Mesangial IgA2 deposits and lectin pathway-mediated complement activation in IgA glomerulonephritis

Mesangial IgA2 Deposits and Lectin Pathway–Mediated Complement Activation in IgA Glomerulonephritis Satoshi Hisano, MD, Misao Matsushita, PhD, Teizo Fujita, MD, Yuzo Endo, MD, and Shigeo Takebayashi, MD ● Three pathways are recognized in complement activation. The aim of our study is to elucidate immunohistologically which complement pathway is associated with complement activation in immunoglobulin A (IgA) glomerulonephritis (GN) and which IgA subclass is related to complement activation. Immunohistological staining was performed on renal biopsy specimens obtained from 36 patients with IgA GN, 10 patients with systemic lupus erythematosus (SLE), and 16 patients with other GNs using polyclonal antibodies of IgG, IgA, IgM, C1q, C3c, and C4 and monoclonal antibodies of IgA1, IgA2, mannose-binding lectin (MBL), and MBL-associated serine protease-1 (MASP-1). Mesangial deposits of both IgA1 and IgA2 were found in 19 of 36 patients with IgA GN. Mesangial deposits of C3c, C4, MBL, and MASP-1 also were detected in these 19 patients, and IgA2, MBL, and MASP-1 deposits were colocalized in the mesangium in these patients. The remaining 17 patients showed mesangial deposits of IgA1 alone. Twelve of these 17 patients showed mesangial deposits of C3c without C4, MBL, or MASP-1. No deposition of C1q was evident in patients with IgA GN. Three of 10 patients with SLE showed glomerular deposition of MBL and MASP-1 without glomerular deposition of IgA2. None of the patients with other GNs showed glomerular deposition of IgA1, IgA2, MBL, or MASP-1. There was no correlation in clinical or pathological indicators between IgA2-positive and IgA2-negative patients with IgA GN. In conclusion, alternative pathway–involved complement activation is associated with mesangial deposits of IgA1 alone in patients with IgA GN. In those with mesangial deposits of both IgA1 and IgA2, both the alternative and lectin pathways are involved in complement activation. We first report that mesangial deposits of IgA2 may activate the lectin pathway in patients with IgA GN. © 2001 by the National Kidney Foundation, Inc. INDEX WORDS: Immunoglobulin A glomerulonephritis (IgA GN); immunoglobulin A1 (IgA1); immunoglobulin A2 (IgA2); mannose-binding lectin (MBL); mannose-binding lectin (MBL)-associated serine protease-1 (MASP-1); lectin pathway.

I

MMUNOGLOBULIN A (IgA) glomerulonephritis (GN) is characterized by predominant IgA deposition and simultaneous C3 deposition in the mesangium on immunofluorescence microscopy (IF).1 Human IgA consists of two subclasses: IgA1 and IgA2.2 In 1980, Andre´ et al3 described that the major constituent of glomerular IgA deposits in IgA GN, Scho¨nlein-Henoch purpura nephritis (SHPN), and alcoholic cirrhosis was IgA2. However, it was reported that in IgA GN, IgA1 was mainly deposited in the mesangium,4 and IgA1containing immune aggregates had an important role in the pathogenesis of IgA GN.5 From the Second Department of Pathology, Fukuoka University, School of Medicine, Fukuoka; Department of Biochemistry, Fukushima Medical University, Fukushima; and The Research of Pathology, The Institute of Medical Science, The University of Tokyo, Japan. Received October 27, 2000; accepted in revised form June 8, 2001. Address reprint requests to Satoshi Hisano, MD, The Second Department of Pathology, School of Medicine, Fukuoka University, Nanakuma 7-45-1, Jonan-ku, Fukuoka 814-0180, Japan. © 2001 by the National Kidney Foundation, Inc. 0272-6386/01/3805-0021$35.00/0 doi:10.1053/ajkd.2001.28611 1082

Three pathways are recognized in complement activation.6 Because of C3 deposits along the capillary wall and/or in the mesangium, the alternative pathway was reported to be activated in IgA GN.7 Recently, a new pathway of complement activation, the lectin pathway, which is initiated by mannose-binding lectin (MBL), MBL-associated serine protease-1 (MASP-1), MASP-2, and a truncated form of MASP-2, was reported.8-11 Endo et al12 and Matsuda et al13 reported that lectin pathway–mediated complement activation was shown immunohistologically in some patients with IgA GN. Their results suggest two different pathways in complement activation in patients with IgAGN. The aim of our current study is to elucidate immunohistologically which complement pathway is associated with complement activation in IgA GN and which IgA subclass is related to complement activation. PATIENTS AND METHODS

