C. Cerqueira

Neutralization of AntiCitrullinated Protein Antibodies in Rheumatoid Arthritis - A Way to Go?

Anticitrullinated protein antibodies (ACPAs) constitute a class of autoantibodies found in 60-70% of patients with rheumatoid arthritis (RA). The most common test for ACPA positivity is based on the occurrence of antibodies that bind to circular citrullinated peptides, so-called CCP, some of which are derived from endogenously citrullinated proteins, like filaggrin. Several lines of evidence suggest that these autoantibodies may confer pathological reactions. They appear years before onset of clinical disease and are associated with worse prognosis and a more erosive disease. Their presence correlates with the most prominent genetic risk factors for RA development, and they were recently described to mediate relevant biological functions such as activation of complement system and induction of osteoclastogenesis. The development of new drugs that specifically target these autoantibodies is an appealing and novel approach. Herein, we briefly review the autoimmune condition of RA, characterized by the presence of ACPA, and we describe how the neutralization of autoantibodies might become a novel pharmacological principle.

A4.17 Anti-citrullinated proteins antibodies promotes osteoclastogenesis and bone destruction in rheumatoid arthritis

Background/Purpose Presence of ACPA is a major risk factor for bone erosion in RA and antibodies against modified citrullinated vimentin induce osteoclast (OC) formation from monocytes. We aimed to identify new molecular mechanisms responsible for ACPA-mediated bone destruction by investigating the direct effect of ACPA (obtained from the synovial fluid (SF) and peripheral blood (PB) of RA patients) on osteoclastogenesis and their influence on distinct osteoclasts precursors (monocytes/macrophages (MΦ) and immature dendritic cells (DC)). Methods ACPA and non-ACPA IgGs were isolated from SF, (n = 26) and PB (n = 38) samples of RA patients. Osteoclasts precursors were generated (DC and MΦ) from ACPA+ RA patient and healthy donors PB and then differentiated into OC in presence of RANKL and M-CSF. In parallel, cells were grown on osteoassay surfaces and bone resorption area was quantified. In-house generated monoclonal anti-citrulline antibodies cloned from SF B-cells were also tested. Cytokines were measured by cytometric bead array in culture supernatants. Mass spectrometry (MS) analysis was performed on different stages of OC maturation. Immunohistochemistry (IHC) was used to stain the OCs with biotinylated ACPA IgG and monoclonal anti-citrulline antibodies. The effect of PAD inhibition (Cl amidine) and IL-8 inhibition was tested in the cultures. Results ACPAs pools enhanced osteoclastogenesis from both DC (fold change of (FC) 1.6 ± 0.14 for OC number) and MΦ (FC 2.0 ± 0.6 for OC and 1.6 ± 0.4 for bone resorption area (erosion)) of healthy donors. Similar effect was observed when the precursor cells were derived from ACPA+ RA patients in both DC (FC 2.3 ± 0.9 OC and 2.6 ± 0.1 erosion) and MΦ (FC 1.8 ± 0.6 OC and 2.3 ± 0.7 erosion). Increased osteoclastogenesis was associated with significantly higher levels of IL-8 levels in culture supernatants measured at all stages of osteoclasts maturation from both DC (FC 2.4 ± 0.5) and MΦ (FC 2.0 ± 0.5). Interestingly SF levels of IL-8 were higher in ACPA+RA patients as compared to disease controls. IL-8 neutralisation completely abolished ACPAs effects while blocking PAD enzymes resulted in a complete inhibition of OC formation. MS analysis showed the presence of actin and vimentin citrullinated during osteoclastogenesis. Two of the tested anti-citrullinated monoclonal antibodies (D10 and B2) but not a third anti-citrulline antibody (C7) nor a negative anti-tetanus control antibody (E2) bound to osteoclasts and promoted osteoclastogenesis and bone destruction, interestingly Fab fragments of the D10 and B2 antibodies retained similar effects. Conclusion In conclusion, SF and PB derived ACPA IgGs with broad specificities enhanced osteoclastogenesis from both DC and MΦ. This effect appears to be restricted to certain ACPAs specificities, IL-8 dependent and at least partially mediated through a Fab mediated mechanism.

Neutralisation of ACPA – A Way to Go?

