Mechanistic basis of m6A-mediated mRNA regulation

The complex organisation of the human body and its response to challenges requires a fine control of the translation of the gene code into proteins in different cells and at different times. The dynamic modification of the messenger RNA (mRNA) molecules play an important part in this process, as this modification is read by regulatory proteins that add a layer to the regulation of gene expression.

Methylation of the Adenosine in position 6, N6-Methyladenosine (m6A) is the most common mRNA modification. The level of m6A methylation is dynamic and is controlled by a set of 'writer' and 'eraser' proteins. In turn, the code created by the modified m6A is read by a set of 'reader' proteins. Mis-function or mis-expression of these proteins is linked to important systemic illnesses, such as diabetes and many forms of cancer. While the canonical pathways for the writing, erasing and reading of m6A methylation have been clarified, recent work has shown that a larger number of non-canonical readers exist. These readers are proteins known to bind and regulate mRNA and the discovery relates m6A regulation to general RNA regulation processes, such as mRNA splicing and enhanced mRNA turnover. However, for most non-canonical readers, we do not understand this relation at the molecular level.

We work on IMP1, a highly conserved RNA-binding protein that regulates mRNA metabolism, transport and translation. It has been recently shown that the regulation of a set of IMP1 targets in cancers cells is dependent on m6A methylation, and this methyl-regulation is important for the synthesis of the cancer protein c-myc. The questions we are asking are i) how the IMP1 protein recognises m6A methylated targets and ii) which is the reach of m6A methyl-regulation of IMP1 in cancer cells.
To answer these questions we will first determine the structure of the IMP1 protein in complex with the methylated RNA target and compared it with the structure with a non-methylated RNA, which we have published in 2017. Then, using the structure, we will design mutations to create IMP1 variants that recognise only non-methylated RNA. Finally we will use these variants to determine, in the cell, which mRNAs are bound by IMP1 in a methyl-dependent fashion, i.e. to define the selectivity of m6A methylation. In addition we will create biophysical models that explain the differences between the recognition of methylated and non-methylated RNAs and how these are differences are related to protein and RNA concentration.

We expect this work will clarify, at the molecular level, how IMP1 recognises the m6A methylated target mRNAs to regulate their stability in cancer cells. Further, we expect the molecular features of recognition will be common to a growing class of related non-canonical m6A methyl readers and that the concepts and tools provided by this study will allow the investigation of the role of m6A in these proteins. Together the work will provide a first paradigmatic molecular insight into how m6A regulation is integrated in global RNA regulation networks in cancer cells.

m6A-mediated regulation of IMP1 promotes the stability of a number of mRNA targets in cancer cells, such as c-myc, and regulates cellular proliferation. The goals of the proposed research are to understand at the molecular level the selectivity of IMP1 m6A-mediated regulation and the reach of this mechanism. We propose to:

1) Understand IMP1 selective recognition of m6A RNA at the structural level. Determine the structure of the IMP1-m6A RNA complex, compare it with the one of the IMP1-non methylated RNA complex and validate the observed determinants of m6A specificity.

2) Define which of the IMP1 targets is recognised and regulated in an m6A- dependent fashion. Design and test IMP1 mutants that bind to the non-methylated RNA target sequences but not the methylated ones. Use these mutants in comparative iCLIP and RNAseq assays to define which of the targets is recognised and regulated in a m6A-dependent fashion. Analyse the result in terms of gene function and regulation pathways.

3) Provide a biophysical framework to understand IMP1 recognition of m6A methylated mRNA in the cell. Using BLI and other biophysical methods analyse the kinetics of the interaction with the well characterized c-myc and beta-actin targets of IMP1, methylated and non-methylated. Explore how methylation changes mRNA re-modelling, and the response to variation in the concentration of protein and RNA.

4) Expand to include IMP1 targets that are bound in m6A-dependent and independent as defined by the iCLIP results. Model the effect of m6A methylation at different concentration of RNA and protein.

Our findings will explain how m6A methylation create an additional layer of regulation that superimpose to the IMP1 regulatory network and will facilitate modelling the phenomenon in other related non-canonical methyl readers.

Fit with BBSRC priorities: The work will provide a key insight in the mechanisms underpinning m6A regulation via non-canonical methyl-readers. Mis-functions of the m6A methylation pathways are linked to a number of long-term pathologies including cancer, diabetes and neuro-pathologies, and the understanding of m6A regulation is important for maintaining lifelong health, a priority for BBSRC. Importantly, our use of molecular and structural information to design new tools that can be used to filter the result of cutting-edge genomics studies is central to the concept of exploring new ways of working, an important enabling theme for the BBSRC.

Public Sector: The almost complete lack of disease modifying therapies for most neuro-developmental disorders and suboptimal therapies for diabetes and some cancers are, at least in part, due to a lack of molecular understanding of the underlying disease mechanisms. Improving the outcome and efficiency of the treatment of this pathologies is currently a challenge for the NHS, which is struggling to cope with an ageing population. The project has the potential to deliver a molecular understanding of a key regulatory process involved in cancer cell invasiveness, and provide tools and concepts to be used in the investigation of similar systems important in neuronal physiology, memory and diabetes. In the medium term (~10 years), this molecular understanding may facilitate the delivery of small molecules to control cancer metastasis and promote neuronal and metabolic health.

General public: A better understanding of the regulation of IMP1 and of its paralogues may guide strategies to contain the risk of tumour metastasis and be of general benefit to the lifelong health of the general population. The concepts / experimental approaches that will be explored here are applicable to a range of RNA binding proteins, which are increasingly implicated in neuropathologies e.g (e.g. IMPs and FMRP proteins). This proposal therefore has significant amplification potential.

Societal Impact: We expect the outcome of this research to have an impact on our understanding of the mechanisms underlying m6A methylation in cancer. This impact will be manifested not only by potentially guiding intervention strategies, but also by highlighting the role of epitranscriptomics in mRNA regulation and cancer. This may help promote further studies in epitranscriptomics in other human health priorities, for example in brain development and function and the neurosciences.

Training and skills development: The work described in this proposal will result in the direct training of the PDRA in cutting-edge NMR and biophysical techniques and, in year two, in transcriptomic analysis, thanks to the work done in collaboration with the Ule group. Importantly, the multi-disciplinary interaction between the collaborating groups will expose all the group members to new ideas and approaches, and will be particularly beneficial to the PhD, master and undergraduate students in the groups, further amplifying the training and development benefits.