Cell Motility and the Cytoskeleton 23:244251 (1992)

Characterization of Smooth Muscle Caldesmon as a Microtubule-Associated Protein Ryoki Ishikawa, Osamu Kagami, Chihiro Hayashi, and Kazuhiro Kohama Department of Pharmacology, Gunma University School of Medicine, Gunma (R. I., C.H., K.K.), and Department of Molecular Biology, Faculty of Science, Nagoya University, Nagoya (0.K.), Japan We have previously shown that nonmuscle caldesmon copurified with brain microtubules binds to microtubules in vitro [Ishikawa et al.: FEBS Lett. 299:54-56, 19921. To explore the role of caldesmon in the functions of microtubules, further characterization was performed using smooth muscle caldesmon, whose molecular structure and function have been best-characterized in all caldesmon species. Smooth muscle caldesmon bound to microtubules with a stoichiometry of five tubulin dimers to one molecule of caldesmon with the binding constant of 1.1 X lo6 M-'. The binding of caldesmon to microtubules was inhibited in the presence of Ca2+ and calmodulin. Partial digestion of the caldesmon with a-chymotrypsin revealed that the binding site of the caldesmon for microtubules lay in the 34-kDa C-terminal domain. When the caldesmon was in the dimeric form in the absence of a reducing agent, the caldesmon cross-linked microtubules to form bundles. Further, the caldesmon potentiated the polymerization of tubulin, and inhibited the in vitro movement of microtubules on dynein. These results suggest that caldesmon may be involved in the regulation by Ca2+ of the functions of microtubules. 0 1992 Wiley-Liss, Inc. Key words: actin, in vitro motility assay, microtubule bundling

INTRODUCTION The dynamic features of microtubules in eukaryotic cells are consequences not only of the inherent properties of microtubules but also of the properties of the proteins that are associated with microtubules. Various kinds of such proteins, designated MAPs, have been purified and characterized [for review see Olmsted, 19861. Their functions are as follows. 1) They modulate the assembly and stability of microtubules. For example, MAP2 and tau, two of the most abundant heat-stable MAPS from the brain, potentiate the assembly of microtubules and stabilize them [Murphy et al., 1977; Kim et al., 1979; Sandoval and Vandekerckhove, 19811. 2) They are candidates for cross-linkers between microtubules and actin filaments. MAP2 has different binding sites for microtubules and actin filaments, cross-linking microtubules and actin filaments in vitro [Sattilaro et al., 1981; Nishida et al., 1981; Brady et al., 19841. Synapsin I 0 1992 Wiley-Liss, Inc.

[Bahler and Greengard, 1987; Petrucci and Morrow, 19871 and spectrin [Ishikawa et al., 19831 are also candidates for cross-linker proteins because they can interact with both actin filaments and microtubules. 3 ) MAPs can serve as motor proteins in microtubule-dependent transport systems. Intracellular vesicles move along the tracks of microtubules in eukaryotic cells, and their movements are driven by the motor proteins, such as kinesin [Brady, 1985; Vale et al., 19851, cytoplasmic dynein [Paschal et al., 19871, and dynamin [Shpetner and Vallee, 19891. These proteins hydrolyze ATP in a microtubule-depen-

Received April 6, 1992; accepted August 7, 1992 Address reprint requests to Dr. Kazuhiro Kohama, Department of Pharmacology, Gunma University School of Medicine, Maebashi, Gunma 37 I , Japan.

Interaction Between Caldesmon and Microtubules

dent manner to produce mechanical energy [for review see Scholey, 19901. Caldesmon, originally isolated from the smooth muscle of chicken gizzard, is an actin-binding protein whose binding activity is regulated by a Ca*+-calmodulin system [Sobue et al., 19811. It has also been suggested that caldesmon potentiates the stability of actin filaments [Ishikawa et al., 1989a,b], bundles the actin filament in the specified conditions [Bretscher, 1984; Yamashiro-Matsumura and Matsumura, 19881, and controls the interaction between actin and myosin [Sobue et al., 1982; Ngai and Walsh, 1984; Smith et al., 1987; Hemric and Chalovich, 1988; Horiuchi and Chacko, 1988; Ishikawa et al., 1991; Sobue and Sellers, 19911. Caldesmon binds directly not only to actin filaments and calmodulin but also to tropomyosin [Graceffa, 1987; Fujii et al., 1988; Watson et al., 19901 and to myosin [Hemric and Chalovich, 1988; Tkebe and Reardon, 19881. Recently, we demonstrated the binding of brain caldesmon to microtubules [Ishikawa et al., 19921. In this report, we confirm the microtubule-binding activity of caldesmon with well-characterized caldesmon from smooth muscle. Further, we show that caldesmon potentiates the tubulin polymerization, bundles the microtubules, and inhibits the interaction between microtubules and dynein. These results are discussed in terms of the possible roles of caldesmon in the physiological functions of microtubules. MATERIALS AND METHODS Proteins

Caldesmon of 150 kDa on sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) [Sobue et al., 1981; Bretscher, 19841 and 87 kDa from the deduced amino acid sequence of cDNA [Bryan et al., 19891 was purified from chicken gizzard by Bretscher's [ 19841 method with slight modification [Yamashiro-Matsumura et al., 19881, and used as caldesmon. Tubulin was purified from porcine brain by the method of Shelanski et al. [ 19731. Dynein was purified from axonemes of Chlamydomonas reinhardtii as described elsewhere [Kagami et al., 19901. Actin was purified from chicken breast muscle as described elsewhere [Kohama, 19811. Calmodulin from bovine brain was purchased from Sigma (St. Louis, MO).

245

pM). In some experiments, 10 pM calmodulin and 0.1 mM CaCl, or 0.5 mM EGTA were also included. After a 30-min incubation at room temperature, solutions were centrifuged at 140,000g for 15 rnin in an airfuge (Beckman, Palo Alto, CA), and supernatants and pellets were carefully separated. The amounts of proteins in supernatants and pellets were determined by SDS-PAGE and subsequent densitometry (Joyce Loebl Chromoscan 3, Vickers Instrument, Malden, MA). Assay for Polymerization of Tubulin

Tubulin at various concentrations (5-30 pM) was incubated with or without 3.4 p M caldesmon in 15 mM KCI, 1 mM DTT, 1 mM MgCI,, 1 mM GTP, and 100 mM MES (pH 6.8) for 30 rnin at 37"C, and then the reaction mixtures were centrifuged at 140,OOOg for 15 min in an airfuge. The amounts of tubulin in supernatants and pellets were determined by SDS-PAGE and densitometry. Assay for Bundling of Microtubules

Tubulin (final concentration, 10 pM) was polymerized in a reaction mixture that contained 15 mM KCI, 1 mM MgCI,, 1 mM GTP, 20 pg/ml taxol, and 100 mM MES (pH 6.8) for 30 rnin at 37"C, and then 3.4 pM (final concentration) of caldesmon and DTT at various concentrations (0-20 mM) were added. After a 30-min incubation at room temperature, the mixtures were examined directly by phase-contrast microscopy (Axioplan, Zeiss, Oberkochen, Germany) or negatively stained with 1% uranyl acetate and observed under a JEM IOOC electron microscope (JEOL, Tokyo, Japan) at a magnification of x 20,000. Limited Proteolysis

Caldesmon (final concentration, 12 pM) was incubated at 37°C for 20 rnin with 0.2 pg/ml a-chymotrypsin in 100 mM KCl, 0.5 mM DTT, and 20 mM imidazoleHC1 (pH 7.0), and then the reaction was stopped by addition of 5 mM PMSF. The solution was used as partially digested caldesmon. Assay for In Vitro Motility of Microtubules on Dynein

In vitro motility assay was performed by the method of Paschal et al. [ 19871 with slight modification [Kagami et al., 19901. Tubulin was first polymerized in a solution of 1 mM MgCI,, 1 mM GTP, 20 pgiml taxol, Assay for Binding to Microtubules and 100 mM MES (pH 6.8) at 37°C for 30 min. PolyTubulin (final concentration, 10 pM) in a reaction merized microtubules (final concentration, 0.2 pM) mixture composed of 15 mM KCI, 1 mM DTT, I mM were incubated with caldesmon at various concentrations MgCl,, 1 mM GTP, 20 pg/ml taxol, and 100 mM MES (0-2.0 pM) in 5 mM MgCI,, 1 mM EGTA, 1 mM DTT, (pH 6.8) was polymerized for 30 min at 37"C, and then and 20 pgiml taxol for 60 rnin at room temperature. caldesmon was added at various concentrations (0-1 1.5 After addition of ATP to a final concentration of 1 mM

246

Ishikawa et al.

and 0.5 mgiml bovine serum albumin, the mixtures were subjected to the in vitro motility assay. Microtubules were monitored with a darkfield microscope equipped with a 100-W high-pressure mercury lamp, an SIT camera (C2400-08, Hamamatsu Photonics, Shizuoka, Japan), and a video recorder (AG-3800, National, Tokyo, Japan). Assay for Binding to Actin Filaments

Various concentrations (0-5.7 p,M) of caldesmon were incubated with 12 p M actin filaments at room temperature for 30 min in 15 mM KCI, 1 mM DTT, 1 mM MgCI,, 1 mM GTP, and 100 mM MES (pH 6.8). Solutions were centrifuged at 140,OOOg for 30 min in an airfuge. Actin and caldesmon in both supernatants and pellets were quantitated by SDS-PAGE and subsequent densitometry . Other Procedures

5

0

10

Gizzard CaD [ p M )

I

Protein concentrations were determined by the methods of Lowry et al. [1951], with bovine serum albumin as a standard. SDS-PAGE [Ishikawa et al., 1989al was performed as described. The following proteins were used as molecular weight markers for SDSPAGE: myosin heavy chain (200 kDa), P-galactosidase ( 1 16 kDa), albumin (66 kDa), aldolase (42 kDa), carbonic anhydrase (30 kDa), and myoglobin (17 kDa) (Daiichi Pure Chemicals, Tokyo, Japan). RESULTS Smooth Muscle Caldesmon Binds to Microtubules We examined whether or not caldesmon could bind to microtubules. Various concentrations of caldesmon were mixed with taxol-stabilized microtubules and then precipitated with the microtubules by centrifugation. As shown in Figure IA, the amount of caldesmon that bound to microtubules gradually increased as the amount of caldesmon was increased. The binding was saturated at a stoichiometry of tubulin dimer to caldesmon of 1: 0.19, if we assume that the molecular weights of tubulin dimer and caldesmon are 100 kDa [Valenzuela et al., 19801 and 87 kDa [Bryan et al., 19891, respectively. The binding constant (K,) was 1.1 X 10' M-I, the value little higher than that obtained with brain caldesmon (K, = 4.5 x lo5 M-') as reported previously [Ishikawa et al., 19921. The previous report also showed that K, was decreased upon phosphorylation [Ishikawa et a]., 19921. We speculate that the difference between K, of smooth muscle and K, of brain caldesmon may be attributable to the difference in their phosphorylated state. K, of caldesmon for actin filaments varied from lo7 M-' to 10' M-' according to the sources, methods

1

2

3

4

5

Gizzard CaD [ p M ) Fig. I . Caldesmon (CaD) binds to microtubules. A: Taxol-stabilized microtubules (10 pM) were incubated with various concentrations (0-1 1.5 )*M) of caldesmon in 100 mM MES (pH 6.8), I mM GTP, 1 mM DTT, 1 mM MgCl,, and 15 mM KC1 for 30 min at room temperature, and then samples were centrifuged (140,0OOg, 20 min) in an airfuge. The amounts of the proteins in supernatants and pellets were analyzed by SDS-PAGE and quantitated by densitometry. B: The binding of caldesmon to actin filaments under the same buffer conditions as those used for microtubule-binding assay. Actin filaments (12 pM) were incubated with various concentrations (0-5.7 p M ) of caldesmon and precipitated by the centrifugation in an airfuge for 30 min.

of preparation, and assay conditions [Smith et a]., 1987; Yamashiro-Matsumura and Matsumura, 1988; Velaz et al., 19891. Therefore, in order to compare K, of caldesmon for actin and for microtubules, we examined the binding of caldesmon to actin filaments under the same buffer conditions as those used for microtubule-binding assay. As shown in Figure l B , caldesmon bound to actin

Interaction Between Caldesmon and Microtubules

Fig. 2. The binding of caldesmon (CaD) to microtubules is inhibited in the presence of C a l f and calmodulin (CaM). Taxol-stabilized microtubules (10 p M ) were incubated with 3.4 &M caldesmon and various concentrations (0-17.2 pM) of calmodulin in the presence of 0.1 m M CaCI, (0)or 0.5 mM EGTA ( 0 ) . Subsequent experimental procedures were the same as those described in the legend to Figure 1.

filaments at a stoichiometry of actin to caldesmon of 1:0.18 with K, of 5.0 X lo6 M-'.

247

Fig. 3. The microtubule-binding site is confined in the 34-kDa Cterminal domain. Partially digested caldesmon (0.35 p M ) by a-chymotrypsin was incubated with taxol-stabilized microtubules (10 p M ) or actin filaments (12 p M ) for 30 min at room temperature. Subsequent experimental procedures were the same as those described in the legend to Figure 1 . Lanes 1, 2: Supernatant and pellet, respectively, in the presence of microtubules. Lanes 3, 4: Supernatant and pellet, respectively, in the presence of actin filaments.

