ASPECTS OF PEPTIDE CONFORMATION

Abstract

The paper consists of a critical review of the potential functions and other data that have been proposed for use in the theory of peptide conformation. It is pointedout that, although they all agree broadly in regard to the low-energy regions of the(¢, si)-map for an alanyl dipeptide, all of them disagree with the experimentally determined probability distribution (Ref 9), in which the region between si = 60^0 and= 120° is relatively unpopulated. From this, and other quantum chemical calculations,as also the observed distribution of .si-values in amino acids and amides, a newform is suggested for V (si), namely V (si) = 2.0 (1 - cos 2si) . Further, the need to have softer potentials for H ....H and H . . . X interactions is pointed out. These changes lead to much better agreement with the observed conformational parameters for the LL-bend. So also, observed characteristic ratios of random coils of polypeptides having a CP-atom attached to Ca agree better with the theoretical value if such modifications are made in the potential functions. The author feels that the potential functions used at present should be carefully reviewed and revised by comparison with authenticated observational data.

Comparison between calculated minimum energy locations and orientations,and those observed, of benzene and N-methyl acetamide in their crystal structuresindicate that the functions proposed from different laboratories lead to widely different values for the stabilization energy, although they all agree as to the location and orientation of the molecule in the case of benzene. Although there is reasonable agreement between theory and experiment for N-methyl acetamide, it is not considered to be good enough, and the potential functions have to be evaluated more critically before they are applied to more complicated examples.

The non-planarity of the peptide unit is considered both from quantum chemical
calculations and its observed distributions. Both indicate that changes in (.0 from planarity of upto 100, and of ON(describing the non-planarity of the threebonds meeting at the nitrogen atom) upto25°, are quite feasible and lead only to destabilizing energies less than RT. Similar resultsare also obtained for formarnide, where INDO calculations agree well with microwave data on non-planarity. The nun-planar peptide unit requires careful consideration in relation to the solution of crystal structures of globular proteins.