Movement is an intrinsic property associated with all living things. Within cells, molecules undergo coordinated and organized movements, and cells themselves may move across a surface. At the tissue level, muscle contraction allows higher organisms to carry out and control crucial internal functions, such as peristalsis in the gut and the beating of the heart. Muscle contraction also enables the organism to carry out organized and sophisticated movements, such as walking, running, flying, and swimming.

17.1 • Molecular Motors

Motor proteins, also known as molecular motors, use chemical energy (ATP) to orchestrate all these movements, transforming ATP energy into the mechanical energy of motion. In all cases, ATP hydrolysis is presumed to drive and control protein conformational changes that result in sliding or walking movements of one molecule relative to another. To carry out directed movements, molecular motors must be able to associate and dissociate reversibly with a polymeric protein array, a surface or substructure in the cell. ATP hydrolysis  drives the process by which the motor protein ratchets along the protein array or surface. As fundamental and straightforward as all this sounds, elucidation of these basically simple processes has been extremely challenging for biochemists, involving the application of many sophisticated chemical and physical methods in many different laboratories. This chapter describes the structures and chemical functions of molecular motor proteins and some of the experiments by which we have come to understand them.

17.2 • Microtubules and Their Motors

One of the simplest self-assembling structures found in biological systems is the microtubule, one of the fundamental components of the eukaryotic cytoskeleton and the primary structural element of cilia and flagella (Figure 17.1).

Figure 17.1   •   Micrographs and electron micrographs of cytoskeletal elements, cilia, and flagella: (a) microtubules, (b) rat sperm tail microtubules (cross-section), (c) Stylonychia, a ciliated protozoan (undergoing division), (d) cytoskeleton of a eukaryotic cell, (e) Pseudomonas fluorescens (aerobic soil bacterium), showing flagella, (f) nasal cilia. ^{(a, K. G. Murti/Visuals Unlimited; b, David Phillips/Visuals Unlimited; c, Eric Grave/Phototake; d, Fawcett and Heuser/Photo Researchers, Inc.; e, Dr. Tony Brain/Custom Medical Stock; f, Veronika Burmeister, Visuals Unlimited)}

Microtubules are hollow, cylindrical structures, approximately 30 nm in diameter, formed from tubulin, a dimeric protein composed of two similar 55-kD subunits known as a-tubulin and b-tubulin. Eva Nogales, Sharon Wolf, and Kenneth Downing have determined the structure of the bovine tubulin ab dimer to 3.7 Å resolution (Figure 17.2a). Tubulin dimers polymerize as shown in Figure 17.2b to form microtubules, which are essentially helical structures, with 13 tubulin monomer “residues” per turn.

Figure 17.2   •  (a) The structure of the tubulin ab heterodimer. (b) Microtubules may be viewed as consisting of 13 parallel, staggered protofilaments of alternating a -tubulin and b -tubulin subunits. The sequences of the a and b subunits of tubulin are homologous, and the ab tubulin dimers are quite stable if Ca²⁺ is present. The dimer is dissociated only by strong denaturing agents.

Microtubules grown in vitro are dynamic structures that are constantly being assembled and disassembled. Because all tubulin dimers in a microtubule are oriented similarly, microtubules are polar structures. The end of the microtubule at which growth occurs is the plus end, and the other is the minus end. Microtubules in vitro carry out a GTP-dependent process called treadmilling, in which tubulin dimers are added to the plus end at about the same rate at which dimers are removed from the minus end (Figure 17.3).

Figure 17.3   •  A model of the GTP-dependent treadmilling process. Both a- and b-tubulin possess two different binding sites for GTP. The polymerization of tubulin to form microtubules is driven by GTP hydrolysis in a process that is only beginning to be understood in detail.

Microtubules Are Constituents of the Cytoskeleton

Although composed only of 55-kD tubulin subunits, microtubules can grow sufficiently large to span a eukaryotic cell or to form large structures such as cilia and flagella. Inside cells, networks of microtubules play many functions, including formation of the mitotic sP_indle that segregates chromosomes during cell division, the movement of organelles and various vesicular structures through the cell, and the variation and maintenance of cell shape. Microtubules are, in fact, a significant part of the cytoskeleton, a sort of intracellular scaffold formed of microtubules, intermediate filaments, and microfilaments (Figure 17.4). In most cells, microtubules are oriented with their minus ends toward the centrosome and their plus ends toward the cell periphery. This consistent orientation is important for mechanisms of intracellular transport.

Figure 17.4 •  Intermediate filaments have diameters of approximately 7 to 12 nm, whereas microfilaments, which are made from actin, have diameters of approximately 7 nm. The intermediate filaments appear to play only a structural role (maintaining cell shape), but the microfilaments and microtubules play more dynamic roles. Microfilaments are involved in cell motility, whereas microtubules act as long filamentous tracks, along which cellular components may be raP_idly transported by specific mechanisms. (a) Cyto­skeleton, double-labeled with actin in red and tubulin in green. (b) Cytoskeletal elements in a eukaryotic cell, including microtubules (thickest strands), intermediate filaments, and actin microfilaments (smallest strands). (a, b, M. Schliwa/Visuals Unlimited)

Microtubules Are the Fundamental Structural
Units of Cilia and Flagella

As already noted, microtubules are also the fundamental building blocks of cilia and flagella. Cilia are short, cylindrical, hairlike projections on the surfaces of the cells of many animals and lower plants. The beating motion of cilia functions either to move cells from place to place or to facilitate the movement of extracellular fluid over the cell surface. Flagella are much longer structures found singly or a few at a time on certain cells (such as sperm cells). They propel cells through fluids. Cilia and flagella share a common design (Figure 17.5).

Figure 17.5   •  The structure of an axoneme. Note the manner in which two microtubules are joined in the nine outer pairs. The smaller-diameter tubule of each pair, which is a true cylinder, is called the A-tubule and is joined to the center sheath of the axoneme by a spoke structure. Each outer pair of tubules is joined to adjacent pairs by a nexin bridge. The A-tubule of each outer pair possesses an outer dynein arm and an inner dynein arm. The larger-diameter tubule is known as the B-tubule.

The axoneme is a complex bundle of microtubule fibers that includes two central, separated microtubules surrounded by nine pairs of joined microtubules. The axoneme is surrounded by a plasma membrane that is continuous with the plasma membrane of the cell. Removal of the plasma membrane by detergent and subsequent treatment of the exposed axonemes with high concentrations of salt releases the dynein molecules (Figure 17.6), which form the dynein arms.

Figure 17.6   •  (a) Diagram showing dynein interactions between adjacent microtubule pairs. (b) Detailed views of dynein crosslinks between the A-tubule of one microtubule pair and the B-tubule of a neighboring pair. (The B-tubule of the first pair and the A-tubule of the neighboring pair are omitted for clarity.) Isolated axonemal dyneins, which possess ATPase activity, consist of two or three “heavy chains” with molecular masses of 400 to 500 kD, referred to as a and b (and g when present), as well as several chains with intermediate (40 to 120 kD) and low (15 to 25 kD) molecular masses. Each outer-arm heavy chain consists of a globular domain with a flexible stem on one end and a shorter projection extending at an angle with respect to the flexible stem. In a dynein arm, the flexible stems of several heavy chains are joined in a common base, where the intermediate- and low-molecular-weight proteins are located.

