Cell-Fate Switch of Synergid to Egg Cell in Arabidopsis eostreMutant Embryo Sacs Arises from Misexpression of theBEL1-Like Homeodomain Gene BLH1 W

Gabriela Carolina Pagnussat,a Hee-Ju Yu,a,1 and Venkatesan Sundaresana,b,2

a Section of Plant Biology, University of California, Davis, California 95616b Department of Plant Sciences, University of California, Davis, California 95616

In Arabidopsis thaliana, the female gametophyte is a highly polarized structure consisting of four cell types: one egg cell and

two synergids, one central cell, and three antipodal cells. In this report, we describe the characterization of a novel female

gametophyte mutant, eostre, which affects establishment of cell fates in the mature embryo sac. The eostre phenotype is

caused by misexpression of the homeodomain gene BEL1-like homeodomain 1 (BLH1) in the embryo sac. It is known that

BELL-KNAT proteins function as heterodimers whose activities are regulated by the Arabidopsis ovate family proteins (OFPs).

We show that the phenotypic effect of BLH1 overexpression is dependent upon the class II knox gene KNAT3, suggesting that

KNAT3 must be expressed and functional during megagametogenesis. Moreover, disruption of At OFP5, a known interactor of

KNAT3 and BLH1, partially phenocopies the eostre mutation. Our study indicates that suppression of ectopic activity of BELL-

KNOX TALE complexes, which might be mediated by At OFP5, is essential for normal development and cell specification in the

Arabidopsis embryo sac. As eostre-1 embryo sacs also show nuclear migration abnormalities, this study suggests that a

positional mechanism might be directing establishment of cell fates in early megagametophyte development.

INTRODUCTION

In Arabidopsis thaliana, the female gametophyte is a seven-cell

structure consisting of four cell types: one egg cell and two

synergids localized at the micropylar end of the mature embryo

sac, one central cell, and three antipodal cells of undetermined

function at the chalazal end (Figures 1A and 1B). All of the cells

within the embryo sac are highly polarized. While the egg cell’s

nucleus is located toward the chalazal end of the embryo sac, the

synergid and central cells have the opposite polarity (Willemse

and van Went, 1984; Huang and Russell, 1992). The formation of

the egg cell, synergids, and one of the polar nuclei can be traced

back to the four-nucleate stage of embryo sac development,

when the two pairs of nuclei migrate to opposite ends of the

coenocytic embryo sac, remaining separated by a large vacuole

(Pagnussat et al., 2005). At the chalazal pole, the nuclei are

positioned one above the other with respect to the micropylar-

chalazal axis. At the micropylar end, the nuclei generally locate

side by side (Christensen et al., 1997; Pagnussat et al., 2005). A

second round of mitotic nuclear division generates the eight

nuclei of the mature embryo sac (Pagnussat et al., 2005). No

genetic predisposition seems to guide the fate for any of the

embryo sac nuclei to form the egg cell. However, the precise

migration and positioning of the nuclei along the embryo sac at

the four to eight nucleate stages suggests an early distinction

between these nuclei as the third mitotic division progresses and

the embryo sac undergoes cellularization (Webb and Gunning,

1994). Although the mechanism of cellularization remains not

well understood (Russell, 1993; Brukhin et al., 2005), the micro-

tubular cytoskeleton appears to establish and maintain organelle

polarity and to function in migration and arrangement of the

nuclei during embryo sac development (Webb and Gunning,

1990, 1994). In agreement with this model, radiating perinuclear

microtubules have been observed during megagametophyte

cellularization (Russell, 1993; Webb and Gunning, 1994).

Although in the last few years a large number of mutants

showing defects during megagametogenesis have been isolated

in Arabidopsis (Christensen et al., 1997, 1998, 2002; Moore et al.,

1997; Pagnussat et al., 2005), the cellular and molecular basis of

cell specification in the embryo sac remains unknown. Recently,

a mutant called lachesis (lis), which affects cell specification, has

been described (Gross-Hardt et al., 2007). In lis mutant embryo

sacs, accessory cells (synergids and antipodals) begin to display

characteristics of gametic cells (egg cell and central cell). It has

been proposed that LIS, which is homologous to the yeast

splicing factor PRP4, might be involved in a signaling pathway

operating in gametic cells that prevents accessory cells from

adopting a gametic cell fate. Although the lis study indicates that

accessory cells in the embryo sac might have the competence to

differentiate into gametic cells, the mechanisms underlying the

initial establishment of cell identity in the female gametophyte are

1 Current address: National Horticultural Research Institute, Rural Devel-opment Administration, I-Mok Dong 475, Jang-An Gu, Suwon, Gyeonggi-Do, 440-706 Republic of Korea.2 Address correspondence to sundar@ucdavis.edu.The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy describedin the Instructions for Authors (www.plantcell.org) is: VenkatesanSundaresan (sundar@ucdavis.edu).W Online version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.107.054890

The Plant Cell, Vol. 19: 3578–3592, November 2007, www.plantcell.org ª 2007 American Society of Plant Biologists

Figure 1. eostre Embryo Sacs Show Two Egg Cells and Only One Synergid Cell.

(A) Differential interference contrast (DIC) image showing a wild-type embryo at stage FG7. Nuclei have been artificially colored in red.

(B) Scheme indicating the different cell types present in a wild-type embryo sac at stage FG7.

(C) DIC image showing eostre mutant embryo sac phenotype at FG7. Nuclei have been artificially colored in red.

(D) Scheme indicating the cells that morphologically resemble egg cells and synergid cells in eostre mutant embryo sacs.

(E) Expression of the specific egg cell marker ET1119 in a wild-type embryo sac.

(F) The egg cell marker ET1119 is expressed in two micropylar cells in eostre embryo sacs.

(G) Distribution of ovules showing different ET1119 expression patterns in eostre-1/EOSTRE pistils. Error bars indicate SE (n ¼ 168). The white bar

represents the percentage of ovules showing no GUS staining. The bar labeled ‘‘Gus (þ) 1 Ec’’ represents the percentage of ovules showing GUS

expression in only one cell of the embryo sac. The bar labeled ‘‘Gus (þ) 2 Ec’’ represents the percentage of ovules showing GUS expression in two cells

of the embryo sac. When ovules from wild-type pistils were analyzed, GUS staining was always detected in only one micropylar cell. Out of 264

observed, 47.8% of the ovules showed GUS staining.

(H) Expression of the specific synergid cell marker ET884 in a wild-type embryo sac.

(I) The synergid cell marker ET884 is expressed in only one micropylar cell in eostre embryo sacs.

(J) Distribution of ovules showing different ET884 expression patterns in eostre-1/EOSTRE pistils. Error bars indicate SE (n ¼ 127). The white bar

represents the percentage of ovules showing no GUS staining. The bar labeled ‘‘Gus (þ) 1 syn’’ represents the percentage of ovules showing GUS

expression in only one cell of the embryo sac. The bar labeled ‘‘Gus (þ) 2 syn’’ represents the percentage of ovules showing GUS expression in two cells

of the embryo sac. When ovules from wild-type pistils where analyzed, GUS staining was detected in two micropylar cells in 46.8% of the ovules

observed and in only one of the micropylar cells in 1.1% of the ovules (n ¼ 247).

Ccn, central cell nucleus; Ec, egg cell; EcL, egg cell–like cell; mi, micropylar pole; S, synergide; Slc, synergid-like cell.

Embryo Sac Cell Specification 3579

not known. In this report, we describe the characterization of a

novel female gametophyte mutant, eostre, which affects estab-

lishment of cell fates in the mature embryo sac.

RESULTS

The Cell Identity of Micropylar Cells Is Altered in the

eostre Mutant

The eostre-1 mutant was identified from a large collection of

female gametophyte mutants generated by Ds insertions and

was originally called EDA12 (for Embryo Sac Development Ar-

rest) (Pagnussat et al., 2005). As the Ds insertion carries an NPTII

gene, gametophytic defects can be identified by the decreased

transmission of resistance to kanamycin. The EDA12/eostre-1

mutant showed no sporophytic defects but exhibited an aber-

rant Kanr:Kans segregation ratio of 0.5513:1 in selfed progeny

(Pagnussat et al., 2005). As this low ratio (<1:1) suggested that

both female and male gametophytes might be affected by the

Ds insertion, we performed reciprocal crosses between eostre-1

and Landsberg erecta (Ler) wild-type plants. Our results show

that both the female and the male gametophytes were severely

affected by the mutation (Table 1). The defect in the male trans-

mission was shown to arise from a defect in pollen development

(L. Boavida and S. McCormick, personal communication). To

analyze the basis of the female transmission deficiency, we

examined the terminal phenotype of the gametophytes by emas-

culating flowers from plants heterozygous for the Ds insertion,

when wild-type embryo sacs are at the final developmental stage

(stage FG7; Christensen et al., 1998; Figure 1) by emasculating

heterozygous flowers. We found that the mutant gametophytes

showed diverse defects ranging from embryo sacs completely

collapsed showing absolute degeneration (see Supplemental

Figure 1 online) or arrested very early (at FG1 to FG2), to embryo

sacs that progressed to stage FG7, but presented an abnormal

display of cells at the micropylar end. Specifically, we found that

29% of the embryo sacs in heterozygous pistils exhibited two

cells with a polarity that resembles wild-type egg cells, while only

one of the micropylar cells showed the morphology that corre-

sponds to a synergid cell (Figures 1A to 1D, Table 2). Due to the

excess of presumptive egg cells, the EDA12 mutant was re-

named eostre, after the Anglo-Saxon goddess of the spring.

