBE, M., KARGACIN,G. J., HARTSHORNE,D. J., KEMP, B. E. and FAY, F. S. (1989) Effects of modulators of myosin light-chain kinase activity in single smooth muscle cells. Nature 338: 164-167. JOHNSON, K. A. and TAYLOR,E. W. (1978) Intermediate states of subfragment-1 and actosubfragment-1 ATPase: Reevaluation of the mechanism. Biochemistry 17: 3432-3442. JULIAN, F. J. (1969) Activation in a skeletal muscle model with a modification for insect fibrillar muscle. Biophys. J. 9: 547-570. JUNG, D. W. G., BLANGE,T., DEGRAAF,H. and TREHTEL,B. W. (1989) Weakly attached crossbridges in relaxed frog muscle fibers. Biophys. J. 55: 605--619. KABSCH,W., MANNHERZ,H. G., SUCK,D., PAI, E. F. and HOLMES,K. C. (1990) Atomic structure of the actin" DNase I complex. Nature 347: 37-44. KAMATH,U. and SHRIVER,J. W. (1989) Characterization of thermotropic state changes in myosin subfragmentl and heavy meromyosin by UV difference spectroscopy. J. biol. Chem. 264: 5586-5592. KING, R. T. and GREENE,L. E. (1985) Regulation of the adenosine triphosphatase activity of cross-linked actin-myosin subfragment 1 by troponin-tropomyosin. Biochemistry 24: 7009-7014. KING, R. T. and GREENE,L. E. (1987) The conformation of cross-linked actin-S-1 in the presence and absence of ATP. J. biol. Chem. 262: 6128-6134. KONRAD,M. and GOODY,R. S. (1982) Kinetic and thermodynamic properties of the ternary complex between F-actin, myosin subfragment 1 and adenosine 5'-[fl,? imido]triphosphate. Fur. J. Biochem. 128: 547-555. KORETZ,J. F. and TAYLOR,E. W. (1975) Transient state kinetic studies of proton liberation by myosin and subfragment 1. J. biol. Chem. 250: 6344-6350. KORN, E. D. (1982) Actin polymerization and its regulation by proteins from nonmuscle cells. Physiol. Rev. 62: 672-737. KRAFT, Tn., YU, L. C. and BRENNER,B. (1990) Effect of Ca~+ on the actin-attachment of weak binding crossbridges in skinned rabbit psoas fibers. Biophys. J. 57 410a. 142 J.M. CHALOVICH KRAFT, TH., CHALOVICH,J. M., YU, L. C. and BRENNER,B. (1991) Weak cross-bridge binding is essential for force generation. Evidence at near physiological conditions. Biophys. J. 59: 375a. KRESS,M., HUXLEY,H. and FARUQI,A. R. (1986) Structural changes during activation of frog muscle studied by time-resolved X-ray diffraction, d. Molec. Biol. 188: 325-342. KRETSINGER,R. H. (1980) Structure and evolution of calcium-modulated proteins. CRC Crit. Rev. Biochem. 8: 119-174. LEADBEATER,L. and PERRY,S. V. (1963) The effect of actin on the magnesium-activated adenosine triphosphatase of heavy meromyosin. Biochem. J. 87: 233-239. LEAVIS,P. C. and GERGELY,J. (1984) Thin filament proteins and thin filament-linked regulation of vertebrate muscle contraction. CRC Crit. Rev. Biochem. 16: 235-305. LEHMAN,W. and SZENT-GYORGYI,A. G. (1972) Activation of the adenosine triphosphatase of limulus polyphemus actomyosin by troponin. J. gen. Physiol. 59: 375-387. LEHRER, S. S. and ISHII, Y. (1988) Fluorescence properties of acrylodan-labeled tropomyosin and tropomyosin-actin: Evidence for myosin subfragment 1 induced changes in geometry between tropomyosin and actin. Biochemistry 27: 5899-5906. LEHRER,S. S. and MORRIS,E. P. (1982) Dual effects of tropomyosin and troponin-tropomyosin on actomyosin subfragment 1 ATPase. J. biol. Chem. 257: 8073-8080. LEHRER,S. S. and MORRIS,E. P. (1984) Comparison of the effects of smooth and skeletal tropomyosin on skeletal actomyosin subfragment 1 ATPase. J. biol. Chem. 259: 2070-2072. LEVlNE,B. A., MOIR,A. J. G. and PERRY,S. V. (1988) The interaction of troponin-I with the N-terminal region of actin. Eur. J. Biochem. 172: 389-397. LEVINE,B. A., MOIR, A. J. G., AUDEMARD,E., MORNET,D., PATCHELL,V. B. and PERRY,S. V. (1990) Structural study of gizzard caldesmon and its interaction with actin---binding involves residues of actin also recognised by myosin subfragment 1. Eur. J. Biochem. 193: 687-696. LIN, J. J.-C., DAVIS-NANTHAKUMAR,E. J., J1N, J.-P., LOURIM,D., NOVY,R. E. and LIN, J. L.-C. (1991) Epitope mapping of monoclonal antibodies against caldesmon and their effects on the binding of caldesmon to Ca2+/calmodulin and to actin or actin-tropomyosin filaments. Cell Motil. Cytoskel. 20: 95-108. LIN, S.