INTERACTOR HUNT PROTOCOL The following procedure describes how to conduct an interactor hunt. For further information on yeast manipulation, see (REFERENCES). The procedure described here utilizes information gained from hunts conducted in the lab by Jeno Gyuris, Russ Finley, Tony Zervos, Andy Mendelsohn, Ze'ev Paroush, Debbie Goff, Steve Hanes, Barak Cohen, Dimitri Krainc, Lakshmi Raj, Claude Sardet, Pierre Leopold, and Herbert Altman. Interaction trap selection/screening protocol - Russ Finley September, 1992 Updated 5/93 Yeast transformation I currently favor using the protocol of Gietz et al., (1992, Nucleic Acids Research vol. 20, pg 1425) which in our hands now gives transformation efficiencies just under 10e5/ug DNA. To get this performance, it is imperative that you use properly prepared carrier DNA (sonicated, boiled salmon sperm DNA; see Schiestl and Gietz, 1989, Current Genetics 16:339-346). Library DNA should be introduced into yeast that already harbor the URA3+ reporter and HIS3+ bait plasmids. Thus, yeast should be grown in ura-his- glucose CM dropout medium before transformation to maintain selection for these plasmids. I perform transformations in several individual eppendorf tubes (one for each 24cm X 24cm transformation plate) to reduce the likelihood of contaminating the entire transformation. It is important not to use excess transforming DNA per aliquot of competent yeast, otherwise each competent cell will take up several plasmids. We have found that adding DMSO to 10% (of the volume) just prior to adding the PEG solution increases efficiencies several fold. Also, it improves efficiency to resuspend the transformed yeast in sterile water rather than TE before plating onto ura- his- trp- CM dropout glucose plates, or to plate PEG along with the yeast. It helps to make the plates with 25g agar/liter to make them extra hard. We use sterile glass beads (e.g. Fisher #11-312B, 4mm diameter) to help spread the transformation mixture on the plates. [Note: An alternative to the following protocol is to plate the transformation directly onto the selection plates (ura-his-trp-leu- gal/raff) after incubation in gal/raff liquid for a few hours to induce the GAL promoter. We are currently comparing these two methods.] To make it easier to determine the number of transformants, make dilutions (with water, not TE) from a few transformations and plate on normal sized ura- his- trp- CM dropout glucose plates. I strive for 200,000 transformants per 24cm X 24cm plate. This corresponds to 1ug per eppendorf tube if your transformation efficiency is 0.5 X 10e5 transformants per ug DNA. Harvest transformants Incubate the plates for about 48 hours at 30 C, or until colonies are about 1 mm in diameter. Cool plates at 4 C for a few hours to harden agar. Use a sterile glass microscope slide (and sterile technique) to scrape the yeast from the plate, trying not to scrape any agar. The yeast can be collected from the glass slide by wiping it on the lip of a sterile 50 ml Falcon tube. Wash the cells 2X with TE. It is best to pellet the cells each time in a sterile round bottom polypropylene tube at 2500 rpm for 4 min. so they may be easily resuspended. The pellet volume for 500,000 transformants will be about 8 ml. Resuspend the cells thoroughly in 1 pellet volume of glycerol solution (Current Protocols in Molecular Biology: 65% glycerol (vol/vol), 0.1 M MgSO4, 25 mM Tris pH 8.0). Freeze ~1ml aliquots at -70oC. Determine plating efficiency Remove an aliquot of transformed yeast and dilute 10-fold with ura-his-trp- CM dropout gal/raff medium (2% galactose, 1% raffinose; in this medium the yeast use raffinose as a carbon source but the galactose still induces transcription from the GAL1 promoter). Incubate, shaking, at 30 C for 4 hours to induce the GAL promoter. Make serial dilutions using the above medium and plate on ura-his-trp- gal/raff plates to determine the number of colony forming units per aliquot of transformed yeast (there should be ~10e8 cfu/100ul. Plate onto selection plates Induce the GAL promoter as above and plate < 10e6 colony-forming-units per 100 mm plate ura-his-trp-leu- CM dropout gal/raff. Pick Leu+ colonies and patch (or better yet, streak for single colonies) onto a new ura-his-trp-leu- CM dropout gal/raff plate. The goal here is to purify the Leu+ yeast away from the many leu- yeast on the original selection plate before putting them on a master plate without leu selection. In order to test for Gal inducibility of the Leu+ phenotype it is necessary to first turn off the GAL promoter by growing on glucose master plates. From the new ura-his-trp-leu- CM gal/raff plate patch to a ura-his-trp- CM dropout glu master plate; growth on this plate will shut off the GAL promoter. You can now replica from this glu master plate to four plates that will determine whether the Leu+ phenotype is galactose-dependent and whether it correlates with galactose-dependent beta-galactosidase activity. The four plates are 1. ura-his-trp- CM dropout glu X-GAL; 2. ura-his-trp- CM dropout gal/raff X-GAL; 3. ura-his-trp-leu- CM dropout glu; 4. ura-his-trp-leu- CM dropout gal/raff. It is important not to transfer too large a mass of yeast in the replica. Isolate library plasmid and test specificity Once you have isolated galactose-dependent LEU+ lacZ+ yeast, the next step is to formally prove that the phenotype is due to the library plasmid and to see that the interacting cDNA-encoded protein is specific for the bait. To do this we isolate the TRP1 library plasmid from the positive yeast and use it to re-transform the original bait strain, and other strains expressing unrelated baits (for example, the LexA-bicoid derivative in pRFHM1). If you have isolated a large number of positive yeast, it is useful to identify those that contain identical library plasmids so that the workload can be reduced. One quick way to determine which yeast have the same cDNA is to do a quick yeast miniprep, PCR the cDNA with primers derived from the vector pJG4-5, and cut the resulting PCR products with HaeIII and/or AluI. It is usually clear from this analysis which cDNAs are the same. We isolate the library plasmid by transforming E.coli strain KC8 (Gyruis et al, Cell, submitted, a gift of Kevin Struhl) with the yeast miniprep DNA and selecting for complementation of the E.coli trpC mutation by yeast TRP1. Russ Finley finley@opal.mgh.harvard.edu September 1992 updated May 1993