Pathogenesis of ANCA-associated Vasculitis




Antineutrophil cytoplasm antibody (ANCA)-associated vasculitis (AAV) comprises a group of systemic inflammatory vasculitides associated with circulating autoantibodies directed against the neutrophil granule components proteinase 3 and myeloperoxidase. ANCA interact with their target antigens on cytokine primed neutrophils, causing neutrophil activation via several signaling pathways that culminates in endothelial interaction, degranulation, cytokine production, and endothelial and tissue damage. The presence of autoantibodies implies the assistance of autoreactive T-helper cells and B cells, and a failure of regulatory mechanisms. This article reviews the current evidence for the pathogenic mechanisms culminating in autoantibody production, the effects of ANCA-neutrophil and neutrophil-endothelial interactions, and the mechanisms of tissue damage.


Wegener’s granulomatosis (WG), microscopic polyangiitis (MPA), and Churg-Strauss syndrome are rare small-vessel vasculitides characterized by necrotizing inflammation of blood vessel walls. Classification criteria for the 3 diseases were agreed on at the Chapel Hill Consensus Conference. The diseases are associated with circulating autoantibodies directed against the neutrophil granule components myeloperoxidase (MPO) and proteinase 3 (PR3). The presence of antineutrophil cytoplasm antibodies (ANCA) and shared pathologic and clinical features have led to MPA and WG often being termed ANCA-associated vasculitis (AAV). This review focuses mainly on WG and MPA.


AAV may cause localized or systemic disease, the commonest organ systems involved being the kidneys and the upper and lower respiratory tract. The clinical presentation is highly variable, but the disease may eventually lead to renal or respiratory failure or failure of other organ systems. Left untreated the disease has high mortality, and current treatment regimens have improved 5-year survival to between 49% and 90%. Current treatments with corticosteroids and cyclophosphamide, while effective, are associated with significant morbidity and mortality.


Much of the research and debate around the pathogenesis of AAV has centered on the role of ANCA in the disease, and this is reviewed herein. The presence of autoantibodies also implies a loss of self tolerance and a role for T-helper cells and B cells in their production, as well as a defect in regulatory T-cell function.


Role of ANCA in ANCA-associated vasculitis


One of the most hotly debated topics in AAV research has been the pathogenic significance of ANCA. There is now a substantial body of evidence supporting a pathogenic role for ANCA in AAV. This evidence consists of in vitro experiments demonstrating that ANCA activate cytokine primed neutrophils, animal models of vasculitis that mimic some of the features of human disease and the transplacental transfer of anti-MPO ANCA, leading to neonatal disease.




In vitro activation of neutrophils by ANCA


The potential involvement of these antibodies in the disease process was suggested in 1990 when Falk and colleagues proposed a pathogenic mechanism whereby ANCA could lead to vascular damage. In vitro experiments demonstrated that neutrophils express the antigens for ANCA on the cell surface. Following cytokine priming, ANCA can activate neutrophils resulting in adhesion to endothelium, superoxide generation, degranulation, and the production of proinflammatory cytokines.


Several cytokines, such as tumor necrosis factor α (TNF-α), interleukin (IL)-18, and granulocyte macrophage colony stimulating factor can prime neutrophils for ANCA activation. The cytokine priming may increase surface expression of ANCA antigens, mobilize components of the NADPH oxidase complex, and perform other functions necessary for subsequent neutrophil activation by ANCA.


