Plant Physiology, February 1999, Vol. 119, pp. 599–607, www.plantphysiol.org © 1999 American Society of Plant Physiologists
Mutations Affecting Induction of Glycolytic and Fermentative
Genes during Germination and Environmental
Stresses in Arabidopsis1
Terry R. Conley2, Hsiao-Ping Peng, and Ming-Che Shih*
Department of Biological Sciences, 204 Chemistry Building, University of Iowa, Iowa City, Iowa 52242
Expression of the alcohol dehydrogenase gene (ADH) of Arabidopsis is known to be induced by environmental stresses and regulated developmentally. We used a negative-selection approach to
isolate mutants that were defective in regulating the expression of
the ADH gene during seed germination; we then characterized
three recessive mutants, aar1–1, aar1–2, and aar2–1, which belong
to two complementation groups. In addition to their defects during
seed germination, mutations in the AAR1 and AAR2 genes also
affected anoxic and hypoxic induction of ADH and other glycolytic
genes in mature plants. The aar1 and aar2 mutants were also
defective in responding to cold and osmotic stress. The two allelic
mutants aar1–1and aar1–2 exhibited different phenotypes under
cold and osmotic stresses. Based on our results we propose that
these mutants are defective in a late step of the signaling pathways
that lead to increased expression of the ADH gene and glycolytic
genes.
Exposure to oxygen deprivation (anoxia) and decreased
oxygen availability (hypoxia) due to flooding are common
plant environmental stresses. After a change from aerobic
to anoxic conditions, carbon metabolism of plant cells
switches from aerobic respiration to fermentation (Drew,
1990). Biochemical studies have shown that oxygen deficiency in maize roots induces synthesis of approximately
20 anaerobic polypeptides (for review, see Sachs et al.,
1980, 1996), several of which have been identified as glycolytic and fermentative enzymes, including ADH and
GAPDH (Freeling and Bennett, 1985; Ricard et al., 1989;
Russell and Sachs, 1989; Xie and Wu, 1989; Yang et al.,
1993). Several cis-acting elements and trans-acting factors
involved in anaerobic induction of the ADH genes in maize
and Arabidopsis have been identified (Ferl and Laughner,
1989; Dolferus et al., 1994; Kyozuka et al., 1994; Hoeren et
al., 1998). Recent reports indicate that Ca21 may play an
important role in the induction of ADH gene expression
during anoxia in maize (Subbaiah et al., 1994a, 1994b;
Sedbrook et al., 1996).
Most crop plants, including barley, maize, sorghum, and
wheat, can tolerate only very transient flooding (Kennedy
1
This work was supported by the U.S. Department of Agriculture National Research Initiative Competitive Grants Program
(grant nos. 9401454 and 9700603 to M.-C.S.).
* Corresponding author; e-mail [email protected];
fax 1–319 –335–3620.
et al., 1992). In contrast, rice plants can survive much
longer under anaerobic conditions. Formation of aerenchyma, characterized by continuous gas spaces in roots
and shoots (Drew et al., 1979; Justin and Armstrong, 1987),
has been proposed to correlate with tolerance to flooding
(Justin and Armstrong, 1987). Campbell and Drew (1983)
have shown that lysogenic aerenchyma formation occurs in
the root cortex of maize during hypoxia. A recent report
indicates that an ethylene signal is required for cell death
during aerenchyma formation induced by hypoxia (He et
al., 1996). These results, in conjunction with reports that
Ca21 is also involved in the induction of the ADH gene in
the early stage of anoxia (Subbaiah et al., 1994a, 1994b;
Sedbrook et al., 1996), suggest that multiple signaling
events may be involved.
In addition to hypoxia or anoxia, other environmental
stresses, including cold and dehydration, also induce the
expression of the ADH gene in Arabidopsis (Dolferus et al.,
1994). Recent genetic evidence suggests that both an ABAdependent and an ABA-independent pathway are involved
in mediating the response to cold and osmotic stress (Ishitani et al., 1997; Zhu et al., 1997). The molecular relationship, if any, between ADH induction by hypoxia or anoxia
and response to cold and osmotic stress is essentially
unknown.
We are taking advantage of the amenability of Arabidopsis to genetic manipulation to dissect these complex signaling pathways. It has been shown that mRNA levels for both
ADH and GAPC genes increase during anoxia (Chang and
Meyerowitz, 1986; Yang et al., 1993). Analysis of transgenic
plants containing ADH- or GAPC-promoter::GUS fusion
genes show that transcriptional control is involved in anaerobic induction of both genes in Arabidopsis (Yang et al.,
1993; Dolferus et al., 1994). We have used a genetic scheme
that allowed us to identify regulatory mutants that were
defective in the induction of the ADH gene during seed
germination. Characterization of some of these mutants
revealed that they were also defective in other stressinduced signal transduction pathways that lead to the activation of several glycolytic genes.
Abbreviations: ADH, alcohol dehydrogenase; GAPDH,
glyceraldehyde-3-phosphate dehydrogenase; PDC, pyruvate decarboxylase; RT, reverse transcriptase.
599
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
600
Conley et al.
