- Email: [email protected]
Antineutrophil Cytoplasmic Autoantibody-Associated Diseases: A Rheumatologist's Perspective
Antineutrophil Cytoplasmic Autoantibody-Associated Diseases: A Rheumatologist's Perspective Wolfgang L. Gross, MD, Wilhelm H. Schmitt, MD, and Elena Csernok, DiplBiol • Autoantibodies against neutrophil cytoplasmic antigens (ANCA) produce two major immunofluorescence (IF) patterns on ethanol-fixed granulocytes: the "classical" (centrally accentuated) C-ANCA, associated with Wegener's granulomatosis (WG), and P-ANCA (perinuclear), which mainly occur in renal vasculitis. Rheumatic manifestations are an important clinical finding in systemic vasculitis, often preceding a fulminant course and sometimes imitating various rheumatic disorders. We analyzed the incidence of ANCA in rheumatic patients and looked for the frequency of rheumatic symptoms in systemic vasculitis. In WG (n = 186), we found rheumatic symptoms in 55% (myalgia, 45%; arthritis, 21%); in 90%, rheumatic complaints were associated with active vasculitis. In 730 patients with various rheumatic conditions (eg, 268 rheumatoid arthritis, 130 systemic lupus erythematosis [SLE], 32 sharp-S, 50 ankylosing spondylitis, 43 systemic sclerosis) no C-ANCA were found. On the contrary, the P-ANCA pattern was seen in seven of 62 giant cell arteritis, five of 27 Felty's/Still's syndrome, and four of 130 SLE patients in addition to renal vasculitis (21/74). We demonstrated that 95% of C-ANCA-positive sera react with proteinase 3 (PR3 or myeloblastin). USing monoclonal antibodies, we showed that PR3 is expressed on the plasma membrane of neutrophil granulocytes and monocytes; thus, PR3 autoantigens are accessible for circulating antibodies. The detection of ANCA in sera from vasculitis and other rheumatic diseases is of immunodiagnostic value and provides new insight in the pathogenesis of systemic vasculitides. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Antineutrophil cytoplasmic autoantibodies; Wegener's granulomatosis; systemic vasculitis; rheumatic diseases; proteinase 3.
HISTORICAL PERSPECTIVE
A
NTIBODIES against leukocytes have been shown to be associated with a variety of autoimmune disorders in general, and with rheumatic diseases in particular. Whereas autoantibodies against granulocytes in the classic forms of collagen vascular diseases (eg, systemic lupus erythematosus [SLED are mainly directed against the CD 12-16 clusters belonging to myeloid antigens, which are expressed as granulocyte plasma membrane proteins, there are others that for a long time have been thought not to be accessible for the antibody because of their intracellular location. The most prominent of this latter group are the so-called granulocyte-specific antinuclear antibodies (GS-ANA). GS-ANA (IgG) are common in rheumatoid arthritis and may be found in drug-related syndromes. A close correlation between the presence of IgG GS-ANA in serum and progression of joint damage was demonstrated by Wiik in 1980. I These autoantibodies have been well known in clinical immunology and rheumatology for at least two decades. In contrast, data on serum IgG antibodies against intracellular, but extranuclear, components of polymorphonuclear neutrophilic granulocytes and monocytes (first called anticytoplasmic autoantibodies [ACPA], later antineutrophil cytoplasmic autoantibodies [ANCAD were first published in the early 1980s. In 1982 and 1984, two
groups from Australia2 ,3 reported on antineutrophil antibodies in patients with segmental necrotizing glomerulonephritis (GN) and with a diffuse systemic disease associated with polyarthralgias and lung involvement. In 1985, a DutchDanish collaboration4 discovered that a distinct cytoplasmic fluorescence pattern (see below: ACPA, C-ANCA) was associated with Wegener's granulomatosis (WG). In 1986, we confirmed their findings on an extended number of well-defined WG patients, characterized both by clinical and histological findings. 5 These data were extended over the years and proved the high immunodiagnostic value of the cytoplasmic ANCA (C-ANCA) in the serological differential diagnosis of vasculitides. 6-8 So, the C-ANCA test still stands the test of time. Later, we and others9 demonstrated the occurrence of other neutrophil fluorescence staining patterns, distinct from the classic C-ANCA asFrom the Department of Clinical Rheumatology, Medical University Lubeck and Rheumaclinic, Bad Bramstedt, Germany. Supported by Bundesministeriumfur Forschung und Technologie Grant No. 01 VM8622. Address reprint requests to Wolfgang L. Gross, MD, Abteilung fur Klinische Rheumatologie der Medizinischen, Universitdt zu Lubeck, and Rheumaklinik, Oskar-Alexander Str 26,2357 Bad Bramstedt, Germany. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1802-0007$3.00/0
