Hard α-Keratin Intermediate Filament Chains: Substructure of the N- and C-Terminal Domains and the Predicted Structure and Function of the C-Terminal Domains of Type I and Type II Chains

Abstract

The quantity of sequence data now available for both Type I and Type II hard α-keratin IF proteins makes it possible to analyze their N- and C-terminal domains and ascertain features of likely structural and/or functional importance. The N-terminal domains of both chain types can be divided into acidic (NA) and basic (NB) subdomains, where NA is 29 and 34 residues long, respectively, for Type I and II chains and is located immediately adjacent to the end of the rod domain. NB constitutes the remainder of the N-terminal domain and is about 27 and 70 residues long for the two chain types, respectively. The glycine residue contents, however, are high in NA(I) and NB(II), but low in NA(II) and NB(I). Subdomain NB(II) contains four consecutive nonapeptide quasirepeats of the form GGGFGYRSX. The C-terminal domain of Type I chains, termed C(I), is characterized by a PCX motif repeated 10 times, 7 of them contiguously. From an analysis of the conformation of like peptides from crystal structures it has been shown that this region will probably adopt a polyproline II left-handed helical structure with three residues per turn. In contrast, the C-terminal domain of Type II hard α-keratin chains (known as C(II)) contains a periodic distribution of hydrophobicities that, together with other predictive techniques, allow its conformation (a twisted four-stranded antiparallel β-sheet) to be predicted with some degree of confidence. In addition, it is possible to suggest two partners with which this domain will interact. The first is with segment L12 in the rod domain and the second is with another C(II) domain in an antiparallel neighboring molecule. The latter possibility appears most likely. In either case the aggregation would likely serve to stabilize the molecular assembly through the interaction of two β-sheets via their apolar faces and, in so doing, would position a number of cysteine residues in external positions that would allow them to form a number of covalent disulfide bonds with other molecules.