The Influence of TGF-β3,EGF,and BGN on SOX9 and RUNX2 Expression in Human Chondrogenic Progenitor Cells

Abstract

Summary

Osteoarthritis (OA) is the most common chronic joint disease and leads to the degradation of the extracellular matrix by an imbalance between anabolic and catabolic processes. TGF-β3 (transforming growth factor beta-3) and epidermal growth factor (EGF) influence the osteochondrogenic potential of chondrocytes. In this study, we compared the expression of mediators and receptors in the TGF-β3 and EGF pathways, as well as biglycan (BGN), in healthy and diseased chondrocytes. Furthermore, we used chondrogenic progenitor cells (CPCs) for in vitro stimulation and knockdown experiments to elucidate the effects of TGF-β3 and EGF on the chondrogenic potential. Our results demonstrate that the expression of TGF-beta receptor type-1 (TGFBRI) and epidermal growth factor receptor (EGFR) is altered in diseased chondrocytes as well as in CPCs. Moreover, TGF-β3 and EGF stimulation influenced the expression levels of BGN, SRY (sex determining region Y)-box 9 (SOX9), and Runt-related transcription factor 2 (RUNX2) in CPCs. Therefore, changes in TGFBRI and EGFR expression likely contribute to the degenerative and regenerative effects seen in late stages of OA.

Keywords

articular cartilage chondrocytes chondrogenesis extracellular matrix

Introduction

Arthritis is projected to affect 78.4 million U.S. adults by 2040.¹ The most common type is osteoarthritis (OA), and its prevalence rises with age, thus affecting a majority of individuals above the age of 65.² OA is characterized by cartilage breakdown and results in joint failure.^3,4 To date, therapeutic interventions can alleviate symptoms but cannot cure the disease.³

One novel strategy is the activation of intrinsic reparative cells.⁵ A migratory population of chondrogenic progenitor cells (CPCs), which exhibit pluripotent capacities, was identified in the late stage of OA.⁶ Like chondrocytes, CPCs are embedded in an extracellular matrix (ECM) containing collagens (COL), proteoglycans, and glycoproteins.^6–8 The pericellular matrix, the innermost part, is important for signal transduction.⁹ Investigations suggest that altered cell-matrix interactions lead to the upregulation of the MMP13 (matrix metallopeptidase 13) in chondrocytes.¹⁰ Furthermore, pericellular matrix proteins, such as agrin, laminin, and nidogen, were found to enhance chondrogenesis by upregulating SRY (sex determining region Y)-box 9 (SOX9).^11,12 Biglycan (BGN) is a small leucine-rich proteoglycan that interacts with members of the TGF-β3 (transforming growth factor beta-3) pathway,^13,14 as well as with members of the epidermal growth factor (EGF) pathway,¹⁵ and binds to mediators and receptors to increase its activity.¹⁶ The TGF-β3 and EGF signaling pathways are both known to influence the differentiation of chondrocytes.^17,18 Furthermore, BGN-deficient mice exhibit an OA phenotype and increased cartilage degradation.¹⁵ In addition, BGN was upregulated in late-stage OA tissues, where it might compensate for the general loss of proteoglycans.⁷

In this study, we investigated if the ECM molecule BGN, as well as the TGF-β3 and EGF signaling pathways, affect the chondrogenic differentiation of CPCs.

Materials and Methods

Tissue Source

Cartilage samples were histopathologically classified according to the Osteoarthritis Research Society International (ORSI) standards¹⁹ and prepared as described elsewhere.⁶ Briefly, we utilized samples from the lateral condyle of knee joints that were collected from a region directly adjacent to the main defect with Grade 4.0 to 4.5.¹⁹ The respective area is depicted in the Supplemental Figure S1 of Koelling et al.⁶ The samples derived from the deep and middle zones exhibiting chondrocyte clusters and deep surface fissures (Mankin grade IV). The patients (n=15 [seven males and eight females]) with a mean age of 67.1 (range = 44 to 85) years met the American College of Rheumatology classification for OA²⁰ and gave their written informed consent, consistent with the ethical regulations of our institution. Healthy specimens from three accident victims revealed macroscopically and histologically intact hyaline cartilage with a smooth surface, and all of the layers were visible and corresponded to an ORSI grade of 0 to 1. These healthy specimens were also taken from the lateral condyle of knee joints to match the OA samples.