American Journal of Kidney Diseases, Vol XVIII, No 2 (August), 1991: pp 175-179
175
GROSS, SCHMITI, AND CSERNOK
176
sociated with WG. Falk and Jennette 9 showed that the most significant of these other patterns is the perinuclear fluorescence (perinuclear ANCA [P-ANCA]), which is frequently associated with renal vasculitis. These findings prompted a world-wide clinical and basic scientific interest in the role of ANCA in WG and other diseases associated with vasculitis. The clinical value of testing for ANCA could be established for a variety of disease entities and the target antigens of the various ANCA specificities were detected. 1O Four years after the first publication of the correlation between ANCA and WG, we have been able to demonstrate that CANCA in at least 95% of subjects are directed against proteinase 3 (PR3).11 PR3 is a serine protease in human polymorphonuclear (PMN) cells and monocytes. P-ANCA is induced by autoantibodies against a variety of myeloid lysozomal enzymes, eg, those against myeloperoxidase (MPO), elastase (HLE), cathepsin G (CTG), lactoferrin, etc. 12,13 Therefore, in addition to the screening procedure of ANCA by the standard immunofluorescence technique, further analyses are required to define the specificity of the autoantibody(ies).14 CLINICAL FINDINGS
Despite their low incidence, the vasculitides represent a major challenge for many disciplines in clinical medicine. Most of them are difficult to recognize early. They are associated with a bad prognosis, especially if treated inadequately. They can mimic a wide variety of clinical entities of other origin. Rheumatic manifestations are observed frequently in the group of primary vasculitides. On the other hand, in the group of collagen vascular diseases (eg, SLE) and the various forms of arthritis (eg, rheumatoid arthritis), vasculitis can severely complicate the course of the disease. Before the introduction of ANCA, no disease-specific laboratory marker was available for the group of vasculitis. Fortunately, ANCA have shed some long awaited light into the darkness of both differential diagnosis and etiopathogenesis of vasculitides. Indeed, C-ANCA-or better anti-PR3 antibodies-have become the third diagnostic criterion for WG. They have extended our understanding of the clinical spectrum of the disease, especially with respect to the rec-
ognition of early forms, variants, and "formes frustes." Musculoskeletal involvement with particular reference to arthralgia and arthritis was reported decades ago. More recently, rheumatic manifestations in a large series ofWG patients have been described as arthralgia, myalgia, and/or arthritis in two thirds of patients. 15 Twenty-eight percent had nonerosive and nondeforming polyarthritis. Sacroiliitis was found in three of 50 and relapsing polychondritis in two of 50 patients with WG. Rheumatoid factor was present in half of the patients tested. In our own series of 186 patients with WG (148 biopsy-proven), we found episodes with arthralgia, myalgia, or arthritis in 55% of all and in 58% of the biopsy-proven cases. Thus, rheumatic complaints have been found to be the fourth main symptom complex (Fig 1). They are more frequent than symptoms of the eye, skin, or nervous system. Mostly, patients suffered from myalgia (45%). However, frank arthritis was seen in 21 %, mainly in the larger joints (monoarthritis, 10 patients; oligoarthritis, five; polyarthritis, six (Fig 2). Approximately 90% of the WG patients suffering from rheumatic symptoms had a generalized form of the disease. In 53%, other signs of disease activity were found. The rheumatic symptoms were usually seen together with constitutional symptoms, which are most typical for active vasculitis (fever, weight loss, night sweats, adynamia). So, the arthralgia-myalgia complex, or even arthritis, has to be seen as a sign of clinical activity in a systemic (or classic) form ofWG. Anamnestic data from WG patients apparently starting with fulminant disease (eg, pulmonary-renal syn-
kidney upper respir. tract lung rheumatic sympt. eyes constitu. sympt. skin
n
186
1===1IIII':1r-
eNS GI-tract - - - ' " heart + large vessels r-P;~-+-+--+----J
Fig 1. Clinical symptoms in WG.
%
177
ANCA: A RHEUMATOLOGIST'S PERSPECTIVE Table 1. ANCA in Collagen Diseases and Other Rheumatic Disorders Arthritis
Collagen Diseases
SLE SLE-vasculitis MCTD Scleroderma Polymyositis/OM
C·ANCA
P-ANCA*
0/109 0/21 0/32 0/43 0/10
4/109 0/21 0/32 0/43 1/10
Rheumatoid arthritis Felty's syndrome Still's disease Sjogren's syndrome Reiter's syndrome Reactive arthritis Ankylosing spondylitis Psoriasis arthritis
C-ANCA
P-ANCA*
0/241 0/14 0/13 0/30 0/9 0/20 0/50 0/32
6/241 3/14 2/13 8/30 0/9 0/20 0/50 5/32
Abbreviations: MCTO, mixed connective tissue disease; OM, dermatomyositis. * Hep-2 control negative.