-H. and CHEUNG,H. C. (1991) Two-state equilibria of myosin subfragment 1 and its complexes with ADP and actin. Biochemistry 30: 4317-4322. LYNN, R. W. and TAYLOR,E. W. (1970) Transient state phosphate production in the hydolysis of nucleotide triphosphates by myosin. Biochemistry 9: 2975-2983. LYNN, R. W. and TAYLOR,E. W. (1971) Mechanism of adenosine triphosphate hydrolysis by actomyosin. Biochemistry 10: 4617--4624. MAK, A. S., WATSON,M. H., LITWIN,C. M. E. and WANG,J. H. (1991) Phosphorylation of caldesmon by cdc2 kinase. J. biol. Chem. 266: 6678-6681. MAKUCH, R., BIRUKOV,K., SHIRINSKY,V. and DABROWSKA,R. (1991) Functional interrelationship between calponin and caldesmon. Biochem. J. 280: 33-38. MARIANNE-PEPIN,T., MORNET,D., BERTRAND,R., LABILE,J. P. and KASSAB,R. (1985) Interaction of the heavy chain of gizzard myosin heads with skeletal F-actin. Biochemistry 24: 3024-3029. MARSTON,S. B. (1978) Complex kinetics of actin-subfragment-1 ATPase at low temperature. FEBS Lett. 92: 147-151. MARSTON,S. B. (1982) The rates of formation and dissociation of actin-myosin complexes. Biochem. J. 203: 453-460. MARSTON,S. (1985) The rates of formation and dissociation of actin-myosin complexes: Effects of solvent, temperature, nucleotide coupling and head-head interactions. Biochem. J. 203: 453-460. MARSTON S. (1988) Aorta caldesmon inhibits actin activation of thiophosphorylated heavy meromyosin Mg2+-ATPase activity by slowing the rate of product release. FEBS Lett. 238: 147-150. MARSTON,S. B. (1991) Properties of calponin isolated from sheep aorta thin filaments. FEBS Lett. 292: 179-182. MARSTON,S. B. and LEHMAN,W. (1985) Caldesmon is a Ca 2+ -regulatory component of native smooth-muscle thin filaments. Biochem. J. 231: 517-522. MARSTON S. B. and REDWOOD,C. S. (1991) The molecular anatomy of caldesmon. Biochem. J. 279: 1-16. MARSTON,S. B. and SMITH,C. W. J. (1984) Purification and properties of Ca 2+-regulated thin filaments and F-actin from sheep aorta smooth muscle. J. Musc. Res. Cell Motil. 5: 559-575. MARSTON, S. B. and TREGEAR,R. T. (1984) Modification of the interactions of myosin with actin and 5'-adenylyl imidodiphosphate by substitution of ethylene glycol for water. Biochem. J. 217: 169-177. MARSTON,S. B., RODGER,C. D. and TREGEAR,R. T. (1976) Changes in muscle crossbridges when fl-~ binds to myosin. J. Molec. Biol. 104: 263-276. MARSTON,S. B., TREGEAR,R. T., RODGER,C. D. and CLARKE,M. L. (1978) Coupling between the enzymatic site and the mechanical output of muscle. J. Molec. Biol. 128:111-126. MARSTON,S. B., REDWOOD,C. S. and LEHMAN,W. (1988) Reversal of caldesmon function by anticaldesmon antibodies confirms its role in the calcium regulation of vascular smooth muscle thin filaments. Biochem. biophys. Res. Commun. 155: 197-202. MARUTA, S., HENRY, G. D., SYKES, B. D. and IKEIIE,M. (1991) Formation of the stable smooth muscle myosin-ADP-A1F3 complex and its analysis using t9F-NMR. Biophys. J. 59: 436a. Actin mediated regulation of muscle contraction 143 MARUYAMA,K. and GERGELY,J. (1962) Interaction of actomyosin with adenosine triphosphate at low ionic strength: I. Dissociation of actomyosin during clear phase. J. biol. Chem. 237: 1095. MARUYAMA,K. and WATANABE,S. (1962) The role of magnesium in the superprecipitation of myosin B. J. biol. Chem. 237: 3437. McKILLOP, D. F. A. and GEEVES,M. A. (1991) Regulation of the actomyosin subfragment l interaction by troponin/tropomyosin. Evidence for control of a specific isomerization between two actomyosin subfragment 1 states. Biochem. J. 279: 711-718. MCLACHLAN,A. D. and STEWART,M. (1976) The 14-fold periodicity in alpha-tropomyosin and the interaction with actin. J. Molec. Biol. 103: 271-298. MEEUSEN, R. L. and CANDE, Z. W. (1979) N-ethylmaleimide-modified heavy meromyosin. A probe for actomyosin interactions. J. Cell. Biol. 82: 57-65. MEHEGAN,J. P. and TOBACMAN,L. S. (1991) Cooperative interactions between troponin molecules bound to the cardiac thin filament. J. biol. Chem. 266: 966-972. MEJEAN,C., BUYER,M., LABBE,J.-P., MARLIER,L., BENYAMIN,Y. and ROUSTAN,C. (1987) Anti-actin antibodies. An immunological approach to the myosin-actin and the tropomyosin-actin interfaces. Biochem. J. 244: 571-577. METZGER, J. M. and Moss, R. L. (1991) Phosphate and the kinetics of force generation in skinned skeletal muscle fibers. Biophys. J. 59: 418a. MIKAWA,T. (1979) 'Freezing' of the calcium-regulated structures of gizzard thin filaments by glutaraldehyde. J. Biochem. 