Calponin and tropomyosin interactions

Biochimica et Biophysica Acta, 1121(1992)41-46

© 1992 Elsevier Science Publishers B.V. All rights remrved 0167-4838/92/$05.00

41

BBAPRO 34187

Calponin and tropomyosin interactions Timothy J. Childs ", Mark H. Watson ~, Robert E. Novy b, Jim J.-C. Lin b and Alan S. Mak ~ a Department of Biochemisto- Queen ~ Univer:,iO'. Kingston, Ontario (Canada) and t, Ek,partment of Biolog% University of hm'a. lo~'a Cio'. IA (USA)

(Received I August 19~1l iRevi.~d manu~ript received 5 December 1991)

Key words: Calponin: Caldesmon; Tropomyosin; Paracrystal: Electron microscopy: Bindingsite The interaction betweer, chicken gizzard calpoain and tropomyosin was examined using vi~osity, light scat lcring, electron microscopy, and affinity chromatography. At neutral pH, l0 mM NaC! and in the abscncc of Mg 2' calponin induced tropomyosin filaments to form paraetystals thus decreasing the viscosity while increasing dramatically the light scattering of the tropomyosin solution. Electron micrographs of the uranyl acetate stained calponin-tropomyosin complex showed the presence of spindle shaped paracrystals with regular ,,,ariation patterns and repeating units of about 400 A. Under similar conditions, smooth mu~le caldesmon al~ induced tropomyo,..,in to form paracrystals. To localize the calponin-binding site on tropomyosin, binding of fragments of trol~myosin, generated by chemical and mutational means, to a calponin-affinity column was studied. The COOH-terminal tropomyosin fragment CnlB(142-281) and the NH,-terminal fragment C S M - ~ ( ! / 8 / 1 2 - 2 2 7 1 bound to a calponin-affinity column with an affinity similar to that of intact tropomyosim while the NH 2-terminal fragment, Cn IA(ll-127), did not bind, indicating that the calponin-binding sitc(s) resides within residues 142-227 of tropomyosin. To dctcrmine the involvement in calponin binding of the area around Cys-190 of tropomyosin, fragmcnts with cleavage sites near or at Cys-19(I were used. Thus, while fragments Cy2(190-284) and CSM-/3(I/8/12-200) bound wcakly to the calponin-affinity column, fragment Cy!(1-1891 did not. These results demonstrate that calponin binds to tropomyosin between residues 142 and 227, and that the integrity of the region around Cys-190 of tropomyosin is important for strong interaction between t h e two proteins.

Introduction

Initiation of contraction in smooth muscle requires the phosphorylation of the 20 kDa myosin light chains by Ca2+/calmodulin-dependent myosin light chain kinose [1]. After tension has developed, however, it is maintained in the muscle at low Ca 2+ concentration by the formation of latch bridges, it has been suggested that the formation of latch bridges represents dephosphorylated myosin to actin cross-bridges which detach slowly. Therefore, myosin phosphorylation may be solely responsible for the regulation of the latch bridges without any need for a thin-filament linked regulatory system [2]. It seems apparent, though, that tension maintenance is regulated, at least in part, by calcium

Abbreviations: D'IT, dithiothreitol; EGTA, [ethylenebis(oxyethylenenitrilo)] tetraacetic acid: SDS-PAGE sodium dodecyl suiratepolyacrylamide gel electrophoresis. Correspondence: A.S. Mak. Queen's University, Kingston, Ontario. Canada, K7L 3N6.

sensitive factors which act on the thin filament. Such factors may include caldesmon/calmodulin [3,4] and the recently discovered protein, calponin [5]. Caiponin has been purified from chicken gizzard [5] and bovine aorta [6]. it is a basic, heat-stable, 34 kDa protein which has been shown to interact with F-actin and tropomyosin, and with calmodulin in a Ca2+-de pendent manner [5]. Calponin has been shown to bind Ca -'+ [7]. It is present in native thin filaments in the molar ratio of 7 . 0 / 0 . 9 / 0 . 6 / 0 . 7 for actin, tropomyosin, caldesmon and calponin, respectively [8]. Evidence supporting calponin's classification as a thin filament prorein include its presence in native thin filaments [8], its localization to the microfilaments of gizzard and aorta smooth muscle cells [9,10], its binding to actin and tropomyosin [11-13], and its decreased susceptibility to calpain degradation when it is bound to thin filaments !14]. In addition, it is related antigenically [6], and in its amino acid sequence [15], to the C-terminal portion of troponin-T. Calponin inhibits the actin-activated MgATPase activity of smooth muscle without affecting the level of myosin phosphorylation and without the need for Ca 2+

