A Novel Arabidopsis Gene Required for Ethanol

Plant Cell Physiol. 45(6): 659–666 (2004) JSPP © 2004

Rapid Paper

A Novel Arabidopsis Gene Required for Ethanol Tolerance is Conserved Among Plants and Archaea Naoko Fujishige 1, Noriyuki Nishimura 1, Satoshi Iuchi 2, Takanori Kunii 1, Kazuo Shinozaki 3 and Takashi Hirayama 1, 3, 4 1

Graduate School of Integrated Science, Yokohama City University, Yokohama, 230-0045 Japan Biological Resource Center, RIKEN Tsukuba Institute, Tsukuba, Ibaraki, 305-0074 Japan 3 Plant Molecular Biology, RIKEN Tsukuba Institute, Tsukuba, Ibaraki, 305-0074 Japan 2

;

However, higher levels of ethanol can be toxic to living cells. The effect of ethanol on living cells has been studied for several decades. Ethanol treatment induces the accumulation of heat shock proteins in various organisms, including plants (Li and Hahn 1978, Sanchez et al. 1992, Kuo et al. 2000). In order to address the ethanol stress response, many ethanol-sensitive mutants have been isolated in budding yeasts. Analysis of these mutants revealed that genes involved in the heat shock response or cell wall integrity are required for ethanol tolerance (Aguilera and Benitez 1986, Piper 1995, Takahashi et al. 2001). The products of ethanol metabolism also affect living cells. Oxidation of ethanol by ADH produces NADH and acetaldehyde. Accumulation of NADH is thought to affect other reactions that produce NADH, and disturb the cellular oxide/redox balance to cause oxidative stress (Fernandez-Checa et al. 1997, Zhang et al. 2002). The other product, acetaldehyde, attacks various biological molecules. Accumulation of endogenous acetaldehyde is toxic to the mammalian cell (Clements et al. 2002). Overproduction of pyruvate decarboxylase, which converts pyruvate to acetaldehyde, in plant tissues conferred a higher-level accumulation of acetaldehyde and caused necrosis (Tadege et al. 1998). In addition, ethanol resistant tobacco mutants turned out to be ADH deficient (Rousselin et al. 1990). These observations indicate that acetaldehyde is also the major contributor to ethanol toxicity in plant cell. Previously, we reported a novel ethanol-hypersensitive mutant, geko1 (gek1), of Arabidopsis (Hirayama et al. 2004). The gek1 mutants showed a significantly enhanced sensitivity to ethanol, 10–100 times greater than that of the wild type. The enhanced sensitivity of gek1 seems restricted to ethanol and its metabolite, acetaldehyde, since gek1 responded normally to other alcohols, plant hormones, high temperature, salt, and anoxic stress. Genetic analysis revealed that the ethanolhypersensitive phenotype of gek1 is largely dependent on ADH activity, but ADH activity was not affected by the gek1 mutations. The GEK1 locus was mapped on the top of chromosome 2 where no ethanol metabolic gene has been reported. Endogenous acetaldehyde levels were not different between gek1 and the wild type even in the presence of ethanol, suggesting that

A novel ethanol-hypersensitive mutant, gek1, of Arabidopsis shows 10–100 times greater sensitivity to ethanol compared to the wild type, while it grows normally in the absence of ethanol, and responds normally to other alcohols and to environmental stresses such as heat shock and high salinity. Mapping of the gek1 locus indicated it is a previously unreported locus. In order to address the GEK1 function, we identified the GEK1 gene by means of mapbased cloning. The GEK1 gene encodes a novel protein without any known functional motifs. Transgenic Arabidopsis plants overexpressing GEK1 displayed an enhanced tolerance to ethanol and acetaldehyde, suggesting that GEK1 is directly involved in the tolerance to those chemicals. By contrast, expression of GEK1 in E. coli and yeasts did not increase their tolerance to ethanol or acetaldehyde. Interestingly, a similarity search revealed that GEK1-related genes are conserved only in plants and archaea. These results might suggest that plants, and presumably archaea, have a novel mechanism for protection from acetaldehyde toxicity. Keywords: Archaea — Arabidopsis — Ethanol hypersensitive mutant — Evolution — Transgenic plant. Abbreviations: ABA; abscisic acid; ACC, 1-aminocyclopropane1-carboxylate; ADH, alcohol dehydrogenase; ALDH, acetaldehyde dehydrogenase; EST, expressed sequence tag; GFP, green fluorescent protein; RT-PCR, reverse transcription-PCR. The nucleotide sequence reported in this paper has been submitted to the DDBJ Data Libraries under accession no. AB091252.

