Author + information
- Received April 19, 2004
- Revision received June 23, 2004
- Accepted July 19, 2004
- Published online October 19, 2004.
- Fiona See, BSc (Hons)*,
- Walter Thomas, PhD†,
- Kerrie Way, PhD‡,
- Alex Tzanidis, PhD*,
- Andrew Kompa, PhD*,
- Dion Lewis*,
- Silviu Itescu, MBBS (Hons), FRACP‡ and
- Henry Krum, MBBS, PhD, FRACP*,* ()
- ↵*Reprint requests and correspondence:
Prof. Henry Krum, NHMRC CCREin Therapeutics, Department of Medicine, Monash Medical School University, Alfred Hospital, Commercial Road, Prahran Victoria 3181, Australia
Objectives The aim of this study was to examine the effect of the p38 mitogen-activated protein kinase (MAPK) inhibitor, RWJ-67657 (RWJ), on left ventricular (LV) dysfunction and remodeling post-myocardial infarction (MI) in rats.
Background p38 MAPK signaling has been implicated in the progression of chronic heart failure.
Methods From day 7 post-MI (coronary artery ligation), rats received either RWJ (50 mg/day, by gavage, n = 8, MI+RWJ) or vehicle (by gavage, n = 8, MI+V) for 21 days. Echocardiography was performed on day 6, before the commencement of treatment, and on day 27. In vivo hemodynamic measurements were made on day 28. Sham-operated rats served as controls.
Results The LV end-diastolic pressure and lung/body weight ratio were reduced, whereas the maximum rate of rise of LV pressure was increased towards sham levels in MI+RWJ compared with MI+V. Baseline echocardiographic studies demonstrated uniform LV remodeling and dysfunction in MI rats. Fractional shortening (FS) further deteriorated in MI+V, whereas FS was preserved in MI+RWJ. Progressive LV dilation and infarct expansion observed in MI+V were inhibited in MI+RWJ. MI+RWJ also demonstrated increased myocyte hypertrophy in the peri-infarct and non-infarct zones, and reduced myocardial collagen and α-smooth muscle actin (SMA) immunoreactivity compared with MI+V. The antifibrotic effects of RWJ in vivo may reflect direct effects on cardiac fibroblasts, because RWJ attenuated transforming growth factor β-1–stimulated collagen synthesis and α-SMA expression in isolated cardiac fibroblasts. RWJ also protected cultured myocytes from hydrogen peroxide-induced apoptosis.
Conclusions RWJ-67657 treatment post-MI had beneficial effects on LV remodeling and dysfunction, supporting a key role for p38 MAPK in pathologic cell signaling in these processes and its inhibition as a novel therapy.
Myocardial infarction (MI) impacts on cardiac output via loss of functional myocardium at the site of injury. The consequences of MI include disturbed loading conditions within the heart, ischemic and oxidative stresses, and activation of various local and systemic neurohormonal systems, such as the renin-angiotensin-aldosterone, endothelin, and sympathetic nervous systems (1). These alterations to the extracellular environment trigger left ventricular (LV) remodeling characterized by necrosis and thinning of the infarcted myocardium, LV chamber dilation, fibrosis both at the site of infarct and in the non-infarcted myocardium, and hypertrophy of viable myocytes. Although early remodeling may be adaptive and sustain LV function in the short term, persistent remodeling may contribute to functional decompensation and eventually the development of the clinical syndrome of heart failure (2). Therefore, identifying the cellular and molecular substrates underlying LV remodeling post-MI is an important strategy in the development of novel therapies to impede heart failure disease progression.
Although the extracellular stimuli activated in response to MI may be diverse, there is emerging evidence that these stimuli may converge upon common intracellular pathways to regulate gene expression and protein function. One such intracellular pathway is the p38 mitogen-activated protein kinase (MAPK) signaling cascade. p38 MAPK is activated in cardiac cells by a range of extracellular stimuli including ischemia (3), hemodynamic stress (4), and neurohormonal factors such as angiotensin II (AngII) (5), endothelin and phenylephrine (6). In vitro, p38 MAPK may regulate myocyte apoptosis (7), hypertrophy (8), inflammation (9), and fibroblast proliferation (5). These p38 MAPK-dependent processes characterize the cellular sequelae post-MI, suggesting that p38 MAPK activation may contribute to progressive LV remodeling post-MI and the transition to heart failure.
Myocardial p38 MAPK is rapidly activated in rats after MI induced by coronary artery ligation (10). Sustained p38 MAPK activation in the heart has been associated with LV remodeling and dysfunction arising from various etiologies both in humans (11) and in animal models (12,13). Cardiac myocyte-specific activation of p38 MAPK in transgenic mice results in LV remodeling marked by interstitial fibrosis, myocyte hypertrophy, systolic and diastolic dysfunction, and ultimately premature death (12). p38 MAPK activation is also implicated in hypertensive cardiac hypertrophy and end-organ damage in spontaneously hypertensive, stroke-prone rats maintained on a high-salt/high-fat diet (13). Long-term pharmacologic blockade of p38 MAPK in these rats reduced hypertrophy and dysfunction and enhanced survival over an 18-week period compared with untreated animals. In the failing human heart owing to ischemic injury, inhibition of p38 MAPK has been shown to improve ex vivo function of ischemic myocardium (14). Together, the data indicate that p38 MAPK may be a common pathway for LV remodeling and dysfunction, and heart failure disease progression arising from diverse etiologic origins.
Pharmacologic inhibitors of p38 MAPK have played an important role in investigations of the significance of this signaling pathway in disease (3,13,15). RWJ-67657 (RWJ) is a highly selective inhibitor of the p38-α and p38-β MAPK isoforms (16,17). Similar to other small molecule inhibitors of p38 MAPK, RWJ contains a pyrindinyl imidazole group, and its mechanism of action is likely to involve blockade of p38 MAPK phosphorylation by upstream kinases and/or autophosphorylation via competition for an adenosine triphosphate binding pocket. The therapeutic potential of RWJ in the treatment of cytokine-mediated inflammatory diseases is currently being investigated (18–20). Although RWJ has not previously been evaluated for the treatment of cardiac disease, existing data on its efficacy, selectivity, and safety in humans and animals (16,17,21) indicate its usefulness as a tool in investigating the role of p38 MAPK in heart failure disease progression.
Based on these observations, we hypothesized that inhibition of the p38 MAPK signaling cascade would attenuate LV dysfunction and pathologic LV remodeling post-MI. Therefore, we evaluated the effect of RWJ on parameters of LV structure and function after coronary artery ligation in rats. We also sought to explore potential mechanisms that may underlie these effects.
This study was conducted with the approval of the Monash University Alfred Hospital Animal Ethics Committee and complied with the “Position of the American Heart Association on Research Animal Use,” adopted by the American Heart Association on November 11, 1984.
Outbred female Sprague-Dawley rats (200 to 220 g, Precinct Animal Centre, Alfred Medical Research and Education Project,Melbourne, Australia) were anesthetized with isoflurane (3%) and underwent either sham surgery or ligation of the left anterior descending coronary artery to induce MI (22).
On day 6 after surgery MI animals, which satisfied electrocardiographic criteria for the presence of an anterior wall infarct affecting ≥30% of the LV (23), were randomized to two treatment groups. One group received the p38 MAPK inhibitor, RWJ (kind gift of Scott Wadsworth, PhD, Johnson & Johnson Pharmaceutical Research & Development, L.L.C.), by daily gavage (50 mg/kg/day in acidified 0.5% methylcellulose; n = 8); the second group received an equivalent volume of acidified 0.5% methylcellulose as vehicle control (n = 8; Sigma, St. Louis, Missouri).The dose of RWJ was based on data provided by the manufacturer, which indicated that oral administration of this compound at a dose of 50 mg/kg/day to rats for three weeks inhibited the development of a chronic inflammatory disease without observable side effects (RWJ-67657 Investigator's Brochure, Johnson & Johnson Pharmaceutical Research & Development, L.L.C.). Sham-operated control animals also received vehicle (n = 8). Animals were treated from day 7 until day 28 post-MI. This delay in the commencement of the treatment regimen aimed to inhibit pathologic fibrosis occurring in regions remote from the infarct and to avoid interference with early reparative fibrosis at the site of infarct. Reparative scar tissue is established within the first week post-MI in rats, whereas fibrosis in the regions remote from the infarct begins to appear from the second week post-MI (24).
Transthoracic two-dimensional and m-mode echocardiography was performed on days 6 and 27 after surgery on anesthetized rats using a Sonos 5500 12-MHz probe (Agilent Technologies, Palo Alto, California)(25). Anterior and posterior wall thicknesses and LV internal dimensions were measured offline over three consecutive cardiac cycles to calculate LV fractional shortening (FS) (26,27). Infarct size was calculated as a percentage of the lengths of the epicardial and endocardial circumferences occupied by the infarct, measured from two-dimensional images (23).
Hemodynamics and tissue harvesting
On day 27, intracardiac and peripheral hemodynamic parameters were measured in anesthetized animals (22). Systemic and LV pressure traces were analyzed using Chart v3.5 (ADInstruments, Castle Hill, NSW, Australia)to derive mean arterial pressure, diastolic blood pressure, systolic blood pressure, left ventricular end-diastolic pressure (LVEDP), and maximum rate of rise of LV pressure (+dP/dtmax). After hemodynamic measurements, the heart and lungs were excised, rinsed in cold 0.9% saline, and weighed. Transverse sections of the heart were fixed in Tissue-Tek O.C.T.-embeddingcompound (Sakura Finetek, Chuo-ku, Tokyo, Japan) for histology or snap-frozen for subsequent molecular analyses.
Immunohistochemistry for collagen types I and III in tissue
Using a three-layer immunoperoxidase technique, paraformaldehyde-fixed tissues were stained for collagen type I or collagen type III and quantitatively assessed as previously described (22).
Western blot analysis of α-smooth muscle actin (α-SMA)
Myocardial tissue protein extracts (20 μg) were resolved by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, then transferred to polyvinylidene fluoride membrane. Membranes were incubated with a monoclonal anti-α-SMA antibody (1:10,000; Sigma)followed by a horseradish peroxidase-conjugated rabbit anti-goat immunoglobulin G (Dako, Glostrup, Denmark). Immunoreactive bands were detected using enhanced chemiluminescence (Amersham, Piscataway, New Jersey).
Measurement of myocyte cross-sectional area
The perimeters of 25 myocytes in each of the non-infarcted myocardium and in the peri-infarct regions in collagen III-stained sections were traced under ×400 magnification using computer-based image analysis (ImagePro Plus 3.0, Media Cybernetics, Silver Spring, Maryland). These data were used to calculate the myocyte mean cross-sectional area in each region of the heart.
Neonatal rat cardiac cell cultures
Cardiac cells were isolated from one- to three-day-old neonatal Sprague-Dawley rats and cultured as described previously (28). Myocytes were seeded in Dulbecco's modified Eagle medium (DMEM)/F12 supplemented with 10% horse serum and then changed 24 h later to define media, which prevents the growth of proliferating cells. Fibroblasts were cultured in DMEM/10% fetal calf serum and used at passages 1 to 3 to ensure relative homogeneity in cell type. After seeding, fibroblasts were allowed to adhere overnight, then serum-deprived for 24 h in DMEM/0.5% bovine serum albumin supplemented with 0.15 mmol/l ascorbic acid before treatment.
Collagen synthesis by cardiac fibroblasts
Neonatal rat cardiac fibroblast collagen synthesis stimulated by transforming growth factor (TGF)-β1 (4 nmol/l; R&D Systems, Minneapolis, Minnesota) in the presence of RWJ (500 pmol/l to 5 μmol/l) was estimated by measuring cellular 3H-proline incorporation as previously described (29).
Collagen gene expression in cardiac fibroblasts
The TGF-β1–stimulated α1(I) and α1(III) procollagen gene expression in neonatal rat cardiac fibroblasts cultured in the presence of RWJ (50 nmol/l) were quantified by Northern blot analysis (22). Procollagen messenger ribonucleic acid (RNA) levels were standardized to the housekeeping gene, glyceraldehyde-3-phosphate dehydrogenaseto control for quantitative variability of isolation, transfer, or loading of total RNA.
α-SMA expression in cardiac fibroblasts
The α-SMA expression was measured in cardiac fibroblasts cultured in the presence of RWJ (50 nmol/l) and TGF-β1 (4 nmol/l) by Western blotting as described previously.
Myocytes were pretreated with RWJ for 30 min before the addition of 250 μmol/l hydrogen peroxide (H2O2) for 24 h. Deoxyribonucleic acid (DNA) fragmentation in myocytes was assessed by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) using a peroxidase-based FragEL DNA Fragmentation Detection Kit (Oncogene, Darmstadt, Germany) followed by double-staining of myocytes with anti-desmin antibody (1:100; Dako). Cells were counted in ten random fields per sample.
Selectivity of RWJ for p38 MAPK
The effects of RWJ on the activation of extracellular signal regulated protein kinase (ERK)1/2 and p38 MAPK and hypertrophy in neonatal rat ventricular myocytes were examined. Myocyte cultures were infected with an adenovirus coding for the rat angiotensin type 1 receptor (MOI 20) for two days (28). These cells with then treated with 1 μmol/l RWJ for 30 min before stimulation with AngII (100 nmol/l). For hypertrophy, cells were cultured for an additional 72 h and protein content was determined in cell extracts as previously described (28). For determination of MAPK activation, stimulation was for 5 min, and cell extracts were resolved on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, then Western blotted for activated p38 MAPK (1:2,000 anti-phospho-p38 MAPK; Cell Signaling, Beverly, Massachusetts) or ERK (1:2,000 anti-phospho-p44/p42 MAPK; Cell Signaling) and were developed using anti-mouse immunoglobulin G-horseradish peroxidase and chemiluminescence.Blots were re-probed for total ERK1/2 and p38 MAPK to confirm equivalent loading.
Results are expressed as mean ± SEM. All data sets were analyzed by one-way analysis of variance (Kruskal-Wallis analysis of variance by ranks) followed by Bonferroni's post hoc comparisons between groups. A value of p < 0.05 was considered statistically significant.
Hemodynamic parameters and organ weights
Vehicle-treated MI animals demonstrated the characteristic profile of intracardiac hemodynamic parameters observed post-MI, with a two-fold elevation in LVEDP and a 33% reduction in +dP/dtmaxcompared with shams (p < 0.05) (Table 1).Mean arterial pressure tended to be lower in vehicle-treated MI compared with shams, but it was not statistically different between these two groups. In comparison to shams, lung weight to body weight ratio and heart weight to body weight ratios were increased by 55% and 17%, respectively, in MI rats treated with vehicle (p < 0.05). In contrast to vehicle treatment of MI animals, RWJ treatment resulted in a significantly lower LVEDP (−50%, p < 0.05), higher +dP/dtmax(+66%, p < 0.05), and reduced lung weight to body weight ratio (−25%, p < 0.05). Mean arterial pressure and heart weight to body weight ratios were not significantly different between vehicle-treated and RWJ-treated MI rats. In a separate study, RWJ had no significant effect on peripheral and intracardiac hemodynamic parameters in sham-operated rats compared with vehicle treatment (mean arterial pressure [mm Hg] in sham+vehicle vs. sham+RWJ: 129.8 ± 4.6 vs. 126.2 ± 3.4, p = NS; LVEDP [mm Hg]: 2.59 ± 1.3 vs. 3.1 ± 0.7, p = NS; +dP/dtmax[mm Hg ms−1]: 5.8 ± 0.3 vs. 5.8 ± 0.2, p = NS).
Myocardial infarction rats demonstrated impaired contractile function, indicated by a significant 52 ± 1.2% reduction in FS compared with shams (p < 0.05) (Table 2).This impairment was associated with a large sized infarct (33.7 ± 0.4% of the LV circumference) and LV enlargement, with increased LV internal dimensions in both diastole and systole compared with shams (diastole: +35.2 ± 2.3%; systole: +96.0 ± 5.0%; p < 0.05 MI+V vs. shams). Importantly, animals randomized to either the vehicle treatment or the RWJ treatment groups were well-matched at baseline, before the commencement of treatment, for LV dysfunction and dimensions and infarct size (MI+V vs. MI+RWJ, p = NS).
Change after treatment
Whereas FS did not change in sham-operated rats, contractile function further deteriorated in MI rats treated with vehicle, indicated by a decrease in FS (−24.2 ± 2.4%) from day 6 to day 27 (Fig. 1a).In contrast, MI rats treated with RWJ demonstrated an increase in FS over the treatment period (−9.9 ± 3.2%, p < 0.05 vs. MI+V), resulting in a significantly higher FS compared with the vehicle-treated MI group at day 27 (Table 3).
Infarct size increased by 17.1 ± 3.0% in vehicle-treated MI rats from day 6 to day 27 (Fig. 1b), resulting in a significantly larger infarct size at the end of the study compared with baseline (percent of LV circumference occupied by infarct on day 6: 33.7 ± 0.38% vs. day 27: 39.47 ± 0.24%, p < 0.05). In comparison to vehicle treatment, infarct expansion was significantly attenuated in MI rats treated with RWJ, in which infarct size increased by 2.6 ± 1.1% over the treatment period (p < 0.05 vs. MI+V). Moreover, RWJ-treated MI rats demonstrated significantly smaller infarcts on day 27 compared with MI rats treated with vehicle (Table 3) and preservation of infarct size (day 6: 35.0 ± 0.46% vs. day 27: 35.92 ± 0.23%, p = NS).
Vehicle-treated MI rats demonstrated LV dilation from day 6 to day 27, with increases in both LV internal diastolic dimensions (+15.1 ± 3.2%) (Fig. 1c) and LV internal systolic dimensions (+25.5 ± 4.6%) (Fig. 1d). In contrast to vehicle-treated MI rats, RWJ-treated MI rats demonstrated significantly less LV dilation (diastole: +6.7 ± 1.7%, p < 0.05 vs. MI+V; systole: +8.7 ± 2.1%, p < 0.05 vs. MI+V), so that LV dimensions were not significantly different on day 6 compared with day 27 in this group (LV internal dimensions in diastole [mm]on day 6: 7.5 ± 0.0 vs. day 27: 8.0 ± 0.1, p = NS; LV internal dimensions in systole [mm] on day 6: 5.7 ± 0.8 vs. day 27: 6.1 ± 0.3, p = NS).
In vehicle-treated MI rats, the percentage area of myocardium immunopositive for collagen type I was increased above sham levels throughout the LV infarct (4.6-fold), peri-infarct (3.1-fold), and non-infarct regions (1.4-fold) (Figs. 2and 3a).In contrast, treatment with RWJ attenuated these increases throughout the LV (infarct: 3.1-fold; peri-infarct: 1.9-fold; non-infarct: 0.04-fold; p < 0.05). Collagen type III immunoreactivity was also increased throughout the myocardium after MI but was not significantly affected by RWJ treatment (Figs. 3band 4).
Consistent with these immunohistochemical data, increases in total myocardial collagen content at 28 days post-MI, assessed by tissue hydroxyproline content, tended to be inhibited in MI animals treated with RWJ in both the infarct zone (μg/mg tissue sham: 1.49 ± 0.14; MI+V: 22.31 ± 1.11; MI+RWJ: 14.15 ± 4.15; p < 0.05, MI+V vs. shams; p = 0.08, MI+RWJ vs. MI+V) and non-infarct zone (sham: 1.58 ± 0.12; MI+V: 3.27 ± 0.18; MI+RWJ: 2.65 ± 0.10; p < 0.05, MI+V vs. shams; p = 0.06, MI+RWJ-67657 vs. MI+V) (n = 4).
Expression of the myofibroblast marker, α-SMA, was increased 2.2 ± 0.2-fold in the infarct zone and 2.5 ± 0.9-fold in the non-infarct zone compared with shams in vehicle-treated MI animals (p < 0.05) (Figs. 5a and 5b).In contrast, MI animals treated with RWJ demonstrated significantly less α-SMA expression in both the infarct (1.1 ± 0.3-fold, p < 0.05 vs. MI+V) and non-infarct regions (1.7 ± 0.7-fold, p < 0.05 vs. MI+V).
Myocyte cross-sectional areas in the peri-infarct and infarct regions increased in vehicle-treated MI rats relative to shams by 1.8 ± 0.3-fold and 1.4 ± 0.2-fold, respectively (p < 0.05) (Fig. 6).In MI rats treated with RWJ, this hypertrophic effect was augmented in both the peri-infarct and the non-infarct regions (p < 0.05), indicated by significantly greater relative myocyte cross-sectional areas in comparison to vehicle-treated MI rats (peri-infarct: 4.1 ± 1.5-fold; non-infarct zone: 1.6 ± 0.2-fold, p < 0.05 vs. MI+V).
Collagen synthesis by neonatal cardiac fibroblasts in vitro
Neonatal rat cardiac fibroblast collagen synthesis, estimated by 3H-proline incorporation, was upregulated by 73% in response to TGF-β1 stimulation (Fig. 7a).This was suppressed in a dose-dependent manner by co-treatment of cells with RWJ (500 pmol/l: −0.9%; 5 nmol/l: −15.5%; 50 nmol/l: −21%; 500 nmol/l: −25%; 5 μmol/l: −42%; p < 0.05 vs. TGF-β1 alone). RWJ had no significant effect on cellular proliferation in the concentration range of 500 pmol/l to 50 nmol/l, assessed by 3H-thymidine incorporation assay under identical conditions to the 3H-proline assay (data not shown). Above this dose range, RWJ had suppressive effects on cellular proliferation. Therefore, a dose of 50 nmol/l RWJ was used in subsequent gene expression studies, because this dose inhibited collagen synthesis independent of effects on cellular proliferation.
The TGF-β1 induced a two-fold increase in α1(I) procollagen gene expression by neonatal rat cardiac fibroblasts at 24 h in comparison to unstimulated cells (p < 0.05) (Fig. 7b). This effect was significantly attenuated by co-treatment of the cells with 50 nmol/l RWJ (1.2-fold increase over untreated) (p < 0.05, vs. untreated). At this dose, RWJ had no significant effect on basal levels of α1(I) procollagen expression.
TGF-β1–induced α-SMA expression in cardiac fibroblasts
The TGF-β1 stimulation of neonatal rat cardiac fibroblasts resulted in a 2.4 ± 0.5-fold induction of α-SMA expression above levels observed in quiescent cells as assessed by Western blot analysis (p < 0.05) (Fig. 8).Whereas RWJ had no significant effect on baseline levels of α-SMA expression (1.1 ± 0.3-fold increase, p = NS vs. unstimulated), the presence of RWJ in culture media significantly reduced α-SMA expression stimulated by TGF-β-1 (1.2 ± 0.1-fold increase, p < 0.05 vs. unstimulated).
H2O2-induced myocyte apoptosis
H2O2(250 nmol/l) induced a significant increase in the proportion of TUNEL-positive cells in culture compared with untreated cells (unstimulated: 0.6 ± 0.4% vs. H2O2: 25 ± 6.7%, p < 0.05) (Fig. 9).Although RWJ had no significant effect on basal levels of apoptosis, RWJ reduced the proportion of H2O2-induced TUNEL-positive cells at both concentrations examined (500 nmol/l: 7.8 ± 2.2%; 1 μmol/l: 3.9 ± 1.7%; p < 0.05 vs. H2O2alone).
Selectivity of RWJ for p38 MAPK
Angiotensin II stimulated a marked increase in phosphorylated p38 MAPK and ERK1/2 in myocytes overexpressing AT1 receptor. Consistent with previous reports characterizing this compound (16), RWJ inhibited p38 MAPK activation independent of the effects on total p38 MAPK, and with negligible effects on phosphorylated and total ERK1/2 expression (Fig. 10).RWJ had no significant effect on AngII-induced myocyte hypertrophy, an effect known to be dependent on ERK1/2 signaling (data not shown).
This is the first study to demonstrate that treatment of rats from seven days after MI with RWJ attenuates the natural progression of pathologic LV remodeling and dysfunction subsequent to MI. Specifically, in comparison to treatment with vehicle, treatment of MI rats with RWJ impeded the development of chronic heart failure (manifested by reduced lung weight to body weight ratio), preserved LV function and dimensions, and inhibited infarct expansion. These ultrastructural and functional changes were associated with reduced myocardial collagen type I and α-SMA expression. RWJ treatment of MI rats augmented myocyte hypertrophy in the peri-infarct and non-infarct regions. The antifibrotic effects observed in vivo appeared to be attributable to direct effects of RWJ on cardiac fibroblasts. In cultured cardiac fibroblasts, RWJ suppressed TGF-β1–stimulated collagen synthesis, α1(I) procollagen gene expression, and expression of α-SMA. Additionally, RWJ protected cultured myocytes against hydrogen peroxide-induced apoptosis. Together, these mechanisms may contribute to the preservation of LV structure and function observed with RWJ treatment after MI in rats. Given the selectivity of this compound for p38 MAPK inhibition, demonstrated in myocytes over-expressing the AT1receptor, the findings of this study imply that p38 MAPK plays an important role in the progression of LV remodeling and dysfunction after MI.
P38 MAPK-mediated mechanisms of LV remodeling and dysfunction
Various cellular processes pertinent to LV remodeling and dysfunction post-MI rely on p38 MAPK signaling, including myocyte apoptosis (7), hypertrophy (6,8), and inflammation (9). Whether p38 MAPK plays a direct role in fibrosis, a hallmark of LV remodeling and dysfunction post-MI, has not been previously examined. In the current study, RWJ treatment of MI animals reduced collagen type I accumulation and α-SMA expression throughout the myocardium. Whereas fibrosis at the site of infarct is reparative, excessive deposition of collagen in regions remote from the infarct is believed to be pathologic. In particular, the accumulation of collagen type I, which is characterized by tensile strength, may contribute to myocardial stiffness by limiting the motion of myocytes, and may promote arrhythmias by electrical isolation of adjacent myocytes (30). Therefore, the inhibitory effect of RWJ on collagen type I, in preference to the more elastic collagen type III, observed in the present study may provide a molecular substrate for the beneficial effects on LV structure and function observed with RWJ treatment post-MI. The reasons for the differential effects of RWJ on deposition of collagen type I versus collagen type III are unclear. Although collagen types I and III are co-expressed in the heart, separate control of the respective procollagen genes may be possible owing to differences in their promoter regions, such as the occurrence of positive and negative regulatory elements (31). However, whether these differences impact on the binding of transcription factors that may be activated by p38 MAPK and thus give rise to discrete regulation of collagen I and III expression remains speculative and awaits comprehensive investigation.
Our in vitro data provide evidence that the antifibrotic effects of RWJ treatment post-MI may be independent of the improvements in hemodynamic status also observed in this study. RWJ had suppressive effects on TGF-β1–stimulated collagen synthesis and α1(I) procollagen gene expression in isolated neonatal cardiac fibroblasts. These findings are consistent with previous reports demonstrating that p38 MAPK may mediate TGF-β1–induced collagen expression in renal mesangial cells (32,33). Our data further point to a potential mechanism for p38 MAPK regulation of collagen expression by cardiac fibroblasts. The TGF-β1 plays a key role in the phenotypic switch of fibroblasts to myofibroblasts, which express α-SMA and demonstrate an increased capacity for extracellular matrix protein synthesis (34). In this study, myocardial α-SMA expression was found to be increased in vehicle-treated MI rats compared with that observed in shams at four weeks post-MI, a time when TGF-β1 expression is also upregulated (35). In contrast to vehicle treatment, RWJ attenuated α-SMA expression in both the infarct and non-infarct regions. In addition, TGF-β1 stimulation of α-SMA expression in isolated cardiac fibroblasts was inhibited in the presence of RWJ compared with cells treated with TGF-β1 alone. Together, these findings suggest that p38 MAPK may be involved in TGF-β1–induced myofibroblast activation and may thus contribute directly to collagen expression and fibrosis post-MI.
Compensatory LV hypertrophy in response to injury such as MI or increased hemodynamic loading conditions may sustain or even improve cardiac output. In this study, RWJ treatment of rats post-MI augmented hypertrophy of the surviving myocytes, indicated by a further increase in myocyte cross-sectional area in the peri-infarct and non-infarct regions, compared with vehicle treatment. This observation indicates an antagonistic role for p38 MAPK in myocyte growth in the post-MI context, and is consistent with recent studies demonstrating that targeted inhibition of p38 MAPK promotes myocyte hypertrophy (36,37). In contrast, Behr et al. (13) reported that SB239063, a selective p38 MAPK inhibitor, attenuated LV hypertrophy in stroke-prone, spontaneously hypertensive rats fed a high-salt/high-fat diet. A possible explanation for these divergent results is that myocyte survival may be differentially regulated by the p38-α and p38-β isoforms. Whereas myocyte apoptosis and growth inhibition are attributable to activation of p38-α, p38-β activation appears to mediate a survival pathway, involving promotion of myocyte growth and reduction of apoptosis (8). Because RWJ inhibits both p38-α and -β isoforms (20), the results of the present study may indicate that the p38-α pathway is preferentially activated in the post-MI setting. Therefore, RWJ treatment post-MI may enhance myocyte growth as an adaptive response to extracellular stress and thus contribute to the observed improvements in LV function compared with vehicle-treated MI rats. Because we observed that RWJ had no significant effect on AngII-induced myocyte growth in vitro, it appears that alternative stimuli reliant on p38 MAPK signaling may be involved in myocyte hypertrophy post-MI.
RWJ also had a protective effect against apoptosis in isolated myocytes exposed to hydrogen peroxide, which mimics oxidative stress in conditions of ischemia. Consistent with our data, previous studies have shown that p38 MAPK inhibition reduces apoptosis in cultured myocytes exposed to oxidative stress (38), whereas p38 MAPK downregulation in transgenic mice confers protection against apoptotic death of myocytes induced by ischemia/reperfusion injury (39). After MI, myocyte apoptosis occurs both acutely in the infarcted region and later in regions remote from the infarct, and has been shown to correlate with LV dysfunction (40). Therefore, the inhibitory effects of RWJ treatment of MI rats on LV dysfunction and remodeling may derive from its anti-apoptotic actions and the consequent enhancement of myocyte survival.
It is possible that the beneficial effects of RWJ treatment on LV function may be independent of the effects on LV remodeling. For example, specific pharmacologic inhibition of p38 MAPK activity can reversibly block the negative-inotropic effects of p38 MAPK activation in cultured adult rat cardiomyocytes (41). However, RWJ had no effect on LV function in sham-operated rats, and our in vitro studies support a role for p38 MAPK in fibrosis and myocyte apoptosis. Collectively, these data demonstrate that the effects of RWJ on LV function are related to direct effects on post-MI remodeling.
p38 MAPK as a therapeutic target post-MI
The results of this study suggest that inhibition of p38 MAPK is a promising therapeutic strategy to impede heart failure disease progression after MI. The anti-remodeling effects of current treatments for heart failure, such as angiotensin-converting enzyme inhibitors (42) and β-blockers (43,44), are postulated to be key mechanisms underlying improvements in parameters of morbidity and mortality observed with such treatments. In the current study, the beneficial effects on LV remodeling and dysfunction with p38 MAPK inhibition post-MI were associated with a decrease in lung/body weight ratio, a marker of “clinical” chronic heart failure. However, larger and longer-term studies are required to determine whether these benefits of p38 MAPK inhibition post-MI translate into enhanced survival.
Whereas there is strong evidence demonstrating the potency of RWJ against p38 MAPK (16,17), at high concentrations the compound also possesses inhibitory activities against other kinases, such as c-src, p56 lck, and ERK-2 (17). Non-specific activity has also been reported for other p38 MAPK inhibitors (45,46). However, it is important to note that the potency of such non-specific inhibitory actions of RWJ is relatively small compared with its actions against p38 MAPK. For example, it is 20-fold less potent for inhibition of JNK2 compared with p38-α (personal communication, Scott Wadsworth, PhD, Johnson & Johnson Pharmaceutical Research & Development, L.L.C., February 22, 2004).Therefore, despite the possibility of non-specific inhibition of other kinases by RWJ, the observations reported in this study are likely to be primarily attributable to p38 MAPK inhibition.
In conclusion, the results of the present study demonstrate that treatment with a p38 MAPK inhibitor, RWJ, from seven days after MI attenuated pathologic cardiac remodeling and LV dysfunction. The findings of the present study therefore suggest that the p38 MAPK signaling cascade may be an important pathway in the progression of LV dysfunction and pathologic remodeling after MI. Furthermore, p38 MAPK inhibition may be a useful treatment strategy in impeding heart failure disease progression.
The authors would like to thank Dr. Ross Hannan (Baker Heart Research Institute) for providing neonatal rat cardiac fibroblasts.
Fiona See is supported by an NHMRC Dora Lush Biomedical Postgraduate Research Scholarship (Canberra, Australia). This study was partially funded by Johnson & Johnson Pharmaceutical Research & Development, L.L.C.
- Abbreviations and acronyms
- angiotensin II
- Dulbecco's modified Eagle medium
- maximum rate of rise of left ventricular pressure
- extracellular signal regulated protein kinase
- fractional shortening
- left ventricle/ventricular
- left ventricular end-diastolic pressure
- mitogen-activated protein kinase
- myocardial infarction
- smooth muscle actin
- transforming growth factor
- terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling
- Received April 19, 2004.
- Revision received June 23, 2004.
- Accepted July 19, 2004.
- American College of Cardiology Foundation
- Swynghedauw B.
- Pfeffer M.A.,
- Braunwald E.
- Ma X.L.,
- Kumar S.,
- Gao F,
- et al.
- Fischer T.A.,
- Ludwig S.,
- Flory E,
- et al.
- Clerk A.,
- Michael A.,
- Sugden P.H.
- MacKay K.,
- Mochly-Rosen D.
- Wang Y.,
- Huang S.,
- Sah V.P,
- et al.
- Gao F.,
- Yue T.L.,
- Shi D.W,
- et al.
- Shimizu N.,
- Yoshiyama M.,
- Omura T,
- et al.
- Liao P.,
- Georgakopoulos D.,
- Kovacs A,
- et al.
- Behr T.M.,
- Nerurkar S.S.,
- Nelson A.H,
- et al.
- Schneider S.,
- Chen W.,
- Hou J.,
- Steenbergen C.,
- Murphy E.
- Wadsworth S.A.,
- Cavender D.E.,
- Beers S.A,
- et al.
- Faas M.M.,
- Moes H.,
- Fijen J.W.,
- Muller Kobold A.C.,
- Tulleken J.E.,
- Zijlstra J.G.
- Schafer P.H.,
- Wadsworth S.A.,
- Wang L.,
- Siekierka J.J.
- Tzanidis A.,
- Lim S.,
- Hannan R.D.,
- See F.,
- Ugoni A.M.,
- Krum H.
- Litwin S.E.,
- Katz S.E.,
- Morgan J.P.,
- Douglas P.S.
- Thomas W.G.,
- Brandenburger Y.,
- Autelitano D.J.,
- Pham T.,
- Qian H.,
- Hannan R.D.
- Tzanidis A.,
- Hannan R.D.,
- Thomas W.G,
- et al.
- Mudryj M.,
- de Crombrugghe B.
- Chin B.Y.,
- Mohsenin A.,
- Li S.X.,
- Choi A.M.,
- Choi M.E.
- Gruden G.,
- Zonca S.,
- Hayward A,
- et al.
- Kaiser R.A.,
- Bueno O.F.,
- Lips D.J,
- et al.
- Palojoki E.,
- Saraste A.,
- Eriksson A,
- et al.
- Liao P.,
- Wang S.Q.,
- Wang S,
- et al.
- Pfeffer M.A.,
- Braunwald E.,
- Moye L.A,
- et al.
- Groenning B.A.,
- Nilsson J.C.,
- Sondergaard L.,
- Fritz-Hansen T.,
- Larsson H.B.,
- Hildebrandt P.R.
- Doughty R.N.,
- Whalley G.A.,
- Walsh H.A,
- et al.