The Mechanism of Ciliary Motion

The motion of cilia results from the ATP-driven sliding or walking of dyneins along one microtubule while they remain firmly attached to an adjacent microtubule. The flexible stems of the dyneins remain permanently attached to A-tubules (Figure 17.6). However, the projections on the globular heads form transient attachments to adjacent B-tubules. Binding of ATP to the dynein heavy chain causes dissociation of the projections from the B-tubules. These projections then reattach to the B-tubules at a position closer to the minus end. Repetition of this process causes the sliding of A-tubules relative to B-tubules. The cross-linked structure of the axoneme dictates that this sliding motion will occur in an asymmetric fashion, resulting in a bending motion of the axoneme, as shown in Figure 17.7

Figure 17.7   •  A mechanism for ciliary motion. The sliding motion of dyneins along one microtubule while attached to an adjacent microtubule results in a bending motion of the axoneme.

Microtubules Also Mediate Intracellular
Motion of Organelles and Vesicles

The ability of dyneins to effect mechano-chemical coupling — i.e., motion coupled with a chemical reaction — is also vitally important inside eukaryotic cells, which, as already noted, contain microtubule networks as part of the cytoskeleton. The mechanisms of intracellular, microtubule-based transport of organelles and vesicles were first elucidated in studies of axons, the long projections of neurons that extend great distances away from the body of the cell. In these cells, it was found that subcellular organelles and vesicles could travel at surprisingly fast rates — as great as 2 to 5 mm/sec — in either direction. Unraveling the molecular mechanism for this raP_id transport turned out to be a challenging biochemical problem. The early evidence that these movements occur by association with specialized proteins on the microtubules was met with some resistance, for two reasons. First, the notion that a network of microtubules could mediate transport was novel and, like all novel ideas, difficult to accept. Second, many early attempts to isolate dyneins from neural tissue were unsuccessful, and the dynein-like proteins that were first isolated from cytosolic fractions were thought to represent contaminations from axoneme structures. However, things changed dramatically in 1985 with a report by Michael Sheetz and his coworkers of a new ATP-driven, force-generating protein, different from myosin and dynein, which they called kinesin. Then, in 1987, Richard McIntosh and Mary Porter described the isolation of cytosolic dynein proteins from Caenorhabditis elegans, a nematode worm that never makes motile axonemes at any stage of its life cycle. Kinesins have now been found in many eukaryotic cell types, and similar cytosolic dyneins have been found in fruit flies, amoebae, and slime molds; in vertebrate brain and testes; and in HeLa cells (a unique human tumor cell line).

Dyneins Move Organelles in a Plus-to-Minus Direction;
Kinesins, in a Minus-to-Plus Direction

The cytosolic dyneins bear many similarities to axonemal dynein. The protein isolated from C. elegans includes a “heavy chain” with a molecular mass of approximately 400 kD, as well as smaller peptides with molecular mass ranging from 53 kD to 74 kD. The protein possesses a microtubule-activated ATPase activity, and, when anchored to a glass surface in vitro, these proteins, in the presence of ATP, can bind microtubules and move them through the solution. In the cell, cytosolic dyneins specifically move organelles and vesicles from the plus end of a microtubule to the minus end. Thus, as shown in Figure 17.8, dyneins move vesicles and organelles from the cell periphery toward the centrosome (or, in an axon, from the synaptic termini toward the cell body). The kinesins, on the other hand, assist the movement of organelles and vesicles from the minus end to the plus end of microtubules, resulting in outward movement of organelles and vesicles. Kinesin is similar to cytosolic dyneins but smaller in size (360 kD), and contains subunits of 110 kD and 65 to 70 kD. Its length is 100 nm. Like dyneins, kinesins possess ATPase activity in their globular heads, and it is the free energy of ATP hydrolysis that drives the movement of vesicles along the microtubules.

Figure 17.8 •  (a) RaP_id axonal transport along microtubules permits the exchange of material between the synaptic terminal and the body of the nerve cell. (b) Vesicles, multivesicular bodies, and mitochondria are carried through the axon by this mechanism. (Adapted from a drawing by Ronald Vale)

            The N-terminal domain of the kinesin heavy chain (38 kD, approximately 340 residues) contains the ATP- and microtubule-binding sites and is the domain responsible for movement. Electron microscopy and image analysis of tubulin-kinesin complexes reveals (Figure 17.9) that the kinesin head domain is compact and primarily contacts a single tubulin subunit on a microtubule surface, inducing a conformational change in the tubulin subunit. Optical trapP_ing experiments (see page 554) demonstrate that kinesin heads move in 8-nm (80-Å) steps along the long axis of a microtubule. Kenneth Johnson and his coworkers have shown that the ability of a single kinesin tetramer to move unidirectionally for long distances on a microtubule depends upon cooperative interactions between the two mechanochemical head domains of the protein.

Figure 17.9   •  The structure of the tubulin-kinesin complex, as revealed by image analysis of cryoelectron microscopy data. (a) The computed, three-dimensional map of a microtubule, (b) the kinesin globular head domain-microtubule complex, (c) a contour plot of a horizontal section of the kinesin-microtubule complex, and (d) a contour plot of a vertical section of the same complex. (Taken from Kikkawa et al., 1995. Nature ^{376:274-277. Photo courtesy of Nobutaka Hirokawa.)}

17.3 • Skeletal Muscle Myosin and Muscle Contraction

The Morphology of Muscle

Four different kinds of muscle are found in animals (Figure 17.10). They are skeletal muscle, cardiac (heart) muscle, smooth muscle, and myoeP_ithelial cells. The cells of the latter three types contain only a single nucleus and are called myocytes. The cells of skeletal muscle are long and multinucleate and are referred to as muscle fibers. At the microscoP_ic level, skeletal muscle and cardiac muscle display alternating light and dark bands, and for this reason are often referred to as striated muscles. The different types of muscle cells vary widely in structure, size, and function. In addition, the times required for contractions and relaxations by various muscle types vary considerably. The fastest responses (on the order of milliseconds) are observed for fast-twitch skeletal muscle, and the slowest responses (on the order of seconds) are found in smooth muscle. Slow-twitch skeletal muscle tissue displays an intermediate response time.

Figure 17.10  •  The four classes of muscle cells in mammals. Skeletal muscle and cardiac muscle are striated. Cardiac muscle, smooth muscle, and myoeP_ithelial cells are mononucleate, whereas skeletal muscle is multinucleate.