To further characterize the identities of the cells present in

eostre-1 embryo sacs, we tested the expression of different cell-

specific molecular markers. The marker lines were crossed with

both wild-type plants and eostre-1 mutant plants, and the F1

progeny was analyzed for expression of the marker gene. The

expression of an egg cell–specific marker (ET119; Gross-Hardt

et al., 2007) was tested both in wild-type and in eostre-1 embryo

sacs. While wild-type embryo sacs express the marker only in the

egg cell, the marker was found to be strongly expressed in two of

the eostre-1 micropylar cells in ;15% of the embryo sacs

(Figures 1E to 1G). This result suggested that two of the eostre-1

embryo sac cells might have egg cell attributes. On the other

hand, while two micropylar cells were b-glucuronidase (GUS)

positive when the expression of a synergid-specific marker

(ET884; Gross-Hardt et al., 2007) was tested in wild-type embryo

sacs, its expression was downregulated in many eostre-1 em-

bryo sacs and ;13% of the female gametophytes of eostre-1

pistils showed GUS expression in just one of the cells composing

the embryo sac (Figures 1H to 1J), indicating that only one of the

micropylar cells of the eostre-1 embryo sac might present

synergid features. Markers for the central cell and the antipodal

cells were also tested in eostre-1 embryo sacs. For both cases,

similar GUS staining patterns were observed in wild-type and

eostre-1 mutant embryo sacs (data not shown).

Fertilization of eostre Embryo Sacs Reveals Two Functional

Egg Cells

In view of the above results, we decided to test whether the

eostre-1 mutation was affecting embryo sac functions such as

pollen tube guidance or fertilization. Fertilization in Arabidopsis

requires controlled growth of the pollen tube until it enters the

micropyle to penetrate the female gametophyte. Since the

synergid cells are responsible for pollen tube attraction (Ray

et al., 1997; Shimizu and Okada, 2000) and eostre-1 mutants

appear to have only one synergid cell instead of the two present

in wild-type embryo sacs, we analyzed the pollen tube growth

patterns in eostre-1 by aniline blue staining. Although ;80% of

the mutant ovules were able to attract pollen tubes, abnormal

pollen tube growth patterns (i.e., spinning and curling pollen

tubes) were observed in ;22% of the ovules in eostre-1/

EOSTRE pistils (Figures 2A and 2B; see Supplemental Table

1 online). Also, ;11% of the ovules showed a pollen tube on the

funiculus but not at the micropyle. (The same situation could also

be observed when wild-type pistils were analyzed, but the

percentage of ovules without pollen tubes was only of 5.72;

see Supplemental Table 1 online). Furthermore, a significantly

higher fraction of mutant ovules was found to attract two pollen

tubes to the micropyle compared with wild-type ovules (4.26

versus 0.96%, respectively; see Supplemental Table 1 online;

P < 0.001, x2 test). Together, the results indicated a defect in

pollen tube attraction and guidance, and as a consequence, part

of the pollen tubes either cannot reach the micropyles of the

ovules or growth toward them following a wavering pattern.

These defects were also observed when wild-type plants were

used as pollen donors (see Supplemental Table 1 online; no

significant differences x2 test), indicating that the problem arises

due to a disruption of the female gametophytic functions.

Table 1. Transmission Efficiency of eostre Mutants

Parental Genotypes

(Female 3 Male)a Kanr Kans

Transmission

Efficiencyb

eostre-1/EOSTRE 3 EOSTRE/EOSTRE 77 137 35.98%

EOSTRE/EOSTRE 3 eostre-1/EOSTRE 56 272 17.07%

eostre-3/EOSTRE 3 EOSTRE/EOSTRE 76 83 47.79%

EOSTRE/EOSTRE 3 eostre-3/EOSTRE 34 144 19.10%

a Plants were crossed manually, and seeds of the resulting cross were

collected and grown on selective Murashige and Skoog (MS) plates to

determine the efficiency in which the mutant allele was transmitted to

the next generation by the male or female gametes.b Transmission efficiency ¼ Kanr/Kans 3 100%.

3580 The Plant Cell

To study if the two micropylar cells expressing the specific egg

cell marker in the eostre-1 embryo sac (Figure 1F) are able to

become fertilized, we studied the phenotype of eostre-1 embryo

sacs after pollinating eostre-1 pistils using wild-type plants as

pollen donors. When pistils were analyzed 24 h after pollination,

49.7% of the ovules showed an elongated zygote and endo-

sperm development, 3.28% of the ovules were found unfertil-

ized, 4.22% showed a zygote but have endosperm development

affected, and ;24% of the ovules looked fertilized, showing two

zygote-like cells, characterized by a relatively rounded shape

and a centrally located nucleus (Figure 2D). There were also

ovules showing only one zygote-like cell and an egg cell that also

exhibited some extent of endosperm development (four to six

endosperm nuclei, Figure 2C). For those ovules showing two

zygote-like cells, they did not show any endosperm development

(Figure 2D). Also, 18.3% of the ovules were found collapsed at

this stage. Although the two micropylar cells observed in the

mutant ovules after fertilization resemble the characteristics of

zygotic cells, we decided to investigate whether those cells were

actually fertilized by pollinating eostre-1 pistils using transgenic

plants carrying the PFAC1IE:GFP-GUS:TFAC1 construct as pol-

len donors. FAC1 (Embryonic Factor 1) encodes a putative AMP

Table 2. Analysis of eostre/EOSTRE Pistils at Different Stages of Development

Stage and Phenotype of Ovules found in the Pistils Analyzed

Pistil Stagea Collapsed Ovules FM FG2-3 FG4 FG5 FG6 FG7 Total Analyzed Abnormal (%)

FG1 0 125 0 0 0 0 0 125 0.00

FG2-3 3 0 96 (0) 12 (0) 0 0 0 111 2.70

FG3-4 0 33 56 (0) 32 (6) 0 0 0 121 4.95

FG4 10 1 5 (3) 21 (7) 2 (1) 0 0 49 42.85

FG4-5 7 0 2 (1) 13 (5) 27 (6) 1 (0) 0 41 46.34

FG6 23 5 5 (5) 17 (14) 27 (14) 55 (6) 0 132 46.96

FG7 47 0 0 0 0 21 (21) 296 (107) 364 48.07

a Assignment of pistil stage was done according to Christensen et al. (1998). An FG stage was assigned to pistils showing the majority (>75%) of the

wild-type female gametophytes at that particular stage. If a single stage does not predominate, a transition between two stages was assigned.

Values within parenthesis indicate number of abnormal embryo sacs observed.

Figure 2. Functional Study of Egg Cells and Synergids in eostre Embryo Sacs.

(A) Abnormal pollen tube growth patterns observed at the micropyle of an eostre mutant ovule.

(B) Pollen tube growing toward the micropyle of a wild-type embryo sac present in the pistil of eostre-1/EOSTRE plants.

(C) eostre embryo sac after fertilization showing one zygote-like cell, an egg cell–like cell, and endosperm development (endosperm nuclei are indicated

by arrowheads).

(D) eostre embryo sac after fertilization showing two zygote-like cells and no endosperm development.

(E) to (H) GUS expression in both wild-type ([E] and [G]) and eostre ([F] and [H]) ovules after an eostre-1/EOSTRE plant was pollinated with a transgenic

plant carrying the PFAC1IE:GFP-GUS:TFAC1 construct.

(E) GUS expression observed in the zygote of a wild-type ovule.

(F) GUS expression observed in both of the zygote-like cells present in this eostre embryo sac.

(G) GUS expression observed in the embryo and in the endosperm of a wild-type ovule.

(H) GUS expression observed in one zygote-like cell and in the endosperm of an eostre embryo sac. The arrows point at endosperm nuclei.

Ec, egg cell; EcL, egg cell–like cell; En, endosperm; EP, embryo proper; F, funiculus; mi, micropylar pole; PT, pollen tube; Z, zygote; ZL, zygote like.

Embryo Sac Cell Specification 3581

deaminase (EC:3.5.4.6) and is expressed in both the developing

embryo and endosperm from the paternal allele (Xu et al., 2005).

Two different patterns of GUS expression were observed in

eostre-1 mutants after pollination: (1) GUS staining was observed

either in both of the zygote- like cells or (2) GUS staining was

found in only one of the zygote-like cells and in the endosperm.

These results suggested that either fertilization of both of the egg

cells or fertilization of only one of the egg cells and of the central

cell can occur (Figures 2F and 2H; see Supplemental Table 2

online). GUS expression was not detected in both zygote-like

cells and endosperm at the same time, suggesting that even

when some eostre-1 mutant sacs are able to attract more than

one pollen tube (see Supplemental Table 1 online), only one of

them would deliver sperm cells into the embryo sac. On the other

hand, no autonomous endosperm development was observed

when the two egg cells are fertilized, suggesting that the

unfertilized central cell is not dividing after the fertilization event.

This result was intriguing since it was previously shown that a

positive signal from the fertilization of the egg cell initiates

proliferation of the unfertilized central cell in wild-type plants

(Nowack et al., 2006). We therefore considered whether the

fertilization of two egg cells rather than one in the embryo sac

was interfering with the signaling after fertilization, preventing

both embryo and endosperm development in the eostre-1 mu-

tant. To test that possibility, we used the cdc-2 mutant, carrying a

mutation in the Arabidopsis Cdc2 homolog CDC2A, to pollinate

eostre-1/EOSTRE siliques. In cdc2a mutant pollen, only one

sperm cell, instead of two, is produced. Mutant pollen is viable

and the single sperm fertilizes only the egg cell, but this single

fertilization event appears to trigger endosperm development in

the central cell (Nowack et al., 2006). If the response of the

central cell of an eostre-1 embryo sac is similar to that of a wild-

type embryo sac, fertilization with cdc2a pollen should result in

an increase in the fraction of ovules that show both embryo

development and some extent of endosperm development. How-

ever, we did not observe any difference in endosperm develop-

ment in the siliques analyzed after pollinating eostre-1 with cdc2a

pollen (see Supplemental Table 3 online).

eostre-1 Shows Nuclei Migration and Cellularization

Abnormalities during Megagametogenesis

To analyze the developmental stage at which eostre-1 female

gametophytes start to show defects, we observed the embryo

sacs of heterozygous pistils at different phases of development.

