G196 epitope tag system: a novel monoclonal antibody, G196, recognizes the small, soluble peptide DLVPR with high affinity

The recognition specificity of monoclonal antibodies (mAbs) has made mAbs among the most frequently used tools in both basic science research and in clinical diagnosis and therapies. Precise determination of the epitope allows the development of epitope tag systems to be used with recombinant proteins for various purposes. Here we describe a new family of tag derived from the epitope recognized by a highly specific mAb G196. The minimal epitope was identified as the five amino acid sequence Asp-Leu-Val-Pro-Arg. Permutation analysis was used to characterize the binding requirements of mAb G196, and the variable regions of the mAb G196 were identified and structurally analyzed by X-ray crystallography. Isothermal titration calorimetry revealed the high affinity (Kd = 1.25 nM) of the mAb G196/G196-epitope peptide interaction, and G196-tag was used to detect several recombinant cytosolic and nuclear proteins in human and yeast cells. mAb G196 is valuable for developing a new peptide tagging system for cell biology and biochemistry research.

Precise determination of the G196 mAb epitope. To confirm that the G196 epitope is located on the C-terminal extension of the 4T-2 fragment, and to further determine the precise location of the G196 mAb epitope, double-stranded oligonucleotides encoding peptides containing the C-terminal extension were inserted into the BamHI/EcoRI restriction sites within the pGEX-6P-1 vector. Total protein lysates were prepared from Escherichia coli that had been transformed with individual GST fusion protein constructs. The lysates were separated using two SDS-PAGE gels: one gel was stained with Coomassie Brilliant Blue to determine the equivalency of expression and loading, and the other (1/10 sample volume applied) was examined by Western blotting with mAb G196. mAb G196 detected 6P-11, 6P-12, 6P-13, 6P-14, 6P-15, and 4T-2 as a positive control ( Fig. 1b and c), whereas mAb G196 did not react with 6P-16, 6P-17, 6P-19, or 6P-1 as a negative control (Fig. 1c). These results identified the minimal epitope as the five amino acid sequence DLVPR.
We conducted alanine scanning mutagenesis on the epitope to determine which amino acid residues were responsible for mAb recognition. mAb G196 detected 6P-27 (Pro to Ala at position 4). In contrast, G196 only faintly detected 6P-26 (Val to Ala at position 3) and did not detect 6P-24, 6P-25, 6P-28, or 6P-29 (Fig. 2a). These results clarified that the epitope contains four critical residues and one nonessential residue (Pro at position 4) under denaturing conditions. The G196 epitope harbors a negatively charged amino acid (Asp at position 1) and a positively charged amino acid (Arg at position 5) at opposite ends of the sequence. To investigate the contributions of the charged amino acid residues of the epitope to mAb binding, we substituted both residues with physicochemically similar amino acids (6P-30: Asp to Glu at position 1; 6P-31: Arg to Lys at position 5). Replacement of either residue did not salvage immunoreactivity (6P-30 or 6P-31, Fig. 2b). Western blot analysis revealed that these charged amino acids at opposite ends of the epitope are critical residues for G196 antibody binding under denaturing conditions.
To evaluate the importance of each amino acid residue of the DLVPR epitope for mAb recognition under non-denaturing conditions, we conducted permutation analysis using peptides coupled through their N-termini to biotin. Single positions of the underlined residues in the peptide SGSGSDLVPRG were substituted individually with the other 19 coded amino acids. The peptides were immobilized onto streptavidin-coated plates, incubated with mAb G196, washed, and then the bound antibody was detected using an HRP-labeled anti-mouse secondary antibody (Fig. 2c,d). The results showed that Asp at position 1 can be replaced by Glu and the flexible amino acids Gly and Ser, whereas Leu at position 2 can be changed to one of several hydrophobic amino acids (Ile, Met, Phe, Asn) but not to Val, and to the hydrophilic amino acid His. Val at position 3 can be exchanged with one of two hydrophobic amino acids (Ile, Ala), and also with the hydrophilic amino acid Thr. Pro at position 4 does not show interaction specificity, whereas Arg at the last position is the most specific: Arg can not be substituted with Lys.
