Maternal control of seed size by EOD3/CYP78A6 in Arabidopsis thaliana

Article (PDF Available)inThe Plant Journal 70(6):929-39 · January 2012with71 Reads
DOI: 10.1111/j.1365-313X.2012.04907.x · Source: PubMed
Abstract
Seed size in higher plants is coordinately determined by the growth of the embryo, endosperm and maternal tissue, but relatively little is known about the genetic and molecular mechanisms that set final seed size. We have previously demonstrated that Arabidopsis DA1 acts maternally to control seed size, with the da1-1 mutant producing larger seeds than the wild type. Through an activation tagging screen for modifiers of da1-1, we have identified an enhancer of da1-1 (eod3-1D) in seed size. EOD3 encodes the Arabidopsis cytochrome P450/CYP78A6 and is expressed in most plant organs. Overexpression of EOD3 dramatically increases the seed size of wild-type plants, whereas eod3-ko loss-of-function mutants form small seeds. The disruption of CYP78A9, the most closely related family member, synergistically enhances the seed size phenotype of eod3-ko mutants, indicating that EOD3 functions redundantly with CYP78A9 to affect seed growth. Reciprocal cross experiments show that EOD3 acts maternally to promote seed growth. eod3-ko cyp78a9-ko double mutants have smaller cells in the maternal integuments of developing seeds, whereas eod3-1D forms more and larger cells in the integuments. Genetic analyses suggest that EOD3 functions independently of maternal factors DA1 and TTG2 to influence seed growth. Collectively, our findings identify EOD3 as a factor of seed size control, and give insight into how plants control their seed size.

Figures

Figure
Figure
Figure
Figure

Full-text (PDF)

Available from: Wenjuan Fang, Mar 23, 2015
Maternal control of seed size by EOD3/CYP78A6 in
Arabidopsis thaliana
Wenjuan Fang
, Zhibiao Wang
, Rongfeng Cui, Jie Li and Yunhai Li
*
State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese
Academy of Sciences, Beijing 100101, China
Received 27 November 2011; revised 9 January 2012; accepted 12 January 2012; published online 2 April 2012.
*
For correspondence (e-mail yhli@genetics.ac.cn).
These authors contributed equally to this work.
SUMMARY
Seed size in higher plants is coordinately determined by the growth of the embryo, endosperm and maternal
tissue, but relatively little is known about the genetic and molecular mechanisms that set final seed size. We
have previously demonstrated that Arabidopsis DA1 acts maternally to control seed size, with the da1-1
mutant producing larger seeds than the wild type. Through an activation tagging screen for modifiers of da1-1,
we have identified an enhancer of da1-1 (eod3-1D) in seed size. EOD3 encodes the Arabidopsis cytochrome
P450/CYP78A6 and is expressed in most plant organs. Overexpression of EOD3 dramatically increases the seed
size of wild-type plants, whereas eod3-ko loss-of-function mutants form small seeds. The disruption of
CYP78A9, the most closely related family member, synergistically enhances the seed size phenotype of eod3-
ko mutants, indicating that EOD3 functions redundantly with CYP78A9 to affect seed growth. Reciprocal cross
experiments show that EOD3 acts maternally to promote seed growth. eod3-ko cyp78a9-ko double mutants
have smaller cells in the maternal integuments of developing seeds, whereas eod3-1D forms more and larger
cells in the integuments. Genetic analyses suggest that EOD3 functions independently of maternal factors DA1
and TTG2 to influence seed growth. Collectively, our findings identify EOD3 as a factor of seed size control, and
give insight into how plants control their seed size.
Keywords: EOD3/CYP78A6,CYP78A9,DA1, seed size, the integument.
INTRODUCTION
Seed size is a key determinant of evolutionary fitness in
plants, and is also an important agronomic trait in crop
domestication (Orsi and Tanksley, 2009). Several studies
suggest that seedlings of large-seeded plants are better able
to tolerate many of the stresses encountered during seedling
establishment, whereas small-seeded plants are considered
to have superior colonization abilities because they produce
large numbers of seeds (Westoby et al., 2002; Moles et al.,
2005). At the same time, seed size is negatively associated
with the number of seeds produced by a plant because of the
limited resources of the mother plant (Harper et al., 1970).
Scientific interest in seed size relates not only to its impor-
tance in plant fitness, but also to crop domestication. Crops
domesticated for consumption of their seeds (e.g. rice and
wheat) often produce seeds significantly larger than their
wild ancestors (Fan et al., 2006; Song et al., 2007; Gegas
et al., 2010).
A seed consists of three major components, the embryo,
the endosperm and the seed coat, which originate from
different cells of the ovule and possess different comple-
ments of maternal and paternal genomes. In angiosperms,
seed development involves a double-fertilization process in
which one sperm nucleus fuses with the egg to produce the
diploid embryo, whereas the other sperm nucleus fuses with
two polar nuclei to form the triploid endosperm (Lopes and
Larkins, 1993). The seed coat differentiates after fertilization
from maternally derived integuments. The embryo is sur-
rounded by the endosperm, which, in turn, is enclosed
within the maternal seed coat. Therefore, the size of a seed is
determined by the coordinated growth of maternal sporo-
phytic and zygotic tissues.
The size of seeds is influenced by a variety of cellular
processes. Seed size is known to be influenced by parent-
of-origin effects. The cross between a diploid female parent
and a tetraploid male parent produces larger F
1
seeds,
whereas the reciprocal cross generates smaller F
1
seeds,
suggesting that a maternal or paternal excess of genome
has a dramatic effect on seed size (Scott et al., 1998). Similar
ª2012 The Authors 929
The Plant Journal ª2012 Blackwell Publishing Ltd
The Plant Journal (2012) 70, 929–939 doi: 10.1111/j.1365-313X.2012.04907.x
Advertisement:
to interploidy crosses, crosses between the wild type and
met1 mutant with hypomethylated genomes show that
larger F
1
seeds are generated when the maternal parent is
met1, whereas smaller F
1
seeds are produced when the
paternal parent is met1 (Xiao et al., 2006), suggesting that
parent-of-origin effects may involve DNA methylation. In
addition, the size of seeds is affected by the maternal and/or
zygotic tissues. Several factors that influence seed size via
the zygotic tissues have been recently identified in Arabi-
dopsis. haiku (iku) and miniseed3 (mini3) mutants form
small seeds because of the reduced growth and early
cellularization of the endosperm (Garcia et al., 2003; Luo
et al., 2005). IKU1,IKU2 and MINI3 function in the same
pathway to promote endosperm growth in Arabidopsis
(Garcia et al., 2003; Luo et al., 2005; Wang et al., 2010).
SHORT HYPOCOTYL UNDER BLUE 1 (SHB1) associates with
both MINI3 and IKU2 promoters in vivo, and may act with
other proteins that bind to MINI3 and IKU2 promoters to
promote endosperm growth in the early phase of seed
development (Zhou et al., 2009). Seed size is also influenced
by maternal tissues. Several factors that act in maternal
tissues to influence seed size have been isolated. Arabidop-
sis TRANSPARENT TESTA GLABRA 2 (TTG2) promotes
seed growth by increasing cell expansion in the integuments
(Garcia et al., 2005; Ohto et al., 2009). APETALA 2 (AP2) may
restrict seed growth by limiting cell expansion in the
integuments (Jofuku et al., 2005; Ohto et al., 2005, 2009).
By contrast, AUXIN RESPONSE FACTOR 2 (ARF2) and the
predicted ubiquitin receptor DA1 limit seed size by restrict-
ing cell proliferation in the integuments (Schruff et al., 2006;
Li et al., 2008). However, CYP78A5/KLU promotes seed
growth by increasing cell proliferation in the integuments
of ovules (Adamski et al., 2009). Therefore, the integument
or seed coat plays a key role in the maternal control of seed
size. In addition, many quantitative trait loci (QTLs) for seed
size have been mapped in Arabidopsis and in crops (Alonso-
Blanco et al., 1999; Li et al., 2004; Fan et al., 2006; Song
et al., 2007; Shomura et al., 2008; Weng et al., 2008). Three
grain-size QTLs have been recently cloned in rice, including
GS3,GW2 and qSW5/GW5 (Fan et al., 2006; Song et al.,
2007; Shomura et al., 2008; Weng et al., 2008). However, it is
not clear whether these three factors act in maternal and/or
zygotic tissues in rice.