MATERIALS AND METHODS
Generation of an ADH-Promoter::GUS Reporter Gene
Construct in the Binary Vector pBI101
The genomic clone pAtAdh-7 served as the source of a
2.4-kb HindIII-SmaI fragment, which included approximately 1 kb of sequence upstream from the transcription
start site of the ADH gene of Arabidopsis (Chang and
Meyerowitz, 1986). The 2.4-kb HindIII-SmaI fragment was
subcloned into pBS (plasmid BlueScript, Stratagene), generating pAdh. This plasmid, along with two oligonucleotide primers, was used in PCR to generate an approximately 1-kb ADH-promoter fragment. The upstream
primer was complementary to the T7 RNA polymerasepromoter site in the pBS sequences flanking the multilinker
region, whereas the downstream primer encompassed the
ADH translation initiation codon. The downstream primer
was designed to engineer a new HincII site by changing the
original 59-GTTGATAATG-39 sequence (the translation initiation codon is underlined) to 59-GTTGACCATG-39 (the
new HincII site is designated by underlining; modified
bases are designated in bold). The HincII site served to
facilitate subsequent manipulations. The 1-kb PCR product
was subcloned into pBS, generating pAdh(Pro). Digesting
pAdh(Pro) with XbaI and HincII produced a 1072-bp fragment that was subcloned directionally into the XbaI and
SmaI sites of the multilinker region of the binary vector
pBI101 (Jefferson et al., 1987). Restriction mapping was
used to confirm the presence and the correct orientation of
the ADH-promoter insert in the resulting plasmid clone
pBI101/Adh-Gus.
Generation of Transgenic Arabidopsis Plants
Root transformation was used to generate transgenic
Arabidopsis plants, essentially as described by Valvekens
et al. (1988). Rooted transformed plantlets were transferred
to Jiffy Mix Plus soil mix (Hummert International, St.
Louis, MO). Plants were grown in environmental chambers
(Percivall, Boone, IA) at 20°C under a mixture of incandescent and fluorescent lamps (fluence rate, 100 mE m22s21)
under long-day conditions (16-h light/8-h dark).
The four independent transformants that were obtained
were propagated to maturity for subsequent analysis. All
results reported here are based on ADH::GUS expression in
F4 progeny of the line designated AG2. The F1 progeny of
the primary transformant line of AG2 had an approximate
3:1 segregation of kanamycin-resistant/kanamycin-sensitive
progeny. Kanamycin-resistant F1 plants were analyzed
through the F2 and F3 generations to identify homozygous
lines. The presence of a single copy of the ADH::GUS construct in line AG2 was confirmed by genomic Southern
analysis.
Plant Physiol. Vol. 119, 1999
sterilized by soaking in 1.5% (v/v) NaOCl with 0.04%
(w/v) sodium dodecyl sulfate for 30 min, followed by five
sterile water rinses. Seeds were then soaked in 0.25% (v/v)
aqueous ethyl methanesulfonic acid for 10 to 12 h (Somerville and Ogren, 1982) and then rinsed several times with
sterile water. The seeds were planted on Jiffy Mix Plus soil
mix and grown to maturity. Approximately 4000 plants
from each of the three 0.5-g aliquots of mutagenized seeds
were grown to maturity. The M2 seeds from each of the
resulting populations of seeds were collected, dried, and
stored in a sealed container at 4°C.
Allyl Alcohol Selection
Approximately 50,000 seeds from each of the three mutagenized populations were surface-sterilized as described
above. The seeds were then soaked in sterile water at room
temperature for 6 h. The imbibed seeds were next exposed
to 45 mm (aqueous) allyl alcohol for 2 h at room temperature (Jacobs et al., 1988), followed by five rinses with sterile
water and treatment overnight with 15 mm GA (aqueous) at
4°C. Each aliquot of seeds was subdivided and transferred
into ten 250-mL flasks, each containing liquid germination
medium (Valvekens et al., 1988). The flasks were incubated
under fluorescent lights in an environmental chamber on a
shaker operated at 150 rpm. Green seedlings that appeared
after several days were transferred to Petri dishes containing germination medium solidified with 0.8% agar.
Growth Conditions and Stress Treatment
Seeds of line AG2 and aar mutants were surfacesterilized and treated with 15 mm GA at 4°C overnight.
Seeds were sown onto plates with Murashige and Skoog
medium containing 1% Suc and 0.8% agar and grown at
20°C with 16-h light/8-h dark cycles. After 1 week, seedlings were transferred to fresh Murashige and Skoog plates
in a vertical position under the same conditions for an
additional 3 weeks. For anoxic treatments, plants were
submerged in liquid Murashige and Skoog medium
through which nitrogen gas (.99% purity) was bubbled to
purge O2. For hypoxic treatments, similar conditions were
used except that pure N2 gas was replaced with gas containing 4.5% to 5% O2 and balanced with N2.
We used a modification of the procedures described by
Dolferus et al. (1994) for osmotic-stress experiments. Arabidopsis plants were submerged in liquid Murashige and
Skoog medium containing 0.6 m mannitol through which
air (18% O2) was bubbled for 24 h at 20°C. For cold treatment, plants were submerged in aerated liquid Murashige
and Skoog medium at 4°C for 24 h.
GUS and ADH Enzymatic Assays
Chemical Mutagenesis
The AG2 line of the transgenic ADH::GUS was mutagenized. Three 0.5-g aliquots of seeds were placed separately into sterile 50-mL centrifuge tubes and surface-
Histochemical staining and enzyme activity assays for
GUS were performed essentially as described by Jefferson
et al. (1987). Fluorescence of the 4-methylumbelliferyl
product was quantified using a minifluorometer (model
TKO-100, Hoefer Scientific Instruments, San Francisco, CA).
