Enhanced Formaldehyde Detoxification by Overexpression
of Glutathione-Dependent Formaldehyde
Dehydrogenase from Arabidopsis1
Hakima Achkor, Maykelis Dı́az, M. Rosario Fernández, Josep Antoni Biosca, Xavier Parés, and
M. Carmen Martı́nez*
Departamento de Bioquı́mica y Biologı́a Molecular, Facultad de Ciencias, Universidad Autónoma de
Barcelona, 08193 Bellaterra, Barcelona, Spain
The ADH2 gene codes for the Arabidopsis glutathione-dependent formaldehyde dehydrogenase (FALDH), an enzyme
involved in formaldehyde metabolism in eukaryotes. In the present work, we have investigated the potential role of FALDH
in detoxification of exogenous formaldehyde. We have generated a yeast (Saccharomyces cerevisiae) mutant strain (sfa1⌬) by
in vivo deletion of the SFA1 gene that codes for the endogenous FALDH. Overexpression of Arabidopsis FALDH in this
mutant confers high resistance to formaldehyde added exogenously, which demonstrates the functional conservation of the
enzyme through evolution and supports its essential role in formaldehyde metabolism. To investigate the role of the enzyme
in plants, we have generated Arabidopsis transgenic lines with modified levels of FALDH. Plants overexpressing the
enzyme show a 25% increase in their efficiency to take up exogenous formaldehyde, whereas plants with reduced levels of
FALDH (due to either a cosuppression phenotype or to the expression of an antisense construct) show a marked slower rate
and reduced ability for formaldehyde detoxification as compared with the wild-type Arabidopsis. These results show that
the capacity to take up and detoxify high concentrations of formaldehyde is proportionally related to the FALDH activity
in the plant, revealing the essential role of this enzyme in formaldehyde detoxification.
Formaldehyde is a toxic compound produced during plant one-carbon (C1) metabolism. Most formaldehyde does not exist in vivo in a free state but is
bound to endogenous nucleophiles, such as glutathione or tetrahydrofolate (Sardi and Tyihak, 1994;
Chen et al., 1997). The main sources of formaldehyde
in plants are dissociation of 5,10-methylenetetrahydrofolate and oxidation of methanol, derived
mainly from pectin demethylation (Fall and Benson,
1996). Formaldehyde can also arise from other oxidative demethylation reactions (Sardi and Tyihak,
1994), from glyoxylate decarboxylation (Prather and
Sisler, 1972), and from P-450-dependent oxidation of
herbicides (Clejan and Cederbaum, 1993). Furthermore, formaldehyde might have an exogenous origin
from industrial waste, being a polluting agent of
residual waters and of air (Wolverton et al., 1989;
Giese et al., 1994). It is one of the main indoor air
pollutants, present in tobacco smoke, furniture, industrial adhesives, and varnishes. It has been classified as a mutagen and suspected carcinogen (Wippermann et al., 1999) and as causing headaches and
1
This work was supported by the Dirección General de
Enseñanza Superior (grant nos. PB96 –1167 and PB98 – 0855), by the
Comissionat per a Universitats i Recerca (grant no. 1999SGR
00103), and by the Commission of the European Union (grant no.
BIO4 –CT97–2123).
* Corresponding author; e-mail [email protected]; fax
34 –93–5811264.
Article, publication date, and citation information can be found
at www.plantphysiol.org/cgi/doi/10.1104/pp.103.022277.
2248
irritation of the mucosa. Formaldehyde is considered
a toxic compound because of its ability to react with
proteins, nucleic acids, and lipids. The World Health
Organization has established an air quality guideline
of 100 ␮g m⫺3 for formaldehyde (European Commission, 1994).
Several studies have demonstrated that exogenous
formaldehyde can be incorporated into the metabolism of photosynthetic cells and be used as a carbon
source. For example, feeding the common spider
plant (Chlorophytum comosun) with 14C-formaldehyde
resulted in 14C-labeled products derived from the C1
metabolism, such as Ser and phosphatidylcholine
(Giese et al., 1994). Removal of formaldehyde can
occur by different in vivo pathways, through the
formation of adducts with glutathione, Arg, Asn, and
tetrahydrofolate (for a review on C1 metabolism in
higher plants, see Hanson and Roje, 2001). However,
biochemical and genetic studies in several eukaryotes
indicate that the main enzyme responsible for the
metabolism of formaldehyde is the glutathionedependent formaldehyde dehydrogenase (FALDH).
This enzyme is universally present in animal and
plant tissues and catalyzes the following reaction:
S-hydroxymethylglutathione ⫹ NAD
FALDH
|
0
S-formylglutathione ⫹ NADH ⫹ H⫹
(1)
Downloaded
from132,
on June
16, 2017 - Published
by www.plantphysiol.org
Plant Physiology, August
2003, Vol.
pp. 2248–2255,
www.plantphysiol.org
© 2003 American Society of Plant Biologists
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
Arabidopsis Formaldehyde Dehydrogenase
where S-hydroxymethylglutathione is a nonenzymatically formed adduct of glutathione and
formaldehyde.
S-formylglutathione is then hydrolyzed to formate
and glutathione by S-formylglutathione hydrolase:
S-formylglutathione ⫹ Water
Hydrolase
|
0
Formate ⫹ Glutathione
RESULTS
Construction of a Yeast sfa1⌬ Strain
(2)
Formate can give rise to C1 folates, but a quantitatively more important fate is oxidation to CO2 mediated by formate dehydrogenase (Cossins, 1964).
NAD-dependent formate dehydrogenase has been
cloned from Arabidopsis, which is predicted to have
a mitochondrial and a chloroplastic localization (Olson et al., 2000). As is the case in the photosynthetic
bacteria Rhodobacter sphaeroides (Barber and Donohue,
1998), it might be that in plants the Calvin cycle
assimilates in part the products of FALDH activity.
