Search

1. Procedures for partial purification of C1 and C1⁻ are described and methods for the isolation of subcomponents C1r and C1s from C1 and C1⁻ preparations have been developed. 2.1. Structural studies of C1r derived from C1 and C1⁻ have shown that C1r, like C1s, exists in serum as a proenzyme, which is activated by limited proteolysis. 2.2. C1r and C1s are both single polypeptide-chain zymogens of 83,000 molecular weight with similar amino acid compositions. They are distinct in carbohydrate composition and antigenic properties. Activation of C1r and C1s occurs without loss of glycopeptides but loss of small peptides may occur. 2.3. C1r⁻ and C1s⁻ bind one mole of DFP per mole of protein. The amino acid sequences of the first 20 residues of the 26,000 molecular weight diisopropylphosphate-binding polypeptide chains of C1r⁻ and C1s⁻ show a high degree of sequence identity with corresponding regions of serine proteases such as trypsin and plasmin. C1r⁻ and C1s⁻ are therefore identified as serine proteaces of similar size and structure to plasmin. The structure and mode of activation of C1r and C1s and other serine protease zymogens are compared. 3.1. The enzymic activities of C1r⁻ and C1s⁻ have been investigated. C1r⁻ cleaves C1s but did not hydrolyse any synthetic substrate tested. C1s⁻ cleaves C2 and C4 and hydrolyses several amino acid esters. The proteolytic specificities of C1r⁻ and C1s⁻ are discussed. 3.2. Rapid and convenient assays for C1r⁻ and C1s⁻, suitable for kinetic studies of zymogen activation, have been developed. 4.1. The interactions of C1q, in solution or bound to antibody-antigen complexes, with C1r and C1s have been examined. 4.2. In solution, C1r and C1s form strong Ca2+-dependent complexes containing 2C1r + 2C1s molecules, or 4C1r + 4C1s molecules. C1r, C1s and the C1r-C1s complexes interact weakly with C1q. 4.3. When C1q is bound to immune complexes, C1r and C1r-C1s complexes bind strongly to C1q. C1s does not bind to C1q in the absence of C1r 4.4. Activation of C1r and C1s does not appear to alter their binding affinity for other subcomponents either in solution or when bound to antibody-antigen. 5.1. Spontaneous activation of C1r and C1s in solution has been studied. C1s does not activate autocatalytically. Absolute stability of C1r preparations was not observed, but it is concluded that auto-activation of C1r does not contribute significantly to the instability of C1r. 5.2. When fixed to antibody-antigen-C1q complexes, C1r does not activate unless C1s is also present. The effect of C1s on C1r activation does not require C1s⁻ enzymic activity and it is apparent that C1s fulfills a binding requirement. 5.3. The mechanism of activation of C1 on interaction with immune complexes is discussed.

The work reported in this thesis is mainly biochemical research on the properties of the human immune system plasma protein, complement factor H (FH). A major function of FH is to bind to “foreign” or “altered-host” surfaces, mainly by recognising charge cluster motifs. When bound to a surface, it down regulates activation of the complement system on that surface. Its binding properties towards a range of other proteins and macromolecules, have been examined, mainly by ELISA-style assays. Research then focused on a smaller number of these FH ligands, which appear to bind FH very strongly, and not, as is the usual situation, by charge interactions. These ligands are Adrenomedullin, Trinitrophenyl-derivatised ligands, and dinitrophenol-derivatised ligands (TNP and DNP). The binding and dissociation characteristics of these ligands have been examined, the binding optimised, and it has been shown that TNP and DNP derivatised ligands can be used for affinity purification of FH from human plasma. A factor H homologue in plasma, C4bp, also binds these ligands, but a number of other FH homologues in plasma do not (eg beta2 glycoprotein1). Expression systems have been obtained from other labs to make recombinant segments of FH and recombinant protein expressed in order to narrow down the binding sites on FH for these ligands. Binding sites in 3 regions of FH have been located, and the effects of ligand binding on the complement-regulatory functions of FH have been assessed.

The complement system in human blood represents a major component of innate immunity. The role of the complement system is to recognise foreign materials coming into contact with the blood, including microorganisms, synthetic particles, or damaged and altered self-components, such as apoptotic and necrotic cells. Complement can be activated via three main pathways: the classical, alternative and lectin pathways. The classical pathway activation is achieved through the binding of the protein C1q to the targets. Factor H is well-known as an inhibitor of the alternative pathway, but as it can bind to many of the same ligands as C1q, it might compete with C1q and, therefore, be involved in the classical pathway control. Different target molecules which activate the classical pathway show variable binding of both of these complement proteins. This thesis explored the binding of C1q and FH to a range of target ligands using the ELISA technique. Previous research has shown that FH can compete directly with C1q binding and inhibit classical pathway activation. Manipulating the relative quantities of C1q and FH in human serum has been shown to influence the extent of classical pathway activation. This role of FH is distinct from its role as a regulator for the alternative pathway. This study measured the FH:C1q molar ratios in human plasmas for the first time, and the results showed a wide range of ratios (1.25:1 to 84:1) as well as widely varying concentrations of C1q and Factor H between individuals. This variation in the molar ratio appeared not only between individuals, but also in single individuals in a longitudinal study. Thus, FH could play an important role in controlling inflammation and have significant involvement in inflammatory diseases.

The complement system is an important immune system mechanism involved in both the recognition and elimination of invading pathogens. It is activated by three different pathways: The classical pathway, which relies on binding of C1, and results in the cleavage of C4 and C2 through activation of C1r and C1s; the alternative pathway that relies on the spontaneous hydrolysis of C3 and the lectin pathway. The lectin pathway is activated by binding of Mannan-binding lectin (MBL) or the ficolins (L-ficolin, H-ficolin and M-ficolin) to microbial binding motifs, and the subsequent activation of the MBL-associated serine proteases (MASP) 1/ 2/ 3. Of these MASP2 has been identified as the enzyme responsible for the activation of complement by C4 and C2 cleavage. The work presented here will focus on four different aspects of the lectin pathway: specificity and stoichiometry of the L-ficolin protein complex, expression of H-ficolin, substrate characterization for MASP1 and investigation of the prothrombin activation potential of MASP2. L-ficolin binding specificity was investigated using glycan array technology, and it was found that L-ficolin, instead of recognizing single monosaccharides like MBL, instead binds to extended oligosaccharide structures. The binding to these was dependent not only on the presence of acetyl groups, but also on their orientation in space. It was also found that L-ficolin in serum is found as a multimeric protein complex composed of 18 polypeptide chains and associated with one MASP dimer. The expression of H-ficolin resulted in the generation of a stable mammalian cell line producing oligomerized and biologically functional H-ficolin. MASP1 substrate specificity was investigated by two different procedures. Firstly fractionated plasma was subjected to MASP1 treatment in an attempt to identify a plasma protein substrate. This did not yield any substrate candidates, since only cleavage of the protease inhibitor α-2-macroglobulin could be detected. Additionally the thrombin-like activity of MASP1 was investigated through cleavage experiments done with factor XIII and fibrinogen. These experiments showed that the factor XIII cleavage site for MASP1 and thrombin is identical. This was also found for the fibrinogen β-chain but not for the α-chain showing that MASP1 interaction with fibrinogen is distinct from that of thrombin. An earlier observation that MASP2 was capable of activating prothrombin and generating thrombin was further characterized. Here it was shown that the activation of prothrombin by MASP2 is identical to that by factor Xa, which is the enzyme undertaking this role in the coagulation system, and that the activation can result in deposition of fibrin on the surface upon which MASP2 is bound. The prothrombin activation potential of MASP2 was also utilized to develop a new MASP2 activity assay, which was shown to be capable of measuring MASP2 activity, when MASP2 is bound, via MBL (or L-ficolin) to appropriate surfaces.