Portal de Periódicos da CAPES

Mapping of myeloperoxidase epitopes recognized by MPO-ANCA using human-mouse MPO chimers

2006; Elsevier BV; Volume: 69; Issue: 10 Linguagem: Inglês

10.1038/sj.ki.5000354

1523-1755

Uta Erdbrügger, Thomas Hellmark, Donna O. Bunch, David A. Alcorta, J. Charles Jennette, Ronald J. Falk, P.H. Nachman,

Eosinophilic Disorders and Syndromes

Myeloperoxidase (MPO) is one of the major target antigens of antineutrophil cytoplasmic autoantibodies (ANCA) found in patients with small-vessel vasculitis and pauci-immune necrotizing glomerulonephritis. To date, the target epitopes of MPO-ANCA remain poorly defined. Human MPO-ANCA do not typically bind mouse MPO. We utilized the differences between human and mouse MPO to identify the target regions of MPO-ANCA. We generated five chimeric MPO molecules in which we replaced different segments of the human or mouse molecules with their homologous counterpart from the other species. Of serum samples from 28 patients screened for this study, 43 samples from 14 patients with MPO-ANCA-associated vasculitis were tested against recombinant human and mouse MPO and the panel of chimeric molecules. Sera from 64 and 71% of patients bound to the carboxy-terminus of the heavy chain, in the regions of amino acids 517–667 or 668–745, respectively. No patient serum bound the MPO light chain or the amino-terminus of the heavy chain. All sera bound to only one or two regions of MPO. Although the pattern of MPO-ANCA binding changed over time (4–27 months) in 6 of 10 patients with several serum samples, such changes were infrequent. Other target regions of MPO-ANCA may not have been detected due to conformational differences between the native and recombinant forms of MPO. MPO-ANCA do not target a single epitope, but rather a small number of regions of MPO, primarily in the carboxy-terminus of the heavy chain. Myeloperoxidase (MPO) is one of the major target antigens of antineutrophil cytoplasmic autoantibodies (ANCA) found in patients with small-vessel vasculitis and pauci-immune necrotizing glomerulonephritis. To date, the target epitopes of MPO-ANCA remain poorly defined. Human MPO-ANCA do not typically bind mouse MPO. We utilized the differences between human and mouse MPO to identify the target regions of MPO-ANCA. We generated five chimeric MPO molecules in which we replaced different segments of the human or mouse molecules with their homologous counterpart from the other species. Of serum samples from 28 patients screened for this study, 43 samples from 14 patients with MPO-ANCA-associated vasculitis were tested against recombinant human and mouse MPO and the panel of chimeric molecules. Sera from 64 and 71% of patients bound to the carboxy-terminus of the heavy chain, in the regions of amino acids 517–667 or 668–745, respectively. No patient serum bound the MPO light chain or the amino-terminus of the heavy chain. All sera bound to only one or two regions of MPO. Although the pattern of MPO-ANCA binding changed over time (4–27 months) in 6 of 10 patients with several serum samples, such changes were infrequent. Other target regions of MPO-ANCA may not have been detected due to conformational differences between the native and recombinant forms of MPO. MPO-ANCA do not target a single epitope, but rather a small number of regions of MPO, primarily in the carboxy-terminus of the heavy chain. Knowledge about the target epitopes of autoantibodies can provide valuable insight into the mechanisms that initiate and regulate the autoimmune response. Epitope mapping can identify molecular mimics and elucidate the relationship between an alloantigen and autoimmune disease. The analysis of changes in target epitopes over time in an individual patient may also provide insight as to whether relapses are associated with reactivity to a new epitope or reactivation of an antibody response to the same epitope(s). Antineutrophil cytoplasmic autoantibodies (ANCA) directed against myeloperoxidase (MPO) or proteinase 3 are associated with pauci-immune necrotizing and crescentic glomerulonephritis, microscopic polyangiitis, and Wegener's granulomatosis.1.Jennette J.C. Falk R.J. Small-vessel vasculitis (see comments).N Engl J Med. 1997; 337: 1512-1523Crossref PubMed Scopus (1193) Google Scholar To date, the epitope specificity of MPO-ANCA remains poorly defined. Characterizations of the MPO-ANCA epitope specificity performed by analysis of competitive binding of various antibodies or antisera to MPO2.Tadros M. Pozzi C. Radice A. et al.Characterization of anti-myeloperoxidase antibodies in vasculitis.Adv Exp Med Biol. 1993; 336: 291-294Crossref PubMed Scopus (2) Google Scholar, 3.Audrain M.A. Baranger T.A. Moguilevski N. et al.Anti-native and recombinant myeloperoxidase monoclonals and human autoantibodies.Clin Exp Immunol. 