Biochimica et Biophysica Acta, 1054 (1990) 103-113
103
Elsevier BBAMCR 12748
Caldesmon, calmodulin and tropomyosin interactions Mark H. Watson, Alison E. Kuhn and Alan S. Mak Department of Biochemistry, Queen's University, Kingston (Canada)
(Received 24 January 1990) (Revised manuscript received10 April 1990)
Key words: Caldesmon;Calmodulin; Tropomyosin;Electron microscopy;Fluorescence Binary complex interactions between caldesmon and tropomyosin, and ealmodulin and tropomyosin, and ternary complex interaction involving the three proteins were studied using viscosity, electron microscopy, fluorescence and affinity chromatography techniques. In 10 m m NaCI, caldesmon decreased the viscosity of chicken gizzard tropomyosin by 7 - 8 fold with a concomitant increase in turbidity (A 33Onto)" Electron micrographs showed spindle-shaped particles in the tropomyosin-caldesmon samples. These results suggest side-by-side aggregation of tropomyosin polymers induced by caldesmon. Binding studies in 10 m m NaCI between caldesmon and chicken gizzard tropomyosin labelled with the fluorescent probe N-(1-anilinonaphthyl-4)maleimide (ANM) gave association constants from 5.3 • 106 to 7.9- 106 m -1 and stoichiometry from 1.0 to 1.4 tropomyosin per caldesmon. Similar binding was observed for rabbit cardiac tropomyosin and caldesmon. Removal of 18 and 11 residues from the C O O H ends of the gizzard and cardiac tropomyosin by carboxypeptidase A, respectively, had no significant effect on their binding to caldesmon. In the presence of Ca 2+, chicken gizzard tropomyosin bound to a calmodulin-Sepharose-4B column and was eluted with a salt concentration of 140 raM. This interaction was weakened in the absence of Ca 2+, and the bound tropomyosin was eluted by 65 m M KCI. ANm-labelled tropomyosin bound calmodulin in the presence of Ca 2 + with a binding constant of 3.5 - 106 M - ! and a binding stoichiometry of 1 to 1.4 tropomyosin per calmodulin. In 10 m m NaCI, calmodulin reduced the specific viscosity of chicken gizzard tropomyosin in the presence of Ca 2+ by 5 fold, while a 1.5-fold reduction in viscosity was observed in the absence of Ca 2+. In either case, no significant increase in turbidity was observed suggesting that calmodulin reduced head-to-tail polymerization of tropomyosin. The interaction of caldesmon with the calmodulin-ANM-tropomyosin complex in the presence and absence of Ca 2+ was also examined. The result is consistent with a model that in the absence of Ca 2+, calmodulin binds weakly to either caldesmon or tropomyosin and has little effect on the tropomyosin-caldesmon interaction; whereas, Ca2+-calmodulin interacts with ealdesmon and reduces its affinity to tropomyosin.
Introduction
Contraction in smooth muscle is initiated by the Ca 2+-dependent phosphorylation of the 20 kDa myosin light chains by the myosin light chain kinase (for review see Ref. 1). This activation process can be reversed by dephosphorylation of the light chain by smooth muscle phosphatases [2]. Although myosin light chain phosphorylation precedes contraction in intact smooth muscle, tension is maintained at a Ca 2+ concentration too low to support light chain phosphorylation [3]. Tension maintenance is a result of formation of slow cycling
Abbreviation: ANM, N-(1-anilinonaphthyl-4)maleimide. Correspondence: A.S. Mak, Department of Biochemistry, Queen's University, Kingston, Ontario, Canada K7L 3N6.
cross-bridges between myosin and actin known as 'latch'. Regulation of this latch mechanism is believed to involve other Ca2+-sensitive factor(s) in the smooth muscle. Intact thin filaments have been shown to activate phosphorylated smooth muscle myosin ATPase in a CaE+-dependent manner raising the possibility that such factors may act on the thin-filament [4,5]. The likely candidate for a thin-filament-linked regulatory protein in smooth muscle is caldesmon, first purified from chicken gizzard by Sobue et al. [6]. The M r of caldesmon is 150000 as determined by SDS-gel electrophoresis [6], and a value of 93 000 was recently reported based on analytical ultracentrifugation data [7]. Recently, a smooth muscle caldesmon has been cloned and expressed; based on the deduced amino acid sequence, a M r of 86 974 was obtained [31]. Caldesmon has been shown to bind to calmodulin [24,51,59], actin [6,51,57], myosin [15] and tropomyosin [25-27,51]; it
0167-4889/90/$03.50 © 1990 Elsevier SciencePublishers B.V. (Biomedical Division)
104 has been found in other smooth muscle and non-muscle cells [8,9,10]. The stoichiometry of actin : tropomyosin : caldesmon is 2 8 : 4 : 1 in native thin filaments [9], Caldesmon inhibits superprecipitation of chicken gizzard actomyosin [11] and the actin-activated Mg-ATPase activity of phosphorylated smooth muscle myosin [12] at low Ca 2+ concentration and the inhibition can be overcome by Ca2+-calmodulin. Caldesmon enhances the binding of smooth muscle H M M [13], S1 and skeletal H M M [14] to actin in the presence of ATP, suggesting that smooth muscle myosin binds to actin at a non-productive site in the presence of caldesmon [14] resulting in the inhibition of, perhaps, the ADP-releasing step in the kinetic cycle [13]. This mechanism may lead to a decrease in the cycling rate of the myosin ATPase activity characteristic of the latch mechanism. Recent findings that caldesmon can cross link actin and myosin by binding directly to the myosin head may also contribute to the enhancement in the binding of myosin to actin by caldesmon [15]. Smooth muscle tropomyosin potentiates actomyosin ATPase 3 - 4 fold even at low S1 to actin ratio. This has been interpreted in terms of the allosteric model of Hill et al. [16], that more of the actin-tropomyosin complex is in the 'turned on' state for the smooth muscle tropomyosin than the skeletal muscle counterpart [17,18]. In the presence of caldesmon, however, the role of smooth muscle tropomyosin is controversial. It has been reported that caldesmon inhibits the tropomyosin potentiation of actomyosin ATPase [19]. While some workers have found that tropomyosin is essential for the caldesmon inhibition [11,9], others have not observed an absolute requirement of tropomyosin for caldesmon inhibition [20-22]. Recently, Chalovich [23] found that at high ionic strength, 50 and 90% of caldesmon inhibition of the Mg 2÷ actomyosin ATPase activity was observed in the absence and presence of tropomyosin, respectively. Direct interaction between tropomyosin and caldesmon has been demonstrated. Tropomyosin binds to a caldesmon-affinity column [24]; and the viscosity of a tropomyosin solution is increased 4.7-fold by caldesmon in the absence of salt and dithiothreitol [25]. Fluorescence analyses of the binding between caldesmon and tropomyosin in the presence and absence of actin was reported using pyrene-labelled caldesmon [26], and interaction between caldesmon and dansyl chloridelabelled tropomyosin has been reported by Fujii et al. [27]. The tropomyosin-binding site has been located on the COOH-terminal 38 kDa fragment of caldesmon, which also contains the actin- and calmodulin-binding sites [28,21,27,50]. D N A sequence analysis [29,31] has shown that this 38 kDa fragment contains a 58 residue domain which has a 43% overall sequence identity with residues Glu-89 to Lys-147 of T n T within a cyanogen
bromide fragment, TnT-CB2, which has been shown to bind close to or at the COOH-terminal end of tropomyosin [30,32]. In this study, we have investigated (1) the binary interactions between caldesmon and tropomyosin, and between tropomyosin and calmodulin, and (2) ternary interaction involving tropomyosin, caldesmon and calmodulin in the presence and absence of Ca 2+, using fluorescence spectroscopy, viscosity measurements, electron microscopy and affinity chromatography. Materials and Methods
Protein preparations Caldesmon from fresh chicken gizzard was prepared according to the method of Bretscher [10]. Tropomyosin from frozen chicken gizzard and rabbit cardiac muscle was prepared as described by Sanders and Smillie [36]. The gizzard tropomyosin was further purified on a DEAE-column to remove nucleic acid contaminations. Non-polymerizable tropomyosin was prepared by treating tropomyosin with carboxypeptidase A (Sigma) as described before [37]. Concentration of tropomyosin was determined by the method of Bradford [38] using rabbit cardiac tropomyosin as standards. Caldesmon and calmodulin concentrations were calculated from 1% __ 1% A 2 8 0 n m - 3.3 [7] and A278,m = 1.9 [40], respectively. The M r for caldesmon, tropomyosin and calmodulin were 87000 [31], 66000 [39] and 18000 [40], respectively. Calmodulin was prepared from frozen bovine brain as in [41]. Fluorescence studies Tropomyosin from either rabbit cardiac and chicken gizzard muscle was reduced with dithiothreitol and the excess dithiothreitol was removed by isoelectric precipitation of tropomyosin at p H 4.5. The reduced tropomyosin was labelled with A N M (Polysciences) as described by Morris and Lehrer [42]. The stoichiometry of labelling was determined by using the extinction coefficient of 10 800 cm -1- M-1 at 345 nm for A N M [43]. For rabbit cardiac tropomyosin, which has 2 cysteine residues per molecule, a label of 1.8 A N M per mol tropomyosin was obtained. A labelling ratio of 2-2.5 was found for the gizzard tropomyosin which contains equal amounts of the B- and "y-isoforms, each containing 1 cysteine residue per 33 kDa chain. Fluorescence measurements were performed on a MPF-66 Perkin-Elmer fluorescence spectrophotometer. The temperature was maintained at 20 ° C by a circulating water thermostat. Binding isotherms for the interaction between caldesmon and tropomyosin or calmodulin ( + C a 2÷) and tropomyosin were obtained by adding known aliquots of caldesmon or calmodulin ( + Ca 2+) to a solution of ANM-tropomyosin (0.05 m g / m l ) in 10 mM imidazole pH 7.0, 10 mM dithiothreitol, 1.0 mM
105 EGTA, 0.01% N a N 3 and 10, 50 or 100 mM NaCI. Corrections were made for dilution during titration. Light scattering due to an identical sample of unlabelled tropomyosin and caldesmon or calmodulin was subtracted from the ANM-tropomyosin fluorescence. Binding isotherms for the ternary complex of ANMtropomyosin, calmodulin and caldesmon were made by mixing ANM-tropomyosin, in the presence or absence of Ca 2÷, with the required amount of calmodulin. This was titrated with known aliquots of caldesmon. Corrections for dilution and light scattering were made with unlabelled proteins as described above. Excitation was at 356 nm and emission was measured at 456 nm. For experiments done in the presence of Ca 2÷, CaC12 was added from a 25 mM stock such that the final solution was 1.1 mM Ca 2÷ and 1.0 mM EGTA. To study the salt effect on the binding of caldesmon or calmodulin to A N M - t r o p o m y o s i n , a solution of caldesmon or calmodulin and ANM-tropomyosin was titrated with NaCI from a stock solution of 5 M NaC1. The protein concentrations are indicated in the figure legends. The association constant ( K ) for a single binding site, the number of caldesmon-binding sites per molecule of tropomyosin (n), and the maximum change in fluorescence at saturation (AFmax) were analyzed similarly to that described by Morris and Lehrer [42]. In the binding between tropomyosin and caldesmon to form a complex, TC,, it can be shown that: K n T b 2 - b ( K T n + K C + 1) + K C = 0
where T = total concentration of tropomyosin, C = total concentration of caldesmon, and b -- A F / A Fm~, (fraction of total sites in tropomyosin with bound-caldesmon) where A F = the change in tropomyosin fluorescence intensity at each addition of caldesmon. This equation assumes non-cooperative binding of caldesmon to tropomyosin if n is not equal to 1. Binding parameters were determined by fitting the values of T, C and b to the above equation by computer using a non-linear least square program.
freshly prepared 1.0 M dithiothreitol stock to make 20 mM. The samples were sealed and left at 20 ° C for 1 h before the viscosity was measured.
Electron microscopy Chicken gizzard tropomyosin, 15 #M, in 10 mM imidazole, 10 mM NaCI, 1.0 mM EGTA, 0.01% NaN 3, 10 m M dithiothreitol (pH 7.0), was incubated with 45 /tM of caldesmon for 30 rain at room temperature. The solution was centrifuged at 280000 × g for 10 min and the pellet was supended in 50/~1 of the same buffer. The protein samples were adsorbed 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. Control samples of tropomyosin and caldesmon alone were examined similarly.
Calmodufin-Sepharose 4B affinity chromatography Purified calmodulin (8 mg) was coupled to CNBractivated Sepharose 4B (Pharmacia) (2 g dry weight) following the manufacturers recommended procedure. The buffer and chromatography conditions are described in the figure legends. Results
Effect of caldesmon on the viscosity of chicken gizzard tropomyosin At a protein concentration of 7.6/xM (0.5 m g / m l ) in 10 m M NaC1 and 20 mM dithiothreitol, the specific viscosity 0Jsp) of gizzard tropomyosin was 3.5, as shown in Fig. 1. Increase in ionic strength reduced the viscosity of the tropomyosin solution in a manner similar to that reported by others [36,44]. Caldesmon alone had very low viscosity under the same conditions, 7/sp = 0. 4.0-
-0.15
3.o ~)
-0.I0
2.0
i
300 m M NaC1, no change in fluorescence was observed 3.0-
2.0
o
1.0 m o3
0.O
O.0
I
I
I
1.0
2.0
3.0
CALMODULIN/TROPOMYOSIN (mol/mol)
Fig. 6. Effect of calmodulin on the viscosity of chicken gizzard tropomyosin at 10 m M NaCl in the presence and absence of Ca 2+. Buffer conditions were as described in the Fig. 1, except that [NaCl] was kept constant at 10 m M . Temperature was 2 0 ° C . Viscosity measurements were repeated three times using different preparations of tropomyosin and calmodulin. Each point in the titration was an average of two measurements. The tropomyosin concentration was kept constant at 7.6 # M while caimodulin was varied from 0-21 #M. Specific viscosity in the presence of Ca 2+, 1.1 m M CaC12/1.0 m M E G T A (e); and in the absence of Ca 2+, 1.0 m M E G T A (A).
o
0.4A 0.3-
&F
0.2-
109
o.ojo 0.40,
______--o
ZxF
01,
. 0
0.0. 0.0
,
1.0
2.0
t
I
i
3.0
4.0
5.0
0.00
e
B
0
o
~
0
1O0
[CALMODULIN] (/zM)
200
300
500
400
[NoCI] (mM)
Fig. 7. Effect of calmodulin on the fluorescence of ANM-l~belled chicken gizzard tropomyosin in the presence and absence of Ca 2+. Buffer conditions were the same as those in Fig. 3. Temperature was 2 0 ° C . Experiments done in the presence of Ca 2÷ were 1.1 mM CaCl2/1.0 mM EGTA while those done in the absence of Ca 2+ were 1.0 mM EGTA. (A) The ANM-tropomyosin was kept constant at 0.76 /,tM and the concentration of calmodulin was varied from 0-5 #M. [NaCI] was 10 mM. z ~ F / Fo of ANM-tropomyosin in the presence of Ca 2+ (0); or in the absence of Ca 2+ (0). (B) The NaC1 concentration was varied from 10-450 mM. ANM-tropomyosin and calmodulin were kept constant at 0.76 and 5.0 #M, respectively. A F / F o of ANM-tropomyosin+calmodulin in the presence of Ca 2+ (©); or in the absence of Ca 2+ (0). F0 was the fluorescence of ANM-tropomyosin alone.
(Fig. 7b). In the absence of C a 2+, calmodulin had a much smaller effect on the fluorescence of A N M tropomyosin. Since the binding was relatively weak, no attempt was made to estimate the binding parameters.
Binding of tropomyosin to a calmodulin-Sepharose 4B affinity column In order to demonstrate direct binding between unmodified tropomyosin and calmodulin, chicken gizzard tropomyosin was applied to a calmodulin-Sepharose 4B column (Fig. 8). In the presence of Ca 2÷, about 70% of the tropomyosin was retained and eluted at about 140 m M NaC1. In the absence of Ca 2÷, about 15% the tropomyosin applied to the column was retained and eluted at 65 m M NaCI. These results confirmed the Ca2+-sensitive interaction between the two proteins. Effect of caldesmon on the fluorescence of the A N M tropomyosin : calmodulin complex in the presence and absence of Ca 2÷ To determine if a ternary complex among the three proteins could be detected using fluorescence measurements, we titrated the A N M - t r o p o m y o s i n : c a l m o d u l i n
complex (2 mol c a l m o d u l i n / m o l tropomyosin) with caldesmon, in the presence and absence of Ca 2÷. In the absence of Ca 2÷ and caldesmon, calmodulin had little effect on the fluorescence of A N M - t r o p o m y o s i n , A F / F 0 = 0 . 0 4 , as shown in Fig. 9. Increasing the caldesmon concentration caused an increase in the fluorescence of A N M - t r o p o m y o s i n : calmodulin in a manner similar to, but to a lesser degree than, that observed for the titration of A N M - t r o p o m y o s i n with caldesmon. In the presence of Ca 2+, caldesmon has no effect on the fluorescence of the A N M - t r o p o m y o s i n : c a l m o d u l i n c o m p l e x up to a m o l a r r a t i o of 3 : 2 : 1 (caldesmon : calmodulin : tropomyosin). Further increase in the caldesmon concentration resulted in an increase in the A F / F o to 0.4 which is similar to the
0.40
500
0.30 -
400 f
E ew
.300 0.20 -
v
103
Elsevier BBAMCR 12748
Caldesmon, calmodulin and tropomyosin interactions Mark H. Watson, Alison E. Kuhn and Alan S. Mak Department of Biochemistry, Queen's University, Kingston (Canada)
(Received 24 January 1990) (Revised manuscript received10 April 1990)
Key words: Caldesmon;Calmodulin; Tropomyosin;Electron microscopy;Fluorescence Binary complex interactions between caldesmon and tropomyosin, and ealmodulin and tropomyosin, and ternary complex interaction involving the three proteins were studied using viscosity, electron microscopy, fluorescence and affinity chromatography techniques. In 10 m m NaCI, caldesmon decreased the viscosity of chicken gizzard tropomyosin by 7 - 8 fold with a concomitant increase in turbidity (A 33Onto)" Electron micrographs showed spindle-shaped particles in the tropomyosin-caldesmon samples. These results suggest side-by-side aggregation of tropomyosin polymers induced by caldesmon. Binding studies in 10 m m NaCI between caldesmon and chicken gizzard tropomyosin labelled with the fluorescent probe N-(1-anilinonaphthyl-4)maleimide (ANM) gave association constants from 5.3 • 106 to 7.9- 106 m -1 and stoichiometry from 1.0 to 1.4 tropomyosin per caldesmon. Similar binding was observed for rabbit cardiac tropomyosin and caldesmon. Removal of 18 and 11 residues from the C O O H ends of the gizzard and cardiac tropomyosin by carboxypeptidase A, respectively, had no significant effect on their binding to caldesmon. In the presence of Ca 2+, chicken gizzard tropomyosin bound to a calmodulin-Sepharose-4B column and was eluted with a salt concentration of 140 raM. This interaction was weakened in the absence of Ca 2+, and the bound tropomyosin was eluted by 65 m M KCI. ANm-labelled tropomyosin bound calmodulin in the presence of Ca 2 + with a binding constant of 3.5 - 106 M - ! and a binding stoichiometry of 1 to 1.4 tropomyosin per calmodulin. In 10 m m NaCI, calmodulin reduced the specific viscosity of chicken gizzard tropomyosin in the presence of Ca 2+ by 5 fold, while a 1.5-fold reduction in viscosity was observed in the absence of Ca 2+. In either case, no significant increase in turbidity was observed suggesting that calmodulin reduced head-to-tail polymerization of tropomyosin. The interaction of caldesmon with the calmodulin-ANM-tropomyosin complex in the presence and absence of Ca 2+ was also examined. The result is consistent with a model that in the absence of Ca 2+, calmodulin binds weakly to either caldesmon or tropomyosin and has little effect on the tropomyosin-caldesmon interaction; whereas, Ca2+-calmodulin interacts with ealdesmon and reduces its affinity to tropomyosin.
Introduction
Contraction in smooth muscle is initiated by the Ca 2+-dependent phosphorylation of the 20 kDa myosin light chains by the myosin light chain kinase (for review see Ref. 1). This activation process can be reversed by dephosphorylation of the light chain by smooth muscle phosphatases [2]. Although myosin light chain phosphorylation precedes contraction in intact smooth muscle, tension is maintained at a Ca 2+ concentration too low to support light chain phosphorylation [3]. Tension maintenance is a result of formation of slow cycling
Abbreviation: ANM, N-(1-anilinonaphthyl-4)maleimide. Correspondence: A.S. Mak, Department of Biochemistry, Queen's University, Kingston, Ontario, Canada K7L 3N6.
