Contractile Proteins

Myosin

Structure

Skeletal muscle contains 70 -100 mg of myosin per gram of fresh muscle weight; this corresponds to 40-50% of the total muscle proteins. Myosin is a globulin, soluble at high salt concentration, e.g. 0.6 M KCl, and insoluble at low salt concentration, e.g. 0.03 M KCl. Thus, myosin can be extracted from the muscle with 0.6 M KCl solution and purified by dilution of the extract with 20 volumes of distilled water. However, under these conditions, some actin remains bound to myosin and special procedures are required to prepare myosin free of actin.

Size and shape of the myosin molecule: Myosin is a large asymmetric molecule, it has a long tail and two globular heads (Fig. M1). The tail is about 1,600 Å long and 20 Å wide. Each head is about 165 Å long, 65 Å wide and 40 Å deep at its thickest part. The molecular weight of myosin is about 500,000. In strong denaturing solutions, such as 5 M guanidine-HCl or 8 M urea, myosin dissociates into six polypeptide chains: two heavy chains (molecular weight of each heavy chain about 200,000) and four light chains (two with a molecular weight of 20,000, one with 15,000 and another with 25,000). The two heavy chains are wound around each other to form a double helical structure. At one end both chains are folded into separate globular structures to form the two heads. In the muscle, the long tail portion forms the backbone of the thick filament and the heads protrude as crossbridges toward the thin filament. Each head contains two light chains.

Fig. M1. Scheme of the myosin molecule

Fig. M2 shows more details of the myosin structure. When myosin is exposed to the proteolytic enzyme trypsin, fragmentation occurs in the middle of the tail yielding heavy meromyosin (HMM, molecular weight about 350,000) and light meromyosin (LMM, molecular weight about 150,000) HMM containing the head and a short tail can be further split by proteolytic enzymes, such as papain, into subfragment 1 (S1, molecular weight about 110,000) and subfragment 2 (S2). The regions of proteolytic fragmentation may serve as hinges. HMM and S1 bind actin, hydrolyze ATP and are water-soluble. LMM has no sites for actin or ATP binding, but inherits the solubility of myosin in 0.6 M KCl and the self-assembling property of myosin in 0.03 M KCl. S2 is water-soluble. Myosin and its proteolytic fragments can be visualized by electron microscopy (Lowey et al., 1969.)

Fig. M2. Details of the myosin structure (Courtesy of Dr. Helen Rarick). Note the drawing of the myosin molecule is not to scale. ELC, essential light chain; RLC, regulatory light chain.

The heavy chain of S1 can still be further split by proteolytic enzymes into three fragments: 25 kDa (N terminal), 50 kDa (central) and 20 kDa (C terminal) fragments. These fragments played a role in the understanding of the three-dimensional structure of subfragment 1.

Myosin light chains: These were discovered by analytical ultracentrifugation; succinylated, or K2CO3 treated myosin exhibited a small slowly sedimenting component behind the main myosin peak. Subsequently, one-dimensional sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) of purified skeletal muscle myosin showed three light chain bands in addition to the heavy chain bands (Fig. M3). The light chain bands were called LC1, LC2, and LC3 with decreasing molecular weight. LC1 and LC3 can be isolated after alkali treatment of myosin, therefore, they are also called Alkali 1 and Alkali 2 light chains; they are also referred to as essential light chains. Alkali 1 and Alkali 2 are identical over their 142 amino acid residues at the C terminal. These isozymes arise from a single gene by alternative splicing. LC2 can be phosphorylated and is also referred to as regulatory light chain. Each myosin head contains 1 essential and 1 regulatory light chain.

 Fig. M3. SDS-PAGE of skeletal muscle myosin (the heavy chain of myosin on the top of the second column is not indicated).

Review: An encyclopedic review about myosin, containing 2,354 references, has been written by Sellers and Goodson (1995).

 Function

Myosin-actin binding: One of the biologically important properties of myosin is its ability to combine with actin. The complex formed is called actomyosin. The actin binding by myosin is highly specific; no other protein can substitute actin. Physiologically, when actin and myosin combine the muscle produces force.

There are several methods to measure the stoichiometry of actin to myosin combination. In solution, analytical ultracentrifugation is the most accurate method: When molecules in solution are subjected to high centrifugal field, they sediment according to their molecular weight. Runs in the analytical ultracentrifuge were performed in our laboratory with solutions containing the same amount of myosin (M) and different amount of fibrous actin. As Fig. M4 shows, without actin only 1 peak, the myosin peak, is seen in the analytical ultracentrifuge. As actin is added to myosin in increasing amount, part of the myosin combines with actin, forming the high molecular weight actomyosin (AM) and 2 peaks are seen. When all the myosin combined with actin, again only 1 peak, the actomyosin peak, is seen.