Despite the importance of seed size, relatively little is
known about the genetic and molecular mechanisms that
control seed size. Here, we describe a dominant enhancer of
da1-1 (eod3-1D), which forms larger seeds than da1-1.EOD3
encodes the cytochrome P450/CYP78A6, and overexpres-
sion of CYP78A6 dramatically increases the seed size of wild-
type plants. An analysis of double eod3-ko cyp78a9-ko
mutants shows that EOD3 acts redundantly with CYP78A9
to control seed size. EOD3 acts maternally to promote seed
growth, but functions independently of maternal factors
DA1 and TTG2 to influence seed size. Our findings identify
EOD3 as a factor of seed size control, and may open future
opportunities for modulating seed size in crop plants.
RESULTS
eod3-1D enhances the seed size phenotype of da1-1
We previously characterized the Arabidopsis da1-1 mutant,
which has larger seeds than the wild type (Li et al., 2008).
DA1, encoding a predicted ubiquitin receptor, sets final seed
size by restricting cell proliferation (Li et al., 2008). To identify
other components in the DA1 pathway or additional factors
of seed size control, we initiated a T-DNA activation-tagging
screen in a da1-1 homozygous genetic background. Seeds
produced from approximate 16 000 T
1
plants were screened
for mutations affecting the seed size phenotype of da1-1.
A dominant enhancer of da1-1 (eod3-1D), which enhanced
the seed size phenotype of da1-1, was identified (Fig-
ure 1A,D). Seeds of the eod3-1D da1-1 double mutant were
dramatically larger and heavier than those of the da1-1
mutant (Figure 1D,E). The embryo constitutes the major
volume of a mature seed in Arabidopsis. The size of eod3-
1D da1-1 embryos was substantially increased, compared
with that of Col-0 and da1-1 embryos (Figure 1B). The chan-
ges in seed size were also reflected in the size of seedlings
(Figure 1C). Cotyledons of eod3-1D da1-1 seedlings were
significantly larger than those of Col-0 and da1-1 seedlings
(Figure 1C,F). In addition, the eod3-1D da1-1 double mutant
had larger flowers and leaves than da1-1 (Figure S1).
(A)
(B)
(D) (E) (F)
(C)
Figure 1. Isolation of an enhancer of da1-1 (eod3-1D).
(A) Seeds from wild-type, da1-1 and eod3-1D da1-1 plants (from left to right).
(B) Mature embryos of wild type, da1-1 and eod3-1D da1-1 (from left to right).
(C) Ten-day-old seedlings of wild type, da1-1 and eod3-1D da1-1 (from left to
right).
(D) Projective area of wild-type, da1-1 and eod3-1D da1-1 seeds.
(E) Weight of wild-type, da1-1 and eod3-1D da1-1 seeds.
(F) Cotyledon area of 10-day-old wild-type, da1-1 and eod3-1D da1-1 seed-
lings.
Values in (D–F) are given as means SEs, relative to the respective wild-type
values, set at 100%. Scale bars: A, B, 0.5 mm; C, 5 mm.
930 Wenjuan Fang et al.
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
eod3-1D sets large seeds
To determine whether the single eod3-1D mutant has an
altered seed size, we identified the eod3-1D mutant among
F
2
progeny derived from a cross between the eod3-1D da1-1
double mutant and the wild type (Col-0). Seeds produced by
eod3-1D were larger and heavier than the wild-type seeds
(Figure 2A,E,F). In addition to the seed phenotype, eod3-1D
plants showed larger flowers and leaves, thicker stems and
higher plants than the wild type (Figures 2B and S2A;
Table S1). However, the number of rosette and cauline
leaves was similar in the wild type and eod3-1D, and the
number of rosette and cauline branches in eod3-1D was also
comparable with that in the wild type (Table S1).
The eod3-1D mutation also caused defects in reproductive
development. For example, the eod3-1D mutant produced
fewer elongated siliques than the wild type (Table S1). First,
several flowers on the primary inflorescences of eod3-1D did
not open normally (Figure S2B,C). Their stamens were much
shorter than those of the wild type (Figure S2B,C). The
dehiscence of eod3-1D anthers was much delayed (Fig-
ure S2C), but their pollens were functional (Figure S2D,E).
The enlarged siliques were more frequently observed on the
latest flowers of old plants. In general, the enlarged siliques
contained few seeds although the number of ovules per
silique in eod3-1D was not reduced (Table S1). We observed
that carpels of the late-developing eod3-1D flowers were
longer than those of wild-type flowers, whereas the length of
stamens was similar to that of wild-type stamens (Fig-
ures 2C and S2A), such that eod3-1D pollen is not able to
directly reach stigmatic papillae; this could, in part, explain
the decreased fertility. Fully elongated eod3-1D mutant
siliques were longer and wider than wild-type siliques
(Figure 2D).
To determine whether the large seed size phenotype
could result from an allocation of extra resources to the few
seeds produced, we hand-pollinated six flowers on primary
inflorescences of wild-type plants, eod3-1D, and a male-
sterile mutant (CS4002). For this set of experiments, flowers
were pollinated with pollens of the same genotypes, with the
exception of male-sterile plants for which wild-type pollens
were used. Thus, each male-sterile plant produced only six
siliques. The average seed size from male-sterile maternal
plants was increased to 116% of that from wild-type mater-
nal plants (Figure 2G), indicating that seed size increased
under conditions of reduced fertility. By contrast, the aver-
age seed size from the eod3-1D mutant was approximately
170% that of the wild type (Figure 2G), indicating that the
effect of eod3-1D on seed size is not primarily a result of its
effect on fertility.
EOD3 encodes a cytochrome P450 monooxygenase
To test whether this T-DNA insertion might cause the
eod3-1D phenotypes, we analyzed the genetic linkage of the
mutant phenotype with Basta resistance, which is conferred
by the selectable marker of the activation-tagging vector
(Fan et al., 2009). All 101 plants with eod3-1D da1-1 pheno-
types in the T
2
population were resistant, whereas the 36
plants with da1-1 phenotypes were sensitive, suggesting
that the insertion is responsible for the eod3-1D mutation. To
identify the EOD3 gene, the DNA flanking the T-DNA inser-
tion was isolated by thermal asymmetric interlaced PCR (Liu
et al., 1995). Sequence analysis indicated that the insertion
was in an intergenic region on chromosome II between the
genes At2g46660 and At2g46670. The T-DNA had inserted
approximately 3.2 kb upstream of the At2g46660 gene, and
about 6.5 kb downstream of the At2g46670 gene
(Figure 3A). The mRNA levels of these two genes were
determined by reverse transcription-polymerase chain
reaction (RT-PCR). Expression levels of the At2g46670 gene
were similar in da1-1 and eod3-1D da1-1 plants (Figure 3B),
suggesting that At2g46670 was unlikely to be the EOD3
gene. The mRNA of At2g46660 accumulated at a higher level
in eod3-1D da1-1 than in da1-1 (Figure 3B), strongly
suggesting that At2g46660 is likely to be the EOD3 gene. To
demonstrate that this gene corresponded to EOD3,we
overexpressed the At2g46660 gene in Col-0 wild-type plants
and isolated 41 transgenic plants. Most transgenic plants
(A)
(E) (F) (G)
(C) (D)
(B)
Figure 2. Seed and organ size in the eod3-1D mutant.
Seeds (A), flowers (B), stamens and carpels (C), and siliques (D) of wild-type
(left) and eod3-1D (right) plants.
(E) Projective area of wild-type and eod3-1D seeds.
(F) Weight of wild-type and eod3-1D seeds.
(G) Projective area of Col-0 ·Col-0 F
1
, CS4002 ·Col-0 F
1
and eod3-
1D ·eod3-1D F
1
seeds.
Values in (E–G) are given as means SEs, relative to the respective wild-type
values, set at 100%.
Scale bars: A, B, C, D, 1 mm.
