Complete cDNA sequence of human complement pro-C5. Evidence of truncated transcripts derived from a single copy gene.

Two truncated human C5 clones, pHC5A and pHC5B, were isolated from an adult human liver cDNA library, and contained inserts of 2930 and 2181 bp, respectively. Both clones were polyadenylated and encoded the 5'-end of the C5 pro-molecule, thereby completing the human pro-C5 cDNA sequence. However, near the 3'-ends, at exon/intron boundaries, the nucleotide sequences of pHC5A and pHC5B diverged from each other and from the full-length 6.0-kb C5 cDNA sequence. Clone pHC5A, which overlapped the first human C5 clone described (J-16), encoded most of the C5 signal peptide, the complete beta-chain, the linker peptide, 177 amino acids of the alpha-chain, and contained 144 bp of Alu family consensus sequence encoding 48 amino acids of divergent protein sequence in an open reading frame. Clone pHC5B encoded the entire C5 signal peptide, the beta-chain, the linker peptide, nine amino acids of the alpha-chain, and six amino acids of divergent protein sequence in an open reading frame. Northern blot experiments demonstrated the presence of a 3.0-kb truncated C5 mRNA in adult human liver and a 4.8-kb truncated C5 mRNA in HepG2 cells in addition to the 6.0-kb full-length transcript. Truncated C5 mRNA were not detected in Raji, MOLT-4, human fibroblast or U937 cells, although the full-length 6.0-kb transcript was seen in MOLT-4 cells. Southern blot analyses indicated that the human C5 structural gene is large, complex, and is present in the human genome in a single copy, thereby demonstrating that the truncated C5 clones and mRNA are derived from a single C5 gene by alternative processing events.