Feb 18, 2016 - express the TnI deletion mutants we used the T7-based PET expression .... 0.01% NaN3); 3) buffer B containing 100 m~ KCI; and 4) 20 m~ imid ...
THEJOURNAL OF B I O ~ I CCHEMISTRY AL 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc
Vol. 269, No. 7, Issue of February 18,pp. 5230-5240, 1994 Printed in U.S.A.
Structural and Regulatory Functionsof the N H 2 - and COOH-terminal Regionsof Skeletal Muscle Troponin I* (Received for publication, July 2, 1993, and in revised form, October 11, 1993)
Chuck S. Farah, CatarinaA. Miyamoto, Carlos H. I. Ramos, Ana Claudia R. da Silva, Ronald0 B. Quaggio, Kumie Fujimori, Lawrence B. SmillieS, and Fernando C. Reinache From the Department Bioquimica, Znstituto de Quimica, Universidade de Scio Paulo, CP 20.780, CEP 01498,Scio Paulo, Brazil and the fMedical Research Council Grouu in Protein Structure and Function, University of Alberta, Edmonton, Alberta T6G 2H7, Canada ~
~
.~
Calcium binding to regulatory sites located in the NH,-terminal domainof troponin C (TnC)induces a conformational change that blocks the inhibitory action of troponin I (%I) and triggers muscle contraction. We used deletion mutants of TnI in conjunction with a series of TnC mutants to understand the structural and functional relationship between different TnI regions and TnC domains. Ourresults indicate that TnI isorganized into structural and regulatory regions which interact in an antiparallel fashion with the corresponding structural and regulatory regions of TnC. Functional studies show that the COOH-terminal region of TnI, when linked to the inhibitory region (TnIlos_18z) can regulate actomyosin ATPase.A TnI lacking the first 57 amino acids (TnId5,) has been shown to have similar properties (Sheng, Z., Pan, B.-S., Miller, T. E.,and Potter, J. D. (1992) J. BioZ. Chem. 267,26407-25413). Regulation was not observed withthe COOH-terminal region alone (TnIlp~18z),with the NH,-terminal region alone (TnIl-s8),or with the NHz-terminallinked to the inhibitory region (TnIl-lle). Binding studies show that the NH,-terminal region of TnI interacts with the COOHterminal domain of TnC in thepresence of Ca2+or Mg2‘ and that the inhibitory plus COOH-terminal region of TnI (TnIlos_18z) interacts with the NH,-terminal domain of TnC in a Ca2+-dependentmanner. Based on these results we propose a model for the Ca2+-induced conformational change. In ourmodel the NHz-terminal domain of TnI is anchored strongly to the COOH-terminal domain of TnC in theabsence and presence of Ca2+ while the inhibitory and COOH-terminal regionsof TnI switch between actin-tropomyosin in the absence of Ca2+to binding sites in both N H 2 - and COOH-terminal domains of TnC in thepresence of Ca2+.
tropomyosin-binding subunit which is necessary for Ca2+-sensitive regulation at physiological ratios of troponin, tropomyosin, and actin. The interactions between these proteins have been studied extensively, as have their interactions with actin and tropomyosin (1-5). The TnI-actin interaction is thought to be the basis of the inhibition of the actomyosin ATPase (6-9). While TnI alone inhibits the actomyosin ATPase, tropomyosin potentiates this inhibition andphysiological regulation is only achieved by the whole troponin-tropomyosin complex (6-8). The mechanism of inhibition by troponin-tropomyosin may be different from that observed with TnI alone. The region of TnI responsible for its inhibitory activity was first localized by Syska et al. (9) to a CNBr cleavage fragment (CN4) corresponding to residues 96116. Other fragmentswhich did not contain thisregion did not possess inhibitory activity. Talbot and Hodges (10, 11)synthesized oligopeptides corresponding to this region anddetermined that amino acids 104-115 contain the minimum sequence necessary for inhibition. Inhibitory peptides, on their own, possess submaximal inhibitory activity suggesting that other regions of TnI may be involved, directly or indirectly, in the interaction withactin-tropomyosin (9-14). 