Polymorphs infiltrate glomeruli in mesangial IgA glomerulonephritis

Kidney International, Vol. 36 (1989), pp. 1108—1111

Polymorphs infiltrate glomeruli in mesangial IgA glomerulonephritis PRISCILLA KINCAID-SMITH, KATHY NICHOLLS, and IAN BIRCHALL Department of Medicine, University of Melbourne, and Department of Nephrology, The Royal Melbourne Hospital, Victoria, Australia

to glomerular damage. In biopsies taken during episodes of macroscopic hematuria in addition to fibrin and crescent formation [4], polymorphonu-

Polymorphs infiltrate glomeruli in mesanglal IgA glomerulonephritis. During episodes of macroscopic hematuria, patients with IgA nephrop— athy commonly have polymorphonuclear neutrophils (PMNs) as well as fibrin and mononuclear cells in glomerular capillaries. We quantitated PMN and macrophage infiltration in glomeruli of 54 patients with IgA

suggesting that auto-immunity contributes

nephropathy in whom renal biopsies were obtained within 30 days of macroscopic hematuria. Control biopsies (N = 22) were from patients with IgA nephropathy and urinary erythrocyte counts below 50,000/mi. PMN monocyte/macrophages were quantitated using the monoclonal antibodies FMC1O and HAMS6, respectively. In patients with heavy 3.4% (mean SE) of glomeruli were positive for hematuria, 45.9 PMNs (control 10.5 2.8%, P 0.001) with a mean PMN count! glomerulus of 1.10 0.20 (control 0.13 0.03, 0.001). 65.6

clear lesicocytes and macrophages are frequently observed in glomerular capillaries. In this study we compare polymorphonuclear leucocyte and macrophage numbers in biopsies taken within 30 days of an episode of macroscopic hematuria with those seen in biopsies taken when the urinary erythrocyte count

9.7%

P<

is below 50,000.

Of the glomeruli were positive for monocytes in the heavy

hematuria group (control 40.0 8.4, P < 0.05) with the mean monocyte count per glomerulus being 1.6 0.2 (control 0.6 0.1, P <0.01). We conclude that, in the acute phase of mesangial IgA nephropathy, PMN

and monocytes are present and presumably participate in glomerular injury.

Methods Between 1979 and 1988, we biopsied 54 patients within 30 days of an episode of macroscopic hematuria. In most patients (45

of 54) urinary erythrocyte counts were still above 106/ml at

the

time of biopsy. None were below l05/ml, and the median

count was

106/mi.

As a control group of patients, we selected from our records

Mesangial IgA glomerulonephritis is the most frequent form of glomerulonephritis in most countries [11 and is an important cause of end-stage renal failure [1—3]. Episodes of macroscopic hematuria are a typical feature of this condition especially in children and young males [2, 31. We have documented that crescents are present on biopsy during episodes of macroscopic hematuria [41, and many authors have noted that the presence of crescents on biopsy is an important prognostic feature in this condition [2, 3, 5—7].

all

patients with IgA nephropathy who had urinary erythrocyte counts below 50,000/mi at the time of biopsy. Over the same time interval 22 such patients were identified.

At biopsy, all specimens

one

were divided into three portions,

each for light microscopy (L.M.), electron microscopy

(E.M.) and fixed in

immunofiuorescence (I.F.). For L.M. the tissue

was

Formal mercury for five minutes, transferred to Dubosq

Brazil fixative for a minimum of 20 minutes, dehydrated, cleared and embedded in Paraplast (Sherwood Medical Corp.).

The mechanism which causes progressive glomerular loss Tissue stored from the time of biopsy was recut for this study. through crescent formation is poorly defined, but during cpi- One to 1.5 tm serial sections were cut and stained with sodes of macroscopic hematuria there is generalized B cell haematoxylin and eosin, Masson trichrome, periodic acidactivation and a polyclonal increase in IgA and IgG secreting Schiff, silver methenamine Masson trichrome and orcein elascells [8,9]. This probably occurs as a response to a mucosal tic. Polymorphs were stained with FMC 10 (Australian Monoantigenic challenge [101, and IgU-containing immune complexes clonal Development Corporation) using the following method, have been detected in the serum [11, 121. Recently Ballardie et Two micron sections were dewaxed in Xylene, rehydrated al [131 documented the presence of IgA auto-antibodies with through graded ethanols (100%, 75%, 50%), and the mercury specificity for glomerular antigenic determinants in the sera of pigment removed with Lugols iodine followed by 5% sodium patients with mesangial IgA glomerulonephritis. The levels of thiosulphate. The sections were washed with TRIS buffer, pH auto-antibodies were raised during macroscopic hematuria, 7.6 prior to the application of FMC 10 diluted Vio with TRIS buffer. The binding of the antibody was demonstrated using a Received for publication February 14, 1989 and in revised form July 13, 1989 Accepted for publication July 18, 1989

© 1989 by the International Society of Nephrology

universal rabbit quick staining kit (DAKO K685) which utilizes a labelled avidin biotin method, Cells were counted as positive if there was any brown cytoplasmic staining (Fig. 1).