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
Arabidopsis Mutants Defective in Stress Responses
ADH enzyme activity assay was performed according to
the procedures described by Freeling (1973) and modified
by Xie and Wu (1989). The assay uses ethanol as the substrate and measures the production of NADH in a spectrophotometer (model DU 64, Beckman). One unit of ADH
enzyme is defined as an increase in A340 of 0.01 per min
(Xie and Wu, 1989).
RNA Isolation and RT-PCR Reactions
Total RNA was isolated by an acidic phenol protocol
adapted from the procedures described in Chomczynski
and Sacchi (1987). Prior to RT-PCR reactions, RNA samples
were treated with DNase I twice to deplete contaminating
genomic DNA. First-strand cDNA synthesis was performed in a 50-mL reaction mixture containing 2 mg of
RNA, 1 mg of random hexamer, 40 mm each of the four
deoxynucleotides, 5 mL of 103 buffer, 200 units of Moloney
murine leukemia virus RT, and 20 units of ribonuclease
inhibitor. The reaction mixture was incubated at 42°C for
60 min and stopped by heating at 65°C for 10 min.
PCR was performed in a DNA Thermal Cycler (Perkin
Elmer/Cetus) programmed for 28 cycles of 1 min at 94°C,
1 min at 55°C, and 1 min at 72°C. To find the amounts of
cDNA suitable for linear amplification, we used 1, 2, 4, 6, 8,
and 10 mL of RT reaction mixture from the hypoxic-treated
AG2 line. For all of the genes shown in Table I, there were
linear increases of PCR products in reactions with 1, 2, and
4 mL of cDNA (data not shown). Therefore, we used 1 or 2
mL of the RT reaction mixture for subsequent PCR and
analyzed the products using agarose-gel electrophoresis.
Sequencing the products from the reactions of the AG2 line
confirmed the identities of the PCR products for each
primer set .
We quantified the PCR products by analyzing the digitized images of agarose gels using the ImageQuant software program, version 1.1 (Molecular Dynamics, Sunnyvale, CA). Nucleotide sequences of the primer pairs for
amplification of each gene and the sizes of the corresponding products appear in Table I. The accession numbers (in
parentheses) for the genes on which the primer sequences
Table I. Primer sequences for RT-PCR
Gene
Primera
Product Length
bp
ADH
PDC1
PDC2
GAPC
bATP
a
TGTCACACCGATGTTTAC
GGTCGAATCTTTTAGAGTTAA
GAAGCTGCGGTGGAAGCAAC
TCCTAGAGTTGCACCAACAG
GGCTCATGAACTTATCGATA
TCTCAGCAAGCACAGCAGAC
GGAGCTGACTACGTTGTTGAGT
CGTTGTCGTACCATGACACCAA
GATCATGACATCTCTCGAGG
TGGTAAGGAGCAAGGAGATC
525
682
740
670
350
The primers shown on top rows represent forward primers used
for each gene, whereas those shown on bottom rows correspond to
the reverse primers.
601
were based are as follows: ADH (M12196), GAPC (M64116),
PDC1 (U71121), and PDC2 (U71122).
Chemical and Enzyme Suppliers
We purchased synthetic plant hormones (benzylaminopurine, GA3, and kinetin), antibiotics (kanamycin, rifampicin,
and vancomycin), and other chemicals from Sigma, and
5-bromo-4-chloro-3-indolyl-b-d-glucuronic acid cyclohexylammonium salt from Gold Biotechnology (St. Louis, MO).
Restriction and modification enzymes were from New England BioLabs and Promega, and the sequenase and radiochemicals used in DNA sequencing were from Amersham.
RESULTS
Isolation of Mutants Affecting ADH Gene Expression
The genetic scheme we adapted uses the ADH enzyme as
a selectable marker. ADH normally functions in alcoholic
fermentation, catalyzing the reduction of acetaldehyde to
ethanol. However, ADH also uses allyl alcohol (2-propen1-ol) as a substrate, forming acrolein (2-propenal), which is
highly toxic to plant cells. Therefore, in the presence of allyl
alcohol, wild-type plants with normal levels of functional
ADH enzyme will die, whereas plants that lack ADH will
survive (Freeling and Bennett, 1985; Jacob et al., 1988). This
property can be exploited to isolate mutants that are defective in regulating the expression of the ADH gene.
Several transgenic Arabidopsis lines that contain one
copy of ADH::GUS were constructed. We chose one of
these lines, AG2, in which the expression pattern of the
ADH::GUS transgene reflected the expression pattern of
the endogenous ADH gene (Chang and Meyerowitz, 1986;
Dolferus et al., 1994), for ethyl methanesulfonic acid mutagenesis. To screen for mutants affecting ADH gene expression, 150,000 M2 seeds were treated with allyl alcohol
(for details, see “Materials and Methods”). A total of 40
seeds germinated and produced green seedlings. These we
designated as aar (allyl alcohol resistant) mutants, which
we further divided into two classes. The first, with mutations to the endogenous ADH gene, were to have normal
ADH::GUS expression. The second, with mutations in
genes involved in the signaling pathway(s) leading to the
induction of ADH, were to have reduced levels of expression of both the endogenous ADH gene and the GUS reporter gene. To distinguish between these two classes, we
stained imbibed seeds of aar mutants for GUS activity.
Thirteen of the 40 aar mutants had pronounced GUS staining after 6 h or less of incubation, similar to the parental
AG2 plants, indicating that the ADH::GUS gene was expressed. In contrast, the remaining 27 aar plants exhibited
no detectable GUS staining even after 24 h of incubation
(data not shown). We characterized three of these mutants,
aar7, aar10, and aar17, which are described below.
