INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN RESPONSE to physiological and pathological stimuli, the mammalian heart undergoes hypertrophic growth. This hypertrophy is associated with a number of phenotypic changes including increased expression of cardiac-specific genes such as atrial natriuretic factor (ANF). In vitro, induction of the hypertrophic program and activation of the ANF promoter in neonatal ventricular myocytes can be induced by agonists such as phenylephrine or angiotensin II, which activate G protein-coupled receptors (15, 24, 27, 30) and increase muscle cell contraction rate (18). These stimuli also enhance the organization of contractile proteins into sarcomeres and significantly increase the size of cardiac myocytes in culture.

A number of studies have indicated that the intracellular signaling mechanisms that mediate cardiac hypertrophy require one or more members of the mitogen-activated protein kinase (MAPK) cascades (2-5, 7, 22, 23, 25, 28, 34, 38). The most recently described member of the MAPK family is the stress-activated protein kinase p38. Activation of p38 normally occurs as a result of cellular stresses such as ultraviolet irradiation and osmotic shock. Oxidative stress and ischemia-reperfusion can also activate p38 in perfused rat hearts (6).

Activation of the p38 MAPK pathway is sufficient to stimulate the ANF promoter in cardiac myocytes (21, 35, 36, 38). Overexpression of MEK6 (the direct upstream activator of p38) has been shown to protect myocytes from apoptosis induced by anisomycin or MEKK-1 (37) and to activate ATF6, which may cooperate with serum response factor to induce expression from the ANF promoter (35). p38 can also phosphorylate and activate the transcription factor MEF2C (11), which has been shown to be important in cardiovascular development and a likely candidate for involvement in cardiac hypertrophy. Whereas there has been some agreement that p38 can be involved in gene expression changes associated with hypertrophy, there are conflicting reports for the role of p38 in mediating the morphological changes of hypertrophy. Infection of cardiomyocytes with adenoviral vectors expressing activated mutants of MEK6 (MKK6DD) or transient transfection of expression constructs for MEK6 caused increased cell size and sarcomere organization (36, 38). However, Clerk et al. (7) showed that inhibition of p38 failed to prevent the hypertrophic morphology induced by phenylephrine and endothelin-1 between 4 and 24 h, although the inhibitor did decrease organization and cell profile at 48 h. Thus it is unclear whether p38 plays a central or minor role in the regulation of morphological and biochemical markers of myocardial hypertrophy.

We examined the role of p38 during induction of the ANF promoter and morphological hypertrophy in response to phenylephrine. Surprisingly, phenylephrine-induced ANF-driven gene expression did not require p38 in low-density myocyte cultures. However, high-density cultures required p38 activity for gene expression. High-density cultures spontaneously contracted at a significantly higher rate than isolated myocytes. Mimicking this effect by rapid electrical pacing of the myocytes induced ANF-driven reporter plasmids, and this induction was sensitive to the p38 inhibitor. Rapid pacing of cultures resulted in small increases in phosphorylated p38 (phospho-p38) levels, and at later time points the kinase was decreased in paced samples. Although p38 appeared to play a role in ANF-driven gene expression in certain conditions, we failed to find that p38 activity was necessary for the morphological features of hypertrophy. These data indicate that p38 may regulate cardiac hypertrophic morphology and gene expression only in certain cell culture conditions.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell culture and transfections. Cardiac muscle cells were cultured and transfected as previously described (12, 32, 33). For low-density myocyte cultures, cells were seeded at a density of 260 cells/mm², and for high-density cultures, cells were seeded at a density of 1,300 cells/mm². For electrical pacing experiments, the cells were plated at 1,000-1,700 cells/mm² in 24-well culture plates, which is required for efficient pacing in vitro (unpublished observations). Myocytes were transiently transfected in duplicate or triplicate 24 h after the cell isolation procedure for all gene expression assays. Approximately 15 h after the calcium phosphate transfection procedure was commenced, cells were washed with serum-free medium and treated with appropriate agonists and inhibitors. For pacing experiments, myocytes were subjected to electrical pacing at this time. Other transfections were conducted in duplicate or triplicate in 3.5-cm tissue culture dishes using the calcium phosphate precipitation method. Before transfection, all plasmids were purified by alkaline lysis followed by polyethylene glycol precipitation. Luciferase and -galactosidase activities were assayed on a Dynex MLX luminometer 48 h after transfection by harvesting cells in reporter lysis buffer (Promega E397a, Madison, WI) and incubating with the appropriate light-emission accelerator [Promega E148a or Tropix (Bedford, MA) reaction buffer, respectively].

