Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Acetate-mediated novel survival strategy against drought in plants

An Erratum to this article was published on 17 July 2017


Water deficit caused by global climate changes seriously endangers the survival of organisms and crop productivity, and increases environmental deterioration1,2. Plants' resistance to drought involves global reprogramming of transcription, cellular metabolism, hormone signalling and chromatin modification38. However, how these regulatory responses are coordinated via the various pathways, and the underlying mechanisms, are largely unknown. Herein, we report an essential drought-responsive network in which plants trigger a dynamic metabolic flux conversion from glycolysis into acetate synthesis to stimulate the jasmonate (JA) signalling pathway to confer drought tolerance. In Arabidopsis, the ON/OFF switching of this whole network is directly dependent on histone deacetylase HDA6. In addition, exogenous acetic acid promotes de novo JA synthesis and enrichment of histone H4 acetylation, which influences the priming of the JA signalling pathway for plant drought tolerance. This novel acetate function is evolutionarily conserved as a survival strategy against environmental changes in plants. Furthermore, the external application of acetic acid successfully enhanced the drought tolerance in Arabidopsis, rapeseed, maize, rice and wheat plants. Our findings highlight a radically new survival strategy that exploits an epigenetic switch of metabolic flux conversion and hormone signalling by which plants adapt to drought.

A genetic screen of the drought stress response using mutants of histone-modifying enzymes identified mutants of Arabidopsis histone deacetylase HDA6 (refs 9,10) with strong drought tolerance without growth retardation under normal conditions (Fig. 1a,b, Supplementary Figs 1 and 2). Prior to drought stress treatments, the hda6 mutant plants showed similar growth to wild-type plants under well-watered conditions (Supplementary Fig. 1). HDA6 is a homologue of human HDAC1 and yeast RPD3 (ref. 11), and has wide-ranging functions, for example gene silencing1214. The hda6 mutants showed enhanced drought tolerance; however, the expression of drought and ABA-responsive genes, water loss and accumulation of osmolytes, myo-inositol and trehalose were similar to wild-type plants3,5,15,16 (Supplementary Fig. 3). Note that the hda6 mutant exhibited greatly reduced proline levels at day 14 of soil drying (Supplementary Fig. 3). Herein, we show that HDA6 has a fundamental role in regulating a novel drought response pathway.

Figure 1: HDA6 correlates with the activation of the acetate biosynthetic pathway, which is essential for plant drought tolerance.

a, Drought tolerance of hda6 mutants (hda6). b, Survival rate of hda6 in the drought stress test. Asterisks indicate significant differences between wild-type (WT) and hda6 plants (mutants axe1-5 and sil1 and their respective WT plants, DR5 (Col-0 background) and Ler (Landsberg erecta) were used). c, Temporal changes in expression profiles of glycolysis and acetate fermentation pathway genes during drought treatment. Boxes indicate relative expression changes in genes encoding enzymes (Supplementary Table 2). d, Drought sensitivity of WT plants (ecotype Nossen) for acetate biosynthesis genes, pdc1 and aldh2b7 mutants. e, Survival rate of pdc1 and aldh2b7 mutants during the drought stress test. Asterisks indicate significant differences between WT and mutants. All error bars denote the s.d., n = 3. *P < 0.01 (Student's t-test).

To identify drought tolerance genes that are regulated by HDA6, we first profiled the genome-wide expression patterns of hda6 during drought stress. The analysis showed that the acetate biosynthesis pathway was upregulated in wild-type plants under drought stress and that this upregulation was greatly enhanced in the hda6 mutant (Fig. 1c, Supplementary Table 1 and 2). The acetate biosynthesis pathway mediated by pyruvate decarboxylase PDC1 and acetaldehyade dehydrogenase ALDH2B7, correlated with anoxia in Arabidopsis17. PDC1 is the key regulator initiating the first step to redirect the metabolic flux from pyruvate in glycolysis to acetaldehyde for the fermentation process under anoxia. ALDH2B7 detoxifies acetaldehyde into acetate and is upregulated during the recovery from anoxia. The expression levels of PDC1 and ALDH2B7 are repressed in the absence of stress, and were gradually increased (PDC1, by eightfold; ALDH2B7, by threefold) during drought conditions in wild-type plants (Supplementary Fig. 4). In hda6, transcripts of both genes accumulated during the early stages of the drought response, suggesting that HDA6 acts as a negative regulator of their expressions under non-stressed conditions. We also analysed the changes in transcript and metabolite levels of glycolysis and its connective pathways (Fig. 1c and Supplementary Fig. 5). Remarkably, the transcription of glycolytic genes was downregulated by drought in wild-type and hda6 plants (Fig. 1c and Supplementary Table 2). From the results of metabolomic analysis, the accumulation of glucose was detected in both wild-type and hda6 mutant plants under drought condition. In contrast, fructose-6-phosphate level was decreased at day 14 of soil drying in hda6 mutant compared with wild-type plants. On the TCA cycle, similarly 2-oxoglutarate and isocitrate levels were reduced in hda6 mutant at day 14 of soil drying (Supplementary Fig. 5). The downregulation of glycolysis and the activation of acetate synthesis under drought indicated that PDC1 and ALDH2B7 genes function not only in anoxia but also in response to drought. To verify the role of PDC1 and ALDH2B7 in the drought response, we tested the sensitivity of their mutants to drought. Indeed, both pdc1 and aldh2b7 (ref. 18) mutants showed severe sensitivity to drought stress (Fig. 1d,e and Supplementary Fig. 6). In addition, PDC1 and ALDH2B7 double overexpression plants exhibited enhanced drought tolerance (Supplementary Fig. 7).

