Kidney International, Vol. 58 (2000), pp. 2473–2477

CLINICAL NEPHROLOGY – EPIDEMIOLOGY – CLINICAL TRIALS

Proteinase 3 gene polymorphisms andWegener’s granulomatosis

MARTIN GENCIK, STEPHAN MELLER, STEFAN BORGMANN, and HARALD FRICKE1

Molecular Human Genetics, Ruhr-University Bochum, and Medizinische Klinik, Klinikum Innenstadt, Universitat Munchen,Munchen, Germany

Proteinase 3 gene polymorphisms and Wegener’s granuloma- Wegener’s granulomatosis (WG) is a rare primary sys-tosis. temic vasculitis characterized by granulomatous lesions

Background. Wegener’s granulomatosis (WG) is a rare sys- of small vessels and the presence of cytoplasmic antineu-temic autoimmune disease characterized by small-vessel vascu-trophil cytoplasmic autoantibodies (c-ANCAs) in seralitis leading to organ damage and the presence of antineutrophilof the patients [1]. ANCAs have been shown to activatecytoplasmic autoantibodies (ANCAs). ANCAs were shown to

be involved in the pathogenesis of the disease by increasing primed polymorphonuclear cells (PMNs) and endothe-adhesion of polymorphonuclear cells (PMNs) to endothelial lial cells, increasing adhesion and subsequent release ofcells and through activation of primed PMN. The main autoan- enzymes, lipid metabolites, and toxic oxygen radicalstigen of ANCA in WG is proteinase 3 (PR3), a neutrophil-

from PMNs [2]. The main autoantigen of c-ANCA isand monocyte-derived neutral serine protease. The associationproteinase 3 (PR3), which is localized in the azurophilicof WG with individuals continuously expressing a high level

of PR3 on the surface of PMNs suggests that PR3 variants or granules as well as in a readily plasma membrane-mobi-altered regulation of PR3 expression might be directly involved lizable pool of secretory vesicles [3] of PMNs and translo-in the pathogenesis of the disease.

cates to the cell surface after activation of PMNs. PR3Methods. We screened the entire coding and promoter se-is involved in the destruction of phagocytosed microor-quences of the PR3 gene for polymorphisms by means of polymer-

ase chain reaction single-strand conformation polymorphism ganisms. It cleaves the connective tissue protein elastin,(PCR-SSCP). Allelic, genotypic, and haplotype frequencies were exhibits chemotactic properties, and regulates the prolif-compared between 79 WG patients and a cohort of 129 healthy eration of granulopoietic progenitor cells [4]. The “defec-controls.

tive” PiZ-allele of the PR3 inhibitor a1-antitrypsin isResults. Seven single-nucleotide polymorphisms (SNPs), oneassociated with WG, another hint for a pathogenic roleamino acid change (Val119Ile), one 84 bp insertion/deletion,

and a microsatellite were identified. An association with WG of PR3 in the development of WG [5–7]. Together withcould be demonstrated for the A-564G polymorphism in the two other neutral proteases, the leukocyte elastasePR3 promoter affecting a putative transcription factor-binding

(ELA2) and azurozidin (AZU), the PR3 gene is localizedsite.within a 50 kb region on chromosome 19p13.3. Recently,Conclusions. This study excludes certain PR3 epitope vari-

ants as autoantigenic stimuli in WG, since the Val119Ile poly- two other serine proteases were identified in near prox-morphism showed no differences between patients and con- imity to the AZU-PR3-ELA2 cluster [8]. All five genestrols. Overexpression of PR3, however, might predispose the code for a family of neutral serine proteases. They arepatient to the development of autoimmune ANCA-associated

transcribed in the same orientation; they have the samevasculitis.principal structure consisting of five exons and similarexon/intron boundaries.

The aim of our study was to investigate whether ge-netic variants of the PR3 gene are associated with thedevelopment of PR3-ANCA1 systemic vasculitis.1 Present address: SmithKline Beecham Pharmaceuticals, Harlow, Es-

sex, CM19 5AW England, United Kingdom.

Key words: ANCA, Val119lle polymorphism, small vessel vasculitis, METHODSsystemic autoimmune disease, transcription factor binding site, granu-

Patients and controlslomatous lesions, azurophilic granules.

A total of 79 patients with WG was included in thisReceived for publication February 24, 2000study. In each patient, vasculitis was diagnosed fromand in revised form May 23, 2000

Accepted for publication June 9, 2000 biopsies taken from affected organs (upper respiratorytract, 12 patients; lung and kidneys, 67 patients). The 2000 by the International Society of Nephrology

2473

Gencik et al: Proteinase 3 gene polymorphisms and WG2474

Table 1. Oligonucleotides used for PCR-SSCP and microsatellite analysis

PR3-region Name Primer sequence

Promotor Ia 34U 59-CTG CAG GGT GCT AGG CAT CA-39148L 59-CAG GAT TCT CAA TCA AGG GGT G-39

Promotor Ib 108U 59-GTG TCC TCA GAC CTC ACC CAG GG-39230L 59-GCC CTC AGT GCC TCA CCT G-39

Promotor II 195U 59-AGA CAA AAC TGG CAG GAG GA-39421L 59-CAC AGC CCT CAG AGC AAC AC-39

Promotor III 391U 59-GGC GTC TCC TTG TGG GTG GC-39573L 59-CCA CTT CCT CCT TTT GCC TTG GG-39

Promotor IV 527U 59-CCA AAG CCC CCA CCT CCT-39731R 59-ACT CAC CGC TCA GCA GCA AG-39

Exon 1 347F 59-AAG AGG AGC TTG ACC GTG GG-39511R 59-TAT TTC CCC ACA ATC CAC CG-39

Intron 1 84 bp ins/del 579F 59-ACC CAG CCT TCA GCA AAT G-391075R 59-TGC TCT CTT CTT CCC CTG ATT-39

Exon 2 2810F 59-CGC CTG GAC TCC CCC CCT GC-393008R 59-CGG GCG GAC AGG GGT GTG GA-39

Exon 3 3220F 59-CCA GGG CGC CGA GGA GTG AC-393418R 59-CTC TGC CGT GCC CCT CCG AG-39

Exon 4 5190F 59-CCT GGA AGC AGT GCT CAC CG-395518R 59-AGA GGT GGC GTG GAA CCC TC-39

Intron 4 microsatellite 6148F 59-CAG AGC GAG ACT CTG TCG CA-396433R 59-CTG TTT CAG AAT AAA AAT GCT TCC-39

Exon 5 6501F 59-TGA CTG GCT GTC CCC ATC TC-396839R 59-TCT GTC CAA CGC TCC GCC TC-39

Abbreviations are: PCR-SSCP, polymerase chain reaction single strand conformation polymorphism; ins/del, insertion/deletion.

diagnosis of WG was determined in accordance with the urea. The gels were run at 48C at 55 W. The V119Ipolymorphism was retyped by demonstrating an addi-definitions for systemic vasculitides by the Chapel Hill

consensus conference (CHC) [9] and the criteria defined tional TaiI restriction site in exon 3. For DNA sequenc-ing, PCR products were gel purified by means of a gelby the American College of Rheumatology (ACR) [10].

All patients originated from Bavaria, gave informed con- extraction kit (Quiaquick; Quiagen, Chatsworth, CA,USA) and were sequenced with the dye terminator cyclesent, and were diagnosed and treated in the departments

of medicine of the following hospitals: Medizinische Klinik sequencing on an ABI377 automatic sequencer.und Poliklinik, Klinikum Innenstadt, Medizinische Klinik I,

StatisticsKlinikum Großhadern, Ludwig-Maximilians-University,Munchen; Department of Medicine, Klinikum Hohe Allele and genotype frequencies were compared be-

tween the groups with the x2 or Fisher’s exact test. Hap-Warte, Bayreuth; and Department of Nephrology, St.Joseph Hospital, Regensburg, Germany. lotype frequencies were estimated using the software

ARLEQUIN using the maximum likelihood methodSerum PR3-ANCAs were demonstrated by indirectimmunofluorescence using ethanol-fixed neutrophils, [13]. Prediction of transcription factor (TF) binding sites

was performed using the program Transcription Elementand their reactivity with PR3 was confirmed in enzyme-linked immunosorbent assays (ELISA) specific for PR3 Search (TES) [14].(Elias, Freiburg, Germany). The clinical parameters ofour cohort have been described previously [11].

RESULTSA matched group of 129 healthy southern German

Polymorphisms in the PR3 genedonors were used as controls.Using SSCP analyses, altogether eight single-nucleo-

DNA analyses tide polymorphisms (SNPs) were detected in the pro-moter and in the exonic sequences of the PR3 gene (Fig. 1).DNA was extracted according to the protocol by

Miller, Dykes, and Poleski [12]. For the promoter se- In the coding sequence, one SNP was found, changingthe first base of the codon 119 from A to G leading toquence analyses, primers were designed in order to am-

plify five overlapping PCR fragments. Each of the five an amino acid change from Valin to Isoleucin. Anotherpolymorphism (C503T) was detected in intron 1, 31 bpexons were amplified, including the relevant splice sites.

The primers used in the study are listed in Table 1. from the 39 end of the first exon. By comparison ofsequences from public databases, an 84 bp deletion/inser-Analyses of single-strand conformation polymorphism

(SSCP) were performed on 6% polyacrylamide (PAA) tion polymorphism was found in intron 1. The sequencesof both alleles are shown in Figure 2. In intron 4, agels with either 10% glycerol or 5% glycerol and 1 mol/L

Gencik et al: Proteinase 3 gene polymorphisms and WG 2475

Fig. 1. Schematic overview of the proteinase 3 (PR3) gene with the corresponding polymorphisms (top) and transcription factor (TF) bindingsite changes (bottom) as estimated by the program, Transcription Element Search (TES).

Fig. 2. Insertion/deletion polymorphism in intron 1 of the PR3 gene.

Table 2. Diallelic polymorphisms in the PR3 gene as well as allelefrequencies in patients with WG and the control groups, including

the corresponding P and OR values

Polymorphism WG N Controls N P OR

119Val/Ile 64.7%/35.3% 75 59.4%/40.6% 117Intron 1 ins/del 67.6%/32.4% 51 56.9%/43.1% 116 .0.05 1.6

503C/T 60.8%/39.2% 74 69.0%/31.0% 79 .0.05 1.4228C/T 84.6%/15.4% 68 77.5%/22.5% 129 .0.05 1.62172C/T 78.2%/21.8% 62 78.7%/21.3% 1012236C/T 84.9%/15.1% 63 78.9%/21.1% 102 .0.05 1.52413G/C 60.2%/39.8% 65 52.0%/48.0% 99 .0.05 1.42536C/T 77.3%/22.7% 75 76.2%/23.8% 1072564A/G 50.8%/49.2% 66 65.6%/34.4% 106 ,0.01 0.5

Abbreviations are: WG, Wegener’s granulomatosis; N, number; OR, odds ratio.

Table 3. Allelic frequencies of the intron 4 microsatellite in patientswith Wegener’s granulomatosis (WG) and controls Fig. 3. PR3 haplotype frequencies [A-564G; C503T] in controls and

patients with Wegener’s granulomatosis as estimated by the programMicrosatellite alleleArlequin. Symbols are: (j) Wegener’s granulomatosis; ( ) controls.

1 2 3 4 5 6 71819 N

WG 3.2% 2.5% 37.3% 32.3% 20.3% 4.4% 0% 79Controls 6.5% 0% 45.2% 25.7% 20.0% 1.3% 1.2% 115

Haplotype frequency estimation with the software Ar-lequin revealed a strong preponderance of the halpotype[2564G; 503C] in patients with WG (39.4%, SD 5 4.3)when compared with the control group (13.4%, SD 5highly polymorphic (ga) microsatellite was identified. Six

additional polymorphisms were detected in the promoter 4.7, P , 0.00001, OR 5 4.2; Fig. 3).region of the PR3 gene: C-28T, C-172T, C-236T, G-413C,

Promoter polymorphisms change potential TFC-536T, and A-564G.binding sites

Allele and haplotype frequencies Analysis of the polymorphisms in the promoter regionof the PR3 gene revealed several changes in sequencesAllele frequencies of diallelic polymorphisms in pa-

tients with WG and in healthy controls are shown in for TF binding sites by means of TES. The A allele ofthe A-564G polymorphism is a CACCC-TF recognitionTable 2. The allele frequency distribution of the (ga)

microsatellite in intron 4 is shown in Table 3. No simple sequence, whereas the G allele represents a GC-boxelement, which is the Sp1-TF binding site. Furthermore,association of a certain allele could be demonstrated.

Gencik et al: Proteinase 3 gene polymorphisms and WG2476

the T allele of the C-536T does not bind the CAC-TF. allele of the intronic polymorphism. The functional sig-Similarly, the T allele of the C-236T polymorphism adds nificance of this latter polymorphism is not obvious. Thea T-cell transcription factor E box (TFEB) recognition vicinity to exon/intron boundaries on both sides mightsite (Fig. 1). influence splicing efficacy. On the other hand, a putative

enhancer/repressor element might be influenced as well.Increased expression of PR3 may be due to an interac-DISCUSSIONtion of SP1 with a protein that selectively recognizes

The development of autoimmune phenomena such as the C allele of the C503T polymorphism. However, theautoantibodies is still not clearly understood, although relevance of the [2564G; 503C] haplotype for the expres-genetic factors are known to play a role in many of sion of PR3 can be evaluated only in functional experi-these processes. In WG and other antibody-associated ments. Such expression studies are currently underwaysystemic vasculitis, genetic factors involved in the patho- in our laboratory. Furthermore, there are additionalgenesis of the diseases have been described recently hints in our patient group for the existence of an ex-[11, 15]. However, the contribution of genetic variants tended haplotype spanning over the entire PR3 locus.of PR3, which is the autoantigen in WG, is entirely un- The [2564G; 503C] haplotype appears to be the statisti-known. Several studies described epitopes on PR3, which cally demonstrable core part of it.are recognized by autoantibodies from sera of patients From the results of our experiments, we conclude thatwith WG [16, 17]. In order to identify genetic factors that the PR3 locus is involved in the pathogenesis of WG.might predispose to the development of PR3-ANCA, we The association of the disease with the [2564G; 503C]have screened the entire coding sequence of PR3 in a haplotype in the PR3 gene represents further evidencecohort of 79 patients. The Val119Ile previously described for the etiologic role of common polymorphisms in the[18] is located in a relevant surface region of PR3 [17];

development of autoimmune vasculitides such as WG.however, no statistically significant over-representationwas obvious for this antigenic variant in our panel of

ACKNOWLEDGMENTSWG patients. However, beside the 119Val, there are atThis work was supported by a FoRUM grant (F 179/99). The authorsleast seven additional major immunodominant amino

thank Professors E. Held, D. Schlondorff, D. Seybold, F.E. Scherber-acid residues on the surface of PR3 [17]. Although our ich, and W. Samtleben for enabling us to study patients treated in theirdata exclude a major contribution of the Val119Ile poly- departments of medicine.morphism in the etiology of WG, it still may be involved

Reprint requests to Dr. Martin Gencik, Molekulare Humangenetik,in the pathogenesis of a subgroup of patients.Ruhr-Universitat Bochum, D-44780 Bochum, Germany.

A second possibility for increased autoantigenicity of E-mail: martin.gencik@ruhr-uni-bochum.dePR3 might be an altered expression on the cell surface.Indeed, in a recent report, the cell surface expression of REFERENCESPR3 was found to differ between individuals [19]. The

1. Savage COS, Harper L, Adu D: Primary systemic vasculitides.authors observed a stable interindividual variation of Lancet 349:553–558, 1997

2. Kallenberg CGM, Brouver E, Mulder AHL, Stegman CA,PR3 expression in freshly isolated nonstimulated as wellWeening JJ, Cohentervaert JW: ANCA: Pathophysiology revis-as stimulated PMNs. Furthermore, the high expressionited. Clin Exp Immunol 100:1–3, 1995

phenotype was found to be associated with ANCA-asso- 3. Witko-Sarsat V, Cramer EM, Hieblot C, Guichard J, Nusbaumciated systemic vasculitis [20]. Our results may provide P, Lopez S, Lesavre P, Halbwachs-Mecarelli L: Presence of

proteinase 3 in secretory vesicles: Evidence of a novel, highlyan explanation for this observation. The polymorphismsmobilisable intracellular pool distinct from azurophil granules.identified in the promoter region of PR3 (C-236T, C-536T, Blood 94:2487–2496, 1999

and A-564G) change TF recognition sites, which may 4. Skold S, Rosberg B, Gullberg U, Olofsson T: A secreted pro-form of neutrophil proteinase 3 regulates the proliferation of gra-lead to altered transcriptional activity of the PR3 genenulopoietic progenitor cells. Blood 93:849–856, 1999and an increased expression of PR3 on PMN. 5. Segelmark M, Elzouki AN, Wieslander J, Eriksson S: The PiZ

The over-representation of the G allele of the A-564G gene of alpha 1-antitrypsin as a determinant of outcome in PR3-ANCA-positive vasculitis. Kidney Int 48:844–850, 1995polymorphism, which leads to the formation of a novel

6. Elzouki AN, Segelmark M, Wieslander J, Eriksson S: StrongGC-box, the binding site for Sp1-TF, also supports thislink between the alpha 1-antitrypsin PiZ allele and Wegener’s

hypothesis. The transcription-enhancing capacity of Sp1 granulomatosis. J Intern Med 236:543–548, 19947. Esnault VL, Testa A, Audrain M, Roge C, Hamidou M, Barrierelements decreases with the distance from the TATA

JH, Sesboue R, Martin JP, Lesavre P: Alpha 1-antitrypsin geneticbox [21], here at position 237. However, several distantpolymorphism in ANCA-positive systemic vasculitis. Kidney Int

and functionally active SP1 sites have been described 43:1329–1332, 19938. Jenne DE: Gene structure of ANCA target antigens: Implicationsfor immunoglobulin k, collagen I, collagen II, b-globin,

for the pathogenesis of vasculitis. Sarcoidosis Vasc Diffuse LungCD13, and gb110 [22–27].Dis 13:209–213, 1996

The frequency estimation of haplotypes revealed a 9. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, GrossWL, Hagen EC, Hoffman GS, Hunder GG, Kallenberg CG:strong association of 2564G combined with the C503

Gencik et al: Proteinase 3 gene polymorphisms and WG 2477

Nomenclature of systemic vasculitides: Proposal of an international for T-cell alloresponses to myeloid leukemia. J Immunother 22:1–6,1999consensus conference. Arthritis Rheum 37:187–192, 1994

19. Halbwachs-Mecarelli L, Bessou G, Lesavre P, Lopez S, Witko-10. Leavitt RY, Fauci AS, Bloch DA, Michel BA, Hunder GG,Sarsat V: Bimodal distribution of proteinase 3 (PR3) surface ex-Arend WP, Calabrese LH, Fries JF, Lie JT, Lightfoot RWpression reflects a constitutive heterogeneity in the polymorphonu-Jr: The American College of Rheumatology 1990 criteria for theclear neutrophil pool. FEBS Lett 374:29–33, 1995classification of Wegener’s granulomatosis. Arthritis Rheum 33:1101–

20. Witko-Sarsat V, Lesavre P, Lopez S, Bessou G, Hieblot C, Prum1107, 1990B, Noel LH, Guillevin L, Ravaud P, Sermet-Gaudelus I, Timsit11. Gencik M, Borgmann S, Zahn R, Albert E, Sitter T, EpplenJ, Grunfeld JP, Halbwachs-Mecarelli L: A large subset of neu-JT, Fricke H: Immunogenetic risk factors for anti-neutrophil cyto-trophils expressing membrane proteinase 3 is a risk factor forplasmic antibody (ANCA)-associated systemic vasculitis. Clin Expvasculitis and rheumatoid arthritis. J Am Soc Nephrol 10:1224–Immunol 117:412–417, 19991233, 199912. Miller SA, Dykes DD, Poleski HF: A simple salting out proce-

21. Segal R, Berk AJ: Promoter activity and distance constraints ofdure for extracting DNA from human nucleated cells. Nucleicone versus two Sp1 binding sites. J Biol Chem 266:20406–20411,Acids Res 16:1215, 1988199113. Schneider S, Kueffer J-M, Roessli D, Excofier L: Arlequin:

22. Costa MW, Atchison ML: Identification of an Spl-like elementA Software for Population Genetic Data Analysis (version 1.1).within the immunoglobulin kappa 39 enhancer necessary for maxi-Geneva, Genetics and Biometry Laboratory, Department of An-mal enhancer activity. Biochemistry 35:8662–8669, 1996thropology, University of Geneva, 1997

23. Caterina JJ, Ciavatta DJ, Donze D, Behringer RR, Townes14. Schug J, Overton GC: TESS: Transcription Element Search Soft-TM: Multiple elements in human beta-globin locus control regionware on the WWW, (Technical Report CBIL-TR-1997–1001-v0.0,59 HS 2 are involved in enhancer activity and position-independent,of the Computational Biology and Informatics Laboratory). Phila-transgene expression. Nucleic Acids Res 22:1006–1011, 1994delphia, School of Medicine, University of Pennsylvania, 1997 24. Pogulis RJ, Freytag SO: Contribution of specific cis-acting ele-15. Gencik M, Meller S, Borgmann S, Sitter T, Menezes-Saecker ments to activity of the mouse pro-alpha 2 (I) collagen enhancer.AM, Fricke H, Epplen JT: The association of CD18 alleles with J Biol Chem 268:2493–2499, 1993

anti-myeloperoxidase subtypes of ANCA-associated systemic vas- 25. Savagner P, Krebsbach PH, Hatano O, Miyashita T, Liebmanculitides. Clin Immunol 94:9–12, 2000 J, Yamada Y: Collagen II promoter and enhancer interact synergis-

16. Bini P, Gabay JE, Teitel A, Melchior M, Zhou JL, Elkon KB: tically through Sp1 and distinct nuclear factors. DNA Cell BiolAntineutrophil cytoplasmic autoantibodies in Wegener’s granulo- 14:501–510, 1995matosis recognize conformational epitope(s) on proteinase 3. J 26. Olsen J, Kokholm K, Troelsen JT, Laustsen L: An enhancerImmunol 149:1409–1415, 1992 with cell-type dependent activity is located between the myeloid

17. Williams RC Jr, Staud R, Malone CC, Payabyab J, Byres L, and epithelial aminopeptidase N (CD 13) promoters. Biochem JUnderwood D: Epitopes on proteinase-3 recognized by antibodies 322:899–908, 1997from patients with Wegener’s granulomatosis. J Immunol 152:4722– 27. Hamann L, Bayer KU, Jensen K, Harbers K: Interaction of sev-4737, 1994 eral related GC-box and GT-box-binding proteins with the intronic

18. Clave E, Molldrem J, Hensel N, Raptis A, Barrett AJ: Donor- enhancer is required for differential expression of the gb110 genein embryonal carcinoma cells. Mol Cell Biol 14:5786–5793, 1994recipient polymorphism of the proteinase 3 gene: A potential target