Calcium-dependent Changes in the Flexibility of the Regulatory Domain of Troponin C in the Troponin Complex
2005; Elsevier BV; Volume: 280; Issue: 23 Linguagem: Inglês
10.1074/jbc.m500574200
ISSN1083-351X
AutoresTharin M. A. Blumenschein, Deborah B. Stone, Robert J. Fletterick, Robert A. Mendelson, Brian D. Sykes,
Tópico(s)Signaling Pathways in Disease
ResumoWith the recent advances in structure determination of the troponin complex, it becomes even more important to understand the dynamics of its components and how they are affected by the presence or absence of Ca2+. We used NMR techniques to study the backbone dynamics of skeletal troponin C (TnC) in the complex. Transverse relaxation-optimized spectroscopy pulse sequences and deuteration of TnC were essential to assign most of the TnC residues in the complex. Backbone amide 15N relaxation times were measured in the presence of Ca2+ or EGTA/Mg2+. T1 relaxation times could not be interpreted precisely, because for a molecule of this size, the longitudinal backbone amide 15N relaxation rate due to chemical shift anisotropy and dipole-dipole interactions becomes too small, and other relaxation mechanisms become relevant. T2 relaxation times were of the expected magnitude for a complex of this size, and most of the variation of T2 times in the presence of Ca2+ could be explained by the anisotropy of the complex, suggesting a relatively rigid molecule. The only exception was EF-hand site III and helix F immediately after, which are more flexible than the rest of the molecule. In the presence of EGTA/Mg2+, relaxation times for residues in the C-domain of TnC are very similar to values in the presence of Ca2+, whereas the N-domain becomes more flexible. Taken together with the high flexibility of the linker between the two domains, we concluded that in the absence of Ca2+, the N-domain of TnC moves independently from the rest of the complex. With the recent advances in structure determination of the troponin complex, it becomes even more important to understand the dynamics of its components and how they are affected by the presence or absence of Ca2+. We used NMR techniques to study the backbone dynamics of skeletal troponin C (TnC) in the complex. Transverse relaxation-optimized spectroscopy pulse sequences and deuteration of TnC were essential to assign most of the TnC residues in the complex. Backbone amide 15N relaxation times were measured in the presence of Ca2+ or EGTA/Mg2+. T1 relaxation times could not be interpreted precisely, because for a molecule of this size, the longitudinal backbone amide 15N relaxation rate due to chemical shift anisotropy and dipole-dipole interactions becomes too small, and other relaxation mechanisms become relevant. T2 relaxation times were of the expected magnitude for a complex of this size, and most of the variation of T2 times in the presence of Ca2+ could be explained by the anisotropy of the complex, suggesting a relatively rigid molecule. The only exception was EF-hand site III and helix F immediately after, which are more flexible than the rest of the molecule. In the presence of EGTA/Mg2+, relaxation times for residues in the C-domain of TnC are very similar to values in the presence of Ca2+, whereas the N-domain becomes more flexible. Taken together with the high flexibility of the linker between the two domains, we concluded that in the absence of Ca2+, the N-domain of TnC moves independently from the rest of the complex. The troponin (Tn) 1The abbreviations used are: Tn, troponin; TnC, troponin C; CTnC, C-domain of troponin C; NTnC, N-domain of troponin C; TnI, troponin I; TnT, troponin T; NMR, nuclear magnetic resonance; TROSY, transverse relaxation optimized spectroscopy; HSQC, heteronuclear single quantum