INTRODUCTION

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Contraction in smooth muscle is driven by the cyclic ATP-driven interaction of myosin and actin. While these proteins are not organized in sarcomeres as they are in skeletal and cardiac muscle, both the actin and myosin are organized into filaments. However, the nature of these filaments is unique to smooth muscle. The actin filaments do not contain troponin. They contain a smooth muscle isoform of tropomyosin, as well as variable amounts of the proteins calponin and caldesmon, that may be involved in regulation, or at least modulation, of contraction (1). The "on-off" switch of smooth muscle is located on the myosin filament itself.

The myosin filaments of smooth muscle are organized in a side polar geometry, which is distinct from the bipolar geometry of the myosin filaments in striated muscle (2). The smooth muscle myosin II shares the same basic features of all myosin II molecules (Figure 1). Each molecule has two globular head domains that are the site of enzymatic activity (i.e., force generation). Within each head two myosin light chains bind to an extended alpha helix that has been proposed to function as a "lever arm," amplifying small movements generated within the myosin head (3, 4). Beyond the light chain domain, the myosin molecule dimerizes by virtue of an alpha-helical coiled-coil. The distal region of this coiled-coil contains sequences involved in filament formation. Smooth muscle myosin is regulated by phosphorylation of one of the two myosin light chains, the regulatory light chain. (Nonmuscle myosin II is regulated in the same manner.) As already noted, this is thought to be the "on- off" switch for smooth muscle contraction. A number of the details of this regulatory mechanism have recently been elucidated.

View larger version (27K): [in this window] [in a new window]   Figure 1.   Domains of the myosin II molecule. The myosin molecule can be divided into different functional pieces. As diagrammed in the figure, these are the myosin head (often referred to as the S1 fragment), the myosin rod (which can be further divided into a light meromyosin [LMM] and S2 fragment), and a soluble fragment containing the head and most of the rod (heavy meromyosin [HMM]).

	THE REGULATORY MECHANISM

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Smooth muscle myosin in vitro is inactive if its regulatory light chains (RLC) are not phosphorylated. Phosphorylation of the RLC in vivo occurs upon elevation of cytoplasmic calcium, which in turn binds to calmodulin. One of the target enzymes for calmodulin is myosin light chain kinase, which phosphorylates the RLCs of smooth muscle myosin, leading to activation of contraction (5) (Figure 2). Dephosphorylation involves a phosphatase whose activity can be modulated (6). The site of phosphorylation is a serine residue, 19 amino acids from the N-terminus. This region of the RLC is highly mobile (7). Mutagenesis studies suggest that a phosphate in this position has two distinct functions: (1) controlling monomer to filament transitions; and (2) turning on the actin-activated adenosine triphosphatase (ATPase) activity. First, the change in local charge may disrupt interactions with the myosin rod that are necessary for smooth muscle myosin to form a folded monomer that cannot participate in filament formation (transition from 10S to 6S conformation) (8). Mutagenesis studies that have removed and reversed positively charged residues that are immediately N-terminal of the phosphorylatable serine of the smooth muscle myosin II RLC demonstrate that the interactions that regulate the folded-to-extended conformational transition in smooth muscle myosin are distinct from those that control ATPase activity (9). However, this role of phosphorylation, while of potential importance for nonmuscle cells, may not have a role in smooth muscle where the myosin concentration is above the critical concentration for filament formation (12). Thus fully differentiated smooth muscle cells have been shown to contain stable myosin filaments, even in the total absence of any RLC phosphorylation (13).

View larger version (16K): [in this window] [in a new window]   Figure 2.   Activation of contraction via myosin regulatory light chain phosphorylation. Phosphorylation of smooth muscle myosin by the Ca2+-calmodulin-activated myosin light chain kinase (MLCK) leads to activation of myosin activity and contraction. Dephosphorylation of the regulatory light chain by phosphatase activity leads to deactivation of myosin activity.

