Working with pea plants in his monastery garden, the Austrian monk Gregor Mendel discovered that they inherit from their parents, what we now know to be genes, which control how they grow. Like peas, the genes in the DNA of our chromosomes have the code for life. But what is that code exactly? DNA is comprised of long chains of chemicals of four different types: A, C, G and T. The genetic code is copied into a related molecule called RNA that is the messenger of this code. RNA is comprised of almost the same chemicals, A, C, and G, but U replaces T. Cellular machines called ribosomes, take the message and use it to build proteins corresponding to this code.

Interestingly, the RNA chemicals can be altered, and by far the most common modification within the messenger RNA chain is m6A. Consequently, messenger RNA is effectively comprised of five different chemicals: A, C, G, U and m6A. You never heard of it? It is surprising how little attention it has had because if humans, flies or plants don't have it, they die. Recently, a human gene called FTO, which is linked to several human diseases, was found to encode a protein able to convert m6A back to A. This revealed that m6A levels in RNA could be controlled, and if this was disrupted, disease could result. It seems that m6A doesn't change the genetic code itself, but it does affect the message and so affects how the code is used in everyday life. This project is all about m6A in plants, but based on what we have done so far, it should tell us about animals and people as well.

Like Mendel, our project results from discoveries we have made with plants. While studying a protein that naturally helps plants flower, Gordon Simpson's team discovered it controlled where messages end. Using a specially developed technique, they discovered that this protein is found close together with enzymes that make m6A. This made some sense because Rupert Fray, an RNA methylation expert, had previously shown that m6A is mostly found near the end of messages. So, using the same techniques to see what proteins were closely associated with the enzymes that make m6A, Gordon Simpson worked with Rupert Fray, and together, they discovered several proteins that were highly related across lots of different plants and animals, that helped these enzymes make m6A not only in plants but in humans as well.

The aim of this project is to understand m6A a lot better by using plants. Plants are vital to our food and energy security so it is important that we know how they work. Because we can make mutant plants in the lab that still live but have altered levels of m6A, we can study them more simply and use that knowledge to try to understand why plants and animals use m6A in the message of their genetic code.

First, we want to know which messages have m6A and where in the message is this found. We want to know if this changes in different situations such as in flowers compared to leaves or when the plant is stressed. Second, we want to know how m6A is made by the factors that help the enzymes we have found. Do they do it to all genes, or only some and only in specific parts of some messages? How do they talk about what they are doing to all the other parts of the cell that are making and reading the code as well? Third, we want to understand exactly what goes wrong when m6A is changed. What happens to individual messages? Finally, we'd like to begin to understand how the m6A code is read. Proteins with YTH domains apparently bind m6A, they are found in plants but we don't know what they do.

We form a hugely experienced team in this area and we hope to learn very basic knowledge about the message of our genetic code. This work will provide state-of-the-art training for early career scientists working as a team on plants, genetics, RNA, proteins and computational analysis of large sequencing datasets - assembling the skills modern plant science needs to ensure future food and energy security.

Technical Abstract:
The most prevalent internal modification of eukaryotic mRNA is the methylation of adenosine at the N6 position (m6A) and there are writers, readers and erasers of this epitranscriptome code. The enzyme that writes this code (MTA in Arabidopsis) is essential for life in Arabidopsis, flies and humans, and specific functions for RNA methylation are emerging. Working with Arabidopsis, we used in vivo interaction proteomics to identify a core set of conserved factors that co-purify with MTA. We have shown that these proteins are required for mRNA methylation in Arabidopsis and HeLa cells, and refer to them as the m6A writer-complex. These breakthroughs suggest that Arabidopsis can be a generally usefully model system for understanding the role and impact of RNA methylation.

The aims of this proposal are to define the Arabidopsis epitranscriptome, determine how it is regulated and assess the impact on gene expression of disrupting individual writer-complex components.

We will use Me-RIP-seq to identify sites of mRNA methylation. We will test whether m6A is dynamically controlled by quantifying shifts in m6A in different tissues and in response to stress. We will examine functional conservation of m6A by Me-RIP-seq of the crop plants rice and tomato. We will assess regulatory roles of the writer-complex by identifying in vivo targets (ChIP-Seq), the impact of disrupted writer-complex component function on specific m6A modifications (Me-RIP) and identify in vivo protein partners. We will analyse the consequences for gene expression in these functionally compromised backgrounds by quantitative RNA-seq. Finally we will begin the first characterization of Arabidopsis YTH domain proteins that can bind m6A.

This collaboration combines expertise in RNA methylation (Fray), the molecular and proteomic analysis of RNA processing (Simpson) and quantitative analysis of high throughput sequencing data (Barton).

Potential Impact:
Public health, Pharmaceutical companies, Taxpayers:
It is now clear that methylation of adenosines in mRNA performs a post-transcriptional gene regulatory role in all Eukaryotes studied. In humans, increased activity of the mRNA de-methylase is associated with increased risk of obesity, diabetes, and Alzheimer's as well as certain cancers. We have shown that MTA, MTB, Fip37, PX1 and a conserved E3 ubiquitin ligase associate together in plants and at least the first four of these are required for mRNA methylation. The same five proteins are also known to associate together in both mammals and Drosophila, although in the case of the last two proteins, a link to methylation has not yet been reported. Thus Arabidopsis, with its ease of transformation, ability to grow even when physiologically compromised and with its lack of welfare issues, is already proving a useful multicellular model organism in which to study fundamentals of the methylation process. Both the E3 ubiquitin ligase and Wilm's Tumor Associated Protein (the human homologue of Fip37) are associated with multiple tumor types and cancer prognosis. In addition, both of the human YTH domain m6A binding proteins have been implicated in various cancers or disease states. Thus scientists and pharmaceutical companies interested in understanding or developing novel therapeutics or screening services for these disease conditions may be users of data and methodologies generated in this research project. Ultimately, improved treatment, diagnosis and understanding of these disease states will benefit the wellbeing and health of the public. It could also reduce the financial timebomb associated with certain age and obesity related conditions. Thus, the taxpayer and UK government would be potential beneficiaries.

Industry and Academia groups using transcriptomics data:
We and others have shown that methylation of mRNA is playing a role in regulating cell differentiation and developmental pathways in plants and yeast and in mammals it has been shown to have a role in stem cell maintenance. The mechanisms involved are currently poorly understood, but any research group in academia or industry with an interest in understanding or interpreting gene expression will be a potential user of this research. mRNA transcript levels are often a poor predictor of the abundance of the encoded protein, and in mice this has been shown to principally be a result of varying translation rates for different transcripts. A clearer understanding of the role of mRNA methylation may thus lead to the ability to use transcriptomics data to build models that more closely match the system being studied.

Following several key publications in the last 18months, the importance of this post-transcriptional regulation is beginning to be appreciated and the research field is poised to expand exponentially in coming years.