Artificial ribonucleases: Quantitative analysis of the structure-activity relationship and a new insight into the strategy of design of highly efficient RNase mimetics

• Published in: Russian journal of bioorganic chemistry Vol. 34; no. 4; pp. 442 - 452
• Main Authors: Koroleva, L. S., Kuz’min, V. E., Muratov, E. N., Artemenko, A. G., Sil’nikov, V. N.
• Format: Journal Article
• Language: English
• Published: Dordrecht SP MAIK Nauka/Interperiodica 01-07-2008
• Subjects: Biochemistry; Biomedical and Life Sciences; Biomedicine; Bioorganic Chemistry; Human immunodeficiency virus; Life Sciences; Organic Chemistry; Review Articles; simplex descriptors; QSAR; artificial ribonucleases
• ISSN: 1068-1620; 1608-330X
• DOI: 10.1134/S1068162008040080

The dependence of hydrolytic activity of artificial ribonucleases toward an HIV-I RNA fragment, a 21-mer oligonucleotide, and tRNA
Asp
on the structure of the RNase mimetic was analyzed. The quantitative structure-activity relationship (QSAR task) was determined by the method of simplex representation of the molecular structure where the amounts of four-atom fragments (simplexes) of fixed structure, symmetry, and chirality served as descriptors. Not only the types of atoms participating in simplexes, but also their physicochemical properties (e.g., partial charges, lipophilicities, etc.) were taken into account. This allowed the determination of the relative role of various factors affecting the interaction of molecules under study with the corresponding biological target. The 2D QSAR models obtained by the method of projection to latent structures have quite satisfactory statistical indices (
R
2
= 0.82−0.96;
Q
2
= 0.73−0.89), which help predict the activities of new compounds. The electrostatic properties of ribonuclease atoms were shown to contribute significantly to the manifestation of the hydrolytic activity of ribonucleases in the case of the 21-mer oligonucleotide and tRNA. In addition, the structural fragments that most greatly contribute to the alteration of the hydrolytic activity of RNases were identified. The models obtained were used for the virtual screening and molecular design of new highly efficient RNase mimetics.