Transformation of S. cerevisiae was performed using the lithium acetate method, as previously described (Adams et al., 1997).

In all experiments other than the fluorescence microscopy, we used the congenic standard strains BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) and BY4742 (MATα his3Δ1 leu2Δ0 lys2Δ0 ura3Δ0) (Brachmann et al., 1998), along with and their derivatives. The kanMX4-based IRE1-knockout mutants of BY4741 and BY4742, named Y01907 and Y11907, respectively, were obtained from EUROSCARF.¹ The plasmid pRS313 is a centromeric low-copy S. cerevisiae vector carrying the HIS3 selectable marker (Sikorski and Hieter, 1989). We previously constructed the plasmid pRS313-IRE1 by inserting the IRE1 gene (coding region and 5′- and 3′-flanking regions) into pRS313 (Kimata et al., 2004). In this study, we used two pairs of IRE1 + and ire1Δ strains. One pair consisted of Y11907 transformed with either pRS313-IRE1 or pRS313. The remaining pair comprised BY4741 and Y01907.

To express green fluorescent protein (GFP)-tagged Ire1 under control of the TEF1 promoter, we used our laboratory strain W303-ire1Δ[pRS313-TEF1p-IRE1-GFP] (MATa leu2-3,112 trp1-1 can1-100 ura3-1 his3-11,15 ire1: TRP1 [pRS313-TEF1p-IRE1-GFP]) (Ishiwata-Kimata et al., 2013; Le et al., 2016). We did not use BY4741, 4,742, or their derivatives for fluorescence microscopy because they seemed to emit somewhat stronger autofluorescence than the W303-based strains.

The plasmid pCM189 is a centromeric S. cerevisiae vector used for the Tet-off system (Garí et al., 1997). We previously inserted the Hac1-coding sequence downstream of the doxycycline-inducible promoter in pCM189 to construct the plasmid pCM189-Hac1 (Ishiwata-Kimata et al., 2025).

The plasmid pML104 is a 2 μ-based multicopy S. cerevisiae vector used for CRISPR-Cas9 genome editing (Laughery et al., 2015). We modified this plasmid to insert a guide sequence targeting the HAC1 gene. The following is a partial sequence of the resulting plasmid pML104-HAC1:

gcagtgaaagataaatgatcTACGACAACAACCGCCACTAGTTTTAGAGCTAGaaatagcaagttaaaataag. The sequences derived from pML104 are shown in lowercase letters. The inserted sequence is in uppercase, and the guide sequence targeting HAC1 is indicated in boldface.

The HAC1 gene was genome-edited by transforming BY4741 with pML104-HAC1 and donor DNA synthesized by Eurofins Genomics (Tokyo, Japan). The resulting strain, YKY-HA-HAC1, carried an in-frame insertion of three tandem copies of the hemagglutinin (3 × HA) tag after the initiation codon of the HAC1 gene. The donor DNA sequence was as follows:

cacctcaatggacaactcgagaaatgaatacagaaatatgttttttagcgaaattttcctttcttcttgtcttcttgttttatttaaacttccaaggctttaactcagtgtcaaacataacaacctcctcctcccccacctacgacaacaaccgccactatgTACCCATACGATGTTCCTGACTATGCGGGCTATCCGTATGACGTCCCGGACTATG CAGGATCCTATCCATATGACGTTCCAGATTACGCTgaaatgactgattttgaactaactagtaattcgcaatcgaacctagctatccctaccaacttcaagtcgactctgcctccaaggaaaagagccaagacaaaagaggaaaaggaacagcgaaggatcgagcgtattttgagaaacagaagagctgctcaccagagcagagagaaaaaaagactacatctgcagtatctcgagagaaaaztgttctcttttggaaaatttactgaacagcgtcaaccttga. The sequences of the homology arms corresponding to the HAC1 5′-flanking and coding regions are shown in lowercase letters. The sequence corresponding to the 3 × HA tag is indicated in uppercase. The initiation codon of HAC1 is shown in boldface.

