Bond distances in polypeptide backbones depend on the local conformation. (1st June 2015)

• Record Type: Journal Article
• Authors: Improta, Roberto; Vitagliano, Luigi; Esposito, Luciana
• Is Part Of: Acta crystallographica. Volume 71:Part 6(2015:Jun.)
• Page Start: 1272
• Page End: 1283
• Publication Date: 2015-06-01
• Subjects: Biomolecules -- Structure -- Periodicals; Physical biochemistry -- Periodicals; X-ray crystallography -- Periodicals; Crystallography -- Periodicals; 572
• DOI: 10.1107/S1399004715005507
• Languages: English
• ISSNs: 0907-4449

Abstract

<abstract abstract-type="main" xml:lang="en"> <title> <x xml:space="preserve">Abstract</x> </title> <p>By combining quantum‐mechanical analysis of small model peptides and statistical surveys of high‐resolution protein structures, a systematic conformational dependence of bond lengths in polypeptide backbones has been unveiled which involves both the peptide bond (C—O and C—N) and those bonds centred on the C<sup>α</sup> atom. All of these bond lengths indeed display a systematic variability in the ψ angle according to both calculations and surveys of protein structures. The overall agreement between the computed and the statistical data suggests that these trends are essentially driven by local effects. The dependence of C<sup>α</sup> distances on ψ is governed by interactions between the σ system of the C<sup>α</sup> moiety and the C—O π system of the peptide bond. Maximum and minimum values for each bond distance are found for conformations with the specific bond perpendicular and parallel to the adjacent CONH peptide plane, respectively. On the other hand, the variability of the C—O and C—N distances is related to the strength of the interactions between the lone pair of the N atom and the C—O π* system, which is modulated by the ψ angle. The C—O and C—N distances are related but their trends are not strictly connected to peptide‐bond planarity, although a correlation amongst all of these parameters is expected on the basis of the classical resonance model.</p> </abstract>