Background and Objectives In a previous study, we have identified endogeneously citrullinated sites in fibrinogen from RA synovial tissue (Hermansson, et al, 2010 in Proteomics-Clin Appl). Within the alpha chain, Arg573 and Arg591 were found citrullinated with an occupancy rate in the range of 1–2% and in the β-chain, Arg72 and Arg74 were also found citrullinated. We now demonstrate that these citrullinated residues are autoantigenic as well as demonstrate that peptides containing these epitopes can be used as probes for development of ACPA neutralising compounds. Materials and Methods The autoantigenic potential was investigated using the Phadia’s ImmunoCAP ISAC® system. Citrullinated and unmodified fibrinogen peptides were immobilised onto a glass slide in an arrayed fashion and serum from 404 CCP positive and 532 CCP negative RA patients and 461 healthy controls from the EIRA cohort were tested. We also assayed the identified citrulline fibrinogen peptides for their ability to prevent purified ACPA (Ossipova, et al, 2012 submitted) to bind to CCP (CCPlus® ELISA, Euro-Diagnostica AB). Peptides were individually or in combinations incubated with different ACPA pools and the blocking efficiency was expressed as percent of inhibition and IC50. Corresponding arginine peptides were used as controls. Results We found that 31% (87% are CCP positive) of patients were positive to Cit573 peptide. For the Cit591 peptide, the corresponding numbers were 10% (65%), for the Cit74 peptide 28% (68%) and for the Cit72 peptide 20% (68%). Interestingly, citrullinated 573 and Cit591 peptides revealed a maximum of 77% and 48% ACPA inhibition, respectively. When equally mixed, these peptides displayed an additive higher degree of ACPA neutralisation (84%). In contrast, Cit74 and Cit72 peptides reached a more modest maximum inhibition of 26% and 30%, respectively. This experiment was repeated using a different set of ACPA pool and then the efficiencies were lower for Cit573 (47%) but similar for Cit591 (51%). Logically, the efficiency of specific citrullinated compounds will depend on the individual ACPA specificities. Conclusions Here we demonstrate extensive autoantibody reactivity against in vivo citrullinated fibrinogen epitopes found in RA synovial membranes. These peptides can now be used as additional biomarkers for studies of ACPA sub-specificity profiles as recently reported (Brink, et al, 2012 A&R, in press). We also demonstrate that these citrullinated peptides can be used as neutralising agents blocking a significant portion of ACPA binding to CCP. These results open novel possibilities for the design of personalised ACPA blockers preventing for instance the osteoclastogenesis and bone loss induced by ACPA (Harre, et al, 2012 JCI).

SAT0043 Identification and Characterization of Novel Molecular Mechanisms for ACPA-Driven Osteoclastogenesis

Background Anti citrullinated protein antibodies (ACPA) associate with bone destruction in RA Objectives We aimed to characterize the molecular mechanisms responsible for APCAs effect. Methods ACPA positive and negative IgGs were isolated from synovial fluid (SF, n=26) and peripheral blood (PB, n=38) samples of RA patients. CD14+ monocytes from PB of ACPA+ RA patients and healthy individuals were first cultured in the presence M-CSF to generate MΦ, and then further differentiated to OC in the presence of RANKL and M-CSF. In parallel, cells were grown on osteoassay surfaces and bone resorption area was quantified by computer assisted image analysis. In house generated anti-citrullinated monoclonal antibodies obtained from SF B-cells were also tested. Cytokines were measured by cytometric bead arrays in cultures supernatants. Immunohistochemistry (IHC) was used to stain the OCs with biotinylated ACPA IgG and monoclonal anti-citrullinated proteins antibodies. The effect of PAD inhibition (C.I.amidine), IL-8 and TNF inhibition was tested in the osteoclasts cultures. Results Polyclonal SF-derived ACPA IgGs increased RANKL-driven osteoclastogenesis from a median of 238 OC/well, IQR 55.5-378.5 to a median of 333.5, IQR 88.8-489.3. PB derived ACPAs had similar effect and equally potency. These changes were paralleled by an increase of bone resorption areas by ACPA IgGs from a median of 4.9%, IRQ 1.9-8.5 to a median of 6.9%, IQR 2.7-10. No significant increase in either osteoclasts numbers or resorption areas was observed with the control ACPA-negative IgGs. Increased osteoclastogenesis was associated with significantly higher levels of IL-8 levels in cultures supernatants (fold increase of 2.0±0.5), while no changes were observed for either TNF or IL-6. Cell exposure to neutralizing anti-IL8 antibodies completely abolished ACPA-mediated osteoclastogenesis. Monoclonal antibodies against cit-vimentin and cit-enolase, but not cit-fibrinogen, had similar effects with the polyclonal ACPAs, promoting osteoclastogenesis. Fab fragments of these monoclonal antibodies retained similar effects. Binding of both polyclonal and monoclonal ACPAs on OC was confirmed using IHC. Conclusions ACPA IgGs directly promote osteoclastogenesis and bone resorption, an effect that is dependent on antibody specificity and mediate through a novel IL-8 mediated mechanisms. IL-8 targeting might represent an alternative therapeutic approach to modulate antibody-mediated bone destruction. Disclosure of Interest None declared

