Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Nov 1;43(2):292–303. doi: 10.1002/jor.26004

Matrikine stimulation of equine synovial fibroblasts and chondrocytes results in an in vitro osteoarthritis phenotype

Rachel Gagliardi 1,2, Drew W Koch 1,2, Richard Loeser 3,4, Lauren V Schnabel 1,2,4,✉
PMCID: PMC11701414  PMID: 39486895

Abstract

Osteoarthritis (OA) is a debilitating disease that impacts millions of individuals and has limited therapeutic options. A significant hindrance to therapeutic discovery is the lack of in vitro OA models that translate reliably to in vivo preclinical animal models. An alternative to traditional inflammatory cytokine models is the matrikine stimulation model, in which fragments of matrix proteins naturally found in OA tissues and synovial fluid, are used to stimulate cells of the joint. The objective of this study was to determine if matrikine stimulation of equine synovial fibroblasts and chondrocytes with fibronectin fragments (FN7‐10) would result in an OA phenotype. We hypothesized that FN7‐10 stimulation of equine articular cells would result in an OA phenotype with gene and protein expression changes similar to those previously described for human chondrocytes stimulated with FN7‐10. Synovial fibroblasts and chondrocytes isolated from four horses were stimulated in monolayer culture for 6 or 18 h with 1 µM purified recombinant 42 kD FN7‐10 in serum‐free media. At the conclusion of stimulation, RNA was collected for targeted gene expression analysis and media for targeted protein production analysis. Consistent with our hypothesis, FN7‐10 stimulation resulted in significant alterations to many important genes that are involved in OA pathogenesis including increased expression of IL‐1β, IL‐4, IL‐6, CCL2/MCP‐1, CCL5/RANTES, CXCL6/GCP‐2, MMP‐1, MMP‐3, and MMP13. The results of this study suggest that the equine matrikine stimulation model of OA may prove useful for in vitro experiments leading up to preclinical trials.

Keywords: equine, fibronectin fragments, in vitro, osteoarthritis

1. INTRODUCTION

Osteoarthritis (OA) is a debilitating disease that impacts millions of individuals and has limited treatment options. 1 , 2 Currently, OA therapeutics are limited to symptom modification rather than disease modification. 3 There are no OA treatments that significantly slow the progression of the disease, and there are none that reverse the pathological changes that occur within the joint. 3 , 4 One significant challenge to the development of therapeutics for the treatment of OA is the lack of reliable and readily translatable in vitro models to screen novel therapeutics for future in vivo preclinical and clinical trials. Historically, in vitro models of OA have utilized stimulation of chondrocytes, and more recently synovial fibroblasts, with supraphysiological levels of inflammatory cytokines such as IL‐1β and TNF‐α or with lipopolysaccharide (LPS). 5 These stimulants invoke an intense, transient, and often inconsistent inflammatory response, rather than the chronic persistent low‐level inflammation that occurs with naturally occurring OA. 6 , 7

In 2021, Reed et al. reported on an alternative in vitro model of OA, utilizing stimulation of human chondrocytes with fibronectin fragments, which are a product of cartilage extracellular matrix degradation found in OA cartilage and synovial fluid. 8 , 9 Importantly, this matrikine model utilized a human recombinant fibronectin fragment containing domains 7‐10 (FN7‐10), which include the RGD cell‐binding motif that is recognized by the α5β1 integrin and plays a role in OA pathogenesis. 10 Previous studies, including one with equine cartilage explant cultures, have utilized fibronectin fragments from the N‐ and C‐terminal portions of fibronectin, containing the heparin‐binding domains I and II respectively, which are not involved in integrin signaling. 11 , 12 Using RNA‐seq, Reed et al. determined differentially expressed genes when comparing normal femoral chondrocytes harvested from aged donors (50–61 years) that were stimulated with FN7‐10 for 3, 6, and 18 h to those that were treated with saline‐ as a control. 8 These differentially expressed genes were then compared to a published chondrocyte gene set from human OA cartilage, revealing that the FN7‐10 model recapitulated many key aspects of the OA phenotype. 8

Because normal human synovial fibroblasts and chondrocytes are difficult to routinely obtain, and because horses are widely accepted and approved by the FDA as a preclinical model for OA, the objective of this study was to determine if matrikine stimulation of equine synovial fibroblasts and chondrocytes with FN7‐10 would result in an OA phenotype. We hypothesized that FN7‐10 stimulation of equine synovial fibroblasts and chondrocytes would result in an OA phenotype with targeted gene and protein expression changes similar to those described for human chondrocytes stimulated with FN7‐10.

2. METHODS

2.1. Samples

Synovium and cartilage samples were obtained postmortem using aseptic technique from the femoropatellar joints of four systemically healthy horses (2 females and 2 castrated males, ages 8–21 years, 3 Thoroughbreds and 1 Warmblood) euthanized for reasons unrelated to this study and as approved by the Institutional Animal Care and Use Committee of North Carolina State University under protocol #20‐454. The age range of the horses was chosen to mimic the middle‐aged nature of the human chondrocyte donors utilized in the previous FN7‐10 study and all femoropatellar joints were free of any gross signs of joint disease.