coherence; CSA, chemical shift anisotropy; SANS, small angle neutron scattering; TFE, trifluoroethanol; DTT, dithiothreitol. complex is responsible for the regulation of contraction in striated skeletal and cardiac muscles. Although many structural studies have been done of pairwise interactions between components of the complex (for reviews see Refs. 1Sykes B.D. Nat. Struct. Biol. 2003; 10: 588-589Crossref PubMed Scopus (25) Google Scholar, 2Gordon A.M. Homsher E. Regnier M. Physiol. Rev. 2000; 80: 854-924Crossref Scopus (1342) Google Scholar, 3Tobacman L.S. Annu. Rev. Physiol. 1996; 58: 447-481Crossref PubMed Scopus (461) Google Scholar, 4Farah C.S. Reinach F.C. FASEB J. 1995; 9: 755-767Crossref PubMed Scopus (476) Google Scholar), the details at an atomic level of how the complex works as a whole have not as yet been fully elucidated. Recent determination of the crystal structure of cardiac troponin (5Takeda S. Yamashita A. Maeda K. Maeda Y. Nature. 2003; 424: 35-41Crossref PubMed Scopus (642) Google Scholar) has revealed an apparently quite flexible complex. This has focused interest on the need for dynamics information on the troponin complex in order to properly interpret the structural data. The three components of Tn, troponins I, C, and T, are responsible for inhibiting muscle contraction in the absence of Ca2+, sensing the change in intracellular Ca2+ levels, releasing the inhibition in the presence of Ca2+, and transmitting this information to the other components of muscle thin filaments, respectively (for a recent review see Ref. 2Gordon A.M. Homsher E. Regnier M. Physiol. Rev. 2000; 80: 854-924Crossref Scopus (1342) Google Scholar). Troponin C (TnC) isaCa2+-binding protein containing two domains connected by a linker. The linker is a straight α-helix in the crystal structure of isolated TnC (6Herzberg O. James M.N.G. J. Mol. Biol. 1988; 203: 761-779Crossref PubMed Scopus (292) Google Scholar, 7Satyshur K.A. Rao S.T. Pyzalska D. Drendel W. Greaser M. Sundaralingam M. J. Biol. Chem. 1988; 263: 1628-1647Abstract Full Text PDF PubMed Google Scholar, 8Houdusse A. Love M.L. Dominguez R. Grabarek Z. Cohen C. Structure (Lond.). 1997; 5: 1695-1711Abstract Full Text Full Text PDF PubMed Scopus (155) Google Scholar) but is flexible in solution (9Slupsky C.M. Sykes B.D. Biochemistry. 1995; 34: 15953-15964Crossref PubMed Scopus (186) Google Scholar). Each domain contains two EF-hand metal-binding sites, and the helices that flank those sites are named A to D in the N-domain and are named E to H in the C-domain. An extra helix at the N terminus is named N. The sites in the N-domain are Ca2+-specific and important for the regulation of muscle contraction, whereas the EF-hand sites in the C-domain have higher affinity, can bind both Ca2+ and Mg2+, and are permanently occupied with one of the two ions. Because of these differences, the two domains are known as regulatory and structural domains, respectively. When the sites are occupied in each domain, conformational changes expose hydrophobic pockets that interact with specific regions of troponin I (TnI). TnI is a mainly helical protein, which interacts with TnC, troponin T (TnT), and actin. The N-terminal region of TnI interacts with the C-domain of TnC (CTnC) and is followed by a region that forms a coiled-coil with a portion of TnT. The inhibitory region, capable of binding to actin and inhibiting muscle contraction, and the switch region follow. The "switch" region is named that because it interacts with the N-domain of TnC (NTnC) in a Ca2+-dependent manner, removing the effect of the inhibitory region (10Tripet B. Van Eyk J.E. Hodges R.S. J. Mol. Biol. 1997; 271: 728-750Crossref PubMed Scopus (184) Google Scholar). The C terminus of TnI also interacts with