Messenger RNA plays a central role in gene expression. Regulation of gene expression is critical to determining phenotype. Disruption of this regulation is a causal or contributory factor in many diseases. Gene expression depends on rates of transcription, the stability of RNA products, the rate of translation of mRNAs in protein production and the stability of these proteins. Furthermore RNAs are often localised to specific cellular regions and their activity may vary in these different molecular environments.Regulation of mRNAs can be affected by other RNAs such as miRNAs and also by interaction with protein binding partners. RNAs do not exist in isolation within living cells - they are decorated with proteins at the site of their creation within the cell nucleus. After nuclear export further proteins bind the RNA and others dissociate to form a heterogenous cellular population of RNA granules. In interaction with the ribosome mRNAs form the active unit of protein translation - the polysome.The binding of the mRNA to other proteins, other RNAs and the loading of ribosomes is affected by sequence features of the mRNA itself. The features of mRNAs that will determine the fate of the transcript can include specific primary sequences of nucleotides, structural elements such as hairpin loops and general features such as transcript length. Many of these features occur in untranslated regions at the 5’ and 3’ ends of the RNA.This thesis explores the translational regulation and localisation of mRNAs. Bioinformatic models are constructed to model the iron-responsive elements (IREs) which function in the 5’ or 3’ untranslated regions (UTRs) of mRNAs as post-transcriptional structured cis-acting RNA regulatory ele- ments. One known functional mechanism is the binding of Iron Regulatory Proteins (IRPs) to 5’ UTR IREs, reducing translation rates at low iron levels. This mechanism has considerable evolutionary conservation - functioning in organisms from sponges to humans. Experimentally proven elements are quite small, have some diversity of sequence and structure, and functional genes have similar pseudogenes in the human genome. These models are used for the detection of known IREs and to predict novel IREs in human genes.De-novo element prediction is then undertaken on a set of yeast mRNAs for nuclear genes encoding mitochondrial proteins. Previous publications showed some of these mRNAs localised to the mitochondria and others did not. The PUF3 element (a binding partner for the Puf3p protein) had been identified as part of one of the mechanisms for the mRNA localisation, but others had not been explained. The search for alternative elements led to a putative primary sequence pattern with potential to bind Pub1p. However, on further investigation collaborators were not able to demonstrate the mitochondrial localisation of these mRNAs despite testing for mRNAs which had shown considerable localisation in the original published data. The motif may be involved in other mitochondrial processes.A more global approach was taken for the analysis of translation efficiency in human genes. Recently large scale transcriptome and proteome datasets for human cells have become available. A bioinformatic analysis of translation efficiency is presented – the rate at which mRNA is translated into protein. We have analysed those human datasets that include genome wide mRNA and protein levels determined in the same study. The analysis comprises five distinct human cell lines that together provide comparable data for 8,170 genes. For each gene we have used levels of mRNA and protein combined with protein stability data from the HeLa cell line to estimate translation efficiency. This was possible for 3,990 genes in one or more cell lines and 1,807 genes in all five cell lines. Interestingly, analysis and modelling shows that for many genes translation efficiency has considerable consistency between cell lines. Some deviations from this consistency likely result from the regulation of protein degradation. Others are likely due to known translational control mechanisms.Gene expression is often measured using mRNA transcript levels. While this is a pragmatic approach it is problematic as expression according to protein level is of direct functional importance. The prediction of protein expression from mRNA expression data is investigated using a dataset for the mouse NIH-3T3 cells. Transcript features such as sequence length and structure in the 5’ UTR are of some use in predicting protein expression. Finally the use of calculated translation efficiency data from the HeLa cell line is shown to be useful in the prediction of protein expression in other human cell lines.
PDF
download
878987797
028-85220240
