TNIIIA2, The Peptide of Tenascin-C, as a Candidate for Preventing Articular Cartilage Degeneration

Abstract

Objective

TNIIIA2 is a peptide of the extracellular matrix glycoprotein tenascin-C. We evaluated whether intra-articular injection of TNIIIA2 could prevent articular cartilage degeneration without inducing synovitis in an osteoarthritis mice model.

Design

Ten micrograms per milliliter of TNIIIA2 were injected into the knee joint of mice (group II) to evaluate the induction of synovitis. The control group received an injection of phosphate buffered saline (group I). Synovitis was evaluated using synovitis score 2 and 4 weeks after injection. The ligaments of knee joints of mice were transected to make the osteoarthritis model. After transection, 10 µg/mL of TNIIIA2 was injected into the knee joint (group IV). The control group received an injection of phosphate buffered saline after transection (group III). Histologic examinations were made using hematoxylin and eosin and safranin-O staining at 2, 4, 8, and 12 weeks postoperatively. An in vitro study was also performed to determine the mechanism by which TNIIIA2 prevents cartilage degeneration. Human chondrocytes were isolated, cultured, and treated with TNIIIA2. The expressions of various mRNAs, including inflammatory cytokines, and anabolic and catabolic factors for cartilage were compared using real-time polymerase chain reaction.

Results

There were no differences between groups in the study of intra-articular injection of mice (group I vs. group II). In the osteoarthritis model, we found development of osteoarthritis was suppressed in group IV at 4 and 8 weeks. TNIIIA2 upregulated the expressions of tumor necrosis factor-α, matrix metalloproteinase 3, and basic fibroblast growth factor.

Conclusion

We demonstrated that TNIIIA2 could prevent cartilage degeneration without synovitis.

Introduction

Tenascin-C (TN-C) is an extracellular matrix (ECM) glycoprotein. It is a hexameric protein of 1.5 million Da. The amino acid sequence of TN-C subunits includes an assembly domain, epidermal growth factor-like repeats, fibronectin type III-like repeats (FNIII), and a fibrinogen-like globe (FBG).^1,2 The sizes of the TN-C monomers vary with the alternative splicing of the FNIII repeats, which are designated A through D; the FNIII domain is inserted between the conserved repeats 5 and 6 in human TN-C.^3,4 Each of these domains can interact with a different subset of binding partners, including cell surface receptors and other extracellular components. ¹ Multifunctional properties have been identified for TN-C, including effects on cell adhesion, migration, proliferation, survival, and differentiation. However, the contributions of each domain to the function of TN-C are still unclear.

In articular cartilage, TN-C expression decreases during maturation of the chondrocyte, and is almost completely absent in adult articular cartilage. In diseased joints such as those with osteoarthritis (OA), high levels of TN-C reappear in cartilage. ⁵ It was reported that cartilage degeneration of a TN-C OA knockout mice model was more conspicuous than in wild-type mice, ⁶ and intra-articular injection of full-length TN-C prevented articular cartilage degeneration in OA mice models. ⁷ However, it was also reported that only the FBG domain of TN-C could induce joint inflammation in mice. ⁸

TNIIIA2 is the 22-mer peptide of TN-C. It positively regulates β1-integrin–mediated cell adhesion, and proliferation activities have been demonstrated in normal fibroblasts. ⁹ With regard to chondrocyte, β1-integrins play an essential role in the regulation of chondrocyte proliferation. ¹⁰

We hypothesized that TNIIIA2 could prevent cartilage degeneration without inducing synovitis. The objectives of this study were to evaluate the effect of TNIIIA2 on chondrocytes in an in vitro study, the effect of intra-articular injection of TNIIIA2 into mice in an in vivo study and the expression of TNIIIA2 in human OA cartilage.

