The interaction trap is a method for cloning cDNAs encoding proteins that interact with a known protein. The method also provides a simple and efficient genetic assay for protein-protein interactions. Each interacting partner is epitope tagged so that the interactions can also be detected physically. Moreover, the individual components have been designed so that the system provides a number of internal controls and gimmicks that simplify its use. The interaction trap was developed to aid our search for G1 specific cell cycle regulatory proteins in higher eukaryotes, primarily by Jeno Gyuris and Erica Golemis, with contributions from other members of the Brent lab. It evolved from attempts in the lab (since 1986) to use LexA fusion proteins and LexA responsive reporter genes to find interacting proteins encoded by cDNA libraries. The trap owes a great debt to the work of Fields and Song, who proposed in 1989 that library encoded proteins that interacted with DNA bound fusion proteins could be detected if the library encoded proteins all carried transcription activation domains. This idea was critical, and the interaction trap can be considered a particular version of the two-hybrid system these workers proposed. Numerous other researchers in other labs have also developed useful implementations of this theme, many of which also contain useful refinements. For the technology described here, we ask you to take note of the names of the postdocs and students who created, over the past 7 years, the various components and ideas that made it possible.
In the interaction trap, the protein of interest, referred to as the "bait", is expressed as a LexA fusion in a "selection strain", a yeast strain containing LexA binding sites ("operators") upstream of a selectable marker gene, LEU2. Expression of a cDNA library with the cDNA-encoded proteins fused to a transcription activation domain allows selection of those that interact with the bait because they activate transcription of LEU2. Yeast containing cDNAs that encode bait interactors grow on media lacking leucine. The yeast strains used also contain one member of a series of different LexA operator-lacZ reporters. Thus, growth of a cell that contains a library plasmid into a colony in the absence of leucine indicates an interaction, and the amount of beta-galactosidase activity produced points to the strength of the interaction.
The following descriptions of the interaction trap are based on our experience with several successful hunts for interacting proteins. We are always tinkering with the interaction technology in response to results from here and from the outside world. We will attempt to update this document frequently with significant improvements as these are made.
Please contact us with any comments or questions about this information. Please also feel free to use and distribute these documents as you wish, as long as you do not change the wording, and keep the names of the authors attached to the information that they contributed..