To determine how HDA6 regulates PDC1 and ALDH2B7 expression, we examined the HDA6 binding activity and the corresponding histone modification status on these loci using a chromatin immunoprecipitation assay (Fig. 2a,b). The results showed that HDA6 binds to the PDC1 and ALDH2B7 loci in wild-type plants, predominantly at the transcribed gene body regions. Notably, the HDA6 binding levels dramatically decreased after drought treatment. This decrease in HDA6 binding mirrored histone H4 acetylation (H4Ac) changes (Fig. 2c,d): H4Ac levels were low in wild-type plants under normal conditions but were greatly increased during the drought response. These changes indicated that the dissociation of HDA6 triggers increased H4Ac and transcriptional upregulation of these loci. In hda6, the H4Ac enrichment occurred earlier and at higher levels than in wild-type plants. Altogether, these results demonstrated that HDA6 represses the acetate biosynthetic pathway directly under normal conditions and that HDA6 dissociates from these regions to trigger the epigenetic and transcriptional changes that activate this pathway under drought stress.

Figure 2: HDA6 directly regulates acetate biosynthetic pathway genes, PDC1 and ALDH2B7.

a, Graphical representation of changes in HDA6 binding/histone H4 acetylation (H4Ac) sites (underlined) on the PDC1 and ALDH2B7 genes, obtained by a chromatin immunoprecipitation assay (ChIP) assay. b, Enrichment of HDA6 binding determined on the PDC1 and ALDH2B7 gene regions by ChIP assays. Drought stress treatment caused a decrease in HDA6 binding to PDC1 and ALDH2B7 in the WT. Asterisks indicate significant differences between non-stressed (0 day) and drought stress conditions (6 and 12 days). c,d, Enrichment of H4Ac on the PDC1 and ALDH2B7 genes under drought stress conditions in WT and hda6 plants. At2g15555 and At1g55750 genes are positive and negative control sites for HDA6 binding and histone H4 acetylation enrichments, respectively13. Grey asterisks indicate significant differences between non-stressed and drought stress conditions. Red asterisks indicate significant differences between WT and hda6. All error bars denote the s.d. The results of independent replicates were averaged (n = 3). *P < 0.01 (Student's t-test).

We next asked whether the activation of PDC1 and ALDH2B7 regulated by HDA6 increases the endogenous acetate levels during drought conditions (Fig. 3a). Mass spectrometry analysis showed that the acetate content increased substantially during the drought response in wild-type plants. In contrast, pdc1 and aldh2b7 did not show any significant increase in acetate levels. Therefore, the endogenous acetate in wild-type plants actually increases in response to drought, and that both PDC1 and ALDH2B7 are required for acetate production. In hda6, acetate accumulated compared with wild-type plants. Acetate abundance correlated with increased plant drought tolerance: in hda6, the acetic acid levels were higher than in wild-type plants and survival was enhanced, but the acetate biosynthesis-deficient pdc1 and aldh2b7 were more sensitive to drought. This is the first indication that acetate biosynthesis is essential for plant drought tolerance. To confirm this conclusion, the drought tolerance of plants pretreated with exogenous acetic acid was assessed. Plants grown in the presence of 10, 20 and 30 mM acetic acid exhibited strikingly increased drought tolerance (Supplementary Figs 8 and 9a), in contrast to other organic acids (Fig. 3b, Supplementary Figs 10 and 11); supporting the specific role of acetic acid in this process.

Figure 3: Acetic acid promotes jasmonate signalling and histone acetylation.

a, Measurement of endogenous acetic acid contents in WT and mutant plants under drought stress conditions after 14 days. b, The effect of several organic acids on plant drought tolerance after 14 days. The lower panel shows the survival rate of WT plants during the drought stress test with several organic acids. c, Clustering display of genes regulated differentially by water or acetic acid and drought-continuous treatments in the WT. Three hundred and fifty-seven genes were highly induced by drought treatment in the WT pretreated with acetic acid. d, Drought sensitivity of the coi1-16B mutant. The lower panel shows the survival rate of the WT and coi1-16B mutant during the drought stress test. e, Measurement of plant hormones, JA, JA-Ile and ABA in WT plants treated with acetic acid. Transient synthesis of JA and JA-Ile by acetic acid treatment was detected on the first day. f, Incorporation of exogenous 14C-AA into the histone H4 protein. Semi-purified histone proteins (50 ng) were loaded in each lane. The incorporation of a substrate including a 14C radiotracer derived from 14C-AA is shown as a 14C image over a time course (0, 1, 3 and 6 h). WB, western blotting analysis using anti-histone H4 antibody; SDS–PAGE, semi-purified histones stained with Coomassie brilliant blue. The error bars denote s.d., n = 6 in a, n = 3 in bd and n = 4 in e. *P < 0.01 (Student's t-test).

The drought tolerance promoted by acetic acid was studied further by comparing transcriptional regulation during drought stress between acetic acid-pretreated and water-pretreated wild-type plants. Microarray analyses showed that 357 genes were induced under drought, specifically in the acetic acid-pretreated plants (Fig. 3c and Supplementary Table 3). Gene ontology enrichment analysis showed that the categories of JA and ethylene responses with defence/immune processes were enriched among these genes (Supplementary Table 4), in agreement with the partial, constitutive JA phenotype of hda6 (ref. 19). Previously, extensive crosstalk was reported between the JA, ethylene and defence response pathways20,21. To determine the functional downstream regulators of the acetic acid-dependent drought tolerance, the drought sensitivity of mutants for the JA receptor COI1 (ref. 22), the JA biosynthetic enzyme AOS (ref. 23) and the essential positive regulator of ethylene responses EIN2 (refs 24,25) were tested. The aos mutant26 showed drought sensitivity (Supplementary Fig. 12). The coi1-16B mutant27 showed drought sensitivity, and acetic acid pretreatment did not ameliorate this phenotype (Fig. 3d and Supplementary Fig. 13). In contrast, the ein2-5 mutant24 did not show altered sensitivity (Supplementary Fig. 9b). Thus, COI1 is essential to mediate acetic acid-dependent drought tolerance. A physical interaction between HDA6 and COI1 proteins using a yeast two-hybrid system was reported19. It suggests that HDA6 regulates JA signalling pathways cooperatively with COI1 and the chromatin status of COI1 target gene regions might be influenced. We next measured the JA levels in wild-type plants treated with acetic acid. Transient biosynthesis of JA and jasmonoyl–isoleucine (JA-Ile) was greatly induced by acetic acid treatment at 24 h, but abscisic acid (ABA) levels remained unchanged (Fig. 3e). The expression of the JA biosynthetic enzyme AOC3 (ref. 28) was induced by acetic acid (Supplementary Fig. 9c). Despite the transient JA increase during acetic acid treatment, the downstream genes of JA signalling, such as the key regulator MYC2 (ref. 29), were not transcriptionally activated (Supplementary Fig. 9d and Supplementary Table 3). However, these genes, involving MYC2 showed higher induction after the drought stimulus in the acetic acid-pretreated plants than in plants pretreated with water. Thus, it suggested that acetic acid primes the activation of a COI1-mediated signalling pathway and the transient production of JA might define the downstream target genes. The other acids, formic acid and lactic acid, also affected AOC3 gene induction; however, these acids could not enhance drought tolerance (Supplementary Figs 10 and 14).

