Jennifer A. Doudna
Structure of Self-Splicing Introns
The discovery that RNA can function as a biological catalyst, or ribozyme, has raised important questions about the chemistry and evolution of enzymes. The primary goal of my research during the next several years is to investigate ribozyme structure and function using X-ray crystallographic and biochemical techniques. Since few RNA structures are currently known at high resolution, this work will impact broadly on our understanding of ribozymes and RNA components of fundamental cellular machinery such as the ribosome and spliceosomes. Atomic resolution structures of RNA will also provide the basis for understanding principles of RNA-mediated chemistry and the evolution of RNA and protein catalysts.
Recent work in my laboratory includes the following: 1) we have solved the crystal structure of a 50 kilodalton domain of a self-splicing intron at 2.8 _ resolution, revealing many novel structural interactions that stabilize tertiary folding and enable RNA to interact with other molecules; 2) we have explored the folding of a second proposed domain of the Tetrahymena self-splicing intron, and are using our crystal structure as the basis for testing interactions within the catalytic core of the intron; 3) we have developed a way of rationally designing crystal contacts into RNA molecules of interest, greatly facilitating crystal growth; 4) we have obtained crystals of several other biologically interesting RNAs including the complete catalytic core of a Group I intron, domains of a Group II self-splicing intron and two other ribozymes. Ongoing research that extends and expands these results is described below.
Group I intron structure and chemistry I. Tertiary domains of the Tetrahymena Group I intron. The crystal structure of the 160-nucleotide P4-P6 domain of the Tetrahymena self-splicing intron, recently solved in my laboratory, is the largest, and one of the few, crystal structures solved to date for RNA. Thus, many fascinating and unexpected chemical interactions have been identified that explain biochemical observations and functional properties of the intron RNA. Several ongoing projects will probe structural interactions in the P4-P6 domain and its contacts to the other domain of the Tetrahymena intron
Metal ion and antibiotic binding sites in P4-P6: Divalent metals and antibiotic drugs that interact specifically with the P4-P6 domain will be soaked into native crystals, and their binding sites determined by difference Fourier analysis. Sites of interaction in each case will be probed by site-directed mutagenesis and activity assays using a domain reconstitution system developed during my postdoctoral work. This work will reveal the modes of ligand binding used by RNA, contributing to its structural stability and biological activity. Since both metals and antibiotics bind to a wide range of known RNAs and affect their function, these experiments are broadly relevant to our understanding of RNA structure and chemistry.
Structural interactions in the catalytic core of the Tetrahymena intron: The phylogenetically-conserved catalytic core of Group I introns has been modeled as two tertiary domains that juxtapose to form the active site. The P4-P6 domain contains an internal loop of adenosines predicted to bind to the substrate helix P1 in the intact intron. The crystal structure revealed two unexpected homopurine base pairs at the putative substrate binding site in P4-P6. In collaboration with Scott Strobel's laboratory we are now substituting various base analogs into this site and the recognition site in P1 to determine the exact chemical nature of the substrate binding interaction. Other regions of contact between the P4-P6 domain and the rest of the intron catalytic core will be determined by site-directed mutagenesis. These experiments will be greatly facilitated by the crystal structure since sites of potential contact in P4-P6 are readily identified. Information about the structural contacts in the intron RNA will be used to construct a detailed model of the intron catalytic core.
P4-P6 as a scaffold for folding of the intron catalytic core: Recent chemical probing experiments in my lab show that the other proposed domain of the intron, P3-P9, does not fold stably until it contacts P4-P6. We are using kinetic experiments, chemical probing and nuclease mapping to determine the structural changes that occur in P3-P9 upon binding to P4-P6. This project directly addresses requirements for RNA folding in the context of the Group I intron. II. Azoarcus Group I intron structure. A long-term goal is to solve a high resolution crystal structure of the complete catalytic core of a Group I intron. Crystals have been obtained for the Group I intron from Azoarcus, a small, thermally-stable ribozyme; they are currently being optimized for data collection.
Group II intron structure and mechanism The Group II class of self-splicing introns have a catalytic mechanism very similar to that of mRNA splicing, which is catalyzed by a large ribonucleoprotein complex called the spliceosome. In order to address questions of function and evolution of these ribozymes, we have obtained crystals of several conserved structural domains of a Group II intron. These crystals are currently being optimized for X-ray diffraction analysis, and we expect to soon begin solving these structures. Ultimately we hope to gain an understanding of the evolutionary relationships, if any, between the Group I and Group II classes of introns by comparing atomic resolution structures of these RNAs. Approaches to RNA crystallization and structure determination
A sparse matrix approach to RNA crystallization, developed during my postdoctoral work, has been improved based on many new results with RNA crystals. We recently published a technique for generating large amounts of very pure RNA of any sequence for biophysical studies. Ongoing work involves engineering specific molecular contacts into RNAs of interest to facilitate crystal packing, an approach that has been extremely successful in our hands so far. Finally, the crystal structure of P4-P6 revealed several specific binding sites for osmium hexammine, a heavy atom derivative used in the structure determination. Thus, it will likely be possible to engineer these sites into other RNAs for the purpose of using osmium hexammine as a derivative to solve the phase problem. These approaches will foster X-ray crystallographic analysis of RNA, a rapidly growing area of biological research.
Additional areas of ongoing research in my laboratory include mapping the solvent-accessible surface of a Group II intron, and using in vitro selection to identify specific nucleic acid ligands for anti-DNA autoantibodies. The intron mapping work complements our structural investigations, and is being used to identify long-range interactions in the intron RNA. These experiments explore the folding of possible tertiary domains within the intron as well, and will be extremely useful for determining appropriate constructs for structural study. The selection work will isolate specific ligands for autoantibodies that recognize DNA, which will then be used to probe the molecular recognition of the antibody combining site. Since the natural epitopes that elicit this abnormal immune response are not known, we hope to explore the biology of this process as well in collaboration with Dr. Mark Shlomchik's laboratory at Yale.