trends in plant science
Reviews
Cytochromes P450 for engineering
herbicide tolerance
Danièle Werck-Reichhart, Alain Hehn and Luc Didierjean
In recent years, genome sequencing has revealed that cytochromes P450 (P450s) constitute
the largest family of enzymatic proteins in higher plants. P450s are mono-oxygenases that
insert one atom of oxygen into inert hydrophobic molecules to make them more reactive and
hydrosoluble. Besides their physiological functions in the biosynthesis of hormones, lipids
and secondary metabolites, P450s help plants to cope with harmful exogenous chemicals
including pesticides and industrial pollutants, making them less phytotoxic. The recovery of
an increasing number of plant P450 genes in recombinant form has enabled their use in
experimentation, which has revealed their extraordinary potential for engineering herbicide
tolerance, biosafening, bioremediation and green chemistry.
O
ne of the striking features emerging from the various plant
genome sequencing projects is the extraordinary diversity
of the cytochrome P450 superfamily of oxygenases in
higher plants: ~300 genes are expected in the diminutive genome
of Arabidopsis (http://drnelson.utmem.edu/CytochromeP450.html;
http://ag.arizona.edu/p450/). It is now clear that P450s form the
largest class of plant enzymes, and that several hundreds of P450
proteins are probably encoded by most plant species. Even if the
number of P450 genes is not proportional to the size of each
genome, this number might be larger in the case of polyploid
crops. Some essential P450 functions are conserved among plant
species1,2, including hormone, sterol and oxygenated fatty acid
synthesis. Others, probably the majority, are involved in aspects
of secondary metabolism that differ from plant to plant. Consequently, P450 number and substrate specificity also differs from
plant to plant. This is one of the reasons for herbicide selectivity.
P450 signatures
Biochemical characteristics
P450s are heme proteins that use electrons from NADPH to catalyse the activation of molecular oxygen. The catalysed reaction is
usually a mono-oxygenation, with the formation of a molecule of
water and an oxygenated product (Eqn 1), but other more atypical
activities, such as dimerizations, isomerizations, dehydrations and
reductions, have also been reported3,4.
R-H 1 O21 NADPH 1 H1 → R-OH 1 H2O 1 NADP1
(1)
Electrons from NADPH are transferred one by one to P450s via
FAD and FMN flavoproteins called cytochrome P450-reductases.
Both plant P450s and their reductases are usually bound via
their N-terminus to the cytoplasmic surface of the endoplasmic
reticulum.
Catalysis is inhibited by carbon monoxide, which is a highaffinity ligand of the reduced enzyme and competes with oxygen.
The reduced-carbon monoxide complex absorbs light at 450 nm.
This is a major characteristic of P450 proteins (P450 means ‘pigment absorbing at 450 nm’). Carbon monoxide can be displaced
by light, with a maximum efficiency at 450 nm.
Common structural features
P450 proteins have molecular masses ranging from 45 to 62 kDa,
and might have as little as 16% amino acid identity. However, their
overall tridimensional structure is conserved3,5, as are a few residues
116
March 2000, Vol. 5, No. 3
on both sides of the heme. A Phe-x-x-Gly-x-Arg-x-Cys-x-Gly
motif near the C-terminus is the most conserved sequence among
P450s (Fig. 1). It includes the cysteine that serves as a fifth ligand
to the heme iron. Another Ala/Gly-Gly-x-Asp-/Glu-Thr-Thr/Ser
consensus, located ~150 residues upstream, corresponds to the
oxygen-binding and activation groove in the I helix on the distal
side of the heme. A few other less conserved motifs, such as a ProPro-x-Pro hinge near the N-terminus, or a Pro-Glu/Asp-Arg/HisPhe/Trp sequence between I helix and heme-binding cysteine are
found in most P450 proteins.
Such conserved sequences, and their location, are considered to
be signatures for the proteins deduced from P450 genes, as are the
biochemical criteria for P450-dependent activities, such as:
• O2-dependence.
• NADPH-dependence.
• Inhibition by carbon monoxide and reversion by light.
• Endoplasmic reticulum location.
• Inhibition by antibodies directed against P450 reductases.
First hints of P450-dependent herbicide metabolism
Formation of oxygenated metabolites
An indication that P450s might be involved in herbicide
metabolism came from the analysis of herbicide residues formed
in vivo. Among the major metabolites of most classes of herbicides are aryl- or alkyl-hydroxylated, and N-, S- or O-dealkylated
products and their glucose conjugates. Dealkylation results from
the oxygenation of the a carbon to the heteroatom, followed
by spontaneous hydrolytic cleavage, and the elimination of aldehyde. Most herbicides, for example prosulfuron, diclofop and
chlortoluron, can be converted by P450s into several metabolites
(Figs 2 and 3). In the case of the phenylurea chlortoluron, detoxification of the herbicide is achieved either via hydroxylation
of the ring-methyl, or via di-N-demethylation. The monoN-demethylated product remains phytotoxic. Chlortoluron provides a good example of metabolism associated with herbicide selectivity in weeds and crops. In the tolerant winter
wheat, the half-life of chlortoluron is ,24 h, and the main
metabolite is ring-methyl hydroxylated. In the susceptible weed
Alopecurus myosuroides, the main metabolite is the monoN-demethylated compound. Phytotoxicity of the metabolite is
associated with a half-life of .24 h. By contrast, in the
tolerant weed Veronica persica, the herbicide has a half-life
of only 6 h, and is converted to the non-phytotoxic
di-N-demethylated product6.