Patients Kidney tissues were obtained by percutaneous renal biopsy from 62 patients visiting our affiliated hospitals who were aged 4 to 75 years and had urinary abnormalities: 36 patients, IgA GN; 10 patients, systemic lupus erythematosus

American Journal of Kidney Diseases, Vol 38, No 5 (November), 2001: pp 1082-1088

IgA2 SUBCLASS AND LECTIN PATHWAY

(SLE); 5 patients, minimal change nephrotic syndrome; 6 patients, membranous GN; and 5 patients, thin basement membrane disease. Informed consent was obtained from patients and/or their parents before a renal biopsy. IgA GN was diagnosed by the predominant deposition of IgA in the mesangium on IF, as well as lack of clinical and pathological evidence of SHPN, SLE, or alcoholic cirrhosis–associated GN.14 SLE was diagnosed on the basis of the 1982 revised criteria established by the American Rheumatism Association.15 Light microscopy was evaluated on sections stained with periodic acid–Schiff and periodic acid–methenamine silver. In IgA GN, histological severity of glomerular lesions was classified into three groups according to our previous report,16 modified by us as follows: mild was considered minor glomerular abnormalities or mild mesangial hypercellularity of at least three cells in the peripheral mesangial area in most glomeruli; moderate, moderate mesangial proliferation with less than 30% of glomeruli showing glomerular sclerosis; and severe, moderate mesangial proliferation with 30% or more of glomeruli showing glomerular sclerosis.

Immunohistological Staining Tissues embedded in ornithine carbamoyltransferase compound (Miles Laboratories Inc, Elkhart, IN) were frozen in acetone dry ice. Fluorescein isothiocyanate (FITC)-labeled anti-IgG, IgA, IgM, C1q, C3c polyclonal antibodies (Dako, Copenhagen, Denmark), and FITC-labeled anti-C4 polyclonal antibody (MBL, Nagoya, Japan) were used. Monoclonal antibodies against human IgA1 (NIF2; Skybio, Bedfordshire, UK; and M4D8; MBL) and IgA2 (2E2; Skybio; and 14-3-26; Becton Dickinson, San Jose, CA) were used. Monoclonal antibodies against human MBL (3E7) and MASP-1 (4C2) were provided by Dr Fujita, coinvestigator, as previously reported.17,18 Sections cut at a 4-␮m thickness from frozen tissues were immersed in Tris-buffered saline (TBS) three times for 5 minutes. Diluted FITC-labeled polyclonal antibodies (IgG, IgA, IgM, C3c, C1q, and C4), IgA1 and IgA2 monoclonal antibodies, MBL monoclonal antibody, and MASP-1 monoclonal antibody were overlaid on the sections, which were incubated at 4°C overnight. Sections on which FITC-labeled polyclonal antibodies were overlaid were observed on IF. Immunohistochemical studies for monoclonal antibodies (IgA1, IgA2, MBL, and MASP-1) were performed using the avidin-biotin peroxidase method, as previously reported.19 After treatment in 0.5% hydrogen peroxide in methanol solution to quench endogenous peroxidase activity, sections were incubated with biotin-labeled antibody against mouse IgG (Dako) at room temperature for 30 minutes. After washing in TBS, sections were incubated with avidin-biotinperoxidase complex (Dako) at room temperature for 30 minutes. After three washes in TBS, sections were incubated with peroxidase-labeled avidin for 30 minutes and treated with 3,3-diaminobenzidine (Dohjin, Kyoto, Japan). Sections were counterstained with hematoxylin and observed by light microscope. The intensity of immunohistological deposits in glomerulus was based semiquantitatively on a scale of 0 to 3⫹, as follows: 0, none; 1⫹, mild; 2⫹, moderate; and 3⫹, intense. Immunohistochemical specificity was confirmed by an ab-

1083

sorption test with antibodies and replacement of primary antibodies to nonimmunized sera. As a negative staining control, nonimmune normal serum or TBS was substituted for primary antibodies according to a previous report.20

Statistical Analysis Data are expressed as mean ⫾ SD. Chi-square test and t-test were used to analyze data.

