Published ahead of print on February 14, 2008, doi:10.1164/rccm.200707-1015OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200707-1015OC
Identification of Target Antigens of Antifibroblast Antibodies in Pulmonary Arterial Hypertension1 Paris Descartes University, Faculty of Medicine, UPRES-EA 4058, Paris, France; 2 Institut Cochin, Université Paris Descartes, CNRS (UMR 8104), Laboratory of Proteomics, Paris, France; 3 Faculty of Medicine, Paris Sud University, Department of Pneumology and French Reference Center for Pulmonary Arterial Hypertension, Antoine-Beclere Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Clamart, France; 4 Faculty of Medicine, Paris Descartes University, Department of Internal Medicine and French Reference Center for Necrotizing Vasculitides and Systemic Sclerosis, Cochin Hospital, AP-HP, Paris, France; and 5 INSERM, U567, Paris, France Correspondence and requests for reprints should be addressed to Dr. Luc Mouthon, M.D., Ph.D., Laboratoire d'Immunologie, UPRES EA 4058, Pavillon Gustave Roussy, 4e étage, Paris-Descartes University, 8 rue Méchain, 75014 Paris, France. E-mail: luc.mouthon{at}cch.aphp.fr
Rationale: Pulmonary arterial hypertension (PAH) may be classified as idiopathic (IPAH) or familial (FPAH) or associated with various conditions and exposures such as dexfenfluramine intake (Dex-PAH) or systemic sclerosis (SSc-PAH). Because fibroblast dysfunction has been identified in SSc and IPAH and antifibroblast antibodies (AFAs) with a pathogenic role have been detected in the serum of SSc patients, we used a proteomic approach combining two-dimensional electrophoresis and immunoblotting to identify the target antigens of AFAs in such patients. Objectives: To identify target antigens of antifibroblast antibodies in pulmonary arterial hypertension. Methods: Sera from 24 patients with IPAH, 6 with FPAH, 6 with Dex-PAH, and 12 with SSc-PAH were collected. We pooled sera from sets of three patients with PAH classification and SSc-PAH based on autoantibody profile. Sera from 14 healthy blood donors were also pooled and used as a control. Measurements and Main Results: Serum IgG antibodies in the pools of patients with IPAH (n = 8), FPAH (n = 2), Dex-PAH (n = 2), and SSc-PAH (n = 4) recognized 103 ± 31, 63 ± 20, 78 ± 11, and 81 ± 12 protein spots, respectively, whereas serum IgG antibodies from healthy control subjects recognized 43 ± 22 protein spots. Twenty-one protein spots were specifically recognized by the serum IgG antibodies from patients with PAH. We identified 16 of the protein spots as vimentin, calumenin, tropomyosin 1, heat shock proteins 27 and 70, glucose-6-phosphate-dehydrogenase, phosphatidylinositol 3-kinase, DAP kinase, and others. These proteins are involved in regulation of cytoskeletal function, cell contraction, oxidative stress, cell energy metabolism, and other key cellular pathways. Conclusions: AFAs detected in patients with PAH recognize cellular targets playing key roles in cell biology and maintenance of homeostasis.
Key Words: fibroblast • antigens • autoantibodies • pulmonary arterial hypertension
Pulmonary arterial hypertension (PAH) is a rare disorder that represents a major cause of progressive right-heart failure leading to premature death (1). PAH is defined on right-heart catheterization by increased mean pulmonary artery pressure greater than 25 mm Hg at rest or greater than 30 mm Hg with exercise in the absence of a postcapillary process (2). PAH occurs as a consequence of chronic obstruction of small pulmonary arteries secondary to dysfunction and proliferation of endothelial cells, vascular smooth muscle cells, and fibroblasts (3). In agreement with the revised classification proposed in 2003, PAH is classified as idiopathic (IPAH) or familial (FPAH) or is associated with various conditions and exposures, including portal hypertension, HIV infection, use of drugs such as dexfenfluramine (Dex-PAH), and connective tissue disorders such as systemic sclerosis (SSc) (4). PAH develops in approximately 8 to 16% of patients with SSc (5, 6) and is responsible for a high mortality rate. PAH has a multifactorial pathobiology contributing to increased pulmonary vascular resistance, including vasoconstriction, remodeling of the pulmonary vessel wall, and thrombosis. The formation of a layer of myofibroblasts and extracellular matrix between the endothelium and the internal elastic lamina, termed neointima, is a hallmark of severe PAH. In hypoxia models, adventitial fibroblasts appear to be the first cells activated to proliferate and synthesize matrix proteins (7). Because dysfunctional fibroblasts have been identified in both SSc and IPAH (3, 8) and because antifibroblast antibodies (AFAs) able to activate and induce collagen synthesis have been detected in the serum of patients with SSc, we recently investigated the presence of AFAs in the serum of patients with PAH. We found that 40% of patients with IPAH and 30% of those with SSc-associated PAH (SSc-PAH) expressed IgG AFAs. IgG AFAs from these patients predominantly bound to 25-, 40-, and 60-kD protein bands, but target antigens of these antibodies remained to be identified (9). Therefore, we aimed to use a proteomic approach combining two-dimensional (2-D) electrophoresis and immunoblotting, with normal human fibroblasts used as a source of self-antigens, to identify the target antigens of AFAs in patients with IPAH, FPAH, Dex-PAH, and SSc-PAH (10).
