Calcium-dependent protein–protein interactions induce changes in proximity relationships of Cys48 and Cys64 in chicken skeletal troponin I Ying-Ming Liou and Ming-Wei Chen Department of Life Science, Institute of Biochemistry, National Chung-Hsing University, Taichung, Taiwan The goal of this study was to relate conformational changes in the N-terminal domain of chicken troponin I (TnI) to Ca 2+ activation of the actin–myosin interaction. The two cysteine residues in this region (Cys48 and Cys64) were labeled with two sulfhydryl-reactive pyrene-containing fluorophores [N-(1-pyrene)maleimide, and N-(1-pyrene)- iodoacetamide]. The labeled TnI showed a typical fluores- cence spectrum: two sharp peaks of monomer fluorescence and a broad peak of excimer fluorescence arising from the formation of an excited dimer (excimer). Results obtained show that forming a binary complex of labeled TnI with skeletal TnC (sTnC) in the absence of Ca 2+ decreases the excimer fluorescence, indicating a separation of the two residues. This reduction in excimer fluorescence does not occur when labeled TnI is complexed with cardiac TnC (cTnC). The latter causes only partial activation of the Ca 2+ -dependent myofibrillar ATPase. The binding of Ca 2+ to the two N-terminal sites of sTnC causes a significant de- crease in excimer fluorescence and an increase in monomer fluorescence in complexes of labeled TnI with skeletal TnC or TnC/TnT, while Ca 2+ binding to site II of cTnC only causes an increase in monomer fluorescence but no change in excimer fluorescence. Thus a conformational change in the N-terminal region of TnI may be necessary for full activation of muscle contraction. Keywords: proximity relationships; cysteine residues; pyrene-containing fluorophores; monomer fluorescence; excimer fluorescence. Ca 2+ activation of striated muscle is mediated by the troponin complex. Troponin is composed of three subunits: troponin C (TnC), a Ca 2+ binding subunit; troponin I (TnI), an inhibitory subunit and troponin T (TnT), a subunit which binds the complex to tropomyosin. Based on several experimental approaches, such as crystallography [1–3], NMR [4–6], neutron scattering [7,8], chemical cross- linking [9–11] and fluorescence resonance energy transfer [12,13], a structural model of the TnC–TnI complex was proposed in which TnI winds around TnC in either a left- handed manner (Model ÔLÕ) or a right-handed manner (Model ÔRÕ) [14]. According to these models, five adjoining segments of TnI, segment I–V, correspond to residues 3–33, 34–53, 54–94, 95–114 and 115–134, respectively. Segment I is an a-helix that binds to the C-terminal hydrophobic cleft of TnC. Segment II does not posses a regular secondary structure [3,7], and is postulated to act as a Ôflexible tetherÕ between segment I and III. Segment III, also an a-helix, is thought to interact with TnT [15,16]. Segment IV, corres- ponding to the so-called inhibitory region of TnI [17], adopts a b-hairpin structure and binds to the central helix of TnC upon Ca 2+ activation. This fragment can fully inhibit actin–myosin ATPase activity and bind alternatively to actin and TnC during the contractile cycle [18]. Segment V, containing an a-helical structure, binds to the hydrophobic cleft of the N-terminal domain of TnC. There are two cysteine residues (Cys48 and Cys64) in the N-terminal end of TnI from both chicken and rabbit fast skeletal muscle and a third cysteine (Cys133) in rabbit TnI [19]. Cys48 is located at segment II, while Cys64 at segment III. Compounds that contain the pyrene group, such as N-(1-pyrene)-maleimide (PM) and N-(1-pyrenyl)- iodoacetamide (PIA) were shown to be suitable for the study of proximity relations between labeled cysteines [20,21]. Strasburg et al. [22] first used PM to label the two N-terminal cysteine residues of rabbit fast skeletal TnI. The fluorescence spectrum of pyrene-labeled TnI exhibits peaks characteristic of pyrene in its monomeric form as well as an additional red-shifted peak due to the formation of an excimer. This spectrum was compared to fluorescence spectra obtained for the TnI–TnC complex, with and without calcium. Conformational changes that occurred during calcium binding to the TnI–TnC complex, were monitored as a change in excimer fluorescence. Thus, the calcium-induced structural changes that took place in the TnI–TnC-binding region upon calcium binding might well be transmitted to TnT as part of the events regulating muscle contraction. However, the rabbit TnI has an additional Cys133. In the experiments of Strasburg et al. [22], this was blocked with iodoacetamide, thereby Correspondence to Y M. Liou, Department of Life Science, National Chung-Hsing University, 250 Kuokang Road, Taichung 402, Taiwan. Fax: + 886 4 22851797, Tel.: + 886 4 22851802, E-mail: ymlion@dragon.nchu.edu.tw Abbreviations: TnI, troponin I; TM, tropomyosin; TnC, troponin C; TnT, troponin T; PM, N-(1-pyrene)-maleimide; PIA, N-(1-pyrenyl)-iodoacetamide; BCA, bicinchoninic acid. (Received 11 March 2003, revised 18 May 2003, accepted 2 June 2003) Eur. J. Biochem. 