Fig. M4. Myosin-actin binding as followed in the analytical ultracentrifuge.

Viscometry is another method to measure myosin-actin binding in solution. The viscosity of actomyosin is much higher than the sum of its individual components actin and myosin. Thus, when constant amount of myosin is titrated with increasing amount of fibrous actin the stoichiometry of the actin binding is reached at the maximal viscosity.

In suspension, at physiological ionic strength, the myosin-actin binding is measured under conditions when both myosin and actomyosin are insoluble, that is they sediment at low centrifugal fields. Thus, adding increasing amount of fibrous actin to constant amount of myosin in suspension will result in actomyosin formation. After myosin becomes saturated with actin, the uncombined actin remains in the supernatant of the centrifugate.

Results from either solution or suspension studies show that 1 myosin molecule binds 2 molecules of globular actin units in the fibrous actin polymer, or since each myosin molecule has two heads (where the actin-binding site resides) each myosin head combines with one molecule of globular actin.

ATPase activity of myosin: A Russian husband wife team, Engelhardt and Lyubimova, made the important discovery in 1939 that myosin is an enzyme that hydrolyzes ATP. It was already known that ATP is the universal energy donor in living cells, thus Engelhardt and Lyubimova created the term mechanochemistry i.e. the contractile protein myosin that carries out the work also liberates the energy necessary for the work. This idea is widely accepted today.

The ATPase activity of rabbit skeletal myosin at 37oC in an ionic medium resembling the intracellular fluid of resting muscle (100 mM K+, 2 mM free Mg2+, and 0.1 µM Ca2+) is low, about 0.2 mole of inorganic phosphate (Pi) is liberated per mole myosin head per second. Actin is the physiological activator of myosin ATPase, the substrate is MgATP2-. During muscle contraction the Ca2+ concentration in the intracellular fluid is increased to about 10 µM and the ATPase activity of myosin activated by actin is increased 50-100 times to about 10-20 mole of Pi per mole myosin head per second.

ATP is hydrolyzed by myosin also in the presence of 10 mM Ca2+; this is not physiological. Furthermore, myosin hydrolyzes ATP in the absence of bivalent metals, i.e. in the presence of EDTA, a strong complexing agent for Mg2+ and Ca2+, but K+ has to be present. Na+ cannot substitute K+ . This (K+ + EDTA) ATPase is not physiological either, but it is useful for detection of myosin in non-muscle systems.

The dependence of various ATPase activities of skeletal muscle myosin on the KCl concentration and pH is sketched in Fig. M5.

Fig. M5. KCl and pH dependency of myosin ATPase activity.

Separate actin-binding and ATPase sites of myosin: Titration of the cysteine residues of myosin revealed that the actin-binding ability of myosin can be separated from its ATPase activity (Fig. M5a). These results indicate that the sites of myosin that interact with actin and ATP are different.

Fig. M5a. Separation of the actin-binding ability of myosin from its ATPase activity. Abscissa shows the remaining cysteine residues (SH groups) of myosin after titration with iodoacetamide. Ordinate shows the percentage of actin-binding ability and ATPase activity of the treated myosin relative to those of the control. Symbols: x-x-x, actin-binding ability; open and filled squares, triangles and circles refer to the Ca2+- and actin-activated ATPase activities at various pH values. (From Bárány and Bárány, 1959).

Intermediates of the ATP hydrolysis: It was shown first by Taylor (Lymn and Taylor, 1970; Taylor et al., 1970) that the ATP hydrolysis catalyzed by myosin involves several intermediates:

Subsequent work by Taylor and others revealed that the above reaction is much more complex than shown in the equation. Part b of Fig. M6 shows a recent version of the ATPase cycle and Part a of Fig. M6 shows the coupling of the ATPase cycle to the crossbridge cycle.

Fig. M6. Crossbridge cycle (Part a) correlated with the ATPase cycle (Part b). (From Perry, 1996).

ATPase activity of myosin and speed of muscle shortening: The ATPase activity of myosin was determined in 25 different muscles with a 250-fold variation in the speed of shortening. A correlation was found between the ATPase activity of myosin and the speed of shortening (Fig. M7). This suggests that the myosin ATPase determines the speed of muscle shortening (Bárány, 1967).

Fig. M7. Relationship between myosin ATPase activity and speed of muscle shortening.

