- Email: [email protected]
HB-EGF induces cardiomyocyte hypertrophy via an ERK5-MEF2A-COX2 signaling pathway
Cellular Signalling 23 (2011) 1100–1109
Contents lists available at ScienceDirect
Cellular Signalling j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c e l l s i g
HB-EGF induces cardiomyocyte hypertrophy via an ERK5-MEF2A-COX2 signaling pathway Kuy-Sook Lee, Jin-Hee Park, Hyun-Joung Lim, Hyun-Young Park ⁎ Division of Cardiovascular and Rare Diseases, Center for Biomedical Sciences, National Institute of Health, Seoul, Korea
a r t i c l e
i n f o
Article history: Received 8 September 2010 Received in revised form 23 December 2010 Accepted 10 January 2011 Available online 16 January 2011 Keywords: HB-EGF MEK5–ERK5 MEF2A COX-2 Cardiomyocyte hypertrophy
a b s t r a c t Heparin-binding EGF-like growth factor (HB-EGF) is a member of the EGF family that binds to and activates the EGF receptor. Transactivated by angiotensin II, ET-1, and various growth factors in cardiomyocytes, HBEGF is known to induce cardiac hypertrophy via the PI3K-Akt, MAP kinase, and JAK–STAT pathways. However, little is known about the potential involvement of the ERK5 pathway in HB-EGF-induced cardiac hypertrophy. In the present report, we identify and characterize a novel MEK5–ERK5 pathway that is involved in HB-EGFinduced cardiomyocyte hypertrophy. HB-EGF (10 ng/ml) significantly increased [3H]-leucine incorporation and atrial natriuretic factor (ANF) mRNA expression in H9c2 cells. In addition, HB-EGF activated a MEK5– ERK5 pathway. Pretreatment with the EGFR inhibitor AG1478 attenuated the activation of ERK5. Blockade of MEK5–ERK5 signaling using MEK5 siRNA reduced the ability of HB-EGF to increase cell size and the expression of ANF mRNA, suggesting the involvement of an EGFR–ERK5 pathway in HB-EGF-induced cardiomyocyte hypertrophy. We further analyzed cyclooxygenase-2 (COX-2). HB-EGF enhanced the expression of COX-2, a response mediated by MEK5–ERK5 signaling, while the COX-2 inhibitor rofecoxib attenuated HB-EGF-induced ANF mRNA expression, suggesting that COX-2 is also associated with HB-EGFinduced cardiomyocyte hypertrophy. It has been known that ERK5 activates the myocyte enhancer factor (MEF) 2 family of transcription factor, we next tested whether activation of MEF2A contributes to HB-EGFinduced COX-2 expression. Inhibition of MEF2A using siRNA attenuated HB-EGF-induced COX-2, ANF expression and cell size. In conclusion, HB-EGF induces cardiomyocyte hypertrophy through an EGFR–ERK5– MEF2A–COX-2 pathway. Our findings will help us to better understand the molecular mechanisms behind HB-EGF-induced cardiomyocyte hypertrophy. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Heparin-binding EGF-like growth factor (HB-EGF), a member of the EGF family that binds to and activates the EGF receptor (EGFR), is expressed in a variety of tissues, notably lung, heart, brain and skeletal muscle [1–4]. HB-EGF is well known to be a potent mitogen for several cell types including fibroblasts, smooth muscle cells, keratinocytes, and breast carcinoma cells [5]. Furthermore, HB-EGF is expressed in heart muscle cells, in which HB-EGF mRNA levels have been shown to be markedly increased by the α-adrenergic agonist phenylephrine, which is known to induce myocardial hypertrophy [6]. In addition, HB-EGF induced a hypertrophic response in rat cardiac myocytes, which suggests that it acts as an autocrine hypertrophic factor [6]. HBEGF is known to activate cell signal transduction networks by auto-
⁎ Corresponding author. Tel.: +82 43 719 8650; fax: +82 43 719 8689. E-mail address: [email protected] (H.-Y. Park). 0898-6568/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2011.01.006
phosphorylation of the EGF receptor, which consequently recruits adapter signaling molecules. MAP kinase, PI3K–Akt, and JAK-STAT pathways are well-known to form part of these cell signal networks [7–9]. MAP kinase pathways play a central role in the signaling network that transduces extracellular stimuli into distinct intracellular responses. Four major MAP kinase pathways have been identified in mammalian cells. The three canonical MAP kinase pathways – ERK1/2, JNK, and p38 MAP kinase – mediate proliferative and stress signaling in various cells [10]. The fourth is Big MAP kinase 1 (BMK1) or ERK5, which was cloned by Lee et al. and Zhou et al. and is activated in response to growth factors and stress [11,12]. Activation of this ERK5 cascade has been implicated not only in physiological functions such as cell survival, proliferation, and differentiation but also in pathological processes such as carcinogenesis, cardiac hypertrophy, and atherosclerosis [13,14]. Downstream targets of ERK5 signaling include the myocyte enhancer factor (MEF) 2 family of transcriptional factors, which has been closely linked to hypertrophy [15]. Moreover, observed MEF2 activation in multiple models of cardiac hypertrophy supports the view that members of this family may play central roles in the
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
regulation of fundamental signaling mechanisms during myocardial hypertrophic growth [16–18]. Cyclooxygenase-2 (COX-2) is an inducible and rate-limiting enzyme in the biosynthesis of prostaglandins. COX-2 is induced in response to various stimuli, including mitogens, cytokines, growth factors, and inflammatory signals [19]. Specifically, it has been shown to be induced by factors with roles in heart function and pathology, including endothelin-1, EGF, and angiotensin II [20–22]. A recent study demonstrated that cardiac-specific COX-2 expression results in elevated prostaglandin levels and mild hypertrophy. Furthermore, the COX-2 inhibitor celecoxib has been reported to modulate hypertrophic signaling and prevent load-induced cardiac dysfunction [23,24]. Recently, the role of COX-2 in human heart disease was brought to the public attention by the finding that chronic use of the popular COX-2specific inhibitor rofecoxib increased heart disease risk [25–27]. Thus, in the present study, we attempted to identify and characterize the novel pathways involved in HB-EGF-induced cardiomyocyte hypertrophy, focusing initially on the MEK5–ERK5 pathway and its relationships to MEF2 and COX-2. We herein report that HB-EGF induces cardiomyocyte hypertrophy via EGFR-dependent induction of COX-2. We further show that
1101
MEK5–ERK5 signaling increases COX-2 expression through phosphorylation of MEF2A.
2. Materials and methods 2.1. Materials Cell culture reagents and oligo(dT)12–18 were purchased from Invitrogen (Carlsbad, CA), the mouse anti-β-actin monoclonal antibody was obtained from Sigma (St. Louis, MO), and HB-EGF was acquired from R&D systems (Minneapolis, MN). Peroxidase-conjugated anti-rabbit and anti-mouse secondary antibodies were obtained from Jackson ImmunoResearch (West Grove, PA), Accupower RT Premix and Accupower PCR Premix were purchased from Bioneer (Daejeon, Korea), and the Alexa Fluor® 488-conjugated goat antimouse IgG was obtained from Molecular Probes (Eugene, OR). RNeasy mini kits were obtained from Qiagen (Valencia, CA), the MEF2A and phospho-MEF2A antibodies from Abcam (Cambridge, UK), the phospho-MEK5 antibody from IMGENEX (San Diego, CA), the phospho-ERK5 antibody from Cell Signaling Technology (Danvers,
Fig. 1. HB-EGF induces hypertrophy in H9c2 cells. H9c2 cells were stimulated with HB-EGF (10 ng/ml) for 24 h. (A) Total mRNA was isolated and the expression of ANF evaluated by RT-PCR using specific primers. (B) Induction of protein synthesis by HB-EGF was assessed by [3H]-leucine incorporation assay. (C) Immunofluorescent microscopy using an αactinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
1102
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
MA), and the COX-2 antibody from Santa Cruz Biotechnology (Santa Cruz, CA). 2.2. Cell culture Embryonic rat heart-derived H9c2 cells were purchased from ATCC (Manassas, VA) and were maintained in Dulbecco's modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Cells were starved for 4 h in DMEM containing 0.1% FBS prior to experimental treatments. Neonatal rat cardiomyocytes (NRCMs) were prepared as described elsewhere with minor modification [28]. Briefly, hearts harvested from 1 to 3 days old Sprague-Dawley rat (obtained from Orient, Seoul, Korea) were washed in PBS (Mg2+, Ca2+ free). Hearts were minced with collagenase type II (1 mg/ml, 262 U/mg, GibcoBRL) in sterile HBSS and incubated for 5 min. The pellet was collected in HBSS containing collagenase type II and incubated for 5 min. The supernatant was collected in α-MEM containing 10% FBS and digestion steps were repeated up to 12 times. Cardiomyocytes were purified from fibroblasts using repeated centrifugation step. Finally, cardiomyocytes were resuspended in α-MEM containing 10% FBS and 0.1 mM BrdU. Cells were then cultured in α-MEM containing 10% FBS at 37 °C for 2– 3 days in a humidified atmosphere containing 5% CO2 and 95 % air. More than 90% of cells were cardiomyocytes (positive for Troponin I and beating feature).
