A transgenic approach to investigate the RNA binding protein T-STAR

Human genes are found on chromosomes, and encoded by DNA. Recently the almost complete DNA sequence of humans has been worked out, and there are between 20-30 thousand human protein coding genes. Although this is a big number, most recent estimates have shown that the protein number in human cells actually far exceeds the number of genes. An important question has now become how does the cell bridge the gap in numbers. One important way seems to be to use the same gene to encode more than one protein. The 1993 Nobel Prize was awarded for the important discovery that the genes of organisms like humans are split between bits that encode proteins (called exons) separated by non-coding regions (called introns). DNA is copied into RNA which in turn is used to make protein. After RNA is made, exons are joined together in the cell, by removing introns to give the template which encodes protein. Frequently different exon combinations are included into RNA from the same gene, resulting in variation. For instance, sometimes some of the exons are removed along with the introns. This process (called alternative splicing) is critically important in development, and might even have been an important evolutionary step in allowing the development of multicellularity. Despite this, it has not been studied as much as the controls (transcription) which decide which genes are turned on and off to make the RNA in the first place. Alternative RNA splicing is controlled by proteins which bind to RNA in the nucleus. Some of these proteins are not made in every part of the body but only in particular tissues such as the brain or the testis. Evidence so far suggests that these are likely to have very important roles. One of these RNA binding proteins, called T-STAR, is of particular interest since it may play roles in splicing and connecting signalling pathways with RNA processing and possibly even transcription during development. A good way of investigating the function of a human gene is to look at the equivalent mouse gene, and we propose to test the role of T-STAR in mouse development. Mice, like humans, have a T-STAR gene which is turned on in the adult testis, developing brain and kidney. Although mouse and human T-STAR proteins are virtually identical, they have an important difference in that they are regulated differently. For this reason we predict that T-STAR protein will regulate the same genes differently in humans and mice, and this might help explain some of the reasons mice are different from humans. We will make a conditional version of the T-STAR gene in mice. Next, we can inactivate this conditional T-STAR by cutting an important part of it out of the chromosome when we want to. The way we do this is by mating the mice with special mice which express another protein called a 'recombinase'. We will first remove T-STAR from every cell in the mouse body by turning on the recombinase in every cell. We expect to see defects in the brain, kidney and germ cells,but it could be that these mice will die while they are developing . For this reason, we will also selectively remove T-STAR in the testis (where the sperm are made). This is the main site of T-STAR expression in the adult, and it is an unusual organ because you can see all the major stages of sperm development occurring in the adult and it is non-essential. The mice might be infertile but they will not die. Hence by removing T-STAR in this tissue, even if we get a block in development we will be able to analyse this in the adult mouse and obtain mutant cells. We will analyse gene expression in these mice to see if different genes are expressed, and if transcripts contain different exons from mice which do not contain the deletion. To find out if these mouse genes are regulated differently in humans, we will then compare transcription and splicing patterns of genes affected by T-STAR deletion in the mouse with their human counterparts..

Landmark genomic sequencing studies have shown that humans and mice have only between twenty and thirty thousand genes, and that most of these genes are similar between the species. This has created a new challenge: understanding how these genes work together to make an organism. It is becoming evident that gene expression controls at the RNA level play an important role in both development, and also in making the full repertoire of human proteins (around 100-fold more than the number of genes). Up to 75% of human genes are alternatively spliced, which means different exons are used in different cells/tissues. This process is hugely important. For example, incorporation of an extra three amino acids by alternative splicing can convert the Wilms Tumour WT1 protein from a splicing regulator within the nucleus to a transcription factor, and this alternative splice is required for kidney development. Alternative splicing controls sexual differentiation in fruit flies and is also implicated in the same process in mammals, and defective alternative splicing can cause human diseases. Since single celled organisms like yeast generally do not have alternative splicing, it has been argued that the development of alternative splicing was one of the key evolutionary steps allowing the development of multicellularity. We propose to investigate a tissue restricted RNA binding protein called T-STAR which is strongly implicated in post-transcriptional control, and particularly alternative splicing. Although almost identical in sequence, T-STAR is also differentially regulated in humans and mice, suggesting the pathways it regulates might be important in the developmental differences between these species. We will use a conditional genetic approach to remove the T-STAR gene from the mouse, and then see what happens to the mouse. The T-STAR protein is highly expressed in germ cells, and at lower levels in the kidney and brain so we predict that by destroying the T-STAR gene, we will affect development of all of these tissues. We will monitor development of these tissues using classical histology and Optical Projection Tomography. This experiment will tell us the critical sites of T-STAR function, but it might also prevent mouse development past an early stage. We will get over this by using different strains of mouse which encode a tissue-specific Cre recombinase which can inactivate the T-STAR gene from any particular cell type that we wish in the body. In particular, we would like to look at the effect of removing T-STAR on germ cell development, which is the main site of adult T-STAR protein expression. Even if germ cells are defective in these mice, they will still be alive. Although the mice might be infertile, we will be able to isolate cell types exactly at the stage they would normally need the T-STAR protein, and test for any effects on alternative RNA splicing or transcription profiles in these cells by comparing them to wild type mice. Since we predict that T-STAR will control the same pathways of gene expression in humans and mice, but with different patterns depending on the expression of the proteins which control the level of human T-STAR protein, we will compare patterns of gene expression and splicing of identified targets between humans and mice.