REVIEW

Pathophysiology of ANCA-associated
vasculitides: are ANCA really pathogenic?
J.W. Cohen Tervaert, P. Heeringa
Department of Clinical and Experimental Immunology, Cardiovascular Research Institute
Maastricht, University Hospital Maastricht, Maastricht, the Netherlands

ABSTRACT

The strong relation between antineutrophil cytoplasmic diseases that is limited to the kidneys, namely idiopathic
autoantibodies (ANCA) and primary vasculitic syndromes necrotising crescentic glomerulonephritis (NCGN), are
suggests a pathophysiological role for ANCA. closely associated with antineutrophil cytoplasmic auto-
Experimental evidence for the pathogenic potential of ANCA antibodies (ANCA). The lesions in those diseases, particu-
has been derived from in vitro studies that demonstrate larly demonstrated in the kidneys, are ‘pauci-immune’,
that ANCA can activate tumour necrosis factor  primed meaning that no immune deposits are found in most cases.
neutrophils, monocytes and/or endothelial cells. The ANCA in these diseases are directed against proteinase 3
binding of ANCA to primed neutrophils results in activation (PR3) or myeloperoxidase (MPO) (table 1). Besides being a
of these cells by a process that is largely dependent on helpful diagnostic tool, determination of ANCA levels can
engagement of -2 integrins and on the interaction of the also be useful for monitoring disease activity, since
Fc portion of ANCA. An Fc-independent mechanism is, relapses of disease are often preceded by rises in ANCA
however, also operative. In experimental animal models, it levels.1,2 Furthermore, persisting high levels of ANCA are
has been demonstrated that immunisation with myeloper- associated with a poor renal outcome.3 The strong relation
oxidase (MPO) induces MPO-ANCA. The induction of between ANCA and primary vasculitic syndromes suggests
ANCA, however, is not sufficient to induce vasculitis in rats an important role for ANCA in the pathophysiology.
since immune complexes first have to be deposited along the Experimental evidence for the pathogenic potential of
vessel wall before lesions develop. When MPO-deficient mice ANCA has been derived from both in vitro and in vivo
are, however, immunised with murine MPO, anti-MPO studies and will be reviewed.
immunoglobulins are purified and subsequently injected into
mice that are not deficient for MPO, systemic vasculitis and
glomerulonephritis is induced. These experiments suggest P AT H O P H Y S I O L O G Y O F A N C A -
that ANCA indeed induces vasculitis. Risk factors for break- A S S O C I AT E D VA S C U L I T I S :
ing self-tolerance to ANCA antigens are genetic factors, IN VITRO D ATA
drugs, chemical substances and/or infectious agents.
In vitro, ANCA can activate neutrophils primed with tumour
necrosis alpha (TNF-) for the production of reactive oxygen
intermediates (ROI), the release of lysosomal enzymes, and
INTRODUCTION the secretion of interleukin-1.4,5 Furthermore, it has been
demonstrated that ANCA are able to stimulate neutrophils
Within the group of small-vessel vasculitides, Wegener’s to adhere to cultured human endothelial cells, a process
granulomatosis (WG), microscopic polyangiitis (MPA), that can be inhibited by anti-CD18 antibodies.6 Johnson et
Churg Strauss Syndrome (CSS) and the form of these al. have clarified the mechanism by which ANCA might

© 2003 Van Zuiden Communications B.V. All rights reserved.

DECEMBER 2003, VOL. 61, NO. 12

404
Table 1
Characteristics of ANCA-associated vasculitides

DISEASE CLINICAL MPO ANCA PR3 ANCA

Churg-Strauss syndrome Asthma, eosinophilia, neuropathy >70% of patients <10% of patients
Wegener’s granulomatosis Nose bleeds, nephritis, lung lesions 10-30% of patients >70% of patients
Microscopic polyangiitis Nephritis, purpura, haemoptysis 30-70% of patients 30-70% of patients
Idiopathic NCGN Nephritis >70% of patients 10-30% of patients

NCGN = necrotising crescentic glomerulonephritis.