Studies of the intracellular signaling cascades of neutrophils following ANCA activation have demonstrated a role for multiple signaling pathways. ANCA IgG Fc binding to the Fcγ receptors appears to activate tyrosine kinases, whereas the F(ab′)2 portion of ANCA activates a separate G-protein pathway, including activation of phosphatidylinositol-3 kinase-γ, and an increased guanosine triphosphatase (GTPase) activity. These distinct pathways appear to converge on a GTPase, p21 ras . p21 ras plays a central role in many cellular processes in the neutrophil, including the respiratory burst. ANCA F(ab′)2 binding and intact ANCA IgG binding activate only 1 of the 2 isoforms of p21 ras present in the neutrophil, with little effect on the other isoform. Karussis and colleagues have successfully used an inhibitor of p21 ras , farnesylthiosalicylic acid, in an animal model of autoimmune encephalitis to selectively inhibit active cells without affecting other immune functions. It is therefore theoretically possible that this particular p21 ras isoform could be inhibited to specifically inhibit the ANCA-activated signaling via p21 ras without affecting other p21 ras functions. The effect of ANCA binding to neutrophils on gene transcription has also been investigated. It has been shown that changes in gene expression can reflect disease activity in vivo. Of note, binding of both intact ANCA and F(ab′)2 portions was found to be capable of inducing gene expression in the neutrophil.




In vitro activation of neutrophils by ANCA


The potential involvement of these antibodies in the disease process was suggested in 1990 when Falk and colleagues proposed a pathogenic mechanism whereby ANCA could lead to vascular damage. In vitro experiments demonstrated that neutrophils express the antigens for ANCA on the cell surface. Following cytokine priming, ANCA can activate neutrophils resulting in adhesion to endothelium, superoxide generation, degranulation, and the production of proinflammatory cytokines.


Several cytokines, such as tumor necrosis factor α (TNF-α), interleukin (IL)-18, and granulocyte macrophage colony stimulating factor can prime neutrophils for ANCA activation. The cytokine priming may increase surface expression of ANCA antigens, mobilize components of the NADPH oxidase complex, and perform other functions necessary for subsequent neutrophil activation by ANCA.


Studies of the intracellular signaling cascades of neutrophils following ANCA activation have demonstrated a role for multiple signaling pathways. ANCA IgG Fc binding to the Fcγ receptors appears to activate tyrosine kinases, whereas the F(ab′)2 portion of ANCA activates a separate G-protein pathway, including activation of phosphatidylinositol-3 kinase-γ, and an increased guanosine triphosphatase (GTPase) activity. These distinct pathways appear to converge on a GTPase, p21 ras . p21 ras plays a central role in many cellular processes in the neutrophil, including the respiratory burst. ANCA F(ab′)2 binding and intact ANCA IgG binding activate only 1 of the 2 isoforms of p21 ras present in the neutrophil, with little effect on the other isoform. Karussis and colleagues have successfully used an inhibitor of p21 ras , farnesylthiosalicylic acid, in an animal model of autoimmune encephalitis to selectively inhibit active cells without affecting other immune functions. It is therefore theoretically possible that this particular p21 ras isoform could be inhibited to specifically inhibit the ANCA-activated signaling via p21 ras without affecting other p21 ras functions. The effect of ANCA binding to neutrophils on gene transcription has also been investigated. It has been shown that changes in gene expression can reflect disease activity in vivo. Of note, binding of both intact ANCA and F(ab′)2 portions was found to be capable of inducing gene expression in the neutrophil.




Effect of ANCA on neutrophil-endothelial adhesion


Neutrophil-mediated damage of endothelial cells plays a key role in the pathogenesis of AAV, as the histologic features of AAV include fibrinoid necrosis of the endothelium. The earliest changes involve swelling, necrosis, and dehiscence of the vascular endothelial cells. This process exposes the basement membrane, leading to platelet adherence, thrombosis, and vessel occlusion. Upregulation of adhesion molecules and the release of chemoattractant proteins by endothelial cells are likely to promote adhesion and transmigration of leukocytes, contributing to the endothelial damage.


In vitro studies have demonstrated that neutrophils exposed to ANCA adhere to TNF-α activated human umbilical vein endothelial cells (HUVEC). ANCA induce adhesion of neutrophils to endothelial cells in vitro via the upregulation and conformational change of adhesion molecules such as CD11b. The key steps involved in the model of ANCA-stimulated neutrophil-endothelial cell interaction are summarized in Fig. 1 .




