Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
1
A Wide Variety of Developmental Mutants and Agronomic Traits in Rice
Yasuo NAGATO and Jun-Ichi ITOH
Graduate School of Agricultural and Life Sciences, The University of Tokyo
1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
Introduction
Almost all agronomic traits are associated with a
particular developmental process; thus, elucidation of
the developmental mechanism underlying each important trait would contribute to a fuller understanding of
the trait and provide clues for manipulating the trait.
Although a large number of rice mutants have been accumulated over many years, developmentally important
mutants such as those affecting shoot meristem development, leaf pattern formation, plastochron length, and
embryogenesis have not been reported until recently.
Mutants associated with panicle development such as
lax panicle (lax) and leafy hull sterile (lhs) were identified several decades ago, but developmental and molecular studies of these genes began much more recently
(KOMATSU et al. 2003, JEON et al. 2000).
In our laboratory, we have accumulated a collection of developmental mutants affecting organs ranging
from embryos to flowers. During embryogenesis, the
plant body axis is established and two apical meristems
differentiate: the shoot apical meristem (SAM) and the
root apical meristem. The SAM is the source of all organs formed post-germination. We have collected and
examined mutants that show aberrant embryonic patterns or fail to differentiate one of the apical meristems
(HONG et al. 1995, SATOH et al. 1999, NAGASAKI et al.
2007, ITOH et al. 2008). We have also identified mutants
defective in SAM maintenance (ABE et al. 2008, ABE
et al. 2010).
The leaf is the major vegetative organ; thus, leaf
arrangement and growth are important determinants of
plant shape. Our attention has been mainly focused on
leaf patterning (apical–basal, central–lateral, and adaxial–abaxial) and on the pattern of leaf primordium initiation (HIBARA et al. 2009, ITOH et al. 1998, MIYOSHI et
al. 2004, KAWAKATSU et al. 2006, 2009).
The panicle and flower are the most important
organs for seed production. A number of mutants affecting these tissues were known before we started our
studies, and we have collected additional mutants affecting panicle size and floral organ identity.
In this review, we focus mainly on two classes of
mutants: those with small panicles and those associated
with leaf initiation rate (plastochron).
Aberrant panicle organization 1 regulates panicle
size and floral organ identity
We have identified several small-panicle mutants from M2 populations of cv. Taichung 65 and cv.
Kinmaze mutagenized with N-methyl-N-nitrosourea.
Among these mutants, three were allelic and caused
similar phenotypes: small panicles with reduced numbers of primary rachis branches and abnormal floral organ identities. The mutants were designated aberrant
panicle organization 1-1 (apo1-1), apo1-2, and apo1-3
(Fig. 1, IKEDA et al. 2005). Since the wild-type gene of
these mutations would suppress the size reduction of
panicle, we expected that overexpression of this gene
would result in large panicles.
In panicles harboring the strongest allele (apo11), panicle length, number of primary branches, and
number of secondary branches were decreased to 75%,
70%, and 10% of levels in the wild-type, respectively.
In addition, the primary rachis branches were shortened
relative to the wild-type. As a result, the total number of
2
Yasuo NAGATO and Jun-Ichi ITOH
Fig. 1. Phenotypes of the apo1 mutants.
(A) Panicles of the wild-type and apo1-3. (B) Wildtype flower with six stamens and one pistil. (C) apo1-1
flower with three stamens and multiple pistils.
spikelets in apo1-1 was reduced to ca. 20% of that in
the wild-type. Developmental analysis revealed that the
rachis meristem of apo1-1 at the initial stage of primary
branch formation was smaller than that of the wild-type
rachis, and the phyllotaxy of primary branches was 1/2
alternate instead of the 2/5 spiral seen in he wild-type.
In addition, the wild-type rachis meristem aborts after
producing ten or more primary branches, but the apo1
rachis meristem is precociously converted to a spikelet meristem after forming several primary branches.
A similar tendency was also observed in the primary
branches: the apical and axillary meristems of the apo1
primary branch tended to be precociously converted to
spikelet meristems, resulting in short primary branches
and almost no secondary branches. All of these phenotypes indicate that conversion of the inflorescence
meristem to a spikelet meristem is temporally accelerated in apo1 mutants.
Floral organ identities were also affected in the
apo1 mutants (Fig. 1B. C). Commonly observed phenotypes were an increased number of lodicules, formation
of mosaic organs containing both lodicule and stamen
tissue, reduction of stamen number, and an increase in
the number of pistils due to a partial loss of meristem
determinacy. These phenotypes suggest that the expression of class-C genes was downregulated in apo1.
Positional cloning revealed that APO1 encodes an
F-box protein and is orthologous to Arabidopsis UFO,
but mutations in these two genes cause somewhat different phenotypes (IKEDA et al. 2007). APO1 was found
to be expressed in the outer few cell layers of the rachis meristem and the primary branch meristems. By
performing a complementation test, we found that the
introduction of an APO1 coding region driven by a 3-kb
section of the native promoter resulted in overexpression phenotypes. On the other hand, the APO1 coding
sequence driven by a 5-kb section of the promoter restored normal phenotypes, indicating that the promoter
region contains a repressor element 3 to 5 kb upstream
of the start codon. Later, dominant Apo1 mutants were
identified that harbored a transposon inserted in the
promoter, approximately 3.5 kb upstream of the start
codon (KAWAKATSU-IKEDA 2009). Interestingly, these
dominant mutants showed enhanced expression of
APO1 and enlarged panicles. Later, it was revealed
that variation in the APO1 gene had been utilized in
practical breeding: a dominant Apo1 allele was found
in high-yielding variety Habataki (TERAO et al. 2010).
Regulation of leaf number
To date, few studies have focused on the temporal regulation of rice plant development except for the
vegetative–reproductive transition. With respect to leaf
development, temporal regulation of the leaf initiation
rate has not been a target of research; in contrast, phyllotactic regulation has fascinated researchers for many
years (SNOW and SNOW, 1931, STEEVES and SUSSEX,
1989). Elucidation of the mechanisms regulating the
rate of leaf primordium formation is agronomically
important, because it defines the number of leaves and
may restrict the production of photosynthate. The rate
of leaf initiation is called the plastochron, which is defined as the time elapsed between two successive primordial initiations. Until our studies, no mutants related to the plastochron had been identified. We identified
the mutants plastochron 1 (pla1) and pla2, which represent independent loci. Both mutants exhibited similar
phenotypes: a short plastochron (around half that of
DEVELOPMENTAL MUTANTS AND AGRONOMIC TRAITS IN RICE
Fig. 2. Phenotypes of the pla1 and pla2 mutants.
(A) Seedlings of the wild-type (left), of pla1-1
(middle), and of pla2-1 (right). The number of
leaves is greater in pla1-1 and pla2-1 than in the
wild-type. (B) Panicles of the wild-type (left) and
of pla1-1 (right), in which primary rachis branches
are converted to vegetative shoots.
the wild-type) and small leaves (Fig. 2A, ITOH et al.
1998, KAWAKATSU et al. 2006). In addition, pla mutants
showed another interesting phenotype. The reproductive phase started normally, but in the young panicles of
pla mutants, the bracts elongated abnormally, and the
primary rachis branches were immediately converted
to vegetative shoots (Fig. 2B). This means that in pla1
mutants, the vegetative program is operating simultaneously with the reproductive program. Thus, PLA1 and
PLA2 have important pleiotropic roles in rice plant development.
Positional cloning revealed that PLA1 encodes
CYP78A11, a member of the cytochrome P450 family,
and PLA2 encodes an RNA-binding protein (MIYOSHI
et al., 2004, KAWAKATSU et al. 2006). So far, the substrate of PLA1 and the target RNA of PLA2 are unknown. In situ hybridization experiments showed that
both genes are expressed in young leaf primordia but
not in the shoot meristem. This result indicates that the
PLA1 and PLA2 proteins in young leaf primordia affect the timing of new leaf primordium formation on
the opposite side of the shoot meristem. We believe
that some signal produced in leaf primordia can move
through the shoot meristem to suppress precocious new
3
Fig. 3. Model of plastochron regulation.
According to this model, leaf initiation is inhibited by the preexisting immature leaf primordium.
(Top row) In the wild-type, PLA genes (PLA1 and
PLA2) limit the rate of leaf maturation. The inhibitory effect becomes weakened as the leaf develops,
and finally allows new leaf initiation. (Bottom row)
In pla mutants, the inhibitory effect is lessened, allowing precocious development of leaves and initiation of new leaf primordia.
leaf production on the opposite side.
The expression pattern of the PLA genes suggests
that their primary functions are related to leaf development. The examination of pla alleles suggests that leaf
size is positively correlated with plastochron length;
that is, the shorter the plastochron, the smaller the leaf.
Detailed examination of leaf growth revealed that the
growth rate of pla leaves at an early stage was comparable to that of the wild-type, but the pla leaves precociously matured and stopped growing much earlier
than wild-type leaves. Thus, the primary function of
the PLA genes appears to be suppression of precocious
leaf maturation. It is widely accepted based on studies
of phyllotaxy that existing leaf primordia produce an
inhibitory signal that moves in the SAM and suppresses the initiation of new leaf primordia in the vicinity
(STEEVES and SUSSEX, 1989). Thus, it is conceivable
that precocious leaf maturation cancels the inhibitory
signal early and allows accelerated production of new
primordia (Fig. 3). In subsequent studies, we found
that overexpression of PLA genes causes enlargement
of leaves and leaf-like organs including the lemma and
palea, and thus produces enlarged seeds.
The apo1 mutant described above also showed a
short plastochron, but the length was not as reduced as in the
pla mutants. It is not yet known whether the PLA genes inter-
4
Yasuo NAGATO and Jun-Ichi ITOH
act with APO1.
Other interesting mutants
During the vegetative stage, plants undergo a juvenile-to-adult phase change. Although this phase change
has not been well studied, differences between the two
phases have been observed for a large number of traits:
shoot meristem size, leaf size, leaf midrib formation,
stem structure (node–internode differentiation, orientation of vascular bundles), and photosynthetic rate. The
mori1 mutant is unable to develop adult characteristics
and perpetually produces juvenile leaves (ASAI et al.
2002). MORI1 could be a master switch controlling
the vegetative–reproductive phase change. We recently
identified another mutant related to the vegetative phase
change, peter pan syndrome (pps), in which the juvenile
phase is prolonged. In spite of its long juvenile phase,
pps starts its reproductive stage much earlier than the
wild-type. We are considering the possibility that the
timing of the juvenile–adult phase change affects plant
development, and thus the adaptability of plants.
To date, tiller number has been studied exclusively
from the viewpoint of axillary bud development. However, a particular kind of embryo mutant may reveal a
novel mechanism for culm number regulation. Embryo
pattern mutants such as aberrant regionalization of embryo 1 (are1) frequently produce a shoot and a radicle
in the dorsal region as well as in the ventral region, resulting in two shoots and two radicles in the embryo.
Upon germination, are1 plants often develop two embryonic shoots, causing a bushy phenotype.
In this review, we have emphasized the importance
of developmental and molecular genetic study of mutants, presenting just a few examples. We hope that developmental and molecular genetic analyses will enrich
our understanding of agronomic traits and lead to novel
approaches for artificial innovation of these traits.
References
1.
ABE, M., KUROSHITA, H., UMEDA, M., ITOH, J.-I. and
NAGATO, Y. (2008) The rice FLATTENED SHOOT MER-
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
ISTEM, encoding chromatin assembly factor-1 p150
subunit, is required for meristem maintenance through
regulating cell-cycle period. Dev. Biol. 319: 384-393.
ABE, M., YOSHIKAWA, T., NOSAKA, M., SAKAKIBARA,
H., SATOH, H., NAGATO, Y. and ITOH, J.-I. (2010) WAVY
LEAF1, an ortholog of Arabidopsis HEN1, regulates
shoot development by maintaining microRNA and transacting siRNA accumulation in rice. Plant Physiol. 154:
1335-1346
ASAI, K., SATOH, N., SASAKI, H., SATOH, H. and NAGATO, Y. (2002) A rice heterochronic mutant, mori1, is defective in the juvenile-adult phase change. Development
129: 265-273.
HIBARA, K., OBARA, M., HAYASHIDA, E., ISHIMARU, T.,
ABE, M., SATOH, H., ITOH, J.-I. and NAGATO, Y. (2009)
The ADAXIALIZED LEAF1 gene functions in leaf and
embryonic pattern formation in rice. Dev. Biol. 334:
345-354.
HONG, S.-K., AOKI, T., KITANO, H., SATOH, H. and NAGATO, Y. (1995) Phenotypic diversity of 188 rice embryo
mutants. Dev. Genet. 16: 298-310.
IKEDA, K., NAGASAWA, N. and NAGATO, Y. (2005) ABERRANT PANICLE ORGANIZATION 1 gene temporally
regulates meristem identity in rice. Dev. Biol. 282: 349360.
IKEDA, K., ITO, M., NAGASAWA, N., KYOZUKA, J. and
NAGATO, Y. (2007) Rice ABERRANT PANICLE ORGANIZATION 1, encoding an F-box protein, regulates meristem fate. Plant J. 51: 1030-1040.
ITOH, J.-I., HASEGAWA, A., KITANO, H. and NAGATO, Y.
(1998) A recessive heterochronic mutation, plastochron
1, shortens the plastochron and elongates the vegetative
phase in rice. Plant Cell 10: 1511-1521.
ITOH, J.-I., HIBARA, K., SATO, Y. and NAGATO, Y. (2008)
Developmental role and auxin responsibility of class III
HD-ZIP gene family members in rice. Plant Physiol.
147: 1960-1975.
JEON, J. S., JANG, S., LEE, S., NAM, J, KIM, C., LEE, S.
H., CHUNG, Y. Y., KIM, S. K., LEE, Y. H., CHO, Y. G., and
AN, G. (2000) leafy hull sterile1 is a homeotic mutation
in a rice MADS box gene affecting rice flower development. Plant Cell 12: 871-884.
KAWAKATSU, T., MIYOSHI, K., ITOH, J.-I., KURATA, N.,
VEIT, B. and NAGATO, Y. (2006) PLASTOCHRON 2 regulates leaf initiation and maturation in rice. Plant Cell
18: 612-625.
KAWAKATSU-IKEDA, K., YASUNO, N., OIKAWA, T., IIDA,
S., NAGATO, Y., MAEKAWA, M. and KYOZUKA, J. (2009)
Expression level of ABERRANT PANICLE ORGANIZATION 1 determines the rice inflorescence form through
DEVELOPMENTAL MUTANTS AND AGRONOMIC TRAITS IN RICE
the control of cell proliferation in the meristem. Plant
Physiol 150: 736-747.
13. KOMATSU, M., MAEKAWA, M., UJIIE, S., SATAKE, Y.,
FURUTANI, L, OKAMOTO, H., SHIMAMOTO, K. and KYOZUK, J. (2003) LAX and SPA: major regulators of shoot
branching in rice. Proc. Natl. Acad. Sci. USA 100:
11765-11770.
14. MIYOSHI, K., AHN, B.-O., KAWAKATSU, T., ITOH, Y.,
ITOH, J.-I., NAGATO, Y. and KURATA, N. (2004) PLASTOCHRON 1, a time keeper of leaf initiation in rice,
encodes cytochrome P450. Proc. Natl. Acad. Sci. USA.
101: 875-880.
15. NAGASAKI, H., ITOH, J.-I., HAYASHI, K., HIBARA, K.,
SATOH-NAGASAWA, N., NOSAKA, M., MUKOUHATA, M.,
ASHIKARI, M., KITANO, K., MATSUOKA, M., NAGATO, Y.
and SATO, Y. (2007) Small interfering RNA production
pathway is required for shoot meristem initiation in rice.
5
Proc. Natl. Acad. Sci. USA. 104: 14867-14871.
16. SATOH, N., HONG, S.-K., NISHIMURA, A., MATSUOKA,
M., KITANO, H. and NAGATO, Y. (1999) Initiation of
shoot apical meristem in rice: characterization of four
SHOOTLESS genes. Development 126: 3629-3636.
17. SNOW, M., and SNOW, R. (1931). Experiments on phyllotaxy. I. The effect of isolating a primordium. Philos.
Trans. Roy. Soc. London. 25: 301-322.
18. STEEVES, T. A., and SUSSEX, I. M. (1989). Pattens in
Plant Development. 2nd ed. Cambridge Univ. Press,
Cambridge, UK.
19. TERAO, T., NAGATA, K., MORINO, K., and HIROSE, T.
(2010) A gene controlling the number of primary rachis
branches also controls the vascular bundle formation and
hence is responsible to increase the harvest index and
grain yield in rice. Theor. Appl. Genet. 120: 875-893.
6
Yasuo NAGATO and Jun-Ichi ITOH
イネの多様な発生変異体と農業形質
長戸　康郎・伊藤　純一
東京大学大学院農学生命科学研究科
〒 113-8657　東京都文京区弥生 1-1-1
ほとんど全ての農業形質は発生過程と結びつい
ていると言ってよい。発生過程の解明は，その形
質の理解を豊かにさせるだけでなく，新たな制御
方法を見いだす助けにもなる。従来，イネの変異
体は数多く蓄積されているが，変異体を利用して
重要な形質の発生過程を明らかにする研究はほと
んどなかった。我々は，農業形質に限らず個体レ
ベルの形質の遺伝的制御機構を解明する目的で，
多くの変異体の同定と解析を行ってきた。
aberrant panicle organization 1 (apo1) では穂が極
端に小さくなるとともに，花器官のアイデンティ
ティーも異常になる。 ₃ つのアリルを同定してい
るが，程度の差はあるが，互いによく似た表現型
を示す。発生過程の解析により，野生型の穂軸メ
リステムは 1 次枝梗原基を分化した後，退化する
が，apo1 変異体では退化せず小穂メリステムに
転換することが明らかになった。同様の傾向は 1
次枝梗においても見られ，1 次枝梗メリステムは
小穂メリステムに野生型より早く転換するため，
1 次枝梗が短くなる。1 次枝梗の腋芽は小穂メリ
ステムにすぐに転換するため，2 次枝梗がほとん
ど形成されない。このことは，apo1 変異体では
花序（穂軸，枝梗）メリステムから小穂メリステ
ムへの転換が時間的に早まっていることを示し
ている。APO1 遺伝子のプロモータ 3kb を用いて
apo1 変異体を形質転換したところ，枝梗が極端
に増加し，APO1 の過剰発現による表現型となっ
た。プロモータを ₅ kb にして形質転換したとこ
ろ，野生型の表現型になった。その後，プロモー
タ領域（上流 3.5Kb 付近）にトランスポゾンが挿
入した優性変異体が見いだされ，それでは発現量
が増加するとともに，穂も大きくなった。
葉を野生型の約2倍の早さで分化する
plastochron 1 (pla1), pla2 変異体を同定した。いず
れも葉を早く分化するが，葉のサイズは小さくな
り，生殖成長期では 1 次枝梗原基が栄養シュート
に転換した。ポジショナルクローニング法により
原因遺伝子を単離したところ，PLA1 は，P450 ファ
ミリーのメンバーであり，PLA2 は RNA 結合タン
パク質をコードしていた。発現パターンを in situ
hybridization 法により調べたところ，PLA1,PLA2
ともにシュートメリステムでは発現せず，葉原
基でのみ発現していた。発生過程の解析により，
pla1,pla2 の葉は急速に成熟することが明らかに
なり，PLA1,PLA2 の第 ₁ の機能は，葉の成熟の
タイミングの制御であると考えられる。pla1,pla2
での葉間期の短縮は，葉が急速に成熟するため，
葉原基分化の抑制シグナルが早く解除されるため
であると考えられる。
変異体を用いた発生過程の研究が，農業形質の
理解を深めるだけでなく，その人為的改変のヒン
トを与えることを期待している。
DEVELOPMENTAL MUTANTS AND AGRONOMIC TRAITS IN RICE
質疑応答
吉田：メリステムの大きさと穂のサイズの話のと
ころで，メリステムのサイズがある程度以上
は大きくならないというのは植物に普遍的な
話なのでしょうか。逆に，メリステムが非常
に大きくて，穂が大型化する植物は少ないの
でしょうか。
長戸：メリステムの話は，植物の細胞間の相互
作用がどれだけの細胞数でできるかというこ
とですが，大体は決まっていると思います。
組織によって多少違いがあるかもしれません
が，多分20細胞以下ぐらいでしょう。正確な
データがあるわけではないですが。small RNA
が12～13とか15細胞までは動くことができる
とか，そういうデータは出来つつあるので，
多分普遍的だろうと思います。細胞間のコミ
ュニュケ－ション能力によるのではないかと
思います。
吉田：それからもう一つ，葉間期の長さと葉の
サイズの相関は植物に普遍的な現象でしょう
か。
長戸：PLA遺伝子と葉のサイズの関係はトウモロ
コシやシロイヌナズナ でも同じなので，多分
普遍的であると思います。というか，両方と
も要するに葉を小さくしているようです。そ
の結果として葉間期が早く，短くなっている
ので，葉間期の関係は二次的な結果だと思い
ます。葉間期が短くなれば，葉のサイズは小
さくなります。
司会：幾つかの発生変異体は代謝系の遺伝子に関
連するということについて，その代謝系その
ものが重要なのか，壊れた遺伝子が別の機能
を持っている可能性が重要なのかを考えるべ
きと思います。例えば代謝系の上下の遺伝子
をノックアウトしてみて，確かに代謝系がか
かわっているのかを調べるような解析はいか
がでしょうか。
長戸：ある代謝系に対する阻害剤を野生型に与え
ると，ミュータントの変化と一部分合致しそ
うなことはあるので，全てかどうかは分から
ないけれども代謝産物が重要なことだろうと
思います。
久保山：花序のところが結局シュートになるとい
7
うのは，葉間期とどういった関係があるよう
に考えられ るのでしょうか。
長戸：pla1は実は包葉で非常に強く発現します。
本来退化するはずの包葉はpla1の発現がなく
なると成長し，成熟します。そうすると，分
裂が中途になってしまいます。多分，ブラク
トでの発現がシュートになることが非常に重
要だろうと思っています。ただし，弱いアリ
ルですと穂ができます。穂ができるというの
は，ブラクトは多少伸びるし，最初の１本か
２本の一次枝梗はシュートになるのですが，
そのあとは穂ができますし，種も作ります。
司会：要するに，juvenileの葉では，光合成活動
が非常に弱いということですが，実際に葉緑
体の構造は違うのか同じなのか，どうでしょ
うか。
長戸：見ていませんが，普通に緑ですから。聞い
た話だと，Rubiscoのアクティベースが若いと
きはしっかりと機能してないようですが，全
然確かではないです。分からないところがあ
ります。
司会：juvenile phaseからadult phaseに移行するに
あたって，多様な遺伝子発現の変化が起こる
というご指摘だったのですが，例えばアレイ
解析とかで比較されたりしたようなことはあ
りますでしょうか。
長戸：まだ実施はしていません。
草場：sem変異体というミュータントが出てきた
のですが，シュートは出るのでしょうか。
長戸：シュートは出ます。
草場：植物ホルモンは定量化されているでしょう
か。
長戸：まだ行っていませんが，普通に成長するこ
とを把握しています。
司会：アラビドシプスでのホモログ，過剰発現さ
せると器官が大きくなるということでよろし
いでしょうか。
長戸：そうなります。
司会：イネでも同じでしょうか。
長戸：イネでも同じになります。
司会：器官は大きくなるが，結局は農業形質にプ
ラスに働かないということもあるのではと思
われます。どの器官が大きくなるかにもよる
8
Yasuo NAGATO and Jun-Ichi ITOH
と思いますが，このあたりはいかがでしょう
か。
長戸：葉のような器官が大きくなるので，籾も
大きくなるということになります。極端に過
剰発現すると，多分マイナスになることもあ
るでしょう。₁コピーか2コピー増やすぐらい
だと，多分ちょうどいいのではないでしょう
か。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
9
Production of Plants that Can Tolerate Various Environmental Stresses Using
Modified Transcriptional Regulators
Masaru TAKAGI
National Institute of Advanced Industrial Science and Technology (AIST)
1-1-1 Higashi, Tsukuba, Ibaraki 305-8602, Japan
Introduction
Plants provide us not only with essential items for
survival, such as oxygen and food, but also with various items that make our life more comfortable, such
as raw materials for medicine and fragrance. Because
the “plant functions” needed to produce these products are provided by the spatiotemporal expression of
genes present in the plant genome, functional analysis
of each gene is necessary to utilize such functions more
effectively. A number of transcription factors have been
shown to act as key regulators for the control of plant
growth and development, and they directly regulate the
plant phenotype. Therefore, functional analysis of transcription factors is an important subject not only for basic research, but also for the manipulation of plant traits
dependent on their activities. Loss-of-function analysis
using gene knockout or antisense RNA techniques is
sometimes an effective strategy for functional characterization of genes. However, single-gene-knockout
lines or transgenic plants expressing antisense RNA often do not exhibit informative phenotypes because plant
genes are frequently duplicated, and this structural and
functional redundancy often interferes with efforts to
identify the function of a gene of interest.
To overcome these difficulties, we developed
a novel gene-silencing system, Chimeric REpressor
gene-Silencing Technology (CRES-T). By this method,
a transcription factor is converted into a strong repressor by fusion with the ERF-associated amphiphilic repression (EAR) motif repression domain, which suppresses the expression of target genes and is dominant
over the activation activity of endogenous and functionally redundant transcription factors (HIRATSU et al.,
2003). As a result, transgenic plants that express the
chimeric repressor exhibit phenotypes similar to lossof-function alleles of the gene for transcription factor
that had been converted into a chimeric repressor (Fig.
1). We have prepared trans genic Arabidopsis lines that
express a chimeric repressor for each of the transcription factors from the Arabidopsis genome. Using these
lines, we are analyzing the functions of transcription
factors in Arabidopsis (MITSUDA and OHME-TAKAGI,
Fig. 1. Transcription factor gene redundancy and the
CRES-T system.
Left panel illustrates the presence of functionally
redundant transcription factors (TF1 and TF2) interfering with the induction of a mutant phenotype
in a single-gene-knockout line. Right panel illustrates a chimeric repressor (TF1-SRDX) suppressing the expression of the target gene dominantly
over the activity of the endogenous and functionally redundant transcription factors, resulting in a
phenotype similar to that of loss-of-functional alleles of the transcription factor gene.
10
Masaru TAKAGI
2009). In addition, we found that the expression of a
chimeric repressor often confers tolerance to various
environmental stresses. Here, we show that the CREST system is a powerful tool for functional analysis of
redundant plant transcription factors and the possibility
of using CRES-T for improving plant stress tolerance
by genetic manipulation.
cDNA probes were then prepared and analyzed with the
Agilent Arabidopsis 2 Oligo Microarray (Agilent Technologies Inc., Palo Alto, CA, USA). All microarray experiments and analysis of data, including calculation of
the P-value, were performed according to the supplier's
manual using the feature-extraction and image-analysis
software (version A.6.1.1; Agilent Technologies Inc.).
Materials and Methods
Results and Discussion
Construction of CRES-T vectors
The protein coding regions of Arabidopsis transcription factors were amplified using appropriate
primers. Each amplified fragment was cloned into the
SmaI site of the p35SSRDXG vector to produce a chi-
The Arabidopsis genome contains around 2000
genes for transcription factors. We fused the SRDX
(LDLDLELRLGFA), a short peptide modified from
the EAR-like motif repression domain of SUPERMAN (HIRATSU et al., 2002), to the coding region of
meric repressor gene, and the region corresponding to
each transgene was transferred into the pBCKH plant
expression vector using the Gateway system (Invitrogen) (MITSUDA et al., 2005).
the gene of each transcription factor (Fig. 2a). We then
transformed the resultant chimeric repressor constructs
driven by the CaMV 35S promoter into Arabidopsis to
generate the CRES-T lines. We collected T2 seeds from
these transgenic plants to form a chimeric repressor library and saved T2 seeds of individual CRES-T lines
for examination of tolerance to various abiotic stresses.
First, we attempted to identify transgenic lines
that exhibited tolerance to salt stress. We found that
wild-type seeds would hardly germinate on MS plates
containing NaCl at more than 200 mM, and the seedlings that were produced did not thrive. We then grew
T2 seeds from Arabidopsis CRES-T lines on MS plates
supplemented with 250 mM NaCl to isolate lines that
exhibited tolerance to high concentrations of NaCl. Under this condition, almost none of the wild-type seeds
germinated. In contrast, the transgenic Arabidopsis
plants that expressed the chimeric repressor against
MYB domain transcription factor, MYB102, NAC domain transcription factor, ANAC047, and GARP family transcription factor, HSR1 (35S:AtMYB102-SRDX,
35S:ANAC047-SRDX, and 35S:HRS1-SRDX) germinated and their seedlings grew vigorously (Fig. 2b).
After transfer from the MS plates into soil, these three
CRES-T lines grew well and developed normally without any morphological defects. To examine the salt tolerance of rosette plants of these CRES-T lines, we treated 4-week-old soil-grown Arabidopsis CRES-T lines
Plant growth conditions, transformation, and isolation of stress-tolerant lines
Arabidopsis plants were grown in soil at 22°C
with 16 h of light daily. For each plant transformation,
a T-DNA vector carrying the appropriate construct
was introduced into Agrobacterium tumefaciens strain
GV3101 by electroporation, and the resultant Agrobacterium was infiltrated into wild-type Arabidopsis plants
(Col-0) by the floral-dip method (CLOUGH and BENT,
1998). Transgenic plants were selected on 30 mg/L hygromycin-containing medium. For the isolation of Arabidopsis transgenic lines tolerant to stress conditions,
plants were grown on Murashige and Skoog (MS) agarsolidified medium with 0.5% sucrose (MS plates) and
with or without the stressor being tested. To examine
the salt tolerance of rosette plants grown from selected
lines, 4-week-old soil-grown plants were treated with
400 mM NaCl solution for 3 weeks.
Isolation of RNA, microarray experiments, and expression analysis
Total RNA was isolated from rosette leaves of
2-week-old Arabidopsis plants. Cy5- and Cy3-labeled
PRODUCTION OF PLANTS TOLERANT TO ABIOTIC STRESS
with 400 mM NaCl solution for 3 weeks. The wild-type
plants were all dead after this treatment, whereas plants
of the three selected CRES-T lines remained viable and
kept green color in their leaves (Fig. 2b; MITO et al.,
2010). The T3 generation of these CRES-T lines also
tolerated salt stress, indicating that the salt-tolerance
phenotype was heritable (MITO et al., 2010).
Next, we attempted to isolate CRES-T lines that
Fig. 2. Isolation of CRES-T lines that exhibit tolerance to
salt and osmotic stresses.
(a) Schematic representation of the construct of a
chimeric repressor. 35S, TF, SRDX, and NOS indicate the CaMV 35S promoter, coding region of a
transcription factor gene, coding sequence of the
SRDX repression domain, and nos terminator, respectively.
(b) Seedlings of the CRES-T lines that germinated
in MS medium containing 225 mM NaCl (upper
panel) and their rosette plants treated with 400 mM
NaCl for 3 weeks (lower panel).
(c) Seedlings of the CRES-T lines that germinated
in MS medium containing 600 mM mannitol.
All figures are modified from MITO et al. (2010).
11
exhibited tolerance to osmotic stress using media containing high concentrations of mannitol. The germination of wild-type seeds was severely inhibited in MS
media containing 600 mM mannitol (Fig. 2c), whereas
the T2 seeds from Arabidopsis CRES-T lines that expressed the chimeric repressor aginst C2H2 zinc-finger
and AP2/ERF domain transcription factors (35S:ZAT6SRDX and 35S:AtERF5-SRDX), respectively, germinated and expanded their cotyledons. As with the salttolerant materials, the T3 generation of these CRES-T
lines tolerated osmotic stresses, similar to the T2 generation, indicating that the phenotype was heritable
(MITO et al., 2010).