Structural Features of Skeletal Muscle

Skeletal muscles in higher animals consist of 100-mm-diameter fiber bundles, some as long as the muscle itself. Each of these muscle fibers contains hundreds of myofibrils (Figure 17.11), each of which spans the length of the fiber and is about 1 to 2 m m in diameter. Myofibrils are linear arrays of cylindrical sarcomeres, the basic structural units of muscle contraction. The sarcomeres are surrounded on each end by a membrane system that is actually an elaborate extension of the muscle fiber plasma membrane or sarcolemma. These extensions of the sarcolemma, which are called transverse tubules or t-tubules, enable the sarcolemmal membrane to contact the ends of each myofibril in the muscle fiber (Figure 17.11). This topological feature is crucial to the initiation of contractions. In between the t-tubules, the sarcomere is covered with a specialized endoplasmic reticulum called the sarcoplasmic reticulum, or SR. The SR contains high concentrations of Ca²⁺, and the release of Ca²⁺ from the SR and its interactions within the sarcomeres trigger muscle contraction, as we will see. Each SR structure consists of two domains. Longitudinal tubules run the length of the sarcomere and are capped on either end by the terminal cisternae (Figure 17.11). The structure at the end of each sarcomere, which consists of a t-tubule and two apposed terminal cisternae, is called a triad, and the intervening gaps of approximately 15 nm are called triad junctions. The junctional face of each terminal cisterna is joined to its respective t-tubule by a foot structure. Skeletal muscle contractions are initiated by nerve stimuli that act directly on the muscle. Nerve impulses produce an electrochemical signal (see Chapter 34) called an action potential that spreads over the sarcolemmal membrane and into the fiber along the t-tubule network. This signal is passed across the triad junction and induces the release of Ca²⁺ ions from the SR. These Ca²⁺ ions bind to the muscle fibers and induce contraction.

Figure 17.11   •  The structure of a skeletal muscle cell, showing the manner in which t-tubules enable the sarcolemmal membrane to contact the ends of each myofibril in the muscle fiber. The foot structure is shown in the box.

The Molecular Structure of Skeletal Muscle

Examination of myofibrils in the electron microscope reveals a banded or striated structure. The bands are traditionally identified by letters (Figure 17.12). Regions of high electron density, denoted A bands, alternate with regions of low electron density, the I bands. Small, dark Z lines lie in the middle of the I bands, marking the ends of the sarcomere. Each A band has a central region of slightly lower electron density called the H zone, which contains a central M disk (also called an M line). Electron micrographs of cross-sections of each of these regions reveal molecular details. The H zone shows a regular, hexagonally arranged array of thick filaments (15 nm diameter), whereas the I band shows a regular, hexagonal array of thin filaments (7 nm diameter). In the dark regions at the ends of each A band, the thin and thick filaments interdigitate, as shown in Figure 17.12. The thin filaments are composed primarily of three proteins called actin, troponin, and tropomyosin. The thick filaments consist mainly of a protein called myosin. The thin and thick filaments are joined by cross-bridges. These cross-bridges are actually extensions of the myosin molecules, and muscle contraction is accomplished by the sliding of the cross-bridges along the thin filaments, a mechanical movement driven by the free energy of ATP hydrolysis.

Figure 17.12   •  Electron micrograph of a skeletal muscle myofibril (in longitudinal section). The length of one sarcomere is indicated, as are the A and I bands, the H zone, the M disk, and the Z lines. Cross-sections from the H zone show a hexagonal array of thick filaments, whereas the I band cross-section shows a hexagonal array of thin filaments. (Photo courtesy of Hugh Huxley, Brandeis University)

The Composition and Structure of Thin Filaments

Actin, the principal component of thin filaments, can be isolated in two forms. Under conditions of low ionic strength, actin exists as a 42-kD globular protein, denoted G-actin. G-actin consists of two principal lobes or domains (Figure 17.13).

Figure 17.13   •  The three-dimensional structure of an actin monomer from skeletal muscle. This view shows the two domains (left and right) of actin.

Under physiological conditions (higher ionic strength), G-actin polymerizes to form a fibrous form of actin, called F-actin. As shown in Figure 17.14, F-actin is a right-handed helical structure, with a helix P_itch of about 72 nm per turn. The F-actin helix is the core of the thin filament, to which tropomyosin and the troponin complex also add. Tropomyosin is a dimer of homologous but nonidentical 33-kD subunits.

Figure 17.14   •  The helical arrangement of actin monomers in F-actin. The F-actin helix has a P_itch of 72 nm and a repeat distance of 36 nm. (Electron micrograph courtesy of Hugh Huxley, Brandeis University)

These two subunits form long a-helices that intertwine, creating 38- to 40-nm-long coiled coils, which join in head-to-tail fashion to form long rods. These rods bind to the F-actin polymer and lie almost parallel to the long axis of the F-actin helix (Figure 17.15 a-c). Each tropomyosin heterodimer contacts approximately seven actin subunits. The troponin complex consists of three different proteins: troponin T, or TnT (37 kD); troponin I, or TnI (24 kD); and troponin C, or TnC (18 kD). TnT binds to tropomyosin, specifically at the head-to-tail junction. Troponin I binds both to tropomyosin and to actin. Troponin C is a Ca²⁺-binding protein that binds to TnI. TnC shows 70% homology with the important Ca²⁺ signaling protein, calmodulin (Chapter 34). The release of Ca²⁺ from the SR, which signals a contraction, raises the cytosolic Ca²⁺ concentration high enough to saturate the Ca²⁺ sites on TnC. Ca²⁺ binding induces a conformational change in the amino-terminal domain of TnC, which in turn causes a rearrangement of the troponin complex and tropomyosin with respect to the actin fiber.

Figure 17.15   •  (a) An electron micrograph of a thin filament, (b) a corresponding image reconstruction, and (c) a schematic drawing based on the images in (a) and (b). The tropomyosin coiled coil winds around the actin helix, each tropomyosin dimer interacting with seven consecutive actin monomers. Troponin T binds to tropomyosin at the head-to-tail junction. (a and b, courtesy of Linda Rost and David DeRosier, Brandeis University; c, courtesy of George Phillips, Rice University)

The Composition and Structure of Thick Filaments

Myosin, the principal component of muscle thick filaments, is a large protein consisting of six polypeptides, with an aggregate molecular weight of approximately 540 kD. As shown in Figure 17.16, the six peptides include two 230-kD heavy chains, as well as two pairs of different 20-kD light chains, denoted LC1 and LC2. The heavy chains consist of globular amino-terminal myosin heads, joined to long a-helical carboxy-terminal segments, the tails. These tails are intertwined to form a left-handed coiled coil approximately 2 nm in diameter and 130 to 150 nm long. Each of the heads in this dimeric structure is associated with an LC1 and an LC2. The myosin heads exhibit ATPase activity, and hydrolysis of ATP by the myosin heads drives muscle contraction. LC1 is also known as the essential light chain, and LC2 is designated the regulatory light chain. Both light chains are homologous to calmodulin and TnC. Dissociation of LC1 from the myosin heads by alkali cations results in loss of the myosin ATPase activity.

Figure 17.16   •  (a) An electron micrograph of a myosin molecule and a corresponding schematic drawing. The tail is a coiled coil of intertwined a -helices extending from the two globular heads. One of each of the myosin light chain proteins, LC1 and LC2, is bound to each of the globular heads. (b) A ribbon diagram shows the structure of the S1 myosin head (green, red, and purple segments) and its associated essential (yellow) and regulatory (magenta) light chains. (a, Electron micrograph courtesy of Henry Slayter, Harvard Medical School; b, courtesy of Ivan Rayment and Hazel M. Holden, University of Wisconsin, Madison)

            Approximately 500 of the 820 amino acid residues of the myosin head are highly conserved between various species. One conserved region, located approximately at residues 170 to 214, constitutes part of the ATP-binding site. Whereas many ATP-binding proteins and enzymes employ a b-sheet-a-helix-b-sheet motif, this region of myosin forms a related a-b-a structure, beginning with an Arg at (approximately) residue 192. The b-sheet in this region of all myosins includes the amino acid sequence

            Gly-Glu-Ser-Gly-Ala-Gly-Lys-Thr

The Gly-X-X-Gly-X-Gly found in this segment is found in many ATP- and nucleotide-binding enzymes. The Lys of this segment is thought to interact with the a-phosphate of bound ATP.

Repeating Structural Elements Are the Secret of Myosin’s Coiled Coils

Myosin tails show less homology than the head regions, but several key features of the tail sequence are responsible for the a-helical coiled coils formed by myosin tails. Several orders of repeating structure are found in all myosin tails, including 7-residue, 28-residue, and 196-residue repeating units. Large stretches of the tail domain are composed of 7-residue repeating segments. The first and fourth residues of these 7-residue units are generally small, hydrophobic amino acids, whereas the second, third, and sixth are likely to be charged residues. The consequence of this arrangement is shown in Figure 17.17. Seven residues form two turns of an a-helix, and, in the coiled coil structure of the myosin tails, the first and fourth residues face the interior contact region of the coiled coil. Residues b, c, and f (2, 3, and 6) of the 7-residue repeat face the periphery, where charged residues can interact with the water solvent. Groups of four 7-residue units with distinct patterns of alternating side-chain charge form 28-residue repeats that establish alternating regions of positive and negative charge on the surface of the myosin coiled coil. These alternating charged regions interact with similar regions in the tails of adjacent myosin molecules to assist in stabilizing the thick filament.

Figure 17.17   •  An axial view of the two-stranded, a-helical coiled coil of a myosin tail. Hydrophobic residues a and d of the seven-residue repeat sequence align to form a hydrophobic core. Residues b, c, and f face the outer surface of the coiled coil and are tyP_ically ionic.

            At a still higher level of organization, groups of seven of these 28-residue units — a total of 196 residues — also form a repeating pattern, and this large-scale repeating motif contributes to the packing of the myosin molecules in the thick filament. The myosin molecules in thick filaments are offset (Figure 17.18) by approximately 14 nm, a distance that corresponds to 98 residues of a coiled coil, or exactly half the length of the 196-residue repeat. Thus, several layers of repeating structure play specific roles in the formation and stabilization of the myosin coiled coil and the thick filament formed from them.

Figure 17.18   •  The packing of myosin molecules in a thick filament. Adjoining molecules are offset by approximately 14 nm, a distance corresponding to 98 residues of the coiled coil.

The Associated Proteins of Striated Muscle

In addition to the major proteins of striated muscle (myosin, actin, tropomy-osin, and the troponins), numerous other proteins play important roles in the maintenance of muscle structure and the regulation of muscle contraction. Myosin and actin together account for 65% of the total muscle protein, and tropomyosin and the troponins each contribute an additional 5% (Table 17.1). The other regulatory and structural proteins thus comprise approximately 25% of the myofibrillar protein. The regulatory proteins can be classified as either myosin-associated proteins or actin-associated proteins.

Human Biochemistry
The Molecular Defect in Duchenne Muscular Dystrophy Involves an Actin-Anchoring Protein Discovery of a new actinin/spectrin - like protein has provided insights into the molecular basis for at least one form of muscular dystrophy. Duchenne muscular dystrophy is a degenerative and fatal disorder of muscle affecting approximately 1 in 3500 boys. Victims of Duchenne dystrophy show early abnormalities in walking and running. By the age of five, the victim cannot run and has difficulty standing, and by early adolescence, walking is difficult or impossible. The loss of muscle function progresses upward in the body, affecting next the arms and the diaphragm. ResPiratory problems or infections usually result in death by the age of 30. Louis Kunkel and his coworkers identified the Duchenne muscular dystrophy gene in 1986. This gene produces a protein called dystrophin, which is highly homologous to a-actinin and spectrin. A defect in dystrophin is responsible for the muscle degeneration of Duchenne dystrophy.             Dystrophin is located on the cytoplasmic face of the muscle plasma membrane, linked to the plasma membrane via an integral membrane glycoprotein. Dystrophin has a high molecular mass (427 kD ), but constitutes less than 0.01% of the total muscle protein. It folds into four principal domains (figure, part a), including an N-terminal domain similar to the actin-binding domains of actinin and spectrin, a long repeat domain, a cysteine-rich domain, and a C-terminal domain that is unique to dystrophin. The repeat domain consists of 24 repeat units of approximately	109 residues each. “Spacer sequences” high in proline content, which do not align with the repeat consensus sequence, occur at the beginning and end of the repeat domain. Spacer segments are found between repeat elements 3 and 4 and 19 and 20. The high proline content of the spacers suggests that they may represent hinge domains. The spacer/hinge segments are sensitive to proteolytic enzymes, indicating that they may represent more exposed regions of the polypeptide.             Dystrophin itself appears to be part of an elaborate protein/glycoprotein complex that bridges the inner cytoskeleton (actin filaments) and the extracellular matrix (via a matrix protein called laminin (see figure)). It is now clear that defects in one or more of the proteins in this complex are responsible for many of the other forms of muscular dystrophy. The glycoprotein complex is composed of two subcomplexes, the dystroglycan complex and the sarcoglycan complex. The dystroglycan complex consists of a-dystroglycan, an extracellular protein that binds to merosin, a laminin subunit and component of the extracellular matrix, and b-dystroglycan, a transmembrane protein that binds the C-terminal domain of dystrophin inside the cell (see figure). The sarcoglycan complex is composed of a-, b-, and g-sarcoglycans, all of which are transmembrane glycoproteins. Alterations of the sarcoglycan proteins are linked to limb-girdle muscular dystrophy and autosomal recessive muscular dystrophy. Mutations in the gene for merosin, which binds to a-dystroglycan, are linked to severe congenital muscular dystrophy, yet another form of the disease.
A model for the actin - dystrophin - glycoprotein complex in skeletal muscle. Dystrophin is postulated to form tetramers of antiparallel monomers that bind actin at their N-termini and a family of dystrophin-associated glycoproteins at their C-termini. This dystrophin-anchored complex may function to stabilize the sarcolemmal membrane during contraction - relaxation cycles, link the contractile force generated in the cell (fiber) with the extracellular environment, or maintain local organization of key proteins in the membrane. The dystrophin-associated membrane proteins (dystroglycans and sarcoglycans) range from 25 to 154 kD . (Adapted from Ahn, A. H., and Kunkel, L. M., 1993. Nature Genetics 3:283-291, and Worton, R., 1995. Science 270:755-756.)

            The myosin-associated proteins include three proteins found in the M disks. The M disks consist primarily of M protein (165 kD ), myomesin (185 kD), and creatine kinase (a dimer of 42-kD subunits). Creatine kinase facilitates raP_id regeneration of the ATP consumed during muscle contraction. The association of M protein, myomesin, and creatine kinase in the M disk maintains the structural integrity of the myosin filaments. Several other myosin-associated proteins have also been identified, including C protein (135 kD ), F protein (121 kD), H protein (74 kD), and I protein (50 kD). The C protein is localized to several regularly spaced stripes in the A band. C protein inhibits myosin ATPase activity at low ionic strength but activates it at physiological ionic strength. The roles of F, H, and I proteins are not yet understood.
            Actin-associated proteins (other than tropomyosin and the troponins) include a-actinin (a homodimer of 95-kD subunits), b-actinin (a heterodimer of 37-kD and 34-kD subunits), g-actinin (a 35-kD monomer), and para-tropomyosin (a homodimer of 34-kD subunits). a-Actinin is found in the Z lines and activates contraction of actomyosin. It is thought to play a role in attachment of actin to the Z lines. a-Actinin consists of three domains: an N-terminal, actin-binding domain; a central domain consisting of four repeats of a 122-residue sequence; and a C-terminal domain that contains two EF-hand, calcium-binding domains (Figure 17.19). The four central repeats in a-actinin are highly homologous with the 106-residue repeat sequences of spectrin, the major structural protein of the red blood cell cytoskeleton.