2.6. siRNA transfection Cells were grown to ~ 20–30% confluence in culture media supplemented with 10% FBS. MEK5 or MEF2A siRNA was introduced into H9c2 cells or NRCMs using LipofectamineTM RNAiMAX (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were harvested. Lysates were prepared for analysis by RT-PCR and Western blotting to evaluate the silencing effect of MEK5 and MEF2A siRNAs. MEK5 and MEF2A siRNAs were purchased from Qiagen. The rat-specific MEK5 and MEF2A target sequences were 5′-CAC AGT CTA CAA AGC ATA TCA-3′ and 5′-CAG CTC TAA CAA GTT GTT TCA-3′, respectively. A nonspecific control siRNA (Invitrogen) was used as a negative control. 2.7. Immunocytochemistry H9c2 cells or NRCMs were plated to 4-well Lab-Tek Chamber Slides (Nunc, New York, NY). Cells were starved for 4 h in DMEM
2.3. [3H]-leucine incorporation assay Protein synthesis assays were performed using cells plated to 96well plates at a density of 3.5 × 103 cells/well. Cells were starved in DMEM with 0.1% FBS for 4 h and then incubated with HB-EGF (10 ng/ ml) for 24 h. Next, [3H]-leucine (1 μCi/ml) was added to the culture medium and the cells were incubated for a further 4 h before being transferred to a filter mat (PerkinElmer, MA) using a Tomtec Harvester 96 harvester (Hamden, CT). [3H]-leucine incorporation was measured using a Wallac Microbeta Trilux 1450 Scintillation Counter (PerkinElmer). 2.4. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis Total cellular RNA was extracted from H9c2 cells and NRCMs using an RNeasy mini kit (Qiagen) according to the manufacturer's instructions. An oligo(dT)12–18 primer and Accupower RT Premix were used to synthesize cDNA from 500 ng of total RNA. Target sequences were amplified using the following primers: ANF, 5′-ACC TGC TAG ACC ACC TGG AG-3′ and 5′-GTA CCG GAA GCT GTT GCA G-3′; MEK5, 5′-GCC AAC GGC CAG ATG AAT GAA C-3′ and 5′-CAC GCC GAA GTC ACA CAG CTT G-3′; MEF2A, 5′-CCG CCT CAG AAC TTC TCA ATG-3′ and 5′-TTG GAG AGG CCC TTG AGT TTA C-3′; COX-2, 5′-GGC AAA GGC CTC CAT TGA CCA G-3′ and 5′-GGA GGC ACT TGC GTT GAT GGT G-3′; β-actin, 5′-CCC ATT GAA CAC GGC ATT GTC-3′and 5′-CGC ACG ATT TCC CTC TCA GC-3′. 2.5. Western blot analysis Cells were starved for 4 h in DMEM supplemented with 0.1% FBS prior to experimental treatments. Following treatment, cells washed with ice-cold phosphate-buffered saline (PBS) and harvested, through scraping, in 1× protein lysis buffer (Cell Signaling Technology). Equal amounts of cell lysates were separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE). Membranes were incubated with primary antibody, and immunopositive bands were subsequently visualized using the ECL detection reagent (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK). Each experiment was performed at least three times.
Fig. 2. HB-EGF activates a MEK5–ERK5 pathway in H9c2 cells. (A) Phosphorylation of MEK5 and ERK5 after HB-EGF treatment in H9c2 cells, detected by Western blotting. (B) Effect of AG1478 on HB-EGF-induced phosphorylation of ERK5. Cells were pretreated with AG1478 (250 nM) for 1 h and then the cells were treated with HB-EGF for 5 min. Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
containing 0.1% FBS prior to experimental treatments. Briefly, after two washes in ice-cold PBS, cells were fixed with 4% paraformaldehyde at room temperature for 30 min and then washed three times in PBS. They were then incubated with anti-α-actinin antibody (Santa Cruz Biotechnology) diluted in 1% bovine serum
1103
albumin (BSA)–PBS. Primary antibody was detected with Alexa Fluor® 488-conjugated goat anti-mouse IgG (Molecular Probes). The images observed under a fluorescence microscope. Cell sizes were measured using AxioVision software (Carl Zeiss Inc. Thornwood, NY).
Fig. 3. Involvement of MEK5–ERK5 signaling in HB-EGF-induced cardiomyocyte hypertrophy. (A) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 5 min. Phospho-ERK5 and MEK5 levels were measured by Western blotting. (B) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 24 h. Cell lysates were analyzed by RT-PCR using ANF- and MEK5-specific primers. (C) Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
1104
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
2.8. Statistical analysis Data are presented as the means ± SEM of three or more individual experiments. Significant differences between treatments and controls were identified using Student's t-test. A p value b0.05 was considered to be statistically significant.
Fig. 4 (continued).
3. Results 3.1. HB-EGF induces cardiomyocyte hypertrophy and activates MEK5– ERK5 signaling via the EGF receptor
Fig. 4. COX-2 is involved in HB-EGF-induced cardiomyocyte hypertrophy. Following HB-EGF treatment, cell lysates were prepared as indicated then analyzed for COX-2 expression through (A) RT-PCR and (B) Western blotting. (C) Effect of AG1478 on HBEGF-induced COX-2 expression. Cells were pretreated with AG1478 (250 nM) for 1 h and then the cells were treated with HB-EGF for 1 h. (D) Analysis of ANF mRNA levels in HB-EGF-treated cardiomyocytes in the presence or absence of the COX-2 inhibitors rofecoxib (10 μM). Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by (E) RT-PCR and (F) Western blotting. Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
We examined the effect of HB-EGF on cardiac hypertrophy using a cardiomyocyte cell line, H9c2. The first, we determined whether HB-EGF induced cardiac hypertrophy. Cardiomyocytes were stimulated with HB-EGF (10 ng/ml) for 24 h. As shown in Fig. 1, we observed an important molecular characteristic of cardiac hypertrophy [29] in that atrial natriuretic factor (ANF) mRNA, [3H]-leucine incorporation and cell size were significantly elevated following HBEGF treatment. To determine whether the MEK5–ERK5 pathway contributes to HB-EGF-induced signal transduction in cardiomyocytes, H9c2 cells were stimulated with 10 ng/ml HB-EGF for 5–180 min and cell lysates analyzed by Western blotting using anti-p-MEK5 and anti-pERK5 antibodies. HB-EGF induced MEK5 and ERK5 phosphorylation
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
within 5 min (Fig. 2A). We did additional experiment at time points between 0–5 min. We observed that phosphorylation of ERK5 was significantly increased by HB-EGF at early time points (1 to 5 min after HB-EGF stimulation, Data not shown), and there are no significant different between each time point. Therefore, we determined the time point as 5 min in subsequent experiments for a steady experimental condition. In addition, the EGFR tyrosine kinase inhibitor AG1478 (250 nM) attenuated HB-EGF-mediated activation of ERK5 (Fig. 2B). 3.2. MEK5–ERK5 signaling is involved in HB-EGF-induced cardiomyocyte hypertrophy To determine whether the MEK5–ERK5 signal pathway is involved in HB-EGF-induced cardiomyocyte hypertrophy, we tested the effect of MEK5 siRNA on cardiomyocyte hypertrophy. Transfection of H9c2 cells with MEK5 siRNA (10–100 nM) abolished MEK5 expression within 48 h (Supplemental Fig. 1). MEK5 siRNA was used at a concentration of 20 nM in all subsequent experiments. Forty-eight hours after MEK5 siRNA transfection, H9c2 cells were stimulated with HB-EGF for 5 min. ERK5 activation decreased significantly in cells transfected with MEK5 siRNA (Fig. 3A). In addition, HB-EGF-increased ANF mRNA expression and cell size were significantly attenuated by MEK5 siRNA transfection (Fig. 3B and C). 3.3. HB-EGF induces cardiomyocyte hypertrophy through induction of COX-2 expression via EGFR We further analyzed COX-2, an activator of hypertrophic signaling that lies downstream of ERK5. HB-EGF induced COX-2 mRNA and protein expression within 1 h (Fig. 4A and B). This increase in COX-2 expression was significantly attenuated by pretreatment with AG1478, indicating that HB-EGF increases COX-2 expression via the EGF receptor (Fig. 4C). The COX-2 inhibitor rofecoxib (10 μM) inhibited HB-EGF-induced ANF mRNA expression (Fig. 4D). Furthermore, induction of COX-2 mRNA and protein expression by HB-EGF decreased in cells transfected with MEK5 siRNA (Fig. 4E and F). These findings support the hypothesis that COX-2, acting downstream of MEK5 and ERK5, mediates HB-EGF-induced cardiomyocyte hypertrophy. 3.4. MEK5–ERK5-dependent activation of MEF2A is involved in HB-EGFinduced COX-2 expression Previous studies have reported that ERK5 activation in cardiac hypertrophy affects downstream transcription factors including MEF2 [30]. Thus, we also examined whether HB-EGF activates MEF2A in cardiomyocytes. HB-EGF induced phosphorylation of MEF2A (T319) within 5 min (Fig. 5A). As shown by Western blot analysis of nuclear fractions, HB-EGF induced phosphorylation of MEF2A (T319) within 15 min. This activation of MEF2A was attenuated by transfection with MEK5 siRNA (Fig. 5B). To investigate the possibility that HB-EGF-induced MEF2A activation increases the expression of COX-2 and thereby induces subsequent cardiomyocyte hypertrophy, we next examined the effects of MEF2A siRNA. MEF2A siRNA (10–50 nM) was introduced into H9c2 cells. Cells were harvested 48 h after transfection and lysates analyzed by Western blotting using an anti-MEF2A antibody. MEF2A siRNA (10–50 nM) greatly reduced MEF2A mRNA and protein expression (Supplemental Fig. 2A and B). MEF2A siRNA was used at a concentration of 20 nM in all subsequent experiments. Forty-eight hours after transfection with MEF2A siRNA, H9c2 cells were stimulated with HB-EGF for 1 h. The ability of HB-EGF to increase COX-2 expression was suppressed in cells transfected with 20 nM MEF2A siRNA (Fig. 5C). These findings indicate that MEF2A activation is involved in HB-EGF-induced COX-2 expression in cardiomyocytes. Finally, the induction of ANF mRNA
1105
expression and increment of cell size by HB-EGF were shown to be significantly attenuated in MEF2A siRNA-transfected cells (Fig. 5D and E).
