Influence of enhanced troponin C Ca²⁺-binding affinity on cooperative thin filament activation in rabbit skeletal muscle

Abstract

We studied how enhanced skeletal troponin C (sTnC) Ca²⁺-binding affinity affects cooperative thin filament activation and contraction in single demembranated rabbit psoas fibres. Three sTnC mutants were created and incorporated into skeletal troponin (sTn) for measurement of Ca²⁺ dissociation, resulting in the following order of rates: wild-type (WT) sTnC–sTn > sTnC^F27W–sTn > M80Q sTnC–sTn > M80Q sTnC^F27W–sTn. Reconstitution of sTnC-extracted fibres increased Ca²⁺ sensitivity of steady-state force (pCa₅₀) by 0.08 for M80Q sTnC, 0.15 for sTnC^F27W and 0.32 for M80Q sTnC^F27W with minimal loss of slope (n_H, degree of cooperativity). Near-neighbour thin filament regulatory unit (RU) interactions were reduced in fibres by incorporating mixtures of WT or mutant sTnC and D28A, D64A sTnC (xxsTnC) that does not bind Ca²⁺ at N-terminal sites. Reconstitution with sTnC: xxsTnC mixtures to 20% of pre-exchanged maximal force reduced pCa₅₀ by 0.35 for sTnC: xxsTnC, 0.25 for M80Q sTnC: xxsTnC, and 0.10 for M80Q sTnC^F27W: xxsTnC. It is interesting that pCa₅₀ increased by ∼0.1 for M80Q sTnC and ∼0.3 for M80Q sTnC^F27W when near-neighbour RU interactions were reduced; these values are similar in magnitude to those for fibres reconstituted with 100% mutant sTnC. After reconstitution with sTnC: xxsTnC mixtures, n_H decreased to a similar value for all mutant sTnCs. Altered sTnC Ca²⁺-binding properties (M80Q sTnC^F27W) did not affect strong crossbridge inhibition by 2,3-butanedione monoxime when near-neighbour thin filament RU interactions were reduced. Together these results suggest increased sTnC Ca²⁺ affinity strongly influences Ca²⁺ sensitivity of steady-state force without affecting near-neighbour thin filament RU cooperative activation or the relative contribution of crossbridges versus Ca²⁺ to thin filament activation.

Contraction in skeletal muscle is initiated by Ca²⁺ binding to the N-terminus of troponin C (TnC), a subunit of the troponin (Tn) complex, which leads to thin filament activation and force generation (reviewed by Gordon et al. (2000)). Following Ca²⁺ binding to TnC, there is increased interaction between the N-terminus of TnC and the inhibitory subunit of Tn (TnI) and between TnC and the tropomyosin (Tm)-binding subunit (TnT). The increased interaction between these Tn subunits results in the release of TnI from actin and greater mobility of Tm. Tm mobility exposes myosin binding sites on actin and allows for cyclic actomyosin (crossbridge) interactions, which result in contractile force and shortening. Strong crossbridge binding increases and stabilizes Tm displacement (Xu et al. 1999; Lehman et al. 2000) allowing further cooperative crossbridge binding (Gordon et al. 2000). This positive crossbridge feedback to increase thin filament activation (defined as the availability of myosin binding sites on actin) occurs with little effect on Ca²⁺ binding to TnC in skeletal muscle (Fuchs & Wang, 1991; Wang & Fuchs, 1994; Martyn & Gordon, 2001). However, how Ca²⁺ binding properties affect cooperative activation of force and spread of activation along the thin filament in skeletal muscle remains unknown.

A number of cooperative mechanisms have been proposed for skeletal muscle to explain the steep force–Ca²⁺ relationship observed experimentally in demembranated muscle fibres (Gordon et al. 2000, 2001; Regnier et al. 2002). We and others have reported that reduction of nearest-neighbour structural regulatory unit (RU; i.e. 7 actins: 1 Tn: 1 Tm) interactions along the length of the thin filament considerably decreases cooperativity in single skinned skeletal fibres (Moss et al. 1985, 1986; Regnier et al. 2002). Completely extracting native skeletal TnC (sTnC) from demembranated rabbit psoas fibres and reconstituting Tn complexes with varying mixtures of sTnC and a non-Ca²⁺-binding sTnC mutant (D28A, D64A; xxsTnC) allows for the study of how near-neighbour RU interactions influence maximal Ca²⁺-activated force, Ca²⁺ sensitivity of force, and slope or Hill coefficient (used as a measure of cooperativity) of the force–log [Ca²⁺] (pCa) relationship (Regnier et al. 2002). In this earlier study we concluded that (1) near-neighbour RU interactions are the dominant form of cooperativity in skeletal muscle (possibly occurring through end-to-end overlap of Tm), (2) these near-neighbour cooperative RU interactions play a major role in setting the maximal force and Ca²⁺ sensitivity of force, and (3) there remains some cooperativity in the system even with isolated RUs (Regnier et al. 2002). Cooperative mechanisms that may occur within these individual isolated RUs include cooperative Ca²⁺ binding between the two EF-hand Ca²⁺-binding sites in the regulatory domain (N-terminus) of sTnC (Gordon et al. 2001) and/or crossbridge-induced increased crossbridge binding, perhaps via compliant realignment of myosin binding sites on actin (Daniel et al. 1998; Gordon et al. 2000; Tanner et al. 2007).

Ca²⁺ sensitivity and cooperativity of force generation may also be influenced by the Ca²⁺-binding properties of sTnC. Engineering single amino acid mutations into the regulatory domain of sTnC alters Ca²⁺-binding kinetics, as measured by changes in Ca²⁺ dissociation rate (k_off) and steady-state Ca²⁺ affinity in solution (Tikunova et al. 2002; Davis et al. 2004). Reconstitution of skinned rabbit psoas fibres with sTnC mutants alters Ca²⁺ sensitivity of steady-state force (pCa₅₀) such that increased pCa₅₀ is correlated with increased Ca²⁺ affinity of sTnC in complex with the TnI peptide TnI_96-148 and vice versa for decreased Ca²⁺ affinity (Davis et al. 2004). Recent work from our group is in agreement with this. Reduced Ca²⁺ affinity of TnC reduced Ca²⁺ sensitivity of force, and comparison with fibres with mixtures of sTnC and xxsTnC suggested that changes in pCa₅₀ were primarily due to the Ca²⁺-binding properties of TnC, but changes in slope were primarily due to the spread of activation between RUs in the thin filament (Moreno-Gonzalez et al. 2005). This demonstrates that Ca²⁺-binding properties of sTnC play a major role in determining the level of thin filament activation and force generation in skeletal muscle fibres. However, how Ca²⁺-binding properties of sTnC are coupled to near-neighbour RU cooperativity remains to be determined.

In the current study, we sought to more fully understand the relationship between the Ca²⁺-binding properties of sTnC, near-neighbour RU cooperativity, and crossbridge binding in activation of skeletal muscle contraction. sTnC proteins were produced with point mutations that progressively increased Ca²⁺-binding affinity and were then reconstituted into demembranated rabbit psoas fibres. To study how these mutants influence the Ca²⁺-activation properties of individual RUs and near-neighbour RU interactions along the thin filament, fibres were reconstituted with mixtures of sTnC (purified, wild-type (WT) or mutant) and xxsTnC (as used by us to reduce the number of Ca²⁺-activatible RUs (Regnier et al. 2002)). Finally, 2,3-butanedione monoxime (BDM) was used to examine how strong crossbridge binding affects cooperative thin filament activation under conditions of altered Ca²⁺ binding (with sTnC mutants) and reduced RU interactions (with mixtures containing xxsTnC). The results of the present studies confirm that near-neighbour RU interactions are the dominant form of cooperativity in skeletal muscle and suggest that while Ca²⁺ affinity of sTnC is a primary determinant of activation within individual RUs, enhancing Ca²⁺ affinity of sTnC does not influence either the spread of activation along the thin filament or the contribution of crossbridges to activation. A preliminary report of this work has been previously published (Kreutziger et al. 2004).

Methods

Preparation of proteins

A phenylalanine (Phe) to tryptophan (Trp) mutation at residue 27 (F27W) and a methionine to glutamine mutation at residue 80 (M80Q) were incorporated into rabbit sTnC cDNA using oligonucleotide primers for site-directed mutagenesis using the QuikChange kit from Stratagene (La Jolla, CA, USA). WT or mutant rabbit sTnC protein was extracted and purified from E. coli according to the method of Dong et al. (1996). Native rabbit sTnC, skeletal TnI (sTnI) and skeletal TnT (sTnT) were purified from ether powder of rabbit skeletal back and leg muscles according to the method of Potter (1982). Whole Tn complexes were formed using recombinant WT or mutant rabbit sTnC and purified native rabbit sTnI and sTnT as previously described with modification (Szczesna et al. 2000). Tn subunits (C, I and T) were dialysed into 10 mm 3-(N-morpholino) propanesulphonic acid (MOPS), 4.6 m urea, 1 mm dithiothreitol (DTT) and 0.01% NaN₃; pH 7.0 at 4°C. TnC, TnI and TnT were mixed at a molar ratio of 1: 1.5: 1.5 and allowed to sit at room temperature (20°C)for 20 min. The concentrations of urea, KCl and MgCl₂ were reduced by serial dialysis against the following three buffers (a, b and c) for > 6 h (with changing concentrations indicated as a–b–c): 10 mm MOPS, 2–0–0 m urea, 3–1–1 mm MgCl₂, 1 mm DTT and 0.01% NaN₃, pH 7.0 at 4°C. Dialysis in the final buffer was repeated twice, followed by centrifugation of precipitated excess TnI–TnT as confirmed by SDS-PAGE. Non-Ca²⁺-binding rabbit mutant sTnC (D28A and D64A; xxsTnC) and TnI peptide (TnI_96-148) were prepared as previously described (Regnier et al. 2002). sTnC concentrations were calculated from peak absorbance at 280 nm with extinction coefficients of 2680 cm⁻¹m⁻¹ for sTnC and M80Q sTnC, and 8370 cm⁻¹m⁻¹ for sTnC^F27W and M80Q sTnC^F27W. Purity of native and recombinant Tn subunits was assessed by SDS-PAGE.

Ca²⁺ dissociation rates from sTnC or sTn

Ca²⁺ dissociation rates (k_off) from rabbit isolated sTnC and whole sTn (sTnC + sTnI + sTnT) were measured at 5.0 and 15.0 ± 0.1°C using an Applied Photophysics Ltd (Leatherhead, UK) model SX-18MV stopped-flow instrument with a dead time of 1.4 ms at 15°C and a 150 W xenon arc source as previously described (Tikunova et al. 2002; Gomes et al. 2004). Two methods were used to measure Ca²⁺k_off: (1) Ca²⁺ was removed from all proteins, including isolated sTnC and whole sTn complexes, using Quin-2 (Calbiochem), a fluorescent Ca²⁺ chelator, that was excited at 330 nm; and (2) Ca²⁺ was removed from isolated sTnC proteins containing Phe to Trp mutations using excess EGTA by following decreases in Trp fluorescence with excitation at 275 nm. Quin-2 fluorescence reports Ca²⁺ binding to Quin-2 as it dissociates from sTnC, whereas Trp fluorescence reports a conformational change within the N-terminus when Ca²⁺ binds to sites I and II of sTnC. Either method – Quin-2 or Trp – can be used with sTnC^F27W proteins, even with addition of TnI_96-148 (which does not contain any Trp residues). However, only the Quin-2 method can be used with non-F27W sTnC proteins (because there is no Trp to report Ca²⁺ kinetics) or with whole sTn complexes (because Trp residues exist in sTnI and sTnT and confound the sTnC Trp signal). k_off was measured with addition of a peptide fragment of TnI, residues 96–148 (TnI_96-148; switch and inhibitory regions) to isolated sTnC using Trp fluorescence because TnI_96-148 does not contain Trp residues. Reactions for Trp and Quin-2 fluorescence were monitored with the appropriate band-pass optical filters, as previously described (Tikunova et al. 2002). Raw data traces were collected and then two to four traces were averaged before being fitted with an exponential curve. Reported k_off values represent an average (±s.e.m.) of the rate reported by the fit from n = 6–8 traces. Whole sTn-containing M80Q sTnC^F27W was fitted with a double exponential equation using fixed parameters for C-terminal rate and endpoint in order to accurately fit the longer time duration (5 s) of these traces that was required to capture the full N-terminal rate of M80Q sTnC^F27W–sTn (variance less than 2.5 × 10⁻⁴). All other traces for both Trp and Quin-2 methods for isolated sTnC and whole sTn were well fitted by a single exponential. Variance was less than 3.5 × 10⁻³ at 15°C and 4.4 × 10⁻³ at 5°C for isolated sTnC and less than 4.3 × 10⁻⁴ at 15°C for whole sTn. Quin-2 also measures Ca²⁺ dissociation from the C-terminal domain of sTnC, requiring a longer time course (10 s for isolated sTnC and 200 s for whole sTn). Collecting traces of different time durations and fitting the data accordingly resulted in reproducibility of results for both isolated sTnC and whole sTn experiments to determine N- and C-terminal Ca²⁺ dissociation rates. Further, we were able to calculate the molar concentration of Ca²⁺ dissociating from sTnC using a calibration where 0, 5, 10 and 20 μm Ca²⁺ rapidly (∼1 ms) mixed with 150 μm Quin-2, giving a linear relationship of [Ca²⁺] to change in voltage (related to change in fluorescence of Quin-2). This calibration indicated that we were detecting dissociation of ∼65% of the Ca²⁺ bound to the N-terminal domain of sTnC (1.3 mol Ca²⁺ per mol protein) during Quin-2 measurements. The buffer used for isolated sTnC or sTnC + TnI_96-148 stopped-flow experiments contained (mm): MOPS 10, KCl 90 and DTT 1; pH 7.0 at experimental temperature as previously described (Tikunova et al. 2002). 30 μm Ca²⁺ was added to 6 μm sTnC for experiments and was mixed with 150 μm Quin-2. The buffer used for whole sTn stopped-flow experiments contained (mm): MOPS 10, KCl 150, MgCl₂ 1 and DTT 1; pH 7.0 at 15°C. This buffer system for whole sTn was chosen so that the ionic strength would be comparable to that used for fibre experiments (0.17 m), which approximates physiological conditions. No additional Ca²⁺ was added to whole sTn as previously described (Gomes et al. 2004) because contaminating Ca²⁺ was sufficient for obtaining both N- and C-terminal rates with Quin-2 (150 μm).

Steady-state Ca²⁺ affinity of rabbit sTnC

The affinity of Ca²⁺ binding to WT and mutant sTnCs was determined using a model LS50B Perkin Elmer Luminescence Spectrometer (Wellesley, MA, USA) with a circulating refrigerated water bath (model 1160 A, PolyScience, Niles, IL, USA) to maintain cuvette temperature at 15.0 ± 0.1°C. Trp fluorescence was measured during Ca²⁺ titration of sTnC^F27W or M80Q sTnC^F27W by using an excitation wavelength of 276 nm and an emission wavelength of 330 nm, as previously described (Gillis et al. 2003). Slit widths were the same for excitation and emission and were set to 9 or 10 nm to maximize the fluorescence signal range. Each titration was normalized by first subtracting the fluorescence at the lowest [Ca²⁺] and then normalizing the resulting values to the maximum fluorescence value. Each data set of Ca²⁺ dependence of fluorescence was fitted with the Hill equation:

(1)

using non-linear least-squares regression analysis (SigmaPlot version 9.0, SPSS, Inc., Chicago, IL, USA). Parameters of slope (n_H), pCa₅₀ (pCa at half-maximal fluorescence) and dissociation constant (K_d) were calculated for each trace, averaged and reported ±s.e.m. Ca²⁺ association rates (k_on) were calculated from K_d = k_off/k_on using average values of K_d and k_off (Quin-2).

Muscle fibre preparation

Rabbits were housed in the Department of Comparative Medicine at the University of Washington (UW) and cared for in accordance with UW Institutional Animal Care and Use Committee procedures. All animal protocols were in accordance with the US National Institutes of Health Policy on Humane Care and Use of Laboratory Animals and were approved by the UW Animal Care Committee. Male New Zealand white rabbits were anaesthetized with an intravenous injection of pentobarbital (40 mg kg⁻¹) in the marginal ear vein and were exsanguinated when all reflexive response were absent. Small bundles of psoas fibres were excised, demembranated and stored at −20°C for up to 6 weeks as previously described (Regnier et al. 2002). Segments of single fibres dissected from fibre bundles were prepared and for most experiments fibre ends were chemically fixed with 1% gluteraldehyde in water (Chase & Kushmerick, 1988) and wrapped in aluminium foil T-clips for attachment to the mechanical apparatus. For all experiments, initial sarcomere length (L_o) was set to 2.5 μm and continuously monitored with helium–neon laser diffraction. Average unfixed L_o of gluteraldehyde-treated fibres was 1.27 ± 0.04 mm (mean ±s.e.m.; n = 47) and diameter was 55 ± 1 μm.

Ca²⁺ solutions for measurements of fibre mechanics

Experimental solutions contained (mm): phosphocreatine 15, EGTA 15, MOPS 80, free Mg²⁺ 1, (Na⁺+ K⁺) 135, ATP 5, DTT 1, and 250 units ml⁻¹ creatine kinase (Sigma, St Louis, MO, USA) and 4% (w/v) Dextran T-500 (Pharmacia, Piscataway, NJ, USA); pH 7.0, 15°C and an ionic strength of 0.17 m. For activating solutions, the Ca²⁺ concentration (expressed as pCa) was varied between pCa 9.0 and 4.0 by adjusting calcium propionate concentration. Some solutions contained 1–50 mm BDM.

Measurements of fibre mechanics

Two experimental setups were used for collecting mechanical data on individual skinned fibre segments. For studies examining recombinant mutant sTnC and xxsTnC mixtures in fibres (Fig. 5) and inhibition of steady-state force by BDM (Fig. 7), fibre segment ends (unfixed) were wrapped around wire hooks to a force transducer and manual micromanipulator as previously described (Martyn et al. 1993) for measurement of steady-state force. All other mechanical data were collected with fixed and clipped fibre segments attached to an Aurora Scientific (Ontario, Canada) force transducer and a General Scanning model G120DT (Watertown, MA, USA) servo-motor (adjusted for 300 μs step time) by minutien pin hooks and mounted on a Nikon (Japan) inverted microscope as previously described (Regnier et al. 2002). At 5 s intervals, fibre segments were shortened by 15%L_o at 10 L_o s⁻¹, then rapidly restretched to initial L_o to maintain fibre integrity (Brenner, 1983; Sweeney et al. 1987; Chase & Kushmerick, 1988). Resulting force transients are visible as vertical lines in the chart records shown in Fig. 3. Measurement of steady-state isometric force was made prior to the release–restretch cycle as [Ca²⁺] in the activating solutions was varied. The fibre preparation was moved between pCa solutions that were held in individual temperature-controlled troughs as previously described (Regnier et al. 2002). Passive force was determined at pCa 9.0 with a release–restretch protocol and subtracted from total force measured in solutions of higher [Ca²⁺] to obtain the active force values reported. Maximal fibre force (F_max), measured at pCa 4.5 just prior to extraction, was 354 ± 11 mN mm⁻² (n = 47; assuming circular cross-sectional area). Fibres with greater than 12% loss of F_max from initial value to just prior to extraction were discarded. sTnC extraction solution contained (mm) MOPS 10, EDTA 5 and trifluoperazine (TFP) 0.5; pH 6.6 at 15°C, and selective sTnC extraction was completed to 1.3 ± 0.1%F_max with repeated incubations in extraction solution (30 s) and pCa 9 solution (10 s) as previously described (Regnier et al. 1999; Moreno-Gonzalez et al. 2005). Reconstitution with 1 mg ml⁻¹ of total protein in pCa 9 relaxing solution was completed with 1–2 min incubations (unless otherwise noted). Reported reconstituted F_max values (Tables 2–4) are from back-to-back measurements comparing just prior to extraction and immediately following reconstitution. Steady-state force–pCa relationships were fitted with the Hill equation (eqn (1)) to obtain pCa at half-maximal force (pCa₅₀; defined as the Ca²⁺ sensitivity of force) and slope (n_H; the apparent cooperativity of contractile activation). Reported pCa₅₀ and n_H values represent the means of the values from the individual fits (±s.e.m). Plots show average (±s.e.m) of each point with curves fitted to the average data. Analysis of variance (ANOVA) was used to compare between groups and when differences between groups were significant, Student's paired or unpaired t tests were used with statistical significance set at P < 0.05.

Figure 5.

Reduction of thin filament near-neighbour cooperativity in fibres by increasing xxsTnC content in protein mixtures for reconstitution A, force–pCa relationships of fibres reconstituted with mixtures of M80Q sTnC^F27W and xxsTnC are shown with pre-extracted native sTnC (•) as control. Mixture ratios of M80Q sTnC^F27W: xxsTnC are shown for each symbol. F_max values for mixtures are slightly less than the consecutive activation values reported in Table 3 due to fibre rundown that occurred with the force–pCa protocol. B, force assay to determine the relative binding affinity of M80Q sTnC^F27Wversus sTnC in sTnC-extracted thin filaments of fibres shows F_max (pCa 4) measured after short interval incubations in 0.1 mg ml⁻¹ sTnC (○) or M80Q sTnC^F27W (□). Black stars indicate F_max after incubations in 1 mg ml⁻¹ purified sTnC. Black diamonds show F_max after fully extracting M80Q sTnC^F27W and reconstituting fibres in 1 mg ml⁻¹ purified sTnC. Force curves were well fit by a single exponential rising curve: f =y₀+a(1−e^−bx) with r² > 0.995 for all proteins. For sTnC, baseline y₀ = 5%, amplitude a = 88%, and the time constant for force rise (1/b) was 2.48 ± 0.11 min. For M80Q sTnC^F27W, y₀ = 8%, a = 69% and 1/b was 8.14 ± 1.59 min. C, plot of F_max (relative to F_max of 100% protein) versus functional sTnC content of reconstitution mixture shows WT sTnC controls (○) and M80Q sTnC^F27W (□) at each mixture ratio (100: 0, 80: 20, 60: 40, 40: 60 and 20: 80; n = 2–8). Transformation of data (ΔM80Q sTnC^F27W; grey stars) to account for unequal binding in the fibre of M80Q sTnC^F27W and xxsTnC (see Appendix) shifts mixture content of M80Q sTnC^F27W (arrows). Continuous line represents the unity line y =x. Some error bars are smaller than symbols.

Figure 7.

Inhibition of strong crossbridge formation with 2,3-butanedione monoxime (BDM) Maximal force (pCa 4) was measured in the presence of increasing concentrations of BDM and with different reconstitution conditions of the thin filament to examine the effect of strongly bound crossbridges on thin filament activation level. F_max in the absence of BDM was normalized to 1.0 for each reconstitution condition for comparison. Thin filament conditions were pre-extracted native sTnC (•), 100% purified sTnC (control; ○), 40: 60 sTnC: xxsTnC (▵), 40: 60 M80Q sTnC^F27W: xxsTnC (grey squares) and 20: 80 sTnC: xxsTnC (grey circles). Inset shows reconstituted F_max for 100% purified sTnC (○; 0.98 ± 0.05; n = 3), 40: 60 sTnC: xxsTnC (▵; 0.39 ± 0.03; n = 3), 20: 80 sTnC: xxsTnC (grey circles; 0.20 ± 0.03; n = 3) and 40: 60 M80Q sTnC^F27W: xxsTnC (grey squares; 0.18 ± 0.05; n = 3). Curves were fit with the inhibition curve F =F_min+ (1 −F_min) × (K_i/(K_i+[BDM])) to determine the asymptote of minimal force (F_min; where amplitude is 1 −F_min) and the inhibition constant is K_i; (see text for values).

Figure 3.

Examples of chart record traces from protocols to measure skinned rabbit psoas fibre mechanics A, trace shows a maximal activation (pCa 4.5) followed by measurement of force–pCa curve ending with maximal activation, F_max, and then sTnC extraction. Force–pCa curve shows relaxation (pCa 9) after first pCa 6.1 activation for readjusting sarcomere length to 2.5 μm. Hill fit parameters were pCa₅₀= 6.04 and n_H= 2.5. All panels have pCa₅₀ value marked by an asterisk to emphasize changes in Ca²⁺ sensitivity between panels. After sTnC extraction in A (first arrow; see Methods), 0.7%F_max remained at pCa 4.5 (second arrow). B, example trace from a separate experiment shows reconstitution of Tn complexes with 100% M80Q sTnC^F27W where 84%F_max was recovered (see Methods), followed by a force–pCa curve where pCa₅₀= 6.35 and n_H= 2.5. The wandering baseline does not effect measurement of steady-state isometric force as true zero force is recorded with each trace recorded. C, example trace (same fibre as A) shows reconstitution of Tn complexes with a mixture of M80Q sTnC^F27W and xxsTnC where 23%F_max was recovered. Signal gain on chart recorder was increased by 2.5 for improved resolution (arrow). pCa₅₀= 5.98 and n_H= 1.4. Fibre diameters were 78 μm (A and C) and 59 μm (B).

Table 2.

Force–Ca²⁺ relationship parameters in skinned psoas fibres fully reconstituted with sTnC

	Pre-extracted		Reconstituted
sTnC (n)	pCa50	nH	Fmax‡	pCa50	nH	ΔpCa50§
Purified sTnC (7)	5.98 ± 0.02	3.1 ± 0.2	1.0	5.96 ± 0.03	3.2 ± 0.4	−0.01 ± 0.02
M80Q sTnC (6)	6.08 ± 0.05	3.4 ± 0.2	0.89 ± 0.03*	6.15 ± 0.05*	2.8 ± 0.2	0.08 ± 0.02*†
sTnCF27W (9)	6.08 ± 0.05	2.9 ± 0.1	0.92 ± 0.02*	6.23 ± 0.04*	2.7 ± 0.1	0.15 ± 0.03*
M80Q sTnCF27W (9)	6.06 ± 0.01	3.4 ± 0.1	0.85 ± 0.01*#	6.38 ± 0.02*#	2.5 ± 0.1	0.32 ± 0.02*#

‡Fraction of reconstituted F_max compared to purified sTnC F_max where purified sTnC was 0.96 ± 0.01 of pre-extracted F_max (P < 0.01). §Change in pCa₅₀versus pre-extracted native sTnC value calculated for each fibre and reported as average ±s.e.m. within group of n = 6–9 fibres. *P < 0.01 versus purified sTnC; #P < 0.01 versus sTnC^F27W; †P < 0.05 versus sTnC^F27W. All reconstituted values are significantly different (for n_H, P < 0.05; for pCa₅₀ and F_max,P < 0.01) by paired t test versus pre-extracted native sTnC within each group of n = 6–9 fibres, except for purified sTnC values for pCa₅₀ and n_H (P > 0.05).

Table 4.

Force–Ca²⁺ parameters for force-matched fibres reconstituted with sTnC: xxsTnC mixtures to ∼0.2 F_max

		Pre-extracted		Reconstituted
sTnC Mixture (n)	pCa50	nH	Fmax#	pCa50	nH	ΔpCa50§
sTnC: xxsTnC (6)	5.97 ± 0.05	2.9 ± 0.1	0.24 ± 0.03*	5.62 ± 0.05**	2.1 ± 0.1**	−0.35 ± 0.03*
M80Q sTnC: xxsTnC (6)	5.98 ± 0.05	3.1 ± 0.3	0.21 ± 0.01*	5.73 ± 0.06*	2.0 ± 0.2*	−0.25 ± 0.04*†
M80Q sTnCF27W: xxsTnC (7)	6.03 ± 0.03	3.1 ± 0.2	0.20 ± 0.01*	5.92 ± 0.05##	2.0 ± 0.2*	−0.10 ± 0.03**‡

# Fraction of reconstituted F_max compared to 100% purified sTnC F_max (reported in Table 2). §Change in pCa₅₀versus pre-extracted native sTnC value calculated for each fibre and reported as average ±s.e.m. within group of n = 6–7 fibres. *P < 0.01 versus purified sTnC value (reported in Table 2); **P < 0.05 versus purified sTnC value (reported in Table 2); ##P < 0.01 versus sTnC: xxsTnC and P < 0.05 versus M80Q sTnC: xxsTnC; †P < 0.05 versus sTnC: xxsTnC; ‡P < 0.01 versus sTnC: xxsTnC and versus M80Q sTnC: xxsTnC. All reconstituted values are significantly different (P < 0.05) by paired t test versus pre-extracted native sTnC within each group of n = 6–7 fibres. All F_max and n_H values do not significantly differ between xxsTnC mixture groups by ANOVA.

Results

Ca²⁺ kinetics of recombinant rabbit sTnC mutants

Ca²⁺ binding properties were determined for isolated sTnC and in complex with TnI_96-148 or sTnI + sTnT. Isolated sTnC measurements allowed us to determine steady-state Ca²⁺ affinity (K_d) and kinetics of Ca²⁺ dissociation (k_off), although determination of K_d was not possible using tryptophan fluorescence for whole sTn complexes. Table 1 summarizes Ca²⁺-binding properties and shows that k_off progressively decreased, in the order of WT sTnC > M80Q sTnC > sTnC^F27W > M80Q sTnC^F27W. k_off was also measured with Trp fluorescence for sTnC^F27W (350 ± 2 s⁻¹) and M80Q sTnC^F27W (83.5 ± 0.7 s⁻¹) as a control, and these values are comparable to those previously reported for the chicken isoform of sTnC (Davis et al. 2004). Measurements at 5°C were made for quantitative comparisons between isolated sTnC mutants because k_off for WT sTnC is too fast to measure at 15°C. Lower temperature decreased k_off for isolated sTnC proteins and was 725 ± 2 s⁻¹ for WT sTnC, and decreased by 5-fold, 17-fold and 40-fold at 5°C for M80Q sTnC, sTnC^F27W and M80Q sTnC^F27W, respectively. Additional measurements were made in the presence of an sTnI peptide (residues 96–148 containing switch and inhibitory regions; TnI_96-148) using Trp fluorescence. k_off decreased by 50-fold for sTnC^F27W–TnI_96-148 (7.0 ± 0.1 s⁻¹) and by 20-fold for M80Q sTnC^F27W–TnI_96-148 (3.5 ± 0.1 s⁻¹) versus isolated sTnC^F27W and M80Q sTnC^F27W, respectively, at 15°C. This trend agrees with previous reports and suggests an important contribution of intermolecular interactions between TnI and TnC to slowing k_off (Davis et al. 2004).

Table 1.

Solution Ca²⁺-binding kinetics of sTnC and sTn at 15°C

	Isolated sTnC		Whole sTn*
Rabbit sTnC Type	koff (s−1)	Kd (μm)	kon× 108 (m−1 s−1)	koff (s−1)
WT sTnC	Too fast	—	—	5.57 ± 0.04
M80Q sTnC	888 ± 44‡†	—	—	3.25 ± 0.01§
sTnCF27W	353 ± 5	2.19 ± 0.16	1.6	4.57 ± 0.02§
M80Q sTnCF27W	70.6 ± 0.7‡	0.75 ± 0.01‡	0.9	2.16 ± 0.02§‡

*Whole sTn complexes contain purified native rabbit sTnI and sTnT. †A slower rate of 24.4 ± 1.0 s⁻¹ was 27% of amplitude signal, and this double exponential signal appeared only with isolated M80Q sTnC. §P < 0.01 versus WT sTn; ‡P < 0.01 versus sTnC^F27W.

Steady-state K_d was measured for isolated sTnC^F27W and M80Q sTnC^F27W at 15°C (Fig. 1) and used to determine Ca²⁺ association rate (k_on). K_d for sTnC^F27W was ∼2.2 μm and decreased for M80Q sTnC^F27W (Table 1). Values of k_on (k_on=k_off/K_d) for both F27W mutants (Table 1) were similar to those previously reported (Johnson et al. 1994; Tikunova et al. 2002), approach diffusion limitations at ∼10⁸m⁻¹ s⁻¹, and were somewhat greater for sTnC^F27W. This calculation suggests that the ∼3-fold increase in steady-state K_d for M80Q sTnC^F27W (compared to sTnC^F27W) is probably not due to differences in k_on but determined primarily by changes in k_off.

Figure 1.

Steady-state Ca²⁺ affinity measurements for isolated sTnC^F27W and M80Q sTnC^F27W at 15°C Ca²⁺ sensitivity of steady-state affinity (pK) was 5.67 ± 0.03 for sTnC^F27W (n = 11) and increased for M80Q sTnC^F27W to 6.13 ± 0.01 (n = 10). The calculated slopes of the binding curves were 2.0 ± 0.1 for sTnC^F27W and 1.6 ± 0.1 for M80Q sTnC^F27W. Error bars (±s.e.m.) are present and often smaller than symbols.

Whole sTn complexes contained WT or mutant sTnC and purified native rabbit sTnI and sTnT. All k_off values in whole sTn were slower than k_off values for isolated sTnC or sTnC–TnI_96-148, suggesting that interactions between all three subunits of troponin slow k_off to its greatest extent. Mutations in sTnC decreased k_off in whole sTn versus WT sTnC–sTn by 42% for M80Q sTnC–sTn, 18% for sTnC^F27W–sTn and 61% for M80Q sTnC^F27W–sTn (Fig. 2 and Table 1). While the order of decreasing k_off changed from isolated sTnC values (with M80Q sTnC–sTn having a slower k_off than sTnC^F27W–sTn), these k_off rates in whole sTn confirmed that these mutations reduced k_off and M80Q sTnC^F27W had the slowest k_off of the three sTnC mutants. It is important to note that these rates at 15°C are 2–6 s⁻¹ which could be slow enough to affect the kinetics of force development or relaxation, especially during submaximal Ca²⁺ activation.

Figure 2.

Example traces of Ca²⁺ dissociation from whole sTn with sTnC mutants Stopped-flow spectroscopy traces show the time course of fluorescence increase in Quin-2 as it binds Ca²⁺ for WT sTnC, sTnC^F27W, M80Q sTnC and M80Q sTnC^F27W in complex with purified native sTnI and sTnT at 15°C. Y-axis is displayed in volts. Traces are fitted with a single exponential curve from which Ca²⁺ dissociation rates (k_off) are determined except for M80Q sTnC^F27W–sTn where a double exponential curve with fixed slow C-terminal rate was used (see Methods). Traces are the average of three runs and are vertically spaced for clarity.

Steady-state force–Ca²⁺ relationships of mutant sTnC-reconstituted fibres

Measurement of the steady-state force–pCa relationship for each skinned rabbit psoas fibre was completed before extraction of native sTnC (see Methods) and after reconstitution of Tn complexes with purified native or recombinant sTnC (Fig. 3). Reconstitution with native purified sTnC (control) resulted in an F_max that was 96 ± 1% of pre-extracted values, demonstrating the ability to completely re-occupy Tn complexes. Fibres reconstituted with 100% recombinant mutant sTnC showed some reduction in F_max compared with control (100% purified sTnC; Table 2). To determine whether all Tn complexes were occupied, recombinant sTnC-reconstituted fibres were incubated for two additional 1 min periods in 100% purified native sTnC. No further change in F_max confirmed that all Tn complexes had been occupied with recombinant sTnC. Possible reasons for lower F_max with recombinant mutant sTnC are discussed below.

Fully reconstituting fibres with the different sTnC mutants clearly altered Ca²⁺ sensitivity of force (pCa₅₀). There were no differences between fibre groups in pCa₅₀ prior to sTnC extraction. Reconstitution with 100% purified sTnC did not change pCa₅₀ or slope (Table 2), demonstrating that the extraction–reconstitution protocol had no effect on Ca²⁺-dependent activation of steady-state force. All sTnC mutants had an increased pCa₅₀ of the force–pCa relationship (Fig. 4). Paired comparisons were made within fibre groups for pre-extraction versus post-reconstitution conditions to determine magnitude changes in pCa₅₀ (ΔpCa₅₀) with individual sTnC mutants. ΔpCa₅₀ increased from 0.08 with M80Q sTnC, to 0.15 with sTnC^F27W and to 0.32 with M80Q sTnC^F27W (Fig. 4 and Table 2). These increases in pCa₅₀ are inversely related to decreases in solution k_off and K_d, indicating that progressively greater Ca²⁺ affinity of sTnC may result in a progressively greater increase in the Ca²⁺ sensitivity of steady-state force. It is important to note that the F27W mutation, which has been used as a reporter of confirmation change with Ca²⁺ binding (Pearlstone et al. 1992; Chandra et al. 1994; Johnson et al. 1994; Tikunova et al. 2002; Davis et al. 2004), slows k_off (by ∼20% at 15°C in whole sTn) and alters fibre mechanics (increases pCa₅₀ by 0.15). It is worth noting that sTnC^F27W is used in this study both as a reporter of sTnC conformational changes with Ca²⁺ (Table 1) and as a tool to study functional effects in fibres (Fig. 4B and Table 2).

Figure 4.

Force–pCa relationships for fibres reconstituted with mutant rabbit sTnC Extraction of native sTnC was followed by reconstitution with M80Q sTnC (A), sTnC^F27W (B) or M80Q sTnC^F27W (C). Each panel shows normalized force as a function of pCa and clearly demonstrates that the Ca²⁺ sensitivity of force (pCa₅₀) gradually increases by 0.08 to 0.15 to 0.32 in A, B and C, respectively, as the Ca²⁺ dissociation rate (k_off) of the mutant sTnC decreased (see Table 1). Some error bars are smaller than symbols; see Table 2 for reconstituted F_max, pCa₅₀ and n_H values.

Fibres fully reconstituted with the different sTnC mutants had slightly reduced n_H, a measure of the cooperativity of force generation. ANOVA between groups reconstituted with M80Q sTnC, sTnC^F27W or M80Q sTnC^F27W showed no statistical difference. However, in paired comparisons (pre-extraction versus post-reconstitution) within fibre groups, n_H was unchanged for purified sTnC (control) and reduced for each mutant sTnC (P < 0.05; Table 2). These data suggest that some loss of cooperative activation probably occurred as a result of the altered Ca²⁺-binding properties or altered quaternary structure of the sTnC mutants. In spite of this slight reduction in n_H, these sTnC mutations had a dramatic Ca²⁺-sensitizing effect on force in skeletal muscle fibres.

Mutant sTnC: xxsTnC mixtures in fibres

To determine whether sTnC Ca²⁺-binding properties influence cooperative interaction between structural RUs along thin filaments during Ca²⁺ activation, fibres were reconstituted with mixtures of WT or mutant rabbit sTnC and xxsTnC. Steady-state force–pCa relationships for mixtures of M80Q sTnC^F27W and xxsTnC (denoted as sTnC: xxsTnC) showed reduced cooperative activation and force generation (Fig. 5A and Table 3). Prior to extraction, pCa₅₀ and n_H were not different between groups of fibres. Following reconstitution, F_max was reduced as the content of M80Q sTnC^F27W in the reconstitution mixture was reduced (and correspondingly, xxsTnC content increased). For these experiments, pCa₅₀ increased (leftward shift) by 0.25 for the 100: 0 mixture (compared with 0.32 in Table 2) and this was progressively reduced for all other mixtures, with a maximum rightward shift of 0.51 for only 20% M80Q sTnC^F27W (80% xxsTnC) in the reconstitution mixture (20: 80; Table 3). n_H also decreased with diminishing M80Q sTnC^F27W content of the mixture by paired comparisons (although no difference was found between groups by ANOVA), similar to previous observations for sTnC: xxsTnC mixtures (Regnier et al. 2002). These data suggest a reduced ability for near-neighbour RU interactions to contribute to cooperative force generation. n_H is not reported for the 20: 80 mixture because these fibres had a very low reconstituted F_max and often exhibited a step-like on/off response to submaximal Ca²⁺ concentrations. For each M80Q sTnC^F27W: xxsTnC mixture, F_max was lower than the percentage of M80Q sTnC^F27W in the protein incubation solution and was only 7% of pre-extracted F_max for 20: 80 M80Q sTnC^F27W: xxsTnC (compared with 23%F_max for 20: 80 sTnC: xxsTnC). Fibres reconstituted in 20: 80 mixtures of M80Q sTnC: xxsTnC or sTnC^F27W: xxsTnC produced only 14% or 17%F_max, respectively (data not shown). Although this result was not expected for sTnC mutants with greater Ca²⁺-binding affinity, it could indicate a reduced ability of the mutant proteins to activate thin filaments compared to native sTnC at maximal [Ca²⁺] (in agreement with slightly reduced F_max in 100: 0 mixtures).

Table 3.

Force–Ca²⁺ parameters for fibres reconstituted with M80Q sTnC^F27W: xxsTnC mixtures

	Pre-extracted		Reconstituted
M80Q sTnCF27W: xxsTnC (n)	pCa50	nH	Fmax*	pCa50	nH	ΔpCa50†
100: 0 (5)	5.92 ± 0.02	2.9 ± 0.3	0.85 ± 0.02	6.17 ± 0.02	2.2 ± 0.1	0.25 ± 0.02
80: 20 (4)	5.97 ± 0.01	2.9 ± 0.1	0.57 ± 0.03	5.91 ± 0.09	1.9 ± 0.1	−0.07 ± 0.09
60: 40 (5)	5.96 ± 0.01	2.8 ± 0.1	0.33 ± 0.06	5.75 ± 0.05	1.8 ± 0.9	−0.21 ± 0.05
40: 60 (3)	6.00 ± 0.01	2.8 ± 0.1	0.19 ± 0.03	5.65 ± 0.03	1.7 ± 0.2	−0.35 ± 0.03
20: 80 (3)	5.98 ± 0.01	3.0 ± 0.1	0.07 ± 0.01	5.47 ± 0.11	—	−0.51 ± 0.11

*Fraction of reconstituted F_max compared to pre-extracted F_max (normalized to 1.0). †Change in pCa₅₀versus pre-extracted native sTnC value calculated for each fibre and reported as average ±s.e.m. within group of n = 3–5 fibres. All reconstituted values are significantly different (P = 0.01) by paired t test versus pre-extracted native sTnC within each group of n = 3–5 fibres, except for pCa₅₀ of 80: 20 ratio. All reconstituted F_max and pCa₅₀ values are different (P < 0.05) between groups, except pCa₅₀ values of 80: 20 versus 60: 40 and 60: 40 versus 40: 60. n_H values do not significantly differ by ANOVA.

An alternative explanation for unexpectedly low F_max values for mixtures containing mutant sTnC and xxsTnC is that different relative binding affinities between the mutants and xxsTnC result in a different ratio of proteins that bind to TnI–TnT complexes in fibres. We previously showed that purified sTnC and xxsTnC bind to TnI–TnT complexes with similar affinity in fibres (Regnier et al. 2002), but this may be altered if sTnC mutants bind with different affinities from sTnC and xxsTnC. To test the relative binding affinity of sTnC mutants for TnI–TnT complexes in fibres, native sTnC was extracted and Tn complexes were reconstituted by short interval incubations in solutions of low [sTnC] (0.1 mg ml⁻¹), each followed by a measurement of F_max (pCa 4; Fig. 5B). Reconstitution with each sTnC was considered complete when successive incubations no longer increased F_max. In a subset of these experiments, additional incubations in a high concentration of purified sTnC (1 mg ml⁻¹; 2 incubations for 2 min and 1 min) did not change F_max, confirming that all TnI-TnT complexes were occupied with M80Q sTnC^F27W (black stars, Fig. 5B). The force plateau occurred at ∼0.8 F_max with M80Q sTnC^F27W, similar to previous measurements with 100% M80Q sTnC^F27W (Tables 2 and 3). Extraction of M80Q sTnC^F27W and reconstitution with 1 mg ml⁻¹ purified sTnC (filled diamonds, Fig. 5B) showed that force recovered to ∼0.8 F_max, suggesting that some rundown occurred because of the long protocol and a second extraction–reconstitution of sTnC. Force curves were well fit by a single exponential rising curve (see legend to Fig. 5). The time constant for force restoration (1/b) was 8.14 ± 1.59 min for M80Q sTnC^F27W, which was approximately three times slower than for purified sTnC (2.48 ± 0.11 min). This force assay demonstrated that M80Q sTnC^F27W had a lower affinity than sTnC for skeletal TnI–TnT complexes in muscle fibre thin filaments. We verified that the single mutant sTnC^F27W also had a reconstitution time constant different from that of purified sTnC (3.01 ± 0.55 min; data not shown). These data suggest that sTnC mutant content is lower in the fibre than in the protein incubation mixture.

Figure 5C shows how F_max varies with the fraction of functional sTnC in fibres. The unity line reflects expected F_max if each functional Tn complex activates a seven actin length of thin filament, equal to the structural RU size. As previously shown for purified sTnC (Regnier et al. 2002), mixtures of WT sTnC: xxsTnC resulted in a curve that was above the unity line, suggesting that more than seven actin monomers are activated by Ca²⁺ binding within an RU. It is odd that data for M80Q sTnC^F27W were below the unity line. This could indicate that the spread of activation along the thin filament with Ca²⁺ binding to individual Tn complexes was reduced. However, the results shown in Fig. 5B suggest the mixture ratio content did not accurately reflect the fibre content of functional sTnC. Thus, a second-order non-linear binding affinity calculation was used to estimate the protein composition in fibres assuming a 3-fold slower protein binding rate for M80Q sTnC^F27Wversus xxsTnC across the range of protein ratio mixtures used for these studies (eqn (3) in Appendix). These calculations transform the F_max data for M80Q sTnC^F27W: xxsTnC mixtures (open squares, Fig. 5C) by adjusting the solution mixture content value (x axis) to reflect predicted content in fibres (grey stars, Fig. 5C). This transformation shifts all data points above the unity line (arrows, Fig. 5C), as we find for sTnC: xxsTnC mixtures. Therefore, the transformed data suggest that similar fractions of either WT or M80Q sTnC^F27W incorporated into thin filaments allow for Ca²⁺ to activate muscle fibre force production to a comparable extent.

To study the effect of sTnC mutants on the Ca²⁺ dependence of force development when near-neighbour RU interactions along thin filaments were greatly reduced, we reconstituted fibres with mixtures of sTnC: xxsTnC that produced ∼0.2 of the pre-extracted F_max. To achieve 0.2 F_max, solution protein mixtures were approximately 20: 80 for sTnC: xxsTnC, 25: 75 for M80Q sTnC: xxsTnC and 40: 60 for M80Q sTnC^F27W: xxsTnC. Force–pCa relations in fibres were determined (Fig. 6) and examination of pCa₅₀ and n_H values between xxsTnC mixtures (Table 4) suggests that sTnC mutants modulate Ca²⁺ sensitivity of force and not cooperativity (as indicated by n_H) at this low level of force. Reconstitution with purified native sTnC: xxsTnC decreased pCa₅₀ by 0.35 and reduced n_H, similar to previous results (Regnier et al. 2002), suggesting that interactions between functional RUs along thin filaments were reduced. However, some interactions may persist, as n_H was not reduced to the value of 1.7 that we obtained previously using 15: 85 sTnC: xxsTnC reconstituted fibres (Regnier et al. 2002). Matching the force level at 0.2 F_max using M80Q sTnC or M80Q sTnC^F27W in mixtures with xxsTnC was important for creating similar thin filament activation levels. Under these conditions, ΔpCa₅₀ was less for fibres reconstituted with M80Q sTnC: xxsTnC (−0.25) and M80Q sTnC^F27W: xxsTnC (−0.10) compared to sTnC: xxsTnC (Table 4). In other words, the force–pCa curve shifted to the left with M80Q sTnC: xxsTnC and M80Q sTnC^F27W: xxsTnC versus sTnC: xxsTnC by +0.13 and +0.28, respectively. This magnitude of increase in pCa₅₀ in xxsTnC mixtures was similar to that of fibres fully reconstituted with each sTnC mutant (Table 2). It is interesting that for mutant sTnC: xxsTnC mixtures to 0.2 F_max, n_H was the same for sTnC and mutants, suggesting that altered Ca²⁺-binding properties of sTnC do not affect cooperative activation within RUs in the absence of near-neighbour RU interactions. Together these data demonstrate that pCa₅₀ is determined (at least partially) by the Ca²⁺-binding properties of sTnC within individual RUs and by the extent of near-neighbour RU interactions, whereas n_H is determined primarily by the extent of near-neighbour RU interactions and much less by the Ca²⁺-binding properties of sTnC.

Figure 6.

Force–pCa relationships for fibres reconstituted with mixtures of mutant sTnC and xxsTnC to 0.2 F_max Normalized force–pCa relationships are shown for 100% sTnC (control; ○) and mixtures with xxsTnC resulting in 0.2 F_max: purified sTnC: xxsTnC (grey circles), M80Q sTnC: xxsTnC (grey triangles) and M80Q sTnC^F27W: xxsTnC (grey squares). Inset shows reconstituted F_max for 100% purified sTnC (○), sTnC: xxsTnC (grey circle), M80Q sTnC: xxsTnC (grey triangles) and M80Q sTnC^F27W: xxsTnC (grey squares). See Table 4 for values.

Inhibition of strong crossbridge formation with BDM

We next sought to determine how increasing Ca²⁺-binding affinity under conditions of reduced near-neighbour RU interactions influences the crossbridge contribution to thin filament activation. We recently demonstrated that 10 mm BDM decreased F_max∼4-fold more in skeletal muscle fibres reconstituted with 20: 80 sTnC: xxsTnC mixtures compared with native sTnC control (Gillis et al. 2007). The conclusion from these experiments was that when near-neighbour RU interactions were minimized, thin filament activation became more dependent on strong crossbridge binding. Figure 7 shows how inhibition by BDM differs between sTnC and M80Q sTnC^F27W when near-neighbour RU interactions are reduced (with xxsTnC to 0.2 F_max). Plotting normalized F_maxversus BDM concentration allows comparison of the extent of force inhibition under each thin filament condition. The sensitivity of F_max to increasing [BDM] prior to extraction (filled circles; inhibition constant (K_i), 12.3 ± 0.4 mm) was unaltered by reconstitution with 100% sTnC (open circles; K_i, 11.0 ± 1.0 mm) and slightly reduced for reconstitution with 100% M80Q sTnC^F27W (K_i, 8.5 ± 2.2 mm; P = 0.09; data not shown). Force inhibition was more sensitive to BDM for fibres reconstituted with 20: 80 sTnC: xxsTnC mixtures (grey circles; K_i, 2.6 ± 0.4 mm) compared with control (open circles), in agreement with previous work. To compare M80Q sTnC^F27W at a similar level of thin filament activation, ∼0.2 F_max was reached by reconstituting fibres with 40: 60 M80Q sTnC^F27W: xxsTnC (inset, Fig. 7). It is surprising that M80Q sTnC^F27W: xxsTnC fibres showed no change in sensitivity to BDM inhibition (grey squares; K_i = 2.0 ± 0.2 mm) compared with 20: 80 sTnC: xxsTnC (grey circles). In comparison, for measurements with 40: 60 sTnC: xxsTnC incubation mixtures, fibre F_max was ∼0.4 (inset, Fig. 7) and sensitivity of force inhibition by BDM was reduced for 40: 60 mixtures (open triangles; K_i, 4.2 ± 0.3 mm; P < 0.02 versus 0.2 F_max mixtures). Together these data indicate that the sensitivity of force inhibition to BDM was correlated with the number of functional RUs in the thin filament and suggest that enhancing the Ca²⁺ affinity of sTnC does not alter the strong crossbridge contribution to thin filament activation.

Discussion

This study was designed to investigate how enhancing the Ca²⁺-binding affinity of sTn influences the myofilament protein interactions in cooperative contractile activation of skeletal muscle. Specifically we studied how near-neighbour RU interactions along the thin filament and strong crossbridge formation were altered when the Ca²⁺ affinity of sTnC was increased. By engineering specific point mutations into sTnC, we were able to increase Ca²⁺ affinity (reported as K_d), which was changed via slowed Ca²⁺k_off. The sTnC mutants were then incorporated into sTn complexes in skinned fibres (following extraction of native sTnC) to examine their influence on the cooperativity of force generation. The main findings were: (1) complete reconstitution of fibres with sTnC containing a mutation that slowed k_off increased the Ca²⁺ sensitivity of steady-state force (pCa₅₀) but did not increase the slope of the force–pCa relationship (n_H); (2) when near-neighbour RU interactions were reduced (mutant sTnC: xxsTnC to produce 0.2 F_max), n_H was reduced and Ca²⁺ sensitivity of steady-state force decreased by 0.35–0.40 for each sTnC mutant from its own 100% reconstitution value; and (3) with reduced RU interactions, enhancing the Ca²⁺ affinity of sTnC did not alter the strong crossbridge contribution to thin filament activation. Below we discuss these findings and how they give insight into cooperative thin filament activation and force development in skeletal muscle.

Ca²⁺-binding properties of sTnC influence the Ca²⁺ sensitivity of force

The point mutations in sTnC were chosen based on previous work with chicken sTnC (Tikunova et al. 2002; Davis et al. 2004) but in this study species isoform (rabbit) was maintained between Tn subunits and other sarcomeric proteins in the fibre. The selected mutations at positions 80 and 27 created a graded decrease in Ca²⁺k_off for isolated sTnC and these rates decreased in the presence of a TnI peptide (TnI_96-148), in agreement with previous work (Tikunova et al. 2002; Davis et al. 2004). Here we report k_off rates measured in whole sTn complex (Fig. 2 and Table 1) which agree well with those previously reported for sTnC in complex with either TnI_96-148 or whole TnI (Davis et al. 2004).

With incorporation of sTnC mutants into skinned rabbit psoas fibres via sTnC extraction and reconstitution, Ca²⁺ sensitivity of steady-state force (pCa₅₀) increased for each sTnC mutant (Fig. 4 and Table 2) demonstrating that reduction of k_off correlates with increased pCa₅₀. However, increasing the Ca²⁺-binding properties of sTnC does not appear to enhance cooperative interactions of myofilament proteins in activation of the thin filament and force development. Indeed, cooperative mechanisms may actually be somewhat reduced for fibres containing these mutants with increased Ca²⁺-binding affinity. One possibility is that communication between crossbridges and sTnC Ca²⁺ binding was disrupted, although this feedback has been shown to be minimal in skeletal muscle (Fuchs & Wang, 1991; Martyn & Gordon, 2001). Another possibility is that the structural change in the N-terminal domain of sTnC associated with decreased K_d (Table 1) results in a somewhat reduced affinity for sTnI in the presence of Ca²⁺, as previously suggested for chicken M82Q sTnC^F29W (Davis et al. 2004). This idea of altered TnC–TnI interaction is supported by a reduced F_max and n_H for fibres reconstituted with M80Q sTnC^F27W.

Ca²⁺-binding properties of sTnC have little effect on near-neighbour RU interactions and the cooperativity of steady-state force generation

By reconstituting fibres with increasingly low fractional content of sTnC mutants (and high xxsTnC content) we were able to study the influence of sTnC mutants with altered Ca²⁺-binding properties on the size and behaviour of increasingly isolated RUs in the sarcomere (Fig. 5). This procedure allows one to study the Ca²⁺-activation properties in the absence of the dominant form of cooperativity in skeletal muscle (i.e. near-neighbour RU interactions). sTnC mutants could alter the Ca²⁺-activation signalling process either by changing the sTnC–sTnI interaction (as stated above) or by reducing the functional unit (FU) size of regulatory units. Here the FU size is defined as the number of myosin binding sites on actin made available when Ca²⁺ binds to individual sTn complexes in the thin filament (also referred to as the spread of activation along the thin filament). We previously determined that the size of the FU is likely to be 10–12 actin monomers in skeletal muscle (Regnier et al. 2002). It is possible that mutant sTnCs could reduce this FU size, which would reduce n_H (the apparent cooperativity of force production). This was suggested previously by Moreno-Gonzalez et al. (2005) when skinned psoas fibres were reconstituted with either D28A sTnC (where Ca²⁺ does not bind to site I) or cardiac TnC to reduce the Ca²⁺ component of thin filament activation. In these fibres, n_H was significantly reduced. When these proteins were complexed with sTnI and sTnT to make whole Tn, k_off was significantly increased, suggesting a reduced Ca²⁺-binding affinity (Moreno-Gonzalez et al. 2007).

In our current experiments, the manipulation of sTnC is essentially the opposite – an increase in the Ca²⁺ component of thin filament activation by increased Ca²⁺ affinity of sTnC (with decreased k_off). Because near-neighbour RU interactions play a dominant role in cooperative activation, we used our previous technique of reducing near-neighbour RU interactions (Regnier et al. 2002) to examine the effects of altered sTnC affinity within a local RU environment. We verified that reconstitution of thin filaments with progressively reduced mixture ratios of M80Q sTnC^F27W: xxsTnC shows a progressive decrease in n_H to a minimum (n_H = 1.7) at a force level of 0.2 F_max (Fig. 5A and Table 3). At maximal [Ca²⁺], F_max data at different ratio mixtures (Fig. 5C) suggest that FU size may be maintained at more than seven actin monomers when values are corrected for unequal relative binding affinities of M80Q sTnC^F27Wversus xxsTnC (Fig. 5B and Appendix). These results suggest that (1) reduced n_H with mixtures of mutant M80Q sTnC^F27W: xxsTnC is caused primarily by loss of near-neighbour RU interactions and not reduction of FU size, and (2) comparison between mutants with reduced near-neighbour RU interactions should be made by matching F_max rather than matching protein mixture ratio with xxsTnC. Additionally, when 0.2 F_max was achieved for each mutant sTnC by reconstituting fibres with mixtures of xxsTnC and sTnC, M80Q sTnC or M80Q sTnC^F27W, RU interactions were reduced (low n_H), but there was no difference in n_H between mutants. This supports the idea that enhancing the Ca²⁺-binding properties of sTnC in fibres with greatly reduced RU interactions does not alter n_H (Fig. 6 and Table 4) or increase FU size. Thus, we demonstrated that the Ca²⁺ sensitivity of force in skeletal muscle fibres can be increased by enhancing sTnC Ca²⁺ affinity, and large increases in Ca²⁺ affinity of sTnC can compromise the apparent cooperative activation of thin filaments.

This idea is further supported by our experiments where, during maximal Ca²⁺ activation, the relative influence of strong crossbridge binding in activating the thin filament was studied by force inhibition with increasing [BDM]. Minimizing near-neighbour RU interactions with sTnC: xxsTnC mixtures to 0.2 F_max allowed us to examine Ca²⁺–crossbridge interactions within RUs along the thin filament. The sensitivity of F_max to BDM (i.e. 1/K_i) was increased for both sTnC: xxsTnC and M80Q sTnC^F27W: xxsTnC mixtures to 0.2 F_max, but did not differ between them (Fig. 7). This suggests the enhanced Ca²⁺ binding of M80Q sTnC^F27W did not help maintain thin filament activation as the number of strong crossbridges was reduced by BDM, and that native sTnC can fully activate individual RUs upon Ca²⁺ binding. Together these data suggest that (1) M80Q sTnC^F27W does not alter the overall level of thin filament activation within individual RUs at saturating Ca²⁺, and (2) the contribution of strong crossbridge binding to thin filament activation is not altered by enhanced Ca²⁺ binding with M80Q sTnC^F27W.

In summary, we have provided the first direct evidence that enhanced Ca²⁺ affinity of sTnC (via decreased k_off) activates the thin filament to set the Ca²⁺ sensitivity of steady-state force with no enhancement of the cooperativity of force generation. Therefore, the Ca²⁺-binding properties of native sTnC may be optimal for providing near-neighbour RU interactions and subsequent strong crossbridge formation to cooperatively activate the thin filaments in skeletal muscle.

Appendix

Different reconstitution time constants (1/b) for the mutant sTnC proteins in a sTnC-extracted fibre (Fig. 5B) suggest different binding affinities to TnI–TnT in the fibre. Changes in steady-state affinity between sTnC with point mutations and the inhibitory fragment of TnI (residues 96–148) support this idea (Davis et al. 2004). These differences in affinity between Tn subunits may alter the ratio of proteins when reconstituted into thin filaments (compared with the protein solution mixture), as with a mixture of sTnC and xxsTnC. The following calculations characterize two proteins with different relative affinities binding into Tn complexes. Relative binding affinity (A_r) is defined as A_A/A_B, where A_A and A_B are the affinities of proteins A and B. Given a reconstitution mixture of protein A and protein B, the fraction of protein A (f_A) is [A]/[A]+[B]. This leads to a second-order binding rate to the thin filament for protein A (r_A) and B (r_B) as a function of f_A during reconstitution:

(2)

Normalizing for the ratio of these rates gives the reconstituted fraction of proteins A (p_A) or B (p_B) in the fibre:

(3)

with p_A(f_A) +p_B(f_A) = 1. We use the example of M80Q sTnC^F27W (protein A) compared to xxsTnC (protein B), which was used for the data transformation in Fig. 5C. Data from Fig. 5B show a 3-fold slower binding of M80Q sTnC^F27W into a sTnC-extracted fibre compared with sTnC. The similar binding affinity between sTnC and xxsTnC in a fibre (Regnier et al. 2002) suggests that M80Q sTnC^F27W has a binding affinity one-third of that for xxsTnC, so that A_r = 1/3. The mixture ratio of A:B for f_A = 0.20, 0.40, 0.60 and 0.80 is rescaled to 0.08, 0.18, 0.33 and 0.57, respectively, for protein A incorporated into the thin filament. These values were used in the data transformation of Fig. 5C. Because this calculation was based on the probability of random binding events, it may better reflect what occurs as M80Q sTnC^F27W and xxsTnC compete for binding sites in the thin filament.