Table II showed that F1 progeny from crosses between
line AG2 and each of the three aar mutants were sensitive
to allyl alcohol treatment. In addition, allyl alcoholsensitive and -resistant plants in the F2 progeny from all
three crosses showed a 3:1 ratio. These results indicated
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
602
Conley et al.
Plant Physiol. Vol. 119, 1999
Table II. Segregation test of aar mutants
Cross
No. of F1
Seeds
aar Seeds in F1a
No. of F2
Seeds
aar Seeds in F2
WT:aarb
x2
AG2 3 aar 7
AG2 3 aar 10
AG2 3 aar 17
42
55
32
0
0
0
293
346
414
73
101
112
3:1
3:1
3:1
1.65
2.93
2.79
a
The aar phenotype corresponds to seeds that can germinate after allyl alcohol treatment.
is defined as seeds that fail to germinate after allyl alcohol treatment.
that each aar mutant contains a monogenic and recessive
mutation. Pair-wise crosses among these mutants showed
that aar7 and aar17 belonged to the same complementation
group, whereas aar10 constituted a second group (Table
III). Therefore, aar7 and aar17 were renamed aar1–1 and
aar1–2, respectively, whereas aar10 was renamed aar2–1.
b
WT
Anoxic and Hypoxic Responses in Mature Plants
To determine whether mutations in the AAR1 and AAR2
genes affected regulation of the ADH::GUS transgene in
mature plants in response to anoxia and hypoxia, 4-weekold plants from line AG2 and from the aar1–1, aar1–2, and
aar2–1 mutants were subjected to hypoxic or anoxic treatments for 8 h and harvested for the assay of GUS
Effects of Mutations in AAR Genes during
Seed Germination
To quantify the effects of mutations on the expression of
the ADH::GUS transgene during seed germination, we
compared the levels of GUS activity of the wild-type AG2
line and the aar mutants (Fig. 1A). In AG2 GUS activity was
highest in imbibing seeds (Fig. 1A, d 0) and 1-d-old germinating seedlings. The activity decreased 50% by 5 d, and
had decreased to a steady-state level by 15 d, which was
less than 10% of the 1-d level. In contrast, GUS activity was
much lower in the imbibing seeds and the germinating
seedlings in all three aar mutants. In 1-d-old germinating
seedlings, GUS activity in line AG2 was 8- to 10-fold higher
than in the aar mutants. In 5-d-old seedlings, GUS activity
in line AG2 was 2- to 3-fold higher than in the mutant
plants. In contrast, at later developmental stages GUS activity was very similar in line AG2 and in all of the mutants
(Fig. 1A, d 15).
We also measured ADH activity in germinating seedlings of AG2 and aar mutants. Figure 1B shows that the
pattern of ADH activity level in AG2 was similar to that of
GUS, except that ADH activity decreased more rapidly,
reaching an undetectable level in 10-d-old seedlings. However, as in the case for GUS, there was much lower ADH
activity in imbibing seeds and germinating seedlings than
in any of the three aar mutants. These results showed that
mutations in the AAR1 and AAR2 genes affected the expression of both the ADH::GUS transgene and the endogenous ADH gene during germination.
Table III. Complementation test
Cross
No. of F1 Seeds
Tested
aar7 3 aar10
aar7 3 aar17
aar10 3 aar7
aar10 3 aar17
aar17 3 aar7
aar17 3 aar10
27
35
17
58
44
30
a
Allyl Alcohol Sensitivitya
Sensitive
Resistant
27
0
17
58
0
30
0
35
0
0
44
0
See footnote in Table II for definition of allyl alcohol sensitivity.
Figure 1. GUS and ADH activities of AG2 and of the aar mutants
during seed germination. Seedlings of AG2 and aar1–1, arr1–2, and
aar2–1 at different developmental stages were harvested and assayed
for GUS and ADH activities as described in “Materials and Methods.”
A, GUS activity is expressed as pmol 4-methylumbelliferone min21
mg21 protein. B, One unit of ADH enzyme is defined as an increase
in A340 of 0.01 per min. The data presented are the averages of three
independent treatments. Plants grown at different times were used for
replicate treatments. Bar graphs at each time point (from left to right)
represent activities for AG2, aar1–1, aar1–2, and aar2–1. Bars indicate SD.
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
Arabidopsis Mutants Defective in Stress Responses
activity. The data showed that after 8 h, there were 10- and
15-fold increases in GUS activity, respectively, in anoxicand hypoxic-treated AG2 plants (Fig. 2). In contrast, there
was no increase in GUS activity in either anoxic- or
hypoxic-treated plants in all three aar mutants (Fig. 2).
To examine the expression pattern of the endogenous
ADH gene, we subjected the 4-week-old AG2 plants and
the aar mutants to 8 h of anoxic or hypoxic treatments.
Total RNA was isolated and analyzed by RT-PCR. The
nuclear gene bATP that encodes the b-subunit of mitochondrial ATP synthase, the expression of which was not affected by either hypoxic or anoxic treatment (T.R. Conley
and M.-C. Shih, unpublished data), was used as an internal
standard. The results from one set of representative RTPCR reactions are illustrated in Figure 3. The data show
that under normoxic conditions, there was little ADH
mRNA in either line AG2 or the three aar mutants (Fig. 3A,
lanes N). Levels of ADH mRNA were greatly induced in
anoxic- or hypoxic-treated AG2 plants (Fig. 3A, lanes N
and H). Quantification of the RT-PCR products indicated
that anoxic treatment resulted in a 25-fold increase in ADH
mRNA level in AG2, whereas hypoxic treatment resulted
in a 50-fold induction (Fig. 3B). In contrast, there was no
increase in the ADH mRNA level in either anoxic- or
hypoxic-treated plants in any of the three aar mutants.
