Angiotensin II and Other Hypertrophic Stimuli Mediated by G Protein–Coupled Receptors Activate Tyrosine Kinase, Mitogen-Activated Protein Kinase, and 90-kD S6 Kinase in Cardiac Myocytes

Many growth stimuli cause tyrosine phosphorylation of various intracellular substrates through receptors that are either directly or indirectly coupled to tyrosine kinases. Tyrosine phosphorylation of the intracellular substrates plays an essential role not only in cellular mitogenic responses but also in many other functions, including differentiation of neuronal, endocrine, and lymphoid cells.¹ In terminally differentiated cardiac muscle, myocytes are known to respond to mechanical load and growth factors by increasing their mass (hypertrophy). This process consists of an increase in protein synthesis and alteration of muscle phenotype but is not accompanied by DNA synthesis or cell division.²³ Although tyrosine phosphorylation has been shown to play an essential role in the mitogenic response, it is not known whether it is also important for the hypertrophic growth response in terminally differentiated cells.

Recently, several protein serine/threonine kinases that may participate in regulating cell growth have been identified. Among these are the mitogen-activated protein (MAP) kinase family. MAP kinases are rapidly activated in response to stimulation of various receptors, including growth factor/tyrosine kinase receptors (such as insulin receptor) and G protein–coupled receptors (such as the thrombin receptor). Importantly, activity of MAP kinase is critically regulated by both tyrosine and threonine phosphorylation.⁴⁵ It has been recently shown that MAP kinase kinases (MEKs) phosphorylate MAP kinase on both tyrosine and threonine residues; this phosphorylation results in activation of MAP kinase.⁶ It has also been shown that raf-1 and MEK kinase phosphorylate and activate MEK.⁷⁸ MAP kinase phosphorylates and activates 90-kD ribosomal S6 protein kinase (RSK), a member of the S6 kinase family.⁵⁹ Interestingly, recent studies suggest that MAP kinase and RSK can phosphorylate nuclear transcriptional factors, such as JUN, MYC, p62^TCF, FOS, and the serum response factor, and regulate expression of downstream genes.⁵¹⁰¹¹ Thus, the protein kinase cascade can potentially regulate a wide variety of cellular phenomena not restricted to mitogenesis. Therefore, it is of interest whether hypertrophic stimuli use this protein kinase cascade, which critically depends on tyrosine phosphorylation.

Many vasoactive substances, including angiotensin II (Ang II), norepinephrine (NE), phenylephrine (Phe), and endothelin 1 (ET-1), cause hypertrophic responses in cardiac myocytes.¹²¹³¹⁴ These humoral factors are known to activate several second-messenger systems through heterotrimeric guanine nucleotide–binding protein (G protein)–coupled receptors. It has been recently shown that Ang II causes tyrosine phosphorylation and activates MAP kinases in vascular smooth muscle cells.¹⁵¹⁶¹⁷ More recently, Bogoyevitch and coworkers¹⁸¹⁹ reported that Phe, ET-1, and phorbol ester activate MAP kinases in cultured ventricular cardiac myocytes. These data suggest that tyrosine kinases or MAP kinases can be activated in cellular growth responses initiated by agonists for G protein–coupled receptors.

Several studies have suggested that Ang II may be a critical factor in mediating cardiac hypertrophy in vivo.²⁰ We and others have reported that Ang II causes a hypertrophic response in neonatal rat cardiac myocytes and a mitogenic response in cardiac fibroblasts in vitro.¹²²¹ We have recently demonstrated that mechanical stretch causes a secretion of Ang II from cardiac myocytes and that Ang II acts as an initial mediator of stretch-induced hypertrophy through activation of AT₁ receptors in neonatal rat cardiac myocytes.²² cDNA cloning of the AT₁ receptor has indicated that the receptor putatively has seven-membrane-spanning regions, a typical feature of a G protein–coupled receptor, but has no apparent tyrosine kinase domain.²³ Interestingly, we²⁴ and others²⁵ have shown that mechanical stretch of neonatal rat ventricular cardiac myocytes causes tyrosine phosphorylation of several proteins²⁴ and activates MAP kinases and RSK.²⁴²⁵ However, it remains to be determined whether Ang II activates tyrosine kinases or other tyrosine phosphorylation–dependent signal transduction pathways in cardiac myocytes and nonmyocytes. Moreover, it has not been clear, in any other systems, how stimulation of G protein–coupled receptors leads to activation of tyrosine phosphorylation and tyrosine phosphorylation–dependent signaling. Therefore, the experiments described in the present study were conducted (1) to examine whether Ang II and other hypertrophic stimuli activate tyrosine kinases in cardiac cells, (2) to examine whether Ang II and other hypertrophic stimuli activate MAP kinases and RSK, and (3) to examine the mechanism of how Ang II activates MAP kinase and RSK. We demonstrate that Ang II and other hypertrophic stimuli cause tyrosine phosphorylation of several intracellular proteins and activate MAP kinases and RSK in both cardiac myocytes and nonmyocytes. Interestingly, a Ca²⁺-dependent signaling mechanism seems to play an essential role in this Ang II–induced activation of the protein kinase cascade.