RESULTS

Glomerular Deposition of Gamma Globulins, Complement Components, MBL, and MASP-1 in Various GNs Specificity of IgA1 and IgA2 monoclonal antibodies was confirmed by immunoabsorption test. In addition, correct identification of IgA2 without cross-reacting with IgA1 was confirmed by the immunoabsorption method using normal serum IgA1 treatment. These studies were performed in serial sections of the same cases. Specificity of IgA1 and IgA2 was confirmed. Specificity of IgA2 and correct identification of IgA2 without cross-reacting with IgA1 are shown in Fig 1A, B, and C. The same distribution of immunostained glomerular deposition was observed in two different epitopes of IgA1 (NIF2 and M4D8) and two different epitopes of IgA2 (2E2 and 14-3-26). The number of cases with glomerular deposition of gamma globulins, complements, MBL, and MASP-1 in various GNs is listed in Table 1. IgA1 was detected in all 36 patients with IgA GN, and IgA2, in 19 patients. Three of 10 patients with SLE had glomerular deposition of MBL and MASP-1, but none of these patients showed glomerular deposition of IgA2. Glomerular deposition of IgA2, MBL, or MASP-1 was not found in patients with other GNs. In 19 of 36 patients with IgA GN, mesangial deposits of both IgA1 and IgA2 were evident, whereas the remaining 17 patients showed mesangial deposits of IgA1 alone (Table 2). Remarkable and diffuse mesangial deposits of IgA1 were seen in all 36 patients with IgA GN. Diffuse and strong mesangial deposits of IgA2 similar to IgA1 deposits was evident in 8 of 19 patients (42.1%) with both IgA1 and IgA2 deposits, shown in Fig 1D and E, whereas segmental and weak mesangial deposits of IgA2 were observed in the remaining 11 patients (Fig 1G and H). The 19 patients with both IgA1 and IgA2 deposits had

1084

HISANO ET AL

Fig 1. (A) IgA2 deposits, (B) no deposit of IgA2 after normal serum IgA2 treatment by the immunoabsorption method, and (C) IgA2 deposits after normal serum IgA1 treatment by the immunoabsorption method. Diffuse and strong mesangial deposits of (D) IgA1, (E) IgA2, and (F) MBL/MASP-1 are seen in a case of IgA GN. Compared with diffuse and strong mesangial deposits of (G) IgA1, (H) IgA2 and (I) MBL/MASP-1 are segmentally and weakly deposited in the mesangium in a case of IgA GN. (J) C4 deposits in the mesangium and along the capillary wall. Glomerular deposition of neither (K) C4 nor (L) MBL/MASP-1 in a case of IgA GN. Immunohistological staining shown in (A-C), (D-F), and (G-I) was performed in serial sections of the same cases. (Original magnification: [A-C, J-L] ⴛ400, [D-I] ⴛ100.)

IgA2 SUBCLASS AND LECTIN PATHWAY Table 1.

1085

Glomerular Deposition of Gamma Globulins, Complement Components, MBL, and MASP-1 in Various GNs No. of Cases

IgG

IgA

IgA1

IgA2

IgM

C1q

C4

C3c

MBL

MASP-1

36 10

11 10

36 10

36 10

19 0

10 6

0 10

19 10

31 10

19 3

19 3

5 6

0 6

0 0

0 0

0 0

0 0

0 0

0 0

0 4

0 0

0 0

5

0

0

0

0

0

0

0

0

0

0

IgA GN Lupus nephritis Minimal change nephrotic syndrome Membranous GN Thin basement membrane disease

mesangial deposits of C4 (Fig 1J), MBL, and MASP-1 (Table 2). Mesangial deposits of MBL and MASP-1 showed similar distribution. IgA2 and MBL/MASP-1 were colocalized in the mesangium (Fig 1E, F, H, and I). Glomerular deposition of C4 and MBL/MASP-1 was not evident in 17 patients with mesangial deposits of IgA1 alone (Fig 1K and L). Comparison of Clinical and Laboratory Findings Between IgA2-Positive and IgG2-Negative Patients With IgA GN Clinical and laboratory findings are compared between patients with and without IgA2 deposits in IgA GN, listed in Table 3. There was no significant difference in clinical findings, except for macrohematuria and the severity of glomerular changes between patients with and without IgA2 deposits. DISCUSSION

On our study, mesangial deposits of IgA1 were evident in all 36 patients with IgA GN. C3c deposits were found in 31 patients, and C1q deposits were not found in any patient. In 17 of 36 patients, mesangial deposits of IgA1 alone were evident. The remaining 19 patients showed mesangial deposits of both IgA1 and IgA2, as Table 2. Characteristics of Glomerular Deposition of Complement Components, MBL, and MASP-1 in Patients With and Without IgA2 Deposition in IgA GN