actin. TnT has two domains, T1 and T2. T1 is an elongated domain that interacts with tropomyosin and actin, the other two main components of muscle thin filaments, whereas T2 is mainly responsible for interacting with the other two components of the troponin complex (2Gordon A.M. Homsher E. Regnier M. Physiol. Rev. 2000; 80: 854-924Crossref Scopus (1342) Google Scholar). There have been extensive studies of the interactions among the troponin subunits in various binary and ternary complexes. These are summarized in diagrammatic form in the steric blocking model of Potter and Gergely (11Potter J.D. Gergely J. Biochemistry. 1974; 13: 2697-2703Crossref PubMed Scopus (215) Google Scholar) and the representations of the thin filament by Smillie and co-workers (12Heeley D.H. Golosinska K. Smillie L.B. J. Biol. Chem. 1987; 262: 9971-9978Abstract Full Text PDF PubMed Google Scholar) and Hodges and co-workers (10Tripet B. Van Eyk J.E. Hodges R.S. J. Mol. Biol. 1997; 271: 728-750Crossref PubMed Scopus (184) Google Scholar). Many recent studies have tended to focus on the interactions between the domains of TnC and peptides from TnI and TnT and have mapped out the interacting regions in some detail. Structures of complexes between TnC domains and TnI peptides (13Vassylyev D.G. Takeda S. Wakatsuki S. Maeda K. Maeda Y. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 4847-4852Crossref PubMed Scopus (193) Google Scholar, 14Li M.X. Spyracopoulos L. Sykes B.D. Biochemistry. 1999; 38: 8289-8298Crossref PubMed Scopus (248) Google Scholar, 15Wang X. Li M.X. Sykes B.D. J. Biol. Chem. 2002; 277: 31124-31133Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 16Mercier P. Ferguson R.E. Irving M. Corrie J.E.T. Trentham D.R. Sykes B.D. Biochemistry. 2003; 42: 4333-4348Crossref PubMed Scopus (33) Google Scholar, 17Lindhout D.A. Sykes B.D. J. Biol. Chem. 2003; 278: 27024-27034Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar) have revealed the nature of these interactions at atomic detail and have been confirmed by the recent x-ray structures of cardiac (5Takeda S. Yamashita A. Maeda K. Maeda Y. Nature. 2003; 424: 35-41Crossref PubMed Scopus (642) Google Scholar) and skeletal troponin (18Vinogradova M.V. Stone D.B. Malanina G.G. Karatzaferi C. Cooke R. Mendelson R.A. Fletterick R.J. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 5038-5043Crossref PubMed Scopus (261) Google Scholar). The general mechanism of muscle regulation is the same in skeletal and cardiac muscles, but different isoforms allow for fine-tuning. In cardiac TnC, the first EF-hand site is not functional, and the N-domain can bind only one Ca2+ (19van Eerd J.-P. Takahashi K. Biochem. Biophys. Res. Commun. 1975; 64: 122-127Crossref PubMed Scopus (116) Google Scholar). The binding of one Ca2+ is not enough to expose the hydrophobic pocket, and the N-domain of isolated cardiac TnC is always in a "closed" state. Interaction with the switch region of TnI causes the opening of the domain (14Li M.X. Spyracopoulos L. Sykes B.D. Biochemistry. 1999; 38: 8289-8298Crossref PubMed Scopus (248) Google Scholar). The second main difference between skeletal and cardiac troponin complexes is that cardiac TnI has an N-terminal extension, absent in skeletal TnI. TnT has multiple isoforms, expressed differently in different tissues, including cardiac and slow and fast skeletal muscles. Despite those differences, the similarities of the troponin complex in the two systems justify the comparison between cardiac and skeletal data. NMR techniques can be used to study internal motions of different time scales in proteins. Relaxation measurements give insight into protein backbone, side chain, and domain dynamics in the picosecond to nanosecond time scale, as well as the presence of conformational or chemical exchange in the microsecond to millisecond time scale (for reviews see Refs. 20Kay L.E. Biochem. Cell Biol. 1998; 76: 145-152Crossref PubMed Scopus (71) Google Scholar and 21Palmer A.G. Chem. Rev. 2004; 104: 3623-3640Crossref PubMed Scopus (705) Google Scholar). Although many nuclei (2H, 13C, 15N, 19F and 31P) can be used for relaxation studies, backbone amide 15N relaxation rates (the longitudinal relaxation time, T1, and the transverse relaxation time, T2, along with nuclear Overhauser effect) are the most common. These relaxation measurements are typically interpreted using the Lipari-Szabo model-free formalism (22Lipari G. Szabo A. J. Am. Chem. Soc. 1982; 104: 4546-4559Crossref Scopus (3405) Google Scholar), from which an overall rotational correlation time, an internal correlation time, and an order parameter can be obtained for each N–HN bond vector (23Mandel A.M. Akke M. Palmer A.G. J. Mol. Biol. 1995; 246: 144-163Crossref PubMed Scopus (906) Google Scholar). To study the dynamics of TnC within the troponin complex, we used a reconstituted 55-kDa troponin complex containing whole skeletal TnC, whole skeletal TnI, and the T2 domain of skeletal TnT. The T1 domain of TnT was not included, because it causes aggregation of the complex (24Stone D.B. Timmins P.A. Schneider D.K. Krylova I. Ramos C.H.I. Reinach F.C. Mendelson R.A. J. Mol. Biol. 1998; 281: 689-704Crossref PubMed Scopus (58) Google Scholar, 25King W.A. Stone D.B. Timmins P.A. Narayanan T. von Brasch A.A.M. Mendelson R.A. Curmi P.M.G. J. Mol. Biol. 2005; 345: 797-815Crossref PubMed Scopus (37) Google Scholar), and it does not interact with TnC and TnI (2Gordon A.M. Homsher E. Regnier M. Physiol. Rev. 2000; 80: 854-924Crossref Scopus (1342) Google Scholar). We used this complex to measure backbone amide 15N NMR relaxation times, which give information about the rotational tumbling and internal dynamics of proteins (20Kay L.E. Biochem. Cell Biol. 1998; 76: 145-152Crossref PubMed Scopus (71) Google Scholar). These results have been used to delineate the relative motions of the regulatory and structural domains of TnC in the apo (EGTA/Mg2+) and calcium-saturated states. These data demonstrate the existence of flexibility of the N-domain of TnC during the contraction cycle. Sample Preparation—Uniformly deuterated cysteineless chicken fast skeletal muscle TnI was expressed and purified as described (24Stone D.B. Timmins P.A. Schneider D.K. Krylova I. Ramos C.H.I. Reinach F.C. Mendelson R.A. J. Mol. Biol. 1998; 281: 689-704Crossref PubMed Scopus (58) Google Scholar), using a construct provided by Dr. Smillie (26Dargis R. Pearlstone J.R. Barrete-Ng I. Edwards H. Smillie L.B. J. Biol. Chem. 2002; 277 (and references within): 34662-34665Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). Substitution of cysteineless TnI for wild-type TnI in the Tn complex does not alter the Ca2+-dependent regulation of myosin ATPase activity (26Dargis R. Pearlstone J.R. Barrete-Ng I. Edwards H. Smillie L.B. J. Biol. Chem. 2002; 277 (and references within): 34662-34665Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). A uniformly deuterated deletion mutant of chicken skeletal muscle TnT (isoform TnT-3), corresponding to the T2 domain, was expressed and purified as described (25King W.A. Stone D.B. Timmins P.A. Narayanan T. von Brasch A.A.M. Mendelson R.A. Curmi P.M.G. J. Mol. Biol. 2005; 345: 797-815Crossref PubMed Scopus (37) Google Scholar). 13C,15N- and 2H,13C,15N-labeled recombinant chicken skeletal TnC were expressed as described (27Slupsky C.M. Kay C.M. Reinach F.C. Smillie L.B. Sykes B.D. Biochemistry. 1995; 34: 7365-7375Crossref PubMed Scopus (62) Google Scholar). After lysis by a French press, salts (CaCl2 to 5 mm, MgCl2 to 1 mm, NaCl2 to 50 mm, and DTT to 1 mm) were added to the soluble fraction of the lysate, which was then applied to a phenyl-Sepharose column, previously equilibrated with 50 mm Tris·Cl, pH 7.5, 50 mm NaCl, 5 mm CaCl2, 1 mm MgCl2, 1 mm DTT, and 0.01% sodium azide. The column was then washed with the same buffer but containing 1 m NaCl and 100 mm CaCl2, and TnC was eluted with a buffer containing 50 mm Tris·Cl, pH 7.5, 1 mm EDTA, 1 mm DTT. For [2H,13C,15N]TnC, Escherichia coli cells conditioned to grow in medium containing D2O were used, and the medium used for expression contained D2O instead of H2O. The troponin complex containing these subunits was assembled and purified as described (25King W.A. Stone D.B. Timmins P.A. Narayanan T. von Brasch A.A.M. Mendelson R.A. Curmi P.M.G. J. Mol. Biol. 2005; 345: 797-815Crossref PubMed Scopus (37) Google Scholar). The resulting complex was concentrated, dialyzed against 100 mm NH4HCO3, 0.1 mm DTT, and lyophilized. For each NMR sample, ∼13 mg of freeze-dried complex were dissolved in 550 μl of buffer containing 10 mm Hepes, 250 mm KCl, 2 mm DTT, 0.03% sodium azide, and 0.2 mm 2,2-dimethyl-2-silapentane-5-sulfonic acid in 90% H2O, 10% D2O. For experiments in the presence of calcium, the buffer also contained 5 mm CaCl2, and the pH was adjusted to 6.8. For experiments in the presence of EGTA/Mg2+, the buffer also contained 5 mm MgCl2 and 10 mm EGTA, and the pH was adjusted to 7.1. The solution was filtered by using a microcentrifuge tube filter with 0.22-μm pore size, and 500 μl of it were transferred to an NMR tube. Expected free and bound concentrations of each ion and binding site were calculated with MaxChelator 2.50 (28Patton C. Thompson S. Epel D. Cell Calcium. 2004; 35: 427-431Crossref PubMed Scopus (329) Google Scholar). This program can be obtained from www.stanford.edu/~cpatton/maxc.html. The dissociation constants used for the N- and C-domains of TnC were determined by Potter and Gergely (29Potter J.D. Gergely J. J. Biol. Chem. 1975; 250: 4628-4633Abstract Full Text PDF PubMed Google Scholar). Under the conditions used, less than 10% of the N-domain sites were expected to be occupied in the presence of EGTA/Mg2+. This should not have affected the measurements, because titration of EGTA into Ca2+-bound complex indicated that Ca2+ binding happened as a slow exchange event in the NMR time scale, and the "free" and "bound" residues appeared as different peaks in the spectrum (data not shown). A control sample ( (TnC·Ca42++TnI-(115−131))) was produced using the same buffer and conditions for the presence of calcium. To 2H,13C,15N-labeled TnC·Ca42+, an equimolar amount of unlabeled TnI-(115–131) peptide was added to prevent the previously described dimerization of TnC through the N domain (27Slupsky C.M. Kay C.M. Reinach F.C. Smillie L.B. Sykes B.D. Biochemistry. 1995; 34: 7365-7375Crossref PubMed Scopus (62) Google Scholar). NMR Spectroscopy—To determine NMR chemical shift assignments of TnC·Ca42+ in the troponin complex, {1H,15N} TROSY-HSQC and three-dimensional TROSY-HNCA spectra were acquired in a Varian INOVA 800-MHz spectrometer, using BioPack (Varian, Inc.) pulse sequences. For the TnC·Ca42++TnI−(115−131) complex, {1H,15N} TROSY-HSQC and three-dimensional TROSY-HNCACB spectra were used. NMR chemical shift assignments for TnC and TnC domains, isolated or bound to TnI peptides, in the presence or absence of Ca2+, were used as a guideline in assigning the spectra (30Mercier P. Li M.X. Sykes B.D. Biochemistry. 