To understand how acetic acid functions in priming gene activation, we monitored the behaviour of acetic acid in wild-type plants. In human cells, exogenous acetic acid treatment promotes intracellular pH increase and elevated histone acetylation associated with cell proliferation30. We detected an analogous pH upshift in Arabidopsis (Supplementary Fig. 15): by adding 10 mM acetic acid to the soil, the pH level of the xylem sap solution increased, which agreed with the observed pH increase of the xylem sap in plants responding to drought31. When 14C-labelled acetic acid (14C-AA) was absorbed through the roots, 14C-AA became evident in the shoots after 1 h (Supplementary Fig. 15). The total amount of absorbed 14C-AA was 500 nmol per shoot, which was approximately fivefold higher than in hda6 under drought conditions (Fig. 3a and Supplementary Fig. 15b,c). From the results of metabolomics analysis using plants treated with 10 mM acetic acid for five days, there was no difference of accumulation in 88 kinds of metabolites compared with plants treated with water for five days (Supplementary Table 5). It suggested that even though acetic acid with a high concentration of 10 mM is given, the amount absorbed by plants is not so high (Supplementary Fig. 15). It seems that the acetic acid absorption by plants correlates to soil pH (Supplementary Fig. 16). Up to 20 mM acetic acid concentration, the soil pH was kept in pH 7. However, the soil pH was shifted down to lower pH by addition of excessed 50 mM acetic acid. In connection with these changes in pH, plant growth seems to be inhibited by the treatment with 50 mM acetic acid (Supplementary Fig. 8 and 16). Interestingly, 14C-AA was incorporated into proteins with a molecular size corresponding to histone H4, detected by western blotting analysis after 3 h (Fig. 3f). Our results suggested that the acetic acid is converted to acetyl-CoA and is used as a substrate for histone acetylation. In addition, from the results of ChIP-seq analysis, histone H4 acetylation (H4ac) of the gene body region was enriched genome-wide by addition of 10 mM acetic acid for nine days (Supplementary Fig. 17 and Supplementary Table 6). The 106 genes, including MYC2, considered to be downstream target genes of the acetic acid–JA pathway to confer plant drought tolerance were involved within the H4ac-enriched genes (Supplementary Table 6). Herein, we propose that in Arabidopsis, the activation of the acetate biosynthetic pathway and the resulting acetate increase are indispensable for acquiring drought tolerance. We considered that HDA6 acts as an ON/OFF switch that controls this pathway directly. The alteration of metabolic flux in response to drought is consistent with the results of transcriptome analyses (Fig. 4a). Our phylogenetic analyses showed that ALDH2B7 is conserved widely from fungi to humans (Fig. 4b and Supplementary Fig. 18). In contrast, PDC1 is only conserved in moss, fungi and plants. In yeast, the acetaldehyde dehydrogenase4 mutant shows oxidative stress sensitivity under glucose starvation, and this sensitivity is recovered by supplying exogenous acetic acid32. As drought also induces oxidative stress in plants3,4, taken together, these results indicated that this novel stress-responsive mechanism is conserved in autotrophic eukaryotes. To test whether acetic acid could enhance the drought tolerance of plants other than Arabidopsis, we assessed its effect when supplied exogenously. We found that concentrations between 20 and 30 mM enhanced drought tolerance in both monocots and dicots, such as rice, wheat, maize and rapeseed plants (Fig. 4c–f and Supplementary Fig. 18). In conclusion, acetate, a basic and simple biochemical compound, has a significant role as an initial factor that orchestrates plants' survival capability, which connects fundamental metabolism, epigenetic regulation and hormone signalling, ultimately conferring plant drought tolerance. Integrative studies of transcription factors and histone modifications will reveal the undiscovered players in the acetic acid-mediated network for environmental adaptation.

Figure 4: Proposed mechanism for acetic acid-induced drought tolerance and its evolutionary conservation.

a, HDA6 acts as a switch during drought stress to redirect the metabolic flux, leading to the accumulation of acetic acid through the activation of specific biosynthetic regulatory genes and the JA signalling pathway in Arabidopsis. The histone acetylation derived from acetate might function in the priming effect of the COI1-mediated JA signalling pathway. b, Schematic representation of the conservation of PDC and ALDH2 homologous genes in eukaryotes (see also Supplementary Fig. 18). cf, Acetic acid confers drought tolerance on maize, rice, wheat and rapeseed plants.


Plant materials and growth conditions

A. thaliana (L.) Heynh plants were used in this study. The hda6 mutants axe1-5 and sil1, and their respective wild types, DR5 (Col-0 background) and Ler (Landsberg erecta)9,10 were used for the drought tolerance tests. The axe1-5 (hda6) mutant and its parental DR5 plants (wild type) were used for other experiments. The pdc1 (No. 13-1034-1) and aldh2b7 (No. 15-2168-1) mutants, and their DS donor line (ecotype: Nossen)18, were used for the drought sensitivity tests. In addition, coi1-16B, aos and ein2-5 mutants2224 were also used for drought sensitivity tests. Seeds were sown in soil (Professional soil No. 2, Dio Chemicals) in 9-cm-diameter pots, stratified for two to four days at 4 °C in the dark, and then grown under a 16 h light/8 h dark regimen at 22 °C.