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01567-3
trends in plant science
Reviews
P450 inducers and inhibitors behave as herbicide safeners
or synergists
Other clues to P450 involvement in vivo can be obtained using
P450 inducers (Fig. 4), such as:
• Herbicide safeners (or antidotes).
• Ethanol.
• Drugs, such as phenobarbital, aminopyrine or clofibric acid.
Or by using inhibitors (Fig. 4), such as:
• Compounds with a methylenedioxo or acetylenic function.
• Azoles7.
The disadvantage of such effectors is that they are usually selective
activators or inhibitors of some P450 isozymes. Their advantage
is that they can be used to differentiate between P450 isoforms
catalysing different reactions8–11.
Herbicide safeners are widely used to protect monocot crops
from herbicide damage, and for more selective weed control12.
Some herbicide safeners bear a structural homology with herbicides.
However, their protective effect appears to result, not from
competition for the target site, but from a general enhancement of
herbicide detoxification processes, including induction of:
• P450 oxygenases.
• Glutathione S-transferases.
• Glucosyl transferases.
• Vacuolar transport.
• Synthesis of glutathione.
• Glutathione peroxidase.
• Sulphate assimilation.
Some other P450 inducers, such as ethanol, phenobarbital, herbicides or MgCl2, have partially additive to synergistic effects when
added together with safeners8,9,13,14. Some of these appear to be
active only during specific stages of plant development9.
One of the P450 inhibitors most commonly used as a synergist
to characterize P450-dependent reactions in vivo is 1-aminobenzotriazole (ABT). This commercially available molecule is a mechanism-dependent irreversible inactivator of P450 enzymes. It shows
relatively little selectivity towards P450 isoforms, especially when
used at high concentration (>100 mM) because of its small size and
the extreme reactivity of the benzyne, which is the product of
the oxygenation reaction15. Synergism between organophosphate
insecticides (e.g. malathion or terbufos and its metabolites) and herbicides is also an indication of the involvement of P450 in herbicide
tolerance, because such insecticides behave as mechanism-based
inhibitors of some P450 enzymes16–18.
Weed resistance
Resistance might result from increased metabolism
In the past ten years, an increasing number of herbicide-resistant
weed populations have been reported, sometimes showing multipleor cross-resistance to herbicides with different chemistries and target
sites. Weeds with several mechanisms of resistance are usually allogamous species: resistance mutations accumulate in a single plant as a
result of cross-pollination. Resistance mechanisms depend on the
weed biotype. They are often related to mutations of target sites, but
can also result from increased metabolism, or from both. An increase
in P450-dependent metabolism was first demonstrated in resistant
Lolium rigidum biotypes from Australia and Alopecurus myosuroides
biotypes that appeared in Europe19,20. More recently, phenylurearesistant populations of Phalaris minor in Asia have been
described21,22. Because the isolation of active microsomal fractions
from weeds is difficult, evidence for P450 involvement has been provided mainly by in vivo experiments using inhibitors. In most cases,
resistance appears to result from an increase in metabolism that can
also be detected in the susceptible biotype. Polar products formed by
resistant weeds are the same as those isolated from the tolerant crops:
(a)
Meander
NADPH
+ H+
NADP+
FAD
Heme-binding
domain
FMN
1e− + 1e−
Fe
O
O−
P450
reductase
P450
I-helix
ER
Hinge
Substrate
(b)
Subfamily
Family
Isoform
CYP1A1
>40% identity
<40% identity
Same family
>55% identity
Same subfamily
Other family
<55% identity
Other subfamily
Trends in Plant Science
Fig. 1. Structure and nomenclature of plant cytochromes P450
(P450s). (a) Plant P450s and P450 reductases are anchored to the
endoplasmic reticulum (ER) via their N-termini. Recent data suggest
that a portion of the globular part of P450 proteins also interacts
strongly with the membrane. Domains that are highly conserved in
most P450s are shown in black. These include: the loop that covalently binds heme via the thiolate of a cysteine (Phe-x-x-Gly-x-Arg-xCys-x-Gly); the portion of the I-helix involved in oxygen activation
and the transfer of protons to the active site (Ala/Gly-Gly-x-Asp/GluThr-Thr/Ser); a structurally highly conserved region probably important for heme-binding called the ‘meander’ (Pro-Glu/Asp-Arg/
His-Phe/Trp)5; and the proline-rich region that forms a hinge between
the membrane-anchored N-terminal helix and the globular part of the
protein. A few residues interacting with the reductases are also conserved. Most other regions are highly variable, especially those
involved in substrate recognition. There are at least two different P450
reductases in higher plants. They channel the electrons from NADPH
to the P450 proteins by transferring them from FAD to FMN. (b) The
nomenclature of P450 genes is based on the amino acid identity
among the proteins they encode. Plant P450 genes have been numbered in chronological order depending on their date of submission to
the P450 nomenclature committee (http://drnelson.utmem.edu/
CytochromeP450.html). Plant P450 families bear numbers from
CYP71 to CYP99, and from CYP701 and above. In August 1999, 417
plant P450 genes distributed among 48 families were recorded.
thus, similar P450s appeared to be involved. From the differential
inhibition of the formation of some metabolites, it was assumed that
several P450s catalysed different reactions on the same herbicide,
or even sometimes the same reaction on a single molecule6,11,21.
