Isolating Interaction-Null/Impaired Mutants Using the Yeast Two-Hybrid Assay

Interactions between biomolecules are the most basic part of biological phenomena. Protein-protein interactions constitute a significant part of such interactions. Therefore, identification of the interaction partner(s) of a protein of interest is critical to further elucidate the function of the protein/gene of interest. The yeast two-hybrid (Y2H) method is a popular technique to identify protein-protein interactions in vivo¹. In this system, two proteins (X and Y) whose interaction is to be tested are fused to the DNA-binding (DB) domain and transcriptional activation domain (AD), respectively. The DB-X fusion protein binds to a recognition sequence of the DB domain; therefore, when proteins X and Y interact, the AD-Y fusion protein comes into the proximity of the recognition sequence. Consequently, transcription of the reporter gene downstream of the recognition sequence is activated. Therefore, the presence or absence of reporter gene activity can be used to determine the presence or absence of the protein-protein interaction¹.

Once a specific interaction partner of the protein of interest is identified, further analyses should be performed to elucidate the biological function of the interaction. For this purpose, if mutants of the proteins that impair or remove the specific protein-protein interaction can be isolated, they will serve as powerful tools. The Y2H system can be used directly to isolate such mutants by screening 'interaction-negative' clones, starting with the wild-type 'interaction-positive' clone. To accelerate this process, 'reverse' Y2H (rY2H) systems were developed^2,3. In rY2H systems, the host yeast strains harbor counter-selectable marker genes as reporter genes, meaning yeast cells grow only when the AD-Y and DB-X proteins do not interact.

Although both the Y2H and rY2H systems allow the isolation of interaction-negative mutants, the process of isolating the mutants is laborious because not all of the candidates obtained by screening carry the desired type of mutations (usually missense mutations). The most serious issue is that a significant fraction of candidates harbor frameshift or nonsense mutations, and it is necessary to perform western blotting to exclude undesired clones. To overcome this problem, new plasmid vectors have been developed⁴. In these vectors, KanMX, a drug resistance marker, is positioned out-of-frame downstream of the DB domain or AD. The marker gene becomes in-frame with the DB domain or AD only when the gene of interest is inserted. When a random mutation(s) is introduced in the gene of interest, undesirable mutants, such as those with frameshift or nonsense mutations, can be easily eliminated by performing drug resistance selection, and candidates carrying desirable missense mutations can be easily identified with the Y2H screen⁴. This article presents a protocol to isolate interaction-null/impaired mutants of a protein of interest using this strategy.

1. Construction of the mutant library

1. Set up the polymerase chain reaction (PCR) (an example is shown below). Generally, prepare 50 µL of the reaction, which is divided into ten aliquots of 5 µL, and amplify the target gene fragment in a PCR machine⁴.
   NOTE: Users should select an appropriate polymerase to efficiently introduce mutations. If the DNA fragment to be amplified is short, then a polymerase with a higher error rate, such as Taq polymerase, which lacks proofreading activity, is needed. If the DNA fragment to be amplified is long, then a higher fidelity polymerase is needed to reduce the number of mutations. Mutations occur every few hundred base pairs after 30 cycles of amplification with regular Taq polymerase⁴. In the Representative results, a different Taq polymerase was used (see Table of Materials), which has higher fidelity than regular Taq polymerase, and the mutation rate was one in 600 base pairs. An example of the PCR mixtures, the primer details, and the reaction conditions are provided in Supplementary File 1.
2. When the PCR is completed, combine all the aliquots of the reaction, confirm that the DNA fragments of interest have been amplified using agarose gel electrophoresis, etc., and purify the DNA⁵.
3. Assemble the DNA fragment generated in step 1.2 with the appropriate Y2H vector using a 'seamless' cloning strategy (Figure 1).
   NOTE: The seamless cloning system, such as Gibson assembly⁶, minimizes contamination of the constructed mutant library by empty vectors generated via self-ligation. A list of Y2H vectors is provided in Table 1. They can be prepared by BamHI digestion before assembly. All plasmids are available from the National BioResource Project - yeast (see Table of Materials).
4. Introduce the reaction mixture generated in step 1.3 into Escherichia coli competent cells prepared according to a highly efficient E. coli transformation protocol (e.g., Inoue et al.⁷). Spread all competent cells on several LB plates (1% tryptone, 0.5% yeast extract, 1% NaCl, and 2% agar) containing appropriate antibiotics (see Table of Materials). At this point, spread a small amount (~1/100 of the total reaction) onto the same selection plate to calculate the library titer. Incubate the plates at 37 °C overnight.
5. Once a large number of E. coli colonies have appeared, scrape all cells from the surface of the plate using a disposable plastic loop. Recover plasmid DNA from these cells using popular methods, such as alkaline lysis⁸. At the same time, count the number of colonies that appear after spreading small aliquots of the transformation mixture on the plate. Using this number, estimate the total number of independent clones in the library.
   NOTE: At this point, pick up a few independent colonies (4-6) that appear on the plate on which the small aliquots of transformation reaction were spread, culture them separately, and isolate plasmids from them to confirm that virtually all clones carry the desired DNA fragments⁵. Many commercial kits are available to isolate plasmids from E. coli. The number of colonies that appear on the plate after spreading small aliquots can be counted manually.
6. Measure the concentration of the library DNA pool. Usually, a few hundred nanograms of library DNA is sufficient to obtain several thousand transformants in the next step.
   NOTE: It is recommended to measure the DNA concentration using a fluorescent dye that binds to DNA because measurement of A₂₆₀ is frequently inaccurate.