Wild-type wheat plants (Triticum aestivum cv. Chinese Spring) were used. Seeds were germinated in Petri dishes on filter paper saturated with water for three days at 4 °C in the dark and then grown for two days at 21 °C. Seedlings were transplanted into soil (Golden planting mix, IRIS Ohyama) in 9-cm-diameter pots under a 16 h light/8 h dark regimen at 21 °C.

Wild-type rice plants (Oryza sativa L. cv. Nipponbare) were used. Seeds were germinated on soil (Bon Sol No.2, Sumitomo Chemical) in pots (6 cm × 6 cm × 4.5 cm) and grown under a 14 h light/10 h dark regimen at 30 °C.

Seeds of maize plants (Zea mays L. cv. Canberra) were germinated on soil (Hanazanmai (Sakata Seed)/Gardening soil No. 1 (Nihon Hiryo): 3:1) in pots (6 cm × 6 cm × 4.5 cm) and the plants were grown for nine days under a 14 h light/10 h dark regimen at 30 °C.

Seeds of rapeseed (Brassica napus cv. Wester) were sown on soil (Professional soil No. 2, Dio Chemicals) in 9-cm-diameter pots, stratified for two to four days at 4 °C in the dark, and then grown under a 16 h light/8 h dark regimen at 22 °C.

The mutant lines of histone modifiers used for initial genetic screen are listed in Supplementary Table 7.

Drought tolerance tests

Watering of three-week-old Arabidopsis plants was withheld for approximately two weeks under a 16 h light/8 h dark regimen at 22 °C, and 40% humidity in the greenhouse. The plants were then rewatered, and the number of surviving plants was counted after five days. The average survival rate was calculated for three independent experiments (n = 40). Pretreatment of Arabidopsis plants with acids or salts was performed as follows: when plants were two weeks old, a solution of organic acids (10 mM of hydrochloric acid, formic acid, acetic acid, butyric acid, lactic acid or citric acid) or 0, 1, 10, 20, 30 and 50 mM acetic acid, was supplied to the soil and the plants were grown for nine days. After removing the treatment solutions (administered from the bottom of the pot by capillarity using a paper towel), plants were subjected to the drought stress test. The average survival rate was calculated for three independent experiments (n = 48).

Four-week-old wheat plants were pretreated with 0, 20 and 50 mM acetic acid (supplied to the soil) and grown for nine days under a 16 h light/8 h dark regimen at 22 °C, and 40% humidity in the greenhouse. After removing the treatment solutions as above, plants were subjected to the drought stress test. The plants were then rewatered for five days to determine plant survival. The average survival rate was calculated for three independent experiments (n = 15).

Two-week-old rice plants were pretreated with 0, 20, 30 and 50 mM acetic acid supplied to the soil and plants were grown on soil for 4 days under a 14 h light/10 h dark regimen at 30 °C in a growth chamber. Removal of treatment solutions and drought treatment (four days) was performed as described above. The plants were then rewatered for ten days to determine plant survival. The average survival rate was calculated for three independent experiments (n = 8).

Ten-day-old maize plants were used for the drought test. Water in the soil was removed from the bottom of the pot by capillarity using a paper towel and the plants were then treated with 0, 10, 20, 30 and 50 mM acetic acid for four days. The acetic acid solution was removed using a paper towel and the plants were incubated under drought conditions for six days. After rewatering for five days, their survival was checked. The average survival rate was calculated for three independent experiments (n = 4).

One-month-old rapeseed plants were pretreated with 0, 20, 30 and 50 mM acetic acid supplied to the soil and the plants were grown on soil for nine days under a 16 h light/8 h dark regimen at 22 °C, and 40% humidity in the greenhouse. After removing the treatment solutions, plants were subjected to the drought stress test. The plants were then rewatered for five days to determine plant survival. The average survival rate was calculated for three independent experiments (n = 12).

RNA preparation

Total RNA was extracted from the aerial parts of Arabidopsis plants, either before or after drought stress, using the Plant RNA Purification Reagent (Invitrogen) according to the manufacturer's instructions. For quantitative real-time reverse transcription polymerase chain reaction (qRT–PCR) and reverse transcription polymerase chain reaction (RT–PCR), cDNA was synthesized using the QuantiTect Reverse Transcription synthesis kit (Qiagen). The primers are listed in Supplementary Table 8.

Microarray analysis

Three-week-old Arabidopsis plants were subjected to drought stress treatment, and the aerial parts were collected after 0, 6, 9, 12 and 15 days. Three biological replicates were performed. Microarray analysis was performed using an Agilent Arabidopsis microarray platform, as described previously13. The microarray data are available on the GEO website (GEO ID: GSE46365). Drought stress-responsive genes were selected using the following criteria: expression level change >1.5 fold and an unpaired t-test false discovery rate (FDR) of <0.05. At each time point, upregulated genes in hda6 were compared with their levels in wild-type plants. Two-week-old wild-type plants (Col-0) were treated with 10 mM acetic acid solutions for nine days, and then subjected to drought stress treatment. The aerial parts were collected after 0, 6, 9, 12 and 15 days of drought stress and used for microarray analysis. Three biological replicates (n = 12) were taken. The microarray data are available on the GEO website (GEO ID: GSE46524). Acetic acid-responsive genes were selected using the following criteria: the genes showed an FDR less than 0.05 at more than one time point, when water-drought treated wild-type plants series and acetic acid-drought treated wild-type plants series were compared at each time point, using an unpaired t-test and the FDR method33 in the R 2.12.1 software. The acetic acid responsive genes were hierarchically clustered for heatmap display.

Clustering of expression patterns in the microarray data

The gene expression heatmap was obtained using heatmap.2 in the gplots package of R ver. 2.1.12 (R Core Team). Hierarchical gene clusters were built using Ward's minimum variance method clustering with Euclidean distance, after which the log2 normalized values were transformed into Z-scores. The heatmap colouring reflects the rank of the Z-scores.

