September 15, 2009 (Vol. 29, No. 16)
Team Advances Knowledge Through a Series of Investigative Studies on the REST Protein
GEN continues its celebration of the “Year of Darwin” by looking at new research that provides important and novel insights on the evolution of genes.
Scientists at the University of Leeds in the U.K. and the Genome Institute of Singapore say they have discovered one of the mechanisms governing how our physical features and behavioral traits have evolved over the centuries. Darwin proposed that such traits are passed from parent to offspring, with natural selection favoring those that provide the greatest advantage for survival. But he did not have a scientific explanation for this process.
The investigators report that a protein known as REST plays a central role in switching specific genes on and off, thereby determining how specific traits develop in offspring. To learn more about the significance of this finding, GEN interviewed lead researcher Ian Wood, Ph.D., a senior lecturer and a member of the University of Leeds’ faculty of biological sciences. He works in the university’s Institute of Membrane and Systems Biology.
You can also hear the entire interview as a podcast by going to www.genengnews.com/genCasts.aspx?aid=3030.
GEN Let’s start by addressing the following question: How have most scientific studies tried to shed light on the genomic basis for phenotypic variation?
Dr. Wood: Most previous studies have looked for differences between sequences within protein-coding regions of genes from different species. A change of even a single nucleotide in a DNA sequence of a gene’s protein coding region can result in an amino acid change of the protein and, from there, possibly a change in the properties of the protein itself. Now that we have genome sequences available for many species, researchers have been able to look for those differences between species, and they’re relatively straightforward to identify.
GEN Is there anything wrong with this approach and is there a better path to follow?
Dr. Wood: There’s nothing wrong with this approach per se. However, it’s likely to only identify a small set of the genomic variation that we actually see and that is responsible for species diversity. It’s been proposed for quite a number of years that most of the variation that we see between different species is not because the proteins expressed by those species are different, but it’s because the proteins are used or expressed in different ways by each of the different species.
As an analogy, imagine a set of houses built of bricks. The bricks themselves are likely to be similar, if not the same. Yet, by putting these bricks together in various ways, we can build houses that look different from each other. Based on this scenario, looking at the changes in protein-coding sequences within genes, which has been the approach taken to date by most researchers, is equivalent to looking at differences between individual bricks from one house to another.
What we’ve done is to ask how the regulation of expression of these genes varies between species. This is equivalent to asking what are the differences in the way bricks are put together to build each individual house. We believe this is the better path. But it’s only been recently that the right tools have become available to be able to do this sort of work. Our studies have concentrated on a gene network controlled by a single protein, known as REST.
REST is a transcription factor found in all vertebrate species. So we looked across all vertebrate species to see how the gene network varies among them.
GEN What was your key finding regarding REST?
Dr. Wood: We provide the first report on the extent to which the network of genes regulated by a single transcription factor varies between species. This has important implications, particularly for a protein like REST, which is involved in several human diseases.
For example mutations of REST are linked with colon cancer while patients with Huntington’s disease are thought to have abnormally high levels of REST in the brain. To understand these types of diseases, researchers create genetically altered animals, mostly mice, which can be studied in a way that humans can’t. Our results indicate that when we look at animal models, we can only expect those animal models to mimic some of the features that we would expect to see in humans. This is not something that always seems to be appreciated.
However, our main finding addressed the process by which genes can acquire new regulatory elements over time. We show that REST binding sites arose throughout the genome, most likely by duplication and insertion of existing DNA sequences, resulting in a change in the regulation of certain genes. Over evolutionary time, these new REST binding sites are consolidated by migration closer to the gene they regulate and by changes in the DNA sequence to increase their affinity for the REST protein itself.
GEN: Please tell us more about the REST protein and what it specifically regulates.
Dr. Wood: As a transcription factor, REST recognizes and binds to specific sequences in genomic DNA and regulates the expression of nearby genes. REST is actually a transcriptional repressor. This means it has a negative effect on gene expression. So recruitment of REST to a gene results in the reduction of the level of RNA that’s transcribed from that gene as well as a reduction of the level of protein of that specific gene being expressed within the cell.
GEN How does REST accomplish its regulatory role?
Dr. Wood: REST itself doesn’t directly control transcription. Rather, it recruits other protein complexes via interactions at its C terminus. The N-terminal region of REST interacts with a corepressor called SIN-3, and the C-terminal region of REST interacts with a corepressor called CoREST. Both of these proteins are part of much larger protein complexes, and they each contain enzymes that repress transcription.
The enzymes do this by altering the post-translational modifications that you find on histone proteins within the chromatin. The enzymes recruited by REST include enzymes that remove acetyl groups from lysines within histones and enzymes that remove methyl groups from some of the lysine residues within histones. Other enzymes add methyl groups of different lysine residues within the histones.
As a result, each of these enzymes has a repressive effect on its own. Put all together, the combination of these enzymes’ activities provides a robust level of repression once REST is recruited to a specific gene.
GEN: Your group’s research on REST has led to the creation of a model where new transcription factor binding sites are constantly generated throughout the genome. Briefly describe this model, and explain how the model helps increase our understanding of the evolutionary process.
Dr. Wood: Our observations of different genomes show that REST binding sites were created often by duplication and insertion of new DNA during meiosis. During meiosis, the exact sequences that are passed from one organism to its offspring are generated. When these sites arise near a gene, that gene can become regulated by REST in the offspring of that organism.
We found that newly created REST binding sites appear some distance from the genes that they regulate. In fact, we would expect this from a random event because regions encoding protein actually only make up a small percentage, about 2 percent, of our genome. Thus DNA inserted at random is likely to be inserted in the 98 percent of the genome that doesn’t encode proteins.
We also found out that new binding sites are imperfect and that they bind REST with quite low affinity. Again, this is something we would expect, as there are more ways to generate an imperfect REST binding site than a perfect REST binding site.
However, older REST binding sites are found much closer to the genes they regulate, and they recruit REST with much higher affinity. This provides us with a molecular mechanism for one evolutionary process in which regulation of genes can change over time.
New REST binding sites start off with weak affinity for REST but become stronger with evolutionary time, probably due to single nucleotide changes. The sites themselves also become closer to the genes they regulate. The effect of this is that the gene and the binding site are less likely to be separated by recombination events that occur during meiosis and thus more likely to be passed on together to future offspring.