Cell Culture

CPCs (241ht cell line) were cultured⁶ and cultivated under standard conditions at 37C and 5% CO₂. Three-dimensional culture was performed as described elsewhere.⁶

Immunohistochemistry (IHC) and Immunocytochemistry (ICC)

Primary antibodies were applied in the following concentrations: BGN, 1:100, polyclonal rabbit H-150, sc-33788 (IHC); BGN, 1:100, monoclonal mouse 3E2, sc-100857 (ICC); EGF, 1:100, polyclonal rabbit Z-12:sc-275; EGFR, 1:100, polyclonal rabbit 1005:sc-03; TGFBRI, 1:100, polyclonal rabbit (V-22):sc-398; TGF-β3, 1:100, polyclonal rabbit V:sc-82; all from Santa Cruz (Heidelberg, Germany). For IHC, a HiDef Detection Alk Phos Polymer System (962D, Cell Marque, Rocklin, CA) with PermaRed/AP-Auto (K049-Auto, Cell Marque) was applied. Polyclonal donkey antirabbit Alexa Fluor 555, ab150074, 1:500 (Abcam, Cambridge, UK) and polyclonal goat anti-mouse DyLight 488, 072-03-18-06, 1:500 (KPL) secondary antibodies were applied for ICC.

Cell Isolation

To obtain chondrocytes from healthy (CC-H) and diseased cartilage (CC-D), tissue samples from the deep zones of 5 to 10 mm³ were treated with 10 mg of collagenase type I (17100-017, Gibco, Karlsruhe, Germany) and 10 mg of collagenase type II (C6885-1G, Sigma-Aldrich, Steinheim, Germany) in 10 mL of medium for 6 hr at 37C. The suspension was filtered, and the cells were distributed into cell culture flasks. In the case of CPCs, a standard explant protocol was performed.⁶

Real-time Polymerase Chain Reaction (RT-PCR)

RNA was isolated using the QIAshredder (79654, Qiagen, Hilden, Germany) and RNeasy Plus Mini Kit (74134, Qiagen) following the manufacturer’s instructions, and the QuantiTect Reverse Transcription Kit (205313, Qiagen) was used to synthesize cDNA. Primers were designed with Primer3web 4.0.0²¹ and ordered from Eurofins MWG Operon. PCR was performed by an Eppendorf RealPlex² system in a total volume of 10 µL with 5 µL of SYBR Green qPCRSuperMix-UDG (KK4600, Invitrogen, Darmstadt, Germany). Determination of the relative mRNA ratio was performed²² using β₂ microglobulin for normalization. The products were sequenced (SeqLab, Goettingen, Germany) and verified using the NCBI library. Primers were ordered from Eurofin Genomics (Ebersberg, Germany). The primers used were as follows: ACAN fw, 5′ ACAGCTGGGGACAT; ACAN rev, 5′ GTGGAATGCAGAGG; BGN fw, 5′ AATGAACTCCACCTAGACCACAA; BGN rev, 5′ GATGTTGTTGGAGTGCAGATAGAC; COL1 fw, 5′ TTCCCCCAGCCACAAAGAGTC; COL1 rev, 5′ CGTCATCGCACAACACCT; COL2 fw, 5′ CTCCTGGAGCATCTGGAGAC, COL2 rev, 5′ ACCACGATCACCCTTGACTC; COL10 fw, 5′ GCTAAGGGTGAAAGGGGTTC; COL10 rev, 5′ CTCCAGGATCACCTTTTGGA; EGFR fw, 5′ CGACAGCTATGAGATGGAGGA; EGFR rev, 5′ GATCCAGAGGAGGAGTATGTGTG; RUNX2 fw, 5′ TTCCAGACCAGCAGCACTC; RUNX2 rev, 5′ CAGCGTCAACACCATCATT; SOX9 fw, 5′ CAGGCTTTGCGATTTAAGGA; SOX9 rev, 5′ CCGTTTTAAGGCTCAAGGTG; TGFBRI fw, 5′ AACCTGCTCTCCTGCTTGCT; TGFBRI rev, 5′ CTCCCTTCCACCTCTAATGACTGA.

TGF-β3 and EGF Stimulation

CPCs were cultured in 3D (three-dimensional) alginate for 3 days before stimulation with 2 ng/mL of TGF-β3 (T5425, Sigma-Aldrich) or using 10, 20, or 40 ng/mL of hrEGF (354052, BD Science, Heidelberg, Germany) for 24 hr. Costimulation was performed with 2 ng/mL of TGF-β3 and 10 ng/mL of hrEGF. For the detection of SOX9 protein, cells were stimulated with 10 ng/mL of hrEGF and additional 10 ng/mL hrEGF for 4 hr before harvesting.

Immunoblotting

Protein amounts were determined by SDS-PAGE and Western blotting (WB) using PVDF (polyvinylidene difluoride) membranes. TBS-T was used for washing and 5% milk powder in TBS-T for blocking. Primary antibodies were diluted—BGN: 1:1000, polyclonal rabbit H-150, sc-33788 (Santa Cruz); EGFR: 1:1000, polyclonal rabbit 1005:sc-03 (Santa Cruz); p-SMAD2: 1:500, polyclonal rabbit #3101 (Cell Signaling, Leiden, Netherlands); RUNX2: 1:2000, polyclonal rabbit ab23981 (Abcam); SOX9: 1:2000, monoclonal mouse H00006662-M02 (Abnova, Heidelberg, Germany)—and incubated overnight. Secondary antibodies were diluted—Polyclonal goat anti-mouse A 9917, 1:40000 (Sigma-Aldrich) and Polyclonal goat antirabbit, A 0545, 1:100000 (Sigma-Aldrich)—and incubated for 1 hr. WesternBright Sirius (K-12043-D20, Biozym, Hess. Oldendorf, Germany) and WesternBright ECL (K-12045-D20, Biozym) were used for detection, and evaluation was performed with ImageJ Software (Version 1.07) with α-tubulin (T6199, 1:10000, Sigma-Aldrich) for normalization.²³

siRNA Knockdown

Cells (5 × 10⁵) were transfected with 10 µmol/mL BGN siRNA (SR300431; OriGENE Technologies, Herford, Germany) via the Human MSC Nucleofector Kit (VAPE-1001, Lonza, Basel, Switzerland) using program U-23 of NucleofectorII (Lonza). Scrambled AllStars-negative siRNA (102784, Qiagen) was used as a negative control. Cells were cultured in 2D (two-dimensional) for 3 days.

Statistical Analysis

SPSS version 13.0 was used (IBM, Ehningen, Germany). qPCR was performed with at least n=6 and WB with n=3 samples. The Shapiro–Wilk test was used to test for normality of distribution, and two-tailed one-sample t-tests were used for significance. Bars show the mean ± SD change compared with the untreated control (set as 1); *p<0.05.

Results

Detection of the Key Players

Initially, we localized BGN, which influences TGF-β3 and EGF signaling by interacting with cell surface receptors. Changes in BGN expression during OA might play a role in disease progression due to altered interactions with these signaling pathways. Occurrence of BGN was confirmed by IHC in healthy (Fig. 1A) and diseased (Fig. 1B) cartilage samples. In the next step, we localized TGF-beta receptor type-1 (TGFBRI) and epidermal growth factor receptor (EGFR) by IHC. The diseased tissue sample lacks the superficial zone (SZ), thus we localized TGFBRI (Fig. 1C–E) and EGFR (Fig. 1F–H) staining of chondrocytes residing in the diseased tissue (CC-D) only in the middle and deep zones.

Figure 1.

Detection of BGN, TGFBRI, and EGFR. Staining of BGN in (A) healthy tissue and (B) late-stage OA tissue. CC-D exhibited positive IHC staining of TGFBRI in the (C) middle and (D) deep zones. (E) Membranous staining of TGFBRI in CC-D. EGFR staining of CC-D residing in the middle (F) and deep (G) zones. (H) Membranous staining of EGFR in CC-D. Relative expression levels of (I) TGFBRI and (J) EGFR mRNA in CC-D and CPCs compared with in CC-H. Arrows indicate duplication of the tidemark. Scale bar in A and B: 750 µm. Scale bar in C, D, F, and G: 150 µm. Scale bar in E and H: 75 µm. Abbreviations: BGN, biglycan; TGFBRI, TGF-beta receptor type-1; EGFR, epidermal growth factor receptor; OA, osteoarthritis; CC-D, chondrocytes residing in diseased tissue; IHC, immunohistochemistry; CC-H, chondrocytes residing in healthy tissue; CPC, chondrogenic progenitor cell; SZ, superficial zone; MZ, middle zone; DZ, deep zone.

We found that TGFBRI expression in the CPCs was significantly increased (Fig. 1I). Contrary to this, EGFR expression was reduced in CC-D and was even less in the CPCs (Fig. 1J).

To further investigate a possible interaction between BGN and both receptors, we performed an ICC double staining in vitro. In both cases, we observed BGN together with either TGFBRI (Fig. 2A–D) or EGFR (Fig. 2E–H), which might indicate a colocalization of BGN with these receptors.

Figure 2.

CPCs were stained for TGFBRI and EGFR by ICC. (A and E) Nuclei were stained with DAPI. (B and F) Staining of BGN. (C) TGFBRI signal. (D) Colocalization of BGN and TGFBRI. (G) EGFR signal. (H) Colocalization of BGN and EGFR. Scale bar: 75 µm. Abbreviations: CPCs, chondrogenic progenitor cells; TGFBRI, TGF-beta receptor type-1; EGFR, epidermal growth factor receptor; ICC, immunocytochemistry; DAPI, 4′,6-diamidino-2-phenylindole; BGN, biglycan.

After detecting the altered receptor expression in OA, we investigated the distribution of TGF-β3 and EGF in healthy and diseased cartilage (Fig. 3). TGF-β3 was expressed in healthy (Fig. 3A) and OA cartilage specimens (Fig. 3B), including the middle (Fig. 3C) and deep (Fig. 3D) zones. Also, EGF was expressed in healthy (Fig. 3E) and diseased cartilage (Fig. 3F). Representative IHC images of the middle zone and deep zone in diseased cartilage are shown in Fig. 3G and Fig. 3H, respectively.

Figure 3.

TGF-β3 and EGF expression by IHC. Staining of TGF-β3 in (A) healthy and (B) diseased cartilage. Higher magnification of the MZ (C) and DZ (D) of diseased cartilage. EGF was found in healthy (E) and diseased (F) cartilage. EGF immunohistology of diseased cartilage MZ (G) and DZ (H). Scale bar in A, B, E, F: 750 µm. Scale bar in C, D, G, and H: 150 µm. Abbreviations: TGF-β3, transforming growth factor beta-3; EGF, epidermal growth factor; IHC, immunohistochemistry; MZ, middle zone; DZ, deep zone; SZ, superficial zone.

TGF-β3, EGF, and BGN Affect Osteochondrogenic Marker Expression

TGF-β3, a mediator of one of the major signaling cascades for chondrogenic differentiation, has been recently described as having a pro-chondrogenic effect in developing cartilage but turns into a major pathological effector in OA.²⁴ Furthermore, the influence of the EGF signaling cascade exhibits positive²⁵ and negative effects¹⁷ on chondrogenesis. In the following experiments, we investigated the role of TGF-β3, EGF, and BGN on CPCs. We considered 2 ng/mL TGF-β3 and 10 ng/mL. High concentrations of EGF may inhibit cell growth²⁶ and EGFR signaling is linked to terminal differentiation and apoptosis in chondrocytes.²⁷

Stimulation with TGF-β3, EGF, or the combination of both increased TGFBRI, mothers against decapentaplegic homolog 2 (SMAD2), EGFR, and BGN expression (Fig. 4A). Next to the upregulation of SMAD2, we detected that TGF-β3 stimulation led to SMAD2 phosphorylation (Fig. 4B) that also has been described for chondrocytes.²⁴ The upregulation of BGN is in strong agreement with BGN upregulation of Mg-63 cells after TGF-β3 treatment.²⁸ Furthermore, we demonstrated that BGN knockdown decreased the amount of EGFR (Fig. 4C, 4D).

Figure 4.

Stimulation and knockdown experiments affect CPCs. (A) Effect of TGF-β3 and EGF stimulation on the TGFBRI, SMAD2, EGFR, and BGN mRNA level. (B). Representative WB of p-SMAD2 after TGF-β3 and EGF stimulation. (C) Effect of BGN knockdown on EGFR protein level. (D) Representative WB of BGN knockdown. (E) Stimulation of CPCs by TGF-β3 and EGF altered SOX9 and RUNX2 mRNA expression. (F) Analysis of SOX9 protein level after stimulation with EGF. (G) Representative WB of SOX9 after EGF stimulation. (H) Chondrogenic and (I) osteogenic (COL1) as well as hypertrophy (COL10) marker expression after TGF-β3 and EGF stimulation. Abbreviations: ACAN, aggrecan; COL, collagen; TGFBRI, TGF-beta receptor type-1; TGF-β3, transforming growth factor beta-3; EGFR, epidermal growth factor receptor; BGN, biglycan; EGF, epidermal growth factor; SMAD2, mothers against decapentaplegic homolog 2; WB, Western blotting; CPC, chondrogenic progenitor cell.

SOX9 and Runt-related transcription factor 2 (RUNX2) are the master regulators of CPC osteochondrogenesis.⁶ Therefore, we investigated the effect of TGF-β3 and EGF stimulation on their expression level. As the previous stimulation with 10 ng/mL EGF resulted in only a minor increase of BGN, we also varied the stimulation dose up to 40 ng/mL EGF. We found that both TGF-β3 and EGF strongly enhanced SOX9 expression (Fig. 4E). Furthermore, RUNX2 was also increased, except that stimulation via 40 ng/mL of EGF left RUNX2 unchanged. This may be due to cell toxicity of the increasing EGF concentration. Furthermore, the combined stimulation of TGF-β3 and EGF strongly increased RUNX2 expression.

The expression of SOX9 is the strongest indicator for chondrogenesis; therefore, we wanted to investigate if the increased mRNA level of SOX9 is capable to increase the protein level. We chose to stimulate cells with only 10 ng/mL EGF, as we have shown that RUNX2 expression was only slightly upregulated under this concentration of EGF. Moreover, to avoid any cytotoxic effects, but to obtain a strong pro-chondrogenic affect, after 24 hr of EGF stimulation, we additionally stimulated for 4 hr with fresh EGF. Indeed, we observed an increased SOX9 protein level (Fig. 4F and Fig. 4G).

Finally, we investigated the expression levels of ECM markers, as the increased SOX9 and RUNX2 expression levels influence the chondrogenic or the osteogenic potential. TGF-β3 and the combined stimulation with EGF increased both aggrecan (ACAN) and COL2 mRNA levels, whereas exclusive EGF stimulation decreased both markers’ expression (Fig. 4H). Furthermore, osteogenic and hypertrophy markers like COL1 and COL10, respectively, were not decreased by any of the stimulations (Fig. 4I), thus neither TGF-β3 nor EGF stimulation displayed only the desired pro-chondrogenic effect but also increased at least one of the osteogenic and hypertrophy markers.

Discussion

We investigated the influence of TGF-β3 and EGF and how they are linked to the ECM component BGN with regard to SOX9 and RUNX2, which are accredited as major regulators of osteochondrogenesis.

CPCs build a fibrocartilaginous repair tissue, its fiber network, mainly containing COL1 and not COL2, is insufficient to regenerate a healthy hyaline cartilage.⁸ However, the chondrogenic potential of CPCs can be enhanced by upregulation of SOX9, thus improving their regeneration attempts.^5,6 The manipulation of signaling cascades is one way to influence SOX9 expression, for example, the TGF-β3 pathway is known to enhance chondrogenesis in chondrocytes as well as in meniscus cells.²⁹ Also the EGF pathway has been linked to osteochondrogenesis,²⁵ and further evidence of an EGFR-SOX9 signaling cascade was found in another developmental context.³⁰

TGF-β3 and EGF Signaling Differs in OA

After we localized BGN, TGFBRI, and EGFR to confirm that these molecules could also be involved in human articular cartilage physiology, we further investigated the role of these players. First, we focused on the differences in TGFBRI expression between CC-H and CC-D as reduced expression of TGFBRI has been linked to OA in a mouse model.³¹ As we did not observe a significant difference, we conclude that either the cells investigated here reacted to the artificial culturing conditions or that other factors including posttranslational modifications of TGFBRI, as well as their ligand concentration, play a role. However, CPCs exhibited an increased expression of TGFBRI, which implies that this receptor takes part in the OA disease phenotype. This multipotent cell population resides in late-stage OA tissues and is capable of differentiating into chondrocytes and other cell lineages; therefore, the decrease in receptor expression may be related to their differentiation potential.^32,33 EGF signaling is strongly involved in chondrogenesis by elevating the level of SOX9.^34,35 The downregulation of EGFR expression in CC-D is likely an indicator of the catabolic changes that the cartilage undergoes in OA, and the lack of EGFR promotes chondrocyte maturation.³⁶ In earlier studies, we demonstrated that CPCs are regulated by the antagonists RUNX2 and SOX9 and take part in the formation of the fibrocartilaginous repair tissue in late-stage OA. Reduced pro-chondrogenic EGFR signaling is in strong agreement with the RUNX2-controlled, fibroblastic phenotype of CPCs in vivo and in vitro.

Influence of TGF-β3, EGF, and BGN Manipulation on the Chondrogenic Potential of CPCs

In articular cartilage, altered ECM homeostasis is a major factor contributing to OA. The ECM component BGN interferes with members of the TGF-β family^13,14 and EGF¹⁵ signaling pathways. We detected TGF-β3 and EGF in human articular cartilage and found evidence that in vitro stimulation with these players also upregulates BGN in CPCs. The upregulation of BGN after TGF-β1 TGF-β2, and TGF-β3 treatment has been shown in various cell types, including osteosarcoma cells and VSMCs (vascular smooth muscle cells).^28,37 Both cell types are strongly regulated by RUNX2 and SOX9,^38–41 as are CPCs.⁶ Furthermore, we observed that BGN influences the EGFR protein level as it has been stated before in a mouse model¹⁵ and, thus, one can speculate about a complex regulation chain in which BGN mediates the TGF-β3 and EGF pathways. This is supported by the colocalization of BGN and the respective receptors as shown here. Recent reviews have focused on the diverse functions of the ECM molecule perlecan,⁴² suggesting that proteoglycans act as central signaling hubs through interactions with other ECM molecules and cell receptors.⁴³ Loss of such a hub may lead to dysbalanced signaling resulting in the initially increased anabolic and finally catabolic processes that have been observed after BGN loss in mandibular condylar chondrocytes.¹³ Also the results of previous studies concerning TGF-β3 and EGF signaling^{31,36,44–46} indicate a highly regulated network of various players at various developmental timepoints necessary to determine the final osteochondrogenic effects. We showed that TGF-β3 and EGF stimulation increased expression of BGN, EGFR, and TGFBRI and seems to upregulate the osteochondrogenic potential in general, considering the rise of SOX9 and RUNX2 expression. In accordance with this, TGF-β3 stimulation upregulated chondrogenic, osteogenic, and hypertrophic markers, which we could not observe after exclusive EGF stimulation. EGF stimulation upregulated SOX9, however also decreased ACAN and COL2 and increased COL1 expression. Thus, TGF-β3 and EGF signaling have a potential for anabolic and catabolic processes, but their specificity is determined in a more complex manner, including further mediators activating alternative signaling cascades, for example, the proposed transition from pro-chondrogenic SMAD2/3 signaling to hypertrophic SMAD1/5/8 signaling^24,46 as shown for TGF-β3. In late-stage OA, dysregulation of this network may promote disease progression. BGN has been linked to angiogenesis during fracture repair by altering the expression and function of endostatin.⁴⁷ Altered TGF and EGF signaling may influence BGN expression and its regulatory functions in the deep zone, resulting in the vascularization of the calcified zone.

Taken together, TGF, EGF, and BGN were found to be expressed in articular cartilage and bear the capacity to influence each other, as well as the expression of the osteochondrogenic regulators SOX9 and RUNX2. The direct stimulation of CPCs in vitro with TGF-β3 and EGF increased the chondrogenic master regulator SOX9 but also increased osteogenic and hypertrophy markers. Further research is needed to determine exclusively chondrogenic mediators in these signaling cascades, which would help to identify attractive targets for future OA therapy.

Footnotes

Competing Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Contributions

JNJ contributed to study conception, study design, analyses and interpretation of data, figure preparation, manuscript preparation, and statistical analyses; SB contributed for study design, analyses and interpretation of data, figure preparation, manuscript preparation, and statistical analyses; BS contributed to manuscript preparation; SS contributed to acquisition of data, analyses, and interpretation of data; CB contributed to acquisition of data; and NM contributed to study conception, study design, interpretation of data, and manuscript preparation.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Deutsche Forschungsgesellschaft (Mi-573/10-2).

References

Hootman

Helmick

Barbour

Theis

Boring

MA.

Updated projected prevalence of self-reported doctor-diagnosed arthritis and arthritis-attributable activity limitation among US adults, 2015-2040. Arthritis Rheumatol. 2016;68(7):1582–7. doi:10.1002/art.39692.

Dahaghin

Bierma-Zeinstra

Ginai

Pols

Hazes

Koes

BW.

Prevalence and pattern of radiographic hand osteoarthritis and association with pain and disability (the Rotterdam study). Ann Rheum Dis. 2005;64(5):682–7. doi:10.1136/ard.2004.023564.

Buckwalter

Mankin

HJ.

Articular cartilage: degeneration and osteoarthritis, repair, regeneration, and transplantation. Instr Course Lect. 1998;47:487–504.

Horton

Jr Bennion

Yang

Cellular, molecular, and matrix changes in cartilage during aging and osteoarthritis. J Musculoskel Neuron. 2006;6(4):379–81.

Schminke

Miosge

Cartilage repair in vivo: the role of migratory progenitor cells. Curr Rheumatol Rep. 2014;16(11):461. doi:10.1007/s11926-014-0461-4.

Koelling

Kruegel

Irmer

Path

Sadowski

Miro

Miosge

Migratory chondrogenic progenitor cells from repair tissue during the later stages of human osteoarthritis. Cell Stem Cell. 2009;4(4):324–35. doi:10.1016/j.stem.2009.01.015.

Bock

Michaeli

Bode

Schultz

Kresse

Herken

Miosge

The small proteoglycans decorin and biglycan in human articular cartilage of late-stage osteoarthritis. Osteoarthritis Cartilage. 2001;9(7):654–63. doi:10.1053/joca.2001.0420.

Miosge

Waletzko

Bode

Quondamatteo

Schultz

Herken

Light and electron microscopic in-situ hybridization of collagen type I and type II mRNA in the fibrocartilaginous tissue of late-stage osteoarthritis. Osteoarthritis Cartilage. 1998;6(4):278–85. doi:10.1053/joca.1998.0121.

Eggli

Herrmann

Hunziker

Schenk

RK.

Matrix compartments in the growth plate of the proximal tibia of rats. Anat Rec. 1985;211(3):246–57. doi:10.1002/ar.1092110304.

10.

Peng

Goldring

Olsen

Activation of the discoidin domain receptor 2 induces expression of matrix metalloproteinase 13 associated with osteoarthritis in mice. J Biol Chem. 2005;280(1):548–55. doi:10.1074/jbc.M411036200.

11.

Schminke

Frese

Bode

Goldring

Miosge

Laminins and nidogens in the pericellular matrix of chondrocytes: their role in osteoarthritis and chondrogenic differentiation. Am J Pathol. 2016;186(2):410–8. doi:10.1016/j.ajpath.2015.10.014.

12.

Eldridge

Nalesso

Ismail

Vicente-Greco

Kabouridis

Ramachandran

Niemeier

Herz

Pitzalis

Perretti

Dell’Accio

Agrin mediates chondrocyte homeostasis and requires both LRP4 and alpha-dystroglycan to enhance cartilage formation in vitro and in vivo. Ann Rheum Dis. 2016;75(6):1228–35. doi:10.1136/annrheumdis-2015-207316.

13.

Embree

Kilts

Ono

Inkson

Syed-Picard

Karsdal

Oldberg

Young

MF.

Biglycan and fibromodulin have essential roles in regulating chondrogenesis and extracellular matrix turnover in temporomandibular joint osteoarthritis. Am J Pathol. 2010;176(2):812–26. doi:10.2353/ajpath.2010.090450.

14.

Hara

Yoshida

Shinkai

Yamamoto

Fujiwara

Kumagai

Kaji

Biglycan intensifies ALK5-Smad2/3 signaling by TGF-beta1 and downregulates syndecan-4 in cultured vascular endothelial cells. J Cell Biochem. 2017;118(5):1087–96. doi:10.1002/jcb.25721.

15.

Iacob

Cs-Szabo

Biglycan regulates the expression of EGF receptors through EGF signaling pathways in human articular chondrocytes. Connect Tissue Res. 2010;51(5):347–58. doi:10.3109/03008200903427695.

16.

Nastase

Young

Schaefer

Biglycan: a multivalent proteoglycan providing structure and signals. J Histochem Cytochem. 2012;60(12):963–75. doi:10.1369/0022155412456380.

17.

Nonaka

Shum

Takahashi

Ikura

Dashner

Nuckolls

Slavkin

HC.

Convergence of the BMP and EGF signaling pathways on Smad1 in the regulation of chondrogenesis. Int J Dev Biol. 1999;43(8):795–807.

18.

Dahlin

Meretoja

Kasper

Mikos

AG.

TGF-beta3-induced chondrogenesis in co-cultures of chondrocytes and mesenchymal stem cells on biodegradable scaffolds. Biomaterials. 2014;35(1):123–32. doi:10.1016/j.biomaterials.2013.09.086.

19.

Pritzker

Gay

Jimenez

Ostergaard

Pelletier

Revell

Salter

van den Berg

WB.

Osteoarthritis cartilage histopathology: grading and staging. Osteoarthritis Cartilage. 2006;14(1):13–29. doi:10.1016/j.joca.2005.07.014.

20.

Altman

Asch

Bloch

Bole

Borenstein

Brandt

Christy

Cooke

Greenwald

Hochberg

Howell

Kaplan

Koopman

Longley

Mankin

McShane

Medsger

Jr Meenan

Mikkelsen

Murphy

Rothschild

Segal

Sokoloff

Wolfe

Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthritis Rheum. 1986;29(8):1039–49.

21.

Untergasser

Cutcutache

Koressaar

Faircloth

Remm

Rozen

SG.

Primer3—new capabilities and interfaces. Nucleic Acids Res. 2012;40(15):e115. doi:10.1093/nar/gks596.

22.

Pfaffl

MW.

A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;29(9):e45.

23.

Gassmann

Grenacher

Rohde

Vogel

Quantifying Western blots: pitfalls of densitometry. Electrophoresis. 2009;30(11):1845–55. doi:10.1002/elps.200800720.

24.

van der Kraan

. The changing role of TGFbeta in healthy, ageing and osteoarthritic joints. Nat Rev Rheumatol. 2017;13(3):155–63. doi:10.1038/nrrheum.2016.219.

25.

Shepard

Jeong

Maihle

O’Brien

Dealy

CN.

Transient anabolic effects accompany epidermal growth factor receptor signal activation in articular cartilage in vivo. Arthritis Res Ther. 2013;15(3):R60. doi:10.1186/ar4233.

26.

Kaplan

Jaroszewski

Faustino

Zugmaier

Ennis

Lippman

Cohen

JS.

Toxicity and effects of epidermal growth factor on glucose metabolism of MDA-468 human breast cancer cells. J Biol Chem. 1990;265(23):13641–9.

27.

Zhang

Zhu

Lin

Siclari

Chandra

Candela

Koyama

Enomoto-Iwamoto

Qin

Epidermal growth factor receptor (EGFR) signaling regulates epiphyseal cartilage development through beta-catenin-dependent and -independent pathways. J Biol Chemistry. 2013;288(45):32229–40. doi:10.1074/jbc.M113.463554.

28.

Heegaard

Xie

Young

Nielsen

KL.

Transforming growth factor beta stimulation of biglycan gene expression is potentially mediated by sp1 binding factors. J Cell Biochem. 2004;93(3):463–75. doi:10.1002/jcb.20189.

29.

Muhammad

Schminke

Bode

Roth

Albert

von der Heyde

Rosen

Miosge

Human migratory meniscus progenitor cells are controlled via the TGF-beta pathway. Stem Cell Reports. 2014;3(5):789–803. doi:10.1016/j.stemcr.2014.08.010.

30.

Grimont

Pinho

Cowley

Augereau

Mawson

Giry-Laterriere

Van den Steen

Waddell

Pajic

Sempoux

Grimmond

Biankin

Lemaigre

Rooman

Jacquemin

SOX9 regulates ERBB signalling in pancreatic cancer development. Gut. 2015;64(11):1790–9. doi:10.1136/gutjnl-2014-307075.

31.

Blaney Davidson

Scharstuhl

Vitters

van der Kraan

van den Berg

WB.

Reduced transforming growth factor-beta signaling in cartilage of old mice: role in impaired repair capacity. Arthritis Res Ther. 2005;7(6):R1338–47. doi:10.1186/ar1833.

32.

Ischenko

Liu

Petrenko

Hayman

MJ.

Transforming growth factor-beta signaling network regulates plasticity and lineage commitment of lung cancer cells. Cell Death Differ. 2014;21(8):1218–28. doi:10.1038/cdd.2014.38.

33.

Droguett

Cabello-Verrugio

Santander

Brandan

TGF-beta receptors, in a Smad-independent manner, are required for terminal skeletal muscle differentiation. Exp Cell Res. 2010;316(15):2487–503. doi:10.1016/j.yexcr.2010.04.031.

34.

Ling

Chang

Schultz

Lee

Chaux

Marchionni

Netto

Sidransky

Berman

DM.

An EGFR-ERK-SOX9 signaling cascade links urothelial development and regeneration to cancer. Cancer Res. 2011;71(11):3812–21. doi:10.1158/0008-5472.can-10-3072.

35.

Chen

Singh

Koenig

Liou

Storz

Zhang

Regul

Nagarajan

Kühnemuth

Johnsen

Hebrok

Siveke

Billadeau

Ellenrieder

Hessmann

NFATc1 links EGFR signaling to induction of Sox9 transcription and acinar-ductal transdifferentiation in the pancreas. Gastroenterology. 2015;148(5):1024–34.e9. doi:10.1053/j.gastro.2015.01.033.

36.

Sibilia

Wagner

Hoebertz

Elliott

Marino

Jochum

Wagner

EF.

Mice humanised for the EGF receptor display hypomorphic phenotypes in skin, bone and heart. Development. 2003;130(19):4515–25. doi:10.1242/dev.00664.

37.

Burch

Yang

Ballinger

Getachew

Osman

Little

PJ.

TGF-beta stimulates biglycan synthesis via p38 and ERK phosphorylation of the linker region of Smad2. Cell Mol Life Sci. 2010;67(12):2077–90. doi:10.1007/s00018-010-0315-9.

38.

Lucero

Vega

Osorio

Tapia

Antonelli

Stein

van Wijnen

Galindo

MA.

The cancer-related transcription factor Runx2 modulates cell proliferation in human osteosarcoma cell lines. J Cell Physiol. 2013;228(4):714–23. doi:10.1002/jcp.24218.

39.

Liu

Chen

Zhou

Jie

Dai

Wang

Zhou

Sox9 regulates hyperexpression of Wnt1 and Fzd1 in human osteosarcoma tissues and cells. Int J Clin Exp Pathol. 2014;7(8):4795–805.

40.

Lin

Chen

Leaf

Speer

Giachelli

CM.

Runx2 expression in smooth muscle cells is required for arterial medial calcification in mice. Am J Pathol. 2015;185(7):1958–69. doi:10.1016/j.ajpath.2015.03.020.

41.

Briot

Jaroszewicz

Warren

Touma

Rudat

Hofmann

Airik

Weinmaster

Lyons

Wang

Kispert

Pellegrini

Iruela-Arispe

ML.

Repression of Sox9 by Jag1 is continuously required to suppress the default chondrogenic fate of vascular smooth muscle cells. Dev Cell. 2014;31(6):707–21. doi:10.1016/j.devcel.2014.11.023.

42.

Douglass

Goyal

Iozzo

RV.

The role of perlecan and endorepellin in the control of tumor angiogenesis and endothelial cell autophagy. Connect Tissue Res. 2015;56(5):381–91. doi:10.3109/03008207.2015.1045297.

43.

Gubbiotti

Neill

Iozzo

RV.

A current view of perlecan in physiology and pathology: a mosaic of functions. Matrix Biol. 2017;57–58:285–98. doi:10.1016/j.matbio.2016.09.003.

44.

Zhang

Zhu

Liu

Chandra

Levin

Beier

Enomoto-Iwamoto

Qin

Reduced EGFR signaling enhances cartilage destruction in a mouse osteoarthritis model. Bone Res. 2014;2:14015. doi:10.1038/boneres.2014.15.

45.

Pombo-Suarez

Castano-Oreja

Calaza

Gomez-Reino

Gonzalez

Differential upregulation of the three transforming growth factor beta isoforms in human osteoarthritic cartilage. Ann Rheum Dis. 2009;68(4):568–71. doi:10.1136/ard.2008.090217.

46.

Bush

Beier

TGF-beta and osteoarthritis—the good and the bad. Nat Med. 2013;19(6):667–9. doi:10.1038/nm.3228.

47.

Myren

Kirby

Noonan

Maeda

Owens

Ricard-Blum

Kram

Kilts

Young

MF.

Biglycan potentially regulates angiogenesis during fracture repair by altering expression and function of endostatin. Matrix Biol. 2016;52–54:141–50. doi:10.1016/j.matbio.2016.03.008.