Structural and functional analysis of SR proteins

The genes in our cells contain the information to produce proteins, which provide the building blocks for our cells and regulate how the cells of our body function. To make a protein, a gene is first copied to produce messenger RNA, which in turns carries the information to make a protein. Different parts of a messenger RNA can be chopped out before it is used to make a protein, which leads to the production of different proteins from the same starting messenger RNA. This means that although humans have around 25,000 genes they can produce many more proteins than this. The different proteins produced from a messenger RNA can have quite different functions in our cells. In this project we are going to work out the molecular structures of some of the important proteins in the cell, known as SR proteins, that regulate which parts of a messenger RNA are chopped out. SR proteins also help the messenger RNA to move from the nucleus to the cytoplasm of a cell, where the mRNA is used to make protein. With these molecular structures we will be able understand how these proteins recognise the bases of messenger RNA to generate specific chops in it and how they recognise other proteins in the cell which help the messenger RNA to leave the nucleus.

Approximately 50% of human genes are alternatively spliced and this generates enormous proteome diversity. Two classes of proteins are largely responsible for making alternative splicing decisions, these are the hnRNP proteins and the SR proteins which recognise specific elements in RNA known as exonic splicing enhancers. SR proteins also play an important role as mRNA export adaptors, binding TAP. They also regulate mRNA stability and stimulate translation in the cytoplasm. Despite their discovery 13 years ago and their pivotal role in eukaryotic gene expression, there are currently no 3D structures for the canonical SR proteins. SR proteins consist of one or two RNA recognition motifs (RRMs) linked to a C-terminal domain rich in serines and arginines. We have defined the minimal domains of three SR proteins, SRp20, 9G8 and SF2 required for stable interaction with the TAP mRNA export factor. The TAP binding domain of SF2 incorporates 4 RS dipeptides which we show regulate the interaction with TAP according to their phosphorylation status. We have devised a method which allows the production of large quantities of functional, labelled, concentrated SR protein domains suitable for structure determination by NMR. We have used this technology to collect 2D spectra for the RRMs from 9G8 and SRp20 together with RRM2 with 4 C-terminal RS dipeptides from SF2, clearly demonstrating that we can determine these protein structures. In this project we will determine the 3D structures using NMR for the 9G8 and SRp20 RRMs which bind TAP, together with the structure for SF2 RRM2-4RS in the hypo and hyperphosphorylated states. We will determine the structure of SRp20 bound to a cognate exonic splice enhancer RNA and we will use chemical shift mapping to define the TAP binding site on all three SR proteins. We will also analyse the molecular basis for remodelling of TAP by SR proteins. For each of these protein:ligand complexes we will carry out a detailed structure/function analysis using site-directed mutagenesis combined with in vitro and in vivo assays. These studies should lead to a better understanding of : a) the molecular basis for recognition of exonic splicing enhancers by SR proteins, which play a major role in determining the complexity of the mammalian proteome, b) the molecular basis for recognition of TAP by SR proteins and c) the structural basis for the phosphoregulation of SF2:TAP interactions. The justification for determining the structures for three SR proteins can be summarised as follows: SRp20 is justified since it is the best target at present for analysis of a protein:RNA complex, SF2 is justified on the basis that it has a non-canonical RRM motif and the construct we are working with can answer questions about the phosphoregulation of SR protein/TAP interactions. 9G8 is justified on the basis that we already have the backbone assignments and the structure calculations and chemical shift mapping experiments with TAP can proceed as soon as the grant starts. Furthermore, the sequences C-terminal of the RRM domains in all three SR proteins are quite divergent, therefore analysis of all three SR protein:TAP interactions may reveal general rules for the molecular recognition in the complexes.