The 3 × HA sequence was also inserted into the same position of the Hac1 gene on pCM189-Hac1 to obtain the plasmid pCM189-HA-Hac1, which was used to express the HA-tagged Hac1 protein (HA-Hac1) under the control of the Tet-off promoter.

Saccharomyces cerevisiae cells were cultured at 30 °C with shaking in synthetic dextrose (SD) medium containing 2% glucose, 0.66% Difco yeast nitrogen base without amino acids (YNB w/o AA; Becton Dickinson, Franklin Lakes, NJ, United States), and appropriate auxotrophic supplements. Optical density of cultures at 600 nm (OD₆₀₀) was measured using a SmartSpec 3,000 spectrophotometer (Bio-Rad Laboratories, Inc., Hercules, CA, United States). The starting OD₆₀₀ of the cultures was approximately 0.2.

Tunicamycin was purchased from Sigma-Aldrich (Merck, Darmstadt, Germany) and prepared as a 2 mg/mL solution in dimethyl sulfoxide solution. Cycloheximide was purchased from Nacalai Tesque (Kyoto, Japan) and was prepared as a 20 mg/mL aqueous solution. DTT was purchased from Tokyo Chemical Industry (Tokyo, Japan) and prepared as a 1 M aqueous solution. Ethanol (99.5%) were purchased from Nacalai Tesque. These reagents were added to cultures during the fast-growing and exponential phase, which were further incubated at 30 °C with shaking before harvest.

Ethanol concentration in the cultures is expressed as volume/volume percentage (v/v). To stepwise increase the ethanol concentration up to 16%, we first added 0.155 mL of ethanol to 5.0 mL of cultures to reach 3% ethanol, followed by a 60-min incubation. Second, we added 0.165 mL of ethanol to reach 6% ethanol, again incubating for 60 min. Third, we added 0.120 mL of ethanol to reach 8% ethanol, again incubating for 60 min. Fourth, we added 0.120 mL of ethanol to reach 10% ethanol, again incubating for 60 min. Fifth, we added 0.120 mL of ethanol to reach 12% ethanol, again incubating for 60 min. Sixth, we added 0.135 mL of ethanol to reach 14% ethanol, again incubating for 60 min. Finally, after addition of 0.135 mL of ethanol, the cultures were incubated for certain durations, followed by various assays.

Based on the Difco manual (Difco Laboratories, 1984), we mixed pure chemicals to prepare YNB w/o AA lacking inositol, which was then used to make SD medium not containing inositol (SD(−inositol) medium). For inositol depletion, cells in the exponential phase grown in SD medium were harvested, washed six times with SD(−inositol) medium, and further cultured at 30 °C in SD(−inositol) medium.

For RNA and protein extraction, cells were harvested from the cultures at the OD₆₀₀ of approximately 1.0.

To assess cell survival, cultures were appropriately diluted in SD medium and spread onto SD agar plates, which were incubated at 30 °C for 3 days prior to colony counting. The survival ratio in the presence of tunicamycin was calculated as the ratio of colony-forming units (CFU) at a given time point to the CFU immediately before tunicamycin addition. Because cells grew even in the presence of tunicamycin, CFU values were normalized to the OD₆₀₀ of the cultures. The survival ratio in the presence of 16% ethanol was calculated as the ratio of the CFU at a given time point to the CFU immediately before the ethanol concentration reached 16%.

