Targeting to C-Terminal Myosin Heavy Chain May Explain Mechanotransduction Involving Focal Adhesion Kinase in Cardiac Myocytes
http://www.100md.com
《循环研究杂志》
From the Department of Internal Medicine (P.M.F., R.Y.I., C.B.K., D.P.C.-A., K.G.F.), School of Medicine, State University of Campinas, Campinas; and Center of Structural Molecular Biology (J.K.), National Synchrotron Light Laboratory, Campinas, S?o Paulo, Brazil.
Focal adhesion kinase (Fak) has been implicated as a signaling molecule involved in the early response of cardiac myocytes to mechanical stress. The mechanism of Fak activation by mechanical stimuli is not clear. In this study, we report the load-induced Fak activation and its association with myosin heavy chain in cardiac myocytes. Pressure overload lasting from 3 to 60 minutes was shown to induce Fak phosphorylation at Tyr-397, -576/7, -861, and -925 as detected by phosphospecific antibodies. This was paralleled by increases of Fak/Src association and Src activity (Tyr-418 phosphorylation). Yeast two-hybrid screening of an adult rat cDNA library revealed an interaction between Fak and C-terminal coiled-coil region of -myosin heavy chain. This was confirmed by pulldown assay with GST-C-terminal myosin fragment and native Fak from rat left ventricle. Such interaction was confirmed by coimmunoprecipitation assay with anti-Fak and anti-heavy chain cardiac myosin antibodies, confocal microscopy of double-labeled isolated cardiac myocytes and immunoelectron microscopy with anti-Fak antibody. Fak activation by mechanical stress was accompanied by a reduction of Fak/myosin heavy chain association and its relocation at subcellular sites such as costameres, Z-discs, and nuclei. Thus, our present data identify Fak interaction with C-terminal region of myosin heavy chain adding comprehensive data on Fak activation by mechanical stress and mechanotransduction in cardiac myocytes.
Key Words: focal adhesion kinase mechanotransduction cell signaling hypertrophy myosin
Mechanical stress is a major factor involved in the development of myocardial adaptive and maladaptive changes in heart diseases. Local mechanical forces activate signaling mechanisms in cardiac myocytes inducing the expression of specific genetic programs linked to myocardial structural and functional remodeling.1,2 Although mechanical forces might directly trigger signaling mechanisms in cardiac myocytes, the mechanism by which they are sensed and converted to biochemical signals remains elusive.
Structures such as sarcomeric lattice, cytoskeleton, and the extracellular matrix operate in the transmission of either passive or active forces in cardiac myocytes.3,4 Studies have confirmed the critical importance of the molecular integrity of Z-disc and cytoskeleton to the expression of genetic program induced by mechanical stress in cardiac myocytes. Z-disc structure is organized by N-terminal titin Z repeats linked to -actinin and associated proteins such as MLP, ALP, telethonin (T-cap), cypher/Zasp, and myotilin.4,5 Notably, MLP null mice were shown to fail to upregulate brain and atrial natriuretic factors mRNA in response to stretch.6 Human titin mutations as well as deletion of the -actinin binding proteins such as ALP or MLP in the mouse causes dilated cardiomyopathy.6–10 However, the mechanisms by which these structures and proteins detect physical forces and initiate biochemical signals are yet unclear.
The link of Z-discs to sarcolemma at sites known as costameres has been implicated as a potential signaling station to mechanotransduction in cardiac myocytes. Costameres are also sites of integrin clusters and share a similarity to focal adhesions.11,12 Focal adhesion kinase (Fak), a primary integrin effector at focal adhesion sites,15–17 is rapidly activated by mechanical stimuli in cultured neonatal rat ventricular myocytes18–20 and in overloaded myocardium of adult animals.21–24 The importance of Fak to the regulation of early gene transcription in response to stretch was demonstrated in neonatal rat cardiac myocytes,20 indicating that this kinase may coordinate signaling pathways involved in the hypertrophic growth induced by mechanical stress. However, the mechanisms responsible for Fak activation by mechanical stress in cardiac myocytes are still unclear. Although Fak is thought to colocalize with integrins at costameres,3,12–14 there is no precise description of Fak localization in cardiac myocytes. We have previously shown20 that stretch induces Fak to cluster at myofilaments in neonatal rat ventricular myocytes, but no evidence linked Fak directly to costameres in such cells. Otherwise, experiments performed with dominant-negative Fak and specific pharmacological Src inhibitor confirmed that Fak activation is strictly dependent on phosphorylation of Tyr397 and cooperation with Src as has been shown in the focal adhesion site. However, the lack of accurate data on localization as well as on the signaling and structural protein partners of Fak in cardiac myocytes preclude a better understanding of Fak activation by mechanical stress in this particular cell type.
Thus, we searched for novel binding partners for Fak by yeast two-hybrid screening of an adult rat left ventricle cDNA library combined with an analysis of Fak localization with immunohistochemistry and immunoelectron microscopy of control and overloaded rat myocardium.
Detailed methods are described in the expanded Materials and Methods in the online data supplement available at http://circres.ahajournals.org.
Animal Model
Male Wistar rats (160 to 200 g) underwent acute pressure overload induced by controlled constriction of transversal aorta as previously described.21 The animals were obtained from the Central Animal House of the University of Campinas and received care in compliance with the principles of laboratory animal care formulated by the Animal Care and Use Committee of the University of Campinas.
Protein Analysis by Immunoblotting
Aliquots of left ventricle extracts (60 μg) containing equal amount of total protein were resolved on SDS-PAGE and transferred to nitrocellulose membranes, which were incubated with specific antibodies and [125I] Protein A. Band intensities were quantified through optical densitometry of the developed autoradiographs.
Subcellular Fractioning
Subcellular fraction enriched in nuclear proteins was obtained from the pellet of left ventricle extracts centrifuged at 1000g. The remaining supernatant fraction that was enriched in soluble, membrane, and myofilament proteins was designated SMM fraction.
Laser Confocal Analysis
Isolated cardiac myocytes were fixed with 4% paraformaldehyde/sucrose and incubated with anti-Fak and anti-heavy chain cardiac myosin primary antibodies. This was followed by incubation with secondary goat anti-rabbit Alexa488-conjugated and rabbit anti-mouse Alexa568-conjugated antibody. Images were obtained with laser confocal microscope (Zeiss LSM510).
Immunoelectron Microscopy
Fragments of left ventricle were fixed in paraformaldehyde/glutaraldehyde (Electron Microscopy Sciences) and embedded in LR White resin at –20°C under UV light. Thin-sections were stained with anti-Fak antibody overnight at 4°C. The sections were then incubated with anti-rabbit IgG-10 nm gold-conjugated antibody diluted 1:20 in 1% BSA (pH 8.2). The grids were stained with 5% uranyl acetate and 0.5% lead citrate, examined, and photographed in a transmission electron microscope (LEO 906). Negative controls were accomplished by incubation with the primary antibody omitted.
cDNA Library
A cDNA library was constructed from of an adult rat LV total RNA. The purified library cDNA was fused to the GAL4 activation domain of pGADT7-rec expressing vector by recombination in L40 yeast strain.
Yeast Two-Hybrid Screening
The pBTM (ADE2)Fak/NX bait construction (provided by Steven K. Hanks, Vanderbilt University School of Medicine, Nashville, Tenn), expressed a fragment spanning 748 amino acid N terminus of Fak protein fused to LEX A DNA binding domain of the pBTM116 vector. L40 yeast cells with left ventricle cDNA library were cotransformed with FAK/NX bait. Positive clones were tested by galactosidase assay, and cDNA cloned in pGADT7-rec was sequenced. Sequences were translated into amino acid sequence by the ORF Finder and submitted to BLAST and Clustal W in the NCBI Database (http://www.ncbi.nlm.nih.gov/).
Protein Expression
The encoding C-terminal region of -myosin heavy chain retrieved after yeast two-hybrid screening was inserted into the vector pGEX5x2 (Amersham Pharmacia Biotech) for expression of recombinant GST-tagged protein in Escherichia coli BL21 (RIL) Codon Plus.
Pulldown Assay
GST-tagged myosin fragment (48.2 kDa) was used for the pulldown assay with LV extracts. GST conjugated glutathione beads were used as negative control for nonspecific binding. The pulldown pellets were resolved on SDS-PAGE and the membranes were stained with anti-Fak antibody.
Statistical Analysis
Data are presented as mean±SEM. Differences between the mean values of the densitometric readings were tested by ANOVA and Bonferroni multiple-range test. A value of P<0.05 indicated statistical significance.
Pressure Overload Induces Fak Phosphorylation at Tyr397, 576/7, 861, and 925
The autophosphorylation of Fak Tyr397 has been shown to be critical for Fak activation. Phosphorylated Tyr397 recruits Src family kinases, which lead to phosphorylation of additional Fak tyrosine residues, which promote the assembly of distinct higher-order individual signaling complexes,25–27 providing a mechanism for coordinating signaling through multiple pathways. We previously demonstrated21,23 a rapid increase in the phosphorylation and activity of Fak in the myocardium of rats in response to mechanical stimuli. In this study, we examined the effects of pressure overload on phosphorylation of myocardial Fak Tyr397, 407, 576/577, 861, and 925 with phosphospecific antibodies. Experiments were performed in rats subjected to constriction of transverse aorta lasting from 3 to 60 minutes. Blood pressure above and below the level of aortic constriction was monitored to ascertain the stability of the experimental preparation. Aortic constriction produced a rapid and sustained increase of blood pressure measured in the ascending aorta with no change in blood pressure in the abdominal aorta, as compared with blood pressure of control period (Figure 1A).
