RESEARCH ARTICLE
303
A calcineurin- and NFAT-dependent pathway regulates Myf5 gene expression in skeletal muscle reserve cells Bret B. Friday and Grace K. Pavlath* Department of Pharmacology, Emory University School of Medicine, Atlanta, GA 30322, USA *Author for correspondence (e-mail:
[email protected])
Accepted 14 November 2000 Journal of Cell Science 114, 303-310 © The Company of Biologists Ltd
SUMMARY Myf5 is a member of the muscle regulatory factor family of transcription factors and plays an important role in the determination, development, and differentiation of skeletal muscle. However, factors that regulate the expression and activity of Myf5 itself are not well understood. Recently, a role for the calcium-dependent phosphatase calcineurin was suggested in three distinct pathways in skeletal muscle: differentiation, hypertrophy, and fiber-type determination. We propose that one downstream target of calcineurin and the calcineurin substrate NFAT in skeletal muscle is regulation of Myf5 gene expression. For these studies, we used myotube cultures that contain both multinucleated myotubes and quiescent, mononucleated cells termed ‘reserve’ cells, which share many characteristics with satellite cells. Treatment of such myotube cultures with the calcium ionophore ionomycin results in an ≈4-fold increase in Myf5 mRNA levels, but similar effects are not observed
in proliferating myoblast cultures indicating that Myf5 is regulated by different pathways in different cell populations. The increase in Myf5 mRNA levels in myotube cultures requires the activity of calcineurin and NFAT, and can be specifically enhanced by overexpressing the NFATc isoform. We used immunohistochemical analyses and fractionation of the cell populations to demonstrate that the calcium regulated expression of Myf5 occurs in the mononucleated reserve cells. We conclude that Myf5 gene expression is regulated by a calcineurin- and NFATdependent pathway in the reserve cell population of myotube cultures. These results may provide important insights into the molecular mechanisms responsible for satellite cell activation and/or the renewal of the satellite cell pool following activation and proliferation.
INTRODUCTION
and differentiation of skeletal muscle using transgenic and knockout mice (Arnold and Braun, 1996; Arnold and Winter, 1998). In the absence of MyoD and Myf5, myoblasts are not formed in the developing embryo (Rudnicki et al., 1993), while in the absence of myogenin, myoblasts are formed, but differentiation is inhibited (Hasty et al., 1993). MRF4 is believed to play a role in later stages of muscle maturation (Zhang et al., 1995). Clearly, the MRFs play a central role in regulating differentiation of skeletal muscle, one of the proposed calcineurin regulated pathways. In addition to differentiation, evidence also exists for the participation of the MRFs in the regulation of hypertrophy and fiber-type determination. In several models of hypertrophy including stretch and electrical stimulation, increases in the expression of MRFs occur within muscle fibers, suggesting a role in the growth of existing myofibers (Jacobs-El et al., 1995; Lowe and Alway, 1999). Evidence for a role of MRFs in fibertype determination is suggested by the fact that MRF expression levels differ between fast and slow muscles (Hughes et al., 1993; Kraus and Pette, 1997). Recently, a transgenic mouse was developed that overexpresses myogenin in postmitotic fast muscle fibers (Hughes et al., 1999). These mice displayed a shift towards metabolic enzymes characteristic of slow fibers. A role for MyoD in the maintenance of fast fibers has also been proposed (Hughes et al., 1997). As the MRFs play important roles in calcineurin regulated pathways, it is
Changes in intracellular calcium have been implicated in regulating diverse processes in skeletal muscle, including differentiation, hypertrophy, and fiber-type determination. In each of these cases, the calcium/calmodulin regulated protein phoshatase calcineurin has been proposed to play a central role in mediating the downstream calcium-dependent signaling (Abbott et al., 1998; Chin et al., 1998; Delling et al., 2000; Dunn et al., 1999; Friday et al., 2000; Musaro et al., 1999; Naya et al., 2000; Semsarian et al., 1999). However, the calcineurin targets that regulate these processes have not been clearly defined. One potential target of these signals is the muscle regulatory factor (MRF) family of bHLH transcription factors that includes MyoD, Myf5, myogenin, and MRF4 (Megeney and Rudnicki, 1995; Rudnicki and Jaenisch, 1995). The MRFs are themselves members of a larger superfamily of transcription factors that contain a basic region which mediates DNA binding to an E-Box consensus DNA element (Davis et al., 1990), and a helix-loop-helix domain that mediates dimerization. A large number of muscle-specific promoters have been found to contain E-Box elements. Each of the family members was initially described based on its ability to transform 10T1/2 fibroblasts to a muscle phenotype, and their roles have been more clearly defined in the determination
Key words: Calcineurin, NFAT, Gene regulation
304
JOURNAL OF CELL SCIENCE 114 (2)
possible that they are directly or indirectly regulated by a calcineurin-dependent pathway. We hypothesize that expression of the Myf5 gene is regulated by the calcineurin-dependent transcription factor nuclear factor of activated T-cells (NFAT). A calcium regulated pathway for controlling Myf5 gene expression has already been proposed. Treatment of skeletal muscle cells in vitro with the physiological peptide arginine-vasopressin induces an increase in cytosolic Ca2+ concentration (Teti et al., 1993), and also results in a strong upregulation of the Myf5 gene (Nervi et al., 1995). However, the downstream mediators of the arginine-vasopressin signal are not known. Regulation of the Myf5 gene is complex, requiring up to 500 kb of genetic sequence to faithfully reproduce the normal pattern of Myf5 expression (Zweigerdt et al., 1997). In addition, genetic elements in distinct locations regulate the expression of Myf5 in different locations during embryonic development (Patapoutian et al., 1993). Few specific DNA elements or transcription factors that bind within the Myf5 promoter have been identified. In one report, an Oct-like binding factor was found to bind to a conserved element within the avian Myf5 promoter, but this element may not be muscle specific, since it was required for expression of a reporter construct in myoblasts as well as fibroblasts (Barth et al., 1998). In the mouse myogenic cell line C2, expression of Myf5 in myoblasts was upregulated by increasing the activity of the glucocorticoid receptor and AP-1 (Aurade et al., 1997). Identification of a calcium-dependent signaling pathway that regulates the expression of Myf5 would help to elucidate the molecular regulation of the Myf5 gene, and may further define the mechanism(s) whereby calcineurin regulates the differentiation, hypertrophy, and fiber type determination of skeletal muscle. In this report, we show that Myf5 mRNA and protein levels are increased by the calcium ionophore ionomycin in the mononucleated cell fraction of myotube cultures. These mononucleated cells have been termed reserve cells, and were shown to share many characteristics with skeletal muscle satellite cells, including quiescence, self-renewal, and the ability to generate multinucleated myotubes (Yoshida et al., 1998). Ionomycin-induced effects on Myf5 gene expression are not found in proliferating myoblast cultures. The increase in Myf5 mRNA levels is dependent on the activity of calcineurin and NFAT, and can be specifically enhanced by overexpressing the NFATc isoform. We conclude that the expression of Myf5 in the reserve cells is regulated by calcineurin and NFAT. MATERIALS AND METHODS Antisera, reagents and statistics Rabbit polyclonal antibodies against Μyf5 were purchased from Santa Cruz Biotech., Inc. (Santa Cruz, CA) and obtained as a gift from Dr Didier Montarras. Secondary antibodies were purchased from Jackson Immunoresearch Laboratories (West Grove, PA). CSA was a gift of Sandoz (Basil, Switzerland). Ionomycin and PMA were purchased from Sigma Chemical Co. Thapsigargin was obtained from the Alexis Corp. (San Diego, CA). Amphotropic retroviral producer cells (SD3443) were obtained from the American Type Culture Collection (Rockville, MD). All cell culture reagents were purchased from Life Technologies (Grand Island, NY), except where noted. Fetal bovine
serum (FBS) was purchased from Atlanta Biologicals (Norcross, GA). Basic fibroblast growth factor (bFGF) was purchased from Promega (Madison, WI). All statistical analyses were performed using a one way analysis of variance and Bonferroni’s pairwise comparisons. Cell culture Primary myoblast cultures were prepared from SJL mice and purified to >99% as previously described (Abbott et al., 1998; Rando and Blau, 1994). Growth medium (GM) consisted of Ham’s F10, 20% FBS, 5 ng/ml bFGF, 200 U/ml penicillin G, and 200 µg/ml streptomycin for primary muscle cells, and a similar medium was used for C2C12 muscle cells with the exception of 15% calf serum and 5% FBS in the place of the 20% FBS. Differentiation was induced by changing cells to a low serum, low mitogen differentiation medium (DM: DME, 2% horse serum, 200 U/ml penicillin G, 200 µg/ml streptomycin) for 24-48 hours. Primary cells were differentiated on E-C-L (Upstate Biotechnology, Lake Placid, NY) coated dishes. Retroviral plasmids, production and infection The retroviral vectors utilized in these experiments have all been previously described, including CAIN and GFP-VIVIT expression vectors (Friday et al., 2000), an NFAT responsive reporter vector (Abbott et al., 1998; Boss et al., 1998), and an NFATc expression vector (Abbott et al., 2000). Production of infectious retrovirus and infection of primary myoblasts were performed as previously described (Abbott et al., 1998). Experiments using GFP-VIVIT were performed on cells that had been selected for GFP expression by flow cytometry (Friday et al., 2000). Northern blotting RNA was prepared from cells using Trizol Reagent (Life Technologies) following the manufacturer’s protocol. RNA was separated on 1% agarose-formaldehyde gels and transferred to Nytran SPC membranes (Schleicher & Schuell, Keene, NH). Membranes were probed with random-primed cDNA (Rediprime II, Amersham Pharmacia Biotech, Buckinghamshire, UK) labeled with 32P in Rapidhyb buffer (Amersham Pharmacia Biotech). After high stringency washing, membranes were visualized by autoradiography. Autoradiographs were scanned and quantitated using Scion Image software. Immunoblotting Cells were lysed with RIPA-2 (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% deoxycholic acid, 0.1% SDS) containing protease inhibitors (Mini Complete, Boehringer Mannheim, Indianapolis, IN). Equal amounts of protein (15 µg per lane) (Bradford, 1976) were separated by SDS-PAGE and transferred electrophoretically to a PVDF membrane (Immobilon P, Millipore, Burlington, MA). After non-specific binding was blocked in 5% nonfat milk in TBS for 30 minutes, the membrane was incubated overnight at 4°C in 0.5% non-fat milk in TBS containing a 1:800 dilution of anti-Myf5 (Santa Cruz Biotech., Inc). Blots were washed extensively in TBS containing 0.1% Tween-20 (TBS-T) and then incubated with a donkey anti-rabbit HRP conjugated secondary antibody (1:10,000) in 0.5% non-fat milk in TBS-T. Blots were washed in TBS-T and antibody binding was detected using ECL reagents (Amersham Pharmacia Biotech). To demonstrate relative protein loading membranes were stained with Coomassie Blue (BioRad, Hercules, CA). Immunohistochemistry Myotube cultures grown in 35 mm dishes were fixed in 3.7% formaldehyde for 10 minutes. After non-specific binding was blocked in TNB buffer (NEN Life Sciences, Boston, MA) for 30 minutes, the cells were incubated overnight at 4°C with a 1:1000 dilution of antiMyf5 (Lindon et al., 1998) in DMEM containing 10% FBS. Cells were washed with PBS containing 0.1% Tween-20 and then incubated
NFAT regulates Myf5 gene expression
305
with a biotinylated donkey anti-rabbit secondary antibody for 1 hour in DMEM containing 10% FBS. Antibody binding was detected using the Vectastain Elite Kit (Vector Laboratories, Burligame, CA). Cells were photographed and analyzed using an Axioplan microscope (Carl Zeiss, Thornwood, NY). Reporter assays Myoblasts containing an NFAT responsive reporter were plated at 4×104 cells per well of 24-well dishes. Cells were infected by two rounds of infection with either control or NFATc retroviruses. After 24 hours, the medium was replaced with DM and the cells were allowed to differentiate for 48 hours. The medium was replaced with the appropriate drug containing medium and incubated for 5-6 hours at 37°C. Cells were washed twice with PBS and 75 µl of Luciferase Cell Culture Lysis Reagent (Promega, Madison, WI) was added to each well. The cell lysates were collected and spun at 12,000 g for 30 seconds. 100 µl of Luciferase Assay Reagent (Promega) was injected into 20 µl of cell lysate and light output was measured after a 5 second delay over a 10 second window using a Turner TD-20e luminometer (Turner Designs, Sunnyvale, CA).
RESULTS Regulation of Myf5 mRNA levels by a calcineurindependent pathway in myotube cultures The possibility that Myf5 gene expression is regulated by calcium has been suggested by work demonstrating an increase in Myf5 mRNA levels after treatment with argininevasopressin, a hormone known to increase cytosolic calcium levels in skeletal muscle cells (Nervi et al., 1995; Teti et al., 1993). To determine if Myf5 mRNA levels could be directly regulated by increasing intracellular calcium, we treated myoblast or myotube cultures with the calcium ionophore ionomycin and determined the effect on the expression of Myf5 mRNA by northern blotting (Fig. 1). In myoblast cultures, ionomycin treatment has no effect on the expression of Myf5 mRNA, but in myotube cultures, ionomycin treatment results in an ≈4-fold increase in Myf5 mRNA levels. To determine if the response of the Myf5 gene to increasing intracellular calcium requires the activity of calcineurin, some cultures were also treated with the calcineurin inhibitor cyclosporine A (CSA). CSA has no effect on Myf5 expression in myoblasts. In myotube cultures, CSA pretreatment blocks the induction of Myf5 mRNA expression following ionomycin treatment. We also find a reduction in the basal expression levels of Myf5 mRNA by ≈30%, but the trend is not statistically significant. We have obtained similar data in human primary muscle cells, C2C12 mouse muscle cells, and primary muscle cells isolated from BALB/C mice (data not shown). The data using CSA suggest that calcineurin regulates Myf5 gene expression, but CSA may have targets distinct from calcineurin in muscle cells (Lo Russo et al., 1996; Lo Russo et al., 1997). To demonstrate conclusively that the increase in Myf5 mRNA levels observed with ionomycin treatment is calcineurin-dependent, we utilized the physiological calcineurin inhibitor CAIN (Lai et al., 1998). In previous work, we have shown that CAIN effectively inhibits calcineurin activity in skeletal muscle cells (Friday et al., 2000). Myoblasts were infected with either control or CAIN retroviruses and induced to differentiate (Fig. 2). Since calcineurin activity is required for the differentiation of skeletal muscle (Friday et al., 2000), we had to place the cells into differentiation medium
Fig. 1. Ionomycin treatment of myotube cultures results in a CSAsensitive increase in Myf5 mRNA levels. Myoblast (Mb) or myotube (Mt) cultures were treated with either vehicle (V), ionomycin (I, 1 µM), CSA (C, 1 µM), or ionomycin plus CSA for 5 hours in DM. RNA was isolated and analyzed by northern blotting for Myf5 mRNA expression. (a) A representative northern blot is shown along with a portion of an ethidium bromide stained gel to demonstrate relative RNA loading. (b) Autoradiographs were quantitated and the data were plotted as fold increase relative to vehicle treated cultures for each cell type. Each bar represents the mean ± s.e.m. of 3-5 independent experiments. *P