Materials and Methods

Materials

All culture reagents were purchased from GIBCO. Glass coverslips were from Fisher Scientific. [γ-³²P]ATP (10 and 6000 Ci/mmol) was obtained from Du-Pont-New England Nuclear. ¹²⁵I–protein A (30 mCi/mg) was from ICN; gel electrophoresis reagents were from Bio-Rad; nitrocellulose membrane was from Schleicher & Schuell; protein A–Sepharose CL-4B was from Pharmacia Biotech; pertussis toxin, pansorbin, and BAPTA-AM were from Calbiochem; H-7, calphostin C, and chelerythrine were from LC Services: okadaic acid was from Moana BioProducts; S6 peptide (RRLSSLRA) was from UBI; phosphocellulose paper (2.5 cm) was from Whatman; and Ang II was from Peninsula. Losartan and PD123319 were gifts from Du Pont–Merck and Parke-Davis, respectively. All other chemicals were purchased from Sigma Chemical Co. Stock solutions of the chemicals were prepared freshly just before experiments as 100- to 1000-fold concentrated solutions. Stock solutions of calphostin C, phorbol 12-myristate 13-acetate (PMA), A23187, and BAPTA-AM were made by dissolving these compounds in dimethyl sulfoxide (DMSO). The final concentration of DMSO was <0.1%, which did not affect basal tyrosine phosphorylation and activities of MAP kinases and RSK. Monoclonal anti-phosphotyrosine antibody (4G10)²⁶ was a gift from Dr T. Roberts (Harvard Medical School, Boston, Mass). Recombinant anti-phosphotyrosine antibody (RC20H)²⁷ was purchased from Transduction Laboratories. Anti-MAP kinase antibody (α-cMAPK)¹⁰ and anti-RSK antibody (αRSK)²⁸ were gifts from Dr J. Blenis (Harvard Medical School). Other anti-MAP kinase polyclonal antibodies (anti–ERK-1 [K-23] and anti–ERK-2 [C-14], where ERK indicates extracellular signal–related kinase-1) were purchased from Santa Cruz Biotechnology, and an anti-rat MAP kinase polyclonal antibody (erk1-CT) was from UBI. Normal rabbit serum (nonimmune serum) was from Jackson Immuno Research.

Cell Culture and Immunofluorescent Staining

Primary cultures of the neonatal rat cardiac myocytes were prepared as previously described.²⁹ After an enzymatic dissociation, the cells were preplated for 1 hour to selectively enrich for cardiac myocytes. The resultant suspension of myocytes was plated onto gelatin-coated 60-mm tissue culture dishes at a density of 1.35×10⁵ cells per cm² and cultured in Dulbecco’s modified Eagle medium/F-12 (GIBCO) (1:1 [vol/vol]) supplemented with 5% horse serum, 2 g/L bovine serum albumin (fraction V), 4 mmol/L glucose, 3 mmol/L pyruvic acid, 15 mmol/L HEPES (pH 7.6), 100 μmol/L ascorbic acid, 100 μg/mL ampicillin, and 100 μmol/L bromodeoxyuridine. The culture medium was changed 24 to 36 hours after seeding to a defined serum-free Dulbecco’s modified Eagle medium/F-12, which had the same composition as described above, except that 5% horse serum and bromodeoxyuridine were not added. Using this method, we routinely obtained myocyte cultures with 90% to 95% myocytes, as assessed by immunofluorescence staining with a monoclonal antibody against sarcomeric myosin heavy chain (MF20).³⁰ For immunofluorescence studies, cardiac myocytes were grown on uncoated glass coverslips, and the indirect immunofluorescent staining was performed as described previously.²⁹ Immunofluorescence studies of MAP kinases were performed as previously described¹⁰ by using 10 μg/mL of an anti–MAP kinase antibody (erk1-CT). Cardiac nonmyocyte (fibroblast) culture was prepared by two passages of the cells adherent to the culture dish during the preplating procedure.²⁹ All experiments were done in serum-free medium 48 hours after changing to the serum-free medium.

Immunoblotting

Cell stimulation was terminated by a rapid aspiration of the medium and addition of 120 μL of ice-cold buffer A containing 25 mmol/L Tris-HCl (pH 7.4), 25 mmol/L NaCl, 1 mmol/L sodium orthovanadate, 10 mmol/L sodium fluoride, 10 mmol/L sodium pyrophosphate, 10 nmol/L okadaic acid, 0.5 mmol/L EGTA, 1 mmol/L phenylmethysulfonyl fluoride (PMSF), 0.8 μg/mL leupeptin, 10 μg/mL aprotinin, and 10 mg/mL p-nitrophenylphosphate. Cell lysates were incubated on ice for 20 minutes and centrifuged for 20 minutes at 4°C. In some experiments, 1% Triton X-100, 1% deoxycholic acid, and 0.1% sodium dodecyl sulfate (SDS) were added to buffer A (modified buffer A). The lysate from one 60-mm dish (3×10⁶ cells) contained 485±28 μg (mean of four randomly picked samples) of total protein. After addition of 5× Laemmli’s SDS–polyacrylamide gel electrophoresis (PAGE) sample buffer, the lysates were boiled for 5 minutes. Lysates containing equal amounts of protein (200 μg) or immunoprecipitates (see below) were electrophoresed on an 8% or 10% polyacrylamide gel and transferred to nitrocellulose membranes. Membranes were blocked by using 5% bovine serum albumin in TBST (20 mmol/L Tris [pH 8.0], 150 mmol/L NaCl, and 0.05% Tween 20) for 2 hours and were then incubated with 4G10 at a concentration of 2 μg/mL,²⁶ RC20H (Transduction Laboratory) at a dilution of 1:2500,²⁷ or polyclonal antisera raised against C-terminal MAP kinase peptide (α-cMAPK) at a dilution of 1:200¹⁰ in TBST. Immunoreactive bands were probed with ¹²⁵I–protein A (30 mCi/mg). In experiments using RC20H, an enhanced chemiluminescence system (Amersham) was used as a detection method, which gave almost identical results as the 4G10 and ¹²⁵I–protein A system. Blots were washed and subjected to autoradiography. Molecular weights of proteins were estimated by using prestained markers (Bio-Rad, 161-0324). Each lane presented in a single panel of the gel picture was from the same gel and the same exposure of the autoradiogram, although in some cases lanes were cut for the final figure production. In some blots, Ponceau S staining was performed according to the manufacturer’s instruction (Sigma, P-7767) to confirm that the equal amount of protein was loaded in each lane.