2.2. Synoviocyte isolation

Synovium was digested for synoviocyte isolation as previously described with modifications. 13 Briefly, synovium was placed in phosphate‐buffered saline (PBS) with penicillin (200 units/mL), and streptomycin (200 µg/mL) and kept on ice for transport to the laboratory. In the laboratory, the synovium was weighed and digested for 2 h at 37°C under constant rotation with digest media (Hank's Balanced Salt Solution (HBSS) with CaCl (0.44 mg/mL)) added at 5 mL/g tissue and containing 1.5 mg/mL Gibco® collagenase type II (ThermoFisher Scientific). The resulting digest was passed through a course stainless steel strainer with HBSS washes and then centrifuged at 800 g for 20 min. The cell pellet was then resuspended in synoviocyte media and centrifuged at 250 g for 10 min. Following this centrifugation, the cell pellet was resuspended in synoviocyte media (high glucose (4.5 g/L) DMEM medium with 10% fetal bovine serum (FBS), 2mM l‐glutamine, 1 mM sodium pyruvate, 25 mM HEPES, penicillin (100 units/mL), and streptomycin (100 µg/mL)) and passed through a 100 µm nylon cell strainer before centrifugation at 500 g for 5 min. This step was repeated, and the cell pellet was then resuspended with fresh synoviocyte media and a live synoviocyte count was determined using a Cellometer® Auto 2000 and ViaStain™ AOPI Staining Solution (Nexcelom Bioscience LLC, Lawrence, MA). Synoviocytes were frozen in synoviocyte freeze media (80% synoviocyte media, 10% FBS, and 10% dimethyl sulfoxide (DMSO)) in aliquots of 10 × 106 cells/mL in liquid nitrogen until use.

2.3. Chondrocyte isolation

Cartilage was digested for chondrocyte isolation as previously described with modifications. 13 Briefly, cartilage shavings were placed in PBS with penicillin (200 units/mL), and streptomycin (200 µg/mL) and kept on ice for transport to the laboratory. In the laboratory, cartilage shavings were minced into smaller pieces using a sterile #10 scalpel blade and digested overnight (16–18 h) at 37°C under constant rotation with chondrocyte media (Ham's F12 medium with 10% FBS, 25 mM HEPES, ascorbic acid (50 μg/mL), α‐ketoglutarate (30 μg/mL), l‐glutamine (300 μg/mL), penicillin (100 units/mL), and streptomycin (100 µg/ml)) added at 5 mL/g tissue and containing 1.5 mg/mL of Gibco® collagenase type II (ThermoFisher Scientific). The resulting digest was passed through a 100 µm nylon cell strainer and centrifuged at 800 g for 20 min. The cell pellet was then washed twice with fresh chondrocyte media and centrifuged at 500 g for 5 min. Cells were resuspended in chondrocyte media, and a live chondrocyte count was determined using a Cellometer® Auto 2000 and ViaStain™ AOPI Staining Solution. Chondrocytes were frozen in chondrocyte freeze media (80% chondrocyte media, 10% FBS, and 10% DMSO) in aliquots of 10 × 106 cells/mL in liquid nitrogen until use.

2.4. Synovial fibroblast and chondrocyte stimulation with fibronectin fragments

Cryopreserved synoviocytes were thawed and grown out in synoviocyte media to P3 to obtain a more homogenous population of synovial fibroblasts. 14 Cryopreserved chondrocytes were thawed and grown out in chondrocyte media to P1. P3 Synovial fibroblasts and P1 chondrocytes were then seeded in monolayer onto 12‐well tissue culture treated plates at 100,000 cells/well in their respective media and maintained at 5% CO2, 90% humidity, and 37°C. Cells were brought to approximately 90% confluency over 48 h before stimulation with media changes performed every 24 h. Two hours before the start of stimulation, cells for each time point and their corresponding control were washed with PBS and switched to serum‐free media (synoviocyte or chondrocyte media devoid of FBS). Serum free media was utilized to eliminate the potential of a confounding effect from fibronectin in serum interfering with the FN7‐10 binding to cell receptors. 8 Cells were stimulated for 6 h or 18 h with 1 µM purified recombinant 42 kD FN7‐10 diluted in PBS as previously described for human chondrocytes 8 or received no stimulation and remained in serum‐free media for the entire duration of stimulation. Following stimulation, culture media was collected from each well, centrifuged at approximately 2000 g for 5 min, and frozen at −20°C until protein analysis. RNA was extracted from cells from each well for gene expression analysis.

2.5. RNA extraction and quality control

Total cellular RNA was extracted from synovial fibroblasts and chondrocytes using the RNeasy Mini Kit (Qiagen Inc.) according to the manufacturer's instructions with the modification that RNAse‐free water for elution was pre‐warmed in a 37°C water bath and allowed to incubate on the column membrane for 2 min before elution to increase RNA yield. The RNA purity (260/280 > 1.8) and quantity (>20 ng/µL) were evaluated using UV microspectrophotometry (NanoDrop 2000 Spectrophotometer, ThermoFisher Scientific) and RNA was stored at −80°C until gene expression analysis was performed.

