Ethanol application enhances freezing stress tolerance in Arabidopsis and sugar beet Authors: Daisuke Todaka, Kentaro Nakaminami, Akihiro Matsui, Seishi Ikeda, Quynh Thi Nhu Do, Maho Tanaka, Satoshi Takahashi, Chieko Torii, Junko Ishida, Tomoyuki Takeda, Atsushi J Nagano, Motoaki Seki Published: 2026 ABSTRACT AbstractCold stress, especially freezing stress, is an important environmental factor that induces physiological damage in plants. In previous studies, we showed that ethanol application increases tolerance to high salinity, heat, high-intensity light, and drought, although the core mechanisms remain unclear. In this study, we investigated whether ethanol application could enhance freezing stress tolerance in Arabidopsis and sugar beet. Arabidopsis seedlings grown on agar plates were pretreated with 20 mM ethanol solution and then subjected to freezing. The survival ratio and growth indicators, including green leaf area, were higher in ethanol-pretreated seedlings than in water-pretreated seedlings. Similarly, in Arabidopsis seedlings grown in pots containing soil, 20 mM ethanol pretreatment increased the green leaf area after freezing treatment. Pretreatment with 80 mM ethanol solution of potted sugar beet seedlings grown in soil before exposure to freezing increased the survival ratio compared with that of water-pretreated seedlings. Transcriptome analysis using potted Arabidopsis plants grown in soil identified differentially expressed genes (DEGs), which included stress-related genes (LEA, DREB1A and ERF6) and anthocyanin biosynthesis-related genes (LDOX/ANS and DFR). Gene ontology analysis showed that DEGs associated with hypoxic response including oxidative stress response and anthocyanin-containing compound metabolic process were enriched in seedlings pretreated with 20 mM ethanol. Collectively, the results reveal that ethanol application effectively increases freezing stress tolerance and provide novel insights into the molecular mechanisms that underlie ethanol-mediated abiotic stress tolerance.Supplementary InformationThe online version contains supplementary material available at 10.1007/s11103-026-01719-5.Keywords: Ethanol, Cold stress, Chemical priming, Arabidopsis, Sugar beet Introduction Exposure to low-temperature stress adversely affects the physiological status of plants. In particular, temperatures below the freezing point can cause severe ultrastructural damage that includes cell membrane disruption (Jahed et al. 2023). In the case of exposure to low temperatures above the freezing point, plants activate a unique mechanism to cope with future freezing stress via regulatory networks (Larran et al. 2023; Kim et al. 2024). For example, the expression levels of genes encoding Dehydration-Responsive Element-Binding Protein 1 (DREB1)/C-repeat Binding Factor (CBF) family proteins are increased at low temperatures above the freezing point, and these proteins function as core regulators in the transcriptional cascades (Thomashow 1999; Shinozaki and Yamaguchi-Shinozaki 2022). Indeed, transgenic plants overexpressing genes encoding DREB1/CBF family proteins show an increase in freezing tolerance (Jaglo-Ottosen et al. 1998; Kasuga et al. 1999). Although genetic engineering is a powerful approach for modification of biological characteristics, its social acceptance remains controversial, especially for food crops. Recently, chemical priming has emerged as a promising alternative approach that can endow plants with favored traits, such as stress tolerance. A growing body of evidence shows that a diverse variety of molecules are useful as chemical priming agents (Savvides et al. 2016; Sako et al. 2021b; Zulfiqar et al. 2022; Bashir et al. 2025). Our previous investigations of the effects on plants of treatment with ethanol have provided evidence that ethanol application is effective for the enhancement of tolerance to several abiotic stress factors. We reported that pretreatment with 0.3% (51 mM) ethanol for 24 h increases the tolerance to high salinity stress of Arabidopsis (Arabidopsis thaliana L.) seedlings grown in liquid medium (Nguyen et al. 2017). Similarly, 2-week-old Arabidopsis seedlings grown in liquid medium containing 10 mM ethanol experience reduced severity of photosynthesis damage caused by high-intensity light treatment (Sako et al. 2021a). Pretreatment with 2–50 mM ethanol increases the heat stress tolerance of Arabidopsis seedlings grown on agar plates through activation of the unfolded protein response mechanism (Matsui et al. 2022). Furthermore, 5–20 mM ethanol pretreatment enhances drought stress tolerance in potted Arabidopsis seedlings grown in soil (Bashir et al. 2022). The increase in drought stress tolerance in response to ethanol application may be caused by multiple factors, including activation of gluconeogenesis, increased stomatal closure, and diminished leaf water loss (Bashir et al. 2022). Similar ethanol application increases drought stress tolerance in cassava (Vu et al. 2022). In addition, we reported that 20 mM ethanol application increases heat stress tolerance with sugar accumulation in tomato (Todaka et al. 2024). Other research groups have