Immune Complex Tyrosine Kinase Assay

Tyrosine kinase assays were performed as previously described.³¹ Cell-free lysates were prepared as above by using buffer A. Lysates containing equal amounts of protein (100 μg) were incubated with 10 vol RIPA buffer (10 mmol/L Tris [pH 7.2], 150 mmol/L NaCl, 1% Triton X-100, 1% deoxycholic acid, and 0.1 % SDS) and 10 μg of 4G10 for 3 hours at 4°C. Protein A–Sepharose CL-4B was then added, and immunoprecipitates were washed once with phosphate-buffered saline, twice with 20 mmol/L Tris (pH 7.8) and 0.5 mol/L LiCl, and once with 50 mmol/L Tris (pH 7.4) and 10 mmol/L MnCl₂. Ten microliters of immunoprecipitate was incubated with 160 μg of an acid-insoluble synthetic tyrosine kinase substrate [EY(4:1), a polymer of glutamate (E) and tyrosine (Y) with a ratio of 4:1; Sigma P0275] and 40 μL of 2× kinase buffer containing 100 mmol/L HEPES (pH 7.6), 60 mmol/L MgCl₂, 2 mmol/L MnCl₂, 0.2 mmol/L sodium orthovanadate, and 0.2% Nonidet P-40. After addition of 5 μCi of [γ-³²P]ATP (6000 Ci/mmol) and 5 μmol/L of cold ATP, samples (80 μL per sample) were incubated at 25°C for 5 minutes. Reactions were terminated by spotting the samples onto Whatman 3MM papers. The papers were immediately washed with 5% trichloroacetic acid (wt/vol) containing 1% sodium pyrophosphate (wt/vol) four times (15 minutes each). The papers were placed into scintillation vials, and the radioactivity was counted. The rates of tyrosine phosphorylation of EY(4:1) were linear for at least 20 minutes in this assay condition. Control immunoprecipitates were prepared by using the anti-phosphotyrosine antibody preabsorbed by phosphotyrosine (50 mmol/L). The ³²P count obtained without EY(4:1) was subtracted from that with EY(4:1) to eliminate background.

MAP Kinase Assays in Myelin Basic Protein–Containing Polyacrylamide Gels

Cell-free lysates were prepared in 60 μL of buffer A, and immune complex was prepared as previously described.¹⁰ Briefly, lysates containing equal amounts of protein (300 μg) were adjusted to 1% SDS, denatured by heating to 85°C to 90°C for 5 minutes, and diluted with 10 vol of RIPA buffer. The lysates were incubated with 2 μL of α-cMAPK,¹⁰ 1 μg of K-23, or 1 μg of C-14 (Santa Cruz Biotechnology) overnight at 4°C. Pansorbin or protein A–Sepharose CL-4B was then added, and the immunoprecipitates were washed twice with buffer A. The immunoprecipitates (immune complex assays) or whole-cell lysates (direct assays) were electrophoresed on a 10% SDS-polyacrylamide gel containing 0.5 mg/mL myelin basic protein (MBP, Sigma M2016). Kinase assays in MBP-containing polyacrylamide gel were performed at room temperature as previously described.³² Briefly, after electrophoresis, SDS was removed by washing the gel with two changes of 20% 2-propanol in 50 mmol/L Tris (pH 8.0) for 1 hour and then with two changes of 50 mmol/L Tris (pH 8.0) containing 5 mmol/L 2-mercaptoethanol for 1 hour. The enzyme was denatured by incubating the gel with two changes of 6 mol/L guanidine-HCl for 1 hour and then renatured with five changes of 50 mmol/L Tris (pH 8.0) containing 0.04% Tween 40 and 5 mmol/L 2-mercaptoethanol for 1 hour (5 times for 12 minutes each). The gel was then incubated with 40 mmol/L HEPES (pH 8.0) containing 2 mmol/L dithiothreitol and 10 mmol/L MgCl₂. The kinase reaction was performed in conditions inhibitory to cyclic nucleotide–dependent protein kinases and Ca²⁺-dependent protein kinases by incubating the gel at 25°C for 1 hour with 40 mmol/L HEPES (pH 8.0) containing 0.5 mmol/L EGTA, 10 mmol/L MgCl₂, 2 μmol/L protein kinase inhibitor peptide (rabbit sequence, Sigma P0300), 40 μmol/L ATP, and 2.5 μCi/mL of [γ-³²P]ATP (6000 Ci/mmol). After incubation, the gel was washed with a 5% (wt/vol) trichloroacetic acid solution containing 1% (wt/vol) sodium pyrophosphate until the radioactivity of the solution became negligible. The gel was dried and then subjected to autoradiography.

RSK Assay

RSK activity was measured by an immune complex kinase assay using S6 peptide (RRLSSLRA) as a substrate as previously described.²⁸³³ The cell-free lysates were prepared as those for MAP kinase assay except that buffer B was used. Buffer B contained 10 mmol/L KPO₄ (pH 7.4), 1 mmol/L EDTA, 5 mmol/L EGTA, 10 mmol/L MgCl₂, 50 mmol/L β-glycerophosphate, 1 mmol/L sodium orthovanadate, 2 mmol/L dithiothreitol, 40 μg/mL PMSF, 10 nmol/L okadaic acid, 0.8 μg/mL leupeptin, 10 mg/mL p-nitrophenylphosphate, and 10 μg/mL aprotinin. The lysate from one 60-mm dish (3×10⁶ cells) contained 490±23 μg (mean of four randomly picked samples) of total protein. The lysates containing equal amounts of protein (300 μg) were diluted with 10 vol of RIPA buffer and were incubated with 2 μL of αRSK²⁸ for 2 hours at 4°C. Pansorbin was then added, and the immunoprecipitates were washed with buffer B without p-nitrophenylphosphate. Ten microliters of immunoprecipitate was incubated with 12.5 μL of 2× kinase buffer containing 50 mmol/L MOPS (pH 7.2), 120 mmol/L β-glycerophosphate, 60 mmol/L p-nitrophenylphosphate, 10 mmol/L EGTA, 30 mmol/L MgCl₂, 2 mmol/L dithiothreitol, 2 mmol/L sodium orthovanadate, 2 μmol/L protein kinase inhibitor (rabbit sequence), 12.5 μCi (6.25 μL) of [γ-³²P]ATP (10 Ci/mmol), and 1.25 μL of 5 mmol/L S6 peptide at 30°C for 15 minutes. To terminate the reaction, samples were spotted onto Whatman P81 phosphocellulose paper (2.5 cm) and washed five times (5 minutes each) with 180 mmol/L phosphoric acid and once with 95% ethanol. The papers were then placed into scintillation vials, and the radioactivity was counted.

