INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ENDOTHELIN-1 (ET) is a potent vasoconstrictor peptide that has been shown to play a role in various clinical forms of cardiovascular disease (27). In addition to its vasoconstrictor effects, ET also has been shown to have growth-promoting effects on the heart and vasculature. Animal models of cardiac hypertrophy have demonstrated increased ET expression and secretion in the heart, and ET has been shown to be a potent stimulus for neonatal rat ventricular myocyte (NRVM) hypertrophy in vitro (34). Chronic ET stimulation produces increased cell size and protein synthesis and increased transcription of myosin light chain-2 and atrial natriuretic factor (ANF) as well as enhanced sarcomeric assembly (11, 17, 20, 32, 35).

Over the past several years, cultured NRVM have been used to delineate the signaling pathways activated by ET. NRVM have been shown to contain predominantly ET_A receptors (14, 21), which, when activated, stimulate a signaling cascade that involves several downstream serine/threonine protein kinases. ET has also been shown to activate a nonreceptor protein tyrosine kinase, focal adhesion kinase (FAK), in mesangial cells (13) and primary astrocytes (5). FAK is one component of the focal adhesion, a structural site at which integrin receptors tether extracellular matrix proteins to intracellular, cytoskeletal proteins (2, 3, 23). At the time of integrin clustering, FAK is activated and phosphorylates itself at tyrosine residue 397. This phosphorylated tyrosine provides a docking site for the protein tyrosine kinase Src and other Src-family protein tyrosine kinases to bind (28). Propagation of the signal continues as Src then phosphorylates other tyrosine residues within FAK and creates sites for still other protein tyrosine kinases and cytoskeletal proteins to bind. In this way signals are transduced via focal adhesions to downstream targets.

In nonmuscle cells, FAK phosphorylation is necessary for focal adhesion and stress fiber formation and may be important in cell adhesion and spreading. In cardiac myocytes, focal adhesions form part of the costameres, which are bandlike structures linking the Z disk to the sarcolemmal membrane (36). There appears to be a coordinated interplay between focal adhesions/costameres and the assembly and maintenance of sarcomeres. For example, normal organization and alignment of sarcomeres were altered in myocytes treated with anti-₁-integrin antibodies (15). In addition, contractile-arrested myocytes not only disassemble myofibrils but also lose focal adhesion and costamere integrity with the loss of ₁-integrin receptors and vinculin (9, 25, 31). Focal adhesion proteins, therefore, play an intimate role in the primary function of cardiac myocytes. In the present study, we hypothesize that ET signals through FAK activation to increase focal adhesion formation and to promote and/or stabilize sarcomere assembly. Therefore, we examined the role of FAK in ET-induced NRVM hypertrophy and further examined whether FAK is necessary and/or sufficient to induce sarcomeric assembly.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents. PC-1 tissue culture medium was obtained from BioWhittaker (Walkersville, MD). DMEM was obtained from GIBCO-BRL (Grand Island, NY). Medium 199 (M199), Ca²⁺-free and Mg²⁺-free Hanks' balanced salt solution (HBSS), acid-soluble calf skin collagen, and antibiotic/antimycotic solution were obtained from Sigma Chemical (St. Louis, MO). Permanox chamber slides were obtained from Nunc (Naperville, IL). Tissue culture plates were obtained from Costar (Cambridge, MA). Sarcomeric myosin heavy chain (MHC) monoclonal antibody (MF20) was obtained from the Developmental Studies Hybridoma Bank (maintained by the Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, and the Department of Biological Sciences, University of Iowa). This MHC antibody recognizes both - and -MHC, which are both expressed in NRVM. FAK, paxillin, and phosphotyrosine antibodies were obtained from Transduction Laboratories (Lexington, KY), Santa Cruz Biotechnology, (Santa Cruz, CA), or Upstate Biotechnologies (Lake Placid, NY). Protein A and protein G beads were obtained from Calbiochem (San Diego, CA) and Sigma. Horseradish peroxidase-conjugated goat anti-rabbit and goat anti-mouse IgG were from Bio-Rad (Hercules, CA). All other reagents were of the highest grade commercially available and were obtained from Sigma and Baxter S/P (McGaw Park, IL).

