Transgenic Research (2005) 14:713–717
Springer 2005

DOI 10.1007/s11248-005-6633-2

Short communication

A model transgenic cereal plant with detoxification activity for the estrogenic
mycotoxin zearalenone

Arisa Higa-Nishiyama1, Naoko Takahashi-Ando1, Tsutomu Shimizu2, Toshiaki Kudo3,

Isamu Yamaguchi1,4 & Makoto Kimura1,3,*
1
Laboratory for Remediation Research, Plant Science Center, RIKEN, 1-7-22 Suehiro, Tsurumi, Yokohama,
Kanagawa 230-0045, Japan
2
Life Science Research Institute, Kumiai Chemical Industry Co., Ltd., Tamari 276, Kakegawa, Shizuoka 436-0011,
Japan
3
Environmental Molecular Biology Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
4
Laboratory for Adaptation and Resistance, Plant Science Center, RIKEN, 1-7-22 Suehiro, Tsurumi, Yokohama,
Kanagawa 230-0045, Japan

Received 7 January 2005; accepted 28 April 2005

Key words: detoxifying gene, Fusarium head blight of wheat, barely, and maize, genetically modified (GM)
cereals, lactonohydrolase, mycotoxin decontamination

Abstract
Zearalenone (ZEN) is an estrogenic mycotoxin produced by the necrotrophic cereal pathogen Fusarium
graminearum. This mycotoxin is detoxified by ZHD101, a lactonohydrolase from Clonostachys rosea, or
EGFP:ZHD101, its fusion to the C-terminus of an enhanced green fluorescence protein. We previously
showed that egfp:zhd101 is efficiently expressed in T0 leaves of rice. In this study, we assessed the feasibility
of in planta detoxification of the mycotoxin using progeny. When protein extract from T1 leaves was
incubated with ZEN, the amount of the toxin decreased significantly as measured by HPLC. ZEN deg-
radation activity was also detected in vivo in transgenic T2 seeds. These results suggest that zhd101 can be
exploited as an efficient and cost-effective system for protection of important cereals that are more sus-
ceptible to the pathogen (e.g., wheat and maize) from contamination with the estrogenic mycotoxin.

Introduction necessarily correlated with the visible disease

Fusarium head blight (FHB) is a significant disease
of small grain cereals (e.g., wheat, barley and
maize) caused by Fusarium species such as
F. culmorum and F. graminearum. In addition to
the yield loss, Fusarium-infected grains are often
contaminated with the mycotoxins trichothecenes
(e.g., deoxynivalenol) and zearalenone (ZEN).
Although several fungicides reduce the disease
severity of FHB, mycotoxin contamination is not
* Author for correspondence
E-mail: mkimura@postman.riken.go.jp
symptoms (Edwards et al., 2001). For example, a
number of field and greenhouse studies demon-
strated that application of the strobilurin fungicide
azoxystrobin increases the amount of deoxynivale-
nol in wheat grains (Simpson et al., 2001). In
addition, sublethal concentrations of certain fungi-
cides were reported to increase trichothecene pro-
duction by Fusarium species (Magan et al., 2002).
To reduce the risks of mycotoxin contamina-
tion, transgenic inactivation using a detoxification
gene is an alternative strategy (Karlovsky, 1999).
Previously, we have isolated zhd101, a ZEN detox-
ifying gene, from a fungal strain of Clonostachys
714