Studies performed on resistant Lolium provided the first evidence
that metabolism of herbicides can be induced by light23.
March 2000, Vol. 5, No. 3
117
trends in plant science
Reviews
Prosulfuron
Trends in Plant Science
(a)
OCH3
CF3
N
N
SO2NHCONH
N
CH3
P450
P450
P450
II
I
HO
N
CH3
CH2OH
CH3
CH3
OH
O
O
OH
O
O
Cl
O
Cl
HO
Cl
CH3
OH
O
O
P450
CH3
CH3
OH
O
O
+
Cl
N
SO2NHCONH
N
N
(b) Diclofop
N
N
SO2NHCONH
N
OH
CF3
N
N
SO2NHCONH
OCH3
CF3
OCH3
CF3
III
O
+
HO
O
HO
O
Cl
Cl
Cl
Cl
Fig. 2. Examples of cytochrome P450-catalysed oxygenations of herbicides. (a) Different reactions can be catalysed by P450 on a single herbicide molecule. In wheat, prosulfuron is metabolized via phenyl-ring hydroxylation, alkyl hydroxylation and O-demethylation of the triazine29.
It is not yet known if these different reactions are catalysed by a single P450 or different isoforms, although metabolism in wheat, maize and
sorghum appear to involve different P450s. (b) Metabolism of diclofop is an example of what is usually called an ‘NIH shift’. The oxidative
attack of halogenated phenyl rings by P450 enzymes often results in a hydroxylation with simultaneous migration of the halogen atom to an
adjacent position. One explanation proposed for this migration is the formation of an intermediate arene oxide.
How do weeds become rapid metabolizers?
The mechanism of acquiring increased metabolism remains to be
determined. Some ultra-rapid metabolizer phenotypes in humans
have been shown to result from the duplication of drug-metabolizing P450 genes24. A point-mutation in a structural gene is also
possible. Another plausible explanation is the altered regulation of
a stress-response pathway, in which one or more P450s are
induced. In insects, constitutive overexpression regulated in trans
is the only mechanism of P450-mediated resistance to an insecticide reported to date25. Recent data on the regulation of glutathione transferases suggest that a similar mechanism might occur
in resistant weed biotypes26, which would explain some cases of
cross-resistance.
In vitro assays
Direct proof of the involvement of plant P450s in herbicide
metabolism was obtained as early as 1969 (Ref. 27), when it was
shown that the phenylurea monuron was actively dealkylated in
cotton microsomes. Together with kaurene hydroxylase, this
was one of the first two P450-dependent reactions characterized
in higher plants. This work was confirmed with other phenylurea
20 years later6, and started a more systematic exploration of
the importance of P450s in the metabolism of other classes of
118
March 2000, Vol. 5, No. 3
herbicides. This work, mostly performed on major crop plants10,28,29,
confirmed the main role of P450s in the oxidation of most classes
of herbicides.
In vitro assays allowed more systematic enzymologic studies,
confirming the mode of action of some inhibitors16,30, and showed
that P450s with catalytic parameters that were different from those
of the constitutive forms (i.e. often with a higher Km) were induced
by chemical treatments8,9. Several P450 isoforms thus appeared to
participate in the metabolism of a single herbicide in the same
plant. Assays performed with plant microsomes showed that
herbicide-metabolizing activities were almost always induced in
much larger proportions than the total P450 content. This indicated
that the detoxification pathways involve P450 isozymes that are
relatively minor, and hence difficult to purify compared with constitutively expressed enzymes. However, the question remained:
whether a few specialized P450s were responsible for herbicide
metabolism, or if agrochemicals are fortuitous substrates of isozymes involved in physiological pathways? Data13 suggesting that
inducers and inhibitors have similar effects on diclofop and lauric
acid hydroxylases in wheat microsomes, and that both substrates
are mutual competitive inhibitors have provided support for the
second hypothesis. In this case, the herbicide and the biological
substrate appear to be metabolized by the same P450 protein.
