Identification of a pathogenic epitope involved in initiation of Heymann nephritis

Advances of the experimental models of idiopathic membranous nephropathy (Review)

Idiopathic membranous nephropathy (IMN) is one of the main types of chronic kidney disease in adults and one of the most common causes of end-stage renal disease. In recent years, the morbidity of IMN among primary glomerular diseases has markedly increased, while the pathogenesis of the disease remains unclear. To address this, a number of experimental models, including Heymann nephritis, anti-thrombospondin type-1 domain-containing 7A antibody-induced IMN, cationic bovine serum albumin, anti-human podocyte antibodies and zymosan-activated serum-induced C5b-9, have been established. This review comprehensively summarized the available animal and cell models for IMN. The limitations and advantages of the current models were discussed and two improved models were introduced to facilitate the selection of an appropriate model for further studies on IMN.

Immunology of membranous nephropathy: from animal models to humans

Membranous nephropathy (MN), the leading cause of nephrotic syndrome in adults, is characterized by the deposition of subepithelial immune deposits that consist mainly of immunoglobulin (Ig)G and complement. Most of the cases are primary or idiopathic (iMN), while only approximately 25% of the cases are secondary to some known disease such as systemic lupus erythematosus, hepatitis B, drugs and malignancies. Most of our knowledge on the pathogenesis of iMN has relied upon old experimental models (i.e. Heymann nephritis) that have shown that immune deposits are formed in situ by the reaction of autoantibodies against the respective podocyte antigen. Recent findings indicate that podocyte proteins also act as an autoantigen in human iMN. The M-type phospholipase A2 receptor (PLA2R) has been identified as the main target antigen, as it can be found in approximately 70% of iMN patients but only rarely in other glomerulonephritides. Podocytes damage in the experimental model of Heymann nephritis is complement-mediated. In humans, the presence of complement within the subepithelial deposits is well established, but IgG4, which does not activate complement by classical or alternative pathways, represents the predominant subclass of IgG anti-PLA2R. Some evidence suggests that IgG4 anti-PLA2R autoantibodies can bind mannan-binding lectin (MBL) and activate the lectin complement pathway. A genetic background for iMN has been demonstrated by genome-wide association studies that have shown highly significant associations of the PLA2R1 and the human leucocyte antigen (HLA)-DQA1 loci with iMN. In addition to their diagnostic value, anti-PLA2R antibodies may be useful to monitor disease activity and predict response to treatment.

2DE maps in the discovery of human autoimmune kidney diseases: the case of membranous glomerulonephritis

The identification of antigens in the autoimmune diseases is a primary point to elucidate the pathogenesis of disease. Here, we propose an "in vivo" proteomics approach to identify the antigens of auto-antibodies in membranous glomerulonephritis. In this approach, podocyte proteins resolved by two-dimensional electrophoresis were semidry blotted to nitrocellulose membrane. Then the antibodies eluted from microdissected glomeruli and serum samples were used as a probe for the detection of podocyte antigens and characterized by means of mass spectrometry. These combined methods allowed us to identify six new antigens in membranous glomerulonephritis.

Treatment of idiopathic membranous nephropathy

Exciting progress recently has been made in our understanding of idiopathic membranous nephropathy, as well as treatment of this disease. Here, we review important advances regarding the pathogenesis of membranous nephropathy. We will also review the current approach to treatment and its limitations and will highlight new therapies that are currently being explored for this disease including Rituximab, mycophenolate mofetil, and adrenocorticotropic hormone, with an emphasis on results of the most recent clinical trials.

Nephrotic syndrome and subepithelial deposits in a mouse model of immune-mediated anti-podocyte glomerulonephritis

Subepithelial immune complex deposition in glomerular disease causes local inflammation and proteinuria by podocyte disruption. A rat model of membranous nephropathy, the passive Heymann nephritis, suggests that Abs against specific podocyte Ags cause subepithelial deposit formation and podocyte foot process disruption. In this study, we present a mouse model in which a polyclonal sheep anti-mouse podocyte Ab caused subepithelial immune complex formation. Mice developed a nephrotic syndrome with severe edema, proteinuria, hypoalbuminemia, and elevated cholesterol and triglycerides. Development of proteinuria was biphasic: an initial protein loss was followed by a second massive increase of protein loss beginning at approximately day 10. By histology, podocytes were swollen. Electron microscopy revealed 60-80% podocyte foot process effacement and subepithelial deposits, but no disruption of the glomerular basement membrane. Nephrin and synaptopodin staining was severely disrupted, and podocyte number was reduced in anti-podocyte serum-treated mice, indicating severe podocyte damage. Immunohistochemistry detected the injected anti-podocyte Ab exclusively along the glomerular filtration barrier. Immunoelectron microscopy localized the Ab to podocyte foot processes and the glomerular basement membrane. Similarly, immunohistochemistry localized mouse IgG to the subepithelial space. The third complement component (C3) was detected in a linear staining pattern along the glomerular basement membrane and in the mesangial hinge region. However, C3-deficient mice were not protected from podocyte damage, indicating a complement-independent mechanism. Twenty proteins were identified as possible Ags to the sheep anti-podocyte serum by mass spectrometry. Together, these data establish a reproducible model of immune-mediated podocyte injury in mice with subepithelial immune complex formation.