Chromatin immunoprecipitation

Arabidopsis chromatin immunoprecipitation (ChIP) assays were performed essentially as described previously34, with three biological replicates. The antibodies used in this study were an anti-HDA6 antibody13 and an anti-H4 tetra-acetylation (06-866) from Millipore. Precipitated DNA was analysed using quantitative PCR (Power SYBR real time reagent and StepOnePlus; Applied Biosystems). We estimated the absolute fraction of DNA recovered from the INPUT (% input DNA) by comparing the reaction threshold cycle of the ChIP sample to a dilution of its own INPUT. Statistical significance was determined using Student's t-test. At1g55750 was used as a negative control for HDA6 binding, and At2g15555 was used as a positive control for HDA6 binding and enrichment of histone H4 acetylation13. The primers used are listed in Supplementary Table 8.

ChIP-seq analysis

ChIPed DNA (250 ng) was used to make sequencing library for a SOLiD 5500 sequencer (Thermo Fisher). Library preparation and sequencing were performed as described in instruction manuals. Sequences were mapped on the TAIR10 genome using bowtie (version 0.12.8) with default options. Since the software had problems with our SOLiD outputs, paired-end sequences were aligned as individual files and aggregated after alignment of either file. Peaks were identified using MACS2 (version 2.1.0) with nomodel option. Whole-genome histone acetylation was evaluated using the normalized read count of all peaks. Mapped reads were aggregated for every 25 bp bin of a peak and statistically evaluated between treated and untreated organisms with the Wilcoxon test. Peaks with q-values <0.05 and a logarithm of fold change (logFC) >0.25 were identified as significant peaks. Associated genes with the peaks of acetylated regions and differentially acetylated peaks were enumerated using our in-house program and gene annotation file provided by the TAIR database. Since the majority of acetylated histones are located downstream of TSS, we identified acetylated genes responding to acetic acid as genes having peaks from 500 bp upstream of TSS to 1.5 kb downstream of TSS. All raw files and processed files were deposited at the NCBI GEO (ID: GSE95817). The Integrative Genome Browser 9.0.0 was used to visualize the processed data ( Sequencing was done using each ChIPed DNA (n = 1) and result of H4ac enrichments was validated by ChIP-qPCR (n = 3) on three target gene regions (NHL3, ZAT10 and At5g42050), a negative gene region (At2g15555) and a positive gene region (UBQ10).

Measurement of acetate

Three-week-old Arabidopsis plants were subjected to a drought stress treatment for 12 (Nos background plants) or 14 days (Col-0 background plants), and rosette leaves were harvested. Well-watered plants of the same age were used as controls. Harvested leaves were immediately weighed and frozen in a plastic tube. Frozen tissues were homogenized in five volumes (w/v) of cold (4 °C) water using a mixer mill (Shake Master Neo; Bio Medical Science) with a zirconia bead for 1 min at 20 Hz. Following centrifugation at 15,000g for 1 min at 4 °C, the supernatants were diluted with an equal amount of acetone. The sample extracts (1 µl) were analysed using a gas chromatography–mass spectrometry (GC–MS) system (GCMS-QP2010 Plus) equipped with a nitroterephthalic acid modified polyethylene glycol (DB-FFAP) column (60 m, 0.25 mm internal diameter, 0.5 µm film thickness; Agilent Technologies). Helium was used as the carrier gas at a flow rate of 1 ml min–1. The oven temperature was held at 100 °C for 5 min and then heated at 10 °C min–1 to 230 °C. The injector temperature was maintained at 250 °C, and the ion source and interface temperatures were 230 and 250 °C, respectively. The sample was injected in split injection mode with a 100:1 split ratio. Acetate was detected using the selected ion monitoring mode at m/z 60.

Metabolite profiling

Metabolite profiling was performed as described previously35, with slight modifications. All chemicals and reagents used were of spectrometric grade. We extracted 25 mg fresh weight (FW) of aerial plant parts per millilitre of extraction medium containing ten stable isotope reference compounds. After centrifugation, a 200 µl aliquot of the supernatant was drawn and evaporated to dryness for derivatization, as described previously35. The equivalent of 56 µg FW of extract was analysed by gas chromatography with time-of-flight mass spectrometry (GC–TOF–MS). (LECO, St Joseph, MI, USA). Data were normalized using the cross-contribution compensating multiple standard normalization (CCMN) algorithm36.

Statistical data analysis of metabolomics data

Statistical data analyses were performed by R statistical software ( Statistical comparison of the metabolome data with log2 transformation between acetate treatment and control was conducted using a Welch's t-test with correction by false discovery rate (FDR)33.

Measurement of xylem sap pH

Xylem sap was collected from six-week-old Arabidopsis plants grown in soil. Before and after acetic acid treatment, shoots corresponding to the hypocotyl of 15–20 plants were cut off. Exudates from the cut ends were collected for 30 min. Approximately 10 µl of the exudate was subjected to pH determination using a pH meter (B-712).

Measurement of soil pH and water content

Ten grams of soil (Professional No. 2) were treated for nine days with 25 ml of an acetic acid–deionized water solution in a 50 ml plastic tube and vigorously shaken for 30 min at 200 rpm with a Shaker NR-30 (Taitec), and then kept at room temperature for 30 min. The pH of the supernatant was measured with a pH meter PH-6011A (Custom). Soil water contents were measured with a soil water content meter DM-18 (Takemura Denki).

Measurement of phytohormones

Approximately 100 mg (fresh weight) of the aerial parts of Arabidopsis plants were subjected to quantification of JA, jasmonoyl–isoleucine (JA-Ile) and abscisic acid (ABA). Extraction and fractionation were performed using the method described previously38. [2H2]JA-Ile was purchased from OlChemim Ltd. The hormones were determined using an ultra high-performance liquid chromatography-Q-Exactive™ system (Thermo Fisher) using an ODS column (Aquity UPLC BEH C18, 1.7 µm, 2.1 × 100 mm; Waters) as previously described36.