42 or tropomyosin [13]. Phosphorylation of calponin with protein kinasc C or Ca-'+/calmodulin-depcndent protein kinase I! releases this MgATPase inhibition [13]. Protein kinase C incorporates I mol of phosphate per mol calponin and is apparently modulated by Ca2+/calmodulin. With Ca :+ present, calmodulin is a noncompetitive inhibitor of protein kinase C phosphorylation, suggesting separate sites for phosphorylation and for calmodulin binding [16]. Phosphorylation is known to inhibit actin binding but does not affect tropomyosin or calmodulin binding [13]. This suggests that calponin exerts its inhibitory effect by mediating actin-myosin interactions. However, it has been reported recently that calponin phosphorylation was not detected in contracting or resting arterial smooth muscle [ 17]. in earlier studies on the interactions between smooth muscle caldesmon and tropomyosin in solution, we observed side-by-side cross-linking of polymerized tropomyosin by caldesmon to form spindle-shaped particles [18]. in this study, we have characterized the interaction between calponin and tropomyosin to determine whether calponin, like caldesmon, can induce tropomyosin to form paracrystals. Calponin binds to tropomyosin and induces tropomyosin polymers to form paracrystals at neutral pH and in the absence of Mg 2+ The calponin-tropomyosin and caldcsmon-tropomyosin paracrystals show the same h;~hly ordered striated patterns under the electron microscope with a repeating unit of 40 nm. This is the first time that it has been shown, as far as we are aware, that tropomyosin can be induced to form paracrystals by another protein in the absence of Mg ~+ and at neutral pH, although it is well known that tropomyosin can form paracrystals at a pH near its p l (4.5) or in the presence of high [Mg2+]. Indeed, by adding calponin to tropomyosin paracrystab previously formed at a pH near the p l of tropomyosin, Takahashi et al. [11] have observed additional bands on the tropomyosin paracrystal structure in a manner similar to that observed in troponin-T decorated tropomyosin paracrystals. Based on this observation, the authors suggested that calponin, like troponin-T, binds near Cys-190 of the tropomyosin molecule. However, troponin-T interacts with tropomyosin at two sites, one at the Cys-190 region of tropomyosin, and the second site at the head-to-tail overlap region of tropomyosin [19]. Therefore, there is some uncertainty in the assignment of the calponin-binding site to Cys-190 rather than to the head-to-tail overlap region of tropomyosin. This fact, aggravated by the low resolution of the electron microscopic analyses, means that the detailed location of the calponin-binding site on tropomyosin cannot be determined solely on the basis of paracrystai results, in order to locate the calponin-binding site(s) more precisely on tropomyosin, we have studied the binding of various chicken breast

and gizzard tropomyosin fragments, generated by chemical and mutational means, to a calponin-affinity column. Our results showed that the calponin-binding site is located within residues 142-227 of tropomyosin and cleavage at :)r near Cys-190 abolished the binding. in contrast to troponin-T, calponin did not bind to the head-to-tail overlap region of tropomyosin. Materials and Methods

Protein preparations Calponin was prepared from fresh chicken gizzard according to Takahashi [5] with several minor modifications. The SP sephadex column was eluted with a step gradient from 50 to 300 mM KC! and gel filtration was carried out in the absence of urea. Caldesmon was prepared from fresh chicken gizzard according to the method of Bretscher [20]. Tropomyosin was prepared as described by Sanders and Smillie [21] with further purification on a DEAE column to remove nucleic acids. Non-polymerizable tropomyosin was prepared by treating tropomyosin with carboxypeptidase A (Sigma) as described previously [22]. Actin was extracted and purified from acetone powder prepared from fresh chicken gizzard by the method of Pardee and Spudich [23]. The M r for calponin, caldesmon, tropomyosin and actin were 34000 [5], 87000 [24], 66000 [25] and 42000 [26] respectively. Concentration of proteins was determined by the method of Lowry [27].