NRVM isolation. Animals used in these experiments were handled in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals [DHHS Publication No. (NIH) 85-23, Revised 1985]. NRVM were isolated from the hearts of 1- to 3-day-old Sprague-Dawley rat pups via collagenase digestion as previously described (25). Dissociated cells were preplated for 1 h in serum-free PC-1 medium to selectively remove nonmuscle cells. Myocytes were then plated in PC-1 medium at a density of 400 or 1,600 cells/mm² onto collagen-coated, plastic 35- or 100-mm dishes or onto chamber slides and were left undisturbed in a 5% CO₂ incubator for 14-18 h. Unattached cells were removed by aspiration, and the attached cells were maintained in a solution of DMEM-M199 (4:1) containing antibiotic/antimycotic solution. The medium was changed daily.

Cellular composition. NRVM were quantitatively scraped from the dishes and used for analysis of total protein by the Lowry method with crystalline human serum albumin as standard and for analysis of DNA with Hoechst 33258 dye and salmon sperm DNA as standard, as previously described (25). For quantitative analysis of - and -MHC content, cells were lysed in sample buffer, and then the concentrations of - and -MHC isoenzymes were assessed by SDS-PAGE and silver staining (25). MHC band intensity was quantified by laser densitometry and compared with the band intensity of purified MHC standards (0-300 ng). Results are the means of each treatment group for each cell isolation (expressed as µg MHC/µg DNA).

Immunolocalization. Cells grown on chamber slides were fixed and permeabilized as previously described (9). Myocytes were then stained with antibodies to MHC, paxillin, phosphotyrosine, and FAK/FRNK (FAK-related nonkinase). Of note, the antibody (no. 06-543, UBI) used to detect FRNK was directed to the COOH terminus of FAK. Thus the staining for adenovirally expressed FRNK and endogenous FAK could not be distinguished. Appropriate FITC- or rhodamine-conjugated secondary antibodies were used to visualize the specific proteins. Fluorescently labeled cells were then viewed using a Zeiss model LSM 410 or LSM 510 laser scanning confocal microscope. Multiple optical sections ~1 µm thick were taken of each sample to eliminate out-of-focus fluorescence.

Immunoprecipitation and Western blotting. NRVM were plated at high density (1,600 cells/mm²) and cultured in medium containing the L-type calcium channel blocker nifedipine (10 µM) to inhibit spontaneous contractile activity. After 48 h, fresh nifedipine-free medium was applied for 1 h with or without various inhibitors, and then the myocytes were stimulated with ET (100 nM) or phorbol 12-myristate 13-acetate (PMA; 2 µM). At the indicated times, the myocytes were rinsed with cold PBS and then scraped in an ice-cold modified lysis buffer according to Schlaepfer and Hunter (29). Equal amounts of protein were immunoprecipitated with anti-FAK, -paxillin, or -phosphotyrosine antibodies, followed by the addition of protein A/G beads. Immune complexes were collected and washed three times and then boiled in 8% SDS sample buffer to solubilize the proteins. To quantify ET-induced FAK and paxillin phosphorylation, proteins were separated by 7.5% SDS-PAGE and transferred to nitrocellulose membranes (Hybond, Amersham, Arlington Heights, IL). Blots containing anti-phosphotyrosine immunoprecipitates were probed with anti-FAK or anti-paxillin antibodies. Blots containing FAK or paxillin immunoprecipitates were probed with an anti-phosphotyrosine antibody. Horseradish peroxidase-conjugated goat anti-mouse or anti-rabbit secondary antibodies were visualized by enhanced chemiluminescence (Amersham, Arlington Heights, IL). The bands corresponding to phosphorylated FAK or paxillin were quantified by laser densitometry.

Subcellular fractionation. High-density NRVM cultures were scraped in homogenization buffer (2 mM EDTA, 2 mM EGTA, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 500 µM sodium orthovanadate, 1 mM PefaBloc, and 20 mM Tris · HCl, pH 7.5), sonicated, and centrifuged (100,000 g, 1 h). The supernatant fraction (S) was considered the cytosolic fraction. The pellet was resuspended in 1% Triton X-100-containing buffer, sonicated, and recentrifuged (10,000 g, 10 min). The supernatant fraction from this spin was considered the P₁ fraction, and the pellet from the second spin, the P₂ fraction. The S and P₁ fractions were lyophilized, and then all three fractions were resuspended in lysis buffer. Each fraction was immunoprecipitated with an anti-FAK monoclonal antibody and separated by SDS-PAGE, and the resolved proteins were transferred to nitrocellulose membrane. The blot was then probed with a monoclonal antibody directed to phosphorylated tyrosine residues.

Adenoviral constructs. A replication-defective adenovirus encoding FRNK (Adv-FRNK) was constructed using a 1.2-kb wild-type chick FRNK cDNA kindly provided by Dr. Thomas Parsons (University of Virginia, Charlottesville, VA). Replication-defective adenoviruses encoding wild-type chick FAK (Adv-FAK) and the -galactosidase gene LacZ (Adv-gal) were constructed as previously described (19). Adenoviruses were amplified and purified using HEK-293 cells (10). Preliminary experiments determined that a concentration of 50-100 particles of Adv-FRNK, Adv-FAK, or Adv-gal per cell increased the expression of these proteins by over 40 times in 48 h (Western blot analysis) and infected virtually every myocyte (X-Gal staining) (data not shown).

Data analysis. Results were expressed as means ± SE. Repeated-measures one-way ANOVA or repeated-measures ANOVA on ranks, followed by either Dunnett's test or the Student-Newman-Keuls test, were used for the statistical comparison of multiple groups. Statistical comparison of two groups was accomplished by paired t-test or Wilcoxon rank sum test where appropriate. Differences among means were considered significant at P < 0.05. Data were analyzed using SigmaStat Statistical Software (ver. 1.0, Jandel Scientific, San Rafael, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ET induces cardiomyocyte hypertrophy. ET (100 nM, 48 h) increased levels of total protein-to-DNA, total MHC protein-to-DNA, and both -MHC protein- and -MHC protein-to-DNA ratios as well as the percentage of myosin that was -MHC (Table 1). In contrast, ET had no effect on DNA content, indicating that growth occurred predominately via cell hypertrophy. Using immunofluorescence confocal microscopy, we found that ET increased cell size, MHC staining intensity, and sarcomeric organization compared with control myocytes (Fig. 1). Low-density control cells contained relatively few organized sarcomeres (Fig. 1a), whereas ET-treated myocytes contained many tightly packed myofibrillar arrays throughout the cell (Fig. 1b). Thus, in agreement with previous studies (11, 17, 20, 32, 35), chronic exposure to ET produced morphological and biochemical alterations indicative of cardiomyocyte hypertrophy.

View larger version (87K): [in this window] [in a new window]   Fig. 1.   Effect of endothelin-1 (ET) on localization of myosin heavy chain (MHC), paxillin, tyrosine-phosphorylated proteins, and focal adhesion kinase (FAK). Neonatal rat ventricular myocytes (NRVM) were cultured in serum-free medium alone (a, c, e, and g) or in serum-free medium containing 100 nM ET (b, d, f, and h) for 48 h. Myocytes were then fixed, permeabilized, and stained using antibodies specific for MHC (a and b), the cytoskeletal protein paxillin (c and d), proteins containing phosphorylated tyrosines (e and f), and FAK (g and h). Myocytes were imaged by laser scanning confocal microscopy. Images depicted in a and b were acquired several micrometers above substratum, whereas images in c-h were obtained at cell-substrate interface. Arrows in f indicate regions of increased tyrosinylated proteins at cell periphery.