In addition to formaldehyde, FALDH can also oxidize long-chain alcohols and ␻-hydroxy fatty acids,
such as octanol and 12-hydroxydodecanoic acid. Furthermore, it has been demonstrated recently that it is
very active in the reduction of S-nitrosoglutathione,
the condensation product of glutathione and NO
(Jensen et al., 1998; Liu et al., 2001; Sakamoto et al.,
2002).
FALDH from Arabidopsis has been characterized,
and its cDNA has been cloned (Martı́nez et al., 1996).
This enzyme exhibits features similar to those of their
animal counterparts and shows a structure highly
conserved in plants and animals. FALDH is expressed in low amounts in Arabidopsis (5 ⫻ 10⫺3
units mg⫺1; Martı́nez et al., 1996), which is approximately the same amount that has been observed in
dry pea (Pisum sativum) seeds (4 ⫻ 10⫺3 units mg⫺1;
Shafqat et al., 1996) and about twice the amount
present in the spider plant (Giese et al., 1994). The
gene is expressed at similar levels in all plant organs,
as assessed by northern-blot analysis (Martı́nez et al.,
1996), in analogy with the ubiquitous distribution of
the corresponding enzyme in animal tissues (Smith,
1988). FALDH is encoded in the Arabidopsis genome
by a single gene, which maps in chromosome 5 (Dolferus et al., 1997).
In the present work, we have investigated the potential role of FALDH in detoxification of exogenous
formaldehyde. We have generated a yeast (Saccharomyces cerevisiae) mutant strain (sfa1⌬) by in vivo deletion of the SFA1 gene that codes for the endogenous
FALDH. We have investigated the effect of the overexpression of Arabidopsis FALDH in this mutant,
regarding its capacity to detoxify exogenous formaldehyde. To investigate the role of the enzyme in
plants, we have generated transgenic Arabidopsis
lines transformed with either sense or antisense conPlant Physiol. Vol. 132, 2003
structs of FALDH gene, and we have measured the
rates for formaldehyde uptake and detoxification by
both transgenic plants overexpressing FALDH and
plants with reduced FALDH levels. Our results confirm the central role of FALDH in formaldehyde
metabolism in plants.
To explore the physiological role of the Arabidopsis FALDH, we made complementation experiments
by transforming a yeast strain lacking FALDH activity. This strain was constructed by in vivo deletion of
the SFA1 gene, coding for the yeast glutathionedependent FALDH (Wehner et al., 1993). A linear
fragment containing the HIS3 gene flanked by the
flanking sequences of SFA1 gene was used to transform a haploid strain (His⫺), derived from the diploid W303D, to prototrophy (His⫹). Isoelectrofocusing analysis stained by activity confirmed that none
of the His⫹ transformants contained glutathionedependent FALDH activity (Fig. 1A, lane 3). The
sfa1::HIS3 replacement was also confirmed by Southern analysis, using as probe a PstI/NcoI fragment
containing the SFA1 gene (Fig. 1, B and C).
Figure 1. Analysis of the yeast sfa1⌬ strain (sfa1::HIS3). A, Isoelectric
focusing and FALDH activity staining on homogenates of wild-type
W303D (lane 1) and sfa1::HIS3 strain transformed with (lane 2) and
without (lane 3) the ADH2-pYes2 plasmid. B, Southern-blot analysis
of genomic DNA digested with PstI and EcoRI from sfa1::HIS3 (lane
1) or W303D (lane 2) strain. The HIS3 gene contains an internal PstI
restriction site that is not present in the SFA1 gene. C, Schema of the
homologous recombination procedure used to construct the sfa1⌬
yeast strain. The PstI/NcoI fragment containing the SFA1 gene (at the
bottom) was used as probe for Southern analysis.
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
2249
Achkor et al.
specific activity of the purified enzyme against
S-hydroxymethylglutathione was found to be 15
units mg⫺1. Fold purification and recovery of the
process are shown in Table I.
The purified enzyme was inoculated into rabbits to
raise polyclonal antibodies. The antibodies recognized a single band of the expected size (45 kD) in
Arabidopsis crude protein extracts that was not detected using the pre-immune sera (Fig. 2B, lanes 2
and 3, respectively). Those immunoreactions were
specific because the binding of the antibodies could
be competed with the protein used for immunization
(Fig. 2B, lanes 4–9).
Figure 2. Electrophoretic analysis of purified Arabidopsis FALDH
and specificity of the anti-FALDH antibodies. A, Purified FALDH (1.4
␮g) was analyzed by SDS-PAGE and revealed by silver staining (lane
1). B, Specificity of the antibodies. Twenty micrograms of Arabidopsis protein extracts was electrophoresed, blotted to a membrane,
stained with Coomassie Blue (lane 1), and developed with the antiFALDH antibodies (lane 2) or the pre-immune antiserum (lane 3). C,
Competition experiments for the antibodies binding. Fifty nanograms
of purified FALDH (lanes 1, 3, and 5) or 20 ␮g of Arabidopsis protein
extracts (lanes 2, 4, and 6) was electrophoresed, and the blotted
membranes were incubated either with the antibodies (lanes 1 and
2), the pre-immune antiserum (lanes 3 and 4), or the antibodies
pre-incubated with 30 ␮g of the protein used for immunization (lanes
5 and 6).
Overexpression of Arabidopsis FALDH in Yeast.
Purification of the Enzyme and Generation of
Specific Antibodies
Arabidopsis FALDH cDNA was cloned into the
yeast pYes2 expression vector, under the control of
the GAL1-GAL10 promoter (ADH2-pYes2 plasmid),
and introduced into the sfa1::HIS3 strain. The transformed strain was grown in the presence of Gal as
carbon source and using His and uracil as markers.
