Nephrol Dial Transplant (1992) 7: 293-299 £ 1992 European Dialysis and Transplant Association-European Renal Association
Nephrology Dialysis Transplantation
Original Article
Presence of circulating antibodies against brush border antigens (FxlA) in a patient with membranous nephropathy and bilateral pyeloureteral stenosis. Comparison with idiopathic membranous nephropathy J. Gonzalez-Cabrero1, R. de Nicolas1, A. Ortiz1, F. Mampaso2, L. Hernando1 and J. Egido1 'Fundacion Jimenez Diaz, Universidad Autonoma, CSIC; 2Hospital Ramon y Cajal, Madrid, Spain
Abstract. In a patient with membranous nephropathy and bilateral pyeloureteral stenosis with hydronephrosis, we examined the possibility that an increase in the intratubular pressure could facilitate the passage of the FxlA antigens to the circulation. Elevated serum anti-FxlA antibodies were detected in this particular patient by ELISA on three occasions during the disease follow-up, even though he was in clinical remission. These antibodies reacted in vitro with the tubular brush border of a normal human kidney. The anti-FxlA antibodies isolated from the patient's sera by affinity chromatography competed with the rabbit anti-FxlA antisera binding to plates coated with human FxlA antigen. In immunoblotting studies the isolated specific IgG antibodies from that patient reacted with a 180 kDa antigen of the human FxlA and with less intensity with 75 kDa and 50-55 kDa polypeptides. In none of 12 patients with idiopathic membranous nephropathy could the circulating anti-FxlA antibodies be demonstrated. On the whole, this particular case suggests that on some occasions increased intratubular pressure could cause the release of FxlA antigens, facilitating an autologous immunocomplex nephritis. These antigens, by contrast, do not seem to play any ro!e in most cases of membranous nephropathy in man. Correspondence and offprint request* to: Dr J. Egido, Servicio de N'efrologia, Fundarion Jimenez Diaz, Avda Reyes Catolicos 2, 28040 Madrid, Spain.
Key words: anti-FxlA antibodies; bilateral pyeloureteral stenosis; membranous nephropathy; renal tubular antigens
Introduction Heymann nephritis in rats, an animal model of membranous nephropathy, has been induced by immunization with various preparations of proximal tubular brush border proteins [1]. In a first set of experiments Edgington et al. [2] demonstrated that the antigen responsible for the disease was essentially located within a membrane preparation (FxlA) derived from tubular brush borders from which a more purified preparation, RTE 5, was isolated and found to be nephritogenic. The immune aggregates in this disease were shown to contain FxlA and anti-FxlA. The same immune histochemical pattern, however, was obtained from passive administration to animals of the anti-FxlA antibody alone [3]. Subsequent studies showed that normal rat glomeruli possessed a fixed tissue antigen with immunological similarity to FxiA, allowing the in situ formation of complexes along the glomerular basement membrane [4]. These observations in rats raised the possibility that some cases of human membranous nephropathy might be associated with autoimmunization to similar antigens. Naruse et al. [5] have described the presence
194
of renal tubular epithelial antigen along the glomerular capillary wall in four of nine cases of idiopathic membranous nephropathy. However, other authors failed to confirm these findings [6-8]. Interestingly, other investigators have reported the presence of tubular epithelial antigens in glomerular lesions associated with a variety of conditions such as renal-vein thrombosis, sickle-cell disease, renal-cell carcinoma, Fanconi syndrome, systemic lupus erythematosus, and in a kidney allograft [9-15]. The existence of circulating antibodies against renal tubular epithelial antigens has also raised controversy. Free antibodies to FxlA were not apparent in the serum of several groups of patients with idiopathic membranous nephropathy when using an indirect immunofluorescence or radioimmunoassay [6,7]. By contrast, its presence has been well documented in isolated cases [15,16]. In this paper we present a patient with membranous nephropathy associated with hydronephrosis secondary to pyeloureteral stenosis in which elevated levels of anti-FxlA antibodies were detected on three occasions during the follow-up. By contrast, these antibodies were not found in a group of twelve patients with idiopathic membranous nephropathy.
Subjects and methods Subjects Serum from thirteen patients with biopsy-proven membranous nephropathy and eight healthy controls was obtained by clotting of blood and stored in aliquots at —70cC until tested. Serum from the propositus was collected in three different periods, and on two occasions in other four patients.
Clinical features A 65-year-old male was first seen in our hospital for abdominal pain consistent with renal colic in 1976. He had been diagnosed previously as having hepatitis, but until his admission he was in excellent health. On examination the patient appeared well. The pulse was 80 and the blood pressure 150/90 mmHg. No murmur was detectable, but dextrocardia was noted. Neither liver nor spleen were palpable. There was no lymphadenopathy, articular abnormality, or peripheral oedema. Laboratory data on admission showed creatinine 97 umol/1 (1.1 mg/dl), urea 8.5 mmol/l (48 mg/dl) and creatinine clearance 1.45 ml/s (87 ml/min). Proteinuria (sulphosalicylic acid) was 0.22 g/day. Addis count evidenced 13x10* leukocytes and 152 x 106 red blood cells. Urine culture was negative. HBsAg was negative, but anti-HBsAg was positive. Chest-roentgenogram revealed dextrocardia. In a urographic examination bilateral pyeloureteral stenosis was evident (Figure 1). The stenosis was considered to be congenital in origin. Because of high blood pressure he was discharged on hydrochlorothiazide and amiloride.
J. Gonzalez-Cabrero et al.
Fig. 1. Urographic examination of the propositus. Note the Bilateral pyeloureteral stenosis producing a certain degree of hydronephrosis.
In December 1979 nephrotic syndrome appeared. On admission proteinuria was 7.5 g/day, creatinine 99 umol/1 (1.13 mg/dl), creatinine clearance 1.35 ml/s (81 ml/min) and serum albumin 32 g/1. A renal biopsy was performed. With the diagnosis of membranous nephropathy (see results) the patient was discharged on propranolol and hydrochlorothiazide. Since then he has maintained a normal creatinine clearance, nephrotic syndrome disappeared in early 1981 and the proteinuria has ranged between 0.2 and 0.35 g/day. Circulating anti-Fxl A antibody was detected in 1985-1986, while the patient was in partial clinical remission (the only period studied). In February 1989 the creatinine was 97 umol/1 (1.1 mg/dl), creatinine clearance 1.56 ml/s (94 ml/min), proteinuria 0.18 g/day, total serum protein 69 g/1, and cholesterol 5.7 mmol/l (221 mg/dl). Urine culture was negative. In May 1991 analytical values were similar and intravenous urography revealed no new findings. A group of 12 patients with idiopathic membranous nephropathy was studied in order to find the prevalence, if any, of circulating anti-FxlA antibodies. The age ranged from 22 to 65 years, and proteinuria from 15 to 5 g/day. All patients had normal renal function. No lmmunological abnormalities were detected that could suggest a particular cause.
Measurement of circulating immune complexes (IC) The presence of serum IgG-IC in 10 patients and in six controls was assessed by the Raji cell assay, as previously published [17]. The mean of the duplicate value of the test sera (Ti) was divided by Ui (upper 95% confidence limit in control sera). If the value Ti/Ui was ;> I the test serum was judged abnormal.
Brush border antigens (FxlA) and antibodies Rat and human FxlA were prepared as previously described [2,18]. Antibodies against human FxlA were raised in rabbits, following the immunization protocol of Zamlanski-Tucker et al. [19].
Anti-FxlA antibodies in membranous nephropathy
Measurement of specific antibodies against FxlA antigens The presence of circulating antibodies against FxlA was studied by enzyme-linked immunosorbsnt assay (ELISA) as follows: Polystyrene microtitre plates (Costar, Broadway, Cambridge. MA, USA) were coated with 100 ng ml (200 ul well) of human FxlA in 0.01 M carbonate-bicarbonate buffer pH 9.6 and was incubated overnight at 4=C. After washing with 0.15 M phosphate-buffered saline pH 7.3 (PBS), 0.1% Tween 20, two-fold serial dilutions, from 1:10 to 1:1280, of the serum collected in three different periods from the propositus, serum from 12 patients with idiopathic membranous nephropathy, and serum from eight healthy control subjects was added. After incubation at room temperature in a moist chamber for 2 h the unbound material was removed by washing with PBS-Tween, and a 1:1000 dilution of peroxidase-labelled goat anti-human IgG (Nordic Immunology, Tilburn, The Netherlands) was added and incubation proceeded for 2 h at room temperature. The enzyme substrate used was 0.4mg/ml of 1,2phenylenediamine (OPD) (Merck) in 0.1 M citric acid, 0.2 M phosphate buffer, pH 5.0 with 0.04 ul of H2O2 30%/ml buffer. The developed colour was determined in a Titertek Multiskan plate reader (Flow Laboratories, Ltd, Scotland, UK). All tests were performed in duplicate.
295
Immunofluorescence studies A normal human kidney specimen was frozen at — 20=C and sections 4 um thick were cut on a microtome cryostat. Slides, after being air-dried for 30 min, were washed three times with PBS pH 7.3 and incubated with various dilutions of serum from the propositus or with 200 ug mi of antiFxlA antibodies isolated by affinity chromatography in a moist chamber at room temperature for 40 min. An indirect immunofluorescence was then performed by incubating with fluorescein isothiocyanate (FIIC)-conjugated rabbit antihuman IgG (Behring, Behringwerke AG, Marburg, Germany)fivefolddiluted for another 40 min. After rinsing with PBS, slides were mounted in PBS-glycerin pH 7.3 and examined.
Poly aery lamide gel electrophoresis (SDS-PAGE) gradient Human FxlA were analysed by sodium dodecyl sulphate (SDS)-PAGE according to Weber and Osborn [21]. To 50 ug and 100 ug of soluble crude fraction in 0.08 TrisHC1 buffer pH 6.8 were added 10% glycerol, 1% SDS and 0.005% of bromophenol blue, and the mixture was run in duplicate simultaneously with various molecular-weight markers, using sled gel of 5-15% polyacrylamide gradient. Gel strips were stained with Coomassie Blue (Merck) or electrophoretically transferred to nitrocellulose.
Isolation of circulating anti-FxlA antibodies Antibodies against human FxlA antigens were isolated from about 22 ml of the propositus serum by using affinity chromatography on AH-Sepharose 4B (Pharmacia Fine Chemicals, Uppsala, Sweden) activated with glutaraldehyde and cross-linked to FxlA [20]. The elution of the fraction that bound to the immunoabsorbent column was performed with 2 M K.SCN in 0.01 M phosphate buffer, pH 7.0. After that it was extensively dialysed against PBS pH 7.3. The specificity of these antibodies was tested by adding the eluted fraction in decreasing amounts, ranging from 50 to 780 ng/well, on polystyrene plates previously coated with human FxlA. Peroxidase-conjugated goat anti-human IgG and the substrate were added in the same conditions as above, and the amount of the enzyme found was measured by absorbance at 495 nm.
Inhibition studies with rabbit anti-FxlA antiserum Studies were performed to determine the mutual blocking abilities of heterologous antibody to FxlA and of patient's antibodies. Anti-FxlA antibodies (500 ug/ml), isolated from the patient's sera by affinity chromatography, and twofold serial dilutions (from 1:20) of rabbit antiserum to FxlA were incubated for 2 h at room temperature in polystyrene plates previously coated with human FxlA. Afterwards a 1:1000 dilution of peroxidase-conjugated goat anti-human IgG was added. The enzyme substrate was also OPD. These tests were also performed in duplicate. Results were expressed as percentage of inhibition according to the equation: Percentage of inhibition ft absorbance of inhibition mixture N = 100 x 1 — absorbance without rabbit antiserum;
Immunoblotting Proteins from human FxlA fraction, separated by SDSPAGE, were transferred to nitrocellulose membrane (0.45 u) (Bio-Rad, Richmond, CA, USA) under electric current as described by Hudson and Hay [22]. Following binding of antigens (FxlA) to porous membrane, the remaining unbound sites were blocked with 3% gelatin (Bio-Rad) in 20 mM Tris, 500 mM NaCl buffer pH 7.5. After washing with the same buffer without gelatin, nitrocellulose was incubated with a 1:10 dilution of serum from the propositus for 3 h at room temperature. Unbound material was removed and a 1:1000 dilution of peroxidase-labelled goat anti-human IgG (Nordic Immunology) was added, and incubation was allowed to proceed for 2 h at room temperature. Colour development of bands was obtained with 4-chloro-l-naphtol (Sigma) as enzyme substrate in methanol Tris-NaCl with H2O2, and the reaction was stopped with distilled water.
Results Light-microscopic examination of the renal biopsy of the patient with bilateral pyeloureteral stenosis showed a normal general architectural pattern, with thickening of the glomerular basement membrane. Fine granular trichrome red deposits were observed on the epithelial surface of the basement membrane (stage I). The vessels, the tubules, and the interstitium appeared normal. The results of immunofluorescence were consistent with membranous nephropathy, revealing IgG and C3 localized in a granular pattern
J. Gonzalez-Cabrero el al.
296
along the glomerular basement membrane. No staining was noted in the proximal tubules epithelium. Since the existence of serum antibodies against FxlA antigens was shown several years after the biopsy, no attempts were made to study the presence of autologous tubular epithelial antigens in the glomerular deposits. As a urographic examination had shown bilateral pyeloureteral stenosis with a certain grade of hydronephrosis (Figure 1), we wondered whether the increased pelvic pressure could have damaged the tubular epithelial cells with the subsequent passage of antigens to the circulation. The sensitive ELISA employed in this study permitted us to observe that the patient's serum contained an antibody, with a titre of 1 :640, that bound specifically to the plates coated with the solubilized crude renal cortical fraction commonly referred to as FxlA. The titre of this antibody remained quite constant on three occasions over the 18 months of follow-up (Figure 2). By contrast, none of the 12 patients with idiopathic membranous nephropathy was above the normal range of eight healthy controls. In a parallel group of experiments we searched for the reactivity of circulating antibodies in the propositus and in controls with rat FxlA antigens. Polystyrene microtitre plates were
coated and incubated with lOOug/ml of rat FxlA and proceeded in the same manner as described above with the human FxlA. At variance with the human FxlA results, the patient's serum only reacted with the rat FxlA antigens at concentrations of 1 : 10 to 1:40 (not shown). By indirect immunofluorescence, both serum and isolated anti-FxlA antibodies stained proximal tubules of normal human kidney (Figure 3). Fluorescence was abolished after absorption with human FxlA, but not by normal human serum. Similar results, but with a higher staining intensity, were noted with rabbit antiserum to human anti-FxlA. Competition studies between patient anti-FxlA antibodies and rabbit anti-FxlA antiserum revealed an inhibition percentage of 67% at 1 : 20 rabbit serum dilution to 11% at 1: 1280 dilution (Figure 4). Analysis by SDS-PAGE showed that soluble FxlA
Fig. 3. Section of normal human kidney reacting with serum of the propositus and fluorescent-labelled rabbit anti-human IgG. The luminal regions of proximal tubules stain brightly. No staining was seen in the glomeruli x 440.
8*r\un dilutions
Fig. 2. Presence of antibodies against FxlA antigens in the patient with membranous nephropathy and bilateral pyeloureteral stenosis ( • ) (mean of determinations on three occasions). The figure also shows mean values of 12 patients with idiopathic membranous nephropathy (C). The space between continuous lines represents mean±l SD of the values found in eight control subjects. The value*, of the propositus arc statistically significant at all strum dilutions in relation to controls.
1/2O
l/BO
1/32O
1/128O
Fig. 4. Competition studies between patient anti-FxlA antibodies and rabbit anti-FxlA antiserum (see Methods).
Anti-FxlA antibodies in membranous nephropathy
297
resolved into several polypeptide bands ranging from 50-55 to 600 kDa. Immunoblotting of this gel after being transferred to nitrocellulose demonstrated that specific IgG antibodies from the patient's serum chiefly reacted with polypeptides of 170 180 kDa, and, with minor intensity, with 75 and 50-55 kDa polypeptides (Figure 5). Levels of circulating IC measured by Raji-cell assay, were slightly elevated (Ti/Ui> 1) in two of 10 (20%) patients with idiopathic membranous nephropathy studied. By contrast, in the patient with membranous nephropathy associated with bilateral pyeloureteral stenosis no complexes were detected on the three occasions studied during the follow-up of the disease.
Discussion Our patient has some features in common with the Heymann nephritis model. Besides typical membranous deposits of IgG, the patient also possessed serum antibodies with avidity for some antigens contained in the human FxlA fraction and in the proximal renal tubular brush border from normal human kidK D
4OO 2OO
-••
116
—
ISO
•~66
75
—
A
B
Fig. 5. SDS-poIyacrylamide gel electrophoresis of soluble human FxlA. The A columns show the molecular weight markers as kDa (Biorad patterns). The B column shows the immunoblotting of this gel, after being transferred to nitrocellulose. The bands observed correspond to the approximate molecular weights of the polypeptides reacting with IgG antibodies from the propositus serum.
ney. Competitive experiments showed that antibodies found in the propositus and the rabbit antihuman FxlA shared antigenic specificities. This observation could suggest that immune complexes formed with FxlA-anti-FxlA antibodies are operative in this patient in a similar manner to that established in Heymann nephritis. The nephritogenic antigens responsible for Heymann nephritis have been relatively well characterized in the last years. Kerjaschki and Farquhar [4] have identified a 330 kDa glycoprotein (gp 330) that is localized in the apical vesicles of the epithelial cells of proximal tubules and in the coated pits of glomerular visceral epithelial cell. Furthermore they have shown that gp 330 directly participates in the formation of immune deposits [23]. In subsequent studies, several authors have isolated antigens that reacted with anti-brush border antibodies, with molecular weights ranging from 95 to 660 kDa [24-26]. Makker and Singh proposed that gp 600 is a precursor of gp 330 [26]. The fact that immunization with FxlA produces a more severe form of Heymann nephritis than gp 330 suggests that FxlA contains at least one nephritogenic antigen in addition to gp 330 [24]. By contrast, the antigens involved in the formation of immune deposits in human membranous nephropathy are unknown. Recently, Kerjaschki et al. [27] have reported that a membrane protein with an apparent molecular weight of 400 kDa is present in human kidney. This molecule, immunologically crossreactive with the rat gp 330, is localized in human proximal tubules, but it is absent from the podocytes of glomeruli. The IgG isolated from our patient's serum reacted with two predominant bands at 170180 kDa and 50-55 kDa of the human FxlA, which contains polypeptides with molecular weights from 600 to 50-55 kDa. Makker et al. [26] showed that gp 300, after its isolation from gp 600, was highly unstable and spontaneously degraded into smaller fragments of 100 and 70 kDa. Therefore it is possible that the patient's IgG detected several gp 300 fragments sharing antigenic determinants. The participation of antibodies against tubular brush border antigens in the pathogenesis of membranous nephropathy in this patient with bilateral pyeloureteral stenosis is not clear because they were found when he was in partial clinical remission (the only period studied). The antibodies were detected when the nephrotic syndrome was in partial remission. Unfortunately, we could not perform studies to detect them in the glomerular immune deposits. Although several studies, including relatively large numbers of patients with idiopathic membranous nephropathy, have failed to demonstrate their presence at the glomerular level, there are weC-described reports supporting the role of tubular antigens in the
298
immunopathogenesis of glomerulonephritis and tubulointerstitial nephritis [9-14,28]. Although the divergent data on detection of brush-border antigens in the immune deposits could be caused by the use of polyspecific antibodies, the existence of circulating antibodies reacting in vitro with kidney brush border is well documented [15,16]. Since these antibodies have also been reported in two transplanted patients without evidence of immune complex glomerulonephritis [29], it is possible that they represent an epiphenomenon. However, in one patient with recurrent membranous nephropathy anti-rat brush-border antibodies were detected and in another patient the presence of anti-brush-border antibodies before transplantation was followed by a drop in their titre after the graft placement [15]. The event initiating the production of these antibodies in our patient is unknown. The association between the development of proteinuria and the existence of reflux nephropathy, a situation in which an increase in the pelvic pressure exists, is well documented. Furthermore, in a recent study the incidence of postransplant membranous nephropathy was greatly increased in patients with ureteral obstruction [30]. However, although the anecdotal association of membranous nephropathy has been described [31], focal and segmental glomerulosclerosis accounts for most of the cases of reflux-associated glomerulopathy [32,33]. The precise mechanism for the development of glomerulosclerosis is still unclear, although the most attractive explanation is that it results from the adaptative changes occurring in glomeruli resulting from reduction in renal mass [34]. Several authors have failed to localize autologous antigens, such as brush-border antigens or TammHorsfall protein, in the mesangium of patients with focal sclerosis and reflux nephropathy [35]. To our knowledge this is the first report suggesting an association between membranous nephropathy and hydronephrosis due to bilateral pyeloureteral stenosis. Since antibodies reacting with human kidney brush-border have been detected at very low titre in some 'normal' human sera [36], as well as in sera from patients with various autoimmune disorders [37], it is possible that some genetic predisposition in our patient could have resulted in an excessive response to the presumed release of tubular antigens due to the increased pelvic pressure, facilitating the development of membranous nephropathy. The stability of the renal function could be regarded as evidence that a major elevation of intratubular pressure was not present. However patients with pyeloureteral stenosis experience episodes of elevated intrapelvic pressure related to increases in diuresis, manifested clinically by renal colicky pain. In summary, this report describes a new case of
J. Gonzalez-Cabrero et ai
membranous nephropathy associated with the presence of circulating anti-brush-border antibodies in a patient with bilateral pyeloureteral stenosis. However, the inability to detect the antibodies in the sera of a group of patients with idiopathic membranous nephropathy suggests, in agreement with previous authors, that renal tubular antigens are rarely involved in the pathogenesis of this disease. Acknowledgements. The study was supported in part by grants from Fondo de Investigaciones Sanitarias de la Seguridad Social (FISss 91/0162), Ministerio de Educacion y Ciencia (PM 89/0065), and Fundacion Inigo Alvarez de Toledo. Dr J. Gonzalez Cabrero is a fellow of Fundaci6n Conchita Rabago de Jimenez Diaz. We thank Dr A Barat for studying the renal biopsy and Ms L. Gulliksen for typing the manuscript.
References 1. Heymann W, Hackel DB, Harwood J, Wilson SGF, Hunter JLP. Production of the nephritic syndrome in rats by Freund's adjuvant and rat kidney suspensions. Proc Soc Exp Biol Med 1959; 100: 660-664 2. Edgington TS, Glassock RJ, Dixon FJ. Autologous immune complex nephritis induced with renal tubular antigen. 1. Identification and isolation of the pathogenetic antigen. J Exp Med 1968; 127: 555-571 3. Barabas AZ, Lanningan R. Induction of autologous immune complex glomerulonephritis in the rat by intravenous injection of heterologous anti-rat kidney tubular antibody. I. Production of chronic progressive immune complex glomerulonephritis. Br J Exp Palhol 1974; 55: 47-55 4. Kerjaschki D, Farquhar MG. The pathogenic antigen of Heymann nephritis is a membrane glycoprotein of the renal proximal tubule brush border. Proc Nail Acad Sci USA 1982; 79: 5557-5561 5. Naruse T, Kitamura K, Miyakawa Y, Shibata S. Deposition of renal tubular epithelial antigen along the glomerular capillary walls of patients with membranous glomerulonephritis. J Immunol 1973; 110: 1163-1166 6 Whitworth JA, Leibowitz S, Kennedy MC el al. Absence of glomerular renal tubular epithelial antigen in membranous glomerulonephritis. Clin Nephrol 1976; 5: 159-162 7. Zager RA, Couser WG, Andrews BS, Bolton WK, Pohl MA. Membranous nephropathy. a radioimmunologic search for anti-renal tubular epithelial antibodies and circulating immune complexes. Nephron 1979; 24: 10-16 8. Collins AB, Andres GA, McCluskey RT. Lack of evidence for a role of renal tubular antigen in human membranous glomerulonephritis. Nephron 1981; 27: 297-301 9. Ozawa T, Boedecker EA, Schon W, Guggenheim S, Mclntosh RM. Immune complex disease with unilateral renal vein thrombosis. Arch Palhol Lab Med 1976; 100: 279-282 10. Strauss J, Pardo V, Koss MN, Griswold W, Mclntosh RM. Nephropathy associated with sickle cell anemia. An autologous immune complex nephritis. I. Studies on nature of glomerularbound antibody and antigen identification in a patient with sickle cell disease and immune deposit glomerulonephntis. Am J Med 1975, 58: 382 11. Pardo V, Strauss J, Kramer H, Ozawa T, Mclntosh RM. Nephropathy associated with sickle cell anemia. An autologous immune complex nephritis. II. Clinicopathologic study of seven patients. Am J Med 1975; 59: 651-659 12. Ozawa T, Pluss R, Lacher J et at. Endogenous immune complex nephropathy associated with malignancy. I. Studies on the nature and immunopathogenic significance of glomerular bound antigen and antibody. Isolation and characterization of tumor specific antigen and antibody and circulating immune complex. Q J Med 1975; 44: 523-541 13. Shwayder M, Ozawa T, Boedecker E, Guggenhaim S, Mcln-
299
Anti-FxlA antibodies in membranous nephropathy
14.
15.
16. 17. 18.
19. 20.
21. 22. 23.
24.
tosh RM. Nephrotic syridrome associated with Fanconi syndrome. Immunopathogemc studies of tubulo-interstitial nephritis with autologous immune complex glomerulonephritis. Ann Intern Med 1976; 84: 433-437 Gilboa N. Durante D, Mclntosh RM. Glomerular deposition of renal tubular epithelial antigen in patients with systemic lupus erythematosus. Its possible role in lupus nephritis. J Rlteumatol 1977; 4: 358-364 Zanetti M, Mandet C, Duboust A, Bedrossian J, Bariety J. Demonstration of a passive Heymann nephritis-like mechanism in a human kidney transplant. Clin S'ephrol 1981; 15: 272-277 Douglas MFS, Rabideau DP, Schwartz MM, Lewis EJ. Evidence of autologous immune complex nephritis. A' Engl J Med 1961; 305: 1326-1329 Egido J, Sancho J, Rivera F, Hernando L. The role of IgA and IgG immune complexes in IgA nephropathy. Nephron 1984; 36: 52-59 Egido J, Alonso F, Sanchez Crespo M, Barat A, Hernando L. Absence of an anaphylactic vasopermeability mechanism for immune complex deposition in the Heymann nephritis of rats. Clin Exp Immunol 1980; 42: 99-106 Zamlanski-Tucker MJ, Van Lien JB, Noble B. The pathophysiology of the kidney in rats with Heymann nephritis. Kidney Int 1985; 28: 504-512 Cambiaso CL, Goffinet A, Vaerman JP, Heremans JF. Glutaraldehyde-activated aminohexyl-derivative of Sepharose 4B as a new versatile immunoabsorbent. Immunochemistry 1975; 12: 273-278 Weber K, Osborn M. The reliability of molecular weight determinations by docecyl sulfate-polyacrylamide gel electrophoresis. J Biol Chem 1969; 244: 4406-4412 Hudson L, Hay FC. Practical Immunology. Oxford, England, Blackwell, 1989 Kerjaschki D, Farquhar MG. Immunocytochemical localization of the Heymann nephritis antigen (gp 330) in glomerular epithelial cells of normal Lewis rats. J Exp Med 1983; 157: 667-686 Kamata K, Baird LG, Erikson ME, Collins AB, McCluskey RT. Characterization of antigen and antibody specificities involved in Heymann nephritis. J Immunol 1985; 135: 24002409
25. Ronco P, Neale TJ, Wilson CB, Galceran M, Verroust P. An immunopathologic study of a 330-kD protein defined by monoclonal antibodies and reactive with anti-RTE 5 antibodies and kidney eluates from active Heymann nephritis. J Immunol 1986; 136: 125-130 26. Makker SP, Singh AK. Characterization of the antigen (gp 600) of Heymann nephritis. Lab Invest 1984; 50: 287-293 27. Kerjaschki A, Horvat R, Binder S et al. Identification of a 400 kd protein in the brush borders in human kidney tubules that is similar to gp 330. the nephritogenic antigen of rat Heymann nephritis. Am J Pathol 1987; 129: 183-191 28. Niles J, Collins B, Baird L et al. Antibodies reactive with a renal glycoprotein and with deposits in membranous nephritis. Kidney Int 1987; 31: 338 29. Paul LC, Stuffers-Heiman M, Van Es LA, de Graff J. Antibodies directed against brush border antigens of proximal tubules in renal allograft recipients. Clin Immunol Immunopathol 1979; 14: 238-243 30. Hoitsma AJ, Kroon AA, Wetzels JFM, Assman FJM, Berden JHM, Koene RAP. Association between ureteral obstruction and de novo membranous nephropathy in renal allografts. Transplant Proc 1990; 22: 1388-1389 31. Woods HF, Walls J. Nephrotic syndrome in veskoureteric reflux. Br Med J 1976; 2: 917-918 32. Zimmerman SW, Uehling DT, Burkholder PM. Vesicoureteral reflux nephropathy: Evidence for immunologically mediated glomerular injury. Urology 1973; 2: 531-538 33. Cotran RS. Glomerulosclerosis in reflux nephropathy. Kidney Int 1983; 21: 528-534 34. Hostetter RH, Olson JL, Rennke HG, Venkatachalan MA, Brenner BM. Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 1981; 241: F85-F93 35. Bhathena DB, Weiss JH, Holland NH et al. Focal and segmental glomerular sclerosis in reflux nephropathy (chronic pyelonephritis) Am J Med 1980; 68: 886-892 36. Makker SP. Brush border antibodies of Heymann nephritis of rats in normal human serum. Proc Soc Exp Biol Med 1977; 154: 9-13 37. Ireton HJC, Muller HK, McGiven AR. Human antibody against gastric parietal cell and kidney brush border. Clin Exp Immunol 1971; 8: 785-791 Received for publication 5.3.91 Accepted in revised form 16.8.91