1997; 107: 127-134Crossref PubMed Scopus (34) Google Scholar, 4.Short A.K. Lockwood C.M. Studies of epitope restriction on myeloperoxidase (MPO), an important antigen in systemic vasculitis.Clin Exp Immunol. 1997; 110: 270-276Crossref PubMed Scopus (12) Google Scholar or by analysis of binding to various fragments5.Pedrollo E. Bleil L. Bautz F.A. et al.Antineutrophil cytoplasmic autoantibodies (ANCA) recognizing a recombinant myeloperoxidase subunit.Adv Exp Med Biol. 1993; 336: 87-92Crossref PubMed Scopus (14) Google Scholar or synthetic overlapping peptides6.Chang L. Binos S. Savige J. Epitope mapping of anti-proteinase 3 and anti-myeloperoxidase antibodies.Clin Exp Immunol. 1995; 102: 112-119Crossref PubMed Scopus (42) Google Scholar of MPO have led to inconclusive results. Furthermore, methods utilizing protein fragments or peptides fail to ‘present’ candidate target epitopes in a preserved three-dimensional conformation. Finally, very limited information is available as to whether the same epitopes are targeted during the onset of disease and relapse.7.Locke I.C. Leaker B. Cambridge G. A comparison of the characteristics of circulating anti-myeloperoxidase autoantibodies in vasculitis with those in non-vasculitic conditions.Clin Exp Immunol. 1999; 115: 369-376Crossref PubMed Scopus (25) Google Scholar Chimeric molecules have been used to determine the immunodominant epitopes of several autoantigens such as thyroid peroxidase8.Nishikawa T. Nagayama Y. Seto P. Rapoport B. Human thyroid peroxidase-myeloperoxidase chimeric molecules: tools for the study of antigen recognition by thyroid peroxidase autoantibodies.Endocrinology. 1993; 133: 2496-2501Crossref PubMed Scopus (23) Google Scholar and the glomerular basement membrane's non-collagenous domain 1 (NC1) domain of the alpha 3 chain of type IV collagen.9.Hellmark T. Segelmark M. Unger C. et al.Identification of a clinically relevant immunodominant region of collagen IV in Goodpasture disease.Kidney Int. 1999; 55: 936-944Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar Despite an 85% amino-acid identity (90% homology) between the human and mouse MPO sequences (Figure 1), the majority of human MPO-ANCA do not bind mouse MPO. We hypothesized that the areas of heterogeneity between human and mouse MPO are candidate target epitopes for MPO-ANCA. Furthermore, we hypothesized that the anti-MPO autoimmune response is directed against a limited number of immunodominant epitopes on MPO and that the same epitopes are targeted during disease onset and relapse. We exploited the difference in ANCA reactivity to human and mouse MPO to identify immunodominant epitopes of human MPO using a panel of recombinant human MPO, recombinant mouse MPO, and five human–mouse chimeric molecules. This approach allows presentation of selected regions of MPO in a preserved three-dimensional configuration. Chimeric molecules were designed to present a region of human MPO on the framework of the mouse MPO molecule. If this human region contains the target epitope of an MPO-ANCA, the antibody binding to the chimeric would be greater than that binding to recombinant mouse MPO. Alternatively, chimeric molecules were generated by replacing a region of the human MPO molecule by its homologous region from mouse MPO. If that region contains the target epitope, the antibody binding to the chimeric would be less than that binding to recombinant human MPO. The heavy chain of MPO was ‘divided’ into four segments, namely A, B, C, and D (Figure 2). We used capital letters to denote a human segment and lower-case letters to denote a mouse segment. ‘L’ or ‘l’ denotes the MPO light chain. Human and mouse recombinant MPO and five of the six chimeric MPO molecules (full description in Materials and Methods) were successfully expressed in human embryonic kidney 293 cells and the purified recombinant proteins verified by Western blot analysis (Figure 3). We were unable to express the ‘labcD’ despite three transfections of human embryonic kidney 293 cells. For each construct, we detected bands