RNase Sa: A LONG STORY OF IMPROVING RESOLUTION

Jozef Sevcik

Institute of Molecular Biology, Slovak Academy of Sciences, Dúbravská cesta 21, 84251 Bratislava, Slovak Republic, e-mail: umbisevc@savba.savba.sk

Keywords: ribonuclease, atomic resolution, accuracy

The structure of a ribonuclease from Streptomyces aureofaciens (RNase Sa) was originally determined by multiple isomorphous replacement at 1.8 Å using synchrotron data collected on films and refined to R = 17 %. It was clear even at that time that the crystals diffracted to about 1.2Å resolution. Subsequently the structure was refined against data collected to 1.2 Å at room temperature (R = 10.5 %) and than to 1.0 at 110 K with cryogenic techniques (R = 11.9 %). The estimated r.m.s. error in the coordinates of the atomic resolution structures is 0.05 Å, less than half that obtained at 1.8 Å resolution. For well ordered part of the main chain the error is lower than 0.02 Å . The accuracy of the structures allowed a detailed analysis of peptide planarity and torsion angles and suggests a possibility of a slight revision of previously established criteria. These features, together with localisation of errors in the structure, flexibility of the main chain observed at atomic resolution will be presented to demonstrate the benefits of more accurately determined structures. The conclusion and main message from this study is that one should not waste the gifts offered by the diffracting power of the crystals and indeed should collect data, whenever possible, to the resolution limits given by the crystals. The extra effort and time spent for data collection and refinement is surely worth having and provides a more reliable structure which may be crucial in disentangling the biological function and may avoid misinterpretations of results.