Fig. 1


Summary of the proposed model for ANCA-stimulated neutrophil-endothelial interaction. Stage 1: Unprimed neutrophils are present in the circulation and the endothelial cells are not activated. Stage 2: Cytokine priming leads to increased surface expression of ANCA antigens (MPO and PR3), integrins, chemokine receptors, and selectin ligands on the neutrophil cell surface. Cytokine activation of endothelial cells leads to increased expression of intercellular adhesion molecules (CAMs), E-selectin, and chemokines. Stage 3: ANCA interacts with its antigens on the neutrophil surface and cross-links the Fc gamma receptors, leading to conformation change in integrins. Stage 4: Rolling adhesion of the neutrophil on the endothelial surface is mediated by the interaction of CAMs and their ligands on the neutrophil (integrins). There may be some differences between various microvascular circulations; in the glomerulus it may be that neutrophils do not undergo rolling adhesion prior to firm adhesion and arrest. Stage 5: Firm adhesion and arrest is mediated by interaction between E-selectin and its ligands and glycosaminoglycan-bound chemokines on endothelial cells and chemokine receptors on the neutrophil. Stage 6: Neutrophils transmigrate the endothelial cell monolayer, undergo respiratory burst and degranulation with release of toxic products, leading to endothelial damage.


This process has been investigated using a flow model whereby neutrophils are perfused over monolayers of endothelial cells replicating the shear stresses found in capillaries. Pretreatment of the neutrophils with ANCA was shown to increase the stability of the adhesion of neutrophils, as well as significantly increasing neutrophil transmigration. This adhesion of neutrophils to endothelium was shown to be dependent on activation of β2 integrins and CXCR2. Of note, a recent study has demonstrated that anti-PR3 ANCA, interacting with adherent cytokine primed neutrophils, increase membrane PR3 expression, providing a positive feedback loop that may amplify anti-PR3 ANCA–induced responses in the small vessels where adhesion occurs.


ANCA have also been shown to induce F-actin polymerization in neutrophils, decreasing neutrophil deformability. This process may lead to impaired passage of neutrophils through capillary beds, such as in the glomerulus, promoting interaction with the endothelium and endothelial damage.


ANCA may play a role in neutrophil-mediated cytotoxicity of endothelial cells. It is known that ANCA can stimulate cytokine primed neutrophils to release both reactive oxygen species and proteases. These factors have been proposed as potential instigators of endothelial cell damage. However, it has recently been shown that endothelial cells inhibit the production of superoxide by neutrophils, and that the release of von Willebrand factor (vWF) by endothelial cells (a well-recognized marker of endothelial cell damage) was more dependent on serine proteases released from neutrophils. Taken together, these findings suggest that the release of intracellular granule contents, including serine proteases, may be more important than the respiratory burst in causing endothelial cell damage. It is also possible that degranulating or necrotic neutrophils release MPO or PR3, which may subsequently be internalized by endothelial cells, causing the release of intracellular reactive oxygen species or endothelial cell apoptosis. Further evidence of endothelial damage during active disease comes from studies reporting increased numbers of circulating endothelial cells and endothelial cell–derived microparticles during active disease that decrease in remission. As well as neutrophil-mediated damage, angiopoietin-2 released from Weibel-Palade bodies may act in a paracrine/autocrine manner to promote further vascular inflammation and endothelial cell detachment. Numbers of circulating endothelial cell progenitor cells are reduced before and during active disease, and it has been suggested that this reflects increased consumption during endothelial repair.




Animal models of ANCA-associated vasculitis


Although there are no good animal models of anti-PR3 AAV, there has been recent interest in murine and rat models of anti-MPO AAV.


MPO −/− mice immunized with purified MPO generate high avidity anti-MPO antibodies. Transfer of splenocytes from these mice to Rag2 −/− mice or transfer of anti-MPO containing IgG to wild-type mice led to the development of necrotizing and crescentic glomerulonephritis, granulomatous inflammation, and systemic necrotizing vasculitis. The addition of lipopolysaccharide (LPS) enhanced renal injury and increased levels of TNF-α. This effect was ameliorated by anti–TNF-α antibodies, demonstrating the importance of cytokines in this pathogenic process. The model has also confirmed the central role of neutrophils in the pathogenesis of AAV. Neutrophil and macrophage infiltration was prominent at the sites of glomerular injury, and neutrophil-depleted mice did not develop crescentic glomerulonephritis.


