Purification and characterization of subcomponent C1q of the first component of mouse complement.

1. Mouse C1q, a subcomponent of the first component of complement, has been purified in a highly haemolytically active form by a combination of precipitation with EGTA, ion-exchange chromatography and gel filtration. Yields ranged from 3 to 5 mg/200 ml of serum, and the activity of final preparations was in the range of 2 X 10(13)-4 X 10(13) C1q effective molecules/mg. 2. The molecular weight of mouse C1q was 439 500 +/- 1586, as determined by polyacrylamide-gel electrophoresis in the presence of sodium dodecyl sulphate. 3. Mouse C1q was shown to be composed of non-covalently linked subunits, all being in the molecular-weight range 45 000-46 000, and three covalently linked chains each having a molecular weight of approx. 23 000 as determined on polyacrylamide-gel electrophoresis in the presence of sodium dodecyl sulphate by using non-covalently and covalently linked subunits of human C1q as markers with known molecular weights calculated theoretically previously [Porter & Reid (1978) Nature (London) 275, 699-704]. 4. Mouse C1q contained hydroxyproline, hydroxylysine, a high percentage of glycine and approx. 9% carbohydrate. The absorption coefficient and nitrogen content of C1q were also determined.     

The purification and characterization of subcomponent C1s of the first component of bovine complement.

Bovine C1s, a subcomponent of the first component of complement, was purified in good yield by a combination of euglobulin precipitation and ion-exchange and molecular-sieve chromatography. Approx. 10 mg can be obtained from 3 litres of serum, representing a yield of 11%. The C1s is obtained in zymogen form, with a mol.wt. of 85000-88000, determined by gel filtration and SDS/polyacrylamide-gel electrophoresis. It is haemolytically active when tested with human C1q and C1r. Activation can be achieved by incubation with human C1r, resulting in cleavage of the C1s chain into two chains of 65000 and 27000 mol.wt. and the generation of an isoleucine N-terminal residue on the smaller chain. Active C1s binds an equimolar amount of di-isopropyl phosphorfluoridate to the smaller chain, which is the C-terminal part in the zymogen. The chains can be separated by ion-exchange in 8 M-urea. All of these characteristics show that bovine C1s is very similar to its human counterpart.     

Subunit composition and structure of subcomponent C1q of the first component of human complement.

1. Unreduced human subcomponent C1q was shown by electrophoresis on polyacrylamide gels run in the presence of sodium dodecyl sulphate to be composed of two types of non-covalently linked subunits of apparent mol.wts. 69 000 and 54 000. The ratio of the two subunits was markedly affected by the ionic strength of the applied sample. At a low ionic strength of applied sample, which gave the optimum value for the 54 000-apparent mol.wt. subunit, a ratio of 1.99:1.00 was obtained for the ratio of the 69 000-apparent mol.wt. subunit to the 5400-apparent-mol.wt. subunit. The amount of the 54 000-apparent-mol.wt. subunit detected in the expected position on the gel was found to be inversely proportional to increases in the ionic strength of the applled sample. 2. Human subcomponent C1q on reduction and alkylation, or oxidation, yields equimolar amounts of three chains designated A, B and C [Reid et al. (1972) Biochem. J. 130, 749-763]. The results obtained by Yonemasu & Stroud [(1972) Immunochemistry 9, 545-554], which showed that the 69 000-apparent-mol.wt. subunit was a disulphide-linked dimer of the A and B chains and that the 54 000-apparent-mol.wt. subunit was a disulphide-linked dimer of the C chain, were confirmed. 3. Gel filtration on Sephadex G-200 in 6.0M-guanidinium chloride showed that both types of unreduced subunit were eluted together as a single symmetrical peak of apparent mol.wt. 49 000-50 000 when globular proteins were used as markers. The molecular weights of the oxidized or reduced A, B and C chains have been shown previously to be very similar all being in the range 23 000-24 000 [Reid et al. (1972) Biochem. J. 130, 749-763; Reid (1974) Biochem. J. 141, 189-203]. 4. It is proposed that subcomponent C1q (mol.wt. 410000) is composed of nine non-covalently linked subunits, i.e. six A-B dimers and three C-C dimers. 5. A structure for subcomponent C1q is proposed and is based on the assumption that the collagen-like regions of 78 residues in each of the A, B and C chains are combined to form a triple-helical structure of the same type as is found in collagens.     

