Complement the opsonization of foreign material.a combination of other serine protease domain-containing molecules are gene duplication in combination with exon shuffling. 8rtooa8emtaladhtonodetiono6teealpe protein mosaics that are made up of different domains or the N-terminus d the pr wo non CIs/UEGF/BMPI (CUB)D characterizes the complement system in vertebrate spe cellular proteins playing roles indevelopmental processes) (Table 1). an epidermal growth factor(EGF)-like domain,two short From a functional point of view,complement proteins consensus repeat(SCR)domains nto three major ulat the ave of th e fo (2)those involved in the regulation of these activation ancestor.since they share very similar domain structures sequences,and (3)those that act as receptors for complement proteins Several e complement proteins Is ar ne proteas hin the compl tein) ded hy in haptoglobin 2 molecule also contains two CCP (compl ment control protein)modules.Therefore.an evolutionary happears toist between haptoglobin and the Components Involved in the Activation and are localized in tandem within the maior histocompat Sequences the SCR structur Complement components Clq and MBL (mannose ng lecun)st cular architect d lectin The CL similaritiesit is believed that factor Band C2 have arisen b molecule has a hexameric structure.and each of the six ule nas sea urcmi chin has ve nature and is thus by far the lobular head region.Modules with AmmcaoecaaoionoTheABsnCeh athway characterized in animal sp cies.It is not known in cule was are also present in the erminal regions of nonco e the C2 molecule.However mple ment protei human type V VIII and ein that s ns to play the ha olved in the binding of Cla to the Fc regionso IgM and IgG antibodies Components C3/C4/C5 tho Inc trast to most of the other complement co subunits is composed of three identical chains.Th C3.C4 and C5 contain neither repeating structures nor terminal region of each chain forms a triple-helical evident modules. the prmary sequences anc collagen-like struct tbchadthcCtcminalreioneonlains co nponer are ver to th ca te-rece mannan gr to the nam cestral molecule C3 C4 and cs are similar in size(200kDa),subunit structure,and order of their chains (a-B in C3 and C5 and a-B-y in C4);all three arginine ween chal th C3 Complement components with serine protease domains The thioesters located in the C3d and C4d region of C3 and C4.respectively,are responsible for the covalent due at p domain but this con group of regulatory proteins.Except for factor D.all the the presence of His 1126 is believed to be essential for the ENCYCLOPEDIA OF LIFE SCIENCES/2001N els.netthe opsonization of foreign material). A combination of gene duplication in combination with exon shuffling, including sequential addition or deletion of several types of modules or domains from various proteins, has produced the functional and structural complexity that characterizes the complement system in vertebrate species (Table 1). From a functional point of view, complement proteins can be grouped into three major categories: (1) those involved in the activation sequences of the three pathways, (2) those involved in the regulation of these activation sequences, and (3) those that act as receptors for complement proteins. Several of the complement proteins may belong to more than one of these three groups, since they have several functional roles within the complement system. Components Involved in the Activation Sequences Complement components C1q and MBL (mannose￾binding lectin) share a similar molecular architecture, and they play analogous roles in the activation of the classical and lectin pathways, respectively. The C1q molecule has a hexameric structure, and each of the six subunits is made of three different chains, A, B and C. The N-terminal portion of each chain has a collagen-like sequence with a triple-helical structure: the C-terminal portion contains a globular head region. Modules with similarity to C-terminal portion of the A, B and C chains are also present in the C-terminal regions of noncomple￾ment proteins, including human type VIII and X collagens and precerebellin. The globular head region of the C1q chains is involved in the binding of C1q to the Fc regions of IgM and IgG antibodies. MBL is a collectin with collagen-like stalks similar to those of C1q; like C1q, MBL also has a hexameric structure. However, in contrast to C1q, each of its six subunits is composed of three identical chains. The N￾terminal region of each chain forms a triple-helical collagen-like structure, and the C-terminal region contains a C-type lectin carbohydrate-recognition domain that is involved in the binding of the MBL molecule to the mannan groups on the surfaces of microorganisms. Complement components with serine protease domains The complement components that contain a serine protease domain are factor D, C1r, C1s, MASP1, MASP2, factor B and C2. Factor I also has a serine protease domain, but this complement component falls within the group of regulatory proteins. Except for factor D, all the other serine protease domain-containing molecules are protein mosaics that are made up of different domains or modules. Beginning from the N-terminus, C1r, C1s, MASP1 and MASP2 consist of two noncontiguous C1r/ C1s/UEGF/BMPI (CUB) modules (found also in extra￾cellular proteins playing roles in developmental processes), an epidermal growth factor (EGF)-like domain, two short consensus repeat (SCR) domains and a serine protease domain at the C-terminal part of the molecule. The four molecules appear to have descended from a common ancestor, since they share very similar domain structures and functional activities. It is interesting to note that MASP2, C1r and C1s are similar to the serine protease domain of haptoglobin (a haemoglobin-binding serum protein) in being encoded by a single exon; in addition, the haptoglobin 2 molecule also contains two CCP (comple￾ment control protein) modules. Therefore, an evolutionary relationship appears to exist between haptoglobin and the MASP2, C1r and C1s molecules. Factor B and C2 share the same genomic organization and are localized in tandem within the major histocompat￾ibility complex (MHC) class III region. They also share the same domain structure and are both composed of three SCR domains, a Von Willebrand domain and the serine protease domain (N- to C-terminus). Because of their similarities it is believed that factor B and C2 have arisen by gene duplication from a common ancestor. A factor B-like molecule has been cloned from the sea urchin (echino￾derm), suggesting that the alternative pathway is very primitive in nature and is thus by far the most ancient pathway characterized in animal species. It is not known in which animal species the ancestral factor B molecule was duplicated and diverged to give the C2 molecule. However, teleost fish (trout) and birds (chickens) are known to have a protein that seems to play the roles of both factor B and C2. Components C3/C4/C5 In contrast to most of the other complement components, C3, C4 and C5 contain neither repeating structures nor evident modules. However, the primary sequences and genomic organizations of the three components are very similar, leading to the belief that the three proteins arose from a common ancestral molecule. C3, C4 and C5 are similar in size (
200 kDa), subunit structure, and order of their chains (a-b in C3 and C5 and a-b-g in C4); all three also possess arginine linkers between chains, and both C3 and C4 contain an active thioester site in the a chain. The thioesters located in the C3d and C4d region of C3 and C4, respectively, are responsible for the covalent binding of these components to their acceptor molecules. The His residue at position 1126 of human C3, as well as the Gln located two amino acids upstream, is very important in determining the binding specificity of the thioester; indeed, the presence of His 1126 is believed to be essential for the Complement ENCYCLOPEDIA OF LIFE SCIENCES / & 2001 Nature Publishing Group / www.els.net 3