85: 879-881. MIKI, M. and HOZUMI,T. (1991) Interaction of maleimidobenzoyl actin with myosin subfragment l and tropomyosin-troponin. Biochemistry 30: 5625-5630. MILLAR,N. C. and GEEVES,M. A. (1983) The limiting rate of the ATP-mediated dissociation of actin from rabbit skeletal muscle myosin subfragment 1. FEBS Lett. 160: 141-148. MILLAR,N. C. and GEEVES,M. A. (1988) Protein fluorescence changes associated with ATP and adenosine 5'-[7-thio]-triphosphate binding to skeletal muscle myosin subfragment 1 and actomyosin subfragment 1. Biochem. J. 249: 735-743. MILLAR,N. C. and HOMSHER,E. (1992) The effect of phosphate and calcium on force generation in glycerinated rabbit skeletal muscle. J. biol. Chem. 265: 20234--20240, MILLER, L., KALNOSKI,i . , YUNOSSI,Z., BULINSKI,J. C. and REISLER,E. (1987) Antibodies directed against N-terminal residues on actin do not block acto-myosin binding. Biochemistry 26: 6064-6070. MILLIGAN,R. A., WHITTAKER,M. and SAFER,D. (1990) Molecular structure of F-actin and location of surface binding sites. Nature 348: 217-221. MOORE, P. B., HUXLEY,H. E. and DEROSIER,D. J. (1970) Three-dimensional reconstruction of F-actin, thin filaments and decorated thin filaments. J. Molec. Biol. 50: 279-295. MORNET, D., BERTRAND,R., PANTEL,P., AUDEMARD,E. and KASSAB,R. (1981) Structure of the actin-myosin interface. Nature 292: 301-306. Moss, R. L., GIULIAN,G. G. and GREASER,M. L. (1985) The effects of partial extraction of TnC upon the tension-pCa relationship in rabbit skinned skeletal muscle fibers. J. gen. Physiol. 86: 585-600. MOSS,R. t., ALLEN,J. n. and GREASER,M. t. (1986) Effects of partial extraction of troponin complex upon the tension-pCa relation in rabbit skeletal muscle. J. gen. Physiol. 871: 761-774. MUHLRAD,A., PEYSER,Y. M. and WERBER,M. M. (1992) Characterization of stable myosin subfragment l(Sl) complexes with beryllium and aluminum fluoride. Biophys. J. 61: A148:848. MULHERN,S. and EISENBERG,E. (1976) Further studies on the interaction of actin with heavy meromyosin and subfragment-1 in the presence of ATP. Biochemistry 15: 5702-5707. MURRAY,J. i . , WEBER,A. and WEGNER,A. (1980) Tropomyosin and the various states of the regulated actin filament. In: Muscle Contraction: Its Regulatory Mechanisms, pp. 221-236, EBASH1,S. (ed.) Springer-Verlag, Berlin. MURRAY,J. M., KNOX,i . K., TRUEBLOOD,C. E. and WEBER,A. (1982) Potentiated state of the tropomyosinactin filament and nucleotide-containing myosin subfragment 1. Biochemistry 21" 906-915. MUSHTAQ, E. and GREENE,L. E. (1989) Effect of ethylene glycol on the interaction of different myosin subfragment 1 nucleotide complexes with actin. Biochemistry 28: 6478-6482. NAGASHIMA,H. and ASAKURA,S. (1982) Studies on cooperative properties of tropomyosin-actin and tropomyosin-troponin-actin complexes by the use of N-ethylmaleimide-treated and untreated species of myosin subfragment 1. J. Molec. Biol. 155: 409-428. NGAI, P. K. and WALSH,M. P. (1984) Inhibition of smooth muscle actin-activated myosin Mg2+-ATPase activity by caldesmon. J. biol. Chem. 259: 13656-13659. NISHIDA,W., ABE, i . , TAKAHASHI,K. and HIWADA,K. (1990) Do thin filaments of smooth muscle contain calponin?: A new method for the preparation. FEBS Lett. 268: 165-168. NOWAK, E., BOROVIKOV,Y. S. and DABROWSKA,R. (1989) Caldesmon weakens the bonding between myosin heads and actin in ghost fibers. Biochim. biophys. Acta 999: 289-292. O'BRIEN,E. J., BENNET,P. M. and HANSON,J. (1971) Optical diffraction studies of myofibrillar structure. Phil. Trans. R. Soc. London Ser. B 261: 201-208. PAN, B.-S., GORDON, A. M. and Luo, Z. (1989) Removal of tropomyosin overlap modifies cooperative binding of myosin S-1 to reconstituted thin filaments of rabbit striated muscle. J. biol. Chem. 264: 8495-8498. 144 J.M. CHALOVICH PARKER,L., PYUN, H. Y. and HARTSHORNE,D. J. (1970) The inhibition of the adenosine triphosphatase activity of the subfragment 1-actin complex by troponin plus tropomyosin, troponin-B plus tropomyosin and troponin B. Biochim. biophys. Acta 223: 453-456. PARRY, D. A. D. (1975) Letter: Double helix tropomyosin. Nature 256: 346-347. PARRY, n . A. D. and SQUIRE,J. M. (1973) Structural role of tropomyosin in muscle regulation: analysis of the X-ray diffraction patterns from relaxed and contracting muscles. J. Molec. Biol. 75: 33-55. PATE,E. and COOKE,R. (1989) Addition of phosphate to active muscle fibers probes actomyosin states within the power stroke. Pfliigers Arch. 414: 73-81. PATE,E., NAKAMAYE,K. L., FRANKS-SK1BA,K., YOUNT,R. G. and COOKE,R. (1991) Mechanics of glycerinated muscle fibers using nonnucleoside triphosphate substrates. Biophys. J. 59: 598~05, PEMRICK, S. and