Cross-innervation studies provided further evidence for a physiological role of the myosin ATPase - muscle shortening relationship (Bárány and Close, 1971). In rat, the fast extensor digitorum longus muscle was cross innervated with the nerve of the slow soleus muscle and vice versa. This transformed the fast extensor to a slow muscle and the slow soleus to a fast muscle. The changes in the myosin ATPase activity closely followed the changes in the speed of shortening of the cross-innervated muscles (Fig. M8). The high actin-activated ATPase activity of myosin from the extensor muscle was reduced to the low ATPase activity of myosin from the soleus. At the same time the myosin ATPase activity of the cross-innervated soleus was elevated to the level of the normal extensor. A new type of myosin is synthesized in the cross-innervated muscle, which carries the genetic information necessary to determine the speed of muscle contraction.

Fig. M8. The relationship between speed of muscle shortening and actin-activated ATPase activity of myosin in cross-innervated muscles. The symbols represent normal or cross-innervated extensor digitorum longus and soleus muscles.

Myosin heavy chains: The myosin heavy chain (MHC) of different skeletal muscles exhibits diversity. MHC from slow muscle has a slightly lower molecular weight than MHC from the fast muscle, as assessed by SDS-PAGE. The MHC isoform expressed in a single muscle fiber is correlated with the contraction speed of the fiber (Reiser et al., 1985).

During development of the muscle, its contractile properties and myosin isozyme composition are changing. As the muscle speed is increasing more of the fast type MHC is incorporated into the fibers. Furthermore, changes in MHC composition were demonstrated upon increased use of muscle, resulting in hypertrophy, or upon denervation causing atrophy.

Three-dimensional structure of subfragment 1: Rayment and collaborators (1993) crystallized myosin subfragment 1 and determined its three-dimensional structure by x-ray diffraction at 2.8-Å resolution. The structure of myosin S1 is illustrated in Fig. M9. The picture shows the head of S1 consisting of a seven-stranded beta-sheet and a C-terminal tail containing the regulatory light chain (magenta) and the essential light chain (yellow). The proteolytic fragments of S1 are color coded as follows: 25K (N-terminal), green; 50K, red; and 20K (C-terminal) blue. The 50K fragment spans two domains: the 50K upper domain and the 50K lower domain or actin-binding domain (grey). Part of the 50-kDa and 20-kDa fragments form the actin-binding site, whereas part of the 50-kDa and 25-kDa fragments of S1 form the ATP-binding site. The ATP-binding site is about 4 nm from the actin-binding site. The ATP-binding site has the sequence GLY-GLU-SER-GLY-ALA-GLY-LYS-THR, which is similar to the sequences found in the active sites of other ATPases. The ATP-binding site was identified as a pocket. There is a cleft in the upper part of the head that extends from under the ATP-binding site to the actin interface. Both portions, above and below the cleft are involved in the actin binding. When ATP binds to S1, the pocket most likely closes and the cleft widens disturbing the S1-actin binding, that is ATP dissociates S1 from actin. When ATP is hydrolyzed by S1 to ADP and Pi, actin recombines with S1. The accompanying structural changes are the narrowing of the cleft and opening of the ATP-binding pocket. These subtle changes are called conformational changes that play a key role in the mechanism of muscle contraction.

The Converter Domain (Houdusse and Cohen, 1996) is seen on the bottom of Fig. M9. This is a small compact domain that functions as socket for the C-terminal tail of S1, which is its regulatory domain.

The regulatory and essential light chains wrap around the 20-kDa fragment of S1, which forms an 85Å-long a helix that spans much of the S1. Rayment suggested that the conformational changes in the nucleotide-binding pocket could be transmitted to the 85Å-long a helix. This structure may play the role of a lever arm, which magnifies small conformational changes to larger movements. The lever arm hypothesis is presented in the Actin-Myosin Interaction chapter.

Fig. M9. Ribbon representation of the structure of S1 (From Geeves and Holmes, 1999). With permission from the Annual Reviews of Biochemistry vol. 68, 1999, by Annual Reviews (http://www.AnnualReviews.org)

Structure-function relatonship in myosin: Two prominent trypsin sensitive surface loops form the border of the 25-, 50-, and 20-kDa subdomains of S1. The first loop, loop 1, spanning the 25/50-kDa junction is situated near the nucleotide binding site and is involved in determining the rate of ADP release. The 50/20-kDa junction, loop 2, is involved in the interaction between actin and myosin. Nine residues were deleted from loop 2 without affecting the nucleotide binding properties of myosin (Knetsch et al., 1999). However, the deletion affected actin binding and the communication between the actin and nucleotide binding sites. Thus, these studies demonstrated different pathways of communication between the actin and nucleotide binding sites. Further studies by a different group (Joel et al., 2001) showed that elimination of two highly conserved lysines at the C-terminal end of loop 2 specifically blocked the ability of HMM to undergo a week to strong binding transition with actin in the presence of ATP. Removal of these lysines had no effect on strong binding in the absence of nucleotide, on the rate of ATP binding or release, or on the basal ATPase activity. The data suggested that the interaction of the two conserved lysines with acidic residues in subdomain 1 of actin either triggers a structural change or stabilizes a conformation that is necessary for actin-activated Pi release and completion of the ATPase cycle.

