Recent Research

We aim to understand nuclear pathways that synthesise and process newly transcribed RNAs, the assembly of RNA-protein complexes and the surveillance activities that monitor their fidelity. Our work applies newly developed techniques (see Figure), with a substantial focus on following the kinetics of changes during time courses, predicting that the most significant interactions will be more evident than at steady-state.

Transcription and the nascent transcript

The CRAC technique maps protein-binding sites on RNA, combining UV-crosslinkling with tandem, denaturing purification, making the technique very stringent, relative to other approaches. High resolution mapping of RNAPI had not previously been reported and showed startlingly uneven distribution. Unexpectedly, we discovered that folding of the nascent pre-rRNA close to the transcribing polymerase has a major, previously unanticipated, effect on the elongation rate. Extending this to yeast RNAPII, we found that unstructured RNA, which favours slowed elongation, is associated with faster cotranscriptional splicing and proximal splice site usage, indicating regulatory significance for transcript folding (Turowski et al., 2020).

We are now focussing on the mechanisms that terminate transcription. For such a key step in gene expression, it seems remarkable that it remains so poorly understood in eukaryotes. We hypothesise that termination is linked to adenylation of the nascent transcript by the TRAMP complex, following polymerase backtracking. We are testing this model – initially in vitro.

RNA quality control – deeply embedded in gene expression

An outstanding question in nuclear RNA surveillance was how vast numbers of defective and spurious RNAs are identified and targeted for degradation. We are characterising surveillance factors, have recently clarified the functional distinctions between different TRAMP isoforms (Delan-Forino et al., 2020; Delan-Forino & Tollervey, 2020). Our analyses led to a new paradigm for RNA surveillance, in which degradation is a default state, from which stable RNAs must escape. This explains how rapidly evolving ncRNAs can be efficiently targeted for degradation by highly conserved proteins. Thermodynamics indicate that spurious transcription is inevitable. Evolution of promiscuous surveillance systems was likely a prerequisite for genome expansion in eukaryotes (Bresson and Tollervey 2018).

RNA-based responses to stress

All cells and organisms face challenges imposed by diverse stresses, with RNA metabolism playing key roles in rapid adaptation. We identified nuclear RNA surveillance as a novel, active step in regulating gene expression (Bresson et al., 2018). To better understand stress responses, we developed total RNA-associated protein purification (TRAPP) to quantify all RNA-protein complexes (Shchepachev et al., 2019). Quantifying proteins strongly altered in RNA-binding immediately following stress, dramatically underlined the key importance of translation regulation. In particular, we discovered how translation shut-down occurs in yeast. Significantly, this is distinct from previously studied pathways (Bresson et al., 2020).

More generally, we currently lack understanding of mRNP architecture and how it changes during the lifetime of a transcript. We will develop techniques to address this major weakness in our understanding of gene expression.

RNA-linked disease

Defects in several ncRNAs and RNA processing factors that we initially characterised, were subsequently found cause human diseases. We selected two of theses for analysis using our newly developed techniques. With the onset of the COVID-19, we included hypothesis-driven research aimed at uncovering SARS-CoV-2/host interactions.

Cartilage Hair Dysplasia (CHH), dwarfism and impaired immune responses are linked to point mutations in the RNA component of the RNA-protein enzyme, RNase MRP. We originally identified the pre-rRNA target for RNase MRP and the first associated protein. Recapitulating the major disease-linked mutation in human cells induced defects in pre-rRNA processing, ribosome accumulation and MRP structure. This established Cartilage Hair Dysplasia as a ribosomopahy, and the first human disorder of rRNA processing to be described (Robertson et al., 2021).

Prader-Willi syndrome is due to loss of genes encoding the imprinted, brain-specific snoRNAs snoRD115 and snoRD116. We constructed pre-neuronal cell lines specifically lacking the expressed genes for each of these snoRNAs. During neuronal differentiation, the cell lines with snoRNA deletions show specific defects in morphology and gene expression. Loss of the neuronal snoRNAs leads to substantial changes, positive and negative, in the abundances of multiple mRNAs and lncRNAs, features not seen on depletion of any other ncRNA. We strongly predict that we will discover novel regulatory mechanisms underlying these changes.

SARS-CoV-2

With the recent COVID-19 pandemic, we altered our research focus to include .  SARS-CoV-2 is a large positive sense, single stranded RNA virus that encodes four structural proteins and sixteen nonstructural proteins (nsp1-16) that carry enzymatic activities important for viral replication, mostly associated with RNA metabolism. Following infection, SARS-CoV-2 is predicted to massively reorganize host cell RNA metabolism, especially protein synthesis, and this is likely to be an important feature of viral replication. With the recent  Here is an outline of some of our work on SARS-CoV-2 . We hope to uncover novel vulnerabilities: Blocking host cell RNA factors that are hijacked by the virus could potentially be therapeutic, as might the protection of host factors that are inhibited by the virus.  

Figure: Techniques developed and applied.

A: Crosslinking and analysis of cDNAs (CRAC). Tagged proteins are tandem-affinity purified under denaturing conditions. Crosslinked RNA fragments are RT-PCR amplified and sequenced.

B: Crosslinking and sequencing of hybrids (CLASH). As CRAC, but RNA-RNA interactions are recovered by ligation of RNA fragment ends prior to sequencing.

C: Total RNA-associated peptide purification (TRAPP). RNA is used as ligand for affinity-purification of crosslinked proteins on a silica column. Following RNA digestion to nucleotide mono-phosphates (NMP), bound proteins are quantified by SILAC mass-spectrometry.

D: Mapping of individual aminoacid-RNA crosslinks (iTRAPP). The phosphate group on the residual NMP are used for peptide enrichment on a TiO₂ column. Sites mapped in mass-spectrometry data using Xi (https://github.com/Rappsilber-Laboratory/XiSearch)

E: Phosphoproteome mapping through purification of phosphorylated peptides on TiO₂ column, with or without prior fractionation for RNA-association. Sites mapped in mass-spectrometry data.