WEBER, A. (1976) Mechanism of inhibition of relaxation by N-ethylmaleimide treatment of myosin. Biochemistry 15: 5193-5198. PERKINS,W. J., WELLS,J. A. and YOUNT, R. G. (1981) Fluorescence studies of the conformation of myosin subfragment 1 (SF1) crosslinked with p-phenylenedimaleimide (pPDM) in the presence of adenine nucleotides. Biophys. J. 33: 149a. PERRY, S. V., COLE, H. A., HEAD, J. F. and WILSON, F. J. (1972) Localization and mode of action of the inhibitory protein component of the tropomyosin complex. CoM Spring Harbor Syrup. Quant. Biol. 37: 251-262. PFITZER,G., FISCHER,W. and CHALOVICH,J. M. (1992) Phosphorylation-contraction coupling in smooth muscle: role of caldesmon. In: Mechanism of Myofilament Sliding in Muscle Contraction, H. SUGI (ed.) Plenum Publishing, New York, in press. PHAN, B. C. and REISLER,E. (1992) Inhibition of myosin ATPase by beryllium fluoride (BeF 3 ). Biophys. J. 61: A300:1721. PHILLIPS,G. N., JR and CHACKO,S. (1992) Mechanical properties of tropomyosin and implications for muscle contraction. J. Molec. Biol., in press. PHILLIPS,G. N., JR, FILLERS,J. P. and COHEN,C. (1986) Tropomyosin crystal structure and muscle regulation. J. Molec. Biol. 192: lll-131. PODOLSKY,R. J. and TEICHHOLZ,L. E. (1970) The relation between calcium and contraction kinetics in skinned muscle fibers. J. Physiol. (Lond.) 211: 19-35. POLLARD,T. D. (1990) AcriD. Curt. Op. Cell Biol. 2: 33-40. POPE,B., WAGNER,P. I~. and WEEDS,A. G. (1981) Studies on the actomyosin ATPase and the role of the alkali light chains. Eur. J. Biochem. 117: 201-206. POPP, D., MAEDA, Y., STEWART,A. A. E. and HOLMES, K. C. (1991) X-ray diffraction studies on muscle regulation. In: Advances in Biophysics, pp. 89-103, KOTANI, M. (ed.) Elsevier, Limerick. POTTER,J. D. and GERGELY,J. (1975) The calcium and magnesium binding sites on tropomyosin and their role in the regulation of myofibrillar adenosine triphosphatase. J. biol. Chem. 250: 4628-4633. PRINGLE,J. W. S. (1967) The contractile mechanism of insect fibrillar muscle. Prog. Biophys. biophys. Chem. 17: 1-60. REEDY, M. K., HOLMES,K. C. and TREGEAR,R. T. (1965) Induced changes in orientation of the crossbridges of glycerinated insect flight muscle. Nature 207: 1276-1280. REISLER,E. (1980) Kinetic studies with synthetic myosin minifilaments show the equivalence of actomyosin and acto-HMM ATPases. J. biol. Chem. 255: 9541-9544. REISLER,E., BURKE,M., HIMMELFARB,S. and HARRINGTON,W. F. (1974) Spatial proximity of the two essential sulfhydryl groups of myosin. Biochemistry 13: 3837-3840. ROSENFELD, S. S. and TAYLOR, E. W. (1984) The ATPase mechanism of skeletal and smooth muscle acto-subfragment 1. J. biol. Chem. 259: 11908-11919. ROSENFELD,S. S. and TAYLOR, E. W. (1987) The mechanism of regulation of actomyosin subfragment l ATPase. J. biol. Chem. 262: 9984-9993. ROUAYRENC,J. F., BERTRAND,R., KASSAB,R., WALZHUNG,n., BOHLER,M. and WALLIMANN,T. (1985) Further characterization of the structural and functional properties of the cross-linked complex between F-actin and myosin S-1. Fur. J. Biochem. 146: 391-401. SCHAUB, M. C. and PERRY, S. V. (1969) The relaxing protein system of striated muscle: Resolution of the troponin complex into inhibitory and calcium-ion sensitizing factors and their relationship to tropomyosin. Biochem. J. 115: 993-1004. SCHNEKENBUHL,S., KRAFT,TH., Yu, L. C., CHALOVICH,J. M. and BRENNER,B. (1991) Characterization of NEM-S-I binding to the thin filament in single skinned rabbit psoas fibers. Biophys. J. 59: 574A. SCHNEKENBUHL,S., KRAFT,TH., Yu, L. C., BRENNER,B. and CHALOVICH,J. M. (1992) Effect of NEM-S-1 on cross-bridge action in skinned rabbit psoas muscle fibers. Biochemical, mechanical and X-ray diffraction studies. Biophys. J. 61: 292A. SCHOENBERG,M. (1985) Equilibrium muscle cross-bridge behavior: theoretical considerations. Biophys. J. 48: 467-475. SCHOENBERG,M. (1988) The kinetics of weakly- and strongly-binding crossbridges: implications for contraction and relaxation. In: Molecular Mechanism of Muscle Contraction, pp. 189-202, SUGI, H. and POLLACK, G. H. (eds) Plenum Publishing Corp., New York. SCHOENBERG,M. (1989) Effect of adenosine triphosphate analogs on skeletal muscle fibers in rigor. Biophys. J. 56: 33-41. Actin mediated regulation of muscle contraction 145 SCHOENBERG,M. and EISENBERG,E. (1985) Characterization of the myosin adenosine triphosphate (M-ATP) crossbridge in rabbit and frog skeletal muscle