Revision of collagen molecular structure.

Based on the fiber diffraction data from native collagen, Rich and Crick proposed the 10/3-helical model with a 28.6 A axial repeat in 1955 (Rich A.; Crick, F. H. C. Nature (Lond) 1955, 176, 915-916). We obtained the 7/2-helical structure with a 20 A axial repeat from the single crystal analysis of (Pro-Pro-Gly)(10). Since the latter structure could explain fiber diffraction patterns from native collagen, we proposed this structure as a new model for collagen in 1977 (Okuyama et al., Polym J 1977, 9, 341-343). These two structural models were refined against observed continuous intensity data from native collagen using a linked-atom least-squares method. It was found that the diffraction data from native collagen could be explained by the 7/2-helical model better than, or at least the same as, the prevailing 10/3-helical model. Together with the evidence that recent single crystal analyses of many model peptides have supported the 7/2-helical model and there was no such active support for the 10/3-helical model, it was concluded that the average molecular structure of native collagen seems to be closer to the 7/2-helical symmetry than the other one. Copyright 2005 Wiley Periodicals, Inc.