270, 3092–3100 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03700.x introducing a possible confounding factor not present with chicken TnI. In this study we have labeled the two N- terminal cysteine residues of chicken pectoral muscle TnI with either PM or PIA. We have used this preparation to further characterize the Ca 2+ -induced proximity changes of Cys48 and Cys64 in TnI during Ca 2+ binding to the ternary complex of TnI–TnC–TnT. In addition, the amino acid sequences of skeletal and cardiac TnC and TnI are different [19,23–26], in particular, at their binding interfaces. These differences might alter the TnI–TnC interaction and thus affect muscle regulation. Chicken fast skeletal TnC is a polypeptide of 159 amino acid residues that contains four Ca 2+ binding sites [24,25]. The two C-terminal sites (III and IV) of TnC bind Ca 2+ with high affinity (K % 10 7 M )1 )andalsoMg 2+ (K % 10 3 M )1 ). Occupation of these sites facilitates the binding of TnC to the thin filament. The other two N-terminal sites (I and II) bind Ca 2+ exclusively but with lower affinity (K % 10 5 M )1 ). Binding of Ca 2+ to the N-terminal sites is responsible for activation of actomyosin ATPase activity. Due to some amino acid substitutions at Ca 2+ binding loop I [23], porcine cardiac TnC binds 3 mol Ca 2+ per mol protein [26]. In this study, we also compared the effects of sTnC and cTnC to test if an isoform-specific interaction of TnI and TnC is required for the Ca 2+ - induced proximity change in the two N-terminal cysteines of TnI. Materials and methods Reagents Unless otherwise specified, all reagents used were ACS grade. N-(1-pyrene)maleimide (PM) and N-(1-pyrene)iodo- acetamide (PIA) were purchased from Molecular Probes. Chromatographic reagents (DEAE-sephadex A-50 and CM-sephadex C-50) were purchased from Amersham Pharmacia Biotech Asia Pacific. Bicinchoninic acid (BCA) protein assay reagent was from Pierce Chemicals. Protein preparation and modification Troponin subunits were isolated from an ether powder prepared from chicken pectoral muscle according to the method of Potter [27]. CTnC was prepared from the left ventricle of porcine hearts according to Szynkiewicz et al. [28]. Purified proteins were stored freeze-dried in a freezer ()80 °C). Before use, the freeze-dried TnI was dissolved in a solution containing 8 M urea, 3 m M dithiothreitol, 100 m M Mops (pH 7.0), 2 m M EDTA and 100 m M KCl to reduce the two thiol groups at room temperature for 4 h. Follow- ing exhaustive dialysis at 4 °C, the reduced TnI was reacted at room temperature for 4 h with a two- to threefold molar excess of PM, or PIA, following the method of Liou and Fuchs [29]. The reaction was terminated with excess dithiothreitol. The latter was removed by solvent exchange with Centricon 10 ultrafiltration cells (Amicon). The pro- teins were then dialyzed against 6 M urea, 25 m M Mops (pH 7.0) and 0.25 M KCl at 4 °C. The dialysis against the same buffer without urea was repeated twice. For the PM- labeled TnI, it was further treated with 8 M urea in alkaline solution (25 m M Tris, pH 8.8) for 24 h at room temperature to cleave the succinimido ring [20]. The protein was then dialyzed against 10 m M Mops, pH 7.0 and 0.25 M KCl. The amount of bound pyrene was determined on the basis of extinction coefficients of 23 000 M )1 Æcm )1 (at 345 nm) and 28 000 M )1 Æcm )1 (at 344 nm) for PM-labeled [22] and PIA- labeled TnI [21], respectively. Concentrations of labeled proteins were measured with the BCA protein assay reagent [29]. The labeling ratio of pyrene to TnI was 1.8–2.0 mol fluorophore per mol protein. Reconstitution of the binary or ternary troponin com- plexes was performed by mixing equimolar amounts of each subunit in denaturing solution containing 6 M urea, 1 M KCl, 50 m M Mops (pH 7.0), 5 m M CaCl 2 ,and5m M dithiothreitol, followed by renaturation by dialysis to remove urea and to lower the salt concentration. The reconstituted protein complex in a buffer solution contain- ing 10 m M Mops (pH 7.0), 1 m M MgCl 2 ,1 m M EGTA, and 0.25 M KCl was stored at )80 °Cbeforeuse. ATPase activity assay The biological activities of the labeled proteins were assayed by determining Ca 2+ -activated myofibrillar ATPase acti- vity. Chicken skeletal myofibrils were prepared at 4 °C according to Liou et al. [30]. The endogenous TnC and TnI were extracted by incubating the myofibrils with purified TnT (myofibril: TnT ¼ 2 : 1, w/w) in a solution containing 100 m M Mops (pH 7.0), 90 m M KCl, 5 m M MgCl 2 ,2m M EGTA and 1 m M dithiothreitol, at room temperature for 1 h, as described by Shiraishi et al.