fibers. Biophys. J. 48: 863-871. SCHOENBERG, M., BRENNER, B., CHALOVICH,J. M., GREENE, L. E. and EISENBERG,E. (1984) Cross-bridge attachment in relaxed muscle. In: Contractile Mechanisms in Muscle, pp 269-284, SUGI,H. and POLLACK, G. H. (eds) Plenum Publishing Corp., New York. SCHWYTER,D., PHILLIPS,M. and REISLER,E. (1989) Subtilisin-cleaved actin: Polymerization and interaction with myosin subfragment 1. Biochemistry 28: 5889-5895. SCOTT-Woo, G. C., SUTHERLAND,C. and WALSH,M. P. (1990) Kinase activity associated with caldesmon is Ca2+/calmodulin-dependent kinase II. Biochem. J. 265: 367-370. SEK1VA,K., TAKEUCHI,K. and TONOMURA,Y. (1967) Binding of H-meromyosin with F-actin at low ionic strength. J. Biochem. (Tokyo) 61: 567-579. SELLERS,J. R., EISENBERG,E. and ADELSTEI~,R. S. (1982) The binding of smooth muscle heavy meromyosin to actin in the presence of ATP: effect of phosphorylation. J. biol. Chem. 23: 13880-13883. SELLERS,J. R., HAtq, Y. J. and KACHAR,B. (1991) Movement of actin filaments by purified molluscan myosin and native thick filaments. Biophys. J. 59: 187a. SEYMOUR,J. and O'BRIEN, E. J. (1980) The position of tropomyosin in muscle thin filaments. Nature 283: 680--682. SnRIVER,J. W. (1986) The structure of myosin and its role in energy transduction in muscle. Biochem. Cell Biol. 64: 265-276. SnRIVER, J. W. and SYKES, B. D. (1981a) Phosphorus-31 nuclear magnetic resonance evidence for two conformations of myosin subfragment-1 nucleotide complexes. Biochemistry 20: 2004-2012. SHRIVER,J. and SYKES, B. D. (1981b) Energetics of the equilibrium between two nucleotide-free myosin subfragment-I states using fluorine-19 nuclear magnetic resonance. Biochemistry 21: 3022-3028. SIEMA~KOWSKI,R. F. and WroTE, H. D. (1984) Kinetics of the interaction between actin, ADP and cardiac myosin-S-l. J. biol. Chem. 259: 5045-5053. SmMANKOWSrd,R. F., WISEMAN,M. O. and WHITE,H. D. (1985) ADP dissociation from actomyosin subfragment 1 is sufficiently slow to limit the unloaded shortening velocity in vertebrate muscle. Proc. natn. Acad. Sci. U.S.A. $2: 658-662. SLEEP, J. A. and BORER,P. D. (1978) Effect of actin concentration on the intermediate oxygen exchange of myosin; relation to the refractory state and the mechanism of exchange. Biochemistry 17: 5417-5422. SLEEP,J. A. and HUTTON,R. U (1980) Exchange between inorganic phosphate and adenosine 5'-triphosphate in the medium by actomyosin subfragment 1. Biochemistry 19: 1276-1283, SLEEP, J. A. and TAYLOR, E. W. (1976) Intermediate states of actomyosin adenosine triphosphatase. Biochemistry 15: 5813-5817. SMITH, C. W. J., PRITCI-IARD,K. and MARSTON,S. B. (1987) The mechanism of Ca 2+ regulation of vascular smooth muscle thin filaments by caldesmon and calmodulin. J. biol. Chem. 262:116-122. SMITH, S. J. and EISENBERG,E. (1990) A comparison of the effect of vanadate on the binding of myosinsubfragment-1 ADP to actin and on actomyosin subfragment 1 ATPase activity. Eur. J. Biochem. 193: 69-73. SOaIESZEK,A. (1982) Steady-state kinetic studies on the actin activation of skeletal muscle heavy meromyosin subfragments. J. Molec. Biol. 157: 275-286. SOBUE,K. and SELLERS,J. R. (1991) Caldesmon, a novel regulatory protein in smooth muscle and nonmuscle ~ actomyosin systems. J. biol. Chem. 266:12115-12118. SOBVE,K., MtraAMOXO,Y., FUJITA,M. and KAKIUCHI,S. (1981) Purification of a calmodulin-binding protein from chicken gizzard that interacts with F-actin. Proc. natn. Acad. Sci. U.S.A. 78: 5652-5655. SOBUE,K., MORIMOTO,K., INUI,M., KANDA,K, and KAKIUCHI,S. (1982) Control of actin-myosin interaction of gizzard smooth muscle by calmodulin- and caldesmon-linked flip-flop mechanism. Biomed. Res. 3: 188-196. SPUDICH,J. A. and WART,S. (1971) The regulation of rabbit skeletal muscle contraction. I. Biochemical studies of the interaction of the tropomyosin-troponin complex with actin and the proteolytic fragments of myosin. J. biol. Chem. 246: 4866-4871. SPUDICH, J. A., HUXLEY,H. E. and FINCH, J. T. (1972) The regulation of skeletal muscle contraction. II. Structural studies of the interaction of the tropomyosin-troponin complex with actin. J. Molec. Biol. 72: 619-632. SQUIRE,J. M. (1981) Muscle regulation: A decade of the steric blocking model. Nature 291: 614-615. STEIN, L. A. (1988) The modeling of the actomyosin subfragment-1 ATPase activity. Cell Biophys. 