Báo cáo khoa học: Comparative studies on the functional roles of N- and C-terminal regions of molluskan and vertebrate troponin-I

Comparative studies on the functional roles of N- and C-terminal regions of molluskan and vertebrate troponin-I Hiroyuki Tanaka1, Yuhei Takeya1, Teppei Doi1, Fumiaki Yumoto2,3, Masaru Tanokura3, Iwao Ohtsuki2, Kiyoyoshi Nishita1 and Takao Ojima1

1 Laboratory of Biotechnology and Microbiology, Graduate School of Fisheries Sciences, Hokkaido University, Japan 2 Laboratory of Physiology, The Jikei University School of Medicine, Tokyo, Japan 3 Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, University of Tokyo, Japan

Keywords invertebrate; mollusk; regulatory mechanism; troponin; troponin-I

Correspondence Takao Ojima, Laboratory of Biochemistry and Biotechnology, Graduate School of Fisheries Sciences, Hokkaido University, Hakodate, Hokkaido 041–8611, Japan Tel ⁄ Fax: +81 138 408800 E-mail: ojima@fish.hokudai.ac.jp

Note The nucleotide sequences of cDNAs enco- ding Akazara scallop 52K-TnI and 19K-TnI are available in DDBJ ⁄ EMBL ⁄ GenBank databases under accession numbers, AB206837 and AB206838, respectively

(Received 24 March 2005, revised 13 June 2005, accepted 15 July 2005)

doi:10.1111/j.1742-4658.2005.04866.x

Vertebrate troponin regulates muscle contraction through alternative bind- ing of the C-terminal region of the inhibitory subunit, troponin-I (TnI), to actin or troponin-C (TnC) in a Ca2+-dependent manner. To elucidate the molecular mechanisms of this regulation by molluskan troponin, we com- pared the functional properties of the recombinant fragments of Akazara scallop TnI and rabbit fast skeletal TnI. The C-terminal fragment of Akaz- ara scallop TnI (ATnI232)292), which contains the inhibitory region (resi- dues 104–115 of rabbit TnI) and the regulatory TnC-binding site (residues 116–131), bound actin-tropomyosin and inhibited actomyosin-tropomyosin Mg-ATPase. However, it did not interact with TnC, even in the presence of Ca2+. These results indicated that the mechanism involved in the alter- native binding of this region was not observed in molluskan troponin. On the other hand, ATnI130)252, which contains the structural TnC-binding site (residues 1–30 of rabbit TnI) and the inhibitory region, bound strongly to both actin and TnC. Moreover, the ternary complex consisting of this frag- ment, troponin-T, and TnC activated the ATPase in a Ca2+-dependent manner almost as effectively as intact Akazara scallop troponin. Therefore, Akazara scallop troponin regulates the contraction through the activating mechanisms that involve the region spanning from the structural TnC- binding site to the inhibitory region of TnI. Together with the observation that corresponding rabbit TnI-fragment (RTnI1)116) shows similar activa- ting effects, these findings suggest the importance of the TnI N-terminal region not only for maintaining the structural integrity of troponin com- plex but also for Ca2+-dependent activation.

understand the molecular mechanisms of this Ca2+ switching, extensive studies of the structure, function, and Ca2+-dependent conformational changes of tropo- nin subunits have been carried out.

Troponin is a Ca2+-dependent regulatory protein com- plex, which constitute thin filaments together with actin and tropomyosin [1]. It is composed of three dis- tinct subunits: troponin-C (TnC), which binds Ca2+, troponin-T (TnT), which binds tropomyosin, and trop- onin-I (TnI), which binds actin and inhibits actin–myo- sin interaction [2–4]. In relaxed muscle, TnI binds to actin and inhibits contraction. Upon muscle stimula- tion, Ca2+ binds to TnC and induces the release of the inhibition by TnI, resulting in muscle contraction. To

In vertebrate muscles, TnC has a dumbbell-like shape with the N- and C-terminal globular domains linked by a central helix [5,6]. Each domain contains two EF-hand Ca2+-binding motifs [7], thus TnC has four possible Ca2+-binding sites, sites I and II in the N-domain and sites III and IV in the C-domain [8,9].

Abbreviations TnC, troponin-C; TnI, troponin-I; TnT, troponin-T; IPTG, isopropyl-1-thio-b-D-galactopyranoside; PMSF, phenylmethylsulfonyl fluoride.

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4475

H. Tanaka et al.

Functional regions of molluskan TnI

responsible not only for Ca2+-binding but also for the interaction with TnI, although the presence of both the N- and C-domains is essential for full regulatory ability [24,25].

In the present study, we compared the functional sites of molluskan and vertebrate TnI by using the recombinant fragments of Akazara scallop Chlamys nipponensis TnI and rabbit fast skeletal TnI. The results provide evidence that molluskan troponin func- tions through a mechanism in which the region span- ning from the structural TnC-binding site to the inhibitory region of TnI plays an important role.

Results

Escherichia coli expression of TnI-fragments

the

(isoforms;

Sites III and IV also show affinity for Mg2+ and are thought to be always occupied by sarcoplasmic Mg2+, whereas Ca2+ binding to site I and ⁄ or II is believed to trigger muscle contraction [10]. TnC interacts with both TnI and TnT. The TnC–TnI interaction and changes in the interaction upon Ca2+ binding to TnC have been intensively studied as the central mecha- nisms of Ca2+ switching. It has been revealed that TnI has three major TnC-binding sites [11–14], namely a structural TnC-binding site (residues 1–30 in rabbit fast skeletal TnI), an inhibitory region (residues 104– 115), and a regulatory TnC-binding site (residues 116– 131). In the relaxed state, the inhibitory region binds to actin and inhibits actin–myosin interaction [11,12], while in the contractile state, Ca2+-binding to site I and ⁄ or II of TnC causes the exposure of a hydropho- bic patch on the surface of the N-domain [15], result- ing in hydrophobic interaction between the N-domain and the regulatory TnC-binding site [16]. This inter- action induces the dissociation of the inhibitory region, which is adjacent to the regulatory TnC-binding site, from actin, resulting in the release of the inhibition and muscle contraction [17]. The structural TnC-bind- ing site interacts with the C-domain of TnC in both the relaxed and contractile states, which plays a role in maintaining the structural integrity of the troponin complex [17,18]. These switching mechanisms were recently confirmed by crystallographic studies of ver- tebrate troponins [19,20], which demonstrated that the Ca2+-saturated N- and C-domains of TnC bind to the regulatory and structural TnC-binding sites, respect- ively, of TnI, and suggested that the C-terminal region of TnI (including the inhibitory region and the regula- tory TnC-binding site) exhibits a positional change from actin-tropomyosin filament to the N-domain of TnC in a Ca2+-dependent manner.