Of the functional sites of S1 there is a helix containing the reactive sulfhydryls, Cys707 (SH1) and Cys697 (SH2). The SH1-SH2 helix can undergo structural changes in the presence of ATP and its derivatives. For instance SH1 and SH2 can be cross-linked in the presence of nucleotides with reagents of spans ranging from 5 to 15 Å, although the distance between the sulfur atoms in the native helix is as much as 19 Å. Therefore, in the presence of nucleotides the helix must undergo some conformational changes in order for SH1 and SH2 to come close to each other. Using the cross-linking approach, it was shown that nucleotides induce a flexibility of the SH1 and SH2 helix, in other words nucleotides shift the equilibria among conformational states of the helix (Nitao and Reisler, 1998).

Assembly

Myosin filament: At low ionic strength, e.g. 0.03 M KCl, myosin precipitates and forms filaments. Electron micrographs reveal the specific structure of the filaments, that is their central shaft and side projections (Fig. M10).

Fig. M10. Electron micrograph of myosin filament (From Huxley, H.E., reprinted with permission from the Structure and Function of Muscle, G. H. Bourne , Ed. , 1972, by Academic Press.).

A model for the arrangement of myosin molecules in the filaments is shown in Fig. M11. Since individual myosin molecules have a globular region (S1) at one end only, the filaments are formed probably by antiparallel association of myosin molecules. All the molecules in one half filament are oriented in one direction and all those in the other half of the filament are oriented in the opposite direction. Thus, in the middle of the filament the tails of antiparallel molecules overlap yielding a bare central shaft, and globular regions are projected at both ends of the filament. During contraction the S1 molecules (the cross bridges) are bending and this is illustrated by the dashed S1 molecules.

Fig. M11. Model of myosin filament (Courtesy of Dr. Helen Rarick).

Electron micrographs of thick filaments from muscle and synthetic thick filaments made from myosin are indistinguishable. However, synthetic thick filaments made from light meromyosin have no projections, as shown by electron microscopy.

To understand how myosin filaments function in contraction, it is necessary to know their native structure, composition, and biochemistry. This requires isolation of native myosin filaments directly from muscle and their purification free of thin filaments and contaminating proteins (Hidalgo et al., 2001).

Muscle fibers, myofibrils: Actomyosin threads produce much less tension than intact muscle and this initiated research on muscle fibers. In his classical experiments Szent-Györgyi divided rabbit psoas muscle in situ into fiber bundles about 1 mm in diameter. These were tied at resting length to a thin stick and placed in 50% glycerol at 0o C for 24 h. After exchanging the glycerol, the fiber bundles were stored at -20o C. Before use, the bundles were transferred to 20% glycerol, then washed with saline. The prolonged glycerol treatment destroys the muscle cell membrane, and the subsequent washing removes the inorganic and organic constituents of the muscle and over half of the sarcoplasmic proteins. Glycerol-treated psoas fibers no longer react to electrical stimulation, but upon addition of ATP produce powerful contraction.

H.H. Weber and Portzehl (1954) prepared single muscle fibers from glycerol treated psoas muscles that developed tension equal to the intact muscle and reproduced the entire contraction-relaxation cycle of the muscle (Fig. M12). Thus, it was proven without doubt that the interaction between actin, myosin and ATP is the basic mechanism for the contraction-relaxation cycle in skeletal muscle.

Fig. M12. Contraction-relaxation-contraction cycle of a single psoas fiber. The fiber was contracted by ATP and at the top of contraction washed with saline. The fiber was relaxed by adding pyrophosphate to the bath, arrow down, then recontracted by the addition of ATP, arrow up. (From Weber and Portzehl, 1954).

Myofibrils are tiny muscle fibers, prepared by homogenization of freshly dissected muscle in physiological salt solution. Their ATP-induced contraction can be followed under the microscope.

Both psoas fibers and myofibrils contain the contractile (myosin and actin) and the regulatory proteins (tropomyosin and troponin) of muscle. The individual component of these systems are well resolved by SDS-PAGE (Fig. M13).