A8.2 Anti Citrullinated Protein Antibodies from Synovial Fluid of Rheumatoid Arthritis Patients Enhance Osteoclastogenesis

Background/Purpose Presence of anti CCP2 antibodies identifies a subgroup of RA patients that are more prone to develop bone erosions. We hypothesised that anti CCP2 IgG might have a direct effect on bone, and thus investigated the effect of anti CCP2 IgG isolated from synovial fluid (SF) of RA patients on osteoclastogenesis and bone destruction in an in vitro system. Methods IgG were isolated on Protein G columns from SF of 26 RA patients and applied on CCP2 affinity columns. Pools of the purified anti-CCP2 and flow through IgG fractions were tested for the ability to influence osteoclastogenesis (TRAP positive multinucleated cells) and bone destruction (% of resorption area on osteologic discs). To do this immature dendritic cells derived from CD14+ cells from peripheral blood of healthy individuals were cultured in the presence of RANKL and M-CSF, with or without CCP2 IgG or flow through IgG (at a final concentration of 100 ng/ml). Results The CCP2 IgG pool induced a significant mean ± SEM of 1.5 ± 0.1 fold increase in the number of osteoclasts formed from immature dendritic cells in the presence of RANKL, while no such effect was observed with flow through IgG fractions. Osteoclasts cultured in the presence of the CCP2 IgG induced a significant mean ± SEM of 3.4 ± 1.3 fold increase of bone resorption while no such effect was observed for the flow through fractions. Conclusions Here, we demonstrate that ACPA IgG, isolated from SF of RA patients, have the ability to enhance the RANKL-driven osteocalstogenesis from immature dendritic cells. Our findings suggest that ACPA might have a direct pathogenic effect in RA associated bone destruction.

A10.14 Identification of Novel ACPA Targets in Rheumatoid Arthritis Synovial Tissues Using 2D Gel Electrophoresis and Mass Spectrometry

Background and Objectives Rheumatoid arthritis (RA) is an inflammatory autoimmune disease characterised by synovial joint inflammation and pannus formation that leads to degradation of cartilage and the underlying bone. Presence of anti-citrullinated protein/peptide antibodies (ACPA) in 60–70% of patients with RA is one of the major characteristics of the disease and associates with a more aggressive disease course, suggesting a direct pathogenic involvement of ACPA in disease initiation and progression. ACPA recognises several citrullinated proteins like fibrinogen, α-enolase, vimentin, and collagen II. In this study, we aim for the identification of novel ACPA targets in synovial tissues of patients with RA. Materials and Methods RA synovial tissues were obtained from patients undergoing joint replacement surgery for rheumatoid arthritis of the knee or elbow at the Karolinska University Hospital, Stockholm, Sweden. Synovial tissues were frozen in liquid nitrogen shortly after resection and stored at –80°C. All procedures were approved by Northern Stockholm Ethical Review Board and tissues were obtained with informed patient consent. Proteins, extracted from pulverised frozen synovial tissues and solubilised in lysis buffer, were resolved in 2D PAGE. Separated proteins were directly transferred to a nitrocellulose membrane and probed with human ACPA pool obtained using CCP2 affinity columns, kindly provided by Euro-Diagnostica, as described previously. [1] Human IgG and CCP2 flow-through fraction were used as control antibodies. Silver stained gel spots, corresponding to WB signals, were extracted from 2D gels, in-gel digested using Lys-C, and resulting peptides were identified using mass spectrometry. Results By combining 2D gel electrophoresis with mass spectrometry, we have identified several novel potential ACPA targets as well as already characterised proteins. It remains to demonstrate if these proteins are citrullinated. Conclusions Here we demonstrate an extensive ACPA reactivity against novel proteins in RA synovial membranes. The results encourage further exploration of the role of these proteins/peptides in rheumatoid arthritis both as additional biomarkers as well as their potential roles in the pathogenesis of RA. Reference Ossipova E, Cerqueira C, Reed E, et al, Affinity purification and characterization of anti-citrullinated protein/peptide antibodies from plasma and synovial fluids of patients with rheumatoid arthritis (submitted).

Neutralization of Acpa in Rheumatoid Arthritis -a Novel Principle of Treatment

Background In a previous study, we have identified endogeneously citrullinated sites in fibrinogen from RA synovial tissue1. Within the alpha chain, Arg573 and Arg591 were found citrullinated with an occupancy rate in the range of 1-2% and in the β-chain, Arg72 and Arg74 were also found citrullinated. Objectives In the present study, we demonstrate that citrullinated fibrinogen peptides are autoantigenic and can be used as probes for development of ACPA neutralizing compounds opening new opportunities for treatment of CCP positive RA patients. Methods The autoantigenic potential was investigated using the Phadia´s ImmunoCAP ISAC® system. Citrullinated and unmodified fibrinogen peptides were immobilized onto a glass slide in an arrayed fashion and serum from 404 CCP positive and 532 CCP negative RA patients and 461 healthy controls from the EIRA cohort were tested. We also assayed the identified citrulline fibrinogen peptides for their ability to prevent purified ACPA2 to bind to CCP (CCPlus® ELISA, Euro-Diagnostica AB). Peptides were individually or in combinations incubated with different ACPA pools and the blocking efficiency was expressed as percent of inhibition and IC50. Corresponding arginine peptides were used as controls. Results We found that 31% (87% are CCP positive) of patients were positive to Cit573 peptide. For the Cit591 peptide, the corresponding numbers were 10% (65%), for the Cit74 peptide 28% (68%) and for the Cit72 peptide 20% (68%). Interestingly, citrullinated 573 and Cit591 peptides revealed a maximum of 77% and 48% ACPA inhibition, respectively. When equally mixed, these peptides displayed an additive higher degree of ACPA neutralization (84%). In contrast, Cit74 and Cit72 peptides reached a more modest maximum inhibition of 26% and 30%, respectively. This experiment was repeated using a different set of ACPA pool and then the efficiencies were lower for Cit573 (47%) but similar for Cit591 (51%). Logically, the efficiency of specific citrullinated compounds will depend on the individual ACPA specificities. Conclusions Here we demonstrate extensive autoantibody reactivity against in vivo citrullinated fibrinogen epitopes found in RA synovial membranes. These peptides can now be used as additional biomarkers for studies of ACPA sub-specificity profiles as recently reported3. We also demonstrate that these citrullinated peptides can be used as neutralizing agents blocking a significant portion of ACPA binding to CCP. These results open novel possibilities for the design of personalized ACPA blockers preventing for instance the osteoclastogenesis and bone loss induced by ACPA4. References Hermansson M. et al. Proteomics: Clinical Applications., 2010, May; 4(5):511-8; Ossipova, E., Cerqueira, C., et al 2013, submitted; Brink, M., et al., Arthritis Rheum. 2013 Jan 10, in press; Harre, U. et al., J Clin Invest. 2012 May 1;122(5):1791-802 Disclosure of Interest None Declared