Experimental Models of Membranous Nephropathy

Membranous nephropathy (MN) is one of the commonest glomerular diseases, typically presenting in older males with nephrotic syndrome. The development and characterization of animal models of MN, in particular, the passive Heymann nephritis model (PHN), has greatly advanced our understanding of this disease. In this review we discuss the different animal models of human MN that are available, with an emphasis on the PHN model, including technical issues, the typical disease course and its application to human disease.

Autoimmunity in membranous nephropathy targets aldose reductase and SOD2

Glomerular targets of autoimmunity in human membranous nephropathy are poorly understood. Here, we used a combined proteomic approach to identify specific antibodies against podocyte proteins in both serum and glomeruli of patients with membranous nephropathy (MN). We detected specific anti-aldose reductase (AR) and anti-manganese superoxide dismutase (SOD2) IgG(4) in sera of patients with MN. We also eluted high titers of anti-AR and anti-SOD2 IgG(4) from microdissected glomeruli of three biopsies of MN kidneys but not from biopsies of other glomerulonephritides characterized by IgG deposition (five lupus nephritis and two membranoproliferative glomerulonephritis). We identified both antigens in MN biopsies but not in other renal pathologies or normal kidney. Confocal and immunoelectron microscopy (IEM) showed co-localization of anti-AR and anti-SOD2 with IgG(4) and C5b-9 in electron-dense podocyte immune deposits. Preliminary in vitro experiments showed an increase of SOD2 expression on podocyte plasma membrane after treatment with hydrogen peroxide. In conclusion, our data support AR and SOD2 as renal antigens of human MN and suggest that oxidative stress may drive glomerular SOD2 expression.

Induction of passive Heymann nephritis in complement component 6-deficient PVG rats

Passive Heymann nephritis (PHN), a model of human membranous nephritis, is induced in susceptible rat strains by injection of heterologous antisera to rat renal tubular Ag extract. PHN is currently considered the archetypal complement-dependent form of nephritis, with the proteinuria resulting from sublytic glomerular epithelial cell injury induced by the complement membrane attack complex (MAC) of C5b-9. This study examined whether C6 and MAC are essential to the development of proteinuria in PHN by comparing the effect of injection of anti-Fx1A antisera into PVG rats deficient in C6 (PVG/C6(-)) and normal PVG rats (PVG/c). PVG/c and PVG/C6(-) rats developed similar levels of proteinuria at 3, 7, 14, and 28 days following injection of antisera. Isolated whole glomeruli showed similar deposition of rat Ig and C3 staining in PVG/c and PVG/C6(-) rats. C9 deposition was abundant in PVG/c but was not detected in PVG/C6(-) glomeruli, indicating C5b-9/MAC had not formed in PVG/C6(-) rats. There was also no difference in the glomerular cellular infiltrate of T cells and macrophages nor the size of glomerular basement membrane deposits measured on electron micrographs. To examine whether T cells effect injury, rats were depleted of CD8+ T cells which did not affect proteinuria in the early heterologous phase but prevented the increase in proteinuria associated with the later autologous phase. These studies showed proteinuria in PHN occurs without MAC and that other mechanisms, such as immune complex size, early complement components, CD4+ and CD8+ T cells, disrupt glomerular integrity and lead to proteinuria.

Glomerular clusterin is associated with PKC-alpha/beta regulation and good outcome of membranous glomerulonephritis in humans

Mechanisms for human membranous glomerulonephritis (MGN) remain elusive. Most up-to-date concepts still rely on the rat model of Passive Heymann Nephritis that derives from an autoimmune response to glomerular megalin, with complement activation and membrane attack complex assembly. Clusterin has been reported as a megalin ligand in immunodeposits, although its role has not been clarified. We studied renal biopsies of 60 MGN patients by immunohistochemistry utilizing antibodies against clusterin, C5b-9, and phosphorylated-protien kinase C (PKC) isoforms (pPKC). In vitro experiments were performed to investigate the role of clusterin during podocyte damage by MGN serum and define clusterin binding to human podocytes, where megalin is known to be absent. Clusterin, C5b-9, and pPKC-alpha/beta showed highly variable glomerular staining, where high clusterin profiles were inversely correlated to C5b-9 and PKC-alpha/beta expression (P=0.029), and co-localized with the low-density lipoprotein receptor (LDL-R). Glomerular clusterin emerged as the single factor influencing proteinuria at multivariate analysis and was associated with a reduction of proteinuria after a follow-up of 1.5 years (-88.1%, P=0.027). Incubation of podocytes with MGN sera determined strong upregulation of pPKC-alpha/beta that was reverted by pre-incubation with clusterin, serum de-complementation, or protein-A treatment. Preliminary in vitro experiments showed podocyte binding of biotinilated clusterin, co-localization with LDL-R and specific binding inhibition with anti-LDL-R antibodies and with specific ligands. These data suggest a central role for glomerular clusterin in MGN as a modulator of inflammation that potentially influences the clinical outcome. Binding of clusterin to the LDL-R might offer an interpretative key for the pathogenesis of MGN in humans.

