Structure of collagen

Abstract

A triple helical structure for collagen with two systematic hydrogen bonds for every three residues was put forward from this laboratory in 1955. This structure was criticized by Rich and Crick on the basis of a few short interatomic contacts occurring in it; they suggested that only structures with one hydrogen bond for every three residues will be free of bad contacts. Recently, we have examined the whole question carefully, and it has been found that it is possible to build up a two-bonded structure (two hydrogen bonds for three residues) while retaining all contacts within permissible values. This has been possible mainly because a study of the observed interatomic contacts in various reported organic structures showed that contact distances actually occur which are much shorter than those mentioned by Rich and Crick, namely, C ... C = 3.6 - 4.0 Å and C ... O = 3.2 - 3.4 Å. In fact, if Rich and Crick's criteria were adopted, some of the accepted structures like the α-helix will become unacceptable. Secondly, the study of the relation between the NH ... O hydrogen bond distance and the infra-red frequency showed that the observed frequency (Ramachandran, G. N., Sasisekharan, V., and Thathachari, Y. T., unpublished work) in collagen, which is higher than that observed in other proteins, should correspond to a value of the order of 3.0 Å. The actual parameters of the minor helix of the collagen structure have also been re-determined. These gave a value of 2.95 Å. for the residue height and 3.28, rather than 3.33 (10/3), for the number of residues per turn. As a consequence, the twist for three residues in the triple helical structure should be 30° rather than 36°.