These results demonstrate that mutations in AAR1 and
AAR2 genes affected anoxic and hypoxic induction of expression of both the endogenous ADH gene and the
ADH::GUS transgene in mature plants.
In addition to ADH, PDC was also required to catalyze
the conversion of pyruvate to ethanol during alcoholic
fermentation. Therefore, we examined whether the expression of PDC1 and PDC2 was affected in the aar1 and aar2
mutants. The data show that there was a low level of PDC1
mRNA present in AG2 plants grown under normoxic conditions (Fig. 4A). However, the PDC1 mRNA level increased by greater than 20- and 40-fold, respectively, in
anoxic- and hypoxic-treated AG2 plants (Fig. 4, top panel).
In contrast, in the three aar mutants, PDC1 mRNA levels
Figure 2. Anoxic and hypoxic induction of GUS activity in mature
plants. Twenty-day-old AG2 and aar1–1, aar1–2, and aar2–1 plants
were subjected to anoxic or hypoxic treatment and assayed for GUS
activity as described in “Materials and Methods.” The data presented
are the averages of three independent treatments. Bars indicate SD.
603
Figure 3. RT-PCR analysis of anoxic and hypoxic induction of ADH.
Total RNA (1 mg) from anoxic- or hypoxic-treated plants was used in
a 100-mL reaction for first-strand cDNA synthesis. One-tenth of the
synthesized cDNA from each treatment was then used in subsequent
PCR reactions. A, RT-PCR products for ADH were analyzed by
agarose-gel electrophoresis. The sizes of PCR products for each gene
are 525 bp for ADH and 350 bp for bATP. The notations on top of
each lane represent treatment conditions of normoxia (N), anoxia (A),
and hypoxia (H). B, Digitized images of the ADH bands in A were
quantified and normalized to the bATP band in each lane. The
normalized mRNA levels from normoxic-treated plants (solid bars) of
AG2 and each aar mutant were used as the reference levels to
calculate magnitudes of induction. Checked and striped bars correspond to anoxic- and hypoxic-treated samples, respectively. The data
presented are the averages of three independent treatments. Bars
indicate SD.
were not induced by either anoxic (lanes A) or hypoxic
(lanes H) treatments. Figure 6 also shows that there was a
low level of PDC2 mRNA present in AG2 and aar mutant
plants grown under normoxic conditions (bottom panels).
However, unlike PDC1, levels of PDC2 mRNA were not
induced in either anoxic- or hypoxic-treated AG2 plants or
in any of the three aar mutants. These results showed that
only the expression of PDC1, not that of PDC2, was inducible by anoxic or hypoxic treatment in Arabidopsis, and
that mutations in AAR1 and AAR2 genes could affect PDC1
induction.
The ADH and PDC genes belong to a class of genes that
function exclusively in alcoholic fermentation. The data
presented here show that these genes were expressed at
extremely low levels under normoxic conditions, but could
be greatly induced by anoxic and hypoxic treatments. Russell and Sachs (1989) and Yang et al. (1993) have shown that
glycolytic genes whose functions are required for both
glycolysis and fermentation were expressed at high levels
under normoxia, but could be induced further by anoxia or
hypoxia. We chose to examine the expression pattern of the
GAPC gene, which encodes cytosolic GAPDH, to determine whether its expression was affected in aar mutants.
Figure 5A shows that there was a detectable mRNA level of
GAPC in AG2 plants grown under normoxic conditions
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
604
Conley et al.
Plant Physiol. Vol. 119, 1999
mRNA levels measured for both ADH and PDC1, whereas
exposure to 0.6 m mannitol resulted in approximately 30fold increases in ADH and PDC1 mRNA levels (Fig. 6, A
[lanes C and M] and B). However, the data presented in
Figure 6 also show that the three mutants exhibited different allele-specific effects on induction of ADH and PDC1
genes by cold and mannitol treatment. In aar1–1 mannitol
treatment resulted in a 15-fold induction of ADH (Fig. 6, A
[top panel] and B). In contrast, the induction of ADH by
cold treatment (Fig. 6A, lanes C [top panel]) and the induction of PDC1 by cold and mannitol treatments (Fig. 6A,
lanes C and M [bottom panel]) were abolished completely
in aar1–1. In aar1–2 the induction of PDC1 by mannitol
treatment and the induction of ADH by cold and mannitol
treatment were abolished, whereas there was a 30-fold
induction of the PDC1 gene by mannitol treatment. In
aar2–1 the induction of both ADH and PDC1 by cold or
mannitol treatment was abolished completely.
DISCUSSION
Figure 4. RT-PCR analysis of anoxic and hypoxic induction of PDC1
and PDC2. Analysis of RT-PCR products (A) and quantification of
levels of induction (B) of PDC1 and PDC2 were performed as described in Figure 3. The symbols used are as described in the Figure
3 legend. The sizes of PCR products are 682 bp for PDC1 and 740 bp
for PDC2.
