Arabidopsis LORELEI, a Maternally-expressed ... - Plant Physiology

Plant Physiology Preview. Published on August 15, 2017, as DOI:10.1104/pp.17.00427

1

Running title: Parent-of-origin dependent expression of LORELEI

2 3

Author for correspondence:

4 5

Ravishankar Palanivelu

6

School of Plant Sciences, University of Arizona,

7

Tucson, AZ 85721, USA

8

[email protected]

9

Phone: 520-626-2229

10 11 12 13 14 15 16 17 18 19 20 21 22

1 Downloaded from on August 15, 2017 - Published by www.plantphysiol.org Copyright © 2017 American Society of Plant Biologists. All rights reserved.

Copyright 2017 by the American Society of Plant Biologists

23

Arabidopsis LORELEI, a Maternally-expressed Imprinted Gene, Promotes Early

24

Seed Development

25 26 27

Yanbing Wang1, Tatsuya Tsukamoto2, Jennifer A. Noble, Xunliang Liu3, Rebecca A.

28

Mosher, and Ravishankar Palanivelu*

29 30

School of Plant Sciences, University of Arizona, Tucson, AZ 85721, USA

31 32 33

One Sentence Summary:

34

LORELEI expression in the seed is paternally-silenced and the maternal-expression is

35

critical for early seed development

36 37

Author contributions:

38

Y.W., T.T., X.L., J.N., R.M., and R.P. designed the experiments; Y.W., T.T., X.L., and

39

J.N. performed the experiments; R.P., Y.W., and R.M. wrote the paper.

40 41

This work was supported by a grant from the US National Science Foundation to RP

42

(IOS- 1146090) and RAM (MCB-1243608).

43 44

Current Address:


45

1

46

Louis, MO 63130

47

2

48

160 Jackson Hall, 321 Church Street SE, Minneapolis, MN 55455.

49

3

Department of Biology, Washington University in St. Louis, One Brookings Drive, St.

Department of Genetics, Cell Biology, and Development, University of Minnesota, 6-

315 Life Sciences Center, 1201 Rollins Road, Columbia, MO 65211

50 51

*Address correspondence to: [email protected]

52 53

The author responsible for distribution of materials integral to the findings presented in

54

this article in accordance with the policy described in the Instructions for Authors

55

(www.plantphysiol.org) is: Ravishankar Palanivelu ([email protected]).

56 57

ABSTRACT

58 59

In flowering plants the female gametophyte controls pollen tube reception immediately

60

before fertilization and regulates seed development immediately after fertilization,

61

although controlling mechanisms remain poorly understood. Previously, we showed that

62

LORELEI (LRE), which encodes a putative glycosylphosphatidylinositol (GPI)-anchored

63

membrane protein, is critical for pollen tube reception by the female gametophyte before

64

fertilization and initiation of seed development after fertilization. Here, we show that LRE

65

is expressed in the synergid, egg, and central cells of the female gametophyte and in

66

the zygote and proliferating endosperm of the seed. Interestingly, LRE expression in the

67

seeds was primarily from the matrigenic LRE allele, indicating that LRE expression is


68

imprinted. However, LRE was biallelically-expressed in 8-day old seedlings, indicating

69

that the patrigenic allele does not remain silenced throughout the sporophytic

70

generation. Regulation of imprinted LRE expression is likely novel, as LRE was not

71

expressed in pollen or pollen tubes of mutants defective either for MET1, DDM1, RNA-

72

dependent DNA methylation, or MSI-dependent histone methylation. Additionally, the

73

patrigenic LRE allele inherited from these mutants was not expressed in seeds.

74

Surprisingly, and contrary to the predictions of the parental conflict hypothesis, LRE

75

promotes growth in seeds, as loss of the matrigenic but not the patrigenic LRE allele,

76

caused delayed initiation of seed development. Our results showed that LRE is a rare

77

imprinted gene that functions immediately after double fertilization and supported the

78

model that a passage through the female gametophyte establishes monoalleleic

79

expression of LRE in seeds and controls early seed development.

80 81

INTRODUCTION

82 83

The female gametophyte in flowering plants controls the transition from the

84

gametophyte to the sporophyte by multiple mechanisms. Before fertilization, gene

85

expression in the female gametophyte (hereafter called maternal expression) controls

86

pollen tube reception and sperm release. After double fertilization, maternally-derived

87

components maintain housekeeping functions, while matrigenic expression (arising from

88

the maternally-transmitted allele in the seed) plays a major role in embryo and

89

endosperm development (Chaudhury et al., 2001). Expression of a gene primarily or

90

exclusively from either the matrigenic or the patrigenic allele is called genomic


91

imprinting (Gehring, 2013). Imprinted genes that control developmental processes

92

through both maternal and matrigenic expression remain poorly characterized;

93

identifying these genes and their roles before and after fertilization will help us

94

understand how the female gametophyte controls early seed development.

95

Mutant analysis led to the identification of initial set of maternally expressed genes

96

(MEGs) and paternally expressed genes (PEGs) and the advent of transcriptomic

97

analysis has revealed numerous MEGs and PEGs in Arabidopsis, rice, and maize

98

(Gehring et al., 2011; Hsieh et al., 2011; Luo et al., 2011; Nodine and Bartel, 2012;

99

Raissig et al., 2013; Xin et al., 2013). Imprinting appears more common in endosperm

100

but also occurs in embryos (Jahnke and Scholten, 2009; Luo et al., 2011; Ngo et al.,

101

2012; Nodine and Bartel, 2012; Raissig et al., 2013; Pignatta et al., 2014). However, the

102

interpretation of transcriptomic studies can be confounded by contamination from the

103

maternal seed coat (Schon and Nodine, 2017). A few of the MEGs or PEGs that

104

function in seeds are also expressed in the mature female or male gametophyte,

105

respectively, where they might mediate double fertilization.

106

Genomic imprinting is controlled by differential epigenetic modification of the

107

matrigenic and patrigenic alleles (Gehring, 2013; Kawashima and Berger, 2014). In

108

case of MEGs, inhibitory epigenetic modifications are maintained on the paternal allele

109

in the male gametophyte and selectively removed from the maternal allele in the female

110

gametophyte. Consequently, in the seed, only the matrigenic allele is expressed and the

111

patrigenic allele remains silenced. Epigenetic modifications that underlie imprinting are

112

typically associated with DNA or histone methylation (Gehring, 2013).


113

The evolutionary and functional significance of imprinting remains unclear. As per

114

the parental conflict hypothesis, parent-of-origin effects are the outcome of conflict

115

between matrigenic and patrigenic alleles in resource allocation: matrigenic alleles favor

116

limited but equitable growth among sibling seeds that have the same mother, while the

117

patrigenic alleles enhance growth at the expense of siblings with different fathers (Haig,

118

2013). Loss of matrigenic expression of some MEGs promotes seed growth, providing

119

support for this hypothesis; however, not all MEGs follow this pattern (Bai and Settles,

120

2015). Identifying all the imprinted genes is critical to test the parental conflict

121

hypothesis and explore alternative theories of imprinting (Spencer and Clark, 2014).

122

LORELEI (LRE), which encodes a putative glycosylphosphatidylinositol (GPI)-

123

anchored membrane protein, is critical for pollen tube reception by the female

124

gametophyte before fertilization (Capron et al., 2008; Tsukamoto et al., 2010; Liu et al.,

125

2016). Early seed development is delayed in lre-5/lre-5 ovules that successfully induce

126

pollen tube reception, indicating that LRE also plays a role in timely initiation of seed

127

development after fertilization (Tsukamoto et al., 2010). Consistent with these LRE

128

functions, using RT-PCR experiments, we previously showed that LRE expression is

129

temporally and spatially regulated during reproduction (Tsukamoto et al., 2010); before

130

fertilization, LRE is expressed in mature unfertilized ovules but not in pollen or pollen

131

tubes. After fertilization, LRE is expressed in ovules up to 24 hours after pollination

132

(HAP) and is not detectable in ovules at 36 and 48 HAP. Yet, important questions

133

remain to be answered. Like some female gametophyte-expressed genes that play a

134

role in seed development after fertilization (Evans and Kermicle, 2001; Yadegari and

135

Drews, 2004; Gehring, 2013; Bai and Settles, 2015; Chettoor et al., 2016), it is not


136

known if only the matrigenic allele of LRE is expressed in the seeds. Additionally, it is

137

not clear if loss of LRE expression in the maternal sporophyte or female gametophyte or

138

seed leads to the delayed seed development.

139

Here, we showed that LRE is expressed in the zygote and in the proliferating

140

endosperm at least up to 24 hours after pollination (HAP). We also showed that LRE

141

expression is imprinted in both zygote and endosperm soon after double fertilization. A

142

novel mechanism might control imprinting of LRE expression, as the patrigenic allele of

143

LRE remained silent when inherited from mutants defective in DNA or histone

144

methylation. Loss of matrigenic LRE, but not the patrigenic LRE, caused delays in

145

initiation of embryo and endosperm development, indicating that LRE is a rare imprinted

146

gene that functions immediately after double fertilization. Our study showed that LRE

147

mediates maternal control of two critical events during the transition from the

148

gametophyte to the sporophyte generation – pollen tube reception and seed

149

development.

150 151

RESULTS

152 153

LRE Expression in Seeds Is Primarily from the Matrigenic Allele

154

Some female gametophyte-expressed genes that play a role in seed

155

development after fertilization are imprinted, resulting in the synthesis of transcripts

156

primarily from the matrigenic allele in the seed (Evans and Kermicle, 2001; Yadegari

157

and Drews, 2004; Gehring, 2013; Bai and Settles, 2015; Chettoor et al., 2016). LRE

158

functions in both the female gametophyte and the seed (Tsukamoto et al., 2010). We


159

therefore tested if LRE expression is monoallelic in seeds by reciprocally crossing lre-5

160

with wild type. We chose lre-5 mutant for this experiment, as lre-5 is a null allele of LRE

161

and LRE transcripts are not produced (Tsukamoto et al., 2010). RT-PCR analysis of

162

ovules isolated 24 HAP showed that ACTIN2 transcripts were expressed whether lre-5

163

or wild type was used as a female parent. However, LRE transcripts were detected only

164

when the female parent was wild type (Fig. 1A), suggesting that primarily the matrigenic

165

LRE allele, but not the patrigenic LRE allele, is expressed after fertilization.

166

To obtain additional evidence in support of this finding, we performed reciprocal

167

crosses between two Arabidopsis accessions (Columbia and C24) with a single

168

nucleotide polymorphism (SNP) in LRE. Twenty four HAP, we isolated ovules and used

169

Allele-Specific PCR (AS-PCR) involving Locked Nucleic Acid (LNA; (Latorra et al.,

170

2003)) primers to SNP genotype and distinguish whether endogenous LRE transcripts

171

originated from the Columbia or the C24 allele. Control PCR reactions using Columbia

172

and C24 genomic DNA identified annealing temperature at which LNA primers can be

173

reliably used to perform SNP genotyping by AS-PCR (Fig. 1B, left panel).

174

We used GRP23 expression as a biallelic expression control in SNP genotyping

175

by AS-PCR (Fig. 1B) because GRP23 is expressed during early seed development,

176

starting from the zygote and endosperm nuclear proliferation stages (Ding et al., 2006).

177

We chose to SNP genotype ovules at 24 HAP, as GRP23 is expressed in seeds at least

178

from 16 HAP, even when transmitted through pollen (Tsukamoto et al., 2010). As

179

expected, in seeds, GRP23 was biallelically expressed (Fig. 1B, top two right panels).

180

Importantly, identification of accession-specific expression of GRP23 from the patrigenic

181

allele in seeds confirmed the sensitivity of this assay in detecting contributions from the


182

patrigenic allele in the embryo sac despite the presence of a large amount of maternal

183

sporophytic tissues in ovules. In this assay, Columbia-specific LNA LRE primers

184

amplified a LRE PCR product only when Columbia was the female parent and not when

185

it was used as a male parent (Fig. 1B, bottom two right panels). Conversely, C24-

186

specific LNA primers amplified a LRE PCR product only when C24 was the female

187

parent and not when it was used as a male parent (Fig. 1B, bottom two right panels).

188

Based on these results, we concluded that endogenous LRE transcripts in 24 HAP

189

ovules primarily originated from the matrigenic LRE allele.

190 191

Matrigenic LRE Allele Is Expressed in Proliferating Endosperm and Zygote-Like Cell

192

Ovules from pollinated pistils used in RT-PCR assays in Figure 1 comprise of

193

both maternal sporophytic tissues and embryo sacs. LRE expression after fertilization

194

can arise from expression in either one or both tissues. Additionally, parent-of-origin

195

dependent gene expression has been reported in both endosperm (Gehring, 2013; Bai

196

and Settles, 2015) or embryo (Jahnke and Scholten, 2009; Nodine and Bartel, 2012;

197

Raissig et al., 2013; Del Toro-De Leon et al., 2014), two distinct cell types within an

198

embryo sac. The RT-PCR and AS-PCR assays in Fig. 1 could not distinguish LRE

199

expression in maternal sporophytic tissues of ovules from expression in the developing

200

embryo sac nor can it rule out contamination of seeds with RNA from maternal tissues,

201

as was reported for endosperm and early embryo transcriptomes in Arabidopsis (Schon

202

and Nodine, 2017). We overcame this shortcoming by performing cell-specific

203

expression analysis using a promoter:reporter fusion. To examine the spatial and

204

temporal expression of LRE expression in seeds, we generated plants carrying a


205

pLRE::GFP reporter transgene (Fig. 2A). Prior to performing expression analysis in

206

seeds, we checked for GFP signal in synergid cells of unfertilized ovules, where LRE

207

functions in pollen tube reception (Capron et al., 2008; Tsukamoto et al., 2010; Liu et

208

al., 2016). The GFP expression was strong in synergid cells of unfertilized ovules (Fig.