of 70 and 90 kDa by Western blot analysis, corresponding to the pro-MPO and mature MPO forms (Figure 3).Figure 3Immunoblot of recombinant mouse and human MPO and human–mouse chimeric molecules. This immunoblot shows the recombinant proteins and chimeric molecules detected by polyclonal rabbit anti-human MPO antibodies. Proteins analyzed are native=native human MPO, r. human=recombinant human MPO; r. mouse=recombinant mouse MPO.View Large Image Figure ViewerDownload (PPT) To assess whether the MPO light chain harbors target epitopes, we compared the binding pattern of each serum sample to the chimerics expressing the murine light chain (‘lABCD’, ‘lAbcd’, and ‘laBcd’) and the molecules expressing the human light chain (‘LABCD’, ‘LABcD’, and ‘LAbcd’). No patient serum demonstrated a consistent binding to the light chain. The reactivity of MPO-ANCA sera to the remaining four chimerics (lAbcd, laBcd, LABcD, and LAbcd) was analyzed. As we did not express the ‘labcD’ chimeric, we were not always able to distinguish whether certain samples bound either or both of the B or D regions or the C and D regions. Based on the binding pattern of sera from our 14 patients, sera from 0 patients bound to region A (amino acids (a.a.) 1–108 of heavy chain), sera from six patients bound to region B (a.a. 109–237), sera from nine patients bound to region C (a.a. 238–388), and sera from 10 patients bound to region D (a.a. 389–469) at some point during the course of disease (Figure 4). Sera from six patients bound to one region, sera from six patients reacted to two regions, and sera from two patients to possibly three regions over time (Table 1). No individual serum sample bound to more than two regions of MPO.Table 1Patient characteristics and pattern of MPO-ANCA binding for each sample testedTarget region: sequential samplesbAntibody target region for each serum sample available from each patient.PatientAgeGenderDiagnosisTime interval (months)aTime interval in months between the first and last serum sample.IIIIIIIVVVIVIIVIII 157FMPA7DC/DcUnable to ascertain if reactivity is to one, the other or both regions of MPO.C 259FWG17DCCC 336FWG16CC 463MMPA3DDDD 555MMPA16B&DD/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO. 645FWG18BBBDDDDD 776FWG0D/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO. 875FMPA20CC 950MRENAL0C1021FRENAL24BB/DcUnable to ascertain if reactivity is to one, the other or both regions of MPO.C/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO.C1146FMPA27DD/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO.D/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO.D/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO.D/BcUnable to ascertain if reactivity is to one, the other or both regions of MPO.CCC1271FRENAL0D1375FMPA11CC1478MMPA0B&DAbbreviations: F, female; M, male; MPA, microscopic polyangiitis; MPO-ANCA, myeloperoxidase-antineutrophil cytoplasmic autoantibodies; renal, pauci-immune necrotizing glomerulonephritis without extra-renal vasculitis; WG, Wegener's granulomatosis.a Time interval in months between the first and last serum sample.b Antibody target region for each serum sample available from each patient.c Unable to ascertain if reactivity is to one, the other or both regions of MPO. Open table in a new tab Abbreviations: F, female; M, male; MPA, microscopic polyangiitis; MPO-ANCA, myeloperoxidase-antineutrophil cytoplasmic autoantibodies; renal, pauci-immune necrotizing glomerulonephritis without extra-renal vasculitis; WG, Wegener's granulomatosis. In 4 of 10 patients with more than one sample, the binding pattern never changed over a period of 3–20 months. In six patients (patient nos. 1, 2, 5, 6, 10, and 11) the binding pattern changed over a period of 7–27 months. For three patients (nos. 1, 2, and 11) the target immunodominant region changed from D to C. For patient no. 6 the target immunodominant region changed from B to D and for patient no. 10 it changed from B to C (Table 1). Due to the small number of patients, we could not correlate the antibody-binding pattern with changes in clinical levels of disease activity. Knowledge about the immunodominant target regions of MPO can provide insight into the MPO-ANCA immune response. Several studies suggested that the immunodominant epitopes of MPO are restricted in number and are conformational in nature,2.Tadros M. Pozzi C. Radice A. et al.Characterization of anti-myeloperoxidase antibodies in vasculitis.Adv Exp Med Biol. 1993; 336: 291-294Crossref PubMed Scopus (2) Google Scholar, 3.Audrain M.A. Baranger T.A. Moguilevski N. et al.Anti-native and recombinant myeloperoxidase monoclonals and human autoantibodies.Clin Exp Immunol. 1997; 107: 127-134Crossref PubMed Scopus (34) Google Scholar, 4.Short A.K. Lockwood C.M. Studies of epitope restriction on myeloperoxidase (MPO), an important antigen in systemic vasculitis.Clin Exp Immunol. 1997; 110: 270-276Crossref PubMed Scopus (12) Google Scholar, 10.Falk R.J. Becker M. Terrell R. Jennette J.C. Anti-myeloperoxidase autoantibodies react with native but not denatured myeloperoxidase.Clin Exp Immunol. 1992; 89: 274-278Crossref PubMed Scopus (61) Google Scholar but failed to identify them. Chimeric molecules have been used to determine the immunodominant epitopes of several autoantigens in a preserved three-dimensional configuration.8.Nishikawa T. Nagayama Y. Seto P. Rapoport B. Human thyroid peroxidase-myeloperoxidase chimeric molecules: tools for the study of antigen recognition by thyroid peroxidase autoantibodies.Endocrinology. 1993; 133: 2496-2501Crossref PubMed Scopus (23) Google Scholar This approach pinpointed seven amino acids in the NC1 domain of the alpha 3 chain of type IV collagen that are essential to the structure of the target epitope of anti-glomerular basement membrane antibodies.9.Hellmark T. Segelmark M. Unger C. et al.Identification of a clinically relevant immunodominant region of collagen IV in Goodpasture disease.Kidney Int. 1999; 55: 936-944Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 11.Gunnarsson A. Hellmark T. Wieslander J. Molecular properties of the Goodpasture epitope.J Biol Chem. 2000; 275: 30844-30848Crossref PubMed Scopus (40) Google Scholar Epitope mapping of proteinase 3 using human–mouse and human proteinase 3–elastase chimeric antibodies determined a restricted number of target epitopes, which varied from patient to patient.12.Selga D. Segelmark M. Wieslander J. et al.Epitope mapping of anti-PR3 antibodies using chimeric human/mouse PR3 recombinant proteins.Clin Exp Immunol. 2004; 135: 164-172Crossref PubMed Scopus (20) Google Scholar The goal of this study was to identify the target immunodominant epitope(s) of MPO. In this respect, our use of recombinant human MPO limits identification of some target epitopes, especially those present only on the dimerized holoenzyme. This limitation is reflected in our ability to analyze the target regions of sera from 50% of the patients originally screened for this study. The recombinant human MPO molecule may also differ from the native molecule as a result of changes in folding and post-translational modifications such as glycosylation or the expression of the ‘pro-form’ as opposed to the ‘mature’ MPO molecule.13.Nauseef W.M. Insights into myeloperoxidase biosynthesis from its inherited deficiency.J Mol Med. 1998; 76: 661-668Crossref PubMed Scopus (51) Google Scholar Nevertheless, using the observation that most human MPO-ANCA patient sera bind human but not mouse recombinant MPO, we were able to define a limited number of antigenic regions of interest. In our study, sera from 71% of patients did not bind recombinant mouse MPO. Similarly, serum from only 1 of 36 (2.7%) patients with MPO-ANCA vasculitis bound rat MPO.14.Patry Y.C. Nachman P.H. Audrain M.A. et al.Difference in antigenic determinant profiles between human and rat myeloperoxidase.Clin Exp Immunol. 2003; 132: 505-508Crossref PubMed Scopus (8) Google Scholar These results confirm that MPO-ANCA recognize epitopes on human MPO that are absent on rat or mouse MPO. Several investigators have concluded that the target epitopes of MPO are conformational in nature by demonstrating that MPO-ANCA bound only the MPO holoenzyme, but not the denatured protein.2.Tadros M. Pozzi C. Radice A. et al.Characterization of anti-myeloperoxidase antibodies in vasculitis.Adv Exp Med Biol. 1993; 336: 291-294Crossref PubMed Scopus (2) Google Scholar, 10.Falk R.J. Becker M. Terrell R. Jennette J.C. Anti-myeloperoxidase autoantibodies react with native but not denatured myeloperoxidase.Clin Exp Immunol. 1992; 89: 274-278Crossref PubMed Scopus (61) Google Scholar Using expressed protein fragments of MPO, researchers have similarly deduced that the epitopes are conformational in nature and demonstrated that the immunodominant epitopes are likely on the heavy chain of MPO.6.Chang L. Binos S. Savige J. Epitope mapping of anti-proteinase 3 and anti-myeloperoxidase antibodies.Clin Exp Immunol. 1995; 102: 112-119Crossref PubMed Scopus (42) Google Scholar, 15.Tomizawa K. Mine E. Fujii A. et al.A panel set for epitope analysis of myeloperoxidase (MPO)-specific antineutrophil cytoplasmic antibody MPO-ANCA using recombinant hexamer histidine-tagged MPO deletion mutants.J Clin Immunol. 1998; 18: 142-152Crossref PubMed Scopus (38) Google Scholar, 16.Fujii A. Tomizawa K. Arimura Y. et al.Epitope analysis of myeloperoxidase (MPO) specific anti-neutrophil cytoplasmic autoantibodies (ANCA) in MPO-ANCA-associated glomerulonephritis.Clin Nephrol. 2000; 53: 242-252PubMed Google Scholar We confirm this observation as all the patient sera bound to the heavy chain, and none bound to the light chain. Based on the crystal structure of MPO,17.Zeng J. Fenna R.E. X-ray crystal structure of canine myeloperoxidase at 3 A resolution.J Mol Biol. 1992; 226: 185-207Crossref PubMed Scopus (270) Google Scholar, 18.Chen J. Anderson J.B. Weese-Scott C. et al.MMDB: Entrez's 3D-structure database.Nucleic Acids Res. 2003; 31: 474-477Crossref PubMed Scopus (146) Google Scholar the three-dimensional model of MPO reveals that, in the dimer form, the light chain is largely ‘hidden’ in the groove between the two MPO monomers and is poorly ‘accessible’ to antibody binding (Figure 5). In contradistinction with anti-glomerular basement membrane autoantibodies where the majority of patients react to a single, well-defined epitope,9.Hellmark T. Segelmark M. Unger C. et al.Identification of a clinically relevant immunodominant region of collagen IV in Goodpasture disease.Kidney Int. 1999; 55: 936-944Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar the MPO-ANCA sera tested in this study most frequently bound epitopes restricted to the C or D region of the heavy chain. Based on the three-dimensional model of MPO, these two regions are ‘intertwined’ on the same ‘hemisphere’ of the molecule (Figure 4). Our results corroborate some of the results of Fujii et al.,16.Fujii A. Tomizawa K. Arimura Y. et al.Epitope analysis of myeloperoxidase (MPO) specific anti-neutrophil cytoplasmic autoantibodies (ANCA) in MPO-ANCA-associated glomerulonephritis.Clin Nephrol. 2000; 53: 242-252PubMed Google Scholar who studied the pattern of reactivity of 20 sera from 20 MPO-ANCA patients against a panel of 10 partially overlapping recombinant fragments of MPO. In their study, 53% of patients bound to the terminal 147 amino acids of the heavy chain (corresponding to a region encompassing part of our C region and the entire D region). However, our results diverge from theirs, in that 84% of their sera bound the first 62 amino acids and 63% the next 68 amino acids near the amino-terminus of the heavy chain (corresponding to our A region and a junction of our regions A and B, respectively). These results most likely derive from the fact that Fujii et al.16.Fujii A. Tomizawa K. Arimura Y. et al.Epitope analysis of myeloperoxidase (MPO) specific anti-neutrophil cytoplasmic autoantibodies (ANCA) in MPO-ANCA-associated glomerulonephritis.Clin Nephrol. 2000; 53: 242-252PubMed Google Scholar expressed fragments of human MPO which were not presented in a preserved three-dimensional molecular structure. The finding of a restricted number of target epitopes was previously suggested based on studies using the method of competition among various antibodies.2.Tadros M. Pozzi C. Radice A. et al.Characterization of anti-myeloperoxidase antibodies in vasculitis.Adv Exp Med Biol. 1993; 336: 291-294Crossref PubMed Scopus (2) Google Scholar, 3.Audrain M.A. Baranger T.A. Moguilevski N. et al.Anti-native and recombinant myeloperoxidase monoclonals and human autoantibodies.Clin Exp Immunol. 1997; 107: 127-134Crossref PubMed Scopus (34) Google Scholar, 4.Short A.K. Lockwood C.M. Studies of epitope restriction on myeloperoxidase (MPO), an important antigen in systemic vasculitis.Clin Exp Immunol. 1997; 110: 270-276Crossref PubMed Scopus (12) Google Scholar Although studies of competitive binding of antibodies to their target antigen are helpful in determining the relative number of epitopes, they generally fail to identify the ‘location’ (target amino acids) of these epitopes. This general approach is further hindered by influences of protein–protein interaction, since the binding of one antibody to MPO may alter the molecule or block the binding of other antibodies to adjacent epitopes, for example.4.Short A.K. Lockwood C.M. Studies of epitope restriction on myeloperoxidase (MPO), an important antigen in systemic vasculitis.Clin Exp Immunol. 1997; 110: 270-276Crossref PubMed Scopus (12) Google Scholar It is important to note that although we report the restriction of ANCA reactivity to relatively large regions of the MPO molecule (C and D regions corresponding to a total of 230 amino acids), the actual number of amino acids that differ between the human and mouse molecules within these regions is much smaller (28 amino acids) (Figure 1). Within the C and D regions, there are four small regions of marked heterogeneity between the mouse and human MPO molecules that are separated by long stretches of amino-acid identity. We will assess whether these non-contiguous divergent amino acids are important to the structure of the target epitopes creating MPO chimerics with smaller areas of differences between the human and mouse molecules. We tested serial serum samples in 10 of our 14 patients. Although we observed changes in the antibody-binding pattern with time in six patients, these changes were infrequent, occurring once over 3–20 months in all but one patient. Owing to the relatively small number of patients in our study, and the inherently variable nature of ANCA vasculitis,19.Falk R.J. Nachman P.H. Hogan S.L. Jennette J.C. ANCA glomerulonephritis and vasculitis: a Chapel Hill perspective.Semin Nephrol. 2000; 20: 233-243PubMed Google Scholar we could not determine a clear correlation between the antibody-binding profile and specific disease manifestations or levels of activity or changes thereof. To do so with confidence requires the prospective analysis of multiple serum samples from a large group of patients. In summary, we employed an approach to the epitope mapping of MPO that preserves its three-dimensional structure. Our data reveal that MPO-ANCA react to one or more epitopes on the recombinant human, but not mouse, MPO molecule. Using our panel of chimeric mouse–human MPO molecules, we demonstrated a restriction of antibody reactivity to two intertwined target regions corresponding to the carboxy-terminus of the heavy chain. We plan to further pinpoint the target epitopes of MPO-ANCA by replacing smaller fragments of the mouse or human MPO with homologous regions from the other species, focusing on the discrete foci of marked heterogeneity in the amino-acid sequences between human and mouse at the carboxy-end of the MPO molecule. Ninety-five archival serum samples from 28 patients with MPO-ANCA vasculitis were screened for use in this study. Of these, 52 samples from 14 patients were excluded from further analysis due to lack of binding to the recombinant human MPO (29 samples), binding to the ‘no-antigen’ control (10 samples) or similar reactivity to both mouse and human recombinant MPO (13 samples from eight patients). Fourteen MPO-ANCA positive patients with microscopic polyangiitis, Wegener's granulomatosis or pauci-immune crescentic glomerulonephritis (age range 21–75 years old; 10 females) provided 43 serum samples (1–8 samples per patient) for testing against our panel of chimeric molecules (Table 1) collected at various times during their disease course and stored at -20°C. These serum samples were derived from patients with different and changing degrees of disease activity, ranging from active vasculitis to remission on and off immunosuppressive therapy. Thirteen of 14 patients provided at least one serum sample, while having signs of active vasculitis. Serum samples from 11 healthy volunteers (age range 21–60 years old; seven females) served as negative controls. This study was conducted in accordance with the Declaration of Helsinki Principles with an Institutional Review Board approved protocol. Informed consent was obtained prior to the blood collections. Full-length cDNA for human MPO was provided by Dr William Nauseef (University of Iowa, Iowa City, IA, USA); cDNA for mouse MPO was obtained by polymerase chain reaction (PCR) cloning from RNA isolated from WEHI 3BD cells20.Warner N.L. Moore M.A. Metcalf D. A transplantable myelomonocytic leukemia in BALB-c mice: cytology, karyotype, and muramidase content.J Natl Cancer Inst. 1969; 43: 963-982PubMed Google Scholar (American Type Culture Collection, Manassas, VA, USA). Using PCR, a HindIII restriction site and a Kozac consensus translation initiation sequence were introduced 5′ to the coding sequence (Figure 2). At the 3′ end of the cDNA, we introduced 18 nucleotides encoding an in-frame StrepTag® peptide (IBA, Göttingen, Germany) followed by a stop codon and an XhoI site. Both the human and murine cDNA for MPO were cleaved by HindIII/XhoI and subcloned into a pcDNA3 vector (Invitrogen, Leek, The Netherlands). Six human–mouse chimeric molecules were generated (Figure 2). In describing the chimeric molecules, we used capital letters to denote a human segment and lower-case letters to denote a mouse segment. ‘L’ or ‘l’ denote the MPO light chain. The heavy chain was ‘divided’ into four segments, namely A, B, C, and D (Figures 3, 4a and b) based on the regions of heterogeneity between the mouse and human molecules and the presence of convenient restriction enzyme sites (ApaI and KpnI) usable on both the human and mouse MPO cDNA. For example, in ‘lAbcd’ only the ‘A’ region of the heavy chain is of human origin and the other parts originate from the mouse sequence. The recombinant mouse and human MPO constructs are referred to as ‘labcd’ and ‘LABCD’, respectively. The L, A, B, C, and D regions correspond to human a.a. numbers, 165–272, 279–386, 387–516, 517–667, and 668–745, and murine a.a. numbers 139–246, 253–360, 361–490, 491–641, and 642–719, respectively. The ‘lABCD’ chimeric was created by sequential PCR technique,21.Ho S.N. Hunt H.D. Horton R.M. et al.Site-directed mutagenesis by overlap extension using the polymerase chain reaction.Gene. 1989; 77: 51-59Crossref PubMed Scopus (6832) Google Scholar where the ‘l’ was amplified from the ‘labcd’ and the ‘ABCD’ part from the ‘LABCD’. The PCR products were purified, combined, and used as template to allow fusion of the two products in a second PCR reaction using the forward murine ‘l’ primer and the reverse human ‘D’ primer. The final PCR product was then cleaved using HindIII/XhoI and inserted into the pcDNA3 vector. ‘lAbcd’ and ‘LAbcd’ were generated from ‘lABCD’ and ‘LABCD’, respectively. The ‘bcd’ segment was generated by standard PCR using ‘labcd’ as template, and subcloned into the previously digested ‘lABCD’ and ‘LABCD’ cDNA using restriction enzymes PshAI (New England Biolabs, Ipswitch, MA, USA) at a.a. 386 and XhoI (New England Biolabs, Ipswitch, MA, USA) 3′ to the MPO coding sequence (Figure 2). The ‘laBcd’ chimeric was generated from ‘labcd’ and a newly generated ‘Bcd’ fragment. Sequential PCR technique was applied to generate the segment ‘Bcd’. The murine sequence contains an ApaI site at the ‘a–b’ join that the human sequence lacks. Thus, a new ApaI site was introduced at the 5′ end of ‘B’ during the PCR generation of the ‘B’ fragment. In a second PCR, the segment ‘cd’ with an ApaI site in the 3′ end was generated using the ‘labcd’ as template. A third PCR used the segments ‘B’ and ‘cd’ as templates to create ‘Bcd’. The ‘bcd’ fragment was excised from ‘labcd’ using the ApaI sites and the new ‘Bcd’ was subcloned into the ‘pcDNA3-la’ DNA. This approach was also used to generate the ‘labcD’ chimeric molecule, using ‘labcd’ as the initial template. The ‘LABcD’ chimeric was generated using ‘LABCD’ as template. The restriction enzyme KpnI (New England Biolabs) was utilized to cut out ‘C’ at a.a. 516 and 667 and replace it by the corresponding mouse ‘c’ segment, which was isolated after similar Kpn1 digestion. For each chimera, the final cDNA construct cloned into the expression vector was sequenced using external and internal sequencing primers to verify that it was complete and corresponded to the desired sequence. Human embryonic kidney 293 cells were cultured in Dulbecco's modified Eagle's medium/F12 medium (Gibco: Invitrogen Corp, Carlsbad, CA, USA) plus 10% fetal bovine serum (Gibco). For each construct, 8 × 105 cells were transfected with 20 μg of plasmid by electroporation (Bio-Rad, Hercules, CA, USA) at 200 V and 960 μF. Geneticin® (400 μg/ml) (Gibco)-resistant cell clones producing recombinant protein, as determined by enzyme-linked immunosorbent assay, were expanded and transferred to serum-free Dulbecco's modified Eagle's medium/F12 1 week prior to collection of cell culture supernatant. Proteins from cell culture supernatant were precipitated using ammonium sulfate at 4°C in the presence of protease inhibitors. The suspension was centrifuged at 18 600 g for 30 min. The pellet was dissolved in and dialyzed against the sample buffer (50 mM Tris buffer, 0.02% azide, pH 7.5). The samples were run through an anion exchange column (UnoQ, Bio-Rad) and the flow-throughs containing the MPO protein were collected and concentrated using ultrafiltration devices (Vivaspin, VivaScience). The purity of the recombinant proteins was determined by analysis of silver-stained sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels and by Western blot analysis. Protein concentration was determined by measuring the absorbance at 280 nm and confirmed by comparing intensities of recombinant protein bands with a known amount of purified human MPO (Calbiochem) on silver-stained sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. A 4–15% Tris-HCl gel (Bio-Rad) was run in Laemmli buffer system at 100 V for 90 min and stained using a Silver Stain Plus kit (Bio-Rad). For immunoblotting, proteins were transferred to a nitrocellulose membrane (Schleicher and Schuell, Keene, NH, USA) (100 V for 60 min). The membranes were blocked with 10% non-fat milk in Tris-buffered saline+0.05% Tween and then incubated overnight at 4°C with a rabbit anti-human MPO antibody diluted 1:5000. The membranes were washed and incubated with a horseradish peroxidase-conjugated goat anti-rabbit IgG (Chemicon, Temecula, CA, USA) for 1 h; proteins were detected with SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA). StrepTactin®-coated plates (IBA, Germany) were coated overnight at 4°C with recombinant human and mouse MPO and chimeric molecules at 0.1 μg/well, the saturation concentration for the StrepTactin® plates. The plates were washed and blocked with fish gelatin buffer (2 g/l in Tris-buffered saline+0.05% Tween) for 1 h at room temperature. MPO-ANCA sera, diluted 1:100 in fish gelatin buffer, were incubated for 3 h at room temperature. Polyclonal rabbit anti-human MPO antibody (1:5000 dilution) served as a positive control. Sera from healthy donors (diluted 1:100) served as negative controls. Secondary antibodies, alkaline phosphatase-conjugated goat anti-rabbit IgG (Pierce) and goat anti-human IgG (Pierce) (diluted 1:5000) were added for 1 h at room temperature. Bound antibodies were detected with an Alkaline Phosphatase Substrate Kit (Bio-Rad). Absorption was measured spectrophotometrically at 405 nm after 1 h. All assays were carried out in duplicate. The reactivity of normal sera to the various recombinant proteins varied substantially. For this reason, we first normalized the reactivity of each serum sample against the ‘background’ by testing each serum sample against the blocking buffer as a no-antigen control (ODbackground) according to the following equation:X=ODchimeric−ODbackgroundODchimeric(1) Equation (1) was used to calculate the reactivity of each sample to each recombinant molecule. The reactivity of 11 normal sera against each chimeric molecule was also determined using Equation (1). For the reactivity of a serum sample to a recombinant molecule to be considered positive, it had to be greater than the average+2 s.d. of the normal sera for that molecule using Equation (1). In order to compare the reactivity of a serum sample to the various chimeric molecules, we normalized its binding to each chimeric molecule relative to its binding to the recombinant human MPO according to Equation (2):X=ODchimeric−ODbackgroundODchimericODrec,human−MPO−ODbackgroundODrec,human−MPO(2) We assumed that the maximal binding of a serum sample would be to the recombinant human MPO. Thus, if the reactivities of a serum sample to the chimeric molecule and recombinant human MPO are the same, the ratio is 1. A ratio of 0.5 would indicate a 50% loss of reactivity to the chimeric molecule compared to the recombinant human MPO. We used a ratio >0.5 as the threshold for determining the substantial preservation of binding to the chimeric molecule when compared to that of the recombinant human MPO. This work was supported by National Institute for Diabetes and Digestive and Kidney Diseases grant PO1DK58335, the Swedish Research Council and the Swedish Association for Medical research. This work was presented in part in abstract form at the 2002 (Erdbrüegger U, Hellmark T, Majure MCM et al. Comparative binding of human anti-MPO antibodies to human and mouse MPO. J Am Soc Nephrol 2002; 13) and 2004 (Erdbrüegger U, Hellmark T, Alcorta DA et al. Most myeloperoxidase (MPO)-ANCA show stable binding to a restricted number of epitopes. J Am Soc Nephrol 2004; 15: 680A) meetings of the American Society of Nephrology.