Total RNA was extracted from S. cerevisiae cells using the hot phenol method (Collart and Oliviero, 2001). Prior to mRNA-sequencing (mRNA-seq) analysis, residual DNA was removed from total RNA samples by DNase I treatment, as described in Ishiwata-Kimata et al. (2025). As previously described (Fauzee et al., 2023), the following procedure was performed by GeomeRead Co. Ltd. (Takamatsu, Japan) for the mRNA-seq analysis. First, total RNA samples were subjected to mRNA purification via the poly(A) method using the KAPA mRNA Capture Kit (KAPA Biosystems, Potters Bar, United Kingdom). Next, DNA libraries were prepared from the mRNA samples using the MGIEasy RNA Directional Library Prep Set (MGI Tech, Shenzhen, China). Finally, DNA sequencing was performed on the DNBSEQ-G400S platform using the DNBSEQ-G400RS High-throughput Sequencing Set (MGI Tech; 2 × 150 bp paired-end reads, 1 Gb data/sample). We then processed the raw FASTQ data using the CLC Genomics Workbench (Qiagen, Venlo, Netherlands). Genes with no or marginal expression (i.e., with the minimum transcript per million (TPM) value of 0.00, or the maximum TPM value less than 1.00) were excluded from further analysis.

To assess the expression levels of selected genes, we performed reverse transcription (RT)-quantitative PCR (qPCR). In accordance with the manufacturer’s instruction, cDNA was synthesized from total RNA samples using the poly(T) primer and ReverTra Ace qPCR RT Master Mix with gDNA Remover (Toyobo, Osaka, Japan). As previously described (Fauzee et al., 2023), cDNA samples were subjected to real-time qPCR analysis using the intercalator method. Table 1 lists the oligonucleotide primers used in this assay. The TAF10 transcript was used as the reference, and the ΔΔCt method was used to calculate relative gene expression levels.

As in our previous publication (Le and Kimata, 2021), RT-competitive PCR was performed to assess the splicing level of HAC1 mRNA. Briefly, cDNA was synthesized from total RNA samples using the poly(T)-primed RT reaction and used as the template for competitive PCR, which amplified products of different sizes from unspliced and spliced HAC1 mRNAs using the same primer set (Table 1). The PCR products were separated by agarose electrophoresis, and the fluorescence band intensity of the ethidium bromide (EtBr)-stained gels was quantified to calculate HAC1 mRNA splicing efficiency using the following formula: 100 × [band intensity of spliced form/(band intensity of spliced form + band intensity of unspliced form)].

Cells equivalent to an OD₆₀₀ of 3–5 (depending on the experiments but consistent within each experiment) were harvested by centrifugation, resuspended in 100 μL of lysis buffer containing 50 mM Tris-Cl (pH 7.9), 5 mM ethylenediaminetetraacetic acid, 1% Triton X-100, and protease inhibitors (2 mM phenylmethylsulfonyl fluoride, 100 μg/mL leupeptin, 100 μg/mL aprotinin, 20 μg/mL pepstatin A, and 1:100-diluted Calbiochem Protease Inhibitor Cocktail Set III (Merck)), and lysed by vortexing (top speed, 30 s × 6 times) with 100 μL of glass beads (425–600 μm; Sigma-Aldrich). After clarification by centrifugation at 8000 × g for 10 min, cell lysates were developed by standard sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by western blot analysis, as described previously (Mai et al., 2018). The primary antibodies used were mouse monoclonal anti-HA antibody (12CA5; Roche, Basel, Switzerland), rabbit anti-yeast BiP antiserum (Kimata et al., 2003), and mouse monoclonal anti-Pgk1 antibody (22C5D8; Abcam, Cambridge, UK).

GFP fluorescence in cells was visualized using a BZ-9000E microscope (Keyence, Osaka, Japan) equipped with a CFI Plan Apo λ100 × H objective lens (Nikon) and a preset fluorescent filter set (excitation: 470/30 nm; dichroic mirror: 495 nm; emission: 535/25 nm).

To monitor the protein synthesis rate, cells were grown in SD medium supplemented with auxotrophic requirements (50 mg/L uracil, 75 mg/L leucine, 75 mg/ lysine, and 50 mg/L histidine-HCl) and exposed to the stress stimuli as described above. Then, 1 mL of the resulting cultures (OD₆₀₀ approximately 0.2) were transferred to 13 mL test tubes and mixed with [³⁵S]-labeled methionine/cysteine (0.65 Mbq; EXPRE³⁵S³⁵S Protein Labeling Mix, Revvity, Waltham, MA, United States), followed by further incubation for 8 min. These procedures were performed at 30 °C under aerobically shaking conditions. Protein synthesis was halted by adding 1/10 volume of 100% trichloroacetic acid (TCA) to the cultures. The TCA precipitates were sequentially washed with 5% TCA (five times) and acetone (two times), and their radioactivity was measured using liquid scintillation counting. The protein synthesis ratio was calculated as follows: (³⁵S radioactivity of the TCA precipitate)/(culture OD₆₀₀).

Polysome profiles were analyzed as previously described (Inada and Aiba, 2005; Uemura et al., 2020; Ando et al., 2023). Cell extracts were subjected to sucrose gradient-based separation using the Gradient Master 107–201 M and Fractionator 152–002 (BioComp Instruments, Tatamagouche, NS, Canada).

Glucose concentration in the medium was measured using an enzymatic method (Enzytec Liqid D-Glucose; Darmstadt, Germany).

Numerical data are presented as the mean ± standard deviation of triplicate independent cultures. For experiments using cells transformed with pRS313-IRE1 or pRS313, three independent transformants of the same genotype were analyzed.

We previously reported that culturing S. cerevisiae cells in the presence of 16% ethanol induces HAC1 mRNA splicing. In the present study, the ethanol concentration in cultures was increased stepwise, as shown in Figure 1A. Using this stepwise addition method, we observed substantial splicing of HAC1 mRNA when the ethanol concentration reached 16% (Figure 1B). In addition, we confirmed that DTT and tunicamycin, two widely used ER stressors, also strongly triggered HAC1-mRNA splicing (Figure 1B).

It should be noted that, in this study, we increased the ethanol concentration stepwise to avoid acute cell death. When ethanol was added all at once to a final concentration of 16%, cells were rapidly and irreversibly damaged, making it impossible to investigate their physiological responses to ethanol stress (Supplementary Figure S1A). However, when ethanol was added to the cultures using the procedure shown in Figure 1A, cells seemed to lose their viability more slowly (Supplementary Figure S1B).

In response to severe ER stress, Ire1 forms self-associated clusters, which can be visualized as Ire1 puncta, to exert strong HAC1 mRNA splicing activity (Kimata et al., 2007; Korennykh et al., 2009; Aragón et al., 2009). As shown in Supplementary Figure S2, we examined the intracellular localization of GFP-tagged Ire1 (Ire1-GFP). Consistent our previous observations (Ishiwata-Kimata et al., 2013), Ire1-GFP exhibited a typical double ring-like ER distribution in non-stressed (NS) cells (Supplementary Figure S2A), whereas in cells stressed with tunicamycin or DTT, it appeared clustered at least partly (Supplementary Figures S2B–D). While 2.0 μg/mL tunicamycin apparently caused the punctate distribution of Ire1-GFP (Supplementary Figure S2B), it seemed to partly cluster when tunicamycin was added to the cultures at a lower concentration of 1.0 μg/mL (Supplementary Figure S2C). A similar punctate distribution of Ire1-GFP was observed in cells exposed to 16% ethanol via the stepwise addition method (Supplementary Figure S2E). Therefore, throughout this study, ethanol was added to cultures using the stepwise addition protocol shown in Figure 1A, which sufficiently activated Ire1 to induce HAC1-mRNA splicing.

To compare the characteristics of cells carrying and not carrying the IRE1 gene, we transformed an IRE1-knockout strain (Y11907) with either a low-copy IRE1 plasmid (pRS313-IRE1) or an empty vector (pRS313), and the resulting transformants were used as IRE1 + and ire1Δ strains, respectively. As shown in Figure 1C, IRE1 + cells exposed to 16% ethanol exhibited a substantial level of HAC1 mRNA splicing, which was comparable to that induced by 1.0 μg/mL tunicamycin. As expected, the HAC1 mRNA splicing was not observed in ire1Δ cells. In the experiment shown in Figure 1D, cell survival after stress induction was measured using colony formation assay. While tunicamycin impaired the viability of ire1Δ cells more severely than that of IRE1 + cells, no significant difference in viability was observed between IRE1 + and ire1Δ cells under ethanol stress. Therefore, under the ethanol exposure condition employed in this study, the IRE1-dependent UPR pathway is unlikely to affect cell survival, although Ire1 is activated to mediate HAC1-mRNA splicing.

As shown in Supplementary Figures S3A,B, cell growth was retarded when ethanol was added to cultures, and completely halted when the ethanol concentration reached 16%. We do not deduce that this is due to glucose consumption or diauxic shift, as cells exhibited similar growth patterns upon ethanol exposure even when incubated at different densities (Supplementary Figure S3A). The medium still contained a substantial concentration of glucose even after cells were cultured with 16% ethanol for 60 min (Supplementary Figures S3A,B). Since cells were subjected to various assays at this time point throughout this study, our observations presented in this paper are unlikely to be caused by glucose depletion.

Next, using mRNA-seq, we investigated IRE1-dependent transcriptome changes induced by tunicamycin treatment and ethanol exposure. IRE1 + and ire1Δ cells were treated with 1.0 μg/mL tunicamycin for 30 min or exposed to ethanol, which was added stepwise to a final concentration of 16%, followed by an additional incubation for 60 min. Both stimuli induced HAC1 mRNA splicing at similar levels (Figure 1C). Supplementary Tables S1, S2 present the TPM values obtained from the mRNA-seq analysis. As shown in Figure 2A, a considerable number of genes were differentially expressed between tunicamycin-treated IRE1 + and ire1Δ cells. The change in gene expression was not as pronounced as that presented in our previous study (Ishiwata-Kimata et al., 2025), due to the lower tunicamycin concentration used in this study. In contrast and unexpectedly, almost no difference in gene expression was observed between IRE1 + and ire1Δ cells under the ethanol exposure condition (Figure 2B).

Figure 2C presents a heat map visualizing the mRNA-seq data for selected genes. Supplementary Table S3 lists the numerical data used to generate Figure 2C. The “prominent UPR target genes” refer to differentially expressed genes (DEGs) that were expressed at least fourfold higher in IRE1 + cells than in ire1Δ cells under the tunicamycin treatment condition. For many of these genes, including SIL1 and JEM1, expression levels in ethanol-exposed IRE1 + and ire1Δ cells were similar to those in tunicamycin-treated ire1Δ cells, suggesting that they were not induced by the ethanol exposure. However, other genes, such as ERO1 and KAR2, exhibited higher expression in ethanol-exposed IRE1 + and ire1Δ cells than in tunicamycin-treated ire1Δ cells. As ERO1 and KAR2 are known to be induced not only by the UPR but also by the heat shock response (HSR) (Kohno et al., 1993; Takemori et al., 2006), we examined the expression of representative heat shock protein (HSP) genes and found that they were induced in both IRE1 + and ire1Δ cells upon ethanol exposure (Figure 2C).

To confirm this observation, we used another pair of IRE1 + and ire1Δ strains and monitored expression levels of the representative UPR-target genes, KAR2, ERO1, SIL1, and JEM1, using RT-qPCR. BY4741 is a wild-type strain carrying the intact IRE1 gene on its genome, whereas Y01907 is an ire1Δ derivative of BY4741. KAR2 encodes the ER-resident molecular chaperone BiP, and SIL1 and JEM1 encode co-chaperones of BiP that are involved in protein quality control in the ER (Hebert et al., 1995; Nishikawa et al., 2001; Rosam et al., 2018). ERO1 is involved in protein disulfide bond formation in the ER (Zito, 2015).

In the experiments shown in Figure 3, ethanol was added stepwise to a final concentration of 16%, followed by further incubation for three different time periods. Consistent with the results in Figures 1B,C, ethanol exposure triggered HAC1-mRNA splicing in IRE1 + cells, but not in ire1Δ cells (Figures 3A,B). The cells were also stressed with 1.0 μg/mL tunicamycin for 30 min. As shown in Figures 3C–F, these UPR-target genes were induced by tunicamycin in IRE1 + cells but not in ire1Δcells. However, the expression of KAR2 and ERO1 was similarly and highly induced in both IRE1 + and ire1Δ cells by ethanol exposure (Figures 3C,D). We deduce that this is due to the HSR, which does not depend on IRE1, rather than the UPR. Other UPR target genes, SIL1 and JEM1, were not (or only poorly) induced by ethanol exposure (Figures 3E,F). Taken together, at least under the conditions employed here, ethanol stress activates Ire1 and induces HAC1 mRNA splicing, but does not trigger downstream gene induction.

The weak and transient induction of JEM1 upon 30-min exposure of cells to 16% ethanol (Figure 3F, the third and fourth columns) is unlikely to result from the UPR because it was similarly observed in both IRE1 + and ire1Δ cells. It should also be noted that the HAC1 mRNA splicing was detectable, but only weak, at this time point (Figure 3A). Since JEM1 is unlikely to be a heat shock gene, another stress-responsive pathway may be responsible for this weak induction of JEM1 at this time point.

Supplementary Figure S4 presents the results of the control experiments in which cells were stressed by conventional ER stress stimuli. To induce HAC1 mRNA splicing for long durations, cells were incubated in the presence of a high concentration (2.5 μg/mL) of tunicamycin, because HAC1 mRNA splicing subsides when cells are sustainedly treated with 1.0 μg/mL tunicamycin. Cells were also incubated in the presence of two different concentrations of DTT. In addition, cells were cultured under the inositol depletion condition. As shown in Supplementary Figures S4A–D, all of these stress stimuli triggered the HAC1 mRNA splicing in IRE1 + cells, but not in ire1Δcells. Supplementary Figure S4E–P shows that these stimuli induced the representative UPR target genes in IRE1 + cells but not in ire1Δcells. It should be particularly noted that the UPR target genes were induced by tunicamycin even 420 min after onset of the stress. Therefore, we do not believe that the ethanol exposure failed to induce the UPR target genes simply because the stress exposure time was too long to sustain induction of these genes.

The observations shown in Figure 3 and Supplementary Figure S4 suggest that the HSR, which induces KAR2 and ERO1 independently of Ire1, is triggered by ethanol but not by the conventional ER stress stimuli such as tunicamycin, DTT, or inositol depletion. Consistent with this, Supplementary Figure S5 shows that, unlike the conventional ER stress stimuli, ethanol exposure drastically induced HSP104, which is a representative heat shock gene.

Next, we investigated why UPR target genes were not induced by ethanol stress. In the experiment shown in Figure 4, we used S. cerevisiae cells carrying the genomic HAC1 gene with an in-frame insertion of an HA epitope-tag sequence (IRE1+/3 × HA-HAC1 cells). As in the previous experiments, cells were treated with 1.0 μg/mL tunicamycin for 30 min or exposed to ethanol added stepwise to a final concentration of 16%, followed by a further 60-min incubation. These two stress stimuli induced the HAC1-mRNA splicing to similar levels (Figure 4A). In the experiment shown in Figure 4B, the HA-tagged Hac1 protein (HA-Hac1) was detected using anti-HA western blotting. While HA-Hac1 was barely detectable in non-stressed (NS) cells, its cellular abundance increased substantially following tunicamycin treatment. This observation is consistent with the well-established view that HAC1 mRNA is translated only when spliced (Rüegsegger et al., 2001). However, HA-Hac1 protein was not induced upon ethanol exposure (Figure 4B), although HAC1 mRNA was well spliced (Figure 4A).

In the experiment shown in Figure 5A, cells were stressed as done in other experiments performed throughout this study, and global protein synthesis was assessed by monitoring the incorporation of [³⁵S]-labeled methionine/cysteine into TCA-insoluble fractions of cells. Unlike conventional ER stress stimuli such as tunicamycin, DTT, and inositol depletion, the ethanol exposure almost completely abolished [³⁵S]-labeled methionine/cysteine incorporation. Consistent with this observation, polysome analysis revealed that polysomal ribosome peaks were lost following the ethanol exposure, indicating the attenuation of global protein synthesis (Figure 5B). Therefore, we presume that under the ethanol exposure condition employed here, cell growth was almost completely halted (Supplementary Figures S3A,B) because protein synthesis was severely suppressed due to the ethanol toxicity. Nevertheless, BiP levels were elevated upon the ethanol exposure (Supplementary Figure S6).

Based on these findings, we deduced that under the ethanol stress condition, HAC1-mRNA splicing does not lead to downstream gene induction because the spliced form of HAC1 mRNA is not efficiently translated. To confirm this idea, we investigated whether inhibition of global protein synthesis using another method would yield a similar outcome. In the experiment shown in Figure 6, the protein synthesis inhibitor cycloheximide was added into cultures of IRE1 + cells in combination with DTT. Figure 6A shows the treatment conditions and the extent of HAC1 mRNA splicing. Cycloheximide alone did not induce HAC1 mRNA splicing (Figure 6Ab). The HAC1 mRNA splicing was also minimal when DTT and cycloheximide were added simultaneously (Figure 6Ad). However, HAC1 mRNA was spliced, albeit weakly, when cycloheximide was added into cultures 10 min after 10 mM DTT addition (Figure 6Ae). This sequential treatment induced HAC1-mRNA splicing at a level comparable to that observed in cells treated with 1.0 mM DTT alone (Figure 6Ac). Figures 6B–E indicate that the UPR target genes were not induced by the sequential addition of DTT and cycloheximide, despite the splicing of HAC1 mRNA (Figure 6Ae).

Finally, we examined the physiological significance of our finding that under certain stress conditions, UPR target genes are not induced despite HAC1-mRNA splicing. Previously, we constructed a YCp-type plasmid, pCM189-Hac1, for the expression of Hac1 from the intronless HAC1 mutant gene under the control of the Tet-off promoter in S. cerevisiae cells (Ishiwata-Kimata et al., 2025). When cells carrying pCM189-Hac1 were cultured without doxycycline, which represses the Tet-off promoter, Hac1 was expressed, inducing UPR target genes, even under non-stress conditions. The expression level of Hac1 from pCM189-Hac1 was insufficient to affect growth rates under non-stress conditions (Ishiwata-Kimata et al., 2025). In the experiment shown in Figure 7, IRE1 + cells carrying either pCM189-Hac1 or the empty vector pCM189 were grown without doxycycline, and ethanol was added stepwise to a final concentration of 16%, followed by a further incubation for monitoring their viability. Cells carrying pCM189-Hac1 exhibited a more rapid loss of viability compared to those carrying pCM189 in the presence of 16% ethanol. Therefore, we assume that cells become highly susceptible to ethanol stress when the UPR is preemptively induced. This observation suggests a negative effect of the UPR and implies the physiological importance of suppressing the UPR under ethanol stress conditions.

The cellular levels of Hac1 under this experimental condition were assessed using the HA epitope-tagging method. Consistent with the result shown in Figure 4B, HA-Hac1 was clearly produced in response to ER stress in cells carrying the genomic HA-HAC1 gene (Supplementary Figure S7; the leftmost and second lanes). On the contrary, HA-Hac1 was substantially produced from the HA epitope-carrying version of pCM189-Hac1 even under non-stress conditions, and its level decreased but was still detectable for at least 120 min of culturing in the presence of 16% ethanol (Supplementary Figure S7; the third, fourth, and rightmost lanes).