Conformation of cyclodipeptides. factor analysis: a convenient tool for simplifying conformational studies of condensed poly-ring systems: prolyltype cyclopeptides

Abstract

AFA analysis (AFA), also known as principal-component analysis, was used in conformational studies of the condensed 2,5-dioxopiperazine/pyrrolydine (DOP/PYR) 6/5-membered two-ring system, a compound typical of cyclodipeptides comprising prolyl-type amino acid residues. The study is based on the analysis of 30 various X-ray conformers of the molecular frame in question. The results were evaluated and discussed using ring-puckering theory (RPT) as a reference. Complete mutual correspondence between the AFA and the RPT results was found, when both rings are considered separately, which is the prerequisite to using RPT. This correspondence allows a clear-cut physical interpretation of the AFA results, which are otherwise abstract in nature. Thus, two or three independent puckering variables were found for the DOP and PYR rings, respectively, and they acquire the physical significance of the absolute amplitudes of the basic pucker modes, typical of the respective rings. The term absolute is used for puckering with respect to the planar conformer of reference. Subsequent AFA treatment of the condensed DOP/PYR system allowed identification of five conformational variables necessary and sufficient to describe the concerted two-ring puckering completely. Each of the respective basic pucker modes defines a unique pattern of conformational variation of the whole two-ring system. In contrast to the separated ring cases, the origin of the five-dimensional conformational space of DOP/PYR is placed at the mean conformer and it spans physically accessible conformational deviations around the mean. When a conformer other than the mean (e.g. planar) one is chosen initially as a reference, AFA automatically brings the system to the basis set of local deviations around the mean. The results obtained demonstrate that AFA may be a very powerful technique in conformational studies based on an evaluation of a large collection of mutually consistent data. Factor analysis is especially useful in analyses of condensed poly-ring systems, not amenable to the RPT treatment.