Role of syndecan-4 side chains in turkey satellite cell growth and development

Authors


Author to whom all correspondence should be addressed.
Email:Velleman.1@osu.edu

Abstract

Syndecan-4 is a cell membrane heparan sulfate proteoglycan that is composed of a core protein and covalently attached glycosaminoglycans (GAG) and N-linked glycosylated (N-glycosylated) chains. Syndecan-4 has been shown to function independent of its GAG chains. Syndecan-4 may derive its biological function from the N-glycosylated chains due to the biological role of N-glycosylated chains in protein folding and cell membrane localization. The objective of the current study was to investigate the role of syndecan-4 N-glycosylated chains and the interaction between GAG and N-glycosylated chains in turkey myogenic satellite cell proliferation, differentiation, and fibroblast growth factor 2 (FGF2) responsiveness. The wild type turkey syndecan-4 and the syndecan-4 without GAG chains were cloned into the expression vector pCMS-EGFP and used as templates to generate syndecan-4 N-glycosylated one-chain and no-chain mutants with or without GAG chains. The wild type syndecan-4, all of the syndecan-4 N-glycosylated chain mutants were transfected into turkey myogenic satellite cells. Cell proliferation, differentiation, and responsiveness to FGF2 were measured. The overexpression of syndecan-4 N-glycosylated mutants with or without GAG chains did not change cell proliferation, differentiation, and responsiveness to FGF2 compared to the wild type syndecan-4 except that the overexpression of syndecan-4 N-glycosylated mutants without GAG chains increased cell proliferation at 48 and 72 h post-transfection. These data suggest that syndecan-4 functions in an FGF2-independent manner, and the N-glycosylated and GAG chains are required for syndecan-4 to regulate turkey myogenic satellite cell proliferation, but not differentiation.

Introduction

Satellite cells are undifferentiated mononuclear myogenic cells that are located between the sarcolemma and basement membrane (Mauro 1961). They play a critical role in posthatch or postnatal muscle growth by proliferating, differentiating, and donating their nuclei to adjacent muscle fibers (Moss & Leblond 1971). In adult skeletal muscle, satellite cells are normally quiescent and can be activated in response to stimuli to proliferate and fuse to form new fibers or repair damaged fibers (Schultz & Jaryszak 1985). For example, the activities of satellite cells can be regulated by growth factors like fibroblast growth factor 2 (FGF2; Dollenmeier et al. 1981), transforming growth factor β (Florini et al. 1986), and insulin-like growth factor (Schmid et al. 1983), and changes in their extracellular environment (Velleman 2004).

The extracellular matrix (ECM) is a dynamic structure that is secreted by cells and regulates cell activity by transducing signals into the cell. The ECM is composed of fibrous and nonfibrous proteins and polysaccharides. The heparan sulfate proteoglycans (HSPG) are an important group of macromolecules in the ECM that have been shown to be important regulators of muscle growth and development (Brandan & Larraín 1998; Velleman et al. 2007). In addition, they are involved in cell-substratum adhesion (Lebaron et al. 1988; Haugen et al. 1992; Sanderson et al. 1992), cell–cell adhesion (Cole et al. 1986; Reyes et al. 1990; Stanley et al. 1995), and signaling mediated by growth factors, which are properties critical in the formation of tissues.

Syndecan-4 is a HSPG that plays an essential role in muscle maintenance and regeneration (Cornelison et al. 2001, 2004; Tanaka et al. 2009). Syndecan-4 has been reported to be a marker for regenerating satellite cells (Tanaka et al. 2009). This is supported by the observation that syndecan-4 knockout mice lose their ability to regenerate damaged muscle (Cornelison et al. 2004).

Syndecan-4 is composed of a transmembrane core protein and covalently attached glycosaminoglycans (GAG) chains and N-linked glycosylated (N-glycosylated) chains. It has been reported that HSPG GAG chains are required for stable binding of FGF2 to its high affinity receptor (Yayon et al. 1991; Aviezer et al. 1994). The GAG chains are attached to the core protein at Ser residues containing Ser-Gly repeats. Turkey syndecan-4 GAG chains are attached to the core protein at Ser38, Ser65, and Ser67, but they are not required for syndecan-4 regulated turkey myogenic satellite cell proliferation, differentiation, and responsiveness to FGF2 during proliferation (Velleman et al. 2007; Zhang et al. 2008).

The N-glycosylated chains are attached to the core protein at the Asn residue at the sequence of Asn-Xaa-Ser/Thr. The Xaa can be any amino acids except proline (Kornfeld & Kornfeld 1985). N-linked glycosylated chains have been reported to function in many biological processes. For example, they are involved in the proper folding of proteins (Parodi 2000; Helenius & Aebi 2001), and localization of membrane proteins to the cell surface (Martinez-Maza et al. 2001; Yan et al. 2002). However, more research efforts have been placed on proteoglycan GAG chains, while the role of N-glycosylated chains has received limited attention. Decorin and glypican-1 are the only two proteoglycans for which studies on the function of their N-glycosylated chains have been reported (Seo et al. 2005; Song et al. 2010). Seo et al. (2005) found that the normal secretion of decorin into the ECM requires at least one GAG chain or N-glycosylated chain attached to the core protein. Glypican-1 N-glycosylated chains are critical in turkey myogenic satellite cell proliferation, differentiation, and responsiveness to FGF2 (Song et al. 2010). Syndecan-4 has two N-glycosylated chains attached to the core protein at Asn124 and Asn136. The deletion of syndecan-4 N-glycosylated chains may lead to the improper folding of the syndecan-4 protein and alter its cell membrane spatial distribution. To investigate the function of each syndecan-4 N-glycosylated chain, site-directed mutagenesis was used to change the Asn that is used in the attachment of N-glycosylated chains to Ala.