[31].Thedegreeof extraction was compared by gel electrophoresis and by the loss of Ca 2+ -activated ATPase activity, as described previously [30]. Unlabeled or pyrene-labeled TnI, in complex with TnC, were mixed with the TnT-treated myofibrils in a ratio of 0.25 mg TnI or TnC per milligram myofibrillar protein. The myofibrils were suspended in 100 m M Mops (pH 7.0), 90 m M KCl, 5 m M MgCl 2 and 2 m M EGTA. After incubation at room temperature for 1 h, the myofibrils were centrifuged at 2600 g to remove free TnI or TnC. The pellet was suspended in the same buffer solution. The actomyosin ATPase activity was measured by suspending the intact, extracted and reconstituted myofibrils (0.4 mgÆmL )1 )in100m M Mops (pH 7.0), 90 m M KCl, 5m M MgCl 2 ,2m M EGTA and various additions as indicated. The reaction mixtures were shaken in a water bath at controlled temperature (30 °C) for 10 min. The reaction was initiated by the addition of MgATP to a concentration of 1 m M and terminated after 10 min by the addition of malachite green reagent (33% malachite green, 16.7% polyvinyl alcohol, and 16.7% ammonium molyb- date). The mixture was then analysed for inorganic phos- phate release [30]. Fluorescence measurements Pyrene-labeled TnI (1–2 l M ), either alone or complexed with other troponin subunits, was dissolved in 100 m M Mops (pH 7.0), 0.25 m M KCl and 2 m M EGTA, with additions as indicated. Corrected fluorescence spectra were recorded with a Perkin-Elmer LS 50 B spectrofluorimeter Ó FEBS 2003 Conformational changes in chicken sTnI (Eur. J. Biochem. 270) 3093 (Beaconsfield, Buckinghamshire, England) at a constant temperature of 25 °C. The solution was excited at 345 nm and the fluorescence emission was scanned from 370– 570 nm, with slit widths of 3 and 10 nm for excitation and emission, respectively. Ca 2+ titration Pyrene-labeled protein solutions were placed in a semimicro cell (1 mL) and Ca 2+ titrations were carried out using a digital micropipette using aliquots of a commercial standard solution (Radiometer) of 0.1 M CaCl 2 . Total volume change with Ca 2+ addition did not exceed 2%. The emitted fluorescence at 385 and 485 nm and the relative ratio of fluorescence emitted at 485 nm to that at 385 nm were taken for data analysis on the Ca 2+ dependent fluorescence changes in pyrene-labeled TnI in I–C complex or in a whole Tn complex. The measured fluorescence at the actual pCa (–log[Ca 2+ ]) was subtracted from that at pCa 8. The subtracted fluorescence (F x ) was normalized to the value (F 0 ) at saturating pCa. If normalized fluorescence (U ¼ F x /F 0 ) is used, then a straight line is obtained with the expression of log[U/(1–U)] vs. the logarithm of the Ca 2+ concentration. This plot was fitted using the Hill equation: log½U=ð1 À UÞ ¼ nðlog½Ca x Þ þ log k where, [Ca x ]istheactualCa 2+ concentration, n (Hill coefficient) is the slope, and k is the x-axis intercept of the fitted line. The Hill coefficient is a measure of cooperativity for the Ca 2+ -induced fluorescence changes in pyrene- labeled TnI complexed with TnC or TnC/TnT. By using the constants derived from the Hill equation, the curves of the normalized fluorescence changes (F x /F 0 )vs.pCawere fitted by computer with the equation: ðF x =F 0 Þ¼½Ca x  n =fðEC 50 Þ n þ½Ca x  n g where EC 50 is the Ca 2+ concentration giving 50% activa- tion of fluorescence changes. Free Ca 2+ concentrations in EGTA buffers were calcu- lated on the basis of constants tabulated by Fabiato and Fabiato [32]. The pCa values were calculated by the computer program EQCAL (Biosoft, Cambridge, UK). Statistics Quantitative values are expressed as mean ± SEM. Com- parisons of statistics were performed by a Student’s t-test where P-values less than 0.05 were considered as being significant [33]. Results Biological activity of pyrene-labeled TnI In interpreting the experimental results it is important to know whether the modified protein retains biological activity. In this case, biological activity is expressed in terms of the ability of the modified TnI to modulate Ca 2+ regulation of myofibrillar ATPase activity. Following extraction of TnC and TnI from skeletal myofibrils, by adding exogenous TnT to replace the endogenous Tn in skeletal myofibrils, either TnC with the native and modified TnI were inserted into the myofibrils as described in Method section. The results are shown in Table 1. The ATPase activity of unextracted myofibrils in the absence of Ca 2+ was 21% of that in the presence of Ca 2+ . As expected, the myofibrillar ATPase activity showed no Ca 2+ dependence following extraction of TnC and TnI. Reinsertion of the modified TnI (PM–TnI and PIA–TnI) into extracted myofibrils restored Ca 2+ -activated ATPase activity as effectively as native TnI, as shown in Table 1. Pyrene-labeled TnI in complex with TnT and TnC When pyrene-labeled TnI formed a binary complex with TnT or TnC, the ratio of excimer fluorescence to monomer fluorescence (E/M ratio) was significantly reduced. A reduction (% 5%) was observed when TnT was complexed with either the PM-labeled or PIA-labeled TnI. As shown in Fig. 1A,C, and Fig. 2, this reduction for the complex of pyrene-labeled TnI with TnT results from significant increases in monomer fluorescence (5–10%) but no appar- ent changes in excimer fluorescence. This result suggests that TnT binding to TnI affects the local environment of the two labeled cysteine residues but does not cause significant changes in distance between these two sites. In contrast, greater reductions in the E/M ratio were observed when TnC was complexed with PIA–TnC (% 20%) and