Since female gametophyte development within a pistil is gener-

ally synchronous (Christensen et al., 1997), we used the wild-

type embryo sacs siblings as a guide to ascribe delays to the

mutant embryo sacs present within the same pistil.

As shown in Table 2, eostre-1 female gametophytes begin to

develop normally, as no differences can be observed at FG1

stage among any of the embryo sacs present in the heterozygous

pistils analyzed. However, from stage FG3 on, the developmental

asynchrony was evident. While the majority of the embryo sacs

was already between FG3 and FG4 stages, ;27% of the embryo

sacs were still at FG1 stage. Moreover, some of the embryo sacs

already at FG4 stage showed abnormal nuclear position. While in

the wild-type embryo sacs at stage FG4 the two pairs of nuclei

are separated by a large central vacuole and showed a typical

arrangement (Figures 3A and 3C), in eostre-1 female gameto-

phytes, the vacuole is positioned on a side of the embryo sac, the

two pairs of nuclei are not completely separated, and they look

aligned along the coenocytic embryo sac (Figure 3I). When pistils

were analyzed at stages FG4 to FG5, 10% of the embryo sacs

showed this defect (Table 2). Approximately 18% of the embryo

sacs also showed to be collapsed at this stage (Table 2) probably

arising from those that were previously observed to be severely

delayed. During the cellularization process that begins following

the third round of mitosis (FG5 and FG6), not only the develop-

mental asynchrony was obvious but also ;30% of the embryo

sacs showed abnormal distribution of nuclei along the embryo

sac or cells with aberrant shapes and polarities at the micropylar

end (Table 2). At mid stage FG5, a wild-type embryo sac shows

four micropylar nuclei with a distinctive distribution, in which the

two nuclei that will further become synergid nuclei are located

side by side at the embryo sac mycropylar edge, while the future

egg cell nuclei is located toward the embryo sac micropylar half

(Figure 3D). However, an atypical arrangement of nuclei at the

micropylar pole was observed in eostre-1 embryo sacs at this

stage (Figure 3J). Only one nuclei instead of the two observed in

wild-type embryo sac is located at the micropylar embryo sac

edge, two nuclei are located side by side at the middle region of

the micropylar pole, and one nucleus is located closer to the

center of the embryo sac (Figures 3D and 3J). As cellularization

progresses in wild-type embryo sacs, cellular morphology in the

cells at the embryo sac’s micropylar pole become clearly iden-

tifiable. At FG6 stage, the egg cell shows a characteristic pear

shape and a highly polarized cytoplasm; its nucleus and most of

its cytoplasm are at its chalazal end and a large vacuole occupies

the remaining space of the cell (Figures 3A and 3E). Similarly,

wild-type synergid cells also have a distinguishing shape and

polarity, which is the opposite of the egg cell cytoplasm (Figures

3A and 3E). By contrast, eostre-1 female gametophytes showed

two cells in which the nuclei occupied a chalazal position and

only one cell in which the nucleus showed a micropylar orienta-

tion. Thus, at the end of FG7, eostre-1 embryo sacs appear to

contain two egg cells instead of one and only one synergid cell

(Figures 3K and 3L). This asynchrony in embryo sac development

was not observed when wild-type pistils were analyzed, and

abnormal or aborted embryo sacs were rarely observed (only 1 to

2.3%; see Supplemental Table 4 online).

A Ds Element Is Inserted Inside an Intergenic Region and

Promotes Misexpression of a BEL1-Like Gene in eostre

Embryo Sacs

Thermal asymmetric interlaced PCR (Liu et al., 1995) was used to

identify the flanking sequence surrounding the Ds insertion in

eostre-1 mutants using Ds-specific primers and arbitrary degen-

erated primers. The flanking sequence obtained revealed that

the single Ds insertion was inserted in the intergenic region

between the genes At2g35940 (BEL1-like homeodomain

1 [BLH1]) and At2g35950 (unknown protein) 1862 nt upstream

of the At2g35950 start codon (Figure 4A). This location was

further confirmed by PCR using sequence-specific primers for

the putative insertion site in combination with the corresponding

3582 The Plant Cell

Figure 3. Wild-Type and eostre Female Gametophyte Development in Arabidopsis.

(A) Scheme showing the developmental stages leading to the wild-type embryo sac formation.

(B) Wild-type embryo sac at FG3 stage, showing two nuclei separated by a vacuole.

(C) Wild-type embryo sac at FG4 stage, showing two pairs of nuclei separated by a vacuole.

(D) Eight-nucleated wild-type embryo sac. One of the chalazal end nuclei is migrating toward the micropylar end (indicated by arrows).

(E) Wild-type embryo sac containing seven cells and eight nuclei.

(F) Wild-type embryo sac containing seven cells and seven nuclei.

(G) Scheme showing the developmental stages leading to the eostre mutant female gametophyte formation.

(H) eostre embryo sac at FG3 stage, showing two nuclei separated by a small vacuole.

(I) eostre embryo sac at FG4 stage. The vacuole is positioned on a side of the embryo sac, the two pairs of nuclei are not completely separated, and they

look aligned along the embryo sac.

(J) eostre embryo sac at FG5 stage showing an atypical arrangement of nuclei at the micropylar end. One of the chalazal end nuclei is migrating toward

the micropylar end (indicated by arrows).

(K) eostre embryo sac containing seven cells and eight nuclei. eostre embryo sacs appear to contain two egg cells instead of one and only one synergid

cell.

(L) eostre embryo sac at FG7 stage. Nuclei have been artificially colored in red.

Cc, central cell; Ccn, central cell nucleus; Ch, chalazal pole; Ec, egg cell; EcL, egg cell–like cell; mi, micropylar pole; Pn, polar nuclei; S, synergide; SL,

synergid-like cell.

Embryo Sac Cell Specification 3583

Ds element primers. A DNA fragment of the predicted size

was amplified from the genomic DNA of eostre-1 but not from

wild-type Arabidopsis genomic DNA. The DNA sequence of the

PCR product verified the junction between the genomic DNA and

the Ds element and showed a direct repeat of 8 bp at the point of

insertion (Figure 4A). To verify if this insertion site was respon-

sible for the phenotype observed in the eostre-1 mutant, we

obtained lines containing T-DNA insertions around this region

from the Salk Institute Genomic Analysis Laboratory collection

(Alonso et al., 2003). We analyzed two independent T-DNA al-

leles with insertions in the intergenic region between At2g35940

and At2g35950 that were named eostre-2 and eostre-3. In

eostre-2, the T-DNA is inserted 1161 nucleotides upstream of

the At2g35950 start codon, and in eostre-3, the T-DNA is in-

serted 3132 nucleotides upstream of the At2g35950 start codon.

No homozygous plants could be detected for either of these

alleles. However, the mutation seemed to be less penetrant in

these mutants. eostre-2 siliques contained ;27% of aborted/

desiccated ovules, while eostre-3 siliques contained ;32% of

aborted ovules (Figures 4B and 4C). To study whether the eostre

allele mutations affect the female gametophyte as in the original

eostre-1 mutation, we analyzed the phenotype of embryo sacs in

both eostre-2 and eostre-3 pistils. In both cases, ;15% of the

mutant ovules showed micropylar cells with abnormal polarities

as was observed for eostre-1 (Figures 4D and 4E). A fraction of

the ovules collapsed very early during development (;30% for

both alleles; see Supplemental Table 5 online). The remaining

fraction, even when they looked normal at stage FG7, seemed to

have problems attracting pollen tubes and remain unfertilized

(see Supplemental Table 5 online). We also noticed that the

number of unfertilized or aborted ovules in eostre-2 and eostre-3

varied widely from silique to silique even within the same plant, a

characteristic that was not observed in the eostre-1 mutant. Also,

eostre-3 exhibited reduced transmission through both the fe-

male and the male gametophyte (Table 1).

As At2g35950 was the nearest gene to these intergenic

insertions, we examined two lines obtained from the Salk

Institute Genomic Analysis Laboratory collection (Alonso et al.,

2003) carrying a T-DNA insertion within this gene to study if a

putative deficiency in At2g35950 expression was causing the

observed phenotype. For both lines, we were able to identify

homozygous null mutants plants for At2g35950, as no tran-

scripts for this gene were detected by RT-PCR (data not shown),

indicating that disruption of At2g35950 does not lead to game-

tophytic defects.

In an attempt to identify genes of noncoding RNAs, the in-

tergenic region was screened for the presence of microRNAs

using the microRNA candidates database (http://sundarlab.

ucdavis.edu/mirna/search_candidates.html/; Adai et al., 2005)

and for the presence of small RNAs using the small RNA

database (http://asrp.cgrb.oregonstate.edu/db), but no matches

were found around the insertion site. We also looked for tran-

scription units across the intergenic region in an effort to detect

the expression of nonannotated genes using the Arabidopsis

Transcriptome Genomic Express Database (http://signal.salk.edu/

cgi-bin/atta; Yamada et al., 2003) and by checking the putative

transcripts by RT-PCR, without positive results. As no new genes

or noncoding RNAs across the region of the insertions studied

Figure 4. The eostre Mutant Phenotype Is Caused by a Ds Insertion within the Intergenic Region between the Genes At2g35940 and At2g35950.

