A Novel Cryptic Binding Motif, LRSKSRSFQVSDEQY, in the C-Terminal Fragment of MMP-3/7-Cleaved Osteopontin as a Novel Ligand for α9β1 Integrin Is Involved in the Anti-Type II Collagen Antibody-Induced Arthritis

Osteopontin (OPN) is a multifunctional protein that has been linked to various intractable inflammatory diseases. One way by which OPN induces inflammation is the production of various functional fragments by enzyme cleavage. It has been well appreciated that OPN is cleaved by thrombin, and/or matrix metalloproteinase-3 and -7 (MMP-3/7). Although the function of thrombin-cleaved OPN is well characterized, little is known about the function of MMP-3/7-cleaved OPN. In this study, we found a novel motif, LRSKSRSFQVSDEQY, in the C-terminal fragment of MMP-3/7-cleaved mouse OPN binds to α9β1 integrin. Importantly, this novel motif is involved in the development of anti-type II collagen antibody-induced arthritis (CAIA). This study provides the first in vitro and in vivo evidence that OPN cleavage by MMP-3/7 is an important regulatory mechanism for CAIA.


Ethics statement
Mice were kept under specific pathogen-free conditions, and provided food and water ad libitum. Every effort was made to minimize suffering during injections, and all surgery was performed in humanely sacrificed mice. All animal experiments were performed in accordance with the guidelines of the Bioscience Committee of Hokkaido University and were approved by the Animal Care and Use Committee of Hokkaido University (Approval license No. 13-0131).

Enzyme cleavage
Mouse OPN protein was purified from supernatant of mOPN cDNA-transfected CHO cells as previously described [22], Mouse OPN was cleaved by MMP-3 and -7 according as previous report [13]. Briefly, 10 ng of MMP was used with 200 ng of OPN in equal volume of cleavage buffer (200 mM NaCl, 50 mM Tris-HCl, pH 7.6, 5 mM CaCl2) for 15 min at 37˚C.

Western Blotting analysis
OPN or OPN treated with MMPs were fractionated by SDS-PAGE, and transferred to polyvinylidene difluoride membrane (PerkinElmer, Boston, MA). The filters were then immunoblotted with various anti-OPN antibodies. Immunoreactive proteins were visualized using an ECL detection system (Millipore, Bedford, MA).

Peptide preparation
All peptides used in this study were purchased from Sigma Aldrich Japan (Hokkaido, Japan), and these were synthesized on an automated Syro II peptide synthesizer (MultiSynTech, Witten, Germany), followed by purification with C18 reversed-phase column chromatography. Cysteine residues were introduced at the C-terminal end of each peptide to conjugate with bovine serum albumin (BSA). In case of SVVYGLR, GRGDS, and GDSLAYGLR, cysteine residues were at the Nterminal end for BSA conjugation.

Cell adhesion assay
Ninety-six-well plates were pre-coated with recombinant OPN in the absence or presence of MMP-3/7, or synthesized OPN peptide as shown in the results section at 37˚C for 1 h, followed by treatment with 0.5% BSA in PBS for 1 h at room temperature. Cells in the presence or absence of various antibodies for integrin, CD44 or OPN were suspended in DMEM containing 0.25% BSA and 200 mL of cell suspension (at a cell density of 5610 4 cells/well) were applied to 96-well plates and incubated for 1 h at 37˚C. The medium was removed from the plates and all wells were washed twice. The adherent cells were fixed and stained by 0.5% crystal violet in 20% methanol for 30 min. All wells were rinsed three times with water, and adherent cells were lysed with 20% acetic acid. The resulting supernatant from each well was analyzed by an immunoreader (Bio-rad, Rochmond, CA), and the absorbance at 595 nm was measured to determine the relative number of cells adhering to each well.

Scratch wound assay
Thirty-five millimeter tissue culture dishes were coated with 10 mg/mL of various OPN peptides overnight at 4˚C, followed by treatment with 0.5% BSA in PBS for 1 h at room temperature. a9/NIH cells were suspended with trypsin/EDTA and plated onto coated dishes at a density of 1610 6 cells/35-mm dish. After 2 h, cell cultures were scratched with a single pass of a pipette tip, and photographed. Cells were incubated at 37˚C for 15 h. Cultures were washed twice with PBS, fixed and stained in 0.5% crystal violet in 20% methanol for 30 min, and photographed. Cells that had migrated into wound spaces in six high-powered fields (HPF) per dish were counted. Cell numbers are expressed as the number of migrated cells per six HPF.