Linking Receptor-mediated Endocytosis and Cell Signaling

Megalin, a member of the low density lipoprotein receptor gene family, is required for efficient protein absorption in the proximal tubule. Recent studies have shown that the low density lipoprotein receptor-related protein, another member of this gene family, is proteolytically processed by gamma-secretase implying a role for low density lipoprotein receptor-related protein in a Notchlike signaling pathway. This pathway has been shown to involve: 1) metalloprotease-mediated ectodomain shedding and gamma-secretase-mediated intramembrane proteolysis of some receptors. Experiments were performed to determine whether megalin undergoes similar processing. By immunocytochemistry, immunoblotting, and a fluorogenic enzyme assay presenilin-1 (required for gamma-secretase activity) and gamma-secretase activity were found in the brush border of proximal kidney tubules where megalin is localized. Using a fluorogenic peptide containing an amyloid precursor protein gamma-secretase cleavage site and Compound E, a specific gamma-secretase inhibitor, we found high levels of gamma-secretase activity in renal brush border membrane vesicles. Immunoblotting analysis of renal microsomes and opossum kidney proximal tubule (OKP) cells using antibodies directed to the cytosolic domain of megalin showed a 35-40-kDa, membrane-associated, carboxyl-terminal fragment of megalin (MCTF). When cells were incubated with 200 nm phorbol 12-myristate 13-acetate, the appearance of the MCTF increased 2.5-fold and was blocked by metalloprotease inhibitors. When the cells were incubated with gamma-secretase inhibitor Compound E, it caused a 2-fold increase in MCTF. Finally, incubating the cells with 1 microm vitamin D-binding protein resulted in a 25% increase in the appearance of the MCTF. In summary, the MCTF is produced by protein kinase C regulated, metalloprotease-mediated ectodomain shedding and is the substrate for gamma-secretase. We postulate that the enzymatic processing of megalin represents part of a novel ligand-dependent signaling pathway in the proximal tubule that links receptor-mediated endocytosis with cell signaling.

Conformation and glycosylation of a megalin fragment correlate with nephritogenicity in Heymann nephritis

Active Heymann nephritis (AHN), a rat model of autoimmune glomerulonephritis, is induced by immunization with autologous megalin, a 600-kDa cell surface glycoprotein isolated from crude renal extracts. Recombinant proteins containing a 563-residue N-terminal sequence of megalin were obtained from Escherichia coli and baculovirus-insect cell expression systems. Rats immunized with the soluble, secreted protein encoded by a baculovirus construct elicited high titer anti-megalin autoantibodies and developed glomerular immune deposits and elevated proteinuria consistent with AHN. Rats treated with the bacterial or nonsecreted insect cell proteins produced a milder anti-megalin response and did not develop the disease. Nephritogenicity appeared to correlate with conformational or other structural features of native megalin. All three recombinant proteins were reactive in Western blots with rabbit anti-megalin antiserum, whereas the insect cell-derived proteins reacted preferentially in Western blot and ELISA with anti-megalin autoantibodies from rats with AHN induced by native megalin. Only the secreted insect cell product was stained in a lectin blot, suggesting its specific glycosylation. These observations provide evidence that a megalin N-terminal domain includes B and T cell epitopes sufficient for a pathogenic autoimmune response and that a native-like conformation and glycosylation are essential for the induction of disease. The importance of conformational B cell epitopes for pathogenic autoantibodies recapitulates observations made in other models of organ-specific autoimmune disease. Glycosidic modifications could influence the presentation of either B or T cell epitopes in AHN, consistent with emerging evidence of the role of post-translational modifications in pathogenic autoimmune responses.

T cell lines specific for a synthetic Heymann nephritis peptide derived from the receptor-associated protein

Pathogenic antigens involved in the induction of Heymann nephritis (HN), an experimental rat model of human membranous nephritis, have been identified in megalin (gp330) and the receptor-associated protein (RAP) [1,2]. A pathogenic epitope has been identified in RAP (amino acid 1-86) that plays a significant role in the formation of immune deposits in glomeruli in HN. A synthetic peptide (P31-53) derived from RAP1-86 contains a pathogenic epitope recognized by antibodies eluted from glomerular immune deposits and includes two putative RT-1B1 MHC class II-binding motifs. We have investigated whether RAP P31-53 can be recognized by T cells. Five peptide-specific T cell lines were generated from regional lymph node (LN) T cells from Lewis rats immunized with P31-53. The T cell lines were characterized by using a T cell proliferation assay for their specificity, FACS and MHC restriction assay for the phenotype, reverse transcription-polymerase chain reaction for TCR Vbeta repertoire and cytokine expression, and cloning and sequencing for the analysis of the CDR3 sequence of TCR. The helper function of the T cell line was confirmed by autoantibody production in vitro. In this study, we clearly identify that the synthetic pathogenic peptide P31-53 contains a T cell epitope recognized by CD4+ Th2 cells in Lewis rats. This recognition was restricted by MHC class II RT1.B1. These CD4+ Th2 cells were able to promote B cells to produce specific antibodies and used a restricted set of TCR Vbeta genes with preferential usage of Vbeta18. A charged amino acid motif at the CDR3 region of predominant TCR Vbeta subfamilies may contribute to the specific ability of these cells to recognize the immunogenic T cell epitope within RAP peptide P31-53.