ET causes focal adhesion formation. We next examined whether ET-induced sarcomeric assembly was accompanied by the formation of focal adhesions. Paxillin is an abundant cytoskeletal protein that has been found in focal adhesions in nonmuscle cells (39). Paxillin localized within control myocytes in distinct punctate areas around the edge of the cell (Fig. 1c). ET-treated myocytes showed a marked increase in the size of the punctate regions and in the distribution of paxillin staining within linear arrays along the sites of cell-substrate contact (Fig. 1d). Using an antibody that recognizes all tyrosine-phosphorylated proteins, we found that focal adhesions contain tyrosine-phosphorylated proteins even under control conditions (Fig. 1e). However, with chronic ET stimulation, tyrosine-phosphorylated proteins increased, particularly along the cell periphery (Fig. 1f, arrows). Finally, we examined whether FAK localized in a similar distribution to paxillin in control and ET-treated myocytes. As shown in Fig. 1g, FAK was readily detected in focal adhesion structures as well as in a more diffuse staining pattern along the cell-substrate interface of control cells. Like paxillin, ET increased FAK staining along the cell periphery as well as within linear arrays along the sites of cell-substrate contact (Fig. 1h). Thus focal adhesions and specific proteins located within focal adhesions are increased in ET-treated NRVM.

ET acutely induces FAK tyrosine phosphorylation. As shown in Fig. 2, FAK phosphorylation increased within 2 min of ET stimulation and returned toward control levels by 60 min. With more chronic exposure, FAK phosphorylation at 24 and 48 h was no different than that in untreated control cells (data not shown). ET-induced FAK tyrosine phosphorylation was completely inhibited by addition to the medium of either a nonspecific (ET_A + ET_B) or an ET_A-specific receptor antagonist (PD-161721 or PD-156707, respectively; 100 nM, 1-h preincubation), indicating that ET-induced FAK phosphorylation occurred via ET_A receptor-dependent signaling (data not shown).

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Effect of ET on FAK phosphorylation. NRVM were stimulated with ET for indicated times (in min). Tyrosine-phosphorylated proteins (pTyr) were immunoprecipitated, size fractionated by SDS-PAGE, and transferred to nitrocellulose membrane. Blots were probed with an anti-FAK polyclonal antibody, and protein bands were visualized using enhanced chemiluminescence (ECL). Cell lysate from 3T3 murine fibroblasts served as a positive control for FAK. A: ET induced FAK tyrosine phosphorylation within 2 min. FAK activation then returned toward baseline by 1 h. Molecular mass is shown at left in kDa. STD, standard; IP, immunoprecipitate; WB, Western blot; Ab, antibody. B: quantitative analysis of blots from 7-17 different experiments. Results are expressed as means ± SE for each time point.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Subcellular fractionation and detection of phosphorylated FAK in response to ET. NRVM were stimulated with ET for indicated times. Myocytes were lysed in absence of detergents and divided into 3 fractions: soluble (S), particulate 1 (P1), and particulate 2 (P2). Each fraction was immunoprecipitated with an anti-FAK monoclonal antibody, separated using SDS-PAGE, and transferred to nitrocellulose membrane. Blot was probed with a monoclonal antibody directed to phosphorylated tyrosine residues. ET caused FAK tyrosine phosphorylation predominantly in P2 fraction, which corresponds to myocyte cytoskeleton. Consistent with data shown in Fig. 2, FAK phosphorylation is rapidly induced by ET and then returns toward baseline by 1 h. Molecular mass is shown at left in kDa.

Signaling pathways involved in the regulation of ET-induced FAK phosphorylation. Many aspects of cardiac myocyte growth and function are dependent on intracellular calcium concentration ([Ca²⁺]_i), so we first examined the dependence of ET-induced FAK phosphorylation on the influx of calcium through L-type voltage-sensitive calcium channels. As was the case with phenylephrine (10), ET increased [Ca²⁺]_i transients in spontaneously contracting NRVM, which were virtually abolished in the presence of the specific calcium channel blocker nifedipine (data not shown). However, basal FAK phosphorylation was not affected by nifedipine (1.3 ± 0.3-fold change in nifedipine-treated vs. untreated control myocytes; n = 7 experiments). More importantly, nifedipine had no effect on ET-induced FAK phosphorylation (1.9 ± 0.2- and 2.4 ± 0.4-fold increase in ET-induced FAK phosphorylation in untreated and nifedipine-treated myocytes, respectively; n = 7 experiments; P < 0.05 for both comparisons).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effect of protein kinase C (PKC) inhibitors on ET-induced FAK phosphorylation. NRVM were cultured in serum-free medium alone or in medium containing the PKC inhibitor chelerythrine or the PKC- isoenzyme-specific inhibitor rottlerin for 1 h (basal) and then stimulated with ET (100 nM, 5 min). Tyrosine-phosphorylated proteins were immunoprecipitated, size-fractionated by SDS-PAGE, and transferred to nitrocellulose membrane. Blots were probed with an anti-FAK polyclonal antibody, and protein bands were visualized using ECL. ET-induced FAK phosphorylation was inhibited by PKC-specific inhibitors, suggesting that FAK activation is downstream of PKC activation. Data are means ± SE for 4-17 different experiments. *P < 0.05 vs. basal control levels of FAK phosphorylation.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   Effect of ET and phorbol 12-myristate 13-acetate (PMA) on FAK phosphorylation. NRVM were cultured in serum-free medium alone for 1 h and then stimulated with ET (100 nM) or PMA (2 µM) for 5 min. A: cell lysates were immunoprecipitated with an anti-FAK antibody, and resulting Western blot was probed with an anti-phosphotyrosine antibody. Similarly, cell lysates were immunoprecipitated with an anti-phosphotyrosine antibody, and resulting Western blot was probed with an anti-FAK antibody. In both cases, ET and the phorbol ester PMA induced FAK tyrosine phosphorylation compared with control cells. A cell lysate from 3T3 murine fibroblasts served as a positive control for phosphorylated FAK. Molecular mass is shown at left in kDa. B: histogram representing quantitative data obtained by immunoprecipitation with an anti-phosphotyrosine antibody and Western blotting with an anti-FAK antibody. Data are means ± SE for 4 different experiments. *P < 0.05 vs. control.

Paxillin, an FAK substrate, is phosphorylated in response to ET treatment. As shown in Fig. 6, control NRVM contained rather high levels of phosphorylated paxillin. However, paxillin phosphorylation further increased within 2-5 min of ET stimulation and then slowly returned toward baseline by 1 h. After 24 and 48 h of continuous ET stimulation, paxillin phosphorylation was no different than in untreated control cells (data not shown). This time course was coincident with ET-induced FAK phosphorylation, suggesting that FAK functions in a signal transduction pathway leading to paxillin phosphorylation. Thus the increased paxillin staining in response to chronic ET treatment shown in Fig. 1 is associated with an acute increase in tyrosine-phosphorylated paxillin.

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Effect of ET on paxillin phosphorylation. NRVM were cultured in serum-free medium alone (1 h) and then stimulated with ET (100 nM) for indicated times (in min). Tyrosine-phosphorylated proteins were immunoprecipitated, size-fractionated by SDS-PAGE, and transferred to nitrocellulose membrane. Western blots were probed with paxillin antibody, and protein bands were visualized using ECL. Cell lysate from A431 cells served as a positive control for paxillin. A: paxillin phosphorylation increased within 2 min of exposure to ET and then returned toward baseline within 1 h. Molecular mass is shown at left in kDa. B: quantitative analysis of blots from 4 different experiments. Data are means ± SE for each time point.

Is FAK necessary for ET-induced neonatal rat ventricular myocyte hypertrophy? To further examine the role of FAK in eliciting specific aspects of the hypertrophic phenotype induced by ET, we created an Adv containing the COOH-terminal region of FAK, known as FRNK. FRNK has been shown in other cell types to compete with FAK for binding to focal adhesions and thus inhibit FAK-dependent signaling (22). Adv-FRNK expression increased to a maximum 48 h after infection (Fig. 7A) and inhibited ET-induced FAK phosphorylation compared with uninfected or Adv-gal-infected myocytes (Fig. 7B). In other experiments, we found that Adv-FRNK also inhibited ET-induced paxillin phosphorylation compared with uninfected myocytes (0.9 ± 0.1- vs. 2.4 ± 0.2-fold increase in ET-induced paxillin phosphorylation in FRNK-infected and uninfected NRVM, respectively; n = 4 experiments; P < 0.05). Thus Adv-FRNK appears to act as a "dominant-negative" inhibitor of FAK-dependent signaling.

View larger version (63K): [in this window] [in a new window]   Fig. 7.   Overexpression of FAK-related nonkinase (FRNK) inhibits FAK phosphorylation. Molecular mass is shown at left in kDa. A: low-density NRVM were infected with FRNK adenovirus (Adv-FRNK, 50 particles/cell) for 1 h in serum-free medium and then changed into fresh medium. Myocytes were lysed in 8% sample buffer at indicated times after infection. Equal amounts of cell protein (50 µg) were separated by SDS-PAGE and immunoblotted for FRNK using an antibody directed to COOH-terminal region of FAK. There was no apparent endogenous FRNK expression in uninfected control NRVM. Expression of FRNK was detected at 25 h after infection and was markedly increased by 48 h. B: in a separate experiment, NRVM remained uninfected or were infected for 1 h with adenovirus encoding -galactosidase (Adv-gal) or FRNK (Adv-FRNK) (100 particles/cell). After 48 h, myocytes were stimulated with ET (100 nM) for 5 min and lysed. Tyrosine-phosphorylated proteins were immunoprecipitated, size-fractionated by SDS-PAGE, and transferred to nitrocellulose membrane. Blots were probed with an anti-FAK polyclonal antibody, and protein bands were visualized using ECL. Cell lysate from 3T3 murine fibroblasts served as a positive control for FAK. ET treatment increased FAK phosphorylation in control and Adv-gal-infected myocytes. However, ET-induced FAK phosphorylation was inhibited in Adv-FRNK-infected cells.

View larger version (49K): [in this window] [in a new window]   Fig. 8.   Overexpression of FRNK inhibits ET-induced sarcomeric assembly. Low-density NRVM were infected (1 h) with 100 particles/cell of Adv-gal (A-D) or Adv-FRNK (E-H). Myocytes were fixed and permeabilized after an additional 23 h of culture in serum-free medium alone (A, B, E, and F) or in medium containing 100 nM ET (C, D, G, and H). Myocytes were double stained using antibodies for FAK/FRNK (A, C, E, and G) or MHC (B, D, F, H). The antibody used to detect FRNK was an antibody directed to the COOH terminus of FAK. Thus staining for adenovirally expressed FRNK and endogenous FAK could not be distinguished. Myocytes were imaged by laser scanning confocal microscopy. Images were acquired several micrometers above substratum.

Is FAK sufficient to cause NRVM hypertrophy? To determine whether FAK could induce NRVM hypertrophy, we overexpressed wild-type chick FAK in NRVM by using Adv-FAK (19). A concentration of 50-100 particles/cell was found to be sufficient to produce marked overexpression within the myocytes (data not shown). Using this concentration of adenovirus, a time course of expression was established. As shown in Fig. 9, A and B, FAK overexpression occurred within 10 h of infection and reached a sustained maximum at 24 h. Compared with uninfected myocytes or myocytes infected with Adv-gal, Adv-FAK-infected myocytes contained approximately four- to fivefold higher levels of phosphorylated FAK (Fig. 9C). Thus myocytes infected with Adv-FAK expressed activated FAK by 24 h.

View larger version (49K): [in this window] [in a new window]   Fig. 9.   Expression of FAK in myocytes infected with an FAK adenovirus (Adv-FAK). A: low-density NRVM were infected with Adv-FAK (50 particles/cell; 1 h) in serum-free medium and then changed into fresh medium. Myocytes were lysed in 8% sample buffer at indicated times after infection. Immunoblots were probed with an anti-FAK polyclonal antibody, and protein bands were visualized using ECL. B: quantitative analysis of FAK expression over time shows a rapid increase in FAK 10-24 h after infection. FAK expression was maximal at 24 h and was sustained at 48 h. C: in a separate set of experiments, NRVM remained uninfected or were infected for 1 h with 50 particles/cell of Adv-gal or Adv-FAK. After 48 h, myocytes were lysed and tyrosine-phosphorylated proteins were immunoprecipitated, size-fractionated by SDS-PAGE, and transferred to nitrocellulose membrane. Blots were probed with an anti-FAK polyclonal antibody, and protein bands were visualized using ECL. Cell lysate from 3T3 murine fibroblasts served as a positive control for FAK. Adv-FAK infection markedly increased FAK phosphorylation compared with uninfected myocytes or myocytes infected with Adv-gal. Molecular mass is shown at left in kDa (A and C).

View larger version (71K): [in this window] [in a new window]   Fig. 10.   Overexpression of FAK does not promote sarcomeric assembly. Low-density NRVM were left uninfected (A and B) or were infected (1 h) with 100 particles/cell of Adv-gal (C and D) or Adv-FAK (E and F ). After an additional 23 h of culture in serum-free medium, myocytes were fixed and permeabilized. Myocytes were double stained using antibodies specific for FAK (A, C, and E) or MHC (B, D, and F). Myocytes were imaged by laser scanning confocal microscopy. All images were acquired several micrometers above substratum.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cardiac myocytes rely on focal adhesions and costameres for the assembly and maintenance of sarcomeres (15). Our laboratory recently demonstrated (9, 31) that spontaneously contracting NRVM contain many focal adhesions and focal adhesion/costamere components such as FAK, vinculin, and ₁-integrin receptors. Contractile-arrested myocytes treated with L-type calcium channel blockers not only disassemble myofibrils but also lose focal adhesion and costamere integrity with the loss of vinculin, ₁-integrin receptors, and FAK from focal contact sites (9, 25, 31, 33). Restoration of contractile activity caused reassembly of focal adhesions and costameres that temporally preceded sarcomeric assembly. Thus focal adhesion and costamere formation play an intimate role in maintaining contractile function of cardiac myocytes.

In the present study, we found that FAK is present in a Triton X-100-insoluble fraction of the cell corresponding to the cytoskeleton and is rapidly activated by ET. FAK has been shown to associate with the cytoskeleton on fibronectin stimulation of NIH/3T3 fibroblasts (30). The present results using NRVM demonstrate that FAK is already present within a detergent-insoluble fraction, where it is rapidly phosphorylated on ET stimulation. However, it should be pointed out that our fractionation scheme does not distinguish whether FAK is associated only at focal adhesions or whether it is also associated with other cytoskeletal elements. Furthermore, previous studies have shown that phenylephrine, ET, and ANG II all induce sarcomeric assembly within 0.5-4 h following agonist stimulation (1, 8). Thus the rapid (2 min) ET-induced FAK phosphorylation observed in our studies must precede sarcomeric assembly and may be a necessary initial component in focal adhesion formation.

Cardiac myocytes are highly dependent on the influx of calcium not only for contractile activity but also for the assembly and maintenance of sarcomeres (4). However, in the present study, basal and ET-induced FAK phosphorylation were found to be independent of calcium influx through L-type voltage-gated calcium channels. It is possible that a small amount of calcium released from inositol 1,4,5-trisphosphate-sensitive stores was sufficient for FAK phosphorylation. Nonetheless, because calcium influx is required for sarcomeric assembly, there must be additional calcium-sensitive events acting downstream of, or parallel to, FAK activation. We have also identified PKCs as potential upstream regulators of FAK in NRVM. Not only did PMA activate FAK but the PKC inhibitor chelerythrine completely blocked ET-induced FAK phosphorylation. In addition, our results with the PKC- isoenzyme inhibitor rottlerin (12) suggest that the novel calcium-independent PKCs are in some way necessary for this response.

Using a replication-defective FRNK adenovirus, we have shown that FAK signaling is necessary for ET-induced NRVM hypertrophy. In some cell types FRNK is endogenously expressed at low levels as a separate 41- to 43-kDa protein (26). FRNK overexpression using a replication-competent avian retrovirus has been shown to inhibit FAK tyrosine phosphorylation and to delay chick embryo fibroblast spreading on fibronectin (22). In addition, FRNK overexpression decreased tensin and paxillin phosphorylation (22). In the present report, ET-induced FAK and paxillin phosphorylation was inhibited in NRVM infected with Adv-FRNK, and no hypertrophy or sarcomeric assembly was observed after chronic ET stimulation. Nevertheless, additional studies are needed to characterize precisely which intermediate signaling pathways were affected by FRNK overexpression and which are required for eliciting specific aspects of the hypertrophic phenotype.

Unlike the Adv-FRNK experiments, the Adv-FAK experiments described in Fig. 10 and Table 3 are more difficult to interpret. Instead of inducing NRVM hypertrophy as we hypothesized, FAK overexpression alone did not promote sarcomeric assembly, had no effect on the total protein-to-DNA ratio, and actually reduced -MHC protein compared with uninfected or Adv-gal-infected myocytes. These changes occurred despite the fact that FAK overexpression resulted in an increase in the amount of phosphorylated FAK that was comparable to that observed in ET-stimulated cells. Clearly, FAK overexpression alone was not sufficient, in the absence of other trophic stimuli, to produce NRVM hypertrophy. However, the mechanisms responsible for these unexpected results are unclear. One potential explanation is that we used wild-type FAK rather than a constitutively active form of FAK (6). Wild-type FAK overexpression resulted in a four- to fivefold increase in active FAK (Fig. 9C) but a 40-fold increase in total FAK (Fig. 9B), which was distributed not only in focal adhesions but throughout the myocyte cytoplasm (Fig. 10e). It is conceivable that nonphosphorylated FAK actually displaced endogenous, active FAK from focal adhesions and thus interfered with basal FAK-dependent signaling. Furthermore, in a previous study we showed that transient transfection of high-density, spontaneously contracting NRVM with an expression vector encoding wild-type FAK along with rat -MHC and ANF promoter-luciferase constructs resulted in a two- to fourfold increase in luciferase activities (9). This was a substantial increase above the already high levels of ANF and -MHC promoter activities observed in these high-density cultures. Thus FAK overexpression under these conditions could indeed stimulate the transcription of fetal genes associated with the hypertrophic phenotype. Interestingly, FAK overexpression did not significantly transactivate ANF or -MHC promoter activity in verapamil-arrested cells, indicating that wild-type FAK overexpression alone was not sufficient to stimulate ANF or -MHC transcription in the absence of [Ca²⁺]_i transients or other trophic signals. The Adv-FAK data described in the present report provide important, complementary information. FAK overexpression in low-density, noncontracting NRVM was also insufficient to induce total protein or MHC accumulation or to induce sarcomeric assembly. Determining which signaling molecules are necessary and/or sufficient to induce NRVM hypertrophy has been the subject of several recent studies (1, 16, 24, 37, 38, 40). Clearly, other signaling components that are stimulated by ET_A-receptor activation in addition to FAK must be necessary to produce sarcomeric assembly and NRVM hypertrophy.