(A) Position of the Ds element in eostre-1 and of the T-DNA insertion in eostre-2 and eostre-3. The arrow indicates the GUS gene within the Ds insertion.

The Ds insertion is located 11,272 bp upstream of the At2g35940 (BLH1) start codon.

(B) DIC image showing eostre-2 mutant embryo sac phenotype at FG7.

(C) DIC image showing eostre-3 mutant embryo sac phenotype at FG7.

(D) Phenotype of an eostre-2/EOSTRE silique. The insert shows an enlarged section of the silique, and the arrow points to an aborted ovule.

(E) Phenotype of an eostre-3/EOSTRE silique. The insert shows an enlarged section of the silique, and the arrowhead points to an aborted ovule.

Ccn, central cell nucleus; EcL, egg cell–like cell; mi, micropylar pole of the embryo sac; SL, synergid-like cell.

3584 The Plant Cell

were detected, the expression of the upstream gene At2g35940

(BLH1) and the expression of two additional downstream genes

(At2g35960 and At2g35970, both encoding a harpin-induced

family protein) was studied in wild-type and eostre-1 ovules by

RT-PCR experiments using ovule total mRNA as a template. For

the downstream genes studied, we were not able to detect any

expression in wild-type or eostre-1 ovules (data not shown).

Although no expression of BLH1 was detected in wild-type

ovules either, its transcripts were consistently detected in

eostre-1 ovules (Figure 5A). This result suggested that BLH1

might be misexpressed in the mutant ovules. Accordingly, ex-

pression of BLH1 was also found in the eostre-1 alleles eostre-2

and eostre-3 (Figure 5B). To study if the overexpression detected

in eostre-1 was ovule specific, BLH1 expression was also ana-

lyzed in leaves. BLH1 expression was higher in eostre-1 leaves

compared with the wild-type ones (Figure 7D), although no

defects were observed in leaf development or morphology. This

result indicated that although the deregulation of BLH1 gene

might be general, it only causes a defect in reproductive tissues.

We used a gene trap line (GT9784) carrying an insertion in BLH1

gene from the Cold Spring Harbor collection (http://genetrap.

cshl.org/) to analyze the BLH1 expression pattern in Arabidopsis.

As observed in Figure 5G, when dissected pistils of heterozygous

plants were analyzed for GUS expression, GUS staining was only

detected in the transmitting tract and in the base of the funiculus

but not in the ovule or the embryo sac. Thus, the normal BLH1

expression pattern seems to exclude both the ovule and the

female gametophyte.

The eostre Phenotype Is Caused by Misexpression of BLH1

in the Embryo Sac

As BLH1 was shown to be misexpressed in eostre ovules, we

further investigated if this might be the cause of the observed

phenotype by driving the expression of BLH1 in the embryo sacs

of wild-type plants using the pOp/LhG4 transactivation system

(Moore et al., 1998). We generated transgenic lines expressing

the chimeric transcription factor LhG4 under the control of two

different promoters: At5g40260 (pES1 [for Embryo Sac promoter

1]), which encodes a protein of the nodulin MtN3 family and is

expressed in the embryo sac from FG1 to FG7 (Yu et al., 2005),

and At1g26795 (pES2 [for Embryo Sac promoter 2]), which

encodes a self-incompatibility protein and it is strongly ex-

pressed from FG3 in the dividing nuclei of the embryo sac to

FG7 (Yu et al., 2005). A reporter construct carrying the BLH1

coding region under the control of the p10Op promoter, which is

activated by LhG4, was introduced in plants that were crossed to

pES:LhG4 plants. As the pES:LhG4 constructs were directly

introduced into transgenic plants carrying a pOp-GUS reporter

construct, we were able to confirm that the promoters used were

directing the expression of the reporter genes to the embryo sacs

by studying GUS staining patterns (Figure 5C). To verify if BLH1

was in fact expressed in these plants, RT-PCR experiments were

performed using total RNA from ovules of plants carrying the

Op:BLH1 reporter construct alone as well as plants carrying both

the driver and reporter constructs. BLH1 expression was only

detected in ovules from plants carrying both constructs (Figure

Figure 5. Misexpression of BLH1 in the Embryo Sac Recapitulates the eostre Phenotype.

(A) RT-PCR showing BLH1 overexpression and actin expression as a positive control (bottom) in eostre-1 ovules.

(B) Overexpression of BLH1 was also observed in eostre-2 and eostre-3 ovules in the Col background. Actin expression is shown as a positive control

(bottom).

(C) GUS staining pattern in a pES1�BLH1 ovule showing that pES1 specifically directs expression of the reporter genes to the female gametophyte.

(D) BLH1 expression in pES1�BLH1 ovules detected by RT-PCR. Actin expression is shown as a positive control (bottom).

(E) Phenotype of a pES1�BLH1 embryo sac at FG4 stage. Arrowheads indicate the unusual position of nuclei.

(F) Phenotype of a pES1�BLH1 embryo sac at FG7 stage.

(G) Expression pattern of BLH1 in the pistil of the BLH1 gene trap line GT9784.

Ccn, central cell nucleus; EcL, egg cell–like cell; Es, embryo sac; F, funiculus; mi, micropylar pole of the embryo sac; SL, synergid-like cell;

Tt, transmitting track.

Embryo Sac Cell Specification 3585

5D). These results indicated that BLH1 was successfully and

specifically expressed in the embryo sac of the plants under

study. When the pistils of F1 plants expressing BLH1 in the

embryo sacs were analyzed, approximately one-fourth of the

ovules showed to be defective, which was the proportion ex-

pected for a double heterozygote for the driver and reporter

construct. The defects observed range, as in eostre-1 mutants,

from early aborted embryo sacs to embryo sacs that progress to

the terminal stage showing two egg cells and a single synergid

(Figures 5E and5F). When the phenotype of these embryo sacs

was studied after fertilization, both embryo sacs showing only

one arrested zygote, one egg cell–like cell and four to six

endosperm nuclei or two zygotic-like structures without endo-

sperm development were observed (see Supplemental Figure 2

online). Plants carrying the reporters or the driver constructs

alone did not show any phenotype, indicating no background

activity.

blh1 Loss-of-Function Mutants Have Normal and

Functional Gametophytes

To examine the effects of loss of function of BLH1, we analyzed

three independent T-DNA lines with insertions in At2g35940

(BLH1) that are predicted to be null alleles. For all three lines, we

were able to identify homozygous mutant plants with blh1 null

alleles (verified by RT-PCR; data not shown), indicating that

BLH1 alone is not essential for gametophyte development or

function. As BLH1 belongs to the BELL family of homeodomain

transcription factors (Chan et al., 1998; Roeder et al., 2003), a

large family that includes 13 proteins, we could not eliminate the

possibility that a functional redundancy might be obscuring

BLH1 function in the embryo sac. As BLH5 is the BEL1-like

protein that shows the closest homology to BLH1 (Roeder et al.,

2003), double gene knockout lines were generated and were

statistically analyzed in segregating F2 populations (blh1 3 blh5).

Homozygous plants for both insertions were recovered, and

plants were found to look like wild-type plants, showing no

obvious defects either in the sporophytic or the gametophytic

organs (data not shown). This result suggested either that BLH1

or BLH5 is not involved in gametophyte development or that a

broader functional redundancy among the members of the

BELL1 family exists, requiring additional mutations in other

BLH genes to reveal their functions in the embryo sac.

A Mutation in the Class II Knox Gene KNAT3 Reverts the

eostre Phenotype

Biochemical studies have shown that BELL proteins associate

with KNOX proteins to form heterodimers that are thought to

compose functional complexes regulating plant development

(Muller et al., 2001). Specifically, BLH1 has been reported to

interact with the KNOX proteins KNAT3, KNAT5, and KNAT6

(Hackbusch et al., 2005). If overexpression of BLH1 in the

embryo sac requires these KNOX proteins to cause the eostre

phenotype, we might expect that loss of function of the corre-

sponding KNAT gene would suppress the eostre mutation. To

test this possibility, we generated double mutants by crossing

knox gene knockout lines to eostre-1. The segregating F1 pop-

ulation was analyzed by genotyping and analysis of the siliques.

No significant differences in the number of aborted seeds

per silique were observed between eostre-1 single mutant and

eostre-1 knat5 or eostre-1 knat6 double mutants (Figures 6A and

6B). On the other hand, heterozygous plants for both a T-DNA

insertion in knat3 gene and for the eostre-1 Ds insertion showed a

significant reduction in the number of aborted seeds per silique

when compared with the eostre-1 single mutant (Figure 6; t test,

P < 0.005). The suppression of eostre-1 cannot be attributed to

the different Arabidopsis ecotype of the knat3 mutant (Columbia

[Col]), as no suppression was detected when eostre-1 (Ler

background) was crossed to the knat5 or knat6 mutants (Col

Figure 6. A Mutation in the Class II Knox Gene KNAT3 Suppresses the eostre Phenotype.

(A) Mature siliques of eostre-1/EOSTRE plants and double mutants eostre1 knat6, eostre1 knat5, and eostre1 knat3. The bottom panel shows a mature

silique of the wild type.

(B) The aborted seeds per silique were quantified for each genetic background. Values represent average number of aborted seeds per silique (mean 6

SE, n¼ 20). Light-gray bars correspond to ovules from plants heterozygous for the insertion in knox genes. The dark-gray bar shows data corresponding

to ovules from plants that were homozygous for the insertion in the knat-3 gene.

