(The FASEB Journal. 1998;12:761-771.)
© 1998 FASEB
A new look at thin filament regulation in vertebrate skeletal muscle
John M. Squirea,1,
and Edward P. Morrisb
a Biophysics Section, Blackett Laboratory
b Biochemistry Department, Imperial College, London SW7 2BZ, U.K.
 |
ABSTRACT
|
|---|
It is 30 years since Ebashi and colleagues showed that Ca2+ ions directly affect regulation of the myosin–actin interaction in muscle through the action of tropomyosin and troponin on muscle thin filaments. It is more than 20 years since the idea was put forward that tropomyosin might act, at least in part, by changing its position on actin, thus uncovering or modifying the myosin binding site on actin when troponin molecules take up Ca2+. Since that time, a great deal of evidence for and against this steric blocking mechanism has been published: a structure for actin filaments at close to atomic resolution has been proposed, and the whole regulation story has become both more complicated and more subtle. Here we review structural and biochemical aspects of regulation in vertebrate skeletal muscle. We show that some basic ideas of the steric blocking mechanism remain valid. We also show that additional factors, such as troponin movements and structural changes within the actin monomers themselves, may be crucial. A number of the resulting regulation scenarios need to be distinguished.—Squire, J. M., Morris, E. P. A new look at thin filament regulation in vertebrate skeletal muscle. FASEB J. 12, 761–771 (1998)
Key Words: actin • tropomyosin • electron microscopy • troponin-T • TnC
|
BACKGROUND AND EARLY IDEAS
|
|---|
Actin filament structure
It is an everyday experience for most of us to be able to move our legs, arms, fingers, toes, and so on, when we wish to. Our skeletal muscles can respond directly to conscious instructions from the brain. Apart from the fundamental question as to how muscular force is actually generated, much of present-day research into the mechanisms of muscular contraction focuses on the related question: How is this force production switched on and off? Although, as it turns out, the answers to both questions are intimately connected, this review primarily addresses the nature of the switch. It has been known for many years that the effect of the arrival of a nervous stimulus at the membrane of a skeletal muscle fiber (a multinucleate cell) is to trigger the release of Ca2+ ions into the interior of the cell. The site of Ca2+ binding was localized in the 1960s by Ebashi and colleagues (1) as being on an actin filament protein, which they termed troponin. Troponin was found to be associated in a one-to-one stoichiometry with another actin binding protein, tropomyosin. Since then it has been shown that tropomyosin molecules are two-chain, 
-helical coiled-coil proteins about 400 Å long (
Fig. 1b;
ref 5) that link end-to-end to form continuous strands along actin filaments (6). Troponin, better referred to as the troponin complex, comprises three separate chains (see review by Tobacman, ref 7): troponin-C, which reversibly binds Ca2+ ions and is the classic E-F hand calcium binding protein; troponin-I, which on its own can inhibit muscular activity; and troponin-T, which binds to I and C and also strongly to tropomyosin.
View larger version (43K):
[in this window]
[in a new window]
|
Figure 1. a) Ribbon diagram (i.e., polypeptide backbone only) of the actin monomer structure as determined by Kabsch et al. (2). Note the existence of discrete subdomains: 1 (red), 2 (green), 3 (blue), and 4 (yellow). b) Ribbon diagram of the actin monomer in panel a placed into a 13/6 helix by Holmes et al. (3) to model the X-ray fiber diffraction data from gels of oriented actin filaments. c) Ribbon diagram of the two-chain, coiled-coil -helical structure of tropomyosin. d) Schematic illustration of the troponin complex, drawn to the same scale as panels a–c, shows troponin-C (red) as a ribbon diagram (4) and troponin-T as the green rod and part of the blue oval, which also contains troponin-I.
|
|
Since the pioneering work of Hanson and Lowy (6), which revealed the basic geometry of actin filaments by the then novel negative staining technique, much subsequent electron microscopy and X-ray diffraction work has characterized the actin filament structure. The filament approximates to a left-handed genetic helix of actin monomers with 13 monomers in 6 turns (a 13/6 helix), each of a pitch of about 59 Å (
Fig. 1b; Fig. 2).
The axial separation of successive monomers along the helix is about 27.5 Å. This structure appears as two slowly twisting strands of actin monomers, with a crossover repeat of the strands equal to about 360 to 380 Å (ref 6; Fig. 2). There are grooves between the two so-called long-period strands; tropomyosin molecules lie along these grooves such that each tropomyosin molecule interacts with seven actin monomers, successive monomers being two apart along the genetic helix ( Fig. 2). There is one troponin complex associated with each tropomyosin molecule.
View larger version (80K):
[in this window]
[in a new window]
|
Figure 2. Two interleaved stereo views of the Holmes et al. (3) actin filament, with the actin subdomains represented as spheres (subdomain 1, dark red), together with the coiled-coil tropomyosin strands (green). These two figures represent the `blocked' (a) and `unblocked' states (b) in the original steric blocking model (8– 10). Note that correct stereo is achieved with a conventional 6-cm spaced stereo viewer: the left and right views of panel a should merge as should the two views of panel b. Having the two images adjacent to each other allows direct structural comparisons.
|
|
The structure of the actin monomer ( Fig. 1a) was determined at atomic resolution by Kabsch et al. (2; see also refs 11, 12), and this structure was positioned by Holmes et al. (3) into the actin filament helix by modeling against fiber X-ray diffraction patterns from aligned gels of actin filaments (13, 14). The resulting structure is shown in Fig. 1b. Each actin monomer in Fig. 1a comprises four subdomains. The Holmes et al. (3) structure positions subdomains 3 and 4 close to the helix axis, where they interact with subdomains 3 and 4 of other actin monomers. Subdomains 1 and 2 are on the outside of the helix. Subdomain 1 contains the amino and carboxyl termini of the molecule. Troponin bound to tropomyosin marks a 385 Å repeat (7x2x27.5=385 Å) along the actin filaments (shown later in Fig. 4). Because this repeat is slightly larger than the crossover repeat of many actin filaments, especially those in vertebrate skeletal muscles, successive pairs of troponin complexes on the 385 Å repeat gradually rotate around the filament axis on a very slow helix.
View larger version (79K):
[in this window]
[in a new window]
|
Figure 4. Interwoven stereo views (a, b) of actin filament models (as in Fig. 2) of an alternative scheme to explain part of the changes observed in X-ray diffraction patterns from relaxed (a) and active muscle (b) (e.g., changes of the second actin layer-line at about 180 Å). In this case, the movement is of troponin (blue) rather than tropomyosin, which remains stationary (J. M. Squire, unpublished results). Note that this is just one example of a possible troponin movement that changes the second actin layer-line. No attempt has been made to define the particular troponin movements that may be involved in regulation.
|
|
The steric blocking mechanism
The working hypothesis on which most muscle researchers base their studies is that muscular force is produced when the projecting myosin heads on the myosin-containing filaments of muscle interact with the adjacent actin filaments. The nature of the interaction is postulated to involve the active tilting or bending of myosin heads on actin, which would tend to move the actin filaments past the myosin filaments (15–17). The whole process is known to be powered by ATP; myosin heads [known as myosin subfragment 1 (S1) when proteolytically cleaved from the rest of the myosin molecule] are ATPases, and the myosin ATPase is activated by actin. It is assumed that the large free energy change associated with hydrolysis of ATP to ADP and inorganic phosphate is somehow linked to the working stroke, the putative `swinging' movement, of all or part of a myosin head on actin (18). The `swinging (or bending) crossbridge model' of muscle contraction is probably the most favored model for how force is generated. Although it has proved to be very difficult to test, a great deal of evidence has recently accumulated in support of this idea (16, 17).
Building on the ideas of the swinging crossbridge model, an early puzzle concerned how troponin, which binds only to actin monomers spaced 385 Å apart axially along the actin filament, could affect the binding of myosin crossbridges to each of the seven actins in that distance. It was apparent that tropomyosin must be implicated in some way; but how? An early idea, based on pioneering X-ray diffraction studies of muscles in various states (19, 20), was that the position of tropomyosin molecules might depend on whether or not Ca2+ ions were bound to troponin (8–10). This idea was helped by early results concerning the location of the site on actin to which the myosin heads might bind. These results were obtained from 3-dimensional reconstructions of electron micrograph images of S1-labeled actin filaments by Moore et al. (21). It is now known that this groundbreaking early 3-dimensional reconstruction work did not have enough detail to locate S1 properly on actin. More recent work has benefited both from the determination of the structure of myosin S1 by protein crystallography (22) and the positioning of this structure within improved 3-dimensional reconstructions of electron micrographs of S1-decorated actin filaments visualized in frozen-hydrated preparations (23–27). From these and other studies (see ref 16), it is known that myosin S1 heads bind mainly to subdomain 1 of actin. It is also known to which side of the groove the heads bind: it is very close to where tropomyosin binds to actin.
Early X-ray diffraction evidence (19, 20) interpreted in terms of changes in the position of tropomyosin concerned the so-called second layer-line of the actin filament fiber X-ray diffraction pattern (8–10, 28, 29). This layer-line was very weak in diffraction patterns from resting muscle, but became much stronger when the muscle was activated or put into rigor. From modeling studies, it was soon found that the second layer-line intensity is very sensitive to the position of tropomyosin (8–10). The intensity changes observed could be explained if the tropomyosin molecules were well out of the groove in the relaxed (low Ca2+) state of the muscle ( Fig. 2a), but much closer to the middle of the groove in the active (high Ca2+) state or rigor ( Fig. 2b). The suggestion was then made that in the relaxed state, tropomyosin was perhaps close enough to the myosin binding site either to physically block attachment or at least modify the actin structure in such a way that attachment was inhibited. Binding of Ca2+ to troponin would then shift the tropomyosin out of the way, to its Ca2+-activated state, so that myosin attachment and force generation could proceed. This idea was soon termed the `steric blocking model', implying that tropomyosin regulates activity by virtue of its position on the thin filaments. Some of the apparently conflicting biochemical results (30) could be accounted for in a reasonable fashion by slight modifications of the original steric blocking ideas (28, 31, 32).
In the original modeling studies in the early 1970s, the actin monomer structure was unknown and each monomer was represented simply as a sphere of appropriate relative mass. Such spheres were located with 13/6 helical symmetry to represent an actin filament. The only other thin filament component in the computations was tropomyosin; therefore, the only structural changes that could be investigated, since the helical symmetry of the filament did not change, were changes in shape and/or position on actin of the tropomyosin strands. The only suitable movement then turned out to be a lateral shift of tropomyosin across the face of the spherical actin monomers, thus exposing what was thought to be the myosin binding site on actin: the steric blocking model was born. When the structure of the actin monomer was determined by X-ray crystallography (see ref 2; Fig. 1a) and located in the actin filament (see ref 3 Holmes et al, 1990; Fig. 1b), the original computations on which the steric blocking model was based were seen to be unreliable (33). Apart from tropomyosin movements, it was equally possible that relative movements of the subdomains of the actin monomer could also contribute to intensity changes in the second layer-line observed from the actin filaments. Indeed, some parts of the actin diffraction pattern to which tropomyosin was thought not to contribute were also observed to change depending on the presence or absence of Ca2+ ions (e.g., refs 34–36). This was true even in muscles stretched to sarcomere lengths, where the myosin and actin filaments do not overlap and myosin head interaction with actin is precluded (37–39). This in itself suggests that structural changes within the actin monomers may occur. For this reason alone, it was necessary to reevaluate the steric blocking model. In addition, many biochemical studies of the thin filament regulatory system have suggested that the simple `steric blocking' story is inadequate. In the next sections we discuss the cooperative nature of actin filament regulation and outline recent structural evidence on tropomyosin position and actin monomer structure in various states; we also discuss the possible structural role of troponin and try to put these ideas into plausible working schemes for thin filament regulation.
|
STRUCTURE AND INTERACTIONS OF TROPONIN
|
|---|
The troponin complex ( Fig. 1d) consists of three polypeptide chains or components: 1) troponin-C (TnC) of Mr ~18 kDa, which contains the regulatory Ca2+ binding sites; 2) troponin-I (TnI) of Mr ~21 kDa, which is capable of inhibiting the ability of actin filaments to activate myosin ATPase and thereby produce force and movement; and 3) troponin-T (TnT) of Mr ~30.5 kDa, which binds strongly to tropomyosin and probably is largely responsible for ensuring the attachment of the troponin complex to the thin filament with the correct geometry and stochiometry (see refs 7, 40, 41).
|
TROPONIN-C (TnC)
|
|---|
Muscle contraction is controlled by modulation of free [Ca2+] within the cytoplasm. TnC is the myofibrillar protein that, by binding Ca2+, transmits the signal to the thin filament. The structure of TnC has been solved to atomic resolution by X-ray crystallography (4) and is illustrated in Fig. 1d. It consists of two globular domains corresponding to the amino and carboxyl termini, and these are linked by a central, nine-turn -helix. Each globular domain contains two divalent metal binding sites formed from the helix-loop-helix EF hand motif. The two metal binding sites in the amino-terminal domain are Ca2+ specific and are believed to be the regulatory sites (i.e., Ca2+ binding to these sites is believed to trigger activation). Conversely, two metal binding sites in the carboxyl terminus bind both Ca2+ and Mg2+, and it seems likely that these sites remain occupied by Ca2+ or Mg2+ throughout the contraction cycle. TnC interacts both with TnI and TnT. TnC interacts more strongly with TnI when Ca2+ is not bound to the TnC regulatory sites; modulation of this interaction is believed to be the regulatory mechanism within the troponin complex.
|
TROPONIN-I (TnI)
|
|---|
Troponin and tropomyosin exert their regulatory effect in the thin filament by causing inhibition when TnC has no Ca2+ bound to the regulatory (amino-terminal) sites ( Fig. 3;
ref 7). The actin filament without its regulatory proteins is intrinsically active; that is, it activates myosin ATPase and supports force production and movement. TnI has a major role in inhibiting this activity. In fact, on its own, TnI has an effect on actin filament activity similar to the whole troponin complex at low [Ca2+]. The detailed structure of TnI is unknown. Models have been generated based on the low to medium resolution structure and antibody labeling studies of the troponin complex (43–45) and on neutron scattering measurements of the TnC–TnI complex (46). The latter indicate that the TnI in the TnI–TnC complex is somewhat more elongated than TnC itself, with a maximum chord of 115 Å. Besides its interaction with TnC, TnI appears to interact with TnT and actin. The interaction with actin is particularly important: it is thought that it is through this link that TnI exerts its regulatory effect and that this interaction is modulated by Ca2+ binding to TnC.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 3. The dual effect of the regulatory proteins tropomyosin and troponin on the actin-activated myosin S1 ATPase. Measurements of ATPase activation by F-actin without regulatory proteins show a linear increase as a function of added S1 in this concentration range (open circles). However, when the regulatory proteins are present, either consisting of tropomyosin alone (solid diamonds), tropomyosin plus troponin plus Ca2+ (open inverted triangles), or tropomyosin plus troponin without Ca2+ (open triangles), graphs of ATPase as a function of [S1] curve upward sharply with increasing S1 concentration. Thus, at lower S1 concentrations (1 µM or less), thin filaments are inhibited in their ability to activate S1 ATPase compared with F-actin whether or not Ca2+ is bound to troponin. Conversely, substantial activity is obtained at higher S1 concentrations even without Ca2+ bound to troponin. These observations demonstrate the central role of bound heads (S1) in determining the state of the thin filament. Redrawn from Lehrer and Morris, 1982 (42).
|
|
|
TROPONIN-T (TnT)
|
|---|
TnT is involved in the attachment of the troponin complex to tropomyosin and thus to actin. It is a rodlike molecule about 185 Å long, with amino and carboxyl termini at opposite ends. The amino-terminal region binds to the head–tail junction of tropomyosin; the carboxyl-terminal region is closely associated with TnC and TnI, which are located about 200 Å away from the head–tail junction near residues 150–180 of tropomyosin in the tropomyosin–troponin complex. TnT can be separated proteolytically into two fragments, TnT1 and TnT2, the relative positions of which have been defined by antibody labeling (43) and in troponin T/tropomyosin crystals (47). TnT2 (Mr 13 kDa) contains the carboxyl-terminal TnT region, which binds to TnC and TnI and is probably part of the globular end of whole troponin. TnT1 (Mr 26 kDa) forms most of a fingerlike extension from the bulk of the troponin structure ( Fig. 1d; ref 48).
|
THIN FILAMENT COOPERATIVITY AND TRANSIENT KINETIC STUDIES PROVIDE EVIDENCE FOR AS MANY AS THREE BIOCHEMICAL STATES
|
|---|
The thin filament shows cooperativity in its interaction with myosin or myosin derivatives (e.g., S1). This is observed ( Fig. 3) both in the binding of myosin to the thin filament and in the actin-activated ATPase, each of which show upward curving or sigmoidal profiles when plotted as a function of myosin concentration (42, 49). The presence of tropomyosin bound to actin is a minimum requirement for this phenomenon. F-actin filaments do not show this type of cooperativity, whereas it is seen both in F-actin + tropomyosin (A+TM) and F-actin + tropomyosin + troponin (A+TM+TN). As well as the transient kinetic studies by McKillop and Geeves (50), cooperative models of myosin binding and activation have been developed to explain this behavior in which there are three states of the thin filament. These are STATE 1: the `off' state (termed `blocked' by Mckillop and Geeves; 50), corresponding to a situation in which a myosin collision complex can form (i.e., nonstereospecific weak binding of myosin to actin), but there is low ATPase activity; STATE 2: the Ca2+-activated state (the `closed' state of McKillop and Geeves; ref 50) in which stereospecific weak binding by myosin can occur, but the ATPase is still low; and STATE 3: the fully `on' state [the `open' state of McKillop and Geeves (50)] corresponding to strong stereospecific myosin binding, high ATPase activity, and high force generation.
As noted by the authors at the time, these three distinguishable biochemical states may correspond to only two different positions of the tropomyosin/troponin complex. We expand on this possibility later.
The cooperative unit seems likely to consist of at least seven actin subunits under the control of a single tropomyosin molecule together with a troponin molecule, if present. Even then there is some evidence that the seven-actin unit may be too small to explain all the data. For this reason, communication between successive seven actin units along the filament has been proposed (51–54).
An important feature of this type of model is that, even in the presence of Ca2+ ions, the thin filaments are predominantly in the `blocked' or `closed' states where the acto-myosin ATPase is low. Transition to the situation where the fully `on' or `open' state predominates requires the presence of, and interaction with, myosin. Binding of Ca2+ to troponin influences the distribution between the `blocked', `closed', and `open' states, increasing the proportion of regulatory units in the `closed' or Ca2+-activated state, but this still does not switch the thin filament to the fully `on' state; this requires interaction with a significant number of myosin heads. In this way, the switching effect of the Ca2+/troponin interaction can be compared to the release of a safety catch on a gun, which makes it possible for the gun to be fired, but firing will actually occur only when the trigger is pulled. Because binding of S1 may be likened to pulling the trigger, we prefer to use the terminology `blocked', `cocked', and `open' for the three states described above, where `cocked' replaces the `closed' terminal of McKillop and Geeves (50). The next section evaluates the molecular movements involved in this triggering mechanism.
|
TROPOMYOSIN POSITION IN DEFINED STATES
|
|---|
Armed with the thin filament model of Holmes et al. (3) and with knowledge of the myosin S1 structure (22), one can test with much greater rigor than was previously possible the original modeling on which the `steric blocking' mechanism of regulation was based ( Fig. 2a). The four-domain structure for G-actin ( Fig. 1a) immediately opens up the possibility that some of the observed X-ray intensity changes of the low-angle actin layer-lines could be due to subdomain movements within each actin monomer. The X-ray changes to be explained from muscles that have been stretched to non-overlap sarcomere lengths, and which report therefore on the `blocked' and `cocked' states only, can easily be summarized (38, 39; data from N. Yagi reported in ref 55). Upon Ca2+-activation, 1) the intensity of the 59 Å layer-line from actin increases; 2) the intensity of the second actin layer-line (at an axial spacing corresponding to 180 Å) increases substantially; whereas 3) the intensity of the first actin layer-line (360 Å) is reduced.
In principle, the intensities on layer-lines 1 and 2 could be affected either by actin subdomain movements, by a shift in tropomyosin position (as proposed before), or (and this is a possibility usually not discussed by structuralists) by a movement of troponin molecules. Of course there may be a combination of all three. Since large intensity changes are observed, it is clear that large structural changes are taking place in the thin filament. It is important to establish as unambiguously as possible the structural events involved in regulation, something that, in principle, can be determined in an objective way. These structural changes should then be correlated with available biochemical and kinetic evidence on regulation.
It seems quite unlikely that the gross structural changes seen are going to explain all aspects of regulation. However, it seems just as unlikely that the structural changes evident from the low-angle X-ray diffraction patterns from Ca2+-activated muscles are not somehow directly involved in the regulation story, even though some biochemists appear to question this. In summary, the observed X-ray data need to be modeled using the known structures of actin, tropomyosin, and troponin to find out what the possible molecular movements might be. The functional implications of these movements can then be assessed. At the same time, a crucial complementary approach to this same structural problem is to try to trap isolated actin filaments in defined states (e.g., Ca2+ free and Ca2+ activated) and to determine their structure by electron microscopy and 3-dimensional reconstruction. Great advances in this direction have, in fact, been achieved recently (56, 57). Fortunately, both the X-ray and electron microscopy approaches seem to be reporting similar observations. However, as described below, what these observations mean is not yet clear.
|
MODELING TROPOMYOSIN POSITION
|
|---|
A starting point in modeling the X-ray diffraction observations described above is to take the Holmes et al. (3) F-actin structure, add to it a model of the tropomyosin -helical coiled-coil strands ( Fig. 1c), and then change the tropomyosin position in a systematic way, each time computing the theoretical diffraction pattern. One such structure was determined by Lorenz et al. (58) for F-actin with tropomyosin but no troponin. They found that a reasonable fit to the high-angle fiber diffraction pattern from oriented gels was obtained with tropomyosin located in a position that was also favorable for electrostatic interactions between tropomyosin and surface residues on subdomain 3 of actin. A related, low-angle X-ray diffraction study by Al-Khayat et al. (55; see also ref 59) used the data obtained by Yagi and Matsubara (ref 38, described earlier) from frog muscles stretched to non-overlap. In this case, the thin filaments carried troponin as well as tropomyosin, but no myosin head interactions were involved. It was not possible to explain all of the observations in this case by moving tropomyosin on the surface of an unchanging (Holmes et al.) actin structure. Another study was therefore carried out in which the actin subdomains were also permitted to move by small amounts around the Holmes et al. (3) positions. Suggestive evidence for such subdomain movements comes from the observed changes of the 59 Å (sixth) layer-line to which the tropomyosin strands appear to make little contribution; they have their main effect on the first few layer-lines, especially 1, 2, and 3 (10, 28). The more general search found that reasonable agreement with the Yagi and Matsubara (38) data could be obtained with structures slightly different from the Holmes et al. structure (3). Crucial features were that: 1) the most satisfactory models all included a movement of the tropomyosin strands, 2) some small movement of subdomain 1 between the off and Ca2+-activated state seemed likely, 3) the positions of subdomains 3 and 4 remained the same (sensible since they appear to define the helical symmetry, which does not appear to change), and 4) the position of subdomain 2 was very hard to define, though quite large movements of this subdomain seemed possible. Evidence for the movement of subdomain 2 has come from other independent studies (e.g., refs 60–65).
After the early electron microscopy studies of actin labeled with myosin S1 (21) and subsequent improvements in technique (24), the location of myosin heads within 3-dimensional reconstructions of decorated thin filaments has become quite reliable. However, for many years it proved difficult to define the position of tropomyosin in the Ca2+-free `blocked' and Ca2+-activated `cocked' states. It has now been shown that there are distinct structural changes induced by Ca2+ binding to thin filaments in Limulus (56) and also in vertebrates (57). The observed changes are consistent with X-ray diffraction modeling (55); the sixth (59 Å) layer-line also showed intensity changes, possibly related to changes in the actin monomers as discussed above. These studies all show that, in the Ca2+-free state, what is taken to be tropomyosin is close to the myosin binding site on actin subdomain 1, actually in a kind of cleft between subdomains 1 and 3 ( Fig. 2). In the Ca2+-activated state, this mass has shifted across the face of subdomain 3 by about 10 to 20 Å. The apparent position of tropomyosin with Ca2+ ions bound to troponin is still close enough to the attachment position of myosin S1 to be able to modify the S1/actin interaction even after myosin head attachment to actin has occurred. It is also possible that further movement of tropomyosin is actually a result of myosin head binding (10, 24, 31, 66, 67).
|
CONTRIBUTION OF TROPONIN TO X-RAY DIFFRACTION DATA
|
|---|
Before trying to piece together the structural and biochemical aspects of thin filament regulation, it would be well to consider an aspect of thin filament structure that is often ignored: the possible structural role of troponin and its contribution to the interpretation of X-ray diffraction data and electron micrograph data discussed in the last section. To put it simply, Is it possible that movements of troponin could explain some or all of the X-ray and electron micrograph observations? Such a possibility has been neglected by structuralists, mainly because knowledge of the complete atomic arrangement in troponin is not yet available. However, as described earlier, we do know the structure of troponin-C (4); there are some ideas from low-angle neutron scattering about the shape of troponin-I (46), and we know roughly the shape of the whole troponin complex (43–45). Therefore, it is possible to model the whole thin filament, as in Fig. 4a,
including a crude but correctly weighted model of troponin. With such a model, one can then move troponin to a new position ( Fig. 4b) and calculate the expected diffraction pattern. The detailed results from this modeling will be presented elsewhere (J. M. Squire, unpublished results), but such a troponin movement does substantially change the computed X-ray diffraction pattern. In summary, 1) Plausible movement of actin subdomains alone, with a fixed tropomyosin/troponin position, is not sufficient to explain the observed X-ray diffraction data (55). 2) Movement of tropomyosin alone on an unchanging Holmes actin is not sufficient to explain the observations (55). 3) It is possible to change the relative intensities of the low-angle actin layer-lines in the observed direction either just by tropomyosin movement (as modeled before) or (and this is the new result) merely by moving troponin while leaving the actin and tropomyosin structures fixed. Obviously, there may be a mixture of actin monomer changes and tropomyosin and troponin movement. 4) Unexpectedly, the movement of troponin in 3) does affect the intensity of the sixth (59 Å) layer-line, a change previously ascribed solely to actin subdomain movements. Actin subdomain movements probably do occur 68–70), but troponin position must now be taken into account as well.
In summary, without further detailed modeling of the X-ray diffraction observations, a number of alternative structural interpretations of the data are possible.
|
INTERPRETATION OF ELECTRON MICROSCOPY RESULTS
|
|---|
As shown above, because of the possible contribution of troponin as well as tropomyosin, the X-ray diffraction evidence is ambiguous. What, then, about the electron microscopy results of Lehman et al. (56, 57)? Unfortunately, these also appear to be ambiguous. The actin filaments they studied also contained troponin, but the 13/6 helical symmetry used in their 3-dimensional reconstructions did not account for the different troponin axial repeat of 385 Å. In the reconstructions, the troponin mass, about the same as the tropomyosin mass, was effectively `averaged' over a structure with 13/6 helical symmetry. What is seen in the reconstructions of the Ca2+-free `blocked' and Ca2+-activated `cocked' filaments is a movement of the center of mass of `something' on the surface of actin. The `something' is an average of the tropomyosin and troponin masses; it cannot be ascertained from current evidence whether this mass shift is solely a tropomyosin movement, solely a troponin movement, or a bit of both. The published electron microscopy work has the same ambiguity problem as the X-ray diffraction work.
Since the tropomyosin position obtained from troponin-free thin filaments (69, 70) seems to be similar to the tropomyosin `off' position, with troponin present in the Lehman et al. (56, 57) reconstructions, it could be argued that if troponin mass contributes to the `off' state, it must be centered in a similar azimuthal position to tropomyosin or it might be disordered and not show up. However, even in this case, early activation effects on the second actin layer-line could be due to troponin and/or tropomyosin movement, as could the effects of myosin binding in strong states. These possibilities are summarized in the next section.
|
ALTERNATIVE SCHEMES TO BE TESTED
|
|---|
The simple three-state scheme
When considering possible regulation scenarios, much has already been written and different models have been analyzed (7–10, 31, 40, 50, 51, 54, 71–76). The discussion here builds on all of these ideas.
The regulation scheme that seems to form the basis of many people's thinking at the present time is as follows. 1) In the Ca2+-free `blocked' state, troponin binds strongly to actin, holding tropomyosin in a blocking position on actin subdomain 1. A collision complex (i.e., a nonstereospecific interaction) between actin and myosin may occur, but nothing else. 2) Binding of Ca2+ to troponin-C releases troponin from actin and allows a movement of tropomyosin. This is the `blocked' to `closed' (cocked) transition of McKillop and Geeves (50) and Head et al. (76). It renders the stereospecific weak binding of heads to actin more likely, but still inhibits strong binding. 3) The gradual transition of some heads to strong binding pushes tropomyosin further across the thin filament and switches the filament more fully on, so that maximal head cycling and increased ATPase activity are observed. The more attached heads there are, the greater the ATPase.
This conventional story, which we term the simple three-state scheme ( Fig. 5),
has tropomyosin located in three different, presumably stable, positions on actin. The central position is that occupied by tropomyosin in the absence of troponin (A+TM) or is the position of A + TM + TN + Ca2+, which is the `cocked' state. With troponin in the absence of Ca2+ (A+TM+TN), tropomyosin is locked further over toward actin subdomain 1 by troponin binding to actin. On the other hand, A + TM + TN + Ca2+ plus significant myosin binding in strong states pushes the tropomyosin farther away from the main myosin binding site on actin.
View larger version (25K):
[in this window]
[in a new window]
|
Figure 5. Schematic illustrations of different possible thin filament regulation schemes as discussed in detail in the text. In scheme 1, the tropomyosin has three distinct positions on the actin filament. In scheme 2, the tropomyosin only has two defined states, `blocked' and `open', but the equilibrium between them varies according to conditions. Without head attachment, the effect of Ca2+ binding to troponin is to shift the equilibrium from almost purely blocked to a mixture (rapid equilibrium) of blocked and open states. The addition of heads shifts the equilibrium to almost fully open. In scheme 3, a major effect of Ca2+ binding is to move troponin, possibly with some small tropomyosin movement as well, thus freeing tropomyosin to move (it is `cocked'), although it takes head binding to move tropomyosin substantially to the open position. Red circles, actin monomers; green rods, tropomyosin; blue shapes, troponin; yellow shapes, myosin heads (S1).
|
|
The dynamic two-state scheme
Another possibility, however, and still based only on the tropomyosin position, is that tropomyosin has only two stable positions, in equilibrium with each other, and that under different conditions the equilibrium is shifted. The two extremes are essentially the `blocked' and `open' states of the thin filament. In the troponin-free state studied by Lorenz et al. (58), there would be a dynamic equilibrium with roughly equal populations in the `blocked' and `open' states. What would be seen structurally is an `average' state halfway between the two extremes. The effect of troponin without bound Ca2+ ions would then be to physically hold tropomyosin out of this equilibrium position, presumably through the TnI–actin interaction, thus shifting the `average' tropomyosin location toward the `blocked' position, as observed by Lehman et al. (56, 57) and Al-Khayat et al. (55). The effect of Ca2+ could simply be to alter the equilibrium between the `blocked' and `open' positions so that something like the halfway, troponin-free `average' position is restored. This average position would be what was seen by Lehman et al. (56, 57) and Al-Khayat et al. (55) as the Ca2+-activated position; the tropomyosin would still partially block attachment. In the presence of Ca2+, only the addition of strongly attached heads would restrain the tropomyosin to be on average mostly in the `open' position, thus allowing other heads to cycle and the full ATPase rate to be observed. Note that in this scheme tropomyosin spends significant time only in the `blocked' and `open' states; the `cocked' state is just a mixture of these two positions. We call this possibility the dynamic two-state scheme.
Modified three-state scheme
If allowance is made for the involvement of troponin, then other kinds of model can be envisaged. We call one of these the modified three-state scheme. Here the effect of Ca2+ activation is to release the binding of troponin to actin so that tropomyosin is no longer tied down in a `closed' position. Some, if not all, of the mass movement involved in this is now a troponin movement. Either a small tropomyosin movement or no tropomyosin movement at all accompanies this `blocked' to `cocked' transition. Although the tropomyosin may not have moved, however, it is now free to move. In our parlance, the safety catch of our gun has been unlocked; the system is `cocked' but it has not yet fired. The tropomyosin equilibrium with an `open' position would shift some way toward `open' (from almost totally `blocked' in the absence of Ca2+). The subsequent addition of heads in strong binding states would then, as before, shift the equilibrium fully over to the `on' position (the trigger has now been pulled).
In this scenario, which has its attractions and seems quite plausible, the early change of the actin second layer-line during the rising phase of a tetanus observed at no-overlap by Kress et al. (77) could be associated mainly with a movement of troponin, not tropomyosin, and only additional changes at full overlap—those associated with strong myosin head attachment—would be associated with tropomyosin movement. Note, however, that attaching heads will themselves affect the second layer-line intensity and that the actin monomer structure will inevitably change with such gross movements of troponin, tropomyosin, and the heads to which it is attached. Evidence for structural changes in troponin on Ca2+ binding has come from other sources (e.g., refs 78–81, and others).
|
CONCLUSIONS
|
|---|
With these alternatives in mind [and there may be more, all of which appear to be consistent with the ideas of McKillop and Geeves (50), Lehrer (75), Head et al. (76), and others], it is appropriate to ask if evidence is available or whether tests can be suggested that might choose between them.
As far as structural experiments are concerned, some key questions come to mind. 1) Is it possible to model the Ca2+-bound state in non-overlap muscle in terms of a mixed population of two distinct tropomyosin positions, as in the dynamic two-state scheme? 2) Is it also possible to account for the observed mass shift in 3-dimensional reconstructions of A + TM + TN in the presence and absence of Ca2+ in terms of the dynamic two-state scheme? 3) Is it possible with the electron microscope or X-ray data to distinguish between the contributions of actin, tropomyosin, and troponin to the observed mass movements? 4) Is it possible to show that the tropomyosin position in the TN-free state (58) and the Ca2+-activated, myosin-free states are really the same? 5) Is it possible to define tropomyosin position in actin filaments labeled with myosin heads (i.e., as in the rigor, nucleotide-free state)? Is this position different from the Ca2+-activated position?
There are other structural questions to pose, but these are some of the most significant ones. The postulation of related biochemical questions we leave to others. The steric blocking mechanism lives on and is evolving, but with new techniques, structural studies should be able to define the mechanism much more thoroughly than has been possible so far. The story has reached a fascinating stage.
|
ACKNOWLEDGMENTS
|
|---|
We are indebted to Dr. Gerald Offer for help with preparation of Fig. 1c and to Dr. Jeff Harford for help with Fig. 2 and Fig. 4.
|
FOOTNOTES
|
|---|
1 Correspondence: Correspondence to John M. Squire 
|
REFERENCES
|
|---|
-
Ebashi, S., Endo, M., and Ohtsuki, I. (1969) Control of muscle contraction. Q. Rev. Biophys. 2, 351–384[Medline]
-
Kabsch, W., Mannherz, H. G., Suck, D., Pai, E. F., and Holmes, K. C. (1990) Atomic structure of the actin: DNaseI complex. Nature (London) 347, 37–44[Medline]
-
Holmes, K. C., Popp, D., Gebhard, W., and Kabsch, W. (1990) Atomic model of the actin filament. Nature (London) 347, 44–49[Medline]
-
Herzberg, O., and James, M. N. G. (1985) Structure of the calcium regulatory muscle protein troponin-C at 2.8 Å resolution. Nature (London) 313, 653–659[Medline]
-
Phillips, G. N., Fillers, J. P., and Cohen, C. (1986) Tropomyosin crystal structure and muscle regulation. J. Mol. Biol. 192, 111–131[Medline]
-
Hanson, E. J., and Lowy, J. (1963) The structure of F-actin and of actin filaments isolated from muscle. J. Mol. Biol. 6, 48–60
-
Tobacman, L. S. (1996) Thin filament-mediated regulation of cardiac contraction. Annu. Rev. Physiol. 58, 447–481[Medline]
-
Huxley, H. E. (1972) Structural changes in the actin- and myosin-containing filaments during contraction. Cold Spring Harbor Symp. Quant. Biol. 37, 361–376
-
Haselgrove, J. C. (1972) X-ray evidence for a conformational change in the actin-containing filaments of vertebrate striated muscle. Cold Spring Harbor Symp. Quant. Biol. 37, 341–352
-
Parry, D. A. D., and Squire, J. M. (1973) Structural role of tropomyosin in muscle regulation: analysis of the X-ray diffraction patterns for relaxed and contracting muscles. J. Mol. Biol. 75, 33–55[Medline]
-
Schutt, C. E., Myslik, J. C, Rozycki, M. D., Goonesekere, N. C., and Lindberg, U. (1993) The structure of crystalline profilin-ß-actin. Nature (London) 365, 810–816[Medline]
-
McLaughlin, P. J., Gooch, J. T., Mannherz, H.-G., and Weeds, A. G. (1993) Structure of gelsolin segment 1-actin complex and the mechanism of filament severing. Nature (London) 364, 685–692[Medline]
-
Popp, D., Lednev, V. V., and Jahn, W. (1987) Methods of preparing well-oriented gels of F-actin containing filaments suitable for X-ray diffraction. J. Mol. Biol. 197, 679–684[Medline]
-
Popp, D., Maeda, Y., Stewart, A. A. E., and Holmes, K. C. (1991) X-ray diffraction studies on muscle regulation. Adv. Biophys. 27, 89–103[Medline]
-
Huxley, H. E. (1969) The mechanism of muscular contraction—recent structural studies suggest a revealing model for cross-bridge action at variable filament spacing. Science 164, 1356–1366[Free Full Text]
-
Holmes, K. C. (1996) Muscle proteins: their actions and interactions. Curr. Opin. Struct. Biol. 6, 781–789[Medline]
-
Squire, J. M. (1997) Architecture and function in the muscle sarcomere. Curr. Opin. Struct. Biol. 7, 247–257[Medline]
-
Lymn, R. W., and Taylor, E. W. (1971) Mechanism of adenosine triphosphate hydrolysis by actomyosin. Biochemistry 10, 4617–4624[Medline]
-
Huxley, H. E. (1971) Structural changes during muscle contraction. Biochem. J. 125, 85P (Abstr.)
-
Vibert, P. J., Haselgrove, J. C., Lowy, J., and Poulsen, F. R. (1972) Structural changes in actin-containing filaments of muscle. J. Mol. Biol. 71, 757–767[Medline]
-
Moore, P., Huxley, H. E., and DeRosier, D. R. (1970) Three-dimensional reconstruction of F-actin, thin filaments and decorated thin filaments. J. Mol. Biol. 50, 279–295[Medline]
-
Rayment, I., Rypniewski, W. R., Schmidt-Base, K., Smith, R., Tomchick, D. R., Benning, M. M., Winkelmann, D. A., Wesenberg, G., and Holden, H. M. (1993) Three-dimensional structure of myosin subfragment-1: a molecular motor. Science 261, 50–58[Abstract/Free Full Text]
-
Taylor, K. A., and Amos, L. A. (1981) A new model for the geometry of the binding of myosin crossbridges to muscle thin filaments. J. Mol. Biol. 147, 297–324[Medline]
-
Milligan, R. A., Whittaker, M., and Safer, D. (1990) Molecular structure of F-actin and location of surface binding sites. Nature (London) 348, 217–221[Medline]
-
Flicker, P. F., Milligan, R. A., and Applegate, D. (1991) Cryo-electron microscopy of S1-decorated actin filaments. Adv. Biophys. 27, 185–196[Medline]
-
Rayment, I., Holden, H. M., Whittaker, M., Yohn, C. B., Lorenz, M., Holmes, K. C., and Milligan, R. A. (1993) Structure of actin-myosin complex and its implications for muscle contraction. Science 261, 58–65[Abstract/Free Full Text]
-
Mendelson, R. A., and Morris, E. P. (1997) The structure of the acto-myosin subfragment 1 complex: results of searches using data from electron microscopy and X-ray crystallography. Proc. Natl. Acad. Sci. USA 94, 8533–8538[Abstract/Free Full Text]
-
Squire, J. M. (1981) The Structural Basis of Muscular Contraction. Plenum Press, New York
-
Harford, J. J., and Squire J. M. (1997) Time resolved diffraction studies of muscle using synchrotron radiation. Rep. Prog. Phys. 60, 1723–1787
-
Weber, A., and Murray, J. M. (1973) Molecular control mechanisms in muscular contraction. Physiol. Rev. 53, 612–673[Free Full Text]
-
Squire, J. M. (1975) Muscle filament structure and muscle contraction. Annu. Rev. Biophys. Bioeng. 4, 137–163[Medline]
-
Squire, J. M. (1981) Muscle regulation: a decade of the steric blocking model. Nature (London) 291, 614–615[Medline]
-
Squire, J. M. (1994) The actomyosin interaction—shedding light on structural events: `Plus ça change, plus c'est la même chose'. J. Musc. Res. Cell. Motil. 15, 227–231[Medline]
-
Matsubara, I., Yagi, N., Miura, H., Ozeki, M., and Izumi, T. (1984) Intensification of the 5.9-nm actin layer line in contracting muscle. Nature (London) 312, 471–473[Medline]
-
Wakabayashi, K., Tanaka, H., Amemiya, Y., Fujishima, A., Kobayashi, T., Hamanaka, T., Sugi, H., and Mitsui, T. (1985) Time-resolved X-ray diffraction studies on the intensity changes of the 5.9 and 5.1 nm actin layer lines from frog skeletal muscle during an isometric tetanus using synchrotron radiation. Biophys. J. 47, 847–850[Abstract/Free Full Text]
-
Wakabayashi, K., Ueno, Y., Amemiya, Y., and Tanaka, H. (1988) Intensity changes of actin-based layer lines from frog skeletal muscles during an isometric contraction. In Molecular Mechanisms of Muscle Contraction (Sugi, H., and Pollack, G. H., eds) pp. 353–367, Plenum Press, New York
-
Yagi, N., and Matsubara, I. (1988) Changes in the 5.9 nm actin layer-line on activation of frog skeletal muscles. In Molecular Mechanism of Muscle Contraction (Sugi, H., and Pollack, G. H., eds) pp. 369–380, Plenum Press, New York
-
Yagi, N., and Matsubara, I. (1989) Structural changes in the thin filament during activation studied by X-ray diffraction of highly stretched skeletal muscle. J. Mol. Biol. 208, 359–363[Medline]
-
Yagi, N. (1991) Intensification of the first actin layer-line during contraction of frog skeletal muscle. Adv. Biophys. 27, 35–43[Medline]
-
Potter, J. D., and Gergely, J. (1974) Troponin, tropomyosin and actin interactions in the Ca2+ regulation of muscle contraction. Biochemistry 13, 2697–2703[Medline]
-
Zot, A. S., and Potter, J. D. (1987) Structural aspects of troponin-tropomyosin regulation of skeletal muscle contraction. Annu. Rev. Biophys. Biophys. Chem. 16, 535–559[Medline]
-
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[Abstract/Free Full Text]
-
Ohtsuki, I. (1979) Molecular arrangement of troponin-T in the thin filament. J. Biochem. 86, 491–497[Abstract/Free Full Text]
-
Flicker, P. F., Phillips, J. G. N., and Cohen, C. (1982) Troponin and its interactions with tropomyosin: an electron microscope study. J. Mol. Biol. 162, 495–501[Medline]
-
White, S. P., Cohen, C., and Phillips, J. G. N. (1987) Structure of co-crystals of tropomyosin and troponin. Nature (London) 325, 826–828[Medline]
-
Olah, G. A., and Trewella, J. (1994) A model structure of the muscle protein complex 4Ca2+-troponin C-troponin I derived from small-angle scattering data: implications for regulation. Biochemistry 33, 12800–12806[Medline]
-
Cabral-Lily, D., Tobacman, L. S., Mehegan, J. P., and Cohen, C. (1991) Polarity of troponin-T in tropomyosin/troponin T co-crystals. Biophys. J. 59, 581a
-
Heeley, D. H., Golosinka, K., and Smillie, L. B. (1987) The effects of troponin T fragments T1 and T2 on the binding on nonpolymerisable tropomyosin to F-actin in the presence and absence of troponin I and troponin C. J. Biol. Chem. 262, 9971–9978[Abstract/Free Full Text]
-
Greene, L. E., and Eisenberg, E. (1980) Cooperative binding of myosin subfragment-1 to the actin-troponin-tropomyosin complex. Proc. Natl. Acad. Sci. USA 77, 2616–2620[Abstract/Free Full Text]
-
McKillop, D. F. A., and Geeves, M. A. (1993) Regulation of the interaction between actin and myosin subfragment 1: evidence for three states of the thin filament. Biophys. J. 65, 693–701[Abstract/Free Full Text]
-
Hill, T. L., Eisenberg, E., and Greene, L. E. (1980) Theoretical model for cooperative equilibrium binding of myosin subfragment-1 to the actin-troponin-tropomyosin complex. Proc. Natl. Acad. Sci. USA 77, 3186–3190[Abstract/Free Full Text]
-
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[Abstract/Free Full Text]
-
Butters, C. A., Willasden, K. A., and Tobacman, L. S. (1993) Cooperative interactions between
TOP
ABSTRACT
BACKGROUND AND EARLY IDEAS