EOD3/CYP78A6 and CYP78A9 and seed size 931
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
showed large seeds and increased plant height (Fig-
ures 3C,D and S3A), as had been seen in the eod3-1D single
mutant, confirming that At2g46660 is the EOD3 gene.
Importantly, the 35S::EOD3#7 transgenic plants exhibited
normal growth and fertility, but produced significantly larger
seeds compared with the wild type (Figure 3C and S3B,C).
The EOD3 gene encodes the putative cytochrome P450
monooxygenase CYP78A6, one of six members of the
CYP78A family in Arabidopsis. Genes in the CYP78A class
belong to the group-A cytochrome P450 in plants and seem
to perform plant-specific functions (Zondlo and Irish, 1999;
Ito and Meyerowitz, 2000; Anastasiou et al., 2007). EOD3/
CYP78A6 exhibits the highest similarity to Arabidopsis
CYP78A9 (Figure 3E) (Ito and Meyerowitz, 2000).
EOD3/CYP78A6 acts redundantly with CYP78A9
to control seed size
In order to further understand the function of EOD3,we
isolated T-DNA inserted loss-of-function mutants for EOD3/
CYP78A6 and CYP78A9, the most closely related family
member. eod3-ko1 and eod3-ko2 were identified with T-DNA
insertions in the first and second exons of the EOD3/
CYP78A6 gene, respectively (Figure 4A). cyp78a9-ko1 had
T-DNA insertion in the second exon of CYP78A9 (Figure 4B).
The T-DNA insertion sites were confirmed by PCR using
T-DNA-specific and flanking primers and sequencing PCR
products (Figure S4). The eod3-ko1,eod3-ko2 and cyp78a9-
ko1 mutants were further backcrossed into Col-0 three times.
Seeds from eod3-ko1,eod3-ko2 and cyp78a9-ko1 mutants
were smaller and lighter than seeds from wild-type plants
(Figure 4F,G). Silique length in eod3-ko1,eod3-ko2 and
cyp78a9-ko1 was reduced, compared with that in the wild
type (Figure 4D,H). By contrast, the size of leaves and petals,
stem thickness and plant height in eod3-ko1 and cyp78a9-
ko1 were comparable with those in wild type (Table S1). In
addition, the number of rosette and cauline leaves, rosette
and cauline branches, siliques per plant and ovules per
silique in eod3-ko1 and cyp78a9-ko1 was similar to that in
the wild type (Table S1). As EOD3/CYP78A6 shows the
highest similarity to the Arabidopsis CYP78A9, we postu-
lated that EOD3 may act redundantly with CYP78A9 to
control seed size. To test this, we generated the double
knock-out mutants eod3-ko1 cyp78a9-ko1 and eod3-ko2 -
cyp78a9-ko1. The seed size and weight phenotype of eod3-
ko mutants was synergistically enhanced by the disruption
of CYP78A9 (Figure 4F,G), suggesting that EOD3 functions
redundantly with CYP78A9 to control seed growth. The
eod3-ko cyp78a9-ko mutations also caused a significant
change in seed shape (Figure 4C). eod3-ko cyp78a9-ko
seeds were shorter than wild-type seeds, whereas seed
width was comparable with that of the wild type
(Figure 4C,I,J), indicating that eod3-ko cyp78a9-ko seeds
are more round in shape than wild-type seeds. eod3-ko
cyp78a9-ko produced fewer siliques per plant than the wild
type (Table S1). The length of siliques in eod3-ko cyp78a9-
ko was dramatically reduced, compared with their parental
lines (Figure 4D,H). Surprisingly, the number of ovules per
silique in eod3-ko1 cyp78a9-ko1 was similar to that in the
wild type, resulting in a higher density of seeds within
siliques (Figure 4E; Table S1). In addition, the primary
inflorescence stem of eod3-ko1 cyp78a9-ko1 was shorter
than that of the wild type, and the size of petals and leaves
was slightly reduced compared with the wild type (Ta-
ble S1). However, the number of leaves and branches in
eod3-ko1 cyp78a9-ko1 was comparable with that observed
in the wild type (Table S1).
EOD3 acts maternally to influence seed size
To obtain clues about the genetic control of seed size, we
asked whether EOD3 functions maternally or zygotically.
(A) (D)
(E)
(B)
(C)
Figure 3. Cloning of the EOD3 gene.
(A) Structure of the T-DNA insertion in the eod3-
1D mutant.
(B) Expression levels of At2g46660 (EOD3) and
At2g46670 in da1-1 and eod3-1D da1-1 seed-
lings.
(C) Projective area of wild-type, 35S::EOD3#7,
35S::EOD3#9 and eod3-1D seeds.
(D) Expression levels of EOD3 in wild-type,
35S::EOD3#7,35S::EOD3#9 and eod3-1D seed-
lings.
(E) Phylogenetic tree of the CYP78A family
members in Arabidopsis thaliana.
Values in (C, D) are given as means SE, relative
to the wild-type value, set at 100%. **P< 0.01
compared with the wild type (Student’s t-test).
932 Wenjuan Fang et al.
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
Reciprocal cross experiments between the wild type and
eod3-ko1 cyp78a9-ko1 were performed. The effect of eod3-
ko1 cyp78a9-ko1 on seed size was observed only when the
eod3-ko1 cyp78a9-ko1 mutant acted as the maternal plant.
Seeds produced by an eod3-ko1 cyp78a9-ko1 mother,
regardless of the genotype of the pollen donor, were con-
sistently smaller than those produced by maternal wild-type
plants, and eod3-ko1 cyp78a9-ko1 mutant pollen in a wild-
type mother produced seeds of wild-type size (Figure 5A).
This indicates that eod3-ko1 cyp78a9-ko1 can act maternally
to control seed size. We further performed reciprocal cross
experiments between the wild type and eod3-1D. Pollinating
wild-type plants with eod3-1D pollen leads to the develop-
ment of eod3-1D/+ embryos within a wild-type seed coat.
However, the size of the resulting seeds was comparable
with that of self-pollinated wild-type seeds (Figure 5B). In
contrast, we could not observed the wild-type sized seeds
from eod3-1D/+ plants fertilized with wild-type pollen,
although half of them contained wild-type embryos. We
further measured the size of individual seeds from eod3-1D/
+plants fertilized with wild-type pollen and genotyped the
eod3-1D mutation. Our results show that the eod3-1D
mutation is not associated with variation in the size of these
seeds (Figure S5). Together, these analyses indicate that the
embryo and endosperm genotype for EOD3 do not influence
seed size, and EOD3 is required in the sporophytic tissue of
the mother plant to promote seed growth.
(A) (F)
(G)
(H)
(B)
(I) (J)
(C) (D) (E)
Figure 4. EOD3 acts redundantly with CYP78A9
to influence seed size.
(A) EOD3 gene structure. The start codon (ATG)
and the stop codon (TAA) are indicated. Closed
boxes indicate the coding sequence, and the line
between the boxes indicates an intron. T-DNA
insertion sites (eod3-ko1 and eod3-ko2) in the
EOD3 gene are indicated.
(B) CYP78A9 gene structure. The start codon
(ATG) and the stop codon (TAA) are indicated.
Closed boxes indicate the coding sequence, and
the line between boxes indicates an intron. The
T-DNA insertion site (cyp78a9-ko1) in the
CYP78A9 gene was shown.
(C) Seeds from wild-type, eod3-ko1 cyp78a9-ko1
and eod3-ko2 cyp78a9-ko1 plants (from top to
bottom).
(D) Siliques from wild-type, eod3-ko1,eod3-ko2,
cyp78a9-ko1,eod3-ko1 cyp78a9-ko1 and eod3-
ko2 cyp78a9-ko1 plants (from left to right).
(E) Opened siliques from wild-type and eod3-
ko1 cyp78a9-ko1 plants (from left to right).
(F) Projective area of wild-type, eod3-ko1,eod3-
ko2,cyp78a9-ko1,eod3-ko1 cyp78a9-ko1 and
eod3-ko2 cyp78a9-ko1 seeds.
(G) Weight of wild-type, eod3-ko1,eod3-ko2,
cyp78a9-ko1,eod3-ko1 cyp78a9-ko1 and eod3-
ko2 cyp78a9-ko1 seeds.
