Marshall University Marshall Digital Scholar
Theses, Dissertations and Capstones
1-1-2002 Actin and Myosin Remodeling in the A7r5 Smooth Muscle Cell Michael E. Fultz [email protected]
Follow this and additional works at: http://mds.marshall.edu/etd Part of the Biochemistry, Biophysics, and Structural Biology Commons, and the Cell Biology Commons
Recommended Citation Fultz, Michael E., "Actin and Myosin Remodeling in the A7r5 Smooth Muscle Cell" (2002). Theses, Dissertations and Capstones. Paper 595.
This Dissertation is brought to you for free and open access by Marshall Digital Scholar. It has been accepted for inclusion in Theses, Dissertations and Capstones by an authorized administrator of Marshall Digital Scholar. For more information, please contact [email protected]. Actin and Myosin Remodeling in the A7r5 Smooth Muscle Cell
______
DISSERTATION
______
Submitted to the Graduate College of Marshall University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
______
by
Michael E. Fultz
Huntington, West Virginia July, 2002 Approval of Examining Committee
______Elsa I. Mangiarua, Ph.D.
______William D. McCumbee, Ph.D.
______Richard Niles, Ph.D.
______Donald Primerano, Ph.D.
______Date Gary L. Wright, Ph.D., Chair
Accepted by Graduate College
______Date Leonard J. Deutsch, Ph.D., Dean
ii Table of Contents
Acceptance Page...... i Signature Page...... ii Table of Contents...... iii Acknowledgments...... v List of Figures...... vi List of Tables...... ix
Page Chapter 1. General Introduction...... 1 Summary...... 1 Dissertation Organization...... 2 Literature Review...... 3 Differences between Smooth and Striated Muscle...... 3 Hypothesis to Explain Differences...... 6 Latch Hypothesis...... 6 Myosin Lability...... 7 Actin Remodeling...... 9 Smooth Muscle Actin...... 11 Actin isoforms in smooth muscle...... 12 Smooth Muscle Myosin...... 14 Smooth Muscle Myosin Heavy Chains...... 16 Smooth Muscle Myosin Light Chains...... 18 References...... 19
Chapter 2. Remodeling of the Actin Cytoskeleton in the Contracting A7r5 Smooth Muscle Cell ...... 30 Summary...... 30 Introduction...... 31 Materials and Methods...... 34 Results...... 37 Discussion...... 54 References...... 59
Chapter 3. Myosin Remodeling in the Contracting A7r5 Smooth Muscle Cell...... 64
iii Abstract...... 64 Introduction...... 65 Materials and Methods...... 67 Results...... 69 Discussion...... 85 References...... 88
Chapter 4. Correlation of Force Development with a-actin Remodeling in Smooth Muscle: Effects of Specific Protein Kinase Inhibitors...... 92 Abstract...... 92 Introduction...... 93 Materials and Methods...... 96 Results...... 100 Discussion...... 112 References...... 115
Chapter 5. General Conclusions...... 119 General Discussion...... 119 Future Research...... 122 References...... 123
Appendix A...... 126
Appendix B...... 127
iv Acknowledgments
To my advisor, Dr. Gary Wright, who taught me not only how to do science, but also how to think like a scientist, and is the best advisor one could ask for...
To my doctoral committee for their advice and suggestions...
To my parents for their constant encouragement and support...
To my sister for advice and support...
To my wife, Crystal, for keeping me maintain my sanity, and for the constant love and support...
Thank you all for making this possible.
v List of Figures
Chapter 2.
Figure 1. The contractile response of A7r5 cells to phorbol 12,13dibutyrate (PDB) and $ the calcium ionophore A23187 as visualized by -actin-GFP expression...... 41
Figure 2. The magnitude of decrease in cell length in A7r5 cells at 30 minutes after the addition of phorbol 12,13 dibutyrate (PDB) or the calcium ionophore A23187...... 42
Figure 3. Time course of PDBu contraction of an A7r5 cell showing the shortening of individual $-actin stress cables...... 43
Figure 4. The effect of cytochalasin B on the TRITC-phalloidin labeled actin cytomatrix of unstimulated and phorbol 12,13 dibutyrate (PDB)-stimulated A7r5 cells...... 44
Figure 5. The effect of cytochalasin-induced dissolution of the actin cytomatrix on cell length in the precontracted A7r5 cell...... 45
Figure 6. A comparison of unstimulated A7r5 cells to cells stimulated with phorbol 12,13 dibutyrate (PDB) for 30 minutes and then stained with TRITC-labeled phalloidin...... 46
Figure 7. A comparison of $-actin structure in unstimulated and phorbol 12, 13 dibutyrate (PDB) stimulated (A’B’) A7r5 cells...... 47
Figure 8. Structure of the "-actin cytomatrix in unstimulated A7r5 cells and those in treated with phorbol 12,13 dibutyrate (PDB) for 10 minutes, and 30 minutes...... 48
Figure 9. Micrographs at magnification 600X and 900X showing the association of "- actin peripheral dense structures with remnant "-actin stress cables...... 49
Figure 10. 4.0 micron increment Z-plane series of "-actin from the cell base toward the cell apex...... 50
Figure 11. Control experiments comparing "-actin remodeling in unstimulated cells; cells treated with phorbol 12,13 dibutyrate; cell treated with the inactive phorbol ester, 4"-phorbol 12,13-didecanoate; and cells treated with norepinephrine...... 51
Figure 12. A comparison of the distribution of "-actinin in unstimulated cells and cells
vi stimulated by phorbol 12,13 dibutyrate for 10 or 30 minutes...... 52
Figure 13. Dual immunostaining for "-actin and "-actinin in unstimulated and phorbol 12,13 dibutyrate-treated (PDB) A7r5 cells...... 53
Chapter 3.
Figure 1. Comparison of the "-actin, $-actin, and myosin cytomatrix in unstimulated and PDBu stimulated A7r5 cells...... 73
Figure 2. Time course of change in myosin localization in the A7r5 cell at intervals up to 60 minutes after stimulation by PDBu...... 74
Figure 3. Dual immunostaining for "-actin and myosin in unstimulated A7r5 cells...... 75
Figure 4. Dual immunostaining for "-actin and myosin in A7r5 at 5 minutes after stimulation with PDBu...... 76
Figure 5. Dual immunostaining for "-actin and myosin in A7r5 cells at 10 minutes after stimulation with PDBu...... 77
Figure 6. Dual immunostaining for "-actin and myosin in A7r5 cells at 30 minutes after stimulation with PDBu...... 78
Figure 7. Dual immunostaining for "-actin and myosin in A7r5 cells at 60 minutes after stimulation with PDBu...... 79
Figure 8. Control experiments comparing myosin distribution in the A7r5 cell prior to stimulation; addition of 10-8M phorbol 12, 13 dibutyrate; or the inactive 4-"-phorbol 12, 13 dideconate (4" PDD)...... 80
Figure 9. Confocal micrographs showing the effect of cytochalasin B on "-actin structure in A7r5 smooth muscle cells...... 81
Figure 10. Confocal micrographs showing the effect of cytochalasin B on the localization of myosin in A7r5 cells...... 82
Figure 11. Micrographs showing the effect of incubation with 2 X 10-6M thapsigargin on "-actin structure in A7r5 smooth muscle cells...... 83
vii Figure 12. Micrographs showing the effect of incubation with 2 X 10-6M thapsigargin on the localization of myosin in A7r5 smooth muscle cells...... 84
Chapter 4.
Figure 1. Tracings from contractile response of rat aortic rings in response to 10-8M PDB, showing the effects of kinase inhibitors...... 103
Figure 2. The effect of genistein, staurosporine, KN-93, and bisindolymaleimide I on the phorbol induced remodeling of "- actin...... 104
Figure 3. Regression analysis of maximal tension development in rat aortic rings and percent control of A7r5 cells showing "-actin remodeling...... 105
Figure 4. Regression analysis of deterioration in force maintenance in rat aortic rings and percent of A7r5 cells showing "-actin remodeling...... 106
Figure 5. Regression analysis of maximum tension developed in rat aortic rings and percent of A7r5 cells showing "-actin remodeling after genistein treatment...... 107
Appendix A
Figure 1. Proposed model explaining force development and maintenance of tension in smooth muscle...... 126
viii List of Tables
Chapter 4.
Table 1. Concentrations of different kinase inhibitors employed in present studies...... 108
Table 2. Effects of protein kinase inhibitors on early force development in rat aortic rings...... 109
Table 3. Effect of genistein on early force development and tension maintenance in rat aortic rings...... 110
Table4. Effects of kinase inhibitors on PDBu-induced "-actin remodeling in A7r5 cells ...... 111
ix Chapter 1. General Introduction
Summary
Active remodeling of the cytoskeleton has been proposed to contribute to the low energy cost and maintenance of the sustained contraction in smooth muscle. Using confocal microscopy and standard immunohistochemical techniques, direct observation of actin remodeling was studied in the contracting A7r5 cell in response to the diacylglycerol
(DAG) analog phorbol 12, 13 dibutyrate (PDBu). $-actin was shown to exist in the resting cell as parallel stress cables that extend across the cell. Stimulation by PDBu resulted in a sustained contraction that occurred in approximately eighty-five percent of the A7r5 cells.
The initial contraction was not uniform, but primarily occurred in the axis parallel to the
$-actin cables. During the latter portion of contraction shortening occurred in all axis.
During contraction, the cells shortened without apparent compression of the $-actin fibers or new cable formation, suggesting that $-actin fibers undergo active remodeling during contraction. Treatment of the cells with cytochalasin resulted in dissolution of fibers in the central region of the cell and elongation of precontracted cells in the axis parallel to the
$-actin fibers, suggesting that the $-actin cytoskeleton may serve to hold the cell in the contracted state. In unstimulated cells, "-actin exists as series of stress fibers similar in appearance to $-actin fibers. In contrast, within ten minutes of PDBu stimulation "-actin disassembled and reassembled into intensely-staining column-like structures that extend from the cell’s base. The "-actin column-like structures demonstrate a strong
1 colocalization with "-actinin suggesting that they may be homologous to dense bodies in differentiated smooth muscle. Smooth muscle myosin also undergoes remodeling, in a manner very similar to "-actin. Dual immunostaining for "-actin and smooth muscle myosin indicate a strong colocalization of the two proteins in the resting cell. However, following PDBu stimulation there was a partial dissociation of myosin from "-actin.
Treatment of cells with cytochalasin B or dissolution of the "-actin fibers with thapsigargin resulted in similar patterns of staining for "-actin and myosin. This suggests that myosin is in association with "-actin and follows changes in "-actin structure during contraction. This work suggests that the "- and $-actin domains in the A7r5 cells undergo specific remodeling in A7r5 cells, and suggests a model in which $-actin filaments undergo remodeling to hold the cell in the shortened state at low energy cost while "-actin and myosin remodel into a configuration that allows maximal mechanical advantage for contraction.
Dissertation Organization
This dissertation is divided into five chapters. This first chapter presents a summary of the dissertation and a general literature review. The next three chapters are manuscripts either published, or prepared for publication and therefore appear in the style of the indicated journal. Chapter two is a manuscript that describes and discusses differential remodeling of actin isoforms in the A7r5 smooth muscle cell. Chapter three details myosin remodeling and its association with the actin cytoskeleton. Chapter four
2 details the effects of specific kinase inhibition on the remodeling phenomenon in order to begin the determination of the biochemical pathways involved in cytoskeletal remodeling.
The final chapter contains a general discussion of the results from the three manuscripts presented in the dissertation as well as recommendations for further research.
Literature Review
Smooth muscle was named due to the lack of striations when visualized by light microscopy. Smooth muscle tissue exists as a closely packed arrangement of the fusiform smooth muscle cells that contain a central nucleus. Smooth muscle is present in several organ systems throughout the body: gastrointestinal, ocular, reproductive, respiratory, urinary, and vascular. The function of smooth muscle is largely directed toward transport or storage. However, this diverse tissue does not function merely as a rigid conduit for transport. This is exemplified by the very dynamic contractile/relaxant responses of smooth muscle of the vasculature in the regulation and maintenance of blood pressure throughout the body. There are many unexplained properties and phenomenon that have been observed in this remarkable tissue.
Differences in Smooth and Striated Muscle
It is widely thought that the basic biochemical and biophysical interactions of actin and myosin are similar in smooth, skeletal and cardiac muscle, as described by the sliding filament theory. However, there are several observations that bring into question the direct
3 applicability of the sliding filament theory in smooth muscle. In striated muscle, active force development is in direct proportion to the overlap of actin and myosin filaments
(Gordon et al., 1966). Smooth muscle, which lacks a typical sarcomere arrangement, is capable of generation of force that is as great as (Murphy, 1976) or greater than (Gabella,
1976) skeletal muscle. Smooth muscle actin and myosin filaments are not as ordered as striated muscle, and appear largely random in distribution within the cells (Rice et al.,
1970; Somylo, 1980). It has been shown that actin filaments may be arranged in bundles that allow up to 15 actin filaments to interact with each myosin filament. The myosin content, however, may be only 20% that of skeletal muscle (Murphy et al., 1974; Murphy
1976). These differences in structure and contractile protein content suggest that there may be mechanisms in smooth muscle in addition to sliding filament that provide biochemical or mechanical advantages during smooth muscle contraction.
In addition to structural differences, smooth muscle demonstrates the unique ability to slowly develop tension, and then maintain the tension for a considerable time period.
This maintenance of tension occurs at energy levels that are about 0.35% that of tetanized striated muscle (Paul, 1983). The principal biochemical and biophysical events underlying the initiation of contraction in smooth muscle have been observed to be similar to striated muscle. Stimulation of smooth muscle results in an elevation of intracellular calcium that leads to activation of calmodulin. The calcium-calmodulin complex then activates myosin light chain kinase which results in phosphorylation of Ser 19 on the myosin light chains.
Upon phosphorylation of the myosin light chains, myosin ATPase activity, and subsequent myosin/actin crossbridge formation and cycling occur. However, it has been shown that
4 slow force development, maintenance of tension (Driska et al., 1981), and velocity of shortening (Merkel et al., 1990) are disassociated from myosin light chain phosphorylation and myosin ATPase activity (Zhang and Moreland, 1994) in smooth muscle. Also noteworthy, it has been demonstrated that the phorbol ester-induced contraction can occur in the absence of a detectable increase of intracellular Ca2+ (Nakajima et al., 1993) or concomitantly increased myosin light chain phosphorylation (Singer and Baker, 1987).
A number of protein kinases not presently associated with the established smooth muscle pathway of contraction have been shown to alter the contractile response. It has been shown that Protein Kinase C (PKC) is tonically activated in response to a variety of contractile agonists (Haller et al., 1990; Singer et al., 1992) whereas, the inhibition or down-regulation of this enzyme blocks or attenuates the contractile response to different agonists (Merkel et al., 1991; Shimamoto et al., 1993; Wright and Hurn, 1994). Also of interest is the finding that the inhibition of tyrosine kinase (Di Salvo et al., 1993 ; Di Salvo et al., 1994; Gokita et al., 1994; Sriuasta and St-Louis, 1997; Tolloczko et al., 2000;
Duarte et al., 1997) blocks contraction in smooth muscle to various agonists. These findings suggest an important role of this kinase in electromechanical coupling or the direct activation of the contractile apparatus by Ca2+ in the cell (Di Salvo et al., 1993; Di
Salvo et al., 1994; Tolloczko et al., 2000; Gokida et al., 1994). Finally, Rokolye and
Singer (2000) have shown that the inhibition of calmodulin kinase II (CAMK II) has variable effects on early force development and tension maintenance depending on the agonist used to activate contraction. They concluded that KN-93, a calmodulin kinase II inhibitor, may play a role in regulating maintained tension in activated smooth muscle and
5 that this inhibition could be abolished by activation of PKC.
Hypotheses to Explain Differences
Several hypotheses have been developed to explain specific smooth muscle properties. Van Buren and colleagues (1994,1995) proposed that smooth muscle myosin may compensate for low density by generation of crossbridge forces that are 3 to 4 times those produced by skeletal muscle under loaded conditions. An increase in the “duty cycle” or the time that myosin remains in contact with actin has also been suggested
(Uyeda et al., 1990). Others have proposed that slow tension development reflects a balance between MLCK and protein phosphatase in the phosphorylation
/dephosphorylation of myosin light chains (Somylo et al., 1989). The identification of a specific myosin light chain phosphatase (Shirazi et al., 1994) lends evidence for this hypothesis. Still others have suggested that specific proteins may regulate the maintained contraction (Lash et al., 1986; Hemric and Chaloric, 1988; Adam et al., 1989). This review will focus on three mechanisms: latch hypothesis, myosin lability, and actin remodeling.
Latch Hypothesis
The latch hypothesis was first proposed by Dillon and colleagues (1981) and further characterized by Hai and Murphy (1988, 1989) and Murphy (1989). The hypothesis suggests a “latch” state in which dephosphorylated myosin cross-bridges remain bound to
6 actin and cycle slowly to account for slow tension generation and force maintenance. Hai and colleagues (1991) describe four states in which myosin can exist: (M), unattached myosin; (AM), myosin bound to actin; (Mp), unbound myosin with phosphorylated MLC
20; and (AMp), phosphorylated myosin bound to actin. The initial contraction is dependent on high intracellular calcium and subsequent activation of myosin light chain kinase (MLCK) and phosphorylation of the regulatory MLC. This results in (Mp) which quickly binds actin to form (AMp). This results in a rapid initial shortening velocity of contraction. As calcium levels and myosin light chain phosphorylation decrease, there is an increase in (AM) which maintains developed force while small amounts of myosin remain in the (AMp) state and contribute to slow velocity development. Mathematical models have been developed that are consistent with the latch hypothesis (Walker et al.,
1994), however, there is little direct evidence for this hypothesis. Recently, investigators demonstrated that the slowing of velocity occurs when phosphorylation levels of myosin are still rising which strongly argues against the validity of the latch hypothesis (Mitchell et al., 2001).
Myosin Lability
Schoenberg (1969) and Rice et al. (1970) first suggested thick filament evanescence in smooth muscle based on electron microscopy and the observation that myosin filaments appeared to be dissolved in the resting cell and formed upon contraction. Several other studies utilizing electron microscopy have indicated that myosin filaments increase in length and number in response to contraction in the anococcygeus muscle but not in the
7 taenia coli (Gillis et al., 1988, Godfrain-Debecker and Gillis, 1988; Xu et al., 1997) suggesting that in some smooth muscle types myosin filament structure may be stable and not labile. However, studies utilizing electron microscopy must be examined carefully as fixation techniques may alter thick filament structure (reviewed in Somlyo and Somlyo,
1992).
The series-to-parallel theory suggests that during smooth muscle contraction, there is a plasticity in the myosin heavy chains. Smooth muscle can adapt to different muscle lengths by variation of the number of filaments in series (Kuo et al., 2001). Lengthening of the myosin fibers will cause a series-to-parallel transition, as numerous filaments in series remodel into longer filaments in parallel (Pratusevich et al., 1995; Seow et al.,
2000).
In the original description of sliding filament theory it was proposed that muscles with long sarcomeres would generate more force and have a slower velocity compared to short sarcomeres (Huxley and Neidergerke, 1954). This inverse relationship between force and velocity is supported by investigation of arthropod muscle which containes sarcomeres of different lengths (Jasper and Pezard, 1934; Jahromi and Atwood, 1969).
Seow et al. (2000) demonstrated that from two seconds after the start of contraction to the tissue force plateau, velocity decreases are exactly matched by force increases when force was corrected for changes in activation as assessed by maximum power. They also concluded that a twofold change in velocity can be explained mathematically by a series- to-parallel change in thick filament orientation without changing the cross-bridge cycling rate. Calculations indicate that a fifty-five percent increase in myosin filament length
8 could accomplish this, which is in agreement with the sixty percent increase in filament density observed by Godfrain-Debecker and Gillis (1988) in contracted cells.
Myosin filament density has also been shown to decrease by twenty percent in response to cyclic stretch and then recover to preoscillation levels during a subsequent tetani (Kuo et al., 2001). It is suggested that mechanical strain facilitates depolymerization of myosin filaments and its subsequent reorganization to allow optimal interaction with the thin filament over a range of lengths.
Actin Remodeling
Recently, considerable interest has focused on the role of actin cytoskeletal organization on smooth muscle contractile properties. The ability to study the effects of actin remodeling was first made possible by isolation of fungal metabolites which inhibit actin polymerization. Cytochalasins, from the Greek cyto, cell and chalasis, relaxation, were isolated from cultures of Helminthosporium dematioideum and Metarchizium anisoplia (Carter, 1967). Cytochalasins were shown to induce dissolution of thin filaments (Schroeder, 1970; Wessels et al., 1971) and these microfilaments were later determined to be actin filaments (Spudich and Lin, 1972; Spudich, 1972). Cooper (1987) determined that the molecular mechanism of cytochalasin action was through binding to the barbed or growing end of the actin filament and preventing the addition of subsequent monomers. Cytochalasins have been shown to inhibit smooth muscle contraction with no effect on calcium currents, myosin light chain phosphorylation, or myosin ATPase activity in the K+ deploarized tissue (Obara and Yabu, 1994).
9 Wright and Hurn (1994) first demonstrated that the use of cytochalasin to inhibit actin polymerization caused a selective blockade of the slow tension increase in rat aortic smooth muscle. They suggested that the slow tension development by smooth muscle required a capability for dynamic remodeling of a portion of the actin cytoskeleton during the interval of contraction. Their conclusion has been supported by recent studies (Mehta and Gunst, 1999) showing that inhibition of actin polymerization by latrunculin-A, which complexes with free monomers and inactivates G-actin (Coue et al., 1987), depressed force development without affecting myosin light chain phosphorylation. Saito and colleagues (1996) demonstrated that cytochalasins can block the contractions induced by
K+ deploarization , norepinepherine, and the Ca2+-independent PDB contraction with no detectable effect on MLC phosphorylation or myosin ATPase activity. Other work
(Wright and Battistella-Patterson, 1998) has shown that inhibition of actin polymerization blocks calcium-dependent stress relaxation in vascular smooth muscle, suggesting that active remodeling of the actin cytoskeleton may also serve to decrease cellular tension following abrupt stretch. Contraction of isolated tracheal smooth muscle has been shown to decrease the amount of G-actin in the cell (Mehta and Gunst, 1999) suggesting increased filament formation. The addition of an actin stabilizing factor, phalloidin, was shown to induce a decrease in isometric force generation and stiffness and increased actomyosin ATPase activity with no change in myosin light chain phosphorylation levels or maximum velocity of contraction (Jones et al., 1999) suggesting that stabilization of the actin filaments increased tension costs. Recently, Barany et al. (2001) demonstrated that even in resting tissue actin is dynamic with polymerization/depolymerization occurring
10 and that stabilization may take place in the plateau of contraction. Taken together, these results suggest that actin polymerization is essential for the contractile and stress relaxant properties of smooth muscle, and that the effects of blocking polymerization cannot be attributed to a simple loss in cytoskeletal structure but more accurately reflects inhibition of dynamic actin cytoskeletal remodeling.
Smooth Muscle Actin
Actin in smooth muscle is structurally similar to actin found in other tissue types.
It has the same molecular weight, ability to polymerize, and ability to bind myosin and activate myosin ATPase (Bray, 1972). The actin content in smooth muscle is thirty to fifty milligrams per gram wet weight or thirty to fifty percent of the noncollagenous proteins present which is 0.9-1.6 millimolar in situ (Murphy et al., 1977; Hartshore, 1987). In muscle, actin exists in two forms. The basic unit is the monomeric globular G-actin which has a molecular weight of 42 KDa (Hartshore, 1987). G-actin undergoes polymerization to form filamentous F-actin and this polymerization, as in skeletal muscle, is dependent on actin concentration, ambient temperature, and cell ionic conditions (Strzelecka-
Golaszewska et al., 1980). Actin monomers contain a divalent cation binding site as well as an adenosine nucleotide binding site. If either of the binding sites are unoccupied, the actin denatures. Polymerization occurs when ATP binds G-actin and is subsequently
hydrolyzed to ADP and Pi and the ADP remains bound to F-actin (Estes et al., 1992). The divalent cation site can bind calcium or magnesium and while in monomeric form the cations can exchange freely. However, after G-actin monomers polymerize, the cation is
11 trapped within the protein (Strzelecka-Golaszewska et al., 1980; Barany et al., 1992).
Smooth muscle and skeletal muscle actin are similar in their ability to activate myosin ATPase as demonstrated by the finding that F-actin purified from either muscle type showed no differences in activation (Marston and Smith, 1984). Also, Harris and
Warshaw (1993) have demonstrated utilizing an in vitro motility assay that there was little functional difference in smooth and skeletal actin and that motility depended on the type of actin used. The rigor complex of skeletal or smooth muscle F-actin and myosin is similar with identical stochiometry (Prochniewicz and Strzelecka-Golaszewska, 1980).
Actin Isoforms in Smooth Muscle
Actin isoforms demonstrate a ninety-five percent homology in their 375 amino acid structure (Pollard and Cooper, 1986) with the most variability between isoforms being in the amino-terminus (Herman, 1993; Vandekerckhove and Webber, 1981). The amino- terminus has been shown to be important structurally and is probably responsible for specific isoform function (Cook et al., 1992). In mammalian tissues, six actin isoforms have been identified: "-skeletal, "-cardiac, "-vascular, (-enteric, (-cytoplasmic, and $- cytoplasmic with each of the proteins expressed from separate genes (Vandekerckhove and
Webber, 1981; Reddy et al., 1990). Four actin isoforms have been identified in vertebrate smooth muscle: "- and (- contractile actin and $- and (- cytoskeletal actin (Kabsch and
Vandekerckhove, 1992). The smooth muscle isoforms have been shown to be compartmentalized, and this compartmentalization may reflect a segregation according to
12 function (Small, 1995). Staining of the smooth muscle cell with an anti-$-actin antibody indicated a diffuse network throughout the cell while "- and (-actin staining suggested that these isoforms form stress fibers in association with myosin filaments (North et al.,
1994). Actin isoform expression does not appear to be species specific, however tissue specific expression has been noted. Visceral smooth muscle has selective expression of (- actin while in vascular smooth muscle the predominate contractile isoform is "-actin
(Fatigati and Murphy, 1984). Alpha-actin expression is considered to be associated with tonic smooth muscle while (-actin is found in phasic smooth muscle. The $-isoform of actin has been found in all smooth muscles examined and appears to be associated with the structural cytoskeleton (North et al., 1994). Isoform expression also changes with development as cytoplasmic actin expression is decreased with an increase in either "- or
(-actin in maturing tissue (Eddinger and Murphy, 1991; Hirai and Hirabayashi, 1983).
The recent development of the smooth muscle "- actin null mouse (Schildmeyer et al.,
2000) has provided several insights into actin isoform specific function. Complete knockout of the smooth muscle "-actin transcript and protein did not have an apparent effect on development of the cardiovascular system. a-Actin knockout mice did not have any apparent difficulty in feeding and reproduction. It was concluded that this was to be expected as the (-actin isoform is predominant in these tissues. However, aortas from the knockout mice did not have the ability to contract in response to potassium depolarization.
Furthermore, reduced blood pressure and blood flow was detected in the a-actin null mouse. Surprisingly, skeletal muscle "-actin expression was activated in these animals, which may help to compensate the loss of smooth muscle a-actin (Schildmeyer et al.,
13 2000).
Smooth Muscle Myosin
The smooth muscle myosin molecule consists of six polypeptide chains: a combination of two heavy chains, two regulatory light chains, and two essential alkali light chains. The two heavy chains form a dimer which comprises the main body of the molecule. The amino terminus of each heavy chain exists as a globular head region that is connected by an "-helical coiled rod. The motor activity of the myosin molecule is contained within the globular head, including the ATP and actin binding sites. Each globular head has an associated regulatory and an essential light chain. The carboxy terminus of the heavy chains end in a short amino acid sequence that is not believed to exhibit an "-helical structure (Reviewed in Gunst, 1999). Enzymatic cleavage of the myosin molecule produces the soluble S-1 subfragment and the rod portion. The S-1 fragment consists of the heavy chain globular heads, and the associated regulatory and essential light chains (Ikebe and Hartsthorne,1985; Adelstein and Sellers, 1996) and has proven invaluable in the study of myosin function.
The control of assembly of monomeric myosin molecules to myosin filaments was determined to reside in a carboxy-terminal fragment of the SM-MHC comprised of 150 amino acid residues (Cross and Vandekerckhove, 1986). New evidence suggests that 28 residues located in the C-terminus end of the coiled-coil domain of the myosin heavy chain are the critical residues mitigating myosin filament formation (Ikebe et al., 2001). It has been suggested that a pool of monomeric myosin may be recruited to form myosin
14 filaments as myosin filament density has been shown to increase in some contracted smooth muscle tissues (Xu et al., 1997;Watanabe et al., 1993; Gillis et al., 1988).
Increases in excess of 61 myosin filaments per square micron were reported in rat annococcygeus smooth muscle upon contraction (Gillis et al., 1988). Others have reported increases in myosin filament densities ranging from 23% (Xu et al., 1997) to 106%
(Watanabe et al., 1993) in contracted muscle. In contrast, Horowitz and colleagues (1994) detected only minute amounts of monomeric myosin in unstimulated muscle and concluded that assembly/disassembly is unlikely to play a role in contraction. Other experimental approaches have implicated myosin assembly/disassembly during contraction. Phosphorylation of the 20KDa myosin light chain has been shown to favor filamentous actin (Trybus, 1994; Suzuki et al. 1978). While dephosphorylation of the of the MLC-20 may lead to disassembly of filaments, dephosphorylated filaments exist in a relaxed smooth muscle (Somlyo et al., 1981). In contract to other reports, Fultz et al.
(unpublished data) directly observed myosin filament disassembly during contraction and subsequent reassembly of filament structure in association with alpha-actin at the termination of contraction in cultured smooth muscle cells. Their finding of closely associated actin/myosin remodeling is supported by observations indicating that F-actin can induce polymerization of monomeric myosin, and the selective disruption of F-actin by sevrin resulted in an immediate disassembly of the myosin filaments (Applegate and
Pardee, 1992). Intracellular calcium levels have also been linked to myosin filament formation. An increase in intracellular calcium induced a 144% increase in fiber density; whereas, in the absence of intracellular calcium, myosin filament density decreased by
15 35% and returned to control levels upon the addition of physiological saline (Herrera et al.,
2002). However, it is also reported that direct calcium binding by myosin leads to a less ordered structure that is more pronounced with unphosphorylated myosin (Podlubnaya et al., 1999). Myosin heavy chain isoform composition has also been shown to play a role in filament formation as SM1 and SM2 MHCs demonstrate variable solubilities and different modes of molecular packaging suggesting that uncoiled tail-pieces may also play a role in filament formation (Rovner et al. 2002).
Smooth Muscle Myosin Heavy Chains
Three distinct types of myosin heavy chains have been shown to be expressed in smooth muscle. Two of these are non-muscle myosin heavy chains (NM-MHC) and the third is smooth muscle specific myosin heavy chain (SM-MHC). The NM-MHCs are separate gene products designated NM-MHCA and NM-MHCB and have a molecular mass of 196KDa and 198-200KDa (Morano, 1992). Alternate splicing 39 base-pairs of the SM-MHC at the 3' end produces two isoforms of SM-MHC that differ in the length of their carboxy terminus, the 204 KDa SM1 and the 200 KDa SM-2 (Nagai et al., 1989;
Babij and Periasamy, 1986). Expression of these isoforms have been shown to be equal in some smooth muscle tissues (Kelley et al., 1991; Kelley et al., 1993) but others have reported unequal expression of the two isoforms (Rovner et al., 1986; Eddinger and Meer,
1997) and highly variable expression between smooth muscle cells from the same tissue
(Eddinger and Meer, 1997). Changes in expression of SM-1 and SM-2 occur during development. The relative amount of SM-2 is increased, while SM-1 and NM-MHC both
16 decrease. The change in the ratio of expression of these various isoforms results in SM-2 being termed the“mature” SM-MHC isoform (Lin et al., 2000; Arens et al., 2000).
Changes in the expression of these different isoforms are believed to occur as the result of a functional need for the specific tissue to be able to develop tension with maturity (Arens et al., 2000). In contrast, it has been reported that there is no relationship between SM-
1/SM-2 isoform expression and Vmax (Morano et al., 1993; Meer and Eddinger, 1997) and that the different isoforms are indistinguishable in their ability to move actin filaments in in vitro motility assays (Kelley et al., 1992). Meer and Eddinger (1997) report that isoform composition may play a role in final cell length after shortening and suggest that
SM-1 and SM-2 isoforms may affect filament length. Ishibashi and Bukaski (1997) correlated lowering of SM-2 isoform levels and a constant level of SM-1 with the loss of force generated by tissue in culture. Other studies have shown that cyclic stretch can induce a four to six fold increase in SM-1 and SM-2 levels with a concurrent lowering of
NM-MHCA protein concentrations (Reusch et al., 1996). The issue of isoform function is further complicated by the finding that the two isoforms may form SM-1/SM-2 heterodimers as well as SM-1 and SM-2 homodimers and that these dimer forms may differ between cells in the same tissue (Tsao and Eddinger, 1993). However, Kelly et al.
(1992) found that only homodimers were formed. Recently, a smooth muscle myosin heavy chain knockout mouse has been developed (Morano et al., 2000). This model did not express SM-1 or SM-2 but did appear to recruit NM-MHC for contraction. In these animals, the initial fast phase of contraction was absent while the slow, maintained phase was identical to that in control mice suggesting that SM-MHC isoforms are only essential
17 for the fast phase of contraction and the tissue has the ability to switch from one type to another at different stages of contraction (Morano et al., 2000). A second set of SM-MHC isoforms have been identified that contains alternate splicing of a twenty-one-bp exon at the 5' end of the transcript that encodes for seven amino acids in the globular head domain near the ATP-binding site. This results in the inserted (SM1B, SM2B) and non-inserted
(SM1A, SM2A) isoforms of the myosin heavy chain (Babij 1993; Kelley et al., 1993;
White et al., 1993). The seven amino acid insertion appears to be mostly absent in vascular smooth muscle. SM-MHCs with the insert have been shown to have a two fold increase in ATPase activity and moves actin filaments twice as fast as SM-MHcs lacking the insert (Kelley et al., 1993; Rovner et al., 1997). However it has been demonstrated that SM-B expression alone cannot completely explain the differences in shortening velocity (Eddinger and Meer, 2001). Distribution of the SM-A and SM-B isoforms has also shown to be nonuniform, even within the same tissue (Parisi and Eddinger, 2002;
Low et al., 1999).
Due to the diversity and heterogenity in myosin heavy chain isoform expression within tissue, the role of the MHC isoforms in the smooth muscle contraction remains uncertain (Murphy et al., 1997).
Smooth Muscle Myosin Light Chains
Two isoforms of the essential 17 KDa smooth muscle light chain have been identified and designated LC17a and LC17b. These isoforms differ in five of the last nine
18 amino acids at their carboxy terminals (Lash et al., 1990; Nabeshima et al., 1987). There appears to be a correlation between the expression of LC17 isoforms and smooth muscle contractile properties. Faster contracting muscles such as those in the gastrointestinal tract contain only LC17a, while the slower contracting smooth muscle of the aorta, pulmonary and carotid arteries contain higher levels of LC17b (Helper et al., 1988; Molmquist and
Arner, 1991). Exchange of the LC17a isoform into avian aortic smooth muscle did not increase filament velocity in the in vitro motility assay (Kelley et al., 1993). However, transfer of LC17a into porcine aorta increased ATPase activity (Hasegawa and Morita,
1992) suggesting the effect of LC17a on filament velocity may be species specific. More recently, recombinant LC17a and LC17b expression has demonstrated that myosin expressing LC17a has an increased rate of force development over tissues expressing
LC17b (Huang et al., 1999). In addition, both an increase in shortening velocity and rate of force development in response to trifluoperazine were shown with forced LC17a expression (Matthew et al., 1988). These studies suggest that the essential light chain 17 isoform composition of tissue may explain some of the differences in contractile properties between phasic and tonic smooth muscle.
References
Adam, L.P., J.R. Haeberle and D.R. Hathaway. 1989. Phosphorylation of caldesmon in arterial smooth muscle. J. Biol. Chem. 264:7698-7703.
Adelstein, R.S. and J.R. Sellers. 1996. Myosin structure and function. In: Biochemistry of smooth muscle contraction. M. Barany ed. Acedemic Press Inc. San Deigo, California. 3-19.
Applegate, D. and J.D. Pardee. 1992. Actin-facilitated assembly of smooth muscle
19 myosin induces formation of actomyosin fibrils. J. Cell. Biol. 117:1223-1230.
Arens, Y.H., C.R. Rosenfeld, and K.E. Kamm. 2000. Maturational differences between vascular and bladder smooth muscle during ovine development. Am. J. Physiol. 278:R1305-1313.
Babij, P. 1993. Tissue-specific and developmentally regulated alternative slicing of a visceral isoform of smooth muscle myosin heavy chain. Nucleic Acids Res. 21:1476-1481.
Babij, P., and M. Periasamy. 1989. Myosin heavy chain isoform diversity in smooth muscle is produced by differential RNA processing. J. Mol. Biol. 210:673-679.
Barany, M., J.T. Barron, L. Gu, and K. Barany. 2001. Exchange of the actin-bound nucleotide in intact arterial smooth muscle. J. Biol. Chem. 276:48398-48403.
Barany, M., E. Polyak, and K. Barany. 1992. Protein phosphorylation during contraction- relaxation-contraction cycle of arterial smooth muscle. Arch. Biochem. Biophys. 294:571-578.
Bray, D. 1972. Cytoplasmic actin: A comparative study. Cold Spring Harbor Symposia on Quantitative Biology. 37:567-571
Carter, S.B. 1967. Effects of cytochalasins on mammalian cells. Nature. 213:261-264.
Cook, R.K., W.T. Blake, and P.A. Rubenstein. 1992. Removal of the amino-terminus acidic residues of yeast actin. Studies in vitro and in vivo. J. Biol. Chem. 267:9430-9436.
Cooper, J.A. 1987. Effects of cytochalasin and phalloidin on actin. J. Cell Biol. 105:1473-1488.
Coue, M., S.L. Brenner, I. Spector, and E.D. Korn. 1987. Inhibition of actin polymerization by latrunculin A. FEBS Letters. 213:316-318.
Cross, R.A. and J. Vandekerckhove. 1986. Solubility-determining domain of smooth muscle myosin rod. FEBS Lett. 200:355-360.
Dillon, P.E., M.O. Aksoy, S.P. Driska, and R.A. Murphy. 1981. Myosin phosphorylation and the cross-bridge cycle in arterial smooth muscle. Science. 211:495-497.
Di Salvo, J., G. Pfitzer, and L.A. Semenchuk. 1994. Protein tyrosine phosphorylation, cellular Ca2+, and Ca2+ sensitivity for contraction of smooth muscle. Can. J. Physiol. Pharmacol. 72:1434-1439.
20 Di Salvo, J., A. Steusloff, L. Semenchuk, S. Satoh, K. Kolguist, and G. Pfitzer. 1993. Tyrosine kinase inhibitors suppress agonist-induced contraction in smooth muscle. Biochem and Biophys. Res. Commun. 190(3):968-974.
Driska, S.P., M.O. Aksoy, and R.A. Murphy. 1981. Myosin light chain phosphorylation associated with contraction in arterial smooth muscle. Am. J. Physiol. 240:C222- C233. Duarte, J., M.A. Ocete, F. Perez-Vizcaino, A. Zarzuelo and J. Tamarso. 1997. Effect of tyrosine kinase and tyrosine phosphatase inhibitors on aortic contraction and induction of nitric oxide synthase. Eur. J. Pharmacol. 338:25-33.
Eddinger, T.J. and D.P. Meer. 1997. Myosin isoform heterogeneity in single smooth muscle cells. Comp. Biochem. B. Biochem. Mol. Biol. 117:29-38.
Eddinger, T.J. and D.P. Meer. 2001. Single rabbit stomach smooth muscle cell myosin heavy chain SMB expression and shortening velocity. Am. J. Physiol. 280:C309- 316.
Eddinger, T.J. and R.A. Murphy. 1991. Developmental changes in actin and myosin heavy chain isoform expression in smooth muscle. Arch. Biochem. Biophys. 284:232-237.
Estes, J.E., L.A. Selden, H.J. Kinosian, and L.C. Gershman. 1992. Tightly-bound divalent cation of actin. J. Musc. Res. Cell Motil. 13:272-284.
Fatigati, V. and R.A. Murphy. 1984. Actin and tropomyosin variants in smooth muscle. Dependence on tissue type. J. Biol. Chem. 259:14383-14388.
Gabella, G. 1976. The force generated by visceral smooth muscle. J. Physiol. (Lond). 263, 199-213.
Gillis, J.M., M.L. Cao, and A. Godfraind-DeBecker. 1988. Density of myosin filaments in the rat anococcygeus muscle at rest, and in contraction II. J. Mus. Res. Cell. Motil. 9:18-28.
Godfrain-DeBecker, A. and J.M. Gillis. 1988. Analysis of the birefringence of the smooth muscle anococcygeus of the rate at rest and in contraction. J. Mus. Res. Cell Motil. 9:9-17.
Gokita, T., Y. Miyauchi and M.K. Uchida. 1994. Effects of tyrosine kinase inhibitor and phosphotyrosine-phosphatase inhibitor, orthovanadate on Ca2+ free contraction of uterine smooth muscle of the rat. Gen. Pharmac. 25:1673-1677.
Gordon, A.M., A.F. Huxley, and F.J. Julian. 1966. The variation in isometric tension
21 with sarcomere length in vertebrate muscle fibers. J. Physiol. (Lond.). 184, 170- 192.
Gunst, S.J. 1999. Applicability of the sliding filament/crossbridge paradigm to smooth muscle. Rev. Physiol. Biochem. Pharmacol. 137:7-61.
Hai, C-M and R.A. Murphy. 1988. Cross-bridge phosphorylation and regulation of latch state in smooth muscle. Am. J. Physiol. 254:C99-106.
Hai, C-M., C.M. Rembold, and R.A. Murphy. 1991. Can different four-state cross-bridge models explain latch and the energetics of vascular muscle? In: Regulation of Smooth Muscle Contraction. Plenum Press, New York, NY. pp 159-171.
Hai, C-M and R.A. Murphy. 1989. Cross-bridge dephosphorylation and relaxation in vascular smooth muscle. Am. J. Physiol. 256:C282-C287.
Haller, H., J.I. Smallwood, and H. Rasmussen. 1990. Protein kinase C translocation in intact vascular smooth muscle strips. Biochem. J. 270:375-381.
Harris, D.E. and D.M. Warshaw. 1993. Smooth and skeletal muscle myosin both exhibit low duty cycles at zero load in vitro. J. Biol. Chem. 268:14764-14768.
Hasegawa, Y. and F. Morita. 1992. Role of the 17-KDa essential light chain isoforms of aorta smooth muscle. J. Biochem. Tokyo. 111:804-809.
Helper, D. J., J. A. Lash, and D.R. Hathaway. 1988. Distribution of isoelectric variants of the 17,000-dalton myosin light chain in mammalian smooth muscle. J. Biol. Chem. 263:15748-15753.
Hemric, M.E. and J.M. Chalovich. 1988. Effect of caldesmon on the ATPase activity and the binding of smooth and skeletal myosin subfragments to actin. J. Biol. Chem. 263:1878-1885.
Herman, I.M. 1993. Actin isoforms. Curr. Opin. Cell Biol. 5:48-55.
Herrera, A.M., K.H. Kuo, and C.Y. Seow. 2002. Influence of calcium on myosin thick filament formation in intact airway smooth muscle. Am. J. Physiol. 282:C310- 316. Hirai, S. and T. Hirabayashi. 1983. Developmental change of protein constituents in chicken gizzards. Dev. Biol. 97:483-493.
Horowitz, A., K.M. Trybus, D.S. Bowman, and F.S. Fay. 1994. Antibodies probe for folded monomeric myosin in relaxed and contracted smooth muscle. J. Cell Biol. 126:1195-1200.
22 Huang, Q.Q., S.A. Fisher, and F.V. Brozovich. 1999. Forced expression of essential myosin light chain isoforms demonstrates their role in smooth muscle force production. J. Biol. Chem. 247:35095-35098.
Huxley, A.F. and R. Neidergerke. 1954. Structural changes in muscle during contraction. Nature. 173:971-973.
Ikebe, M. and D.J. Hartsthorne. 1985. Proteolysis of smooth muscle myosin by Staphlyococcus areus protease: preparation of heavy meromyosin and subfragment 1 with intact 20 000-dalton light chains. Biochem. 24:2380-2387.
Ikebe, M., S. Komatsu, J.L. Woodhead, K. Mabuchi, R. Ikebe, J. Saito, R. Craig, and M. Higashihara. 2001. The tip of the coiled-coil rod determines the filament formation of smooth muscle and nonmuscle myosin. J. Biol. Chem. 276:30293- 30300.
Ishibashi, K. and R.D. Bukoski. 1997. Myosin isoform expression and force generation in cultured resistance arteries. Am. J. Physiol. 272:C1144-1150.
Jahromi, S.S. and H.L. Atwood. 1969. Correlation of structure, speed of contraction, and total tension in fast and slow abdominal muscle fibers of the lobster (Homarus americanus). J. Exp. Zool. 171:25-37.
Jasper, H.H. and A. Pezard. 1934. Relation entre la rapidite d’un muscle strie sa structure histologique. C. R. Acad. Sci. Hebd. Seances Acad. Sci. D. 198:499-501.
Jones, K.A., W.J. Perkins, R.R. Lorenz, Y.S. Prakash, G.C. Sieck and D.O. Warner. 1999. F-actin stabilization increases tension cost during contraction of permeabilized airway smooth muscle in dogs. J. Physiol. (Lond.). 519:527-538.
Kabsch, W. and J. Vandekerckhove. 1992. Structure and function of actin. Annu. Rev. Biophys. Biomol. Str. 21:49-76.
Kelley, C.A., S. Kawamoto, M.A. Conti, and R.S. Adelstein. 1991. Phosphorylation of vertebrate smooth muscle and nonmuscle myosin heavy chains in vitro and in intact cells. J. Cell Sci. Suppl. 14:49-54.
Kelley, C.A., J.R. Sellers, P.K. Goldsmith, and R.S. Adelstein. 1992. Smooth muscle myosin is composed of homodimeric heavy chains. J. Biol.Chem. 267:2127-2130.
Kelley, C.A., M. Takahasi, J.H. Yu, and R.S. Adelstein. 1993. An insert of seven amino acids confers functional differences between smooth muscle myosins from the intestines and vasculature. J. Biol. Chem. 268:12848-12854.
23 Kuo, K., L. Wang, P.D. Pare, L.E. Ford, and C. Y. Seow. 2001. Myosin thick filament lability induced by mechanical strain in airway smooth muscle. J. Appl. Physiol. 90:1811-1816.
Lash, J.A., J.R. Sellers and D.R. Hathaway. 1986. The effects of caldesmon on smooth muscle heavy actomeromyosin ATPase activity and binding of heavy meromyosin to actin. J. Biol. Chem. 261:16155-16160.
Lash, J.A., D.J. Helper, M. Klug, A.W. Nicolozakes, and D.R. Hathaway. 1990. Nucleotide and deduced amino acid sequence of cDNAs encoding two isoforms for the 17,000 dalton myosin light chain in bovine aortic smooth muscle. Nucleic Acid Res. 18:71-76.
Lin, V.K., J.B. Robertson, I.L. Lee, P.E. Zimmern, and J.D. McConnell. 2000. Smooth muscle myosin heavy chains are developmentally regulated in the rabbit bladder. J. Urol. 164:1376-1380.
Low, R.B., J. Mitchell, J. Woodcock-Mitchell, A.S. Rovner, and S.L. White. 1999. Smooth-muscle myosin heavy-chain SM-B isoform diversity in developing and adult rat lung. Am. J. Res. Cell Mol. Biol. 20:651-657.
Malmquist, U. and A. Arner. 1991. Correlation between isoform composition of the 17KDa myosin light chain and maximal shortening velocity in smooth muscle. Eur. J. Physiol. 418:523.530.
Marston, S.B. and C.W.J. Smith. 1990. Purification and properties of Ca2+-regulated thin filaments and F-actin from sheep aorta smooth muscle. J. Musc. Rec. Cell. Motil. 5:559-575.
Matthew, J.D., A.S. Khromov, K.M. Trybus, A.P. Somlyo, and A.V. Somlyo. 1998. Myosin essential light chain isoforms modulate the velocity of shortening propelled by nonphosphorylated cross-bridges. J. Biol. Chem. 273:31289-31296.
Meer, D.P. and T.J. Eddinger. 1997. Expression of smooth muscle myosin heavy chains and unloaded shortening in single smooth muscle cells. Am. J. Physiol. 273:C1259-C1266.
Mehta, D. and S.J. Gunst. 1999. Actin polymerization stimulated by contractile activation regulates force development in canine tracheal smooth muscle. J. Physiol. (Lond). 519(3):829-840.
Merkel, L., W.T. Gerthoffer, and T.J. Torphy. 1990. Dissociation between myosin phosphorylation and shortening velocity in canine trachea. Am. J. Physiol. 258:C524-C532.
24 Merkel, I.A., L.M. Rivera, D.J. Colussi, and M.H. Perione. 1991. Protein kinase C and vascular smooth muscle contractility: Effects of inhibition and down-regulation. J. Pharmacol. Exp. Ther. 257:134-140.
Mitchell, R.W., C.Y. Seow, T. Burdyga, R. Maass-Moreno, V.R. Pratusevich, J. Ragozzino, and L.E. Ford. 2001. Relationship between myosin phosphorylation and contractile capability of canine airway smooth muscle. J. Appl. Physiol. 90:2460-2465.
Morano, I. 1992. Molecular biology of smooth muscle. J. Hypertens. 10:411-416.
Morano, I., G. Erb, B. Sogl. 1993. Expression of myosin and light chains changes during pregnancy in the rat uterus. Pflugers Archive. 423:434-441.
Morano, I., G. Chai, L.G. Baltas, V. Lamounier-Zepter, G. Lutsch, M. Kott, H. Haase, and M. Bader. 2000. Smooth-muscle contraction without smooth muscle myosin. Nature Cell Biol. 2:371-375
Murphy, R.A. 1976. Contractile system function in mammalian smooth muscle. Blood Vessels. 13, 1-23.
Murphy, R.A., J.T. Herliky, and J. Meyerman. 1974. Force-generating capacity and contractile protein content of arterial smooth muscle. J. Gen. Physiol. 64, 691-705
Murphy, R.A., J.S. Walker, and J.D. Strauss. 1997. Myosin isoforms and functional diversity in vertebrate smooth muscle. Comp. Biochem. Physiol. 117B:51-60.
Nabeshima, Y., Y. Nonomura, and Y. Fiji-Kurijama. 1987. Nonmuscle and smooth muscle myosin light chain mRNAs are generated from a single gene by the tissue- specific alternative RNA splicing. J. Biol. Chem. 262:10608-10612.
Nagai, R., M.. Kuro, M. Babij, and M. Periasamy. 1989. Identification of two types of smooth muscle myosin heavy chain isoforms by cDNA cloning and immunoblot analysis. J. Biol. Chem. 264:9734-9737.
Nakajima, S., M. Fujimoto, M. Veda. 1993. Spatial changes of [Ca2+]i and contraction caused by phorbol esters in vascular smooth muscles. Am. J. Physiol. 265(4):C1138-C1145.
North, A.J., M. Gimona, Z. Lando, and J.V. Small. 1994. Actin isoform compartments in chicken gizzard smooth muscle. J. Cell Sci. 107:445-455.
Obara, K. and H. Yabu. 1994. Effect of cytochalasin B on intestinal smooth muscle cells.
25 Eur. J. Pharm. 255:139-147.
Parisi, J.A. and T.J. Eddinger. 2002. Smooth muscle myosin heavy chain isoform distribution in the swine stomach. J. Histochem. Cytochem. 50:385-394.
Paul, R.J. 1983. Functional compartmentalization of oxidative and glycolytic metabolism in vascular smooth muscle. Am. J. Physiol. 244:C399-C409.
Pollard, T.D. and J.A. Cooper. 1986. Actin and actin-binding proteins. A critical evaluation of mechanisms and fuctions. Ann. Rev. Biochem. 55:987-1035.
Pratusevich, V.R., C.Y. Seow, and L.E. Ford. 1995. Plasticity in canine airway smooth muscle. J. Gen. Physiol. 105:73-94.
Prochniewicz, E. and H. Strzelecka-Golaszewska. 1980. Chicken-gizzard actin: interaction with skeletal-muscle myosin. Eur. J. Biochem. 106:305-312.
Reddy, S., K. Ozgur, M. Lu, W. Chang. S.R. Mohan, C.C. Kumar, anf H.E. Ruley. 1990. Structure of the human smooth muscle alpha-actin gene. Analysis of a cDNA and 5' upstream region. 265:1683-1687.
Reusch, P., H. Wagdy, R. Reusch, E. Wilson, and H.E. Ives. 1996. Mechanical strain increases smooth muscle and decreases nonmuscle myosin expression in rat vascular smooth muscle cells. Circ. Res. 79:1046-1053.
Rice, R.V., J.A. Moses, G.M. McCanus, A.C. Brady, and L.M. Blask. 1970. The organization of contractile filaments in mammalian smooth muscle. J. Cell Biol. 47, 183-196.
Rokolya, A. and H.A. Singer. 2000. Inhibition of CaM Kinase II activation and force maintenance by KN-93 in arterial smooth muscle. Am. J. Cell. Physiol. 278:C537-C545.
Rovner, A.S., R.A. Murphy, and G.K. Owens. 1986. Expression of smooth muscle and nonmuscle myosin heavy chains in cultured vascular smooth muscle. J. Biol. Chem. 261:14740-14745.
Rovner, A.S., Y. Freyzon, and K.M. Trybus. 1997. An insert in the motor domain determines the functional properties of expressed smooth muscle myosin isoforms. J. Mus. Res. Cell Motil. 18:103-110.
Rovner, A.S., P.M. Fagnant, S. Lowey, and K.M. Trybus. 2002. The carboxy-terminal isoforms of smooth muscle myosin heavy chain determine thick filament assembly properties. J. Cell Bio. 156:113-123.
26 Saito, S., H. Masatoshi, H. Ozaki, and H. Kavaki. 1996. Cytochalasin D inhibits smooth muscle contraction by directly inhibiting the contractile apparatus. J. Smooth Mus. Res. 32:51-60.
Seow, C.Y., V. R. Pratusevich and L.E. Ford. 2000. Series-to-parallel transition in the filament lattice of airway smooth muscle. J. Appl. Physiol. 89:869-876.
Schildmeyer, L.A., R. Braun, G. Taffet, M. Debiasi, A.E. Burns, A. Bradley, and R. J. Schwartz. 2000. Impaired vascular contractility and blood pressure homeostasis in the smooth muscle "-actin null mouse. FASEB J. 14:2213-2220.
Schoenberg, C.F. 1969. An electron microscope study of the influence of divalent ions on myosin filament formation in chicken gizzard extracts and homogenate. Tissue Cell. 1:83-96.
Schroeder, T.E. 1970. The contractile ring. I. Fine structure of dividing mammalian (HeLa) cells and the effects of cytochalasin B. Z. Zellforsch. Mikrosk. Anat. 109:431-449.
Shimamoto, Y., H. Shimamoto, C.Y. Kwan, and E.E. Daniel. 1993. Differential effects of putative protein kinase C inhibitors on contraction of rat aortic smooth muscle. Am. J. Physiol. 264(33):H1300-H1306.
Shirazi A., K. Iizuka, P. Fadden, C. Mosse, A.P. Somlyo, A.V. Somlyo, and T.A.J. Haystead. 1994. Purification and characterization of the mammalian myosin light chain phosphatase holoenzyme. The differential effects of the holoenzyme and its subunits on smooth muscle. J. Biol. Chem. 269:31598-31606.
Singer, H.A., C.M. Schworer, C. Sweeley, and H. Benscoter. 1992. Activation of protein kinase C isozyme by contractile stimuli in arterial smooth muscle. Arch. Biochem. Biophys. 299:320-329.
Small, J.V. 1995. Structure-function relationships in smooth muscle: The missing links. Bioessays. 17:785-792.
Somlyo, A.V. 1980. Ultrastructure of vascular smooth muscle. In: Handbook of Physiology, the Cardiovascular System vol. II, Vascular Smooth Muscle, Bethesda, MD, Am. Physiol. Soc. pp 33-68.
Somlyo, A.V., T.M. Butler, M. Bond, and A.P. Somlyo. 1981. Myosin filaments have non-phosphorylated light chains in relaxed smooth muscle. Nature Lond. 294:567-569.
Somlyo, A.P., T. Kitazawa, B. Himpens, G. Matthijs, K. Horviti, S. Kobayashi, Y.E.
27 Goldman and A.V. Somlyo. 1989. Modulation of Ca2+ sensitivity and the time course of contraction in smooth muscle: A major role of protein phosphatases. Adv. Protein Phosphatases. 5:181-195.
Somlyo, A.P. and A.V. Somlyo. 1992. Smooth muscle structure and function. In: The heart and cardiovascular system. Fozzard, H.A. ed. Raven Press. NY, New York. pp 1295-1324.
Srivastava, A.K., and J. St-Louis. 1997. Smooth muscle contractility and protein tyrosine phosphorylation. Mol. Cellular Biochem. 176:47-51.
Suzuki, H., H. Onishi, K. Takahashi, and S. Watanabe. 1978. Structure and function of chicken gizzard myosin. J. Biochem. 84:1529-1542.
Spudrich, J.A. 1972. Effects of cytochalasin B on actin filaments. Cold Spring Harbor Symposium on Quantitative Biology. 37:585-593.
Spudrich, J.A. and S. Lin. 1972. Cytochalasin B, its interaction with actin and actomyosin from muscle. Proc. Nat. Acad. Sci. 69:442-446.
Tolloczko, B., F.C. Tao, M.E. Zacour and J.G. Martin. 2000. Tyrosine kinase-dependent calcium signaling in airway smooth muscle. Am. J. Physiol. 278:L1138-1145.
Trybus, K. 1994. Regulation of expressed truncated smooth muscle myosins. Role of the essential light chain and tail length. J. Biol. Chem. 269:20819-20822.
Tsao, A.E. and T.J. Eddinger. 1993. Smooth muscle myosin heavy chains combine to form native myosin isoforms. Am. J. Physiol. 264:H1653-1662.
Uyeda, T.Q.P., S.J. Cron, and J.A. Spudich. 1990. Myosin step size. Estimation from slow sliding movement of actin over low densities of heavy meromyosin. J. Mol. Biol. 214:699-710.
VanBuren, P., W.H. Guilford, G. Kennedy, J. Wu, and D.M. Warshaw. 1995. Smooth muscle myosin: a high force-generating motor. Biophys. J. 68(4):2588-2598.
VanBuren, P., S.S. Work, and D.M. Warshaw. 1994. Enhanced force generation by smooth muscle myosin in vitro. Proc. Natl. Acad. Sci. 91:202-205.
Vandekerckhove, J. and K. Webber. 1978. At least six different actins are expressed in a higher mammal: An analysis based on the amino acid sequence of the amino- terminal tryptic peptide. J. Mol. Biol. 126:783-802.
Vandekerckhove, J. and K. Webber. 1981. Actin typing on total cellular extracts: A
28 highly sensitive protein-chemical procedure able to distinguish different actins. Eur. J. Biochem. 113:595-603.
Walker, J.S., C.J. Wingard and R.A. Murphy. 1994. Energetics of cross-bridge phosphorylation and contraction in vascular smooth muscle. Hypertension. 23:1106-1112.
Wessells, S.A. 1971. Microfilaments in cellular and developmental processes. Science. 171:135-143.
White, S., A.F. Martin, and M. Periasamy. 1993. Identification of a novel smooth muscle myosin heavy chain cDNA: isoform diversity in the S1 head region. Am. J. Physiol. 264:125-128.
Wright, G.L. and A.S. Battistella-Pattersom. 1998. Involvement of the cytoskeleton in calcium-dependent stress relaxation of rat aortic smooth muscle. J. Mus. Res. Cell Motil. 19:405-414.
Wright, G. and E. Hurn. 1994. Cytochalasin inhibition of slow tension increase in rat aortic rings. Am. J. Physiol. 267: H1437-H1447.
Xu, J., J. Gillis, and R. Craig. 1997. Polymerization of myosin on activation of rat anococcygeus smooth muscle. J. Mus. Res. Cell Motil. 18:381-393.
Zhang, Y., and R.S. Moreland. 1994. Regulation of Ca2+-dependent ATPase activity in detergent skinned vascular smooth muscle. Am. J. Physiol. 36:H1032-H1039.
29 Chapter 2. Remodeling of the Actin Cytoskeleton in the Contracting A7r5 Smooth Muscle Cell
A manuscript published in the Journal of Muscle Research and Cell Motility
M. E. Fultz, C. Li, W. Geng, and G. L. Wright
reproduced with the kind permission of Kluwer Academic Publishers
Summary
It has been proposed that the reorganization of components of the actin cytomatrix could contribute to force development and the low energy cost of sustained contraction in contractile cells which lack a structured sarcomere (Battistella-Patterson et al., 1997).
However, there has been no direct evidence of an apropos actin reorganization specifically linked to the contractile response in cells of this type. Remodeling of the "- and $-actin domains was studied in A7r5 smooth muscle cells during phorbol 12, 13 dibutyrate
(PDB)-induced contraction using immunohistologic staining and $-actin-green fluorescent protein ($-actin-GFP) fusion protein expression. Cells stained with phalloidin as well as cells expressing $-actin-GFP showed densely packed actin stress cables, arranged in parallel and extending across the cell body. PDB caused approximately 85% of cells to contract with evidence of forcible detachment from peripheral adhesion sites seen in many cells. The contraction of the cell body was not uniform but occurred along a principal axis parallel to the system of densely packed $-actin cables. During the interval of contraction,
30 the $-actin cables shortened without evidence of disassembly or new cable formation. The use of cytochalasin to inhibit actin polymerization resulted in the dissolution of the actin cables at the central region of the cell and caused the elongation of precontracted cells. In unstimulated cells, "-actin formed cables similar in arrangement to the cell spanning $- actin cables. Within a short interval after PDB addition; however, the majority of "-actin cables disassembled and reformed into intensely fluorescing column-like structures extending vertically from the cell base at the center of clusters of "-actin filaments. The
"-actin columns of contracting cells showed strong colocalization of "-actinin suggesting they could be structurally analogous to the dense bodies of highly differentiated smooth muscle cells. The results indicate that the "- and $-actin domains of A7r5 cells undergo a highly structured reorganization during PDB-induced contraction. The extent and nature of this restructuring suggest that remodeling could play a role in contractile function.
Introduction
It is widely accepted that the mechanism underlying the generation of force by cells centers on the interaction of actin with myosin and that the principal biochemical and biophysical events determining cellular contractile properties are generally similar among different cell types. Much of our understanding of how cells develop force stems from the study of skeletal muscle in which the basic contractile unit, the sarcomere, is highly structured and active force development is in direct proportion to the degree of overlap of actin and myosin filaments (Gordon et al., 1966). However, other types of contractile cells
31 may lack a typical sarcomere arrangement but still exhibit the capability to generate force at a level equal (Murphy, 1976) or greater (Gabella, 1976) than in skeletal muscle. Within this category of contractile cells, smooth muscle has been intensely studied and a wealth of information is available pertaining to its contractile properties. For example, smooth muscle shows a uniform distribution of actin and myosin filaments within the cell
(reviewed in Small, 1995). Other notable differences include the ability of smooth muscle to slowly develop force and to sustain maximal tension for extended intervals at an energy cost of about 0.35% that expended in comparable contractions of skeletal muscle (Paul,
1983). Moreover, there is evidence suggesting that force development and maintained tension (Driska et al., 1981) as well as velocity of shortening (Merkel et al., 1990) are disassociated from myosin light chain phosphorylation and actin-activated myosin ATPase activity (Zhang and Moreland, 1994) in smooth muscle. Hence, there has been considerable recent interest in additional factors that could contribute in modifying the contractile properties of smooth muscle from that predicted solely on the basis of sliding filament theory.
Several reports have documented the finding that different agents which inhibit actin filament polymerization either block or significantly alter the contractile properties of smooth muscle (Wright and Hurn, 1994; Battistella-Patterson et al., 1997; Wright and
Batistella-Patterson, 1998; Mehta and Gunst, 1999; Jones et al., 1999). One explanation is that the remodeling of the contractile domain of the actin cytomatrix could contribute to the characteristics of active force development by effectively modulating this component
32 at optimal mechanical advantage prior to and during cell shortening (Battistella-Patterson et al., 1997). It has been further suggested that crosslinking (Small et al., 1986) and remodeling of the actin tension bearing fibers (Battistella-Patterson et al., 1997) could contribute to the low energy cost of sustained tension maintenance by holding the cell in the contracted configuration.
Despite speculation that cytoskeletal remodeling may play an important role in contractile function, there is little direct evidence of active structural remodeling in smooth muscle
(Togashi et al., 1998). To a considerable extent this paucity of information reflects the small cell size and the apparent lack of structural order of the cytoskeleton in the differentiated smooth muscle cell (Small and Gimona, 1998). One approach to the study of contractile function-related cytoskeletal organization is through the use of cell lines that retain the ability to contract in cell culture. For example, a number of structural and biochemical similarities can be demonstrated in the actin cytoskeletal and contractile domains of smooth muscle and fibroblasts (Sappino et al., 1990; Buoro et al., 1993) suggesting that the study of different cell types could provide useful information concerning contraction-linked remodeling phenomenon in the cytoskeleton. In the present study we utilized confocal microscopy with $-actin-enhanced green fluorescent protein ($- actin-GFP) fusion protein expression and standard immunohistologic staining techniques to investigate changes in the actin cytoskeleton in contracting A7r5 cells. The results indicate extensive and highly different modes of remodeling of the $-actin and "-actin domains during phorbol 12,13 dibutyrate-induced cell contraction. The findings provide
33 initial insight into the extent and nature of the reorganization of the actin cytoskeleton of contractile cells that do not have a highly structured arrangement of contractile proteins.
Materials and Methods
Cell Culture. A7r5 cells, originally derived from embryonic rat aorta and exhibiting an adult smooth muscle phenotype (Firulli et al., 1998) were obtained from American Type
Culture Collection (Manassas, VA) and cultured in Dulbecco’s modified Eagle’s medium
(DMEM) that was supplemented with 10% fetal bovine serum, 100 units/ml penicillin G, and 100 µg/ml streptomycin. Cells were plated onto 75 mm2 culture flasks and grown to approximately 85% confluence. Cultures were maintained at 37°C in a humidified
atmosphere of 5% CO2 in air. The medium was changed every two days and the cells were passaged at least once a week. Cells were detached from the culture flask by the addition of a trypsin/EDTA solution in HBSS.
$-actin-GFP Plasmid Transfection. A $-actin-EGFP expression plasmid was purchased from Clontech (Palo Alto, CA). The green fluorescent protein-actin fusion protein has been successfully utilized in a wide range of cell types (Noegel and Schleicher, 2000;
Okada et al., 1999; Verkhusa et al., 1999; Faix et al., 1998; Aizawa et al., 1997) and has been shown to be a suitable probe for the study of actin organization and dynamics in several mammalian cell lines (Choidas et al., 1998). Cells were seeded at a density of 4 x
106 cells/100 mm2 culture flask and transfected with 2 to 6 µg of plasmid using
34 lipofectamine (Life Technologies, Rockville, MD) according to the manufacturer’s standard protocol. Fluorescence of the $-actin-GFP fusion protein was typically observed within 2 days and experiments were performed within 3 or 4 days after transfection.
Confocal Microscopy. In experiments employing cell immunostaining, A7r5 cells were plated onto 22 mm by 22 mm #1 glass coverslips placed in 6 well culture plates and returned to the incubator for a minimum of 24 hours to allow for cell attachment and spreading. Cells were fixed and permeabilized by the addition of ice cold acetone for 1 minute. The cells were then washed multiple times (3X) with phosphate-buffered saline
(PBS) containing 0.5% TWEEN-20 (PBS-T); pH 7.5, and incubated for 10 minutes in blocking solution (5% nonfat dry milk in PBS-T). Cells were stained for "-actin by incubation in a 1:500 dilution of monoclonal anti-"-smooth muscle actin clone 1A4 FITC- labeled antibody (Sigma Chemical Co., St. Louis, MO) for 30 minutes at room temperature. For total F-actin staining, the fixed cells were incubated with 2.5 x 10-6M
TRITC-labeled phalloidin (Sigma Chemical Co., St. Louis, MO) for 30 minutes at room temperature. Cells were stained for $-actin by incubation in a 1:500 dilution of monoclonal anti-$-actin clone AC-15 primary antibody (Sigma Chemical Co., St. Louis,
MO) for 30 minutes at room temperature followed by incubation with an Alexa 488 labeled secondary antibody (Molecular Probes, Eugene, OR). The crosslinking protein "- actinin was visualized by incubation of cells in a 1:200 dilution of monoclonal Anti-"-
Actinin primary antibody (Sigma Chemical Co., St. Louis, MO) followed by the addition of Alexa 488-labeled secondary antibody (Molecular Probes, Eugene, OR). In dual
35 immunostaining experiments to study the colocalization of "-actin and "-actinin, Alexa
594-labeled secondary antibody (Molecular Probes, Eugene, OR) was substituted for the
Alexa 488-labeled antibody. Stained cells or living cell preparations expressing $-actin-
GFP were visualized by mounting on a Nikon Diaphot Microscope and confocal microscopy was performed with a BioRad Model 1024 Scanning system equipped with a krypton/argon laser. Micrographs were built by projecting serial Z-plane image acquisitions and were analyzed using Lasersharp and Confocal Assistant software
(BioRad, Hercules, CA).
Cell Treatments. Actin cytoskeletal reorganization was studied in A7r5 cells contracted by the addition of 10-8 M phorbol 12,13 dibutyrate (PDB) to the incubation medium. In order to ensure that the results were not due to a nonspecific effect of PDB, control studies were also conducted using the inactive 4"-phorbol 12,13-didecanoate (4" PDD) at concentrations ranging to 10-6 M. In ancillary experiments to examine the effects of
-5 different agonists, the contractile response of A7r5 cells to A23187 (10 M), norepinephrine
(10-6 M), and angiotensin II (10-6 M) was also investigated. In order to study the effects of inhibiting actin polymerization, cytochalasin B (0.1 µg/ml) was added to the medium at varying intervals prior to cell fixation or imaging.
Data Analysis. Each experiment was performed a minimum of three times with at least
50 cells from fixed preparations and 3 to 6 cells from live preparations evaluated per experiment. Results were confirmed by at least two observers and phenomenon were
36 judged significant if observed in 80% or more cells. Contractile shortening of cells in the x- and y-axis was analyzed by two way ANOVA followed by Students t-test. P < 0.05 was considered significant.
Results
Figure 1 compares the contractile response to phorbol 12,13 dibutyrate (PDB) and the
$ calcium ionophore A23187 in A7r5 cells expressing -actin-GFP fusion protein. Both agents were effective agonists, consistently causing approximately 85% of cells to visibly contract. Cell shortening in response to PDB could first be observed at about 10 minutes and extended over an interval of approximately 60 minutes to completion. Cell constriction to PDB was extremely strong such that by 35 to 45 minutes the resolution of individual actin filaments was no longer possible. In consequence, cell imaging was performed during the first 30 minutes following PDB addition, while individual actin stress cables were clearly observed. During this interval the pattern of cell constriction was acentric with the principal axis of shortening (x-axis) parallel to the densely packed $- actin stress cables spanning the cell body. PDB-treated cells showed an average 34% reduction of length in the x-axis (p < 0.05) while shortening in the y-axis was only 17% at
30 minutes (Fig. 2). Veritcal displacement of cell volume was verified from analysis of Z- plane images which indicated an average 46 ± 8% increase in cell height of PDB-treated
cells at 30 minutes. By comparison, cells treated with A23187 showed a more rapid onset of
37 contraction with detectable shortening noted as early as 2.0 minutes and completion of response recorded by 30 minutes in the majority of cells (Fig. 1). The average length in
the x- and y- axes of A23187 treated cells at 30 minutes was 48 ± 10% and 62 ± 2% of the pretreatment value, respectively. Cells contracted by either agonist exhibited thread-like extrusions and detached cell remnants on the glass substrate, indicating the forcible detachment from peripheral adhesion sites. Interestingly, A7r5 cells did not contract in the presence of 10-6 M norepinephrine or 10-6 angiotensin II (data not shown) suggesting that,
while these cells were clearly responsive to the elevation of intracellular calcium (A23187) and activation of PKC (PDB), they lacked the capability for receptor-mediated contraction to at least these two agonists.
Figure 3 shows the time course of PDB-stimulated cell shortening with the accompanying changes in the $-actin stress cables in A7r5 cell expressing $-actin-GFP. The $-actin cables shortened without evidence of significant compression throughout the interval in which the length of the x-axis decreased. Moreover, individual cables were identified throughout the interval of contraction indicating that shortening was not accomplished by disassembly and reassembly of new structure. The use of cytochalasin B to block actin filament polymerization caused the loss of actin stress cable integrity at the central region of the cell (Fig. 4). It was further noted that the treatment of precontracted cells to disrupt actin stress cables caused a visible cellular elongation in the x-axis (Fig. 5), suggesting that these structures contributed to holding the cell in the contracted configuration.
38 Staining PDB-stimulated cells with phalloidin verified the presence of an extensive system of actin stress cables in contracted cells (Fig. 6). In addition, these experiments revealed the occurrence of discrete peripheral bodies that were absent in unstimulated cells.
Because these structures were also not observed in contracted cells expressing $-actin-
GFP or in the cells stained with anti-$-actin antibody (Fig. 7), the results suggested that they most likely represented a component of the "-actin domain. Immunostaining of unstimulated cells indicated that "-actin formed thick stress cables (Fig. 8) similar in appearance to $-actin cables. Upon stimulation with PDB; however, "-actin underwent a dramatic reorganization that began to develop within 10 minutes, thus preceeding the first observable shortening of PDB-treated cells. Addition of PDB was followed by the loss of the major portion of densely packed "-actin stress cables observed prior to cell stimulation
(Fig. 8). Concurrent with the disassembly of "-actin stress cables, large numbers of intensely staining bodies were observed to form at the cell periphery just inside the membrane (Figs. 8, 9). These peripheral structures were further observed to occur at intervals along some of the few remaining "-actin cables or appeared as the focal point to clusters of randomly oriented "-actin fibers. An analysis of Z-series images indicated that they extended column-like from the cell base toward the dorsal cell surface (Fig. 10).
Interestingly, there was little or no evidence of compression of the "-actin cables and filaments in the interval between 10 and 30 minutes (Fig. 8) suggesting that this domain continued to remodel during the interval of cell constriction. Figure 11 demonstrates that neither the inactive phorbol ester 4" PDD nor norepinephrine, which did not cause contraction of A7r5 cells, had an effect on "-actin remodeling.
39 Immunostaining for "-actinin further indicated that this important actin crosslinking protein underwent PDB-induced remodeling closely associated with that of "-actin (Fig.
12). In unstimulated cells "-actinin staining was highly granular in appearance with the protein organized into filamentous strands. Within 10 minutes after treatment with PDB,
"-actinin reformed into intensely fluorescing peripheral bodies (Fig. 12) with some cells also showing high levels of "-actinin at the subplasmalemma. Dual immunostaining for
"-actin and "-actinin showed partial colocalization of the proteins in unstimulated cells and clear evidence of close association of the two components in the column-like peripheral structures of PDB-contracted cells (Fig. 13).
40 Figure 1 The contractile response of A7r5 cells to phorbol 12,13dibutyrate (PDB) $ and the calcium ionophore A23187. Cells were transfected with a -actin-GFP expression plasmid three days prior to experimentation. Individual cells are shown prior to (A, C) and 30 min after (B, D) the addition of the contractile agents to the media. Large arrows indicate the designated x-axis parallel to the majority of $-actin stress cables. Small arrows show cell debris resulting from the forceful constriction of the cell. Micrographs represent the results from 22 experiments in which a total of 74 and 67 cells were
examined throughout the interval of contraction induced by PDB and A23187, respectively. Original magnification at 400X, scale bar (B) indicates 50 microns.
41 Figure 2 The magnitude of decrease in cell length in the x-axis (hatched bar) parallel to the system of actin stress cables and the perpendicular y-axis (solid bar) in A7r5 cells at 30 minutes after the addition of phorbol 12,13 dibutyrate
(PDB) or the calcium ionophore A23187. Data are calculated as the percent of the resting cell length prior to stimulation. Bars represent the average ± SEM of results from 9 experiments in which a total of 74 and 67 cells were
studied throughout the interval of contraction induced by PDB and A23187, respectively. An (*) or (+) indicates a significant difference from the
unstimulated control value or between corresponding PDB and A23187 values, respectively (P <0.05).
42 Figure 3 The contraction of an A7r5 cell showing the shortening of individual $- actin stress cables. A) The cell prior to stimulation and at B) 5 minutes, C) 15 minutes, and D) 30 minutes after the addition of 10-8 M PDB to the medium. The cell was transfected with a $-actin-GFP expression plasmid three days prior to experimentation. The large arrow indicates the x-axis of cell constriction while the small arrow points to an individual cable which can be seen to shorten and appears to increase in diameter during cell constriction. Original magnification at 400X, scale bar (D) indicates 50 microns.
43 Figure 4 The effect of cytochalasin B on the actin cytomatrix of unstimulated and phorbol 12,13 dibutyrate (PDB)-stimulated A7r5 cells. Control cells were untreated (A) or stimulated by the addition of 10-8 PDB for 10 minutes (D) prior to staining with TRITC-labeled phalloidin. Cytochalasin B (0.1 µg/ml) was added to the medium of unstimulated cells for 10 minutes (B) or 20 minutes (C) prior to fixing the cells for staining. Stimulated cells were incubated in the presence of PDB for 10 minutes and the cytochalasin B added for an additional 10 minutes (E) or 20 minutes (F). The micrographs are representative of images collected in 4 individual experiments in which a total of 425 cells were examined. Original magnification at 600X, scale bar (F) indicates 50 microns.
44 Figure 5 The effect of cytochalasin-induced dissolution of the actin cytomatrix on cell length in the precontracted A7r5 cell. Cells were contracted by 10-8 M PDB 30 minutes prior to the addition of 0.1 µg/ml cytochalasin B to the media for up to 40 minutes. Images are from a single cell transfected with a $-actin-GFP expression plasmid three days prior to experimentation. The results show that within 10 minutes the cell began to elongate in the x-axis (arrow) parallel to the visible $-actin stress cables. Micrographs are representative of images collected in 3 separate experiments in which 8 cells were evaluated. Original magnification at 600X.
45 Fig ure 6 A comparison of unstimulated A7r5 cells (A, B) to cells stimulated with phorbol 12,13 dibutyrate (PDB) for 30 minutes (A’, B’) and then stained with TRITC- labeled phalloidin to demonstrate the preservation of an extensive system of actin stress cables and the formation of peripheral actin dense structures in the stimulated cells. Due to the limited dynamic range of the imaging system, the fluorescent signal is saturated when the image is presented at sufficient brightness to display the peripheral bodies. Micrographs were taken from a series of 8 separate experiments in which a total of 263 control and 280 PDB-stimulated cells were evaluated. Original magnification at 600X, scale bar (A’) indicates 50 microns.
46 Figure 7 A comparison of unstimulated (A, B) and phorbol 12, 13 dibutyrate (PDB) stimulated (A’B’) A7r5 cells stained with a monoclonal anti-$-actin clone AC-15 primary antibody and an Alexa 488-labeled secondary antibody to demonstrate the preservation of an extensive system of $-actin stress cables and the lack of formation of peripheral dense structures in the stimulated cells. Micrographs were taken from a series of 3 separate experiments in which a total of 94 control and 130 PDB-stimulated cells were evaluated. Original magnification at 600X, scale bar indicates 50 microns.
47 Figure 8 A comparison of the "-actin cytomatrix in unstimulated A7r5 cells and those treated with phorbol 12,13 dibutyrate (PDB) for 10 minutes (B, B’) and 30 minutes (C, C’). Cells were stained with a monoclonal anti-"- smooth muscle actin clone 1A4 FITC-labeled antibody. The results show the loss of "-actin stress cables and formation of peripheral dense structures in PDB-treated cells. Micrographs are representative of images collected in 6 separate experiments in which a total of 432 PDB-stimulated cells were evaluated. An average of 80 ± 3% of cells showed significant loss of "- actin cables and formation of peripheral dense structures. Original magnification at 600X, scale bar (C’) indicates 50 microns.
48 Figure 9 Micrographs at magnification 600X (A, B) and 900X (A’, B’) showing the association of "-actin peripheral dense structures with remnant "-actin stress cables (arrow) (A, A’) and the aster-like formation of these structures with associated "-actin filaments (B, B’). Cells were incubated with 10-8 M phorbol 12,13 dibutyrate for 30 minutes and then stained with a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. Micrographs are exemplary examples of images collected in 6 separate experiments. Scale bars (B, B’) indicate 50 microns.
49 Figure 10 A Z-plane series of sections taken in 4.0 micron increments from the cell base (A) toward the cell apex (D) showing that peripheral dense structures extend vertically as rod-or column-like bodies in the phorbol 12,13 dibutyrate (PDB)-treated cell. Cells were incubated with 10-8 M PDB for 30 minutes and then stained with a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. Micrographs are representative of results obtained in 2 experiments in which 60 cells were evaluated. Original magnification at 600X, scale bar (D) indicates 50 microns.
50 Figure 11 Control experiments comparing "-actin remodeling in unstimulated cells (Control; A, E); cells treated with 10-8 M phorbol 12,13 dibutyrate (PDB; B, F); cell treated with the inactive phorbol ester, 4"-phorbol 12,13- didecanoate (4" PDD; C, G); and cells treated with norepinephrine which was shown to not stimulate constriction of A7r5 cells (NE; D, H). Cells were incubated with the indicated agents for 30 minutes and then stained with a monoclonal anti-"-actin smooth muscle actin clone 1A4 FITC- labeled antibody. Micrographs are representative of results obtained in 3 experiments in which an average of 90 cells were evaluated for each treatment group. Original magnification at 600X, scale bar (H) indicates 50 microns.
51 Figure 12 A comparison of the distribution of "-actinin in unstimulated cells and cells stimulated to constrict by the addition of phorbol 12,13 dibutyrate (PDB) to medium for 10 minutes or 30 minutes. Cells were stained with a Anti-"- Actinin primary antibody followed by the addition of an Alexa 488-labeled secondary antibody. The results show that the highly granular filamentous organization of "-actinin in unstimulated cells undergoes remodeling to form intensely fluorescing peripheral bodies in PDB-stimulated cells. Micrographs are representative of images obtained in 4 separate experiments in which a total of 168 cells were studied. An average of 82 ± 4% of PDB-treated cells showed "-actinin peripheral bodies. Original magnification at 600X, scale bar (30 minutes) indicates 50 microns.
52 Figure 13 Dual immunostaining for "-actin and "-actinin in unstimulated (control) and phorbol 12,13 dibutyrate-treated (PDB) A7r5 cells. Yellow color indicates colocalization of the two proteins. Cells were stained for "-actin using a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. "-Actinin was visualized using a monoclonal Anti-"-Actinin primary antibody followed by an Alexa 594-labeled secondary antibody. Micrographs are representative of images obtained in 4 experiments in which a total of 44 cells were evaluated. Original magnification at 600X, bar scale indicates 50 microns.
53 Discussion
There are four isoforms of actin that have been identified in mammalian and avian smooth muscle: "- and (-smooth muscle and $- and (- nonmuscle actin. The $-actin isoform is expressed in all smooth tissues examined thus far; whereas, the expression of the remaining isoforms is tissue specific (Hartshorne, 1987). For example, (-actin is found at high concentration primarily in visceral smooth muscle (Hartshorne, 1987;
Vanderkerckhove and Weber, 1981; North et al., 1994) while "-actin is highly expressed in vascular smooth muscle (Fatigati and Murphy, 1984; Hartshorne, 1987). Despite the fact that these different isoforms exhibit 95% homology in their amino acid sequence
(Pollard and Cooper, 1986), there is evidence to suggest that they are segregated into separate functional domains (Rasmussen et al., 1987). Small and coworkers (Small, 1986;
Draeger et al., 1990; North et al., 1994; Small, 1995) first described the separation of actin filaments into “cytoskeletal” and “contractile” domains. Cytoskeletal $-actin filaments were observed to colocalize with cell cytoplasmic dense bodies, membrane dense plaques, the intermediate filament component desmin, and the actin crosslinking protein filamin.
By comparison, contractile filaments containing "- or (-actin are thought to be excluded from the cytoskeletal domain and are found in association with myosin filaments.
The diacylglycerol analogue, PDB induced a robust, slowly developed cellular constriction consistently observed in 85% or more of treated cells. In the 30 minute interval following
PDB addition, cellular constriction was acentric, with greater shortening along the axis
54 described by the densely packed, parallel array of actin stress cables extending across the cell body (Figs. 1, 2, 3). Nevertheless, there was significant constriction along the cell periphery indicating a capability for force development in multiple axes. During the interval of constriction, the actin cytomatrix was observed to undergo extensive and highly organized forms of remodeling. Moreover, the $-actin and "-actin components could be identified as functionally separate entities on the basis of their remodeling characteristics.
Through use of cells expressing $-actin-GFP fusion protein, it was determined that this isoform is incorporated into stress cables aligned in parallel with the principal axis of constriction and that the $-actin cables shorten throughout the interval of cell constriction
(Fig. 3). The localized effect of cytochalasin (Fig. 4) suggests that the length of individual cables could be modified by polymerization/depolymerization and crosslinking of filaments at the cell center. The observation that cytochalasin-induced dissolution of stress cables resulted in an elongation in the x-axis of precontracted cells (Fig. 5) further suggests that the $-actin cables play an important role in maintaining the cell in the contracted configuration in the x-axis.
Staining cells with phalloidin verified the preservation of the extensive system of actin cables in PDB-stimulated cells (Fig. 6) previously noted in the $-actin-GFP studies. In addition, many of the phalloidin-stained cells showed evidence for the formation of discrete peripheral structures not observed in experiments employing $-actin-GFP expressing cells nor in cells stained with anti-$-actin antibody (Fig. 7). These structures were subsequently determined to result from the remodeling of "-actin (Fig. 8).
55 Immunostaining for "-actin revealed that, similar to $-actin, the " isoform was incorporated into densely packed stress cables in unstimulated cells. Stimulation of cell constriction by PDB; however, caused a remarkable disassembly/reassembly of "-actin resulting in the loss of most of the "-actin stress cables and formation of intensely fluorescing rod-like bodies located at the cell periphery (Figs. 8, 9) and extending vertically from the cell base (Fig. 10). These peripheral dense structures showed strong colocalization with "-actinin (Figs. 12, 13) suggesting they are analogous to cytoplasmic dense bodies which anchor oppositely polarized actin filaments of the contractile apparatus (Bond and Somlyo, 1982) in differentiated smooth muscle cells. Consistent with this idea, the peripheral dense structures were associated with remaining "-actin filaments and appeared as focal points from which clusters of randomly oriented filaments originated. Furthermore, the peripheral ring of "-actin/"-actinin dense structures with associated "-actin filaments formed within the first 10 minutes after PDB addition, during the interval cells first begin to constrict. The rapidity of remodeling of the "-actin domain to this specific configuration concurrent with the onset of cell constriction may suggest an active role in force development in PDB-induced contraction.
In the present study, we investigated changes in the actin cytomatrix of A7r5 cells during
PDB-induced contraction. The A7r5 clonal smooth muscle cell line was derived from embryonic rat aorta (Kimes and Brandt, 1976) and has been recently demonstrated to exhibit an adult smooth muscle phenotype (Firulli et al., 1998). These cells show expression and promoter activity of several highly restricted smooth muscle cell markers including: smooth muscle myosin heavy chain, "-actin, calponin (Firulli et al., 1998) and
56 exhibit a high level of expression of SM22 alpha mRNA (Solway et al., 1995) which is found exclusively in smooth muscle tissues and is considered one of the earliest markers of differentiated smooth muscle. A7r5 smooth muscle cells exhibit an inability to proliferate in serum-free media and the absence of PDGF-$ mRNA (Firulli et al., 1998).
They retain their responsiveness to the vasoactive agent arginine vasopressin (Byron,
1996) and their ability to contract in culture. Nakajima et al. (1993) have utilized the polydimethyl siloxane substrate “skin” method (Murray et al., 1990) to demonstrate the development of contractile force by A7r5 cells in response to either high K+ or phorbol esters. They concluded that the contractile response to phorbol ester occurred by two distinct mechanisms, one dependent and one independent of increased intracellular calcium. In our hands, A7r5 cells showed a strong contractile response to the calcium
ionophore A23187 and PDB but were not responsive to either angiotensin II or norepinephrine. This verifies the observation of a functionally intact contractile apparatus while suggesting that some component of the receptor/transduction pathway of the ligand- activated contractile response was absent in A7r5 cells. Hence, many indicators suggest that A7r5 cells represent a suitable model for the study of contractile function in vascular smooth muscle. It must be emphasized; however, that the actin cytomatrix of these cells show many similarities to fibroblasts much like other smooth cell types which revert to a less differentiated phenotype in culture (Chamley-Campbell et al., 1979). In particular, we note that "-actinin is not localized into discreet cytoplasmic dense bodies as seen in highly differentiated smooth muscle cells. Therefore, it is unlikely that present results can be applied directly to contractile phenomenon in highly differentiated smooth muscle.
Nevertheless, the results may provide some initial insight into the potential for actin
57 remodeling relevant to contractile activity.
In summary, A7r5, smooth muscle cells retain the ability to undergo robust cellular constriction to phorbol 12,13 dibutyrate. Prior to stimulation both the "- and $-actin domains of these cells are found primarily in large stress cables than extend the length of the cell body. Upon stimulation both domains undergo characteristic remodeling activity.
$-Actin is retained in stress cables which shorten during the interval of cell constriction.
Cytochalasin-induced dissolution of these stress cables in precontracted cells results in cell elongation in the axis parallel to the system of stress cables leading us to speculate that these structures serve to hold the cell in the contracted configuration. By comparison, "- actin stress cables disassemble and reform in association with "-actinin into column-like structures which ring the cell periphery and appear to be associated with remaining "-actin filament structure. Because of the temporal relationship between the completion of remodeling of the "-actin domain and the onset of observable cell constriction, we further speculate that this specific configuration functions in active force development. One explanation is that the remodeling of the "-actin contractile domain increases mechanical advantage or provides a structural arrangement allowing force development in multiple axes. Because there was no evidence of significant compression of "-actin stress cables and filaments between 10 and 30 minutes after PDB addition (Fig. 8), the results further suggest that these elements continued to remodel during the interval of cell constriction, supporting the contention that remodeling of the contractile domain could modulate optimal actin/myosin interaction and mechanical advantage of the system throughout the interval of contraction. Although not studied in the present investigation, the results
58 further imply significant myosin redistribution in association with "-actin remodeling.
The results demonstrate that the actin cytomatrix undergoes extensive and highly organized remodeling which may play an important role in the contractile response to PDB in A7r5 cells. The present finding may provide initial insight into the extent and characteristics of actin remodeling in contracting cells that do not have a well-defined sarcomere.
References
1. AIZAWA, H., SAMESHIMA, M., & YAHARA, I. (1997) A green fluorescent
protein-actin fusion protein dominantly inhibits cytokinesis, cell spreading, and
locomotion in Dictyostelium. Cell Struct. Funct. 22(3), 335-345.
2. BATTISTELLA-PATTERSON, A. S., WANG, S., & WRIGHT, G. L. (1997) Effect
of disruption of the cytoskeleton on smooth muscle contraction. Can. J. Physiol.
Pharmacol. 75, 1287-1299.
3. BOND, M. & SOMLYO, A. V. (1982) Dense bodies and actin polarity in vertebrate
smooth muscle. J. Cell Biol. 95, 403-413.
4. BUORO, S., FERRARESE, P., CHIAVEGATO, A., ROELOFS, M., SCATENA, M.,
PAULETTO, P., PASSERINI-GLAZEL, G., PAGANO, F., & SARTORE, S. (1993)
Myofibroblast-derived smooth muscle cells during remodeling of rabbit urinary
bladder wall induced by partial outflow obstruction. Lab. Invest. 69, 589-602.
5. BYRON, K. L. (1996) Vasopressin stimulates Ca2+- spiking activity in A7r5 vascular
smooth muscle cells via activation of phospholipase A2. Circ. Res. 78, 813-820.
59 6. CHAMLEY-CAMPBELL, J. H., CAMPBELL, G. R., & ROSS, R. (1979) The
smooth muscle in culture. Physiol. Rev. 59, 1-61.
7. CHOIDAS, A., JUNGBLUTH, A., SECHI, A., MURPHY, J., ULLRICH, A., &
MARRIOTT, G. (1998) The suitability and application of a GFP-actin fusion protein
for long-term imaging of the organization and dynamics of the cytoskeleton in
mammalian cells. Eur. J. Cell Biol. 77(2), 81-90.
8. DRAEGER, A., AMOS, W. B., IKEBE, M., & SMALL, J. V. (1990) The
cytoskeletal and contractile apparatus of smooth muscle:contraction band and
segmentation of the contractile elements. J. Cell Biol. 111, 2463-2473.
9. DRISKA, S. P., AKSOY, M. O., & MURPHY, R. A. (1981) Myosin light chain
phosphorylation associated with contraction in arterial smooth muscle. Am. J.
Physiol. 240, C222-C223.
10. FAIX, J., CLOUGHERTY, C., KONZOK, A., MINTERT, U., MURPHY, J.,
ALBRECHT, R., MUHLBAUER, B., & KUHLMANN, J. (1998) The IQGAP-
related protein DGAP1 interacts with Rac and is involved in the modulation of the F-
actin cytoskeleton and control of cell motility. J. Cell Sci. 111(20), 3059-3071.
11. FATIGATI, V., & MURPHY, R. A. (1984) Actin and tropomyosin variants in
smooth muscles. Dependence on tissue types. J. Biol. Chem. 259, 14383-14388.
12. FIRULLI, A. B., HAN, D., KELLY-ROLOFF, L., KATELIANSKY, V. E.,
SCHWARTZ, S. M., OLSON, E. N., & MIANO, J. M. (1998) comparative
molecular analysis of four rat smooth muscle cell lines. In Vitro Cell Dev. Biol. -
Animal. 34, 217-226.
13. GABELLA, G. (1976) The force generated by visceral smooth muscle. J. Physiol.
60 (Lond). 263, 199-213.
14. GORDON, A. M., HUXLEY, A. F., & JULIAN, F. J. (1966) The variation in
isometric tension with sarcomere length in vertebrate muscle fibers. J. Physiol.
(Lond). 184, 170-192.
15. HARTSHORNE, D. J. (1987) Biochemistry of the contractile process in smooth
muscle. In. Johnson, L(ed) Physiology of the gastrointestinal tract. Raven press.
NY, pp 423-482.
16. JONES, K. A., PERKINS, W. J., LORENZ, R. R., PRAKASH, Y. S., SIEK, G., &
WARNER, D. O. (1999) F-actin stabilization increases tension cost during
contraction of permeabilized airway smooth muscle in dogs. J. Physiol. 519(2), 527-
538.
17. KIMES, B. W. & BRANDT, B. L. (1976) Characterization of two putative smooth
muscle cell lines from rat thoracic aorta. Exp. Cell Res. 98, 349-366.
18. MEHTA, D. & GUNST, S. J. (1999) Actin polymerization stimulated by contractile
activation regulates force development in canine tracheal smooth muscle. J. Physiol.
(Lond). 519(3), 829-840.
19. MERKEL, L., GERTHOFFER, W. T., & TORPHY, T. J. (1990) Disassociation
between myosin phosphorylation and shortening velocity in canine trachea. Am. J.
Physiol. 258, C524-C532.
20. MURPHY, R. A. (1976) Contractile system function in mammalian smooth muscle.
Blood Vessels. 13, 1-23.
21. MURRAY, T. R., MARSHALL, B. E. & MACARAK, E. J. (1990) Contraction of
vascular smooth muscle in cell culture. J. Cell Physiol. 143, 26-38.
61 22. NAKAJIMA, S., FUJIMOTO, M., & UEDA, M. (1993) Spatial changes of [Ca2+]i
and contraction caused by phorbol esters in vascular smooth muscles. Am. J. Physiol.
265(4), C1138-C1145.
23. NOEGEL, A. A. & SCHLEICHER, M. (2000) The actin cytoskeleton of
Dictyostelium:a story told by mutants. J. Cell Sci. 113(5), 759-766.
24. NORTH, A. J., GIMONA, M., LANDO, Z. & SMALL, J. V. (1994) Actin isoform
compartments in chicken gizzard smooth muscle cells. J. Cell Science 107, 445-455.
25. OKADA, A., LANSFORD, R., WEIMANN, J. M., FRASER, S. E. & McCONNELL,
S. K. (1999) Imaging cells in the developing nervous system with retrovirus
expressing modified green fluorescent protein. Exp. Neurol. 156(2), 394-406.
26. PAUL, R. J. (1983) Functional compartmentalization of oxidative and glycolytic
metabolism in vascular smooth muscle. Am. J. Physiol. 244, C399-C409.
27. POLLARD, T. D. & COOPER, J. A. (1986) Actin and actin-binding proteins. A
critical evaluation of mechanisms and functions. Ann. Rev. Biochem. 55, 987-1035
28. RASMUSSEN, H., TAKUWA, Y. & PARK, S. (1987) Protein kinase C in the
regulation of smooth muscle contraction. FASEB J. 1, 177-185.
29. SAPPINO, A. P., MASOUYE, I., SAURAT, J. H. & GABBIANI, G. (1990) Smooth
muscle differentiation in scleroderma fibroblastic cells. Am. J. Pathol. 137, 585-591.
30. SMALL, J. V., FURST, D. O. & DeMAY, J. (1986) Localization of filamin in
smooth muscle. J. Cell Biol. 102, 210-220.
31. SMALL, J. V. (1995) Structure-function relationships in smooth muscle: the missing
links. Bio Essays. 17(9), 785-792.
32. SMALL, J. V. & GIMONA, M. (1998) The cytoskeleton of the vertebrate smooth
62 muscle cell. Acta Physiol. Scand. 164, 341-348.
33. SOLWAY, J., SELTZER, J., SAMAHA, F. F., KIM, S., ALGER, L. E., NIV, Q.,
MORRISEY, E. E., IP, H. S. & PARMACEK, M. S. (1995) Structure and
expression of a smooth muscle cell-specific gene, SM22 alpha. J. Biol. Chem.
270(22), 13460-13469.
34. TOGASHI, H., EMALA, C. W., HALL, I. P. & HIRSHMAN, C. A. (1998)
Carbachol-induced actin reorganization involves Gi activation of Rho in human
airway smooth muscle cells. Am. J. Physiol. 18, L803-L809.
35. VANDEKERCKHOVE, J. & WEBER, K. (1981) Actin typing on total cellular
extracts:a highly sensitive protein-chemical procedure able to distinguish different
actins. Eur. J. Biochem. 113, 595-603.
36. VERKHUSHA, V. V., TSUKITA, S. & ODA, H. (1999) Actin dynamics in
lamellipodia of migrating border cells in the Drosophila ovary revealed by a GFP-
actin fusion protein. FEBS Lett. 445(2-3), 395-401.
37. WRIGHT, G & HURN, E. (1994) Cytochalasin inhibition of slow tension increase
in rat aortic rings. Am. J. Physiol. 267: H1437-H1447.
38. WRIGHT, G. L. & BATTISTELLA-PATTERSON, A. S. (1998) Involvement of the
cytoskeleton in calcium-dependent stress relaxation of rat aortic smooth muscle. J.
Mus. Res. and Cell Motil. 19, 405-414.
39. ZHANG, Y. & MORELAND, R. S. (1994) Regulation of Ca2+ dependent ATPase
activity in detergent-skinned vascular smooth muscle. Am. J. Physiol. 267, H1032-
H1039.
Chapter 3. Myosin Remodeling in the Contracting A7r5 Smooth Muscle Cell
A manuscript submitted to Acta. Physiol. Scand.
M. E. Fultz and G. L. Wright
Abstract
Using confocal microscopy and standard immunohistochemical techniques, changes in the structure of "-actin and $-actin as well as the distribution of myosin II were studied in the contracting A7r5 smooth muscle cell. Prior to stimulation, each of the three proteins were incorporated into filamentous structures extending the length of the cell body. At thirty minutes after stimulation by phorbol 12, 13 dibutyrate (PDBu), the system of densely packed $-actin fibers was retained. By comparison, "-actin and myosin were observed to undergo significant remodeling, characterized by loss in fiber structure and the formation of brightly fluorescing peripheral bodies. Dual immunostaining for "-actin and myosin II revealed strong co-localization of the two proteins prior to stimulation. Following stimulation, myosin II was observed to partially disassociate from "-actin structure beginning at five to ten minutes after PDBu addition and continuing throughout the interval of contraction. The use of cytochalasin B to block actin polymerization or the selective dissolution of "-actin cable structure by thapsigargin produced similar patterns of change in "-actin structure and the localization of myosin. The results support the
64 concept of myosin evanescence and potential for remodeling. The results suggest that myosin undergoes extensive relocalization in association with "-actin remodeling during
PDBu-mediated contraction in the A7r5 smooth muscle cell.
Introduction
Studies of airway smooth muscle indicate that abrupt changes in tissue length result in a dynamic alteration in contractile properties (Pratusevich et al., 1995; Wang et al., 2000).
Immediately following the length change, ability to develop force is decreased. However, if the tissue is held at the new length it recovers the ability to generate force over an interval of minutes. It is proposed that the “adaptation” of smooth muscle to length changes is due to the remodeling of myosin thick filaments to increase the length of individual filaments and/or the number of contractile units in series (Ford et al., 1994;
Pratusevich et al., 1995). The series-to-parallel transition in the myosin thick filament lattice invoked by the concept of myosin remodeling (Seow et al., 2000) could explain potentiation of isometric force and the inverse relationship between force and velocity of isotonic shortening.
Early work (Kelly and Rice, 1968; Schoenberg, 1969) suggested a potential physiological transience of myosin filamentous structure based on variable results obtained by different preparative methods for tissue ultrastructure studies. However, the bulk of subsequent work provided little evidence of significant change in myosin thick filament density in
65 contracted smooth muscle (Somlyo and Somlyo, 1992; Horowitz et al., 1994). One possible explanation is that myosin remodeling does not occur in all smooth muscles.
Several studies of rat anococcygeus muscle, using different experimental approaches have indicated that myosin filaments increase in length or number during contraction
(Godfraind-De Becker and Gillis, 1988; Gillis et al., 1988; Wantanabe et al., 1993; Xu et al., 1997). By comparison, no structural evidence for changes in myosin were observed in guinea pig taenia coli (Godfraind- De Becker, 1988; Xu et al., 1997). In a recent report
Kuo et al. (2001) have utilized conventional electromicroscopy to compare myosin filament density at different intervals in the recovery of force generation in porcine airway smooth muscle subjected to length oscillation. Their results suggest that the initial loss in force generating capacity is correlated with myosin depolymerization and approximate
27% decrease in thick filament density. Subsequently, contractile force and myosin filament density returned to preoscillation values suggesting a causal link between the two parameters.
The finding of myosin lability and remodeling in anococcygeus and airway smooth muscle but not taenia coli opens the question of how broadly this concept applies to different smooth muscle types. As pointed out by Xu et al. (1997), smooth muscles that do not show assembly of myosin on activation may be influenced by additional factors that alter the critical concentration for myosin assembly or act to stabilize filament structure.
Alternatively, the characteristics of myosin remodeling could vary between cell types reflecting subtle differences in contractile function or responsiveness to different stimuli.
In the present study, we investigated the remodeling of myosin in the contracting A7r5
66 smooth muscle cell using standard immunohistochemical techniques and confocal microscopy. The results confirm previous observations that "-actin undergoes extensive remodeling during PDBu-induced contraction. The results further suggest that myosin is closely associated with "-actin during the interval of contraction.
Materials and Methods
Cell Culture. A7r5 smooth muscle cells were obtained from American Type Culture
Collection (Manassas, VA) and cultured in Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 10% fetal bovine serum, 100 units/ml penicillin G, and 100
µg/ml streptomyocin sulfate. Cells were plated onto 75 mm2 culture flasks and grown to approximately 85% confluence. Cultures were maintained at 37°C in a humidified
atmosphere of 5% CO2 in air. The medium was changed every two days and the cells were passaged at least once a week. Cells were detached from culture flasks by addition of a trypsin/EDTA solution in HBSS.
Confocal Microscopy. In experiments employing cell immunostaining, A7r5 cells were plated onto 22 mm by 22 mm #1 glass coverslips placed in six well culture plates and returned to the incubator for a minimum of 24 h to allow for cell attachment and spreading. Cells were fixed and permeabilized by the addition of ice cold acetone for 1 min. The cells were then washed multiple times (3X) with phosphate-buffered saline
(PBS) containing 0.5% TWEEN-20 (PBS-T); pH 7.5, and incubated for 10 min in
67 blocking solution (5% nonfat dry milk in PBS-T). Cells were stained for "-actin by incubation in a 1:500 dilution of monoclonal anti-"-smooth muscle actin clone 1A4 FITC- labeled antibody (Sigma Chemical Co., St. Louis, MO) for 30 min at room temperature.
Cells were stained for $-actin by incubation in a 1:500 dilution of monoclonal anti-$-actin clone AC-15 primary antibody (Sigma Chemcial Co.) for 30 min at room temperature followed by incubation with an Alexa 488-labeled secondary antibody (Molecular Probes,
Eugene, OR). Myosin was visualized using an anti-myosin smooth muscle heavy chain
(G-4): SC-6956 antibody (Sigma Chemical Co) followed by Alexa 488-labeled secondary antibody (Molecular Probes). In dual immunostaining experiments to study the co- localization of "-actin and myosin, Alexa 594-labeled secondary antibody (Molecular
Probes, Eugene, OR) was substituted for Alexa 488-labeled antibody. Stained cells were visualized by mounting on a Nikon Diaphot Microscope and confocal microscopy was performed with a BioRad Model 1024 Scanning System equipped with a krypton/argon laser. Micrographs were built by projecting serial Z-plane image acquisitions and were analyzed using Lasersharp and Confocal Assistant Software (BioRad, Hercules, CA).
Cell Treatments. Actin and myosin cytoskeletal reorganization was studied in A7r5 cells contracted by the addition of 10-8M phorbol 12, 13 dibutyrate (PDBu) to the incubation medium. In order to ensure that the results were not due to a non-specific effect of PDBu, control experiments were also conducted using the inactive 4"-phorbol 12, 13-didecanoate
(4" PDD) at concentrations ranging to 10-6M. Potential effects of nonspecific binding were examined in additional experiments in which cells were incubated with secondary antibody in the absence of primary antibody or in which a 5X molar excess of purified
68 myosin protein (Sigma Chemical Co., St. Louis, MO) was mixed with primary antibody prior to incubation. In ancillary experiments to examine the effect of disruption of the actin cytoskeleton on myosin distribution, cells were treated with 0.1 µg/ml of cytochalasin B prior to and following stimulation with PDBu. Because it has been observed that a sustained high level increase in intracellular calcium ion concentration causes a selective dissolution of "-actin cable structure (Li et al., in preparation), a final series of experiments was conducted to examine the effect of 10-5M thapsigargin on myosin localization in the A7r5 cell.
Results
Previous work (Fultz et al., 2000) has demonstrated that PDBu induces a slowly developed but robust contractile response in A7r5 smooth muscle cells. In excess of 85% of cells respond with visible cell shortening which is first observed at about 8 minutes and continues over an interval of approximately 60 minutes to completion. PDBu-induced cell constriction was extremely robust with the resolution of individual actin fibers often lost by 45 minutes.
Figure 1 shows a comparison of A7r5 cells immunostained for "-actin, $-actin, and myosin II before stimulation and 30 minutes after the addition of PDBu. Prior to stimulation each of these proteins were present in filaments or cables which extended across the cell body. At 30 minutes after addition of 10-8M PDBu, cells stained for $-actin showed an extensive system of densely packed cables similar in appearance to
69 unstimulated control cells. By comparison, "-actin and myosin were observed to have undergone extensive reorganization characterized by the formation of brightly fluorescing peripheral bodies and an apparent reduction in the number of fibers.
Figure 2 shows the time course of change in smooth muscle specific myosin II distribution in contracting A7r5 cells. The results indicate significant reorganization of myosin by 10 minutes. Because there was no evidence of compression of remnant filaments during the interval between 10 and 60 minutes, the results further suggest that the reorganization continued throughout the period of contraction.
Co-localization studies indicated that myosin II is associated with "-actin in unstimulated and contracting A7r5 cells (Figs. 3-7). However, the association is less complete during the interval of contraction and active "-actin remodeling. Unstimulated control cells exhibit clear stress fiber resolution and strong co-localization of "-actin and myosin (Fig.
3). Within 5 minutes after the addition of PDBu, the resolution of myosin stained fibers is reduced although there is continued evidence of a high level of association between the two proteins (Fig. 4). The first indication of visible cell constriction (Fultz et al., 2000) and active "-actin remodeling with initial formation of peripheral bodies (Fig. 5) may be observed at about 10 minutes after PDBu. At this time point, myosin staining was clearly diffuse with evidence of strong co-localization at peripheral bodies but reduced association of myosin with "-actin stress fibers. This pattern of dispersed myosin staining with high levels of co-localization at peripheral bodies was also noted at 30 minutes (Fig. 6) suggesting incomplete myosin association with "-actin during the period of cell
70 contraction and active cytoskeletal remodeling. At 60 minutes after the addition of PDBu, the majority of cells cease to contract or demonstrate only minor further constriction. In those cells in which individual cytoskeletal structure could be resolved, the results suggest a high level of "-actin/myosin association at peripheral bodies and increased co- localization at stress fibers compared to earlier time points during active remodeling (Fig.
7).
Control experiments indicated that the partial dissociation and relocalization of myosin induced by PDBu did not occur in cells treated with inactive 4"-phorbol 12, 13- didecanoate (4" PDD) (Fig. 8). Other experiments in which cells were incubated with secondary antibody in the absence of primary antibody or in which the primary antibody was mixed with a 5X molar excess of purified myosin protein (data not shown) indicated that nonspecific binding had no significant effect on results.
Cytochalasin B caused distinct change in the structural organization of "-actin. The use of cytochalasin B to block actin polymerization resulted in the loss of "-actin stress fiber integrity in unstimulated cells and prevented remodeling formation of peripheral bodies in cells stimulated with PDBu (Fig. 9). Cytochalasin-treated control and PDBu-stimulated cells stained for myosin showed changes in structural organization similar to "-actin (Fig.
10).
Li et al. (in press) have recently shown that sustained elevation of [Ca2+]i by thapsigargin results in partial to complete depolymerization of "-actin but does not affect $-actin stress
71 fibers. Similar to their findings, thapsigargin caused a significant loss in "-actin stress fibers (Fig. 11). Thapsigargin-treated cells stained for myosin showed dissolution of fiber structure similar to "-actin (Fig. 12).
72 Figure 1. A comparison of the "-actin, $-actin, and myosin cytomatrix in unstimulated A7r5 cells (control) and cells stimulated by the addition of phorbol 12, 13 dibutyrate (PDBu) for 30 minutes. "-Actin was visualized using monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. $-Actin was visualized by incubation with anti-$-actin clone AC-15 primary antibody followed by Alexa 488-labeled secondary antibody. Myosin was visualized with anti-myosin smooth muscle heavy chain (G-4):SC-6956 antibody followed by Alexa 488-labeled secondary antibody. Micrographs are representative of images collected in six separate experiments in which a total of 620 PDBu-treated cells were evaluated. In excess of 80% of cells stained for "-actin and myosin showed loss of fibers and relocalization of fluorescence at peripheral bodies after PDBu. Evidence of peripheral bodies was observed in less than 2% of cells stained for $-actin. Original magnification 400X, scale bar indicates 50 microns.
73 Figure 2. The time course of change in myosin localization in the A7r5 cell at intervals up to 60 minutes after stimulation by PDBu. Myosin was visualized using an anti-myosin smooth muscle heavy chain (G-4):SC-6956 antibody followed by Alexa 488-labeled secondary antibody. The results show an early loss of well- defined cable structure at 5 minutes and the first indications of myosin localization at peripheral bodies by 10 minutes. Micrographs are representative of images collected in four separate experiments in which approximately 400 cells were examined at each time point. An average of 78 ± 5% of cells showed evidence of significant loss of cable structure and formation of peripheral bodies by 30 minutes after PDBu. Original magnification at 400X, scale bar indicates 50 microns.
74 Figure 3. Dual immunostaining for "-actin and myosin in unstimulated A7r5 cells. Cells were stained for "-actin using a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. Myosin was visualized using a monoclonal myosin heavy chain primary antibody followed by an Alexa 594-labeled secondary antibody. Yellow color indicates co-localization of the two proteins. Magnification of upper panels at 400X with boxed areas shown in lower panels at 784X. Micrographs are representative of images obtained in 4 separate experiments in which a total of 243 cells were evaluated.
75 Figure 4. Dual immunostaining for "-actin and myosin in A7r5 at 5 minutes after stimulation with PDBu. Cells were stained for "-actin using a monclonal anti- "-smooth muscle actin clone 1A4 FITC-labeled antibody. Myosin was visualized using a monoclonal myosin heavy chain primary antibody followed by an Alexa 594-labeled secondary antibody. Yellow color indicates co- localization of the two proteins. Magnification of upper panels at 400X with boxed areas shown in lower panels at 784X. The results show strong co- localization of the two proteins. Note, however, the diffuse character of myosin staining which has become apparent in the central region of the cell. Micrographs are representative of images collected in 4 separate experiments in which a total of 95 cells were evaluated.
76 Figure 5. Dual immunostaining for "-actin and myosin in A7r5 cells at 10 minutes after stimulation with PDBu. At this time point cells may be visibly observed to have shortened while many show evidence of initial formation of brightly fluorescent peripheral bodies. Cells were stained for "-actin using a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. Myosin was visualized using a monoclonal myosin heavy chain primary antibody followed by an Alexa 594-labeled secondary antibody. Yellow color indicates colocalization of the two proteins. Magnification of upper panels at 400X with boxed areas shown in lower panels at 784X. Although there is clear evidence of "-actin/myosin at developing peripheral bodies the results indicate that myosin association with "-actin cable structure is significantly reduced. Micrographs are representative of images collected in 4 separate experiments in which 125 cells were evaluated.
77 Figure 6. Dual immunostaining for "-actin and myosin in A7r5 cells at 30 minutes after stimulation with PDBu. At this time point, the majority of cells exhibit extensive "-actin remodeling with evidence of reduced cable structure and the appearance of numerous brightly fluorescing peripheral bodies. Cells were stained for "-actin using a monoclonal anti-smooth muscle actin clone 1A4 FITC-labeled antibody. Myosin was visualized using a monoclonal myosin heavy chain primary antibody followed by an Alexa 594-labeled secondary antibody. Yellow color indicates co-localization of the two proteins. Magnification of upper panels is at 400X with boxed areas shown in lower panels at 784X. The results show strong co-localization of "-actin and myosin at peripheral bodies and significant association of myosin with remaining "- actin cables. However, myosin staining did indicate that a significant portion of this protein was diffusely distributed in the cell. Micrographs are representative of images collected in 4 separate experiments in which 122 cells were evaluated.
78 Figure 7. Dual immunostaining for "-actin and myosin in A7r5 cells at 60 minutes after stimulation with PDBu. Cells were stained for "-actin using a monoclonal anti-"-smooth muscle actin clone 1A4 FITC-labeled antibody. Myosin was visualized using a myosin heavy chain primary antibody followed by an Alexa 594-labeled secondary antibody. Yellow color indicates co-localization of the two proteins. Magnification of upper panels is at 400X with boxed areas shown in lower panels at 784X. The results indicate highly significant co- localization of "-actin and myosin at peripheral bodies and remaining "-actin cables. Micrographs are exemplary examples of images obtained in 4 separate experiments in which 165 cells were examined.
79 Figure 8. Control experiments comparing myosin distribution in the A7r5 cell prior to stimulation and at 30 minutes after addition of 10-8M phorbol 12, 13 dibutyrate (PDBu) or the inactive 4-"-phorbol 12, 13 dideconate (4" PDD). Original magnification 400X, scale bar indicates 50 microns.
80 Figure 9. Confocal micrographs showing the effect of cytochalasin B on "-actin structure in unstimulated A7r5 smooth muscle cells and cells activated by the addition of 10-8M phorbol 12, 13 dibutyrate (PDBu) for 30 minutes, "-Actin was visualized using a FITC-labeled anti-"-smooth muscle antibody. Cytochalasin B (0.1 µg/ml) was added 10 minutes prior to the addition of PDBu. Micrographs are representative of images obtained in 3 separate experiments in which at least 150 cells were evaluated for each treatment. Original magnification at 400X, scale bar indicates 50 microns.
81 Figure 10. Confocal micrographs showing the effect of cytochalasin B on the localization of myosin in unstimulated A7r5 cells and cells activated by the addition of 10- 8M phorbol 12, 13 dibutyrate (PDBu) for 30 minutes. Myosin was visualized using an anti-myosin heavy chain primary antibody followed by an Alexa 488- labeled secondary antibody. Cytochalasin B (0.1 µg/ml) was added 10 minutes prior to the addition of PDBu. Micrographs are representative of images obtained in 3 separate experiments in which at least 150 cells were evaluated per treatment. Original magnification at 400X, scale bar indicates 50 microns.
82 Figure 11. Micrographs showing the effect of incubation with 2 X 10-6M thapsigargin on "-actin structure in A7r5 smooth muscle cells. "-Actin was visualized using a FITC-labeled anti-"-smooth muscle antibody. Micrographs are representative of results from 6 separate experiments in which a total of 495 thapsigargin- treated cells were evaluated. It was judged that 88 ± 3% of cells showed partial to complete dissolution of "-actin structure. Original magnification at 400X, scale bar indicates 50 microns. Boxed portions of cells indicate areas shown at 800X in lower panels.
83 Figure 12. Micrographs showing the effect of incubation with 2 X 10-6M thapsigargin on the localization of myosin in A7r5 smooth muscle cells. Myosin was visualized using an anti-myosin heavy chain primary antibody followed by an Alexa 488-labeled secondary antibody. Micrographs are representative of results from 4 separate experiments in which 405 thapsigargin-treated cells were evaluated. It was judged that 83 ± 5% of treated cells exhibited partial to complete loss of filamentous structure. Original magnification at 400X, scale bar indicates 50 microns. Boxed portions of cells indicate area chosen for magnification at 800X in lower panels.
84 Discussion
The results support earlier observations (Fultz et al., 2000) that "-actin and $-actin cellular structures undergo distinctly different modes of remodeling during phorbol ester-induced contraction of A7r5 smooth muscle cells (Fig. 1). They showed that after activation, $- actin remained incorporated into stress fibers which shortened during cell contraction without evidence of individual fiber disassembly. By comparison, "-actin stress fibers underwent disassembly and reformed into numerous discrete peripheral bodies associated with reduced numbers of randomly oriented "-actin fibers. Because of the highly structured nature of active actin remodeling and the temporal relationship of "-actin remodeling to the onset of visible cell contraction, they proposed that the two modes of remodeling reflected different functions. It was suggested that $-actin fibers served to hold the cell in the contracted configuration thus preserving tension gains; whereas, remodeling of "-actin was directly associated with force development. The present finding of "-actin/myosin association throughout the interval of active remodeling is consistent with the proposed role of "-actin remodeling in force development.
Based on the diffuse character of myosin staining, the results suggest that myosin is partially dissociated from actin during active "-actin remodeling (Figs. 5, 6). It is unlikely that these observations represent an artifact of sample preparation or imaging technique because the dispersal of myosin in the cytosol was only observed in the interval following
PDBu and was not apparent in cells treated with inactive 4"PDD (Fig. 8). The results indicate significant remodeling and relocalization of myosin but suggest that this activity
85 may be permissive, depending on "-actin structural dynamics. This conclusion is consistent with the further observation that myosin localization closely follows distinctive changes in "-actin structure produced by low concentrations of cytochalasin B (Fig. 9) or incubation of cells with thapsigargin (Fig. 12).
The present findings support the concept of myosin evanescence (Ford et al., 1994) but argue against increased thick filament density as a primary goal of contractile remodeling
(Kuo et al., 2001). Instead, they suggest a highly coordinated remodeling of the "- actin/myosin lattice consistent with the proposal of continued reorganization of the contractile apparatus during contraction (Wright and Hurn, 1994; Battistella-Patterson et al., 1997; Fultz et al., 2000) of the A7r5 cell. It must be emphasized, however, that there is presently no evidence to indicate that a common mechanism of contractile remodeling occurs among cells lacking a well-defined sarcomere. For example, based on controversy surrounding myosin thick filament densities in contracted tissues, Kuo et al. (2001) have suggested that remodeling of the contractile lattice differs between cell types or may be absent in some cells. Irrespective of differences between cell types, we consider it likely that major differences will be observed in the contractile remodeling in isolated cells versus studies with tissues. Cells in vivo would be expected to experience some degree of limitation in contractile shortening due to their preload and the constraining influences of the extracellular matrix and cell/cell adhesion. It has been reported that some types of smooth muscle can decrease their length in vitro by as much as 75% (Gabella, 1984) and isolated cells have been observed to contract into a spherical shape (Mamose and Gomi,
1977). We recently reported a similar phenomenon in A7r5 cells, noting that within 45 to
86 60 minutes after additions of phorbol ester the majority of cells constricted to a point that the resolution of individual actin fibers was lost (Fultz et al., 2000). This suggests that contraction and cytoskeletal remodeling in the unloaded cell may continue unabated until constrained by coalescence of the cytoskeleton. Thus, contractile remodeling in the isolated, unloaded cell may represent a greatly exaggerated response compared to that of preloaded tissue.
Hence, it is uncertain to what extent findings in A7r5 cells are applicable to mechanisms underlying the contractile properties of highly differentiated smooth muscle. The A7r5 clonal smooth muscle cell line was originally derived from embryonic rat aorta (Kimes and Brandt, 1976) and exhibits an adult smooth muscle phenotype (Firulli et al., 1998).
These cells show expression and promoter activity of highly restricted smooth muscle cell markers including: smooth muscle myosin heavy chain, "-actin, calponin (Firulli et al.,
1998) and exhibit a high level of expression of SM22 " mRNA (Solway et al., 1995) which is expressed exclusively in smooth muscle and is considered one of the earliest markers of differentiated smooth muscle. A7r5 cells are unable to proliferate in serum- free media and in the absence of PDGF-$ mRNA (Firuilli et al., 1998). They retain responsiveness to vasoactive arginine vasopressin (Byron, 1996) and the ability to contract by both increased intracellular calcium-dependent and-independent mechanisms
(Nakajima et al., 1993; Fultz et al., 2000). Therefore, there are many indicators to suggest that A7r5 may serve as a model for the study of contractile function in smooth muscle.
Nevertheless, the actin cytomatrix of these cells shows many structural similarities to fibroblasts much like other smooth muscle cell types which revert to a less differentiated
87 phenotype in culture (Chamley-Campbell et al., 1979). In particular, we note significant gross structural similarity between the peripheral bodies of contracting A7r5 cells and
“aster-like assemblies” (Verkhovsky et al.1997) in fibroblasts contracted by treatment with cytochalasin. Verkhovsky et al. (1997) explained the formation of these structures as due to actin/myosin interaction resulting in actin filament polarity sorting and formation of radial arrays of actin filaments with barbed ends oriented toward clusters of myosin.
These findings are consistent with a network arrangement of actin/myosin and could explain nonsarcomeric tension development if dynamic uncoupling of actin filament crosslinking and filament remodeling are presumed.
In summary, the use of confocal microscopy and standard immunohistochemical staining techniques have enabled a relatively detailed evaluation of changes in the localization of smooth muscle specific myosin II in A7r5 cells during PDBu-induced contraction. The results indicate that myosin undergoes dramatic reorganization in association with "-actin but not $-actin structural remodeling. The results are consistent with earlier conclusions that the remodeling of "-actin is an important factor in force development by cells which lack a highly structured sarcomere.
References
Battistella-Patterson AS, Wang S and Wright GL. 1997. Effect of disruption of the
cytoskeleton on smooth muscle contraction. Can J Physiol 75: 1287-1299.
Byron KL. 1996. Vasopressin stimulates Ca2+-spiking activity in A7r5 vascular smooth
88 muscle cells via activation of phospholipase A2. Circ Res 78: 813-820
Chamley-Campbell JH, Campbell GR and Ross R. 1979. The smooth muscle cell in
culture. Physiol Rev 59: 1-61.
Firulli AB, Han D, Kelly-Roloff L, Kateliansky VE, Schwartz SM, Olson EN and Miano
JM . 1998. Comparative molecular analysis of four rat smooth muscle cell lines.
In Vitro Cell Dev Biol Anim 34: 217-226.
Ford LE, Seow CY and Pratusevich VR. 1994. Plasticity in smooth muscle. Can J
Physiol Pharmacol 72: 1320-1324.
Fultz ME, Li C, Geng W and Wright GL. 2000. Remodeling of the actin cytoskeleton in
the contracting A7r5 smooth muscle cell. J Mus Res Cell Motil 21: 775-787.
Gabella G. 1984. Smooth muscle cell junctions and structural aspects of contraction.
Brit Med Bull 35: 213-218.
Gillis JM, Cao ML and Godfraind-De Becker A. 1988. Density of myosin filaments in
the rat anococcygeus muscle, at rest and in contraction. II. J Mus Res Cell Motil 9:
18-28.
Godfraind-De Becker A and Gillis JM. 1988. Analysis of the birefringence of the smooth
muscle anococcygeus of the rat, at rest and in contraction. I. J Mus Res Cell Motil
9: 9-17.
Horowitz A, Trybus KM, Bowman DS and Fay FS. 1994. Antibodies probe for folded
monomeric myosin in relaxed and contracted smooth muscle. J Cell Biol 126:
1195-1200.
Kelly RE and Rice RV. 1968. Localization of myosin filaments in smooth muscle. J Cell
Biol 37: 105-116.
89 Kimes BW and Brandt BL. 1976. Characterization of two putative smooth muscle cell
lines from rat thoracic aorta. Exp Cell Res 98: 349-366.
Kuo K-H, Wang L, Paré PD, Ford LE and Seow CY. 2001. Myosin thick filament lability
induced by mechanical strain in airway smooth muscle. J Appl Physiol 90: 1811-
1816.
Li C, Fultz ME, Parkash J, Rhoten WB and Wright GL Ca2+-dependent actin remodeling
in the contracting A7r5 cell. J Mus Res Cell Motil (in press).
Momose K and Gomi Y. 1977. Studies on isolated smooth muscle cells. I. Continuous
observation of single smooth muscle cells isolated from the vas deferens of guinea-
pig. Chem Pharm Bull 25: 2449-2451.
Nakajima S, Fujimoto M and Veda M. 1993. Spatial changes of [Ca2+]i and contraction
caused by phorbol esters in vascular smooth muscles. Am J Physiol 265: C1138-
C1145.
Pratusevich VR, Seow CY and Ford LE. 1995. Plasticity in canine airway smooth
muscle. J Gen Physiol 105: 73-94.
Schoenberg CF. 1969. An electronmicroscope study of the influence of divalent ions on
myosin filament formation in chicken gizzard extracts and homogenate. Tiss Cell
1: 83-96.
Seow CY, Pratusevich VR and Ford LE. 2000. Series-to parallel transition in the filament
lattice of airway smooth muscle. J Appl Physiol 89: 869-876.
Solway J, Seltzer J, Samaha FF, Kim S, Alger LE, Niv Q, Morrisey EE, IP HS and
Parmacek MS. 1995. Structure and expression of smooth muscle cell-specific
gene, SM22". J Biol Chem 270: 13460-13469.
90 Somlyo AP and Somlyo AV. 1992. Smooth muscle structure and function. In: The Heart
and Cardiovascular System. H.A. Fozzard, ed., Raven Press, NY pp 1293-1324.
Verkhovsky AB, Sritkina TM and Bousy GG. 1997. Polarity sorting of actin filaments in
cytochalasin-treated fibroblasts. J Cell Sci 110: 1693-1704.
Wang L, Paré PD and Seow CY. 2000. Effect of length oscillation on the subsequent
force development in swine tracheal smooth muscle. J Appl Physiol 88: 2246-
2250.
Watanabe M, Takemori S and Yagi N. 1993. X-ray diffraction study on mammalian
visceral smooth muscles in resting and activated states. J Mus Res Cell Motil 14:
469-475.
Wright G and Hurn E. 1994. Cytochalasin inhibiton of slow tension increase in rat aortic
rings. Am J Physiol 267: H1437-H1446.
Xu J-Q, Gillis JM and Craig R. 1997. Polymerization of myosin on activation of rat
anococcygeus smooth muscle. J Mus Res Cell Motil 18: 381-393.
91 Chapter 4. Correlation of Force Development with a-actin Remodeling in Smooth Muscle: Effects of Specific Protein Kinase Inhibitors.
A manuscript prepared for submission
M.E. Fultz and G.L. Wright
Summary
Remodeling of "-actin and $-actin is markedly different in the contracting A7r5 cell (Fultz et al., 2000) suggesting that the two isoforms may play different functional roles. Based on remodeling characteristics it was proposed that "-actin is involved in force development while $-actin stress cables serve to hold tension in the cell. In the present study, we examined the effect of seven protein kinase inhibitors on the relationship between force developed, tension maintenance, and "-actin remodeling. Inhibition of
PKC (10-9M bisindolymaleimide I, -75%) and tyrosine kinase (2.5 x 10-6M genistein, -
62%) significantly decreased early force development but had little subsequent effect on tension maintenance. By comparison, inhibition of calmodulin kinase II (3.7 x 10-6M KN-
93) only moderately reduced force development (-32%) but significantly decreased tension maintenance (-51%). At concentrations expected to block both PKC and calmodulin kinase II (7 x 10-7M), the ser/thr kinase inhibitor staurosporine significantly decreased both force development (-70%) and tension maintenance (-74%) in rat aortic rings. Each of the
92 active kinase inhibitors reduced "-actin remodeling in A7r5 cells with regression analysis indicating a significant r = 0.77, p < 0.01) relationship between force development and "- actin reorganization. By comparison, the correlation between tension maintenance and "- actin remodeling (r = 0.50) was not statistically significant. The results suggest that force development and tension maintenance are regulated differently and support the proposal that force development is linked to "-actin remodeling.
Introduction
The principal biochemical and biophysical pathway of smooth muscle contraction is thought to be initiated in a manner similar to skeletal muscle. Stimulation of smooth muscle results in an elevation of intracellular calcium which then leads to activation of calmodulin. The calcium-calmodulin complex activates myosin light chain kinase which results in the phosphorylation of Ser 19 on the myosin light chain. Phosphorylation of the myosin light chains results in myosin ATPase activity and subsequent myosin/actin crossbridge formation and cycling occurs. However, it has been shown that slow force development, maintenance of tension (Driska et al., 1981) and velocity of shortening
(Merkel et al., 1990) are disassociated from myosin light chain phosphorylation and myosin ATPase activity (Zhang and Moreland, 1994) in smooth muscle. Also noteworthy, it has been demonstrated that phorbol ester-induced contraction can occur in the absence of a detectable increase of intracellular Ca2+ (Nakajima et al., 1993) or concomitant increased myosin light chain phosphorylation (Singer and Baker, 1987).
A number of protein kinases not presently associated with the established smooth
93 muscle pathway of contraction have been shown to alter the contractile response. It has been shown that Protein Kinase C (PKC) is tonically activated in response to a variety of contractile agonists (Haller et al., 1990; Singer et al., 1992); whereas, the inhibition or down-regulation of this enzyme blocks or attenuates the contractile response to different agonists (Merkel et al., 1991, Shimamoto et al., 1993, Wright and Hurn, 1994). Also of interest is the finding that the inhibition of tyrosine kinase (Di Salvo et al., 1993; Di Salvo et al., 1994; Gokita et al., 1994; Sriuasta and St-Louis, 1997; Tolloczko et al., 2000;
Duarte et al., 1997) blocks contraction in smooth muscle to various agonists which has been suggested to reflect an important role of this kinase in electromechanical coupling or the direct activation of the contractile apparatus by Ca2+ in the cell (Di Salvo et al., 1993,
Di Salvo et al., 1994, Tolloczko et al., 2000, Gokida et al., 1994). Finally, Rokolye and
Singer (2000) have shown that the inhibition of calmodulin kinase II (CAMK II) has variable effects on early force development and tension maintenance depending on the agonist used to activate contraction. They concluded that KN-93, a calmodulin kinase II inhibitor, may play a role in regulating maintained tension in activated smooth muscle and that this inhibition could be abolished by activation of PKC.
Recent interest has focused on the role of cytoskeletal organization on smooth muscle contractile properties. Wright and Hurn (1994) first demonstrated that the use of cytochalasin to inhibit actin polymerization caused a selective blockade of the slow tension increase in rat aortic smooth muscle. They suggested that the slow tension development by smooth muscle required a capability for dynamic remodeling of a portion of the actin cytoskeleton during the interval of contraction. Their conclusion has been supported by recent studies showing that inhibition of actin polymerization by latrunculin-
94 A depressed force development without affecting myosin light chain phosphorylation, further suggesting that actin polymerization contributes directly to force development
(Mehta and Gunst, 1999). Other work (Wright and Battistella-Patterson, 1998) has shown that inhibition of actin polymerization blocks calcium-dependent stress relaxation in vascular smooth muscle, suggesting that remodeling of the actin cytoskeleton may also serve to decrease cellular tension following abrupt stretch. Taken together, the results suggest that the effects of inhibition of actin polymerization on the contractile and stress relaxant properties of smooth muscle cannot be attributed to a simple loss in cytoskeletal structure but more accurately reflects the inhibition of dynamic actin cytoskeletal remodeling.
Battistella-Patterson et al. (1997) have proposed that actin remodeling could serve to modulate the optimal positioning of actin and myosin filaments during the interval of force development and hold the cell in the constricted configuration during tension maintenance. Recently Fultz et al. (2000) have defined the A7r5 smooth muscle cell line as a model for actin remodeling in cells that lack a highly structured sarcomere. They showed that "-actin and $-actin undergo highly structured and distinctly different modes of remodeling during contraction. Their findings have provided support for the concept of a direct role of actin remodeling in force development (Wright and Hurn, 1994; Battistella-
Patterson et al., 1997; Mehta and Gunst, 1999) and further suggest that $-actin stress cables maintain contractile tension while "-actin remodeling is linked to force development.
In the present study, we utilized various kinase inhibitors to demonstrate that attenuated force development and tension maintenance may also alter the remodeling of
95 the "-actin cytoskeleton. Our results indicate a correlation between "-actin remodeling and force development, further implicating the reorganization of the actin cytoskeleton as an important component in smooth muscle contraction.
Materials and Methods
Animals. Male, 12 week old Sprague Dawley rats were utilized in cell experiments including aortic ring segment contractility. All rats were housed in rooms maintained at
23 ± 2°C with a 12:12 hour light-dark cycle. Animals were allowed unrestricted access to
Purina Rat Chow and tap water.
Contractility Measurements. The evaluation of in vitro contractile response was performed as previously described (Huang et al., 1988). Rats were anesthetized with a
Ketamine-xylazine mixture (21.9 mg/kg) and the thoracic aorta was removed, placed into
warm Krebs buffer (118mM NaCl, 4.7mM KCl, 1.5mM CaCl2, 25mM NaHCO3, 1.1mM
MgCl, 1.2 mM KH2PO4 and 5.6mM glucose, pH 7.4), maintained at 37°C and aerated with
a 95% O2/5% CO2 mixture. Tissues were cleaned of adherent tissue and cut into rings approximately 0.3 cm in length. Rings were denuded of endothelium and mounted on stainless steel hooks under 1.0g of passive tension in glass organ baths containing KREBS
buffer maintained at 37°C and aerated with 95% O2 /15% CO2. The tissues were allowed to equilibrate for a minimum of 1 hour before undergoing a reference contraction to 80mM
K+. After the reference contraction, the tissues were returned to baseline tension by
96 rinsing with fresh buffer. The tissues were then allowed to equilibrate for 30 minutes, kinase inhibitor was then added for an additional 30 minutes and the tissue contracted by addition of 10-8M phorbol 12, 13 dibutyrate (PDBu). The force generated was recorded at the plateau in tension (approximately 30 minutes) as a measure of early force development
(Fig. 1). Force was again recorded at 90 minutes for comparison with the 30 minute value as a measure of ability to maintain tension. Isometric tension was recorded using a force- displacement transducer (Grass FTD3) and a Grass Model 7D polygraph.
A total of seven protein kinase inhibitors were investigated: staurosporine, a general serine/threonine kinase inhibitor; bisindolymaleimide II, highly specific for protein kinase
C(PKC); KN-93, highly specific for calmodulin kinase II (CAMKII); H-89, selective for protein kinase A (PKA); ML-7, selective for myosin light chain kinase (MLCK); genistein, a general tyrosine kinase inhibitor; and AG1478, a tyrosine kinase inhibitor selective for epidermal growth factor receptor kinase (EGFR). To ensure effective inhibition, each
4 compound was added at a final concentration of 10X its reported IC50 or K value (Table
1). Because it was reasoned that inhibition of contractile responsiveness by different kinase pathways might have variable effects on "-actin remodeling, a single inhibitor
(genistein) was selected for study at 0.1X, 1.0X and 10.0X of its IC50 concentration.
Cell Culture. A7r5 cells, derived from embryonic rat aorta, expressing smooth muscle markers (Firulli et al., 1998), and maintaining the ability to contract in response to phorbol esters (Fultz et al., 2000; Nakajima et al., 1993), were obtained from American Type
Culture Collection (Manassas, VA). Cells were cultured on 75 mm2 flasks and grown to
97 approximately 85% confluence at a maintained temperature of 37°C in a humidified
atmosphere of 5% CO2 in air. The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) that was supplemented with 10% fetal bovine serum, 100 units/ml penicillin G, and 100 µg/ml streptomycin. Media was changed every other day and cells were passaged at least once per week. Passaging was accomplished by addition of a trypsin/EDTA solution in HBSS to the culture flask and subsequent collection of the detached cells.
Confocal Microscopy. In experiments where immunostaining was performed, A7r5 cells were seeded onto glass coverslips, placed in 6 well culture plates and returned to the incubator for a minimum of 24 hours to allow for cell attachment and spreading. After experimental treatment, cells were fixed and permeabilized by the addition of ice cold acetone for 1 minute. The cells were washed several times in phosphate-buffered saline
(PBS) containing 0.5% TWEEN-20 (PBS-T) pH 7.5 followed by incubation with a blocking solution containing 5% nonfat dry milk in PBS-T for 10 minutes. "-Actin staining was accomplished by the incubation of the cells with a 1:500 dilution of monoclonal-"-smooth-muscle actin, clone 1A4, FITC-labeled antibody (Sigma Chemical
Co., St. Louis, MO) for 30 minutes at room temperature. Cells stained for $-actin were incubated in a 1:500 dilution of monoclonal anti-$-actin clone AC-15 primary antibody
(Sigma) for 30 minutes followed by incubation with an Alexa 488 labeled secondary antibody (Molecular Probes, Eugene, OR). Cells were visualized by mounting on a Nikon
Diaphot Microscope and confocal microscopy performed with a BioRad Model 1020 scanning system with a Krypton/Argon laser. Micrographs were constructed by projection
98 of Z-plane acquisitions and analyzed by Lasersharp and Confocal Assistant software
(BioRad, Hercules, CA).
Cell Treatments. The effect of various kinase inhibitors on actin remodeling was studied
in A7r5 cells by the addition of either 0.1X, 1.0X or 10.0X the reported IC50, or the Ki of the particular kinase of interest (Table 1) at 30 minutes prior to cell stimulation. Control cells received vehicle only. The actin cytoskeletal reorganization and cell contraction was then stimulated by the addition of phorbol 12, 13 dibutyrate (PDB) to the incubation media at a final concentration of 10-8M. Cells were typically allowed to undergo phorbol- induced contraction for 15 minutes before fixation and immunohistochemistry.
Data Analysis and Statistics. Each experiment evaluating "-actin remodeling was performed a minimum of 3 times with a total of approximately 300 cells evaluated per experiment. Cells were scored as having undergone "-actin remodeling if there was a significant loss of stress fibers and the presence of "-actin peripheral dense bodies. Cells were evaluated independently by 2 viewers and the counts averaged. Analysis of data was performed by one factor ANOVA followed by Tukey test utilizing Sigma Plot and Sigma
Stat Software (SPSS, Chicago, IL). Differences were considered significant if p < 0.05.
Data are presented as mean ± SEM.
99 Results
Contractile effects of protein kinase inhibitors. Individual protein kinase inhibitors affected early force development and tension maintenance by aortic rings in a variable fashion (Table 2, Fig. 1). PDB caused robust force development with maximum tension at about 158% of that obtained in reference contractions to high K+ (Table 2). Of seven protein kinase inhibitors investigated staurosporine (ser/thr, -70%), bisindolymaleimide I
(PKC, -75%), and genistein (tyrosine kinase, -98%) significantly reduced early force development. The CAMKII inhibitor KN-93 caused a moderate (-31%) reduction in force development compared to controls but the difference was not statistically significant. The remaining inhibitors of PKA, MLCK, and the highly specific tyrosine kinase EGFR had no detectable effect on early force development.
A comparison of control values (Table 2) from experiments with aortic rings indicates that the force generated in response to PDBu was well maintained, averaging less than a 7% decrease in tension between 30 and 90 minutes. Of the two kinase inhibitors shown to partially block early force development, staurosporine caused a significant (-75%) loss in tension maintenance between 30 and 90 minutes; whereas, the specific PKC inhibitor bisindolymaleimide I had no effect on this parameter. Similarly, it was observed that low concentrations of the tyrosine kinase inhibitor genistein which partially blocked early force development had no effect on subsequent maintenance of tension (Table 3). Interestingly, the CAMKII inhibitor KN-93, which only moderately reduced early force development, caused a significant (-52%) loss in ability to maintain tension (Table 2). The remaining
100 protein kinase inhibitors; H-89, ML-7, and AG1478 had no detectable effect on tissue tension maintenance (Table 2).
Effects of protein kinase inhibitors on "-actin remodeling in A7r5 cells. Figure 2 shows PDB-induced remodeling of "-actin in the A7r5 cells and the appearance of these cells treated with protein kinase inhibitor shown to block early force development in rat aortic smooth muscle. Prior to stimulation, "-actin forms stress cables which span the cell body (Fig. 2A). Within 8 to10 minutes after the addition of phorbol, the "-actin stress cables disassemble and reform into brightly fluorescing peripheral bodies with a reduced number of associated filaments (Fig. 2B). By comparison, cells treated with kinase inhibitors (Fig. 2 C-F) did not show loss of cable structure or formation of peripheral bodies, providing an unambiguous indicator of lack of "-actin remodeling.
Each of the protein kinase inhibitors shown to alter smooth muscle contractile responsiveness (staurosporine, -69.8%; bisindolymaleimide I, -95.7%; KN-93, -92.2%; genistein, -92.2%) significantly reduced the percent of A7r5 cells undergoing "-actin remodeling in response to PDB (Table 3, 4). Regression analysis of percentage of cells showing remodeling against the percentage of early force development in tissues treated with the different kinase inhibitors indicated a significant correlation (r = 0.77) between the two parameters (Fig. 3). A similar analysis of results, from studies of the individual kinase inhibitor genistein, further indicated a high level of correlation (r = 0.95) between force development and "-actin remodeling (Fig. 4). In contrast, regression analysis of the reduction in tension between 30 and 90 minutes in aortic rings and the percentage of cells
101 showing "-actin remodeling indicated no significant relationship (r = -0.50, NS) between these parameters (Fig. 5).
102 Figure 1 Tracings from contractile response of rat aortic rings in response to 10-8M PDB, showing the effects of kinase inhibitors (broken line) on isometric force generation and maintenance of tension compared to control (solid line) tissues. The inhibitors shown are (A) staurosporine, (B) bisindolymaleimide I, (C) KN-93, (D) ML-7, (E) H-89, and (F) genistein.
103 Figure 2 The effect of (C) genistein, (D) staurosporine, (E) KN-93, and (F) bisindolymaleimide I on the phorbol induced remodeling of "-actin. Control cells were given vehicle (A) or stimulated by the addition of 10-8M
PDB for 10 minutes (B). Inhibitors were added (10X IC50, 30 minutes) before stimulation by PDB. Inhibitor treated cells did not undergo the characteristic "-actin remodeling and the formation of peripheral dense bodies observed in control cells. Original magnification at 400X, scale bar (F) indicates 50 microns.
104 Figure 3 Regression analysis of maximal tension development in rat aortic rings and percent control of A7r5 cells showing "-actin remodeling after treatment with different kinase inhibitors. Data are calculated as percent of control values from untreated tissues or cells in each experiment. Each point represents the average of 3 experiments with a single compound studied at
10X its IC50 or its Ki value.
105 Figure 4 Regression analysis of maximum tension developed in rat aortic rings and percent of A7r5 cells showing "-actin remodeling after treatment with the
protein tyrosine kinase inhibitor genistein at 0, 0.1X, 1X, and 10X its IC50 concentration. Data are calculated as percent of the control value.
106 Figure 5 Regression analysis of deterioration in force maintenance in rat aortic rings and percent of A7r5 cells showing "-actin remodeling. After treatment with different kinase inhibitors "-actin remodeling is calculated as % controls values from untreated cells. Deterioration in force is calculated as the change in maximum tension and tension maintenance at 90 minutes after PDB stimulation. The data point for genistein was dropped due to the total inhibition of contractile response. Each point represents the mean of 3 experiments with a single compound at its 10X concentration.
107 Table 1. The IC50, Ki concentration of different kinase inhibitors employed in present studies.
Inhibitor (Target Kinase) IC50 , Ki
Staurosporine (Ser/Thr) 7.0 x 10-8 M Bisindolymaleimide I (PKC) 1.0 x 10-10 M KN-93 (CAM II) 3.7 x 10-7 M H-89 (PKA) 5.0 x 10-10 M ML-7 (MLCK) 3.0 x 10-11 M Genistein (Tyrosine Kinase) 2.5 x 10-6 M AG1478 (EGFR) 3.0 x 10-9 M
108 Table 2. Effects of different protein kinase inhibitors on early force development (30 minutes) and tension maintenance (90 minutes) in control aortic rings and
rings incubated with different protein kinase inhibitors (10X IC50) for 30 -6 minutes prior to contraction with 10 M PDBu.
30 Minute 90 Minute Inhibitor (Target) Control Inhibitor Control Inhibitor
Staurosporine 141 ± 20 43 ± 12* 134 ± 23 11 ± 4†
(Ser/Thr) Bisindolymaleimide I 163 ± 22 41 ± 18* 134 ± 37 35 ± 19
(PKC) KN-93 (CAMK II) 158 ± 13 109 ± 16 149 ± 15 53 ± 15† H-89 (PKA) 129 ± 13 131 ± 10 120 ± 15 104 ± 17 ML-7 (MLCK) 155 ± 11 136 ± 15 148 ± 11 117 ± 21 Genistein (Tyrosine 150 ± 18 3 ± 2* 139 ± 20 3 ± 2
Kinase) AG1478 (EGFR) 209 ± 12 179 ± 29 198 ± 19 158 ± 36
Values are calculated as the percent of the reference contraction to 80 mM potassium. An
asterisk (*) or cross (†) indicates a significant difference from the 30 minute control value and
between the values obtained for inhibitor treated tissues at 30 and 90 minutes, respectively, p <
0.05.
109
Table 3. The effect of the tyrosine kinase inhibitor genistein on early force development (30 minutes) and tension maintenance (90 minutes) in rat aortic rings contracted with 10-8 PDB. Genistein concentrations were tested
at 0.1X, 1X, and 10X of the published IC50 value.
Control 0.1X 1X 10X
Early Force Development (30 minutes) 150 ± 18 116 ± 19 57 ± 21* 3 ± 2*
Tension Maintenance 139 ± 20 105 ± 20 44 ± 19 3 ± 2 (90 minutes) (7.4%) (9.5%) (22.9%) (0)
Values are calculated as percent of the reference contraction to 80 mM potassium. An asterisk indicates a
significant difference in early force development, p < 0.05. Values in parenthesis indicate % decline in
tension between 30 and 90 minutes.
110 Table 4. Effect of several kinase inhibitors (10X IC50) on the number of cells showing successful "-actin remodeling after PDB stimulation. Data are presented as percentage of values obtained for vehicle controls in each experiment.
Inhibitor (Target Kinase) "-Actin Remodeling, %
Staurosporine (Ser/Thr) 30.2 ± 2.0* Bisindolymaleimide I (PKC) 4.3 ± 1.6* KN-93 (CAMK II) 6.6 ± 3.3* H-89 (PKA) 97.5 ± 11.9 ML-7 (MLCK) 97.9 ± 6.3 Genestein (Tyrosine Kinase) 7.8 ± 3.1* AG1478 (EDGFR) 77.0 ± 9.3
* Indicates significant difference from control, p < 0.05.
111 Discussion
Battistella-Patterson et al. (1997) proposed that the broad length-tension relationship, slow force development and the low energy cost of maintained tension in smooth muscle could be explained by remodeling of the actin cytoskeleton to modulate optimal actin/myosin overlap and to form a stable system of tension bearing elements to hold the cell in the contracted configuration. They modeled this concept in terms of asynchronous contraction in which it was proposed that different portions of the actin cytoskeleton undergo alternating active contraction, tension maintenance, and remodeling. Subsequently, Fultz et al. (2000) have shown "-actin and $-actin remodel differently in the contracting A7r5 cell. Both actin isoforms are incorporated into stress cables in the unstimulated cell.
Upon stimulation, $-actin cables shorten without evidence of disassembly or new cable formation. By comparison, "-actin cables were observed to undergo dramatic disassembly and reform into a ring of column-like structures located at the cell periphery, in association with randomly oriented "-actin cables and filaments. Importantly, neither the $-actin or
"-actin components showed evidence of compression indicating that remodeling of the system continued throughout the interval of cell contraction. Based on these findings,
Fultz et al. (2000) concluded that their results supported the earlier proposal by Battistella-
Patterson et al. (1997) but suggested that $-actin remodeling served to hold tension, whereas, "-actin remodeling was directly linked to active force development. Present findings confirm the work of others demonstrating that several classes of protein kinase inhibitors may significantly attenuate the contractile response in vascular smooth muscle.
112 The results show further that specific inhibitors may selectively reduce early force development or tension maintenance, suggesting the two activities may be differently regulated. Inhibition of PKC (bisindolymaleimide I) and tyrosine kinase (genistein) both reduced early force development with little evidence of a subsequent effect on tension maintenance (Table 2, 3). By comparison, the inhibition of CAMKII only moderately reduced force development but significantly decreased tension maintenance. The conclusion of separately determined force development and tension maintenance was further strengthened by the observation that the ser/thr kinase inhibitor staurosporine, which was expected to block both PKC and CAMKII at the concentration used, significantly decreased both contractile parameters. Finally, several of the kinase inhibitors investigated showed no activity in reducing either force development or tissue maintenance, suggesting the results were due to specific kinase inhibition.
Regression analysis indicated a significant correlation in the relationship between early force development and "-actin remodeling among tissues and cells treated with different kinase inhibitors (Fig. 3) and those treated with several concentrations of the individual inhibitor, genistein (Fig. 4). The finding that inhibition of three classes of kinase which reduce contractile force also block "-actin remodeling may be suggesting that several biochemical pathways regulate "-actin remodeling. However, it is difficult to envision a mechanism based solely on enzymatic regulation that would precisely modulate cytoskeletal remodeling to accommodate the mechanical requirements of the cell. In studies of stress relaxation of activated smooth muscle, Wright and Battistella-Patterson
(1998) concluded that the isometric loss in tension was due to remodeling of the actin
113 cytoskeleton and that the magnitude of relaxation was associated with the degree of strain placed on the tension-bearing elements of the system. Noteworthy from present results,
A7r5 cells treated with kinase inhibitors did not show compression or dissolution of "- actin structure but were similar in appearance to unstimulated control cells. This suggests that direct biochemical regulation of "-actin remodeling may play a permissive role while perturbations in mechanical strain are obligatory for initiation of reorganization in the system. Hence, the surprising level of correlation between force development and "-actin remodeling among tissues and cells treated with different classes of kinase inhibitors may reflect either a direct effect on actin remodeling or an indirect effect through attenuation of initial force development and perturbation in the distribution of tension in the system.
Present findings indicate no significant relationship between "-actin remodeling and tension maintenance. Based on the observation that the addition of cytochalasin prior to initiation of contraction blocked slow force development but had little effect in precontracted tissue, Wright and Hurn (1994) concluded the results were consistent with action against a semistable structure undergoing remodeling during tension development with stabilization and cessation of remodeling at the plateau in tension. It was subsequently proposed that tension maintenance could be explained by the remodeling of actin to a stable structure holding the cell in the constricted configuration (Battistella-
Patterson et al., 1997; Fultz et al, 2000). Because of the marked effect of KN-93 on the loss in tension maintenance in aortic rings, the results may be suggesting a role of calmodulin kinase II in the stabilization of tension bearing actin filaments in PDBu contracted cells.
114 In summary, the inhibition of PKC and tyrosine kinase markedly reduced early force development without affecting tissue maintenance of tension. By comparison, the inhibition of CAMKII only moderately reduced force development but significantly reduced tension maintenance. Force development but not tension maintenance was significantly correlated with "-actin remodeling in similarly treated A7r5 smooth muscle cells. The results suggest that the mechanisms which regulate force development and tension maintenance are different and are consistent with earlier proposals for a direct role of "-actin remodel in force development.
References
Aizawa, H., M. Sameshima, and I. Yahara. 1997. A green fluorescent protein-actin fusion protein dominantly inhibits cytokinesis, cell spreading, and locomotion in Dictyostelium. Cell Struct. Funct. 22:335-345.
Akiyama, T., Ishida, J., Makagawa S., Ogawara, H., Watanabe S., Itoh, N., Shibuya M., and Fukami Y. 1987. Genistein, a specific inhibitor of tyrosine-specific protein kinase. J. Biol. Chem. 262:5592-5595.
Batistella-Patterson, A.S., S. Wang, and G.L. Wright. 1997. Effect of disruption of the cytoskeleton on smooth muscle contraction. Canad. J. Physiol. Pharmacol. 75:1287-1299.
Bond, M. and Somlyo, A.V. 1982. Dense bodies and actin polarity in vertebrate smooth muscle. J. Cell Biol. 95:403-413.
Byron, K.L. 1996. Vasopressin stimulates Ca2+-spiking activity in A7r5 vascular smooth
muscle cells via activation of phospholipase A2. Circ. Res. 78:813-820.
Choidas, A., A. Jungbluth, A. Sechi, J. Murphy., A. Ullrich, and G. Marriott. 1998. The suitability and application of a GFP-actin fusion protein for long-term imaging of the organization and dynamics of the cytoskeleton in mammalian cells. Eur. J. Cell Biol. 77:81-90.
115 Di Salvo, J., G. Pfitzer, and L.A. Semenchuk. 1994. Protein tyrosine phosphorylation, cellular Ca2+, and Ca2+ sensitivity for contraction of smooth muscle. Can. J. Physiol. Pharmacol. 72:1434-1439.
Di Salvo, J., A. Steusloff, L. Semenchuk, S. Satoh, K. Kolguist, and G. Pfitzer. 1993. Tyrosine kinase inhibitors suppress agonist-induced contraction in smooth muscle. Biochem. and Biophys. Res. Commun. 190:968-974.
Driska, S.P., M.O. Aksoy, and R.A. Murphy. 1981. Myosin light chain phosphorylation associated with contraction in arterial smooth muscle. Am. J. Physiol. 240:C222- 233.
Duarte, J., M.A. Ocete, F. Perez-Vizcaino, A. Zarzuelo, and J. Tamarso. 1997. Effect of tyrosine kinase and tyrosine phosphatase inhibitors on aortic contraction and induction of nitric oxide synthase. Eur. J. Pharmacol. 338:25-33.
Faix, J., C. Clougherty, A. Konzok, U. Mintert, J. Murphy, R. Albrecht, B. Muhlbauer, and J. Kuhlmann. 1998. The IQGAP-related protein DGAP 1 interacts with Rac and is involved in the modification of the F-actin cytoskeleton and control of cell motility. J. Cell Sci. 111:3059-3071.
Firulli, A.B., D. Han, L. Kelly-Roloff, V.E. Katelionsky, S.M. Schwartz, M. Olson, and J.M. Miano. 1998. Comparative molecular analysis of four rat smooth muscle cell lines. In Vitro Cell Dev. Biol.-Animal. 34:217-226.
Fultz, M.E., C. Li, W. Geng, and G.L. Wright. 2000. Remodeling of the actin cytoskeleton in the contracting A7r5 smooth muscle cell. J. Mus. Res. Cell Motil. 21:775-787.
Gabella, G. 1976. The force generated by visceral smooth muscle. J. Physiol. (Lond). 263: 199-213.
Gokita, T., Y. Miyauchi and M.K. Uchida. 1994. Effects of tyrosine kinase inhibitor and phosphotyrosine-phosphatase inhibitor, orthovanadate on Ca2+ free contraction of uterine smooth muscle of the rat. Gen. Pharmac. 25:1673-1677.
Gordon, A.M., A.F. Huxley, and F.J. Julian. 1966. The variation in isometric tension with sarcomere length in vertebrate muscle fibers. J. Physiol. (Lond.). 184: 170- 192.
Haller H., J.I. Smallwood, and H. Rasmussen. 1990. Protein kinase C translocation in intact vascular smooth strips. Biochem. J. 270:375-381.
Kimes, B.W. and B.L. Brandt. 1976. Characterization of two putative smooth muscle cell lines from rat thoracic aorta. Exp. Cell Res. 98:349-366.
116 Mehta, D. and S.J. Gunst. 1999. Actin polymerization stimulated by contractile activation regulates force development in canine tracheal smooth muscle. J. Physiol. (Lond). 519:829-840.
Merkel, I.A., L.M. Rivera, D.J. Colussi, and M.H. Perione. 1991. Protein Kinase C and vascular smooth muscle contractility: Effects of inhibition and down-regulation. J. Pharmacol. Exp. Ther. 257:134-140.
Merkel, L., W.T. Gerthoffer, and T.J. Torphy. 1990. Dissociation between myosin phosphorylation and shortening velocity in canine trachea. Am. J. Physiol. 258:C524-C532.
Murphy, R.A. 1976. Contractile system function in mammalian smooth muscle. Blood Vessels. 13: 1-23.
Murphy, R.A., J.T. Herliky, and J. Meyerman. 1974. Force-generating capacity and contractile protein content of arterial smooth muscle. J. Gen. Physiol. 64: 691- 705.
Nakajima, S., M. Fujmoto, and M. Veda. 1993. Spatial changes of [Ca2+]i and contraction caused by phorbol esters in vascular smooth muscles. Am. J. Physiol. 265:C1138- C1145.
Noegel, A.A. and M. Schleicher. 2000. The actin cytoskeleton of Dictostelium: a story told by mutants. J. Cell Sci. 113:759-766.
Okada, A., R. Lansford, J.M. Weimann, S.E. Fraser, and S.K. McConnell. 1999. Imaging cells in the developing nervous system with retrovirus expressing modified green fluorescent protein. Exp. Neural. 156:394-406.
Paul, R.J. 1983. Functional compartmentalization of oxidative and glycolytic metabolism in vascular smooth muscle. Am. J. Physiol. 244:C399-409.
Rice, R.V., J.A. Moses, G.M. McCanus, A.C. Brady, and L.M. Blask. 1970. The organization of contractile filaments in mammalian smooth muscle. J. Cell Biol. 47: 183-196.
Rokolya, A. and H.A. Singer. 2000. Inhibition of CaM Kinase II activation and force maintenance by KN-93 in arterial smooth muscle. Am. J. Cell. Physiol. 278:C537-C545.
Shimamoto, Y., H. Shimamoto, C.Y. Kwan, and E.E. Daniel. 1993. Differential effects of putative protein kinase C inhibitors on contraction of rat aortic smooth muscle. Am. J. Physiol. 264:H1300-H1306.
Singer, H.A., C.M. Schworer, C. Sweeley, and H. Benscoter. 1992. Activation of protein
117 kinase C isozyme by contractile stimuli in arterial smooth muscle. Arch. Biochem. Biophys. 299:320-329.
Singer, H.A. and K.M. Baker. 1987. Calcium dependence of phorbol 12, 13 dibutyrate- induced force and myosin light chain phosphorylation in arterial smooth muscle. J. Pharmacol. Exp. Therapeutics. 243:814-821.
Solway, J., J. Seltzer, F.F. Samaha, S. Kim, L.E. Alger, Q. Niv, E.E. Morrisey, H.S. Ip, H, and M.S. Parmacek. 1995. Structure and expression of a smooth muscle cell- specific gene SM22 alpha. J. Biol. Chem. 270:13460-13469.
Somlyo, A.V. 1980. Ultrastructure of vascular smooth muscle. In Handbook of Physiology, the Cardiovascular System vol. II, Vascular Smooth Muscle, Bethesda, MD, Am. Physiol. Soc. pp 33-68.
Srivastava, A.K., and J. St-Louis. 1997. Smooth muscle contractility and protein tyrosine phosphorylation. Mol. Cellular Biochem. 176:47-51
Tolloczko, B., F.C. Tao, M.E. Zacour and J.G. Martin. 2000. Tyrosine Kinase-dependent calcium signaling in airway smooth muscle. Am. J. Physiol. 278:L1138-L1145.
VanBuren, P., W.H. Guilford, G. Kennedy, J. Wu, and D.M. Warshaw. 1995. Smooth muscle myosin: a high force-generating motor. Biophys. J. 68:2588-2598.
Vandekerckhoue, J. and K. Weber. 1981. Actin typing on total cellular extracts:a highly sensitive protein-chemical procedure able to distinguish different actins. Eur. J. Biochem. 113:595-603.
Wright, G.L. and A.S. Battistella-Patterson. 1998. Involvement of the cytoskeleton in calcium-dependent stress relaxation of rat aortic smooth muscle. J. Mus. Res. Cell Motil. 19:405-414.
Wright, G. and E. Hurn. 1994. Cytochalasin inhibition of slow tension increase in rat aortic rings. Am. J. Physiol. 267: H1437-H1447.
Yamada, H. and K.N. Bitar. 1995. Phospholipase A2-activating peptide induced contraction of smooth muscle is mediated by protein kinase C-MAP kinase cascade. Biochem. Biophys. Res. Comm. 217:203-210.
Zhang, Y., and R.S. Moreland. 1994. Regulation of Ca2+-dependent ATPase activity in detergent skinned vascular smooth muscle. Am. J. Physiol. 36:H1032-H1039. Chapter 5. General Conclusions
General Discussion
Battistella-Patterson et al. (1997) proposed that the broad length-tension relationship, slow force development, and the low energy cost of maintained tension in smooth muscle could be explained by remodeling of the actin cytoskeleton to modulate optimal actin/myosin overlap and to form a stable system of tension bearing elements to hold the cell in the contracted configuration. They modeled this concept in terms of asynchronous contraction in which it was proposed that different portions of the actin cytoskeleton undergo alternating active contraction, tension maintenance, and remodeling.
However, there was no direct evidence that actin reorganization was linked to the contractile response. Our work has lead to a number of significant conclusions that expand upon this base of knowledge. We have demonstrated that alpha and beta actin undergo active remodeling during the contraction of the A7r5 smooth muscle cell. b-
Actin is shown to undergo shortening without evidence of fiber compression or disassembly, while a-actin undergoes extensive disassembly and subsequent reassembly into intensely fluorescing column-like structures that extend from the base to the apex of the cell. We have further demonstrated that the actin cross-linking protein a-actinin shows strong colocalization with a-actin throughout the interval of contraction and remodeling. The timing and extent of remodeling further suggests that active remodeling may play a role in contraction. We have further demonstrated that there is strong colocalization of a-actin and type II smooth muscle myosin prior to stimulation. Upon
119 stimulation, myosin reorganization follows a-actin remodeling. However, there is a partial dissociation of the two proteins. Agents that affect the structure of a-actin such as cytochalasin and thapisigargin also affect the localization of myosin indicating a close association between the two proteins. These data suggest that the remodeling of a-actin/ myosin may serve to provide a structural arrangement to give maximal mechanical advantage and allow force development in multiple directions. The results further suggest that a-actin is responsible for force development. The last portion of our work begins to elucidate the biochemical pathway involved in the PDBu induced contraction. Inhibition of protein kinase C (PKC) and tyrosine kinase decreased early force development in rat aortic tissue, but had no significant effect on tension maintenance. However, inhibition of calcium/calmodulin kinase II (CAM kinase II) only moderately affected force development, but significantly decreased tension maintenance in aortic rings. At concentrations that inhibit PKC and CAM kinase II, staurosporine significantly decreased both tissue force development and force maintaince. Treatment of A7r5 cells with kinase inhibitors indicate a significant correlation between a-actin remodeling in cells and force development in tissue, with no significant correlation between force maintaince and a- actin remodeling. Taken together our findings suggest that force development may be linked to a-actin remodeling and is controlled separately from b-actin remodeling.
This work has produced a modified model of cytoskeletal remodeling during vascular smooth muscle contraction (Appendix A, Li et al., 2001) from that first proposed by Batistella-Patterson et al. (1997). In the stimulated cell, a-actin/ myosin and b-actin are segregated into separate functional domains (Appendix A, A). Immediately following agonist stimulation, increased intracellular-calcium levels induce a-actin/myosin
120 contraction (Appendix A, B) and is responsible for the fast-phase of contraction. This changes the stress on the b-actin fibers and, in turn, initiates b-actin cable shortening
(Appendix A, C) to preserve the tension gained. a-Actin and myosin may then remodel to allow maximum mechanical advantage of the contractile proteins (Appendix A, E-F).
Eventually, equilibrium between force generation and the opposing load would result in a plateau of active tension generation (Appendix A, G) and would be characterized by a low level of remodeling and lock-up of the proteins. This model predicts a high capacity to generate force by modulation of mechanical advantage of the contractile proteins, slow force development, and low energy cost during tension maintenance.
121 Future Research
The kinase inhibitor studies performed in our laboratory have primarily focused upon the a-actin cytoskeleton. Work is still needed to characterize which pathways are involved in remodeling of the b-actin cyctoskeleton. CAM kinase II is a likely candidate to affect b-actin is based on its effects on tension maintaince. If a- and b-actin are functionally compartmentalized, one would expect that distinct biochemical pathways are involved in the remodeling of each isoform.
Contraction of tracheal smooth muscle has been shown to induce the tyrosine phosphorylation and subsequent activation of focal adhesion kinase (FAK) (Tang et al.,
1999). Phosphorylation levels of FAK was also shown to correlate with tension development during muscarinic stimulation. Antisense to FAK, to deplete protein levels, reduced force development, intracellular calcium levels, and myosin light chain phosphorylation levels in response to muscarinic stimulation (Tang and Gunst, 2001).
Preliminary work in our laboratory indicates that FAK is present in A7r5 cells and colocalizes with "-actin and possibly $-actin before and after phorbol contraction. In collaboration with Dr. Donald Primerano, we have developed a fluorescently labeled antisense oligonucleotide to FAK. This will allow studies on individual cells and aid in the understanding of the role of FAK in A7r5 smooth muscle cells during contraction.
Several actin binding proteins have been identified that can affect the polymerization state of actin filaments. Gelsolin is a Ca2+-regulated actin severing protein that is present in relatively high amounts in smooth muscle. (Hinnsen et al., 1984). After severing the actin filament, gelsolin remains bound to the rapidly polymerizing end and
122 may promote nucleation (Janmay et al., 1985; Hinnsen et al., 1984). Gelsolin has also been shown to stimulate myosin ATPase activity (Dabrowska et al., 1996). Recently,
Grem and Wegner (2000) concluded that the capping and severing functions of gelsolin differ in their calcium affinity. Thus, gelsolin may be a major contributor to actin remodeling during contraction. Other actin binding proteins, especially the actin depolyermerizating factor (ADF) /cofilin family, may be responsible for actin remodeling in motile cells and are deserving of further study (Bamburg, 1999).
In addition to it kinase activity, myosin light chain kinase (MLCK) is reported to bind smooth muscle actin (Dabrowska et al., 1982; Sellers et al., 1984, Ye et al., 1997).
Two actin binding sites may serve to cross-link actin in a Calcium/Calmodulin (Ca2+/CaM) sensitive manner (Hayakawa et al., 1999). Utilizing peptide fragments, the actin binding sites were further characterized, indicating one is Ca2+/CaM-sensitive and one is
Ca2+/CaM-insensitive (reviewed in Gao et al., 2001). Our laboratory is currently collaborating with Dr. Kazuhiro Kahoma to further investigate MLCK’s involvement in actin bundling and its possible role in active remodeling during contraction. Preliminary studies indicate that selectively blocking the Ca2+/CaM-sensitive actin binding site results in disassembly of actin bundles in resting and contracted tissues.
References
Bamburg, J.R. 1999. Proteins of the ADF/cofilin family: essential regulators of actin dynamics. Annu. Rev. Cell Dev. Biol. 15:185-230.
Batistella-Patterson, A.S., S. Wang, and G.L. Wright. 1997. Effect of disruption of the cytoskeleton on smooth muscle contraction. Can. J. Physiol. Pharmacol. 75:1287- 1299.
123 Dabrowska, R., H. Hinssen, B. Galazkiewicz, and E. Nowak. 1996. Modulation of gelsolin-induced actin-filament severing by caldesmon and tropomyosin and the effects of these proteins on the actin activation of Mg(2+)-ATPase activity. Biochem. J. 315:753-759
Dabrowska, R., S. Hinks, M.P. Walsh, and D.J. Hartshorne. 1982. The binding of smooth muscle myosin light chain kinase to actin. Biochem. Biophys. Res. Comm. 107:1524-1531.
Fultz, M.E., C.Li, W. Geng, and G.L. Wright. 2000. Remodeling of the actin cytoskeleton in the contracting A7r5 smooth muscle cell. J. Mus. Res. Cell Motil. 21:775-787.
Gao, Y., L. Ye, H. Kishi, T. Okagaki, K. Samizo, A. Nakamura, and K. Kohama. 2001. Myosin light chain kinase as a multifunctional regulatory protein of smooth muscle contraction. IUBMB Life. 51:337-344.
Gremm, D. and A. Wegner. 2000. Gelsolin as a calcium-regulated actin filament-capping protein. Eur. J. Biochem. 267:4339-4345.
Hayakawa, K., T. Okagiki, L. Li, K. Samizo, S. Higashi-Fujime, T. Takagi, and K. Kohama. 1999. Characterization of the myosin light kinase from smooth muscle as an actin-binding protein that assembles actin filaments in vitro. Biochem. Biophys. Acta. 1450:12-24.
Hinssen, H., J. V. Small, and A. Sobieszek. 1984. A Ca2+-dependent actin modulator from vertebrate smooth muscle. FEBS Lett. 166:90-95.
Janmey, P.A., C. Chaponnier, S.E. Lind, K.S. Zaner, T.P. Stossel, and H.L. Yin. 1985. Interactions of gelsolin and gelsolin-actin complexes with actin. Effects of calcium on actin nucleation, filament severing, and end blocking. Biochemistry. 24:3714- 3723.
Li, C., M.E. Fultz, J. Parash, W.B. Rhoten, and G.L. Wright. 2001. Ca2+-dependent actin remodeling in the contracting A7r5 cell. J. Mus. Res. Cell Motil. 22:521-534.
Sellers, J.R. and M.D. Pato. 1984. The binding of smooth muscle myosin light chain kinase and phosphatases to actin and myosin. J. Biol. Chem. 259:7740-7746.
Tang, D.D. and S.J. Gunst. 2001. Depletion of focal adhesion kinase by antisense oligonucleotides depresses contractile activation of smooth muscle. Am. J. Physiol. 280:C874-C883.
124 Tang, D.D., D. Mehta, and S.J. Gunst. 1999. Mechanosensitive tyrosine phosphorylation of paxillin and focal adhesion kinase in tracheal smooth muscle. Am. J. Physiol. 276:C250-C258.
Ye, L., K. Hayakawa, H. Kishi, M. Immamura, A. Nakamura, T. Okagaki, T. Takagi, and K. Kohama. 1997. The structure and function of the actin-binding domain of myosin light chain kinase of smooth muscle. J. Biol. Chem. 272:32182-32189.
125 Appendix A:
Figure 1. Proposed model explaining force development and maintenance of tension in smooth muscle.
126 Appendix B
127 128