When I started my own group in 1986 at the University of Zürich (Switzerland), I asked how developmental selector genes, such as the HOX genes Ultrabithorax, Abdominal-B and labial, are transcribed in response to positional information along the anteroposterior body axis of the embryo. Ultimately, this boils down to the question of how transcription factors bind to cis-acting control elements (transcriptional enhancers and silencers) to decode and respond to positional cues from within and outside cells. We used P-element transformation to insert reporters into the Drosophila genome comprising lacZ linked to short sections from the ~30kb upstream or ~45kb of intronic sequences of Ultrabithorax, as previous genetic analysis pointed to cis-acting control elements within these sequences. We thus discovered discrete segment-specific transcriptional enhancers that contain binding sites for positively-acting homeodomain proteins such as Fushi Tarazu and Even-skipped. We also identified silencers with binding sites for the Zn finger protein Hunchback, which acts in a hit-and-run fashion to promote recruitment of Polycomb complexes. Upon stable tethering to their cognate silencers, these complexes shut down Ultrabithorax permanently outside its normal realm of expression (e.g. at both ends of the anteroposterior body axis), whereby this OFF state is inherited throughout subsequent cell divisions.
When I arrived at the MRC Laboratory of Molecular Biology (LMB) in Cambridge in 1991, I reduced my group to four members (for space reasons), and thus decided to wind down the silencing work. Instead, we focussed our efforts on the transcriptional regulation of Ultrabithorax and labial in the embryonic midgut in response to positional signalling (see next paragraph). However, the silencing work was pursued by one of my first students, Jürg Müller and his team (at the Max Plank Institute of Biochemistry in Munich, Germany).
Image Gallery