1106
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
Fig. 5. HB-EGF induced COX-2 expression involves MEK5–ERK5-dependent activation of MEF2A. (A) Following HB-EGF treatment, cell lysates were prepared as indicated and then analyzed for phopho-MEF2A expression by Western blotting. (B) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF and then harvested at the indicated times. Nuclear fractions were subjected to Western blot analysis using a phopho-MEF2A-specific antibody. (C) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by Western blotting. (D) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 24 h. Cell lysates were analyzed by RT-PCR using ANF-specific primers. (E) Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
3.5. Effects of HB-EGF on cardiomyocyte hypertrophy in neonatal rat cardiomyocytes (NRCMs) To eliminate the possibility that embryonic heart derived cell H9c2 behaves differently from physiological cardiomyocyte in response to HB-EGF, we performed additional experiments using cultured primary cardiomyocytes. As shown in Fig. 6A and B, phosphorylation of ERK5 and induction of COX-2 expression by HB-EGF decreased in cells transfected with MEK5 siRNA. Also, HB-EGF induced phosphorylation of MEF2A (T319) and this activation of MEF2A was attenuated by transfection with MEK5 siRNA (Fig. 6C). Furthermore, transfection with MEF2A siRNA decreased the expression of HB-EGF-induced COX-
2 (Fig. 6D). Lastly, HB-EGF-increased cell size and ANF gene expression were also attenuated by MEK5 or MEF2A siRNA in primary cardiomyocytes (Fig. 6E and F). 4. Discussion HB-EGF, which is expressed in various cell types including heart muscle cells, induces a hypertrophic response in rat cardiac myocytes by activating PI3K-Akt, MAP kinase, and JAK–STAT signaling, suggesting that it acts as an autocrine hypertrophic factor [7–9]. Once cardiac hypertrophy sets in motion, it induces various molecular characteristic changes. These changes include increases in total
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
protein content, changes in cell size and morphology, and reexpression of fetal genes such as ANF [29,31]. In the present study, we confirmed, using H9c2 cells and NRCMs, that HB-EGF induces cardiomyocyte hypertrophy. In addition, we showed that it activates the MEK5–ERK5 pathway, as well as the ERK1/2 pathway (data not shown). Furthermore, we showed that blockade of MEK5–ERK5 signaling using MEK5 siRNA inhibited HB-EGF-induced cardiomyocyte hypertrophy, suggesting that the MEK5–ERK5 pathway is also involved in HB-EGF-induced cardiomyocyte hypertrophy. In recent years, considerable progress has been made in uncovering the role of ERK5 signaling in the heart [32]. Although reports concerning the role of ERK5 in cardiac hypertrophy have been contradictory, they raised the exciting possibility of targeting ERK5 (or downstream mediators) to promote myocyte survival and prevent dysfunction [14,32–35]. Nevertheless, relatively little is known about the contribution of MEK5–ERK5 signaling to HB-EGF-induced cardiomyocyte hypertrophy. Thus, in the present study, we attempted to elucidate the mechanisms underpinning the activation of MEK5–ERK5 signaling by HB-EGF. After transverse aortic constriction, increased hypertrophy has been reported to occur in mice with cardiac-specific overexpression of MEF2A [36]. Elsewhere, a recent study demonstrated that
1107
targeted deletion of ERK5 attenuates hypertrophic response and promotes pressure overload-induced apoptosis in the heart [30]. Moreover, ERK5 and p38 MAP kinase are able to phosphorylate and activate MEF2A and MEF2C [37]. However, p38 MAP kinases fail to compensate for the defect in MEKK2/3-induced, MEF2A-mediated transcription caused by deletion of the Mek5 gene, indicating that ERK5 signaling is essential for regulating MEF2A activity [38]. These findings suggest that ERK5 activation triggers the phosphorylation of MEF2A during the induction of cardiac hypertrophy. Thus, we next examined, using MEK5 and MEF2A siRNAs, whether MEF2A, a downstream target of ERK5, is involved in HB-EGF-induced cardiomyocyte hypertrophy. In agreement with previous reports, our data showed that HB-EGF induced cardiomyocyte hypertrophy through activation of ERK5 and subsequent phosphorylation of MEF2A. Various studies have reported that COX-2 is induced by factors with roles in cardiac function and pathology, including endothelin1, EGF, and angiotensin II, and that inhibition of COX-2 affects hypertrophic signaling and cardiac dysfunction [20–24]. We showed that COX-2 expression increased significantly within 1 h of HB-EGF treatment, before returning to basal levels by 4 h. Our finding that COX-2 induction was transient is consistent with the
Fig. 6. Effects of HB-EGF on cardiomyocyte hypertrophy in NRCMs. (A) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 5 min. Phospho-ERK5 and MEK5 levels were measured by Western blotting. (B) NRCMs were stimulated with HB-EGF for 1 h and then COX-2 and MEK5 levels were measured by Western blotting. (C) NRCMs were stimulated with HB-EGF and then harvested at the indicated times. Nuclear fractions were subjected to Western blot analysis using a phopho-MEF2A-specific antibody. (D) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by Western blotting. (E) Forty-eight hours after MEK5 or MEF2A siRNA transfection, cells were stimulated with HB-EGF for 24 h. Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Right panel: the results of measurement of cell area are represented in a bar graph format. (F) Cell lysates were analyzed by RT-PCR using ANF-specific primers. Data represent the mean ± SEM of at least three independent experiments. * and **, p b 0.05.
1108
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
Fig. 6 (continued).