drome) show that these patients had rheumatic symptoms before the dramatic course of the disease. Furthermore, many patients with renal vasculitis (eg, rapidly progressive GN) suffer very early from rheumatic manifestations. This observation indicates that, even in this "organ-lim-
B
A
ited vasculitis," there is an occult systemic process. When WG starts with rheumatic symptoms in the absence of the classic WG triad, usually other rheumatic diseases have to be ruled out. Since-at least in a few cases-the clinical picture is indistinguishable from early manifestations of spondarthritis, rheumatoid arthritis, or SLE, immunological studies must be performed. In acute generalized WG, one can find IgM rheumatoid factor in more than half of patients. ANA are seen in about 30%. By contrast, CANCA fluorescence (or anti-PR3 antibodies measured by enzyme-linked immunosorbent assay [ELISA]) does not occur in the disease group under differential diagnosis. We could not find one case out of the collagen vascular disease group or the various arthritides to be false-positive for
14%
!
Table 2. ANCA in Vasculitic Syndromes
19% _/~
1-I--iII!!i--
~
00'
9%
\L.tr-~-9%
Wegener's granulomatosis Polyarteritis nodosa Churg-Strauss syndrome
, - - 43%
47%
14% _ _ 00
Fig 2. Localization of arthritis in WG. (A) Small jOints; (8) large joints (n = 186).
Polymyalgia rheumatica Takayasu's arteritis Temporal arteritis Thrombangitis obliterans Behcet's disease Henoch-SchOpnlein purpura Unclassified vasculitis Cryoglobulinemia * Hep-2 control negative. t PR3 ELISA
C-ANCA (+/all)
P-ANCA" (+/all)
295/383 14/49 (5/49)t 4/13 (1/13)t 0/62 0/3 0/24 0/6 0/21 0/18 8/110 0/5
20/383 2/49 1/13 5/62 0/3 2/24 0/6 0/21 0/18 9/110 0/5
GROSS, SCHMITT, AND CSERNOK
178
this test (Tables 1 and 2). For practical reasons, we have to repeat that in approximately one third of cases WG begins with an initial stage oflimited locoregional disease. This form can last months and occasionally several years, generally without causing any life-threatening complications. 6 The stable phase can lead into the generalized stage (classic WG), in which the disease can rapidly run a fatal course if left untreated. Initial symptoms are usually those of the upper respiratory tract, eg, chronic (hemorrhagic) rhinitis, sinusitis, etc. In the lungs, single or multiple nodular infiltrations may be noted. Because of the locoregional symptomatology, the initial stage is puzzling to the diagnostician. Often it takes years until the disease is suspected and finally histologically proven. The transition from the initial stage into the active generalized (systemic) phase is usually initiated by the well-known constitutional symptoms. They are frequently associated with rheumatologic symptoms. Simultaneously, the localized lesions (eg, hemorrhagic rhinitis) intensify. In particular, there is a systemic reaction due to the vasculitis, which becomes more and more prominent. This process includes the "red eye" syndrome (eg, episcleritis), palpable purpura (leukocytoclastic vasculitis), and abnormal laboratory tests indicating severe inflammation (erythrocyte sedimentation rate, C-reactive protein, leukocytosis and thrombocytosis). From the clinical point of view, one wonders whether a more benign granulomatous process is now (secondarily?) complicated by vasculitis. Although the C-ANCA can be detected in sera from patients with initial-phase disease in about 50% of cases, the strong sensitivity of approximately 95% will be reached only in the active generalized WG. In addition, the C-ANCA titer correlates with disease activity, which is mainly dependent on the vasculitic process. PATHOPHYSIOLOGICAL STUDIES
The etiology of all primary vasculitides is unknown. The pathogenesis of the necrotizing vasculitides seems to be closely related to the leukocyte-endothelial interaction. Furthermore, the association between disease activity of WG and levels of anti-PR3 antibodies may suggest that ANCA are a pathogenetic factor and that immunological mechanisms are important in WG and other vasculitides. In addition to the clinical
work depicted, two major questions had to be answered before further speculations could be given. First, the target antigen ("Wegener's autoantigen") ofthe C-ANCA had to be characterized. Second, the accessibility of this autoantigen had to be worked out. Our biochemical and functional characterization studies delineated that Wegener's autoantigen is PR3. 11 More recently, we published the nucleotide and revised amino acid sequence of PR3. 