| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
, G203S, K206Q) enhance filament sliding1
1 Molekular- und Zellphysiologie, Medizinische Hochschule, D-30625 Hannover, Germany
2 Departments of Radiology
3 Physiology and Biophysics
4 Bioengineering, University of Washington, Seattle, Washington
5 Department of Biological Science, Florida State University, Tallahassee, Florida 32306-4370
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
, G203S, and K206Q in cTnI affect Ca2+ regulation. All three mutants enhanced Ca2+ sensitivity and maximum speed (smax) of filament sliding of in vitro motility assays. Enhanced smax (pCa 5) was observed with rabbit skeletal and rat cardiac (
-MHC or ß-MHC) heavy meromyosin (HMM). We developed a passive exchange method for replacing endogenous cTn in permeabilized rat cardiac trabeculae. Ca2+ sensitivity and maximum isometric force did not differ between preparations exchanged with cTn(cTnI,K206Q) or wild-type cTn. In both trabeculae and motility assays, there was no loss of inhibition at pCa 9. These results are consistent with COOH terminus of TnI modulating actomyosin kinetics during unloaded sliding, but not during isometric force generation, and implicate enhanced cross-bridge cycling in the cTnI-related pathway(s) to hypertrophy. heart; calcium regulation; systole; in vitro motility; troponin exchange
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
Kimura et al. (31) reported genetic linkage studies identifying mutations in the cardiac troponin I (cTnI) gene that lead to FHC. Their initial study identified six mutations in five residues, all in the COOH-terminal third of cTnI (exons 7 and 8). Subsequent reports have identified four additional FHC-related mutations in cTnI, all but one of which are located near the COOH terminus (43, 44, 47). Troponin I (I = inhibitory) is crucial for turning myofilaments "off" when intracellular [Ca2+] is low during diastole (21, 48). Central to this function is the inhibitory peptide region of cTnI in which two mutations were identified at residue R145 (31). Recent work has established that the R145G variant has functional effects consistent with early peptide studies by Van Eyk and Hodges (70) and also the hypertrophic phenotype (9, 19, 29, 35, 67). Although considerable effort has gone into studies of the R145G mutation and structure of the inhibitory region, only limited work has been reported on the other mutations in cTnI, and little is known about the structure of the COOH-terminal region. The work of Takahashi-Yanaga et al. (67) showed that all but the G203S mutation increased Ca2+ sensitivity of ATPase activity and isometric force production, although the reported changes with the K206Q mutation were small. These results contrasted with Burton et al. (9), who showed that G203S enhanced Ca2+ sensitivity of force generation but not ATPase activity, whereas R145G did not affect Ca2+ sensitivity of force. Both Takahashi-Yanaga et al. (67) and Burton et al. (9), and additionally Lang et al. (35), agree that the R145 mutations resulted in significant Ca2+-independent force. Studies that examined isometric force generation, however, incorporated recombinant cTnI into the myofilament lattice using procedures that involve prolonged activation during extraction/reconstitution, which can result in significant decreases in maximum force and/or increases in force at "relaxing" Ca2+ levels with wild-type (WT) cTnI. In one other study, Elliott et al. (19) showed that the R162W mutation increased both Ca2+ sensitivity of actin-Tn-Tm-S1-ATPase activity in solution and also ATPase at very low [Ca2+].
Because only limited and in some instances contradictory information is currently available, we sought to identify changes in myofilament function caused by the most COOH-terminal FHC-related mutations in cTnI which would enable us to infer a mechanism(s) by which these mutations lead to cardiac hypertrophy. Cardiac troponin subunits from rat, WT or with site-directed mutations equivalent to those found in FHC, were expressed in bacterial systems. Recombinant rat cTns were evaluated using Ca2+-regulated in vitro motility assays to quantitatively examine the sliding of individual actin filaments. A procedure was also developed for exchanging the recombinant cTns into permeabilized cardiac ventricular trabeculae to examine effects of mutations on force generation. Our results suggest that the K183, G203S, and K206Q mutations enhance filament sliding and thus may cause hypertrophy via a signaling pathway that responds to increased systolic ATPase activity in cardiomyocytes.
Portions of this work have been published in abstract form (14, 15).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
(codon deletion), G203S or K206Q and also at the x position of the low-affinity, NH2-terminal Ca2+ binding site II, cTnC (D65A) (also referred to as xcTnC), by site-directed mutagenesis using the T7-GEN In Vitro Mutagenesis Kit (US Biochemicals, Cleveland, OH). A vector pET-24 (Novagen, Madison, WI) containing the T7 promoter, lac operator, and a kanamycin resistance gene was used for the expression of the respective WT or mutant clones in Escherichia coli, and the protein was extracted from bacterial cells as described for rat cardiac cTnC (17) and purified according to methods described for native cTnC, cTnI, or cTnT (52). Although the COOH-terminal residue numbering is different between rat and human cTnI because of the additional residue A24 in the NH2 terminus of rat cTnI, we retain the nomenclature of the human sequence to avoid confusion with clinical literature. For a limited set of control studies, bovine cTnT was purified from fresh heart tissue obtained from a local abattoir according to previously described methods (52). Concentration and purity of troponin subunits and all other proteins described below were evaluated by UV absorbance and SDS-PAGE, respectively (20).
Rhodamine labeling of cTnT and xcTnC.
The fluorescent probe, the 5' isomer of iodoacetamidotetramethylrhodamine (5'-IATR), was a generous gift of Dr. John Corrie (National Institute for Medical Research, Mill Hill, London). The recombinant cTnC mutant, xcTnC, was labeled with 5'-IATR under denaturing conditions as previously described for sTnC (41). Native bovine cTnT was similarly labeled at Cys39 under denaturing conditions. Labeling was 0.2 mol of rhodamine per mol of xcTnC or 0.4 mol of rhodamine per mol of cTnT.
Troponin complex.
cTn was purified from frozen rat hearts (Pel-Freez, Rogers, AZ) as described for bovine heart (52). Cardiac Tn complex was reconstituted from isolated (recombinant or native) subunits (1:1:1 ratio) as described for native Tn subunits (52).
Myosin, actin, and tropomyosin.
Skeletal myosin and heavy meromyosin (sHMM) were prepared from rabbit back and leg muscles as described previously (13, 20). Myosin was stored in 50% vol/vol glycerol at -20°C for up to 6 wks; sHMM was stored at 0–4°C for up to 1 wk. Cardiac myosin and HMM (cHMM) were prepared as previously described (57) from control rat hearts or from rats treated with 0.8 mg/ml propylthiouracil (PTU) in their drinking water. Freshly prepared cardiac myosin was used immediately to make cHMM, and cHMM was used within 3 days of preparation. Cardiac myosin and HMM were stored at 0–4°C. At the beginning of each day of motility experiments, ATP-insensitive heads were removed from an aliquot of HMM by ultracentrifugation (20, 34). A 1.5-fold molar excess of F-actin was added to HMM, followed by 1 mM ATP and ultracentrifugation at 513,000 g and 4°C (model TLX 120.2; Beckman, Fullerton, CA). Competent HMM (supernatant) was then diluted to 250 µg/ml (determined using the Bradford assay due to the presence of ATP and ADP), the concentration used in motility experiments. Actin and tropomyosin (Tm) were prepared from rabbit skeletal muscle ether powder as previously described (13, 20, 57) using the methods of Pardee and Spudich (50) and Smillie (62), respectively. F-actin was labeled with rhodamine-phalloidin (RhPh) as described by Kron et al. (34) for visualization by fluorescence microscopy.
In vitro motility assay.
In vitro motility assays were carried out with regulated actin as previously described (20, 56) with minor modifications. Fundamental aspects of the motility assay with rabbit skeletal HMM (13, 20, 56) or rat cardiac HMM (57) and data analysis were performed according to established procedures in our laboratory. Glass microscope slides and no. 1 thickness coverslips were cleaned by sonication in 1 mM KOH and then rinsed, sonicated in, and rinsed again with deionized H2O, and finally oven dried. The coverslips were coated on one side with a thin layer of 0.1% nitrocellulose in amyl acetate (Ernest Fullam, Latham, NY). Flow cells were constructed on microscope slides by placing the nitrocellulose-coated coverslips on no. 1
thickness glass spacers with silicone grease (34). After the chamber was readied, the flow cell was completed by infusing a series of solutions (
2x chamber volume, each), the majority of which were left for 1 min in the chamber and then flushed with actin buffer (AB) made without ATP (25 mM KCl, 25 mM imidazole, 4 mM MgCl2, 1 mM EGTA, 1 mM DTT, pH 7.4) (34) before infusing the next solution. All solutions were allowed to equilibrate to room temperature before infusion into the flow cell to minimize formation of bubbles. HMM was applied to the flow cell first for 1 min. HMM application was immediately followed by 0.5 mg/ml BSA in AB to block nonspecific protein binding. After the chamber was flushed with AB, unlabeled F-actin (
100 µg/ml, sheared by at least 15 rapid passages through a 23-gauge needle) was added. Unbound F-actin was flushed out of the flow cell with AB, then AB with 0.5 mM ATP was added to dissociate remaining unlabeled F-actin from competent HMM on the nitrocellulose-coated surface, thus leaving residual "dead heads" blocked by unlabeled F-actin (34, 59). After again flushing the chamber with AB, 8 nM RhPh-labeled F-actin was applied in the absence of ATP. Labeled actin filaments that did not bind to HMM on the surface were flushed from the chamber with a "wash buffer" that was either AB for assays with unregulated RhPh F-actin or was AB plus 50–250 nM each Tn and Tm for regulated filaments. The concentrations of Tn and Tm in the wash buffer and motility buffer (see below) were the same and were chosen as the minimum needed to maintain regulation of the filaments, i.e., little or no movement at pCa 9 as previously described (20). A difference from our previous methods (20) was that regulated filaments were reconstituted on the flow cell surface by incubating unregulated RhPh-labeled F-actin filaments with AB plus 50–250 nM each of Tn and Tm for 3 min, similar to our recent work with filaments regulated with sTn plus Tm (36).