11) TnC has two low-affinity Ca2+-specificbinding sites (I and located in its NH2-terminal domain and two high-affinity Ca2+/ Mg2+-bindingsites (I11 and IV)located in its COOH-terminal domain (15, 16). Sites I11 and IV are believed to be always occupied by metal ions andprobably perform a structural role in binding TnC to the thin filament under relaxingconditions (17). Upon muscle stimulation, the transitionfrom the resting is initiated by calcium binding to sites to the contracting state I and I1 of TnC (18, 19).This is believed to modify the TnI-TnC interaction in such a manner as to remove TnI’s inhibitory action. TnC mutants in which sites I or I1 have been destroyed have reduced ability to restore force when incorporated into skinned muscle fibers (20). Syska et al. (9) identified two reTroponin regulates skeletal and cardiacmuscle contraction. gions of TnI which possess affinityfor TnC: the inhibitoryCN4 Troponin has three subunits: troponin I (Tn1)l which inhibits fragment and two NH2-terminal fragments corresponding to the actomyosin M$+-ATPase, troponin C (TnC) which removes residues 1-21 a n d 1 4 7 .Cross-linking and other studies have TnI inhibition upon binding Ca2+, and troponin T (TnT), the suggested that the TnI inhibitory region interacts withboth the structural COOH-terminal domain (21-26) and with the regu* This work was supported by grants from FundagHo de Amparo A latory NH2-terminal domain of TnC (26-28). Recently, Ngai an correspondPesquisa do Estado de SHo Paulo, Conselho Nacional de Pesquisa, Pro- and Hodges (14) demonstrated that oligopeptide grama de Apoio a0 DesenvolvimentoCientifico e Technolbgico, The ing to amino acids 1-40 of TnI can compete with TnI or TnI Rockefeller Foundation, and The Medical Research Council of Canada. inhibitory peptidein binding toTnC and diminishTnC’s ability The costs of publication of this article were defrayed in part by the to remove TnI inhibitionof the actomyosin ATPase. No role has payment of page charges. This articlemust therefore behereby marked been assigned to the COOH-terminal region of TnI (residues “advertisement” in accordancewith 18 U.S.C.Section 1734 solelyto 11C182, see Fig. 1) although this region is more conserved indicate this fact. 5 To whom correspondence should be addressed: Dept. Bioquimica, than the NHz-terminal half of the protein (29-35). Instituto de Quimica, Universidade de SHo Paulo, CP 20.780, CEP Although proteolytic and synthetic TnI peptides have been 01498,SHo Paulo SP, Brazil. Fax: 55-11-815-5579. The abbreviations used are: Tn,troponin; WT, wild-type; Dm, powerful tools in revealing information about these proteins’ functional domains, it is important t o perform studies with dithiothreitol;PAGE, polyacrylamide gel electrophoresis.
5230
Functional Domains of TnI
5231
The larger fragments. To this end, Sheng et al. (36) and our labo- change the Val157 codonto a stop codon to produce clone M13-R131. ratory (37) recently reported the bacterial expression of rabbit sequences of all mutated TnI cDNAs were confirmed by dideoxy DNA sequencing (43). and chicken fast skeletal TnIs. Sheng et al. (36) used an interConstructionof Plasmids for Expression of TnI Deletion Mutants-% nal restriction site to generate a TnI mutant (TnId57) lacking express the TnIdeletion mutants we used the T7-based PET expression the first 57 amino-terminal residues. This protein's inhibitory system (44).The NheI-EcoRI fragment of MI3 mpl8TnINhe was cloned properties were similar to those of wild-typeTnI and it retained between the same restriction sites of the PET-3a vector (44)to create Ca2+-dependentinteractions with TnC. TnId57'S Ca2+/Mg2+-de- pANlOaTnI, which encodes TnIlo3_18z,amino acids 103-182 preceded pendent interactions with TnC were abolished, suggestingthat by the tripeptide Met-Ala-Ser.The NdeI-BamHI fragment of M13-R131 the NHz-terminal domain of TnI is important for Ca2+/Mg2+- was cloned in between the same two sites of pET-3a, to create PET-R131 which encodes Tn11-156,the first 156 amino acids of TnI. The BamHIdependent interactions with TnC. EcoRI fragment of MI3 mpl8TnI*S was cloned in between the same two In this work we report the synthesis of recombinant deletion sites of PET-3a to create clone pETslOFXTnI*S. Deletion of the NheImutants of chicken fast skeletal troponin I corresponding to SpeI fragment resulted in clone pANI119TnI which encodes Tn112%182, residues 1-98 (TnIl-9s), residues 1-116(Tn11-116), residues amino acids 120-182 of TnI preceded by the tripeptide Met-Ala-Ser. 1-156 (Tn11-156), residues 120-182 (TnIlzo_lsz),residues 103- Deletion of the small NdeI fragment of pETslOFXTnI*S produced the 182 (TnIlo3-lsa),and a polypeptide in which the inhibitory re- clone pAC117TnI which encodes TnIl-l16, the first 116 amino acids of gion (residues 99-119) has been deleted (AI-TnI) (Fig. 1).We TnI. The BamHI-EcoRI fragment of M13 mpl8TnIB*A was cloned in have also expressedthe TnC NH,-terminal (residues 1-90, N- between the same sites of PET-3a to produce pETslOFXTnIB*A. Deletion of the small NdeI fragment produced pAIC99TnI which encodes TnC) and COOH-terminal domains (residues 88-162, C-TnC) the first98 amino acids of TnI plus a COOH-terminalArg. The and four TnC mutants in which each of the four calcium bind- BamHI-EcoRI fragment of M13mp18A99/119TnIwascloned in being sites have been sequentially disrupted by mutating the tween the same sites of PET-3ato produce clone pETslOF'XA99/119TnI. aspartic acids at position 1 of each site to alanine, a substitu- Removal of the small NdeI fragment produced pAI99/119TnI which tion knownto disrupt calcium bindingto EF-hands (38).These encodes AI-TnI, residues 1-98 followed by an Arg followed by residues mutants are similar to TnC mutants previously described and 120-182. Expression and Purification of Recombinant Tnls-Escherichia coli characterized in detail (17,20). We used these recombinant strain BLal(DE3)pLysS (44) was used to express WT-TnI,Td1-156, wild-type and mutant TnIs and TnCs, plus recombinant TnT, to TnIlo3_182, T n I l z ~ l s zand , AI-TnI whileBLal(DE3)pLysE(44) was used investigate the structural and functional interactions of the to express TnII-118and Tn11-98. Bacterial expression of recombinant different regions of TnI with actin-tropomyosin,TnC and TnT. TnIs was performed as described for recombinant wild-type TnI (37). We found that, like TnC,TnI may be dividedinto two function- Purification of WT-TnI and Tn11-156was as described for recombinant TnIlz,182, Tn11-116, TnIl-g8, and ally distinct regions. The NHz-terminal region of TnI plays a TnI (37). Cells expressing TnIlo3_182, structural role in the troponin complex as itis responsible for AI-TnI fromeach liter of culture were resuspended in 12.5 ml of 20 rn Tris-HC1 (pH 8.0), 1 m~ EDTA, 4% (v/v) glycerol, 1 rn phenylmethylbinding to the COOH-terminal domainof TnC in the presence sulfonyl fluoride, 10 rn 2-mercaptoethanol and lysed by freeze-thawof Ca2+or Mg2' and stabilizes the incorporation of TnT into a ing. Another 12.5 ml of the above buffer wasadded and the mixture was ternary complex. TheNHz-terminalregion is not necessary for made 10 rn in MgSO, and 10 &ml in DNase I and kept for 90 min on inhibitory activity norfor the reconstitution of a Ca2+-sensitive ice with frequent gentle agitation. EDTA (25 rn)was added, the mixthin filament. A TnI fragment containing the inhibitory and ture homogenized manually on ice with a Potter homogenizer (Thomas COOH-terminal regions possesses the major regulatory hnc- Scientific), and centrifuged (193,000 x g, 1 h, 4 "C). TnI1-98, Tn11-116, tions of the molecule as itinteracts with boththe NHz-terminal and Tn1103-182partitioned mostly to the insoluble fraction while Tn112%182and AI-TnI were found exclusively in the supernatant. In the and COOH-terminal domainsof TnC in a Ca2+-dependentman- case of TnIl-98,TnIl-I16,and TnIlo3_18z, the pellets were resuspended in ner, inhibits the actomyosin ATPase as well as wild-type TnI, 25 mVliterof original culture of 50 rn sodium acetate (pH 5.0), 8 M urea, and can be reconstituted into a functional Ca2+-sensitivethin 1 rn EDTA,10 m~ 2-mercaptoethanol, homogenized as above, and centrifuged (27,000 x g, 1 h, 4 "C).The supernatant contained the filament. solubilized proteins. In the case of TnIlz%18zand AI-TnI, the lysis supernatant was made 40% saturated in ammonium sulfate (0"C) by MuscleProteins-Chicken actin and myosinwere purified from the slow addition of 100% saturated ammonium sulfate solution. Prechicken pectoralis muscles as described (39, 40). Chicken cardiac cipitated proteins were removed by centrifugation (16,000 x g, 30 min, a-tropomyosin was purified by the method of Smillie (41) and was a gift 4 "C) and the supernatant brought to 60% saturation by the further addition of 100% saturated ammonium sulfate solution. Precipitated from Dr. Jesus A. Fern. Mutagenesis of the TnI cDNA-The EcoRI-BamHI fragment of M13- proteins were collected as above, resuspended in 20 mVliter of original R18 (37) containing the whole modified TnI cDNA was subcloned be- culture of 50 rn sodium acetate (pH 5.01, 1 rn EDTA, 10 rn 2-mercaptoethanol for Tn112%182 or 50 rn Tris-HC1(pH 7.01, 1rn EDTA, 10 tween the EcoRI and BamHI sites of M13 mp18 to produceM13 mpl8TnI. The XbaI-BamHI fragment of PET-R18 (37) was subcloned rn 2-mercaptoethanol for AI-TnI, and dialyzed against the same buffbetween the XbaI and BamHI sites of M13 mp18to produce M13-R75. ers, after which the TnIs precipitate. Proteins were collected by cenThese were used as templates for oligonucleotide-mediated site-directed trifugation (27,000 x g, 1h, 4 "C), resuspended in 