Cell treatments. Cells were treated as required with phenylephrine (50 µM) (Sigma, St. Louis, MO) and the p38 inhibitor SB-203580 (20 µM; Calbiochem, La Jolla, CA) and kept from direct light to prevent degradation. Propranolol (1 µM) was included in culture medium used in these experiments to block -receptors. For electrical stimulation to induce muscle cell contraction, the cells were plated in 24-well plates that were electrically connected via an agarose salt bridge and stimulated with a custom-built electric stimulator. In this case, controls were from the same plate of cells but from wells that were not electrically connected. Media were changed every 24 h for all experiments. Electrical pacing experiments were conducted in a manner similar to those described by McDonough et al. (18). Briefly, either paced or unpaced cardiac cells were kept in serum-free medium while electric pulses were delivered at a frequency of 2-3 Hz for 24-72 h. The polarity of electrodes was reversed every other pulse to prevent the accumulation of net electric charge and pH drift of medium.

p38 Western blots. Assays for p38 activity were performed using the PhosphoPlus p38 MAP Kinase Antibody Kit (New England BioLabs, Beverly, MA). Cardiac cells were kept in serum-free medium for 24 h before agonist addition or pacing. The medium was first changed again to serum-free medium (±inhibitors), and cells were then allowed to equilibrate undisturbed for 6 h. At that time, either phenylephrine (50 µM) was added or electrical pacing at 2-3 Hz was begun. Cells were scraped in ice-cold phosphate-buffered saline (PBS) and resuspended in radioimmunoprotective assay buffer containing protease inhibitors (10 mM Tris, pH 7.4, 0.15 M NaCl, 1% Nonidet P-40, 1% SDS, and 10 µg/ml each of aprotinin, leupeptin, and pepstatin with 1 mM Na-vanadate). Protein content from cell extracts was quantified using the bicinchoninic acid protein assay kit (Pierce, Rockford, IL). Sample aliquots (25-40 µg cellular protein) were subjected to SDS-PAGE and Western blotting. For pacing experiments, this represented protein from ~4-16 wells from 24-well culture dishes. Phospho-p38 and total p38 were detected by chemiluminescence as described by the manufacturer. Real-time quantitation of band intensity was analyzed on a Lumi-Imager (Boehringer Mannheim, Indianapolis, IN), and the images were quantitated using the Lumi-Analyst program. Quantitative densitometry was conducted on a ScanMaker IISP and analyzed using Kodak Digital Science Image Analysis software.

p38 kinase assays. Cardiac cells were first kept in serum-free medium as described in p38 Western blots. Approximately 1.6 × 10⁶ cardiac cells were either unpaced or paced at 2 Hz for 5 min to 4 h in 24-well plates. Total p38 was immunoprecipitated from whole cell extracts by first washing paced or unpaced cells with ice-cold PBS. Protein was harvested for immunoprecipitation kinase assays and incubated with 15 µl of anti-p38 antibody in kinase buffer in a protocol similar to that described previously (31, 32). Immunoprecipitates were finally used in kinase assays with myelin basic protein (MBP; Sigma). Reactions were stopped by addition of 2× Laemmli buffer, and phosphorylated products were subjected to SDS-PAGE and autoradiography.