209

2B-D and Supplemental Table S1 and Supplemental Movie S1 and Movie S2),

210

consistent with LRE function in pollen tube reception in the synergid cell. No GFP

211

expression was detected in pollen or pollen tubes carrying the pLRE::GFP transgene (n

212

>1000), consistent with RT-PCR analysis and phenotypic analysis that showed no

213

function for LRE in pollen tubes (Tsukamoto et al., 2010).

214

When the pLRE::GFP transgene was maternally transmitted in a cross with wild-

215

type pollen, GFP expression was detected in the proliferating endosperm (Fig. 2E-I and

216

Supplemental Table S1 and Supplemental Movie S3). Additionally, when the

217

pLRE::GFP transgene was maternally transmitted, GFP expression was detected in the

218

zygote-like cell in the micropylar end of seeds (Supplemental Movie S3; also see

219

below). LRE promoter activity is dynamic in the seed, as GFP expression level

220

increased (Fig. 2E-G and Supplemental Table S1) and then decreased (Fig. 2G-I and

221

Supplemental Table S1) over time after pollination. GFP expression in the endosperm

222

and the zygote-like cell was not detected when the pLRE::GFP transgene was

223

paternally contributed (Fig. 2J-O), confirming that LRE expression in the seeds is

224

primarily from the matrigenic allele. These results also show that the LRE promoter

225

used in the pLRE::GFP transgene is sufficient to recapitulate the matrigenic allele-

226

specific expression of LRE in seeds and likely contains all of the cis-elements required

227

for the monoallelic expression of LRE in seeds. Additionally, there was no GFP


228

expression in the maternal sporophytic tissues of seeds such as integuments (seed

229

coat) and funiculus, either before or after fertilization (Fig. 2E-2I). Based on these

230

results, we infer that the LRE transcripts detected in RT-PCR experiments using seeds

231

(Fig. 1) must have been primarily from the embryo sac rather than the maternal

232

sporophytic tissues. Taken together, our results indicate that LRE expression lasts only

233

for a short duration after double fertilization and that LRE is a MEG in the both

234

fertilization products.

235 236

LRE Is Expressed in the Zygote

237

The GFP expression in the zygote-like cell in the micropylar end of the seed

238

could be either from the zygote or continued expression in the persistent synergid cell

239

that briefly lingers after double fertilization (Volz et al., 2013; Maruyama et al., 2015). To

240

distinguish between these two possibilities, we performed co-localization of LRE

241

expression from pLRE::DsRed with a zygote marker (pWOX8::gWOX8-YFP; (Ueda et

242

al., 2011). The co-localization analysis was done at a developmental stage (seeds with

243

≥4 endosperm nuclei) when the persistent synergid in most of the seeds has already

244

degenerated (Volz et al., 2013) and any residual pLRE:DsRed expression in a single

245

persistent synergid cell will not confound our analysis of pLRE:DsRed expression in the

246

zygote. We reasoned that in ovules with ≥4 endosperm nuclei, any DsRed expression in

247

the micropylar end must be from a cell in the embryo sac of a seed (i.e. zygote).


248

We established a pLRE::DsRed transgene (Fig. 3A), in which DsRed reporter

249

was expressed from the same LRE promoter used in pLRE::GFP, and transformed it

250

into plants carrying the pWOX8::gWOX8-YFP transgene. In the transgenic plants

251

carrying both markers, we chose three lines for co-localization analysis. Unlike in

252

pLRE:GFP transgenic lines, in pLRE::DsRed lines, the DsRed levels were barely above

253

the background autofluorescence in the endosperm nuclei of the seed. However, in the

254

zygote-like cell in the micropylar end of the seed, the DsRed signal was clearly above

255

the background autofluorescence (Fig. 3C and Supplemental Table S2), which is

256

sufficient to complete the co-localization experiments with WOX8-YFP and determine if

257

LRE is expressed in the zygote. Lack of pLRE::DsRed transgene expression in the

258

proliferating endosperm did not prevent us from staging the seed development, as we

259

were able to overcome this shortcoming by scoring WOX8-YFP expression, which is

260

also expressed in proliferating endosperm nuclei (our analysis of the pWOX8::gWOX8-

261

YFP lines in this study), in addition to the zygote (Ueda et al., 2011).

262

Pistils carrying both reporter genes were pollinated with wild-type pollen and

263

ovules were scored for co-localization of DsRed and YFP expression 13.5 HAP, a time

264

point at which ~70% of the ovules (314/478) are fertilized, as indicated by developing

265

endosperm in them. Of these 314 seeds, 210 contained ≥4 endosperm nuclei. Our

266

analysis showed that 89% of these ovules with ≥4 endosperm nuclei (187/210)

267

expressed both DsRed and YFP in a single cell in the micropylar end of the ovule (Fig.

268

3C-D), indicating that LRE is expressed in the zygote. Similar to the pLRE::GFP

269

construct, pLRE::DsRed expression was also dynamic in the seed, as DsRed


270

expression level in the zygote increased and then decreased over time after pollination

271

(Supplemental Table S2). DsRed expression in the the zygote was not detected when

272

the pLRE::DsRed transgene was paternally contributed (Supplemental Table S3),

273

confirming that LRE expression in the zygote is primarily from the matrigenic allele.

274

Based on these results, in conjunction with the matrigenic allele-specific expression of

275

LRE (Figs. 1 and 2 and Supplemental Table S3), we concluded that LRE expression is

276

paternally imprinted in both the zygote and the proliferating endosperm of seeds.

277 278

de novo transcription after fertilization likely results in increased expression of LRE in

279

the zygote and proliferating endosperm

280

LRE expression in proliferating endosperm and zygote (Fig. 2 and Fig. 3) could

281

be due to transcripts that were transcribed in the two female gametes and inherited into

282

the corresponding fertilization products and/or due to de novo transcription of LRE after

283

fertilization in the zygote and proliferating endosperm. To distinguish between these

284

possibilities, we examined pLRE::GFP expression in the two female gametes before

285

fertilization. We found that the LRE promoter is active in the egg (Fig. 2, B and C and

286

Supplemental Movie S1 and Movie S2) and central cells (Fig. 2, B and C), although the

287

level of GFP expression in these cells was noticeably lower than that in synergid cells

288

(Fig. 2, B and C and Supplemental Table S1 and Supplemental Movie S1 and Movie

289

S2) but clearly above the background fluorescence in non-transgenic Columbia ovules

290

(Supplemental Fig. S1). GFP expression in central cells was observed in noticeably

291

fewer ovules compared to those that expressed GFP in synergid and egg cells (Fig. 2D

292

and Supplemental Table S1). Importantly, the GFP expression was higher in the


293

proliferating endosperm than in the central cell, the cell from which endosperm is

294

derived after fertilization (compare Fig. 2G with Fig. 2C or Fig. 2E). Similarly, the GFP

295

expression was higher in the zygote than in the egg cell, the cell from which zygote is

296

derived after fertilization (compare Fig. 2G with Fig. 2C or Fig. 2E). These results

297

suggest that there is de novo expression of the matrigenic pLRE::GFP allele in the

298

proliferating endosperm and zygote after fertilization.

299

We also examined pLRE::DsRed expression in the female gametophyte and

300

found that DsRed levels were barely above the background autofluorescence in the

301

central cell of the female gametophyte (Fig. 3 and Supplemental Table S2). Since the

302

DsRed expression was also barely detectable in the proliferating endosperm, DsRed

303

expression could not be used to examine the de novo transcription in the proliferating

304

endosperm. However, the DsRed expression was higher in the zygote than in the egg

305

cell, the cell from which zygote is derived after fertilization (compare Fig. 3C with Fig.

306

3B). Increased pLRE::GFP expression in the zygote and proliferating endosperm and

307

the increased pLRE::DsRed expression in the zygote indicate that de novo transcription

308

after fertilization is likely a major contributing factor towards the increased LRE

309

expression in the seed.

310 311

LRE Is Biallelically Expressed In 8 Day-Old Seedlings

312

Our results showed that the paternal and patrigenic allele is silenced in the male

313

gametophyte and during early seed development, respectively, raising the possibility

314

that the patrigenic allele remains silent throughout the sporophytic generation.

315

Examining LRE expression in vegetative tissues is one way to test this possibility.


316

Previous RT-PCR experiments showed that LRE is expressed in 8 day-old seedlings

317

(Tsukamoto et al., 2010). Therefore, we examined if the patrigenic LRE allele is

318

expressed in 8-day old seedlings. We reciprocally crossed wild type and lre-5 and did

319

RT-PCR experiments using 8 day-old seedlings of F1 progeny from these crosses. As

320

in Fig. 1A, for this experiment also we chose a null allele of LRE to reliably identify the

321

source of detected LRE transcripts. RT-PCR experiments showed that LRE is

322

expressed in 8 day-old seedlings of the F1 progeny regardless if wild type is used as a

323

male or female parent (Fig. 4), indicating that LRE is biallelically-expressed in 8 day-old

324

seedlings. Similar results were obtained when a second null allele (lre-7) was used in

325

this experiment (Fig. 4). This finding is in marked contrast to the monoallelic expression

326

of LRE in seeds (Fig. 1A). These results combined with our observation that LRE is

327

expressed in the female gametophyte but not the male gametophyte indicate that LRE

328

expression is imprinted during the gametophytic generation (at some point after male

329

gametogenesis) and that restoration of biallelic LRE expression occurs during the

330

sporophytic generation (at some point during embryogenesis or after seed development

331

or germination).

332 333

DNA Methylation Pathways that Regulate MEGs Do Not Control Imprinted Expression

334

of LRE

335

During development, DNA methylation of both maternal and paternal alleles of

336

many genes is primarily maintained by METHYLTRANSFERASE1 (MET1, CG sites)

337

(Kishimoto et al., 2001) and CHROMOMETHYLASE3 (CMT3, CHG sites) (Lindroth et

338

al., 2001). The chromatin remodeler DECREASE IN DNA METHYLATION 1 (DDM1) is


339

also required for the maintenance of CG and non-CG (CHG and CHH) methylation

340

(Johnson et al., 2002; Stroud et al., 2013). At many MEG loci, selective removal of DNA

341

methylation at the maternal allele is achieved by the activity of the DNA glycosylase

342

DEMETER (DME) in the female gametophyte, resulting in differentially methylation of

343

maternal and paternal alleles (Choi et al., 2002; Gehring et al., 2006). At some MEG

344

loci, RNA-directed DNA methylation silences the paternal allele and thereby sets up the

345

MEG expression pattern (Vu et al., 2013). Such methylation differences in the two

346

alleles are the basis of differences in transcription, as the hypermethylated patrigenic

347

allele and hypomethylated matrigenic allele are usually associated with inactive and

348

active transcriptional states, respectively.

349

To test if differential DNA methylation establishes imprinting of LRE, we examined

350

whether the patrigenic allele of LRE is expressed in seeds when inherited from pollen of

351

three DNA hypomethylation mutants: met1-1 (defective in CG methylation), ddm1-2

352

(defective in CG and non-CG methylation), and drm1-2 drm2-2 cmt3-11 (ddc) (defective

353

in CHG and CHH methylation) (Kakutani et al., 1996; Yadegari et al., 2000; Kankel et

354

al., 2003; Chan et al., 2006). We pollinated lre-5 pistils with hypomethylated mutant

355

pollen and examined endogenous LRE expression in the ovules from pollinated pistils in

356

a RT-PCR assay. We examined expression of PHERES1, a gene that is primarily

357

expressed from the patrigenic allele after fertilization (Kohler et al., 2005) as a control

358

for the sensitivity of this RT-PCR assay to detect expression from a patrigenic allele in

359

seeds. ACTIN11 expression was examined as a control for a gene that is biallelically-

360

expressed in seeds (Huang et al., 1997). Our results showed that hypomethylation is

361

not sufficient to induce expression from the patrigenic allele of LRE (Fig. 5A and


362

Supplemental Fig. 2A), even though we detected PHERES1 (Fig. 5A) or ACTIN11

363

(Supplementary Fig. S2A) expression in these crosses.

364

When each of the three hypomethylated mutant pollen carrying pLRE::GFP was

365

crossed onto wild-type pistils, we did not detect pLRE::GFP expression in the seeds,

366

indicating that the patrigenic pLRE::GFP allele remained silent following demethylation

367

(Supplemental Table S4). GFP expression from pLRE::GFP was detected in seeds from

368

corresponding reciprocal crosses in which the hypomethylated pollen carrying

369

pLRE::GFP was used as a female parent (Supplemental Table S4). We performed

370

these crosses with pollen from heterozygous hypomethylated mutant carrying

371

pLRE::GFP, as (i) it will help examine potential roles of the male gametophyte in the

372

silencing of the patrigenic allele and (ii) sibling wild-type pollen can serve as an internal

373

control for the GFP expression in seeds in each cross. Furthermore, there was no

374

ectopic expression of the pLRE::GFP transgene in mature pollen (n > 1000) or pollen

375

tubes (n > 200) grown through a cut pistil (Palanivelu and Preuss, 2006) in any of the

376

three hypomethylated mutants, indicating that demethylation is not sufficient to express

377

paternal pLRE::GFP. Additionally, we also found that demethylation mediated by DME

378

in the female gametophyte is not required to activate LRE expression from the

379

matrigenic allele in the seed (Supplemental Table S5).

380

Female gametophyte-derived factors are also known to regulate expression of

381

patrigenic alleles in seeds. GLAUCE function in the female gametophyte is required for

382

the expression of the patrigenic allele of RPS5a and FAC1 in developing embryos (Ngo

383

et al., 2007). Recently, it was shown that maternal RdDM pathway mediates silencing of

384

the patrigenic allele at MEG loci; loss of the maternal DCL3 resulted in the activation of


385

the patrigenic AtBMI1C allele in the endosperm, but not in the embryo, of hybrid seeds

386

from a cross between dcl3-1 and wild type (Bratzel et al., 2012). To investigate if the

387

function of maternal RdDM pathways is required for repressing expression from the

388

patrigenic LRE allele, we crossed wild-type pollen carrying pLRE::GFP onto the DNA

389

hypomethylation mutant pistils. To increase the chance of identifying a seed with

390

pLRE::GFP expression from the patrigenic allele, we performed these crosses between

391

homozygous hypomethylation mutant pistils and pollen from plants that are

392

homozygous for the pLRE::GFP transgene. At 13.5 HAP, we scored GFP expression in

393

the seeds of these crosses. None of the hypomethylation mutants when used as a

394

female parent induced the expression of pLRE::GFP in seeds (Table 1). These results

395

demonstrate that DNA methylation by pathways known for some MEGs do not control

396

the imprinting of LRE.