reported ethanol-mediated drought tolerance in soybean (Rahman et al. 2022) and cold tolerance in sorghum (Ghosh et al. 2024). These studies have shown that ethanol application activates the antioxidant defense system and increases the accumulation of metabolites, including soluble sugars. These findings raise the question of whether ethanol treatment is of general utility for the alleviation of cellular and physiological damage to plants from diverse environmental stresses. Arabidopsis is typically grown under conditions at 18—23 °C, which corresponds to the temperature range where growth and development are optimized (Casal and Balasubramanian 2019). It was observed that root meristem activity was decreased at 15 °C (Yang et al. 2017). Transcriptome analysis showed that expression levels of genes related to growth were reduced at 10 °C while expression levels of genes associated with cold stress tolerance were increased at 4 °C (Ruiz-Aguilar et al. 2024). Exposure to below 0 °C temperatures resulted in freezing damage, as evidenced by reduced survival and increased injury after freezing treatments (Hannah et al. 2006). Sugar beet (Beta vulgaris L.) is an important crop cultivated for sugar production in temperate areas. When sugar beet plants were grown at 10, 17, 24, and 31 °C temperatures, sugar concentrations in the storage root were greatest at 17 °C (Terry 1968). Similar to Arabidopsis plants, sugar beet plants are adversely affected by the exposure of below 0 °C temperatures (Yolcu et al. 2022). For example, when sugar beet seedlings were exposed to -3 °C for 2 h, electrolyte leakage was increased (Kulan and Kaya 2024). In field-grown sugar beet cultivation, frost events during early spring and autumn frequently damage seedlings, resulting in reduced plant density, delayed development and costly re-sowing and yield damage was estimated by modeling in a range of 62.5 to 100% from 1993 to 2009 in northeastern Iran (Deihimfard et al. 2019). In the present study, we examined the effects of ethanol application on tolerance to freezing (< 0 °C) stress in Arabidopsis and sugar beet. The results showed that ethanol pretreatment increased the survival of seedlings of both plant species exposed to freezing stress. Transcriptome analysis revealed that many genes associated with hypoxic response were induced by ethanol application. The findings support the hypothesis that ethanol application is broadly useful to mitigate physiological damage to plants caused by abiotic stress. Materials and methods Plant materials and growth conditions Arabidopsis thaliana wild-type (ecotype Columbia; Col-0, RIKEN Bioresource Research Center, Japan), ein2—1 mutant (CS3071, Arabidopsis Biological Resource Center, USA), eto1—1 mutant (CS3072, Arabidopsis Biological Resource Center, USA) plants and sugar beet (Beta vulgaris subsp. vulgaris ‘Monohikari’, National Agriculture and Food Research Organization, Japan) plants were used in the present study. For plate-based experiments, seedlings were grown in petri dishes on solid medium comprising 1 × Murashige and Skoog medium, 1% sucrose, and 0.8% agar for 14 days at 22 °C under a 16 h/8 h (light/dark) photoperiod in a growth chamber (LPH-350S, Nippon Medical & Chemical Instruments Co., Ltd., Osaka, Japan). For pot-based experiments, seeds were sown in pots containing Professional Potting Soil (Dio Chemicals Ltd., Tokyo, Japan). After sowing, plants were grown for 2–3 weeks at 22 °C under a 16 h/8 h (light/dark) photoperiod in a growth room. Ethanol pretreatment and freezing treatment In experiments using seedlings grown in liquid medium, seedlings were pretreated with water or 20 mM ethanol for 3 d and then subjected to freezing (− 6.5 °C) overnight in a low-temperature incubator (LU-123, ESPEC Corporation, Osaka, Japan). After freezing, the seedlings were grown under normal conditions for 14 d. In experiments involving gaseous ethanol application, seedlings were pretreated with water or gaseous ethanol for 3 d and then subjected to freezing (− 9.0 °C) overnight in the same low-temperature incubator. After freezing treatment, seedlings were grown under normal conditions for 14 d. In experiments using potted seedlings grown in soil, seedlings were pretreated with water or 20–80 mM ethanol for 3–6 d by immersing the base of each pot into the relevant solution and then subjected to freezing (− 4 °C) overnight in the same low-temperature incubator. After freezing treatment, the seedlings were grown under normal conditions for 7 − 14 d. Measurement of green leaf area Green leaf area was measured using OpenCV (version 4.0.1) with Python 3.8.5. The threshold range for the hue, saturation, and value color space was set from [31, 70, 10] to [95, 255, 255]. Transcriptome analysis Four to eight biological replicates per treatment were used. For the transcriptome analysis using pot-grown Arabidopsis seedlings, plants were sampled at 14 days after stratification (DAS; 0), after water pretreatment for 6 d (W6d), after ethanol pretreatment for 6 d (E6d), after water pretreatment for 6 d and freezing treatment for 2 h (W6dF2h), and after ethanol pretreatment for 6 d and freezing treatment