The selected colonies exhibited FALDH activity in
the crude extracts, confirmed by the presence of a
band of the expected size in electrofocusing gels
stained by activity (Fig. 1A, lane 2), indicating both
that it was correctly expressed and that it was functional in this heterologous system. To purify the
enzyme from this source, we used three consecutive
chromatographic steps. From 500 g of yeast cell
pellet, 2.24 mg of purified protein was obtained that
appeared as a single band of Mr 45,000 on SDSPAGE (Fig. 2A). This band was recognized by an
anti-rat FALDH antiserum (result not shown). The
Enzymatic Properties of Arabidopsis FALDH
The Arabidopsis enzyme expressed in yeast
showed a Km value of 7 ␮m. The deduced turnover
number (Kcat) was 1,351 min⫺1, calculated for a Mr ⫽
90,000 (dimer), and the catalytic efficiency (Kcat/Km)
was 193,000 mm⫺1 min⫺1. The enzyme exhibited high
activity at pH 10 toward several plant alcohols, such
as farnesol and geraniol. However, at more physiological pH (pH 7.5), only farnesol showed a significant activity, with a Kcat/Km of 780 mm⫺1 min⫺1.
Other medium-chain alcohols and ␻-hydroxy fatty
acids were also substrates of the enzyme, though
their activity was much lower (Table II).
Overexpression of Arabidopsis FALDH in Yeast
Greatly Improves Its Ability to Grow in the
Presence of Formaldehyde
The yeast sfa1::HIS3 strain could grow at the same
rate as the wild type in rich medium but could not
grow in the presence of 0.6 mm formaldehyde. In
contrast, the wild-type yeast strain could grow in the
presence of 0.6 mm formaldehyde, though after a
long lag phase (data not shown). These results confirm the essential role of FALDH for yeast formaldehyde metabolism. The sfa1::HIS3 strain transformed
with the ADH2-pYes2 plasmid was then grown both
in the absence and in the presence of different concentrations of formaldehyde added exogenously. Figure 3A shows that both the wild type and the transformed strains grew equally well in the absence of
formaldehyde, reaching saturation at around 35 h
after subculturing. In contrast, at 1 mm formalde-
Table I. Purification of Arabidopsis recombinant FALDH
Activity was measured in 0.1 M sodium phosphate (pH 8) at 25°C using 1 mM formaldehyde, 1 mM
glutathione, and 4 mM NAD.
Step
Homogenate
DEAE-Sepharose
Hydroxyapatite
Blue-Sepharose
2250
Activity
Protein
Specific Activity
Purification
Yield
units
mg
units mg⫺1
-fold
%
300
240
120
62
510
60.3
27.3
4.1
0.58
3.98
4.4
15
1
6.86
7.6
25.8
100
80
40
20.6
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
Plant Physiol. Vol. 132, 2003
Arabidopsis Formaldehyde Dehydrogenase
Table II. Kinetic properties of Arabidopsis FALDH
Activity versus S-hydroxymethylglutathione was measured in 0.1 M sodium phosphate (pH 8).
Activity with other substrates was measured in 0.1 M sodium phosphate (pH 7.5) or in 0.1 M glycine
buffer (pH 10). In all cases, 2.4 mM NAD was used.
Substrate
S-hydroxymethylglutathione (pH 8)
Farnesol (pH 7.5)
Geraniol (pH 7.5)
Cinnamylalcohol (pH 7.5)
12-Hydroxydodecanoic acid (pH 7.5)
Farnesol (pH 10)
Geraniol (pH 10)
Cinnamylalcohol (pH 10)
Octanol (pH 10)
12-Hydroxydodecanoic acid (pH 10)
a
Km
Kcat
Kcat/Km
␮M
min⫺1
mM⫺1 min⫺1
7
7.7
n.s.a
22,000
n.s.a
3
800
3,500
n.s.a
4,700
1,351
6
–
324
–
126
1,200
1,220
–
335
193,000
780
40
15
20
42,000
1,500
350
130
70
n.s., Non-saturation reached with this substrate.
hyde, yeast wild type was unable to grow during the
time of the experiment (60 h), whereas the transformed strain showed the same growth rate as in the
absence of formaldehyde. At 2 mm formaldehyde,
the transformed strain still grew very efficiently.
We also determined the kinetics of formaldehyde
elimination from the culture medium (Fig. 3B). Our
results show that exogenous formaldehyde disappeared very rapidly in the case of the transformed
yeast strain and was completely metabolized in 50 to
60 h, which is consistent with the cell’s growth rate
shown in Figure 3A. However, in the case of the
wild-type strain, the concentration of formaldehyde
in the culture medium remained constant for long
periods of time, which proves its inability to metabolize it efficiently and, as a consequence, to grow in
the presence of this toxic compound.
To test the ability of the transgenic plants to metabolize exogenous formaldehyde, we selected three
of the homozygous lines (T3 generation) showing the
highest expression of the transgene (lines 1, 4, and 6).
Equal amounts of seeds (3 mg) were germinated and
grown in liquid medium (5 mL), and formaldehyde
at the desired concentration was added to 6-d-old
plantlets. Subsequently, aliquots were taken to measure the variation of formaldehyde concentration
Detoxification of Exogenous Formaldehyde by
Transgenic Arabidopsis Plants
We addressed the question of whether plants overexpressing FALDH might be able to cope better with
moderately high concentrations of environmental
formaldehyde. We generated transgenic Arabidopsis
lines transformed with FALDH cDNA under the control of the 35S cauliflower mosaic virus promoter. A
total of 40 independently transformed plants were
obtained (T1 progeny) after infiltration of Columbia
plants. The T1 plants were self-fertilized, and 16 individuals of the T2 progeny showing a Mendelian
segregation of kanamycin resistance were brought to
homozygosis (T3 generation). Western-blot analysis
of seven individuals from the T3 generation is shown
in Figure 4A. One of them is a cosuppression line
(line 13), whereas the others show from a moderate to
a high increase of FALDH protein levels. In the same
figure, it can be seen that the increase in FALDH
levels correlated well with increased values of
FALDH enzymatic activity.