Introduction Organisms can use ethanol as a carbon source. Ethanol is oxidized to acetate through acetaldehyde by alcohol dehydrogenase (ADH) and acetaldehyde dehydrogenase (ALDH). Acetate is metabolized as acetyl-CoA in various metabolic pathways such as the citric acid cycle and the glyoxylate cycle. 4

Corresponding author: E-mail, [email protected]; Fax, +81-45-508-7363. 659

660

GEK1, a novel gene required for ethanol tolerance

the defect in GEK1 does not affect acetaldehyde metabolism. Taken together, it can be concluded that the defect in GEK1 somehow enhances the cellular sensitivity to acetaldehyde toxicity. To address the GEK1 function, we identified the GEK1 gene by map-based cloning and found that it is a novel gene. A similarity search revealed that GEK1-related genes are conserved only in plants and archaea. The overexpression of GEK1 conferred an enhanced tolerance to ethanol or acetaldehyde in plant tissues but not in E. coli or yeast cells. Our results suggest the existence of an unknown system, unique to plants and probably archaea, for protecting cells from acetaldehyde toxicity.

Results Identification of the GEK1 gene To examine how the gek1 mutation confers an enhanced sensitivity to ethanol, we identified the GEK1 gene by mapbased chromosomal walking (Fig. 1A). GEK1 was localized on BAC clone F19B11 on chromosome 2 (Arabidopsis Genome Initiative 2000). We isolated three independent gek1 alleles, gek1-1, gek1-2 and gek1-3 (Hirayama et al. 2004). In gek1-2, we found a deletion going from At2g03800 at nt1977 (counting from the A of the predicted first ATG codon as 1) to At2g03810, making an in-frame stop codon adjacent to the deletion site. Single base substitutions in gek1-1 (G135 to A) and gek1-3 (G1685 to A) were found in the same gene. In gek11, the base substitution causes an amino acid change from Met45 to Ile. The base substitution in gek1-3 disrupts the splicing acceptor site of the seventh intron (Fig. 1A). We constructed several transgenic gek1-2 plants carrying a genomic DNA segment containing At2g03800. All transgenic plants examined were tolerant to 0.1% (v/v) ethanol. From these results, we conclude that At2g03800 is GEK1. The exon-intron organization of GEK1 was confirmed by sequencing a GEK1 cDNA clone, although we could not isolate cDNA clones possessing the predicted first ATG codon. There is no other related gene in the sequence data of the Arabidopsis genome. To see the effect of the mutations on the transcription of GEK1, we performed two RT-PCR experiments to amplify the 5′-half or the entire length of the GEK1 coding region. In both experiments, the same amount of PCR product was detected in wild type and gek1-1. By contrast, in gek1-2, the PCR product for the entire coding region was hardly detected because of the deletion, and a reduced amount of that for the 5′-half was observed (Fig. 1B). In gek1-3, a reduced amount of the PCR product was obtained in both PCR experiments. Sequencing of the RT-PCR product obtained from gek1-3 revealed that G1715 is used as the splicing acceptor site of the seventh intron instead of G1695 (A in gek1-3), causing a frame-shift and the truncation of the open reading frame at 24 bp downstream of this pseudointron-acceptor site. These results suggest that gek1-2 and gek1-3 can produce truncated GEK1 proteins with reduced expression levels.

Fig. 1 Identification of the GEK1 gene by map-based gene cloning. (A) Schematic representations of the mapping and structure of GEK1 are shown. Numbers indicate the number of recombinant chromosomes found among the number of F2 progeny. The mutation sites of the gek1 alleles are shown on the lowest bar. “atg” indicates the first ATG codon that is predicted by the Aguilera and Benitez 1986. Small open triangles indicate the position of corresponding sequences for oligonucleotide primers used in RT-PCR experiments shown in panel B. (B) Expression of GEK1 in the mutants. RT-PCR was performed to amplify the 5′ half of, or all of the GEK1 coding region using oligonucleotides F19-25.1-F19-25.11 or F19-25.1-F19-25.R as primers, respectively. PCR products obtained from two independent RNA preparations are shown on each line. β-tubulin was used as a control.

GEK1 encodes a novel protein conserved in plants and archaea The GEK1 gene could encode a polypeptide consisting of 361 amino acids according to the gene prediction. The predicted GEK1 polypeptide has no putative motifs described before. Interestingly, a similarity search against public databases revealed that GEK1 has significant similarity to proteins predicted from archaea genome sequences (Fig. 2) and ESTs of various plant species, including tomato, rice, maize, wheat, and barley. According to the database, the genomes of 17 archaea have been sequenced already. Among them, 16 species carry GEK1-related genes (Methanothermobacter thermoautotrophicus is the only exception). There are no related genes in other organisms whose genomes have been sequenced so far, including E. coli, Synechocystis PC6803, Giardia, S. cerevisiae, S.