Figure 1. A, Representative blood pressure recording in ascending and abdominal aorta. Aortic constriction was adjusted to increase the ascending aorta systolic pressure by 40 mm Hg, without reducing the mean blood pressure levels in the abdominal aorta. B, Representative blots showing Fak phosphorylation at tyrosines 397, 576/7, 861, and 925, and the amount of total Fak from control (CT) and overloaded myocardium (AoCo). C, Graphic showing the average values (n=4 individual experiments) of densitometric readings for anti-Fak phosphospecific antibodies. D, top row, Anti-Src immunoblotting of anti-Fak immunoprecipitates (Fak/Src). Middle row, Representative immunoblot of myocardial extracts stained with anti-phosphospecific antibody against Src-Tyr418. Bottom row, Anti-Src immunoblotting of myocardial extracts. In the immunoprecipitates of anti-Fak antibodies, a band of immunoglobulin staining was detected at 50 kDa, besides the 60 kDa Src. E, Graphic showing the average values (n=4 individual experiments) of densitometric readings for anti-Src immunoblotting of anti-Fak immunoprecipitates and anti-Src phosphospecific antibody. P<0.05 compared with controls.
Pressure overload induced a rapid phosphorylation of Fak Tyr397, beginning at 3 (2-fold), extending up to 60 minutes (3-fold) after aortic constriction (Figure 1B and 1C). Similar results were observed with antibodies against tyrosine residues 576/577, 861, and 925. Notably, we were unable to detect changes in the phosphorylation of Tyr407 in myocardial extracts from control and acutely overloaded rats (data not shown). No change was observed in the amount of Fak in the immunoblottings of anti-Fak antibody.
Activation of Src by Fak autophosphorylation is central for triggering downstream cellular events.25 Phosphorylation of Tyr418 increases, whereas dephosphorylation decreases Src kinase activity.28 Coimmunoprecipitation assays with anti-Fak and anti-c-Src antibodies showed only a weak binding of Src to Fak in the myocardium of control rats (Figure 1D and 1E). Pressure overload increased the Src binding to Fak, in parallel with Fak Tyr397 phosphorylation, as assessed by coimmunoprecipitation assay with anti-Fak and anti-Src antibodies. Activation of myocardial Src by mechanical stress was examined by Western blotting of myocardial extracts performed with Src phosphospecific antibody against Src-Tyr418. Pressure overload increased Tyr418 phosphorylation, indicating its activation. This occurred while the amount of myocardial Src remained constant, as detected by anti-Src immunoblotting of myocardial extracts.
Myosin Heavy Chain Interacts With Fak
We next screened a rat left ventricle cDNA library with yeast two-hybrid system to identify the molecular components that underpin Fak activation pathway. The screening was performed with a bait of N-terminal Fak fragment spanning amino acids 1 to 748 fused to LEX-A DNA binding domain of the pBTM116 vector (pBTM(ADE2)Fak/NX) (Figure 2A). Approximately 3x104 clones were screened. A coding sequence of 161 amino acids was found as a strongly interacting protein and characterized as be homologous (98%) to the coiled-coil C-terminal region of -cardiac myosin heavy chain (NP 058935-gi: 8393804) (Figure 2B). As a first step toward the confirmation of this interaction, the positive clone containing this fragment was plated in minimal medium without histidine and grown against negative and positive controls (Figure 2C). The positive clone was also tested for ?-galactosidase turning blue (data not shown).
Figure 2. A, Mapping of full-length Fak and the region of Fak protein (FakNX) used as a bait in the two-hybrid screening. Bait consists of Fak N-terminal domain, including FERM domain and the kinase domain. B, Mapping of full-length myosin heavy chain and the fragment found in the two-hybrid screening. Fragment showed 98% homology to the coiled-coil C-terminal region of -cardiac myosin heavy chain (NP 058935-gi: 8393804). C, Fak interacts with myosin on two-hybrid screen. Yeast strain L40 was cotransformed with negative controls: C1(–) pGADT7Rec+pBTM116, C2(–) pGADT7Rec+FakNX, C3(–) pVP16fynSH2+pBTMADE2FAK(F393); positive control: C(+) pVP16fynSH2+FakNX and Fak+myosin: pGADT7REC(cDNA library myosin)+FakNX.
A pulldown assay with the fusion protein GST-C-terminal myosin fragment (Figure 3A) conjugated to sepharose beads and left ventricle extracts was performed to assess the interaction of the C-terminal myosin fragment with Fak. GST-tagged protein binding assay followed by immunoblotting analysis showed the presence of Fak in the GST-C-terminal myosin fragment precipitates obtained from control and overloaded myocardium (10 and 60 minutes) (Figure 3B). The specificity of this interaction was demonstrated by the lack of Fak in GST precipitates. The relative amount of myocardial Fak able to be targeted to GST-C-terminal myosin fragment was examined by comparing the amount of Fak immunoprecipitated by excess of anti-Fak antibody (Figure 3B, lane 1) with the amount of Fak precipitated by GST-C-terminal myosin fragment (Figure 3B, lanes 2 to 4). Interestingly, the amount of Fak targeted by GST-C-terminal myosin fragment slightly increased at 10 but it was reduced in 60-minute overloaded myocardium compared with extracts of control rats. Next, precipitates of the pulldown assays were blotted with anti-Fak-Tyr397 antibody. As shown in the representative example of Figure 3B, staining with anti-Fak-Tyr397 antibody paralleled the changes in the amount of Fak, but was barely detected in the GST-C-terminal myosin fragment precipitates, suggesting that GST-C-terminal myosin fragment might bind preferentially nonphosphorylated Fak. As shown in Figure 3C, assays performed with isolated GST did not retrieve Fak from myocardial extracts.
Figure 3. A, Coomassie dyed 12% SDS PAGE showing GST (26 kDa) and GST-C-terminal myosin fragment (48 kDa) hybrid protein (arrows) used to perform the protein binding assay. B, Representative immunoblottings obtained with anti-Fak and anti-Fak-Tyr397 antibodies of anti-Fak immunoprecipitates (IP) from extracts of control left ventricle and precipitates GST-C-terminal myosin fragment from control, 10- and 60-minute overloaded left ventricles. C, Representative immunoblottings obtained with anti-Fak antibody from extracts treated with GST.
The relative amount of Fak associated with myosin heavy chain was estimated by comparing the amount of Fak immunoprecipitated by excess of anti-Fak antibody with the amount of Fak in the anti-heavy chain cardiac myosin immunoprecipitates. The ability of anti-Fak immunoprecipitation to deplete Fak from myocardial extracts from control and overloaded hearts was demonstrated by the reduction of the amount of Fak in the supernatants to almost an undetectable level (see online expanded Material and Methods). As demonstrated in the representative example and in Figure 4, the amount of Fak precipitated by anti-heavy chain cardiac myosin antibody was 40% of total Fak immunoprecipitated with anti-Fak antibody. Pressure overload markedly reduced the amount of Fak precipitated by anti-heavy chain cardiac myosin antibody, beginning already at 3 minutes and reaching the lowest value in the extracts of 30- and 60-minute overloaded left ventricles. As shown in the representative example of Figure 4, the immunoprecipitates of anti-heavy chain cardiac myosin antibody were barely stained with anti-Fak-Tyr397 antibody, further indicating that Fak associated with myosin heavy chain is mainly nonphosphorylated at Tyr397.
Figure 4. Representative immunoblottings obtained with anti-Fak and anti-Fak-Tyr397 antibodies of anti-Fak and anti-heavy chain cardiac myosin immunoprecipitates (IP) from extracts of control, 3-, 10-, and 60-minute overloaded left ventricles. Graphic shows the average values (n=4 individual experiments) of densitometric readings for anti-Fak staining of anti-Fak and anti-heavy chain cardiac myosin antibodies immunoprecipitates. P<0.05 compared with the amount of Fak in anti-Fak immunoprecipitates.
FAK Colocalizes With Myosin in Cardiac Myocytes
Next, we examined the subcellular localization of Fak by confocal microscopy of isolated cardiac myocytes doubled-stained with anti-Fak and anti-heavy chain cardiac myosin antibodies. Cardiac myocytes stained with anti-heavy chain cardiac myosin antibody (red) revealed striations, suggesting that the labeled structures represent the myosin in the sarcomeric A-band. As shown in the representative example of Figure 5A, in cardiac myocytes from control left ventricles anti-Fak staining overlaps the anti-heavy chain cardiac myosin staining (yellowish staining), indicating that Fak and myosin heavy chain are colocalized in cardiac myocytes from control left ventricles. In cardiac myocytes from 60-minute overloaded left ventricles, cytoplasmic anti-Fak staining (green) was substantially decreased, but it was still regularly distributed along the myofilaments. Interestingly, the cytoplasmic anti-Fak staining of these cells was no longer overlapping the anti-heavy chain cardiac myosin staining, suggesting a relocation of Fak to distinct sarcomeric sites. Notably, in these cardiac myocytes, we found a consistent staining of cell nuclei with anti-Fak antibody. This nuclear localization of Fak was further examined by immunoblotting performed with anti-Fak antibody and tissue fractions of rat left ventricle (Figure 5B). In extracts from control myocardium, Fak was found preferentially in the supernatant fraction (SMM) of low-speed centrifugation, which was enriched in myofibrils, membranes, and soluble proteins, whereas in the 60-minute overloaded myocardium, Fak was found mainly in the pellet of low-speed centrifugation, which is enriched with nuclear proteins. The relative amounts of Fak in SMM and nuclear fractions were estimated by comparison with the amount of Fak from total myocardial extracts. As shown in the representative example and graphic of Figure 5B, after 60 minutes of pressure overload, 70% of Fak was found in the nuclear extracts. We also performed immunoblotting analysis of myocardial fractions with the anti-heavy chain cardiac myosin to examine whether nuclear fractions were free of myofilaments. Although some myosin was detected in nuclear fractions, its amount was similar in the nuclear extracts of control and 60-minute overloaded myocardium, indicating that alterations of Fak levels in nuclear fractions were not related to contamination of nuclear extracts with myofibrils.