trends in plant science
Reviews
10 µM
O
H
CYP76B1
(46 min–1)
O
N
CH3
CYP76B1
(803 min–1)
O
NH
NH2
CH3
CH3
Cl
CH3
NH
Cl
Non-phytotoxic
Phytotoxic
N
CH3
NH
CH3
Cl
Control
CYP71A10
(0.7 min–1)
1 µM
O
Chlortoluron
CYP76B1
CH3
N
OHCH2
CYP73A1
(0.014 min–1)
CYP81A1
(0.28 min–1)
NH
CH3
Cl
Non-phytotoxic
Control
CYP71A10
Fig. 3. Plant cytochromes P450 (P450s) metabolizing phenylurea. P450s metabolize the phenylurea chlortoluron either via ring-methyl hydroxylation or via N-dealkylation. The mono-N-dealkylated product remains phytotoxic, but ring-methyl hydroxylation and di-N-demethylation lead to
non-phytotoxic compounds. The genes of several plant P450s that catalyse these reactions were isolated recently. Their respective turnover rates
are indicated next to each reaction in blue boxes. The CYP73A1 and CYP81B1 genes were isolated from Jerusalem artichoke (Helianthus tuberosus)
– their main function is phenylpropanoid and oxygenated fatty acid biosynthesis, respectively. Both proteins expressed in yeast catalyse ringmethyl hydroxylation of chlortoluron35,36, but the efficiency of the reaction is probably too low to confer a significant resistance. CYP71A10 was
isolated from soybean. The recombinant protein catalyses chlortoluron ring-methyl hydroxylation at a higher rate38 and catalyses the monoN-dealkylation of several phenylurea. When over-expressed in tobacco, CYP71A10 increases its tolerance to chlortoluron. The lower part of the
figure shows control plants and CYP71A10-transformed plants growing on a medium containing 1 mM chlortoluron. CYP76B1 is a gene from
Jerusalem artichoke. The protein expressed in yeast catalyses the double dealkylation of phenylurea37. The second dealkylation is slower than the
first, but the turnovers of both reactions are high, and comparable to those measured with physiological substrates. When over-expressed in
tobacco, CYP76B1 also confers an increased tolerance to chlortoluron. The upper part of the figure shows excised leaf pieces of control plants
and CYP76B1-transformed plants that were aged for ten days under light on a medium containing 10 mM chlortoluron. Leaf bleaching caused by
the inhibition of photosynthetic electron transport by the herbicide is suppressed in the recombinant plants. Both CYP71A10 and CYP76B1 also
increase resistance to linuron, which can be detoxified by a single N-dealkylation. Figure reproduced, with permission, from Ref. 39.
P450 genes for engineering herbicide tolerance
Bacterial genes
Once the involvement of P450s in herbicide selectivity had been
demonstrated, the isolation of their genes became a commercial
goal for the control of herbicide tolerance in crops and weeds, and
to help the optimization of new active compounds. The purification
of minor plant P450s being difficult, the first genes encoding
herbicide-metabolizing P450s were not isolated from plants, but
from bacteria and mammals. The genes of two inducible sulfonylurea metabolizing P450s, SU1 and SU2, were isolated from the soil
bacterium Streptomyces griseolus. These bacterial P450s are soluble enzymes that need two electron transporters, a reductase and a
ferredoxin, to obtain reducing equivalent from NAD(P)H. A
chloroplast transit sequence was added to the SU1 (CYP105A1)
gene to target the protein to the chloroplast stroma, a compartment
providing an adequate redox partner (ferredoxin). It was then
expressed in tobacco under the control of promoters directing
ectopic expression, or a promoter specific for the tapetum31. When
the gene product is targeted to the chloroplast in this way, it activates the harmless sulfonylurea R7402, which becomes a highly
phytotoxic herbicide. When SU1 is expressed in the whole plant,
R7402 treatment results in plant death. When the gene is expressed
from a tapetum-specific promoter, R7402 treatment of immature
flowers results in male sterility. Currently, the SU1/R7402 system
is widely used as a negative selection marker (Fig. 5).
Another bacterial P450 gene has been isolated from a
Rhodococcus strain that is able to grow on thiocarbamates as sole
carbon and nitrogen sources32. This P450 confers on other bacteria
the ability to degrade the herbicide S-ethyl-dipropylcarbamothioate (EPTC), and to exert a biosafening activity on inoculated
maize seedlings: bacteria accumulated in the crop rhizosphere
detoxify EPTC present in the soil, thus protecting maize from
herbicide injury.
Mammalian genes
Only a few mammalian P450s have been shown to play a significant role in the metabolism of drugs and other xenobiotics. Their
herbicide-metabolizing ability has been explored to create transgenic plants. This has been achieved largely by exploiting rat
CYP1A1 to generate herbicide-tolerant tobacco and potato
plants33–35. The enzyme, which was reported previously to metabolize various planar xenobiotics, shows a low substrate specificity,
metabolizing herbicides of different chemistries, including atrazine,
chlortoluron and pyriminobac-methyl. In addition, CYP1A1
March 2000, Vol. 5, No. 3
119
trends in plant science
Reviews
(a) P450 inducers: herbicide safeners
Cl
COOC2H5
O
N
Cl
O
O
CH2C
O
N
CH(CH2)4CH3
C
N
O
C
Cl3C
O
CH3
Fenchlorazole-ethyl
Cloquintocet-mexyl
1,8-Naphthalic anhydride (NA)
(b) P450 inactivators: herbicide synergists
H2O
N
N
P450 + NADPH + H+ + O2 +
N
N
2N2
NH2
N
1-Aminobenzotriazole
Benzyne
CH3O
N
N
Heme adduct
CH3O
P
CH3O
Fe
S
S
CH
P
COOC2H5
CH3O
CH2COOC2H5
Malathion
S
S
O
CH
COOC2H5
CH2COOC2H5
Malaoxon
Protein adduct
Trends in Plant Science
Fig. 4. Effectors of cytochrome P450 enzymes. (a) Herbicide safeners, such as naphthalic anhydride, cloquintocet-mexyl and fenchlorazoleethyl, are used as seed coatings or for co-treatment with herbicidal molecules to selectively increase the resistance of crops relative to that of
weeds. In spite of a structural similarity to herbicides, they have no effect on herbicide target sites, but instead activate herbicide metabolism.
Their effect is pleiotropic, and they increase P450s, glutathione transferases and glucosyl transferases, as well as glutathione content in the
treated plants. They apparently specifically up-regulate the P450 isoforms involved in detoxification. (b) P450 inhibitors can be used as herbicide
synergists. By inhibiting herbicide metabolism, they increase the proportion of toxic molecules reaching the target site. Competitive and tightbinding inhibitors are effective, but mechanism-based irreversible inhibitors are the most efficient. The metabolism of such molecules generates
a reactive intermediate that forms a covalent adduct with the P450 heme or apoprotein. They are, therefore, P450-specific (although some might
also inactivate some peroxidases), and they are selective for some P450 isoforms. Different classes of insecticides have been shown to behave
as herbicide synergists. Some (e.g. carbamates) might be just competitive inhibitors. Others, such as organophosphates, when oxygenated by
P450s, release atomic sulphur that covalently binds to the apoprotein, leading to its inactivation. When fields are treated simultaneously with a
herbicide and an insecticide that are recognized by the same P450, it could lead to crop destruction.
metabolizes herbicides with low regio-specificity, catalysing both
chlortoluron ring-methyl hydroxylation and N-dealkylation, as
well as N-dealkylation of atrazine on two different nitrogens. To
improve the enzyme efficiency, and input of reducing equivalents,
two strategies have been used. The first strategy involves expressing a fusion of CYP1A1 with the yeast P450-reductase in plants,
mimicking bacterial P450s with high catalytic turnovers. The second strategy targets the fusion proteins to the chloroplast, the major
site of NADPH production under light, to improve coupling of
P450 reduction with photosynthetic electron transfer. The fusion
protein located in the thylakoid membranes shows enhanced activity under light irradiation. However, the transfer of mammalian
genes into plants poses an ethical problem, in particular when
they encode a protein such as CYP1A1, which has been shown to
activate polyaromatic hydrocarbons into procarcinogens.
120
March 2000, Vol. 5, No. 3
Plant P450 genes
The first plant P450s available in recombinant systems that were
assayed for herbicide metabolism were enzymes involved in phenolic or lipid pathways36,37. They catalyse extremely low, but
detectable, chlortoluron hydroxylation. However, the kinetic constants of the reaction indicate that these P450s do not play a significant role in herbicide detoxification in vivo. Two plant P450s
that metabolize herbicides more efficiently have been reported in
the past year. Both metabolize phenylurea, but none of the other
classes of herbicides tested. One gene, CYP76B1, was isolated
from Jerusalem artichoke (Helianthus tuberosus) on the basis of
its inducibility by Mn21 ions or drugs such as aminopyrine and
phenobarbital, which also induce mammalian P450 enzymes. When
expressed in yeast, together with an Arabidopsis P450 reductase,
the CYP76B1 protein catalyses the dealkylation of several planar
trends in plant science
Reviews
Proherbicide activation
N
H
Active in plant
chloroplasts
SO2
OCH3
N
N
SO2
OCH3
SU1
N
SO2NHCONH
SO2NHCONH
N
N
OCH3
OCH3
R7402
proherbicide
R7402 metabolite
= herbicide
Negative selection system
Control plants
SU1 expression
in all plant tissues
Control plants
R7402 treatment
SU1 tissue-specific
expression in the tapetum
R7402 treatment
Inhibition of
plant growth
Inhibition of
pollen germination
Fig. 5. SU1 of Streptomyces griseolus: a cytochrome P450 that metabolizes sulfonylurea. The SU1 (CYP105A1) gene was isolated from the soil
bacterium Streptomyces griseolus, which metabolizes herbicides of the class of sulfonylurea. Expression of this P450 is induced by sulfonylurea
and phenobarbital in the bacteria. SU1 detoxifies some phenylurea herbicides, such as chlorimuron or chlorsulfuron, but with only low efficiency. At a higher rate, it activates the proherbicide R7402 via N-dealkylation. This property has been used to develop negative selection systems that are now widely used. To be active in plants, the bacterial protein needs to be targeted to the chloroplast where ferredoxin provides an
appropriate electron donor. When the protein is expressed in all plant tissues, proherbicide treatment leads to plant death or severe growth inhibition. Control and SU1-transformed plants (left) were grown for 21 days on a medium containing 50 nM R7402. Selection is also possible on
soil-grown plants. For example, SU1 is used to select stable insertion events after transposon mutagenesis (http://www.jic.bbsrc.ac.uk/
Sainsbury-lab/jonathan-jones/SINS-database/sins.htm). When the protein is expressed under the control of a promoter that is specifically activated
in the tapetum of the anther, the treatment of immature flowers produces pollen that is unable to germinate, resulting in male sterility. Germination
of the pollen of control and SU1-transformed plants seven days after R7402 treatment (right). Figure reproduced, with permission, from Ref. 31.
xenobiotics, including phenylurea, with a high turnover comparable to that of physiological substrates38. Phenylurea are converted
to non-phytotoxic di-N-dealkylated products. A second gene,
CYP71A10, was isolated from soybean on the basis of P450conserved sequences39. The CYP71A10 protein expressed in yeast
converts chlortoluron both to ring-methyl hydroxylated and, to a
lesser extent, mono-N-demethylated compounds, as such behaving like rat CYP1A1. Other phenylurea were exclusively monoN-demethylated. Catalytic turnover of CYP71A10 with phenylurea is more than ten times lower than that of CYP76B1, which is
not so surprising considering its looser regio-specificity. Both genes
confer increased tolerance to phenylurea when over-expressed in
tobacco (Fig. 3).
Only preliminary data are available concerning plant P450s
metabolizing other classes of herbicides. Genetic analysis has
suggested that, in maize, chlortoluron and bentazon metabolism
might be under the control of the same gene. A CYP72, recently
isolated from a tolerant line, appears to encode a protein capable
of metabolizing both chlortoluron and bentazon, as well as chlorimuron and the insecticide malathion (M. Barrett, unpublished).
Data concerning the turnover rates of such a multi-metabolizer
P450 are not available.
Concluding remarks
It is clear that P450 differences in substrate specificity, redundance
or regulation from plant to plant are responsible for herbicide tolerance and selectivity. They are also one of the keys to herbicide
cross-resistance, which is becoming more and more frequent in
weeds. So far, only a few recombinant plant P450s have been
examined for their ability to metabolize exogenous molecules,
and those assayed have been tested for a few herbicides, and even
fewer pesticides and pollutants. Recent data show that several
P450 families can metabolize the same class of herbicides, some
being poor metabolizers, and others being much more efficient.
March 2000, Vol. 5, No. 3
121
trends in plant science
Reviews
This is a situation reminiscent of that observed for drug metabolism
in humans – P450s showing sometimes broad, sometimes narrower,
and overlapping substrate specificities. It is not clear if the poor and
non-metabolizers are enzymes with essential physiological functions. It is not clear either, if there are P450s specialized in stress
response, metabolizing a broad range of exogenous molecules, and
if these P450s have a physiological function. The information
needed for efficient engineering of herbicide tolerance includes the
relative efficiencies of the different isoforms that are able to catalyse detoxification of a given molecule, their substrate specificities,
and their physiological impacts.
In the case of human P450s, it is usually considered that 20 out of
~50 P450 genes play a role in the metabolism of drugs and other
xenobiotics. It is possible that a similar proportion of plant genes
participate in herbicide metabolism. Considering the huge number
and diversity of P450 genes in the different plant species, the P450
family might provide an amazing source of catalysts for engineering
pesticide tolerance by plant transformation, but also for biosafening
using soil bacteria. Other possible applications, which have not been
exploited to date, include positive markers for plant transformation,
soil and waste-water bioremediation and green chemistry. Because
P450s are active in the chloroplasts, it is possible to envisage their
integration in the chloroplast genome. This would, at the same time,
increase the number of copies of relevant genes per plant cell,
improve the coupling of electron transfer with photosynthesis, and
provide gene containment because plastid genes usually cannot be
transmitted by pollen to crop-related weeds.
To date, little is known about the regulation of plant P450 genes
or about the mechanisms of acquisition of herbicide tolerance.
These mechanisms might be similar in tolerant crops and weeds.
It is interesting to note that tolerance appears more frequent in
monocot weeds and crops than in dicot plants. This might be
related to the higher GC content of their genome, which would
favour recombinational events and gene duplication.
Acknowledgements
We are grateful to Barbara Sears for her critical reading of the
manuscript. A.H. and L.D. thank Rhône-Poulenc Agro and DuPont
Agrochemicals, respectively, for financial support. This paper was
written during D.W-R.’s sabbatical at the Carnegie Institution of
Washington, Stanford, CA, USA.
References
1 Schuler, M.A. (1996) Plant cytochrome P450 monooxygenases. Crit. Rev.
Plant. Sci. 15, 235–284
2 Chapple, C. (1998) Molecular genetics analysis of plant cytochrome P450dependent monooxygenases. Annu. Rev. Plant Physiol. Mol. Biol. 49, 311–343
3 Halkier, B.A. (1996) Catalytic reactivities and structure/function relationships
of cytochrome P450 enzymes. Phytochemistry 43, 1–21
4 Mansuy, D. (1998) The great diversity of reactions catalysed by cytochromes
P450. Comp. Biochem. Physiol. C 121, 5–14
5 Hasemann, C.A. et al. (1995) Structure and function of cytochromes P450: a
comparative analysis of three crystal structures. Structure 3, 41–62
6 Mougin, C.P. et al. Plant and fungal cytochromes P450: their role in pesticide
transformation. In Pesticide Biotransformation in Plants and Microorganisms:
Similarities and Divergences (ACS Series) (Hoagland, R.E. et al., eds), Oxford
University Press (in press)
7 Werck-Reichhart, D. (1995) Herbicide metabolism and selectivity: role of
cytochrome P450. Proc. Brighton Crop Prot. Conf. Weeds 7B-1, 813–821
8 Frear, D.S. et al. (1991) Induced microsomal oxidation of diclofop, triasulfuron,
chlorsulfuron, and linuron in wheat. Pestic. Biochem. Physiol. 41, 274–287
9 Persan, M.W. and Schuler, M.A. (1995) Differential induction of cytochrome
P450-mediated triasulfuron metabolism by naphthalic anhydride and
triasulfuron. Plant Physiol. 109, 1483–1490
122
March 2000, Vol. 5, No. 3
10 Moreland, D.E. et al. (1995) Partial characterization of microsomes isolated
from mung bean cotyledons. Pestic. Biochem. Physiol. 52, 98–108
11 Preston, C. et al. (1996) Multiple resistance to dissimilar herbicide chemistries
in a biotype of Lolium rigidum due to enhanced activity of several herbicide
degrading enzymes. Pestic. Biochem. Physiol. 54, 123–134
12 Hatzios, K.K. (1997) Regulation of xenobiotic degrading enzymes with
herbicide safeners. In Regulation of Enzymatic Systems Detoxifying
Xenobiotics in Plants (NATO ASI Series 3. High Technology) (Vol. 37)
(Hatzios, K.K., ed.), pp. 275–288, Kluwer
13 Zimmerlin, A. and Durst, F. (1992) Aryl hydroxylation of the herbicide diclofop
by a wheat cytochrome P450 monooxygenase. Plant Physiol. 100, 874–881
14 Frear, D.S. and Swanson, H.R. (1996) Cytochrome P450-dependent
hydroxylation of prosulfuron (CGA 152005) by wheat seedlings microsomes.
J. Agric. Food Chem. 44, 3658–3664
15 Ortiz de Montellano, P.R. and Mattews, J.M. (1981) Autocatalytic alkylation of
the cytochrome P450 prosthetic heme group by 1-aminobenzotriazole: isolation
of an N,N-bridged benzyne-protoporphyrin IX adduct. Biochem. J. 195, 761–764
16 Kreuz, K. and Fonné-Pfister, R. (1992) Herbicide–insecticide interaction in
maize: malathion inhibits cytochrome P450-dependent primisulfuron
metabolism. Pestic. Biochem. Physiol. 43, 232–240
17 Christoffer, J.T. et al. (1994) Malathion antagonizes metabolism-based
chlorsulfuron resistance in Lolium rigidum. Pestic. Biochem. Physiol. 49, 172–182
18 Baerg, R.J. et al. (1996) Insecticide and insecticide metabolite interactions with
cytochrome P450 mediated activities in maize. Pestic. Biochem Physiol. 55, 10–20
19 Holt, J.S. et al. (1993) Mechanisms and agronomic aspects of herbicide
resistance. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44, 203–229
20 De Prado, R. and Menendez, J. (1997) Cross-resistance and herbicide
metabolism. In Regulation of Enzymatic Systems Detoxifying Xenobiotics in
Plants (NATO ASI Series 3. High Technology) (Vol. 37) (Hatzios, K.K., ed.),
pp. 351–366, Kluwer
21 Singh, S. et al. (1998) Effect of ABT on the activity and rate of degradation of
isoproturon in susceptible and resistant biotypes of Phalaris minor and in
wheat. Pestic. Sci. 53, 123–132
22 Singh, S. et al. (1998) Effect of the monooxygenase inhibitor piperonyl
butoxide on the herbicidal activity and metabolism of isoproturon in herbicide
resistant and susceptible biotypes of Phalaris minor and wheat. Pestic.
Biochem. Physiol. 59, 143–153
23 Preston, C. and Powles, S.B. (1997) Light-dependent enhanced metabolism of
chlortoluron in a substituted urea herbicide-resistant biotype of Lolium rigidum
Gaud. Planta 201, 202–208
24 Lundqvist, E. et al. (1999) Genetic mechanisms for duplication and
multiduplication of the human CYP2D6 gene and methods for detection of
duplicated CYP2D6 genes. Gene 226, 327–338
25 Feyereisen, R. (1999) Insect P450 enzymes. Annu. Rev. Entomol. 44, 507–533
26 Cummins, I. et al. (1999) A role for glutathione transferases functioning as
glutathione peroxidases in resistance to multiple herbicides in black-grass.
Plant J. 18, 285–292
27 Frear, D.S. et al. (1969) N-demethylation of substituted 3-(phenyl)-1-methylureas:
isolation and characterization of a microsomal mixed function oxidase from
cotton. Phytochemistry 8, 2157–2169
28 Barrett, M. (1995) Metabolism of herbicides by cytochrome P450 in corn.
Drug Metab. Drug Interact. 12, 299–316
29 Frear, S. (1995) Wheat microsomal cytochrome P450 monooxygenases:
characterization and importance in the metabolic detoxification and selectivity
of wheat herbicides. Drug Metab. Drug Interact. 12, 329–358
30 Helvig, C. et al. (1996) Selective inhibition of a cytochrome P450 enzyme in
wheat that oxidizes both the natural substrate lauric acid and the synthetic
herbicide diclofop. Pestic. Biochem. Physiol. 54, 161–171
31 O’Keefe, D.P. et al. (1994) Plant expression of a bacterial cytochrome P450 that
catalyses activation of a sulfonylurea pro-herbicide. Plant Physiol. 105, 473–482
32 Nagy, I. et al. (1995) Degradation of the thiocarbamate EPTC
(S-ethyl-dipropylcarbamothioate) and biosafening by Rhodococcus sp. strain
NI86/21 involve an inducible cytochrome P450 system and aldehyde
dehydrogenase. J. Bacteriol. 177, 676–687
trends in plant science
Reviews
33 Ohkawa, H. et al. (1998) Molecular mechanisms of herbicide resistance. Rev.
Toxicol. 2, 245–252
34 Inui, H. et al. (1999) Herbicide metabolism and cross-tolerance in transgenic
potato plants expressing human CYP1A1. Pestic. Biochem. Physiol. 64, 33–46
35 Lacour, T. and Ohkawa, H. (1999) Engineering and biochemical
characterization of the rat microsomal cytochrome P450 to ferredoxinNADP(1) reductase from plant chloroplasts. Biochim. Biophys. Acta 1433,
87–102
36 Pierrel, M.A. et al. (1994) Catalytic properties of the plant cytochrome P450
CYP73 expressed in yeast. Substrate specificity of a cinnamate hydroxylase.
Eur. J. Biochem. 224, 835–844
37 Cabello-Hurtado, F. et al. (1998) Cloning, expression in yeast and functional
characterization of CYP81B1, a plant P450 which catalyses in-chain
hydroxylation of fatty acids. J. Biol. Chem. 273, 7260–7267
38 Robineau, T. et al. (1998) The chemically inducible plant cytochrome P450
CYP76B1 actively metabolizes phenylurea and other xenobiotics. Plant
Physiol. 118, 1049–1056
39 Siminsky, B. et al. (1999) Expression of a soybean cytochrome P450
monooxygenase cDNA in yeast and tobacco enhances the metabolism of
phenylurea herbicides. Proc. Natl. Acad. Sci. U. S. A. 96, 1750–1755
Danièle Werck-Reichhart*, Alain Hehn and Luc Didierjean are
at the Dept of Cellular and Molecular Enzymology, Institute of Plant
Molecular Biology, Centre National de la Recherche Scientifique
UPR 406, 28 rue Goethe, 67083 Strasbourg Cedex, France.
*Author for correspondence (tel 133 3 88 35 83 32;
fax 133 3 88 35 84 84; e-mail [email protected]).
Programmed cell death and
aerenchyma formation in roots
Malcolm C. Drew, Chuan-Jiu He and Page W. Morgan
Lysigenous aerenchyma contributes to the ability of plants to tolerate low-oxygen soil
environments, by providing an internal aeration system for the transfer of oxygen from the
shoot. However, aerenchyma formation requires the death of cells in the root cortex. In maize,
hypoxia stimulates ethylene production, which in turn activates a signal transduction
pathway involving phosphoinositides and Ca21. Death occurs in a predictable pattern, is
regulated by a hormone (ethylene) and provides an example of programmed cell death.
R
oots, rhizomes and other plant organs in the soil usually
obtain O2 for respiration directly from their immediate
environment – the soil gaseous atmosphere. But when the
soil becomes excessively wet, transfer of O2 from the air into the
soil is effectively blocked because the larger soil pores, which are
usually air-filled, are water-filled instead. Without replenishment
from the air, any dissolved O2 remaining in the soil is quickly consumed by microorganisms and plants, and the soil is then no
longer able to supply O2. Aerenchyma – tissue comprising a high
proportion of gas-filled spaces or lacunae – provides the plant
with an alternative strategy for obtaining O2. The interconnected
lacunae, extending from below the ground up into the stems and
leaves, make up an internal aeration system, enabling parts of
the plant to survive or grow for a time in environments that are
O2-deficient or even completely devoid of O2.
Spaces form within aerenchymatous organs, either by cell
separation at the middle lamella during development (shizogeny),
or by cell death and dissolution (lysigeny). Sometimes, as in
Sagittaria lancifolia, both processes occur, with lysigenous
aerenchyma in the roots and shizogenous aerenchyma in the
leaf petiole1. In lysigenous aerenchyma formation in the root
cortex of maize, rice and Sagittaria, cell death is first detected
at a distance of a cm or less from the root apical meristem, in the
zone where cell elongation has just been completed2–4. The
space created by the death of cells becomes increasingly prominent in older zones behind the tip. The system of gas-filled
lacunae therefore develops acropetally, extending into the root
towards the tip, simultaneous with the root’s extension into
the soil.
Aerenchyma function
Lysigenous aerenchyma provides not only an internal pathway for
O2 transfer, but also simultaneously reduces the number of O2consuming cells, a feature that might assist in low O2 environments. It has been suggested that aerenchymatous organs are often
water- or fluid-filled5, a feature that would prevent them from
functioning in O2 transfer. However, evidence from the measurement of gas-filled porosity, microscopic observation (Fig. 1) and
O2 transport6 is overwhelming – the lacunae are indeed usually
gas-filled and therefore are able to transfer gases by convection
and diffusion.
Aerenchyma formation is inducible by flooding in maize7,8, in
the coastal grass Spartina patens9 and in many other native
species, both monocot and dicot, that occupy wetland habitats10.
By contrast, in the maize relatives Tripsacum dactyloides (eastern
gamagrass) and Zea luxurians (teosinte)11, as well as in wetland
species, such as rice3,12 and S. lancifolia1,4, aerenchyma forms constitutively, apparently without any requirement for an external
stimulus. Eastern gamagrass possesses notable tolerance to
drought and flooding, which is thought to be because of its deep
rooting pattern and constitutive aerenchyma13. The roots have
been found to exceed 1.8 m in depth, penetrating a resistant clay
pan at or below 0.90 m, apparently when the soil was wet and
mechanically least resistant. Therefore aerenchyma was likely to
have assisted oxygenation of the root at that time. In general, soil
compaction can lower the oxygen concentration of the soil (fewer
large, gas-filled pores for gas exchange with the atmosphere).
Therefore, aerenchyma could be potentially beneficial for growth or
function of roots in a densely packed soil layer. These considerations
1360 - 1385/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01570-3
March 2000, Vol. 5, No. 3
123