661

Fig. 2 GEK1-related proteins are conserved among plants and archaea. Alignment of GEK1 with GEK1-related proteins. The amino acid sequence of Arabidopsis GEK1 (A.t; starts at Met45) is compared with GEK1-related proteins from tomato (L.e; deduced from two tomato EST sequences, BE432622 and AW650002), rice (O.s; GenBank accession BAD03648), archaea Archaeoglobus fulgidus DSM4304 (A.f; NP_069459), Pyrococcus horikoshii OT3 (P.h; NP_142029), Methanopyrus kandleri AV19 (M.k; NP_614222), Thermoplasma acidophilum DSM1728 (T.a; NP_393501), and Sulfolobus solfataricus P2 (S.s; NP_343614). Identical amino-acid residues are indicated by reversed characters, and conservative changes among four or more sequences are shaded. The gek1 mutation sites are also shown. Sequences were aligned with CLUSTAL X (Thompson et al. 1994).

pombe, C. elegans, and Drosophila, and there are no related ESTs from organisms other than plants. It is likely that GEK1 has been transferred horizontally. Genes transferred horizontally from one organism to another often have distinct structures, such as different codon usage or GC content, but we could not find any significant differences for GEK1. Subcellular localization of GEK1 and expression pattern of GEK1 TargetP, a computer program that predicts the cellular localization of proteins (Emanuelsson et al. 2000), predicted a mitochondrial localization of GEK1. The predicted GEK1 has an additional 44 amino acids at its N-terminus compared to archaeal proteins. It is possible that this N-terminal extension functions as a signal peptide and localizes GEK1 to mitochondria or plastids. To confirm this, and to obtain insight into the in vivo function of GEK1, we transiently expressed recombinant GEK1–green-fluorescent-protein (GFP) fused proteins in epidermal cells of Arabidopsis petioles by using a particle bombardment technique. We constructed two types of plasmid, in which a GEK1 open reading-frame from Met1 or Met45 was fused translationally to the GFP gene. After a 20-h incubation

following the bombardment, green fluorescence was observed in several cells with each construct. The green fluorescence was detected in the whole cell, including the nucleus, whichever construct was used (Fig. 3). This fluorescence pattern was the same as that observed when GFP alone was used. We also examined a GFP–GEK1 fusion protein and obtained the same fluorescence pattern (data not shown). These results suggest that GEK1 is localized in the cytoplasm and nucleus, and that the N-terminal extension does not function as a signal peptide. It is worth noting that the fluorescence patterns of GEK1–GFP were not affected by ethanol treatment (data not shown). To see whether the N-terminal extension is required for the GEK1 function, we constructed two types of 35S:: GEK1cDNA transgene that had a GEK1 open reading frame starting at Met1 or Met45 and introduced each into the gek1-2 plants. Both types of transgenic plant grew at the same rate on medium containing 0.1% (v/v) ethanol (data not shown), suggesting that the protein translated from Met45 has the GEK1 function. Amino acid sequences upstream of Met45 are not conserved among plant GEK1-related proteins. In addition, several plant cDNAs for the GEK1-related genes have an in-frame stop codon upstream of the Met codon corresponding to Met45

662


Fig. 3 Expression of the GEK1 gene. Northern blot analysis of GEK1. Total RNAs were isolated from tissues of wild-type (Col) plants grown under normal conditions (A), or from plants treated with NaCl (250 mM), sorbitol (0.6 M), low-temperature shock (0°C), hightemperature shock (37°C), low oxygen stress (N2 gas), paraquat (100 µM), abscisic acid (ABA, 50 µM), 1-aminocyclopropane-1carboxylic acid (ACC, 25 µM), or water for 15 h (B), and fractionated on agarose gel. The GEK1 cDNA was used as a probe. Ethidium bromide staining of rRNAs is shown beneath as loading controls.