14C-Labelled acetic acid incorporation

Acetic acid transport to shoots was examined using 14C-labelled acetic acid. Ten millilitres of 10 mM acetic acid containing 12MBq of [1-14C]-acetic acid was supplied from the bottom of the pot containing 18-day-old seedlings. A shoot was excised at each treatment time and placed on cardboard at 4 °C. After being covered with a thin plastic film, the shoots were exposed to an imaging plate (BAS–IP-MS, GE Healthcare) for one day, and then the 14C image was acquired by an FLA-5000 image analyser (FujiFilm). The amount of acetic acid in the aerial part was calculated from the signal intensity, using imaging analysis software (Image Gauge ver. 4.0, FujiFilm).

Western blotting

Three Arabidopsis plants for each sample were fixed with liquid nitrogen and disrupted by shaking for 1 min using an SK mill (Tokken). Histone proteins were purified from the crude protein fraction using a Histone purification kit (Active motif). Protein concentrations were measured using a Qubit protein assay kit and Qubit 2.0 Fluorometer (Thermo Fisher). Histone proteins were separated by SDS–PAGE on a 15% gel (SuperSep Ace, 15%, Wako Pure Chemical Industries). Histone proteins were transferred to a polyvinylidene difluoride (PVDF) membrane using an iBlot Dry Blotting System (Thermo Fisher). To detect the histone H4 proteins, an antibody specific for the histone H4 C-terminus (ab10158, Abcam) was used as the primary antibody.

Development and characterization of transgenic lines expressing PDC1 and ALDH2B7 under control of the TSPO promoter

Arabidopsis TSPO (At2g47770) is a small intron less gene and its expression is up regulated by drought stress39. A 1.4 kb promoter region upstream of TSPO gene was cloned with forward and reverse primers as 5′-ATAGTAGACTGCGGCCGCACCATGGCA-3′ and 5′-TGTCCTGAGATCTAGATACAAACGAAACGTCCAAAAC-3′, respectively. A NotI site was introduced into the forward primer, while a XbaI site was introduced into the reverse primer and a 1.4 kb region of the Arabidopsis TSPO promoter was cloned into the pGreenII 0029 and pGreenII 0229 vectors. Then, the ALDH2B7 and PDC1 genes were cloned downstream of the TSPO promoter in pGreenII 0229 and pGreenII 0029, respectively. The primers used for the cloning of PDC1 were 5′-CTCGAATTCCCCGGGTTAATTAAATTAATC-3′ and 5′-ATCAGGCCGTCGACGCCGGATC-3′ containing SmaI and SalI sites, respectively. The primers used for the cloning of ALDH2B7 were 5′-AATCGGCCCCCGGGTTAATTAAATTAATC -3′ and 5′-TGGAGCTAGTCGACTTATGGCCGA-3′ containing SmaI and SalI sites, respectively. Transgenic Arabidopsis plants expressing both of these genes were developed through Agrobacterium tumefaciens (GV3101)-mediated transformation of plants by the floral dip method40. The both plasmids containing PDC1 or ALDH2B7 were transformed into Arabidopsis wild-type (Col-0) plants at one time. Neomycin phosphotransferase II (NPT II), and the bialaphos resistance (bar) genes were used to screen transgenic plants using kanamycin (km) and bialaphos for PDC1 and ALDH2B7, respectively.

Homozygous lines for plants expressing ALDH2B7, PDC1 or both were developed, and the drought-inducible expression of these genes was confirmed through RT–qPCR analysis. cDNA for each sample was synthesized from 200 ng RNA using the QuantiTect Reverse transcription kit according to the manufacturer's instructions (Qiagen). For PDC1, the forward and reverse primers were 5′-AGAGCACCATCCTCATTTCATT-3′ and 5′-TGGCTTTTTCTTTCTTGAGGAG-3′, respectively. Forward and reverse primers used for ALDH2B7 were 5′-CATGCTCATAGCAACAGACGA-3′ and 5′-CAGCAGCTAAACCGTACCTTG-3′, respectively. Actin 2, used as an internal control, was amplified using the forward and reverse primers 5′-TGAAGTGTGATGTGGATATCAGG-3′ and 5′-GATTTCTTTGCTCATACGGTCAG-3′, respectively.

For the drought treatment, seeds were sown in Dio propagation mix no. 2 professional soil (Dio Chemicals) at 22 °C (16 h light/8 h dark cycle, 60 µmol m−2 s−1 photon flux density). Two weeks after germination, plants were subjected to the drought treatment by removing excess water from the trays and ceasing any subsequent watering. The survival ratio of the wild-type and the transgenic plants was calculated by counting the number of survived plants at 48 h after rewatering the plants. For measurement of biomass, plants were grown in soil for two weeks and shoots were subsequently harvested and fresh weight was measured. The samples were then subjected to heat drying at 55 °C for five days and sample dry weights were measured. The weight of pots containing two-week-old plants was recorded and changes in the pot weight was expressed as percentage change in soil water content.

Data availability

The transcriptome and epigenome data are available at the GEO website under the accession numbers GSE46365, GSE46524 and GSE95817. The data that support the findings of this study are available from the corresponding authors upon request.

Additional information

How to cite this article: Kim, J.-M. et al. Acetate-mediated novel survival strategy against drought in plants. Nat. Plants 3, 17097 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1

    Schiermeier, Q. Water: a long dry summer. Nature 452, 270–273 (2008).

    CAS  Article  Google Scholar 

  2. 2

    Lesk, C., Rowhani, P. & Ramankutty, N. Influence of extreme weather disasters on global crop production. Nature 529, 84–87 (2016).

    CAS  Article  Google Scholar 

  3. 3

    Xiong, L., Schumaker, K. S. & Zhu, J. K. Cell signalling during cold, drought, and salt stress. Plant Cell 14, S165–S183 (2002).

    CAS  Article  PubMed  Google Scholar 

  4. 4

    Urano, K. et al. Characterization of the ABA-regulated global responses to dehydration in Arabidopsis by metabolomics. Plant J. 57, 1065–1078 (2009).

    CAS  Article  Google Scholar 

  5. 5

    Pei, Z. M., Ghassemian, M., Kwak, C. M., McCourt, P. & Schroeder, J. I. Role of farnesyltransferase in ABA regulation of guard cell anion channels and plant water loss. Science 282, 287–290 (1998).