12: 29-58. STEIN, L. A. and CHALOVICH,J. M. (1991) Activation of skeletal S-I ATPase activity by actin-tropomyosintroponin. Effect of Ca 2+ on the fluorescence transient. Biophys. J. 60: 399-407. STEIN,L. A., Sch'wAgz, R. P., JR, CHOCK,P. B. and EISENBERG,E. (1979) Mechanism of actomyosin adenosine triphosphatase: evidence that adenosine 5'-triphosphate hydrolysis can occur without dissociation of the actomyosin complex. Biochemistry 18: 3895-3909. STEIN,L. A., CHOCK,P. B. and EISENBERG,E. (1981) Mechanism of the actomyosin ATPase: effect of actin on the ATP hydrolysis step. Proc. natn. Acad. Sci. U.S.A. 78: 1346-1350. STEIN, L. A., CHOCK, P. B. and EISENBERG,E. (1984) The rate-limiting step in the actomyosin adenosine triphosphatase cycle. Biochemistry 23: 1555-1563. 146 J.M. CHALOVICH STEIN, L. A., GREENE,L. E., CHOCK,P. B. and EISENBERG,E. (1985) Rate-limiting step in the actomyosin adenosine triphosphatase cycle: studies with myosin subfragment 1 cross-linked to actin. Biochemistry 24: 1357-1363. STRZELECKA-GOLASZEWSKA,H., KLIMASZEWSKA,U. and DYDYNSKA,M. (1979) Polyphasic character of ATP hydrolysis in actomyosin system. Fur. J. Biochem. 101: 523-530. SUEMATSU,E., RESNICK,M. and MORGAN,K. G. (1991) Change of Ca 2÷ requirement for myosin phosphorylation by prostaglandin F2~. Am. J. Physiol. 261: C253-C258. SUNDARALINGAM,M., BERGSTROM,R., STRASBURG,G., RAO, S. T., ROYCHOWDHURY,P., GREASER,M. L. and WANG, B. C. (1985) Molecular structure of troponin C from chicken skeletal muscle at 3 /~ resolution. Science 227: 945-948. SUTHERLAND,C. and WALSH,M. P. (1989) Phosphorylation of caldesmon prevents its interaction with smooth muscle myosin. J. biol. Chem. 264: 578-583. SUTOH, K., ANDO, M., SUTOH,K. and TOYOSHIMA,Y. Y. (199 l) Site-directed mutations of Dictyostelium actin: disruption of a negative charge cluster at the N terminus. Proc. HatH. Acad. Sci. U.S.A. 88:7711-7714. SVENSSON,E. C. and THOMAS,D. D. (1986) ATP induces microsecond rotational motions of myosin heads crosslinked to actin. Biophys. J. 50: 999-1002. SWARTZ,D. R. and Moss, R. L. (1991) Influence of a tight-binding myosin analog, N-ethylmaleimide S1, on calcium-sensitive mechanical properties of rabbit skinned psoas fibers. Biophys. J. 59: 587A. SWARTZ, O. R., HOEMANN,P. A. and Moss, R. L. (1992) Differential effects of NEM-S1 on the calcium sensitivity of tension in skinned preparations of myocardium and fast and slow skeletal muscles. Biophys. J. 61: A293. SWENSON, C. A. and FREDRICKSEN,R. S. (1992) Interaction of troponin C and troponin C fragments with troponin I and the troponin I inhibitory peptide. Biochemistry 31" 3420-3429. SWENSON,C. A. and STELLWAGEN,N. C. (1989) Flexibility of smooth and skeletal tropomyosins. Biopolymers 28: 955-963. SYSKA,H., WILKINSON,J. M., GRAND,R. J. A. and PERRY, S. V. (1976) The relationship between biological activity and primary structure of Troponin I from white skeletal muscle of rabbit. Biochem. J. 153: 375-387. SZPACENKO,A. and DABROWSKA,R. (1986) Functional domain of caldesmon. FEBS Lett. 202: 182-186. TAKAHASHI,K., HIWADA,K. and KOKUBU,T. (1988) Vascular smooth muscle calponin: a novel troponin T-like protein. Hypertension 11: 620~26. TALBOT, J. A. and HODGES, R. S. (1979) Synthesis and biological activity of an icosapeptide analog of the actomyosin ATPase inhibitory region of Troponin I. J. biol. Chem. 254: 3720-3723. TALBOT,J. A. and HODGES, R. S. (1981) Synthetic studies on the inhibitor region of rabbit skeletal troponin I. J. biol. Chem. 256:2798 2802. TANAKA,T., OHTA,H., KANDA,K., HIDAKA,H. and SOBUE,K. (1990) Phosphorylation of high MW caldesmon by protein kinase C modulates the regulatory function of this protein on the interaction between actin and myosin. Eur. J. Biochem. 188: 495-500. TANSEY,M. G., HORI, M., KARAKI,n., KAMM,K. E. and STULL,J. T. (1990) Okadaic acid uncouples myosin light chain phosphorylation and tension in smooth muscle. FEBS Lett. 270: 219-221. TAYLOR,E. W. (1977) Transient phase of adenosine triphosphate hydrolysis by myosin, heavy meromyosin and subfragment-1. Biochemistry 16: 732-740. TAYLOR,E. W. (1989) Actomyosin ATPase mechanism and muscle contraction. Prog. clin. Biol. Res. 315: 9-14. TAYLOR,E. W. (1991) Kinetic studies on the association and dissociation of myosin subfragment 1 and actin. J. biol. Chem. 266: 294-302. TAYLOR,E. W., LYMN,R. W. and MOLL, G. (1970) Myosin-product complex and its effect on the steady-state rate of nucleoside triphosphate hydrolysis. Biochemistry 9: 2984-2991. TAYLOR,K. A. and AMOS,L. (1981) A new model for the geometry of the binding of myosin crossbridge to muscle thin filaments. J. Molec. Biol. 147: 297-324. TESI, C., BARMAN,T. and TRAVERS,F. (1990) Is a four-state model sufficient to describe actomyosin ATPase. FEBS Lett. 260: 229-232. THOMAS, D. D. (1987) Spectroscopic probes of muscle cross-bridge rotation. A. Rev. Physiol. 49: 691-709. TOBACMAN,L. S. and ADELSTEIN,R. S. (1986) Mechanism of regulation of cardiac actin-myosin subfragment 1 by troponin-tropomyosin. Biochemistry 25: 798-802. TOBACMAN,L. S., BUTTERS,C., HILL,L., WILLADSEN,K. and MEHEGAN,J. P. (1992) Effects of tropomyosin acetylation and of the N-terminal region of troponin T on thin filament assembly and cooperativity. Biophys. J. 61: A157. TOZEREN,A. and SCHOENBERG,M. (1986) The effect of cross-bridge clustering and head-head competition on the mechanical response of skeletal muscle under equilibrium conditions. Biophys. J. 50: 875-884. TRAYER,H. R. and TRAYER,I. P. (1988) Fluorescence resonance energy transfer within the complex formed by actin and myosin subfragment 1. Comparison between weakly and strongly attached states. Biochemistry 27: 5718-5727. Tavaus, K. M. and TAYLOR,E. W. (1980) Kinetic studies of the cooperative binding of subfragment l to regulated actin. Proc. natn. Acad. Sci. U.S.A. 77: 7209-7213. TRYBUS,K. M. and TAYLOR,E. W. (1982) Transient kinetics of adenosine 5'-diphosphate and adenosine 5'-(fl, y-imidotriphosphate) binding to subfragment-1 and actosubfragment-l. Biochemistry 21: 1284-1294. VANDEKERCKHOVE,J. (1990)Actin-binding proteins. Curr. Op. Cell Biol. 2: 41-50. Actin mediated regulation of muscle contraction 147 VELAZ,L., HEMRIC,M. E., BENSON,C. E. and CHALOVICH,J. M. (1989) The binding of caldesmon to actin and its effect on the ATPase activity of soluble myosin subfragments in the presence and absence of tropomyosin. J. biol. Chem. 264: 9602-9610. VELAZ, L., INGRAHAM,R. H. and CHALOVICH,J. M. (1990) Dissociation of the effect of caldesmon on the ATPase activity and on the binding of smooth heavy meromyosin to actin by partial digestion of caldesmon. J. biol. Chem. 265: 2929-2934. VIBERT,P. J., HASELGROVE,J. C., LOWY,J. and POULSEN,F. R. (1972) Structural changes in actin-containing filaments in muscle. J. Molec. Biol. 716: 757-767. VOROTNIKOV,A. V., SHRINSKV,V. P. and GUSEV,N. B. (1988) Phosphorylation of smooth muscle caldesmon by three protein kinases: implication for domain mapping. FEBS Lett. 236: 321-324. WAGNER,P. D. (1984) Effect of skeletal muscle myosin light chain 2 on the Ca 2+-sensitive interaction of myosin and heavy meromyosin with regulated actin. Biochemistry 23: 5950-5956. WAGNER,P. D. (1986) Unphosphorylated calf thymus and aorta myosins contract ghost myofibrils. J. biol. Chem. 261: 14863-14866. WAGNER,P. D. and GEORGE,J. N. (1986) Phosphorylation of thymus myosin increases its apparent affinity for actin but not its maximum adenosine triphosphatase rate. Biochemistry 25: 913-918. WAGNER,P. D. and GINIGER,E. (1981) Calcium sensitive binding of heavy meromyosin to regulated actin in the presence of ATP. J. biol. Chem. 256: 12647-12650. WAGNER,P. D. and STONE,D. B. (1983) Calcium-sensitive binding of heavy meromyosin to regulated actin requires light chain 2 and the head-tail junction. Biochemistry 22: 1334-1342. WAGNER,P. D. and WEEDS,A. G. (1979) Determination of the association of myosin subfragment-1 with actin in the prescence of ATP. Biochemistry 18: 2260-2266. WAKABAYASHI,Z. and TOYOSHIMA,C. (1981) Three-dimensional image analysis of the complex of the thin filaments and myosin molecules from skeletal muscle II: The multi domain structure of actin-myosin Sl complex. J. Biochem. 90: 683-701. WAKABAYASHI,T., HUXLEY,H. E., AMOS,L. A. and KLUG,A. (1975) Three-dimensional image reconstruction of actin-tropomyosin complex and actin-tropomyosin-TropT TropI complex. J. Molec. Biol. 93: 477-497. WALKER,G., KERRICK,W. G. L. and BOURGUIGNON,Y. W. (1989) The role of caldesmon in the regulation of receptor capping in mouse T-lymphoma cell. J. biol. Chem. 264: 496-500. WALKER, J. W., LU, Z., SWARTZ,D. R. and Moss, R. L. (1991) Thin filament modulation of cross-bridge transitions measured by photogeneration of Pi in skeletal muscle fibers. Biophys. J. 59: 418a. WANG, C.-L. A., CHALOVICH,J. M., GRACEFFA,P., LU, R. C., MABUCHI,K. and STAFFORD,W. F. (1991) A long helix from the central region of smooth muscle caldesmon. J. biol. Chem. 266: 13958-13963. WEEKS, R. A. and PERRY, S. V. (1978) Characterization of a region of the primary sequence of troponin-c involved in calcium ion-dependant interaction with troponin-i. Biochem. J. 173: 449-457. WEGNER,A. (1979) Equilibrium of the actin-tropomyosin interaction. J. Molec. Biol. 131: 839-853. WEGNER,A. and WALSH,T. P. (1981) Interaction of tropomyosin-troponin with actin filaments. Biochemistry 20: 5633-5642. WEIGT,C., WEGNER,A. and KOCH,M. H. J. (1991) Rate and mechanism of the assembly of tropomyosin with actin filaments. Biochemistry 30: 10700-10707. WELLS,C. and BAGSHAW,C. R. (1984) The characterization of vanadate-trapped nucleotide complexes with spin-labeled myosins. J. Musc. Res. Cell MotiL 5:97-112. WELLS,J. A. and YOUNT,R. G. (1982) Chemical modification of myosin by active-site trapping of metal-nucleotides with thiol crosslinking reagents. Meth. Enzymol. 85:93-116. WHITBY,F. G., KENT,H., STEWART,F., STEWART,M., XIE, X., HATCH,V., COHEN,C. and PHILLIPS,G. N., JR (1992) Structure of tropomyosin at 9/~ resolution. J. Molec. Biol., in press. WHITE,H. D. and TAYLOR,E. W. (1976) Energetics and mechanism of actomyosin adenosine triphosphatase. Biochemistry 15: 5818-5826. WHITE, S. P., COHEN,C. and PHILLIPS,G. N., JR (1987) Structure of co-crystals of tropomyosin and troponin. Nature 325: 826-828. WILKINSON,J. M. and GRAND,R. J. A. (1978) Comparison of amino acid sequence of troponin I from different striated muscles. Nature 271: 31-35. WILKINSON,J. M., PERRY,S. V., COLE,H. A. and TRAYER,I. P. (1972) The regulatory proteins of the myofibril: separation and biological activity of the components of inhibitory factor preparations. Biochem. J. 127: 215-228. WILLIAMS,D. L., GREENE,L. E. and EISENaERG,E. (1984) Comparison of effects of smooth and skeletal muscle tropomyosins on interactions of actin and myosin subfragment 1. Biochemistry 23: 4150-4155. WILLIAMS,D. L., JR, GREENE,L. E. and EISENBERG,E. (1988) Cooperative turning on of myosin subfragment 1 adenosine triphosphatase activity by the troponin-tropomyosin-actin complex. Biochemistry 27: 6897-6993. WINDER, S. J. and WALSH,M. P. (1990) Smooth muscle calponin. Inhibition of actomyosin MgATPase and regulation by phosphorylation. J. biol. Chem. 265: 10148-10155. WINKELMANN,D. A., BAKER,T. S. and RAYMENT,I. (1991) Three-dimensional structure of myosin subfragment1 from electron microscopy of sectioned crystals. J. Cell Biol. 114: 701-713. WOLEDGE,R. C., CURTIN,N. A. and HOMSHER,E. 0985) Energetic Aspects of Muscle Contraction, Academic Press, New York. 148 J.M. CHALOVICH YAGI, K., NAKATA,T. and SAKAKIBARA,I. (1965) Dissociation of a complex of F-actin and H-meromyosin by addition of adenosine triphosphate. J. Biochem. (Tokyo) 58: 236-242. YAGI, N. and MATSUBARA,I. (1980) Myosin heads do not move on activation in highly stretched vertebrate striated muscle. Science 207: 307-308. YAGI,N. and MATSUBARA,I. (1989) Structural changes in the thin filament during activation studied by X-ray diffraction in highly stretched skeletal muscle. J. Molec. Biol. 208: 359-363. YAMASHIRO,S. and MATSUMURA,F. (199 l) Mitosis-specific phosphorylation of caldesmon- Possible molecular mechanism of cell rounding during mitosis. Bioessays 13: 563-568. YAMASHIRO,S., YAMAKITA,Y., ISHIKAWA,R. and MATSUMURA,F. (1990) Mitosis-specific phosphorylation causes 83K nonmuscle caldesmon to dissociate from microfilaments. Nature 344: 675~578. YAMASHIRO, S., YAMAKITA,Y., HOSOYA, H. and MATSUMURA,F. (1991) Phosphorylation of nonmuscle caldesmon by p34cdc2kinase during mitosis. Nature 349: 169-172. YANAGIDA,T. and OOSAWA,F. (1978) Polarized flourescence from eATP incorporated into F-actin in a myosin-free single fiber: Conformation of F-actin and changes induced in it by heavy meromyosin. J. Molec. Biol. 126: 507-524. YANAGIDA,T. and OOSAWA,F. (1980) Conformational changes of F-actin-e-ADP in thin filaments in myosin-free muscle fibers induced by Ca 2+ . J. Molec. Biol. 140: 313--320. YOUNT, R. G., OJALA,D. and BABCOCK,D. (1971) Interaction of P-N-P and P-C-P analogs of adenosine triphosphate with heavy meromyosin, myosin and actomyosin. Biochemistry 10: 2490-2495. YF, L. C. and BRENNER,B. (1989) Structures of actomyosin crossbridges in relaxed and rigor muscle fibers. Biophys. J. 55: 441-453. ZOT, A. S. and POTTER,J. D. (1987) Structural aspects of troponin-tropomyosin regulation of skeletal muscle contraction. A. Rev. Biophys. biophys. Chem. 16: 535-539.