0022-202X/ 80/ 7405-0333$02.00/ 0 THE .J OURNAL OF I NVESTIGATIVE DERMATOLOGY, 74 :333-338, 1980 Vol. 74. No. 5 Copyright © 1980 by T he Williams & Wilkins Co. Printed in U. S.A. Physicochemical and Functional Relationships of Immune Complexes

MART MANNIK, M.D. Division of Rheumatology, Department of Medicine, University of Washington, Seattle, Washington, U.S.A .

The biological properties of immune complexes depend immune complexes. In any given immune complex the number on the nature of antigens and antibodies comprising of antigen and antibody molecules may vary, depending on the these complexes. The lattice of immune complexes influ­ characteristics of each of the reactants and the nature of the ences t h eir tissue deposition, complement activation, antigen-antibody union. The biological properties of immune and interaction with Fe receptors. The lattice of immune complexes are related to the properties of the molecules forming complexes d epends on the valence of antigens, the anti­ the complexes, as well as the number of reactants in each gen-antibody ratio, the association constant of these complex. reactants and the concentration of antig ~n and antibody. Antigens are defined as substances that interact specifically Kupffer cells effectively remove large-latticed immune with availa ble a ntibodies or sensitized l ymphocytes. The term complexes from the circulation due to their Fe receptors. immunogen, on the other hand, is reserved for a substance that This system is saturable, leading to prolonged circula­ upon administration to a s uitable host will elicit a n immune tion of the complexes a nd enhanced deposition in tissues. response. This distinction is made because all substances that Small-latticed immune complexes are slowly removed react with antibodies or with sensitized lymphocytes do not from the circulation by yet unidentified mechanisms. necessarily induce an immune response. The actual portion of The renal glomerulus serves as an example of immune an antigenic molecule that reacts with the a ntibody combining complex deposition from the circulation. Only large-lat­ site or with the s pecific receptor on a sensitized lymphocyte is ticed complexes are deposited in the glomerular capil­ defined as the antigenic determinant. The number of antigenic lary wall in the subendothelial area and the m esangial determinants on a molecule defines its valence for the interac­ matrix. An influx of bone marrow-derived monocytes tion wi th a specific antibody. An antigenic molecule may have participates in the disposal of immune complexes depos­ the valence of one for a given antibody specificity, but a number ited in these areas of the glomerulus, the resident mes­ of naturally occurring molecules have repeating chemical struc­ angial cells do not phagocytize these substances. The tuTes, thus g iving a n antigenic valence of more t han one. Fur­ subepithelial d eposits of immune complexes appear to th ~ rmor e, a given molecule may have different antigenic deter­ be locally formed and not deposited from the circulation. minants and thus able to react with antibodies of different specificities. Thus, a macromolecule may be mutivalent with respect to one or more specificities. The number of antigenic Immune complexes cause a variety of clinical disorders, af­ determinants of a molecule profoundly influences the kind of fecting various sizes of blood vessels, renal glomeruli, renal antigen-antibody complexes that may form with specific anti­ tubules, thyroid gland, c horoid plexus and other organs. The bodies. For detailed discussion of antigens and antigenic deter­ pathogenic immune complexes are either deposited from the minants t he interested r eader should consult immunochemical circulation or locally formed. In local formation of immune texts (1, 2]. complexes the antigens are part of the· organ or unrelated Antibodies, as the other essential constituents of immune antigens are selectively deposited in the organ and antibodies complexes, may belong to IgG, IgA, IgM, IgD or IgE class of from the circulation react with these antigens. In this patho­ immunoglobulins. IgG, monomeric IgA, IgD and IgE have a genic mechanism characteristically one organ system i s in­ valence of 2, i. e., each of these molecules has 2 binding sites volved and immune complexes are not present in the circula­ for a given antigen. Dimeric IgA has a valence of 4. lgM has a tion. On the other hand, when immune complexes are present valence of 5 or 10, depending on th e nature in the circulation, usually more than one organ is involved due of antigen molecules. For detailed discussion to deposition of these materials from circulation. In this a1·ticle of various a ntibody molecules, the reader should consult appropriate reviews [1,3,4]. the nature and biological