Binding properties of the calcium-activated F2 isoform of Lethocerus troponin C.

While in most muscles contraction is triggered by calcium effluxes, insect flight muscles are also activated by mechanical stretch. We are interested in understanding the role that the troponin C protein, usually the calcium sensor, plays in stretch activation. In the flight muscles of Lethocerus, a giant water bug often used as a model system, there are two isoforms of TnC, F1 and F2, present in an approximately 10:1 ratio. F1 TnC is responsible for activating the muscle following a stretch, whereas F2 TnC produces a sustained contraction, the magnitude of which depends on the concentration of Ca(2+) in the fiber. We have previously shown that F1 TnC binds only one Ca(2+) ion in its C-terminal domain and that interaction with troponin H, the insect ortholog of troponin I, is insensitive to Ca(2+). Here, we have studied the effect of Ca(2+) and Mg(2+) on the affinities of the interaction of F2 TnC with troponin H peptides. We show that the presence of two Ca(2+) ions, one in each of the globular domains, increases the affinity for TnH by at least 1 order of magnitude. The N lobe has a lower affinity for Ca(2+), but it is also sensitive to Mg(2+). The C lobe is insensitive to Mg(2+) as previously demonstrated by mutations of the individual EF-hands. The interaction with TnH seems also to have significant structural differences from that observed for the F1 TnC isoform. We discuss how our findings could account for stretch activation.

Muscle contraction is initiated by impulses from the motor neurons to the muscle that cause depolarization of the innervated fibers (see refs (1−3) for reviews). In synchronous muscles, an action potential triggers release of Ca2+ from an internal store within the fiber, the sarcoplasmatic reticulum. The released Ca2+ diffuses into the myofibrils, binds to control sites on the contractile filaments, and initiates contraction. This contraction is terminated when the sarcoplasmatic reticulum resequesters the released Ca2+ and reduces its cytoplasmatic concentration to a level below the threshold required for contractile activity.

Abbreviations: BAPTA, 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid; CD, circular dichroism; Kd, equilibrium dissociation constant; LB, Luria-Bertani; NMR, nuclear magnetic resonance; Tm, tropomyosin; Tn, troponin; TnC, troponin C; TnH, troponin H.

The Ca2+ sensor that regulates this process resides in the tropomyosin−troponin complex. In the absence of Ca2+, tropomyosin, a rodlike protein that extends over seven actin monomers, blocks the myosin-binding sites present on the thin filament (3−5). Troponin is a complex of three regulatory proteins: a Ca2+ binding protein (TnC), an inhibitory protein (TnI), and a tropomyosin binding protein (TnT) (6). TnC senses Ca2+ through up to four Ca2+ binding motifs (EF-hands) distributed across two globular domains, the N- and C-terminal domains (7−11). The C-terminal domain (C lobe) usually interacts with an N-terminal region of TnI and anchors it into a specific conformation (12−15). When the N-terminal domain (N lobe) of TnC is able to bind Ca2+, it undergoes a conformational transition that allows it to bind to the “switch peptide” of TnI (residues 115−131 of skeletal TnI) (12), initiating the tropomyosin movement that causes the generation of force.

In some muscles, however, such as in the flight muscles of many insects, contraction is not simply regulated by the Ca2+ switch. The wing beat frequency is too high for individual contractions of the flight muscles to be activated by Ca2+[16−19]. To adapt to this requirement, indirect flight muscles (IFM) have evolved two distinct forms of activation, which are finely tuned to the type of movement required (20,21): a stretch-activated mechanism (asynchronous contraction) is used for flying, whereas Ca2+ regulation remains important during the “warm-up” contractions that precede flight in large insects (synchronous contraction). During flight, stretch and Ca2+ activation are finely balanced, depending on the Ca2+ concentration produced by the intermittent nerve impulses (20−22).