Methods

TNIIIA2 Synthesis

TNIIIA2 was synthesized using the Boc and Fmoc solid phase strategy, in which cysteine was added to the C terminus of the peptide to increase its activity by dimerization and to facilitate coupling to silica gel beads, as described previously. ⁹ For the photoaffinity cross-linking of a putative cell surface receptor of TNIIIA2, a photosensitive derivative of TNIIIA2, RSTDLPGLKAAT-p-benzoyl-phenylalanine (Bpa)-YTITIRGVK (biotin) C (biotinylated TNIIIA2-Bpa) was synthesized using standard solid-state methodology. These synthetic peptides were purified by reverse-phase high performance liquid chromatography and characterized by mass spectrometry.

Animals

Ninety-four male 8-week-old BALB/c strain mice weighing about 22 g were used and maintained according to guidelines approved by the animal experiment and care committee of our institution. The Institutional Ethics Review Board approved the study protocol (Department of Mie University Medical and Hospital Management, approval number 28-8).

Intra-Articular Injection of TNIIIA2

Mouse knee joints were analyzed microscopically to evaluate the inflammatory effect of TNIIIA2 on cartilage. Mice were anesthetized with an intramuscular injection of sodium pentobarbital (0.05 mg/g body weight). Ten micrograms per milliliter of TNIIIA2 (10 µL) were injected into the knee joint of mice (group II, n = 12) using the trance patella tendon approach. The control group received an injection of phosphate buffered saline (PBS; 10 µL) (group I, n = 12). All mice were randomly divided into groups alternately.

After surgery, all mice could walk freely without any splintage. They were kept in the laboratory animal house at 24°C to 25°C, housed 5 per cage, provided with standard mouse chow and water ad libitum, and maintained under a 12-hour light-dark cycle. Mice were sacrificed by cervical dislocation at 2 weeks (group I, n = 8; group II, n = 8) and 4 weeks (group I, n = 4; group II, n = 4) after injection.

Intra-Articular Injection of TNIIIA2 in the OA Mice Model

The knee joints were analyzed microscopically to evaluate the preventive effect of TNIIIA2 on cartilage in the OA model. All mice were anesthetized with an intramuscular injection of sodium pentobarbital (0.05 mg/g body weight). Both knee joints were exposed following a medial capsular incision, and the anterior cruciate ligament and medial collateral ligament were transected. The articular capsule and skin were closed independently. After the articular capsule was closed, 10 µg/mL of TNIIIA2 was injected into the knee joint (group IV, n = 35). The control group received an injection of PBS (group III, n = 35). Mice were randomly allocated to each group. Mice were sacrificed by cervical dislocation at 2 weeks (group III, n = 5; group IV, n = 5), 4 weeks (group III, n = 10; group IV, n = 10), 8 weeks (group III, n = 10; group IV, n = 10), and 12 weeks (group III, n = 10; group IV, n = 10) postoperatively.

Evaluation

All samples were fixed in 10% formalin at room temperature for 2 days, decalcified with 10% ethylenediaminetetraacetic acid, dehydrated, embedded in paraffin, and sliced coronally at 4 μm. Sections were stained with hematoxylin and eosin (H&E) and safranin-O (Saf-O). All specimens were scored blindly by a single investigator. Synovitis was evaluated using the synovitis score at 2 and 4 weeks after injection. The scoring of the synovial membranes was carried out, according to the 3 synovial membrane features (synovial lining cell layer, stroma cell density, and inflammatory infiltrate), the ranking of alterations being on a scale from none (0), slight (1), moderate (2), to strong (3). Grade 9 is the most severe/high-grade synovitis. ¹¹ We compared the synovitis score of groups I and II at 2 and 4 weeks after injection. Cartilage degeneration was evaluated using the Mankin score at 2, 4, 8, and 12 weeks postoperatively. The Mankin score provides a combined score for cartilage structure (normal [0], surface irregularities [1], pannus and surface irregularities [2], clefts to transitional zone [3], clefts to radial zone [4], clefts to calcified zone [5], complete disorganization [6]), cellular abnormalities (normal [0], diffuse hypercellularity [1], cloning [2], hypocellularity [3]), matrix staining (normal [0], slight reduction [1], moderate reduction [3], severe reduction, no dye noted [4]), and tidemark integrity (intact [0], crossed by blood vessels [1]) graded. Higher scores show increased severity of osteoarthritis. ¹² This score was generated separately from the medial tibial plateau and the medial femoral condyle. Mean scores were reported.

We compared the Mankin scores of groups III and IV at 2, 4, 8, and 12 weeks postoperatively.

Chondrocyte Isolation and Culture

An in vitro study was performed to determine the mechanism of preventing cartilage degeneration with TNIIIA2. Chondrocytes were isolated from human articular cartilage during knee replacements for advanced OA under sterile conditions. Cartilage specimens were obtained from the femoral condyles of 25 patients (Kellgren and Lawrence grade 3, 20 patients; grade 4, 5 patients) who were undergoing total knee joint replacement for the treatment of OA. All patients gave informed consent, and the local ethics committee approved the study (Department of Mie University Medical and Hospital Management, approval number 2891). Cartilage fragments, damaged by fibrillation resulting from OA changes, were removed from the femoral condyles of knee joints with a sharp curette. The cartilage fragments were incubated in 0.8% Pronase solution (Calbiochem, Darmstadt, Germany) and dissolved in Dulbecco’s modified Eagle’s medium/Ham F12 (DMEM/F12) (Gibco, Grand Island, NY, USA) for 30 minutes at 37°C with continuous agitation in an atmosphere of 5% CO₂. After washing with DMEM/F12, the cartilage pieces were incubated with 0.4% collagenase (Roche Diagnostics, Penzberg, Germany) in DMEM/F12 for 90 minutes at 37°C with orbital mixing. The cell suspension was filtered using a 70-μm pore size nylon filter (BD Biosciences, Bedford, MA, USA) to remove tissue debris. The filtrate was centrifuged for 5 minutes at 1200 rpm. The cells were washed 3 times with DMEM/F12 containing 10% fetal bovine serum (FBS) and plated on 96- or 6-well tissue culture plates (Becton Dickinson Labware, Franklin Lakes, NJ, USA) in DMEM/F12 supplemented with 10% FBS, amphotericin B solution 0.25μg/mL (Sigma Chemical Co., St. Louis, MO, USA), kanamycin 110 µg/mL (Gibco), penicillin-streptomycin (penicillin 100 IU/mL, streptomycin100 µg/mL) (Gibco), and 25 μg/mL ascorbic acid (Sigma). Chondrocytes were grown at 37°C in a humidified atmosphere of 5% CO₂ and 95% air.

Proliferation Assay

To evaluate the effect of TNIIIA2 on chondrocyte proliferation, cells were estimated with the proliferation assay. Cell proliferation was determined using CellTiter 96 (Promega KK, Tokyo, Japan) according to the manufacturer’s directions. The chondrocytes were plated at a density of 1.0 × 10⁴ cells/well in 96-well plates. After incubation in DMEM/F12 containing 10% FBS for 24 hours, the cells were untreated or treated with 10 μg/mL of TNIIIA2 and further cultured for 1 to 8 days before the proliferation assay. The untreated samples were administered the same volume of culture solution as the treated samples. CellTiter 96 AQ One Solution Reagent was added to each well of the plate at 20 μL per well and then incubated for another 2 hours. Optical density (OD) could be read directly at 490 nm by the microplate reader. The optical density concentrations were standardized between groups.

RNA Extraction and cDNA Synthesis

After the cells were 80% to 90% confluent on 6-well plates, chondrocytes were untreated or treated with 10 or 100 μg/mL under a serum-free condition with 0.1% bovine serum albumin. Total RNA was isolated using Isogen (Nippon Gene, Toyama, Japan) 12 hours after administration of TNIIIA2. Complementary DNA (cDNA) synthesis was performed by oligo (dT) 15 priming from 1 µg of total RNA using a cDNA synthesis kit (Roche). Quantitative analysis of the cDNA was performed using the Sequence Detection System (Applied Biosystems, Foster City, CA, USA) and TaqMan Universal PCR Master Mix (Roche). The thermal cycling conditions consisted of 2 minutes at 50°C, 10 minutes at 95°C, 40 cycles at 95°C for 15 seconds, and 1 minute at 60°C. Glyceraldehyde-3 phosphate dehydrogenase (GAPDH) was used as the housekeeping gene for internal control.

Real-Time Polymerase Chain Reaction (PCR)

We evaluated the expression of inflammatory cytokines (tumor necrosis factor [TNF]-α, n = 8); anabolic factors (transforming growth factor β [TGFβ], n = 10; tissue inhibitor of metalloproteinase [TIMP]-3, n = 10; basic fibroblast growth factor [bFGF], n = 10; and catabolic factors (a disintegrin and metalloproteinase with thrombospondin motifs [ADAMTS]-4, n = 6; ADAMTS5, n = 6; matrix metalloproteinase [MMP]-3, n = 6; and MMP13, n = 6). TaqMan gene expression assay primer-probe pairs were ordered for detection of TNF-α (assay no. Hs01113624-g1), nuclear factor-κB (NFκB; assay no. Hs00765730-m1), TGFβ (assay no. Hs00998133-m1), TIMP3 (assay no. Hs00165949-m1), bFGF (assay no. Hs00256645-m1), ADAMTS4 (assay no. Hs00192708-m1), ADAMTS5 (assay no.Hs00199841-m1), MMP3 (assay no. Hs00968305-m1), MMP13 (assay no. Hs00233992-m1), and GAPDH (assay no.Hs99999905-ml). Quantitative analysis of the cDNA was performed using the ABI Prism 7000 Sequence Detector System (Applied Biosystems, Foster City, CA, USA) and TaqMan Universal PCR Master Mix (Roche). The thermal cycling conditions consisted of 2 minutes at 50°C, 10 minutes at 95°C, and 40 cycles at 95°C for 15 seconds, and 1 minute at 60°C. The mRNA levels of these cytokines were normalized by GAPDH levels. The levels were compared with TNIIIA2-treated cells and -untreated cells.

Purification of TN-C

TN-C was purified from culture supernatant of the U-251MG glioma cell line by ammonium sulfate precipitation, Sephacryl S-500 gel filtration, Mono Q ion exchange chromatography, and a hydroxyapatite column, as described previously. ¹³

Anti-TNIIIA2 Antibody

Rabbits were immunized with a synthetic peptide containing the active sequence coupled with thyroglobulin. The IgG fraction of rabbit serum was applied to Sepharose beads coupled with the synthetic peptide immunogen. Eluted IgG was used as the anti-TNIIIA2 antibody. ¹⁴

Ant-TN-C Antibody

Rabbit polyclonal anti-TN-C antibody (4F10TT; IBL, Takasaki, Japan), which reacts with all TN-C variants, was used.

Immunolabeling of Human Articular Cartilage

Human cartilage specimens were obtained from the tibial plateau of 3 patients who underwent total knee arthroplasty for the treatment of OA. The OA grade of the 3 patients was Kellgren-Lawrence grade III. Non-OA cartilage specimens were obtained from the tibia plateau of a patient with no history of joint disease. No evidence of macroscopic articular degeneration was noted for patients who underwent TKA for femur neoplasm. Specimens were immediately fixed in 4% paraformaldehyde in 0.1% PBS (pH 7.4) at room temperature for 7 days, decalcified in treated K-CX (Falma, Tokyo, Japan), and embedded in paraffin. The sections were cut at 5-μm thickness and placed on Silane-coated glass-slides (Matsunami, Osaka, Japan). Sections were stained with H&E or Saf-O. We also immunohistochemically analyzed TN-C and TNIIIA2. Briefly, sections were treated with 0.04% Proteinase K (Sigma) for 10 minutes at room temperature to retrieve the antigens. Sections were then incubated with monoclonal anti-TN-C antibody or anti-TNIIIA2 antibody overnight at 4°C. After washing, sections were incubated with peroxidase-conjugated anti-rabbit IgG Fab’ (1:100 dilution; DAKO, Glostrup, Denmark) for 1 hour at 37°C. Finally, the immune reaction was developed with diaminobenzidine/H₂O₂ solution. Immunohistochemical analysis of TNIIIA2 was carried out using the CSAII Biotin-free Tyramide Signal Amplification System (DAKO, Glostrup, Denmark) according to the manufacturer’s protocol.

Statistical Analysis

Statistical significance was determined using the Kruskal-Wallis test and the Mann-Whitney U test. A P value <0.05 was considered significant.

Results

Intra-Articular Injection of Mice

At 2 weeks, the cellularity was increased in the synovium, and the synovial lining cell layer was enlarged in groups I and II. At 4 weeks, the increased cellularity and enlarged lining cell layer was improved in both groups. Low-grade synovitis occurred at 2 weeks but improved at 4 weeks in both groups. There were no significant differences in average synovitis scores between group I and group II at 2 weeks or 4 weeks ( Figs. 1 and 2 ).

Figure 1.

Histologic analysis of synovitis in knee joints of mice after administration of TNIIA2 or phosphate buffered saline (PBS) control. Hematoxylin and eosin staining.

Figure 2.

Both phosphate buffered saline (PBS) and TNIIIA2 induced low-grade synovitis at 2 weeks, but it was improved at 4 weeks in both groups. There was no difference between both groups. The error bars represent standard deviation.

Intra-Articular Injection in the OA Mice Model

The articular cartilage had a smooth surface and was evenly stained with Saf-O in groups III and IV at 2 weeks. At 4 and 8 weeks, OA development was suppressed in group IV compared with in group III. Changed surface structure and proteoglycan loss were noticeable in group III. TNIIIA2 administration protected the articular cartilage from proteoglycan loss in group IV. At 12 weeks, changed surface structure and proteoglycan loss were observed in both groups. In the Mankin score, group III was significantly different compared to grope IV at multiple time point (P = 0.019). At 2 weeks, there were no significant differences between groups, and no group showed development of OA. However, group III was significantly higher than group IV at 4 weeks (P = 0.043). At 8 weeks, group III was significantly higher than group IV (P = 0.029). These results indicated that progressive cartilage damage was seen in group III compared with group IV at 4 and 8 weeks. At 12 weeks, OA development was found in both groups with no significant differences between groups ( Figs. 3 and 4 ).

Figure 3.

Histologic analysis of surgically induced osteoarthritis in knee joints of mice after administration of TNIIA2 or phosphate buffered saline (PBS) control: Hematoxylin and eosin (H&E) and safranin-O staining. Osteoarthritis developed only in group III at 4 and 8 weeks.

Figure 4.

TNIIIA2 inhibits cartilage degeneration of osteoarthritis model mice at 4 and 8 weeks. There was no significant change in both group at 12 weeks. The error bars represent standard deviation.

Proliferation Assay

Compared with control cells not treated with TNIIIA2, the OD of cells treated with 10 μg/mL TNIIIA2 was significantly increased at day 6 (P = 0.005) ( Fig. 5 ).

Figure 5.

Proliferation assay of chondrocyte. TNIIIA2 promotes chondrocyte proliferation at day 6.

Gene Expression of Chondrocytes

Many kinds of gene expressions were measured with real-time quantitative PCR ( Fig. 6 ). A dose of 100 μg/mL of TNIIIA2 upregulated the expression of the inflammatory cytokine TNF-α (P = 0.001), the anabolic factor bFGF (P = 0.037), and the catabolic factor MMP3 (P = 0.009). However, there were no significant differences in any gene expression of cells treated with 10 μg/mL of TNIIIA2.

Figure 6.

The expression of mRNA. TNIIIA2 upregulated the inflammatory cytokine (tumor necrosis factor-α [TNF-α]), the catabolic factor (matrix metalloproteinase-3 [MMP3]), and the anabolic factor for cartilage (basic fibroblast growth factor [bFGF]).

Immunofluorescence of Human Cartilage

In normal cartilage specimens, strong staining was observed with Saf-O. OA cartilage specimens showed degeneration, including an irregular cartilage surface. Saf-O staining was weak in degenerated cartilage. Immunohistochemistry showed expression of both anti-TN-C and anti-TNIIIA2 in degenerated cartilage, but not in normal cartilage ( Fig. 7 ).

Figure 7.

Immunostaining of cartilage specimens. Both anti-tenascin-C (anti-TN-C) and anti-TNIIIA2 staining was observed in the same area of the osteoarthritic cartilage, but not in normal cartilage.

Discussion

TN-C has a cryptic functional site that stimulates cell adhesion to the ECM. TNIIIA2, the 22-mer peptide of FNIII, appears to become functionally exposed by processing TN-C molecules with MMP2. TNIIIA2 regulates β1 integrin-mediated cell adhesion to the ECM by inducing a conformational change in β1 integrins, which is necessary for their functional activation. Furthermore, TNIIIA2 sustains this activated status. ⁹ An in vitro study showed that β1 integrin increases the expression levels of aggrecan and type II collagen, thus promoting chondrocyte proliferation and inhibiting chondrocyte apoptosis. ¹⁰ FGF regulates the proliferation of chondrocytes based on cell adhesion between fibronectin and integrin α5β1. ¹⁵ In the present study, TNIIIA2 upregulated the expression of bFGF on chondrocytes, and TNIIIA2 promoted chondrocyte proliferation. TNIIIA2-induced β1 integrin activation may promote chondrocyte proliferation.

The present study demonstrated that a single injection of full-length TN-C into the knee joint could prevent articular cartilage degeneration in the OA mice model. Furthermore, we observed that labeled full-length TN-C provided via an intra-articular injection remained in the cartilage matrix and synovium 1 week after injection. These findings indicate that TN-C affects the cartilage surface and protects the articular cartilage from proteoglycan depletion. ⁷ In the present study, TNIIIA2 prevented articular cartilage degeneration in the OA mice model. TNIIIA2 may work as a key fragment of TN-C for the preventing the articular cartilage degeneration. It was reported that intra-articular TN-C injection promoted joint inflammation in wild-type mice, and TN-C-null mice showed rapid resolution of acute joint inflammation. ⁶ However, the FBG domain of TN-C was demonstrated to induce joint inflammation in vivo. ⁵ Our previous study showed that full-length TN-C did not cause the severe inflammation that was seen in experiments with TN-C fragments. ¹⁶ In the present study, intra-articular-injection of TNIIIA2, the peptide of the domain of TN-C, did not induce joint inflammation in mice models.

In our previous study, the expression of TN-C was observed at the early phase of cartilage repair and completely disappeared in regenerated mature cartilage in the cartilage defect mice model. TN-C has been expressed at high levels in conjunction with cartilage remodeling such as occurs with degeneration and repair. ⁷ In addition, TN-C has been shown to promote proliferation of cultured chondrocytes. ¹⁷ A review article reported that TN-C was a key molecule in tissue remodeling and its deregulated high expression is causally linked to joint diseases. ¹⁸ In the present study, both anti-TN-C and anti-TNIIIA2 staining were observed on the surface of the OA cartilage, but not on normal cartilage. TN-C and TNIIIA2 were expressed in the same area of the OA cartilage. The present study demonstrated that TNIIIA2 promoted the proliferation of cultured chondrocytes. TNIIIA2 may work as a key fragment of TN-C during cartilage remodeling.

The in vitro study demonstrated that TNIIIA2 upregulates inflammatory cytokines (TNFα), anabolic factors (bFGF), and catabolic factors (MMP13). TNF-α, a major pro-inflammatory cytokine, stimulates the inflammatory and degradative cascades in OA. ¹⁹ In contrast, one report demonstrated that pro-inflammatory responses in adipose tissue are essential for both proper ECM-remodeling and angiogenesis. ²⁰ However, TNFα was upregulated in the current study, and a certain degree of inflammatory cytokines might be necessary for cartilage repair. ²¹

We demonstrated that 10 µg/mL of TNIIIA2 could prevent articular cartilage degeneration without synovitis for 8 weeks. Our hypothesis that TNIIIA2, a key fragment of TN-C, could prevent cartilage degeneration without synovitis was verified. TNIIIA2 could be an important candidate in the prevention of articular cartilage degeneration.