properties of immune complexes will When the antigen-antibod be briefly reviewed. The relationship between the composition y union occurs, a variety of im­ mune complexes m ay form, ranging from of immune complexes and their fate in cil:culation will be an immune complex discussed with particular emphasis on the role of the mononu­ consisting of 1 antigen m olecule a nd 1 sntibody m olecule (Ag ~, Abt) to immune complexes clear phagocyte system in removal of circulating immune com­ consisting of many molecules of each r eacta plexes. Since detailed experimental data ar e not available on n t. The lattice of immune complexes thus reflects the relationship between the composition of immune complexes the number of antigen and number of antibody molecules in each complex. T he lattice and their deposition in skin, this relationship will be examined of an immune complex in turn for renal glomeruli. influences its biological properties. The lattice of immune complexes is influenced by several THE NATURE AND PROPERTIES OF IMMUNE variables. As all·eady pointed out, the valence of most antibodies COMPLEXES is 2. Detailed studies with polymeric IgA and IgM have not been conducted. The valence of antigen molecules significantly Antigens and antibodies aTe t he essential constituents of all influences the lattice of immune complexes that can be formed with given antibodies. A monovalent a ntigenic molecule can The work from our laboratories was in part supported by Resea1·ch only form Ag2Ab1 complexes and larger lattices or immune Grant AM 11476 and Research Training Grant T32AM7108, b oth from precipitates can not the National Institute of Arthritis, Metabolism and Digestive Diseases. be generated. Bivalent antigens, depending Reprint req uests to: Mart Man nik, M.D., Division of Rheumatology, on the distance between the antigenic determinants, may form Department of Medicine, Uni versity of Washington, Seattle, Washing­ Ag t Ab ~, Ag2Abz, cil'cular Ag3Ab3 or larger open or closed com­ to n 98195. plexes [5]. Such small-latticed immune complexes are soluble. Abbreviations: Only multivalent an tigens can form immune complexes with HSA: human serum albumin high degrees of lattice and undergo immune precipitation. MPS: mononuclear phagocyte system With multivalent antigens the molar r atio of antigens and 333 334 MANNIK Vol. 74, No.5 antibodies influence t he degree of lattice formation as best molecules attach to these receptors weakly, but do not lead to illustrated by the classic precipitin curves. At the point of interiorization of the molecules. Attachment of large-latticed equivalence maximum amount of antigen- antibody precipitate complexes (defined as containing more than 2 antibody mole­ is formed and free antigen and free antibody are not detectable cules) results in firmer attachment of the complexes and cul­ in the supernatant. Addition of excess antigen beyond the point minates in their phagocytosis. Evidence for these conclusions of equivalence leads to formation of soluble immune complexes has been obtained by examining complexes with oligovalent and a decrease in the a mount of immune precipitate. At rela­ and multivalenc antigens [14], immune complexes with known tively low degrees of antigen excess soluble large-latticed im­ degrees of lattice formation [15, 16], and with covalently cross­ mune complexes are formed, but with increasing a ddition of linked immune complexes with discrete degrees of lattice for­ excess antigen only small-latticed immune complexes are ob­ mation [17]. Complement is not essential for the attachment of tained. For example, with high degrees of excess human serum soluble immune complexes to the Fe receptor of monocytes albumin (subsequently abbreviated HSA) to a given amount of [15]. The phagocytosis and degradation of soluble immune antibodies to HSA only Ag,Ab, and Ag2Ab2 complexes were complexes, however, was facilitated by the presence of comple­ formed [6]. Fw·thermore, the absolute concentration of antigen ment [18]. and antibody will influence the degree of lattice formation. At The interaction of soluble immune complexes, containing IgG a given degree of antigen excess more small-latticed immune molecules, with the Fe receptors is of fu ndamental importance complexes are formed at microgram concentrations of the reac­ to the fate of immune complexes in circulation, as will be tants than at the same antigen-antibody ratio in milligram discussed in the next section. Furthermore, during the phago­ concentration of the reactants. cytosis of large-latticed immune complexes, particularly when The association constant between the a ntigen and antibody these substances are attached to nonphagocytizable surface, will influence lattice formation. When the association constant lysosomal enzymes spill in to