(lanes N) and that the expression of GAPC was induced by
both anoxic and hypoxic treatments (lanes A and H). Quantification of the RT-PCR products indicated that the anoxic
treatment resulted in a 3-fold increase and the hypoxic
treatment resulted in a 5-fold increase in the GAPC mRNA
level in AG2 plants (Fig. 5B). Figure 5 also shows that there
were detectable and similar-to-wild-type GAPC mRNA
levels in the three aar mutants grown under normoxic conditions. In contrast to the wild-type AG2 line, the expression
level of GAPC remained unchanged after the anoxic or hypoxic treatments in all three mutants. Therefore, mutations in
AAR1 or AAR2 genes affect anoxic and hypoxic induction of
the GAPC gene, as well as induction of ADH and PDC1 genes.
We have identified and characterized trans-regulatory
mutations that affect the expression of the ADH gene in
Arabidopsis. The genetic approach used here takes advantage of the fact that ADH catalyzes the conversion of allyl
alcohol into a toxic compound (acrolein) that is lethal to
wild-type plants. In addition, we have used a transgenic
line bearing an ADH::GUS construct as the starting point
for mutagenesis. The GUS marker gene allowed us to distinguish between the two classes of mutants that were
predicted to be isolated: mutations affecting the promoter
or coding region of the endogenous ADH gene, and mutations affecting any step in the signal transduction pathway
leading to the activation of the ADH gene. In our screening,
Allelic-Specific Effects of aar1 Mutants under Cold and
Osmotic Stress
The expression of the ADH gene is also affected by other
environmental conditions, such as cold and osmotic stress,
in several plant species (Xie and Wu, 1989; Dolferus et al.,
1994). Our RT-PCR analysis showed that in AG2 plants,
levels of ADH and PDC1 mRNA were greatly increased by
cold or mannitol treatment (Fig. 6). However, the mRNA
levels for PDC2 and GAPC were not affected by either
treatment (data not shown). Therefore, we examined
whether the induction of ADH and PDC1 by cold or mannitol treatment was affected in the aar mutants. In AG2,
following 24 h at 4°C, there were 35-fold increases in
Figure 5. Effect of aar mutations on anoxic and hypoxic induction of
GAPC. A, Levels of GAPC mRNA from normoxic-treated (lanes N),
anoxic-treated (lanes A), and hypoxic-treated (lanes H) AG2 plants
and aar mutants were analyzed by relative RT-PCR. The sizes of the
PCR products are 670 bp for GAPC and 350 bp for bATP. B,
Quantification of levels of induction of GAPC by anoxia and hypoxia
in AG2 and aar mutants were calculated as described for Figure 3B.
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
Arabidopsis Mutants Defective in Stress Responses
Figure 6. Effects of cold and mannitol on ADH and PDC1 expression. A, RNA samples from AG2 and aar mutants grown under
normal conditions (lanes N), cold treatment (lanes C), and mannitol
treatment (lanes M) were subjected to RT-PCR analysis. B, Quantification of cold and mannitol induction of ADH and PDC1 was
performed as described for Figure 3B. Solid bars correspond to
samples from control experiments, whereas checked and striped bars
correspond to samples from cold and mannitol treatments, respectively. The data presented are the averages of three independent
treatments. Bars indicate SD.
about one-third of the putative mutants were potentially
members of the first group, whereas the remaining twothirds potentially belong to the latter group of mutants.
Characterization of three of these regulatory mutants has
provided us with information for understanding the molecular mechanism of induction of ADH and PDC1 genes
by three different environmental stresses. Li et al. (1995)
used a similar screening scheme successfully to isolate
trans-regulatory mutants of the CAB gene in Arabidopsis.
Expression of ADH in Arabidopsis is known to be induced by several environmental stresses and to be regulated developmentally (Dolferus et al., 1994). The selection
scheme we have adapted allowed us to identify mutants
that were defective in regulating the expression of the ADH
gene in germinating seedlings. Three such mutants, aar1–1,
aar1–2, and aar2–1, were also defective in regulating the
anoxic and hypoxic induction of the ADH gene in mature
plants. These results can be interpreted in two ways. First,
605
the regulatory pathways controlling the expression of ADH
during germination and anoxia/hypoxia are identical. Second, two signaling pathways share some common steps
and the aar mutants characterized here affect these steps. If
the second hypothesis is correct, it is conceivable that some
of the other aar mutants identified in these screens may
affect ADH expression during germination, but exhibit normal anoxic or hypoxic induction of ADH. Experiments are
under way to further characterize the other aar mutants
that we have isolated.
In addition to the ADH gene, glycolytic and other fermentative genes are induced by anoxia and hypoxia (Yang
et al., 1993; Sachs et al., 1996; Rivoal et al., 1997). We have
also examined the effects of the aar mutations on the expression of other glycolytic genes. Among these genes,
GAPC gene products are required for both aerobic respiration and anaerobic fermentation, whereas ADH and PDC1
gene products are required only for alcoholic fermentation.
The expression patterns of these genes reflect their physiological roles. ADH and PDC1 genes are expressed at extremely low levels in plant cells under normal growth
conditions and their expression is strongly induced by
hypoxia or anoxia. In contrast, the GAPC gene is expressed
constitutively at a fairly high level under normal growth
conditions, but its expression can be induced further by
hypoxia or anoxia. This raises the interesting question of
whether similar regulatory elements are involved in the
activation of these two different classes of genes. Our data
show that mutations in AAR1 and AAR2 genes affected
induction of both classes of glycolytic genes by anoxia and
hypoxia. These results suggested that similar regulatory
elements mediated the induction of these two different
classes of genes in response to anoxic and hypoxic stresses.