Plasmids. The MEK6 expression plasmid (MKK6) was provided by Bernd Stein (Signal Pharmaceuticals, La Jolla, CA). Luciferase reporter plasmids were normalized to a cotransfected RSV-LacZ plasmid provided by Michael Kapiloff (Oregon Health Services University, Portland, OR). CL100 was provided by Steve Keyse (Imperial Cancer Research Fund, University of Dundee, Nethergate, Dundee, Scotland). ANF-luciferase is a 638-bp fragment of the rat ANF promoter cloned into pGL3 basic. pFA-CHOP and pFR-luciferase were obtained from Stratagene (La Jolla, CA).

Measurement of contraction rate. Low- or high-density myocyte cultures were treated with 50 µM phenylephrine and 1 µM propranolol for 24 h in serum-free medium. Contraction was measured visually on a Zeiss Axiovert microscope in a preequilibrated 37°C, 10% CO₂ stage. Myocyte cultures were at equilibrium conditions during these measurements and were viewed under low-intensity light to prevent heating. Thirty measurements at 1 min each were made for isolated cells, and ten independent measurements were made for five cell clusters and dense cultures.

[³H]phenylalanine protein measurements. Rates of protein synthesis were conducted by a method similar to that of Abdellatif et al. (1). Briefly, cardiac myocytes were serum starved for 24 h after isolation and then exposed to 0.4 mM unlabeled phenylalanine for 30 min to expand intracellular pools of the amino acid. The medium was then removed and replaced with medium containing 2 µCi/ml L-[ring-2,3,4,5,6-³H]phenylalanine and appropriate agonists (pacing 2 Hz or 50 µM phenylephrine) or inhibitors (20 µM SB-203580) along with 1 µM propranolol. The medium was changed every 24 h. Total protein/RNA and DNA were simultaneously harvested using the RNA/DNA extraction kit (Qiagen, Santa Clarita, CA). Protein pellets were washed twice in ice-cold trichloroacetic acid (10%) and resuspended in 0.3 N NaOH, and radioactivity was quantitated by liquid scintillation counting in 5.0 ml of Opti-Fluor reagent (Packard, Meriden, CT). DNA pellets were resuspended in distilled H₂0 and quantitated by spectrophotometry with absorbance at 260 nm. Newly synthesized protein (in counts/min) was normalized to DNA yields. For pacing experiments, the ³H-labeled protein levels were also normalized to cell number by direct counting of the number of living cells per high-power field. Cells were counted from three random fields for each sample. Each data point represents two samples, thus resulting in six independent measurements of cell number per data point. ³H-labeled protein was also harvested in TRIzol reagent (GIBCO, Grand Island, NY) with similar results. In this case, 300 µl EtOH was added to reduce quenching of the samples during scintillation counting.

Time-course p38-inhibition studies. To inhibit p38 activity during various time points during cell culture, cardiac myocytes were transiently transfected as described in Cell culture and transfections and washed with serum-free medium. Cells were then treated with SB-203580 (20 µM) for the first 24 h of culture (hours 1-24) or the last 24 h of culture (hours 48-72), or the p38 inhibitor was kept in the medium chronically (hours 1-72). All media were changed every 24 h. ANF-luciferase gene expression assays were conducted at the 72-h time point after the end of treatment. Cells treated with p38 inhibitor for the first 24 h were washed thoroughly with serum-free medium before the addition of medium containing either DMSO or SB-203580.

Statistics. The data presented for each experiment represent the mean ± SE for a single experiment. All transfection experiments presented were conducted in either duplicate or triplicate. These experiments were representative of other experiments conducted during the study. When appropriate, statistical methods were applied including ANOVA (single factor). Two sample t-tests for independent means, assuming equal variances, were also performed where required, indicating significance at P < 0.05. For all experimental data points, significance has been indicated either between unstimulated controls and stimulated values or between nonactivated and activated cultures within similar drug treatment categories or between drug groups. Bonferroni's method was used for making multiple pairwise comparisons when the hypothesis of the equality of several means was rejected.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

p38 activity is not required for ANF-driven gene expression stimulated by phenylephrine in low-density cultures but is required in dense cultures. To determine whether the MAP kinase p38 was involved in induction of the ANF promoter in cardiac myocytes, we performed transient transfection assays in primary ventricular myocyte cultures using an ANF-luciferase reporter plasmid. Figure 1A shows that inhibition of p38 had no effect on phenylephrine-induced ANF-luciferase expression. These cells were plated at a density of 260 cells/mm², which is that of our routine cultures. The inhibitor failed to prevent ANF-luciferase expression even at the limiting doses of phenylephrine treatment, which did not induce maximal gene expression. Figure 1B shows a similar experiment in which active MEK6, which induces p38 but not other MAP kinases, can induce ANF expression, confirming results from others (21, 36, 38), and the p38 inhibitor SB-203580 strongly inhibits this gene expression.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   p38 inhibitor SB-203580 (SB) blocks phenylephrine-induced atrial natriuretic factor-luciferase (ANF-Luc) expression only in dense myocyte cultures. A: activation of ANF promoter by phenylephrine is not inhibited by SB. At all phenylephrine concentrations (0.1-100 µM) p38 inhibitor failed to reduce activation by phenylephrine, even at limiting doses of agonist. -gal, -galactosidase. * Significant difference (P < 0.05) between nonactivated and phenylephrine (100 µm)-activated myocytes within same treatment group. B: expression vector encoding constitutively active MEK6 (MKK6DD) was transiently transfected in neonatal rat cardiac myocytes along with ANF-Luc reporter plasmid and RSV-driven -gal for normalization. MEK6 potently activates ANF promoter (DMSO included as vehicle control). p38 inhibitor SB (20 µM) strongly blocks MEK6-activated ANF-Luc expression, indicating that inhibitor is working as expected in these culture conditions. * Significant difference (P < 0.05) between nonactivated and MEK6 (3 µg)-activated myocytes within same treatment group. # Significant difference (P < 0.05) between MEK6 (3 µg)-activated control and SB-treated myocytes. C: high-density (1,300 cells/mm2) or low-density (260 cells/mm2) cardiac myocytes were transfected with an ANF-Luc reporter plasmid and treated with phenylephrine (50 µM) and SB (20 µM) or with DMSO as a control. p38 inhibitor only inhibited ANF-driven gene expression in response to phenylephrine in high-density cell cultures. * Significant difference (P < 0.05) between nonactivated and phenylephrine (50 µg)-activated myocytes within same treatment group. # Significant difference (P < 0.05) between phenylephrine (50 µg)-activated DMSO- and SB-treated high-density myocytes.

Increasing contraction rate by rapid electrical pacing induces ANF gene expression that requires p38 activity. To determine whether cell density affected metabolic activity when cells were treated with phenylephrine, we measured the pH of the medium of low- and high-density cultures with or without phenylephrine treatment. We reasoned that the pH of the medium might differ with both cell density and phenylephrine treatment because of the liberation of lactic acid. Figure 2A shows that this is the case. When low-density cultures were treated with 50-100 µM phenylephrine, no discernible change from untreated controls in pH of the medium was measured. However, high-density cultures showed a marked acidification of the medium in cells treated with phenylephrine, presumably due to the increased contraction rate and metabolic activity of these cells. Interestingly, the acidification of the medium was most apparent in cells treated with 50 rather than 100 µM phenylephrine. Similar chronotropic effects of phenylephrine have been observed previously (9). When the contraction rate of phenylephrine-treated cells in low- and high-density cultures was measured directly, we found that high-density myocyte cultures contracted at a significantly higher rate than those at low density (Fig. 2B). Isolated myocytes contracted at a mean of 5.7 beats/min (median 0, range 58, SD 15, n = 30), five myocytes grouped together contracted at a mean of 55.7 beats/min (median 58, range 79, SD 23.5, n = 10), and cells in a field of dense culture contracted at a mean of 115 beats/min (median 114, range 20, SD 5.9, n =10). Although occasional isolated myocytes had high contraction rates, the majority of cells displayed virtually no contractile activity, as evidenced by the value of the median (0) in these measurements.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Contraction-induced gene expression is dependent on p38. A: high- or low-density myocyte cultures were incubated for 24 h with 50 or 100 µM phenylephrine. pH of medium was then determined in duplicate measurements. Only high-density cultures treated with phenylephrine showed significant acidification of medium, an indicator of high metabolic activity and lactate levels. B: contraction rate was measured for phenylephrine-treated isolated myocytes, cells in groups of 5 myocytes, or cells electrically connected in fields of dense cultures. Higher cell density increased spontaneous contraction of myocytes in culture. C: cardiac myocytes were transiently transfected with ANF-Luc and subjected to electrical pacing to stimulate contraction at 2 Hz. Addition of p38 inhibitor SB (20 µM) blocked pacing-induced ANF-driven gene expression. * Significant difference (P < 0.05) between unpaced (control) and paced untreated myocytes. # Significant difference (P < 0.05) between paced untreated and paced SB-treated myocytes.

Rapid electrical pacing induces phosphorylation and activation of p38. Because pacing-induced ANF-luciferase expression was inhibited by SB-203580, we wanted to determine whether increasing the contraction rate was sufficient to activate p38. Figure 3A shows that this is the case. Cardiac myocytes driven to contract at 2 Hz showed increased phosphorylation of p38 after 30 min to 2 h by Western blot using a phospho-specific antibody. Three independent pacing experiments are shown. This activation was small compared with phenylephrine (Fig. 3A) or anisomycin controls (data not shown). However, in six of seven experiments, the level of phospho-p38 was higher in paced samples from 30 min to 1 h and was generally increased approximately twofold when assayed by a chemiluminescence imager (data not shown) or densitometry [30-min activation was 2.3-fold (median 2.15, range 1.6-3.2, SD 0.61, SE 0.3, n = 5) and 1-h activation was 2.4-fold (median 1.46, range 1.37-5.8, SD 1.65, SE 0.6, n = 7)]. We also performed in vitro kinase assays using anti-p38 antibody to immunoprecipitate p38 from cell extracts. We then tested the ability of the immunoprecipitated kinase to phosphorylate the MBP substrate. As shown in Fig. 3B, paced samples from 5 min to 4 h had only small increases in p38 activity compared with controls. This result confirmed the small but consistent increase in phospho-p38 detected in Western blots. We also tested the levels of phosphorylated and nonphosphorylated forms of p38 at later time points. Surprisingly, as shown in Fig. 3C, in four of four experiments, both the phosphorylated and nonphosphorylated forms of p38 were reduced at all time points after 24 h. In most cases, there was a more significant decrease in phospho-p38 in paced cells, compared with total p38 (see, e.g., 72-h time point). The same blot was also probed for total actin, confirming approximately equal loadings of total protein and that paced samples containing hypertrophied myocytes showed the expected increases in actin levels.

View larger version (39K): [in this window] [in a new window]   Fig. 3.   Myocyte contraction induces weak p38 phosphorylation. A: cardiac myocytes were electrically driven to contract at 2 Hz and harvested after 30 min or 1, 2, or 4 h. Three different pacing experiments are shown. A: phenylephrine-stimulated p38. p38 activation was determined by immunoblotting with a phospho-specific or total p38 antibody. Contraction induced p38 phosphorylation (phospho-p38) by 30 min to 1 h (but not 24 h) in 6 of 7 experiments. B: immunoprecipitated p38 from paced samples at time points ranging from 5 min to 4 h was used in a kinase assay using myelin basic protein (MBP) as substrate. Pacing induced weak phosphorylation of MBP. C: experiments similar to that in A were conducted for samples paced for 1-72 h. p38 Western blot shows that pacing decreases levels of both phosphorylated and nonphosphorylated forms of p38 at later time points. Both total p38 and actin are also shown. Identical results were obtained in 3 similar experiments.

Inhibition of p38 during pacing suggests both direct and indirect actions of p38 on ANF promoter. The experiments shown in Fig. 3 show that myocyte contraction induces small increases in p38 phosphorylation at early time points and that the kinase is then downregulated starting after ~24 h of pacing. We then tested whether inhibition of p38 during the first 24 h would affect the activity of the ANF promoter when assayed at 72 h of electrically driven contraction. Figure 4A shows a time-course activation of the ANF promoter by pacing. Activation of the promoter occurs after ~24 h. Virtually no activation of the promoter was seen at the 6-h time point. Interestingly, inhibition of p38 with SB-203580 during hours 1-24 of contraction had no effect on ANF expression when assayed at 72 h (Fig. 4B). Likewise, inhibition of p38 during hours 48-72 of contraction also had no effect on ANF promoter activity. Only the continuous presence of SB-203580 during hours 1-72 inhibited ANF-luciferase induced by contraction. We then tested the ability of SB-203580 to inhibit ANF-luciferase activation by constitutively active MEK6 (MKK6DD). As shown in Fig. 4C, MEK6 induction of the ANF promoter is inhibited by the p38 inhibitor when included in the medium at all time points. Inhibition of p38 chronically resulted in the most significant inhibition. These data suggest that the effect of p38 on ANF gene expression may be both direct and indirect.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Chronic p38 inhibitor treatment is required to inhibit pacing-induced ANF-driven luciferase expression. A: cardiac myocytes were transiently transfected with ANF-Luc and RSV -gal plasmids. Time-course activation of ANF promoter by electrical pacing was determined. B: cultures similar to those in A were treated with SB during the first (hrs 1-24) or last (hrs 48-72) 24 h of pacing or were included chronically in medium (hrs 1-72). Only chronic inactivation of p38 was sufficient to inhibit pacing-induced ANF expression. * Significant difference (P < 0.05) between unpaced and paced myocytes within same treatment group. # Significant difference (P < 0.05) between paced DMS0-treated and paced SB-treated myocytes in hrs 1-72. C: cells were transiently transfected with ANF-Luc along with constitutively active MEK6 (MKK6DD). SB was added in a manner identical to that in experiments conducted in B. Presence of p38 inhibitor during any of the time periods reduced MEK6-induced activation of ANF promoter. This effect was greatest when p38 inhibitor was included throughout 72-h time period. * Significant difference (P < 0.05) between nonactivated (control) and MEK6-activated myocytes within same treatment group. # Significant difference (P < 0.05) between MEK6-activated DMSO-treated and MEK6-activated SB-treated myocytes within same treatment group.

Inhibition of p38 fails to prevent increases in protein synthesis and morphological features of hypertrophy induced by electrical pacing or phenylephrine. To determine whether p38 activity was necessary for cardiac myocytes to increase protein synthesis rates induced by phenylephrine and pacing, we measured the change in [³H]phenylalanine incorporation in cells that were treated with phenylephrine or electrical pacing in the presence or absence of the p38 inhibitor. These experiments are shown in Figs. 5, A and B. It was determined that in both low-density (250 cells/mm²) and high-density (1,240 cells/mm²) cultures an increase in protein synthesis occurred in response to phenylephrine, even in the presence of the inhibitor SB-203580. In several experiments we noted that the p38 inhibitor decreased the protein content in unstimulated control samples. However, even in these cases, the change from basal protein incorporation rates to the increased hypertrophied state in response to phenylephrine still occurred. This is particularly apparent in Fig. 5B, where, although the p38 inhibitor decreased basal protein metabolism in the 72-h time point, these cells still markedly hypertrophied in response to electrically driven contraction. It was also noted that in sparse myocytes at the 24-h time point the p38 inhibitor seemed to reduce the basal and induced levels of protein synthesis. However, these myocytes still markedly hypertrophied at the 48-h time point.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   SB fails to significantly inhibit [3H]phenylalanine incorporation in phenylephrine-treated or electrically paced myocytes. Cardiac myocytes were cultured in presence of 2 µCi/ml L-[ring-2,3,4,5,6-3H]phenylalanine and appropriate agonists: 50 µM phenylephrine (A) or electrical pacing at 2 Hz (B) for 24, 48, or 72 h. Cells were exposed to either SB (20 µM) or DMSO carrier along with 1 µM propranolol. Although p38 inhibitor decreased total 3H-labeled protein levels for some samples, change from control to hypertrophied phenotype still occurred for both agonists. * Significant difference (P < 0.05) between nonactivated (carrier) and activated myocytes within same drug treatment group (A) or between unpaced and paced myocytes within same drug treatment group (B). # Significant difference (P < 0.05) between DMSO- and SB-treated nonactivated low-density myocytes during 1st 24 h.

View larger version (140K): [in this window] [in a new window]   Fig. 6.   Inhibition of p38 does not prevent myocyte hypertrophy induced by phenylephrine or contraction. Cardiac myocytes were treated with either DMSO (A-C) or SB (20 µM) (D-F) to inhibit p38 and subjected to phenylephrine (50 µM) treatment (B and E) or electrical pacing at 2 Hz (C and F). Low-power (×320) phase photographs are shown. Cells treated with SB hypertrophy normally in response to both stimuli.

View larger version (114K): [in this window] [in a new window]   Fig. 7.   Inactivation of p38 does not inhibit formation of myofibrils and sarcomeres. A-F: high-power views of cardiac myocytes treated with either DMSO (A-C) or SB (D-F) and treated with phenylephrine (50 µM; B and E) or electrically paced (C and F). Phenylephrine-treated and electrically paced cells form apparently normal sarcomeres when p38 is inactivated with SB.

View larger version (39K): [in this window] [in a new window]   Fig. 8.   Inhibition of mitogen-activated protein kinases by the dual specificity phosphatase CL100 fails to prevent myofibrillar organization induced by phenylephrine. Myocytes were transiently transfected with expression vectors encoding CL100 (3 µg) and green fluorescent protein (GFP; 1 µg). A: GFP fluorescence. B: phalloidin staining of same field as in A. CL100-containing cell displays similar myofibrillar organization as its nontransfected neighbors indicating that p38 activity is not required for organization. C: a control experiment was conducted in which myocytes were transiently transfected with p38-responsive transcription factor GAL4-CHOP (pFA-CHOP), along with a GAL4-luciferase reporter plasmid (pFR-luciferase) and expression vectors for wild type MEK6 (MKK6wt) or constitutively active MEK6 (MKK6DD) and CL100. CL100 is able to inhibit p38-dependent transcriptional activation. * Significant difference (P < 0.05) between MKK6wt- and MKK6DD-transfected cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A number of diverse stimuli induce hypertrophy of neonatal rat ventricular cardiac myocytes in tissue culture. Some of these include mechanical stretch and angiotensin II (23, 26), ₁-adrenergic agonists (16, 30), growth factors such as endothelin (14, 29), rapid electrical pacing (18, 19), and calcineurin-dependent pathways (20). Recent data from a number of groups have shown that p38 can stimulate the ANF promoter (21, 35, 36, 38). However, it is not clear whether all the above stimuli require p38 for morphological changes and induction of ANF gene expression. Despite the diversity of these stimuli, the end phenotype is very similar, i.e., increased myofibrillar organization and cell size along with induction of fetal cardiac genes such as ANF.

We investigated the role of p38 in induction of gene expression and morphology in response to two physiological stimuli, phenylephrine and rapid electrical pacing. Our results indicate that p38 is required for ANF-driven gene expression stimulated by phenylephrine only in higher density cultures. We speculated that one explanation for our observations is that cells in dense myocyte cultures contract more frequently than in low-density cultures. When we investigated cell cultures that were electrically paced to induce contraction, we found that contraction caused small increases in phosphorylation of p38 but that p38 activity was required for ANF-driven gene expression when assayed 72 h later. Whereas we have found that phenylephrine can activate p38, confirming the results from other investigators (7), our data suggest that chronotropic effects of the drug might also contribute to p38 activation. Sugden and colleagues (7) have found that p38 is induced in varying degrees by different hypertrophic agonists. For example, the phorbol ester 12-O-tetradecanoylphorbol-13-acetate only slightly elevates phospho-p38 levels but strongly activates the ANF promoter and hypertrophic organization. Endothelin also activated p38 but did so less efficiently than phenylephrine. It is possible, therefore, that the different potencies of p38 activation by hypertrophic agonists dictate the precise molecular response of the cell. Nevertheless, it is difficult to reconcile how different degrees of p38 activation can lead to very similar phenotypes if the cellular response is simply a direct result of p38 activity.

Because induction of the ANF promoter by contraction was sensitive to the p38 inhibitor, p38 may be important for gene expression mediated by contraction. Under some conditions, i.e., relatively dense cells that increase their contraction rate in response to phenylephrine treatment, activation of ANF gene expression is highly dependent on the contractile activity, as noted recently by Samarel and his colleagues (8, 9) and as indicated by our unpublished data. Thus we believe that the discrepancy between our results (Fig. 1) and those of other investigators regarding the requirement for p38 activity for phenylephrine-induced ANF expression might be reconciled by concluding that some p38 sensitivity could be mediated through an increased contraction rate. This conclusion implies that, in sparse cultures, the signaling pathways that regulate phenylephrine-induced ANF gene expression might be distinct from those that regulate contraction-induced expression, at least in regard to their dependence on p38. Recent data from our laboratory supports this view, because we have found that some promoter elements that are required for p38-dependent ANF gene expression can be differentiated from those that are required for phenylephrine-induced expression (13).

It is shown in Fig. 4 that, although blocking p38 in contracting cells with the p38 inhibitor reduced ANF-driven gene expression, this inhibition was critically dependent on the time points when SB-203580 was included in the medium. A chronic presence of the inhibitor during hours 1-72 was required to reduce pacing-induced ANF gene expression. Including the p38 inhibitor during times when p38 was increased in its activity (hours 1-24) failed to decrease ANF gene expression when assayed 48 h after the cessation of p38 inhibition. The half-lives of firefly-luciferase mRNA and protein are 6 and 3 h, respectively. If contraction activated a p38-dependent kinase cascade and activation of this cascade was required for gene expression 72 h later, we should find that inhibition in the first 24 h would be sufficient to block ANF expression. However, as shown in Fig. 4, this did not appear to be the case. Alternatively, the ANF promoter might be directly activated by contraction [e.g., by causing the phosphorylation of a transcription factor that is p38 dependent (35)]. If this were responsible for ANF expression, we would find that including the p38 inhibitor for the last 24 h of gene expression would decrease contraction-induced ANF-luciferase expression. However, this was also not the case. Interestingly, although ANF gene expression was unaffected by the p38 inhibitor at these early and late time points in paced cells, a different result was seen in MEK6-transfected cells. Cells transfected with constitutively active MEK6 and the ANF-luciferase construct showed inhibition of ANF-driven gene expression at all time points (early, late, and chronic). Together, these data suggest that p38 activity induced by MKK6 may have both direct and indirect effects on the ANF promoter. Thus, although the activity of p38 is required for gene expression by pacing, it might not be a direct requirement. This model is also supported by our observations that levels of both phospho-p38 and total p38 are decreased in paced samples from hours 24-72. It is difficult to imagine how p38 could be directly activating transcription factors on the ANF promoter when the kinase is present at a low level in the cells.

In this study we also showed that inhibition of p38 using the inhibitor SB-203580 did not significantly alter cell shape and morphological hypertrophy in response to phenylephrine or electrical pacing at any of the cell densities tested. Similar results were obtained when we transfected the potent MAP kinase phosphatase CL100 into the cells. Because two distinct methods of inhibiting p38 both failed to significantly inhibit the hypertrophic morphology of the cells, we conclude that p38 is either dispensable or at best a minor component of the regulatory pathway that controls this phenotype in both electrically paced and phenylephrine-treated cells.