Role of Myostatin (GDF-8) Signaling in the Human Anterior Cruciate Ligament

Abstract

Myostatin, also referred to as growth and differentiation factor-8 (GDF-8), is expressed in muscle tissue where it functions to suppress myoblast proliferation and myofiber hypertrophy. Recently, myostatin and its receptor, the type IIB activin receptor (ActRIIB), were detected in the leg tendons of mice, and recombinant myostatin was shown to increase cellular proliferation and the expression of type 1 collagen in primary fibroblasts from mouse tendons. We sought to determine whether myostatin and its receptor were present in human anterior cruciate ligament (ACL) tissue, and if myostatin treatment had any effect on primary ACL fibroblasts. ACL tissue samples were obtained from material discarded during ACL reconstruction surgery. Real-time PCR and immunohistochemistry demonstrate that both myostatin and its receptor are abundant in the human ACL. Primary fibroblasts isolated from human ACL specimens were treated with recombinant myostatin, and myostatin treatment increased fibroblast proliferation as well as the expression of tenascin C, type 1 collagen, and transforming growth factor-β1. Real-time PCR analysis of tenascin C and type 1 collagen expression in ACL specimens from normal mice and mice lacking myostatin supported these findings by showing that both tenascin C and type 1 collagen were downregulated in ACL tissue from myostatin-deficient mice. Together, these data suggest that myostatin is a pro-fibrogenic factor that enhances cellular proliferation and extracellular matrix synthesis by ACL fibroblasts. Recombinant myostatin may therefore have therapeutic applications in the area of tendon and ligament engineering and regeneration.

Growth and differentiation factor 8 (GDF-8), also known as myostatin, is a member of the TGF beta superfamily of growth and differentiation factors that is highly expressed in skeletal muscle.¹ Myostatin binds the type IIB activin receptor (ActRIIB) and type I co-receptor (Alk4/5) to regulate the expression of downstream target genes such as myogenic differentiation protein (MyoD) and myogenic factor-5 (Myf-5) via a TGF beta signaling pathway.^2,3 Congenital absence of myostatin (GDF-8) is known to significantly increase muscle mass,⁴ and myostatin inhibitors can improve muscle mass and strength in laboratory animals.⁵ Recently it was found that myostatin and its receptor were highly expressed in the tendons of mice.⁶ Surprisingly, the tendons of mice lacking myostatin were observed to be relatively smaller in their cross-sectional area than those of normal mice, suggesting that myostatin-deficiency may alter the integrated growth, development, and adaptation of the myotendon complex. ⁶ Indeed, the tendons of mice lacking myostatin were also found to be relatively hypocellular compared to those of wild-type mice, and myostatin treatment of primary cells isolated from mouse tendon was found to increase fibroblast proliferation and type I collagen expression.⁶ Although it has previously been shown that blocking myostatin signaling in injured muscle decreases fibrosis during muscle regeneration,⁷ the fact that myostatin deficiency was associated with tendon hypoplasia was unexpected given the fact that the muscles of these same mice were increased in mass by more than 50%. These interesting findings suggest that myostatin might be involved in the development of multiple musculoskeletal tissues aside from just muscle, a hypothesis that is also supported by our earlier studies indicating a role for myostatin in bone formation,⁸ fracture healing,⁹ and cranial suture biology.¹⁰

The majority of studies examining myostatin signaling in musculoskeletal tissues have been performed in animal models. The goal of this study is first to determine whether or not myostatin and its receptor are expressed in human connective tissue, specifically the human anterior cruciate ligament (ACL), and second to determine how myostatin alters the proliferation and expression of extracellular matrix proteins in primary ACL fibroblasts. We focus here on the ACL because ACL tears are a common injury encountered by orthopaedic surgeons that are associated with considerable morbidity and healthcare cost. In terms of extracellular matrix molecules this study examines the role of myostatin in regulating tenascin C expression in the ACL. Tenascins are a family of extracellular matrix (ECM) glycoproteins that contribute to matrix structure, and tenascin expression is upregulated with mechanical loading.^11,12 Furthermore, tenascins are thought to play a direct role in the pathophysiology of tendon injury, as sequence variants of tenascin C (TNC) are associated with tendinopathies and Achilles tendon ruptures.¹³ The overarching hypothesis of the paper is that myostatin is a pro-fibrogenic, anti-myogenic factor that promotes fibroblast proliferation and extracellular matrix synthesis in tendons and ligaments, thereby regulating the size and mass of these fibrous connective tissues.

Methods

Preparation of human ACL samples

ACL tissue specimens were retrieved during ACL reconstruction surgery in 12 patients under the approval of Medical College of Georgia HAC protocol 05-12-135. Patients included 8 males and 6 females ranging in age from 14-31 years. Six samples were used for isolation of mRNA, cDNA synthesis, and real time PCR analysis of gene expression. Two samples were used for isolation of primary cells for cell culture studies, and four samples were used for immunohistochemistry.

Real-time PCR

Ligament tissue was wrapped in sterile gauze soaked in saline following surgery, placed on ice, and later homogenized in Trizol. Isolation of mRNA using Trizol was also used for primary fibroblasts described below. The samples were assayed for absorbance at 260 nm (Helios-Gamma, Thermo Spectronic, Rochester, NY), then reverse transcribed using iScript reagents from Bio-Rad on a programmable thermal cycler (PCR-Sprint, Thermo Electron, Milford, MA). 50 ng complementary deoxyribonucleic acid (cDNA) was amplified in each 40-cycle real-time polymerase chain reaction using a Bio-Rad iCycler, ABgene reagents (distributed by Fisher) and custom designed primers and probes specific to the human genome (Table 1). Data were analyzed as described previously.^{8, 9}

Table 1.

Primer sequences and accession numbers for human primers used in real time PCR studies.

Gene	Sequence	Product Size	Accession Number
Type IIB Activin Receptor (ACVR2B)	Fwd gggttcacctgtttctcacagt Rev agagcttccttgcttctacagc	98	NM_001106
Tenascin C (TNC)	Fwd gagatatggggacaataaccacag Rev atttctgaagttgcttggtctcag	112	NM_002160
Myostatin (MSTN)	Fwd caacggtgctaatacgataggc Rev gaggtgtaggaaaatgcacctg	113	NM_005259
Collagen type 1 alpha 1 (Col1A1α)	Fwd tctctcctctgaaaccctcctc Rev gtgctttgggaagttgtctctg	108	NM_000088
TGF beta 1 (TGFβ1)	Fwd ccctggacaccaactattgctt Rev cttgcggaagtcaatgtacagc	72	NM_000660.3
GAPDH	Fwd gagccacatcgctcagacac Rev catgtagttgaggtcaatgaagg	150	NM_002046.3

Immunohistochemistry

Two ACL samples were snap frozen in OCT for cryostat sectioning and two were fixed in paraformaldehyde for paraffin embedding. Slides were stained with antibodies to myostatin (rabbit anti-human, Santa Cruz Biotechnology clone H-109) and the myostatin receptor (rabbit anti-human ACVR2B, Santa Cruz Biotechnology clone H-70). Primary antibodies were conjugated with goat-anti rabbit secondary antibodies AlexaFluor488, which fluoresces in the FITC spectrum, and AlexaFluor546, which fluoresces in the Cy3 spectrum (Molecular Probes, Inc.). Slides were counterstained with DAPI (4′,6-diamidino-2-phenylindole), a fluorescent stain that binds DNA, to label nuclei and slides were mounted using aqueous medium.

Isolation and treatment of primary cells from ACL

ACL tissue was carefully trimmed of muscle and fat tissue, finely minced, and placed in 0.05% in type II collagenase (Worthington Biochemical, Lakewood, NJ) at 37°C overnight using Ham F12 medium (Mediatech, Herndon, with 4.5 g/L glucose) modified with 4.8 mM CaCl₂ (Sigma) and 40 mM HEPES buffer (Sigma). The cells were washed in phosphate buffered saline (PBS), plated in 100-mm tissue culture plates (Becton Dickinson Labware), then grown for one week with changes in Ham F12 medium containing 50 U/ml penicillin, 50 ug/ml streptomycin, 1% l-glutamine, and 10% fetal bovine serum. Fibroblasts were passaged once upon reaching 70% confluence and the experiments were carried out at 2^nd passage. Fibroblasts were plated on 12-well culture dishes for real time analysis and 24-well dishes for the proliferation assay and expanded until reaching 70% confluence. Fibroblasts were then starved of serum for 24 h before treatment by replacing serum containing media with Ham F12 media plus containing 50 U/ml penicillin,50 ug/ml streptomycin, and 1× ITS (Insulin Transferrin Selenium supplement, BD Biosciences). Recombinant myostatin (R&D Systems, Minneapolis) was dissolved into the serum-free media at a final concentration of 50 and 500 ng/ml. In a separate set of experiments, vehicle (serum-free media)-treated and myostatin-treated ACL fibroblasts were grown on glass coverslips and stained with an antibody to TGFβ1 (Santa Cruz Biotechnology SC-146, rabbit anti-human) and labeled with a biotynilated goat anti-rabbit secondary antibody (Vector Laboratories).

Proliferation assays

The number of viable ACL cells was determined using a Promega CellTiter 96® AQueous One MTS Cell Proliferation Assay. Briefly, ACL cells were plated in triplicate at an initial density of 15,000/cm² in 24-well plates (BD Labware) using supplemented Hams F12 medium containing 5% FBS to support overnight attachment. The following day, fibroblasts were starved of serum for 24 h before treatment by replacing serum containing media with Ham F12 media plus containing 50 U/ml penicillin, 50 ug/ml streptomycin plus 1× ITS (Insulin Transferrin Selenium supplement (BD Biosciences, Bedford, MA). The next day cells were fed with fresh supplemented Hams F12, substituting the FBS with 1% ITS and adding 0, 50 and 500 ng/ml of recombinant myostatin for 24hr ,48hrs and 72hrs. Cells were washed twice with PBS and 150ul of MTS (CellTiter 96® AQueous One Solution Reagent, Promega) assay buffer added. Cells were then incubated for 3 hr at 37°C in a humidified, 5% CO2 incubator. Optical density (OD) was read at 490 nm.

Mouse ACL Tissue Samples

Mice utilized in this study include 12 wild-type (+/+) mice and 12 myostatin-deficient (−/−) mice, six males and six females per genotype, on a CD-1 background. Animals were sacrificed according to IACUC-approved procedures. ACL tissue was harvested from the right knee of mice, tissue was homogenized, and the resultant RNA isolated. The RNA was reverse transcribed into cDNA, and this cDNA library was used in real-time PCR. A total of three genes were used in RT-PCR analysis: GAPDH (housekeeping gene), Tenascin C, and Type I collagen. Each gene was amplified for 45 cycles. A melt curve was used to assess purity of gene products and the data was normalized by GAPDH using the cycle threshold (CT) comparative method.

Statistical Analysis

Data are expressed as the mean ± SD. Differences in measured variables between experimental and control groups were assessed using Student's t-test. A p-value <0.05 was considered statistically significant in between-group comparisons.

Results

Myostatin and its receptor are expressed in the human ACL

Real-time PCR assays show that tenascin C, myostatin, and ActRIIB are all expressed in the human ACL (Fig. 1A). Myostatin expression is relatively greater than that measured for ActRIIB, and these results are supported by the immunohistochemical findings (Fig. 1B). Specifically, immunostaining shows that myostatin is abundant in the extracellular matrix of the ACL (Fig. 1B), whereas ActRIIB staining is primary localized to areas surrounding DAPI-positive nuclei (Fig. 1B).

Figure 1.

A. Real-time PCR data, shown relative to GAPDH expression, for tenascin C, myostatin (GDF-8), the myostatin receptor (ActRIIB), and col1 in anterior cruciate ligament (ACL) tissue from two patients following ACL reconstruction surgery. B. Immunofluorescent staining of ACL tissue using antibodies specific for human myostatin (GDF-8) and the myostatin receptor (ActRIIB), ×400.

Myostatin increases the proliferation of ACL fibroblasts

Treatment of primary fibroblasts isolated from patient ACL samples dose-dependently increased fibroblast proliferation after 24 hours (Fig. 2). The lower dose of myostatin significantly increased proliferation after both 48 and 72 hours (Fig. 2). The higher myostatin dose also significantly increased fibroblast proliferation relative to controls at 48 and 72 hours, but this increase was similar to that observed with the lower myostatin dose (Fig. 2).

Figure 2.

MTS proliferation assay of primary human ACL fibroblasts treated with increasing concentrations (0-500 ng/ml) of recombinant myostatin. Data were recorded 24, 48, and 72 hours following treatment. Means with different superscripts (a, b, c) differ significantly (P<.05) from one another in pairwise comparisons.

Myostatin treatment increases TNC, ColA1α, and TGFβ1 expression in ACL fibroblasts

Analysis of gene expression in fibroblasts treated with myostatin shows that there is no difference in Col1 expression after 24 hours of treatment, but the higher dose of myostatin increased Col1 expression after 48 and 72 hours (Fig. 3A). Likewise, tenascin C is expressed at lower levels than the housekeeping gene GAPDH after 24 hours of treatment, but the higher myostatin dose increased tenascin C expression relative to GAPDH after 48 and 72 hours (Fig. 3B). A similar pattern was observed with TGFβ1 expression, where the higher dose of myostatin increased TGFβ1 expression after 48 and 72 hours (Fig. 3C). ACL fibroblasts grown on coverslips and stained with an antibody to TGFβ1 further demonstrate that myostatin treatment not only increased cell proliferation but also increased TGFβ1 production (Fig. 3D).

Figure 3.

Real-time PCR analysis of type 1 collagen (A), tenascin C (B), and TGFβ (C) expression in primary human ACL fibroblasts treated with increasing concentrations (0-500 ng/ml) of recombinant myostatin. Data were recorded 24, 48, and 72 hours following treatment. Fibroblasts were also stained with an antibody for TGFβ1 (D), showing that myostatin treatment increased not only cell number but also abundance of TGFβ1 in the cytosol (brown stain; nuclei are stained blue with hematoxylin).

Tenascin C and type 1 collagen are downregulated in ACL tissue from myostatin-deficient mice

Real-time PCR analyses of anterior cruciate ligaments isolated from normal and myostatin-deficient mice show that congenital absence of myostatin is associated with a decrease in Col1 and tenascin C expression (Fig. 4).

Figure 4.

Real-time PCR data from the anterior cruciate ligaments of normal (WT) and myostatin-deficient (KO) mice showing decrease in type 1 collagen expression (A) and tenascin C expression (B) in the absence of myostatin. Data are expressed relative to the housekeeping gene GAPDH. * P<.01.

Discussion

The TGF-β superfamily of growth and differentiation factors includes a number of genes that are highly expressed in musculoskeletal tissues. In particular, GDF-5, -6, and -7 have all been demonstrated to play a role in the biology of tendons and ligaments. GDF-5 deficiency is associated with a decrease in tendon strength and delayed tendon healing^14,15. GDF-6 deficiency results in decreased collagen content in tail tendons of mail mice¹⁶, whereas absence of GDF-7 has much more minor effects on tendon composition and material properties¹⁷. The role of these factors in tendon biology is consistent with recent data from myostatin (GDF-8) deficient mice showing that tendons of myostatin knockouts are small, stiff, and hypocellular⁶. Together, the studies from animal models demonstrate that GDF signaling in tendon is critical for cellular proliferation, extracellular matrix formation, and tissue regeneration. To our knowledge, ours is the first study to report a role for growth and differentiation factors in the human anterior cruciate ligament. The finding that myostatin treatment increased the proliferation of ACL fibroblasts as well as the expression of genes involved in extracellular matrix synthesis supports findings from mouse primary tendon fibroblasts, in which exogenous myostatin increased cell proliferation and production of type 1 collagen⁶. Recombinant GDF-5 is already being employed to improve the healing of ligaments and tendons^18-20, and it is likely that other GDFs such as GDF-6 and -8 may have similar therapeutic potential. Indeed, a recent report indicated that exogenous myostatin increased extracellular matrix production and cross-sectional area of injured rat Achilles tendons²¹. In addition, the myostatin antagonist follistatin is decreased with loading^21-22, providing additional support for the hypothesis presented here that myostatin signaling plays an important role in regulating cell proliferation and extracellular matrix synthesis in tendons and ligaments.

One of the more significant findings in this regard is that myostatin treatment increased the expression of tenascin C in primary ACL fibroblasts, whereas myostatin deficiency was associated with a decline in tenascin C expression in the mouse ACL. Tenascins are a family of extracellular matrix (ECM) glycoproteins that contribute to matrix structure,²³ and tenascin expression is upregulated with mechanical loading.¹¹ Furthermore, tenascins are thought to play a direct role in the pathophysiology of tendon injury, as sequence variants of tenascin C (TNC) are associated with tendinopathies and Achilles tendon ruptures.¹² The tenascin C molecule is a hexabrachion-shaped structure that regulates cell adhesion and the attachment of cells to components of the extracellular matrix (ECM).²⁴ It is thought that during tensile loading tenascin C secretion releases contacts between the cell and the ECM, preventing potentially damaging overstretching and cell deformation.²³ Tenascin C expression is known to be induced by TGFβ1,²⁵ widely recognized as a pro-fibrotic growth factor. TGFβ1 expression is, in turn, induced by myostatin treatment in both C2C12 myoblasts and multipotent mouse C3H 10T1/2 mesenchymal stem cells.^26,27 These data suggest that myostatin may be an upstream regulator of tenascin C expression via its stimulatory effect on TGFβ1. The myostatin:follistatin ratio in connective tissues would therefore have added significance for functional adaptation, as the abundance of free myostatin would presumably mediate tenascin C secretion and hence the adhesion of fibroblasts to the ECM during tensile loading.

Growth factors such as FGF-2 and TGFβ1 have both been shown to enhance proliferation and ECM synthesis in cells isolated from the human ACL.²⁸ Future research is needed to gain a better understanding of how different growth factors interact to modulate ACL strength and adaptation under conditions of mechanical loading, disuse, and injury. It is clear from our experiments that myostatin, like other GDFs, plays a key role in ligament and tendon biology, and our data further suggest that exogenous delivery of recombinant myostatin may have therapeutic potential for tendon and ligament regeneration.