3586 The Plant Cell

background). When the F2 population was analyzed, plants that

were homozygous for the T-DNA insertion in the knat3 gene and

heterozygous for the eostre-1 Ds insertion were recovered and

its siliques studied. Only ;13% of the ovules observed looked

aborted in these siliques, a number significantly lower than the

one observed for pistils from heterozygous plants for both the

knat3 gene and eostre-1 insertions (Figure 6; t test, P < 0.05).

Thus, the knat3 mutation is able to suppress the eostre pheno-

type, but the suppression is not complete, suggesting the

possibility that other KNATs might be interacting with BLH1 in

the embryo sac. Previous results have reported KNAT3 expres-

sion during early organ development in leaves, buds, and ped-

icels at the junction between two organs during development,

including the ovule-funiculus boundaries, and in maturing tis-

sues, such as siliques, petioles of maturing leaves, and the root

(Serikawa et al., 1997; Truernit et al., 2006). Although no data

regarding KNAT3 expression in Arabidopsis embryo sac have

been reported so far, the closely related class II knox gene

KNOX6 was shown to be expressed in the maize (Zea mays)

embryo sac after cellularization (Evans, 2007). To examine

KNAT3 expression in Arabidopsis ovules, total RNA from wild-

type ovules and from spl mutant ovules, which lack embryo sacs

(Yang et al., 1999), was used to study KNAT3 expression by RT-

PCR. KNAT3 expression was consistently detected in both wild-

type and spl mutant ovules, indicating that KNAT3 is at least

expressed sporophytically in the ovule, although expression in

the embryo sac remains to be confirmed (Figure 7). Furthermore,

and to test the possibility of an upregulation of KNAT3 in eostre

ovules, the level of KNAT3 expression was assessed in eostre-1,

wild-type, and spl ovules by RT-PCR. No differences were found

among the samples, indicating that KNAT3 is not upregulated in

eostre-1 ovules (Figure 7B). As BLH1 overexpression was also

detected in somatic tissues (Figure 7D), we also studied KNAT3

expression in wild-type and eostre-1 leaves by RT-PCR. Again,

no differences were detected (Figure 7E).

At OFP5, a Putative Regulator of BLH1 Activity, Is Required

for Normal Embryo Sac Development

An investigation of Arabidopsis TALE protein interactions re-

vealed a heavily connected network of interactions of TALE

proteins with each other and with members of the recently

discovered OVATE protein family (At OFP) (Hackbusch et al.,

2005). Particularly, it was shown that members of this family

interact both with KNAT3 and BLH1 in a yeast two-hybrid system

(Hackbusch et al., 2005). Furthermore, At OFP1 and At OFP5

regulate subcellular localization of BLH1, relocalizing it from the

nucleus, where it would be active as a transcription factor, to the

cytoplasmic space (Hackbusch et al., 2005). Although little

information is available about the specific expression patterns

of both TALE and At OFP genes in the female gametophyte, the

results with eostre and knat3 suggested that complexes involv-

ing both TALE and At OFP might be functional during embryo sac

development. To study this possibility, we analyzed Arabidopsis

lines with T-DNA insertions in four At OFP genes (At OFP1-1,

At OFP1-2, At OFP2-1, At OFP4-1, and At OFP5-1) obtained

from the Salk Institute Genomic Analysis Laboratory collection

(Alonso et al., 2003). Although homozygous plants were recov-

ered from selfing of heterozygous plants for At OFP1-1, At

OFP1-2, At OFP2-1, and At OFP4-1 mutants, no homozygous

plants were detected for At OFP5-1 mutants, suggesting that

At OFP5 might be required for essential processes in gameto-

phyte or sporophyte development. When female gametophyte

development was investigated in At OFP5-1 pistils, out of 256

embryo sacs studied, 54% looked normal, while ;38% of the

ovules seemed to collapse very early during development

around stage FG2. The remaining ;8% of the embryo sacs

(20/256) showed abnormal micropylar cells, with embryo sacs

showing two egg cells as was observed in eostre-1 mutants

(Figure 7). These results strongly indicate a requirement for At

OFP5 during female gametophyte development and also that

TALE-OVATE protein complexes might be functional during

embryo sac early developmental stages. RT-PCR experiments

were performed to evaluate At OFP5 expression in the embryo

sac. At OFP5 expression levels were found to be higher in wild-

type ovules compared with spl ovules, indicating that At OFP5 is

expressed both in the embryo sac and in the sporophytic tissues

of the ovule (Figure 7).

Figure 7. At OFP5, an Interactor of BLH1 and KNAT3, Is Required for

Normal Embryo Sac Development.

(A) At OFP5-1 mutant embryo sac at FG7 showing a phenotype similar to

the one observed in eostre-1 ovules.

(B) RT-PCR showing KNAT3 expression in wild-type, eostre-1, and spl

ovules. Actin expression is also shown as a positive control (bottom). The

picture corresponds to 20 PCR cycles.

(C) RT-PCR showing At OFP5 expression in wild-type and spl ovules.

Actin expression is shown as a positive control (bottom). The picture

corresponds to 20 PCR cycles. At OFP5 expression was detected in spl

ovules after 25 PCR cycles.

(D) RT-PCR showing BLH1 expression in wild-type and eostre-1 leaves.

The picture corresponds to 25 PCR cycles. BLH1 expression was

detected in wild-type leaves at 30 cycles.

(E) RT-PCR showing KNAT3 expression in wild-type and eostre-1 leaves.

The picture corresponds to 25 PCR cycles.

Ccn, central cell nucleus; EcL, egg cell–like cell; mi, micropylar pole of

the embryo sac; SL, synergid-like cell.

Embryo Sac Cell Specification 3587

DISCUSSION

Regulation of BELL-KNOX TALE Complexes Is Essential for

Normal Development of the Arabidopsis Embryo Sac

The eostre gametophytic mutant shows synergid to egg cell fate

switch in Arabidopsis embryo sacs, a phenotype that could be

recapitulated when BLH1 was misexpressed in the embryo sac.

Hence, misexpresion of BLH1 seems sufficient to generate the

eostre phenotype, probably by interfering with the normal pat-

terning during female gametophyte development. Although no

loss-of-function mutants were found for BELL proteins that might

be involved in the normal embryo sac development pathway,

functional redundancy might be obscuring a substantial role for

these proteins during normal megagametogenesis. BELL and

KNOX proteins are members of the MEINOX-TALE superfamily

of eukaryotic transcriptional regulators that function as hetero-

dimers and are known to regulate developmental processes both

in plants and animals (Knoepfler et al., 1997; Rieckhof et al.,

1997; Bellaoui et al., 2001; Muller et al., 2001; Smith et al., 2004).

Biochemical studies have shown that BELL proteins associate

with KNOX proteins to form functional complexes regulating

plant development (Muller et al., 2001). In particular, BLH1 has

been reported to interact with the KNOX proteins KNAT3,

KNAT5, and KNAT6 (Hackbusch et al., 2005). As a common

feature of KNOX-BELL interactions is that the KNOX protein

partner often interacts with a subset of BELL proteins (Cole et al.,

2006), BLH1 misexpression in the embryo sac might be prevent-

ing functional KNOX–BELL interactions to take place and, as a

result, interfering with normal patterning. Supporting this idea,

KNAT3, a gene belonging to the class II subfamily of knox genes

(Reiser et al., 2000) and an already described interactor of BELL

proteins (including BLH1; Hackbusch et al., 2005), appears to be

essential for the phenotypic effect of BLH1 overexpression in the

embryo sac (Figure 6). This result suggested that KNAT3 must be

expressed and functionally competent in the embryo sac. In

agreement, KNAT3 expression was detected in Arabidopsis

ovules (Figure 7B). In addition, the class II gene Knox6, which

is the nearest maize homolog of KNAT3, was the only Knox gene

found to be consistently expressed in isolated maize embryo

sacs (Evans, 2007). As no functions for class II Knox genes in

plant development have been reported in any plant species to

date (Reiser et al., 2000), knat3 suppression of the embryo sac

defects of the eostre mutant revealed a phenotype for a class II

Knox mutant. Emphasizing the importance of regulation of knox

genes during megagametophyte development, the maize gene

INDETERMINATE GAMETOPHYTE1 (IG1), which represses the

expression of meristem-specific knox genes in leaf primordia,

has been shown to be expressed during early embryo sac

development, and ig1 mutants exhibit embryo sac abnormalities

that include extra egg cells, extra polar nuclei, and extra syner-

gids (Evans, 2007). In Arabidopsis, mutation of the putative Ig1

ortholog AS2 does not result in embryo sac defects (Semiarti

et al., 2001). However, At OFP5, a member of the Arabidopsis

ovate family, which was shown to interact with KNAT3 and to

regulate subcellular localization of BLH1 (Hackbusch et al.,

2005), appears to play an essential role during embryo sac

development (Figure 7). At ofp5 embryo sacs displayed pheno-

types that resemble those observed in eostre-1 mutants, consistent

with At OFP5 acting as a key negative regulator of BELL-KNOX

activity during early embryo sac development in Arabidopsis.

Factors Affecting Cell Fate Specification during Female

Gametophyte Development

The atypical migration and positioning of nuclei observed in

eostre mutant embryo sacs lead to a variety of morphological

and functional defects, including the production of an extra

functional egg cell and only one synergid cell. Our observations

suggest that the specification of cell fate, as egg cell or synergid,

appears to rely on a position-based mechanism. However, we

cannot rule out that the altered migration of nuclei could also be a

consequence of an altered fate. At FG5, only one nucleus is

located at the micropylar edge of eostre mutant embryo sacs, a

position at which the pair of nuclei that will further form part of the

synergid cells are normally positioned in wild-type embryo sacs.

