Dose-dependent activation of distinct hypertrophic pathways by serotonin in cardiac cells

C. Villeneuve, A. Caudrillier, C. Ordener, N. Pizzinat, A. Parini, J. Mialet-Perez

Abstract

There is substantial evidence supporting a hypertrophic action of serotonin [5-hydroxytryptamine (5-HT)] in cardiomyocytes. However, little is known about the mechanisms involved. We previously demonstrated that 5-HT-induced hypertrophy depends, in part, on the generation of reactive oxygen species by monoamine oxidase-A (MAO-A) (see Ref. 3). Cardiomyocytes express 5-HT2 receptors, which may also participate in hypertrophy. Here, we analyzed the respective contribution of 5-HT2 receptors and MAO-A in H9C2 cardiomyoblast hypertrophy. 5-HT induced a dose-dependent increase in [3H]leucine incorporation and stimulation of two markers of cardiac hypertrophy, ANF-luc and αSK-actin-luc reporter genes. Experiments using 1 μM 5-HT showed that hypertrophic response occurred independently from MAO-A. Using pharmacological inhibitors (M100907 and ketanserin), we identified a novel mechanism of action involving 5-HT2A receptors and requiring Ca2+/calcineurin/nuclear factor of activated T-cell activation. The activation of this hypertrophic pathway was fully prevented by 5-HT2A inhibitors and was unaffected by MAO inhibition. When 10 μM 5-HT was used, an additional hypertrophic response, prevented by the MAO inhibitors pargyline and RO 41-1049, was observed. Unlike the 5-HT2A-receptor-mediated H9C2 cell hypertrophy, MAO-A-dependent hypertrophic response required activation of extracellular-regulated kinases. In conclusion, our results show the existence of a dose-dependent shift of activation of distinct intracellular pathways involved in 5-HT-mediated hypertrophy of cardiac cells.

  • 5-hydroxytryptamine
  • 5-hydroxytryptamine 2A
  • monoamine oxidase type A
  • hypertrophy

cardiac ventricular remodeling is a hallmark feature in the progression of heart failure. Increased myocardial mass characteristic of the compensatory ventricular remodeling process is primarily attributed to hypertrophy of individual cardiomyocytes. This process is initiated by a variety of stimuli, including paracrine and endocrine factors, such as catecholamines, angiotensin II, endothelin, growth factors, and inflammatory cytokines (1). In addition, serotonin [5-hydroxytryptamine (5-HT)] is believed to play a role in the regulation of cardiac growth in pathological conditions. For instance, previous reports demonstrated higher blood 5-HT levels in heart failure patients, which may be due to either clearance defect or enhanced secretion (7, 17). Furthermore, chronic administration of 5-HT in rodents causes cardiac hypertrophy (9, 13).

A large body of evidence supports a role for cell surface 5-HT2 receptors in cardiac hypertrophy. 5-HT2A and 5-HT2B receptors are expressed in the heart and are present on both cardiomyocytes and cardiac fibroblasts (10). A role for 5-HT2B receptors in cardiac hypertrophy has been demonstrated using knockout mice (11). However, it is now clear that hypertrophic signaling mediated by 5-HT2B occurs in cardiac fibroblasts, and not on myocytes, through the release of pro-hypertrophic cytokines (10, 11). In addition, cardiac fibroblasts express 5-HT2A receptors, which mediate fibroblast migration, differentiation into myofibroblasts, and transforming growth factor-β1 secretion in response to 5-HT (24). Altogether, these studies have provided evidence for a role of 5-HT2A and 5-HT2B receptors present on cardiac fibroblasts in mediating prohypertrophic and profibrotic effects of 5-HT in the heart.

On the other hand, 5-HT can also act directly on cardiac myocytes, but its mechanism of action remains unclear. One study demonstrated that activation of 5-HT2 receptors produced robust hypertrophy in neonatal rat cardiomyocytes (5). However, the subtype of receptor involved (5-HT2A or 5-HT2B) and the signaling mechanisms associated were not clearly defined. In addition, a receptor-independent pathway has been characterized in cardiomyocytes (3). This mechanism required 5-HT uptake into cardiomyocytes, its degradation by monoamine oxidase type A (MAO-A), and the generation of reactive oxygen species (ROS) (3).