Despite its importance for understanding the mechanisms of muscle contraction, relatively little is known about stretch activation and the way it coexists with synchronous regulation, although current evidence all seems to indicate that the muscle regulation in insects differs markedly from that of vertebrates. Most of our knowledge about insect muscles comes from studies of Lethocerus, a giant water bug of the Belostomatidae family, native to Southeast Asia, which is commonly used as a model system because its large muscle fibers are easily manageable. In Lethocerus, two isoforms of TnC coexist in the same myofibril, F1 and F2, in a ratio of ∼10:1 (23,24). F1 TnC, which has a single Ca2+ binding site in the C lobe, is responsible for activating the muscle following a stretch. F2 TnC has two Ca2+ binding sites, one in each lobe, and produces a sustained contraction, the magnitude of which is dependent on the Ca2+ concentration (20). In previous studies, we have shown that the C lobe of F1 TnC binds to peptides spanning the N-terminal and inhibitory regions of TnH [the insect equivalent of TnI (25)], although with very different affinities (26). The equilibrium dissociation constant for the complex with the N-terminal TnH peptide is in the nanomolar range, which is comparable to the values reported for the skeletal TnC−TnI complex (27,28). The affinity for the peptide from the inhibitory region is much lower, in the micromolar range, which is comparable to the affinity of the inhibitory peptide of skeletal TnI for the C lobe of TnC (29). Both interactions are essentially Ca2+ independent, and the N lobe of F1 TnC appears to play no role. It was therefore suggested that competition between two distinct regions of TnH, which alternately occupy the C lobe of F1 TnC, could be the basis of stretch activation. The relative affinity of F1 TnC and F2 TnC for the whole TnI sequence in TnH, in the presence of Ca2+, has been determined (24). The overall affinity of F2 TnC is greater than that of F1 TnC, which is likely caused by the two binding sites for TnI in F2 TnC and the single site in F1 TnC.

Here, we have used a range of biophysical techniques to dissect the interaction of F2 TnC into specific contributions using synthetic peptides from TnH and studied how the complex affinities are modulated by the presence of Ca2+ and/or Mg2+. We show that F2 TnC binds to the same regions of TnH as F1 TnC but with two important differences. First, the interaction of F2 TnC with the peptides shows significant Ca2+ dependence, and second, interaction with the peptides does not involve just the TnC C lobe but causes a more generalized conformational change. Our data improve our understanding of Ca2+ regulation of insect flight muscle.

Experimental Procedures

Protein Production

Lethocerus F2 TnC cDNA was inserted into the NcoI and NotI sites of a modified pET24d (M11) expression vector (Novagen) containing an N-terminal hexahistidine tag followed by a tobacco etch virus (TEV) protease cleavage site. The construct was further subcloned to produce individually the two lobes (residues 1−88 and 89−158) of F2 TnC into the pEDTDuet-1 vector using BamHI and PstI sites. The following forward and reverse primers were used: 5′-atggatccagaaaacctgtacttccaaggaatggtcgccatggatgatctggac-3′ (forward primer) and 5′-cggtcatgctgcagttatcaggcggccgaatcttcttcgttgtcttc-3′ (reverse primer) for the N-terminal lobe and 5′-atggatccagaaaacctgtacttccaaggaatgatggaagagctcagggaggc-3′ (forward primer) 5′-cggtcatgctgcagttatcactcgtccccaccgatcatcacttc-3′ (reverse primer) for the C-terminal lobe.

The constructs were transformed through a heat shock protocol using BL21(DE3) or BL21(DE3) pLysS competent cells (Stratagene). The plasmid DNA (1 μL at concentrations of 50−100 ng/μL) was added to 50 μL of competent cells kept on ice for 30 min and then heat shocked at 42 °C for 45−50 s followed by incubation on ice for 2 min. Luria-Bertani (LB) broth (1 mL) was added to the cells, which were left at 37 °C for 30 min and then allowed to grow on LB agar plates containing kanamycin (0.1 mg/mL) at 25 °C for 24 h. The growth was used to inoculate 1 L of medium, and the cells were grown at 25 °C. When the OD600 reached 0.7−0.8, isopropyl β-d-thiogalactopyranoside (IPTG) was added at a final concentration of 0.2 mM to induce protein production. Isotopically labeled samples were produced by growing the cells in minimum medium supplemented with [13C6]glucose and 15NH4Cl as the sole carbon and nitrogen sources, respectively.

