Nucleotide sequence of cDNA covering the complete coding part of the human vimentin gene.

Seven clones containing all or parts of the coding sequence of the human vimentin gene (1398 bp) were isolated from a Xgtl 1 human cDNA library constructed by reverse transcription of mRNA purified from SV40 transformed MRC-5 fibroblasts (MRC-5 V2) (1). Together, the clones span a cDNA sequence of 1765 bp. Previous attempts to isolate cDNA for human vimentin have resulted in partial lengths below 1100 bp (2, 3). A monoclonal antibody, 98C1, recognizing vimentin among other proteins, was used for preliminary screening of 2X10 plaques. Two positive clones were obtained. One clone contained a 1.4 kb insert. Both strands were sequenced with the dideoxy chain termination technique (4). This clone was used to re-screen 2 X10 plaques giving approximately 80 positive clones. Five clones were isolated and end-sequenced. Two of these latter clones extended the preliminary sequence in both directions. Both strands of the extensions were sequenced. The overlapping sequences were identical except for 44 bp on one of the clones considered to be a cloning artefact. The complete cDNA sequence (1765 bp) is shown in Fig. 1. The sequence contained an open reading frame coding for a 466 amino acid protein found to be almost identical with the published sequence for human vimentin based on a combination of partial cDNA sequences and genomic DNA sequences (2, 3). Our sequence, however, shows some mismatches in the coding sequence with the previously published partial sequences. We have thus found Ser-42 and Leu-442 instead of Asp and Phe respectively (2), and Asn-201, Leu-265, Ser-278, Ser-339 and Asn-350 instead of Ser, Ser, lie, Cys and Lys respectively (3). The 3' untranslated region contains 322 bp from the stop codon to the poly A stretch. In agreement with previous findings (3) we found two putative polyadenylation signals, the first is located 40 bp downstream from the stop codon, the second is located 301 bp from the stop codon.