RNA Conformational Classes

David Micallef¹, Laura Murray², Helen M. Berman¹, Jane Richardson², Zdeněk Morávek³,
Bohdan Schneider⁴

¹Rutgers, The State University of New Jersey, Department of Chemistry and Chemical Biology, NJ-08854, USA. ²Duke University, Durham NC, USA. ³Faculty of Mathematics and Physics, Charles University, Ke Karlovu 2, Prague, Czech Republic. ⁴Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo n. 2, CZ-16602 Prague, Czech Republic, bohdan@rcsb.rutgers.edu

RNA conformations have been analyzed by different knowledge-based approaches. Duarte et al. [1] reduced nucleotide six dimensional (6-D) conformational space to two pseudo torsions around pseudo bonds P_i-C4’_i and C4’_i-P_i+1. This approach, while appealing by its simplicity, neglects correlations between nucleotide torsion angles and therefore fails to recognize NA conformational families. Hershkovitz et al. [2] have analyzed nucleotide torsional space. An automated pattern-recognition approach concentrated on binning of torsions with the highest variability, namely a, g, d, and z. The analysis of correlations within a nucleotide precludes analysis of correlations of torsions between nucleotides, especially the key correlation between the torsions at the phosphodiester link.

Murray et al. [3] have studied a ribose-to-ribose unit called “suite”, not a nucleotide. Ingenious analysis of two 3-D torsional spaces of “heminucleotides”, d—e—z and a—b—g, have lead to identification of 47 distinct conformations of suites. Schneider, Morávek and Berman [4] have studied relationships between nucleotide torsion angles with the main emphasis on the torsions at the phosphodiester link, z and a, and considering the other main descriptors of RNA conformations, namely d, describing ribose pucker, g, and c, the torsion describing base orientation around the glycosidic bond.

The approaches of Murray et al. [3] and Schneider, Morávek and Berman [4] are to some extent complementary and their combination brings a more detailed knowledge about the RNA conformational behavior. The talk presents across-the-database analysis of RNA structures combining these two approaches. The original data containing all dinucleotides (di-nt) from the NDB public archive were filtered by crystallographic criteria, resolution and temperature factor, and by checking their stereo chemical quality, mainly atom-atom close contacts as described by Murray et al. [3]; the resulting data matrix contains about four thousand di-nt fragments. The fragments were divided into two groups, the majority of “A-RNA like” with z ~ a ~ 300°, and the “non-A-like” rest. In each group, seventeen 3-D scatter grams (“maps”) were analyzed by the Fourier averaging technique developed previously [5]. The information content of the maps was estimated by Shannon’s entropy, S ~ - SP_ilnP_i, where P_i is a fraction of di-nt fragments which can be assigned to the i-th peak. The maps with the highest information content (the largest S) were then used to cluster the data points. Conformational clusters were independently determined using protocols described in both [3] and [4] and their geometries compared and validated. The final RNA conformational families were then determined as consensus geometries determined by [3] and [4].

BS is grateful to support by grant LN00A032 from the Ministry of Education of the Czech Republic.

[1] Duarte, C.M., Wadley, L.M. and Pyle, A.M. RNA structure comparison, motif search and discovery using a reduced representation of RNA conformational space. Nucl.Acids Res., 31 (2003) 4755-4761.

[2] Hershkovitz, E., Tannenbaum, E., Howerton, S.B., Sheth, A., Tannenbaum, A. and Williams, L.D. Automated identification of RNA conformational motifs: Theory and application to the HM LSU 23S rRNA. Nucl.Acids Res., 31 (2003) 6249-6257.

[3] Murray, L.J.W., Arendall III, W.B., Richardson, D.C. and Richardson, J.S. RNA backbone is rotameric. Proc.Natl.Acad.Sci USA, 100 (2003) 13904-13909.

[4] Schneider, B., Morávek, Z. and Berman, H.M. RNA conformational classes. Nucl. Acids Res. 32 (2004) 1666-1677.

[5] Schneider, B., Cohen, D.M., Schleifer, L., Srinivasan, R., Olson, W.K. and Berman, H.M. A systematic method to study the spatial distribution of water molecules around nucleic acid bases. Biophys.J., 65 (1993) 2291-2303.