Abstract
The primary amino acid sequence of contactin, a neuronal cell surface glycoprotein of 130 kD that is isolated in association with components of the cytoskeleton (Ranscht, B., D. J. Moss and C. Thomas 1984. J. Cell Biol. 99:1803-1813) was deduced from the nucleotide sequence of cDNA clones and is reported here. The cDNA sequence contains an open reading frame for a 1.071-amino acid transmembraned protein with 962 extracellular and 89 cytoplasmic amino acids. In its extracellular portion the polypeptide features six type 1 and two type 2 repeats. The six amino-terminal type 1 repeats (I-VI) each consist of 81-99 amino acids and contain two cysteine residues that are in the right context to form globular domains as described for molecules with immunoglobulin structure. Withing the proposed globular region, contactin shares 31% identical amino acids with the neural cells adhesion molecule NCAM. The two type 3 repeats (I-II) are each composed of 100 amino acids and lack cysteine residues. They are 20-31% identical to fibronectin type III repeats. Both the structural similarity of contactin to molecules of the immunoglobulin supergene family, in particular the amino acid sequence resemblance to NCAM, and its relationship to fibronectin indicate that contactin could be involved in some aspect of cellular adhesion. This suggestion is further strengthed by its localization in neuropil containing axon fascicles and synapses.