Transcriptional activation of the early Drosophila genome

The broad goal of my research program is to understand the molecular mechanisms that underlie early embryonic development. We use a combined genetic and molecular approach to study gene regulatory networks in the early Drosophila embryo. Recently we discovered a key transcriptional regulator, Zelda, which is the long-sought after factor that activates the early zygotic genome. Initially the embryo relies on maternally deposited gene products to begin developing, and the transition to dependence on its own zygotic genome is called the maternal to zygotic transition (MZT). Two hallmark events that occur during this transition are zygotic gene transcription and maternal RNA degradation, and interestingly Zelda appears to be involved in both processes. Zelda protein appears very early, and persists through to the cellular blastoderm stage when cells are actively becoming fated (Figure 1). Zelda activates batteries of genes that in turn prepare the embryo for major developmental events such as sex determination and dosage compensation, cellularization and gastrulation, and axis patterning. Zelda also activates several microRNAs, at least one of which is involved in maternal RNA degradation (Bushati et al., 2007). Zelda is a zinc-finger transcription factor, and target genes share the motif CAGGTAG, and related sites, collectively referred to as TAG-team sites (ten Bosch et al., 2006; de Renzis et al., 2007). We are currently using genomic approaches to further study the role of Zelda during the MZT. Zygotic gene transcription doesn’t begin until one hour after fertilization – is Zelda involved in this timing? Is Zelda protein accumulation or activity a factor? Does Zelda interact with other major regulators such as Dorsal (see below) to regulate downstream genes? If so, what is the molecular nature of this interaction? What RNAs are targets of the Zelda-activated microRNAs? What happens if they are not degraded?

 

My lab is also interested in dorsoventral patterning.  Initially, the Dorsal transcription factor subdivides the DV axis into three main domains (ventral, lateral, dorsal) by regulating the spatial expression of downstream zygotic genes that in turn give further instructions to cells.  For example, sna is activated by high levels of Dorsal in the ventral region of the embryo, while sog can be activated by lower levels of Dorsal in lateral regions, as long as Zelda is present. dpp is activated by Zelda, but repressed by Dorsal in the ventral half of the embryo. We have further focused on how Dpp, a secreted molecule that belongs to the TGF-b superfamily of growth factors, acts to further subdivide the dorsal region into the dorsal-most amnioserosa, a squamous extraembryonic membrane, and the broader dorsal epidermis. We have identified several downstream target genes that respond differentially to the Dpp gradient. Thus far we have learned that the broadly-expressed genes are regulated by a combination of the transcription factors Mad and Medea, the Dpp signal transducers, and the transcriptional repressor Brinker, which is expressed ventrally. It appears that Brinker and Mad define the precise border of target-gene expression by competing for DNA binding to target enhancers. However, the Dpp target genes that are localized to the presumptive amnioserosa require Mad in combination with Zen, homeodomain protein, for transcriptional activation, and Brinker is not involved. We propose that Dpp and Zen act via a feed forward mechanism to globally regulate all amnioserosa-specific target genes, and we are interested in defining the regulatory network that leads from these master regulators to the genes that effect differentiation of blastoderm columnar epithelial cells to a squamous epithelium capable of folding upon itself during germ band extension.