Identification and localization of a skeletal muscle secrotonin 5-HT2A receptor coupled to the Jak/STAT pathway.

The neurotransmitter serotonin mediates a wide variety of peripheral and central physiological effects through the binding to multiple receptor subtypes (Wilkinson, L. O., and Dourish, C. T. (1991) in Serotonin Receptor Subtypes: Basic and Clinical Aspects (Peroutka, S. J., ed) Vol. 15, pp.147-210, Wiley-Liss, New York). Among them, serotonin 5-HT2A receptors are known to activate the phospholipase C-beta second messenger pathway (Peroutka, S. J. (1995) Trends Neurosci. 18, 68-69). We identified and localized in rat skeletal muscle myoblasts a functional serotonin 5-HT2A receptor. This receptor was detected on the plasma membrane, in myoblasts, and at the level of T-tubules in contracting myotubes. Binding of serotonin to its receptor increases the expression of genes involved in myogenic differentiation. Unexpectedly, the 5-HT2A receptor is able to activate another signaling pathway; it triggers a rapid and transient tyrosine phosphorylation of Jak2 kinase in response to serotonin. Jak2 auto-phosphorylation is followed by the tyrosine phosphorylation of STAT3 (signal transducers and activators of transcription) and its translocation into the nucleus. We also find that the 5-HT2A receptor and STAT3 co-precipitate with Jak2, indicating that they are physically associated. We conclude that the serotonin 5-HT2A receptor identified in skeletal muscle myoblasts is able to activate the intracellular phosphorylation pathway used by cytokines. The presence of serotonin receptors in T-tubules suggests a role for serotonin in excitation-contraction coupling and (or) an effect in skeletal muscle fiber repairing.

Serotonin (5-hydroxytryptamine, 5-HT) 1 is a neurotransmitter that mediates diverse central and peripheral physiological responses by interacting with multiple serotonin receptor subtypes (1). The physiological effects of serotonin are mediated by at least four families of 5-HT receptors that have been distinguished pharmacologically, depending on the second messenger they are coupled to. The first family, including 5-HT 1 and 5-HT 5 receptor subtypes, interacts negatively with adenylate cyclase, whereas the second family (5-HT 2 ) is coupled to the activation of phospholipase C-␤/protein kinase C. The 5-HT 3 receptor is a ligand-gated ion channel, and the family including 5-HT 4 , 5-HT 6 , and 5-HT 7 receptor subtypes activates adenylate cyclase (2). The receptors of the 5-HT 2 subfamily are implicated in many central physiological functions of serotonin, such as neuronal sensitization to tactile stimuli and mediation of hallucinogenic effects of lysergic acid diethylamide, as well as in peripheral cardiovascular effects, e.g. contraction of blood vessels and shape change in platelets. The 5-HT 2A (formerly 5-HT 2 ) receptor shares an overall 49% sequence identity with the 5-HT 2C (formerly 5-HT 1C ) receptor, but they present different patterns of expression in the brain (3). Transfection of both 5-HT 2C and 5-HT 2A receptors in NIH3T3 fibroblasts results in cellular transformation and focus formation (3). The mouse 5-HT 2B receptor, mainly expressed in the cardiovascular system, gut, and developing brain (4), shares the highest degree of homology with the other 5-HT 2B receptors cloned from the rat fundus or from human libraries (5). The 5-HT 2B receptor has been classified as a ligand-dependent proto-oncogene, since its expression is necessary and sufficient to induce tumor formation in nude mice (6).
Serotonin 5-HT 2A receptors have been mainly localized in frontal cortex (3), blood platelets (7), and aortic smooth muscle cells (8). To date, serotonin receptors have never been isolated in skeletal muscle cells. The present study reports the identification and localization of a functional skeletal muscle serotonin 5-HT 2A receptor in rat fetal myoblasts. This receptor mediates the serotonin-induced up-regulation of the transcription factor myogenin, and the neuronal glucose transporter isoform GLUT3, both expressed during myogenic differentiation. Furthermore, we show that the skeletal muscle serotonin 5-HT 2A receptor is capable of stimulating the Jak/STAT pathway.

