Validation of a New Biomarker in Patients with Kawasaki Disease Identified by Proteomics

Abstract

Systemic vasculitis is a heterogeneous disorder characterized by chronic or acute inflammation, and is classified into groups by the size of the affected blood vessels: large vessel vasculitis, medium vessel vasculitis, or small vessel vasculitis. The pathogenesis of systemic vasculitis is not yet understood and except for Anti-neutrophil Cytoplasmic Antibodies (ANCA) for a subset of small vessel vasculitis, biomarkers for diagnosis and disease activity in systemic vasculitis have still to be discovered. Kawasaki Disease (KD) is an acute systemic vasculitis of infancy and early childhood. It mainly affects small and medium size blood vessels, and can cause cardiovascular complications like coronary artery aneurysm. Anti-endothelial Cell Antibodies (AECA) have been detected in a variety of diseases associated with vascular injury,and KD is one of them. Further, the presence of AECA has been correlated with disease activity, and/or its clinical manifestations [1,2]. A number of mechanisms for AECA, for example, activation and apoptosis of vascular Endothelial Cells (ECs), have been proposed [3]. In KD patients, IgM AECA mediate complement-dependent cytotoxicity against Human Umbilical Vein Endothelial Cells (HUVEC) and gammaglobulin, a standard therapy for KD, may reduce their effect [4]. Grunebaum et al. [5] demonstrated that AECA increased the secretion of Interleukin-6 (IL-6) from HUVEC, in KD patients. However, a significant role for AECA in systemic vasculitis has yet to be found. To evaluate the roles of AECA in detail, it is essential that their target antigens are identified and that the role of individual target antigens is assessed. Various approaches, such as expression libraries and proteomic, have been developed to identify biomarkers [6-11]. In KD, two target antigens for the AECA, tropomyosin and T-plastin, were identified using serological analysis of a recombinant cDNA expression library [11]. However, target antigens suitable for clinical applications, including Myeloperoxidase (MPO) and proteinase 3 of ANCA, remain to be discovered. Proteomics is a powerful tool for detecting and identifying biomarker proteins, because proteomics techniques allow the global analyses of protein function, modifications, composition and dynamics to be performed. In addition, current proteomic techniques can be used to analyze several samples from, for example, serum, urine, cerebrospinal fluid, and various tissues and cells. These kinds of analyses may lead to clinical applications. To identify the target proteins for AECA in patients with vasculitis, we used a proteomics approach that included 2-dimensional electrophoresis and Western Blotting, followed by mass spectrometry [9]. Briefly, we extracted the proteins from HUVEC and HeLa cells (control), and separated them by 2-dimensional electrophoresis to detect AECA antigens specific for ECs. We then analyzed the antigens by Western Blotting against serum samples from patients with vasculitis. We selected the spots that were detected only in the HUVEC samples, and not in HeLa cell samples. This procedure should identify candidate antigens, specific for ECs. We identified the detected proteins by peptide mass fingerprinting. Currently, we have detected more than 150 candidate antigenic spots, and identified more than 50 protein spots. In our earlier study, we identified Peroxiredoxin-2 (Prx2), a member of the Peroxiredoxin (Prx) family of peroxidases, as a target antigen for AECA [9]. In mammalian cells, the Prx family has at least six members (Table 1), and Prx2 is one of the most rapid and potent responders to oxidative stress. Furthermore, oxidative stress has been implicated as a pathogenic factor, and/or a progression-related factor in various diseases, including vasculitis. For these reasons, Prx2 was selected for further investigation. We investigated the clinical significance of anti-Prx2 antibodies in patients with KD. The titers of IgG antibodies to recombinant Prx2 were evaluated by ELISA (Enzymelinked Immunosorbent Assay), using 30 untreated patients with KD, including three patients with Coronary Artery Lesions (CALs), and 15 age- and sex-matched controls. The optical density value of the average ± 2 Standard Deviations (SDs) in the control individuals was defined as 100 AU, and values more than 100 AU, were regarded as positive. IgG antibodies to recombinant Prx2 were detected in 60% (18 of 30) of the untreated patients with KD, whereas no IgG antibodies were detected in the control individuals (Figure 1). We also evaluated the titers of IgM and IgA antibodies to recombinant Prx2 by ELISA, in patients with KD and in controls. IgM antibodies to recombinant Prx2 were