Microarray analysis using RNA of 3-week-old
seedlings revealed that the expression of a number of stress-response genes was upregulated in the
35S:AtMYB102-SRDX plants compared to the wildtype. Those included the Dehydration-Responsive
Fig. 3. Upregulated genes in AtMYB102-SRDX plants.
Expression profiles of downstream genes regulated
by AtMYB102-SRDX, analyzed by qRT-PCR. Grey
and black bars indicate relative levels of expression of each gene in 3-week-old wild-type control
(WT) and AtMYB102-SRDX plants, respectively.
Treated plants were incubated on MS medium plus
200 mM NaCl for 24 h. Values for the WT without
salt treatment were set at 1. All panels are modified
from MITO et al. (2010).
12
Masaru TAKAGI
Element Binding transcription factor (DREB)-related
genes DREB1A, DREB2B, At1g22810 and At1g19210,
and several genes for zinc-finger transcription factors,
ZAT11, ZAT12, and ZAT7. RT-PCR analyses showed
that the expression of DREB1A, At1g22810, and
ZAT11 was enhanced by AtMYB102-SRDX, and salt
stress synergistically enhanced their expression (Fig.
3). Interestingly, the expression of At1g19210 was not
induced in the wild-type under either normal or saltstress conditions, but its expression was upregulated in
35S:AtMYB102-SRDX plants under both conditions. It
is likely that those upregulated genes play critical roles
in tolerance to salt stress in AtMYB102-SRDX plants
(MITO et al., 2010).
Because AtMYB102-SRDX acts as a transcriptional repressor, the upregulated genes in
35S:AtMYB102-SRDX plants are unlikely to be direct
targets of AtMYB102-SRDX. These results indicate
that the plant genome may contain negative regulators that suppress the expression of the genes related
to stress responses, and that AtMYB102 may activate
these negative regulators, thus maintaining the suppression of the stress response. By applying the chimeric
AtMYB102-derived artificial repressor, AtMYB102SRDX, the expression of the negative regulators was
suppressed and the downstream genes related to stress
response were upregulated. This finding suggests that,
under normal conditions, plants may suppress genes
that confer abiotic tolerance. Our results strongly suggest that negative regulators play an important role in
the acquisition of stress tolerance in plants, possibly
by repressing expression of genes that are detrimental
to fitness under non-stress conditions. As shown in the
case of ectopic expression of DREB genes, overexpression of such genes might have adverse effects on plant
development (GILMOUR et al., 2000).
Our results show that application of the CRES-T
system is an effective method to manipulate plant traits.
Furthermore, the chimeric repressors identified in Arabidopsis may be applicable to various crop and horticulture species (TANAKA et al., 2011).
Summary
Transcription factors are effective tools for the
manipulation of plant traits because a number of transcription factors act as master regulators of various
phenotypes and usually regulate a group of genes. We
show here that the expression of a chimeric repressor,
in which a transcriptional activator is converted into
a repressor by fusion with the EAR-motif repression
domain, confers tolerance to various abiotic stresses
in transgenic Arabidopsis. The chimeric repressors
derived from the AtMYB102, ANAC047, and HRS1
transcription factors induced significant tolerance to
high concentrations of NaCl. In addition, the transgenic
Arabidopsis lines that expressed the chimeric repressors derived from ZAT6 and AtERF5 were tolerant to
osmotic stress and grew in MS medium containing 600
mM mannitol. Expression profiling by microarray analysis revealed that the expression of DREB1A, DREB2B,
and several genes for ZAT transcription factors rose 10to 100-fold in the transgenic plants expressing the gene
for the chimeric AtMYB102-derived repressor. We
conclude that the expression of a chimeric repressor
provides an effective strategy for enhancing the tolerance of plants to abiotic stresses.
Acknowledgments
A part of this work was supported by a grant-inaid from the Research and Development Program for
New Bio-industry Initiatives at the Bio-oriented Technology Research Advancement Institutions.
References
1. G
ILMOUR, S.J., SEBOLT, A.M., SALAZAR, M.P., EVERARD, J.D., and THOMASHOW, M.F. (2000) Overexpression of the Arabidopsis CBF3 transcriptional activator
mimics multiple biochemical changes associated with
cold acclimation. Plant Physiol. 124: 1854-1865.
2. HIRATSU, K., OHTA, M., MATSUI, K., and OHME-TAKAGI, M. (2002) The SUPERMAN protein is an active repressor whose carboxy-terminal repression domain is
required for the development of normal flowers. FEBS
PRODUCTION OF PLANTS TOLERANT TO ABIOTIC STRESS
Lett. 514: 351-354.
3. H
IRATSU, K., MATSUI, K., KOYAMA, T., and OHME-TAKAGI, M. (2003) Dominant repression of target genes by
chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant J. 34: 733-739.
4. LIU, Q., KASUGA, M., SAKUMA, Y., ABE, H., MIURA, S.,
YAMAGUCHI-SHINOZAKI, K. and SHINOZAKI, K. (1998)
Two transcription factors, DREB1 and DREB2, with an
EREBP⁄AP2 DNA binding domain separate two cellular signal transduction pathways in drought- and lowtemperature-responsive gene expression, respectively, in
Arabidopsis. Plant Cell 10: 1391-1406.
5. MITO T., SEKI, M,, SHINOZAKI, K., OHME-TAKAGI, M.,
and MATSUI, K. (2010) Generation of chimeric repressors that confer salt tolerance in Arabidopsis and rice.
13
Plant Biotechnol. J. (in press).
6. MITSUDA, N., and OHME-TAKAGI, M. (2009) Functional
analysis of transcription factors in Arabidopsis. Plant
Cell Physiol. 50: 1232-1248.
7. MITSUDA, N., SEKI, M, SHINOZAKI, K, and OHME-TAKAGI, M. (2005) The NAC transcription factors NST1 and
NST2 of Arabidopsis regulate secondary wall thickenings and are required for anther dehiscence. Plant Cell
17: 2993-3006.
8. TANAKA, Y, YAMAMURA, T, OSHIMA, Y, MITSUDA, N,
KOYAMA, T, OHME-TAKAGI, M., and TERAKAWA, T.
(2011) Creating ruffled flower petals in Cyclamen persicum by expression of the chimeric cyclamen TCP repressor. Plant Biotechnology. 28: 141-147.
14
Masaru TAKAGI
植物転写因子機能を利用した環境ストレス耐性植物作出の試み
高　木　優
独立行政法人産業技術総合研究所
〒305−8562　茨城県つくば市東1−1−1
植物では転写レベルの制御が，遺伝子発現制御
に中心的な役割を果たしていることが知られてい
る。また，一つの転写因子が，複数の標的遺伝子
の発現を制御すること，および，数多くの転写因
子が形質のマスター因子として働くことが報告さ
れていることから，植物機能の改変に転写因子を
利用することは，有効な手段であると考えられて
いる。ところが，植物の転写因子遺伝子は，大き
なファミリーを形成し，重複遺伝子が数多く存在
することから，遺伝子破壊や相補的なRNA導入
等の従来の方法では，転写因子の機能を十分活用
できないことが判ってきた。そこで，我々は，標
的遺伝子の発現を内在性の転写因子に優先して抑
制するキメラリプレッサーを用いることによって
（CRES-T法)，様々な環境ストレスに対して耐性
を持つ形質転換体が作出できることを明らかに
した。これまでにAtMYB102，ANAC047および
HRS1転写因子に対するキメラリプレッサーを発
現させたシロイヌナズナ形質転換体では，高濃
度（400mM）の塩に対して耐性を示し，ZAT6あ
るいはAtERF5転写因子に対するキメラリプレッ
サーを発現する植物では，600mMマンニトール
が含まれた培地においても発芽し，強い浸透圧耐
性が付与されることを明らかにした。興味ある事
に，強い塩耐性を示したAtMYB102キメラリプレ
ッサー発現体では，ストレス関連遺伝子である
DREB1A，DREB2Bおよび ZAT遺伝子の発現が10
～100倍上昇していることが判った。これらのこ
とからキメラリプレッサーを用いることは，スト
レス耐性植物の作出や，多様な形質の改変に有効
な手段であると考えられる。
PRODUCTION OF PLANTS TOLERANT TO ABIOTIC STRESS
質疑応答
草場：あるCRES-TラインでDREBがオーバーエ
クスプレスしているというお話でしたが，植物
体は矮化しなかったのですね。そのメカニズム
について，どういうことを考えてらっしゃるの
でしょうか。
高木：メカニズムも非常に難しく，推察なのです
が，このDREBをやるナチュラルな状態で発現
させているからじゃないでしょうか。キメラリ
プレッサーの場合に，矮化などの不健康な植物
はとれにくい。ナチュラルな状態で発現し，仮
の遺伝子がコンディションしているからじゃな
いかというふうに自分では考えています。この
キメラリプレッサーのいいところは，ネイティ
ブプロモーターでドライブできるところです。
どんなプロモーターでもドライブできるという
ところが強みなんです。ある組織と特異的なプ
ロモーターでいろんなキメラリプレッサーを発
現させて，人工的に植物を作ることができる
し，逆にネイティブなプロモーターでリプレッ
サーを発現させることによって，もともとのバ
イオロジカルファンクションにもっと非常に近
づけられるというふうに考えます。何かどうい
うことかというのは難しいなと思っています。
遺伝子一個の問題だけじゃなく，マルチなやつ
がかなり同時に関与していて，そういう相乗効
果があるのかもしれないですね。
中川：最後のリグニンのところ，非常に面白いと
思ったのですけれど，多分リグニンの合成系と
いうのはCADだとか三つか四つの合成遺伝子
が使われていると思います。今回の場合は，リ
グニンができるときに確かに合成系が壊れてリ
15
グニンができないというケースと，リグニンは
できているけれども，構造上非常に消化しやす
いリグニンになっているというのがあるようで
すが，これはどういうふうに考えたら良いでし
ょうか。
高木：このNSTの場合はリグニンは作られていま
せん。リグニンの合成系の遺伝子は軒並み全
部抑制されています。それでNSTというのは，
こういう二次木部を作るというスイッチのアイ
デンティティーの遺伝子だったらしいんです。
だから，セルロースも作らない。セルロースと
リグニンの両方を作らない。その上のほうでス
イッチをオフにしているというふうに考えてい
ます。逆にもっと下流でも遺伝子が働いている
はずですから，リグニンだけを作るもの，セル
ロースだけを制御するもの，とすればリグニン
を抑えながら，セルロース量というのをもっと
増やせるのではないかと考えて下流をやったの
ですが，面白いことにリグニンとセルロースの
マスターレギュレーターって，下でトップドラ
イブしているんですよ。必ず一つ抑制すれば，
片一方も抑制するということが分かってきまし
た。
西村：篠崎先生のグループでDREBのマイクロア
レイ解析をやっていると思いますが，今回とら
れたマイクロアレイデータや下流因子など，予
想されているものはオーバーラップしていたの
でしょうか。
高木：かなりオーバーラップしていますね。DREB
が制御している遺伝子も同様にDREBが上がっ
てました。矮性の話については，全くそれは説
明できないですね。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
17
Analysis and Utilization of a Cleistogamous Mutant of Rice
Hitoshi YOSHIDA
National Institute of Crop Science, National Agriculture and Food Research Organization
2-1-18 Kannondai, Tsukuba, Ibaraki 305-8518, Japan
Introduction
Until more information about the environmental
and agronomic impact of genetically modified (GM)
crops is available, gene containment is necessary to
prevent the spread of transgenes into the environment
(DANIELL, 2002). Cultivated rice in Japan is essentially
autogamous, with an outcrossing frequency of less than
1%, but its flowers still open at anthesis and release pollen grains into the environment. Therefore, the possibility of outcrossing with wild relatives or non-GM rice has
raised public concern regarding the environmental impact of GM rice. Several approaches have been proposed
to minimize or eliminate gene flow through pollen, but
no practical method has yet been developed in rice.
One possible approach, cleistogamy (self-fertilization without flower opening), would minimize the
risk of transgene spread. In rice, cleistogamous mutants
and transgenic plants have been reported; however,
there have been no commercially practical cleistogamous rice lines identified or developed so far. In collaboration with the University of Tokyo and Kyushu
University, we identified a novel cleistogamous mutant
with a missense mutation in the floral homeotic gene
SPW1, and we found that the mutation does not affect
other agronomic traits (YOSHIDA et al., 2007). In this
chapter, we describe the molecular and physiological
characterization of this mutant and discuss strategies
for its utilization.
Mechanistic view of rice flower opening: the important role of the lodicule
To develop and utilize cleistogamous rice, we first
need to understand the mechanism of flower opening in
rice, and then to inhibit the flower-opening machinery
without affecting other physiological processes. The
flower of rice, also known as the floret, comprises one
lemma, one palea, two lodicules, six stamens, and one
pistil (Fig. 1).
Fig. 1. Structure of the rice spikelet.
(Left panel) Floral organs of rice. The front halves
of the lemma and palea have been removed.
(Right panel) Schematic representation of a transverse section of the rice floret.
Each stamen further consists of a filament and an
anther, in which more than 1000 pollen grains are produced. The spreading of the tips of the lemma and palea
is called flower opening or anthesis. The driving force
behind flower opening is generated by the lodicule, the
least conspicuous floral organ in rice. The lodicule is a
small and roundish scale-like organ located inside the
flower at the base of the lemma. The lodicules start to
swell immediately before flower opening and push the
lemma outward to make the flower open (HOSHIKAWA,
1989). Based on comparative observations of its location in the flower, surface cell structure, and expression
patterns of floral genes, the lodicule is regarded as a homolog of the dicot petal (KYOZUKA and SHIMAMOTO,
2002). Its appearance is quite different from that of the
18
Hitoshi YOSHIDA
beautiful petals in roses or cherry blossoms; however,
its role in flower opening is the same. The lodicule possesses a number of vascular bundles that allow water
uptake, which causes the lodicule to swell. At the same
time, the filament begins to elongate and lift the anther toward the top of the flower. Immediately before
the onset of flower opening, the anthers dehisce and
disperse the pollen grains onto the stigmas within the
same floret. After dehiscence, the filaments continue
to elongate and move the anthers outside the glumes
(i.e., the lemma and palea). Finally, the lodicules stop
swelling and begin to shrink, closing the glumes. Once
the lodicules shrink, they lose the ability to swell, thus
the flower never opens again. At the sametime, filament
elongation stops, and the nearly empty anthers are left
outside the closed glumes. Therefore, it is possible to
evaluate the frequency of flower opening by counting
numbers of florets with anthers left outside the glumes.
As a method to produce fertilized seeds while at
the same time preventing pollen dispersal into the environment, cleistogamy would be an efficient strategy, at
least in theory. It is unclear whether creating a practical
type of cleistogamous rice is really possible, but several reports provide useful information relevant to this
question. For example, it has long been known that rice
flowers do not open on rainy days, but the anthers dehisce inside the closed flowers and disperse the pollen
grains onto the stigma, self-fertilizing the ovule (HOSHIKAWA, 1989). Therefore, rice appears to have the potential to become cleistogamous. There have also been
several studies on cleistogamous rice mutants such as
d7; however, these mutants have not been of practical use because they also have undesirable agronomic
characters such as dwarfism (NAGAO and TAKAHASHI,
1963). On the other hand, cleistogamous varieties and
landraces are already available in other gramineous
species such as barley and wheat. Cleistogamous barley varieties have smaller lodicules than those of noncleistogamous varieties; therefore, it is suggested that
their lodicules are less able to push the lemma outward
and cause flower opening (TURUSPEKOV et al., 2004).
These observations suggest that it would be possible to
make practical cleistogamous rice if the morphology or
identity of the lodicule could be modified.
One of the possible strategies to create cleistogamous rice would be to transform the identity of
the lodicule into that of another floral organ. B-class
MADS-box genes are involved in the specification of
lodicules and stamens in grass species (AMBROSE et al.,
2000; NAGASAWA et al., 2003). In Arabidopsis, B-class
MADS-box proteins comprise two major subfamilies,
APETALA3 (AP3) and PISTILLATA (PI). Members
of these two subfamilies combine to form a heterodimer (AP3-PI) that exerts B-class activity and specifies
petal and stamen identity (JACK et al., 1992; GOTO and
MEYEROWITZ, 1994; RIECHMANN et al., 1996). Rice
has one AP3 ortholog, SUPERWOMAN1/MADS16
(SPW1), and two PI orthologs, MADS2 and MADS4
(KANG et al., 1998; NAGASAWA et al., 2003; PRASAD
and VIJAYRAGHAVAN, 2003; YADAV et al., 2007; YOSHIDA et al., 2007; YAO et al., 2008). Therefore, it would
be theoretically possible to create cleistogamy by manipulating expression of these B-class genes.
Identification of spw1-cls, a practical cleistogamous
mutant in rice
We performed screening for cleistogamous mutants at the experimental farm of the University of Tokyo, using a mutated population of ssp. japonica cv.
Taichung 65 generated by the research group at Kyushu
University. From an M2 population induced by MNU
(N-methyl-N-nitrosourea) treatment, we selected several lines that segregated for plants in which anthers
were not observed outside the glumes at the flowering
stage. Further examination revealed that most of these
lines had underdeveloped stamens and were male-sterile. However among these lines, we identified one line
in which the cleistogamy segregated as a single-gene
recessive trait, and we tentatively named the mutant
cleistogamy (cls) (YOSHIDA et al., 2007). As described
above, the stamens of wild-type flowers elongated out
of the glumes upon flower opening and remained outside of the glumes for several days. In contrast, cls mutant flowers did not open and no stamens were observed
outside the glumes (Fig. 2). Consequently, the pollen
grains of this mutant were theoretically not dispersed
ANALYSIS AND UTILIZATION OF A CLEISTOGAMOUS MUTANT OF RICE
Fig. 2. Panicle phenotype of the cleistogamous rice mutant
spw1-cls.
Panicles of the wild-type (left) and the spw1-cls
mutant (right). Stamens were not observed outside
the glumes of the mutant.
into the environment. However, it will be necessary to
confirm whether pollen grains that do not land where
they can fertilize an ovule survive long enough to fall
from the flower and enter the environment. When this
mutation was introduced into “Kasalath” (an indica
variety), the cleistogamous trait was expressed stably
despite the differences in genetic background between
Kasalath and Taichung 65.
The cls mutant would not be of practical use if the
causal gene caused pleiotropic effects on other agronomic traits, as in the case of d7. We also wondered if
the stamens remaining inside the glumes might cause
a malformed grain shape. Therefore, we cultivated this
mutant at the experimental farms at Hokuriku Research
Center (Jo-etsu, Niigata, Japan) and at the University
of Tokyo (Tanashi, Tokyo, Japan), and examined the
agronomic traits and grain shape of the cls mutant.
The heading date, plant height, culm number, panicle
length, flower number, seed fertility, grain weight, and
grain shape of cls did not differ significantly from those
of the wild-type (YOSHIDA et al., 2007). Thus, cls appears to affect specifically the mechanism of flower
opening and is therefore a promising allele for practical
gene containment in rice.
When we examined the florets of cls in detail, the
stamens and pistil of cls were normal, but its lodicules
19
Fig. 3. Structure of the inner floral organs in the cleistogamous rice mutant spw1-cls.
The lemma and palea of each flower have been removed. Left, wild-type; right, spw1-cls mutant. In
spw1-cls, the stamens are normal but the lodicules
(arrows) are elongated.
were elongated compared to those of the wild-type (Fig.
3). The number of vascular bundles in the lodicules of
cls was also reduced compared to wild-type lodicules. A
series of morphological observations suggested that the
cls lodicules partially acquire the identity of the glume;
therefore, the cleistogamy of cls appears to result from
a partial homeotic conversion of lodicules into glumes.
This defect seems to inactivate lodicule swelling and
thus prevent flower opening.
Molecular characterization of the spw1-cls mutation
The cls mutation was mapped to the long arm of
chromosome 6 using molecular markers. This region
contains the SPW1 gene, which (as described above) is
a B-class MADS-box gene involved in the specification
of lodicule and stamen identity. Two recessive alleles
of SPW1, spw1-1 and spw1-2, both cause the homeotic
transformation of lodicules into glume-like organs and
of stamens into carpels. Therefore, the flowers of spw11and spw1-2 do not open and are sterile (NAGASAWA et
al., 2003). The flowers of cls somewhat resemble spw11 and spw1-2 flowers, in that their lodicules become
elongated like glumes. Examination of the nucleotide
sequence of the SPW1 gene in cls revealed a single
base change leading to an amino acid change (isoleucine-45 to threonine; I45T) in the MADS-box domain.
A complementation test further revealed that the cleistogamous phenotype of cls could be rescued to that of
20
Hitoshi YOSHIDA
the wild-type by introducing a genomic DNA fragment
of the wild-type SPW1 (gSPW1wt), whereas introduction of the same genomic region containing the I45T
mutation (gSPW1I45T) failed to complement the cleistogamous phenotype. Therefore, whereas spw1-1 and
spw1-2 are thought to be severe null alleles, cls is a
novel weak allele of SPW1 and was thus renamed as
superwoman1-cleistogamy (spw1-cls; YOSHIDA et al.,
2007). In SPW1 RNAi lines, a wide range of phenotypes from Spw– (the phenotype of spw1-1 and spw1-2)
to Cls– (the phenotype of spw1-cls) was observed. In
addition, introduction of gSPW1I45T into spw1-1 phenocopied the Cls– phenotype. These results suggest that
lodicule and stamen development have differential sensitivity to the level of B-class protein activity. They further suggest the possibility of producing cleistogamous
plants by engineering one or more B-class floral organ
identity genes and thus transforming lodicule identity.
As in Arabidopsis, the proper functioning of SPW1
in rice would require heterodimerization with two PI
orthologs (MADS2 and MADS4). The 45th amino acid
of SPW1 is located in the b-strand of the MADS domain and is thought to be involved in the dimerization
of MADS proteins. The MADS-box proteins possess
a conserved cluster of small hydrophobic amino acids
(I, L, V, A, or F) at or around amino acid 45, whereas
spw1-cls has a hydrophilic amino acid (T) in this position. In a yeast two-hybrid system, the wild-type SPW1
strongly interacted with both MADS2 and MADS4,
suggesting formation of two types of heterodimers: SPW1WT-MADS2 and SPW1WT-MADS4. However, SPW1I45T showed much weaker interaction with MADS2
than did wild-type SPW1. SPW1I45T interacted even
less with MADS4: almost no interaction was detected,
suggesting the critical role of a hydrophobic amino
acid at this position for the dimerization of MADS-box
proteins. Therefore, the SPW1I45T protein is defective
in dimerization with the counterpart B-class proteins
(YOSHIDA et al., 2007; YAO et al., 2008).
In situ hybridization analysis revealed that MADS2
was expressed in both lodicule and stamen primordia,
whereas MADS4 was expressed in stamen primordia
but not in lodicule primordia (YOSHIDA et al., 2007),
suggesting that the SPW1-MADS2 heterodimer is the
one that is important for lodicule formation. In contrast,
both the SPW1-MADS2 and SPW1-MADS4 heterodimers are involved in stamen development. Thus, it is
presumed that the low level of the SPW1I45T-MADS2
heterodimer present in spw1-cls is insufficient to allow
the establishment of complete lodicule identity, but sufficient for the specification of stamen identity, because
SPW1I45T failed to interact with MADS4, which is normally expressed in stamen primordia.
Taken together, the lodicule-specific phenotype of
spw1-cls is likely caused by a combination of differential requirements of B-class protein activity in lodicule and stamen, differential expression of MADS2 and
MADS4 in lodicule and stamen, and differential interaction affinity of SPW1I45T for MADS2 and MADS4.
Toward the effective utilization of cleistogamy in
rice
The occurrence of outcrossing in rice, even at a low
frequency, has caused concern about transgene contamination from GM rice in both feral rice (crop-to-wild)
and cultivated rice (crop-to-crop). Therefore, gene containment technologies are necessary in rice to suppress
gene flow and enable the coexistence of GM rice with
non-GM rice or wild relatives in both experimental and
commercial cultivation. As described above, spw1-cls
is a promising genetic resource for developing a practical gene containment technology in rice. With that goal
in mind, we have been introducing the spw1-cls mutation into several Japanese cultivars using DNA markers to make near-isogenic lines (OHMORI and YOSHIDA,
unpublished data). These lines can be used not only for
gene containment in GM rice, but also for other situations in which outcrossing is unfavorable. For example,
it would be useful for suppressing outcrossing between
purple/red rice cultivars or cultivars with modified levels of specific seed contents and normal cultivars. It
would be also useful for maintenance of purity during
foundation seed production.
However, because spw1-cls is a weak mutation
based on a reduced level of protein–protein interaction,
it is likely to be a temperature-sensitive mutant. Indeed,
ANALYSIS AND UTILIZATION OF A CLEISTOGAMOUS MUTANT OF RICE
yeast two-hybrid experiments revealed that low temperature caused an increase in the binding activity between
SPW1I45T and MADS2 (YOSHIDA et al., 2007). These
results suggest that when the temperature at the booting
stage is cool, increased B-class heterodimer formation
in the spw1-cls mutant may lead to partial recovery of
lodicule morphology, resulting in flowers that open at
the usual flowering time. Therefore, we have been examining the stability of the cleistogamy in spw1-cls for
several years in several regions of Japan with a range
of environmental conditions (from Hokkaido to Okinawa). In collaboration with many laboratories in Japan,
we found that spw1-cls flowers open at a certain rate in
the northern area of Japan (i.e., in Hokkaido and part
of the Tohoku region; Ohmori et al., unpublished data).
Further investigation should allow us to define detailed
conditions to regulate flower opening/non-opening of
spw1-cls. By applying such knowledge, guidelines for
use of this mutation in different environments and in
various genetic backgrounds can be established.
Because of the potential instability of spw1-cls,
screening for more stable cleistogamous mutants is
required. In addition, molecular biological techniques
that artificially induce cleistogamy should be pursued
and could be useful in cereal crops other than rice.
Recently, several novel findings on molecular genetic
mechanisms regulating lodicule development have
been published, such as identification of the barley
cleistogamy gene Cly1/HvAP2 (NAIR et al., 2010) and
the rice MFO1 gene, which regulates floral morphology, including lodicule development (OHMORI et al.,
2009). Such basic studies would facilitate our understanding of lodicule development and enhance the development of cleistogamous cereal crops.
Cleistogamy is a novel character that has not yet
been used in commercial rice but would provide a number of benefits. It is anticipated that further studies will
lead to various agronomic applications of this novel
and intriguing character.
Acknowledgements
The author thanks Dr. Yasuo Nagato (the Univer-
21
sity of Tokyo), Dr. Hikaru Satoh (Kyushu University),
and many staff members of the National Agricultural
Research Center (Hokuriku Research Center), University of Tokyo and Kyushu University. The author also
thanks the many rice researchers who kindly assisted us
with cultivation and evaluation of spw1-cls in various
regions of Japan. This work was supported by grants
from the Ministry of Agriculture, Forestry and Fisheries (MAFF) of Japan, and by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture,
Sports, Science and Technology (MEXT) of Japan.
References
1. AMBROSE, B.A., LERNER, D.R., CICERI, P., PADILLA,
C.M., YANOFSKY, M.F., and SCHMIDT, R.J. (2000). Molecular and genetic analyses of the silky1 gene reveal
conservation in floral organ specification between eudicots and monocots. Mol Cell 5: 569-579.
2.DANIELL, H. (2002). Molecular strategies for gene containment in transgenic crops. Nature Biotechnol 20: 581586.
3.GOTO, K., and MEYEROWITZ, E.M. (1994). Function and
regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes Dev. 8: 1548-1560.
4.HOSHIKAWA, K. (1989). The Growing Rice Plant. (Tokyo: Nosan Gyoson Bunka Kyokai (Nobunkyo)), p. 310.
5. JACK, T., BROCKMAN, L.L., and MEYEROWITZ, E.M.
(1992). The homeotic gene APETALA3 of Arabidopsis
thaliana encodes a MADS box and is expressed in petals
and stamens. Cell 68: 683-697.
6. KANG, H.G., JEON, J.S., LEE, S., and AN, G. (1998).
Identification of class B and class C floral organ identity
genes from rice plants. Plant Mol Biol. 38: 1021-1029.
7. KYOZUKA, J., and SHIMAMOTO, K. (2002). Ectopic expression of OsMADS3, a rice ortholog of AGAMOUS,
caused a homeotic transformation of lodicules to stamens in transgenic rice plants. Plant Cell Physiol. 43:
130-135.
8. NAGAO, S., and TAKAHASHI, M. (1963). Trial construction of twelve linkage groups in Japanese rice. J Fac Agric Hokkaido Univ. 53: 72-130.
9. NAGASAWA, N., MIYOSHI, M., SANO, Y., SATOH, H., HIRANO, H., SAKAI, H., and NAGATO, Y. (2003). SUPERWOMAN1 and DROOPING LEAF genes control floral
organ identity in rice. Development 130: 705-718.
10. NAIR, S.K., WANG, N., TURUSPEKOV, Y., POURKHEIRAN-
22
Hitoshi YOSHIDA
DISH,
M., SINSUWONGWAT, S., CHEN, G., SAMERI, M.,
TAGIRI, A., HONDA, I., WATANABE, Y., KANAMORI, H.,
WICKER, T., STEIN, N., NAGAMURA, Y., MATSUMOTO, T.,
and KOMATSUDA, T. (2010). Cleistogamous flowering in
barley arises from the suppression of microRNA-guided
HvAP2 mRNA cleavage. Proc Natl Acad Sci U S A 107:
490-495.
11. OHMORI, S., KIMIZU, M., SUGITA, M., MIYAO, A., HIROCHIKA, H., UCHIDA, E., NAGATO, Y., and YOSHIDA, H.
(2009). MOSAIC FLORAL ORGANS1, an AGL6-like
MADS box gene, regulates floral organ identity and
meristem fate in rice. Plant Cell 21: 3008-3025.
12. PRASAD, K., and VIJAYRAGHAVAN, U. (2003). Doublestranded RNA interference of a rice PI/GLO paralog,
OsMADS2, uncovers its second-whorl-specific function
in floral organ patterning. Genetics 165: 2301-2305.
13. RIECHMANN, J.L., KRIZEK, B.A., and MEYEROWITZ, E.M.
(1996). Dimerization specificity of Arabidopsis MADS
domain homeotic proteins APETALA1, APETALA3,
14.
15.
16.
17.
PISTILLATA, and AGAMOUS. Proc Natl Acad Sci U S
A 93: 4793-4798.
TURUSPEKOV, Y., MANO, Y., HONDA, I., KAWADA, N.,
WATANABE, Y., and KOMATSUDA, T. (2004). Identification and mapping of cleistogamy genes in barley. Theor
Appl Genet. 109: 480-487.
YADAV, S.R., PRASAD, K., and VIJAYRAGHAVAN, U.
(2007). Divergent regulatory OsMADS2 functions control size, shape and differentiation of the highly derived
rice floret second-whorl organ. Genetics 176: 283-294.
YAO, S.G., OHMORI, S., KIMIZU, M., and YOSHIDA, H.
(2008). Unequal genetic redundancy of rice PISTILLATA orthologs, OsMADS2 and OsMADS4, in lodicule and
stamen development. Plant Cell Physiol. 49: 853-857.
YOSHIDA, H., ITOH, J., OHMORI, S., MIYOSHI, K.,
HORIGOME, A., UCHIDA, E., KIMIZU, M., MATSUMURA,
Y., KUSABA, M., SATOH, H., and NAGATO, Y. (2007).
Superwoman1-cleistogamy, a hopeful allele for gene
containment in GM rice. Plant Biotechnol J. 5: 835-846.
ANALYSIS AND UTILIZATION OF A CLEISTOGAMOUS MUTANT OF RICE
23
閉花受粉性イネ突然変異体の解析とその利用
吉　田　均
農業・食品産業技術総合研究機構　作物研究所
〒 305-8518　茨城県つくば市観音台 2-1-18
日本で栽培されているイネの大部分は自殖性が
高く，花（小花）が開くときにおしべが小花の外
まで伸び出て花粉を飛散させるため，低頻度では
あるものの，自然交雑が起きる。花粉飛散による
遺伝子組換えイネと一般イネとの自然交雑を抑制
するための技術の一つとして，閉花受粉性（開花
せずに受粉・稔実する性質）の利用が有効と考え
られる。本稿では，筆者らが発見した閉花受粉性
イネ突然変異体の特徴，原因遺伝子の作用機構，
さらに今後の利用に向けた取り組みなどについて
紹介する。
イネの小花は，外穎と内穎， 2 つの鱗被， ₆ 本
の雄蕊， ₁ 本の雌蕊，によって構成される。この
うち，開花を引き起こす原動力となるのが「鱗被」
である。鱗被は外穎の基部にある丸く小さな器官
であるが，開花の直前から急激に膨らみ始め，外
穎を外側に押し出すため，開花が起きる。開花時
に花の外に抽出した葯は，穎が閉じた後にも外部
に取り残される。一度閉じた穎は二度と開かない
ため，穎にはさまれた葯の有無によって，開花の
有無を判断することができる。鱗被の形態だけを
変化させることができれば，イネにおいても実用
的な閉花受粉性を付与することが可能であると
考えられたため，鱗被の形態形成に着目し，「台
中 65 号」の突然変異体集団を用いて閉花受粉性
突然変異体のスクリーニングを行った。まず，開
花期に穎の外に葯が出ていないものを数系統選抜
したところ，多くのものは不稔であったものの，
花を開かずに正常に稔実するものを ₁ 系統見出
し，「閉花受粉性」を意味する「cleistogamy」（略
称：cls）と名付けた。出穂日，草丈，穂数，穂長，
₁ 穂粒数，稔実率，粒重，粒の形状や外観などの
農業形質について，cls と原品種の農業特性には，
顕著な差がなく，cls は実用的な閉花受粉性イネ
を開発するための有望な遺伝資源と考えられた。
cls の花では，雄蕊や雌蕊には変化がないが，鱗
被が平らで細長い穎状の器官に変化し，内部の維
管束数も減少していた。こうした変化により鱗被
が膨潤できず，開花しなくなったものと考えられ
る。
閉花受粉性の原因は，SUPERWOMAN1(SPW1)
遺伝子のアミノ酸置換変異であったため，この
閉花受粉性イネを superwoman1-cleistogamy(spw1cls) と呼ぶことにした。SPW1 は鱗被と雄蕊の形
作りに関わる遺伝子群の発現を制御する MADS
ボックス型転写因子をコードしており，spw1-1，
spw1-2 などの機能欠失型アリルでは，鱗被が穎
状器官に変化するだけでなく，雄蕊も雌蕊へと変
化し，完全不稔となる。同じ遺伝子の変異が原
因であるにもかかわらず，なぜ spw1-cls では雄
蕊が正常に形成され，正常に稔実するのだろう
か？転写因子である SPW1 タンパク質は，DNA
結合ドメインである MADS ドメインなどを介し
てヘテロ二量体を形成し，標的 DNA に結合す
る。酵母ツーハイブリッドアッセイの結果などか
ら，cls 型 SPW1 タンパク質では，MADS ドメイ
ン内の保存性の高い第 45 位アミノ酸が疎水性か
ら親水性に変異しているため，二量体形成能が低
下し，転写因子としての機能が低下するものと
考えられた。また，SPW1 のパートナータンパク
質をコードする 2 つの遺伝子のうち，OsMADS2
は鱗被と雄蕊の両方の形成に関与するのに対し，
OsMADS4 は主に雄蕊形成に関与するため，これ
らパートナー遺伝子の発現パターンおよび，cls
型 SPW1 タンパク質との相互作用能の違いなど
によって，鱗被にだけ形態変化を生じるものと考
24
Hitoshi YOSHIDA
えられた。
spw1-cls 変異体は実用的な閉花受粉性遺伝資源
として期待されるが，アミノ酸置換による弱い変
異が原因であるため，穂の形成時期に冷涼となる
地域では温度感受性を示し，野生型と同様に開花
する可能性が考えられた。実際，日本国内の環境
の異なる地域で栽培を行い，閉花受粉性の安定性
を検討したところ，北海道や東北などでは spw1cls は一定の割合で開花することが明らかとなり，
spw1-cls を安定的に利用するためには，開花を引
き起こす詳細な条件などの知見を積み重ねていく
必要があると考えられた。また，新規閉花受粉
性突然変異体の探索や，人為的に閉花受粉性イネ
を作出する技術を開発していくことも重要であろ
う。鱗被形成機構に関わる知見は急速に蓄積して
きており，閉花受粉性研究の加速が期待される。
一方では，spw1-cls 変異体をさまざまな実用品
種と交配し，閉花受粉性準同質遺伝子系統の育成
を進めている。こうした系統は，遺伝子組換えイ
ネの母本としてだけでなく，紫黒米や赤米と言っ
た有色素米品種，特定の成分含量を変化させた機
能性品種，あるいは原種や原原種段階で種子の純
度維持が強く求められる品種など，花粉飛散によ
る交雑に対して注意が必要とされる場面への応用
も考えられる。イネにおいて，閉花受粉性はこれ
までになかった新しい形質である。今後，さらな
る研究の進展により，多様な場面において閉花受
粉性イネが利用されていくことを期待したい。
ANALYSIS AND UTILIZATION OF A CLEISTOGAMOUS MUTANT OF RICE
質疑応答
司会：オオムギの閉花性のミュータントは，メカ
ニズムとしては同じものだとお考えでしょう
か。
吉田：かなり違っていて，AP2というクラスAに
分類されているんですが，全然別の構造を持
つ遺伝子です。これのミューテーションだと
いうことが分かっています。このAP2の遺伝子
の３′側にmicro-RNAのターゲットサイトが
あります。ここにミューテーションが起きる
と，micro-RNAの標的にならなくなって，発現
パターンが変わって，その結果，鱗被が小さく
なって閉花受粉性になるというふうに報告され
ています。ですから，イネでもこういうホモロ
グがあると思いますので，そういったものの利
用なども今後考えてもいいのではないかと思い
ます。
司会：アミノ酸置換ということで温度感受性のこ
とが問題になっていますが，このパターンの場
合にはどういったことが生じるのでしょう。
吉田：温度のことは特に報告はされていません。
どこでも閉花受粉性になりますよというふうに
は伺っています。日本のあちこちで栽培したと
きにどうかとか，世界中でどうなのか，そうい
25
う厳密なデータは私はまだお聞きしていないの
で，興味深いところです。
高木：ほかの植物でこのような閉花受粉性という
のはありますか。
吉田：閉花受粉というのは，いろいろあって，ダ
イズも基本的に閉花受粉と言われてます。多少
開いて，花粉を飛ばしたりするのもあります
し，スミレとか暑くなると閉花受粉するとかい
うのもあります。いろいろな植物で，閉花受粉
というのは環境適応として利用されているのか
なと思います。
草場：イネは閉花受粉すると稔実率が落ちたりし
ないわけですね。そうすると，イネは開花して
花を咲かせる必要は特にないわけで，それは
昔，自家不和合性だった時代の名残と考えれば
いいのですか。
吉田：そこまで厳密には分からないですけれど
も，インド型は大きく開花するとか，野生種は
さらに程度が大きいかなという気がします。交
雑したがるというか，そういう習性の名残があ
ると思います。栽培品種ではそういう必要がな
いので，必ずしも開花する必要はないのかもし
れないなとも思います。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
27
The Rice esp2 Mutant Accumulates Protein Aggregates in the Endosperm and
Has Better Qualities for Rice Bread
Yasushi KAWAGOE
National Institute of Agrobiological Sciences
2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan
Introduction
Rice (Oryza sativa) is one of the most widely grown
crops in the world. Rice has been consumed mostly as
cooked rice. In recent years, however, breadlike products using rice flour have attracted significant attention
as a way to increase rice consumption in Japan. In addition, rice bread is considered a desirable substitute for
regular bread made from wheat flour because it is less
allergenic (YANO 2010). Although the demand for rice
bread, and hence rice flour, is expected to increase in
the near future, there are several limitations to the use
of rice flour as a substitute for wheat flour. The main
limitation is that the seed storage molecules, mainly
starch and proteins, are markedly different between
the two cereals. In particular, the rice genome does
not contain homologs of the wheat genes for the high
molecular weight (HMW) subunits of glutenin, which
strongly affect the rheological properties of the dough
(BRANLARD et al. 2001). In addition, the compositions
and physicochemical properties of the storage proteins
of rice (glutelins and prolamins) are significantly different from those of wheat (Fig. 1). Furthermore, the
starch granule types are different between the two cereals: wheat granules are classified as the simple type,
whereas rice granules are classified as the compound
type (Fig. 2). We recently reported that the synthesis
of compound-type starch granules and amyloplast division are closely related processes that share common
protein factors (YUN and KAWAGOE 2009; 2010).
Low-molecular-weight thiols such as glutathione (γ-glutamylcysteinylglycine) are known to affect
the rheological properties of wheat dough (HAHN and
GROSCH 1998; JOYE et al. 2009). Glutathione is widely
distributed in most cells and acts as an antioxidant in
the protection of sulfhydryl groups. YANO (2010) found
that the addition of glutathione to rice batter improves
its gas-retaining properties, which thus significantly increases the volume of the bread. This outstanding advance in rice bread making has shown the importance
of redox (oxidation/reduction) properties not only for
dough made from wheat flour but also for dough containing rice flour.
Fig. 1. Seed proteins. Proteins were extracted from
wheat and rice flour purchased from a local supermarket in extraction buffer consisting of 50mM
Tris-HCl (pH 6.8), 8 M urea, 4% SDS, 10% glycerol, and 5% 2-mercaptoethanol. The proteins
were separated on an SDS-polyacrylamide gel
(5–20% gradient gel) and stained with Coomassie
Brilliant Blue. Lane 1, wheat; lane 2, rice. Note
that the wheat proteins are larger and much more
heterogeneous in size.
28
Yasushi KAWAGOE
Fig. 2. Starch granules. Starch granules were purified
from wheat and rice flour purchased from a local
supermarket. Purified granules were dried and observed with a scanning electron microscope. Bars
= 10μm. Note that the wheat granules (simple
type) are rounded: granules larger than 10μm are
called the A type and smaller ones are called the B
type. The rice starch granules (compound type) are
polygonal, sharp-edged, and much smaller than
the wheat starch granules.
We have been studying the molecular mechanisms of redox regulation of proteins using the rice
endosperm as a model system. We found that how the
disulfide bond is formed in the endoplasmic reticulum
(ER) plays a significant role in protein sorting to protein bodies (PB) in the rice endosperm (KAWAGOE et al.
2005), and that the ER membrane-localized oxidoreductase Ero1 is required for the disulfide bond formation of proglutelins in the ER (ONDA et al. 2009). We
have recently revealed that the protein disulfide isomerase (PDI) family of proteins, PDIL1;1 (also called
Esp2) and PDIL2;3, play distinct roles in the oxidation
of storage proteins and PB development in the rice endosperm (ONDA et al. 2011).
The esp2 mutant, a PDIL1;1-knockout mutant,
was initially identified as a mutant that accumulates
large amounts of proglutelins and develops unusual
PBs (TAKEMOTO et al. 2002). In vitro analysis with recombinant Esp2 (PDIL1;1) and PDIL2;3 showed that
Esp2 facilitates the oxidative folding of reduced, denatured α-globulin and ribonuclease, whereas PDIL2;3
facilitates intermolecular disulfide bond formation
when α- globulin mutated at the conserved second
cysteine residue in the CCxQL motif is used as the substrate (ONDA et al. 2011). Consistent with these notable
differences in the redox activities between Esp2 and
PDIL2;3 in vitro, PDIL2;3 is significantly upregulated
Fig. 3. Schematic representation of disulfide bonds. In
the wild-type rice endosperm, proglutelins form
intramolecular disulfide bonds assisted by Esp2
(PDIL1;1), whereas in the esp2 mutant, proglutelins form aggregates through intermolecular
disulfide bonds assisted by PDIL2;3 (ONDA et al.
2011). It is most likely that Ero1, which is upregulated in the esp2 endosperm, is involved in the oxidation reactions (ONDA et al. 2009).
in the endosperm of the esp2 mutant, and proglutelins
form large protein aggregates through intermolecular
disulfide bonds (Fig. 3, ONDA et al. 2009).
The distinct characteristics of esp2 flour are well
suited for bread making
The apparently unique redox properties of storage
proteins in the esp2 mutant prompted us to test the
flour from this mutant for bread-making quality. Because esp2 flour contains protein aggregates formed
through intermolecular disulfide bonds, it may form a
unique type of gluten-like structure or protein network
when mixed with wheat flour and water. In addition,
PDIL2;3, which is significantly upregulated in the esp2
mutant (ONDA et al. 2009), may effectively facilitate
intermolecular disulfide bond formation during mixing
and fermentation. Our initial experiments on a small
scale consistently demonstrated that the esp2 flour exhibited better dough qualities than wild-type rice flour
when each was mixed with wheat flour. We then asked
a professional baker, an expert in making rice bread,
to evaluate the bread-making quality of the esp2 flour
compared to wild-type flour. As shown in Fig. 4, the
bread made from wild-type rice flour (80%) and wheat
gluten (20%) wrinkled while it cooled after baking,
which is a major problem encountered with bread con-
RICE ESP2 MUTANT FOR BREAD MAKING
Wild type
29
esp2
References
1.
Fig. 4. Rice bread. Rice flour (80%) from either wildtype or esp2 mutant grain was mixed with wheat
gluten (20%), and the mixed flours were evaluated
for their bread-making qualities. The bread made
with wild-type rice flour wrinkled considerably
while it cooled after baking (left panel), whereas
the bread made with esp2 flour (right panel) kept
its expanded shape.
taining a large amount of rice flour. In contrast, the
bread made from esp2 flour (80%) and wheat gluten
(20%) wrinkled much less. These results are consistent
with the idea that protein networks formed through
intermolecular disulfide bonds significantly improve
bread quality, not only for regular wheat bread but also
for rice bread. It is likely that some rice mutants that
have been considered inadequate for cooked rice may
indeed contain better flour qualities for bread making.
It is my hope that rice breeding programs in the future
include targets of improving flour qualities for bread
making.
Acknowledgements
This work was supported by the Research and Development Program for New Bio-Industry Initiatives
from the Bio-Oriented Technology Research Advanced
Institution and by a grant from the Ministry of Agriculture, Forestry and Fisheries of Japan (Genomics for
Agricultural Innovation, IPG-0023). I sincerely thank
Mr. Akio WATANABE of Namisato, Inc., Sano, Tochigi,
for making rice bread using wild-type and esp2 flour.
BRANLARD, G., DARDEVET, M., SACCOMANO, R., LAGOUTTE, F. and GOURDON, J. (2001) Genetic diversity
of wheat storage proteins and bread wheat quality. Euphytica 119: 59-67.
2. HAHN, B., and GROSCH, W. (1998) Distribution of glutathione in Osborne fractions as affected by additions of
ascorbic acid, reduced and oxidized glutathione. J. Cereal Sci. 27: 117-125.
3. JOYE, I.J., LAGRAIN, B. and DELCOUR, J.A. (2009) Endogenous redox agents and enzymes that affect protein
network formation during breadmaking – A review. J.
Cereal Sci. 50: 1-10.
4. KAWAGOE, Y., SUZUKI, K., TASAKI, M., YASUDA, H., AKAGI, K., KATOH, E., NISHIZAWA, N.K., OGAWA, M., and
TAKAIWA, F. (2005) The critical role of disulfide bond
formation in protein sorting in the endosperm of rice.
Plant Cell 17: 1141-1153.
5. ONDA, Y., KUMAMARU, T., and KAWAGOE, Y. (2009) ER
membrane-localized oxidoreductase Ero1 is required for
disulfide bond formation in the rice endosperm. Proc.
Natl. Acad. Sci. USA 106: 14156-14161.
6. ONDA, Y., NAGAMINE, A., SAKURAI, M., KUMAMARU, T.,
OGAWA, M., and KAWAGOE, Y. (2011) Distinct roles of
protein disulfide isomerase and P5 sulfhydryl oxidoreductases in multiple pathways for oxidation of structurally diverse storage proteins in rice. Plant Cell 23: 210223.
7. TAKEMOTO, Y., COUGHLAN, S.J., OKITA, T.W., SATOH,
H., OGAWA, M., and KUMAMARU, T. (2002) The rice mutant esp2 greatly accumulates the glutelin precursor and
deletes the protein disulfide isomerase. Plant Physiol.
128: 1212-1222.
8. YANO, H. (2010) Improvements in the bread-making
quality of gluten-free rice batter by glutathione. J. Agric.
Food Chem. 58: 7949-7954.
9. YUN, M.-S., and KAWAGOE, Y. (2009) Amyloplast division progresses simultaneously at multiple sites in the
endosperm of rice. Plant Cell Physiol. 50: 1617-1626.
10. YUN, M.-S. and KAWAGOE, Y. (2010) Septum formation
in amyloplasts produces compound granules in the rice
endosperm and is regulated by plastid division proteins.
Plant Cell Physiol. 51: 1469-1479.
30
Yasushi KAWAGOE
米粉パンに適した有用突然変異～蛋白質変異体の利用
川　越　靖
農業生物資源研究所
〒 305-8602 茨城県つくば市観音台 2-1-2
小麦粉の代替原材料として米粉の利用促進は，
輸入小麦への依存を低下させ，我が国の食料自給
率を 10 年間で 50％に引き上げる目標達成のため
の有効な手段の一つとして期待されている。米粉
と小麦粉を食材として比較したとき最も大きく異
なる点は，小麦粉からはグルテンが形成されるの
に対し，米粉からはグルテンが形成されないこと
である。イネの種子はコムギ高分子量グルテニン
に類似した貯蔵蛋白質を含まず，グルテリン，プ
ロラミン，αグロブリンが主な貯蔵蛋白質である。
細胞イメージング技術を用いた実験で，貯蔵蛋白
質のジスルフィド架橋の形成様式は分別輸送に重
要であることが明らかになった。胚乳細胞での蛋
白質のジスルフィド架橋形成の分子機構は不明な
点が多いため，ジスルフィド架橋形成に関与する
酵素の解析を行った。PDI はジスルフィド架橋の
形成（酸化），還元，つなぎ換え（異性化）反応
を触媒する。蛋白質ジスルフィドイソメラーゼ
（PDIL1;1）欠損変異体 esp2 はグルテリン前駆体
を過剰に蓄積する。加えて，esp2 のグルテリン
前駆体は分子間ジスルフィド架橋による複合体を
形成して蓄積する。種子蛋白質の組成（グルテリ
ン前駆体やプロラミンの含量），及びジスルフィ
ド架橋の結合状態の違い（分子内架橋または分子
間架橋とその程度など）は，米粉の特性に少なか
らず影響を及ぼすと推察し，この作業仮説の検証
実験を行った。米粉とコムギグルテンからパン生
地を作製したところ，esp2 の生地はまとまりが
良く，作業性が良いことが明らかになった。加え
て，esp2 の生地は膨らんだ後の凹みが少ないので，
生地の可塑性が高いことが明らかになった。これ
らの結果から，種子蛋白質の組成，及びジスルフィ
ド架橋の結合状態の違いは，製パン特性に大きな
影響を及ぼすことが明らかになった。今後，米粉
用に適したイネの育種が活発になることが期待さ
れる。
RICE ESP2 MUTANT FOR BREAD MAKING
質疑応答
濱田：ジスルフィド結合がその後のタンパク質局
在に重要ということですが，ジスルフィドを
作る場所は同じ小胞体であって，グルテリン
だけでなく，プロラミンにもジスルフィド結
合は存在します。ジスルフィド結合の局在は
どういうメカニズムで起こっているのでしょ
うか。
川越：私たちの持っているデータを基に考えます
と，PDIL2-3 というのは PB-I の表面に局在
してプロラミンに作用し，PDIL1-1 は ER 内
腔に局在してグルテリンやアルファグロブリ
ンのフォールディングを助けていると考えら
れます。これらの PDI の局在が小胞体の中で
違うので，プロラミンは PB-I に集積し、グ
ルテリン，アルファグロブリンは ER 内腔で
フォールディングして，そこからゴルジ体に
輸送される，そのように考えています。それ
だけではメカニズム全体を説明できないと思
いますが，PDI の局在の違いは大きな要因で
はないかと考えています。
濱田：関連して，esp2 の高重合グルテリン複合
体には，プロラミンと融合しているものもあ
るのでしょうか。
川越：あると思います。
濱田：製パン性についてですが，グルテリンが高
分子化したのが，膨らんだ一つの原因になっ
ている感じです。粉質米は恐らくデータにも
あったように，すごく粒度が小さくなって，
多分恐らく損傷でんぷんも少ないと思います。
製パン性に大きくかかわってくるのは，むし
31
ろそういったデンプン特性のほうに理由があ
るような気もします。esp2 は粉質性ですが，
製パン特性の向上はその粉質性が寄与してい
るのではないでしょうか。例えば，esp2 をわ
ざと還元して分子化した状態でパンを作って
みたら，本当に膨らみがどうなるのか，そう
いったデータはありませんでしょうか。
川越：そのデータはありませんけれども，最近，
食品総合研究所の矢野さんがグルタチオンを
生地に入れると大きく膨らむことを示しまし
た。これは普通に考えると，グルタチオンが
タンパク質を還元したのではないかと考えら
れますから，還元剤を入れると米粉生地は大
きく膨らむということになります。この現象
を私も理論的に説明できないのですが、一つ
の可能性としては、デンプンを分解する酵素
の中には還元されて活性化されるものがあり
ますので，グルタチオンを入れることによっ
て，そのようなデンプン分解酵素が活性化さ
れ，デンプンの特性が変化して生地が膨らむ
ようになったと考えることもできると思いま
す。今のご質問と直接関係があるか分かりま
せんが，グルテンのようなタンパク質のネッ
トワークを還元することで生地が膨らむよう
になると考えるのは難しいと思います。米粉
の場合には、タンパク質に加えてデンプンも
生地特性に大きく影響することが分かってい
ます。それでどういった変異体を使っていけ
ばいいのかというのは，これから探索してい
くことではないかと思います。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
33
Mutation Breeding with Ion Beams and Gamma Rays
Hiroyasu YAMAGUCHI
National Institute of Floricultural Science
2-1 Fujimoto, Tsukuba, Ibaraki 305-8519, Japan
Introduction
Mutation breeding is a useful method for crop improvement. The type of mutagenic treatment and the
treatment methods used are important factors in obtaining successful results from mutation breeding.
Although mutagenic treatment induces mutations,
it also causes treatment damage to the plant. KONZAK
et al. (1965) suggested that the usefulness of any mutagen in plant breeding depends on not only the mutation
induction effect but also on the plant damage caused
by the treatment should be considered. In this study,
the effects of ion beam and gamma ray dose rates are
described in terms of obtaining useful mutants with low
plant damage by the irradiation treatment.
To obtain mutants efficiently, a wide mutated sector is desirable. The present study shows that the width
of the mutated sector was greater when produced with
ion beam irradiation than that with gamma ray irradiation. In addition, I introduce the results and discuss the
factors that caused the differences between ion beam
and gamma ray irradiation.
Comparison of ion beam and gamma ray irradiation
Ion beams have high linear energy transfer (LET),
and thus have greater biological effects compared with
low LET radiation such as gamma rays and X-rays
(YANG and TOBIAS 1979, TANAKA 1999). Therefore,
ion beams are expected to produce higher mutation frequencies and the number of novel mutants. Mutation
induction using ion beams and a variety of plants have
been attempted since the 1990s in Japan. The charac-
teristics of ion beams have been gradually clarified,
and ion beam irradiation has evolved as a new mutation
method. As a result, new varieties have been released
in chrysanthemum (NAGATOMI et al. 2003, UENO et
al. 2005), carnation, verbena etc. However, few studies have compared the mutagenic characteristics of ion
beams in terms of the mutagen for mutation breeding.
1) Efficiency
To compare the mutagenic effects of different mutagens, the terms effectiveness and efficiency are used
(KONZAK et al. 1965, MIKAELSEN et al. 1971, NILAN
et al. 1965). Effectiveness is defined as the number of
mutations produced per unit dose of radiation, whereas
efficiency is defined as the ratio of specific, desirable
mutagenic changes to plant damage. KONZAK et al.
(1965) suggested that the usefulness of any mutagen in
plant breeding depends not only on its mutagenic effectiveness but also on its mutagenic efficiency.
In the present study, we compared the efficiency
of ion beams and gamma rays using rice (Oryza sativa
L.) and chrysanthemum (Chrysanthemum morifolium).
Rice seeds were irradiated with 220 MeV (LET =
107 keV/µm) and 320 MeV carbon ions (76 keV/µm)
and with 100 MeV helium ions (9 keV/µm), all generated with an azimuthally varying field cyclotron (Japan Atomic Energy Agency, Takasaki, Japan) and with
gamma rays in a gamma room (Institute of Radiation
Breeding, National Institute of Agrobiological Sciences, Hitachi-omiya, Japan). Survival and fertility were
examined in the M1 generation. The frequency of the
chlorophyll mutations were examined in the M2 generation using the M1-plant progeny method.
34
Hiroyasu YAMAGUCHI
320 MeV carbon ion beams and gamma rays was lower
(1.1%). Thus, based on fertility, the mutation frequency
induced by ion beams equaled or exceeded that induced
by gamma rays. Moreover, based on lethality, ion beams
induced more mutations than gamma rays within the
relationship between mutation frequency and lethality.
Thus, it was suggested that the efficiency of ion beam
irradiation is equal to or greater than that of gamma ray
irradiation in rice (YAMAGUCHI et al. 2009a).
In an irradiation experiment of leaf explants of
chrysanthemum carried out using the same ion beams,
the effects of irradiation treatments were investigated
using mutations in flower color and nuclear DNA content as indices of radiation damage. Efficiency was
determined as the relationship between the frequency
of flower color mutation and the reduction in nuclear
DNA content, as shown in Fig. 2. The efficiency differed according to the type of ion beam used. The efficiency was lower with 100 MeV helium ions than with
gamma rays (YAMAGUCHI et al., 2010).
Fig. 1. Relationship between fertility and frequency of
chlorophyll mutation with ion beam and gamma
ray irradiations in rice.
The mutation frequency is determined as the number of chlorophyll mutants divided by the number of M2 plants investigated using the M1-plant
progeny method. 220 MeV carbon ion beam, r =
−0.743**; 320 MeV carbon ion beam, r = −0.644*;
100 MeV helium ion beam, r = −0.911**; gamma
rays, r = −0.587. * and ** Significant at 5 and 1%
levels, respectively.
To evaluate the efficiency based on fertility, the relationship between the mutation frequency per M2 plant
and fertility was investigated (Fig. 1). Fertility and mutation frequency per M2 plant showed a negative linear
relationship for each type of radiation. The mutation
frequency of ion beams increased significantly with decreasing fertility, and the same tendency was observed
for gamma rays. Mutation frequencies at 60% fertility
using 220 MeV carbon ion beams and 100 MeV helium ion beams were 1.4%; the frequency found using
Fig. 2. Relationship between the reduction in nuclear DNA
content and the frequency of flower color mutation
with ion beam and gamma ray irradiations in chrysanthemum.
Relative nuclear DNA content is expressed as the
ratio of the nuclear DNA content of the investigated
plants divided by that of the control plants maintained by vegetative propagation. ●: 220 MeV
carbon ion beam, r = −0.997*; ○: 320 MeV carbon ion beam, r = −0.875; ▲: 100 MeV helium ion
beam, r = −0.995; △: gamma rays, r = −0.996. *
Significant at 5% level.
35
MUTATION BREEDING WITH ION BEAMS AND GAMMA RAYS
In conclusion, the efficiencies of ion beams appear to be higher than those of gamma rays; however,
some types of ion beams (e.g., 100 MeV helium ions)
produced lower efficiencies than those produced by
gamma rays.
2) Mutated sector size
To obtain mutants through the mutagenic treatment of seeds or buds, the seed or bud organs must arise
from the mutated cells. The possibility that the flowers
or buds are not formed within the mutated sector rises
if the mutated sector is narrow, even if the mutation occurs within the cells of the apical shoot meristem. Thus,
if the mutated sector is narrow, the mutant cannot be
isolated. Therefore, a wide mutated sector is desirable.
The results of segregation frequency of chlorophyll mutants in the M2 generation of rice showed that
wider mutated sectors were produced by ion beams
compared with gamma rays. With 220 MeV carbon
ions, 320 MeV carbon ions, 100 MeV helium ions,
and gamma rays, the strain showed a higher average
segregation frequency with irradiation at a higher dose.
The segregation frequencies produced by ion beams
were higher than those produced by gamma rays; the
segregation frequencies for the chlorophyll mutants
induced by gamma rays did not exceed 0.17. In contrast, segregation frequencies for mutants induced by
ion beams were approximately 0.20. Thus, the segregation frequencies of ion beams were higher than those
of gamma rays, suggesting that wider mutated sectors
were produced by ion beams compared with gamma
rays (YAMAGUCHI et al. 2009a).
In addition, expansion of the mutated sectors
through the layers (LI, LII, and LIII) was found only
with ion beam irradiation treatment to the lateral buds
of chrysanthemum. Axillary buds were irradiated with
220 MeV carbon ions at 2 Gy, 100 MeV helium ions at
10 Gy, and gamma rays at 80 Gy, all of which had similar effects on survival. The lower five nodes of shoots
elongated from irradiated buds were cut individually
and new shoots were allowed to grow from the axillary
buds. This procedure was repeated twice, producing
a maximum of 25 plants derived from each irradiated
bud, and the color of the flowers from those plants was
investigated. There were the stains of which all progenies were mutated both in ion beams and gamma rays.
This result shows that the cells of the LI layer of all
progenies from one bud originated from a single mutated cell, suggesting the possibility that the LI, LII, and
LIII layers of the mutant plants were produced from
only one mutated cell, that is, solid mutant. Therefore,
the chimeric structure was analyzed by comparing the
flower color of the mutants to that of plants regenerated
from the roots of mutants. In all ten strains induced by
gamma rays, the flower color of the plant derived from
the root was different from that of the mutant, indicating that all the flower color mutants induced by gamma
rays were periclinal chimeras (Table 1). In contrast,
some mutants obtained with ion beams had the same
flower color as plants derived from the roots. This suggests that they were solid mutants, where both LI and
LIII layers were derived from the same mutated cell
Table 1 Comparison of the flower color in plants derived from the roots of mutants with that of mutants induced by ion beam and gamma ray irradiation to the buds of chrysanthemum.
Radiation
Carbon ion
Helium ion
Helium ion
Gamma rays
(1)
(2)
Dose
No. of strains
No. of same
(Gy)
investigated
colored strains
2
10
5
80
9
9
7
10
5
1
3
0
No. of different colored strains
Pink(1)
Mutated(2)
4
7
4
8
0
1
0
2
The same flower color as the original cultivar “Taihei.”
The flower color differs not only from the flower color of the mutant but also from the original
cultivar “Taihei.”
36
Hiroyasu YAMAGUCHI
(YAMAGUCHI et al. 2009b).
Because the mutated sectors produced by ion
beams were wider than those produced by gamma rays
in rice, it was thought that there were fewer initial cells
in the seeds after irradiation with ion beams than with
gamma rays, whereas it appeared that solid mutants in
chrysanthemum were produced from sectoral chimeras
formed from only one or a few cells. Thus, it is supposed that there is difference between ion beams and
gamma rays with regard to their effectiveness of altering only a few cells. Ion beams induce DNA damage
in a limited region in the nucleus (YANG and TOBIAS
1979). Thus, there is a large difference in radiation
damage caused by ion beams between cells constructing the apical meristem, with only a few cells surviving.
In contrast, gamma rays tended to produce lesions that
were distributed uniformly throughout the cells (YANG
and TOBIAS 1979), and there was little difference in radiation damage between the cells, and the possibility of
survival of only a few cells was low.