these proteins except

for all

Figure 1A shows a schematic representation of the recombinant TnI-fragments used in this study. ATnI- recombinant 52K, ATnI-19K and RTnI are Akazara scallop 52K-TnI, 19K-TnI see Experimental procedures section and [27]), and rabbit fast skeletal TnI, respectively. ATnI1)128 is the frag- ment corresponding to the N-terminal extending region of 52K-TnI. ATnI130)252 and RTnI1)116 are the frag- ments, corresponding to the regions spanning from the structural TnC-binding sites to the inhibitory regions of Akazara scallop and rabbit TnI, respectively. ATnI232)292 and RTnI96)181 correspond to the regions spanning from the inhibitory regions to the C-termini of these TnI. Figure 1B shows an SDS ⁄ PAGE of these purified recombinant proteins. ATnI-52K and ATnI1)128 showed anomalously low mobility due to the high fraction of hydrophilic residues in the N-ter- minal extending region as described previously [26]. The initiator Met at the N-terminus was removed by the bacterial cell for RTnI96)181.

Inhibition of Mg-ATPase of actomyosin by TnI-fragments

revealed the importance of

The inhibition of actomyosin-tropomyosin Mg-ATPase by TnI fragments was compared. The inhibitory effects of RTnI, RTnI1)116 and RTnI96)181 differed greatly from one another, although all of these proteins contained the inhibitory region (Fig. 2A). RTnI1)116 inhibited only 33% of rabbit-actomyosin–rabbit-tropo- myosin Mg-ATPase at a 3 : 1 molar ratio with tropo- myosin, compared with 82% for RTnI. As has been reported previously [18,28,29], weaker inhibitory effects residues of RTnI1)116 117–181 for maximal inhibition. In particular, residues

However, a significant discrepancy exists between the above schemes and the structural and functional features of some invertebrate troponins. Molluskan TnC binds only one mole of Ca2+ per mole of protein at site IV in the C-domain because of amino acid sub- stitutions at sites I–III [21,22]. Nevertheless, ternary troponin complex combined with molluskan tropomyo- sin can regulate the Mg-ATPase activity of vertebrate actomyosin in a physiologically significant Ca2+- dependent manner [21]. Moreover, the troponin regu- lates the ATPase of molluskan myofibril together with a well known myosin light chain-linked regulatory sys- tem, especially under low temperature conditions [23]. Therefore, the molecular mechanisms of regulation by molluskan troponin are expected to be somewhat dif- ferent from those described above. A previous study the C-domain of molluskan TnC is revealed that

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4476

H. Tanaka et al.

Functional regions of molluskan TnI

A

B

Fig. 1. (A) Schematic representation of recombinant TnI-fragments. The numbers preceding and following each box indicate the amino acid positions of Akazara scallop 52K-TnI (Swiss-Prot #Q7M3Y3) and rabbit fast skeletal TnI (Swiss-Prot #P02643). The N-terminal extending region of 52K-TnI and the functional regions that have been previously identified in vertebrate TnI are indicated by bars. The inhibitory regions are shaded. (B) SDS ⁄ PAGE of recombinant TnI-fragments used in this study. Each protein (1.5 lg) was run on a 10% (w/v) acryl- amide gel. Molecular mass markers are also shown (M).

to the N-terminal

the N-terminal extending region of 52K-TnI could decrease the inhibitory effects, although ATnI1)128, which corresponds extending region, on its own, exhibited neither activation nor inhibition.

140–148 had been proven to bind to actin-tropomyosin and thus are referred to as the second actin-tropo- myosin-binding site [14]. Moreover, in our results, the inhibition by RTnI96)181 was the strongest (94% of the ATPase was inhibited), suggesting that residues 1–95 may decrease the inhibitory effects of residues 96–181.

the

(51%)

On the other hand, Akazara scallop TnI isoforms and their fragments showed somewhat different pro- perties (Fig. 2B). ATnI130)252, which corresponds to RTnI1)116, inhibited about 70% of rabbit-actomyosin- scallop-tropomyosin Mg-ATPase at a 3 : 1 molar ratio with tropomyosin. Moreover, inhibition by ATnI232)292, which corresponds to RTnI96)181, was weaker than that by ATnI-19K (88%) or ATnI130)252. Therefore, the effects of the N- or C-ter- minal region of TnI on the function of the inhibitory region appeared to differ between rabbit and Akazara scallop TnI. Interestingly, ATnI-52K showed weaker suggesting that than ATnI-19K, inhibition (65%)

To determine whether the inhibitory effect correlates with the binding affinity to actin-tropomyosin, we examined each TnI for its ability to cosediment with actin-tropomyosin. When TnI-fragments were mixed at 2 : 1 molar ratios with tropomyosin, RTnI, RTnI1)116 and RTnI96)181 cosedimented with molar ratios of approximately 0.23, 0.048, and 0.35, respectively, to actin. On the other hand, ATnI-19K, ATnI130)252 and ATnI232)292 cosedimented with molar ratios of 0.49, 0.44, and 0.065, respectively, to actin (the extent of the cosedimentation of ATnI-52K could not be deter- mined because it precipitated even in the absence of actin-tropomyosin in a control experiment due to the low solubility). Therefore, the observed difference in the inhibitory effects of TnI-fragments might be

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4477

H. Tanaka et al.

Functional regions of molluskan TnI

Interactions of TnI-fragments with TnC

corresponds

We compared the ability of TnI-fragments to form a complex with TnC by alkaline urea PAGE. The experi- ments were performed under either 6 or 3 m urea condi- tions in the presence of either 2 mm EDTA or 2 mm CaCl2. RTnI and both rabbit TnI-fragments formed a complex with rabbit TnC in 2 mm CaCl2 but not in 2 mm EDTA under both urea conditions (Fig. 3A). These results agreed with those reported by Farah et al. for chicken skeletal TnI-fragments [18], and were com- patible with the fact that all of these proteins have at least two of three known TnC-binding sites, namely the structural TnC-binding site, the inhibitory region, and the regulatory TnC-binding site. On the other hand, ATnI1)128 and ATnI232)292 did not form a complex with Akazara scallop TnC under any of the tested conditions, whereas ATnI-52K, ATnI-19K, and ATnI130)252 did under both urea concentrations in the presence of Ca2+ (Fig. 3B). It was interesting that ATnI232)292 did not to form a complex, as ATnI232)292 RTnI96)181 and should have two TnC-binding sites, the inhibitory region and the regulatory TnC-binding site. Therefore, this suggests that TnC-binding affinities of these regions of the Akazara scallop TnI were much weaker than those of rabbit TnI. Moreover, under the 3 m urea condition, ATnI-52K, ATnI-19K, and ATnI130)252 showed complex formation even in the absence of Ca2+ (Fig. 3B, upper panels), suggesting that in the absence of Ca2+, the Akazara scallop TnI binds to TnC more strongly than rabbit due to the properties of the interaction between residues 130–252 and TnC.

