What we have learned so far.

An unexpected finding in the set of genetic experiments analyzing the PB-SCR interaction was an explanation to a genetic contradiction. When SCR is expressed in every cell, the normal aristae are transformed into tarsi, but when the Scr gene in the ectoderm cells of normal tarsi is inactivated, the tarsi are not transformed to aristae. This is a contradictory result because ectopic expression of SCR results in tarsus determination of arista, but SCR is seemingly not required for normal tarsus determination. I was able to resolve this contradiction by showing that SCR activity is not required in the ectodermal cells of the tarsus for tarsus determination (Percival-Smith et al., 1997). Instead, SCR activity is required in the mesoderm of the tarsus for tarsus determination. From this result, I proposed that SCR is required in the mesoderm for the synthesis of the Tarsus-inducer, and when the Tarsus-inducer is secreted from the mesoderm cells it acts on the overlaying ectoderm cells determining tarsus identity. In subsequent experiments, we found that a target of the Tarsus-inducer is whether a protein Extradenticle (EXD) is in the nucleus or the cytoplasm. When EXD is in the nucleus, arista identity is determined, but in the presence of the Tarsus-inducer, EXD protein is localized to the cytoplasm and now tarsus identity is determined (Percival-Smith and Hayden 1998).

Where we are going.

The Tarsus-inducer is an SCR-regulated gene whose activity is required for tarsus determination and formation. Hence, the Tarsus-inducer gene represents a HOX-regulated gene that is clearly required for the formation of a body part. This is why we want to isolate the gene encoding the Tarsus-inducer. Using a genetic screen, we have identified potential cytological regions that the Tarsus-inducer may reside, and knowing that the Tarsus-inducer is expressed in the mesoderm of legs and not antennae, we are trying to identify the gene encoding the Tarsus-inducer.