2. Transformation of yeast with the mutant library and replica plating

1. Introduce the mutant library into the host yeast strain (TAT-7 (L40-ura3)^9,10 MATa, leu2-3,112, trp1-901, his3-Δ200, ade2-101, gal80Δ, LYS2:(lexAop)₄-HIS3, ura3:(lexAop)₈-lacZ) for the Y2H assay using about 100 ng of DNA. Pre-transform the host cells with the 'bait' plasmid.
   1. After the transformation procedure, suspend yeast cells in sterile water and spread aliquots of different amounts on appropriate plates (usually SC medium lacking leucine and tryptophan: SC-LW [0.67% yeast nitrogen base, 2% glucose, 1.546 g/L SC double drop-out mix -Leu -Trp, and 2% agar, see Table of Materials]). Incubate these plates overnight at 30 °C.
      NOTE: The standard protocol, such as the lithium acetate-PEG method¹¹ with ~5 × 10⁶ logarithmically growing cells, is sufficient for the transformation of yeast. Another method with a high transformation efficiency, such as electroporation, can also be used.
2. The next day, when tiny colonies appear, select the plates with an appropriate number of cells (500-2000) and make replicates from them as follows.
3. Set two sheets of filter paper on a replica block to make several replicas (medium is SC-LW and SC-LW containing kanamycin) (see Table of Materials). Incubate at 30 °C for 1-2 days.
   NOTE: The concentration of kanamycin (G418) is 600 µg/mL. Do not use velvet for replica plating because the resolution of colonies will be lost.
4. Once colonies have grown, compare plates and pick candidates. Perform a Y2H color assay (see step 3) if lacZ is used as the reporter.
   NOTE: For the Y2H reporter, lacZ is usually better at detecting a weak interaction than HIS3.

3. Y2H color assay

1. Place a sheet of filter paper cut in advance to the appropriate size on the replica plate prepared using step 2. Be careful not to allow air bubbles to form between the filter paper and the plate.
   NOTE: Use No. 4A filter paper or Grade 50 filter paper (see Table of Materials). Any SC-LW or SC-LW containing kanamycin plates, which are made by replica plating, can be used for the color assay.
2. When the filter paper on the plate is completely moistened (when the color of the filter paper changes completely), place several sheets of paper towel on top and make it come into contact with the filter paper to remove excess moisture. Remove the paper towel when it absorbs the water and changes color. Repeat this process two or three times.
3. Pick up the edges of the filter paper with tweezers, peel it off the plate quickly, immerse it in liquid nitrogen, and freeze it. If liquid nitrogen is not available, place the filter paper on a plastic tray with the colony side up and immediately place it in a freezer (-20 °C to -80 °C) to freeze completely.
4. Remove the frozen filter paper and place it, with the colony side up, on a dry paper towel to thaw. Immediately after thawing, place the filter paper, with the colony side up, on another filter paper that has been placed on a plastic tray or sealable container and soaked with Z-buffer (60 mM Na₂HPO₄, 40 mM NaH₂PO₄, 10 mM KCl, and 1 mM MgSO₄, pH 7.0) containing X-gal (5-bromo-5-chloro-3-indolyl-β-D-galactoside) and 2-mercaptoethanol⁹ (see Table of Materials). Be careful not to allow air bubbles to form between the top and bottom filter papers.
5. Cover the plastic tray containing the filter paper and put it into an airtight container or plastic bag. Close the airtight container or plastic bag and incubate it at 37 °C until a blue color appears.
6. When a blue color appears, place the filter paper with the colony side up on about 500 µL of stop solution (1 M Na₂CO₃) placed on a clean plastic tray. Stop solution soaks in quickly; therefore, remove the filter immediately, place it on a fresh paper towel with the colony side up, and allow it to dry.
7. After the paper towel is replaced and the filter is dry enough, use a scanner or other device to capture the data.

4. Recovery and confirmation of candidate clones

1. Select white and Kan^R candidate clones from the replica's original plate (or the replicated plate not used for the color assay). Examples are shown in Figure 1C. Confirm that candidate clones are white and Kan^R by re-streaking them as a small patch on SC-LW medium containing kanamycin and incubating them at 30 °C for 1-2 days. Make at least two sets or make replicates the next day.
   NOTE: White colonies should be selected because pale blue colonies might retain the interaction.
2. Once candidate clones have re-grown well, repeat the color assay with at least one plate generated in step 4.1 as described in step 3 and confirm that they remain white and do not turn blue (Figure 2A).
   NOTE: When re-streaking the candidates, do not carry over large amounts of cells. Too many cells make it impossible to distinguish Kan^R and Kan^S clones.
3. Take clones that are still white and Kan^R generated in step 4.2 from the plate remains and grow them independently in 1 mL of selection medium (SC-L or SC-W, depending on which vector is used for the library construction) overnight at 30 °C.
4. Transfer the culture to a microtube and collect cells by centrifugation (~15,000 × g, for 10-30 seconds at room temperature). Isolate whole DNA from the cell pellet as follows.
5. Suspend the cell pellet in 400 µL of lysis buffer (2% Triton X-100, 1% SDS, 100 mM NaCl, 10 mM Tris-Cl (8.0), and 1 mM EDTA) containing 1 µL each of 2-mercaptoethanol and 20 mg/mL Zymolyase 100T (see Table of Materials), and incubate at 37 °C for 30 min to digest cell walls. At this time, mix gently by inverting the tube every 5-10 min.
6. When the cell suspension has become opaque or translucent, add an equal volume of a mixture of phenol-chloroform-isoamyl alcohol (25:24:1) and mix well by vortexing at moderate speed for 10-20 s.
7. Centrifuge the microtubes in a microcentrifuge for 5 min at full speed (~15,000 × g, at room temperature) and recover the aqueous phase containing DNA in a new microtube. Add 0.8 volume of isopropanol and mix well by inverting the tube.
8. Leave the tube at room temperature for 2-3 min and then centrifuge in a microcentrifuge for 4-5 min at full speed (~15,000 × g, at room temperature) to precipitate the DNA.
9. Remove the supernatant and wash the precipitate with about 500 µL of 70% ethanol. Remove ethanol completely and dry the precipitate briefly.
10. Add 20 µL of TE buffer (10 mM Tris-Cl (8.0) and 1 mM EDTA) to the precipitate, mix briefly, and leave the tube overnight at room temperature to dissolve the precipitate.
    NOTE: If users are in a hurry, keep tubes at 37-50 °C. The DNA precipitate will dissolve in a few hours.
11. Once the pellet is completely dissolved, transform E. coli with a small amount of this DNA solution.
    NOTE: About 0.5 µL of DNA solution is sufficient if E. coli with a high transformation efficiency is used.
12. When E. coli colonies appear, pick several independent colonies, culture them, and recover plasmids by standard methods, such as alkaline lysis.
13. Introduce the plasmid DNA obtained in step 4.12 into Y2H host cells harboring the bait plasmid. When transformants appear, check them by the color assay (they should appear white) and western blotting⁴ (they should express a protein of the desired size).
    NOTE: The post-alkaline method¹² is an easy and rapid way to prepare a whole-cell extract from yeast.
14. If the above two criteria are met, determine the mutation site of the clones by nucleotide sequencing⁴.

Recently, it was found that the C-terminal half of Pol2 protein (Pol2-C) interacts with Mcm10. Both proteins are essential for the initiation of DNA replication and, hence, for cell growth in the budding yeast Saccharomyces cerevisiae^13,14,15. To help understand the biological significance of this interaction, Pol2-C mutants that have no/diminished interaction with Mcm10 were isolated using the method described here.

A mutation library was constructed following the method described in step 1. Pol2-C DNA fragments were amplified with GoTaq polymerase and cloned into the Gal4AD (pST2525) vector using the In-Fusion enzyme. The number of independent clones in the library was estimated to be 5000.

As described in step 2, DNA of the library plasmid was introduced into the Y2H host strain TAT-7, which was pre-transformed with the LexA-Mcm10 plasmid. Cells were spread on an SC-LW plate and replicated to SC-LW and SC-LW+G418 plates the following day. The replicates were subjected to the color assay as described in step 3. Some of the results of the color assay are shown in Figure 1C.

Forty-seven colonies that were white in the color assay and G418-resistant were isolated as the first candidates from about 8500 transformants. They were tested again with the color assay, and 25 of the 47 clones were confirmed to be white (Figure 2A). These 25 clones were retained as candidates. Candidate plasmid DNAs were recovered from each of the 25 candidates as described in step 4. They were re-introduced into Y2H yeast host cells harboring LexA-Mcm10 and tested with the color assay, and 16 clones that lacked color were selected. To further confirm that these were Pol2-C missense mutants, the expression of Pol2-C protein was confirmed by western blotting. Twelve candidates exhibited a band at a position that was almost the same as that of the positive control (Gal4AD-HA-Pol2-C-KanMX fusion protein, calculated m.w.: 158.8 kDa) (Figure 2B). The color assay showed that these 12 clones did not interact with Mcm10 (Figure 2C). After determining the nucleotide sequences of these clones, it was confirmed that all of them had a missense mutation(s). The number of missense mutations varied from one to five (data not shown). On average, about one missense mutation occurred in every 1000 bases.

Figure 1: Schematics of the Y2H-based approach for screening interaction-null/impaired mutants. (A) The Y2H system. (B) Strategy for isolating interaction-null/impaired mutants. 1: The newly constructed Y2H vector contains a copy of the KanMX gene downstream of the Y2H tag, DB domain, and AD. Importantly, this KanMX gene lacks a start codon and is out-of-frame with the Y2H tag. Therefore, the KanMX gene is not expected to be expressed from this vector. When the PCR-amplified DNA fragment is inserted into the vector, the KanMX gene is in-frame with the Y2H tag and expressed. Consequently, only plasmids harboring the wild-type insert or an insert with a missense mutation(s) can express the KanMX gene, which confers resistance to G418. Plasmids with a nonsense or frameshift mutation(s) cannot express the KanMX gene. 2: The mutant library constructed in step 1 is introduced into the Y2H host yeast cells. When transformants appear, replicas are made. By comparing the expression of a reporter gene(s) and resistance to G418, candidates of interaction-null/impaired mutants can be selected. (C) An example of screening. The plasmid library, which harbored Gal4-AD and a PCR-mutagenized Pol2-C DNA fragment, was introduced into the Y2H host strain TAT-7, which was pre-transformed with the LexA-Mcm10 plasmid. Images of cells before (left) and after (middle) the color assay and of cells grown on an SC-LW+G418 plate (right) are shown. Examples of candidates of interaction-null/impaired mutants and false candidates are indicated by white and black arrowheads, respectively. (D) DNA sequence around the cloning site of Y2H vectors, AD (top), and DB (bottom) vectors. The sequence from the last ten amino acids (aa) of the Y2H tag (red) to the portion corresponding to the first ten aa of KanMX (green) is shown. Recognition sites of SmaI and BamHI are also shown. The positions of PCR primers are shown at the bottom when the BamHI site is used as the cloning site. The same primers apply to clone the gene of interest into the other plasmids (pST2303/2523). This figure has been modified from Tanaka et al.⁴. Please click here to view a larger version of this figure.

Figure 2: An example of the screening process of interaction-null/impaired mutants. (A) The 47 colonies isolated as candidate clones, as shown in Figure 1C, were again grown on SC-LW medium containing G418 and subjected to the color assay. +: positive control (Wt Pol2-C), -: negative control (vector). Candidate clones numbered 1-25 were cultured to recover plasmids. (B) Plasmids were recovered from each candidate clone in (A), introduced into Y2H host cells, and again subjected to the color assay. Sixteen of them were white (lacked color). Whole-cell extracts were prepared from them, and western blotting with an anti-HA monoclonal antibody was performed. The number on the panel corresponds to the number in (A). Analyses were performed of several independent plasmid clones recovered from each candidate except #6. (C) Results of the color assay for the final 12 clones. Numbers correspond to those in (A). #16, which retained the interaction with Mcm10, was used as the positive control in the color assay. Please click here to view a larger version of this figure.

Table 1: List of Y2H vectors containing KanMX out-of-frame with the Y2H tag.

Supplementary File 1: An example of the PCR mixtures, the primer details, and the reaction conditions. Please click here to download this File.

This article describes how to isolate interaction-null/impaired mutants using the Y2H assay. Such mutants are powerful tools to analyze the function of a protein of interest. To isolate such mutants, rY2H assays were developed previously by modifying the Y2H host strain^2,3. However, they have not greatly reduced the amount of labor. By contrast, mutants can be isolated with this method without a significant amount of labor. In this method, modification of the Y2H vectors ensures that all candidates are wild-type or missense mutants, in theory. In addition, this method can efficiently eliminate empty vectors from the candidates. Because of these desirable characteristics, this method may become a standard way to isolate interaction-null/impaired mutants.

Although the methods are simple and straightforward, there are a few critical steps to obtain good results: replica plating (step 2.3) and colony lift for the color assay (steps 3.2 and 3.3). Therefore, these steps are shown in the video. For step 2.3, users must use appropriate filter paper and not traditional velvet, as detailed in the NOTE of this step, to obtain discrete and high-resolution replicas. For steps 3.2 and 3.3, users should prepare at least two replica plates as a backup because sometimes colonies cannot be fully lifted. If the lift is incomplete with the first plate, users should repeat the removal of excess moisture with a paper towel from the backside of the filter paper one extra time with the second plate.

In the previous work, Y2H vectors, pST2302 (AD plus KanMX, which is out-of-frame) and pST2303 (DB plus KanMX, which is out-of-frame) were constructed to isolate interaction-null/impaired mutants efficiently⁴. The leakiness of the KanMX gene differs in these plasmids; cells harboring pST2302 grew on a G418-containing plate even without the inserted fragment. DNA sequences around the cloning sites are the same in the two plasmids (Figure 1D). Therefore, the reason for this difference in G418 resistance is unknown. Due to this difference, the leak-free vector, pST2303, was used for screening in the previous analysis. At that time, western blotting showed that all eight candidate clones expressed proteins with sizes consistent with that of the full-length protein. In the analysis shown here, the AD-Pol2-C construct was desirable; therefore, the leaky AD vector pST2525 (a variant of pST2302, but identical in sequence from AD to KanMX) was used. When protein expression of the 16 candidate clones was tested by western blotting, 12 clones expressed proteins with sizes consistent with that of the full-length protein. These results indicate that candidate clones can be efficiently obtained even with leaky vectors. Thus, while leak-free DB vectors are optimal for screening, AD vectors can also be used without a significant increase in labor.

The vectors used here contain the KanMX gene for drug resistance selection to ensure the mutation(s) is missense. AD vectors have leaked as described above. Although they are acceptable to screen mutants, as shown here, a leak-free vector is obviously better. In the currently used vectors, no initiation codon, ATG, resides upstream of KanMX in the same reading frame. Therefore, translation of the KanMX gene, which is responsible for the leak, might start from the ATG codon located in the KanMX ORF. Therefore, a plasmid was constructed in which the first ATG codon to appear in the KanMX ORF was mutated to ATC, causing the M18I substitution, but this did not affect KanMX leakage (data not shown). The next ATG codon in the KanMX ORF is the 68th methionine, and mutation of this codon may suppress leakage. Another way to construct leak-free vectors might be to change drug resistance markers. Selection by hygromycin and nourseothricin resistance genes is popular in yeast; therefore, the use of these genes instead of KanMX may be a good option in the future.

In the screening shown here, the Pol2-C fragment was amplified using GoTaq polymerase, which should have higher fidelity than wild-type Taq polymerase. Sequencing of the 12 mutants finally obtained showed that one missense mutation occurred in about 1000 bases on average. In addition to missense mutations, several silent mutations occurred, resulting in 59 base substitutions in the 12 clones, with a final mutation rate of approximately one in 600 base pairs. The amplified Pol2-C fragment was 2868 bp; therefore, there were more than four mutations in each clone on average, which is not an ideal mutation rate. Therefore, if users want to obtain mutants very efficiently, selecting an appropriate PCR enzyme is crucial by considering the lengths of DNA fragments to be amplified.