397 398

Histone Methylation Pathways That Regulate MEGs Do Not Control Imprinted

399

Expression of LRE

400

Gene imprinting can also be mediated by differential histone modification. In

401

Arabidopsis endosperm and embryo, differential methylation of lysine 27 on histone H3

402

(H3K27me3) establishes monoallelic expression of some MEGs (Jullien et al., 2006;

403

Raissig et al., 2013). At the paternal alleles of these loci, H3K27me3 is selectively

404

maintained by Polycomb Repressive Complex 2 (PRC2) to repress patrigenic

405

expression after fertilization (Jullien et al., 2006; Raissig et al., 2013). To determine if

406

H3K27me3 silences the paternal or patrigenic allele of LRE, we used the msi1 mutant,

407

which is defective in one of the four core subunits of the PRC2 complex (Köhler et al.,


408

2003; Guitton et al., 2004). RT-PCR experiments, in which lre-5 pistils were pollinated

409

with pollen from msi1 mutant, revealed that the patrigenic LRE allele is not expressed

410

even when inherited from msi1 mutant pollen (Fig. 5B and Supplemental Fig. S2B). Our

411

results also showed that msi1 does not cause ectopic expression from the paternal

412

pLRE::GFP transgene in mature pollen (n > 1000) or pollen tubes grown through a cut

413

pistil (n > 200), or from the patrigenic pLRE::GFP allele in seeds (Supplemental Table

414

S6).

415

Loss of SUVH4 histone methyltransferase KRYPTONITE (KYP) function in the

416

female gametophyte led to increased and earlier expression of the patrigenic alleles of

417

RPS5a, AGP18, PROLIFERA, and GRP23 in the seeds of crosses between kyp mutant

418

pistils and wild-type pollen (Autran et al., 2011). In other instances, maternal PRC2

419

activity is required for continued repression of the silent patrigenic allele of two MEGs

420

after fertilization (Raissig et al., 2013). To investigate if maternal PRC2 activity is

421

required for repressing expression from the patrigenic LRE allele, we crossed wild-type

422

pollen carrying pLRE::GFP onto msi1 mutant pistils. At 13.5 HAP, we scored seeds in

423

these crosses for GFP expression. Even when msi1 mutant was used as a female

424

parent, there was no induction in expression of pLRE::GFP in the seeds (Table 1).

425

Based on these results, we concluded that LRE is not imprinted through known histone

426

modification pathways that are known to regulate some MEGs in Arabidopsis.

427 428

Loss of Expression from the Matrigenic LRE Allele Results in Delayed Early Seed

429

Development


430

Previously, using embryo and/or endosperm markers, we showed that initiation

431

of embryo and endosperm development is delayed in homozygous lre mutant seeds

432

(Tsukamoto and Palanivelu, 2010; Tsukamoto et al., 2010). Importantly, we showed that

433

this late start was not caused from a delay in i) pollen tube arrival at the female

434

gametophyte, ii) completion of pollen tube reception, or iii) double fertilization; instead,

435

the delay was in the initiation of early seed development after double fertilization

436

(Tsukamoto et al., 2010). RT-PCR and genetic assays also established that LRE is not

437

expressed and does not function, respectively, in the male gametophyte (Tsukamoto et

438

al., 2010). Above we demonstrated that patrigenic allele of LRE is silenced in seeds

439

(Figs. 1 and 2), indicating that loss of the matrigenic LRE allele, rather than the

440

patrigenic LRE allele, leads to the delayed initiation of seed development. Still, this

441

possibility remains to be tested. Additionally, complementation experiments to

442

conclusively demonstrate that LRE is required for timely initiation of seed development

443

have not been performed.

444

To address these questions, we examined endosperm and embryo development

445

in seeds using a well-established chloral hydrate-based clearing assay (Yadegari et al.,

446

1994), as it offered two advantages over marker-based analysis of seed development.

447

First, by facilitating counting of cells/nuclei in developing embryo and endosperm, this

448

clearing assay allowed accurate determination of embryo and endosperm development

449

stage (Supplemental Fig. 3A and 3B). Second, this assay also allowed scoring pollen

450

tube reception (manifested as coiled tubes) in every ovule, even if it is fertilized. By

451

scoring both pollen tube reception and seed development simultaneously in an ovule,

452

we identified seeds that initiated embryo development after undergoing normal pollen


453

tube reception (type 1) and distinguished them from two other types of ovules:

454

unfertilized lre mutant ovules that do not undergo pollen tube reception (type 2;

455

Supplemental Fig. S3C) and those in which embryo development has initiated but had a

456

coiled tube in it (type 3; Supplemental Fig. S3D). In this study, to examine embryo and

457

endosperm development analysis without any confounding effects from pollen tube

458

reception defects, we excluded types 2 and 3 ovules from our analysis and used only

459

type 1 ovules (reported in Figs. 6 and 7; Supplemental Figs. S4 and S5).

460

Using chloral hydrate-based clearing assay, we first confirmed delayed initiation

461

of seed development in lre seeds (reported previously using GRP23:GUS reporter-

462

based experiments; (Tsukamoto et al., 2010)) by crossing lre-7 pistil with wild type

463

pollen and monitoring development of endosperm and embryo in seeds at 48 HAP.

464

Loss of LRE delayed endosperm nuclear proliferation in 48 HAP ovules, as evident from

465

significantly higher number of lre-7 seeds with ≤ 5 nuclear divisions compared to wild-

466

type ovules (Supplemental Fig. S4). Embryo development was also delayed in 48 HAP

467

ovules; significantly more lre-7 seeds contained an embryo with ≤ 1 cell division (zygote

468

or proembryos with 1 or 2 cells) compared to wild-type ovules (Fig. 6A). Next, we used

469

this assay to perform in-depth analysis of seed development by focusing only on

470

embryo development. The delay in embryo development was not observed if the lre-7

471

mutation was paternally contributed to the seeds; instead, the delay was observed in 48

472

HAP ovules only if they carried the matrigenic lre-7 allele (Fig. 6A and Fig. 6C). Similar


473

delay in embryo development was also observed when we used lre-5, another loss of

474

function LRE allele (Tsukamoto and Palanivelu, 2010; Tsukamoto et al., 2010); like our

475

observations using lre-7, delay in embryo development was seen only with a loss of the

476

matrigenic lre-5 allele but not the patrigenic lre-5 allele (Supplemental Fig. S5).

477

To complement the delay in initiation of embryo development defect in lre-7

478

seeds, we generated pLRE::LRE-HA transgene, transformed into lre-7 plants, identified

479

single insertion lines, and demonstrated first that the transgene is functional; presence

480

of pLRE::LRE-HA transgene restored seed set in lre-7 plants to wild-type levels,

481

presumably by rescuing pollen tube reception defect in these plants (Supplemental

482

Tables S7-S9). We then reciprocally crossed transgenic line carrying pLRE::LRE-HA

483

with wild-type pollen and found that delayed embryo development defect can be

484

complemented by supplying the matrigenic, but not the patrigenic pLRE::LRE-HA

485

transgene (Figs. 6A and 6C), indicating that loss of expression from the matrigenic LRE

486

allele results in delayed early seed development.

487 488

Delay in Seed Development is Not Caused by the Loss of LRE Expression in the

489

Maternal Sporophyte

490

An alternative explanation for the observations in Fig. 6 is that loss of LRE

491

expression in the maternal sporophyte resulted in delayed seed development, as

492

homozygous lre ovules were used in these experiments and they contain mutant

493

sporophytic tissues enclosing mutant gametophytes. However, this possibility is unlikely

494

because LRE is not expressed in the maternal sporophyte (Fig. 2). To confirm the effect

495

on embryo development is due to maternal or matrigenic LRE expression and not due


496

to LRE expression in the female sporophyte, we crossed lre heterozygous pistils with

497

wild-type pollen and scored embryo development in cleared ovules. In heterozygous

498

pistils, there is a functional LRE allele only in 50% of female gametophytes even though

499

diploid female sporophyte surrounding every ovule has one functional LRE allele. To

500

survey the effects of loss of LRE in the female sporophyte on seed development, we

501

first crossed pollen carrying pGRP23::GUS transgene (Ding et al., 2006) with wild-type

502

and lre-5 heterozygous pistils. In wild-type pistils crossed with pGRP23::GUS pollen,

503

98.4% of seeds showed normally-developed embryos (Fig. 7A). However, in lre-5

504

heterozygous pistils (Fig. 7B), among all the seeds, 18.8% (70/373) of seeds were

505

delayed in embryo development (Fig. 7D), while the remainder had normal embryo

506

development (Fig. 7C). Detection of seeds with delayed embryo development in lre-5

507

heterozygous pistils indicates that loss of LRE in the maternal sporophyte is not related

508

to delayed embryo development.

509

To confirm these results, we performed this experiment using the chloral hydrate

510

assay and lre-7 heterozygous pistils. The number of seeds with delayed embryo

511

development in crosses with lre-7 heterozygous pistils is significantly higher than in

512

crosses with wild-type pistils (Fig. 7E), indicating that delayed early embryo

513

development is not caused by the female sporophyte. Still, these experiments cannot

514

distinguish between loss of maternal LRE expression in the female gametophyte and

515

loss of matrigenic LRE expression in the developing seed. However, our observations of

516

increased LRE expression in the zygote and endosperm after fertilization compared to

517

the LRE expression in the egg and central cell before fertilization (Fig. 2) suggest that


518

loss of expression from the matrigenic LRE allele in seeds causes delayed early seed

519

development.

520 521

DISCUSSION

522 523

LRE Expression is Imprinted in the Zygote and Endosperm Immediately After Double

524

Fertilization

525

Soon after fertilization, in the zygote and proliferating endosperm, there is

526

preferential expression from the matrigenic allele of LRE. This expression is likely from

527

de novo LRE transcripts generated after fertilization, as the GFP expression level in the

528

zygote and endosperm is higher than that in the egg cell and the central cell,

529

respectively (Fig. 2). LRE expression in the seeds is not only primarily from the

530

matrigenic allele but is also detectable only for a short duration. LRE transcripts were

531

detected at 24 HAP but not detected at 36 HAP or 48 HAP in seeds of manually selfed

532

crosses (Tsukamoto et al., 2010) or up to 24 HAP in reciprocal crosses between

533

pLRE::GFP and wild type (Fig. 2), suggesting that sometime between 24 and 36 HAP,

534

LRE expression ceases in seeds. Additionally, based on publicly available RNA-seq

535

data, LRE is not expressed in embryos at approximately 40, 64, and 78 HAP (1-cell/2-

536

cell, 8-cell, and 32-cell) (Nodine and Bartel, 2012) and in the endosperm of seeds 6-8

537

DAP (Gehring et al., 2011; Hsieh et al., 2011). Subsequently in development, LRE is

538

expressed in 8-day old seedlings, at which point it is bialleically expressed. Still, without

539

determining the expression of LRE in seeds from 9 DAP to mature stage, the precise


540

stage of plant development when biallelic expression of LRE is restored cannot be

541

established.

542

Imprinted expression of LRE shares similarities and differences with other MEGs

543

in Arabidopsis. Monoallelic expression during reproduction (Fig. 2) and biallelic

544

expression of LRE in 8 day-old seedlings (Fig. 4) indicate that like other MEGs,

545

epigenetic reprogramming for imprinted expression and subsequent restoration of

546

biallelic LRE expression must occur during gametogenesis and vegetative growth,

547

respectively (Jahnke and Scholten, 2009; Kawashima and Berger, 2014; Boavida et al.,

548

2015). Additionally, like other MEGs, LRE is not expressed in pollen [(Tsukamoto et al.,

549

2010; Loraine et al., 2013); this study] or pollen tubes [(Tsukamoto et al., 2010); this

550

study]. LRE is also distinct from other MEGs in other aspects, as its expression is

551

imprinted in both zygote and proliferating endosperm. Additionally, monoallelic

552

expression of LRE expression after fertilization in the developing seed is earlier than

553

reported for other MEGs (Jahnke and Scholten, 2009; Ngo et al., 2012; Nodine and

554

Bartel, 2012; Raissig et al., 2013). Because the patrigenic allele of LRE remains silent in

555

various stages of seed development after 24 HAP (Tsukamoto et al., 2010; Gehring et

556

al., 2011; Hsieh et al., 2011; Nodine and Bartel, 2012; Raissig et al., 2013), our

557

observations also cannot be explained by delayed paternal genome activation (Autran

558

et al., 2011; Del Toro-De Leon et al., 2014; Garcia-Aguilar and Gillmor, 2015). Based on

559

these observations, we conclude that LRE expression is imprinted in both zygote and

560

endosperm.

561 562 563

Silencing of the Paternal Allele of LRE May Be Controlled by a Novel Pathway


564

Imprinting of LRE likely occurs at some point during female gametogenesis,

565

which suggests that passing through the female gametophyte could result in differential

566

modification of the maternal LRE allele compared to the paternal LRE allele. For

567

example, the paternal allele of LRE could be preferentially hypermethylated, either in

568

the UTR or the gene body region, as is the case in many MEGs (Gehring, 2013; Bai and

569

Settles, 2015). However, our data showed that this is not the case for LRE, as paternal

570

imprinting of LRE is not affected in DNA methylation pathway mutants, including met1-

571

1, ddm1-2, and ddc. Conversely, maternal imprinting of LRE is also not affected in the

572

dme mutant (Supplemental Table S2), which disrupts DNA de-methylation pathway in

573

the female gametophyte. Known pathways that differentially modify histone methylation

574

are also likely not involved in imprinting LRE expression, as silencing of the paternal

575

allele of LRE is not reversed in the msi1 mutant, which disrupts PRC2-dependent

576

histone modification (H3K27me3). These results indicate that imprinting of LRE

577

expression is controlled by yet to be characterized novel pathway. For example, it could

578

be due to an imprinting-like phenomenon proposed for certain transcripts in early

579

embryos of Arabidopsis (Nodine and Bartel, 2012), as imprinted LRE expression in

580

seeds is short-lived. Nevertheless, our results point to the fact that maternal LRE allele

581

is imprinted at some point during female gametogenesis, which then sets up monoallelic

582

expression of the matrigenic allele in the seeds.