Effects of dose and dose rate of gamma ray irradiation on mutation induction and nuclear DNA content in chrysanthemum
The dose rate of gamma rays influences the radiation damage and also affects somatic mutation. However, to the best of our knowledge, there is no data
available regarding the effect of total radiation dose and
dose rate on mutation induction and radiation damage.
Information on the effect of dose rate on radiation damage and mutation induction and the interaction between
dose and dose rate is useful for radiation breeding, especially in radiation breeding of vegetatively propagated
crops, because the obtained mutants would be directly
used as new cultivars. Thus, I investigated the effect of
total radiation dose and dose rate on flower color mutation and nuclear DNA content in chrysanthemum.
C. morifolium cv. “Taihei” plants grown by in vitro
culture were gamma-irradiated with a total dose of 15,
30, or 60 Gy at a rate of 0.5, 1, or 2 Gy/h in the gamma
room. Leaf explants cut from the irradiated plants were
cultured, and the frequency of flower color mutation
and the nuclear DNA content were investigated.
The frequency of flower color mutation increased
significantly when the total dose increased from 15 Gy
to 30 Gy at each dose rate, although similar differences
were not observed when the dose was increased from
30 Gy to 60 Gy. On the other hand, the frequency of
mutation did not differ significantly among dose rates
at each total dose. These results show that mutation frequency was independent of dose rate and was mainly
dependent on the total radiation dose.
Comparison of the average nuclear DNA content
with each treatment revealed that the nuclear DNA content was influenced by both dose rate and total dose and
that the reduction in nuclear DNA content was less at
low dose rates, even when the total dose was high.
The relationship between the reduction of nuclear
DNA content and the mutation frequency at each dose
rate is shown in Fig. 3. The dose rate of gamma rays
influenced the nuclear DNA content but did not affect
mutation frequency. The same mutation frequencies
were obtained without large reductions in nuclear DNA
content by 0.5 Gy/h compared with 2 Gy/h. Consequently, I conclude that gamma ray irradiations with
high total doses at low dose rates efficiently induce mu-
Fig. 3. Relationship between the reduction in nuclear DNA
content and the frequency of flower color mutation
with gamma ray irradiation at different dose rates
in chrysanthemum.
Relative nuclear DNA content is expressed as the
ratio of the nuclear DNA content of the investigated
plants divided by that of the control plants maintained by vegetative propagation. ●: 2 Gy/h, r =
−0.999*; ▲: 1 Gy/h, r = −0.679; ▼: 0.5 Gy/h, r =
−0.898. * Significant at 5 % level.
MUTATION BREEDING WITH ION BEAMS AND GAMMA RAYS
tations with less radiation damage in chrysanthemum
(YAMAGUCHI et al. 2008).
In chronic radiation with a gamma field, dose rates
are lower and the exposure period is generally longer;
therefore, the total dose would be higher than that used
in the present study. However, the results from the present study should be applicable to chronic radiation exposure in a gamma field and support the usefulness of
chronic radiation.
Acknowledgments
This study was supported, in part, by the Budget for
Nuclear Research of the Ministry of Education, Culture, Sports, Science and Technology, based on screening and counseling by the Atomic Energy Commission.
References
1.
2.
3.
KONZAK, C. F., NILAN, R. A., WAGNER, J., and FOSTER,
R. J. (1965). Efficient chemical mutagenesis. In: The
Use of Induced Mutations in Plant Breeding. Report of
the FAO/IAEA technical meeting organized by the food
and agriculture organization of the United Nations and
the International Atomic Energy Agency in cooperation
with the European Association for Research on Plant
Breeding, Rome, Italy, 25 May 1964, Pergamon Press,
Oxford, pp 49-70.
MIKAELSEN, K., KARUNAKARAN, K., and KISS, I. S.
(1971). Mutagenic effectiveness and efficiency of gamma rays, fast neutrons and ethyl methane sulphonate
in rice. In: Rice breeding with induced mutations. III.
Report of an FAO/IAEA research co-ordination meeting
on the use of induced mutations in rice breeding, India,
September 1969. Technical Reports Series No. 131. International Atomic Energy Agency, Vienna, pp 91-96.
NAGATOMI, S., WATANABE, H., TANAKA, A., YAMAGUCHI, H., DEGI, K., and MORISHITA, T. (2003). Six mutant
37
varieties induced by ion beams in chrysanthemum. Institute of Radiation Breeding, Technical News No. 65.
4. NILAN, R. A., KONZAK, C. F., WAGNER, J., and LEGAULT,
R. R. (1965). Effectiveness and efficiency of radiations
for inducing genetic and cytogenetic changes. In: The
Use of Induced Mutations in Plant Breeding. Report of
the FAO/IAEA technical meeting organized by the Food
And Agriculture Organization of the United Nations and
the International Atomic Energy Agency in cooperation
with the European Association for Research on Plant
Breeding, Rome, Italy, 25 May 1964, Pergamon Press,
Oxford, pp 71-89.
5. TANAKA, A. (1999). Mutation induction by ion beams in
Arabidopsis. Gamma Field Symp. 38: 19-27.
6. UENO, K., SHIRAO, T., NAGAYOSHI, S., HASE, Y., and
TANAKA, A. (2005). Additional improvement of Chrysanthemum using ion beam re-irradiation. TIARA Ann.
Rep. 2004, 60.
7. YAMAGUCHI, H., HASE, Y., TANAKA, A., SHIKAZONO, N.,
DEGI, K., SHIMIZU, A., and MORISHITA, T. (2009a). Mutagenic effects of ion beam irradiation on rice. Breed.
Sci. 59: 169-177.
8. YAMAGUCHI, H., SHIMIZU, A., HASE, Y., DEGI, K., TANAKA, A., and MORISHITA, T. (2009b). Mutation induction
with ion beam irradiation of lateral buds of chrysanthemum and analysis of chimeric structure of induced mutants. Euphytica. 165: 97-103.
9. YAMAGUCHI, H., SHIMIZU, A., DEGI, K., and MORISHITA,
T. (2008). Effects of dose and dose rate of gamma ray
irradiation on mutation induction and nuclear DNA content in chrysanthemum. Breed. Sci. 58: 331-335.
10. YAMAGUCHI, H., SHIMIZU, A., HASE, Y., TANAKA, A.,
SHIKAZONO, N., DEGI, K., and MORISHITA, T. (2010). Effects of ion beam irradiation on mutation induction and
nuclear DNA content in chrysanthemum. Breed. Sci. 60:
398-404.
11. YANG, T. C., and TOBIAS, C. A. (1979). Potential use of
heavy-ion radiation in crop improvement. Gamma Field
Symp. 18: 141-154.
38
Hiroyasu YAMAGUCHI
イオンビーム照射とガンマ線照射による突然変異育種
山　口　博　康
（独）農業・食品産業技術総合研究機構　花き研究所
〒 305-8519　茨城県つくば市藤本 2-1
突然変異育種では，変異原の選択，処理方法は
重要な要素であり，変異率が高く，一方で変異原
処理により付随する障害が少ないことが望まれ
る。そのため，変異原の評価は，変異原処理によ
る障害の程度に対する変異率の高さと定義される
「効率」ですべきとされている。イオンビームは，
高い変異率や変異スペクトルが異なることを期待
され，様々な植物でイオンビームを使った突然変
異育種が行われている。しかし，突然変異育種の
ための変異原としての観点から，従来用いられて
きたガンマ線などと比較した研究はほとんどな
い。そこで，「効率」を中心にイオンビームとガ
ンマ線とを比較した。また，突然変異頻度や照射
による障害に及ぼすガンマ線の線量率の影響，お
よび照射線量と線量率の相互作用についても知見
が少ないことから，それについて検討した。
イネ種子に炭素イオン 2 種，ヘリウムイオンを
照射し，生存率，稔実率と葉緑素変異率との関係
から効率を求め，ガンマ線と比較した結果，生存
率と稔実率のいずれを基準にした場合にも，イオ
ンビームのほうがガンマ線よりも高い，あるいは
同等であった。同じイオン種を用いたキクの培養
葉片に対する照射試験では，核 DNA 量への影響
と花色変異頻度の関係から求めた効率はイオン種
によって異なり，イオン種によってはガンマ線よ
りも効率が低かった。
また，照射によって生じる変異セクターの大き
さにも，イオンビームとガンマ線とで差がみられ
た。キクの腋芽にイオンビームを照射して得られ
た変異体において，層構造（ＬⅠ，ＬⅡ，ＬⅢ）
を越えた変異セクターの拡大がみられた。照射
後， 2 回の切り戻しを行い作出した花色変異体の
キメラ構造を解析した結果，炭素イオン，ヘリウ
ムイオンでは完全変異体と推測される変異体も
あった。それに対し，ガンマ線ではすべて周縁キ
メラだった。その原因として，イオンビームでは，
茎頂分裂組織の層構造の破壊と少数の細胞からの
その再構築が起こったためと推測された。イネ種
子への照射による葉緑素変異の分離頻度からも，
イオンビームでガンマ線よりも広い変異セクター
が入ることが伺われた。
ガンマ線を総線量 15，30，60Gy，線量率 0.5，1，
2Gy/h で照射したキクの葉片を培養し，花色変異，
および照射による障害として核 DNA 量を調査し
た。その結果，花色変異頻度は総線量が同じであ
れば線量率に関わらず同じであった。一方，核
DNA 量は総線量と線量率の両者の影響を受け，
高線量でも低線量率で照射することで核 DNA 量
の減少が抑えられた。核 DNA 量の減少程度と花
色 変 異 頻 度 の 関 係 は 線 量 率 に よ っ て 異 な り，
0.5Gy/h では，1 および 2Gy/h と比べて核 DNA 量
の減少程度が小さくても同程度の変異が得られ
た。このように，キクでは高い線量を低い線量率
で照射することで，変異頻度を低下させずに照射
による障害の小さい変異体の獲得が可能であるこ
とが示された。
MUTATION BREEDING WITH ION BEAMS AND GAMMA RAYS
質疑応答
司会：キクの変異セクターのサイズに関連して，
イオンビームでは完全変異体がとれたことの理
由として細胞間のダメージの差がイオンビーム
の場合は大きいという考察だったのですが，例
えば，ある特定の層がほかの層に比べて感受性
が高いといったことを想定すると，もう少しう
まく説明できるのかなと思ったのですが，その
あたりいかがでしょうか。
山口：L-1，L-2，L-3層の放射線感受性の違いに
つきましては，リンゴやクワなどで感受性が違
うということが報告されております。そういう
ことの違いも考慮していくこともあるのかもし
れませんが，一方でイオンビームですと直接的
な影響が強くて，放射線の感受性に差があるの
か，感受性の差が出にくいのではないかという
ことを考えていて，今回はそのようなことを原
因として考えた次第です。
西尾：ガンマ線とイオンビームの比較で，稔実率
とか生存率あたりの突然変異率をみてみるとイ
オンビームのほうがガンマ線より高いというこ
とでしたが，イオンビームは集中して当たると
いうようなことが理由になるのでしょうか。集
39
中して当たっても，そこの部分が結局同じよう
に障害を受けるのではないかと思います。障害
あたりの突然変異率がイオンビームのほうが高
いということに対する明確な理由付けというの
は考えておられますでしょうか。もう一つ，先
ほどキメラの問題で，イオンビームだとほかの
ところが死んでしまい，一部が残るからという
ふうに言われたのは，それだと逆にはならない
でしょうか。障害が大きいということを示して
いることになるのではないでしょうか。
山口：まず最初のご質問についてですが，同じ障
害を基準として横にとって並立比較したとき
に，なぜイオンビ－ムのほうが高いというの
かについては，原因のイメージが私もありませ
ん。キメラのほうの質問についてですが，確か
に障害という点では大きいですが，その細胞間
で差があるがために大きい障害が効果的に働い
たというか，そういう結果ではないかと思いま
す。結局，イオンビームのほうが効果が高いか
らというよりも，イオンビームのほうが細胞間
で差ができるというところが，先ほどのような
キメラ構造になった原因ではないかと考えてい
ます。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
41
Neo-Domestication of Vigna (Leguminosae) Species and Identification of a Multiple
Organ Gigantism (mog) Mutant of Cultivated Black Gram (Vigna mungo)
Norihiko TOMOOKA, Akito KAGA, Takehisa ISEMURA, Ken NAITO and Duncan VAUGHAN
National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan
Introduction
Vigna is a pan-tropical genus in the Leguminosae
family. Nine wild species have been domesticated from
98 species in the genus Vigna (MAXTED et al., 2004).
The genus Vigna is subdivided into 6 subgenera: subgenera Vigna and Haydonia in Africa, Plectotropis in
Eurasia, Ceratotropis in Asia, and Sigmoidotropis and
Lasiospron in America (Fig. 1). Among the 9 domesticated Vigna species, 2 (cowpea [V. unguiculata] and
bambara groundnut [V. subterranea]) belong to the subgenus Vigna, 1 (tuber cowpea [V. vexillata]) belongs to
the subgenus Plectotropis, and 6 (mungbean [V. radiata], black gram [V. mungo], moth bean [V. aconitifolia],
rice bean [V. umbellata], azuki bean [V. angularis], and
creole bean [V. reflexo-pilosa]) belong to the subgenus
Ceratotropis.
The subgenus Ceratotropis is a morphologically
homogeneous taxonomic group that consists of 21 species (TOMOOKA et al., 2002). This subgenus is also referred to as the Asian Vigna based on its natural distribution. Multiple and independent domestication events
make the Asian Vigna species unique and interesting
materials for studies of domestication genetics (TOMOOKA et al., 2006). In the present paper, we compare
the genetic changes accumulated during the process of
domestication using 4 economically important Asian
Vigna crops: mungbean, black gram, rice bean, and
azuki bean.
Comparison of quantitative trait loci (QTLs) for domestication among 4 Asian Vigna crops
Numerous common differences in morphological
and physiological traits associated with domestication
are observed between the domesticated and wild forms
of different species. These differences, collectively
called the domestication syndrome, result from selection over several thousands of years of adaptation to
cultivated environments, human nutritional requirements, and human preferences (HAWKES, 1983). The 4
cultivated species mngbean, black gram, rice bean and
azuki bean are ideal materials for improving our understanding of the gene evolution related to domestication among Vigna species and for characterizing useful
traits as QTLs for use in breeding.
Fig. 1. Relationships among Phaseolus, African Vigna
(subgenus Vigna), Eurasian Vigna (subgenus Plectotropis), and Asian Vigna (subgenus Ceratotropis) (after TOMOOKA et al., 2010).
To compare the genomic structures and genomic
regions associated with domestication among these
4 species, we previously constructed genetic linkage
maps of azuki bean (HAN et al., 2005), black gram
(CHAITIENG et al., 2006), rice bean (ISEMURA et al.,
2010a), and mungbean (ISEMURA et al., 2012), mainly
using simple sequence repeat (SSR) markers. We have
42
Norihiko TOMOOKA, Akito KAGA, Takehisa ISEMURA, Ken NAITO and Duncan VAUGHAN
also reported several QTLs for domestication-related
traits of azuki bean (ISEMURA et al., 2007, KAGA et al.,
2008), rice bean (ISEMURA et al., 2010a, 2010b) and
mungbean (ISEMURA et al., 2012) in populations derived from crosses between cultivated and wild forms.
The order of common markers in the linkage groups
(LGs) was highly conserved between the 4 species.
Among the domestication syndrome traits, we
compared the QTLs for 3 important traits in the present
study: seed size (100-seed weight), pod dehiscence,
and water absorption by seeds (Fig. 2). We found several differences in the genetics of domestication of these
closely related Asian Vigna crops.
First, the number of QTLs detected differed among
traits and among species. A total of 5 to 7 QTLs per species were detected for seed size increase, but only 1 to
3 QTLs were detected for pod dehiscence (Fig. 2). For
water absorption by seeds, 2 to 5 QTLs were detected.
Second, the percentages of common QTLs (red
ovals) and species-specific QTLs (green ovals) differed
among the 3 domestication traits (Fig. 2). For seed size
increase, a total of 16 loci were detected in the 4 species;
among these 16 loci, 6 (37.5%) were common among
2 or more species. In particular, the QTL detected in
linkage group 2 (LG2) was associated with seed size
increase in all 4 species. For pod dehiscence, a total of
4 loci were detected among the 4 species. Among these
4 loci, one QTL in LG7 was found in azuki bean, rice
bean, and mungbean but not in black gram. In contrast,
a pod dehiscence QTL located in LG5 was detected
only in black gram. Only 1 pod dehiscence QTL was
detected in azuki bean, rice bean and black gram, but 3
were found in mungbean. Interestingly and unexpectedly, no common QTLs were detected for water absorp-
Fig. 2. Comparison of QTLs for 3 domestication-related traits among 4 Vigna crop species.
NEO-DOMESTICATION OF VIGNA AND IDENTIFICATION OF MOG MUTANT
43
tion by seeds among the 13 loci detected.
Third, several of the QTLs showed effects opposite those that would be expected based on the parental
phenotypes (blue ovals, Fig. 2). For example, the cultivated-parent allele in LG6 of mungbean was associated
with pod dehiscence in spite of the fact that the pods of
the cultivated parent showed phenotypically lower pod
dehiscence than those of the wild parent. Likewise, the
wild-parent alleles in LG6 of azuki bean and in LG2
and LG10 of rice bean facilitated water absorption by
seeds, contrary to the expectation based on the phenotypic evaluations of the parents.
MOG (Multiple Organ Gigantism) phenotype of
cultivated black gram
Black gram (V. mungo) is an ancient food legume
that was domesticated in India (Fuller, 2007). Seeds of
black gram are exported to Japan from Southeast Asia
to produce bean sprouts. Thailand is one of the exporting countries, and has been conducting a black gram
breeding program at the Chai Nat Field Crops Research
Center. To develop hairless and large-seeded cultivars,
mutation breeding using gamma rays was applied to
the promising breeding line ‘BC48’ (later released as
a registered cultivar in Thailand, ‘Phitsanulok 2’). As a
result, a mutant line which has larger seeds, pods, and
vegetative organs was generated (Fig. 3; Chinchest and
Nakeeraks, 1991). Based on its prominent gigantism effects on multiple organs, the line was named ‘MOG’
(Multiple Organ Gigantism).
MOG showed higher biomass production than
BC48 under a 12-hour day length (Table 1). The total
leaf area and total leaf dry weight of MOG were significantly higher than those of BC48 at 20, 40, and 80 DAP.
The total stem dry weight and total dry weight of MOG
were significantly higher than those of BC48 at 20 and
40 DAP, but not significantly different at 80 DAP. The
maturity of MOG was slightly later than that of BC48.
At 90 DAP, MOG plants had a mean of 36 immature
pods and 47 mature pods (43% immature pods), whereas BC48 plants had a mean of 12 immature pods and
62 mature pods (16% immature pods). The mean number of mature seeds per plant at 90 DAP was 226.5 for
Fig. 3. Seed and leaf size of wild ancestor of black gram,
BC48 (cultivar), and MOG (mutant line derived
from BC48).
MOG and 397.5 for BC48. The mean number of mature seeds per pod was 4.8 for MOG and 6.4 for BC48.
Since the 100-seed weight of MOG (7.9 g) was higher
than that of BC48 (4.7 g), total seed weight per plant at
90 DAP was not significantly different between MOG
(17.9 g) and BC48 (18.6 g). Since a higher percentage
of immature pods remained on MOG than on BC48 at
90 DAP, the potential seed yield per plant will be higher
for MOG if the growth duration is longer than 90 days.
Domestication, super-domestication and neo-domestication
A locus (mog) linked to the MOG phenotype,
which behaves as a single-gene recessive trait, was
mapped to the upper part of LG8 (Fig. 2). The 100-seed
44
Norihiko TOMOOKA, Akito KAGA, Takehisa ISEMURA, Ken NAITO and Duncan VAUGHAN
Table 1. Comparison of growth traits between BC48 and MOG plants under 12 hrs day length
Short day (12 hrs)
BC48
(%)
MOG
(%)
P3
20 days
Stem length (cm)
No. of nodes
Leaf area (cm2)
Stem DW (g)
Leaf DW (g)
Total DW (g)
18.6
3.0
133.7
0.1
0.3
0.4
100
100
100
100
100
100
24.2
3.0
219.8
0.2
0.4
0.7
130
100
164
185
162
169
**
n.s.
**
**
*
**
40 days
Stem length (cm)
No. of nodes
Leaf area (cm2)
Stem DW (g)
Leaf DW (g)
Flower + immature pod
DW (g)
Total DW (g)
97.1
10.8
1368.2
2.1
3.0
100
100
100
100
100
131.0
9.0
2115.9
2.7
4.7
135
84
155
130
156
**
**
**
*
**
0.4
5.5
100
100
0.1
7.6
29
136
*
*
Stem length (cm)
No. of nodes
No. of branches
Leaf area (cm2)
Stem DW (g)
Leaf DW (g)
Stem + Leaf DW (g)
Flower + immature pod
DW (g)
Mature pod + seed DW
(g)
Reproductive DW (g)
Total DW (g)
125.5
10.8
1.8
1279.5
7.0
3.9
10.8
100
100
100
100
100
100
100
134.8
8.8
0.3
2205.9
8.4
6.8
15.2
107
81
14
172
121
176
141
n.s.
n.s.
**
**
n.s.
**
*
1.8
100
3.6
197
n.s.
17.9
19.7
30.5
100
100
100
12.4
15.9
31.2
69
81
102
**
*
n.s.
No. of mature pods
No. of seeds/pod
No. of mature seeds
Total seed weight (g)
100 seed wt (g)
62.0
6.4
397.5
18.6
4.7
100
100
100
100
100
47.0
4.8
226.5
17.9
7.9
76
75
56.98113
96
169
*
*
**
n.s.
**
DAP
1
80 days
90 days
Traits
BC48 : A black gram breeding line released as recommended cultivar“Phitsanulok 2”in Thailand
MOG : A mutant line of BC48 produced by irradiation of γ ray and shows Multiple Organ Gigantism
1. DAP : Days After Planting
2. DW : Dry Weight
3. Significance level :
**: at 1% level; *: at 5% level ; n.s.: not significant, based on the analysis of variance
weight of black gram cultivar BC48 was 4.7 g, which is
5.2 times that of the wild parent (0.9 g). In azuki bean,
the 100-seed weight of the domesticated parent ‘Kyoto
Dainagon’ was 24.0 g, 9.6 times that of the wild parent (2.5 g). In rice bean, the 100-seed weight of the
domesticated parent was 27.4 g, 15.2 times that of the
wild parent (1.8 g). In mungbean, the 100-seed weight
of the domesticated parent was 6.9 g, 5.8 times that of
the wild parent (1.2 g). In spite of the fact that the domesticated azuki bean, rice bean, and mungbean have
larger seeds than domesticated black gram, a mutant
mog gene gene does not appear to be responsible for
the larger seed size in these three Vigna crops because
no QTL at the location of mog was detected in the map-
NEO-DOMESTICATION OF VIGNA AND IDENTIFICATION OF MOG MUTANT
ping populations. Therefore, if the recessive mog allele
can be introduced into these crops, or derived by mutation, further seed size increase could be expected.
Among the 4 Asian Vigna crops compared here,
fertile F1 plants can be obtained between azuki bean
and rice bean, and between mungbean and black gram.
If the rice bean cultivar QTL allele for 100-seed weight
detected in LG4 could be introduced into azuki bean,
development of a novel azuki bean cultivar having larger seeds than Kyoto Dainagon might be possible. Kyoto
Dainagon is a very popular cultivar because of its large
seed size and therefore commands a high market price.
Low water absorption by seeds sometimes decreases
the quality of azuki bean paste; thus, there is demand
for an azuki bean cultivar having higher seed water permeability. For this purpose, the rice bean cultivar allele
for water absorption by seeds in LG4 could be used for
azuki bean breeding.
For mungbean, demand for cultivars suitable for
machine harvesting is high. The black gram cultivarspecific QTL allele for decreasing pod dehiscence in
LG5 could be incorporated into mungbean for this
purpose. These approaches to detect species-specific
domestication QTL alleles by comparative genomic
studies of closely related crop species and use them to
develop new cultivars of related species could be called
“super-domestication”, as proposed by VAUGHAN et al.
(2007).
We are planning to isolate several of the domestication QTLs (including the mog gene) described in
the present study by map-based cloning. Recent studies have revealed that most of the useful phonotypic
changes accumulated during domestication are based
on loss-of-function mutations (DOEBLEY et al., 2006).
Such mutations have already proved to be useful for
agriculture and can be induced by gamma radiation or
chemical mutagen treatment. Therefore, once the sequence of the target gene becomes available, screening for mutations in that gene could be conducted by
reverse-genetics methods such as TILLING (TILL et al.,
2003). Hence, these useful mutations can be applied to
related species even though there is no cross-compatibility and, more importantly, without using genetic
45
transformation.
Using this strategy, it should be possible to incorporate basic phenotypic changes that occurred during domestication such as seed size increase, low pod
dehiscence, and high water absorption by seeds into
non-domesticated wild species. In other words, “neodomestication” should be possible within a relatively
short time span. In the genus Vigna, there are many
wild species that are adapted to marginal land with
stressful environments. In some of these marginal land
areas, wild Vigna species are used as an occasional food
for humans. For example, Vigna marina, which grows
on tropical seaside beaches, can grow in soils with
NaCl concentrations of 400 mM, while Vigna trilobata,
which grows in sandy soils under drought conditions,
develops a long tap root; both species are reported to be
used as human food (TOMOOKA et al., 2011).
Over many years, attempts to develop stress-resistant cultivars by using stress-resistance genes from
wild species have met with little success. This failure is
mainly because the mechanisms of physiological adaptation to stressful environments are complex and therefore cannot be achieved by incorporating only a few
genes. Compared with genetic adaptations to stressful
environments, the genetic changes that have occurred
during the domestication process are relatively simple.
We anticipate that by changing a few domestication
genes, wild Vigna species could be domesticated to the
primitive-cultivar level. To develop crops for sustainable use under marginal conditions, “neo-domestication” of wild species using domestication genes might
be a better choice than attempting to transfer stresstolerance genes into cultivated species.
References
1.
CHAITIENG, B., KAGA A, TOMOOKA N., ISEMURA T.,
KURODA Y., and VAUGHAN DA. (2006) Development of
a black gram [Vigna mungo (L.) Hepper] linkage map
and its comparison with an azuki bean [Vigna angularis (Willd.) Ohwi and Ohashi] linkage map. Theoretical
and Applied Genetics 113: 1261-1269. DOI 10.1007/
s00122-006-0380-5.
2. CHINCHEST A., and NAKEERAKS P. (1991) Mutation
46
3.
4.
5.
6.
7.
8.
9.
10.
11.
Norihiko TOMOOKA, Akito KAGA, Takehisa ISEMURA, Ken NAITO and Duncan VAUGHAN
breeding of black gram (Vigna mungo). In THAVARASOOK C et al. (eds.) Proceedings of the Mungbean
Meeting 90, pp. 43-46.
DOEBLEY JF., GAUT BS., and SMITH BD. (2006) The molecular genetics of crop domestication. Cell 127: 13091321.
FULLER DQ. (2007) Contrasting patterns in crop domestication and domestication rates: Recent archaeological
insights from the Old World. Annals of Botany 100: 903924.
HAN OK., KAGA A., ISEMURA T., WANG XW., TOMOOKA
N., and VAUGHAN DA. (2005) A genetic linkage map for
azuki bean [Vigna angularis (Willd.) Ohwi & Ohashi].
Theoretical and Applied Genetics 111: 1278-1287.
HAWKES JG. (1983) The diversity of crop plants. Cambridge, MA: Harvard University Press.
ISEMURA T., KAGA A., KONISHI S., ANDO T., TOMOOKA
N., HAN OK., and VAUGHAN DA. (2007) Genome dissection of traits related to domestication in azuki bean
(Vigna angularis) and comparison with other warm season legumes. Annals of Botany 100: 1053-1071.
ISEMURA T., KAGA A., TOMOOKA N., SHIMIZU T., and
VAUGHAN DA. (2010a) The genetics of domestication of rice bean (Vigna umbellata). Annals of Botany,
doi:10.1093/aob/mcq188.
ISEMURA T., TOMOOKA N., KAGA A., and VAUGHAN DA.
(2010b) Comparison of the pattern of crop domestication between two Asian beans, azuki bean (Vigna angularis) and rice bean (V. umbellata). Japan Agricultural
Research Quarterly 45(1): 23-30.
ISEMURA T., KAGA A., TABATA S., SOMTA P., SRINIVES
P., SHIMIZU T., JO U., VAUGHAN DA., and TOMOOKA N.
(2012) Construction of a genetic linkage map and genetic analysis of domestication related traits in mungbean
(Vigna radiata). PlosOne in press.
KAGA A., ISEMURA T., TOMOOKA N., and VAUGHAN
12.
13.
14.
15.
16.
17.
DA. (2008) The genetics of domestication of the azuki
bean (Vigna angularis). Genetics 178(2): 1013-1036.
MAXTED N., MABUZA-DLAMINI P., MOSS H., PADULOSI
S., JARVIS A., and GUARINO L. (2004) An ecogeographic
study. African Vigna. International Plant Genetic Resources Institute, Rome, Italy.
TILL BJ., REYNOLDS SH., GREENE EA., CODOMO CA.,
ENNS LC., JOHNSON JE., BURTNER C., ODDEN AR.,
YOUNG K., TAYLOR NE., HENIKOFF JG., COMAI L., and
HENIKOFF S. (2003) Large-scale discovery of induced
point mutations with high-throughput TILLING. Genome Research 13: 524-530.
TOMOOKA N., VAUGHAN DA., MOSS H., and MAXTED N.
(2002) The Asian Vigna: Genus Vigna subgenus Ceratotropis genetic resources. Kluwer Academic Publishers.
270 pages.
TOMOOKA N., KAGA A., and VAUGHAN DA. (2006) The
Asian Vigna (Vigna subgenus Ceratotropis) biodiversity and evolution. Pages 87-126 In SHARMA AK and
SHARMA A (eds.) “Plant Genome: Biodiversity and
evolution” Vol. 1, Part C Phanerogams (AngiospermsDicotyledons). Science Publishers, Enfield, New Jersey.
TOMOOKA N., KAGA A., ISEMURA T., VAUGHAN DA.,
SRINIVES P., SOMTA P., THADAVONG S., BOUNPHANOUSAY C., KANYAVONG K., INTHAPANYA P., PANDIYAN M,
SENTHIL N., RAMAMOORTHI N., JAIWAL PK., JING T.,
UMEZAWA K., and YOKOYAMA T. (2011) Vigna genetic
resources. In TOMOOKA N and VAUGHAN DA (eds.) “Genetic resources and comparative genomics of Legumes
(Glycine and Vigna)” Proceedings of the 14th NIAS
International Workshop on Genetic Resources. pp.1121. http://www.gene.affrc.go.jp/pdf/misc/internationalWS_14_11.pdf
VAUGHAN DA., BALAZS E., and HESLOP-HARRISON JS.
(2007) From crop domestication to super-domestication.
Annals of Botany 100: 893-901.
NEO-DOMESTICATION OF VIGNA AND IDENTIFICATION OF MOG MUTANT
47
アジア原産ササゲ属マメ科植物のネオ・ドメスティケーションと
ケツルアズキの多器官大型化突然変異
友岡　憲彦・加賀　秋人・伊勢村　武久・内藤　健・ダンカン ヴォーン
農業生物資源研究所　遺伝資源センター　多様性活用研究ユニット
〒 305-8602　茨城県つくば市観音台 2-1-2
我々は，マメ科作物とその近縁野生種を対象
に，進化と栽培化の過程で生み出された遺伝的多
様性を解明し，それを利用することを目的に研究
を進めている。
本稿では，アジアにおいて独立に栽培化が進行
した ₄ 種のササゲ（Vigna）属マメ科作物である
アズキ，ツルアズキ，リョクトウおよびケツルア
ズキの栽培化に伴う遺伝変異について考察を行
う。栽培化の過程で，これらの作物は，多くの器
官とくに種子の大型化，裂莢性と種子休眠性の消
失という共通の形質変異を獲得し，人間による栽
培に適した作物となった。そこで，これら共通
の形質変異をもたらした遺伝的背景を，QTL 解
析を行うことによって明らかにし，検出された
QTL の種間比較を行った。その結果，共通形質
獲得の遺伝的背景には，種を超えて共通して利用
された遺伝変異と，ある種の栽培化においてのみ
利用された遺伝変異とが存在し，その数と割合は
形質によって大きく異なることが明らかになっ
た。検出した QTL のうち，形質変化への寄与率
が高いものに関して，現在遺伝子の単離に向けた
解析を進めている。
ケツルアズキにおいては，栽培化で大型化した
栽培品種へのガンマー線照射によって，さらに
多器官が顕著に大型化した突然変異体（MOG：
Multiple Organs Gigantism）が得られている。この
突然変異遺伝子のマッピングを行った結果，本突
然変異は，近縁種の栽培化過程における器官大型
化には用いられなかった突然変異であることが明
らかになった。本突然変異は，種子やその他の器
官を約 2 倍に大型化する極めて作用力が大きな突
然変異であり，しかも種子稔性の低下がほとんど
見られないことから，産業利用上有望な遺伝子で
あると考えられ，単離に向けた解析を行ってい
る。
これらの研究で明らかになってきた，近縁作物
では利用されなかった栽培化関連有用変異をも
たらす遺伝子が単離できれば，交雑親和性がな
い種に対しても，化学物質や放射線照射によっ
て 誘 起 さ れ た 変 異 体 を 用 い た TILLING に よ る
スクリーニングによって効率的に有用変異体を
獲得し，作物のさらなる改良を行ったり（Super
-domestication），限界環境に適応した野生種を新
規に作物化（Neo-domestication）したりできるよ
うになると期待している。
48
Norihiko TOMOOKA, Akito KAGA, Takehisa ISEMURA, Ken NAITO and Duncan VAUGHAN
質疑応答
久保山：種子が大きくなる MOG という突然変
異体において，葉がシワになるという写真で，
細かく小さな細胞がたくさん集まる話をされ
ましたが，それは気孔細胞のように見えたの
ですが。
友岡：そうですね。気孔がたくさん見えますね。
こういうふうなところどころにシワの入った
葉ではあるんですね。
久保山：気孔が一カ所に固まるというのは，普通
はないと思うので，一種の突然変異だと思い
ます。シロイヌナズナのそういう気孔の突然
変異体との表現型とかも比較されると，面白
いことが分かるかもしれないと思いました。
高木：ナチュラルバリエーションで耐塩性などに
関連する遺伝子についての情報はありますか。
友岡：今のところ，うちの研究では耐塩性は始め
たばかりですので，遺伝子には迫っていませ
ん。
高木：将来はどうでしょうか。
友岡：現在，強いものと弱いものとで交配ができ
るものについて雑種集団を作っていて，まず
QTL から遺伝子に迫るという手法でやってい
こうと思っています。
一番強い Vigna marina，ハ
マササゲ，あれは種内全部が強いんです。こ
れらでは，強いのと弱いのとの雑種集団がで
きないので，それらについてはいわゆる遺伝
解析，QTL から遺伝子単離という方法は使え
ないと思っています。別種でルテオラという
のがありまして，それと何とか去年交配がで
きましたので，その集団が使えればいいかな
というふうに思っています。
高木：こういうのを掛け合わせたりできるんです
ね。いわゆる栽培品種と掛け合わせて，実際
に耐塩性は付与されるのですか。
友岡：掛け合わせができる野生種もあります。例
えば，アズキとかけられる野生種も５～６種
ありますが，その中で一番強い耐塩性を持っ
ているのは沖縄にあるビグナリュウキュウエ
ンシスです。ところが，ビグナリュウキュウ
エンシスの耐塩性は野生種の中で特別に強い
とは言えません。しかし，栽培種と交雑でき
ない野生種の中には特別に強い耐塩性のもの
があったりするんです。なので，アズキと交
雑できる程度の耐塩性のものは普通の交雑で
耐塩性を付与できますが，アズキを極端に強
くするというのは交雑ではできないので，いっ
そのこと耐塩性野生種を栽培化して新作物に
したらどうかなというのが発想です。
司会：MOG の分離集団ではなく，QTL 解析をやっ
たときの集団は MOG のケツルアズキのものは
MOG とは親品種ということですか。
友岡：祖先種になります。
司会：QTL がもう一つありますよね。LG3 にも，
その種子に関して。
友岡：そうですね。
司会：MOG の，元の MOG の変異体は葉にシワ
がないように思えますが。
友岡：あります。
司会：分かりました。QTL のもう一つのほうも，
この大粒化には寄与しているのですか。
友岡：そうですね。ケツルアズキの集団は実は二
つとも一挙に解析したかったので，祖先野生
種の小さい ₁ グラムのものと， ₅ グラムの栽
培種でやらずに 10 グラムの突然変異体の雑種
集団で行ってみたのです。なので，こういう
ところに出てきている大型からの遺伝子は ₁
グラムから ₅ グラムまでいく栽培化の過程で
利用された突然変異です。 ₈ 番に見つかった
その巨大なものが，その栽培種から 10 グラム
に変化した，放射線で変異した遺伝子だとい
うふうに理解しています。
司会：この MOG の突然変異は大きくなりますが，
生育上のデメリットはどうでしょうか。
友 岡：12 時 間 以 下 の 短 い 日 長 で 作 る と， 同 じ
ように生育して，バイオマス生産はミュータ
ントのほうが大きいです。種子収量も，やや
ミュータントが多くなる。でも，長日で作ると，
親品種も長日では花を咲かせないのですが，
ずっと栄養成長となります。長いこと，例え
ば夏につくばで作ると，大きく大きくなるん
です。大きく大きくなっていったときに，突
然変異体のほうよりも，元品種のほうが葉を
たくさん作り続けて，そちらのほうが大きい。
ただ，栽培上はそういうふうに作ると，もと
もとの元品種でもいつまでたっても種がとれ
NEO-DOMESTICATION OF VIGNA AND IDENTIFICATION OF MOG MUTANT
ないので，それがデメリットにはならないか
と思いますけれども，そういう差はあります。
あと葉が大型化するのと同時に節間も伸びる
し，葉柄も伸びるんです。葉が重く，葉が大
49
きくなりすぎて垂れて，ジョイント部分がそ
の重さに耐えきれずに，ちょっと強い風が吹
いたりすると変異体の方は葉が取れるという
のはあります。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
51
Genetic and Hormonal Control of Parthenocarpy in a Newly Selected Line of Eggplant
Kaori KIKUCHI, Koji MIYATAKE and Hiroyuki FUKUOKA
Molecular Genetics and Physiology Research Team, National Institute of Vegetable and Tea Science,
National Agriculture and Food Research Organization, 360 Kusawa, Ano, Tsu, Mie 514-2392, Japan
Introduction
To minimize yield losses under unfavorable conditions, the adoption of parthenocarpic cultivars may be
a cost-effective and labor-saving solution. In tomato,
a species in which parthenocarpy has been well characterized, genetic parthenocarpy is the result of lossof-function mutations in Pat genes, pat, pat-2, and the
pat-3/pat-4 (FOS and NUEZ, 1996, 1997; GEORGE et al.,
1984; PECAUT and PHILOUZE, 1978). The parthenocarpy is reported to be caused by recessive mutations with
pleiotropic effects (PHILOUZE and PECAUT, 1986). In
addition to parthenocarpy, these mutants have abnormal floral organs and fruit with fewer seeds (FOS et al.,
2000, 2001; JOHKAN et al., 2010; KATAOKA et al., 2008;
MAZZUCATO et al., 1998). Recently, the parthenocarpic
eggplant cultivar ‘Anominori’ was developed at the
National Institute of Vegetable and Tea Science (SAITO
et al., 2009). The parthenocarpic trait of ‘Anominori’
and several true-bred parthenocarpic lines derived during the breeding process is reported to be controlled by
a semi-dominant locus (SAITO et al., 2005). The floral
organ development and seed formation of these parthenocarpic eggplant lines are almost normal, suggesting
that the mechanism controlling parthenocarpy differs
between these eggplants and the pat mutants of tomato.
Here, we introduce our research on the mechanism of
parthenocarpy in these eggplant lines.
Breeding of parthenocarpic eggplant and research
on control of fruiting
It is known that most Japanese eggplant cultivars
can set fruit without pollination, but the fruit fails to
grow and is often small or malformed. Malformed fruit
is often produced under conditions of unusually high
or low temperature, so phytohormone (auxin) applications to flowers at anthesis have been widely performed
in commercial eggplant production under unfavorable
conditions and climates (NOTHMANN and KOLLER,
1975; OLYMPIOS,1976). However, hormone treatment
requires a great deal of labor (typically one-fourth to
one-third of the total working hours for cultivation) and
is therefore expensive. To mitigate the defects in fruit
set under unfavorable conditions, the use of parthenocarpic cultivars may be a cost-effective alternative, because fruit set in such cultivars does not depend on the
presence of favorable pollination conditions.
A parthenocarpic eggplant F1 cultivar, ‘Talina’,
has been developed in Europe. In an eggplant-breeding
program in Japan, the parthenocarpic trait from ‘Talina’ was introduced into Japanese cultivars in 1994 and
‘Anominori’, a parthenocarpic eggplant cultivar with
fruit qualities preferred by Japanese consumers, was
released in 2006 (SAITO et al., 2009). Many true-bred
parthenocarpic lines (designated “AE-P lines”) were
obtained during the breeding of ‘Anominori’. It was
notable that the fruit set rate was higher in these AE-P
lines than in the parent cultivar ‘Talina’. Therefore,
we further evaluated the fruiting ability of ‘AE-P03’,
which had the highest fruit set rate of all the AE-P lines.
‘AE-P03’, ‘Talina’, ‘Senryo Nigou’ (a non-parthenocarpic F1 cultivar; Takii & Co., Ltd.), and ‘LS1934’
(a non-parthenocarpic germplasm from tropical Asia;
YOSHIDA et al., 2001) were grown under various conditions, and the set rates of normal and malformed fruit
were assessed. In the field, the total fruit set rate and
52
Kaori KIKUCHI, Koji MIYATAKE and Hiroyuki FUKUOKA.
normal fruit set rate of ‘AE-P03’ and ‘Senryo Nigou’ did
not differ significantly, but both were significantly higher
than those of ‘Talina’ and ‘LS1934’. In the greenhouse,
the total fruit set rate and normal fruit set rate of ‘AEP03’ were significantly higher than those of the other
three lines. In a growth chamber experiment, fruit set rate
was highest in the 25/15 ºC (day/night) condition, and
decreased at higher temperature. Although the total fruit
set rate was higher in ‘Senryo Nigou’ than in ‘Talina’,
most of the ‘Senryo Nigou’ fruit was malformed. On the
other hand, the normal fruit set of ‘AE-P03’ was higher
than that of the other lines under almost every growing
condition. Notably, high temperatures such those in the
30/20 ºC growth chamber treatment have an inhibitory
influence on fruit set, but ‘AE-P03’ exhibited a higher
fruit set rate than the other lines even under these conditions. These results indicated that the fruit set rate of
‘AE-P03’ is higher than that of ‘Talina’ under a wide
range of conditions.
To elucidate the reason for the high fruit set rate
in ‘AE-P03’, the fruiting characteristics of ‘Nakate
Shinkuro’, one of the parents used for breeding of
‘AE-P03’, were compared to those of ‘AE-P03’, ‘Talina’, and ‘LS1934’. These lines were grown in the
greenhouse in mid-summer, and the fruit set rates from
emasculated flowers were compared. Under these conditions, ‘LS1934’ and ‘Talina’ produced no fruit. ‘Nakate Shinkuro’ was able to set fruit (20% total fruit set),
but all were malformed. ‘AE-P03’ was most successful, with total fruit set rate and normal fruit set rate of
34% and 27% of flowers, respectively. These results
suggested that, in the absence of pollination, ‘Nakate
Shinkuro’ can bear only malformed fruit, but its total
fruit set rate is higher than that of ‘Talina’. In contrast,
the parthenocarpic variety ‘Talina’ shows low fruit set
rate, but fruit that do set grow to normal size. We therefore hypothesized that parthenocarpy and high fruit set
are controlled by different mechanisms and contribute
independently to ovary growth and fruit set in eggplant.
It is also possible that some parthenocarpic ‘AE-P’
lines such as ‘AE-P03’ exhibit both traits: high fruit set
rate provided by the ‘Nakate Shinkuro’ and the ability
of unpollinated fruit to grow to normal size provided by
the parthenocarpic trait from ‘Talina’.
Relationship between parthenocarpy and plant hormone levels
In eggplant, parthenocarpic fruit-set and growth
can be induced by the application of an auxin. ROTINO
et al. (1997) reported that the ectopic overexpression
of iaaM, a gene involved in indole-3-acetic acid (IAA)
biosynthesis, induced parthenocarpy in eggplant when
expressed in ovules under the control of the DefH9 promoter. On the basis of this finding, we investigated the
role of endogenous IAA in the fruit set and growth of
‘AE-P03’ and ‘LS1934’. At anthesis, endogenous IAA
content in these ovaries was low and did not differ between the two lines under the emasculated condition.
An increase in IAA content was observed during fruit
development, indicating that IAA might have a role
in fruit enlargement. Further quantitative analysis of
various phytohormones during fruit development suggested that a decrease of ABA content in the ovary is
correlated with fruit set. Taken together, our observations indicated that fruit set is induced by decreases in
ABA content and that fruit enlargement is induced by
increases in IAA content.
Transcriptome analysis of ‘AE-P03’ using microarrays
The parthenocarpy of ‘AE-P03’ includes two
events: fruit set and fruit enlargement. Our observations suggested that those two events were regulated by
ABA and IAA, respectively. In tomato, much is known
about the hormonal interactions that take place after
pollination and the molecular mechanisms involved in
the control of parthenocarpy (de JONG et al., 2009a, b;
NITSCH et al., 2009; VRIEZEN et al., 2008). In the parthenocarpic eggplant line ‘AE-P03’, we expected that
close interactions between multiple hormones would
regulate the parthenocarpic process. To identify candidate genes that are essential for parthenocarpy in
‘AE-P03’, comprehensive and comparative analysis of
gene expression profiles was performed using microarrays of ‘LS1934’ treated with various hormones and of
untreated ‘AE-P03’. Untreated ‘Nakate Shinkuro’ was
used as a second non-parthenocarpic reference. The ex-
GENETIC AND HORMONAL CONTROL OF PARTHENOCARPY IN EGGPLANT
perimental agents included five basic phytohormones
(auxin, cytokinin, gibberellin, brassinosteroid, and abscisic acid) and an inhibitor of each compound.
First, genes that showed either three-fold higher
or three-fold lower expression level in the ovary of
‘AE-P03’ than in the ovaries of ‘Nakate Shinkuro’
and ‘LS1934’ were selected. From this gene set, genes
which showed more than a three-fold difference in expression level between ‘LS1934’ and ‘Nakate Shinkuro’ were removed, leaving 283 genes identified as
candidate parthenocarpy-specific genes. Based on the
functional categories determined by using Gene Ontology (FUKUOKA et al., 2010), the proportion of genes
related to signal transduction and secondary metabolism in the set of 283 candidates was remarkably higher
than that in the whole transcriptome. Cluster analysis
of the expression profiles of the 283 candidate genes
revealed that non-parthenocarpic ‘LS1934’ treated with
fluridone, IAA, t-zeatin, N-(2-chloro-pyridin-4-yl)N’-phenylurea (CPPU), 4-chloro-2-cyclobutylamino6-ethylamino-s-triazine (CCET), or brassinosteroids
exhibited relatively similar expression profiles to untreated ‘AE-P03’. It is suggested that auxin, cytokinin
and brassinosteroid were positive and abscisic acid
was negative in relation to expression profile of ‘AEP03’ specific genes. Although 703 genes were found
to be differently expressed between untreated ‘AE-P03’
and untreated ‘LS1934’, 766 genes were also significantly over- or underexpressed in ‘Nakate Shinkuro’
compared with ‘LS1934’ and approximately half of
these (388 genes) were found in both sets. The result
suggested that the differential expression of these 388
genes was due to varietal differences and not related to
parthenocarpy; varietal differences can therefore have
a considerable effect on the analysis of expression profiles. To improve the accuracy and specificity of this
type of comprehensive analysis of gene expression,
near-isogenic lines are required.
Breeding of chromosome segment substitution lines
The results of comparative gene expression analysis using parthenocarpic and non-parthenocarpic cultivars of independent origins revealed that the differences
53
in genetic background prevented accurate identification of parthenocarpy-specific genes. Thus, a series of
chromosome segment substitution lines (CSSLs) was
developed in which the genome regions containing the
parthenocarpy-associated QTLs were derived from
‘AE-P03’ and most other regions of the genome were
from ‘LS1934’. Using an F2 population derived from
a single cross between ‘AE-P03’ and ‘LS1934’, three
main QTLs affecting parthenocarpy (designated as A,
B, and C) were identified. For QTLs A and B from
‘AE-P03’ had positive effects on parthenocarpic fruit
development, and the contribution of QTL A was much
larger than that of QTL B. For QTL C from ‘AE-P03’
had a positive effect on parthenocarpic fruit set.
To confirm the effect of the three QTLs, three
CSSLs were selected, each containing one of the three
(A, B, or C) regions derived from ‘AE-P03’ and denoted as line A, line B and line C, respectively. Under the
unpollinated condition, the expected differences among
the normal and malformed fruit set rates of ‘LS1934’
and the three CSSLs were confirmed. Although line
A showed normal fruit set and development, line B
stopped fruit development after fruit set. Line C showed
fruit set, but the fruit did not develop at all. The effect
of the cultivation environment such as temperature or
light intensity on fruiting was larger in the CSSLs than
in ‘AE-P03’. This indicates that the expression of parthenocarpy induced by a single QTL was unstable.
In future experiments, we plan to estimate the effect of interactions among the three QTLs on fruit set
and development. It will be necessary to clarify the
gene clusters and phytohormones that are involved in
fruit set or development by performing experiments
with this new set of CSSLs. Moreover, additional studies on the interaction between fruit set and fruit development are required. Through these experiments, new
understanding of parthenocarpic mechanisms in eggplant will be obtained.
References
1. de JONG, M., MARIANI, C., and VRIEZEN, W. H. (2009a)
The role of auxin and gibberellin in tomato fruit set. J.
54
Kaori KIKUCHI, Koji MIYATAKE and Hiroyuki FUKUOKA.
Exp. Bot. 60: 1523-1532.
2. d e JONG, M., WOLTERS-ARTS, M., FERON, R., MARIANI,
C., and VRIEZEN,W. H. (2009b) The Solanum lycopersicum Auxin Response Factor 7 (SlARF7) regulates auxin
signalling during tomato fruit set and development. Plant
J. 57: 160-170.
3. FOS, M., and NUEZ, F. (1996). Molecular expression
of genes involved in parthenocarpic fruit set in tomato.
Physiol. Plant. 98:165-171.
4. FOS, M., and NUEZ, F. (1997). Expression of genes associated to natural parthenocarpy in tomato ovaries. J.
Plant Physiol. 151: 235-238.
5. FOS, M., NUEZ, F., and GARCIA-MARTINEZ, J. L. (2000).
The gene pat-2, which induces natural parthenocarpy, alters the gibberellin content in unpollinated tomato ovaries. Plant Physiol. 122: 471-479.
6. FOS, M., PROANO, K., NUEZ, F., and GARCIA-MARTINEZ,
J. L. (2001). Role of gibberellins in parthenocarpic fruit
development induced by the genetic system pat-3/pat-4
in tomato. Physiol. Plant. 111: 545-550.
7. FUKUOKA, H., YAMAGUCHI, H., NUNOME, T., NEGORO,
S., MIYATAKE, K., and OHYAMA, A. (2010). Accumulation, functional annotation, and comparative analysis of
expressed sequence tags in eggplant (Solanum melongena L.), the third pole of the genus Solanum species after
tomato and potato. Gene 450: 76-84.
8. GEORGE, W. L., SCOTT, J. W., and SPLITTSTOESSER, W. E.
(1984). Parthenocarpy in tomato. Hort. Rev. (Amer. Soc.
Hort. Sci.) 6: 65-84.
9. JOHKAN, M., CHIBA, T., KAZUHIKO, M. YAMASAKI, S.,
TANAKA, H., MISHIBA, K., MORIKAWA, T., and ODA, M.
(2010). Seed production enhanced by antiauxin in the
pat-2 parthenocarpic tomato mutant. J. Amer. Soc. Hort.
Sci. 135: 3-8.
10. KATAOKA, K., SAKAKIBARA, T., NISHIKAWA, K., KUSUMI,
K., and YASAWA, S. (2008). Seed formation is affected
by uniconazole, an inhibition of gibberellins biosynthesis, in a parthenocarpic cultivar Severianin of tomato
(Lycopersicon esculentum Mill.). J. Jpn. Soc. Hort. Sci.
77: 283-288.
11. MAZZUCATO, A., TADDEI, A. R., and SORESSI., G. P.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
(1998). The parthenocarpic fruit (pat) mutant of tomato
(Lycopersicon esculentum Mill.) sets seedless fruit and
has aberrant anther and ovule development. Development. 125: 107-114.
NITSCH, L. M. C., OPLAAT, C., FERON, R., MA QIAN.,
WOLTERS-ARTS, M., HEDDEN, P., MARIANI, C., and
VRIEZEN, W. H. (2009). Abscisic acid levels in tomato
ovaries are regulated by LeNCED1 and SlCYP707A1.
Planta. 229: 1335-1346.
NOTHMANN, J., and KOLLER, D. (1975). Effects of
growth regulators on fruit and seed development in eggplant (Solanum melongena L.). J. Hort. Sci. 50: 23-27.
OLYMPIOS, C.M. (1976) Effect of growth regulators on
fruit-set and fruit development of the eggplant (Solanum
melongena L.). Hort. Res. 16: 65-70.
PECAUT, P., and PHILOUZE, J. (1978). A sha pat line obtained by natural mutation. Rep. Tom. Genet. Coop. 28:
12.
PHILOUZE, J., and PECAUT, P. (1986) Natural parthenocarpy in tomato. 3. Study of Parthenocarpy due to the
pat (Parthenocarpic fruit) gene in line Montfavet-191.
Agronomie. 6: 243-248.
ROTINO, G. L., PERRI, E., ZOTTINI, M., SOMMER., H., and
SPENA, A. (1997) Genetic engineering of parthenocarpic
plants. Nat. Biotechnol. 15: 1398-1401.
SAITO, T., YOSHIDA, T., and MORISHITA, M. (2005).
Breeding strategy for labor-saving cultivation in fruit
vegetables. Proc. Veg. Tea Sci. 2: 29-35 (in Japanese).
SAITO, T., YOSHIDA, T., MONMA, S., MATUNAGA, H.,
SATO, T., SAITO, A., and YAMADA, T. (2009). Development of the parthenocarpic eggplant cultivar ‘Anominori’. JARQ. 43: 123-127.
VRIEZEN, W. H., FERON, R, MARETTO, F., KEJIMAN, J.,
and MARIANI, C. (2008). Changes in tomato ovary transcriptome demonstrate complex hormonal regulation of
fruit set. New Phytol. 177: 60-76.
YOSHIDA, T., MATSUNAGA, S., and SAITO, T. (2001). Effect of seasonal condition and genotype on fully developed parthenocarpy in eggplant. J. Jpn. Soc. Hort. Sci.
70 (2): 388 (in Japanese).
GENETIC AND HORMONAL CONTROL OF PARTHENOCARPY IN EGGPLANT
55
新規単為結果性ナスの着果および果実発達機構の解明
菊地　郁・宮武　宏治・福岡　浩之
（独）農研機構　野菜茶業研究所　野菜ゲノム研究チーム
〒 514-2392　三重県津市安濃町草生 360
野菜茶業研究所ではヨーロッパの単為結果性
ナス品種 ＇Talina＇ の単為結果性を日本型ナスに導
入する事で新規の日本型単為結果性ナスを育成
した。この新規単為結果性ナスは ＇Talina＇ よりも
正常果の結実率が高い傾向がある事が分かった。
そこで固定系統の中でも特に高い結実率を示す
＇AE-P03＇ 系統について ＇Talina＇ と日本型交配親の
＇ 中生真黒 ＇，非単為結果性ナス ＇LS1934＇ 等と結実
率の比較を未受粉条件で行った。その結果 ＇Talina＇
は着果した果実は正常に肥大するが，その着果
率は日本型のナスに比べて著しく低い事，＇ 中生
真黒 ＇ は着果した果実はその後正常に肥大しない
が，着果率は高い事，そして両者を交配親にもつ
＇AE-P03＇ は着果率も高く，着果した果実も正常に
肥大する事が明らかになった。この事から ＇AEP03＇ の単為結果性には高着果性と果実肥大性とい
う異なる形質が関与していると推察した。ナスは
オーキシンによって単為結果が誘導されることか
ら，＇AE-P03＇ と ＇LS1934＇ の開花時における内生
IAA 量の測定を行った。しかし両系統間に差異は
見られなかった。＇AE-P03＇ の未受粉果実は受粉果
実と比較して肥大開始時期が遅い事から，果実が
肥大し始める開花後 ₉ 日目に測定を行ったところ
IAA 量は上昇した。この事から IAA 量の上昇が
果実肥大に関与している可能性が示唆された。し
かしナスの落花は開花前後に引き起こされる事が
多く，着果との関連性は不明なままであった。そ
こで着果時における他の植物ホルモンについても
定量を行ったところ，＇AE-P03＇ では ABA 量が低
下している事が明らかとなった。これらの結果か
ら ＇AE-P03＇ の着果には ABA 量の低下が，果実肥
大には IAA 量の上昇が関与していると推察した。
＇AE-P03＇ の単為結果過程には ABA と IAA，ある
いは複数の植物ホルモンによる精密な制御関係が
関与している可能性が考えられたので，マイクロ
アレイを用いた網羅的な遺伝子発現プロフィール
分析を行い ＇AE-P03＇ 特異的な遺伝子発現と植物
ホルモンとの関連性の解明を試みた。＇LS1934＇ の
子房に種々のホルモン及びそれらの阻害剤を施
用し発現が変動する遺伝子群と ＇AE-P03＇ の結実
過程で特異的に発現する遺伝子群との比較解析
を行った。その結果 ＇AE-P03＇ 特異的な遺伝子発
現プロフィールにはオーキシン，サイトカイニ
ン，ブラシノライドが正方向に，ABA が負の方
向に関わっていることが示唆された。これまで
に ＇AE-P03＇ の単為結果性は主に ₃ つの QTL が関
与している事を明らかにしており，現在これら
QTL を ＇AE-P03＇ 型に置換した染色体部分置換系
統群（CSSLs）の育成を行っている。今後はこの
CSSLs を解析に用いる事で，着果と果実肥大に関
するより詳細な機構を明らかにしたいと考えてい
る。
56
Kaori KIKUCHI, Koji MIYATAKE and Hiroyuki FUKUOKA.
質疑応答
司会：着果性と肥大に関した二つの遺伝子がある
ようですが，質的に関係する遺伝子だけでなく，
量的に関与する遺伝子の存在があるかどうかと
いうのと，気温が大きく影響しているようです
が，このあたりを伺います。
菊地：ご指摘のとおり，量的なものが関与してい
ると思います。育成した系統は，温度による影
響を AEP-03 や LS1934 よりも顕著に受けるよ
うな形質を持っています。この A 系統は AEP03 だったら単為結果するような温度環境下で
あっても，単為結果しないというようなことが
分かってきています。やはり温度の作用が大き
いということもありますし，何か量的なものが
かかわってきているのだろうと考えています
が，今のところ，いろいろなものが相互作用し
ているというふうに考えられますので，そこを
一つ一つ探索していくのが少し難しいかなとい
うふうに考えています。
長戸：後の展望の中で，三つの QTL について原
因遺伝子を単離するというような計画はどうで
しょうか。
菊地：マッピングのほうで単離するというような
計画があります。また，アレイによる発現解析
から幾つか候補を絞り込み，QTL 上に座乗し
ているかどうかというのも順次見ていて，そこ
から単為結果に大きく関与するものを同定する
というのが一番の目標ではあります。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
57
Genomics Approaches for the Molecular Analysis of Citrus Mutants
Tokurou SHIMIZU
Okitsu Citrus Research Station, National Institute of Fruit Tree Science
485-6 Okitsu Nakacho, Shimizu, Shizuoka 424-0292, Japan
1. History and importance of citrus mutants
History and brief introduction of citrus cultivars
The citrus industry represents the major part of
the fruit industry in many countries from tropical to
temperate regions (KHAN and KENDER 2007). Both
fresh fruit and processed fruit, including juice, jam, and
canned fruit, are important agricultural products. About
1800 citrus varieties have been collected in Japan and
are maintained at the National Institute of Fruit Tree
Science (NIFTS). Among them, Satsuma (Citrus unshiu Marc., synonym C. reticulata Blanco) is the most
widely produced citrus fruit in Japan and was worth as
much as 129.2 billion yen in 2009.
Despite the breadth of the available germplasm
collections, a limited number of varieties of major citrus
varieties are distributed worldwide. Some of the commercially important types of citrus varieties include
sweet orange, pummelo, grapefruit, tangerine (Satsuma, Ponkan, or other mandarins), and lemon. Each
of these varieties comprises many horticultural cultivars of agronomical and economic importance, most
of which can be traced back to a chance seedling in
the past (HODGSON 1967; KREZDORN 1970; REUTHER
et al. 1967). For example, sweet orange (C. sinensis
(L.) Osbeck, synonym C. × aurantium L.) is the most
widely produced citrus variety in the world. It covers
many unique horticultural varieties, including common
oranges (e.g., Hamlin, Parson Brown, Salustiana, Pineapple, Cadenera, Macetera, Shamouti, Berna, Pera, and
Valencia), navel oranges (e.g., Washington, Navelina,
Navelate, and Cara cara), blood oranges (e.g., Moro,
Tarocco, Sanguinelli, and Maltese Blood), and acidless
oranges (e.g., Lima and Succari). All of these varieties
are mutants of original trees that were selected for their
superior quality during vegetative propagation by grafting or from nucellar seedlings. Contrary to common
misconception, “sour orange” is a different species and
is not a mutant of sweet orange cultivar.
Based on recent phylogenetic studies using molecular markers, it appears that citrus varieties emerged by
iterative hybridization of several ancestral cultivars and
stable maintenance of chance seedlings (MOORE 2001;
NICOLOSI 2007; NICOLOSI et al. 2000). Furthermore, a
new classification system has been proposed based on
these observations (DIANXIANG et al. 2008; MABBERLEY 1997).
Origin of citrus mutants
Various types of mutants affecting maturation period, acid content profile, sugar content, rind and flesh
color, fruit shape, rind smoothness, tree vigor, fruit
yield, seed number, parthenocarpy, pest resistance, and
self-incompatibility have been identified within a wide
range of citrus varieties; some of these were spontaneous mutations whereas others were induced (GMITTER
1995; GULSEN et al. 2007; IWAMASA 1985; IWAMASA
1984; KREZDORN 1970; LIU et al. 2007; NISHIURA
1968; NISHIURA 1964; ROBINSON 1933; ROUSE et al.
2001; TANAKA 1922; TANAKA 1925).
Spontaneous citrus mutants can be derived from
either limb sports (i.e., bud mutants) or nucellar seedlings. A limb sport is a branch with a trait differing from
the original tree that originates from a mutated cell. Another type of spontaneous mutant can appear in nucellar seedlings. A considerable number of citrus varieties,
58
Tokurou SHIMIZU
including Satsuma mandarin, sweet orange, grapefruit,
ponkan, tankan, lemon, and natsudaidai produce polyembryonic seeds that are apomictically developed from
nucellar tissue (IWAMASA et al. 1967; KOBAYASHI et al.
1967; KOBAYASHI et al. 1979; NISHIURA 1967). These
polyembryonic seeds occasionally develop seedlings
that have novel traits; each seedling is regarded as a
new mutant. Mutations have also been induced by artificial irradiation of grapefruit (GMITTER 1995; ROUSE
et al. 2001), Satsuma (SPIEGEL-ROY et al. 1972), and
lemon (GULSEN et al. 2007).
Mutation rates in citrus
Although the emergence of a mutant is considered
a random event, the mutation frequency has not been
well studied. Tanaka first surveyed the occurrence of
early-type mutants from the original Satsuma in Shizuoka prefecture and reported the frequency as 1 tree
out of every 40 000 (TANAKA 1925). IWAMASA (1984)
carried out a large-scale survey to determine the occurrence of very-early-type mutants from early-type Satsuma, as assessed by earlier fruit coloring, from 1977 to
1978 in Saga Prefecture. Very-early-type mutants were
recognized in 35 trees among 1 648 696, or 1 tree out of
every 47 106, which is close to the frequency reported
by Tanaka (TANAKA 1925). These observations suggest
that limb sports occur in approximately 1 out of every
40 000 adult trees. IWAMASA (1984) also reported that
the apparent frequency of very early mutants increased
from 1 out of every 172 557 in young trees (<5 years
old) to 1 out of every 16 136 in aged trees (>16 years
old). Proportional changes in the mutation frequency
according to tree age possibly reflect the fact that the
larger limb sports on older trees make recognition easier.
Phenotypic variation among citrus mutants
Mutation events can accumulate over time in vegetatively propagated plants including fruit trees, producing progeny that gradually diverge from the original
tree. In the case of maturation period, the original Satsuma tree is considered to have developed as a chance
seeding on Kyushu Island (Fig. 1). The fruit of the
original tree is a late type that matures in December.
Fig. 1. Emergence and selection of Satsuma varieties
Varieties in blue boxes surrounded by a double line represent the original Satsuma or initial selections. Varieties in
gray boxes with dashed outlines, black boxes with white characters, white boxes with bold black lines, and dark gray
boxes with black lines represent very early (gokuwase) types, early (wase), middle (nakate), and late/common, respectively. Rounded and rectangular boxes indicate the type of mutation (i.e., nucellar seedling or limb sport, respectively).
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
Starting from the original tree, many mutants that show
early coloring of the rind have been identified. Common or mid types of Satsuma that mature fruit from
mid- to late November have been found as mutants
from among late-type selections. Similarly, early-type
Satsuma, with fruit that matures from late October to
early November, has been found as a mutant among
common and mid types. The very early types, with
fruit that matures 3 weeks earlier than the early type,
have been found among various early types. Many of
these mutants that have altered time of coloring exhibit
rapid decreases in acidity compared to their progenitors. All of these mutants are clonal, although some of
them have been recognized as “cultivars” and patented
due to their commercial importance. Red-color mutants
have also been identified for Satsuma, sweet orange,
grapefruit, natsudaidai, and hyuganatsu. High-sugarcontent mutants are rather difficult to identify due to
the requirement of measuring Brix content for all fruits
set upon an individual branch. Although hyuganatsu is
a self-incompatible plant, several lines that have lost
self-incompatibility have been identified by recognizing significant and stable production of fruit compared
to the originals. Several mutants that have lost their juvenility and bloom at a very young stage have been reported in trifoliate orange (TONG et al. 2009; WAKANA
et al. 2005; WAKANA et al. 1995; YADAV et al. 1980).
Practical importance of citrus mutants
Cross-breeding of fruit trees usually involves selection of candidate scions from a single cross. The
process of cross-breeding has been used to alter a wide
variety of traits, although a single round of the process
requires a long time to complete -16 years or more in
the case of citrus varieties. A cross-breeding strategy
requires huge efforts for selection, as just a few undesirable traits within a cross-bred cultivar can often restrict
its use for commercial production. However, in a crossbreeding strategy, it is not possible to alter more than
one undesirable trait at a time because of the high heterozygosity of citrus. Promising new cultivars should
be maintained for at least 5 years by citrus growers until
the fruit is first shipped after planting; therefore, such
59
cultivars are a risky investment. In contrast, spontaneous mutation plays an important role in improving a
limited number of traits gradually without changing the
other characteristic features of a cultivar. Mutants of
excellent time-tested cultivars that have improved traits
would be a more attractive investment for most citrus
growers than unfamiliar new cultivars. Therefore, new
varieties derived from mutation-based breeding appeal
to many citrus growers because they confer a practical
benefit by minimizing the risk of investment loss compared to the risk of growing new cross-bred cultivars.
For these reasons, mutants of major citrus cultivars
have been sought and patented.
2. C
himerism and other challenges to molecular
characterization of citrus mutants
Stability and chimerism of citrus mutants
Citrus mutants selected from limb sports are regarded as chimeras of mutant and progenitor cells (IMAI
1935; ROOSE and WILLIAMS 2007). A model for chimerism involving different histogenic layers explains the
origin of red-type grapefruit mutants (Fig. 2) and can be
applied to other mutants. Several Satsuma cultivars of
mutant origin are known to revert to their original form.
Fig. 2. Model of chimerism in citrus based on histogenic
layers
A. Diagram of histogenic layers in the shoot. B.
Model for color changes in grapefruit mutants
based on chimerism between histogenic layers.
Pictures were compiled from Iwamasa (IWAMASA
1976).
60
Tokurou SHIMIZU
For example, Aoe Wase of Satsuma is considered to be
one such unstable chimera (TANAKA 1925), and additional similar unstable selections are known. In the case
of Jutaro, the tree occasionally sets a branch that would
be regarded as an intermediate between Jutaro and
Aoshima, which is the original cultivar of Jutaro (TERAOKA, personal communication). On the other hand,
Miyagawa Wase, which was derived from a limb sport
of Zairai found in 1915 (Figure 1), showed good stability during propagation without reversion and became a
major variety of Satsuma (IWAMASA 1988; NISHIURA
1964). An example of another type of spontaneous chimera is Bizzarria, a line that developed by conjugation
of 2 obviously different tissues by grafting (TANAKA
1927) and has been maintained stably for more than
300 years.
Problems of instability due to chimerism are often
observed in induced mutants. Mutants from irradiated
branches are often unstable, possibly reflecting chimerism in the shoots grown from it (IMAI 1935). Consecutive cut-back treatment to elongate a new branch for
several cycles is recommended to resolve chimerism,
but this takes a long time. Selecting promising individuals from nucellar seedlings of chimeric mutants is
another effective process for eliminating chimerism in
citrus.
Problems regarding the molecular identification of
mutants
Citrus cultivars of mutant origin represent 59.1%
of the patented citrus cultivars in Japan; citrus cultivars
developed by cross-breeding represent 34%, and the
remaining 6.9% includes selections from germplasm
collections, grafting chimera, or polyploids of known
accessions or obtained from cellular fusion (SHIMIZU
2010). Unique DNA markers have been developed for
molecular identification of these patented cultivars in
order to protect the breeders’rights. Simple sequence
repeat (SSR) markers are now widely used for cultivar identification (BARKLEY et al. 2006; LURO et al.
2008; OLLITRAULT et al. 2010), and single-nucleotide
polymorphism (SNP) markers have been evaluated for
the same purpose. While these technologies are simple,
reliable, and cost-effective, they are inefficient for the
molecular identification of mutant-origin cultivars because of their least frequency to the progenitor cultivar
as a somatic variant.
Detection of polymorphism within citrus mutants
has been hindered by three limitations. The first is the
absence of technology for the efficient detection of rare
polymorphisms. Several types of DNA markers have
been applied for such purposes, but the polymorphisms
obtained with them are insufficient for cultivar identification (CORAZZA-NUNES et al. 2002; FANG and ROOSE
1997; NATIVIDADE TARGON et al. 2000; YANO and
HARADA 1999). The second is that most technologies
to detect polymorphism do not verify the influence of
chimerism in mutant tissues. Several types of chimeras, including periclinal, sectorial, and mericlinal, have
been proposed for citrus mutants (IMAI 1935; ROOSE
and WILLIAMS 2007). These observations suggest that
the composition, localization, and stability of chimerism would vary among these mutants. The third reason
is a combination of the first two. Molecular identification of chimeric mutants should consider the ratio of
mutant cells to progenitor cells. This ratio could vary
according to the tissue sampled, and will range from
near 0 to close to 1. Molecular identification for such
purposes should have sufficient capabilities to validate
polymorphisms among mutants by quantitative determination of mutated cells within chimeric tissue. These
three problems are present in most if not all fruit trees.
A technical breakthrough solving these issues would be
very valuable for use with all patented mutant cultivars.
Application of next-generation sequencing technologies for mutant analysis
Several types of next-generation sequencing technologies are now common for various research purposes (ANSORGE 2009; MARDIS 2008; VOELKERDING et
al. 2009). These technologies provide the capability to
perform high-throughput sequencing on the megabase
to gigabase scale at a moderate cost, and they enable us
to consider various aspects of genomic interpretation
on any scale. At first glance, these technologies appear
to be suitable for the detection of infrequent polymor-
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
61
phisms and thus might be applicable to citrus mutants.
However, when applied to mutation analysis, the high
heterozygosity of fruit trees interferes with the detection of polymorphisms between mutants and their progenitors. The heterozygosity of citrus genomes is much
higher than that of typical model plants and severely
interferes with polymorphism detection. Chimerism of
mutants also disturbs the interpretation of frequencybased analysis performed with next-generation sequencing. Therefore, a huge amount of sequencing data
is required to detect compositional changes between
mutants and their progenitors. A statistical approach to
detect slight differences in the compositions of the mutant and its progenitor is also required. Consequently,
next-generation sequencing approaches are not yet well
cilitate detailed study.
In contrast to the established techniques and advanced tools available for herbaceous plants, the available approaches are quite limited for fruit trees, despite
their economic importance, because of their large size
and long juvenility period. For example, in the case of
citrus, the average juvenility period is about 7 years,
and modern guidelines for Satsuma production recommend keeping the canopy size as large as 10 m3. These
features make it difficult to evaluate physiological traits
under controlled conditions or to perform experimental verification by genetic approaches. Another aspect
of the problem for many fruit tree plants in contrast
to an herbaceous plant species is that the product of
agricultural importance is not grain but fruit. Many
established for chimeric mutant analysis. According
to the chimeric model of citrus mutants, however, this
technology would be useful for evaluation of mutants
isolated from nucellar seedlings, which effectively
separate mutated cells from non-mutant progenitor
cells during embryogenesis. Careful selection of plant
materials and combinations of comparisons would provide a unique opportunity to conduct a genome-wide
heterozygosity analysis and mutation surveys using
next-generation sequencing technologies. Another application of next-generation sequencing technologies is
the identification of genes that have altered expression
as a result of mutation (LIU et al. 2009; XU et al. 2009).
physiological studies have aimed to identify key factors
determining the development, maturation, quality, and
stable production of fruit. Although previous studies in
model plants are expected to provide clues to various
phenomena in fruit trees, we should note that the origin
and developmental processes of fruit tissue are entirely
different from those of cereals. Limited, incomplete genomic knowledge and fewer tools for evaluation have
restricted the use of whole-genome approaches in fruit
trees. On the other hand, molecular evaluation of altered features by comparing the fruit among citrus mutants would be an alternative approach to overcome the
lack of a genetic approach.
3. M
olecular approaches for evaluation of citrus
mutants
Citrus mutants as tools for physiological study
Linkage analysis of spontaneous and induced
mutants is a common technique used to identify the
genes responsible for an altered trait in an herbaceous
plant species. Environmental treatments such as alterations in day length, light intensity, and temperature and
chemical treatments applied to tissues, whole plants,
and soil under controlled conditions are other useful
approaches for characterizing a mutant in detail. The
well-maintained databases of genome information and
mutant collections available for model plants also fa-
Expression profiling of mutants for the identification of genes controlling traits of interest
Expressed sequence tag (EST) analysis of citrus
varieties has been conducted to build fundamental resources for comparative genomics, polymorphism
surveying, and promotion of physiological research
(CHEN et al. 2006; FORMENT et al. 2005; FUJII et al.
2006; JIANG et al. 2006; TEROL et al. 2007). More
than 500 000 ESTs collected from 15 citrus accessions
have been obtained and deposited into a public database (dbEST, http://www.ncbi.nlm.nih.gov/dbEST/).
We recently built a non-redundant unigene set using a
clustering analysis of citrus ESTs (SHIMIZU et al. 2009;
SHIMIZU 2011). Polymorphisms within the obtained
62
Tokurou SHIMIZU
unigenes were identified and used for SSR, SNP, and
indel marker development (SHIMIZU 2011). The newly
developed DNA markers were successfully applied to
linkage map construction, development of selection
markers, and molecular identification techniques aimed
at protecting breeder’rights. Functional annotation of
the unigenes by comparison with manually curated
gene sets of model organisms was used to deduce their
molecular functions according to gene ontology, InterPro, or KEGG pathway classifications (SHIMIZU 2011).
Some of the annotated unigenes were then used to
construct a citrus DNA microarray. The first version of
the citrus oligo DNA microarray contained 21 495 nonredundant 60-mer oligos and was useful for identifying
genes controlling the responses to phytohormone treatment (FUJII et al. 2008; FUJII et al. 2007). This microarray platform is advantageous due to its wide dynamic
range, and we used it to compare gene expression profiles in juice sacs during fruit development among four
Satsuma cultivars (SHIMIZU et al. 2010). Most genes in
the early-ripening-type cultivar Ueno Wase and Haraguchi wase, early-type Miyagawa Wase, and late-type
Otsu No. 4 showed similar expression profiles. However, we identified three genes that showed higher expression at an early developmental stage in Ueno Wase
than in the other cultivars. While the expression of
these genes remained low from 12 to 20 weeks after
flowering (WAF) in Miyagawa Wase, Haraguchi Wase,
and Otsu No. 4, their expression levels transiently increased until approximately 15 WAF in Ueno Wase.
Identical expression profiles were also observed in a
different year. Two of these genes were citrus homologs
of transcriptional factor genes encoding dehydrationresponsive element-binding protein 1D (DREB1D) and
an NAM-ATAF1/2-CUC2 (NAC) domain–containing
protein. The third gene exhibited weak similarity to a
CCAAT-box-binding transcriptional factor. In Ueno
Wase, the expression levels of these three genes were
highest at the time of the rapid decrease in acid content.
This transient increase in gene expression may be involved in the transcriptional regulation of the decrease
in acid at the early stage of fruit development.
Restriction landmark genomic scanning analysis
used to identify polymorphisms among Satsuma varieties
Expression profiling of citrus mutants with largescale DNA microarrays is a valuable approach for exploring and identifying candidate genes that could be
involved in traits differing between mutants and their
progenitors, or between different mutants. While this
technology will also be useful for understanding the
molecular functions of various traits responsible for
fruit production, it is not suitable for directly identifying mutated genomic regions.
We recently evaluated amplified fragment length
polymorphism (AFLP) markers for the identification of
polymorphisms in citrus, but they could not detect sufficient polymorphic loci (SAKANISHI 2007), as was also
described by Yano and Harada (YANO and HARADA
1999). Several reports claim the successful detection
of many polymorphisms among mutants with several
types of DNA markers, although these reports do not
account for the possible influence of chimerism; furthermore, the reproducibility of these markers has not
been adequately assessed (CHAO et al. 2004; FANG and
ROOSE 1997; NATIVIDADE TARGON et al. 2000).
We applied restriction landmark genomic scanning (RLGS) for identification of polymorphism among
Satsuma mutants. In RLGS analysis, genomic DNA is
digested by two different restriction endonucleases,
the 5′ termini are labeled with radionucleotides, and
the digested DNA fragments are separated by agarose
gel electrophoresis (ANDO and HAYASHIZAKI 2006;
COSTELLO et al. 2002; HATADA et al. 1991; HAYASHIZAKI et al. 1993; OKAZAKI et al. 1995). The separated
genomic DNA is digested with a third restriction endonuclease in agarose gel and is subsequently subjected
to a second round of acrylamide gel electrophoresis in
a direction perpendicular to the first. After the second
gel electrophoresis, a profile for the separation of labeled DNA fragments is obtained by autoradiography.
Since the methylation status at restriction endonuclease
recognition sites can affect the efficiency of digestion,
it can also alter the apparent positions in the RLGS profile, thus revealing an additional polymorphism. The
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
performance of RLGS analysis is expected to be higher
than that of AFLP or SSR markers because it does not
only determine polymorphisms by using a combination of three restriction endonucleases but it also detect DNA methylation status. Comparison of RLGS
profiles among Satsuma mutant cultivars showed that
about 10% of the detected spots were polymorphic
(SHIMIZU, in preparation). Several polymorphic spots
were cloned, and RFLP analysis with the cloned polymorphic DNA as a probe confirmed the polymorphisms
observed on the profile. A faint band detected on RFLP
analysis confirmed the chimerism of mutants obtained
from limb sports.
The polymorphism frequencies observed in RLGS
analysis were higher than our initial expectations. The
high frequencies may reflect mutations accumulated
during the long period of vegetative propagation of
these varieties, although most of the mutations would
be neutral with respect to visible traits. Therefore, this
technique is well suited for polymorphism surveys.
Each polymorphic spot was converted to a simple DNA
marker, which was used to confirm the polymorphism
observed on the profile. Since RLGS analysis is performed using radionucleotides, we recently developed a
modified protocol using fluorescent labeling (SHIMIZU,
unpublished results). The fluorescent labeling method
for RLGS (designated F-RLGS) also successfully detected polymorphisms among Satsuma and hyuganatu
mutants. However, F-RLGS method was also laborious
process than usual DNA marker analysis.
Very recently, we developed an entirely new technology for polymorphism surveys of mutants using
DNA microarrays (SHIMIZU et al. 2011). This technology is substantially more powerful than either RLGS
or F-RLGS due to its simple analysis process, ease in
scale-up (up to 1 million loci simultaneously), flexibility for individual cultivars regarding design changes,
and direct estimation of chimeric composition by elimination of the PCR amplification step. Because genomic
sequences around each microarray probe had already
been obtained and stored in our database, the detected
polymorphic loci were easily converted to sequencetagged site (STS) markers.
63
Impact of citrus genome sequencing on mutation
studies
The citrus genome is about 360 Mb in size, and
most citrus accessions are diploid (ROOSE and CLOSE
2008; TALON and GMITTER 2008). Despite the economic importance of citrus fruits, there are few researchers
of citrus genomics, and efforts for citrus genomic sequencing have been limited. In April 2003, researchers
of citrus genomics from nine countries, including Japan, gathered in Valencia, Spain to establish an international citrus genome consortium, which was formalized
through the Valencia Declaration. Through cooperative
partnership and annual meetings, the Sanger shotgun
sequence of sweet orange with 1.2× coverage of the
genome was determined by a community sequencing
program, JGI (DoE), and released in 2007 (http://www.
jgi.doe.gov/sequencing/why/3128.html). The Spanish
research group at IVIA also released a set of BAC-end
sequences of Clementine in 2008 (TEROL et al. 2008).
The release of these sequences greatly enhanced the
identification of candidate sequences of DNA markers (OLLITRAULT et al. 2010) and novel genes (SHIMIZU 2011), although they were not long, assembled
sequences. Subsequent citrus genome sequencing efforts have targeted haploid Clementine (ALEZA et al.
2009) and diploid sweet orange through a combination
of Sanger sequencing and 454 sequencing. These sequences were released in 2011.
Detailed analysis of mutated regions is expected
to clarify the molecular basis of these mutation events.
The development of a high-performance polymorphism
detection technique will help to effectively identify
mutated loci. In addition, the citrus genome sequence
should be valuable when trying to postulate a model of
mutation. Such a model would contribute toward improving mutation breeding and promoting rapid and effective identification of mutants of interest. All of these
efforts will facilitate mutation breeding approaches
to complement the cross-breeding of citrus varieties.
Further approaches such as expression profiling of mutants using genomics technologies are also expected to
complement the identification of genes that determine
64
Tokurou SHIMIZU
fruit quality.
4. Conclusion
We introduced the history, background, and recent
approaches for polymorphism detection and physiological analysis of citrus mutants. Mutation has played a
very important role in citrus breeding by enabling continuous improvement of varieties with excellent qualities. However, cultivars derived from limb sports retain
chimerism, which complicates mutation detection and
further analysis. A number of mutation-derived cultivars are known for most types of fruit trees (PREDIERI
2001; SANADA and AMANO 1998), and similar problems due to chimerism exist in these as well.
Polyembryony, which is present in some citrus
varieties, provides a method to reduce or eliminate
chimerism by selection of a nucellar seedling that
has developed from a mutated cell. However, it is impossible to apply this to other fruit trees that have no
polyembryony. This technique has a benefit in that it
minimizes disturbances due to chimerism by eliminating the progenitor cells within the mutant progeny. Although polyembryony provides an effective method for
reduction or elimination of chimerism, the long-term
vegetative maintenance of the isolated mutant would
accumulate novel mutation over time. These unique
features of citrus are advantageous for understanding mutation events that occur over extended periods.
Many mutants have also been developed in various agricultural crops of commercial importance (AHLOOWALIA et al. 2004). The emergence of new technologies
for effective identification of mutated loci in conjunction with high-throughput DNA sequencing technology
and large-scale expression profiling will facilitate the
detailed study of mutation events and enhance mutation
breeding in various plants.
Acknowledgements
The authors thank Mr. Toshio YOSHIDA, Mr. Hirohisa NESUMI, and Mr. Terutaka YOSHIOKA for providing the plant samples as well as Dr. Kanako YANO
and Ms. Chiaki HIRANO for their excellent technical assistance. The authors also appreciate Drs. Makoto KAWASE, Hiroshi YANO and Toshiya YAMAMOTO for their
continuous support and encouragement. This work was
partly supported by research grants from the NARO
Research Project No. 184, “Database development
and large-scale analysis of expressed genes in fruit,”
and a research grant, “Development of evaluation and
management methods for supply of safe, reliable, and
functional food and farm produce,” from the Japanese
Ministry of Agriculture, Forestry, and Fisheries.
References
1.
AHLOOWALIA, B.S., M. MALUSZYNSKI and K. NICHTERLEIN. (2004). Global impact of mutation-derived varieties. Euphytica. 135: 187-204.
2. ALEZA, P., J. JUAREZ, M. HERNANDEZ, J.A. PINA, P. OLLITRAULT and L. NAVARRO. (2009). Recovery and characterization of a Citrus clementina Hort. ex Tan. ‘Clemenules’ haploid plant selected to establish the reference
whole Citrus genome sequence. BMC Plant Biol. 9: 110.
3. ANDO, Y. and Y. HAYASHIZAKI. (2006). Restriction landmark genomic scanning. Nat Protoc. 1: 2774-83.
4. ANSORGE, W.J. (2009). Next-generation DNA sequencing techniques. New Biotechnol. 25: 195-203.
5. BARKLEY, N.A., M.L. ROOSE, R.R. KRUEGER and C.T.
FEDERICI. (2006). Assessing genetic diversity and population structure in a citrus germplasm collection utilizing simple sequence repeat markers (SSRs). Theor Appl
Genet. 112: 1519-31.
6. CHAO, C.-C.T., P.S. DEVANAND and B.R. CAO. (2004).
Identification of satsuma mandarin (Citrus unshiu) cultivars in California using amplified fragment length polymorphism (AFLP) markers. Acta Hort 634:159-165.
7. CHEN, C., P. ZHOU, Y. CHOI, S. HUANG and F. GMITTER.
(2006). Mining and characterizing microsatellites from
citrus ESTs. Theor Appl Genet. 112: 1248-1257.
8. Corazza-Nunes, M.J., M.A. MACHADO, W.M.C. NUNES,
M. CRISTOFANI and M.L.P.N. TARGON. (2002). Assessment of genetic variability in grapefruits (Citrus paradisi Macf.) and pummelos (C. maxima (Burm.) Merr.)
using RAPD and SSR markers. Euphytica. 126:169-176.
9. COSTELLO, J.F., C. PLASS and W.K. CAVENEE. (2002).
Restriction landmark genome scanning. Methods Mol
Biol. 200: 53-70.
10. DIANXIANG, Z., T.G. HARTLEY and D.J. MABBERLEY.
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
(2008). Rutaceae. In Flora of China Eds. Z.Y. Wu, P.H.
Raven and D.Y. Hong. Science Press, Beijing, China and
Missouri Botanical Garden Press, St. Louis, Missouri,
USA, pp 51-97.
FANG, D.Q. and M.L. ROOSE. (1997). Identification
of closely related citrus cultivars with inter-simple sequence repeat markers. Theor Appl Genet. 95: 408-417.
FORMENT, J., J. GADEA, L. HUERTA, L. ABIZANDA, J.
AGUSTI and S. ALAMAR. (2005). Development of a citrus
genome-wide EST collection and cDNA microarray as
resources for genomic studies. Plant Mol Biol. 57: 375391.
FUJII, H., T. SHIMADA, T. ENDO, T. SHIMIZU and M. OMURA. (2006). 29,228 Citrus ESTs - collection and analysis
toward the functional genomics phase. Plant Animal Genomes XIV
FUJII, H., T. SHIMADA, A. SUGIYAMA, T. ENDO, F. NISHIKAWA, M. NAKANO, Y. IKOMA, T. SHIMIZU and M.
OMURA. (2008). Profiling gibberellin (GA3)-responsive
genes in mature mandarin fruit using a citrus 22K oligoarray. Sci Hortic. 116: 291-298.
FUJII, H., T. SHIMADA, A. SUGIYAMA, F. NISHIKAWA, T.
ENDO, M. NAKANO, Y. IKOMA, T. SHIMIZU and M. OMURA. (2007). Profiling ethylene-responsive genes in mature mandarin fruit using a citrus 22K oligoarray. Plant
Sci. 173: 340-348.
GMITTER, F. (1995). Origin, evolution and breeding of
the grapefruit. Plant Breed Rev. 13: 345-363.
GULSEN, O., A. UZUN, H. PALA, E. CANIHOS and G.
KAFA. (2007). Development of seedless and Mal Secco
tolerant mutant lemons through budwood irradiation. Sci
Hortic. 112: 184-190.
HATADA, I., Y. HAYASHIZAKI, S. HIROTSUNE, H. KOMATSUBARA and T. MUKAI. (1991). A genomic scanning
method for higher organisms using restriction sites as
landmarks. Proc Natl Acad Sci U S A. 88: 9523-7.
HAYASHIZAKI, Y., S. HIROTSUNE, Y. OKAZAKI, I. HATADA,
H. SHIBATA, J. KAWAI, K. HIROSE, S. WATANABE, S. FUSHIKI, S. WADA and et al. (1993). Restriction landmark
genomic scanning method and its various applications.
Electrophoresis. 14: 251-8.
HODGSON, R.W. (1967). Horticultural varieties of Citrus. In The Citrus Industry Ed. W. Reuther, Webber, H.J.,
Batchelor, L.D. University of California, California, pp
431-591.
IMAI, Y. (1935). The mechanism of bud variation. Am
Nat. 69: 587-595.
IWAMASA, M. (1976). Varieties of citrus (Kankitsu no
Hinshu). Shizuoka Citrus Growers Association, Shizuo-
65
ka.
23. IWAMASA, M. (1985). Role of mutation in fruit breeding.
Gamma Field Symposia 22: 31-48.
24. IWAMASA, M. (1988). Citrus cultivars in Japan. HortScience 23: 687-690.
25. IWAMASA, M., I. UENO and M. NISHIURA. (1967). Inheritance of nucellar embryony in citrus. Bull. Fruit Tree
Res. Stn. E. 2: 9-24.
26. IWAMASA, M., YAMAGUCHI, S., KURIYAMA, T. NAKAMUTA, T., EHARA, T., NITO, N. and KATAYAMA, Y. (1984).
Occurrence of very early mutants from the wase (early
ripening) Satsumas and their characteristics. Bull. Fac.
Agr., Saga Univ. 56: 99-107.
27. JIANG, D., G. ZHONG and Q. HONG. (2006). Analysis of
microsatellites in citrus unigenes. Yi Chuan Xue Bao.
33: 345-353.
28. KHAN, I.A. and W.J. KENDER. (2007). Citrus breeding:
Introduction and objectives. In Citrus Genetics, Breeding and Biotechnology Ed. I.A. Khan. CABI, CAB International, Oxfordshire, UK, pp 1-8.
29. KOBAYASHI, S., I. IKEDA and M. NAKATANI. (1967).
Studies on the nucellar embryogenesis in Citrus I. Formation of nucellar embryo and development of ovule.
Bull. Fruit Tree Res. Stn. E. 2: 9-24.
30. KOBAYASHI, S., I. IKEDA and M. NAKATANI. (1979). Studies on nucellar embryogenesis in Citrus. II. Formation of
the primordium cell of the nucellar embryo in the ovule
of the flower bud, and its meristematic activity. Journal
of the Japanese Society for Horticultural Science. 48:
179-185.
31. KREZDORN, A.H. (1970). Citrus cultivars for the tropics.
Proceedings of the Florida State Horticultural Society.
83: 336-341.
32. LIU, Q., J. XU, Y. LIU, X. ZHAO, X. DENG, L. GUO and J.
GU. (2007). A novel bud mutation that confers abnormal
patterns of lycopene accumulation in sweet orange fruit
(Citrus sinensis L. Osbeck). J Exp Bot. 58: 4161-71.
33. LIU, Q., A. ZHU, L. CHAI, W. ZHOU, K. YU, J. DING, J.
XU and X. DENG. (2009). Transcriptome analysis of a
spontaneous mutant in sweet orange [Citrus sinensis (L.)
Osbeck] during fruit development. J. Exp. Bot. 60: 801813.
34. LURO, F.L., G. COSTANTINO, J. TEROL, X. ARGOUT, T.
ALLARIO, P. WINCKER, M. TALON, P. OLLITRAULT and
R. MORILLON. (2008). Transferability of the EST-SSRs
developed on Nules clementine (Citrus clementina Hort
ex Tan) to other Citrus species and their effectiveness for
genetic mapping. BMC Genomics. 9: 287.
35. MABBERLEY, D.J. (1997). A classification for edible Cit-
66
Tokurou SHIMIZU
rus (Rutaceae). Telopea. 7: 167-172.
36. MARDIS, E.R. (2008). Next-generation DNA sequencing
methods. Annu Rev Genomics Hum Genet. 9: 387-402.
37. MOORE, G.A. (2001). Oranges and lemons: clues to the
taxonomy of Citrus from molecular markers. Trends
Genet. 17: 536-540.
38. NATIVIDADE TARGON, M.L.P., M.A. MACHADO, H.D.
COLETTA FILHO and M. CRISTOFANI. (2000). Genetic
polymorphism of sweet orange (Citrus sinensis [l.] Osbeck) varieties evaluated by random amplified polymorphic DNA. Acta Hort. 535: 51-54.
39. NICOLOSI, E. (2007). Origin and taxonomy. In Citrus
genetics, Breeding and Biotechnology Ed. I.A. Khan.
CABI, CAB International, Oxfordshire, UK, pp 19-43.
40. NICOLOSI, E., Z. DENG, A. GENTILE, S. LA MALFA, G.
CONTINELLA and E. TRIBULATO. (2000). Citrus phylogeny and genetic origin of important species as investigated by molecular markers. Theor Appl Genet. 100:
1155-1166.
41. NISHIURA, M. (1964). Citrus breeding and bud selection
in japan. Proceedings of the Florida state horticultural
society. 77: 79-83.
42. NISHIURA, M. (1968). Mutation in Citrus. JARQ. 3:1014.
43. NISHIURA, M. (1967). Citrus breeding through nucellar
seedling selection. JARQ. 2: 15-19.
44. OKAZAKI, Y., H. OKUIZUMI, N. SASAKI, T. OHSUMI, J.
KUROMITSU, N. HIROTA, M. MURAMATSU and Y. HAYASHIZAKI. (1995). An expanded system of restriction landmark genomic scanning (RLGS Ver. 1.8). Electrophoresis. 16: 197-202.
45. OLLITRAULT, F., J. TEROL, J.A. PINA, L. NAVARRO, M.
TALON and P. OLLITRAULT. (2010). Development of SSR
markers from Citrus clementina (Rutaceae) BAC end sequences and interspecific transferability in Citrus. Am. J.
Bot. 97: e124-129.
46. PREDIERI, S. (2001). Mutation induction and tissue culture in improving fruits. Plant Cell, Tissue Organ Culture. 64: 185-210.
47. REUTHER, W., H.J. WEBBER and L.D. BATCHELOR.
(1967). History, World Distribution, Botany, and Varieties. The Citrus Industry. Volume I
48. ROBINSON, T.R. (1933). The origin of the marsh seedless
grapefruit. J Hered. 24: 437-439.
49. ROOSE, M.L. and T.J. CLOSE. (2008). Genomics of Citrus, a Major Fruit Crop of Tropical and Subtropical Regions. In Genomics of Tropical Crop Plants Eds. P.H.
Moore and R. Ming. Springer New York, pp 187-202.
50. ROOSE, M.L. and T.E. WILLIAMS. (2007). Mutation
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
breeding. In Citrus Genetics, Breeding and Biotechnology Ed. I.A. Khan. CABI, CAB International, Oxfordshire, UK, pp 345-352.
ROUSE, R.E., H.K. WUTSCHER and C.O. YOUTSEY.
(2001). Tracing the development of currently planted
grapefruit cultivars. Subtropical Plant Science. 53: 1-3.
SAKANISHI, M., T. UEDA, T. SHIMIZU, H. FUJII, T. SHIMADA, T. ENDO, T. YOSHIDA and M. OMURA. (2007). AFLP
analysis among satsuma mandarin (Citrus unshiu Marc.)
varieties. DNA Polymorphism. 15: 138.
SANADA, T. and E. AMANO. (1998). Induced mutation
in fruit trees. In Somaclonal Variation and Induced Mutations in Crop Improvement Ed. S.M.B. Jain, D.S.;
Ahloowalia, B.S. (Eds.) Springer-Verlag, New York, pp
401-419.
SHIMIZU, T. (2010). Advances of genotyping technology
for citrus fruits with processed products of them. Nogyo
Gijutsu (in Japanese). 65: 470-477.
SHIMIZU, T., H. FUJII, F. NISHIKAWA, T. SHIMADA, N.
KOTODA, K. YANO and T. ENDO. (2009). Data mining
of citrus sequence data sets to develop microarrays for
expression and genomic analysis. In Plant Animal Genomes XVII, San Diego, CA, p 414.
SHIMIZU, T., H. FUJII, K. YANO, T. SHIMADA, F. NISHIKAWA, T. ENDO and M. OMURA. (2010). Identification of
transcriptional factor genes that show different expression profiles among cutlivars of Satsuma mandarin. In
Proceedings of the International Society of Citriculture
Eds. X. Deng, J. Xu, S. Lin and R. Guan, Wuhan, China.
SHIMIZU, T., FUJII, H., KOTODA, N., YANO, K. and ENDO,
T. (2011a). Data mining of Citrus expression sequence
data sets and application for functional genomic study.
Acta Hortic. 892: 29-36.
SHIMIZU, T., T. YOSHIOKA, H. FUKAMACHI, M. KITA, A.
IMAI, K. NONAKA, S. OHTA and K. YANO. (2011b). Large
scale genotyping for identification of polymorphism
among mutation-derived citrus cultivars. Plant Animal
Genome. XIX: 131.
Spiegel-Roy, P., A.H. TEICH and J. KOCHBA. (1972).
Gamma irradiation and pollen cultivar influence on
polyembryony of “satsuma” (Citrus unshiu Marc.). Radiation Botany. 12: 365-367.
TALON, M. and F.G. GMITTER, Jr. (2008). Citrus genomics. Int J Plant Genomics. 2008: 528361.
TANAKA, T. (1922). Citrus fruits of Japan: with notes
on their history and the origin of varieties through bud
variation. J Hered. 13: 243-253.
TANAKA, T. (1925). Further data on bud-variation in citrus. Jpn J Genet. 3:131-143.
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
63. TANAKA, T. (1927). Bizzarria—A clear case of periclinal
chimera. J Genet. 18: 77-85.
64. TEROL, J., A. CONESA, J. COLMENERO, M. CERCOS, F.
TADEO and J. AGUSTI. (2007). Analysis of 13000 unique
Citrus clusters associated with fruit quality, production
and salinity tolerance. BMC Genomics. 8: 31.
65. TEROL, J., M.A. NARANJO, P. OLLITRAULT and M. TALON. (2008). Development of genomic resources for Citrus clementina: characterization of three deep-coverage
BAC libraries and analysis of 46,000 BAC end sequences. BMC Genomics. 9: 423.
66. TONG, Z., B. TAN, J. ZHANG, Z. HU, W. GUO and X.
DENG. (2009). Using precocious trifoliate orange (Poncirus trifoliata [L.] Raf.) to establish a short juvenile
transformation platform for citrus. Sci Hortic. 119: 335338.
67. VOELKERDING, K.V., S.A. DAMES and J.D. DURTSCHI.
(2009). Next-Generation Sequencing: From Basic Research to Diagnostics. Clin Chem. 55: 641-658.
68. WAKANA, A., Y. FUJIWARA, K. SAKAI, I. FUKUDOME
69.
70.
71.
72.
67
and K. YASUKOCHI. (2005). Frequent occurrence of
precocious flowering in zygotic seedlings derived from
crosses with a monoembryonic and male sterile acid citrus hybrid between Yuzu (C. junos Sieb. ex Tanaka) and
Hanayu (C.hanaju Hort. ex Shirai). J Fac Agr, Kyushu
Univ. 50: 615-624.
WAKANA, A., N. HANADA and K. YASUKOCHI. (1995).
Precocious flowering in zygotic seedlings of trifoliate
orange ‘Flying Dragon’. J Jpn Soc Hort Sci. 64:786 (in
Japanese).
XU, Q., Y. LIU, A. ZHU, X. WU, J. YE, K. YU, W. GUO and
X. DENG. (2009). Discovery and comparative profiling
of microRNAs in a sweet orange red-flesh mutant and its
wild type. BMC Genomics. 11: 246.
YADAV, I.S., S.H. JALIKOP and H.P. SINGH. (1980). Recognition of short juvenility in Poncirus. Curr. Sci. 49:
512-513.
YANO, H. and K. HARADA. (1999). Fingerprinting of
plant genomic DNAs using AFLP technique. DNA Polymorphism. 7: 219-222.
68
Tokurou SHIMIZU
カンキツにおける突然変異の利用とゲノム研究との接点
清　水　徳　朗
（独）農研機構・果樹研究所
〒424-0292　静岡県静岡市清水区興津中町485−6
果樹はヘテロ性が高く，単交雑でも従来にない
新規な特性を有する品種の育成が可能である反
面，育成には 20 年前後を要し，少数の不良形質
が有望系統の普及の妨げになる例も少なくない。
となっていると考えられ，それら組織の存在比が
不明であることも突然変異の検出を困難にしてい
る。これとは別に，果樹では樹体の大きさから栽
培試験や遺伝解析が容易でないため重要形質の評
交雑育種では特定形質の部分改良が困難であるこ
とから，有用な突然変異の探索と栄養繁殖が果樹
育種では重要な位置を占めている。これまでに枝
変わりや放射線照射による人為的な変異誘発から
多くの変異が見出されて品種化されており，カン
キツでは多胚性種子を利用した珠心胚実生からの
変異個体選抜も利用されている。変異体由来の登
録品種は数多く，カンキツでは現在の登録品種の
約 60% を占める。
これまでに，果実の減酸が早く進むことで早期
収穫可能となった変異体が多数見出され，また，
果実糖度や果皮・果肉色，耐病性，果面の粗滑，
収量性，種子数，自家不和合性，浮皮性や樹体，
果実，葉の形状など多くの形質で変異が見出され，
カンキツの交雑育種では困難な，特定形質に限定
した部分的な改良に効果を発揮している。 近年，
価が困難であることから，変異した原因遺伝子を
直接特定することで変化した形質との関連を解明
する研究にも期待が寄せられている。
これまでに多くの果樹でもゲノム情報の収集や
全ゲノム配列の解読が進み，ゲノム情報やそれを
元にした多面的な解析が可能となってきた。 カ
ンキツでも高密度の DNA マイクロアレイを開発，
利用することで，変異体で発現特性が変化した遺
伝子の推定が行われ，近年では大規模シーケンス
解析技術を利用した解析も進められつつある。
変異系統間多型の検出は非常に困難であったが，
我々は高感度のゲノムスキャン技術を利用するこ
とで多数の多型の検出に成功している。 これら
の新しい技術やその解析の過程で蓄積された情報
は，従来困難であった突然変異体の多型検出と発
現遺伝子レベルの解析に新たなブレークスルーを
もたらしつつある。 今後は高速シーケンサも活
用したゲノム研究の成果を積極的に利用すること
で，突然変異育種の高度化，高精度化や，栽培技
術の分子レベルでの検証などに貢献すると期待さ
れる。
変異体の登録品種の育成者権保護を目的に，品種
判別に有効な DNA マーカー開発が求められるよ
うになったが，変異系統間での多型頻度が低いた
めに通常の DNA マーカーでは変異を十分に検出
することが困難である。また，枝変わり由来の変
異体では，元となった品種由来の組織とのキメラ
GENOMICS APPROACHES FOR THE MOLECULAR ANALYSIS OF CITRUS MUTANTS
質疑応答
西尾：自然の突然変異で新しい品種が多く出てい
るというお話でしたが，そういう突然変異は恐
らくほとんどキメラになるのではないかなと思
います。それから珠心胚実生を育てたら親品種
に戻るというケースはどれぐらいの率なので
しょうか。
清水：厳密にすべてについてやっているわけでは
ないのですが，やってみたらうまくいったとい
う例と，やってみたらうまくいかなった，いく
らやってもうまくいかなかったという例があり
ます。厳密には分からないですが，戻るのは多
分半々ぐらいかなと思います。
西尾：珠心胚実生でキメラを解消されることがあ
るということでありましたが，それはすべて解
消するのではないのですか。
清水：そうですね。変異した祖先由来の種子であ
れば，後代に遺伝するわけですが。そうでなけ
れば，変異した祖先由来でなければ，いくら種
をまいても後代でそれが再現できないという例
は確かにあり，先祖返りしてしまいます。
松尾：いろいろな解析が枝変わりなどでもうまく
いくというふうにおっしゃられていました。枝
変わりを増殖するときに採穂部位により樹勢に
違いが出てしまうか，熟期が変わるのではない
かというところが，生産者のほうから問題とし
て上がってきてます。そういうときに，戻って
いるか変異体なのかどうかの確認というのは，
もうできるということと考えてよろしいでしょ
うか。
清水：技術的には多分できるだろうと思います。
栽培的な観点からすると，枝変わりというもの
が見つかった時点で，先ほど西尾先生にご指摘
いただいたのですが，まず１回珠心胚実生を経
由して，枝変わりの中のキメラ性を解消して安
定な系統として普及を図るというのが，多分一
番確実だろうと思っています。ただ，やはり見
つかったらすぐに現場に下ろして普及してほし
69
いというのが生産関係者の願いですので，なか
なかそうはいってないところが難しいですが。
司会：産業的には温州ミカンの需要が縮小してお
りますので，糖度の高いミカンを売るのは，至
上命題になってきています。生産現場では木の
下にマルチを敷いて糖度を上げるなどの努力を
している状況です。特に極早生では糖度が低い
という問題がありまして，高糖度化が求められ
ています。糖度の高い変異体を選抜できる手法
開発について展望はどうでしょうか。
清水：そうですね。温州に限らず，他のものでも
糖度が若干高くなる系統というのは，いくつか
あります。圃場で調べてみると，大体平均糖度
が１度高くなるのです。そこも何とかならない
かと思って，いろいろと調べてきてはいます
が。実際のところ，非常に環境変動の要因が大
きくて，確かに糖の蓄積に非常に重要だと自信
を持って言えるものというのが，実はあまりあ
りません。突然変異系統の間の比較も実はやっ
ているのですが，非常によく分からない結果に
なっている。非常に多くのものがかかわってき
ていて，これは突然変異の結果だけではなくて，
糖代謝のプロファイルも見て，少し超えたモデ
ルを作ってみているのですが。例えば，糖代謝
の例として，液胞に輸送されているパスウエー
をモデルにしてみたりしたのですが。こういっ
た幾つかの中で，特に重要ではないかなという
ステップを考えています。少し重要だろうと考
えていますが，幾つかホモロガスな遺伝子が
あって，それがそれぞれ発現のプロファイルが
違うということが分かってきてます。ただ，実
際にそれがどのように糖蓄積に貢献しているか
ということは，まだ実験的な証明ができていな
い段階です。これをもう少し，連鎖地図の上に
マッピングをして，分離集団で何とか連鎖解析
までもっていけないかということを考えている
段階です。
Gamma Field Symposia, No. 49, 2010 Institure of Radiation Breeding
NIAS, Japan
71
Studies of Genes Regulating Seed Size in Rice
Yukimoto IWASAKI, Yuki IZAWA, Mizuki KONO, Yuki ABE and Kotaro MIURA
Department of Bioscience, Fukui Prefectural University
4-1-1 Matsuoka Kenjyojima, Eiheiji-cho, Yoshida-gun, Fukui 910-1195, Japan
Introduction
Rice yield potential is determined by multiple factors, including seed size (or weight), number of panicles
per plant, and number of seeds per panicle (TAKEDA
and MATSUOKA, 2008; SONG and ASHIKARI, 2008). We
have focused on seed size regulation and have started
to unravel the regulatory mechanisms that control seed
formation through a forward-genetics approach using
seed size mutants. A better understanding of the molecular mechanisms of seed formation will be useful
for improving rice grain production by manipulating
the genes regulating seed size.
Seed size mutants
The seed size mutants described here in our laboratory, namely, the heterotrimeric G protein signaling
mutants, the brassinosteroid (BR) signaling mutants,
and the novel small and round seed (srs) mutants, are
shown in Fig. 1. The d1 mutants have mutations in
D1 (also named RGA1), which encodes the heterotrimeric G protein a subunit (Ga) (FUJISAWA et al., 1999;
ASHIKARI et al., 1999; OKI et al., 2009). The d11 and d2
mutants have mutations in D11 (TANABE et al., 2005)
and D2 (HONG et al., 2003) respectively, which encode
different types of cytochrome P450s that are involved
in BR biosynthesis. The d61 mutants have mutations
in D61 (also named OsBRI1), which encodes the BR
receptor (YAMAMURO et al., 2000). The six srs mutants
are of particular interest to us. Among them, srs3 was
recently identified as a mutation in SRS3, a kinesin 13
protein gene (KITAGAWA et al., 2010).
Chromosome mapping of the genes regulating seed size
Through genetic analysis of 42 local lines producing small or short seeds, we have mapped the mutant
alleles present in 32 of the lines (Fig. 2). We identified 4
d11 mutant alleles (TANABE et al., 2005), 10 d1 mutant
alleles (OKI et al., 2009), 3 srs3 mutant alleles (KITAGAWA et al., 2010), and 6 srs1 mutant alleles (ABE et al.,
2010). The srs1 mutants have mutations in SRS1, which
encodes a novel protein without any known functional
domains (ABE et al., 2010). Although the other four srs
mutants, srs2, srs4, srs5, and srs6, remain to be characterized, the mapping results indicate that the causal
genes of these mutants encode novel factors for seed
size regulation. The chromosomal positions of the BRrelated genes, namely, the OsDWARF gene (HONG et
al., 2002), which encodes a cytochrome P450, and the
Fig. 1. Phenotypes of seed size mutants described in this
report. Growth phenotype (top panels) and seed
morphology (lower panels) are shown. The name
of a local line containing each mutant allele is indicated below the corresponding allele name.
72
Yukimoto IWASAKI, Yuki IZAWA, Mizuki KONO, Yuki ABE and Kotaro MIURA
Fig. 2. Map locations of causal genes corresponding to
the BR-related mutants, d1 and the srs mutants.
Mutant names enclosed by circles are those for
which the causal genes have been cloned. Mutant
names enclosed by boxes are those for which the
chromosomal positions of the causal genes were
roughly identified but which have not yet been
cloned.
DIM/DWF1 gene, which encodes an enzyme functioning in BR biosynthesis (HONG et al., 2005), are also
shown in Fig. 2.
Heed to Morphological analysis of mutant seeds
To determine whether the small and round seed
phenotype is due to a reduction in cell length, cell number, or both factors, we have analyzed the length of epidermal cells in the lemma of d1 (IZAWA et al., 2010) and
srs3 (KITAGAWA et al., 2010) by scanning electron microscopy. The cell number in the longitudinal direction
was estimated by dividing the length of the lemma by
the average length of the epidermal cells in the central
region of the lemma.
Although valuable information can be obtained by
this approach, it remains to be investigated whether the
length of the epidermal cells is proportional to that of
the cells in the inner layers of the lemma. It may also be
important to correlate the regulation of cell length and
cell number in the lemma to that of the palea. Finally,
the measurement of total cell number in the lemma and
the palea may prove to be necessary to make the findings more conclusive.
Genes regulating cell number in the lemma
The d1 mutants, which are deficient for the rice
heterotrimeric G protein a subunit (Ga) gene (D1 or
RGA1), show the typical phenotype of short-seed mu-
tants: shortened leaf sheath, shortened internode, and
small and round seeds (FUJISAWA et al., 1999). In the
d1-1 mutant, the lemma length, the average cell length
in the lemma, and the deduced cell number in the lemma were 66%, 132%, and 51%, respectively, of the
wild-type (WT) values (IZAWA et al., 2010). Thus, the
size reduction of the organs was due to a reduction in
cell number.
Genes regulating the cell length in the lemma
The BRI1 gene, which encodes a BR receptor,
is known to regulate both cell length (YAMAMURO et
al., 2000; OKI et al., 2009) and cell number in the internodes through fine-tuning of the direction and rate
of cell division (NAKAMURA et al., 2006). By analogy,
the cause of the short-seed phenotypes in the BR-related mutants was assumed to be both a reduction in
cell number and impaired cell elongation. The lemma
length, the average cell length in the lemma, and the
deduced cell number in the lemma were 83.0%, 84.1%,
and 99.2% in the d61-2 mutant and 92.0%, 92.3%,
and 99.8% in the d2-2 mutant, respectively, compared
with those in the WT (IZAWA and KONO, unpublished).
These results suggested that the small or short seeds
in the d61-2 and d2-2 mutants were mainly due to cell
length reduction in the seeds.
Recently, the causal gene of the srs3 mutant in rice
was identified by map-based cloning and named SRS3
(KITAGAWA et al., 2010). SRS3 was classified as a member of the kinesin 13 subfamily. The lemma length, the
average cell length in the lemma, and the deduced cell
number in the lemma in the srs3 mutant were 87.2%,
84.4%, and 103.3%, respectively, compared with those
in the WT. Thus, the epidermal cell length in the lemma
of srs3 is shorter than that in the WT, but the epidermal
cell number in the longitudinal direction of the lemma
in srs3 was not significantly different from that in the
WT. These results suggest that the small and round seed
phenotype of the srs3 mutant is caused by a reduction
in cell length along the longitudinal axis of the seeds.
Mode of action of genes regulating seed size
Based on the mutant seed phenotypes, the BR-
73
GENES REGULATING SEED SIZE IN RICE
der light, the elongation of the second internode was
impaired in d61-2, d2-2, d11-1, and d1-1, whereas it
was normal in srs3. When d61-2, d2-2, d11-1 and d1-1
were grown in the dark, the elongation of the internode
was also impaired, hence the phenotype was called
abnormal skotomorphogenesis (i.e., abnormal etiolation). In contrast to these mutants, the internode in srs3
elongated in the dark, similar to WT (Fig. 3). Based on
the similar phenotypes exhibited by the BR-deficient
mutants and the d1-1 mutant, it appears that there may
be crosstalk between the BR and G protein signaling
pathways in regulating the elongation of the internodes.
Fig. 3. Comparison of phenotypic characteristics between
BR-related mutants, d1 and srs3. The seed morphology, the cell morphology in seeds, the elongation patterns of the second internodes, and the
skotomorphogenesis characteristics are compared
among the BR-related mutants (d61-2, d2-2, and
d11-2), d1-1, and srs3.
related genes and SRS3 could be classified as genes that
regulate cell length during seed development (Fig. 3).
SRS3 may function mainly in seed size determination
through a regulatory mechanism that involves the BR
signaling pathway. On the other hand, D1 may belong
to a class of genes that regulate mainly cell number during seed development.
Comparison of the elongation pattern of internodes
in the BR-related mutants, d1 and srs3
The BR-related genes, D1, and SRS3 exert pleiotropic effects on the rice body plane. Comparison of
different organs among these mutants will provide insights into the function of these genes in the regulation
of rice body plane determination. Analysis of the responses to external signals in these mutants could also
help us to dissect the functional interaction between
external signals and these genes.
In Fig. 3, we compare the elongation patterns of
the internodes among the mutants. When grown un-
Potential use of seed-size-regulation genes for the
improvement of grain productivity
Enhancement of the G protein signaling pathway
could enable rice to set larger seeds. We have generated
a chimeric gene that corresponds to a constitutive active form of the heterotrimeric G protein a subunit (QL
gene). The QL gene has a single amino acid substitution (Q223L) that reduces GTPase activity and leads
to constitutive binding of GTP by the encoded protein
(IWASAKI et al., 1997). When the QL gene was introduced into the rice mutant d1, which is defective for
the a-subunit gene, both seed length and weight were
increased by about 20% in the transformants compared
to WT (OKI et al., 2005).
Enhancement of the BR signaling pathway may
also enable rice to set larger seeds. When a chimeric
gene of a sterol C-22 hydroxylase for maize BR biosynthesis (Zm-CYP) was introduced into rice plants,
the seeds were heavier in the transformants than in
WT (WU et al., 2008). These results suggest that the
enhancement of BR levels could lead to increased grain
yield in rice.
Discussion
Many of the seed size mutants showed pleiotropic phenotypes (Fig. 1). In order to make use of the
seed-size-regulating genes for breeding, it will be important to understand the mode of action of these genes
in regulating plant body plan and the traits affected by
74
Yukimoto IWASAKI, Yuki IZAWA, Mizuki KONO, Yuki ABE and Kotaro MIURA
each mutant. For example, it will be important to understand the relationships between seed size and internode length and between seed size and leaf erectness
in the seed size mutants. There are groups of mutants
in which seed length and internode length are proportionally reduced and others in which these traits are not
tightly associated. There are mutants showing or not
showing leaf erectness among these srs mutants.
To elucidate these gene functions, analysis of null
mutants will be important because mutant phenotypes
could differ depending on the severity of the mutation
in a specific causal gene. As shown in Figure 1, the phenotypes of the d1 alleles vary widely; for example, the
morphology of d1-1 (Taichung 65 cultivar) is very different from those of d1-7 (Kinmaze cultivar) and d1-8
(Blue Rose cultivar). The differences in seed phenotype
may be due to the different genetic backgrounds or to
altered function of the gene products caused by different mutations. Selection of appropriate alleles will be
important for use in breeding.
Two studies showed that larger seeds were produced in transgenic plants by enhancing the BR or heterotrimeric G protein signaling pathways, evoking the
potential for manipulating seed size genes as a means
to improve rice yield and grain characteristics. We expect that the more we learn about the function of these
seed size genes, the better efficiency we will achieve in
generating plants with the desired phenotypes. Further
investigations into the relationship between the BR and
G protein signaling pathways will be essential. If these
two signaling pathways are independent from each other, simultaneous enhancement of both signaling pathways may produce even larger seeds than manipulation
of either one alone.
The morphological analyses available at present
seem to indicate that cell length is regulated by the BR
signaling pathway whereas cell number is regulated by
G protein signaling. However, the existence of genes
regulating both cell length and cell number has not
been ruled out. To unravel the mechanisms of seed formation, it would be helpful to identify the causal genes
of the uncharacterized seed size mutants srs2, srs4,
srs5, and srs6. Furthermore, generation of double mu-
tants could provide essential information regarding the
relationships among the genes regulating seed size.
Acknowledgements
We thank Drs. Hidemi Kitano (Nagoya University), Yasuo Nagato (University of Tokyo), and Hikaru
Sato (Kyushu University) for the gift of rice mutants
and Dr. Masahiro Yano, Mrs. Tsuyu Ando, and Mrs.
Izumi Kono (NIAS) for the rough mapping of the seed
size mutants. This work was supported by the Ministry
of Education, Culture, Sports, Science and Technology,
Japan [Grant-in-Aid for Scientific Research on Priority
Areas (No. 19060003)] and the Ministry of Agriculture,
Forestry and Fisheries of Japan [Genomics for Agricultural Innovation IPG-0002].
References
1. ABE, Y., MIEDA, K., ANDO, T., KONO, I., YANO, M.,
KITANO, H. and IWASAKI, Y. (2010). The SMALL AND
ROUND SEED1 (SRS1/DEP2) gene is involved in the
regulation of seed size in rice. Genes Genet. Syst. 85:
327-339.
2. ASHIKARI, M., WU, J., YANO, M., SASAKI, T. and YOSHIMURA, A. (1999). Rice gibberellin-insensitive dwarf
mutant gene Dwarf 1 encodes the α-subunit of GTPbinding protein. Proc. Natl. Acad. Sci. USA 96: 1028410289.
3. FUJISAWA, Y., KATO, T., OHKI, S., ISHIKAWA, A., KITANO,
H., SASAKI, T., ASAHI, T. and IWASAKI, Y. (1999). Suppression of the heterotrimeric G protein causes abnormal morphology, including dwarfism, in rice. Proc. Natl.
Acad. Sci. USA 96: 7575-7580.
4. HONG, Z., UEGUCHI-TANAKA, M., SHIMIZU-SATO, S.,
INUKAI, Y., FUJIOKA, S., SHIMADA, Y., TAKATSUTO, S.,
AGETSUMA, M., YOSHIDA, S., WATANABE, Y., UOZU, S.,
KITANO, H., ASHIKARI, M. and MATSUOKA, M. (2002).
Loss-of-function of a rice brassinosteroid biosynthetic
enzyme, C-6 oxidase, prevents the organized arrangement and polar elongation of cells in the leaves and stem.
Plant J. 32: 495-508.
5. HONG, Z., UEGUCHI-TANAKA, M., UMEMURA, K., UOZU,
S., FUJIOKA, S., TAKATSUTO, S., YOSHIDA, S., ASHIKARI,
M., KITANO, H. and MATSUOKA, M. (2003). A rice
brassinosteroid-deficient mutant, ebisu dwarf (d2), is
GENES REGULATING SEED SIZE IN RICE
6.
7.
8.
9.
10.
11.
caused by a loss of function of a new member of cytochrome P450. Plant Cell 15: 2900-2910.
HONG, Z., UEGUCHI-TANAKA, M., FUJIOKA, S., TAKATSUTO, S., YOSHIDA, S., HASEGAWA, Y., ASHIKARI, M.,
KITANO, H. and MATSUOKA, M. (2005). The rice brassinosteroid-deficient dwarf2 mutant, defective in the rice
homolog of Arabidopsis DIMINUTO/DWARF1, is
rescued by the endogenously accumulated alternative
bioactive brassinosteroid, dolichosterone. Plant Cell 17:
2243-2254.
IWASAKI, Y., KATO, T., KAIDOH, T., ISHIKAWA, A and
ASAHI, T. (1997). Characterization of the putative α
subunit of a heterotrimeric G protein in rice. Plant Mol.
Biol. 34: 563-572.
IZAWA, Y., TAKAYANAGI, Y., INABA, N., ABE, Y., MINAMI,
M., FUJISAWA, Y., KATO, H., OHKI, S., KITANO, H. and
IWASAKI, Y. (2010). Function and expression pattern of
the α subunit of the heterotrimeric G protein in rice.
Plant Cell Physiol. 51: 271-281.
KITAGAWA, K., KURINAMI, S., OKI, K., ABE, Y., ANDO,
T., KONO, I., YANO, M., KITANO, H. and IWASAKI, Y.
(2010). A novel kinesin 13 protein regulating rice seed
length. Plant Cell Physiol. 51: 1315-1329.
NAKAMURA, A., FUJIOKA, S., SUNOHARA, H., KAMIYA,
N., HONG, Z., INUKAI, Y., MIURA, K., TAKATSUTO, S.,
YOSHIDA, S., UEGUCHI-TANAKA, M., HASEGAWA, Y.,
KITANO, H. and MATSUOKA, M. (2006). The role of OsBRI1 and its homologous genes, OsBRL1 and OsBRL3,
in rice. Plant Physiol. 140: 580-590.
O
KI, K., FUJISAWA, Y., KATO, H. and IWASAKI, Y. (2005).
Study of the constitutively active form of the α subunit
12.
13.
14.
15.
16.
17.
75
of rice heterotrimeric G proteins. Plant Cell Physiol. 46:
381-386.
OKI, K., INABA, N., KITANO, H., TAKAHASHI, S., FUJISAWA, Y., KATO, H. and IWASAKI, Y. (2009). Study of novel
d1 alleles, defective mutants of the α subunit of heterotrimeric G-protein in rice. Genes Genet. Syst. 84: 35-42
SONG, X.J. and ASHIKARI, M. (2008). Toward an optimum return from crop plants. Rice 1: 135-143.
TAKEDA, S. and MATSUOKA, M. (2008). Genetic approaches to crop improvement: responding to environmental and population changes. Nat. Rev. Genet. 9: 444457.
TANABE, S., ASHIKARI, M., FUJIOKA, S., TAKATSUTO, S.,
YOSHIDA, S., YANO, M., YOSHIMURA, A., KITANO, H.,
MASTUOKA, M., FUJISAWA, Y., KATO, H. and IWASAKI,
Y. (2005). A novel cytochrome P450 is implicated in
brassinosteroid biosynthesis via the characterization of
a rice dwarf mutant, dwarf11, with reduced seed length.
Plant Cell 17: 776-790.
WU, C., TRIEU, A., RADHAKRISHNAN, P., KWOK, S.F.,
HARRIS, S., ZHANG, K., WANG, J., WAN, J., ZHAI, H.,
TAKATSUTO, S., MATSUMOTO, S., FUJIOKA, S., FELDMANN, K. and PENNELL, R. (2008). Brassinosteroids regulate grain filling in rice. Plant Cell 20: 2130-2145.
YAMAMURO, C., IHARA, Y., WU, X., NOGUCHI, T., FUJIOKA, S., TAKATSUTO, S., ASHIKARI, M., KITANO, H.
and MATSUOKA, M. (2000). Loss of function of a rice
brassinosteroid insensitive1 homolog prevents internode
elongation and bending of the lamina joint. Plant Cell
12: 1591-1605.
76
Yukimoto IWASAKI, Yuki IZAWA, Mizuki KONO, Yuki ABE and Kotaro MIURA
イネ種子形を制御する遺伝子群の研究
岩崎　行玄，井沢　有希，香野　みずき，安部　優樹，三浦　孝太郎
福井県立大学生物資源学部
〒910-1195　福井県吉田郡永平寺町松岡兼定島4-1-1
穀物の収量は，種子の大きさ，穂当たり種子数，
穂数，栽培条件など，多様な要因の総和として決
定される。私どもは，このような要因の内，種子
形に着目し，種子形を決定するメカニズムの解明
を進めている。種子形研究のアプローチとして，
はじめに，種子形を規定する主要な遺伝子群を見
出すことが重要と考え，80 系統を超える粒形変
異体（small and round seed mutant，srs と略す）を
収集し，42 系統をラフマッピングにより解析し，
異なる遺伝子座に座上する ₆ 種類の新規な粒形制
御遺伝子の存在を見出した。
現在までに明らかにされた種子形を制御する遺
伝子は，ブラシノステロイド（BR）シグナリン
グと ₃ 量体Ｇタンパク質シグナリングに関与する
遺伝子である。ここに，新規な ₆ 種類の粒形制御
遺伝子が加わることにより，種子形を制御するた
めのメカニズムの理解が深まることを期待してい
る。
種子の縦方向の細胞数を制御する遺伝子：₃ 量
体 G タンパク質αサブユニット（Ｇα）遺伝子
(D1 or RGA1) が欠失すると，外頴，節間，葉鞘な
どにおいて，細胞長は野生型と同じであるが，細
胞数が約半減していた。この結果より，３量体 G
タンパク質αサブユニット遺伝子は細胞数を正に
制御する遺伝子と考えた。
種子の縦方向の細胞長を制御する遺伝子群：
（Ａ）BR 関連変異体，d2-2 と d61-2 の外頴の内表
皮の細胞長を，SEM を用いて観察し，細胞数を
見積もったところ，両者とも，細胞長が短くなる
ことが原因で種子が小さくなることが示された。
細胞数は野生型と同じであった。以上の結果より，
BR 関連変異体は細胞長を正に制御する遺伝子と
考えた。
（Ｂ）新規粒形制御遺伝子 SRS3 は，キネシン
13 タンパク質をコードする。この遺伝子欠失変
異体，srs3 は，細胞長が短くなることが原因で種
子が小さくなることが示された。以上の結果より，
BR 関連遺伝子と同様に SRS3 遺伝子は細胞長を
正に制御する遺伝子と考えた。なお，BR 関連変
異体 d61 は，節間や葉鞘の伸長において，細胞長
を正に制御するとともに，細胞分裂方向の異常を
介して細胞数を制御することが報告されている。
よって，これらの遺伝子は，器官間で機能発現に
相異がある可能性がある。
イネ G α遺伝子を恒常的活性型に改変した遺
伝子（QL）を，イネに導入した場合，形質転換
イネの種子が大きくなることを見出した。Pennell
らのグループは，トウモロコシ由来の BR 生合成
に関与する P450 (Zm-CYP-1) をイネに導入したと
ころ，種子重量が増加することを見出した。これ
らの結果は，種子形制御遺伝子の利用は，収量を
増大することを示唆した。
今後も粒形制御遺伝子群の同定を進め，これら
の遺伝子を機能に基づいて，種子形の形成機構の
理解を深めたいと考えている。種子形を制御する
新しい遺伝子の同定は，この目的を達成する上で
重要な研究と考えている。
GENES REGULATING SEED SIZE IN RICE
質疑応答
長戸：Ｇαは三量体で機能するのでしょうか。
岩崎：植物は複雑です。動物は三量体で生成して
ますが，イネもアラビドプシスも三量体じゃな
くて，もっと巨大な形になっているということ
と，モノマーでも動きそうであるということで
す。αとβが必ずしも強くついていないのです。
今の質問は，しばらく正確な答えが出ないので
すが，多分違うと思います。
長戸：G βのミューテーションは d1 のように矮
性にならないのでしょうか。
岩崎：植物は複雑です。d1 は G βのミューテー
ションは恐らく致死になる。ここではそういう
データは示さなかったのですけど。αのノック
ダウンとβのノックダウンの表現系は決定的に
違うので，そのことからも一緒に働いていない
部分が結構たくさんあると思います。
長戸：そうすると，いつも三量体 G タンパクと
いうのはイネではあまりふさわしくないことで
しょうか。
岩崎：そう思います。だけど，αサブユニット自
体がヘテロトライメリグ G プロテインαサブ
ユニットという名前が定着しているので。一応
つけてはいるんですけど。植物型 G αと書い
たほうがいいかもしれません。
長戸：もう一点。d1 の発現に関して，節間伸長
DM タイプになりますね。
岩崎：はい，そうなります。
長戸：上から第二節間だけの伸長が抑制されるの
ですか。
岩崎：第二節間が顕著です。
長戸：ワイルドタイプの発現プロファイルから
DM 型になる説明ができますか。
77
岩崎：第二節間でも G αはちゃんと強く発現し
ますし，第一，第二，第三と同じように発現し
ているので，発現プロファイルからは第二節間
が矮化する説明はできません。
長戸：ほかのことからは説明できますか。
岩崎：第二節間だけが小さくなることの説明は私
たちにはできません。
高木：（産総研）大粒種というのは劣性の形質な
のでしょうか。　岩崎：遺伝子の変異ごとにタイプが分かれていて，
半優性や劣性などが混じっています。
高木：Gαのミュータントのプロファイリングを
やられましたか。
岩崎：アレイ解析をしています。
高木：何か興味ある遺伝子は動いていましたか。
岩崎：多くのものが動きすぎてて，よく分からな
いところがあります。
高木：やはり発現プロファイルは，種子のところ
でやらなくてはいけないでしょうか。
岩崎：Gαが大量に発現している組織なら，どこ
でもいいと思います。今年は，個体で発現のプ
ロファイリングを見ました。そうすると，L4
期では４が多いですよね。だからワイルドタイ
プの４とミュータントの４だけで比べないと，
多分分からないとかいうふうな気がしていて。
G αの発現が一番強い，いつもその強い組織ご
とのそのステージの比較を今年やっています。
高木：シグナルのレギュレーションにかかわって
いるということですが，最終的な遺伝子発現に
制御しているということですよね。
岩崎：はい。だから，ターゲットは見たいと思っ
ています。
79
総合討論
座長　北野英己（名古屋大学）
北野： ₉ 名の方のご講演者が，それぞれの分野の
最先端の研究成果を発表されております。そうい
う中で今回のシンポジウムのテーマである「新た
な育種と有用突然変異」ということについて，本
来ならばまとめて議論をしたいと思いますが，そ
れぞれの分野に非常に造詣が深くて，様々な方向・
視点からの報告をされております。また，質疑を
改めてされたいこともあると思いますので，個別
にご議論をお伺いしていきながら，最後に時間が
残るようでしたら，全体にかかわったところで少
し議論を進めていきたいというふうに思っており
ます。それでよろしいでしょうか。
それでは発表の順番に進めていきたいと思いま
す。まず，特別講演の長戸先生のご報告に対し
て，何かご質問あるいはコメント等ございました
ら，ぜひ挙手をされて発表をお願いしたいと思い
ます。
久保山：長戸先生の研究は多くの突然変異をスク
リーニングされて，それに基づいて行われていた
ということですけれども，どちらかというと遺伝
子が壊れる方向の突然変異が多いわけですが，実
際に植物を栽培化したりして，それを品種改良を
していく中で，ピックアップされてきたものの中
にハバタキのように実際に使える，収量を上げる
ような，そういう突然変異も実際には起こってい
るわけですけれども，そういったものを直接拾い
上げるということというのは，頻度としては壊れ
るよりも非常に難しいということなのでしょう
か。
長戸：優性の突然変異が実際に，例えば栽培化の
過程で使われてきたという例は幾つかあります
が，そういうのを変異体として見つけることがで
きるかということですか。
久保山：優性，劣性ということではなくて，結局
異常なものとしてターニングされてきた apo1 と
いうものは，実際にはハバタキみたいにアリルに
よっては収量が増大するような方向の変異もある
わけですけれども。そういったものというのは，
やはり単純に突然変異を見るという中で，出現し
てくる頻度というのは極めてまれだということな
のでしょうか。
長戸：例えば，穂が大きくなるような変異はあり
ます。
久保山：直接利用可能な変異はどうでしょう。
長戸：それは頻度が少ないですけれども出ていま
す。例えば半矮性が有用だとしたら，それは出て
きます。ただ僕がやりたいのは直接に有用なケー
スではなくて，個体レベルの形質の制御過程が本
当にどういうものがあるかということです。多
分，普通のイネを見ただけでは分からないような
制御過程が当然あるはずなので。そういうのを見
いだすことによって，新しい育種の可能性が出て
くるだろうということでやっています。よく分か
るのは，多分ノックアウトの系統ですが，意識的
といえば意識的なのですが，そういうものを中心
にとっています。
久保山：栽培か何かで使われるものというのは，
割とアミノ酸置換とかそういったものが多いと思
うのですが。やはりそれは頻度的に少ないと考え
てよいのでしょうか。
長戸：アミノ酸置換が多いですけれど，突然変異
は。全部とは言いませんが，結構，栽培化の過程
で使われています。例えば，プロモーターに変異
がある。ハバタキの場合もそうですけど，apo1 も。
そういうのを特定するというのは，とても大変で
す。そういうことからは多分分からないだろうと。
多分ハバタキの場合も apo1 のプロモーターに変
異があるのは分かるのですが，どの塩基の変異が
重要かというのはよく分からないのです。幾つか
変異がありますから。
北野：続きまして高木先生のご講演に対して何か
ご質問等ございましたら。
草場：先生が最近ブッシェルがリプレッサーでも
あり，アクティベ－タ－でもあるという論文を出
80
されていると思います。それは遺伝子に特異的で，
その違いは相手のターゲット遺伝子の違いでそう
なるのかどうかということをお聞きします。もう
一つは，これに関連して言えば，この CRES-T 法
の場合，そういったことがあり得るかと。ブッシェ
ルで見られたことは非常に珍しいことであって，
普通は起こらないことかどうかということをお聞
きしたいのですけれど。
高木：ブッシェルは，WUS-box という領域をもっ
ており，これがこれまで使っている SRDX と違
うリプレッションドメインとして機能することが
判りました。これまで調べたところ，ブッシェル
の機能は，リプレッサーです。ところが，アガマ
ス（AG）遺伝子にのみアクティベーターとして
機能し，その活性化は，WUS-box が担っている
ようです。つまり，先生のおっしゃるように，ブッ
シェルは，遺伝子に特異的に抑制因子と活性化因
子の使い分けをしています。何故，リプレッショ
ンドメインが，活性化能を付与するのか，判って
いません。バイファンクショナルな転写因子とい
うのは，ある意味では初めての転写因子です。ま
た，これまで使っている SRDX リプレッション
ドメインによる CRES-T 法では，WUS-box のよ
うなバイファンクショナル機能は見つかっており
ませんので，そのような機能は無いと考えており
ます。
廣近：キメラリプレッサーを異種植物で発現した
場合に，機能する割合というのはどの程度でしょ
うか。
高木：非常に重要な質問です。TCP などですが，
それは非常に保存性の高い植物質で保存性の高い
ものはいろんな植物にシロイヌナズナの遺伝子が
作用しました。同様にキクとアサガオにも作用し
ました。ところが，イネに対しては，シロイヌナ
ズナ由来の転写因子の作用効果は，かなり低いこ
とが判りました。ところが，イネのホモログで作
成したキメラリプレッサーを用いると，シロイヌ
ナズナで見られたのと同様な表現型を誘導する事
が出来ました。転写因子遺伝子の保存性とそれら
と相互作用するタンパク因子との関係が効果的な
作用に関与しているのはないかと考えています。
例を挙げれば，シロイヌナズナの ABC 遺伝子を
用いたリプレッサーは，トレニアやアサガオ等シ
ロイヌナズナ以外の植物は，あまり効果を示しま
せんでしたが，ホモログを用いると，大変効果的
に ABC 遺伝子の変異体の表現型を誘導しました。
おそらく，カウンターパートの違いが決め手じゃ
ないかというふうに考えています。それ故，多様
な植物で CRES-T システムを用いる場合，できる
だけオルソログ，ホモログを単離して使われるこ
とを勧めています。
北野：それでは次に吉田先生のご講演に対して，
何かご質問等ございましたら。
西尾：とられました突然変異体の閉花性のものが
温度感受性であるということで，それは前から
も，とれたときから予測されてたというようなこ
とを聞いて，なかなか興味深かったのですが。ど
ういう遺伝子の変異をとれば，温度依存性でない
閉花性のものがとれるだろうというふうな予測を
ご講演で一つ言っておられたように思います。温
度依存性でない閉花性のものがとれる可能性につ
いて，どういうふうに考えておられますでしょう
か。
吉田：結局，ミスセンスのものだったわけですけ
れども，spw1-cls の場合は。ミスセンスの場合は
どうしてもやはりある程度温度感受性というの
は，可能性としてはついてまわると思うんです。
ただ，その程度がどうかということで。例えば，
その SPW1 遺伝子の中で別のタイプのミスセンス
ミューテーション。ノックアウトするとおしべも
変わるので駄目なんですけれども。SPW1 の場合
だったら，別のタイプのミスセンスというのは可
能性はあるかもしれないと思っていますが，一番
いいのを探すというのは，宝物探しみたいなこと
で，できるかもしれないし，でもかなり難しいか
なという気がします。もしかすると，鱗被の形づ
くりにかかわるほかの遺伝子，ノックアウトをし
てもほかの器官には影響が出ないような遺伝子が
多分どこかにあると思うので，そういったものが
とれてくるとか。開花，開穎というのにかかわる
のは鱗被がもちろん一番大事なのですけれども，
それ以外の現象というのも多分かなり大事なこと
があって，そういう遺伝子のミューテーションと
いうこともあり得るかなと。それを実際とってく
るのを理論的に考えていくというアプローチもあ
ると思います。僕たちが今やっているのは実際圃
81
場で新しいミュータントをとって，例えば去年は
低温年ですけども，とれてきた中で，去年も開穎
しなかったようなものが実際あるので，そういっ
たものに可能性があるかもしれないなというふう
に考えています。
北野：この閉花受粉にかかわる研究というのは，
そのスタートの動機といいますか，組み換え圃場
での安全性とか，そういうことに対して要求が
あってということだと思うのですけれども。この
性質そのものが，もう少し何か積極的に生産に役
立つとか，あるいは採種に役立つとか，そういう
ような流れから何かアイデアみたいなものはある
のでしょうか。
吉田：私も育種体系のほうは詳しくはないので，
素人的なお答えになってしまうかもしれないです
が，一つにはやはり原種とか原原種というところ
で，混じってしまうということは，大問題だとお
聞きしています。そういったところで利用してい
ただくということが一つの可能性かなと思いま
す。もうちょっと広げると，例えばハイブリッド
ライスなどで，系統の維持というのを考えたとき
に，温度感受性が逆にあるので，開穎もさせられ
るし閉じることもできるということで，そういう
システムの中で使っていっていただくということ
も，一つ面白いのかもしれないなと。あとは昨日
のお話で話したような病害虫対策などですね。
北野：それでは次に進みまして，川越先生のご発
表に対してです。私，米粉の利用というのは，す
ごく今社会でもいろいろ話題になっていますし，
すごくいい話が出てきたなというふうに思ってい
ました。お話をお聞きすると，なかなか奥が深い
テーマで，やらなきゃいけないことがいっぱいあ
るという，そういう印象を受けました。この需要
をもっと広げていくために，今回お話しされたこ
とと，それからもう少し現場の要求に合わせてど
んなことが考えられるのか。これをこうすればブ
レークスルーできるとか，そういうようなお話が
あればぜひお聞かせ願いたいと思います。
川越：私も現場の要求というのをそれほど知らな
いのですが，製粉会社の方にお話をお聞きします
と，米は小麦に比べて硬いのが製粉工程での技術
的な問題だということです。本当に米粉に向いた
米の開発をするときには，製粉特性が今後大きな
一つのポイントになるのではないかと思います。
従来育種では製粉特性は全く考えられていません
ので，どのような変異体が製粉特性に優れている
かを明らかにしていくといいのではないかと思い
ます。
北野：それでは山口先生のご講演に対してご質問
等ございましたらよろしくお願いします。
栗田：イオンビームやガンマ線照射によって変異
体が得られますが，どの遺伝子が壊れたかという
のを効率的に探す方法というのは，どういった方
法がよろしいでしょうか。
山口：私は今回キクにしてもイネにしましても，
非常に見やすい葉緑素変異ですとか花色変異を使
いましたが，個々の形質についてそれぞれ別の事
情とかがあるかと思います。それぞれ違うのじゃ
ないかというところで，ご回答させていただきた
いと思います。
栗田：あと，2 ｎの状況で当てると一つの遺伝子
が壊れても，もう一つのほうが残っていると表現
型として効果が出にくいと思うのですが，表現型
を出やすくする方法というのは何かあるのでしょ
うか。
山口：2 ｎで片方を検出したようなものでは，自
殖できるものはそれを自殖させて後代で見て評価
しますし，自殖できないものの場合でヘテロの場
合はもう片方の第一遺伝子のほうの表現型が見え
てくると思います。それを見やすくする方法は今
の DNA 的な手法を使ったら，あるのかもと思い
ます。
北野：ユーザーの立場からすると，自分たちが材
料を作りたいときに，どういう変異原を使用した
ら最も自分たちの目的にかなうかということで，
よく議論になるのですけれども。例えばイオン
ビームの場合は点突然変異が多いとか，ガンマ線
なんかを使うとデリーションがいっぱいあるから
駄目だとか，いろいろなことをお聞きするのです
が。どういうふうな方法を選んだらいいかという
ようなことで，われわれが参考になるようなこと
はないでしょうか。
山口：目的によってバックグラウンドがいくら壊
れても構わないので特定の遺伝子だけが変わっ
て，目的としては変わっているものだけをとりた
いというのであれば，ガンマ線の強い線量で十分
82
かと思います。目的別にそれぞれあるとは思いま
すが，ケミカルですとかイオンとかの違いと目
的とするケースとそれが合っているかどうかは，
個々にやってみないと分からないのではないかと
思います。
北野：友岡先生のご講演に対してのご質問です。
野生種の持っているストレス耐性とか，そういう
ものはいわゆる栽培種が持っているレベルをはる
かに超えている。その中から，そういうものを逆
に栽培化するという話，非常に感動的でした。例
えば生殖的な障壁とか，そういうものがクリアで
きれば，栽培化をしたあとに，その持っている有
用遺伝子を組み換えの方法ではなくて品種に導入
するような，そういうことを志向した研究も進め
られるわけですね。
友岡：これまでは野生種が持っている耐病性やス
トレス耐性を近縁の栽培種に導入するという方向
で研究を進めてきました。野生種の中に広い交雑
親和性を持つような種が実はあるのです。例えば，
アズキと A という栽培種は，直接は交雑できな
いけれども，その間に B という野生種をかませる。
まず栽培種 A の耐性遺伝子を野生種 B に入れて，
野生種 B からアズキに入れるというようなこと
ができる例があったのです。そういうことをもっ
と詳しく調べていけば，近縁なものから徐々に有
用遺伝子を移していくというようなことは可能性
としてはあると思っています。
西尾：私も野生の植物から栽培化に適するような
ケースを突然変異でとっていって，それを利用す
るという考えはなかなか面白いと思いました。そ
うした場合に，食経験がない作物を開発していく
ことになるわけですね。そういう問題は想定して
おられますでしょうか。
友岡：今は，野生種であっても食経験のあるもの
に限定して栽培化を考えています。幸い Vigna 属
には野生の状態で食料として利用されている種が
非常に多いということも関係しています。さらに，
アズキ亜属には 2₁ 種ありますが，その中から ₆
種が既に栽培植物になっているのです。栽培植物
にまでなっていない野生植物も大抵は野生の状態
で現地の住民は採集して食べている例が多いので
す。限界環境に生育している植物というのは結構
たくさんあると思うんですが，その中で既に食料
として利用されているような，そういう植物を対
象に栽培化をおこなっていくという戦略が，最初
はいいかなと思っています。
中川：私はマメ科の牧草をやったことがあって，
マメ科というのは，かなり毒性を持って害虫に対
抗していくというふうな戦略を持っている種が多
いですよね。Vigna 属の中で，例えば人間が食べ
て体に悪いというふうなものを持っている種もあ
るのでしょうか。そういうものだったら，多分人
間が栽培化という以前に，毒素をとって栽培化す
るかもしれませんが，その前に捨ててしまうこと
もあるでしょう。だけど，そっちのほうにひょっ
としたら面白い耐寒性などの遺伝子を持っている
可能性もありますよね。そのときに，昨日言われ
たような戦略をとったときにどうなっていくか
と。一つはそういう毒性を持つものがあるのかと
いうことと，あるならば，それをどうしていくか。
そちらのほうに非常に有用な遺伝子があって，そ
れをまず栽培化してからというようなことを考え
たときに，どういう戦略をとるか伺いたいと思い
ます。
友岡：毒性を持つものはあります。程度によりま
すが，やはりある程度インヒビター類が多い野生
種で，そのまま食べると良くないというような報
告があります。毒性のあるような野生種に有用形
質が隠れているような例を，どう拾っていくかと
いうことについてですが，食べる器官に何か毒性
があるけれども，そうじゃないところ，例えば葉っ
ぱの耐病性が強いとか耐塩性が強いとか，そうい
う有用特性については，目的とする形質ごとにス
クリーニングをしています。その場合の材料につ
いては，食経験のあるなしにかかわらず供試する
戦略をとっています。
北野：続きまして，菊地先生のご発表に対して何
かご質問等ございましたら。
久保山：ナスの単為結果性を扱われていますけれ
ども，これを明らかとした知見というのは，どの
程度の範囲で適用できるものでしょうか。その植
物ごとに分類群ごとに単為結果のメカニズムとい
うのは違う可能性があるのかどうかというあたり
をお聞きしたいのですが。
菊地：確かに植物種によって単為結果機構という
のはかなり変わってくるものもあれば，ナスに近
83
いものもあると考えています。トマトとナスは単
為結果を誘起する植物ホルモンが同じですし，あ
る程度オーバーラップしているメカニズムも多い
です。しかし単為結果機構というのは，いろいろ
な要因によって引き起こされるということが分
かっていまして，ある一つの要因だけが単為結果
を引き起こすというものではなく，いろいろな植
物ホルモン関連や，植物ホルモンとは異なる関連
遺伝子であっても，それをオーバーエクスプレス
すれば単為結果するという現象も報告されていま
す。ある程度今回の研究で分かったようなものを
トマトなりピーマンなり，同じナス科のものに導
入してやることによって，単為結果が誘起される
という可能性は十分考えられると思います。しか
し例えば，キュウリの品種のほとんどが単為結果
性であるということが分かっていますが，キュウ
リの単為結果を誘起する植物ホルモンはナスとは
異なっています。なので，キュウリにナスと同じ
ものが応用できるかどうかというような事は，今
後もうちょっと検討していかないといけないと考
えています。
吉田：三つの QTL を取得されて，それぞれを入
れた系統を作ってらしたと思うのですけれども，
A だけを入れたやつは高温に弱く，高温だと結実
しにくいということでしたが，B，C に関しては
どうだったんでしょうか。
菊地：A が一番果実肥大性に対する QTL の寄与
が高いのですが，実際はあまり着果が良くないよ
うです。B，C に対してはどうかというと，B，C
自体も単為結果親の AE-P 系統に比べると非常に
結実率というのは弱まっている傾向があります。
やはり三つの QTL が協調して働くことによって
結実率というのが高くなっている，プロモートさ
れているという現象が多く見られるようです。た
だ，C はやはり着果性の寄与が一番高いので，A
や B に比べると，着果率はいいような傾向があ
ります。
吉田：結実に対する温度の QTL というのは，こ
の三つの領域にも含まれているのでしょうか。
菊地：温度に対する QTL というよりは，実際に
この単為結果を選抜する過程で使ってきた選抜法
というのが，温度を制御して単為結果の強いもの
を拾ってきているので，温度感受性というよりは
単為結果の強弱とリンクしたものを選抜している
と思います。単為結果自体は温度以外にも光量な
どに左右され，あるいは最近分かってきたことな
のですが，CSSLs を今まで人工気象室で育ててい
たんですが，どうもそれだと結実率があまり良く
なくて，圃場やハウスに地植えにしてやると結実
率が高くなるというようなことがありました。や
はり地に下ろしてやると結実率が高くなるという
ような傾向もあるので，温度以外にもそういった
要因が，結実や果実形成に関与しているんだなと
考えています。
草場：着果しないという現象は，離層ができたり
して落ちてしまうという，そういうことでしょう
か。
菊地：そうですね。離層ができて，そこから落ち
るという現象です。
草場：そうすると，そこに ABA が関与している
のじゃないかという，そういうお話と考えていい
ですか。
菊地：実際に ABA を測ったのは離層が形成され
る果柄のところと子房と両方を測ったのですが，
両方とも上昇しています。多分離層形成のところ
に関与しているのだろうとは推察しているのです
が，今のところ，そこは詰めていないところです。
草場：理解としては，ABA 含量が下がるので受
粉しなくても，着果すると考えていいわけですね。
菊地：そうですね。
高原：正常な果実肥大の場合は受粉が刺激になっ
て果実肥大が起こるということだと思うのです
が，その受粉から果実肥大に至る情報伝達という
のは，どのくらい分かっているのでしょうか。
菊地：一般的には，受粉をすることによって種子
からオーキシンが合成されて，そのオーキシンの
シグナルによって，トマトですとジベレリンの生
合成が促進されたりします。さらにそれらの植物
ホルモンが糖代謝関連酵素遺伝子に働いて果実肥
大が引き起こされるといったような大まかなモデ
ルは提唱されています。ただ，実際にはまだよく
分かっていないことも多く，例えば，本当に種子
からオーキシンが合成されているのかどうか，そ
ういった根本的なところ自体がまだ分かっていな
いことも多いのが現状だと思います。
北野：それでは清水先生のご講演のほうに移って
84
いきたいと思います。
廣近：枝変わり変異がたくさん見つかってきて，
それが品種の育成等にも利用されているというこ
とですが，普通に考えると，その変異というのは
優性でないと表現型に現れてこないと思います。
その遺伝学的な解析をされているケースというの
はあるのでしょうか。
清水：実はそれをやった例はあるのですが，はっ
きりした結果というのが出ていません。質的な形
質に関しては後代に遺伝して，確か優性だったと
いう報告が幾つかあったと思います。ただ，ほと
んどの形質が量的な形質でありまして，評価自体
が非常に難しいものですから，実際それをやった
例というのは，特に果実形質に関してはないで
す。
廣近：ということは，遺伝子が特定されて機能が
ある程度分かったというのは，カンキツではない
ということですか。
清水：そうですね。表現型としていろいろ面白い
ものは見つかっているですが。カロチノイド系の
着色に関係するカロチノイドの代謝の遺伝子の一
つが，確かに着色系では変化しているということ
が報告としてはあります。ただ，実際それがゲノ
ムレベルでどういった変異が起こっているかとい
うことに関しては，まだ報告はなかったと思いま
す。
髙品：山形県園芸試験場の髙品です。例えば，枝
変わりの木についてクローンで増やしたものが
あって，その元品種もまた別のラインでクローン
で増やしたものがあって，それと元の木の多型比
較をすると，果樹の場合ライフサイクルが長いの
で変異が蓄積していると思うのですが，そこの差
を拾うと原因遺伝子などの当たる確率とかは，結
構高まりますでしょうか。それともやはりまだ難
しいテクニックで，原因を多型を検出するのは難
しいでしょうか。
清水：なかなか難しいご質問ですが，確かにおっ
しゃるとおりです。見つかって直後のものを探す
というのが，やはり一番いいだろうと思っていま
す。ただ，実際にそれをやる中で，やはりキメラ
であるということをどう評価するかということが
実際問題非常に難しく思います。それを解消する
ためには，カンキツの場合ですと，例えば珠心胚
実生のようなクローン胚を作って，後代を作って，
そこで順化してみるということが研究としては必
要だろうと思っています。ただ，それをやると実
際に果実で見られたもともとの枝変わりの形質の
変化というものが，本当に後代で再現したかとい
うことを確認しないといけないわけです。それに
は少なくとも ₇ ～ ₁₀ 年かかるということで，そ
れぐらいのスパンでやってしまうと，その間に一
定の確率で突然変異が蓄積していくということも
考えられます。非常にクリアな違いであって，な
おかつ全層が変異したようなものでないとなかな
か明快な結論というのは出せないのかなと思って
います。ただ，一方で今のゲノムレベルの話です
が，発現レベルの解析であれば，そこはある程度
キメラ性があったとしても，ある程度はそういっ
たものをキャンセルして解析する，遺伝的なバッ
クグラウンドはほとんど同じですので，そういっ
たところは無視して解析できるのではないかと
思っています。それを両方やっていくというのが，
現時点でカンキツでできる方法なのかなと考えて
います。
北野：最後の岩崎先生のご発表に対して何かご質
問等ございましたら。
久保山：粒形のこととは違うのですが，G タンパ
ク質を上昇させると耐病性が上昇するという話で
すが，これはいわゆる全身獲得抵抗性のようなも
のが上昇するというお話と思います。その辺のメ
カニズムというのは，少し分かったようなことが
あるのでしょうか。
岩崎：まだ断定できる状態ではないのですが，
PR タンパク質などが若干上がっているという傾
向があります。そのことが実際の抵抗性にリンク
しているかどうかというのは，まだわかりません。
北野：全体にかかわった話をしないといけないと
思いますが，そういう観点から特に今回のシンポ
ジウムのテーマにかかわって，どなたから何かコ
メントを頂けたらと思います。あるいはどなたか
の講演者に対して，さらに聞いてみたいというこ
とがございましたら，ぜひ積極的にご発言をお願
いしたいと思います。
廣近：長戸先生にお聞きしたいのですが，遺伝学
的なアプローチで様々な遺伝子を見つけて，遺伝
子なり遺伝子の機能を見つけられていますが，そ
85
の遺伝学的なアプローチの限界というのもあるか
と思います。要するに，アプローチで見落とすも
のがあり，実は非常にキーとなるものを結構見落
としているのかもしれないと思われているのか，
もうこのまま進めば理解できるだろうというふう
に考えられているのか，ご意見の形で伺えれば幸
いです。
長戸：例えば，変異体に関しては，要するに重複
した遺伝子に関しては多分とれないですね。偶然
にとれる可能性はもちろんあります。でも，それ
以外のシングルコピーの遺伝子なら，多分とれる
と思います。あまり今やっていないのは遺伝子間
のネットワークの話です。相互作用の話は一部始
めてますけれども，ほとんどやっていません。多
分それは非常に重要な問題で，農業形質としても
非常に重要なので，その辺をどうやって効率的に
うまく進めるかというのがこれからの問題だろう
と思っています。
北野：どうもありがとうございました。私の今回
の発表を聞いての感想ですが，大学の基礎研究か
ら始めて応用研究へという流れと，現場の重要な
テーマを解決するために，それを進めていくとい
う両側面の研究があるし，皆様方はそのどちら
か，あるいは両方にかかわって進められていると
思います。今回の話，例えば長戸先生の話は研究
は研究というようなところからスタートしたよう
なところもあって，いろいろなことが分かってく
ると，そういうものの中に実用的なものが存在し
ていたということを思いました。例えば，遺伝子
がとれたとしても，それがどういうふうに位置づ
いて，なぜそうなるかということにはなかなかつ
ながらないだろうし，効果としてもドラスチック
なものが一方であるから，その微妙な違い，例え
ば数％上げるとか，₁₀％上げるとかという，育種
のところでの重要な仕組みみたいなものを見る上
ですごく役に立つとか，そういうようなことにつ
ながってきているのかなと考えられます。高木先
生のやっておられるような流れの研究というの
は，今のネットワークの話とすごく関係していて
いるように思えました。我々は一つのミュータン
トをとったら，それを突破口にしてそのリンケー
ジを見ていこうという発想で一本釣りみたいなと
ころがありますが，高木先生がやられる話は地引
き網のようにしてネットワーク全部を見ていくよ
うな，そういう流れの研究が進みつつあるという
ことを感じました。これからいろんな方向の研究
はもちろんあると思うのですけれども。そういう
基礎科学と実学，それが新しい育種の方法にもつ
ながったり，あるいは有用突然変異を見つけると
きのヒントにもなったり，つながっていったり，
私としてはそういうきっかけを与えていただいた
というような気がしております。

第3回スポーツ講座 - 岡山大学スポーツ教育センター

キジツ（枳実）

寒い季節の馬体管理

説明練習

フルーツセーフティ 新・輸入果物図鑑 【メロゴールド】 【学 名】 Citrus

ジャズディナーの夕宴

タイの価格統制品リスト - DATAMAX CO.,LTD.

∼世界のお米料理∼

シンガポールにおける 輸入米・米加工品の規制等に 関する実態について