Puffs and PCR: the in vivo dynamics of early gene expression during ecdysone responses in Drosophila

Abstract

The steroid hormone ecdysone orchestrates insect development by regulating gene networks. In Drosophila the most detailed description of ecdysone action is the sequential activation of early and late puffs in the polytene chromosomes of the late larval salivary gland. A number of these early puffs (2B5, 74EF and 75B) contain complex transcription units (Broad-Complex, E74 and E75 respectively) encoding families of regulatory proteins which are expressed in most if not all tissues. In vitro, transcripts of the different isoforms of these early genes as well as the ecdysone receptor (EcR) present varying dose response characteristics (Karim and Thummel, 1992, EMBO J. 11, 4083-4093). We have developed an in vivo approach using a reverse transcription-polymerase chain reaction assay (RT-PCR) so as to visualise these transcripts in the RNA extracted from a single salivary gland. Using one salivary gland lobe for developmental puff staging and the sister lobe for RT-PCR, we have obtained precise developmental profiles for these transcripts and have extended our study to other tissues and stages where puffing studies were not possible. In the salivary gland we have characterised three distinct ecdysone responses. For the mid and late third larval instar responses our results confirm and extend the conclusions of the in vitro studies concerning the temporal expression of the early gene isoforms. The relatively brief prepupal response contains elements in common with each of the larval responses and all three can be explained by the profiles of the respective ecdysone peaks. Interestingly EcR transcripts respond differently during each response. The analysis of different tissues of the same animal reveals subtle differences in the timing of the ecdysone response and isoform expression and suggests that this may reflect tissue differences in the ecdysone profiles. As these molecules have homologues in vertebrates, our analysis may have general implications for the organisation of hormonal responses in vivo.