Mapping the domain of troponin T responsible for the activation of actomyosin ATPase activity. Identification of residues involved in binding to actin.

The in vitro Ca(2+) regulation of the actomyosin Mg(2+)-ATPase at physiological ratios of actin, tropomyosin, and troponin occurs only in the presence of troponin T. We have previously demonstrated that a polypeptide corresponding to the first 191 amino acids of troponin T (TnT-(1-191)) activates the actomyosin Mg(2+)-ATPase in the presence of tropomyosin. In order to further characterize this activation domain, we constructed troponin T fragments corresponding to residues 1-157 (TnT-(1-157)), 1-76 (TnT-(1-76)), 77-157 (TnT-(77-157)), 77-191 (TnT-(77-191)), and 158-191 (TnT-(158-191)). Assays using these fragments demonstrated the following: (a) residues 1-76 do not bind to tropomyosin or actin; (b) residues 158-191 bind to actin cooperatively but not to tropomyosin; (c) the sequence 77-157 is necessary for troponin interaction with residue 263 of tropomyosin; (d) TnT-(77-191) on its own activates the actomyosin ATPase activity as described previously for TnT-(1-191). TnT-(1-157), TnT-(1-76), TnT-(77-157), TnT-(158-191), and combinations of TnT-(158-191) with TnT-(1-157) or TnT-(77-157) showed no effect on the ATPase activity. We conclude that the activation of actomyosin ATPase activity is mediated by a direct interaction between amino acids 77 and 191 of troponin T, tropomyosin, and actin.


INTRODUCTION
The contractile interaction between myosin and actin in skeletal muscle is regulated by changes in intracellular Ca 2+ concentration and by the effects of such changes on the interactions between thin filament proteins. In skeletal muscle, tropomyosin (Tm) and the troponin complex (Tn) regulate this process. The troponin complex is formed by three subunits: troponin C (TnC), troponin I (TnI) and troponin T (TnT) (1). In the absence of Ca 2+ , the TnI subunit inhibits the actomyosin ATPase activity. The binding of Ca 2+ to the regulatory sites of TnC removes TnI inhibition and leads to activation of actomyosin ATPase activity (for review see ref. 2 -4). In addition to the central role played by TnI and TnC in this Ca 2+ -dependent regulation, an increasing amount of evidence points to the important role of TnT in the regulation of muscle contraction. Besides its role in attaching the troponin complex to the thin filament (for review see ref. 5), TnT is required for the full inhibition of actomyosin ATPase activity in the absence of Ca 2+ in vitro at physiological Tn:Tm:actin molar ratios (6)(7)(8). Furthermore, only in the presence of TnT does the binding of Ca 2+ to TnC lead to an increase of the level of the actomyosin ATPase activity above the level of activity observed in the presence of tropomyosin and actomyosin (1,(6)(7)(8).
The elongated shape of TnT (9) and its Ca 2+ -dependent and Ca 2+ -independent interactions with the thin filament (10) indicate its importance in the control of the position occupied by tropomyosin in the thin filament, a central theme in the proposed molecular mechanisms of muscle contraction regulation (11)(12). Experiments performed with chymotryptic fragments T1 and T2 of TnT (13) demonstrated that the NH 2 -terminal fragment T1 binds strongly to the COOH-terminal region of tropomyosin (14,15,16). The affinity of T1 for tropomyosin is mediated by residues present in cyanogen bromide fragment CB2 (17). The role of the NH 2 -region of TnT adjacent to CB2 in the regulation of muscle contraction is not clearly defined, although it has been suggested that this region overlaps the head-to-tail junction region of two adjacent tropomyosins (18). The COOHterminal region of TnT (chymotryptic fragment T2) binds to the TnI/TnC dimer and to tropomyosin (15,(19)(20)(21). The interaction of T2 with tropomyosin in the presence of TnI and TnC is sensitive to Ca 2+ concentration. In the absence of Ca 2+ , T2 binds to a region near to amino acid 190 of tropomyosin (14). In the presence of Ca 2+ , it is believed that modifications in the interaction between the T2 region and TnI/TnC lead to detachment of this region of TnT from tropomyosin (10,22).
Using recombinant fragments of TnT, our group has previously shown that in the absence of the TnI/TnC dimer, an isolated polypeptide corresponding to the first 191 amino acids of skeletal chicken muscle TnT (TnT1-191) is able to activate the Mg 2+ -ATPase activity of actomyosin, whereas isolated wild type TnT is not (22). The level of activation observed (~30%) was similar to those levels observed in experiments performed under similar conditions with the whole Tn complex in the presence of Ca 2+ , indicating that the amino acid sequence between residues 1 and 191 of TnT encompasses an activation domain. This domain must be responsible for the activation of actomyosin ATPase at high Ca 2+ concentrations. To explain the observation that TnT1-191 is able to activate the ATPase activity while full-length TnT is not, we proposed a refinement of the original two-site binding model of Pearlstone and Smillie (10) used to describe troponin-binding to the thin filament. In the refined model (22), TnT was divided into three domains: a) an activation domain localized in the NH 2 -terminal region of TnT; b) a central domain necessary for the full inhibition by TnI in the absence of Ca 2+ ; and c) a COOH-terminal domain that anchors the TnI/TnC dimer to the filament. We proposed that binding of Ca 2+ to TnC would promote the dissociation of the globular domain of troponin (formed by TnC, TnI and the COOH-terminal region of TnT) from the thin filament, thereby abolishing the inhibitory interactions. This would release the activation effects of the NH 2 -terminal region of TnT.
In the present study we further map the activation domain of TnT in order to better understand its interactions with the thin filament. We constructed TnT fragments encompassing the region between the amino acids 1-191. Functional mapping was performed by examining the effect of TnT fragments on the actomyosin Mg 2+ -ATPase activity. Direct interactions of these fragments with tropomyosin and actin were also studied.

Expression and purification of recombinant TnT (TnTwt) and fragments TnT1-191 and
TnT157-163 were carried out as previously described by Malnic et al. (22).

TnT158-191 Synthesis
The peptide TnT158-191 and the ABZ-containing fragment were synthesized using the solid phase method (28). Optimized coupling reaction conditions were applied according to peptidyl-resin-solvation theory (29). The peptides were purified by semipreparative reverse phase-HPLC column and characterized by amino acid analysis, analytical HPLC and mass spectrometry.
The amplified product was subcloned into pET-3a previously digested with BstEII and EcoRI. The nucleotide sequence was confirmed (26). Expression and purification of the mutant tropomyosin containing 5-hydroxytryptophan at position 263 (5OH263W) was performed as described by Farah and Reinach (32). 5OH263WTm concentration was determined according to Hartree (27).

Other Muscle Proteins
α-Tropomyosin was purified from chicken heart as described by Smillie (33). Actin (34) and myosin (35) were prepared from chicken pectoralis major and minor muscles.
Recombinant TnI and TnC were expressed and purified as described (36). Reconstitution of the troponin complex was performed as described previously (6).

CD Spectra
TnT fragments dialyzed against 10 mM sodium phosphate (pH 7.0), 100 mM KCl and 1 mM DTT were used in circular dichroism studies performed on a Jasco 720 spectropolarimeter at 20 o C. The apparent secondary structure content was calculated as described (37,38).

Actomyosin ATPase Assays
The ATPase activity measurements were performed by mixing actin (4 µM), tropomyosin (0.57 µM), TnT or TnT fragments (concentrations are indicated in the figure legends) and myosin (0.2 µM) in 20 mM imidazole-HCl (pH 7.0), 3.5 mM MgCl 2 , 0.5 mM EGTA, 60 mM KCl, 1 mM EDTA and 1 mM DTT as described (7,22). Some assays were performed in the absence of tropomyosin and actin (see legend Figure 3 and Table 2). The samples were incubated for 15 minutes at 25 o C before initiating the reaction by the addition of 2 mM Na 2 ATP (pH 7.0). Reactions were stopped after 15 minutes and ATPase activity was determined by measuring the amount of inorganic phosphate released (39).

Effects of Actin and Tropomyosin on the Fluorescence Spectra of ABZTnT158-191
The effect of the presence of actin (1 ìM) and tropomyosin (1 ìM) on the emission spectra of ABZTnT158-191 (1 ìM) was determined in a solution containing 20 mM MOPS (pH 7.0), 3.5 mM MgCl 2, 1 mM EDTA, 1 mM DTT and 60 mM KCl or 1M KCl (specified in figure legends). Samples were excited at 319 nm. The emission spectra were monitored at 25 o C with slit widths of 5 nm. All spectra were corrected for dilution.

Gel Filtration Assays
Gel filtration of mixtures of tropomyosin and ABZTnT158-191 were carried out using a Superose 6HR 10/30 column (Pharmacia Biotech) coupled to an AKTA-Pharmacia were identified using 15% SDS-PAGE (40). Fractions containing ABZTnT158-191 were identified by fluorescence spectroscopy as described above.

Effects of Troponin Complex and TnT Fragments on the Fluorescence Spectra of 5OH263WTm
The effect of the addition of isolated TnT fragments or troponin complexes reconstituted with wild type TnT or TnT157-263 (see figure 7 and Table 3 for concentrations) on the emission spectra of 5OH263WTm was determined in 25 mM MOPS (pH 7.0), 60 mM NaCl, 5 mM MgCl 2 and 1 mM DTT in the presence and in the absence of 7 ìM actin using a F-4500 Hitachi spectrofluoremeter at 25 o C. Samples were excited at 312 nm and the fluorescence detected between 320 nm and 420 nm using slit bandwidths of 5 nm. In parallel, binding was confirmed by analyzing TnT-5OH263WTm-actin or Tn-5OH263WTm-actin mixtures in co-sedimentation assays as described above except that 25 mM MOPS (pH 7.0) was used as a buffer instead of 20 mM imidazole-HCl (pH 7.0).

TnT Fragments
We have previously shown that residues 1-191 of TnT are able to activate the actomyosin ATPase in the presence of tropomyosin and in the absence of TnC and TnI (22). In order to identify the sequences responsible for this effect we constructed five fragments of chicken skeletal TnT-3 ( Figure 1). We produced three recombinant fragments based on the known chymotryptic fragment T1 (fragment TnT1-157) and on the cyanogen bromide fragments CB2 and CB3 (fragments TnT77-157 and TnT1-76 respectively) (42).
In addition, recombinant fragment TnT77-191 (amino acids 77-191) and a synthetic fragment TnT158-191 were produced. In contrast to wild type TnT, all five fragments constructed in this work were soluble at low ionic strength (60 mM KCl) (data not shown).
Insert Figure 2 and Table 1.

Activation Properties of TnT Fragments
The effect of the isolated TnT fragments on the actomyosin Mg 2+ -ATPase activity was studied in the presence of tropomyosin and in the absence of TnC and TnI. Our These results demonstrated that the activation domain is located between amino acids 77-191 of TnT.
Insert Figure 3 and Table 2 We also studied the effects of these fragments on actomyosin ATPase activity in the absence of tropomyosin (Table 2). Our results showed no activation caused by the presence of TnT1-191 and TnT77-191. In fact, we observed an inhibition of ATPase activity in these assays. In the absence of tropomyosin, the fragments inhibit to the same extent that full-length TnT does in the presence of tropomyosin. To exclude the possibility that activation activity of TnT77-191 and TnT1-191 is due to a non-specific effect, we analyzed the direct influence of TnT77-191 and TnT1-191 on the Mg 2+ -ATPase activity of myosin in the absence of actin. These results (summarized in Table 2) demonstrate that activation depends on the presence of both tropomyosin and actin.

Interaction of TnT Fragments with Actin and Tropomyosin
All TnT fragments constructed in this present work are soluble in 60 mM KCl, permitting the study of their interactions with thin filaments at physiological salt concentrations. Performing qualitative co-sedimentation binding assays with TnT1-76, TnT77-157 and TnT1-157, we confirmed that the region of TnT that encompasses

Insert Figure 5.
We could not demonstrate a direct interaction between ABZTnT158-191 and tropomyosin in the absence of actin. The addition of tropomyosin has no effect on ABZTnT158-191 fluorescence at 60 mM KCl ( Figure 5A). When a mixture of ABZTnT158-191 and tropomyosin were applied to a gel filtration column, tropomyosin and ABZTnT158-191 appeared as two distinct peaks ( Figure 5B). The peaks corresponding to tropomyosin and to ABZTnT158-191 eluted at the same positions as that of tropomyosin alone and ABZTnT158-191 alone, indicating a lack of interaction between tropomyosin and the region correspondent to residues 158-191 of TnT under the conditions tested. This result is in agreement with previous studies that showed that a fragment of TnT corresponding to residues 159-227 of rabbit skeletal TnT did not bind to immobilized tropomyosin (17).

Insert Figure 6.
To explore the binding of TnT158-191 to actin and to tropomyosin-actin, we performed quantitative co-sedimentation assays with increasing concentrations of ABZTnT158-191. In control experiments performed with ABZTnT158-191 alone, no significant amount of TnT fragment was detected in the pellet. Figure 6 shows the binding curve of ABZTnT158-191 to actin and to tropomyosin-actin. Assuming that the binding of TnT158-191 to actin occurs at a molar ratio 1:1, the calculated apparent dissociation constant is 8.1 X 10 -6 M -1 . Interestingly, the S shape of the binding curve shown in Figure 6 suggests that this binding is cooperative. The Hill coefficient was calculated to be 2.1. The binding curve of ABZTnT158-191 to tropomyosin-actin (molar ratio 1:7) was very similar to the binding curve of ABZTnT158-191 to actin ( Figure 6).

Effects of Troponin Complex and TnT fragments on the Fluorescence Spectra of 5OH263WTm
To study the region of tropomyosin that interacts with the amino terminal half of TnT, we produced a recombinant tropomyosin with an intrinsic fluorescent 5-hydroxytryotophan probe replacing the original glutamine at position 263 (5OH263WTm).
While 5OH263WTm binds to F-actin in co-sedimentation assays (data not shown), its emission spectrum is unaltered by this interaction (Figure 7). On the other hand, the addition of troponin to 5OH263WTm or actin-5OH263WTm causes a significant decrease in 5OH263WTm fluorescence intensity (Figure 7). Therefore, 5OH263WTm fluorescence can be used as a specific probe for the TnT-Tm interaction. Figure 7 and Table 3.  Table 3. These assays demonstrate that: i) while the binding of actin to 5OH263WTm had no effect on its emission spectra, significant decreases in fluorescence intensity were caused by the presence of wild-type troponin complex, TnT1-191 and TnT77-191, both in the presence and in the absence of actin.

Insert
These decreases were more significant in the presence of actin (~ 20%) than in its absence (~12%) ( Table 3) (Figure 7). iii) As expected, fragment TnT1-76 did not affect the fluorescence of 5OH263WTm since it does not bind to actin-Tm (Table 3). iv) All fragments of TnT that contain the region corresponding to fragment CB2 (TnT1-191, TnT1-157, TnT77-157 and TnT77-191) were able to bind to 5OH263WTm-actin in cosedimentation binding assays (data not shown). Surprisingly, all these TnT fragments except TnT77-157 (CB2) promoted a decrease in fluorescence intensity of 5OH263WTm.
These data suggest that the binding of TnT residues 77-157 (CB2) to tropomyosin-actin is modulated by neighboring amino acid sequences along the primary structure of TnT. Only in presence of these sequences is a change in the fluorescence intensity of 5OH263WTm observed. It is noteworthy that the two fragments that are able to activate the ATPase activity of actomyosin (TnT1-191 and TnT77-191) encompass both the tropomyosin binding (residues 77-157) and the actin-binding (residues 158-191) sites.

DISCUSSION
The aim of this study was to investigate the phenomena of activation of actomyosin ATPase activity. We studied the properties of fragments of TnT that encompass the region corresponding to the first 191 amino acid residues of TnT. The fact that all of the TnT fragments constructed here were soluble allowed us to study and map the interactions of the first 191 amino acid residues of TnT with tropomyosin and actin. In this study, we: i) confirmed that TnT is responsible for the activation of actomyosin ATPase activity; ii) demonstrated that the activation domain is located between the residues 77 and 191 of TnT; iii) demonstrated that there is an actin-binding site between residues 158 and 191 of TnT; iv) demonstrated that amino acid residue 263 of tropomyosin interacts with residues 77-157 of TnT.
In agreement with previous studies (15,17), our results demonstrated that the region between residues 77-157 (CB2) is a tropomyosin-binding site. However, our binding assays performed with TnT fragments and 5OH263WTm (Figure 7) suggested that neighboring amino acid sequences along the primary structure of TnT modulate this interaction. The hypothesis that the CB2 region (residues 77-157) interacts nonspecifically with tropomyosin and that CB3 (residues 1-76) confers specificity to binding of T1 (1-157) to tropomyosin has been suggested previously (44). These authors observed that CB2 interacts weakly and non-specifically with tropomyosin while the T1tropomyosin interaction is stronger and more specific. Our results suggest that residues 158-191 of TnT have a similar effect on the modulation of the interaction of amino acid residues 77-157 with tropomyosin.
Studies of the binding of TnT to actin have been impaired due the insolubility of full-length TnT. However, a set of studies has demonstrated that the presence of TnT increases the affinity of tropomyosin for actin (45 -47). Hill et al. (46) showed that a deletion of the first 69 residues of cardiac TnT did not affect this property, while the affinity was significantly reduced when the first 158 amino acid residues of cardiac TnT (50) demonstrated that the removal of the first 11 amino acids of tropomyosin, which results in a non-polimerizable tropomyosin that does not bind to actin in the absence of troponin, has no effect on the cooperativity of Ca 2+ -activation of the actomyosin ATPase.
This observation was interpreted as evidence that TnT would play a central role in the cooperativity of the activation process perhaps through its interactions with two adjacent tropomyosins. Studies using troponin complexes reconstituted respectively with cardiac TnT lacking the first 94 amino acids (51) and rabbit skeletal TnT lacking the first 45 amino acids (52), showed that the removal of the NH 2 -terminal extremity of TnT has no significant effect on the cooperativity of activation of the actomyosin ATPase. Since the fragments studied in the present study lack TnC and TnI binding sites (amino acids 216-263 of TnT) (22), assays of Ca 2+ regulation by the troponin complex were not performed.
However, our results are consistent with the notion that the CB3 region of TnT does not play an essential role in the activation process.
Interestingly, we observed that only fragments of TnT that preserve both the tropomyosin binding site (amino acids 77-157) (15,17)          Secondary structure composition was estimated (37,38) from the CD spectra shown in Figure 2. Values are expressed as percentages. * "Other" is the sum of the predictions for β-turn and "remainder" structures as defined in (37,38). In the parenthesis are previously reported values for cyanogen bromide fragments and chymotryptic fragments of rabbit skeletal TnT (20,43  The data shown are presented as percentage of ATPase activity observed for tropomyosin-actomyosin (100%  The data shown were obtained from the spectra presented in Figure 7 (Tn:5OH263WTm:actin = 3:1:7) and represent typical results obtained from at least 2 independent assays. ()*: Results obtained in assays performed at TnT77-157:5OH263WTm:actin = 10:1:7.
34 by guest on July 24, 2018