previous report stating that various stimuli, including bradykinin and IL-1β, rapidly and transiently induce COX-2 expression [39]. Next, we tested whether the induction of COX-2 expression by HBEGF involves ERK5 signaling and/or is associated with cardiomyocyte hypertrophy. HB-EGF-induced COX-2 mRNA and protein expression decreased in cells transfected with MEK5 siRNA. In addition, the COX-2 inhibitor rofecoxib inhibited HB-EGF-induced expression of mRNA encoding ANF, another factor linked to cardiac hypertrophy. A second COX-2 inhibitor, NS398, produced the same effect (data not shown). These data supported the hypothesis that COX-2 is an activator of hypertrophic signaling that lies downstream of MEK5 and ERK5. Consistent with this conclusion, a previous study reported that activation of MAP kinases including JNK, p38 MAP kinase, and ERK5 is required for gastrin-mediated induction of COX2 gene expression during gastrointestinal carcinogenesis [36]. The authors of that report observed that the COX-2 response required ligand-independent activation of EGFR [40]; we found HB-EGF-increased COX-2 expression to be significantly attenuated by pretreatment of the EGFR inhibitor AG1478. As a result, we speculate that HB-EGF-increases cardiomyocyte COX-2 expression via the EGF receptor. Based on previous reports that MEF2-binding sequences exist in the promoters of the human, rat, and mouse COX2 genes, we hypothesized that ERK5, MEF2A, and COX-2 may be functionally related. In further experiments, we investigated whether activation of MEF2A contributes to HB-EGF-induced COX-2 expression. Our data clearly show that the induction of COX-2 expression by HB-EGF was suppressed in cells transfected with MEF2A siRNA. These findings
indicate that MEF2A activation is involved in HB-EGF-induced COX-2 expression in cardiomyocytes. Furthermore, our proposed sequence of events – activation of EGFR by HB-EGF, activation of MEK5–ERK5 signaling, phosphorylation of MEF2A, induction of COX-2 expression, cardiomyocyte hypertrophy – was indirectly confirmed by the observation that pretreatment with rofecoxib significantly inhibited HB-EGF-induced cardiomyocyte hypertrophy. Based on the findings of the present study, we conclude that HBEGF stimulation induces cardiomyocyte hypertrophy via an EGFR– ERK5–MEF2A–COX2 cascade. Our data will help us better understand the molecular mechanism of HB-EGF-induced cardiac hypertrophy, and identifies a new potential therapeutic target in the treatment of this condition. This topic warrants further investigation. Acknowledgments This work was supported by a Korea National Institute of Health intramural research grant (2007-N63001-00). Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.cellsig.2011.01.006. References [1] T.J. Vaughan, J.C. Pascall, K.D. Brown, Biochem. J. 287 (Pt 3) (1992) 681.
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109 [2] J.A. Abraham, D. Damm, A. Bajardi, J. Miller, M. Klagsbrun, R.A. Ezekowitz, Biochem. Biophys. Res. Commun. 190 (1993) 125. [3] M. Yoshizumi, S. Kourembanas, D.H. Temizer, R.P. Cambria, T. Quertermous, M.E. Lee, J. Biol. Chem. 267 (1992) 9467. [4] T.G. Kennedy, K.D. Brown, T.J. Vaughan, Endocrinology 132 (1993) 1857. [5] G. Raab, M. Klagsbrun, Biochim. Biophys. Acta 1333 (1997) F179. [6] M.A. Perrella, T. Maki, S. Prasad, D. Pimental, K. Singh, N. Takahashi, M. Yoshizumi, A. Alali, S. Higashiyama, R.A. Kelly, J. Biol. Chem. 269 (1994) 27045. [7] K. Jin, X.O. Mao, G.G. Del Rio, L. Jin, D.A. Greenberg, J. Neurosci. Res. 81 (2005) 497. [8] C.M. Reynolds, S. Eguchi, G.D. Frank, E.D. Motley, Hypertension 39 (2002) 525. [9] B.H. Shah, K.J. Catt, Trends Pharmacol. Sci. 24 (2003) 239. [10] J. Garaude, S. Cherni, S. Kaminski, E. Delepine, C. Chable-Bessia, M. Benkirane, J. Borges, A. Pandiella, M.A. Iniguez, M. Fresno, R.A. Hipskind, M. Villalba, J. Immunol. 177 (2006) 7607. [11] J.D. Lee, R.J. Ulevitch, J. Han, Biochem. Biophys. Res. Commun. 213 (1995) 715. [12] G. Zhou, Z.Q. Bao, J.E. Dixon, J. Biol. Chem. 270 (1995) 12665. [13] M. Hayashi, J.D. Lee, J. Mol. Med. 82 (2004) 800. [14] N. Takahashi, Y. Saito, K. Kuwahara, M. Harada, K. Tanimoto, Y. Nakagawa, R. Kawakami, M. Nakanishi, S. Yasuno, S. Usami, A. Yoshimura, K. Nakao, J. Mol. Cell. Cardiol. 38 (2005) 185. [15] W. Nadruz Jr., C.B. Kobarg, S.S. Constancio, P.D. Corat, K.G. Franchini, Circ. Res. 92 (2003) 243. [16] R. Passier, H. Zeng, N. Frey, F.J. Naya, R.L. Nicol, T.A. McKinsey, P. Overbeek, J.A. Richardson, S.R. Grant, E.N. Olson, J. Clin. Invest. 105 (2000) 1395. [17] J.D. Molkentin, B.E. Markham, J. Biol. Chem. 268 (1993) 19512. [18] S.M. Kolodziejczyk, L. Wang, K. Balazsi, Y. DeRepentigny, R. Kothary, L.A. Megeney, Curr. Biol. 9 (1999) 1203. [19] K. Xu, H.K. Shu, Cancer Res. 67 (2007) 6121. [20] M.C. Rebsamen, R. Capoccia, M.B. Vallotton, U. Lang, J. Mol. Cell. Cardiol. 35 (2003) 81. [21] R. Rocha, C.L. Martin-Berger, P. Yang, R. Scherrer, J. Delyani, E. McMahon, Endocrinology 143 (2002) 4828. [22] L. Molero, J. Farre, A. Garcia-Mendez, P. Jimenez Mateos-Caceres, M.C. Carrasco, I. Millas, F. Navarro, M. Cordoba, S. Casado, A. Lopez-Farre, Cardiovasc. Res. 57 (2003) 109.
1109
[23] C. Jacobshagen, M. Gruber, N. Teucher, A.G. Schmidt, B.W. Unsold, K. Toischer, V.P. Nguyen, L.S. Maier, H. Kogler, G. Hasenfuss, Eur. J. Heart Fail. 10 (2008) 334. [24] J.M. Streicher, K. Kamei, T.O. Ishikawa, H. Herschman, Y. Wang, J. Mol. Cell. Cardiol. 49 (2010) 88. [25] R.S. Bresalier, R.S. Sandler, H. Quan, J.A. Bolognese, B. Oxenius, K. Horgan, C. Lines, R. Riddell, D. Morton, A. Lanas, M.A. Konstam, J.A. Baron, N. Engl. J. Med. 352 (2005) 1092. [26] C. Bombardier, L. Laine, A. Reicin, D. Shapiro, R. Burgos-Vargas, B. Davis, R. Day, M.B. Ferraz, C.J. Hawkey, M.C. Hochberg, T.K. Kvien, T.J. Schnitzer, N. Engl. J. Med. 343 (1520–8) (2000) 2. [27] J.M. Streicher, Y. Wang, Cardiovasc. Hematol. Agents Med. Chem. 6 (2008) 69. [28] G.W. Wang, D.A. Schuschke, Y.J. Kang, Am. J. Physiol. 276 (1999) H167. [29] R.D. Hannan, A. Jenkins, A.K. Jenkins, Y. Brandenburger, Clin. Exp. Pharmacol. Physiol. 30 (2003) 517. [30] T.E. Kimura, J. Jin, M. Zi, S. Prehar, W. Liu, D. Oceandy, J.I. Abe, L. Neyses, A.H. Weston, E.J. Cartwright, X. Wang, Circ. Res. 106 (2010) 961. [31] P.H. Sugden, J. Card. Fail. 8 (2002) S359. [32] Y. Wang, Circulation 116 (2007) 1413. [33] Y. Nakaoka, K. Nishida, Y. Fujio, M. Izumi, K. Terai, Y. Oshima, S. Sugiyama, S. Matsuda, S. Koyasu, K. Yamauchi-Takihara, T. Hirano, I. Kawase, H. Hirota, Circ. Res. 93 (2003) 221. [34] R.L. Nicol, N. Frey, G. Pearson, M. Cobb, J. Richardson, E.N. Olson, EMBO J. 20 (2001) 2757. [35] S.J. Cameron, S. Itoh, C.P. Baines, C. Zhang, S. Ohta, W. Che, M. Glassman, J.D. Lee, C. Yan, J. Yang, J. Abe, FEBS Lett. 566 (2004) 255. [36] J. Xu, N.L. Gong, I. Bodi, B.J. Aronow, P.H. Backx, J.D. Molkentin, J. Biol. Chem. 281 (2006) 9152. [37] X. Wang, C. Tournier, Cell. Signal. 18 (2006) 753. [38] X. Wang, A.J. Merritt, J. Seyfried, C. Guo, E.S. Papadakis, K.G. Finegan, M. Kayahara, J. Dixon, R.P. Boot-Handford, E.J. Cartwright, U. Mayer, C. Tournier, Mol. Cell. Biol. 25 (2005) 336. [39] Y.J. Kang, U.R. Mbonye, C.J. DeLong, M. Wada, W.L. Smith, Prog. Lipid Res. 46 (2007) 108. [40] Y.S. Guo, J.Z. Cheng, G.F. Jin, J.S. Gutkind, M.R. Hellmich, C.M. Townsend Jr., J. Biol. Chem. 277 (2002) 48755.