Figure 5. A, Double staining of isolated cardiac myocytes from control and 60-minute overloaded (AoCo) rat heart with anti-Fak and anti-heavy chain cardiac myosin antibodies. B, Representative immunoblottings obtained with anti-heavy chain cardiac myosin and anti-Fak from whole myocardium extracts and subcellular fractioning samples. SMM indicates myocardial extract fraction enriched in soluble, membrane, and myofilaments proteins. P<0.05 compared with the amount of Fak in myocardial crude extracts.
Fak subcellular localization was further examined by immunoelectron microscopy. As shown in the representative examples of Figure 6A and 6B, in cardiac myocytes from control left ventricles, immunogold particles were mostly found in the region of sarcomeric A-band. As shown in Figure 6C and 6D, 3 minutes after the onset of pressure overload, anti-Fak immunolabeling was preferentially found as aggregates along the myofilaments, Z-discs, and costameres. Similar localization of Fak was detected in cardiac myocytes from left ventricles subjected to 10-minute pressure overload (data not shown). These results indicated that pressure overload leads to Fak clustering at different subcellular sites.
Figure 6. A and B, Anti-Fak antibody immunogold labeling of myocardium from control hearts. Arrows (B) indicate specific anti-Fak antibody staining close to myofilaments. C through E, Immunogold labeling of Fak in myocardium obtained from 3-minute overloaded hearts. Arrows indicate Fak clusters are found at costameres (D), Z-discs (E), and myofilaments (E). Magnification: A, 31 620x; C, 25 300x.
Fak is rapidly activated and triggers the assembly of a multicomponent signaling complex that has been considered to occupy a central position in the transduction and coordination of the earlier responses of cardiac myocytes to mechanical stress. In the present study, we showed comprehensive data on Fak tyrosine phosphorylation, subcellular distribution, and interaction with C-terminal region of myosin heavy chain, providing insight on the mechanism of its activation by mechanical forces in cardiac myocytes.
By using phosphospecific antibodies against tyrosine residues of Fak and Src, we extended previous demonstration of Fak/Src signaling complex activation by mechanical stress, to show the load-induced phosphorylation of specific Fak tyrosine residues Tyr397, 576/7, 861, and 925 in the rat myocardium. These results agree with a proposed model for Fak activation triggered by autophosphorylation at Tyr397 followed by the engagement of Src, which phosphorylates additional Fak tyrosine residues.29 Consistent with this, we have shown previously that the activation of Fak/Src complex by mechanical stress in cardiac myocytes is strictly dependent on phosphorylation of Fak at Tyr397.20 The phosphorylation of additional tyrosine residues supports the notion that Fak may work as a docking protein providing a scaffold to other signaling molecules related to the activation of downstream pathways involved in multiple cell functions. Phosphorylation of Tyr397 not only starts cooperation with Src, but it also appears to be important for the recruitment of other SH2-containing proteins.30–33 Otherwise, the phosphorylation of Tyr397, as well as Tyr925, creates binding sites for the Grb2-SOS complex.26,34,35 Accordingly, we have previously shown21,23 that pressure overload increases the association of Fak with PI3-kinase and Grb2, which occurs simultaneously to Akt and Erk1/2 activation. Additional studies will be necessary to dissect out the potential role of the activation of these signaling systems to Fak influence on early gene expression in response to mechanical stress in cardiac myocytes.
By using yeast-two hybrid screening of a rat left ventricle cDNA library, pulldown, and coimmunoprecipitation assays, we demonstrated that Fak interacts with a C-terminal region of myosin heavy chain. Furthermore, data from confocal immunofluorescence and immunoelectron microscopy with specific antibodies supports the notion that Fak is localized in the sarcomeric A-band, associated with myosin heavy chain. Comparisons of the amount of Fak immunoprecipitated by excess of anti-heavy chain cardiac myosin with that of anti-Fak antibody allowed us to estimate that 40% of total myocardial Fak is associated with myosin heavy chain. This data raises the question of the localization of the remaining 60% of myocardial Fak. Although we did not perform additional quantitative characterization of Fak distribution in the myocardium of rat left ventricle, it is plausible to assume that this remaining Fak was localized elsewhere in distinct subcellular compartments of cardiac myocytes and other myocardial cell types. Indeed, the immunoelectron microscopy imaging of cardiac myocytes from left ventricles of control rats showed that anti-Fak staining was detected in structures such as costameres and Z discs, although less frequently than in sarcomeric A band. This is the first demonstration of a protein partner of Fak peculiar to muscle cells. The amino acid sequence of C-terminal myosin heavy chain identified as a target to Fak is part of the long C-terminal myosin heavy chain sequence that drives the assembly of two molecules of myosin into a coiled-coil that constitutes the structural backbone of the thick filament. Although Fak/myosin interaction was not previously known, the ability of Fak to directly bind the coiled-coil regions of the p190RhoGEF36 and the GIT1 ArfGAP37 protein were recently demonstrated. In both cases the interaction occurred via a helical bundle structure of the C-terminal Fak FAT domain by means of an unknown motif. However, in the present study, the two-hybrid screening was performed with Fak N-terminal sequence including the FERM and kinase, but not the FAT domain. Although the N-terminal domain of Fak has also been shown to interact with receptor and cytoplasmic tyrosine kinases, it is not clear yet the specific region responsible for this interaction. Further studies will be needed to identify the surface residues of N-terminal Fak required for myosin heavy chain binding and how its C-terminal region functions to promote Fak association.
Fak phosphospecific antibody against Tyr397 could not detect a significant amount of Fak in the immunoprecipitates of anti-heavy chain cardiac myosin antibody, indicating that most Fak associated with this sarcomeric protein is inactive. Conversely, Fak activation by mechanical stress was paralleled by a drastic reduction in the association of Fak with myosin heavy chain, indicating that such interaction is subjected to regulation by mechanical stress. Notably, a low amount of phosphorylated Fak was found in the precipitates of the pulldown assays, even in the extracts of overloaded myocardium, indicating a preferential interaction of GST-C-terminal cardiac myosin to nonphosphorylated Fak. Together, these results raised the interesting possibility of a direct connection between myosin heavy chain and Fak activation by mechanical stress in cardiac myocytes. Incidentally, myosin heavy chain contains small folded segments that make it a truly elastic protein.38 On the other hand, a recent report39 indicates that Fak activation is dependent on the release of an autoinhibitory interaction of the N-terminal FERM and the kinase domains. Thus, it is conceivable that myosin stretch might cause Fak to assume a conformation that favors Tyr397 autophosphorylation. This implies that C-terminal region of myosin heavy chain filaments might work as a true mechanotransducer, transmitting mechanical forces to activate Fak. However, further studies are needed to confirm the assumptions of this model, as well as to exclude the contribution of agonist mediated Fak activation in overloaded myocardium.
We also found that in the early period after the beginning of pressure overload sarcoplasmic Fak-specific immunofluorescence, although reduced, was no longer overlapping the anti-myosin staining, suggesting a relocation of Fak to distinct sarcomeric sites. This was supported by the findings that immunogold staining was frequently seen as aggregates close to Z-discs and costameres. These findings, together with the demonstration that Fak is rapidly phosphorylated by pressure overload, is consistent with the idea that Fak activation by mechanical stress is accompanied by clustering and translocation to diverse subcellular structures of cardiac myocytes. This agrees with data from our previous studies in isolated cardiac myocytes of neonatal rats,20 indicating that controlled stretch induces Fak to aggregate at the myofilaments, Z-discs, and costameres. Moreover, it is interesting to note that Fak phosphorylation at tyrosine residues have been shown to regulate the subcellular localization of Fak.40 The functional significance of Fak clustering is not completely understood at present, but clustering has been considered to optimize Fak signaling because the molecular proximity in clusters may serve to enhance and sustain Fak signaling.41 Thus, Fak cluster and target to Z-discs and costameres, which also serve as sites for mechanical sensing, might account for the amplification of Fak signaling triggered by mechanical stress in cardiac myocytes. This implies that the lack of Z-disc integrity would impair the full activation of Fak by mechanical stress. Therefore, it will be interesting to determine whether Fak plays a role in the pathogenesis of cardiomyopathy associated with the lack of structural integrity of Z discs.
Translocation of activated Fak was further supported by our finding in this study that after 60 minutes of pressure overload, almost 70% of Fak was found in the myocardial nuclear extracts. This was strengthened by the demonstration of a consistent staining with anti-Fak antibody of nuclei of cardiac myocyte isolated from 60-minute overloaded left ventricles. Similar translocation of Fak to nucleus was recently reported in cardiac myocytes of spontaneously hypertensive heart failure rats, suggesting that Fak might play a role in the regulation of nuclear processes in response to mechanical stress.42 However, the implications of such findings remain to be determined.
In conclusion, the demonstration in this study of the subcellular localization of Fak within the sarcomeric A-bands and its interaction with myosin heavy chain have implications for our understanding of how cardiac myocytes sense and transmit mechanical stimuli. Mechanistic insights of these findings may disclose new pathogenetic aspects of mechanical stress in myocardial hypertrophy and failure.
References
Katz AM. The cardiomyopathy of overload: an unnatural growth response in the hypertrophied heart. Ann Intern Med. 1994; 121: 363–371.
Chien KR. Stress pathways and heart failure. Cell. 1999; 98: 555–558.