12.5 mVliterof original culture of the above buffers plus 8 M urea, homogenized as above, mutagenesis (42). M13 mpl8TnI was mutated with oligo*S (5"GGGTand centrifuged as above. The supernatant contained the solubilized GCGCATGTAAACTAGTGCCATGCTGCG-3') to producecloneM13 mpl8TnI*S in which an SpeI site was created and codons Serl", Ala118, proteins. Tn11-116, TnIl-98, TnIlo3-182and TnI12,182 were loaded (200 and Asp'19 were changed to Stop117, Thr'l8,Ser1l9.M13 mpl8TnI was ml/h, room temperature) onto a 100-ml CM-Sepharose fast-flow column mutated with oligoB*A (5'-TGAGCCAGAAGCGGTAACCGCCTAG- equilibrated with 50 rn sodium acetate (pH 5.0), 8 M urea, 1rn EDTA, GGGCAAGTTCA-3') to producecloneM13 mpl8TnIB*A in which and 10 m~ 2-mercaptoethanol. TnIl-l16 was washed with 50 rn NaCl BstEII and AurII sites were created and the codonsLeug9, PheIoo, and eluted with a 50-300 rn NaCl gradient (500 ml). TnI,_,, was Asp'O', and LeuloZwere changed to A rp,Stop'", Prolo', and Pro'o2. eluted with a 0-100 m~ NaCl gradient. TnIlo3_lszwas washedwith 100 Mutagenesis of MI3 mplBTnIB*Awith oligo*S produced the clone M13 rn NaCl and eluted with a l o w 5 0 rn NaCl gradient. TnIlzwlszwas mpl8TnIB*A*S. Codons 99 to 119 were deleted (and replaced with an washed with 50 rn NaCl and eluted with 85 m~ NaC1. AI-TnI was equilibrated with 50 rn arginine codon) bydigesting M13 mpl8TnIB*A*S with BstEII and SpeI, loaded ontoa CM-Sepharose fast-flow column creation of blunt ends with S1 nuclease, and ligation; to produce clone Tris-HC1(pH 7.0), 8 M, urea, 1rn EDTA, 10 m~ 2-mercaptoethanoland M13 mp18A99/119TnI. M13mpl8TnI was mutated with oligo Nhe (5'- eluted with a G140 m~ NaCl gradient. The purified proteins were GCTG'ITTGCTAGCCGTGGCAAGTTCAAGCGTCCACC-3')to intro- dialyzed against 10 rn imidazole-HC1 (pH 7.0), 1 M KCl, 2 rn MgCl,, duce an NheI site at codons 101 and 102 and to substitute two rare AGG 10 rn 2-mercaptoethanol and stored at -20 "C. All protein concentraArg codonsat positions 103 and 108 with the more common CGT codons tions were determined by the method of Hartree (45). Final yield of the different TnIs ranged from 5 to 20 mg of pure proteiditer of culture. (37),thereby producing clone M13 mpl8TnINhe. M13-R75 wasmutated with oligoV157* (5'-CTCCGCGATTAGGGTGACTG-3') in order to NH,-terminal amino acid sequencing of the purified proteins by Edman MATERIALSANDMETHODS
5232
Functional Domains of TnI
degradation was performedon an Applied Biosystems Model473A Protein Sequencer. RecombinantTnCs-Recombinantchicken skeletal TnC was expressed (37) and purified (46) as described. The chicken fast skeletal TnC cDNA (47) was used as a template for site-directed mutagenesis to produce the calcium-binding deficient mutants D30ATnC, D66ATnC, DlOGATnC, and D142ATnC in which aspartic acid codons at positions 30, 66, 106, and 142 were changed to alanine codons. Mutated TnC cDNAs were cloned into pET-ga, expressed, and purified as previously described for wild-type chicken TnC (37). Recombinant NH,-terminal domain (N-TnC,amino acids 1-90) and COOH-terminaldomain of TnC (C-TnC,amino acids 88-162) were expressed using the chicken skeletal muscle cDNA (37). The details are described elsewhere., Recombinant TnT-Site-directed mutagenesis was used to create an NdeI site at codon Met1 of the chicken fast skeletal TnT-3 cDNA (48). The cDNA was clonedinto NdeI-EcoRI cut pET-3a, expressed in E. coli strain BL2l(DEB)pLysSand purified with a procedure similar to that used for the expression of TnTs from rat and rabbit (49, 50). Cells expressing TnT were collected and resuspended in 25 mVliter of the original culture, 50 m~ Tris-HC1 (pH 8.0), 5 m~ EDTA, 1m~ phenylmethylsulfonylfluoride,lysed ina French Press, and centrifuged (193,000xg, 1h, 4"C). TnTis found mostlyin the supernatant.Sucrose was added to a final concentration of 25 g/lOO ml and then solid ammonium sulfate was added to 33% saturation (0 "C). Precipitated proteins were removed by centrifugation (16,000 x g, 30 min, 4 "C), resuspended in 30 ml of 50 m~ Tris-HC1 (pH 8.01, 1 m~ EDTA, 5 m~ 2-mercaptoethanol, 1 m~ phenylmethylsulfonyl fluoride, and dialyzed against the same buffer, after which the TnT precipitates. TnT was resuspended by the addition of solid urea to a final concentration of 6 M and loaded (200 ml/h, room temperature) onto a 100-ml DEAE-Sepharose fast-flow column equilibrated with 50 m~ Tris-HC1 (pH 8.0), 6 M urea, 1m~ EDTA, and 1m~ 2-mercaptoethanol.TnT was eluted with a 0-450 m~ NaCl gradient (500ml). Fractions containing TnT were dialyzed against 20 m~ imidazole-HC1 (pH 7.0), l M KCl, 2 m~ MgCI,, 10 m~ 2-mercaptoethanol, and stored at -20 "C. The identity of the purified TnT was confirmedby NH,-terminal sequencing by Edman degradation. Gel Electrophoresis-SDS-polyacrylamide gel electrophoresis was performed as described (51). TnC-TnI complexes werevisualized on 6 M urea-polyacrylamide (52) or 10%glycerol-polyacrylamide (53) gels where their mobilities differ significantly from free TnC, which has a relatively high electrophoretic mobility, and free TnI, which remains at the origin. To prepare TnI-TnC complexes forurea or glycerol gel electrophoresis, recombinant TnIs and TnCs were mixed in a mole ratio of 2:l and diluted with 10 m~ 2-mercaptoethanol to a final KC1 concen. of these complexes werediluted further with tration of 300 m ~Aliquots 1.5 volumesof 41.5 m~ Tris, 133 m~ glycine (pH 8.61, 2 m~ D m , 0.02% bromphenol blue, plus either 10 M urea or 16.7% (v/v) glycerol plus either 10 mM CaCl, or 10 m~ EDTA or 50 mM MgClJl m~ EGTA to the final concentration indicated in the figure legends. Aliquots (20 pl) of these mixtures were analyzed in 8% polyacrylamide gelscontaining 25 m~ Tris base, 80 m~ glycine (titrated to pH 8.6 with HCI) plus either 6 M urea or 10% (v/v) glycerolplus either 0.5 m~ CaCl, or 0.5 m~ EDTA or 10 m~ MgClJl m~ EGTA. The reservoir buffer was the same as the gel buffer exceptthat it contained no urea or glycerol. , Actin-binding Assays-Actin (23.8 m), tropomyosin (6.8 p ~ ) and recombinant TnIs (6.8 or 23.8 p ~ were ) combined in 10 m~ imidazoleHCl (pH 7.0), 200 m~ NaCl, 2.5 m~ MgCl,, 25 m~ 2-mercaptoethanol. Binding of the various TnI fragments to actin or actin-tropomyosin was assayed by centrifugation of the mixtures in an Airfuge (Beckman)for 20 min at 23 p.s.i. (120,000 x g). The mixtures before centrifugation and the supernatant andpellets aRer centrifugation were analyzed by SDSpolyacrylamide gelelectrophoresis. Doponin Reconstitution-Recombinant TnIs, TnC, and TnTwere combined in a 1:l:l molar ratio in 4.6 M urea, 25 m~ Tris-HC1 (pH 8.0), 1 M KCI, 0.5 m~ CaCl,, 10 m~ 2-mercaptoethanol, 0.01% NaN,, and dialyzed against bufferA (4.6 M urea, 50 m~ Tris-HC1 (pH 8.01, 1M KCI, 50 p~ CaCl,, 10 m~ 2-mercaptoethanol,0.01% NaN3).The urea andKC1 concentrations were reduced stepwise by the following changes of dialysis buffers: 1) buffer A containing 2 M urea; 2) buffer B (20 m~ imidazole (pH 7.01, 1 M KCl, 50 p~ CaCI,, 10 m~ 2-mercaptoethanol, 0.01% NaN3); 3) buffer B containing 100 m~ KCI; and 4) 20 m~ imidazole-HC1 (pH 7.0), 6.5 m~ KCI, 3.5 m~ MgCl,,50 p~ CaCl,, 2 m~ dithiothreitol, 0.01%NaN3. The binary WT-TnI-TnC and Tn1103-182Li, M. X., Chandra, M., Pearlstone, J. R., Racher, K. I., Trigo-Gonzalez, G., Borgford, T., Kay, C. M.,and Smillie, L. B. (1994)Biochemistry, in press.
TnC complexes used in Figs. 5 and 6 were also reconstituted in this manner. After dialysis the reconstituted complexes(5-12 p ~ were ) stored a t -70 "C. After centrifugation (12,000 x g, 10 min) to remove precipitated proteins, the reconstituted complexes were analyzed by passage through a Superdex-75 HR 10/30 gel filtration column (Pharmacia LKB Biotechnology Inc.) equilibrated with 20 m~ imidazole-HC1 (pH 7.01, 100 m~ KCI, 3.5 m~ MgCI,, 50 p~ CaCl,, 5 m~ 2-mercaptoethanol, 0.01%NaN,. The flow rate was 0.5 muminand eluting proteins were detected by monitoring A,,, and collected in 0.2-ml fractions. Proteins were precipitated with trichloroaceticacid foranalysis by SDSPAGE. MgZ+-ATPase Measurements-Actin (4 p ~ ) tropomyosin , (1.16 p ~ ) , recombinant TnIs (see figure legends for concentrations), recombinant TnC (see figure legends for concentrations),reconstituted troponin com, myosin (0.2 p ~ were ) combined on ice in 20 m~ plexes (1.16 p ~ ) and imidazole-HC1 (pH 7.0), 60 m~ KC1 (20.4 m~ KC1 for reconstituted troponins), 3.5 m~ MgCl,, 1 mM D m , 0.5 m~ EGTA. In Fig. 3, 0.6 m~ CaCI, was added to give a free Ca2+concentration of 0.1 m ~ in; Fig. 4, CaCl, was added to give the free CaZ+concentrations indicated. The samples were preincubated at 25 "C for 10 min, then sodium ATP (pH . 25 (Figs. 2 and 3) or 10 min (Fig. 4), 7.0) was added (2 m ~ ) After liberated phosphate was determined (54). Control experiments demonstrated that theATPase rate under these conditions wasconstant for 30 (Figs. 2 and 3)or 15 min (Fig. 4). RESULTS Recombinant TnI Fragments-The amino acid sequence of recombinant chickenWT-TnI (37) is presented inFig. lA and a schematic representationof the recombinant TnI deletion mutants is shown in Fig. Ut. NHz-terminal sequencing of t h e recombinant TnIs by Edman degradation confirmed the identity of the purified proteins. As with WT-TnI (37) and native chicken fast skeletal muscle TnI (291, all of the recombinant TnI deletion mutants had their initiation formyl-methionine removed by the bacterial cell. Fig. 1C shows an SDS-polyacrylamide gel of the purified recombinant TnIs. These proteins, especially those containing the NHz-terminal region, require in absence of TnC. high ionicstrength to remain in solution the Regulation of Actomyosin ATPase-Fig. 2 shows the inhibition of the actomyosin ATPase by the different TnI fragments in the presence of tropomyosin. Fragmentsthat do not containthe inhibitory region (Tn11-98,TnIlzcrlsz, and AI-TnI) do not show inhibitory activity even when present in a 2:l molar ratio with actin. Fragments that contain the inhibitory region were capable of inhibiting the ATPase. Fragments TnIlos-lsz a n d TnI1156 inhibited the ATPase in a concentration-dependent manner that is indistinguishable from WT-TnI. These three proteins inhibit 80-90% of the actomyosin ATPase at a 1:l molar ratio with actin. The concentration necessary to obtain 50% inhibition was 1.5 p.Higher molar ratiosof TnIl-116:actin are necessary to obtain maximum inhibition. A 2.5 molar excess of Tn11-116 overactin resulted in only 70% inhibition of t h e ATPase. At 1:lmolar ratioof TnI1-116:aCtin (4 p Tn11-116) only 50% of the maximum inhibition was observed. These results show that the presence of the inhibitory region is essential for inhibition and that sequences to the COOH-terminal side of this region are necessary for maximum inhibition. It is interesting to notethat the curves of the strongly inhibiting proteins are slightly sigmoidal in nature (see also Fig. 2A i n Ref. 37); the maximum slopes inthe curves being achievedat intermediate TnI concentrations instead of at t h e lowest concentrations.Extrapolation of the linear portions of these curves gives maximum inhibitionat between 2 a n d 3 p inhibitor, that is, at less than one molecule of inhibitor per moleculeof actin. This may be indicative of a cooperative inhibitory effect between neighboring TnI molecules a n d o r neighboringTnI-boundactin monomers as the sites onthe actin filament become saturated. Maximum inhibition with less than 1:l Tn1:actin has been reported before (11, 13, 36, 37, 55). To determine whether inhibitory activity correlates with the
5233
Functional Domains of TnI I0
1
30
20
50
40
6()
SDEEKKRRAATARRQHLKSAMLQLAVTEIEKEAMKEVEKQNYLAEHCPPLSLPGSMQEL 70
IO0
90
80
1 IO
120
QELCKKLHAKIDSVDEERYDTEVKLQKTNKELEDLSQKLFDLRGKFKRPPLRRVRMSADA 130
150
140
160
170
180
MLRALLGSKHKVNMDLRANLKQVKKEDTEKEKDLRDVGDWRKNIEEKSGMEGRKKMFEAGES
B
I
103
I
I
C
I
C
120
C
I
182 1 82
I 1r- 1
L””
4
- -
I
”
c
--*
a “”
-
FIG.1. A, amino acid sequence of recombinant chicken fast skeletal TnI (37). The inhibitoryregion (amino 103-116) acids has been underlined. B , schematic representationof recombinant TnI deletion mutants. We constructed three COOH-terminal truncations of TnI. Tn11-ls6contains the NH,-terminal region, the inhibitory region, and thefirst half of the COOH-terminal region up to residue 156. Tn11-l16,contains theNH,-tednal region plus the inhibitory region up to residue 116. Tnll-9Rcontains the NH,-terminal region uptoresidue98. We also constructed two NH,-terminal truncations of TnI. Tnll,3_,Rzbegins a t residue 103 and contains the inhibitoryregion and the COOH-terminalregion. T ~ ~ I Z C L I R Z begins a t residue 120 and contains only the COOH-terminal region. Finally we constructed AI-Tnl, in which the inhibitory region, residues 99-119, was deleted and residues 98 and 120 are separated by a single arginine.C , SDS-polyacrylamide gel of purified recombinant TnI fragments and TnC mutants used in this study. Recombinant TnCs: N , N-TnC (the NH,-terminal domainof TnC residues1-90]; C , C-TnC (the COOH-terminaldomain of TnC residues88-162); WT,wt-TnC; D30A (Asp 30 substitutedby Ala), D66A, D106A, and D142A are four TnC mutants where theAsp residue present in thefirst coordination position of each of the four metal binding loops was substituted for an Ala. Recombinant TnIs: WT, WT-TnI ( M , 21,106); AI, AI-TnI ( M , 18,721); Tn11-1s6 ( M , 18,091); TnIl-llfi(M,13,545); TnIl_,8 ( M , 11,433); TnIlo3_,R2( M , 9,526); Tn112CL182 ( M , 7463). Molecular mass markers( M ) of 92.5, 66,45, 31, 21.5, and 14.4 kDa are also shown. Wild-type TnI and the TnI fragments migrate SDS-PAGE in with apparent molecular masses greater than their calculated molecular mass.
ability to bind actin we tested each TnI fragmentfor its ability to co-sediment with actin and actin-tropomyosin in a qualitative sedimentation assay. Using actin:TnI fragment molar ratios of 7:2,WT-TnI, Tn11-156, Tn1103-182, and Tn11-116 all cosediment with actin or actin-tropomyosin in conditions where the proteins alone do not sediment (data not shown). Tn11-98, TnIlzo_18z,and AI-TnI do not sediment with actin or actintropomyosin under these conditions (data not shown). These results correlate well with inhibitory activitiesof the TnIfragments (Fig. 2). We next investigated the ability of TnC to remove the inhi-
bition of WT-TnI, Tn1103-182r Tn11-156, and TnI,-116 in the presence and absence of Ca2+. In these experiments, we used an Tn1:actin ratio that produced maximal inhibition in the absence of TnC and titrated the mixture with TnC in the absence and presence of calcium up to a TnC:TnI ratio of 2:l. The TnC induced release of WT-TnI and Tn1103-182inhibition is dependent on the presence of calcium (Fig. 3.4). The differences between the ATPase activity inthe presence and absence of calcium is the same for WT-TnI and TnIlo3-18z. However, there is a smalldifference in their sensitivities to TnC addition, as reflected by the maximal ATPase activity achieved at 1:l
Functional D(gmains of TnI
5234
hibit the ATPase to 30% at pCa 9.0 and activate the ATPase to 120%at pCa 4.5. The midpoint of this transitionoccurs at pCa 100 6.3.TnI103-182, in which the NHz-terminal region is absent, behaves exactly like wild-type TnI (Fig. 4A). At an actin: troponin ratio of7:2, the inhibitory and activation effects of 80 K these complexes are dependent on the presence of TnT since reconstituted binary complexes lacking TnT inhibited poorly and did not activate the ATPase (Fig. 4B). This is expected since WT-TnI only inhibits 25% of ATPase activity at anactin: TnI ratio of 7:2 (1.16 WT-TnI in Fig. 2). A partial deletion of the COOH-terminal region of TnI (Tr1I1-~56,Fig. 4A)results in a reduction of inhibition at low calcium concentrations, a small increase in activation at high calcium concentrations and an increase in calcium sensitivity (midpoint of transition at pCa 20 6.7). Complete removal of the COOH-terminal region of TnI (TnI1-116, Fig. 4A) or the inhibitory region (AI-TnI, Fig. 4B) results in troponin complexes that are insenkitive to calcium 0 2 4 6 8 10 concentrations and that activate the ATPase 25% above the inhibitor concentration (pM) actin-tropomyosin level. The NHz-terminal region of TnI FIG.2. Inhibition of actomyosin ATPase by recombinant WT- (Tn11-98, Fig. 4B) when incorporated in a troponin complex TnI and T n I fragments. Steady-state actomyosin Mg2+-ATPasewas activates the actomyosin ATPase 50% above the actin-tropomeasured as a function of added TnI fragments. The results were ex- myosin alone at low calcium concentrations and activates 30% pressed as a percentage of the ATPase activity obtained in the absence at high calcium concentrations. The midpoint of this transition of TnI. The average * S.E. of a t least threeindependent determinations a t each TnI concentration are shown. Assay conditions: actin (4 p), occurs at much higher calcium concentrations (pCa 5.6). A trotropomyosin (1.16p), WT-TnI or TnI fragments (0-10 p), and myosin ponin complex reconstituted with TnIlzo_lsz does not signifi(0.2 p)in 20 IIM~imidazole-HC1 (pH 7.01, 60 m KCl, 3.5 l l l ~MgC12, 1 cantly inhibit or activate the actomyosin ATPase at the Ca2+ m D m ,0.5 II~MEGTA, 2 m Na2ATP. Under these conditions, 100% concentrations tested (Fig. 4B). These results confirm the reactivity corresponds to 2.2 s-l. Symbols:WT-TnI, 0; Tn11-,66,A; sults obtained with TnI-TnC complexes, and suggest that the TnI1-116, 0 TnIl-s8, V; AI-TnI, Y T~IIOSISZ, 0;TnIlz+-18z... inhibitory and COOH-terminal regions of TnI are involved in the calcium-regulated inhibitory activity of TnI. TnC:TnI ratio in the absence and presence of calcium. For To investigate to what extent the regulatory ability of the WT-TnI, the addition of TnC in the absence of calcium results different TnI fragments could becorrelated with their ability to in a small release of inhibition while in thepresence of calcium form stable complexes with TnC and TnT,we analyzed the the release of inhibition is complete. In comparison, the inhibi- reconstituted complexes in analytical gel filtration columns. tory effect of Tn1103-1~2, in theabsence of calcium, is completely TnIs containing the NHz-terminal region (Td-156, Tn11-116, insensitive to the presence of TnC while in the presence of Tn11-98, and AI-TnI)formed stableternary complexes that calcium a large butincomplete release of inhibition (80%) was eluted as single peaks with apparent molecular weights larger than that of WT-TnI-TnC binary complex (Fig. 5 and data not obtained (Fig. 3A). In contrast to WT-TnI and Tn1103-182,deletion mutants in shown). SDS-PAGE gels of the proteins eluting at the different which part or all of the COOH-terminal region was removed peaks confirmed the presence of the three troponin subunits (Fig. 5 and data not shown). The product of the reconstitution have lost the ability to maintain inhibition when complexed with TnC in theabsence of calcium (Fig. 3B 1. Both Tn11-116 and ofTn1103"1s2 with TnC and TnT elutes as one peak (Fig. 5). Tn11-156 have their inhibitory effect removed by TnC in the Analysis bySDS-PAGE shows that this peak contains only absence or presence of calcium at a 1:l molar ratio of TnC:TnI TnIlo3-lsz and TnC (Fig. 51,showing that TnT doesnot remain passage through the (Fig. 3B). The inhibition of Tn11-156can be completely released associated with TnC and TnIlo3-l~z during by TnC in the presence of calcium. It is important to note, column. This result is consistent with protection of lysine resihowever, that therelease in theabsence of calcium is not com- dues in theNHz-terminal region of TnI (residues 40-98) by TnT plete, reaching only 80% ATPase activity even at a 2:l molar (56). In addition, Cheung et al. (57) found that the free energy excess of TnC over TnI1-156. The small difference observed in coupling for the formation of troponin from the three binary the presence or absence of calcium in experiments with complexes was positive and large and that virtually all of the Tn11-156was completely abolished whenthe whole COOH-ter- Ca2+-inducedstabilization of the troponin complex is due to a minal region was deleted in Tn11-116 (Fig. 3B). The maximum tightening of the TnI-TnC interaction. This suggests that the release of inhibition in both conditions with Td1-116 is 80% and TnI-TnT and TnC-TnT interactions may be destabilized when may be due to the higher TnIlPll6 concentration used to obtain the binary complexes are incorporated into the three-subunit troponin structure (57). If the TnI-TnT interaction is diminmaximum inhibition (Fig. 2). Taken together, the results in (Fig. 5 and Ref.561, then Figs. 2 and 3 show that while deletions in the COOH-terminal ished inthe case ofTnI103-182 region of TnI do not abolish its inhibitory activity, the mainte- the inclusion of TnT into a ternary troponin complex with nance of inhibition in the presence of TnC in the absence of TnIlo3-18zand TnC maybe energetically unfavorable under our Ca2+ is dependent on the presence of both the inhibitory conditions. These results (Figs. 4 and 5) show that the NHz-terminal COOH-terminal regions of TnI. In addition, the deletion of the first 98 residues of TnI makes its inhibition completely insen- region of TnI is necessary for the incorporation of TnT into a stable troponin complex in vitro but is not necessary for the sitive to the presence of TnC in the absence of Ca2+. Next, we analyzed the Ca2+-dependentregulatory properties calcium regulation of the inhibitory activity of TnI in a reconof reconstituted actin-tropomyosin thin filaments with tropo- stituted thin filament. In contrast, theCOOH-terminal region nin complexes containing the different TnI fragments. Fig. 4 and the inhibitory region of TnI are necessary for calciumregushows the ATPase activity as a function of calcium concentra- lation but cannot, on their own, form a ternary complex with tion (pCa). Ternary troponin complexes containing WT-TnI in- TnC and TnT. Since a reconstitution mixture containing Y
5235
Functional Domains of TnI
tB
t 0.01.0
0.5
1.5
0.0
2.0
0.5
1.5
1 .O
2.0
Molarratio(TnC/inhibitor)
Molarratio(TnC/inhibitor)
of TnI fragments.The effect of TnI fragments (A, WT-TnI and TnIloslsz; B , FIG.3. Effect of TnC and calcium on the inhibitory action was measured as a function of added TnC in the absence (filledsymbols) and Tn11_156and TnIl-I16) on the steady-state actomyosin Mg2+-ATPase presence (hollow symbols) of calcium. The results are expressed as a percentage of the ATPase activity obtained in the absence of TnI and TnC. The average * S.E. of at least three independent determinations at each TnC concentration is shown. Assay conditions: actin (4 p), tropomyosin (1.16 p),TnI (4 p for WT-, TnIlo,182, and TnIl-l,6 and 10 p d for TnIl-l16), TnC (0-20 p ~ and ) myosin (0.2 phf) were combined in 20 m imidazole-HC1(pH 7.0), 60 m KCl, 3.5m MgCl,, 1 m D’IT, 0.5 II~MEGTA, 2 m Na,ATP, with or without 0.6m CaC1,. Under these conditions, 100% activity corresponds to 2.2 s-l. Symbols: A: WT-TnI, H, 0;Tn1103-182,A, A. B : TnIl-156,H, 0; Tn11_116, A, A.
G
v
120
>r
Y
.-
100
Y
0
80 0
a 0 n I-
60
0 Tn(1-98)