Binding of Caldesmon to Microtubules Is Regulated by Ca*+-Calmodulin

We examined the binding of caldesmon to microtubules in the presence of Ca2+ and various concentrations of calmodulin (Fig. 2). Without calmodulin, caldesmon bound to microtubules with a stoichiometry of tubulin dimers to caldesmon of 1:O. 19. When we increased the concentration of calmodulin, the amount of caldesmon that bound to microtubules gradually decreased in the presence of Ca2+. Calmodulin at 18 p M reduced the amount of caldesmon bound to microtubules to a molar ratio of tubulin dimers to caldesmon of 1 :0.05. We interpret this reduction to be caused by the decrease of the affinity of caldesmon for microtubules by calmodulin. In the absence of C a 2 + , however, the amount of caldesmon that bound to microtubules did not decrease in the presence of a large amount, i.e., 18 p,M, of calmodulin (filled circles in Fig. 2). These results suggest that the binding of caldesmon to microtubules is regulated by Ca2+ and calmodulin. Microtubule-Binding Site of Caldesmon Localizes in Its 34-kDa C-Terminal Domain

We examined the microtubule-binding site of caldesmon by limited proteolysis with a-chymotrypsin followed by microtubule-binding assay. As shown in Figure 3 (lane 2), intact caldesmon and the 40-kDa frag-

ment on SDS-PAGE could bind to microtubules. The fragment is known to be the C-terminal domain, whose true molecular weight is 34 kDa from its amino acid sequence [Bryan et al., 19891. This 34-kDa C-terminal domain binds to actin filaments (Fig. 3, lane 4) [Szpacenko and Dabrowska, 1986; Fuji et a]., 1987; Wang et al., 1991; Hayashi et al., 1991; for review see also Sobue and Sellers, 19911 and calmodulin [Yazawa et al., 19871. Thus the binding site for microtubules should lie in the 34-kDa C-terminal domain, i.e., in the vicinity of the actin-binding site and calmodulin-binding site. Potentiation of the Polymerization of Tubulin by Caldesmon

MAPS have diverse effects on the assembly of microtubules. For example, MAP2 and tau enhance the polymerization of tubulin [Murphy et al., 1977; Kim et al., 1979; Sandoval and Vandekerckhove, 19811 while synapsin I has no such effect [Baines and Bennett, 19861. We examined the effect of caldesmon on the polymerization of tubulin. As shown in Figure 4, tubulin was polymerized to form polymeric microtubules at various concentrations of tubulin monomers. The critical concentration of tubulin, namely, the minimal concentration of monomers required for formation of polymers,

Ishikawa et al.

248

10

A

E a Y

5

I-

E

/* , / 10

/

Q 1

I

20

30

TUBULIN ( P M I Fig. 4. Potentiation of the polymerization of tubulin by caldesmon. Various concentrations (5-30 p M ) of tubulin were polymerized in the presence or absence of 3.4 p M caldesmon at 37°C for 30 min and then centrifuged at 140,OOOg for 15 min. The amount of tubulin in both supernatants and pellets was determined by SDS-PAGE and subsein the presquent densitometry. ( 0 ) In the absence of caldesmon; (0) ence of caldesmon. Note that caldesmon decreased the critical concentration of tubulin polymerization. MT indicates polymeric tubulin (microtubules).

was 12 pM. This concentration decreased to about 2 pM when caldesmon was present, indicating that caldesmon potentiates the polymerization of tubulin. Caldesmon Bundles Microtubules in the Absence of a Reducing Agent

Caldesmon causes the formation of bundles of actin filaments, an activity that is detectable only in the absence of reducing agents such as DTT. The formation of caldesmon dimers through S-S bonds makes caldesmon divalent and gives it the ability to cross-link actin filaments [Bretscher, 1984; Yamashiro-Matsumura and Matsumura, 19881. Can caldesmon bundle microtubules when it is present as dimers'? Figure 5 shows that large bundles of microtubules were formed when microtubules were mixed with caldesmon in the absence of DTT (Fig. 5B). By contrast, few bundles were observed in the presence of DTT (Fig. 5D). Electron microscopic observations revealed that more than 100 microtubules can be loosely

Fig. 5 . Caldesmon bundles microtubules in the absence of reducing agents. Taxol-stabilized microtubules { I0 pM) were incubated with 3.4 pM caldesmon in the presence or absence of DTT. The solutions were observed by phase-contrast microscopy. A: No additions. B: Plus caldesmon without DTT. C: Plus caldesmon and 2 mM DTT. D: Plus caldesmon and 20 mM DTT. Bar = 100 pm,

bundled together in the absence of DTT (data not shown). These results suggest that the mechanism of the bundling of microtubules by caldesmon may be the same as that of the bundling of actin filaments by caldesmon. It appears that caldesmon monomers have a single site for interaction with microtubules and when caldesmon molecules form dimers through an S-S bond, they can then cross-link microtubules. Inhibition of the Sliding Movement of Microtubules on Dynein by Caldesmon

Caldesmon binds to actin filaments and regulates the in vitro motility between actin and myosin [Okagaki

Interaction Between Caldesmon and Microtubules TABLE I. Effects of Caldesmon (CaD) on the In Vitro Motility of Microtubules on Dynein Sliding velocity (p,m/s + S . D . , n = 30) Control (-CaD) +0.2 (*M CaD f 2 . 0 (*M CaD

Percentage of sliding microtubules (%)"

1.80 t 0.95

90

*

29

0.37 0.67 -b

Ob

"The filaments that slid within 4 s with random pickup were defined as sliding filaments. The number of sliding filaments as a percentage of the total number of filaments examined is presented. bNo filaments attached to the surface of coverslips.

et al., 1991; Ishikawa et al., 19911. Does not caldesmon affect the in vitro motility of microtubules on dyneincoated coverslips? As shown in Table I, microtubules slid at an average velocity of 1.80 pm/s, and 90% of microtubule filaments moved smoothly. In the presence of 0.2 pM caldesmon, the average sliding velocity dropped to 0.37 p d s . Only 29% of microtubule filaments moved, and the rest of the filaments were stuck to the surface of coverslips. When the concentration of caldesmon was further increased to 2.0 pM, the attachment of microtubule filaments to the surface of coverslips was completely inhibited. These results suggest that caldesmon inhibits the interaction between microtubules and dynein. DISCUSSION

In this report, we describe a novel phenomenon, namely, the modulation by caldesmon of the functions of microtubules. It appears that caldesmon has 1 ) the ability to potentiate the polymerization of tubulin (Fig. 4), 2) the microtubule-bundling activity (Fig. 3, and 3) the ability to inhibit the interaction between microtubules and dynein (Table I). If these properties were all the result of nonspecific binding of caldesmon to microtubules, it is unlikely that caldesmon would be subject to physiological regulation, namely, a Ca2+-calmodulin system (Fig. 2) [see also Ishikawa et al., 19921 and/or phosphorylation by cdc2 kinase [Ishikawa et al., 19921. Thus, we are confident that these are inherent properties of native caldesmon. Estimation of the Amount of Caldesmon That Binds to Microtubules In Vivo

The in vivo concentrations of tubulin [Hiller and Weber, 19781, actin [Ishikawa et al., 19911, and caldesmon [Ishikawa et al., 19911 in tissue cultured fibroblast are 20, 27, and 0.1 p M , respectively. If 60% of tubulin [Hiller and Weber, 19781 and 85% of actin [Lin et al., 19841 are polymerized in vivo, the amounts of microtubules and actin filaments in fibroblast are 12 and 23 pM,

249

respectively. From these values and binding constants, we can estimate the amount of caldesmon that binds to microtubules. If we assume that caldesmon does not bind to actin filaments and microtubules at the same time, the following five equations are obtained: CM = K, x C x M CA = K, X C X A CM CA C = c C M + M = m C A + A = a

+

+

where C is the concentration of free caldesmon after equilibrium; M is the concentration of free microtubules after equilibrium; A is the concentration of free actin filaments after equilibrium; CM is the concentration of the caldesmon-microtubule complex after equilibrium; CA is the concentration of the caldesmon-actin complex after equilibrium; c is the initial concentration of caldesmon ( = 0.10 pM); m is the initial concentration of microtubules ( = 12 pM); a is the initial concentration of actin filaments ( = 23 pM); K, is the binding constant of caldesmon for microtubules ( = 1.1 x lo6 M- '); and K, is the binding constant of caldesmon for actin filaments ( = 5.0 X lo6 M-I). From these equations, we calculated that 0.089 p,M caldesmon will bind to actin filaments and 0.010 pM caldesmon to microtubules. The immunofluorescence studies, however, show that caldesmon is not detectable in association with microtubules but with actin filaments [for review see Sobue and Sellers, 19911. Our preliminary studies also fail to detect the colocalization of caldesmon with microtubules, although caldesmon is copurified with microtubules from the brain [Ishikawa et al., 19921. The reason for the disagreement between the immunofluorescence studies and the present in vitro study remains to be solved. Possible Role of Caldesmon in the Regulation of the Polymerization of Tubulin by a Ca2+-Calmodulin System

The polymerization of tubulin is inhibited by a Ca2+-calmodulin system [Marcum et al., 19781. The Caz+ sensitivity of such inhibition is more evident with crude preparations of MAPS [Lee and Wolff, 19821. MAP2 and tau bind to calmodulin in a Ca2+-dependent manner [Lee and Wolff, 1984; Kumagai et a]., 19861. This observation suggests that both proteins are targets of calmodulin. Caldesmon potentiated the polymerization of tubulin (Fig. 4), which may be inhibited by a C a 2 + calmodulin system. Our crude preparation of microtubules contained caldesmon as an MAP [Ishikawa et al., 19921. Thus, caldesmon may also regulate the polymerization of tubulin in a Ca2+-dependent manner via calmodulin.

250

Ishikawa et al.

Possible Role of Caldesmon in the Organization of Microtubules

Fully reduced caldesmon exists in a monomeric form without actin-bundling activity. The oxidized caldesmon that forms dimers through S-S bonds has actin-bundling activity [Bretscher, 1984; Lynch et al., 1987; Yamashiro-Matsumura and Matsumura, 19881. Bretscher [ 19841 speculated that some caldesmon might exist as dimers in vivo because the S-S bond between caldesmons cannot be broken very easily in vitro. Cross et al. [ 19871 showed that some caldesmon forms dimers at physiological salt concentrations. Therefore, the actinbundling activity of caldesmon may be of physiological significance. Caldesmon bundled microtubules in the absence of DTT and did not do so in the presence of DTT (Fig. 5 ) . This result suggests that the transition between monomers and dimers controls the bundling of microtubules. Thus, the transition may control not only the organization of actin but also that of microtubules in vivo. Possible Role of Caldesmon in the Regulation of the Activity of Microtubule-Dependent Motor Proteins

The ATP-dependent sliding of microtubules on kinesin was not observed with a crude preparation of kinesin. A purified preparation of kinesin was required for sliding, suggesting that some inhibitor(s) of kinesin activity is present in crude preparations [Vale et al., 19851. MAP2, which binds along the sides of microtubules, was recently shown to inhibit the sliding of microtubules on kinesin in vitro [Massow et al., 19891. This result indicates that MAP2 is one of the factors that regulate kinesin activity. Similar regulation by MAP2 has been reported in the interaction between dynein and microtubules [Paschal et al., 19891. The observation from our in vitro motility assay that caldesmon inhibits the sliding and the attachment of microtubules to dynein (Table I) suggests that caldesmon may also be a regulator of microtubule-dependent transport systems. It will be of considerable interest to examine the effects of caldesmon on the activities of other motor-protein systems. ACKNOWLEDGMENTS

This work was supported by grants from the Japan Research Foundation for Clinical Pharmacology, the Life Science Foundation of Japan, the Naito Foundation, the Uehara Memorial Foundation, Nervous and Mental Disorders from the Ministry of Health and Welfare of Japan, and by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan. We thank Dr. Ritsu Kamiya at Nagoya University

for his kind introduction to the in vitro motility assay. Taxol is kindly supplied by the Drug Synthesis and Chemistry Branch, Division of Cancer Treatment, National Cancer Institute, Bethesda, MD. REFERENCES Bahler, M . , and Greengard, P. (1987): Synapsin I bundles F-actin in a phosphorylation-dependent manner. Nature 326:704-707. Baines, A.J., and Bennett, V. (1986): Synapsin I is a microtubulebundling protein. Nature 319:145-147. Brady, S.T. (1985): A novel brain ATPase with properties expected for the fast axonal transport motor. Nature 317:73-75. Brady, S . T . , Lasek, R . J . , Allen, R.D., Yin, H . L . , and Stossel, T.P. ( I 984): Gelsolin inhibition of fast axonal transport indicates a requirement for actin microfilaments. Nature 310:56-58. Bretscher, A . (1 984): Smooth muscle caldesmon: Rapid purification and F-actin cross-linking properties. J . Biol. Chem. 259: 1287312880. Bryan, J., Imai, M., Lee, R., Moore, P., Cook, R . G . , and Lin, W.-G. (1989): Cloning and expression of a smooth muscle caldesmon. J . Biol. Chem. 264:13873-13879. Cross, R.A., Cross, K.E., and Small, J.V. (1987): Salt dependent dimerization of caldesmon. FEBS Lett. 219:306-3 10. Fujii, T., Imai, M . , Rosenfeld, G.C., and Bryan, J . (1987): Domain mapping of chicken gizzard caldesmon. J. Biol. Chem. 262: 2757-2763. Fujii, T., Ozawa, J . , Ogoma, Y., and Kondo, Y. (1988): Interaction between chicken gizzard caldesmon and tropomyosin. J. Biochem. (Tokyo) 104:734-737. Graceffa, P. (1987): Evidence for interaction between smooth muscle tropomyosin and caldesmon. FEBS Lett. 2 18: 139-142. Hayashi, K., Fujio, Y . , Kato, I., andSobue, K. (1991): Structural and functional relationships between h- and I-caldesmons. J. Biol. Chem. 266:355-361. Hemric, M.E., and Chalovich, J.M. (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. Hiller, G . , and Weber, K. (19783: Radioimmunoassay for tubulin: A quantitative comparison of the tubulin content of different established tissue culture cells and tissues. Cell 14:795-804. Horiuchi, K., and Chacko, S . (1988): Interaction between caldesmon and tropomyosin in the presence and absence of smooth muscle actin. Biochemistry 273388-8393. Ikebe, M . , and Reardon, S . (1988): Binding of caldesmon to smooth muscle myosin. J. Biol. Chem. 263:3055-3058. Ishikawa, M . , Murofushi, H., and Sakai, H . (1983): Bundling of microtubules in vitro by fodrin. J . Biochem. 94:1209-1217. Ishikawa, R., Yamashiro, S . , and Matsumura, F. (1989a): Differential modulation of actin-severing activity of gelsolin by mu.ltiple isoforms of cultured rat cell tropomyosin: Potentiation of protective ability of tropomyosins by 83-kDa nonmuscle caldesmon. J . Biol. Chem. 264:7490-7497. Ishikawa, R., Yamashiro, S . , and Matsumura, F. (1989b): Annealing of gelsolin-severed actin fragments by tropomyosin in the presence of Ca2+: Potentiation of the annealing process by caldesmon. J. Biol. Chem. 264:16764-16770. Ishikawa, R., Okagaki, T., Higashi-Fujime, S . , and Kohama, K. (1991): Stimulation of the interaction between actin and myosin by Physarum caldesmon-like protein and smooth muscle caldesmon. J . Biol. Chem. 266:21784-21790. Ishikawa, R . , Kagami, O . , Hayashi, C . , and Kohama, K. (1992): The binding of nonmusde caldesmon from brain to microtubules:

Interaction Between Caldesrnon and Microtubules Regulations by Ca’+-calmodulin and cdc2 kinase. FEBS Lett. 299:54-56. Kagami, O . , Takada, S . , and Kamiya, R. (1990): Microtubule translocation caused by three subspecies of inner-arm dynein from Chhmydomonas flagella. FEBS Lett. 264: 179-1 82. Kim, H., Binder, L.I., and Rosenbaum, J.L. (1979): The periodic associations of MAP2 with brain microtubules in vitro. J . Cell Biol. 80:266-276. Kohama, K. (1981): Amino acid incorporation rates into myofibrillar proteins of dystrophic chicken skeletal muscle. J . Biochem. (Tokyo) 90:497-501. Kumagai, H., Nishida, E . , Kotani, S . , and Sakai, H. (1986): On the mechanism of calmodulin-induced inhibition of microtubule assembly in vitro. J. Biochem. (Tokyo) 99:521-525. Lee, Y.C., and Wolff, J. (1982): Two opposing effects of calmodulin on microtubule assembly depend on the presence of microtubule-associated proteins. J. Biol. Chem. 257:6306-63 10. Lee, Y . C . , and Wolff, J. (1984): Calmodulin binds to both microtubule-associated protein 2 and tau proteins. J . Biol. Chem. 259: 1226-1230. Lin, J.J.-C., Matsumura, F . , and Yamashiro-Matsumura, S. (1984): Tropomyosin-enriched and alpha-actinin-enriched microfilaments isolated from chicken embryo fibroblasts by monoclonal antibodies. J . Cell Biol. 98:116-127. Lowry, O.H., Rosebrough, N.J., F a r , A.L., and Randall, R.J. (1951): Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. Lynch, W.P., Riseman, V.M., and Bretscher, A. (1987): Smooth muscle caldesmon is an extended flexible monomeric protein in solution that can readily undergo reversible intra- and intermolecular sulfhydryl cross-linking: Mechanism for caldesmon’s F-actin bundling activity. J . Biol. Chem. 262:7429-7437. Marcum, J.M., Dedman, J.R., Brinkley, B.R., and Means, A.R. ( 1978): Control of microtubule assembly-disassembly by calcium-dependent regulator protein. Proc. Natl. Acad. Sci. U. S .A. 75 :317 1-3775. Massow, A , , Mandelkow, E.-M., and Mandelkow, E. (1989): Interaction between kinesin, microtubules, and microtubule-associated protein 2. Cell Motil. Cytoskeleton 14:562-57 I. Murphy, D.B., Vallee, R.B., and Borisy, G.G. (1977): Identity and polymerization-stimulatory activity of the nontubulin protein associated with microtubules. Biochemistry 16:2598-2605. Ngai, P., and Walsh, P. (1984): Inhibition of smooth muscle actinactivated myosin Mg’ -ATPase activity by caldesmon. J. Biol. Chem. 259: 13656-13659. Nishida, E . , Kuwaki, T., and Sakai, H. (1981): Phosphorylation of microtubule-associated proteins (MAPs) and pH of the medium control interaction between MAPs and actin filaments. J . Biochem. (Tokyo) 90:575-578. Okagaki, T . , Higashi-Fujime, S . , Ishikawa, R . , Takano-Ohmuro, H . , and Kohama, K. (1991): In vitro movement of actin filaments on gizzard smooth muscle myosin: Requirement of phosphorylation of myosin light chain and effects of tropomyosin and caldesmon. 3. Biochem. (Tokyo) 109:858-866. Olmsted, J.B. ( 1986): Microtubule-associated protein. Annu. Rev. Cell Biol. 2:421-457. Paschal, B . M . , Shpetner, H.S., and Vallee, R.B. (1987): MAP 1C is a microtubule-activated ATPase which translocates microtubules in vitro and has dynein-like properties. J. Cell B i d . 105: 1273-1 282. Paschal, B . M . , Obar, R.A., and Vallee, R.B. (1989): Interaction of brain cytoplasmic dynein and MAP with a common sequence at the C terminus of tubulin. Nature 342569-572. Petrucci, T . , and Morrow, J.S. (1987): Synapsin I and actin-bundling +

251

protein under phosphorylation control. J . Cell Biol. 105: 13551363. Sandoval, T.V., and Vandekerckhove, J. (1981): A comparative study of the in vitro polymerization of tubulin in the presence of microtubule-associated protein MAP2 and tau. J . B i d . Chem. 256:8795-8800. Sattilaro, R.F., Dentler, W . L . , and LeCluyse, E.L. (1981): Microtubule-associated proteins (MAPs) and organization of actin filaments in vitro. J. Cell Biol. 90:467-473. Scholey, J.M. (1990): Multiple microtubule motors. Nature 343: I 18120. Shelanski, M.L., Gaskin, I.,and Cantor, C . R . (1973): Microtubule assembly in the absence of added nucleotides. Proc. Natl. Acad. Sci. U.S.A. 70:765-768. Shpetner, H.S., and Vallee, R.B. (1989): Identification of dynamin. a novel mechanochemical enzyme that mediates interaction between microtubule. Cell 59:421-432. Smith, C . W . , Pritchard, K., and Marston, S. (1987): The mechanism of CaZ+ regulation of vascular smooth muscle thin filaments by caldesmon and calmodulin. J. Biol. Chem. 262:116-122. Sobue, K., and Sellers. J.R. (1991): Caldesmon, a novel regulatory protein in smooth muscle and nonmuscle actomyosin system. J . Biol. Chem. 266:12115-12118. Sobue, K . , Muramoto, Y . , Fujita, M., and Kakiuchi, S. (1981): Purification of a calmodulin-binding protein from chicken gizzard that interacts with F-actin. Proc. Natl. Acad. Sci. U.S.A. 78: 5652-5655. Sobue, K., Morimoto, K., Inui, M . , Kanda, K . , and Kakiuchi, S. (1982): Control of actin-myosin interaction of gizzard smooth muscle by calmodulin- and caldesmon-linked flip-flop mechanism. Biomed. Res. 3:188-196. Szpacenko, A , , and Dabrowska, R . (1986): Functional domain ot caldesmon. FEBS Lett. 202:182-186. Vale, R.D., Reese, T.S., and Sheetz, M.P. (1985): Identification of a novel force-generating protein, kinesin, involved in microtubule-based motility. Cell 42:39-50. Valenzuela, P., Quiroga, M . , Zaldiver, J., Rutter, W.J., Kirschner, M.W., and Cleveland, D.W. (1980): Nucleotide and corresponding amino acid sequences encoded by a and p tubulin mRNAs. Nature 289:650-655. Velaz, L., Hemric, M.E., Benson, C.E., and Chalovich, J . M . (1989): The binding of caldesmon to actin and its effect on the ATPase activity of soluble myosin subfragments in the presence and absence of tropomyosin. J . Biol. Chem. 2649602-9610. Wang, C.-L.A., Wang, L.-W.C., Xu, S . , Lu, R.C., Saavedra-Alanis. V., and Bryan, J . (199 1): Localization of the calmodulin- and the actin-binding sites of caldesmon. J. Biol. Chem. 266: 9 166-9 172. Watson, M . H . , Kuhn, A . E . , Novy, R . E . , Lin, J.J.-C., and Mak, A.S. (1990): Caldesmon-binding sites on tropomyosin. J . B i d . Chem. 265: 18860-18866. Yamashiro-Matsumura, S., and Matsumura,.F. ( 1988): Characterization of 83-kilodalton nonmuscle caldesmon from cultured rat cells: Stimulation of actin binding of nonmuscle tropomyosin and periodic localization along microfilaments like tropomyosin. J. Cell Biol. 106:1973-1983. Yamashiro-Matsumura, S.. Ishikawa, R . , and Matsumura, F. (1988): Purification and characterization of 83 kDa nonmuscle caldesmon from cultured rat cells: Changes in its expression upon L6 myogenesis. Protoplasma [Suppl. 2]:9-2I. Yazawa, M., Yagi, K., and Sobue, K. (1987): Isolation and chiracterization of a calmodulin binding fragment of chicken gizzard caldesmon. J. Biochem. (Tokyo) 102:1065-1073.