Furthermore, two nuclei are located side by side at the middle

region of the micropylar pole, where typically the future egg cell

nucleus is placed in wild-type embryo sacs. As cellularization

progresses, eostre embryo sacs clearly showed an abnormal

pattern of cells at their micropylar edge, with only one cell

displaying synergid features and two cells showing egg cell

characteristics (Figure 1). These specific cells were also shown to

be functional, as the synergid cell was able to attract pollen tubes

and both egg cells in eostre mutant embryo sacs are able to

become fertilized (Figure 2). Recently, the description of the lis

mutant, which shows extra egg cells in detriment of both syn-

ergids and central cell identities, has led to a model that upon

differentiation, gametic cells would generate an inhibitory signal

that transmitted to the adjacent cells would prevent excess

gametic cell formation (Gross-Hardt et al., 2007). Although this

mechanism of lateral inhibition can explain the maintenance of

only one egg cell in the embryo sac after the initial specification

of cell fate, the mechanisms involved in the early establishment

of different cell fates in the embryo sac are not addressed by the

model. A positional mechanism would explain how cell fate is

specified in early megagametophyte development. Although

when all the nuclei present in the embryo sac might be compe-

tent to form gametic cells upon cellularization, those nuclei that

are at the micropylar or chalazal poles acquire the accessory cell

fates. However, once the egg cell is differentiated, a lateral in-

hibition mechanism seems to be necessary to maintain cell fates,

as accessory cells can yet differentiate later on into gametic cells

if this inhibitory mechanism is not present (Gross-Hardt et al.,

2007). The idea of a positional mechanism is also supported by

studies in maize, where ig1 embryo sacs undergo extra rounds of

free nuclear divisions, resulting in extra egg cells, extra central

cells, and extra polar nuclei (Guo et al., 2004; Evans, 2007).

The migration and position of nuclei during megagameto-

genesis in Arabidopsis has been shown to be highly regular

(Webb and Gunning, 1990, 1994). Previous studies by Brown and

Lemmon (1991, 1992) suggested that cytoplasmic domains may

determine the fate of cells during cell partitioning. This idea is

supported by our findings studying eostre mutants, as the

unusual position of the nuclei along the embryo sac might

expose them to different cytoplasmic domains compared with

3588 The Plant Cell

the wild-type embryo sac nuclei. An abnormal microtubular

behavior (i.e., a change in microtubule dynamics) during and

following the second and third rounds of mitotic divisions (FG2 to

FG4) or a failure in the nuclei interactions with components of the

microtubular cytoskeleton might explain the unusual pattern of

nuclear migration and the resulting abnormal nuclear positioning

observed in eostre embryo sacs. The importance of specific

positions of nuclei along the embryo sac for cell fate determina-

tion might rely on the asymmetric distribution of still unknown

morphogenetic determinants, perhaps in the form of gradients,

that might be responsible for determining the fate of the different

cells composing the embryo sac as described for numerous

multicellular systems (Adler, 2000; Betschinger and Knoblich,

2004; Harris and Peifer, 2005).

METHODS

Plant Growth Conditions

The origin of the Ds insertion lines used in this study has been previously

described (Sundaresan et al., 1995). Typically 50 to 200 seeds were

sterilized in 20% (v/v) sodium hypochlorite, washed with sterile water, and

plated on MS medium with 50 mg L�1 kanamycin in Percival growth

chambers (Percival Scientific), with a 16-h-light/8-h-dark cycle at 228C.

Resistant (green) seedlings were then transferred onto soil and grown

under the conditions described above with 60% relative humidity. For

crosses, flowers of the female parent were manually emasculated 2 d

before anthesis and cross-pollinated 2 d later.

Histology and Microscopy

To prepare cleared whole-mount preparations, pistils containing at least

20 ovules were dissected and cleared overnight in Hoyer’s solution (Liu

and Meinke, 1998). The dissected pistils were observed on a Zeiss

Axioplan imaging 2 microscope under DIC optics. Images were captured

on an Axiocam HRC CCD camera (Zeiss) using the Axiovision program

(version 3.1). For GUS staining, developing carpels and siliques were

dissected and incubated in GUS staining buffer [5 mM EDTA, 0.1% Triton

X-100, 5 mM K4Fe(CN)6, 0.5 mM K3Fe (CN)6, and 1 mg mL�1 X-Gluc [Rose

Scientific] in 50 mM sodium phosphate buffer, pH 7.0) overnight at 378C.

Individual ovules were dissected from the pistils/siliques and cleared

overnight with Hoyer’s solution. The ovules were observed under DIC

optics. For pollen tube staining, 10 pistils containing at least 20 ovules

each were manually pollinated and opened longitudinally 24 h after

pollination for each mutant. The pistils were cleared in 10% chloral

hydrate at 658C for 5 min and washed with water, softened with 5 M NaOH

at 658C for 5 min, and washed again with water. The pistils were then

treated with 0.1% aniline blue in 0.1 M K3PO4 buffer, pH 8.3, for 3 h in

darkness and washed with 0.1 M K3PO4 buffer. The pistils were mounted

on a microscope slide using a drop of glycerol and carefully squashed

under a cover slip. The pistils were observed using a fluorescence

microscope.

Image Processing

All images were processed for publication using Adobe Photoshop CS

(Adobe Systems).

Segregation Analysis

For self-cross analysis, heterozygous plants were allowed to self-pollinate

and progeny seed was collected. The progeny seed was germinated on

growth medium containing the proper markers (kanamycin, BASTA, or

sulfadiazine) as indicated in each case. The F1 seed was germinated on

growth medium containing 50 mg/mL kanamycin, and the number of

resistant and sensitive plants was scored. For analysis of reciprocal

crosses, we crossed the heterozygous mutant lines carrying a resistant

marker gene, with wild-type plants using the insertional lines as egg donor

and wild-type Ler as the sperm donor and vice versa. Quantification of

resistant plants in the progeny was performed by collecting the seeds

derived from these crosses and plating them in selective media according

to the selection marker carried by each of the insertional lines tested.

Molecular Analysis of the Sequences Flanking the Ds Insertion

The sequences flanking the Ds element were isolated with thermal

asymmetric interlaced PCR as described by Liu et al. (1995) and se-

quenced with an Applied Biosystems sequencer. The flanking sequences

obtained were run against BLASTN to identify the genomic location of the

Ds element. The junctions were then confirmed by PCR using sequence-

specific primers for the putative insertion site in combination with the

corresponding Ds element primers. To verify the left border–genomic

sequence junction, we used the DS-specific primer Ds59-1A (59-ACG-

GTCGGGAAACTAGCTCTAC-39) combined with the genomic sequence

primer EDA12-5 (59-CCATCCTATGTATTTAGAGTTCCTGC-39). For the

right border–genomic sequence junction, we used the DS-specific primer

Ds39-2A (59-CGATTACCGTATTTATCCCGTTTC-39) combined with the

genomicsequenceprimerEDA12-3 (59-TGATACGATGGTTAGATCACG-39).

Both junctions were corroborated by sequencing.

Molecular Characterization of Insertional Lines around

the Ds Insertion

eostre-2 (SALK_033866) and eostre-3 (SAIL_582_F02) insertional lines

were obtained from the Salk Institute Genomic Analysis Laboratory

collection (Alonso et al., 2003). For eostre-2, the left border–genomic

sequence junction was determined by PCR using the T-DNA–specific

primer LBb1 (59-GCGTGGACCGCTTGCTGCAACT-39) combined with

the genomic sequence-specific primer eostre1RP (59-CAGTGGATAT-

GGGAATGCAAC-39). For eostre-3, the left border–genomic sequence

junction was determined also by PCR in plants showing BASTA

resistance using the T-DNA–specific primer 59-GCCTTTTCAGAAATG-

GATAAATAGCCTTGCTTCC-39) combined with the genomic sequence-

specific primer eostre2 RP (59-AATCCGATCGGTATTACGAGG-39). Lines

35950-1 (SALK_086362) and 35950-2 (SALK_086364) were obtained

from the Salk Institute Genomic Analysis Laboratory collection (Alonso

et al., 2003). For both lines, the left border–genomic sequence junction

was determined by PCR using the T-DNA–specific primer LBb1 (59-GCG-

TGGACCGCTTGCTGCAACT-39) combined with genomic sequence-

specific primers. For both lines, the specific genomic primers were

35950-1/2 RP (59-TTGTACCGAGAAAGGCTCAAG-39) and 35950-/21 LP

(59-TGGGAATTTTGGATTTAGCCC-39). Line blh1-1 (SALK_089095) was

obtained from the Salk Institute Genomic Analysis Laboratory collection

(Alonso et al., 2003), while blh1-2 (GK-114D09) and blh1-3 (GK-475C05)

were obtained from the Nottingham Arabidopsis Stock Centre (Scholl

et al., 2000). For blh1-1, the left border–genomic sequence junction was

determined by PCR using the T-DNA–specific primer LBb1 (59-GCG-

TGGACCGCTTGCTGCAACT-39) combined with genomic sequence-

specific primers blh1-1RP (59-TTCCAGCCGCTTAAGCATAC-39) and

blh1-1LP (59-TATGAATCCCAATCACAACGG -39). For blh1-2, the left

border–genomic sequence junction was determined by PCR using the

T-DNA–specific primer LBGK (59-TGGTTCACGTAGTGGGCCATCG-39)

combined with genomic sequence-specific primers blh1-2RP (59-CCG-

TCGGATCCGGCAGAGATCT-39) and blh1-2LP (59-GATCTTTGAGTCT-

GACACAGAGACC -39). For blh1-3, the left border–genomic sequence

junction was determined by PCR using the T-DNA–specific primer LBGK

Embryo Sac Cell Specification 3589

(59-TGGTTCACGTAGTGGGCCATCG-39) combined with genomic

sequence-specific primers blh1-3RP (59-CCAGCCGCTTAAGCATA-

CATGT-39) and blh1-3LP (59-GGCGTCACTGGAATGCAAGGAA-39).