The second role of RLC phosphorylation is to activate the enzymatically inactive state of smooth muscle myosin. Studies indicate that this involves interaction of the phosphate with specific residues in the C-terminal lobe of the light chain (14, 15). This activation of ATPase activity and motility, unlike folding, does not depend solely on alteration of net charge at the N-terminus but also has spatial constraints that are completely satisfied only by a phosphate moiety. Studies of chimeric RLCs composed of the N- or C-terminal half of each skeletal muscle myosin RLC (skRLC) and smooth muscle myosin RLC (smRLC) indicate that it is the C-terminal half of skRLC that lacks structural elements necessary for phosphorylation-mediated regulation (14). C-terminal truncation studies on the smooth RLC indicate that this region of the light chain must be present for tight binding, normal coupling between ATPase activity and motility and contains a region that is necessary to confer regulation (16, 17). Based on the phosphorylation-mediated regulation of glycogen phosphorylase (18), one study examined the possibility of residues in the C-terminus providing coordination of the phosphoryl serine in the N-terminal domain of the RLC (15). Substitution of four arginine residues that are present in the C-terminal half of the smRLC (conserved in all regulated myosin RLCs) but missing in nonregulated myosin RLCs was able to restore regulation to the skRLC. (While the skRLC has maintained the ability to inhibit the activity of smooth muscle myosin, regulation, i.e., activation, of activity via phosphorylation of the RLC has been lost.) The four arginines are located in four different domains: the E, F, G, and H helices, using calmodulin nomenclature for the helices, as revealed in the crystal structures of chicken skeletal myosin S1 (3) and the scallop myosin regulatory domain (19). A functional role in regulation for a subset of the missing arginine residues was postulated by analogy to the case of glycogen phosphorylase, wherein coordination of the phosphoserine is by arginines residing in different helices (18). Global rearrangement of helices is associated with the coordination of the phosphoserine. Based on the orientation of the corresponding residues in chicken skeletal structure (3), only the arginines found in helices E and H would appear to be positioned appropriately to coordinate the phosphate of the phosphoserine by hydrogen bonds. The coordination of the phosphoserine may cause conformational changes that lead to regulation of myosin through altered interactions that involve the two RLCs and the myosin heavy chain.

What is the nature of the interactions that lock myosin into an "off" state and that are overcome by structural rearrangements in the RLC that are brought about by its phosphorylation? To answer this question, one must first define the minimal number of domains of the smooth muscle myosin molecule that are necessary for complete regulation. Two studies have proven that regulation requires the presence of two heads (20, 21). The latter study (21) demonstrated that not only do the two heads have to dimerize in order to be regulated, but the minimal length of myosin rod that allows complete regulation is on the order of the length of the myosin head. This suggests a model in which complete regulation requires a number of weak interactions involving the light chain domains of the myosin heads, the myosin rod, and perhaps even regions within the motor domain of the myosin head. The inhibition is broken when the RLC confirmation is altered by phosphorylation, presumably breaking interactions between the myosin heads and rod.

	LATCH

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

While phosphorylation operates as a simple on-off switch at the level of the single molecule, the effect of phosphorylation is more complicated in smooth muscle cells. While it is generally thought that some level of phosphorylation is necessary to activate smooth muscle myosin, once activated the relationship between the level of phosphorylation and the degree of activation can be modulated. Low levels of phosphorylation can support high levels of force maintenance (22). However, under these conditions, the shortening speed of the muscle is greatly reduced as a function of slower actin-myosin cycling. This economical maintenance of force at the expense of shortening speed is termed "latch" (22). The basis of this remains unclear, but it would appear that dephosphorylated cross-bridges can cycle under some conditions (23). It has been suggested that a mechanism that interferes with the interaction of the myosin heads with the rod would lead to slow cycling of the dephosphorylated heads. Perhaps cooperative activation of the filament involves such a mechanism, or perhaps there are accessory proteins in smooth muscle that regulate these sorts of interactions. Whether any conditions exist in vivo that allow dephosphorylated crossbridges to cycle is an unanswered question. An alternative mechanism, suggested by Hai and Murphy (26), involves dephosphorylation of crossbridges after attachment but prior to detachment. However, this would require the alteration of the kinetics of attached crossbridges, which kinetic measurements on smooth heavy meromyosin (HMM) have failed to show.

Another feature of smooth muscle myosin that may or may not contribute to latch, but certainly contributes to the economy of smooth muscle contraction, was revealed by cryo-electron microscopy (4). Unlike fast skeletal muscle myosin (27), smooth muscle myosin undergoes a large rotation of its light chain binding domain (~ 23°) upon release of ADP (Figure 3). Since ADP release is necessary before ATP can bind and detach crossbridges, detachment of smooth muscle crossbridges involves an extra, strain-dependent step that is not found in fast skeletal muscle. Thus, under strain, ADP release will be slowed in smooth muscle, prolonging the crossbridge cycle and increasing the economy of force generation. Furthermore, if ADP levels rise during smooth muscle contraction, then the tendency to latch would be increased (28).

View larger version (103K): [in this window] [in a new window]   Figure 3.   Structural changes in the smooth muscle myosin head associated with ADP release. Shown side by side are three-dimensional reconstruction of cryo-electron micrographs of smooth muscle myosin bound to actin with ADP bound to the myosin (left) or with the myosin in rigor (right). Density associated with actin is labeled A, the motor domain of myosin is labeled Mt, the essential light chain of myosin is labeled E, and the regulatory light chain is labeled R.

	MYOSIN ISOFORMS OF SMOOTH MUSCLE

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Vertebrate smooth muscle cells contain a number of myosin isoforms, based on alternative exon splicing of two regions of the myosin heavy chain and via the expression of two variants of the essential light chain. These variations are not linked, and the expression of the isoform variants does not appear to be coordinately regulated. Variation in smooth muscle content of each of these isoforms has been correlated with differential contractile properties of smooth muscle (29). However, biochemical studies, to date, have only demonstrated functional alterations associated with the alternative insert in the myosin head. This insert adds seven amino acids to a flexible loop that is known as the 25/50 kilodalton (kD) loop (Figure 4).

View larger version (12K): [in this window] [in a new window]   Figure 4.   Amino acid sequences of the 25 /50 kD loops (underlined ) found in different smooth muscle myosins. The seven amino acid insert that is alternatively spliced into either the chicken or rabbit loop is shown in italics. The inserted form of rabbit smooth muscle myosin differs from the chicken and rat inserted form by two amino acids (bold ).

	25/50 kD LOOP

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Three studies have combined to provide definitive evidence that the insertion of seven additional amino acids into a flexible loop (the 25 / 50 kD junctional loop) on the surface of the smooth muscle myosin head near the nucleotide binding pocket results in increases in the ADP release rate, steady-state actin-activated ATPase activity, and the rate of actin filament sliding in an in vitro motility assay (30). Thus, the inserted (i.e., additional seven amino acids) isoform should allow smooth muscle to shorten at a higher speed when compared to the noninserted isoform. Data derived from mutagenesis of chicken smooth muscle myosin (32) show that the seven amino acid insert found in rabbit smooth muscle results in less of an increase than the insert found in rat and chicken smooth muscle (Figure 4) (Table 1).

	ESSENTIAL LIGHT CHAIN

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Two essential light chain isoforms are expressed in smooth muscle, LC17a and LC17b. As shown in Figure 5, they differ only in four amino acids that are near the C-terminus of the light chain. While correlations between the ratio of the two light chains and the contractile properties of smooth muscle have been noted (29, 30), interpretation of these correlations is complicated by the fact that there are variations in the inserted and noninserted form of the motor domain (i.e., differing 25 /50 kD loops) in smooth muscles. Biochemical studies that have varied the essential light chain (ELC) isoform on a common motor domain have failed to show any kinetic difference that is attributable to the light chain (30). However, additional studies must be performed under loaded conditions to ascertain whether the ELC isoform has a role in altering the contractile properties of smooth muscle myosin in vivo.

View larger version (4K): [in this window] [in a new window]   Figure 5.   C-terminal sequences of smooth muscle essential light chain (ELC) isoforms. The two essential light chain isoforms of smooth muscle myosin differ by four amino acids (highlighted in bold ) near the C-terminus of the light chains. The numbers above the sequences refer to the position relative to the initiation methionine, with 151 being the last amino acid in the ELC.

	ALTERNATIVE TAIL

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

The third and final isoform variation found in vertebrate smooth muscle myosin is at the end of the myosin rod (33). An alternatively spliced exon can insert a nonhelical tail at the end of the rod. The form with the additional tail is called SM1 (204 kD), while the form without the additional sequence is SM2 (200 kD) (33, 34). Such an addition could serve two possible functions. First, this piece may alter the type of filament that is formed by affecting the nature and/or stability of the assembly intermediates in filament formation. There is no evidence for such a role, but an analogous function has been attributed to the nonhelical tailpiece of a nonmuscle myosin (35). The role of the nonhelical tailpiece of the vertebrate nonmuscle myosin II rod was investigated using Escherichia coli to express rod fragments (35). The presence of the tailpiece greatly facilitated assembly. The analysis suggested that the nature of the effect was not via specific site interaction with the tailpiece, but more likely involved the tailpiece sterically blocking aggregations that are not productive for filament assembly. In altering the nature of the thick filaments in a muscle, the tail could have a major impact on the contractile properties of smooth muscle. Alternatively, it has been proposed that this tail may interact with the S2 segments of other myosin molecules in a filament, thus directly altering the myosin kinetics (36). Whether one or both of these mechanisms exist, correlations have been drawn between the shortening speed of smooth muscle and the amount of myosin that contains the longer tail (37).

	CONCLUSION

TOP ABSTRACT INTRODUCTION THE REGULATORY MECHANISM LATCH MYOSIN ISOFORMS OF SMOOTH MUSCLE 25/50 kD LOOP ESSENTIAL LIGHT CHAIN ALTERNATIVE TAIL CONCLUSION REFERENCES

Phosphorylation of the RLC of smooth muscle myosin triggers a cascade of structural rearrangements that allows activation of the enzymatic activity of the myosin. In vivo the majority of the smooth muscle myosin is in filamentous form. It is likely that the activation of the myosin in the filaments is highly cooperative, thus allowing high forces to be generated when a small percentage of the RLCs are phosphorylated. The shortening velocity and crossbridge cycling rate of smooth muscle decreases as phosphorylation levels fall, allowing economical maintenance of isometric force (the latch state). While isoform variations in smooth muscle myosin underlie some of the differential contractile properties of various smooth muscle tissues, the isoform differences likely result in less profound contractile differences than do alterations in the steady-state levels of RLC phosphorylation.