- Email: [email protected]
IgA-associated glomerular deposits in liver disease
IgA-associated Glomerular Deposits in Liver Disease CARLOS ABRAMOWSKY,MD, BEVERLYDAHMS, MD, AND GARY SWINEHART,BA IgA-associated immunopathologic renal injury has been reported in patients with cirrhosis. In an effort to elucidate the pathogenesis of this phenomenon, the livers and kidneys obtained at autopsy from a group of patients with cystic fibrosis were studied; these patients were selected because of their broad spectrum of liver abnormalities. On the basis of histologic examination of sections of liver, the patients were divided into two groups: Group I (20 patients) included patients with focal biliary cirrhosis and multilobular biliary cirrhosis; all had anatomic distortion of the biliary system, and many had cholestasis. Group II (28 patients) showed n u b i l e duct anomalies. Immunofluorescence studies of the corresponding kidneys for immunoglobulin, complement, and free' secretory component (FSC) revealed significantly more numerous IgA-containing glomerular deposits in group I (P < 0.02). Although FSC was virtually absent in these deposits, significant in vitro binding of this protein revealed the polymeric nature of the glomerular IgA. This is consistent with previous observations of elevated serum levels of polymeric IgA, which forms the dominant component of glomerular deposits in cirrhotic patients. Since IgA glomerular deposition occurred in patients with focal biliary and no hepatocellular dysfunction, it seems that the source of this polymeric lgA is related to its impaired serum clearance by a distorted and stagnant bile duct system. Ilowever, the mechanism that leads to the deposition of this immunoglobulin in the glomeruli and other tissues remains conjectural. HUM PATHOL 16:1243--1246, 1985.
Polymeric s e r u m IgA has been shown to be transported selectively across hepatobiliary barriers, where it is coupled with free secretory component (FSC). The resuhing secretory immunoglobulin is excreted in the bile and along with its locally synthesized congener, contributes to the protection of the intestinal barrier. It has been demonstrated that in patients with cirrhosis polymeric IgA levels are elevated in the serum and that deposits of this immunoglobulin can be found i n the glomertfli as well as in other sites. In the present study we sought to deternfine the degree and pattern of hepatic attatomic distortion that are required to produce IgA renal deposition. Patients with cystic fibrosis were selected for the study, since these patients have a spectrum of liver conditions, ranging from norntal and fatty liver through focal biliary cirrhosis and severe muhilobular biliary cirrhosis.
Received from the Institute of Pathology, UniversityHospitals of Cleveland and Case Western Reserve University, Cleveland, Ohio. Revision accepted for publicationJuly 1, 1985. Address correspondence and reprint requests to Dr. Abramowsky: Institute of l'atholo~y, Case Western Reserve University, 2085 Adelbert Road, Cleveland, OIl 44106.
MATERIALSAND METHODS
Samples of liver and kidney were obtained at the time of autopsy front 48 patients with cystic fibrosis. Most of the autopsies were performed within a few honrs after deatit, and tissues were imnlediately frozen in isopentane liquid nitrogen. Additional samples were taken randomly for routine histologic evaluation, h n m u n o f l u o r e s c e n c e studies were performed in kidney and selected liver specimens with antisera to IgG, IgM, IgA, C3, and albuntin control (Cappell Laboratory, Malvern, Pennsylvania). T h e monospecificity of these reagents was confirmed by immtmoelectrophoresis and by immunohistologic studies of known tissue substrates. Frozen sections of the specimens were also stained by indirect immttnofluorescence with rabbit anti-FSC (Dako, Westbury, New York). This antibody was affinity-purified and trod been shown by gel difft/sion and immunoblotting techniques to react monospecifically with human milk whey, saliva, and bound secretory component (secretory IgA), but not with heavy or light chains of immunoglobulins or other serum proteins. For the latter tests sera from patients with myeloma and cirrhosis (and elevated IgA levels) were used. In a binding assay in microtiter plates the antibody did not react with lactoferrin. By immunofluorescence the anti-FSC stained lactating breast and intestinal epithelial, but not mncosal, plasma cells and did not stain frozen sections of known plasmocytomas or lymphomas (heavy mad light chains) or a preparation of lacto ferrin-containing granulocytes. In vitro binding of FSC was deternfined by overlaying frozen sections of kidney with a 200 I.tg/ml solution o f human FSC l obtained from human milk by the method of Kobayashi'-' (provided by Dr. Micltael E. Lamm, Institute of Pathology, Case Western Reserve University). A paired section was alternatively treated with a bovine serum albumin buffer solution, and both were subsequently stained with anti-FSC, as described earlier. Histologic evaluation of tire liver was performed blindly, and each case was assigned to one of the major categories of liver morphology associated with cystic fibrosis. 3 Patients with either severe nmhilobular biliary cirrhosis or milder focal biliary cirrhosis (with or without cholestasis) constituted the g r o u p with abnormal bile ducts (group I). Patients who had fatty metamorphosis alone, or normal livers, constituted the normal bile duct group (group II). Kidney sections were evahmted with hentatox-
1243
HUMAN PATHOLOGY TABLE t.
