Secondary membranoproliferative glomerulonephritis

Kidney International, Vol. 47 (1995), pp. 643—656

NEPHROLOGY FORUM

Secondary membranoproliferative glomerulonephritis Principal discussant: HELMUT G. RENNKE Brigham and Women's Hospital, Boston, Massachusetts

a normal differential count; hematocrit, 27.2%; and a platelet count of 171,000/mm3. The serum sodium was 132 mEq/liter; potassium, 4.6

Editors

mEq/liter; chloride, 97 mEq/liter; and bicarbonate, 19 mM/liter. The BUN was 121 mg/dl and the serum creatinine, 4.6 mg/dl. The glucose level was 176 mg/dl. The alanine aminotransferase was 76 U/liter (reference range, 7 to 52 U/liter); aspartate aminotransferase, 24 U/liter (reference range, 9 to 30 U/liter); and lactic dehydrogenase, 418 U/liter (reference range, 107 to 231 U/liter). Total serum protein was 5.1 Wdl with a serum albumin of 3.5 gldl. The serum calcium was 8.7 mg/dl. Total cholesterol was 149 mg/dl. The prothrombin time was 12,5 sec and the partial thromboplastin time, 23.6 sec. Urinalysis disclosed a pH of 5, and dipstick evaluation revealed 3+ protein and 3+ blood. The urinary sediment contained red blood cells too numerous to count, with numerous dysmorphic forms and occasional white blood cells. Several serologic tests were ordered, and the patient was hospitalized for a renal biopsy. A percutaneous renal biopsy was performed under local anesthesia and

JORDAN J. COHEN JOHN T. HARRINGTON

NtcorAos E. MADIAS

Managing Editor CHERYL J. Zu5MAN

Tufts University School of Medicine

ultrasound guidance. Two cores of renal tissue were examined by the pathologist in the ultrasound suite for the presence of adequate cortical

Case presentation

tissue using a stereoscopic microscope. A small fragment of cortical tissue

A 37-year-old man was transferred to Brigham and Women's Hospital for evaluation of progressive renal failure. His medical history was notable for intravenous drug use and heavy ethanol consumption but he denied their use for the last 7 years. He had had at least one bout of alcoholic hepatitis, which was diagnosed about 8 years previously.

paraformaldehyde mixture for ultrastructural analysis. A similar fragment,

with identifiable glomeruli was immediately fixed in glutaraldehyde/

The patient originally presented to his physician 2 years ago with complaints of mild anorexia and fatigue. Workup at that time was remarkable for slightly elevated liver enzymes and a serum creatinine of 1.2 mg/dl. Six weeks later he noted the onset of periorbital edema followed

by edema of the legs and a 20-pound weight gain. He consulted the emergency room of a local hospital for these symptoms one year ago,

protease inhibitors, subsequently was processed for direct immunofluorescence microscopy. The remaining tissue was fixed in buffered formaldehyde solution and processed for conventional light microscopy. The sample processed for light microscopy included renal cortex and

outer medulla; 8 glomeruli were present. The tufts of all glomeruli appeared enlarged and showed prominent hypercellularity (Fig. IA). Most cells within the tuft were mononuclear, but several neutrophils also were

identified, The inflammatory cells appeared within the confines of the capillary loops; others were located in the expanded mesangium. In the sections stained by the PAS technique, several loops showed thickened

where he was noted to have elevated liver enzymes and a serum creatinine that had risen to 1.7 mg/dl. Urinalysis revealed 4+ protein and 2+ blood by dipstick; microscopic examination of the urinary sediment revealed red blood cells too numerous to count, 5 to 10 white blood cells/high-power

capillary walls and double basement membranes (Fig. 18). Tubules generally were well preserved; 2 small foci showed thickening of the

field, and numerous hyaline and granular casts. He was referred to a nephrologist. At physical examination performed by the nephrologist 2 days later, his pulse was 72 beats/mm; blood pressure, 160/110 mm Hg; and respirations were normal. No oral lesions were present, and funduscopic examination showed no abnormalities. The jugular venous pressure was estimated at 9

cm H20, and the thyroid gland was palpable without nodules. The chest was clear; cardiac examination revealed a normal rhythm with a soft S4. A grade II/IV systolic ejection murmur was audible. The abdomen was distended and soft but without tenderness, and no masses were palpable; bowel sounds were normal. The spleen was not detectable, but the liver edge was palpable 4 cm below the right costal margin. The extremities were remarkable for 3 + pitting edema up to the mid-portion of the calf. The skin revealed a discrete, papular, erythematous rash. Laboratory data included a white blood cell count of 30,200/mm3 with

The Nephrology Forum is funded in part by grants from Amgen, Incorporated; Merck & Co., Incorporated; Marion Merrell Dow, Incorporated; Dialysis Clinic, Incorporated; and R & D Laboratories.

© 1995 by the International Society of Nephrology

placed in transport medium containing a buffered salt solution and

tubular basement membranes accompanied by atrophy of the epithelium. The interstitium was expanded in these areas, in part by connective tissue and by minimal infiltration of mononuclear inflammatory cells. Small arteries and arterioles revealed prominence of the endothelium, and rare inflammatory cells could be identified in the intima. The sample processed for immunofluorescence microscopy contained renal cortex snd 3 glomeruli. All glomeruli showed irregular and coarse granular immune deposits along the peripheral capillary wall and in the mesangium; the deposits

were reactive fur IgG (3+ to 4+/4+), 1gM (3+/4+) (Fig. 1C), and C3

(1 +/4+). A similar distribution of deposits was visualized with reagents specific for kappa and lambda light chains. Isolated glomerular capillary loops contained small aggregates of fibrin. The electron micrographs revealed capillary loops occluded by inflammatory cells and deformed red blood cells (Figs. 2A and 28). The endothelial cells appeared swollen and had lost most of their fenestrae. Electron-dense deposits were identified in

the lamina rara interna in most capillaries; the deposits had a finely granular and homogeneous appearance. Ultrastructural characteristics frequently associated with cryoglobulins such as organized cylindrical structures, short curved parallel lines, or "fingerprint-like" images could not be identified with certainty within the deposits. Several loops showed a thickened capillary wall with double basement membrane, interposition of cellular projections, and few electron-dense deposits in close apposition to the original basement membrane (Fig. 28). The epithelial cells showed

643

644

Nephrology Fomm: Secondaiy MPGN

Fig. 1. Light microscopic and immunohistochemical findings in a patient with hepatitis C-associated membranoproliferative glomerulonephntis. A Glomerulus shows global endocapillary hypercellularity; the inflammatory cells comprise mononuclear cells and polymorphonuclear neutrophils (arrows). Isolated loops show a thickened glomerular capillary wall (PAS stain). B Glomerulus illustrates a slightly more advanced stage of the injury, with frequent double-contoured or split basement membrane (single arrows) and moderate hypercellularity and expansion of the mesangium (double arrows). These abnormalities define the membranoproliferative pattern of injury (PAS stain). C Immunofluorescence micrograph obtained from sections incubated with FITC-labeled anti-human 1gM; it reveals irregular granular peripheral capillary wall deposits. Some of the deposits show a sharp outer peripheral boundary, which suggest a subendothelial location. The reactivity for IgG revealed a similar pattern and intensity (not shown), a finding that suggests the participation of an antibody (IgG) and an anti-immunoglobulin antibody with characteristics of a rheumatoid factor (1gM). The complement

component C3 also was present within these deposits (FITC-labeled anti-IgM). D Giomerulus depicts the peripheral capillary walls with the characteristic double contours or tram-track pattern (arrows) after silver-methenamine impregnation (Jones' silver stain).

focal degeneration and loss of foot process interdigitations, a change that was particularly prominent over capillary loops with thickened walls. The lesion was classified as an immune-complex-mediated glomerulonephritis with a diffuse exudative and membranoproliferative pattern of injury. The structural and immunohistochemical characteristics and the distribution of the deposits suggested the possibility of a process with circulating immune complexes, such as that seen in patients with infections and persistent or episodic antigenemia. Alternatively, and in the absence of a demonstrable systemic bacterial, parasitic, or viral infection, the lesion could be classified under the category of idiopathic membranoproliferative or mesangiocapillary glomerulonephritis, type I.

Discussion DR. HELMUT G. RENNK.E (Director, Renal Pathology Laboratory, Brigham and Women's Hospital; and Associate Professor of Pathol-

ogy, Harvard Medical School, Boston, Massachusetts): Today's patient illustrates extremely well the fallacy, and often the dangers, of equating a particular lesion or structural pattern of injury with a disease entity. In the past, before testing for hepatitis C had been developed [1], this case unquestionably would have been Additional laboratory test results included: a negative ASO titer, classified as "idiopathic membranoproliferative glomerulonephritis (MPGN), type I" [2—4], also referred to as mesangiocapillary negative ANCA serology, trace ANA, negative anticardiolipin antibodies, and negative hepatitis-B surface antigen and antibody. Hepatitis C giomerulonephritis. Several studies over the past 2 years, howserology was positive, and an HIV test was negative. The cryoglobulins ever, have uncovered an association between hepatitis C infection, were considered equivocal. The CH50 was 20 U/mi (reference range, 150 the presence of circulating immune complexes often with the to 250 U/mI), the serum level of C1q was 9 mg/dl (reference range, 35 to characteristics of cryoglobulins [5—8]; and a membranoprolifera56 mg/dl), the C3 level was 138 mg/dl (reference range, 91 to 198 mg/dl), tive glomerulonephritis [9, 10]. I am using the term membranoand the C4 level was 1 mg/dl (reference range, 26 to 83 mg/dl). The serum protein electrophoresis showed a faint monocional 1gM kappa band. proliferative glomerulonephritis here to denote a general pattern

Is'

-.

p

I

c:t

Ep71

r Fig. 2. Electron microscopic abnonnalities in hepatitis C-associated membranoproliferative glomendonephritis. A High-power electron micrograph shows characteristic subendothelial electron-deose deposits (marked by the *). The capillary lumen is occupied by a mononuclear inflammatory cell (Mon). The endothellal cell (End) is focally displaced by the deposits and a new, irregular, and incomplete basement membrane is beginning to be laid down (arrows). Ep = visceral epithelial cell; US = urinary space (conventional electron microscopy after uranyl acetate and lead citrate staining). B This glomerular capillary loop illustrates the characteristic findings of the membranoproliferative pattern at a later stage of healing. Few electron-dense deposits are still present in the expanded subendothelial space (white arrowhead). The endotheial cell is markedly displaced, and it appears to have lost many of its fenestrae. A new basement membrane is clearly visible (arrows) along almost the entire length of the markedly thickened peripheral capillary wall. A projection of a lysosome-rich cell has insinuated itself between the displaced endothelium and the original basement membrane (indicated as CI = cell interposition). CL = capillary lumen; Mes = mesangium; Hp = visceral epithellal cell (conventional electron microscopy after uranyl acetate and lead citrate staining).

646

Nephrology Forum: Secondaiy MPGN

of tissue injury seen in a variety of disease processes that share a Table 1. Conditions associated with a membranoproliferative pattern of injury common pathogenetic mechanism, and not to convey the more restricted meaning of a particular disease entity that causes the Immune-complex-mediated disease nephrotic syndrome (idiopathic MPGN, type I). Unfortunately, Idiopathic forms of MPGN or of unknown association

the precise meaning to be conveyed by the term is not always made clear either in the medical literature or in biopsy reports.

Confusion similar to that besetting the use of the term membranoproliferative glomerulonephritis also surrounds the use of other

renal pathologic terms. Thus, the term focal and segmental glomerulosclerosis is used not only to describe a particular lesion seen on renal biopsies of patients with a great number of different disease entities [11, 12] (the more generalized meaning), but also to refer to a particular pathologic entity of unknown etiology that

causes the idiopathic nephrotic syndrome [13] (the more restricted meaning). Other examples of confusing terminology are membranous nephropathy, acute or proliferative glomerulonephritis, mesangial proliferative glomerulonephritis, and crescentic glomerulonephritis. The varying usage of the terms membranoproliferative glomerulonephritis, focal glomerulosclerosis, and others often falls along lines of professional specialties: nephrolo-

gists and internists tend to use these terms with the more restricted meaning of a particular clinical disease entity. Pathol-

MPGN type I [14I MPGN type II or dense-deposit disease and partial lipodystrophy [15, 16]

MPGN type III (Strife and Anders variant, Burkholder variant) [17-19] Autoimmune diseases SLE [20] Sjogren syndrome [21] Rheumatoid arthritis [221 Inherited complement deficiencies, in particular C2 deficiency [23, 24] Chronic infections Viral: hepatitis B [25], hepatitis C and cryoglobulinemia type II (formerly called essential mixed cryoglobulinemia) [9, 10] Bacterial: endocarditis [26], infected ventriculoatrial (or jugular) shunt [27], multiple visceral abscesses [26, 281, leprosy [291 Protozoal: malaria (30), schistosomiasis [31] Other infections: mycoplasma [32] Miscellaneous Chronic liver disease (cirrhosis and alpha 1-antitrypsin deficiency) [33, 341

ogists usually apply the same names to indicate a particular Chronic and recovered thrombotic microangiopathies pattern of tissue injury. The ambiguous use of these terms often can be traced back to the time when a group of patients was split off from others that presented with a common clinical syndrome

(for example, the nephrotic syndrome) by the occurrence of a particular lesion or pattern of tissue injury (that is, focal and segmental glomerulosclerosis). This characteristic finding on bi-

opsy often would correlate with a particular clinical course,

Healing phase of HUSITFP [35] The syndrome of circulating anti-phospholipid (anti-cardiolipin) antibodies [36] Radiation nephritis [37] Nephropathy associated with bone marrow transplantation [38] Sickle cell anemia [39] and polycythemia [40] Transplant glomerulopathy [41]

Paraprotein deposition diseases

specific prognosis for the renal condition, or particular response Glomerulopathies associated with cryoglobulinemia type I [42, 43] to a therapeutic modality [2]; the newly identified clinicopathoWaldenstrflm macroglobulinemia [44] logic disease entity then was named according to its apparently Immunotactoid glomerulopathy [45, 46] Immunoglobulin light-chain or heavy-chain deposition diseases [47, unique histologic features. 48] Confusing a pattern of injury with the particular disease entity Fibrillary glomerulonephritis [45, 46, 49—52] traditionally defined by such a pattern sometimes has the unina Number in brackets refers to references. tended effect of giving a false sense of security that comes when one classifies or "pigeonholes" a given patient. Additional serologic or bacteriologic studies that would further define the disease process and its correct therapeutic approach often are omitted. This sometimes occurs in patients with a membranoproliferative such circumstances, most conditions can reliably be placed into pattern of injury secondary to a potentially treatable but con- one of three basic categories, however, depending on the findings cealed infectious process. The misunderstanding also can have of the renal biopsy (Table 1). This classification is purely descripother regrettable consequences. For instance, a systemic vasculitis tive; it is based on a combination of objective findings obtained by can be mistakenly overlooked and the patient not treated appro- light, immunofluorescence, and electron microscopy that reflect priately because the biopsy showed "focal and segmental glomer- the fundamental pathophysiologic processes underlying the tissue injury. The precise classification of a given case and the correct ulosclerosis." Such a focal scierosing lesion can result from diffuse epithelial cell injury or in the setting of secondary functional and identification of an associated condition are often only possible, structural adaptations that ensue as a consequence of a reduced therefore, after results of additional studies on the patient are number of functioning nephrons, but also as the end result of the obtained, studies often suggested by the findings of the renal

healing process that follows an active focal necrotizing and biopsy. An in-depth discussion of the clinical manifestations of all crescentic glomerulitis, as seen during the active phase of several conditions that have been associated with a membranoproliferavasculitides. In most textbooks and in reviews on MPGN, authors tradition- tive injury clearly is beyond the scope of this discussion. Suffice it ally have divided this type of injury into idiopathic and secondary to say that the renal manifestations in most patients with this form

forms that have a recognizable etiologic agent or a known associated disease [2—4]. This classification is often of limited value to the pathologist, who is confronted with this pattern of injury upon examination of the renal biopsy specimen, when the clinical and serologic workup of the patient is incomplete. Under

of injury include hematuria, proteinuria, and some degree of functional impairment. The proteinuria is of variable severity, often reflecting the extent of the structural changes in the peripheral capillary wall and the variable obliteration and simplification of foot processes. A certain percentage of patients within

Nephrology Forum: Secondwy MPGN

647

all three groups of diseases (Table 1) can present with nephrotic- deposition disease, lupus nephritis, or SjOgren's syndrome. I range proteinuria or overt nephrotic syndrome. The urine sedi- would like to emphasize, however, that the light microscopic ment contains red cells, often red cell casts, and white cells, a findings I have described are generally not unique for any of the reflection of the inflammatory component seen at some stages of disease entities listed in Table 1. the development of the disease. The acute nephritic syndrome is Immunofluorescence microscopy or immunohistochemical more frequent in patients with dense-deposit disease or early in studies usually help to define more precisely the type of disease the course of a postinfectious condition. Systemic hypertension process that has caused the histologic changes I have described.

can be observed at some stage in the course of the disease, a

Depending on the type of disease responsible for this form of

complication that is especially prominent in patients with fibrillary glomerulonephritis [45, 50].

injury, immunoglobulins, complement components, paraproteins, or products of the coagulation cascade can be demonstrated with variable intensity, usually along the peripheral capillary wall, but

Structural features

The membranoproliferative pattern of injury, defined predominantly by the abnormalities revealed by light microscopic examination, is characterized by two basic changes: diffuse expansion of

most notably also in a mesangial distribution. The presence of granular deposits of polyclonal immunoglobulins and complement components usually indicates one of the immune-complex-medi-

ated diseases. The presence of fibrin-related antigens, either in the form of occlusive microthrombi or diffusely outlining the lary wall. The expansion of the mesangium involves variable peripheral capillary wall, suggests a thrombotic angiopathy. The degrees of proliferation of resident mesangial cells, infiltration of finding of monotypic light chains or heavy chains alone suggests the area by blood-borne mononuclear inflammatory cells [28, 53], the involvement of a paraprotein in the disease process. and increased mesangial matrix. These changes can result in an The electron microscopic evaluation is often the only reliable exaggeration of the lobular pattern of the glomerulus; early way of correctly identifying the process responsible for the investigators referred to this variant of the process as "lobular glomerular disease. In the immune-complex diseases, the endoglomerulonephritis." The increase in the thickness of the periphthelial cell, normally attached to the basement membrane by a the mesangial elements, and thickening of the glomerular capil-

eral capillary wall can be segmental rather than global and diffuse in nature. Capillary loops with a thickened appearance often, but not invariably, display a characteristic split or reduplicated basement membrane; this feature gives the glomerular capillary wall a double-contoured, "tram track," or chain-like appearance. Such changes are particularly well demonstrated when the light microscopy sections are stained by Jones' methenamine-silver impreg-

nation (Fig. 1D) or the periodic acid Schiff reaction. The space between the reduplicated basement membrane, which contains immune complexes or other plasma-derived protein components and cytoplasmic extensions of mononuclear cells, does not stain with these reagents. In all classic textbooks, this latter phenomenon is referred to as mesangial cell interposition, although direct or even circumstantial evidence that would indicate that resident mesangial cells have the capacity to move into the peripheral capillary wall is lacking. It is reasonable to suggest, instead, that mononuclear inflammatory cells, which have been demonstrated in the mesangial regions and within the capillaries of the glomerulus [54] and contribute to the mesangial hypercellularity [55], also extend into the subendothelial space. These monocytes and macrophages likely accumulate in the glomerulus, attracted by chemotactic factors locally generated following the initial injury to

the endothelium or the deposition of immune complexes or paraproteins within the proximal layers of the glomerular capillary

very thin lamina rara interna, is separated from this structure by an expanded subendothelial space that contains cellular projections (interposition of mononuclear cells) and discrete electrondense deposits. A new and often irregular and incomplete base-

ment membrane is present immediately under the displaced endothelium (Fig. 2). Mononuclear and polymorphonuclear inflammatory cells often are seen within the capillaries. The chronic thrombotic angiopathies also are characterized by a widening of the subendothelium, which contains finely granular and electronlucent, "fluffy" material, occasional strands of cross-linked fibrin, platelet fragments, cell membrane debris, and other poorly identified elements; prominent interposition of cellular elements and a newly formed second basement membrane also are present with this form of injury. In the deposition diseases, the peripheral capillary wall is thickened more uniformly by irregular aggregates of fine, granular, electron-dense material (light-chain-deposition diseases), amyloid-like fibrillary deposits (fibrillary glomerulonephritis), or highly organized microtubular elements (immunotactoid glomerulopathy, Waldenstrom's macroglobulinemia, and cryoglobulinemia). Often, this material also infiltrates the basement membrane diffusely; the lamina densa appears less well defined and often is destroyed. Classic double-contoured basement membranes and cellular interposition are less obvious in the deposition disease group. Foot processes show degenerative changes that vary greatly among cases; some degree of simplifi-

wall. A membranoproliferative pattern associated with vascular lesions such as fibrin thrombi in glomerular capillaries, fibrinoid necrosis of arterial walls, or myointimal proliferation affecting cation of the normal interdigitating arrangement of the foot interlobular arteries strongly suggests one of the conditions processes and an increase in the number of cytoplasmic vacuoles, associated with a thrombotic microangiopathy. The presence of lysosomes, and cell surface microvilli are routinely observed. In all PAS-positive hyaline masses occluding the lumen of glomerular cases with a membranoproliferative pattern of injury, one obcapillaries or small arteries and arterioles usually indicates in- serves signs of endothelial cell damage with loss of fenestrae and volvement by cryoglobulins or other paraproteins. The participa- swelling of the normally tenuous cell body. The endothelial cell tion of light chains in this pattern can be documented reliably by also can be missing altogether during the more active phase of a the use of light-chain-specific antibodies, which probably should thrombotic angiopathy. The mesangium is also expanded by be used routinely during the evaluation of a renal biopsy by electron-dense deposits or fibrillaiy material, a variable increase immunofluorescence microscopy. Significant interstitial fibrosis or in amorphous or finely fibrillar extracellular matrix material, and inflammation is often seen in patients with systemic light-chain an increase in cellular elements.

648

Nephrology Forum: Secondaiy MPGN

Pathogenesis

infectious process. Complement components are invariably found

within the glomerular capillary wall in these patients. One can Let me repeat: three fundamentally different pathologic pro- assume that various peptide fragments, such as C3a and C5a, or cesses can result in a membranoproliferative pattern of glomeru- heteromultimers that result from complement activation, such as lar injuly: disease processes with circulating immune complexes, C5-9, are locally generated and act respectively as chemotactic thrombotic angiopathies, and some paraproteinemias and depo- factors for inflammatory cells or as mediators of direct injury to sition diseases (Table 1). The basic pathologic processes common the cell membrane of endothehial, visceral epithelial, and mesanto all the diseases associated with this pattern of injury include gial cells. We have gained many insights into the pathogenesis of these first, a form of active injury to the more proximal layers of the glomerular capillary wall and the mesangium, often accompanied disorders over the past few decades from the analysis of various by active inflammation, and second, a phase of cellular prolifera- models of experimental serum sickness. Most active models tion and repair that results in expansion of the mesangial matrix involve single or repeated injections of a foreign protein that serves both as an immunogen and a target for the subsequently and reduplication of the basement membrane. During the active phase of the injury, the endothelial cell often produced antibodies. Immune deposits in the tissues result from is directly damaged by toxins or products of complement activa- entrapment within the walls of arteries and capillaries of immune tion and by the ensuing accumulation and infiltration of blood- complexes assembled in the circulation during the phase of active borne inflammatory cells. Although inflammatory cells are rea- immune elimination of the circulating antigen [58]. Alternatively, sonably efficient in removing immune complexes from the more immune complexes in the tissues also can result from the interproximal layers of the glomerular capillary wall [561, in some action of the foreign antigen planted at such sites during the capillary loops immune complexes trapped upstream of the pre-immune phase and antibodies delivered to such sites after the lamina densa may prevent the regenerating endothelium from immune elimination of the antigen is completed and free antibodmaking appropriate contact with the original basement mem- ies are available in circulation and in the tissues. Immune brane. Similarly, the regenerating endothelium may be prevented complexes within the lamina rara interna and in the mesangium from reaching the necessary matrix components normally present also can be induced by sequential administration of antigens and in the lamina densa of the basement membrane by paraproteins specific isologous or heterologous antibodies [56, 59, 60]. Some deposited in the lamina rara interna, fibrin, cell, and platelet glomerular abnormalities also can be reproduced in animals by debris, or other products produced during activation of the the administration of immune complexes produced in the test coagulation cascade. Under such adverse conditions, the endothe- tube by the interaction of antigens of defined physicochemical hal cell is stimulated to generate a new basement membrane characteristics and specific antisera or purified fractions of immuseparated from the original lamina densa by the deposited noglobulins; these experimental diseases constitute the passive material and trapped inflammatory cells. The endothelium's models of serum sickness [61]. The glomerular capillary wall functions as an effective size- and ability to generate a basement-membrane-like structure has been well documented in cell culture experiments. charge-selective filter [62, 63]. Negatively charged residues are At different stages of the disease process, either the inflamma- concentrated predominantly on endothelial and epithelial cell surfaces and in both laminae rarae [62]. Antigens, antibodies, and immune complexes with positively charged residues therefore can larity and active inflammation (the proliferative or exudative interact with these negatively charged structural elements of the component of the pattern), while the glomeruhi during later stages capillary. Using ultrastructural tracer studies, several investigators show more signs of tissue repair. This latter phase is evidenced by have also concluded that the lamina densa and the filtration slit thickening of the peripheral capillary wall with characteristic diaphragms can restrict the filtration of macromolecules on the double-contoured or split basement membranes and an expansion basis of their size [62, 64]. One can reasonably assume, therefore, of the mesangium by infiltrating mononuclear cells and prolifer- that several molecular characteristics of the antigen, the antibody, ated resident mesangial cells and by an increase in extracellular or the immune complexes are important factors that determine the site of localization of the deposits. The site of immunematrix. Immune-complex-mediated diseases. The active tissue injury in complex deposition or assembly in turn determines the clinical immune-complex-mediated diseases occurs after in-situ formation and structural expression of the disease. Molecular characteristics of immune complexes or deposition by entrapment of pre-formed, that are relevant in this context include the charge and the circulating complexes within the lamina rara interna or subendo- effective filtration size of the antigen or the pre-formed comthelial layer of the glomerular capillary wall. Most patients plexes, the antibody affinity and avidity, and other physicochemiprobably suffer from persistent or episodic antigenemia derived cal characteristics that endow some antigens with special affinity from an infectious agent. A similar process of immune-complex for extracehlular matrix components or with lectin-hike properties. A substantial body of often conflicting experimental data deals deposition also probably occurs in the idiopathic membranoproliferative glomerulonephritides types I and III, although we have with the effect of the molecular charge of antigens, antibodies, and no clue as to the nature or source of the antigens involved in these immune complexes on the resultant disease. The bulk of the diseases. As Appel recently pointed out in an editorial comment experimental evidence indicates that models involving sequential [57], the precise identity of the antigens involved in any of the administration of cationic macromolecular antigens and specific glomerular diseases that occur in association with a known antibodies and acute serum sickness models induced with cationic tory attack or the repair process can dominate. As a result of these two activities, some renal biopsies demonstrate more hypercellu-

infectious process is still unknown; most secondary forms of antigens produce a membranous pattern of injury with diffuse membranoproliferative glomerulonephritis are defined exclusively subepithehial electron-dense deposits [56, 65, 66]. Active inflam-

by the frequency of the temporal association with a particular

mation can occur transiently during the first few hours after

Nephrology Forum: Seconda.'y MPGN

649

form under these circumstances can be expected to be large, as they are formed continuously in antibody excess. An identical daily antigenic load given as a single intravenous bolus injection yields a typical membranous pattern of injury when the cationic antigen is used. Intermittent (daily bolus) injection of native, negatively charged bovine serum albumin produces a membranoproliferative type of injury with subendothelial and mesangial electron-dense deposits, focal cellular interposition in the peripheral capillary wall, and basement membrane splitting [78]. The lack of immune-complex deposition and subsequent injury deposits show no signs of active inflammation; a potential gradient in the tissues in experimental settings that favor large immuneof complement-derived chemotactic agents (C3a and C5a) is not complex formation in the circulation has largely been ascribed to likely to be generated upstream given the high unidirectional an effective removal of circulating complexes by the mononuclear convective flow across the glomerular capillary wall driven by the phagocyte system of the liver, spleen, and bone marrow. This imbalance of Starling's forces [56, 69]. The absence of an inflam- clearing function is greatly assisted in rodents by platelets and in matory response in disease models of the membranous type with primates by erythrocytes and leukocytes with CR1 complement immune complexes present exclusively in the lamina rara externa receptors [80]. These cells effectively shuttle immune complexes

induction of the disease but is generally absent. Clinically these models are characterized by significant proteinuria and a normal glomerular filtration rate [56, 65]. Removal of these subepithelial deposits is slow, and the proteinuric phase lasts several months after a single episode of immune-complex deposition [56]. In this setting, the damage seen in the visceral epithelium is likely the result of complement activation and generation of C5-9 near the podocytes, as demonstrated for the Heymann model of membranous nephropathy by Salant and Couser and their collaborators [67, 68]. Immune-complex-disease models with epimembranous

also could be ascribed to a failure of the endothelium to be between the peripheral circulation and the organs that are actively engaged in the process or to sustain significant damage, because immune complexes and complement activation occur at a

responsible for immune-complex degradation. Immune complexes of intermediate size usually remain within

distance in the lamina rara externa of the glomerular capillary the proximal layers of the glomerular capillary wall and in the wall. Factors that could upregulate specific inflammatory adhe- mesangium [81] and potentially could induce an inflammatory sion molecules on the endothelium [70] are not likely to reach, in response. Only the smallest immune complexes assembled in a retrograde fashion, the inner layers of the capillary wall. The antigen excess or prepared with antibodies of low affinity can filter lack of glomerular inflammation is also in keeping with the beyond the lamina densa and form epimembranous electronrelatively bland urinary sediment observed in patients with a dense deposits [82]. It has been argued, however, that all subepithelial deposits result from an in-situ assembly of complexes [83] membranous pattern of injury. Active serum sickness models that utilize negatively charged and that small, soluble immune complexes trapped in the lamina antigens [65, 71] and passive models induced with pre-formed rara interna are likely to dissociate before traversing the lamina immune complexes with weakly cationic, anionic, or heteroge- densa and re-associating in the lamina rara externa. In a series of neous molecular charge [72] result in mild glomerular pathology elegant studies, Gallo and collaborators have clearly shown that with electron-dense deposits seen predominantly in the mesan- small, cross-linked cationic immune complexes, and hence nongium and, more rarely, in the subendothelial space. Dense depos- dissociating complexes, can filter beyond the lamina densa and its within the subendothelial layer of the capillary wall often result induce a membranous form of injury [72, 84]. These findings are from an in-situ mechanism of immune-complex formation. Such in keeping with the observation that cationized ferritin with a immune complexes can be induced with greater consistency with molecular weight close to 500 kD and equivalent in size to an antibodies specific for local, endothelium-derived antigens [73] or Ag2:Ab2 immune complex filters across the lamina densa and with antibodies directed against an exogenous antigen with lectin- accumulates in the lamina rara externa [62]. It is also conceivable that several molecular characteristics— like activity planted on or near the endothelium [60, 74]. In-vitro studies also have shown a special binding capacity of DNA for such as charge, size, and affinity of the antigen or the circulating collagen molecules [75]; such interactions also could occur be- immune complex for components of the glomerular capillary tween the glomerular capillary wall and viral RNA present in the circulating immune complexes detected in patients with hepatitis C-associated cryoglobulinemia [6, 9, 76], the presumed cause of renal injury in today's patient.

wall—act in concert to determine the likelihood that deposits will form in the lamina rara interna. Antigens and immune complexes

that localize in the subendothelial layer of the basement membrane are likely to have a net positive charge or a strong binding capacity for elements of the lamina rara interna. Such character-

The size of immune complexes is in direct proportion to the molecular weight of the reactants and the number of molecules istics would be essential to overcome the effective charge-selective within the complex; the latter, in turn, is determined by the barrier offered by the negatively charged endothelium. Further-

relative concentration or ratio of antigen to antibody in the more, such antigens or immune complexes also need be of solution and in the circulation. The level of antibody in the relatively large size to be restricted from filtering beyond the circulation is a function of the animal's immune response and lamina densa. The antigen or an immune complex initially deposantibody production and the rate of antibody disappearance due to binding to circulating or tissue-bound antigen. Animals with high levels of antibody production, and presumably with large circulating immune complexes, suffer minimal or no renal disease [77]. Preliminary studies from our laboratory performed in an active serum sickness model using cationic and native bovine serum albumin in rats have shown that continuous infusion of the

antigen via osmotic minipumps does not result in immunecomplex deposition in the kidney [78]; immune complexes that

ited in the more proximal layers of the glomerular capillary wall then could interact with additional free antigen, antibodies, or antigen-antibody complexes [85], or bind anti-idiotypic antibodies and rheumatoid factor (antibodies, usually of 1gM class, directed against the Fe portion of IgG) and elements of the complement system.

Immune complexes deposited or generated in the more proximal layers of the glomerular capillary wall, that is, on or near the endothelium and in the subendothelial space of the glomerular

650

Nephrology Fonsm: Seconda,y MPGN

capillary wall, usually elicit a brisk but transient inflammatory lesion often associated with these conditions has a striking resemresponse [56]. Animal models with subendothelial immune depos- blance to the changes seen in diabetic nephropathy [47]; thickits also are characterized by an early phase of platelet accumula- ened basement membranes are routinely seen, but duplicated or tion [56, 86], significantly less (or no) proteinuria compared with double-contoured basement membranes are infrequent. In many patients with this category of disorders, the overpromodels of subepithelial immune deposition, and often an early phase of diminished glomerular filtration rate [56]. The removal duction of light chains, heavy chains, or a fully assembled but of subendothelial deposits is usually swift, occurring within a few abnormal immunoglobulin is the result of a plasma cell dyscrasia hours or days in disease models induced by a single episode of or an overt multiple myeloma or B-cell lymphoma; in others, a neoplastic source is not immediately apparent at the time of the immune-complex formation [56]. The idiopathic type-Il membranoproliferative glomerulone- renal involvement [47]. It is not clear why only a small proportion phritis, or dense-deposit disease, is in many aspects a completely of patients who overproduce light chains manifest systemic depdifferent process than the immune-complex-mediated entities I osition diseases. Furthermore, the molecular characteristics that just described. It is an autoimmune disease with circulating would predict whether a given light chain is amyloidogenic (most antibodies directed at C3bBb, the alternate pathway C3 conver- often lambda chains) or would result in non-fibrillary, fine, and tase; the autoantibody is usually an IgG, also known as the C3 irregular dense deposits (most often kappa chains) are unknown nephritic factor. The antibody stabilizes the enzyme, prevents its [48, 951. The biochemical and molecular analysis of light chains normal degradation, and increases complement activation and from patients with deposited paraproteins has not yielded a consumption. In addition, there is a decrease in hepatic synthesis unique structure or sequence that would explain a potential of complement of poorly understood pathogenesis. Electron- affinity of these light chains for components of the extracellular dense material is deposited in the basement membranes of matrix, and the basement membrane in particular (reviewed in glomeruli, tubules, and vessels. The nature of the dense material Ref. 96). Likewise, the mechanisms and mediators or growth factors involved in the proliferation of cells and the overproducdeposited in the capillary is unknown [87]. Chronic thrombotic microangiopathies. The membranoprolifera- tion of matrix components remain unexplored. The nature and sources of the paraproteins responsible for tive pattern of injury also develops in some patients during the recovery phase of acute thrombotic microangiopathies or during fibrillary glomerulonephritis are not known. Its structural features the course of some of the more indolent forms of these disorders and clinical associations have led to the suggestion that this entity (see Table 1). The term "thrombotic microangiopathy" is used is distinct from immunotactoid glomerulopathy [46, 97, 98]. here only to indicate a type of lesion that occurs in several Others have argued strongly that these two disorders merely conditions having a common pathophysiologic mechanism, rather than to define a particular clinical entity. The pathophysiology of these processes is somewhat confusing [88]. An important step in the pathophysiology is platelet activation and consumption. The mechanism by which this occurs is not completely understood. In general, three basic mechanisms have been postulated as operating in these clinical settings: endothelial cell injury, the presence

of a platelet aggregator in the circulation, and a reduction in a normally occurring platelet inhibitor. Direct injury to the endothelial cell, or indirect injury via neutrophil activation [89], is thought to be responsible for the injury in the childhood form of the hemolytic-uremic syndrome (HUS) associated with E. coli infections (particularly 0157:H7 type), radiation nephritis, the nephropathy associated with bonemarrow transplantation, and several forms of cancer and antirejection therapy. An immunologic basis for endothelial cell injury also has been proposed, in particular for the conditions associated with organ rejection and in some children with HUS [90]. Some patients with these disorders have an increased circulating level of a platelet aggregating factor, such as unusually large von Willebrand factor [91], a circulating cysteine protease (calpain) [92], or increased levels of plasminogen activator inhibitor, type 1 [93].

represent different expressions of a similar biochemical abnormality [45, 99]. The deposits in fibrillary glomerulonephritis usually are reactive for polyclonal IgG [45, 46, 49, 52, 100]. Although the

fibrils in this condition are thought by some to derive from an abnormal immunoglobulin, the infrequent observation of an associated plasma cell dyscrasia strongly argues against this possibility [46, 50]. Such an association, however, does seem to exist in patients with immunotactoid glomerulopathy [46, 101]. The relatively constant finding of IgG4 in the glomerular deposits in a recently published series of patients with fibrillary glomerulonephritis again has raised the possibility of a neoplastic B-cell

origin of the protein [52]. The fibrillary deposits, however, are usually not monoclonal, as both lambda and kappa reactivities are present simultaneously. The alternative possibility—that IgG4, a negatively charged immunoglobulin, binds avidly to fibrils on the basis of charge interactions—also should be considered seriously. A strongly cationic fibrillary component also would explain the high predilection of the deposits for the glomerular capillary wall. Summary

The term membranoproliferative glomerulonephritis has been

Finally, a decreased level of a normal inhibitor of platelet used in the literature and in practice to convey several different aggregation also could explain the occasional reversal of the meanings: it can describe a pattern of structural glomerular disease in patients after plasma infusion [94]. Paraprotein deposition diseases. A membranoproliferative pattern of injury is also a common structural finding in paraproteinemias with renal involvement (see Table 1). In renal biopsies of patients with these conditions, the glomeruli often show prominent expansion of the mesangium by excess matrix accumulation, mild to moderate mesangial hypercellularity, and variable thick-

abnormalities or refer to an immune-complex-mediated disease process of unknown cause, often identified as idiopathic membranoproliferative glomerulonephritis. The structural changes that characterize this form of injury include an increase in the

ening of the peripheral capillary wall. The distinctive nodular

of the lobular glomerular pattern and formation of distinctive

number of cells within the glomerular capillaries or in the mesangium and an increase of the extracellular matrix. These changes often result in expansion of the mesangium, accentuation

Nephrology Forum: Secondaiy MPGN

nodules, and the production of double-contoured basement membranes. These structural glomerular abnormalities often complicate the course of diseases of totally diverse pathogenesis with different clinical manifestations. Common to all these processes is an acute or persistent injury to the more proximal layers of the

651

toid factor" but without the ability to induce the immune complex to precipitate in the cold. Alternatively, it is also likely that some

patients respond to the hepatitis C virus infection without producing this anti-immunoglobulin 1gM antibody. Such patients then could develop an immune-complex disease similar in nature to the glomerulonephritis seen in the course of other chronic or persistent infections. Such patients would not have detectable circulating cryoglobulins and would not present with the unique

glomerular capillary wall and the mesangium with subsequent healing or repair. A component of active inflammation or microvascular thrombosis is present during the early phases of the disease process, while an increase in extracellular matrix and clinical findings associated with cryoglobulinemia or the peculiar basement membrane reduplication characterizes the more estab- ultrastructural features within the electron-dense deposits. lished and chronic lesions. These changes are nonspecific at the DR. HARRINUTON: I am fascinated by your studies showing that continuous infusion of bovine serum albumin resulted in no light microscopic level. Three groups of diseases should be considered in the differen- detectable structural abnormalities. Recently several papers have tial diagnosis of patients whose biopsy findings reveal a mem- suggested that immunoglobulin infusions could benefit various branoproliferative pattern of glomerular injury. (1) Immune- glomerular diseases. A recent article in fact argued that treatment complex-mediated diseases. Most notable in this group are several with high-dose immunoglobulin favorably affected Henochchronic or persistent infectious processes that result in glomerular SchOnlein purpura [102]. Can you connect your experimental lesions indistinguishable from those in patients with idiopathic observation with those preliminary, clinical results? DR. Reri: It has been proposed that patients with IgA type-I membranoproliferative glomerulonephritis. (2) Chronic nephropathy have an underlying relative immune deficiency state, and healed thrombotic microangiopathies. (3) Paraprotein deposition diseases. The precise diagnosis of these diseases often as evidenced by a partial deficit of IgG antibodies (mainly requires thorough immunohistochemical and ultrastructural stud- affecting the IgGI subclass) [102] and a high renal allograft survival rate [103]. The administration of immunoglobulins is ies or additional serologic and chemical laboratory tests. intended to correct this deficit and modify the basic mechanism Questions and answers underlying the disease process. Some have suggested that IgA DR. JOHN T. I-IARRINGT0N (Academic Dean, Tufts University nephropathy could result from the deposition in the mesangium of immune complexes that develop in response to an uncontrolled School of Medicine, Boston, Massachusetts): You began by discussing patients with hepatitis C and cryoglobulinemia. What is the mucosal immune response to chronic exposure to an environmen"gold standard" for making a diagnosis of a disease due to tal or dietary antigen. If one assumes that the aforementioned cryoglobulinemia? Is the diagnosis based on serum findings, or is suggestions are correct, I could argue that immunoglobulin therit based on tissue examination? My followup question is: In apy results in the formation of large IgG complexes (formed in patients with hepatitis C and MPGN who have neither tissue nor great antibody excess) that can effectively be cleared by the serologic evidence of cryoglobulins, do you think we're simply reticuluni/phagocyte system and thereby result in removal of the offending antigen(s) from the circulation. The deposition of missing the ciyoglobulins? DR. RBNNKE: The issue of a "gold standard" for making the IgA-containing immune complexes then could be avoided. There diagnosis of a disease due to cryoglobulinemia has not been is, of course, no direct evidence for this hypothesis. DR. AJAY K. SINGH (Division of Nephrology, New England resolved entirely. In part, this is due to a scarcity of longitudinal serologic results and a lack of ultrastructural examination of the Medical Center, Boston, Massachusetts): As you know from Johntissue in a sufficient number of patients that would allow us to son's work in Seattle and Border's work in Utah, in the Thyl.1 correlate these findings. The reports in the literature often deal model of mesangioproliferative glomerulonephritis, binding of either with the serologic and clinical findings or the morphologic IgG autoantibody to Thyl.1 on the surface of mesangial cells leads manifestations, but seldom with both. I believe a tissue diagnosis to mesangial cell injury and the induction of a variety of growth is not always necessary for us to make the diagnosis of a factors, including PDGF and TGF/3. What can you tell us about cryoglobulinemia-induced renal disease, particulary when con- the profile of growth factors when the endothelium is the initial vincing serologic documentation is available in a patient with a target of immunologic injury, as in MPGN? Specifically, have clear and consistent clinical picture. I also have seen renal biopsies there been in-situ hybridization analyses of cytokine gene expreswith ultrastructural findings characteristic for cryoglobulins and a sion in renal biopsies from patients with MPGN? Dn. RENNKE: Most of the experimental work you mention has negative cryoprecipitation test done shortly after the tissue had

been obtained. In such cases I often have assumed that the concentrated on a model of mesangial injury induced by cytotoxic appearance of the cryoglobulin in the circulation must have been antibodies. I am not aware of any study in human membranoprotransient. Your followup question has not been answered com- liferative glomerulonephritis or in an appropriate animal model pletely either, because the association among hepatitis C infec- that has looked for the expression of growth factors or cytokine

tion, cryolobulinemia, and glomerulonephritis has only been production. The exact source of cytokines and growth factors recognized recently. However, I do not think we are simply could be difficult to pinpoint in such a study, as platelets and missing the cryoglobulins in the serum of some patients with monocytes/macrophages are important participants in the disease hepatitis C-associated glomerulonephritis. I cannot discount the process. Under appropriate stimulatory conditions, these cell possibility that in some patients the production of cryoglobulins elements release a number of cytokines and growth factors. The might not reach a level in the serum that would be easily detected relative contribution of the intrinsic glomerular cells in the by standard laboratory techniques. Other patients might produce production of such factors might be obscured, unless cumbersome a monoclonal 1gM antibody with the characteristics of "rheuma- electron microscopic studies are performed.

652

Nephrology Forum: Seconda,y MPGN

DR. ANDREW S. LEVEY (Division of Nephrology, New England composition of most antigen-antibody complexes that localize in Medical Center): Your definition of the membranoproliferative the glomerulus, as emphasized recently by Appel [571. pattern emphasizes the endothelial injury and the subendothelial The pathogenicity of the autoantibody associated with densesite of deposits or accumulation of material. What term do you deposit disease, the C3 nephritic factor, was believed to relate to use to classify the pattern of injury frequently observed in patients the capacity of this antibody to block the enzymatic function of the With lupus in which the renal lesion is characterized by the C3 convertase of the alternative complement pathway. How such superimposition of a proliferative and a membranous lesion? a biochemical defect with its ensuing consequences on compleDR. RENNKE: The situation you mention occurs relatively ment activation could result in tissue injury so characteristic for frequently. The clinical course in such patients is often character- this disease is entirely unknown. An animal model for denseized by many months or even years of proteinuria or the nephrotic deposit disease does not exist. Several reports of conditions with syndrome with a bland urinalysis followed by a more recent detectable C3 nephritic factor but without nephritis have apdecline in glomerular filtration rate and the appearance of peared in the literature; these include herpes gestationis [104], significant hematuria and red cell casts in the urine sediment. I meningococcal meningitis, and chronic urticaria with mild leuko-

personally refer to such a process as "lupus nephritis with a cytoclastic vasculitis [105, 106]. It even has been suggested that the membranous and proliferative pattern of injury." In such cases we find not only prominent and diffuse subepithelial electron-dense deposits, as one expects in a membranous pattern of injury, but

appearance of C3 nephritic factor is a normal physiologic event

England Medical Center): In some patients with type-I MPGN and most of those with type-TI, a renal biopsy reveals only complement but no immunoglobulin. What is the significance of this observation regarding pathogenesis?

cannot do this, and the infection persists in the liver with

[107].

DR. BRJ PEREIRA (Division of Nephrology, New England also subendothelial deposits accompanied by accumulation of Medical Center): In the hepatitis C-associated nephritides, why do inflammatory cells within the capillaries or in the expanded some people develop MPGN, some membranous, and others IgA mesangial compartment. As you imply in your question, in such nephropathy? patients we assume that a change in the type and specificity of the DR. RENNKE: The answer to your question has not been antibodies produced in the patient leads to a more aggressive elucidated, It certainly has to do with the appropriateness of the form of glomerular injury, superimposed on a background of immune response to the hepatitis C virus, which varies among membranous lupus nephritis. patients. Some patients develop an immune response capable of DR. NIcoLAos E. MAD!AS (Chief Division of Nephrology, New effectively eradicating the virus from the system. Other patients

DR. RENNKE: I do not know the exact significance of this observation, and there is no good explanation in the literature for this phenomenon. The apparent loss of tissue-bound immunoglobulin also occurs in patients with other postinfectious glomerulonephritides, in particular, in patients with serologically docu-

mented poststreptococcal glomerulonephritis. These patients often are biopsied relatively late in the course of the disease. Characteristic electron-dense deposits, or "humps," are seen along the glomerular capillary wall by electron microscopy, but by

immunfluorescence microscopy one is able to uncover complement components but no immunglobulins. In the past, I have interpreted such findings as indicative of a relatively early loss of the antigenicity of the immunglobulin(s) in the setting of active inflammation. It is possible that complement components, for reasons unknown to me, retain the antigenic structure for a longer time. DR. MADIAs: Do we know anything about the composition of dense deposits in type-Il MPGN? Specifically, do they contain C3 nephritic factor? Also, is there evidence for the pathogenicity of

continuous hepatic damage. Products of the virus can enter the circulation and set the stage for an immune-complex-mediated disease. As is the case for other viral infections, and in particular for hepatitis B infection, the renal condition most often associated with persistent antigenemia is a membranoproliferative glomerulonephritis. The proposed mechanisms responsible for the development of a membranous glomerulopathy in response to a viral infection are different from those that produce MPGN. It is likely that the virus induces a limited autoimmune response in some human hosts. Autoantibodies could develop through so-called molecular mimicry, whereby a viral antigen induces the formation of antibodies that cross-react with an endogenous antigen on the glomerular visceral epithelial cell. Alternatively, such autoantibodies could be the result of a nonspecific polyclonal B-cell stimulation by the virus. The target of such autoantibodies is antigens normally expressed on the cell surface of the glomerular visceral epithelial cell. Immune complexes form in situ by the interaction of endogenous epithelial cell antigens and filtered autoantibodies [83]. A third alternative mechanism for the development of a membranous pattern of injury in the setting of a persistent viral infection implies the participation of a cationic viral antigen in the formation of subepithelial complexes. As I

this factor? Can it exist in the circulation in the absence of discussed before, such a mechanism would require that the nephritis? DR. RENNKE: We do not know the exact composition of the dense deposits. A recent review suggests that the dense deposits have a biochemical composition that resembles the normal basement membrane, and that they probably contain glycoproteins and lipids rich in unsaturated fatty acids [4]. Immunoglobulins, C3 nephritic factor, complement, and binding sites recognized by

exogenous antigen be planted in the subepithelial space of the glomerular capillary wall during phases of relative antigen excess in the circulation. The precise mechanism reponsible for a virus-

induced membranous nephropathy in humans has not been elucidated to date, and the possibilities I mentioned remain speculative. DR. PEREIRA: Although the hepatitis C virus has not yet been discovered, a fair length of the genomic sequences is known. Do

Goodpasture antibodies have not been identified within this material. The lack of a definitive answer to this question is not you know which particular antigens are associated with these entirely surprising, because we do not know the exact biochemical cryoglobulin deposits? Is there a possibility that only patients who

Nephrology Forum: Secondaiy MPGN

produce non-neutralizing antibodies are vulnerable to glomerulonephritides?

653

with a fingerprint pattern. A fingerprint pattern also has been

described in patients with lupus [108]. Do these patients also have cryoglobulins? Dn. RENNKE: It has been my experience that many patients with non-neutralizing antibodies develop some form of glomerular injury, is probably correct, at least for the membranoproliferative lupus nephritis who have electron-dense deposits with fingerprint type of disease, a condition that depends on the persistence of an patterns also have circulating ctyoglobulins. Often these patients antigen or immune complexes in the circulation. At present, not also present on biopsy with cryoprecipitates within capillary enough information is available on the immune status of patients lumena, the so-called microthrombi of active lupus nephritis. with hepatitis C infection and membranous nephropathy for us to Unfortunately, I have not rigorously recorded these observations, determine whether this complication can develop after the patient and I have not seen a published study that has addressed this has successfully eradicated the viral infection through an effective association. DR. MAmAs: Is there evidence for platelet involvement in the antibody response. DR. RENNICE: Your speculation, that only patients who produce

DR. PEREIRA: Does treatment with interferon modulate the histology in hepatitis C-associated glomerulonephritides? Is the renal injury worse in those who have liver disease compared with those who don't have liver disease? DR. RENNKE: I do not know whether the occurrence of renal

idiopathic or immune-complex types of MPGN, as opposed to that associated with thrombotic syndromes? DR. RENNICE: Several investigators have reported the finding of platelets within glomerular capillaries during the early phases of experimental immune-complex-mediated glomerulonephritis [56, disease correlates with the severity of the liver damage. Treatment 60, 86]. Most of these studies, however, were performed on with interferon appears to arrest the renal disease, at least models of acute glomerulonephritis and not in experimental serum sickness, the more appropriate model of membranoproliftemporarily [9]. DR. ANDREW KING (Division of Nephrology, New England Med- erative glomerulonephritis. The evidence that platelets are inical Center): Your classification of patients with membranoprolif- volved in membranoproliferative glomerulonephritis in humans is erative glomerulonephritis contains a wide range of urinary indirect and comes mostly from studies on platelet survival and findings associated with these diseases—from a very active ne- from clinical trials on the effect of platelet-inhibitor therapy phritic-type sediment to a relatively bland sediment. Could you (reviewed in Ref. 4). comment on the morphologic findings that lead to the differing urinary sediment composition? Dn. RENNXE: My own bias would suggest that patients with Reprint requests to Dr. HG. Rennke, Department of Pathology, Brigham active inflammation, defined by the presence of polymorphonu- and Women's Hospital, 75 Francis Street, Boston, Massachusetts 02115, USA clear and mononuclear inflammatory cells within the capillaries, would tend to have a more active urine sediment, occurring References predominantly during the phases of active immune-complex deposition. The sediment would become less active during phases of repair; morphologically this stage of the disease would be char1. Kuo G, Caoo Q, ALTER HJ, GrrNIcIc GL, REDEICER AG, PURCELL acterized by a lack of active inflammation within the capillary RH, MIYAMURA T, DIENSTAO JL, ALTER MJ, STEvENs CE, TEGTEloops, frequent double-contoured basement membranes, and MEIER GE, B0NIN0 F, CoLoMao M, LEE W, Kuo C, BERGER K, SHU5TER JR, OvERBY LR, BRADLEY DW, Houom7oN M: An assay moderate mesangial hypercellularity. A patient with this pattern for circulating antibodies to a major etiologic virus of human non-A, of injury can be anticipated to have heavy proteinuria or the non-B hepatitis. Science 244:362, 1989 nephrotic syndrome. Unfortunately, a rigorous study that corre2. GLkssocic RJ, ADLER SG, WARD HJ, COHEN AH: Primary glomerlates the urinary sediment findings and the biopsy abnormalities in

membranoproliferative glomerulonephritis has not been performed to my knowledge. DR. MrnAs: You expressed your conviction that the interposed

cells between the reduplicated basement membrane in MPGN represent blood-borne monocytic cells rather than glomerular mesangial cells. Has this issue been examined carefully using appropriate cell markers? DR. RENNKE: This issue has not been examined very carefully.

Most studies performed on biopsy specimens from patients with membranoproliferative glomerulonephritis and on experimental animals have employed immuno-enzymatic techniques at the light microscopic level. These techniques have the shortcoming of a limited resolution. Identification of the cells within the capillary wall at the ultrastructural level by appropriate markers is clearly required before we can settle this issue. DR. HARRINGTON: Has anyone tried to block ICAMs in an experimental model of MPGN? DR. RENNXE: I am not aware of such a study.

DR. LEvEY: You mentioned that cryoglobulins can present ultrastructurally in a number of different ways, one of which is

ular diseases, in The Kidney (vol I), edited by BRENNER BM, RECrOR

FC, Philadelphia, Saunders, 1991, p 1182 3. Swv FG: Membranoproliferative glomerulonephritis, in Pathology of the Kidney (vol I), edited by HEPTINSTALL RH, Boston, Little, Brown, 1992, p 477 4. HOLLEY KE, D0NAnI0 JV JR: Membranoproliferative glomerulonephritis, in Renal Pathology, With Clinical and Functional Correlations (vol I), edited by TI5HER CC, BRENNER BM, Philadelphia, Lippincott, 1994, p 294 5. FERRI C, GRECO F, LONGOMBARDO 0, PALLA P, MORETrI A, MARzO E, FO5ELLA PV, PASERO G, BOMBARDIER! 5: Antibodies to hepatitis

C virus in patients with mixed cryoglobulinemia. Arthritis Rheum 34:1606—1610, 1991

6. AGNELLO V, CHUNG RT, K.&nsi'i LM: A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 327:1490—1495, 1992 7, MIsIANI R, BELLAvITA P, FENILI D, BORELLI 0, MARCHES! D, MAssARzA M, VENDRAMIN 0, Corvoni B, TAlC! F, SCUDELLER 0,

ZANnnI A: Hepatitis C virus infection in patients with essential mixed cryoglobulinemia. Ann Intern Med 117:573—577, 1992 8. FERRI C, LA CIvITA L, LONGOMBARDO 0, GRECO F, BOMBARDIER! S:

Hepatitis C virus and mixed cryoglobulinaemia. Eur J Clin Invest 23:399—405, 1993

9. JOHNSON RJ, GRETCH DR, YAMABE H, HART J, BACCHI CE, HARTWELL P, CouseR WG, COREY L, WENER MH, ALPERS CE:

Nephrology Forum: Secondary MPGN

654

Membranoproliferative glomerulonephritis associated with hepatitis C virus infection. N Engl J Med 328:465—470, 1993 10. PASQUARIELLO A, FERRI C, MoRicoNi L, LA CIvITA L, L0NGOMBARDO G, LOMBARDINI F, GRECO F, ZIGNEGO AL: Ciyoglobu-

32. VoN BONSDORFF M, PONKA A, TORNROTH T: Mycoplasmal pneumo-

nia associated with mesangiocapillary glomerulonephritis type II (dense deposit disease). Acta Med Scand 216:427—429, 1984 33. NOCHY D, CALLARD P, BELLON B, BARIETY J, DRUET P: Association

linemic membranoproliferative glomerulonephritis associated with

of overt glomerulonephritis and liver disease: a study of 34 patients.

hepatitis C virus (letter). Am J Nephrol 13:300—304, 1993 11. RENNKE HG, KLEIN PS: Pathogenesis and significance of nonprimaly focal and segmental glomeruloscierosis. Am J Kidney Dis 13:443—456, 1989 12. D'AGATI V: Nephrology Forum: The many masks of focal segmental glomerulosclerosis. Kidney mt 46:1223—1241, 1994 13. RIcH AR: A hitherto undescribed vulnerability of the juxtamedullary glomeruli in lipoid nephrosis. Bull Johns Hopkins Hosp 100:173—186, 1957

Clin Nephrol 6:422—427, 1976 34. MORZ SP, CUTZ E, BALFE JW, SASS-KORTSAK A: Membranoprolif-

14. WEST CD, MCADAMS AJ, MCCONVILLE JM, DAVIS NC, HOLLAND

NH: Hypocomplementemic and normocomplementemic persistent (chronic) glomerulonephritis: Clinical and pathologic characteristics. JPediatr 67:1089—1112, 1965 15. HABIB R, GUBLER MC, LOIRAT C, MAIz HB, LEVY M: Dense deposit

disease: a variant of membranoproliferative glomerulonephritis. Kidney mt 7:204—215, 1975 16. POLLOCK J, WOOD B, KELLY JP: Membranoproliferative glomerulo-

nephritis, type II and partial lipodystrophy in an adult. Am J Kidney Dir 8:274—276, 1986 17. STRIFE CF, MCENERY PT, McAIi.Ms A.J, WEST CD: Membranopro-

liferative glomerulonephritis with disruption of the glomerular basement membrane. Clin Nephrol 7:65—72, 1977 18. ANDERS D, AGRICOLA B, SIPPEL M, THOENES W: Basement mem-

brane changes in membranoproliferative glomerulonephritis. II. Characterization of a third type by silver impregnation of ultra thin sections. Virchows Arch [Pathol Anatl 376:1—19, 1977 19. BURKHOLDER PM, HYMAN LR, KRUEGER RP: Characterization of

mixed membranous and proliferative glomerulonephritis: recognition of three varieties. Perspect Nephrol Hypertens 1:557—589, 1973 20. AUSTIN HA, MUENZ LR, JOYCE KM, ANTONOVYCH TF, BALOW JE:

Diffuse proliferative lupus nephritis: identification of specific pathologic features affecting renal outcome. Kidney mt 25:689—695, 1984 21. MOUTSOPOULOS HM, BAJ.ow JE, LAWLEY TJ, STAHL NI, ANTONOVYCH TT, CHUSED TM: Immune complex glomerulonephritis in sicca syndrome. Am J Med 64:955—960, 1978

22. TING HC, WANG R: Mesangiocapillary (membranoproliferative) glomerulonephritis and rheumatoid arthritis. Br Med J 1:270—271, 1977 23. KIM Y, FRIEND PS, DRESNER IG, YUNIS EJ, MICHAEL AF: Inherited

deficiency of the second component of complement (C2) with membranoproliferative glomerulonephritis. Am J Med 62:765—771, 1977

24. BORZY MS, HOUGHTON D: Mixed-pattern immune deposit glomer-

ulonephritis in a child with inherited deficiency of the third component of complement. Am J Kidney Dis 5:54—59, 1985 25. COLLINS AB, BHAN AK, DIENSTAG JL, COLVIN RB, HAUPERT GT JR,

MUSHAHWAR 1K, MCCLUSKEY RT: Hepatitis B immune complex glomerulonephritis: simultaneous glomerular deposition of hepatitis B surface and e antigens. Gun Immunol Immunopathol 26:137—153, 1983 26. BEAUFILS M, GIBERT C, MOREL-MAROGER L, Si&R JD, KANFER A,

MEYRIER A, KOURILSKY 0, VACHON F, RICHET G: Glomerulone-

phritis in severe bacterial infections with and without endocarditis. Adv Nephrol Necker Hosp 7:217—234, 1977 27. STRIFE CF, MCDONALD BM, RULEY EJ, MCADAMS AJ, WEST CD:

Shunt nephritis: the nature of the serum cryoglobulins and their relation to the complement profile. J Pediatr 88:403—413, 1976 28. MAGIL AB: Monocytes and glomerulonephritis associated with remote visceral infection. Clin Nephrol 22:169—175, 1984 29. WEINER ID, NORTHCUTr AD: Leprosy and glomerulonephritis: case report and review of the literature. Am J Kidney Dis 13:424—429, 1989

30. WHITE RH: Quartan malarial nephrotic syndrome. Nephron 11:147— 162, 1973 31. ANDRADE ZA, ROCHA H: Schistosomal glomerulopathy. Kidney mt 16:23—29, 1979

erative glomerulonephritis in childhood cirrhosis associated with alpha 1-antitrypsin deficiency. Pediatrics 57:232—238, 1976 35. FIABIB R, COURTECUISSE V, LECLERC F, MATHIEU H, ROYER P:

Etude anatomo-pathologique de 35 observations de syndrome hemolytique et uremique de l'enfant. Arch Fr Pediatr 26:391—416, 1969 36. KINCAID-SMITH P, NICHOLLS K: Renal thrombotic microvascular disease associated with lupus anticoagulant. Nephron 54:285—288, 1990 37. KEANE WF, CROSSON JT, STAEEY NA, ANDERSON WR, SHAPIRO FL:

Radiation-induced renal disease. A clinicopathologic study. Am J Med 60:127—137, 1976

38. ZAGER RA: Nephrology Forum: Acute renal failure in the setting of bone marrow transplantation. Kidney mt 46:1443—1458, 1994 39. PARDO V, STRAUSS J, KRAMER H, OZAWA T, MCINTOSH RM:

Nephropathy associated with sickle cell anemia: an autologous immune complex nephritis. II. Clinicopathologic study of seven patients. Am J Med 59:650—659, 1975

40. SPEAR GS: Glomerular alterations in cyanotic congenital heart disease. Bull Johns Hopkins Hosp 106:347—367, 1960

41. HSU HC, SUZUKI Y, CHURG J, GRISHMAN E: Ultrastructure of transplant glomerulopathy. Histopathology 4:351—367, 1980 42. FEINER H, GALLO G: Ultrastructure in glomerulonephritis associated

with cryoglobulinemia. A report of six cases and review of the literature. Am J Pathol 88:145—162, 1977 43. MOULIN B, RoNco PM, MOUGENOT B, FRANGIONE B, FINBERG R,

MIGNON F: Glomerulonephritis in chronic lymphocytic leukemia and related B-cell lymphomas. Kidney mt 42:127—135, 1992 44. LIN JH, OROFINO D, SHERLOCK J, LETFERI J, DUFFY JL: Walden-

strom's macroglobulinemia, mesangiocapillary glomerulonephritis, angiitis and myositis. Nephron 10:262—270, 1973 45. KORBET SM, SCHWARTZ MM, ROSENBERG BF, SIBLEY RK, LEWIS

EJ: Immunotactoid glomerulopathy. Medicine (Baltimore) 64:228— 243, 1985 46. FoGo A, QURESHI N, HORN RG: Morphologic and clinical features of fibrillary glomerulonephritis versus immunotactoid glomerulopathy. Am J Kidney Dis 22:367—377, 1993 47. GALLO GR, FEINER HD, KATZ LA, FELDMAN GM, CORREA EB, CHUBA JV, BUXBAUM JN: Nodular glomerulopathy associated with

nonamyloidotic kappa light chain deposits and excess immunoglobulin light chain synthesis. Am J Pathol 99:621—644, 1980 48. BUXBAUM JN, CHUBA JV, HELLMAN GC, SOLOMON A, GALW GR:

Monoclonal immunoglobulin deposition disease: light chain and light and heavy chain deposition diseases and their relation to light chain

amyloidosis. Clinical features, immunopathology, and molecular analysis. Ann Intern Med 112:455—464, 1990 49. STURGILL BC, BOLTON WK, GRIFFITH KM: Congo red-negative amyloidosis-like glomerulopathy. Hum Pathol 16:220—224, 1985 50. ALPERS CE, RENNKE HG, HOPPER JJ, BIAVA CG: Fibrillary glomer-

ulonephritis: an entity with unusual immunofluorescence features. Kidney mnt 31:781—789, 1987

51. ESPARZA AR, CH&zAN JA, NAYAK RN, CAVALLO T: Fibrillary

(immunotactoid) glomerulopathy. A possible role for kappa light chain in its etiology and/or pathogenesis. Am J Surg Pathol 15:632— 643, 1991 52. ISKANDAR SS, FALK RJ, JENNETTE JC: Clinical and pathologic features of fibrillary glomerulonephritis. Kidney Int 42:1401—1407, 1992 53. MONGA G, MAzzucco G, DIBELGIOJOSO GB, BUSNACH G: The presence and possible role of monocyte infiltration in human chronic proliferative glomerulonephritides: Light microscopic, immunofiuorescence, and histochemical correlations. Am J Pathol 94:271—284, 1979

Nephrology Forum: Secondary MPG/SI 54. SCHREINER OF, COTRAN RS, PARDO V, UNANUE ER: A mononu-

clear cell component in experimental immunological glomerulonephritis. J Exp Med 147:369—384, 1978 55. HUNSICKER LU, SHEARER TP, PLAnNER SB, WEISENaUROER D: The

role of monocytes in serum sickness nephritis. J Exp Med 150:413— 425, 1979 56. FRIES JWU, MENORICK DL, RENNKE HG: Determinants of immune complex-mediated glomerulonephritis. Kidney mt 34:333—345, 1988 57. Anu. GB: Immune-complex glomerulonephritis deposits plus interest (editorial). N Engi J Med 328:505—506, 1993 58. DixoN FJ, VAZQUEZ JJ, WEIGLE WO, COCHRANE CU: Pathogenesis of serum sickness. Arch Pathol 65:18—28, 1958 59. MAUER SM, SUTI-IERLANO DE, HowAEn RJ, FISH AJ, NAJARIAN JS,

MICHAEL AF: The glomerular mesangium. III. Acute immune me-

sangial injury: a new model of glomerulonephritis. J Exp Med 137:553—570, 1973

60. GoLauS SM, WiLsoN CB: Experimental glomerulonephritis induced by in situ formation of immune complexes in glomerular capillary wall. Kidney lot 16:148—157, 1979 61. EDDY AA, MICHAEL AF: Immunopathogenic mechanisms of glomer-

655

77. DIxON FJ, FELDMAN JD, VAZQUEZ JJ: Experimental glomerulone-

phritis. The pathogenesis of a laboratory model resembling the spectrum of human glomerulonephritis. J FLrp Med 113:899—919, 1961

78. AvILA-CA5ADO C, SANDSTROM DJ, RENNKE HG: Immune complex

mediated glomerulonephritis (ICON) in the rat; the effect of molec-

ular charge and mode of antigen delivery (abstract). J Am Soc Nephrol 4:-953, 1993 79. BENACERRAF B, SEBESTYEN M, COOPER NS: The clearance of antigen-antibody complexes from the blood by the reticuloendothehal system. J Immunol 82:13 1—137, 1959

80. HEBERT LA: The clearance of immune complexes from the circulation of man and other primates. Am J Kidney Dis 17:352—361, 1991 81. MCCLUSKEY RT, BENACERRAF B, POTrER JL, MILLER F: The

pathologic effect of intravenously administered soluble antigenantibody complexes. J Exp Med 111:181—194, 1960 82. GERMUTH FG JR, RODRIGUEZ E, LORELLE CA, TRUMP El, Mn&ro

L, WISE 0: Passive immune complex glomerulonephritis in mice: models for various lesions found in human disease. II. Low avidity complexes and diffuse proliferative glomerulonephritis with subepi-

ular injury, in Renal Pathology with Clinical and Functional Correlations, edited by TISHER CC, BRENNER BM, Philadelphia, Lippincott, 1994, p 162

thehial deposits. Lab Invest 41:366—371, 1979 83. COU5ER WG: Nephrology Forum: Mechanisms of glomerular injury in immune-complex disease. Kidney lot 28:569—583, 1985

62. RENNKE HG, COTRAN RS, VENKATACHALAM MA: Role of molecular

84. CAULIN-GLASER T, GALLO OR, LAMM ME: Nondissociating cationic

charge in glomerular permeability. Tracer studies with cationized

tivity of the glomerular capillary wall. III. Restricted transport of

immune complexes can deposit in glomerular basement membranes. JExp Med 158:1561—1572, 1983 85. MCCLUSKEY RT: Modification of glomerular immune complex deposits (editorial). Lab Invest 48:24 1—244, 1983

polyanions. Kidney mt 8:212—218, 1975

86. JOHNSON RJ, ALPERS CE, PRITZL P, SCHULZE M, BAKER P, PRUCHNO

ferritins. J Cell Biol 67:638—646, 1975 63. CsANG RLS, DEEN WM, ROBERTSON CR, BRENNER BM: Permselec-

64. RoDEwan R, KARNOVSKY MJ: Porous structure of the glomerular slit diaphragm in the rat and mouse. J Cell Biol 60:423—433, 1974

C, COLISER WG: Platelets mediate neutrophil-dependent immune complex nephritis in the rat. J Clin Invest 82:1225—1235, 1988

65. BORDER WA, KAMIL ES, WARD HI, COHEN AH: Antigenic charge as

87. HOLLEY ICE, DONADIO JV JR: Membranoproliferative glomerulonephritis, in Renal Pathology with Clinical and Functional Correlations, edited by TISHER CC, BRENNER BM, Philadelphia, Lippincott, 1994, p 294 88. REMUZZI U, RUGGENENTI P, BERTANI T: Thrombotic microangiopa-

a determinant of immune complex localization in the rat glomerulus. Lab Invest 40:442—449, 1981 66. VOGT A, ROHRBACH R, SHIMIzu F, TAKAMIYA H, BATSFORD 5:

Interaction of cationized antigen with rat glomerular basement membrane: in situ immune complex formation. Kidney mt 22:27—35, 1982 67. A0LER 5, SALANT DJ, DITrMER JE, RENNKE HG, MADAJO MP, COUSER WG: Mediation of proteinuria in membranous nephropathy due to a planted glomerular antigen. Kidney lot 23:807—815, 1983 68, CYBULSKY AV, RENNKE HG, FEINTZEIG ID, SALANr DJ: Comple-

ment-induced glomerular epithelial cell injury. Role of the membrane attack complex in rat membranous nephropathy. J Clin Invest 77:1096—1107, 1986

thy, in Renal Pathology with Clinical and Functional Correlations, edited by TISHER CC, BRENNER BM, Philadelphia, Lippincott, 1994, p 1154 89. FITZPATRICK MM, SHAD V, TROMPETER RS, DILLON MJ, BARRATI-

TM: Interleukin-8 and polymorphoneutrophil leukocyte activation in hemolytic uremic syndrome of childhood. Kidney Int 42:951—956, 1992

90. LEUNG DY, MOAKE JL, HAVENS PL, KIM M, POBER JS: Lytic

aoti-endothelial cell antibodies in haemolytic-uraemic syndrome.

69. MADDOx DA, BRENNER BM: Glomerular ultrafiltration, in The Kidney, edited by BRENNER BM, RECrOR FC JR, Philadelphia, Saunders, 1991, p 205 70. BRADY HR: Leukocyte adhesion molecules and kidney diseases.

Lancet 2(8604):183—186, 1988 91. MOATCE IL, RUDY CK, TROLL JH, WEINSTEIN MJ, COa.At'IINO NM, AZOCAR I, SEDER RH, HONG SL, DEYKIN D: Unusually large factor

Kidney Int 45:1285—1300, 1994 71. GALLO GR, CAULIN-GLASER T, EMANCIPATOR SN, LAMM ME:

botic thrombocylopenic purpura. N EngI J Med 307:1432—1435, 1982 92. MURPHY WG, MOORE JC, KELTON JO: Calcium-dependent cysteine

Nephritogenicity and differential distribution of glomerular immune complexes related to immunogen charge. Lab Invest 48:353—362,

protease activity in the sera of patients with thrombotic thrombocy-

1983 72. GALLO GR, CAULIN-GLASER T, LAMM ME: Charge of circulating

immune complexes as a factor in glomerular basement membrane localization in mice. J Clin Invest 67:1305—1313, 1981 73. MATSUO 5, FUKATSU A, TAUB ML, CALDWELL RB, BRENTJENS JR,

ANDRES GA: Glomerulonephritis induced in the rabbit by antiendothelial antibodies. J Clin Invest 79:1798—1811, 1987 74. MATSUO 5, YOSHIDA F, YUZAwA Y, FLkIEk 5, FUKATSU A, WATANABE Y, 54&Ic.4MOTO N: Experimental glomerulonephritis induced

in rats by a lectin and its antibodies. Kidney Int 36:1011—1021, 1989 75. Izu 5, LAMBERT PH, MIESCHER PA: In vitro demonstration of a

particular affinity of glomerular basement membrane and collagen for DNA: A possible basis for a local formation of DNA-anti-DNA

VIII:von Willebraod factor multimers in chronic relapsing throm-

topenic purpura. Blood 70:1683—1687, 1987 93. BERGSTEIN JM, RILEY M, BANG NU: Role of plasminogen-activator

inhibitor type 1 in the pathogenesis and outcome of the hemolytic uremic syndrome. N EngI J Med 327:755—759, 1992 94. RLJGGENENTT P, GALBUSERA M, CORNEJO RP, BELLAVITA P, RE-

MUZZI U: Thrombotic thrombocytopenic purpura: Evidence that infusion rather than removal of plasma induces remission of the disease, Am J Kidney Dis 21:314—318, 1993 95. FANG LS: Nephrology Forum: Light-chain nephropathy. Kidney Int 27:582—592, 1985

96. MOREL-MAROGER STRIKER LI, PREUD'HOMME J-L, D'AMICO U, STRIKER GE: Monoclonal gammopathies, mixed cryoglobuhinemias,

and lymphomas, in Renal Pathology with Clinical and Functional Correlations, edited by TISHER CC, BRENNER BM, Philadelphia,

complexes in systemic lupus erythematosus. J Exp Med 144:428—443, 1976 76. MISIANI R, BELLAVITA P, FENTLI D, BORELLI U, MARCHESI D,

Lippincott, 1994, p 1442 97. ALPERS CE: Immunotactoid (microtubular) glomerulopathy: An entity distinct from fibrillaiy glomerulonephritis? Am J Kidney Dis

MASSAZZA M, VENDRAMIN G, CoMoni B, TANzI E, SCUDELLER U,

19:185—191, 1992 98. DEVANEY K, SABNIS SG, ANTONOVYCH Ti': Nonamyloidotic fibrillasy

ZANETrI A: Hepatitis C virus infection in patients with essential mixed cryoglobulinemia. Ann totem Med 117:573—577, 1992

glomerulopathy, immunotactoid glomerulopathy and the differential

656

Nephrology Fomm: Secondaiy MPGN

diagnosis of filamentous glomerulopathies. Mod Pathol 4:36—43, 1991

99. SCHWARTZ MM: Immunotactoid glomerulopathy: the case for Occam's razor (editorial). Am J Kidney Dis 22:446—447, 1993

100. YrG GC, NIET0 R, STACHURA I, GALLO GR: Ultrastructural immunohistochemical localization of polyclonal lgG, C3, and amybid P component on the congo red-negative amyboid-like fibrils of fibrillary glomerulopathy. Am J Pathol 141:409—419, 1992 101. SCHWARTZ MM, LEwIs EJ: The quarterly case: Nephrotic syndrome in a middle-aged man. Ultrastruct Pathol 1:575, 1980 102. ROSTOKER G, DESVAUZ-BELGHITI D, PILATrE Y, PETIT-PHAR M, PHILIPPON C, DEFORGES L, TERZIDIS H, INTRATOR L, ANDRE C,

LAGRUE G, Lo P, WElL B: High-dose immunogbobulin therapy for severe IgA nephropathy and Henoch-SchOnlein purpura. Ann Intern Med 120:476—484, 1994 ADNOT 5, BONIN P, BIERLING P, REMY P,

103. HIRATA M, TERASAKI PT: The long-term effect of primary disease on

cadaver-donor renal transplant recipients, in Clinical Transplants

1993, edited by TERASAKI P1, CECKA JM, Los Angeles, UCLA Tissue

Typing Laboratory, 1994, p 485 104. GRIMWOOD R, ARROYAVE CM, WESTON WL, AELING JL: Herpes

gestationis associated with the C3 nephritic factor. Arch Dermatol 116: 1045—1047, 1980

105. BORRADORI L, RYBOJAD M, MOREL P, PuIssANT A, WEISS L: Chronic urticaria and moderate leukocytoclastic vasculitis associated with C3 nephritic factor activity (letter). Arch Der,natol 125:1589— 1590, 1989 106. CARMICHAEL AJ, MARSDEN JR: Urticarial vasculitis: a presentation of C3 nephritic factor (letter). Br J Dermatol 128:589, 1993 107. SPITZER RE, STITZEL AE, TsoKos GC: Evidence that production of autoantibody to the alternative C3 convertase is a normal physiologic event.JPediatr 116:S103—S108, 1990 108. AUSTIN HA III, MUENZ LR, JOYCE KM, ANTONOVYCH VF, BALOW

JE: Diffuse proliferative lupus nephritis: Identification of specific pathologic features affecting renal outcome. Kidney mt 25:689—695, 1984