Autoantibodies to myeloperoxidase aggravate mild anti-glomerular-basement-membrane-mediated glomerular injury in the rat

Autoantibodies to myeloperoxidase (MPO) are present in sera from patients with various forms of vasculitis-associated glomerulonephritis. Evidence for a pathogenic role of anti-MPO antibodies has been provided mainly by in vitro studies. We studied the pathogenic role of autoantibodies to MPO in a rat model of mild immune-mediated glomerular injury. Brown Norway rats were immunized with human MPO in complete Freund's adjuvant or with complete Freund's adjuvant alone. At 2 weeks after immunization, rats had developed antibodies to human and rat MPO as detected by indirect immunofluorescence, enzyme-linked immunosorbent assay, and immunoprecipitation. At this time point, rats were intravenously injected with a subnephritogenic dose of 150 micrograms of rabbit anti-rat GBM. Rats were sacrificed at 4 hours, 24 hours, 4 days, and 10 days after antibody administration. Control immunized rats developed mild glomerulonephritis characterized by slight proteinuria at day 10 (14.8 +/- 8.1 mg/24 hours) and moderate intraglomerular accumulation of ED1+ macrophages. Crescent formation, tuft necrosis, and tubular atrophy were not observed in those rats. In contrast, rats immunized with MPO developed severe glomerulonephritis characterized by the early occurrence of severe hematuria, marked proteinuria at day 10 (76.2 +/- 18.2 mg/24 hours), and massive glomerular deposition of fibrin. Complement and rat IgG were present in insudative lesions, but no linear pattern along the glomerular capillary wall was observed. By light microscopy, severe glomerular lesions were found at day 10 consisting of crescent formation and fibrinoid necrosis of capillary loops. In the interstitium, tubular necrosis and atrophy and marked interstitial mononuclear infiltration were found in conclusion, autoantibodies to MPO severely aggravate subclinical anti-GBM disease demonstrating their in vivo pathogenic potential.

Elastase, but not proteinase 3 (PR3), induces proteinuria associated with loss of glomerular basement membrane heparan sulphate after in vivo renal perfusion in rats

Elastase, but not PR3, induces proteinuria associated with loss of glomerular basement membrane (GBM) heparan sulphate after in vivo renal perfusion in rats. PR3 and elastase are cationic neutral serine proteinases present in the azurophilic granules of polymorphonuclear leucocytes. Release of these proteolytic enzymes along the glomerular capillary wall may induce glomerular injury. Here, we investigated the effects of PR3 and elastase on the induction of proteinuria and glomerular injury after renal perfusion of these enzymes in Brown-Norway rats. Perfusion of active elastase induced a dose-dependent proteinuria 24h after perfusion, while inactivated elastase did not. Perfusion of comparable amounts of active PR3 did not induce proteinuria. Light and electron microscopy showed no morphological abnormalities in any experimental group. However, immunohistology revealed that proteinuria occurring after perfusion of active elastase was associated with a strong reduction in intraglomerular expression of the heparan sulphate side chain and, to a lesser extent, of the protein core of heparan sulphate proteoglycans (HSPG). In vitro, both elastase and PR3 digested HSPG. However, PR3 bound to a lesser extent to HSPG than elastase. We conclude that elastase, but not PR3, induces proteinuria after in vivo renal perfusion. This differential effect probably relates to different binding to the GBM of those enzymes due to differences in their isoelectric points. Degradation of heparan sulfate proteoglycans, leading to the disappearance of their side chains that contribute to the polyanionic structure of the GBM, appears to be involved in the induction of proteinuria after perfusion of elastase.

