Structural biology of macromolecular complexes involved in RNA metabolism and transcriptional silencing
The expression of individual genes is controlled at the levels of mRNA transcription and also post-transcriptionally, by processes such as splicing, localization, modification or editing, and degradation. To gain a mechanistic understanding of these processes it is important to understand the interactions between the individual players, including both protein and nucleic acid components, at the molecular level. We have used structural approaches to tackle mechanistic questions about how protein-RNA interactions can control RNA maturation and RNA editing and how transcriptional repressors are recruited to methylated DNA. By combining structural studies with biochemical, biophysical and cell-based functional assays we can gain powerful insights into these molecular processes.
Previously, we focused on proteins that control RNA metabolism during eukaryotic ribosome biogenesis and RNA turnover. By solving the structure of yeast Tsr1, an essential ribosome biogenesis factor, we were able to model it into low-resolution maps of immature 40S ribosomal. This gave a key insight into how this protein controls the timing of events during maturation of the ribosomal small subunit.
We also solved several structures of an essential vertebrate protein complex that is thought to be an RNA chaperone on many types of transcript including rRNA. This complex, made up of nuclear factors 90 and 45 (NF90/NF45) specifically recognizes stretches of dsRNA. We have shown that NF90 has an evolutionary relationship to proteins involved in adenosine-to-inosine editing. In future work we hope to better understand how the full-length form of this protein recognizes dsRNA and how this may impact on RNA editing in cells.
More recently, in collaboration with Adrian Bird’s laboratory, we gained new insights into the molecular basis for a genetic autism spectrum disorder known as Rett syndrome (RTT) (Figure 1). Mutations to the methylated DNA binding protein MeCP2 cause RTT and fall in to two clusters, one in the DNA binding domain and one in a region that recruits the transcriptional co-repressor complex called NCoR/SMRT. We mapped the MeCP2 binding to the C-terminal WD40 domain TBLR1, a tetrameric core protein of NCoR/SMRT. The crystal structure of a complex between TBLR1 and a fragment of MeCP2 reveals that the only residues that make extensive contacts to TBLR1 are exactly those mutated in RTT. This suggests that a functional interaction between this region of MeCP2 and TBLR1 is required for the development of a healthy brain.
Kruusvee, V., Lyst M.J., Taylor, C., Tarnauskaite, Z. and Bird, A.P. (2017). Structure of the MeCP2-TBLR1 complex reveals a molecular basis for Rett syndrome and related disorders. Proc Natl Acad Sci USA 114, E3243-E3250
McCaughan, U.M., Jayachandran, U., Shchepachev, V., Chen, Z.A., Rappsilber, J., Tollervey, D., and Cook, A.G. (2016). Pre-40S ribosome biogenesis factor Tsr1 is an inactive structural mimic of translational GTPases. Nat Commun 7, 11789.
Jayachandran, U., Grey, H., and Cook, A.G. (2016). Nuclear factor 90 uses an ADAR2-like binding mode to recognize specific bases in dsRNA. Nucleic Acids Res 44, 1924-1936.