Volume 16, No. 12 [December 1985]
Hepatic Renal Immunofluorescence and Histopathologic Findings In Patients with Cystic Fibrosis
Group 1: abnormal bile ducts Muhilobular biliary cirrhosis Focal biliary cirrhosis Total Group II: normal bile ducts Fatty liver No pathologic diagnosis Total
Total Positive Fluorescence
ImmunoglobulinPositive Fluorescence
No. of Patients
IgG
lgM
IgA
C3
7 13 20
1 4 5
3 8 11
2 6 8*
6 7 13
6 10 16
3 8 11
17 11 28
3 1 4
6 5 II
1 2 3*
5 6 11
6 7 13
6 6 12
* P < 0.02.
ylin-eosin, periodic acid-Schiff, and Jones' silver stain for histopathologic aherations. 4
these sites. In the livers in group II, no IgA or secretory component deposits were seen.
RESULTS
DISCUSSION
T h e res'ults o f the renal i m m u n o f l u o r e s c e n c e studies are shown in table 1. Eight of the 20 specimens from group I showed IgA deposits, predominantly in the mesangium (fig. 1), with prominent subendothelial localization in two specimens. In contrast, IgA deposits were seen in only three of the 28 specimens from group II (P < 0.02). This significant segregation o f glomerular IgA del)osition in patients with abnormal bile ducts was not confirmed for IgG, IgM, or C3. Histopathologic changes in the IgA-positive renal specimens included mild mesangial expansion with minimal cellular proliferation. In four specinaens additional lesions, such as splitting and segmental thickening o f the capillary loops, were observed. In group I, glomerular lgA was found not only in patients with multilobular biliary cirrhosis but also in those with the milder focal biliary cirrhosis, with or without cholestasis (table 1). In the 11 kidneys with IgA deposits, immunofluorescence studies with anti-FSC antibodies (table 2) showed only trace glomerular deposition in four specimens and 1 + staining in one specimen. In contrast, the FSC-binding study demonstrated noticeable binding in nine of the 11 cases in which IgA deposits were present. O n e o f three IgA-positive patients from group II (patient 9, table 2) and one of eight from group I (patient 7, table 2) showed no binding of FSC, suggesting that in those cases the glomerular IgA may not have been polylneric. Specimens that were negative tbr IgA, with or without IgM, showed no binding of FSC. h n n a u n o f l u o r e s c e n c e studies o f the liver in group I showed IgA and secretory component in cholestatic plugs in the lumen and epithelial cytoplasm of bile ducts, cholangioles, or canaliculi in four of 12 specimens. In addition, IgA and secretory component were seen occasionally in hepatic sinusoids, hepatocellular cytoplasm, and Kupffer cells. IgA- and IgM-containingplasnm cells were seen only rarely in> portal areas in some specimens. Other immunoglobulins or albumin were not substantially present in
This study showed a significant deposition of lgA in 40 per cent of the kidneys obtained from patients with cystic fibrosis who had either focal or multilobular biliary cirrhosis. T h e ill vitro capacity of most of these IgA deposits to bind FSC indicates their polymeric (dimeric) nature, a requisite for such binding. Binding of FSC also occurred in the presence of IgM, but only if Ig~,k was also present. Furthermore, since IgM was not significantly segregated in group I, these observations suggest that IgM had no intportant role in binding FSC, and it was probably present ibr other reasons, as reported previously. 5 IgA deposits in the kidney and other tissues have b6en observed in association with liver disease. Berger et al., 6 in an autopsy study of 100 cirrhotics, found IgA, predominantly in the glomerular mesangium, in 61 per cent of the patients. 7 Woodroffe found mesangial IgA in 50 per cent of the kidneys from another group of cirrhotic patients. In this same report IgA deposits were also described in the skin and choroid plexus. Other studies with similar resuhs have been published recently, 8-11 and deposition of IgA along hepatic sinusoids has been mentioned as characteristic of alcoholic liver disease. 12,13 Immunochemical serum studies in patients with liver disease have shown elevated levels of polymeric IgA 14-16 and even secretory IgA, 17 which is normally present in negligible quantities. Polymerization of IgA occurs at the level of mucosal, mainly gastrointestinal, plasma cells in preparation for its binding with FSC as the immunoglobulin is translocated across the surface epithelium. Is However, some polymeric IgA ends up in the circulation, where it is norntally cleared by hepatobiliary transport. This system has been well studied in the ra0 o-':l and has been confirmed in humans. 22 It provides a mechanism for the removal o f circulating polymeric IgA, which is bound to FSC in hepatocytes in rats or, in humans, in bile canaliculi and bile ducts, with the resulting secretory immunoglobulin
4244
IGA4kSSOCIATED RENAL DEPOSITSIN LIVER DISEASE (Abramowsky et al.]
FIGURE 1 (left). Immunofluorescence photograph illustrating IgA deposition In the mesangial region of a glomerulus from a patient with focal biiiary cirrhosis. ( x 250.) FIGURE 2 [right). Immunofluorescence preparation of the liver stained with anti-free secretory component showing a cholestatic plug in a portal bile duct (lower right of field). 1he branching structures in the center are proliferating bile ductules, or cholangioIes. This preparation shows staining of the ductular cytoplasm as well as the inspissated luminal contents. The lafler were also shown to stain intensely with antiIgA antibodies. [ x 250.]
draining into bile. z3 Additional local production of secretory IgA in biliary mucosa has been shown. 24 In the presence of liver disease, tiffs hepatobiliary transport is impaired, resulting in rising levels of polymeric IgA. The mechanisms that lead to the deposition of polymeric IgA in the mesangium are less well underTABLE 2.
Glomerular Secretory Component Binding In IgA-positive Renal Specimens
Patient
Liver Disease
lgM
1 2 3 4 5 6 7 8 9 I0 11
MBC MBC FBC-C FBC-C FBC-G FBC FBG FBC Fat None None
+ + + + + + + + + + +
IgA
Indirect FSC
Secretory Component Binding
+ + + + + + + + + + +
Tr I+ Tr Tr Tr
4+ 1+ 3+ 3+ 1+ 3+ 1+ 1+ 2+
ABBREVIATIONS: M B C . ' m u h i l o b u l a r biliary cirrhosis; FBC. focal biliar)' c i r r h o s i s ; C, c h o l e s t a s i s ; Fat. fatt)' liver; T r , t r a c e ; FSC, free secretory component.
stood. A specific polymeric IgA receptor, similar to that described for complement, may be present, or the deposits may represent immune or nonimmune aggregates of native or d e n a t u r e d polymeric IgA. Our observations of the secretory immunoglobulin exposed to a detergent, such as bile, within cholestatic plugs lead us to favor the latter possibility. However, evidence has been submitted in support of tile hypothesis that most polymeric IgA, which should be cleared by the hepatobiliary system, is in the form of cryoglobulins s,25 or i m m u n e complexes 26-31 and, therefore, subject to tissue deposition if appropriate conditions exist. The patients in most previous studies had severe forms of liver disease, namely cirrhosis. Our studies showed that IgA renal deposits also occur with the pathologically milder focal biliary cirrhosis, in tile absence o f laboratory evidence of liver dysfunction. This suggests that severe hepatocelhdar disease is not a requisite for decreased clearance of polymeric IgA and implies that regurgitation in distorted and stagnant biliary pathways may account for re-entry of the immnnoglobulin into the serum. This view is supported further by the recent report of elevation of secretory IgA levels in the sera of patients with liver disease. 17 Regurgitation is also s u p p o r t e d by the 1245
HUMAN PATHOLOGY
Volume 16, No. '12 (December 1985)
r a p i d rise in s e r u n t levels o f I g A a n d tissue d e p o s i t i o n w i t h i n a few days a f t e r l i g a t i o n o f e x t r a h e p a t i c bile ducts, with r e v e r s a l o f this process o n t-elease o f t h e o b s t r u c t i o n . 32 T h e c o n s i s t e n t lack o f s e c r e t o r y c o m p o n e n t associated with t h e I g A tissue d e p o s i t s r e p o r t e d i n p r e vious w o r k a n d the c u r r e n t s t u d y s u g g e s t s t h a t m o s t o f this I g A r e v e r t s to tlte c i r c u l a t i o n b e f o r e t r a v e r s i n g t h e bile d u c t s .
Acknowledgments. T h e attthors express their gratitude to Mary J o h n s o n and Robert Langlotz for their help in tire preparation of this study. REFERENCES 1. Bene MC, Faure G, Duheille J: IgA nephropathy: characterization of the polymeric nature of mesangial deposits by in vitro binding of free secretory component. Clin Exp Immunol 47:527, 1982 2. Kobayashi K: Studies on human secretory IgA. Comparative studies on the IgA bound and free secretory component. Innnunochemistry 8:785, 1971 3. Oppenheimer Ett, EsterlyJR: Pathology of cystic fibrosis. Review of the literature and comparison with 146 autopsied cases. Perspect Pediatr Pathol 2:24 l, 1975 4. Luna LG: Manual of I listologic Staining Methods of the AFIP, ed 3. New York, McGraw Hill, 1968 5. Davis CA, Abramowsky CR, Swinehart G: Circtdating iummne complexes and the nephropathy of cystic fibrosis, llum Pathol 15:244, 1984 6. Berger, J, Yaneva H, Nabarra B: Glmnerular changes its patients with cirrhosis of the liver. Adv Nephrol 7:3, 1977 7. Woodroofe AJ: IgA, glomerulonephritis and liver disease. Aust NZJ Med ll:109, 1981 8. Nochy D, Callard P, Bellon B, et al: Association of overt glomerulonephritis and liver disease: a study of 34 patients. Clin Nephrol 6:422, 1976 9. Nakamoto Y, Iida II, Kobayaski K, et al: Ilepatic glomerulonephritis: characteristics of hepatic IgA glomerulonephritis as the major part. Virchows Arch [A] 392:45, 1981 I0. Endo Y, Matsushita H, Nozawa Y, et al: Glomerulonephritis associated with liver cirrhosis. Acta PatholJpn 32:333, 1983 11. Lomax-Smith JD, Zabrowarn)' LA, Howarth GS, et al: The immunochemical characterization of mesangial IgA deposits. Am J Pathol 113:359, 1983 12. Kater L, Jobsis AC, Baart de la Faille-Kuyper EIt, et ah Alcoholic hepatic disease: specificity of IgA deposits in liver. AmJ Clin Pathol 71:51, 1979 13. Swerdlow MA, Chowdhury LN: lgA subclasses in liver tissues in alcoholic liver disease. Am J Clin Pathol 80:283, 1983 14. Chandy KG, Hubscher SG, Elias E, et ah Dual role of the liver its regulating circulating polymeric lgA in man: studies on patients with liver disease. Clin Exp lmmunol 52:207, 1983 I 1. Lomax-Smith JD, Zabrowarny LA, llowarth GS, et al: The immunocllemical characterization of mesangial IgA deposits. AmJ Pathol 113:359, 1983 12. Kater L, Jobsis AC, Baart de la Faille-Kuyper EH, et al: Alcoholic hepatic disease: specificity of IgA deposits in liver. A m j Clin Pathol 71:51, 1979 13. Swerdlow MA, Chowdhury LN: IgA subclasses in liver tissue in alcoholic liver disease. Am J Clin Pathol 80:283, 1983
14. Chandy KG, llubscher SG, Elias E, et al: Dual role of the liver in regulating circnlating pol)meric IgA ill man:'studies on patients with liver disease. Clin Exp hnmunol 52:207, 1983 15. Delacroix DL, Elkon KB, Geubel AP, et al: Changes in size, subclass, and metabolic properties of serum imnumoglobt,lin A in liver disease and in other diseases with Ifigh serum imntunoglobuIinA. J Clin Invest 71:358, 1983 16. Newkirk MM, Klein bill, Katz A, et ah Estimation of pol).meric IgA in lmtnan serum: an assay based on binding of radiolabeled human secretory component with applications in the study of IgA nephropathy, IgA monoclonal gammopathy, and liver disease. J Immunol 130:1176, 1983 17. Farris MA, Chandy KG, Ashdown K, et al: Serum secretory component: a potential marker of biliary obstruction. Clin Chim Acta 133:65, 1983 18. Brandtzaeg P, Prydz II: Direct evidence for an integrated function of J chain attd secretory component in epithelial transport of inmmnoglobulins. Nature 311:71, 1984 19. Fisher M*I, Nag) B, Bazin II, et al: Biliary transport of IgA: role of secretory component. Proc Natl Acad Sci 76:2008, 1979 20. I'eppard J, Orlans E, Payne AWR, et al: The elimination of circulating complexes containing polymeric IgA by excretion in the bile. Immunol 42:83, 1981 21. Socken DJ, Simms ES, Nag} BR, et al: Secretory componentdependent hepatic transport of IgA antibody-antigen complexes. J Immunol 127:316, 198 I 22. Delacroix DL, ttodgson HJF, Mcl'herson A, et al: Selective transport of polymeric immunoglobulin A in bile. J Clin Invest 70:230, 1982 23. Nagura II, Smith PD, Nakane PK, et ah IgA in Imman bile and liver. J hmntmol 126:587, 1981 24. Nagura II, Tsutsumi Y, llasegawa II, et al: IgA plasma cells in biliary mucosa: a likely course of locally synthesized IgA in human hepatic lille. Clin Ex]~ Immunol 54:671, 1983 25. Kaufman RL, Hoers JC, Quismorio FP: Immune complexes in the portal and systenfic circulation of patients with alcoholic liver disease. Clin Immtmol Immtmopathol 22:44, 1982 26. Sancho J, Egido J, Sanchez-Crespo M, et al: Detection of monomeric and polymeric IgA containing imnttme com" plexes in serum and kidney from patients with alcoholic liver disease. Clin Exp lmmunol 47:327, 1981 27. Sancho J, Egido J, Rivera F, et al: Immune complexes in IgA nephropathy: presence of antibodies against diet antigens and delayed clearance of specific polymeric IgA immune complexes. Clin Exp Immtmol 54:194, 1983 28. Brown TA, Russell MW, Mestecky J: llepatobiliary transport of IgA immune complexes: molecular and celhdar aspects. J hnmunol 128:2183, 1982 29. Burns J, D'Ardenne AJ, Morton JA, et al: Immune complex nephritis in alcoholic cirrhosis: detection of Mallory body antigen in complexes by means of monoclonal antibodies to Mallor)" bodies. J Clin Pathol 36:751, 1983 30. Emancipator SN, Gallo GR, Razaboni R, et ah Experimental cholestasis promotes the deposition of glomerular IgA immune complexes. Am J Pathol 113:19, 1983 31. Ohshio G, Furukawa F, Sekita K, et al: IgA containing circulating immune complexes and IgA anti-single stranded DNA antibodies in patients with obstructive jaundice. Clin Exp lmmunol 59:435, 1985 32. Cuadrado E, Arenas j I , Echaniz P, et al: Rapid decrease of secretory lgA serum levels in extrahepatic obstructive jaundice after surgical relief of the bile duct obstruction (letter). Gastroenterology 84:203, 1983
1246
Polymeric s e r u m IgA has been shown to be transported selectively across hepatobiliary barriers, where it is coupled with free secretory component (FSC). The resuhing secretory immunoglobulin is excreted in the bile and along with its locally synthesized congener, contributes to the protection of the intestinal barrier. It has been demonstrated that in patients with cirrhosis polymeric IgA levels are elevated in the serum and that deposits of this immunoglobulin can be found i n the glomertfli as well as in other sites. In the present study we sought to deternfine the degree and pattern of hepatic attatomic distortion that are required to produce IgA renal deposition. Patients with cystic fibrosis were selected for the study, since these patients have a spectrum of liver conditions, ranging from norntal and fatty liver through focal biliary cirrhosis and severe muhilobular biliary cirrhosis.
Received from the Institute of Pathology, UniversityHospitals of Cleveland and Case Western Reserve University, Cleveland, Ohio. Revision accepted for publicationJuly 1, 1985. Address correspondence and reprint requests to Dr. Abramowsky: Institute of l'atholo~y, Case Western Reserve University, 2085 Adelbert Road, Cleveland, OIl 44106.
MATERIALSAND METHODS
Samples of liver and kidney were obtained at the time of autopsy front 48 patients with cystic fibrosis. Most of the autopsies were performed within a few honrs after deatit, and tissues were imnlediately frozen in isopentane liquid nitrogen. Additional samples were taken randomly for routine histologic evaluation, h n m u n o f l u o r e s c e n c e studies were performed in kidney and selected liver specimens with antisera to IgG, IgM, IgA, C3, and albuntin control (Cappell Laboratory, Malvern, Pennsylvania). T h e monospecificity of these reagents was confirmed by immtmoelectrophoresis and by immunohistologic studies of known tissue substrates. Frozen sections of the specimens were also stained by indirect immttnofluorescence with rabbit anti-FSC (Dako, Westbury, New York). This antibody was affinity-purified and trod been shown by gel difft/sion and immunoblotting techniques to react monospecifically with human milk whey, saliva, and bound secretory component (secretory IgA), but not with heavy or light chains of immunoglobulins or other serum proteins. For the latter tests sera from patients with myeloma and cirrhosis (and elevated IgA levels) were used. In a binding assay in microtiter plates the antibody did not react with lactoferrin. By immunofluorescence the anti-FSC stained lactating breast and intestinal epithelial, but not mncosal, plasma cells and did not stain frozen sections of known plasmocytomas or lymphomas (heavy mad light chains) or a preparation of lacto ferrin-containing granulocytes. In vitro binding of FSC was deternfined by overlaying frozen sections of kidney with a 200 I.tg/ml solution o f human FSC l obtained from human milk by the method of Kobayashi'-' (provided by Dr. Micltael E. Lamm, Institute of Pathology, Case Western Reserve University). A paired section was alternatively treated with a bovine serum albumin buffer solution, and both were subsequently stained with anti-FSC, as described earlier. Histologic evaluation of tire liver was performed blindly, and each case was assigned to one of the major categories of liver morphology associated with cystic fibrosis. 3 Patients with either severe nmhilobular biliary cirrhosis or milder focal biliary cirrhosis (with or without cholestasis) constituted the g r o u p with abnormal bile ducts (group I). Patients who had fatty metamorphosis alone, or normal livers, constituted the normal bile duct group (group II). Kidney sections were evahmted with hentatox-
1243
HUMAN PATHOLOGY TABLE t.
Volume 16, No. 12 [December 1985]
Hepatic Renal Immunofluorescence and Histopathologic Findings In Patients with Cystic Fibrosis
Group 1: abnormal bile ducts Muhilobular biliary cirrhosis Focal biliary cirrhosis Total Group II: normal bile ducts Fatty liver No pathologic diagnosis Total
Total Positive Fluorescence
ImmunoglobulinPositive Fluorescence
No. of Patients
IgG
lgM
IgA
C3
7 13 20
1 4 5
3 8 11
2 6 8*
6 7 13
6 10 16
3 8 11
17 11 28
3 1 4
6 5 II
1 2 3*
5 6 11
6 7 13
6 6 12
* P < 0.02.
ylin-eosin, periodic acid-Schiff, and Jones' silver stain for histopathologic aherations. 4
these sites. In the livers in group II, no IgA or secretory component deposits were seen.
RESULTS
DISCUSSION
T h e res'ults o f the renal i m m u n o f l u o r e s c e n c e studies are shown in table 1. Eight of the 20 specimens from group I showed IgA deposits, predominantly in the mesangium (fig. 1), with prominent subendothelial localization in two specimens. In contrast, IgA deposits were seen in only three of the 28 specimens from group II (P < 0.02). This significant segregation o f glomerular IgA del)osition in patients with abnormal bile ducts was not confirmed for IgG, IgM, or C3. Histopathologic changes in the IgA-positive renal specimens included mild mesangial expansion with minimal cellular proliferation. In four specinaens additional lesions, such as splitting and segmental thickening o f the capillary loops, were observed. In group I, glomerular lgA was found not only in patients with multilobular biliary cirrhosis but also in those with the milder focal biliary cirrhosis, with or without cholestasis (table 1). In the 11 kidneys with IgA deposits, immunofluorescence studies with anti-FSC antibodies (table 2) showed only trace glomerular deposition in four specimens and 1 + staining in one specimen. In contrast, the FSC-binding study demonstrated noticeable binding in nine of the 11 cases in which IgA deposits were present. O n e o f three IgA-positive patients from group II (patient 9, table 2) and one of eight from group I (patient 7, table 2) showed no binding of FSC, suggesting that in those cases the glomerular IgA may not have been polylneric. Specimens that were negative tbr IgA, with or without IgM, showed no binding of FSC. h n n a u n o f l u o r e s c e n c e studies o f the liver in group I showed IgA and secretory component in cholestatic plugs in the lumen and epithelial cytoplasm of bile ducts, cholangioles, or canaliculi in four of 12 specimens. In addition, IgA and secretory component were seen occasionally in hepatic sinusoids, hepatocellular cytoplasm, and Kupffer cells. IgA- and IgM-containingplasnm cells were seen only rarely in> portal areas in some specimens. Other immunoglobulins or albumin were not substantially present in
This study showed a significant deposition of lgA in 40 per cent of the kidneys obtained from patients with cystic fibrosis who had either focal or multilobular biliary cirrhosis. T h e ill vitro capacity of most of these IgA deposits to bind FSC indicates their polymeric (dimeric) nature, a requisite for such binding. Binding of FSC also occurred in the presence of IgM, but only if Ig~,k was also present. Furthermore, since IgM was not significantly segregated in group I, these observations suggest that IgM had no intportant role in binding FSC, and it was probably present ibr other reasons, as reported previously. 5 IgA deposits in the kidney and other tissues have b6en observed in association with liver disease. Berger et al., 6 in an autopsy study of 100 cirrhotics, found IgA, predominantly in the glomerular mesangium, in 61 per cent of the patients. 7 Woodroffe found mesangial IgA in 50 per cent of the kidneys from another group of cirrhotic patients. In this same report IgA deposits were also described in the skin and choroid plexus. Other studies with similar resuhs have been published recently, 8-11 and deposition of IgA along hepatic sinusoids has been mentioned as characteristic of alcoholic liver disease. 12,13 Immunochemical serum studies in patients with liver disease have shown elevated levels of polymeric IgA 14-16 and even secretory IgA, 17 which is normally present in negligible quantities. Polymerization of IgA occurs at the level of mucosal, mainly gastrointestinal, plasma cells in preparation for its binding with FSC as the immunoglobulin is translocated across the surface epithelium. Is However, some polymeric IgA ends up in the circulation, where it is norntally cleared by hepatobiliary transport. This system has been well studied in the ra0 o-':l and has been confirmed in humans. 22 It provides a mechanism for the removal o f circulating polymeric IgA, which is bound to FSC in hepatocytes in rats or, in humans, in bile canaliculi and bile ducts, with the resulting secretory immunoglobulin
4244
IGA4kSSOCIATED RENAL DEPOSITSIN LIVER DISEASE (Abramowsky et al.]
FIGURE 1 (left). Immunofluorescence photograph illustrating IgA deposition In the mesangial region of a glomerulus from a patient with focal biiiary cirrhosis. ( x 250.) FIGURE 2 [right). Immunofluorescence preparation of the liver stained with anti-free secretory component showing a cholestatic plug in a portal bile duct (lower right of field). 1he branching structures in the center are proliferating bile ductules, or cholangioIes. This preparation shows staining of the ductular cytoplasm as well as the inspissated luminal contents. The lafler were also shown to stain intensely with antiIgA antibodies. [ x 250.]
draining into bile. z3 Additional local production of secretory IgA in biliary mucosa has been shown. 24 In the presence of liver disease, tiffs hepatobiliary transport is impaired, resulting in rising levels of polymeric IgA. The mechanisms that lead to the deposition of polymeric IgA in the mesangium are less well underTABLE 2.
Glomerular Secretory Component Binding In IgA-positive Renal Specimens
Patient
Liver Disease
lgM
1 2 3 4 5 6 7 8 9 I0 11
MBC MBC FBC-C FBC-C FBC-G FBC FBG FBC Fat None None
+ + + + + + + + + + +
IgA
Indirect FSC
Secretory Component Binding
+ + + + + + + + + + +
Tr I+ Tr Tr Tr
4+ 1+ 3+ 3+ 1+ 3+ 1+ 1+ 2+
ABBREVIATIONS: M B C . ' m u h i l o b u l a r biliary cirrhosis; FBC. focal biliar)' c i r r h o s i s ; C, c h o l e s t a s i s ; Fat. fatt)' liver; T r , t r a c e ; FSC, free secretory component.
stood. A specific polymeric IgA receptor, similar to that described for complement, may be present, or the deposits may represent immune or nonimmune aggregates of native or d e n a t u r e d polymeric IgA. Our observations of the secretory immunoglobulin exposed to a detergent, such as bile, within cholestatic plugs lead us to favor the latter possibility. However, evidence has been submitted in support of tile hypothesis that most polymeric IgA, which should be cleared by the hepatobiliary system, is in the form of cryoglobulins s,25 or i m m u n e complexes 26-31 and, therefore, subject to tissue deposition if appropriate conditions exist. The patients in most previous studies had severe forms of liver disease, namely cirrhosis. Our studies showed that IgA renal deposits also occur with the pathologically milder focal biliary cirrhosis, in tile absence o f laboratory evidence of liver dysfunction. This suggests that severe hepatocelhdar disease is not a requisite for decreased clearance of polymeric IgA and implies that regurgitation in distorted and stagnant biliary pathways may account for re-entry of the immnnoglobulin into the serum. This view is supported further by the recent report of elevation of secretory IgA levels in the sera of patients with liver disease. 17 Regurgitation is also s u p p o r t e d by the 1245
HUMAN PATHOLOGY
Volume 16, No. '12 (December 1985)
r a p i d rise in s e r u n t levels o f I g A a n d tissue d e p o s i t i o n w i t h i n a few days a f t e r l i g a t i o n o f e x t r a h e p a t i c bile ducts, with r e v e r s a l o f this process o n t-elease o f t h e o b s t r u c t i o n . 32 T h e c o n s i s t e n t lack o f s e c r e t o r y c o m p o n e n t associated with t h e I g A tissue d e p o s i t s r e p o r t e d i n p r e vious w o r k a n d the c u r r e n t s t u d y s u g g e s t s t h a t m o s t o f this I g A r e v e r t s to tlte c i r c u l a t i o n b e f o r e t r a v e r s i n g t h e bile d u c t s .
Acknowledgments. T h e attthors express their gratitude to Mary J o h n s o n and Robert Langlotz for their help in tire preparation of this study. REFERENCES 1. Bene MC, Faure G, Duheille J: IgA nephropathy: characterization of the polymeric nature of mesangial deposits by in vitro binding of free secretory component. Clin Exp Immunol 47:527, 1982 2. Kobayashi K: Studies on human secretory IgA. Comparative studies on the IgA bound and free secretory component. Innnunochemistry 8:785, 1971 3. Oppenheimer Ett, EsterlyJR: Pathology of cystic fibrosis. Review of the literature and comparison with 146 autopsied cases. Perspect Pediatr Pathol 2:24 l, 1975 4. Luna LG: Manual of I listologic Staining Methods of the AFIP, ed 3. New York, McGraw Hill, 1968 5. Davis CA, Abramowsky CR, Swinehart G: Circtdating iummne complexes and the nephropathy of cystic fibrosis, llum Pathol 15:244, 1984 6. Berger, J, Yaneva H, Nabarra B: Glmnerular changes its patients with cirrhosis of the liver. Adv Nephrol 7:3, 1977 7. Woodroofe AJ: IgA, glomerulonephritis and liver disease. Aust NZJ Med ll:109, 1981 8. Nochy D, Callard P, Bellon B, et al: Association of overt glomerulonephritis and liver disease: a study of 34 patients. Clin Nephrol 6:422, 1976 9. Nakamoto Y, Iida II, Kobayaski K, et al: Ilepatic glomerulonephritis: characteristics of hepatic IgA glomerulonephritis as the major part. Virchows Arch [A] 392:45, 1981 I0. Endo Y, Matsushita H, Nozawa Y, et al: Glomerulonephritis associated with liver cirrhosis. Acta PatholJpn 32:333, 1983 11. Lomax-Smith JD, Zabrowarn)' LA, Howarth GS, et al: The immunochemical characterization of mesangial IgA deposits. Am J Pathol 113:359, 1983 12. Kater L, Jobsis AC, Baart de la Faille-Kuyper EIt, et ah Alcoholic hepatic disease: specificity of IgA deposits in liver. AmJ Clin Pathol 71:51, 1979 13. Swerdlow MA, Chowdhury LN: lgA subclasses in liver tissues in alcoholic liver disease. Am J Clin Pathol 80:283, 1983 14. Chandy KG, Hubscher SG, Elias E, et ah Dual role of the liver its regulating circulating polymeric lgA in man: studies on patients with liver disease. Clin Exp lmmunol 52:207, 1983 I 1. Lomax-Smith JD, Zabrowarny LA, llowarth GS, et al: The immunocllemical characterization of mesangial IgA deposits. AmJ Pathol 113:359, 1983 12. Kater L, Jobsis AC, Baart de la Faille-Kuyper EH, et al: Alcoholic hepatic disease: specificity of IgA deposits in liver. A m j Clin Pathol 71:51, 1979 13. Swerdlow MA, Chowdhury LN: IgA subclasses in liver tissue in alcoholic liver disease. Am J Clin Pathol 80:283, 1983
14. Chandy KG, llubscher SG, Elias E, et al: Dual role of the liver in regulating circnlating pol)meric IgA ill man:'studies on patients with liver disease. Clin Exp hnmunol 52:207, 1983 15. Delacroix DL, Elkon KB, Geubel AP, et al: Changes in size, subclass, and metabolic properties of serum imnumoglobt,lin A in liver disease and in other diseases with Ifigh serum imntunoglobuIinA. J Clin Invest 71:358, 1983 16. Newkirk MM, Klein bill, Katz A, et ah Estimation of pol).meric IgA in lmtnan serum: an assay based on binding of radiolabeled human secretory component with applications in the study of IgA nephropathy, IgA monoclonal gammopathy, and liver disease. J Immunol 130:1176, 1983 17. Farris MA, Chandy KG, Ashdown K, et al: Serum secretory component: a potential marker of biliary obstruction. Clin Chim Acta 133:65, 1983 18. Brandtzaeg P, Prydz II: Direct evidence for an integrated function of J chain attd secretory component in epithelial transport of inmmnoglobulins. Nature 311:71, 1984 19. Fisher M*I, Nag) B, Bazin II, et al: Biliary transport of IgA: role of secretory component. Proc Natl Acad Sci 76:2008, 1979 20. I'eppard J, Orlans E, Payne AWR, et al: The elimination of circulating complexes containing polymeric IgA by excretion in the bile. Immunol 42:83, 1981 21. Socken DJ, Simms ES, Nag} BR, et al: Secretory componentdependent hepatic transport of IgA antibody-antigen complexes. J Immunol 127:316, 198 I 22. Delacroix DL, ttodgson HJF, Mcl'herson A, et al: Selective transport of polymeric immunoglobulin A in bile. J Clin Invest 70:230, 1982 23. Nagura II, Smith PD, Nakane PK, et ah IgA in Imman bile and liver. J hmntmol 126:587, 1981 24. Nagura II, Tsutsumi Y, llasegawa II, et al: IgA plasma cells in biliary mucosa: a likely course of locally synthesized IgA in human hepatic lille. Clin Ex]~ Immunol 54:671, 1983 25. Kaufman RL, Hoers JC, Quismorio FP: Immune complexes in the portal and systenfic circulation of patients with alcoholic liver disease. Clin Immtmol Immtmopathol 22:44, 1982 26. Sancho J, Egido J, Sanchez-Crespo M, et al: Detection of monomeric and polymeric IgA containing imnttme com" plexes in serum and kidney from patients with alcoholic liver disease. Clin Exp lmmunol 47:327, 1981 27. Sancho J, Egido J, Rivera F, et al: Immune complexes in IgA nephropathy: presence of antibodies against diet antigens and delayed clearance of specific polymeric IgA immune complexes. Clin Exp Immtmol 54:194, 1983 28. Brown TA, Russell MW, Mestecky J: llepatobiliary transport of IgA immune complexes: molecular and celhdar aspects. J hnmunol 128:2183, 1982 29. Burns J, D'Ardenne AJ, Morton JA, et al: Immune complex nephritis in alcoholic cirrhosis: detection of Mallory body antigen in complexes by means of monoclonal antibodies to Mallor)" bodies. J Clin Pathol 36:751, 1983 30. Emancipator SN, Gallo GR, Razaboni R, et ah Experimental cholestasis promotes the deposition of glomerular IgA immune complexes. Am J Pathol 113:19, 1983 31. Ohshio G, Furukawa F, Sekita K, et al: IgA containing circulating immune complexes and IgA anti-single stranded DNA antibodies in patients with obstructive jaundice. Clin Exp lmmunol 59:435, 1985 32. Cuadrado E, Arenas j I , Echaniz P, et al: Rapid decrease of secretory lgA serum levels in extrahepatic obstructive jaundice after surgical relief of the bile duct obstruction (letter). Gastroenterology 84:203, 1983
1246