stimulate neutrophil adherence by showing that ANCA be operative, antibody-dependent cell-mediated cytotoxicity,
stimulate the upregulation of CD11b on neutrophils and cell-mediated immune responses.17
in vitro.7 Finally, it has been demonstrated that ANCA- Pathogenic immune complexes (ICX) deposition-mediated
stimulated primed neutrophils can lyse cytokine-pretreated vasculitis is best depicted in the serum sickness animal
cultured endothelial cells.8,9 Apart from inducing the model. In this model, rabbits are injected with bovine
activation of neutrophils, ANCA also activate monocytes10 serum albumin and about seven to ten days later immune
and/or endothelial cells.11,12 complexes are found that may induce vasculitis and/or
The mechanisms involved in ANCA-mediated neutrophil glomerulonephritis. In ANCA-associated vasculitis, however,
activation are not completely understood. Upon priming immune complexes are generally not found in the lesions.
with TNF-, neutrophils express PR3 and MPO on the Therefore, the classic renal lesion in ANCA glomerulo-
cell surface which then become accessible for interaction nephritis is labelled ‘pauci-immune’. In kidney biopsies
with ANCA.4 It is thought that PR3 and MPO binding to of patients with ANCA-associated glomerulonephritis, we
the cell membrane is both through charge interactions found no IgA or IgG deposits and only nonspecific IgM
and receptor mediated. Binding of ANCA to primed deposits in a minority of the patients.18 Complement
neutrophils may result in activation of neutrophils, a deposition, however, is often present (in about 50% of the
process that is largely dependent on engagement of 2 cases).18 This may point to prior ICX deposition, but there
integrins and the interaction of the Fc portion of ANCA.13 is no proof for this hypothesis.19
A Fc-independent mechanism, however, has also been To test the hypothesis that ANCA themselves may induce
described to be operative in vitro.4,14 Recently, Ben Smith vasculitis,20 we immunised BN rats with human MPO
et al. demonstrated that ligation of FcRIIa and FcRIIIb which induced antibodies to human MPO which also
is necessary for ANCA-induced neutrophil activation. cross-reacted with rat MPO.21 Furthermore, in these rats
Since the signalling cascades that are used by ANCA a cellular response to MPO could be detected.21 To our
are different from the signal pathways used by FcR surprise, these rats appeared completely normal and no
engagement only, it is suggested that apart from Fc vasculitic lesions were found at autopsy. So, the induction
engagement also other membrane co-factors are used by of ANCA is not sufficient to induce vasculitis in rats. We
ANCA for neutrophil activation.15 These other membrane hypothesised that there must first be ICX deposition at
co-factors have not been identified yet. The signals involved vessel walls. These ICX then attract neutrophils and these
in neutrophil activation have been recently dissected and neutrophils then express MPO on their cell surface that
include p38 mitogen-activated protein kinase (MAPK) may bind anti-MPO, which results in an overstimulation
and extracellular signal-regulated kinase (ERK) as well as of the neutrophils resulting in vasculitis and also the rapid
phosphatidylinositol 3 kinase control systems.16 disappearance of ICX. To test this hypothesis, we injected
MPO-immunised rats with an extract of neutrophils
containing MPO, and hydrogen peroxide (H2O2). In this
P AT H O P H Y S I O L O G Y O F A N C A - context we predicted that ICX deposition and subsequently
A S S O C I AT E D VA S C U L I T I S : A N I M A L vasculitis would occur. We observed vasculitis of the lungs
STUDIES and the gut in the rats that were immunised with MPO
but not in the rats that were not MPO immunised and
Several types of pathophysiological events that may lead to received the neutrophil extract only.22 Unfortunately, no
vessel wall damage in vasculitis are currently recognised. glomerulonephritis was found. However, after unilateral
These include pathogenic immune complex deposition or perfusion of the left kidney with the neutrophil extract
in situ formation, a ‘Shwartzman-like’ phenomenon in which and H2O2 we saw a severe form of necrotising crescentic
intravascular activation and aggregation of neutrophils may glomerulonephritis in rats that had been immunised with

Cohen Tervaert, et al. ANCA-associated vasculitides.