2.6. Gene expression analysis

A limited, targeted transcriptomic analyses of 28 genes (Supporting Information S1: Table 1) was performed on a NanoString (NanoString Technologies, Inc.) custom code set of equine reporter probes designed for our laboratory based on previously published work from Reed et al. as well as previously published work on gene expression changes in human and equine OA (Supporting Information S1: Table 1). 8 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 Following the manufacturer's instructions, 100 ng of RNA from each sample was hybridized with the custom codeset using the NanoString nCounter™ GX CodeSet RNA Hybridization Setup instructions in a thermocycler at 65°C with a heated lid at 70°C for 18 h, then set at 4°C until the next step for a maximum of 12 h. After hybridization, samples were placed in NanoString cartridges on the nCounter™ Prep Station for purification. Cartridges were then moved to the nCounter™ Digital Analyzer for fluorescence imaging and data collection, following which RNA counts were analyzed using the nSolver Analysis software. All samples were assessed for quality control, and data was normalized using the geometric mean of counts for the three reference genes beta‐actin (ACTB), hypoxanthine phosphoribosyltransferase 1 (HPRT1) and glyceraldehyde 3‐phosphate dehydrogenase (GAPDH). Normalized mRNA counts were then utilized for statistical comparisons for each gene between treatment groups.

2.7. Protein production analysis

A truncated version (IL‐1β, IL‐4, IL‐6, MCP‐1, TNF‐α, RANTES) of the commercially available equine multiplex assay (MILLIPLEX MAP Equine Cytokine/Chemokine Magnetic Bead Panel, EMD Millipore) and the commercially available MILLIPLEX MAP Human TGF‐β Magnetic Bead 3 Plex Kit (EMD Millipore) previously validated for the horse 16 were used to quantify selected cytokine and chemokine production in response to FN7‐10 stimulation on a MAGPIX R System (Luminex Corp.). All samples were analyzed in duplicate using a 96‐well platform performed per manufacturer's instructions. A minimum bead count of 50 for each cytokine was acquired for data analysis using Milliplex Analyst 5.1 software (Luminex Corporation). For the purpose of statistical analysis, undetectable values were set to zero, and values below the lower limit of quantification (LLOQ) were set to 1/10 of the LLOQ when being compared to values above the LLOQ.

2.8. Statistical analysis

Statistical analysis was performed with Prism 9 software for gene and protein expression. All data were normally distributed and reported as means in table format and as means ± standard deviation for all graphs with individual data points also displayed. Multiple two‐tailed paired t‐tests were used to assess the significance between each stimulation time point and its corresponding unstimulated control with significance set at p < 0.05.

3. RESULTS

3.1. Fibronectin fragment stimulation of equine synovial fibroblasts and chondrocytes results in significant changes in the expression of genes relevant to OA pathogenesis

We hypothesized that FN7‐10 stimulation of equine synovial fibroblasts and chondrocytes would significantly alter the gene expression of many of the genes examined and result in an OA phenotype with gene expression changes similar to those described for human chondrocytes stimulated with FN7‐10. Consistent with our hypothesis, 27 of 28 genes examined were significantly altered in synovial fibroblasts stimulated with FN7‐10 compared to unstimulated controls at either or both time points examined, and 26 of 28 genes examined were significantly altered in chondrocytes stimulated with FN7‐10 compared to unstimulated controls at either or both time points examined (Table 1). In particular, IL‐6, CCL2/MCP1, CCL5/RANTES, CXCL6/GCP‐2, MMP1, MMP13, and MMP3 were markedly upregulated. Interestingly, FN7‐10 stimulation significantly decreased the expression of TGFβ1, TGFβ2, and TGFβ3 isoforms in both synovial fibroblasts and chondrocytes.

Table 1.

Mean gene expression fold change of 28 genes assessed in equine synovial fibroblasts (SF) and chondrocytes (C) stimulated with fibronectin fragments (FN7‐10) for 6 or 18 hours (h) relative to each corresponding serum free timepoint control.

Gene 6 h FN‐f SF 18 h FN‐f SF 6 h FN‐f C 18 h FN‐f C
ACAN 1.57 3.81 −1.5 −2.88
ADAMTS4 23.11 40.03 19.47 17.07
ADAMTS5 5.05 5.46 4.85 2.68
CCL2/MCP1 391.33 132 16.65 6.78
CCL5/RANTES 17.98 42.04 18.96 48.45
CCR2 1.9 2.36 2.6 3.07
COL1A1 −1.06 −1.57 1.09 1.08
COL1A2 −1.08 −1.85 −1.09 −1.51
COL2A1 4.59 4.04 2.04 3.81
COL3A1 1.47 1.26 1.91 2.39
COMP −1.01 1.72 1.06 −1.71
CXCL6 151.64 265.69 31.03 22.55
HAS1 1.93 1.76 5.19 2.24
HAS2 1.4 2.07 1.9 2.06
HAS3 3.19 3.47 3.54 4.7
IL‐1β 1.37 2.61 3.48 7.23
IL‐4 37.16 29.62 21.18 10.8
IL‐6 2890.24 2638.73 103.91 22.68
IL‐10 1.23 1.17 1.96 1.04
MMP‐1 518.56 5123.98 40.66 63.56
MMP‐3 426.81 1786.52 49.08 65.91
MMP‐13 69.88 162.93 47.4 28.18
PIEZO1 1.15 −1.09 1.03 −1.04
PRG4 1.58 3.25 1.26 −1.66
TGF‐β1 −1.3 −1.42 −1.06 −1.63
TGF‐β2 −2.62 −6.42 −2.03 −7.4
TGF‐β3 −5.45 −11.26 −3.12 −4.6
TNF‐α 2.85 1.72 4.24 2.15
Open in a new tab

Note: Significant differences are shown as gray highlighted blocks and defined as p < 0.05.