583 584 585 586

Loss of Expression from the Matrigenic LRE Allele Causes Delayed Seed Development

587

functional significance remain unclear. The parental-conflict/kinship-conflict theory

Imprinting is well established in several genes; however, its evolutionary and


588

provides a plausible reason for prevalence of imprinting (Haig and Westoby, 1989,

589

1991; Haig, 2013). As per this theory, parent of origin effects is the outcome of the

590

conflict between patrigenic and matrigenic alleles in influencing resource allocation:

591

matrigenic alleles favor equitable resource allocation among all sibling seeds and thus

592

tend to promote smaller seeds, while the patrigenic alleles favor larger seeds and are

593

less constrained by costs to sibling seeds (Haig and Westoby, 1989, 1991; Haig, 2013).

594

Endosperm proliferation and increase in seed growth when mutations in maternally-

595

expressed imprinted genes including FIS2, FIE, and MEA are maternally inherited

596

(Grossniklaus et al., 1998; Kohler and Makarevich, 2006), provides support for this

597

hypothesis.

598

Contrary to the expectations of this theory, loss of function mutations in lre

599

resulted in delayed seed development, revealing a positive role for maternal expression

600

of LRE during seed development. LRE is expressed in the zygote and the proliferating

601

endosperm and therefore we hypothesize that loss of LRE in the zygote and the

602

proliferating endosperm caused the delay in zygote and endosperm development.

603

Alternatively, the delay of the early embryogenesis could be due to an indirect effect of

604

loss of LRE in the endosperm. Nevertheless, our study adds LRE to a growing list of

605

MEGs whose loss negatively impacts seed development, including ZIX, FH5, and

606

NUWA in Arabidopsis (Ingouff et al., 2005; Gerald et al., 2009; Ngo et al., 2012; He et

607

al., 2017) and MEG1 in maize (Gutiérrez-Marcos et al., 2004; Costa et al., 2012).

608

Additionally, mutations in MEGs like FWA and AGL36 do not result in endosperm

609

phenotypes (Kinoshita et al., 2004; Shirzadi et al., 2011). These MEGs highlight the

610

need to explore alternative theories for imprinting. For example, maternal-offspring


611

coadaptation theory suggests that imprinting evolved to increase the adaptive

612

integration of offspring and maternal genomes, leading to higher offspring fitness (Wolf

613

and Hager, 2006).

614 615

CONCLUSIONS

616 617

Double fertilization in flowering plants occurs in the female gametophyte, which is

618

located within an ovule. During this critical step in flowering plant reproduction, the two

619

female gametes (the egg cell and the central cell) in the female gametophyte fuse with

620

the two male gametes (two sperm cells) delivered by the male gametophyte. The fusion

621

of egg with a sperm cell results in the embryo and the fusion of the central cell with the

622

second sperm cell gives rise to endosperm. Initiation of the embryo and the endosperm

623

development in the seed after double fertilization. Since double fertilization and seed

624

development occurs in the female gametophyte of an ovule, the female gametophyte

625

controls events immediately before and soon after fertilization.

626

Experiments reported in this study lead to three major conclusions. First, we

627

show that LRE expression is imprinted, as matrigenic LRE allele contributes nearly all

628

the LRE expression after fertilization. Second, it is likely that imprinting of LRE is

629

mediated by a novel pathway, as histone and DNA Methylation pathways known to

630

regulate MEGs do not control imprinted expression of LRE. Finally, we show that the

631

loss of the matrigenic but not the patrigenic LRE allele, caused delayed embryo and

632

endosperm development and revealed a growth promoting role for LRE in seeds. Our

633

study shows that LRE is a rare imprinted gene that functions immediately after double


634

fertilization. Coupled with our prior study on the role of LRE in pollen tube reception (Liu

635

et al., 2016), this study demonstrates that maternal and matrigenic expression of LRE in

636

the female gametophyte and seeds, respectively, allows the female gametophyte to

637

exert control over pollen tube reception before fertilization and seed development after

638

fertilization.

639

Many interesting questions remain to be addressed. It needs to be confirmed if

640

loss of matrigenic LRE expression in the seed results in the delayed initiation of seed

641

development. The pathway that controls imprinting in the gametophyte generation also

642

needs to be deciphered. The molecular mechanism by which LRE controls early seed

643

development is another important area of future research. LRE might be part of a

644

signaling complex in embryo and endosperm analogous to signaling that induces pollen

645

tube reception in the synergid (Li et al., 2015; Liu et al., 2016). Expression of the

646

matrigenic LRE allele in both fertilized products perhaps allows the female gametophyte

647

to extend control of seed development beyond fertilization and points to LRE’s utility as

648

a marker to characterize the maternal control of molecular events taking place during

649

this critical yet poorly characterized developmental phase in sexual plant reproduction –

650

the transition from the gametophytic to the sporophytic generation.

651 652 653

MATERIALS AND METHODS

654

Plant Material and Growth Conditions

655

Seeds were plated on Murashige and Skoog media (MS, Carolina Biological

656

Supply Company, 195703) and incubated in the growth chamber at 20℃ and 24 hours


657

illumination. For segregation analysis, seedlings were grown for two weeks in MS

658

supplemented with Hygromycin B (20 μg mL−1, PhytoTechnology Laboratories, H397)

659

and Basta (10 μg mL−1, Fisher Scientific, 50-240-693). For other experiments, 7-10

660

day-old seedlings were transplanted from plates to soil and grown in chambers at 20℃

661

and 24 hours illumination. The wild type and mutant accession used in this study is

662

Columbia (Col). In AS-PCR, wild-type Col and C24 accessions were used. demeter

663

(dme-1) mutant is in Landsberg erecta (Ler) background. All mutant plants were

664

confirmed by genotyping and/or phenotyping: lre-5 and lre-7 (Tsukamoto et al., 2010),

665

dme-1 (Choi et al., 2002), met1-1 (Kankel et al., 2003), ddm1-2 (Kakutani et al., 1996;

666

Yadegari et al., 2000), drm1-2 drm2-2 cmt3-11 (Chan et al., 2006), and msi1

667

(SAIL_429_B08) (Köhler et al., 2003).

668 669

RT-PCR

670

To analyze LRE gene expression (Fig. 1), RT-PCR was performed as described

671

(Tsukamoto et al., 2010). RT-PCR involving hypomethylation mutants were performed

672

as follows: Indicated mutant pollen was crossed onto lre-5 pistils. In case of met1-1 and

673

ddm1-2, only first generation homozygous plants from segregating progeny of a

674

heterozygote were used. The genotype of ddc triple mutant plants used in this

675

experiment was drm1-2/+ drm2-2/drm2-2 cmt3-11/cmt3-11. For msi1, pollen from only

676

heterozygous plants was used in the crosses, as homozygous mutants cannot be


677

established (Köhler et al., 2003). In each case, 25-30 pollinated pistils were excised

678

from the plant 12-14 HAP and ovary walls were removed before freezing. RNeasy Plant

679

Mini Kit (QIAGEN, 74904), DNase I (RNase-free, Life Technologies, AM2222) and

680

RNeasy MinElute Cleanup Kit (QIAGEN, 74204) were used for RNA isolation, digestion

681

of contaminating genomic DNA, and RNA purification, respectively. ThermoScript RT-

682

PCR System (Life Technologies, 11146-024) was used for reverse transcription. In all

683

crosses, 5 μg of total RNA was used as template for each RT reaction. TaKaRa Ex Taq

684

DNA Polymerase (Fisher Scientific, TAK_RR01BM) was used to carry out RT-PCR

685

reactions as follows: 94 oC (denaturation) for 15 seconds, 55 oC (annealing) for 15

686

seconds and 72oC (extension) for 1 minute for 41 cycles. RNA isolation and RT-PCR for

687

LRE expression in 8 day-old seedlings, including the roots, was performed similarly

688

(Fig. 4).

689 690

Allele Specific RT-PCR

691

Seeds at 24 HAP were collected from reciprocal crosses between Col and C24

692

ecotypes. RNA isolation, purification, and cDNA synthesis were performed as above

693

and AS-PCR was performed as follows: 3 minutes at 94oC, 50 cycles of 30 seconds at

694

94oC, 30 seconds at the chosen annealing temperature (below), 1 minute at 72oC,

695

followed by 10 minutes at 72oC. LNA primers (Latorra et al., 2003) used to amplify Col

696

or C24 allele of LRE and GRP23 are listed (Supplemental Table S10). Optimal

697

annealing temperature, at which the LNA primer pair can only amplify a PCR product

698

from one allele, but not the other, was determined using genomic DNA as a template in

699

a 50 cycle PCR reaction and conditions as described above: 67oC and 66oC for


700

Columbia and C24 alleles of LRE, respectively; 68oC for both Columbia and C24 alleles

701

of GRP23, respectively.

702 703

pLRE::GFP and pLRE::DsRed Expression Analysis

704

The pLRE::GFP construct was generated by amplifying the LRE promoter – 959

705

bp upstream of the LRE start codon that was sufficient to complement reproductive

706

defects in lre-7 (Liu et al., 2016 and this study using pLRE::LRE-HA, see below) – and

707

cloning it into the pCAMBIA1300-GFP vector between BamH I and Sal I sites. The

708

sequence-verified pLRE::GFP construct was introduced into wild-type plants (Clough

709

and Bent, 1998). Transformants were selected on MS plates with hygromycin. GFP

710

fluorescence in mature ovules (24-48 HAE) and developing seeds (0-48 HAP) was

711

observed and imaged in a confocal laser scanning microscope (LEICA TCL SP5) using

712

488 nm excitation and 510 emission. Plants homozygous for the transgene from three

713

independent transformants, with at least two plants per line and two pistils per plant,

714

were scored for GFP expression. The sequence-verified pLRE::DsRed construct was

715

introduced into plants carrying pWOX8::gWOX8-YFP (Ueda et al., 2011). The

716

pLRE::DsRed construct was generated by cloning 959 bp fragment upstream of the

717

LRE start codon and PCR amplified DsRed sequence (678 bp from a pLAT52::DsRed

718

construct) into the pH7WG vector cut with the Spe I (NEB, Catalog # R0133S) and AscI

719

(NEB, R5558S) using In-Fusion HD Cloning Plus System (Clontech, 638909).

720

Transformants were selected on MS plates with hygromycin. Expression of both

721

gWOX8-YFP and pLRE::DsRed were first confirmed in unfertilized ovules 24-48 hours

722

after emasculation in an epifluorescence microscope (ZEISS Axiophot) using FITC and


723

RHOD filters. Wild-type pollen was crossed onto pistils expressing both markers, seeds

724

were collected at 13.5 HAP, mounted on slides using 5% glycerol, and visualized under

725

the epifluorescence microscope. Three independent transformants, with at least two

726

plants per line and two pistils per plant, were scored for DsRed expression.

727 728

Complementation of Delayed Seed Development Phenotype Using pLRE::LRE-HA

729

The pLRE::LRE-HA construct was generated as follows: the 959 bp sequences

730

upstream of the LRE start codon, LRE coding sequence (888 bp), and 3’UTR sequence

731

(131 bp) were PCR amplified (primers listed in Supplementary Table 7). Primers used to

732

amplify LRE coding sequence contained the HA sequence (27bp). The fragments were

733

fused by overlapping PCR and cloned into pENTR™/D-TOPO® (Invitrogen, K2400-20)

734

and then swapped into gateway vector pH7WG using Clonase II enzyme mix (Life

735

Technologies, 11791020). Sequence verified pLRE::LRE-HA construct was transformed

736

into lre-7 plants. Transformants were selected on MS plates containing both hygromycin

737

(pLRE::LRE-HA transgene) and Basta (lre-7 mutation).

738

Candidate single insertion lines were selected from among the T1 transformants

739

as follows: if the T1 transformant generated in the lre-7/lre-7 background had a single

740

insertion of the transgene (pLRE::LRE-HA/+, lre-7/lre-7) and if it complemented the

741

partially-penetrant female gametophytic defects in lre, then seed set frequency (~65%)

742

in the selfed pistils will resemble that of lre-7/+ pistils (Supplemental Table S7). Seed

743

set was scored in 5-8 siliques/plant in 10 plants for every line. Based on this criterion,

744

we selected four T1 lines (12, 13, 15, and 20) as putative single insertion lines

745

(Supplemental Table S7). To further test the presence of single insertion in these lines,


746

we plated T2 seeds from these lines on hygromycin. If the transgenic plant is expected

747

to contain a single transgene and is heterozygous in that loci, the hygromycin resistance

748

to susceptibility ratio of the T2 selfed seeds from that plant is expected to be ~8-13:1

749

provided the transgene complements partially-penetrant lre female gametophytic

750

defects in inducing pollen tube reception (Liu et al., 2016). If there is no

751

complementation, the single insertion line will be expected to produce T2 seeds that

752

have a segregation ratio of 3:1. Based on this test, line #12 (335 resistant : 32 sensitive;

753

10.5 : 1) and line # 13 ( 484 resistant : 37 sensitive; 13.1 : 1) were identified as single

754

insertion lines.