Identification of megalin/gp330 as a receptor for lipoprotein(a) in vitro

Lipoprotein(a) [Lp(a)] is an atherogenic lipoprotein of unknown physiological function. The mechanism of Lp(a) atherogenicity as well as its catabolic pathways are only incompletely understood at present. In this report, we show that the low density lipoprotein receptor (LDLR) gene family member megalin/glycoprotein (gp) 330 is capable of binding and mediating the cellular uptake and degradation of Lp(a) in vitro. A mouse embryonic yolk sac cell line with native expression of megalin/gp330 but genetically deficient in LDLR-related protein (LRP) and a control cell line carrying a double knockout for both LRP and megalin/gp330 were compared with regard to their ability to bind, internalize, and degrade dioctadecyltetramethylindocarbocyanine perchlorate (DiI)-fluorescence-labeled Lp(a) as well as equimolar amounts of 125I-labeled Lp(a) and LDL. Uptake and degradation of radiolabeled Lp(a) by the megalin/gp330-expressing cells were, on average, 2-fold higher than that of control cells. This difference could be completely abolished by addition of the receptor-associated protein, an inhibitor of ligand binding to megalin/gp330. Mutual suppression of the uptake of 125I-Lp(a) and of 125I-LDL by both unlabeled Lp(a) and LDL suggested that Lp(a) uptake is mediated at least partially by apolipoprotein B100. Binding and uptake of DiI-Lp(a) resulted in strong signals on megalin/gp330-expressing cells versus background only on control cells. In addition, we show that purified megalin/gp330, immobilized on a sensor chip, directly binds Lp(a) in a Ca2+-dependent manner with an affinity similar to that for LDL. We conclude that megalin/gp330 binds Lp(a) in vitro and is capable of mediating its cellular uptake and degradation.

Immortalized rat proximal tubule cells produce membrane bound and soluble megalin

Megalin (gp330), a glycoprotein receptor found on renal proximal tubule cells and several other epithelial cells, is deduced to be a type I integral membrane protein, but may also exist as a cell surface form lacking a cytoplasmic domain. Furthermore, soluble megalin products have been detected in urine, and in culture medium of a rat yolk sac carcinoma cell line, combined with receptor associated protein (RAP). Permanent renal cell lines expressing megalin were unavailable until the recent description of two immortalized rat proximal tubule cell lines (IRPTC). The present study demonstrated megalin on IRPTC surface by immunofluorescence, without surface staining for RAP, which was, however, readily detected within cells. Antibodies to ectodomain megalin epitopes immunoprecipitated megalin products both from cell lysates and culture medium, whereas antibodies to cytoplasmic domain epitopes precipitated megalin only from lysates. Western blots showed two major megalin products in medium, a prominent band at approximately 200 kDa, and a fainter band above 400 kDa, slightly below intact megalin in cell lysates. Anti-receptor associated protein antibodies immunoprecipitated megalin from IRPTC lysates, but not from media. We propose that portions of megalin are spontaneously produced by IRPTC, probably either by cleavage in the ectodomain or release of forms lacking a cytoplasmic domain.

Receptor-associated protein and members of the low density lipoprotein receptor family share a common epitope. An extended model for the development of passive Heymann nephritis

Heymann nephritis is an experimental rat model for human membranous glomerulonephritis. Two target antigens have been identified in the proximal tubule brush border of rat kidneys. One of them is megalin, a 600-kDa membrane protein that belongs to the family of low density lipoprotein receptor (LDLR)-related proteins. The other one is receptor-associated protein (RAP), a polypeptide of 40 kDa that associates with members of the LDLR family. Here we show that antibodies produced against recombinant human RAP strongly cross-react with the chicken oocyte receptor for very low density lipoprotein and vitellogenin (LR8), and with two other members of the LDLR family, LDLR-related protein and megalin. The interaction of this antibody with LR8 showed binding characteristics exactly as those demonstrated for the physiological ligands of this receptor, in that binding of the antibody: (i) is Ca2+-dependent; (ii) is abolished by unfolding of the cysteine-rich binding domain by reduction; and (iii) interferes with the binding of very low density lipoprotein and vitellogenin. Immunopurification of the LR8-specific subpopulation of the polyclonal antiserum yielded an IgG fraction strongly reacting with LR8 as well as with RAP. Using recombinant fragments of RAP and peptide mapping, the cross-reacting epitope(s) could be narrowed down to three short sequences (5-7 residues) in the COOH-terminal part of the protein. After immunization with RAP, anti-LR8 antibodies and anti-RAP antibodies arise simultaneously, indicating that the receptor-specific activity is not due to anti-idiotypic antibodies. These findings suggest the existence of a common epitope(s) on RAP and members of the LDL receptor family. Based on these results, we present an extended molecular model for the development of passive Heymann nephritis.

Mapping rat megalin: the second cluster of ligand binding repeats contains a 46-amino acid pathogenic epitope involved in the formation of immune deposits in Heymann nephritis

Megalin (gp330), an epithelial endocytic receptor, is a major target antigen of Heymann nephritis (HN), an autoimmune disease in rats. To elucidate the mechanisms of HN, we have mapped a pathogenic epitope in megalin that binds anti-megalin antibodies. We focused our attention on four clusters of cysteine-rich, low density lipoprotein receptor (LDLR) ligand binding repeats in the extracellular domain of megalin because they represent putative ligand binding regions and therefore would be expected to be exposed in vivo and to be able to bind circulating antibodies. Rat megalin cDNA fragments I through IV encoding the first through fourth clusters of ligand-binding repeats, respectively, were expressed in a baculovirus system. All four expression products were detected by immunoblotting with two antisera capable of inducing passive HN (pHN). When antibodies eluted from glomeruli of rats with pHN were used for immunoblotting, only the expression product encoded by fragment II was detected. This indicates that the second cluster of LDLR ligand binding repeats is directly involved in binding anti-megalin antibodies and in the induction of pHN. To narrow the major epitope in this domain, fragment II was used to prepare proteins sequentially truncated from the C- and N-terminal ends by in vitro translation. Analysis of the truncated translation products by immunoprecipitation with anti-megalin IgG revealed that the fifth ligand-binding repeat (amino acids 1160-1205) contains the major epitope recognized. This suggests that a 46-amino acid sequence in the second cluster of LDLR ligand binding repeats contains a major pathogenic epitope that plays a key role in pHN. Identification of this epitope will facilitate studies on the pathogenesis of HN.

