STRUCTURAL STUDIES OF THE RNA COMPONENT OF RNASE P
Xiaojing Yang , Valerie Grum and Alfonso Mondragón*, Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, Evanston, IL, *Corresponding author: a-mondragon@nwu.edu
Ribonuclease P, or RNase P, is the endonuclease responsible for processing the 5' end of tRNA by cleaving a precursor and leading to tRNA maturation. It is made of an RNA component and a protein component. RNase P has been identified in all organisms and cells that need to process tRNA. Although it has been found in all kingdoms (Archaea, Bacteria, and Eucarya), the best characterized RNase P come from two different bacteria, Escherichia coli and Bacillus subtilis. The bacterial RNase P RNA component (or P RNA) consists of between 350-400 nucleotides while the protein component is a small, basic protein of around 120 amino acids. Both molecules are needed for activity in vivo. In vitro the RNA can catalyze RNA cleavage in the absence of protein. Unlike other catalytic RNAs that perform one cleavage reaction, RNase P is a real enzyme as it turns over and can catalyze several cleavages.
Recognition of the tRNA by RNase P involves the acceptor stem and the common arm of the tRNA and the CCA sequence at the 3' end. Short sequences can be used as substrate as long as they have some common features: a short stem and the CCA sequence at the 3' end. The cleavage reaction is not fully characterized, but it is probably similar to the reactions catalyzed by other catalytic RNA molecules. It cleaves the phosphodiester bond producing a 5' phosphate at one end and a 3' hydroxyl at the other end. The involvement of magnesium in the cleavage and binding of precursor tRNA has been established and some of the regions involved in cleavage and recognition have been mapped.
Understanding of the catalytic mechanism and the recognition process requires structural information and considerable efforts have been devoted to obtaining a secondary and tertiary structure of P RNA. Until now, the most successful approach has been phylogenetic analysis. The primary sequence of many P RNAs is now known and although there is little sequence similarity among them, it is possible to align them by looking at covariant changes. In this way, it is possible to recognize a common fold for all P RNAs. The sequence data together with chemical footprinting data suggest some predictions of the tertiary structure. Three dimensional models for E. coli P RNA (called M1 RNA) have been proposed and are being constantly refined based on more data. Nevertheless, these models are not very accurate and a three dimensional structure of a P RNA has not yet been obtained by physical methods.
Similar to proteins, large RNA molecules can be formed by different domains that fold independently. The first molecule characterized in this manner was the catalytic core of the group I intron, which has two independent domains (P4-P6 and P3-P9). The two domains can be produced separately, and they recover activity when mixed. Furthermore, the identification of these domains led to the structure of the P4-P6 domain which, at 160 nucleotides, is the largest RNA structure known to date. The structural domains of both E. coli and B. subtilis P RNA have been investigated and characterized. Eubacterial P RNA consists of at least two domains, domain I and II, that can be produced separately. In B. subtilis the first domain comprises nucleotides 86 to 239 while the second domain comprises the rest of the molecule (and can be produced as a circularly permuted molecule). In E. coli domain I spans nucleotides 87 to 242 while the rest form domain II. The overall domain structure in both molecules is very similar despite the marked differences in secondary structure. The two domains are responsible for different activities. Domain I is known to interact with precursor tRNA, can bind tRNA with micromolar affinity, and probably contacts tRNA directly. Domain II can catalyze cleavage of in vitro selected substrates with close to wild type efficiency and contains most of the residues considered to form the active site. The characterization of these two domains is a major step forward toward obtaining a three dimensional structure, as it is now possible to focus attention not only on the intact molecule but also on smaller domains, as was done for the Tetrahymena group I intron.
Results:
The goal of this work is to elucidate the structures of a 154-nucleotide fragment of P RNA from B. subtilis and E. coli in order to understand in atomic detail the structure of this particular RNA and also to add important new structural information on RNA structure. In the past year we have conducted several data collection experiments at DND-CAT using the MAR CCD detector and radiation from the insertion device. We have collected data on several heavy atom derivative candidates as well as on native crystals. As a result of these experiments, we identified a new crystal form of the B. subtilis P RNA that diffracts to 4.0C. The crystals belong to space group C2221. Self rotation functions indicate the presence of a non-crystallographic two-fold axis, and this is consistent with the expected number of molecules in the asymmetric unit. We have now also identified two closely related heavy atom derivatives. The two derivatives produce interpretable Patterson functions and have a set of common sites and hence we are looking for additional heavy atom derivatives. We have now data from several native crystals in two crystal forms and derivatives in one of these forms. We plan to continue our heavy atom search in order to solve the structure by MIR or MAD approaches. A diffraction pattern from one of the RNA crystals is shown below.
|
|
Acknowledgments. We thank H. Feinberg, D. Li, and K. Perry for help with data collection and D. Keane, J. Quintana and other members of DND-CAT for all their help