AB0043 Affinity purification and characterization of human acpas

Background Autoimmunity in rheumatoid arthritis (RA) is characterized by autoantibodies to citrullinated proteins/peptides (ACPA)1. These antibodies, present in 60-70% of patients, antedate clinical onset and associate with an erosive disease course, suggesting a direct pathogenic involvement in disease initiation and progression2. Objectives With this study, we aimed to develop an efficient method for the purification of human ACPA, and to characterize their frequency and fine-specificity pattern in synovial fluid (SF) and plasma of RA patients. Methods SF and plasma samples were collected with informed consent and ethical approval from patients (fulfilling the ACR criteria for RA) with high anti-CCP antibody levels. SF samples (n=36) were first treated with hyaluronidase to decrease viscosity, then proteins were precipitated with ammonium sulfate, dissolved and further dialysed against phosphate buffered saline (PBS), before the IgG fractions were purified on ProteinG columns (GE Healthcare, Uppsala, Sweden). Plasma samples (n=10) were diluted in PBS before applied to the ProteinG column. ACPAs were further purified using CCP2 affinity columns, kindly provided by Euro-Diagnostica. Recovery and purity of total IgG and anti-CCP IgG were analysed using SDS-PAGE, Nanodrop (Thermo Scientific, Wilmington, DE, USA) and the CCP2-ELISA kit. Fine-specificity of the purified ACPAs were investigated using in-house ELISAs, with peptides from citrullinated α-enolase (CEP-1), -vimentin (Cit-vim), -fibrinogen (Cit-fib) and -collagen type II (Cit-C1). Results Anti-CCP IgG could efficiently be purified from SF and plasma, using ProteinG-, followed by CCP2-, columns. No CCP IgG response could be detected in the flow-through fractions. Higher concentrations of total IgG were found in plasma (13,6 mg/ml) compared to SF (4,2 mg/ml), while a higher percentage of CCP-specific IgG was detected in SF (3%), compared to plasma (2%). The purified anti-CCP IgG fractions cross-reacted with CEP-1, Cit-vim, Cit-fib and Cit-C1, while no reactivity to these citrullinated antigens were detected in the IgG flow-through fractions. Anti-CCP IgG dilution curves (starting at 10 μg/ml of purified antibodies) demonstrated differences in affinity between patients, which may correspond to the different ACPA-fine specificity patterns seen in patients. Conclusions The described methodology efficiently purifies ACPAs with multiple specificities, which will allow for their use in in vivo and in vitro studies, to further elucidate their arthritogenic and pathogenic capacity. In addition, the ACPAs will be tools for future immunoprecipitation-, immunoblotting- and immunohistochemistry experiments. References Schellekens, G. A. et al., The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citrullinated peptide. Arthritis Rheum 43(1), 155 (2000). van der Helm-van Mil AH, Verpoort KN, Breedveld FC, Toes RE, Huizinga TW, Antibodies to citrullinated proteins and differences in clinical progression of rheumatoid arthritis. Arthritis Res Ther 7(5), R949 (2005). Disclosure of Interest None Declared

Affinity purification and characterisation of human ACPAs

Backgroundand objectives Autoimmunity in rheumatoid arthritis (RA) is characterised by autoantibodies to citrullinated proteins/peptides (ACPA). These antibodies, present in 60–70% of patients, antedate clinical onset and associate with an erosive disease course, suggesting a direct pathogenic involvement in disease initiation and progression. With this study, the authors aimed to develop an efficient method for the purification of human ACPA, and to characterise their frequency and fine-specificity pattern in synovial fluid (SF) and plasma of RA patients. Materials and methods SF and plasma samples were collected with informed consent and ethical approval from patients (fulfilling the American college of rheumatology criteria for RA) with high anti-CCP antibody levels. SF samples (n=36) were first treated with hyaluronidase to decrease viscosity, then proteins were precipitated with ammonium sulphate, dissolved and further dialysed against PBS, before the IgG fractions were purified on Protein G columns (GE Healthcare, Uppsala, Sweden).Plasma samples (n=10) were diluted in PBS before applied to the Protein G column. ACPAs were further purified using CCP2 affinity columns, kindly provided by Euro-Diagnostica. Recovery and purity of total IgG and anti-CCP immunoglobulin G (IgG) were analysed using SDS-PAGE, Nanodrop (Thermo Scientific, Wilmington, DE, USA) and the CCP2-ELISA kit. Fine-specificity of the purified ACPAs were investigated using inhouse ELISAs, with peptides from citrullinated α-enolase (CEP-1), -vimentin (Cit-vim), -fibrinogen (Cit-fib) and -collagen type II (Cit-C1). Results Anti-CCP IgG could efficiently be purified from SF and plasma, using ProteinG-, followed by CCP2-, columns. No CCP IgG response could be detected in the flow-through fractions. Higher concentrations of total IgG were found in plasma (13.6 mg/ml) compared to SF (4.2 mg/ml), while a higher percentage of CCP-specific IgG was detected in SF (3%), compared to plasma (2%). The purified anti-CCP IgG fractions cross-reacted with CEP-1, Cit-vim, Cit-fib and Cit-C1, while no reactivity to these citrullinated antigens were detected in the IgG flow-through fractions. Anti-CCP IgG dilution curves (starting at 10 µg/ml of purified antibodies), demonstrated differences in affinity between patients, which may correspond to the different ACPA-fine specificity patterns seen in patients. Conclusions The described methodology efficiently purifies ACPAs with multiple specificities, which will allow for their use in in vivo and in vitro studies, to further elucidate their arthritogenic and pathogenic capacity. In addition, the ACPAs will be tools for future immunoprecipitation-, immunoblotting- and immunohistochemistry experiments.