Monocytes and macrophages were stained with HAM 56

(Enzo biochem M.A. 935) after the same pretreatment as for FMC 10. The sections were washed with TRIS buffer, pH 7.6, 1108

1109

Kincaid-Smith et a!: Polymorphs in IgA glo,nerulonephritis

:O

,

-S

1'

S

t,

$

a

I,,

,t V

'S

. St.

4:

*51

4p and the binding of the antibody was demonstrated using the following immunoperoxidase method. 1. TRIS buffer bath, pH 7.6, 10 mm 2. 0.1% trypsin, 37°C, 10 mm 3. Wash TRIS buffer 4. 20% Normal swine serum, 10 mm 5. HAM 56, Vioo, 60 mm 6. Wash TRIS buffer 7. Rabbit anti-mouse (Dako Z 259), '/2oo, 15 mm 8. Wash TRIS buffer 9. Swine anti-rabbit (Dako Z 196), '/50, 15 mm 10. Wash TRIS buffer 11. Rabbit P.A.P. (Dako Z 113), ¼o, 15 mm 12. Wash TRIS buffer 13. 0.05% 3,3'Diamobenzidine (Sigma D 5637) 4 mm 14. Wash distilled water 15. Harris hematoxylin 2 mm 16. Dehydrate, clear and mount in DePex

The presence of fibrin was demonstrated using both the labelled avidin biotin system (DAKO K685) to detect rabbit antihuman fibrinogen (DAKO A080) and the Masson trichrome stain. The specificity of the staining of FMC 10 and HAM 56

$

•

as

tilt:

Fig. 1. Paraffin section showing a glomerulus

with positive staining with FMC 10 demonstrating intracapil!ary po!ymorphs. (x400).

Results

A biopsy taken during an episode of macroscopic hematuria usually shows an increase in polymorphs and macrophages in glomerular capillaries (Fig. 2). Macrophages are recognized with difficulty unless stained by the immunoperoxidase technique using HAM 56. Polymorphs, although recognizable on hematoxylin and eosin sections (Fig. 2) are more easily counted after immunoperoxidase staining using FMC 10. Table 1 shows that biopsies taken during or within 30 days of episodes of macroscopic hematuria show significantly more polymorphs, macrophages, fibrin and crescents than those from in patients with low urinary erythrocyte counts. Discussion

We have been very impressed by the value of the urinary erythrocyte count in predicting the activity of renal biopsy lesions in mesangial IgA glomerulonephritis. This study supports our previous evidence showing that polymorphs, macro-

phages, fibrin and crescents are all more frequent during macroscopic hematuria, fibrin and crescents were not observed when the urinary erythrocyte count was below 50,000 per ml.

We believe that the major mechanism of progression in this was confirmed by staining lymph node and spleen sections disease is through destruction of glomeruli by crescent formaprepared by the same method as the renal biopsy specimens. tion which occurs both during episodes of macroscopic hemaFor each monoclonal antibody, all glomeruli in a single coded tuna and when there is a continuing high count of urinary slide (usually 3 tissue sections containing up to 40 glomeruli) erythrocytes. The highest relative risk for progression in this were counted independently by two observers. The mean, form of glomerulonephritis is a persisting high urinary erythromedian, minimum and maximum number of polymorphs and of cyte count, and there is an excellent correlation between the monocytes/macrophages were recorded, and averaged between

the two observers. The numbers of glomeruli with positive staining for fibrin on Masson stain were also recorded.

Statistics Statistical analysis used the Mann-Whitney non-parametric test, adjusted for multiple comparisons.

urinary erythrocyte count and crescent formation [3]. In this study we document high counts of polymorphs and macrophages in glomeruli in biopsies taken within 30 days of an episode of macroscopic hematuria. Macrophages have been known for many years to participate in glomerular injury in crescentic glomerulonephritis [14, 15] but multiple descriptions of the pathology in mesangial IgA glomerulonephritis fail to

Kincaid-Smith et a!: Po!ymorphs in IgA glomerulonephritis

1110

• S

4$

-

'fl

'r

a..

Table 1. Quantitative comparison of polymorphs (FMC 10) monocytes/macrophages (HAM 56), fibrin and crescents between control patients with low-grade hematuria and patients with recent macroscopic hcmaturia

Control

N=22 Mean FMC 10 + ye

Recent macroscopic hematuria

N=54

0.20

0.13

0.03

1.10

10.5

2.9

45.9

3.4°

0.59

0.14

1.58

0.201)

40.0

8.4

65.6

9•7a

0

10.0

2.0°

0

16.5

2.2°

cells/glomerulus

% Glomeruli ÷ ye for FMC 10 cells Meaa Ham 56 + ye cells/glomerulus % Glomeruli + ye for Ham 56 cells

% Glomeruli + ye for Fibrin % Glomeruli with crescents Median % glomeruli with crescents Patients with crescents Results are expressed as x

a c 0.05 i c 0.01

0

12.5°

0/22

45/54

SE.

I

Fig. 2. Glomerulus (Hematoxylin-eosin )< 800) at time of macroscopic hematuria. Polymorphs are prominent in the capillary loops.

lar granulocyte and total leucocyte counts were found. These findings differ from our study, where the control group had only 0.12 0.03 (mean su) polymorphs per glomerulus, while the heavy-hematuria group showed 1.10 0.20 polymorphs. The biopsies of our control group may provide better baseline data

than operative tissue specimens, particularly as glomerulonephritis is well documented in patients with tumors [17, 18], including those of renal origin. A computer-assisted literature search reveals the sole previous article connecting polymorphs with IgA nephropathy to be an electron microscopic study of

the glomerular basement membrane in children with IgA nephropathy and Henoch-SchOnlein purpura [19]. Lysis of the glomerular basement membrane, which correlated with. disease severity, was frequently associated with subepithelial dense deposits and polymorphonuclear leucocytes. This is suggestive evidence that polymorphs cause basement membrane disruption. In the study of Hooke, Gee and Atkins [16], glomerular fibrin deposition in the total patient group encompassing a wide range of glomerulonephritides was associated with a significant increase in total leucocyte and granulocytes within glomeruli.

This is fully consistent with our findings but our different

process of patient selection precludes further comparison. As heavy hematuria is strongly associated with crescent formation [3, 4] our group of patients with heavy hematuria would be expected to have more fibrin in glomeruli. Participation of polymorphs in several experimental models mention polymorphonuclear leucocytes [1, 2, 5—7]. In a similar study analyzing leucocytes in human glomerulonephritis, of glomerulonephritis is well documented, particularly in the Hooke et al [16] used the FMC1O antibody to quantitative heterologous phase of anti-GBM disease [20] where passage of glomerular granulocytes in 18 patients with IgA nephropathy of polymorphs through gaps in the basement membrane has been unspecified clinical status. They did not find an increase in documented during macroscopic hematuria [21]. There has granulocytes but reported 0.8 0.3 (mean SE) granulocytes been much recent interest in release of reactive oxygen species by polymorphonuclear leucocytes as a mechanism of injury in per glomerular cross-section, of a total leucocyte count of 3.8 0.7. In control sections from either cadaver kidneys not used for glomerulonephritis. Neutrophils as well as monocytes can also transplantation or kidneys removed for localized tumors, simi- release many other toxic and vasoactive substances including P C 0.001

Kincaid-Smith et a!: Po/ymorphs in IgA g/omerulonephritis

1111

tients with active idiopathic IgA nephropathy. C/in Nephrol 26(4): proteases, prostaglandins, leucotrienes and platelet activating 163—168, 1986 factor [22]. Protean mechanisms probably interact to produce 9. FEEHALLY J, BEATTIE TJ, BRENCHLEY PEC, CouPEs BM, MALglomerular injury. Furthermore, human polymorphs are known LICK NP, POSTLETHWAITE Ri: Sequential study of the IgA system to possess specific IgA Fe receptors which are markedly up in relapsing IgA nephropathy. Kidney mt 30:924—931, 1986 regulated by IgA [23]. Increased receptor expression is accom- 10. COUSER WG: Mesangial IgA nephropathies—steady progress. West JMed 140:89—91, 1984 panied by enhanced antibody-dependent cell cytotoxicity, but 11. CZERKINSKY C, KOOPMAN Wi, JACKSON S: Circulating immune an intra-glomerular role for this has not been defined. complexes and immunoglobulin A rheumatoid factor in patients This study demonstrates that both polymorphs and macrowith mesangial immunoglobulin A nephropathies. J Clin Invest phages are present and presumably participating in glomerular 77:1931—1938, 1986 12. SINICO RA, FORNASIERI A, ORENI N, BENUZZI S. D'AMICO G: injury in mesangial IgA glomerulonephritis.

Polymeric IgA rheumatoid factor in idiopathic IgA mesangial Acknowledgments The authors thank Mr. R. Wong for technical assistance in the histological preparations, and Mrs. Janet Barr for preparing this manuscript.

nephropathy (Berger's disease). J Immunol 137:536—541, 1986 13. BALLARDIE FW, WILLIAMS S, BRENCHLEY PE, O'DONOGHUE DJ: Auto-immunity in IgA nephropathy. Lancet 11:588—592, 1988 RC, HOLDSWORTH SR, HANcocic WM, THOMSON NM, GLASGOW EF: Cellular immune mechanisms in human glomerulo-

14. ATKINS

Reprint requests to Professor P. Kincaid-Smith, Director of Nephrology, The Royal Melbourne Hospital, Parkville, 3050, Victoria, Australia.

nephritis: The role of mononuclear leucocytes. Springer Semin Immunopathol 5:269—296, 1982 15. D'AMICO G, IMBASCIATI E, DI BELGI0I0S0 GB, BERTOLI S: Idiopathic IgA mesangial nephropathy: Clinical and histological

References

16. HOOKE DH, GEE DC, ATKINS RC: Leucocyte analysis using monoclonal antibodies in human glomerulonephritis. Kidney mt

1. D'Asnco G: The commonest glomerulonephntis in the world: IgA nephropathy. Q J Med 245:707—727, 1987

31:964—972, 1987 17. ZIMMERMAN SW, MOORTHY AV, BURKHOLDER PM, JENKINS PG:

study of 374 patients. Medicine 64(1):49—60, 1985

2. LEVY M, GONZALEZ-BURCHARD G, BROYER M, DOMMERGUES JP,

Glomerulopathies associated with neoplastic disease, in Cancer

FOULARD M, SOREZ JP, HABIB R: Berger's disease in children: Natural history and outcome. Medicine 64:157—180, 1985

and the Kidney, edited by RIESELBACH RE, GARNICK MG, Phila-

3. NICHOLLS K, FAIRLEY KF, DOWLING J, KINCAID-SMITH P: The

18. ALPERS CE, COTRAN RS: Neoplasia and glomerular injury. Kidney mt 30:465—473, 1986 19. YOSHIKAwA N, YOSHIARA 5, YoSHIYA K, MATSUO T, OKADA S:

clinical course of mesangial IgA associated nephropathy in adults.

Q J Med 4.

210:227—250, 1984

BENNETT WM, KINCAID-SMITI-I: Macroscopic haematuria in IgA nephropathy. Kidney mt 23:293—400, 1983

5. DROZ D: Natural history of primary glomerulonephritis with 6.

mesangial deposits of IgA. C/in Nephrol 2:150—157, 1976 ABE T, KIDA M, YosHIMu M, YOKOYAMA H, KosMiNo Y, ToMosua N, NATTORI N: Participation of extracapillary lesions

(ECL) in progression of IgA nephropathy. Clin

delphia, Lea and Febiger, 1982

Lysis of the glomerular basement membrane in children with IgA nephropathy and Henoch-Schdnlein nephritis. J Patho/ 150:119— 126, 1986 20. COMEANG CG, UNANUE ER, DIXON FJ: A role of polymorphonu-

clear leucocytes and complement in nephrotoxic nephritis. J Exp Med 122:95—1 10, 1965

Nephrol 25(1):

21. MAKINO H, NISHIMURA 5, TAKAOKA M, OTA Z: Mechanism of

7. ABUELO JG, ESPARZA AR, MATARESE RA, ENDRENY RG, CAR-

Haematuria. II. A scanning electron microscopic demonstration of the passage of blood cells through a glomerular capillary wall in rabbit Masugi nephritis. Nephron 50:142—150, 1988

37—41,

1986

VALHO JS, ALLEGRA SR: Crescentic IgA nephropathy. Medicine 63:396—406, 1984 8. SCHENA FP, MASTROLITTI G, FRANcAsso AR, PASTORE A, LADISA

N: Increased immunoglobulin-secreting cells in the blood of pa-

22. WEISSMAN G, SMOLEN JE, KORCUAU HM: Release of inflamma-

tory mediators from stimulated neutrophils. N EngI J Med 303: 27—34, 1980