PM–TnI (% 15%) as the results of an increase in monomer fluores- cence (5–10%) and a decrease in excimer fluorescence (10–15%) (Figs 1B,D and 2). This finding suggests that the Cys48–64 distance is increased in the TnI–TnC complex. The formation of a ternary complex with TnT and TnC was also associated with an increase in monomer fluorescence and a decrease in excimer fluorescence (Figs 1 and 2). These data suggest that TnI contains different interaction sites for TnC and TnT and that interactions among these subunits can cause conformational changes in the N-terminal domain of TnI. Metal binding When the pyrene-labeled TnI formed a binary complex with TnC or a ternary complex with TnC and TnT, addition of 5 m M Mg 2+ did not cause significant changes in the fluorescence emission spectrum (Fig. 3A), whereas, addition of Ca 2+ (pCa 4) elicited an increase in monomer fluores- cence (5–10%) and a decrease in excimer fluorescence (10–45%) (Fig. 3B,C,D). Results obtained with Ca 2+ titration to pyrene-labeled TnI in binary complex with TnC or ternary complex with TnC and TnT are shown in Fig. 4. It appears that the binding of Ca 2+ to the two low affinity sites of TnC in a binary complex (pK Ca % 6.61) or in a ternary complex (pK Ca % 6.59) would account for the proximity changes of the two cysteines in TnI. Ca 2+ effect on pyrene-labeled TnI complex with cardiac troponin C Morimoto & Ohtsuki [34] and Putkey et al. [35] have shown that cTnC, when substituted into fast skeletal myofibrils, was only partially effective in activating Ca 2+ -dependent ATPase activity. In agreement with these reports, the 3094 Y M. Liou and M W. Chen (Eur. J. Biochem. 270) Ó FEBS 2003 present study shows that reinsertion of cardiac TnC with modified TnI (PIA–TnI) into extracted myofibrils only partially restored Ca 2+ -activated ATPase activity (Table 1). The degree of reinsertion of pyrene-labeled TnI in complex with both sTnC and cTnC into TnT-treated myofibrils was compared by running the SDS/PAGE. Densitometric analyses (Fig. 5A) show that both sTnC and cTnC in complex with labeled TnI have the same degree of re-incorporation (% 90%) into TnT-treated skeletal myo- fibrils. Experiments were then carried out to determine if the binding of cTnC to TnI altered the proximity between Cys48–64 of TnI. In contrast to the results obtained with the PIA-TnI/sTnC complex (Figs 1B and 2A), the Table 1. ATPase activities of control, TnT-treated and reconstituted myofibrils. Myofibrils (0.2–0.4 mgÆmL )1 ) were dissolved in 100 m M Mops (pH 7.0) 90 m M KCl, 5 m M MgCl 2 ,2m M EGTA and Ca 2+ concentration as indicated. The ATPase assay was described in Method section. Values are expressed as nmol P i Æmg proteinÆmin )1 . Each value is the mean ± SEM, with number of measurements (n), as indicated. Myofibrils Measurements (n) Calcium concentration pCa 8 pCa 4 Control 9 63.29 ± 3.83 286.37 ± 22.0 TnT-Treated 9 107.46 ± 17.63 110.42 ± 15.7 TnT-Treated + TnI + TnC 6 66.43 ± 5.23 303.83 ± 11.32 TnT-Treated + PIA-TnI + TnC 6 64.8 ± 3.23 277.25 ± 14.01 TnT-Treated + PM-TnI + TnC 6 59.07 ± 2.22 301.18 ± 1.73 TnT-Treated + PIA-TnI + cTnC 3 54.72 ± 2.86 102.35 ± 12.53 Fig. 1. Effect of TnT (A, C) and TnC (B, D) on the fluorescence emis- sion spectrum of PIA-labeled (A, B), and PM-labeled TnI (C, D). Pyrene fluorescence emission spectra were recorded from solutions of 1 l M pyrene-labeled TnI, and 1 l M pyrene-labeled TnI in complex with 3 l M TnC or/and 3 l M TnT. Solution composition is 100 m M Mops (pH 7.0), 0.25 M KCl and 2 m M EGTA. Fig. 2. Changes in monomer and excimer fluorescence of (A) PIA- labeled and (B) PM-labeled TnI incorporated into the complexes with TnT and TnC. Fluorescence changes were calculated by subtracting the fluorescence of monomer (at 385 nm) and excimer for the binary or ternary complex from each intensity for the labeled TnI alone, and then normalizing the difference of each to the value for the labeled TnI alone. Solution composition is as in Fig. 2. Each histogram shows mean (± SEM) of five measurements. Ó FEBS 2003 Conformational changes in chicken sTnI (Eur. J. Biochem. 270) 3095 formation of a binary complex of PIA-TnI with cTnC (in the absence of Ca 2+ )causedanincrease(% 12%) in monomer fluorescence but no change in excimer fluores- cence (Fig. 5B,C). Upon the addition of Ca 2+ there was a further increase (% 25%) in monomer fluorescence but no significant change in excimer fluorescence (Fig. 5B,C,D). The half maximal increase in monomer fluorescence occurred at pCa % 6.11, as compared to % 6.34 for sTnC complexed with PIA-labeled TnI (Fig. 4D). Thus, both with and without Ca 2+ , sTnC and cTnC elicit different con- formational changes in the N-terminal region of TnI. Discussion According to the model of the TnC–TnI complex proposed by Tung et al. [14], the predicted Cys48–64 distances are approximately 26.8 A ˚ and 21.3 A ˚ for the L and R models, respectively. For PM- and PIA-labeled TnI, the estimated pyrene-SH bond length is approximately 11.7 and 11.5 A ˚ , respectively. Apparently, the two pyrene labels at Cys48 and 64 have no