Figure 17.19   •   a -Actinin exists as a homo­dimer of antiparallel subunits, illustrated here in terms of their primary structure. The N-terminal, actin-binding domain and the C-terminal, EF-hand domains are separated by a central domain consisting of four repeats of a 122-residue sequence.

The repeating segments of both a-actinin and spectrin are thought to consist of bundles of four a-helices (Figure 17.20). b-Actinin acts as an actin-capP_ing protein, specifically binding to the end of an actin filament. g-Actinin also inhibits actin polymerization, but its location in thin filaments is not known with certainty. Paratropomyosin is similar to tropomyosin, but appears to be located only at the A band-I band junction.

Figure 17.20   •  A schematic drawing of the four-helix cluster model for a-actinin and spectrin. Helix 1 is long and is postulated to lie at an angle with respect to the long axis of the repeated domain.

            Two cytoskeletal proteins, titin (also known as connectin) and nebulin, account for 15% of the total protein in the myofibril. Together these proteins form a flexible filamentous network that surrounds the myofibrils. Titin is an elastic protein and can stretch under tension. Its discovery and characterization ended a century-long debate over the possible existence of an elastic component in muscle.
            Titin is the largest known protein, consisting of nearly 27,000 amino acids and having a molecular mass of 2,993 kD. The sequencing of titin was accomplished by Siegfried Labeit and Bernhard Kolmerer in 1995. (To accomplish this prodigious feat, Labeit and Kolmerer P_ieced together the full sequence from 50 overlapP_ing cDNA fragments!) Titin’s sequence is composed mainly (90%) of 244 repeats of the immunoglobulin (Ig) and fibronectin 3 (FN3) domains. In the center of the titin molecule is a novel protein motif (not known in other proteins), consisting of repeats of the sequence PEVK (proline-glutamate-valine-lysine). PEVK motifs may act as spring devices to pull muscles back into shape after they have been stretched and may also play a role in regulating the stiffness and elasticity of muscle fibers. In characteristically stiff muscles, such as cardiac muscle, the PEVK region is only 163 residues in length, whereas in the more elastic skeletal muscle the PEVK domain is over 2000 residues in length.
            Titin forms long, flexible, thin filaments in muscle fibers. A single titin filament spans 1000 nm in its relaxed state, and a titin filament under tension can stretch to a length of over 3000 nm! Titin filaments in muscle originate at the periphery of the M band and extend along the myosin filaments all the way to the Z line (Figure 17.21). They appear to function by linking the myosin filaments to the Z lines and by acting as a template to regulate the assembly of myosin filaments and the spacing of myosin monomers in the filaments. When myofibrils are stretched beyond the overlap of the thick and thin filaments, titin filaments passively generate tension. This effect is provided by a relatively small number of titin molecules. The ratio of myosin to titin filaments is approximately 24 to 1. With 300 myosin molecules per thick filament, only 6 or so titin filaments are present in each half of a myosin thick filament.

Figure 17.21   •  A drawing of the arrangement of the elastic protein titin in the skeletal muscle sarcomere. Titin filaments originate at the periphery of the M band and extend along the myosin filaments to the Z lines. These titin filaments produce the passive tension existing in myofibrils that have been stretched so that the thick and thin filaments no longer overlap and cannot interact. (Adapted from Ohtsuki, I., Maruyama, K., and Ebashi, S., 1986. Advances in Protein Chemistry 38:1-67.)

The Mechanism of Muscle Contraction

When muscle fibers contract, the thick myosin filaments slide or walk along the thin actin filaments. The basic elements of the sliding filament model were first described in 1954 by two different research groups, Hugh Huxley and his colleague Jean Hanson, and the physiologist Andrew Huxley and his colleague Ralph Niedergerke. Several key discoveries paved the way for this model. Electron microscoP_ic studies of muscle revealed that sarcomeres decreased in length during contraction, and that this decrease was due to decreases in the width of both the I band and the H zone (Figure 17.22). At the same time, the width of the A band (which is the length of the thick filaments) and the distance from the Z disks to the nearby H zone (that is, the length of the thin filaments) did not change. These observations made it clear that the lengths of both the thin and thick filaments were constant during contraction. This conclusion was consistent with a sliding filament model.

Figure 17.22   •  The sliding filament model of skeletal muscle contraction. The decrease in sarcomere length is due to decreases in the width of the I band and H zone, with no change in the width of the A band. These observations mean that the lengths of both the thick and thin filaments do not change during contraction. Rather, the thick and thin filaments slide along one another.

The Sliding Filament Model

The shortening of a sarcomere (Figure 17.22) involves sliding motions in opposing directions at the two ends of a myosin thick filament. Net sliding motions in a specific direction occur because the thin and thick filaments both have directional character. The organization of the thin and thick filaments in the sarcomere takes particular advantage of this directional character. Actin filaments always extend outward from the Z lines in a uniform manner. Thus, between any two Z lines, the two sets of actin filaments point in opposing directions. The myosin thick filaments, on the other hand, also assemble in a directional manner. The polarity of myosin thick filaments reverses at the M disk. The nature of this reversal is not well understood, but presumably involves structural constraints provided by proteins in the M disk, such as the M protein and myomesin described above. The reversal of polarity at the M disk means that actin filaments on either side of the M disk are pulled toward the M disk during contraction by the sliding of the myosin heads, causing net shortening of the sarcomere.

Albert Szent-Györgyi’s Discovery of the Effects of Actin on Myosin

The molecular events of contraction are powered by the ATPase activity of myosin. Much of our present understanding of this reaction and its dependence on actin can be traced to several key discoveries by Albert Szent-Györgyi at the University of Szeged in Hungary in the early 1940s. Szent-Györgyi showed that solution viscosity is dramatically increased when solutions of myosin and actin are mixed. Increased viscosity is a manifestation of the formation of an actomyosin complex.
            Szent-Györgyi further showed that the viscosity of an actomyosin solution was lowered by the addition of ATP, indicating that ATP decreases myosin’s affinity for actin. Kinetic studies demonstrated that myosin ATPase activity was increased substantially by actin. (For this reason, Szent-Györgyi gave the name actin to the thin filament protein.) The ATPase turnover number of pure myosin is 0.05/sec. In the presence of actin, however, the turnover number increases to about 10/sec, a number more like that of intact muscle fibers.
            The specific effect of actin on myosin ATPase becomes apparent if the product release steps of the reaction are carefully compared. In the absence of actin, the addition of ATP to myosin produces a raP_id release of H1, one of the products of the ATPase reaction:

            ATP^4- + H₂O ® ADP^3- + P_i^2- + H⁺

However, release of ADP and P_i from myosin is much slower. Actin activates myosin ATPase activity by stimulating the release of P_i and then ADP. Product release is followed by the binding of a new ATP to the actomyosin complex, which causes actomyosin to dissociate into free actin and myosin. The cycle of ATP hydrolysis then repeats, as shown in Figure 17.23a. The crucial point of this model is that ATP hydrolysis and the association and dissociation of actin and myosin are coupled. It is this coupling that enables ATP hydrolysis to power muscle contraction.

