Main

Sometimes taking the old and giving it a few twists can yield surprising results. Michael DeVit, at the time in Stanley Fields' laboratory at the University of Washington, experienced this when he explored new ways to investigate the effects of protein-protein interactions. DeVit and Fields took the basic premise of a yeast two-hybrid screen—finding proteins that can bind to a bait protein—and reversed it: they forced proteins to bind to their bait and screened for those that activated a pathway of interest (DeVit et al., 2005). This approach proved to be well-suited to dissect components of biological processes such as signal transduction.

The first step in this assay, forcing the protein interaction, was the most technically challenging one. DeVit used the leucine zipper sequences of the transcriptional activators Jun and Fos, which are known to bind to each other with high affinity, and fused one to his bait and the other to his library of yeast proteins. In the first trial of the new system, he used Jun-GFP as bait and bound it to a library of Fos-proteins. By following the localization of GFP in the cell, DeVit could identify the intracellular destination of the target proteins. Although this provided the basis for a screen to find membrane-associated proteins, the real interest of DeVit and Fields was in identifying components of signal transduction pathways.

They focused their attention on a yeast MAP kinase pathway responsible for filamentous growth. DeVit transformed yeast with a fusion of Jun to Msb2, the initiator of the MAP kinase signaling cascade, and a library of Fos-labeled proteins (Fig. 1). Only library proteins that activated the pathway when interacting with Msb2 triggered signaling, which resulted in the growth of yeast; proteins that did not normally bind to Msb2 did not lead to signaling when the interaction was forced, and consequently the yeast did not grow. Using this screen, they identified a previously uncharacterized yeast protein to be involved in Msb2 signaling. Notably, it contained a peptide motif implicated in protein degradation, a finding that could add an interesting dimension to the regulation of the pathway and is now being investigated. Fields pointed out that this result underscored both the strength of the assay—the identification of new partners in a biological process—as well as its potential pitfalls. He cautioned: “Because of the artificial nature of the study, you cannot be sure that anything you come up with will be necessarily meaningful in vivo, so you have to carry out additional studies.”

Figure 1: Schematic of the assay.
figure 1

Msb2, a protein in the MAP kinase pathway, and library proteins, tagged with complementary leucine zippers (Jun or Fos-Lzip), are overexpressed in yeast. If a library protein is involved in the signaling pathway, its interaction with Msb2 can trigger signaling and growth; if a protein is not part of the pathway, interaction with Msb2 will not facilitate growth.

Full size image

Taken as an initial screen for candidate proteins though, the applications of the assay are numerous. Asked for examples, Fields responded: “It would seem that for almost any process, such as signaling or transcription, DNA replication, protein transport and so forth, you can find a key protein that you put your zipper on, and then bring all other proteins to that site and look for phenotypes.” He added, “Effectively the applications are limited only by the number of cell biological assays you can come up with.” With only the imagination of scientists being the limit, this assay is likely to see wide use in exploring the intricate communication in protein networks.