Functional Effects of Rho-KinaseeCDependent Phosphorylation of Specific Sites on Cardiac Troponin
http://www.100md.com
循环研究杂志 2005年第4期
the University of Illinois at Chicago Department of Physiology and Biophysics and Center for Cardiovascular Research College of Medicine, Chicago, Ill.
Abstract
We tested the hypothesis that activation of Rho-AeCdependent kinase (ROCK-II) alters cardiac myofilament response to Ca2+ by mechanisms involving phosphorylation of thin filament proteins. We determined effects of a constitutively active form of ROCK-II on ATPase activity and tension development in detergent-extracted (skinned) fiber bundles isolated from mouse left ventricular papillary muscles. ROCK-II induced a depression in maximum ATPase rate and tension, which was associated with phosphorylation of troponin T (TnT), troponin I (TnI), and myosin-binding protein C (C-protein). This effect of ROCK-II was retained in fiber bundles isolated from transgenic (TG) mice in which phosphorylation sites (S14, S15, and S19) of myosin light chain 2 were mutated to alanine. Moreover, exchange of ROCK-IIeCphosphorylated Tn complex with the native Tn complex in the fiber bundles resulted in inhibition of maximal Ca2+ activation of tension and ATPase activity. Mass spectrometric analysis demonstrated that ROCK-II phosphorylated cardiac TnI (cTnI) at S23, S24, and T144 and cardiac TnT (cTnT) at S278 and T287. An important role for these cTnT sites is indicated by results demonstrating that ROCK-II induced a depression in tension and ATPase activity in skinned fiber bundles from a TG model in which cTnI is replaced by slow skeletal TnI, which lacks S23 and S24 and in which T144 is replaced by proline. Our data provide the first evidence that ROCK-II phosphorylation of the Tn complex, most likely at cTnT, has an important role in functional effects of signaling through the Rho-A pathway.
Key Words: Rho-kinase G-proteins heart failure troponin T
Introduction
There is evidence that functionally significant phosphorylation of sarcomeric proteins is associated with activation of G-proteineCcoupled receptors and the stimulation of signaling cascades, resulting in hypertrophy.1 For example, Gq signaling and activation of protein kinase C (PKC) not only result in phosphorylation of transcriptional factors but also in altered phosphorylations of myofilament proteins that affect contractility.2 The altered sarcomeric function may give rise to a vicious cycle in which hypertrophy is associated with depressed contractility, eventually leading to malignant growth. Gq is implicated in the activation of a number of signaling pathways, including the small GTPase Rho-A, which signals through its effector Rho-AeCdependent protein kinase (ROCK-II), the focus of this study. Rho-A activation and hypertrophy occur with the major G-proteineCcoupled agonists norepinephrine (NE), endothelin-1 (ET-1), and angiotensin II (Ang II).3,4 Moreover, Rho/ROCK activation occurs with stressors such as rapid pacing, acute pressure overload, and the chronic pressure overload of hypertension in Dahl salt-sensitive rats.5eC7 In the case of cardiac hypertrophy in Dahl salt- sensitive rats, a specific inhibitor of ROCK-II, Y-27632, has been demonstrated to improve contractile function and regress hypertrophy, independently of changes in blood pressure.8 The changes in cardiac myocyte growth associated with Rho-A activation occur with increased phosphorylation of myosin light chain 2 (MLC2), apparently independent of the PKC pathway.5,7 The mechanism appears to involve ROCK phosphorylation of myosin phosphatase targeting subunit 1, inhibition of MLC phosphatase, and thus enhanced phosphorylation of MLC2.9
In the present experiments, we tested the hypothesis that activation of ROCK-II results in functionally significant phosphorylation of thin filament regulatory proteins of cardiac sarcomeres. A rationale behind this hypothesis was the identification of a number of consensus sequences10,11 (RXXS/T or RXS/T) for ROCK-II in cardiac troponin T (cTnT) and cardiac troponin I (cTnI). A consensus sequence in cTnI is located in the inhibitory peptide that is important in tethering cTnI to actin during diastole. We have shown that phosphorylation of this site is a significant determinant of cross-bridge cycling rate and myofilament response to Ca2+.12 We also identified a consensus sequence in the C-terminal region of cTnT. We reported previously that phosphorylation in this region has significant effects on tension generation and the ATPase rate of cardiac myofilaments.13
Thus, our aim in the present study was to determine whether ROCK-II phosphorylates cTnI or cTnT, and, if so, to determine the functional effects of these phosphorylations on Ca2+ dependence of myofilament tension and ATPase rate. Our data provide the first evidence that ROCK-II depresses maximum myofilament tension and ATPase activity as well as Ca2+ sensitivity by a mechanism that appears to be dependent on phosphorylation of cTnT.
Materials and Methods
Animal Model and Fiber Bundle Preparation
We studied preparations from 3-month-old nontransgenic mouse hearts (Charles River) and from hearts of transgenic (TG) mice (generously provided by Dr Jeff Robbins, University of Cincinnati) in which ventricular regulatory myosin light chain 2 (RLC2v) was replaced with a nonphosphorylatable mutant.14 We also used transgenic (TG) mouse hearts in which cTnI was replaced with slow skeletal TnI (ssTnI),15 the embryonic/neonatal isoform. Hearts were excised under deep anesthesia (pentobarbital sodium; 100 mg/kg IP). Fiber bundles (1 to 3 mm in length and 150 to 250 e in diameter) were dissected from left ventricular papillary muscles and treated with 1% Triton X-100 for 2 hours as described previously.16
Myofilament Mechano-Energetic Measurements
We measured isometric tension and actomyosin Mg-ATPase activity simultaneously as described previously.13,16 Sarcomere length was adjusted to 2.2 e. Fiber bundles were equilibrated for 5 minutes in relaxing buffer followed by 3 minutes in preactivating buffer and then in maximal activating solution. The composition of all solutions was as described previously.13 Only fiber bundles retaining 95% of their initial maximum tension were used for analysis. The isometric tension and ATPase activity were determined simultaneously at 20°C in the presence of variable Ca2+ concentrations as described.13 After a series of measurements over a range of values from eClog [Ca2+](pCa) 8.0 to pCa 4.5, we added 1 unit of constitutively active ROCK-II (Upstate) to the fiber bundle in relaxing solution and incubated for 45 minutes at 20°C in the presence of calyculin A (100 ng/mL). We then measured isometric tension and ATPase activity again over the same range of pCa values.
Control experiments showed that this incubation protocol in the absence of ROCK-II had no significant effect on any of the mechano-energetic parameters. Alternatively, we exchanged the native Tn complex in the fiber bundles with recombinant Tn complex13,17,18 with or without previous in vitro phosphorylation by ROCK-II, following a previously described exchange protocol.13 The extent of recombinant Tn exchange into the fibers was determined by SDS-PAGE immunoblot analysis.13
Phosphoprotein Determination
Proteins were separated by 1D-PAGE on 15% polyacrylamide gels as described previously.19 Phosphorylated proteins were detected by PRO Q Diamond stain following the protocol of the supplier (Molecular Probes). Briefly, the gels were fixed in 10% trichloroacetic acid/50% methanol and stained with Pro Q Diamond (1.5 hours). The gel was destained overnight and scanned using an FX Pro Plus imager (Bio-Rad) in conjunction with a CY3 filter set. Subsequently, the gel was stained with 80 ml of SYPRO Ruby for 3 hours (Molecular Probes) to visualize total protein. The gel was destained with 10% methanol, 10% acetic acid for at least 30 minutes, and scanned on the FX Pro Plus imager (Bio-Rad) using a SYPRO filter set.
Matrix-Assisted Laser Desorption/IonizationeCTime of Flight Analysis
ROCK-treated and untreated Tn complexes were digested with either lysyl endopeptidase from Acromobactor (Wako) or Staphylococcus aureus V8 protease (Roche), and phosphopeptides were enriched with an immobilized metal affinity column (IMAC) as described.20 Matrix-assisted laser desorption/ionizationeCtime of flight (MALDI-TOF) analysis was performed using 2',4',6'-trihydroxyacetophenone as a matrix. The presence of phosphates group in a peptide was confirmed by postsource decay mass spectrometry analysis.
Data Analysis and Statistical Evaluation
Individual tensioneCCa2+ and ATPaseeCCa2+ relationships were fit to a modified Hill equation (Graphpad Prism Software) as described previously.13 Tension cost was derived by linear regression of the tensioneCATPase relationship. Statistical evaluation was by 1-way ANOVA and unpaired t test as appropriate; P<0.05 was considered significant.