Crystal Structure of G196 Fab. The structure of G196 Fab was determined as the antigen-free form at 2.0 Å resolution using the molecular replacement method. Data and refinement statistics are summarized in Table 1. The R and R free values of the final model of G196 Fab were 0.191 and 0.228, respectively. The Fab fragment displays a conventional immunoglobulin fold characterized by an anti-parallel β -sheet sandwich architecture, with an r.m.s.d. value of 1.76 Å to anti-phenobarbital mAb Fab structure (Protein Data Bank code 1IGY). Trp99 in complementarity-determining region (CDR)-H3 interacts with His35 in framework region (FR)-H2 through π-π aromatic -stacking and forms hydrophobic core with Tyr32 and Trp33 in CDR-H1, and Phe103 in CDR-H3 (Fig. 4). Asn-31 in CDR-H1, Asn52 and Asn55 in CDR-H2, are located at one side of the hydrophobic core.
G196-tag application for scientific research. Short peptides and paired mAbs specific for the sequences of interest are powerful and irreplaceable tools for scientific research. To ascertain the possible versatility of G196-tag as a versatile fusion tag, we generated several human and yeast expression constructs in which the G196-tag was fused to reporter proteins. First, we expressed FLAG-and HA-tagged Emerald GFP (hereafter abbreviated FHG) as a reporter protein, C-terminally fused with a G196-tag (6P12, SDLVPRGSP) in HeLa cells. As shown in Fig. 5a, reporter protein fused with the G196-tag (FHG-G196) was detected as a single band in cell extracts by Western blotting with both anti-FLAG and the G196 mAbs (lanes 2, 4), whereas, as expected, the reporter without G196-tag (FHG) was not detected by mAb G196 (lane 3). It is noteworthy that mAb G196 did not cross-react with cellular proteins, and reacted solely with the G196-tagged protein (lanes 3, 4). In addition, mAb G196 efficiently immunoprecipited reporter protein fused with the G196-tag (lane 10) from cell extracts, comparable to that of FLAG immunoprecipitates (lane 8), but did not immunoprecipitate reporter protein lacking the G196-tag (lane 9).
Second, we applied the G196 epitope tag system to immunofluorescence assays with HeLa cells expressing the G196-tagged reporter protein to evaluate its utility. We utilized the cancer-related transcription regulator protein NAC1 as a nuclear reporter protein 18,19 . mAb G196 successfully detected the nuclear reporter protein, as did the control polyclonal anti-GFP antibody (Fig. 5b).
Lastly, we applied the G196 epitope tag system to yeast cells. The Atf1/Pcr1 heterodimeric transcription factor binds the cyclic AMP-responsive element (CRE)-like hexanucleotide sequence 5′ -TGACGT-3′ 20 . We expressed C-terminally 3 × G196-tagged Atf1 as a reporter protein in fission yeast. Western blotting showed that mAb G196 detected the Atf1 protein in yeast cell extracts, and chromatin immunoprecipitation (ChIP) enriched the Atf1 protein at the promoter of the tdh1 gene, which harbors a CRE consensus site 21 (Fig. 5c). Furthermore, mAb G196 recognized G196-and GFP-tagged Atf1, as shown by immunofluorescent staining of yeast, comparable to that of a polyclonal anti-GFP antibody (Fig. 5d). These results indicate that the G196 epitope tag system is suitable for Western blotting, immunoprecipitation, ChIP, and immunofluorescence assay in both yeast and human.

Discussion
Here we describe the characterization of a new G196 epitope tag system exhibiting defined properties. The G196 epitope tag system was characterized using Western blotting, immunofluorescence, and immunoprecipitation using human cells (this study and ref. 22) and Western blotting, immunofluorescence, immunoprecipitation, and chromatin immunoprecipitation using yeast cells (this study and refs 23 and 24). The new G196 epitope tag system will thus be useful for a broad range of studies in cell biology and biochemistry.
The minimal epitope of the G196 mAb is the five amino acid sequence DLVPR. We typically add a glycine-serine linker sequence upstream and downstream of the minimal epitope to minimize the influence of the tag on the target protein and maximize its accessibility for antibody binding, and therefore we used the nine amino acid sequence GSDLVPRGS as the original G196-tag. mAb G196 detected both N-and C-terminally G196-tagged protein (this study and refs. 22 and 23). The Grand Average of Hydropathicity score of the peptide was − 0.444 (http://www.bioinformatics.org/sms2/protein_gravy.html), indicating that the peptide was hydrophilic; indeed, 1 mg of the synthetic peptide was dissolved in 1 mL of PBS and used for competitive elution of bound G196-tagged proteins from a G196 affinity column (data not shown).
A K d value of ~10 nM or less between the anti-tag mAb and the tag is desirable for rapid and robust purification of modest to low abundance proteins of interest, and such levels are common when the proteins are expressed at the endogenous level 25 . It is noteworthy that the K d between mAb G196 and the G196-tag peptide is 1.25 nM (Fig. 3), which is comparable to or higher than the affinity of the anti-FLAG M2 mAb/FLAG-tag interaction (K d = 3-28 nM, and typically considered high affinity) [26][27][28] . It was previously reported that the K d of the anti-c-Myc 9E10 mAb/ c-Myc-tag is 2.2-560 nM [28][29][30] . Although the affinity of the anti-FLAG M2 mAb/FLAG-tag interaction exhibited a low-nM K d as described above, the 3 × tag version was superior in Western blotting, with a sensitivity reportedly increased by over an order of magnitude 7,31 . In our hands, the 3 × G196-tag version exhibited higher affinity compared to G196-tag in yeast (this study and refs 23 and 24).
Western blotting is widely used to ascertain an antibody's specificity and is an appropriate first validation step. mAb G196 detected the G196-tagged reporter protein as a single band at the proper molecular weight by Western blotting in HeLa cells, a human cell line (Fig. 5a, lane 4). Given that decreasing the length of a peptide tag has the disadvantage of increasing its reactivity with endogenous proteins containing the same sequence, we were surprised to observe that mAb G196 did not (cross-)react with cellular proteins, and only reacted with G196-tagged protein in HeLa cells (Fig. 5a, lanes 3) and several other human cell lines (data not shown). Use of the G196 epitope DLVPR as a search sequence for scanning the human proteome, disregarding splice variants, with ScanProsite 32 results in 11 hits on all UniProtKB/Swiss-Prot database sequences (release 2016_08 of 07-Sep-16: 551987 entries) (Supplementary Table 1). It is probable that the transfected G196-tagged reporter protein was overexpressed at a higher level than these 11 proteins were endogenously expressed. Alternatively, these 11 proteins were weakly expressed or did not express in these cell lines. mAb G196 must be used with caution when G196-tagged proteins are immunoprecipitated in order to search for binding partners in human cells, or when G196-tagged proteins are being investigated in cells for the first time. Cross-reactions of specific cellular proteins with anti-tag mAbs are observed universally. For example, anti-FLAG M2 mAb showed non-specific Western blot bands in Indian mustard plant 33 , tobacco 34 and rat brain tissue 35 . ScanProsite is an informative tool for checking (cross-)reactions of specific cellular proteins with anti-tag mAbs, and is available free of charge at http://prosite. expasy.org/scanprosite/.
Although not described in detail in the literature, many researchers have noticed that anti-FLAG antibody M2 is inadequate for the immunofluorescence staining of cells to detect FLAG-tagged proteins expressed in the cell nucleus. Previously, we studied the subcellular localization of myocardin-related transcription factor (MRTF)-B, a shuttle protein between the nucleus and cytoplasm that is expressed in response to various stimuli. FLAG-tagged MRTF-B in the nuclei of A431 cells was detectable, but stained more faintly compared to FLAG-tagged MRTF-B in the cytoplasm reacted with anti-FLAG mAb M2. We then changed the epitope tag system from FLAG-tag to G196-tag and could detect G196-tagged MRTF-B in the nuclei as well as in the cytoplasm, thus allowing evaluation of the relative subcellular distribution of MRTF-B 22 . In this study, we clearly stained G196-tagged NAC1 and ATF1 in the nuclei of human and yeast cells, respectively (Fig. 5b,d), thus demonstrating the utility of the G196 epitope tag system for the immunofluorescence staining of cells to detect G196-tagged proteins expressed where I i is the intensity of an observation and < I > in the mean value for that reflection and the summations are over all reflections. free R-factor was calculated with 5% of the data.
where Fo and Fc are the observed and calculated structure factor amplitudes, respectively.
in the cell nucleus. It has been reported that Arg at mild pH (pH 3.5-4.4) can effectively dissociate FLAG-tagged proteins from an anti-FLAG M2 mAb affinity column for use in fusion protein purification technology 36 . Arg and Lys are abundant (comprising more than 20% of the amino acids) in histone proteins 37 and histones proteins have been estimated to represent as much as 55% of the nuclear proteins 38 . The high abundance of Arg and Lys in the cell nucleus might interfere with anti-FLAG mAb M2 recognizing FLAG-tagged nuclear proteins. Recently, a novel anti-FLAG mAb (2H8, mouse IgG 2a ) with a sensitivity higher than that of M2 was generated and shown to be suitable for immunofluorescence and immunocytochemistry studies 39 , but mAb 2H8 can recognize only amino-terminal FLAG tags. Although the structure of G196 Fab shown in this study does not contain the epitope peptide, it nonetheless shows remarkable features of the antigen binding site (Fig. 4). Tyr32 and Trp33 in CDR-H1, and Trp99 and Phe103 in CDR-H3, together form a hydrophobic core. Leu at position 2 and Val at position 3 of the G196 epitope might be recognized in the region. In particular, Leu can bind through stacking interaction with aromatic residues, given that it can be replaced with His or Phe (Fig. 2c,d). Arg at position 5 might interact with an acidic residue, such as Asp104 in CDR-H3. However, Asp104 is too close to the hydrophobic region for this interaction with Arg to occur, as demonstrated by the finding that the residue at position 4 in the epitope can be replaced with any other residue (Fig. 2c). Also, Arg can be replaced with His but not with Lys. Therefore, it is likely that Arg at position 5 is recognized by Tyr102 in CDR-H3 rather than by Asp104. On the other hand, the Asp residue at position 1 of the epitope can be replaced not only with Glu but also Ser and Gln. Asn31 in CDR-H1, and Asn52 and Asn55 in CDR-H2, are located at the side of the hydrophobic core and are good candidates for recognizing Asp. Structural analysis of G196 complexed with the epitope peptide will allow precise elucidation of this antibody's high affinity recognition mechanism for the epitope. Each epitope tag system, including G196 epitope tag system, has its own distinct advantages and drawbacks, and it is important to consider these advantages and drawbacks prior to final selection of the tag to be used. In  the future, we intend to apply the G196 epitope tag system 1) as a bispecific antibody for drug targeting and 2) to structural studies of membrane proteins, including G-protein-coupled receptors.
Plasmid construction. All short 4T-2 fragments were generated by annealing complementary oligos. The oligos used to synthesize the indicated fragments are summarized (Fasmac, Kanagawa, Japan) in Supplementary Table 2. Annealing reaction mixtures contained equimolar amounts of each oligo and the reactions were performed in a thermal cycler, with initial denaturation at 98 °C for 3 min, followed by a temperature decrease of 2 °C every 20 s down to 4 °C to allow the oligos to anneal slowly. The oligos contained the BamHI/EcoRI restriction sites and the 4T-2 fragments were cloned into these restriction sites within the pGEX-6P-1 vector (GE Healthcare, Buckinghamshire, England).
A fragment of the GST gene, encoding residues 2-212, was prepared by polymerase chain reaction (PCR) amplification using the Pfu polymerase (BioAcademia) and the pGEX-6P-1 plasmid as the template. The primers contained the BamHI and XhoI restriction sites (underlined) and their sequences were 5′ -GGGGATCCTCCCCTATACTAGGTTAT-3′ (GST-6P22-S) and 5′ -GGCTCGAGTTAACCACCAAACGTGGCTTG-3′ (GST-6P21-AS). The amplified fragments were digested with BamHI and XhoI, then cloned into the pET28a vector (Merck Millipore, Darmstadt, Germany) digested with the same two enzymes. BL21 (DE3) E. coli were transformed with the plasmids, and protein expression was induced with 0.1 mM isopropyl-β -D-1-thiogalactopyranoside (IPTG). The final constructs were sequenced to ensure that no mutations had occurred during the PCR and cloning processes. The expressed proteins were purified as described previously 40 . mAb generation. Mouse mAb G196 was generated by immunizing mice with GST protein bacterially expressed using the pGEX-2T vector, and serum titers were monitored by immunoblotting using the same GST protein. Clonal populations of fusion cells were screened by ELISA for antibody production against this GST protein. Productive cells were cloned to monoclonal lines by serial dilution screening. Highly concentrated mAbs were isolated from murine ascites after an intraperitoneal injection of hybridoma cells. All animal experiments were performed in compliance with the standards established by the International Guiding Principles for Biomedical Research Involving Animals and were approved by the animal study committee of Shimane University.
Cell culture and transfection. The HeLa human cervical epithelioid carcinoma cell line was purchased from the Japanese Collection of Research Bioresources (JCRB) Cell Bank (JCRB9004). HeLa cells and hybridomas were grown in DMEM (Nissui, Tokyo, Japan) and RPMI1640 medium, respectively, supplemented with 10% FBS (Sigma-Aldrich). HeLa cells were transiently transfected with each plasmid using Lipofectamine 2000 (ThermoFisher Scientific) according to the manufacturer's instructions.
Permutation ELISA analysis. The biotinylated 11-aa peptide (biotin-SGSGSDLVPRG) was synthesized by Mimotopes (Clayton, VIC, Australia). The sequence of the G196 epitope was DLVPR and the 5-aa oligopeptide of SGSGS was used as a spacer between biotin and the G196 epitope. Single positions of the peptide biotin-SGSGSDLVPRG (epitope sequence underlined) were substituted with the other 19 coded amino acids. The biotinylated peptides had > 70% purity as judged by high-performance liquid chromatography.
Permutation ELISA analysis was performed according to the manufacturer's instruction. Briefly, the biotinylated peptide was applied to NeutrAvidin-coated wells of a 96-well plate (15125, ThermoFisher Scientific) (5 μ g/ mL, 1 h, 37 °C). The plate was blocked (2% BSA in PBS, 1 h, 22 °C), then washed 4 times with PBST (PBS buffer containing 0.1% Tween 20). Subsequently, 100 μ L of a 1:2,000 dilution of purified mAb G196 (1 mg/mL) was added to the wells for 1 h at 22 °C. After washing 4 times with PBST, 100 μ L of HRP-conjugated goat F(ab') 2 anti-mouse antibody (1:5,000 diluted with 2% BSA/PBS) was added and the plate was incubated at 22 °C for 1 h. After washing 4 times with PBST and twice with PBS, 100 μ L of o-phenylenediamine dihydrochloride peroxidase substrate (P1526, Sigma-Aldrich) was added and the plate was incubated at 22 °C for 15 min. The reaction was stopped by the addition of 100 μ L of 2 M H 2 SO 4 , and the OD value at 492 nm was measured using a microplate reader.

Isothermal Titration Calorimetry (ITC). ITC was performed with a VP-ITC titration calorimeter
(MicroCal, GE Healthcare, Buckinghamshire, England) at 25 °C. Calorimetric measurements were carried out with purified G196 Fab and the synthetic peptide GSDLVPRGS (Tufts University Core Facility, Boston, MA, USA). The ITC cell (1.4 mL) contained 25 μ M antibody G196 Fab in 20 mM Tris-HCl (pH 8.0), 150 mM NaCl, and was stirred constantly at 307 rpm throughout the experiment. The peptide (500 μ M) in the same buffer was added using 28 injections: the first injection volume was 3 μ L and all subsequent injections were 10 μ L. The first data point was not used for fitting. Binding parameters were determined using the Origin software package provided with the instrument by fitting the data to a single-site model. Sequencing of the variable region of the mAb G196. Total RNA was extracted from 3 × 10 6 cells of the murine hybridoma clone G196-3 by using Sepasol-RNA I Super G (Nacalai Tesque, Kyoto, Japan). Total RNA was reverse transcribed into cDNA by using oligo-dT 14 and ReverTra Ace (mutated M-MLV reverse transcriptase, RNaseH negative) (TOYOBO, Osaka, Japan). DNA fragments encoding the heavy and light chain were PCR amplified using the Pfu polymerase (BioAcademia). The primers used, containing the BamHI and EcoRI restriction sites, for amplifying heavy chain genes were VH1-1S and IgG2-1AS; for light chain genes the primers were VK-1S and CK-2AS (Supplementary Table 2). The amplified fragments were digested with BamHI and EcoRI and cloned into the pBlueScript II SK(+ ) vector (Agilent Technologies, Santa Clara, CA, USA) digested with the same two enzymes. The DNA sequences of the heavy and light chain genes were determined by automated sequencing using T7 and T3 sequencing primers with BigDye sequencing reagent (ThermoFisher Scientific) and an ABI 3100 automated capillary DNA sequencer (ThermoFisher Scientific).
The isotype of mAb G196 was identified as mouse IgG 1 using an IsoStrip mouse monoclonal antibody isotyping kit (Roche Diagnostics, Basel, Switzerland).
Purification of Fab fragments from mAb G196. mAb G196 IgG was purified from ascites of hybridoma-bearing mice as follows. The ascites were diluted 1:40 with 20 mM Tris-HCl (pH 8.0), 150 mM NaCl, and were purified by passage though four chromatography columns (HiTrap Q HP, GE Healthcare; hydroxyapatite, Bio-Rad Laboratories, Hercules, CA, USA; HiTrap SP HP, GE Healthcare; HiTrap Heparin HP, GE Healthcare). The flow-through was precipitated using ammonium sulfate. The resulting precipitate was suspended in 1/100 of the starting volume with 20 mM sodium phosphate (pH 7.0), and was dialyzed against 20 mM Tris-HCl (pH 8.0). The material was further fractionated by FPLC on a HiTrap Q HP column (GE Healthcare). Proteins were eluted using a stepwise NaCl gradient, with the final concentration being 150 mM NaCl. Purified antibody was fragmented using papain, and the resulting Fab fragments were purified by protein A chromatography (Pierce Fab preparation kit, ThermoFisher Scientific). The purity and homogeneity of the fragment were evaluated by means of SDS-PAGE followed by Coomassie Brilliant Blue staining.
