Treatment of experimental arthritis by targeting ... - Wiley Online Library

0 downloads 0 Views 502KB Size Report
Feb 7, 2012 - Objective. To show that a new recombinant pro- tein (MT07) obtained by fusing a synovial-homing pep- tide to a neutralizing antibody to C5 can ...
ARTHRITIS & RHEUMATISM Vol. 64, No. 8, August 2012, pp 2559–2567 DOI 10.1002/art.34430 © 2012, American College of Rheumatology

Treatment of Experimental Arthritis by Targeting Synovial Endothelium With a Neutralizing Recombinant Antibody to C5 Paolo Macor,1 Paolo Durigutto,1 Luca De Maso,1 Chiara Garrovo,2 Stefania Biffi,2 Andrea Cortini,1 Fabio Fischetti,1 Daniele Sblattero,3 Costantino Pitzalis,4 Roberto Marzari,1 and Francesco Tedesco1 demonstrated by changes in joint swelling, polymorphonuclear cell counts in synovial washes, release of interleukin-6 and tumor necrosis factor ␣, and tissue damage. A similar therapeutic effect was obtained testing MT07 in a CAIA model. Conclusion. Our findings show that the novel recombinant molecule MT07 has the unique ability to selectively target inflamed joints and to exert local control of the inflammatory process by neutralizing the complement system without interfering with circulating C5 levels. We believe that this approach can be extended to other antiinflammatory drugs currently used to treat patients with rheumatoid arthritis.

Objective. To show that a new recombinant protein (MT07) obtained by fusing a synovial-homing peptide to a neutralizing antibody to C5 can be selectively delivered to inflamed synovium and can effectively control joint inflammation in experimental models of arthritis. Methods. Binding of MT07 to human, rat, and mouse synovial tissue was evaluated in vitro by immunofluorescence, and selective localization in the inflamed joints of rats was documented in vivo using time-domain optical imaging. The antiinflammatory effect of MT07 was tested in a rat model of antigeninduced arthritis (AIA) and in a mouse model of collagen antibody–induced arthritis (CAIA). Results. MT07 was able to bind to samples of inflamed synovium from humans, mice, and rats while failing to recognize uninflamed synovium as well as inflamed mouse lung or rat kidney. In vivo analysis of the biodistribution of MT07 confirmed its preferential homing to inflamed joints, with negligible inhibition of circulating C5 levels. MT07 prevented and resolved established inflammation in a rat model of AIA, as

Rheumatoid arthritis (RA) is an autoimmune disease characterized by chronic inflammation of synovial tissues, infiltration by T cells, B cells, dendritic cells, and macrophages, and production of proinflammatory cytokines and autoantibodies, leading to cartilage destruction and bone erosion and, eventually, joint deformities and disability (1). Substantial progress has been made over the last 15 years in the treatment of RA, with early institution of disease-modifying antirheumatic drugs (DMARDs), followed by initiation of biologic agents that target the molecules and cells involved in disease pathogenesis in patients with an inadequate response to DMARDs. Among the cytokines involved in RA, tumor necrosis factor ␣ (TNF␣) plays a critical role in promoting joint inflammation and damage, and TNF␣ is the target of biologic agents that have been administered in an effort to alter the natural history of the disease in RA patients (2). Current standard therapy, consisting of the combination of methotrexate (MTX) and anti-TNF agents, has resulted in marked amelioration of disease severity and a significant reduction in radiographic damage (3). Several biologic agents that target other

Supported by grants from the Italian Ministry of Education, University, and Research (PRIN MFXE7L_004) and the Fondazione Casali, Trieste, Italy. 1 Paolo Macor, PhD, Paolo Durigutto, BS, Luca De Maso, MLT, Andrea Cortini, PhD, Fabio Fischetti, MD, Roberto Marzari, PhD, Francesco Tedesco, MD: University of Trieste, Trieste, Italy; 2 Chiara Garrovo, PhD, Stefania Biffi, PhD: Cluster in Biomedicine, Trieste, Italy; 3Daniele Sblattero, PhD: University of Eastern Piedmont, Novara, Italy; 4Costantino Pitzalis, MD, PhD, FRCP: Barts and the London School of Medicine and Dentistry at Queen Mary, University of London, London, UK. Dr. Tedesco has received consulting fees, speaking fees, and/or honoraria from Adienne Pharma & Biotech (less than $10,000). Address correspondence to Francesco Tedesco, MD, Department of Life Sciences, University of Trieste, Via Valerio, 28-34127 Trieste, Italy. E-mail: [email protected]. Submitted for publication September 28, 2011; accepted in revised form February 7, 2012. 2559

