Prediction and Modeling of the Structure of 16S rRNA

Abstract

The ribosome is a complex macromolecule responsible for the translation of genetic information into proteins. Experimental data has provided clear evidence that the ribosome undergoes conformational changes during the process of translation. The size of the ribosome coupled with its dynamic nature has made it difficult to determine the detailed structure of the ribosome using traditional structural analysis techniques (e.g. NMR or crystallography). Recently, crystal structures of the ribosome of several organisms have been determined. The crystal structures are consistent with ~90% of the data obtained experimentally. The areas of the crystal structure of the ribosome which are inconsistent with experimental data are concentrated primarily in the areas of the ribosome known to be actively involved in the process of translation. The computational techniques utilized in this project provide an alternative method for generating structures that are consistent with all the experimental data.

In the first phase of the project, MC-SYM, a constraint satisfaction algorithm, was used to generate detailed three-dimensional models for two regions in the RNA portion of the small subunit of the ribosome known to be actively involved in translation: 921- 930/1387-1396 and 1399-1410/1490-1504. The models generated are consistent with the predicted secondary structure, chemical reactivity data, mutagenesis data and crosslinking data.

The second phase of the project used statistical methods to identify the highest probability base triples in the small subunit of the ribosome. The most likely candidates for base triples in the ribosome were identified as base triples involving the single stranded nucleotide 121and the 124:237 or the 125:236 base pair. Isomorphic modeling using the algorithm ISOPAIR indicated that a base triple consisting of an interaction between 121 and the 124:237 was the most likely to produce an isomorphic structure for the most frequently occurring sequences for this region. MC-SYM was used then used to construct a model for the 122-127/234-239 base paired region that included a base triple interaction between the single stranded nucleotide 121 and the 124:237 base pair.