Measurement of [Ca²⁺]_i

Cardiac myocytes were grown on coverslips in serum-free medium. The cells were loaded with indo 1 by incubation with 10 μmol/L of acetoxymethyl ester indo 1 for 30 minutes at 37°C as described previously.³⁴ After three washes with serum-free culture medium, the coverslips were incubated at 37°C for 1 hour and then mounted on the stage of a fluorescence microscope. Ca²⁺ monitoring was performed as described previously.³⁴ The excitation wavelength was 360 nm. Fluorescence signals at 405- and 480-nm emissions from single cells were monitored continuously. Free [Ca²⁺]_i was calculated by the ratio method (405/480 nm), with correction for background fluorescence as described previously.³⁴

Statistics

Data are given as mean±SEM. Statistical analysis was performed by using ANOVA or unpaired t test as appropriate. Significance was accepted at P<.05.

Results

Ang II Causes Tyrosine Phosphorylation in Cardiac Myocytes and Nonmyocytes

We examined whether Ang II induces protein tyrosine phosphorylation in neonatal rat cardiac myocytes by immunoblot analysis using 4G10.²⁶ Ang II (100 nmol/L) caused an increase in phosphotyrosine content of several proteins with apparent molecular sizes of ≈42-, 75- to 80-, and 120- to 130-kD in cardiac myocytes (Fig 1A, left). Ponceau S staining of the filter confirmed that almost equal amounts of protein were loaded in each lane (Fig 1A, right). These signals are specific to phosphotyrosyl residues, because preabsorption of the antibody with 50 mmol/L phosphotyrosine abolished the signals (Fig 1B; see figure legend). The bands at 75- to 80- and 120- to 130-kD are broad and may consist of multiple proteins. A similar pattern of tyrosine phosphorylation was observed when RC20H²⁷ or a polyclonal anti-phosphotyrosine antibody³⁵ was used (data not shown). The tyrosine phosphorylation of p120 to p130 was observed even in the control state, and the level increased after Ang II stimulation, having a peak at ≈1 minute. On the other hand, p42 and p75 to p80 have low or no baseline tyrosine phosphorylation, and their levels reached a peak after 5 minutes. Although less evident in Fig 1A, an increase in tyrosine phosphorylation of p44 by Ang II was clearly observed in multiple experiments (for example see Figs 1B and 1C, 3B, and 4A). When detergents were included in a lysis buffer (modified buffer A), tyrosine phosphorylation of p90 was observed in the control state, and Ang II caused a decrease in the phosphotyrosine content of p90 (eg, see Fig 3B).

We and others have previously shown that Ang II increases both DNA synthesis and cell number (hyperplasia) in cardiac nonmyocytes, the majority of which are cardiac fibroblasts.¹²²¹ In nonmyocyte culture, a treatment with Ang II (100 nmol/L) caused an increase in the phosphotyrosine content of several proteins similar to those observed in myocytes culture, including p75 to p80 and p120 to p130. In a longer exposure of the autoradiogram, tyrosine phosphorylation of p42 and p44 was also observed (Fig 1C, right).

A similar pattern of tyrosine phosphorylation (p42, p44, p75 to p80, and p120 to p130) was observed when cardiac myocytes or fibroblasts were stimulated with 20% fetal calf serum (FCS, Fig 1C), except that in FCS-stimulated cardiac fibroblasts tyrosine phosphorylation of ≈p170 was also observed (Fig 1C, right).

Ang II–Induced Increase in Tyrosine Phosphorylation Is Accompanied by an Increase in Tyrosine Kinase Activity

Increased tyrosine phosphorylation can be caused by activation of tyrosine kinases or by suppression of tyrosine phosphatases. To confirm that tyrosine kinases are activated by Ang II, tyrosine kinase activity in the cell lysates from cardiac myocytes was measured by using an acid-insoluble synthetic tyrosine kinase substrate, EY(4:1).³¹ Because most, if not all, tyrosine kinases are autophosphorylated when they are activated,¹ the kinase assay was performed after immunoprecipitation of cell lysates with 4G10. This removes high background activities of serine/threonine kinases toward endogenous substrates.³¹ As shown in Fig 2, Ang II (100 nmol/L) treatment significantly increased the incorporation of ³²P into EY(4:1) within 1 minute (open circles). This increase in ³²P incorporation was specific to tyrosine-phosphorylated protein, because no significant increase in ³²P incorporation was observed when 4G10 was preabsorbed with 50 mmol/L phosphotyrosine (filled circle). This suggests that the Ang II–induced increase in phosphotyrosine content in cardiac myocytes is due, at least in part, to an Ang II–induced increase in tyrosine kinase activity.

Ang II–Induced Tyrosine Phosphorylation Is AT₁ Receptor–Mediated and Mimicked by Ca²⁺ and Protein Kinase C Activation

We next examined which Ang II receptor subtype (AT₁ or AT₂) mediates protein tyrosine phosphorylation. The Ang II–induced increase in tyrosine phosphorylation of p42 and p120 to p130 was inhibited by the AT₁ receptor antagonist losartan (10 μmol/L) (Fig 3A). Losartan also suppressed Ang II–induced increase in tyrosine phosphorylation of p75 to p80, although losartan slightly increased the baseline tyrosine phosphorylation (Fig 3A). In contrast, the AT₂ receptor antagonist PD123319 (10 μmol/L) did not suppress Ang II–induced tyrosine phosphorylation of p42 and p75 to p80 (Fig 3A). The effect of PD123319 on p120 to p130 was less clear, because PD123319 itself also increased the basal level of tyrosine phosphorylation of p120 to p130 (Fig 3A). This result suggests that Ang II–induced tyrosine phosphorylation of p42 and p75 to p80 is mediated predominantly by the AT₁ receptor subtype. In experiments using a detergent-containing lysis buffer (modified buffer A), Ang II caused a decrease in phosphotyrosine content of p90 (Fig 3B, open triangle). This Ang II–induced dephosphorylation of p90 was also inhibited by losartan but not by PD123319, suggesting that it is also AT₁ receptor–mediated (data not shown).