rosea (Kakeya et al., 2002; Takahashi-Ando et al.,
2002). Although the molar activity of ZHD101 was
quite low and its optimal pH inadequate for living
organisms (i.e., pH = 10.5) (Takahashi-Ando et
al., 2004), the genetically modified (GM) microbes
completely converted 2 lg/mL of ZEN to a cleaved
product that is devoid of estrogenic activity
(Takahashi-Ando et al., 2005). This result suggests
the feasibility of the transgenic strategy for efficient
decontamination of ZEN in important cereal crops.
The success of the detoxification approach for
mycotoxins depends in part on the extent to which the
GM plant-produced detoxification enzyme reaches
its substrate mycotoxin. We previously showed that
egfp:zhd101, a ZEN detoxification gene fused to the
C-terminus of an enhanced green fluorescence protein
gene (egfp), is efficiently expressed in T0 leaves of rice.
In this study, the ZEN degradation activity of the
GM crops was characterized in grains using the seeds
of the T2 generation.
3¢-egfp: zhd101 (5¢-TCAAAGATGCTTC TGCG-
TAGTTTCCA-3¢). Probe-target hybrids were
detected using the DIG Nucleic Acid Detection kit
(Roche). The Southern blot of the T0 line and its T1
progenies revealed the same hybridization patterns
comprising of five bands (see Figure 1a), suggesting a
multiple-copy integration of the transgene at a single
locus in line Z54. The transgene showed a stable
pattern of integration through the three generations
investigated in this study (data not shown).
Expression of egfp:zhd101 was visually exam-
ined by monitoring the bright-green EGFP fluo-
rescence under a Leica MZFLIII fluorescence
stereomicroscope. For the observation of shoots,
roots, and flowers, filter sets GFP-Plus (460–
500 nm excitation and >510 nm barrier filter)
were used. Expression of egfp:zhd101 in the
transgenic lines could easily be confirmed by the
presence of bright-green EGFP fluorescence,
which was clearly visible in the pistil and stamen
inside the flowering spikelets (Figure 1b).

Materials, results and discussion

Transgene integration and inheritance

We previously obtained six T0 rice plants express-

ing egfp:zhd101 on the basis of the bright-green
EGFP fluorescence of regenerated shoots (Higa
et al., 2003), the intensity of which is proportional
to the ZEN detoxification activity (Takahashi-Ando
et al., 2004). The transgenic rice plants flowered
normally and produced viable seeds. In this com-
munication, we focus on description of molecular
and genetic analyses of one promising line, Z54.
The genomic DNA of T0 line Z54 and its
fluorescent T1 progenies were examined by South-
ern blot analysis. The DNA (10 lg) was digested
with PstI, whose recognition sequences are present
in the polylinker upstream of the Act1 promoter
cassette and in the coding region of zhd101
(between the BsrGI and NcoI sites) in pAct1-
egfp:zhd101 (Higa et al., 2003), separated on
an agarose gel, and transferred to a Nytran N
membrane (Schleicher and Schell, Dassel, Figure 1. Analyses of T0 and T1 rice plants transformed with
egfp:zhd101. (a) Southern blot analysis of PstI-digested geno-
Germany). As a probe, the entire coding region mic DNA from wild-type and transgenic rice plants. Z54 is
of egfp:zhd101 was labeled with digoxigenin (DIG) the name of the T0 plant, and the extended names joined by
using the PCR DIG Probe Synthesis kit hyphens represent T1 progenies derived from the T0 parents
(e.g., Z54-8 implies a T1 progeny ‘‘8’’ from the T0 parent
(Roche Applied Science GmbH, Manheim,
Z54). (b) Expression of egfp:zhd101 in flowering spikelets.
Germany) with primers 5’-egfp:zhd101 (5¢-ATG- The flowers of wild-type and T1 line Z54-9 were dissected and
GTGAGCAAGGGCGAGGAGCTGTT-3¢) and observed under an epifluorescence stereomicroscope.
 715