spatial constraints to the formation of the excimer. When pyrene-labeled TnI forms a binary complex with TnC in the absence of Ca 2+ , excimer fluorescence is decreased and monomer fluorescence is increased (Figs 1 and 2). This reduction in excimer fluorescence indicates separation between Cys48 and 64 caused by the formation of a binary complex with TnC. This result is consistent with the notion that the antiparallel arrangement of TnI and TnC would lead to a more elongated configuration of the troponin complex [36–38]. There are multiple interactions between TnI and TnC, involving regions located in the N-terminal hydrophobic cleft (residues 51–62), central helix (residues 89–100), and the C-terminal hydrophobic cleft (residues 127–138) of TnC and regions located in the N-terminal part (residues 1–47) and the C-terminal part (residues 96–117 and 118–129) of TnI [3,13,17,39,40]. However, Strasburg et al. [22] observed that formation of a complex between pyrene-labeled TnI and TnC in the absence of Ca 2+ had little effect on the spectrum. It is possible that the reaction of iodoacetamide with Cys133 of rabbit TnI altered the TnI–TnC interaction in the absence of Ca 2+ . Addition of TnT to pyrene-labeled TnI caused a small increase in monomer fluorescence but no apparent change in excimer fluorescence (Figs 1 and 2). This small change in fluorescence emission might be a consequence of the bulky pyrene groups causing some interference with the binding of TnT to TnI [22]. There is also evidence that TnI segment 53–106 that has properties of a heptad repeat, may bind to a similar heptad repeat region on TnT segment 205–255 in a coiled–coil interaction [41,42]. It has been suggested that the N-terminal region next to the inhibitory segment (96–116) of TnI does not play a regulatory role in the activation of contraction. Rather, it plays a structural role by anchoring the troponin complex to the thin filament [38]. Studies with a TnI N-terminal Fig. 3. Effect of divalent cation binding on fluorescence emission spectra of pyrene-labeled TnI in a binary complex with TnC or in a ternary complex with TnC and TnT. (A) Effect of Mg 2+ on spectrum of PIA-labeled TnI–TnC–TnT. Solutions contained 1 l M PIA-labeled TnI, 3 l M TnC, 3 l M TnT, 100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, with or without 5 m M Mg 2+ . (B) Effect of Ca 2+ on spectrum of PIA-labeled TnI–TnC. Solutions contained 1 l M PIA-labeled TnI, 3 l M TnC, 100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, with or without 2.1 m M Ca 2+ . (C) Effect of Ca 2+ on spectrum of PIA-labeled TnI–TnC–TnT. Solutions contained 1 l M PIA-labeled TnI, 3 l M TnC, 3 l M TnT, 100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, with or without 2.1 m M Ca 2+ . (D) Effect of Ca 2+ binding on monomer (at 385 nm), excimer (at 485 nm) fluorescence peaks, and the ratio of excimer fluorescence to monomer fluorescence (E/M ratio) for PM-labeled TnI in a binary complex with TnC or in a ternary complex with TnC and TnT. Solutions contained 1 l M PM-labeled TnI, 3 l M TnC or 3 l M TnT, 100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, with or without 2.1 m M Ca 2+ . 3096 Y M. Liou and M W. Chen (Eur. J. Biochem. 270) Ó FEBS 2003 deletion mutant (TnI 58)182 ) showed that absence of the N-terminal residues 1–57 caused a modified interaction with TnT and loss of the ability to form a ternary complex with TnT and TnC [43]. However, when TnI 58)182 was incorporated into the troponin complex there was a Ca 2+ - dependence of actomyosin ATPase activity that was essen- tiallythesameasthatseeninthepresenceofintactTnI[44]. As shown in this study, the binding of Ca 2+ to the regulatory sites of TnC clearly affects the relative proximity between the two cysteine residues in the N-terminal region of TnI, both in the TnC–TnI and TnC–TnI–TnT complexes (Fig. 4). The same conclusion was also drawn in earlier work with rabbit muscle proteins [22]. This conformational change in the N-terminal region of TnI may be attributed to the Ca 2+ -induced flexibility increase in either TnI–TnC complex [45] or in a whole troponin complex [46]. Although evidence has been presented that the N-terminus of TnI interacts with the C-terminal end of TnT in a Ca 2+ - independent manner [11,36], the Ca 2+ -induced proximity changes that occur in the N-terminal region of TnI may also contribute to the Ca 2+ -activation of muscle contraction. Studies with recombinant TnT and TnI fragments by Malnic et al. [47] showed that the N-terminal region of TnI together with TnC was stabilized in the troponin complex through a Ca 2+ -independent interaction with the C-termi- nus of TnT. Their studies also indicated that this interaction is essential for full Ca 2+ -activation by the ternary complex. Recently, Rarick et al. [48] have addressed the same question with respect to the importance of the N-terminus of cardiac TnI in the activation and regulation of cardiac myofilaments. They prepared a wild-type mouse cardiac TnI, and two N-terminal deletion mutants: cTnI 54)211 (deleting the binding site for the C-terminal