2, Massachusetts General Hospital, Charlestown, MA, USA
Counter-selectable markers can be used in two-hybrid systems to search libraries for a protein or compound that interferes with a macromolecular interaction or to identify macromolecules from a population that cannot mediate a particular interaction. In this report, we describe the adaptation of the yeast URA3/5-FOA counter-selection system for use in bacterial interaction trap experiments. Two different URA3 reporter systems were developed that allow robust counter-selection: (i) a single copy F′ episome reporter and (ii) a co-cistronic HIS3-URA3 reporter vector. The HIS3-URA3 reporter can be used for either positive or negative selections in appropriate bacterial strains. These reagents extend the utility of the bacterial two-hybrid system as an alternative to its yeast-based counterpart.
Many in vivo interaction trap systems have been developed for the identification of protein-protein, protein-DNA, and protein-RNA interactions. These methods typically involve the use of a “bait” (a specific protein, DNA, or RNA sequence) to identify an interacting partner from a library of different “prey.” These experiments can take the form of one-hybrid (1,2), two-hybrid (3), or three-hybrid (4) selections and can be performed in yeast (3,5) or bacteria (6,7,8,9). Interaction trap systems were originally developed for use in Saccharomyces cerevisiae, with interacting partners identified by positive selection. Positive selection is typically achieved by inactivating a gene, such as HIS3, required for survival under certain specific growth conditions, and then introducing a reporter system that drives expression of a functional copy of the deleted gene if a desired interaction occurs between the bait and prey (1). For HIS3, the stringency of the selection can be manipulated by varying the concentration of 3-amino triazole (3-AT), a competitive inhibitor of HIS3, in the media.
The URA3 gene (orotidine-5′ -phosphate decarboxylase) was originally developed as a counter-selectable marker for yeast (10,11). The endogenous URA3 gene can be inactivated, and a reporter cassette introduced in which expression of URA3 is dependent on the presence of a bait/prey interaction. If URA3 is expressed, the uracil biosynthesis pathway will be functional, and 5-fluoro-orotic acid (5-FOA), which is added to the media, is metabolized into a suicide substrate for the essential thymidylate synthase enzyme. URA3/5-FOA systems have been used successfully in yeast interaction trap systems to counter-select against bait/prey interactions in vivo (5,12). The stringency of this counter-selection can be titrated by varying expression of the URA3 gene; higher levels of URA3 expression increase the sensitivity of a cell to 5-FOA (12).
pyrF is the Escherichia coli homolog of URA3, and deletion of pyrF can also be complemented by expression of the yeast URA3 gene (13,14). The URA3/5-FOA system has been used to select against the expression of a functional URA3 gene in bacteria (14), but it has not been employed in the context of an interaction trap selection system in bacteria. For certain applications, bacteria provide a superior in vivo selection system to yeast (7,15,16). The transformation efficiency of bacteria is approximately 1000 times greater than yeast, which permits the interrogation of much larger libraries. Furthermore, fundamental differences in mechanism between prokaryotic and eukaryotic transcriptional regulation reduce the possibility of interactions between baits or preys of eukaryotic origin and components of the bacterial transcriptional system that could lead to constitutive activation or repression of the reporter gene.
A counter-selection system for bacteria would provide a mechanism for eliminating self-activating baits/preys from libraries (5,17) and would facilitate the identification of mutations that inhibit a protein-protein, protein-DNA, or protein-RNA interaction, thereby permitting the mapping of important functional residues at a recognition interface (5,12). Given the potential utility of a counter-selectable marker in bacteria, we developed a URA3 reporter system for bacterial-based interaction trap systems ((Figure 1)). This counter-selectable marker eliminates functional bait-prey interactions with a low breakthrough rate. We also constructed a tandem HIS3-URA3 reporter system that provides the flexibility of performing positive or negative selections using a single reporter construct.Figure 1.
The pyrF gene was deleted from two different E. coli strains, KJ1C and US0, which bear a preexisting deletion in the hisB gene (Reference 7, and J. Hurt, J.K. Joung, and C.O. Pabo, unpublished data). The US0 strain is isogenic to KJ1C, except that it lacks a Tn10 insertion positioned near the lac operon deletion present in both strains. The pyrF gene was inactivated in both strains using a λ red recombination system (18). Primers 5′-GCCCATCATCAAGAAGGTCTGGTCATGACGTTAACTGCTTCATCTTCTTCCGTGTAGGCTGGAGCTGCTTCG-3′ and 5′-TCATGCACTCCGCTGTAAAGAGGCGTTGATCGCTTTCAGCGTCTGCGCTGGACATATGAATATCCTCCTTAG-3′ were used to amplify a region of the pKD4 recombination plasmid with about 50 bp of homology to pyrF (bolded letters in the primer sequence) at each end. The PCR product was digested with DpnI to remove any plasmid template, gel-purified, and transformed into electrocompetent KJ1C and US0 cells containing a λ red helper plasmid (pKD46). The transformants were recovered in SOB media with 10 mM L-arabinose and screened for recombination on kanamycin plates. Candidate insertions in the pyrF gene were confirmed by PCR using primers that amplify pyrF (5′-AGGGGAGAAATCGCAACTGT-3′, 5′-CTTATCGCCCTGGATTTCAA-3′). The kanamycin gene was removed from positive clones by FLP-mediated recombination by introducing plasmid pCP20. The scars remaining from the recombination process in the KJ1C and US0 strains were confirmed by sequencing the PCR-amplified region of the genome. The F′ episome from XL1-Blue (Stratagene, La Jolla, CA, USA), which contains lacIq was introduced into US0ΔhisBΔpyrF through conjugation followed by selection on minimal medium containing 0.1% histidine, 0.1 mM uracil, 10 µg/mL tetracycline, and 2.5 mM 5-FOA.