Preparation and purification of tropomyosin fragments Cyanogen bromide cleavage of a-tropomyosin from chicken breast muscle and separation of the fragments, CnlA(II-127) and CnlB(142-281)was performed as reported previously by Watson et al. [28]. Cleavage at Cys-190 of a-tropomyosin by DTNB/KCN and purification of the resulting fragments, Cy1(1-189) and Cy2(190-284), was performed as described by Watson et al. [28]. Preparation of the recombinant mutants of chicken gizzard tropomyosin, CSM-/3(1/8/12-200) and CSM-~(1/8/12-227), was performed as recorded in Watson et al. [28]. The fragments were judged to be at least 90% pure based on SDS-gel electrophoresis patterns and assignments of the fragments were obtained by amino-terminal sequence analyses as described before [28].

Viscosity measurements The viscosity of chicken gizzard tropomyosin solutions with arid without calponin, in 10 mM imidazole (pH 7.0), 10 mM NaCI, 1 mM DTF, 1 mM EGTA and 0.02% NaN 3 was measured using a Cannon-Mamiing semi-micro type AS0 viscometer at 20°C. The time for the flow of buffer alone requires approx. 4 min. To determine the effect of caiponin on tropomyosin viscosity, equimolar quantities of both proteins were mixed

43 (7.6 #M). The protein stock solutions were centrifuged to remove dust and aggregated material. All samples were made up to 0.60 ml, and D'VI" was added from a freshly prepared 1.0 M stock ,solution to reach a final concentration of 10 raM. To further examine the interaction, calponin was titrated into the tropomyosin ~ l u tion up to a molar ratio of 5:1 (calponin per tropomyosin).

Electron microscopy Chicken gizzard tropomyosin or its non-polymerizable counterpart, 7.6 #M, in l0 mM imidazole (pH 7.0), i mM DTr, 1 mM EGTA, 0.02% NaN 3, and 10, 50 or 100 mM NaC! was incubated with 15/.tM calponin or caldesmon for 30 min at room temperature. All solutions were centrifuged at 280000 × g for 10 rain and each pellet was suspended in 50 #1 of the same buffer. Control samples of individual proteins were also examined. In addition, a solution of 5 mg/ml chicken gizzard tropomyosin was prepared in 10 mM Tris (pH 7.5) and subsequently incubated in 50 mM "Iris (pH 8.0), 50 mM MgCI z for 24 h at 4°C to produce Mg 2+ paracrystals. Each protein sample was then ad~rbed to formvar coated grids, stained with 2% aqueous uranyl acetate and examined in a Hitachi H-500 electron microscope with an accelerating voltage of 75 kV. The periodicity of caiponin-tropomyosin and caldesmon-tropomyosin paracrystals was determined by comparing the length of their period (on electron micrographs) to the length of the period of Mg-"+-gizzard tropomyosin paracrystals prepared and photographed at the same time. Mg -'+tropomyosin paracrystals have been well characterized and the generally accepted value for their repeat is 400 ,~ [291.

Light scattering measurements Light scattering measured at a right angle was determined on the MPF-66 Perkin-Elmer fluorescence spec-

trophotometer at 330 nm. The temperature was maintained at 20°C with a circulating water thermostat. The buffer contained 10 mM imidazole (pH 7.0), 10 mM NaCI, 1 mM DTT, 1 mM EGTA and 0.02% NaN 3. A binding isotherm for the interaction between tropomyosin and calponin was obtained by adding known aliquots of calponin to a solution of tropomyosin (0.4 /zM). Upon each addition of calponin to the tropomyosin solution, the light scattering signal was followed for 20 to 40 min to reach a stable level at equilibrium. Corrections were made for dilution during titration.

Calponhz-Affigei 15 affinity chromatography Purified calponin was coupled to 3.0 ml of Affigel 15 (Bio-Rad) according to the procedure recommended by the manufacturer. The buffer used for binding studies was 10 mM imidazole (pH 7.0), 10 mM NaC1, 1.0 mM EGTA, 0.02% NaN 3 and 1.0 mM DTI'. Elution of the tropomyosin fragments was monitored by A_,30,m and the gel electrophoretic pattern of the eluted fractions.