Isolation and characterization of C1q, a subcomponent of the first component of complement, from human and rabbit sera

1. C1q, a subcomponent of the first component of complement, has been isolated, in a haemolytically active and soluble form, by ion-exchange chromatography and gel filtration, from human and rabbit sera. Yields ranged from 10 to 25mg/litre of serum and the activity of final preparations was consistently in the range 5x10(3)-15x10(3) C1qH(50) units/mg. 2. The molecular weights of human and rabbit subcomponent C1q were 409600 and 417600, as determined by sedimentation equilibrium studies. 3. Subcomponent C1q from both species was shown to be composed of non-covalently linked subunits of approximately 57000 molecular weight as determined by gel-filtration or sedimentation equilibrium studies in 5.3m-guanidinium chloride. Reduction or oxidation of human and rabbit subcomponent C1q yielded three chains each having a molecular weight of approximately 23000 and which differed slightly in amino acid composition but markedly in carbohydrate content. The oxidized chains were separated, on a preparative scale, by ion-exchange chromatography in 8m-urea on DEAE-cellulose. 4. Both human and rabbit subcomponent C1q contained hydroxyproline, hydroxylysine, a high percentage of glycine and approximately 8% carbohydrate. Glutamic acid and aspartic acid were the free N-terminal amino acids of human subcomponent C1q whereas only serine was found in rabbit subcomponent C1q. 5. Collagenase digestion of human or rabbit subcomponent C1q caused a rapid loss of haemolytic activity which correlated with the breakdown of collagenous regions in the molecule.     

Autocatalytic activation of C1r subcomponent of the first component of human complement

Autoactivation of the proenzyme form of a subunit of the first component (C1r) was performed in the presence and absence of diisopropyl fluorophosphate (DFP). The time-course of autoactivation of zymogen C1r followed a sigmoidal curve and was accelerated by addition of the enzyme C1r and by increasing the concentration of C1r, suggesting that autoactivation of C1r consists of two intermolecular reactions, i.e. zymogen(C1r)- and enzyme(C1r)-catalyzed reactions. In the presence of 10 mM DFP, the enzyme-catalyzed autoactivation of C1r was completely inhibited, while the zymogen-catalyzed autoactivation still proceeded depending upon C1r concentration. These results suggested that the zymogen-catalyzed autoactivation of C1r is a DFP-insensitive second-order reaction and is mediated by an active site generated in a single chain C1r through a conformational change (Kassahara et al. (1982) FEBS lett. 141, 128-131). Based on these results, a possible reaction process of autoactivation of C1r was proposed, as follows: (formula; see text) where C1r represents a conformational isomer which catalyzes the autoactivation of C1r, and the rate constants, k2 and k3, are of second-order. Utilizing a computer, we simulated the autoactivation of C1r and found the above scheme to be a reasonable model of C1r autoactivation. Evidence which supports the formation of a conformational isomer of C1r, C1r, as an intermediate in its autoactivation was also obtained by a surface radiolabeling method.     

Studies on the structure and activity of rabbit C1q (a subcomponent of the first component of complement)

1. The subunit structure of rabbit subcomponent C1q was examined in a previous publication (Reid et al., 1972). The present paper describes some aspects of the structure of the polypeptide chains derived from the molecule. 2. The three polypeptide chains, produced by performic oxidation, of rabbit subcomponent C1q were isolated by ion-exchange chromatography in 8m-urea on DEAE-cellulose. 3. Each chain was found to contain 15-18% glycine and significant amounts of the amino acids hydroxyproline and hydroxylysine. 4. By means of collagenase digestion it was shown that all three chains of rabbit subcomponent C1q contain collagen-like sequences of amino acids which constitute about 40% of each chain. 5. By use of carboxypeptidase A it was established, indirectly, that the collagen-like sequences, in one of the chains, are probably located near, or at, the N-terminal end of the chain. 6. Collagenase digestion and heating at 52 degrees C (but not at 49 degrees C) caused rapid loss of native rabbit subcomponent C1q haemolytic activity.     

Monoclonal anti-mouse macrophage antibodies recognize the globular portions of C1q, a subcomponent of the first component of complement

One of seven monoclonal antibodies generated against mouse macrophages (M phi) was found to recognize isolated heterologous C1q. This antibody was shown to be cytotoxic and to react in a strain-independent way with mouse M phi derived from bone marrow cells as well as with M phi from the peritoneal cavity; it did not react, however, with mouse granulocytes, thymocytes, or T and B lymphocytes. The hemolytic activity of fluid phase C1q was inhibited to 50% at a 2 X 10(-4) dilution of hybridoma supernatant, whereas a 100-fold higher concentration was required to inhibit C1q bound to immune complexes ( EAC1q ) to the same extent. It was demonstrated that this antibody recognizes the isolated globular, Fc-binding portions of the C1q molecule and reacts with the A and B chains. Because M phi have been shown to synthesize C1q, the Fc-recognizing subcomponent of the first component of complement, evidence was provided that endogeneous C1q can serve as an Fc receptor on M phi during secretion. This fact was demonstrated by a dose-dependent inhibition of Fc-receptor activity for EIgG by the F(ab')2 fragment of this monoclonal antibody. These experiments further support the concept that C1q produced by M phi functions on the surface as an Fc-recognizing molecule before it is released and incorporated into the macromolecular complex of serum C1.     