A similar purification strategy was used for all constructs. Cells were centrifuged at 4 °C, and the pellet was resuspended in lysis buffer containing 150 mM NaCl, 10 mM imidazole, 20 mM Tris-HCl (pH 8), lysozyme, DNase I, MgCl2, the COMPLETE EDTA free protease inhibitor cocktail (Roche), and PMSF. The lysate was sonicated on ice, and the supernatant was collected after centrifugation at 20000 rpm for 40 min at 4 °C. The His6-tagged protein was purified with a Ni2+-nitrilotriacetate (Ni-NTA)−agarose column (Qiagen) and eluted with 100 mM imidazole in the same buffer. All fractions were checked on a 12% acrylamide gel. The His6 tag was removed using TEV protease (30) incubation at 4 °C for 16 h. The cleaved His6 tag and the residual TEV were removed when the sample was loaded onto Ni2+-nitrilotriacetate (Ni-NTA)−agarose columns (Qiagen) equilibrated in 20 mM Tris-HCl (pH 8) and 150 mM NaCl. The proteins were further purified by size exclusion chromatography using a Superdex 75 26/60 column (Pharmacia) equilibrated with a buffer containing 100 mM NaCl and 20 mM Tris-HCl (pH 6.8).

Ca2+-free (apo) F2 TnC was prepared by mixing concentrated samples of the protein with 5 mM EGTA. These samples were then desalted by two passes through Pharmacia G-25 PD10 columns equilibrated with Ca2+-free (Chelex-100-treated) buffer [20 mM Tris-HCl (pH 6.8) and 100 mM NaCl]. The residual Ca2+ concentration in the buffer was estimated to be <1 μM using the chromophoric Ca2+ chelator 5,5′-Br2BAPTA.

Peptides

Peptides TnH(30−61) (YKAKKAKKGFMTPERKKKLRLLLRKKAAEELKK) and TnH(126−159) (YDLNSQVNDLRGKFVKPTLKKVSKYENKFAKLQKK) were purchased from Peptide 2.0 Inc. Melittin from honey bee venom was purchased from Sigma-Aldrich Co. (Dorset, U.K.). The purity of the peptides was greater than 95% as assessed by HPLC, and the sequences were verified using MALDI mass spectroscopy.

Fluorescence Measurements

Uncorrected fluorescence emission spectra were recorded using a Jasco FP-6300 spectrofluorimeter equipped with an ETC-273T temperature controller with excitation at 280 nm (bandwidth of 1.5 nm) and emission scanned from 290 to 400 nm (bandwidth of 5 nm). All measurements were taken at 20 °C in 20 mM Tris-HCl (pH 6.8) and 100 mM NaCl.

CD Measurements

Near- and far-UV CD spectra were recorded on a Jasco J-715 spectropolarimeter equipped with a PTC-348WI temperature controller. Near-UV CD intensities are presented as the CD absorption coefficient calculated using the molar concentration of protein (ΔεM) and far-UV CD intensities as the CD absorption coefficient calculated on a mean residue weight basis (ΔεMRW). Thermal denaturation was followed by monitoring the far-UV CD signal at 222 nm while changing the temperature from 2 to 95 °C at a rate of 1 °C/min. All CD measurements were taken in 10 mM Hepes (pH 6.8) and 100 mM NaCl.

Ca2+ Binding Studies

Stoichiometric Ca2+ association constants for F2 TnC in the presence and absence of the TnH peptides were determined from Ca2+ titrations of apoprotein samples performed in the presence of the chromophoric Ca2+ indicators, 5,5′-dibromo BAPTA and 5-nitro BAPTA, using published methods (31). Measurements were performed at 20 °C in 20 mM Tris-HCl (pH 6.8) and 100 mM NaCl using a JASCO V-550 UV−vis spectrophotometer equipped with an ETC-505T temperature controller. The equilibrium association constants for 5,5′-dibromo BAPTA and 5-nitro BAPTA under these conditions were determined to be (6.3 ± 0.5) × 105 and (4.1 ± 0.3) × 104 M−1, respectively.

NMR Measurements

All NMR experiments were carried out at 25 °C and 600 MHz on a Varian Inova or an Advanced Bruker spectrometer equipped with a cryoprobe. The concentrations of highly concentrated stock solutions of the peptides, in 100 mM KCl and 20 mM Tris-HCl, were measured in triplicate by amino acid analysis. The pH was adjusted to 6.8 with small additions of a 0.01 M NaOH solution. The pH was checked after the final peptide addition to make sure that it had remained constant throughout the titration. The final F2 TnC:TnH(30−61) and TnC:TnH(126−159) molar ratios were 1:3 and 1:2.5, respectively. At these molar ratios, two successive additions of peptides did not produce further changes in the spectra.

Results

CD Studies of F2 TnC

The far-UV CD spectra of the apo and Ca2+-loaded forms of F2 TnC (Figure 1A) are those of a protein with a high α-helical content (∼50%), indicating that the protein is folded also in the absence of the cation. The Mg2+-induced change in CD intensity is similar to, though slightly smaller than, that induced by Ca2+ (not shown), indicating that both metal ions perturb the secondary structure (as reflected by CD) in a similar way. The near-UV CD spectra of the apo and Ca2+-loaded forms of F2 TnC (Figure 1B) show that Ca2+ also causes a change in the tertiary structure of the protein close to one or more of the three tyrosine residues. The near-UV CD spectrum of the Mg2+-loaded form is essentially identical to that of the apoprotein, showing that Mg2+ binding does not perturb the environment of the tyrosine residues.

Figure 1

Probing the secondary structure of F2 TnC by CD spectroscopy. CD spectra of F2 TnC in the absence (black lines) and presence (gray lines) of 1 mM Ca2+. We estimated ∼50% helical content using the methods described in ref (41). (A) Far-UV spectra. (B) Near-UV spectra.

Although the change in the far-UV CD signal is small, it can be used to monitor binding of a metal ion to F2 TnC. The far-UV CD titration of apo F2 TnC with Ca2+ (Figure 2A) could not be satisfactorily analyzed with a one-site binding model (χ2 = 4.8). Analysis with a two-site binding model with the assumption that the binding at each site generates the same change in CD signal improved the χ2 value to 1.2. The equilibrium dissociation constants were determined to be 4.8 ± 1.4 and 120 ± 28 μM. Given the assumptions involved, these values are in reasonable agreement with values of 2.6 and 180 μM, respectively, previously determined (20) using the same technique but under slightly different buffer conditions. The corresponding titration with Mg2+ (Figure 2B) could be analyzed with a one-site binding model to give a dissociation constant (Kd) of 0.81 ± 0.12 mM, which is in agreement with the value of 0.7 mM estimated using competition with Ca2+ binding described in ref (20).

Figure 2

Effects of Ca2+ and Mg2+ on F2 TnC. Far-UV CD titrations of apo F2 TnC with Ca2+ (A) and Mg2+ (B). The solid lines are fits to two-site (A) and one-site (B) binding models. The dashed line in panel A is the best fit to a one-site model (χ2 of 4.8 compared with 1.2 for the fit to the two-site model).

Far-UV CD thermal unfolding studies are a useful method for screening for interaction partners because any ligand that binds to only the folded form of the protein will increase its stability toward thermal denaturation. The higher the affinity of the ligand, the greater the shift toward higher unfolding temperatures. We therefore performed thermal unfolding studies with F2 TnC in the presence and absence of Ca2+ and/or TnH peptides. Because the concentration of Mg2+ in muscle is in the millimolar range, we also probed the effects of this ion. The synthetic peptides used here, TnH(30−61) and TnH(126−159), correspond to the N-terminus and to a region containing the inhibitory and switch regions in vertebrates, respectively. F2 TnC in the absence of metal ions and peptides undergoes a single transition, presumably caused by unfolding of the C lobe, with a transition midpoint temperature (Tm) of 40.6 °C. When temperature unfolding was repeated in the presence of cations and/or peptides, the results (Table 1) suggest that Ca2+ binds significantly more strongly than Mg2+, that the TnH peptides bind to both apo and metal-loaded forms of F2 TnC, and that TnH(30−61) binds more strongly than TnH(126−159).

Table 1

Transition Midpoint Temperatures (Tm) for Thermal Unfolding of F2 TnC (18 μM) in the Presence and Absence of Metal Ions and TnH Peptides

conditions	Tm (°C)
apo F2 TnC	40.6
35 μM TnH(30−61)	>70a
35 μM TnH(126−159)	56.4
1 mM Ca2+	74.6
1 mM Ca2+ and 35 μM TnH(30−61)	>90
1 mM Ca2+ and 35 μM TnH(126−159)	79.0
1 mM Mg2+	49.4
1 mM Mg2+ and 35 μM TnH(30−61)	>85
1 mM Mg2+ and 35 μM TnH(126−159)	59.6

F2 TnC complexes with peptides precipitate at >70 °C in the absence of metal ions.

Competition Studies of the Interaction of F2 TnC with TnH Peptides by Fluorescence

As in a previous study of the interaction of F1 TnC with the TnI peptides (26), we quantified the interactions by exploiting the fact that F2 TnC also binds with high affinity to melittin, a 26-residue peptide containing a single tryptophan. Binding of melittin was studied by monitoring the changes in tryptophan fluorescence associated with binding to F2 TnC, which contains no tryptophans (Figure 3A). The equilibrium dissociation constants for binding of melittin to F2 TnC are 20−25-fold weaker than those for F1 TnC, which have been remeasured in this study (Table 2). Equilibrium Kd values for the binding of the TnH(30−61) and TnH(126−159) peptides were determined using competition with melittin binding as previously described (26) (Figure 3B). The Kd values for the binding of TnH(30−61) to apo and holo F2 TnC [37.7 and 7.8 nM, respectively (see Table 2)] are some 10-fold higher than the corresponding values for F1 TnC. Interestingly, however, in this case, we observe that the effect of Ca2+ in increasing the strength of the interaction is significantly greater than for F1 TnC. The TnH(126−159) peptide binds much more weakly than does the TnH(30−61) peptide (Table 2). In this case, however, the affinities for F2 TnC and F1 TnC are very similar, as are the effects of Ca2+.

Figure 3

Competition experiments to determine the affinities of TnH peptides for F2 TnC. (A) Titration of melittin (0.33 μM) with F2 TnC in the presence (filled symbols) and absence (empty symbols) of 1 mM Ca2+. (B) Titrations of melittin (0.33 μM) and F2 TnC with TnH(30−61) (squares) and TnH(126−159) (circles). F2 TnC concentrations were 1.4 and 2.4 μM for experiments performed in the presence (filled symbols) and absence (empty symbols) of 1 mM Ca2+, respectively.

Table 2

Equilibrium Dissociation Constants for the Binding of Melittin and the TnH Peptides to F1 TnC and F2 TnC in the Presence and Absence of Metal Ions

peptide	metala	Kd (F1 TnC)b	Kd (F2 TnC)	Kd (F2 TnC)/Kd (F1 TnC)
melittin	none	41 ± 9 nM (26 ± 7 nM)	0.84 ± 0.18 μM	20.5 ± 6.3
	Ca2+	8.3 ± 2.1 nM (5.2 ± 1.2 nM)	0.19 ± 0.03 μM	22.9 ± 6.8
	Mg2+	32 ± 6 nM	0.77 ± 0.12 μM	24.0 ± 5.8
TnH(30−61)	none	1.9 ± 0.6 nM (1.5 ± 0.5 nM)	37.7 ± 4.5 nM	19.8 ± 3.1
	Ca2+	0.9 ± 0.2 nM (1.3 ± 0.3 nM)	7.8 ± 0.15 nM	8.7 ± 2.5
	Mg2+	2.2 ± 0.4 nM	38.5 ± 5.2 nM	17.5 ± 3.3
	Ca2+ and Mg2+	1.3 ± 0.4 nM	6.7 ± 0.9 nM	5.2 ± 1.2
TnH(126−159)	none	12.2 ± 3.4 μM (8.5 ± 3.7 μM)	15.5 ± 4.5 μM	1.3 ± 0.5
	Ca2+	3.7 ± 1.1 μM (5.4 ± 1.5 μM)	3.1 ± 0.8 μM	0.85 ± 0.35
	Mg2+	9.1 ± 3.2 μM	13.5 ± 3.5 μM	1.5 ± 0.6
	Ca2+ and Mg2+	not determined	3.5 ± 0.8 μM	not determined

Experiments were performed in the presence of 0.2 mM EGTA and 1 mM Ca2+ or 5 mM Mg2+.

Data in parentheses are from ref (26).

Values measured in the presence of 5 mM Mg2+ are very similar to those measured with the apoproteins for both melittin and the TnH peptides. Values measured in the presence of 5 mM Mg2+ and 1 mM Ca2+ are essentially the same as those measured in the presence of 1 mM Ca2+.

Ca2+ Binding Studies with Chromophoric Ca2+ Chelators

Initial experiments involving titration of a mixture of the chelator 5,5′-dibromo BAPTA [KCa = (6.3 ± 0.5) × 105 M−1] and apo F2 TnC (1:1.5 ratio) with a stock solution of Ca2+ demonstrated that F2 TnC competed poorly with this high-affinity chelator. We therefore performed all further experiments with a lower-affinity chelator, 5-nitro BAPTA [with a KCa of (4.1 ± 0.3) × 104 M−1]. Solutions of 5-nitro BAPTA (22 μM) and apo F2 TnC (32 μM) were titrated with Ca2+ in the presence and absence of the TnH peptides (65 μM), and the absorbance was measured at 430 nm (Figure 4). Stoichiometric Ca2+ dissociation constants were determined by direct fits to the experimental data as described in ref (31). The results are summarized in Table 3. Titrations performed in the presence of Mg2+ demonstrate that this metal does not compete effectively with Ca2+ binding, although there is some evidence for weak competition at one site (Kd1 increases from 6.3 to 9.3 μM). In the case of the TnH peptides, there is clear evidence that Ca2+ binding is enhanced in the presence of the peptides. This is the expected result because we have shown that the peptides bind more strongly in the presence of Ca2+.

Figure 4

Titrations of 5-nitro BAPTA with Ca2+. Spectrum of 5-nitro BAPTA (22 μM) alone (○), with 31.5 μM apo F2 TnC (◻), or with 31.5 μM apo F2 TnC and 65 μM TnH(30−61) (◇).

Table 3

Stoichiometric Dissociation Constants for the Binding of Ca2+ to F2 TnC Measured in the Presence and Absence of Mg2+ and the TnH Peptides

sample	Kd1 (μM)	Kd2 (μM)	Kd1Kd2 (μM2)
F2 TnC	6.3 ± 1.1	93 ± 15	580 ± 140
F2 TnC and 1 mM Mg2+	9.3 ± 1.4	83 ± 11	780 ± 160
TnH(30−61)	2.4 ± 0.3	45 ± 8	110 ± 20
TnH(126−159)	3.7 ± 0.5	59 ± 12	220 ± 50

TnH Peptide Binding Involves Both Lobes of F2 TnC

We then mapped the sites of interaction of the two peptides on F2 TnC by NMR spectroscopy. The NMR 1H−15N HSQC spectra of apo and holo F2 TnC have an excellent spectral dispersion indicating that both domains are intrinsically structured even in the absence of Ca2+, in agreement with the CD data. This is at variance with what is observed for the C lobe of skeletal muscle TnC, which has in the apo form a spectrum with low chemical shift dispersion typical of random coils (29). Comparison of the HSQC spectra of the apo and holo forms of F2 TnC shows conclusively binding of two Ca2+ ions based on the 15N chemical shift of the amide nitrogen of the residue in position 8 of a canonical EF-hand loop. The value of this observable is highly diagnostic of the presence of Ca2+ binding and provides a direct identification of the residues involved (32). The two other coordination sites are inactive even at high (millimolar) Ca2+:protein molar ratios.

We titrated labeled Ca2+-loaded F2 TnC with both TnH(30−61) (Figure 5A) and TnH(126−159) (Figure 5B) to reach saturation. In both cases, the spectra show progressive chemical shifts during the titration that are typical of a fast exchange regime on the NMR time scale. Similar resonances are affected by the two peptides, but overall, TnH(30−61) produces a stronger chemical shift perturbation at comparable peptide:protein molar ratios. To identify more specifically which lobe of F2 TnC is involved in the interaction without achieving full assignment of the spectrum of F2 TnC, we engineered two constructs spanning the sequence of each of the two lobes: NTnC (residues 1−88) and CTnC (residues 89−158). The NMR spectra of the two fragments are almost completely additive, and their superposition reproduces the spectrum of the full-length protein (Figure 5C). This additivity is observed for most of the individual domains of calmodulin-like proteins and tells us that the two lobes are independently folded and do not significantly influence each other structurally. Comparison of the spectra at the final points of the titrations (each titration was stopped when no further change was observed upon further addition of peptide) with those of the individual domains indicates that the interaction involves resonances of both lobes (Figures 5B,C), implying an overall structural rearrangement of F2 TnC upon TnH binding. This is at strong variance with what is observed for the same peptides with F1 TnC where interaction seems to involve only the C lobe (26) but is in good agreement with the behavior of vertebrate TnCs (33).

Figure 5

Titration of F2 TnC with TnH peptides as followed by NMR spectroscopy and mapping of the interactions onto the individual lobes. (A) Superposition of the HSQC spectrum of Ca2+-loaded F2 TnC (black) with the final point of a titration with TnH(126−159) (gray) (1:2.5 molar ratio). (B) Same as panel A but for the TnH(30−61) peptide (1:1.3 molar ratio). Red and green boxes are examples of resonances from the N and C lobes, respectively, which can be judged by comparing the spectra in panel C. (C) Comparison of the spectra of full-length F2 TnC (black) and of the N- and C-terminal lobes (red and green, respectively). All spectra were recorded at 600 MHz and 25 °C.

The TnH Peptides Compete for the Same Binding Site on F2 TnC

To test whether the two peptides compete for the same site and to detect any potential extra sites of interaction, we first saturated F2 TnC with the high-affinity TnH(30−61) peptide and then titrated this complex with the low-affinity TnH(126−159) peptide. The spectrum recorded at a 1:1 TnH(126−159):F2 TnC molar ratio is mostly superposable with that recorded in the absence of TnH(126−159) (Figure 6A), suggesting that, once TnH(30−61) is bound, the second peptide is not able to occupy additional binding sites. In contrast, when F2 TnC was first saturated with the low-affinity TnH(126−159) peptide and this complex was then titrated with TnH(30−61), most of the features observed in the presence of just TnH(30−61) reappeared (Figure 6B). These results support the hypothesis that the two peptides compete for the same site.

Figure 6

Testing whether the peptides compete for the same site of F2 TnC. (A) Superposition of the HSQC spectra recorded at the final points of the titrations of F2 TnC with TnH(30−61) (green) and with both TnH(30−61) and TnH(126−159) (blue). (B) Superposition of the HSQC spectra recorded at the final points of the titrations of F2 TnC with TnH(126−159) (red) and with both TnH(126−159) and TnH(30−61) (black). All spectra were recorded at 600 MHz and 25 °C.

Discussion

We have characterized here the structural features of the F2 isoform of Lethocerus TnC and the properties of its binding to TnH. We have also explored how the affinities of TnC for TnH are affected by the presence of Ca2+ and/or Mg2+. It is interesting to compare our findings with what is known for vertebrate TnCs and for the Lethocerus isoform F1 TnC, responsible for stretch activation (20). Structurally, F2 TnC seems to share most of the properties already described for F1 TnC (26,34): both lobes are structured and are mainly helical also in the absence of Ca2+, a property at variance with that of the C lobe of skeletal TnC in which this domain is completely unstructured in the absence of Ca2+(29). Interaction with TnH was tested using two peptides spanning the N-terminus and the region of TnH that corresponds to the inhibitory and switch peptide in vertebrate muscle (35). These are the regions that anchor TnI and activate contraction in vertebrate muscles (see ref (5) and references cited therein); the same peptides were also found to interact with F1 TnC (26). We obtain a Kd in the nanomolar range for binding of F2 TnC to TnH(30−61) that reveals a somewhat tighter interaction compared to that observed for vertebrate skeletal TnC, i.e., 8 nM compared to 48 nM in the presence of Ca2+(28). The corresponding value for binding of TnH(126−159) to F2 TnC is in the micromolar range and again approximately comparable to that of vertebrate TnC.

An even more important comparison of our findings with the Kd values observed for F1 TnC using the same peptides (26) should be conducted. The interaction of F2 TnC with TnH has several important different features compared to that of the F1 TnC isoform. First, the interaction is modulated by Ca2+, while F1 TnC seems to be entirely insensitive to Ca2+(26). We could confirm this observation using three independent approaches that go from probing the effect of Ca2+ on the melting points of F2 TnC in the absence and in presence of the peptides to a quantification of the dissociation constants by competition experiments. A Ca2+ dependence of the interaction for F2 TnC but not for F1 TnC can be explained by the presence of an extra Ca2+ binding site in the regulatory N lobe of F2 TnC that is absent in F1 TnC. The current evidence suggests that the calcium concentration in Lethocerus should not be sufficiently high to allow full occupancy of the two sites (20,21). We also probed the interactions in the presence of an excess of Mg2+ to mimic the muscle ion concentrations. Although Mg2+ is constantly present at high concentrations in muscles and is therefore an important component that is customarily used in mechanical measurements, most of the in vitro assays aiming at quantifying the binding affinities of TnI fragments have been conducted in the absence of this cation. This choice was probably dictated by the reasonable assumption that, because Mg2+ does not alter dramatically the structure of TnC (36), it should not be of primary importance. Interestingly, we observe for both F1 and F2 TnCs that the presence of Mg2+ does modulate the affinities but does not substantially alter their ratios.

The second noticeable difference from F1 TnC is that for this isoform the peptide−TnC interaction involves only the C lobe, leaving the N lobe idle (26). This behavior is very different from what we observed here for F2 TnC where the peptides have profound effects on the whole spectrum and affect resonances of residues from both the N lobe and the C lobe, indicating that the mode of binding of the two isoforms to the same TnH peptides is very different.

Finally, whether in an ion-loaded or ion-free form, F2 TnC has overall lower affinities for TnH(30−61) compared to those of F1 TnC by approximately 1 order of magnitude, whereas the affinities for TnH(126−159) are comparable. This seems to be at variance with recent estimates by isothermal titration calorimetry of the Kd values of F1 and F2 TnCs with a long N-terminal TnH fragment [TnH(1−224)] that contains both peptides and with the observation that, upon replacement of F1 and F2 TnC in different proportions to reproduce the behavior of native fibers, the maximal work is produced at excess F1:F2 molar ratios (i.e., 50−100:1) (24). While accurate quantification of the interactions at the level of the longer TnH construct might be biased by the strong tendency of the protein to aggregate, the relative differences in Kd values observed for the longer TnH(1−224) (1 μM for F1 TnC and 23 nM for F2 TnC) could well reflect the presence of additional interactions formed by F2 TnC with regions of TnH outside the peptides tested here and/or conformational effects that can be observed only at the level of long TnH constructs.

Taken together, our results all suggest a much greater similarity of F2 TnC to the vertebrate muscle orthologs than to F1 TnC. In agreement with this hypothesis is the observation that TnH(1−224) has an affinity for F2 TnC closer to that of cardiac TnI for TnC [i.e., ∼8 nM (24,37)]. We therefore propose that the structure and the main mechanism of action of F2-mediated contraction are also more similar to those of vertebrate muscle, whereas we expect that the F1 TnC−TnH−TnT complex will be significantly different.

What do these observations tell us about the mechanism operating in insect flight muscles responsible for stretch activation?

The role of binding of Ca2+ to F2 TnC seems to be, as in vertebrate muscle, twofold. On one hand, it confers specific properties that facilitate the anchoring of the C lobe to the N-terminus of TnI and/or TnH. Especially tight affinities of F1 TnC could well be required because of the need for greater resistance to forces arising from the greater stiffness of flight muscle fibers (17). On the other hand, we expect that the Ca2+ bound to the N lobe of F2 TnC but not of F1 TnC will act as the sensor that triggers and regulates the contraction during the preparatory phases of the flight. When the system has been sufficiently warmed up, the much more abundant F1 TnC isoform will come into action. The extent to which F2 TnC is activated by Ca2+ also affects the work done during oscillatory contraction by limiting the range of Ca2+ concentration over which the muscle performs maximal work (24). Recently, EM and X-ray diffraction studies on Lethocerus flight muscle have confirmed the presence of “troponin bridges” extending from troponin on the thin filament to myosin on the thick filament (38,39). It was proposed that during stretch activation, troponin bridges pull on tropomyosin to expose myosin binding sites on actin, according to the steric blocking model (39). If troponin bridges contain the extended C-terminal tail of TnH (24,40), stretch may act through the troponin complex to displace tropomyosin on actin as long as Ca2+ is bound to F1 TnC. Alternatively, the N lobe of F1 TnC may interact with an as yet unidentified partner extending from the thick filament in a troponin bridge. In this case, mechanical stress on F1 TnC might have the same effect as Ca2+ binding to the N lobe of F2 TnC.

While much work is still needed to clarify the exact mechanism of stretch activation, this paper contributes to our comprehension of the phenomenon and provides clear suggestions about where to direct our future efforts. We expect that a real breakthrough may come only when the structure of the F1 troponin complex becomes available.