Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2003 Jul;132(3):1292–1302. doi: 10.1104/pp.103.022244

Enhanced Low Oxygen Survival in Arabidopsis through Increased Metabolic Flux in the Fermentative Pathway1

Kathleen P Ismond 1, Rudy Dolferus 1,2, Mary De Pauw 1, Elizabeth S Dennis 1, Allen G Good 1,*
PMCID: PMC167069  PMID: 12857811

Abstract

We manipulated the enzyme activity levels of the alcohol fermentation pathway, pyruvate decarboxylase (PDC), and alcohol dehydrogenase (ADH) in Arabidopsis using sense and antisense overexpression of the corresponding genes (PDC1, PDC2, and ADH1). Transgenic plants were analyzed for levels of fermentation and evaluated for changes in hypoxic survival. Overexpression of either Arabidopsis PDC1 or PDC2 resulted in improved plant survival. In contrast, overexpression of Arabidopsis ADH1 had no effect on flooding survival. These results support the role of PDC as the control step in ethanol fermentation. Although ADH1 null mutants had decreased hypoxic survival, attempts to reduce the level of PDC activity enough to see an effect on plant survival met with limited success. The combination of flooding survival data and metabolite analysis allows identification of critical metabolic flux points. This information can be used to design transgenic strategies to improve hypoxic tolerance in plants.

Plants are constantly challenged by environmental stresses that reduce crop yield (e.g. salinity, low temperature, drought, and flooding). Soils with excess water account for 15.7% of the arable land in the United States, and flooding accounted for 16.4% of the crop insurance claims, making it the second highest cause of crop loss in the United States (Boyer, 1982). Traditionally, plant breeders have selected directly for increased stress tolerance, however, selection for flooding tolerance has been performed only in rice (Oryza sativa), where a genetic mapping approach was used to identify major and minor genes involved in submergence tolerance (Sripongpangkul et al., 2000; Xu et al., 2000). Transgenic plants with increased tolerance to drought, salinity, low temperatures, or a combination of these stresses (Nelson et al., 1998; Huang et al., 2000) were obtained by introducing genes for the overaccumulation of benign compounds (e.g. Gly betaine, mannitol, and Pro) that are believed to act as osmolytes. Other groups have increased stress tolerance in Arabidopsis by overexpressing transcription factors (Jaglo-Ottosen et al., 1998; Kasuga et al., 1999; Yamaguchi-Shinozaki and Shinozaki, 2001) or signal transduction components (Piao et al., 2001; Tamminen et al., 2001).

There is substantial variation among terrestrial crop species in their ability to tolerate waterlogging conditions. Even rice, which is adapted to life in a flooded environment, incurs damage if the shoots are completely submerged for periods of time (Drew, 1997; Quimio et al., 2000). Genetic variation in flooding tolerance has been found in rice, maize (Zea mays), barley (Hordeum vulgare), and Arabidopsis. In some cases, this variation results from specific mutants (Harberd and Edwards, 1982; Dolferus et al., 1985; Lemke-Keyes and Sachs, 1989; Ricard et al., 1998). Several studies have focused on the effect of overexpression of genes on flooding tolerance. Zhang et al. (2000) overexpressed a gene involved in cytokinin biosynthesis and demonstrated improved flooding tolerance in Arabidopsis. Grichko and Glick (2001) overexpressed a bacterial ACC deaminase under the control of three different promoters and were able to produce transgenic tomato (Lycopersicon esculentum) plants that had increased flooding tolerance.

Because higher plants are sessile and obligate aerobes, they have evolved a number of mechanisms to survive the hypoxic conditions that are generated by flooding. These mechanisms include morphological adaptations, such as formation of adventitious roots (Lorbiecke and Sauter, 1999; Mergemann and Sauter, 2000) and aerenchyma, and a number of complex metabolic adaptations (for review, see Drew, 1997; Liao and Lin, 2001). Studies on maize root tips have led to the conclusion that several factors are involved in improved hypoxic tolerance, all of which are associated with the fermentation pathway. These include better energy status due to higher rates of glycolysis and ethanol fermentation, limitations of lactate accumulation, and better regulation of cytoplasmic pH (Roberts et al., 1984a, 1984b, 1985; Ricard et al., 1998). The de novo synthesis of the fermentative enzymes (Fig. 1; pyruvate decarboxylase [PDC]; alcohol dehydrogenase [ADH]; lactate dehydrogenase [LDH]; and Ala aminotransferase [AlaAT]) and other anaerobic polypeptides occurs upon exposure to low oxygen conditions within the plant root (Sachs et al., 1980). The fermentative products of pyruvate include the end products lactate, ethanol, and Ala, which are produced by three separate pathways (Fig. 1).

Figure 1.

Figure 1.

Open in a new tab

Under low oxygen stress conditions, plants activate three main fermentation pathways that use pyruvate as starting substrate.

This study focuses on the manipulation of the ethanol fermentation pathway. PDC (EC 4.1.1.1) catalyzes the first step, which is responsible for the irreversible conversion of pyruvate to acetaldehyde. ADH (EC 1.1.1.1) then converts acetaldehyde to ethanol, with the concomitant regeneration of NAD+ (Fig. 1). The regeneration of NAD+ from NADH is thought to be the most important function of the alcohol fermentation pathway under low oxygen stress conditions, because this function is impaired by the inactivation of oxidative phosphorylation. This function is vital, because in the absence of NAD+, glycolysis ceases (Kennedy et al., 1992; Robertson et al., 1994). Although the importance of the fermentation pathway in hypoxia survival is known, the switch from respiration to fermentation is not completely understood. A large amount of research has been conducted on the role of ADH, however, to date, no one has overexpressed ADH to determine whether increased ADH activity would improve flooding tolerance. Less research has been done on PDC, but several groups have overexpressed this gene in different plants (Tadege et al., 1998; Quimio et al., 2000), and its importance as a critical control point is becoming apparent. In Arabidopsis, four genes encode PDC activity. PDC1 is the main anaerobically induced gene and is expressed primarily in the roots, whereas PDC2 is constitutively expressed at low levels in both roots and leaves (R. Dolferus, unpublished data). Only one ADH gene (ADH1) has been identified in Arabidopsis (Dolferus and Jacobs, 1984). The anaerobic induction and promoter function of this gene has been well documented (Dolferus et al., 1994; Dennis et al., 2000).