for 2 h (E6dF2h) (Fig. 3a). For the transcriptome analysis using pot-grown sugar beet seedlings, plants were sampled at 7 DAS (0), after water pretreatment for 3 d (W3d), after ethanol pretreatment for 3 d (E3d), after water pretreatment for 3 d and freezing treatment for 2 h (W3dF2h), and after ethanol pretreatment for 3 d and freezing treatment for 2 h (E3dF2h) (Fig. 3a and S10a). For RNA-sequencing (RNA-seq) analysis, the RNA library was prepared using the Lasy-Seq method (Kamitani et al. 2019). Specifically, 200 ng total RNA per sample was used. The library was sequenced on a HiSeq X Ten platform (Illumina, San Diego, CA, USA) using the 151 bp paired-end mode. The RNA-seq analyses were performed with the R1 reads. Low-quality reads and adapters were trimmed using Trimmomatic version 0.39 (http://www.usadellab.org/cms/?page=trimmomatic) with the settings ‘ILLUMINACLIP:TruSeaq3-SE.fa:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36’. HISAT2 version 2.2.1 (http://daehwankim-lab.github.io/hisat2/) was used to map the reads to the Arabidopsis thaliana reference genome (TAIR10) with the ‘–max-intronlen 5000’ option. Aligned reads that were within gene models were counted using featureCounts version 2.0.1 (http://subread.sourceforge.net/) with the ‘–fracOverlap 0.5 -O -t gene -g ID -s 1 –primary’ options. Differentially expressed genes (DEGs) were identified using R version 4.0.5 (https://www.r-project.org/) and the DESeq2 version 1.30.1 R package (https://bioconductor.org/packages/release/bioc/html/DESeq2.html). For the transcriptome analysis using pot-grown Arabidopsis and sugar beet seedlings, up- and down-regulated genes were identified based on the criteria log2 fold change ≥ 1.0 with false discovery rate < 0.05 and log2 fold change ≤ -1.0 with false discovery rate < 0.05, respectively, in each comparison. For the transcriptome analysis using plate-grown Arabidopsis seedlings, we set the different values of criteria log2 fold change because up- and down-regulated genes were not sufficiently identified. Up- and down-regulated genes were identified based on the criteria log2 fold change ≥ 0.585 with false discovery rate < 0.05 and log2 fold change ≤ -0.585 with false discovery rate < 0.05, respectively, in each comparison. Transcriptome analysis of Arabidopsis seedlings grown in pots. (a) Schematic diagram of sampling time points for transcriptome analysis. Fourteen-day-old seedlings were pretreated with 20 mM ethanol for 6 d and treated with freezing at − 4 °C for 2 h. Stars represent the sampling time points. (b) PCA of normalized read counts of 0, W6d, E6d, W6dF2h, and E6dF2h samples. (c) Volcano plots of comparative data between E6d and W6d (E6d vs. W6d) and between E6dF2h and W6dF2h (E6dF2h vs. W6dF2h). Horizontal green lines indicate the threshold q-value (q < 0.05). Vertical green lines indicate the threshold values of up-regulated DEGs (log2 fold change ≥ 1.0) and down-regulated DEGs (log2 fold change ≤ − 1.0). Characters in red and blue indicate numbers of up- and down-regulated DEGs, respectively. (d) Venn diagrams of numbers of DEGs in comparisons as indicated. (e) Representative enriched GO terms of DEGs in E6d vs. W6d and E6dF2h vs. W6dF2h. GO IDs are shown in Table S6 and S8 Quantitative reverse transcription PCR (RT-qPCR) Four to eight biological replicates per treatment were used. Samples were placed in a 2 mL tube, frozen in liquid nitrogen, and pulverized using the Multi-Beads Shocker system (Yasui Kikai, Osaka, Japan). cDNA synthesis and PCR were performed as previously described (Matsui et al. 2022). Primer sequences used are listed in Table S1. Statistical data analysis Statistical analyses were performed using Fisher’s exact test and Tukey’s multiple comparison test. Fisher’s exact test was used for categorical data. Differences among groups were evaluated by one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test. All analyses were performed using R (version 4.4.2), and P < 0.05 was considered statistically significant. Information on biological replicates is provided in the respective figure legends. Results Ethanol-induced freezing stress tolerance in Arabidopsis First, we investigated the effects of ethanol application on freezing tolerance in Arabidopsis plants using plastic well plates (Fig. 1a). When seedlings pretreated with water were subjected to freezing, only three of the 12 plants survived (Well_1 in Fig. 1b). In contrast, all seedlings pretreated with ethanol survived after exposure to freezing (Well_2 in Fig. 1b). Figure 1c shows representative plates of seedlings grown under the normal condition for 14 d after freezing treatment. The green leaf area of the ethanol-pretreated seedlings (Well_2 in Fig. 1d) was higher than that of the water-pretreated seedlings (Well_1 in Fig. 1d). The same trend was observed for seedling fresh weight and chlorophyll content (Fig. 1e, f). To confirm the increase in freezing tolerance in response to ethanol pretreatment, we performed freezing tolerance tests using pot-grown Arabidopsis seedlings. As expected, the freezing tolerance of pot-grown seedlings pretreated with ethanol was increased compared with that of pot-grown seedlings pretreated with water (Fig. 2a, b). Furthermore, the ethanol-induced freezing tolerance of pot-grown seedlings was reproducible (Fig. S1a, b). These results suggest that ethanol application can increase freezing tolerance in Arabidopsis. Enhancement of freezing stress tolerance in Arabidopsis by ethanol pretreatment through liquid-based cultivation using plastic well plates. Seedlings were pretreated with water or 20 mM ethanol for 3 d (five plants per well) and then subjected to freezing treatment (− 6.5ºC, overnight). After the freezing treatment, the seedlings were grown under normal conditions for 14 d. (a) Schematic diagram of liquid cultivation. (b) Survival ratios of wells containing seedlings treated with freezing. The green leaf area value in a well was used for this analysis. A value higher than 10 mm2 was defined as a well containing surviving plants. Well_1 to Well_4 show each well in (a). Differences between survival ratios were evaluated with Fisher’s exact test. (c) Appearance of seedlings grown under normal conditions for 14 d after freezing treatment. (d) Box plot of green leaf areas of seedlings in (c). A value from one well was regarded as one sample (n = 9–12). (e) Box plot of fresh weights of seedlings in (c) (n = 3). (f) Box plot of chlorophyll contents of seedlings in (c) (n = 3). (d) to (f) Different letters indicate significant differences at P < 0.05 (Tukey–Kramer test) Freezing stress tolerance test of soil-grown Arabidopsis plants. Fifteen-day-old seedlings were pretreated with water or 20 mM ethanol for 6 d and then subjected to freezing treatment (− 4ºC, 16 h). After the freezing treatment, the seedlings were grown under normal conditions for 7 d. (a) Appearance of seedlings grown under normal conditions for 7 d after freezing treatment. (b) Box plot of green leaf areas of seedlings in (a) (n = 11–12). Different letters indicate significant differences at P < 0.05 (Student’s t-test) Transcriptome analysis of Arabidopsis plants pretreated with ethanol To elucidate the molecular mechanisms responsible for the ethanol-induced freezing tolerance, we performed transcriptome analysis using pot-grown Arabidopsis plants. After sequencing and data normalization, a principal component analysis (PCA) scatterplot indicated that the gene expression patterns differed strongly between the 0 samples and the other samples (Fig. 3b). Next, we aimed to identify up- and down-regulated DEGs. In the comparison E6d vs. W6d, 114 and 113 genes were identified as up- and down-regulated DEGs (Fig. 3c, Tables S2 and S3). In the comparison E6dF2h vs. W6dF2h, 69 and 93 genes were identified as up- and down-regulated DEGs (Fig. 3c, Tables S4 and S5). The expression patterns of 14 DEGs were confirmed by RT-qPCR analysis (Figs. S2 and S3), validating the results of the transcriptome analysis. The identified DEGs partially overlapped between the E6d vs. W6d and E6dF2h vs. W6dF2h comparisons (Fig. 3d). To explore the functions of the identified DEGs, we conducted an enriched gene ontology (GO) analysis. A variety of GO terms were enriched among up- and down-regulated DEGs of E6d vs. W6d and E6dF2h vs. W6dF2h (Tables S6–S9). Little involvement of ethylene response in ethanol-induced freezing stress tolerance In the enriched GO analysis, we noted that GO terms associated with ethylene response predominated among the up-regulated DEGs of E6d vs. W6d (Table S6). These included GO terms associated with leaf senescence, hypoxia, and wounding (Fig. 3e). Thus, we considered that ethylene-related signaling pathways might be involved in the ethanol-mediated freezing stress-tolerance mechanism. For clarification, we assessed the ethanol-mediated freezing stress tolerance of the Arabidopsis ethylene-overproduction mutant, ethylene overproducer 1 (eto1) and ethylene insensitive 2 (ein2). The stress tolerance test analysis showed that growth indicators including green leaf area and fresh weight of these mutants pretreated with ethanol and exposed to freezing were not significantly different from those of Col-0 (Figs. S4 and S5). In addition, we investigated ethanol-mediated freezing stress tolerance by the application of an ethylene precursor 1-aminocyclopropane-1-carboxylate (1-ACC) or an ethylene biosynthesis inhibitor AgNO3. Both applications did not change significantly ethanol-mediated freezing stress tolerance (Fig. S6a, b). These findings suggest that ethylene-related signaling pathways are not involved in ethanol-mediated freezing stress tolerance. In comparisons between water-pretreated Col-0 and water-pretreated eto1 and between water-pretreated Col-0 and water-pretreated ein2, green leaf area and fresh weight after freezing treatment were not significantly different (Figs. S4 and S7). Enhanced freezing stress tolerance in response to gaseous ethanol In the freezing stress tolerance tests using plastic well plates in Arabidopsis (Fig. 1), the survival ratio of seedlings in Well_3 was increased compared with that of seedlings in Well_1 (Fig. 1b). Well_3 contained seedlings in water, and Well_2 contained seedlings in ethanol solution (Fig. 1a). Each plate was covered with a lid, but there was a slight gap between the lid and the upper edge of the well. Thus, gaseous molecules could diffuse to the neighboring well of the plate. Therefore, we considered that gaseous ethanol from Well_2 might diffuse and affect the seedlings of Well_3. To test this possibility, Well_4 contained seedlings in water, and the neighboring wells were filled with ethanol solution in the same plate (Fig. 1a). As expected, the survival ratio, green leaf area, seedling fresh weight, and chlorophyll content of seedlings in Well_4 were increased compared with those of seedlings in Well_1 (Fig. 1b, d–f). To confirm the effects of gaseous ethanol on freezing stress tolerance, Arabidopsis seedlings were grown on agar plates, which were divided into two halves by a semi-partition (Fig. 4a). Seedlings were cultivated on one half and water or different concentrations of ethanol solution were poured onto the other half. For the plates containing ethanol, it was expected that gaseous ethanol could diffuse to the air space of the half in which seedlings were cultivated. After pretreatment with water or ethanol, the seedlings were subjected to freezing and the survival ratio was determined. The survival ratio of seedlings pretreated with 10 mM ethanol was significantly increased compared with that of seedlings pretreated with water (Fig. 4b). Figure 4c shows representative plates of seedlings grown under the normal condition for 14 d after freezing treatment. Taken together, these results indicate that gaseous ethanol could affect the plants and increase their freezing stress tolerance. Enhancement of freezing stress tolerance in Arabidopsis by gaseous ethanol pretreatment. Eleven-day-old seedlings were pretreated with water or gaseous ethanol for 3 d and then subjected to freezing treatment (− 9.0 ºC, overnight). After the freezing treatment, seedlings were grown under normal conditions for 14 d. (a) Schematic diagram of gaseous ethanol pretreatment. In one plate, seedlings were cultivated in the left half of the plate, and water or ethanol solution was poured onto the right half of the plate. (b) Survival ratio of seedlings treated with freezing. Differences between survival ratios were evaluated with Fisher’s exact test. (c) Appearance of seedlings grown under normal conditions for 14 d after freezing treatment Transcriptome analysis of Arabidopsis plants treated with gaseous ethanol To examine the molecular mechanisms responsible for gaseous ethanol-induced freezing stress tolerance, we performed a transcriptome analysis of Arabidopsis seedlings treated with the method described in Fig. 4a. Figure S8a shows a schematic diagram of the sampling time points, which were set at 1 d (W1d or gE1d) and 3 d (W3d or gE3d). Three and 13 DEGs up-regulated in the comparisons gE1d vs. W1d and gE3d vs. W3d, respectively, were identified (log2 fold change ≥ 0.585, q-value < 0.05), whereas zero and two DEGs down-regulated in the same comparisons were identified (log2 fold change ≤ -0.585, q-value < 0.05) (Fig. S8b, c, Tables S10–S12). To verify the expression changes of the DEGs in the transcriptome analysis, we performed RT-qPCR analysis. Of 13 DEGs, the expression levels of nine DEGs in gE3d were higher than those in W3d (Fig. S9). Thus, the results of the transcriptome analysis were validated. Although an enriched GO analysis of the DEGs was performed, no significant GO terms were detected. Ethanol-induced freezing stress tolerance in sugar beet Next, we examined whether ethanol application could enhance freezing tolerance in sugar beet. Seedlings were pretreated with 80 mM ethanol and then subjected to freezing. The ethanol pretreatment improved the seedling growth after freezing treatment (Fig. 5a). The survival ratio of seedlings pretreated with ethanol (21/24) was significantly higher than that of seedlings pretreated with water (6/24) (Fig. 5b). Therefore, ethanol application could enhance freezing stress tolerance in sugar beet as well as in Arabidopsis. Enhancement of freezing stress tolerance by ethanol pretreatment in sugar beet. Seven-day-old seedlings were pretreated with water or 80 mM ethanol for 3 d and then subjected to freezing treatment (− 2.1ºC to − 3.5ºC, overnight). After the freezing treatment, the seedlings were grown under normal conditions. Four plants were used in each treatment. (a) Appearance of seedlings grown under normal conditions after freezing treatment. (b) Survival ratios of seedlings treated with freezing. Color tiles show the results of each test. Differences between survival ratios were evaluated with Fisher’s exact test Transcriptome analysis of sugar beet plants pretreated with ethanol Transcriptome analysis was performed using pot-grown sugar beet plants. After sequencing and data normalization, we aimed to identify up- and down-regulated DEGs. In the comparison W3dF2h vs. W3d, 401 and 334 genes were identified as up- and down-regulated DEGs (Fig. S10b, Tables S13 and S14). The up-regulated DEGs included stress-related genes, ex. heat shock protein 18.2 (Bevul.1G135600), chaperone DnaJ-domain superfamily protein (Bevul.2G006400), and low temperature and salt responsive protein