IgA1 ⫹ IgA2 IgA1 alone

No. of Cases

C3c

C4

MBL

MASP-1

19 17

19 12

19 0

19 0

19 0

NOTE. The number 19 in all columns represents the same 19 patients.

well as MBL/MASP-1. Complement activation through the classical pathway is not evident in our patients with IgA GN. Complement activation through the alternative pathway is considered to be associated with mesangial deposits of IgA1 in IgA GN. In addition to C3c deposits, mesangial deposits of C4 were found in only 19 patients with mesangial deposits of both IgA1 and IgA2. In these patients, a complement pathway other than the alternative pathway is activated. The lectin pathway is initiated by MBL, which has a high affinity to terminal mannose and N-acetyl glucosamine moieties on surfaces of various pathogens, including bacteria, yeast, fungi, and viruses.8-11,21 Human MBL forms complexes with two types of serine protease, MASP-1 and MASP-2.8,9,22 Binding to these sugar chains leads to activation of C4 and C2 without C1 component.22 IgA2 and MBL/MASP-1 were colocalized in the mesangium in these 19 patients. Therefore, complement activation through the lectin pathway may be related to mesangial deposits of IgA2 in IgA GN. In patients with both IgA1 and IgA2 deposits, complement activation is considered to be mediated by both the alternative and lectin pathways. However, Endo et al12 described that C3c was positive in 11 patients with MBL/MASP-1 deposition in IgA GN, but C4c was positive in only 4 of 11 patients. We used anti-C4 antibody in our immunohistological study. This antibody possesses immunologic activity against entire chains of C4a, C4b, C4c, and C4d (MBL product information). Therefore, it is considered that this antibody reacted with C4 component other than C4c. Matsuda et al13 reported that IgA1 and MBL

1086

HISANO ET AL Table 3.

Clinical and Laboratory Findings in Patients With and Without IgA2 Deposits in IgA GN

Age at onset (y) Time from onset to biopsy (mon) Initial presentation (no. of cases) Asymptomatic hematuria and proteinuria Macrohematuria Nephrotic syndrome Upper respiratory tract infection before biopsy (no. of cases) Proteinuria at biopsy (mg/dL) Serum concentration Creatinine (mg/dL) IgA (mg/dL) C3 (mg/dL) C4 (mg/dL) Severity of glomerular abnormalities (no. of cases) Mild Moderate Severe

were colocalized in the mesangium in 7 of 42 patients with IgA GN. In our 36 patients with IgA GN, IgA1 was deposited diffusely in the mesangium. In 8 of 19 patients with both IgA1 and IgA2 deposits, IgA2 and MBL/MASP-1 deposits were diffusely colocalized in the mesangium. In 11 patients who showed segmental IgA2 deposits, MBL/MASP-1 and IgA2 deposits were segmentally colocalized in the mesangium. Therefore, our study shows that IgA2 and MBL/ MASP-1 deposits were colocalized in the mesangium in patients with IgA GN. Because mesangial deposits of IgA1/IgA2 and MBL/MASP-1 are colocalized in a diffuse distribution pattern in approximately 40% of patients with both IgA1 and IgA2 deposits, IgA1 and MBL/MASP-1 are determined to be colocalized in the mesangium when IgA2 deposits are not evaluated, as reported by Matsuda at al.13 In our study, MBL and MASP-1 deposits were evident in 3 patients with SLE, but IgA2 deposits were not found. In our patients with SLE, complement activation is through the classical pathway because C1q, C3c, and C4 were deposited in the glomerulus. Why MBL and MASP-1 deposits were present in the 3 patients with SLE is unknown. Lhotta et al23 reported previously that MBLassociated GN was not IgA GN, but lupus GN, membranous GN, and membranoproliferative GN, in contrast to our results and those of Endo et al12 and Matsuda et al.13 MBL and MASP-1 monoclonal antibodies used on our study and

IgA1 ⫹ IgA2 (n ⫽ 19)

IgA1 Alone (n ⫽ 17)