Intravital microscopy studies of the cremaster muscle of wild-type mice (pretreated with cytokines), showed that anti-MPO IgG enhances leukocyte adhesion and transmigration. This effect was shown to be dependent on Fcγ receptors and β2 integrins (CD18), as there was no increase in adherence or transmigration in Fc receptor γ chain −/− mice, or following coadministration of anti-CD18 antibodies. Others have found that the mechanisms involved in anti-MPO–induced neutrophil adhesion is dose dependent; low-dose anti-MPO in the presence of LPS induced endothelial adhesion in a β2-integrin dependent manner whereas high-dose anti-MPO in the absence of LPS induced adhesion in a β4-integrin dependent manner.


Other studies of ANCA activation of neutrophils in mice and humans have suggested a role for complement in the development of inflammation and glomerular damage, with immune complex deposition in renal biopsies demonstrated by electron microscopy. In vitro studies demonstrate ANCA activation of cytokine-primed neutrophils results in activation of the alternative complement pathway and that C5a acting via the neutrophil C5a receptor further primes neutrophils for ANCA activation. C5a-deficient mice were resistant to induction of glomerulonephritis following transfer of murine anti-MPO antibodies. Recent studies in human renal biopsy tissue have demonstrated the presence of the membrane attack complex, C3d, and factors B and P in the glomeruli and microvasculature, and that complement deposition was associated with more severe renal disease.


A recent rat model of anti-MPO ANCA vasculitis, termed experimental autoimmune vasculitis (EAV), has shown interesting results. Wistar-Kyoto rats immunized with purified human MPO in adjuvant develop anti-MPO antibodies as well as pauci-immune crescentic glomerulonephritis and lung hemorrhage. In this model intravital microscopy demonstrated increased leukocyte adhesion and transmigration in response to a rat homologue of IL-8 (CXC Ligand-1, a chemokine that acts on neutrophils), as well as in naïve rats after passive transfer of anti-MPO IgG from animals with EAV.


In humans, the most direct evidence for the pathogenicity of ANCA comes from a case study in which an infant was born to a mother with active MPA. The infant developed pulmonary-renal syndrome 48 hours after delivery and was found to have serum MPO-ANCA titers similar to the mother. It is hypothesized that the neonate’s disease developed due to transplacental transfer of pathogenic ANCA (and possibly cytokines). The neonate was treated with steroids and plasma exchange, and recovered.




Effect of ANCA on neutrophil apoptosis


Neutrophil activation usually ends in cell death by apoptosis, which should not trigger an inflammatory response or expose the internal components of the cell to the immune system. It has been shown that when primed neutrophils interact with ANCA, the neutrophils undergo dysregulated and accelerated apoptosis, which can lead to secondary necrosis and the release of toxic intracellular contents. Opsonization of neutrophils by ANCA may lead to enhanced macrophage uptake, and the release of proinflammatory cytokines such as IL-1 and IL-8 to perpetuate inflammation. Furthermore, apoptosis of nonprimed neutrophils leads to translocation of granule contents to the cell surface, which may thus provide an additional mechanism whereby ANCA can interact with the intracellular granule contents PR3 and MPO.




Other functions of the ANCA antigens


The membrane expression of PR3 (mPR3) is bimodal, with an mPR3-positive neutrophil subset distinguishable from a membrane PR3-negative neutrophil subset within a given individual. Although the proportion of mPR3-expressing neutrophils appears to be relatively stable within an individual, there is a wide variation between individuals (0%–100%). This variation has been shown to be genetically determined, perhaps related to the major histocompatibility complex HLA region. High levels of mPR3 expression have been associated with susceptibility to vasculitis and disease relapse in WG. It has also been demonstrated that patients with AAV show increased transcription of PR3 and MPO compared with healthy controls. During the normal process of neutrophil maturation, the transcription of granule constituents (including MPO and PR3) is terminated by the time the neutrophil leaves the bone marrow. In patients with AAV, mRNA transcripts for PR3 and MPO were found in mature circulating monocytes and neutrophils.


