ARTHRITIS & RHEUMATISM Vol. 65, No. 7, July 2013, pp 1753–1763 DOI 10.1002/art.37932 © 2013, American College of Rheumatology
A Matrix Metalloproteinase 1–Cleavable Composite Peptide Derived From Transforming Growth Factor –Inducible Gene h3 Potently Inhibits Collagen-Induced Arthritis Eon Jeong Nam,1 Jin Hee Kang,1 Shijin Sung,1 Keum Hee Sa,1 Kyung Hoon Kim,1 Jae Seok Seo,1 Jong-Ho Kim,2 Seung Woo Han,1 In San Kim,3 and Young Mo Kang1 Objective. Transforming growth factor  – inducible gene h3 (IG-H3), which is abundantly expressed in rheumatoid synovium, and matrix metalloproteinases (MMP) play important roles in the pathogenesis of rheumatoid arthritis (RA). The aim of this study was to determine the therapeutic efficacy of IG-H3–derived peptides using MMP-1–dependent target tissue delivery in chronic inflammatory arthritis. Methods. Peptides developed from IG-H3 derivatives, including the second and fourth YH peptides, the fourth fas-1 domain, the fourth fas-1 domain truncated for H1 and H2 sequences (dhfas-1), and an MMP-1– cleavable composite peptide (MFK24), were cloned. We confirmed the specificity of MFK24 cleavage by immunoblot analysis after treatment with different proteases. Results. The YH18 peptide in the fourth fas-1 domain of IG-H3 was weakly effective in suppressing arthritis severity in mice with collagen-induced arthritis (CIA). Treatment with higher-dose dhfas-1 (30 mg/kg) showed remarkable efficacy, whereas treatment with a lower dose (10 mg/kg) resulted in only partial improve-
ment. MFK24, a composite peptide consisting of dhfas-1 and RGD peptide linked by MMP-1 substrate, was cleaved specifically by MMP-1. The adhesion and migration of NIH3T3 cells mediated by IG-H3 were inhibited by MFK24 at a low concentration. MFK24 suppressed the adhesion of NIH3T3 cells more efficiently compared with murine dhfas-1 (MFK00) or RGD, either alone or in combination. The therapeutic efficacy of MFK24 in mice with CIA was remarkably enhanced, with consistently reduced expression of inflammatory mediators within joint tissue. Conclusion. This proof-of-concept study showed that an MMP-cleavable composite peptide, based on IG-H3 derivatives, had markedly improved therapeutic efficacy in chronic inflammatory arthritis, implicating a new expandable strategy for enhancement of the efficacy of 2 different active molecules in RA. Rheumatoid arthritis (RA) is a chronic destructive arthritis characterized by proliferating pannus, in which synoviocyte proliferation, increased expression of extracellular matrix (ECM) proteins, immune cell infiltration, and neovascularization are characteristic findings (1–3). Fibroblast-like synoviocytes (FLS) serve as one of the primary players in joint destruction by producing proteolytic enzymes, including matrix metalloproteinases (MMPs) (4), via interactions with abundantly expressed ECM proteins (5–7) and inflammatory cytokines (8). Transforming growth factor –inducible gene h3 (IG-H3) is an ECM protein that is secreted by several cell types, such as FLS, in synovial tissue and is highly expressed in rheumatoid synovium (9,10). The IG-H3 protein, which is composed of 4 homologous fas-1 domains and an Arg-Gly-Asp (RGD) motif at the carboxyl terminal, mediates cellular adhesion, migration, proliferation, differentiation, and survival, thereby mod-
Supported by the Korea Research Foundation (grant KRF2006-311-E00313), the Basic Research Program of the Korea Science and Engineering Foundation (grant R01-2007-000-11155-0), and Kyungpook National University Hospital (Medical Research Institute grant 2007). 1 Eon Jeong Nam, MD, PhD, Jin Hee Kang, PhD, Shijin Sung, MS, Keum Hee Sa, PhD, Kyung Hoon Kim, MD, MS, Jae Seok Seo, MD, MS, Seung Woo Han, MD, PhD, Young Mo Kang, MD, PhD: Kyungpook National University School of Medicine, Daegu, South Korea; 2Jong-Ho Kim, PhD: Kyung Hee University College of Pharmacy, Seoul, South Korea; 3In San Kim, MD, PhD: Kyungpook National University School of Medicine, Daegu, South Korea, and Kyung Hee University College of Pharmacy, Seoul, South Korea. Drs. Nam and J. H. Kang contributed equally to this work. Address correspondence to Young Mo Kang, MD, PhD, Division of Rheumatology, Department of Internal Medicine, Kyungpook National University Hospital, 101 Dongin-2ga, Jung-gu, Daegu 700-721, South Korea. E-mail:
[email protected]. Submitted for publication September 18, 2012; accepted in revised form March 5, 2013. 1753
1754
ulating inflammatory processes and tumor progression (9,11–15). Each fas-1 domain contains 2 highly conserved sequences of ⬃10 amino acids, with H1 and H2 motifs at both ends and one YH18 motif composed of at least 18 amino acids flanking tyrosine and histidine (11,16). The YH18 motif comprises the basic element required for interaction with integrins during regulation of cellular adhesion and migration (11). It is a versatile ligand that binds to different integrins on the cell surface according to cell type, such as ␣v3 on FLS (9) and endothelial cells (12), ␣M2 on monocytes (13), and ␣IIb3 on platelets (15), thereby playing a multifunctional role. Another active component, the RGD motif, plays a central role in cell recognition through diverse integrins. In solution, the RGD sequence acts as a decoy, preventing the adhesion of cells (17). Because integrinmediated cell adhesion regulates cell function and fate, combining the RGD and non-RGD ligands, leading to integrin blocking, may be effective from a therapeutic standpoint. Overexpression of MMPs is a hallmark of RA as well as cancer (18,19). The expression of MMP-1, which is produced by synovial lining cells (20), is increased in synovial tissue during the early stages of RA (21). The active form of MMP-1 is also abundant along with increased expression of MMP-3, which is an activator of proMMP-1 (19,22). Based on these findings, we hypothesized that a composite peptide, which may produce smaller effector molecules by specific cleavage of this key MMP in rheumatoid synovium, may provide a novel therapeutic strategy for RA. In the present study, we attempted to develop IG-H3 derivative–based peptides that suppress the inflammatory processes in chronic inflammatory arthritis. Fragments of IG-H3, including the fourth YH18 motif and the fourth fas-1 domain truncated for H1 and H2 sequences (dhfas-1), were effective in inhibiting cellular functions and suppressing arthritis severity in mice with collagen-induced arthritis (CIA). Because of rapid clearance of the soluble RGD peptide, which has been shown to engage ␣v and 1 integrins, modifications were applied to improve stability, including cyclization and insertion of the RGD peptide into a protein (17). We hypothesized that a composite peptide consisting of dhfas-1 and RGD peptide linked by MMP-1 substrate (designated MFK24) may improve targeted delivery of effector molecules into inflamed joint tissue. We observed that MFK24 was specifically cleaved by MMP-1 in vitro and showed improved therapeutic efficacy in mice with CIA. Therefore, we propose a novel strategy for enhancing the efficacy of 2 different effectors in chronic inflammatory arthritis.
