Search

The complete amino acid sequence (673 residues plus 15 residues of leader sequence) of human complement component C1s has been determined by nucleotide sequencing of cDNA clones from a human liver library probed with synthetic oligonucleotides. Much of the sequence is supported by independent amino acid sequence information. The cDNA sequence contains an anomalous "intron-like" sequence, including a stop codon, that can be discounted because of the amino acid sequence evidence. The N-terminal chain (422 residues) of C1s, like that of C1r with which it is broadly homologous, contains five domains: domains I and III are homologous to one another and to similar regions in C1r, domain II is homologous to the epidermal growth factor sequence found in C1r and several other proteins, and domains IV and V are homologous to one another and to the 60-residue repeating sequence found in C1r, C2, factor B, C4-binding protein and some apparently unrelated proteins. The sequence of the C-terminal chain (251 residues) agrees with that already established to be the "serine protease" domain of C1s.

Overlapping molecular clones encoding the complement subcomponent Cls were isolated from a human liver cDNA library. The nucleotide sequence reconstructed from these clones spans about 85% of the length of the liver Cls messenger RNAs, which occur in three distinct size classes around 3 kilobases in length. Comparisons with the sequence of Clr, the other enzymatic subcomponent of Cl, reveal 40% amino acid identity and conservation of all the cysteine residues. Beside the serine protease domain, the following sequence motifs, previously described in Clr, were also found in Cls: (a) two repeats of the type found in the Ba fragment of complement factor B and in several other complement but also noncomplement proteins, (b) a cysteine-rich segment homologous to the repeats of epidermal growth factor precursor, and (c) a duplicated segment found only in Clr and Cls. Differences in each of these structural motifs provide significant clues for the interpretation of the functional divergence of these interacting serine protease zymogens. Hybridizations of Clr and Cls probes to restriction endonuclease fragments of genomic DNA demonstrate close physical linkage of the corresponding genes. The implications of this finding are discussed with respect to the evolution of Clr and Cls after their origin by tandem gene duplication and to the previously observed combined hereditary deficiencies of Clr and Cls.

Normal human C1q is a serum glycoprotein of 460 kDa containing 18 polypeptide chains (6A, 6B, 6C) each 226 amino acids long and each containing an N-terminal collagen-like domain and a C-terminal globular domain. Two unusual forms of C1q have been described: a genetically defective form, which has a molecular mass of approximately 160 kDa and is found in the sera of homozygotes for the defect who show a marked susceptibility to immune complex related disease; a fibroblast form, shown to be synthesized and secreted, in vitro, with a molecular mass of about 800 kDa and with chains approximately 16 kDa greater than those of normal C1q. A higher than normal molecular mass form of C1q has also been described in human colostrum and a form of C1q has been claimed to represent one of the types of Fc receptor on guinea-pig macrophages. To initiate studies, at the genomic level, on these various forms of C1q, and to investigate the possible relation between the C1q genes and the procollagen genes, the complementary DNA corresponding to the B chain of normal C1q has been cloned and characterized.

We have isolated cDNA clones covering the complete B chain of the complement subunit Clq from mouse; this subunit initiates the classical complement pathway. Deoxynucleotide sequence analysis shows that these clones contain 156 nucleotides of the 5' untranslated region, followed by sequences coding for the 25 amino acids of the signal peptide; all of the 228 amino acids of the mature protein; and 140 nucleotides of the 3' untranslated region, including a poly A additional signal. The coding region for the mature protein contains 261 nucleotides for the Gly-X-Y repeat and 408 nucleotides for the globular portion of Clq. By comparing the nucleotide sequence of the mouse B chain of Clq with the human B chain (Reid, K.B.M., 1985, Biochem. J. 231, 729), we find a high homology (80%) within the mature protein, a lower homology within the signal peptide (59%) and the 3' untranslated region (47%) and no homology (26%) in the 5' untranslated region.

C1r is a zymogen of a serine protease that is involved in the activation of the first component of the classical pathway of the complement system. cDNAs coding for human C1r have been isolated from libraries prepared from poly(A) RNA from human liver and Hep G2 cells. From DNA sequence analysis, the overlapping cDNA inserts were shown to span 2493 nucleotides of the C1r mRNA, not including the poly(A) tail. The cDNA sequence coding for C1r contained a 5' noncoding region, 2115 nucleotides coding for a polypeptide precursor of 705 amino acids, and a 3' noncoding region. Some variability in the length of the 3' noncoding sequence was observed with the cDNA inserts, although most contained a polyadenylation signal followed by a poly(A) tail. The A or noncatalytic chain of C-1r, which originates from the amino-terminal end of the precursor molecule, contains a potential growth factor domain and two different pairs of internal repeats. One pair of these internal repeats is closely related to the amino-terminal sequence of C1s, while the other pair of repeats is homologous to the tandem repeats present in beta 2-glycoprotein I, complement factor B, the b subunit of factor XIII, and a single region present in the alpha 1 chain of haptoglobin. The B chain of C-1r contains the catalytic portion of the enzyme and is homologous to the trypsin family of serine proteases.

Our study was undertaken to determine if native DNA (dsDNA), which is known to bind to collagen, also binds to the collagen-like segment of Clq (CLS). Three methods were employed to demonstrate the binding of dsDNA to CLS: (1) Six human sera and a mouse monoclonal antibody to dsDNA were employed in an enzyme linked immunosorbent assay to detect CLS bound dsDNA. When applied to CLS bound dsDNA, sera with antibodies to dsDNA and monoclonal antibody to DNA yielded mean OD values of greater than or equal to 0.6, significantly higher than those values obtained from control experiments (mean OD less than or equal to 0.2, p less than or equal to 0.003). (2) In a solid phase immunoassay radiolabelled DNA (3H DNA) was exposed to increasing amounts of solid phase adherent CLS. The binding of 3H DNA to the solid phase was substantially greater (more than a 30-fold increase) when the solid phase had been precoated with CLS. (3) dsDNA binding to CLS was demonstrated further by incubating electrophoretically resolved CLS with dsDNA and localizing the bound dsDNA by ethidium bromide staining. Our results indicate the dsDNA binds to CLS. Since CLS is the binding site for Clr and then Clr2Cls2, molecules which bind to CLS, such as dsDNA, could be important factors affecting Cl activation.

Plasmid clones containing cDNA coding for the B-chain of human Clq were isolated from a liver cDNA library. The longest cDNA insert isolated contained all the coding sequence for amino acid residues B1 to B226 plus a 3' non-translated region of 264 nucleotides that extended into the poly(A) tail, thus accounting for 950 nucleotides of the mRNA. The B-chain mRNA was estimated by Northern-blot analysis to be 1.46 kb (kilobases) long, which indicated that approx. 500 bases were not accounted for in the cDNA clone. A cosmid clone containing the C1q-B chain gene was isolated from a human genomic DNA library. The precise 5' limit of gene was not established, but from the data available it appears that the gene is approx. 2.6 kb long. The coding sequence for residues B1 to B226 in the gene is interrupted by one intron, of 1.1 kb, which is located within the codon coding for glycine at position B36. This glycine residue is located in the middle of the triple-helical regions found in C1q at exactly the position where there is an unusual structural feature, i.e. a bend in each of the helical regions brought about by the interruption of the Gly-Xaa-Yaa repeating triplet sequences in the A- and C-chains and the presence of an 'extra' triplet in the B-chain. Nucleotide sequencing of the 5' end of the gene indicates the presence of a predominantly hydrophobic stretch of 29 amino acids, immediately before residue B1, which could serve as a signal peptide.

We have isolated cDNA clones covering the complete B chain of the complement subunit C1q from mouse; this subunit initiates the classical complement pathway. Deoxynucleotide sequence analysis shows that these clones contain 156 nucleotides of the 5' untranslated region, followed by sequences coding for the 25 amino acids of the signal peptide, all of the 228 amino acids of the mature protein and 140 nucleotides of the 3' untranslated region, including a poly A addition signal. The coding region for the mature protein contains 261 nucleotides for the Gly-X-Y repeat and 408 nucleotides for the globular portion of C1q. By comparing the nucleotide sequence of the mouse B chain of C1q with the B chain from human (Reid, K. B. M., 1985, Biochem. J. 231, 729), we find a high homology (80%) within the mature protein, a lower homology within the signal peptide (59%) and the 3' untranslated region (47%) and no homology (26%) in the 5' untranslated region.