2000; 39: 2902-2911Crossref PubMed Scopus (49) Google Scholar, 31McKay R.T. Tripet B.P. Hodges R.S. Sykes B.D. J. Biol. Chem. 1997; 272: 28494-28500Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar, 32Gagné S.M. Tsuda S. Li M.X. Chandra M. Smillie L.B. Sykes B.D. Protein Sci. 1994; 3: 1961-1974Crossref PubMed Scopus (176) Google Scholar). {1H,15N} HSQC NMR spectra were also acquired, using the BioPack (Varian, Inc.) pulse sequence, for comparison with the TROSY pulse sequences. For the backbone amide 15N relaxation measurements, spectra were acquired on Varian INOVA 600- and 800-MHz spectrometers, using the T1 or T2 options in the gNhsqc BioPack (Varian, Inc.) pulse sequence, based on pulse sequences developed by Kay and co-workers (33Kay L.E. Nicholson L.K. Delaglio F. Bax A. Torchia D.A. J. Magn. Reson. 1992; 97: 359-375Google Scholar, 34Farrow N.A. Muhandiram R. Singer A.U. Pascal S.M. Kay C.M. Gish G. Shoelson S.E. Pawson T. Forman-Kay J.D. Kay L.E. Biochemistry. 1994; 33: 5984-6003Crossref PubMed Scopus (2018) Google Scholar). These pulse sequences use 1H 180° pulses during the relaxation period, to eliminate by averaging the effects of cross-correlation between dipolar and CSA relaxation mechanisms, and were chosen to facilitate comparison with previously published data. Relaxation delays were 10, 30, 50, 70, and 90 ms for T2 at both fields; 0.1, 0.5, 0.9, 1.3, 1.7, and 2.1 s for T1 at 600 MHz; and 0.1, 0.25, 0.4, 0.7, 1.0, and 1.3 s for T1 at 800 MHz. Spectral width was 12,000 Hz in the 1H dimension, 2500 in the 15N dimension, and 6034 Hz (HNCA) or 11,000 Hz (HNCACB) in the 13C dimension. Two-dimensional NMR spectra were collected with 2048 (t2) × 96 (t1) complex points in TROSY mode, and either 720 (t2) × 96 (t1) or 672 (t2) × 96 (t1) complex points when decoupling the indirectly detected dimension during acquisition. Three-dimensional NMR spectra were collected with 2048 (t3) × 64 (t1) × 36 (t2) (for HNCACB) or 2048 (t3) × 62 (t1) × 32 (t2) (for HNCA) complex points. All spectra were acquired at 30 °C. All experimental free induced decays were processed with NMRPipe (35Delaglio F. Grzwsiek S. Vuister G.W. Zhu G. Pfeifer J. Bax A. J. Biomol. NMR. 1995; 6: 277-293Crossref PubMed Scopus (11630) Google Scholar) and analyzed with NMRView (36Johnson B. Blevins R. J. Biomol. NMR. 1994; 4: 603-614Crossref PubMed Scopus (2686) Google Scholar), except for T1 envelope measurements, which were processed and analyzed with Vnmr 6.1 (Varian, Inc.). Linear prediction for up to half the number of experimental points was used in the indirectly detected dimension of experiments used for assigning. No linear prediction was used in the relaxation experiments. Data were zero-filled to the next power of 2 and multiplied by a sine-bell apodization function shifted by 60–90° before Fourier transformation. Data used for assigning were also multiplied by a Gaussian function before Fourier transformation. Backbone amide 15N T2 relaxation data were fitted to a two-parameter exponential decay using the "Rate Analysis" function of NMRView. The error bars in Fig. 5 represent the confidence intervals calculated by NMRView, using the noise level as an estimate for the standard deviation in intensity. For backbone amide 15N T1 relaxation spectra, the spectral region containing HN amide signals was integrated, and the resulting areas were fitted to a single exponential curve using the in-house program xcrvfit (by Robert Boyko, available at www.pence.ca/software/xcrvfit/). The expected values for T1 and T2 were calculated by using an awk script provided by Dr. Stéphane Gagné, a Mathematica script provided by Dr. Leo Spyracopoulos, or the program HYDRONMR (37García de la Torre J. Huertas M.L. Carrasco B. J. Magn. Reson. 2000; 147: 138-146Crossref PubMed Scopus (452) Google Scholar). For HYDRONMR calculations, the parameters used were temperature = 303 K, viscosity = 0.8 centipoise, effective radius of the atomic elements = 2 Å, γH = 2.654 × 104 radians/s·G, γN =–2.713 × 103 radians/s·G, N–HN bond length = 1.04 Å, and 15N chemical shift anisotropy = –160 ppm. NMR studies of proteins and protein complexes larger than 30 kDa are hindered by fast transverse relaxation of the spins, which broadens the spectral lines, decreasing resolution and limiting the use of many modern and complex pulse sequences commonly used to characterize and assign protein NMR spectra. TROSY techniques (38Pervushin K. Riek R. Wieder G. Wüthrich K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12366-12371Crossref PubMed Scopus (2084) Google Scholar) reduce transverse relaxation by taking advantage of the interference between its two main contributing sources, dipole-dipole interactions and chemical shift anisotropy, which results in narrower spectral lines, especially for the amide NH resonances. Extensive deuteration of the non-exchangeable protons in a protein can also reduce transverse relaxation, by reducing dipole-dipole interactions between 15N, 1HN, and remote (not covalently bound) protons. We compared the independent effects of TROSY and deuteration on the 55-kDa skeletal troponin complex by acquiring {1H,15N} HSQC and {1H,15N} TROSY-HSQC two-dimensional NMR spectra of troponin complex containing uniformly deuterated TnI and TnT-T2 and either 13C,15N- or 2H,13C,15N-labeled TnC in the presence of Ca2+. The simultaneous use of both TROSY techniques and deuteration greatly increased the resolution. A larger improvement was seen by deuteration of TnC, even with non-TROSY pulse sequences, than by using TROSY pulse sequences with non-deuterated TnC (Fig. 1). A comparison of the {1H,15N} TROSY-HSQC NMR spectra of 2H,13C,15N-labeled TnC·Ca42+, in the troponin complex or bound to the switch peptide (TnI-(115–131)), revealed the differences in the position of the cross-peaks (Fig. 2). TnI-(115–131) was added to minimize the dimerization of isolated TnC. Some peaks were in very different positions in the two spectra, whereas others were in the same position. This is true mostly for residues in the N-domain, due to the presence of TnI-(115–131), which binds to the hydrophobic pocket in the N-domain. Backbone chemical shift assignments for TnC in the troponin complex were based on three-dimensional TROSY-HNCA and three-dimensional TROSY-HNCO NMR spectra acquired at 800 MHz. Assignments were aided by comparison with the previously determined NMR chemical shift values for the C-domain of TnC bound to TnI-(1–40) (30Mercier P. Li M.X. Sykes B.D. Biochemistry. 2000; 39: 2902-2911Crossref PubMed Scopus (49) Google Scholar) and either the N-domain of TnC bound to TnI-(115–131) (31McKay R.T. Tripet B.P. Hodges R.S. Sykes B.D. J. Biol. Chem. 1997; 272: 28494-28500Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar) (for troponin in the presence of Ca2+) or the apo N-domain of TnC (32Gagné S.M. Tsuda S. Li M.X. Chandra M. Smillie L.B. Sykes B.D. Protein Sci. 1994; 3: 1961-1974Crossref PubMed Scopus (176) Google Scholar) (for troponin in the presence of EGTA/Mg2+). 15N, 1HN, and C-α chemical shifts were assigned for more than two-thirds of the residues of TnC in each condition. Fig. 3 compares TnC and TnC domains in different environments to TnC in the troponin complex, by subtracting the NMR chemical shift values of TnC in each environment from the values for TnC in the troponin complex. The chemical shift values used include isolated TnC·Ca42+ (39Slupsky C.M. Reinach F.C. Smillie L.B. Sykes B.D. Protein Sci. 1995; 4: 1279-1290Crossref PubMed Scopus (38) Google Scholar), the regulatory domain of TnC ( (NTnC·Ca22+)) bound to the switch peptide (TnI-(115–131)) (31McKay R.T. Tripet B.P. Hodges R.S. Sykes B.D. J. Biol. Chem. 1997; 272: 28494-28500Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar), the structural domain of TnC ( (CTnC·Ca22+)) bound to the first 40 residues of TnI (30Mercier P. Li M.X. Sykes B.D. Biochemistry. 2000; 39: 2902-2911Crossref PubMed Scopus (49) Google Scholar), and CTnC·Ca22+ bound to a TnT peptide (TnT-(160–193)) (40Blumenschein T.M.A. Tripet B.P. Hodges R.S. Sykes B.D. J. Biol. Chem. 2001; 276: 36606-36612Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar). These peptides represent the main regions of TnI and TnT known to interact with TnC in the troponin complex. Whereas TnC·Ca42+ by itself and CTnC·Ca22+ -TnT-(160–193) show significant differences from TnC in the complex, NTnC·Ca22+ -TnI-(115–131) and CTnC·Ca22+ -TnI-(1–40) show very good agreement with those for the respective regions of TnC in the complex. The {1H,15N} TROSY-HSQC NMR spectrum of 2H,13C,15N-labeled TnC in the troponin complex in the presence of EGTA/Mg2+ shows significant differences from the spectrum in the presence of Ca2+. EGTA competes with TnC for the Ca2+, removing it from the EF-hand sites in TnC. The concentrations used were too low to completely remove Ca2+ from the C-domain high affinity sites, and because the buffer also contained Mg2+, the C-domain sites remain occupied. The N-domain sites, however, have a lower Ca2+ affinity, are Ca2+-specific, and were ≥90% empty in the concentrations of EGTA used, as calculated with MaxChelator 2.50 (28Patton C. Thompson S. Epel D. Cell Calcium. 2004; 35: 427-431Crossref PubMed Scopus (329) Google Scholar). Fig. 4 shows a region of the corresponding {1H,15N} TROSY-HSQC NMR spectra. Residues in the C-domain suffered almost no change, consistent with the metal-binding sites remaining occupied, whereas the NMR chemical shift values for residues in the N-domain changed, becoming more similar to the values for the apo N-domain of TnC. This confirms that in the troponin complex in the presence of EGTA/Mg2+, the EF-hand sites in the N-domain of TnC become empty, whereas the ones in the C-domain of TnC remain occupied. The Ca2+ binding occurs as a slow exchange event in the NMR time scale during the titration between states (data not shown), with the apo residues and calcium-bound residues appearing as different peaks in the spectrum, which were independently assigned (see above). Backbone amide 15N T2 relaxation times were measured for 2H,13C,15N-labeled TnC in the troponin complex in the presence of Ca2+ and EGTA/Mg2+, at both 600 and 800 MHz, and are presented on a residue-specific basis in Fig. 5. These values were obtained by using pulse sequences that average the cross-correlation between dipolar and CSA relaxation mechanisms (33Kay L.E. Nicholson L.K. Delaglio F. Bax A. Torchia D.A. J. Magn. Reson. 1992; 97: 359-375Google Scholar, 34Farrow N.A. Muhandiram R. Singer A.U. Pascal S.M. Kay C.M. Gish G. Shoelson S.E. Pawson T. Forman-Kay J.D. Kay L.E. Biochemistry. 1994; 33: 5984-6003Crossref PubMed Scopus (2018) Google Scholar), chosen to facilitate comparison of the relaxation measurements with previously published data on TnC and TnC-TnI complexes, from this laboratory and under similar conditions. In the presence of Ca2+, T2 values for residues in the N- and C-domains are roughly constant and equal, suggesting that the complex tumbles as a single unit. The T2 values for residues in the N-domain become significantly longer in the presence of EGTA/Mg2+, suggesting that the N-domain moves independently from the remainder of the complex in the presence of EGTA. Table I shows the average of backbone amide 15N T2 values for residues assigned in each domain of TnC in the troponin complex, in the presence of Ca2+ or EGTA/Mg2+, in both fields. Backbone amide 15N T2 relaxation times were also measured using 15N HSQC filtered 1H NMR spectra. These envelope measurements determine th
Referência(s)