Crystallization and Structure Determination. mAb G196 Fab was crystallized using the hanging drop vapor diffusion method by mixing 1 μ L sample solution (8 mg/mL) and 1 μ L reservoir solution (100 mM sodium citrate (pH 5.0), 20% PEG 20000) equilibrated against 1 mL reservoir solution. The crystals grew at 20 °C in 15% PEG 20000, 100 mM sodium citrate (pH 5.0) over a period of one week. The crystals were briefly soaked in a cryoprotectant solution consisting of the reservoir solution plus 20% (v/v) glycerol before cryocooling. X-ray data were collected using an ADSC Q270 CCD detector at beam-line BL17A of the Photon Factory, Tsukuba, Japan. The wavelength of the incident X-rays was 0.98 Å. Diffraction data sets were processed with HKL2000 41 and scaled with SCALEPACK 41 . The crystals belonged to the space group P212121, with unit cell parameters of a = 38.65 Å, b = 82.27 Å, c = 127.31 Å. The structures were solved by molecular replacement using PHASER 42 and the previously reported structures of the heavy and light chain of anti-(4-hydroxy-3-nitrophenyl)acetate and anti-phenobarbital mAbs (PDB codes 1NGQ and 1IGY, respectively) as starting models. COOT 43 and REFMAC5 44 were subsequently employed for iterative cycles of model building and refinement. The later stages of crystallographic refinement were carried out using PHENIX 45 . After an initial round of simulated annealing refinement, several macrocycles that included bulk solvent correction, anisotropic scaling of the data, individual coordinate refinement with minimization, and individual isotropic ADP (atomic displacement parameters) refinement, were carried out with maximum likelihood as the target function. In the course of the refinement, water molecules were added to the models by manual inspection of their positions in both the 2Fo -Fc and Fo -Fc maps, and combined TLS (translation, libration, and screw-rotation) and individual ADP refinement were carried out in the final stages. Manipulation of the model and validation were performed with COOT 43 . The stereochemistry of the final model was also assessed using PROCHECK 46 . Data collection and refinement statistics are summarized in Table 1.
The atomic coordinates and experimental structure factors of mAb G196 Fab have been deposited with the Protein Data Bank under code 5H2B.
Yeast cell manipulation. The fission yeast strains used are listed in Supplementary Table 3. Yeast extract medium (YES) was supplemented with 250 mg/L each of adenine, leucine and uracil. Atf1 and Iws1 were tagged by modifying the 2-step PCR method to create a fragment for gene targeting. The primers used for genetic manipulations are listed in Supplementary Table 2. C-terminal tagging of the atf1 gene ( Supplementary Fig. 1) was accomplished using the G196t2 primer. This primer contains a sequence homologous to a portion of the atf1 coding sequence, a sequence for 3 × G196-tag, and a sequence homologous to another primer, com5. In the first PCR step, two sequences flanking the translational stop codon of the atf1 gene were amplified with the primers t1 and G196t2, and with the primers t3 and t4, using the genomic DNA as a template. The GFP-kanMX cassette was also amplified with the primers com5 and com6, using pFA6a-GFP(S65T)-kanMX as a template. The three PCR products were fused together in the second PCR step with the primers t1 and t4 to produce a DNA fragment for homologous recombination.
N-terminal tagging of the iws1 gene ( Supplementary Fig. 2) was accomplished by amplifying the kanMX cassette from pFA6a-kanMX with the primers Nt3 and Nt4 in the first PCR step. The EmGFP coding sequence of EmGFP/pCDNA3.1-neo was amplified with the primers Nt7 and Nt8. We also amplified three sequences from the iws1 locus: one homologous to the upstream region of the iws1 promoter (upstream), one homologous to the promoter (promoter), and one homologous to a portion of the iws1 coding region (CDS). The three sequences were amplified as follows: upstream sequence using the primers Nt1 and Nt2, the promoter sequence using Nt5 and NT6, and the CDS sequence using Nt9 and Nt10. The Nt2, Nt5, Nt6 and Nt9 sequences contained a sequence homologous to Nt3, Nt4, Nt7 and Nt8, respectively. The five fragments produced in the first PCR step were fused together in the second PCR step with the primers Nt1 and Nt10 to produce a DNA fragment for homologous recombination.
Yeast transformation was conducted using a yeast transformation kit for Saccharomyces pombe (WAKO Pure Chemical Industries, Osaka, Japan) according to the manufacturer's instructions. G418 sulfate (Merck) was added to YES at 100 mg/L to select the recombinants. The genomic DNA of the strains was Sanger-sequenced to confirm that proper recombination had occurred and that no additional mutation had been introduced during the procedure.
Chromatin immunoprecipitation. Yeast cells were grown in YES medium at 30 °C to mid-log phase.
After formaldehyde crosslinking and quenching, 1 × 10 8 cells were harvested to obtain 2 mL of soluble input extract. M-280 anti-mouse sheep antibody-conjugated magnetic beads (112-02, ThermoFisher Scientific) were used for immunoprecipitation: 200 μ L of the beads were washed twice with extraction buffer (50 mM HEPES-KOH (pH 7.5), 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, and 0.1% sodium deoxycholate), and then incubated in 1 mL of extraction buffer containing 3 μ L of mAb G196 for 1 h at 4 °C. The beads were again washed twice with extraction buffer to remove unbound antibodies and incubated for 1 h with 800 μ L of input extract at 4 °C. The relative concentrations of the target sequences were determined using SYBR Premix ExTaq (RR041A, TaKaRa Bio, Shiga, Japan) and a Thermal Cycler Dice Real Time System TP800 (TaKaRa Bio), according to the manufacturer's instructions. The primers for detecting the − 1 kb region were 5′ -GAATCGATGCTCATTTGAGTC-3′ and 5′ -TACTCCACCATAGTTCTTGTC-3′ , and for the tdh1 CRE region the primers were 5′ -GTGCTAGCAATTCCTCCTTC-3′ and 5′ -TTCGTTGGTAATCTGCCTCG-3′ .
Immunofluorescence microscopy. HeLa cells were simultaneously fixed and permeabilized for 10 min with 3.7% formaldehyde and 0.2% Triton X-100 in PBS, washed 3 times with PBS, and then blocked for 30 min with 5% skim milk in PBS. Next, the cells were incubated with primary antibodies for 1 h at room temperature. After four washes with PBS, secondary antibody incubations were performed on coverslips for 1 h at room temperature, then the coverslips were mounted (mounting medium; Vector Laboratories, Burlingame, CA, USA) with DAPI.
Yeast cells exponentially growing in YES medium at 30 °C were fixed with 3% formaldehyde and incubated at 18 °C for 30 min. After quenching with 250 mM glycine, the cells were washed with PBSEMS (PBS buffer with 1 mM EGTA, 1 mM MgSO 4 and 1.2 M sorbitol), resuspended in PBSEMS containing 5 mg/mL Zymolyase 20T (07663-91, Nacalai Tesque) and 0.1% 2-mercaptoethanol, and incubated at 37 °C for 1 h. The cells were washed three times with PBSEMS and resuspended in PBSEMT (PBS with 1 mM EGTA, 1 mM MgSO 4 and 1% Triton X-100) and incubated at room temperature for 30 sec. After washing three times with PBSEM (PBS with 1 mM EGTA and 1 mM MgSO 4 ), the cells were incubated in PBSEMALM (PBS with 1 mM EGTA, 1 mM MgSO 4 , 0.1% sodium azide, 0.1 M L-lysine and 5% skim milk) for 30 min. The cells were resuspended in PBSEMALM containing G196 mAb (1:1000) and anti-GFP rabbit polyclonal antibody (1:1000). After overnight incubation at room temperature, cells were washed with PBSEMALM three times, then resuspended in PBSEMALM containing anti-mouse antibody conjugated with Alexa 594 (1:100) and anti-rabbit conjugated with Alexa 488 (1:100). After one-hour incubation, the cells were washed with PBSEMAL containing 5% skim milk three times and resuspended in PBS containing 0.1% sodium azide. The cell suspension was mixed with an equal amount of 1% low-melting point agarose with DAPI on glass plates.