Fig. M13. SDS-PAGE of purified skeletal muscle myofibrils.(Reprinted from Biochim. Biophys. Acta vol. 490, Porzio, M.A. and Pearson, A.M. " Improved resolution of myofibrillar proteins with sodium dodecyl sulfate gel electrophoresis," pp. 27-34, Copyright 1977, with permission from Elsevier Sciences).

Not shown in Fig. M13 are the thick filament proteins, titin, H protein, M band protein, and the thin filament protein nebulin.

Localization of myosin in the structure of muscle: Myofibrils were extracted under the microscope with a solution selective for myosin dissolution. The A-band disappeared as a result of the extraction; hence the conclusion was reached that myosin is localized in the A-band, the darkly staining part of the muscle. Antibodies, specific for myosin that were deposited in the A-band also confirmed the localization.

References

Bárány, M. (1967). ATPase activity of myosin correlated with speed of muscle shortening. J. Gen. Physiol. 50, 197-218.

Bárány, M. and Bárány, K. (1959). Studies on "active centers" of L-myosin. Biochim. Biophys. Acta, 35, 293-309.

Bárány, M. and Close, R. I. (1971). The transformation of myosin in cross-innervated rat muscles. J. Physiol. 213, 455-474.

Engelhardt, V.A. and Lyubimova, M.N. (1939). Myosin and adenosinetriphosphatase. Nature, 144, 668.

Geeves, M.A. and Holmes, K.C. (1999). Structural mechanism of muscle contraction. Annu. Rev. Biiochem. , 68, 687-728.

Hidalgo, C., Padron, R., Horowitz, R., Zhao, F-G., and Craig, R. (2001). Purification of native myosin filaments from muscle. Biophys. J. 81, 2817-2826.

Houdusse, A. and Cohen, C. (1996). Structure of the regulatory domain of scallop myosin at 2 angstrom resolution. Structure, 4, 21-32.

Huxley, H.E. (1972). Molecular basis for contraction in cross-striated muscles. In The structure and function of muscle, ( G.H. Bourne, Ed.) second edition, vol. I, Part 1, pp. 301-387.

Joel, P.B., Trybus, K.M., and Sweeney, H.L. (2001). Two conserved lysines at the 50/20-kDa junction of myosin are necessary for triggering actin activation. J. Biol. Chem. 276, 2998-3003.

Knetsch, M. L.W., Uyeda, T.Q. P., and Manstein, D.J. (1999). Disturbed communication between actin- and nucleotide-binding sites in a myosin II with truncated 50/20-kDa junction. J. Biol. Chem. 274, 20133-20138.

Lowey, S., Slayter, H.S., Weeds, A.G., and Baker, H. (1969). Substructure of the myosin molecule I. Subfragments of myosin by enzymic degradation. J. Mol. Biol., 42, 1-20.

Lymn, R.W. and Taylor, E.W. (1970). Transient state phosphate production in the hydrolysis of nucleoside triphosphates by myosin. Biochemistry, 9, 2975-2983.

Nitao, L.K., and Reisler, E. (1998). Probing the conformational states of the SH1-SH2 helix in myosin: a cross-linking approach. Biochemistsry, 37, 16704-16710.

Perry, S.V. (1996). Molecular mechanisms in striated muscle. Cambridge University Press.

Porzio, M.A. and Pearson, A.M. (1977). Improved resolution of myofibrillar proteins with sodium dodecylsulfate polyacrylamide gel electrophoresis. Biochim. Biophys. Acta, 490, 27-34.

Rayment, I., Rypniewski, W.R., Schmidt-Bäse, K., Smith, R., Tomchick, D.R., Benning, M.M., Winkelman, D.A., Wesenberg, G., and Holden, H.M. (1993). Three-dimensional structure of myosin subfragment-1: a molecular motor. Science, 261, 50-58.

Reiser, P.J., Moss, R.L., Giulian, G.G., and Greaser, M.L. (1985). Shortening velocity in single fibers from adult rabbit soleus muscles is correlated with myosin heavy chain composition. J. Biol. Chem. 260, 9077- 9080.

Sellers, J.R. and Goodson, H.V. (1995). Motor proteins 2: myosin. Protein Profile, 2, 1323-1423.

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.

Weber, H.H. and Portzehl, H. (1954). The transference of the muscle energy in the contraction cycle. Progress in Biophysics and Biophysical Chemistry, 4, 60-111.


Back to the Home

Back to the Beginning of the Chapter

Go to the Next Chapter