16 Our revised PR3 sequence agrees with the sequence of p29, 17 and is identical to AGP7 (an antimicrobial protein described by Gabay et aI 18 ), except for GIn 19 instead of Glu. In addition, we were surprised to find that PR3 is identical to myeloblastin, a serine protease whose complementary DNA has been cloned from promyelocytic leukemia HL-60 cells. Inhibition of myeloblastin expression by antisense oligonucleotides induces terminal differentiation of monocyte-granulocyte precursor cells and growth arrest. Because most auto antibodies against PR3 interfere with its enzymatic function, excessive neutrophilic leukocytosis seen in fulminant WG could be a result of enhanced granulocyte differentiation induced by C-ANCA. Furthermore, we have investigated the distribution of PR3 within the organelles of resting normal human PMN and monocytes using immunocytochemical techniques on thin frozen sections. 19 To obtain valuable tools for immunolabeling, two murine monoclonal antibodies directed against PR3 were produced. In PMN, we observed immunogold label for PR3, predominantly in azurophil granules and in smaller amounts on the plasma membrane. In monocytes, Wegener's autoantigen could be detected in small granules, which occasionally also contained myeloperoxidase, and rare labeling was found on the monocyte membrane. The finding that PR3 is expressed on the plasma membrane of PMN and monocytes, thereby becoming accessible to circulating C-ANCA, supports the supposition that anti-PR3 autoantibodies are not only markers of disease activity, but also are involved in the pathogenesis ofWG. The circulating autoantibodies may activate neutrophils to release oxygen radicals and lysosomal enzymes, as has been demonstrated in vitro for C-ANCA and antimyeloperoxidase antibodies. This could lead to necrotizing inflammation in the blood vessels. Recently, it has been speculated that the bind-
179
ANCA: A RHEUMATOLOGIST'S PERSPECTIVE
ing of autoantibodies to lysosomal enzymes prevents these enzymes from being inactivated by its natural inhibitor, which could lead to extensive tissue destruction in WG and other vasculitides. We investigated this attractive hypothesis and found that the addition of anti-PR3 autoantibodies of several patients inhibited the elastin-
olytic activity of PR3. Furthermore, the autoantibodies did not influence the inactivation of the enzyme by aI-antitrypsin? I Thus, this hypothesis does not seem relevant as an explanation of inflammatory process in WG. So, the pathophysiological role of the autoantibodies to different myeloid lysosomal enzymes still has to be proven.
REFERENCES I. Wiik A: Granulocyte-specific antinuclear antibodies, possible significance for the pathogenesis, clinical features and diagnosis of rheumatoid arthritis. Allergy 35:263-289, 1980 2.. Davies OJ, Moran JE, Niall JF, et al: Segmental necrotising glomerulonephritis with antineutrophil antibody: Possible arbovirus aetiology? Br Med J 285:606, 1982 3. Hall JB, Wadham BMcN, Wood CJ, et al: Vasculitis and glomerulonephritis: A subgroup with an antineutrophil cytoplasmic antibody. Aust N Z J Med 14:277, 1984 4. van der Woude FJ, Rasmussen N, Lobatto S, et al: Autoantibodies against neutrophils and monocytes: Tool for diagnosis and marker of disease activity in Wegener's granulomatosis. Lancet 1:425, 1985 5. Gross WL, Ludemann G, Kiefer G, et aI: Anti-cytoplasmic antibodies in Wegener's granulomatosis. Lancet 1: 806, 1986 6. Gross WL: Wegener's granulomatosis: New aspects of the disease course, immunodiagnostic procedures, and stageadapted treatment. Sarcoidosis 6:15-29, 1989 7. Nolle B, Specks U , Ludemann J, et al: Anticytoplasmic autoantibodies: Their immunodiagnostic value in Wegener granulomatosis. Ann Intern Med III :28-40, 1989 8. Gross WL, Ludemann J, Csernok E: Anticytoplasmic auto-antibodies ANCA/ACPA in vasculitis. Immunol Newslett 10: 159-163, 1990 9. Falk RJ, Jennette JC: Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Eng! J Med 318:1651-1657, 1988 10. Scott DGI, Sharrack BR, Webb FWS, et al: Antineutrophil cytoplasmic antibodies and rheumatoid arthritis. C1in Exp RheumatoI8:F24, 1990 (suppI4) II. Ludemann J, Utecht B, Gross WL: Anti-neutrophil cytoplasm antibodies in Wegener's granulomatosis recognize an elastinolytic enzyme. J Exp Med 171:357-362, 1990
12. Nlissberger L, Jonsson H, Sjholm AG, et al: Circulating anti-elastase in systemic lupus erythematosus. Lancet 1:509, 1989 13. Cohen Tervaert JW, Goldschmeding R, Elema JD, et