We also performed affinity chromatography to con- firm the interaction of TnI-fragments with immobilized rabbit or Akazara scallop TnC under nondenaturing conditions (Fig. 4). ATnI232)292 binding to Akazara scallop TnC was not observed, even in the absence of both urea and KCl and the presence of 0.5 mm CaCl2, whereas ATnI130)252, RTnI1)116, and RTnI96)181 strongly bound to TnCs. These results suggested that the inhibitory region and the regulatory TnC-binding site of Akazara scallop TnI essentially cannot interact with TnC.

Fig. 2. Inhibition of actomyosin-tropomyosin Mg-ATPase by rabbit (A) or Akazara scallop (B) TnI-fragments. The actomyosin-tropo- increasing ratios of TnI myosin Mg-ATPase was measured at or TnI-fragments to tropomyosin as indicated on the abscissa. The measurements were performed at 15 (cid:1)C. The results were expressed as a percentage of the ATPase activity obtained in the absence of TnI. Each point is an average of three determinations. (A) RTnI, d; RTnI1)116, n; RTnI96)181, h. (B) ATnI-52K, d; ATnI- 19K, s; ATnI1)128, e; ATnI130)252, n; ATnI232)292, h.

Ca2+-dependent alternative binding of C-terminal TnI fragments to actin-tropomyosin and TnC

although

region

this

attributable to the difference in their binding affinities for actin-tropomyosin. In addition, ATnI1)128 did not cosediment and remained in the supernatant (data not shown). This suggested that the N-terminal extending region of 52K-TnI was not involved in binding to showed actin-tropomyosin, sequence homology to the N-terminal tropomyosin binding site of vertebrate TnT [26].

To understand the biological significance of the differ- ence in TnI–TnC interactions, we compared the ability of TnC to neutralize the inhibitory effects of the C-ter- minal fragments in the presence and absence of Ca2+. As has been reported for similar vertebrate TnI frag- ments [14,18,29], the inhibitory effect of RTnI96)181 in

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4478

H. Tanaka et al.

Functional regions of molluskan TnI

A

B

Fig. 3. Complex formation between TnI-fragments and TnC detected by alkaline urea PAGE. TnI-fragments were combined with TnC as des- cribed under ‘Experimental procedures’. The final concentration of the proteins was 13.8 lM. Twenty-microliter aliquots of the mixture were electrophoresed on the gel containing either 6 or 3 M urea and either 2 mM EDTA (– Ca; upper panels) or 2 mM CaCl2 (+ Ca; lower panels). (A) Rabbit TnI or TnI-fragments were run on the gels in the absence (lanes a–c) or presence (lanes d–f) of equimolar amounts of rabbit TnC. Lanes a and d, RTnI; lanes b and e, RTnI1)116; lanes c and f, RTnI96)181; lane g, rabbit TnC. (B) Akazara scallop TnI or TnI-fragments were run in the absence (lanes h–l) or presence (lanes m–q) of equimolar amounts of Akazara scallop TnC. Lanes h and m, ATnI-52K; lanes i and n, ATnI-19K; lanes j and o, ATnI1)128; lanes k and p, ATnI130)252; lanes l and q, ATnI232)292; lane r, Akazara scallop TnC. Complex forma- tion was detected by the bands of the TnI–TnC complex (arrowheads) and weakening of the free TnC bands. Free RTnI, RTnI1)116, RTnI96)181, ATnI-19K, ATnI130)252, and ATnI232)292 did not migrate into the gels, while free ATnI-52K and ATnI1)128 exhibited a band near the origin and at the middle of the gel, respectively. The bands corresponding to the free rabbit or Akazara scallop TnC were found in the middle to bottom of the gels (indicated as RTnC or ATnC, respectively).

Fig. 4. TnC-affinity chromatography of TnI- fragments. The fragments of rabbit or Akaz- ara scallop TnI were applied onto the affinity columns prepared by immobilizing either rabbit (A) or Akazara scallop (B) TnC on Formyl-Cellulofine. The fragments were eluted with a stepwise gradient of KCl concentrations indicated at the top of the figures. Each fraction contains 1.0 mL. Eluted protein was detected by the method of Bradford [40] and identified by SDS ⁄ PAGE (data not shown). Due to low solubility, RTnI1)116 was applied at a KCl concentration of 0.1 M.

a 2 : 1 molar ratio with tropomyosin was effectively neutralized by rabbit TnC in the presence of Ca2+, but not in its absence (Fig. 5A, upper panel). In addi-

tion, the cosedimentation experiment performed under a 4 : 4 : 2 : 7 molar ratio of RTnI96)181–TnC–tropo- myosin–actin showed that the amount of RTnI96)181

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4479

H. Tanaka et al.

Functional regions of molluskan TnI

A

B

Fig. 5. Functional differences between RTnI96)181 (A) and ATnI232)292 (B). Upper panels, effects of TnC on inhibition by the C-terminal TnI- fragments. TnI-fragments were present at a 2 : 1 molar ratio of TnI-fragments ⁄ tropomyosin. The Mg-ATPase activity was measured at increasing ratios of TnCs to the fragments in the presence (d) or absence (s) of Ca2+. The measurements were performed at 15 (cid:1)C. The results were expressed as a percentage of the ATPase activity obtained in the absence of both TnI and TnC. Lower panels, change in C-ter- minal TnI-fragment affinity for actin-tropomyosin tested by cosedimentation experiments. The fragments were added to actin-tropomyosin at a molar ratio of 4 : 2 : 7 (fragment ⁄ tropomyosin ⁄ actin) with or without an equimolar amount of TnC in the presence or absence of Ca2+. The pellets (P) and supernatants (S) were redissolved in equivalent volumes of 5 M urea solution and then run on SDS ⁄ PAGE. Lanes a and d, in the absence of both TnC and Ca2+; lanes b and e, in the presence of TnC and the absence of Ca2+; lanes c and f, in the presence of both TnC and Ca2+. Ac, actin; Tm, tropomyosin; RTnC, rabbit TnC; ATnC, Akazara scallop TnC. The relative staining intensities of the C-terminal TnI-fragments on lanes a–c were expressed as a percentage of that on lane a and were shown on the right.

concentrations (Fig. 5B, upper panel). Moreover, the amount of ATnI232)292 cosedimented with actin-tropo- myosin was unaffected by the presence and absence of TnC and Ca2+ (Fig. 5B, lower panel). Therefore, the Ca2+-switching mechanisms involving the alternative binding of the C-terminal region of TnI were not pre- sent in Akazara scallop troponin.

Ca2+-regulatory effects of troponins containing TnI fragments

cosedimented with actin-tropomyosin was greatly reduced in the presence of Ca2+ but not in its absence. The amount that remained with TnC in the super- natant was greater in the presence of Ca2+ than in its absence (Fig. 5A, lower panel). Therefore, this sugges- ted that RTnI96)181 bound actin and TnC in the absence and presence, respectively, of Ca2+. These phenomena should directly reflect the mechanism of Ca2+ switching involving the alternative binding of the C-terminal region of TnI to actin or TnC in a Ca2+- dependent manner [17,19]. On the other hand, the inhibitory effect of ATnI232)292 was not neutralized by adding Akazara scallop TnC, irrespective of Ca2+

The Ca2+-regulatory effects of troponins composed of TnI-fragments, native TnT, and TnC on actomyosin-

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4480

H. Tanaka et al.

Functional regions of molluskan TnI

Fig. 6. Ca2+-regulation of actomyosin-tropo- myosin Mg-ATPase by rabbit (A and C) and Akazara scallop (B and D) reconstituted tropo- nins. The effects of the troponin containing TnI or TnI fragments on the actomyosin- tropomyosin Mg-ATPase were measured as a function of pCa ()10g[Ca2+]). The assays were performed at 15 (cid:1)C (A and B) or 25 (cid:1)C (C and D). A and C: RTn, d; RTn1)116, n; RTn96)181, h. B and D: ATn-52K, d; ATn- 19K, s; ATn130)252, n; ATn232)292, h. The activities in the absence of troponin are indi- cated by dashed lines.

regulate

inhibitory

and

the

the

that the region spanning from the regulatory TnC- binding site to the C-terminus of Akazara scallop TnI is not important for this regulation, and that Akazara scallop troponin acts through mechanisms in which the region spanning from the structural TnC-binding site to the inhibitory region plays an important role. It should also be mentioned that ATn-52K more strongly activated the ATPase than ATn-19K under high Ca2+ concentrations. Thus, the N-terminal extending region of ATnI-52K may be involved in the activation of the ATPase in the presence of Ca2+. When we performed similar assays at 25 (cid:1)C, the regulation by RTn1)116, which was observed at 15 (cid:1)C, became unremarkable, effectively regulated the whereas RTn96)181 more ATPase results than at 15 (cid:1)C (Fig. 6C). These obtained at 25 (cid:1)C were essentially the same as those reported by Farah et al. [18] for the chicken skeletal troponins containing similar TnI fragments. On the other hand, the regulatory ability of Akazara scallop troponins dramatically decreased (Fig. 6D), suggesting that Akazara scallop troponin does not function at the temperature appropriate for vertebrate troponins.

Discussion

as

tropomyosin Mg-ATPase were compared. The assays were performed at different temperatures, 15 (cid:1)C, which is the normal ambient temperature for Akazara scal- lops and is suitable for functionalizing the molluskan troponin [23], and 25 (cid:1)C, at which many assays of Ca2+ regulation by vertebrate troponin have been con- ducted [14,18,28–30]. At 15 (cid:1)C, all the ternary com- plexes consisting of rabbit TnI or TnI fragments, rabbit TnT and TnC, regulated the ATPase, although they exhibited quite different Ca2+-dependence curves (Fig. 6A). The complex containing RTnI1)116 (repre- sented as RTn1)116) showed no inhibition, even under low Ca2+ concentrations, although it strongly activa- ted the ATPase at Ca2+ concentrations higher than pCa 4.5. RTn96)181 did not activate the ATPase beyond the level observed in the absence of troponin, even at pCa 4.0. On the other hand, the complex con- sisting of ATnI232)292, Akazara scallop TnT and TnC (ATn232)292) inhibited the ATPase irrespective of Ca2+ concentration, and could not it at all (Fig. 6B). This property could be explained by the fact that regulatory region TnC-binding site of Akazara scallop TnI bind to actin- irrespective of Ca2+ tropomyosin, but not to TnC, above. Moreover, described concentration, ATn130)252 regulated the ATPase almost as effectively as intact troponins (ATn-52K or ATn-19K), suggesting

The vertebrate TnI is known to interact with TnC in an antiparallel manner such that the regulatory and

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4481

H. Tanaka et al.

Functional regions of molluskan TnI

integrity of

structural TnC-binding sites of TnI interact with the N- and C-domains, respectively, of TnC [18,19]. The inhibitory region is known to interact with both the N- and C-domains, but preferentially with the C-domain [18,20,31]. In the present study, we revealed a striking difference in the TnI–TnC interactions of vertebrate and mollusk. We showed that ATnI232)292, which is the Akazara scallop TnI-fragment containing the inhibitory region and the regulatory TnC-binding site, does not bind to Akazara scallop TnC, whereas ATnI130)252, which contains the structural TnC-bind- ing site and the inhibitory region, strongly binds to TnC. The antiparallel structural features of vertebrate TnI–TnC complex and previous observations that the N-domain of Akazara scallop TnC did not bind to TnI while the C-domain bound strongly [24], suggest a single interaction between the structural TnC-binding site of TnI and the C-domain of TnC in Akazara scal- lop TnI–TnC complex. Although the further verifica- tion under nondenaturing conditions is required, the results of the alkaline urea gel electrophoresis indicate that this interaction is strengthened by Ca2+ and is stronger than the corresponding interaction in rabbit TnI–TnC in the absence of divalent cation. Therefore, this interaction potentially participates in both the Ca2+-dependent activation of the contraction and the maintenance of structural the troponin complex in the relaxed state.

molluskan and vertebrate TnI and revealed for the first time that (a) the alternative binding of the TnI C-terminal region is not observed in molluskan tropo- nin, as the C-terminal region of molluskan TnI does not interact with TnC; and (b) molluskan troponin regulates the ATPase by a mechanism in which the TnI N-terminal region (from the structural TnC-bind- ing site to the inhibitory region) participates in the Ca2+-dependent activation. In addition, at 15(cid:1)C, sim- ilar activation is observed for the troponin containing the corresponding vertebrate TnI-fragment, suggesting the presence of a common activating mechanism In molluskan between vertebrates and mollusks. troponin, the activation is probably induced by streng- thening of the interaction between the structural TnC- binding site and the C-domain of TnC accompanying Ca2+ binding to site IV of TnC. In vertebrate tropo- nin, the activation may be a result of the interaction between the inhibitory region and TnC accompanying Ca2+ binding to site I or II of TnC. However, we can- not rule out the possibility that the substitution of Mg2+ at site III or IV of vertebrate TnC with Ca2+ causes the activation in vitro. Several observations have indicated that the N-terminal region of vertebrate TnI is involved in the activating process [14,28,30]. In particular, Malnic et al. [30] suggested that the activa- ting effects of the N-terminal region of TnT are exer- ted in the presence of Ca2+ by the TnI N-terminal region (from the structural TnC-binding site to the TnT-binding site) and TnC.

In summary, we propose a novel view of the general architecture of TnI. In vertebrate muscles, the C-ter- minal region plays a role in the inhibition ⁄ removal of inhibition by alternative binding, while the N-terminal region is responsible for the Ca2+-dependent activa- tion. This view replaces the general and conventional view that the N-terminal region of TnI only plays a role in maintaining the structural integrity of the tro- ponin complex. In molluskan muscles, the C-terminal region does not function and troponin regulates contraction only through the activation exerted by the N-terminal region of TnI.

Experimental procedures

Muscle proteins

Tropomyosin, TnT, and TnC from Akazara scallop striated adductor muscle or rabbit fast skeletal muscle were pre- pared by the method of Ojima and Nishita [21,34]. Rabbit fast skeletal myosin and F-actin were prepared by the method of Perry [35] and Spudich and Watt [36], respect- ively. All measures were taken to minimize pain and

Troponin-tropomyosin based regulation exhibits two components [32]: inhibition and removal of inhibition in the absence and presence, respectively, of Ca2+, and Ca2+-dependent activation. The regulatory mech- anism involving the alternative binding of the C-ter- minal region of TnI to actin or TnC should be responsible for the former. However, it cannot account for the latter, namely the phenomenon that, in the presence of Ca2+, troponin activates actomyosin- tropomyosin Mg-ATPase beyond the level observable in the absence of troponin. This activation is promin- ent, especially for molluskan troponin, which confers Ca2+ sensitivity on the ATPase predominantly through its activation in the presence of Ca2+, rather than by inhibition due to its absence. In contrast, the vertebrate troponin regulates the ATPase mainly by inhibition in the absence of Ca2+ (Fig. 6 and [21,32]). The difference in Ca2+ sensitization between verte- brates and mollusks should also be closely related to the difference in the inhibitory effects of vertebrate and molluskan tropomyosins [33], which inhibit rab- bit actomyosin Mg-ATPase activity to 0.043 and 0.021 lmolÆmin)1Æmg myosin)1, respectively, at 15(cid:1)C (Fig. 6A,B). In the present study, we compared the functional roles of the N- and C-terminal regions of

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4482

H. Tanaka et al.

Functional regions of molluskan TnI

discomfort of animals. The procedures were conducted in accordance with the institutional guidelines by Hokkaido University.

Construction of plasmids expressing TnI fragments

two cDNA clones

mutated DNA was cut out with NcoI and BamHI and ligated into pET-16b for the construction of the plasmid expressing RTnI (recombinant rabbit fast skeletal TnI; resi- dues 1–181). The expression plasmid was also used as a template for PCR to amplify the DNA encoding RTnI1)116 (fragment; residues 1–116 of rabbit fast skeletal TnI) and RTnI96)181 (fragment; residues 96–181), using the primer sets RTnI1F and RTnI116R (5¢-GAGCATGGCGGGAT CCTACATGCGCAC-3¢) and RTnI96F (5¢-GCTGGAGG CCATGGACCAGAAGC-3¢) and RTnI181R, respectively (BamHI ⁄ NcoI sites and termination ⁄ initiation codons are indicated by underlines and bold type face, respectively). In RTnI96F the Asn96 of the template was replaced by Asp96, and an NcoI site was introduced. The PCR prod- ucts were used for the construction of expression plasmids by the method described above.

Expression and purification of recombinant TnI fragments

(8% (w/v) sucrose, 50 mm Tris ⁄ HCl

and ATnI292R,

termination codons (bold), except or

Based on the partial nucleotide sequence (GenBank acces- sion number AB009368), we cloned the cDNA including the entire coding region for Akazara scallop TnI by 5¢-RACE [37] from the striated adductor muscle. As a result, encoding isoforms, namely 52K-TnI and 19K-TnI [27], were obtained. The deduced amino acid sequence of 19K-TnI was identical to that of C-terminal 163 residues of 52K-TnI. The 52K-TnI-cDNA was subcloned into pCR2.1-TOPO (Invitrogen, Carlsbad, CA, USA), and used as a template for PCR to amplify the DNAs encoding various regions of 52K-TnI. For the amplification of the DNAs encoding ATnI-52K (recombin- ant 52K-TnI; residues 1–292), ATnI1)128 (recombinant frag- ment consisting of residues 1–128 of 52K-TnI), ATnI-19K (recombinant 19K-TnI; residues 130–292), ATnI130)252 (fragment; residues 130–252), and ATnI232)292 (fragment; residues 232–292), combinations of the forward and reverse primers, ATnI1F (5¢-CATATCACCATGGGTTCCCTTG-3¢) and ATnI292R (5¢-CTTGATTTGGATCCTTTAAGGTA TAGC-3¢), ATnI1F and ATnI128R (5¢-GTTCCGGATC CTATCTTCTGGCTTCC-3¢), ATnI130F (5¢-GCCAGAA CCATGGCGGAGGAAC-3¢) and ATnI292R, ATnI130F and ATnI252R (5¢-CAAGTTTGGGATCCTATTTGTTAA CTTTTC-3¢), and ATnI232F (5¢-CGAGATTAATGCC respectively, ATGGCACTTAAGG-3¢) were used. These forward and reverse primers introduced NcoI and BamHI restriction sites (underlined), respectively, into the PCR products. These primers also introduced the initiation in ATnI292R, which would anneal to the 3¢-noncoding region. It should be noted that in ATnI1F and ATnI232F, the Ser1 and Thr232 codons in the template were replaced by Gly1 and Ala232, respectively, in addition to introducing the NcoI site. The PCR products were digested with NcoI and BamHI and then ligated into the NcoI-BamHI site of the expression vector, pET-16b (Novagen, Madison, WI, USA).

The expression plasmids were introduced into E. coli BL21(DE3) cells (Novagen) and cultivated at 37 (cid:1)C for 9 h in LB medium, and then TnI fragments were expressed by induction with 1 mm IPTG. The cells were harvested by centrifugation (10 000 g, 10 min), and resuspended in STET buffer (pH 8.0), 50 mm EDTA, and 5% (v/v) Triton X-100), and then lysed by three freeze-thaw cycles. After centrifugation (10 000 g, 10 min), ATnI1)128, ATnI232)292, and RTnI96)181 were found in the supernatant, and purified by CM-Toyopearl 650 m (Tosoh, Tokyo, Japan) column chromatography in the presence of 6 m urea [34]. ATnI-52K, ATnI-19K, ATnI130)252, RTnI, and RTnI1)116, which were found in the precipitate, were dissolved in 7 m guanidine hydrochlo- ride, 10 mm Tris ⁄ HCl (pH 7.6), 1 mm EDTA, and 5 mm 2- mercaptoethanol, and then subjected to CM-Toyopeal col- umn chromatography as described above. ATnI-52K was further purified by DEAE-Toyopearl 650 m (Tosoh) col- umn chromatography under the conditions used for CM- Toyopeal chromatography. RTnI, RTnI1)116, and ATnI- 19K were also purified by hydroxyapatite (Wako Pure Chemicals, Osaka, Japan) column chromatography per- formed using 6 m urea, 10 mm KH2PO4 (pH 7.0), 5 mm 2- mercaptoethanol, and a linear gradient of 0–500 mm KCl. The N-terminal sequences of these recombinant proteins were analyzed on an ABI 492HT protein sequencer (Applied Biosystems, Foster City, CA, USA).

Polyacrylamide gel electrophoresis

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4483

SDS ⁄ PAGE was carried out using the method of Porzio and Pearson [38] on a 10% (w/v) acrylamide and 0.1% bis- acrylamide slab gel. Alkaline urea PAGE was performed by the method of Head and Perry [39] on a 6% (w/v) acryl- We also cloned the cDNA encoding rabbit fast skeletal TnI from the back muscle of rabbit by RT-PCR using the primer set, RTnI1F (5¢-CAAACCTCACCATGGGAGAT GAAG-3¢) and RTnI181R (5¢-CCCCGGAGCCGGATCC CCAGCCCC-3¢). These primers were designed based on the sequence retrieved from the GenBank database under accession number L04347, and NcoI or BamHI sites (under- lined) and the initiation codon (bolded) were introduced into the sequences. The cDNA subcloned into pCR2.1- TOPO was first subjected to mutagenesis for deactivating the native NcoI site in the coding region by using Mutan- (Takara-bio, Ohts, Japan). The Super Express Km kit

H. Tanaka et al.

Functional regions of molluskan TnI

(pH 8.9), 0.5% (w ⁄ v) SDS, and 5% (v ⁄ v) 2-mercaptoetha- nol, and then analyzed by SDS ⁄ PAGE. The amount of the TnI-fragment bound to actin-tropomyosin was estimated by densitometry, using known amounts of protein run on the same gel, as a standard. The amount of nonspecific precipitation of the TnI-fragment was also monitored by simultaneous centrifugation of the sample containing no actin-tropomyosin under the same conditions.

Reconstitution of troponins

amide and 0.48% (w/v) bis-acrylamide slab gel containing either 6 m or 3 m urea and either 2 mm CaCl2 or 2 mm EDTA. The samples were prepared as follows: TnI-frag- ment and TnC were mixed to a 1 : 1 molar ratio in the 10 mm Tris ⁄ HCl 0.125 m KCl, medium containing (pH 7.6), and either 5 mm CaCl2 or 5 mm EDTA, and then diluted with 1.5 volumes of either 10 or 5 m urea, 41.5 mm Tris, 133 mm glycine (pH 8.6), 0.02% (w/v) bromophenol blue, and 8% (v/v) 2-mercaptoetanol. The samples were allowed to stand for 2 h on ice before application to the gels. The electrophoresis was carried out at room tempera- ture by using 25 mm Tris and 80 mm glycine (pH 8.6) as a running buffer.

The gels were stained with 0.2% (w/v) Coomassie brilli- ant blue R250. Fluorescent staining using SYPRO Red (Cambrex, East Rutherford, NJ, USA) was also performed for densitometric analysis on a fluorescent imager, FLA- 3000G (Fuji Photo Film, Tokyo, Japan).

Affinity chromatography

Recombinant TnI-fragment and native TnC and TnT were mixed at a 1 : 1 : 1 molar ratio and dialyzed against 6 m (pH 7.6), and 5 mm urea, 0.5 m KCl, 10 mm Tris ⁄ HCl 2-mercaptoethanol. The urea and KCl concentrations were reduced stepwise by the following changes of dialysis buf- fer: (a) buffer B (3 m urea, 0.5 m KCl, 10 mm Tris maleate (pH 6.8), 2 mm MgCl2, 0.2 mm EGTA, 0.3 mm CaCl2, 0.01% NaN3 (w/v), and 5 mm 2-mercaptoethanol); (b) buf- fer B containing 1 m urea and 0.5 m KCl; (c) buffer B con- taining 0.5 m KCl; and (d) buffer B containing 0.25 m KCl. After dialysis, the complexes were centrifuged and the sup- ernatants were used immediately.

Measurements of Mg2+-ATPase activity

immobilized on Rabbit or Akazara scallop TnC was Formyl-Cellulofine (Chisso, Tokyo, Japan) according to the procedure suggested by the manufacturer. The TnC- Cellulofine was packed into a column (0.8 · 4.0 cm) and equilibrated with 10 mm Tris ⁄ HCl (pH 7.6) and 0.5 mm CaCl2. About 50 nmol of TnI-fragment was dialyzed against the same solution and then applied onto the col- umn. The fragment was eluted with a stepwise gradient of KCl at a flow rate of 0.16 mLÆmin)1. The fragment that was not eluted under these conditions was removed with 6 m urea, 0.5 m KCl, 10 mm Tris ⁄ HCl (pH 7.6), and 1 mm EGTA. The proteins in the effluents were detected by the method of Bradford [40], and identified by SDS ⁄ PAGE. RTnI1)116, which was insoluble in 10 mm Tris ⁄ HCl (pH 7.6) and 0.5 mm CaCl2, was applied at a KCl concentration of 0.1 m.

rabbit F-actin, (0.71 lm)

Actin-tropomyosin centrifugation studies

The inhibition of actomyosin-tropomyosin Mg2+-ATPase by the TnI-fragment and the release of the inhibition by TnC were measured in the presence of 0.05 mgÆmL)1 (1.2 lm) rabbit F-actin, 0.1 mgÆmL)1 (0.19 lm) rabbit myo- sin, 0.025 mgÆmL)1 (0.38 lm) rabbit or Akazara scallop tropomyosin, and various concentrations of TnI-fragment and TnC. The assays were performed at 15 (cid:1)C in a medium containing 50 mm KCl, 2 mm MgCl2, 20 mm Tris maleate (pH 6.8), 1 mm ATP, and 0.2 mm EGTA (in the absence of Ca2+) or 0.2 mm EGTA plus 0.3 mm CaCl2 (in the pres- ence of Ca2+). The Ca2+ regulatory effect of the recon- stituted troponin was measured in the presence of 0.03 mgÆmL)1 0.06 mgÆmL)1 (0.11 lm) rabbit myosin, 0.015 mgÆmL)1 (0.23 lm) rabbit or Akazara scallop tropomyosin, and 0.23 lm reconstituted troponin. The assays were performed at 15 or 25 (cid:1)C in a medium containing 50 mm KCl, 2 mm MgCl2, 20 mm Tris maleate (pH 6.8), 1 mm ATP, 0.1 mm CaCl2 and 0–3.84 mm EGTA. The concentrations of EGTA required to attain the desired final free Ca2+ concentrations (pCa 7.5–4.0) were calculated by using the stability constant of 8.45 · 105 m)1 for the Ca2+–EGTA complex [41].

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4484

The reaction was initiated by adding 0.5 mL of 10 mm ATP to 4.5 mL of the solution containing all the compo- nents except for ATP. After 2, 4, 6, and 8 min incubation, 1 mL aliquots were withdrawn from the reaction mixture and added to 4 mL of acidic malachite green solution to determine the liberated inorganic phosphate concentrations by the method of Chan et al. [42]. The binding of the TnI-fragment to actin-tropomyosin was analyzed by a cosedimentation assay. The assay conditions were as follows: 0.15 mgÆmL)1 (3.6 lm) rabbit F-actin, 0.075 mgÆmL)1 (1.1 lm) rabbit or Akazara scallop tropo- myosin, 2.2 lm recombinant TnI-fragment with or without equimolar amount of TnC, 50 mm KCl, 20 mm Tris maleate (pH 6.8), 2 mm MgCl2, and 0.2 mm EGTA (in the absence of Ca2+) or 0.2 mm EGTA plus 0.3 mm CaCl2 (in the pres- ence of Ca2+). The proteins were mixed in the presence of 0.3 m KCl and then diluted to the above conditions. The samples (0.5 mL) were incubated at 15 (cid:1)C for 30 min and then centrifuged at 100 000 g for 30 min on an Optima TL-100 ultracentrifuge (Beckman Coulter, Fullerton, CA, USA). The pellets and supernatants were redissolved in equivalent volumes (0.1 mL) of 5 m urea, 5 mm Tris ⁄ HCl

H. Tanaka et al.

Functional regions of molluskan TnI

Acknowledgements

This study was supported by Special Coordination from the Ministry of Education, Culture, Funds the Japanese Sports, Science and Technology, of Government.

interaction between troponin C and the N-terminal region of troponin I. J Cell Biochem 83, 99–110. 14 Tripet B, Van Eyk JE & Hodges RS (1997) Map- ping of a second actin-tropomyosin and a second troponin C binding site within the C terminus of tro- ponin I, and their importance in the Ca2+-dependent regulation of muscle contraction. J Mol Biol 271, 728–750.

References

1 Ebashi S & Kodama A (1965) A new protein factor promoting aggregation of tropomyosin. J Biochem (Tokyo) 58, 107–108.

15 Strynadka NC, Cherney M, Sielecki AR, Li MX, Smil- lie LB & James MN (1997) Structural details of a cal- cium-induced molecular switch: X-ray crystallographic analysis of the calcium-saturated N-terminal domain of troponin C at 1.75 A˚ resolution. J Mol Biol 273, 238– 255. 2 Ohtsuki I, Maruyama K & Ebashi S (1986) Regulatory and cytoskeletal proteins of vertebrate skeletal muscle. Adv Protein Chem 38, 1–67.

16 McKay RT, Tripet BP, Hodges RS & Sykes BD (1997) Interaction of the second binding region of troponin I with the regulatory domain of skeletal muscle troponin C as determined by NMR spectroscopy. J Biol Chem 272, 28494–28500. 17 Vassylyev DG, Takeda S, Wakatsuki S, Maeda K & 3 Zot AS & Potter JD (1987) Structural aspects of tropo- nin-tropomyosin regulation of skeletal muscle contrac- tion. Annu Rev Biophys Biophys Chem 16, 535–559. 4 Farah CS & Reinach FC (1995) The troponin complex and regulation of muscle contraction. FASEB J 9, 755– 767. 5 Sundaralingam M, Bergstrom R, Strasburg G, Rao ST, Maeda Y (1998) Crystal structure of troponin C in com- plex with troponin I fragment at 2.3-A˚ resolution. Proc Natl Acad Sci USA 95, 4847–4852.

Roychowdhury P, Greaser M & Wang BC (1985) Molecular structure of troponin C from chicken skeletal muscle at 3-angstrom resolution. Science 227, 945–948.

18 Farah CS, Miyamoto CA, Ramos CH, da Silva AC, Quaggio RB, Fujimori K, Smillie LB & Reinach FC (1994) Structural and regulatory functions of the NH2- and COOH-terminal regions of skeletal muscle troponin I. J Biol Chem 269, 5230–5240. 6 Herzberg O & James MN (1985) Structure of the cal- cium regulatory muscle protein troponin-C at 2.8 A˚ resolution. Nature 313, 653–659. 19 Takeda S, Yamashita A, Maeda K & Maeda Y (2003)

Structure of the core domain of human cardiac troponin in the Ca2+-saturated form. Nature 424, 35–41. 7 Kretsinger RH & Nockolds CE (1973) Carp muscle cal- cium-binding protein. II. Structure determination and general description. J Biol Chem 248, 3313–3326.