hydrophobic cleft of cTnC), cTnI 80)211 (deleting the binding site for the cTnT). Their data provided evidence in support of a significant function of a cTnT-binding domain in cTnI, and they suggested that cTnT binding to the N-terminus of cTnI is a negative regulator of cardiac activation. More recently, Ngai et al. have characterized a minimum sequence of the N-terminus of TnI (residues 1–30) that retains full biological activity to bind the C-terminal domain of TnC [49], and showed that the N- and C-terminal domains of TnI act as a Ca 2+ -dependent switch for TnC [50]. Thus, the N-terminal region of TnI may participate in the calcium activation of muscle contraction in cardiac and skeletal muscle. Earlier work by Morimoto and Ohtsuki [34] examined the effect of cTnC on the pCa-ATPase activity in CDTA- replaced skeletal myofibrils. Their results showed that cTnC is only partially effective in restoring ATPase activity to the CDTA-treated skeletal myofibrils. Using recombinant cTnC (wild type) and mutants (Site II inactive, Site I active, and Site I active but Site II inactive), Putkey et al. [35] found that substitution with all of these cTnCs can only partially restore the calcium-dependent myofibrillar ATPase Fig. 4. Ca 2+ + -dependent fluorescence changes of pyrene-labeled TnI in the binary complex with TnC or in the ternary complex. (A) PM-labeled TnI– TnC complex and (B) PM-labeled TnI–TnC–TnT complex show normalized monomer and excimer fluorescence as a function of pCa. Relative fluorescence of monomer (at 385 nm) and excimer (at 485 nm) at each pCa was obtained by subtracting the fluorescence measured at each pCa from that at pCa 8, and then normalizing the value to the difference at pCa 4. The curve was fitted to the data using the Hill equation as described in Materials and methods. Hill coefficients (n) and the pCa values for giving 50% fluorescence change (pK Ca ) are indicated. Solutions contained 1 l M PM-labeled TnI, 3 l M TnC, or 3 l M TnT, in 100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, 5 m M Mg 2+ , and varying pCa. (C) Pyrene fluorescence of PM-labeled TnI–TnC and PM-labeled TnI–TnC–TnT vs. pCa. 1 l M PM-labeled TnI formed a binary complex with 3 l M TnC or a ternary complex with 3 l M TnC and TnT in 100 m M Mops buffer. Pyrene fluorescence was expressed as the ratio of monomer (at 385 nm) to excimer (at 485 nm) fluorescence peaks. (D) Ca 2+ titration on monomer (at 385 nm), excimer (at 485 nm) fluorescence peaks, and the ratio of excimer fluorescence to monomer fluorescence (E/M ratio) for PIA-labeled TnI–TnC. 1 l M PIA-labeled TnI formed a binary complex with 3 l M TnC in100 m M Mops (pH 7.0), 0.25 M KCl, 2 m M EGTA, 5 m M Mg 2+ , and varying pCa. Each point is mean ± SEM of five measurements. Ó FEBS 2003 Conformational changes in chicken sTnI (Eur. J. Biochem. 270) 3097 activity for sTnC-extracted skeletal myofibrils. In this study, when cTnC was substituted for sTnC, no excimer fluores- cence was found in the complex with pyrene-labeled TnI (Fig. 5). The binding of Ca 2+ to cTnC (pK Ca ¼ 6.11) only caused the monomer fluorescence to increase but had no changes in excimer (Fig. 5C,D). The inability of cTnC to efffect a change in Cys48–64 distance suggests that a conformational change in the N-terminal region of TnI is essential for full activation of skeletal muscle contraction. We are currently carrying out more experiments with TnI fragments and segment-deletion TnI mutants to test if the TnI C-terminal end is required to mediate this Ca 2+ - induced structural change in the N-terminal domain of TnI. Future studies will also relate such changes to the functional roles of TnI in skeletal muscle contraction. References 1.Houdusse,A.,Love,M.L.,Dominguez,R.,Grabarek,Z.& Cohen, C. (1997) Structures of four Ca 2+ -bound troponin C at 2.0 A ˚ resolution: further insights into the Ca 2+ -switch in the cal- modulin superfamily. Structure 5, 1695–1711. 2. Strynadka, N.C.J., Cherney, M., Sielecki, A.R., Li, M.X., Smillie, L.B. & James, M.N.G. (1997) Structural details of a calcium- 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. 3. Vassylyev, D.G., Takeda, S., Wakatsuki, S., Maeda, K. & Maeda, Y. (1998) Crystal structure of troponin C in complex with tro- ponin I fragment at 2.3-A ˚ resolution. Proc. Natl Acad. Sci. USA 95, 4847–4852. 4. McKay, R.T., Tripet, B.P., Hodges, R.S. & Sykes, B.D. (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. 5.McKay,R.T.,Pearlstone,J.R.,Corson,D.C.,Gagne,S.M., Smillie, L.B. & Sykes, B.D. (1998) Structure and interaction site of the regulatory domain of troponin-C when complexed with the 96–148 region of troponin-I. Biochemistry 37, 12419–12430. 6. Mercier, P., Li, M.X. & Sykes, B.D. (2000) Role of the structural domain of troponin C in muscle regulation: NMR studies of Ca 2+ binding and subsequent interactions with regions 1–40 and 96–115 of troponin I. Biochemistry 39, 2902–2911. 7. Olah, G.A. & Trewhella, J. (1994) A model structure of the muscle protein complex 4Ca 2+ ÆTroponin CÆTroponin I derived from small-angle scattering data: Implications for regulation. Bio- chemistry 33, 12800–12806. 8. Stone, D.B., Timmins, P.A., Schneider, D.K., Krylova, I., Ramos, C.H.I.,Reinach,F.C.&Mendelson,R.A.(1998)Theeffectof regulatory Ca 2+ on the in situ structures of troponin C and troponin I: a neutron scattering study. J. Mol. Biol. 281, 689–704. 9. Kobayashi, T., Tao, T., Grabarek, Z., Gergely, J. & Collins, J.H. (1991) Cross-linking of residue 57 in the regulatory domain of a mutant rabbit skeletal troponin C to the inhibitory region of troponin I. J. Biol. Chem. 266, 13746–13751. 10. Luo,Y.,Leszyk,J.,Li,B.,Gergely,J.&Tao,T.(2000)Proximity relationships between residue 6 of troponin I and residues in tro- ponin C: Further evidence for extended conformation of troponin C in the troponin complex. Biochemistry 39, 15306–15315. 11. Luo,Y.,Wu,J.L.,Li,B.,Langsetmo,K.,Gergely,J.&Tao,T. (2000) Photocrosslinking of benzophenone-labeled single cysteine troponin I mutants to other thin filament proteins. J. Mol. Biol. 296, 899–910. Fig. 5. Effect of Ca 2+ + on pyrene fluorescence of PIA-labeled TnI in a complex with cardiac troponin C. (A) Quantitative analysis of reinsertion of PIA-labeled TnI in complex with sTnC and cTnC into TnT-treated skeletal myofibrils. SDS/PAGE and densitometic analysis were as described previously [30]. Myofibril-bound TnI, TnT, sTnC and cTnC were estimated by normalizing the ratio of the area under the peaks of TnI, or TnT, or sTnC or cTnC to that under the LC1 (myosin light chain 1) peak. Each column is shown as the mean ± SEM of three different gel scans. (B) Fluorescence emission spectra of PIA–TnI and PIA–TnI/cTnC at pCa 8 and pCa 4. (C) Effects of cTnC and Ca 2+ on monomer (at 385 nm), excimer (at 485 nm) fluorescence peaks, and the ratio of excimer fluorescence to monomer fluorescence (E/M ratio) for PIA-labeled TnI. (D) Relative excimer and monomer fluorescence of PIA-labeled TnI-cTnC as a function of pCa. PM-labeled TnI (1 l M ) formed a binary complex with cTnC (3 l M ) from porcine heart in 100 m M Mops (pH 7.0), 0.1 M KCl, 2 m M EGTA, 5 m M Mg 2+ and varying pCa. Each value is mean ± SEM of five measurements. 3098 Y M. Liou and M W. Chen (Eur. J. Biochem. 270) Ó FEBS 2003 12. Luo,Y.,Wu,J.L.,Gergely,J.&Tao,T.(1997)TroponinTand Ca 2+ dependence of the distance between Cys48 and Cys133 of troponin I in the ternary troponin complex and reconstituted thin filaments. Biochemistry 36, 11027–11035. 13. Luo,Y.,Wu,J.L.,Gergely,J.&Tao,T.(1998)Localizationof Cys133 of rabbit skeletal troponin-I with respect to troponin-C by resonance energy transfer. Biophys. J. 74, 3111–3119. 14. Tung, S.C., Wall, M.E., Gallagher, S.C. & Trewhella, J. (2000) A model of troponin-I in complex with troponin-C using hybrid experimental data: The inhibitory region is a b-hairpin. Protein. Sci. 9, 1312–1326. 15. Chong, P.C. & Hodges, R.S. (1982) Proximity of sulfhydryl groups to the sites of interaction between components of the troponin complex from rabbit skeletal muscle. J. Biol. Chem. 257, 2549–2555. 16. Hitchcock, S.E., Zimmerman, C.J. & Smakkey, C. (1981) Study of the structure of troponin-T by measuring the relative reactivities of lysines with acetic anhydride. J. Mol. Biol. 147, 125–151. 17. Syska, H., Wilkinson, J.M., Grand, R.J. & Perry, S.V. (1976) The relationship between biological activity and primary structure of troponin I from white skeletal muscle of the rabbit. J. Biochem. (Tokyo) 153, 375–387. 18. Van Eyk, J.E. & Hodges, R.S. (1988) The biological importance 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. 19. Wilkinson, J.M. & Grand, R.J.A. (1978) Comparison of amino acid sequence of troponin I from different striated muscles. Nature 271, 31–35. 20. Betcher-Lange, S.L. & Lehrer, S.S. (1978) Pyrene excimer fluor- escence in rabbit skeletal aatropomyosin labeled with N-(1-py- rene) maleimide. J. Biol. Chem. 253, 3757–3760. 21. Ishii, Y. & Lehrer, S.S. (1990) Excimer fluorescence of pyr- enyliodoacetamide-labeled tropomyosin: a probe of the state of tropomyosin in reconstituted muscle thin filaments. Biochemistry 29, 1160–1166. 22. Strasburg, G.M., Leavis, P.C. & Gergely, J. (1985) Troponin-C- mediated calcium-sensitive changes in the conformation of troponin I detected by pyrene excimer fluorescence. J. Biol. Chem. 260, 366–370. 23. Van Eerd, J.P. & Takahashi, K. (1975) The amino acid sequence of bovine cardiac troponin-C: comparison with rabbit skeletal troponin-C. Biochem. Biophys. Res. Commun. 64, 122–127. 24. Wilkinson, J.M. (1976) The amino acid sequence of troponin C from chicken skeletal muscle. FEBS Lett. 70, 254–256. 25. Reinach, F.C. & Karlsson, R. (1988) Cloning expression, and site- directed mutagenesis of chicken skeletal muscle troponin C. J. Biol. Chem. 263, 2371–2376. 26. Kobayashi, T., Tagagi, T., Konishi, K., Morimoto, S. & Ohtsuki, I. (1989) Amino acid sequence of porcine cardiac muscle troponin C. J. Biochem. (Tokyo) 106, 55–59. 27. Potter, J.D. (1982) Preparation of troponin and its subunits. Methods Enzymol. 85, 241–263. 28. Szynkiewicz, J., Stepkowski, D., Brzeska, H. & Drabikowski, W. (1985) Cardiac troponin-C: a rapid and effective method of puri- fication. FEBS Lett. 181, 281–285. 