755

T2 plants from line 12 were raised and scored for seed set. Those plants that

756

showed complete seed set were considered to be homozygous for the pLRE::LRE-HA

757

transgene and that the seed set defect in these plants have been completely rescued

758

(Supplemental Table S8). T3 seeds from two plants (12-7 and 12-17) were further

759

tested if they are homozygous for the pLRE::LRE-HA transgene; indeed, T3 seeds from

760

both of these plants were all resistant to hygromycin (Supplemental Table S9). Progeny

761

of 12-7 and 12-17 were then used in parent-of-origin complementation experiments

762

(Fig. 6).

763 764

Analysis of Embryo and Endosperm Development

765

GUS assay and microscopic analysis using pGRP23::GUS in 48 HAP seeds

766

were performed as described (Tsukamoto et al., 2010). For seed clearing assay, stage

767

12c flowers (Smyth et al., 1990) were emasculated and pollinated 24 hours later.

768

Crossed pistils were excised from the plant 48 HAP, silique walls were removed, and


769

fixed in ethanol/acetic acid (9:1) solution overnight, followed by successive incubations

770

in 90% ethanol and 70% ethanol each for 30 minutes. Clearing of seeds was performed

771

as described (Yadegari et al., 1994) after minor modifications: siliques were cleared

772

overnight in a clearing solution (chloral hydrate:glycerol:water = 4:1:2, m/v/v) (Chloral

773

hydrate, SIGMA-ALDRICH, C8383), mounted using the same clearing solution, and

774

scored under DIC microscope (ZEISS Axiophot) within a week after mounting the slides.

775

Embryo development in Arabidopsis is asynchronous and the different stages of

776

embryo development were scored at 48 HAP as zygote, 2-cell proembryo, proembryo

777

with 2-cell, 4-cell or 8-cell embryo proper (EP), and embryo with 16-cell EP as described

778

(Goldberg et al., 1994) based on the number of cells/nuclei in the developing embryo.

779

Wild-type and lre-7 single mutant was included in every experiment to account

780

for variation in the penetrance of the delayed embryo development phenotype which is

781

caused by the variation in the seed set (14.29 - 42.44%, n=3561 from 10 plants;

782

Supplemental Table S7). Accounting for this variability, in every experiment we

783

performed, we included as controls a fresh set of WT X WT and lre X WT crosses using

784

wild type and lre mutant plants grown simultaneously under the same conditions. As a

785

result, in each experiment, we only compared results from the experimental crosses

786

with that in concurrently performed control crosses and experimental crosses in one

787

experiment were never compared with the control crosses performed as part of another

788

experiment. This strategy allowed direct comparisons between wild type and mutant or

789

wild type and complemented lines or mutant and complemented lines.

790 791

Statistical Analysis


792

In WT and lre crosses, majority of embryos were at proembryo with 4-cell EP and

793

proembryo with 2-cell EP, respectively. Therefore, we grouped embryo developmental

794

stages into two categories. Embryos in the ‘earlier’ developmental stages, including

795

zygote, 2-cell proembryo, and proembryo with 2-cell EP (after 1 cell division in embryo

796

proper) are part of the first category (black column in graphs of Fig. 6, Fig. 7, and Fig.

797

S5). Embryos in the ‘later’ developmental stages, including proembryos with 4-, 8-cell

798

EP, and embryos with 16-cell EP (i.e., ≥ 2 cell divisions in EP) were grouped in the

799

second category (gray column in graphs of Fig. 6, Fig. 7, and Fig. S5).

800

Endosperm development stage was scored at 48 HAP only based on the number

801

of endosperm nuclei, as cellularization in endosperm does not initiate until the embryo

802

reaches the heart stage (Berger, F 2003). In WT and lre crosses, majority of endosperm

803

were at 33-128 endosperm nuclei (6-7 nuclear divisions of the primary endosperm

804

nucleus) and at 32 or less endosperm nuclei (≤ 5 nuclear divisions of the primary

805

endosperm nucleus), respectively. Therefore, endosperm developmental stages were

806

grouped into two categories. Endosperm with 32 or less endosperm nuclei was included

807

in the first category (‘earlier’, black column in graphs of Fig. S4). Later developmental

808

stages of endosperm, including those with 33-128 endosperm nuclei were grouped in

809

the second category (‘later’, gray column in graphs in Fig. S4).

810

‘Fisher’s Exact Test for Count Data’ was performed for the earlier and later

811

category data in 2x2 contingency table using R package (version 3.2.3, http://www.r-

812

project.org/) with fisher.test() function (R Core Team, 2015). We hypothesized that true

813

odds ratio is equal to 1. If p-value < 0.05, the hypothesis was rejected. In graphs (Fig. 6,

814

Fig. 7, Fig. S4, and Fig. S5), *** was used to represent statistically significant difference


815

in the Fisher's Exact Test (p-value < 0.001) and NS was indicated when there was no

816

statistical difference in the Fisher's Exact Test (p-value > 0.05).

817

Binomial test was performed using R package (version 3.1.1, http://www.r-

818

project.org/). Number of samples in each experiment was determined to be large

819

enough

820

(http://powerandsamplesize.com/Calculators/Compare-2-Proportions/2-Sample-

821

Equality). Seed number analyzed in each categories of every cross reported in this

822

study is provided in Supplemental Table S11.

for

a

statistical

analysis

using

‘Power

and

Sample

Size’

823 824 825 826

Image Processing Photoshop

CS4

(Adobe,

https://www.adobe.com/)

and

ImageJ

(http://imagej.nih.gov/ij/) were used for assembling image panels and preparing figures.

827 828

Accession Numbers

829

Accession numbers of genes studied in this work: LRE (At4g26466), GRP23

830

(At1g10270), ACTIN2 (At3g18780), ACTIN11 (At3g12110), LLG1 (At5g56170),

831

PHERES1

832

(AT1G69770), DRM2 (AT5G14620), DRM1 (AT5G15380), and MET1 (At5g49160).

(At1g65330),

MSI1

(AT5G58230),

DME1

(AT5G04560),

833 834

Competing interests

835

The authors declare no competing financial interests.

836 837


CMT3

838 Table 1. The paternal silencing of LRE is not affected by maternally inherited mutants in the DNA methylation pathway and in PRC2 function. Crosses (♀ × ♂ ) a

Endosperm expression (n) b

%

Total (n) c

WT × WT pLRE::GFP × WT

0 240

0.00 76.92

156 312

WT × pLRE::GFP

0

0.00

228

met1-1 × pLRE::GFP

0

0.00

412


0 181

0.00 69.88

125 259

WT × pLRE::GFP

0

0.00

350

ddm1-2 × pLRE::GFP

0

0.00

546


0 245

0.00 69.41

170 353

WT × pLRE::GFP

0

0.00

45

d

0

0.00

193

e

ddc × pLRE::GFP

0

0.00

194


0 262

0.00 72.18

143 363

WT × pLRE::GFP

0

0.00

266

msi1/+ × pLRE::GFP

0

0.00

493

ddc × pLRE::GFP

a

Line 6 of pLRE::GFP (Figure 2O; homozygous) was used in the crosses reported in this table. b Number of seeds with GFP expression in endosperm (≥ 2n) 13.5 HAP. c Number of total ovules analyzed. d Genotype of ddc triple mutant was drm1-2 drm2-2 cmt3-11. e Genotype of ddc triple mutant was drm1-2/+ drm2-2 cmt3-11.

839 840 841 842 843 844 845 846

SUPPLEMENTAL DATA


847

The following supplemental materials are available

848

Supplemental Figure S1. LRE is expressed in the female gametophyte.

849

Supplemental Figure S2. Defects in DNA and histone methylation pathway genes did

850

not lead to LRE expression from the patrigenic allele in seeds.

851

Supplemental Figure S3. A clearing procedure to monitor embryo and endosperm

852

development in seeds.

853

Supplemental Figure S4. Loss of matrigenic LRE allele leads to delayed endosperm

854

development.

855

Supplemental Figure S5. Loss of matrigenic LRE, but not the patrigenic LRE, causes

856

delayed embryo development in 48 HAP seeds.

857

Supplemental Table S1. GFP expression in the seeds from crosses of pLRE::GFP ♀ x

858

Col ♂

859

Supplemental Table S2. DsRed expression in the seeds from crosses of

860

pLRE::DsRed, pWOX8::WOX8-YFP ♀ x Col ♂

861

Supplemental Table S3. DsRed expression in the seeds from crosses of

862

pWOX8::WOX8-YFP ♀ x pLRE::DsRed, pWOX8::WOX8-YFP ♂

863

Supplemental Table S4. Defects in DNA methylation pathway genes does not lead to

864

expression from the patrigenic pLRE::GFP allele in fertilized ovules

865

Supplemental Table S5. demeter mutation does not result in decrease in the

866

expression of matrigenic pLRE::GFP allele in fertilized ovules

867

Supplemental Table S6. Defects in a histone methylation pathway gene (MSI1) does

868

not lead to expression from the patrigenic pLRE::GFP allele in fertilized ovules


869

Supplemental Table S7. Seed set defect in lre-7/lre-7 plants is complemented* if they

870

carry the pLRE::LRE-HA transgene, establishing that pLRE::LRE-HA is functional

871

Supplemental Table S8. Reduced seed set defect seen in lre-7/lre-7 plants is rescued

872

in T2 segregants establishing that pLRE::LRE-HA is functional

873

Supplemental Table S9. T3 segregation on plates containing both hygromycin and

874

basta confirm that tested lines are homozygous for the pLRE::LRE-HA transgene

875

Supplemental Table S10. List of primers used in this study

876

Supplementary Table S11. Seed number analyzed in each categories of every cross

877

reported in this study.

878

Supplemental Movie S1. LRE is expressed in the synergid and egg cells of the female

879

gametophyte.

880

Supplemental Movie S2. LRE is expressed in the synergid, egg, and central cells of

881

the female gametophyte.

882

Supplemental Movie S3. LRE is expressed in the zygote-like cell and the proliferating

883

endosperm of a developing seed.

884 885

Acknowledgements

886

We thank R. Yadegari (University of Arizona) for the epifluorescence microscope

887

(Zeiss Axiophot) and a modified clearing protocol to visualize seed development. We

888

thank Di Ran (University of Arizona) for help with statistical analysis and Nathaniel

889

Ponvert for technical assistance. We thank Gregory Copenhaver (University of North

890

Carolina) and Thomas Laux (University of Freiburg) for the pLAT52::DsRed and

891

pWOX8::gWOX8-YFP marker lines, respectively.


892 893

Figure legends

894 895

Figure 1. LRE is a maternally expressed imprinted gene. A, RT-PCR analysis of LRE

896

expression in 24 HAP ovules from indicated crosses between wild type (WT) and lre-5.

897

ACTIN2 (ACT2) expression was used as a positive control (An et al., 1996). B, LNA

898

primers-based Allele-specific (AS) RT-PCR analysis of LRE expression in 24 HAP

899

ovules from indicated crosses involving two accessions of Arabidopsis, Columbia (Col)

900

and C24. Amplification of target genes from Col and C24 genomic DNA (gDNA) in an

901

accession-specific manner (Latorra et al., 2003) confirmed allele specificity of AS-PCR

902

assay. cDNA, complementary DNA. Expression of GRP23 in seeds was used as a

903

control gene that is expressed from both matrigenic and patrigenic alleles. Marker sizes

904

(in kb) are shown on the left. ♀, female parent; ♂, male parent.

905 906

Figure 2. LRE is expressed in the female gametophyte and seeds. A-D, LRE is

907

expressed in the mature female gametophyte. A, Diagram of the pLRE::GFP construct.

908

B-D, LRE expression in unfertilized ovules. B, top panel is a fluorescent image, bottom

909

panel is a merged image of bright field (not shown) and fluorescent image. Zoomed in

910

version of the portion within red rectangle in B is shown in C. B-C, LRE is expressed in

911

the synergid cell (sc, white dashed line). To a lower extent, LRE is expressed in the egg

912

cell (ec, yellow dashed line). In some ovules, LRE is also expressed in the central cell

913

(cc, arrow). The embryo sac is outlined with a blue dashed line. D, Quantification of

914

microscopic observations of LRE expression, as shown in B-C. Scale bar, 15 μm. E-N,


915

pLRE::GFP expression up to 24 HAP in the unfertilized mature ovule (E) and in

916

developing seeds (F-I) carrying only maternally-transmitted pLRE::GFP. Double

917

fertilization is asynchronous in Arabidopsis and hence a crossed pistil contains seeds at

918

various stages of development. Only representative images of seeds that have

919

developmentally progressed the furthest in a crossed pistil at the indicated HAP are

920

shown; complete distribution of seeds at different stages of development in these

921

crosses is provided in Supplemental Table S1. Asterisks in G point to developing

922

endosperm in the embryo sac. Arrow in G refers to zygote-like cell in the embryo sac. J-

923

N, pLRE::GFP signal was not detected either in the unfertilized mature ovule (J) or in

924

seeds carrying only paternally-transmitted pLRE::GFP (K-N). In E-N, top panels are

925

fluorescent images while bottom panels are merged images of bright-field images (not

926

shown) and fluorescent images. Images in B,C,E-N were captured using identical

927

camera settings. Scale bars, 30 μm. O, Quantification of microscopic observations of

928

LRE expression in 13.5 HAP ovules from crosses involving three independent

929

pLRE::GFP transgenic lines. Only the developing seeds are reported here; unfertilized or just

930

fertilized ovules in the crossed pistil are not shown.

931 932

Figure 3. LRE is expressed in the zygote. A, Diagram of the pLRE::DsRed construct. B,

933

Unfertilized ovules with noticeable LRE expression in the synergid cells. The left panel

934

is YFP channel fluorescent image of the micropylar end of the unfertilized ovule

935

showing gWOX8-YFP expression in the two synergid cell nuclei (white arrowhead), the

936

egg cell nucleus (white arrow), and the central cell nucleus (white asterisk). The panel in

937

the middle is red channel fluorescent image of the micropylar end of the unfertilized

938

ovule showing pLRE::DsRed expression in the synergid cells (white arrowhead). 42 Downloaded from on August 15, 2017 - Published by www.plantphysiol.org Copyright © 2017 American Society of Plant Biologists. All rights reserved.