Cell proliferation assay
Cell proliferation assay was performed with Cell Counting Kit-8 (Dojindo, Kumamoto, Japan). 96-well plates were coated with 10 mg/mL of various OPN peptides overnight at 4˚C, followed by blocking with 0.5% BSA in PBS for 1 h at room temperature. a9/NIH cells were suspended in DMEM containing 10% FCS and plated in 96-well plates at 2610 4 cells/well, then incubated for 2 h at 37˚C. The medium was replaced with DMEM/0.25% BSA without FCS after washing once with the same medium and incubated for another 13 h. WST-8 (2-(2methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-isulfophen-yl)-2H-tetrazolium, monosodium salt) was then added and absorbance of reduced WST-8 was measured after 2 h the absorbance at 450 nm with a reference wavelength of 620 nm by an immunoreader (Bio-Rad).

Flow cytometry
For a9b1 integrin expression, B16-BL6 cells were blocked by normal goat serum then sequentially incubated with biotinylated anti-mouse a9b1 integrin (55A2C) antibody, followed by streptavidin-phycoerythrin. Biotinylated hamster IgG was used as a negative control. For CD44 expression, a4/NIH cells or B16-BL6 cells after blocking were incubated with anti-mouse CD44 (KM81) antibody, followed by anti-rat IgG-phycoerythrin. Rat IgG2a was used as a negative control. Analyses were performed on FACSCalibur flow cytometer (BD Biosciences).

Isolation of ankle joint synovial tissues
The ankle joint portion of a leg was cut and then the dermal, subcutaneous, tendinous, and muscle tissues were surgically taken off from joints under a dissecting microscope. The remaining soft tissues were surgically separated from bone tissues and were used as synovial tissues.

Real-time PCR
Total RNA from synovial tissues from arthritic joints was extracted using Trizol (Life Technologies) and first-strand cDNA was generated with a first-strand cDNA synthesis kit (GE Healthcare Biosciences). Real-time quantitative PCR was performed using LightCycler FastStart DNA Master SYBR Green I Systems (Roche Diagnostics). The specific primers used were: MMP-3: 59-ATGAAAATGAAGGGTC-39 and 59-ACCAGCTATTGCTCTTC-39; MMP-7: 59-ATGCAGCTCACCCTGTT-39 and 59-TCACAGCGTGTTCCTCTTTCC-39. The expression level of mRNA was calculated using the calibration curve method using LightCycler software version 3. Data were standardized to G3PDH.

Mouse experiments
Arthritis was induced using an arthritogenic 4-clone monoclonal antibody cocktail kit containing clone A20, F10-21, D8-6, and D1-2G, which recognize the epitopes on various species of type II collagen. We purchased this kit (Cat#62200) from IBL, Gunma, Japan, but it is no longer available from IBL. The same kit (Cat# 62100) is available from Chondrex Inc, Redmond, WA. Now, Chondrex Inc developed 5-clone cocktail, clone A20, F10-21, D8-6, D1-2G, and D2-112, in order to induce arthritis effectively compared to 4-clone cocktail. Briefly, 7-week-old female BALB/c mice (Charles River, Yokohama, Japan) were injected intravenously with a mixture of four anti-type II collagen monoclonal antibodies (2 mg each) on day 23, followed by an intraperitoneal injection of 50 mg of LPS (0111:B4) on day 0 as reported previously [15]. Anti-OPN (LRS-EQY) antibody was administered intraperitoneally at doses of 400 mg per mouse on days 24 and 0. The clinical severity of arthritis was graded up to 9 days after LPS administration in each of the four paws on a 0-4 scale. The disease severity was recorded for each limb as follows: 0, normal; 1, focal slight swelling and/or redness in one digit; 2, moderate swelling and erythema of .2 digits; 3, marked swelling and erythema of the limb; and 4, maximal swelling, erythema, deformity, and/or ankylosis. Mice were scored in a double-blind manner.

Histology
Histological assessment of the arthritic joints on day 14 was carried out using sections stained with fast green/Safranin-O or hematoxylin and eosin as described previously [21].

Statistical analysis
Data are presented as means ¡SEM and are representative of at least three independent experiments. The statistical significance of differences between groups was calculated with a Student's t test for single comparisons and one-way ANOVA followed by Student's t test with Bonferroni's correction for multiple comparisons. Differences were considered to be significant when P,0.05 (*) or 0.005 (**). The analysis was performed with the statistical software JMP11 for Windows (SAS Institute Inc).

Results
A C-terminal region of MMP-3/7-cleaved OPN binds to a9b1 integrin OPN can be cleaved by MMP-3/7 between G 151 and L 152 and by thrombin between R 153 and S 154 [13]. The MMP-3/7 and thrombin cleavage sites in OPN are shown Fig. 1A. To confirm mouse OPN can be cleaved by MMP-3 or -7, we performed western blot analysis using antibody against N-terminal or C-terminal sequences of OPN. Then, we detected cleaved fragments of OPN by MMP-7, but hardly detected cleaved form by MMP-3 (Fig. 1B). To determine whether MMPcleaved OPN has cell adhesion ability, we used a B16-BL6 mouse melanoma cell adhesion assay with OPN or OPN treated with MMP-3 or -7. Interestingly, binding effect of B16-BL6 cells to plate coated with MMP-3 as well as MMP-7 is enhanced (Fig. 1C). This result suggests that MMP-3 affects OPN function weakly. Thus, mouse OPN, like human and rat OPN, is a substrate for MMP-7 and MMP-3. Next, we asked whether cleavage site by MMP is involved in cell adhesion. For this experiment, we synthesized 145 GDSLAYG 151 and 152 LRSKSRSFQVSDEQYPDATDE 172 peptides, (referred to as LRS-TDE peptide, hereafter) which are peptides mimicking the C-terminal end in the N-terminal fragment or the N-terminal end in the C-terminal fragment of MMP-cleaved mouse OPN, respectively. We found that LRS-TDE peptide, but not GRSLAYG showed strong cell adhesion to B16-BL6 cells (Fig. 1D).
We then tried to identify the cell surface receptor on B16-BL6 cells binding with LRS-TDE peptide. We hypothesized that receptor(s) should be an integrin because the main OPN receptors are integrins. Integrins require divalent cations for activity [24][25][26]. Therefore, we tested the effect of divalent cations in cell adhesion assay by using EDTA used for depletion of cations. We found that the cell binding was completely inhibited by EDTA, but augmented along with the dose of exogenously added divalent cations ( Fig. 2A), indicating that the receptor on B16-BL6 cells for LRS-TDE peptide is indeed an integrin.
To identify the integrin receptor, the inhibitory effects of various antibodies against integrins were examined in cell adhesion assay. As shown in Fig. 2B, the adhesion of LRS-TDE peptide to B16-BL6 cells was inhibited by specific antibodies against a9b1 integrin and b1 integrin. Surface expression of a9b1 integrin on B16-BL6 cells are analyzed by flow cytometer (Fig. 2C). We therefore concluded that the receptor for LRS-TDE peptide is a9b1 integrin.
To confirm specific binding of LRS-TDE peptide with mouse a9b1 integrin, a cell adhesion assay was performed using NIH3T3 cells expressing a9 integrin (a9/ NIH) or a4 integrin (a4/NIH). SVVYGLR peptide was used as the positive control for binding to a9 integrin and a4 integrin, because the SVVYGLR sequence in OPN can bind both integrins [4,27]. We found that LRS-TDE peptide binds to a9/NIH cells but not a4/NIH (Fig. 2D). Nevertheless, both a9/NIH and a4/NIH cells bind to SVVYGLR, as expected. Previous report indicates that C-terminal domain of OPN interacts with CD44 [28]. Therefore, CD44 is one of candidate receptor on B16-BL6 for the binding of LRS-TDE peptide. Flow cytometric analysis revealed CD44 expression on B16-BL6 cells (Fig. 2E). However, we found the interaction between LRS-TDE peptide and B16-BL6 cells was not inhibited by anti-CD44 (KM81) antibody (Fig. 2F), which is the neutralizing antibody for chemotaxis between C-terminal domain of OPN and CD44 [28]. It is noted that LRS-TDE peptide does not bind to a4/NIH cells, despite expressing CD44 on the  (Fig. 2E). These results indicate that the receptor of LRS-TDE peptide is a9 integrin, but not CD44.

LRS-TDE peptide promotes a9b1 integrin-mediated in vitro wound healing
We tested the effect of LRS-TDE peptide in a scratch wound healing assay. Scratch wound closure after 15 h was quantified by phase-contrast microscopy. Cells on LRS-TDE peptide showed significantly stronger scratch wound closure than cells on BSA, GRGDS, or GDSLAYGLR (Fig. 3A). The study of cell proliferation revealed that LRS-TDE peptide did not affect a9/NIH cell growth (Fig. 3B), indicating that promoted wound closure on LRS-TDE peptide was not caused by cell growth. We then tested the blocking effect of anti-a9b1 integrin antibody in scratch wound closer on LRS-TDE peptide. Anti-a9b1 integrin antibody almost completely canceled scratch wound healing on LRS-TDE peptides (Fig. 3C). We therefore concluded that a novel binding region, LRS-TDE, is involved in a9b1 integrin-mediated cell migration and cell adhesion, as reflected in scratch wound healing assay. The essential requirement of peptide sequence within LRS-TDE peptide for binding to a9b1 integrin To determine whether OPN cleavage by MMP-3/7 is required to bind a9b1 integrin, we synthesized three different peptides as follows. 149 AYG-TDE 172 peptide, which contains the MMP-3/7 and thrombin cleavage sites, and is designed to mimic full length OPN. 152 LRS-TDE 172 peptide starts at 152 L, where the MMP-3/7 cleavage site is, and is designed to have the N-terminal end in the Cterminal fragment of MMP-3/7-cleaved OPN. 154 SKS-TDE 172 peptide starts at 154 S, where the thrombin-cleavage site is, and is designed to have the N-terminal end in the C-terminal fragment of thrombin cleaved OPN. We found that LRS-TDE peptide, but not AYG-TDE and SKS-TDE binds to a9/NIH cells (Fig. 4A), suggesting that MMP-3/7 cleavage is critical for creating the peptide, which is recognized by a9b1 integrin.
We then prepared another three different peptides in which some of peptides were deleted in the C-terminal end of LRS-TDE peptide. We found that 152 LRS-TDE 172 , 152 LRS-PDA 169 , and 152 LRS-EQY 166 peptide bind to a9b1 integrin, but not 152 LRS-VSD 163 peptide (Fig. 4B). We therefore concluded that LRSKSRSFQVSDEQY (LRS-EQY) in the C-terminal fragment of MMP-3/7cleaved OPN is a binding motif for a9b1 integrin. Next, we asked whether the same region of human OPN is involved in the binding to a9 integrin. To answer this question, the homologous 15 aa peptide in human OPN, LRSKSKKFRRPDIQY peptide, was synthesized and cell adhesion assay was performed. Then, we found that LRSKSKKFRRPDIQY peptide does not bind to a9 integrin (Fig. 4C).

Attenuation of collagen antibody-induced arthritis (CAIA) by administration of anti-LRS-EQY antibody
We previously reported that a9b1 integrin is critical for the development of CAIA [21]. We now considered whether the interaction between LRS-EQY motif in MMP-3/7-cleaved OPN and a9b1 integrin is involved in the development of CAIA. The mouse model of inflammatory arthritis was induced by intravenous administration of a mixture of four anti-type II collagen monoclonal antibodies followed by intraperitoneal administration of LPS. We have also shown that upregulated OPN expression is critically involved in the development of this murine arthritis model [15].
We first analyzed MMP-3 and MMP-7 expression in synovial tissues of arthritic joints by real-time polymerase chain reaction (PCR). Gene expression of MMP-3 was significantly up-regulated on day 3 and day 6 after LPS injection. MMP-7 expression was induced on day 6 (Fig. 5A). This result suggests that MMP-cleaved OPN may be induced after 3 days in arthritic joints. Next, we generated an antibody against LRS-EQY peptide, anti-LRS-EQY antibody, which successfully detected MMP-cleaved OPN as well as full length OPN (Fig. 5B). It should be noted that an anti-LRS-EQY antibody can inhibit the adhesion of LRS-TDE peptide to a9b1 integrin (Fig. 2B). Using this antibody, we tested the arthritis blocking effect. Mice were treated twice with an anti-LRS-EQY antibody before the onset of arthritis on days 24 and 0. Joint swelling in control Ig-treated mice became evident on day 2 and reached a peak on day 6, whereas the severity of arthritis was significantly reduced by the anti-LRS-EQY antibody after day 5 (Fig. 5C). Fig. 5D is a representative image of the gross appearances of an arthritic joint treated with an anti-LRS-EQY antibody on day 8. In concert with the clinical score, histological analysis demonstrated marked differences between anti-LRS-EQY antibody-treated and control Ig-treated mice (Fig. 5E). Cartilage destruction was assessed by Safranin-O staining. As shown in Fig. 5E, faint Safranin-O staining was detected in the arthritic joints of control Ig-treated mice on day 14 after LPS injection, indicating that cartilage destruction was present. In contrast, Safranin-O staining of articular cartilage in anti-LRS-EQY antibody-treated mice was similar to that of normal mice (Fig. 5E), indicating that articular cartilage was well preserved in anti-LRS-EQY antibody-treated mice. We also detected remarkable proliferation of synovial cells and pannus formation at the base of the metatarsal bones in control Ig-treated mice (Fig. 5F, arrowhead). Furthermore, an irregular bone surface was apparent in the arthritic joints of control Ig-treated mice, which indicated intensive bone erosion (Fig. 5F, arrow). The appearance of these pathological features of CAIA clearly diminished with administration of an anti-LRS-EQY antibody.

Discussion
Osteopontin can be cleaved by thrombin [9] and MMP-3 (stromelysin-1) or -7 (matrilysin) [13] at positions 153 R/S 154 and 151 G/L 152 , respectively. A cleavage site for MMP-3/7 resides just two amino acid residues towards the N-terminal from the thrombin-cleavage site. Thrombin-cleavage of OPN generates a N-terminal fragment that exposes a cryptic adhesive motif, SVVYGLR, on its C-terminal end, which has been reported to be recognized by both a4 integrin [29] and a9b1 Role of MMP-Cleaved OPN in an Inflammatory Arthritis Murine Model integrin [4]. In contrast, the N-terminal fragment cleaved by MMP-3/7 does not reveal an additional a9b1 integrin binding site [7,13], but not a4 integrin [7,30].
The cell binding motif in the C-terminal fragment of thrombin-or MMP-3/7cleaved OPN has not yet been identified. In this study using B16-BL6 mouse melanoma cells, we found that LRSKSRSFQVSDEQYPDATDE (LRS-TDE) peptide in the C-terminal fragment of OPN cleaved by MMP-3/7 binds to a9b1 integrin (Fig. 2D). A cell adhesion assay using a synthesized peptide showed that MMP-3/7 cleavage of OPN leads to exposure of novel a9b1 integrin-binding motif, LRSKSRSFQVSDEQY (Fig. 4A).
In a previous study, we found that a9b1 integrin is critical in the development of CAIA [21]. In the current study, we used an antibody for LRS-EQY peptide (anti-LRS-EQY) and found that the novel interaction is involved in inflammatory arthritis. Real-time PCR analysis showed that MMP-3 and MMP-7 mRNA in ankle joints were up-regulated on day 3 and day 6 after LPS injection, respectively (Fig. 5A), suggesting that MMP-3/7-cleaved OPN is not generated until at least day 3. In an in vivo experiment, we observed that anti-LRS-EQY antibody significantly protected mice from inflammatory arthritis after 5 days (Fig. 5C). This 5th day, when the antibody is functional, is consistent with the induction of MMP-3/7-cleaved OPN.
We reported that treatment with antibodies for SLAYGLR (M5 Ab) [15] and a9b1 integrin [21] delayed the onset of arthritis as well as attenuating its severity, suggesting that the interaction between thrombin-cleaved OPN and a9b1 integrin is important for the development of arthritis in the early phase. However, treatment with anti-a9b1 integrin antibody reduced joint swelling to a normal state after reaching its peak on day 3 (clinical score of 2), whereas treatment with M5 Ab exacerbated the arthritis gradually after onset, suggesting that perhaps other a9b1 integrin ligands are involved in the late phase of arthritis. Treatment with an anti-LRS-EQY antibody attenuated the severity of arthritis from day 5 after LPS injection. This strongly suggests the critical involvement of the interaction between a9b1 integrin and MMP-3/7-cleaved OPN via a LRS-EQY motif in the development of inflammatory arthritis in the late phase.
OPN undergoes post-translational modification by transglutaminase 2 (TG2), which is a protein cross-linking enzyme, as well as thrombin and MMP-3/7, and OPN is polymerized by TG2. These modifications provide a new function for OPN, which is binding to a9b1 integrin. It is conceivable that polymeric OPN may be involved in inflammatory arthritis since TG2 is up-regulated in inflammatory diseases [31]. The generation of antibody against polymeric OPN to inhibit the interaction between polymeric OPN and a9b1 integrin may be useful. Further research is required to understand how polymeric OPN is involved in inflammatory arthritis.
The LRSKSRSFQVSDEQY novel binding motif is within the mouse OPN sequence. The homologous sequence in human OPN is LRSKSKKFRRPDIQY (underlines indicate mismatch amino acids compared with mouse OPN). We tested whether this human homolog peptide has a binding ability to human a9b1 integrin using a cell adhesion test, but we were not able to conclude that the peptide bound to a9b1 integrin (Fig. 4C). However, we cannot exclude the possibility that the longer homolog peptide directed to the C-terminal in the Cterminal fragment of MMP-cleaved human OPN binds to a9b1 integrin.
In conclusion, we identified a novel interaction between the LRS-EQY motif in the C-terminal fragment of mouse OPN cleaved by MMP-3/7 and a9b1 integrin, and provide evidence for a potential role of this novel binding motif in the regulation of a9b1 integrin-mediated cell adhesion and migration and in the pathogenesis of inflammatory arthritis. Future studies will be directed at understanding the mechanisms by which LRS-EQY motif and its interactions with a9b1 integrins affect other cellular functions.