    CAS  Article  Google Scholar 

  6. 6

    Mirouze, M. & Paszkowski, J. Epigenetic contribution to stress adaptation in plants. Curr. Opin. Plant Biol. 14, 267–274 (2011).

    CAS  Article  Google Scholar 

  7. 7

    Clauw, P. et al. Leaf responses to mild drought stress in natural variants of Arabidopsis. Plant Physiol. 165, 800–816 (2015).

    Article  Google Scholar 

  8. 8

    Ahmad, R. et al. Jasmonates: multifunctional roles in stress tolerance. Front. Plant Sci. 7, Article 813 (2016).

    Google Scholar 

  9. 9

    Murfett, J., Wang, X. J., Hagen, G. & Guilfoyle, T. J. Identification of Arabidopsis histone deacetylase HDA6 mutants that affect transgene expression. Plant Cell 13, 1047–1061 (2001).

    CAS  Article  PubMed  Google Scholar 

  10. 10

    Aufsatz, W., Mette, M. F., van der Winden, J., Matzke, M. & Matzke, A. J. HDA6, a putative histone deacetylase needed to enhance DNA methylation induced by double-stranded RNA. EMBO J. 21, 6832–6841 (2002).

    CAS  Article  PubMed  Google Scholar 

  11. 11

    Pandey, R. et al. Analysis of histone acetyltransferare and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Res. 30, 5036–5055 (2002).

    CAS  Article  PubMed  Google Scholar 

  12. 12

    Earley, K. W. et al. Mechanisms of HDA6-mediated rRNA gene silencing: suppression of intergenic Pol II transcription and differential effects on maintenance versus siRNA-directed cytosine methylation. Genes Dev. 24, 1119–1132 (2010).

    CAS  Article  PubMed  Google Scholar 

  13. 13

    To, K. T. et al. Arabidopsis HDA6 regulates locus-directed heterochromatin silencing in cooperation with MET1. PLoS Genet. 7, e1002055 (2011).

    CAS  Article  PubMed  Google Scholar 

  14. 14

    Luo, M. et al. Histone deacetylase HDA6 is functionally associated with AS1 in repression of KNOX genes in Arabidopsis. PLoS Genet. 8, e1003114 (2012).

    CAS  Article  PubMed  Google Scholar 

  15. 15

    Kavi Kishor, P. B., Hong, Z., Miao, G.-H., Hu, C.-A.A. & Verma, D. P. S. Overexpression of Δ1-pyrroline-5-carboxylate synthetase increases proline production and confers osmotolerance in transgenic plants. Plant Physiol. 108, 1387–1394 (1995).

    Article  Google Scholar 

  16. 16

    Holmström, K. O. et al. Drought tolerance in tobacco. Nature 379, 683–684 (1996).

    Article  Google Scholar 

  17. 17

    Kürsteiner, O., Dupuis, I. & Kuhlemeier, C. The pyruvate decarboxylase1 gene of Arabidopsis is required during anoxia but not other environmental stresses. Plant Physiol. 132, 968–978 (2003).

    Article  PubMed  Google Scholar 

  18. 18

    Kuromori, T. et al. A collection of 11800 single-copy Ds transposon insertion lines in Arabidopsis. Plant J. 37, 897–905 (2004).

    CAS  Article  Google Scholar 

  19. 19

    Devoto, A. et al. COI1 links jasmonate signalling and fertility to the SCF ubiquitin-ligase complex in Arabidopsis. Plant J. 32, 457–466 (2002).

    CAS  Article  Google Scholar 

  20. 20

    Kazan, K. Diverse roles of jasmonates and ethylene in abiotic stress tolerance. Trends Plant Sci. 20, 219–229 (2015).

    CAS  Article  Google Scholar 

  21. 21

    Balvi, V. & Devoto, A. Jasmonate signalling network in Arabidopsis thaliana: crucial regulatory nodes and new physiological scenarios. New Phytol. 177, 301–318 (2008).

    Google Scholar 

  22. 22

    Thines, B. et al. JAZ repressor proteins are targets of the SCF(COI1) complex during jasmonate signaling. Nature 448, 661–665 (2007).

    CAS  Article  Google Scholar 

  23. 23

    Park, J. H. et al. A knock-out mutation in allene oxide synthase results in male sterility and defective wound signal transduction in Arabidopsis due to a block in jasmonic acid biosynthesis. Plant J. 31, 1–12 (2002).

    Article  Google Scholar 

  24. 24

    Cheng, M. C., Liao, P. M., Kuo, W. W. & Lin, T. P. The Arabidopsis ETHYLENE RESPONSE FACTOR1 regulates abiotic stress-responsive gene expression by binding to different cis-acting elements in response to different stress signals. Plant Physiol. 162, 1566–1582 (2013).

    CAS  Article  PubMed  Google Scholar 

  25. 25

    Alonso, J. M., Hirayama, T., Roman, G., Nourizadeh, S. & Ecker, J. R. EIN2, a bifunctional transducer of ethylene and stress responses in Arabidopsis. Science 284, 2148–2152 (1999).

    CAS  Article  Google Scholar 

  26. 26

    von Malek, B., van der Graaff, E., Schneitz, K. & Keller, B. The Arabidopsis male-sterile mutant dde2-2 is defective in the ALLENE OXIDE SYNTHASE gene encoding one of the key enzymes of the jasmonic acid biosynthesis pathway. Planta 216, 187–192 (2002).

    CAS  Article  Google Scholar 

  27. 27

    Noir, S. et al. Jasmonate controls leaf growth by repressing cell proliferation and the onset of endoreduplication while maintaining a potential stand-by mode. Plant Physiol. 161, 1930–1951 (2013).

    CAS  Article  PubMed  Google Scholar 

  28. 28

    He, Y., Fukushige, H., Hildebrand, D. F. & Gan, S. Evidence supporting a role of jasmonic acid in Arabidopsis leaf senescence. Plant Physiol. 128, 876–884 (2002).

    CAS  Article  PubMed  Google Scholar 

  29. 29

    Chini, A. et al. The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448, 666–671 (2007).