Syndecan-4 N-glycosylated one-chain and no-chain mutants were transfected into turkey myogenic satellite cells, following which the proliferation, differentiation, and responsiveness to FGF2 were measured. Based on N-glycosylated chains affecting protein secondary structure, the deletion of the syndecan-4 N-glycosylated chains may lead to the improper folding of the syndecan-4 core protein and change the interaction of syndecan-4 GAG chains with other molecules. Thus, the syndecan-4 N-glycosylated one-chain and no-chain mutants without GAG chains were also generated and transfected into turkey myogenic satellite cells to study the interaction between syndecan-4 N-glycosylated chains and GAG chains on cell proliferation, differentiation, and FGF2 responsiveness.

The results from the current study provide new information about the function of syndecan-4 N-glycosylated chains and the interaction between syndecan-4 GAG chains and N-glycosylated chains during myogenesis.

Materials and methods

Clone generation

A full length wild type turkey syndecan-4 cDNA (GenBank accession number AY852251; Velleman et al. 2006) and syndecan-4 without GAG chains have been previously cloned into the mammalian expression vector pCMS-EGFP (BD Biosciences Clontech) (Velleman et al. 2006). These two clones were used as templates to generate syndecan-4 N-glycosylated one-chain and no-chain mutants with or without GAG chains using the Quick Change Multi Site-Directed Mutagenesis kit (Stratagene Corporation). Figure 1 illustrates the site-directed mutagenesis strategy used to generate syndecan-4 all possible N-glycosylated one-chain and no-chain mutations. The primers used to change N1 and N2 chains were: N1, 5′-AGC TTC ACC TGT TGA AGA GGC CCT GTC CAA CAA GAT CTC CAT GGC-3′ and N2, 5′-TCC ATG GCA AGC ACA GCC GCC AGC AGC ATC TTT GAA AG-3′. All of the mutations were confirmed by DNA sequencing.

Figure 1.

 Site-directed mutagenesis strategy used to generate syndecan-4 N-linked glycosylated (N-glycosylated) chain one-chain and no-chain mutants with or without glycosaminoglycan (GAG) chains. Syndecan-4 has three GAG chains attached to the core protein at Ser38, Ser65, and Ser67. The two N-glycosylated chains are attached to the core protein at Asn124 and Asn139. The Asn124 is referred to as N1 chain, and Asn139 is N2 chain. To distinguish different forms of syndecan-4, the following nomenclature was developed: S4 is syndecan-4; the number after N means the potential N-glycosylated chain attachment sites unaltered, for example: N1 means the syndecan-4 has the N1 chain site unaltered. S0 and N0 indicate all of the GAG chain attachment sites or all of the N chain attachment sites were altered.

Satellite cell culture and transfection

Satellite cells were isolated from the pectoralis major muscle of 7-week-old male randombred control 2 (RBC2) line turkeys (Velleman et al. 2000). The turkey RBC2 line is representative of a 1967 turkey and maintained at the Ohio Agricultural Research and Development Center without selection for any growth traits (Nestor 1977).

Satellite cells were plated and transfected with syndecan-4 N-glycosylated one-chain and no-chain mutants with or without GAG chains, wild type syndecan-4, or pCMS-EGFP empty vector as described by Song et al. (2010). At 0, 24, 48, and 72 h post-transfection and at 0, 24, 48, 72, and 96 h of differentiation, cell cultures were removed and stored at −70°C until analysis.

Real-time quantitative PCR

Total RNA was extracted from the cell cultures at 48 h post-transfection using TRIzol (Invitrogen) according to the manufacturer’s protocol (Song et al. 2010). Real-time quantitative polymerase chain reaction (PCR) was performed with a DNA Engine Opticon 2 real-time system (MJ Research) using the DyNAmo Hot Start SYBR Green qPCR kit (Finnzymes). The PCR reaction consisted of 2 μL cDNA, 10 μL 2× master mix, 250 nmol/L of each of the forward and reverse primers, and nuclease-free water up to 20 μL. The primers for syndecan-4 were: forward primer 5′-CTACCCTGGCTCTGGAGACCT-3′ and reverse primer 5′-TCATTGTCCAGCATGGTGTTT-3′. The primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were: forward primer 5′-GAGGGTAGTGAAGGCTGCTG-3′ and reverse primer 5′-CCACAACACGGTTGCTGTAT-3′. For both of the genes, the cycling parameters were denaturation at 95°C for 15 min, followed by 34 cycles of 94°C for 30 s, 58°C for 30 s and at 72°C for 30 s with a final elongation of 72°C for 5 min. The melting curve program was 52–95°C, 0.2°C per read, and a 1 s hold. The PCR products were analyzed on a 1% agarose gel to check the amplification specificity. Standard curves were constructed with serial dilutions of purified PCR products from each gene. The PCR products were purified by agarose gel electrophoresis using the QIAquick Gel Extraction Kit (Qiagen). The amount of sample cDNA for each gene was determined by comparing the results to the syndecan-4 standard curve, and then normalized to GAPDH expression. The PCR products were verified by DNA sequencing for specificity.