Cold and osmotic stress also induce the expression of
ADH in several plant species (Xie and Wu, 1989; Dolferus
et al., 1994). Our results show that both ADH and PDC1
genes could be induced by cold and mannitol treatment in
Arabidopsis. Recent genetic and molecular studies suggested that both ABA-dependent and -independent pathways interact and converge to activate the expression of
downstream genes (Ishitani et al., 1997, 1998; Zhu et al.,
1998). Our results show that the aar2–1 mutant was defective in the induction of ADH and PDC1 by anoxic/hypoxic,
cold, and osmotic stresses (Fig. 6). This suggests that the
three stress-induced signaling pathways shared some common intermediate steps and that aar2–1 was defective in
one of these steps. However, our results also show that the
induction of ADH and PDC1 by these three stresses was
affected differently in the two allelic mutants, aar1–1 and
aar1–2. There, the induction of ADH and PDC1 by the
three stresses was abolished, with two exceptions: in
aar1–1 there was a 15-fold induction of ADH in response
to the mannitol treatment, which was about 50% of the
induction level of the wild-type AG2 line (Fig. 6); second,
the induction of PDC1 by cold treatment was unaffected
in aar1–2.
A report by de Bruxelles et al. (1996) indicated that
although induction of the ADH gene by osmotic and cold
stresses required some common cis-acting promoter elements, additional, but distinct, regulatory elements were
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
606
Conley et al.
required for each pathway. Their results also showed that
regulatory elements different from those involved in cold
and osmotic responses were required for induction of ADH
by anoxia and hypoxia. This implies that transcription
complexes for the ADH gene under cold, osmotic, and
low-oxygen stresses may share some common factors, but
probably also embody some discrete factors. If this hypothesis is true, one can propose that the AAR1 gene may
encode a transcription factor shared by these complexes,
and that mutations in aar1–1 and aar1–2 have different
effects on the interaction between the AAR1 factor and
factors that are specific to each stress condition.
Current information suggests complex patterns of interactions among the signal transduction pathways elicited by cold, osmotic, and low-oxygen stresses in plants
(de Bruxelles et al., 1996; Stockinger et al., 1997; Zhu et al.,
1997). Ishitani et al. (1997) have isolated mutants that
affect cold- and osmotic-stress signaling pathways by a
high-throughput screening approach. Some of these mutants are specifically defective in cold- or osmotic-stressinduced signaling pathways (Ishitani et al., 1997; Zhu et
al., 1998). It is likely that some of the aar mutants we
have isolated are specifically defective in the signaling pathway that corresponds to decreased oxygen. Combining studies of these different classes of mutants
should greatly facilitate the elucidation of these signaling
pathways.
ACKNOWLEDGMENTS
We thank Drs. Joseph Frankel and Andy Wang for comments on
the manuscript. We also thank the Arabidopsis Biological Resource Center at the Ohio State University, Columbus, for providing us with expressed sequence tag clones.
Received September 8, 1998; accepted October 28, 1998.
LITERATURE CITED
Campbell R, Drew MC (1983) Electron microscopy of gas space
(aerenchyma) formation in adventitious roots of Zea mays L.
subjected to oxygen shortage. Planta 157: 350–357
Chang C, Meyerowitz E (1986) Molecular cloning and DNA sequence of the Arabidopsis thaliana alcohol dehydrogenase gene.
Proc Natl Acad Sci USA 83: 1408–1412
Chomczynski P, Sacchi N (1987) Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform
extraction. Anal Biochem 162: 156–160
de Bruxelles GL, Peacock WJ, Dennis ES, Dolferus R (1996)
Abscisic acid induces the alcohol dehydrogenase gene in Arabidopsis. Plant Physiol 111: 381–391
Dolferus R, Jacobs M, Peacock W, Dennis E (1994) Differential
interactions of promoter elements in stress responses of the
Arabidopsis Adh gene. Plant Physiol 105: 1075–1087
Drew MC (1990) Sensing soil oxygen. Plant Cell Environ 13:
681–693
Drew MC, Jackson MB, Giffard S (1979) Ethylene-promoted adventitious rooting and development of cortical air spaces (aer-
Plant Physiol. Vol. 119, 1999
enchyma) may be adaptive response to flooding in Zea mays L.
Planta 153: 217–224
Ferl R, Laughner B (1989) In vivo detection of the regulatory factor
binding sites of Arabidopsis thaliana Adh. Plant Mol Biol 12:
357–366
Freeling MM (1973) Simultaneous induction by anaerobiosis or
2,4-D of multiple enzymes specified by two unlinked genes:
differential ADH1-ADH2 expression in maize. Mol Gen Genet
127: 215–227
Freeling MM, Bennett DC (1985) Maize Adh1. Annu Rev Genet
19: 297–323
He C-J, Page WM, Drew MC (1996) Transduction of an ethylene
signal is required for cell death and lysis in the root cortex of
maize during aerenchyma formation induced by hypoxia. Plant
Physiol 112: 463–472
Hoeren FU, Dolferus R, Wu Y, Peacock WJ, Dennis ES (1998)
Evidence for a role for AtMYB2 in the induction of the Arabidopsis alcohol dehydrogenase gene (ADH1) low oxygen. Genetics 149: 479–490
Ishitani M, Xiong L, Lee H, Stevenson B, Zhu JK (1997) Genetic
analysis of osomotic and cold stress signal transduction in Arabidopsis: interactions and convergence of abscisic acid-dependent and abscisic acid-independent pathways. Plant Cell 9:
1935–1949
Ishitani M, Xiong L, Lee H, Stevenson B, Zhu JK (1998) HOS1, a
genetic locus involved in cold-responsive gene expression in
Arabidopsis. Plant Cell 10: 1151–1162
Jacobs M, Dolferus R, Van Den Bossche D (1988) Isolation and
biochemical analysis of ethyl methanesulfonate-induced alcohol
dehydrogenase null mutants of Arabidopsis thaliana (L.) Heynh.
Biochem Genet 26: 105–122
Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusion:
b-glucuronidase as a sensitive and versatile gene fusion marker
in higher plants. EMBO J 6: 3901–3907
Justin SHF, Armstrong W (1987) The anatomical characteristics of
roots and plant response to soil flooding. New Phytol 106:
465–495
Kennedy RA, Rumpho M, Fox TC (1992) Anaerobic metabolism in
plants. Plant Physiol 100: 1–6
Koornneef M, Stam P (1992) Genetic analysis. In C Koncz, N-H
Chua, J Schell, eds, Methods in Arabidopsis Research. World
Scientific, Singapore, pp 83–99
Kyozuka J, Olive M, Peacock W, Dennis E, Shimamoto K (1994)
Promoter elements required for developmental expression
of the maize Adh1 gene in transgenic rice. Plant Cell 6:
799–810
Li H-M, Culligan K, Dixon RA, Chory J (1995) CUE1: a mesophyll
cell-specific positive regulator of light-controlled gene expression in Arabidopsis. Plant Cell 7: 1599–1610
Ricard B, Rival J, Pradet A (1989) Rice alcohol cytosolic glyceraldehyde-3-phosphate dehydrogenase contains two subunits
differentially regulated by anaerobiosis. Plant Mol Biol 12:
131–139
Rivoal J, Thind S, Pradet A, Ricard B (1997) Differential induction
of pyruvate decarboxylase subunits and transcripts in anoxic
rice seedlings. Plant Physiol 114: 1021–1029
Russell DA, Sachs MM (1989) Differential expression and sequence analysis of the maize glyceraldehyde-3-phosphate dehydrogenase gene family. Plant Cell 1: 793–803
Sachs MM, Freeling M, Okimoto R (1980) The anaerobic proteins
of maize. Cell 20: 761–767
Sachs MM, Subbaiah CC, Sabb IN (1996) Anaerobic gene expression and flooding tolerance in maize. J Exp Bot 47: 1–15
Sedbrook JC, Kronebusch PJ, Borisy GG, Trewavas AJ, Masson
PH (1996) Transgenic aequorin reveals organ-specific cytosolic
Ca21 responses to anoxia in Arabidopsis thaliana seedlings. Plant
Physiol 111: 243–257
Somerville CR, Ogren WL (1982) Isolation of photorespiration mutants in Arabidopsis thaliana. In M Edelman, RB Hallick, N-H Chua, eds, Methods in Chloroplast Molecular Biol-
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
Arabidopsis Mutants Defective in Stress Responses
ogy. Elsevier Biomedical Press, Amsterdam, The Netherlands,
pp 129–138
Stockinger EJ, Gilmour SJ, Thomashow MF (1997) Arabidopsis
thaliana CBF1 encodes an AP2 domain containing transcriptional
factor that binds to the C-repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in response to low
temperature and water deficit. Proc Natl Acad Sci USA 94:
1035–1040
Subbaiah CC, Bush DS, Sachs MM (1994a) Elevation of cytosolic
calcium precedes anoxic gene expression in maize suspensioncultured cells. Plant Cell 6: 1747–1762
Subbaiah CC, Zhang J, Sachs MM (1994b) Involvement of
intracellular calcium in anaerobic gene expression and survival
of maize seedlings. Plant Physiol 105: 369–376
607
Valvenkens D, Van Montagu M, van Lijsebttens M (1988)
Agrobacterium tumefaciens-mediated transformation of Arabidopsis thaliana root explants by using kanamycin selection. Proc
Natl Acad Sci USA 85: 5536–5540
Xie Y, Wu R (1989) Rice alcohol dehydrogenase genes: anaerobic
induction, organ specific expression and characterization of
cDNA clones. Plant Mol Biol 13: 53–68
Yang Y, Kwon HB, Peng H-P, Shih M-C (1993) Stress responses and
metabolic regulation of glyceraldehyde-3-phosphate dehydrogenase genes in Arabidopsis. Plant Physiol 101: 209–216
Zhu J-K, Hasegawa PM, Bressan RA (1997) Molecular aspects of
osmotic stress in plants. Crit Rev Plant Sci 16: 253–277
Zhu J-K, Liu J, Xiong L (1998) Genetic analysis of salt tolerance in
Arabidopsis: evidence for a critical role of potassium nutrition.
Plant Cell 10: 1181–1192
Downloaded from on June 15, 2017 - Published by www.plantphysiol.org
Copyright © 1999 American Society of Plant Biologists. All rights reserved.
Plant Physiology, April 1999, Vol. 119, pp. 1569, www.plantphysiol.org © 1999 American Society of Plant Physiologists
Correction
Vol 119: 599–608, 1999
Conley, T.R., Peng, H-P., and Shih, M-C. Mutations affecting induction of glycolytic and
fermentative genes during germination and environmental stresses in Arabidopsis.
On page 599, the present address for Terry Conley is:
Department of Biology, Oklahoma City University, 2501 N. Blackwelder, Oklahoma City,
OK 73106.
1569