DECEMBER 2003, VOL. 61, NO. 12

405
MPO and no lesions in nonimmunised rats.21 More INDUCTION OF AUTOIMMUNITY TO
importantly, immediately after perfusion, ICX deposits were P R 3 A N D/ O R M P O
seen in the kidneys, but these disappeared very quickly
and when the glomerulonephritis was at its maximum The central mechanism in autoimmunity is the breaking
no further immune deposits were detected.21 So, in the of self-tolerance. It is now well established that autoreactive
presence of ANCA, severe vasculitis and ‘pauci-immune’ T and B cells exist in the blood of healthy individuals and
glomerulonephritis can be induced in rats when immune that these cells can potentially induce autoimmunity if
complexes are first deposited along the vessel wall. The activated beyond a certain threshold.25 A combination of
next question was what would happen if ICX other than risk factors may be present in patients who develop ANCA.
ANCA/MPO ICX are deposited along the vessel wall. To Inherited determinants have been sought, i.e. associations
study this, Heeringa et al. injected rats with an antibody with certain HLA class I or class II molecules and/or with
to the rat glomerular basement membrane (GBM) and the C3F component of complement, but have not been
compared MPO-immunised rats with non-immunised convincingly found in patients with ANCA-associated
rats. For these studies, a low dose of anti-GBM antibody vasculitis and/or glomerulonephritis. Other genetic factors,
was used that binds to the GBM but is not enough to however, were found to be involved in ANCA-associated
induce a glomerulonephritis.23 In rats with anti-MPO a vasculitis. These genes include the genes for PR3, MPO,
severe glomerulonephritis developed whereas no lesions FcR, 1 antitrypsin, CD18 and/or CTLA-4.26 In addition,
were found in the non-immunised rats. environmental factors are probably important modulators.
In mice, ANCA have been identified in MRL-lpr-/-lpr and Among these, drugs such as propylthiouracil and/or
in SCG-/-Kj mice. In these models the role of ANCA is, hydralazine27 and/or exposure to chemical substances
however, difficult to tease out from the complex back- such as silicon28 have been incriminated. Infectious
grounds of polyclonal B cell activation. Recently, however, agents are, however, the most likely candidates to cause
convincing evidence was obtained that ANCA are sufficient autoimmunity. Several mechanisms by which infectious
to cause systemic ‘pauci-immune’ vasculitis and glomeru- agents might induce autoimmunity have been postulated.29
lonephritis in vivo.24 Two major strategies were used to These include molecular mimicry, abnormal presentation of
demonstrate this. In the first, MPO-deficient mice were self-proteins, and/or abnormal stimulation of autoreactive
immunised with murine MPO and developed anti-MPO. T or B cells by agents such as superantigens. This latter
Adoptive transfer of splenocytes from these mice into mechanism has our special attention, since superantigens
immune deficient RAG2-/- mice (lacking functioning B produced by staphylococci that may activate autoreactive
lymphocytes and T lymphocytes) resulted in anti-MPO B cells, in a T-cell dependent way, to produce ANCA
and the development of glomerulonephritis and capillaritis. depositions are often present in patients with ANCA-
In contrast, transfer of splenocytes from mice that were associated vasculitis.30
immunised with BSA into RAG2-/- mice resulted in a mild
form of immune complex glomerulonephritis without
crescents. The nature of the background immune complex CONCLUSION
disease found in RAG2-/- mice that received either
splenocytes from mice immunised with MPO or BSA is Vascular damage in ANCA-associated vasculitis and/or
unclear. It was hypothesised that this relatively nonspecific glomerulonephritis results predominantly from activation
response may represent a form of graft versus host disease. of neutrophils when these cells adhere to endothelial
In the second strategy, purified anti-MPO was intravenously cells. These neutrophils may be initially attracted by
injected into RAG2-/- mice or wild type. ‘Pauci-immune’ immune complexes formed in situ. Once these adherent
necrotising and crescentic glomerulonephritis and systemic activated neutrophils are ‘over-stimulated’ by ANCA they
vasculitis, closely resembling the human disease, were may cause necrotising vasculitis and glomerulonephritis
observed.24 These experiments indicate that ANCA can and, in addition, stimulate the rapid disappearance of
produce vasculitis without the further participation of T immune complexes, thus explaining the absence of
lymphocytes and/or B lymphocytes. This suggests that immune complexes in tissue biopsies.
ANCA indeed induce vasculitis.
From these experiments, we come to our current working
hypothesis. ANCA induce activation of neutrophils and REFERENCES
monocytes resulting in ICX deposition in vessel walls.
Other antigens, however, may also be involved in ICX 1. Cohen Tervaert JW, Woude FJ van der, Fauci AS, et al. Association between
formation. In the presence of ANCA this ICX deposition active Wegener’s granulomatosis and anticytoplasmic antibodies. Arch
results in persistent activation of neutrophils and monocytes Intern Med 1989;149:2461-5.
and subsequently severe glomerulonephritis or vasculitis. 2. Boomsma MM, Stegeman CA, Leij MJ van der, et al. Prediction of relapses

Cohen Tervaert, et al. ANCA-associated vasculitides.

DECEMBER 2003, VOL. 61, NO. 12

406
 in Wegener’s granulomatosis by measurement of antineutrophil cytoplasmic ventional Fc gamma receptor ligation. Blood 2001;98:1448-55.
antibody levels: a prospective study. Arthritis Rheum 2000;43:2025-33. 16. Kettritz R, Choi M, Butt W, et al. Phosphatidylinositol 3-kinase controls
3. Franssen CF, Stegeman CA, Oost-Kort WW, et al. Determinants of renal antineutrophil cytoplasmic antibodies-induced respiratory burst in
outcome in anti-myeloperoxidase-associated necrotizing crescentic human neutrophils. J Am Soc Nephrol 2002;13:1740-9.
glomerulonephritis. J Am Soc Nephrol 1998;9:1915-23. 17. Cohen Tervaert JW, Kallenberg CGM. The role of autoimmunity to
4. Falk RJ, Terrell RS, Charles LA, Jennette JC. Anti-neutrophil cytoplasmic myeloid lysosomal enzymes in the pathogenesis of vasculitis. In:
autoantibodies induce neutrophils to degranulate and produce oxygen Hansson GK, Libby P (ed). Immune functions of the vessel wall. London:
radicals in vitro. Proc Natl Acad Sci USA 1990;87:4115-9. Harwoord Academic Publishers, 1996:99-120.
5. Brooks CJ, King WJ, Radford DJ, Adu D, McGrath M, Savage CO. IL-1 beta 18. Brouwer E, Huitema MG, Mulder AH, et al. Neutrophil activation in vitro
production by human polymorphonuclear leucocytes stimulated by anti- and in vivo in Wegener’s granulomatosis. Kidney Int 1994;45:1120-31.
neutrophil cytoplasmic autoantibodies: relevance to systemic vasculitis. 19. Brons RH, Kallenberg CG, Tervaert JW. Are antineutrophil cytoplasmic
Clin Exp Immunol 1996;106:273-9. antibody-associated vasculitides pauci-immune? Rheum Dis Clin North
6. Ewert BH, Becker ME, Jennette JC, Falk RJ. Antimyeloperoxidase antibodies Am 2001;27:833-48.
induce neutrophil adherence to cultured human endothelial cells. Ren 20. Heeringa P, Brouwer E, Cohen Tervaert JW, Weening JJ, Kallenberg CG.
Fail 1995;17:125-33. Animal models of anti-neutrophil cytoplasmic antibody associated vasculitis.
7. Johnson PA, Alexander HD, McMillan SA, Maxwell AP. Up-regulation of Kidney Int 1998;53:253-63.
the endothelial cell adhesion molecule intercellular adhesion molecule-1 21. Brouwer E, Huitema MG, Klok PA, et al. Antimyeloperoxidase-associated
(ICAM-1) by autoantibodies in autoimmune vasculitis. Clin Exp Immunol proliferative glomerulonephritis: an animal model. J Exp Med
1997;108:234-42. 1993;177:905-14.
8. Ewert BH, Jennette JC, Falk RJ. Anti-myeloperoxidase antibodies stimulate 22. Heeringa P, Foucher P, Klok PA, et al. Systemic injection of products of
neutrophils to damage human endothelial cells. Kidney Int 1992;41:375-83. activated neutrophils and H2O2 in myeloperoxidase-immunized rats
9. Savage CO, Pottinger BE, Gaskin G, Pusey CD, Pearson JD. Autoantibodies leads to necrotizing vasculitis in the lungs and gut. Am J Pathol
developing to myeloperoxidase and proteinase 3 in systemic vasculitis 1997;151:131-40.
stimulate neutrophil cytotoxicity toward cultured endothelial cells. Am J 23. Heeringa P, Brouwer E, Klok PA, et al. Autoantibodies to myeloperoxidase
Pathol 1992;141:335-42. aggravate mild anti-glomerular-basement-membrane-mediated glomerular
10. Weidner S, Neupert W, Goppelt-Struebe M, Rupprecht HD. Antineutrophil injury in the rat. Am J Pathol 1996;149:1695-706.
cytoplasmic antibodies induce human monocytes to produce oxygen 24. Xiao H, Heeringa P, Hu P, et al. Antineutrophil cytoplasmic autoantibodies
radicals in vitro. Arthritis Rheum 2001;44:1698-706. specific for myeloperoxidase cause glomerulonephritis and vasculitis in
11. Johnson PA, Alexander HD, McMillan SA, Maxwell AP. Up-regulation of the mice. J Clin Invest 2002;110:955-63.
granulocyte adhesion molecule Mac-1 by autoantibodies in autoimmune 25. Kotb M. Short analytical review. Infections and autoimmunity: a story of
vasculitis. Clin Exp Immunol 1997;107:513-9. the host, the pathogen, and the copathogen. Clin Immunol
12. Muller Kobold AC, Wijk RT van, Franssen CF, Molema G, Kallenberg CG, Immunopathol 1995;74:10.
Cohen Tervaert JW. In vitro up-regulation of E-selectin and induction of 26. Kallenberg CG, Rarok A, Stegeman CA. Genetics of ANCA-associated
interleukin-6 in endothelial cells by autoantibodies in Wegener’s granulo- vasculitides. Cleve Clin J Med 2002;69:SII61-3.
matosis and microscopic polyangiitis. Clin Exp Rheumatol 1999;17:433-40. 27. Merkel PA. Drugs associated with vasculitis. Curr Opin Rheumatol
13. Reumaux D, Vossebeld PJ, Roos D, Verhoeven AJ. Effect of tumor necrosis 1998;10:45-50.
factor-induced integrin activation on Fc gamma receptor II-mediated signal 28. Cohen Tervaert JW, Stegeman CA, Kallenberg CG. Silicon exposure and
transduction: relevance for activation of neutrophils by anti-proteinase 3 vasculitis. Curr Opin Rheumatol 1998;10:12-7.
or anti-myeloperoxidase antibodies. Blood 1995;86:3189-95. 29. Cohen Tervaert JW, Popa ER, Bos NA. The role of superantigens in vasculitis.
14. Kettritz R, Jennette JC, Falk RJ. Crosslinking of ANCA-antigens stimulates Curr Opin Rheumatol 1999;11:24-33.
superoxide release by human neutrophils. J Am Soc Nephrol 1997;8:386-94. 30. Popa ER, Stegeman CA, Kallenberg CG, Cohen Tervaert JW.
15. Ben-Smith A, Dove SK, Martin A, Wakelam MJ, Savage CO. Antineutrophil Staphylococcus aureus and Wegener’s granulomatosis. Arthritis Res
cytoplasm autoantibodies from patients with systemic vasculitis activate 2002;4:77-9.
neutrophils through distinct signaling cascades: comparison with con-

Cohen Tervaert, et al. ANCA-associated vasculitides.

DECEMBER 2003, VOL. 61, NO. 12

407