RNA molecules are uniquely capable of encoding and controlling the expression of genetic information, often as a consequence of their three-dimensional structures. We are interested in understanding RNA-mediated initiation of protein synthesis, and RNA-protein complexes involved in targeting proteins for export out of cells. We are also investigating the early steps in gene regulation by RNA interference.
Internal Ribosome Entry Site RNAs
Most eukaryotic and viral messages initiate translation by a mechanism involving recognition of a 7-methylguanosine cap at the 5' end of the mRNA. In a few cases, however, translation occurs via a cap-independent mechanism in which an internal ribosome entry site (IRES) in the 5'-untranslated region of the mRNA recruits the ribosome. In hepatitis C virus (HCV), the ~400-nucleotide IRES folds into a magnesium-dependent structure in which loops thought to interact with the ribosome are exposed on the surface of the RNA. Point mutations that destroy IRES activity disrupt the folded structure of the RNA. The IRES is formed from two independently folding structural domains. One of these, the "core," binds specifically to the 40S subunit of eukaryotic ribosomes, while the other domain interacts with initiation factor eIF3. Structures of the IRES-40S subunit complex, determined by cryo-electron microscopy in collaboration with Joachim Frank (HHMI, Health Research Inc., Wadsworth Center), revealed that the IRES induces a significant conformational change in the 40S subunit upon binding. This conformational change helps lock the start of the viral mRNA protein-coding sequence into the correct site on the 40S subunit.
In collaboration with Eva Nogales (HHMI, University of California, Berkeley), we determined structures of the HCV IRES in complex with the human translational machinery, showing how the IRES can functionally replace proteins that help position most cellular mRNAs on the ribosome. The use of affinity-purified samples and mass spectrometry has revealed the full composition and post-translational modification states of IRES-bound complexes that assemble in human cell extracts (in collaboration with Julie Leary, University of California, Davis). Working with Carol Robinson (Cambridge University), we are using mass spectrometry to analyze intact IRES-ribosome complexes to determine how the HCV IRES induces assembly of active human 80S ribosomes. (This work is supported in part by the National Institutes of Health.)
Structure and Mechanism of the Signal Recognition Particle
The signal recognition particle (SRP) is a highly conserved ribonucleoprotein responsible for transport of nascent polypeptides targeted for secretion or membrane insertion. In prokaryotes, the SRP consists of one protein (Ffh) and one RNA molecule (4.5S RNA), and both are required for SRP activity. The RNA sequence corresponding to the Ffh-binding site has been maintained through evolution and is virtually identical in organisms from the three kingdoms of life—bacteria, archaea, and eukaryotes. The RNA plays a key, yet undetermined, role in the protein-targeting pathway.
In 2000 we determined the crystal structure of the complex at 1.5-Å resolution, revealing a fascinating network of contacts at the RNA-protein interface that explain the observed evolutionary conservation. Using site-directed hydroxyl radical probing, we discovered that the association of the SRP with its receptor triggers a dramatic conformational change in the complex, localizing the SRP RNA and the adjacent signal peptide–binding site at the SRP-receptor heterodimer interface. The orientation of the RNA explains how peptide binding and GTP hydrolysis can be coupled through direct structural contact during cycles of SRP-directed protein translocation.
Recent experiments show that the position of the SRP RNA within the SRP-receptor complex enhances the rate of GTP hydrolysis in the complex above a critical threshold required in vivo. Work toward a crystal structure of the SRP-receptor complex (a ~130-kDa assembly) has been aided by the use of phage display technology to select and purify multiple antibody proteins.
RNA Recognition by Dicer Enzymes
Double-stranded RNA induces potent and specific gene silencing in a broad range of eukaryotic organisms. This mode of gene silencing, called RNA interference (RNAi), acts at the transcriptional level through formation of heterochromatin and at the post-transcriptional level through mRNA degradation and translational suppression. In all cases, RNAi begins with the processing of endogenous or introduced precursor RNA into micro-RNAs (miRNAs) and small interfering RNAs (siRNAs) 21–25 nucleotides in length by the enzyme Dicer. We recently solved the crystal structure of an intact Dicer enzyme, revealing how Dicer functions as a molecular ruler to measure and cleave duplex RNAs of a specific length. The structure has now been refined to higher resolution, and a series of mutant forms of Dicer have been used to delineate the roles of various domains and interactions, both in vitro and in vivo. Ongoing work focuses on determining how Dicer interacts with other components of the RNAi pathway and how diced RNAs are targeted to specific mRNAs. (This work is supported by the National Institutes of Health.)