Raabe, T., J. Riesgo-Escovar, X. Liu, B. Bausenwein, P. Deak, P. Maröy, and E. Hafen. 1996. DOS, a novel pleckstrin homology domain-containing protein required for signal transduction between Sevenless and Ras1 in Drosophila. Cell 85: 911-920. The authors screen for mutations that can suppress a gain of function sevenless allele and identify a new gene, daughter of sevenless. Epistasis analyses reveal its function in a parallel pathway upstream of Ras but independent of several other known pathway components. We will emphasize the methods and logic used for identifying and interpreting parallel pathways.

Questions for Thought

1. What is the logic behind this screen? What was the signaling pathway that was generally accepted and why was that pathway called into question? How did the screen work? What genes might be mutated by this approach?

Imagine you are just starting in Ernst Hafen's lab and you want to impress him with your knowledge of signaling pathways. You suggest a screen that would use an activated Sevenless receptor, but this gain of function allele works by increasing the ability of receptors to form dimers. How would the mutations you recover in this screen differ from those recovered here?

2. The dos gene was identified by P element insertion; how did the presence of the transposon facilitate cloning the gene?

The P allele is lethal. The authors used two tests to determine that the gene mutated by the P element was involved in eye development, one specific for the P allele and one used to examine the EMS allele as well. What were these tests? How do they work? Could the authors use their second method to test tissue requirements of dos function? If so, how?

3. DOS shares homology with adaptor proteins thought to interact with RTKs and SH2/SH3-containing molecules. How do the authors test this idea?

What additional support is provided by antibody studies?

4. Background: Note that the authors use a NEW method for generating clones in the eye (Golic and Lindquist, 1989) . They take advantage of a yeast enzyme, FLP, that binds to FRT (Flp Recognition Target) sequences and catalyzes recombination at only the site of the FRT. Xu and Rubin (1993) created strains in which FRT sites are placed near the centromeres of each arm of each chromosome (5 different strains: X, 2L, 2R, 3L, 3R). If you can't picture how this technique works, take a look at the Xu and Rubin paper. We will discuss the advantages of and potential problems with this sort of method for generating clones.

When making clones in the eye, even with this fancy new method, the authors obtained disappointing results. Why? If dos really represents a parallel pathway that provides redundant function with drk, why is it lethal? Why does one have trouble making clones?

How could they get around the problem they encountered in making clones?

5. Draw a schematic representation of the sev pathway, including all the components we have discussed so far (boss, csw, drk, dos, Raf, Ras, rl, sev, sina, Sos). How could dos act in this pathway? How strong is the evidence that it acts downstream of sev and upstream of Ras? Could we draw alternative pathway structures?

How does dos relate to drk? Do these functions complement each other (that is, are they redundant functions)? Why or why not?

How would you test the accuracy of your hypothetical pathway?