In this paper, we report the effect of overexpressing the enzymes of ethanol fermentation, PDC and ADH, in Arabidopsis. The combination of enzyme activity data and metabolite data from all enzymes involved in the hypoxic response allows us to critically examine carbon flow along the glycolytic and ethanol fermentation pathway. In conjunction with the survival assay, we provide a comprehensive look at overexpression of ADH and PDC and their role in hypoxic tolerance. We show that overexpression of either PDC1 or PDC2 improved plant survival under low oxygen conditions. This confirms that PDC is the metabolic control point in the alcohol fermentation pathway. ADH1 overexpression had no effect on flooding tolerance, but normal levels of ADH1 expression were shown to be critical to plant survival under low oxygen conditions.

RESULTS

Generation and Assay of Transgenic Lines for Enzyme Activity

Transgenic Arabidopsis lines containing the PDC1, PDC2, and ADH1 sense and antisense constructs (under the control of the 35S promoter) were screened in the T2 generation by northern-blot hybridizations (data not shown). The line with the highest mRNA expression levels was selected. For the antisense constructs, we selected those lines with the most significant reduction in mRNA levels. We then used enzyme activity assays to confirm the mRNA data and to select the best overexpressing/antisense lines. Homozygous T3 lines were selected from these transgenic lines, and the levels of mRNA and enzyme activity were again confirmed in this generation. The lines overexpressing PDC1, PDC2, or ADH1 were assigned the prefix “Ox” (Overexpressing). Lines showing a reduced enzyme activity were generated using antisense constructs for PDC1 and PDC2 and were assigned the prefix “Ue” (Under-expressing). In the case of ADH1, an adh1– null mutant was selected in C24 background using allyl alcohol selection (Jacobs et al., 1988). Under normal growth conditions, no obvious phenotypic differences were observed between any of the over- and under-expressing transgenic plants, or the adh1– mutant compared with C24 control plants.

Enzyme activities were determined for PDC (Fig. 2, A and B), ADH (Fig. 2, C and D), LDH (Fig. 2, E and F), and AlaAT (data not shown), under aerobic and hypoxic conditions in the selected lines. In roots and shoots, the levels of PDC activity were six to ten times higher for OxPDC1 and OxPDC2, respectively, under either aerobic or hypoxic conditions. The least expressing PDC lines, UePDC1 and UePDC2, were originally selected because they had mRNA levels reduced to 30% to 50% of wild-type levels under aerobic and hypoxic conditions (data not shown). However, in terms of activity (Fig. 2, A and B), these lines had only marginally lower PDC activity compared with the wild type under both control and hypoxic conditions.

Figure 2.

Figure 2.

Open in a new tab

Activities of PDC, ADH, and LDH in roots and shoots of wild-type, transgenic, and null mutant Arabidopsis plants. Plants were grown for 4 weeks and then treated under aerobic (white column) or hypoxic (hatched column) conditions for 24 h as described (see “Materials and Methods”). Results shown are the mean of three replicates with ses from one representative experiment. A, PDC activity in roots. B, PDC activity in shoots. C, ADH activity in roots. D, ADH activity in shoots. E, LDH activity in roots. F, LDH activity in shoots.

The OxADH1 line showed a 2- to 3-fold increase in ADH activity under hypoxic conditions in both roots and shoots compared with wild-type roots and shoots, whereas activities were increased 7- to 8-fold under aerobic conditions (Fig. 2, C and D). The ethyl methanesulfonate-induced adh1– mutant had negligible ADH activity compared with the C24 wild type (Fig. 2, C and D). In this adh1 null mutant, Glu99 and His101 were mutated to Asp and Lys, respectively, in a highly conserved area of the ADH enzyme (Dolferus et al., 1990).

The over- or under-expression of PDC1 or PDC2 had no effect on the levels of ADH activity in either roots or shoots (Fig. 2, C and D), nor was the total PDC activity in roots and shoots significantly affected by overexpression of ADH1 (Fig. 2, A and B). The UePDC2 and OxADH1 lines displayed increased levels of LDH activity in the roots, but in shoots, only OxADH1 showed slightly increased LDH activity levels (Fig. 2, E and F). In the adh1– line, LDH levels were decreased in roots but not in shoots (Fig. 2, E and F). No significant changes of LDH activity were observed by overexpressing the PDC1 or PDC2 genes. No differences in AlaAT activity were detected between any of the transgenic or mutant genotypes and the C24 genotype (data not shown).

Manipulation of PDC Levels

To examine the effect of the introduced PDC transgenes, metabolites for the key compounds in the fermentative pathway were analyzed in roots, shoots, and the medium surrounding the roots under hypoxic conditions (Figs. 3 and 4). For acetaldehyde, only the liquid Murashige and Skoog medium was measured, because previous experience has indicated that levels in the medium accurately reflect intracellular levels, because acetaldehyde passes easily through intracellular membranes into the medium.

Figure 3.

Figure 3.

Open in a new tab

Effect of PDC and ADH manipulation on metabolic products of ethanol fermentation under hypoxic conditions. Plants were grown for 4 weeks and then hypoxically induced for 24 h. The levels of pyruvate, acetaldehyde, and ethanol were then determined (see “Materials and Methods”). Results shown are the mean of three replicates with ses from one representative experiment. Metabolite levels are presented for pyruvate in roots (A) and shoots (B), acetaldehyde content in the Murashige and Skoog growth medium (C), and ethanol in the medium (D), roots (E), and shoots (F). Values have been normalized against C24 wild type.

Figure 4.

Figure 4.

Open in a new tab

Effect of PDC and ADH manipulation on other fermentative products under hypoxic conditions. Plants were grown for 4 weeks and hypoxically induced for 24 h, and then the levels of lactate and Ala were determined (see “Materials and Methods”). Results shown are the mean of three replicates with ses from one representative experiment. Metabolite levels are presented for lactate in roots (A), shoots (B), and Murashige and Skoog growth medium (C); and Ala in the medium (D), roots (E), and shoots (F). Values have been normalized against the C24 wild type.

OxPDC2, OxPDC1, and UePDC2 lines had decreased levels of pyruvate in roots and shoots under hypoxic conditions (Fig. 3, A and B). The decrease in pyruvate levels was associated with a 2- or 4-fold increase of acetaldehyde in the medium for OxPDC1 and OxPDC2 (Fig. 3C). Both lines also showed a corresponding increase in ethanol concentration in the medium (1.6 times; Fig. 3D). OxPDC1 was the only line that showed significantly higher ethanol levels in roots (1.8 times) and shoots (Fig. 3, D and E).

