Role of Circulating Immune Complexes in Renal Diseases

J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

J Clin Pathot 1981 ;34:1214-1222

Role of circulating immune complexes in renal diseases

ROLAND J LEVINSKY From the Department of Immunology, Institute of Child Health, 30 Guilford Street, London WCJ

The elimination of a foreign antigen is a function of induced by repeated antigen administration.7 The the antibody, complement and phagocyte systems. most widely quoted experimental model of immune When an immune complex is formed in the circu- complex injury is serum sickness, in which the lation, clearance is effected by cells of the reticulo- animal develops nephritis and vasculitis due to endothelial system. Macrophages and polymorpho- deposition of immune complexes 8-10 days after the nuclear cells have both IgG Fc and C3b complement injection of antigen. When the antigen is still in receptors so that opsonisation by either the classical excess of antibody within the circulation, the small or the alternative pathway of complement activation immune complexes formed remain in solution but facilitates antigen elimination. This system normally can become trapped in vessel walls at sites of provides a most efficient protection against disease, turbulence or at capillary membranes where filtration since man is continually challenged by a variety of occurs. Tissue damage at these sites results from the inhaled and ingested antigens. biological effects of complexes which activate the

Why then does a normally protective system complement system, resulting in immune adherence,copyright. become self-damaging, resulting in immune complex polymorph chemotaxis, release of lysosomal pro- disease? It is unlikely that most immune complex teolytic enzymes and kinins, and platelet aggregation. diseases are due to rare or as yet unidentified The latter causes release of vasoactive amines8 which antigens. It is much more likely that host variation in increase vascular permeability and so further immune response to a common antigen resuits in localise complexes to the site of the injury. effective clearance and health in one individual, but If sections of the kidneys are examined at the persistent circulating antigen and immune complex height of injury, the glomeruli show evidence of of with cell disease in another. Increased concentrations nephritis, endothelial swelling and lifting http://jcp.bmj.com/ circulating soluble immune complexes are found in a of the underlying basement membrane. Immuno- variety of diseases; these include systemic lupus fluorescence shows immunoglobulins of different erythematosus, various forms of glomerulonephritis, classesandvariouscomplementcomponents deposited rheumatoid arthritis, and chronic inflammatory in an irregular granular manner along the capillary bowel diseases.' High concentrations of immune basement membrane.9 A similar distribution of complexes are also found in many forms of dis- electron-dense deposits may be shown by electron seminated malignancy2 where it is thought that they microscopy.10 Despite the similarities between may have an enhancing effect on tumour growth, by animal and human nephritis, there is still controversy on September 30, 2021 by guest. Protected preventing the host from mounting an effective about the role of circulating soluble immune immune response to his own tumour.3 complexes in man; indeed in membranous nephro- Paradoxically, immune complexes are also found, pathy and acute poststreptococcal nephritis it may in low concentrations, in the sera of apparently be difficult to demonstrate immune complexes in the healthy people, particularly after food;4 they have circulation, yet the renal biopsies show the classical also been described in normal pregnancy,5 and in granular deposition of immunoglobulins and com- pre-eclampsia.6 The finding of immune complexes plement components. In animals it is possible to in so many different diseases as well as in normal induce local formation of immune complexes within biological states raises the question whether they do glomeruli by the intravenous injection of bacterial indeed play a crucial role in pathological processes. lipopolysaccharide.1" It is certainly possible that in many forms of glomerulonephritis an extrinsic Evidencefor a damaging role ofimmune complexes antigen provokes the initial injury, possibly releasing Much of the evidence comes from animal studies, in hidden self-antigens which elicit an immune response particular those in which chronic nephritis is resulting in the formation of immune complexes at 1214 J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

Role of circulating immune complexes in renal diseases 1215 the site of damage. Local formation of complexes or of viruses, bacteria, parasites, moulds and ingested deposition of circulating immune complexes are not food antigens, but some are intrinsic such as DNA, mutually exclusive mechanisms of tissue damage, RNA, immunoglobulins and other proteins. The but it must be emphasised that the mere demonstra- portal of entry for the antigen may be important in tion of soluble immune complexes in the circulation the type of immune response elicited; entry via the does not necessarily mean that they are directly gastrointestinal tract may be morelikely to producean responsible for the disease. IgA antibody response than entry by respiratory or parenteral routes. If the antigen is a virus or a Host factors contributing to variation in susceptibility parasite, its mode of replication within the host and to immune complex disease the subsequent release of reproductive antigenic In human poststreptococcal glomerulonephritis, as material may be important in producing different in serum sickness, the majority of individuals types of immune complexes. recover from the disease rapidly and completely. The different immunoglobulin classes and sub- Only a very small minority develop chronic glom- classes of antibodies impart different biological erulonephritis with ensuing renal failure. Such properties to the immune complexes. Theoretically, variation in immune response to an antigen may be immune complexes may be produced by all the demonstrated in experimental nephritis in rabbits immunoglobulin classes, and some evidence has where only those animals producing non-precipit- been provided for complexes involving IgM, IgG, ating (poor quality) antibody develop chronic IgA and also IgE.22-24 The complement-binding disease; those producing precipitating antibody get characteristic of the immune complex is determined acute nephritis and recover.'2 Oldstone and Dixon'3 by the class of immunoglobulin involved, and the demonstrated that only certain strains of mice were size of the complex depends on factors such as the susceptible to chronic nephritis when infected with antibody :antigen ratio, strength of antibody lymphocytic choriomeningitis virus in the neonatal binding and whether secondary interaction has period. In response to antigen injected in saline, the occurred with rheumatoid factors or immuno- strains of mice to chronic nephritis a human susceptible conglutinin, naturally occurring 1gM copyright. produced lower-affinity antibody-that is, with antibody to activated C3bi. Another factor is the weaker antigen-antibody binding than the nephritis- action of complement in solubilising membrane- resistant animals,'4 suggesting that susceptibility was bound immune complexes.25 genetically determined. Low affinity antibody is poor Thus immune complexes vary not only in the at antigen elimination'5 and is one example of a antigen but also in the immunoglobulin class and defective rather than an overactive immune system subclass, in size, in complement-binding capacity and failing to eliminate antigen and so predisposing to in the antigen:antibody ratio (Fig. 1). All these chronic immune complex disease. There are many features may be important in the localisation of examples in man of immune complex disease with complexes in one organ rather than another, and in http://jcp.bmj.com/ underlying primary immune deficiency-for example, the different syndromes and diseases which result. the higher incidence of juvenile chronic polyarthritis in children with selective IgA deficiency,'6 the (Site of entry ) (Quantity) arthritis complicating hypogammaglobulinaemia,'7 Antgen (Rate of release) (Affinity) Antiby and the increased incidence of diseases such as sys- (Replication) (Subclass) temic lupus erythematosus in people with inherited (Structure ) (Duration) deficiencies of the complement system, in particular on September 30, 2021 by guest. Protected homozygous C2 deficiency.'8 The genetic differences in susceptibility to immune Immune complexes Varying in antigen, antibody class and subclass, size, complex disease in animals is antigen non-specific. It complement binding capacity as well as Ag/Ab ratio is likely that this is true also for man since a common antigen such as hepatitis B can occur in the healthy Solubilisation Local anatomy. carrier state, yet has been incriminated in such dif- by complement permeability factors. ferent diseases as chronic active hepatitis,'9 mem- cell receptors branous nephropathy,20 and polyarteritis nodosa.21 Reticuloendothelial system Tissue deposition capacity and funct ion Factors affecting immune-complexformation, deposition and elimination There are many variables involved in the formation Removal by BES.- of immune complexes. There is the wide variety of Fig. 1 Factors affecting immune complex formation and antigens available, mostly extrinsic such as products elimination. J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

1216 Levinsk.y The factors underlying the deposition of immune complexes into smaller ones.25 Complement- complexes in tissues are not fully understood. They solubilised immune complexes have a low binding must include the rate of formation, the rate of affinity for platelet and neutrophil membranes,37 38 clearance and the efficiency of elimination by the and hence may be less able to mediate the release of reticuloendothelial system, as well as the rate of vasoactive amines and lysozomal enzymes from these blood flow through the organ. Furthermore the size cells. of the complex was shown to be important in renal It is thought that a local increase in vascular localisation in animals by Germuth and Rodrigues,26 permeability facilitates immune complex deposition; who demonstrated that medium sized complexes of histamine injected into an animal intravenously at bovine serum albumin (BSA)/anti-BSA were the time of immune complex injection localises nephrotoxic in rabbits whereas larger ones were not. complexes to lung capillaries.39 Platelets are a rich Similarly, when well-defined dinitrophenol (DNP)/ source of histamine and other vascular permeability anti-DNP immune complexes were injected into rats, factors. In acute serum sickness in rabbits, basophil complexes of intermediate size (greater than 19S) depletion and degranulation occur at the time of became localised in the mesangium of the renal glomerular injury and it has been suggested that tgE glomerulus whereas very large complexes became antibodies are produced and bind to mast cells localised transiently as microemboli in the lumen of causing degranulation in the presence of antigen and capillaries in the lung, skin, and heart.27 release of platelet-activating factor which causes Elimination of immune complexes is dependent on platelet clumping with subsequent release ofvasoactive the functional capacity of the reticuloendothelial amines.40 Human platelets may be aggregated by system which may easily become overloaded. platelet-activating factor and it is certainly possible Blockage of the macrophage system in animals by that this mechanism is involved in immune complex injection of carbon particles facilitated glomerular nephritis in man. deposition of preformed complexes injected subse- Macrophages and polymorphs have receptors for quently.28 Localisation in this instance was in tgG Fc and C3b, so clearance of IgG complexes by