33.6 ⫾ 20.5 40.5 ⫾ 39.0

31.0 ⫾ 23.6 34.2 ⫾ 45.1

14 4 1 4 125.1 ⫾ 130.0

12 0 5 4 470.1 ⫾ 874.2

0.9 ⫾ 0.5 378.4 ⫾ 163.8 99.5 ⫾ 27.2 30.8 ⫾ 10.3

0.9 ⫾ 0.6 378.8 ⫾ 192.0 83.7 ⫾ 22.1 26.6 ⫾ 7.6

5 9 5

P

⬍0.05

7 6 4

that of Endo et al12 were provided by Dr Fujita, our coinvestigator. Therefore, the discrepancy between our results and those of Lhotta et al23 is caused by the difference in MBL/MASP-1 monoclonal antibodies. However, the prevalence of patients with MBL/MASP-1 deposition was greater than that of Endo et al12 despite using the same MBL/MASP-1 monoclonal antibody. The difference in prevalence observed between the studies may reflect the difference in the patients investigated. Endo et al12 and Matsuda et al13 reported a contradictory clinicopathological correlation between MBL-positive and MBL-negative patients. Based on our results, there was no correlation in clinical or pathological severity between MBL/MASP-1–positive and MBL/ MASP-1–negative patients with IgA GN. Therefore, glomerular deposition of MBL or MASP-1 is not a prognostic indicator in IgA GN. How are mesangial deposits of IgA2-MBL/ MASP-1 formed? In some patients with IgA GN, macrohematuria often presents after viral or bacterial infection in the upper respiratory tract, including tonsils.13 On our study, macrohematuria after upper respiratory tract infection was present in 4 of 19 patients with mesangial deposits of IgA2 and MBL/MASP-1, but the pathogen of upper respiratory tract infection before renal biopsy was unknown. Suzuki et al24 and Endo25 reported that certain bacterium might be the etiologic candidate in IgA GN. After infection by certain bacterium or a virus, MBL/MASP-1 may

IgA2 SUBCLASS AND LECTIN PATHWAY

bind to mannose or N-acetyl glucosamine of the bacterium or virus.21 IgA likely is produced preferentially by carbohydrate antigens.26 Therefore, it is potentially possible that IgA2 production is enhanced by carbohydrates of some bacteria or viruses. Wold et al27 reported that IgA possessed carbohydrate receptors for mannose-specific lectin. Endo et al28 further described small amounts of a high mannose-type carbohydrate chain found in both IgA1 and IgA2 myeloma proteins. Therefore, increased IgA2 after infection may bind to MBL/MASP-1 and form an IgA2-MBL/MASP-1 complex. Increased IgA1 in IgA GN may form a polymeric complex of IgA1/IgA2-MBL/MASP-1 in the circulation, and these macromolecules may be trapped in the mesangium. The second possibility is that the IgA2-MBL/MASP-1 complex produced in the circulation after infection may be trapped in the mesangium, where IgA1 has already deposited. The lectin pathway is considered to be activated in the mesangium. From the results of our study, mesangial deposits of IgA1 were evident in the 36 patients with IgA GN, and there was no correlation in clinical or pathological indicators between IgA2-positive and IgA2-negative patients with IgA GN. Therefore, IgA1-containing immune complexes and IgA1associated complement activation through the alternative pathway are considered to have a major pathogenic role in IgAGN, rather than IgA2associated immune complexes and IgA2-associated complement activation through the lectin pathway. When other factors, such as bacterial or viral infection, are involved in patients with IgAGN, IgA2 and MBL/MASP-1 may be activated. REFERENCES 1. Berger J: IgA glomerular deposits in renal disease. Transplant Proc 1:939-944, 1969 2. Williams GD: Pathogenesis of idiopathic IgA nephropathy. Pediatr Nephrol 7:303-311, 1993 3. Andre´ C, Berthoux FC, Andre´ F, Gillon J, Genin C, Sabatier J-C: Prevalence of IgA2 deposits in IgA nephropathies. A clue to their pathogenesis. N Engl J Med 303:13431346, 1980 4. Conley ME, Cooper MD, Michael AF: Selective deposition of immunoglobulin A1 in immunoglobulin A nephropathy, anaphylactoid purpura nephritis, and systemic lupus erythematosus. J Clin Invest 66:1432-1436, 1980 5. Czerkinsky C, Koopman WJ, Jackson S, Collins JE, Crago SS, Schroehenloher RE, Julian BA, Galla JH, Mestecky J: Circulating immune complexes and immunoglobulin A rheumatoid factor in patients with mesangial

1087

immunoglobulin nephropathies. J Clin Invest 77:1931-1938, 1986 6. Del Zoppo GJ: In stroke, complement will get you nowhere. Nat Med 5:995-996, 1999 7. Emancipator SN, Lamm ME: IgA nephropathy: Pathogenesis of the most common form of glomerulonephritis. Lab Invest 60:168-183, 1989 8. Matsushita M, Fujita T: Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease. J Exp Med 176:1492-1502,1992 9. Thiel S, Vorup-Jensen T, Stover CM, Schwaeble W, Laursen SB, Poulsen K, Willis AC, Eggleton P, Hansen S, Holmskov U, Reid KB, Jensenius JC: A second serine protease associated with mannose-binding lectin that activates complement. Nature 386:506-510, 1997 10. Takahashi M, Endo Y, Fujita T, Matsushita M: A truncated form of mannose-binding lectin-associated serine protease (MASP)-2 expressed by alternative polyadenylation is a component of the lectin complement pathway. Int Immunol 11:859-863, 1999 11. Stover CM, Thiel S, Thelen M, Lynch NJ, VorupJensen T, Jensenius JC, Schwaeble WJ: Two constituents of the initiation complex of the mannan-binding lectin activation pathway of complement are encoded by a single structural gene. J Immunol 162:3481-3490, 1999 12. Endo M, Ohi H, Ohsawa I, Fujita T, Matsushita M, Fujita T: Glomerular deposition of mannose-binding lectin (MBL) indicates a novel mechanism of complement activation in IgA nephropathy. Nephrol Dial Transplant 13:19841990, 1998 13. Matsuda M, Shikata K, Wada J, Sugimoto H, Shikata Y, Kawasaki T, Makino H: Deposition of mannan-binding protein and mannan-binding protein-mediated complement activation in the glomeruli of patients with IgA nephropathy. Nephron 80:408-413, 1998 14. Kusumoto Y, Takebayashi S, Taguchi T, Harada T, Naito S: Long-term prognosis and prognostic indices of IgA nephropathy in juvenile and adult Japanese. Clin Nephrol 28:118-124, 1987 15. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, Schaller JG, Talal N, Winchester RJ: The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 25:1271-1277, 1982 16. Takebayashi S, Yanase K: Asymptomatic urinary abnormalities found via the Japanese school screening program: A clinical, morphological and prognostic analysis. Nephron 61:82-88, 1992 17. Matsushita M, Takahashi A, Hatsuse H, Kawakami M, Fujita T: Human mannose-binding protein is identical to a component of Ra-reactive factor. Biochem Biophys Res Commun 183:645-651, 1992 18. Terai I, Kobayashi K, Matsushita M, Fujita K: ␣2Macroglobulin binds to and inhibits mannose-binding protein-associated serine protease. Int Immunol 7:1579-1584, 1995 19. To´th T, To´th-Jakatics R, Jimi S, Ihara M, Urata H, Takebayashi S: Mast cells in rapidly progressive glomerulonephritis. J Am Soc Nephrol 10:1498-1505, 1999 20. Nagata M, Akioka Y, Tsunoda Y, Komatsu Y, Kawagu-

1088

chi H, Yamaguchi Y, Ito K: Macrophage in childhood IgA nephropathy. Kidney Int 48:527-535, 1995 21. Turner MW: Mannose-binding lectin: The pluripotent molecule of the innate immune system. Immunol Today 17:532-540, 1996 22. Matsushita M, Endoh Y, Fujita T: MASP1 (MBLassociated serine protease 1). Immunobiology 199:340-347, 1998 23. Lhotta K, Wu¨rzner R, Ko¨nig P: Glomerular deposition of mannose-binding lectin in human glomerulonephritis. Nephrol Dial Transplant 14:881-886, 1999 24. Suzuki S, Nakatomi Y, Sato H, Tsukada H, Arakawa M: Haemophilus parainfluenzae antigen and antibody in renal biopsy samples and serum of patients with IgA nephropathy. Lancet 343:12-16, 1994

HISANO ET AL

25. Endo Y: IgA nephropathy—Human disease and animal model. Ren Fail 19:347-371, 1997 26. Russell MW, Lue C, Van Den Wall Bakke AWL, Moldoveanu Z, Mestecky J: Molecular heterogeneity of human IgA antibodies during immune response. Clin Exp Immunol 87:1-6, 1992 27. Wold AE, Mestecky J, Tomana M, Kobata A, Ohbayashi H, Endo T, Svanborg-Eden C: Secretary immunoglobulin A carries oligosaccharide receptors for Escherichia coli type 1 fimbrial lectin. Infect Immun 58:30733077, 1990 28. Endo T, Mestecky J, Kulhavy R, Kobata A: Carbohydrate heterogeneity of human myeloma proteins of the IgA1 and IgA2 subclass. Mol Immunol 31:1415-1422, 1994