--> Back to the Genome Discovery Home Page
CREB/TRANSCRIPTION FACTOR
BACKGROUNDER, Version 1
The transcription factor CREB plays a prominent role in the mechanisms underlying memory consolidation, addiction, circadian rhythms, and developmental plasticity. In Drosophila, expression of an inhibitory form of CREB impairs memory formation, while expression of an activating form enhances memory consolidation. Similarly, genetic perturbations in CREB impair memory consolidation and long-term synaptic plasticity in a variety of vertebrate models.
The appeal of the CREB pathway for study stems not only from its importance in synaptic plasticity, but also from its simplicity. As a result, this pathway provided some of the first insights into the control of gene expression by protein phosphorylation, one of the central themes in transcriptional regulation.
The most widely held model is that CREB binds in a non-regulated manner to a DNA element termed the cAMP-regulated enhancer (CRE), found in a large number of gene promoters. Gene activation is mediated by a multiplicity of kinases that modify the CREB protein, allowing recruitment of the coactivators, CBP and p300. Thus, which genes become activated is determined by the presence or absence of a CREB binding site in the gene promoter and activation per se is mediated by the functions of CBP/p300. This model developed largely from studies in the Goodman lab and has been extended by a larger number of other research groups. Remarkably, this paradigm is based on analysis of only a very limited number of genes, most of which have been examined in fairly artificial contexts. Evaluation of the model as it relates to the population of endogenous promoters is clearly needed if this pathway is to be targeted for therapeutic intervention.
The existence of CRE sequences in multiple gene promoters allows the potential for coordinate regulation, that is, the simultaneous activation of families of genes in response to a given signal. In the parlance used to describe bacterial genomes, this would be considered the CREB regulon. It probably would not be beneficial for a cAMP signal, for example, to activate all genes in a given cell that contain a CRE, however, and this scenario becomes even more difficult to imagine when one considers that at least a dozen distinct signaling systems have been shown to activate CREB. That each of these signaling systems activates exactly the same populations of genes is unlikely. One of the most important pathways, at lest for synaptic signaling, involves calcium. It is entirely unknown whether the same set of genes is activated by cAMP and calcium because studies to date have focused on the expression of individual genes (in highly artificial contexts) rather than on populations. Additionally, calcium signaling provides the potential for regulation that might not be shared by other signal transduction pathways. For example, particular aspects of the calcium signal, either its intensity or duration, could determine which sets of genes become CREB targets and, ultimately, what programs of gene transcription are activated. It is difficult to imagine how this could occur very precisely at a level other than CREB binding.
Previous studies have generally assumed that the presence of a CRE in a gene promoter is synonymous with CREB binding. This assumption is unlikely to be correct, however. For example, a database search for binding sites for the yeast transcriptional activator Ga14 identified more than 200 potential target genes, of which only 1-0 were supported by experimental analysis. Thus, despite their potential predictive value, computational analysis of transcription factor binding sites still requires verification by more labor-intensive experimental techniques. Because of the central role of CREB in synaptic signaling, information generated in our project will help close the gap between signaling events at the cell membrane and the long-term changes in the nucleus that underlie plasticity.
The possibility that different types or patterns of extracellular signals direct CREB to distinct populations of promoters has not been explored and would have profound implications on our understanding of neuronal gene regulation. To address this hypothesis, one must interrogate the entire genome rather than isolated genes. We will accomplish this using a novel method for analyzing CREB binding sites. Our approach allows us to establish not only which genes contain promoters capable of interacting with CREB (i.e., the CREB regulon), but also whether different signals promote CREB binding to distinct targets. This proposal describes an unprecedented combination of approaches to address a question that as implications for investigators in many fields of neurobiology.
Our specific goal is to characterize the family of CREB-regulated genes. Bioinformatics can predict potential CREB targets but cannot identify which promoters actually bind CREB in vivo. We have already shown that our technique, which combines chromatin immunoprecipitation (ChIP) with the analysis of genomic signature tags (GSTs), small fragments of DNA capable of specifying unique sequences in the genome, will be successful but are currently limited by the lack of a high-throughput DNA sequencing facility and adequate bioinformatics. Nonetheless, our preliminary data prove that our approach will allow us to determine whether CREB binds specifically to different CRE elements in brain as opposed to other tissues and whether binding to particular promoters is induced differentially by activation of specific types or patterns of neuronal signaling. Although we are using the method initially to identify CREB target genes, it should be applicable to any transcription factor. Thus, we next plan to use this approach to identify the target genes for transcription factors that direct neuronal development. One such factor is Phox2, which directs the differentiation of cells that produce dopamine. Abnormalities in dopamine-producing cells underlie disorders such as Parkinson’s disease and the identification of Phox2-driven target genes may be useful for stem cell therapy.
|
Richard Goodman, M.D., Ph.D.
Professor of cell and developmental biology, and biochemistry and molecular biology, OHSU School of Medicine.
Soren Impey Ph. D.
|
