The molecular cloning of cDNAs for the cytoskeletal protein vinculin.

The complete primary amino acid sequence of chick vinculin has been determined from two overlapping cDNAs, cVin5 and a 2.89 kb cDNA, isolated from a chick embryo fibroblast lambda gtll cDNA library. In the region of overlap, the nucleotide sequences of the two cDNAs were identical apart from nine single base differences and the absence, in cVin5, of 123 bp corresponding to amino acid residues 167-207 encoded by the 2.89 kb cDNA. N-terminal vinculin polypeptides, encoding the region of difference between the two cDNAs, were generated using an in vitro transcription/translation approach. To this end, residues 167-207 were shown to be important in defining the talin-binding activity of vinculin. Site-directed mutagenesis has further defined this activity to residues 167-201. The origin of the heterogeneity between the two cDNAs has been investigated by analysis of the genomic organisation of the vinculin gene. Analysis of the domain structure of vinculin showed that the molecule could be cleaved by V8 proteinase into two polypeptides of 90 kDa and 32kDa. The 32 kDa fragment could be further cleaved to a 27 kDa fragment. The 90 kDa polypeptide corresponded to the globular head domain and contained the N- terminus of vinculin, the talin-binding site and three 112 amino acid residue repeats. The 32 kDa polypeptide corresponded to the extended tail region of vinculin and was located at the C- terminus. The head and tail domains were apparently separated by a proline-rich region which contained the V8 proteinase cleavage sites and a candidate tyrosine phosphorylation site.