Activation of granulocytes by anti-neutrophil cytoplasmic antibodies (ANCA): a Fc gamma RII-dependent process

ANCA have been demonstrated to induce the respiratory burst in primed neutrophils. In this study we have extended the investigations on neutrophil activation by ANCA directed against proteinase 3 (PR3), myeloperoxidase (MPO) and lactoferrin (LF), and we have analysed the underlying mechanisms. All three ANCA antigens were expressed on the cell surface of primed neutrophils. Superoxide production assayed by both cytochrome c reduction and oxidation of dihydrorhodamine 123, was induced by heterologous polyclonal anti-MPO and anti-LF antibodies, and ANCA-positive plasma samples. Induction of superoxide production was dose-dependent. F(ab')2 fragments did not induce the respiratory burst. Blockade of Fc receptors by specific MoAbs showed that anti-Fc gamma RII antibodies were able to turn off the ANCA-induced respiratory burst, whereas anti-Fc gamma RIII antibodies did not. Plasma samples that induced the respiratory burst did not differ from samples that did not induce superoxide production with respect to ANCA titre, but had higher levels of the IgG3 subclass of ANCA. Levels of the other subclasses of ANCA were comparable between those samples. We conclude that ANCA-induced activation of primed neutrophils is Fc gamma RII-dependent, and appears to be facilitated by antibodies of the IgG3 subclass.

Podocyte expression of MHC class I and II and intercellular adhesion molecule-1 (ICAM-1) in experimental pauci-immune crescentic glomerulonephritis

We examined immunopathological changes of podocytes in vivo which, based on in vitro studies, are thought to be relevant for the pathogenesis of renal diseases. We investigated the alterations of podocytes in local inflammation in a recently developed model of pauci-immune necrotizing crescentic glomerulonephritis (NCGN) in the rat. Frozen and plastic embedded kidney sections at different time points of the disease were incubated with antibodies directed to MHC class I, MHC class II, ICAM-1 and to relevant cytokines. Strong glomerular expression of MHC class I, II and ICAM-1 was found within 4 days, and plastic embedded sections clearly demonstrated increased cell membrane staining of podocytes. Increased glomerular interferon-gamma (IFN-gamma) was detected within 24 h of induction of NCGN, and IL-1 beta and tumour necrosis factor-alpha (TNF-alpha) were found from day 4. The potency of these cytokines to induce adhesion molecules on podocytes was investigated on rat glomerular epithelial cells in vitro. By using FACS analysis and electron microscopical techniques, we found that the in vivo expression of MHC class I, II and ICAM-1 by podocytes could in vitro be simulated by IFN-gamma. IFN-alpha weakly induced MHC class I, while IL-1 beta and TNF-alpha were ineffective. We hypothesize that podocytes in this in vivo model are important to maintain the local inflammatory process in the glomerulus by expression of relevant adhesion molecules and MHC molecules upon stimulation with specific cytokines.

Neutrophil activation in vitro and in vivo in Wegener's granulomatosis

The mechanisms underlying glomerular capillary wall injury in Wegener's granulomatosis (WG) are not well understood. Anti-neutrophil cytoplasmic antibodies (ANCA), present in sera from patients with WG, are known to stimulate respiratory burst and degranulation of primed polymorphonuclear neutrophils (PMN) in vitro. Experimental studies have shown that oxygen radical production and lysosomal enzymes are important mediators of glomerular capillary wall injury. In the present study we investigated the presence of activated PMN and the extracellular localization of lysosomal enzymes in 28 consecutive renal biopsies from patients with WG. The presence of activated PMN within the renal biopsies was compared with the capacity of ANCA, isolated from simultaneously drawn serum samples, to activate primed PMN obtained from a normal donor. Both parameters were also related to renal function. Renal biopsies were obtained from newly diagnosed WG patients before therapy had started. Activation of PMN in the biopsies was assessed by measuring hydrogen peroxide production in situ. The number of activated PMN in the biopsy correlated with the extent of impairment of renal function. Proteinase 3, myeloperoxidase, and elastase, all targets of ANCA, were localized extracellularly in renal tissue and were also found within tubular epithelial cells. All ANCA positive samples were capable of activating primed PMN. The amount of activation correlated with the ANCA titer in those samples. No correlation, however, was found between the in vitro capacity of ANCA-positive IgG fractions to activate primed PMN and the number of activated PMN present in the renal biopsy. We conclude that activated PMN producing toxic oxygen metabolites and releasing lysosomal enzymes, are present in renal biopsies from patients with WG. The amount of activated PMN present within the kidney, and not the capacity of the corresponding ANCA to activate PMN, correlates with renal tissue damage as assessed by serum creatinine levels.