(H) Silique length of wild-type, eod3-ko1,eod3-
ko2,cyp78a9-ko1,eod3-ko1 cyp78a9-ko1 and
eod3-ko2 cyp78a9-ko1 seeds.
Seed length (I) and seed width (J) of wild type,
eod3-ko1,cyp78a9-ko1 and eod3-ko1 cyp78a9-
ko1 plants.
Values in (F–J) are given as means SEs, rela-
tive to the respective wild-type values, set at
100%. *P< 0.05 and **P< 0.01, compared with
the wild type (Student’s t-test).
Scale bars: C, D, E, 1 mm.
(A) (B)
Figure 5. EOD3 acts maternally to control seed size.
(A) Projective area of Col-0 ·Col-0 F
1
, Col-0 ·eod3-ko1 cyp78a9-ko1(d) F
1
,
eod3-ko1 cyp78a9-ko1(d) ·eod3-ko1 cyp78a9-ko1(d) F
1
and eod3-ko1 cy-
p78a9-ko1(d) ·Col-0 F
1
seeds.
(B) Projective area of Col-0 ·Col-0 F
1
, Col-0 ·eod3-1D F
1
,eod3-1D/+ ·eod3-
1D/+ (e/e) F
1
and eod3-1D/+ ·Col-0 (e/c) F
1
seeds.
Values (A and B) are given as means SEs, relative to the respective wild-
type values, set at 100%.
EOD3/CYP78A6 and CYP78A9 and seed size 933
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
eod3-ko1 cyp78a9-ko1 reduces cell expansion
in the integuments of developing seeds
The reciprocal crosses indicate that EOD3 acts maternally to
influence seed growth. The integuments surrounding the
ovule form the seed coat after fertilization, which may phys-
ically restrict seed growth. The integument size of ovules is
known to influence seed size (Garcia et al., 2005; Schruff et
al., 2006). We therefore asked whether EOD3 functions
through the maternal integument to affect seed size. To test
this, we characterized mature ovules from the wild type and
eod3-ko1 cyp78a9-ko1 at 2 days after emasculation. Sur-
prisingly, the size of eod3-ko1 cyp78a9-ko1 ovules was not
significantly altered, compared with that of the wild-type
ovules (Figures 6A and S6). We further investigated the outer
integument length of wild-type and eod3-ko1 cyp78a9-ko1
seeds at specific times after pollination. The size of wild-type
and eod3-ko1 cyp78a9-ko1 outer integuments showed a
significant difference at 2 days after pollination (DAP), and at
subsequent time points (Figure 6B). A previous study
showed that the integument of a developing seed could
completely stop cell division at 4 days after pollination (Gar-
cia et al., 2005). To assess the contribution of cell proliferation
and cell expansion in the integuments of developing seeds in
eod3-ko1 cyp78a9-ko1, we measured the outer integument
cell number and cell size at 6 DAP. The outer integument cell
number in eod3-ko1 cyp78a9-ko1 was similar to that in the
wild type (Figure 6C), whereas cells in eod3-ko1 cyp78a9-ko1
outer integuments were significantly smaller than those in
wild-type outer integuments (Figure 6D). These results indi-
cate that eod3-ko1 cyp78a9-ko1 restricts cell expansion in the
integuments of developing seeds.
eod3-1D promotes both cell proliferation and cell expansion
in the integuments
As the gain-of-function eod3-1D mutant had large seeds, we
further asked whether the eod3-1D mutant affects the
integument size of ovules and developing seeds. The size of
eod3-1D ovules was significantly larger than wild-type
ovules (Figures 6A and S6). eod3-1D also had dramatically
larger outer integuments than the wild type during the
whole process of seed development (Figure 6B). We further
investigated the outer integument cell number and cell size
of developing seeds at 6 DAP and found that eod3-1D had
more and larger outer integument cells than the wild type
(Figure 6C, D).
Effects of eod3-ko1 cyp78a9-ko1 and eod3-1D mutations on
embryo development
eod3-ko1 cyp78a9-ko1 and eod3-1D had smaller and larger
seed coats, respectively. The maternal integument or seed
coat acts as a physical constraint on embryo growth. We
therefore investigated whether eod3-ko1 cyp78a9-ko1 and
eod3-1D integuments could indirectly influence embryo
development. To test this, we manually pollinated wild-type,
eod3-ko1 cyp78a9-ko1 and eod3-1D plants with their own
pollen grains and examined developing embryos at specific
times after pollination. In the siliques of wild-type plants, the
majority of embryos reached the globular stage at 2 DAP,
the heart and torpedo stages at 4 DAP, the bent-cotyledon
stage at 6 DAP and the stage of the fully filled seed cavity
from 10 DAP onwards (Table S2). The developmental pro-
gression of eod3-ko1 cyp78a9-ko1 embryos was almost
similar to that of the wild type. However, the morphological
development of eod3-1D embryos was slightly slower than
the wild type at 4 DAP. At 6 DAP, most embryos reached the
bent-cotyledon stage, as seen in wild-type plants (Table S2).
This phenomenon of embryo development has also been
observed in other Arabidopsis mutants (Schruff et al., 2006;
Ohto et al., 2009; Zhou et al., 2009). Interestingly, the
majority of wild-type embryos fully filled the seed cavity at
12 DAP, whereas most eod3-1D embryos completely filled
the seed cavity at 14 DAP. It is plausible that eod3-1D forms a
larger seed cavity than the wild type; therefore, eod3-1D
embryos need to grow for a longer period of time to fill the
large seed cavity than wild-type embryos.
Effects of eod3-ko1 cyp78a9-ko1 and eod3-1D mutations
on embryo cell number and cell size
We isolated and visualized embryos from mature eod3-
ko1 cyp78a9-ko1 and eod3-1D seeds. eod3-ko1 cyp78a9-ko1
embryos were significantly smaller than those of the wild
(A) (B)
(C) (D)
Figure 6. Cell size and cell number in the integuments of wild-type, eod3-
ko1 cyp78a9-ko1 and eod3-1D developing seeds.
(A) Mature ovule perimeter.
(B) The outer integument length at specific times after pollination, as
measured from the insertion point at the funiculus to the tip at the micropyle.
(C) The number of cells in the outer integument at 6 DAP.
(D) Average length of cells in the outer integument at 6 DAP, calculated from
the outer integument length and cell number for individual seeds.
Values in (A–D) are given as means SEs. **P< 0.01 compared with the wild
type (Student’s t-test).
934 Wenjuan Fang et al.
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
type, whereas eod3-1D produced large mature embryos
compared with the wild type (Figure 7A). The average cot-
yledon area of eod3-ko1 cyp78a9-ko1 and eod3-1D embryos
was about 72 and 196% that of wild-type embryos, respec-
tively (Figure 7B). The size of embryos is determined by both
cell number and cell size. We measured palisade cells in the
central regions of wild-type, eod3-ko1 cyp78a9-ko1 and
eod3-1D cotyledons to learn which parameter is affected.
The average size of eod3-ko1 cyp78a9-ko1 cotyledon cells
was 79% that of wild-type cotyledon cells, whereas the
average size of eod3-1D cotyledon cells was 1.36-fold that of
the wild-type cotyledon cells (Figure 7C). The magnitude of
the changes in the areas of eod3-ko1 cyp78a9-ko1 and wild-
type cotyledons (0.72 times) closely parallels the differences
in the areas of cotyledon cells (0.79 times), suggesting that
eod3-ko1 cyp78a9-ko1 mainly affects embryo cell expan-
sion. Given the differences in the areas of eod3-1D and
wild-type cotyledons (1.96 times) and cells (1.36 times), we
conclude that eod3-1D had approximately 1.44 times more
cells than the wild type (1.96/1.36 = 1.44). These results
indicate that eod3-ko1 cyp78a9-ko1 formed small embryos
as a result of the reduced embryo cell expansion, and that
eod3-1D had large embryos because of increases in both
embryo cell proliferation and cell expansion. Thus, EOD3
could act maternally to influence embryo cell proliferation
and cell expansion because EOD3 is solely required in the
sporophytic tissue of the mother plant to control seed
growth (Figure 5).
Expression pattern of EOD3/CYP78A6
To examine the expression pattern of EOD3, RT-PCRs were
performed with total RNA from various tissues with EOD3-
specific primers, including roots, stems, leaves, seedlings
and inflorescences. EOD3 mRNA can be detected in all
plant organs tested (Figure 8A). To monitor the expression
pattern of EOD3 during development, the pEOD3::GUS
vector was constructed and transformed into wild-type
plants. Tissues at different development stages were
stained with GUS solution. In 14-day-old seedlings, GUS
(A)
(B) (C)
Figure 7. Cell size and cell number in cotyledons of mature wild-type, eod3-
ko1 cyp78a9-ko1 and eod3-1D embryos.
(A) Mature embryos of wild-type, eod3-ko1 cyp78a9-ko1 and eod3-1D plants.
(B) Cotyledon area of wild-type, eod3-ko1 cyp78a9-ko1 and eod3-1D embryos.
(C) Average area of palisade cells in cotyledons of wild-type, eod3-
ko1 cyp78a9-ko1 and eod3-1D embryos.
Values in (B and C) are given as means SEs, relative to the respective wild-
type values, set at 100%. **P< 0.01, compared with the wild type (Student’s t-
test). Scale bar: 0.25 mm.
(A)
(B)
(F)
(J) (K) (L) (M) (N)
(G) (H) (I)
(C) (D) (E)
Figure 8. Expression pattern of EOD3.
(A) RT-PCR analysis of EOD3 gene expression. Total RNA was isolated from
stems, roots, 10-day-old seedlings, leaves and inflorescences.
(B–L) EOD3 expression activity was monitored by pEOD3::GUS transgene
expression.
Three GUS-expressing lines were observed, and all showed a similar pattern,
although they differed slightly in the intensity of the staining. Histochemical
analysis of GUS activity in a 14-day-old seedling (B), a sepal (C), a petal (D), a
stamen (E), a carpel (F), an inflorescence (G), the valve of a silique (H) and
embryos (I–L). No GUS activity was detected in developing seeds.
(M, N) Results of in situ hybridization with an EOD3 antisense probe. Cross
section of the carpel of a stage-8 flower (M). Cross-section of the carpel of a
stage-12 flower (N).
The blue arrow indicates the central region of the septum, and the red arrow
shows the funiculus.
Scale bars: B, 2 mm; G, 1 mm; C, E, F, I, J, K, L, 100 lm; D, M, N, 50 lm; H,
200 lm.
EOD3/CYP78A6 and CYP78A9 and seed size 935
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
activity was detected in leaves. Relatively higher GUS
activity was observed in old leaves compared with young
leaves (Figure 8B). In flowers, GUS expression was
detected in sepals, petals, stamens and carpels (Figure 8C–
H). Surprisingly, there was no EOD3 expression during the
development of seeds (Figure 8I–L; Figure S7). We further
performed in situ hybridization experiments to investigate
expression of EOD3.EOD3 accumulated in the medial
gynoecial domains at stage 8 (Figure 8M). During
stage 12, the EOD3 transcript was found within the central
region of the septum (Figure 8N). Expression was also
seen in the funiculus (Figure 8N). However, EOD3
expression was not detected in integuments, embryos and
endosperms during seed development (Figure S8A–D),
consistent with the GUS staining results. Similarly,
CYP78A9 was also not observed in developing seeds
(Figure S8E–H). These analyses indicate that EOD3 is a
temporally and spatially expressed gene.
EOD3 may function independently of DA1 and TTG2
to influence seed size
The da1-1 mutant had large seeds because of the increased
cell proliferation in maternal integuments (Li et al., 2008),
whereas eod3-ko mutants produced small seeds as a result
of the reduced cell expansion in the integuments after fer-
tilization, suggesting that EOD3 and DA1 might function in
different pathways. However, the gain-of-function eod3-1D
mutant promotes both cell proliferation and cell expansion
in the integuments. We therefore asked whether there are
any genetic interactions between eod3-1D and da1-1. To test
this, we measured the size of seeds in wild-type, da1-1,eod3-
1D and eod3-1D da1-1 plants. The genetic interaction
between eod3-1D and da1-1 was essentially additive for
seed size, compared with their parental lines (Figure 9A),
further suggesting that EOD3 might function independently
of DA1 to control seed size.
The TTG2 gene acts maternally to promote cell expansion
in the integuments. ttg2 mutants produced small seeds as a
result of the reduced cell elongation in the integuments
(Garcia et al., 2005). To determine the genetic interaction
between EOD3 and TTG2, we generated the ttg2-3 eod3-ko1
double mutant. The genetic interaction between eod3-ko1
and ttg2-3 was additive for seed size, compared with their
parental lines (Figure 9B), suggesting that EOD3 functions to
control seed growth separately from TTG2.
DISCUSSION
EOD3 promotes seed growth by increasing maternal
integument size
In this study, we identified the role of EOD3/CYP78A6 in
seed size control. The eod3-1D gain-of-function mutant
formed larger seeds, whereas eod3-ko loss-of-function
mutants exhibited smaller seeds. In addition, mutations
in its most closely related family member CYP78A9 syn-
ergistically enhanced the seed size phenotype of eod3-ko
mutants (Figure 4C,F,G), indicating that EOD3/CYP78A6
acts redundantly with CYP78A9 to influence seed growth.
However, the eod3-1D mutant exhibited partial sterility,
although eod3-ko mutants had normal fertility. The trade-
off between seed number and size in many species (Harper
et al., 1970), including Arabidopsis (Alonso-Blanco et al.,
1999), has been observed. Our results show that the effect
of eod3-1D on seed size is not primarily caused by its
effect on fertility. Similarly, recent studies show that ap2
and arf2 mutations increase seed size, partly because of
reduced fertility, but also through a separate maternal
effect on seed growth (Jofuku et al., 2005; Ohto et al.,
2005; Schruff et al., 2006).
Reciprocal cross experiments show that EOD3 acts mater-
nally to affect seed growth. The integuments surrounding
the ovule are maternal tissues, and form the seed coat after
fertilization. An altered maternal integument size, such as
those seen in arf2,da1-1 and klu ovules, is known to
contribute to changes in seed size (Schruff et al., 2006;
Li et al., 2008; Adamski et al., 2009). However, the size of
mature eod3-ko1 cyp78a9-ko1 ovules was similar to that of
wild-type ovules, suggesting that the size difference
between the wild-type and eod3-ko1 cyp78a9-ko1 seeds
happens after fertilization. Consistent with this idea, eod3-
kocyp78a9-ko1 integuments were smaller than wild-type
integuments from 2 DAP onwards (Figure 6B). By contrast,
eod3-1D formed large integuments in mature ovules and
developing seeds (Figure 6A, B). Thus, a general theme
emerging from these studies is that the control of maternal
integument size is one of the critical mechanisms for
determining final seed size.
The size of the integument or seed coat is determined by
cell proliferation and cell expansion. The cell number in the
integuments of the mature ovule sets the growth potential of
the seed coat after fertilization. For example, arf2 and da1-1
mutants had large ovules with more cells, resulting in large
(A) (B)
Figure 9. Genetic interactions of eod3 with da1-1 and ttg2-3 mutants.
(A) Projective area of wild-type, eod3-1D,da1-1 and da1-1 eod3-1D seeds.
(B) Projective area of wild-type, ttg2-3,eod3-ko1 and ttg2-3 eod3-ko1 seeds.
Values in (A, B) are given as means SEs, relative to the respective wild-type
values, set at 100%.
936 Wenjuan Fang et al.
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
seeds (Schruff et al., 2006; Li et al., 2008), whereas klu
mutants formed small ovules with less cells, leading to small
seeds (Adamski et al., 2009). After fertilization, cells in
integuments mainly undergo expansion. Our results indi-
cate that the eod3-ko1 cyp78a9-ko1 mutant formed normal-
sized ovules, but smaller developing seeds, as a result of the
reduced cell expansion in the integuments after fertilization
(Figure 6). However, eod3-1D promoted both cell prolifera-
tion and cell elongation in the integuments of developing
seeds, resulting in a seed cavity of large volume. Therefore,
integument growth is driven by both cell proliferation and
cell expansion; these two processes are assumed to be
coordinated. In addition, our reciprocal cross experiments
provide a demonstration of maternal sporophytic control of
seed growth (Figures 5, S5). It is plausible that the maternal
integument or seed coat, which acts as a physical constraint
on embryo and endosperm growth, sets an upper limit to
final seed size.