Plant Physiol. Vol. 132, 2003
Figure 3. Growth kinetics and formaldehyde detoxification by yeast
strains expressing Arabidopsis FALDH. Growth (A) and formaldehyde
concentration (B) were measured in culture medium containing no
added formaldehyde (circles) or 1 (triangles) or 2 (squares) mM
formaldehyde. White symbols, Yeast strain overexpressing the Arabidopsis FALDH; black symbols, W303D yeast strain.
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
2251
Achkor et al.
structs (lines 5a, 10a, and 17a, Fig. 4B) were incubated
with 2 mm formaldehyde added exogenously. Aliquots from the culture medium were removed during a time course to measure the formaldehyde concentration. The results in Figure 5A demonstrate that
the lines with reduced levels of FALDH show a 20%
decrease in their ability to detoxify exogenous
formaldehyde.
DISCUSSION
Figure 4. Analysis of Arabidopsis transgenic lines. Western-blot
analysis and FALDH-specific activity (SA) values of several independent transgenic lines overexpressing Arabidopsis FALDH (A) or bearing FALDH antisense constructs (B). Ten-day-old homozygous seedlings (T3 progeny) grown in liquid medium were used. Schematic
diagrams of the constructs used for transformation are also shown.
n.d., Not determined. CaMV35S, 35S cauliflower mosaic virus promoter. p-NOS, Nopaline synthase promoter. NPTII, neomycin phosphotransferase gene-coding sequence.
during a time course. A control of liquid medium
with formaldehyde but without plants was also performed to discard losses of formaldehyde by evaporation. At least three independent experiments were
carried out with each line.
At 2 mm formaldehyde, we observed significant
differences in the rate for formaldehyde uptake (Fig.
5A). The three transgenic lines tested were able to
completely detoxify the formaldehyde in 48 h,
whereas Arabidopsis wild-type metabolized it at a
slower rate (80% of the total in 48 h). In view of the
fact that the plants could cope well with this concentration of exogenous formaldehyde (no visible defects were observed as a consequence of the addition
of this compound), we decided to perform the same
experiment using a concentration of 5 mm formaldehyde. As can be seen in Figure 5B, more significant
differences were observed in this case. Line 1, showing the highest rate for formaldehyde uptake, could
achieve a 50% decrease in formaldehyde concentration in the medium after 48 h (in contraposition to the
25% by the Arabidopsis wild type). This higher rate
was evidenced very early (at 6 h, there was already a
13% differential rate), and subsequently increased (at
24 and 48 h, there was a 25% differential rate).
To corroborate the above data that suggest a correlation between the amount of FALDH expressed by
the plant and its ability to detoxify exogenous formaldehyde, we performed the same experiments with
transgenic lines showing reduced levels of FALDH.
One line showing a cosuppression phenotype (line
13, Fig. 4A) and three lines bearing antisense con2252
Plant FALDHs have been isolated from pea seeds
(Uotila and Koivusalo, 1979; Shafqat et al., 1996),
Arabidopsis (Martı́nez et al., 1996), and maize (Zea
mays; Wippermann et al., 1999). Their molecular and
kinetic properties are remarkably similar to those
found in the homolog enzymes from mammals, invertebrates, and microorganisms, providing evidence
of the high degree of structural and functional conservation through evolution. The enzyme shows very
low expression levels in Arabidopsis plants (Martı́nez et al., 1996); thus, a system to produce a recombinant enzyme was desirable. We have used a yeast
expression system to purify the enzyme to homogeneity and to study its functional significance in the
detoxification of exogenous formaldehyde.
Arabidopsis FALDH, purified as a recombinant
enzyme from yeast, has a specific activity of 15 units
mg⫺1 and a Km value of 7 ␮m. The polypeptide chain
Figure 5. Kinetics of formaldehyde detoxification by Arabidopsis
transgenic lines. Seedlings germinated and grown in liquid medium
were subjected to formaldehyde treatment at an initial concentration
of 2 (A) or 5 (B) mM, respectively. Transgenic lines are those shown
in Figure 4. Lines 1 (E), 4 (ƒ), and 6 () express high levels of
FALDH. Line 13 (〫) is a cosuppression line. Lines 5a (f), 10a (䡺),
and 17a (⽧) are antisense lines. F, Wild-type Arabidopsis. Data are
shown as percentage formaldehyde remaining in the liquid culture at
different times after inoculation. Values are expressed as the mean of
at least two separate experiments. SEs of means were ⬍12%.
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
Plant Physiol. Vol. 132, 2003
Arabidopsis Formaldehyde Dehydrogenase
of the enzyme has a molecular mass of 45 kD, calculated by SDS-PAGE, which is consistent with both
the molecular mass deduced from the cDNA sequence and with that of the enzyme purified from
Arabidopsis plants (Martı́nez et al., 1996). The
FALDH purified from a variety of different sources
consists of two identical subunits and, thus, is a
homodimer (Uotila and Koivusalo, 1989). Based on a
dimeric structure for Arabidopsis FALDH, we calculated a Kcat value of 1,351 min⫺1 and a Kcat/Km of
193,000 mm⫺1 min⫺1. This last value, which is the
catalytic efficiency, is comparable with those of pea
and other eukaryotes (Shafqat et al., 1996; Fernández
et al., 1997).
SFA1 deletion in yeast is not lethal but impairs its
growing in the presence of formaldehyde (Fernández
et al., 1999; this work). We have demonstrated in this
work that Arabidopsis FALDH is able to complement
a knockout yeast mutant that has the SFA1 gene
replaced by the HIS3 marker gene. Our results show
that overexpression of Arabidopsis FALDH in this
mutant results in high resistance to formaldehyde,
demonstrating the functionality of the Arabidopsis
enzyme and its essential role in formaldehyde metabolism. This effect was quite remarkable because at
1 mm formaldehyde, yeast wild type was unable to
grow, whereas the strain overexpressing Arabidopsis
FALDH grew at the same rate as in the absence of
formaldehyde. At 2 mm formaldehyde, the growing
rate of the transformed strain still was only slightly
affected.
Additional support for the important role of
FALDH in detoxifying formaldehyde was obtained
by the generation of transgenic Arabidopsis lines.
Three independent lines, exhibiting from 11- to 18fold the wild-type FALDH activity, were used to test
the capacity of plants overexpressing FALDH to take
up and detoxify exogenous formaldehyde. We observed an increase in the detoxification rate of 25%,
both at 2 and 5 mm formaldehyde (64 and 132.5 mL
L⫺1, respectively). At 2 mm formaldehyde, transgenic
plants could completely detoxify the exogenous
formaldehyde in 48 h or less, without any visible
damage. These results were corroborated by generating transgenic lines with decreased levels of
FALDH enzyme (cosuppression and antisense lines).
At 2 mm formaldehyde, these plants showed a 20%
decrease in their ability to detoxify formaldehyde, as
compared with the wild-type plants.
At 5 mm formaldehyde, which is an extremely toxic
concentration, the transgenic lines showed a good
rate for formaldehyde uptake during the first 24 h.
From this time on, we observed important phytotoxic
effects, the leaves starting to display chlorotic
patches that finally spread to the whole tissue. At the
same time, the rate decreased significantly, and the
plants could not achieve a complete detoxification of
the exogenous formaldehyde. This phytotoxic effect
of formaldehyde at such high concentration is not
Plant Physiol. Vol. 132, 2003
surprising and might be due to the formaldehyde
itself or to the products of its oxidation. The chlorotic
phenotype of the plants and the observation of leaf
sections under electron microscopy (data not shown)
suggest that chloroplasts are seriously damaged after
long expositions to 5 mm formaldehyde. This might
be due to a process of photooxidation caused by an
accumulation of FALDH oxidation products that are
toxic and have effects on the glutathione/redox
homeostasis.
It has been reported that foliar application of methanol can stimulate plant growth (Nonomura and
Benson, 1992) and that methanol is quickly oxidized
in most plant tissues successively to formaldehyde,
formic acid, and CO2 (Cossins, 1964). Formaldehyde
can be incorporated to plant C1 metabolism mainly
by two metabolic pathways that involve folatemediated and -independent reactions (Hanson and
Roje, 2001). In the pathway involving folatemediated reactions, formaldehyde can be converted
to 5,10-methylene-tetrahydrofolate by a nonenzymatic reaction with tetrahydrofolate or to formate by
the successive reactions catalyzed by FALDH and
S-formylglutathione hydrolase. The formate is then
incorporated into the C1-tetrahydrofolate pool via the
10-formyl-tetrahydrofolate. In the folate-independent
pathway, formaldehyde is successively converted to
formic acid and CO2, and CO2 is then assimilated by
the Calvin-Benson cycle in the photosynthetic tissues
(Cossins, 1964). Formate might also give rise to
glyoxylate, in a reaction catalyzed by glyoxylate synthetase that is exclusively found in plants (Janave et
al., 1993). Our results showing that the capacity to
detoxify exogenous formaldehyde is proportionally
related to the level of FALDH activity in the plants
strongly suggest that the major metabolic flux of formaldehyde is via formate. However, our data do not
allow the elucidation of the final fate of this formate.
S-formylglutathione hydrolase and formate dehydrogenase have been cloned from Arabidopsis (Olson et
al., 2000; Kordic et al., 2002), demonstrating that all the
enzymes necessary for this metabolic pathway are
present.
In summary, our results show that overexpression
of the enzyme FALDH in plants confer a high resistance to formaldehyde, thus supporting the “green
liver” concept of plant xenobiotic metabolism (Sandermann, 1992). The availability of transgenic plants
with modified amounts of FALDH will allow us to
investigate the mechanisms of formaldehyde toxicity
at the cellular level in the future. This knowledge is
necessary to improve the detoxification properties of
the plants and to better understand C1 metabolism in
higher plants.
MATERIALS AND METHODS
Biological Material
Yeast (Saccharomyces cerevisiae) W303D strain was grown at 30°C in
Wickerham’s medium (United States Biochemical Corp., Cleveland). Arabi-
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
2253
Achkor et al.
dopsis ecotype Columbia was grown in soil under a 16-h-light/8-h-dark
regime at 22°C. For the formaldehyde treatments, plant seeds were surface
sterilized and sown in six-well tissue culture clusters (Costar Corp., Cambridge, MA), containing 5 mL well⫺1 of sterile Murashige and Skoog media
(Duchefa, Harlem, The Netherlands) supplemented with 0.5% (w/v) Suc
and grown with shaking (150 rpm) under the same light regime as above.
Gene Deletion
A null mutant of the SFA1 gene, coding for the yeast FALDH, was
constructed by one-step gene replacement (Rothstein, 1983). In brief, an
internal fragment containing the SFA1 coding region was removed (from a
pBR322-derived construct containing the SFA1 gene) and replaced by the
yeast HIS3 gene (Berben et al., 1991). A linear fragment, including the
selectable HIS3 gene flanked by sequences homolog to the flanking sequences of the SFA1 coding region, was obtained from the previous construct and used to transform a haploid yeast strain derived from W303.
Screening of transformants was performed in Wickerham’s medium without His, and the SFA1 deletion was confirmed by FALDH activity and
Southern analysis.
Expression in Yeast
ADH2 cDNA was amplified using two primers based on the reported
sequence (Martı́nez et al., 1996). The upstream primer included the ATG
initiation codon and had the following sequence: 5⬘-GCCGGTACCATGGCGATCAAGGTAGG-3⬘. The downstream primer included the termination codon and had the following sequence: 5⬘-CGCCGTCGACTCATTTGCGGTATCGAGG-3⬘. PCR was carried out in two steps: (a) 93°C for
1 min, 54.5°C for 2 min, and 72°C for 2 min for five cycles, and (b) 93°C for
1 min, 61.5°C for 2 min, and 72°C for 2 min for 25 cycles, followed by an
elongation step of 72°C for 10 min. One band of the expected size was
obtained, purified, and cloned into the SmaI site of pUC18 vector. The insert
was completely sequenced by the Sanger method and then subcloned into
the KpnI site of the yeast pYes2 expression vector. The recombinant plasmid
(ADH2-pYes2) was introduced into the Escherichia coli MC1061 strain to
verify insert orientation and into the haploid sfa1::HIS3 yeast strain for
expression. Screening of transformants was made by growing in the absence
of both His and uracil. Transformed yeast strain was grown at 30°C in
Wickerham’s medium without His and uracil with 2% (w/v) Gal as carbon
source to induce the GAL1-GAL10 promoter and a starting density of 3 ⫻ 106
cells mL⫺1.
Purification of Arabidopsis Recombinant FALDH
Yeast sfa1::HIS3 strain transformed with the ADH2-pYes2 plasmid (50 L)
was grown until the early stationary phase, and cell pellets were collected
by centrifugation and resuspended in buffer A, containing 10 mm Tris/HCl
(pH 7.5), 0.5 mm dithiothreitol (DTT), 1 mm phenylmethylsulfonyl fluoride
(PMSF), and 1 mm benzamidine (1:1 [w/v]). Cells were lysed with a beadbeater homogenizer (Biospec Products, Inc., Bartlesville, OK), using 0.5-mm
diameter glass beads. The homogenate was centrifuged at 29,000g for 1 h,
filtered through glass wool, and loaded onto a DEAE-Sepharose CL-6B
column (36.5 ⫻ 4 cm) equilibrated with the buffer A. A 1,200-mL linear
gradient of 0 to 0.25 m NaCl was used to eluate the bound enzyme. The
active material was concentrated to 60 mL with Diaflo PM-10 membrane
(Amicon, Billerica, MA), dialyzed against 10 mm KH2PO4 (pH 6.8), 0.5 mm
DTT, 1 mm PMSF, and 1 mm benzamidine, and loaded onto a hydroxiapatite
Bio-Gel HT (Bio-Rad Laboratories, Hercules, CA) column (16 ⫻ 2 cm)
equilibrated with the same buffer. A 1,200-mL linear gradient of 10 to 400
mm KH2PO4 was applied, and the active fractions were collected, concentrated to 25 mL, dialyzed against buffer A, and loaded onto a BlueSepharose column (15.5 ⫻ 1.8 cm) equilibrated with the same buffer. The
active material was eluted with a 600-mL linear gradient of 0 to 0.75 mm
NADH, concentrated to 16 mL, and the excess of NADH was eliminated
with a PD-10 gel filtration column (Amersham-Pharmacia Biotech, Uppsala)
equilibrated with 10 mm Tris/HCl (pH 7.5) and 0.5 mm DTT. The active
material was equilibrated with 20 mm Tris/acetate buffer (pH 6.6), 0.5 mm
DTT, 1 mm PMSF, and 1 mm benzamidine, using a PD-10 gel filtration
column, and loaded onto a Protein Pack Q 8HR column (Millipore, Billerica,
MA) equilibrated with the same buffer. Bound proteins were eluted with a
2254
40-mL linear gradient of 0 to 0.2 m sodium acetate. Protein concentrations
were determined by the Bradford assay (Bio-Rad), using bovine serum
albumin as a standard.
Construction of Transgenic Arabidopsis Plants
To prepare the sense construct, an EcoRI/XhoI restriction fragment containing the Arabidopsis FALDH coding sequence (Martı́nez et al., 1996) was
purified and subcloned downstream of the cauliflower mosaic virus 35S
promoter into the SmaI site of pDH51 (Pietrzak et al., 1986). To prepare the
antisense construct, a PstI/BglII fragment (425 bp) was purified and subcloned into pDH51 cut with BamHI and PstI. After verifying the correct
orientation of the insert, the resulting 35S::ADH2 expression cassettes were
excised as EcoRI restriction fragments and subcloned into the EcoRI site of
the Agrobacterium tumefaciens binary plasmid pBin19 (Bevan, 1984). The
recombinant plasmids were then introduced into A. tumefaciens C58CI strain
by the freeze-thaw method. Arabidopsis plants were transformed by the
vacuum infiltration method (Bechtold et al., 1993). Transgenic plants were
selected on Murashige and Skoog solid media containing 40 mg L⫺1 kanamycin and later transferred to soil for seed production. Transformants with
pBin19 empty vector were also produced to be used as controls. Segregation
tests were performed with the T2 seeds obtained by self-fertilization of
independent primary transformants. Homozygous, monogenic T3 seeds
were obtained by self-fertilization of T2 plants using kanamycin resistance
as the marker.
Electrophoretic Analysis and
Immunochemical Procedures
SDS-PAGE was performed as described (Laemmli, 1970), with subsequent Coomassie Blue or silver staining (Merril et al., 1981). Isoelectric
focusing was performed on polyacrylamide gels in a pH 3 to 9 gradient
(PhastSystem, Amersham-Pharmacia Biotech), and bands were visualized
by staining for glutathione-dependent FALDH activity with Nitro Blue
tetrazolium/phenazine methosulphate, 4.8 mm formaldehyde, and 1 mm
glutathione. For production of the antibodies, Arabidopsis FALDH purified
as a recombinant enzyme from yeast was used to inoculate rabbits by
conventional methods (Harlow and Lane, 1988). For western blotting, proteins were electrophoresed on 12.5% (w/v) SDS-PAGE gels, transferred to
immobilon-P membranes (Millipore), and incubated with the antiserum
(dilution 1:3,000 [v/v]). The immunocomplexes were revealed using the
Immun-Star detection kit system (Bio-Rad). Loading of equal amounts of
proteins was controlled by Bradford analysis and by gel and membrane
staining. Competition experiments to prove the specificity of the antibodies
were performed by pre-incubation of the antibodies with 30 ␮g of the same
protein used for rabbit inoculation for 1 h at room temperature.
Enzymatic Activity
Enzyme activity was determined spectrophotometrically at 25°C by monitoring the production of NADH at 340 nm (⑀340 ⫽ 6.22 mm⫺1 cm⫺1) with a
Hewlett-Packard 8452A diode array spectrophotometer (Hewlett-Packard,
Palo Alto, CA). In some experiments, 0.6 mm 3-acetylpyridine adenine
dinucleotide (⑀363 ⫽ 9.10 mm⫺1 cm⫺1; Sigma, St. Louis) was used as a
cofactor instead of NAD. One unit of activity corresponds to 1 ␮mol reduced
cofactor formed per minute. FALDH activity was measured in 0.1 m sodium
phosphate (pH 8) as described by Martı́nez et al. (1996). Kinetic constants for
FALDH were determined with S-hydroxymethylglutathione. Concentrations of S-hydroxymethylglutathione were calculated using its dissociation
constant (Koivusalo et al., 1989) and keeping free glutathione at a constant
concentration of 1 mm, changing accordingly the formaldehyde concentration. Kinetic constants for alcohols were determined with 2.4 mm NAD, in
0.1 m sodium phosphate at pH 7.5, or in 0.1 m Gly buffer at pH 10. Seven
substrate concentrations, in duplicate, were used for each kinetic measurement. Kinetic results were analyzed using the nonlinear regression data
analysis program Enzfitter (Elsevier, Biosoft, Cambridge, UK), and the
constants were expressed as the mean of two independent determinations.
Formaldehyde concentration was determined by the conversion of NAD to
NADH in the presence of FALDH from Pseudomonas putida with an excess of
NAD.
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
Plant Physiol. Vol. 132, 2003
Arabidopsis Formaldehyde Dehydrogenase
Distribution of Materials
Upon request, all novel materials described in this publication will be
made available in a timely manner for noncommercial research purposes,
subject to the requisite permission from any third party owners of all or
parts of the material. Obtaining any permission will be the responsibility of
the requestor.
Received February 22, 2003; returned for revision March 24, 2003; accepted
April 28, 2003.
LITERATURE CITED
Barber RD, Donohue TJ (1998) Function of a glutathione-dependent formaldehyde dehydrogenase in Rhodobacter sphaeroides formaldehyde oxidation and assimilation. Biochemistry 37: 530–537
Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium mediated gene
transfer by infiltration of adult Arabidopsis thaliana plants. C R Acad Sci
Paris 316: 1194–1199
Berben G, Dumont J, Gilliquet V, Bolle PA, Hilger F (1991) The YDp
plasmids: a uniform set of vectors bearing versatile gene deletion cassettes for Saccharomyces cerevisiae. Yeast 7: 475–477
Bevan M (1984) Binary Agrobacterium vectors for plant transformation.
Nucleic Acids Res 12: 8711–8721
Clejan LA, Cederbaum AI (1993) Stimulation by paraquat of microsomal
and cytochrome P-450-dependent oxidation of glycerol to formaldehyde.
Biochem J 295: 781–786
Chen L, Chan SY, Cossins EA (1997) Distribution of folate derivatives and
enzymes for synthesis of 10-formyltetrahydrofolate in cytosolic and mitochondrial fractions of pea leaves. Plant Physiol 115: 299–309
Cossins EA (1964) The utilization of carbon-1 compounds by plants. I. The
metabolism of methanol-C14 and its role in amino acid biosynthesis. Can
J Biochem 42: 1793–1802
Dolferus R, Osterman JC, Peacock WJ, Dennis ES (1997) Cloning of the
Arabidopsis and rice formaldehyde dehydrogenase genes: implications
of the origin of plant ADH enzymes. Genetics 146: 1131–1141
European Commission (1994) Environmental chemicals. In G Rossi, G
Realini, eds, Annual Report, Environment Institute. Joint Research Centre, Ispra, Italy, pp 53–59
Fall R, Benson AA (1996) Leaf methanol: the simplest natural product from
plants. Trends Plant Sci 1: 296–301
Fernández MR, Biosca JA, Martı́nez MC, Achkor H, Farrés J, Parés X (1997)
Formaldehyde dehydrogenase from yeast and plant: implications for the
general functional and structural significance of class III alcohol dehydrogenase. In H Weiner, ed, Advances in Experimental Medicine and
Biology, Vol 414. Plenum Press, New York, pp 373–381
Fernández MR, Biosca JA, Torres D, Crosas B, Parés X (1999) A double
residue substitution in the coenzyme binding site accounts for the different kinetic properties between yeast and human formaldehyde dehydrogenases. J Biol Chem 274: 37869–37875
Giese M, Bauer-Doranth U, Langebartels C, Sandermann H Jr (1994)
Detoxification of formaldehyde by the spider plant (Chlorophytum comosum L.) and by soybean (Glycine max L.) cell-suspension cultures. Plant
Physiol 104: 1301–1309
Hanson AD, Roje S (2001) One-carbon metabolism in higher plants. Annu
Rev Plant Physiol Plant Mol Biol 52: 119–137
Harlow E, Lane D (1988) Antibodies, a Laboratory Manual. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY
Janave MT, Ramaswami NK, Nair PM (1993) Purification and characterization of glyoxylate synthetase from greening potato-tuber chloroplasts.
Eur J Biochem 214: 889–896
Jensen DE, Belka GK, Du Bois GC (1998) S-nitrosoglutathione is a substrate for rat alcohol dehydrogenase class III isoenzyme. Biochem J 331:
659–668
Koivusalo M, Baumann M, Uotila L (1989) Evidence for the identity of
glutathione-dependent formaldehyde dehydrogenase and class III alcohol dehydrogenase. FEBS Lett 257: 105–109
Plant Physiol. Vol. 132, 2003
Kordic S, Cummins I, Edwards R (2002) Cloning and characterization of an
s-formylglutathione hydrolase from Arabidopsis thaliana. Arch Biochem
Biophys 399: 232–238
Laemmli UK (1970) Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227: 680–685
Liu L, Hausladen A, Zeng M, Que L, Heitman J, Stamler JS (2001) A
metabolic enzyme for S-nitrosothiol conserved from bacteria to humans.
Nature 410: 490–494
Martı́nez MC, Achkor H, Persson B, Fernández MR, Shafqat J, Farrés J,
Jörnvall H, Parés X (1996) Arabidopsis formaldehyde dehydrogenase:
molecular properties of plant class III alcohol dehydrogenase provide
further insight into the origins, structure and function of plant class P and
liver class I alcohol dehydrogenase. Eur J Biochem 241: 849–885
Merril CR, Goldman D, Sedman SA, Ebert MH (1981) Ultrasensitive stain
for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211: 1437–1438
Nonomura AM, Benson AA (1992) The path of carbon in photosynthesis:
improved crop yields with methanol. Proc Natl Acad Sci USA 89:
9794–9798
Olson BJ, Skavdahl M, Ramberg H, Osterman JC, Markwell J (2000)
Formate dehydrogenase in Arabidopsis thaliana: characterization and possible targeting to the chloroplast. Plant Sci 159: 205–212
Pietrzak M, Shillito RD, Hohn T, Potrykus I (1986) Expression in plants of
two bacterial antibiotic resistance genes after protoplast transformation
with a new plant expression vector. Nucleic Acids Res 14: 5857–5868
Prather CW, Sisler EC (1972) Glycine and glyoxylate decarboxylation in
Nicotiana rustica roots. Phytochemistry 11: 1637–1647
Rothstein RJ (1983) One-step gene disruption in yeast. Methods Enzymol
101: 202–211
Sakamoto A, Ueda M, Morikawa H (2002) Arabidopsis glutathionedependent formaldehyde dehydrogenase is an S-nitrosoglutathione reductase. FEBS Lett 515: 20–24
Sardi E, Tyihak E (1994) Simple determination of formaldehyde in dimedone adduct form in biological samples by high performance liquid
chromatography. Biomed Chromatogr 8: 313–314
Sandermann H (1992) Plant metabolism of xenobiotics. Trends Biochem Sci
17: 82–84
Shafqat J, El-Ahmad M, Danielsson O, Martı́nez MC, Persson B, Parés X,
Jörnvall H (1996) Pea formaldehyde-active class III alcohol dehydrogenase: common derivation of the plant and animal forms but not of the
corresponding ethanol-active forms (classes I and P). Proc Natl Acad Sci
USA 93: 5595–5599
Smith M (1988) Molecular genetic studies on alcohol and aldehyde dehydrogenase: individual variation, gene mapping and analysis of regulation. Biochem Soc Trans 16: 227–230
Uotila L, Koivusalo M (1979) Purification of formaldehyde and formate
dehydrogenases from pea seeds by affinity chromatography and
S-formylglutathione as the intermediate of formaldehyde metabolism.
Arch Biochem Biophys 196: 33–45
Uotila L, Koivusalo M (1989) Glutathione-dependent oxidoreductases:
formaldehyde dehydrogenase. In D Dolphin, R Poulson, O Avramovic,
eds, Coenzymes and Cofactors. Glutathione: Chemical, Biochemical and
Medical Aspects, Vol III, Part A. John Wiley and Sons, New York, pp
517–55
Wehner EP, Rao E, Brendel M (1993) Molecular structure and genetic
regulation of SFA, a gene responsible for resistance to formaldehyde in
Saccharomyces cerevisiae, and characterization of its protein product. Mol
Gen Genet 237: 351–358
Wippermann U, Fliegmann J, Bauw G, Langebartels C, Maier K, Sandermann H Jr (1999) Maize glutathione-dependent formaldehyde dehydrogenase: protein sequence and catalytic properties. Planta 208: 12–18
Wolverton BC, Johnson A, Bounds K (1989) Interior Landscape Plants for
Indoor Air Pollution Abatement: Final Report. National Aeronautics and
Space Administration, John C. Stennis Space Center, Pearl River, MS
Downloaded from on June 16, 2017 - Published by www.plantphysiol.org
Copyright © 2003 American Society of Plant Biologists. All rights reserved.
2255