Induction of passive Heymann nephritis with antibodies specific for a synthetic peptide derived from the receptor-associated protein

Passive Heymann nephritis (pHN) is an experimental rat model for human membranous glomerulopathy. In pHN, the formation of subepithelial immune deposits (ID) involves as antigenic targets the membrane glycoprotein gp330/megalin and the 44-kD receptor-associated protein (RAP). A single binding site for ID- inducing antibodies (Abs) was previously mapped to the 86 NH2-terminal amino acids of RAP (RAP1-86). To further narrow this epitope, Abs eluted from the glomeruli were immunoblotted on membranes that were loaded with overlapping synthetic peptides representing the amino acid sequence of RAP (SPOTs system). Two adjacent Ab-binding domains with the sequences PVRLAF, (amino acids 39-44) and HSD-LKIQE (amino acids 46-53), which were separated by a single L residue at amino acid 45, were detected. Rabbit Abs raised against synthetic peptides containing these domains individually (P31-44 and P46-53) failed to procedure glomerular IDs. By contrast, Abs raised against a larger composite peptide (P31-53) induced IDs within 3d that were firmly cross linked to the glomerular basement membrane. These data suggest that Ab binding in vivo depends on the conformation of the antigenic target sequence that is preserved in the synthetic peptide P31-53, which covers the entire Ab-binding domain of RAP but not in its subdomains, P31-44 and P46-53. Collectively, these results locate the sole ID-inducing epitope of RAP to amino acids 39-53.

Molecular analysis of the pathological autoimmune antigens of Heymann nephritis

Images

Induction of Heymann nephritis with a gp330/megalin fusion protein

There is considerable evidence that glomerular deposits in Heymann nephritis, a rat model of membranous nephritis, result from shedding of immune complexes formed on podocytes and that the principal antigen is part of the extracellular domain of a cell surface glycoprotein receptor called gp330 or megalin. It has also been reported that the immunogen that induces Heymann nephritis is a complex formed between gp330 and the receptor-associated protein RAP. The recent elucidation of the primary structure of gp330 makes it possible to investigate the ability of defined portions of gp330, devoid of RAP, to induce Heymann nephritis. In the present study we show that a gp330-glutathione-S-transferase fusion protein, containing 137 amino acid residues (1114 to 1250) of the ectodomain, induces active Heymann nephritis and that heterologous antibodies against this fusion protein produce passive Heymann nephritis. By immunofluorescence, typical glomerular immunoglobulin deposits were found, but complement components were lacking and the rats did not develop proteinuria. In the active model, we obtained evidence indicating that the deposits contained portions of the ectodomain of gp330, including regions other than those of the fusion protein. Thus, the deposits were stained by polyclonal antibodies to gp330 and to the gp330 fusion protein, as well as by two monoclonal antibodies reactive with portions of the ectodomain of gp330, only one of which reacted with the fusion protein in vitro. Antibodies against the cytoplasmic domain of gp330 did not stain. Furthermore, we found that RAP was able to bind to gp330 in the glomerular deposits but not to the gp330 fusion protein in vitro. The results show that the region of gp330 spanning amino acid residues 1114 to 1250 contains peptides capable of inducing pathogenic antibodies of Heymann nephritis without a contributory role of RAP.

A novel class of autoantigens of anti-neutrophil cytoplasmic antibodies in necrotizing and crescentic glomerulonephritis: the lysosomal membrane glycoprotein h-lamp-2 in neutrophil granulocytes and a related membrane protein in glomerular endothelial cells

Necrotizing and crescentic glomerulonephritis (NCGN) is frequently associated with circulating antineutrophil cytoplasmic autoantibodies (ANCA). It is established that ANCA are specific for soluble enzymes of granules of polymorphonuclear neutrophil granulocytes (PMN), such as myeloperoxidase (MPO) or protease 3 (PR3). The purpose of this study was to identify membrane proteins of PMNs, and/or glomerular cells, as additional autoantigenic ANCA targets. When membrane protein fractions were prepared from PMNs and isolated human glomeruli, and immunoblotted with ANCA sera of NCGN patients, two bands with apparent molecular masses of 170 and 80-110 kD (gp170/80-110) were labeled in PMNs, and a 130-kD glycoprotein (gp130) in glomeruli. Gp130 was purified, and monoclonal and rabbit antibodies (Abs) were produced which showed the same double specificity as the patient's ANCA. Using these probes, evidence was provided that gp170/80-110 is identical with human lysosomal-associated membrane protein 2 (h-lamp-2), because both proteins were immunologically cross-reactive and screening of a cDNA expression library from human promyelocytic leukemia cells with anti-gp130 Ab yielded a clone derived from h-lamp-2. Gp170/80-110 was localized primarily in granule membranes of resting PMNs, and was translocated to the cell surfaces by activation with FMLP. By contrast, gp130 was localized in the surface membranes of endothelial cells of human glomerular and renal interstitial capillaries, rather than in lysosomes, as found for h-lamp-2. Potential clinical relevance of autoantibodies to gp170/80-110 and gp130 was assessed in a preliminary trial, in which ANCA sera of patients (n = 16) with NCGN were probed with purified or recombinant antigens. Specific reactivity was detected in approximately 90% of cases with active phases of NCGN, and frequently also in combination with autoantibodies specific for PR3 or MPO. Collectively, these data provide evidence that h-lamp-2 in PMNs and a different, structurally related 130-kD membrane protein on the cell surface of renal microvascular endothelial cells are autoantigenic targets for ANCA in patients with active NCGN.