The effect of PDC overexpression on the metabolites of the other fermentation pathways that are derived from pyruvate was also measured (Fig. 4). Decreased levels of both lactate and Ala were found in the roots, shoots, and Murashige and Skoog medium for OxPDC1. These metabolites remained unchanged in OxPDC2 (Fig. 4, A–F). No significant differences were observed in Glu or 2-oxoglutarate levels for OxPDC2, but OxPDC1 had significantly lower levels of Glu in the medium (data not shown). These data show that overexpression of the Arabidopsis PDC1 and PDC2 enzymes, although overexpressed to similar total activity levels (Fig. 2), had different effects on metabolite accumulation, suggesting that they may function differently.

If overexpression of PDC leads to increased carbon flow through the ethanol pathway, then repression of PDC could act to decrease carbon flow and reduce ethanol levels. But the decrease in PDC activity that we observed in the UePDC lines was not nearly as dramatic as the increase in activity of the overexpressing lines (Fig. 2). Residual PDC activity in the UePDC lines could still account for a significant flow of carbon through the alcohol fermentation pathway. The levels of pyruvate, acetaldehyde, and ethanol production of UePDC1 and UePDC2 were not significantly different from wild type (Fig. 3, A–F), but UePDC1 and UePDC2 both showed increased levels of lactate production in roots but not in leaves (Fig. 4, A and B). This suggests that under hypoxic conditions even a minor reduction in PDC activity may be sufficient to increase the amount of carbon flow from pyruvate to lactate. Only small differences were observed in lactate and Ala (Fig. 4) and in Glu and 2-oxoglutarate levels (data not shown).

Manipulation of ADH Activity Levels

Pyruvate levels were not affected in roots, but they decreased in shoots of the OxADH1 and adh1– lines (Fig. 3, A and B). Acetaldehyde levels were reduced to 0.1% of wild-type levels in roots in OxADH1, whereas the adh1– null line had a 9-fold increase in acetaldehyde production in the medium (Fig. 3C). Although there were no significant changes in tissue ethanol levels in roots or shoots of the OxADH1 line (Fig. 3, E and F), a strong reduction in ethanol production (80%) was observed in the roots of the adh1– mutant; no change was found in the shoots (Fig. 3, D and E). No significant changes were found in lactate or Ala levels in roots, shoots, or the medium for OxADH1 transgenic lines, whereas the adh1– mutant showed a 3- to 4-fold increase in lactate levels in the medium and in roots (Fig. 4, A–C).

PDC Overexpression Enhances Low Oxygen Stress Survival

Survival assays were conducted on each genotype to assess the contribution of the transgene to whole-plant survival under low oxygen conditions. We used two experimental conditions that were previously shown to discriminate between wild-type and adh1– null mutants (Ellis et al., 1999): a 24-h non-hypoxic pretreatment (24-h NHPT), in which the plants were subjected to 0.1% (v/v) O2 for 24 h, and a 48-h hypoxic pretreatment (48-h HPT), in which the plants received a 24-h pretreatment of 5% (v/v) O2 before receiving 48 h of 0.1% (v/v) O2. Plants were then allowed to recover on vertical plates for 2 weeks, as illustrated in Figure 5. Survival, based on the ability to recover from the treatments, was quantified by measuring root tip growth, root weight, and shoot weight (Table I). For each transgenic line, we carried out a minimum of two survival assays with three repeats for each experiment.

Figure 5.

Figure 5.

Open in a new tab

Survival assay results of Arabidopsis seedlings subjected to a low oxygen treatment (0.1% [v/v] O2) with or without hypoxic (5% [v/v] O2) pretreatment. Seedlings were exposed to two treatments: low oxygen (0.1% [v/v] O2) conditions for 24 h without HPT (24H NHPT) or low oxygen for 48 h with a 24-h 5% (v/v) O2 HPT (48H HPT). After these treatments, seedlings were transferred to vertical plates for a 2-week recovery period under normal growth chamber conditions. A representative plate for each environment was selected. Column 1, wild type (C24); column 2, PDC overexpressor (Ox-PDC1); and column 3, adh1– mutant. The treatments are in row 1, 24-h NHPT; in row 2, 48-h NHPT; and in row 3, 48-h HPT.

Table I.

Effect of PDC and ADH manipulation on survival and growth of roots and shoots of wild-type, transgenic, and null mutant Arabidopsis plants presented as a percentage of C24 control

Seedlings were exposed to two treatments: low oxygen (0.1% [w/v] O2) conditions for 24 h without HPT (24H NHPT) or low oxygen conditions for 48 h with a 24-h 5% (w/v) O2 HPT (48H HPT). After these treatments, seedlings were transferred to vertical plates for a 2-week recovery period under normal growth chamber conditions. Root tip growth and root and shoot weight were quantified for six lines used in this study. The performance of each of the lines is presented as an averaged percentage compared to the C24 wild type. The results of two independent experiments are presented; each experiment consists of three repeats, and statistical analysis results are presented. t test confidence levels are indicated as follows: †, P < 0.05; ‡, P < 0.01.

Root Tip Growth
Root Weight
Shoot Weight
Line 24H NHPT t Test 48H HPT t Test 24H NHPT t Test 48H HPT t Test 24H NHPT t Test 48H HPT t Test
mm mg
OxPDC1 128.9 0.059 157.2 0.022† 113.1 0.026† 212.3 0.008‡ 139.9 0.052 108.7 0.635
137.2 0.137 202.3 0.029† 151.2 0.019† 156.4 0.357 105.2 0.131 179.4 0.036†
UePDC1 88.6 0.726 96.6 0.853 79.7 0.106 105.8 0.744 59.7 0.112 128.1 0.231
66.9 0.200 102.1 0.818 75.9 0.118 110.5 0.531 99.3 0.985 97.1 0.809
OxPDC2 109.0 0.151 157.9 0.042† 46.8 0.017† 132.2 0.012† 113.6 0.031† 176.5 0.052
109.0 0.432 136.3 0.000‡ 107.7 0.546 134.8 0.000‡ 183.9 0.001‡ 112.1 0.489
UePDC2 113.5 0.520 108.4 0.655 73.8 0.026† 210.0 0.021† 62.4 0.019† 127.6 0.299
120.0 0.324 108.4 0.228 111.2 0.005‡ 113.4 0.008‡ 112.8 0.399 101.4 0.361
OxADH1 118.4 0.604 48.7 0.096 73.3 0.165 91.1 0.475 70.8 0.109 77.4 0.070
65.0 0.084 48.7 0.096 89.5 0.719 81.7 0.545 79.2 0.436 113.1 0.720
ADH1- 19.7 0.017† 0.0 0.002‡ 120.0 0.581 26.0 0.008‡ 64.9 0.025† 121.3 0.455
60.5 0.014† 46.7 0.007‡ 109.8 0.475 91.5 0.032† 176.7 0.020† 36.6 0.000‡
Open in a new tab