family (Bevul.2G186400) (Table S13) while the down-regulated DEGs included photosynthesis-related genes, ex. RUBISCO accumulation factor 1 (Bevul.9G145800), chloroplast thylakoid lumen protein (Bevul.9G051900), and Photosystem II reaction center PsbP family protein (Bevul.8G167800) (Table S14). These suggested the reliability of the transcriptome analysis of sugar beet. The comparison E3d vs. W3d found 3 and 3 genes as up- and down-regulated DEGs (Fig. S10b, c, Table S15 and S16). The up-regulated DEGs were alpha/beta-Hydrolases superfamily protein (Bevul.5G073300), nodulin MtN21 /EamA-like transporter family protein (Bevul.4G117700), and beta vacuolar processing enzyme (Bevul.5G008000) (Table S15) and the down-regulated DEGs were HSP20-like chaperones superfamily protein (Bevul.1G116300), Chaperone DnaJ-domain superfamily protein (Bevul.5G150600), and CHY-type/CTCHY-type/RING-type Zinc finger protein (Bevul.9G040600) (Table S16). In the comparison E3dF2h vs. W3dF2h, 4 and 12 genes were extracted as up- and down-regulated DEGs (Fig. S10b, c, Table S17 and S18). They included Integrase-type DNA-binding superfamily protein (Bevul.8G144000) (up-regulated DEG) and Phosphate-responsive 1 family protein (Bevul.1G104100) (down-regulated DEG) and CHY-type/CTCHY-type/RING-type Zinc finger protein (Bevul.9G040600) (down-regulated DEG) (Table S17 and S18). The expression patterns of some of the DEGs were confirmed by RT-qPCR analysis (Fig. S10d). Enriched GO terms were not found in the DEGs of E3d vs. W3d and E3dF2h vs. W3dF2h. Discussion In this study, we demonstrated that ethanol application can increase the freezing stress tolerance of Arabidopsis and sugar beet. For Arabidopsis seedlings cultivated in liquid medium, the survival ratio, green leaf area, seedling fresh weight, and chlorophyll content of ethanol-pretreated seedlings after freezing treatment were significantly increased compared with those of water-pretreated seedlings (Fig. 1b–f). Similarly, ethanol-pretreated Arabidopsis and sugar beet seedlings grown in pots containing soil showed an increase in freezing stress tolerance (Figs. 2 and 5). In the present freezing stress tolerance experiments using potted seedlings grown in soil, ethanol pretreatment for 4 d did not enhance the stress tolerance in Arabidopsis (data not shown). Subsequently, we extended the ethanol pretreatment duration to 6 d, which successfully led to an increase in stress tolerance (Figs. 2 and S1). During the pretreatment in the present experiments, the base of each pot in a tray was continuously immersed into the relevant solution. The results imply that mild hypoxic stress might be important in the ethanol-induced stress tolerance mechanism. In addition, ethanol is produced under similar conditions. For example, stacked seeds show an increase in ethanol content compared with that of seeds spread out during imbibition (Chen et al. 2020). It is well known that ethanolic fermentation in plants is activated under hypoxia by enzymes, including alcohol dehydrogenase and pyruvate decarboxylase (Jardine and McDowell 2023). It has been reported that under hypoxic stress conditions endogenous ethanol levels were increased in Arabidopsis root tissues, inducing the production of superoxide (Kagenishi et al. 2023). Our transcriptome analysis using potted Arabidopsis seedlings revealed DEGs that may participate in the molecular mechanism by which ethanol pretreatment enhances freezing stress tolerance (Fig. 3, Tables S2–S5). In the set of up-regulated DEGs of the E6d vs. W6d comparison, GO terms associated with hypoxic response were enriched (Fig. 3e, Table S6). Taking these studies into consideration, we considered that there might be synergistic effects of hypoxic responses with ethanol-induced abiotic stress tolerance mechanisms. Combined application of ethanol with reactive oxygen species or other molecules that trigger mild hypoxic stress conditions might be more effective in increasing abiotic stress tolerance. Our transcriptome analysis using potted Arabidopsis seedlings also revealed that GO terms associated with ethylene signaling pathways were enriched in the set of up-regulated DEGs of the E6d vs. W6d comparison (Fig. 3e, Table S6). Chen et al. (2020) reported the involvement of ethylene signal transduction in ethanol responses where seed germination is inhibited by ethanol accumulation. It has also been reported that ethylene signaling negatively controls freezing stress tolerance through the decrease of the cold responsive pathway (Shi et al. 2012). These findings led us to hypothesize that ethylene signal transduction is implicated in the ethanol-mediated freezing stress tolerance mechanism. To examine this hypothesis, we compared the ethanol-mediated freezing stress tolerance between Col-0 and the ethylene-overproduction mutant eto1 and between Col-0 and the ethylene-insensitive mutant ein2. The freezing stress tolerance of eto1 and ein2 mutants was not significantly different from that of Col-0 (Figs. S4 and S5). In addition, the applications of an ethylene precursor 1-ACC and an ethylene biosynthesis inhibitor AgNO3 did not significantly affect ethanol-mediated freezing stress tolerance (Fig. S6). These results suggest that the ethanol-mediated freezing stress tolerance may be involved in the mechanisms other than ethylene signaling. Recently, Ghosh et al. (2024) reported that ethanol application induces cold tolerance in sorghum plants. In their study, the cold-stress treatment was performed at 8 °C. This implies that ethanol treatment might be useful to elicit cold-stress resilience under both chilling (> 0 °C) and freezing (< 0 °C) temperatures. In plants subjected to low temperatures above the freezing point, the key regulatory network of DREB1/CBF family proteins is activated (Larran et al. 2023; Kim et al. 2024). In the present transcriptome analysis, DREB1A was identified among the up-regulated DEGs in the E6dF2h vs. W6dF2h comparison (Table S4). Further analysis is needed to elucidate the involvement of DREB1A in ethanol-mediated freezing stress tolerance. In addition to DREB1A, Late Embryogenesis Abundant (LEA) (AT5G53820), Leucoanthocyanidin Dioxygenase (LDOX) / anthocyanidin synthase (ANS) (AT4G22880), and dihydroflavonol 4-reductase (DFR) (AT5G42800) were also identified as up-regulated DEGs in the E6dF2h vs. W6dF2h comparison (Table S4). LEA is well documented as one of the important proteins that contribute to enhancing abiotic stress tolerance (Hao et al. 2025). LDOX/ANS is a key enzyme that converts leucoanthocyanidins to anthocyanidins while DFR is responsible for the biosynthesis of leucoanthocyanidins (Koes et al. 2005). Anthocyanins function as antioxidants and contribute to enhancing abiotic stress tolerance (Buitrago et al. 2025). Ethylene Responsive Factor 6 (ERF6) (AT4G17490) was found as an up-regulated DEG in the E6d vs. W6d comparison (Table S2). ERF6 is a transcriptional regulator that regulates leaf growth and stress-related gene expression under osmotic stress condition (Dubois et al. 2015) and ROS-responsive gene expression under oxidative stress conditions (Sewelam et al. 2013; Wang et al. 2013). In the both E6d vs. W6d and E6dF2h vs. W6dF2h comparisons, three genes encoding class III CC-type glutaredoxins, ROXY8 (AT3G62960), ROXY9/CEPD2 (AT2G47880), and ROXY21 (AT4G33040) were identified as up-regulated DEGs (Tables S2 and S4). The expression levels of these ROXYs were increased under nitrogen-deficient conditions (Jung et al. 2018). Shoot-derived ROXY9/CEPD2 up-regulates the expression level of the nitrate transporter gene in roots and increases nitrogen uptake (Ohkubo et al. 2017; Kobayashi et al. 2024). These suggests that ethanol application might be effective for the improved growth under nitrogen-deficient conditions. We considered that ethanol application might also cause depression at the freezing point in biological tissues. From a physicochemical perspective, the higher the concentration of ethanol in a solution, the lower the freezing point of the solution (The Engineering ToolBox: https://www.engineeringtoolbox.com/ethanol-water-d_989.html). Further analysis is needed to elucidate our hypothesis. For example, a cryomicroscopic system equipped with a high-speed camera to capture the process of intracellular ice crystal formation (Ninagawa et al. 2016) might be useful for such an evaluation. Plants produce diverse volatile compounds. In most cases, volatile compounds convey important messages (Loreto and D’Auria 2022). For example, wounded plants emit gaseous methanol, and the emitted molecules can activate the defense response not only in the wounded plant but also in neighboring plants (Dorokhov et al. 2018). In the present study, gaseous ethanol increased the freezing stress tolerance in Arabidopsis plants (Figs. 1 and 4). A previous study showed that ethanol accumulates in rice roots under chilling stress (Kato-Noguchi 2008). Under such a condition, gaseous molecules might be advantageous for movement from the roots to the shoots owing to the low transpiration rate. In the present transcriptome analysis of seedlings pretreated with gaseous ethanol, small sets of genes were up- or down-regulated (Fig. S8). The set of up-regulated DEGs included AT1G76590. This gene encodes AtPLATZ2, a member of the PLATZ transcription factor family. AtPLATZ2 is involved in up-regulation of the Na+ content in the leaves and roots (Liu et al. 2020). This raises the possibility that ethanol application increases the Na+ content in those organs. An increase in Na+ content might cause depression of the freezing point. The DEGs identified by the transcriptome analysis of seedlings pretreated with gaseous ethanol (Tables S10–S12) were not overlapped with the DEGs identified by the transcriptome analysis of pot-grown seedlings pretreated with ethanol solution (Tables S2–S5). This might be due to different growth conditions (agar or soil) or different ethanol pretreatment times (Figs. 3a and S8a). Another point is that ethanol molecules in the experiment of gaseous ethanol application might be majorly incorporated via leaf surface parts while those in the experiment of pot-grown seedlings were incorporated from root parts. The expression level of the Arabidopsis gene encoding Alcohol Dehydrogenase 1, which had been recognized as a key enzyme that converts ethanol to acetaldehyde, was different between leaves and roots (Dolferus et al. 1994). This might cause the different activity of the ethanol-related metabolic process between different organs. We also performed the transcriptome analysis of sugar beet pretreated with ethanol. The results showed that only a small set of DEGs was identified in the E3d vs. W3d and E3dF2h vs. W3dF2h comparisons (Fig. S10 and Table S15 to S18). Although we tried to perform comparative analysis of the acquired transcriptome data between Arabidopsis and sugar beet, the limited number of DEGs in sugar beet precluded a robust assessment of similarities and differences. It was also difficult to identify any notable features based on the annotation information of individual DEGs. Further analysis is needed to elucidate molecular mechanisms governing the ethanol-induced freezing stress tolerance of sugar beet and whether the molecular mechanism of sugar beet is similar to those of other plant species or not. We also speculate that ethanol molecules may affect cellular structure such as cell membrane and organelle membrane although we have no direct evidence. It is interesting to note that F-actin organization is altered by ethanol application (Kagenishi et al. 2023). Conclusion Ethanol application increases the survival of seedlings exposed to freezing stress in Arabidopsis and sugar beet. Transcriptome analysis revealed that numerous genes associated with hypoxic response were induced by ethanol application. Ethanol-mediated freezing stress tolerance was not significantly changed in the ethylene-related Arabidopsis mutant eto1 and ein2 and by the applications of an ethylene precursor 1-ACC and an ethanol biosynthesis inhibitor AgNO3, suggesting that hypoxic response, not ethylene signaling, is involved in ethanol-mediated freezing stress tolerance. Figure 6 shows the model for the mechanism of ethanol-mediated freezing stress tolerance in Arabidopsis. DREB1A, LEA, ERF6, H2O2, and anthocyanins might be involved in the regulatory mechanisms. Experiments using mutants associated with these genes and metabolic pathways will be important to elucidate the mechanisms in future. In addition, other signaling cascades might be involved in the ethanol-mediated freezing stress tolerance mechanism. For example, in ongoing research, we are investigating mutants that show decreased or increased ethanol-mediated abiotic stress tolerance. Functional analysis of these mutants will provide insights into the underlying mechanisms. To date, we have demonstrated that ethanol application activates adaptive responses to high salinity (Nguyen et al. 2017), high-intensity light (Sako et al. 2021a), heat (Matsui et al. 2022; Todaka et al. 2024), and drought stress (Bashir et al. 2022; Vu et al. 2022). The present findings underscore that ethanol application is useful for mitigating physiological damage under diverse abiotic stress conditions. Model for the mechanism of ethanol-mediated freezing stress tolerance in Arabidopsis Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (PPTX 88176 kb) Supplementary file2 (XLSX 11 kb) Supplementary file3 (XLSX 41 kb) Supplementary file4 (XLSX 19 kb) Supplementary file5 (XLSX 12 kb) Supplementary file6 (XLSX 71 kb) Acknowledgements This work was supported by grants from RIKEN, Japan, including the RIKEN–AIST Joint Research Fund (full research) to MS, CREST (JPMJCR13B4), A-STEP (JPMJTM19BS), GteX (JPMJGX23B0), and ASPIRE (JPMJAP24A3) programs of the Japan Science and Technology Agency (JST) to MS. We thank Kaori Sako, Kyoko Y. Mogami, Minoru Ueda, Yoshinori Utsumi, Akihiro Ezoe, and Anh Thu Vu for their comments and technical support. We thank Edanz (https://jp.edanz.com/english-editing-c) for editing a draft of this manuscript. We thank Dr. Yusuke Saijo (NAIST) and Dr. Shigetaka Yasuda (NAIST) for gifting ein2-1 seeds. We thank Dr. Kazuyuki Okazaki for gifting sugar beet seeds. Author contributions DT, KN, AM, and MS designed the research. DT, KN, AM, DTNQ, MT, CT, JI, and AJN performed the experiments. DT, KN, TT and ST analyzed the data. KN, TT and SI reviewed the data. DT and MS wrote the manuscript. Funding This work was supported by grants from RIKEN, Japan, including the RIKEN–AIST Joint Research Fund (full research) to MS, CREST (JPMJCR13B4), A-STEP (JPMJTM19BS), GteX (JPMJGX23B0), and ASPIRE (JPMJAP24A3) programs of the Japan Science and Technology Agency (JST) to MS. Data availability The raw RNA-seq data generated for this study have been deposited into the GenBank database with the accession numbers GSE297488, GSE297489 and GSE317383. Declarations Conflict of interest MS is Editor-in-Chief of PMB but was not involved in the evaluation of this manuscript. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary file1 (PPTX 88176 kb) Supplementary file2 (XLSX 11 kb) Supplementary file3 (XLSX 41 kb) Supplementary file4 (XLSX 19 kb) Supplementary file5 (XLSX 12 kb) Supplementary file6 (XLSX 71 kb) Data Availability Statement The raw RNA-seq data generated for this study have been deposited into the GenBank database with the accession numbers GSE297488, GSE297489 and GSE317383.