Contents lists available at ScienceDirect
Cellular Signalling j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c e l l s i g
HB-EGF induces cardiomyocyte hypertrophy via an ERK5-MEF2A-COX2 signaling pathway Kuy-Sook Lee, Jin-Hee Park, Hyun-Joung Lim, Hyun-Young Park ⁎ Division of Cardiovascular and Rare Diseases, Center for Biomedical Sciences, National Institute of Health, Seoul, Korea
a r t i c l e
i n f o
Article history: Received 8 September 2010 Received in revised form 23 December 2010 Accepted 10 January 2011 Available online 16 January 2011 Keywords: HB-EGF MEK5–ERK5 MEF2A COX-2 Cardiomyocyte hypertrophy
a b s t r a c t Heparin-binding EGF-like growth factor (HB-EGF) is a member of the EGF family that binds to and activates the EGF receptor. Transactivated by angiotensin II, ET-1, and various growth factors in cardiomyocytes, HBEGF is known to induce cardiac hypertrophy via the PI3K-Akt, MAP kinase, and JAK–STAT pathways. However, little is known about the potential involvement of the ERK5 pathway in HB-EGF-induced cardiac hypertrophy. In the present report, we identify and characterize a novel MEK5–ERK5 pathway that is involved in HB-EGFinduced cardiomyocyte hypertrophy. HB-EGF (10 ng/ml) significantly increased [3H]-leucine incorporation and atrial natriuretic factor (ANF) mRNA expression in H9c2 cells. In addition, HB-EGF activated a MEK5– ERK5 pathway. Pretreatment with the EGFR inhibitor AG1478 attenuated the activation of ERK5. Blockade of MEK5–ERK5 signaling using MEK5 siRNA reduced the ability of HB-EGF to increase cell size and the expression of ANF mRNA, suggesting the involvement of an EGFR–ERK5 pathway in HB-EGF-induced cardiomyocyte hypertrophy. We further analyzed cyclooxygenase-2 (COX-2). HB-EGF enhanced the expression of COX-2, a response mediated by MEK5–ERK5 signaling, while the COX-2 inhibitor rofecoxib attenuated HB-EGF-induced ANF mRNA expression, suggesting that COX-2 is also associated with HB-EGFinduced cardiomyocyte hypertrophy. It has been known that ERK5 activates the myocyte enhancer factor (MEF) 2 family of transcription factor, we next tested whether activation of MEF2A contributes to HB-EGFinduced COX-2 expression. Inhibition of MEF2A using siRNA attenuated HB-EGF-induced COX-2, ANF expression and cell size. In conclusion, HB-EGF induces cardiomyocyte hypertrophy through an EGFR–ERK5– MEF2A–COX-2 pathway. Our findings will help us to better understand the molecular mechanisms behind HB-EGF-induced cardiomyocyte hypertrophy. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Heparin-binding EGF-like growth factor (HB-EGF), a member of the EGF family that binds to and activates the EGF receptor (EGFR), is expressed in a variety of tissues, notably lung, heart, brain and skeletal muscle [1–4]. HB-EGF is well known to be a potent mitogen for several cell types including fibroblasts, smooth muscle cells, keratinocytes, and breast carcinoma cells [5]. Furthermore, HB-EGF is expressed in heart muscle cells, in which HB-EGF mRNA levels have been shown to be markedly increased by the α-adrenergic agonist phenylephrine, which is known to induce myocardial hypertrophy [6]. In addition, HB-EGF induced a hypertrophic response in rat cardiac myocytes, which suggests that it acts as an autocrine hypertrophic factor [6]. HBEGF is known to activate cell signal transduction networks by auto-
⁎ Corresponding author. Tel.: +82 43 719 8650; fax: +82 43 719 8689. E-mail address: [email protected] (H.-Y. Park). 0898-6568/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2011.01.006
phosphorylation of the EGF receptor, which consequently recruits adapter signaling molecules. MAP kinase, PI3K–Akt, and JAK-STAT pathways are well-known to form part of these cell signal networks [7–9]. MAP kinase pathways play a central role in the signaling network that transduces extracellular stimuli into distinct intracellular responses. Four major MAP kinase pathways have been identified in mammalian cells. The three canonical MAP kinase pathways – ERK1/2, JNK, and p38 MAP kinase – mediate proliferative and stress signaling in various cells [10]. The fourth is Big MAP kinase 1 (BMK1) or ERK5, which was cloned by Lee et al. and Zhou et al. and is activated in response to growth factors and stress [11,12]. Activation of this ERK5 cascade has been implicated not only in physiological functions such as cell survival, proliferation, and differentiation but also in pathological processes such as carcinogenesis, cardiac hypertrophy, and atherosclerosis [13,14]. Downstream targets of ERK5 signaling include the myocyte enhancer factor (MEF) 2 family of transcriptional factors, which has been closely linked to hypertrophy [15]. Moreover, observed MEF2 activation in multiple models of cardiac hypertrophy supports the view that members of this family may play central roles in the
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
regulation of fundamental signaling mechanisms during myocardial hypertrophic growth [16–18]. Cyclooxygenase-2 (COX-2) is an inducible and rate-limiting enzyme in the biosynthesis of prostaglandins. COX-2 is induced in response to various stimuli, including mitogens, cytokines, growth factors, and inflammatory signals [19]. Specifically, it has been shown to be induced by factors with roles in heart function and pathology, including endothelin-1, EGF, and angiotensin II [20–22]. A recent study demonstrated that cardiac-specific COX-2 expression results in elevated prostaglandin levels and mild hypertrophy. Furthermore, the COX-2 inhibitor celecoxib has been reported to modulate hypertrophic signaling and prevent load-induced cardiac dysfunction [23,24]. Recently, the role of COX-2 in human heart disease was brought to the public attention by the finding that chronic use of the popular COX-2specific inhibitor rofecoxib increased heart disease risk [25–27]. Thus, in the present study, we attempted to identify and characterize the novel pathways involved in HB-EGF-induced cardiomyocyte hypertrophy, focusing initially on the MEK5–ERK5 pathway and its relationships to MEF2 and COX-2. We herein report that HB-EGF induces cardiomyocyte hypertrophy via EGFR-dependent induction of COX-2. We further show that
1101
MEK5–ERK5 signaling increases COX-2 expression through phosphorylation of MEF2A.
2. Materials and methods 2.1. Materials Cell culture reagents and oligo(dT)12–18 were purchased from Invitrogen (Carlsbad, CA), the mouse anti-β-actin monoclonal antibody was obtained from Sigma (St. Louis, MO), and HB-EGF was acquired from R&D systems (Minneapolis, MN). Peroxidase-conjugated anti-rabbit and anti-mouse secondary antibodies were obtained from Jackson ImmunoResearch (West Grove, PA), Accupower RT Premix and Accupower PCR Premix were purchased from Bioneer (Daejeon, Korea), and the Alexa Fluor® 488-conjugated goat antimouse IgG was obtained from Molecular Probes (Eugene, OR). RNeasy mini kits were obtained from Qiagen (Valencia, CA), the MEF2A and phospho-MEF2A antibodies from Abcam (Cambridge, UK), the phospho-MEK5 antibody from IMGENEX (San Diego, CA), the phospho-ERK5 antibody from Cell Signaling Technology (Danvers,
Fig. 1. HB-EGF induces hypertrophy in H9c2 cells. H9c2 cells were stimulated with HB-EGF (10 ng/ml) for 24 h. (A) Total mRNA was isolated and the expression of ANF evaluated by RT-PCR using specific primers. (B) Induction of protein synthesis by HB-EGF was assessed by [3H]-leucine incorporation assay. (C) Immunofluorescent microscopy using an αactinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
1102
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
MA), and the COX-2 antibody from Santa Cruz Biotechnology (Santa Cruz, CA). 2.2. Cell culture Embryonic rat heart-derived H9c2 cells were purchased from ATCC (Manassas, VA) and were maintained in Dulbecco's modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Cells were starved for 4 h in DMEM containing 0.1% FBS prior to experimental treatments. Neonatal rat cardiomyocytes (NRCMs) were prepared as described elsewhere with minor modification [28]. Briefly, hearts harvested from 1 to 3 days old Sprague-Dawley rat (obtained from Orient, Seoul, Korea) were washed in PBS (Mg2+, Ca2+ free). Hearts were minced with collagenase type II (1 mg/ml, 262 U/mg, GibcoBRL) in sterile HBSS and incubated for 5 min. The pellet was collected in HBSS containing collagenase type II and incubated for 5 min. The supernatant was collected in α-MEM containing 10% FBS and digestion steps were repeated up to 12 times. Cardiomyocytes were purified from fibroblasts using repeated centrifugation step. Finally, cardiomyocytes were resuspended in α-MEM containing 10% FBS and 0.1 mM BrdU. Cells were then cultured in α-MEM containing 10% FBS at 37 °C for 2– 3 days in a humidified atmosphere containing 5% CO2 and 95 % air. More than 90% of cells were cardiomyocytes (positive for Troponin I and beating feature).