3.2. Fibronectin fragment stimulation significantly increases gene expression of select inflammatory cytokines in equine synovial fibroblasts and chondrocytes

IL‐6 was one of the genes with the greatest increase in expression on our codeset following FN7‐10 stimulation for both equine synovial fibroblasts and chondrocytes (Table 1). The increase in IL‐6 expression was particularly profound for equine synovial fibroblasts with 2890‐ and 2639‐fold higher levels than the unstimulated controls at the 6 and 18 h timepoints, respectively, compared to 104 and 23 fold increases from controls for equine chondrocytes at the same timepoints (Table 1, Figure 1B,E). Stimulation of equine synovial fibroblasts with FN7‐10 also resulted in significant increases in gene expression of IL‐1β at the 18 h timepoint and of TNF‐α at the 6 h timepoint (Table 1, Figure 1A,C). In contrast, stimulation of equine chondrocytes with FN7‐10 resulted in significant increases in gene expression of both IL‐1β and TNF‐α at both the 6 and 18 h timepoints (Table 1, Figure 1D,F).

Figure 1.

Figure 1

Open in a new tab

Fibronectin fragment stimulation significantly increases gene expression of select inflammatory cytokines in equine synovial fibroblasts and chondrocytes. Comparison of gene expression of inflammatory cytokines IL‐β, IL‐6, and TNF‐α in synovial fibroblasts (A–C) and chondrocytes (D–F) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.000. Created with BioRender.com.

3.3. Fibronectin fragment stimulation significantly increases gene expression of the immunomodulatory cytokines IL‐4 and 1L‐10 in equine synovial fibroblasts and chondrocytes

Significant increases in IL‐4 expression were observed in both equine synovial fibroblasts and chondrocytes following FN7‐10 stimulation at both the 6 and 18 h timepoints (Table 1, Figure 2A,C). Significant increases in IL‐10 gene expression were observed in both equine synovial fibroblasts and chondrocytes following FN7‐10 stimulation at the 6 h timepoint only (Table 1, Figure 2B,D).

Figure 2.

Figure 2

Open in a new tab

Fibronectin fragment stimulation significantly increases gene expression of the immunomodulatory cytokines IL‐4 and 1L‐10 in equine synovial fibroblasts and chondrocytes. Comparison of gene expression of IL‐4 and IL‐10 in synovial fibroblasts (A, B) and chondrocytes (C, D) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001. Created with BioRender.com.

3.4. Fibronectin fragment stimulation significantly increases gene expression of select chemokines in both equine synovial fibroblasts and chondrocytes

Stimulation of equine synovial fibroblasts with FN7‐10 resulted in a significant increase in CCL2/MCP‐1, CCL5/RANTES, and CXCL6/GCP‐2 gene expression at both the 6 and 18 h timepoints (Table 1, Figure 3A–C). Similarly, in chondrocytes, gene expression of CCL2/MCP‐1, CCL5/RANTES, and CXCL6/GCP‐2 was significantly increased at both stimulation timepoints, although generally to a lesser extent than synovial fibroblasts (Table 1, Figure 3D–F).

Figure 3.

Figure 3

Open in a new tab

Fibronectin fragment stimulation significantly increases gene expression of select chemokines in both synovial fibroblasts and chondrocytes. Comparison of gene expression of chemokines CCL2/MCP‐1, CCL5/RANTES, and CXCL6/GCP‐2 in synovial fibroblasts (A–C) and chondrocytes (D–F) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001. Created with BioRender.com.

3.5. Fibronectin fragment stimulation significantly increases gene expression of matrix metalloproteinases (MMPs) in both equine synovial fibroblasts and chondrocytes

FN7‐10 stimulation of both equine synovial fibroblasts and chondrocytes resulted in significantly increased expression of MMP‐1, MMP‐3, and MMP‐13 at both stimulation timepoints (Table 1, Figure 4A–F). Similar to the trends observed for cytokines and chemokines, increases in gene expression were generally more pronounced for synovial fibroblasts compared to chondrocytes.

Figure 4.

Figure 4

Open in a new tab

Fibronectin fragment stimulation significantly increases gene expression of matrix metalloproteinases (MMPs) in both synovial fibroblasts and chondrocytes. Comparison of gene expression of MMP‐1, MMP‐3, and MMP‐13 in synovial fibroblasts (A–C) and chondrocytes (D–F) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001. Created with BioRender.com.

3.6. Fibronectin fragment stimulation significantly increases gene expression of ADAMTS in both equine synovial fibroblasts and chondrocytes

FN7‐10 stimulation of both equine synovial fibroblasts and chondrocytes resulted in significantly increased expression of ADAMTS4 and ADAMTS5 at both stimulation timepoints (Table 1, Figure 5A,B,D,E). The gene for Aggrecan (ACAN) was also significantly upregulated in equine synovial fibroblasts following FN7‐10 stimulation at the 18 h timepoint (Table 1, Figure 5C). In contrast, significantly decreased ACAN expression was observed in equine chondrocytes following FN7‐10 stimulation at both timepoints (Table 1, Figure 5F).

Figure 5.

Figure 5

Open in a new tab

Fibronectin fragment stimulation significantly increases gene expression of ADAMTS in both synovial fibroblasts and chondrocytes. Comparison of gene expression of ADAMTS4, ADAMTS5, and Aggrecan (ACAN) in synovial fibroblasts (A–C) and chondrocytes (D–F) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001. Created with BioRender.com.