29. Liou, Y.M. & Fuchs, F. (1992) Pyrene-labeled cardiac troponin C: Effects of Ca 2+ on monomer and excimer fluorescence in solution and myofibrils. Biophys. J. 61, 892–901. 30. Liou, Y.M., Jiang, M.J. & Wu, M.C. (2002) Changes in thiol reactivity and extractability of myofibril bound cardiac troponin C in porcine malignant hyperthermia. J. Biochem. (Tokyo) 132, 317–325. 31. Shiraishi, F., Kambara, M. & Ohtsuki, I. (1992) Replacement of troponin components in myofibrils. J. Biochem. (Tokyo) 111, 61–65. 32. Fabiato, A. & Fabiato. F. (1979) Calculator programs for com- puting the composition of the multiple metals and ligands used for experiments in skinned muscle cells. J. Physiol. (Paris) 75, 463–505. 33. Steel, R.G.D. & Torrie, J.H. (1980) Principles and Procedures of Statistics: A Biometrical Approach, pp. 39–121. McGraw-Hill, New York. 34. Morimoto, S. & Ohtsuki, I. (1987) Ca 2+ - and Sr 2+ -sensitivity of the ATPase activity of rabbit skeletal myofibrils: Effect of the complete substitution of troponin C with cardiac troponin C, calmodulin, and parvalbumins. J. Biochem. (Tokyo) 101, 291–301. 35. Putkey,J.A.,Liu,W.&Sweeney,H.L.(1991)Functionofthe N-terminal calcium-binding sites in cardiac/slow troponin C assessed in fast skeletal muscle fibers. J. Biol. Chem. 266, 14881–14884. 36. Farah, C.S., Miyamoto, C.A., Ramos, C.H.I., da Silva, C.R., Quaggio, R.B., Fujimori, K., Smillie, L.B. & Reinach, F.C. (1994) Structural and regulatory functions of the NH 2 -andCOOH- terminal regions of skeletal muscle troponin I. J. Biol. Chem. 269, 5230–5240. 37. Tripet, B., Van Eyk, J.E. & Hodges, R.S. (1997) Mapping of a second actin-tropomyosin and a second troponin C binding site within the C terminus of troponin I, and their importance in the Ca 2+ -dependent regulation of muscle contraction. J. Mol. Biol. 271, 728–750. 38. Perry, S.V. (1999) Troponin I: Inhibitor or facilitator. Mol. Cell Biochem. 190, 9–32. 39.Leavis,P.C.,Rosenfeld,S.S.,Gergely,J.,Grabarek,Z.& Drabikowski, W. (1978) Proteolytic fragments of troponin C: Localization of high and low affinity Ca 2+ binding sites and interactions with troponin I and troponin T. J. Biol. Chem. 253, 5452–5459. 40. Grabarek,Z.,Drabikowski,W.,Leavis,P.C.,Rosenfeld,S.S.& Gergely, J. (1981) Proteolytic fragments of troponin C. Inter- actions with the other troponin subunits and biological activity. J. Biol. Chem. 256, 13121–13127. 41. Pearlstone, J.R. & Smillie, L.B. (1978) Troponin T fragments: physical properties and binding to troponin C. Can. J. Biochem. 56, 521–527. 42. Stefancsik, R., Jha, P.K. & Sarkar, S. (1998) Identification and mutagenesis of a highly conserved domain in troponin T responsible for troponin I binding: Potential role for coiled coil interaction. Proc. Natl Acad. Sci. USA 95, 957–962. 43. Sheng, Z., Pan, B.S., Miller, T.E. & Potter, J.D. (1992) Isolation, expression, and mutation of a rabbit skeletal muscle cDNA clone for troponin I. The role of the NH 2 terminus of fast skeletal muscle troponin I in its biological activity. J. Biol. Chem. 267, 25407– 25413. 44. Potter, J.D., Sheng, Z., Pan, B.S. & Zhao, J. (1995) A direct reg- ulatory role for troponin T and a dual role for troponin C in the Ca 2+ regulation of muscle contraction. J. Biol. Chem. 270, 2557–2562. 45. Zhao, X., Kobayashi, T., Malak, H., Gryczynski, Z., Gryczynski, I., Lakowicz, J., Wade, R. & Collins, J.H. (2000) Calcium-induced flexibility changes in the troponin C-troponin I complex. Biochim. Biophys. Acta 1494, 247–254. 46. Zhao, X., Kobayashi, T., Malak, H., Gryczynski, I., Lakowicz, J., Wade, R. & Collins, J.H. (1995) Calcium- induced troponin flexibility revealed by distance distribution measurements between engineered sites. J. Biol. Chem. 270, 15507–15514. 47. Malnic, B., Farah, C.S. & Reinach, F.C. (1998) Regulatory properties of the NH 2 - and COOH-terminal domains of troponin T. ATPase activation and binding to troponin I and troponin C. J. Biol. Chem. 273, 10594–10601. Ó FEBS 2003 Conformational changes in chicken sTnI (Eur. J. Biochem. 270) 3099 48. Rarick, H.M., Tang, H.P., Guo, X.D., Martin, A.F. & Solaro, R.J. (1999) Interactions at the NH 2 -terminal interface of cardiac troponin I modulate myofilament activation. J. Mol. Cell. Cardiol. 31, 363–375. 49. Ngai, S.M., Pearlstone, J.R., Smillie, L.B. & Hodges, R.S. (2001) Characterization of the biologically important interaction between troponin C and the N-terminal region of troponin I. J. Cell Bio- chem. 83, 99–110. 50. Ngai, S.M., Pearlstone, J.R., Farah, C.S., Reinach, F.C., Smillie, L.B. & Hodges, R.S. (2001) Structural and functional studiesontroponinIandtroponinCinteractions.J. Cell Biochem. 83, 33–46. 3100 Y M. Liou and M W. Chen (Eur. J. Biochem. 270) Ó FEBS 2003 . Calcium-dependent protein–protein interactions induce changes in proximity relationships of Cys48 and Cys64 in chicken skeletal troponin I Ying-Ming Liou. conformational changes in the N-terminal domain of chicken troponin I (TnI) to Ca 2+ activation of the actin–myosin interaction. The two cysteine residues in