subendothelial areas whereas without carbon block- the former and IgM complexes by the latter (ifcopyright. ade the immune complexes were deposited only in complement is activated) can be envisaged. Recently the mesangium,29 30 altnough at higher dosage some a transport system has been proposed for the passage capillary localisation does occur.27 31 32 In contrast, of dimeric IgA and IgA complexes through the in human disease and in animals with chronic hepatocyte into the biliary system, which recognises nephritis produced by prolonged antigen administra- the secretory piece linking the IgA molecules.4' tion, complexes may be found within the glomerular Since IgA immune complexes appear to constitute a basement membrane.32 Mannik(1 979)33 has suggested normal physiological clearance mechanism for food that are carried across the basement antigens entering across the gut mucosa (R Paganelli complexes http://jcp.bmj.com/ membrane from endothelium to epithelial cell to and RJ Levinsky unpublished data, 1981), it is mesangial cell, and elimination occurs at the latter possible that in diseases where deposition of igA site. The alternative view is that localisation in the immune complexes in the kidney are a predominant basement membrane or subepithelially is due to feature, the defect lies in the liver transport mechan- local formation and not due to deposition of isms. circulating complexes.34 C3b receptors have been found on the epitnelial cell surface in the glomeruli,35 and these receptors are blocked in the glomerular Antigens involved in renal immune-complex diseases on September 30, 2021 by guest. Protected diseases which are thought to be mediated by immune Table 1 gives some of the antigens which have been complexes.36 It is an attractive suggestion that implicated in human immune complex nephritis. In immune complexes become localised in an organ most cases of nephritis in man, however, no antigen such as the kidney because of specific receptors, but is identified either in the circulation or in the kidney the site of these receptors does not coincide with the biopsy material or indeed even suspected. Antibodies distribution of complexes found by electron micro- against extrinsic antigens may cross-react with scopy in human disease. self-protein so that the continuing presence of the The finding of various complement components in antigen is no longer necessary for a disease process to tissues by immunofluorescence indicates the im- continue, since the antibody formed then reacts portance of complement activation in the mechanism against the host tissue. Systemic lupus erythematosus of injury. Complement also has a clearance function (SLE) may well be such an example where antibody since membrane-bound immune complexes may be produced against viral DNA cross-reacts with host solubilised by the insertion of C3b and possibly C4b DNA. Such a mechanism could also lead to local into the antigen-antibody lattice, so splitting large formation of immune complexes if further antigens J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

Role of circulating immune complexes in renal diseases 1217

Table I Antigens involved in renal immune complex Immunity deficiency diseases

Extrinsic Intrinsic Persistance of microbial or viral products Bacteria Streptococcus spp DNA Host cell disrruption Adjuvant effect Polydonal Bcell Staphylococcus spp Tumour antigens activation Mycobacterium leprae Carcinoembryonic antigen Treposenia pallidumyi Thyroglobulin Viruses Hepatitis B Epstein-Barr Varicel la Release to-antigens Autoantibodies

Mumps % Rickettsia spp Failure of Parasites suppressor Malaria T cell Schistosoma regulation Drugs (examples) Penicillin Immune complexes Rheumatoid factitor Penicillamine inin Gold t immunoconglutir Tissue injury Fig. 2 Non-specific induction of immune complexes. are uncovered at the site of a damaged basement membrane. A further intriguing possibility is that many of the Table 2 Principles of methods of inmmune complex immune complexes are non-specific, induced by an detection

antigen which causes polyclonal activation of Morphology Electron microscopy copyright. B lymphocytes with the result that rheumatoid Separation by size Ultracentrifugation Gel chromatography factors and other autoantibodies suchas those to DNA Physicochemical Cryoprecipitation are formed. Such a mechanism occurs in mice injected Polyethylene glycol precipitation Complement binding Anticomplementary activity with bacterial lipopolysaccharide,42 or infected Clq experimentally with trypanosomiasis.43 Furthermore, Conglutinin Antiglobulin binding Monoclonal rheumatoid factor Roberts and Lewis4 demonstrated both DNA and Polyclonal rheumatoid factor its antibodies in cryoprecipitates in patients with IgM antibodies or Antiantibody SLE and also with other forms of nephritis http://jcp.bmj.com/ Cell receptor binding Raji cells (IgG Fc, C3b, Clq) intercurrent bacterial infections. A hypothetical Macrophages (IgG Fc, C3b) model for the non-specific induction of immune Polymorphs (IgG Fc, C3b) B lymphocytes (IgG Fc, C3b or d) complexes is illustrated in Fig. 2. Immune deficiency K cells (IgG Fc) is proposed as the underlying trigger which allows Platelets (IgG Fc) persistence of antigens within the circulation and tissues, leading to host cell disruption and auto- antibody formation. The system becomes self- recognises the conformational change occurring at generating when there is a failure of control mech- the antigen-binding site when the antibody combines on September 30, 2021 by guest. Protected anisms such as a lack of suppressor cell control of with antigen, is the only one of these antigen non- polyclonal B cell activation in this model. specific techniques capable of distinguishing aggregated IgG from an antigen-antibody complex. Principles ofimmune complex detection Although the assays have usually been shown to All the techniques for detecting immune complexes detect preformed immune complexes, false-positives are based upon either their physicochemical prop- due to aggregated immunoglobulins may easily be erties which are mainly due to size, or their biological obtained. properties which depend upon their binding to other Separation of immune complexes by ultracentri- humoral proteins or to specific cell receptors (Table fugation or gel chromatography is relatively crude 2). and requires further analysis of the macromolecular With the exception of the electron microscope, by fraction by immunochemical methods. The simplest which it is occasionally possible to identify the technique for detecting immune complexes is by specific antigen, all the other techniques are antigen cryoprecipitation. Cryoprecipitates have been de- non-specific. The antiantibody method, which scribed in many diseases, and in certain instances J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

1218 Levinsky have been shown to contain antigen, antibody and for immune complex detection produced widely activated complement components.45 discrepant results for the same sera in different Polyethylene glycol (PEG), in common with other diseases.50 Table 3 gives the results of 10 commonly uncharged water-soluble polymers such as dextran or polyvinyl pyrolidone, will precipitate proteins Table 3 Comparisoni f 1(3 different inuilnine comnplex without denaturation according to their molecular ass5as.5 0)1 saiiie SLE sera weight.46 The reaction is not specific and the amount co- T(.,.vt °, .lsowi1,,' of uncomplexed monomeric immunoglobulin (Ib),iornlolit v precipitated depends upon the concentration of PEG (2-3 is optimal for immune complex precipitation). Clql'2 Clq binding in polveths,lene glycol"2 58 % Clq SP Clq solid phase" 5X The techniques based on complement interaction Clq DV Clq deviationi 19) depend either upon activation of additional comple- Clq RI Inhibition of Clq bindiilg to IgG sepharose"' 2) ment (anticomplementary method) upon binding to KGB SP Conglutinin binding solid phase9 33 Clq, the recognition protein of the classical comple- RAJI Raji cell binding6786 ment which binds to the Fc region of tgM, MRF-I Inhibition of solid phase monoclonal pathway rheumatoid factor binding to "2I-aggregated IgGi, lgG2 and of IgG3, or binding by bovine con- IgG 04 29 glutinin to the activated C3 (C3bi) within the immune Plat Agg Platelet aggregation9 65 K cell Inhibition of K cell cytotoxicity 68 complex (conglutinin is a bovine protein of non- MRFRI Inhibition of soluble monoclonal immunological origin which binds C3bi). Whereas the rheumatoid factor binding to IgG sepharose 5 4 first two methods require that the immune complex activates the classical pathway, the reaction with Data 'adapted froni WHO collaborative stUds. -` bovine conglutinin can follow complement activation by either the classical or the alternative pathway. It is used assays applied to the same SLE sera; it is possible that conglutinin only binds to those readily apparent that in this disease some assays complexes solubilised by complement and released give a higher percentage of positives than others with the Raji cell assay ranking highest, due in part to back into the circulation where they no longer play copyright. any pathological role. false-positive reactions with antilymphocyte anti- Antiglobulin binding (whether by monoclonal bodies. However by understanding the specificities of rheumatoid factors, polyclonal rheumatoid factors the different assays and applying several with or IgM antibodies to human immunoglobulins) different probes to the same serum, some degree of depends upon the multivalency of IgM antibodies in characterisation can be achieved. Since immune which each of the binding sites in the Fab portion of complexes are clearly heterogeneous and only some the molecule is of low affinity, but stable binding to are of pathological significance, it is important to apply this type of finger-printing exercise in order to aggregated or complexed immunoglobulins occurs http://jcp.bmj.com/ because of multipoint attachment.47 48 Antiantibody assess whether the immune complexes play any role is also an IgM antibody, but differs from rheumatoid in a particular disease. Table 4 indicates how the size factor in recognising the antigen-binding site at the Fab portion of the immunoglobulin molecule.49 Table 4 Size of IgG immlunie complexes in relationt to Immune complex detection by cell receptor clinical signs in SLE58 binding differs according to the specific receptor Signs NO stildied Size of coniplex (approximiiate molecular involved (Table 2). Where IgG Fc receptors are used, weight in daltons) on September 30, 2021 by guest. Protected monomeric IgG is distinguished from IgG in Sinall Mediurn Large complexed form by the stronger binding of the latter, 312 000 1 0-1 5 x 106 25-40 x 106 owing to their attachment to the cells at multiple Renal only 1 1 1 sites. Cells with complement receptors only react with Renal plus activatedcomplement components, either C3b or C3d. extrarenal 7 7 7 7 Extrarenal None of the techniques is ideal; all give false- only 5 5 0 5 positives and false-negatives and none of them is capable of characterising the immune complex in terms of its constituents. However, when several of IgG- and Clq-binding immune complexes may be different tests with different biological specificities important for renal damage in lupus nephritis. Those are used to analyse sera, a great deal of information of medium size are associated with the renal lesions, can be obtained. whereas extravascular lesions are associated with much larger immune complexes. These observations, Heterogeneity ofimmune complexes in renal disease confirmed by others,51 extend those of Germuth TheWHO collaborative study of 18 different methods et al.52 who showed that in the chronic nephritis J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

Role of circulating immune complexes in renal diseases 1219 induced in animals, nephrotoxicity was also related The normal response to the entry of food antigen to the medium-sized immune complexes. Yet it is ap- into the circulation after eating is the formation of parent from observations on the size of complexes in IgA immune complexes which are presumably Behcet's syndrome,53 in which renal lesions are rare cleared bythishepaticroute. That the type of immune although medium-sized IgG complexes do occur, complex is important for the development of that immune complex size alone is not the only factor disease symptoms is supported by the fact that the governing renal localisation. It is likely that specific response differs in subjects allergic to food where the receptors are involved, but the immune deposits seen complexes after food challenge contain IgG and IgE, on the endothelial surface of the GBM cannot be are Clq-binding and are associated with symptoms explained by our current views on localisation, as of skin itching and wheezing.57 C3b receptors have not been demonstrated on the In Behcet's syndrome, in HSP and in idiopathic endothelial surface. recurrent haematuria of childhood, diseases all The finding of IgG complexes which do not bind associated with high concentrations of IgA com- Clq in the steroid-responsive nephrotic syndrome4 plexes, the more damaging elements of these suggests involvement of one of the IgG subclasses diseases only occur either when there is a switch to which is poor at activating complement (IgG4 and IgG- and Clq-binding complexes, as in Behcet's IgG2). These complexes apparently are not deposited syndrome, or when IgA and IgG complexes occur in the kidney and it is not known whether they are together, as in HSP, nephritis, and recurrent in any way responsible for the heavy proteinuria. The haematuria.58 It is possible that in these diseases the concept of Lagrue et al.,55 who suggested that IgA complexes inhibit the clearance of IgG immune proteinuria was due to a lymphokine causing complexes by polymorphs.59 If this is so, there is no increased vascular permeability, is so far unsub- reason to believe that the same antigen is involved in stantiated, but this type of interrelationship between these two types of immune complexes. immune complexes and T cell immune responses is Table 5 illustrates certain aspects of this "finger- still unexplored in human diseases. The particular printing" exercise for renal diseases in which immune type of immune complex found in the steroid- complexes may play a role. copyright. responsive nephrotic syndrome suggests that complement binding may be a prerequisite for renal Table 5 Different types of immune complexes found in localisation. The complexes found in this disease are certain forms ofnephritis58 large (2-5 x 106 daltons), yet no evidenceof deposition IgG IgA Clq binding can be demonstrated by immunofluorescent techniques. In addition, the findings in Henoch-Schonlein SLE nephritis + + + Acute nephritis + - i purpura (HSP), where the nephritis is associated Steroid-responsive nephrotic syndrome in + i with IgG complexes of similar size to those of the relapse http://jcp.bmj.com/ Henoch-Schonlein purpura without steroid-responsive nephrotic syndrome,56 suggest nephritis - + - that additional factors operate in causing renal Henoch-Schfnlein purpura with localisation and damage. In HSP, although the nephritis + + - IgA nephropathy in remission - + - complexes do not bind Clq, there is evidence of IgA nephropathy during haematuria + + - complement activation since C3, properdin, IgG, Subacute bacterial endocarditis nephritis + - 4 and IgA may be demonstrated by immunofluorescent staining of renal biopsy material. Here the evidence + Denotes raised concentrations in most patients. on September 30, 2021 by guest. Protected suggests that, if complement activation is necessary ± Denotes raised concentrations in a few patients only. for the localisation of complexes in the kidneys in - Denotes concentrations not raised. HSP, it is by the alternative pathway. IgA immune complexes have been demonstrated Is measurement of immune complexes clinicallji in several diseases. Their role is unclear but it is useful ? possible that their formation is the normal response It is important to distinguish between a test which to an antigen entering via a mucosal surface and that may be of diagnostic value and one which is useful theyonlybecomedeposited in tissues as "bystanders" for monitoring the clinical course of patients during when their clearance is impaired. The elegant different treatments. transport system in which the secretory piece on the At present, the detection of soluble immune hepatocyte binds to dimeric or complexed IgA (only complexes in serum does not provide any diagnostic demonstrated in the rat so far4l) and transports the help. It is possible however that in the future, when secretory IgA into the bile with antigen degradation immune complex characterisation is more precise, occurring intracellularly, provides a normal physio- the finding of a particular type of immune complex in logical clearance mechanism for such complexes. a disease will provide diagnostic distinction. J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

1220 Levinsky The question of whether measurement of immune such as lupus nephritis (Fig. 3), though in others, complex concentrations is clinically useful for such as steroid-responsive nephrotic syndrome and monitoring treatment of patients is also difficult to HSP nephritis, no such correlation is obtained answer. Immune complexes, by their very nature, are (Fig. 4). difficult to quantify. They represent a dynamic equilibrium of antigens and antibodies, with _ Abdominal continual immune complex formation and elimin- - Rash ation; in addition the secondary binding of other - Arthralgia Proteinur'ia proteins such as complement components, rheuma- - toid factors and immunoconglutinin, adds to the 80 Plasmapheresis problem of quantification. Furthermore, we have little idea how various forms of treatment alter this c immune complex equilibrium in terms of individual ° 50 constituents, or how their formation and elimination D IgA complexes is affected. Nonetheless, immune complex con- cZ centrations, measured semiquantitatively, may 201------correlate well with disease activity in some diseases nI Igcmplexes C 100 o Cyclophosphamide -! ; s.-.. -f 0 o 0 \77 7 .4* V44

-L-j 1975 ; 76 77 7u) copyright. Fig. 4 Clinical course in a boy with Henoch- Schonlein purpura and nephritis who eventually required haemodialysis for his end-stage renal failure.56 There was no correlation of his progressive renal deterioration with concentrations of circulating IgG or IgA complexes. (Reprinted with permission of the Editor of the Lancet.) http://jcp.bmj.com/ Conclusions The differences shown in the immune complexes in the different renal diseases illustrate how a single technique is not of great use for their detection. Characterisation in terms of size, immunoglobulin class and complement components as well as the on September 30, 2021 by guest. Protected antigen are all necessary in order to assess their role in disease. It must also be emphasised that their mere presence in the circulation does not necessarily mean that they have a causal role. It has become apparent that not only are immune complexes important for normal clearance mechanisms, but they play a vital part in modulating immune responses; indeed some of those detected may be idiotype/anti-idiotype complexes.60 There is also Fig. 3 Clinical course andplasma concentrations of much evidence to indicate how they modulate T cell immune complexes and complement components in an and B cell function.6' Table 6 lists some of the known 18-year-old woman with a post-partum SLE crisis. The Clq-binding immune biological effects of immune complexes. It is only by concentrations of IgG-containing and in complexes correlated very well with her clinical condition. careful characterisation that we may be able, (Reprinted with permission of the editor of the Lancet.) time, to distinguish between those which cause J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

Role of circulating immunie complexes in renal diseases 1221 Table 6 Effects of immune complexes affinity antibody. Clin Exp Immunol 1972;12:121. 16 Cassidy JT, Burt A, Petty R, Sullivan D. Selective IgA Localised inflammatory foci-nephritis deficiency in connective tissue diseases. N Engl J Med Systemic tissue lesions-SLE 1969 ;280:275. Impairment of effector mechanisms-polymorph, eosinophil, K cell 17 Lawrence JS. Rheumatic disease in hypogamma- Modulation of immune response-help, suppression, anti-idiotype globulinaemia patients and their relatives. In: Hypo- network gammaglobulinaemia in the United Kingdom. M RC Special Report Series. No 310. London: HMSO, 1971: 35-44. disease and those which are essential for immune 18 Day NK, Geiger H, McLean R, Michael A, Good RA. C2 regulation. deficiency; recognition of systemic lupus erythematosus. J Clin Invest 1973 ;52:1601. 9 Almeida JD, Waterson AP. Immune complexes in I thank the National Fund for Research into hepatitis. Lancet 1969;ii:983. Crippling Diseases and the Nuffield Foundation for 20 Combes B, Stastny P, Shoney J, et al. Glomerulonephritis generous grant support. with deposition of Australia antigen-antibody complexes in glomerular basement membrane. Lancet 1971 ;ii :234. 21 Gocke DJ, Hsu K, Morgan C. Association between polyarteritis and Australia antigen. Lancet 1970;ii:: 149. References 22 Soothill JF. Detection and measurement of circulating soluble antigen-antibody complexes and anti-DNA WHO Scientific Group. The role of immune complexes in antibodies. Ann Rheum Dis 1977;36, suppl 64. diseases. WHO Tech Rep Ser 1977:606. 23 Levinsky RJ, Soothill JF. The heterogeneity of immune 2Theofilopoulos AN Andrews BS, Urist BM, Morton DL, complexes in disease. In: Peeters H, ed. Proteins and Dixon FJ. The nature of immune complexes in human related subjects. Vol 26. Protides of the Biological Fluids cancer sera. J Immunol 1977;119:657. 26 Colloquium. Pergamon Press, 1979:243. 3Hellstrom KE, Hellstrom 1. Lymphocyte mediated 24 Brostoff J, Johns P, Stanworth DS. Complexed IgE in cytotoxicity and blocking serum activity to tumour atopy. Lancet 1977 ;ii :741. antigens. Adv Immunol 1974;18:209. 25 Miller GW, Nussenzweig V. A new complement function: Paganelli R, Levinsky RJ, BrostoffJ, Wraith DG. Immune solubilization of antigen-antibody aggregates. Proc Natl complexescontainingfood proteins in normal and atopic AcadSci USA 1975;72:418. subjects after oral challenge and effect of sodium 26 Germuth FG, Rodrigues A. Immunopathology of the renal copyright. cromoglycate on antigen absorption. Lancet 1979 ;i :1270. glomerulus. Boston: Little, Brown & Co, 1973. Masson PL, Delire M, Cambiaso CC. Circulating immune 27 Kijlstra A, Knutson DW, V d Lelij A, Van Es LA. Charac- complexes in normal human pregnancy. Nature 1977; teristics of soluble immune complexes prepared from 266:542. oligovalent DNA conjugates and anti-DNA antibodies. 6 Stirrat GM, Redman CWG, Levinsky RJ. Circulating J Immunol Methods 1977 ;17 :263. immune complexes in pre-eclampsia. Br MedJ 1978;i: 28 Ford PM. The effect of manipulation of reticuloendothelial 1450. system activity on glomerular deposition of aggregated 7Unanue ER, Dixon FJ. Experimental glomerulonephritis: protein and immune complexes in two different strains of immunological events and pathogenetic mechanisms. mice. BrJ Exp Pathol 1975 ;56:523. Adv Immunol 1967 ;6: 1. 29 McCluskey RT, Benacerraf B, Potter JL, Miller F. The http://jcp.bmj.com/ 8 Henson PM, Spiegelberg HL. Release of serotonin from pathologic effects of intravenously administered soluble human platelets induced by aggregated immuno- antigen-antibody complexes. 1. Passive serum sickness in globulins of different classes and subclasses. J Clin Invest mice. J Exp Med 1960;11l1: 181. 1973 ;52:1282. 30 Okumura K, Kondo Y, Tada T. Studies on passive serum 9 Cochrane CG, Dixon FJ. In: Miescher PA, Muller- sickness nephritis induced by preformed antigen- Eberhard HJ, eds. Textbook of immunopathology. Vol 1. antibody complexes in the mouse. Lab Invest 1971 ;24: New York: Grune and Stratton, 1968 :94. 383. 0 Fish AJ, Michael AF, Vernier RL, Good RA. Acute 31 Haakenstad AO, Striker GE, Mannik M. The glomerular serum sickness nephritis in the rabbit. Am J Pathol 1966; deposition of soluble immune complexes prepared with 49:997. reduced and alkylated antibodies and with intact on September 30, 2021 by guest. Protected Izui S, Lambert PH, Miescher PA. In vitro demonstration antibodies in mice. Lab Invest 1976;35:293. of a particular affinity of glomerular basement membrane 32 Andres GA, Seegal BC, Hsu KC, Rothenberg MS, and collagen for DNA. A possible basis for a local Chapeau ML. Electron microscopic studies of experi- formation of DNA/anti-DNA complexes in SLE. J Exp mental nephritis with ferritin-conjugated antibody. Med 1976;144:428. Localisation of antigen-antibody complexes in rabbit 12 Pincus T, Haberkern R, Christian CL. Experimental glomeruli following repeated injections of bovine serum chronic glomerulitis. J Exp Med 1968;127:819. albumin. J Fxp Med 1963;117:691. 13 Oldstone MBA, Dixon FJ. Pathogenesis of chronic 33 Mannik M. Clearance and glomerular deposition of disease associated with persistent lymphocytic chorio- circulating immune complexes. In: Peeters H, ed. meningitis viral infection. 1. Relationship of antibody Proteins and related subjects. Vol 26. Protides of the production to disease in neonatally infected mice. J Exp Biological Fluids 26th Colloquium. location: Pergamon Med 1969;129:483. Press, 1979:265. 'Soothill JF, Steward MW. The immunopathological 3- Kijlstra A, Daha MR, Van Es LA. Immune complexes in significance of the heterogeneity of antibody affinity. experimental glomerulonephritis. Behring Inst Mitt 1979; Clin Exp Immunol 1971 ;9:193. 64:45. 15 Alpers JH, Steward MW, Soothill JF. Differences in 35 Gelfand MC, Frank MM, Green 1. A receptor for the immune elimination in inbred mice. The role of low 3rd component of complement in the human renal J Clin Pathol: first published as 10.1136/jcp.34.11.1214 on 1 November 1981. Downloaded from

1222 Levinsky glomerulus. J Exp Med 1975;142:1029. Circulating immune complexes in steroid-responsive 36 Moran J, Colasanti G, Amos N, Peters DK. C3b receptors nephrotic syndrome. N EngIJ Med 1978;298:126. in glomerular disease. Clin Exp Immunol 1977;28:212. 5 Lagrue G, Xheneumont S, Branellec A, et al. A vascular 3 Miller GW, Saluk PH, Nussenzweig V. Complement permeability factor elaborated from lymphocytes. dependent release of immune complexes from the Demonstration in patients with nephrotic syndrome. lymphocyte membrane. J Exp Med 1973;138:495. Biomedicine 1975 ;23:37. 38 Miller GW, Nussenzweig V. Complement as a regulator of 56 Levinsky RJ, Barratt TM. IgA immune complexes in interactions between immune complexes and cell Henoch-Schonlein purpura. Lancet 1979 ;ii:100. membranes. J Immunol 1974;113:454. 57 Brostoff J, Carini C, Wraith DG, Paganelli R, Levinsky 3 Cochrane CG, Hawkins D. Studies on circulating immune RJ. Immune complexes in atopy. In: Pepys J, Edwards complexes.,llI. Factors governing the ability of circulating AM, eds. The Mast Cell-its role in health and disease. complexes to localize in blood vessels. J Exp Med 1968; Pitman Medical, 1979:380. 127:137. 58Levinsky RJ. Immune complexes in health and disoase. 40 Benveniste J, Egido J, Gutierrez-Millet U. Evidence for the University of London: MD Thesis, 1980. involvement of the IgE basophil system in acute serum 59 Wilton JMA. Suppression by IgA of lgG mediated sickness. Clin Exp Immunol 1976 ;26 :449. phagocytosis by human polymorphonuclear leucocytes. 41Peppard J, Orlans E, Payne AWR, Andrew E. The Clin Exp Immlunol 1978;34:423. elimination of circulating complexes containing poly- 60 Rose LM, Lambert PH. The natural occurrence of meric IgA by excretion in the bile. Immunology 1981 ;42: circulating idiotype-anti-idiotype complexes during a 83. secondary immune response to phosphorylcholine 42 Izui S, Lambert PH, Fournie GJ, Turler H, Miescher PA. (abstract). Proc 4th Int Congress Immunology 1980. Features of SLE in mice injected with bacterial lipo- 61 Theofilopoulos AN. Immune complexes in humoral polysaccharide. Identification of circulating DNA and responses; suppressive and enhancing effects. Immunol- renal localisation of DNA/anti-DNA complexes. J Exp ogy Today 1980;1:1. Med 1977;145:1115. 62 Zubler RH, Lange G, Lambert PH, Miescher PA. Detec- 13 Kobayakama T, Louis J, lzui S, Lambert PH. Autoimmune tion of immune complexes by modified 12,1 Clq binding response to DNA, red blood cells and thymocyte test. Effect of heating on the Clq binding by immune antigens in association with polyclonal antibody synthesis complexes and application of the test to systemic lupus during experimental African trypanosomiasis. J Immunol erythematosus. J Immunol 1976 ;116 :232. 1979 ;122:296. 63 Hay FC, Nineham LJ, Roitt IM. Routine assay for the "4 Roberts JL, Lewis EJ. Identification of antinative DNA detection of immune complexes of known immuno- antibodies in cryoglobulinaemic states. Ani J Med 1978; globulin class using solid phase Clq. Clin Exp Immuinol 65:437. 1976;24:396. copyright. 1 McIntosh RM, Kulvinskas C, Kaufman DB. Cryoglobulins 64 Sobel AT, Bokisch VA, Muller-Eberhard HJ. C1 q deviation Il: the biological and chemical properties of cryoproteins test for the detection of immune complexes, aggregates in acute post-streptococcal glomerulonephritis. Int of IgG and bacterial products in human serum. J Exp Arch Allergy Appl Immunol 1971 ;41 :700. Med 1975;142:139. '6 Harrington JC, Fenton JW, Pert JH. Polymer-induced 65 Gabriel A, Agnello V. Detection of immune complexes. precipitation of antigen-antibody complexes: "'Perciflex" The use of radio-immunoassays with Clq and mono- reactions. Immunochemistry 1971 ;8 :413. clonal rheumatoid factor. J Clin Invest 1977;59:990. 17 Eisenberg R. The specificity and polyvalency of binding of 66 Casali P, Bossus A, Carpentier NA, Lambert PH. Solid a monoclonal rheumatoid factor. Immu(nochemistry 1976; phase enzyme immunoassay or radioassay for the 13:355. detection of immune complexes based on their rec- http://jcp.bmj.com/ 48 Dissanayake S, Hay FC, Roitt IM. The binding constants ognition by conglutinin: conglutinin binding test. Cli,, of IgM rheumatoid factors and their univalent fragments Exp Immunol 1977 ;29 :342. for native and aggregated human IgG. Immuinology 1977; 67 Theofilopoulos AN, Wilson CB, Dixon FJ. The Raji cell 32:309. radio-immunoassay for detecting immune complexes in 19 Milgrom F, Dubiski S, Wozniczko G. Human sera with human sera. J Clin Invest 1976;57:169. "anti-antibody". VoxSang1956;1:172. 68 Luthra HS, MacDuffie FC, Hunder GG, Samayoa EA. ;"6 Lambert PH, Dixon FJ, Zubler RH, et al. WHO col- Immune complexes in sera and synovial fluids of laborative study for the evaluation of eighteen methods patients with rheumatoid arthritis. Radioimmunoassay for detecting immune complexes in serum. J Clin Lab with monoclonal rheumatoid factor. J Clin Invest 1975; on September 30, 2021 by guest. Protected Immunol 1978 ;1: l. 56:458. 5 Van Aarden L. Immunology of DNA. University of Amster- 69 Penttinen K, Makela 0, Vaheri A. Soluble antigen- dam: MD Thesis, 1977:141. antibody complexes and platelet aggregation. Acta 52 Germuth FG, Senterfit LB, Dreesman GR. Immune Pathol Microbiol Scand 1969;77:309. complex disease V. The nature of the circulating com- 70 MacLennan ICM. Competition for receptors for immuno- plexes associated with glomerular alteration in the globulin on cytotoxic lymphocytes. Clin Exp Immunnol chronic BSA-rabbit system. Johns Hopkins Med J 1972; 1972 ;10:275. 130:344. Levinsky RJ, Paganelli R, Lehner T. Immune complexes and their characterisation in Behqet's syndrome and recurrent oral ulcers. In: Lehner T, Barnes CG, eds. of Beh(et's syndrome-clinical and immunological features. Requests for reprints to: Dr RJ Levinsky, Department New York: Academic Press, 1979. Immunology, Institute of Child Health, 30 Guilford 'Levinsky RJ, Malleson PN, Barratt TM, Soothill JF. Street, London WC1, England.