the surroundings. This event is is low, small-latticed immune complexes tend to be formed, thought to be important in leading to tissue damage after whereas large-latticed immune complexes can form when the immune complex deposition or formation in tissues. The phag­ association constant is high and the antigen-antibody ratio is ocytosis of immune complexes also causes activation of macro­ appropriate [7]. phages with additional biologic properties as discussed earlier The immunochemical characteristics of immune complexes in this workshop. in human diseases have not been examined in sufficient detail to relate their features to the disease processes and disease outcome. As will be discussed below, the lattice of immune THE FATE OF CIRCULATING IMMUNE COMPLEXES complexes influences the fate and tissue deposition of these The concentration of immune complexes in circulation at any materials in experimental animals. given time depends o n the rate of immune complex formation The biological properties of immune complexes relevant to and on the rate of immune complex removal. The rate of immune complex diseases are deposition in tissues, activation immune complex formation is a function of the availabili ty of of the complement systems and interaction with cell receptors. specific a ntigens a nd antibodies to these antigens. In human These properties of immune complexes depend on the class and diseases and in spontaneous animal models these variables have subclass of antibodies in the complexes, the degree of lattice not been quantified. The rate of immune complex removal from formation and to a lesser degree, on the properties of antigens circulation is a function of uptake of immune complexes by the in the complexes. Other biologic properties of immune com­ mononuclear phagocyte system (formerly termed the reticulo­ plexes, including their influence on the immune response of the endothelial system) and deposition of immune complexes in host and alterations of lymphocyte functions, are less well tissues. Quantitative studies in antigen-induced acute or chronic understood on the molecular level at this time, but have been serum sickness models disclosed that only very small propor­ reviewed elsewhere [8]. tions of antigen were deposited in renal glomeruli during the Deposition of immune complexes in tissues will be discussed disease process [19], yet these small amounts of immune com­ below. A detailed discussion of the mechanisms of complement plexes sufficed to cause tissue damage and organ failure. activation are beyond the scope of this review, but other sources The fate of immune complexes in circulation h as been ex­ are available [9,10]. Human IgG1, IgG2, and lgG3 are effective amined quantitatively by the injection of preformed immune in activating the classical complement pathway, whereas lgG4 complexes with known degrees of lattice formation. These is ineffective. Among the other classes of im.munoglobulins only studies have been confined to the IgG class of antibodies. The IgM can activate the complement system through the classical fate of circulating immune complexes was examined in unim­ pathway. The alternative complement pathway is also activated munized mice [20,21], rabbits [22,23], a nd monkeys [24] by by the same subclasses and classes of antibodies that activate injection of preformed immune complexes. The lattice of im­ the system through the early complement components. This mune complexes, the status of the mononuclear phagocyte activation occurs through the C3b-dependent loop of the alter­ system (subsequently abbreviated MPS), the nature of antigens native pathway. Effective complement activation occurs when in immune complexes and the nature of antibodies in immune the lattice of immune complexes contains more that 2 or 3 IgG complexes alter the uptake of immune complexes by the MPS. molecules. The molecular mechanisms for complement com­ Large-latticed immune complexes, defined as containing ponent binding and activation an not yet settled [11,12]. more than 2 a ntibody molecules (greater than Ag 2Ab2) were Fe receptors (also called IgG receptors) mediate the interac­ removed relatively rapidly when injected into unimmunized tion of immune complexes with neutrophils, monocytes (includ­ mice [20,21], rabbits [22,23] or monkeys [24]. The disappear­ ing tissue macrophages and Kupffer cells), platelets and certain ance of these complexes was described by a single exponential T and B lymphocyte populations. These receptors are specific function. The half-life of these large-latticed complexes was for immunoglobulin G and do not interact with other classes of related to the dose of injected materials; with increasing dose immunglobulins. The Cy3 domain (third constant homology the rate of removal of these complexes decreased, and the region) of the lgG molecules interacts with this receptor [ 4]. In clearance velocity of large-latticed immune complexes reached addition to the Fe receptors, phagocytic cells a nd some lym­ a plateau, indicating saturation of the system [20]. The major phocytes possess complement receptors. Mouse monocytes pos­ organ for the uptake of the large-latticed circulating immune sess 2 distinct Fe receptors, one for aggregated IgG and one for complexes was the liver in mice, rabbits, and monkeys the IgG2a subclass of mouse IgG [13]. Such differences have [20,23,24]. As increasing doses were injected, also specific he­ not yet been defined for human monocyte receptors. patic uptake reached a plateau, providing independent evidence The lattice of immune complexes influences their interaction for the saturation of the uptake of circulating immune com­ with the Fe receptors of phagocytic cells. Monomeric IgG plexes by the liver [20]. Kupffer cells are responsible for the May 1980 IMMUNE COMPLEXES 335 hepatic uptake of immune complexes and with increasing doses nized mice as a function of the epitope density [35]. Small­ of complexes the Fe receptors are saturated. Kupffer cells are latticed immune complexes prepared with highly conjugated derived from bone-marrow monocytes [25] and uniquely suited antigen were also removed faster from circulation than similar for cleru·ance of circulating materials since the hepatic sinusoids complexes prepared with minimally col).jugated antigen [36]. lack endothelial cells, thus leaving the Kupffer cells directly These observations indicate that the antigen can influence the exposed to circulating substances [26]. The saturability of the fate of immune complexes independent of the lattice structure. MPS by other circulating materials, not dependent on Fe The fate of immune complexes in circulation can be altered receptors, is well established [27 ,28]. by the nature of antibodies in the complexes. Human IgA and The role of the complement system in hepatic removal of IgG2 and IgG4 are ineffective in interacting with the Fe receptor circulating immune complexes was examined by depleting com­ of phagocytic cells [9]. Therefore, immune complexes contain­ plement components with cobra venom factor or with aggre­ ing these antibodies might persist longer in circulation than gated human IgG in rabbits and administering small doses of complexes containing antibodies that react effectively with Fe preformed soluble immune complexes. The clearance kinetics receptors. Support for this possibility was obtained by exam­ [22] and quantitative hepatic uptake [23] were not altered as ining the fate of immune complexes prepared with IgG mole­ compru·ed with control animals. Furthermore, the administra­ cules with reduced and alkylated interchain disulfide bonds. tion of large doses of immune complexes to mice treated with Such antibodies formed lattice structures complarable to com­ cobra venom factor showed no alteration of the kinetics of plexes made with intact antibody molecules [22]. Immune com­ removal of preformed, soluble immune complexes, (Bockow B, plexes prepared with reduced and alkylated IgG molecules Mannik M: unpublished observations), as compru·ed to un­ reacted ineffectively with monocyte receptors in vitro [15]. treated mice. When complexes with reduced and alkylated antibodies were With the injection of small doses of immune complexes into injected into mice [21], rabbits [22,23], or monkeys [24], the rabbits, the splenic uptake accounted for less than 1% of im­ large-latticed immune complexes persisted in circulation due to mune complexes removed from the circulation [23]. With lru·ge, decreased hepatic uptake and were removed at rates compa­ saturating doses of immune complexes the specific splenic up­ rable to small-latticed immune complexes. As a result of pro­ take increased to 10% of the amount taken up by the liver longed circulation of these materials, increased glomerular dep­ [21]. On the other hand, antibody-sensitized red cells are pri­ osition occurred [33]. marily taken up by the spleen, mediated through the Fe recep­ tors [29]. With this system saturation of splenic uptake has been found in human immune complex diseases [30] (reviewed GLOMERULAR DEPOSITION OF IMMUNE by Lawley in this workshop). The saturation of the hepatic COMPLEXES uptake of immune complexes, however, has not yet been ex­ The development of glomerulonephritis due to immune com­ amined in human disorders but has been documented in a plexes is well established experimentally. The presence of im­ chronic immune complex disease of mice with aggregated IgG mune complexes in glomeruli in human diseases is ext~msive l y as a surrogate for immune complexes [31]. documented. About 80 percent of glomerulonephritis appears The small-latticed immune complexes, defined as containing to be caused by immunologic mechanisms. Only about 10 1 or 2 antibody molecules (i.e., Ag2Abz, Ag2Ab1 or Ag1Ab1) were percent of these result from antibodies to the glomerular base­ removed more slowly from the circulation than large-latticed ment membrane (Goodpasture's syndrome) and the remainder immune complexes, but faster than antibodies alone ru·e caused by immune complexes [37]. In Goodpastme's syn­ [21,22,24]. The fate of these materials was best described by drome the antibodies to the glomerular basement membrane exponential curves composed of 2 components. The flrst expo· ru·e diffusely localized on the basement membrane by immu­ nential component was attributed to equilibration between the nofluorescence microscopy. In contrast, immunofluorescence intra- and extravascular spaces. The second exponential com­ microscopy reveals lumpy-bumpy deposits along the glomerular ponent was ascribed to catabolism. The site of catabolism and basement membrane in glomerulonephritis caused by immune the physiological consequences of these complexes have not complexes. been fully explored. Of note is that even the catabolic site of In human diseases and in experimental models immune com­ IgG has not yet been established. The conclusion on the fate of plexes ha~e be~n recognized in the subendothelial, mesangial, immune complexes, reached by injection of preformed immune and subep1thehal areas of glomeruli. In a given disease, such as complexes with known degrees of lattice formation, have been systemic lupus erythematosus, complexes may exist in all these supported by experiments where isolated complexes with vru-y­ areas to varying degrees. Furthermore, in the chronic serum ing degrees of lattice formation were injected into animals. In sickness I_Dlysosomes in the glomerular mesangium and not in the resident cells of the mesangium that lack the giant lyso­ somes. Therefore, the conclusion was reached that the mesan­ Resolution of glomerular immune co mplexes by administration of gial cells do not participate· in phagocytosis of excess antigen, demonstrated by immunofluorescence microscopy. At deposited im­ mune time 0 mice received HSA-antiHSA immune complexes prepared with complexes, fmthermore, the hypercellularity of the glo­ reduced and ·a]kylated rabbit antibodies. The kidney sections in (A) merular area resulted from influx of marrow-derived mono­ and (C) were obtained from a mouse sacrificed at 24 hr for control cytes. It is of interest that the accumulation of bone-marrow purpose. The kidney sections in (B) and (D) were obtained from a derived monocytes in glomeruli of experimental animals was mouse given 40-fold excess antigen (HSA) at 12 hr and sacrificed at 24 associated with the development of proteinuria [ 42]. Thus, the hr. Section (A) and (B) were stained with sheep antibodies to rabbit marrow-derived Kupffer cells in the liver play an important lgG, conjugated with flu orescein isothiocyanate-(A) shows 4+ staining role in clearing large-latticed, pathogenic immune complexes and (B) shows no staining. Sections (C) and (D) were stained with from the circulation. When these complexes are deposited in rabbit antibodies to HSA, conjugated wtih flu erescein isot hiocyanate­ glomeruli, bone-marrow derived monocytes (C) shows 3+ staining a nd (D) shows no staining. With the excess participate in dis­ posal of these antigen administration both antigen and antibody were removed from deposited complexes. During this process of the glomeruli. phagocytosis they may release lysosomal enzymes that cause damage to structural components of the glomeruli. The mechanisms for development of subepithelial deposits remained unclear. The recent and elegant experiments of tors in sim.ilar studies [7,39], suggesting that circulating small­ Couser et al [ 43], however, indicate that at least with the Fx1A latti ced immune complexes did not reach this area during the antigen, derived from the brush border of proximal renal tu­ course of experiments up to 96 hr. Experiments with adminis­ bules, the immune complexes with this antigen are locally tration of excess antigen after the administration of immune formed in the subepithelial space and not deposited from the complexes showed that when the deposited immun e complexes circulation. This possibility is supported by the finding of no were converted in vivo to small-latticed immune complexes, circulating immune complexes in patients with m embranous they were released from the glomeruli [ 40]. In these experi­ glomerulonephritis, characterized by only subepithelial deposits ments soluble immune complexes were injected into unimmu­ of immune complexes [ 44,45]. These findings point again to the nized mice, 12 hr was allowed for glomerular deposition of need to examine the nature and fate of antigens in human immune complexes and then a 40-fold excess dose of antigen immune complex diseases. was given and mice were serially sacrificed. By 12 hr after the The deposition of circulating immune complexes in the skin injection of excess antigen no antibody or antigen could be may be governed by other principles, but experimental methods detected in glomeruli by immunofluorescence microscopy (see are on hand to examine this problem. Figure). Electron microscopy showed no electron dense deposits in these specimens. Control mice sacrificed at the same time showed abundant deposits. These experiments clearly indicate REFERENCES that the subendothelial area and mesangial matrix are accessi­ J. Day ED: Advanced Immunochemistry. Baltimore,· Williams a 1~d ble to some circulating antigens. The molecular mechanism of Wilkens Co., 1972 entrapment of large-latticed immune complexes in subendo­ 2. Kabat EA: Structural Co ncepts in Immunology and Immunochem­ thelial area and mesangial matrix remain unclear, but when istry , 2nd edition, New York, Holt, Rinehart and Winston, 1976 3. Natvig JB, Kunke these complexes are converted to small-latticed complexes l HG: Human immunoglobulins: classes, sub­ classes, genetic variants, and idiotypes. Adv lmmunol 16:1-59, (Ag,Ab1 or Ag2Ab2}, they are readily released from these areas. 1973 The sequence of glomerular deposition of immune complexes 4. Dorrington KJ: The stru ctural basis for the fun ctional versatility was examined by transmission electron microscopy [33]. After of immunoglobulin G. Can J Biochem 56:1087-1101, 1978 5. Hyslop NE Jr, t he injection of a single dose of immune complexes, electron Dourmashkin RR, Greene NM, Porter RR: The fixation of complement a nd the activated fu·st component (C1) dense deposits first appeared in the fenestrae of the endothelial of complement by co mplexes form ed b etween antibody and di­ cells. Deposits in the s ubendothelial area and in the m esangial valent hapten. J Exp Med 131:783-802, 1970 area increased later. When the large-latticed complexes were 6. Arend WP, Mannik M: Determination of soluble immune co mplex molar com no longer present in the circulation, the endothelial f enestrae positio n and antibody association constants by a m­ monium sulfate precipitation. J Immunol 112:451-461, 1974 became clear and subendothelia l deposits disappeared leaving 7. Koyama A, Niwa Y, Shigematsu H, Taniguchi M, Tada T: Studies only deposits in the mesangial matrix. By 96 hr after the on passive serum sickness. II. Factors determin ing t he loca liza- 337 May 1980 IMMUNE COMPLEXES tion of antigen-antibody complexes in the murine renal glomer­ ceedings of the 6th International Convocation on Immunology. ulus. Lab Invest 38:253- 262, 1978 Edited by JF Mohn. S. Krager, 1979, pp 212-216 8. H aakenstad AO, Mannik M: The biology of immune complexes. 37. Wilson CB, Dixon FJ: Diagnosis of immunopathologic renal disease Autoimmunity. Edi ted by N Tala!. New Yo rk, Academic Press, Kidney lnt 5:389- 401, 1974 1977, pp 277- 360 38. Germuth FG Jr., Senterfit LB, Dreesman GR. Immune complex 9. Spiegelberg HL: Biological activities of immunoglobulins of differ­ disease. V. T he nature of the circulating complexes associated ent classes and subclasses. Adv Immunol 19:259-294, 1974 with glomerular alterations in the c hronic BSA-rabbit system. 10. Gotze 0, Mi.iller-Eberhard HJ: The alternative pathway of comple­ J ohns H opkins Med J 130:344 - 357. 1972 ment activation. Adv Immunol 24: 1-35, 1976 39. Okumura K, Kondo Y, Tada T : Studies on passive serum sickness. 11. Metzger H : Effect of antigen binding on the properties of antibody. I. The glomerulru· fine structure of serum sickness nephritis Adv Immuno\18:1 69- 207, 1974 induced b y preformed antigen-antibody complexes in the mouse. 12. Goers JW, Ziccardi RJ, Schumaker VN, Glovsky MM: The mech­ Lab Invest 24:383- 391, 1971 anism of activation of the first component of complement by a 40. Mannik M, Striker GE: Deposition and removal of glomerular univalent hapten-IgG antibody complex. J Immunol 118:2182- immune complexes; Relationships with the mononucleru· phago­ 2191 , 1977 cyte system. Immune Mechanisms in Renal Disease. Edited by 13. H eusser CH, Anderson CL, Grey HM: Receptors for IgG: subclass AF Michael, NB Cummings. P lenum Press, in press specificity of receptors on different mouse ce ll types and the 41. Striker GE, Mannik M, T ung MY: Role of mruTow-derived mono­ definition of two distinct receptors o n a macrophage cell line. J cytes and mesangial cells in removal of immune complexes from Exp Med 145:1316-1327, 1977 renal glomeruli. J Exp Med 149:127-136, 1979 14. P hillips-Quagli ata JM, Levine BB, Quagliata F, Uhr JW: Mecha­ 42. Schreiner GF, Cotran RS, Pardo V, Unanue ER: A mononucleru· nisms underlying binding of immune complexes to macrophages. cell component in experimental immunological glomerulonephri­ J Exp Med 133:589-601, 1971 . tis. J Exp Med 14 7:369- 384, 1978 15. Arend WP, Mannik M: In vitro adherence of soluble immune 43. Couser WG, Steinmuller DR, Stilmant MM, Salant DJ, Lowenstein complexes to macrophages. J Exp Med 136:514- 531, 1972 LH: Experimental glomerulonephritis in the isolated perfused rat 16. Hawkins D, Peeters S: The response of p olymorphonuclear leuko­ kidney. J Clin Invest 62: 1275-1287, 1978 cytes to immune complexes in vitro. Lab Invest 24:483-491, 1971 44. Ooi YM, Ooi BS, Pollak VE: Relationship of levels o f circulating 17. Segal DM, Hurwitz E: Binding of affini ty cross-linked oligomers of immune complexes to histologic patterns of nephritis: A compru·­ IgG to cells bearing Fe receptors. J Immunol 118:1 338-1347, 1977 ative study of membranous g lomerulonephropathy a nd diffuse 18. Van Snick JL, Masson PL: The effect of complement o n the proliferative glomerulonephritis. J Lab Clin Med 90:891-898, inge tion of soluble antigen-antibody complexes and lgM aggre­ 1977 gates by mouse peritoneal macrophages. J Exp Med 148:903-914, 45. Tung KSK, Woodroffe AJ, Ahlin TD, Williams RC Jr., Wilson CB: 1978 Application of the solid phase Clq and Raji cell radioimmune 19. Wilson CB, Dixon FJ: Quantitation of acute and chronic serum assays fo r the detection of circulating immune complexes in sickness in the rabbi t. J Exp Med 134:7s-18s, 1971 glomerulonephritis. J Cli n Invest 62:61-72, 1978 20. Haakenstad AO, Mannik M: Saturation of the reticuloendothelial system with soluble immune complexes. J Immunol 112:1939- 1948, 1974 DISCUSSION , M: T he disappearance kinetics of solu ble 21. Haakenstad AO, Mannik of solubilization of immune complexes prepared with reduced and alkylated anti­ DR . GIGLI: What is the s ignificance in vivo bodies and with intact antibodies in mice. Lab Invest 35:283-292, immune-complexes by C3? 1976 DR. MANNJK: Oul' feeling is t hat it may not have much sign'ificance 22 . Mannik M, Arend WP, Hall AP, Gilliland BC: Studies on antigen­ in the whole animal. We have not, however, examined this question. antibody complexes. I. Elimination of soluble complexes from DR. GIGL J: The concept about the in-situ formation of immune rabbit c irculation. J Exp M ed 133:713-739, 1971 complexes is very interesting. Could you please comment on t he obser­ 23. Arend WP, Mannik M: Studies on antigen-antibody complexes. II. vation that DNA combines to the basement membrane in both the e uptake of soluble complexes in normal Quantification of tissu glomerulus a nd also in the s kin. Is the rapid disappearance of DNA and complement-depleted r abbits. J Immunol 107:63-75, 1971 pointed out when it is injected alone, due 24. Mannik M , Arend WP: Fate of preformed immune complexes in from the c irculation, as you rabbits and rhesus monkeys. J Exp Med 134:19s- 31s, 1971 to binding to these areas? 25. Crofton RW, Diesselhoff-den Dulk MMC, van Furth R: T he origin, DR. MANNIK: This is a n ru·ea we ar e very much interested in. We kinetics, and characteristics of the Kupffler cells in the normal looked quantitatively where DNA goes and found that over 90% of it is steady state. J Exp Med 148: 1-17, 1978 accounted for in the liver. The c urious t hing is t hat, most likely, the 26. Wisse E, Daems WT: Fine structural study on the sinusoidal living breakdown occm·s not in the phagocytic cells, but on the cell surface by cells of rat liver, Mononucleru· Phagocytes. Edited by R van an exoenzym e. . ia, F.A. Davis, Co., 1970, pp 200- 210 Furth. P hiladelph It is interesting t hat in some patients who have SLE acerraf B, Halpern BN: Quantitative study of the DR. PROVOST: 27. Biozzi G, Ben e complexes, e.g., by the granulopectic activity of the reticuloendothelial system. II. A one finds very high quantities of serum immun study of the kinetics of the granulopectic activity of the R.E.S. in Raji cell assay, yet the clinical evidence of renal disease, as well as relation to the dose of cru·bon injected. Relationship between the other manifestations of immune complexes, is really nonexistent. It weight of the organs and their activity. Br J Exp Pathol 34:44 1- may very well be t hat in-situ complex formation does play a big role in 457, 1953 the clinical manifestation of thi disease. 28. Norman SJ: Kinetics of phagocytosis. II. Analysis of in vivo clear­ DR. GIGLT: Something that is very interesting is that nobody who etitive inhibition between sim­ ance wit h demonstration of comp Clq binding assay takes into account the fact that DNA is a es. Lab Invest 31:161-169, 1974 uses the ilar and dissimilar foreign par ticl fantastic binder of Clq. ank MM, Schreiber AD, Atkinson JP, J affe CJ: Pathophysiology 29. Fr do not use Clq of immune hemolytic anemia. Ann Intern Med 87:210- 222, 1977 DR. MAN NIK : That is t he reason why many of us 30. Frank MM, Hamburger MI, Lawley TJ, Kimberley RP, Plotz PH: binding assays. Defective reticuloendotheli al system F c- receptor function in sys­ DR. PROVOST: We're all measuring immune complexes in sera but temic lupus e rythematosus. N E ng! J M ed 300:518-523, 1979 what we just heru·d is t hat maybe t here is some in-situ complex 31. Hoffstein PE, Swerdlin A, Bertell M, Hill CL, Venverloh J, Broth­ formation which we have no way of m easuring with the tools we erson K, Klahr S: Reticuloendothelial and mesangial function in presently use to evaluate immune complex disease. murine immune co mplex glomerulonephritis. Kidney Int 15:144- DR. JORDON: I'd better make a comment to defend the C lq binding 159, 1979 assay. We did some studies with Fred McDuffy a few yeru·s ago, RL, Segal OM: Stable model 32. Plotz PH, Kimberly RP, Guyer comparing several of these sensitive radio assays for measuring immune nity labelling hap­ immune complexes produced b y bivalent affi cell and Clq binding assay Mol Immunol, 16:72 1-729 complexes. In lupus, for instance, the Raji tens: in vivo survival. play 33. Haakenstad AO, Striker GE, Mannik M: The glomerular deposition were almost identical. Do complement receptors in the kidneys of soluble immune complexes prepared with reduced and alkyl­ any role in the deposition of complexes there? ated antibodies and with intact antibodies in mice. La b Invest DH. MANNIK: That has not been carefully sorted out. When you ask 35:293-301, 1976 what cells have the receptors; thei1· presence has been claimed on all 34. Klaus GGB, Mitchell GF: The influence of epitope density on the cells, but the best studies s how them to be the epithelial cell, and that immunological properties of hapten-protein conjugates. II. The is, of course, outside t he g lomerular basement membrane. The com­ tabolism of heavily a nd lightly conjugated in vivo and in vitro me plexes, as I illustrated above, which h ad deposited in the lru·gest ein. Immunology 27:699-710, 1974 prot he ones that were ineffective in complement 35. Mannik M, Haakenstad AO: Circulation and glomerular d eposition am ounts in glomeruli, ru·e t of immune complexes. Arth Rheum 20:s148- s157, 1977 fixation, and therefore their deposition is not mediated by complement 36. Mannik M, Jimenez RAH: T he mononuclear phagocyte system receptors on glomerular structures. Others have suggested an Fe recep­ (MPS) and immune complex diseases. Immunopat hology. Pro- tor in glomeruli. Obviously some mechanism k eeps the complexes, 338 MANNIK Vol. 74, No.5

when they are large enough, entrapped in glomeruli and they are its to see them. Some of the examples, I showed, in l hr after injection released when converted to small complexes. of immune complexes, one expects t hat some complexes are deposited; DR. WUEPPE R: Many of us see deposits of the size and shape of . yet by immunofluorescence or by e lectronmicroscopy we did not see immune c omplexes at the locations where t hey may land in tissue, them. Considered in terms o f what must happen in the glomerulus, the including skin, and frequently see complement components but have complexes in the circulation were between 11 and 19 Svedberg units. great difficul ty demonstrating immunoglobins, which I presume a re One can not really see these complexes by the kind of resolution used buried beneath these complement components. Curtis Wilson, for ex­ fo r electron microscopy of tissues. The e lectron dense deposits t hat one ample, at times has looked at kidneys with 125-I antigen and has been recognizes homs later are t he result of further accumulation of com­ able to measure the label in the kidney. Yet, by no fluorescent antibody plexes or actual immune precipitate formation due to local concentra­ technique could he find antigen. tion of complexes. DR. MANNIK: Obviously, one must have a minimal amount of depos-

Acknowledgment

In order for the Society for Investigative Dermatology to generate additional funds and further expand its activities in the fie ld of dermatology, a new class of membership, known as Corporate Sustaining Membership, has been established. The Society wishes to acknowledge the support of the following companies, who are Corporate Sustaining Members: BURROUGHS WELLCOME COMPANY DERMIK LABORATORIES, INC. ELI LILLY AND CoMPANY JOHNSON & JOHNSON N EUTROGENA CORPORATION OWEN LABORATORIES PROCTER AND GAMBLE REED & CARNRICK PHARMACEUTICALS SCHERING LABORATORIES STIEFEL LABORATORIES, !NC. SQUIBB INSTITUTE FOR MEDICAL RESEARCH SYNTEX LABORATORIES TEXAS PHARMACAL CoMPANY UPJOHN COMPANY The Society also wishes to acknowledge Westwood Pharmaceuticals for its support of the Resident­ Fellow memberships.