Abstract

Well-controlled gene expression is critical for the proper development and function of many traits. Highly-specific temporal and spatial expression patterns are often due to the overlapping activities of activator and repressor sequences that form cis-regulatory elements called enhancers. While many studies have shown that evolutionary changes in enhancers can result in novel traits, few studies illuminate how enhancers originate, how activator and repressor sequences interact during enhancer evolution, and the order in which they evolve. Here, we traced the evolutionary origin of a recently evolved enhancer that drives the expression of the fatty acyl-CoA elongase, bond, specifically in the semicircular wall epithelium (swe) of the Drosophila male ejaculatory bulb (EB). We show that this enhancer consists of two activator regions that drive bond expression in the entire EB and a repressor region that restricts expression specifically to the EB swe. Interestingly, the repressor region preceded the evolution of the two activator regions. The evolution of the first activator region, consisting of two putative Abdominal-B sites, did not drive expression in the EB due to the action of the repressor region. Expression of bond in the EB swe requires the evolution of the second activator region, which does not drive expression on its own, but synergizes with the first activator region and the repressor region to produce a highly-specific spatial expression pattern. Our results show that the origin and evolution of a novel enhancer require multiple steps and the evolution of repressor sequences can precede the evolution of activator sequences.