Projects - Overview

For a more complete overview description of the project, click on "About us", and then choose "Research". Here we briefly summarize the overall goal of the project and then describe the individual subprojects.

The overall question posed by the CycliX project is how and when, on a genome-wide scale, are the circadian, cell-division, and nutrient-response cycles regulated and connected. Indeed, although there is ample evidence indicating interconnections, we still know little about the regulatory mechanisms involved. We know of some regulators controlling genes whose genomic state and transcription activity vary during at least two of the three cycles, referred to here as cyclic-nodes, and we have deciphered some limited pieces of circuitry, but we are completely missing a global and exhaustive view of these phenomena. We will use an unbiased approach for the discovery of an extensive set of cyclic-node regulators. The objectives of this proposal are (i) a quantification of the genomic states, in terms of transcriptional activity and localization of certain histone marks and transcription factors, at the different stages of the three cycles, so as to establish a comprehensive set of cyclic-nodes; (ii) the identification of cyclic-node regulators, and (iii) the testing of specific functions of such regulators in cycle-cycle interconnections.

As the experimental system, we will use the mouse and more specifically the mouse liver, as it offers the opportunity to study a normal tissue in the context of the organism, as well as mouse fibroblasts, as they offer the convenience of a cell culture system and the possibility of isolating genetically altered cells without the artifacts of highly transformed established cell lines.

• Subproject 1. Identification of cyclic-nodes
  
  We will first identify cyclic nodes by determining, on a genome-wide scale, the transcription state of all genes at different times during the three cycles. Because existing datasets consist mostly of microarray mRNA expression data which (i) reflect accumulation levels of RNAs rather than ongoing transcription, (ii) are often derived from an eclectic collection of cell lines and organisms, and (iii) lack any information about RNA polymerase (pol) III transcripts, we will generate sets of high quality data reflecting ongoing transcription derived from standardized liver and fibroblast samples by chromatin immunoprecipitation followed by deep sequencing (ChIP-Seq). These datasets will describe occupancy by pol II and pol III as well as localization of histone H3 lysine 4 trimethylation (H3K4me3), which marks active pol II promoters, and histone H3 lysine 36 trimethylation (H3K36me3), which marks transcribed regions, at different stages of the three cycles. We will then identify those genes (cyclic nodes) whose activity is regulated in more than one cycle.
• Subproject 2. Interaction network at cyclic-nodes
  
  Having identified cyclic-nodes, we will want to identify cyclic-node regulators, particularly those regulators that interconnect and coordinate the cycles. To achieve this goal, we will first identify the cis-acting elements that control expression of these node genes, and the trans-regulatory factors binding to these elements. This will identify cyclic-node regulators, a subset of which will be involved in cycle-cycle interconnections and might be "master regulators".
  
  In this subproject, we will answer the following questions:
  - Which are the active regulatory elements in mouse liver and fibroblasts at different phases of the three cycles as determined by genome-wide mapping of (i) DNase I hypersensitive sites, (ii) p300/CBP occupancy, and (iii) H3K4me1?
  - Which are the silenced regulatory regions as determined by genome-wide occupancy of NCOR1 and NCOR2?
  - Which are the cis-regulatory elements controlling node genes?
  - Which candidate cyclic-node regulators are transcription factors that can bind to these elements directly as identified in a yeast one-hybrid approach?
  - Which of the candidate cyclic-node regulators indeed bind to the expected sites in mouse fibroblasts and liver?
  
  
  The identification of cyclic-node regulatory elements and regulators as well as the interactions between them will allow us to generate a first draft of the core regulatory network linking the three cycles. With this draft, we will be able to proceed to Subproject 3, where we will test the functions of cyclic-node regulators within the network of interconnected cycles.
• Subproject 3. Functional consequences of cyclic-node regulator perturbations.
  
  Subproject 2 will provide us with a list of candidate transcription factors interconnecting two or three cycles. In the last subproject, we will test directly the function of such factors in interconnecting cycles. We will up- and down-regulate these factors and determine the effects in individual cycles and in cycle-cycle interconnections. We will perform the experiments in cultured cells and start the experiments in mice, prioritizing the nutrient-response cycle as it is not possible to mimic properly its complexity in cultured cells.