MATERIALS AND METHODS
Cells and Culture Conditions-Primary cultures of myoblasts from 19-day-old rat fetuses were grown as described previously (9) with the following modifications. Myoblasts were separated from muscle fibers by a treatment (20 min, 37°C) with 0.15% protease (Sigma) in Ham's F-12 medium containing 10% fetal calf serum. Myoblasts were plated at a density of 1.5 ϫ 10 4 cells/ml on gelatin-coated flasks, in minimal essential medium/199 medium (2/1) containing 10% horse serum and 0.2% Matrigel (Becton-Dickinson). Serum and Matrigel were removed 24 h before the beginning of the experiments. Media were purchased from Life Technologies, Inc.; fetal calf serum and horse serum were obtained from Boehringer Mannheim.
Cloning of a cDNA Coding for a Skeletal Muscle 5-HT 2A Receptor-Using total RNA isolated from 3-day-cultured fetal myoblasts as a template, first strand cDNA was synthesized by reverse transcription, using a cDNA synthesis kit according to the manufacturer's instructions (Perkin Elmer). For amplification of the cDNA fragments by the PCR, 21-mer oligonucleotide primers were designed, based on a portion of the third intracytoplasmic loop of the rat 5-HT 2A receptor cDNA (3) (GenBank™ accession no. M30705). Forward (sense) primer was 5Ј-ACC TAC TTC CTG ACT ATC AAG-3Ј (nucleotides 816 -837). Reverse (antisense) primer was 3Ј-GCC CAG CAC CTT GCA CGC CTT-5Ј (nucleotides 1016 -1037). PCR fragments were subcloned into the BamHI-EcoRI sites of a Bluescript pBS-SK(ϩ) plasmid (Promega), amplified in * This work was supported in part by a grant from the Association Française contre les Myopathies. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
RNA Extraction and Northern Blotting-Myoblasts were treated with 5⅐10 Ϫ6 M serotonin hydrochloride (Sigma) in the presence of the 5-HT 2A receptor antagonist ketanserin (10 Ϫ6 M) in some experiments (Research Biochemicals Inc.) Total RNA was extracted from the cells by the guanidine thiocyanate method (10), then electrophoresed and transferred as described previously (9). Northern blots were hybridized with 32 P-labeled cDNA probes coding for GLUT3, GLUT1 (a gift of Dr. G. Bell, Howard Hughes Medical Institute, University of Chicago, Chicago, IL), myogenin (provided by Dr. W. Wright, Southwestern Medical Center, University of Texas, Dallas, TX), and serotonin 5-HT 2A receptor (cDNA cloned by reverse transcription-PCR).
Localization of the 5-HT 2A Receptor by Immunofluorescence-Fetal myoblasts were grown on gelatin-coated Permanox four-chamber slides (Lab-Tek, Nunc) for 3 days (fusioning myoblasts) or 8 days (contracting myotubes) as described previously (9). The serotonin 5-HT 2A receptor was detected using a monoclonal anti-5-HT 2A receptor antibody (PharMingen). Cells were incubated overnight at 4°C in diluted antibody (dilution 1/500 in PBS, 0.2% gelatin), then washed three times with PBS, and treated for 1 h with fluorescein isothiocyanate-conjugated sheep anti-mouse IgG (dilution 1/128, Sigma). The slides were mounted in a glycerol/PBS mounting medium (Cityfluor), and confocal laser scanning microscopy was performed, using a Leica confocal imaging system (TCS-4D) and an immersion lens (63ϫ, numerical aperture 1.4 plan Apochromat). A focal series of up to 18 sections apart (0.5 m between two adjacent sections) was collected for each specimen and then processed to produce single composite images (extended focus). Micrographs were printed directly from the computer on a dye sublimation printer (Colorease, Eastman Kodak Co.).
Antibodies-Polyclonal anti-mouse Jak2 and anti-human/mouse Jak2/protein A-agarose were purchased from Upstate Biotechnology Inc. Monoclonal anti-STATs antibodies and biotinylated recombinant anti-phosphotyrosine (RC 20-B) were purchased from Transduction Laboratories. Anti-rabbit and anti-mouse Ig-peroxidase-linked whole antibodies were obtained from Amersham.
Nuclear Protein Extract Preparation-Myoblasts were exposed to 5⅐10 Ϫ6 M serotonin for 0, 1, 5, 15, 30, or 60 min. Nuclei were isolated according to the protocol described by Shapiro et al. (11), with the following modifications. Cells were washed with ice-cold phosphatebuffered saline, recovered by scraping, then pelleted and resuspended in 300 l of hypotonic buffer (10 mM Hepes, pH 7.9, 0.1 mM EDTA, 0.1 mM EGTA, 0.75 mM spermidine, 0.15 mM spermine, 10 mM KCl, 1 mM dithiothreitol, 0.5 mM PMSF, 0.5 g⅐ml Ϫ1 leupeptin, 0.5 g⅐ml Ϫ1 pepstatin A, 1% aprotinin). Cells were broken by 3 strokes of the tight pestle of a Dounce homogenizer, after addition of hypotonic buffer containing 0.6% Nonidet P-40. Sucrose restore buffer (50 mM Hepes, pH 7.9, 0.2 mM EDTA, 10 mM KCl, 1 mM dithiothreitol, 0.5 mM PMSF, 1 g⅐ml Ϫ1 pepstatin A, 2 g⅐ml Ϫ1 leupeptin, and 5 g⅐ml Ϫ1 aprotinin, 67.5% (w/v) sucrose) was rapidly added and the homogenate mixed with 2 strokes of the loose pestle of the homogenizer. After centrifugation (1 min at 12,000 ϫ g, 4°C), the nuclear pellet was resuspended in 10% glycerol nuclear resuspension buffer. The nuclear resuspension was treated according to Shapiro. Nuclear proteins (40 g) were electro-phoresed on a 7.5% polyacrylamide-SDS gel, transferred to nitrocellulose, and immunoblotted as described for immunoprecipitation assays. 2A Receptor-Using total RNA isolated from 3-day-cultured myoblasts as a template, the first strand cDNA was synthesized by reverse transcription, as described under "Materials and Methods." Since the homology within the transmembrane domains of 5-HT 2 receptors is over 80%, we chose, for amplification of the cDNA fragment by PCR, two oligonucleotide primers in a portion of the nonconserved domain of the 5-HT 2A receptor, the third intracytoplasmic loop (3). Sequencing of the amplified 221-base pair cDNA fragment presented 100% homology with nucleotide 816 -1037 of the rat 5-HT 2A receptor cDNA (3). Although poly(A) ϩ RNA from the cortex revealed two discrete transcripts between 5 and 6 kb, corresponding to 5-HT 2A receptor mRNAs (3), this cDNA recognized in total RNA (40 g) from rat fetal myoblasts a major 6.0-kb and a minor 4.0-kb transcript (Fig. 1B).

Cloning of a cDNA Coding for the Serotonin 5-HT
Time Course of Serotonin Effect on GLUT3, GLUT1, and Myogenin mRNA Expression-The neuronal isoform GLUT3 is the main glucose transporter isoform expressed during fusion of myoblasts into myotubes, e.g. between day 3 and day 4 of culture (9). Preliminary dose-response experiments showed that 5⅐10 Ϫ6 M serotonin induced a maximal increase in mRNA , GLUT3, or myogenin cDNA probes. Blots were exposed overnight, except for the 5-HT 2A receptor, which was exposed for 3 nights. expression, in 3-day-cultured myoblasts (data not shown). A time course of serotonin effect was performed after treatment of 3-day-cultured myoblasts with 5⅐10 Ϫ6 M serotonin, for times ranging from 1.5 to 24 h. In control cells, GLUT3 mRNA expression increased transiently between day 3 and day 4, as expected (9). Nevertheless, a 2-3-fold increase in GLUT3 mRNA level was observed as soon as 1.5 h after addition of serotonin to the culture medium. The effect, maximal after 3 h, decreased to reach the control level after 9 h. A similar time course was observed on myogenin mRNA expression, whereas the ubiquitous glucose transporter isoform GLUT1 was quite unaffected (Fig. 1A).
Inhibition of Serotonin-induced mRNA Up-regulation-To determine if serotonin-induced up-regulation of mRNAs was mediated by the 5-HT 2A receptor, myoblasts were exposed to serotonin in the absence or presence of serotonin receptor antagonists, and Northern blots were hybridized with 5-HT 2A receptor, GLUT3 and myogenin cDNA probes. Within 6 h, serotonin induced a 5-fold increase in 5-HT 2A receptor mRNAs, and a 2-3-fold increase in GLUT3 and myogenin mRNAs (Fig.  1B). This effect was completely abolished in the presence of 1 M ketanserin, a specific 5-HT 2A receptor antagonist (Fig. 1B), whereas zacopride, a 5-HT 3 inhibitor, was inefficient (data not shown). Taken together, these results show that serotonin is able to increase the expression of genes expressed during myogenic differentiation, through the skeletal muscle 5-HT 2A receptor.
Cellular Localization of the 5-HT 2A Receptor-Fetal myoblasts and contracting myotubes were stained for the serotonin 5-HT 2A receptor by immunofluorescence, and images were analyzed by confocal laser scanning microscopy. The 5-HT 2A receptor was localized on the plasma membrane, in fusioning myoblasts (Fig. 2a). Later on during myogenic differentiation, 5-HT 2A receptors appeared at the level of deep invaginations of the plasma membranes, the transverse tubules (T-tubules), in contracting myotubes (Fig. 2, b-h). T-tubules are known to transmit the electrical impulses that trigger muscular contraction from the cell surface to the sarcoplasmic reticulum, and to facilitate the distribution of substrates to the myofibers (12). The presence of serotonin receptors in T-tubules of contracting myotubes suggests a role for serotonin in excitation-contraction coupling, and (or) a trophic role for the skeletal muscle fiber.
Co-precipitation of the 5-HT 2A Receptor with Jak2 Kinase-Serotonin increased gene expression through a very rapid transduction pathway, since a 1-min stimulation of myoblasts by serotonin was sufficient to induce a 50% increase in GLUT3 mRNA levels 3 h later (data not shown). The intracellular Jak/STAT pathway transduces very quickly the signal of many cytokines and peptide growth factors (13). Ligand binding rapidly triggers tyrosine phosphorylation of STATs (14, 15) by receptor-associated tyrosine kinases of the Jak family (16 -18). To investigate whether serotonin was able to stimulate the Jak/STAT pathway, myoblasts were incubated with serotonin for 0, 5, 15, 30, or 60 min, and the association of Jak2 with the 5-HT 2A receptor was assessed. Cell lysates from serotonintreated myoblasts were immunoprecipitated with anti-Jak2 antibody, and, after SDS-PAGE electrophoresis, immunoblotted with anti-5-HT 2A receptor antibody. This antibody revealed in rat brain, used as a positive control, a strong band at 53 kDa, and a minor band at 58 kDa, probably due to posttranslational modifications (glycosylation or phosphorylation) of the receptor. In myoblasts, a major band corresponding to 55 kDa was detected. The association of Jak2 with the 5-HT 2A receptor was observed within 5 min after addition of serotonin, and was maximal after 15 min (Fig. 3A).
Serotonin-induced Phosphorylation of Jak2 Kinase-To de-termine whether Jak2 auto-phosphorylation was induced by serotonin stimulation, Western blots of cell lysates were immunoblotted with anti-phosphotyrosine antibody, then stripped and reprobed with anti-Jak2 antibody (Fig. 3B, a). In a second protocol, cell lysates were immunoprecipitated with anti-Jak2 antibody, then immunoblotted with anti-phosphotyrosine antibody (Fig. 3B, b). Tyrosine phosphorylation of Jak2 was maximal after 30 min and then decreased until 60 min. These results show that the 5-HT 2A receptor cloned in skeletal muscle myoblasts is functional and activates a pathway never described for serotonin. mediated Signal Transduction-The activation of Jak kinases leads to the recruitment and phosphorylation of STATs proteins (19,20). To determine which STATs proteins were phosphorylated after serotonin treatment, cell lysates were immunoprecipitated with anti-phosphotyrosine antibody and then immunoblotted with monoclonal antibodies raised against STAT proteins. STAT1 ␣/␤ (p91/84), STAT3 (p92), and STAT6 (interleukin-4-STAT) were expressed in fetal myoblasts (Fig. 4, A-C). A431 cells, stimulated for 5 min with epidermal growth factor (100 ng/ml), were used as a positive control for STAT1 and STAT3, and mouse 3T3 cells as a positive control for STAT6. Tyrosine phosphorylation of STAT3 occurred within 5 min of exposure of myoblasts to serotonin, and was maximal after 30 min (Fig. 4B). No tyrosine phosphorylation of STAT1 (Fig. 4A) or STAT6 (Fig. 4C) was observed. In our study, only STAT3 was tyrosine-phosphorylated in response to binding of serotonin to the 5-HT 2A receptor.

Phosphorylation of STAT Protein(s) Involved in Serotonin-
Association of Jak2, 5-HT 2A Receptor, and STAT3-To determine whether STAT3 was associated with the complex Jak2 kinase-5-HT 2A receptor in response to serotonin stimulation, the same blot as in Fig. 3A was stripped and then immunoblotted with anti-STAT3 antibody. Serotonin induced the coprecipitation of STAT3 with Jak2 within 5 min (Fig. 5A). The simultaneous co-precipitation of the 5-HT 2A receptor and STAT3 with Jak2 suggests that STAT3 is associated to the receptor complex, in response to serotonin stimulation.
Nuclear Translocation of STAT3-The activated STATs dimerize and translocate to the nucleus, where they directly activate target genes (13). To test whether serotonin induced the nuclear translocation of STAT3 in myoblasts, cells were exposed to serotonin for 0, 1, 5, 15, 30, or 60 min, and nuclear protein extracts were electrophoresed. Western blots were immunoblotted with anti-STAT antibodies. Serotonin induced the nuclear translocation of STAT3 within 5 min, the presence of STAT3 in the nucleus remaining at least 60 min (Fig. 5B). In contrast, no translocation of STAT1 or STAT6 was observed (data not shown). In fetal myoblasts, serotonin triggers the recruitment of STAT3 to the Jak2-5-HT 2A receptor complex, followed by the tyrosine phosphorylation of STAT3 and its translocation into the nucleus. DISCUSSION In this study, we tried to elucidate the mechanisms by which serotonin increased the expression of genes in relation with myogenic differentiation. We have shown that the effects of serotonin on skeletal muscle cells are mediated by a receptor of the 5-HT 2 receptor subfamily, namely the 5-HT 2A receptor. This receptor is located at the level of T-tubules in contracting myotubes.
The junctional T-tubules exert a dual role in muscle fibers, since they are involved both in facilitating substrate uptake, such as glucose (12), and excitation-contraction coupling. The presence of serotonin receptors in T-tubules underlies a trophic effect for serotonin, in skeletal muscle fibers, since serotonin increases glucose transport rate 2 and GLUT3 glucose transporter expression in fetal myoblasts. Trophic effects of serotonin have already been described. Serotonin, detected early in embryonic development (21), plays a role in cranofacial (22) and cardiovascular morphogenesis (23,24) by unknown molecular mechanisms. Some trophic functions of serotonin during embryogenesis have been related to the mitogenic and transforming properties of the 5-HT 2B receptor (6). Nevertheless, the hypothesis that serotonin could play a role in excitation-contraction coupling must be considered.
The serotonin 5-HT 2A receptor is a member of the G proteinlinked receptor superfamily known to activate the phospholipase C-␤/protein kinase C pathway (2). In fetal myoblasts, serotonin binding to the 5-HT 2A receptor resulted in the stimulation of another signaling pathway, the Jak/STAT pathway. The serotonin-induced association of Jak2 to the 5-HT 2A receptor allowed the recruitment of STAT3 to the receptor complex and its subsequent phosphorylation by the phosphorylated Jak2 kinase, and led to STAT3 nuclear translocation. A similar result has been reported in human interleukin-6-treated NJBC-T cells, where the rapid tyrosine phosphorylation of STAT3 correlated with the rapid nuclear translocation of STAT3, and the formation of STAT3-STAT3-DNA complexes (25). STAT3 was reported to increase the expression of genes that are expressed in response to tissue injury and inflammation, and are referred as acute phase response genes (26). Some fetal myoblasts persist in adult skeletal muscle as quiescent stem cells (muscle satellite cells) that are capable of differentiation to repair muscle fibers, in case of injury. In myoblasts, the nuclear translocation of STAT3, in response to serotonin, might be related to the up-regulation of myogenin, a transcription factor involved in myogenic differentiation. This point requires further studies.
The serotonin 5-HT 2A receptor we cloned in skeletal muscle is the second member of the seven-transmembrane domain G protein-coupled receptor superfamily able to activate the Jak/ STAT pathway, after the angiotensin AT1 receptor, in rat aortic smooth muscle cells (27,28). It is noteworthy that contractions of rat aortic smooth muscle cells are stimulated by serotonin through 5-HT 2A receptors. It could be of great inter-est to determine if serotonin stimulates also the Jak/STAT pathway in these cells. Recently, a mouse 5-HT 2B receptor transfected in fibroblasts was reported to induce serotonin stimulation of mitogen-activated protein kinase cascade (ERK2/ERK1) through the Ras pathway (6). Furthermore, serine/threonine phosphorylation mediated by the mitogen-activated protein kinases was shown to play a role in the activation of STATs, linking STAT and Ras pathways (25,29).
In conclusion, our results show that the Jak/STAT pathway is implicated in the serotonin-induced increase in mRNAs coding for proteins involved in myogenic differentiation, suggesting that this pathway may play a role in skeletal muscle ontogenesis, and (or) repairing. This novel localization and signal transduction for serotonin 5-HT 2A receptors deserve further studies, in particular a search for implication of serotonin in excitation-contraction coupling.