Figure 17.23   •  The mechanism of skeletal muscle contraction. The free energy of ATP hydrolysis drives a conformational change in the myosin head, resulting in net movement of the myosin heads along the actin filament. (Inset) A ribbon and space-filling representation of the actin-myosin interaction. (S1 myosin image courtesy of Ivan Rayment and Hazel M. Holden, ^{University of Wisconsin, Madison.)}

The Coupling Mechanism: ATP Hydrolysis Drives
Conformation Changes in the Myosin Heads

The only remaining P_iece of the puzzle is this: How does the close coupling of actin-myosin binding and ATP hydrolysis result in the shortening of myofibrils? Put another way, how are the model for ATP hydrolysis and the sliding filament model related? The answer to this puzzle is shown in Figure 17.23b. The free energy of ATP hydrolysis is translated into a conformation change in the myosin head, so that dissociation of myosin and actin, hydrolysis of ATP, and rebinding of myosin and actin occur with stepwise movement of the myosin S1 head along the actin filament. The conformation change in the myosin head is driven by the hydrolysis of ATP.
            As shown in the cycle in Figure 17.23a, the myosin heads — with the hydro-lysis products ADP and P_i bound — are mainly dissociated from the actin filaments in resting muscle. When the signal to contract is presented (see following discussion), the myosin heads move out from the thick filaments to bind to actin on the thin filaments (Step 1). Binding to actin stimulates the release of phosphate, and this is followed by the crucial conformational change by the S1 myosin heads — the so-called power stroke — and ADP dissociation. In this step (Step 2), the thick filaments move along the thin filaments as the myosin heads relax to a lower energy conformation. In the power stroke, the myosin heads tilt by approximately 45 degrees and the conformational energy of the myosin heads is lowered by about 29 kJ/mol. This moves the thick filament approximately 10 nm along the thin filament (Step 3). Subsequent binding (Step 4) and hydrolysis (Step 5) of ATP cause dissociation of the heads from the thin filaments and also cause the myosin heads to shift back to their high-energy conformation with the heads’ long axis nearly perpendicular to the long axis of the thick filaments. The heads may then begin another cycle by binding to actin filaments. This cycle is repeated at rates up to 5/sec in a tyP_ical skeletal muscle contraction. The conformational changes occurring in this cycle are the secret of the energy coupling that allows ATP binding and hydro-lysis to drive muscle contraction.
            The conformation change in the power stroke has been studied in two ways: (1) cryoelectron microscopy together with computerized image analysis has yielded low-resolution images of S1-decorated actin in the presence and absence of MgADP (corresponding approximately to the states before and after the power stroke), and (2) feedback-enhanced laser optical trapP_ing experiments have measured the movements and forces exerted during single turnovers of single myosin molecules along an actin filament. The images of myosin, when compared with the X-ray crystal structure of myosin S1, show that the long a -helix of S1 that binds the light chains (ELC and RLC) may behave as a lever arm, and that this arm swings through an arc of 23 degrees upon release of ADP. (A glycine residue at position 770 in the S1 myosin head lies at the N-terminal end of this helix/lever arm and may act as a hinge.) This results in a 3.5-nm (35-Å) movement of the last myosin heavy chain residue of the X-ray structure in a direction nearly parallel to the actin filament. These two imaging “snapshots” of the myosin S1 conformation may represent only part of the working power stroke of the contraction cycle, and the total movement of a myosin head with respect to the apposed actin filament may thus be more than 3.5 nm.

Control of the Contraction-Relaxation Cycle by Calcium Channels and Pumps

The trigger for all muscle contraction is an increase in Ca²⁺ concentration in the vicinity of the muscle fibers of skeletal muscle or the myocytes of cardiac and smooth muscle. In all these cases, this increase in Ca²⁺ is due to the flow of Ca²⁺ through calcium channels (Figure 17.24). A muscle contraction ends when the Ca²⁺ concentration is reduced by specific calcium pumps (such as the SR Ca²⁺ -ATPase, Chapter 10). The sarcoplasmic reticulum, t-tubule, and sarcolemmal membranes all contain Ca²⁺ channels. As we shall see, the Ca²⁺ channels of the SR function together with the t-tubules in a remarkable coupled process.

Figure 17.24  •  Ca²⁺ is the trigger signal for muscle contraction. Release of Ca²⁺ through voltage- or Ca²⁺-sensitive channels activates contraction. Ca²⁺ pumps induce relaxation by reducing the concentration of Ca²⁺ available to the muscle fibers.

            Ca²⁺ release in skeletal and heart muscle has been characterized through the use of specific antagonist molecules that block Ca²⁺ channel activity. The dihydropyridine (DHP) receptors of t-tubules, for example, are blocked by dihydropyridine derivatives, such as nifediP_ine (Figure 17.25). The purified DHP receptor of heart muscle can be incorporated into liposomes, whereupon it shows calcium channel activity. The channel displays voltage-dependent gating and is selective for divalent cations over monovalent cations. Thus, the heart muscle DHP receptor is a voltage-dependent Ca²⁺ channel. Other evidence suggests that the skeletal muscle DHP receptor is a voltage-sensing protein; it presumably undergoes voltage-dependent conformation changes.

Figure 17.25   •  The structures of nifediP_ine and ryanodine. NifediP_ine binds with high affinity to the Ca²⁺-release channels of t-tubules. Ryanodine binds with high affinity to the Ca²⁺ channels of SR terminal cisternae.

            The DHP receptor from t-tubules consists of five different polypeptides, designated a₁ (150 to 173 kD ), a₂ (120 to 150 kD), b (50 to 65 kD), d (30 to 35 kD), and d (22 to 27 kD). The a₂- and D -subunits are linked by a disulfide bond. The a₁, a₂- d, b, and g stoichiometry is 1;1;1;1. The a₂-subunit is glycosylated, but a1 is not. a1 is homologous with the a-subunit of the voltage-sensitive sodium channel (Chapter 34). The sequence of a₁ contains four internal sequence repeats, each containing six transmembrane helices, one of which is positively charged and is believed to be a voltage sensor (Figure 17.26). The loop between helices 5 and 6 contributes to the pore. These six segments share many similarities with the corresponding segments of the sodium channel. The a₁-subunit of the DHP receptor in heart muscle is implicated in channel formation and voltage-dependent gating.

Figure 17.26   •  The a ₁-subunit of the t-tubule Ca²⁺ channel/DHP receptor contains six peptide segments that may associate to form the Ca²⁺ channel. This Ca²⁺ channel polypeptide is homologous with the voltage-sensitive Na⁺ channel of neuronal tissue.

            The Ca²⁺-release channel from the terminal cisternae of sarcoplasmic reticulum has been identified by virtue of its high affinity for ryanodine, a toxic alkaloid (Figure 17.25). The purified receptor consists of oligomers, containing four or more subunits of a single large polypeptide (565 kD). Electron microscopy reveals that the purified ryanodine receptor (Figure 17.27) is in fact the foot structure observed in native muscle tissue.

Figure 17.27   •  (a) Electron micrograph images of foot structures of terminal cisternae. (b, c) Foot structures appear as trapezoids and diamonds on the surface of the membrane. The central canal (CC), radial canals (RC), and peripheral vestibules (PV) are indicated. (d) The relationship between the foot structures, t-tubule, terminal cisternae, and muscle fiber. (Photo courtesy of Sidney Fleischer, Vanderbilt University)

Image reconstructions reveal that the receptor is a square structure with fourfold symmetry , containing a central pore with four radially extending canals (Figure 17.28). These radial canals each extend to openings in the periphery of the structure and are therefore contiguous with the myoplasm.

Figure 17.28   •  Image reconstructions of the junctional channel complex of a foot structure. (Photo courtesy of Sidney Fleischer, Vanderbilt University)

            So how do the foot structures effect the release of Ca²⁺ from the terminal cisternae of the SR? The feet that join the t-tubules and the terminal cisternae of the SR are approximately 16 nm thick. The feet apparently function by first sensing either a voltage-dependent conformation change (skeletal muscle) or the transport of Ca²⁺ across the voltage-sensitive Ca²⁺ channel (heart muscle) of the t-tubule and then facilitating the release of large amounts of Ca²⁺ through the foot structure from the SR. The reconstructed image (Figure 17.28) for the foot structure suggests a possible pathway for Ca²⁺ transport from the lumen of the SR to the myoplasm via the ryanodine receptor. A Ca²⁺ - or voltage-dependent conformation change may serve to gate open the central canal of the foot structure. On entering the central canal, calcium ions move outward through the radial canals to the outer vestibule regions and into the myoplasm adjacent to the triad junctions, where binding to the muscle fibers induces contraction.

Regulation of Contraction by Ca²⁺

Early in this chapter, the importance of Ca²⁺ ion as the triggering signal for muscle contraction was described. Ca²⁺ is the intermediary signal that allows striated muscle to respond to motor nerve impulses (Figure 17.24). The importance of Ca²⁺ as a contraction signal was understood in the 1940s, but it remained for Setsuro Ebashi, a P_ioneer of muscle research, to show in the early 1960s that the Ca²⁺ signal is correctly interpreted by muscle only when tropomyosin and the troponins are present. Specifically, actomyosin prepared from pure preparations of actin and myosin (thus containing no tropomyosin and troponins) was observed to contract when ATP was added, even in the absence of Ca²⁺ . However, actomyosin prepared directly from whole muscle would contract in the presence of ATP only when Ca²⁺ was added. Clearly the muscle extracts contained a factor that conferred normal Ca²⁺ sensitivity to actomyosin. The factor turned out to be the tropomyosin-troponin complex.
            Actin thin filaments consist of actin, tropomyosin, and the troponins in a 7 ;1 ;1 ratio (Figure 17.15). Each tropomyosin molecule spans seven actin molecules, lying along the thin filament groove, between pairs of actin monomers. As shown in a cross-section view in Figure 17.29, in the absence of Ca²⁺ , troponin I is thought to interact directly with actin to prevent the interaction of actin with myosin S1 heads. Troponin I and troponin T interact with tropomyosin to keep tropomyosin away from the groove between adjacent actin monomers. However, the binding of Ca²⁺ ions to troponin C appears to increase the binding of troponin C to troponin I, simultaneously decreasing the interaction of troponin I with actin. As a result, tropomyosin slides deeper into the actin-thin filament groove, exposing myosin-binding sites on actin, and initiating the muscle contraction cycle (Figure 17.23). Because the troponin complexes can interact only with every seventh actin in the thin filament, the conformational changes that expose myosin-binding sites on actin may well be cooperative. Binding of an S1 head to an actin may displace tropomyosin and the troponin complex from myosin-binding sites on adjacent actin subunits.

Figure 17.29   •  A drawing of the thick and thin filaments of skeletal muscle in cross-section showing the changes that are postulated to occur when Ca²⁺ binds to troponin C.

The Interaction of Ca²⁺ with Troponin C

There are four Ca²⁺ -binding sites on troponin C — two high-affinity sites on the carboxy-terminal end of the molecule, labeled III and IV in Figure 17.30, and two low-affinity sites on the amino-terminal end, labeled I and II. Ca²⁺ binding to sites III and IV is sufficiently strong (K_D = 0.1 mM) that these sites are presumed to be filled under resting conditions. Sites I and II, however, where the K_D is approximately 10 mM, are empty in resting muscle. The rise of Ca²⁺ levels when contraction is signaled leads to the filling of sites I and II, causing a conformation change in the amino-terminal domain of TnC. This conformational change apparently facilitates a more intimate binding of TnI to TnC that involves the C helix, and also possibly the E helix of TnC. The increased interaction between TnI and TnC results in a decreased interaction between TnI and actin.

Figure 17.30   •  (a) A ribbon diagram and (b) a molecular graphic showing two slightly different views of the structure of troponin C. Note the long a-helical domain connecting the N-terminal and C-terminal lobes of the molecule.

The Structure of Cardiac and Smooth Muscle

The structure of heart myocytes is different from that of skeletal muscle fibers. Heart myocytes are approximately 50 to 100 m m long and 10 to 20 m m in diameter. The t-tubules found in heart tissue have a fivefold larger diameter than those of skeletal muscle. The number of t-tubules found in cardiac muscle differs from species to species. Terminal cisternae of mammalian cardiac muscle can associate with other cellular elements to form dyads as well as triads. The association of terminal cisternae with the sarcolemma membrane in a dyad structure is called a peripheral coupling. The terminal cisternae may also form dyad structures with t-tubules that are called internal couplings (Figure 17.31). As with skeletal muscle, foot structures form the connection between the terminal cisternae and t-tubule membranes.

Figure 17.31   •  Electron micrograph of a dog heart muscle. The terminal cisterna of the SR (TC-SR) is associated with the t-tubule (TT) by means of foot structures (FS), forming a dyad junction. MF indicates the location of myofilaments. LT-SR signifies the longitudinal tubule of the SR. (From Fleischer, S., and Inui, M., 1989. Annual Review of Biophysics and Biophysical Chemistry 18:333-364.)

            In higher animals, large percentages of the terminal cisternae of cardiac muscle are not associated with t-tubules at all. For SR of this type, Ca²⁺ release must occur by a different mechanism from that found in skeletal muscle. In this case, it appears that Ca²⁺ leaking through sarcolemmal Ca²⁺ channels can trigger the release of even more Ca²⁺ from the SR. This latter process is called Ca²⁺ -induced Ca²⁺ release (abbreviated CICR).

The Structure of Smooth Muscle Myocytes

The myocytes of smooth muscle are approximately 100 to 500 m m in length and only 2 to 6 m m in diameter. Smooth muscle contains very few t-tubules and much less SR than skeletal muscle. The Ca²⁺ that stimulates contraction in smooth muscle cells is predominantly extracellular in origin. This Ca²⁺ enters the cell through Ca²⁺ channels in the sarcolemmal membrane that can be opened by electrical stimulation, or by the binding of hormones or drugs. The contraction response time of smooth muscle cells is very slow compared with that of skeletal and cardiac muscle.

The Mechanism of Smooth Muscle Contraction

Vertebrate organisms employ smooth muscle myocytes for long, slow, and involuntary contractions in various organs, including large blood vessels; intestinal walls; and, in the female, the uterus. Smooth muscle contains no troponin complex; thin filaments consist only of actin and tropomyosin. DesP_ite the absence of troponins, smooth muscle contraction is dependent on Ca²⁺ , which activates myosin light chain kinase (MLCK), an enzyme that phosphorylates LC2, the regulatory light chain of myosin. Contraction of smooth muscle is initiated by phosphorylation of LC2, and dephosphorylation causes relaxation of smooth muscle tissue.
            The mechanism of this contraction process is shown in Figure 17.32. Smooth muscle myocytes have a resting [Ca²⁺] of approximately 0.1 mM. Electrical stimulation (by the autonomic or involuntary nervous system) opens Ca²⁺ channels in the sarcolemmal membrane, allowing [Ca²⁺] to rise to about 10 mM, a concentration at which Ca²⁺ binds readily to calmodulin (see Chapter 34). Binding of the Ca²⁺ -calmodulin complex to MLCK activates the kinase reaction, phosphorylating LC2 and stimulating smooth muscle contraction. Export of Ca²⁺ by the plasma membrane Ca²⁺-ATPase returns Ca²⁺ to its resting level, deactivating MLCK. Smooth muscle relaxation then occurs through the action of myosin light chain phosphatase, which dephosphorylates LC2. This reaction is relatively slow, and smooth muscle contractions are tyP_ically more sustained and dissipate more slowly than those of striated muscle.

Figure 17.32   •  A model for the control of contraction in smooth muscle. (Calmodulin is abbreviated CaM.)

            Smooth muscle contractions are subject to the actions of hormones and related agents. As shown in Figure 17.32, binding of the hormone eP_inephrine to smooth muscle receptors activates an intracellular adenylyl cyclase reaction that produces cyclic AMP (cAMP). The cAMP serves to activate a protein kinase that phosphorylates the myosin light chain kinase. The phosphorylated MLCK has a lower affinity for the Ca²⁺-calmodulin complex and thus is physiologically inactive. Reversal of this inactivation occurs via myosin light chain kinase phosphatase.

Human Biochemistry
Smooth Muscle Effectors Are Useful Drugs
The action of ePinephrine and related agents forms the basis of therapeutic control of smooth muscle contraction. Breathing disorders, including asthma and various allergies, can result from excessive contraction of bronchial smooth muscle tissue. Treatment with ePinephrine, whether by tablets or aerosol inhalation, inhibits MLCK and relaxes bronchial muscle tissue. More specific bronchodilators, such as albuterol (see	figure), act more selectively on the lungs and avoid the undesirable side effects of ePinephrine on the heart. Albuterol is also used to prevent premature labor in pregnant women, owing to its relaxing effect on uterine smooth muscle. Conversely, oxytocin, known also as Pitocin, stimulates contraction of uterine smooth muscle. This natural secretion of the Pituitary gland is often administered to induce labor.
	The structure of oxytocin.

17.4  •  A Proton Gradient Drives the Rotation of Bacterial Flagella

Bacterial cells swim and move by rotating their flagella. The flagella of Escherichia coli are helical filaments about 10,000 nm (10 mm) in length and 15 nm in diameter. The direction of rotation of these filaments affects the movements of the cell. When the half-dozen filaments on the surface of the bacterial cell rotate in a counterclockwise direction, they twist and bundle together and rotate in a concerted fashion, propelling the cell through the medium. (On the other hand, clockwise-rotating flagella cannot bundle together and under such conditions the cell merely tumbles and moves erratically.)
            The rotations of bacterial flagellar filaments are the result of the rotation of motor protein complexes in the bacterial plasma membrane. The flagellar motor consists of at least two rings (including the M ring and the S ring) with diameters of about 25 nm assembled around and connected rigidly to a rod attached in turn to the helical filament (Figure 17.33). The rings are surrounded by a circular array of membrane proteins. In all, at least 40 genes appear to code for proteins involved in this magnificent assembly. One of these, the motB protein, lies on the edge of the M ring, where it interacts with the motA protein, located in the membrane protein array and facing the M ring.

Figure 17.33   •  A model of the flagellar motor assembly of Escherichia coli. The M ring carries an array of about 100 motB proteins at its periphery. These juxtapose with motA proteins in the protein complex that surrounds the ring assembly. Motion of protons through the motA/motB complexes drives the rotation of the rings and the associated rod and helical filament.

            In contrast to the many other motor proteins described in this chapter, a proton gradient, not ATP hydrolysis, drives the flagellar motor. The concentration of protons, [H⁺], outside the cell is tyP_ically higher than that inside the cell. Thus, there is a thermodynamic tendency for protons to move into the cell. The motA and motB proteins together form a proton shuttling device that is coupled to motion of the motor disks. Proton movement into the cell through this protein complex or “channel” drives the rotation of the flagellar motor. A model for this coupling has been proposed by Howard Berg and his coworkers (Figure 17.34). In this model, the motB proteins possess proton exchanging sites — for example, carboxyl groups on aspartate or glutamate residues or imidazole moieties on histidine residues. The motA proteins, on the other hand, possess a pair of “half-channels,” with one half-channel facing the inside of the cell and the other facing the outside. In Berg’s model, the outside edges of the motA channel protein cannot move past a proton-exchanging site on motB when that site has a proton bound, and the center of the channel protein cannot move past an exchange site when that site is empty. As shown in Figure 17.34, these constraints lead to coupling between proton translocation and rotation of the flagellar filament. For example, imagine that a proton has entered the outside channel of motA and is bound to an exchange site on motB (Figure 17.34a). An oscillation by motA, linked elastically to the cell wall, can then position the inside channel over the proton at the exchange site (Figure 17.34b), whereupon the proton can travel through the inside channel and into the cell, while another proton travels up the outside channel to bind to an adjacent exchange site. The restoring force acting on the channel protein then pulls the motA/motB complex to the left as shown (Figure 17.34c), leading to counterclockwise rotation of the disk, rod, and helical filament. The flagellar motor is driven entirely by the proton gradient. Thus, a reversal of the proton gradient (which would occur, for example, if the external medium became alkaline) would drive the flagellar filaments in a clockwise direction. Extending this P_icture of a single motA/motB complex to the whole motor disk array, one can imagine the torrent of protons that pass through the motor assembly to drive flagellar rotation at a tyP_ical speed of 100 rotations per second. Berg estimates that the M ring carries 100 motB proton-exchange sites, and various models predict that 800 to 1200 protons must flow through the complex during a single rotation of the flagellar filament!

Figure 17.34   •  Howard Berg’s model for coupling between transmembrane proton flow and rotation of the flagellar motor. A proton moves through an outside channel to bind to an exchange site on the M ring. When the channel protein slides one step around the ring, the proton is released and flows through an inside channel and into the cell, while another proton flows into the outside channel to bind to an adjacent exchange site. When the motA channel protein returns to its original position under an elastic restoring force, the associated motB protein moves with it, causing a counterclockwise rotation of the ring, rod, and helical filament. (Adapted from Meister, M., Caplan, S. R., and Berg, H. C., 1989. Dynamics of a tightly coupled mechanism for flagellar rotation. Biophysical Journal 55:905-914)