aI: Autoantibodies against myeloid lysosomal enzymes in crescentic glomerulonephritis. Kidney Int 37:799-806, 1990 14. Ludemann J, Utecht B, Gross WL: Laboratory methods for detection of anti-neutrophil cytoplasm antibodies. Immunol Newslett 10:163-168, 1990 15. Noritake DT, Weiner SR, Bassett LW, et aI: Rheumatic manifestations of Wegener's granulomatosis. J Rheumatol 14: 949-951 , 1987 16. Jenne DE, Tschopp J, Ludemann J, et aI: Wegener's autoantigen decoded. Nature 346:520, 1990 17. NilesJL, McCluskey RT, Ahmad MF, etal: Wegener's granulomatosis autoantigen is a novel neutrophil serine proteinase. Blood 74:1888-1893,1989 18. Gabay JE, Scott RW, Campanelli 0 , et al: Antibiotic proteins of human polymorphonuclear leukocytes (bacteria/ Iysosomes/neutrophils). Proc Nat! Acad Sci USA 86:56105614,1989 19. Csernok E, Ludemann J, Gross WL, et al: Ultrastructural localization of proteinase 3, the target antigen of anticytoplasmic antibodies circulating in Wegener's granulomatosis. Am J Pathol 137:5, 1990 20. Falk RJ, Terrell RS, Charles LA, et al: Anti-neutrophil cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Nat! Acad Sci NSA 87:4115-4119, 1990 21. Ludemann J, Utecht B, Gross WL: Anti-cytoplasmic antibodies in Wegener's granulomatosis are directed against proteinase 3, in Horl WH, Heiland A (eds): Proteasis III. New York, NY, Plenum, 1991 (in press)
HISTORICAL PERSPECTIVE
A
NTIBODIES against leukocytes have been shown to be associated with a variety of autoimmune disorders in general, and with rheumatic diseases in particular. Whereas autoantibodies against granulocytes in the classic forms of collagen vascular diseases (eg, systemic lupus erythematosus [SLED are mainly directed against the CD 12-16 clusters belonging to myeloid antigens, which are expressed as granulocyte plasma membrane proteins, there are others that for a long time have been thought not to be accessible for the antibody because of their intracellular location. The most prominent of this latter group are the so-called granulocyte-specific antinuclear antibodies (GS-ANA). GS-ANA (IgG) are common in rheumatoid arthritis and may be found in drug-related syndromes. A close correlation between the presence of IgG GS-ANA in serum and progression of joint damage was demonstrated by Wiik in 1980. I These autoantibodies have been well known in clinical immunology and rheumatology for at least two decades. In contrast, data on serum IgG antibodies against intracellular, but extranuclear, components of polymorphonuclear neutrophilic granulocytes and monocytes (first called anticytoplasmic autoantibodies [ACPA], later antineutrophil cytoplasmic autoantibodies [ANCAD were first published in the early 1980s. In 1982 and 1984, two
groups from Australia2 ,3 reported on antineutrophil antibodies in patients with segmental necrotizing glomerulonephritis (GN) and with a diffuse systemic disease associated with polyarthralgias and lung involvement. In 1985, a DutchDanish collaboration4 discovered that a distinct cytoplasmic fluorescence pattern (see below: ACPA, C-ANCA) was associated with Wegener's granulomatosis (WG). In 1986, we confirmed their findings on an extended number of well-defined WG patients, characterized both by clinical and histological findings. 5 These data were extended over the years and proved the high immunodiagnostic value of the cytoplasmic ANCA (C-ANCA) in the serological differential diagnosis of vasculitides. 6-8 So, the C-ANCA test still stands the test of time. Later, we and others9 demonstrated the occurrence of other neutrophil fluorescence staining patterns, distinct from the classic C-ANCA asFrom the Department of Clinical Rheumatology, Medical University Lubeck and Rheumaclinic, Bad Bramstedt, Germany. Supported by Bundesministeriumfur Forschung und Technologie Grant No. 01 VM8622. Address reprint requests to Wolfgang L. Gross, MD, Abteilung fur Klinische Rheumatologie der Medizinischen, Universitdt zu Lubeck, and Rheumaklinik, Oskar-Alexander Str 26,2357 Bad Bramstedt, Germany. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1802-0007$3.00/0
American Journal of Kidney Diseases, Vol XVIII, No 2 (August), 1991: pp 175-179
175
GROSS, SCHMITI, AND CSERNOK
176
sociated with WG. Falk and Jennette 9 showed that the most significant of these other patterns is the perinuclear fluorescence (perinuclear ANCA [P-ANCA]), which is frequently associated with renal vasculitis. These findings prompted a world-wide clinical and basic scientific interest in the role of ANCA in WG and other diseases associated with vasculitis. The clinical value of testing for ANCA could be established for a variety of disease entities and the target antigens of the various ANCA specificities were detected. 1O Four years after the first publication of the correlation between ANCA and WG, we have been able to demonstrate that CANCA in at least 95% of subjects are directed against proteinase 3 (PR3).11 PR3 is a serine protease in human polymorphonuclear (PMN) cells and monocytes. P-ANCA is induced by autoantibodies against a variety of myeloid lysozomal enzymes, eg, those against myeloperoxidase (MPO), elastase (HLE), cathepsin G (CTG), lactoferrin, etc. 12,13 Therefore, in addition to the screening procedure of ANCA by the standard immunofluorescence technique, further analyses are required to define the specificity of the autoantibody(ies).14 CLINICAL FINDINGS
Despite their low incidence, the vasculitides represent a major challenge for many disciplines in clinical medicine. Most of them are difficult to recognize early. They are associated with a bad prognosis, especially if treated inadequately. They can mimic a wide variety of clinical entities of other origin. Rheumatic manifestations are observed frequently in the group of primary vasculitides. On the other hand, in the group of collagen vascular diseases (eg, SLE) and the various forms of arthritis (eg, rheumatoid arthritis), vasculitis can severely complicate the course of the disease. Before the introduction of ANCA, no disease-specific laboratory marker was available for the group of vasculitis. Fortunately, ANCA have shed some long awaited light into the darkness of both differential diagnosis and etiopathogenesis of vasculitides. Indeed, C-ANCA-or better anti-PR3 antibodies-have become the third diagnostic criterion for WG. They have extended our understanding of the clinical spectrum of the disease, especially with respect to the rec-
ognition of early forms, variants, and "formes frustes." Musculoskeletal involvement with particular reference to arthralgia and arthritis was reported decades ago. More recently, rheumatic manifestations in a large series ofWG patients have been described as arthralgia, myalgia, and/or arthritis in two thirds of patients. 15 Twenty-eight percent had nonerosive and nondeforming polyarthritis. Sacroiliitis was found in three of 50 and relapsing polychondritis in two of 50 patients with WG. Rheumatoid factor was present in half of the patients tested. In our own series of 186 patients with WG (148 biopsy-proven), we found episodes with arthralgia, myalgia, or arthritis in 55% of all and in 58% of the biopsy-proven cases. Thus, rheumatic complaints have been found to be the fourth main symptom complex (Fig 1). They are more frequent than symptoms of the eye, skin, or nervous system. Mostly, patients suffered from myalgia (45%). However, frank arthritis was seen in 21 %, mainly in the larger joints (monoarthritis, 10 patients; oligoarthritis, five; polyarthritis, six (Fig 2). Approximately 90% of the WG patients suffering from rheumatic symptoms had a generalized form of the disease. In 53%, other signs of disease activity were found. The rheumatic symptoms were usually seen together with constitutional symptoms, which are most typical for active vasculitis (fever, weight loss, night sweats, adynamia). So, the arthralgia-myalgia complex, or even arthritis, has to be seen as a sign of clinical activity in a systemic (or classic) form ofWG. Anamnestic data from WG patients apparently starting with fulminant disease (eg, pulmonary-renal syn-
kidney upper respir. tract lung rheumatic sympt. eyes constitu. sympt. skin
n
186
1===1IIII':1r-
eNS GI-tract - - - ' " heart + large vessels r-P;~-+-+--+----J
Fig 1. Clinical symptoms in WG.
%
177
ANCA: A RHEUMATOLOGIST'S PERSPECTIVE Table 1. ANCA in Collagen Diseases and Other Rheumatic Disorders Arthritis
Collagen Diseases
SLE SLE-vasculitis MCTD Scleroderma Polymyositis/OM
C·ANCA
P-ANCA*
0/109 0/21 0/32 0/43 0/10
4/109 0/21 0/32 0/43 1/10
Rheumatoid arthritis Felty's syndrome Still's disease Sjogren's syndrome Reiter's syndrome Reactive arthritis Ankylosing spondylitis Psoriasis arthritis
C-ANCA
P-ANCA*
0/241 0/14 0/13 0/30 0/9 0/20 0/50 0/32
6/241 3/14 2/13 8/30 0/9 0/20 0/50 5/32
Abbreviations: MCTO, mixed connective tissue disease; OM, dermatomyositis. * Hep-2 control negative.
drome) show that these patients had rheumatic symptoms before the dramatic course of the disease. Furthermore, many patients with renal vasculitis (eg, rapidly progressive GN) suffer very early from rheumatic manifestations. This observation indicates that, even in this "organ-lim-
B
A
ited vasculitis," there is an occult systemic process. When WG starts with rheumatic symptoms in the absence of the classic WG triad, usually other rheumatic diseases have to be ruled out. Since-at least in a few cases-the clinical picture is indistinguishable from early manifestations of spondarthritis, rheumatoid arthritis, or SLE, immunological studies must be performed. In acute generalized WG, one can find IgM rheumatoid factor in more than half of patients. ANA are seen in about 30%. By contrast, CANCA fluorescence (or anti-PR3 antibodies measured by enzyme-linked immunosorbent assay [ELISA]) does not occur in the disease group under differential diagnosis. We could not find one case out of the collagen vascular disease group or the various arthritides to be false-positive for
14%
!
Table 2. ANCA in Vasculitic Syndromes
19% _/~
1-I--iII!!i--
~
00'
9%
\L.tr-~-9%
Wegener's granulomatosis Polyarteritis nodosa Churg-Strauss syndrome
, - - 43%
47%
14% _ _ 00
Fig 2. Localization of arthritis in WG. (A) Small jOints; (8) large joints (n = 186).
Polymyalgia rheumatica Takayasu's arteritis Temporal arteritis Thrombangitis obliterans Behcet's disease Henoch-SchOpnlein purpura Unclassified vasculitis Cryoglobulinemia * Hep-2 control negative. t PR3 ELISA
C-ANCA (+/all)
P-ANCA" (+/all)
295/383 14/49 (5/49)t 4/13 (1/13)t 0/62 0/3 0/24 0/6 0/21 0/18 8/110 0/5
20/383 2/49 1/13 5/62 0/3 2/24 0/6 0/21 0/18 9/110 0/5
GROSS, SCHMITT, AND CSERNOK
178
this test (Tables 1 and 2). For practical reasons, we have to repeat that in approximately one third of cases WG begins with an initial stage oflimited locoregional disease. This form can last months and occasionally several years, generally without causing any life-threatening complications. 6 The stable phase can lead into the generalized stage (classic WG), in which the disease can rapidly run a fatal course if left untreated. Initial symptoms are usually those of the upper respiratory tract, eg, chronic (hemorrhagic) rhinitis, sinusitis, etc. In the lungs, single or multiple nodular infiltrations may be noted. Because of the locoregional symptomatology, the initial stage is puzzling to the diagnostician. Often it takes years until the disease is suspected and finally histologically proven. The transition from the initial stage into the active generalized (systemic) phase is usually initiated by the well-known constitutional symptoms. They are frequently associated with rheumatologic symptoms. Simultaneously, the localized lesions (eg, hemorrhagic rhinitis) intensify. In particular, there is a systemic reaction due to the vasculitis, which becomes more and more prominent. This process includes the "red eye" syndrome (eg, episcleritis), palpable purpura (leukocytoclastic vasculitis), and abnormal laboratory tests indicating severe inflammation (erythrocyte sedimentation rate, C-reactive protein, leukocytosis and thrombocytosis). From the clinical point of view, one wonders whether a more benign granulomatous process is now (secondarily?) complicated by vasculitis. Although the C-ANCA can be detected in sera from patients with initial-phase disease in about 50% of cases, the strong sensitivity of approximately 95% will be reached only in the active generalized WG. In addition, the C-ANCA titer correlates with disease activity, which is mainly dependent on the vasculitic process. PATHOPHYSIOLOGICAL STUDIES
The etiology of all primary vasculitides is unknown. The pathogenesis of the necrotizing vasculitides seems to be closely related to the leukocyte-endothelial interaction. Furthermore, the association between disease activity of WG and levels of anti-PR3 antibodies may suggest that ANCA are a pathogenetic factor and that immunological mechanisms are important in WG and other vasculitides. In addition to the clinical
work depicted, two major questions had to be answered before further speculations could be given. First, the target antigen ("Wegener's autoantigen") ofthe C-ANCA had to be characterized. Second, the accessibility of this autoantigen had to be worked out. Our biochemical and functional characterization studies delineated that Wegener's autoantigen is PR3. 11 More recently, we published the nucleotide and revised amino acid sequence of PR3. 16 Our revised PR3 sequence agrees with the sequence of p29, 17 and is identical to AGP7 (an antimicrobial protein described by Gabay et aI 18 ), except for GIn 19 instead of Glu. In addition, we were surprised to find that PR3 is identical to myeloblastin, a serine protease whose complementary DNA has been cloned from promyelocytic leukemia HL-60 cells. Inhibition of myeloblastin expression by antisense oligonucleotides induces terminal differentiation of monocyte-granulocyte precursor cells and growth arrest. Because most auto antibodies against PR3 interfere with its enzymatic function, excessive neutrophilic leukocytosis seen in fulminant WG could be a result of enhanced granulocyte differentiation induced by C-ANCA. Furthermore, we have investigated the distribution of PR3 within the organelles of resting normal human PMN and monocytes using immunocytochemical techniques on thin frozen sections. 19 To obtain valuable tools for immunolabeling, two murine monoclonal antibodies directed against PR3 were produced. In PMN, we observed immunogold label for PR3, predominantly in azurophil granules and in smaller amounts on the plasma membrane. In monocytes, Wegener's autoantigen could be detected in small granules, which occasionally also contained myeloperoxidase, and rare labeling was found on the monocyte membrane. The finding that PR3 is expressed on the plasma membrane of PMN and monocytes, thereby becoming accessible to circulating C-ANCA, supports the supposition that anti-PR3 autoantibodies are not only markers of disease activity, but also are involved in the pathogenesis ofWG. The circulating autoantibodies may activate neutrophils to release oxygen radicals and lysosomal enzymes, as has been demonstrated in vitro for C-ANCA and antimyeloperoxidase antibodies. This could lead to necrotizing inflammation in the blood vessels. Recently, it has been speculated that the bind-
179
ANCA: A RHEUMATOLOGIST'S PERSPECTIVE
ing of autoantibodies to lysosomal enzymes prevents these enzymes from being inactivated by its natural inhibitor, which could lead to extensive tissue destruction in WG and other vasculitides. We investigated this attractive hypothesis and found that the addition of anti-PR3 autoantibodies of several patients inhibited the elastin-
olytic activity of PR3. Furthermore, the autoantibodies did not influence the inactivation of the enzyme by aI-antitrypsin? I Thus, this hypothesis does not seem relevant as an explanation of inflammatory process in WG. So, the pathophysiological role of the autoantibodies to different myeloid lysosomal enzymes still has to be proven.
REFERENCES I. Wiik A: Granulocyte-specific antinuclear antibodies, possible significance for the pathogenesis, clinical features and diagnosis of rheumatoid arthritis. Allergy 35:263-289, 1980 2.. Davies OJ, Moran JE, Niall JF, et al: Segmental necrotising glomerulonephritis with antineutrophil antibody: Possible arbovirus aetiology? Br Med J 285:606, 1982 3. Hall JB, Wadham BMcN, Wood CJ, et al: Vasculitis and glomerulonephritis: A subgroup with an antineutrophil cytoplasmic antibody. Aust N Z J Med 14:277, 1984 4. van der Woude FJ, Rasmussen N, Lobatto S, et al: Autoantibodies against neutrophils and monocytes: Tool for diagnosis and marker of disease activity in Wegener's granulomatosis. Lancet 1:425, 1985 5. Gross WL, Ludemann G, Kiefer G, et aI: Anti-cytoplasmic antibodies in Wegener's granulomatosis. Lancet 1: 806, 1986 6. Gross WL: Wegener's granulomatosis: New aspects of the disease course, immunodiagnostic procedures, and stageadapted treatment. Sarcoidosis 6:15-29, 1989 7. Nolle B, Specks U , Ludemann J, et al: Anticytoplasmic autoantibodies: Their immunodiagnostic value in Wegener granulomatosis. Ann Intern Med III :28-40, 1989 8. Gross WL, Ludemann J, Csernok E: Anticytoplasmic auto-antibodies ANCA/ACPA in vasculitis. Immunol Newslett 10: 159-163, 1990 9. Falk RJ, Jennette JC: Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Eng! J Med 318:1651-1657, 1988 10. Scott DGI, Sharrack BR, Webb FWS, et al: Antineutrophil cytoplasmic antibodies and rheumatoid arthritis. C1in Exp RheumatoI8:F24, 1990 (suppI4) II. Ludemann J, Utecht B, Gross WL: Anti-neutrophil cytoplasm antibodies in Wegener's granulomatosis recognize an elastinolytic enzyme. J Exp Med 171:357-362, 1990
12. Nlissberger L, Jonsson H, Sjholm AG, et al: Circulating anti-elastase in systemic lupus erythematosus. Lancet 1:509, 1989 13. Cohen Tervaert JW, Goldschmeding R, Elema JD, et
aI: Autoantibodies against myeloid lysosomal enzymes in crescentic glomerulonephritis. Kidney Int 37:799-806, 1990 14. Ludemann J, Utecht B, Gross WL: Laboratory methods for detection of anti-neutrophil cytoplasm antibodies. Immunol Newslett 10:163-168, 1990 15. Noritake DT, Weiner SR, Bassett LW, et aI: Rheumatic manifestations of Wegener's granulomatosis. J Rheumatol 14: 949-951 , 1987 16. Jenne DE, Tschopp J, Ludemann J, et aI: Wegener's autoantigen decoded. Nature 346:520, 1990 17. NilesJL, McCluskey RT, Ahmad MF, etal: Wegener's granulomatosis autoantigen is a novel neutrophil serine proteinase. Blood 74:1888-1893,1989 18. Gabay JE, Scott RW, Campanelli 0 , et al: Antibiotic proteins of human polymorphonuclear leukocytes (bacteria/ Iysosomes/neutrophils). Proc Nat! Acad Sci USA 86:56105614,1989 19. Csernok E, Ludemann J, Gross WL, et al: Ultrastructural localization of proteinase 3, the target antigen of anticytoplasmic antibodies circulating in Wegener's granulomatosis. Am J Pathol 137:5, 1990 20. Falk RJ, Terrell RS, Charles LA, et al: Anti-neutrophil cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Nat! Acad Sci NSA 87:4115-4119, 1990 21. Ludemann J, Utecht B, Gross WL: Anti-cytoplasmic antibodies in Wegener's granulomatosis are directed against proteinase 3, in Horl WH, Heiland A (eds): Proteasis III. New York, NY, Plenum, 1991 (in press)