Primary sequence and domain structure of chicken vinculin.

We have determined the complete sequence of chick vinculin from two overlapping cDNA clones. The vinculin mRNA consists of 262 bp of 5' untranslated sequence, an open reading frame of 3195 bp (excluding the initiation codon) and a long 3' untranslated sequence (greater than 2 kb). Chick vinculin contains 1066 amino acid residues, and has a deduced molecular mass of 116,933 Da. Analysis of the domain structure of vinculin shows that the molecule can be cleaved by V8 proteinase into a 90 kDa globular head and a 32 kDa tail region, the latter of which could further be cleaved into a 27 kDa polypeptide. The 90 kDa globular head contains the N-terminus of vinculin, three 112-residue repeats (residues 259-589), and extends to approximately residue 850. Gel overlay experiments show that it also contains a binding site for the cytoskeletal protein talin. The talin-binding domain was further localized to the N-terminal 398 amino acid residues of the protein by expression in vitro of this region from a vinculin cDNA cloned into the Bluescript SK+ vector. The head and tail domains are apparently separated by a proline-rich region that contains V8-proteinase-cleavage sites and a candidate tyrosine (822)-phosphorylation site. Secondary-structure prediction suggests that the head and tail domains contain alpha-helical regions separated by short stretches of turn/coil. Comparison of the chick with a partial human sequence reveals that vinculin is a highly conserved protein. In chickens Southern-blot analysis is consistent with a single vinculin gene, and it is therefore likely that vinculin, and its higher-molecular-mass isoform termed metavinculin, arise through alternative splicing.