The CYP78A family members have overlapping
and distinct functions in seed growth
EOD3 encodes a cytochrome P450 CYP78A6, one of the
CYP78A family members. The other CYP78A subfamily
member genes have been isolated as growth regulators.
Overexpression of CYP78A9, which is most closely related to
EOD3/CYP78A6, induced large and seedless silique in Arabi-
dopsis (Ito and Meyerowitz, 2000). To a certain extent, plants
overexpressing EOD3/CYP78A6 and CYP78A9 exhibited
similar growth phenotypes, such as large siliques and short
stamens (Figure 2C and D) (Ito and Meyerowitz, 2000),
suggesting that these two genes might affect the same or
related metabolic network. In line with this idea, our genetic
analyses demonstrate that the cyp78a9-ko1 mutation syn-
ergistically enhanced the seed size phenotype of eod3-ko
mutants (Figure 4C,F). This suggests that EOD3 and
CYP78A9 may have overlapping functions in seed size con-
trol.
Another CYP78A subfamily member KLU/CYP78A5 also
affects seed size by promoting cell proliferation in the
integuments of ovules (Adamski et al., 2009). klu mutants
produced smaller seeds than the wild type because their
small ovules contained less cells (Adamski et al., 2009). By
contrast, eod3-ko1 cyp78a9-ko1 mutants did not signifi-
cantly affect the size of ovules, but restricted cell expansion
in the integuments of developing seeds. These findings
suggest that KLU may act in the cell proliferation phase at
the early stages of integument development, and EOD3 may
mainly function in the cell expansion phase at the later
stages of integument growth.
EOD3 and CYP78A9 may control seed growth
in a non-cell-autonomous manner
Another interesting feature of the CYP78A subfamily mem-
bers is to generate mobile factors mediating organ growth
(Miyoshi et al., 2004; Anastasiou et al., 2007). Rice PLA1/
CYP78A11 affected cell division in the shoot apical meristem
(SAM), but CYP78A11 expression was not detected in the
shoot apical meristem, suggesting that CYP78A11 probably
acts through its non-cell-autonomous function (Miyoshi et
al., 2004). Arabidopsis CYP78A5 has been proposed to be
involved in generating a mobile signal distinct from the
classical phytohormones (Anastasiou et al., 2007). However,
mobile growth substances remain to be discovered. Inter-
estingly, EOD3 and CYP78A9 were not detected in the
maternal integuments of developing seeds (Figure S8) (Ito
and Meyerowitz, 2000), but eod3-ko,cyp78a9-ko and eod3-
ko cyp78a9-ko mutants produced small seeds (Figure 4C,F).
This suggests that EOD3 and CYP78A9 might control seed
growth in a non-cell-autonomous manner, as proposed for
other CYP78A subfamily members (Miyoshi et al., 2004;
Anastasiou et al., 2007). However, EOD3 expression was
detected in other organs, such as leaves and carpels (Fig-
ure 8B,F), suggesting that EOD3 might promote leaf and
carpel growth in a cell-autonomous manner. Several Ara-
bidopsis mutants with large organs also exhibited large
seeds (Schruff et al., 2006; Li et al., 2008), suggesting a
possible link between organ size and seed growth. By con-
trast, several other mutants with large organs produced
normal-sized seeds (Szecsi et al., 2006; White, 2006), indi-
cating that organ size is not always positively related to seed
growth. 35S::EOD3#7 plants exhibited normal growth and
fertility, but produced significantly larger seeds than the wild
type (Figure 3C,D; Figure S3), suggesting that the effect of
EOD3 on seed size might not linked to its effect on organ
size. CYP78A9 has been suggested to be involved in pro-
ducing an undiscovered plant growth substance (Ito and
Meyerowitz, 2000). One of the functions of EOD3 might be
the production of a signal that promotes integument growth.
Eventually, the elucidation of the biochemical function of
these gene products may lead to the discovery of one or
more new plant-growth substances that can be used to
control seed size.
The size of seeds is one of the most important
agronomic traits that affect seed yield and biomass.
Improving seed yield remains an important challenge for
growers of many agricultural crops worldwide. In this
study, EOD3 was identified as an important player influ-
encing seed size. Our current knowledge of EOD3 func-
tions suggest that the combination of EOD3 and DA1
genes [and their homologs in crops such as Glycine max
(soybean), Brassica napus (oilseed rape) and Oryza sativa
(rice)] could make a significant contribution to the future
improvement of these key crops. However, it should be
noted that crop plants have undergone selection for large
seed size, which is not the case in Arabidopsis. Therefore,
it could be interesting to know whether beneficial alleles of
EOD3 and DA1 genes have already been used by crop
breeders.
EOD3/CYP78A6 and CYP78A9 and seed size 937
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
EXPERIMENTAL PROCEDURES
Activation tagging screening
The Agrobacterium tumefaciens strain GV3101 was transformed
with the activation-tagging vector pJFAT260 (Fan et al., 2009), and
the resulting strain was used for floral-dip transformation of
Arabidopsis da1-1 mutant plants (Li et al., 2008). T
1
plants were
selected by using the herbicide Basta. Seeds produced from T
1
plants were passed through a fine wire sieve (425 lm) (Fisher Sci-
entific, http://www.fishersci.com). Seeds retained by the sieve were
kept for further characterization.
Plant materials and growth conditions
Arabidopsis thaliana Columbia (Col-0) was the wild-type line used.
All mutants were in the Col-0 background. Plant materials and
growth conditions are available in Appendix S1.
Morphological and cellular analysis
Area measurements of fully expanded cotyledons, petals (stage 14)
and leaves were made by flattening the organs, scanning to produce
a digital image and then calculating area by using
IMAGEJ
. Embryo
cell sizes were measured on the adaxial side of cotyledons from
differential interference contrast (DIC) images.
For analysis of whole-mount seeds, seeds were dissected from
siliques and placed in a drop of clearing solution [30 ml H
2
O, 80 g
chloral hydrate (C8383; Sigma-Aldrich, http://www.sigmaaldrich.
com), 10 ml 100% glycerol (G6279; Sigma-Aldrich)]. Samples were
photographed under a microscope (LEICA DM2500; Leica, http://
www.leica.com), with differential interference contrast optics using
a SPOT FLEX Cooled CCD Digital Imaging System.
Seed size and seed mass analysis
The average seed mass was determined by weighing mature dry
seeds in batches of 500 using an electronic analytical balance
(AL104; Mettler Toledo, http://us.mt.com). The weights of five
sample batches were measured for each seed lot. The wild-type
and mutant seeds were photographed under a Leica microscope
(LEICA S8APO) using Leica CCD (DFC420). The length, width and
projective area of wild-type and mutant seeds were measured
with
IMAGEJ
.
Cloning of the EOD3 gene
The flanking region of the T-DNA insertion of the eod3-1D mutant
was isolated by the thermal asymmetric interlaced PCR (TAIL-PCR)
(Liu et al., 1995). Detailed protocols are described in Appendix S1.
Constructs and transformation
The EOD3 CDS was subcloned into the PstI site of the binary vector
35S::pGreen to generate the transformation plasmid 35S::EOD3.
The specific primers for the EOD3 CDS are EOD3CDS-F and
EOD3CDS-R (Table S3).
The 1878-bp EOD3 promoter was subcloned into SacI and NcoI
sites of the binary vector pGreen-GUS to generate the transforma-
tion plasmid pEOD3::GUS. The specific primers for the EOD3
promoter are EOD3PROM-F and EOD3PROM-R (Table S3).
GUS staining
Samples (pEOD3::GUS) were stained in a solution of 1 m
M
X-gluc,
50 m
M
NaPO
4
buffer, 0.4 m
M
of K
3
Fe(CN)6/K
4
Fe(CN)6 and 0.1% (v/v)
Triton X-100, and incubated at 37C for 8 h. After GUS staining,
chlorophyll was removed using 70% ethanol.
RT-PCR, quantitative real-time RT-PCR and RNA in situ
hybridization
Total RNA was extracted from Arabidopsis seedlings using an
RNAprep Pure Plant Kit (Tiangen, http://www.tiangen.com). Reverse
transcription (RT)-PCR was performed as described by Li et al.
(2006). cDNA samples were standardized on actin transcript using
the primers ACTIN7-F and ACTIN7-R. Quantitative real-time RT-PCR
analysis was performed with a Lightcycler 480 machine (Roche,
http://www.roche.com) using the Lightcycler 480 SYBR Green I
Master (Roche). ACTIN2 mRNA was used as an internal control, and
relative quantities of mRNA were calculated using the comparative
threshold cycle method. The RNA in situ hybridization method is
described in Appendix S1. The primers used for RT-PCR, quantita-
tive real-time RT-PCR and RNA in situ hybridization are described in
Table S3.
ACKNOWLEDGEMENTS
We thank: the anonymous reviewers and the editor for their critical
comments on the article; Fiona Corke, Caroline Smith, Jun Fan and
Michael W. Bevan for their suggestions on the activation tagging
method; Jun Fan for the pJFAT260 vector; and the Arabidopsis
Stock center ABRC for ttg2-3,eod3-ko and cyp78a9-ko mutants. This
work was supported by the National Basic Research Program of
China (2009CB941503) and the National Natural Science Foundation
of China (91017014; 30921003).
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
Figure S1. eod3-1D enhances the organ size phenotype of da1-1.
Figure S2. Organ size and reproductive development in eod3-1D.
Figure S3. Phenotypes of wild-type, eod3-1D and 35S::EOD3 plants.
Figure S4. Identification of eod3-ko1,eod3-ko2 and cyp78a9-ko1
mutants.
Figure S5. EOD3 acts maternally to influence seed size.
Figure S6. Mature ovules from Col-0, eod3-ko1 cyp78a9-ko1 and
eod3-1D plants.
Figure S7. Expression of EOD3 in developing seeds.
Figure S8. EOD3 and CYP78A9 expression in developing seeds.
Table S1. Phenotypes of wild-type, eod3-ko1, cyp78a9-ko1,eod3-
ko1 cyp78a9-ko1 and eod3-1D plants.
Table S2. Developmental stages of embryogenesis.
Table S3. List of primers used in this study.
Appendix S1. Plant materials and growth conditions, cloning of the
EOD3 gene, cellular analysis and RNA in situ hybridization
Please note: As a service to our authors and readers, this journal
provides supporting information supplied by the authors. Such
materials are peer-reviewed and may be re-organized for online
delivery, but are not copy-edited or typeset. Technical support
issues arising from supporting information (other than missing
files) should be addressed to the authors.
REFERENCES
Adamski, N.M., Anastasiou, E., Eriksson, S., O’Neill, C.M. and Lenhard, M.
(2009) Local maternal control of seed size by KLUH/CYP78A5-dependent
growth signaling. Proc. Natl. Acad. Sci. USA,106, 20115–20120.
Alonso-Blanco, C., Blankestijn-de Vries, H., Hanhart, C.J. and Koornneef, M.
(1999) Natural allelic variation at seed size loci in relation to other life history
traits of Arabidopsis thaliana.Proc. Natl. Acad. Sci. USA,96, 4710–4717.
Anastasiou, E., Kenz, S., Gerstung, M., MacLean, D., Timmer, J., Fleck, C. and
Lenhard, M. (2007) Control of plant organ size by KLUH/CYP78A5-depen-
dent intercellular signaling. Dev. Cell,13, 843–856.
938 Wenjuan Fang et al.
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
Fan, C., Xing, Y., Mao, H., Lu, T., Han, B., Xu, C., Li, X. and Zhang, Q. (2006)
GS3, a major QTL for grain length and weight and minor QTL for grain
width and thickness in rice, encodes a putative transmembrane protein.
Theor. Appl. Genet. 112, 1164–1171.
Fan, J., Hill, L., Crooks, C., Doerner, P. and Lamb, C. (2009) Abscisic acid has a
key role in modulating diverse plant-pathogen interactions. Plant Physiol.
150, 1750–1761.
Garcia, D., Saingery, V., Chambrier, P., Mayer, U., Jurgens, G. and Berger, F.
(2003) Arabidopsis haiku mutants reveal new controls of seed size by
endosperm. Plant Physiol. 131, 1661–1670.
Garcia, D., Fitz Gerald, J.N. and Berger, F. (2005) Maternal control of inte-
gument cell elongation and zygotic control of endosperm growth are
coordinated to determine seed size in Arabidopsis. Plant Cell,17, 52–60.
Gegas, V.C., Nazari, A., Griffiths, S., Simmonds, J., Fish, L., Orford, S., Sayers,
L., Doonan, J.H. and Snape, J.W. (2010) A genetic framework for grain size
and shape variation in wheat. Plant Cell,22, 1046–1056.
Harper, J.L., Lovell, P.H. and Moore, K.G. (1970) The shapes and sizes of
seeds. Annu. Rev. Ecol. Syst. 1, 327–356.
Ito, T. and Meyerowitz, E.M. (2000) Overexpression of a gene encoding a
cytochrome P450, CYP78A9, induces large and seedless fruit in arabidopsis.
Plant Cell,12, 1541–1550.
Jofuku, K.D., Omidyar, P.K., Gee, Z. and Okamuro, J.K. (2005) Control of seed
mass and seed yield by the floral homeotic gene APETALA2. Proc. Natl.
Acad. Sci. USA,102, 3117–3122.
Li, J., Thomson, M. and McCouch, S.R. (2004) Fine mapping of a grain-weight
quantitative trait locus in the pericentromeric region of rice chromosome 3.
Genetics,168, 2187–2195.
Li, Y., Lee, K.K., Walsh, S., Smith, C., Hadingham, S., Sorefan, K., Cawley,
G. and Bevan, M.W. (2006) Establishing glucose- and ABA-regulated
transcription networks in Arabidopsis by microarray analysis and pro-
moter classification using a relevance vector machine. Genome Res. 16,
414–427.
Li, Y., Zheng, L., Corke, F., Smith, C. and Bevan, M.W. (2008) Control of final
seed and organ size by the DA1 gene family in Arabidopsis thaliana.Genes
Dev. 22, 1331–1336.
Liu, Y.G., Mitsukawa, N., Oosumi, T. and Whittier, R.F. (1995) Efficient isola-
tion and mapping of Arabidopsis thaliana T-DNA insert junctions by ther-
mal asymmetric interlaced PCR. Plant J. 8, 457–463.
Lopes, M.A. and Larkins, B.A. (1993) Endosperm origin, development, and
function. Plant Cell,5, 1383–1399.
Luo, M., Dennis, E.S., Berger, F., Peacock, W.J. and Chaudhury, A. (2005)
MINISEED3 (MINI3), a WRKY family gene, and HAIKU2 (IKU2), a leucine-
rich repeat (LRR) KINASE gene, are regulators of seed size in Arabidopsis.
Proc. Natl. Acad. Sci. USA,102, 17531–17536.
Miyoshi, K., Ahn, B.O., Kawakatsu, T., Ito, Y., Itoh, J., Nagato, Y. and Kurata,
N. (2004) PLASTOCHRON1, a timekeeper of leaf initiation in rice, encodes
cytochrome P450. Proc. Natl. Acad. Sci. USA,101, 875–880.
Moles, A.T., Ackerly, D.D., Webb, C.O., Tweddle, J.C., Dickie, J.B. and West-
oby, M. (2005) A brief history of seed size. Science,307, 576–580.
Ohto, M.A., Fischer, R.L., Goldberg, R.B., Nakamura, K. and Harada, J.J. (2005)
Control of seed mass by APETALA2. Proc. Natl. Acad. Sci. USA,102, 3123–
3128.
Ohto, M.A., Floyd, S.K., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (2009)
Effects of APETALA2 on embryo, endosperm, and seed coat development
determine seed size in Arabidopsis. Sex. Plant Reprod. 22, 277–289.
Orsi, C.H. and Tanksley, S.D. (2009) Natural variation in an ABC transporter
gene associated with seed size evolution in tomato species. PLoS Genet. 5,
e1000347.
Schruff, M.C., Spielman, M., Tiwari, S., Adams, S., Fenby, N. and Scott, R.J.
(2006) The AUXIN RESPONSE FACTOR 2 gene of Arabidopsis links auxin
signalling, cell division, and the size of seeds and other organs. Develop-
ment,133, 251–261.
Scott, R.J., Spielman, M., Bailey, J. and Dickinson, H.G. (1998) Parent-of-
origin effects on seed development in Arabidopsis thaliana. Development,
125, 3329–3341.
Shomura, A., Izawa, T., Ebana, K., Ebitani, T., Kanegae, H., Konishi, S. and
Yano, M. (2008) Deletion in a gene associated with grain size increased
yields during rice domestication. Nat. Genet. 40, 1023–1028.
Song, X.J., Huang, W., Shi, M., Zhu, M.Z. and Lin, H.X. (2007) A QTL for rice
grain width and weight encodes a previously unknown RING-type E3
ubiquitin ligase. Nat. Genet. 39, 623–630.
Szecsi, J., Joly, C., Bordji, K., Varaud, E., Cock, J.M., Dumas, C. and
Bendahmane, M. (2006) BIGPETALp, a bHLH transcription factor is involved
in the control of Arabidopsis petal size. EMBO J. 25, 3912–3920.
Wang, A., Garcia, D., Zhang, H., Feng, K., Chaudhury, A., Berger, F., Peacock,
W.J., Dennis, E.S. and Luo, M. (2010) The VQ motif protein IKU1 regulates
endosperm growth and seed size in Arabidopsis. Plant J. 64, 670–679.
Weng, J., Gu, S., Wan, X. et al. (2008) Isolation and initial characterization of
GW5, a major QTL associated with rice grain width and weight. Cell Res. 18,
1199–1209.
Westoby, M., Falster, D.S., Moles, A.T., Vesk, P.A. and Wright, I.J. (2002) Plant
ecological strategies: some leading dimensions of variation between
species. Annu. Rev. Ecol. Syst. 33, 125–159.
White, D.W. (2006) PEAPOD regulates lamina size and curvature in Arabi-
dopsis. Proc. Natl. Acad. Sci. USA,103, 13238–13243.
Xiao, W., Brown, R.C., Lemmon, B.E., Harada, J.J., Goldberg, R.B. and Fischer,
R.L. (2006) Regulation of seed size by hypomethylation of maternal and
paternal genomes. Plant Physiol. 142, 1160–1168.
Zhou, Y., Zhang, X., Kang, X., Zhao, X. and Ni, M. (2009) SHORT HYPOCOTYL
UNDER BLUE1 associates with MINISEED3 and HAIKU2 promoters in vivo
to regulate Arabidopsis seed development. Plant Cell,21, 106–117.
Zondlo, S.C. and Irish, V.F. (1999) CYP78A5 encodes a cytochrome P450 that
marks the shoot apical meristem boundary in Arabidopsis. Plant J. 19, 259–
268.
EOD3/CYP78A6 and CYP78A9 and seed size 939
ª2012 The Authors
The Plant Journal ª2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 929–939
    • "Evidence information in Arabidopsis indicates that TTG2 mutants have smaller seeds and that TTG2 is involved in seed coat development and epidermal cell fate specification. PubMed references (PMID:22251317 and PMID:15598800) are provided within the knowledge network as an additional source of evidence [38,39]. This example highlights the potential benefits of data integration and linked data to establish associations between distant concepts such as traits/QTL on one side and genes/biological processes on the other side. "
    [Show abstract] [Hide abstract] ABSTRACT: The chances of raising crop productivity to enhance global food security would be greatly improved if we had a complete understanding of all the biological mechanisms that underpinned traits such as crop yield, disease resistance or nutrient and water use efficiency. With more crop genomes emerging all the time, we are nearer having the basic information, at the gene-level, to begin assembling crop gene catalogues and using data from other plant species to understand how the genes function and how their interactions govern crop development and physiology. Unfortunately, the task of creating such a complete knowledge base of gene functions, interaction networks and trait biology is technically challenging because the relevant data are dispersed in myriad databases in a variety of data formats with variable quality and coverage. In this paper we present a general approach for building genome-scale knowledge networks that provide a unified representation of heterogeneous but interconnected datasets to enable effective knowledge mining and gene discovery. We describe the datasets and outline the methods, workflows and tools that we have developed for creating and visualizing these networks for the major crop species, wheat and barley. We present the global characteristics of such knowledge networks and with an example linking a seed size phenotype to a barley WRKY transcription factor orthologous to TTG2 from Arabidopsis, we illustrate the value of integrated data in biological knowledge discovery. The software we have developed (www.ondex.org) and the knowledge resources http://knetminer.rothamsted.ac.uk we have created are all open-source and provide a first step towards systematic and evidence-based gene discovery in order to facilitate crop improvement.
    Full-text · Article · Nov 2016
    • "The embryo is surrounded by the endosperm, which, in turn, is enclosed within the maternal seed coat. The grain size is regulated by the coordinated growth of the embryo, endosperm, and maternal tissues (Fang et al., 2012; Xia et al., 2013). In recent years, the knowledge about grain development improved considerably, and some genetic and molecular mechanisms are now known, mainly in model plants (Sundaresan, 2005; Sun et al., 2010; Orozco-Arroyo et al., 2015). "
    [Show abstract] [Hide abstract] ABSTRACT: Grain size is the result of the coordinated growth of the embryo, endosperm and maternal tissues. Understanding the clues of the development and growth of these tissues is essential for increasing grain weight, a key component of sunflower yield and quality. This research was aimed at evaluating the effect of pre-anthesis shading (source-sink ratio reduction) on grain growth and the expression of genes associated with grain size between R3 and physiological maturity in sunflower. Two sunflower genotypes contrasting in grain weight were sown in a split plot design with three replicates. Shading treatments (nets intercepting 80% of incident radiation) were set over the plots from R3 to R5 stage. Ovaries and grains (the last divided in pericarp and embryo) were sampled from R3 to R9 stage. RNA was extracted from ovary and grain tissues. The time-course of the expression of putative orthologous genes for sunflower of HaGW2 (RING-type E3 ubiquitin ligase-like) and HaAP2 (EREBP-like), were assessed by qPCR. Grain weight was affected (P< 0.05) by both genotype and shading treatments. The lower source-sink ratio decreased final grain weight. Interestingly, the expression of HaGW2 and HaAP2 genes was affected by the genotypes and the source-sink ratio in flowers and grains tissues across the developmental stages. Results presented here suggest that HaGW2 and HaAP2 genes act in the pericarp and might be involved in driving the growth of grains in this crop.
    Full-text · Conference Paper · May 2016 · Current opinion in plant biology
    • "This notion originates from observations that mutations that influence seed coat growth lead to variations in seed size, which normally coincide with altered endosperm proliferation rates. While mutants such as apetala2 (ap2) and megaintegumenta/auxin response factor 2 (mnt/arf2) that positively impact on seed coat development produce large seeds with increased number of endosperm nuclei [37,38], mutants with reduced seed coat growth, such as transparent testa glabra 2 (ttg2) or enhancer of da1-1 (eod3) [39,40], produce smaller seeds, with fewer endosperm nuclei. Whether this effect on endosperm development is due to an undiscovered signaling pathway or purely due to spatial limitation of endosperm growth, remains to be investigated. "
    [Show abstract] [Hide abstract] ABSTRACT: In seed plants, as in placental animals, gamete formation and zygotic development take place within the parental tissues. To ensure timely onset and to coordinate the development of the new generation, communication between the parent plant with the filial tissues and its precursors is of utmost importance. During female gametogenesis the maternal tissues tightly regulate megagametophyte formation and the interplay between the sporophyte and the fertilization products, embryo and endosperm, has major implications in the formation of a viable seed. We review the current knowledge on these interactions and highlight the many questions that still remain unanswered, in particular the nature of the pathways involved in these signaling events.
    Article · Feb 2016
Show more

Recommended publications

Discover more