Serum of Patients Sera were obtained from 48 patients with PAH: 24 with IPAH, 6 with FPAH, 6 with Dex-PAH, and 12 with SSc-PAH. PAH was confirmed by right-heart catheterization. In all subjects, mean pulmonary artery pressure at rest was greater than 25 mm Hg. By convention, patients with PAH were considered to have IPAH if they showed PAH with no evidence of familial PAH, Dex exposure, or associated disease. Patients with SSc fulfilled the LeRoy and Medsger (11) and/or the American Rheumatism Association criteria (12). Fourteen healthy blood donors were recruited as control subjects. Healthy individuals did not differ significantly from patients with PAH in terms of age (44 ± 13 vs. 45 ± 19 yr, respectively; P = 0.99) or sex (37 females, 11 males vs. 8 females, 6 males, respectively; P = 0.18). All patients gave their written, informed consent according to the ethics committee of the La Pitié-Salpêtrière Hospital Group (Paris, France). None of the patients received corticosteroids or immunosuppressants and none had cancer or another connective tissue disease. Patients were grouped by distinct phenotype: (1) IPAH, (2) FPAH, (3) Dex-PAH, (4) SSc-PAH with anticentromeric antibodies, (5) SSc-PAH with anti–topoisomerase 1 antibodies, or (6) SSc-PAH with antinuclear antibodies without specificity. Sera from 14 healthy blood donors were studied as controls. Therefore, we tested eight pools from patients with IPAH, two from patients with FPAH, two from patients with Dex-PAH, four from patients with SSc-PAH; we pooled the sera for the 14 healthy blood donors.
Cell Culture
Detection of Antibody Reactivity with Cell Antigens The 2-D gel electrophoresis (2-DE), 2-D blots, and protein identification by mass spectrometry on 2-DE gels were as described (15) (see the online supplement). The study protocol is depicted in Figure 1.
Statistical Analysis Data are presented as mean values. The Mann-Whitney test was used to compare quantitative values. P  0.05 was considered significant. Statistical analyses used InStat (version 3.00; GraphPad Software, San Diego, CA). Peptide masses were searched by treating raw spectra automatically according to the Mascot wizard algorithm (Matrix Science, Ltd., London, UK) on the NCBI nonredundant GenBank database (www.ncbi.nlm.nih.gov). When nonsignificant scoring was obtained, a manual treatment was performed according to the following explanation. First, raw spectra were treated using a noise filter algorithm (correlation factor 0.7) and the default advanced baseline correction. Then monoisotopic masses were generated on fully detected spectrum after internal calibration using autodigestion tryptic peptides or using external calibration using a mixture of five external standards (PepMix 1; LaserBio Labs, Sophia Antipolis, France). Main peaks were selected according to local background and molecular weight of the protein, known contamination peaks from trypsin and keratin using PeakErazor software (Lighthouse Data, Odense, Denmark). Peptide masses were performed using four different algorithms for protein identification (Mascot [Matrix Science], ProFound [Proteometrics Winnipeg, MB, Canada], MS-Fit [ProteinProspector, University of California, San Francisco, CA], and Aldente [Expasy, Swiss Institute of Bioinformatics, Geneva, Switzerland] software) on rodent proteins from a comprehensive nonredundant protein sequence database (NCBInr database, version October 2006 and later). Allowable variable modifications were oxidation of methionine, acrylamide-modified cystein, and carbamidomethylation of cystein. Up to one missed tryptic cleavage was considered, and a mass accuracy in the range of 25 to 50 ppm was used for all tryptic mass searches. Identified proteins, defined as proteins detected with the most elevated score with at least three algorithms, are shown in Tables 2 and 3.
Clinical and immunologic characteristics of patients with IPAH, FPAH, Dex-PAH, and SSc-PAH are summarized in Table 1 and detailed in the online supplement.
Fibroblast Proteome We first analyzed the fibroblast proteome. We have used the same source of normal human fibroblast proteins in all experiments. From proteins extracted from normal human fibroblasts and separated on 2-DE (isoelectric point [pI] 3–10; 7–18% polyacrylamide gel electrophoresis [PAGE]), 859 protein spots were detected on silver nitrate staining. Most of the protein spots were detected at pI 4.5–8 and 15–100 kD (Figure 2). A total of 422 ± 129 spots were successfully transferred onto polyvinylidene fluoride (PVDF) membranes (n = 17).
Serum IgG Reactivity in Healthy Control Subjects and Patients with PAH with Fibroblast Antigens Serum IgG antibodies in the sera pool for the 14 healthy control subjects incubated in three different experiments recognized 43 ± 22 protein spots, significantly less than that recognized in the pools for patients with PAH (89 ± 27 spots). Serum IgG antibodies in the eight pools for patients with IPAH, two pools for patients with FPAH, two pools for patients with Dex-PAH, and four pools for patients with SSc-PAH recognized 103 ± 31, 63 ± 20, 78 ± 11, and 81 ± 12 spots, respectively, which were not significantly different from each other.
Comparison of the Binding of Serum IgG Antibodies between Patients with PAH and Healthy Control Subjects
Identification of Target Antigens of AFAs Specifically Recognized in Patients with PAH For the 21 spots, we could identify 16 AFA target antigens on mass spectrometry. The proteins identified and indications regarding the reliability of these assignments are shown in Table 2. Details regarding the reactivity expressed in the different pools of sera are shown in Table 3.
In previous work, we detected AFAs in 40% of patients with IPAH and 30% of patients with SSc-PAH (9). In the present study, we have identified target antigens of AFAs by a proteomic approach. We used pools of sera from phenotypically identical patients to screen a large panel of patients with PAH. Patients with PAH expressed greater IgG reactivity against AFAs than healthy control subjects. Next, we identified target antigens of AFAs in patients with PAH that were recognized in the sera pools of more than 75% of all patients with PAH, in non-SSc PAH patients, in patients with IPAH alone, and in patients with SSc-PAH alone and not in the healthy control pool. These target antigens were involved in three main cellular systems: regulation of cytoskeletal organization, cell contraction, and oxidative stress. The other antigens are involved in various other pathways.
The target antigens involved in the regulation of cytoskeletal organization include vimentin, calumenin, and phosphatidylinositol 3-kinase. Vimentins are intermediary filaments highly expressed in fibroblasts, especially myofibroblasts. Because of the overexpression of vimentin in myofibroblasts, the detection of a humoral immune response could reflect increased myofibroblastic differentiation as in severe lesions in fibrosis. Calumenin, produced by activated platelets, is involved in the modulation of cytoskeletal organization and the development of lesions in thrombosis and atherosclerosis (16). Antibodies against calumenin could perturb the cytoskeleton homeostasis and antithrombotic system. Phosphatidylinositol 3-kinase, known to play an important role in the signal transduction of cell growth, is also involved in regulation of cytoskeletal reorganization by binding to Tropomyosin 1 is involved in cell contraction. Indeed, the tropomyosins play a central role in the regulation of smooth and skeletal muscle cells. Antitropomyosin antibodies were already reported in Behçet disease and could play a pathogenic role, because vasculitis features developed in rats immunized with this protein (18). However, the mechanisms by which these antibodies exert their pathogenic role remain to be determined. The target antigens involved in oxidative stress include heat shock protein (HSP) 27, HSP 70, and glucose-6-phosphate dehydrogenase. HSPs are expressed in response to stress and could be overexpressed in PAH. Glucose-6-phosphate dehydrogenase is a key enzyme involved in the production of reduced nicotinamide adenine dinucleotide phosphate (NADPH). NADPH is necessary for the production of reduced glutathione, which is involved in protection against oxidative stress–induced apoptosis (19). Interestingly, oxidative stress was found associated with vascular endothelial dysfunction in SSc (20) and PAH (21). The presence of a humoral immune response against these proteins could reflect their overexpression in PAH as part of the pathogenic process. The remaining target antigens are involved in various other pathways. Alanine-glyoxylate aminotransferase 2 and glutaminase play a role in the regulation of protein metabolism. Interestingly, glutamate carboxy-peptidase, a transmembrane protein, regulates angiogenesis by modulating integrin signal transduction in solid tumors (22), and vascular changes and defective angiogenesis are a hallmark in the pathogenesis of SSc (23). We identified other minor antigens, but their function and/or pathogenic role remains to be characterized. Globally, we detected 21 proteins and identified 16 potential targets of AFAs in PAH. These findings further support the existence of additional molecules that can contribute to the development of autoimmunity and/or pathophysiology in PAH. These proteins are involved in regulation of cytoskeletal regulation, cell contraction, oxidative stress, cell energy metabolism, and other pathways that play key roles in cell biology and maintenance of homeostasis. Recent studies showed that AFAs from patients with SSc induced fibroblast activation (24) and acquisition of a proinflammatory and proadhesive phenotype (13). Furthermore, in patients with SSc, AFA could activate platelet-derived growth factor receptor and exert a pathogenic role by stimulating the production of reactive oxygen species (25). Finally, AFAs also induce type I collagen–gene expression and conversion of normal human fibroblasts to myofibroblasts (13). Thus, AFAs could have a pathogenic role by interacting with cell-surface proteins but also by being internalized by fibroblasts via a caveolin-linked pathway and then interacting with cytoplasmic proteins (26). Two-dimensional gel electrophoresis is mainly used to analyze proteomes and investigate differential patterns of qualitative and quantitative protein expression. The combination of 2-DE and immunoblotting offers a new approach for the identification of target antigens of antibodies. Pooling sera from patients for analysis has been adopted by other authors (27, 28). Kao and colleagues used a pool of four patients for the identification of an allergen from garlic (27), and Mutapi and coworkers used pools of 62 to 112 individuals to identify and compare proteins recognized by serum samples from Schistosoma hematobium–exposed individuals before and after curative praziquantel treatment (28). Using different approaches with antibody microarrays or magnetic bead separation for proteome profiling of human blood by MALDI-TOF MS (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry), pools of 8 (29) and 20 (30) serum samples were used, respectively. Also, this method allows for identifying reactivity against different protein isoforms, which reflects post-translational modifications. Taken together, our results confirm the presence of AFAs in patients with PAH and provide evidence that these antibodies recognize cellular targets playing key roles in cell biology and maintenance of homeostasis. Functional analysis will be necessary to demonstrate the potential pathogenic role of these antibodies and confirm that autoimmune mechanisms could take part in the disease process in PAH. Finally, use of recombinant proteins will be necessary to validate our results. The usefulness of the identified targets of AFAs for PAH screening, diagnosis, or follow-up needs further confirmation by extensive laboratory screening with large groups of patients and control subjects.
Supported by grants from the Association des Sclérodermiques de France (ASF); the Legs Poix, Chancellerie des Universités, Académie de Paris, France; Actelion Pharmaceuticals France; and a "Contrat d'Investigation et de Recherche Clinique" (CIRC 05066) from the Assistance Publique-Hôpitaux de Paris. Also supported in part by the French Network on Pulmonary Arterial Hypertension. B.T. received financial support from the Direction Régionale de l'Action Sanitaire et Sociale d'Ile de France. M.C.T. is a recipient of grants from AMPLI (Avenir Mutualiste des Professions Libérales & Indépendantes), ASF, and Actelion Pharmaceuticals. M.C.T., M.H., L.G., and L.M. are members of the Groupe Français de Recherche sur la Sclérodermie.
* These authors contributed equally to this work. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200707-1015OC on February 14, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 10, 2007; accepted in final form February 13, 2008
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
-actinin (