Molecular Characterization of Insertional Lines for KNOX and

OVATE Genes

Insertional lines for KNAT3 (SALK_136464), KNAT5 (SALK_088589), and

KNAT6 (SALK_140566) genes were obtained from the Salk Institute

Genomic Analysis Laboratory collection (Alonso et al., 2003). For the

three lines, the left border–genomic sequence junction was determined

by PCR using the T-DNA–specific primer LBb1 (59-GCGTGGACCGC-

TTGCTGCAACT-39) combined with genomic sequence-specific primers.

For knat3, the DNA-specific primers used were knat3LP (59-GTTAAACA-

CAGCGCTTCTTCG-39) and knat3RP (59-TCACAGGATTCATTTTCTC-

ACC-39). For knat5, the DNA-specific primers used were knat5LP

(59-ATTCGGGGTTTTGATTACCAG-39) and knat5RP (59-GAACAGCAAC-

TCTTCCACGTC-39). For knat6, the DNA-specific primers used were

knat6LP (59-GTTTTAGCCATGGGATTAGGG-39) and knat6RP (59-TGA-

TTTGTTATGGCCACAATG-39). T-DNA insertional lines for At OFP1

(SALK_111492 and SALK_127550), At OFP2 (SALK_122550), At OFP4

(SALK_014905), and At OFP5 (SALK_010386) were obtained from the

Salk Institute Genomic Analysis Laboratory collection (Alonso et al.,

2003). For all the lines, the left border–genomic sequence junction was

determined by PCR using the T-DNA–specific primer LBb1 (59-GCG-

TGGACCGCTTGCTGCAACT-39) combined with genomic sequence-

specific primers. For At OFP1-1, the DNA-specific primers used were

At OFP1-1LP (59-AGAGATCCCAGATCTCGAAGC-39) and At OFP1-1RP

(59-AAGGTTGCGGTTTTGGATAAC-39). For At OFP1-2, the DNA-specific

primers used were At OFP1-2LP (59-ATTGGGCCGAAAACATATAGG-39)

and At OFP1-2RP (59-AAGGTTGCGGTTTTGGATAAC-39). For At OFP2-1,

the DNA-specific primers used were At OFP2-1LP (59-ACCAAATTCAAA-

GAAGCATCG-39) and At OFP2-1RP (59-TGGTGAGTTATGGTGAG-

GAGG-39). For At OFP4-1, the DNA-specific primers used were At

OFP4-1LP (59-TCGTTAGGGTTGTGACTGACC-39) and At OFP4-1RP

(59-CCTAGATTCAAAGGATGTGCG-39). For At OFP5-1, the DNA-specific

primers used were At OFP5-1LP (59-GCTTCTATATCCCCATTTTAA-

TGTG-39) and At OFP4-1RP (59-CCTAGATTCAAAGGATGTGCG-39).

Molecular Characterization of a Gene Trap Line with an

Insertion in BLH1

The gene trap line GT9784, carrying an insertion in the BLH1 gene, was

obtained from the Cold Spring Harbor collection (http://genetrap.cshl.

org/). The junctions of the Ds element with the genomic DNA were

confirmed by PCR using sequence-specific primers for the insertion site

in combination with Ds element primers. The DS-specific primer Ds59-1A

(59-ACGGTCGGGAAACTAGCTCTAC-39) was used combined with the

genomic sequence primer GT9784RP (59-TCATCAATCTTGTAAGTTGT-

CA-39), and the DS-specific primer Ds39-2A (59-CGATTACCGTATT-

TATCCCGTTTC-39) was used combined with the genomic sequence

primer GT9784LP (59-GGTGGCGTACTAAGGACTGTGGGA-39).

Expression Analysis

Total RNA was isolated from dissected ovules using TRIzol (Invitrogen).

cDNA was generated by SuperScript II reverse transcriptase (Invitrogen).

This was used for RT-PCR analysis using 0.5 mL of cDNA template, and

intron-spanning primer pairs were used for 30 cycles at 948C for 30 s,

588C for 30 s, and 728C for 30 s. RT-PCR results were normalized using

Arabidopsis thaliana actin-2 controls for PCR. BLH1 cDNA was amplified

using the following primers: 59-CCGGGTCGGGTCTGGCTCTA-39 and

59-CGGCTCGTTCTGGGAGACCACG-39. ACTIN2 was amplified using

59-CCTGAAAGGAAGTACAGTG-39 and 59-CTGTGAACGATTCCTGGAC-39.

Constructs and Plant Transformation

Total RNA was isolated from dissected Ler ovules using TRIzol (In-

vitrogen). cDNAs of BLH1 were generated by SuperScript II reverse

transcriptase (Invitrogen). After sequence verification, 10op::BLH1 was

constructed by inserting BLH1 cDNA behind an OP array (10OP-TATA-

BJ36) and subsequently subcloned into the binary vector pCAMBIA 1300.

The plasmid was introduced into Agrobacterium tumefaciens strain

GV3101 by electroporation, and Ler wild-type plants were transformed

using the floral dip method (Clough and Bent, 1998). Transformants were

selected based on their ability to survive in MS medium with 20 mg L�1

hygromicyn. Resistant (green seedlings with true leaves) were then trans-

ferred onto soil and grown under the conditions described above.

At5g40260 (pES1), At1g26795 (pES2), and promoter fragments were

previously cloned and its activity characterized as described (Yu et al.,

2005). pES:Lhg4 constructs were prepared by inserting the promoter

fragments upstream the coding sequence of the quimeric transcription

factor LhG4 and then subcloned into the binary vector pMLBART. The

plasmid was introduced into A. tumefaciens strain GV3101 by electropora-

tion, and wild-type Ler plants containing the transgene 6OP:GUS were

transformed using the floral dip method (Clough and Bent, 1998). Trans-

formants were selected based on their ability to survive in MS medium

containing 50 mg L�1 kanamycin and 6 mg L�l of ammonium glufosinate.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis Genome

Initiative database under the following accession numbers: At2g35940

(BLH1), At5g40260 (pES1), At1g26795 (pES2), AT4G18830 (ATOFP5),

and AT5G25220 (KNAT3).

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. DIC Image Showing a Collapsed Embryo Sac.

Supplemental Figure 2. Misexpression of BLH1 Expression in

pES2�BLH1 Ovules Recapitulates the eostre Phenotype.

Supplemental Table 1. Pollen Tube Growth Patterns toward eostre

and Wild-Type Female Gametophytes.

Supplemental Table 2. Patterns of GUS Expression Observed when

Both Wild-Type Plants and eostre-1 Mutants Were Pollinated Using

Transgenic Plants Carrying the PFAC1IE:GFP-GUS:TFAC1 Construct

as Pollen Donors.

Supplemental Table 3. Embryo and Endosperm Development in

Plants Fertilized with cdc2a Pollen.

Supplemental Table 4. Analysis of Wild-Type Pistils at Different

Stages of Development.

Supplemental Table 5. Phenotype of Embryo Sacs in eostre-2 and

eostre-3 Pistils.

ACKNOWLEDGMENTS

We thank John Bowman for providing 10OP-TATA-BJ36 and LhG4-

BJ36 vectors and seeds of 6OP-GUS transgenic plants, Joe Simorowski

for providing seeds of the BLH1 gene trap line, and members of the

Gasser and Sundaresan labs for helpful discussions. This work was

funded by National Science Foundation 2010 Grant 0313501.

Received August 8, 2007; revised October 29, 2007; accepted November

8, 2007; published November 30, 2007.

3590 The Plant Cell

REFERENCES

Adai, A., Johnson, C., Mlotshwa, S., Archer-Evans, S., Manocha, V.,

Vance, V., and Sundaresan, V. (2005). Computational prediction of

miRNAs in Arabidopsis thaliana. Genome Res. 15: 78–91.

Adler, R. (2000). A model of retinal cell differentiation in the chick

embryo. Prog. Retin. Eye Res. 19: 529–557.

Alonso, J.M., et al. (2003). Genome-wide Insertional mutagenesis of

Arabidopsis thaliana. Science 301: 653–657.

Bellaoui, M., Pidkowich, M.S., Samach, A., Kushalappa, K., Kohalmi,

S.E., Modrusan, Z., Crosby, W.L., and Haughn, G.W. (2001). The

Arabidopsis BELL1 and KNOX TALE homeodomain proteins interact

through a domain conserved between plants and animals. Plant Cell

13: 2455–2470.

Betschinger, J., and Knoblich, J.A. (2004). Dare to be different:

Asymmetric cell division in Drosophila, C. elegans and vertebrates.

Curr. Biol. 14: R674–R685.

Brown, R.C., and Lemmon, B.C. (1991). Pollen development in orchids

2: The cytokinetic apparatus in simultaneous cytokinesis. Protoplasma

165: 155–166.

Brown, R.C., and Lemmon, B.C. (1992). Cytoplasmic domain: A model

for spatial control of cytokinesis in reproductive cells of plants. EMSA

Bull. 22: 48–53.

Brukhin, V., Curtis, M.D., and Grossniklaus, U. (2005). The angio-

sperm female gametophyte: No longer the forgotten generation. Curr.

Sci. 89: 1844–1852.

Chan, R.L., Gago, G.M., Palena, C.M., and Gonzalez, D.H. (1998).