of Arabidopsis GEK1. These findings indicate that Met45 is the initiation codon for GEK1. This idea is consistent with the phenotype of gek1-1. The gek1-1 mutation causes an amino acid substitution, Met45 to Ile. Although the conversion Met to Ile does not seem to cause a drastic conformational change around this residue, the gek1-1 mutant has a clear phenotype. Given that Met45 is the initiation codon, the clear phenotype of gek1-1 is explicable. The expression of GEK1 in the wild-type plant was examined by Northern blot analysis. The mRNA of GEK1 was ubiquitously detected in all tissues, relatively higher in siliques and root, and lower in leaves (Fig. 4A). Genes involved in ethanol metabolism, such as ADH (de Bruxelles et al. 1996), PDC1 (Kursteiner et al. 2003), and AtALDH3 (Kirch et al. 2001), have been shown to be stress-inducible. We assessed the expression pattern of GEK1 under several stress conditions by Northern blotting. The expression of GEK1 did not seem to respond to treatment with ethanol (0.5% [v/v]), NaCl (250 mM), sorbitol (0.6 M), low temperature (0°C), high temperature (37°C), low oxygen stress (nitrogen gas), paraquat (100 µM), ABA (50 µM), or ACC (25 µM) (Fig. 4B). Overexpression of GEK1 confers an enhanced tolerance to ethanol and acetaldehyde in plants but not in E. coli or yeast cells As described previously, mutations in the GEK1 gene cause the ethanol- or acetaldehyde-hypersensitive phenotype (Hirayama et al. 2004). To determine whether overproduction of GEK1 increases ethanol tolerance, we checked the growth of transgenic Col lines carrying 35S::GEK1cDNA in the presence of ethanol. Four-day-old seedlings of wild type and these transgenic lines were placed on medium containing 3% (v/v) ethanol. Several transgenic lines (#500, 502, and 512) grew better than the wild type (Fig. 5A). The increased ethanol tolerance of these lines was more evident in the root growth (Fig. 5B). RTPCR experiments revealed that these ethanol-tolerant lines had higher GEK1 mRNA levels (Fig. 5C). In contrast, the lines with low GEK1 expression levels (#449 and 451) exhibited an

Fig. 4 Subcellular localization of GEK1–GFP fusion protein. Subcellular localization of GEK1–GFP fusion proteins. Two types of recombinant plasmids that encoded GEK1–GFP fusion proteins were introduced into epidermal cells of Arabidopsis petioles by particle bombardment. The GEK1 (Met1) construct contained the N-terminal extension, whereas GEK1 (Met45) did not. Fluorescence was observed under a microscope 20 h after the bombardment.

enhanced ethanol-sensitivity compared with the wild type. We examined whether GEK1 overexpression increases acetaldehyde tolerance by using gek1-2 transgenic lines. Transgenic lines possessing 35S::GEK1cDNA displayed an enhanced tolerance to acetaldehyde when compared to the wild type (Fig. 5D, E). These results suggest that the overexpression of GEK1 confers an enhanced tolerance to ethanol and acetaldehyde in plant tissues. It should be interesting to see whether overexpression of GEK1 improves the ethanol and acetaldehyde tolerance of organisms that do not possess a GEK1-related gene. We tried to determine the effect of GEK1 on the ethanol tolerance of E. coli, S. cerevisiae, and S. pombe. The high-level GEK1 expression of those cells was confirmed by the detection of recombinant GEK1 protein in the soluble fraction of an E. coli extract separated on SDS-PAGE gel or by the detection of transcripts by Northern blot analysis using total RNA isolated from yeast cells (data not shown). As shown in Fig. 6, expression GEK1 did not seem to improve their growth in the presence of ethanol.

Discussion As described previously, mutations in the GEK1 gene causes enhanced sensitivity to acetaldehyde (Hirayama et al. 2004). Here we showed that overexpression of GEK1 in plant tissues conferred an increased tolerance to ethanol and acetaldehyde, confirming that GEK1 is required for tolerance to those substances. Ethanol and acetaldehyde tolerance seems common since ethanol metabolic pathway is found almost all the organisms. Therefore, the GEK1-like gene or GEK1-related protein has been expected to be common. Surprisingly, con-


663

Fig. 5 Overexpression of GEK1 confers an enhanced tolerance to acetaldehyde and ethanol. (A) Phenotype of 35S:: GEK1cDNA transgenic Col plants in the presence of ethanol. Four-day-old seedlings of each line were transferred to medium containing 3% (v/v) ethanol (upper panels) and grown for 10 d (lower panels). (B) Root morphology of plants shown in panel A. Roots of the wild type (Col) and a transgenic line, #512, are shown. (C) Analysis of the expression of GEK1. RT-PCR experiments were performed using total RNA from the lines indicated. β-tubulin was used as a control. (D) Phenotype of 35S::GEK1cDNA transgenic gek1-2 plants in the presence of acetaldehyde. Five-day-old plants were moved to medium containing 0.3% (v/v) acetaldehyde (upper panels) and grown for 10 d (lower panels). (E) Root morphology of plants shown in panel D. The roots of gek1-2, wild type (Col), and a transgenic line, TP#3, are shown.

