Medicine, Science & Technology

Defining the region of troponin-I that binds to troponin-C

Description
Defining the region of troponin-I that binds to troponin-C
Published
of 12
6
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  Defining the Region of Troponin-I that Binds to Troponin-C † Ryan T. M c Kay, Brian P. Tripet, Joyce R. Pearlstone, Lawrence B. Smillie, and Brian D. Sykes*  MRC Group in Protein Structure and Function, Department of Biochemistry, 474 Medical Sciences Building,Uni V  ersity of Alberta, Canada T6G 2H7  Recei V  ed December 17, 1998; Re V  ised Manuscript Recei V  ed February 23, 1999 ABSTRACT : The kinetics and energetics of the binding of three troponin-I peptides, corresponding to regions96 - 131 (TnI 96 - 131 ), 96 - 139 (TnI 96 - 139 ), and 96 - 148 (TnI 96 - 148 ), to skeletal chicken troponin-C wereinvestigated using multinuclear, multidimensional NMR spectroscopy. The kinetic off-rate and dissociationconstants for TnI 96 - 131  (400 s - 1 , 32  µ M), TnI 96 - 139  (65 s - 1 ,  < 1  µ M), and TnI 96 - 148  (45 s - 1 ,  < 1  µ M)binding to TnC were determined from simulation and analysis of the behavior of   1 H, 15 N-heteronuclearsingle quantum correlation NMR spectra taken during titrations of TnC with these peptides. Two-dimensional  15 N-edited TOCSY and NOESY spectroscopy were used to identify 11 C-terminal residuesfrom the  15 N-labeled TnI 96 - 148  that were unperturbed by TnC binding. TnI 96 - 139  labeled with  13 C at fourpositions (Leu 102 , Leu 111 , Met  121 , and Met 134 ) was complexed with TnC and revealed single bound speciesfor Leu 102 and Leu 111 but multiple bound species for Met 121 and Met 134 . These results indicate that residues97 - 136 (and 96 or 137) of TnI are involved in binding to the two domains of troponin-C under calciumsaturating conditions, and that the interaction with the regulatory domain is complex. Implications of these results in the context of various models of muscle regulation are discussed. The transient release of calcium in a muscle cell inresponse to a neural signal results in a cascade of changingprotein ‚ protein interactions and eventually muscle contraction(for review, see refs  1 - 4 ). The sliding of the thick and thinfilaments past one another constitutes the actual physicalmechanism of contraction produced as a result of thehydrolysis of ATP by myosin. The regulatory target forcalcium in skeletal muscle cells is the troponin complexconsisting of troponin-C (TnC), 1 troponin-I, and troponin-T. TnC is the calcium-binding component and is the onlymember of the complex presently resolved at an atomic level.TnI inhibits the ability of myosin to hydrolyze ATP, thuspreventing muscle contraction. Troponin-T anchors thetroponin complex to actin/tropomyosin, interacts with TnCdirectly, and is involved in the activation of contraction inthe presence of calcium (see ref   5  and references therein).Once calcium is released in the cell, the three-dimensionalstructure of TnC and subsequent protein ‚ protein interactionsare altered. Specifically, TnC in the presence of calciuminteracts with TnI more strongly and allows muscle contrac-tion to occur (see refs  1 ,  2 ,  6  ,  7  , and references therein).However, the exact mechanism of how TnI inhibits contrac-tion in the absence of calcium, and conversely participatesin the activation of the actomyosin complex in the presenceof calcium, is not presently understood.TnC is an 18 kDa protein of 162 amino acids comprising2 separate domains (N- and C-domains each  ∼ 80 aminoacids) covalently attached by a flexible “helical” linker(residues  ∼ 80 - 95). Both the C- and N-terminal domainsare predominantly  R  -helical with each individual domaincontaining two EF hand calcium-binding sites that interactthrough a short bisecting   -sheet. The calcium-binding sitesare labeled sequentially I - IV on the basis of their respectivepositions in the primary amino acid sequence of the protein.The C-terminal sites have a higher affinity for Ca 2 + thanthe N-terminal sites, and the C-terminal sites also haveaffinity for Mg 2 + while the N-terminal sites are calcium-specific. Sites I and II have calcium affinities ( ∼ 7  ×  10 4 and 5  ×  10 5 M - 1 for  K  a , respectively) in the range of thetransient Ca 2 + signal ( 8  ). It has not been determined if theC-terminal sites are bound with Mg 2 + at all times, or if theCa 2 + signal is of sufficient duration to displace Mg 2 + in thecontractile state.There is presently no high-resolution structure for TnCwith Mg 2 + bound in sites III and IV, but there are severalcrystal structures of skeletal TnC with calcium present inthe C-domain, and absent from ( 9 ,  10 ) or present in theN-domain ( 11 - 13 ). NMR solution structures include that † This work was supported by the Medical Research Council Groupin Protein Structure and Function, a Faculty of Medicine 75thAnniversary Studentship (R.T.M.), and by two Alberta HeritageFoundation for Medical Research Studentships (R.T.M. and B.P.T.).* To whom correspondence should be addressed. Tel: (403) 492-5460. Fax: (403) 492-0886. E-mail: brian.sykes@ualberta.ca andrtm@polaris.biochem.ualberta.ca. 1 Abbreviations: TnC, whole skeletal chicken troponin-C; TnI,skeletal troponin-I; TnI x , synthetic skeletal troponin-I peptide containingresidues designated by the subscript value “x”; N-TnC, N-domain of skeletal chicken troponin-C; TnC ‚ TnI 96 - 131 , [U- 15 N, 13 C]TnC ‚ TnI 96 - 131 ;TnC ‚ TnI 96 - 139 , [U- 15 N,  13 C-Ala]TnC ‚ TnI 96 - 139 ; TnC ‚ TnI 96 - 148 , [U- 15 N, 13 C-Ala]TnC ‚ TnI 96 - 148 ;  15 N-TnI 96 - 148 , [U- 15 N]TnI 96 - 148 ;  13 C-TnI 96 - 139 , [U- 13 C-(100%)Leu 102 ,-(50%)Leu 111 ,  13 CH 3 -(100%)Met 121 ,-(50%)Met 134 ]TnI 96 - 139 ; 15 N-TnI 96 - 148 ‚ TnC, [U- 2 H]TnC ‚ [U- 15 N]TnI 96 - 148 ;  13 C-TnI 96 - 139 ‚ TnC,[U- 2 H, 15 N]TnC ‚ [U- 13 C-(100%)Leu 102 ,-(50%)Leu 111 ,  13 CH 3 -(100%)-Met 121 ,-(50%)Met 134 ]TnI 96 - 139 ; HSQC, 2D- 15 N-edited heteronuclearsingle quantum correlation NMR spectroscopy;  13 C-HSQC, 2D- 13 C-edited heteronuclear single quantum correlation NMR spectroscopy; ∆ δ , change (   ∆ δ 1 H2 + ∆ δ 1 5N2 ) in chemical shift (Hz) of a backboneamide cross-peak as monitored by HSQC spectroscopy; ∆ δ total , the  ∑ ∆ δ i over all monitored residues at each point of the titration;  ∆ G off  q ,activation energy of the reverse reaction. 5478  Biochemistry  1999,  38,  5478 - 5489 10.1021/bi9829736 CCC: $18.00 © 1999 American Chemical SocietyPublished on Web 04/03/1999  of apo and Ca 2 + -saturated N-TnC ( 14 ), Ca 2 + -saturated E41AN-TnC ( 15 ), Ca 2 + -saturated TnC ( 16  ), both apo and Ca 2 + -saturated cardiac N-TnC ( 17  ), cardiac TnC ( 18  ), and Ca 2 + -saturated N-TnC while bound to TnI 96 - 148  ( 19 ) (see ref   20 for an overall review and comparison of the solution andX-ray troponin-C structures). The structures have revealedimportant information regarding the mechanism of bothcardiac and skeletal muscle regulation. In the skeletal systemthe binding of calcium in the low-affinity sites causes themovement of the B and C helices away from the N, A, andD helices ( 14 ), exposing a hydrophobic pocket that has beenshown to bind TnI ( 19 ,  21 ). Interestingly, this large structuralchange is not observed in cardiac TnC upon binding calcium.Recently, the X-ray crystal structure of 2Ca 2 + (sites III andIV occupied) skeletal TnC while bound with a TnI peptide(residues 1 - 47) was reported ( 22 ,  23 ).Calmodulin and the myosin light chains are very highlyhomologous, multiple EF hand motif-containing, dumbbell-shaped, calcium-binding proteins that have also been resolvedat an atomic level (see ref   24  for review). The structures of CaM, bound to skeletal or smooth muscle myosin light chainkinase peptides ( 25 ,  26  ), a brain CaM-dependent proteinkinase II R  peptide ( 27  ), and the regulatory and essential lightchains of myosin bound to a portion of the heavy chain ( 28  ),have revealed different manners in which the N- andC-terminal domains of calcium regulatory proteins interactwith protein targets (see ref   29  and references therein). Thesestructures and other experiments involving various targetpeptides suggest that calcium-binding proteins may be ableto bind using both domains in an extended structure, or bothdomains in a collapsed orientation to grasp target peptides.Despite the structural information presently available forTnC and the homologous CaM and myosin systems, thereis relatively little known about TnI either in isolation or incomplex with TnC and/or TnT. We do not know exactlywhich residues of TnI interact with TnC under conditionsnormally associated with the contractile or relaxed states,nor do we fully understand the mechanism or nature of theinteraction. However, on the basis of a variety of experi-mental approaches, evidence for an antiparallel arrangementof TnC and TnI molecules ( 2 ) in an extended, partiallyextended, or compact structure has been deduced ( 23 ,  30 - 32 ) and a number of interaction sites identified (see below).Studies have included cross-linking ( 33 - 38  ), low-angleX-ray diffraction of TnC ‚ TnI ( 30 ,  39 ,  40 ), ATPase assaysof various combinations of both intact proteins and theirfragments in the presence or absence of Ca 2 + ( 41 - 43 ), andbinding studies of intact TnC, N-TnC, C-TnC with TnI, andits peptide fragments (i.e., proteolytic, synthetic, or recom-binant) by gel electrophoresis, fluorescence, and NMRmeasurements ( 19 ,  21 ,  44 - 48  ).These studies have identified two regions (1 - 21 and 96 - 148) of the TnI polypeptide chain important in Ca 2 + -dependent interaction with TnC. TnI residues 1 - 40 appearedto bind exclusively to the C-domain of TnC with a lowdissociation constant ( K  d < 10 - 7 M) in the presence of Ca 2 + ( 49 ). This affinity is weakened when Ca 2 + is replaced byMg 2 + ( 50 ). These observations are consistent with otherstudies ( 5 ,  41 ) using longer fragments of TnI (i.e., residues1 - 98 and 1 - 116). The recently reported 2.3 Å X-raystructure of 2 Ca 2 + TnC in complex with TnI 1 - 47  shows a“compact” TnC structure with multiple proposed contactsfrom both N- and C-terminal domains of TnC to the peptide.On the basis of their structure and other data they haveproposed a model. It is presently unclear how representativethis model is of the complex of calcium-saturated TnC withintact TnI (or fragments derived from its central and/orC-terminal regions).The latter segments of TnI are the major focus of thispresent investigation. Residues 96 - 116 have long beenrecognized as containing the major inhibitory activity of TnI(i.e., residues 104 - 115 known as the inhibitory region) ( 7  , 51 ,  52 ). Interaction of the inhibitory region with TnC hasbeen localized to the C-domain and possibly the linker (i.e.,the D/E helix in X-ray structure) between the N- andC-domains. More recently a section of TnI (i.e., residues ∼ 115 - 148) has been recognized as contributing significantlyto inhibition and to Ca 2 + sensitivity of the ATPase activityin the reconstituted troponin ‚ tropomyosin actomyosin system( 41 ,  43 ). In a comparison of binding affinities of TnI 96 - 116 and TnI 96 - 148  to calcium-saturated TnC and its isolated N-and C-domains, the binding to the N-domain of TnC wassignificantly increased with the extended fragment, and a3-fold repeated sequence motif has been reported, specificallyinvolving TnI residues 101 - 114, 121 - 132, and 135 - 146( 47  ). By using a variety of synthetic peptides encompassingresidues 96 - 148, Tripet et al. ( 32 ) have demonstrated theimportance of lysine residues 141, 144, and 145 for fullinhibitory activity (i.e., binding to tropomyosin/actin) andthe importance of the 116 - 126 region in binding to TnC.Two models have been suggested. The first contains a“switching mechanism” that is dependent on the presenceof calcium in the N-domain of TnC (refs  30 ,  32 , andreferences therein), while the second model has the N-terminal TnI residues ( ∼ 1 - 47) specifically involved incalcium-independent binding to the hydrophobic pocket of C-TnC ( 22 ,  23 ). There is presently no definitive experimentto decide between either model.In an attempt to provide more detailed information on theinterface between the two components of the TnC ‚ TnIcomplex, this laboratory has applied multinuclear, multidi-mensional NMR spectroscopy to the interaction of TnC withTnI peptides, specifically in the 96 - 148 region. Studies todate have provided dissociation constants for TnI 115 - 131  andTnI 96 - 148  in complex with calcium-saturated N-TnC andidentified residues perturbed by binding of these peptides( 19 ,  21 ). The data confirm that TnI peptide binding involvesinteractions with residues in the hydrophobic pocket of N-TnC. The present report extends these studies to acomparison of the binding kinetics and energetics of TnI 96 - 131 , TnI 96 - 139 , and TnI 96 - 148  to calcium-saturated, intactTnC. In addition to providing the dissociation and kineticoff-rates for these peptides, the NMR spectral data of thelabeled TnI peptide show that residues 97 - 136 are theresidues within the 96 - 148 region primarily involved inbinding to calcium-saturated TnC. Also, a specifically 13 C-labeled TnI 96 - 139  peptide when complexed to TnC indicatesthat the N-terminal region of the TnI peptide appears to bindin a single orientation (presumably to the C-domain of TnC),while the C-terminal region of TnI assumes multiple boundconformations.Defining the Region of TnI that binds TnC  Biochemistry, Vol. 38, No. 17, 1999  5479  EXPERIMENTAL PROCEDURES Proteins and Peptides . The cloning, expression, andpurification of [U- 15 N, 13 C-Ala] TnC, [U- 15 N, 13 C] TnC, and[U- 2 H (88%)] TnC were performed as described previously( 21 ,  53 ). Synthetic N R  -acetyl peptides corresponding to rabbitskeletal troponin-I regions 96 - 131, 96 - 139, and 96 - 148were prepared as described by Tripet et al. ( 32 ). SyntheticN R  -acetyl TnI 96 - 139  incorporating [U- 13 C]- L -leucine (residues102 and 111) and [ 13 CH 3 ]- L -methionine (residues 121 and134) was prepared as the other peptides except for manualsynthesis runs at each labeled position. Leucine at position102 was 100% labeled while the synthesis of position 111was performed with only 50% labeled leucine, as was thecase for methionine positions 121 (100%) and 134 (50%),respectively. The uniformly  15 N-labeled TnI 96 - 148  peptide wasproduced from the cloning and transformation of pAED4 ‚ TnI 96 - 148  into  Escherichia coli  strain BL21 ‚ DE3 ‚ pLysS(Novagen) for inducible protein expression with 0.4 mMIPTG. The purification of expressed TnI 96 - 148  from an extractof dried acetone powder cell pellet, on a CM-cellulosecolumn at pH 7.5, has been described previously ( 47  ). Toproduce uniformly  15 N-labeled protein, we initially grew cellsin ZB medium ( 54 ) supplemented with ampicillin andchloramphenicol (both at 0.1 mg/mL) to  A 600  ≈  0.7. Tenmilliliters of this culture was inoculated into each of four, 1L volumes of M9 medium ( 55 ), in which NH 4 Cl wasreplaced with 99.4%  15 N-enriched (NH 4 ) 2 SO 4  (Isotec Inc.).The M9 minimal medium was supplemented with filter-sterilized solutions of minerals (final concentration of 2 mMMgSO 4 , 1  µ M FeCl 3 , 25  µ M ZnSO 4 , and 0.1 mM CaCl 2 ),ampicillin, chloramphenicol, and vitamins (1 mg each of  D -biotin, choline chloride, folic acid, niacimamide,  D -pantothenic acid, and pyridoxine, 5 mg of thiamine, and 0.1mg of riboflavin per liter). The concentration and primarysequence of proteins and peptides were confirmed by aminoacid analysis done in quadruplicate, the correct mass verifiedby electrospray mass spectrometry, and the overall purityconfirmed by reverse-phase HPLC.  NMR Sample Preparation . NMR samples were preparedfor each of the [U- 15 N, 13 C]TnC ‚ TnI 96 - 131 , [U- 15 N,  13 C-Ala]-TnC ‚ TnI 96 - 139 , [U- 15 N, 13 C-Ala]TnC ‚ TnI 96 - 148 , [U- 2 H]TnC ‚ [U- 15 N]TnI 96 - 148 , and [U- 2 H, 15 N]TnC ‚ [U- 13 C (100%) Leu 102 ,(50%) Leu 111 ,  13 CH 3  (100%) Met 121 , (50%) Met 134 ]TnI 96 - 139 complexes. All titration samples started with 500  µ Lvolumes. The [U- 2 H]TnC ‚ [U- 15 N]TnI 96 - 148  complex samplehad a volume of 500  µ L with peptide and protein concentra-tions of 5  ×  10 - 4 and 6  ×  10 - 4 M, respectively. The  13 C-TnI 96 - 139 ‚ TnC complex sample had a volume of 500  µ L withpeptide and protein concentrations of 1.2 mM each. All NMRsamples consisted of 90% H 2 O, 10% D 2 O, 10 mM deuteratedimidazole, 100 mM KCl, 0.1 mM 2,2-dimethyl-2-silapentane-5-sulfonate as an internal reference standard, and slightamounts of HCl and/or NaOH as necessary to change thepH of the samples to 6.8 (uncorrected for isotope effects).Each sample contained 8 equiv of calcium per molecule of TnC (i.e., 2 mol per Ca 2 + -binding site). Imidazole has beencalibrated and was used as an internal pH reference standard(data not shown). Titrations of TnC with TnI Peptides . Three separatetitrations were performed. In each case TnC was titrated withthe addition of a TnI peptide, and 2D- 1 H, 15 N-HSQC NMRspectra were taken at each point. In the first titration 5  µ Lof stock TnI 96 - 131  (5.1 × 10 - 8 mol/addition) were added to500  µ L of TnC (0.65 mM) achieving a final ratio of 1.54/1,peptide to TnC. The second titration involved adding 5  µ Lof stock TnI 96 - 139  (6.3  ×  10 - 8 mol/addition) to 500  µ L of TnC (0.85 mM) to a final ratio of 1.48/1, peptide to TnC. Inthe first two titrations the final addition of peptide was only4  µ L because 1  µ L had been set aside for amino acid analysis.The third titration employing the TnI 96 - 148  peptide was foundto have precipitation upon complex formation, and thereforeamino acid analysis was performed before and after titration.Stock TnI 96 - 148  was added 10 times in 5  µ L aliquots (2.25 × 10 - 8 mol/addition) to 500  µ L of TnC (0.4 mM) to a finalratio of 1.1/1, peptide to protein.  NMR Spectroscopy and Assignment  . Experiments wereconducted on a Varian Unity-600 (titrations, TOCSY, andNOESY) or an Inova 500 ( 13 C-HSQC) spectrometer, andspectra were referenced according to Wishart et al. ( 56  ). TheHSQC ( 57  ,  58  ) spectra for the TnC ‚ TnI 96 - 131 , TnC ‚ TnI 96 - 139 ,and TnC ‚ TnI 96 - 148  titrations were acquired at 31  ° C withsweep widths of 8000 Hz (all spectra acquired at 512complex  t  2  points) and 1650.2 Hz (128, 96, and 96 complex t  1  points for the first through third titrations, respectively)for the directly and indirectly detected dimensions, respec-tively, and with 24 transients/increment. The 2D- 15 N-edited-TOCSYHSQC ( 59 ) spectra collected on the  15 N-TnI 96 - 148 ‚ TnC complex was acquired at 25  ° C with 6500 Hz sweepwidths in the indirectly (352 complex  t  1  points) and directly(512 complex  t  2  points) detected dimensions, respectively,and with 256 transients/increment. Both the 150 and 75 msmixing time 2D- 15 N-edited-NOESYHSQC ( 59 ) experimentswere collected at 25  ° C with 7000 Hz sweep widths in boththe directly and indirectly detected dimensions. The 75 msmixing time NOESYHSQC had 512 complex points in bothdimensions with 80 transients/increment, while the 150 msexperiment had 256  t  1  complex points, 512  t  2  complex points,and 256 transients/increment. The  13 C-HSQCs for the  13 C-TnI 96 - 139  peptide and  13 C-TnI 96 - 139 ‚ TnC complex were doneat 31  ° C, and had 7000 Hz (2048  t  2  complex points) and5500 Hz sweep widths (1536  t  2  complex points) for thedirectly detected dimension, respectively. Both  13 C-HSQCswere acquired with a sweep width of 2000 Hz (56 complex t  1  points) for the indirectly detected dimension, and mirror-image linear prediction was used to double the number of complex points. Carbon decoupling was not performedduring the extended, directly detected acquisition period (293ms) to prevent probe damage.All directly and indirectly detected data sets were zerofilled to twice the number of acquired (plus predicted whenused) points, and spectra were apodized using a shifted sinebell before Fourier transformation. The acquisition time forHSQC experiments was approximately 4 h while ∼ 1.5 dayswas required for each of the TOCSY and NOESY experi-ments. All experiments were processed and analyzed usingthe software packages NMRPipe and NMRDraw ( 60 ).Assignment of the TOCSY and NOESY experiments wasperformed as described previously ( 61 ,  62 ).  Dissociation Constants . Two procedures were used todetermine the equilibrium dissociation constants for thereaction of TnC with the TnI peptides. For TnI 96 - 131  the totalchemical shift change of the well-resolved amide protonHSQC cross-peaks was monitored during the titration as a5480  Biochemistry, Vol. 38, No. 17, 1999  McKay et al.  function of added TnI 96 - 131 . The changes were fit to bothsimple 1:1 bindingand to the case where two peptides bind to TnCwhere P designates TnC, L the TnI peptide ligand, and PLand PL 2  stand for the protein ‚ peptide and protein ‚ 2peptidecomplexes, respectively. Fitting was performed using anonlinear least-squares technique (see ref   21  and referencestherein). For the TnI 96 - 139  and TnI 96 - 148  titrations thechemical shift changes were not in the intermediate - fastNMR chemical exchange limit required for this approach.For the longer peptides, spectra taken during the titrationswere analyzed using a full line shape analysis (see below)to determine the dissociation rate constant (under an as-sumption of 1:1 binding). The dissociation constant is aparameter of this fitting process. However, because of therelatively high concentrations used for NMR spectroscopy,the fitting is not sensitive to the value of   K  d  once thedissociation constant is tighter than approximately 1  µ M. Thisis the case for the TnI 96 - 139  and TnI 96 - 148  titrations, and thusonly an upper limit for the  K  d  is determined. A lower limitfor  K  d  can be determined from the fitted value of   k  off   and anupper limit for the value of   k  on  ( e 1 × 10 8 M - 1 s - 1 ; see refs 63 - 65  and references therein) using eq 2.  Line Shape Analysis . The  15 N-labeled TnC was monitoredby 2D- 15 N, 1 H-HSQC NMR spectroscopy during each pointof the titration, for each of the different TnI peptides. Spectralcross-peaks of backbone amides were then analyzed forchanges in chemical shift and line shape during the titrations.Residues from each of the N- and C-terminal domains of troponin-C (Asp 32 and Lys 107 , respectively) were chosen thatspecifically had chemical shift changes only in the protondimension. The lack of chemical shift changes in the nitrogendimension allowed for easier cross-peak simulation andcross-peak display. The program Mathematica ( 66  ) was usedas previously described ( 21 ) to simulate the spectral lineshapes of the selected residues during each titration, exceptthat the script file was modified from the previous study tobetter simulate the effect of dilution on TnC peak intensity. 2 The  k  off   was modified manually in an iterative manner untilthe best observable fit was obtained. The backbone amidecross-peaks for Glu 16 , Val 65 , and Gly 119 were also checked(data not shown), to ensure that the line width behavior wasconsistent with other residues. RESULTS Two-dimensional  1 H, 15 N-HSQC NMR spectroscopy wasused to elucidate the interaction of three different length TnIpeptides with labeled, calcium-saturated chicken skeletalTnC. The three TnI peptides corresponded to regions 96 - 131, 96 - 139, and 96 - 148, respectively. Initially the TnIpeptides were unlabeled, while the TnC protein was  15 N-labeled. This was done to allow the specific NMR observa-tion of TnC in the complex without interference from TnIresonances. The HSQC NMR spectra display backbone andside chain amide cross-peaks that are sensitive to changesin their local environment and thus allow the monitoring of the titration at the amino acid residue level of resolution.Subsequently,  15 N-labeling was incorporated into the TnI 96 - 148 peptide, while the TnC was deuterated to minimize signalloss and focus on peptide residues. In the case of the  15 N-labeled TnI 96 - 148 , 2D  15 N edited TOCSY and NOESY NMRspectroscopy was used to identify peptide amino acid spinsystems after addition of one equivalent of deuterated TnC.A TnI 96 - 139  peptide ( 13 C-TnI 96 - 139 ) was synthesized incor-porating uniformly  13 C-labeled leucine (position 102 and 111at 100% and 50% label, respectively) and  13 C-methyl-labeledmethionine (positions 121 and 134 at 100% and 50% label,respectively).  13 C-HSQC was used to monitor the effect of TnC addition to the  13 C-TnI 96 - 139  peptide.Titrations of TnC with TnI 96 - 131 , TnI 96 - 139 , and TnI 96 - 148 are shown in Figure 1 panels A - C, respectively, usingcontour representations of HSQC spectra. Not all cross-peaksare affected equally, indicating that some TnC residues havea greater change in their local environment than others;however, changes experienced by individual cross-peaks arevery similar when comparing the effect of different peptides.The spectral changes induced in the N-domain of intact TnCupon addition of the TnI 96 - 131  peptide resemble quite closelythe spectra seen for the TnI 115 - 131  and TnI 96 - 148  peptides whenadded to the isolated N-terminal domain of troponin-C ( 19 , 21 ). The largest difference between the three titrations shownin Figure 1 is the change in NMR cross-peak behavior fromintermediate - fast exchange with the TnI 96 - 131  peptide, tointermediate - slow exchange with the TnI 96 - 139  and TnI 96 - 148 peptides. In the NMR fast exchange limit a single cross-peak is observed with a chemical shift that is a weightedaverage of the free and bound chemical shifts. In the slowexchange limit two cross-peaks are observed, one with thechemical shift of the free species and one with the chemicalshift of the bound species. In the slow limit the intensity of each peak represents the relative abundance of each species.We do not observe spectra in either the extreme slow or fastexchange limits (see detailed line shape and chemical shiftanalysis below), but instead see a mixture dependent uponthe total chemical shift and respective rates of complexdissociation. Chemical Shift Analysis and Determination of DissociationConstants . The observed total chemical shift change ( 21 ) wasused to determine a dissociation constant for the reaction of TnI 96 - 131  with TnC. This titration exhibited spectra in theintermediate - fast exchange limit. Chemical shift assignmentsfor TnC were taken from Slupsky et al. ( 67  ). A total of 85backbone amide HSQC cross-peaks (out of 162 possible)were followed during each point of the titration (Figure 2)from which ∆ δ total  ( 21 ) was determined. Figure 2, panels Aand B, show the best fit for a 1:1 and 1:2 binding of TnC toTnI 96 - 131 , respectively. The 1:1 binding was best fit with a K  d  of 32  (  16  µ M, while the 1:2 binding analysis yielded a K  d1  of 1 - 50  µ M and a  K  d2  of  ∼ 2 mM. The binding of twopeptides to TnC was considered because a second binding 2 The new and previous Mathematica scripts used are available uponrequest from brian.sykes@ualberta.ca or ryan.mckay@ualberta.ca. P + L { \ }  k  on k  off  PL (1) K  d ) k  off  k  on (2)P + L [ \ ]  k  off  k  on PL + L { \ }  k  on2 k  off 2 PL 2  (3) Defining the Region of TnI that binds TnC  Biochemistry, Vol. 38, No. 17, 1999  5481  event was observed when excess peptide was added (Figure3). In Figure 3 (especially Panel G) initial peptide binding(i.e., cross-peak broadening) can be seen. Subsequently,coupling of peptide binding and breakup of the TnC dimerresults in sharpening of the N-TnC cross-peak line width upto a 1:1 ratio. In the presence of excess TnI peptide, TnCcross-peaks again broaden. This is probably due to competi-tion for both the N- and C-terminal TnC domains by morethan one TnI peptide. Since the initial binding event is somuch stronger (i.e.,  K  d1  ,  K  d2 ), the second event can beassumed to be negligible and a value of 32  µ M was usedduring the line shape analyses (see below).  Line Shape Analysis and Determination of Exchange Rates . The dissociation rate constants for all three complexes F IGURE  1: Contour plots of an expanded region of the 2D- 1 H, 15 N-HSQC NMR spectra taken of   15 N-labeled TnC upon titration with unlabeledpeptide: (A) TnI 96 - 131 , (B) TnI 96 - 139 , and (C) TnI 96 - 148 , respectively. Panel A shows the addition of 0, 0.2, 0.3, 0.5, 0.6, 0.8, 1.0, 1.1, 1.3,1.4, and 1.5 molar equiv of TnI 96 - 131 , B shows the addition of 0, 0.1, 0.3, 0.4, 0.6, 0.7, 0.9, 1.0, 1.2, 1.3, and 1.5 molar equiv of TnI 96 - 139 ,and C shows the addition of 0, 0.1, 0.2, 0.3, 0.45, 0.6, 0.7, 0.8, 0.9, 1.1 molar equiv of TnI 96 - 148 . Uncomplexed TnC cross-peaks are plottedwith more contours while overlaying single contour cross-peaks show the effect of peptide addition. The different effect on cross-peaksexperiencing fast-intermediate or slow exchange is very evident in residues such as Gly 43 , Asp 32 , and Gly 50 when comparing the additionof TnI 96 - 131  (A) and the two larger peptides (B and C). Residues experience changes of magnitude and direction similar to those of theiramide chemical shift in all three titrations. The one letter code is used for marked residues in the figure.F IGURE  2: Binding curves derived from the 2D- 1 H, 15 N-HSQC spectra of labeled TnC upon addition of TnI 96 - 131 . The ∆ δ total  for the 85 TnCbackbone amide pairs that were monitored throughout the TnI 96 - 131  titration as a function of the molar ratio of TnI 96 - 131 ‚ TnC with a 1:1binding curve fit analysis (see text) is shown in panel A, while (B) is the same data fit with a 1:2 binding of TnC to TnI 96 - 131 . 5482  Biochemistry, Vol. 38, No. 17, 1999  McKay et al.
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x