Cloning, characterization, and chromosomal localization to 4p16 of the human gene (LRPAP1) coding for the alpha 2-macroglobulin receptor-associated protein and structural comparison with the murine gene coding for the 44-kDa heparin-binding protein

We report the molecular cloning of the human gene (symbol LRPAP1) coding for the alpha 2-macroglobulin receptor-associated protein (A2MRAP), as well as the gene coding for the 44-kDa heparin-binding protein (HBP-44), its murine counterpart. For both, genomic cosmid clones were isolated, and for the human gene a bacteriophage P1 clone containing the entire A2MRAP gene was also retrieved. The genes were characterized after subcloning: in both species, the known coding part of the cDNA is encoded by eight exons, and the position of the boundaries of the exons was conserved. The human LRPAP1 locus was assigned to chromosome 4 by PCR of human-hamster hybrid cell lines and by fluorescence in situ hybridization to band 4p16.3. This maps closely to the variable constitutional deletions of the short arm of chromosome 4, observed cytogenetically in patients with the Wolf-Hirschhorn syndrome. Metaphase spreads of two such patients were analyzed by fluorescence in situ hybridization with an LRPAP1 genomic probe. The first patient, with karyotype 46,XY,del4(p14-p16.1), had retained both copies of the LRPAP1 gene. In contrast, the other patient, with karyotype 46,XY,del4(p15.3-pter), displayed no signal for LRPAP1 on the deleted chromosome.

Proteinuria in passive Heymann nephritis is associated with lipid peroxidation and formation of adducts on type IV collagen

Passive Heymann nephritis (PHN) is a model of human membranous nephropathy that is characterized by formation of granular subepithelial immune deposits in the glomerular capillary wall which results in complement activation. This is causally related to damage of the filtration barrier and subsequent proteinuria. The local accumulation of injurious reactive oxygen species (ROS) is a major effector mechanism in PHN. ROS may induce tissue damage by initiating lipid peroxidation (LPO). In turn, this leads to adduct formation between breakdown products of LPO with structural proteins, such as formation of malondialdehyde (MDA) or 4-hydroxynonenal-lysine adducts. To examine the role of LPO in the development of proteinuria we have localized MDA and 4-hydroxynonenal-lysine adducts in glomeruli of PHN rats by immunofluorescence microscopy, using specific monoclonal antibodies. By immunogold electron microscopy, MDA adducts were localized to cytoplasmic vesicles and cell membranes of glomerular epithelial cells, to the glomerular basement membrane (GBM), and also to immune deposits. Type IV collagen was specifically identified as being modified by MDA adducts, using a variety of techniques. Collagenase pretreatment of GBM extracts indicated that the NC-1 domain of type IV collagen was a site of adduct formation. When LPO was inhibited by pretreatment of PHN rats with the antioxidant probucol, proteinuria was reduced by approximately 85%, and glomerular immunostaining for dialdehyde adducts was markedly reduced, even though the formation of immune deposits was not affected. By contrast, lowering of the serum cholesterol levels had no influence on the development of proteinuria. These findings are consistent with the premise that ROS-induced glomerular injury in PHN involves LPO and that this results not only in damage of cell membranes but in modification of type IV collagen in the GBM as well. The close temporal correlation of the occurrence of LPO with proteinuria and the ability of probucol to inhibit proteinuria support a causal role for LPO in the the alteration of the glomerular permselectivity which results in proteinuria.

gp330 and RAP: the Heymann nephritis antigenic complex

It is apparent that significant progress has been made in the characterization of gp330 in the years that have elapsed since its initial identification as the nephritogenic antigen of Heymann nephritis. However, there are still many gaps in our knowledge and we do not yet have a full picture of the molecular events leading to the formation of immune deposits in glomerular capillaries. Moreover, we still do not have direct information on the normal function(s) of gp330 and RAP and their trafficking in renal and other epithelia. The availability of the yolk sac and other cell lines that express gp330 and RAP together with the identification of the functional domains of RAP should greatly facilitate experimental studies designed to elucidate these problems. Progress will also be greatly facilitated in the future when the complete amino-acid sequence of gp330 becomes available, making possible further structural studies. It is our hope that new knowledge obtained on the molecular mechanisms of HN will provide insights into the molecular pathogenesis of human membranous nephropathy and will provide a strategy for the design of appropriate treatments to interrupt the process.