cross-bridges between myosin and actin known as 'latch'. Regulation of this latch mechanism is believed to involve other Ca2+-sensitive factor(s) in the smooth muscle. Intact thin filaments have been shown to activate phosphorylated smooth muscle myosin ATPase in a CaE+-dependent manner raising the possibility that such factors may act on the thin-filament [4,5]. The likely candidate for a thin-filament-linked regulatory protein in smooth muscle is caldesmon, first purified from chicken gizzard by Sobue et al. [6]. The M r of caldesmon is 150000 as determined by SDS-gel electrophoresis [6], and a value of 93 000 was recently reported based on analytical ultracentrifugation data [7]. Recently, a smooth muscle caldesmon has been cloned and expressed; based on the deduced amino acid sequence, a M r of 86 974 was obtained [31]. Caldesmon has been shown to bind to calmodulin [24,51,59], actin [6,51,57], myosin [15] and tropomyosin [25-27,51]; it
0167-4889/90/$03.50 © 1990 Elsevier SciencePublishers B.V. (Biomedical Division)
104 has been found in other smooth muscle and non-muscle cells [8,9,10]. The stoichiometry of actin : tropomyosin : caldesmon is 2 8 : 4 : 1 in native thin filaments [9], Caldesmon inhibits superprecipitation of chicken gizzard actomyosin [11] and the actin-activated Mg-ATPase activity of phosphorylated smooth muscle myosin [12] at low Ca 2+ concentration and the inhibition can be overcome by Ca2+-calmodulin. Caldesmon enhances the binding of smooth muscle H M M [13], S1 and skeletal H M M [14] to actin in the presence of ATP, suggesting that smooth muscle myosin binds to actin at a non-productive site in the presence of caldesmon [14] resulting in the inhibition of, perhaps, the ADP-releasing step in the kinetic cycle [13]. This mechanism may lead to a decrease in the cycling rate of the myosin ATPase activity characteristic of the latch mechanism. Recent findings that caldesmon can cross link actin and myosin by binding directly to the myosin head may also contribute to the enhancement in the binding of myosin to actin by caldesmon [15]. Smooth muscle tropomyosin potentiates actomyosin ATPase 3 - 4 fold even at low S1 to actin ratio. This has been interpreted in terms of the allosteric model of Hill et al. [16], that more of the actin-tropomyosin complex is in the 'turned on' state for the smooth muscle tropomyosin than the skeletal muscle counterpart [17,18]. In the presence of caldesmon, however, the role of smooth muscle tropomyosin is controversial. It has been reported that caldesmon inhibits the tropomyosin potentiation of actomyosin ATPase [19]. While some workers have found that tropomyosin is essential for the caldesmon inhibition [11,9], others have not observed an absolute requirement of tropomyosin for caldesmon inhibition [20-22]. Recently, Chalovich [23] found that at high ionic strength, 50 and 90% of caldesmon inhibition of the Mg 2÷ actomyosin ATPase activity was observed in the absence and presence of tropomyosin, respectively. Direct interaction between tropomyosin and caldesmon has been demonstrated. Tropomyosin binds to a caldesmon-affinity column [24]; and the viscosity of a tropomyosin solution is increased 4.7-fold by caldesmon in the absence of salt and dithiothreitol [25]. Fluorescence analyses of the binding between caldesmon and tropomyosin in the presence and absence of actin was reported using pyrene-labelled caldesmon [26], and interaction between caldesmon and dansyl chloridelabelled tropomyosin has been reported by Fujii et al. [27]. The tropomyosin-binding site has been located on the COOH-terminal 38 kDa fragment of caldesmon, which also contains the actin- and calmodulin-binding sites [28,21,27,50]. D N A sequence analysis [29,31] has shown that this 38 kDa fragment contains a 58 residue domain which has a 43% overall sequence identity with residues Glu-89 to Lys-147 of T n T within a cyanogen
bromide fragment, TnT-CB2, which has been shown to bind close to or at the COOH-terminal end of tropomyosin [30,32]. In this study, we have investigated (1) the binary interactions between caldesmon and tropomyosin, and between tropomyosin and calmodulin, and (2) ternary interaction involving tropomyosin, caldesmon and calmodulin in the presence and absence of Ca 2+, using fluorescence spectroscopy, viscosity measurements, electron microscopy and affinity chromatography. Materials and Methods