Last, ATP-containing motility buffer was infused into the flow cell. Motility buffer for regulated actin filaments consisted of 2 mM Mg-ATP, 1 mM Mg2+, 65 mM Na+ + K+, 10 mM EGTA, 8–29 mM propionate, 28–70 mM 3-(N-morpholino)propanesulfonic acid (MOPS), 0.085 M ionic strength (
/2), and 0.5–0.75% (wt/vol) methylcellulose, pH 7.0 at 30°C, the experimental temperature (20). Calcium dipropionate was altered to change the pCa (-log[Ca2+]) between 9.2 and 4.6 as calculated using the National Institute of Standards and Technology (NIST) Critically Selected Stability Constants of Metal Complexes Database. Motility buffer for assays of regulated actin contained 50–250 nM each of Tm and Tn. For all motility buffers, 3 mg/ml glucose, 100 µg/ml glucose oxidase (Sigma, St. Louis, MO), 18 µg/ml catalase (Boehringer-Mannheim, Indianapolis, IN), and 40 mM DTT (Bio-Rad, Hercules, CA) were added to minimize photo-oxidation and photobleaching (34). Motility assays at 30°C were imaged and recorded on videocassettes as previously described (20).
Motion analysis.
Edge-detection hardware and Expert Vision software from Motion Analysis Systems (Santa Rosa, CA) were used to obtain filament speed statistics from videocassette recordings (20, 25, 59). Typically, six fields were analyzed for 1 min each in every flow cell. Data were sampled at 10 frames per second (fps) and individual filament paths, calculated from filament centroids, were retained only when they could be unambiguously tracked for at least 2 s. Speed statistics were calculated for each retained path using the Motion Analysis algorithm. The ratio of standard deviation (SD) to mean speed was calculated for each path as an indicator of uniformity of motion (13, 20, 25, 56, 59). A filament was accepted as moving uniformly when this ratio was <0.5 for 10 fps sampling. For each flow cell (one condition), the fraction of filaments moving uniformly and the unweighted mean speed (±SD) of those uniformly moving filaments (su) was obtained by combining information from all filament paths. If su was < 5 µm/s, then the position data along each path were smoothed with an unweighted moving average filter (5-frame window) and every fifth point retained, yielding an effective sampling rate of 2 fps; a filament was accepted as moving uniformly when the ratio of SD to mean speed was <0.3 for 2 fps sampling. Speed-pCa relations were fit to Eq. 1 by nonlinear least squares regression (SigmaPlot; SPSS, Richmond, CA):
 | (1) |
Cardiac Trabecular Mechanics
Trabecular dissection and mechanical apparatus.
Mechanical experiments were conducted on permeabilized trabeculae from the right ventricular free walls of rat hearts obtained as described previously (22, 57). Adult male Sprague-Dawley rats were euthanized with pentobarbital (50 mg/kg ip). Hearts were rapidly excised and rinsed of blood, and the right ventricle splayed open in oxygenated physiological saline (in mM, 94 NaCl, 24 NaCO3, 5 KCl, 1 MgSO4, 1 Na2HPO4, 0.7 CaCl2) on a chilled dissection stage. The free wall was pinned out and incubated, with one change of solution, overnight at 4°C in skinning buffer (in mM: 100 KCl, 10 MOPS, 5 dipotassium EGTA, 9 MgCl2, 4 ATP, pH 7.0 at 4°C, 1% vol/vol Triton X-100, and 50% vol/vol glycerol). Solution was then changed to skinning buffer without Triton X-100 glycerol for dissection and storage. Individual trabeculae were dissected, and the ends were wrapped in photochemically etched T-clips and stored at -20°C for up to 4 days. Trabeculae were attached via T-clips to a force transducer (model 400A, 2.2-kHz resonant frequency; Cambridge Technology, Watertown, MA) at one end and a servo-motor (model 300; Cambridge Technology) tuned for a 300-µs step response at the opposite end; the mechanical apparatus was mounted on the stage of an inverted microscope (Leitz Diavert, Wetzlar, Germany) equipped for digital imaging (XR-77 CCD; Sony).
Solutions.
Solutions for experiments on permeabilized trabeculae contained (in mM) 15 phosphocreatine (PCr), 15 EGTA, at least 40 MOPS, 1 free Mg2+, 135 Na+ + K+, 1 DTT, 250 U/ml creatine kinase (CK; Sigma), and 5 ATP at pH 7.0 and 15 ± 1°C, the temperature at which mechanical measurements were made. Ionic strength was 0.2 M. For activation solutions, the Ca2+ level (expressed as pCa = -log[Ca2+]) was set to pCa 4.5 or pCa 4.0 by adjusting calcium dipropionate (propionate was the anion used to adjust ionic strength) (40, 57). Solutions were contained in 200-µl wells mounted on a temperature-controlled base. The base holds a total of 12 solution wells. The temperature of groups of four wells could be independently maintained by Peltier thermoelectric chips controlled by an ATR-4 adaptable thermoregulator (Quest Scientific, North Vancouver, BC, Canada) to aid in the cTn exchange protocol.
Isometric force.
Relaxed trabecular sarcomere length (Ls) was set to 2.2 µm with helium-neon laser diffraction (11). Trabecular length was 1.49 ± 0.35 mm and diameter was 147 ± 59 µm (mean ± SD, n = 23). Steady-state isometric force measurements were obtained under relaxing and activating conditions using our previously described data acquisition and control system (11, 12, 57). Trabecular length was shortened (by 20%) and rapidly restretched at 5-s intervals to maintain structural and mechanical integrity of the preparation (6, 10, 65); isometric force measurements were made during the steady-state period between shortening ramps, and the passive force (pCa 9) was subtracted to obtain active force (pCa 
6). Force measurements (F) were then fit by nonlinear least squares regression (SigmaPlot; SPSS) to the Hill equation (Eq. 2)
 | (2) |
Exchange of recombinant cTn into permeabilized trabeculae.
To exchange purified troponins for endogenous troponin in permeabilized cardiac trabeculae, we adapted the method of Brenner et al. (8) devised for rabbit skeletal muscle. After initial control force data had been obtained, the trabecula was transferred first briefly (<1 min) to "pre-rigor" solution (in mM: 10 imidazole, 2.5 EGTA, and 15 EDTA) at 5°C. The fiber was then transferred into rigor solution [relaxing solution with no ATP, no PCr, no CK, and 5 mM 2,3-butanedione monoxime (BDM) added]. BDM was added to inhibit actomyosin force generation (1, 18), and the trabecula was also shortened to further eliminate rigor force. After 30 min in rigor solution, the trabecula was transferred into troponin exchange buffer (in mM: 20 MOPS, 5 MgCl2, 5 EGTA, 240 KCl, 5 DTT, 5 BDM, and 0.02 pepstatin, pH 6.5, plus 1 mg/ml cTn) at 10°C and covered to minimize evaporation or condensation. Typically, the trabecula was incubated in troponin exchange buffer for 120 min to achieve complete exchange (see RESULTS). The incubation time was abbreviated in some experiments to evaluate extent of exchange. At the end of the cTn exchange period, the trabecula was returned to pCa 9 relaxing solution at 15°C, and Ls was reset to 2.2 µm prior to data acquisition.
Fluorescence microscopy and laser-scanning confocal microscopy of trabeculae.
Fluorescence of trabeculae exchanged with cTn (5'-IATR-cTnT) was obtained as previously described for cardiac and skeletal muscle preparations following TnC replacement with 5'-IATR-labeled TnC (41, 42). To avoid nonspecific binding of fluorescently labeled cTn, the exchange protocol was modified by incubating trabeculae with 1 mg/ml BSA (in relaxing solution) for 10 min. Trabeculae that were to be examined by laser-scanning confocal microscopy were chemically fixed at the end of the experiment. For double-labeling experiments, trabeculae were incubated with phalloidin green (Molecular Probes, Eugene, OR) in relaxing solution prior to fixation. Trabeculae were dunked into two washes of rigor solution before being placed in 25 mM glutaraldehyde (in rigor) for 10 min. Confocal images were acquired with a Bio-Rad MRC-600 laser-scanning confocal microscope.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
|
or cTnI,G203S (Fig. 2). Filament sliding was halted at low [Ca2+] for all regulated filaments containing either WT or mutant cTnI (Fig. 2A). The concentration of mutant cTns required to stop motility at pCa 9 was less than or equal to that for WT cTn, suggesting that the mutations do not reduce affinity of cTn for actin-Tm filaments. The maximum speed at high [Ca2+] (obtained from nonlinear regression of the data using Eq. 1) for all three mutants was elevated 47–61% above that for WT (Fig. 2A) and also above the average sliding speed for unregulated F-actin, which does not vary with [Ca2+] (20). Ca2+ sensitivity, as indicated by pCa50, was altered with all three mutants such that less Ca2+ was required to activate filament sliding than with WT cTn (Fig. 2B).
|
- or ß-cardiac myosin is the motor protein rather than the faster rabbit skeletal HMM used in the initial experiments (Figs. 1 and 2). To test this possibility, we measured the maximum sliding speed obtained at low pCa (high [Ca2+]) using HMM purified from control rat hearts [predominantly -myosin heavy chain (-MHC)] or from hearts from PTU-treated rats (predominantly ß-MHC). The maximum speed for regulated filaments was normalized to that obtained with WT cTn for each of the HMM types: 4.14 ± 0.61 µm/s (n = 8) with rabbit skeletal HMM, 1.40 ± 0.51 µm/s (n = 5) with untreated rat cardiac HMM, and 1.02 ± 0.03 µm/s (n = 2) with PTU-treated rat cardiac HMM. The high [Ca2+] data with rabbit skeletal HMM from Fig. 2A was replotted in Fig. 3A after normalization to WT. Figure 3, B and C, illustrates that a similar pattern was obtained with both - and ß-cardiac MHC, although the small sample size with PTU-treated rat cardiac HMM (Fig. 3C) reduced the level of statistical significance. The maximum speed for regulated filaments containing mutant cTnI (K183, G203S, or K206Q) cTn was increased over WT cTn with either skeletal or cardiac HMM by 1.3- to 2.4-fold.
|
Incorporation of cTn(5'-IATR-cTnT).
We first followed the time course of incorporation of cTn composed of recombinant rat WT cTnC and cTnI with purified bovine cTnT labeled with 5'-IATR (Fig. 4). We chose to work initially with cTn containing fluorescently labeled cTnT because of the central role of TnT for anchoring the troponin complex to Tm (21, 51); incorporation of fluorescent cTnT into the sarcomere shows that the entire troponin complex has been incorporated. Bovine cTnT was used because it contains a Cys residue 39, whereas rat cTnT does not. The time course of cTn incorporation, as measured by monitoring total fluorescence over time, is shown in Fig. 4A. Fluorescence increased over the first 90 min, then decreased at 120 min, must likely due to slight photobleaching of the fluorescent label during repeated measurements. Maximum Ca2+-activated force was lower than the initial control in these preparations (27 ± 16%, n = 3), suggesting that the fluorescent probe (or possibly the substitution of bovine for rat cTnT) interfered with activation.
|
2.4 x 10-9 cm2/s. At high magnification, fluorescent bands at the periphery of the trabecula were significantly shorter than the 2 µm expected for complete labeling of the I-band (Fig. 4C). This incomplete labeling at short times is consistent with Tn in the overlap zone being preferentially replaced first (60). In contrast, at 90 min, there was no obvious radial gradient of fluorescence across the trabecula (Fig. 4D), although there are presumptive interstitial spaces that are not labeled. I-bands were fully labeled along their entire length after 90 min of incubation (Fig. 4E). The latter point is particularly evident in a highly stretched region of another preparation labeled for 120 min (Fig. 4F). Z-lines are evident as thin dark stripes within the fluorescent bands of Fig. 4, E and F. This pattern suggests that cTn binding is specific to thin filaments and that nonspecific binding is minimal. We therefore chose 2 h as the routine time for incubation with cTn, similar to that used in the protocol for skeletal muscle (8), to ensure full incorporation of the desired protein.
Recombinant rat WT cTn.
After demonstrating incorporation of endogenous cTn into trabecular preparations, we measured steady-state isometric force-pCa relations before and after exchange with WT cTn (Fig. 5A). These data were acquired to control for effects of the exchange procedure and serve as the baseline against which cTn containing mutant subunits is compared (each preparation, prior to exchange also serves as its own control). WT cTn was prepared from recombinant WT cTnC, WT cTnI, and WT cTnT with rat sequences (METHODS) and thus was similar to the endogenous complex except for presence of NH2-terminal Met residues combined with the absence of an acetylated NH2 terminus, absence of phosphorylated residues, and presence of only a single, adult isoform of cTnT. Figure 5B shows that trabecular structure and striation pattern were well maintained throughout the entire experimental protocol. Following the exchange procedure, Fmax (pCa 4) was 86 ± 10% (mean ± SD) of the initial control (Fig. 5C, inset). Before exchange, pCa50 (Eq. 2) was 5.37 ± 0.10 (mean ± SD, n = 10) and was 5.33 ± 0.09 after WT cTn exchange (Fig. 5C). Hill coefficient n was 6.6 ± 1.5 before and 4.7 ± 1.8 after exchange. Comparable results were obtained in five preparations using native cTn purified from rat heart, suggesting that changes were not due to differences between recombinant and native proteins (Fig. 5C). Overall, these control measurements demonstrate that the procedure for cTn exchange in cardiac trabeculae causes only small changes in Ca2+-activated, steady-state force.
|
2% of the initial control force is present after exchange with cTn containing Rh-xcTnC (Fig. 6A) demonstrating that exogenous cTn functionally replaces almost all of the endogenous cTn during the exchange procedure. Localization of the exogenous cTn to actin filaments was verified by confocal microscopy (Fig. 6, B–D). Colocalization of Rh-labeled cTn (Fig. 6C) and phalloidin-green-labeled actin (Fig. 6B) is clearly evident in the average sarcomere scans of Fig. 6, B and C (Fig. 6D).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
, G203S. and K206Q substantially enhanced Ca2+ sensitivity and, surprisingly, maximum speed of filament sliding in the in vitro motility assay using regulated F-actin containing recombinant subunits of rat cTn. 2) Maximum sliding speed was enhanced irrespective of whether HMM was derived from rabbit skeletal muscle, control rat cardiac muscle (primarily -MHC), or cardiac muscle from PTU-treated rats (primarily ß-MHC). 3) At very low [Ca2+], there was no change in effectiveness of the mutant cTnIs examined to inhibit filament sliding in the in vitro motility assay and force generation of permeabilized trabeculae. 4) Exogenous cTn can be efficiently exchanged for endogenous cTn in permeabilized trabeculae from cardiac muscle. 5) cTnI mutation K206Q did not affect the Ca2+ sensitivity or Fmax of permeabilized trabecular preparations. Taken together, these results suggest an important functional role for the COOH terminus of TnI in determining the duration of actomyosin transitions that limit filament sliding speed, but not isometric force generation, and implicate enhanced contractility in the beating heart as the initial signal in the cTnI-related pathway(s) to hypertrophy.
Procedure for cTn exchange into permeabilized cardiac preparations.
One significant aspect of this study is methodological. We have adapted a method for Tn complex exchange established for skeletal muscle (8) to permeabilized cardiac muscle. This novel method is a preferable alternative to methods using troponin subunits, particularly TnT (23, 24, 61) or orthovanadate (64). Both of these other methods lead to significant active force in the absence of Ca2+ during the exchange procedure, and the latter method does not allow replacement of cTnT. Thus the method reported here will be generally useful for introduction of mutant proteins and proteins modified with a variety of labels (fluorescent, etc.) into cardiac preparations.
Figures 4–6 show that exogenous cTn can be stably incorporated into the correct, functional binding sites on actin filaments of intact cardiac sarcomeres over a time course of 2 h, with minimal change in function or trabecula structure due to the exchange procedure itself or to the minor differences between recombinant troponin subunits and the endogenous proteins. The small decrease in Ca2+ sensitivity observed after exchange with WT cTn (Fig. 5) could not be explained by the absence of phosphorylation of recombinant cTnI. Replacing endogenous cTnI with unphosphorylated, recombinant cTnI would be expected to have either no effect or to increase Ca2+ sensitivity (63), whereas we observed a small decrease in Ca2+ sensitivity (Fig. 5). The changes in Ca2+ sensitivity and force related to the procedure may have resulted from a small loss of TnC (5, 46), although the more likely explanation is a general loss of function during the course of experiments with permeabilized preparations. The key modifications that minimize such loss of function involve reducing rigor force in the exchange buffer (METHODS).
Changes in myofilament function due to cTnI mutations.
The three COOH-terminal mutations in rat cTnI studied in the motility assay (K183, G203S, and K206Q) all caused substantial increases in the Ca2+ sensitivity of filament sliding (Figs. 1 and 2B), a result that is in accord with observed enhancements of Ca2+ sensitivity of solution ATPase activity by the R145G, R145Q, R162W, K183, and K206Q mutants (19, 67) but not with reports of the G203S mutation having no effect on ATPase (9, 67). A reduction in the effectiveness of TnI inhibition (i.e., less Ca2+ is required to turn on the filament) under unloaded conditions (filament sliding or acto-S1-ATPase activity in solution) is consonant with the studies of COOH-terminal truncation mutants of cTnI by Rarick et al. (54) and of skeletal TnI by Van Eyk et al. (71). Figures 1, 2A, and 7 all illustrate that the single residue mutations in this study were all effective at turning off actomyosin in the absence of Ca2+, also indicating complete reconstitution of regulated thin filaments, which differs from the effects observed in TnI truncation studies. Results with K183, G203S, and K206Q mutants also differ from those with R145G that suggested the possibility of diastolic dysfunction for that specific mutation (9, 19, 35, 70).
The most dramatic and surprising effect of cTnI mutants K183, G203S, and K206Q introduced into recombinant rat cTn was the large enhancement of smax (Figs. 1, 2A, and 3). Intriguingly, the enhancement was not only above smax for WT Tn but was also above the speed of unregulated F-actin alone (Fig. 2A). The speed of unregulated actin is similar to smax for filaments regulated with rat WT cTn (Fig. 2A) (25), although only the speed of regulated F-actin varies with [Ca2+] (20, 25).
At first glance, it seems surprising that the large enhancements obtained in the motility assay with rat cTnI,K206Q (Figs. 1–3) were not also observed in force-pCa relations of trabeculae (Fig. 7). There is a related example of such disparate effects on isometric force vs. unloaded sliding in the literature. An FHC-related mutation in troponin T, cTnT,I79N, also has been shown to have different effects on maximum sliding speed and isometric force. At saturating [Ca2+], this cTnT mutation increases smax of single filaments in motility assays by 7–50% (26, 37) and also increases Vmax of single fibers by 70% (66). Under isometric conditions, Fmax of single filaments was decreased by 26% (26) and by 27% in single fibers (66). This example in which isometric force decreased differs from the cTnI,K206Q mutation, which did not affect isometric force even though both mutations increased unloaded sliding speed. Significantly, this example shows that differences in isometric vs. unloaded parameters observed in the intact sarcomere persist at the single filament level.
Enhancements of smax observed in this study were not paralleled by increases in maximum ATPase activity in previous studies of the K183 and K206Q mutations (67); although Takahashi-Yanaga et al. (67) observed a small increase in the maximum ATPase with the G203S mutation, Burton et al. (9) did not. In our experiments, this enhancement with mutant cTns was observed with HMM from both skeletal and cardiac muscle sources (Fig. 3) and thus is not an artifact of using fast skeletal myosin. The enhancement of smax by mutant rat cTns is reminiscent of the effect of sTn (20) and supports the suggestion that regulatory proteins can uniquely modulate the kinetics of interaction of myosin cross-bridges with actin (20, 26, 66); it highlights a role for charged residues, particularly basic residues that are mutated in FHC, with G203S being an exception, and thus electrostatic interactions in this process.
Mechanism of increased sliding speed without affecting isometric force.
The example of cTnT,I79N mutation described above suggests that disparity between effects on isometric force vs. unloaded sliding is not unique to mutations in cTnI. How can this apparent contradiction be explained? Isometric force is proportional to f/(f + g), where f is the apparent rate constant for the transition from weak (non-force generating) states to strong (force generating) states, and g is the apparent rate constant for the return of cross-bridges to weak-binding states (7). The latter rate (g) is limited by ADP release. Under isometric conditions when cross-bridges are strained, g is slow (g1x/h in the framework of Ref. 27, where x is the displacement between a cross-bridge’s equilibrium position and the binding site on actin, and h is the maximum work-producing displacement of a cross-bridge); under these conditions, g is typically smaller than f and thus does not have a large effect on Fmax. On the other hand, if we assume that sliding speed in motility assays is proportional to the maximum shortening speed, Vmax, of the intact sarcomere, then filament sliding would be proportional to g2 (again in the framework of Ref. 27). Detachment rate g2 applies to force generating cross-bridges after strain is released during movement. The two detachment rates for isometric and unloaded conditions, g1x/h and g2, respectively, are not the same, and evidence points to g2 in unloaded conditions being substantially faster than g1x/h (27). To summarize, the step in the cross-bridge cycle most likely to increase filament sliding speed without influencing isometric tension, would be an increase in detachment rate g2 (but not g1) (27), and such a modulation of the cross-bridge cycle during unloaded filament sliding is a previously unappreciated function of the COOH terminus of cTnI.
The combined enhancements of smax and pCa50 for filament sliding imply that systolic pumping activity would be markedly enhanced throughout the entire Ca2+ transient. The data in Fig. 7 suggest that the amount of Ca2+ bound to the myofilaments, and by implication, free Ca2+ because TnC is a major site of myoplasmic Ca2+ binding, would not likely change, because Ca2+ affinity as reflected by Ca2+ sensitivity of isometric force was not significantly affected, at least for the cTnI (K206Q) mutation. [Ca2+ sensitivity of force is a better indicator of Tn affinity than filament sliding, because maximum speed for filament sliding is achieved with only partial activation of regulated thin filaments, as evidenced by motility speed-pCa relations (Fig. 1) being shifted leftward by 0.5 (WT) or 0.9 (K206Q) pCa units relative to comparable force-pCa curves (Fig. 7), in agreement with previous studies (20, 26, 36) that used other troponins]. If, as suggested above, g2 is increased by changes in the COOH terminus of TnI, then filament sliding speed would also increase at submaximal [Ca2+] without influencing isometric tension, because g2 is only relevant during filament sliding (27). Clearly, it is important to investigate both isometric conditions and more dynamic conditions, particularly filament sliding, because the heart muscle cells shorten during the ejection period of each beat. Thus the changes observed in the motility assay due to cTnI mutations are not in conflict with the lack of change in isometric force and are particularly relevant to the beating heart of affected individuals.
Mechanism of hypertrophy.
Initial studies of FHC-related mutations in myosin suggested that hypertrophy was the result of a compensatory response to reduced contractility (4, 48, 55). Subsequent studies, particularly those on thin filament Ca2+-regulatory protein mutations cTnT,I79N, cTnT,R92Q, -Tm, D175N, -Tm,E180G, and the cTnI mutations examined in this study, illustrate that hypertrophy could result from enhanced contractility as evidenced by increased filament sliding speed at saturating [Ca2+] (Figs. 1–3 and Refs. 2, 26, 37, 66). Furthermore, the majority of FHC-related mutations in cTnT (32), cTnI (Figs. 1 and 2; and Refs. 9, 19, 35, 67), and -Tm (3, 30) lead to enhanced Ca2+ sensitivity of force and/or filament sliding, thus leading to enhanced cardiac function at [Ca2+] that is physiologically relevant to the cardiac contractile cycle. Such an effect could be mediated directly through elevated ATPase activity, or it could result from kinetically mediated alterations in cooperativity in the thin filament.
The possibility that there could be two or more pathways to muscular hypertrophy is plausible given the wide variety of stimuli that can cause a hypertrophic response in vitro and in vivo (68). Interestingly, recent examinations of MHC mutants R403Q and L908V suggest the possibility of convergence in the mechanisms underlying these seemingly different routes to hypertrophy. Tyska et al. (69) reported that MHC,R403Q and Palmiter et al. (49) reported that both MHC,R403Q and MHC,L908V mutations increase filament sliding speed of unregulated F-actin. MHC,R719W has also been reported to increase isometric force and stiffness in the intact sarcomere (33). These observations have not been reconciled with earlier reports that FHC-related mutations in MHC result in dominant-negative, inhibitory effects on contractility. They support the likelihood that mutations in ß-MHC cause hypertrophy through enhanced activity as is reported here for cTnI mutations. It remains to be determined whether multiple signaling pathways are involved in FHC, or whether there is a common underlying route to hypertrophy for the mutations identified in a wide variety of sarcomeric proteins.
|
DISCLOSURES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
Present address for Y. Chen: UMDNJ RWJ Medical School, Dept. Pathology, Piscataway, NJ 08854.
|
FOOTNOTES |
|---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: P. B. Chase, Florida State Univ., Dept. of Biological Science and Program in Molecular Biophysics, Biology Unit One, Tallahassee, FL 32306-4370 (E-mail: chase{at}bio.fsu.edu).
10.1152/physiolgenomics.00101.2002. 
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESReferences |
|---|
-tropomyosin, Asp175Asn and Glu180Gly, on Ca2+ regulation of thin filament motility. Biochem Biophys Res Commun 236: 760–764, 1997.[Web of Science][Medline]
-tropomyosin (Asp175Asn and Glu180Gly) on the regulatory properties of human cardiac troponin determined by in vitro motility assay. J Mol Cell Cardiol 32: 1489–1498, 2000.[Web of Science][Medline]
-tropomyosin mutation (V95A) is associated with mild cardiac phenotype, abnormal calcium binding to troponin, abnormal myosin cycling, and poor prognosis. Circulation 103: 65–71, 2001.- and ß-tropomyosins. Methods Enzymol 85: 234–241, 1982.
This article has been cited by other articles:
|
|
 |
|
  Y. Wen, J. R. Pinto, A. V. Gomes, Y. Xu, Y. Wang, Y. Wang, J. D. Potter, and W. G. L. Kerrick Functional Consequences of the Human Cardiac Troponin I Hypertrophic Cardiomyopathy Mutation R145G in Transgenic Mice J. Biol. Chem., July 18, 2008; 283(29): 20484 - 20494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Iorga, N. Blaudeck, J. Solzin, A. Neulen, I. Stehle, A. J. L. Davila, G. Pfitzer, and R. Stehle Lys184 deletion in troponin I impairs relaxation kinetics and induces hypercontractility in murine cardiac myofibrils Cardiovasc Res, March 1, 2008; 77(4): 676 - 686. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Gillis, D. A. Martyn, A. J. Rivera, and M. Regnier Investigation of thin filament near-neighbour regulatory unit interactions during force development in skinned cardiac and skeleta muscle J. Physiol., April 15, 2007; 580(2): 561 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Moreno-Gonzalez, T. E. Gillis, A. J. Rivera, P. B. Chase, D. A. Martyn, and M. Regnier Thin-filament regulation of force redevelopment kinetics in rabbit skeletal muscle fibres J. Physiol., March 1, 2007; 579(2): 313 - 326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Schoffstall, N. M. Brunet, S. Williams, V. F. Miller, A. T. Barnes, F. Wang, L. A. Compton, L. A. McFadden, D. W. Taylor, M. Seavy, et al. Ca2+ sensitivity of regulated cardiac thin filament sliding does not depend on myosin isoform J. Physiol., December 15, 2006; 577(3): 935 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M Kruger, S Zittrich, C Redwood, N Blaudeck, J James, J Robbins, G Pfitzer, and R Stehle Effects of the mutation R145G in human cardiac troponin I on the kinetics of the contraction-relaxation cycle in isolated cardiac myofibrils J. Physiol., April 15, 2005; 564(2): 347 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Schoffstall, A. Kataoka, A. Clark, and P. B. Chase Effects of Rapamycin on Cardiac and Skeletal Muscle Contraction and Crossbridge Cycling J. Pharmacol. Exp. Ther., January 1, 2005; 312(1): 12 - 18. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
. B–F: confocal microscope images of cardiac trabeculae incubated for 15 min (B and C), 90 min (D and E), or 120 min (F) with 5'-IATR-labeled cTn. Scale bars represent 10 µm (B and D) and 2 µm (C, E, and F). Insets in B and D show florescence intensity profiles across diameter of the trabecular preparations.