Sussman MA, McCulloch A, Borg TK. Dance band on the titanic: biomechanical signaling in cardiac hypertrophy. Circ Res. 2002; 91: 888–898.
Epstein ND, Davis JS. Sensing stretch is fundamental. Cell. 2003; 112: 147–150.
Clark KA, McElhinny AS, Beckerle MC, Gregorio CC. Striated muscle cytoarchitecture: an intrincate web of form and function. Ann Rev Cell Dev Biol. 2002; 18: 637–706.
Knoll R, Hoshijima M, Hoffman HM, Person V, Lorenzen-Schmidt I, Bang ML, Hayashi T, Shiga N, Yasukawa H, Schaper W, McKenna W, Yokoyama M, Schork NJ, Omens JH, McCulloch AD, Kimura A, Gregorio CC, Poller W, Schaper J, Schultheiss HP, Chien KR. The cardiac mechanical stretch sensor machinery involves a Z disc complex that is defective in a subset of human dilated cardiomyopathy. Cell. 2002; 111: 943–955.
Gerull B, Gramlich M, Atherton J, McNabb M, Trombitas K, Sasse-Klaassen S, Seidman JG, Seidman C, Granzier H, Labeit S, Frenneaux M, Thierfelder L. Mutations of TTN, encoding the giant muscle filament titin, cause familial dilated cardiomyopathy. Nat Genet. 2002; 30: 201–204.
Arber S, Hunter JJ, Ross J Jr, Hongo M, Sansig G, Borg J, Perriard JC, Chien KR, Caroni P. MLP-deficient mice exhibit a disruption of cardiac cytoarchitectural organization, dilated cardiomyopathy, and heart failure. Cell. 1997; 88: 393–403.
Faulkner G, Lanfranchi G, Valle G. Telethonin and other new proteins of the Z-disc of skeletal muscle. IUBMB Life. 2001; 51: 275–282.
Vatta M, Mohapatra B, Jimenez S, Sanchez X, Faulkner G, Perles Z, Sinagra G, Lin JH, Vu TM, Zhou Q, Bowles KR, Di Lenarda A, Schimmenti L, Fox M, Chrisco MA, Murphy RT, McKenna W, Elliott P, Bowles NE, Chen J, Valle G, Towbin JA. Mutations in Cypher/ZASP in patients with dilated cardiomyopathy and left ventricular non-compaction. J Am Coll Cardiol. 2003; 42: 2014–2027.
Terracio L, Rubin K, Gullberg D, Balog E, Carver W, Jyring R, Borg TK. Expression of collagen binding integrins during cardiac development and hypertrophy. Circ Res. 1991; 68: 734–744.
Borg TK, Goldsmith EC, Price R, Carver W, Terracio L, Samarel AM. Specialization at the Z line of cardiac myocytes. Cardiovasc Res. 2000; 46: 277–285.
Ross RS, Pham C, Shai SY, Goldhaber JI, Fenczik C, Glembotski CC, Ginsberg MH, Loftus JC. ?1 Integrin participates in the hypertrophic response of rat ventricular myocytes. Circ Res. 1998; 82: 1160–1172.
Shai S-Y, Harpf AE, Babbitt CJ, Jordan MC, Fishbein MC, Chen J, Omura M, Leil TA, Becker KD, Jiang M, Smith DJ, Cherry SR, Loftus JC, Ross RS. Cardiac myocyte-specific excision of the 1 integrin gene results in myocardial fibrosis and cardiac failure. Circ Res. 2002; 90: 458–464.
Hanks SK, Calalb MB, Harper MC, Patel SK. Focal adhesion protein-tyrosine kinase phosphorylated in response to cell attachment to fibronectin. Proc Natl Acad Sci U S A. 1992; 89: 8487–8491.
Wang H-B, Dembo M, Hanks SK, Wang Y-L. Focal adhesion kinase is involved in mechanosensing during fibroblast migration. Proc Natl Acad Sci U S A. 2001; 98: 11295–11300.
Li S, Butler P, Wang Y, Hu Y, Han DC, Usami S, Guan J-L, Chien S. The role of the dynamics of focal adhesion kinase in the mechanotaxis of endothelial cells. Proc Natl Acad Sci U S A. 2002; 99: 3546–3551.
Seko Y, Takahashi N, Tobe K, Kadowaki T, Yazaki Y. Pulsatile stretch activates mitogen-activated protein kinase (MAPK) family members and focal adhesion kinase [p125(FAK)] in cultured rat cardiac myocytes. Biochem Biophys Res Commun. 1999; 259: 8–14.
Aikawa R, Nagai T, Kudoh S, Zou Y, Tanaka M, Tamura M, Akazawa H, Takano H, Nagai R, Komuro I. Integrins play a critical role in mechanical stress-induced p38 MAPK activation. Hypertension. 2002; 39: 233–238.
Torsoni AS, Constancio SS, Nadruz, Jr W, Hanks SK, Franchini KG. Focal adhesion kinase is activated and mediates the early hypertrophic response to stretch in cardiac myocytes. Circ Res. 2003; 93: 140–147.
Franchini KG, Torsoni AS, Soares PHA, Saad MJA. Early activation of the multicomponent signaling complex associated with focal adhesion kinase induced by pressure overload in the rat heart. Circ Res. 2000; 87: 558–565.
Laser M, Willey CD, Jiang W, Cooper C, Menick DR, Zile MR, Kuppuswamy D. Integrin activation and focal complex formation in cardiac hypertrophy. J Biol Chem. 2000; 275: 35624–35630.
Domingos PP, Fonseca PM, Nadruz W Jr., Franchini KG. Load-induced focal adhesion kinase activation in the myocardium: role of stretch and contractile activity. Am J Physiol. 2002; 282: H556–H564.
Bayer AL, Heidkamp MC, Patel N, Porter MJ, Engman SJ, Samarel AM. PYK2 expression and phosphorylation increases in pressure overload-induced left ventricular hypertrophy. Am J Physiol. 2002; 283: H695–H706.
Calalb MB, Polte TR, Hanks SK. Tyrosine phosphorylation of focal adhesion kinase at sites in the catalytic domain regulates kinase activity: a role for Src family kinases. Mol Cell Biol. 1995; 15: 954–963.
Schlaepfer DD, Hanks SK, Hunter T, van der Geer P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature. 1994; 372: 786–791.
Polte TR, Hanks SK. Complexes of focal adhesion kinase (FAK) and Crk-associated substrate (p130Cas) are elevated in cytoskeletal-associated fractions following adhesion and Src transformation: requirements for Src kinase activity and FAK proline-rich motifs. J Biol Chem. 1997; 272: 5501–5509.
Schaller MD, Hildebrand JD, Shannon JD, Fox JW, Vines RR, Parsons JT. Autophosphorylation of the focal adhesion kinase, pp125FAK, directs SH2-dependent binding of pp60src. Mol Cell Biol. 1994; 14: 1680–1688.
Parsons JT. Focal adhesion kinase: the first ten years. J Cell Sci. 2003; 116: 1409–1416.
Akagi T, Murata K, Shishido T, Hanafusa H. v-Crk activates the phosphoinositide 3-kinase/AKT pathway by utilizing focal adhesion kinase and H-Ras. Mol Cell Biol. 2002; 22: 7015–7023.
Chen HC, Appeddu PA, Isoda H, Guan JL. Phosphorylation of tyrosine 397 in focal adhesion kinase is required for binding phosphatidylinositol 3-kinase. J Biol Chem. 1996; 271: 26329–26334.
Chen HC, Guan JL. Stimulation of phosphatidylinositol 3'-kinase association with foca adhesion kinase by platelet-derived growth factor. J Biol Chem. 1994; 269: 31229–31233.
Han DC, Guan JL. Association of focal adhesion kinase with Grb7 and its role in cell migration. J Biol Chem. 1999; 274: 24425–24430.
Chen Q, Kinch MS, Lin TH, Burridge K, Juliano RL. Integrin-mediated cell adhesion activates mitogen-activated protein kinases. J Biol Chem. 1994; 269: 26602–26605.
Schlaepfer DD, Hunter T. Evidence for in vivo phosphorylation of the Grb2 SH2-domain binding site on focal adhesion kinase by Src-family protein-tyrosine kinases. Mol Cell Biol. 1996; 16: 5623–5633.
Zhai J, Lin H, Nie Z, Wu J, Canete-Soler R, Schlaepfer WW, Schlaepfer DD. Direct interaction of focal adhesion kinase with p190RhoGEF. J Biol Chem. 2003; 278: 24865–24873.
Zhao ZS, Manser EL, Loo TH, Lim L. Coupling of PAK-Interacting Exchange Factor PIX to GIT1 Promotes Focal Complex Disassembly. Mol Cell Biol. 2000; 20: 6354–6363.
Schwaiger I, Sattler C, Hostetter DR, Rief M. The myosin coiled-coil is a truly elastic protein structure. Nat Mater. 2002; 1: 232–235.
Cooper LA, Shen TL, Guan JL. Cooper Regulation of focal adhesion kinase by its amino-terminal domain through an autoinhibitory interaction. Mol Cell Biol. 2003; 23: 8030–8041.
Katz B-Z, Romer L, Miyamoto S, Volberg T, Matsumoto K, Cukierman E, Geiger B, Yamada KM. Targeting membrane-localized focal adhesion kinase to focal adhesions: roles of tyrosine phosphorylation and SRC family kinases. J Biol Chem. 2003; 278: 29115–29120.
Katz B-Z, Miyamoto S, Teramoto H, Zohar M, Krylov D, Vinson C, Gutkind JS, Yamada KM. Direct transmembrane clustering and cytoplasmic dimerization of focal adhesion kinase initiates its tyrosine phosphorylation. Biochim Biophys Acta. 2002; 1592: 141–152.
Yi XP, Wang X, Gerdes AM, Li F. Subcellular redistribution of focal adhesion kinase and its related nonkinase in hypertrophic myocardium. Hypertension. 2003; 41: 1317–1323.(Priscila M. Fonseca, Rosa)
Focal adhesion kinase (Fak) has been implicated as a signaling molecule involved in the early response of cardiac myocytes to mechanical stress. The mechanism of Fak activation by mechanical stimuli is not clear. In this study, we report the load-induced Fak activation and its association with myosin heavy chain in cardiac myocytes. Pressure overload lasting from 3 to 60 minutes was shown to induce Fak phosphorylation at Tyr-397, -576/7, -861, and -925 as detected by phosphospecific antibodies. This was paralleled by increases of Fak/Src association and Src activity (Tyr-418 phosphorylation). Yeast two-hybrid screening of an adult rat cDNA library revealed an interaction between Fak and C-terminal coiled-coil region of -myosin heavy chain. This was confirmed by pulldown assay with GST-C-terminal myosin fragment and native Fak from rat left ventricle. Such interaction was confirmed by coimmunoprecipitation assay with anti-Fak and anti-heavy chain cardiac myosin antibodies, confocal microscopy of double-labeled isolated cardiac myocytes and immunoelectron microscopy with anti-Fak antibody. Fak activation by mechanical stress was accompanied by a reduction of Fak/myosin heavy chain association and its relocation at subcellular sites such as costameres, Z-discs, and nuclei. Thus, our present data identify Fak interaction with C-terminal region of myosin heavy chain adding comprehensive data on Fak activation by mechanical stress and mechanotransduction in cardiac myocytes.
Key Words: focal adhesion kinase mechanotransduction cell signaling hypertrophy myosin
Mechanical stress is a major factor involved in the development of myocardial adaptive and maladaptive changes in heart diseases. Local mechanical forces activate signaling mechanisms in cardiac myocytes inducing the expression of specific genetic programs linked to myocardial structural and functional remodeling.1,2 Although mechanical forces might directly trigger signaling mechanisms in cardiac myocytes, the mechanism by which they are sensed and converted to biochemical signals remains elusive.
Structures such as sarcomeric lattice, cytoskeleton, and the extracellular matrix operate in the transmission of either passive or active forces in cardiac myocytes.3,4 Studies have confirmed the critical importance of the molecular integrity of Z-disc and cytoskeleton to the expression of genetic program induced by mechanical stress in cardiac myocytes. Z-disc structure is organized by N-terminal titin Z repeats linked to -actinin and associated proteins such as MLP, ALP, telethonin (T-cap), cypher/Zasp, and myotilin.4,5 Notably, MLP null mice were shown to fail to upregulate brain and atrial natriuretic factors mRNA in response to stretch.6 Human titin mutations as well as deletion of the -actinin binding proteins such as ALP or MLP in the mouse causes dilated cardiomyopathy.6–10 However, the mechanisms by which these structures and proteins detect physical forces and initiate biochemical signals are yet unclear.
The link of Z-discs to sarcolemma at sites known as costameres has been implicated as a potential signaling station to mechanotransduction in cardiac myocytes. Costameres are also sites of integrin clusters and share a similarity to focal adhesions.11,12 Focal adhesion kinase (Fak), a primary integrin effector at focal adhesion sites,15–17 is rapidly activated by mechanical stimuli in cultured neonatal rat ventricular myocytes18–20 and in overloaded myocardium of adult animals.21–24 The importance of Fak to the regulation of early gene transcription in response to stretch was demonstrated in neonatal rat cardiac myocytes,20 indicating that this kinase may coordinate signaling pathways involved in the hypertrophic growth induced by mechanical stress. However, the mechanisms responsible for Fak activation by mechanical stress in cardiac myocytes are still unclear. Although Fak is thought to colocalize with integrins at costameres,3,12–14 there is no precise description of Fak localization in cardiac myocytes. We have previously shown20 that stretch induces Fak to cluster at myofilaments in neonatal rat ventricular myocytes, but no evidence linked Fak directly to costameres in such cells. Otherwise, experiments performed with dominant-negative Fak and specific pharmacological Src inhibitor confirmed that Fak activation is strictly dependent on phosphorylation of Tyr397 and cooperation with Src as has been shown in the focal adhesion site. However, the lack of accurate data on localization as well as on the signaling and structural protein partners of Fak in cardiac myocytes preclude a better understanding of Fak activation by mechanical stress in this particular cell type.
Thus, we searched for novel binding partners for Fak by yeast two-hybrid screening of an adult rat left ventricle cDNA library combined with an analysis of Fak localization with immunohistochemistry and immunoelectron microscopy of control and overloaded rat myocardium.
Detailed methods are described in the expanded Materials and Methods in the online data supplement available at http://circres.ahajournals.org.
Animal Model
Male Wistar rats (160 to 200 g) underwent acute pressure overload induced by controlled constriction of transversal aorta as previously described.21 The animals were obtained from the Central Animal House of the University of Campinas and received care in compliance with the principles of laboratory animal care formulated by the Animal Care and Use Committee of the University of Campinas.
Protein Analysis by Immunoblotting
Aliquots of left ventricle extracts (60 μg) containing equal amount of total protein were resolved on SDS-PAGE and transferred to nitrocellulose membranes, which were incubated with specific antibodies and [125I] Protein A. Band intensities were quantified through optical densitometry of the developed autoradiographs.
Subcellular Fractioning
Subcellular fraction enriched in nuclear proteins was obtained from the pellet of left ventricle extracts centrifuged at 1000g. The remaining supernatant fraction that was enriched in soluble, membrane, and myofilament proteins was designated SMM fraction.
Laser Confocal Analysis
Isolated cardiac myocytes were fixed with 4% paraformaldehyde/sucrose and incubated with anti-Fak and anti-heavy chain cardiac myosin primary antibodies. This was followed by incubation with secondary goat anti-rabbit Alexa488-conjugated and rabbit anti-mouse Alexa568-conjugated antibody. Images were obtained with laser confocal microscope (Zeiss LSM510).
Immunoelectron Microscopy
Fragments of left ventricle were fixed in paraformaldehyde/glutaraldehyde (Electron Microscopy Sciences) and embedded in LR White resin at –20°C under UV light. Thin-sections were stained with anti-Fak antibody overnight at 4°C. The sections were then incubated with anti-rabbit IgG-10 nm gold-conjugated antibody diluted 1:20 in 1% BSA (pH 8.2). The grids were stained with 5% uranyl acetate and 0.5% lead citrate, examined, and photographed in a transmission electron microscope (LEO 906). Negative controls were accomplished by incubation with the primary antibody omitted.
cDNA Library
A cDNA library was constructed from of an adult rat LV total RNA. The purified library cDNA was fused to the GAL4 activation domain of pGADT7-rec expressing vector by recombination in L40 yeast strain.
Yeast Two-Hybrid Screening
The pBTM (ADE2)Fak/NX bait construction (provided by Steven K. Hanks, Vanderbilt University School of Medicine, Nashville, Tenn), expressed a fragment spanning 748 amino acid N terminus of Fak protein fused to LEX A DNA binding domain of the pBTM116 vector. L40 yeast cells with left ventricle cDNA library were cotransformed with FAK/NX bait. Positive clones were tested by galactosidase assay, and cDNA cloned in pGADT7-rec was sequenced. Sequences were translated into amino acid sequence by the ORF Finder and submitted to BLAST and Clustal W in the NCBI Database (http://www.ncbi.nlm.nih.gov/).
Protein Expression
The encoding C-terminal region of -myosin heavy chain retrieved after yeast two-hybrid screening was inserted into the vector pGEX5x2 (Amersham Pharmacia Biotech) for expression of recombinant GST-tagged protein in Escherichia coli BL21 (RIL) Codon Plus.
Pulldown Assay
GST-tagged myosin fragment (48.2 kDa) was used for the pulldown assay with LV extracts. GST conjugated glutathione beads were used as negative control for nonspecific binding. The pulldown pellets were resolved on SDS-PAGE and the membranes were stained with anti-Fak antibody.
Statistical Analysis
Data are presented as mean±SEM. Differences between the mean values of the densitometric readings were tested by ANOVA and Bonferroni multiple-range test. A value of P<0.05 indicated statistical significance.
Pressure Overload Induces Fak Phosphorylation at Tyr397, 576/7, 861, and 925
The autophosphorylation of Fak Tyr397 has been shown to be critical for Fak activation. Phosphorylated Tyr397 recruits Src family kinases, which lead to phosphorylation of additional Fak tyrosine residues, which promote the assembly of distinct higher-order individual signaling complexes,25–27 providing a mechanism for coordinating signaling through multiple pathways. We previously demonstrated21,23 a rapid increase in the phosphorylation and activity of Fak in the myocardium of rats in response to mechanical stimuli. In this study, we examined the effects of pressure overload on phosphorylation of myocardial Fak Tyr397, 407, 576/577, 861, and 925 with phosphospecific antibodies. Experiments were performed in rats subjected to constriction of transverse aorta lasting from 3 to 60 minutes. Blood pressure above and below the level of aortic constriction was monitored to ascertain the stability of the experimental preparation. Aortic constriction produced a rapid and sustained increase of blood pressure measured in the ascending aorta with no change in blood pressure in the abdominal aorta, as compared with blood pressure of control period (Figure 1A).
Figure 1. A, Representative blood pressure recording in ascending and abdominal aorta. Aortic constriction was adjusted to increase the ascending aorta systolic pressure by 40 mm Hg, without reducing the mean blood pressure levels in the abdominal aorta. B, Representative blots showing Fak phosphorylation at tyrosines 397, 576/7, 861, and 925, and the amount of total Fak from control (CT) and overloaded myocardium (AoCo). C, Graphic showing the average values (n=4 individual experiments) of densitometric readings for anti-Fak phosphospecific antibodies. D, top row, Anti-Src immunoblotting of anti-Fak immunoprecipitates (Fak/Src). Middle row, Representative immunoblot of myocardial extracts stained with anti-phosphospecific antibody against Src-Tyr418. Bottom row, Anti-Src immunoblotting of myocardial extracts. In the immunoprecipitates of anti-Fak antibodies, a band of immunoglobulin staining was detected at 50 kDa, besides the 60 kDa Src. E, Graphic showing the average values (n=4 individual experiments) of densitometric readings for anti-Src immunoblotting of anti-Fak immunoprecipitates and anti-Src phosphospecific antibody. P<0.05 compared with controls.
Pressure overload induced a rapid phosphorylation of Fak Tyr397, beginning at 3 (2-fold), extending up to 60 minutes (3-fold) after aortic constriction (Figure 1B and 1C). Similar results were observed with antibodies against tyrosine residues 576/577, 861, and 925. Notably, we were unable to detect changes in the phosphorylation of Tyr407 in myocardial extracts from control and acutely overloaded rats (data not shown). No change was observed in the amount of Fak in the immunoblottings of anti-Fak antibody.
Activation of Src by Fak autophosphorylation is central for triggering downstream cellular events.25 Phosphorylation of Tyr418 increases, whereas dephosphorylation decreases Src kinase activity.28 Coimmunoprecipitation assays with anti-Fak and anti-c-Src antibodies showed only a weak binding of Src to Fak in the myocardium of control rats (Figure 1D and 1E). Pressure overload increased the Src binding to Fak, in parallel with Fak Tyr397 phosphorylation, as assessed by coimmunoprecipitation assay with anti-Fak and anti-Src antibodies. Activation of myocardial Src by mechanical stress was examined by Western blotting of myocardial extracts performed with Src phosphospecific antibody against Src-Tyr418. Pressure overload increased Tyr418 phosphorylation, indicating its activation. This occurred while the amount of myocardial Src remained constant, as detected by anti-Src immunoblotting of myocardial extracts.
Myosin Heavy Chain Interacts With Fak
We next screened a rat left ventricle cDNA library with yeast two-hybrid system to identify the molecular components that underpin Fak activation pathway. The screening was performed with a bait of N-terminal Fak fragment spanning amino acids 1 to 748 fused to LEX-A DNA binding domain of the pBTM116 vector (pBTM(ADE2)Fak/NX) (Figure 2A). Approximately 3x104 clones were screened. A coding sequence of 161 amino acids was found as a strongly interacting protein and characterized as be homologous (98%) to the coiled-coil C-terminal region of -cardiac myosin heavy chain (NP 058935-gi: 8393804) (Figure 2B). As a first step toward the confirmation of this interaction, the positive clone containing this fragment was plated in minimal medium without histidine and grown against negative and positive controls (Figure 2C). The positive clone was also tested for ?-galactosidase turning blue (data not shown).
Figure 2. A, Mapping of full-length Fak and the region of Fak protein (FakNX) used as a bait in the two-hybrid screening. Bait consists of Fak N-terminal domain, including FERM domain and the kinase domain. B, Mapping of full-length myosin heavy chain and the fragment found in the two-hybrid screening. Fragment showed 98% homology to the coiled-coil C-terminal region of -cardiac myosin heavy chain (NP 058935-gi: 8393804). C, Fak interacts with myosin on two-hybrid screen. Yeast strain L40 was cotransformed with negative controls: C1(–) pGADT7Rec+pBTM116, C2(–) pGADT7Rec+FakNX, C3(–) pVP16fynSH2+pBTMADE2FAK(F393); positive control: C(+) pVP16fynSH2+FakNX and Fak+myosin: pGADT7REC(cDNA library myosin)+FakNX.
A pulldown assay with the fusion protein GST-C-terminal myosin fragment (Figure 3A) conjugated to sepharose beads and left ventricle extracts was performed to assess the interaction of the C-terminal myosin fragment with Fak. GST-tagged protein binding assay followed by immunoblotting analysis showed the presence of Fak in the GST-C-terminal myosin fragment precipitates obtained from control and overloaded myocardium (10 and 60 minutes) (Figure 3B). The specificity of this interaction was demonstrated by the lack of Fak in GST precipitates. The relative amount of myocardial Fak able to be targeted to GST-C-terminal myosin fragment was examined by comparing the amount of Fak immunoprecipitated by excess of anti-Fak antibody (Figure 3B, lane 1) with the amount of Fak precipitated by GST-C-terminal myosin fragment (Figure 3B, lanes 2 to 4). Interestingly, the amount of Fak targeted by GST-C-terminal myosin fragment slightly increased at 10 but it was reduced in 60-minute overloaded myocardium compared with extracts of control rats. Next, precipitates of the pulldown assays were blotted with anti-Fak-Tyr397 antibody. As shown in the representative example of Figure 3B, staining with anti-Fak-Tyr397 antibody paralleled the changes in the amount of Fak, but was barely detected in the GST-C-terminal myosin fragment precipitates, suggesting that GST-C-terminal myosin fragment might bind preferentially nonphosphorylated Fak. As shown in Figure 3C, assays performed with isolated GST did not retrieve Fak from myocardial extracts.
Figure 3. A, Coomassie dyed 12% SDS PAGE showing GST (26 kDa) and GST-C-terminal myosin fragment (48 kDa) hybrid protein (arrows) used to perform the protein binding assay. B, Representative immunoblottings obtained with anti-Fak and anti-Fak-Tyr397 antibodies of anti-Fak immunoprecipitates (IP) from extracts of control left ventricle and precipitates GST-C-terminal myosin fragment from control, 10- and 60-minute overloaded left ventricles. C, Representative immunoblottings obtained with anti-Fak antibody from extracts treated with GST.
The relative amount of Fak associated with myosin heavy chain was estimated by comparing the amount of Fak immunoprecipitated by excess of anti-Fak antibody with the amount of Fak in the anti-heavy chain cardiac myosin immunoprecipitates. The ability of anti-Fak immunoprecipitation to deplete Fak from myocardial extracts from control and overloaded hearts was demonstrated by the reduction of the amount of Fak in the supernatants to almost an undetectable level (see online expanded Material and Methods). As demonstrated in the representative example and in Figure 4, the amount of Fak precipitated by anti-heavy chain cardiac myosin antibody was 40% of total Fak immunoprecipitated with anti-Fak antibody. Pressure overload markedly reduced the amount of Fak precipitated by anti-heavy chain cardiac myosin antibody, beginning already at 3 minutes and reaching the lowest value in the extracts of 30- and 60-minute overloaded left ventricles. As shown in the representative example of Figure 4, the immunoprecipitates of anti-heavy chain cardiac myosin antibody were barely stained with anti-Fak-Tyr397 antibody, further indicating that Fak associated with myosin heavy chain is mainly nonphosphorylated at Tyr397.
Figure 4. Representative immunoblottings obtained with anti-Fak and anti-Fak-Tyr397 antibodies of anti-Fak and anti-heavy chain cardiac myosin immunoprecipitates (IP) from extracts of control, 3-, 10-, and 60-minute overloaded left ventricles. Graphic shows the average values (n=4 individual experiments) of densitometric readings for anti-Fak staining of anti-Fak and anti-heavy chain cardiac myosin antibodies immunoprecipitates. P<0.05 compared with the amount of Fak in anti-Fak immunoprecipitates.
FAK Colocalizes With Myosin in Cardiac Myocytes
Next, we examined the subcellular localization of Fak by confocal microscopy of isolated cardiac myocytes doubled-stained with anti-Fak and anti-heavy chain cardiac myosin antibodies. Cardiac myocytes stained with anti-heavy chain cardiac myosin antibody (red) revealed striations, suggesting that the labeled structures represent the myosin in the sarcomeric A-band. As shown in the representative example of Figure 5A, in cardiac myocytes from control left ventricles anti-Fak staining overlaps the anti-heavy chain cardiac myosin staining (yellowish staining), indicating that Fak and myosin heavy chain are colocalized in cardiac myocytes from control left ventricles. In cardiac myocytes from 60-minute overloaded left ventricles, cytoplasmic anti-Fak staining (green) was substantially decreased, but it was still regularly distributed along the myofilaments. Interestingly, the cytoplasmic anti-Fak staining of these cells was no longer overlapping the anti-heavy chain cardiac myosin staining, suggesting a relocation of Fak to distinct sarcomeric sites. Notably, in these cardiac myocytes, we found a consistent staining of cell nuclei with anti-Fak antibody. This nuclear localization of Fak was further examined by immunoblotting performed with anti-Fak antibody and tissue fractions of rat left ventricle (Figure 5B). In extracts from control myocardium, Fak was found preferentially in the supernatant fraction (SMM) of low-speed centrifugation, which was enriched in myofibrils, membranes, and soluble proteins, whereas in the 60-minute overloaded myocardium, Fak was found mainly in the pellet of low-speed centrifugation, which is enriched with nuclear proteins. The relative amounts of Fak in SMM and nuclear fractions were estimated by comparison with the amount of Fak from total myocardial extracts. As shown in the representative example and graphic of Figure 5B, after 60 minutes of pressure overload, 70% of Fak was found in the nuclear extracts. We also performed immunoblotting analysis of myocardial fractions with the anti-heavy chain cardiac myosin to examine whether nuclear fractions were free of myofilaments. Although some myosin was detected in nuclear fractions, its amount was similar in the nuclear extracts of control and 60-minute overloaded myocardium, indicating that alterations of Fak levels in nuclear fractions were not related to contamination of nuclear extracts with myofibrils.
Figure 5. A, Double staining of isolated cardiac myocytes from control and 60-minute overloaded (AoCo) rat heart with anti-Fak and anti-heavy chain cardiac myosin antibodies. B, Representative immunoblottings obtained with anti-heavy chain cardiac myosin and anti-Fak from whole myocardium extracts and subcellular fractioning samples. SMM indicates myocardial extract fraction enriched in soluble, membrane, and myofilaments proteins. P<0.05 compared with the amount of Fak in myocardial crude extracts.
Fak subcellular localization was further examined by immunoelectron microscopy. As shown in the representative examples of Figure 6A and 6B, in cardiac myocytes from control left ventricles, immunogold particles were mostly found in the region of sarcomeric A-band. As shown in Figure 6C and 6D, 3 minutes after the onset of pressure overload, anti-Fak immunolabeling was preferentially found as aggregates along the myofilaments, Z-discs, and costameres. Similar localization of Fak was detected in cardiac myocytes from left ventricles subjected to 10-minute pressure overload (data not shown). These results indicated that pressure overload leads to Fak clustering at different subcellular sites.
Figure 6. A and B, Anti-Fak antibody immunogold labeling of myocardium from control hearts. Arrows (B) indicate specific anti-Fak antibody staining close to myofilaments. C through E, Immunogold labeling of Fak in myocardium obtained from 3-minute overloaded hearts. Arrows indicate Fak clusters are found at costameres (D), Z-discs (E), and myofilaments (E). Magnification: A, 31 620x; C, 25 300x.
Fak is rapidly activated and triggers the assembly of a multicomponent signaling complex that has been considered to occupy a central position in the transduction and coordination of the earlier responses of cardiac myocytes to mechanical stress. In the present study, we showed comprehensive data on Fak tyrosine phosphorylation, subcellular distribution, and interaction with C-terminal region of myosin heavy chain, providing insight on the mechanism of its activation by mechanical forces in cardiac myocytes.
By using phosphospecific antibodies against tyrosine residues of Fak and Src, we extended previous demonstration of Fak/Src signaling complex activation by mechanical stress, to show the load-induced phosphorylation of specific Fak tyrosine residues Tyr397, 576/7, 861, and 925 in the rat myocardium. These results agree with a proposed model for Fak activation triggered by autophosphorylation at Tyr397 followed by the engagement of Src, which phosphorylates additional Fak tyrosine residues.29 Consistent with this, we have shown previously that the activation of Fak/Src complex by mechanical stress in cardiac myocytes is strictly dependent on phosphorylation of Fak at Tyr397.20 The phosphorylation of additional tyrosine residues supports the notion that Fak may work as a docking protein providing a scaffold to other signaling molecules related to the activation of downstream pathways involved in multiple cell functions. Phosphorylation of Tyr397 not only starts cooperation with Src, but it also appears to be important for the recruitment of other SH2-containing proteins.30–33 Otherwise, the phosphorylation of Tyr397, as well as Tyr925, creates binding sites for the Grb2-SOS complex.26,34,35 Accordingly, we have previously shown21,23 that pressure overload increases the association of Fak with PI3-kinase and Grb2, which occurs simultaneously to Akt and Erk1/2 activation. Additional studies will be necessary to dissect out the potential role of the activation of these signaling systems to Fak influence on early gene expression in response to mechanical stress in cardiac myocytes.
By using yeast-two hybrid screening of a rat left ventricle cDNA library, pulldown, and coimmunoprecipitation assays, we demonstrated that Fak interacts with a C-terminal region of myosin heavy chain. Furthermore, data from confocal immunofluorescence and immunoelectron microscopy with specific antibodies supports the notion that Fak is localized in the sarcomeric A-band, associated with myosin heavy chain. Comparisons of the amount of Fak immunoprecipitated by excess of anti-heavy chain cardiac myosin with that of anti-Fak antibody allowed us to estimate that 40% of total myocardial Fak is associated with myosin heavy chain. This data raises the question of the localization of the remaining 60% of myocardial Fak. Although we did not perform additional quantitative characterization of Fak distribution in the myocardium of rat left ventricle, it is plausible to assume that this remaining Fak was localized elsewhere in distinct subcellular compartments of cardiac myocytes and other myocardial cell types. Indeed, the immunoelectron microscopy imaging of cardiac myocytes from left ventricles of control rats showed that anti-Fak staining was detected in structures such as costameres and Z discs, although less frequently than in sarcomeric A band. This is the first demonstration of a protein partner of Fak peculiar to muscle cells. The amino acid sequence of C-terminal myosin heavy chain identified as a target to Fak is part of the long C-terminal myosin heavy chain sequence that drives the assembly of two molecules of myosin into a coiled-coil that constitutes the structural backbone of the thick filament. Although Fak/myosin interaction was not previously known, the ability of Fak to directly bind the coiled-coil regions of the p190RhoGEF36 and the GIT1 ArfGAP37 protein were recently demonstrated. In both cases the interaction occurred via a helical bundle structure of the C-terminal Fak FAT domain by means of an unknown motif. However, in the present study, the two-hybrid screening was performed with Fak N-terminal sequence including the FERM and kinase, but not the FAT domain. Although the N-terminal domain of Fak has also been shown to interact with receptor and cytoplasmic tyrosine kinases, it is not clear yet the specific region responsible for this interaction. Further studies will be needed to identify the surface residues of N-terminal Fak required for myosin heavy chain binding and how its C-terminal region functions to promote Fak association.
Fak phosphospecific antibody against Tyr397 could not detect a significant amount of Fak in the immunoprecipitates of anti-heavy chain cardiac myosin antibody, indicating that most Fak associated with this sarcomeric protein is inactive. Conversely, Fak activation by mechanical stress was paralleled by a drastic reduction in the association of Fak with myosin heavy chain, indicating that such interaction is subjected to regulation by mechanical stress. Notably, a low amount of phosphorylated Fak was found in the precipitates of the pulldown assays, even in the extracts of overloaded myocardium, indicating a preferential interaction of GST-C-terminal cardiac myosin to nonphosphorylated Fak. Together, these results raised the interesting possibility of a direct connection between myosin heavy chain and Fak activation by mechanical stress in cardiac myocytes. Incidentally, myosin heavy chain contains small folded segments that make it a truly elastic protein.38 On the other hand, a recent report39 indicates that Fak activation is dependent on the release of an autoinhibitory interaction of the N-terminal FERM and the kinase domains. Thus, it is conceivable that myosin stretch might cause Fak to assume a conformation that favors Tyr397 autophosphorylation. This implies that C-terminal region of myosin heavy chain filaments might work as a true mechanotransducer, transmitting mechanical forces to activate Fak. However, further studies are needed to confirm the assumptions of this model, as well as to exclude the contribution of agonist mediated Fak activation in overloaded myocardium.
We also found that in the early period after the beginning of pressure overload sarcoplasmic Fak-specific immunofluorescence, although reduced, was no longer overlapping the anti-myosin staining, suggesting a relocation of Fak to distinct sarcomeric sites. This was supported by the findings that immunogold staining was frequently seen as aggregates close to Z-discs and costameres. These findings, together with the demonstration that Fak is rapidly phosphorylated by pressure overload, is consistent with the idea that Fak activation by mechanical stress is accompanied by clustering and translocation to diverse subcellular structures of cardiac myocytes. This agrees with data from our previous studies in isolated cardiac myocytes of neonatal rats,20 indicating that controlled stretch induces Fak to aggregate at the myofilaments, Z-discs, and costameres. Moreover, it is interesting to note that Fak phosphorylation at tyrosine residues have been shown to regulate the subcellular localization of Fak.40 The functional significance of Fak clustering is not completely understood at present, but clustering has been considered to optimize Fak signaling because the molecular proximity in clusters may serve to enhance and sustain Fak signaling.41 Thus, Fak cluster and target to Z-discs and costameres, which also serve as sites for mechanical sensing, might account for the amplification of Fak signaling triggered by mechanical stress in cardiac myocytes. This implies that the lack of Z-disc integrity would impair the full activation of Fak by mechanical stress. Therefore, it will be interesting to determine whether Fak plays a role in the pathogenesis of cardiomyopathy associated with the lack of structural integrity of Z discs.
Translocation of activated Fak was further supported by our finding in this study that after 60 minutes of pressure overload, almost 70% of Fak was found in the myocardial nuclear extracts. This was strengthened by the demonstration of a consistent staining with anti-Fak antibody of nuclei of cardiac myocyte isolated from 60-minute overloaded left ventricles. Similar translocation of Fak to nucleus was recently reported in cardiac myocytes of spontaneously hypertensive heart failure rats, suggesting that Fak might play a role in the regulation of nuclear processes in response to mechanical stress.42 However, the implications of such findings remain to be determined.
In conclusion, the demonstration in this study of the subcellular localization of Fak within the sarcomeric A-bands and its interaction with myosin heavy chain have implications for our understanding of how cardiac myocytes sense and transmit mechanical stimuli. Mechanistic insights of these findings may disclose new pathogenetic aspects of mechanical stress in myocardial hypertrophy and failure.
References
Katz AM. The cardiomyopathy of overload: an unnatural growth response in the hypertrophied heart. Ann Intern Med. 1994; 121: 363–371.
Chien KR. Stress pathways and heart failure. Cell. 1999; 98: 555–558.
Sussman MA, McCulloch A, Borg TK. Dance band on the titanic: biomechanical signaling in cardiac hypertrophy. Circ Res. 2002; 91: 888–898.
Epstein ND, Davis JS. Sensing stretch is fundamental. Cell. 2003; 112: 147–150.
Clark KA, McElhinny AS, Beckerle MC, Gregorio CC. Striated muscle cytoarchitecture: an intrincate web of form and function. Ann Rev Cell Dev Biol. 2002; 18: 637–706.
Knoll R, Hoshijima M, Hoffman HM, Person V, Lorenzen-Schmidt I, Bang ML, Hayashi T, Shiga N, Yasukawa H, Schaper W, McKenna W, Yokoyama M, Schork NJ, Omens JH, McCulloch AD, Kimura A, Gregorio CC, Poller W, Schaper J, Schultheiss HP, Chien KR. The cardiac mechanical stretch sensor machinery involves a Z disc complex that is defective in a subset of human dilated cardiomyopathy. Cell. 2002; 111: 943–955.
Gerull B, Gramlich M, Atherton J, McNabb M, Trombitas K, Sasse-Klaassen S, Seidman JG, Seidman C, Granzier H, Labeit S, Frenneaux M, Thierfelder L. Mutations of TTN, encoding the giant muscle filament titin, cause familial dilated cardiomyopathy. Nat Genet. 2002; 30: 201–204.
Arber S, Hunter JJ, Ross J Jr, Hongo M, Sansig G, Borg J, Perriard JC, Chien KR, Caroni P. MLP-deficient mice exhibit a disruption of cardiac cytoarchitectural organization, dilated cardiomyopathy, and heart failure. Cell. 1997; 88: 393–403.
Faulkner G, Lanfranchi G, Valle G. Telethonin and other new proteins of the Z-disc of skeletal muscle. IUBMB Life. 2001; 51: 275–282.
Vatta M, Mohapatra B, Jimenez S, Sanchez X, Faulkner G, Perles Z, Sinagra G, Lin JH, Vu TM, Zhou Q, Bowles KR, Di Lenarda A, Schimmenti L, Fox M, Chrisco MA, Murphy RT, McKenna W, Elliott P, Bowles NE, Chen J, Valle G, Towbin JA. Mutations in Cypher/ZASP in patients with dilated cardiomyopathy and left ventricular non-compaction. J Am Coll Cardiol. 2003; 42: 2014–2027.
Terracio L, Rubin K, Gullberg D, Balog E, Carver W, Jyring R, Borg TK. Expression of collagen binding integrins during cardiac development and hypertrophy. Circ Res. 1991; 68: 734–744.
Borg TK, Goldsmith EC, Price R, Carver W, Terracio L, Samarel AM. Specialization at the Z line of cardiac myocytes. Cardiovasc Res. 2000; 46: 277–285.
Ross RS, Pham C, Shai SY, Goldhaber JI, Fenczik C, Glembotski CC, Ginsberg MH, Loftus JC. ?1 Integrin participates in the hypertrophic response of rat ventricular myocytes. Circ Res. 1998; 82: 1160–1172.
Shai S-Y, Harpf AE, Babbitt CJ, Jordan MC, Fishbein MC, Chen J, Omura M, Leil TA, Becker KD, Jiang M, Smith DJ, Cherry SR, Loftus JC, Ross RS. Cardiac myocyte-specific excision of the 1 integrin gene results in myocardial fibrosis and cardiac failure. Circ Res. 2002; 90: 458–464.
Hanks SK, Calalb MB, Harper MC, Patel SK. Focal adhesion protein-tyrosine kinase phosphorylated in response to cell attachment to fibronectin. Proc Natl Acad Sci U S A. 1992; 89: 8487–8491.
Wang H-B, Dembo M, Hanks SK, Wang Y-L. Focal adhesion kinase is involved in mechanosensing during fibroblast migration. Proc Natl Acad Sci U S A. 2001; 98: 11295–11300.
Li S, Butler P, Wang Y, Hu Y, Han DC, Usami S, Guan J-L, Chien S. The role of the dynamics of focal adhesion kinase in the mechanotaxis of endothelial cells. Proc Natl Acad Sci U S A. 2002; 99: 3546–3551.
Seko Y, Takahashi N, Tobe K, Kadowaki T, Yazaki Y. Pulsatile stretch activates mitogen-activated protein kinase (MAPK) family members and focal adhesion kinase [p125(FAK)] in cultured rat cardiac myocytes. Biochem Biophys Res Commun. 1999; 259: 8–14.
Aikawa R, Nagai T, Kudoh S, Zou Y, Tanaka M, Tamura M, Akazawa H, Takano H, Nagai R, Komuro I. Integrins play a critical role in mechanical stress-induced p38 MAPK activation. Hypertension. 2002; 39: 233–238.
Torsoni AS, Constancio SS, Nadruz, Jr W, Hanks SK, Franchini KG. Focal adhesion kinase is activated and mediates the early hypertrophic response to stretch in cardiac myocytes. Circ Res. 2003; 93: 140–147.
Franchini KG, Torsoni AS, Soares PHA, Saad MJA. Early activation of the multicomponent signaling complex associated with focal adhesion kinase induced by pressure overload in the rat heart. Circ Res. 2000; 87: 558–565.
Laser M, Willey CD, Jiang W, Cooper C, Menick DR, Zile MR, Kuppuswamy D. Integrin activation and focal complex formation in cardiac hypertrophy. J Biol Chem. 2000; 275: 35624–35630.
Domingos PP, Fonseca PM, Nadruz W Jr., Franchini KG. Load-induced focal adhesion kinase activation in the myocardium: role of stretch and contractile activity. Am J Physiol. 2002; 282: H556–H564.
Bayer AL, Heidkamp MC, Patel N, Porter MJ, Engman SJ, Samarel AM. PYK2 expression and phosphorylation increases in pressure overload-induced left ventricular hypertrophy. Am J Physiol. 2002; 283: H695–H706.
Calalb MB, Polte TR, Hanks SK. Tyrosine phosphorylation of focal adhesion kinase at sites in the catalytic domain regulates kinase activity: a role for Src family kinases. Mol Cell Biol. 1995; 15: 954–963.
Schlaepfer DD, Hanks SK, Hunter T, van der Geer P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature. 1994; 372: 786–791.
Polte TR, Hanks SK. Complexes of focal adhesion kinase (FAK) and Crk-associated substrate (p130Cas) are elevated in cytoskeletal-associated fractions following adhesion and Src transformation: requirements for Src kinase activity and FAK proline-rich motifs. J Biol Chem. 1997; 272: 5501–5509.
Schaller MD, Hildebrand JD, Shannon JD, Fox JW, Vines RR, Parsons JT. Autophosphorylation of the focal adhesion kinase, pp125FAK, directs SH2-dependent binding of pp60src. Mol Cell Biol. 1994; 14: 1680–1688.
Parsons JT. Focal adhesion kinase: the first ten years. J Cell Sci. 2003; 116: 1409–1416.
Akagi T, Murata K, Shishido T, Hanafusa H. v-Crk activates the phosphoinositide 3-kinase/AKT pathway by utilizing focal adhesion kinase and H-Ras. Mol Cell Biol. 2002; 22: 7015–7023.
Chen HC, Appeddu PA, Isoda H, Guan JL. Phosphorylation of tyrosine 397 in focal adhesion kinase is required for binding phosphatidylinositol 3-kinase. J Biol Chem. 1996; 271: 26329–26334.
Chen HC, Guan JL. Stimulation of phosphatidylinositol 3'-kinase association with foca adhesion kinase by platelet-derived growth factor. J Biol Chem. 1994; 269: 31229–31233.
Han DC, Guan JL. Association of focal adhesion kinase with Grb7 and its role in cell migration. J Biol Chem. 1999; 274: 24425–24430.
Chen Q, Kinch MS, Lin TH, Burridge K, Juliano RL. Integrin-mediated cell adhesion activates mitogen-activated protein kinases. J Biol Chem. 1994; 269: 26602–26605.
Schlaepfer DD, Hunter T. Evidence for in vivo phosphorylation of the Grb2 SH2-domain binding site on focal adhesion kinase by Src-family protein-tyrosine kinases. Mol Cell Biol. 1996; 16: 5623–5633.
Zhai J, Lin H, Nie Z, Wu J, Canete-Soler R, Schlaepfer WW, Schlaepfer DD. Direct interaction of focal adhesion kinase with p190RhoGEF. J Biol Chem. 2003; 278: 24865–24873.
Zhao ZS, Manser EL, Loo TH, Lim L. Coupling of PAK-Interacting Exchange Factor PIX to GIT1 Promotes Focal Complex Disassembly. Mol Cell Biol. 2000; 20: 6354–6363.
Schwaiger I, Sattler C, Hostetter DR, Rief M. The myosin coiled-coil is a truly elastic protein structure. Nat Mater. 2002; 1: 232–235.
Cooper LA, Shen TL, Guan JL. Cooper Regulation of focal adhesion kinase by its amino-terminal domain through an autoinhibitory interaction. Mol Cell Biol. 2003; 23: 8030–8041.
Katz B-Z, Romer L, Miyamoto S, Volberg T, Matsumoto K, Cukierman E, Geiger B, Yamada KM. Targeting membrane-localized focal adhesion kinase to focal adhesions: roles of tyrosine phosphorylation and SRC family kinases. J Biol Chem. 2003; 278: 29115–29120.
Katz B-Z, Miyamoto S, Teramoto H, Zohar M, Krylov D, Vinson C, Gutkind JS, Yamada KM. Direct transmembrane clustering and cytoplasmic dimerization of focal adhesion kinase initiates its tyrosine phosphorylation. Biochim Biophys Acta. 2002; 1592: 141–152.
Yi XP, Wang X, Gerdes AM, Li F. Subcellular redistribution of focal adhesion kinase and its related nonkinase in hypertrophic myocardium. Hypertension. 2003; 41: 1317–1323.(Priscila M. Fonseca, Rosa)