In the present study, we decided to evaluate the respective contribution of receptor-mediated and MAO-mediated mechanisms in cardiac hypertrophy in vitro and to delineate the intracellular determinants involved. Using H9C2 cardiomyoblasts, we provide evidence for a dual mechanism of action of 5-HT through 5-HT2A receptors and MAO-A, which signal through distinct hypertrophic pathways.

MATERIALS AND METHODS

Materials.

Culture medium, antibiotics, and serum were purchased from Invitrogen (Cergy Pontoise, France). Chemicals [5-HT, phenylephrine, pargyline, N-acetyl-l-cysteine (NAC), EGTA, ketanserin, SB206553, cyclosporin A (CsA), PD98059, 2,5-dimethoxy-4-iodoamphetamine hydrochloride (DOI), and BW723C86] were purchased from Sigma Aldrich (L'Isles d'Abeau Chesnes, France). M100907 {R-(+)-a-(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenyl)ethyl-4-piperidinemethanol]} was from Sanofi-Aventis (Paris, France). RO-41–1049 was from Roche (Neuilly-sur-Seine, France).

Cell culture.

The H9c2 cardiomyocyte-like cell line (American Type Culture Collection, Molsheim, France) is a rat embryonic myoblast-derived cell line commonly used as an in vitro model of cardiomyocyte biology that shows similar hypertrophic and apoptotic responses as those seen in primary adult and neonatal cardiomyocytes (12, 22). Cells were maintained in MEM, supplemented with 10% fetal calf serum and 4 mM glutamine. Cells were grown at subconfluence in an atmosphere of 5% CO2 and 37°C. They were used below the 20th passage.

Primary cultures of rat cardiomyocytes.

Cardiomyocytes were isolated from Sprague-Dawley rats weighing 200 g, using a protocol based on previously described procedures (6). Rats were handled in accordance with the procedures outlined in Council Directive 86/609/EEC. Rats were anesthetized with 45 mg/kg ip pentobarbital, and the heart was rapidly excised, mounted in a Langendorff apparatus, and perfused for 20 min with low-calcium solution (LCS) prewarmed at 37°C [117 mM NaCl, 5.7 mM KCl, 4.4 mM NaHCO3, 1.5 mM KH2PO4, 1.7 mM MgCl2, 11.7 mM d(+)-glucose, 21 mM HEPES, 20 mM taurine, 10 mm creatine (pH 7.2)]. The solution was then quickly changed to LCS plus 1 mg/ml collagenase type I, 0.03 mg/ml dispase (Worthington Biochemicals), and 1 mg/ml albumin for 10 min. The heart was minced, and the pieces were stirred in LCS. Cardiomyocytes present in the supernatant were purified by gravity sedimentation, collected, and stored in LCS supplemented with 1 mg/ml BSA and 4% penicillin-streptomycin at room temperature. Calcium was then added stepwise up to a concentration of 312 μM at room temperature. Cardiomyocytes were collected after sedimentation at RT and plated on culture plate covered with 100 μg/ml laminin in DMEM containing penicillin-streptomycin. After 2 h, the dishes were washed twice with DMEM to eliminate nonadherent cells, and cells were stimulated for 36 h.

Plasmid constructs and transfection.

Plasmid construct with luciferase reporter gene linked to promoter for skeletal muscle α-actin (SK-act-Luc) was provided by Dr M. D. Schneider (18). The luciferase reporter plasmid driven by four nuclear factor of activated T-cell (NFAT) consensus binding sites (NFAT-Luc) and the control plasmid pFR-Luc were obtained from Stratagene (Agilent Technologies, Massy, France). ANF-Luc reporter gene was made by amplification of atrial natriuretic factor (ANF) promoter (−638 to +65) from rat genomic DNA, sequencing, and cloning into pGL3-basic vector (Promega France, Charbonnières, France). Stable cell lines were obtained by transfection of H9C2 cells using lipofectamine 2000 reagent (Invitrogen, Cergy Pontoise, France). Mass pools of stable transfectants were selected in growth medium containing 0.75 mg/ml G418. Cells were serum starved in MEM supplemented with 1% dialyzed fetal bovine serum 24 h before experiments.

Luciferase detection assay.

Expression of luciferase was monitored with the Luciferase Assay System (Promega France) on a luminometer (Mithras LB940, Berthold Technologies, Thoiry, France).

Assays of MAO activity.