Results
ROCK-Dependent Phosphorylation Depresses Cardiac Myofilament Response to Ca2+
In the first series of experiments, we determined the effect of ROCK-IIeCdependent phosphorylation on Ca2+ activation of tension and ATPase activity of detergent-extracted fiber bundles of wild-type (WT) mouse. Figure 1A illustrates results from measurements of tension, and Figure 1B illustrates the simultaneously determined ATPase activity. Figure 1C shows a plot of ATPase activity at each level of tension generation and thus gives a measure of the tension cost. The Table summarizes all the data obtained before and after treatment with ROCK-IIeCdependent phosphorylation. Data plotted in Figure 1A illustrate that treatment of cardiac fiber bundles with ROCK-II depressed maximum developed tension by 15% and also resulted in a decrease in Ca2+ sensitivity (pCa50=0.2). Data plotted in Figure 1B indicate that ROCK-dependent phosphorylation also resulted in a decrease in Ca2+ sensitivity (pCa50=0.2) and depression of maximal actomyosin Mg-ATPase activity by 15% (Table). Figure 1C indicates there was a proportional depression in tension and ATPase activity, resulting in no change in tension cost.
A likely mechanism by which the signaling process engaging ROCK-II might affect myofilament activity is by altered levels of myofilament protein phosphorylation. To determine the substrate specificity for ROCK-II in detergent-extracted fiber bundles, we visualized phosphorylated proteins using Pro Q Diamond phospho-specific stain (Figure 2A and 2B). Figure 2A shows protein staining after SDS-PAGE analysis of skinned fiber bundles before and after treatment with ROCK-II, calyculin A, and protein phosphatase 2A (PP2A). Figure 2B shows results of Pro-Q phospho-specific staining of the same gel. Proteins resolved in lane 1 were from control skinned fibers with no treatment. Proteins in lane 2 are from skinned fibers treated with ROCK-II and demonstrate ROCK-IIeCdependent phosphorylation of C-protein, cTnI, and cTnT. Proteins in lane 3 are from skinned fibers incubated with ROCK-II and phosphatase inhibitor calyculin A (100 ng/mL) and in lane 4 from skinned fibers incubated with PP2A (0.01 U) for 45 minutes at 30°C with the reaction stopped by addition of okadaic acid and calyculin A (100 ng/mL). Proteins in lane 5 are from skinned fibers treated as in lane 4 but followed by addition of ROCK-II, indicating phosphorylation of C-protein, cTnT, and cTnI.
To identify the specific sites of ROCK-IIeCdependent phosphorylation, we analyzed complexes of recombinant troponin by mass spectrometric analysis (Figure 3). Our data show that ROCK-II phosphorylation sites of cTnT (TnT4 isoform) are located at the C terminus at S278 and T287, and phosphorylation sites of cTnI are at S23, S24, and T144 (Figure 3A and 3B). These sites of phosphorylation are not present in the control untreated recombinant proteins used in the reconstitution of the troponin complexes.
ROCK-Dependent Inhibition of Myofilament Tension Is Independent of MLC2 Protein Phosphorylation
In view of evidence that activated ROCK-II leads to phosphorylation of MLC2,5 in a second series of experiments, we tested fiber bundles isolated from TG-RLC2v(PeC) mice in which Ser14, Ser15, and Ser19 of MLC2 were mutated to Ala.5 Figure 4A illustrates results from measurements of tension; Figure 4B, the simultaneously determined Mg-ATPase activity; and Figure 4C, a plot of tension as a function of ATPase activity (tension cost). The inset in Figure 4A shows the Pro Q Diamond phospho-specific stain from TG-RLC2v(PeC) mice skinned fibers treated with ROCK-II (lane 1) and nontreated (lane 2) in vitro and indicates an increases in phosphorylation of C-protein, cTnT, and cTnI but no phosphorylation of MLC2. The Table summarizes all the data obtained before and after treatment with ROCK-IIeCdependent phosphorylation. ROCK-II phosphorylation resulted in a significant decrease in the maximum tension (17%) in TG preparations (Figure 4A). As shown in Figure 4A, the Ca2+ sensitivity was altered by ROCK-II treatment (pCa50=0.2). Similarly, ROCK-IIeCdependent phosphorylation resulted in a significant depression of (15%) maximum Mg-ATPase activity and altered Ca2+ sensitivity (Figure 4B). There were no significant differences in the tension cost in fiber bundles from TG-RLC2v(PeC) mouse hearts after treatment with ROCK-II (Figure 4C; Table).
Effects of ROCK-Dependent Phosphorylation of Tn Complex on Myofilament Function
The results reported in Figure 4 indicate that MLC phosphorylation by ROCK-II is not responsible for decreased tension. Thus, we examined whether phosphorylation of proteins in the cTn complex by ROCK-II accounts for its functional effect on cardiac myofilaments. We compared the mechano-energetics of fiber bundles in which native cTn was exchanged with recombinant cTn-WT with fiber bundles in which the native cTn was exchanged with recombinant cTn-WT that was phosphorylated by ROCK-II (p-cTn) in vitro (Figure 5A through 5C). As documented by Sumandea et al,11 the exchange protocol outlined in Materials and Methods results in essentially 100% replacement of the Tn. The fiber bundles subjected to this protocol retained 80% to 90% of the tension generated by untreated controls. Figure 5A through 5C illustrates data obtained from exchange experiments using either WT-Tn complex or p-cTn. The inset in Figure 5A (lane 1) shows phospho-specific staining of skinned fiber bundle exchange with WT-Tn and fiber bundles exchanged with p-cTn (lane 2). Quantification of the staining indicated an increase in cTnI and cTnT phosphorylation but no change in C-protein phosphorylation. The Table summarizes all the data from exchange experiments. The fiber bundles in which native Tn was exchanged with p-cTn complex demonstrated a significantly depressed (29%) maximum developed tension when compared with tension developed by skinned fiber bundles exchanged with control WT-Tn. Similarly, exchange of p-cTn complex resulted in a decrease in Ca2+ sensitivity (pCa50=0.1). ROCK-IIeCdependent phosphorylation also significantly decreased (25%) maximal actomyosin Mg-ATPase activity and altered Ca2+ sensitivity (pCa50=0.1; Figure 5B) without appreciably affecting tension cost (Figure 5C).
ROCK Induces Desensitization to Ca2+ in Myofilaments Containing ssTnI in Place of cTnI
Our mass spectrometry data indicated that the sites for ROCK-II phosphorylation of cTnI are at S23, S24, and T144 sites. Our approach toward understanding the functional significance of phosphorylation of these sites was to use a TG-ssTnI model in which cTnI is replaced by ssTnI.15 ssTnI lacks protein kinase A sites (S23 and S24) and the PKC site at T144, which is replaced by proline. Figure 6A through 6C depicts effects of ROCK-II phosphorylation on mechano-energetics of fiber bundles isolated from TG-ssTnI mouse hearts. Figure 6A and the Table indicate that treatment with ROCK-II induces desensitization to Ca2+ and depression in maximal tension. The inset in Figure 6A shows the Pro Q Diamond phospho-specific stain from untreated TG-ssTnI skinned fibers (lane 1) and fibers treated with ROCK-II in (lane 2). These results demonstrate an increase in phosphorylation of cTnT, but not ssTnI, and C-protein. ROCK-II also decreased Ca2+ sensitivity and depressed maximal ATPase activity (Figure 6B; Table), with no significant changes in tension cost (Figure 6C; Table).
Discussion
Our data are the first to provide evidence that cardiac troponin is a potential downstream effector of Rho signaling in heart. Strong evidence for this conclusion comes from results demonstrating a depression in tension and ATPase in skinned fiber bundles lacking MLC2 phosphorylation sites and in bundles in which native Tn was replaced with Tn phosphorylated previously in vitro by ROCK-II. Although C-protein and troponin may be phosphorylated by ROCK-II, our data demonstrated that the effects of ROCK-II on tension and ATPase rate could occur with no change in phosphorylation of C-protein, as was the case in the preparations in which we exchanged native troponin with WT and p-cTn. Previous studies have emphasized either direct phosphorylation of MLC2 or indirect phosphorylation of MLC2 by inhibition of myosin phosphatase.9,21 In general, these studies have not examined potential modifications in thin filament proteins. Moreover, phosphorylation of MLC2 would be expected to increase myofilament Ca2+ sensitivity with little effect on maximum tension.22,23 In our studies, ROCK-II phosphorylation of MLC2 would offset the depression in Ca2+ sensitivity by MLC2 phosphorylation, with little effect on the decrease in maximum tension induced by phosphorylation of cTnI and cTnT.
The mechanism by which ROCK-IIeCdependent phosphorylation of skinned fiber bundles inhibited tension and ATPase activity appears to involve functionally significant and specific sites on cardiac troponin. In the case of cTnI, the sites, as identified with mass spectrometry, involved S23 and S24 in the unique N-terminal extension and in T144. However, our studies (Figure 6) with fiber bundles missing these phosphorylation sites indicate that the effects of ROCK-II phosphorylation on maximum tension are not likely to be attributable to cTnI phosphorylation. In the case of cTnT, we identified by mass spectrometry a C-terminal peptide containing S278 and T287 as the sites of phosphorylation by ROCK-II. This region of cTnT interacts with cTnC, cTnI, and tropomyosin (Tm) and forms a Ca2+-sensitive domain.24 With binding of Ca2+ to cTnC, cTnI moves a substantial distance on the thin filament, shifting its binding from actin to cTnC.25 These changes are sensed by cTnT in a signaling cascade that is ultimately responsible for movement of Tm and actin subdomains. There is evidence from studies with reconstituted thin filament preparations of fast skeletal muscle that the interaction of TnT with TnC not only acts in concert with TnI to release the thin filament from inhibition but also acts to activate the actineCmyosin interaction.26 Previous studies13,26 indicate that phosphorylation in the C-terminal region of cTnT would be expected to depress maximum tension and ATPase rate, as we have reported here for ROCK-IIeCdependent phosphorylation. Early studies using reconstituted thin filament preparations demonstrated that PKC-dependent phosphorylation of C-terminal residues in cTnT induced a depression in maximum actomyosin ATPase activity.27 More recently, in studies using direct phosphorylation and charge modifications of C-terminal residues of cTnT,13 we reported that these modifications have significant and specific effects on maximum tension and Ca2+ sensitivity of skinned fiber bundles. Results of work presented here support these earlier findings and extend them to include a potential role of regulation of maximum tension via Rho/ROCK signaling.
An important question is: what is the physiological or pathophysiological significance of ROCK-IIeCdependent phosphorylation of troponin This question is based on evidence that the Rho/ROCK pathway is involved in regulation of cardiac function. Pathways upstream of Rho/ROCK are well recognized to be significant in the long-term regulation of cardiac function in the hypertrophic/failure process. NE, ET-1, and Ang II all induce cell growth and bind to receptors that engage a signal transduction path involving Gq.28eC30 For example, treating cardiac myocytes with Ang II induced cardiac hypertrophy associated with activation of Rho-A.4 A specific role of Gq in promoting cardiac hypertrophy and apoptosis associated with decompensated contractile failure has been demonstrated in a TG mouse model with Gq overexpression in the cardiac compartment.31,32 Additionally, TG overexpression of Rho-A in hearts resulted in a cardiomyopathy and ventricular failure associated with atrial enlargement and abnormalities of the sino-atrial and atrio-ventricular nodes.33 Stresses that induce hypertrophy and failure are also associated with Rho/ROCK activation and alterations in myofilament response to Ca2+. One study5 reported increased Rho-A expression in dog hearts stressed by a rapid pacing protocol that induced heart failure. In this model, there was an increase of MLC2 phosphorylation in the failing heart.5 Evidence for an enhanced signaling through the Rho/ROCK pathway was also presented by stimulation of permeabilized single myocytes with phenylephrine, which induced increased myofilament Ca2+ sensitivity in cells from failing hearts but not from the controls. Maximum tension was not measured in that study. A role for Rho-A/ROCK in hypertrophy is also strongly indicated by reports that the specific ROCK-II inhibitorY-27632 is able to ameliorate functional changes in hypertrophic and failing heart preparations induced by hypertension in the Dahl salt-sensitive rat7 by Ang II administration8,34 and by treatment of myocytes with ET-1.35 A role for Rho-A is also indicated from results using adenoviral gene transfer approach and showing that overexpression of dominant-negative mutants of Rho-A inhibited stress-induced hypertrophic response.36 It is also important with regard to the results presented here that after acute pressure overload in hearts, there is an early activation of ROCK-II. Torsoni et al6 reported that acute constriction of the aorta in rats induced a rapid binding of Rho-A with ROCK-II that transiently rose 3.5-fold above the levels in controls within 30 minutes and then fell to a steady-state level of 2-fold. It is also of significance to the data reported here that the pressure overload induced an association of ROCK-II with the Z-disk as well as T-tubules and subsarcolemal sites.6
Although a role for TnT phosphorylation in hearts with activation of the Rho-A signaling cascade/pathway has yet to be demonstrated, on the basis of data presented here, it will be important to re-examine some of the models described above for alterations in Tn phosphorylation. We think existing data do not rule out this possibility because neither determination of Tn phosphorylation nor maximum sarcomeric tension has generally been studied in models in which ROCK-II activation is specifically inhibited, for example by chronic administration of Y-27632.7 On the other hand, there are data indicating that increased thin filament phosphorylation is associated with heart failure in animal models37 and in humans.38,39 At the present time, the specific kinases and phosphatases that give rise to these changes in thin filament phosphorylation are not clear. Our data indicate that a complete investigation of this requires determination of the possible involvement of ROCK-II not only in the end stages but also in the progression to heart failure.
Acknowledgments
This research was supported in part by National Institutes of Health (NIH) grants 5PO1 HL62426-05 Proj1 (R.J.S), 5 RO1 HL64035-05 (R.J.S), and 5 PO1 HL62426-05 Proj4 (P.P.dT.), American Heart Association (AHA) scientist development grant 0230038N (T.K), AHA-Midwest Affiliate predoctoral fellowship award 0315209Z (S.V), and NIH training grants 5 T32 HL07692-14 to 5 T32 HL07692-15 (R.J.S). These studies were submitted in partial fulfillment for the degree of doctor of philosophy. We thank Drs A.F. Martin and B.M. Wolska for their support.
References
Solaro RJ, Montgomery DM, Wang L, Burkart EM, Ke Y, Vahebi S, Buttrick P. Integration of pathways that signal cardiac growth with modulation of myofilament activity. J Nucl Cardiol. 2002; 5: 523eC533.
Solaro RJ. Modulation of cardiac myofilament activity by protein phosphorylation. In: Page E, Fozzard HE, Solaro RJ, eds. Hand Book of Physiology. New York, NY: Oxford University Press; 2001: 264eC300.
Eto M, Barandier C, Rathgeb L, Kozai T, Joch H, Yang Z, Luscher TF. Thrombi suppresses endothelial nitric oxide synthase and upregulates endothelin-converting enzyme-1 expression by distinct pathways: role of Rho/ROCK and mitogen-activated protein kinase. Circ Res. 2001; 89: 583eC590.
Aoki H, Izumo S, Sadoshima J. Angiotensin II activates RhoA in cardiac myocytes: a critical role of RhoA in angiotensin II-induced premyofibril formation. Circ Res. 1998; 82: 666eC676.
Suematsu N, Satoh S, Kinugawa S, Tsutsui H, Hayashidani S, Nakamura R, Egashira K, Makino N, Takeshita A. Alpha1-adrenoceptor-Gq-RhoA signaling is upregulated to increase myofibrillar Ca2+ sensitivity in failing hearts. Am J Physiol Heart Circ Physiol. 2001; 281: H637eCH646.
Torsoni AS, Fonseca PM, Crosara-Alberto DP, Franchini KG. Early activation of p160ROCK by pressure overload in rat heart. Am J Physiol Cell Physiol. 2003; 284: C1411eCC1419.
Satoh S, Ueda Y, Koyanagi M, Kadokami T, Sugano M, Yoshikawa Y, Makino N. J Chronic inhibition of Rho kinase blunts the process of left ventricular hypertrophy leading to cardiac contractile dysfunction in hypertension-induced heart failure. J Mol Cell Cardiol. 2003; 35: 59eC70.
Kobayashi N, Horinaka S, Mita S, Nakano S, Honda T, Yoshida K, Kobayashi T, Matsuoka H. Involvement of Rho-kinase pathway for angiotensin II-induced plasminogen activator inhibitor-1 gene expression and cardiovascular remodeling in hypertensive rats. J Pharmacol Exp Ther. 2002; 30: 459eC466.
Feng J, Ito M, Kureishi Y, Ichikawa K, Amano M, Isaka N, Okawa K, Iwamatsu A, Kaibuchi K, Hartshorne DJ, Nakano T. Rho-associated Kinase of Chicken Gizzard Smooth Muscle. J Biol Chem. 1999; 274: 3744eC3752.
Fukata Y, Oshiro N, Kinoshita N, Kawano Y, Matsuoka Y, Bennett V, Matsuura Y, Kaibuchi K. Phosphorylation of adducin by Rho-kinase plays a crucial role in cell motility. J Cell Biol. 1999; 145: 347eC361.
Matsui T, Maeda M, Doi Y, Yonemura S, Amano M, Kaibuchi K, Tsukita S, Tsukita S. Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J Cell Biol. 1998; 140: 647eC657.
Burkart EM, Sumandea MP, Kobayashi T, Nili M, Martin AF, Homsher E, Solaro RJ. Phosphorylation or glutamic acid substitution at protein kinase C sites on cardiac troponin I differentially depress myofilament tension and shortening velocity. J Biol Chem. 2003; 278: 11265eC11272.
Sumandea MP, Pyle WG, Kobayashi T, de Tombe PP, Solaro RJ. Identification of a functionally critical protein kinase C phosphorylation residue of cardiac troponin T. J Biol Chem. 2003; 278: 35135eC35144.
Sanbe A, Fewell JG, Gulick J, Osinska H, Lorenz J, Hall DG, Murray LA, Kimball TR, Witt SA, Robbins J. Abnormal cardiac structure and function in mice expressing nonphosphorylatable cardiac regulatory myosin light chain 2. J Biol Chem. 1999; 274: 21085eC21094.
Fentzke RC, Buck SH, Patel JR, Lin H, Wolska BM, Stojanovic MO, Martin AF, Solaro RJ, Moss RL, Leiden JM. Impaired cardiomyocyte relaxation and diastolic function in transgenic mice expressing slow skeletal troponin I in the heart. J Physiol. 1999; 517: 143eC157.
de Tombe PP, Steinen GJ. Protein kinase A does not alter economy of tension maintenance in skinned rat cardiac trabeculae. Circ Res. 1995; 76: 734eC741.
Kobayashi T, Dong WJ, Burkart EM, Cheung HC, Solaro RJ. Effects of protein kinase C dependent phosphorylation and a familial hypertrophic cardiomyopathy-related mutation of cardiac troponin I on structural transition of troponin C and myofilament activation. Biochemistry. 2004; 4.
Chandra M, Kim JJ, Solaro RJ. An improved method for exchanging troponin subunits in detergent skinned rat cardiac fiber bundles. Biochem Biophys Res Commun. 1999; 263: 219eC223.
Fritz J, Swartz DR, Greaser ML. Factors affecting polyacrylamide gel electrophoresis and electroblotting of high-molecular weight myofibrillar proteins. Anal Biochem. 1998; 180: 205eC210.
Yang X, Wu H, Kobayashi T, Solaro RJ, van Breemen RB. Enhanced ionization of phosphorylated peptides during MALDI TOF mass spectrometry. Anal Chem. 2004; 76: 1532eC1536.
Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M, Yamamori B, Feng J, Nakano T, Okawa K, Iwamatsu A, Kaibuchi K. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science. 1996; 273: 245eC248.
Noland TA Jr, Kuo JF. Phosphorylation of cardiac myosin light chain 2 by protein kinase C and myosin light chain kinase increases Ca(2+)-stimulated actomyosin MgATPase activity. Biochem Biophys Res Commun. 1993; 193: 254eC260.
Morano I, Hofmann F, Zimmer M, Ruegg JC. The influence of P-light chain phosphorylation by myosin light chain kinase on the calcium sensitivity of chemically skinned heart fibres. FEBS Lett. 1985; 189: 221eC224.
Fisher D, Wang G, Tobacman LS. NH2-terminal truncation of skeletal muscle troponin T does not alter the Ca2+ sensitivity of thin filament assembly. J Biol Chem. 1995; 270: 25455eC25460.
Tao T, Gong BJ, Leavis PC. Calcium-induced movement of troponin-I relative to actin in skeletal muscle thin filaments. Science. 1990; 247: 1339eC1341.
Potter JD, Sheng Z, Pan BS, Zhao J. A direct regulatory role for troponin T and a dual role for troponin C in the Ca2+ regulation of muscle contraction. J Biol Chem. 1995; 270: 2557eC2562.
Noland TA Jr, Kuo JF. Protein kinase C phosphorylation of cardiac troponin I or troponin T inhibits Ca2(+)-stimulated actomyosin MgATPase activity. J Biol Chem. 1991; 266: 4974eC4978.
Knowlton KU, Michel MC, Itani M, Shubeita HE, Ishihara K, Brown JH, Chien KR. The alpha 1A-adrenergic receptor subtype mediates biochemical, molecular, and morphologic features of cultured myocardial cell hypertrophy. J Biol Chem. 1993; 268: 15374eC15380.
LaMorte VJ, Thorburn J, Absher D, Spiegel A, Brown JH, Chien KR, Feramisco JR, Knowlton KU. Gq- and ras-dependent pathways mediate hypertrophy of neonatal rat ventricular myocytes following alpha 1-adrenergic stimulation. J Biol Chem. 1994; 269: 13490eC13496.
Sadoshima J, Izumo S. Signal transduction pathways of angiotensin IIeCinduced c-fos gene expression in cardiac myocytes in vitro. Roles of phospholipid-derived second messengers. Circ Res. 1993; 73: 424eC438.
Adams JW, Sakata Y, Davis MG, Sah VP, Wang Y, Liggett SB, Chien KR, Brown JH, Dorn GW II. Enhanced Galphaq signaling: a common pathway mediates cardiac hypertrophy and apoptotic heart failure. Proc Natl Acad Sci U S A. 1998; 95: 10140eC10145.
Mende U, Kagen A, Cohen A, Aramburu J, Schoen FJ, Neer EJ. Transient cardiac expression of constitutively active Galphaq leads to hypertrophy and dilated cardiomyopathy by calcineurin-dependent and independent pathways. Proc Natl Acad Sci U S A. 1998; 95: 13893eC13898.
Sah VP, Minamisawa S, Tam SP, Wu TH, Dorn GW 2nd, Ross J Jr., Chien KR, Brown JH. Cardiac-specific overexpression of RhoA results in sinus and atrioventricular nodal dysfunction and contractile failure. J Clin Invest. 1999; 103: 1627eC1634.
Kobayashi N, Horinaka S, Mita S, Nakano S, Honda T, Yoshida K, Kobayashi T, Matsuoka H. Critical role of Rho-kinase pathway for cardiac performance and remodeling in failing rat. Cardiovasc Res. 2002; 57eC67.
Kuwahara K, Saito Y, Nakagawa O, Kishimoto I, Harada M, Ogawa E, Miyamoto Y, Hamanaka I, Kajiyama N, Takahashi N, Izumi T, Kawakami R, Tamura N, Ogawa Y, Nakao K. The effects of the selective ROCK inhibitor, Y27632, on ET-1-induced hypertrophic response in neonatal rat cardiac myocytes-possible involvement of Rho/ROCK pathway in cardiac muscle cell hypertrophy. FEBS Lett. 1999; 452: 314eC318.
Sah VP, Hoshijima M, Chien KR, Brown JH. Rho is required for Galphaq and alpha1-adrenergic receptor signaling in cardiomyocytes. Dissociation of Ras and Rho pathways. J Biol Chem. 1996; 271: 31185eC31190.
Takeishi Y, Walsh RA. Cardiac hypertrophy and failure: lessons learned from genetically engineered mice. Acta Physiol Scand. 2001; 173: 103eC111.
Noguchi T, Hunlich M, Camp PC, Begin KJ, El-Zaru M, Patten R, Leavitt BJ, Ittleman FP, Alpert NR, LeWinter MM, VanBuren P. Thin filament-based modulation of contractile performance in human heart failure. Circulation. 2004; 110: 982eC987.
Knott A, Purcell I, Marston S. In vitro motility analysis of thin filaments from failing and non-failing human heart: troponin from failing human hearts induces slower filament sliding and higher Ca(2+) sensitivity. J Mol Cell Cardiol. 2002; 34: 469eC482.(Susan Vahebi, Tomoyoshi K)
Abstract
We tested the hypothesis that activation of Rho-AeCdependent kinase (ROCK-II) alters cardiac myofilament response to Ca2+ by mechanisms involving phosphorylation of thin filament proteins. We determined effects of a constitutively active form of ROCK-II on ATPase activity and tension development in detergent-extracted (skinned) fiber bundles isolated from mouse left ventricular papillary muscles. ROCK-II induced a depression in maximum ATPase rate and tension, which was associated with phosphorylation of troponin T (TnT), troponin I (TnI), and myosin-binding protein C (C-protein). This effect of ROCK-II was retained in fiber bundles isolated from transgenic (TG) mice in which phosphorylation sites (S14, S15, and S19) of myosin light chain 2 were mutated to alanine. Moreover, exchange of ROCK-IIeCphosphorylated Tn complex with the native Tn complex in the fiber bundles resulted in inhibition of maximal Ca2+ activation of tension and ATPase activity. Mass spectrometric analysis demonstrated that ROCK-II phosphorylated cardiac TnI (cTnI) at S23, S24, and T144 and cardiac TnT (cTnT) at S278 and T287. An important role for these cTnT sites is indicated by results demonstrating that ROCK-II induced a depression in tension and ATPase activity in skinned fiber bundles from a TG model in which cTnI is replaced by slow skeletal TnI, which lacks S23 and S24 and in which T144 is replaced by proline. Our data provide the first evidence that ROCK-II phosphorylation of the Tn complex, most likely at cTnT, has an important role in functional effects of signaling through the Rho-A pathway.
Key Words: Rho-kinase G-proteins heart failure troponin T
Introduction
There is evidence that functionally significant phosphorylation of sarcomeric proteins is associated with activation of G-proteineCcoupled receptors and the stimulation of signaling cascades, resulting in hypertrophy.1 For example, Gq signaling and activation of protein kinase C (PKC) not only result in phosphorylation of transcriptional factors but also in altered phosphorylations of myofilament proteins that affect contractility.2 The altered sarcomeric function may give rise to a vicious cycle in which hypertrophy is associated with depressed contractility, eventually leading to malignant growth. Gq is implicated in the activation of a number of signaling pathways, including the small GTPase Rho-A, which signals through its effector Rho-AeCdependent protein kinase (ROCK-II), the focus of this study. Rho-A activation and hypertrophy occur with the major G-proteineCcoupled agonists norepinephrine (NE), endothelin-1 (ET-1), and angiotensin II (Ang II).3,4 Moreover, Rho/ROCK activation occurs with stressors such as rapid pacing, acute pressure overload, and the chronic pressure overload of hypertension in Dahl salt-sensitive rats.5eC7 In the case of cardiac hypertrophy in Dahl salt- sensitive rats, a specific inhibitor of ROCK-II, Y-27632, has been demonstrated to improve contractile function and regress hypertrophy, independently of changes in blood pressure.8 The changes in cardiac myocyte growth associated with Rho-A activation occur with increased phosphorylation of myosin light chain 2 (MLC2), apparently independent of the PKC pathway.5,7 The mechanism appears to involve ROCK phosphorylation of myosin phosphatase targeting subunit 1, inhibition of MLC phosphatase, and thus enhanced phosphorylation of MLC2.9
In the present experiments, we tested the hypothesis that activation of ROCK-II results in functionally significant phosphorylation of thin filament regulatory proteins of cardiac sarcomeres. A rationale behind this hypothesis was the identification of a number of consensus sequences10,11 (RXXS/T or RXS/T) for ROCK-II in cardiac troponin T (cTnT) and cardiac troponin I (cTnI). A consensus sequence in cTnI is located in the inhibitory peptide that is important in tethering cTnI to actin during diastole. We have shown that phosphorylation of this site is a significant determinant of cross-bridge cycling rate and myofilament response to Ca2+.12 We also identified a consensus sequence in the C-terminal region of cTnT. We reported previously that phosphorylation in this region has significant effects on tension generation and the ATPase rate of cardiac myofilaments.13
Thus, our aim in the present study was to determine whether ROCK-II phosphorylates cTnI or cTnT, and, if so, to determine the functional effects of these phosphorylations on Ca2+ dependence of myofilament tension and ATPase rate. Our data provide the first evidence that ROCK-II depresses maximum myofilament tension and ATPase activity as well as Ca2+ sensitivity by a mechanism that appears to be dependent on phosphorylation of cTnT.
Materials and Methods
Animal Model and Fiber Bundle Preparation
We studied preparations from 3-month-old nontransgenic mouse hearts (Charles River) and from hearts of transgenic (TG) mice (generously provided by Dr Jeff Robbins, University of Cincinnati) in which ventricular regulatory myosin light chain 2 (RLC2v) was replaced with a nonphosphorylatable mutant.14 We also used transgenic (TG) mouse hearts in which cTnI was replaced with slow skeletal TnI (ssTnI),15 the embryonic/neonatal isoform. Hearts were excised under deep anesthesia (pentobarbital sodium; 100 mg/kg IP). Fiber bundles (1 to 3 mm in length and 150 to 250 e in diameter) were dissected from left ventricular papillary muscles and treated with 1% Triton X-100 for 2 hours as described previously.16
Myofilament Mechano-Energetic Measurements
We measured isometric tension and actomyosin Mg-ATPase activity simultaneously as described previously.13,16 Sarcomere length was adjusted to 2.2 e. Fiber bundles were equilibrated for 5 minutes in relaxing buffer followed by 3 minutes in preactivating buffer and then in maximal activating solution. The composition of all solutions was as described previously.13 Only fiber bundles retaining 95% of their initial maximum tension were used for analysis. The isometric tension and ATPase activity were determined simultaneously at 20°C in the presence of variable Ca2+ concentrations as described.13 After a series of measurements over a range of values from eClog [Ca2+](pCa) 8.0 to pCa 4.5, we added 1 unit of constitutively active ROCK-II (Upstate) to the fiber bundle in relaxing solution and incubated for 45 minutes at 20°C in the presence of calyculin A (100 ng/mL). We then measured isometric tension and ATPase activity again over the same range of pCa values.
Control experiments showed that this incubation protocol in the absence of ROCK-II had no significant effect on any of the mechano-energetic parameters. Alternatively, we exchanged the native Tn complex in the fiber bundles with recombinant Tn complex13,17,18 with or without previous in vitro phosphorylation by ROCK-II, following a previously described exchange protocol.13 The extent of recombinant Tn exchange into the fibers was determined by SDS-PAGE immunoblot analysis.13
Phosphoprotein Determination
Proteins were separated by 1D-PAGE on 15% polyacrylamide gels as described previously.19 Phosphorylated proteins were detected by PRO Q Diamond stain following the protocol of the supplier (Molecular Probes). Briefly, the gels were fixed in 10% trichloroacetic acid/50% methanol and stained with Pro Q Diamond (1.5 hours). The gel was destained overnight and scanned using an FX Pro Plus imager (Bio-Rad) in conjunction with a CY3 filter set. Subsequently, the gel was stained with 80 ml of SYPRO Ruby for 3 hours (Molecular Probes) to visualize total protein. The gel was destained with 10% methanol, 10% acetic acid for at least 30 minutes, and scanned on the FX Pro Plus imager (Bio-Rad) using a SYPRO filter set.
Matrix-Assisted Laser Desorption/IonizationeCTime of Flight Analysis
ROCK-treated and untreated Tn complexes were digested with either lysyl endopeptidase from Acromobactor (Wako) or Staphylococcus aureus V8 protease (Roche), and phosphopeptides were enriched with an immobilized metal affinity column (IMAC) as described.20 Matrix-assisted laser desorption/ionizationeCtime of flight (MALDI-TOF) analysis was performed using 2',4',6'-trihydroxyacetophenone as a matrix. The presence of phosphates group in a peptide was confirmed by postsource decay mass spectrometry analysis.
Data Analysis and Statistical Evaluation
Individual tensioneCCa2+ and ATPaseeCCa2+ relationships were fit to a modified Hill equation (Graphpad Prism Software) as described previously.13 Tension cost was derived by linear regression of the tensioneCATPase relationship. Statistical evaluation was by 1-way ANOVA and unpaired t test as appropriate; P<0.05 was considered significant.
Results
ROCK-Dependent Phosphorylation Depresses Cardiac Myofilament Response to Ca2+
In the first series of experiments, we determined the effect of ROCK-IIeCdependent phosphorylation on Ca2+ activation of tension and ATPase activity of detergent-extracted fiber bundles of wild-type (WT) mouse. Figure 1A illustrates results from measurements of tension, and Figure 1B illustrates the simultaneously determined ATPase activity. Figure 1C shows a plot of ATPase activity at each level of tension generation and thus gives a measure of the tension cost. The Table summarizes all the data obtained before and after treatment with ROCK-IIeCdependent phosphorylation. Data plotted in Figure 1A illustrate that treatment of cardiac fiber bundles with ROCK-II depressed maximum developed tension by 15% and also resulted in a decrease in Ca2+ sensitivity (pCa50=0.2). Data plotted in Figure 1B indicate that ROCK-dependent phosphorylation also resulted in a decrease in Ca2+ sensitivity (pCa50=0.2) and depression of maximal actomyosin Mg-ATPase activity by 15% (Table). Figure 1C indicates there was a proportional depression in tension and ATPase activity, resulting in no change in tension cost.
A likely mechanism by which the signaling process engaging ROCK-II might affect myofilament activity is by altered levels of myofilament protein phosphorylation. To determine the substrate specificity for ROCK-II in detergent-extracted fiber bundles, we visualized phosphorylated proteins using Pro Q Diamond phospho-specific stain (Figure 2A and 2B). Figure 2A shows protein staining after SDS-PAGE analysis of skinned fiber bundles before and after treatment with ROCK-II, calyculin A, and protein phosphatase 2A (PP2A). Figure 2B shows results of Pro-Q phospho-specific staining of the same gel. Proteins resolved in lane 1 were from control skinned fibers with no treatment. Proteins in lane 2 are from skinned fibers treated with ROCK-II and demonstrate ROCK-IIeCdependent phosphorylation of C-protein, cTnI, and cTnT. Proteins in lane 3 are from skinned fibers incubated with ROCK-II and phosphatase inhibitor calyculin A (100 ng/mL) and in lane 4 from skinned fibers incubated with PP2A (0.01 U) for 45 minutes at 30°C with the reaction stopped by addition of okadaic acid and calyculin A (100 ng/mL). Proteins in lane 5 are from skinned fibers treated as in lane 4 but followed by addition of ROCK-II, indicating phosphorylation of C-protein, cTnT, and cTnI.
To identify the specific sites of ROCK-IIeCdependent phosphorylation, we analyzed complexes of recombinant troponin by mass spectrometric analysis (Figure 3). Our data show that ROCK-II phosphorylation sites of cTnT (TnT4 isoform) are located at the C terminus at S278 and T287, and phosphorylation sites of cTnI are at S23, S24, and T144 (Figure 3A and 3B). These sites of phosphorylation are not present in the control untreated recombinant proteins used in the reconstitution of the troponin complexes.
ROCK-Dependent Inhibition of Myofilament Tension Is Independent of MLC2 Protein Phosphorylation
In view of evidence that activated ROCK-II leads to phosphorylation of MLC2,5 in a second series of experiments, we tested fiber bundles isolated from TG-RLC2v(PeC) mice in which Ser14, Ser15, and Ser19 of MLC2 were mutated to Ala.5 Figure 4A illustrates results from measurements of tension; Figure 4B, the simultaneously determined Mg-ATPase activity; and Figure 4C, a plot of tension as a function of ATPase activity (tension cost). The inset in Figure 4A shows the Pro Q Diamond phospho-specific stain from TG-RLC2v(PeC) mice skinned fibers treated with ROCK-II (lane 1) and nontreated (lane 2) in vitro and indicates an increases in phosphorylation of C-protein, cTnT, and cTnI but no phosphorylation of MLC2. The Table summarizes all the data obtained before and after treatment with ROCK-IIeCdependent phosphorylation. ROCK-II phosphorylation resulted in a significant decrease in the maximum tension (17%) in TG preparations (Figure 4A). As shown in Figure 4A, the Ca2+ sensitivity was altered by ROCK-II treatment (pCa50=0.2). Similarly, ROCK-IIeCdependent phosphorylation resulted in a significant depression of (15%) maximum Mg-ATPase activity and altered Ca2+ sensitivity (Figure 4B). There were no significant differences in the tension cost in fiber bundles from TG-RLC2v(PeC) mouse hearts after treatment with ROCK-II (Figure 4C; Table).
Effects of ROCK-Dependent Phosphorylation of Tn Complex on Myofilament Function
The results reported in Figure 4 indicate that MLC phosphorylation by ROCK-II is not responsible for decreased tension. Thus, we examined whether phosphorylation of proteins in the cTn complex by ROCK-II accounts for its functional effect on cardiac myofilaments. We compared the mechano-energetics of fiber bundles in which native cTn was exchanged with recombinant cTn-WT with fiber bundles in which the native cTn was exchanged with recombinant cTn-WT that was phosphorylated by ROCK-II (p-cTn) in vitro (Figure 5A through 5C). As documented by Sumandea et al,11 the exchange protocol outlined in Materials and Methods results in essentially 100% replacement of the Tn. The fiber bundles subjected to this protocol retained 80% to 90% of the tension generated by untreated controls. Figure 5A through 5C illustrates data obtained from exchange experiments using either WT-Tn complex or p-cTn. The inset in Figure 5A (lane 1) shows phospho-specific staining of skinned fiber bundle exchange with WT-Tn and fiber bundles exchanged with p-cTn (lane 2). Quantification of the staining indicated an increase in cTnI and cTnT phosphorylation but no change in C-protein phosphorylation. The Table summarizes all the data from exchange experiments. The fiber bundles in which native Tn was exchanged with p-cTn complex demonstrated a significantly depressed (29%) maximum developed tension when compared with tension developed by skinned fiber bundles exchanged with control WT-Tn. Similarly, exchange of p-cTn complex resulted in a decrease in Ca2+ sensitivity (pCa50=0.1). ROCK-IIeCdependent phosphorylation also significantly decreased (25%) maximal actomyosin Mg-ATPase activity and altered Ca2+ sensitivity (pCa50=0.1; Figure 5B) without appreciably affecting tension cost (Figure 5C).
ROCK Induces Desensitization to Ca2+ in Myofilaments Containing ssTnI in Place of cTnI
Our mass spectrometry data indicated that the sites for ROCK-II phosphorylation of cTnI are at S23, S24, and T144 sites. Our approach toward understanding the functional significance of phosphorylation of these sites was to use a TG-ssTnI model in which cTnI is replaced by ssTnI.15 ssTnI lacks protein kinase A sites (S23 and S24) and the PKC site at T144, which is replaced by proline. Figure 6A through 6C depicts effects of ROCK-II phosphorylation on mechano-energetics of fiber bundles isolated from TG-ssTnI mouse hearts. Figure 6A and the Table indicate that treatment with ROCK-II induces desensitization to Ca2+ and depression in maximal tension. The inset in Figure 6A shows the Pro Q Diamond phospho-specific stain from untreated TG-ssTnI skinned fibers (lane 1) and fibers treated with ROCK-II in (lane 2). These results demonstrate an increase in phosphorylation of cTnT, but not ssTnI, and C-protein. ROCK-II also decreased Ca2+ sensitivity and depressed maximal ATPase activity (Figure 6B; Table), with no significant changes in tension cost (Figure 6C; Table).
Discussion
Our data are the first to provide evidence that cardiac troponin is a potential downstream effector of Rho signaling in heart. Strong evidence for this conclusion comes from results demonstrating a depression in tension and ATPase in skinned fiber bundles lacking MLC2 phosphorylation sites and in bundles in which native Tn was replaced with Tn phosphorylated previously in vitro by ROCK-II. Although C-protein and troponin may be phosphorylated by ROCK-II, our data demonstrated that the effects of ROCK-II on tension and ATPase rate could occur with no change in phosphorylation of C-protein, as was the case in the preparations in which we exchanged native troponin with WT and p-cTn. Previous studies have emphasized either direct phosphorylation of MLC2 or indirect phosphorylation of MLC2 by inhibition of myosin phosphatase.9,21 In general, these studies have not examined potential modifications in thin filament proteins. Moreover, phosphorylation of MLC2 would be expected to increase myofilament Ca2+ sensitivity with little effect on maximum tension.22,23 In our studies, ROCK-II phosphorylation of MLC2 would offset the depression in Ca2+ sensitivity by MLC2 phosphorylation, with little effect on the decrease in maximum tension induced by phosphorylation of cTnI and cTnT.
The mechanism by which ROCK-IIeCdependent phosphorylation of skinned fiber bundles inhibited tension and ATPase activity appears to involve functionally significant and specific sites on cardiac troponin. In the case of cTnI, the sites, as identified with mass spectrometry, involved S23 and S24 in the unique N-terminal extension and in T144. However, our studies (Figure 6) with fiber bundles missing these phosphorylation sites indicate that the effects of ROCK-II phosphorylation on maximum tension are not likely to be attributable to cTnI phosphorylation. In the case of cTnT, we identified by mass spectrometry a C-terminal peptide containing S278 and T287 as the sites of phosphorylation by ROCK-II. This region of cTnT interacts with cTnC, cTnI, and tropomyosin (Tm) and forms a Ca2+-sensitive domain.24 With binding of Ca2+ to cTnC, cTnI moves a substantial distance on the thin filament, shifting its binding from actin to cTnC.25 These changes are sensed by cTnT in a signaling cascade that is ultimately responsible for movement of Tm and actin subdomains. There is evidence from studies with reconstituted thin filament preparations of fast skeletal muscle that the interaction of TnT with TnC not only acts in concert with TnI to release the thin filament from inhibition but also acts to activate the actineCmyosin interaction.26 Previous studies13,26 indicate that phosphorylation in the C-terminal region of cTnT would be expected to depress maximum tension and ATPase rate, as we have reported here for ROCK-IIeCdependent phosphorylation. Early studies using reconstituted thin filament preparations demonstrated that PKC-dependent phosphorylation of C-terminal residues in cTnT induced a depression in maximum actomyosin ATPase activity.27 More recently, in studies using direct phosphorylation and charge modifications of C-terminal residues of cTnT,13 we reported that these modifications have significant and specific effects on maximum tension and Ca2+ sensitivity of skinned fiber bundles. Results of work presented here support these earlier findings and extend them to include a potential role of regulation of maximum tension via Rho/ROCK signaling.
An important question is: what is the physiological or pathophysiological significance of ROCK-IIeCdependent phosphorylation of troponin This question is based on evidence that the Rho/ROCK pathway is involved in regulation of cardiac function. Pathways upstream of Rho/ROCK are well recognized to be significant in the long-term regulation of cardiac function in the hypertrophic/failure process. NE, ET-1, and Ang II all induce cell growth and bind to receptors that engage a signal transduction path involving Gq.28eC30 For example, treating cardiac myocytes with Ang II induced cardiac hypertrophy associated with activation of Rho-A.4 A specific role of Gq in promoting cardiac hypertrophy and apoptosis associated with decompensated contractile failure has been demonstrated in a TG mouse model with Gq overexpression in the cardiac compartment.31,32 Additionally, TG overexpression of Rho-A in hearts resulted in a cardiomyopathy and ventricular failure associated with atrial enlargement and abnormalities of the sino-atrial and atrio-ventricular nodes.33 Stresses that induce hypertrophy and failure are also associated with Rho/ROCK activation and alterations in myofilament response to Ca2+. One study5 reported increased Rho-A expression in dog hearts stressed by a rapid pacing protocol that induced heart failure. In this model, there was an increase of MLC2 phosphorylation in the failing heart.5 Evidence for an enhanced signaling through the Rho/ROCK pathway was also presented by stimulation of permeabilized single myocytes with phenylephrine, which induced increased myofilament Ca2+ sensitivity in cells from failing hearts but not from the controls. Maximum tension was not measured in that study. A role for Rho-A/ROCK in hypertrophy is also strongly indicated by reports that the specific ROCK-II inhibitorY-27632 is able to ameliorate functional changes in hypertrophic and failing heart preparations induced by hypertension in the Dahl salt-sensitive rat7 by Ang II administration8,34 and by treatment of myocytes with ET-1.35 A role for Rho-A is also indicated from results using adenoviral gene transfer approach and showing that overexpression of dominant-negative mutants of Rho-A inhibited stress-induced hypertrophic response.36 It is also important with regard to the results presented here that after acute pressure overload in hearts, there is an early activation of ROCK-II. Torsoni et al6 reported that acute constriction of the aorta in rats induced a rapid binding of Rho-A with ROCK-II that transiently rose 3.5-fold above the levels in controls within 30 minutes and then fell to a steady-state level of 2-fold. It is also of significance to the data reported here that the pressure overload induced an association of ROCK-II with the Z-disk as well as T-tubules and subsarcolemal sites.6
Although a role for TnT phosphorylation in hearts with activation of the Rho-A signaling cascade/pathway has yet to be demonstrated, on the basis of data presented here, it will be important to re-examine some of the models described above for alterations in Tn phosphorylation. We think existing data do not rule out this possibility because neither determination of Tn phosphorylation nor maximum sarcomeric tension has generally been studied in models in which ROCK-II activation is specifically inhibited, for example by chronic administration of Y-27632.7 On the other hand, there are data indicating that increased thin filament phosphorylation is associated with heart failure in animal models37 and in humans.38,39 At the present time, the specific kinases and phosphatases that give rise to these changes in thin filament phosphorylation are not clear. Our data indicate that a complete investigation of this requires determination of the possible involvement of ROCK-II not only in the end stages but also in the progression to heart failure.
Acknowledgments
This research was supported in part by National Institutes of Health (NIH) grants 5PO1 HL62426-05 Proj1 (R.J.S), 5 RO1 HL64035-05 (R.J.S), and 5 PO1 HL62426-05 Proj4 (P.P.dT.), American Heart Association (AHA) scientist development grant 0230038N (T.K), AHA-Midwest Affiliate predoctoral fellowship award 0315209Z (S.V), and NIH training grants 5 T32 HL07692-14 to 5 T32 HL07692-15 (R.J.S). These studies were submitted in partial fulfillment for the degree of doctor of philosophy. We thank Drs A.F. Martin and B.M. Wolska for their support.
References
Solaro RJ, Montgomery DM, Wang L, Burkart EM, Ke Y, Vahebi S, Buttrick P. Integration of pathways that signal cardiac growth with modulation of myofilament activity. J Nucl Cardiol. 2002; 5: 523eC533.
Solaro RJ. Modulation of cardiac myofilament activity by protein phosphorylation. In: Page E, Fozzard HE, Solaro RJ, eds. Hand Book of Physiology. New York, NY: Oxford University Press; 2001: 264eC300.
Eto M, Barandier C, Rathgeb L, Kozai T, Joch H, Yang Z, Luscher TF. Thrombi suppresses endothelial nitric oxide synthase and upregulates endothelin-converting enzyme-1 expression by distinct pathways: role of Rho/ROCK and mitogen-activated protein kinase. Circ Res. 2001; 89: 583eC590.
Aoki H, Izumo S, Sadoshima J. Angiotensin II activates RhoA in cardiac myocytes: a critical role of RhoA in angiotensin II-induced premyofibril formation. Circ Res. 1998; 82: 666eC676.
Suematsu N, Satoh S, Kinugawa S, Tsutsui H, Hayashidani S, Nakamura R, Egashira K, Makino N, Takeshita A. Alpha1-adrenoceptor-Gq-RhoA signaling is upregulated to increase myofibrillar Ca2+ sensitivity in failing hearts. Am J Physiol Heart Circ Physiol. 2001; 281: H637eCH646.
Torsoni AS, Fonseca PM, Crosara-Alberto DP, Franchini KG. Early activation of p160ROCK by pressure overload in rat heart. Am J Physiol Cell Physiol. 2003; 284: C1411eCC1419.
Satoh S, Ueda Y, Koyanagi M, Kadokami T, Sugano M, Yoshikawa Y, Makino N. J Chronic inhibition of Rho kinase blunts the process of left ventricular hypertrophy leading to cardiac contractile dysfunction in hypertension-induced heart failure. J Mol Cell Cardiol. 2003; 35: 59eC70.
Kobayashi N, Horinaka S, Mita S, Nakano S, Honda T, Yoshida K, Kobayashi T, Matsuoka H. Involvement of Rho-kinase pathway for angiotensin II-induced plasminogen activator inhibitor-1 gene expression and cardiovascular remodeling in hypertensive rats. J Pharmacol Exp Ther. 2002; 30: 459eC466.
Feng J, Ito M, Kureishi Y, Ichikawa K, Amano M, Isaka N, Okawa K, Iwamatsu A, Kaibuchi K, Hartshorne DJ, Nakano T. Rho-associated Kinase of Chicken Gizzard Smooth Muscle. J Biol Chem. 1999; 274: 3744eC3752.
Fukata Y, Oshiro N, Kinoshita N, Kawano Y, Matsuoka Y, Bennett V, Matsuura Y, Kaibuchi K. Phosphorylation of adducin by Rho-kinase plays a crucial role in cell motility. J Cell Biol. 1999; 145: 347eC361.
Matsui T, Maeda M, Doi Y, Yonemura S, Amano M, Kaibuchi K, Tsukita S, Tsukita S. Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J Cell Biol. 1998; 140: 647eC657.
Burkart EM, Sumandea MP, Kobayashi T, Nili M, Martin AF, Homsher E, Solaro RJ. Phosphorylation or glutamic acid substitution at protein kinase C sites on cardiac troponin I differentially depress myofilament tension and shortening velocity. J Biol Chem. 2003; 278: 11265eC11272.
Sumandea MP, Pyle WG, Kobayashi T, de Tombe PP, Solaro RJ. Identification of a functionally critical protein kinase C phosphorylation residue of cardiac troponin T. J Biol Chem. 2003; 278: 35135eC35144.
Sanbe A, Fewell JG, Gulick J, Osinska H, Lorenz J, Hall DG, Murray LA, Kimball TR, Witt SA, Robbins J. Abnormal cardiac structure and function in mice expressing nonphosphorylatable cardiac regulatory myosin light chain 2. J Biol Chem. 1999; 274: 21085eC21094.
Fentzke RC, Buck SH, Patel JR, Lin H, Wolska BM, Stojanovic MO, Martin AF, Solaro RJ, Moss RL, Leiden JM. Impaired cardiomyocyte relaxation and diastolic function in transgenic mice expressing slow skeletal troponin I in the heart. J Physiol. 1999; 517: 143eC157.
de Tombe PP, Steinen GJ. Protein kinase A does not alter economy of tension maintenance in skinned rat cardiac trabeculae. Circ Res. 1995; 76: 734eC741.
Kobayashi T, Dong WJ, Burkart EM, Cheung HC, Solaro RJ. Effects of protein kinase C dependent phosphorylation and a familial hypertrophic cardiomyopathy-related mutation of cardiac troponin I on structural transition of troponin C and myofilament activation. Biochemistry. 2004; 4.
Chandra M, Kim JJ, Solaro RJ. An improved method for exchanging troponin subunits in detergent skinned rat cardiac fiber bundles. Biochem Biophys Res Commun. 1999; 263: 219eC223.
Fritz J, Swartz DR, Greaser ML. Factors affecting polyacrylamide gel electrophoresis and electroblotting of high-molecular weight myofibrillar proteins. Anal Biochem. 1998; 180: 205eC210.
Yang X, Wu H, Kobayashi T, Solaro RJ, van Breemen RB. Enhanced ionization of phosphorylated peptides during MALDI TOF mass spectrometry. Anal Chem. 2004; 76: 1532eC1536.
Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M, Yamamori B, Feng J, Nakano T, Okawa K, Iwamatsu A, Kaibuchi K. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science. 1996; 273: 245eC248.
Noland TA Jr, Kuo JF. Phosphorylation of cardiac myosin light chain 2 by protein kinase C and myosin light chain kinase increases Ca(2+)-stimulated actomyosin MgATPase activity. Biochem Biophys Res Commun. 1993; 193: 254eC260.
Morano I, Hofmann F, Zimmer M, Ruegg JC. The influence of P-light chain phosphorylation by myosin light chain kinase on the calcium sensitivity of chemically skinned heart fibres. FEBS Lett. 1985; 189: 221eC224.
Fisher D, Wang G, Tobacman LS. NH2-terminal truncation of skeletal muscle troponin T does not alter the Ca2+ sensitivity of thin filament assembly. J Biol Chem. 1995; 270: 25455eC25460.
Tao T, Gong BJ, Leavis PC. Calcium-induced movement of troponin-I relative to actin in skeletal muscle thin filaments. Science. 1990; 247: 1339eC1341.
Potter JD, Sheng Z, Pan BS, Zhao J. A direct regulatory role for troponin T and a dual role for troponin C in the Ca2+ regulation of muscle contraction. J Biol Chem. 1995; 270: 2557eC2562.
Noland TA Jr, Kuo JF. Protein kinase C phosphorylation of cardiac troponin I or troponin T inhibits Ca2(+)-stimulated actomyosin MgATPase activity. J Biol Chem. 1991; 266: 4974eC4978.
Knowlton KU, Michel MC, Itani M, Shubeita HE, Ishihara K, Brown JH, Chien KR. The alpha 1A-adrenergic receptor subtype mediates biochemical, molecular, and morphologic features of cultured myocardial cell hypertrophy. J Biol Chem. 1993; 268: 15374eC15380.
LaMorte VJ, Thorburn J, Absher D, Spiegel A, Brown JH, Chien KR, Feramisco JR, Knowlton KU. Gq- and ras-dependent pathways mediate hypertrophy of neonatal rat ventricular myocytes following alpha 1-adrenergic stimulation. J Biol Chem. 1994; 269: 13490eC13496.
Sadoshima J, Izumo S. Signal transduction pathways of angiotensin IIeCinduced c-fos gene expression in cardiac myocytes in vitro. Roles of phospholipid-derived second messengers. Circ Res. 1993; 73: 424eC438.
Adams JW, Sakata Y, Davis MG, Sah VP, Wang Y, Liggett SB, Chien KR, Brown JH, Dorn GW II. Enhanced Galphaq signaling: a common pathway mediates cardiac hypertrophy and apoptotic heart failure. Proc Natl Acad Sci U S A. 1998; 95: 10140eC10145.
Mende U, Kagen A, Cohen A, Aramburu J, Schoen FJ, Neer EJ. Transient cardiac expression of constitutively active Galphaq leads to hypertrophy and dilated cardiomyopathy by calcineurin-dependent and independent pathways. Proc Natl Acad Sci U S A. 1998; 95: 13893eC13898.
Sah VP, Minamisawa S, Tam SP, Wu TH, Dorn GW 2nd, Ross J Jr., Chien KR, Brown JH. Cardiac-specific overexpression of RhoA results in sinus and atrioventricular nodal dysfunction and contractile failure. J Clin Invest. 1999; 103: 1627eC1634.
Kobayashi N, Horinaka S, Mita S, Nakano S, Honda T, Yoshida K, Kobayashi T, Matsuoka H. Critical role of Rho-kinase pathway for cardiac performance and remodeling in failing rat. Cardiovasc Res. 2002; 57eC67.
Kuwahara K, Saito Y, Nakagawa O, Kishimoto I, Harada M, Ogawa E, Miyamoto Y, Hamanaka I, Kajiyama N, Takahashi N, Izumi T, Kawakami R, Tamura N, Ogawa Y, Nakao K. The effects of the selective ROCK inhibitor, Y27632, on ET-1-induced hypertrophic response in neonatal rat cardiac myocytes-possible involvement of Rho/ROCK pathway in cardiac muscle cell hypertrophy. FEBS Lett. 1999; 452: 314eC318.
Sah VP, Hoshijima M, Chien KR, Brown JH. Rho is required for Galphaq and alpha1-adrenergic receptor signaling in cardiomyocytes. Dissociation of Ras and Rho pathways. J Biol Chem. 1996; 271: 31185eC31190.
Takeishi Y, Walsh RA. Cardiac hypertrophy and failure: lessons learned from genetically engineered mice. Acta Physiol Scand. 2001; 173: 103eC111.
Noguchi T, Hunlich M, Camp PC, Begin KJ, El-Zaru M, Patten R, Leavitt BJ, Ittleman FP, Alpert NR, LeWinter MM, VanBuren P. Thin filament-based modulation of contractile performance in human heart failure. Circulation. 2004; 110: 982eC987.
Knott A, Purcell I, Marston S. In vitro motility analysis of thin filaments from failing and non-failing human heart: troponin from failing human hearts induces slower filament sliding and higher Ca(2+) sensitivity. J Mol Cell Cardiol. 2002; 34: 469eC482.(Susan Vahebi, Tomoyoshi K)