Chemical and molecular biology of a eukaryotic riboswitch

The genome of an organism is composed of all of its genes, which can range from a few thousand in bacteria to 30,000 in higher organisms such as plants and animals. The genes are made of DNA, which is transcribed (copied) into messenger RNA (mRNA), which is then used as a template to assemble amino acids into proteins. Even in the simplest cells not all of the genes are expressed all of the time. In order to control cellular metabolism, and respond to changes in the environment (such as the availability of different nutrients), a cell must be able to regulate the expression of its genes. Until recently, this control was thought to be exclusively under the influence of protein factors, which, when exposed to different metabolites, are able to bind DNA or RNA, and alter gene expression. In the last few years a different mechanism has been shown to operate, whereby a metabolite can directly interact with mRNA and, by changing the structure of the mRNA, interfere with the expression of the gene. The regions of mRNA that bind the metabolites have been called riboswitches, because they act as metabolic switches, turning gene expression on or off. Most of the research that has been carried out on riboswitches has focused on bacteria, whose DNA is free in the cytoplasm of the cell. Much less is known about the phenomenon in eukaryotes, cells where the DNA is contained in the nucleus. We have found evidence for a riboswitch in Chlamydomonas reinhardtii, which is a eukaryotic green alga - a simple plant. Like all plants, Chlamydomonas needs to make all its amino acids, one of which is called methionine. Chlamydomonas has two enzymes to make methionine: MetE and MetH. MetE can work on its own, but MetH needs vitmain B12 as a cofactor. However, Chlamydomonas cannot make vitamin B12, it has to take it up from the medium in which it grows. So when vitamin B12 is absent, the alga uses MetE, but if vitamin B12 is present it can use MetH, which is a more active enzyme. We have found that vitamin B12 causes the metE mRNA to disappear, and furthermore the metE mRNA binds to vitamin B12 immobilised on small beads. It is therefore very likely to be regulated by a riboswitch mechanism. In this proposal we intend to use a number of different techniques to investigate the affinity of different derivatives of vitamin B12 (cobalamins) for different parts of the mRNA sequence, determine which nucleotides within the mRNA are important for binding cobalamin, and then mutate these to see if this diminishes metabolite binding. We will test the effect of these mutations in the cell by monitoring the levels of mRNA, and by linking the riboswitches to the gene for an enzyme called luciferase. The expression of luciferase can be monitored by the fact that it gives off light. We will also investigate the structure of the cobalamin-mRNA complex. We will then determine how these riboswitches actually function in the cell. Riboswitches could interfere with a number of different processes, such as transcription, mRNA processing or translation, as well as the degradation of RNA. We will carry out several experiments to determine which processes are controlled by riboswitches. In these experiments we will use the natural riboswitch sequence, as well as the mutant sequences that we will have created in the first part of the proposal. Finally, we will investigate the possibility that two other genes in Chlamydomonas are also regulated by a riboswitch mechanism.

Riboswitches are short sequences within mRNA molecules that are able to bind metabolites directly with a high degree of specificity, without the need for intermediary proteins. The binding of the metabolite causes a conformational change within the secondary structure of the RNA molecule, which then affects the expression of the gene. One of the first riboswitches to be characterised was the adenosylcobalamin riboswitch in the 5`UTR of the E. coli btuB gene, which encodes a protein for the transport of cobalamin (vitamin B12). Binding of cobalamin to the mRNA sequesters the Shine-Dalgano sequence in a stem-loop, thus interfering with ribosome binding and inhibiting translation. Almost all of the research on naturally occurring riboswitches has been carried out in bacteria. The only documented report of an endogenous riboswitch is in the thiA gene from Aspergillus oryzae, where the binding of pyrithiamine (a thiamine analogue) to the mRNA is thought to interfere with splicing. The presence of a riboswitch has also been inferred in a thiamine biosynthesis gene from Arabidopsis thaliana by bioinformatics. Recently, we have made the exciting discovery of potential cobalamin binding riboswitches in genes involved in vitamin B12 metabolism from the eukaryotic green alga Chlamydomonas reinhardtii. We found that one of these genes, encoding the vitamin B12-independent methionine synthase (metE) is repressed when C. reinhardtii is grown in media containing vitamin B12. The mRNA for metE binds specifically to vitamin B12 immobilised on agarose beads, and there is a sequence at the 5' end of the mRNA that is similar to the consensus cobalamin riboswitch sequence found in bacteria. In this proposal we intend to characterise the potential metE B12-riboswitch in detail, and to investigate the possibility of riboswitches in two other genes of vitamin B12 metabolism, metH (vitamin B12-dependent methionine synthase) and pheB (a cell-wall associated vitamin B12 binding protein). In the first part of the proposal we aim to determine how different cobalamin derivatives are able to bind to the metE mRNA. We will use equilibrium dialysis to determine the binding affinity for cobalamin for different regions of the 5' end of the mRNA. We will then study the binding characteristics in more detail using Isothermal Titration Calorimtery (ITC) and surface plasmon resonance. Lastly we will use a technique called in-line probing to establish the stem-loop structures that form and which bases interact with the ligand. Once key nucleotides have been identified they will be mutated, and the effect of the mutations on ligand binding will be determined by ITC and in-line probing. We will also determine the consequence of mutating these riboswitches on gene expression in vivo by using RT-PCR and luciferase reporter assays. We will initiate crystallisation trials, and NMR and EPR spectroscopy to start to build a molecular portrait of the interactions of cobalamin with RNA. The other major part of the proposal will investigate the mechanism of riboswitch function. Gene expression in eukaryotes has many more stages than in prokaryotes, because the RNA must be spliced, capped, polyadenylated and transported before it is translated and finally degraded, so riboswitch mechanisms are likely to differ from those observed in prokaryotes. We will study the effect of cobalamin on a range of processes including RNA turnover and degradation in vitro and in vivo, translational efficiency, rate of transcription, and investigate if an RNAi mechanism is involved.