As demonstrated with the T0 leaves in our in wheat, barely, and maize (Goswami and Kistler,
previous report (Higa et al., 2003), individual fluo- 2004), infections of FHB pathogens are quite rare
rescent T1 lines showed bands for EGFP:ZHD101 on in rice. In addition, ZEN was reported not to
the western blots (data not shown). T1 line Z54-9 and accumulate in Fusarium-infected rice grains (Sugi-
their T2 seeds Z54-9-n (n = 1, 2, 3, . . .) were used for ura et al., 1993; Yoshida et al., 2004). In agreement
the in vitro and in planta ZEN detoxification assay, with these reports, we were not able to detect any
respectively (see below). ZEN from the rice grains infected with F. grami-
nearum. Thus, to evaluate the usefulness of the
GM grains, we exogenously added ZEN solution
Degradation of ZEN by crude enzyme fraction
to mature seeds and quantified the residual
from leaves
amount of the mycotoxin.
T2 rice seeds (Z54-9-n) were immersed in a
To evaluate the ZEN degradation activity of the
5 lg/mL ZEN solution containing 0.1% Tween-20
transgenic rice plants, crude protein extracts were
and allowed to absorb the mycotoxin for 2 days.
prepared from the leaves of T1 line Z54-9 and used
All progenies from Z54-9 (i.e., 60 T2 seeds) showed
for the in vitro detoxification assay. Rice leaves
bright-green EGFP fluorescence (examples are
(about 0.1 g FW) ground in liquid nitrogen were
shown in Figure 2b insert), suggesting that the
vigorously shaken in extraction buffer (50 mM
T1 line is homozygous for the transgene.
Tris–HCl, pH7.5, 10 mM EDTA, 0.1% Triton
The contaminated seeds were lightly tapped to
X-100 and 0.1% ß-mercaptoethanol) and the super-
remove excess ZEN solution from the surface,
natant enzyme fraction was recovered. The assay
disrupted with a commercial coffee mill, and then
was initiated by adding 89 lL of this crude protein
ground to a fine powder in a mortar in liquid
extract (protein concentration adjusted to 1 lg/lL)
nitrogen. Four ml of acetonitrile–water (21:4) was
to 11 lL of substrate solution (1 lg of ZEN in
added to the powdered samples (1 g) and the
0.1 M Tris–HCl, pH9.5) at 37 C. The enzyme
mixtures were vigorously shaken for 30 min. The
reaction was terminated by adding 10 lL of 1 N
ZEN contents in the supernatants were quantified
HCl. The samples were diluted 2-fold with meth-
using a RIDASCREENFAST Zearalenon kit
anol, filtered through a 0.2 lm Millex-LG filter
(R-Biopharm AG, Darmstadt, Germany). As
(Millipore, Billerica, MA), and quantified by high-
shown in Figure 2b, the wild-type seeds contained
performance liquid chromatography (HPLC) as
approximately 30 lg of ZEN per gram of seeds. In
described previously (Takahashi-Ando et al., 2004).
contrast, the T2 seeds contained a reduced amount
As shown in Figure 2a, ZEN degradation was not
of the mycotoxin, 16 lg ZEN per gram of seeds.
observed with the protein extract prepared from the
This implies that 14 lg of ZEN was decontami-
wild-type leaves. In contrast, the amount of ZEN in
nated in the GM seeds. Considering the previously
the reaction mixture of the transgenic protein
reported levels of mycotoxin contamination,
extract decreased significantly compared to the
namely 0.04–12 lg of ZEN per gram of cereal
control reaction with the wild-type. Approximately
grains (Luo et al., 1990; Kim et al., 1993; Park
two-thirds of the ZEN was removed from the
et al., 1996), a significantly greater amount of the
reaction mixture of Z54-9. The cleavage product of
mycotoxin initially existed in the rice grains in our
ZEN was also detected from the reaction mixture
experiment. It may thus be possible to reduce the
of the transgenic lines by the HPLC analysis (data
practical level of contaminated ZEN in the trans-
not shown). These results suggest that ZEN detox-
genic seeds regardless of the low catalytic efficiency
ification activity can be conferred to cereal crops by
and high optimal pH of the detoxification enzyme
expression of zhd101 and that this activity can be
(Takahashi-Ando et al., 2004).
transmitted to the subsequent generation.
Conclusion and perspective
Degradation of exogenously absorbed ZEN
in mature seeds Using model cereal rice plants, we have demon-
strated that zhd101 can be exploited to develop
While decreased yield and mycotoxin contamina- an efficient and cost-effective system for reducing
tion are the major problems associated with FHB the amount of ZEN in GM grains. Transgene
716

Figure 2. Detoxification of ZEN by GM cereal crops. (a) Degradation of ZEN by protein extracts from transgenic rice leaves. Pro-
tein extract was prepared from T1 line Z54-9 as described in Materials and methods. Time course of the change in the ZEN con-
centration (lg/mL) in the reaction mixture was quantified by HPLC. (b) In planta detoxification of ZEN. Wild-type and T2 rice
seeds Z54-9-n (n = 1, 2, 3. . .) were immersed in ZEN solution (5 lg/mL) for 2 days to absorb ZEN inside the grains. The seeds of
wild-type and T2 lines were observed under an epifluorescence stereomicroscope. The amount of ZEN in the rice seeds was quanti-
fied using the RIDASCREEN FAST Zearalenon kit. Statistical significance between wild-type and transgenic lines were analyzed
by Student’s t-test: *p = 0.0151.

mediated detoxification of the mycotoxin could
provide an efficient means of protecting other
cereals, especially maize, which are actually ex-
posed to serious threats of ZEN contamination.
Acknowledgements
We thank K. Mimori for excellent technical
assistance. This research was supported in part
by a grant for the ‘‘Integrated Research Program
for Functionality and Safety of Food Toward an
Establishment of Healthy Diet’’ to M.K. from
the Ministry of Agriculture, Fishery, and For-
estry of Japan (MAFF).
References
Edwards SG, Pirgozliev SR, Hare MC and Jenkinson P (2001)
Quantification of trichothecene-producing Fusarium species
in harvested grain by competitive PCR to determine efficacies
of fungicides against Fusarium head blight of winter wheat.
Appl Environ Microbiol 67: 1575–1580.
Goswami RS and Kistler HC (2004) Heading for disaster:
Fusarium graminearum on cereal crops. Mol Plant Pathol 5:
515–525.
Higa A, Kimura M, Mimori K, Ochiai-Fukuda T, Tokai T,
Takahashi-Ando N, Nishiuchi T, Igawa T, Fujimura M,
Hamamoto H, Usami R and Yamaguchi I (2003) Expression
in cereal plants of genes that inactivate Fusarium mycotoxins.
Biosci Biotechnol Biochem 67: 914–918.
Kakeya H, Takahashi-Ando N, Kimura M, Onose R, Yam-
aguchi I and Osada H (2002) Biotransformation of the my-
cotoxin, zearalenone, to a non-estrogenic compound by a
fungal strain of Clonostachys sp. Biosci Biotechnol Biochem
66: 2723–2726.
Karlovsky P (1999) Biological detoxification of fungal toxins
and its use in plant breeding, feed and food production. Nat
Toxins 7: 1–23.
Kim JC, Kang HJ, Lee DH, Lee YW and Yoshizawa T (1993)
Natural occurrence of Fusarium mycotoxins (trichothecenes
and zearalenone) in barley and corn in Korea. Appl Environ
Microbiol 59: 3798–3802.
Luo Y, Yoshizawa T and Katayama T (1990) Comparative
study on the natural occurrence of Fusarium mycotoxins
(trichothecenes and zearalenone) in corn and wheat from
high- and low-risk areas for human esophageal cancer in
China. Appl Environ Microbiol 56: 3723–3726.
Magan N, Hope R, Colleate A and Baxter ES (2002) Rela-
tionship between growth and mycotoxin production by
Fusarium species, biocides and environment. Eur J Plant
Pathol 108: 685–690.
Park JJ, Smalley EB and Chu FS (1996) Natural occurrence of
Fusarium mycotoxins in field samples from the 1992 Wis-
consin corn crop. Appl Environ Microbiol 62: 1642–1648.
Simpson DR, Weston GE, Turner JA, Jennings P and Nich-
olson P (2001) Differential control of head blight pathogens
of wheat by fungicides and consequences for mycotoxin
contamination of grain. Eur J Plant Pathol 107: 421–431.
Sugiura Y, Fukasaku K, Tanaka T, Matsui Y and Ueno Y
(1993) Fusarium poae and Fusarium crookwellense, fungi
responsible for the natural occurrence of nivalenol in Hok-
kaido. Appl Environ Microbiol 59: 3334–3338.
Takahashi-Ando N, Kimura M, Kakeya H, Osada H and
Yamaguchi I (2002) A novel lactonohydrolase responsible for
the detoxification of zearalenone: enzyme purification and
gene cloning. Biochem J 365: 1–6.
Takahashi-Ando N, Ohsato S, Shibata T, Hamamoto H,
Yamaguchi I and Kimura M (2004) Metabolism of

zearalenone by genetically modified organisms expressing the
detoxification gene from Clonostachys rosea. Appl Environ
Microbiol 70: 3239–3245.
Takahashi-Ando N, Tokai T, Hamamoto H, Yamaguchi I and
Kimura M (2005) Efficient decontamination of zearalenone, the
mycotoxin of cereal pathogen, by transgenic yeasts through the
717
expression of a synthetic lactonohydrolase gene. Appl Micro-
biol Biotechnol: DOI: 10.1007/s00253-004-1816-y in press.
Yoshida M, Nakajima T, Arai M and Suzuki F (2004) Myco-
toxin productivity of Fusarium graminearum distributed in
Western part of Japan and pathogenicity of the nivalenol-
producing strains. Jpn J Phytopathol 70: 27.