The role of ANCA target antigens may extend beyond cross-linking of ANCA on the surface of neutrophils. The serum levels of PR3 have also been shown to be elevated in AAV in both active disease and remission, and in MPO-ANCA and PR3-ANCA vasculitis. Circulating serum PR3 may contribute to the pathogenesis of disease by interaction with endothelial cells, leading to IL-8 and MCP-1 production, recruiting neutrophils and macrophages, or causing endothelial cell apoptosis. PR3 may also increase the activity of multiple inflammatory cytokines.


Anti-MPO ANCA have also been shown to interact with MPO directly, leading to generation of hypochlorous acid, which was strongly cytolytic in culture.


α-1 Antitrypsin is the main physiologic inhibitor of PR3, and anti-PR3 ANCA has been shown to inhibit the inactivation of PR3 by α-1 antitrypsin. In one study, this inhibitory effect of anti-PR3 ANCA correlated more closely with disease activity than the serum levels of ANCA. However, the main evidence that the inhibition of PR3 by α-1 antitrypsin is important in the pathogenesis of AAV stems from the observation that deficiency of α-1 antitrypsin is associated with AAV. The z allele (which leads to reduced α-1 antitrypsin activity) for α-1 antitrypsin is more common in patients with WG, especially in those with more severe disease.


The mechanisms by which the interaction of ANCA and neutrophils may contribute to microvascular inflammation are varied, and may include inappropriate degranulation at the endothelial surface, altered interaction with endothelial cells (with increased transmigration across the vessel wall and endothelial cell damage), and altered apoptosis and necrosis of neutrophils, leading to the release of proinflammatory cytokines.




Lysosomal membrane protein 2


Antibodies against lysosomal membrane protein 2 (LAMP-2) have recently been reported in AAV. LAMP-2 ANCA are capable of damaging human microvascular endothelium and inducing focal necrotizing glomerulonephritis in rats. Of note, LAMP-2 antibodies recognize an epitope that has homology with a bacterial adhesion molecule, “Fim H.” It is postulated that bacteria expressing the Fim H antigen could act as a molecular mimic to LAMP-2, thus triggering the production of LAMP-2 ANCA alongside anti-Fim H antibodies.




Antiendothelial cell antibodies


ANCA are the autoantibodies most closely associated with WG and MPA, but other autoantibodies have been reported in patients with these diseases, including antibodies against endothelial cell components. Some in vivo evidence of a pathogenic role for antiendothelial cell antibodies (AECA) exists, whereby mice were immunized with purified IgG AECA (from a patient with WG) and subsequently developed perivascular lymphocytic infiltrate of venules and arterioles but no glomerulonephritis. AECA are prevalent in systemic vasculitis in humans as well as other inflammatory conditions, and levels may fluctuate with disease activity. AECA also demonstrate some organ specificity: AECA specific to human nasal, kidney, and lung endothelial cells are found more frequently than AECA specific to HUVEC in patients with WG. Mechanisms for a pathogenic role for AECA have been proposed. It is suggested that AECA may up-regulate endothelial surface adhesion molecules and secretion of inflammatory cytokines, promoting adhesion of activated neutrophils and monocytes as well as cytotoxicity. It is conceivable that some of the pathogenic activity of AECA is due to anti–LAMP-2 antibodies.




The role of lymphocytes cells in ANCA-associated vasculitis


T Cells


The role of lymphocytes in AAV is less well defined than the role of ANCA in activating neutrophils. T cells are a predominant infiltrating cell in the interstitium of the kidney with crescentic glomerulonephritis, and are found in other sites of inflammation. ANCA are class switched high-affinity autoantibodies implying a role for T-cell help in their production, the presence of autoreactive T cells, and a failure of regulatory T-cell function. In addition, patients with resistant disease have been reported to respond to T-cell–depleting treatment with antithymocyte globulin. Similarly, depleting CD4+ T cells in a mouse model of anti-MPO vasculitis attenuated the glomerular crescent formation, despite the MPO ANCA titer being unaffected.


Following activation T-helper cells differentiate into 1 of 4 distinct lineages depending on, amongst other things, the cytokine environment. Each lineage is characterized by the presence of specific transcription factors and patterns of cytokine.


Both Th1 and Th2 cell populations have been described in the tissues in AAV. High levels of interferon-γ production following nonspecific stimulation of these T cells suggest a predominantly Th1-driven response. However, this conclusion is challenged by other studies reporting a predominantly Th2 response in WG. MPO- and PR3-specific T cells have been reported in patients with AAV, with proliferative responses to MPO or PR3 in culture and strong production of IL-6 and IL-10. At the time it was suggested that this may represent a Th2 response; however, in light of better understanding of Th17 cells (whose differentiation depends on IL-6 among other cytokines) and Treg cells, it may be possible to interpret these data differently. Recently a population of PR3-specific Th17 cells has been described in ANCA-positive patients. IL-17 may be involved in neutrophil migration and stimulate the generation of other proinflammatory cytokines.


The role of regulatory T cells has also been studied. The number and function of CD4+ CD25+ Treg cells has been shown to be reduced in AAV, especially in active disease, and reduced numbers of Treg correspond to slower times to remission and increased relapse rates.


Peripheral blood T cells in patients with AAV have an abnormal phenotype and appear to be persistently activated, during periods of both active disease and remission. Overall, there appears to be a peripheral lymphopenia, with low levels of CD4+ T-helper cells but a relative increase in the subset of CD4+ memory cells. It is not understood how these differences in T-cell phenotype between healthy individuals and AAV patients relate to disease pathogenesis. Many previous studies were done before Th17 cells were recognized as a separate lineage and before human Treg cells were characterized. Many of the apparent inconsistencies in previous studies (the different reporting of Th1 and Th2 associations) and apparent persistent activation (such as high CD25 expression) may become more comprehensible as our understanding of T-cell biology improves.


B Cells


The importance of B cells in AAV is evident, due to their role in the production of ANCA. Histologically B cells are found in affected tissues in WG, such as nasal lesions and the renal interstitial infiltrate of those with renal involvement. The proportion of active circulating B cells is also increased in active WG compared with disease in remission or healthy controls.


The presence of autoreactive B cells implies a breakdown of normal mechanisms that ensure tolerance to self antigens. CD19 is a B-cell coreceptor that enhances B-cell receptor signal transduction, and surface expression of this coreceptor has been shown to be approximately 20% lower in AAV than in healthy controls. It is possible that autoreactive B cells with low surface expression of CD19 may escape normal tolerance mechanisms due to the reduced strength of signaling achieved via the B-cell receptor.


Granuloma formation is a feature of WG often found in the lungs and upper airways, and may be a feature of early or “localized” disease. Histologic examination of granulomatous inflammation in WG demonstrated B-cell clusters in close proximity to numerous PR3-positive cells. An analysis of the functional immunoglobulins from WG tissue was supportive of an antigen-driven selection process. It is therefore possible that this early granulomatous inflammation provides a site for the selection and maturation of anti-PR3 ANCA producing B cells, contributing to progression to systemic vasculitis.


Recently there has been interest in the role of the anti-CD20 antibody rituximab in the treatment of autoimmune diseases, including AAV. CD20 is expressed on mature B cells but not plasma cells, yet there are several case series now describing rapid remission induction and reduction in ANCA titers following peripheral B-cell depletion with rituximab treatment. A recently completed randomized controlled trial comparing cyclophosphamide with rituximab suggests that rituximab is as effective as cyclophosphamide (in conjunction with corticosteroids) at inducing disease remission.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Oct 1, 2017 | Posted by in RHEUMATOLOGY | Comments Off on Pathogenesis of ANCA-associated Vasculitis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access