Protein preparations Caldesmon from fresh chicken gizzard was prepared according to the method of Bretscher [10]. Tropomyosin from frozen chicken gizzard and rabbit cardiac muscle was prepared as described by Sanders and Smillie [36]. The gizzard tropomyosin was further purified on a DEAE-column to remove nucleic acid contaminations. Non-polymerizable tropomyosin was prepared by treating tropomyosin with carboxypeptidase A (Sigma) as described before [37]. Concentration of tropomyosin was determined by the method of Bradford [38] using rabbit cardiac tropomyosin as standards. Caldesmon and calmodulin concentrations were calculated from 1% __ 1% A 2 8 0 n m - 3.3 [7] and A278,m = 1.9 [40], respectively. The M r for caldesmon, tropomyosin and calmodulin were 87000 [31], 66000 [39] and 18000 [40], respectively. Calmodulin was prepared from frozen bovine brain as in [41]. Fluorescence studies Tropomyosin from either rabbit cardiac and chicken gizzard muscle was reduced with dithiothreitol and the excess dithiothreitol was removed by isoelectric precipitation of tropomyosin at p H 4.5. The reduced tropomyosin was labelled with A N M (Polysciences) as described by Morris and Lehrer [42]. The stoichiometry of labelling was determined by using the extinction coefficient of 10 800 cm -1- M-1 at 345 nm for A N M [43]. For rabbit cardiac tropomyosin, which has 2 cysteine residues per molecule, a label of 1.8 A N M per mol tropomyosin was obtained. A labelling ratio of 2-2.5 was found for the gizzard tropomyosin which contains equal amounts of the B- and "y-isoforms, each containing 1 cysteine residue per 33 kDa chain. Fluorescence measurements were performed on a MPF-66 Perkin-Elmer fluorescence spectrophotometer. The temperature was maintained at 20 ° C by a circulating water thermostat. Binding isotherms for the interaction between caldesmon and tropomyosin or calmodulin ( + C a 2÷) and tropomyosin were obtained by adding known aliquots of caldesmon or calmodulin ( + Ca 2+) to a solution of ANM-tropomyosin (0.05 m g / m l ) in 10 mM imidazole pH 7.0, 10 mM dithiothreitol, 1.0 mM
105 EGTA, 0.01% N a N 3 and 10, 50 or 100 mM NaCI. Corrections were made for dilution during titration. Light scattering due to an identical sample of unlabelled tropomyosin and caldesmon or calmodulin was subtracted from the ANM-tropomyosin fluorescence. Binding isotherms for the ternary complex of ANMtropomyosin, calmodulin and caldesmon were made by mixing ANM-tropomyosin, in the presence or absence of Ca 2÷, with the required amount of calmodulin. This was titrated with known aliquots of caldesmon. Corrections for dilution and light scattering were made with unlabelled proteins as described above. Excitation was at 356 nm and emission was measured at 456 nm. For experiments done in the presence of Ca 2÷, CaC12 was added from a 25 mM stock such that the final solution was 1.1 mM Ca 2÷ and 1.0 mM EGTA. To study the salt effect on the binding of caldesmon or calmodulin to A N M - t r o p o m y o s i n , a solution of caldesmon or calmodulin and ANM-tropomyosin was titrated with NaCI from a stock solution of 5 M NaC1. The protein concentrations are indicated in the figure legends. The association constant ( K ) for a single binding site, the number of caldesmon-binding sites per molecule of tropomyosin (n), and the maximum change in fluorescence at saturation (AFmax) were analyzed similarly to that described by Morris and Lehrer [42]. In the binding between tropomyosin and caldesmon to form a complex, TC,, it can be shown that: K n T b 2 - b ( K T n + K C + 1) + K C = 0
where T = total concentration of tropomyosin, C = total concentration of caldesmon, and b -- A F / A Fm~, (fraction of total sites in tropomyosin with bound-caldesmon) where A F = the change in tropomyosin fluorescence intensity at each addition of caldesmon. This equation assumes non-cooperative binding of caldesmon to tropomyosin if n is not equal to 1. Binding parameters were determined by fitting the values of T, C and b to the above equation by computer using a non-linear least square program.
freshly prepared 1.0 M dithiothreitol stock to make 20 mM. The samples were sealed and left at 20 ° C for 1 h before the viscosity was measured.
Electron microscopy Chicken gizzard tropomyosin, 15 #M, in 10 mM imidazole, 10 mM NaCI, 1.0 mM EGTA, 0.01% NaN 3, 10 m M dithiothreitol (pH 7.0), was incubated with 45 /tM of caldesmon for 30 rain at room temperature. The solution was centrifuged at 280000 × g for 10 min and the pellet was supended in 50/~1 of the same buffer. The protein samples were adsorbed 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. Control samples of tropomyosin and caldesmon alone were examined similarly.
Calmodufin-Sepharose 4B affinity chromatography Purified calmodulin (8 mg) was coupled to CNBractivated Sepharose 4B (Pharmacia) (2 g dry weight) following the manufacturers recommended procedure. The buffer and chromatography conditions are described in the figure legends. Results
Effect of caldesmon on the viscosity of chicken gizzard tropomyosin At a protein concentration of 7.6/xM (0.5 m g / m l ) in 10 m M NaC1 and 20 mM dithiothreitol, the specific viscosity 0Jsp) of gizzard tropomyosin was 3.5, as shown in Fig. 1. Increase in ionic strength reduced the viscosity of the tropomyosin solution in a manner similar to that reported by others [36,44]. Caldesmon alone had very low viscosity under the same conditions, 7/sp = 0. 4.0-
-0.15
3.o ~)
-0.I0
2.0
i
300 m M NaC1, no change in fluorescence was observed 3.0-
2.0
o
1.0 m o3
0.O
O.0
I
I
I
1.0
2.0
3.0
CALMODULIN/TROPOMYOSIN (mol/mol)
Fig. 6. Effect of calmodulin on the viscosity of chicken gizzard tropomyosin at 10 m M NaCl in the presence and absence of Ca 2+. Buffer conditions were as described in the Fig. 1, except that [NaCl] was kept constant at 10 m M . Temperature was 2 0 ° C . Viscosity measurements were repeated three times using different preparations of tropomyosin and calmodulin. Each point in the titration was an average of two measurements. The tropomyosin concentration was kept constant at 7.6 # M while caimodulin was varied from 0-21 #M. Specific viscosity in the presence of Ca 2+, 1.1 m M CaC12/1.0 m M E G T A (e); and in the absence of Ca 2+, 1.0 m M E G T A (A).
o
0.4A 0.3-
&F
0.2-
109
o.ojo 0.40,
______--o
ZxF
01,
. 0
0.0. 0.0
,
1.0
2.0
t
I
i
3.0
4.0
5.0
0.00
e
B
0
o
~
0
1O0
[CALMODULIN] (/zM)
200
300
500
400
[NoCI] (mM)
Fig. 7. Effect of calmodulin on the fluorescence of ANM-l~belled chicken gizzard tropomyosin in the presence and absence of Ca 2+. Buffer conditions were the same as those in Fig. 3. Temperature was 2 0 ° C . Experiments done in the presence of Ca 2÷ were 1.1 mM CaCl2/1.0 mM EGTA while those done in the absence of Ca 2+ were 1.0 mM EGTA. (A) The ANM-tropomyosin was kept constant at 0.76 /,tM and the concentration of calmodulin was varied from 0-5 #M. [NaCI] was 10 mM. z ~ F / Fo of ANM-tropomyosin in the presence of Ca 2+ (0); or in the absence of Ca 2+ (0). (B) The NaC1 concentration was varied from 10-450 mM. ANM-tropomyosin and calmodulin were kept constant at 0.76 and 5.0 #M, respectively. A F / F o of ANM-tropomyosin+calmodulin in the presence of Ca 2+ (©); or in the absence of Ca 2+ (0). F0 was the fluorescence of ANM-tropomyosin alone.
(Fig. 7b). In the absence of C a 2+, calmodulin had a much smaller effect on the fluorescence of A N M tropomyosin. Since the binding was relatively weak, no attempt was made to estimate the binding parameters.
Binding of tropomyosin to a calmodulin-Sepharose 4B affinity column In order to demonstrate direct binding between unmodified tropomyosin and calmodulin, chicken gizzard tropomyosin was applied to a calmodulin-Sepharose 4B column (Fig. 8). In the presence of Ca 2÷, about 70% of the tropomyosin was retained and eluted at about 140 m M NaC1. In the absence of Ca 2÷, about 15% the tropomyosin applied to the column was retained and eluted at 65 m M NaCI. These results confirmed the Ca2+-sensitive interaction between the two proteins. Effect of caldesmon on the fluorescence of the A N M tropomyosin : calmodulin complex in the presence and absence of Ca 2÷ To determine if a ternary complex among the three proteins could be detected using fluorescence measurements, we titrated the A N M - t r o p o m y o s i n : c a l m o d u l i n
complex (2 mol c a l m o d u l i n / m o l tropomyosin) with caldesmon, in the presence and absence of Ca 2÷. In the absence of Ca 2÷ and caldesmon, calmodulin had little effect on the fluorescence of A N M - t r o p o m y o s i n , A F / F 0 = 0 . 0 4 , as shown in Fig. 9. Increasing the caldesmon concentration caused an increase in the fluorescence of A N M - t r o p o m y o s i n : calmodulin in a manner similar to, but to a lesser degree than, that observed for the titration of A N M - t r o p o m y o s i n with caldesmon. In the presence of Ca 2+, caldesmon has no effect on the fluorescence of the A N M - t r o p o m y o s i n : c a l m o d u l i n c o m p l e x up to a m o l a r r a t i o of 3 : 2 : 1 (caldesmon : calmodulin : tropomyosin). Further increase in the caldesmon concentration resulted in an increase in the A F / F o to 0.4 which is similar to the
0.40
500
0.30 -
400 f
E ew
.300 0.20 -
v