Autoimmune myocarditis induced in mice by cardiac C-protein. Cloning of complementary DNA encoding murine cardiac C-protein and partial characterization of the antigenic peptides

Autoimmune myocarditis is considered to play a major role in the pathogenesis of dilated cardiomyopathy. A new autoimmune myocarditis model was attained by repeated immunization using murine cardiac C-protein with the immunological adjuvant, Klebsiella pneumoniae O3 lipopolysaccharide. For further analysis of a pathological epitope, the cDNA encoding C-protein was isolated; a fusion protein encoded by part of this cDNA induced myocarditis in SMA mice as well as in three other strains: DBA/1J (H-2q), O20/A (H-2pz1), and SJL (H-2s). The nucleotide sequence and its deduced amino acid analysis revealed that this protein had immunoglobulin-like and fibronectin-like repeats. This study provides a new animal model of autoimmune myocarditis which may shed light on the pathogenesis of dilated cardiomyopathy.

Functional domains of the receptor-associated protein (RAP)

The receptor-associated protein (RAP) specifically associates with gp330 and the low density lipoprotein (LDL) receptor-related protein (LRP), the two newest members of the LDL receptor gene family. Results obtained by ligand blotting, affinity chromatography, and density-gradient sedimentation demonstrate that RAP binds to both receptors with high affinity and that the binding is Ca2+ dependent. RAP also binds heparin and is identical to a mouse heparin binding protein (HBP-44) identified in a teratocarcinoma cell line (F9). While biochemical studies have shown that RAP is present on the cell surface and is an effective inhibitor of ligand binding to gp330 and LRP, immunocytochemical findings indicate that RAP is most abundant in the endoplasmic reticulum lumen and may function in receptor folding and/or trafficking. To facilitate the characterization of RAP's function(s) we have mapped its gp330 and heparin binding sites by performing direct binding studies on fusion proteins representing overlapping domains of RAP. gp330 was found to bind to two separate sites on RAP--i.e., between amino acids 85-148 and 178-248. Binding studies with radiolabeled heparin indicate that the heparin binding site is between amino acids 261 and 323, which is consistent with our previously proposed site (residues 287-306) based on the amphipathic nature of the C terminus of RAP. These data demonstrate that the gp330 and heparin binding sites and the Heymann nephritis pathogenic epitope (amino acids 1-86) demonstrated earlier are represented by distinct domains of the RAP polypeptide.

Antibodies to glycolipids activate complement and promote proteinuria in passive Heymann nephritis

Passive Heymann nephritis is an experimental rat model of human membranous nephropathy induced by injection of antisera against crude renal cortical fractions such as Fx1A or rat tubular microvilli. This results in the formation of subepithelial immune deposits, the activation of the C5b-9 membrane attack complex of complement, and severe proteinuria. While the formation of immune deposits is attributed to in situ immune complex formation with antibodies specific for the gp330-Heymann nephritis antigenic complex (HNAC), activation of complement and proteinuria appear to be caused by at least one additional antibody species present in anti-Fx1A sera. We have separated by affinity absorption polyspecific antisera against Fx1A and rat microvilli into one IgG fraction directed specifically against microvillar proteins (anti-Fx1A-prot) and another IgG fraction specific for glycolipids (ant-Fx1A-lip) of tubular microvilli. When injected into rats, the anti-Fx1A-prot fraction induced immune deposits but failed to activate complement or produce proteinuria, similar to results obtained with affinity-purified anti-gp330 IgG. When the antibodies of the anti-Fx1A-lip fraction were injected alone they did not bind to glomeruli. By contrast, when the IgGs specific for the Fx1A-prot fraction (or for gp330-HNAC) were combined with those directed against the Fx1A-lip glycolipid preparation, immune deposits were formed, in situ complement activation was observed, and also proteinuria was induced. It is concluded that within anti-Fx1A and anti-microvillar sera there are at least two IgG fractions of relevance for the development of PHN: one directed against the gp330-HNAC complex which is responsible for the development of immune deposits, and a second specific for glycolipid antigen(s) which activate(s) the complement cascade.

Distribution of mRNA coding for alpha-2-macroglobulin, the murinoglobulins, the alpha-2-macroglobulin receptor and the alpha-2-macroglobulin receptor associated protein during mouse embryogenesis and in adult tissues

The distribution of mRNA coding for the members of the wide-spectrum proteinase scavenging system of the alpha-2-macroglobulin family was examined in the mouse: Alpha-2-macroglobulin (MAM), the murinoglobulins (MUG), the alpha-2-macroglobulin receptor (alpha 2MR) and the receptor associated protein, the heparin binding protein-44 (alpha 2MRAP/HBP-44), a component of unknown function. The results demonstrate that MAM is expressed in the mouse embryo exclusively in the liver and not before day 13 of gestation. MUG mRNA was never detected during embryogenesis. On the other hand, both the alpha 2MR and the alpha 2MRAP/HBP-44 messages were present throughout all embryonal stages examined. The distribution of the alpha 2MR mRNA was widespread in most tissues, with stronger signals observed in developing mouse brain, in whisker follicles and in the perifollicular mesenchyme, in lung, liver, kidney, intestine and placenta. The alpha 2MRAP/HBP-44 mRNA was detected predominantly in brain, lung, liver, kidney and placenta. Interestingly, within each tissue the cellular distribution of the alpha 2MR and alpha 2MRAP/HBP-44 mRNA was quite different with the most remarkable extremes observed in kidney and in placenta. The implication of these observations for receptor expression and function are discussed. Northern analysis of adult tissues extended these observations: major signals for MAM and MUG were seen only in liver, while the expression of the alpha 2MR and the alpha 2MRAP/HBP-44 was widespread with highest levels of the 15-kb alpha 2MR mRNA in liver. Kidney was the most abundant source of alpha 2MRAP/HBP-44 mRNA with the 1.8- and 3.6-kb mRNAs, derived from the same gene by alternative mRNA splicing, present in nearly constant ratios in most tissues, except in testis. The notable absence of expression of MAM in the first half of gestation indicates that during this period the receptor is scavenging for proteinases complexed to MAM derived from the maternal circulation or is being used for endocytosis of the other documented ligands, such as plasminogen activator complexes or apolipoprotein E-containing lipoprotein particles.

Effects of laminin monoclonal antibodies on the development of cultured rat embryos

Whole rat embryos (9.5 days of gestation) were exposed to six different monoclonal antibodies to laminin during 48 hr of culture. Four (LAM I, LAM V, 5A2, 9D2) of the six were teratogenic or lethal and two (LAM II, 5D3) were not toxic at comparable levels. Teratogenicity and lethality were not related to antibody level, subclass or affinity for whole laminin. Indirect immunofluorescence studies using mouse embryo sections revealed that the toxic antibodies bound in a diffuse manner, while the nontoxic antibodies showed distinct labeling of tissues. These observations suggest that previous varied responses seen in cultured rat embryos exposed to laminin antibodies obtained from humans, monkeys, and rats were the result of differences in the epitope specificity of those antibodies.

Immunocytochemical and biochemical characterization of the Heymann nephritis antigenic complex in rat L2 yolk sac cells

Heymann nephritis in the rat is the most widely used model of human membranous glomerulonephritis. Glycoprotein (gp)330, a large (M(r) > 550,000) membrane-associated glycoprotein, has been identified as the main antigen in this autoimmune disease. Studies of gp330 and receptor-associated protein (RAP), its 44-kd subunit, have been restricted largely to rat kidney, as no stable cultured cell line has been available that expresses gp330. We have recently identified a rat yolk sac carcinoma cell line (L2) that expresses both gp330 and RAP. In this report, we have carried out detailed morphological, immunocytochemical, and biochemical studies characterizing the biosynthesis and localization of gp330 and RAP in the L2 rat yolk sac cell line. At the electron microscope level, the L2 cells are seen to be attached by cell junctions, and their predominant morphological features include extensive networks of rough endoplasmic reticulum (ER) and numerous clathrin-coated pits found on the cell membrane. By immunocytochemistry, gp330 was localized primarily to clathrin-coated pits at the cell surface, whereas RAP was localized predominantly to the lumen of the rough ER. Pulse-chase experiments indicated that gp330 spends a prolonged time maturing in the ER of L2 cells, as transport of gp330 to the Golgi complex (based on acquisition of endoglycosidase H resistance) is slow (t1/2 = 90 to 120 minutes). Gp330 reached the L2 cell surface beginning at 2 hours after synthesis, where it could be detected by cell surface immunoprecipitation. RAP was found to be an N-linked glycoprotein, and it remained endoglycosidase H-sensitive up to 4 hours after synthesis. Co-precipitation and co-sedimentation experiments demonstrated that gp330 and RAP form a large heterodimer (M(r) approximately 669,000) immediately after biosynthesis and are further assembled into a large hetero-oligomer in the ER. These findings demonstrate that the localization and the kinetics of assembly of gp330 and RAP into the Heymann nephritis antigenic complex are similar in both L2 cells and rat kidney. They also provide new information on the intracellular processing of these two molecules and their delivery to the cell surface. Thus, the L2 cell system should facilitate further characterization of the functions and interactions of gp330 and RAP, which may shed light on the cellular and molecular mechanisms of Heymann nephritis.

Identification of a cell line that expresses a cell surface and a soluble form of the gp330/receptor-associated protein (RAP) Heymann nephritis antigenic complex

gp330 is a large glycoprotein located in clathrin-coated pits at the surface of the glomerular and proximal tubule epithelia in the rat kidney. It was originally identified as the target of autoimmune antibodies in Heymann nephritis (HN) and has since been shown to be a member of the low density lipoprotein receptor gene family and to form a stable association with receptor-associated protein (RAP), which together constitute the HN antigen complex (HNAC). Progress in defining the normal functions of gp330 as well as the molecular mechanisms of HN has been hampered by the lack of an available kidney cell line that expresses this protein. We here report the identification of a rat yolk sac carcinoma cell line (L2) that synthesizes HNAC and expresses it in coated pits at the cell surface. gp330 and RAP from L2 cells are immunologically identical to their kidney counterparts, and peptide maps of gp330 yielded identical peptide fragments. Characterization of the cell line revealed that there are 3.3 x 10(4) gp330 molecules per L2 cell and that the cells produce a soluble form of gp330 that is released into the medium. Heparin ligand blot analysis demonstrated that RAP but not gp330 binds heparin. By heparin affinity chromatography, gp330 and RAP copurify, indicating that the glycosaminoglycan binding site within RAP is accessible when the subunit is complexed with gp330. These results indicate that the L2 cell line provides a valid and useful model for studies on the function of HNAC and the pathogenesis of HN.