NAM ET AL
MATERIALS AND METHODS Reagents. Bovine type II collagen, Freund’s complete adjuvant (CFA), and Freund’s incomplete adjuvant (IFA) were obtained from Chondrex. Recombinant human MMP-1, MMP-2, and MMP-3 were purchased from Chemicon. Recombinant human cathepsin D, cathepsin L, and cathepsin K were obtained from Calbiochem. Polyclonal antibody was produced by immunizing rabbits with recombinant IG-H3. Antibodies against murine CD3 (Dako), CD31 (clone MEC13.3; BD Biosciences), intercellular adhesion molecule 1 (ICAM-1; clone 166623), RANKL (clone 88227; R&D Systems), and GAPDH (clone 6C5; Abcam), and against human MMP-1 (clone 41-1E5; Chemicon) and histadine tag were used. Secondary antibodies included horseradish peroxidase– conjugated goat anti-rat, rabbit anti-goat, biotinylated rabbit anti-rat, and swine anti-rabbit antibodies (Dako). Peptide preparation. The YH18 control peptide (designated KELANIHGIKLYDEILVS), the second YH18 peptide (designated EALRDLLNNHILKSAMCA), the fourth YH18 peptide (designated KELANILKYHIGDEILVS), and RGD peptide (GGRGDSP) were synthesized by AnyGen. Bacterial expression plasmids for wild-type IG-H3 and the fourth fas-1 domain have been described previously (12). Complementary DNA (cDNA) for the IG-H3 fragments dhfas-1 (human) and murine dhfas-1 (MFK00) encoding amino acids 548–614 were generated by polymerase chain reaction (PCR) and cloned into the Eco RV and Xho I sites of pET-29b(⫹) plasmid (Novagen). In order to produce a composite peptide, MFK24 (MFK00 and RGD motif with MMP-1 substrate [GPQGIAG] cDNA and MFK12 (MFK00 and RGD motif without MMP-1 substrate) cDNA were cloned into the Eco RI, Sal L, and Xho I sites of pET-29b(⫹) plasmid. Endotoxin was removed using Triton X-114 in lysis buffer, followed twice by endotoxin removal using an EndoTrap Blue system (Cambrex). Lipopolysaccharide was tested using a QCL-1000 Chromogenic Endpoint LAL Kit (Cambrex), and the concentration was maintained at 0.2–0.4 EU/ml (⬍0.1 EU/mouse). Cell culture. FLS were isolated by enzymatic dispersion of synovial tissue from patients with RA, as previously described (9), and NIH3T3 cells (CRL-1658), a murine fibroblast line, were obtained from ATCC. Cells were cultured in Dulbecco’s modified Eagle’s medium containing 4.5 gm/liter glucose supplemented with 100 units/ml penicillin, 100 g/ml streptomycin, and 10% fetal calf serum at 37°C in an atmosphere of 5% CO2. FLS between passages 3 and 8 were used for the experiments. Induction and assessment of arthritis. A murine model of CIA was generated as previously described, with minor modifications (23). Briefly, male DBA/1J mice (7–9 weeks of age) were intradermally injected with 100 g of bovine type II collagen emulsified in CFA (1:1 weight/volume) at the base of the tail. Twenty-one days after the primary immunization, the mice were administered a booster injection of 100 g type II collagen emulsified in IFA (1:1 w/v). The mice were treated intraperitoneally with peptides, including YH control, YH18 peptide, MFK00, MFK12, and MFK24, at different concentrations that fit volumes of 100–400 l according to the dosage. Clinical arthritis was scored using a previously described 0–4-point scale, where 0 ⫽ no arthritis, 1 ⫽ 1 inflamed digit, 2 ⫽ 2 inflamed digits, 3 ⫽ ⬎2 digits and foot pad
THERAPEUTIC POTENTIAL OF IG-H3–DERIVED PEPTIDES IN INFLAMMATORY ARTHRITIS
inflammation, and 4 ⫽ all digits. The incidence of arthritic paws was defined as the incidence of inflamed paws with a clinical arthritis score of ⱖ2 among 4 paws. Mice were maintained under specific pathogen–free conditions at the animal facility of Kyungpook National University School of Medicine. All the procedures were performed in accordance with an animal protocol approved by the Kyungpook National University Institutional Animal Care and Use Committee. Adhesion assay. A cell adhesion assay was performed as described previously (9). Briefly, 96-well non–tissue culture plates (Costar) were coated with human or murine recombinant wild-type IG-H3 protein (1 g/well) at 4°C for 16 hours. FLS (3 ⫻ 105/ml) and NIH3T3 cells (5 ⫻ 105/ml) were added to the coated wells and allowed to adhere for 2 hours at 37°C. For the inhibition assay, peptides were preincubated with cells at 37°C for 30 minutes. Adherent cells were lysed by the addition of 0.5% Triton X-100 and detached with 3.75 mM p-nitrophenyl-N-acetyl--D-glucosaminide (hexosaminidase substrate) in citrate buffer. Migration assay. Transwell migration assays were performed as described previously (9). A Transwell plate of 8-m pore size (Costar) was settled in a 24-well culture plate (Costar). The undersurface of the membrane was coated with recombinant human or murine wild-type IG-H3 protein (1 g/well) and incubated at 4°C for 16 hours. FLS (5 ⫻ 105/ml) or NIH3T3 cells (1 ⫻ 106/ml) were added to the upper chamber. For the inhibition assay, modified IG-H3–derived peptides were preincubated with the cells at 37°C for 30 minutes. After allowing migration for 4 hours at 37°C, cells were fixed with 4% paraformaldehyde and stained with crystal violet. The number of migrated cells was determined in 9 randomly selected high-power fields by using light microscopy. Immunohistochemical analysis. The hind paws of the mice were fixed in 10% neutral buffered formalin for 2 days and decalcified in 10% EDTA for 30 days at 4°C. The decalcified paws were then dehydrated in a series of ethanol gradients (70–100%), washed twice with xylene, and embedded in paraffin. Paraffin-embedded tissue sections (3 m thick) were deparaffinized. Antigen retrieval was performed at 60°C overnight in EDTA buffer (pH 9.0) for CD3 or with trypsin digestion for CD31. The sections were treated with 0.3% H2O2 to quench endogenous peroxidase activity. The slides were then overlaid with primary antibodies followed by the biotinylated secondary antibodies and Vectastain Elite ABC reagents (Vector). Histologic scoring of synovial tissue hyperplasia, cartilage destruction, bone erosion, and pannus formation was performed using an arbitrary 3-point scoring system. In vivo optical imaging. Following intravenous administration of Cy5.5-labeled MFK12 or MFK24 (5 mg/kg), fluorescent images were captured using an eXplore Optix system (Advanced Research Technologies) and overlaid onto a white light image of each mouse. The total fluorescence intensity (4 paws of each mouse) was calculated using the region of interest function of Analysis Workstation software (Advanced Research Technologies). Based on the imaging results, the fluorescence intensity in the paws (photons/unit area [mm2]) was evaluated. Semiquantitative reverse transcription PCR (RT-PCR). Total RNA was extracted from whole paws after removal of skin, using an RNeasy extraction kit (Intron). Reverse transcription was performed with 1 `g ı of total RNA, using oligo(dT) and a PrimeScript first-strand cDNA synthesis kit
1755
(Takara Bio). Gene expression was quantified by real-time semiquantitative PCR, which was performed using primers and TaqMan probes and a LightCycler 480 System (Roche) accord-
Figure 1. Inhibition of fibroblast-like synoviocyte (FLS) adhesion and attenuation of arthritis severity by the fourth YH18 peptide of transforming growth factor –inducible gene h3 (IG-H3). A, Inhibitory capability of the second and fourth YH18 peptides on IG-H3– mediated adhesion of FLS. The second and fourth YH18 peptides and the YH control peptide were preincubated with FLS at the indicated concentrations and added to a plate coated with IG-H3 (10 g/ml). Adherent cells were quantified by hexosaminidase assay. Values are the mean ⫾ SEM of 3 independent experiments. ⴱ ⫽ P ⬍ 0.05 versus control, by analysis of variance (ANOVA) with Tukey’s post hoc test. B, Clinical arthritis index in mice with collagen-induced arthritis (CIA) treated with YH control, the second YH18 peptide, or the fourth YH18 peptide. Mice with CIA were intraperitoneally injected with YH control and the second and fourth YH18 peptides (10 mg/kg/day for each peptide) beginning on day 23 after the first immunization. Values are the mean ⫾ SEM. ⴱ ⫽ P ⬍ 0.05 by repeated-measures ANOVA.
1756
NAM ET AL
Figure 2. Effect of the fourth fas-1 domain truncated for H1 and H2 sequences (dhfas-1) and the whole human fourth fas-1 domain on IG-H3–mediated adhesion and migration of FLS, and in vivo efficacy of recombinant murine dhfas-1 in mice with CIA. A and C, Adhesion of FLS. FLS preincubated with fourth fas-1 domain (A) or dhfas-1 (C) were added to IG-H3–coated plates, and adherent cells were quantified by hexosaminidase assay. B and D, Migration of FLS. Transwell migration assays using FLS were performed using 2-compartment Transwells in which the undersurface of the filter was coated with IG-H3. FLS preincubated with fourth fas-1 domain (B) or dhfas-1 (D) were added to the upper chambers. Migrated cells were stained and quantified by counting the number of cells in 9 high-power fields. E and F, In vivo efficacy of recombinant murine dhfas-1 in mice with CIA. MFK00 was preincubated with NIH3T3 cells at different concentrations for 1 hour, and IG-H3–mediated adhesion (E) and migration (F) of the cells were determined. Values are the mean ⫾ SEM of at least 3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001 versus control. See Figure 1 for other definitions.
ing to the manufacturer’s instructions. Relative transcript levels for inflammatory mediators were calculated as target gene:reference gene (18S ribosomal RNA) ratios and normalized to the relative expression level in nonarthritic mice. The primer and probe sequences used are available from the corresponding author upon request. Western blot analysis. Protein was solubilized at 100°C for 5 minutes, fractionated by 15% sodium dodecyl sulfate– polyacrylamide gel electrophoresis, and transferred to the nitrocellulose membranes using a semi-dry transfer system (TransBlot; Bio-Rad). The membranes were incubated with specific primary antibodies followed by horseradish peroxidase– conjugated secondary antibodies. Immunoreactive bands were then detected using ECL Plus chemiluminescence reagents (Amersham). Quantification of protein band intensity was performed using ImageJ software (National Institutes of Health). Statistical analysis. Quantitative data are presented as the mean ⫾ SEM. The difference between the treatment and control groups was analyzed by Student’s t-test or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Repeated-measures ANOVA with Tukey’s post hoc test were performed to compare differences between treatment groups at multiple time points. P values less than 0.05 were
considered significant. Statistical analysis was performed using SPSS 12.0 software.
RESULTS Efficacy of the fourth YH18 peptide for the treatment of arthritis. We previously demonstrated that IG-H3 mediated the adhesion and migration of FLS, which were blocked by both the fourth YH18 peptide and RGD peptide (9). We compared the inhibitory capability of the second and fourth YH18 peptides on IG-H3–mediated adhesion and migration of FLS. The fourth YH18 peptide inhibited these functions at a relatively high concentration (500 M), while the second YH18 peptide did not inhibit these functions at all (Figure 1A). Consistent with the results of the adhesion study, cell migration was also suppressed only by the fourth YH18 peptide (results not shown). To identify the in vivo efficacy of the fourth YH18 peptide in a murine model of CIA, we synthesized
THERAPEUTIC POTENTIAL OF IG-H3–DERIVED PEPTIDES IN INFLAMMATORY ARTHRITIS
1757
Figure 3. MFK00-induced suppression of synovial inflammation in mice with collagen-induced arthritis (CIA). Mice with CIA were intraperitoneally treated with MFK00 and phosphate buffered saline beginning on day 23 after the first immunization. A, Dose-response relationship between MFK00 and the severity of clinical arthritis. B, Incidence of paw involvement over time in mice with CIA. In A and B, ⴱⴱ ⫽ P ⬍ 0.001. C, Immunoblot analysis of intercellular adhesion molecule 1 (ICAM-1) in the joint tissue of control and MFK00-treated mice. GAPDH was used as loading control. The graph shows quantitative measurement of ICAM-1 expression. D, Representative synovial tissue sections from control and MFK00-treated mice showing immunohistochemical staining for CD31, ICAM-1, and CD3 expression. Positivity is indicated by diaminobenzidine precipitation (brown). Bar ⫽ 20 m. E, Semiquantitative measurement of inflammatory mediator transcript levels. To quantify the transcript levels of the indicated genes, semiquantitative real-time polymerase chain reaction was performed using TaqMan probes and a LightCycler480, as described in Materials and Methods. The comparison of transcript levels was performed using 18S ribosomal RNA as the denominator. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001 versus control. Values in A–C and E are the mean ⫾ SEM. IL-1 ⫽ interleukin-1; TNF␣ ⫽ tumor necrosis factor ␣; MMP-1 ⫽ matrix metalloproteinase 1; VCAM-1 ⫽ vascular cell adhesion molecule 1; MCP-1 ⫽ monocyte chemotactic protein 1.
murine YH control, second YH18, and fourth YH18 peptides and treated mice with CIA intraperitoneally (10 mg/kg/day) for 3 weeks beginning on day 23 after the first immunization. On day 44, the clinical arthritis index in the group treated with the fourth YH18 peptide was significantly lower than that in the control group (Figure 1B). Effect of the truncated and whole fourth fas-1 domains on cell adhesion and migration. Because the longer stretches containing the YH18 motif may be more efficient for the inhibition of cellular adhesion and migration (11,24), we sought to determine whether recombinant human fourth fas-1 domain could block the adhesion and migration of FLS. Contrary to our expectations, the fourth fas-1 protein did not inhibit IG-H3–mediated adhesion and migration of FLS (Figures 2A and B). To circumvent the possibility of self-aggregation or structural change induced by the H1 and H2 segments,
we cloned recombinant human dhfas-1 and tested its efficacy in inhibiting IG-H3–mediated adhesion and migration of FLS. Both adhesion and migration were significantly inhibited by dhfas-1 at 10 M (Figures 2C and D), which is substantially lower than the efficient concentration (500 M) of the fourth YH18 peptide. In vivo therapeutic efficacy of recombinant murine dhfas-1 in a CIA model. We next cloned recombinant murine dhfas-1 (designated MFK00), examined its capability to inhibit cellular functions, and evaluated its in vivo therapeutic efficacy. MFK00 significantly inhibited both adhesion and migration of NIH3T3 cells mediated by IG-H3, in a dose-dependent manner (Figures 2E and F). Treatment with MFK00 in mice with CIA resulted in dose-dependent inhibition of arthritis severity, as measured by the clinical arthritis index. In the group treated with 10 mg/kg MFK00, the clinical arthritis index
1758
NAM ET AL
Figure 4. Confirmation of matrix metalloproteinase 1 (MMP-1)–specific cleavage of the composite peptide MFK24. Recombinant murine MFK24 was produced by combining MFK00 (amino acids 548–614), which lacks the H1 and H2 regions of the fourth fas-1 domain (amino acids 502–632), and an RGD motif linked by MMP-1 substrate (GPQGIAG). A, Schematic representation of MFK24 structure and Western blot of purified recombinant MFK24. B, Schematic representation of the pET-29b(⫹) vector containing the MFK24 sequence. C, Cleavage study to confirm MMP-1– specific cleavage of MFK24. MFK24 (3 g/ml) was dissolved in assay buffer (150 mM NaCl, 20 mM Tris, 5 mM CaCl2, pH 7.5) at 37°C for 3 hours and mixed with different concentrations of MMPs or cathepsins (CTPs). Immunoblotting using anti–His-tag and IG-H3 antibodies was performed. D, Cleavage study to determine whether MFK24 was cleaved by MMP-1 secreted from interleukin-1–stimulated FLS. MFK24 was incubated with FLS for 3 hours, and MMP-1 secretion and MFK24 cleavage were confirmed by immunoblotting using anti–MMP-1, anti–IG-H3, and anti–His-tag antibodies. The Western blots shown are representative of those from at least 3 independent experiments. See Figure 1 for other definitions.
showed a dual phase of response in which the index remained similar to that in the higher-dose group until day 35, followed by a sharp increase reaching ⬎70% of that in the control group on day 48. Treatment with a higher dose (30 mg/kg) resulted in remarkable improvement in the clinical arthritis index (72% reduction) compared with that in the control group on day 48, whereas treatment with a lower dose (10 mg/kg) resulted in only partial improvement (Figure 3A). The patterns showing the difference in the incidence of paw involvement were similar in the lower- and higher-dose treatment groups (Figure 3B).
To analyze the mechanism of the in vivo effect of MFK00, inflammatory mediators were measured at both the protein and messenger RNA (mRNA) levels. Quantitative measurement of ICAM-1 expression, which was up-regulated on activated endothelia, was performed using whole extracts of ankle joints and revealed markedly reduced expression after treatment with MFK00 (Figure 3C). Immunohistologic analysis showed that joint destruction and synovial hyperplasia were remarkably inhibited, with mild T cell recruitment and minimal angiogenesis, in the group treated with 30 mg/kg (Figure 3D).
THERAPEUTIC POTENTIAL OF IG-H3–DERIVED PEPTIDES IN INFLAMMATORY ARTHRITIS
The transcript expression of inflammatory mediators including interleukin-1 (IL-1), IL-6, tumor necrosis factor ␣ (TNF␣), MMP-3, vascular cell adhesion molecule 1 (VCAM-1), and monocyte chemotactic protein 1 (MCP-1), as measured by semiquantitative RT-PCR, was significantly reduced in both treatment groups, while that of MMP-1 and RANKL was significantly inhibited only in the group treated with 30 mg/kg (Figure 3E). These data indicated that a higher dose of murine dhfas-1 is required for sufficient efficacy in the treatment of arthritis. Confirmation of MMP-1–specific cleavage of the composite peptide MFK24. Based on the results showing that the dhfas-1 and RGD motifs had therapeutic effects, and that MMPs, especially MMP-1, are abundantly expressed in inflamed synovial tissue (18–21), we designed a composite peptide (designated MFK24) consisting of the dhfas-1 and RGD motifs that is linked by a substrate of MMP-1. We selected the amino acid sequence of MMP-1 substrate (GPQGIAG), which is specifically cleavable by active MMP-1 (25). The molecular weight was 10.7 kd (Figures 4A and B). The integrity and purity of MFK24 were assessed by immunoblotting (Figure 4A). To confirm whether MFK24 was cleaved specifically by MMP-1, we treated MFK24 with different proteases including MMP-1, MMP-2, MMP-3, and cathepsin D, cathepsin L, and cathepsin K. After cleavage by active MMP-1, the expression of an intact composite peptide, detected by both monoclonal anti–His-tagged antibodies and polyclonal anti–IG-H3 antibodies, substantially decreased at MMP-1 concentrations ⬎5 ng/ml. However, MFK24 was not cleaved by the other proteases tested, even at the highest concentrations used for the experiments (Figure 4C). We then tested whether MFK24 was cleaved by secreted MMP-1 from FLS that were stimulated with IL-1 (0.1 ng/ml). Activated FLS produced a significant amount of MMP-1 and cleaved MFK24 efficiently (Figure 4D). Effect of lower concentrations of MFK24 on cellular function and arthritis severity. We assessed the capability of MFK24 to inhibit IG-H3–mediated adhesion and migration of murine fibroblasts. MFK24 significantly inhibited adhesion even at a concentration of 0.5 M (Figure 5A) and inhibited migration at a concentration of 1 M (Figure 5B). We then compared the adhesion-blocking efficiency of MFK24 with that of MFK00 and RGD, alone or in combination. MFK24 suppressed the adhesion of NIH3T3 cells more efficiently than MFK00 and RGD peptide, either alone or in combination (Figure 5C). To evaluate whether MFK12 and MFK24 are
1759
Figure 5. Inhibition of adhesion and migration of NIH3T3 cells by MFK24. A, Quantification of adhesion. NIH3T3 cells were preincubated with different concentrations of MFK24 for 1 hour, and the percentage of adherent cells was determined. B, Quantification of migration by Transwell assay. NIH3T3 cells were incubated with different concentrations of MFK24, and cells were allowed to migrate for 4 hours. After staining with crystal violet, the percentage of migrated cells was determined. C, Quantification of the efficiency of MFK24 and that of MFK00 and RGD, alone or in combination, to block transforming growth factor –inducible gene h3 (IG-H3)–mediated adhesion of murine fibroblasts. NIH3T3 cells were preincubated with MFK24, MFK00, and RGD motif, either alone or in combination, for 1 hour, and the percentage of adherent cells was determined. Values are the mean ⫾ SEM of at least 3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001 versus control; † ⫽ P ⬍ 0.001 versus MFK00 and RGD, alone or in combination.
1760
NAM ET AL
Figure 6. Amelioration of clinical arthritis by an MMP-1–cleavable composite peptide derived from transforming growth factor –inducible gene h3. A, Accumulation of Cy5.5-labeled MFK12, Cy5.5-labeled MFK24, and free Cy5.5 in mice with active CIA, 0.5 and 4 hours after intravenous injection. Representative color-coded near-infrared images and total photon counts in 4 paws are shown. B and C, Clinical arthritis index (B) and incidence of paw involvement (C) in mice with CIA treated with different doses of MFK24. Control groups received either vehicle or MFK12. In A–C, ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001. D, Histologic scoring of arthritis severity (hyperplasia, pannus formation, cartilage destruction, and bone erosion) in control and MFK24-treated mice with CIA. E, Semiquantitative measurement of inflammatory mediator transcript levels in joint tissue, using reverse transcription–polymerase chain reaction. Relative expression was normalized to 18S ribosomal RNA detected within that sample. In D and E, ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001 versus control. F, Immunoblot analyses of ICAM-1 and RANKL expression in joint tissue from MFK24-treated and control mice. Values in A–E are the mean ⫾ SEM; values in F are the mean ⫾ SD of at least 3 independent experiments. See Figure 3 for other definitions.
delivered to inflamed joint tissue, Cy5.5-labeled MFK12 and MFK24 were intravenously injected into mice with active CIA. Compared with MFK12 or free Cy5.5, MFK24 accumulated efficiently in arthritic joints, and the total photon counts in 4 paws remained significantly higher in MFK24-injected mice (Figure 6A). In subsequent in vivo experiments, the therapeu-
tic efficacy of MFK24 was evaluated using a murine model of CIA. Four groups of mice were injected intraperitoneally with different doses of MFK24 (0.1, 1, 10, and 30 mg/kg) from day 23 through day 48; control groups were injected with either vehicle (phosphate buffered saline) or MFK12 (10 mg/kg). Clinical disease severity as measured using the clinical arthritis index
THERAPEUTIC POTENTIAL OF IG-H3–DERIVED PEPTIDES IN INFLAMMATORY ARTHRITIS
(Figure 6B) and the incidence of paw involvement (Figure 6C) were significantly alleviated at all doses. Treatment with even the lowest dose (0.1 mg/kg) showed significant clinical efficacy, which was comparable with that of higher doses (1 mg/kg and 10 mg/kg) and was more effective than MFK12 (10 mg/kg). In contrast, treatment with the highest dose (30 mg/kg) did not completely abolish the development of joint inflammation (Figure 6B). Histologic scoring of synovial tissue revealed significant improvement in hyperplasia, cartilage destruction, bone erosion, and pannus formation in MFK24-treated mice (Figure 6D). We then quantified inflammatory mediators in joint tissue at both the mRNA and protein levels. Transcripts of mediators including RANKL, VCAM-1, TNF␣, MMP-1, MMP-3, IL-1, IL-6, and MCP-1 were significantly reduced by MFK24 (Figure 6E), and immunoblotting also revealed a substantial decrease in the expression of ICAM-1 and RANKL in the treatment groups (Figure 6F). DISCUSSION This study demonstrated the therapeutic potential of IG-H3–derived peptides, including the fourth YH18 peptide, dhfas-1, and a strategically modified dhfas-1–RGD composite peptide incorporating MMP-1– specific substrate (MFK24), in chronic inflammatory arthritis. The fourth YH18 peptide showed weak efficacy in the treatment of mice with CIA, as expected from the results of the in vitro study. Compared with the fourth YH18 peptide, dhfas-1 was more effective in the treatment of murine arthritis. An MMP-1–cleavable composite peptide potently inhibited IG-H3–mediated cellular functions and markedly reduced the severity of arthritis. Treatment with MFK24 showed histologic evidence of arthritis amelioration and reduced the expression of inflammatory mediators. The expression of IG-H3 is abundant within RA synovial tissue, where other ECM proteins such as fibronectin and vitronectin are also present at increased levels (9,26,27). The IG-H3 protein may mediate inflammation by participating in the regulation of adhesion, migration, and differentiation of major subsets of cells in synovial tissue, including FLS and monocytes (9,13). The interaction mediated by IG-H3 involves multiple cell adhesion motifs within the fas-1 domains that can bind to different cell types through distinct integrins (9,11,12). The selection of the integrin may depend on the activation state of the integrins in each cell type. Furthermore, different ECM proteins may be engaged in the interaction with cells using a common
1761
integrin (11), implicating the possible application of IG-H3 derivatives for blocking cellular adhesion on other ECM proteins. MFK24 was designed as a composite molecule that can be efficiently cleaved by MMP-1 into 2 effective moieties, dhfas-1 and RGD. MMP-1, which degrades fibrillar type I collagen and other matrix molecules such as aggrecan and perlecan, plays a pivotal role in the physiologic and pathologic remodeling of ECM (28). Whereas the levels of MMP-1 in normal tissue are usually low, both the activity and expression of MMP-1 are highly up-regulated in rheumatoid synovium (19,28), especially at the cartilage–pannus junction, and in hyperplastic synovial lining cells (29,30). Furthermore, the serum level of MMP-1 is up-regulated in patients with early RA, correlates with disease activity, and predicts radiographic outcome (31). MMP-1, which is produced mainly by chondrocytes and synovial fibroblasts, may be the foremost collagenase that plays a critical role in the invasion of cartilage by pannus tissue (32). In addition, endothelial cells from rheumatoid synovium showed higher MMP-1 activity in response to phorbol myristate acetate (33). Based on the therapeutic potential of dhfas-1 and RGD peptides and elevated MMP-1 expression in inflamed tissue, we adopted a tissue-targeting drug strategy in which up-regulated MMP-1 activity is used to split a composite peptide into 2 active components within the target tissue. MFK24 was composed of dhfas-1 and RGD peptides as the main effectors. Adhesion of synovial fibroblasts was significantly inhibited by dhfas-1 compared with the fourth YH18 motif, but the therapeutic effect of dhfas-1 in mice with CIA was obvious only in the higher-dose treatment group. The difference in inhibitory capability among these IG-H3 derivatives on cellular function has been observed in previous studies, wherein the 18-residue YH peptide showed lower antiangiogenic activity compared with that of the fourth fas-1 domain (12,24). The fourth YH motif, which comprises the minimal fragment required for IG-H3– mediated function (11), may not be structurally complete enough to perform effective blocking activity compared with the longer fragments of the fas-1 domain (24). Moreover, the fourth fas-1 domain suppressed the adhesion and migration of cells mediated by fibronectin and vitronectin (24), which may provide an additional explanation for the lower inhibitory capability of the fourth YH motif. These peptides inhibit the ECM–cell interaction, which provides a potent anoikis resistance factor (34), thus leading to apoptosis. The other effector, RGD peptide, is the cell attachment motif of a large number of adhesive ECM
1762
and blood and cell surface proteins. In rheumatoid synovium, extensive angiogenesis occurs, with up-regulation of ␣v3 integrin, which is a marker as well as a crucial effector for blood vessels undergoing angiogenesis (35,36). The synthetic RGD peptide has been used to inhibit cell adhesion, migration, growth, and differentiation. In addition, the RGD peptide can directly induce apoptosis by triggering procaspase 3 autoprocessing and activation (17). The RGD peptide from MFK24 cleaved by active MMP-1 in inflamed joint tissue may play an additional role, in concert with dhfas-1, in regulating cellular functions. The mechanisms of the much greater therapeutic efficacy of MFK24 compared with that of either dhfas-1 or RGD peptide alone may be explained by distinct structural characteristics. A peptide with low molecular weight, such as RGD peptide, is easily degraded in the circulation and may have a higher degree of redistribution in the body. Systemically administered linear RGD peptide is rapidly cleared from the circulation, especially through the kidney (37). Treatment with RGD in mice with CIA showed weak therapeutic efficacy (Nam EJ, et al: unpublished observations). Furthermore, in comparison with free RGD peptides, RGD peptides chemically coupled to a protein backbone showed a prolonged half-life and higher therapeutic efficacy (38). Considering these results, RGD peptides in MFK24 may be efficiently delivered to inflamed synovial tissue, thereby effectively modulating inflammation. Other explanations may include effective inhibition of the invasiveness of FLS, which is a pathogenic process that leads to joint destruction in RA. The invasive activity of tumor cells involves directional delivery of MMP-1 to the lamellipodia, where focal adhesion through an ECM–integrin interaction occurs (39). MFK24 may bind to integrins on the invasive front of FLS and can be cleaved by MMP-1, thereby leading to the suppression of intracellular signaling by focal adhesion; this possibility needs further study. Because MMPs are responsible for the turnover and degradation of ECM proteins and act on proinflammatory cytokines, chemokines, and other proteins to regulate varied aspects of inflammation and immunity, they have been targeted for therapy in cancer and inflammation (40,41). Whereas short-term treatment with MMP inhibitors demonstrated efficacy in acute inflammation, long-term blockade of MMPs is associated with unacceptable risks of side effects; this may be circumvented by developing highly selective inhibitors (40). Until now, however, strategies to use MMPs for preferential delivery to inflamed tissue have rarely been attempted. Doxorubicin prodrug incorporating
NAM ET AL
an MMP-2 substrate that binds to circulating albumin showed superior efficacy for the treatment of melanoma (42) compared with the parent drug doxorubicin. Recently, synovial endothelium–targeting peptide (CKSTHDRLC) fused with IL-4 by an MMP-cleavable sequence was delivered specifically to human rheumatoid synovium transplanted into SCID mice (43). Therefore, there is a need to develop composite peptide prodrugs that can be cleavable and activated by enzymes that are specifically enriched in inflamed tissue. Diverse strategies for targeted delivery of drugs in inflammatory disorders have been developed. Antibodies and peptide ligands have been used to actively target angiogenic endothelial cells, adhesion molecules, and ECM at sites of inflammation (43–45). In the present study, a proof-of-concept design of the MMPcleavable composite peptide based on IG-H3 derivatives showed markedly improved therapeutic efficacy for chronic inflammatory arthritis. Therefore, the approach provides a new expandable strategy for enhancement of the efficacy of 2 different active molecules for the treatment of RA. Furthermore, this strategy could lead to the convergence with MMP-activatable imaging techniques such as fluorescence resonance energy transfer, thereby enabling tracking of this drug. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Y. M. Kang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. I. S. Kim, Y. M. Kang. Acquisition of data. Nam, J. H. Kang, Sung, Sa, K. H. Kim, Seo, J.-H. Kim, Han. Analysis and interpretation of data. Nam, J. H. Kang, Y. M. Kang.
REFERENCES 1. Tak PP, Bresnihan B. The pathogenesis and prevention of joint damage in rheumatoid arthritis: advances from synovial biopsy and tissue analysis [review]. Arthritis Rheum 2000;43:2619–33. 2. Koch AE. The role of angiogenesis in rheumatoid arthritis: recent developments. Ann Rheum Dis 2000;59:65–71. 3. Huber LC, Distler O, Tarner I, Gay RE, Gay S, Pap T. Synovial fibroblasts: key players in rheumatoid arthritis. Rheumatology (Oxford) 2006;45:669–75. 4. Firestein GS. Invasive fibroblast-like synoviocytes in rheumatoid arthritis: passive responders or transformed aggressors? [review]. Arthritis Rheum 1996;39:1781–90. 5. Barilla ML, Carsons SE. Fibronectin fragments and their role in inflammatory arthritis. Semin Arthritis Rheum 2000;29:252–65. 6. Shiozawa K, Hino K, Shiozawa S. Alternatively spliced EDAcontaining fibronectin in synovial fluid as a predictor of rheumatoid joint destruction. Rheumatology (Oxford) 2001;40:739–42. 7. Yumoto K, Ishijima M, Rittling SR, Tsuji K, Tsuchiya Y, Kon S, et al. Osteopontin deficiency protects joints against destruction in
THERAPEUTIC POTENTIAL OF IG-H3–DERIVED PEPTIDES IN INFLAMMATORY ARTHRITIS
8. 9.
10.
11.
12. 13.
14. 15.
16.
17. 18. 19. 20.
21.
22. 23.
24. 25.
26.
anti-type II collagen antibody-induced arthritis in mice. Proc Natl Acad Sci 2002;99:4556–61. McInnes IB, Schett G. Cytokines in the pathogenesis of rheumatoid arthritis. Nat Rev Immunol 2007;7:429–42. Nam EJ, Sa KH, You DW, Cho JH, Seo JS, Han SW, et al. Up-regulated transforming growth factor –inducible gene h3 in rheumatoid arthritis mediates adhesion and migration of synoviocytes through ␣v3 integrin: regulation by cytokines. Arthritis Rheum 2006;54:2734–44. Kim JE, Kim SJ, Lee BH, Park RW, Kim KS, Kim IS. Identification of motifs for cell adhesion within the repeated domains of transforming growth factor--induced gene, ig-h3. J Biol Chem 2000;275:30907–15. Kim JE, Jeong HW, Nam JO, Lee BH, Choi JY, Park RW, et al. Identification of motifs in the fasciclin domains of the transforming growth factor--induced matrix protein ig-h3 that interact with the ␣v5 integrin. J Biol Chem 2002;277:46159–65. Nam JO, Kim JE, Jeong HW, Lee SJ, Lee BH, Choi JY, et al. Identification of the ␣v3 integrin-interacting motif of ig-h3 and its anti-angiogenic effect. J Biol Chem 2003;278:25909. Kim HJ, Kim IS. Transforming growth factor--induced gene product, as a novel ligand of integrin ␣M2, promotes monocytes adhesion, migration and chemotaxis. Int J Biochem Cell Biol 2008; 40:991–1004. Kim JE, Kim SJ, Jeong HW, Lee BH, Choi JY, Park RW, et al. RGD peptides released from ig-h3, a TGF--induced cell-adhesive molecule, mediate apoptosis. Oncogene 2003;22:2045–53. Kim HJ, Kim PK, Bae SM, Son HN, Thoudam DS, Kim JE, et al. Transforming growth factor--induced protein (TGFBIp/ig-h3) activates platelets and promotes thrombogenesis. Blood 2009;114: 5206–15. Kawamoto T, Noshiro M, Shen M, Nakamasu K, Hashimoto K, Kawashima-Ohya Y, et al. Structural and phylogenetic analyses of RGD-CAP/ig-h3, a fasciclin-like adhesion protein expressed in chick chondrocytes. Biochim Biophys Acta 1998;1395:288–92. Ruoslahti E. RGD and other recognition sequences for integrins. Annu Rev Cell Dev Biol 1996;12:697–715. Malemud CJ. Matrix metalloproteinases (MMPs) in health and disease: an overview. Front Biosci 2006;11:1696–701. Rannou F, Francois M, Corvol MT, Berenbaum F. Cartilage breakdown in rheumatoid arthritis. Joint Bone Spine 2006;73: 29–36. Hembry RM, Bagga MR, Reynolds JJ, Hamblen DL. Immunolocalisation studies on six matrix metalloproteinases and their inhibitors, TIMP-1 and TIMP-2, in synovia from patients with osteoand rheumatoid arthritis. Ann Rheum Dis 1995;54:25–32. Yoshihara Y, Nakamura H, Obata K, Yamada H, Hayakawa T, Fujikawa K, et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59:455–61. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151–60. Brownlie RJ, Myers LK, Wooley PH, Corrigall VM, BodmanSmith MD, Panayi GS, et al. Treatment of murine collageninduced arthritis by the stress protein BiP via interleukin-4– producing regulatory T cells: a novel function for an ancient protein. Arthritis Rheum 2006;54:854–63. Nam JO, Jeong HW, Lee BH, Park RW, Kim IS. Regulation of tumor angiogenesis by fastatin, the fourth FAS1 domain of ig-h3, via ␣v3 integrin. Cancer Res 2005;65:4153–61. Verheijen JH, Nieuwenbroek NM, Beekman B, Hanemaaijer R, Verspaget HW, Ronday HK, et al. Modified proenzymes as artificial substrates for proteolytic enzymes: colorimetric assay of bacterial collagenase and matrix metalloproteinase activity using modified pro-urokinase. Biochem J 1997;323:603–9. Clemmensen I, Holund B, Andersen RB. Fibrin and fibronectin in rheumatoid synovial membrane and rheumatoid synovial fluid. Arthritis Rheum 1983;26:479–85.
1763
27. Nikkari L, Haapasalmi K, Aho H, Torvinen A, Sheppard D, Larjava H, et al. Localization of the ␣v subfamily of integrins and their putative ligands in synovial lining cell layer. J Rheumatol 1995;22:16–23. 28. Pardo A, Selman M. MMP-1: the elder of the family. Int J Biochem Cell Biol 2005;37:283–8. 29. Woolley DE, Crossley MJ, Evanson JM. Collagenase at sites of cartilage erosion in the rheumatoid joint. Arthritis Rheum 1977; 20:1231–9. 30. Okada Y, Gonoji Y, Nakanishi I, Nagase H, Hayakawa T. Immunohistochemical demonstration of collagenase and tissue inhibitor of metalloproteinases (TIMP) in synovial lining cells of rheumatoid synovium. Virchows Arch B Cell Pathol Incl Mol Pathol 1990;59:305–12. 31. Green MJ, Gough AK, Devlin J, Smith J, Astin P, Taylor D, et al. Serum MMP-3 and MMP-1 and progression of joint damage in early rheumatoid arthritis. Rheumatology (Oxford) 2003;42:83–8. 32. Murphy G, Knauper V, Atkinson S, Butler G, English W, Hutton M, et al. Matrix metalloproteinases in arthritic disease. Arthritis Res 2002;4 Suppl 3:S39–49. 33. Jackson CJ, Arkell J, Nguyen M. Rheumatoid synovial endothelial cells secrete decreased levels of tissue inhibitor of MMP (TIMP1). Ann Rheum Dis 1998;57:158–61. 34. Frisch SM, Vuori K, Ruoslahti E, Chan-Hui PY. Control of adhesion-dependent cell survival by focal adhesion kinase. J Cell Biol 1996;134:793–9. 35. Brooks PC, Clark RA, Cheresh DA. Requirement of vascular integrin ␣v3 for angiogenesis. Science 1994;264:569–71. 36. Storgard CM, Stupack DG, Jonczyk A, Goodman SL, Fox RI, Cheresh DA. Decreased angiogenesis and arthritic disease in rabbits treated with an ␣v3 antagonist. J Clin Invest 1999;103: 47–54. 37. Noiri E, Goligorsky MS, Wang GJ, Wang J, Cabahug CJ, Sharma S, et al. Biodistribution and clearance of 99mTc-labeled Arg-GlyAsp (RGD) peptide in rats with ischemic acute renal failure. J Am Soc Nephrol 1996;7:2682–8. 38. Schraa AJ, Kok RJ, Moorlag HE, Bos EJ, Proost JH, Meijer DK, et al. Targeting of RGD-modified proteins to tumor vasculature: a pharmacokinetic and cellular distribution study. Int J Cancer 2002;102:469–75. 39. Garamszegi N, Garamszegi SP, Scully SP. Matrix metalloproteinase-1 contribution to sarcoma cell invasion. J Cell Mol Med 2012;16:1331–41. 40. Hu J, Van den Steen PE, Sang QX, Opdenakker G. Matrix metalloproteinase inhibitors as therapy for inflammatory and vascular diseases. Nat Rev Drug Discov 2007;6:480–98. 41. Ryu JH, Lee A, Chu JU, Koo H, Ko CY, Kim HS, et al. Early diagnosis of arthritis in mice with collagen-induced arthritis, using a fluorogenic matrix metalloproteinase 3–specific polymeric probe. Arthritis Rheum 2011;63:3824–32. 42. Mansour AM, Drevs J, Esser N, Hamada FM, Badary OA, Unger C, et al. A new approach for the treatment of malignant melanoma: enhanced antitumor efficacy of an albumin-binding doxorubicin prodrug that is cleaved by matrix metalloproteinase 2. Cancer Res 2003;63:4062–6. 43. Wythe SE, DiCara D, Taher TE, Finucane CM, Jones R, Bombardieri M, et al. Targeted delivery of cytokine therapy to rheumatoid tissue by a synovial targeting peptide. Ann Rheum Dis 2013;72:129–35. 44. Schwager K, Kaspar M, Bootz F, Marcolongo R, Paresce E, Neri D, et al. Preclinical characterization of DEKAVIL (F8-IL10), a novel clinical-stage immunocytokine which inhibits the progression of collagen-induced arthritis. Arthritis Res Therapy 2009;11: R142. 45. Crielaard BJ, Lammers T, Schiffelers RM, Storm G. Drug targeting systems for inflammatory disease: one for all, all for one. J Control Release 2012;161:225–34.
