The role of Ultrabithorax negative autoregulation in Drosophila melanogaster

One of the more striking features of animal development is that a limited set of developmental control genes is used repeatedly, in different contexts (within an organism and between species), to form different structures. To achieve this, gene regulatory networks must be versatile. Transcription factors regulate target genes by acting combinatorially, and must be deployed with spatial, temporal, and quantitative precision. In addition to being versatile, gene regulatory networks are robust, enabling animal development to yield reproducible outcomes despite environmental and genetic variation. Focusing on the D. melanogaster Hox gene Ultrabithorax (Ubx), I explore how cis-regulatory elements of developmental control genes contribute to these two hallmarks of developmental biology: versatility and robustness. Ubx specifies the identity of the third thoracic (T3) segment along the anterior-posterior axis of the developing fly. It is required for the development of T3 appendages including the haltere - a dorsal appendage that helps the fly balance during flight. Not only is Ubx presence required, but its levels are also important: Ubx levels are inversely correlated with haltere size. In Chapter 2, we describe how Ubx negative autoregulation establishes different Ubx levels in two different spatial domains of the developing haltere: the proximal haltere (which forms the joint and body wall in the adult) and the distal haltere (which forms the capitellum - the appendage proper). Ubx directly represses its own transcription with the aid of Homothorax (Hth) and Extradenticle (Exd) in the developing proximal haltere. Distally, Hth is absent, Exd is cytoplasmic, and Ubx levels are high. We identify an enhancer that captures this regulation and identify a binding site for Ubx/Exd/Hth. In Chapter 3, we describe another function for Ubx negative autoregulation: promoting developmental robustness by buffering haltere size against changes in Ubx levels. Haltere size is inversely correlated with Ubx levels, but neither haltere size nor Ubx levels change in step with changes in Ubx copy number, suggesting the possibility of phenotypic buffering. Consistently, certain Ubx enhancer traps are silenced in response to increases in Ubx gene dose. Here, we show that functional Ubx protein must exceed a certain threshold to silence Ubx enhancer traps, confirming the idea that it reflects Ubx negative autoregulation at work. Together with the results from Chapter 2, this shows that a single gene can employ the same mechanism to achieve two seemingly opposing purposes: conferring variation and robustness to its expression. Finally, we investigate Ubx enhancer trap silencing in response to naturally occurring genetic variation. We previously described that the same Ubx enhancer traps that are silenced by increases in Ubx copy number are also silenced in F1 offspring of outcrosses to certain wild populations of D. melanogaster. Although it is unclear if this is due to Ubx negative autoregulation or an independent mechanism, our data argue that the Ubx locus, and not the P-element insertions themselves, are being silenced. Interestingly, we find that i) silencing is suppressed by a gain-of-function mutation in a gene that opposes the spread of heterochromatin and ii) the expression of Position Effect Variegation reporters also changes when outcrossed to certain wild populations of D. melanogaster. Together, these results suggest that there are considerable fluctuations in the transcriptional landscape between different populations of a given species