We next examined how stimulation of the AT₁ receptor by Ang II leads to an increase in phosphotyrosine content in cardiac myocytes. Ang II–induced tyrosine phosphorylation was not inhibited but augmented (Fig 3B) by pretreating cardiac myocytes with pertussis toxin (100 ng/mL) for 24 hours, which has been shown to fully ADP-ribosylate the G_iα subunit in neonatal rat ventricular cardiac myocytes.³⁶ This suggests that pertussis toxin–sensitive G protein may not be involved in the Ang II–induced tyrosine phosphorylation. We have previously shown that Ang II activates multiple phospholipases, including phospholipase C (PLC), through an AT₁ receptor–mediated mechanism in cardiac myocytes.³⁷ Activation of PLC leads to production of inositol trisphosphate (IP₃) and diacylglycerol through hydrolysis of phosphatidylinositol 4,5-disphosphate. These second messengers are known to cause release of Ca²⁺ from intracellular stores and activation of protein kinase C (PKC).³⁸³⁹ A 5-minute treatment of cardiac myocytes with PMA (1 μmol/L), a direct activator of PKC, or A23187 (30 μmol/L), a Ca²⁺ ionophore, caused an increase in phosphotyrosine content of proteins that have molecular sizes similar to those observed with Ang II stimulation, such as p42, p44, p75 to p80, and p120 to p130 (Fig 3C). These data suggest that activation of each component of the PLC-derived second-messenger system seems to be sufficient to cause an increase in phosphotyrosine content of proteins that are tyrosine-phosphorylated in response to Ang II stimulation.

p42 and p44 Are Immunologically Related to ERK-1

Among several tyrosine-phosphorylated proteins, p42 and p44 may correspond to MAP kinases.⁴⁵⁴⁰ An immunoblot of cell lysates from cardiac myocytes with α-cMAPK, directed against the rat 44-kD form of MAP kinase (also called ERK-1),¹⁰ showed the existence of two proteins immunologically related to ERK-1 (Fig 4A, right). Comparison of phosphotyrosine and MAP kinase blots of the same cell lysates electrophoresed on the same SDS-PAGE gel shows that the proteins recognized by α-cMAPK comigrate with tyrosine-phosphorylated p42 and p44 (Fig 4A). The ERK-1–related proteins were observed in the control state. However, after stimulation with Ang II, parts of p42 and p44 bands shifted slightly toward slower mobility (Fig 4A, right), in accordance with the appearance of tyrosine-phosphorylated proteins (Fig 4A, left). Because phosphorylated forms of MAP kinase are known to migrate more slowly in SDS–polyacrylamide gel than the unphosphorylated form,¹⁰ these results are consistent with the notion that Ang II induces tyrosine phosphorylation of ERK-1–related 42- and 44-kD MAP kinases in cardiac myocytes.

To further confirm that both 42- and 44-kD MAP kinases are tyrosine-phosphorylated after Ang II stimulation, MAP kinases were immunoprecipitated by α-MAPK, and anti-phosphotyrosine immunoblotting was performed. As shown in Fig 4B, in cardiac myocytes both 42- and 44-kD MAP kinases were tyrosine-phosphorylated after stimulation with Ang II.

Ang II Activates MAP Kinase

To examine whether MAP kinases are activated by Ang II, we next measured MAP kinase activity. To estimate the molecular sizes of the MAP kinases at the same time, the “in-gel MAP kinase assay” was performed by using an SDS–polyacrylamide gel containing MBP.³² As shown at the left in Fig 5, a 5-minute treatment with Ang II (100 nmol/L) increased MAP kinase activity of proteins at 42- and 44-kD in cardiac myocytes, although the kinase activity of p42 seems to have a higher background level than that of p44 in this assay condition. An additional band was observed at ≈62 kD, and intensity of this band increased by treatment with Ang II. Boulton et al⁴⁰ have reported that neonatal rat hearts express mRNA of ERK-3, a member of the ERK family that has a predicted molecular size of 62.6 kD. It is possible that the band at ≈62 kD may correspond to ERK-3. FCS (20%) also increased MAP kinase activity of p42, p44, and p62 (Fig 5, left). In cardiac nonmyocytes (fibroblasts), Ang II (100 nmol/L) and FCS (20%) increased p42 MAP kinase activity (Fig 5, right). Interestingly, activation of p44 MAP kinase was much less pronounced, and p62 MAP kinase activity was not detectable in nonmyocytes.

To confirm the identity of the 42- and 44-kD kinase activity, the in-gel MAP kinase assay was performed after immunoprecipitation with MAP kinase antibodies. In immunoprecipitates with the anti–ERK-1 antibody (α-MAPK), a significant increase in kinase activity at 42 and 44 kD was observed after a 5-minute treatment of cardiac myocytes with Ang II (Fig 6A). Immunoprecipitates with nonimmune serum had no kinase activity to MBP, indicating that kinase activity in the MAP kinase immunoprecipitates was in fact due to MAP kinase (Fig 6A, lane 6). The 42- and 44-kD forms of MAP kinase activities were also specifically immunoprecipitated by other polyclonal antibodies, anti–ERK-2 (C-14) and anti–ERK-1 (K-23), respectively (Fig 6B). These results suggest that Ang II activates MAP kinase activity of p42^erk2 and p44^erk1, members of the ERK family.⁴⁰ The Ang II–induced increase in MAP kinase activity was observed at 100 pmol/L in p44^erk1 and at 1 nmol/L in p42^erk2 and reached a peak around 100 nmol/L in both cases (Fig 6B). Ang II–induced activation of 42- and 44-kD MAP kinases are inhibited by losartan but not by PD 123319, indicating that Ang II–induced MAP kinase activation is mediated predominantly by the AT1 receptor (Fig 6C). A 5-minute treatment with PMA (1 μmol/L) or A23187 (30 μmol/L) also caused a significant increase in MAP kinase activity at 42 and 44 kD, suggesting that both PKC- and Ca²⁺-dependent stimuli can activate 42- and 44-kD MAP kinases in cardiac myocytes (Fig 6D).

Fig 7A shows time courses of Ang II (100 nmol/L)–induced activation of 42- and 44-kD MAP kinases in cardiac myocytes. Both MAP kinases were activated within 1 minute with treatment with Ang II. Their activity reached a peak at ≈5 minutes and then decreased to below the control level within 30 minutes. It is interesting to note that the activity at 30 minutes after Ang II stimulation is consistently lower than that before Ang II stimulation (p42, 44±9%, P<.05, n=3; p44, 57±6%, P<.05, n=3; percent MAP kinase activity at 30 minutes compared with that at 0 minute). Panels B and C of Fig 7 show the time course of PMA- and A23187- induced MAP kinase activation in cardiac myocytes, respectively. PMA (1 μmol/L) rapidly activated MAP kinases, and their activation persisted for >45 minutes. A23187 (30 μmol/L) also rapidly activated p42 and p44 MAP kinases, but their activity returned to the control level in ≈30 minutes.

Ang II Activates RSK

The 40S ribosomal protein (S6) is phosphorylated in response to mitogens. S6 phosphorylation has been correlated with growth and increased protein synthesis.⁹ The S6 kinase consists of at least two distinct families, the 70- to 85-kD S6 kinase (pp70^S6K) and 90-kD S6 kinase (RSK). Although the upstream kinase capable of activating the pp70^S6K has not been identified, it has been demonstrated that the RSK is activated by p42 and p44 MAP kinases in vitro.⁵⁹ To examine whether Ang II activates RSK, we performed an immune complex RSK assay using an S6 peptide (RRLSSLRA) as a substrate.¹⁰³³ As shown in Fig 8 (circles), Ang II (100 nmol/L) activated RSK in cardiac myocytes. The time course of activation of the RSK followed that of p42 and p44 MAP kinases, being activated within 1 minute, and it reached a peak at ≈10 minutes, but it decreased much more slowly than that of the MAP kinases. Phosphorylation of the S6 peptide in this assay condition was specific to RSK, because no significant increase in S6 peptide phosphorylation was observed when anti-RSK antibody was preabsorbed with an excess amount of antigen peptide (Fig 8, triangles). This Ang II–induced S6 peptide phosphorylation was inhibited by losartan (10 μmol/L) but not by PD123319 (10 μmol/L), suggesting that RSK activation is AT₁ receptor−mediated (data not shown). A 10-minute treatment of cardiac myocytes with FCS (20%), PMA (1 μmol/L), or A23187 (30 μmol/L) also caused a significant increase in RSK activity (Table). The increase in RSK activity was also observed when cardiac nonmyocytes (fibroblasts) were stimulated with Ang II (100 nmol/L) or FCS (20%) (Table).

Roles of PKC in Ang II–Induced Tyrosine Phosphorylation, MAP Kinase, and RSK Activation in Cardiac Myocytes

The precise mechanism of how agonists for G protein–coupled receptors activate the protein tyrosine kinase cascade has not been fully elucidated. The data presented so far indicate that activation of PKC or an increase in intracellular Ca²⁺ is sufficient to cause an increase in tyrosine phosphorylation and activation of MAP kinases and RSK. Therefore, we examined whether PKC is essential for Ang II–induced activation of these protein kinases in cardiac myocytes. We have previously shown that a 48-hour treatment of cardiac myocytes with PMA (2 μmol/L) causes complete functional downregulation of PKC activity, as assessed by PKC assay using a synthetic peptide as a substrate.³⁷ We first examined the effect of a 48-hour treatment with PMA on Ang II–induced tyrosine phosphorylation. As shown in Fig 9A, prolonged pretreatment with PMA (2 μmol/L) alone increased the basal level of tyrosine phosphorylation of p75 to p80 and p120 to p130, but it did not block the Ang II–induced increase in protein tyrosine phosphorylation, including p42 and p44 (arrows). The immune complex in-gel MAP kinase assay showed that the 48-hour pretreatment with PMA prevented both 42- and 44-kD MAP kinase activation by a subsequent addition of the direct PKC activators, phorbol 12,13-dibutyrate (2 μmol/L) or PMA (2 μmol/L) (see Fig 9B, upper panel, and 9C), confirming that at least the phorbol ester–sensitive component of PKC has been completely downregulated in this experiment. However, under this condition, Ang II–induced activation of 42- and 44-kD MAP kinases was not inhibited (Fig 9B and 9C). General protein kinase inhibitors that are reported to be relatively specific to PKC, such as H-7 (50 μmol/L), calphostin C (1 μmol/L), and chelerythrine (10 μmol/L),⁴¹⁴²⁴³ did not block Ang II–induced MAP kinase activation (data not shown), which is consistent with the results of the PMA pretreatment experiments. Similarly, pretreatment with PMA did not affect FCS (20%)–induced MAP kinase activation (Fig 9B). A 48-hour pretreatment of myocytes with PMA slightly increased the baseline RSK activity but did not block subsequent Ang II–induced RSK activation (Fig 9D). These results suggest that Ang II seems to use PKC-independent pathways to induce tyrosine phosphorylation and activate MAP kinases and RSK. However, some isoforms of PKC, the activities of which cannot be detected in our PKC assay, may not have been downregulated. For example, it has been shown that the ξ form of PKC is resistant to downregulation by the prolonged treatment with PMA in neonatal rat cardiac myocytes.¹⁹ Therefore, we cannot exclude the possibility that a PMA-insensitive component of PKC may play a role in Ang II–induced activation of MAP kinases.

Roles of Ca²⁺ in Ang II–Induced Tyrosine Phosphorylation, MAP Kinase, and RSK Activation in Cardiac Myocytes

It is known that Ang II not only increases Ca²⁺ influx through the L-type Ca²⁺ channel⁴⁴ but also induces Ca²⁺ release from intracellular Ca²⁺ stores.⁴⁵⁴⁶ As mentioned earlier, treatment of cardiac myocytes with A23187, a Ca²⁺ ionophore, induced an increase in protein tyrosine phosphorylation and activated 42- and 44-kD MAP kinases and RSK. To block an increase in [Ca²⁺]_i, irrespective of its origin, cardiac myocytes were incubated with a membrane-permeable Ca²⁺-chelating compound, BAPTA-AM (10 μmol/L), for 30 minutes.⁴⁷ To confirm that the concentration of BAPTA-AM used was sufficient to inhibit the Ang II–induced rise in [Ca²⁺]_i, continuous measurements of [Ca²⁺]_i by fluorescence microscopy were performed at the single-cell level. Fig 10A shows a typical time course of the Ang II (100 nmol/L)–induced change in [Ca²⁺]_i. A 30-minute pretreatment with BAPTA-AM significantly suppressed the Ang II–induced increase in [Ca²⁺]_i. Fig 10B shows absolute values of peak [Ca²⁺]_i obtained by each treatment. BAPTA-AM (10 μmol/L) significantly lowered the resting level of [Ca²⁺]_i and suppressed the Ang II–induced increase in [Ca²⁺]_i to the control resting level. A 30-minute treatment of cardiac myocytes with BAPTA-AM (10 μmol/L) completely inhibited the Ang II–induced increase in tyrosine phosphorylation of multiple proteins, including p42 and p44 (Fig 9A, arrows). BAPTA-AM also markedly suppressed the Ang II–induced activation of 42- and 44-kD MAP kinases (Fig 9B and 9C). However, BAPTA-AM did not affect PMA-induced MAP kinase activation (Fig 9B, lower panel, and 9C), suggesting that BAPTA-AM, at the concentration used (10 μmol/L), did not affect the PKC-dependent pathway of MAP kinase activation. BAPTA-AM also significantly suppressed Ang II–induced RSK activation, as assessed by the immune complex RSK assay (Fig 9D). These results suggest that Ang II–induced protein tyrosine phosphorylation and activation of MAP kinases and RSK are critically dependent on [Ca²⁺]_i.

Other G_q Protein–Coupled Hypertrophic Stimuli Also Activate Tyrosine Kinase, MAP Kinase, and RSK in Cardiac Myocytes

Finally, we examined whether other hypertrophic stimuli acting through G protein–coupled receptors, such as Phe, NE, ET-1, and isoproterenol (ISO) activate tyrosine kinase, MAP kinases, and RSK in cardiac myocytes. As shown in Fig 11A, a 5-minute treatment of cardiac myocytes with NE (1 μmol/L), Phe (100 μmol/L), or ET-1 (50 nmol/L), agonists for G_q protein–coupled receptors, produced an increase in phosphotyrosine content of p42, p75 to p80, and p120 to p130. Interestingly, ISO (10 μmol/L), an agonist for G_s protein–coupled receptor, induced an increase in phosphotyrosine content of p55, p75 to p80, and p120 to p125, but it did not increase that of p42 and p130 (Fig 11A, arrows with asterisk). NE, Phe, and ET-1 also increased 42- and 44-kD forms of MAP kinase activity in cardiac myocytes, as assessed by the immune complex in-gel kinase assay (Fig 11B), in agreement with recent reports that Phe and ET-1 stimulate MAP kinase activity in cardiac myocytes.¹⁸ However, ISO (10 μmol/L) did not increase MAP kinase activity (Fig 11B), corresponding to the lack of tyrosine phosphorylation of p42. Similarly, a 10-minute treatment of cardiac myocytes with NE, Phe, or ET-1 caused a significant increase in RSK activity, as assessed by the immune complex RSK assay using S6 peptide as a substrate, but stimulation with ISO did not (Table).

Discussion

We have demonstrated that Ang II and other hypertrophic stimuli known to act through G_q protein–coupled receptors cause tyrosine phosphorylation of several intracellular proteins and activate MAP kinases and RSK. Activation of these protein kinases was observed in both cardiac myocytes and nonmyocytes (fibroblasts), suggesting that these second-messenger systems may be used in both hypertrophic and mitogenic responses to Ang II. Interestingly, a Ca²⁺-dependent, rather than PKC-dependent, mechanism seems to be critical in Ang II–induced tyrosine phosphorylation and activation of MAP kinases and RSK in cardiac myocytes.

Because our cardiac myocyte culture contains 5% to 10% of contaminating nonmyocytes, which are primarily fibroblasts,¹²²⁹ one may argue that the Ang II–induced increase in tyrosine phosphorylation and activation of MAP kinases and RSK might have originated solely from the “contaminating” fibroblasts without contribution from myocytes. To address this question, we examined Ang II–induced activation of these protein kinases in both myocyte-rich culture and cardiac fibroblast culture. We performed experiments on myocytes and fibroblasts in identical conditions and observed comparable levels of activation of these second-messenger systems in both cell types. If the fibroblasts were the sole source of the activation of protein kinases, the contaminating fibroblasts (which are <10% of the total cells) in the myocyte culture could not have accounted for the levels of activation seen in the myocyte culture.

Our results also indicate that a downstream signaling molecule of PLC (either elevation of intracellular Ca²⁺ or activation of PKC) is sufficient to cause activation of MAP kinases and RSK in cardiac myocytes. However, among these two signaling pathways, the Ca²⁺-dependent pathway seems to play a more critical role in Ang II–induced activation of MAP kinases and RSK in cardiac myocytes. The following lines of evidence support this conclusion. First, downregulation of PKC by prolonged treatment with PMA had no effect on Ang II–induced activation of MAP kinases or RSK. In contrast, phorbol ester–induced MAP kinase activation was completely inhibited by PKC downregulation. Second, three different classes of protein kinase inhibitors reported to be (relatively) specific to PKC also failed to inhibit Ang II–induced activation of MAP kinases. Third, BAPTA-AM, at a concentration that completely suppressed the Ang II–induced increase in [Ca²⁺]_i, fully suppressed activation of these protein kinases. Because treatment with BAPTA-AM did not completely inhibit Ang II–induced production of IP₃³⁷ or phorbol ester–induced activation of MAP kinase (Fig 9B and 9C), the effect of BAPTA-AM was unlikely to be due to a direct inhibition of PLC, PKC, or MAP kinases. Fourth, the time course of Ang II–induced MAP kinase activation was more similar to an A23187-induced MAP kinase activation than a PMA-induced one. MAP kinase activation by Ang II was rapid and transient, and its activity returned to the levels lower than the control level within 30 minutes. The mechanism of this more rapid inactivation of MAP kinases observed in Ang II–treated or A23187-treated cells remains to be determined. Ca²⁺-dependent activation of phosphatase may be one of the mechanisms, as has been suggested by deactivation of epidermal growth factor–activated MAP kinase in human foreskin fibroblasts.⁴⁸ It is interesting to note that Duff et al⁴⁹ recently reported that Ang II rapidly upregulates mRNA of 3CH134, a protein tyrosine phosphatase that specifically dephosphorylates MAP kinase in vascular smooth muscle cells.

It has been shown that activation of MAP kinase by thapsigargin also requires Ca²⁺, and the Ca²⁺ released from internal stores appears to be sufficient for mediating the MAP kinase activation in human foreskin fibroblasts.⁴⁸ At present, we do not know what step in Ang II–induced MAP kinase activation is critically dependent on [Ca²⁺]_i. It is interesting to note that Huckle and coworkers³¹⁵⁰ have reported that Ang II–induced tyrosine kinase activation is Ca²⁺ dependent in WB and GN4 liver epithelial cells. This raises the possibility that there exists an intermediate tyrosine kinase(s) that plays an important role in Ang II–induced MAP kinase activation. Although our results suggest that Ca²⁺-dependent signaling is essential, we found that the magnitudes of tyrosine phosphorylation and activation of MAP kinases and RSK were higher in Ang II stimulation than in A23187 stimulation alone. In physiological conditions, both Ca²⁺ and PKC are likely to act synergistically for the Ang II–induced activation of MAP kinases and RSK in cardiac myocytes.

We have shown that MAP kinases are tyrosine-phosphorylated by Ang II treatment. However, we do not know the molecular identities of other proteins that are tyrosine-phosphorylated by hypertrophic stimuli. In the case of the platelet-derived growth factor and epidermal growth factor receptors, several cellular proteins have been identified as substrates of the receptor tyrosine kinases. These include phospholipase Cγ (molecular mass, ≈148 kD), PI3 kinase (85 and 110 kD), and GTPase activating protein (120 kD).¹ Recently, Molloy et al¹⁵ reported that p75 and p120, which are tyrosine-phosphorylated by Ang II in vascular smooth muscle, are different from these known tyrosine kinase substrates. We also observed tyrosine phosphorylation at ≈75 to 80 kD and 120 to 130 kD in cardiac myocytes and fibroblasts. Further studies are necessary to identify whether either or both of these proteins are known substrates.

In the present experiments, we have shown that in cardiac myocytes both hypertrophic stimuli such as Ang II, NE, Phe, and ET-1¹²¹³¹⁴ and mitogenic stimuli such as FCS⁵¹ seem to use a similar tyrosine kinase–MAP kinase cascade. This is reminiscent of PC12 cells, in which agonists promoting differentiation (nerve growth factor) and proliferation (epidermal growth factor) both activate the tyrosine kinase–MAP kinase cascade.⁵² This indicates that differences in the cellular response to stimuli that cause hypertrophy or hyperplasia may depend on differences in cellular processes downstream from these protein kinases. Alternatively, each specific form of receptor activation may stimulate other parallel signaling pathways that determine the specificity of the stimulation.

We have previously shown that Ang II–induced activation of various phospholipid-derived second-messenger systems is mediated by the AT₁ receptor subtype.³⁷ Our results indicate that tyrosine phosphorylation and activation of MAP kinases and RSK are also mediated by the AT₁ receptor subtype. The AT₂ receptor in rat adrenal glomerulosa and PC12W cells has recently been suggested to modulate tyrosine phosphatase activity.⁵³⁵⁴ In our experiments, tyrosine phosphorylation of p90 was observed by using a detergent-containing lysis buffer (modified buffer A), and the tyrosine phosphorylation of p90 was decreased by the treatment with Ang II (Fig 3B). However, this decrease was not affected in the presence of PD123319, indicating that the AT₂ receptor does not seem to play a major role in the regulation of phosphotyrosine content in cardiac myocytes.

We have previously shown that after prolonged treatment with PMA, Ang II–induced c-fos expression is significantly, if not completely, suppressed, indicating that the classical PKC pathway plays an essential role in Ang II–induced c-fos expression in cardiac myocytes.³⁷ This indicates that Ang II activates c-fos and MAP kinases mainly through different signaling mechanisms. However, it should be noted that this Ang II–induced c-fos expression was not completely inhibited by the prolonged treatment with PMA.³⁷ This suggests that there may be a PKC-independent mechanism in Ang II–induced c-fos expression, although we cannot exclude a contribution from a PMA-insensitive component of PKC.¹⁹³⁹ We have previously shown that a binding site for p62^TCF in the c-fos promoter is necessary for the Ang II–induced c-fos expression in cardiac myocytes.³⁷ Interestingly, a recent report showed that MAP kinase can phosphorylate the nuclear transcription factor p62^TCF and regulate transcription of c-fos.¹¹ Ang II rapidly activates MAP kinase, and our preliminary results using immunofluorescence staining suggest that Ang II causes translocation of MAP kinase into the nucleus of cardiac myocytes (J. Sadoshima and S. Izumo, unpublished data). Thus, MAP kinase may play a role in Ang II–induced immediate-early gene expression in cardiac myocytes, although further investigation is necessary to prove this hypothesis.

It is known that some agonists for G protein–coupled receptors, such as thrombin and lysophosphatidic acid, are capable of stimulating DNA synthesis in some cell types. However, activation of the PLC pathway is neither required nor sufficient for lysophosphatidic acid–induced and thrombin-induced DNA synthesis, and the importance of other downstream signaling mechanisms has been suggested.⁵⁵ Some other agonists for G protein–coupled receptors (such as bombesin and serotonin) are capable of stimulating DNA synthesis only in the presence of other growth factors that activate tyrosine kinase (such as insulin and epidermal growth factor).⁵⁵ Because Ang II alone induces a mitogenic response in cardiac fibroblasts, it would be interesting to examine whether tyrosine kinase activation is critical in Ang II–induced mitogenic response in these cells. Moreover, by analogy to mitogenic response, it would be interesting to examine whether tyrosine kinase activation is necessary for hypertrophic response of cardiac myocytes. Further studies are necessary to identify the role of tyrosine kinases and downstream signaling in the cardiac hypertrophic response caused by G protein–coupled receptors.

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Footnotes