trary to this expectation, GEK1 turned out to be a novel protein that is unique to plants and archaea. Structure and biochemical function of GEK1 At this moment, there is no information on the biochemical function of GEK1. There is no GEK1 homologue whose function has been elucidated. Although histidine residues are considerably highly conserved among GEK1-related proteins, reminiscent of metal binding property, the predicted GEK1 protein does not have any known motifs that offer us the information on its biochemical function. As shown in Fig. 2, when plant GEK1-related proteins are compared to the archaeal one, several sequence insertions were detected. It might be possible that the function of archaeal GEK1-like proteins is different

from those of plants. It should be determined whether GEK1like proteins from plants and archaea have the same function. Without the information on the biochemical function of GEK1, complementation analysis is the only approach we can take. However, it will be difficult to test if archaeal GEK1-like proteins complement gek1 since archaeal proteins prefer extreme conditions, such as higher temperature. Since the defect in GEK1 causes an enhanced sensitivity to acetaldehyde, it might be possible that GEK1 is involved in the removal of acetaldehyde adducts, or in a metabolic reaction producing a certain metabolite whose accumulation causes an enhanced acetaldehyde sensitivity (Hirayama et al. 2004). However, proteins with such a function have not been described so far. To identify the biochemical function of GEK1,

664


correlation between GEK1 expression and ethanol tolerance, and strongly suggest that GEK1 is involved in the tolerance to ethanol and acetaldehyde in plants. The GEK1-related genes, however, have not been found in organisms other than plants and archaea. In addition, overexpression of Arabidopsis GEK1 in E. coli, budding or fission yeast did not increase their tolerance to ethanol (Fig. 6). These results imply that GEK1 functions in a plant-specific mechanism for ethanol and acetaldehyde tolerance, or requires some plant-specific cofactors for its functions. If GEK1 functions with other enzymes or requires cofactors, it is possible that those enzymes or cofactors are also specific to plants and archaea. We systematically searched for Arabidopsis genes whose orthologs existed in archaeal genomes but not in the genomes of E. coli, S. cerevisiae, Drosophila, and C. elegans, or in ESTs from human and mouse. We found AtTOP6B in addition to GEK1 fell into this category. AtTOP6B encodes a topoisomerase 6B subunit and has been reported already for its uniqueness to plants and archaea (Hartung and Puchta 2001). Recently, AtTOP6B was shown to be involved in endoreduplication and brassinosteroid-dependent cell elongation (Hartung et al. 2002, Sugimoto-Shirasu et al. 2002, Yin et al. 2002). From the obvious difference between the phenotypes of GEK1 and AtTOP6B defective mutants, any interactions between these two genes are implausible.

Fig. 6 Effect of GEK1 on the ethanol sensitivity of E. coli, S. cerevisiae and S. pombe. The ethanol sensitivity of E. coli (A), S. cerevisiae (B) and S. pombe (C) expressing the GEK1 gene were examined. Over-night culture was sequentially diluted and spotted on the medium containing ethanol. The cells harboring vacant vector plasmid were used as control. In E. coli, GEK1 was expressed as a (His)6-tag-fusion protein or a native form (see Materials and Methods). The results obtained using the latter one are shown in panel A.

detailed analysis of a recombinant GEK1 protein will be required. Furthermore, isolation of suppressor mutants for gek1 will also help us to address the GEK1 function. Such approaches are underway. Transgenic plants overexpressing the GEK1 gene showed an increased ethanol and acetaldehyde tolerance (Fig. 5). On the other hand, transgenic plants with low GEK1 expression showed a less ethanol tolerance. In the latter lines, presumably, co-suppression or post-transcriptional gene silencing of GEK1 occurred (for reviews, Fagard and Vaucheret 2000, Sijen and Kooter 2000). The expressions of GEK1 in those lines were not lower than wild type. We think that post-transcriptional gene silencing did not occur in all cells but some cells still expressed the GEK1 gene in these lines. These results revealed the clear

Physiological relevance of GEK1 We observed no abnormalities even in the stronger gek1 mutants, gek1-2 and gek1-3, other than their increased sensitivity to ethanol and acetaldehyde. In plants, ethanol and acetaldehyde production are involved in low oxygen stress tolerance and pollen development (Tadege et al. 1999). As we reported previously, gek1 had no significant effect in low oxygen stress tolerance. Based on the segregation ratio, it is apparent that gek1 does not have any defect in pollen development (Hirayama et al. 2004). We also failed to observe any effect of GEK1 overexpression on the growth, low oxygen stress tolerance and fertility (data not shown). Since plants are thought not to be exposed to exogenous ethanol at high concentration in the field, at this moment we cannot suggest a physiological relevance for GEK1. However, because of its remarkable conservation in the plant kingdom, we think that GEK1 has an important role in plant life. One possible explanation is that GEK1 is required under limited conditions that we have not examined. For example, ADH is required for survival under anoxic stress but is dispensable for growth under laboratory conditions. Alternatively, these stronger gek1 alleles (gek1-2 and gek1-3) are leaky, as is gek1-1, since both alleles can express truncated GEK1 proteins. All three gek1 alleles were obtained in the screening of individuals that could not germinate on medium containing 0.04% ethanol over 5–7 d, but that germinated in a few days after transfer to normal medium. In such a screening method, stronger alleles might be missed, since they might be killed or


damaged too badly during ethanol treatment to grow after transfer to normal medium. Isolation and characterization of additional gek1 alleles and of suppressor or enhancer mutants for gek1 will allow us to understand the physiological function of GEK1. Evolutionary aspects GEK1-related genes are found only in plants and archaea. As far as we know, there are no other such genes except for AtTOPB6. How AtTOPB6 has been conserved only in plants and archaea remains to be elucidated. There are two explanations for conservation of a gene in two organisms that are phylogenetically divergent from each other, horizontal and vertical gene transfer. Since no GEK1-related genes can be found in organisms other than plants and archaea, it is more likely that GEK1 was transferred horizontally from the archaea to the plant ancestor(s). If this is the case, plant ancestors, probably unicellular, had to have lived alongside archaea, because horizontal gene transfer requires physical interaction between the organisms. Horizontal gene transfer among eukaryotes, archaea, and bacteria is now considered to have happened relatively frequently (Koonin et al. 2001, Andersson et al. 2003). Given that no GEK1-related genes have been found in the bacterial genomes sequenced so far, the system that involves the GEK1 function appears to be unique. It is still possible that GEK1 has been inherited vertically from ancestral organisms to plants and archaea through evolution. If this is the case, organisms other than plants and archaea have lost this gene during evolution. Whichever the case, the presence of GEK1 in plants emphasizes the importance of its function in plants and the uniqueness of the plant kingdom among eukaryotes. Biochemical characterization of the GEK1 protein and deep studies of the system in which GEK1 is involved will allow us to uncover more secrets that plants hide.

Materials and Methods Plant materials and growth conditions Arabidopsis thaliana Columbia (Col) or Landsberg erecta (Ler) ecotype was used. Plants were either grown on MS plates that contained Murashige and Skoog salt mix, 2% (w/v) sucrose, 0.5 mM MES (pH 5.8), and 0.8% (w/v) agar, or grown on soil, under a 16-h light/8-h dark cycle at 22–23°C. Sown seeds were kept at 4°C for 3 or 4 d to induce germination before incubation in growth chambers. The adh mutant R002 (Bensheim ecotype) was obtained from the Arabidopsis Biological Resource Center. Mapping of the GEK1 gene For mapping GEK1, we used PCR-based markers, such as simple-sequence-length polymorphisms (Bell and Ecker 1994). gek1-1 or gek1-2 (Col background) was crossed with Ler, and F2 progeny were obtained. Ethanol-hypersensitive individuals were selected on medium containing 0.1% (v/v) ethanol, and germinated and grown on normal MS medium. Methods for genomic DNA isolation and PCR conditions are described elsewhere (Hirayama et al. 1999).

665

Isolation of the cDNA and genomic clones, DNA sequencing and RTPCR A GEK1 cDNA clone was isolated from a cDNA library constructed on a lambda Zap vector by using a DNA fragment amplified by PCR from the GEK1 gene as a probe. The nucleotide sequence of the cDNA was determined by a Sequencer 3100 (Applied Biochemicals) according to the manufacturer’s instructions. The cDNA clones obtained did not contain “the predicted first ATG codon” (Met1) suggested by the Arabidopsis Genome Initiative (2000). To construct a cDNA containing the predicted first ATG codon, we replaced the DNA segment from the 5′ end of the GEK1 cDNA to the MscI site with the genomic DNA fragment spanning from 83 bp upstream of the predicted first ATG codon to the MscI site. A 3.5-kb genomic DNA fragment containing GEK1 was obtained by PCR reaction using oligonucleotides GEK1gEco and GEK1gBam as primers with a highfidelity Taq polymerase, KOD-Plus (Toyobo, Osaka, Japan). Sequencing of the DNA fragment amplified by PCR showed that no change had occurred. For RT-PCR experiments, complementary DNA was synthesized from 1 to 2 µg of total RNA using a poly-T primer with ReverTraAce (Toyobo) according to the manufacturer’s suggestion. One- to two-fiftieths of the synthesized cDNA was used for PCR reactions. Oligonucleotide primers used in this study, F19-25.1; TTACGTTGGGAAGAAGCTACCG, F19-25.11; CTGCCTCCTAGCCCAAGACC, F19-25.R; ATGTTTTTAAGGGGGTCTCTCTAGA, GEK1gEco; GAGACGAATTCAATAGGTGGGTG, GEK1gBam; ATATGGATCCGTGGACTTACTTGGTC, βTUB-F; ATCCCACCGGACGTTACAACG, βTUB-R; TTCGTTGTCGAGGACCATGC. Transient expression of GEK1–GFP fusion proteins The cDNA for GEK1 was introduced into a GFP expression vector derived from pTH2 (Chiu et al. 1996) so as to express GEK1–GFP fusion proteins under the control of the CaMV 35S promoter. The plasmid DNAs were adsorbed onto gold microcarrier particles and bombarded into 2-week-old rosette plants with a Helios particle gun (BioRad) according to the manufacturer’s instructions. Around 20 h after the bombardment, fluorescence was observed under a fluorescence microscope (Olympus) with a U-MNIBA2 mirror unit. Construction of transgenic plants The GEK1 cDNA and the GEK1 genomic DNA were inserted between the BamHI and Acc65I sites of the pROK2 binary vector and between the BamHI and HindIII sites of the pBI101 binary vector, respectively. Agrobacterium GV3101 cells were transformed with the plasmid and used for infection of Arabidopsis plants by the flower dipping method (Clough and Bent 1998). Transgenic lines were screened by kanamycin tolerance in the next generation. Overexpression of GEK1 in E. coli, S. cerevisiae and S. pombe For expression in E. coli cells, two plasmids were constructed using pET21d. One was intended to express a recombinant GEK1 protein fused to a (His)6-tag at its C-terminus and the other to express GEK1 alone. The GEK1 cDNA was placed downstream of the ADH1 promoter of the pSCW231 vector for expression in S. cerevisiae, or downstream of the adh+ promoter derived from the pART1 vector for expression in S. pombe cells. These recombinant plasmids were introduced into S. cerevisiae DBY746 (α, leu2-3.112, ura3-52, trp1-289, his3-∆1) or S. pombe HM123 (h–, leu1–) by the lithium acetate method (Ito et al. 1983, Okazaki et al. 1990). These cells harboring recombinant GEK1 plasmids or parental plasmids were grown on media containing various concentrations of ethanol and their growth rates were observed.

666


Acknowledgments We thank Drs. T. Kuromori, J. Kikuchi, H. Nakamura, M. Ikeguchi, and H. Hiroaki for their helpful discussions and suggestions; Dr. M. Noji for information on the usage of the particle gun; and Drs. T. Ito and Y. Niwa for the plasmid DNAs for the GFP fusion proteins. This work was partly supported by a Grant-in-Aid from the Ministry of Education, Sports, Culture, Science and Technology of Japan and the RIKEN President’s Special Research Grant to T.H.

References Aguilera, A. and Benitez, T. (1986) Ethanol-sensitive mutants of Saccharomyces cerevisiae. Arch. Microbiol. 143: 337–344. Andersson, J.O., Sjorgen, A.M., Davis, L.A.M., Embley, T.M. and Roger, A.J. (2003) Phylogenetic analysis of diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes. Curr. Biol. 13: 94–104. Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796–815. Bell, C. and Ecker, J.R. (1994) Assignment of thirty microsatellite loci to the linkage map of Arabidopsis. Genomics 19: 137–144. Chiu, W.-L., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H. and Sheen, J. (1996) Engineered GFP as a vital reporter in plants. Curr. Biol. 6: 325–330. Clements, D.I., Forman, A., Jerrells, T.R., Sorrell, M.F. and Tuma, D.J. (2002) Relationship between acetaldehyde levels and cell survival in ethanolmetabolizing hepatoma cells. Hepatology 35: 1196–1204. Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16: 735– 743. de Bruxelles, G.L., Peacock, W.J., Dennis, E.S. and Dolferus, R. (1996) Abscisic acid induces the alcohol dehydrogenase gene in Arabidopsis. Plant Physiol. 111: 381–391. Emanuelsson, O., Nielsen, H., Brunak, S. and von Heijne, G. (2000) Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300: 1005–1016. Fagard, M. and Vaucheret, H. (2000) (Trans)Gene silencing in plants; How many mechanisms? Annu. Rev. Plant Physiol. Plant Mol. Biol. 51: 167–194. Fernandez-Checa, J.C., Kaplowitz, M.D., Colell, A. and Garcia-Ruiz, C. (1997) Oxidative stress and alcoholic liver disease. Alcohol Health & Research World 21: 321–324. Hartung, F., Angelis, K.J., Meister, A., Schubert, I., Meizer, M. and Puchta, H. (2002) An archaebacterial topoisomerase homolog not present in other eukaryotes is indispensable for cell proliferation of plants. Curr. Biol. 12: 1787–1791. Hartung, F. and Puchta, H. (2001) Molecular characterization of homologues of both subunits A (SPO11) and B of the archaebacterial topoisomerase 6 in plants. Gene 271: 81–86. Hirayama, T., Fujishige, N., Kunii, T., Nishimura, N., Iuchi, S. and Shinozaki, K. (2004) A novel ethanol-hypersensitive mutant of Arabidopsis. Plant Cell Physiol. 45: ???–???. Hirayama, T., Kieber, J.J., Hirayama, N., Kogan, M., Guzman, P., Nourizadeh, S., Alonso, J.M., Dailey, W.P., Dancis, A. and Ecker, J.R. (1999) RESPON-

SIVE-TO-ANTAGONIST1, a Menkes/Wilson disease-related copper transporter, is required for ethylene signaling in Arabidopsis. Cell 97: 383–393. Ito, H., Fukuda, Y., Murata, K. and Kimura, A. (1983) Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153: 163–168. Kirch, H.-H., Nair, A. and Bartels, D. (2001) Novel ABA- and dehydrationinducible aldehyde dehydrogenase genes isolated from resurrection plant Craterostigma plantagineum and Arabidopsis thaliana. Plant J. 28: 555–567. Koonin, E.V., Makarova, K.S. and Aravind, L. (2001) Horizontal gene transfer in prokaryotes: quantification and classification. Annu. Rev. Microbiol. 55: 709–742. Kuo, H.-F., Tsai, Y.-F., Young, L.-S. and Lin, C.-Y. (2000) Ethanol treatment triggers a heat shock-like response but no thermotolerance in soybean (Glycine max cv. Kaohsiung No. 8) seedlings. Plant Cell Environ. 23: 1099– 1108. Kursteiner, O., Dupuis, I. and Kuhlemeier, C. (2003) The Pyruvate decaroboxylase1 gene of Arabidopsis is required during anoxia but not other environmental stresses. Plant Physiol. 132: 968–978. Li, G.C. and Hahn, G.M. (1978) Ethanol-induced tolerance to heat and to adriamycin. Nature 274: 699–701. Okazaki, K., Okazaki, N., Kume, K., Jinno, S., Tanaka, K. and Okayama, H. (1990) High-frequency transformation method and library transducing vectors for cloning mammalian cDNAs by trans-complementation of Schizosaccharomyces pombe. Nucleic Acids Res. 18: 6485–6489. Piper, P.W. (1995) The heat shock and ethanol stress response of yeast exhibit extensive similarity and functional overlap. FEMS Microbiol. Lett. 134: 121– 127. Rousselin, P., Lepingle, A., Faure, J.D., Bitoun, R. and Caboche, M. (1990) Ethanol-resistant mutants of Nicotiana plumbaginifolia are deficient in the expression of pollen and seed alcohol dehydrogenase activity. Mol. Gen. Genet. 222: 409–415. Sanchez, Y., Taulien, J., Borkovich, K.A. and Lindquist, S. (1992) Hsp104 is required for tolerance to many forms of stress. EMBO J. 11: 2357–2364. Sijen, T. and Kooter, J.M. (2000) Post-transcriptional gene-silencing; RNAs on the attack or on the defense? Bioessays 22: 520–531. Sugimoto-Shirasu, K., Stacey, N.J., Corsar, J., Roberts, K. and McCann, M.C. (2002) DNA topoisomerase VI is essential for endoreduplication in Arabidopsis. Curr. Biol. 12: 1782–1786. Tadege, M., Bucher, M., Stahli, W., Suter, M., Dupuis, I. and Kuhlemeier, C. (1998) Activation of plant defense responses and sugar efflux by expression of pyruvate decarboxylase in potato leaves. Plant J. 16: 661–671. Tadege, M., Dupuis, I. and Kuhlemeier, C. (1999) Ethanolic fermentation: new functions for an old pathway. Trends Plant Sci. 4: 320–325. Takahashi, T., Shimoi, H. and Ito, K. (2001) Identification of genes required for growth under ethanol stress using transposon mutagenesis in Saccharomyces cerevisiae. Mol. Gen. Genet. 265: 1112–1119. Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673–4680. Yin, Y., Cheong, H., Friedrichsen, D., Zhao, Y., Hu, J., Mora-Garcia, S. and Chory, J. (2002) A crucial role for the putative Arabidopsis topoisomerase VI in plant growth and development. Proc. Natl Acad. Sci. USA 99: 10191– 10196. Zhang, Q., Piston, W.D. and Goodman, R.H. (2002) Regulation of corepressor function by nuclear NADH. Science 295: 1896–1897.

(Received March 19, 2004; Accepted April 2, 2004)