Reactive oxygen products in heterologous anti-glomerular basement membrane nephritis in rats.

The effect of ''scavengers'' of reactive oxygen products (ROPs) was studied in the heterologous phase of anti-glomerular basement (anti-GBM) nephritis induced in rats. Glomerulonephritis was induced by the intravenous administration of sheep anti-GBM antibody (5 mg/100 g) to rats on day 0. The intraperitoneal administration of superoxide dismutase (SOD) 30 mg/kg/day or 150 mg/kg/day leads to a significant reduction in proteinuria on day 1 and also on day 3 in animals given SOD 30 mg/kg/day. Proteinuria was not significantly reduced by the intraperitoneal administration of inactivated SOD (150 mg/kg/day). In rats given polyethylene glycol coupled catalase (PEG-catalase) intraperitoneally at a dose of 10,000 iu/kg/day and 100,000 iu/kg/day proteinuria was lower than in rats with unmodified anti-GBM nephritis. These differences were significant on day 1 (P less than 0.05) in rats given PEG-catalase 100,000 iu/kg/day and on days 3 and 5 in rats treated with either dose of PEG-catalase (P less than 0.01). These data suggest a role for superoxide anion and hydrogen peroxide, or a product of their interaction such as hydroxyl radical, in glomerular injury induced by anti-GBM antibody.     

Induction of autoimmunity to antigens of the glomerular basement membrane in inbred Brown-Norway rats.

Induction of autoimmune antibodies against antigens of glomerular basement membrane (GBM) was studied in nine inbred strains of rats each with a different major histocompatibility complex H-1. Brown-Norway (BN) (H-1n), Lewis (H-1(1)), PVG/c (H-1c), AS2 (H-1f), AVN (H-1a), BD V (H-1d), DA (H-1a) and F344 (H-1(1)) rats were immunized with bovine GBM and Freund's complete adjuvent (CFA). A pronounced linear deposition of host IgG (IgG1 and IgG2a) along the GBM was found in BN rats. No deposition of C3 could be detected in the glomeruli nor did the animals develop proteinuria. The quantity of autoimmune antibodies fixed to the GBM was low (48 microgram +/- 14) which could explain the absence of C3 deposition and proteinuria. The antigenic specificity of the antibodies deposited along the GBM in BN rats was shown by the fixation in vitro of the eluted antibodies to the GBM and tubular basement membrane (TBM) of normal kidneys. A much weaker and irregular deposition of host IgG along the GBM was observed in PVG/c, AS2. AVN, BD V, DA and F344 rats. Of these strains, eluates from the glomeruli of PVG/c, AVN, BD V and DA rats fixed very weakly to the GBM of normal kidneys whereas eluates from AS2 and F344 rats did not fix to GBM or TBM. No deposition of host IgG was found in Lewis rats, and the eluates did not fix to normal kidneys. Congenic L.BN rats with the BN H-1n haplotype and a Lewis background did not respond. This study shows a genetic predisposition in rats to an autoimmune anti-GBM response which is not, or not exclusively, controlled by genes linked to the H-1 histo-compatibility complex.     

Crescentic glomerulonephritis induced in the goat by immunization with homologous or heterologous glomerular basement membrane antigen.

Crescentic glomerular changes developed in goats 16 to 46 weeks after immunization with homologous or heterologous glomerular basement membrane (GBM) antigens incorporated in Freund's complete adjuvant. Mesangial widening, accumulation of migrant cells, fibrin deposition, and GBM thickening became manifest with time. The GBM was then attenuated and ruptured at a local area. Through the ruptured GBM, the influx of fibrin and blood cells occurred in the Bowman's space (BS) accompanying proliferation of epithelial cells lining the BS. The crescent thus formed was mainly composed of the capsular epithelial cells. The visceral epithelial cells (podocytes) and migrated blood cells contributed to the crescent formation to a lesser extent. Basement menbrane-like material appeared later between the crescentic cells and the crescent underwent a fibroepithelial form. The results indicate that the crescent develops in relation to the characteristic GBM damage with a resultant severe exudative change in the BS.     

Reactive oxygen products in heterologous anti-glomerular basement membrane nephritis in rats

The effect of 'scavengers' of reactive oxygen products (ROPs) was studied in the heterologous phase of anti-glomerular basement (anti-GBM) nephritis induced in rats. Glomerulonephritis was induced by the intravenous administration of sheep anti-GBM antibody (5 mg/100 g) to rats on day 0. The intraperitoneal administration of superoxide dismutase (SOD) 30 mg/kg/day or 150 mg/kg/day leads to a significant reduction in proteinuria on day 1 and also on day 3 in animals given SOD 30 mg/kg/day. Proteinuria was not significantly reduced by the intraperitoneal administration of inactivated SOD (150 mg/kg/day). In rats given polyethylene glycol coupled catalase (PEG-catalase) intraperitoneally at a dose of 10,000 iu/kg/day and 100,000 iu/kg/day proteinuria was lower than in rats with unmodified anti-GBM nephritis. These differences were significant on day 1 (P less than 0.05) in rats given PEG-catalase 100,000 iu/kg/day and on days 3 and 5 in rats treated with either dose of PEG-catalase (P less than 0.01). These data suggest a role for superoxide anion and hydrogen peroxide, or a product of their interaction such as hydroxyl radical, in glomerular injury induced by anti-GBM antibody.     

Immune complex glomerulonephritis is induced in rats immunized with heterologous myeloperoxidase.

Anti-neutrophil cytoplasmic antibodies (ANCA), including anti-myeloperoxidase (MPO) antibodies, are associated with pauci-immune necrotizing small vessel vasculitis or glomerulonephritis. In order to substantiate a pathogenic role for ANCA, an animal model of pauci-immune ANCA-induced glomerulonephritis or vasculitis is required. Brouwer et al. reported pauci-immune glomerulonephritis in rats immunized with human MPO followed by perfusion of kidneys with lysosomal enzyme extract combined with H2O2, and suggested that this could serve as a model of ANCA-induced disease. We repeated these studies in spontaneously hypertensive rats (SHR) and Brown Norway rats (BNR). We immunized rats with human MPO. When circulating anti-MPO antibodies were detectable by indirect immunofluorescence microscopy and ELISA, blood pressure was measured, then perfusion of the left kidney of each rat was done via the renal artery in a closed, blood-free circuit with either MPO + H2O2, MPO, H2O2 alone or MPO + H2O2 + neutral protease. Rats were killed on day 4 or day 10 after perfusion, and specimens were examined by light and immunofluorescence microscopy. Pathological lesions and deposits of IgG, C3, and MPO were found in immunized rats perfused with MPO + H2O2 with or without neutral protease, or MPO alone, in both rat strains and on both day 4 and day 10. The degree of histologic injury was proportional in intensity to the amount of IgG immune deposits. Spontaneously hypertensive rats sustained more damage and higher blood pressure than Brown Norway rats. No lesion was observed in immunized rats perfused with H2O2 or in the non-perfused right kidneys. Some of the non-immunized rats perfused with MPO + H2O2 developed pathological lesions. In conclusion, these rat models are examples of immune complex-mediated glomerulonephritis, and therefore are not similar to human ANCA-associated disease.     

Cellular reactivity to altered glomerular basement membrane in glomerulonephritis.

Glomerular basement membrane may be altered during glomerulonephritis, exposing antigens that are recognized as foreign. Immunochemical studies suggest that removal of peripheral glycopeptides from the basement membrane with glycosidase mimics this pathogenetic event. To examine these hypotheses, we studied 24 patients with biopsy-proved glomerulonephritis by means of the lymphocyte-blast-transformation assay. Three preparations of normal glomerular basement membrane were used: two mimicked the native state for the peripheral glycopeptides, and one was altered by glycosidases. Results showed minimal differences in responses to native glomerular basement-membrane preparations among patients with glomerulonephritis and control groups. However, patients with glomerulonephritis had a significant blastogenic response to the glycosidase-treated glomerular basement membrane as compared to patients with nonglomerular renal disease and normal controls (P less than 0.0005). These studies suggest that cellular reactivity to altered glomerular basement-membrane antigens can be detected in certain forms of progressive glomerulonephritis.     

Immunological response to glomerular basement membrane and streptococcal antigensin immunized rabbits.

Cellular and humoral responses were ivestigated following human glomerular basement membrane a-d streptococcol Type 12 membrane injections in rabbits. Using the leucocyte migration inhibition test and double diffusion in agar gel, cross-reactivity between the antigens was evident, though neither antigen induced significant proteinuria or nephritis in the dosages given.     