Immunohistochemistry

Satellite cells were cultured and transfected in 24 well cell culture plates. After 24 h of transfection, the cell cultures were fixed with 3% paraformaldehyde for 30 min at room temperature. The cells were then incubated with goat anti-syndecan-4 antibody (Santa Cruz, Sc-33913) 1:100 dilution in blotto (5% nonfat milk dissolved in 20 mmol/L Tris-HCl, pH 7.4, and 150 mmol/L NaCl) for 1 h. The secondary antibody used was donkey anti-goat antibody conjugated with rhodamine (Santa Cruz, SC-2904) 1:200 dilution in 5% nonfat milk dissolved in blotto. An Olympus XI 70 microscope and an Optronics digital camera (Optronics) were used to view and record the images.

Proliferation assay

The proliferation assay was performed as described by Velleman et al. (2006). Cell proliferation was measured by the DNA content in each well by the method of McFarland et al. (1995). The Hoechst 33258 fluorochrome (Sigma-Aldrich) was used to determine DNA concentration on a Fluoroskan Ascent FL plate reader (ThermoElectron Co.) using double-stranded calf thymus DNA as the standard.

Cell count

Satellite cells were plated and transfected as described in the proliferation assay. At 48 h post-transfection, plates were removed from the incubator and rinsed three times with phosphate-buffered saline (PBS). Cells were treated with 0.1% trypsin in PBS, scraped off the well and then transferred into 1.5 mL centrifuges. Cell pellets were washed with PBS, and then resuspended in 100 μL of PBS. Ten microliters of suspension was stained with equal amounts of trypan blue. The cell count measurement was repeated three times for each sample using a Countess Automated Cell Counter (Invitrogen).

Differentiation assay

Satellite cells were cultured and transfected for differentiation as described in Zhang et al. (2008). Differentiation was determined by measuring the muscle specific creatine kinase level by a modified method from Yun et al. (1997). In brief, all plates were brought to room temperature for 10 min. Creatine phosphokinase (Sigma-Aldrich) was used to generate a standard curve.

Two hundred microliters of creatine kinase assay buffer consisting of 20 mmol/L glucose (Fisher Scientific), 10 mmol/L Mg acetate (Fisher Scientific), 1.0 mmol/L adenosine diphosphate (Sigma-Aldrich), 10 mmol/L adenosine monophosphate (Sigma-Aldrich), 20 mmol/L phosphocreatine (Calbiochem), 0.5 U/mL of hexokinase (Worthington Biochemical), 1 U/mL of glucose-6-PO4 dehydrogenase (Worthington Biochemical), 0.4 mmol/L thio-nicotinamide adenine dinucleotide (Sigma-Aldrich), and 1 mg/mL of bovine serum albumin prepared in 0.1 mol/L glycylglycine (Sigma-Aldrich, pH 7.5) was added to each well. After a 20 min incubation at room temperature, cell differentiation was determined by measuring the rate of thio-nicotinamide adenine dinucleotide reduction with a Dynex MRX Revelation Microtiter Plate Reader (Dynex Technologies Inc.) at 405 nm at 5 min intervals.

Nuclei number counting

Satellite cells were plated in gelatin-coated 24 well cell culture plates at a density of 12 500 cells per well, and were transfected and cultured as described above. At 72 h of fusion, the cells were fixed with 3% paraformaldehyde for 30 min at room temperature. The dye, 4′, 6-diamidino-2-phenylindole (Invitrogen) at a 1:10 000 dilution in PBS was then added to the cell cultures to stain the cell nuclei. Images were viewed with an Olympus XI 70 microscope and recorded with an Optronics digital camera. For each of the syndecan-4 mutants, wild type syndecan-4, and the empty pCMS-EGFP vector, the number of nuclei in 25 individual myotubes was counted and the average number of nuclei per myotube was calculated.

Responsiveness to fibroblast growth factor 2 assay

Cellular responsiveness to FGF2 during proliferation was determined by measuring the DNA concentration in each well (Velleman et al. 2006). Satellite cells were plated in gelatin-coated 24 well cell culture plates at a density of 12 500 cells per well, and transfected as described above. After 6 h of transfection, the Dulbecco’s Modified Eagle Medium (DMEM) was removed and changed to serum-free defined media (McFarland et al. 2006) containing 0, 2.5, or 10.0 ng/mL of FGF2 (Pepro Tech). Cell responsiveness was measured at 72 h post-transfection by the DNA content as described above in the proliferation assay.

Cellular responsiveness to FGF2 was also measured at 0, 48, and 72 h of differentiation. Cells were plated and cultured as described in the differentiation assay except at the beginning of differentiation FGF2 was added at 0, 2.5 or 10 ng/mL. Cell responsiveness to FGF2 was determined by measuring the creatine kinase level as described above in the differentiation assay.

Syndecan-4 fusion protein analysis

Full length syndecan-4 was amplified from wild type syndecan-4 and syndecan-4 without N-glycosylated chains in pCMS-EGFP vector. The following primers were used for PCR reaction. The forward primer was 5′-ATA ATG CCG CTG CTC CGC G-3′ (Operon), and the reverse primer was 5′-AGC GTA GAA CTC ATT TGT AGG GGC-3′ (Operon). The PCR condition was denaturation at 95°C for 15 min, and 35 cycles of 30 s at 94°C, 30 s at 60°C, 30 s at 72°C followed by 15 min at 72°C. The ligation reaction contained 3 μL of fresh PCR product from the above reaction, 1 μL of salt solution (1.2 mol/L NaCl and 0.06 mol/L MgCl2), and 1 μL of pcDNA3.1/V5-His-TOPO Vector (Invitrogen). After being incubated for 5 min at room temperature, 2 μL of the reaction was used to transform TOP10 Chemically Competent Escherichia coli (Invitrogen) according to the manufacturer’s instruction. Luria-Bertani-ampicillin agar plates were used to select positive clones. The positive clones were grown overnight at 37°C in LB broth containing 100 μg/mL ampicillin. The plasmids were isolated using QIAquick Miniprep kit (Qiagen) and confirmed by DNA sequencing.

Wild type syndecan-4 and syndecan-4 without N-glycosylated chains in pcDNA3.1/V5-His-TOPO Vector were then transfected into turkey satellite cells as described above. Control cells were treated through the transfection process but without any plasmid. At 72 h post-transfection, proteins were extracted from cells using cell lysis buffer A containing 50 mmol/L Tris–HCl, pH 7.4, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 150 mmol/L NaCl, 1 mmol/L ethylenediaminetetraacetic acid (EDTA), and 1 mmol/L Na3VO4, and freshly prepared protease inhibitor cocktail (Roche Applied Science). Protein concentration in the supernatant was determined by the Bradford assay (Bradford 1976). Equal amounts of protein were mixed with reducing sample buffer (0.125 mol/L Tris–HCl, pH6.8, 4.1% SDS, 20% glycerol, 2%ß-mercaptoethanol, and 0.001% bromphenol blue), boiled for 5 min and then separated on an 8% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). Proteins were transferred to a polyvinylidene difluoride membrane (PVDF; Millipore). After being blocked for 2 h with 5% nonfat milk dissolved in tris-buffered saline containing 20 mmol/L Tris–HCl, pH7.4, 150 mmol/L NaCl, and 0.05% Tween 20 (TBS-T), the PVDF membrane was incubated with the indicated antibodies in blocking buffer for 2 h at room temperature. Antibodies used for immunoblotting were mouse anti-V5 conjugated with alkaline phosphatase (1:5000 dilution; Invitrogen) and mouse monoclonal β-actin (1:10 000 dilution; Sigma). Secondary antibody used for β-actin was goat anti-mouse IgG alkaline phosphatase conjugated (1:10 000 dilution; Santa Cruz). After incubating with chemiluminescent alkaline phosphatase substrate (Millipore) for 5 min at room temperature, immunoreactive bands were visualized with a Bio-Rad ChemiDoc XRS imaging system (Bio-Rad). The size of the bands was determined with Quantity One 4.6.6 Software (Bio-Rad).

Western analysis

The pCMS-EGFP empty vector, wild type syndecan-4 and all of the syndecan-4 N-glycosylated chain mutants were transfected into turkey satellite cells. At 48 h post-transfection, proteins were extracted from cells using cell lysis buffer A. Protein concentration in the supernatant was determined by the Bradford assay. Equal amounts of protein were loaded on an 8% SDS–PAGE. Proteins were transferred to a PVDF membrane and screened with goat anti-human syndecan-4 antibody (1:2000 dilution; Santa Cruz) and mouse monoclonal β-actin antibody (1:10 000 dilution; Sigma). The size and quantity of the bands was determined with Quantity One 4.6.6 Software (Bio-Rad).

Statistical analysis

Cell proliferation, differentiation, and FGF2 responsiveness assays were independently repeated at least three times. Within each experiment, there were four replicates of each treatment for proliferation and FGF2 responsiveness measurements during proliferation, and six replicates for differentiation and FGF2 responsiveness during differentiation. For the real time PCR, two replicates for each treatment in each assay were performed. The data from each repeat were used to calculate a mean and the standard error of the mean (SEM). Data are graphed as the mean ± SEM. The SAS PROC GLM (SAS Institute Inc.) was used for statistical analyses. Differences among means were detected using the Fisher’s least significance method. The cell proliferation assays, differentiation assays, and real-time quantitative PCR data were analyzed using SAS Proc GLM procedures. For the FGF2 responsiveness assays, the statistical model included the effect of FGF2 treatment, mutant transfection, and their interaction. The main factors of FGF2 treatment and variant mutant transfection and the interaction between two-factors were analyzed using SAS Proc GLM procedures. Differences among means in each experiment were evaluated using an anova and detected using Fisher’s least-significant-difference. Two-sided P-values of < 0.05 were considered statistically significantly.

Results

Site-directed mutagenesis and the overexpression in satellite cells

Six syndecan-4 N-glycosylated chain mutants were generated by site-directed mutagenesis. The N-glycosylated chains attached to Asn124 and Asn136 are defined as N1 and N2 chain, respectively. The syndecan-4 mutants with N1 or N2 chains without changing the GAG chain attachment site are referred to as S4-N1 and S4-N2, respectively. The syndecan-4 mutants with the N1 or N2 chains with the GAG chain attachment sites eliminated are termed as S4-S0-N1 and S4-S0-N2, respectively. The syndecan-4 mutants without N-glycosylated chains attached to the core protein with or without GAG chains are defined as S4-N0 and S4-S0-N0, respectively. Immunoblotting of the fusion protein confirmed the deletion of syndecan-4 N-glycosylated chains (Fig. 2). The wild type syndecan-4 had a molecular mass of 32 kDa, whereas S4-N0 had a molecular mass of 31 kDa.

Figure 2.

 Syndecan-4 fusion protein analysis. Proteins were extracted from randombred control 2 line male turkey skeletal muscle satellite cells that were transfected with wild type syndecan-4 (S4), syndecan-4 without N-glycosylated chains (S4-N0) in pcDNA3.1/V5-His-TOPO vector or without any plasmid (NC). The anti-V5 antibody was used to detect the V5 tag expressed from the pcDNA3.1/V5-His-TOPO vector. The lanes are NC, S4, and S4-N0. Molecular mass is highlighted with an arrow. β-actin was used to confirm protein migration and had a molecular mass of 42 kDa.

All of the syndecan-4 N-glycosylated one chain and no-chain mutants with or without GAG chains and wild type syndecan-4 had significantly higher expression of syndecan-4 compared to the empty pCMS-EGFP vector (Fig. 3). The overexpression of syndecan-4 was also confirmed at the protein level by immunohistochemistry at 24 h post-transfection (Fig. 4). Figure 4a–c illustrates turkey satellite cells transfected with the pCMS-EGFP empty vector. Figure 4d–f illustrates turkey satellite cells transfected with wild type syndecan-4. Figure 4a,d shows brightfield images to indicate the cell morphology. Figure 4b,e shows the enhanced green fluorescence protein marker expressed from the pCMS-EGFP vector indicating the transfection of the cells. Figure 4c,f shows the cells stained with syndecan-4 antibody. The cells transfected with syndecan-4 and all of the syndecan-4 mutants had higher protein levels of syndecan-4 compared to the cells transfected with pCMS-EGFP empty vector as assayed by immunohistochemistry and western blot (Fig. 5).

Figure 3.

 Real-time quantitative polymerase chain reaction (PCR) analysis of mRNA expression at 48 h post-transfection. Randombred control 2 line male turkey myogenic satellite cells were transfected with the empty pCMS-EGFP vector (control), wild type syndecan-4 (S4), syndecan-4 N-glycosylated chain no-chain mutant (S4-N0), syndecan-4 N-glycosylated chain one-chain mutants (S4-N1 and S4-N2), syndecan-4 glycosaminoglycan (GAG) no-chain mutant (S4-S0), syndecan-4 N-glycosylated chain no-chain mutant without GAG chains (S4-S0-N0), and syndecan-4 N-glycosylated chain one-chain mutants without GAG chains (S4-S0-N1 and S4-S0-N2). The error bars represent standard error of the mean. *Indicates a significant difference from the control (< 0.05).

Figure 4.

 Syndecan-4 expression in randombred control 2 line male turkey myogenic satellite cells transfected with the pCMS-EGFP empty vector, wild type syndecan-4, and syndecan-4 mutants at 24 h post-transfection. Panel (a) cells transfected with pCMS-EGFP. Panel (b) cells transfected with syndecan-4. Panel (c) cells transfected with syndecan-4 N-glycosylated chain no-chain mutant. Panels (d) and (e) cells transfected with syndecan-4 N-glycosylated chain one-chain mutants attached to core protein at the Asn124 and Asn139, respectively. Panel (f) cells transfected with syndecan-4 glycosaminoglycan (GAG) no-chain mutant. Panel (g) cells transfected with syndecan-4 N-glycosylated chain no-chain mutant without GAG chains, and panels (h) and (i) cells transfected with syndecan-4 N-glycosylated chain one-chain mutants attached to the core protein at the Asn124 and Asn139 without GAG chains, respectively. Black and white images were bright field (BF) views of the cell (left column). Green fluorescence indicates the cells expressing enhanced green fluorescent protein (EGFP) from the pCMS-EGFP vector (middle column). The red fluorescent images were stained with goat anti-human syndecan-4 antibody and a donkey anti-goat IgG rhodamine conjugated secondary antibody (right column).

Figure 5.

 Western analysis of syndecan-4 expression in randombred control 2 line male turkey myogenic satellite cells transfected with pCMS-EGFP empty vector (control), wild-type syndecan-4 (S4), syndecan-4 mutants (syndecan-4 N-glycosylated chain no-chain mutant [S4-N0], syndecan-4 N-glycosylated chain one-chain mutants attached to core protein at the Asn124 [S4-N1] and Asn139 [S4-N2], respectively, syndecan-4 glycosaminoglycan (GAG) no-chain mutant [S4-S0], syndecan-4 N-glycosylated chain no-chain mutant without GAG chains [S4-S0-N0], syndecan-4 N-glycosylated chain one-chain mutants attached to the core protein at the Asn124 and Asn139 without GAG chains [S4-S0-N1 and S4-S0-N2], respectively) at 48 h post-transfection. Each lane contains 50 μg of protein and was hybridized to anti-human syndecan-4 and anti-mouse β-actin antibodies. A ratio was calculated with the average density of syndecan-4 and its mutants relative to the control after normalization with β-actin.

The effect of syndecan-4 N-glycosylated chains with glycosaminoglycan chains attached to the core protein on turkey satellite cell proliferation, responsiveness to fibroblast growth factor 2, and differentiation

Cells transfected with wild type syndecan-4 and all of the syndecan-4 N-glycosylated chain mutants with GAG chains had decreased proliferation at 72 h post-transfection compared to the cells transfected with the control pCMS-EGFP empty vector. There was no difference in proliferation between the cells transfected with syndecan-4 N-glycosylated one-chain and no-chain mutants with the GAG chains attached compared to the wild type syndecan-4 (Fig. 6a). As shown in Figure 6b, cell replication was not affected by the transfection at 48 h post-transfection. Cell number was not significantly different among wild type syndecan-4, syndecan-4 N-glycosylated chain mutants with GAG chains, and pCMS-EGFP empty vector transfection, ranging from 80.0 to 92.3 cells/μL. At 72 h post-transfection, with 0, 2.5, or 10 ng/mL of FGF2, the cells transfected with S4-N2 had a significantly higher proliferation rate compared to the cells transfected with wild type syndecan-4, and other N-glycosylated chain mutants with GAG chains except for S4-N1 at 2.5 ng/mL of FGF2 (Fig. 6c). Without FGF2 treatment, compared to the wild type syndecan-4 transfection, the S4-N2 transfected cells had significantly higher proliferation; with 2.5 and 10 ng/mL FGF2 the S4-N2 group also had significantly higher proliferation compared to the wild type syndecan-4 transfection. Statistical analysis was used to analyze the effect of FGF2 treatment, the effect of mutant transfection, and their interaction. Based on this analysis, the higher proliferation of the cells transfected with S4-N2 was due to a gene effect not the FGF2 treatment. These data suggested that syndecan-4 N-glycosylated chains have no effect on cell responsiveness to FGF2 during proliferation.

Figure 6.

 Proliferation analysis and fibroblast growth factor 2 (FGF2) responsiveness of the randombred control 2 line male turkey myogenic satellite cells transfected with pCMS-EGFP empty vector (pCMS) (inline image), wild type syndecan-4 (S4) (inline image), syndecan-4 N-glycosylated chain no-chain mutant (S4-N0) (inline image), and syndecan-4 N-glycosylated chain one-chain mutants (S4-N1 [inline image] and S4-N2 [inline image]). (a) Proliferation analysis; (b) cell concentration at 48 h post-transfection; and (c) cellular response to FGF2 at concentrations of 0, 2.5, and 10 ng/mL at 72 h post-transfection. The error bars represent standard error of the mean. Bars without a common letter within times in figure (a) and within the same FGF2 concentration in figure (c) were significantly different (< 0.05).

At 48 and 72 h of differentiation, cells transfected with wild type syndecan-4 and syndecan-4 N-glycosylated mutants had decreased cell differentiation compared to those transfected with pCMS-EGFP empty vector, except for S4-N1 at 72 h of differentiation (Fig. 7a). There was no significant difference in differentiation from 0 to 96 h between the cells transfected with syndecan-4 N-glycosylated chain mutants containing GAG chains and the wild type syndecan-4. The number of nuclei in the myotubes at 72 h of differentiation was similar in trend to the differentiation assay results (Fig. 7b). At 72 h of differentiation, the average number of nuclei per myotube in the empty vector transfected cells was 27 compared to those transfected with wild type syndecan-4, S4-N0, S4-N1 and S4-N2 with 19.7, 22.9, 23.4, and 21.4, respectively.

Figure 7.

 Differentiation of the randombred control 2 line male turkey myogenic satellite cells after transfection with pCMS-EGFP empty vector (pCMS) (inline image), wild type syndecan-4 (S4) (inline image), syndecan-4 N-glycosylated chain no-chain mutant (S4-N0) (inline image), and syndecan-4 N-glycosylated chain one-chain mutants (S4-N1 [inline image] and S4-N2 [inline image]). (a) Differentiation analysis; and (b) average number of nuclei counted in 25 individual myofibers at 72 h differentiation. The error bars represent the standard error of the mean. Bars without common letters were significantly different (< 0.05).

The effect of syndecan-4 N-glycosylated chain mutants with the deletion of glycosaminoglycan chains on turkey satellite cell proliferation, responsiveness to fibroblast growth factor 2, and differentiation

At 48 h post-transfection, satellite cells transfected with syndecan-4 N-glycosylated one-chain and no-chain mutants without the GAG chains had a higher level of proliferation compared to the cells transfected with wild type syndecan-4 (Fig. 8a). At 72 h post-transfection, all the syndecan-4 N-glycosylated one-chain and no-chain mutants without the GAG chains also had a higher proliferation compared to the wild type syndecan-4. Transfection with S4-S0 did not influence cell proliferation compared to the cells transfected with wild type syndecan-4 at all sampling times. Cell number at 48 h post-transfection was not significantly affected by the overexpression of the syndecan-4 constructs (Fig. 8b). At 0, 2.5, or 10 ng/mL of FGF2, cells transfected with syndecan-4 N-glycosylated one-chain and no-chain mutants without the GAG chains had no significant difference in proliferation (< 0.05) compared to those transfected with wild type syndecan-4 (Fig. 8c).

Figure 8.

 Proliferation and fibroblast growth factor 2 (FGF2) responsiveness of the randombred control 2 line male turkey myogenic satellite cells transfected with pCMS-EGFP empty vector (pCMS) (inline image), wild type syndecan-4 (S4) (inline image), syndecan-4 without glycosaminoglycan (GAG) chains (S4-S0) (inline image), syndecan-4 N-glycosylated chain no-chain mutant without GAG chains (S4-S0-N0) (inline image), and syndecan-4 N-glycosylated chain one-chain mutants without GAG chains (S4- S0-N1 [inline image] and S4- S0-N2 [inline image]). (a) Proliferation analysis; (b) cell concentration at 48 h post-transfection; and (c) cellular response to exogenous FGF2 at 0, 2.5, and 10 ng/mL at 72 h post-transfection. The error bars represent standard error of the mean. Bars without a common letter within a time in figure (a) and within the same FGF2 concentration in figure (c) were significantly different (< 0.05).

At 0, 24, 48, 72, and 96 h of differentiation, cells transfected with syndecan-4 N-glycosylated one-chain and no-chain mutants without GAG chains showed no difference in differentiation compared to the cells transfected with wild type syndecan-4 (Fig. 9a). The average number of nuclei per myotube at 72 h of differentiation was similar to the differentiation assay data except the differentiation of S4-S0-N0 and S4-S0-N1 was significantly higher than the wild type syndecan-4 (Fig. 9b). The average number of nuclei per myotube in cells transfected with pCMS-EGFP was 27.0, wild type syndecan-4 was 19.7, S4-S0 was 22.2, S4-S0-N0 was 24.7, S4-S0-N1 was 23.3, and S4-S0-N2 was 23.1.

Figure 9.

 Differentiation of the randombred control 2 line male turkey myogenic satellite cells transfected with pCMS-EGFP empty vector (pCMS) (inline image), wild type syndecan-4 (S4) (inline image), syndecan-4 without GAG chains (S4-S0) (inline image), syndecan-4 N-glycosylated chain no-chain mutant without GAG chains (S4-S0-N0) (inline image), and syndecan-4 N-glycosylated chain one-chain mutants without GAG chains (S4-S0-N1 [inline image] and S4-S0-N2 [inline image]). (a) Differentiation analysis; and (b) average number of nuclei at 72 h of differentiation. The error bars represent standard error of the mean. Bars without common letters indicate a significant difference (< 0.05).

The effect of syndecan-4 glycosaminoglycan and N-glycosylated chains on turkey satellite cell responsiveness to fibroblast growth factor 2 during differentiation

Cell responsiveness to FGF2 was measured at 0, 48 and 72 h of differentiation (Fig. 10, respectively). Cells treated with FGF2 showed decreased differentiation after 48 h in differentiation medium, but was not significantly decreased at 72 h of differentiation compared to the cells without FGF2 treatment. At 0 and 48 h of differentiation, the differentiation of the cells transfected with wild type syndecan-4, S4-S0, S4-N0, and S4-S0-N0 was not significantly different compared to those transfected with the pCMS-EGFP empty vector within the same treatment of FGF2. Without FGF2 treatment, compared to the wild type syndecan-4 and pCMS-EGFP empty vector transfection, the cells transfected with S4-S0-N0 had significantly higher differentiation; with 2.5 and 10 ng/mL FGF2 treatment cells transfected with S4-S0-N0 also having significantly higher differentiation compared to the wild type syndecan-4 and pCMS-EGFP empty vector transfections. Statistical analysis was used to analyze the effect of FGF2 treatment, effect of mutant transfection, and their interaction. The analysis showed that the higher differentiation of the cells transfected with S4-S0-N0 was due to a gene effect, not the FGF2 treatment. These data indicated that syndecan-4 N-glycosylated chains and GAG chains are not required for cell responsiveness to FGF2 during differentiation.

Figure 10.

 Fibroblast growth factor 2 (FGF2) responsiveness of the randombred control 2 line male turkey myogenic satellite cells transfected with pCMS-EGFP empty vector (pCMS) (inline image), wild type syndecan-4 (S4) (inline image), syndecan-4 N-glycosylated chain no-chain mutant with or without GAG chains (S4-N0 [inline image] and S4-S0-N0 [inline image]), and syndecan-4 without GAG chains (S4-S0) (inline image). (a) Creatine kinase level at 0 h differentiation without FGF2 treatment (b) cellular response to FGF2 at 0, 2.5, and 10 ng/mL at 48 h of differentiation; and (c) cellular response to exogenous FGF2 at 0, 2.5, and 10 ng/mL at 72 h of differentiation. The error bars represent standard error of the mean. Bars without common letters within a time or FGF2 concentration were significantly different (< 0.05).

Discussion

Syndecan-4 is a membrane associated HSPG that plays an important role in satellite cell maintenance, activation, proliferation, and differentiation (Cornelison et al. 2001, 2004; Tanaka et al. 2009). However, the mechanism of how syndecan-4 functions in these processes is still not well understood. Zhang et al. (2008) reported that syndecan-4 GAG chains are not required for syndecan-4 to affect turkey satellite cell proliferation, differentiation, and FGF2 responsiveness during proliferation. In the present study, focus was placed on the function of syndecan-4 N-glycosylated chains and the interaction between syndecan-4 N-glycosylated chains and GAG chains on turkey satellite cell proliferation, differentiation, and responsiveness to FGF2.

Syndecan-4 GAG chains and N-glycosylated chains may affect cell proliferation through regulating focal adhesion formation. Focal adhesions form where the cells attach to the substrate. Focal adhesions are composed of a large number of proteins including but not limited to integrins (Hynes 1992; Schwarta et al. 1992), cadherins (Takeichi 1988, 1990, 1991), and selectins (Bevilacqua & Nelson 1993). They act not only as anchorage points for the cells, but also function in signal transduction pathways. Integrins are involved in cell proliferation, adhesion, and apoptosis in muscle cells (Liu et al. 2008). Saoncella et al. (1999) reported that syndecan-4 is required for integrin mediated fibronectin-induced focal adhesion formation. Syndecan-4 is an important regulator of the composition of focal adhesions (Woods & Couchman 1994). The overexpression of syndecan-4 in Chinese hamster ovary K1 cells resulted in increased focal adhesion formation and decreased cell migration (Longley et al. 1999). It is possible that when syndecan-4 is overexpressed, more focal adhesions form and cell migration is decreased. Because migration is important for satellite cells to interact with each other to receive signals to promote proliferation, the increased number of focal adhesions may inhibit cell proliferation.

Oligomerization of syndecan-4 plays a critical role in the regulation of focal adhesion formation (Choi et al. 2005). Syndecan-4 has the tendency to form SDS-resistant dimers or oligomers (Rapraeger 2000). Except for the transmembrane domain, syndecan-4 cytoplasmic domain and extracellular domain participate in syndecan-4 dimer formation (Carey 1997; Oh et al. 1997). Whether syndecan-4 GAG chains and N-glycosylated chains are involved in dimer formation is not well understood at the present time. Sareneva et al. (1994) reported that the N-glycosylated chains are required for proteins to form dimers. Syndecan-4 N-glycosylated chains may influence syndecan-4 dimer formation. Syndecan-4 GAG chains have been reported to play an important role in the initial binding of syndecan-4 to focal adhesions (Woods & Couchman 2001). It is possible that when both GAG chains and N-glycosylated chains are deleted, syndecan-4 regulated focal adhesion formation is decreased, thus, the cells may have increased migration and a higher rate of proliferation. When only the GAG chains or N-glycosylated chains are deleted, syndecan-4 still has the ability to increase focal adhesion formation. As a result, cells will have decreased migration and proliferation.

Liu et al. (2006) found that syndecan-4 expression is higher during satellite cell proliferation but not during differentiation, which suggested that syndecan-4 may play an important role in regulating satellite cell proliferation but not differentiation. Zhang et al. (2008) reported that syndecan-4 GAG chains are not required for the initial differentiation. In the current study, syndecan-4 N-glycosylated chain mutants with or without GAG chains did not influence turkey myogenic satellite cell differentiation compared to wild type syndecan-4, which suggests that both GAG chains and N-glycosylated chains are not required for syndecan-4 to regulate turkey myogenic satellite cell differentiation.

The addition of exogenous FGF2 to satellite cell cultures increased proliferation and decreased differentiation reflecting its stimulating effect on proliferation and inhibitory effect on differentiation. However, the cellular responsiveness to FGF2 during both proliferation and differentiation suggest that syndecan-4 GAG and N-glycosylated chains are not required for the cellular response to FGF2. In other studies, syndecan-4 has been shown to function in an FGF2-independent manner (Velleman et al. 2007; Zhang et al. 2008). Syndecan-4 may regulate satellite cell growth and development through other signaling pathways like protein kinase C alpha (PKCα). In the presence of phosphatidylinositol 4,5-bisphosphate, the syndecan-4 cytoplasmic tail can bind to PKCα and activate PKCα (Horowitz & Simons 1998; Oh et al. 1998). The activated PKCα can stimulate downstream signaling pathways to regulate cell growth and development. In summary, the data from the current study suggested that syndecan-4 N-glycosylated chains and GAG chains are required for satellite cell proliferation, but not for differentiation and cellular FGF2 responsiveness during proliferation and differentiation.

Acknowledgments

Salary and partial research support to S.G.V. were provided by state and federal funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University, the Midwest Poultry Consortium to S.G.V. and D.C.M., and this project was supported by National Research Initiative Competitive Grant no. 2009-35503-05176 to S.G.V. from the USDA National Institute of Food and Agriculture.

Ancillary