939

Location of the egg cell (white arrow) is also shown. Due to autofluorescence, LRE

940

expression in the central cell (asterisk) is not visible. The right panel is the merged

941

image of the two panels on the left. C, Seed with LRE expression in the zygote (at the

942

elongating stage). The left panel is YFP channel fluorescent image of the micropylar

943

end of the seed showing gWOX8-YFP expression in the zygote nucleus (white

944

arrowhead). The panel in the middle is red channel fluorescent image of the micropylar

945

end of the seed showing pLRE::DsRed expression in the zygote cell (white arrowhead).

946

The right panel is the merged image of the two panels on the left. D, Quantification of

947

co-localization of gWOX8-YFP and pLRE::DsRed expression in 13.5 HAP ovules from

948

crosses involving three independent pLRE::DsRed lines carrying the pWOX::gWOX8-

949

YFP transgene. Only those ovules with gWOX8-YFP expression in ≥ 4 endosperm

950

nuclei were included in the co-localization analysis and reported in this table. Scale bar,

951

20 μm.

952 953

Figure 4. Both matrigenic and patrigenic LRE are expressed in 8 day-old seedlings.

954

WT, lre-5, and lre-7 indicate 8 day-old seedlings from selfed seeds of indicated

955

genotypes. WT ♀ x lre-5 ♀ or WT ♀ x lre-7 ♂ indicates 8 day-old F1 seedlings raised

956

from a cross in which WT was the female parent and lre-5 (or lre-7) was the male

957

parent. Marker sizes shown in kb.

958 959

Figure 5. Defects in DNA and histone methylation pathway genes does not lead to LRE

960

expression from the patrigenic allele in seeds. A, RT-PCR analysis of LRE expression in

961

12 HAP seeds from crosses between wild type (WT) and lre-5. B, RT-PCR analysis of


962

LRE expression in 13.5 HAP seeds from indicated crosses. PHERES1 (PHE1), whose

963

patrigenic allele is preferentially expressed in seeds (Kohler et al., 2005; Makarevich et

964

al., 2008), was used as a positive control. ♀ and ♂, female and male parent,

965

respectively. 0.83% reverse transcription (RT) reaction was used as template in each of

966

the PCR reaction. gDNA, genomic DNA; Marker sizes (in kb) is shown on the left.

967

Genotype of ddc used in this experiment was drm1-2/+ drm2-2/drm2-2 cmt3-11/cmt3-

968

11.

969 970

Figure 6. Loss of matrigenic LRE causes delayed embryo development in 48 HAP

971

seeds. A, Graph showing that the embryo development is delayed in 48 HAP seeds

972

from a cross in which lre-7 was the female (♀) parent. This delay in embryo

973

development was rescued in crosses with female lre-7 mutant plants carrying the

974

pLRE::LRE-HA transgene (progeny of Line 12-7 and 12-17; Supplemental Table S5). B,

975

Graph showing that the embryo development was not delayed in 48 HAP seeds from a

976

cross in which wild type (WT) was the female (♀) parent. C, Graph showing that the

977

delayed embryo development in 48 HAP seeds using lre-7 mutant as the female parent

978

was not rescued if pollinated with pollen from lre-7 mutant plants carrying the

979

pLRE::LRE-HA transgene. In graphs shown in A-C, black column shows the first

980

category of embryo development stages (earlier), which includes zygote and

981

proembryos with 1 or 2 cells, while gray column represents the second category (later)

982

that includes proembryos with 4, 8, or 16 cells. Wild-type and lre-7 single mutant was

983

included in every experiment to account for the variation in the penetrance of the

984

delayed embryo development phenotype and allow direct comparisons between wild


985

type and mutant or wild type and complemented lines or mutant and complemented

986

lines. Fisher’s Exact Test p-values in A (*** and NS) are 1.022e-12 and 0.6998,

987

respectively. Fisher’s Exact Test p-value in B (NS) is 0.6143. p-values in C (*** and NS)

988

are 3.835e-10 and 0.6644, respectively.

989 990

Figure 7. Loss of LRE in the female sporophyte does not result in delayed embryo

991

development in 48 HAP seeds. A, A seed with normal embryo development from

992

crosses using WT as the female parent (♀) and pGRP23::GUS as the male parent (♂).

993

B-D, Seeds from crosses using lre-5/+ as the female parent (♀) and pGRP23::GUS as

994

the male parent (♂). B, A portion of the heterozygous silique containing a seed

995

undergoing normal (purple rectangle) and delayed embryo development (pink oval). C,

996

A zoomed-in view of the seed in purple rectangle in B. D, A zoomed in view of the seed

997

in pink oval in B. EM, embryo (black arrowhead); CE, chalazal endosperm (red

998

arrowhead); DEM, delayed embryo (orange arrowhead). Scale bars: 100 μm. E, Graph

999

showing that the embryo development is delayed in 48 HAP seeds from a cross in

1000

which lre-7/+ was the female (♀) parent and wild type (WT) was the male (♂) parent. In

1001

this graph, black column shows the first category of embryo development stages

1002

(earlier), which includes zygote and proembryos with 1 or 2 cells, while gray column

1003

represents the second category (later) that includes proembryos with 4, 8, or 16 cells.

1004

Fisher’s Exact Test p-value (***) in the graph is 0.0001038.

1005 1006

Legends for Supplemental Movies

1007


1008

Supplemental Movie S1. LRE is expressed in the synergid and egg cells of the female

1009

gametophyte. A z-stack of images of optical sections of an unfertilized ovule carrying

1010

pLRE::GFP transgene showing that LRE promoter is active in the synergid and egg

1011

cells of the female gametophyte. Imaging was performed on a Leica Confocal SP5

1012

microscope 24 hours after emasculation. GFP expression in the egg cell is visible in

1013

frames 1-5; expression in the first of the two synergid cells is visible to the right from

1014

frame1, while the expression in the second of the two synergid cells is visible to the left

1015

from frame 7. In this ovule, GFP expression in the central cell is not detected.

1016 1017

Supplemental Movie S2. LRE is expressed in the synergid, egg, and central cells of

1018

the female gametophyte. A z-stack of images of optical sections of an unfertilized ovule

1019

carrying pLRE::GFP transgene showing that the LRE promoter is active in the synerigd,

1020

egg, and central cells of the female gametophyte. Imaging was performed on a Leica

1021

Confocal SP5 microscope 24 hours after emasculation. GFP expression in the central

1022

cell is visible in frames 1-6 (faint expression); GFP expression in the egg cell is visible in

1023

frames 1-6; GFP expression in the first of the two synergid cells is visible to the right

1024

from frame 6, while the GFP expression in the second of the two synergid cells is visible

1025

to the left from frame 9.

1026 1027

Supplemental Movie S3. LRE is expressed in the zygote-like cell and the proliferating

1028

endosperm of a developing seed. A z-stack of images of optical sections of a seed

1029

formed when an ovule carrying pLRE::GFP transgene was fertilized by wild-type pollen

1030

showing that the maternally-transmitted LRE promoter is active in the zygote-like cell


1031

and proliferating endosperm. Imaging was performed on a Leica Confocal SP5

1032

microscope 24 hours after emasculation. Zygote-like cell expressing GFP is visible in

1033

frames 4 to 12. Proliferating endosperm expressing GFP is visible in frames 10 to 18.

1034 1035

REFERENCES

1036 1037

An YQ, McDowell JM, Huang SR, McKinney EC, Chambliss S, Meagher RB (1996) Strong,

1038

constitutive expression of the Arabidopsis ACT2/ACT8 actin subclass in vegetative

1039

tissues. Plant Journal 10: 107-121

1040

Autran D, Baroux C, Raissig MT, Lenormand T, Wittig M, Grob S, Steimer A, Barann M,

1041

Klostermeier UC, Leblanc O, Vielle-Calzada JP, Rosenstiel P, Grimanelli D,

1042

Grossniklaus U (2011) Maternal epigenetic pathways control parental contributions to

1043

Arabidopsis early embryogenesis. Cell 145: 707-719

1044 1045

Bai F, Settles AM (2015) Imprinting in plants as a mechanism to generate seed phenotypic diversity. Frontiers in Plant Science 5

1046

Berger F (2003) Endosperm: the crossroad of seed development. Curr Opin Plant Biol 6: 42-50

1047

Boavida LC, Hernandez-Coronado M, Becker JD (2015) Setting the stage for the next

1048

generation: Epigenetic reprogramming during sexual plant reproduction. In O Pontes, H

1049

Jin, eds, Nuclear Functions in Plant Transcription, Signaling and Development. Springer

1050

New York, pp 93-118

1051

Bratzel F, Yang C, Angelova A, Lopez-Torrejon G, Koch M, del Pozo JC, Calonje M (2012)

1052

Regulation of the new Arabidopsis imprinted gene AtBMI1C requires the interplay of

1053

different epigenetic mechanisms. Mol Plant 5: 260-269

1054

Capron A, Gourgues M, Neiva LS, Faure JE, Berger F, Pagnussat G, Krishnan A, Alvarez-Mejia

1055

C, Vielle-Calzada JP, Lee YR, Liu B, Sundaresan V (2008) Maternal control of male-


1056

gamete delivery in Arabidopsis involves a putative GPI-anchored protein encoded by the

1057

LORELEI gene. Plant Cell 20: 3038-3049

1058

Chan SWL, Henderson IR, Zhang XY, Shah G, Chien JSC, Jacobsen SE (2006) RNAi, DRD1,

1059

and histone methylation actively target developmentally important non-CG DNA

1060

methylation in Arabidopsis. PLoS genetics 2: 791-797

1061

Chaudhury AM, Koltunow A, Payne T, Luo M, Tucker MR, Dennis ES, Peacock WJ (2001)

1062

Control of early seed development. Annual Review of Cell and Developmental Biology

1063

17: 677-699

1064 1065

Chettoor AM, Phillips AR, Coker CT, Dilkes B, Evans MM (2016) Maternal gametophyte effects on seed development in maize. Genetics 204: 233-248

1066

Choi Y, Gehring M, Johnson L, Hannon M, Harada JJ, Goldberg RB, Jacobsen SE, Fischer RL

1067

(2002) DEMETER, a DNA glycosylase domain protein, is required for endosperm gene

1068

imprinting and seed viability in Arabidopsis. Cell 110: 33-42

1069 1070

Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant Journal 16: 735-743

1071

Costa Liliana M, Yuan J, Rouster J, Paul W, Dickinson H, Gutierrez-Marcos Jose F (2012)

1072

Maternal control of nutrient allocation in plant seeds by genomic imprinting. Current

1073

Biology 22: 160-165

1074 1075

Del Toro-De Leon G, Garcia-Aguilar M, Gillmor CS (2014) Non-equivalent contributions of maternal and paternal genomes to early plant embryogenesis. Nature 514: 624-627

1076

Ding Y-H, Liu N-Y, Tang Z-S, Liu J, Yang W-C (2006) Arabidopsis GLUTAMINE-RICH

1077

PROTEIN23 is essential for early embryogenesis and encodes a novel nuclear PPR

1078

motif protein that interacts with RNA polymerase II subunit III. Plant Cell 18: 815-830

1079 1080

Evans MM, Kermicle JL (2001) Interaction between maternal effect and zygotic effect mutations during maize seed development. Genetics 159: 303-315


1081 1082 1083 1084

Garcia-Aguilar M, Gillmor CS (2015) Zygotic genome activation and imprinting: parent-of-origin gene regulation in plant embryogenesis. Current Opinion in Plant Biology 27: 29-35 Gehring M (2013) Genomic imprinting: Insights from plants. Annual Review of Genetics 47: 187208

1085

Gehring M, Huh JH, Hsieh T-F, Penterman J, Choi Y, Harada JJ, Goldberg RB, Fischer RL

1086

(2006) DEMETER DNA glycosylase establishes MEDEA polycomb gene self-imprinting

1087

by allele-specific demethylation. Cell 124: 495-506

1088 1089

Gehring M, Missirian V, Henikoff S (2011) Genomic analysis of parent-of-origin allelic expression in Arabidopsis thaliana seeds. Plos One 6

1090

Gerald JNF, Hui PS, Berger F (2009) Polycomb group-dependent imprinting of the actin

1091

regulator AtFH5 regulates morphogenesis in Arabidopsis thaliana. Development 136:

1092

3399-3404

1093 1094

Goldberg, RB, de Paiva G, Yadegari R (1994) Plant embryogenesis: zygote to seed. Science 266: 605-14

1095

Grossniklaus U, Vielle-Calzada J-P, Hoeppner MA, Gagliano WB (1998) Maternal control of

1096

embryogenesis by MEDEA, a polycomb group gene in Arabidopsis. Science 280: 446-

1097

450

1098

Guitton A-E, Page DR, Chambrier P, Lionnet C, Faure J-E, Grossniklaus U, Berger F (2004)

1099

Identification of new members of Fertilisation Independent Seed Polycomb Group

1100

pathway involved in the control of seed development in Arabidopsis thaliana.

1101

Development 131: 2971-2981

1102

Gutiérrez-Marcos JF, Costa LM, Biderre-Petit C, Khbaya B, O'Sullivan DM, Wormald M, Perez

1103

P, Dickinson HG (2004) maternally expressed gene1 is a novel maize endosperm

1104

transfer cell–specific gene with a maternal parent-of-origin pattern of expression. The

1105

Plant Cell 16: 1288-1301


1106 1107 1108 1109

Haig D (2013) Kin conflict in seed development: An interdependent but fractious collective. Annual Review of Cell and Developmental Biology, Vol 29 29: 189-211 Haig D, Westoby M (1989) Parent-specific gene expression and the triploid endosperm. The American Naturalist 134: 147-155

1110

Haig D, Westoby M (1991) Genomic imprinting in endosperm: Its effect on seed development in

1111

crosses between species, and between different ploidies of the same species, and its

1112

implications for the evolution of apomixis. Philosophical Transactions of the Royal

1113

Society of London B: Biological Sciences 333: 1-13

1114

He S, Sun Y, Yang Q, Zhang X, Huang Q, Zhao P, Sun M, Liu J, Qian W, Qin G, Gu H, Qu LJ

1115

(2017) A novel imprinted gene NUWA controls mitochondrial function in early seed

1116

development in Arabidopsis. PLoS Genet 13: e1006553

1117

Hsieh T-F, Shin J, Uzawa R, Silva P, Cohen S, Bauer MJ, Hashimoto M, Kirkbride RC, Harada

1118

JJ, Zilberman D, Fischer RL (2011) Regulation of imprinted gene expression in

1119

Arabidopsis endosperm. Proceedings of the National Academy of Sciences 108: 1755-

1120

1762

1121

Huang S, An Y-Q, McDowell J, McKinney E, Meagher R (1997) The Arabidopsis ACT11 actin

1122

gene is strongly expressed in tissues of the emerging inflorescence, pollen, and

1123

developing ovules. Plant Molecular Biology 33: 125-139

1124

Ingouff M, Gerald JNF, Guerin C, Robert H, Sorensen MB, Damme DV, Geelen D, Blanchoin L,

1125

Berger F (2005) Plant formin AtFH5 is an evolutionarily conserved actin nucleator

1126

involved in cytokinesis. Nat Cell Biol 7: 374-380

1127 1128 1129 1130

Jahnke S, Scholten S (2009) Epigenetic resetting of a gene imprinted in plant embryos. Current Biology 19: 1677-1681 Johnson LM, Cao X, Jacobsen SE (2002) Interplay between two epigenetic marks: DNA methylation and histone H3 lysine 9 methylation. Current Biology 12: 1360-1367


1131

Jullien PE, Katz A, Oliva M, Ohad N, Berger F (2006) Polycomb group complexes self-regulate

1132

imprinting of the polycomb group gene MEDEA in Arabidopsis. Current Biology 16: 486-

1133

492

1134

Kakutani T, Jeddeloh JA, Flowers SK, Munakata K, Richards EJ (1996) Developmental

1135

abnormalities and epimutations associated with DNA hypomethylation mutations.

1136

Proceedings of the National Academy of Sciences 93: 12406-12411

1137

Kankel MW, Ramsey DE, Stokes TL, Flowers SK, Haag JR, Jeddeloh JA, Riddle NC, Verbsky

1138

ML, Richards EJ (2003) Arabidopsis MET1 cytosine methyltransferase mutants.

1139

Genetics 163: 1109-1122

1140 1141

Kawashima T, Berger F (2014) Epigenetic reprogramming in plant sexual reproduction. Nat Rev Genet 15: 613-624

1142

Kinoshita T, Miura A, Choi Y, Kinoshita Y, Cao X, Jacobsen SE, Fischer RL, Kakutani T (2004)

1143

One-way control of FWA imprinting in Arabidopsis endosperm by DNA methylation.

1144

Science 303: 521-523

1145

Kishimoto N, Sakai H, Jackson J, Jacobsen SE, Meyerowitz EM, Dennis ES, Finnegan EJ

1146

(2001) Site specificity of the Arabidopsis METI DNA methyltransferase demonstrated

1147

through hypermethylation of the superman locus. Plant Molecular Biology 46: 171-183

1148

Köhler C, Hennig L, Bouveret R, Gheyselinck J, Grossniklaus U, Gruissem W (2003)

1149

Arabidopsis MSI1 is a component of the MEA/FIE Polycomb group complex and

1150

required for seed development. The EMBO Journal 22: 4804-4814

1151 1152

Kohler C, Makarevich G (2006) Epigenetic mechanisms governing seed development in plants. EMBO Rep 7: 1223-1227

1153

Kohler C, Page DR, Gagliardini V, Grossniklaus U (2005) The Arabidopsis thaliana MEDEA

1154

Polycomb group protein controls expression of PHERES1 by parental imprinting. Nat

1155

Genet 37: 28-30


1156

Latorra D, Campbell K, Wolter A, Hurley JM (2003) Enhanced allele-specific PCR discrimination

1157

in SNP genotyping using 3' locked nucleic acid (LNA) primers. Human Mutation 22: 79-

1158

85

1159

Li C, Yeh F-L, Cheung AY, Duan Q, Kita D, Liu M-C, Maman J, Luu EJ, Wu BW, Gates L, Jalal

1160

M, Kwong A, Carpenter H, Wu H-M (2015) Glycosylphosphatidylinositol-anchored

1161

proteins as chaperones and co-receptors for FERONIA receptor kinase signaling in

1162

Arabidopsis. eLife 4: e06587

1163

Lindroth AM, Cao X, Jackson JP, Zilberman D, McCallum CM, Henikoff S, Jacobsen SE (2001)

1164

Requirement of CHROMOMETHYLASE3 for maintenance of CpXpG methylation.

1165

Science 292: 2077-2080

1166

Liu XL, Castro C, Wang YB, Noble J, Ponvert N, Bundy M, Hoel C, Shpak E, Palanivelu R

1167

(2016) The role of LORELEI in pollen tube reception at the interface of the synergid cell

1168

and pollen tube requires the modified eight-cysteine motif and the receptor-like kinase

1169

FERONIA. Plant Cell 28: 1035-1052

1170 1171

Loraine AE, McCormick S, Estrada A, Patel K, Qin P (2013) RNA-seq of Arabidopsis pollen uncovers novel transcription and alternative splicing. Plant Physiology 162: 1092-1109

1172

Luo M, Taylor JM, Spriggs A, Zhang HY, Wu XJ, Russell S, Singh M, Koltunow A (2011) A

1173

genome-wide survey of imprinted genes in rice seeds reveals imprinting primarily occurs

1174

in the endosperm. PLoS genetics 7

1175

Maruyama D, Volz R, Takeuchi H, Mori T, Igawa T, Kurihara D, Kawashima T, Ueda M, Ito M,

1176

Umeda M, Nishikawa S, Gross-Hardt R, Higashiyama T (2015) Rapid elimination of the

1177

persistent synergid through a cell fusion mechanism. Cell 161: 907-918

1178

Ngo QA, Baroux C, Guthörl D, Mozerov P, Collinge MA, Sundaresan V, Grossniklaus U (2012)

1179

The armadillorepeat gene ZAK IXIK promotes Arabidopsis early embryo and endosperm

1180

development through a distinctive gametophytic maternal effect. The Plant Cell 24:

1181

4026-4043 52 Downloaded from on August 15, 2017 - Published by www.plantphysiol.org Copyright © 2017 American Society of Plant Biologists. All rights reserved.

1182

Ngo QA, Moore JM, Baskar R, Grossniklaus U, Sundaresan V (2007) Arabidopsis GLAUCE

1183

promotes fertilization-independent endosperm development and expression of paternally

1184

inherited alleles. Development 134: 4107-4117

1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196

Nodine MD, Bartel DP (2012) Maternal and paternal genomes contribute equally to the transcriptome of early plant embryos. Nature 482: 94-97 Palanivelu R, Preuss D (2006) Distinct short-range ovule signals attract or repel Arabidopsis thaliana pollen tubes in vitro. BMC Plant Biol 6: 7 Pignatta D, Erdmann RM, Scheer E, Picard CL, Bell GW, Gehring M (2014) Natural epigenetic polymorphisms lead to intraspecific variation in Arabidopsis gene imprinting. eLife 3 R Core Team (2015) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Raissig MT, Bemer M, Baroux C, Grossniklaus U (2013) Genomic imprinting in the Arabidopsis embryo is partly regulated by PRC2. PLoS Genet 9: e1003862 Schon MA, Nodine M (2017) Widespread contamination of Arabidopsis embryo and endosperm transcriptome datasets. Plant Cell

1197

Shirzadi R, Andersen ED, Bjerkan KN, Gloeckle BM, Heese M, Ungru A, Winge P, Koncz C,

1198

Aalen RB, Schnittger A, Grini PE (2011) Genome-wide transcript profiling of endosperm

1199

without

1200

AGAMOUS-LIKE36. PLoS genetics 7

1201 1202 1203 1204

paternal

contribution

identifies

parent-of-origin-dependent

regulation

of

Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis Plant Cell 2: 755-767 Spencer HG, Clark AG (2014) Non-conflict theories for the evolution of genomic imprinting. Heredity 113: 112-118

1205

Stroud H, Greenberg Maxim VC, Feng S, Bernatavichute Yana V, Jacobsen Steven E (2013)

1206

Comprehensive analysis of silencing mutants reveals complex regulation of the

1207

Arabidopsis methylome. Cell 152: 352-364 53 Downloaded from on August 15, 2017 - Published by www.plantphysiol.org Copyright © 2017 American Society of Plant Biologists. All rights reserved.

1208

Tsukamoto T, Palanivelu R (2010) Loss of LORELEI function in the pistil delays initiation but

1209

does not affect embryo development in Arabidopsis thaliana. Plant Signal. Behav. 5:

1210

1487-1490

1211

Tsukamoto T, Qin Y, Huang Y, Dunatunga D, Palanivelu R (2010) A role for LORELEI, a

1212

putative glycosylphosphatidylinositol-anchored protein, in Arabidopsis thaliana double

1213

fertilization and early seed development. The Plant Journal 62: 571-588

1214 1215

Ueda M, Zhang Z, Laux T (2011) Transcriptional activation of Arabidopsis axis patterning genes WOX8/9 links zygote polarity to embryo development. Developmental Cell 20: 264-270

1216

Volz R, Heydlauff J, Ripper D, von Lyncker L, Gross-Hardt R (2013) Ethylene signaling is

1217

required for synergid degeneration and the establishment of a pollen tube block.

1218

Developmental Cell 25: 310-316

1219

Vu TM, Nakamura M, Calarco JP, Susaki D, Lim PQ, Kinoshita T, Higashiyama T, Martienssen

1220

RA, Berger F (2013) RNA-directed DNA methylation regulates parental genomic

1221

imprinting at several loci in Arabidopsis. Development 140: 2953-2960

1222 1223

Wolf JB, Hager R (2006) A maternal-offspring coadaptation theory for the evolution of genomic imprinting. Plos Biology 4: 2238-2243

1224

Xin M, Yang R, Li G, Chen H, Laurie J, Ma C, Wang D, Yao Y, Larkins BA, Sun Q, Yadegari R,

1225

Wang X, Ni Z (2013) Dynamic expression of imprinted genes associates with maternally

1226

controlled nutrient allocation during maize endosperm development. The Plant Cell 25:

1227

3212-3227

1228 1229

Yadegari R, Drews GN (2004) Female gametophyte development. Plant Cell 16 Suppl: S133141

1230

Yadegari R, Kinoshita T, Lotan O, Cohen G, Katz A, Choi Y, Katz A, Nakashima K, Harada JJ,

1231

Goldberg RB, Fischer RL, Ohad N (2000) Mutations in the FIE and MEA genes that

1232

encode interacting polycomb proteins cause

1233

development by distinct mechanisms. The Plant Cell 12: 2367-2381

parent-of-origin effects on seed


1234

Yadegari R, Paiva G, Laux T, Koltunow AM, Apuya N, Zimmerman JL, Fischer RL, Harada JJ,

1235

Goldberg RB (1994) Cell differentiation and morphogenesis are uncoupled in

1236

Arabidopsis raspberry embryos. The Plant Cell 6: 1713-1729

1237 1238 1239


Downloaded from on August 15, 2017 - Published by www.plantphysiol.org Copyright © 2017 American Society of Plant Biologists. All rights reserved.







Parsed Citations An YQ, McDowell JM, Huang SR, McKinney EC, Chambliss S, Meagher RB (1996) Strong, constitutive expression of the Arabidopsis ACT2/ACT8 actin subclass in vegetative tissues. Plant Journal 10: 107-121 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Autran D, Baroux C, Raissig MT, Lenormand T, Wittig M, Grob S, Steimer A, Barann M, Klostermeier UC, Leblanc O, Vielle-Calzada JP, Rosenstiel P, Grimanelli D, Grossniklaus U (2011) Maternal epigenetic pathways control parental contributions to Arabidopsis early embryogenesis. Cell 145: 707-719 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Bai F, Settles AM (2015) Imprinting in plants as a mechanism to generate seed phenotypic diversity. Frontiers in Plant Science 5 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Berger F (2003) Endosperm: the crossroad of seed development. Curr Opin Plant Biol 6: 42-50 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Boavida LC, Hernandez-Coronado M, Becker JD (2015) Setting the stage for the next generation: Epigenetic reprogramming during sexual plant reproduction. In O Pontes, H Jin, eds, Nuclear Functions in Plant Transcription, Signaling and Development. Springer New York, pp 93-118 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Bratzel F, Yang C, Angelova A, Lopez-Torrejon G, Koch M, del Pozo JC, Calonje M (2012) Regulation of the new Arabidopsis imprinted gene AtBMI1C requires the interplay of different epigenetic mechanisms. Mol Plant 5: 260-269 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Capron A, Gourgues M, Neiva LS, Faure JE, Berger F, Pagnussat G, Krishnan A, Alvarez-Mejia C, Vielle-Calzada JP, Lee YR, Liu B, Sundaresan V (2008) Maternal control of male-gamete delivery in Arabidopsis involves a putative GPI-anchored protein encoded by the LORELEI gene. Plant Cell 20: 3038-3049 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Chan SWL, Henderson IR, Zhang XY, Shah G, Chien JSC, Jacobsen SE (2006) RNAi, DRD1, and histone methylation actively target developmentally important non-CG DNA methylation in Arabidopsis. PLoS genetics 2: 791-797 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Chaudhury AM, Koltunow A, Payne T, Luo M, Tucker MR, Dennis ES, Peacock WJ (2001) Control of early seed development. Annual Review of Cell and Developmental Biology 17: 677-699 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Chettoor AM, Phillips AR, Coker CT, Dilkes B, Evans MM (2016) Maternal gametophyte effects on seed development in maize. Genetics 204: 233-248 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Choi Y, Gehring M, Johnson L, Hannon M, Harada JJ, Goldberg RB, Jacobsen SE, Fischer RL (2002) DEMETER, a DNA glycosylase domain protein, is required for endosperm gene imprinting and seed viability in Arabidopsis. Cell 110: 33-42 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant Journal 16: 735-743 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title


Costa Liliana M, Yuan J, Rouster J, Paul W, Dickinson H, Gutierrez-Marcos Jose F (2012) Maternal control of nutrient allocation in plant seeds by genomic imprinting. Current Biology 22: 160-165 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Del Toro-De Leon G, Garcia-Aguilar M, Gillmor CS (2014) Non-equivalent contributions of maternal and paternal genomes to early plant embryogenesis. Nature 514: 624-627 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ding Y-H, Liu N-Y, Tang Z-S, Liu J, Yang W-C (2006) Arabidopsis GLUTAMINE-RICH PROTEIN23 is essential for early embryogenesis and encodes a novel nuclear PPR motif protein that interacts with RNA polymerase II subunit III. Plant Cell 18: 815-830 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Evans MM, Kermicle JL (2001) Interaction between maternal effect and zygotic effect mutations during maize seed development. Genetics 159: 303-315 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Garcia-Aguilar M, Gillmor CS (2015) Zygotic genome activation and imprinting: parent-of-origin gene regulation in plant embryogenesis. Current Opinion in Plant Biology 27: 29-35 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gehring M (2013) Genomic imprinting: Insights from plants. Annual Review of Genetics 47: 187-208 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gehring M, Huh JH, Hsieh T-F, Penterman J, Choi Y, Harada JJ, Goldberg RB, Fischer RL (2006) DEMETER DNA glycosylase establishes MEDEA polycomb gene self-imprinting by allele-specific demethylation. Cell 124: 495-506 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gehring M, Missirian V, Henikoff S (2011) Genomic analysis of parent-of-origin allelic expression in Arabidopsis thaliana seeds. Plos One 6 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gerald JNF, Hui PS, Berger F (2009) Polycomb group-dependent imprinting of the actin regulator AtFH5 regulates morphogenesis in Arabidopsis thaliana. Development 136: 3399-3404 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Goldberg, RB, de Paiva G, Yadegari R (1994) Plant embryogenesis: zygote to seed. Science 266: 605-14 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Grossniklaus U, Vielle-Calzada J-P, Hoeppner MA, Gagliano WB (1998) Maternal control of embryogenesis by MEDEA, a polycomb group gene in Arabidopsis. Science 280: 446-450 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Guitton A-E, Page DR, Chambrier P, Lionnet C, Faure J-E, Grossniklaus U, Berger F (2004) Identification of new members of Fertilisation Independent Seed Polycomb Group pathway involved in the control of seed development in Arabidopsis thaliana. Development 131: 2971-2981 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gutiérrez-Marcos JF, Costa LM, Biderre-Petit C, Khbaya B, O'Sullivan DM, Wormald M, Perez P, Dickinson HG (2004) maternally expressed gene1 is a novel maize endosperm transfer cell-specific gene with a maternal parent-of-origin pattern of expression. The Plant Cell 16: 1288-1301 Pubmed: Author and Title


CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Haig D (2013) Kin conflict in seed development: An interdependent but fractious collective. Annual Review of Cell and Developmental Biology, Vol 29 29: 189-211 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Haig D, Westoby M (1989) Parent-specific gene expression and the triploid endosperm. The American Naturalist 134: 147-155 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Haig D, Westoby M (1991) Genomic imprinting in endosperm: Its effect on seed development in crosses between species, and between different ploidies of the same species, and its implications for the evolution of apomixis. Philosophical Transactions of the Royal Society of London B: Biological Sciences 333: 1-13 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

He S, Sun Y, Yang Q, Zhang X, Huang Q, Zhao P, Sun M, Liu J, Qian W, Qin G, Gu H, Qu LJ (2017) A novel imprinted gene NUWA controls mitochondrial function in early seed development in Arabidopsis. PLoS Genet 13: e1006553 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hsieh T-F, Shin J, Uzawa R, Silva P, Cohen S, Bauer MJ, Hashimoto M, Kirkbride RC, Harada JJ, Zilberman D, Fischer RL (2011) Regulation of imprinted gene expression in Arabidopsis endosperm. Proceedings of the National Academy of Sciences 108: 1755-1762 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Huang S, An Y-Q, McDowell J, McKinney E, Meagher R (1997) The Arabidopsis ACT11 actin gene is strongly expressed in tissues of the emerging inflorescence, pollen, and developing ovules. Plant Molecular Biology 33: 125-139 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ingouff M, Gerald JNF, Guerin C, Robert H, Sorensen MB, Damme DV, Geelen D, Blanchoin L, Berger F (2005) Plant formin AtFH5 is an evolutionarily conserved actin nucleator involved in cytokinesis. Nat Cell Biol 7: 374-380 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Jahnke S, Scholten S (2009) Epigenetic resetting of a gene imprinted in plant embryos. Current Biology 19: 1677-1681 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Johnson LM, Cao X, Jacobsen SE (2002) Interplay between two epigenetic marks: DNA methylation and histone H3 lysine 9 methylation. Current Biology 12: 1360-1367 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Jullien PE, Katz A, Oliva M, Ohad N, Berger F (2006) Polycomb group complexes self-regulate imprinting of the polycomb group gene MEDEA in Arabidopsis. Current Biology 16: 486-492 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kakutani T, Jeddeloh JA, Flowers SK, Munakata K, Richards EJ (1996) Developmental abnormalities and epimutations associated with DNA hypomethylation mutations. Proceedings of the National Academy of Sciences 93: 12406-12411 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kankel MW, Ramsey DE, Stokes TL, Flowers SK, Haag JR, Jeddeloh JA, Riddle NC, Verbsky ML, Richards EJ (2003) Arabidopsis MET1 cytosine methyltransferase mutants. Genetics 163: 1109-1122 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kawashima T, Berger F (2014) Epigenetic reprogramming in plant sexual reproduction. Nat Rev Genet 15: 613-624 Pubmed: Author and Title


CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kinoshita T, Miura A, Choi Y, Kinoshita Y, Cao X, Jacobsen SE, Fischer RL, Kakutani T (2004) One-way control of FWA imprinting in Arabidopsis endosperm by DNA methylation. Science 303: 521-523 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kishimoto N, Sakai H, Jackson J, Jacobsen SE, Meyerowitz EM, Dennis ES, Finnegan EJ (2001) Site specificity of the Arabidopsis METI DNA methyltransferase demonstrated through hypermethylation of the superman locus. Plant Molecular Biology 46: 171-183 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Köhler C, Hennig L, Bouveret R, Gheyselinck J, Grossniklaus U, Gruissem W (2003) Arabidopsis MSI1 is a component of the MEA/FIE Polycomb group complex and required for seed development. The EMBO Journal 22: 4804-4814 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kohler C, Makarevich G (2006) Epigenetic mechanisms governing seed development in plants. EMBO Rep 7: 1223-1227 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kohler C, Page DR, Gagliardini V, Grossniklaus U (2005) The Arabidopsis thaliana MEDEA Polycomb group protein controls expression of PHERES1 by parental imprinting. Nat Genet 37: 28-30 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Latorra D, Campbell K, Wolter A, Hurley JM (2003) Enhanced allele-specific PCR discrimination in SNP genotyping using 3' locked nucleic acid (LNA) primers. Human Mutation 22: 79-85 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Li C, Yeh F-L, Cheung AY, Duan Q, Kita D, Liu M-C, Maman J, Luu EJ, Wu BW, Gates L, Jalal M, Kwong A, Carpenter H, Wu H-M (2015) Glycosylphosphatidylinositol-anchored proteins as chaperones and co-receptors for FERONIA receptor kinase signaling in Arabidopsis. eLife 4: e06587 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Lindroth AM, Cao X, Jackson JP, Zilberman D, McCallum CM, Henikoff S, Jacobsen SE (2001) Requirement of CHROMOMETHYLASE3 for maintenance of CpXpG methylation. Science 292: 2077-2080 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Liu XL, Castro C, Wang YB, Noble J, Ponvert N, Bundy M, Hoel C, Shpak E, Palanivelu R (2016) The role of LORELEI in pollen tube reception at the interface of the synergid cell and pollen tube requires the modified eight-cysteine motif and the receptor-like kinase FERONIA. Plant Cell 28: 1035-1052 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Loraine AE, McCormick S, Estrada A, Patel K, Qin P (2013) RNA-seq of Arabidopsis pollen uncovers novel transcription and alternative splicing. Plant Physiology 162: 1092-1109 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Luo M, Taylor JM, Spriggs A, Zhang HY, Wu XJ, Russell S, Singh M, Koltunow A (2011) A genome-wide survey of imprinted genes in rice seeds reveals imprinting primarily occurs in the endosperm. PLoS genetics 7 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Maruyama D, Volz R, Takeuchi H, Mori T, Igawa T, Kurihara D, Kawashima T, Ueda M, Ito M, Umeda M, Nishikawa S, Gross-Hardt R, Higashiyama T (2015) Rapid elimination of the persistent synergid through a cell fusion mechanism. Cell 161: 907-918 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title


Ngo QA, Baroux C, Guthörl D, Mozerov P, Collinge MA, Sundaresan V, Grossniklaus U (2012) The armadillorepeat gene ZAK IXIK promotes Arabidopsis early embryo and endosperm development through a distinctive gametophytic maternal effect. The Plant Cell 24: 4026-4043 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ngo QA, Moore JM, Baskar R, Grossniklaus U, Sundaresan V (2007) Arabidopsis GLAUCE promotes fertilization-independent endosperm development and expression of paternally inherited alleles. Development 134: 4107-4117 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Nodine MD, Bartel DP (2012) Maternal and paternal genomes contribute equally to the transcriptome of early plant embryos. Nature 482: 94-97 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Palanivelu R, Preuss D (2006) Distinct short-range ovule signals attract or repel Arabidopsis thaliana pollen tubes in vitro. BMC Plant Biol 6: 7 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Pignatta D, Erdmann RM, Scheer E, Picard CL, Bell GW, Gehring M (2014) Natural epigenetic polymorphisms lead to intraspecific variation in Arabidopsis gene imprinting. eLife 3 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

R Core Team (2015) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Raissig MT, Bemer M, Baroux C, Grossniklaus U (2013) Genomic imprinting in the Arabidopsis embryo is partly regulated by PRC2. PLoS Genet 9: e1003862 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Schon MA, Nodine M (2017) Widespread contamination of Arabidopsis embryo and endosperm transcriptome datasets. Plant Cell Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Shirzadi R, Andersen ED, Bjerkan KN, Gloeckle BM, Heese M, Ungru A, Winge P, Koncz C, Aalen RB, Schnittger A, Grini PE (2011) Genome-wide transcript profiling of endosperm without paternal contribution identifies parent-of-origin-dependent regulation of AGAMOUS-LIKE36. PLoS genetics 7 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis Plant Cell 2: 755-767 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Spencer HG, Clark AG (2014) Non-conflict theories for the evolution of genomic imprinting. Heredity 113: 112-118 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Stroud H, Greenberg Maxim VC, Feng S, Bernatavichute Yana V, Jacobsen Steven E (2013) Comprehensive analysis of silencing mutants reveals complex regulation of the Arabidopsis methylome. Cell 152: 352-364 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Tsukamoto T, Palanivelu R (2010) Loss of LORELEI function in the pistil delays initiation but does not affect embryo development in Arabidopsis thaliana. Plant Signal. Behav. 5: 1487-1490 Pubmed: Author and Title CrossRef: Author and Title


Google Scholar: Author Only Title Only Author and Title

Tsukamoto T, Qin Y, Huang Y, Dunatunga D, Palanivelu R (2010) A role for LORELEI, a putative glycosylphosphatidylinositol-anchored protein, in Arabidopsis thaliana double fertilization and early seed development. The Plant Journal 62: 571-588 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ueda M, Zhang Z, Laux T (2011) Transcriptional activation of Arabidopsis axis patterning genes WOX8/9 links zygote polarity to embryo development. Developmental Cell 20: 264-270 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Volz R, Heydlauff J, Ripper D, von Lyncker L, Gross-Hardt R (2013) Ethylene signaling is required for synergid degeneration and the establishment of a pollen tube block. Developmental Cell 25: 310-316 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Vu TM, Nakamura M, Calarco JP, Susaki D, Lim PQ, Kinoshita T, Higashiyama T, Martienssen RA, Berger F (2013) RNA-directed DNA methylation regulates parental genomic imprinting at several loci in Arabidopsis. Development 140: 2953-2960 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Wolf JB, Hager R (2006) A maternal-offspring coadaptation theory for the evolution of genomic imprinting. Plos Biology 4: 2238-2243 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Xin M, Yang R, Li G, Chen H, Laurie J, Ma C, Wang D, Yao Y, Larkins BA, Sun Q, Yadegari R, Wang X, Ni Z (2013) Dynamic expression of imprinted genes associates with maternally controlled nutrient allocation during maize endosperm development. The Plant Cell 25: 3212-3227 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Yadegari R, Drews GN (2004) Female gametophyte development. Plant Cell 16 Suppl: S133-141 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Yadegari R, Kinoshita T, Lotan O, Cohen G, Katz A, Choi Y, Katz A, Nakashima K, Harada JJ, Goldberg RB, Fischer RL, Ohad N (2000) Mutations in the FIE and MEA genes that encode interacting polycomb proteins cause parent-of-origin effects on seed development by distinct mechanisms. The Plant Cell 12: 2367-2381 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Yadegari R, Paiva G, Laux T, Koltunow AM, Apuya N, Zimmerman JL, Fischer RL, Harada JJ, Goldberg RB (1994) Cell differentiation and morphogenesis are uncoupled in Arabidopsis raspberry embryos. The Plant Cell 6: 1713-1729 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title