3.7. Fibronectin fragment stimulation significantly decreases gene expression of TGF‐β isoforms in both equine synovial fibroblasts and chondrocytes

Stimulation of both equine synovial fibroblasts and chondrocytes with FN7‐10 resulted in significant decreases in TGF‐β1, TGF‐β2, and TGF‐β3 isoform gene expression at both timepoints (Table 1, Figure 6A–F).

Figure 6.

Figure 6

Open in a new tab

Fibronectin fragment stimulation significantly decreases gene expression of TGF‐β isoforms in both equine synovial fibroblasts and chondrocytes. Comparison of gene expression of TGF‐β1, TGF‐β2, and TGF‐β3 isoforms in synovial fibroblasts (A–C) and chondrocytes (D–F) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001. Created with BioRender.com.

3.8. Fibronectin fragment stimulation resulted in significantly increased protein production of IL‐6 in both equine synovial fibroblasts and chondrocytes

Within the relatively short timeframe of this 18 h study, significant changes in protein secretion into the media following FN7‐10 stimulation of equine synovial fibroblasts and chondrocytes were observed for IL‐6 for both cell types and at both timepoints (Figure 7A,B). This is consistent with IL‐6 having the greatest increases in gene expression following FN7‐10 stimulation in this study. The remaining proteins examined (IL‐1β, IL‐4, MCP‐1, TNF‐α, RANTES, TGF‐β1, TGF‐β2, and TGF‐β3) were all either undetectable or below quantifiable levels in media from both unstimulated and stimulated cells and at both time points and therefore did not undergo statistical analyses.

Figure 7.

Figure 7

Open in a new tab

Fibronectin fragment stimulation significantly increases protein production of IL‐6 in both equine synovial fibroblasts and chondrocytes. Comparison of media concentration of IL‐6 from synovial fibroblasts (A) and chondrocytes (B) following stimulation with FN7‐10 for 6 h or 18 h compared to the no stimulation corresponding timepoint control. A paired t‐test was performed between each stimulation timepoint and its corresponding no‐stimulation control. Statistical significance is denoted by *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Created with BioRender.com.

4. DISCUSSION

The objective of this study was to determine if matrikine stimulation of equine synovial fibroblasts and chondrocytes using FN7‐10 would result in an in vitro OA phenotype. While prior studies have investigated the effects of differing types of fibronectin fragments on human and equine chondrocytes, studies to date have not examined the effects on equine synovial fibroblasts. 8 , 12 , 17 , 24 , 25 Furthermore, no studies to date have utilized FN7‐10 specific to α5β1 integrin activation 26 on equine synovial fibroblasts or chondrocytes. Consistent with our hypothesis, FN7‐10 stimulation of equine synovial fibroblasts and chondrocytes resulted in a targeted OA‐like phenotype similar to that previously reported for human chondrocytes following FN7‐10 stimulation. 8 Given the ease of availability of equine cells compared to human cells and the widespread acceptance of the horse as an FDA‐approved model of OA, the results of this study suggest that FN7‐10 stimulation of equine synovial fibroblasts and chondrocytes could prove very useful for in vitro experiments leading up to in vivo preclinical trials.

In the present study, equine synovial fibroblasts and articular chondrocytes in monolayer were stimulated with FN7‐10 for either 6 h or 18 h based on the previous RNAseq study by Reed et al. with FN7‐10 stimulation of human chondrocytes. 8 Within the context of this relatively short time frame, FN7‐10 stimulation significantly impacted the expression of many genes related to OA pathogenesis for both cell types, but with the greatest changes in gene expression noted in synovial fibroblasts. This finding was most remarkable for IL‐6, CCL2/MCP‐1, MMP‐1, and MMP‐3 gene expression suggesting the synovium is an important source of these OA mediators within the joint. It has been widely documented that IL‐6 is present in increased concentrations in the synovial fluid of human OA patients as well as equine OA patients. 27 , 28 , 29 , 30 IL‐6 interacts with its membrane‐bound IL‐6 receptor, which results in downstream activation of the JAK/STAT signaling pathway and ultimately leads to increased production of MMPs and ADAMTS. 27 Consistent with this, we observed that FN7‐10 stimulation of equine synovial fibroblasts resulted in marked upregulation of IL‐6 as well as MMP‐1, MMP‐3, MMP‐13, ADAMTS4 and to a lesser extent ADAMTS5. The same trends were observed for FN7‐10 stimulation of equine chondrocytes, but to a lesser degree, and are also consistent with the trends observed for human chondrocytes stimulated with FN7‐108. Thus, in vitro matrikine stimulation of equine synovial fibroblasts and chondrocytes appears to replicate induction of a critical cluster of molecular pathways associated with OA.

In addition to its effects on MMPs and ADAMTSs, the presence of IL‐6 in areas of inflammation has been shown to cause the recruitment of circulating monocytes through upregulation of CCL2/MCP‐1. 31 , 32 As such, CCL2/MCP‐1 has been shown to be increased in the synovial fluid of both human and equine OA patients compared to non‐OA patients 33 , 34 , 35 and is now being investigated as a biomarker for OA. 34 , 36 Menarim et al. 34 found that synovial fluid CCL2/MCP‐1 concentrations were significantly higher in OA joints compared to healthy joints and that increased CCL2/MCP‐1 concentrations were associated with synovial intimal hyperplasia on histology. 34 Consistent with these findings in both human and equine OA patients, we have shown that gene expression of CCL2/MCP‐1 is markedly upregulated in equine synovial fibroblasts and, to a lesser extent, equine chondrocytes following FN7‐10 stimulation. In addition to CCL2/MCP‐1, the chemokines CCL5/RANTES and CXCL6/GCP‐2 are also key contributors to inflammation and immune cell recruitment in OA and have been previously found to be upregulated following fibronectin fragment stimulation of human chondrocytes consistent with our current results. 8 , 19

Although IL‐4 and IL‐10 have not been a focus of previous in vitro fibronectin fragment stimulation studies, the role of these regulatory cytokines in naturally occurring and experimentally induced OA is under continued investigation. There is increasing evidence that IL‐4 expression is increased in OA synovium compared to normal synovium while IL‐10 expression is in increased in OA cartilage compared to normal cartilage and that these regulatory cytokines are chondroprotective but can have dual anti‐inflammatory and pro‐inflammatory effects. 18 Notably, we have shown a significant increase in IL‐4 gene expression following FN7‐10 stimulation in both equine synovial fibroblasts and chondrocytes at both timepoints as well as a significant increase in IL‐10 gene expression in both cell types at the 6 h timepoint.

Interestingly, we observed a significant decrease in gene expression of the TGF‐β isoforms (TGF‐β1, TGF‐β2, and TGF‐β3). TGF‐β plays an important role in both healthy joints and those with OA, but its role changes as disease pathogenesis progresses. Although the literature conclusions vary on exactly when and how the role of TGF‐β signaling changes in OA pathogenesis, some studies suggest that decreased TGF‐β concentrations may be associated with increased OA susceptibility 37 while increased TGF‐β concentrations have primarily been documented in chronic well‐established clinical OA cases. 23 High levels of TGF‐β result in the activation of the Smad 1/5/8 pathway, resulting in chondrocyte hypertrophy, synovial fibrosis, osteophyte formation, and many other pathological changes to the joint. 23 While a previous study of FN7‐10 stimulation of human chondrocytes found increased expression of TGF‐β3, it found decreased expression of TGF‐β2 consistent with our results. 8

While FN7‐10 stimulation of equine synovial fibroblast and chondrocytes in this study resulted in marked alterations in gene expression of several inflammatory cytokines, chemokines, and matrix degradation genes as described above, more modest and mixed alterations were detected for COL, COMP, PRG4 (lubricin), and HAS genes as shown in Table 1. This could either be due to the matrikine stimulation model itself or due to the relatively short time frame of this study. There is existing evidence in the literature that HAS1 and HAS3 are more likely to be upregulated in the presence of inflammation than HAS2, which was corroborated with our findings.

Several limitations to this study must be acknowledged. A relatively small number of donors (n = 4) were used to obtain synovium and cartilage samples. Additionally, gene and protein expression analyses in this study were targeted and not unbiased. While our NanoString custom codeset was largely based on published literature from FN7‐10 stimulation of human chondrocytes, it does not compare to an unbiased approach such as RNA‐seq. The time frame of this study was limited and only observed the effects of FN7‐10 stimulation up to 18 h in duration. This study did not assess for any prolonged effects of FN7‐10 stimulation on synovial fibroblasts or chondrocytes that may persist after the end of the stimulation period. Lastly, culture media samples used for protein production analysis were stored at −20°C rather than −80°C, which may have affected protein quality over time. Protein degradation may be the reason that only IL‐6 levels in the media were increased following FN7‐10 stimulation compared to unstimulated controls. It is also possible the treatment time of 18 h was not sufficient for the detection of the other examined OA mediators in the media.

In conclusion, the findings of this study support further investigation into the use of this novel in vitro equine matrikine stimulation model of OA. Future studies are needed to examine the effects of FN7‐10 stimulation on equine synovial fibroblasts and chondrocytes in coculture and to compare FN7‐10 stimulation in coculture to traditional equine in vitro IL‐1β and LPS stimulation models of OA. A longer culture duration following FN7‐10 stimulation may also be useful for detecting additional gene and protein expression changes not identified in this current study. Results of future studies will be critical for validating which in vitro stimulation model most reliably mimics OA pathogenesis and will provide an avenue for in vitro testing of OA therapeutics for both veterinary and translational uses.

AUTHOR CONTRIBUTIONS

Rachel Gagliardi acquired, analyzed and interpreted data, and drafted the manuscript. Drew W. Koch assisted in the analysis and interpretation of data and critically revised the manuscript for intellectual content. Richard Loeser and Lauren V. Schnabel contributed to the conception and design of the project, to the interpretation of data, and critically revised the manuscript for intellectual content. Rachel Gagliardi and Lauren V. Schnabel take full responsibility for the integrity of this work from inception to the finished article. All authors have read and approved the final manuscript.

Supporting information

Supporting information.

JOR-43-292-s001.docx (50.2KB, docx)

ACKNOWLEDGMENTS

The authors would like to acknowledge the North Carolina State University Comparative Medicine Institute and the Thurston Arthritis Research Center at the University of North Carolina Chapel Hill for fostering the collaborations for this manuscript. The authors would also like to thank Dr. Kristen M. Messenger for her assistance with the multiplex protein expression assays. Research reported in this publication was supported by a North Carolina State University College of Veterinary Medicine Intramural Research Grant (LVS and RG) and the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under award number R37 AR049003 (RL) as well as the Fund for Orthopedic Research in honor of Gus and Equine athletes (F.O.R.G.E; LVS). Stipend support for RG was provided by the NCSU GAANN Biotech Fellowship; Stipend support for DWK was provided by NIH T32OD011130. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of North Carolina State University, the University of North Carolina at Chapel Hill, or the National Institutes of Health.

Gagliardi R, Koch DW, Loeser R, Schnabel LV. Matrikine stimulation of equine synovial fibroblasts and chondrocytes results in an in vitro osteoarthritis phenotype. J Orthop Res. 2025;43:292‐303. 10.1002/jor.26004

REFERENCES

  • 1. Katz JN, Arant KR, Loeser RF. Diagnosis and treatment of hip and knee osteoarthritis: a review. JAMA. 2021;325(6):568. 10.1001/jama.2020.22171 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Hunter DJ, Schofield D, Callander E. The individual and socioeconomic impact of osteoarthritis. Nat Rev Rheumatol. 2014;10(7):437‐441. 10.1038/nrrheum.2014.44 [DOI] [PubMed] [Google Scholar]
  • 3. Cho Y, Jeong S, Kim H, et al. Disease‐modifying therapeutic strategies in osteoarthritis: current status and future directions. Exp Mol Med. 2021;53(11):1689‐1696. 10.1038/s12276-021-00710-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Li S, Cao P, Chen T, Ding C. Latest insights in disease‐modifying osteoarthritis drugs development. Ther Adv Musculoskelet Dis. 2023;15:1759720X2311698. 10.1177/1759720X231169839 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Johnson CI, Argyle DJ, Clements DN. In vitro models for the study of osteoarthritis. Vet J. 2016;209:40‐49. 10.1016/j.tvjl.2015.07.011 [DOI] [PubMed] [Google Scholar]
  • 6. Palmer JL, Bertone AL. Experimentally‐induced synovitis as a model for acute synovitis in the horse. Equine Vet J. 1994;26(6):492‐495. 10.1111/j.2042-3306.1994.tb04056.x [DOI] [PubMed] [Google Scholar]
  • 7. Colbath AC, Frisbie DD, Dow SW, Kisiday JD, McIlwraith CW, Goodrich LR. Equine models for the investigation of mesenchymal stem cell therapies in orthopaedic disease. Oper Tech Sports Med. 2017;25(1):41‐49. 10.1053/j.otsm.2016.12.007 [DOI] [Google Scholar]
  • 8. Reed KSM, Ulici V, Kim C, Chubinskaya S, Loeser RF, Phanstiel DH. Transcriptional response of human articular chondrocytes treated with fibronectin fragments: an in vitro model of the osteoarthritis phenotype. Osteoarthritis Cartilage. 2021;29(2):235‐247. 10.1016/j.joca.2020.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Xie DL, Meyers R, Homandberg GA. Fibronectin fragments in osteoarthritic synovial fluid. J Rheumatol. 1992;19(9):1448‐1452. [PubMed] [Google Scholar]
  • 10. Goldring MB. Integrin‐dependent recruitment of Src to ROS‐producing endosomes in osteoarthritic cartilage. Sci Signaling. 2023;16(809):eadj9760. 10.1126/scisignal.adj9760 [DOI] [PubMed] [Google Scholar]
  • 11. Hwang HS, Park SJ, Cheon EJ, Lee MH, Kim HA. Fibronectin fragment‐induced expression of matrix metalloproteinases is mediated by MyD88‐dependent TLR‐2 signaling pathway in human chondrocytes. Arthritis Res Ther. 2015;17:320. 10.1186/s13075-015-0833-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Johnson A, Smith R, Saxne T, Hickery M, Heinegård D. Fibronectin fragments cause release and degradation of collagen‐binding molecules from equine explant cultures. Osteoarthritis Cartilage. 2004;12(2):149‐159. 10.1016/j.joca.2003.10.008 [DOI] [PubMed] [Google Scholar]
  • 13. Gilbertie JM, Long JM, Schubert AG, Berglund AK, Schaer TP, Schnabel LV. Pooled platelet‐rich plasma lysate therapy increases synoviocyte proliferation and hyaluronic acid production while protecting chondrocytes from synoviocyte‐derived inflammatory mediators. Front Vet Sci. 2018;5:150. 10.3389/fvets.2018.00150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Rosengren S, Boyle DL, Firestein GS. Acquisition, culture, and phenotyping of synovial fibroblasts. Methods Mol Med. 2007;135:365‐375. 10.1007/978-1-59745-401-8_24 [DOI] [PubMed] [Google Scholar]
  • 15. McCoy AM, Kemper AM, Boyce MK, Brown MP, Trumble TN. Differential gene expression analysis reveals pathways important in early post‐traumatic osteoarthritis in an equine model. BMC Genomics. 2020;21(1):843. 10.1186/s12864-020-07228-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Koch DW, Berglund AK, Messenger KM, Gilbertie JM, Ellis IM, Schnabel LV. Interleukin‐1β in tendon injury enhances reparative gene and protein expression in mesenchymal stem cells. Front Vet Sci. 2022;9:963759. 10.3389/fvets.2022.963759 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Loeser RF, Goldring SR, Scanzello CR, Goldring MB. Osteoarthritis: a disease of the joint as an organ. Arthritis Rheum. 2012;64(6):1697‐1707. 10.1002/art.34453 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Van Helvoort EM, Van Der Heijden E, Van Roon JAG, Eijkelkamp N, Mastbergen SC. The role of Interleukin‐4 and Interleukin‐10 in osteoarthritic joint disease: a systematic narrative review. Cartilage. 2022;13(2):194760352210981. 10.1177/19476035221098167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Scanzello CR. Chemokines and inflammation in osteoarthritis: insights from patients and animal models. J Orthop Res. 2017;35(4):735‐739. 10.1002/jor.23471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Grillet B, Pereira RVS, Van Damme J, Abu El‐Asrar A, Proost P, Opdenakker G. Matrix metalloproteinases in arthritis: towards precision medicine. Nat Rev Rheumatol. 2023;19(6):363‐377. 10.1038/s41584-023-00966-w [DOI] [PubMed] [Google Scholar]
  • 21. Verma P, Dalal K. ADAMTS‐4 and ADAMTS‐5: key enzymes in osteoarthritis. J Cell Biochem. 2011;112(12):3507‐3514. 10.1002/jcb.23298 [DOI] [PubMed] [Google Scholar]
  • 22. Li T, Peng J, Li Q, Shu Y, Zhu P, Hao L. The mechanism and role of ADAMTS protein family in osteoarthritis. Biomolecules. 2022;12(7):959. 10.3390/biom12070959 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Van Der Kraan PM. Differential role of transforming growth factor‐beta in an osteoarthritic or a healthy joint. J Bone Metab. 2018;25(2):65. 10.11005/jbm.2018.25.2.65 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Hwang HS, Lee MH, Kim HA. Fibronectin fragment inhibits xylosyltransferase‐1 expression by regulating Sp1/Sp3‐dependent transcription in articular chondrocytes. Osteoarthritis Cartilage. 2019;27(5):833‐843. 10.1016/j.joca.2019.01.006 [DOI] [PubMed] [Google Scholar]
  • 25. Lu HT, Lu JW, Lee CH, et al. Attenuative effects of platelet‐rich plasma on 30 kda fibronectin fragment‐induced MMP‐13 expression associated with TLR2 signaling in osteoarthritic chondrocytes and synovial fibroblasts. J Clin Med. 2021;10(19):4496. 10.3390/jcm10194496 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Miao MZ, Su QP, Cui Y, et al. Redox‐active endosomes mediate α5β1 integrin signaling and promote chondrocyte matrix metalloproteinase production in osteoarthritis. Sci Signaling. 2023;16(809):eadf8299. 10.1126/scisignal.adf8299 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Akeson G, Malemud C. A role for soluble IL‐6 receptor in osteoarthritis. J Funct Morphol Kinesiology. 2017;2(3):27. 10.3390/jfmk2030027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Mihailova A. Interleukin 6 concentration in synovial fluid of patients with inflammatoryand degenerative arthritis. Curr Rheumatol Rev. 2022;18(3):230‐233. 10.2174/1874471015666220128113319 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Pearson MJ, Herndler‐Brandstetter D, Tariq MA, et al. IL‐6 secretion in osteoarthritis patients is mediated by chondrocyte‐synovial fibroblast cross‐talk and is enhanced by obesity. Sci Rep. 2017;7(1):3451. 10.1038/s41598-017-03759-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Tsuchida AI, Beekhuizen M, Rutgers M, et al. Interleukin‐6 is elevated in synovial fluid of patients with focal cartilage defects and stimulates cartilage matrix production in an in vitro regeneration model. Arthritis Res Ther. 2012;14(6):R262. 10.1186/ar4107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Gschwandtner M, Derler R, Midwood KS. More than just attractive: how CCL2 influences myeloid cell behavior beyond chemotaxis. Front Immunol. 2019;10:2759. 10.3389/fimmu.2019.02759 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Wang XM, Hamza M, Wu TX, Dionne RA. Upregulation of IL‐6, IL‐8 and CCL2 gene expression after acute inflammation: correlation to clinical pain. Pain. 2009;142(3):275‐283. 10.1016/j.pain.2009.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Ni F, Zhang Y, Peng X, Li J. Correlation between osteoarthritis and monocyte chemotactic protein‐1 expression: a meta‐analysis. J Orthop Surg. 2020;15(1):516. 10.1186/s13018-020-02045-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Menarim BC, Gillis KH, Oliver A, et al. Macrophage activation in the synovium of healthy and osteoarthritic equine joints. Front Vet Sci. 2020;7. 10.3389/fvets.2020.568756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Haraden CA, Huebner JL, Hsueh MF, Li YJ, Kraus VB. Synovial fluid biomarkers associated with osteoarthritis severity reflect macrophage and neutrophil related inflammation. Arthritis Res Ther. 2019;21(1):146. 10.1186/s13075-019-1923-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Xu Y, Ke Y, Wang B, Lin J. The role of MCP‐1‐CCR2 ligand‐receptor axis in chondrocyte degradation and disease progress in knee osteoarthritis. Biol Res. 2015;48(1):64. 10.1186/s40659-015-0057-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Shen J, Li S, Chen D. TGF‐β signaling and the development of osteoarthritis. Bone Res. 2014;2(1):14002. 10.1038/boneres.2014.2 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting information.

JOR-43-292-s001.docx (50.2KB, docx)

Articles from Journal of Orthopaedic Research are provided here courtesy of Wiley

RESOURCES