Homeoboxes in plant development. Biochim. Biophys. Acta. 1442: 1–19.

Christensen, C.A., Gorsich, S.W., Brown, R.H., Jones, L.G., Brown, J.,

Shaw, J.M., and Drews, G.N. (2002). Mitochondrial GFA2 is required

for synergid cell death in Arabidopsis. Plant Cell 14: 2215–2232.

Christensen, C.A., King, E.J., Jordan, J.R., and Drews, G.N. (1997).

Megagametogenesis in Arabidopsis wild type and the Gf mutant. Sex.

Plant Reprod. 10: 49–64.

Christensen, C.A., Subramanian, S., and Drews, G.N. (1998). Identi-

fication of gametophytic mutations affecting female gametophyte

development in Arabidopsis. Dev. Biol. 202: 136–151.

Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method

for Agrobacterium-mediated transformation of Arabidopsis thaliana.

Plant J. 16: 735–743.

Cole, M., Nolte, C., and Werr, W. (2006). Nuclear import of the

transcription factor SHOOT MERISTEMLESS depends on heterodi-

merization with BLH proteins expressed in discrete sub-domains of

the shoot apical meristem of Arabidopsis thaliana. Nucleic Acids Res.

34: 1281–1292.

Evans, M.M.S. (2007). The indeterminate gametophyte1 gene of maize

encodes a LOB domain protein required for embryo sac and leaf

development. Plant Cell 19: 46–62.

Gross-Hardt, R., Kagi, C., Baumann, N., Moore, J.M., Baskar,

R., Gagliano, W.B., Jurgens, G., and Grossniklaus, U. (2007).

LACHESIS restricts gametic cell fate in the female gametophyte of

Arabidopsis. PLoS Biol. 5: e47.

Guo, F.L., Huang, B.Q., Han, Y.Z., and Zee, S.Y. (2004). Fertilization in

maize indeterminate gametophyte1 mutant. Protoplasma 223: 111–120.

Hackbusch, J., Richter, K., Muller, J., Salamini, F., and Uhrig, J.F.

(2005). A central role of Arabidopsis thaliana ovate family proteins in

networking and subcellular localization of 3-aa loop extension homeo-

domain proteins. Proc. Natl. Acad. Sci. USA 102: 4908–4912.

Harris, T.J.C., and Peifer, M. (2005). The positioning and segregation of

apical cues during epithelial polarity establishment in Drosophila. J.

Cell Biol. 170: 813–823.

Huang, B.Q., and Russell, S.D. (1992). Female germ unit - Organization,

isolation, and function. Int. Rev. Cytol. 140: 233–293.

Knoepfler, P.S., Calvo, K.R., Chen, H.M., Antonarakis, S.E., and

Kamps, M.P. (1997). Meis1 and pKnox1 bind DNA cooperatively with

Pbx1 utilizing an interaction surface disrupted in oncoprotein E2a-Pbx1.

Proc. Natl. Acad. Sci. USA 94: 14553–14558.

Liu, C.M., and Meinke, D.W. (1998). The titan mutants of Arabidopsis

are disrupted in mitosis and cell cycle control during seed develop-

ment. Plant J. 16: 21–31.

Liu, Y.G., Mitsukawa, N., Oosumi, T., and Whittier, R.F. (1995).

Efficient isolation and mapping of Arabidopsis thaliana T-DNA insert

junctions by thermal asymmetric interlaced PCR. Plant J. 8: 457–463.

Moore, I., Galweiler, L., Grosskopf, D., Schell, J., and Palme, K.

(1998). A transcription activation system for regulated gene expres-

sion in transgenic plants. Proc. Natl. Acad. Sci. USA 95: 376–381.

Moore, J.M., Calzada, J.P.V., Gagliano, W., and Grossniklaus, U.

(1997). Genetic characterization of hadad, a mutant disrupting female

gametogenesis in Arabidopsis thaliana. Cold Spring Harb. Symp.

Quant. Biol. 62: 35–47.

Muller, J., Wang, Y.M., Franzen, R., Santi, L., Salamini, F., and

Rohde, W. (2001). In vitro interactions between barley TALE homeo-

domain proteins suggest a role for protein-protein associations in the

regulation of Knox gene function. Plant J. 27: 13–23.

Nowack, M.K., Grini, P.E., Jakoby, M.J., Lafos, M., Koncz, C., and

Schnittger, A. (2006). A positive signal from the fertilization of the egg

cell sets off endosperm proliferation in angiosperm embryogenesis.

Nat. Genet. 38: 63–67.

Pagnussat, G.C., Yu, H.J., Ngo, Q.A., Rajani, S., Mayalagu, S.,

Johnson, C.S., Capron, A., Xie, L.F., Ye, D., and Sundaresan, V.

(2005). Genetic and molecular identification of genes required for

female gametophyte development and function in Arabidopsis. De-

velopment 132: 603–614.

Ray, S., Park, S.S., and Ray, A. (1997). Pollen tube guidance by the

female gametophyte. Development 124: 2489–2498.

Reiser, L., Sanchez-Baracaldo, P., and Hake, S. (2000). Knots in the

family tree: Evolutionary relationships and functions of knox homeo-

box genes. Plant Mol. Biol. 42: 151–166.

Rieckhof, G.E., Casares, F., Ryoo, H.D., AbuShaar, M., and Mann,

R.S. (1997). Nuclear translocation of extradenticle requires homo-

thorax, which encodes an extradenticle-related homeodomain pro-

tein. Cell 91: 171–183.

Roeder, A.H.K., Ferrandiz, C., and Yanofsky, M.F. (2003). The role of

the REPLUMLESS homeodomain protein in patterning the Arabidop-

sis fruit. Curr. Biol. 13: 1630–1635.

Russell, S.D. (1993). The egg cell - Development and role in fertilization

and early embryogenesis. Plant Cell 5: 1349–1359.

Scholl, R.L., May, S.T., and Ware, D.H. (2000). Seed and molecular

resources for Arabidopsis. Plant Physiol. 124: 1477–1480.

Semiarti, E., Ueno, Y., Tsukaya, H., Iwakawa, H., Machida, C., and

Machida, Y. (2001). The asymmetric leaves2 gene of Arabidopsis

thaliana regulates formation of a symmetric lamina, establishment of

venation and repression of meristem-related homeobox genes in

leaves. Development 128: 1771–1783.

Serikawa, K.A., Martinez-Laborda, A., Kim, H.S., and Zambryski,

P.C. (1997). Localization of expression of KNAT3, a class 2 knotted1-

like gene. Plant J. 11: 853–861.

Shimizu, K.K., and Okada, K. (2000). Attractive and repulsive interac-

tions between female and male gametophytes in Arabidopsis pollen

tube guidance. Development 127: 4511–4518.

Smith, H.M.S., Campbell, B.C., and Hake, S. (2004). Competence to

respond to floral inductive signals requires the homeobox genes

PENNYWISE and POUND-FOOLISH. Curr. Biol. 14: 812–817.

Sundaresan, V., Springer, P., Volpe, T., Haward, S., Jones, J.D.G.,

Dean, C., Ma, H., and Martienssen, R. (1995). Patterns of gene

Embryo Sac Cell Specification 3591

action in plant development revealed by enhancer trap and gene trap

transposable elements. Genes Dev. 9: 1797–1810.

Truernit, E., Siemering, K.R., Hodge, S., Grbic, V., and Haseloff, J.

(2006). A map of KNAT gene expression in the Arabidopsis root. Plant

Mol. Biol. 60: 1–20.

Webb, M.C., and Gunning, B.E.S. (1990). Embryo sac development in

Arabidopsis thaliana. 1. Megasporogenesis, including the microtubu-

lar cytoskeleton. Sex. Plant Reprod. 3: 244–256.

Webb, M.C., and Gunning, B.E.S. (1994). Embryo sac development in

Arabidopsis thaliana. 2. The cytoskeleton during megagametogene-

sis. Sex. Plant Reprod. 7: 153–163.

Willemse, M.T.M., and van Went, J.L. (1984). The Female Gameto-

phyte. (Berlin: Springer-Verlag).

Xu, J., Zhang, H.Y., Xie, C.H., Xue, H.W., Dijkhuis, P., and Liu, C.M.

(2005). EMBRYONIC FACTOR 1 encodes an AMP deaminase and is

essential for the zygote to embryo transition in Arabidopsis. Plant J.

42: 743–756.

Yamada, K., et al. (2003). Empirical analysis of transcriptional activity in

the Arabidopsis genome. Science 302: 842–846.

Yang, W.C., Ye, D., Xu, J., and Sundaresan, V. (1999). The

SPOROCYTELESS gene of Arabidopsis is required for initiation of

sporogenesis and encodes a novel nuclear protein. Genes Dev. 13:

2108–2117.

Yu, H.J., Hogan, P., and Sundaresan, V. (2005). Analysis of the female

gametophyte transcriptome of Arabidopsis by comparative expres-

sion profiling. Plant Physiol. 139: 1853–1869.

3592 The Plant Cell

DOI 10.1105/tpc.107.054890; originally published online November 30, 2007; 2007;19;3578-3592Plant Cell

Gabriela Carolina Pagnussat, Hee-Ju Yu and Venkatesan SundaresanBLH1Misexpression of the BEL1-Like Homeodomain Gene

Mutant Embryo Sacs Arises fromArabidopsis eostreCell-Fate Switch of Synergid to Egg Cell in

 This information is current as of April 17, 2020

 

Supplemental Data /content/suppl/2007/11/16/tpc.107.054890.DC1.html

References /content/19/11/3578.full.html#ref-list-1

This article cites 46 articles, 20 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists