Final Report Summary - INTRICATE (Infectious triggers of chronic autoimmunity)
Executive Summary:
INTRICATE is a multidisciplinary consortium comprised of scientists and clinical investigators who joined together to develop a deeper insight into why infection predisposes to autoimmune disease. Autoimmune injury in rodents can be triggered by molecular mimicry with microbial proteins, non-specific activation of auto-reactive lymphocytes by bystander activation or bacterial superantigens, and by activation of innate immunity by LPS or other pathogen associated molecular patterns (PAMPs); but little is known about the involvement of these mechanisms in human autoimmune disease. INTRICATE addressed this problem by systematically analysing a single candidate disease, ANCA-associated vasculitis (AAV), a severe disease associated with autoantibodies specific for the neutrophil enzymes myeloperoxidase (MPO) and proteinase 3 (PR3) and for lysosome membrane associated membrane protein 2 (LAMP-2). AAV is a model disease because its known association with bacterial infection: specifically chronic dysbiotic expansion of S. aureus in upper airways is associated with relapse in AAV associated with autoantibodies to PR3; and autoantibodies to LAMP-2 recognise an epitope common to the bacterial adhesion FimH with which they cross-react. The work was separated into four interlocking Work packages:
Work Package 01 developed novel antigen arrays suitable for high-throughput screening for serum antibodies and applied them to prospectively assembled cohorts of individuals with AAV and controls. The results establish the ability of this approach for large-scale characterisation of autoantibodies and the ability of hierarchical clustering to segregate patients into groups. The analysis of the prospective cohorts demonstrated that antibodies to a range of staphylococcal proteins are rare in AAV even in those with dysbiotic colonisation. Conversely, antibodies specific for FimH are significantly more common in AAV than in healthy controls but their concentration is not influenced by urinary tract infections.
Work Package 02 characterised the diversity S. aureus in AAV and characterised their antigen independent effects on autoantibody generation. The results identified a high degree of genetic diversity in S. aureus strains from individuals with GPA, albeit with the emergence of predominant spa-types at particular times, that did not differ significantly from the healthy population; discovered antibodies to numerous S. aureus proteins in patients with GPA; and demonstrated PR3-specific IgD and IgA in patients with GPA which suggests a vigorous mucosal immune response.
Work Package 03 generated and characterised unique mouse strains that express human MPO or PR3 in instead of their murine orthologues; and has used them to determine how the fine specificity of antibodies to them influences the injury they cause, and whether infection provokes autoimmunity to them. Antibodies to human MPO or human PR3 in the circulation enhanced the inflammatory response after intra-tracheal instillation of LPS and the results demonstrate that antibody isotype is a much greater determinant than fine specificity and in particular that injury is not influenced by whether or not the antibodies interfere with enzymatic activity. Consistent with this latter notion was the clinical finding that the epitope-dependent inhibitory profile of ANCA towards active PR3 did not correlated with disease manifestations, disease severity or onset of disease. Infections with fimbriated bacteria did not elicit ANCA against the endogenous mouse LAMP-2 sequence and even recombinant strains of fimbriated E. coli expressing LAMP-2-related sequences failed to trigger a cross-reactive immune response to the endogenous murine LAMP-2 sequence confirming firm mechanisms of maintaining to tolerance to self antigens under physiological conditions.
Work Package 04 explored the interactions between FimH, LAMP-2 and antibodies to them together with host genetic factors that influence susceptibility to AAV. A unique panel of fluorescently tagged E. coli strains expressing normal or mutated FimH was generated and used to show that E. coli adhere more efficiently to microvascular endothelium than to bladder epithelium, and to identify specific amino acid residues responsible the difference. LAMP-2 was then identified as a previously unsuspected receptor for FimH on the surface of endothelial cells where it is also accessible to anti-LAMP-2 antibodies. Once ligated by specific antibodies, cell surface LAMP-2 is rapidly internalised with major consequences for cell function, including antigen presentation. Host susceptibility was examined by genome wide association study (GWAS) that revealed that susceptibility to AAV with autoantibodies to PR3 and MPO are associated with different HLA class II polymorphisms, whilst the PTEN and SEPINA1 loci - that respectively encode PR3 and its major inhibitor 1-anti-trypsin - are uniquely associated anti-PR3 associated disease. This provides the first evidence that the two disorders are genetically distinct.
INTRICATE has succeeded in providing new insights into the links between infection and AAV as well as a wealth of novel clinical and experimental resources that will be available to the community.
Project Context and Objectives:
Background and Aims
INTRICATE is a multidisciplinary group of scientists, clinical investigators, pathologists, (clinical) immunologists, world-class geneticists, microbiologists, developmental biologists, cell biologists and bio-informaticians each of whom has made a substantial individual contribution to understanding the links between infection and autoimmunity. They have collaborated extensively in the past but join now together to form a Consortium specifically constructed to meet the aims of FP7- HEALTH.2010.2.4.5-2: Infection and dysbiosis as the triggers of the development of inflammatory processes in allergies and autoimmune diseases. In order to maintain focus the INTRICATE project investigated autoimmunity because, despite common initial pathways, allergy and autoimmunity engage different effector pathways - resulting in immunity to self rather than foreign antigens. The in-depth analysis of those mechanisms forms an integral part of the project. The aim of the INTRICATE consortium is to prosecute a programme of Translational Research that delineates the role of infection in the induction and perpetuation of severe systemic autoimmune disease with the ultimate objective of identifying new therapeutic strategies based on knowledge of pathogenesis.
Autoimmune disease is a severe cause of morbidity whose prevalence in the Western world and has increased relentlessly over the past fifty years. This trend has been especially of late and epidemiological studies from Europe and United States give the prevalence as 8-10% compared to 3% ten years ago [Jacobson DL, 1997; http://www3.niaid.nih.gov/topics/autoimmune/]. This change has been attributed directly to the pace of societal change over the last century; for example to the consequences of improved housing and vaccination policies that have virtually eliminated the major childhood infectious diseases, and to the use of antibiotics that cut short the effects of common bacterial infections. The result has been an unprecedented pressure on the balance between the human immune system and the bacterial universe with which it has co-evolved over millions of years. This led to the proposal that one consequence of the reduced infective load was the increased predisposition to allergic and autoimmune disease – this is the "Hygiene Hypothesis" [Strachan DP et al. 1989; Strachan DP. 2000] – however in allergy and autoimmunity common initial pathways result in entirely different effector mechanism. These ideas stimulated a general reconsideration of the relation between the microorganisms and the immune system not only with regard to the long-term consequences of acute infection but also the underlying role of commensal bacteria in shaping the immune system [Sansonetti PJ. and Medzhitov R. 2009].
The INTRICATE proposal is based on the premise that research into complex human disorders such as autoimmune disease must be undertaken primarily using samples and tissue taken from human subjects. Accordingly we investigated the interaction between infection and autoimmunity in a “model” human disease: Anti-neutrophil cytoplasmic antibody (ANCA) associated systemic vasculitis (AAV). AAV is ideally suited because it is known to be caused by autoantibodies of defined specificity and second because it is strongly linked to infection. AAV is a clinically well defined autoimmune disease that destroys the small blood vessels and most of those affected fall into one of two broad clinical groups: those with Granulomatosis with Polyangitis (GPA) in which the systemic vasculitis is accompanied by (and often preceded by) granulomas involving the nasal sinuses and upper and lower airways often against a background of chronic sinus disease; and microscopic polyangiitis (MPA) which has neither granulomas nor upper airway involvement. Most other aspects are similar in both groups including involvement of the kidney to cause pauci-immune focal necrotizing glomerulonephritis (piFNGN) with rapid irreversible kidney failure and the lung associated with severe haemorrhage. There is general agreement that the disease is caused by ANCA that bind the soluble lysosomal enzymes myeloperoxidase (MPO) and proteinase 3 (PR3) or the more recently defined target, lysosomal associated membrane protein-2 (LAMP-2) [Jennette JC et al. 2006; Kallenberg CG. 2006; Kain R et al. 2008]. INTRICATE uses mouse models but only to answer the specific question whether infection with clinically relevant bacteria induces autoimmune disease in transgenic mice that express the human autoantigen. Our concentration on human samples and “humanised” mouse models should mean that we are be able to translate the results swiftly into the clinic.
The INTRICATE project had four specific aims, namely to:
• Use novel high-throughput antigen array technology and well-characterized patient cohorts to determine whether acute infection with specific microorganisms triggers the induction or re-activation of AAV; and to ascertain whether antibody responses to microbial proteins cross-react with native or epigenetically modified self-proteins
• Elucidate the reasons why dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised and systemic granulomatous vasculitis in AAV; and in particular to analyse the roles of microbial superantigens and the local adaptive immune response to them.
• Analyse the mechanisms of molecular mimicry in transgenic mice expressing the human forms of LAMP-2, PR3 and MPO – the major targets of autoantibodies in AAV – by determining whether infection with bacteria that express molecular mimics induce AAV and, if so, to define under which circumstances they do so.
• Characterise disease-associated genes identified in the European Vasculitis Genetics Consortium's genome wide association study (GWAS) of AAV and to examine their effect on gene expression and function; and to determine whether the genetic variants that predispose to AAV have been maintained in the gene pool because of a beneficial effect on resistance to infection.
Work strategy and general description
The INTRICATE Consortium consists of a multidisciplinary consortium of scientists and clinical investigators with diverse expertise who have joined together to elucidate the reasons why infection predisposes to autoimmune disease and to analyse of the links between bacterial infection and the development of ANCA-associated vasculitis (AAV) as a model autoimmune disease more generally. AAV was chosen because of its long-recognised links with bacterial infection; and because its pathogenesis sufficiently well understood. The proposed programme is based on: (i) the secure knowledge from epidemiological studies that infection is a common trigger for autoimmune disease in man; (ii) the results of rodent models of autoimmunity that establish at least four non-exclusive potential triggering pathways including molecular mimicry, bystander activation, superantigens, and activation of innate immunity; and (iii) current understanding of the immune system.
The immunological mechanisms that lead from infection or dysbiosis to autoimmunity and the processes involved are both diverse and complex and consequently the INTRICATE research analysed them systematically using a „model disease“. We have selected ANCA associated systemic vasculitis because it is a common autoimmune disease, the nature of the autoimmune response involved is well characterised and there are clear links between AAV and specific bacterial pathogens. AAV is therefore a uniquely tractable disease in which to analyse infectious triggers in the development of inflammatory processes in autoimmune disease.
The INTRICATE research generated new knowledge in four inter-dependent areas – represented in the four scientific Work packages – with the ultimate goal to translate the results into clinically applicable tools to predict, intersect and monitor processes that lead infection to trigger autoimmune disease. Each of the four complementary work packages focused on a different host and pathogen mechanisms that could be responsible for loss of tolerance to self-antigens and the development of tissue injury.
Work package 01 Infection and Dysbiosis in clinical AAV has two main objectives: first to develop novel antigen arrays suitable for high-throughput screening for serum antibodies specific for self-antigens or bacterial proteins in AAV; and second to assemble cohorts of prospectively followed patients with AAV suitable to determine the frequency of antibodies detected and their relation to disease activity.
Work package 02 S. aureus diversity and antigen independent effects in immune responses addresses the reasons for the well-characterized epidemiological association between dysbiotic nasal carriage of S.aureus and clinical relapse in granulomatosis with polyangiitis (GPA); and in particular to analyse the roles of microbial superantigens and the local adaptive immune response to them.
Work package 03 Chronic infection and Dysbiosis in murine models of AAV is generating transgenic mouse lines that express the human MPO, PR3 and LAMP-2, the targets of autoantibodies, to study pathogenesis and molecular mimicry in AAV; and developing mouse models of urinary tract infections.
Work package 04 Host-pathogen interactions in generation of AAV has three discrete objectives: to determine how genetic variation of FimH – an adhesin and major virulence factor of gram negative pathogens that commonly cause urinary tract infections (UTI) - influences its interaction with LAMP-2, its molecular mimic; to determine whether uptake of antibodies to LAMP-2 influence antigen processing and presentation; and finally to use genome wide association studies to identify genes that influence susceptibility to AAV.
Management structure and procedures
The Project Coordinator ensured the smooth operation of the project and guaranteed that all efforts were focused towards the objectives. She submitted all required progress reports, deliverables, financial statements to the European Commission, and, with the assistance of GABO:mi she was responsible for the proper use of funds and their transfers to participants. The INTRICATE office was established by and based at the coordinator in Medizinische Universitaet Wien and at GABO:mi in Munich. The Project Office at the Coordinator was concerned with the scientific management and the co-ordination of all research activities. The Project Office at GABO:mi was responsible for administrative, financial and contractual management and the organisational co-ordination of the project activities.
The Project General Assembly (GA) was in charge of the political and strategic orientation of the project and acted as the arbitration body. It met once a year unless the interest of the project required intermediate meetings. The Project Steering Committee (STC) consisted of all work package leaders and the Coordinator and was in charge of monitoring all activities towards the objective of the project in order to deliver as promised, in due time and in the budget. The Project Steering Committee met every six months during the funding period. Furthermore, a scientific advisory board (SAB) was implemented to ensure a high standard of research and monitor the progress of the project by taking part in the annual General Assembly.
Project website: www.intricate.eu
The News section of INTRICATE website contains the publishable summaries, several press releases and also the video with Prof. Kain.
This video was already in social media channels (Facebook, Twitter, Google+, LinkedIn) announced.
Project Meetings:
Type of meeting Date Venue
Steering Committee Meeting 30th November 2010 Vienna, Austria (INTRICATE kick off meeting)
WP01, WP02 Investigator Meeting 16th May 2011 Chapel Hill (at 15th international Vasculitis & ANCA Workshop)
Steering Committee Meeting 4th May 2011 Telephone conference
Steering Committee Meeting 22nd September 2011 Cambridge, UK (INTRICATE 2nd general assembly)
Steering Committee Meeting 28-29th October 2012 Stanford, USA (INTRICATE 3rd general assembly)
WP01 Clinical Investigator Meeting 17th April 2013 Paris, France (at 16th international Vasculitis & ANCA Workshop)
Steering Committee and 4th GA Meeting 13-15th October 2013 Munich Germany (INTRICATE 4th general assembly)
Steering Committee Meeting 14th January 2014 Telephone conference
Steering Committee and 5th GA Meeting 3-5th September 2014 Vienna, Austria
Steering Committee Meeting 3rd September 2014 Vienna, Austria
WP01 Clinial Investigator Meeting 13th November 2014 Philadelphia, USA (at the American Society Nephrology renal week meeting)
Final Steering Committee and GA Meeting 17-18th April 2015 Cambridge, UK
WP Leader meetings via TCs regularly Telephone conferences
WP meetings via TCs regularly Telephone conferences
Objectives of INTRICATE:
The objectives of the four INTRICATE work packages are defined in its specific tasks:
WP01 uses a novel high-throughput technology and well-characterized patient cohorts to profile the serum antibodies to test the hypothesis that acute infection with specific microorganisms triggers the development or reacti¬vation of ANCA-associated systemic vasculitis; and to ascertain whether antibody responses to individual microbial proteins cor¬relate with those to native or epigenetically modified self-proteins from the tissues affected in vasculitis. Specifically, its objectives are to
1. Assemble the required cohorts of prospectively followed patients with AAV, infection and relevant controls,
2. Synthesize and utilize antigen microarrays and combinatorial peptide libraries designed to identify infections associated with AASV,
3. Characterize bacterial proteins expressed by bacteria associated with AAV,
4. Develop tools to monitor antibodies to bacterial peptides and proteins in AAV.
WP02 investigates the reasons why the dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised granulomatous inflammation in Wegener’s Granulomatosis that often presages the development of systemic disease. The overall purpose of this work package is to elucidate possible mechanisms that explain their localisation or association with disease. Specifically it:
1. Analyses the genetic diversity of S. aureus strains isolated from patients with AAV to ascertain whether they have particular properties,
2. Ascertains whether staphlococcal proteins induce lymphocyte and neutrophil responses likely to induce or maintain injury in AAV,
3. Analyses the local B and T cell responses to S. aureus in the upper airways of patients with AAV.
WP03 aims to overcome one of the main barriers limiting analysis the role of molecular mimicry in loss of tolerance to self-proteins human autoimmune disease, by creating transgenic mouse lines that express the human forms of LAMP-2, PR3 and MPO – the major targets of autoantibodies in AAV. These mice will then be used to analyse the conditions required to induce autoimmunity by molecular mimicry. Its specific aims are to:
1. Generate transgenic mouse lines expressing human LAMP-2, PR3 and MPO and to analyse the consequences of immunity to them,
2. Ascertain whether infections with fimbriated bacteria or S. aureus induce ANCA and whether they induce autoimmune disease,
3. Produce natively folded LAMP-2 and PR3 in order to analyse mimicry with three dimensional epitopes rather that linear.
The aim of WP 04 is to examine what determines and influences the development of autoimmune disease when a genetically susceptible individual is exposed to a sufficiently powerful environmental challenge using FimH/LAMP-2 molecular mimicry in AAV as a model. It combines bacterial genetic studies, cell biology with data from the first genome wide association study (GWAS) in AAV three specific aims which are to:
1. Ascertain how genetic difference in FimH influences its uptake by cells and the resulting effects,
2. Determine whether antibodies to FimH/LAMP-2 affect presentation of PR3 and MPO antigens to T cells,
3. Use GWAS data to identify immune-related genes that influence susceptibility to AAV or infection.
Project Results:
WP01 Infection and dysbiosis in clinical AAV
Work package 01’s main objectives were to develop novel antigen arrays suitable for high-throughput screening for serum antibodies specific for self-antigens or bacterial proteins in AAV; and to assemble cohorts of prospectively followed patients with AAV suitable to determine the frequency of antibodies detected and their relation to disease activity.
Work completed includes the (i) generation of libraries of purified, synthetic and recombinant proteins for array development, (ii) generation of novel antigen arrays for large-scale antibody screening – including a novel protein microarray for discovery of anti-cytokine antibodies in AAV - and the (iii) assembly of large prospectively followed patient cohorts and their serum samples that were applied to the arrays and analysed by necessary bioinformatics approaches.
GENERATION AND ANALYSIS OF NOVEL MICROARRAYS
INTEL Microarray Partner 08 (Stanford), in collaboration with Intel Inc., developed two silicon-based peptide microarrays featuring sequences from the known autoantigens histone H2B and U1-70K, respectively. The peptides were synthesized using maskless photolithography on the surface of silicon wafers. Each microarray contained every possible overlapping peptide within linear protein sequences covering the N-terminal tail of human histone H2B or the RNA binding domain of U1-70K. The H2B microarrays included peptides with acetyl and methyl modified lysine residues (as well as unmodified control peptides). These modifications are observed in vivo, and are hypothesized to be involved in loss of tolerance to the H2B antigen. Using the H2B microarrays allowed to identify the peptide epitopes of serum IgG autoantibodies in individuals with systemic lupus erythematosus. The U1-70K microarrays were used to identify a minimal epitope (116-121) that was cross-reactive with a known epitope in the M1 protein of influenza B viruses. Molecular mimicry has been hypothesized as a mechanism of development of autoantibodies to RNP (a protein/RNA complex that includes U1-70K). A recently developed technique to develop next generation microarrays on the surface of silicon wafers with an underlying integrated circuit allows, by combining these new microarrays with magnetic detection methods, to perform real-time measurement of antibody binding.
Histone microarrays Partner 08 (Stanford) developed peptide microarrays, featuring histone pepitdes derived from H2A, H2B, H3 and H4, for profiling histone post-translational modifications (PTMs) of neutrophil extracellular traps (NETs). Modifications included mono-, di- and tri-methylated lysine, citrulline, acetyl-lysine, and phosphorylation. Antibodies to histones are observed in several autoimmune disorders, including systemic lupus erythematosus (SLE) and the work elucidate the hypothesis that specific histone PTMs within NETs are involved in loss of tolerance and induction of histone autoantibodies. The results showed that SLE patients with reactivity to histones had serum antibodies that bound acetyl-modified histone H2B and other histone PTMs. Further investigation confirmed that distinct histone PTMs - indicative of transcriptionally silent DNA - are present in NETs using high-content immunoblotting. Immunization of Balb/c mice with NETs induced a weak autoantibody response, with partial overlap in autoantibody profiles to those seen in patients with SLE.
Cytokine autoantibody microarrays Recently, anti-cytokine autoantibodies have been found to cause increased susceptibility to infections that are not typically seen in the general population. These include anti-IFNg autoantibodies in patients with diffuse mycobacterial infections, and anti-Th17-associated cytokine autoantibodies in APECED patients with chronic mucocutaneous candidiasis. Partner 08 (Stanford) developed a novel protein microarray for the high-throughput discovery of anti-cytokine autoantibodies in autoimmune diseases and immunologic deficiencies. The microarray was customised to address the specific question whether anti-cytokine autoantibodies could be responsible for the increased prevalence of Staphylococcus aureus (SA) infections observed in AAV patients and consists of 65 commercially available recombinant cytokines, chemokines and immune mediators - including IL8, TNF, IL1b and MIP1a, that are produced by neutrophils - and 34 well-characterized autoantigens, including ANCA proteins MPO, PR3 and LAMP2. The array was used to analyse whether tolerance to these self-proteins could have also been broken in parallel with or subsequent to known ANCA proteins, and represent previously unidentified autoantigens in AAV. Beyond this targeted screen of neutrophil-associated cytokines and chemokines, a complimentary, unbiased approach by including a large number of theoretically unreleated cytokines and chemokines was included on the microarray.
In collaboration with partner 07 (UCAM), Rainer Doffinger and Ken Smith, partner 08 evaluated over 50 serum samples from patients with a variety of immunodeficiencies, including a recently identified patient population that suffers from diffuse mycobacterial infections due to anti-IFNg autoantibodies, using the anti-cytokine microarrays. Hierarchical clustering of the patient samples (blinded) based on their reactivity patterns, identified two distinct patient groups, or clusters. The first cluster had high reactivity to GMCSF, which has been previously reported in pulmonary alveolar proteinosis (PAP), while the second had high reactivity to type I IFN and Th17-associated cytokines (IL17F, IL22), which have been previously associated with APECED (autoimmune polyendocrine syndrome type I). Unblinding revealed that the results had identified patients with PAP and APECED with 100% accuracy, demonstrating that cytokine microarrays were capable of identifying anti-cytokine autoantibodies in immunodeficiencies.
Autoantigen microarrays Nitrocellulose 16-pad microarray slides were printed with 27 known or suspected AAV antigens and controls in triplicate using a robotic arrayer. ANCA-associated Vasculitis (AAV) serum, plasmapheresis samples and controls provided by INTRICATE partners were diluted and used to probe the microarrays in duplicate, followed by a fluorescently-labeled anti-human IgG secondary. Microarrays were scanned using a fluorescent microarray scanner, and the mean fluorescent intensity (MFI) of each protein feature was used as a measure of antibody binding.
In total, 127 samples from 5 INTRICATE clinical investigators were analyzed using autoantigen microarrays in the laboratory of partner 08. AAV patients and controls who were selected because likely to be informative have been assayed in a series of INTRICATE arrays. This work has provided an insight into the antibody responses in AAV that can be translated into the development of novel biomarkers.
The centers were: partner 01 (MUW, Kain), partner 05 (Mayo, Specks), partner 06 (UMCG, Kallenberg), partner 07 (UCAM, Smith/Lyons) and partner 08 (Stanford, Utz). Partner 08 was unblinded to the diagnosis and clinical features of the collaborator samples after circulating the coded microarray results to all collaborators.
Hierarchical clustering successfully divided patients samples based on whether they had p or c-ANCA. FimH reactivity was observed in a number of patient samples and did not appear to be associated with p or c-ANCA. Interestingly, reactivity to partner 08’s commercial preparation of LAMP-2 was observed in patient samples from both p and c-ANCA groups, while little reactivity was observed to the other LAMP-2 preparations. Reactivity to the less common ANCA antigen azurocidin was observed in a cluster of c-ANCA positive patient samples that were positive for PR3 and HNE. A second cluster of patient samples has reactivity to PR3, but lacks reactivity to the S195A mutated version of PR3. This suggests that their binding to the PR3 antigen may be dependent on this residue.
In a parallel analysis, epoxy-coated microarray slides were printed with 30 known or suspected SLE antigens, AAV antigens, and controls proteins in triplicate using a robotic arrayer. Serum samples derived from SLE patients that have been characterized by SNP typing, gene expression analysis, and flow cytometry, were provided by partner 07 (UCAM, Smith/Lyons).
Microbial microarrays Partner 01 (MUW) used 19 selected serum samples, one at untreated disease presentation and one during remission each, from nine patients with AAV and one healthy control to investigate IgG and IgM antibody responses to microbial peptide antigens in a customized, commercial peptide array that contained 4,347 linear B-cell epitopes (max. 16 aa length) derived from 159 pathogens associated with infectious diseases and control peptides. Analysis to date showed that, both IgG and IgM responses to microbial antigens in general were significantly higher in patients with active disease compared to remission and that antibodies from patients with active disease reacted to a broader range of antigens indicative of epitope spreading.
INTRICATE COHORT
In total, 328 patients with AAV, 50 patients with UTIs and 60 healthy controls were collected to the INTRICATE cohort. Their serum, blood and DNA samples were, in addition to samples of 170 patients with AAV from the INTRICATE collection, available for the studies proposed. Recruitment of patients and collection of their samples thus provides a biorepository of over 1000 serum samples and DNA as well as microbial collections of 105 urinary isolates and 271 nasal isolates from clinically well characterised patients and controls that were available for the INTRICATE research. The clinical data from patients recruited at partner 01 have already been entered into the INTRICATE registry at MUW: http://cemsiis-akim.meduniwien.ac.at/plattformen/rda/projekte/intricate-register/.
In conclusion, the results from WP01 have resulted in the
1. Generation of novel resources (INTRICATE arrays) that provide at the same time a diverse but specific platform for the analysis of antibody responses in large patient cohorts.
2. Assembly of the INTRICATE cohort that provides a uniquely large cohort of over 300 patients with AAV and controls and their serum samples and DNA as well as microbial collections
3. Prove of concept for the use of microarrays for antibody profiling
4. Provision of novel insights into antibody responses in sera from patients with AAV, SLE, autoimmune inflammatory disorder and infection that can be translated to other autoimmune diseases
WP02: Antigen independent effects of S. aureus on immune responses in AAV
Work package 02 addresses the reasons for the well-characterized epidemiological association between nasal carriage of S. aureus and clinical relapse in granulomatosis with polyangiitis (GPA). The work aims at elucidating the reasons why dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised and systemic granulomatous vasculitis in AAV. In particular the work analysed the roles of microbial superantigens and the local adaptive immune response to them. The results identified (i) a high degree of genetic diversity in S. aureus strains from individuals with GPA, albeit with the emergence of predominant spa-types at particular times; (ii) discovered antibodies to numerous S. aureus proteins in patients with GPA; and (iii) demonstrated PR3-specific IgD and IgA in patients with GPA which demonstrates a vigorous mucosal immune response.
Dysbiotic expansion of Staphylococcus aureus (S. aureus) in the upper airways occurs in about 60% of patients with Granulomatosis with Polyangiitis (GPA, formerly Wegener’s).This contrasts with the occurrence of chronic nasal carriage in only 20% of healthy controls. The persistence of S. aureus in the nasal cavity in GPA patients is associated with recurrent relapses of the disease.
S. aureus genetic diversity In this part of project partner 06 (UMCG) first analyzed the genetic diversity of S. aureus in patients with ANCA-associated vasculitis (AAV). The work investigated the hypothesis that particular strains of S. aureus could be related to the effect of S. aureus carriage on relapsing disease. The results from testing 210 isolates from the nasal cavity from 71 GPA patients in multiple techniques showed that the S. aureus population structure in GPA patients proved highly diverse without showing a particular subset being dominant and associated with relapsing disease. In fact, these isolates did not differ in the genetic diversity of S. aureus from isolates from healthy controls. The S. aureus population structure changed over time. Interestingly, increased resistance to co-trimoxazole and ciprofloxacin developed over time in the isolates from the GPA patients and this increased resistance seemed to coincide with the increased treatment of patients with co-trimoxazole. Co-trimoxazole resistant isolates were only obtained from patients treated with this antibiotic, and co-trimoxazole resistance was not observed for isolates from healthy controls.
Humoral and cellular responses to S. aureus antigen Next, partner 06 (UMCG) evaluated in depth the humoral response against 59 S. aureus antigens. The humoral antibody response against S. aureus appears deficient in GPA patients which may, at least in part, explain the persistence of S. aureus carriage in the patients. This seems to be a specific defect as the humoral response against influenza antigens was not disturbed in GPA patients.
Analysis of the effects of S. aureus derived stimulatory molecules on innate and adaptive immunity in AAV revealed that there appeared to be a shift towards inflammatory monocytes in GPA as a more pronounced decrease in classical monocytes was present in patients with GPA who were chronic carrier of S. aureus compared to non-carrier. This suggests that S. aureus carriage induces a shift in monocytes towards the pro-inflammatory type. No effect of S. aureus carriage was seen on functional capacities of neutrophils including the functional capacity of CD62L-low and CD16-high expressing neutrophils to downregulate T-cell proliferation.
Partner 06 (UMCG) also analyzed in depth the adaptive immunity in relation to S.aureus carriage and found that S. aureus carriage appears to stimulate an important subset of T-cells, namely follicular helper T-cells, in GPA patients. This subset drives the adaptive humoral immune response including the autoimmune response and activates macrophages, dendritic cells, NK-cells and NKT-cells. Interestingly, the Toxic Shock Syndrome Toxin-1 (TSST-1) from certain S. aureus strains appeared to have an amplifying role in autoantibody production.
Analysis of the link between IgD-positive B-cells and S. aureus in the upper airways demonstrated that IgD-ANCA do occur but are infrequently demonstrated in the peripheral blood in contrast to IgA-class ANCA. B-cell clusters are certainly present in the lesions in patients with GPA in the upper airways but it was not clear whether IgD-positive B-cells are related to carriage of S. aureus.
In conclusion, the results show that
1. The S. aureus population of patients with GPA is highly diverse and mirrors the general S. aureus population as seen in controls.
2. Chronic use of co-trimoxazole induces co-trimoxazole resistance and the immune response to multiple staphylococcal antigens is decreased in patients with GPA despite chronic exposure so explaining the persistence of S. aureus in the nasal cavity. Consequently, these important findings should guide to more targeted therapies: S. aureus nasal isolates should be frequently tested for co-trimoxazole resistance if co-trimoxazole is used to eliminate S. aureus carriage and antibiotic therapy should be either changed when resistance develops or preemtively changed periodically. In order to improve the humoral response against S. aureus vaccination should be considered and effective vaccines developed.
3. S. aureus carriage in patients with GPA mainly stimulates T-follicular helper cells, which are the primary effector memory T-cells. This finding implies that these cells could be specifically targeted in AAV/GPA. Development of therapeutic monoclonal antibodies to these cells should be considered and trials which such antibodies designed. Furthermore, the role of the TSST-1 superantigen of S. aureus should be taken into account. This means that S. aureus isolates should be tested for the presence of this superantigen. If present, elimination of these strains of S. aureus should be in the forefront of therapy.
4. S. aureus carriage induces pro-inflammatory monocytes. These results indicate that this cell population should be considered as targets of treatment. Presently, most attention is given to neutrophils but targeting monocytes could also be essential in the therapeutic approach in patients with AAV/GPA.
WP03 Analysis of molecular mimicry in murine models of AAV
Work package 03 is designed to analyse the mechanisms of molecular mimicry by determining whether infection with bacteria that express molecular mimics induce AAV and, if so, to define under which circumstances they do so. This is achieved by generating transgenic mouse lines that express the human MPO, PR3 and LAMP-2 – the major targets of autoantibodies in AAV – to study pathogenesis of AAV and molecular mimicry and by developing mouse models of urinary tract infections. As part of WP04 (i) mouse lines that express human PR3 or MPO and a mouse line that expresses human LAMP-2 driven by the Tie2 promoter/enhancer had been generated. (ii) Human PR3 and MPO expressing mouse lines and a murine model of pyelonphritis has been used to evaluate antibody responses and molecular mimicry in the development of AAV and (iii) newly generated, appropriately folded ANCA antigens were used to characterize the responses of antibodies to ANCA antigens in human.
Generation of humanized mouse lines expressing MPO, PR3 and LAMP-2 As for many autoimmune diseases, it is currently not understood how autoimmune responses are induced by external triggers. More than 90% of all autoantibodies are directed against distinct 3D structures of natively folded host proteins, called conformational epitopes. The 3D-surface of human proteins is highly divergent from those of commonly used rodents excluding the cross-reactivity of human autoantibodies with endogenous proteins of rodents. According to current theories, there are two principal ways in which external agents could induce autoimmunity. External components may coincidentally share structural, topological similarities with one of the host proteins or may elicit a strong immune response of antibody molecules which themselves invoke the production of anti-antibodies, as the antigen combining regions of these primary antibodies are also regarded as foreign. The foreign primary trigger in the latter case should be complementary to the host protein and should interact with it like the anti-idiotypic antibody. The major target self-antigens in autoantibody-associated vasculitis are PR3, MPO and LAMP2. The human LAMP2 protein (residues 41 to 49), but not the murine homolog, was reported to share a high sequence homology with the bacterial protein FIMH (residues 72-80) of type 1 fimbriae which are expressed by various bacterial species. WP03 therefore explored whether an experimentally induced bacterial infection of kidneys caused the formation of cross-reacting autoantibodies.
Partner 12 (HMGU) successfully established genomically modified knock-in mouse lines expressing the human PR3 and human MPO vasculitis antigens from the endogenous mouse promoters by homologous gene replacement. Both antigens were demonstrated to be correctly processed and functionally expressed in mouse neutrophils from bone marrow. The PR3-knock-in mice were distributed to partner 03 (Rfw-Bonn), the MPO knock-in mouse colony is currently being expanded in the animal facility of partner 12 (HMGU). A humanized LAMP-2-expressing transgenic mice could not be established during the funding period.
Pathogenicity of antibodies to ANCA antigens in humanized mouse lines Several monoclonal antibodies to human MPO (partner 06 UMCG and 10 HB) and newly produced (partner 12 HMGU) monoclonal antibodies to human PR3 were evaluated by intraperitoneal application in knock-in mice in combination with intratracheal instillation of low dose lipopolysaccharide (LPS) to see whether these antibodies are pathogenic under conditions of bronchial infections. Using the respective isotype controls we discovered that the pathogenic activity of mouse monoclonal antibodies to self-antigens of neutrophils was highly isotype dependent and not a matter of their binding specificity to a structurally distinct epitope. Two antigen-specific monoclonal antibodies of the Ig2a and Ig2b isotype class were found to enhance the low inflammatory effects of LPS in the lungs of mice suggesting that pathogenic effector functions of neutrophils, macrophages and dendritic cell are triggered by distinct interactions of certain IgG subclasses and activating Fc gamma receptors. In the future, the humanized knock-in mice will permit investigators to identify the Fc gamma receptors on major target cells that mediate the proinflammatory effects of autoantibodies under pauci-immune conditions of chronic inflammatory diseases.
Molecular mimicry in murine models of infection Alternative experimental approaches were chosen to study the pathogenic role of murine LAMP-2-FIMH cross-reacting antibodies in mice during experimentally induced pyelonephritis, e.g. after direct immunization with the murine LAMP-2-derived peptide AVPDKATHDGSSC. In a second approach, a mutant E. coli strain was engineered to achieve a local sequence identity between a mutated FIMH and the endogenous murine LAMP-2 protein. Although antibodies to murine LAMP-2 were generated after peptide immunization or in the course of experimentally induced bacterial infections, no differences of the inflammatory response in murine kidneys were noticed. To clarify, why no differences were observed during the infection with the Escherichia coli, peripheral blood was taken one day before infection, and serum was analysed by ELISA to check if LAMP-2/FimH-specific antibodies were produced after infection. The FimH-specific antibody titer was strongly increased 2 weeks after the first infection in contrast to the non-infected control. Surprisingly, 2 weeks after the second infection the specific LAMP-2/FimH-antibody titer decreased dramatically and unexpectedly. The results from these infection models indicated that pathogenic cross-reactive autoantibodies to the native murine LAMP-2 on murine neutrophils and activated Gr-1 positive myeloid cells are not easily induced by infections with mutant FimH-expressing bacteria or by immunization with a mLAMP2-derived linear peptide. Decrease of antibody titres in patients with urinary tract infections (UTI), but not AAV, with type 1 fimbriated pathogens investigated in WP04 exhibited a similar course after re-infection or resistant disease and indicates complex mechanisms of tolerance maintenance under physiological conditions in individuals that do not develop AAV after UTI.
Responses of antibodies to ANCA antigens Apart from directly cross-reacting antibodies, cross-reactive anti-idiotypes may be formed as a secondary response to antibodies (idiotypes) that are induced by an exogenous protein of potential bacterial origin. The primary exogenous immunogen should tightly interact with a folded host protein, e. g. PR3, and may, therefore, belong to the large family of serine protease inhibitors. Bacterial pathogens secrete a multitude of pathogenic factors in order to interfere with host-specific functions, in particular with endogenous proteolytic activities. Recently a new structural class of proteins, called EAPs (extracellular adhesion proteins) from Staphylococcus aureus were identified as inhibitors of PR3. PR3 inhibitors of bacterial origin could trigger the formation of anti-idiotypes that are directly interfering with the activity of PR3 similarly as the primary immunogen. We therefore searched for activity changing autoantibodies in vasculitis patients and developed a highly sensitive quantitative assay to detect these antibody populations in homogenous solution under equilibrium conditions. A large proportion of patients were shown to produce PR3-targeting autoantibodies with inhibitory capacity during the course of disease. However, no correlation between the changes in these autoantibody subpopulations and disease severity or disease progression was found in our clinical study in collaboration with partner 05 (Mayo).
Partner 02 (MPG, HMGU) has discovered and characterized a new serine protease with significant homology to human PR3 and neutrophil elastase. NSP4 is a low abundant neutrophil elastase- (NE-) with PR3-like biophysical properties, almost identical molecular weight and unusual substrate specificity. Like the other three abundant serine proteases of neutrophils (NE, cathepsin G, PR3), NSP4 is processed at the N-terminus and converted to its functionally active mature form by dipeptidylpeptidase I (cathepsin C). It is stored in the azurophil (primary) granules of neutrophils together with MPO, NE, cathepsin G and PR3 and released upon stimulation.
In conclusion, the results
1. Demonstrate the validity of using humanized murine models for the investigations of autoimmunity and the mouse lines established provide an invaluable resource for future studies investigating autoimmunity to ANCA antigens. One key finding is that target specificity of autoantibodies is less relevant for their pathogenicity than the isotype of autoantibodies. These result should lead to the development of assays that measure pathogenic autoantibody subsets to specific ANCA antigens
2. Indicate complex mechanisms of tolerance in molecular mimicry that could be explored to develop tolerizing immuno-therapies in individuals predisposed to developing cross-reactive antibodies
3. Identify novel potential ANCA antigens for clinical evaluation.
WP04 Host-pathogen interactions in generation of AAV
Work package 04 was based on the demonstration that the autoantibodies to LAMP-2 found in over 80% of patients with AAV commonly recognise an epitope shared with the bacterial adhesin FimH and that and that rodents immunised with FimH develop antibodies to LAMP-2 and AAV. This is one of the clearest examples of autoimmune disease induced by molecular mimicry and the central purpose of the work package was to examine the hypothesis that this mechanism applies to AAV clinically as well as in rodent models. Specifically, it was designed to provide insights into why only a small fraction of individuals infected with type 1 fimbriated bacteria develop antibodies to the shared FimH/LAMP-2 epitope AAV. It combined genetic studies of the relevant bacteria, functional studies of their effects on host cells and genetic studies of patients with AAV into and work in all three areas has produced important insights into (i) clinically relevant mechanisms in infections with type 1 fimbriated pathogens relevant to AAV and the injury it causes, (ii) the effects of antibody ligation on antigen presentation and (iii) genetic variances underlying the susceptibility to developing AAV, as well as providing the foundation for new avenues of research.
Molecular characterisation of FimH and its interaction with host cells Using a novel ELISA to quantify antibodies to FimH, Partner 01 (MUW) has shown that most patients presenting with AAV have IgG antibodies specific for IgG and that rising concentrations of these antibodies anticipate clinical relapses. However, urinary infections did not increase antibody titres either in patients with AAV or those without. A library of these pathogenic E. coli strains from individuals with and without AAV has been created and characterised by whole genome sequencing. Partner 04 (DTU) generated a unique panel of reagents for determining at a molecular level how FimH engages host cells: these include recombinant His-tagged FimH and a panel of GFP-tagged E. coli strains that express wild type or mutated FimH. Using these reagents, Partner 01 demonstrated that E. coli bound directly to glomerular endothelium in tissue sections, and that they adhere more efficiently to glomerular endothelial cells (GEnC) than to bladder urothelium in vitro, including under shear stress. Using the E. coli strains with mutated FimH, they identified two FimH residues responsible for enhanced bacterial adhesion to endothelium, and a further two that facilitated the initial adhesive contact under shear stress These studies emphasised the need to characterise the FimH receptors on microvascular endothelium responsible for adhesion and using a combination of affinity purification and mass spectrometry. Partner 01 identified three novel FimH binding partners on GEnC, including LAMP-2 which was then shown to mediate FimH-dependent adhesion to ldlD cells that stably express LAMP-2 on their cell surface (ldlD/hLAMP-2H). The discovery that LAMP-2 a cellular receptor for its molecular mimic FimH has important implications for the pathogenesis of microvascular injury in AAV but also for immune responses to LAMP-2 as indicated by the work undertaken in Task 2.
The effect of antibodies to LAMP-2 on antigen presentation Partner 01 has used human monocyte derived macrophages (MDM) and dendritic cells (moDC) to analyse how autoantibodies to LAMP-2 disturb its function in antigen presenting cells. They demonstrated that LAMP-2 was much more abundant on the surface of myeloid cells than previously suspected, even reaching 40% in activated moDC and was rapidly internalised after ligation by H4B4, a mouse monoclonal antibody specific for human LAMP-2. In macrophages, H4B4 trafficked rapidly to lysosomes and specifically depleted the lysosomal membrane of LAMP-2 and this in turn abrogated chaperone mediated autophagy (CMA) – a process critical for presentation of intracellular antigens to T cells. This established the principal that exposure to specific antibodies could interfere with LAMP-2’s intracellular functions. The high surface expression on moDC raised the question whether LAMP-2 was a surface receptor for antigen. As with macrophages, moDC rapidly internalised H4B4 that then trafficked rapidly to the MHC class II loading compartment (MIIC). Conjugates of a nominal antigen keyhole limpet haemocyanin (KLH) with H4B4 (H4B4-KLH) or isotype control (ISO-KLH) were used to analyse the effect of antigen delivery by LAMP-2. This confirmed H4B4-KLH was rapidly internalised and KLH trafficked significantly faster to MIIC than after pulsing with ISO-KLH. However, localisation to MIIC was transient and the overall effect was to re-route antigen from MIIC and into extracellular vesicles that were released into the surrounding medium. Analysis of peptides eluted from HLA-DR molecules displayed on the cell surface showed that fewer KLH-derived peptides when KLH was delivered by LAMP-2, whereas significantly more KLH and HLA-DR molecules were incorporated into EV when moDC were pulsed with H4B4-KLH than ISO-KLH. These changes had profound effect on CD4 T cell responses since proliferation induced by H4B4-KLH pulsed moDC was markedly attenuated whilst proliferation induced by EV released from them was considerably augmented: in both cases the differences were highly significant. These results identify LAMP-2 as a DC receptor for antigen and show that antigen internalised by it re-routed and presented to T cells selectively by EV rather than as antigen-derived peptides bound to HLA class II molecules on the cell surface. The antigens normally presented in this way have yet to be identified but may include MPO if initial experiments are confirmed, and the demonstration that LAMP-2 as a cellular receptor for FimH in Task 1 is also highly relevant.
Identification and characterisation of genes that influence susceptibility to AAV Partner 07 (UCAM) analysed patient cohorts collected by the European Genetics of Vasculitis Consortium which was originally established by Partners 07 (UCAM) and 01 (MUW) to identify genetic loci that influence susceptibility to AAV. Genome wide data was available from a UK discovery cohort of 1,233 patients with AAV and 5,884 UK controls and a replication cohort from Continental Northern Europe consisting of 1,454 additional patients and 1,666 controls. Three genetic loci, were HLA-DP, PRTN3 and SERPINA1, were associated with AAV at levels of genome-wide significance (p<5X10-8) together with a number of other promising candidates that did not reach the required level of significance. PRTN3 and SERPINA1 are the genes that encode PR3 and its major inhibitor a1-anti-trypsin, and were exclusively associated with individuals with AAV who had PR3-ANCA, as was HLA-DP. Separate analysis of individuals with MPO and PR3 ANCA revealed that those with autoantibodies to MPO were associated with HLA-DQ. The genetic loci were fine mapped using a custom made Immunochip designed specifically autoimmune and inflammatory diseases which enabled us to impute the HLA-DP alleles associated with the PR3-ANCA. This confirmed the previous association with HLA-DP0401 but in addition identified two novel alleles that conferred dominant resistance to the disease. Further analysis identified three residues within the HLA-DP1*0401 allele that are critical for susceptibility. Similar analysis of the HLA-DQ* A1 locus associated with MPO-ANCA identified one susceptibility allele and one that conferred dominant protection. Similar approaches identified two discrete susceptibility regions in the non-coding region of the PRTN3; and two separate susceptibility regions in the SERPINA1 gene, again located in non-coding regions. This provides the foundation for studies analyzing the physiological consequences of these non-coding genetic variants in PR3 ANCA-associated AAV. These data provide the first independent evidence that autoantibodies to PR3 are directly involved in pathogenesis and not simply an epiphenomena. Equally importantly, they demonstrate that the two commonest forms of AAV, granulomatous polyangiitis (GPA) and microscopic polyangiitis (MPA), are genetically distinct and that the primary genetic association is with the autoantibody specificity rather than the clinical syndrome. The initial results were published in the New England Journal of Medicine in July 2012 (Lyons et al NEJM. 267: 214-223) and immediately established the European group as the world leaders in the genetics of AAV.
In conclusion, the results
1. Generated a unique resource of fluorescently tagged E. coli strains expressing normal or mutated FimH and library of uropathogenic E. coli found in patients with AAV and control subjects
2. Identified LAMP-2 as receptor for FimH
3. Dissected the effects of autoantibody ligation on cellular function and translation of both findings into the development of novel therapies is currently explored.
4. Confirmed host susceptibility in AAV with autoantibodies to PR3 and MPO are associated with different HLA class II polymorphisms, whilst the PTEN and SEPINA1 loci - that respectively encode PR3 and its major inhibitor 1-anti-trypsin - are uniquely associated anti-PR3 associated disease. This provides the first evidence that the two disorders are genetically distinct and established INTRICATE researchers as world leaders in the genetics of AAV.
Potential Impact:
Socio-economic impact and the wider societal implications of the project
Contribution to Community and social objectives
Impact on the general population
The partners of INTRICATE aimed to disseminate the results of their research not only to researchers at scientific conferences but actively informed through their connections with patients’ representatives and by providing information for patients on INTRICATE web page.
Impact on resources available for future studies
In addition to its research output INTRICATE research has generated a unique set of resources available to the investigators and collaborators for the investigation of AAV.
Thus INTRICATE clinical collectors have assemble a large and unique cohort of patients, controls and their samples, established microbiological libraries of pathogens from patients with AAV and controls, generated research tools, like assays and peptide libraries through the engagement of SMEs.
Impact on clinical practice
The important findings from all work packages has already resulted in the extension of autoantibody profiling in AAV to novel antigens and the use of newly developed assays that promise will guide to development of more targeted therapies: S. aureus nasal isolates should be frequently tested for co-trimoxazole resistance if co-trimoxazole is used to eliminate S. aureus carriage and antibiotic therapy should be either changed when resistance develops or preemtively changed periodically. In order to improve the humoral response against S. aureus vaccination should be considered and effective vaccines developed.
Impact on Europe-wide and international collaborations
INTRICATE research has received international recognition and sparked several new collaborations that strengthen the European and international research in AAV. Thus partner 01 MUW has collaborated with investigators at East & North Herts NHS Trust, UK and Charles University Prague, Czech republic to build the large INTRICATE patient cohort. Partner 06 UMCG collaborated with the department of microbiology UMCG and with the department of microbiology Erasmus medical center Rotterdam and thus ensured success of WP02. Partner 08 Stanford and partner 09 EMC Prof Karl-Heinz Wiesmüller initiated collaborations with Science for Life Laboratory Stockholm, the Institute of Immunology, University of Veterinary Medicine Vienna and the Department of Neurology, University of Regensburg to extend the generation and use of microarray platforms and recombinant peptide methodology in cellular immunology and autoimmunre research. Partner 12 HMGU established new collaborations with Wroclaw University of Technology, Faculty of Chemistry, Division of Medicinal Chemistry and Microbiology in Poland, the Institute of Lung Biology and Disease (iLBD), Neuherberg, Germany to optimize the generation of synthetic protease substrates and inhibitors and models of lung injury and intranasal instillation of elastase.
Impact on training
INTRICATE partners not only trained over 20 researcher at all levels (PhD and postdoc level) in the methods of translation medicine, the project also allowed exchange of researchers an three fellowships had been successfully concluded.
Impact on women in science
The INTRICATE consortium is composed of 10 scientific partners coordinated by a female coordinator. Two partners are led by female principle investigators and the percentage of female staff members in all research groups involved in the proposed projects and already in employment is 28% at advanced academic level.
These figures indicate that the INTRICATE partners already have a strong commitment and INTRICATE put measures in place to promote women in science. With recruitment of equal numbers of 6 female and 6 male Phd students and 10 female and 10 male postdocs the INTRICATE consortium’s has met its goal to increase female researchers at all levels to at least 40% (preferred 50%).
Main dissemination activities and exploitation of results
Resulted in the publication of 48 original articles in high ranking biomedical journals , 5 of those in top medical Journals (Nature Medicine, Nature communications, Nature Reviews Immunology, Cell, NEJM) and several publications have received broad attention and were widely publicised also to the lay public:
Perera NC, Schilling O, Kittel H, Back W, Kremmer E, Jenne DE. NSP4, an elastase-related protease in human neutrophils with arginine specificity. Proc Natl Acad Sci U S A. 2012 Apr 17;109(16):6229-34.
http://www.neuro.mpg.de/133434/news_publication_5589778?c=22167
Perera NC, Jenne DE. Perspectives and potential roles for the newly discovered NSP4 in the immune system. Expert Rev Clin Immunol. 2012 Aug;8(6):501-3.
Hinkofer LC, Seidel SA, Korkmaz B, Silva F, Hummel AM, Braun D, Jenne DE, Specks U. A monoclonal antibody (MCPR3-7) interfering with the activity of proteinase 3 by an allosteric mechanism. J Biol Chem. 2013 Sep 13;288(37):26635-48.
http://www.biomedexperts.com/Abstract.bme/23902773/A_monoclonal_antibody_MCPR3-7_interfering_with_the_activity_of_proteinase_3_by_an_allosteric_mechanism
Schiwon M, Weisheit C, Franken L, Gutweiler S, Dixit A, Meyer-Schwesinger C, Pohl JM, Maurice NJ, Thiebes S, Lorenz K, Quast T, Fuhrmann M, Baumgarten G, Lohse MJ, Opdenakker G, Bernhagen J, Bucala R, Panzer U, Kolanus W, Gröne HJ, Garbi N, Kastenmüller W, Knolle PA, Kurts C, Engel DR. Crosstalk between sentinel and helper macrophages permits neutrophil migration into infected uroepithelium. Cell. 2014 Jan 30;156(3):456-68.
http://www3.uni-bonn.de/Press-releases/immune-cells-need-a-second-opinion
Tadema H, Abdulahad WH, Lepse N, Stegeman CA, Kallenberg CG, Heeringa P. Bacterial DNA Motifs Trigger ANCA
Production in ANCA-Associated Vasculitis in Remission Rheumatology (Oxford). 2011;50:689-696.
http://www.medscape.com/viewarticle/740011
Outlook and future research
INTRICATE legacy
The results of the INTRICATE research will not only result in additional publications of the work that had been recently concluded during the last period of funding but will impact also on future research. Thus INTRICATE research has made a major impact on research in AAV and autoimmunity more generally. The methods and approaches adopted have been adopted to investigating other autoimmune disease like SLE, Rheumatoid arthritis and Crohn’s disease by either INTRICATE investigators alone or in – often newly established – collaborations. These have been facilitated by the availability of existing INTRICATE resources, like samples from well characterised patients, research reagents and assays developed by INTRICATE’s SME partners and resulted in novel sources of funding and grant income.
List of Websites:
www.intricate.eu
Ao.Prof. PhD. Dr. Renate Kain
Clinical Department of Pathology
Medical University Vienna
Währinger Gürtel 18-20
A-1090 Vienna
Austria
Tel: 0043-1-40400-3681
Fax: 0043-1-40400-5193
email: renate.kain@meduniwien.ac.at
INTRICATE is a multidisciplinary consortium comprised of scientists and clinical investigators who joined together to develop a deeper insight into why infection predisposes to autoimmune disease. Autoimmune injury in rodents can be triggered by molecular mimicry with microbial proteins, non-specific activation of auto-reactive lymphocytes by bystander activation or bacterial superantigens, and by activation of innate immunity by LPS or other pathogen associated molecular patterns (PAMPs); but little is known about the involvement of these mechanisms in human autoimmune disease. INTRICATE addressed this problem by systematically analysing a single candidate disease, ANCA-associated vasculitis (AAV), a severe disease associated with autoantibodies specific for the neutrophil enzymes myeloperoxidase (MPO) and proteinase 3 (PR3) and for lysosome membrane associated membrane protein 2 (LAMP-2). AAV is a model disease because its known association with bacterial infection: specifically chronic dysbiotic expansion of S. aureus in upper airways is associated with relapse in AAV associated with autoantibodies to PR3; and autoantibodies to LAMP-2 recognise an epitope common to the bacterial adhesion FimH with which they cross-react. The work was separated into four interlocking Work packages:
Work Package 01 developed novel antigen arrays suitable for high-throughput screening for serum antibodies and applied them to prospectively assembled cohorts of individuals with AAV and controls. The results establish the ability of this approach for large-scale characterisation of autoantibodies and the ability of hierarchical clustering to segregate patients into groups. The analysis of the prospective cohorts demonstrated that antibodies to a range of staphylococcal proteins are rare in AAV even in those with dysbiotic colonisation. Conversely, antibodies specific for FimH are significantly more common in AAV than in healthy controls but their concentration is not influenced by urinary tract infections.
Work Package 02 characterised the diversity S. aureus in AAV and characterised their antigen independent effects on autoantibody generation. The results identified a high degree of genetic diversity in S. aureus strains from individuals with GPA, albeit with the emergence of predominant spa-types at particular times, that did not differ significantly from the healthy population; discovered antibodies to numerous S. aureus proteins in patients with GPA; and demonstrated PR3-specific IgD and IgA in patients with GPA which suggests a vigorous mucosal immune response.
Work Package 03 generated and characterised unique mouse strains that express human MPO or PR3 in instead of their murine orthologues; and has used them to determine how the fine specificity of antibodies to them influences the injury they cause, and whether infection provokes autoimmunity to them. Antibodies to human MPO or human PR3 in the circulation enhanced the inflammatory response after intra-tracheal instillation of LPS and the results demonstrate that antibody isotype is a much greater determinant than fine specificity and in particular that injury is not influenced by whether or not the antibodies interfere with enzymatic activity. Consistent with this latter notion was the clinical finding that the epitope-dependent inhibitory profile of ANCA towards active PR3 did not correlated with disease manifestations, disease severity or onset of disease. Infections with fimbriated bacteria did not elicit ANCA against the endogenous mouse LAMP-2 sequence and even recombinant strains of fimbriated E. coli expressing LAMP-2-related sequences failed to trigger a cross-reactive immune response to the endogenous murine LAMP-2 sequence confirming firm mechanisms of maintaining to tolerance to self antigens under physiological conditions.
Work Package 04 explored the interactions between FimH, LAMP-2 and antibodies to them together with host genetic factors that influence susceptibility to AAV. A unique panel of fluorescently tagged E. coli strains expressing normal or mutated FimH was generated and used to show that E. coli adhere more efficiently to microvascular endothelium than to bladder epithelium, and to identify specific amino acid residues responsible the difference. LAMP-2 was then identified as a previously unsuspected receptor for FimH on the surface of endothelial cells where it is also accessible to anti-LAMP-2 antibodies. Once ligated by specific antibodies, cell surface LAMP-2 is rapidly internalised with major consequences for cell function, including antigen presentation. Host susceptibility was examined by genome wide association study (GWAS) that revealed that susceptibility to AAV with autoantibodies to PR3 and MPO are associated with different HLA class II polymorphisms, whilst the PTEN and SEPINA1 loci - that respectively encode PR3 and its major inhibitor 1-anti-trypsin - are uniquely associated anti-PR3 associated disease. This provides the first evidence that the two disorders are genetically distinct.
INTRICATE has succeeded in providing new insights into the links between infection and AAV as well as a wealth of novel clinical and experimental resources that will be available to the community.
Project Context and Objectives:
Background and Aims
INTRICATE is a multidisciplinary group of scientists, clinical investigators, pathologists, (clinical) immunologists, world-class geneticists, microbiologists, developmental biologists, cell biologists and bio-informaticians each of whom has made a substantial individual contribution to understanding the links between infection and autoimmunity. They have collaborated extensively in the past but join now together to form a Consortium specifically constructed to meet the aims of FP7- HEALTH.2010.2.4.5-2: Infection and dysbiosis as the triggers of the development of inflammatory processes in allergies and autoimmune diseases. In order to maintain focus the INTRICATE project investigated autoimmunity because, despite common initial pathways, allergy and autoimmunity engage different effector pathways - resulting in immunity to self rather than foreign antigens. The in-depth analysis of those mechanisms forms an integral part of the project. The aim of the INTRICATE consortium is to prosecute a programme of Translational Research that delineates the role of infection in the induction and perpetuation of severe systemic autoimmune disease with the ultimate objective of identifying new therapeutic strategies based on knowledge of pathogenesis.
Autoimmune disease is a severe cause of morbidity whose prevalence in the Western world and has increased relentlessly over the past fifty years. This trend has been especially of late and epidemiological studies from Europe and United States give the prevalence as 8-10% compared to 3% ten years ago [Jacobson DL, 1997; http://www3.niaid.nih.gov/topics/autoimmune/]. This change has been attributed directly to the pace of societal change over the last century; for example to the consequences of improved housing and vaccination policies that have virtually eliminated the major childhood infectious diseases, and to the use of antibiotics that cut short the effects of common bacterial infections. The result has been an unprecedented pressure on the balance between the human immune system and the bacterial universe with which it has co-evolved over millions of years. This led to the proposal that one consequence of the reduced infective load was the increased predisposition to allergic and autoimmune disease – this is the "Hygiene Hypothesis" [Strachan DP et al. 1989; Strachan DP. 2000] – however in allergy and autoimmunity common initial pathways result in entirely different effector mechanism. These ideas stimulated a general reconsideration of the relation between the microorganisms and the immune system not only with regard to the long-term consequences of acute infection but also the underlying role of commensal bacteria in shaping the immune system [Sansonetti PJ. and Medzhitov R. 2009].
The INTRICATE proposal is based on the premise that research into complex human disorders such as autoimmune disease must be undertaken primarily using samples and tissue taken from human subjects. Accordingly we investigated the interaction between infection and autoimmunity in a “model” human disease: Anti-neutrophil cytoplasmic antibody (ANCA) associated systemic vasculitis (AAV). AAV is ideally suited because it is known to be caused by autoantibodies of defined specificity and second because it is strongly linked to infection. AAV is a clinically well defined autoimmune disease that destroys the small blood vessels and most of those affected fall into one of two broad clinical groups: those with Granulomatosis with Polyangitis (GPA) in which the systemic vasculitis is accompanied by (and often preceded by) granulomas involving the nasal sinuses and upper and lower airways often against a background of chronic sinus disease; and microscopic polyangiitis (MPA) which has neither granulomas nor upper airway involvement. Most other aspects are similar in both groups including involvement of the kidney to cause pauci-immune focal necrotizing glomerulonephritis (piFNGN) with rapid irreversible kidney failure and the lung associated with severe haemorrhage. There is general agreement that the disease is caused by ANCA that bind the soluble lysosomal enzymes myeloperoxidase (MPO) and proteinase 3 (PR3) or the more recently defined target, lysosomal associated membrane protein-2 (LAMP-2) [Jennette JC et al. 2006; Kallenberg CG. 2006; Kain R et al. 2008]. INTRICATE uses mouse models but only to answer the specific question whether infection with clinically relevant bacteria induces autoimmune disease in transgenic mice that express the human autoantigen. Our concentration on human samples and “humanised” mouse models should mean that we are be able to translate the results swiftly into the clinic.
The INTRICATE project had four specific aims, namely to:
• Use novel high-throughput antigen array technology and well-characterized patient cohorts to determine whether acute infection with specific microorganisms triggers the induction or re-activation of AAV; and to ascertain whether antibody responses to microbial proteins cross-react with native or epigenetically modified self-proteins
• Elucidate the reasons why dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised and systemic granulomatous vasculitis in AAV; and in particular to analyse the roles of microbial superantigens and the local adaptive immune response to them.
• Analyse the mechanisms of molecular mimicry in transgenic mice expressing the human forms of LAMP-2, PR3 and MPO – the major targets of autoantibodies in AAV – by determining whether infection with bacteria that express molecular mimics induce AAV and, if so, to define under which circumstances they do so.
• Characterise disease-associated genes identified in the European Vasculitis Genetics Consortium's genome wide association study (GWAS) of AAV and to examine their effect on gene expression and function; and to determine whether the genetic variants that predispose to AAV have been maintained in the gene pool because of a beneficial effect on resistance to infection.
Work strategy and general description
The INTRICATE Consortium consists of a multidisciplinary consortium of scientists and clinical investigators with diverse expertise who have joined together to elucidate the reasons why infection predisposes to autoimmune disease and to analyse of the links between bacterial infection and the development of ANCA-associated vasculitis (AAV) as a model autoimmune disease more generally. AAV was chosen because of its long-recognised links with bacterial infection; and because its pathogenesis sufficiently well understood. The proposed programme is based on: (i) the secure knowledge from epidemiological studies that infection is a common trigger for autoimmune disease in man; (ii) the results of rodent models of autoimmunity that establish at least four non-exclusive potential triggering pathways including molecular mimicry, bystander activation, superantigens, and activation of innate immunity; and (iii) current understanding of the immune system.
The immunological mechanisms that lead from infection or dysbiosis to autoimmunity and the processes involved are both diverse and complex and consequently the INTRICATE research analysed them systematically using a „model disease“. We have selected ANCA associated systemic vasculitis because it is a common autoimmune disease, the nature of the autoimmune response involved is well characterised and there are clear links between AAV and specific bacterial pathogens. AAV is therefore a uniquely tractable disease in which to analyse infectious triggers in the development of inflammatory processes in autoimmune disease.
The INTRICATE research generated new knowledge in four inter-dependent areas – represented in the four scientific Work packages – with the ultimate goal to translate the results into clinically applicable tools to predict, intersect and monitor processes that lead infection to trigger autoimmune disease. Each of the four complementary work packages focused on a different host and pathogen mechanisms that could be responsible for loss of tolerance to self-antigens and the development of tissue injury.
Work package 01 Infection and Dysbiosis in clinical AAV has two main objectives: first to develop novel antigen arrays suitable for high-throughput screening for serum antibodies specific for self-antigens or bacterial proteins in AAV; and second to assemble cohorts of prospectively followed patients with AAV suitable to determine the frequency of antibodies detected and their relation to disease activity.
Work package 02 S. aureus diversity and antigen independent effects in immune responses addresses the reasons for the well-characterized epidemiological association between dysbiotic nasal carriage of S.aureus and clinical relapse in granulomatosis with polyangiitis (GPA); and in particular to analyse the roles of microbial superantigens and the local adaptive immune response to them.
Work package 03 Chronic infection and Dysbiosis in murine models of AAV is generating transgenic mouse lines that express the human MPO, PR3 and LAMP-2, the targets of autoantibodies, to study pathogenesis and molecular mimicry in AAV; and developing mouse models of urinary tract infections.
Work package 04 Host-pathogen interactions in generation of AAV has three discrete objectives: to determine how genetic variation of FimH – an adhesin and major virulence factor of gram negative pathogens that commonly cause urinary tract infections (UTI) - influences its interaction with LAMP-2, its molecular mimic; to determine whether uptake of antibodies to LAMP-2 influence antigen processing and presentation; and finally to use genome wide association studies to identify genes that influence susceptibility to AAV.
Management structure and procedures
The Project Coordinator ensured the smooth operation of the project and guaranteed that all efforts were focused towards the objectives. She submitted all required progress reports, deliverables, financial statements to the European Commission, and, with the assistance of GABO:mi she was responsible for the proper use of funds and their transfers to participants. The INTRICATE office was established by and based at the coordinator in Medizinische Universitaet Wien and at GABO:mi in Munich. The Project Office at the Coordinator was concerned with the scientific management and the co-ordination of all research activities. The Project Office at GABO:mi was responsible for administrative, financial and contractual management and the organisational co-ordination of the project activities.
The Project General Assembly (GA) was in charge of the political and strategic orientation of the project and acted as the arbitration body. It met once a year unless the interest of the project required intermediate meetings. The Project Steering Committee (STC) consisted of all work package leaders and the Coordinator and was in charge of monitoring all activities towards the objective of the project in order to deliver as promised, in due time and in the budget. The Project Steering Committee met every six months during the funding period. Furthermore, a scientific advisory board (SAB) was implemented to ensure a high standard of research and monitor the progress of the project by taking part in the annual General Assembly.
Project website: www.intricate.eu
The News section of INTRICATE website contains the publishable summaries, several press releases and also the video with Prof. Kain.
This video was already in social media channels (Facebook, Twitter, Google+, LinkedIn) announced.
Project Meetings:
Type of meeting Date Venue
Steering Committee Meeting 30th November 2010 Vienna, Austria (INTRICATE kick off meeting)
WP01, WP02 Investigator Meeting 16th May 2011 Chapel Hill (at 15th international Vasculitis & ANCA Workshop)
Steering Committee Meeting 4th May 2011 Telephone conference
Steering Committee Meeting 22nd September 2011 Cambridge, UK (INTRICATE 2nd general assembly)
Steering Committee Meeting 28-29th October 2012 Stanford, USA (INTRICATE 3rd general assembly)
WP01 Clinical Investigator Meeting 17th April 2013 Paris, France (at 16th international Vasculitis & ANCA Workshop)
Steering Committee and 4th GA Meeting 13-15th October 2013 Munich Germany (INTRICATE 4th general assembly)
Steering Committee Meeting 14th January 2014 Telephone conference
Steering Committee and 5th GA Meeting 3-5th September 2014 Vienna, Austria
Steering Committee Meeting 3rd September 2014 Vienna, Austria
WP01 Clinial Investigator Meeting 13th November 2014 Philadelphia, USA (at the American Society Nephrology renal week meeting)
Final Steering Committee and GA Meeting 17-18th April 2015 Cambridge, UK
WP Leader meetings via TCs regularly Telephone conferences
WP meetings via TCs regularly Telephone conferences
Objectives of INTRICATE:
The objectives of the four INTRICATE work packages are defined in its specific tasks:
WP01 uses a novel high-throughput technology and well-characterized patient cohorts to profile the serum antibodies to test the hypothesis that acute infection with specific microorganisms triggers the development or reacti¬vation of ANCA-associated systemic vasculitis; and to ascertain whether antibody responses to individual microbial proteins cor¬relate with those to native or epigenetically modified self-proteins from the tissues affected in vasculitis. Specifically, its objectives are to
1. Assemble the required cohorts of prospectively followed patients with AAV, infection and relevant controls,
2. Synthesize and utilize antigen microarrays and combinatorial peptide libraries designed to identify infections associated with AASV,
3. Characterize bacterial proteins expressed by bacteria associated with AAV,
4. Develop tools to monitor antibodies to bacterial peptides and proteins in AAV.
WP02 investigates the reasons why the dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised granulomatous inflammation in Wegener’s Granulomatosis that often presages the development of systemic disease. The overall purpose of this work package is to elucidate possible mechanisms that explain their localisation or association with disease. Specifically it:
1. Analyses the genetic diversity of S. aureus strains isolated from patients with AAV to ascertain whether they have particular properties,
2. Ascertains whether staphlococcal proteins induce lymphocyte and neutrophil responses likely to induce or maintain injury in AAV,
3. Analyses the local B and T cell responses to S. aureus in the upper airways of patients with AAV.
WP03 aims to overcome one of the main barriers limiting analysis the role of molecular mimicry in loss of tolerance to self-proteins human autoimmune disease, by creating transgenic mouse lines that express the human forms of LAMP-2, PR3 and MPO – the major targets of autoantibodies in AAV. These mice will then be used to analyse the conditions required to induce autoimmunity by molecular mimicry. Its specific aims are to:
1. Generate transgenic mouse lines expressing human LAMP-2, PR3 and MPO and to analyse the consequences of immunity to them,
2. Ascertain whether infections with fimbriated bacteria or S. aureus induce ANCA and whether they induce autoimmune disease,
3. Produce natively folded LAMP-2 and PR3 in order to analyse mimicry with three dimensional epitopes rather that linear.
The aim of WP 04 is to examine what determines and influences the development of autoimmune disease when a genetically susceptible individual is exposed to a sufficiently powerful environmental challenge using FimH/LAMP-2 molecular mimicry in AAV as a model. It combines bacterial genetic studies, cell biology with data from the first genome wide association study (GWAS) in AAV three specific aims which are to:
1. Ascertain how genetic difference in FimH influences its uptake by cells and the resulting effects,
2. Determine whether antibodies to FimH/LAMP-2 affect presentation of PR3 and MPO antigens to T cells,
3. Use GWAS data to identify immune-related genes that influence susceptibility to AAV or infection.
Project Results:
WP01 Infection and dysbiosis in clinical AAV
Work package 01’s main objectives were to develop novel antigen arrays suitable for high-throughput screening for serum antibodies specific for self-antigens or bacterial proteins in AAV; and to assemble cohorts of prospectively followed patients with AAV suitable to determine the frequency of antibodies detected and their relation to disease activity.
Work completed includes the (i) generation of libraries of purified, synthetic and recombinant proteins for array development, (ii) generation of novel antigen arrays for large-scale antibody screening – including a novel protein microarray for discovery of anti-cytokine antibodies in AAV - and the (iii) assembly of large prospectively followed patient cohorts and their serum samples that were applied to the arrays and analysed by necessary bioinformatics approaches.
GENERATION AND ANALYSIS OF NOVEL MICROARRAYS
INTEL Microarray Partner 08 (Stanford), in collaboration with Intel Inc., developed two silicon-based peptide microarrays featuring sequences from the known autoantigens histone H2B and U1-70K, respectively. The peptides were synthesized using maskless photolithography on the surface of silicon wafers. Each microarray contained every possible overlapping peptide within linear protein sequences covering the N-terminal tail of human histone H2B or the RNA binding domain of U1-70K. The H2B microarrays included peptides with acetyl and methyl modified lysine residues (as well as unmodified control peptides). These modifications are observed in vivo, and are hypothesized to be involved in loss of tolerance to the H2B antigen. Using the H2B microarrays allowed to identify the peptide epitopes of serum IgG autoantibodies in individuals with systemic lupus erythematosus. The U1-70K microarrays were used to identify a minimal epitope (116-121) that was cross-reactive with a known epitope in the M1 protein of influenza B viruses. Molecular mimicry has been hypothesized as a mechanism of development of autoantibodies to RNP (a protein/RNA complex that includes U1-70K). A recently developed technique to develop next generation microarrays on the surface of silicon wafers with an underlying integrated circuit allows, by combining these new microarrays with magnetic detection methods, to perform real-time measurement of antibody binding.
Histone microarrays Partner 08 (Stanford) developed peptide microarrays, featuring histone pepitdes derived from H2A, H2B, H3 and H4, for profiling histone post-translational modifications (PTMs) of neutrophil extracellular traps (NETs). Modifications included mono-, di- and tri-methylated lysine, citrulline, acetyl-lysine, and phosphorylation. Antibodies to histones are observed in several autoimmune disorders, including systemic lupus erythematosus (SLE) and the work elucidate the hypothesis that specific histone PTMs within NETs are involved in loss of tolerance and induction of histone autoantibodies. The results showed that SLE patients with reactivity to histones had serum antibodies that bound acetyl-modified histone H2B and other histone PTMs. Further investigation confirmed that distinct histone PTMs - indicative of transcriptionally silent DNA - are present in NETs using high-content immunoblotting. Immunization of Balb/c mice with NETs induced a weak autoantibody response, with partial overlap in autoantibody profiles to those seen in patients with SLE.
Cytokine autoantibody microarrays Recently, anti-cytokine autoantibodies have been found to cause increased susceptibility to infections that are not typically seen in the general population. These include anti-IFNg autoantibodies in patients with diffuse mycobacterial infections, and anti-Th17-associated cytokine autoantibodies in APECED patients with chronic mucocutaneous candidiasis. Partner 08 (Stanford) developed a novel protein microarray for the high-throughput discovery of anti-cytokine autoantibodies in autoimmune diseases and immunologic deficiencies. The microarray was customised to address the specific question whether anti-cytokine autoantibodies could be responsible for the increased prevalence of Staphylococcus aureus (SA) infections observed in AAV patients and consists of 65 commercially available recombinant cytokines, chemokines and immune mediators - including IL8, TNF, IL1b and MIP1a, that are produced by neutrophils - and 34 well-characterized autoantigens, including ANCA proteins MPO, PR3 and LAMP2. The array was used to analyse whether tolerance to these self-proteins could have also been broken in parallel with or subsequent to known ANCA proteins, and represent previously unidentified autoantigens in AAV. Beyond this targeted screen of neutrophil-associated cytokines and chemokines, a complimentary, unbiased approach by including a large number of theoretically unreleated cytokines and chemokines was included on the microarray.
In collaboration with partner 07 (UCAM), Rainer Doffinger and Ken Smith, partner 08 evaluated over 50 serum samples from patients with a variety of immunodeficiencies, including a recently identified patient population that suffers from diffuse mycobacterial infections due to anti-IFNg autoantibodies, using the anti-cytokine microarrays. Hierarchical clustering of the patient samples (blinded) based on their reactivity patterns, identified two distinct patient groups, or clusters. The first cluster had high reactivity to GMCSF, which has been previously reported in pulmonary alveolar proteinosis (PAP), while the second had high reactivity to type I IFN and Th17-associated cytokines (IL17F, IL22), which have been previously associated with APECED (autoimmune polyendocrine syndrome type I). Unblinding revealed that the results had identified patients with PAP and APECED with 100% accuracy, demonstrating that cytokine microarrays were capable of identifying anti-cytokine autoantibodies in immunodeficiencies.
Autoantigen microarrays Nitrocellulose 16-pad microarray slides were printed with 27 known or suspected AAV antigens and controls in triplicate using a robotic arrayer. ANCA-associated Vasculitis (AAV) serum, plasmapheresis samples and controls provided by INTRICATE partners were diluted and used to probe the microarrays in duplicate, followed by a fluorescently-labeled anti-human IgG secondary. Microarrays were scanned using a fluorescent microarray scanner, and the mean fluorescent intensity (MFI) of each protein feature was used as a measure of antibody binding.
In total, 127 samples from 5 INTRICATE clinical investigators were analyzed using autoantigen microarrays in the laboratory of partner 08. AAV patients and controls who were selected because likely to be informative have been assayed in a series of INTRICATE arrays. This work has provided an insight into the antibody responses in AAV that can be translated into the development of novel biomarkers.
The centers were: partner 01 (MUW, Kain), partner 05 (Mayo, Specks), partner 06 (UMCG, Kallenberg), partner 07 (UCAM, Smith/Lyons) and partner 08 (Stanford, Utz). Partner 08 was unblinded to the diagnosis and clinical features of the collaborator samples after circulating the coded microarray results to all collaborators.
Hierarchical clustering successfully divided patients samples based on whether they had p or c-ANCA. FimH reactivity was observed in a number of patient samples and did not appear to be associated with p or c-ANCA. Interestingly, reactivity to partner 08’s commercial preparation of LAMP-2 was observed in patient samples from both p and c-ANCA groups, while little reactivity was observed to the other LAMP-2 preparations. Reactivity to the less common ANCA antigen azurocidin was observed in a cluster of c-ANCA positive patient samples that were positive for PR3 and HNE. A second cluster of patient samples has reactivity to PR3, but lacks reactivity to the S195A mutated version of PR3. This suggests that their binding to the PR3 antigen may be dependent on this residue.
In a parallel analysis, epoxy-coated microarray slides were printed with 30 known or suspected SLE antigens, AAV antigens, and controls proteins in triplicate using a robotic arrayer. Serum samples derived from SLE patients that have been characterized by SNP typing, gene expression analysis, and flow cytometry, were provided by partner 07 (UCAM, Smith/Lyons).
Microbial microarrays Partner 01 (MUW) used 19 selected serum samples, one at untreated disease presentation and one during remission each, from nine patients with AAV and one healthy control to investigate IgG and IgM antibody responses to microbial peptide antigens in a customized, commercial peptide array that contained 4,347 linear B-cell epitopes (max. 16 aa length) derived from 159 pathogens associated with infectious diseases and control peptides. Analysis to date showed that, both IgG and IgM responses to microbial antigens in general were significantly higher in patients with active disease compared to remission and that antibodies from patients with active disease reacted to a broader range of antigens indicative of epitope spreading.
INTRICATE COHORT
In total, 328 patients with AAV, 50 patients with UTIs and 60 healthy controls were collected to the INTRICATE cohort. Their serum, blood and DNA samples were, in addition to samples of 170 patients with AAV from the INTRICATE collection, available for the studies proposed. Recruitment of patients and collection of their samples thus provides a biorepository of over 1000 serum samples and DNA as well as microbial collections of 105 urinary isolates and 271 nasal isolates from clinically well characterised patients and controls that were available for the INTRICATE research. The clinical data from patients recruited at partner 01 have already been entered into the INTRICATE registry at MUW: http://cemsiis-akim.meduniwien.ac.at/plattformen/rda/projekte/intricate-register/.
In conclusion, the results from WP01 have resulted in the
1. Generation of novel resources (INTRICATE arrays) that provide at the same time a diverse but specific platform for the analysis of antibody responses in large patient cohorts.
2. Assembly of the INTRICATE cohort that provides a uniquely large cohort of over 300 patients with AAV and controls and their serum samples and DNA as well as microbial collections
3. Prove of concept for the use of microarrays for antibody profiling
4. Provision of novel insights into antibody responses in sera from patients with AAV, SLE, autoimmune inflammatory disorder and infection that can be translated to other autoimmune diseases
WP02: Antigen independent effects of S. aureus on immune responses in AAV
Work package 02 addresses the reasons for the well-characterized epidemiological association between nasal carriage of S. aureus and clinical relapse in granulomatosis with polyangiitis (GPA). The work aims at elucidating the reasons why dysbiotic expansion of S. aureus in nasal sinuses and upper airways is linked to localised and systemic granulomatous vasculitis in AAV. In particular the work analysed the roles of microbial superantigens and the local adaptive immune response to them. The results identified (i) a high degree of genetic diversity in S. aureus strains from individuals with GPA, albeit with the emergence of predominant spa-types at particular times; (ii) discovered antibodies to numerous S. aureus proteins in patients with GPA; and (iii) demonstrated PR3-specific IgD and IgA in patients with GPA which demonstrates a vigorous mucosal immune response.
Dysbiotic expansion of Staphylococcus aureus (S. aureus) in the upper airways occurs in about 60% of patients with Granulomatosis with Polyangiitis (GPA, formerly Wegener’s).This contrasts with the occurrence of chronic nasal carriage in only 20% of healthy controls. The persistence of S. aureus in the nasal cavity in GPA patients is associated with recurrent relapses of the disease.
S. aureus genetic diversity In this part of project partner 06 (UMCG) first analyzed the genetic diversity of S. aureus in patients with ANCA-associated vasculitis (AAV). The work investigated the hypothesis that particular strains of S. aureus could be related to the effect of S. aureus carriage on relapsing disease. The results from testing 210 isolates from the nasal cavity from 71 GPA patients in multiple techniques showed that the S. aureus population structure in GPA patients proved highly diverse without showing a particular subset being dominant and associated with relapsing disease. In fact, these isolates did not differ in the genetic diversity of S. aureus from isolates from healthy controls. The S. aureus population structure changed over time. Interestingly, increased resistance to co-trimoxazole and ciprofloxacin developed over time in the isolates from the GPA patients and this increased resistance seemed to coincide with the increased treatment of patients with co-trimoxazole. Co-trimoxazole resistant isolates were only obtained from patients treated with this antibiotic, and co-trimoxazole resistance was not observed for isolates from healthy controls.
Humoral and cellular responses to S. aureus antigen Next, partner 06 (UMCG) evaluated in depth the humoral response against 59 S. aureus antigens. The humoral antibody response against S. aureus appears deficient in GPA patients which may, at least in part, explain the persistence of S. aureus carriage in the patients. This seems to be a specific defect as the humoral response against influenza antigens was not disturbed in GPA patients.
Analysis of the effects of S. aureus derived stimulatory molecules on innate and adaptive immunity in AAV revealed that there appeared to be a shift towards inflammatory monocytes in GPA as a more pronounced decrease in classical monocytes was present in patients with GPA who were chronic carrier of S. aureus compared to non-carrier. This suggests that S. aureus carriage induces a shift in monocytes towards the pro-inflammatory type. No effect of S. aureus carriage was seen on functional capacities of neutrophils including the functional capacity of CD62L-low and CD16-high expressing neutrophils to downregulate T-cell proliferation.
Partner 06 (UMCG) also analyzed in depth the adaptive immunity in relation to S.aureus carriage and found that S. aureus carriage appears to stimulate an important subset of T-cells, namely follicular helper T-cells, in GPA patients. This subset drives the adaptive humoral immune response including the autoimmune response and activates macrophages, dendritic cells, NK-cells and NKT-cells. Interestingly, the Toxic Shock Syndrome Toxin-1 (TSST-1) from certain S. aureus strains appeared to have an amplifying role in autoantibody production.
Analysis of the link between IgD-positive B-cells and S. aureus in the upper airways demonstrated that IgD-ANCA do occur but are infrequently demonstrated in the peripheral blood in contrast to IgA-class ANCA. B-cell clusters are certainly present in the lesions in patients with GPA in the upper airways but it was not clear whether IgD-positive B-cells are related to carriage of S. aureus.
In conclusion, the results show that
1. The S. aureus population of patients with GPA is highly diverse and mirrors the general S. aureus population as seen in controls.
2. Chronic use of co-trimoxazole induces co-trimoxazole resistance and the immune response to multiple staphylococcal antigens is decreased in patients with GPA despite chronic exposure so explaining the persistence of S. aureus in the nasal cavity. Consequently, these important findings should guide to more targeted therapies: S. aureus nasal isolates should be frequently tested for co-trimoxazole resistance if co-trimoxazole is used to eliminate S. aureus carriage and antibiotic therapy should be either changed when resistance develops or preemtively changed periodically. In order to improve the humoral response against S. aureus vaccination should be considered and effective vaccines developed.
3. S. aureus carriage in patients with GPA mainly stimulates T-follicular helper cells, which are the primary effector memory T-cells. This finding implies that these cells could be specifically targeted in AAV/GPA. Development of therapeutic monoclonal antibodies to these cells should be considered and trials which such antibodies designed. Furthermore, the role of the TSST-1 superantigen of S. aureus should be taken into account. This means that S. aureus isolates should be tested for the presence of this superantigen. If present, elimination of these strains of S. aureus should be in the forefront of therapy.
4. S. aureus carriage induces pro-inflammatory monocytes. These results indicate that this cell population should be considered as targets of treatment. Presently, most attention is given to neutrophils but targeting monocytes could also be essential in the therapeutic approach in patients with AAV/GPA.
WP03 Analysis of molecular mimicry in murine models of AAV
Work package 03 is designed to analyse the mechanisms of molecular mimicry by determining whether infection with bacteria that express molecular mimics induce AAV and, if so, to define under which circumstances they do so. This is achieved by generating transgenic mouse lines that express the human MPO, PR3 and LAMP-2 – the major targets of autoantibodies in AAV – to study pathogenesis of AAV and molecular mimicry and by developing mouse models of urinary tract infections. As part of WP04 (i) mouse lines that express human PR3 or MPO and a mouse line that expresses human LAMP-2 driven by the Tie2 promoter/enhancer had been generated. (ii) Human PR3 and MPO expressing mouse lines and a murine model of pyelonphritis has been used to evaluate antibody responses and molecular mimicry in the development of AAV and (iii) newly generated, appropriately folded ANCA antigens were used to characterize the responses of antibodies to ANCA antigens in human.
Generation of humanized mouse lines expressing MPO, PR3 and LAMP-2 As for many autoimmune diseases, it is currently not understood how autoimmune responses are induced by external triggers. More than 90% of all autoantibodies are directed against distinct 3D structures of natively folded host proteins, called conformational epitopes. The 3D-surface of human proteins is highly divergent from those of commonly used rodents excluding the cross-reactivity of human autoantibodies with endogenous proteins of rodents. According to current theories, there are two principal ways in which external agents could induce autoimmunity. External components may coincidentally share structural, topological similarities with one of the host proteins or may elicit a strong immune response of antibody molecules which themselves invoke the production of anti-antibodies, as the antigen combining regions of these primary antibodies are also regarded as foreign. The foreign primary trigger in the latter case should be complementary to the host protein and should interact with it like the anti-idiotypic antibody. The major target self-antigens in autoantibody-associated vasculitis are PR3, MPO and LAMP2. The human LAMP2 protein (residues 41 to 49), but not the murine homolog, was reported to share a high sequence homology with the bacterial protein FIMH (residues 72-80) of type 1 fimbriae which are expressed by various bacterial species. WP03 therefore explored whether an experimentally induced bacterial infection of kidneys caused the formation of cross-reacting autoantibodies.
Partner 12 (HMGU) successfully established genomically modified knock-in mouse lines expressing the human PR3 and human MPO vasculitis antigens from the endogenous mouse promoters by homologous gene replacement. Both antigens were demonstrated to be correctly processed and functionally expressed in mouse neutrophils from bone marrow. The PR3-knock-in mice were distributed to partner 03 (Rfw-Bonn), the MPO knock-in mouse colony is currently being expanded in the animal facility of partner 12 (HMGU). A humanized LAMP-2-expressing transgenic mice could not be established during the funding period.
Pathogenicity of antibodies to ANCA antigens in humanized mouse lines Several monoclonal antibodies to human MPO (partner 06 UMCG and 10 HB) and newly produced (partner 12 HMGU) monoclonal antibodies to human PR3 were evaluated by intraperitoneal application in knock-in mice in combination with intratracheal instillation of low dose lipopolysaccharide (LPS) to see whether these antibodies are pathogenic under conditions of bronchial infections. Using the respective isotype controls we discovered that the pathogenic activity of mouse monoclonal antibodies to self-antigens of neutrophils was highly isotype dependent and not a matter of their binding specificity to a structurally distinct epitope. Two antigen-specific monoclonal antibodies of the Ig2a and Ig2b isotype class were found to enhance the low inflammatory effects of LPS in the lungs of mice suggesting that pathogenic effector functions of neutrophils, macrophages and dendritic cell are triggered by distinct interactions of certain IgG subclasses and activating Fc gamma receptors. In the future, the humanized knock-in mice will permit investigators to identify the Fc gamma receptors on major target cells that mediate the proinflammatory effects of autoantibodies under pauci-immune conditions of chronic inflammatory diseases.
Molecular mimicry in murine models of infection Alternative experimental approaches were chosen to study the pathogenic role of murine LAMP-2-FIMH cross-reacting antibodies in mice during experimentally induced pyelonephritis, e.g. after direct immunization with the murine LAMP-2-derived peptide AVPDKATHDGSSC. In a second approach, a mutant E. coli strain was engineered to achieve a local sequence identity between a mutated FIMH and the endogenous murine LAMP-2 protein. Although antibodies to murine LAMP-2 were generated after peptide immunization or in the course of experimentally induced bacterial infections, no differences of the inflammatory response in murine kidneys were noticed. To clarify, why no differences were observed during the infection with the Escherichia coli, peripheral blood was taken one day before infection, and serum was analysed by ELISA to check if LAMP-2/FimH-specific antibodies were produced after infection. The FimH-specific antibody titer was strongly increased 2 weeks after the first infection in contrast to the non-infected control. Surprisingly, 2 weeks after the second infection the specific LAMP-2/FimH-antibody titer decreased dramatically and unexpectedly. The results from these infection models indicated that pathogenic cross-reactive autoantibodies to the native murine LAMP-2 on murine neutrophils and activated Gr-1 positive myeloid cells are not easily induced by infections with mutant FimH-expressing bacteria or by immunization with a mLAMP2-derived linear peptide. Decrease of antibody titres in patients with urinary tract infections (UTI), but not AAV, with type 1 fimbriated pathogens investigated in WP04 exhibited a similar course after re-infection or resistant disease and indicates complex mechanisms of tolerance maintenance under physiological conditions in individuals that do not develop AAV after UTI.
Responses of antibodies to ANCA antigens Apart from directly cross-reacting antibodies, cross-reactive anti-idiotypes may be formed as a secondary response to antibodies (idiotypes) that are induced by an exogenous protein of potential bacterial origin. The primary exogenous immunogen should tightly interact with a folded host protein, e. g. PR3, and may, therefore, belong to the large family of serine protease inhibitors. Bacterial pathogens secrete a multitude of pathogenic factors in order to interfere with host-specific functions, in particular with endogenous proteolytic activities. Recently a new structural class of proteins, called EAPs (extracellular adhesion proteins) from Staphylococcus aureus were identified as inhibitors of PR3. PR3 inhibitors of bacterial origin could trigger the formation of anti-idiotypes that are directly interfering with the activity of PR3 similarly as the primary immunogen. We therefore searched for activity changing autoantibodies in vasculitis patients and developed a highly sensitive quantitative assay to detect these antibody populations in homogenous solution under equilibrium conditions. A large proportion of patients were shown to produce PR3-targeting autoantibodies with inhibitory capacity during the course of disease. However, no correlation between the changes in these autoantibody subpopulations and disease severity or disease progression was found in our clinical study in collaboration with partner 05 (Mayo).
Partner 02 (MPG, HMGU) has discovered and characterized a new serine protease with significant homology to human PR3 and neutrophil elastase. NSP4 is a low abundant neutrophil elastase- (NE-) with PR3-like biophysical properties, almost identical molecular weight and unusual substrate specificity. Like the other three abundant serine proteases of neutrophils (NE, cathepsin G, PR3), NSP4 is processed at the N-terminus and converted to its functionally active mature form by dipeptidylpeptidase I (cathepsin C). It is stored in the azurophil (primary) granules of neutrophils together with MPO, NE, cathepsin G and PR3 and released upon stimulation.
In conclusion, the results
1. Demonstrate the validity of using humanized murine models for the investigations of autoimmunity and the mouse lines established provide an invaluable resource for future studies investigating autoimmunity to ANCA antigens. One key finding is that target specificity of autoantibodies is less relevant for their pathogenicity than the isotype of autoantibodies. These result should lead to the development of assays that measure pathogenic autoantibody subsets to specific ANCA antigens
2. Indicate complex mechanisms of tolerance in molecular mimicry that could be explored to develop tolerizing immuno-therapies in individuals predisposed to developing cross-reactive antibodies
3. Identify novel potential ANCA antigens for clinical evaluation.
WP04 Host-pathogen interactions in generation of AAV
Work package 04 was based on the demonstration that the autoantibodies to LAMP-2 found in over 80% of patients with AAV commonly recognise an epitope shared with the bacterial adhesin FimH and that and that rodents immunised with FimH develop antibodies to LAMP-2 and AAV. This is one of the clearest examples of autoimmune disease induced by molecular mimicry and the central purpose of the work package was to examine the hypothesis that this mechanism applies to AAV clinically as well as in rodent models. Specifically, it was designed to provide insights into why only a small fraction of individuals infected with type 1 fimbriated bacteria develop antibodies to the shared FimH/LAMP-2 epitope AAV. It combined genetic studies of the relevant bacteria, functional studies of their effects on host cells and genetic studies of patients with AAV into and work in all three areas has produced important insights into (i) clinically relevant mechanisms in infections with type 1 fimbriated pathogens relevant to AAV and the injury it causes, (ii) the effects of antibody ligation on antigen presentation and (iii) genetic variances underlying the susceptibility to developing AAV, as well as providing the foundation for new avenues of research.
Molecular characterisation of FimH and its interaction with host cells Using a novel ELISA to quantify antibodies to FimH, Partner 01 (MUW) has shown that most patients presenting with AAV have IgG antibodies specific for IgG and that rising concentrations of these antibodies anticipate clinical relapses. However, urinary infections did not increase antibody titres either in patients with AAV or those without. A library of these pathogenic E. coli strains from individuals with and without AAV has been created and characterised by whole genome sequencing. Partner 04 (DTU) generated a unique panel of reagents for determining at a molecular level how FimH engages host cells: these include recombinant His-tagged FimH and a panel of GFP-tagged E. coli strains that express wild type or mutated FimH. Using these reagents, Partner 01 demonstrated that E. coli bound directly to glomerular endothelium in tissue sections, and that they adhere more efficiently to glomerular endothelial cells (GEnC) than to bladder urothelium in vitro, including under shear stress. Using the E. coli strains with mutated FimH, they identified two FimH residues responsible for enhanced bacterial adhesion to endothelium, and a further two that facilitated the initial adhesive contact under shear stress These studies emphasised the need to characterise the FimH receptors on microvascular endothelium responsible for adhesion and using a combination of affinity purification and mass spectrometry. Partner 01 identified three novel FimH binding partners on GEnC, including LAMP-2 which was then shown to mediate FimH-dependent adhesion to ldlD cells that stably express LAMP-2 on their cell surface (ldlD/hLAMP-2H). The discovery that LAMP-2 a cellular receptor for its molecular mimic FimH has important implications for the pathogenesis of microvascular injury in AAV but also for immune responses to LAMP-2 as indicated by the work undertaken in Task 2.
The effect of antibodies to LAMP-2 on antigen presentation Partner 01 has used human monocyte derived macrophages (MDM) and dendritic cells (moDC) to analyse how autoantibodies to LAMP-2 disturb its function in antigen presenting cells. They demonstrated that LAMP-2 was much more abundant on the surface of myeloid cells than previously suspected, even reaching 40% in activated moDC and was rapidly internalised after ligation by H4B4, a mouse monoclonal antibody specific for human LAMP-2. In macrophages, H4B4 trafficked rapidly to lysosomes and specifically depleted the lysosomal membrane of LAMP-2 and this in turn abrogated chaperone mediated autophagy (CMA) – a process critical for presentation of intracellular antigens to T cells. This established the principal that exposure to specific antibodies could interfere with LAMP-2’s intracellular functions. The high surface expression on moDC raised the question whether LAMP-2 was a surface receptor for antigen. As with macrophages, moDC rapidly internalised H4B4 that then trafficked rapidly to the MHC class II loading compartment (MIIC). Conjugates of a nominal antigen keyhole limpet haemocyanin (KLH) with H4B4 (H4B4-KLH) or isotype control (ISO-KLH) were used to analyse the effect of antigen delivery by LAMP-2. This confirmed H4B4-KLH was rapidly internalised and KLH trafficked significantly faster to MIIC than after pulsing with ISO-KLH. However, localisation to MIIC was transient and the overall effect was to re-route antigen from MIIC and into extracellular vesicles that were released into the surrounding medium. Analysis of peptides eluted from HLA-DR molecules displayed on the cell surface showed that fewer KLH-derived peptides when KLH was delivered by LAMP-2, whereas significantly more KLH and HLA-DR molecules were incorporated into EV when moDC were pulsed with H4B4-KLH than ISO-KLH. These changes had profound effect on CD4 T cell responses since proliferation induced by H4B4-KLH pulsed moDC was markedly attenuated whilst proliferation induced by EV released from them was considerably augmented: in both cases the differences were highly significant. These results identify LAMP-2 as a DC receptor for antigen and show that antigen internalised by it re-routed and presented to T cells selectively by EV rather than as antigen-derived peptides bound to HLA class II molecules on the cell surface. The antigens normally presented in this way have yet to be identified but may include MPO if initial experiments are confirmed, and the demonstration that LAMP-2 as a cellular receptor for FimH in Task 1 is also highly relevant.
Identification and characterisation of genes that influence susceptibility to AAV Partner 07 (UCAM) analysed patient cohorts collected by the European Genetics of Vasculitis Consortium which was originally established by Partners 07 (UCAM) and 01 (MUW) to identify genetic loci that influence susceptibility to AAV. Genome wide data was available from a UK discovery cohort of 1,233 patients with AAV and 5,884 UK controls and a replication cohort from Continental Northern Europe consisting of 1,454 additional patients and 1,666 controls. Three genetic loci, were HLA-DP, PRTN3 and SERPINA1, were associated with AAV at levels of genome-wide significance (p<5X10-8) together with a number of other promising candidates that did not reach the required level of significance. PRTN3 and SERPINA1 are the genes that encode PR3 and its major inhibitor a1-anti-trypsin, and were exclusively associated with individuals with AAV who had PR3-ANCA, as was HLA-DP. Separate analysis of individuals with MPO and PR3 ANCA revealed that those with autoantibodies to MPO were associated with HLA-DQ. The genetic loci were fine mapped using a custom made Immunochip designed specifically autoimmune and inflammatory diseases which enabled us to impute the HLA-DP alleles associated with the PR3-ANCA. This confirmed the previous association with HLA-DP0401 but in addition identified two novel alleles that conferred dominant resistance to the disease. Further analysis identified three residues within the HLA-DP1*0401 allele that are critical for susceptibility. Similar analysis of the HLA-DQ* A1 locus associated with MPO-ANCA identified one susceptibility allele and one that conferred dominant protection. Similar approaches identified two discrete susceptibility regions in the non-coding region of the PRTN3; and two separate susceptibility regions in the SERPINA1 gene, again located in non-coding regions. This provides the foundation for studies analyzing the physiological consequences of these non-coding genetic variants in PR3 ANCA-associated AAV. These data provide the first independent evidence that autoantibodies to PR3 are directly involved in pathogenesis and not simply an epiphenomena. Equally importantly, they demonstrate that the two commonest forms of AAV, granulomatous polyangiitis (GPA) and microscopic polyangiitis (MPA), are genetically distinct and that the primary genetic association is with the autoantibody specificity rather than the clinical syndrome. The initial results were published in the New England Journal of Medicine in July 2012 (Lyons et al NEJM. 267: 214-223) and immediately established the European group as the world leaders in the genetics of AAV.
In conclusion, the results
1. Generated a unique resource of fluorescently tagged E. coli strains expressing normal or mutated FimH and library of uropathogenic E. coli found in patients with AAV and control subjects
2. Identified LAMP-2 as receptor for FimH
3. Dissected the effects of autoantibody ligation on cellular function and translation of both findings into the development of novel therapies is currently explored.
4. Confirmed host susceptibility in AAV with autoantibodies to PR3 and MPO are associated with different HLA class II polymorphisms, whilst the PTEN and SEPINA1 loci - that respectively encode PR3 and its major inhibitor 1-anti-trypsin - are uniquely associated anti-PR3 associated disease. This provides the first evidence that the two disorders are genetically distinct and established INTRICATE researchers as world leaders in the genetics of AAV.
Potential Impact:
Socio-economic impact and the wider societal implications of the project
Contribution to Community and social objectives
Impact on the general population
The partners of INTRICATE aimed to disseminate the results of their research not only to researchers at scientific conferences but actively informed through their connections with patients’ representatives and by providing information for patients on INTRICATE web page.
Impact on resources available for future studies
In addition to its research output INTRICATE research has generated a unique set of resources available to the investigators and collaborators for the investigation of AAV.
Thus INTRICATE clinical collectors have assemble a large and unique cohort of patients, controls and their samples, established microbiological libraries of pathogens from patients with AAV and controls, generated research tools, like assays and peptide libraries through the engagement of SMEs.
Impact on clinical practice
The important findings from all work packages has already resulted in the extension of autoantibody profiling in AAV to novel antigens and the use of newly developed assays that promise will guide to development of more targeted therapies: S. aureus nasal isolates should be frequently tested for co-trimoxazole resistance if co-trimoxazole is used to eliminate S. aureus carriage and antibiotic therapy should be either changed when resistance develops or preemtively changed periodically. In order to improve the humoral response against S. aureus vaccination should be considered and effective vaccines developed.
Impact on Europe-wide and international collaborations
INTRICATE research has received international recognition and sparked several new collaborations that strengthen the European and international research in AAV. Thus partner 01 MUW has collaborated with investigators at East & North Herts NHS Trust, UK and Charles University Prague, Czech republic to build the large INTRICATE patient cohort. Partner 06 UMCG collaborated with the department of microbiology UMCG and with the department of microbiology Erasmus medical center Rotterdam and thus ensured success of WP02. Partner 08 Stanford and partner 09 EMC Prof Karl-Heinz Wiesmüller initiated collaborations with Science for Life Laboratory Stockholm, the Institute of Immunology, University of Veterinary Medicine Vienna and the Department of Neurology, University of Regensburg to extend the generation and use of microarray platforms and recombinant peptide methodology in cellular immunology and autoimmunre research. Partner 12 HMGU established new collaborations with Wroclaw University of Technology, Faculty of Chemistry, Division of Medicinal Chemistry and Microbiology in Poland, the Institute of Lung Biology and Disease (iLBD), Neuherberg, Germany to optimize the generation of synthetic protease substrates and inhibitors and models of lung injury and intranasal instillation of elastase.
Impact on training
INTRICATE partners not only trained over 20 researcher at all levels (PhD and postdoc level) in the methods of translation medicine, the project also allowed exchange of researchers an three fellowships had been successfully concluded.
Impact on women in science
The INTRICATE consortium is composed of 10 scientific partners coordinated by a female coordinator. Two partners are led by female principle investigators and the percentage of female staff members in all research groups involved in the proposed projects and already in employment is 28% at advanced academic level.
These figures indicate that the INTRICATE partners already have a strong commitment and INTRICATE put measures in place to promote women in science. With recruitment of equal numbers of 6 female and 6 male Phd students and 10 female and 10 male postdocs the INTRICATE consortium’s has met its goal to increase female researchers at all levels to at least 40% (preferred 50%).
Main dissemination activities and exploitation of results
Resulted in the publication of 48 original articles in high ranking biomedical journals , 5 of those in top medical Journals (Nature Medicine, Nature communications, Nature Reviews Immunology, Cell, NEJM) and several publications have received broad attention and were widely publicised also to the lay public:
Perera NC, Schilling O, Kittel H, Back W, Kremmer E, Jenne DE. NSP4, an elastase-related protease in human neutrophils with arginine specificity. Proc Natl Acad Sci U S A. 2012 Apr 17;109(16):6229-34.
http://www.neuro.mpg.de/133434/news_publication_5589778?c=22167
Perera NC, Jenne DE. Perspectives and potential roles for the newly discovered NSP4 in the immune system. Expert Rev Clin Immunol. 2012 Aug;8(6):501-3.
Hinkofer LC, Seidel SA, Korkmaz B, Silva F, Hummel AM, Braun D, Jenne DE, Specks U. A monoclonal antibody (MCPR3-7) interfering with the activity of proteinase 3 by an allosteric mechanism. J Biol Chem. 2013 Sep 13;288(37):26635-48.
http://www.biomedexperts.com/Abstract.bme/23902773/A_monoclonal_antibody_MCPR3-7_interfering_with_the_activity_of_proteinase_3_by_an_allosteric_mechanism
Schiwon M, Weisheit C, Franken L, Gutweiler S, Dixit A, Meyer-Schwesinger C, Pohl JM, Maurice NJ, Thiebes S, Lorenz K, Quast T, Fuhrmann M, Baumgarten G, Lohse MJ, Opdenakker G, Bernhagen J, Bucala R, Panzer U, Kolanus W, Gröne HJ, Garbi N, Kastenmüller W, Knolle PA, Kurts C, Engel DR. Crosstalk between sentinel and helper macrophages permits neutrophil migration into infected uroepithelium. Cell. 2014 Jan 30;156(3):456-68.
http://www3.uni-bonn.de/Press-releases/immune-cells-need-a-second-opinion
Tadema H, Abdulahad WH, Lepse N, Stegeman CA, Kallenberg CG, Heeringa P. Bacterial DNA Motifs Trigger ANCA
Production in ANCA-Associated Vasculitis in Remission Rheumatology (Oxford). 2011;50:689-696.
http://www.medscape.com/viewarticle/740011
Outlook and future research
INTRICATE legacy
The results of the INTRICATE research will not only result in additional publications of the work that had been recently concluded during the last period of funding but will impact also on future research. Thus INTRICATE research has made a major impact on research in AAV and autoimmunity more generally. The methods and approaches adopted have been adopted to investigating other autoimmune disease like SLE, Rheumatoid arthritis and Crohn’s disease by either INTRICATE investigators alone or in – often newly established – collaborations. These have been facilitated by the availability of existing INTRICATE resources, like samples from well characterised patients, research reagents and assays developed by INTRICATE’s SME partners and resulted in novel sources of funding and grant income.
List of Websites:
www.intricate.eu
Ao.Prof. PhD. Dr. Renate Kain
Clinical Department of Pathology
Medical University Vienna
Währinger Gürtel 18-20
A-1090 Vienna
Austria
Tel: 0043-1-40400-3681
Fax: 0043-1-40400-5193
email: renate.kain@meduniwien.ac.at