    CAS  Article  Google Scholar 

  30. 30

    McBrian, M. A. et al. Histone acetylation regulates intracellular pH. Mol. Cell 49, 310–321 (2013).

    CAS  Article  Google Scholar 

  31. 31

    Wilikinson, S. & Davies, W. J. Xylem sap pH increase: a drought signal received at the apoplastic face of the guard cell that involves the suppression of saturable abscisic acid uptake by the epidermal symplast. Plant Physiol. 113, 559–573 (1997).

    Article  Google Scholar 

  32. 32

    Baron, J. A., Laws, K. M., Chen, J. S. & Culotta, V. C. Superoxide triggers an acid burst in Saccharomyces cerevisiae to condition the environment of glucose-starved cells. J. Biol. Chem. 288, 4557–4566 (2013).

    CAS  Article  Google Scholar 

  33. 33

    Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. Roy. Stat. Soc. Ser. B 57, 289–300 (1995).

    Google Scholar 

  34. 34

    Kim, J. M. et al. Highly reproducible ChIP-on-chip analysis to identify genome-wide protein binding and chromatin status in Arabidopsis thaliana. Methods Mol. Biol. 1062, 405–426 (2014).

    Article  Google Scholar 

  35. 35

    Kusano, M. et al. Unbiased characterization of genotype-dependent metabolic regulations by metabolomic approach in Arabidopsis thaliana. BMC Syst. Biol. 1, 53 (2007).

    Article  PubMed  Google Scholar 

  36. 36

    Redestig, H. et al. Compensation for systematic cross-contribution improves normalization of mass spectrometry based metabolomics data. Anal. Chem. 81, 7974–7980 (2009).

    CAS  Article  Google Scholar 

  37. 37

    Tamura, K., Stecher, G., Peterson, D., Filipski, A. & Kumar, S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725–2729 (2013).

    CAS  Article  PubMed  Google Scholar 

  38. 38

    Kojima, M. et al. Highly-sensitive and high-throughput analysis of plant hormones using MS-probe modification and liquid chromatography-tandem mass spectrometry: an application for hormone profiling in Oryza sativa. Plant Cell Physiol. 50, 1201–1214 (2009).

    CAS  Article  PubMed  Google Scholar 

  39. 39

    Guillaumot, D. et al. The Arabidopsis TSPO-related protein is a stress and abscisic acid-regulated, endoplasmic reticulum-Golgi-localized membrane protein. Plant J. 60, 242–246 (2009).

    CAS  Article  Google Scholar 

  40. 40

    Zhang, X., Henriques, R., Lin, S.-S., Niu, Q.-W. & Chua, N.-H. Agrobacterium-mediated transformation of Arabidopsis thaliana using the floral dip method. Nat. Protoc. 1, 641–646 (2006).

    CAS  Article  Google Scholar 

Download references


We thank C. Pikaard for the hda6 mutant seeds and T. Hirayama for the ein2-5 mutant seeds. This work was supported by RIKEN; the Japan Science and Technology Agency (JST), Core Research for Evolutionary Science and Technology (CREST) (grant no. JPMJCR13B4) to M.S.; and Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan, (Innovative Areas 23119522 and 25119724 to M.S.; Innovative Areas 24113523 and (C) 24570065 to J.M.K.). JST, PRESTO 15665950 to K.T. A.D. was supported by a Japan Society for the promotion of Science (JSPS) Invitation Fellowship for Research in Japan (L10551) and by a Royal Society International Joint Project (JP091348, to A.D and M.S).

Author information




J.M.K., T.K.T., and M.S. conceived the project. J.M.K., and T.K.T. designed the experiments. J.M.K. and T.K.T. carried out all drought stress and growth tests in Arabidopsis. J.M.K. performed the ChIP assay. T.K.T. and J.I. performed the qRT–PCR and RT–PCR expression analyses. J.M.K., K.T. and N.I.K. performed the radioactive incorporation assay. F.M. measured the acetic acid concentration by GC–MS. M.K., A.F. and K.S. carried out the metabolomic analyses. Y.T. and H.S. measured the phytohormone levels. J.I., M.T. and T.M. supported the microarray analyses. A.M. analysed the microarray data. S.M. and T.S. measured the xylem sap pH. J.M.K., D.O. and Y.H. carried out the drought stress test in rice and maize. J.M.K., M.A., H.T., K.K. and Y.O. carried out the drought stress test in wheat and rapeseed. T.A.E carried out data analysis for ChIP-seq. C.T. supported the management of plants and seeds. J.M.K., T.K.T. and A.D. identified the link with JA and conceived the experiments using mutants of the jasmonate signalling pathway genes. J.M.K., M.A., S.R. and K.B. analysed the transgenic plants expressing PDC1 and ALDH2B7. T.K., K.S. and A.D. supplied the pdc1, aldh2b7 and coi1-16B mutant seeds, respectively. J.M.K., T.K.T., A.D. and M.S. wrote, reviewed and edited the manuscript.

Corresponding authors

Correspondence to Jong-Myong Kim or Motoaki Seki.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Information

Supplementary Figures 1–17, Supplementary Table 4 and 8. (PDF 20829 kb)

Supplementary Table 1

Expression profiles of 248 genes whose expressions were higher in hda6 than in wild-type plants under drought stress conditions. (XLS 171 kb)

Supplementary Table 2

Expression changes in glycolysis and acetate fermentation pathway genes under drought treatment in wild-type plants and hda6 mutants. (XLSX 27 kb)

Supplementary Table 3

Drought-induced, highly expressed 357 genes in wild-type Arabidopsis plants pretreated with acetic acid. (XLS 189 kb)

Supplementary Table 5

Metabolite profiles of Arabidopsis plants exposed to acetate treatment. (XLSX 21 kb)

Supplementary Table 6

Expression profiles of 3,914 genes with histone H4 acetylation enriched by acetic acid treatment during acetic acid and drought treatments. (XLSX 3209 kb)

Supplementary Table 7

List of histone modifier gene mutants used in this study. (XLSX 12 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kim, JM., To, T., Matsui, A. et al. Acetate-mediated novel survival strategy against drought in plants. Nature Plants 3, 17097 (2017).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing