The precise control of gene expression is required for the development of all multicellular animals. When control mechanisms are disrupted, genes are expressed in the wrong time or place, which can lead to serious defects and disease states. However, defining the cis-regulatory sequences that control developmentally regulated genes is still a major challenge, even in the age of complete genome sequences. Here, the cis-regulatory information that controls body patterning in the Drosophila embryo will be studied, with particular focus on genes activated by the morphogenetic protein, Bicoid (Bed). Bed target genes are expressed in a myriad of patterns, which divide the embryo into regions that will form specific body structures later in development. Previous work focused on the transcriptional activities of a cis-regulatory module (CRM) that controls a single stripe of the pair-rule gene even-skipped [eve stripe 2 (eve2)]. This 500 bp module contains 13 clustered binding sites for genetically defined regulators including a cluster of five Bed- binding sites. Mutations in individual sites predictably alter eve2 expression, suggesting that its position in the embryo is controlled by its unique number, combination, and arrangement of sites. Here, an integrated combination of molecular genetic, bio-informatic, and genomics approaches is proposed to study how the patterning information in the Bed gradient is interpreted at the level of the whole genome.
In Aim 1, genetics and misexpression experiments will be used to rigorously classify 21 known Bed-dependent CRMs according to their expression patterns and their transcriptional inputs. Co-regulated CRMs will be """"""""mined"""""""" for sequence signatures that might reflect function.
In Aim 2, bio-informatics approaches will be used in genome-wide searches for novel CRMs, which will be tested by in vivo reporter genes. Iterations of the experiments in Aims 1 and 2 will refine recognition models for predicting functional CRMs.
In Aim 3, the Bed morphogen hypothesis will be rigorously tested by creating embryos with different uniform levels of Bed protein. Expression micro-arrays will monitor genome-wide transcription profiles, which will identify target genes missed by the bio-informatics predictions. ChIP assays will also be performed to assay the role of binding site occupancy in target gene activation and repression. These experiments will identify new Bed target genes, and lead to a cis-regulatory code for predicting expression patterns from genomic sequences.