2.6. siRNA transfection Cells were grown to ~ 20–30% confluence in culture media supplemented with 10% FBS. MEK5 or MEF2A siRNA was introduced into H9c2 cells or NRCMs using LipofectamineTM RNAiMAX (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were harvested. Lysates were prepared for analysis by RT-PCR and Western blotting to evaluate the silencing effect of MEK5 and MEF2A siRNAs. MEK5 and MEF2A siRNAs were purchased from Qiagen. The rat-specific MEK5 and MEF2A target sequences were 5′-CAC AGT CTA CAA AGC ATA TCA-3′ and 5′-CAG CTC TAA CAA GTT GTT TCA-3′, respectively. A nonspecific control siRNA (Invitrogen) was used as a negative control. 2.7. Immunocytochemistry H9c2 cells or NRCMs were plated to 4-well Lab-Tek Chamber Slides (Nunc, New York, NY). Cells were starved for 4 h in DMEM
2.3. [3H]-leucine incorporation assay Protein synthesis assays were performed using cells plated to 96well plates at a density of 3.5 × 103 cells/well. Cells were starved in DMEM with 0.1% FBS for 4 h and then incubated with HB-EGF (10 ng/ ml) for 24 h. Next, [3H]-leucine (1 μCi/ml) was added to the culture medium and the cells were incubated for a further 4 h before being transferred to a filter mat (PerkinElmer, MA) using a Tomtec Harvester 96 harvester (Hamden, CT). [3H]-leucine incorporation was measured using a Wallac Microbeta Trilux 1450 Scintillation Counter (PerkinElmer). 2.4. Reverse transcriptase polymerase chain reaction (RT-PCR) analysis Total cellular RNA was extracted from H9c2 cells and NRCMs using an RNeasy mini kit (Qiagen) according to the manufacturer's instructions. An oligo(dT)12–18 primer and Accupower RT Premix were used to synthesize cDNA from 500 ng of total RNA. Target sequences were amplified using the following primers: ANF, 5′-ACC TGC TAG ACC ACC TGG AG-3′ and 5′-GTA CCG GAA GCT GTT GCA G-3′; MEK5, 5′-GCC AAC GGC CAG ATG AAT GAA C-3′ and 5′-CAC GCC GAA GTC ACA CAG CTT G-3′; MEF2A, 5′-CCG CCT CAG AAC TTC TCA ATG-3′ and 5′-TTG GAG AGG CCC TTG AGT TTA C-3′; COX-2, 5′-GGC AAA GGC CTC CAT TGA CCA G-3′ and 5′-GGA GGC ACT TGC GTT GAT GGT G-3′; β-actin, 5′-CCC ATT GAA CAC GGC ATT GTC-3′and 5′-CGC ACG ATT TCC CTC TCA GC-3′. 2.5. Western blot analysis Cells were starved for 4 h in DMEM supplemented with 0.1% FBS prior to experimental treatments. Following treatment, cells washed with ice-cold phosphate-buffered saline (PBS) and harvested, through scraping, in 1× protein lysis buffer (Cell Signaling Technology). Equal amounts of cell lysates were separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE). Membranes were incubated with primary antibody, and immunopositive bands were subsequently visualized using the ECL detection reagent (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK). Each experiment was performed at least three times.
Fig. 2. HB-EGF activates a MEK5–ERK5 pathway in H9c2 cells. (A) Phosphorylation of MEK5 and ERK5 after HB-EGF treatment in H9c2 cells, detected by Western blotting. (B) Effect of AG1478 on HB-EGF-induced phosphorylation of ERK5. Cells were pretreated with AG1478 (250 nM) for 1 h and then the cells were treated with HB-EGF for 5 min. Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
containing 0.1% FBS prior to experimental treatments. Briefly, after two washes in ice-cold PBS, cells were fixed with 4% paraformaldehyde at room temperature for 30 min and then washed three times in PBS. They were then incubated with anti-α-actinin antibody (Santa Cruz Biotechnology) diluted in 1% bovine serum
1103
albumin (BSA)–PBS. Primary antibody was detected with Alexa Fluor® 488-conjugated goat anti-mouse IgG (Molecular Probes). The images observed under a fluorescence microscope. Cell sizes were measured using AxioVision software (Carl Zeiss Inc. Thornwood, NY).
Fig. 3. Involvement of MEK5–ERK5 signaling in HB-EGF-induced cardiomyocyte hypertrophy. (A) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 5 min. Phospho-ERK5 and MEK5 levels were measured by Western blotting. (B) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 24 h. Cell lysates were analyzed by RT-PCR using ANF- and MEK5-specific primers. (C) Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
1104
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
2.8. Statistical analysis Data are presented as the means ± SEM of three or more individual experiments. Significant differences between treatments and controls were identified using Student's t-test. A p value b0.05 was considered to be statistically significant.
Fig. 4 (continued).
3. Results 3.1. HB-EGF induces cardiomyocyte hypertrophy and activates MEK5– ERK5 signaling via the EGF receptor
Fig. 4. COX-2 is involved in HB-EGF-induced cardiomyocyte hypertrophy. Following HB-EGF treatment, cell lysates were prepared as indicated then analyzed for COX-2 expression through (A) RT-PCR and (B) Western blotting. (C) Effect of AG1478 on HBEGF-induced COX-2 expression. Cells were pretreated with AG1478 (250 nM) for 1 h and then the cells were treated with HB-EGF for 1 h. (D) Analysis of ANF mRNA levels in HB-EGF-treated cardiomyocytes in the presence or absence of the COX-2 inhibitors rofecoxib (10 μM). Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by (E) RT-PCR and (F) Western blotting. Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
We examined the effect of HB-EGF on cardiac hypertrophy using a cardiomyocyte cell line, H9c2. The first, we determined whether HB-EGF induced cardiac hypertrophy. Cardiomyocytes were stimulated with HB-EGF (10 ng/ml) for 24 h. As shown in Fig. 1, we observed an important molecular characteristic of cardiac hypertrophy [29] in that atrial natriuretic factor (ANF) mRNA, [3H]-leucine incorporation and cell size were significantly elevated following HBEGF treatment. To determine whether the MEK5–ERK5 pathway contributes to HB-EGF-induced signal transduction in cardiomyocytes, H9c2 cells were stimulated with 10 ng/ml HB-EGF for 5–180 min and cell lysates analyzed by Western blotting using anti-p-MEK5 and anti-pERK5 antibodies. HB-EGF induced MEK5 and ERK5 phosphorylation
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
within 5 min (Fig. 2A). We did additional experiment at time points between 0–5 min. We observed that phosphorylation of ERK5 was significantly increased by HB-EGF at early time points (1 to 5 min after HB-EGF stimulation, Data not shown), and there are no significant different between each time point. Therefore, we determined the time point as 5 min in subsequent experiments for a steady experimental condition. In addition, the EGFR tyrosine kinase inhibitor AG1478 (250 nM) attenuated HB-EGF-mediated activation of ERK5 (Fig. 2B). 3.2. MEK5–ERK5 signaling is involved in HB-EGF-induced cardiomyocyte hypertrophy To determine whether the MEK5–ERK5 signal pathway is involved in HB-EGF-induced cardiomyocyte hypertrophy, we tested the effect of MEK5 siRNA on cardiomyocyte hypertrophy. Transfection of H9c2 cells with MEK5 siRNA (10–100 nM) abolished MEK5 expression within 48 h (Supplemental Fig. 1). MEK5 siRNA was used at a concentration of 20 nM in all subsequent experiments. Forty-eight hours after MEK5 siRNA transfection, H9c2 cells were stimulated with HB-EGF for 5 min. ERK5 activation decreased significantly in cells transfected with MEK5 siRNA (Fig. 3A). In addition, HB-EGF-increased ANF mRNA expression and cell size were significantly attenuated by MEK5 siRNA transfection (Fig. 3B and C). 3.3. HB-EGF induces cardiomyocyte hypertrophy through induction of COX-2 expression via EGFR We further analyzed COX-2, an activator of hypertrophic signaling that lies downstream of ERK5. HB-EGF induced COX-2 mRNA and protein expression within 1 h (Fig. 4A and B). This increase in COX-2 expression was significantly attenuated by pretreatment with AG1478, indicating that HB-EGF increases COX-2 expression via the EGF receptor (Fig. 4C). The COX-2 inhibitor rofecoxib (10 μM) inhibited HB-EGF-induced ANF mRNA expression (Fig. 4D). Furthermore, induction of COX-2 mRNA and protein expression by HB-EGF decreased in cells transfected with MEK5 siRNA (Fig. 4E and F). These findings support the hypothesis that COX-2, acting downstream of MEK5 and ERK5, mediates HB-EGF-induced cardiomyocyte hypertrophy. 3.4. MEK5–ERK5-dependent activation of MEF2A is involved in HB-EGFinduced COX-2 expression Previous studies have reported that ERK5 activation in cardiac hypertrophy affects downstream transcription factors including MEF2 [30]. Thus, we also examined whether HB-EGF activates MEF2A in cardiomyocytes. HB-EGF induced phosphorylation of MEF2A (T319) within 5 min (Fig. 5A). As shown by Western blot analysis of nuclear fractions, HB-EGF induced phosphorylation of MEF2A (T319) within 15 min. This activation of MEF2A was attenuated by transfection with MEK5 siRNA (Fig. 5B). To investigate the possibility that HB-EGF-induced MEF2A activation increases the expression of COX-2 and thereby induces subsequent cardiomyocyte hypertrophy, we next examined the effects of MEF2A siRNA. MEF2A siRNA (10–50 nM) was introduced into H9c2 cells. Cells were harvested 48 h after transfection and lysates analyzed by Western blotting using an anti-MEF2A antibody. MEF2A siRNA (10–50 nM) greatly reduced MEF2A mRNA and protein expression (Supplemental Fig. 2A and B). MEF2A siRNA was used at a concentration of 20 nM in all subsequent experiments. Forty-eight hours after transfection with MEF2A siRNA, H9c2 cells were stimulated with HB-EGF for 1 h. The ability of HB-EGF to increase COX-2 expression was suppressed in cells transfected with 20 nM MEF2A siRNA (Fig. 5C). These findings indicate that MEF2A activation is involved in HB-EGF-induced COX-2 expression in cardiomyocytes. Finally, the induction of ANF mRNA
1105
expression and increment of cell size by HB-EGF were shown to be significantly attenuated in MEF2A siRNA-transfected cells (Fig. 5D and E).
1106
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
Fig. 5. HB-EGF induced COX-2 expression involves MEK5–ERK5-dependent activation of MEF2A. (A) Following HB-EGF treatment, cell lysates were prepared as indicated and then analyzed for phopho-MEF2A expression by Western blotting. (B) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF and then harvested at the indicated times. Nuclear fractions were subjected to Western blot analysis using a phopho-MEF2A-specific antibody. (C) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by Western blotting. (D) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 24 h. Cell lysates were analyzed by RT-PCR using ANF-specific primers. (E) Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Data represent the mean ± SEM of at least three independent experiments. *, p b 0.05.
3.5. Effects of HB-EGF on cardiomyocyte hypertrophy in neonatal rat cardiomyocytes (NRCMs) To eliminate the possibility that embryonic heart derived cell H9c2 behaves differently from physiological cardiomyocyte in response to HB-EGF, we performed additional experiments using cultured primary cardiomyocytes. As shown in Fig. 6A and B, phosphorylation of ERK5 and induction of COX-2 expression by HB-EGF decreased in cells transfected with MEK5 siRNA. Also, HB-EGF induced phosphorylation of MEF2A (T319) and this activation of MEF2A was attenuated by transfection with MEK5 siRNA (Fig. 6C). Furthermore, transfection with MEF2A siRNA decreased the expression of HB-EGF-induced COX-
2 (Fig. 6D). Lastly, HB-EGF-increased cell size and ANF gene expression were also attenuated by MEK5 or MEF2A siRNA in primary cardiomyocytes (Fig. 6E and F). 4. Discussion HB-EGF, which is expressed in various cell types including heart muscle cells, induces a hypertrophic response in rat cardiac myocytes by activating PI3K-Akt, MAP kinase, and JAK–STAT signaling, suggesting that it acts as an autocrine hypertrophic factor [7–9]. Once cardiac hypertrophy sets in motion, it induces various molecular characteristic changes. These changes include increases in total
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
protein content, changes in cell size and morphology, and reexpression of fetal genes such as ANF [29,31]. In the present study, we confirmed, using H9c2 cells and NRCMs, that HB-EGF induces cardiomyocyte hypertrophy. In addition, we showed that it activates the MEK5–ERK5 pathway, as well as the ERK1/2 pathway (data not shown). Furthermore, we showed that blockade of MEK5–ERK5 signaling using MEK5 siRNA inhibited HB-EGF-induced cardiomyocyte hypertrophy, suggesting that the MEK5–ERK5 pathway is also involved in HB-EGF-induced cardiomyocyte hypertrophy. In recent years, considerable progress has been made in uncovering the role of ERK5 signaling in the heart [32]. Although reports concerning the role of ERK5 in cardiac hypertrophy have been contradictory, they raised the exciting possibility of targeting ERK5 (or downstream mediators) to promote myocyte survival and prevent dysfunction [14,32–35]. Nevertheless, relatively little is known about the contribution of MEK5–ERK5 signaling to HB-EGF-induced cardiomyocyte hypertrophy. Thus, in the present study, we attempted to elucidate the mechanisms underpinning the activation of MEK5–ERK5 signaling by HB-EGF. After transverse aortic constriction, increased hypertrophy has been reported to occur in mice with cardiac-specific overexpression of MEF2A [36]. Elsewhere, a recent study demonstrated that
1107
targeted deletion of ERK5 attenuates hypertrophic response and promotes pressure overload-induced apoptosis in the heart [30]. Moreover, ERK5 and p38 MAP kinase are able to phosphorylate and activate MEF2A and MEF2C [37]. However, p38 MAP kinases fail to compensate for the defect in MEKK2/3-induced, MEF2A-mediated transcription caused by deletion of the Mek5 gene, indicating that ERK5 signaling is essential for regulating MEF2A activity [38]. These findings suggest that ERK5 activation triggers the phosphorylation of MEF2A during the induction of cardiac hypertrophy. Thus, we next examined, using MEK5 and MEF2A siRNAs, whether MEF2A, a downstream target of ERK5, is involved in HB-EGF-induced cardiomyocyte hypertrophy. In agreement with previous reports, our data showed that HB-EGF induced cardiomyocyte hypertrophy through activation of ERK5 and subsequent phosphorylation of MEF2A. Various studies have reported that COX-2 is induced by factors with roles in cardiac function and pathology, including endothelin1, EGF, and angiotensin II, and that inhibition of COX-2 affects hypertrophic signaling and cardiac dysfunction [20–24]. We showed that COX-2 expression increased significantly within 1 h of HB-EGF treatment, before returning to basal levels by 4 h. Our finding that COX-2 induction was transient is consistent with the
Fig. 6. Effects of HB-EGF on cardiomyocyte hypertrophy in NRCMs. (A) Forty-eight hours after MEK5 siRNA transfection, cells were stimulated with HB-EGF for 5 min. Phospho-ERK5 and MEK5 levels were measured by Western blotting. (B) NRCMs were stimulated with HB-EGF for 1 h and then COX-2 and MEK5 levels were measured by Western blotting. (C) NRCMs were stimulated with HB-EGF and then harvested at the indicated times. Nuclear fractions were subjected to Western blot analysis using a phopho-MEF2A-specific antibody. (D) Forty-eight hours after MEF2A siRNA transfection, cells were stimulated with HB-EGF for 1 h. Cell lysates were analyzed for COX-2 expression by Western blotting. (E) Forty-eight hours after MEK5 or MEF2A siRNA transfection, cells were stimulated with HB-EGF for 24 h. Immunofluorescent microscopy using an α-actinin-specific antibody (green: α-actinin, blue: DAPI). Right panel: the results of measurement of cell area are represented in a bar graph format. (F) Cell lysates were analyzed by RT-PCR using ANF-specific primers. Data represent the mean ± SEM of at least three independent experiments. * and **, p b 0.05.
1108
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109
Fig. 6 (continued).
previous report stating that various stimuli, including bradykinin and IL-1β, rapidly and transiently induce COX-2 expression [39]. Next, we tested whether the induction of COX-2 expression by HBEGF involves ERK5 signaling and/or is associated with cardiomyocyte hypertrophy. HB-EGF-induced COX-2 mRNA and protein expression decreased in cells transfected with MEK5 siRNA. In addition, the COX-2 inhibitor rofecoxib inhibited HB-EGF-induced expression of mRNA encoding ANF, another factor linked to cardiac hypertrophy. A second COX-2 inhibitor, NS398, produced the same effect (data not shown). These data supported the hypothesis that COX-2 is an activator of hypertrophic signaling that lies downstream of MEK5 and ERK5. Consistent with this conclusion, a previous study reported that activation of MAP kinases including JNK, p38 MAP kinase, and ERK5 is required for gastrin-mediated induction of COX2 gene expression during gastrointestinal carcinogenesis [36]. The authors of that report observed that the COX-2 response required ligand-independent activation of EGFR [40]; we found HB-EGF-increased COX-2 expression to be significantly attenuated by pretreatment of the EGFR inhibitor AG1478. As a result, we speculate that HB-EGF-increases cardiomyocyte COX-2 expression via the EGF receptor. Based on previous reports that MEF2-binding sequences exist in the promoters of the human, rat, and mouse COX2 genes, we hypothesized that ERK5, MEF2A, and COX-2 may be functionally related. In further experiments, we investigated whether activation of MEF2A contributes to HB-EGF-induced COX-2 expression. Our data clearly show that the induction of COX-2 expression by HB-EGF was suppressed in cells transfected with MEF2A siRNA. These findings
indicate that MEF2A activation is involved in HB-EGF-induced COX-2 expression in cardiomyocytes. Furthermore, our proposed sequence of events – activation of EGFR by HB-EGF, activation of MEK5–ERK5 signaling, phosphorylation of MEF2A, induction of COX-2 expression, cardiomyocyte hypertrophy – was indirectly confirmed by the observation that pretreatment with rofecoxib significantly inhibited HB-EGF-induced cardiomyocyte hypertrophy. Based on the findings of the present study, we conclude that HBEGF stimulation induces cardiomyocyte hypertrophy via an EGFR– ERK5–MEF2A–COX2 cascade. Our data will help us better understand the molecular mechanism of HB-EGF-induced cardiac hypertrophy, and identifies a new potential therapeutic target in the treatment of this condition. This topic warrants further investigation. Acknowledgments This work was supported by a Korea National Institute of Health intramural research grant (2007-N63001-00). Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.cellsig.2011.01.006. References [1] T.J. Vaughan, J.C. Pascall, K.D. Brown, Biochem. J. 287 (Pt 3) (1992) 681.
K.-S. Lee et al. / Cellular Signalling 23 (2011) 1100–1109 [2] J.A. Abraham, D. Damm, A. Bajardi, J. Miller, M. Klagsbrun, R.A. Ezekowitz, Biochem. Biophys. Res. Commun. 190 (1993) 125. [3] M. Yoshizumi, S. Kourembanas, D.H. Temizer, R.P. Cambria, T. Quertermous, M.E. Lee, J. Biol. Chem. 267 (1992) 9467. [4] T.G. Kennedy, K.D. Brown, T.J. Vaughan, Endocrinology 132 (1993) 1857. [5] G. Raab, M. Klagsbrun, Biochim. Biophys. Acta 1333 (1997) F179. [6] M.A. Perrella, T. Maki, S. Prasad, D. Pimental, K. Singh, N. Takahashi, M. Yoshizumi, A. Alali, S. Higashiyama, R.A. Kelly, J. Biol. Chem. 269 (1994) 27045. [7] K. Jin, X.O. Mao, G.G. Del Rio, L. Jin, D.A. Greenberg, J. Neurosci. Res. 81 (2005) 497. [8] C.M. Reynolds, S. Eguchi, G.D. Frank, E.D. Motley, Hypertension 39 (2002) 525. [9] B.H. Shah, K.J. Catt, Trends Pharmacol. Sci. 24 (2003) 239. [10] J. Garaude, S. Cherni, S. Kaminski, E. Delepine, C. Chable-Bessia, M. Benkirane, J. Borges, A. Pandiella, M.A. Iniguez, M. Fresno, R.A. Hipskind, M. Villalba, J. Immunol. 177 (2006) 7607. [11] J.D. Lee, R.J. Ulevitch, J. Han, Biochem. Biophys. Res. Commun. 213 (1995) 715. [12] G. Zhou, Z.Q. Bao, J.E. Dixon, J. Biol. Chem. 270 (1995) 12665. [13] M. Hayashi, J.D. Lee, J. Mol. Med. 82 (2004) 800. [14] N. Takahashi, Y. Saito, K. Kuwahara, M. Harada, K. Tanimoto, Y. Nakagawa, R. Kawakami, M. Nakanishi, S. Yasuno, S. Usami, A. Yoshimura, K. Nakao, J. Mol. Cell. Cardiol. 38 (2005) 185. [15] W. Nadruz Jr., C.B. Kobarg, S.S. Constancio, P.D. Corat, K.G. Franchini, Circ. Res. 92 (2003) 243. [16] R. Passier, H. Zeng, N. Frey, F.J. Naya, R.L. Nicol, T.A. McKinsey, P. Overbeek, J.A. Richardson, S.R. Grant, E.N. Olson, J. Clin. Invest. 105 (2000) 1395. [17] J.D. Molkentin, B.E. Markham, J. Biol. Chem. 268 (1993) 19512. [18] S.M. Kolodziejczyk, L. Wang, K. Balazsi, Y. DeRepentigny, R. Kothary, L.A. Megeney, Curr. Biol. 9 (1999) 1203. [19] K. Xu, H.K. Shu, Cancer Res. 67 (2007) 6121. [20] M.C. Rebsamen, R. Capoccia, M.B. Vallotton, U. Lang, J. Mol. Cell. Cardiol. 35 (2003) 81. [21] R. Rocha, C.L. Martin-Berger, P. Yang, R. Scherrer, J. Delyani, E. McMahon, Endocrinology 143 (2002) 4828. [22] L. Molero, J. Farre, A. Garcia-Mendez, P. Jimenez Mateos-Caceres, M.C. Carrasco, I. Millas, F. Navarro, M. Cordoba, S. Casado, A. Lopez-Farre, Cardiovasc. Res. 57 (2003) 109.
1109
[23] C. Jacobshagen, M. Gruber, N. Teucher, A.G. Schmidt, B.W. Unsold, K. Toischer, V.P. Nguyen, L.S. Maier, H. Kogler, G. Hasenfuss, Eur. J. Heart Fail. 10 (2008) 334. [24] J.M. Streicher, K. Kamei, T.O. Ishikawa, H. Herschman, Y. Wang, J. Mol. Cell. Cardiol. 49 (2010) 88. [25] R.S. Bresalier, R.S. Sandler, H. Quan, J.A. Bolognese, B. Oxenius, K. Horgan, C. Lines, R. Riddell, D. Morton, A. Lanas, M.A. Konstam, J.A. Baron, N. Engl. J. Med. 352 (2005) 1092. [26] C. Bombardier, L. Laine, A. Reicin, D. Shapiro, R. Burgos-Vargas, B. Davis, R. Day, M.B. Ferraz, C.J. Hawkey, M.C. Hochberg, T.K. Kvien, T.J. Schnitzer, N. Engl. J. Med. 343 (1520–8) (2000) 2. [27] J.M. Streicher, Y. Wang, Cardiovasc. Hematol. Agents Med. Chem. 6 (2008) 69. [28] G.W. Wang, D.A. Schuschke, Y.J. Kang, Am. J. Physiol. 276 (1999) H167. [29] R.D. Hannan, A. Jenkins, A.K. Jenkins, Y. Brandenburger, Clin. Exp. Pharmacol. Physiol. 30 (2003) 517. [30] T.E. Kimura, J. Jin, M. Zi, S. Prehar, W. Liu, D. Oceandy, J.I. Abe, L. Neyses, A.H. Weston, E.J. Cartwright, X. Wang, Circ. Res. 106 (2010) 961. [31] P.H. Sugden, J. Card. Fail. 8 (2002) S359. [32] Y. Wang, Circulation 116 (2007) 1413. [33] Y. Nakaoka, K. Nishida, Y. Fujio, M. Izumi, K. Terai, Y. Oshima, S. Sugiyama, S. Matsuda, S. Koyasu, K. Yamauchi-Takihara, T. Hirano, I. Kawase, H. Hirota, Circ. Res. 93 (2003) 221. [34] R.L. Nicol, N. Frey, G. Pearson, M. Cobb, J. Richardson, E.N. Olson, EMBO J. 20 (2001) 2757. [35] S.J. Cameron, S. Itoh, C.P. Baines, C. Zhang, S. Ohta, W. Che, M. Glassman, J.D. Lee, C. Yan, J. Yang, J. Abe, FEBS Lett. 566 (2004) 255. [36] J. Xu, N.L. Gong, I. Bodi, B.J. Aronow, P.H. Backx, J.D. Molkentin, J. Biol. Chem. 281 (2006) 9152. [37] X. Wang, C. Tournier, Cell. Signal. 18 (2006) 753. [38] X. Wang, A.J. Merritt, J. Seyfried, C. Guo, E.S. Papadakis, K.G. Finegan, M. Kayahara, J. Dixon, R.P. Boot-Handford, E.J. Cartwright, U. Mayer, C. Tournier, Mol. Cell. Biol. 25 (2005) 336. [39] Y.J. Kang, U.R. Mbonye, C.J. DeLong, M. Wada, W.L. Smith, Prog. Lipid Res. 46 (2007) 108. [40] Y.S. Guo, J.Z. Cheng, G.F. Jin, J.S. Gutkind, M.R. Hellmich, C.M. Townsend Jr., J. Biol. Chem. 277 (2002) 48755.