2560

cytokines or cytokine receptors or that control immune cells, including B and T lymphocytes, are being used or are under investigation in clinical trials and are recommended for use in patients whose RA fails to respond to TNF inhibitors (4–6). The consensus on the use of combination therapies for control of disease progression is that the goal is to neutralize the proinflammatory effects of different mediators (4). However, the biologic agents that are currently available for the management of RA do not cover the full spectrum of inflammatory mediators that are responsible for the onset, persistence, and amplification of the inflammatory process in RA. The biologically active products of the complement system are among the mediators that are not selectively neutralized by these agents. Several studies in humans have revealed the important contribution of complement activated through the classical, alternative, and lectin pathways to chronic synovitis in RA, as suggested by the finding of complement activation products in the synovial fluid and the detection of deposits of complement components in synovial tissue of RA patients (7). The resistance of mice with deficiency in various complement components and related proteins, including C3 (8), factor B (8), C5 (9), C5a receptor (10), and mannosebinding lectin–associated serine protease 1 (MASP-1) and MASP-3 (11), to the development of experimental model of arthritis further supports the role of complement in joint inflammation. Based on these findings, therapeutic strategies have been devised to prevent the development of arthritis or to treat established arthritis by inhibiting complement activation or blocking the interaction of complement activation products with the cell targets (12,13). C5 appears to be an ideal target to control because its activation leads to the release of the potent proinflammatory peptide C5a and to the assembly of the terminal complement complex, which is responsible for tissue damage and inflammation. Inhibition of C5 activation through neutralizing antibodies to C5 (12) or through C5a receptor antagonists that interfere with the interaction between C5a and CD88 (13) has the advantage of controlling the attack phase of complement activation and leaving unimpaired the critical defense function that is mainly associated with the activation of C3. The importance of C5 in the pathogenesis of RA is also highlighted by the observation that variants in the TNF receptor–associated factor/C5 locus on chromosome 9 are associated with rheumatoid arthritis susceptibility and severity (14–16) and that the development of

MACOR ET AL

arthritis in mice is regulated by a locus on chromosome 2 containing the C5 gene (17). A problem that has not yet been solved in the management of RA patients is how to reduce and possibly avoid the side effects, particularly the increased risk of common and opportunistic infections, that may be associated with long-term administration of therapeutic drugs (3). Intraarticular injection of neutralizing antibody to C5 has been used successfully to prevent and to control the development of experimental arthritis (18), but this treatment is not easily applicable to patients. We propose an alternative approach that was taken following the identification of a peptide that binds selectively to synovial microvascular endothelium (19). The aim of the present investigation was to show that a recombinant protein obtained by fusing this peptide to a neutralizing antibody to C5 has a distinctive homing property for inflamed synovium and is effective in controlling joint inflammation in experimental models of arthritis. MATERIALS AND METHODS Animals. Wistar rats were obtained from a colony kept in the animal house at the University of Trieste. Male animals weighing 230–260 gm were used in this study. Male BALB/c mice weighing 20–24 gm were purchased from Charles River Italy and maintained in our university facilities. The in vivo experiments were performed in compliance with the guidelines of the European (86/609/EEC) and Italian (D.L.116/92) laws and were approved by the Italian Ministry of University and Research and by the University Institutional Committee. All treatments were performed with the animals under total anesthesia that was induced with 25 mg/kg of bromoethanol (Avertin; Sigma). Source of tissues. Human synovial tissue was obtained from patients undergoing surgery for posttraumatic injury of a knee ligament who had no evidence of synovitis or from patients undergoing surgery for an aggressive form of RA. The tissue samples were collected after obtaining informed consent from the patients and the approval of the Institutional Ethics Board. Uninflamed synovium tissue from rat and mouse joints was removed from healthy animals, and inflamed synovial tissue was obtained from mice and rats developing arthritis that was induced as described in detail below. Samples of inflamed kidney were obtained from rats developing experimental renal ischemia-reperfusion injury (20,21), and samples of inflamed lung were obtained from mice experiencing ovalbumin-induced acute asthma (22). Production of recombinant proteins. The modified anti-C5 antibody specific for the synovial epithelium (MT07) was generated by cloning the sequence coding for peptide 3.1 (CKSTHDRLC) (19) downstream of the complementary DNA (cDNA) for the anti-C5 minibody MB12/22 (rat version of Mubodina; Adienne Pharma & Biotech) (23). Briefly, a