Activation of the first component of human complement, C1, by monoclonal antibodies directed against different domains of subcomponent C1q

Two monoclonal antibodies directed against C1q, and their (Fab)2 and Fab fragments, were used to study the mechanism of C1 activation. Monoclonal antibody 2A10, an IgG2a, was digested by pepsin to yield fully immunoreactive (Fab')2. Monoclonal antibody 1H11, an IgG1, was digested by papain to yield fully immunoreactive, bivalent (Fab)2. Previously 1H11 had been shown to bind to the C1q "heads," whereas 2A10 bound to stalks. Activation of C1 was followed by the cleavage of 125I-C1s in the presence of C1 inhibitor (C1-Inh) at 37 degrees C. Spontaneous activation was minimal at inhibitor concentrations above 0.4 micron (1.3 X physiologic inhibitor concentration); all results were corrected for the spontaneous activation background. Heat-aggregated IgG activated completely in this system and was taken as 100% activation. Monoclonal antibody 2A10 caused precipitation of C1 and slow activation; neither the (Fab')2 nor the Fab' derived from 2A10-caused activation. Probably, aggregates of intact 2A10 and C1 were serving as immune complexes to activate other molecules of C1. In contrast, both 1H11 and its (Fab)2 activated completely and stoichiometrically; that is, maximal activation was achieved at a ratio of one C1q head to one antibody combining site. The monovalent Fab derived from 1H11 bound well to C1q, but no activation of C1 was observed. Thus, bivalent binding of this head-binding monoclonal is required for C1 activation, but not the presence of the antibody Fc portion. Neither 1H11 nor its (Fab)2 fragments caused C1 precipitation; however, the 1H11 did form complexes composed of two C1q cross-linked by multiple 1H11, which were visualized by electron microscopy. The presence of these dimeric complexes correlated well with activation. A model for C1 activation is proposed in which two C1q subcomponents are held together by multiple (Fab)2 bridging C1q heads. The model is roughly analogous to touching opposing pairs of fingers and thumb tips, the two hands representing the two C1q, forming a cage. C1-Inh, which probably binds to C1r through the open end of the C1 cone, is too long asymmetric to be included within the cage. Thus, according to this model, the dimers of C1 are released from the inhibitory action of C1-Inh, and activation proceeds spontaneously and rapidly at 37 degrees C.     

Isolation of the globular region of the subcomponent q of the C1 component of complement.

Digestion after heat treatment of the subcomponent q of the C1 component of complement by collagenase leads to the isolation of the globular region of the protein. This product ('heads') is composed of three chains giving an overall molecular weight of about 57000. About half of the collagen-like region present in C1 q is lost after digestion. The 'heads' are shown to be soluble and hemolytically active products.     

Chemical and hydrodynamic characterization of the human leucocyte receptor for complement subcomponent C1q.

A procedure for preparation of the receptor for complement subcomponent Clq from human tonsil lymphocytes and the monocytic cell line U937 was developed. The procedure is suitable for isolation of several hundred micrograms of the receptor, Clq-R, and has yielded sufficient material for chemical and hydrodynamic characterization. Clq-R from tonsil lymphocytes behaves identically with that from U937 cells. Clq-R has a monomer Mr of 56,000, and is an acidic glycoprotein containing about 17% carbohydrate. The polypeptide chain length is estimated to be 416-448 amino acid residues, with two or three sites for N-linked glycosylation. Detergent-solubilized Clq-R exists as an elongated dimer (f/fo = 1.8), and does not bind a significant weight of detergent. The radioiodinated isolated receptor binds specifically and saturably to solid-phase Clq, but not to collagen, IgG, bovine serum albumin or complement component C3.     

Complete amino acid sequences of the three collagen-like regions present in subcomponent C1q of the first component of human complement.

Possible interactions between polymerized (F-) actin and insulin-storage granules from rat islets of Langerhans were examined in vitro by comparing the sedimentation of the granules in the presence of various actin concentrations. Actin in the concentration range 0.1--0.5 mg/ml produced a retardation in granule-sedimentation rates consistent with binding of the granules to the actin filaments. The interaction was increased by addition of ATP (2mM), but was decreased by CaCl2 (0.1 mM). Binding of granules to actin was unaffected by cyclic AMP or by preincubation of the granules with phospholipase C. Specificity of the interaction was confirmed by the use of depolymerized (G-) actin and of myosin to provide a solution of comparable viscosity; neither of these caused any alteration of granule sedimentation. Possible implications of this interaction of insulin-storage granules with actin for the mechanism of insulin secretion are briefly discussed.     

Interaction between complement subcomponent C1q and bacterial lipopolysaccharides.

The heptose-less mutant of Escherichia coli, D31m4, bound complement subcomponent C1q and its collagen-like fragments (C1qCLF) with Ka values of 1.4 x 10(8) and 2.0 x 10(8) M-1 respectively. This binding was suppressed by chemical modification of C1q and C1qCLF using diethyl pyrocarbonate (DEPC). To investigate the role of lipopolysaccharides (LPS) in this binding, biosynthetically labelled [14C]LPS were purified from E. coli D31m4 and incorporated into liposomes prepared from phosphatidylcholine (PC) and phosphatidylethanolamine (PE) [PC/PE/LPS, 2:2:1, by wt.]. Binding of C1q or its collagen-like fragments to the liposomes was estimated via a flotation test. These liposomes bound C1q and C1qCLF with Ka values of 8.0 x 10(7) and 2.0 x 10(7) M-1; this binding was totally inhibited after chemical modification of C1q and C1qCLF by DEPC. Liposomes containing LPS purified from the wild-strain E. coli K-12 S also bound C1q and C1qCLF, whereas direct binding of C1q or C1qCLF to the bacteria was negligible. Diamines at concentrations which dissociate C1 into C1q and (C1r, C1s)2, strongly inhibited the interaction of C1q or C1qCLF with LPS. Removal of 3-deoxy-D-manno-octulosonic acid (2-keto-3-deoxyoctonic acid; KDO) from E. coli D31m4 LPS decreases the binding of C1qCLF to the bacteria by 65%. When this purified and modified LPS was incorporated into liposomes, the C1qCLF binding was completely abolished. These results show: (i) the essential role of the collagen-like moiety and probably its histidine residues in the interaction between C1q and the mutant D31m4; (ii) the contribution of LPS, particularly the anionic charges of KDO, to this interaction.     

Effect of the C1q complement subcomponent on blood coagulability and fibrinolysis

The in vitro experiments on human plasma have shown that C1q addition in a concentration of 120 micrograms/ml led to a substantial shortening of coagulation time of test-plasma, as well as kaolin- and cephalin time. The effect is preserved in plasma deficient in factors V, X and VII. It is assumed that C1q has properties similar to those of thromboplastin.     

Proteolytic cleavage of an activated subcomponent of the first component of rabbit complement, C1s

When rabbit C1 purified by affinity chromatography on IgG-Sepharose 6B was chromatographed on DEAE-cellulose in the presence of ethylenediaminetetraacetate, C1s was isolated as two forms, C1s(I) and C1s(II), having different molecular weights. On the other hand, incubation of the C1 with soybean trypsin inhibitor before the chromatography resulted in the isolation of C1s(I) alone, indicating that, during the purification, C1s(II) was derived from C1s(I) by proteolytic cleavage of C1s(I) by a contaminating protease, probably plasmin [EC 3.4.21.7]. In fact, C1s(I) was completely converted to C1s(II) or a C1s(II)-like fragment by highly purified plasmin. Analysis of the polypeptide chain structures revealed that C1s(I), which consisted of H and L chains with molecular weights of 70,000 and 36,000, respectively, was converted to C1s(II) by cleavage of the H chain, since C1s(II) consisted of two chains each with a molecular weight of 37,000. This conversion proceeded without any alteration in C1 esterase activity, but was accompanied by loss of the ability to form C1r-C1s complex.     

Completion of the amino acid sequences of the A and B chains of subcomponent C1q of the first component of human complement.

The sequences of amino acid residues 109--224 of the A chain, and residues 109--22 of the B chain, of human subcomponent C1q are given. These results, along with previously published sequence data on the N-terminal, collagen-like, regions of the A and B chains [Reid (1979) Biochem. J. 179, 367--371] yield the complete amino acid sequences of the A and B chains of subcomponent C1q. The asparagine residue at position A-124 has been identified as the major site of asparagine-linked carbohydrate in subcomponent C1q. When the sequences of the C-terminal, 135-residue-long, 'globular' regions of A and B chains are compared they show 40% homology. The degree of homology over certain stretches of 15--20 residues, within the C-terminal regions, rises up to values of 73%, indicating the presence of strongly conserved structures. Structure prediction studies indicate that both the A and B chain C-terminal regions may adopt a predominantly beta-type structure with apparently little alpha-helical structure.     

Enzymatic alteration of C1q, the collagen-like subcomponent of the first component of complement, leads to cross-reactivity with type II collagen

Native serum C1q, the collagenous-like subcomponent of the first component of complement, is not recognized by polyclonal anti-collagen type II antibodies. However, when purified C1q was subjected to limited proteolysis by collagenase it showed antigenic cross-reactivity with collagen type II. The same cross-reactivity was observed with hemolytically active C1q in synovial fluids of patients with rheumatoid arthritis (RA), whereas C1q from synovial fluids of patients with osteoarthritis (OA), villo-nodular synovitis and ankylosing spondylitis was not recognized by this antibody. However, incubation of synovial fluid C1q of OA patients with synovial fluid leucocytes from RA patients led to an alteration of OA-C1q which was now recognized by the anti-collagen type II antibody.     

Antigenic interspecies cross-reaction of subcomponent C1q of the first component of complement: evidence for its cross-reaction in both collagen-like and non-collagen-like regions

Human, bovine, and mouse C1q, a subcomponent of the first complement component, were purified, and both globular (GF) and collagen-like fragments (CLF) were isolated from human and bovine C1q. Antisera were produced in rabbits with these C1q or fragments, and F(ab')2 of immunoglobulin G (IgG) was purified from the antisera in order to avoid the possible non-specific binding of C1q of these animals to the Fc portion of rabbit IgG. Immunodiffusion analyses and radioimmune inhibition tests with these F(ab')2 showed that the definitive antigenic cross-reactivity was among C1q molecules of these animals, and that the regions participating in interspecies cross-reactions were located in both GF and CLF of C1q. These results suggest that both the C-terminal non-collagenous globular and the N-terminal collagen-like domains of C1q molecules may have remained highly conserved during evolution.     

Inhibition of the reconstitution of the haemolytic activity of the first component of human complement by a pepsin-derived fragment of subcomponent C1q.

1. A fragment of subcomponent C1q, which contained all the collagen-like features present in the intact molecule, was isolated by pepsin digestion as described by Reid [Biochem. J. (1976) 155, 5-17]. 2. The pepsin-derived fragment of subcomponent C1q did not bind to antibody-coated erythrocytes under conditions where complete binding of sub-component C1q took place. 3. The peptic fragment blocked the reconstitution of C1 haemolytic activity by competing with intact subcomponent C1q in the utilization of a mixture of the other two subcomponents, C1r and C1s. 4. Reduction and alkylation of the interchain disulphide bonds in the pepsin fragment did not markedly affect its inhibitory effect, whereas heating at 56 degrees C for 30min completely abolished the effect. 5. Lathyritic rat skin collagen and CNBr-derived peptides of pig type II collagen showed no ability to mimic the inhibitory effect of the pepsin fragment when tested over the same concentration range as used for the peptic fragment. 6. The peptic fragment was unable to block efficiently the reconstitution of C1 haemolytic activity unless it was added to the mixture of subcomponents C1r and C1s before the attempt to reconstitute C1 haemolytic activity, in solution, or on the surface of antibody-coated erythrocytes. 7. Evidence was obtained that suggested that subcomponent C1q bound the subcomponent C1r-C1s complex more efficiently when the subcomponent C1q was bound to antibody than when it was free in solution.     

A collagen-like amino acid sequence in a polypeptide chain of human C1q (a subcomponent of the first component of complement)

1. A partial amino acid sequence of 95 residues of the 191 residues in the oxidized A chain of human subcomponent C1q was determined. The partial nature of the sequence is because one overlapping peptide is missing in the proposed sequence, also the proof of some of the overlapping peptides depends partly on their amino acid composition and not on their complete sequence. 2. This region of the A chain contained a repeating sequence of glycine-X-Y (where X is often proline and Y is often hydroxyproline) for 78 residues. 3. The five hydroxylysine residues and the five hydroxyproline residues present in the oxidized A chain were all in these 78 residues and only in the Y position of the repeating sequence. 4. Prolonged collagenase digestion of the oxidized A chain yielded a large, apparently C-terminal, peptide which contained most of the non-collagenous sequences present in the chain. 5. It is concluded that there is a collagen-like region in the A chain of subcomponent C1q which constitutes most of the N-terminal half of the chain and that similar collagen-like regions will be found in the B and C chains.