Results

Effect of calponin on the riscosity and light scatterhlg of chicken gizzard tropomyosin sohttions The specific viscosity (7/~) of tropomyosin (7.6 p.M) in 10 mM NaCI and 10 mM DTT was about 4, as shown in Fig. la. Increasing the ionic strength reduced the vi~osity of the tropomyosin ,solution in a manner similar to that reported by others [21,30]. Addition of an equal molar concentration of calponin (7.6 pM) to the tropomyosin .solution did not affect the vi~osity significantly; caiponin alone had a very low vi~osity under the same conditions (-r/~, = 0). To determine the effect of higher ratios of calponin to tropomyosin at a fixed ionic strength (10 mM NaCi), calponin was titrated into a 7.6 #M solution of tropomyosin to a 40

5.0

5.0

A

E 30 o ¢:

¢L 4.0

4.0

ee3

m

(-9

3°I

3.0

0 U

2:

20

2.0

~

2.o

(J

1o ~

(11

~F

1.0 A

0

50

1O0 150 [NoCI] (mM)

200

•

~b/

k0

v

"

i

1.0

2.0

_

i

a

i

3.0

4.0

5.0

0 6.0

CALPONIN,/TROPOMYOSIN(mol/mol)

Fig. 1. Effect of salt on the viscosity and I;' . ~.a,tering (330 nm) of chicken giT~.ard tropomyosin and calponin. Buffer condilions w e r e l{I mM imidazole (pH 7.0), ] mM EGTA. 0.02% NaN~ and I mM D T T and the temperature was 20°C. In Fig. ]A, sail concentration ranged from 1O to 200 mM NaCI. Specific viscosity (~,,p) of tropomyosin alone, 7.6 # M (o); specific viscosity of 7.6 p M tropomyosin + 7.6 p M calponin (e). in Fig. IB, the salt concentration was fixed at 10 mM NaCI and the tropomyosin concentration was kept constant while the concentration of calponin was varied. Si~cific viscosity (rlw) using 7.6 # M tropomyosin (e) and light scattering (330 nm) using 0.4/zM tropomyofin (o).

44 maximum molar ratio of 5: 1 (calponin to tropomyosin), as depicted in Fig. lb. Specific viscosity decreased from an initial value of 3.6 with no added calponin to a value of 1.2 with 5 mol calponin per mol tropomyosin accompanied by an increase in turbidity which was detectable visually. To study the possible aggregation of tropomyosin by ealponin, light scattering of a 0.4 #,M tropomyosin solution in 10 mM NaCI with various amounts of added calponin was recorded at 330 nm with constant stirring, as shown in Fig. lb. A sigmoidal binding curve was obtained indicating the formation of large protein aggregates (because the light scattering signal is proportional to the M~ of the calponin-tropomyosin complex) rather than positive coopcrativity in the interaction between the proteins. The light ~attering signal reached a maximum at about 3 to 4 mol calponin per tool tropomyosin. When the molar ratio of calponin:tropomyosin exceeded 5:1. aggregates became so large that they began to settle even in the presence of moderate stirring, resulting in a decrease in light scattering (not shown).

A

Bw

b

C

Electron microscopy of the caiponbi-tropomyosin complex To examine the ultrastructure of the tropomyosincalponin complex formed in 10 mM NaCI (pH 7.0) and in the absence of Mg -'~, the pelleted sample (280000 x g) was stained with uranyl acetate and examined under the electron microscope. At low magnification (6000 × ), a large number of spindle-shaped particles were observed as shown in Fig. 2a. In 50 and RI0 mM NaCI, similar but fewer particles could be found (not shown). Upon closer examination (80000 x , Fig. 2b), a periodic arrangement of light and dark striations was observed, indicative of formation of tropomyosin paracrystals. These paracrystals exhibited a periodicity of about 400 ,~, approx, the length of one tropomyosin molecule, and contained a 2-fold axis of symmetry (dyad symmetry). Under similar conditions, chicken gizzard caldesmon also induced tropomyosin to form paracrystals as shown in Fig. 2c, with similar staining patterns as that found in the calponin-tropomyosin complex. When the pelleted paracrystals were run on SDS-PAGE an apparent stoiehiometry of 1-2 mol of ealponin per tool of tropomyosin, and 0.5-1 mol of caldesmon per mol of tropomyosin was observed based on laser densitometry of Coomassie blue-stained gels. These paracrystals likely represent side-by-side aggregates of tropomyosin polymers cross-linked by calponin or caldesmon. Tropomyosin has been shown to form ordered paracrystal structures in 50 mM Mg 2+ at neutral pH (as shown in Fig. 2d and in Ref. 30) or at a pH close to its isoelectric point of 4.5 [11]. However, both calponin-tropomyosin (Fig. 2b) and caidesmontropomyosin (Fig. 2c) co-crystals have different stria-

•

.

.

Fig, 2. Striation pattern and periodicity of chicken gizzard tropomyosin paracrystals. All ~mples were stained with 2% uranyl acetate. Fig. 2A represents a low magnification ((-dXM)×)electron mtcrograph of tropomyosin-calponin paracwstals. The conditions for production of these paracrystals were i0 mM imidazole (pH 7.0), l0 mM NaCI, 1 mM EGTA, 0.02% NaN.~, I mM DTI" and incubation [or ,h'~ min at 20°C. All subsequent electron micrographs are at 80000x magnification. Fig. 2B and C represent tropomyosincalponin and tropomyosin-caldesmon (tool/tool) paracrystals, respectively, produced using the same conditions listed for Fig 2A. Fig. 2D represents tropomyosin paracrystals produced after a 24 h incubation in 50 mM Tris {pH 8.0) and 50 mM MgCI2 at 4°C. Bars represent 100 nm.

tion patterns from the tropomyosin paracrystal formed in Mg 2+ (Fig. 2d).

Binding of tropomyoshl fi'agments to hnmobilized calponin In an attempt to locate the calponin-binding region on tropomyosin, intact chicken gizzard tropomyosin or

45 TABLE I

Summary of the bimting of m~pomyosin and tropomyosin fragments to immobilized caiponbi Tropomyosin or fragment

[NaCi] (raM) at which the protein peak is elutcd

Chicken gizzard or breast tropomyosin

(I-284) CnlB (142-281) CSM-p(!/8/12-227) Non-pol~nnerizable chicken gizzard tropomyosin (1-273) CY2 (190-284) CSM-p(I/8/12-20~, C n l A (! 1-127) CY! (I-189) Bovine .~rum albumin Control ~

il5 II)I I I t~

96

85 78 55 53 54 54

a Elution of intact chicken gir.,,~rd tropomyosin from Affigel which was previously blocked by ethanolamine.

tropomyosin fragments derived from either chicken breast a-tropomyosin or chicken gizzard /]-tropomyosin were applied to a calponin-Affigel 15 column .and eluted with a linear NaCI gradient (10-500 mM). Chicken gizzard and chicken breast tropomyosin bind with similar affinity to calponin (data not shown), and thus using fragments from chicken breast tropomyosin was deemed justifiable. Intact gizzard tropomyosin, non-polymerizable tropomyosin NPTM(l-273), CSM/](1/8/12-227) and the COOH-terminal fragment CnlB(142-281) bound to the immobilized calponin with similar affinity and were eluted at about 100 mM NaCl, as shown in Table i. The smaller COOH-terminal fragment, Cy2(190-284), and the smaller NH 2terminal fragment CSM-/](I/8/12-200) bound with an intermediate strength and were eluted at 85 mM NaCI. Both amino-terminal fragments, Cy1(1-189) and CnlA(ll-127), bound weakly to the immobilized calponin and were eluted at approx. 50 mM NaCI. Bovine serum albumin, presumably interacting nonspecifically with the calponin-column, was eluted at 54 mM Nacl; intact tropomyosin was eluted at 54 mM NaCi from an Affigel column which had been blocked by ethanolamine (i.e. no calponin present). Discussion

Results from viscosity, light scattering and electron microscopy studies indicate that calponin reduces the viscosity of a tropomyosin solution at low salt, not by depolymerizing the tropomyosin polymers, but rather by inducing the tropomyosin filaments to form sideby-side aggregates by a mechansim which is not clear at present. Electron micrographs of the uranyl acetatestained ealponin-tropomyosin aggregates showed clearly the presence of spindle-shaped particles. At

high magnifications, these particles showed aoregular pattern of striations with a periodicity of 4(K) A, about the length of a tropomyosin molecule, indicating that calponin, in the absence of Mg 2+ and at neutral pH, induced tropomyosin to form paracrystals. Under the same conditions, chicken gizzard caldcsmon also induced the formation of tropomyosin paracrystals. Although it is known that tropomyosin forms paracrystals either in 50 mM Mg 2÷ at neutral pH or at a pH close to its isoelectric point (pH 4.5), in the absence of calponin in both cases, this is the first time, as far as we are aware, that tropomyosin has been cross-linked by another protein to form paracrystals in the absence of Mg 2+ at neutral pH. The striation pattern of the calponin-tropomyosin co-crystal we have observed is different from the polar form of paracrystal produced by Takahashi et al. [I l] in Na acetate, pH 5.5. Based on their crystal structure, Takahashi et al. [l I] suggest that calponin binds tropomyosin at a site about 17 nm from the COOH-end, placing the calponin around Cys-190 of tropomyosin. This assumption is based on the similarity between the troponin-T, and the calponin decorated paracrystals they produced. Since it is known that troponin-T binds to both the Cys-190 and the head-to-tail overlap region of tropomyosin [19], similar troponin-T-tropomyosin and calponin-tropomyosin paracrystal structures may also suggest that calponin can bind to Cys-190 and/or the overlap region of tropomyosin. To delineate the calponin-binding site(s) more precisely on tropomyosin, binding of tropomyosin fragments from chicken breast and gizzard muscle to a calponin-affinity column was performed. We found that intact tropomyosin and its fragments, NPTM(1-273), CSM-g( 1/8/12-227) and Cn I B(142-281 ), had similar affinity for the immobilized calponin, indicating that the calponin-binding site is located within residues 142-227 of tropomyosin, but does not involve the head-to-tail overlap region. Cleavage at Cys-190 of tropomyosin reduced significantly the binding of the resulting fragments, C y l ( l - 189) and Cy2(190-284), to calponin. Similar to our previous results for the caldesmon binding site on tropomyosin [28], this indicates that the calponin-binding region around Cys-190 on tropomyosin must be kept intact for strong binding. it has been suggested that calponin and caldesmon may represent inhibitory components similar to the troponin-T/! complex in striated muscle [13,31]. Thc tryptic fragment T2 of troponin-T and intact troponin-I have been shown to bind to a region around Cys-190 of tropomyosin, which also constitutes an essential part of the binding-sites for calponin and caldesmon, it appears that this region of tropomyosin, highly conserved in amino acid sequence [32], may represent the location of a thin-filament regulated switch for both striated and smooth muscle contraction. The T1 fragment, residues 1-158, of troponin-T interacts with the head-

46 to-tail overlap region of tropomyosin [19]; however, caldesmon and calponin do not interact with this region of tropomyosin, which is highly variable in amino acid sequence between striated and smooth music tropomyosin [32]. It appears, therefore, that the headto-tail overlap region of tropomyosin may possess different functions in striated and smooth muscle. Winder and Waish [13] have reported that calponin is capable of inhibiting the actomyosin ATPase to similar degrees in the presence: or absence of tropomyosin and that phosphorylated calponin, while binding to immobilized tropomyosin, does not bind to tropomyosin-actin filaments. This raises concerns about the relevance of the calponin-tropomyosin interaction in the regulation of smooth muscle contraction. Smooth muscle tropomyosin is known to potentiate the actinactivated myosin ATPase activity [33,34]. Recently, Horiuchi and Chacko [35] obsen'ed that tropomyosin potentiated actin-activated ATPase activity by increasing the Vm,,, 2-3-fold; in the presence of calponin, however, the tropomyosin potentiation was eliminated. The mechanism by which calponin attenuates the tropomyosin potentiation of actin-activated ATPase activity is unclear. Calponin may achieve its effect by interacting with tropomyosin and/or actin. Although it has been shown that binding of calponin to actin appeared to be unaffected by the presence of tropomyosin [35], it is not clear whether actin may in fact weaken the binding of calponin to tropomyosin. Therefore, the nature of the 3-way interaction among calponin-tropomyosinactin remains to be explored. It is interesting to note that Vancompernolle et al. [12.] have suggested that the F-actin and tropomyosin binding sites on calponin may be close enough to interact cooperatively. Taking all the evidence together, the regulation of the smooth muscle thin-filament must involve an intricate interplay amongst actin, tropomyosin, caiponin and/or caldesmon. Towards a better understanding of these protein-protein interactions, it is important to locate the sites of interaction. Acknowledgements We would like to thank Dr. M.E. Nesheim for the use of the Perkin-Elmer MPF-66 fluorescence spectrophotometer. This study was supported [-,~ grants from the Medical Research Council of C; ~,da and the Ontario Heart and Stroke Foundation (to ASM) and by National Institutes of Health Grants HD 18577 and GM 40580 (to JJ-CL). T.J.C. and M.H.W. are recipients of Studentships from the Medical Research Council of Canada. References I Adelstein, R.S. and Sellers, J.R. 11987) Am. J. Cardioi. 59, 4B- I0B.

2 ltai. C.-M. and Murphy. R.A. (1989) Annu. Rev. Physiol. 51. 285-298. 3 Marston, S., Lehman, W,. Mondy, C., Pritchard, K. and Smith, C. (1988) in Calcium and Calcium Binding Proteins (Gerday, Ch., Gilles. R. and Bolls, L,, eds.), pp. 69-81, Springcr-Verlag, Berlin. 4 Walsh, M,P. (1987) ProB. Clin. Biol. Res, 245. 119-141. 5 Takahashi, K., Lliwada. K. and Kokubu, T. (1986) Biochem. Biophys. Res. Commun. 141, 20-26, 6 Takahashi. K., Hiwada, K. and Kokubuo T. (1988) Hypertension I I, 620-626. 7 Takahashi. K., Hiwada, K. and Kokubu. T. (1987) Hypertension 10, .36(I (abstr.). 8 Nishida, W., Abe, M., Takahashi, K. and ltJwada, IC, (1990) FEBS Letl. 268, | 65-168. 9 Gimona, M., Herzog. M, Vandekerckhove, J. and Small, V. (1990} FEBS Lett. 274, 159-162. 10 Takeuchi, K., Takahashi, K., Abe. M., Nishida, W., Hiwada, K., Nabeya. T. and Maruyama, K. (1991)J. Biochem. l[lg, 311-316. I I Takahashi. K., Abe, M.. Hiwada. K. and Kokubu, T. 11988) J. Hypertension 6. $40-$43. 12 Vancompernolle, K., Gimona, M., Herzog, M., Van Damme, J,0 Vandekerckhove, J. and Small, V. (1990) FEBS Lett. 274, 146150. 13 Winder, S.J. and Walsh, M.P. (19~0) J. Biol. Chem. 265, 1014810155. 14 Tsunckawa, S., Takahashi, K., Abe, M., Hiwada, K., Ozawa, K. and Murachi, T. 11989) FEBS Lett. 250, 493-496. 15 Takahashi0 K. and NadaI-Ginard, B, (1991) J. Biol. Chem. 266, 13284-13288. 16 Naka, M.. Kureishi, Y., Muroga, Y., Takahashi, K., lto, M. and Tanaka. 1. (19~0) Biochem. Biophys. Res. Commun. 171, 933937. 17 BfirAny, M., Rokolya. A. and 13dr~ny, K. (1991) FEBS Lett. 279, 65 -68 18 Watson, M.H.. Kuhn, A.E. and Mak, A.S. (1990) Biochim. Biophys. Acta 1054, 1[)3-113. 19 Sanders, C. and Smillie, L.B. (1985) J. Biol. Chem. 260, 72647275. 20 Brelcher, A. (1984)J. Biol. Chem. 259, 12873-1288{). 21 Sanders, C. and Smillie. L.B. (1984) Can, J. Biochem. Cell Biol. 62, 443-448. 22 Mak, A.S. and Smillic, L.B. 11981) Biochem. Biophys. Res. Commun. 101,208-214. 23 Pardee, J.D. and Spudich, J.A. 11982) Melhods Enzymol. 85, 164-181. 24 Bryan, J., Imai, M, Lee, R., Moore, P., Cook, R.G. and Lin, W-G. (1989) J. Biol. Chem. 264, 13873-13879. 25 Mak, A,S.. Smillie. L.B. and Stewart, G. 11981)) J, Biol. Chem. 255, 3647-3655. 26 Clarke, M, and Spudich, J.A. 11977) Annu. Rev. Biochem. 46, 797-822. 27 Lowry, O.H., Rosebrough, N.J,, Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 28 Walson, M.H.. Kuhn, A.E., Novy, R.E., Lin, I.J.-C and Mak, A.S. (199[l) J. Biol. Chem. 265, 18860-18866. 29 Cohen, C. and Longley, W., 11966) Science 152, 794-796. 3() Dabrowska, R., Nowak, E. and Drabikowski, W. 11980) Comp. Biochem. Physiol. 65h, 75-83. 31 Abe, M., Takahasi, K. and Hiwada, K. (1990)J. Biochern. 108, 835-838. 32 Mak, A.S. and Smillie, L.B. 11981) J. Mol. Biol. 149. 541-550. 33 Chacko. S, 11981) Biochemistry 20, 702-707. 34 Chacko, S. and Eisenberg. E. (1990) J. Biol. Chem. 265, 2105211D. 35 Horiuchi, K.Y. and Chacko, S. 11991) Biochem. Biophys. Res. Commun, 176, 1487-1493.