Tubulointerstitial nephritis induced by monoclonal anti-proximal tubular basement membrane antibodies in mice

Tubulointerstitial nephritis (TIN) was induced by monoclonal anti-tubular basement membrane (TBM) antibodies. Hybridomas producing anti-TBM antibodies were produced by the fusion of a mouse parental cell line with BALB/c mice which had been immunized with Wistar rat renal cortices. Three hybridoma cell lines were selected for production of antibodies against proximal TBM by indirect immunofluorescence. The isotypes of these monoclonal antibodies (MoAbs) were determined to be of the IgM class by double immunodiffusion. Subsequently, 6-week-old female BALB/c mice were injected intraperitoneally with cells (1 x 10(7)) of anti-TBM antibody-producing hybridomas, 39-1, 39-4, or 339-3. As a control, a monoclonal IgM-producing myeloma cell line was used. Proteinuria developed from Day 8, reaching 200 to 300 mg% in mice from the experimental groups, while in the control mice, urinary protein did not exceed 50 mg%. By sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis, the excreted urinary proteins proved to be of low molecular weight. An immunofluorescent study revealed a linear localization of IgM along the proximal TBM from Day 4, and light microscopy showed focal degenerative alteration in proximal tubules and focal round cell infiltration in the interstitium. Electron microscopy revealed dense deposits on some proximal TBM. These results indicate that monoclonal anti-TBM IgM antibodies can induce TIN as a result of persistent production of antibodies from intraperitoneally injected hybridomas.     

Cyclosporin A and anti-glomerular basement membrane antibody glomerulonephritis in rats.

In a telescoped model of antiglomerular basement membrane (GBM) antibody induced nephritis, Lewis strain rats were injected in the footpad with rabbit IgG on day 0 and then given a single intravenous injection of rabbit anti-rat GBM antibody on day 5. Proteinuria developed within 24 h and renal histology 7 days later showed a focal or diffuse proliferative glomerulonephritis. In this study rats treated as above were given Cyclosporin A (CyA) 20 mg/kg daily by intraperitoneal injection from day 0 or from day 5. Rats given CyA plus anti-GBM antibody developed extensive glomerular infiltration with polymorphs and glomerular thrombosis, lesions not seen with unmodified anti-GBM nephritis or in rats who received CyA alone. The mechanism by which CyA given prior to or at the onset of immunological insult in this model worsens glomerular injury is unclear.     

Hereditary nephritis: immunoblotting studies of the glomerular basement membrane

Hereditary nephritis (HN) is associated with antigenically abnormal glomerular basement membrane (GBM) manifest by reduced or absent binding of MCA-P1, a mouse monoclonal antibody which recognizes Goodpasture antigen. In the present studies, immunoblotting has been used to analyze antigenic and biochemical composition of renal tissue from patients with HN in whom binding of MCA-P1 could not be demonstrated by indirect immunofluorescence (IIF). Pooled normal collagenase-solubilized GBM (CS-GBM) and CS-GBM from three patients with either end-stage renal failure (ESK1-3) or HN (HNK1-3), were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel and, after transfer by Western blotting to nitrocellulose, reacted with sera from six patients with anti-GBM disease (GPS1-6) or anti-GBM antibody containing sera from three transplanted HN patients (HNS1-3), different from those patients with HN contributing HNK1-3. We found that GPS1-6 and HNS1-2 recognized the same six major bands in CS-GBM and ESK1-3 (between 54 and 24 kilodalton (kd) but only three bands (48, 42 and 24 kd) in HNK1-3. HNS3 only bound strongly to bands of 54 and 26 kd in CS-GBM or ESK1-3 and not all to HNK1-3. Immunoblotting studies of HNK1-3 have shown a partial rather than absolute loss of Goodpasture antigenicity (54, 28 and 26 kd bands). Studies with HNS1-3 suggest heterogeneity of antibody responses to allografted kidneys between patients with HN; HNS-3 showed restricted antibody specificity with recognition in CS-GBM of some bands antigenically absent from HN kidney. The abnormality in HN kidneys appears closely related to, but distinct from, the Goodpasture determinant and the inherited defect in HN may involve an essential modifying enzyme.     

Mechanisms of T cell-induced glomerular injury in anti-glomeruler basement membrane (GBM) glomerulonephritis in rats

The effector mechanisms of T cell-dependent acute glomerular injury were studied in autologous phase anti-GBM glomerulonephritis (GN) in rats. Acute proliferative GN was induced in sensitized rats by a subnephritogenic dose of sheep anti-rat GBM antibody. Injury was manifested by proteinuria and glomerular leucocyte infiltration composed predominantly of macrophages but also CD4⁺ and CD8⁺ T cells. T cell depletion, using an anti-CD5 MoAb, demonstrated that glomerular leucocyte infiltration and proteinuria were T cell-dependent. Inhibition of T helper cell function using an anti-CD4 MoAb prevented proteinuria and glomerular macrophage and CD4⁺ T cell influx, but not accumulation of CD8⁺ T cells. Depletion of CD8⁺ T cells also prevented proteinuria and the influx of macrophages and CD8⁺ T cells, but not accumulation of CD4⁺ T cells. Macrophage depletion, using micro-encapsulated clodronate, prevented proteinuria and glomerular macrophage infiltration, but not the accumulation of CD4⁺ or CD8⁺ T cells, indicating that macrophages are the common cellular effectors for both CD4 and CD8 T cell-dependent injury. Evidence for cytotoxic mechanisms of injury (increased numbers of apoptotic cells or accumulation of natural killer (NK) cells in glomeruli) could not be demonstrated. These data suggest that acute glomerular injury in anti-GBM GN is the result of macrophage recruitment, which is dependent on both CD4 and CD8 T cells, and that direct T cell-mediated injury (cellular cytotoxicity) is not involved.     

Cyclosporin A and anti-glomerular basement membrane antibody glomerulonephritis in rats

In a telescoped model of antiglomerular basement membrane (GBM) antibody induced nephritis, Lewis strain rats were injected in the footpad with rabbit IgG on day 0 and then given a single intravenous injection of rabbit anti-rat GBM antibody on day 5. Proteinuria developed within 24 h and renal histology 7 days later showed a focal or diffuse proliferative glomerulonephritis. In this study rats treated as above were given Cyclosporin A (CyA) 20 mg/kg daily by intraperitoneal injection from day 0 or from day 5. Rats given CyA plus anti-GBM antibody developed extensive glomerular infiltration with polymorphs and glomerular thrombosis, lesions not seen with unmodified anti-GBM nephritis or in rats who received CyA alone. The mechanism by which CyA given prior to or at the onset of immunological insult in this model worsens glomerular injury is unclear.     

Decrease of heparan sulfate staining in the glomerular basement membrane in murine lupus nephritis.

Recently we found in biopsies of human lupus nephritis a nearly complete loss of heparan sulfate (HS) staining in the glomerular basement membrane (GMB). To clarify the relationship between HS staining and albuminuria in lupus nephritis, we studied MRL/lpr mice with short (< 7 days) or prolonged duration of albuminuria (14-21 days) and compared these with mice of different ages without albuminuria. Kidney sections were stained for mouse immunoglobulin (Ig), HS, heparan sulfate proteoglycan (HSPG)-core protein and laminin in immunofluorescence. In mice with prolonged albuminuria HS staining in the glomerular capillary loops had almost completely disappeared, whereas staining was unaltered in non-albuminuric mice (P = 0.001). In mice with short duration of albuminuria, there was a tendency toward a decrease of HS staining (P = 0.06). The expression of HSPG-core protein and other extra cellular matrix (ECM) components was unaltered in all groups. HS staining correlated inversely with albuminuria (rs = -0.55; P < 0.001) and with staining of Ig deposits in the capillary loops (rs = -0.74; P < 0.001). Despite the nearly complete loss of HS staining in the GBM in mice with prolonged albuminuria, there was no change in glomerular HS content as assessed by agarose electrophoresis and HS inhibition ELISA. We conclude that the development of albuminuria in MRL/lpr mice is accompanied by a loss of HS staining in the GBM, probably due to the masking of HS by deposits of Ig. In vitro studies revealed that autoantibodies complexed to nucleosomal antigens can inhibit the binding of the anti-HS monoclonal antibody to HS. Whether this also occurs in vivo remains to be determined.     

Loss and rearrangement of glomerular basement membrane laminin during acute nephrotoxic nephritis in the rat.

Many earlier studies have shown that the intravenous injection into rats of sheep antibodies against rat glomerular basement membrane (GBM) induces a rapid influx of neutrophils and proteinuria (nephrotoxic nephritis or NTN). The GBM antigens recognized by nephrotoxic antibodies (NTAbs) have not been identified conclusively. Our experiments presented here, however, showed that NTAbs did not significantly reduce binding of anti-laminin IgGs to laminin-coated enzyme-linked immunosorbent assay (ELISA) plates or to the GBM in vivo, indicating little cross-reactivity between the NTAbs and laminin. To evaluate possible changes in GBM architecture during acute stages of NTN, the ultrastructural distribution of laminin was determined by postfixation, postembedding immunogold labeling, and compared between normal and nephritic rats. The density of immunoreactive GBM laminin was significantly reduced in rats with acute NTN. In addition, conjugates of anti-laminin IgG and horseradish peroxidase were intravenously injected into rats that then received injections of NTAbs. Anti-laminin peroxidase conjugates were also injected after administering NTAbs. In both cases, an overall decrease in anti-laminin peroxidase reaction product was observed as compared to normal controls. The densest labeling was seen in the lamina rara interna, especially in areas of endothelial cell detachment. Some immunoperoxidase reaction product was also bound to basal surfaces of detaching endothelial cells, demonstrating the removal of at least some laminin from the GBM. A decrease in GBM binding of intravenously injected anti-laminin IgG, both before and after injection of rats with NTAbs, was also confirmed by postembedding immunogold labeling. Furthermore, morphometry showed that the GBM was significantly wider in nephritic rats than in controls, indicating a redistribution of laminin over a greatly increased area. These immunoultrastructural findings show, therefore, that GBM architecture is altered in the early phase of NTN.     

Ultrastructural changes of glomerular basement membrane in IgA nephritis: relationship to hematuria

Ultrastructural changes in the glomerular basement membrane (GBM) of 415 samples from 344 patients with IgA nephritis, were examined for potential relationship to hematuria. The GBM showed various alteration: splitting, thinning, membranolysis with swelling of the lamina rara externa and interna, forming of small projections, and rupture. These lesions were present in 48% of IgA nephritis and in 16% of the controls. In the IgA nephritis group, the patients with hemispherical mesangial dense deposits had the highest rate (60%) of capillary wall abnormalities. Such lesions were more frequent in patients biopsied during severe hematuria (80%) than in those biopsied without hematuria (33%), (p less than 0.01). It is assumed that the ultrastructural abnormalities of GBM may contribute to the clinical evidence of severe hematuria.     

Experimental autoimmune glomerulonephritis induced by anti-glomerular basement membrane antibody. II. Effects of injecting heterologous, homologous, or autologous glomerular basement membranes and complete Freund's adjuvant into sheep.

The effects of injecting human, rabbit, rat, or single-kidney homologous glomerular basement membrane (GBM) or autologous GBM, each in complete Freund's adjuvant (CFA), into 15- to 18-month-old sheep are compared. All sheep receiving heterologous GBM and 3 of 6 sheep receiving homologous GBM had anti-GBM nephritis, but such sheep did not bind autoantibodies or have Goodpasturelike lesions in their lungs. Sheep given injections of human GBM had autoantibodies to antigenic determinants shared by fetal or adult sheep and human GBM, by lung basement membranes, and by certain nonvascular basement membranes. Sheep given homologous GBM had two populations of autoantibodies: one was neither species- nor organ-specific; the other was sheep-specific. No sheep given autologous GBM had any evidence of anti-GBM autoantibodies or nephritis. Their kidneys were indistinguishable by histologic, immunohistologic, and functional studies from CFA-treated controls. Thus, sheep seem very tolerant to autologous GBM. These findings suggest that human anti-GBM nephritis may occur if the GBM is altered so that it becomes cross-reacting and induces autoantibodies, as does homologous GBM.     

Scanning electron microscopy of the acellular glomerular and tubular basement membrane in lupus nephritis

After using a cellular digestion technic to extract cells from the basement membranes of frozen kidney tissue, we used scanning electron microscopy to examine the acellular glomerular basement membranes (AGBM) and acellular tubular basement membranes (ATBM) from normal kidneys and from the kidneys of patients with lupus nephritis. This method revealed, in the AGBM, previously unrecognized three-dimensional patterns of pathologic changes. These patterns correlated with the World Health Organization (WHO) subclass of lupus nephritis and with the quantity of immune-complex deposition seen with two-dimensional microscopy. These pathologic changes included epimembranous, crater-like deformities with and without material resembling immune complexes; severely distorted, "moth-eaten" glomerular basement membrane; and the formation of secondary basement membrane within glomerular capillaries. We did not see similar abnormalities in the ATBM.     

Amyloid P component is not present in the glomerular basement membrane in Alport-type hereditary nephritis.

Serum amyloid P component (SAP) is a normal plasma glycoprotein immunochemically indistinguishable from amyloid P, a glycoprotein found in all tissue amyloid deposits. SAP has also been shown to be a constituent of normal glomerular basement membrane (GBM). In this study the authors discovered a unique association between SAP and Goodpasture (GP) antigen. In those patients whose GBM lack GP antigen (those with Alport-type hereditary nephritis) SAP is also uniformly absent. Although the relationship between these two components is unknown, this association may provide clues to the abnormality of GBM in Alport-type hereditary nephritis.