Perinuclear Antineutrophil Cytoplasmic Antibodies in Inflammatory Bowel Disease

Abstract: Antineutrophil cytoplasmic antibodies (ANCAs) are a group of autoantibodies which are specific for granulocyte antigens. By indirect immunofluorescence (llF) on ethanol-fixed neutrophils, ANCAs can be divided into two types: a cytoplasmic staining pattern (C-ANCA) or a perinuclear staining pattern (P-ANCA). Their pathogenic role and initiating stimuli are unknown.This study analysed P-ANCA in monozygotic twins with inflammatory bowel disease. In ulcerative colitis (UC), P-ANCA occurred in 9 of 14 (64.3%) monozygotic twins and in 13 of 21 (61.9%) non-twin subjects, which was significantly different compared with healthy controls who were positive in three of 52 (5.8%) subjects (p<O.OOOl). P-ANCA was found in two of 10 (20%) healthy twin siblings to twins with UC, which was not significantly different from the healthy controls (p=0.18). The results in Crohn's disease (CD) did not differ from healthy controls. This finding does not support the hypothesis that P-ANCA is a subclinical marker of genetic susceptibility to UC.Seventy-six UC patients after proctocolectomy with ileal pouch-anal anastomosis (JP AA) including 28 patients who had had pouchitis and 48 patients who had not, were analysed regarding the correlation between P-ANCA and pouchitis. P-ANCA was found in 49n6 (64.5%) iu UC patients after the operation. Furthermore, we found that patients with recent (512 months) or ongoing pouchitis were all P-ANCA positive and pouchitis patients with higher P-ANCA titres were more prone to have frequent relapses.To screen the occurrence of P-ANCA and detect the target antigen(s), 36 patients with UC, 37 patients with CD, 38 patients with collagenous colitis (CC) and 190 controls were studied. P-ANCA was found in a higher frequency in UC (23/36. 63.9%) than in CD (2/37. 5.4%). CC (4/38, 10.5%) or controls (4/190. 2.1 %). The antigens of P-ANCA were not found to be associated with reactivity to lactoferrin (L:f), !3-glucuronidase (j3- Glc), myeloperoxidase (M:PO) or proteinase 3 (PR3).ANCAs were analysed by IIF on unfixed neutrophils or cells fixed by either ethanol, acetone or parafonnaldehyde in 21 sera from patients with UC, 17 sera from patients with vasculitides including 8 with PR3-positive C-ANCA and 9 with MPO-positive P-ANCA. Established ANCA patterns were most clear with ethanol-fixed neutrophils. The PR3-positive C-ANCA pattern was the same on unfixed or fixed cells irrespective of fixation method. However, the staining was brightest and the serum titres were higher with ethanol-fixed cells. Furthermore, we confirmed that the antigen of UC associated P-ANCA was better exposed on ethanol- than on acetone- or paraformaldehyde-fixed cells. In addition, anti-MPO positive sera tested on the very same day with freshly prepared paraformaldehyde-fixed neutrophils gave a C-ANCA pattern. Two or more days after preparation of the neutrophils anti-MPO-positive sera gave a mixed C- and P-ANCA pattern. Neutrophils kept unfixed for a few days or more and then treated with parafonnaldebyde gave a P-ANCA pattern, indicating diffusion of MPO from the azurophilic granules to the periphery of the nucleus, a process that is strongly enhanced by ethanol fixation.Six anti-MPO-negative P-ANCA-positive sera from patients with UC, six antiMMPO-positive P-ANCA, five anti-PR3-positive C-ANCA and ten antinuclear antibody (ANA)-positive serum samples were tested with 15 different GramHnegative and Gram-positive bacterial strains. We found that soluble material from live E. coli and P. mirabilis has the capacity to decompose the anti. genic substrate of neutrophils responsible for both MPOpositive and MPO-negative P-ANCA. most probably brought about by enzymatic activity. Anti-PR3-positive CANCA was not affected which suggests substrate specificity of the proposed enzymatic activity.