Both the OxPDC1 and OxPDC2 lines performed better than the control line in the 24-h NHPT and 48-h HPT treatments. These data suggest that PDC enzyme activity levels are a limiting factor for survival of moderately severe oxygen conditions. We did not see any significant differences between C24 and the two UePDC lines that we tested. The Ox-ADH1 line did not perform better than the C24 wild-type line, but as was previously shown (Ellis et al., 1999), survival of the adh1– line was reduced compared with the C24 wild type (Fig. 5). In the recovery period after the 24-h NHPT treatment, the adh1– line showed a reduction in root tip growth, whereas none of the other lines showed any significant change (Table I). When HPT was used before low oxygen stress (48-h HPT), both PDC overexpressing lines showed improved root tip growth, whereas the adh1– line still performed more poorly than the C24 wild type (Table I).

The data for the root and shoot weight measurements were less consistent than those for root tip growth (Table I), which could partially be due to the variation in stringency of the stress treatment between the two independent experiments. OxPDC1 showed increased root formation in the 24-h NHPT and the 48-h HPT treatments. In contrast, root development of the OxPDC2 line was higher in the 48-h HPT treatment, but in one of the 24-h NHPT treatments, results were significantly worse than for C24. No significant changes were found for the UePDC1, UePDC2, and OxADH1 lines, whereas the adh1– showed decreased root formation in the 48-h HPT treatment. For the shoot weight measurements, only the OxPDC2 line showed significantly higher shoot weights in the 24-h NHPT treatment. The data for the 24-h NHPT (OxPDC1) and 48-h HPT (OxPDC1 and OxPDC2) treatments showed increased shoot weights but were not significant due to variation within each experiment.

DISCUSSION

In the absence of oxidative phosphorylation, the glycolytic conversion of Glc to pyruvate is the only means to generate limited amounts of ATP under anaerobic conditions. Plants then switch to fermentation pathways to regenerate NAD+ (Roberts et al., 1984a, 1984b). Of the three main fermentation pathways in plants, only two—alcohol and lactic acid fermentation—regenerate NAD+ (Fig. 1). Ala fermentation does not produce NAD+, but Ala accumulation under low oxygen conditions is thought to play a role in nitrogen assimilation (Good and Muench, 1993).

The three fermentation pathways are relatively simple, enabling us to start a metabolic engineering approach in Arabidopsis. The three final metabolites of these pathways are also easily measured, and because they have a common starting point, pyruvate, we can monitor changes in metabolites and study the impact on carbon flow in the branches of the fermentative pathway. We established a survival assay to study the influence of metabolic changes on low oxygen stress survival (Ellis et al., 1999).

We have successfully developed plants with modified levels of PDC and ADH activities. Six different lines were analyzed for over- and under-expression of these enzymes using enzymatic, metabolic, and survival data. This allows us to study their role in ethanol fermentation and their impact on the other branches in the fermentation pathway. Overexpression of any of these genes using the cauliflower mosaic virus 35S promoter resulted in high levels of constitutive enzyme activity that was insensitive to hypoxia (Fig. 2). Overexpression of either PDC1 or PDC2 enhanced plant tolerance to low oxygen conditions (Fig. 5; Table I). Similarly, Quimio et al. (2000) found that enhancement of PDC levels in rice corresponded with an increase in submergence tolerance. Our results indicate that increased PDC activity results in increased carbon flow through the ethanol pathway, as reflected by an increase in both acetaldehyde and ethanol pool sizes (Fig. 3). This ultimately results in increased ATP and NAD+ production. However, overexpression of a bacterial PDC gene in transgenic tobacco (Nicotiana tabacum) did not result in improved hypoxic tolerance (Tadege et al., 1998). There may be several reasons for the differences between our study and theirs. First, different plant species have different tolerances to flooding, consequently what is observed in rice may vary from what is observed Arabidopsis or tobacco. Second, our levels of PDC overexpression were significantly higher than those observed by Tadege et al. (1998). Third, the use of a bacterial gene may be inappropriate in this instance, because the enzyme may be regulated in a different fashion. This may also explain the their observation that in roots, the ethanol fermentation pathway is substrate limited. Tadege et al. (1998) suggests that in roots, oxygen levels play an important role in controlling flux through the ethanolic pathway, rather than PDC, and they suggest different regulatory mechanisms in roots, shoots, and pollen. Therefore, although the role of PDC as a control point in ethanol fermentation has been recognized, overexpression of PDC and its contribution to plant survival seems to depend on the species being studied, on the tissue of interest, and possibly on the origin of the transgene used in the experiments.

The PDC overexpressors produce much higher levels of acetaldehyde (Fig. 3), as has been observed in PDC-overexpressing tobacco plants (Bucher et al., 1994). Acetaldehyde is a highly reactive chemical effecting cellular damage by forming acetaldehyde-protein adducts (Perata et al., 1992; Braun et al., 1995). It has been suggested that acetaldehyde accumulation in adh1 mutants is the major cause of cellular damage during anoxic and postanoxic conditions (Jackson and Armstrong, 1999). Despite this, the OxPDC1 and OxPDC2 performed better than C24 in the survival assay (Fig. 5). This could mean that endogenous ADH levels in Arabidopsis are sufficient for removing harmful levels of acetaldehyde or that acetaldehyde leaks out into the medium efficiently enough to prevent internal injury. Maize mutants with reduced levels of ADH were still able to withstand hypoxia, suggesting that ADH in normal plants is in excess (Roberts et al., 1989).

The UePDC1 and UePDC2 lines displayed no visible changes in low oxygen tolerance relative to the wild type. These transformants also did not show significant differences in pyruvate, acetaldehyde, and ethanol production, although there are small differences at the level of lactate and Ala (Table I; Figs. 3 and 4). Given our current knowledge about gene silencing (Waterhouse et al., 2001b), it is unlikely that the PDC2 antisense could silence expression of the other PDC genes in Arabidopsis, but PDC1 antisense expression could silence the closely related PDC3 and PDC4 genes. Approaches to reduce PDC levels include more effective antisense constructs or the use of T-DNA or transposon-tagged Arabidopsis mutants.

The adh1– mutant had minimal ADH activity, resulting in a dramatic increase in acetaldehyde in the medium surrounding the roots (Fig. 3). The mutant also had poor tolerance to low oxygen conditions (Fig. 5). These results confirm earlier studies showing that at least low to normal levels of ADH activity are necessary for short-term, waterlogging tolerance (Johnson et al., 1994; Ellis et al., 1999). Overexpression of ADH1 did not affect carbon flow and ethanol levels did not change, suggesting that wild-type Arabidopsis ADH levels are already in excess in anoxic or hypoxic roots and that ADH levels above a threshold do not correlate with the ability to survive anoxia. Overexpression of ADH therefore does not improve hypoxic survival rates (Fig. 5). High levels of ADH activity may be of greater importance when plants are returned to aerobic conditions (Thomson and Greenway, 1991; Drew, 1997). Accumulated ethanol in roots can be remetabolized to acetaldehyde and then converted to acetate and acetyl-CoA with the enzymes acetaldehyde dehydrogenase (EC 1.2.1.4/EC 1.2.1.5) and acetyl-CoA synthase (EC 6.2.1.1). This results in a reduction of carbon loss in the recovery period after anoxia (Thomson and Greenway, 1991).

PDC activity occurs at a critical branch point between aerobic and anaerobic metabolism (Roberts et al., 1989; Waters et al., 1991a, 1991b), and PDC activity is closely related to the in vivo rates of fermentation (Bouny and Saglio, 1996). In yeast, pyruvate flows preferentially into the tricarboxylic acid cycle because the pyruvate dehydrogenase complex has a much higher affinity for pyruvate. However, in conditions of sugar over-supply, pyruvate dehydrogenase is bypassed, and much of the pyruvate is channeled through PDC (Pronk et al., 1996). In plants, the acidification of the cytoplasm during anoxia may be responsible for this metabolic switch. According to the Davies-Roberts hypothesis, cytoplasmic pH drops during the transition from normoxia (pH 7.3–7.4) to anoxia (pH 6.8) in maize root tips (Davies et al., 1974; Davies, 1980; Roberts et al., 1984a, 1984b, 1985; Gancedo and Serrano, 1989). Below pH 7.0, LDH is inhibited (Rivoal and Hanson, 1994), and the activity of PDC increases dramatically, due to its acidic pH optimum (Morrell et al., 1990; Fox et al., 1994). What is not known is whether the different PDC isoforms have different pH optima. Our data show that overexpression of the PDC1 gene—but not PDC2—leads to a reduction in metabolites of the other fermentation pathways (lactate and Ala; Fig. 4). This could indicate that the Km of PDC1 for pyruvate would favor entry of the substrate into the ethanol fermentation pathway.

Although our lines overexpressing the anaerobically inducible (PDC1) and constitutively expressed (PDC2) gene had similar PDC activity levels and low oxygen survival properties, our results indicate that PDC1 and PDC2 may have different enzymatic properties. Overexpression of the PDC1 gene, but not PDC2, leads to a reduction in metabolites of the other fermentation pathways (lactate and Ala; Fig. 4). Biochemical characterizations have demonstrated that, like yeast and some bacterial PDC enzymes, plant PDC has nonlinear kinetics and is activated by its substrate pyruvate (Mücke et al., 1996; Dietrich and König, 1997). The PDC2 enzyme, which is constitutively expressed at low levels in roots, is quite different from the anaerobically induced enzyme (R. Dolferus, unpublished data) and may have different kinetic properties. This could result in very different in vivo kinetics within the cell and corresponding changes in the levels of different metabolites.

The experiments described in this paper were carried out with plants grown in Murashige and Skoog medium containing 3% (w/v) Suc. Therefore our data showing that PDC overexpression increases metabolic flux through the alcohol fermentation pathway and improves low oxygen stress survival may only be true when sugar supply is adequate. Under sugar-limiting conditions (e.g. in the field), a faster depletion of carbohydrate stores may lead to decreased survival. An exogenous supply of Glc has been shown to ameliorate the effects of anoxia (Xia and Saglio, 1990; Waters et al., 1991a, 1991b; Hole et al., 1992; Xia et al., 1995; Ricard et al., 1998; Tadege et al., 1998). Thus, the flow of carbon through to ethanol may only be effective when high levels of sugars are available.

In conclusion, our data support the idea that PDC is a critical flux control point in hypoxic metabolism and that PDC activity levels can determine plant survival under low oxygen conditions. PDC activity controls carbon flow through the ethanol fermentation pathway and affects flow through the lactate and Ala pathways. On the basis of this study, improved flooding survival may be achieved by overexpressing a plant PDC1 or PDC2 gene. Whether overexpression of both PDC and ADH activity would further improve low oxygen stress survival remains to be established. Additional work is also required to determine first, whether continual overexpression has significant negative effects on productivity and second, whether the results of the survival assay can vary, depending on the conditions of evaluation.

MATERIALS AND METHODS

Plant Material, Growth Conditions, and Stress Treatment

Arabidopsis ecotype C24 was used in all our experiments. Seeds were germinated, and seedlings were grown as outlined by Ellis et al. (1999) on Murashige and Skoog medium containing 3% (w/v) Suc and 0.5% (w/v) agar. After 3 weeks growth, plants were transferred to liquid Murashige and Skoog medium and placed on a rotary shaker for 7 d to promote root growth. These plants were then used for all further tests. Hypoxic treatment of plants was carried out for 24 h in a 5% (v/v) O2/95% (v/v) N2 environment in the dark as described previously (Howard et al., 1987), and then tissue was harvested for enzyme, metabolite, and northern analyses.

Development of Transgenic Plants

Standard cloning and recombinant DNA methods were used throughout (Maniatis et al., 1982). Sense and antisense constructs expressing PDC1 and PDC2 (GenBank accession nos. U71121 and U71122, respectively; R. Dolferus, unpublished data), and a sense construct expressing ADH1 were developed as previously described (Dennis et al., 2000). The cauliflower mosaic virus 35S promoter was used to drive overexpression of all three genes. Constructs were introduced into plants using a modified root transformation protocol (Dolferus et al., 1994). The ethyl methanesulfonate adh1– mutant was isolated using allyl alcohol selection as described previously (Jacobs et al., 1988). Sequencing of the adh1 allele was performed using synthetic oligonucleotides created to identify mutations within the ADH1 gene. The University of Wisconsin GCG program was used for sequence analysis.

RNA Extractions and Northern-Blot Analysis

RNA extractions, blot hybridizations using RNA probes, and filter washings were performed as described previously (Dolferus et al., 1994). Signals were detected by laying filters upon phosphor imager screens (Molecular Dynamics, Sunnyvale, CA) and then quantified. To normalize for sample loading differences, signal strengths were divided by their respective ubiquitin signals (Dolferus et al., 1994).

Enzyme and Metabolite Assays

Each overexpressing line was assessed alongside its matching under-expressing line and the wild-type C24 line, in both aerobic and hypoxic environments (Howard et al., 1987). Three plates with 50 plants of each plant line were used in each environment. Roots and shoots were collected separately along with 1 mL of liquid medium from each plate and then quick frozen in liquid nitrogen and stored at –70°C for further processing. Enzymes were extracted with a 3:1 ratio (milliliters per gram fresh weight) of extraction buffer to tissue. The extraction buffer for ADH, LDH, and AlaAT was made fresh each day (5 mm EDTA, 1 mm dithiothreitol, 10 mm Cys, 0.1 mm phenylmethylsulfonyl fluoride, 5 μm leupeptin, and 0.1 m Tris-HCl, pH 7.7). Acid-washed sand and polyvinylpolypyrrolidone (20%–50% of tissue weight) was added during tissue grinding. ADH and AlaAT activity were measured spectrophotometrically using a SpectraMax+ ELISA plate reader (Good and Crosby, 1989a, 1989b). LDH activity was measured as an increase in A490 using the protocol of Korzeniewski and Callewaert (1983). The PDC extraction buffer and assay (Laszlo and Saint Lawrence, 1983) was performed with the addition of proteinase inhibitors (Cys, leupeptin, and phenylmethylsulfonyl fluoride) to the PDC extraction buffer in the same final concentrations as in the ADH extraction buffer.

Gas chromatography (GC) was used to determine acetaldehyde and ethanol concentrations in the liquid medium. Roots of 3-week-old plants were harvested, transferred into 10-mL syringes containing 10 mL of growing medium gassed with 100% argon to exclude N2. The syringes were outfitted with 18-gauge needles containing glass wool at the air lock to prevent roots from clogging the needle. After the plunger had been inserted into the syringe, a 1-mL argon gas bubble was pulled into the syringe's chamber, and the needle tip was fitted with a rubber stopper to prevent gas exchange or leakage. The syringes were left on a rotary shaker at room temperature. Within a 24-h period, 1-mL samples were taken at regular intervals and stored at 4°C. Fresh weights of the roots were taken at the experiment's conclusion. Measurement of acetaldehyde and ethanol in the medium was performed using a Varian 3700 GC, fitted with a Varian 800 series auto-sampler and flame ionization detector. The stainless steel column (2-m × 2.16-mm i.d.) was packed with Poropak QS. The GC set-up is as follows: injector, 170°C; detector, 250°C; column programmed between 110°C to 130°C; and N2 carrier gas flow, 30 mL min–1. Calibration was done using samples spiked with known amounts of acetaldehyde and ethanol. All other metabolites were assayed as described previously. In brief, roots were ground in sand and perchloric acid in a 3:1 (w/v) ratio as described by Good and Muench (1993). Shoots were ground in a Corning homogenizer using a methanol-chloroform-water mixture and then separated as described by Lerma et al. (1988). Pyruvate, ethanol, lactate, Glu, 2-oxoglutarate, and Ala levels were measured spectrophotometrically as described by Good and Muench (1993), but adapted for a 96-well ELIZA plate reader (SpectraMax+). All assays were carried out at room temperature.

Survival Assay

The survival assay protocol of Ellis et al. (1999) was used to assess the flooding tolerance of each transgenic line. The only modifications included gassing the medium for 15 min instead of the original 5 min, and plants were transferred to plates containing 0.8% (w/v) agar instead of 1.0% (w/v) agar for the recovery period. The experimental and the wild-type plants were put into one of two different oxygen treatments: 24 h at 0.1% (v/v) O2 without hypoxic pretreatment (24-h NHPT) and 48 h at 0.1% (v/v) O2 after a 24-h HPT at 5% (v/v) O2 (48-h HPT). Shoot and root tip survival, shoot and root weights, and growth of the primary root tip were recorded at the end of the recovery period.

Data Analysis

Two separate survival assay experiments were completed for each transgenic or mutant line. The data for each measurement for the two experiments were analyzed using a two-tailed Student's t test. For each experiment the measured means between the transgenic/mutant and wild-type plants were used for the t test. Differences were considered significant providing that differences from both experiments had P < 0.05. Presented values in the table indicate significant differences where 0.01 ≤ P < 0.05 and where P < 0.01. Data for all enzyme and metabolite experiments are based on a minimum of two separate experiments. Each experiment consisted of three replicate plates per treatment with a C24 control included in each hypoxic chamber as a control. Student's t test (two-tailed) was used to assess whether there was a significant difference between the means of the wild-type and transgenic/mutant plants.

Acknowledgments

We thank Dr. Marc Ellis for valuable assistance in all aspects, particularly the survival assay and experimental set-up for GC analysis. Special thanks go to Drs. Lorraine Tonnet (GC analysis), Nic Savidov, and Peter Hunt for their discussions, comments, and assistance. We also appreciate the excellent technical work performed by Sandra Stops.

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.103.022244.

1

This work was supported in part by a Natural Sciences and Engineering Research Council of Canada (NSERC) grant to A.G.G.

References

  1. Bouny JM, Saglio PH (1996) Glycolytic flux and hexokinase activities in anoxic maize root tips acclimated by hypoxic pre-treatment. Plant Physiol 111: 187–194 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Boyer JS (1982) Plant productivity and environment. Science 218: 443–448 [DOI] [PubMed] [Google Scholar]
  3. Braun KP, Cody RB, Jones JDR, Peterson CM (1995) A structural assignment for a stable acetaldehyde-lysine adduct. J Biol Chem 270: 11263–11266 [DOI] [PubMed] [Google Scholar]
  4. Bucher M, Brändle R, Kuhlemeier C (1994) Ethanolic fermentation in transgenic tobacco expressing Zymomonas mobilis pyruvate decarboxylase. EMBO J 13: 2755–2763 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Davies DD (1980) Anaerobic metabolism and production of organic acids. In PK Stumpf, EE Conn, eds, The Biochemistry of Plants: A Comprehensive Treatise, Vol 2. Academic Press, New York, pp 581–611 [Google Scholar]
  6. Davies DD, Grego S, Kenworthy P (1974) The control of the production of lactate and ethanol by higher plants. Planta 118: 297–310 [DOI] [PubMed] [Google Scholar]
  7. Dennis ES, Dolferus R, Ellis M, Rahman M, Wu Y, Hoeren FU, Grover A, Ismond KP, Good AG, Peacock WJ (2000) Molecular strategies for improving waterlogging tolerance in plants. J Exp Bot 51: 89–97 [PubMed] [Google Scholar]
  8. Dietrich A, König S (1997) Sustrate activation behaviour of pyruvate decarboxylase from Pisum sativum cv. Miko. FEBS Lett 400: 42–44 [DOI] [PubMed] [Google Scholar]
  9. Dolferus R, Jacobs M (1984) Polymorphism of alcohol dehydrogenase in Arabidopsis thaliana (L.) Heynh.: genetical and biochemical characterization. Biochem Genet 22: 817–838 [DOI] [PubMed] [Google Scholar]
  10. Dolferus R, Marbaix B, Jacobs M (1985) Alcohol dehydrogenase in Arabidopsis: analysis of the induction phenomenon in plants and tissue cultures. Mol Gen Genet 199: 256–264 [Google Scholar]
  11. Dolferus R, Peacock WJ, Dennis ES (1994) Differential interactions of promoter elements in stress responses of the Arabidopsis ADH gene. Plant Physiol 105: 1075–1078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dolferus R, Van den Bossche D, Jacobs M (1990) Sequence analysis of two null-mutant alleles of the single Arabidopsis Adh locus. Mol Gen Genet 224: 297–302 [DOI] [PubMed] [Google Scholar]
  13. Drew MC (1997) Oxygen deficiency and root metabolism: injury and acclimation under hypoxia and anoxia. Annu Rev Plant Physiol Plant Mol Biol 48: 223–250 [DOI] [PubMed] [Google Scholar]
  14. Ellis MH, Dennis ES, Peacock WJ (1999) Arabidopsis roots and shoots have different mechanisms for hypoxic stress tolerance. Plant Physiol 119: 57–64 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fox GG, McCallan NR, Ratcliffe RG (1994) Manipulating cytoplasmic pH under anoxia: a critical test of the role of pH in the switch from aerobic to anaerobic metabolism. Planta 195: 324–330 [Google Scholar]
  16. Gancedo C, Serrano R (1989) Energy yielding metabolism. In AH Rose, JS Harrison, eds, The Yeasts, Vol 3. Academic Press, London, pp 205–259 [Google Scholar]
  17. Good AG, Crosby WL (1989a) Anaerobic induction of alanine aminotransferase in barley root tissue. Plant Physiol 90: 1305–1309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Good AG, Crosby WL (1989b) Induction of alcohol dehydrogenase and lactate dehydrogenase in hypoxically induced barley. Plant Physiol 90: 860–866 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Good AG, Muench DG (1993) Long term anaerobic metabolism in root tissue: metabolic products of pyruvate metabolism. Plant Physiol 101: 1163–1168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Grichko VP, Glick BR (2001) Flooding tolerance of transgenic tomato plants expressing the bacterial enzyme ACC deaminase controlled by the 35S, rolD or PRB-1b promoter. Plant Physiol Biochem 39: 19–25 [Google Scholar]
  21. Harberd NP, Edwards KJR (1982) The effect of a mutation causing alcohol dehydrogenase deficiency on flooding tolerance in barley Hordeum vulgare. New Phytol 90: 631–644 [Google Scholar]
  22. Hole DJ, Cobb BG, Hole PS, Drew MC (1992) Enhancement of anaerobic respiration in root tips of Zea mays following low-oxygen (hypoxic) acclimation. Plant Physiol 99: 213–218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Howard EA, Walker JC, Dennis ES, Peacock WJ (1987) Regulated expression of an alcohol dehydrogenase-1 chimeric gene introduced into maize protoplasts. Planta 170: 535–597 [DOI] [PubMed] [Google Scholar]
  24. Huang J, Hirji R, Adam L, Rozwadowski KL, Hammerlindl JK, Keller WA, Selvaraj G (2000) Genetic engineering of glycine betaine production toward enhancing stress tolerance in plants: metabolic limitations. Plant Physiol 122: 747–756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jackson MB, Armstrong W (1999) Formation of aerenchyma and the processes of plant ventilation in relation to soil flooding and submergence. Plant Biol 1: 274–287 [Google Scholar]
  26. Jacobs M, Dolferus R, Van Den Bosshe VB (1988) Isolation and biochemical analysis of ethyl methyl sulfonate induced alcohol dehydrogenase null mutants of Arabidopsis thaliana (L.) Heynh. Biochem Genet 26: 102–112 [DOI] [PubMed] [Google Scholar]
  27. Jaglo-Ottosen KR, Gilmour SJ, Zarka DG, Schabenberger O, Thomashow MF (1998) Arabidopsis CBF1 over-expression induces COR genes and enhances freezing tolerance. Science 280: 104–106 [DOI] [PubMed] [Google Scholar]
  28. Johnson JR, Cobb BG, Drew MC (1994) Hypoxic induction of anoxia tolerance in roots of ADH1 null Zea mays L. Plant Physiol 105: 61–67 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kasuga M, Liu Q, Miura S, Yamaguchi-Shinozaki K, Shinozaki K (1999) Improving plant drought, salt and freezing tolerance by gene transfer of a single stress-inducible transcription factor. Nat Biotechnol 17: 287–291 [DOI] [PubMed] [Google Scholar]
  30. Kennedy RA, Rumpho ME, Fox TC (1992) Anaerobic metabolism in plants. Plant Physiol 100: 1–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Korzeniewski C, Callewaert DM (1983) An enzyme-release assay for natural cytotoxicity. J Immunol Methods 64: 313–320 [DOI] [PubMed] [Google Scholar]
  32. Laszlo A, Saint Lawrence P (1983) Parallel induction and synthesis of PDC and ADH in anoxic maize roots. Mol Gen Genet 192: 110–117 [Google Scholar]
  33. Lemke-Keyes CA, Sachs MM (1989) Genetic variation for seedling tolerance to anaerobic stress in maize germplasm. Maydica 34: 329–337 [Google Scholar]
  34. Lerma C, Hanson AD, Rhodes D (1988) Oxygen-18 and deuterium labelling studies of choline oxidation by spinach and sugar beet. Plant Physiol 88: 695–702 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Liao C-T, Lin C-L (2001) Physiological adaptation of crop plants to flooding stress. Proc Natl Sci Counc Repub China B 25: 148–157 [PubMed] [Google Scholar]
  36. Lorbiecke R, Sauter M (1999) Adventitious root growth and cell-cycle induction in deepwater rice. Plant Physiol 119: 21–30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Maniatis T, Fritsch EF, Sambrook J (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  38. Mergemann H, Sauter M (2000) Ethylene induces epidermal cell death at the site of adventitious root emergence in rice. Plant Physiol 124: 609–614 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Morrell S, Greenway H, Davies DD (1990) Regulation of pyruvate decarboxylase in vitro and in vivo. J Exp Bot 41: 131–139 [Google Scholar]
  40. Mücke U, Wohlfarth T, Fiedler U, Bäumlein H, Rücknagel KP, König S (1996) Pyruvate decarboxylase from Pisum sativum: properties, nucleotide and amino acid sequences. Eur J Biochem 237: 373–382 [DOI] [PubMed] [Google Scholar]
  41. Nelson DE, Rammesmayer G, Bohnert HJ (1998) Regulation of cell-specific inositol metabolism and transport in plant salinity tolerance. Plant Cell 10: 753–764 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Perata P, Vernieri P, Armellini D, Bugnoli M, Tognoni F, Alpi A (1992) Immunological detection of acetaldehyde-protein adducts in ethanol-treated carrot cells. Plant Physiol 98: 913–918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Piao HL, Lim JH, Kim SJ, Cheong G-W, Hwang I (2001) Constitutive overexpression of AtGSK1 induces NaCl stress responses in the absence of NaCl stress and results in enhanced tolerance in Arabidopsis. Plant J 27: 305–314 [DOI] [PubMed] [Google Scholar]
  44. Pronk JT, Yde Steensma H, Van Dijken JP (1996) Pyruvate metabolism in Saccharomyces cerevisiae. Yeast 12: 1607–1633 [DOI] [PubMed] [Google Scholar]
  45. Quimio CA, Torrizo LB, Setter TL, Ellis M, Grover A, Abrigo EM, Oliva NP, Ella ES, Carpena AL, Ito O et al. (2000) Enhancement of submergence tolerance in transgenic rice overproducing pyruvate decarboxylase. J Plant Physiol 156: 516–521 [Google Scholar]
  46. Ricard B, Toai TV, Chourey P, Saglio P (1998) Evidence for the critical role of sucrose synthase for anoxic tolerance of maize roots using a double mutant. Plant Physiol 116: 1323–1331 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rivoal J, Hanson AD (1994) Metabolic control of anaerobic glycolysis. Plant Physiol 106: 1179–1185 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Roberts JKM, Andrade FH, Anderson IC (1985) Further evidence that cytoplasmic acidosis is a determinant of flooding intolerance in plants. Plant Physiol 77: 492–494 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Roberts JKM, Callis J, Jardetzky O, Walbot V, Freeling M (1984a) Cytoplasmic acidosis as a determinant of flooding intolerance in plants. Proc Natl Acad Sci USA 81: 6029–6033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Roberts JKM, Callis J, Wemmer D, Walbot V, Jardetzky O (1984b) Mechanisms of cytoplasmic pH regulation in hypoxic maize root tips and its role in survival under anoxia. Proc Natl Acad Sci USA 81: 3368–3372 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Roberts JKM, Chang K, Webster C, Callis J, Walbot V (1989) Dependence of ethanolic fermentation, cytoplasmic pH regulation, and viability on the activity of alcohol dehydrogenase in hypoxic maize root tips. Plant Physiol 89: 1275–1278 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Robertson AJ, Ishikawa M, Gusta LV, MacKenzie SL (1994) Abscisic acid-induced heat tolerance in Bromus inermis Leyss cell-suspension cultures: Heat-stable, abscisic acid-responsive polypeptides in combination with sucrose confer enhanced thermostability. Plant Physiol 105: 181–190 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Sachs MM, Freeling M, Okimoto R (1980) The anaerobic proteins of maize. Cell 20: 761–767 [DOI] [PubMed] [Google Scholar]
  54. Sripongpangkul K, Posa GBT, Senadhira DW, Brar D, Huang N, Khush GS, Li ZK (2000) Genes/QTLs affecting flood tolerance in rice. Theor Appl Genet 101: 1074–1081 [Google Scholar]
  55. Tadege M, Brandle R, Kuhlemeier C (1998) Anoxia tolerance in tobacco roots: effect of overexpression of pyruvate decarboxylase. Plant J 14: 327–335 [Google Scholar]
  56. Tamminen I, Makela P, Heino P, Palva ET (2001) Ectopic expression of ABI3 gene enhances freezing tolerance in response to abscisic acid and low temperature in Arabidopsis thaliana. Plant J 25: 1–8 [DOI] [PubMed] [Google Scholar]
  57. Thomson CJ, Greenway H (1991) Metabolic evidence for stelar anoxia in maize roots exposed to low O2 concentrations. Plant Physiol 96: 1294–1301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Waterhouse PM, Wang M, Finnegan EJ (2001a) Role of short RNAs in gene silencing. Trends Plant Sci 6: 297–301 [DOI] [PubMed] [Google Scholar]
  59. Waterhouse PM, Wang MB, Lough T (2001b) Gene silencing as an adaptive defense against viruses. Nature 411: 834–842 [DOI] [PubMed] [Google Scholar]
  60. Waters I, Kuiper PJC, Watkin E, Greenway H (1991a) Effects of anoxia on wheat seedlings: I. Interaction between anoxia and other environmental factors. J Exp Bot 42: 1427–1435 [Google Scholar]
  61. Waters I, Morrell S, Greenway H, Colmer TD (1991b) Effects of anoxia on wheat seedlings: II. Influence of O2 supply prior to anoxia on tolerance to anoxia, alcoholic fermentation and sugar levels. J Exp Bot 42: 1437–1447 [Google Scholar]
  62. Xia J-H, Saglio P (1990) H+ efflux and hexose transport under imposed energy status in maize root tips. Plant Physiol 93: 453–459 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Xia J-H, Saglio PH, Roberts JKM (1995) Nucleotide levels do not critically determine survival of maize root tips acclimated to a low-oxygen environment. Plant Physiol 108: 589–595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Xu K, Ronald PC, Mackill DJ (2000) A high-resolution linkage map of the vicinity of the rice submergence tolerance locus Sub1. Mol Gen Genet 263: 681–689 [DOI] [PubMed] [Google Scholar]
  65. Yamaguchi-Shinozaki K, Shinozaki K (2001) Improving plant drought, salt and freezing tolerance by gene transfer of a single stress-inducible transcription factor. Novartis Found Symp 236: 176–186 [PubMed] [Google Scholar]
  66. Zhang J, Van Toai T, Huynh L, Preiszner J (2000) Development of flooding-tolerant Arabidopsis thaliana by autoregulated cytokinin production. Mol Breed 6: 135–144 [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES