The Clq subunit of the first component of complement binds to laminin : A mechanism for the deposition and retention of immune complexes in basement membrane ( connective tissue / glomerulonephritis / vascultis
J. Bohnsack,A. Tenner,Gordon W. LAURIEt,Hynda K. KLEINMANt,George R. MARTINt,E. Brown
Abstract:The Clq subunit of complement component C1 is known to bind to immune complexes, which often are deposited in basement membrane. We investigated the possibility that this deposition is a result of binding to laminin, a large basement membrane glycoprotein. Clq showed soturable binding to immobilized laminin; this binding was increased at reduced ionic strength. Intact C1 did not bind laminin. A ternary complex was formed by laminin, Clq, and aggregated IgG. This complex formation was dependent on and proportional to the amount of Clq bound to the aggregated IgG. Binding of laminin to Clq occurred with a Kd of 2 nM and was stronger than the binding of Clq to fibronectin. Preliminary data, including electron micrographs of rotary-shadowed preparations, suggest that laminin binds to the collagen-like tail of Clq. Electron microscopy localized the site of interaction with Clq to a short arm of laminin. Since laminin is found only in basement membranes, the interaction between laminin and Clq could be involved in the deposition and retention of immune complexes in these structures. Basement membrane is a thin matrix that lines epithelia and endothelia in addition to surrounding muscle cells and the whole nervous system. Basement membranes are frequent sites of immune-complex deposition, which could be due to an affinity of immune complexes for some as yet unidentified basement membrane component. Laminin, a major and specific component of basement membrane, is a large glycoprotein (Mr 106). It was originally isolated from the basement membrane of a mouse tumor (1) but has since been found by immunocytological techniques in all basement membranes investigated (2). Purified laminin, which binds to basement membrane collagen (type IV) and basement membrane heparan sulfate (3), can mediate the attachment of a variety of cell types to type IV (basement membrane) collagen substrates (4). Fibronectin is another connective tissue glycoprotein that is also associated with basement membranes, though in lesser amounts. Unlike laminin, fibronectin is also present in fibrous connective tissues (2). Fibronectin is also found in high concentrations in the circulation (_300 itg/ml), whereas there is little or no circulating laminin (<1 pg/ml). Fibronectin binds to diverse substances including collagens I-IV, glycosaminoglycans, and receptors on certain cell surfaces (5). The binding of immune complexes to a basement membrane component may occur via Clq, since Clq binds to immune complexes and its structure includes a collagenous tail (6). Clq exists in plasma as a subunit of the complex macromolecule C1, the first component of complement. C1 consists of Clq and the tetrameric complex Clr2s2 of the proenzymes Clr and Cls. In the presence of Ca2', Clr and Cls bind to the collagenous region to form hemolytically active C1 (7). Binding of C1 to the Fc portion of IgG or IgM in an antigen-antibody complex, which occurs via the globular domains of Clq (8), initiates the classical pathway of complement activation through conversion of Clr and Cls to their active-enzyme forms, CiT and C1h. After activation, Cl? and C13 are displaced from Clq through the action of Cl-inhibitor (9), leaving Clq bound to the immune complex with its collagen-like tail exposed for interaction with other molecules. Fibronectin has been shown to bind directly to Clq (10, 11) and to Clq-bearing immune complexes in vitro (12). In this report, we demonstrate a high-affinity interaction between Clq and laminin. Further, Clq can mediate the binding of aggregated IgG, a model immune complex, to laminin. Addition of excess purified ClF2competitively inhibits this interaction of laminin with Clq-bearing immune complexes, suggesting that laminin interacts with the collagenlike region of Clq. MATERIALS AND METHODS Protein Purification and Iodination. Laminin (13), Clq (14), human IgG (15), and fibronectin (16) were purified as described. IgG was aggregated by heat treatment at 630C for 30 min and the large aggregates (.1.8 x 106 daltons) (aggIgG) obtained from the void volume of a Bio-Gel A-5m column were stored in 0.145 M NaCl/5 mM sodium phoaphate, pH 7.2 (P1/NaCl) at 40C until used. Clr2s was obtained by depleting human C1 of Clq (17); this preparation contained about 430 jAg of C1i2and 7 ,ug ofClq per ml, as determined by hemolytic titration. Bovine serum albumin (BSA) and human serum albumin (HSA) were purchased from Sigma. Clq was radiolabeled with Na'1I (specific activity of labeled Clq, 5.4 x 105 cpm/4g) as described (14). Laminin, fibronectin, and aggIgG were iodinated with Na1251 to specific activities of 7.7 x 105, 5.4 x 105, and 3.2 x 106 cpm/pg, respectively, by using Iodobeads (Pierce). Preparation of Protein-Coated Surfaces. Wells of Immulon II Remov-a-well strips (Dynatech, Alexandria, VA) were coated overnight at 40C with protein in 0.1 M NaHCO3, pH 9.5 (coating buffer; 100 A.l per well). Lamininor BSA-coated plates were prepared using the protein at 50 pug/ml in coating buffer; for aggIgG-coated wells, the protein concentration was 90 jug/ml. Immediately before use, the wells were washed three times with P1/NaCl containing 0.05% Tween 20 (Fisher) (P1/NaCl/Tween). For experiments done at low ionic strength, the wells were washed with 0.1 M sodiAbbreviations: C1, first component of complement, consisting of subunits Clq, Cir, and Cis; aggIgG, aggregated IgGBSA, bovine serum albumin; HSA, human serum albumin; 125I.., 'lI-labeled. 3824 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Proc. NatL Acad Sci USA 82 (1985) 3825 um acetate, pH 7.4, containing 0.05% Tween 20 (NaOAc/Tween). In both cases, the wells were incubated in wash buffer containing 1% BSA for 30 min at room temperature to quench their remaining nonspecific binding sites. Binding Assays. Radiolabeled aggIgG, Clq, or laminin, diluted in either physiologic (P,/NaCl; conductance = 8.5 mS) or low ionic strength buffer (NaOAc; conductance = 3.7 mS), were incubated in Immulon II wells coated with various proteins for 2 hr at room temperature in the presence of 1% BSA. To quantitate association, wells were washed extensively after the incubation, and bound radioactivity in individual wells was determined in a gamma counter. In some experiments, unlabeled Clq was substituted for radiolabeled Clq, and the amount of Clq bound after 2 hr was determined by ELISA. This ELISA assay is capable of detecting from 10 to 150 ng of human Clq per ml (15). Each binding curve was determined in duplicate in all experiments, and data are presented as the mean of duplicates (which always agreed within 10%) of representative experiments. Further details of individual experiments are given in the text. Electron Microscopy. Rotary-shadowed replicas of aggregates of Clq and laminin were examined in the electron microscope according to the method of Shotton et al. (18). Briefly, equal volumes of Clq and laminin at 30 ug/ml in P,/NaCl were mixed together, dialyzed overnight at 40C vs. 0.155 M ammonium acetate at pH 7.4, sprayed in 50%o glycerol onto freshly cleaved mica discs, and rotary-shadowed with platinum at 2 mPa in an Edwards 306A coater. The replica was examined in a JEOL 100C electron microscope.