Crude extracts of proteins from H9C2 cells (40 μg) were incubated at 37°C for 20 min, in the presence of 400 μM of [14C]serotonin or 100 μM of β-[14C]phenylethylamine to measure MAO-A and MAO-B activities, respectively (16). To define nonspecific activities, MAO-A inhibitor clorgyline and MAO-B inhibitor deprenyl were used (0.1 μM). The reaction was ended by the addition of 0.1 ml of HCl, 4 N at 4°C. The reaction product was extracted (efficiency 92%) with 1 ml of ethyl acetate/toluene (vol/vol), and the radioactivity contained in the organic phase was counted in a liquid scintillation spectrometer.

RT-PCR experiments.

Extraction of RNA was performed using column affinity purification (Masherey-Nagel, Hoerdt, France). First-strand cDNA was synthesized using the superscript II RT-PCR system (Invitrogen, Cergy Pontoise, France) with random hexamers. Negative controls without reverse transcriptase were made to verify the absence of genomic DNA contamination. PCR amplification was performed using platinum-taq DNA enzyme (Invitrogen) in a volume of 20 μl. The primers were as followed: GAPDH-F, ATGGTGAAGGTCGGTGTGAACG; GAPDH-R, CTTCCACGATGCCAAAGTTGTC; 5-HT2A-F, TCCAGAACCAAAGCCTTCCTGA; 5-HT2A-R, TGCAGGATTCTTTGCAGATGAC; 5-HT2B-F, GTCTGGAACTGGACTGAGTCAC; 5-HT2B-R, GGGAAATGGCACAGAGATGCATGA; 5-HT2C-F, CCAGTAGCAGCTATAGTAACTGAC; 5-HT2C-R, ATGGCCTTAGTCCGCGAATTGAAC; 5-HT4-F, GGGACAGGCAGCTCAGGAAAATAA; 5-HT4-R, GGCATGCTCCTTAGCAGTGACATA; 5-HT7-F, GAGCAGATCAACTATGGCAGAG; 5-HT7-R, GTGGAGTAGATCGTGTAGCCAA; 5-HTT-F, AGTGCTGTCAGAGTGTAAGGA; 5-HTT-R, GCGCCCAGGCTATGATGGTGTT; MAOA-F, ATGACGGATCTGGAGAAGCC; MAOA-R, TGCCTCACATACCACAGGAAC. PCR amplicons were resolved on an agarose gel (1%). To verify primer efficiencies, PCR experiments were performed in parallel on rat brain cDNAs. All couples of primers demonstrated expression in the brain. Real-time PCR was performed in an ABI prism 7000 system (Applied Biosystem, Courtaboeuf, France) in 96-well plates. A 1/10 dilution of cDNA (5 μl) from RT reaction was mixed with specific primers and SYBRgreen mix (Eurogentec, Angers, France). Among different housekeeping genes, GAPDH gene was chosen for normalization by the threshold cycle (2ΔCt) method based on its stability (GeNorm tool). The primers were as followed: ANF-F, GGAAGTCAACCCGTCTCAGAGA; ANF-R, GCAGAGCCCTCAGTTTGCTT; αSK-actin-F, CCCAGGGCCAGAGTCAGA; αSK-actin-R, GAGCCGTTGTCACACACAAGAG.

Hypertrophy experiments.

For [3H]leucine experiments, H9C2 cells were plated in 24-well plates at subconfluence and treated with agonists for 24 h in the presence of 1 μCi/ml of [3H]leucine (Amersham, Les Ullis, France). At the end of the incubation period, cells were washed with PBS, incubated in 5% trichloroacetic acid (TCA), and neutralized with 0.25 M NaOH. Cell extracts were counted in a scintillation counter.

Proliferation experiments.

For [3H]thymidine experiments, H9C2 cells were plated in 24-well plates at subconfluence and treated with agonists for 24 h in the presence of 0.5 μCi/ml of [3H]thymidine (Amersham, Les Ullis, France). At the end of the incubation period, cells were washed with PBS, incubated in 5% TCA, and neutralized with 0.25 M NaOH. Cell extracts were counted in a scintillation counter. For cell counting, H9C2 were plated on six-well plates, treated with agonists for 24 h, and counted with a cell counter (Coulter Cell Counter, Beckman, Roissy, France).

Protein immunoblotting.

Cells grown in 10-mm plates were washed and lysed in RIPA buffer (25 mM Tris, pH 7.6, 150 mM NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, 0.1% SDS) with protease inhibitors. Proteins (30 μg) were electrophoresed through 10% Tris/glycine gels and transferred to polyvinylidene difluoride. A polyclonal ERK2 antibody (Santa Cruz Biotechnology, TEBU, Le Perray en Yvelines, France) and a monoclonal phosphorylated ERK1/2 antibody (Santa Cruz Biotechnology, TEBU) were used to determine expression of these peptides in H9C2 cells. The proteins were visualized using a horseradish peroxidase-linked secondary antibody and ECL detection (GE Healthcare, Saclay, France).

Statistical analysis.

Results are expressed as means ± SE. Statistical comparison of the data was performed using the t-test for comparison between two groups, or one-way analysis of variance and the post hoc Tukey test for comparison of more than two groups. A value of P < 0.05 was considered significant.

RESULTS

5-HT stimulates protein synthesis and hypertrophic gene expression in cardiomyoblasts.

Chronic exposure of H9C2 cells to 5-HT dose-dependently increased [3H]leucine incorporation (Fig. 1A), total protein content (Fig. 1B), ANF mRNA expression (Fig. 1C), and ANF-luc promoter construct (Fig. 1D), indicating that H9C2 cells exposed to 5-HT exhibit several major characteristics of hypertrophy. αSK-actin mRNA and luciferase constructs were also dose-dependently activated by 5-HT (data not shown). Maximal hypertrophic signal was observed with 10 μM 5-HT and was comparable to phenylephrine, an α1-adrenergic receptor agonist used as a positive control. 5-HT (10 μM) did not trigger proliferation of cardiomyoblasts, as measured by [3H]thymidine incorporation and cell number (Supplemental Fig. 1). (The online version of this article contains supplemental data.)

Fig. 1.

Protein synthesis and hypertrophic gene expression by 5-hydroxytryptamine (5-HT) in cardiomyoblasts. [3H]leucine incorporation (A) and total proteins (B) were used to measure dose-dependent effects of 5-HT compared with 10 μM phenylephrine (Phe). Atrial natriuretic factor (ANF) mRNA expression (C) and ANF-luc reporter construct (D) were used to determine the effects of 5-HT on the hypertrophic gene program in cardiomyoblast, compared with 10 μM Phe. For promoter construct, results are expressed as relative light units (RLU) in percentage over unstimulated control (CT). The minimal promoter construct pFR-luc is used as a negative CT for 5-HT stimulation. Values are means ± SE of 3–4 independent experiments performed in triplicate. *P < 0.05, **P < 0.01, ***P < 0.001 compared with CT or vs. indicated values.

Participation of MAO-A in hypertrophic response mediated by 5-HT.

H9C2 express MAO-A but not MAO-B, as known for adult rat cardiomyocytes (Fig. 2A). To assess the importance of MAO-A in protein synthesis and hypertrophic gene expression mediated by 5-HT, we measured [3H]leucine incorporation and αSK-actin-luc in the presence of an irreversible MAO inhibitor, pargyline (10 μM). Pargyline blocked both parameters at high doses of 5-HT (10 μM), but not at low doses (1 μM) (Fig. 2B). Hydrogen peroxide generation by MAO-A has been shown to be responsible for the hypertrophic response mediated by 5-HT in adult rat ventricular myocytes. Indeed, application of the ROS scavenger NAC (1 mM), and application of the selective MAO-A antagonist RO-41–1049 (1 μM), blocked [3H]leucine incorporation and ANF-luc stimulation with 10 μM 5-HT (Fig. 2C).

Fig. 2.

Effect of monoamine oxidase (MAO)-A inhibition on protein synthesis and hypertrophic gene expression mediated by 5-HT. A: MAO-A activity was determined using 400 μM [14C]5-HT as a specific substrate, and MAO-B activity using 100 μM β-[14C]phenylethylamine. Results are expressed as difference between total and nonspecific activities, with the latter being defined in the presence of selective inhibitors clorgyline (10−7 M) and deprenyl (10−7 M) for MAO-A and MAO-B, respectively. B: effect of MAO inhibitor pargyline (Parg; 10 μM) on the hypertrophic response to 5-HT at low (1 μM) and high doses (10 μM) using [3H]leucine incorporation and skeletal muscle (SK)-actin-luc stimulation. C: effect of MAO-A selective inhibitor RO-41–1049 (1 μM) and N-acetyl-cysteine (NAC; 1 mM) on 5-HT mediated [3H]leucine incorporation and ANF-luc stimulation. Results are expressed as percentage of CT cells and are means ± SE of 4 independent experiments performed in triplicate. *P < 0.05, **P < 0.01, ***P < 0.001 compared with CT or vs. indicated values.

Contribution of 5-HT receptors in cardiac cell hypertrophy.

Results observed in Fig. 2 suggest that another mechanism, activated by low levels of 5-HT (1 μM), may be present in cardiac cells. In addition to MAO-A and 5-HT transporter (5-HTT), H9C2 cells express mRNA for serotonin 5-HT2A and 5-HT2B receptors (Fig. 3A). Real-time PCR analysis demonstrated that Ct values were 23.5 for 5-HT2A, 30.7 for 5-HT2B, 27.9 for MAO-A, and 31.5 for 5-HTT in H9C2 cells compared with GAPDH (15.8), whereas 5-HT2C, 5-HT4, and 5-HT7 were undetectable. Since 5-HT2A receptors are predominantly expressed in H9C2 cells, we measured protein synthesis and hypertrophic gene expression using M100907 (0.1 μM), a selective 5-HT2A receptor antagonist. 5-HT2A receptor blockade antagonized [3H]leucine incorporation (Fig. 3B) and ANF-luc stimulation (Fig. 3C) at low and high doses of 5-HT. Another 5-HT2A receptor antagonist ketanserin had the same effect (Fig. 1, D and E). On the other hand, 5-HT2B selective antagonist SB-206553 (0.1 μM) induced small but nonsignificant inhibition of the 5-HT response.

Fig. 3.

Role of serotonin receptors in 5-HT-mediated response. A: RT-PCR expression of 5-HT receptors, 5-HT transporter (5-HTT), and MAO-A in H9C2 cells. This is a representative gel of three independent experiments. Negative CTs were performed without reverse transcriptase (RT−). B and C: effect of selective 5-HT2A antagonist M100907 (0.1 μM) on the hypertrophic response mediated by 5-HT (1 and 10 μM) using [3H]leucine incorporation (B) and ANF-luc stimulation (C). D and E: effect of 5-HT2A antagonist ketanserin (Ket; 0.1 μM) and 5-HT2B antagonist SB206553 (SB; 0.1 μM). Results are expressed as percentage of CT cells and are means ± SE of 4 independent experiments performed in triplicate. *P < 0.05, ***P < 0.001 compared with CT or vs. indicated values.

Participation of 5-HT2A receptors in the hypertrophic response mediated by 5-HT was verified in primary culture of adult ventricular myocytes. RT-PCR analysis indicated that ventricular myocytes, as cardiomyoblasts, expressed 5-HT2A (Ct = 25.3), 5-HT2B (Ct = 33.9), 5-HTT (Ct = 31.5), and MAO-A (Ct = 29.9) (Fig. 4A). Incubation of cardiomyocytes with 5-HT (10 μM) for 36 h promoted increase in cell area, which was fully inhibited by 5-HT2A antagonist M100907 (Fig. 4B).

Fig. 4.

Hypertrophic effect of 5-HT on adult rat ventricular myocytes. A: RT-PCR expression of 5-HT receptors, 5-HTT, and MAO-A in adult ventricular myocytes. This is a representative gel of three independent experiments. Negative CTs were performed without RT (RT−). B: mean cell surface area of ventricular myocytes incubated with 5-HT (10 μM) or Phe (10 μM) for 36 h. Photographic images of cells were digitized, and the area of 200 cells per condition from 4 independent experiments were analyzed with computer-assisted planimetry. Values are means ± SE and are expressed as the percentage of CT value. *P < 0.05, **P < 0.01 compared with CT or vs. indicated values.

5-HT2A receptor activates the calcineurin/NFAT pathway.

5-HT2A receptors are Gαq-coupled receptors, which are known to promote intracellular Ca2+ elevation, followed by calcineurin/NFAT pathway activation. In H9C2 cells, application of the calcium scavenger EGTA (5 mM) or the calcineurin inhibitor CsA (1 μM) blocked 5-HT-mediated hypertrophic gene expression (Fig. 5A). These results prompted us to measure direct activation of NFAT, using a reporter construct harboring four NFAT binding sites in front of the luciferase gene. 5-HT produced robust activation of NFAT, which was reversed by EGTA and CsA (Fig. 5B). Activation of NFAT by 5-HT was blocked by M100907, but not by SB-206553, indicating that 5-HT2A receptors were responsible for this activation (Fig. 5C). In addition, 2,5-dimethoxy-4-iodoamphetamine hydrochloride (DOI), a 5-HT2A receptor agonist, but not BW723C86 (a 5-HT2B receptor agonist), was sufficient to promote NFAT activation (Fig. 5D). Interestingly, application of pargyline failed to inhibit NFAT activation by 5-HT, which confirms that this pathway is independent of MAO-A.

Fig. 5.

Role of the calcineurin/nuclear factor of activated T cell (NFAT) pathway in 5-HT-mediated response in H9C2 cells. A: ANF-luc reporter stimulation by 5-HT (10 μM) or 5-HT in DMSO (1%) in the presence of cyclosporine A (CsA; 1 μM) or EGTA (5 mM). B and C: NFAT-luc reporter stimulation by 5-HT (10 μM) or 5-HT in DMSO (1%) in the presence of CsA (1 μM), EGTA (5 mM), M100907 (0.1 μM), SB206553 (SB; 0.1 μM) or Parg (10 μM). D: NFAT stimulation with 2,5-dimethoxy-4-iodoamphetamine hydrochloride (DOI; 1 μM) and BW723C86 (BW; 1 μM). Values are means ± SE and are expressed as the percentage of CT value. *P < 0.05, **P < 0.01, ***P < 0.001 compared with CT or vs. indicated values.

Role of MAPK pathway in hypertrophic response mediated by 5-HT.

ERK protein kinases, in coordination with calcineurin/NFAT, constitute a major route for hypertrophic growth of cardiomyocytes. In our experiments, we demonstrated that MEK inhibitor PD98059 fully blocked 5-HT (10 μM)-induced [3H]leucine incorporation and ANF-luc stimulation in cardiomyoblasts (Fig. 6A). 5-HT produced robust phosphorylation of ERK1/2 kinases, with an early peak at 5 min and an attenuated stimulation, which remained stable until 60 min (Fig. 5B). We investigated whether ERK phosphorylation by 5-HT was mediated through 5-HT2A receptor-dependent or MAO-dependent pathway. Blockade of MAO-A with pargyline or NAC completely prevented ERK phosphorylation by 5-HT, whereas M100907 had no effect (Fig. 6, C and D). Therefore, as previously demonstrated in adult ventricular myocytes, ERK MAPK pathway is dependent on MAO-A activation in our model.

Fig. 6.

MEK inhibition prevents hypertrophic effect of 5-HT in H9C2 cells. A: [3H]leucine experiments and ANF-luc stimulation were performed to assess the impact of MEK inhibition by PD98059 (50 μM) in H9C2 cells stimulated by 5-HT in EtOH (1%). B: immunoblotting experiments performed with ERK phosphorylation (pERK) antibody demonstrated that 5-HT induced strong activation of ERK with a pick at 5 min. Immunoblots are representative of four experiments. Histogram is mean ± SE of 4 experiments. C and D: activation of pERK2 with 5-HT (5 min) is inhibited by Parg (10 μM) and NAC (5 mM), but not M100907 (0.1 μM). Immunoblots are representative of 4 experiments, as depicted in histograms. *P < 0.05, **P < 0.01 compared with CT or vs. indicated values.

DISCUSSION

In the present study, we demonstrated that extracellular 5-HT concentration was determinant for activation of receptor-dependent or MAO-A-dependent hypertrophic pathways. At low doses, 5-HT stimulated cardiac cell hypertrophy through activation of 5-HT2A receptors and stimulation of the Ca2+/calcineurin/NFAT pathway. Higher doses of 5-HT produced an additional hypertrophic effect that was dependent on MAO-A/H2O2/ERK pathway (Fig. 7).

Fig. 7.

Schematic representation of 5-HT-mediated signaling pathways to produce hypertrophy.

Experiments performed at low 5-HT concentrations allowed us to identify, for the first time, 5-HT2A receptors as the mediators of the receptor-dependent hypertrophic activity of 5-HT in cardiac cells. The role of these receptors in cardiac hypertrophy has been supported by various results in vivo. Indeed, we showed that, in a mouse model of increased plasma 5-HT, 5-HT2A but not 5-HT2B receptors were involved in pressure overload hypertrophy (15). Other studies showed that 5-HT2A receptors mediated positive inotropic response in ventricular myocytes and were overexpressed in different models of cardiac hypertrophy, along with human heart failure (4, 20). In addition, 5-HT2A receptor blockade reverses hypertrophy during pressure overload and reduces left ventricular hypertrophy during hypertension (8, 15). 5-HT2B receptors have also been implicated in ventricular hypertrophy. However, converging evidence suggests that 5-HT2B receptors on nonmyocytes contribute to ventricular hypertrophy (10, 11). Indeed, it has been shown that 5-HT2B receptors on cardiac fibroblasts mediate the release of IL-6 and indirectly induce cardiomyocyte hypertrophy. Our results supply new insight on the cell mechanisms involved in the receptor-mediated hypertrophy of cardiac cells by 5-HT. Indeed, in addition to the recognition of 5-HT2A receptors as the main target of 5-HT, we identified the Ca2+/calcineurin/NFAT pathway as a major determinant in cardiac cell hypertrophy. Calcineurin, acting via its downstream effector NFAT, plays a key role in cardiac hypertrophy and is sufficient to initiate and propagate hypertrophic growth of the heart in animal models (23).

Higher doses of 5-HT promoted stronger hypertrophic stimulation that involved an additional pathway requiring ROS generation by MAO-A and activation of ERK MAPK. We found that 5-HT2A receptor stimulation was still required for hypertrophic response generated with high doses of 5-HT. Therefore, ERK activation may not be sufficient to propagate the hypertrophic response alone and may act in cooperation with the calcineurin/NFAT pathway. At present, the potential interconnections between calcineurin and ERK signaling in our model are uncertain. Since MAO-A inhibitor did not prevent 5-HT-induced NFAT activation, we can assume that cooperation did not occur directly on calcineurin activation, NFAT dephosphorylation, or NFAT translocation into the nucleus. One possibility would be that ERK activation allows the formation of transcriptional complexes with NFAT in cardiomyocytes. In an interesting study, Sanna et al. (21) demonstrated that ERK1/2 signaling could enhance hypertrophic gene expression by the induction of the transcription factor activator protein-1, which functioned as an essential NFAT-interacting partner in cardiac cells.

Our results show that extracellular concentrations of 5-HT are critical for discrimination between 5-HT2A receptor and MAO-A hypertrophic pathways. At present, the relative weight of 5-HT2A and MAO-A pathways in ventricular hypertrophy in vivo is still unclear. Peripheral 5-HT is mainly stored in platelets and is released during platelet activation. Therefore, the extent of platelet activation and 5-HT release may determine the activation of 5-HT2A receptor or MAO-A pathways in vivo. This hypothesis is supported by our laboratory's recent results showing that, in the case of a moderate increase in circulating 5-HT, 5-HT2A receptor but not MAO-A are involved in pressure overload hypertrophy (15). On the other hand, we have also shown that, in the model of cardiac ischemia-reperfusion, which could be considered as an extreme model of platelet activation and 5-HT release, ventricular damage is fully dependent on MAO-A/H2O2 pathway (2, 19). Based on our results, it is conceivable that MAO-A hypertrophic pathway may be relevant in pathological situations associated with sustained platelet activation and/or increase in cardiac MAO-A activity, such as aging or during the transition to heart failure (14, 16).

In conclusion, we identified a dual mechanism of action of 5-HT in cardiac hypertrophy via a MAO-A and receptor-dependent pathway. We produced the first evidence for a role of 5-HT2A receptor in cardiomyocyte hypertrophy. Moreover, we found that MAO-A activation generated a hypertrophic response through ERK stimulation, whereas the 5-HT2A receptor acted through calcineurin/NFAT activation in cardiomyoblasts. These results open new avenues for the comprehension of the mechanisms of action of 5-HT in the heart.

GRANTS

This work was supported by grants from Institut National de la Santé et de la Recherche Médicale, Agence Nationale pour la Recherche, Région Midi-Pyrénées, and Fondation pour la Recherche Médicale.

REFERENCES

View Abstract