20 Vinogradova MV, Stone DB, Malanina GG, Karatzaferi C, Cooke R, Mendelson RA & Fletterick RJ (2005) Ca2+-regulated structural changes in troponin. Proc Natl Acad Sci USA 102, 5038–5043. 8 Potter JD & Gergely J (1975) The calcium and magne- sium binding sites on troponin and their role in the reg- ulation of myofibrillar adenosine triphosphatase. J Biol Chem 250, 4628–4633. 9 Leavis PC, Rosenfeld SS, Gergely J, Grabarek Z & 21 Ojima T & Nishita K (1986) Troponin from Akazara scallop striated adductor muscles. J Biol Chem 261, 16749–16754. 22 Ojima T, Tanaka H & Nishita K (1994) Cloning

Drabikowski W (1978) Proteolytic fragments of troponin C: localization of high and low affinity Ca2+ binding sites and interactions with troponin I and troponin T. J Biol Chem 253, 5452–5459. and sequence of a cDNA encoding Akazara scallop troponin C. Arch Biochem Biophys 311, 272–276. 23 Shiraishi F, Morimoto S, Nishita K, Ojima T &

10 Johnson JD, Charlton SC & Potter JD (1979) A fluores- cence stopped flow analysis of Ca2+ exchange with troponin C. J Biol Chem 254, 3497–3502. 11 Syska H, Wilkinson JM, Grand RJ & Perry SV (1976)

Ohtsuki I (1999) Effects of removal and reconstitution of myosin regulatory light chain and troponin C on the Ca2+-sensitive ATPase activity of myofibrils from scallop striated muscle. J Biochem (Tokyo) 126, 1020– 1024. The relationship between biological activity and primary structure of troponin I from white skeletal muscle of the rabbit. Biochem J 153, 375–387.

24 Ojima T, Koizumi N, Ueyama K, Inoue A & Nishita K (2000) Functional role of Ca2+-binding site IV of scal- lop troponin C. J Biochem (Tokyo) 128, 803–809. 25 Doi T, Satoh A, Tanaka H, Inoue A, Yumoto F,

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4485

12 Van Eyk JE & Hodges RS (1988) The biological impor- tance of each amino acid residue of the troponin I inhibitory sequence 104–115 in the interaction with troponin C and tropomyosin-actin. J Biol Chem 263, 1726–1732. 13 Ngai SM, Pearlstone JR, Smillie LB & Hodges RS (2001) Characterization of the biologically important Tanokura M, Ohtsuki I, Nishita K & Ojima T (2005) Functional importance of Ca2+-deficient N-terminal lobe of molluscan troponin C in troponin regulation. Arch Biochem Biophys 436, 83–90.

H. Tanaka et al.

Functional regions of molluskan TnI

35 Perry SV (1955) Myosin adenosine triphosphatase. In Methods in Enzymology (Colowick SP & Kaplan NO, eds), 2, pp. 582–588. Academic Press, New York. 36 Spudich JA & Watt S (1971) The regulation of rabbit skeletal muscle contraction. I. Biochemical studies of the interaction of the tropomyosin–troponin complex with actin and the proteolytic fragments of myosin. J Biol Chem 246, 4866–4871.

26 Tanaka H, Ojima T & Nishita K (1998) Amino acid sequence of troponin-I from Akazara scallop striated adductor muscle. J Biochem (Tokyo) 124, 304–310. 27 Ojima T & Nishita K (1991) A binary complex of tro- ponin-I and troponin-T from Akazara scallop striated adductor muscle. J Biochem (Tokyo) 110, 847–850. 28 Van Eyk JE, Thomas LT, Tripet B, Wiesner RJ, Pearl- stone JR, Farah CS, Reinach FC & Hodges RS (1997) Distinct regions of troponin I regulate Ca2+-dependent activation and Ca2+ sensitivity of the acto-S1-TM ATPase activity of the thin filament. J Biol Chem 272, 10529–10537. 37 Frohman MA, Dush MK & Martin GR (1988) Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci USA 85, 8998–9002.

29 Ramos CH (1999) Mapping subdomains in the C-ter- minal region of troponin I involved in its binding to troponin C and to thin filament. J Biol Chem 274, 18189–18195. 38 Porzio MA & Pearson AM (1977) Improved resolution of myofibrillar proteins with sodium dodecyl sulphate– polyacrylamide gel electrophoresis. Biochim Biophys Acta 490, 27–34.

39 Head JF & Perry SV (1974) The interaction of the cal- cium-binding protein (troponin C) with bivalent cations and the inhibitory protein (troponin I). Biochem J 137, 145–154.

40 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein util- izing the principle of protein-dye binding. Anal Biochem 72, 248–254.

30 Malnic B, Farah CS & Reinach FC (1998) Regulatory properties of the NH2- and COOH-terminal domains of troponin T. ATPase activation and binding to tropo- nin I and troponin C. J Biol Chem 273, 10594–10601. 31 Pearlstone JR & Smillie LB (1995) Evidence for two-site binding of troponin I inhibitory peptides to the N and C domains of troponin C. Biochemistry 34, 6932–6940. 32 Ohtsuki I, Morimoto S & Takahashi-Yanaga F (2003) Several aspects of calcium regulator mechanisms linked to troponin. Adv Exp Med Biol 538, 221–229. 33 Nishimura Y, Ojima T & Nishita K (1997) Bivalve

41 Harafuji H & Ogawa Y (1980) Re-examination of the apparent binding constant of ethylene glycol bis (beta-aminoethyl ether)-N,N,N’,N’-tetraacetic acid with calcium around neutral pH. J Biochem (Tokyo) 87, 1305–1312. tropomyosins exhibit strong inhibition of actomyosin Mg-ATPase and high viscosity. Fish Sci 63, 802–806. 34 Ojima T & Nishita K (1988) Separation of Akazara 42 Chan KM, Delfert D & Junger KD (1986) A direct

FEBS Journal 272 (2005) 4475–4486 ª 2005 FEBS

4486

colorimetric assay for Ca2+-stimulated ATPase activity. Anal Biochem 157, 375–380. scallop and rabbit troponin components by a single-step chromatography on CM-Toyopearl. J Biochem (Tokyo) 104, 9–11.