TARGETED ANTI-C5 ANTIBODY AS ARTHRITIS THERAPY

miniconstruct coding for the SV5-tag and the peptide 3.1 (SV5-3.1) was generated using the following oligos: for SV5pep3.1 forward, 5⬘-CTAGTGGCAAACCAATCCCAAACCCACTGCTGGGCCTGGATAGTACTTGCAAGAGCACCCACGACCGCCTGTGCTAAA-3⬘ and for SV5-pep3.1 reverse, 5⬘-AGCTTTTAGCACAGGCGGTCGTGGGTGCTCTTGCAAGTACTATCCAGGCCCAGCAGTGGGTTTGGGATTGGTTTGCCA-3⬘. The original tag (SV5) from the cDNA of the anti-C5 antibody was substituted with the SV53.1 using specific restriction sites (Hind III and Spe I). The resulting construct was transformed into XL1 Escherichia coli cells, and the clones obtained were analyzed by sequencing. MT07, MB12/22, and Fc-HP fusion protein, which contains the hinge–CH2–CH3 domains of IgG1 fused to a synovial-homing peptide, were then subcloned as the Eco RI–Hind III fragment in the pcDNA3 (Invitrogen) vector. Purified plasmid DNA was transfected with Lipofectamine 2000 (Invitrogen) in Chinese hamster ovary S (CHO-S) cells grown in ProCHO 5 medium (Lonza). The recombinant proteins were purified from cell-conditioned medium loaded on protein A columns and eluted with 0.1M citric acid, pH 3. Fractions containing the recombinant proteins were selected by enzyme-linked immunosorbent assay (ELISA) and checked for purity by sodium dodecyl sulfate–polyacrylamide gel electrophoresis. A schematic diagram of the 3 recombinant molecules is shown in Figure 1A. Generation of antigen-induced arthritis (AIA) in rats. The animals were injected at the base of the tail with 2 intradermal injections (given 1 week apart) consisting of an emulsion containing an equal volume of 150 ␮g of methylated bovine serum albumin (mBSA) in 200 ␮l of sterile saline and Freund’s complete adjuvant (both from Sigma). Fourteen days after the second injection, joint inflammation was induced by intraarticular administration of mBSA (100 ␮g in 100 ␮l of saline) into the right knee joint. Saline was injected into the left knee joint, which served as a control. The development of arthritis was monitored by using calipers to measure the degree of swelling of the right and left knees. At the time the rats were euthanized, the intraarticular cavity was surgically exposed and lavage was performed with 2 ml of saline. The number of polymorphonuclear neutrophils (PMNs) was assessed, and their myeloperoxidase content was measured, as previously reported (18). Part of the anterior capsule of the joint containing synovial tissue was removed, embedded in OCT compound (Miles), snap-frozen in liquid nitrogen, and kept at –80°C until used for the immunofluorescence analysis. The remaining part of the knee joint was fixed for 6 days in 10% buffered formalin, decalcified for 5 days in Decalcifier I (Bio-Optica), and embedded in paraffin. Immunofluorescence analysis. Binding of recombinant proteins to synovium, lung, and kidney samples from patients, rats, and mice was evaluated using frozen sections (7 ␮m) that had been incubated with the proteins (10 ␮g/ml) for 60 minutes at room temperature and treated with biotin-labeled goat anti-rat IgG (Sigma) and fluorescein isothiocyanate (FITC)–labeled streptavidin (Dako). Tissue deposition of C3 was assessed following incubation of frozen tissue sections with goat IgG anti-rat C3 (Cappel/MP Biomedicals) at a 1:200 dilution for 60 minutes at room temperature, followed by FITC-labeled rabbit anti-goat

2561

Figure 1. A, Schematic representation of the recombinant proteins used in the study. MT07 is a recombinant anti-C5 miniantibody fused to a synovial-homing peptide (HP). MB12/22 is a recombinant anti-C5 miniantibody that lacks the synovial-homing peptide. Fc-HP fusion protein consists of the hinge–CH2–CH3 domains of IgG1 fused to a synovial-homing peptide. B, Immunofluorescence analysis of knee joints for deposition of MT07 (top and middle). Shown are representative images from 6 samples of human synovium, 8 of rat synovium, and 4 of mouse synovium obtained from inflamed (arthritic) and uninflamed (control) joints incubated with the recombinant molecule MT07, followed by biotin-labeled goat anti-rat IgG and fluorescein isothiocyanate–labeled streptavidin. Original magnification ⫻ 200. ImageJ software was used to determine the mean fluorescence intensity (MFI) of the images for each experimental group (bottom). Values are the mean ⫾ SD. ⴱ ⫽ P ⬍ 0.01 versus controls.

IgG at a 1:200 dilution (Dako) for an additional 60 minutes at room temperature. A similar approach was used to examine synovial tissue for the presence of C9, using rabbit IgG anti-rat C9 (a kind gift from Prof. P. Morgan, Cardiff University, Cardiff, UK) at a 1:1,000 dilution, followed by FITC-labeled swine anti-rabbit IgG (Dako) at a 1:40 dilution. The fluorescence intensity was evaluated in 10 different areas of synovial tissue, each of which measured 0.07 mm2, that was selected at random. ImageJ image analysis software (National Institutes of Health; online at http://rsbweb.nih.gov/ ij/) was used.

2562

Evaluation of MT07 distribution using time-domain near-infrared optical imaging. Purified recombinant antibodies were labeled with the N-hydroxysuccinimide ester of Cy5.5 (FluoroLink Cy5.5 monofunctional dye; Amersham Biosciences), that binds the free amino groups of amino acids. A freshly prepared solution of dye (0.05 mg/ml) in 0.1M sodium carbonate buffer, pH 9.3, was added to recombinant antibodies (1 mg/ml) in phosphate buffer (1:1 volume/volume). The reagents were incubated with gentle shaking for 1 hour at room temperature. Excess unconjugated dye was removed by overnight dialysis against phosphate buffer at pH 7.4. A small-animal time-domain Optix MX preclinical near-infrared imager (Advanced Research Technologies) equipped with a pulsed laser diode and a time-correlated single-photon detector was used for the in vivo and ex vivo evaluation of labeled MT07 distribution. Two-dimensional scanning regions of interest were selected, and laser power, integration time, and scan step were optimized according to the signal emitted. The data were recorded as temporal point-spread functions, and the images were reconstructed as fluorescence intensity. All in vivo analyses were preceded by native scans of the rats prior to injection of labeled MT07 in order to provide a baseline for later analysis. At the end of the in vivo evaluation, the animals were euthanized to perform the ex vivo analyses. The knees and other organs of interest, such as the liver, kidney, spleen, heart, and lung, were collected, washed in phosphate buffered saline (PBS), and analyzed. Measurement of TNF␣ and interleukin-6 (IL-6) concentrations in synovial tissue. Snap-frozen specimens of synovial tissue were homogenized in PBS and centrifuged at 12,000g for 20 minutes at 4°C. The levels of TNF␣ and IL-6 were measured with ELISA kits (PeproTech), and the data were expressed as picograms of TNF␣ or IL-6 per milligram of total protein, as measured by the Bradford method. Generation of collagen antibody–induced arthritis (CAIA) in mice. Mice were injected with a cocktail of anti–type II collagen monoclonal antibodies (Arthrogen; Chemicon) followed 3 days later by lipopolysaccharide (LPS; from Escherichia coli strain O111:B4) according to the manufacturer’s instructions. Swelling of the knees, ankles, and wrists was measured at different time points with the use of calipers. Fourteen days after injection of the anticollagen antibodies, the mice were euthanized, and the knees, ankles, and wrists joints were removed, fixed for 6 days in 10% buffered formalin, decalcified for 5 days in Decalcifier I, and embedded in paraffin. Histomorphologic evaluation. Sections (6–8 ␮m) of paraffin-embedded synovial tissue obtained from the rats with AIA and mice with CAIA were stained with hematoxylin and eosin and examined by 2 independent observers (PD and FF). Tissue damage was graded using a cumulative score resulting from the evaluation of thickness and hyperplasia of the synovial membrane, leukocyte infiltration in the synovial tissue, microvascular lesions, and cartilage erosions, as previously described (18). Statistical analysis. Results are expressed as the mean ⫾ SD. Data were compared by analysis of variance using post hoc analysis for paired multiple comparisons with Fisher’s corrected t-test. A nonparametric Mann-Whitney test was used to determine the significance of differences between tissue

MACOR ET AL

damage scores in the tested groups. P values less than or equal to 0.01 were considered significant.

RESULTS In vitro characterization of MT07. A recombinant molecule MT07 containing a neutralizing minibody to C5 MB12/22 fused with a peptide targeting the synovial endothelium (19) was developed with the aim of preventing the onset of arthritis through the use of a therapeutic tool preferentially delivered to inflamed joints (Figure 1A). We initially analyzed the ability of MT07 to bind to inflamed and uninflamed synovial tissue from various sources, including human, rat, and mouse joints. The results presented in Figure 1B show strong immunofluorescence staining of sections of RA synovium incubated with MT07, as revealed by FITClabeled anti-human Ig antibodies, whereas binding of the same molecule to uninflamed synovium was undetectable. We observed a similar difference in the interaction of MT07 with inflamed and uninflamed mouse and rat synovial tissue obtained from arthritic and healthy animals, respectively (Figure 1B). The anti-C5 minibody MB12/22, which lacks the homing peptide, and the Fc-HP, which contains the peptide but fails to recognize C5, were both tested for their reactivity with synovium (Figure 1B). Unlike MB12/22, Fc-HP shared with MT07 the ability to bind to inflamed tissue (data not shown), suggesting that MT07 targets inflamed synovium in humans, mice, and rats via the homing peptide. To ascertain if the recombinant protein binds to other inflamed tissues besides the synovium, sections of asthmatic lung and ischemic kidney were incubated with MT07 and the bound molecule was revealed by FITClabeled secondary antibody. Immunofluorescence analysis failed to reveal staining of these inflamed tissues (data not shown). Biodistribution of MT07. Having found that MT07 recognized rat and mouse as well as human synovium, we decided to follow the in vivo distribution of the labeled recombinant molecule injected intravenously into rats with AIA. This animal model of arthritis was selected because it involves a single joint and does not affect the remaining joints, which can be used as uninflamed control joints. MT07 was labeled with nearinfrared dye, and its functional activity was confirmed by analyzing the C5 binding capacity (ELISA) and by analyzing the binding to human inflamed synovium (immunofluorescence) (data not shown). The tissue distribution of labeled MT07 was

TARGETED ANTI-C5 ANTIBODY AS ARTHRITIS THERAPY

Figure 2. In vivo distribution of Cy5.5-labeled MT07. A, Time-domain optical images of the right knee joint of a representative arthritic rat (of 4 rats) injected with Cy5.5-labeled MT07 into the tail vein. The joint distribution of the labeled molecule was assessed at the indicated times over a period of 30 days. Signal was detected in the joint at 5 hours after injection, reached a maximum after 4 days, and persisted for at least 30 days. The distribution pattern was similar in all treated animals. Color bar is shown at the right for comparison. The average fluorescence intensity is indicated by the normalized counts (NC). B, Time course of fluorescence intensity in the right and left knee joints. Results are expressed as the average fluorescence intensity (normalized counts) of a selected area located over the knee joint at the indicated time point. C, Ex vivo analysis of joints and various organs isolated from a representative rat on day 30 after injection of the labeled molecule, as analyzed with eXplore Optix software. All animals showed a similar organ distribution. Results are expressed as the average fluorescence intensity (normalized counts) in the indicated organ.

monitored every few hours during the first 24 hours, every 48–72 hours during the first week, and weekly thereafter until the end of the month, using time-domain optical imaging. As shown in Figure 2A, MT07 started to be seen in measurable amounts in the inflamed knee joint 5 hours after intravenous injection, peaking on day 3–4, and still detectable on day 30 after injection (Figures 2A and B). Ex vivo analysis confirmed the preferential localization of the labeled molecule in the inflamed joint, with minimal deposition in the contralateral knee and in the liver and spleen and undetectable in all other organs and joints examined (Figures 2B and C). MT07 injected into healthy rats did not show selective homing to the joints or several organs examined. This was also the case for MB12/22 injected both into arthritic and control rats. The distribution pattern of Fc-HP was essentially the same as that of MT07 (data not shown).

2563

Prevention of AIA in the rat model. To evaluate the ability of targeted recombinant anti-C5 minibody to prevent inflammation in the rat model of AIA, 300 ␮g of MT07, MB12/22, or Fc-HP was injected intravenously 2 hours after the induction of arthritis (day 0), and the animals were euthanized 3 days later, when arthritis was fully developed. The complement activity in serum samples collected at various time intervals (from 10 minutes to 24 hours) after injection of MT07 or MB12/22 remained unchanged as compared to the preinjection values (data not shown). Conversely, MT07 inhibited the deposition of C9 in synovial tissues obtained from treated rats, but as expected, it did not prevent the binding of C3. Neither MB12/22 nor Fc-HP had an inhibitory effect on C3 or C9 deposition in synovial tissue (data available upon request from the author). We next sought to determine whether neutralization of C5, resulting in inhibition of C5a release and C5b–9 assembly, by MT07 targeted to inflamed synovial tissue was also associated with reduced or absent development of arthritis. Untreated animals that were euthanized 3 days after injection manifested joint swelling, high PMN counts, increased IL-6 and TNF␣ concentrations in synovial washings, and extensive tissue damage, as demonstrated by the degree of synovial thickening, cell infiltration, and cartilage erosion (Figure 3). The extent of the inflammatory process in the joint remained essentially unchanged in rats treated with MB12/22 or Fc-HP, suggesting that neither the soluble anti-C5 minibody nor the homing peptide was able to prevent the onset of AIA (data available upon request from the author). In contrast, a single injection of the targeted anti-C5 recombinant antibody caused a significant reduction in joint swelling, PMN counts, and IL-6 and TNF␣ concentrations (Figure 3). Histologic analysis of the joints obtained from untreated, MB12/22-treated, and Fc-HP–treated rats revealed hyperplasia of synovial lining cells and marked leukocyte infiltration in the synovial tissue. Intravenous injection of MT07 2 hours after AIA induction resulted in a marked decrease in synovial hyperplasia and leukocyte infiltration. Treatment of established AIA in the rat model. To evaluate the beneficial effects of MT07 on established arthritis, the recombinant molecule was injected intravenously 3 days after intraarticular administration of mBSA, which was at the peak of joint inflammation, and 4 days later, the animals were euthanized and examined. As shown in Figure 4, control animals still exhibited a strong inflammatory response 7 days after antigen injection, as documented by joint swelling, in-

2564

Figure 3. Effect of MT07, MB12/22, and Fc-HP fusion protein on the development of antigen-induced arthritis in rats. Three groups of rats were injected intravenously with MT07, MB12/22, or Fc-HP (final concentration 300 ␮g) 2 hours after intraarticular administration of methylated bovine serum albumin to induce arthritis. A fourth group of rats were injected with saline and served as a positive control. Three days later, animals were euthanized and examined for joint swelling, polymorphonuclear neutrophil (PMN) counts in synovial lavage fluids, levels of interleukin-6 (IL-6) and tumor necrosis factor ␣ (TNF␣) in the synovium, and tissue damage score (representing the degree of synovial thickening, cell infiltration, and cartilage erosion). Values are the mean ⫾ SD of 5 rats per group. ⴱ ⫽ P ⬍ 0.01 versus saline-treated arthritic rats.

creased PMN counts, and increased IL-6 and TNF␣ levels. These levels were substantially lower in rats that received a single injection of MT07 (300 ␮g). Immunofluorescence analysis of synovial tissues showed no difference in C3 deposition in control and MT07-treated animals. C9 deposition was restricted to control rats and was not observed in animals treated with MT07 (data available upon request from the author). A significant difference in the histomorphologic changes expressed as a tissue damage score was observed between the inflamed joints of untreated rats presenting with hyperplasia of synovial lining cells and marked leukocyte infiltration in the synovial tissue and the joints of MT07-treated rats with a substantial decrease in the degree of the synovial inflammation (Figure 5).

MACOR ET AL

Treatment of CAIA in the mouse model. To test the therapeutic efficacy of MT07 in a model of polyarthritis that more closely resembles the clinical features of RA in humans, we evaluated the effects of the recombinant molecule in a BALB/c mouse model of polyarthritis induced by injection of anti–type II collagen antibodies on day 0, followed 3 days later, by the injection of LPS. Saline-treated animals manifested slight swelling of the ankles, knees, and wrists around day 3, which reached maximum levels associated with redness of the ankles and knees on days 5–7 and of the wrists on day 12 (Figures 6A–C). The swelling continued to increase until day 14, when the mice were euthanized for histologic analysis. A single intravenous injection of 50 ␮g of MT07 given 24 hours after induction of arthritis was able to significantly reduce swelling and tissue damage (Figures 6A–D) as well as hyperplasia of synovial cells and leukocyte infiltration observed in saline-treated mice (Figures 6E and F). The antiinflammatory effect of the targeted recombinant antibody was already apparent in the first 2–3 days prior to LPS injection, particularly in the ankles and knees, and persisted throughout the entire observation period. Administration of MT07 did not prevent binding of C3 to synovial tissues, but it completely inhibited the deposition of C9 (data available upon request from the author).

Figure 4. Effect of MT07 on established antigen-induced arthritis in rats. Two groups of rats were injected intravenously with MT07 (300 ␮g) or saline 3 days after intraarticular administration of methylated bovine serum albumin to induce arthritis. Seven days later, animals were euthanized and examined for joint swelling, polymorphonuclear neutrophil (PMN) counts in synovial lavage fluids, and levels of interleukin-6 (IL-6) and tumor necrosis factor ␣ (TNF␣) in the synovium. Values are the mean ⫾ SD of 6 rats per group. ⴱ ⫽ P ⬍ 0.01 versus saline-treated arthritic rats.

TARGETED ANTI-C5 ANTIBODY AS ARTHRITIS THERAPY

Figure 5. Photomicrographs of knee joint sections obtained from a representative rat with antigen-induced arthritis treated intravenously with saline (A) or MT07 (B) 3 days after intraarticular administration of methylated bovine serum albumin to induce arthritis. Note the synovial hyperplasia and leukocyte infiltration in the saline-treated rat, as compared with the apparently normal synovium in the MT07treated rat. Original magnification ⫻ 250. A tissue damage score (C) was determined as the degree of synovial thickening, cell infiltration, and cartilage erosion. Values are the mean ⫾ SD of 6 rats per group. ⴱ ⫽ P ⬍ 0.01 versus saline-treated arthritic rats.

2565

(26). However, neither the liposomes used as vehicle for the antiinflammatory drugs nor the CR2-fH fusion protein exhibit preferential binding to synovial tissue, whereas the peptide of the fusion protein MT07 used in the present study does exhibit such preferential binding. This peptide, which was isolated by in vivo selection from a phage peptide library, was found to target the synovial endothelium in samples from RA patients or osteoarthritis patients that had been transplanted into SCID mice, as well as to be tissue-specific (19). We now show that this peptide maintains the ability to home to the synovium even when fused to the anti-C5 minibody. An important feature of the recombinant molecule is its selective binding to inflamed synovial tissue, while being unable to bind to uninflamed tissue. The specificity of this peptide was unexpected, but 3 additional peptides encoded by phage clones selected from a peptide phage display library have recently been reported to show preferential binding to inflamed joints and skin (27). Although inflammation is a prerequisite for the synovial homing of MT07, the interaction with the inflamed synovium is tissue-specific, since the recombinant molecule failed to bind to inflamed kidney and lung. Our finding that the peptide binds to human synovium as well as to rat and mouse synovium is

DISCUSSION Maximum therapeutic efficacy and minimal side effects represent the gold standard for the management of RA. The first objective can be obtained by widening the range of inflammatory targets and using suitable drugs in various combinations. An important requirement for effective treatment of RA involving multiple joints is the preferential accumulation of antiinflammatory drugs in the inflamed joints, such that the undesired side effects of treatment can be reduced. We believe that these goals can be achieved by using the recombinant antiinflammatory molecule that selectively targets the synovial tissue, as described herein. Various attempts to deliver antiinflammatory drugs encapsulated in liposomes, including methotrexate or glucocorticoids, to inflamed joints have yielded satisfactory results (24). Liposomes coupled with RGD peptides on their surface that target endothelial cells have been used to deliver dexamethasone to sites of inflammation (25). More recently, the fusion protein CR2-fH, which contains the complement receptor 2 binding site for the C3 activation products iC3b/C3d and the aminoterminus of factor H, the inhibitor of the alternative pathway, was shown to target iC3b/C3d in the joint and to decrease the disease activity score in mice with CAIA

Figure 6. Effect of MT07 on established collagen antibody–induced arthritis in mice. Two groups of mice were injected intravenously with MT07 (300 ␮g) or saline 24 hours after intraperitoneal administration of anti–type II collagen antibody to induce arthritis. Swelling of the ankles (A), knees (B), and wrists (C) was monitored for 14 days. Mice were then euthanized, and the tissue damage score (D) was determined as the degree of synovial thickening, cell infiltration, and cartilage erosion. Values are the mean ⫾ SD of 6 mice per group. ⴱ ⫽ P ⬍ 0.01 versus saline-treated arthritic mice. Photomicrographs of knee joint sections obtained from representative mice treated intravenously with saline (E) or with MT07 (F) show synovial hyperplasia and leukocyte infiltration in the saline-treated mouse that are not seen in the MT07-treated mouse. Original magnification ⫻ 125.

2566

evidence against the species specificity of the synovialhoming peptide (19). A likely explanation for these conflicting results lies in the different status of the tissues analyzed as the peptide targets, since we used inflamed mouse and rat synovial tissue, whereas Lee and colleagues in their original work (19) examined normal uninflamed mouse kidney. AIA was selected as an in vivo model by which to follow the biodistribution of MT07 because it mainly involves a single joint in immunized mice following intraarticular injection of mBSA. Failure of MB12/22 to localize to the joint rules out the possibility that either the variable portion or the Fc fragment of the antibody component of MT07 was responsible for the homing of the recombinant molecule. A signal of the labeled molecule was clearly visualized in the joint by timedomain optical imaging 5 hours after injection of MT07, but we cannot exclude the possibility that the molecule starts localizing in the joint at an earlier time point after infusion. It is clear that the signal could still be detected in inflamed joints for several days and as long as 1 month after injection. This finding has important therapeutic implications that are particularly relevant to the treatment of RA patients who require repeated administrations of drug. MT07 proved to be effective in reducing the disease severity in the AIA and CAIA animal models, as revealed by the decreased joint swelling and the amelioration of synovitis. Similar results were obtained using a neutralizing monoclonal antibody to mouse C5 in a model of collagen-induced arthritis (12). These findings led investigators at Alexion Pharmaceuticals to investigate the therapeutic efficacy of eculizumab, a humanized antibody that binds to human C5 and prevents its cleavage (28), on a fairly large number of RA patients. That treatment yielded a satisfactory response in terms of reduced disease activity (29). However, MT07 has several advantages over the available antibodies. First, it can easily be produced as a fully human neutralizing antibody and is therefore less likely to be immunogenic after repeated administration to patients who require long-term treatment. In addition, MB12/22, which is used to construct the fusion molecule, binds to human C5 and cross-reacts with C5 from several species, including mice, rats, rabbits, and pigs (23). The antibody can, as a consequence, be tested and validated in preclinical models prior to its therapeutic use in patients. The selective delivery to inflamed joints also allows the drug to exert its antiinflammatory activity locally, without affecting circulating C5. In this way, the risk of increased susceptibility to meningococ-

MACOR ET AL

cal infection that is usually associated with C5 deficiency is virtually abolished. Finally, treatment with the recombinant molecule would be cost effective, since 300 ␮g was sufficient to prevent inflammation in the rat model of AIA, whereas a 1-mg dose of the soluble anti-C5 minibody MB12/22 was entirely ineffective. Failure of the soluble molecule to control joint inflammation at a 3-fold higher dose than that of the targeted antibody may be explained by the insufficient level reached by the antibody in the joint as a result of its neutralization by circulating C5. Treatment with MT07 was efficacious in controlling established inflammation in AIA, as revealed by the decrease in joint swelling and the reduced signs of inflammation. This effect was also observed in mice receiving the recombinant molecule 1 day after administration of Arthrogen to induce CAIA at a time when complement is already deposited and synovitis is in progress. Essentially the same data were obtained by Wang and colleagues (12) in mice with collagen-induced arthritis treated with a monoclonal antibody against mouse C5, suggesting that complement activation products contribute not only to the onset of the disease, but also to its maintenance and progression. In conclusion, we report the beneficial effects of a recombinant molecule consisting of a synovial-homing peptide fused to a neutralizing antibody to C5 that was selectively delivered to inflamed joints, where it was able to prevent and control the development of AIA in rats and CAIA in mice. We believe that this is a prototype of molecules that can be constructed by fusing the peptide not only with anticomplement antibodies, but also with different molecules that neutralize TNF and other proinflammatory cytokines. 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. Tedesco 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. Macor, Tedesco. Acquisition of data. Durigutto, De Maso, Garrovo, Biffi, Cortini, Fischetti. Analysis and interpretation of data. Sblattero, Pitzalis, Marzari.

REFERENCES 1. Weyand CM, Fujii H, Shao L, Goronzy JJ. Rejuvenating the immune system in rheumatoid arthritis. Nat Rev Rheumatol 2009;5:583–8. 2. Imboden JB. The immunopathogenesis of rheumatoid arthritis. Annu Rev Pathol 2009;4:417–34. 3. Taylor PC, Feldmann M. Anti-TNF biologic agents: still the

TARGETED ANTI-C5 ANTIBODY AS ARTHRITIS THERAPY

2567

therapy of choice for rheumatoid arthritis. Nat Rev Rheumatol 2009;5:578–82. Van Vollenhoven RF. Treatment of rheumatoid arthritis: state of the art 2009. Nat Rev Rheumatol 2009;5:531–41. Senolt L, Vencovsky J, Pavelka K, Ospelt C, Gay S. Prospective new biological therapies for rheumatoid arthritis. Autoimmun Rev 2009;9:102–7. Smolen JS, Aletaha D, Grisar JC, Stamm TA, Sharp JT. Estimation of a numerical value for joint damage-related physical disability in rheumatoid arthritis clinical trials. Ann Rheum Dis 2009;69: 1058–64. Corvetta A, Pomponio G, Rinaldi N, Luchetti MM, Di Loreto C, Stramazzotti D. Terminal complement complex in synovial tissue from patients affected by rheumatoid arthritis, osteoarthritis and acute joint trauma. Clin Exp Rheumatol 1992;10:433–8. Hietala MA, Nandakumar KS, Persson L, Fahlen S, Holmdahl R, Pekna M. Complement activation by both classical and alternative pathways is critical for the effector phase of arthritis. Eur J Immunol 2004;34:1208–16. Wang Y, Kristan J, Hao L, Lenkoski CS, Shen Y, Matis LA. A role for complement in antibody-mediated inflammation: C5-deficient DBA/1 mice are resistant to collagen-induced arthritis. J Immunol 2000;164:4340–7. Ji H, Ohmura K, Mahmood U, Lee DM, Hofhuis FM, Boackle SA, et al. Arthritis critically dependent on innate immune system players. Immunity 2002;16:157–68. Banda NK, Takahashi M, Levitt B, Glogowska M, Nicholas J, Takahashi K, et al. Essential role of complement mannose-binding lectin-associated serine proteases-1/3 in the murine collagen antibody-induced model of inflammatory arthritis. J Immunol 2010; 185:5598–606. Wang Y, Rollins SA, Madri JA, Matis LA. Anti-C5 monoclonal antibody therapy prevents collagen-induced arthritis and ameliorates established disease. Proc Natl Acad Sci U S A 1995;92: 8955–9. Woodruff TM, Strachan AJ, Dryburgh N, Shiels IA, Reid RC, Fairlie DP, et al. Antiarthritic activity of an orally active C5a receptor antagonist against antigen-induced monarticular arthritis in the rat. Arthritis Rheum 2002;46:2476–85. Kurreeman FA, Goulielmos GN, Alizadeh BZ, Rueda B, Houwing-Duistermaat J, Sanchez E, et al. The TRAF1-C5 region on chromosome 9q33 is associated with multiple autoimmune diseases. Ann Rheum Dis 2009;69:696–9. Plenge RM, Seielstad M, Padyukov L, Lee AT, Remmers EF, Ding B, et al. TRAF1-C5 as a risk locus for rheumatoid arthritis—a genomewide study. N Engl J Med 2007;357:1199–209. Andersson L, Stahl F. Distribution of candidate genes for experimentally induced arthritis in rats. BMC Genomics 2010;11:146. Lindqvist AK, Johannesson M, Johansson AC, Nandakumar KS,

Blom AM, Holmdahl R. Backcross and partial advanced intercross analysis of nonobese diabetic gene-mediated effects on collageninduced arthritis reveals an interactive effect by two major loci. J Immunol 2006;177:3952–9. Fischetti F, Durigutto P, Macor P, Marzari R, Carretta R, Tedesco F. Selective therapeutic control of C5a and the terminal complement complex by anti-C5 single-chain Fv in an experimental model of antigen-induced arthritis in rats. Arthritis Rheum 2007;56: 1187–97. Lee L, Buckley C, Blades MC, Panayi G, George AJ, Pitzalis C. Identification of synovium-specific homing peptides by in vivo phage display selection. Arthritis Rheum 2002;46:2109–20. Nakajima A, Ueda K, Takaoka M, Kurata H, Takayama J, Ohkita M, et al. Effects of pre- and post-ischemic treatments with FK409, a nitric oxide donor, on ischemia/reperfusion-induced renal injury and endothelin-1 production in rats. Biol Pharm Bull 2006;29: 577–9. Yamashita J, Itoh M, Kuro T, Kobayashi Y, Ogata M, Takaoka M, et al. Pre- or post-ischemic treatment with a novel Na⫹/Ca2⫹ exchange inhibitor, KB-R7943, shows renal protective effects in rats with ischemic acute renal failure. J Pharmacol Exp Ther 2001;296:412–9. Cortez-Retamozo V, Swirski FK, Waterman P, Yuan H, Figueiredo JL, Newton AP, et al. Real-time assessment of inflammation and treatment response in a mouse model of allergic airway inflammation. J Clin Invest 2008;118:4058–66. Marzari R, Sblattero D, Macor P, Fischetti F, Gennaro R, Marks JD, et al. The cleavage site of C5 from man and animals as a common target for neutralizing human monoclonal antibodies: in vitro and in vivo studies. Eur J Immunol 2002;32:2773–82. Garrood T, Pitzalis C. Targeting the inflamed synovium: the quest for specificity [editorial]. Arthritis Rheum 2006;54:1055–60. Koning GA, Schiffelers RM, Wauben MH, Kok RJ, Mastrobattista E, Molema G, et al. Targeting of angiogenic endothelial cells at sites of inflammation by dexamethasone phosphate–containing RGD peptide liposomes inhibits experimental arthritis. Arthritis Rheum 2006;54:1198–208. Banda NK, Levitt B, Glogowska MJ, Thurman JM, Takahashi K, Stahl GL, et al. Targeted inhibition of the complement alternative pathway with complement receptor 2 and factor H attenuates collagen antibody-induced arthritis in mice. J Immunol 2009;183: 5928–37. Yang YH, Rajaiah R, Ruoslahti E, Moudgil KD. Peptides targeting inflamed synovial vasculature attenuate autoimmune arthritis. Proc Natl Acad Sci U S A 2011;108:12857–62. Matis LA, Rollins SA. Complement-specific antibodies: designing novel anti-inflammatories. Nat Med 1995;1:839–42. Kaplan M. Eculizumab (Alexion). Curr Opin Investig Drugs 2002;3:1017–23.

4. 5. 6.

7.

8.

9.

10. 11.

12.

13.

14.

15. 16. 17.

18.

19. 20.

21.

22.

23.

24. 25.

26.

27. 28. 29.