Transcriptomes of eight Arabidopsis thaliana ...

Plant Physiology Preview. Published on May 15, 2016, as DOI:10.1104/pp.16.00472

1

Short title: Natural variation in submergence tolerance

2

Corresponding author: Rashmi Sasidharan ([email protected])

3 4

Transcriptomes of eight Arabidopsis thaliana accessions reveal core

5

conserved, genotype- and organ-specific responses to flooding stress

6 7

Hans van Veen*a,b, Divya Vashisht*a,1, Melis Akman*c,2, Thomas Girked,

8

Angelika Mustrophe, Emilie Reinena, Sjon Hartmana, Maarten Kooikera, Peter

9

van Tienderenc, M. Eric Schranzf, Julia Bailey-Serresa,d, Laurentius ACJ

10

Voeseneka and Rashmi Sasidharana

11 12

*these authors contributed equally

13

a

14

3584CH, Utrecht, The Netherlands

15

b

Institute of Life Sciences, Scuola Superiore Sant’Anna, Pisa, Italy

16

c

Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam,

17

The Netherlands

18

d

19

California, Riverside, USA

20

e

Department of Plant Physiology, Bayreuth University, Bayreuth, Germany

21

f

Biosystematics Group, Wageningen University, The Netherlands

Plant Ecophysiology, Institute of Environmental Biology, Utrecht University,

Center for Plant Cell Biology, Botany and Plant Sciences University of

22 23 24

Summary: A study of eight Arabidopsis accessions reveals novel insights into

25

early transcriptional and post-transcriptional responses to starvation and

26

flooding stress.

27 28

Footnotes:

29

Author contributions: HvV, DV, MA, JB-S, MES, PvT, LACJV and RS

30

conceived the research plans. JB-S, MES, PvT, LACJV and RS supervised

31

the experiments. HvV, DV, MA did most of the experiments. ER and SH

32

provided technical assistance. HvV, DV, MA, TG and RS analyzed the data.

33

RS and LACJV conceived the project and HvV and RS wrote the article with

Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Copyright 2016 by the American Society of Plant Biologists

1

34

contributions of all the authors. MA, JB-S, MES, PvT, LACJV supervised and

35

complemented the writing.

36 37

Funding information: This research was supported by an NWO-VENI

38

(86312013) and NWO-ALW (82201007) grant to RS and a Centre for

39

Biosystems Genomics (CBSG) 2012 grant to LACJV, RS and DV.

40 41

Present address:

42

1

43

Sciences, 1030 Vienna, Austria

44

2

Gregor Mendel Institute of Molecular Plant Biology, Austrian Academy of Plant and Environmental Sciences, University of California, Davis, USA

45 46

Corresponding author email: [email protected]

47


2

48 49

Abstract

50

Climate change has increased the frequency and severity of flooding events

51

with significant negative impact on agricultural productivity. These events

52

often submerge plant aerial organs and roots, limiting growth and survival due

53

to a severe reduction in light reactions and gas exchange necessary for

54

photosynthesis and respiration, respectively. To distinguish molecular

55

responses to the compound stress imposed by submergence, we investigated

56

transcriptomic adjustments to darkness in air and under submerged

57

conditions using eight Arabidopsis thaliana accessions differing significantly in

58

sensitivity to submergence. Evaluation of root and rosette transcriptomes

59

revealed an early transcriptional and post-transcriptional response signature

60

that

61

susceptibility-associated

62

uncovered. Post-transcriptional regulation encompassed darkness- and

63

submergence-induced alternative splicing of transcripts from pathways

64

involved in alternative mobilization of energy reserves. The organ-specific

65

transcriptome adjustments reflected the distinct physiological status of roots

66

and

67

upregulation of chloroplast-encoded photosynthesis and redox-related genes,

68

whereas those of the rosette were related to regulation of development and

69

growth processes. We identified a novel set of ‘tolerance-genes’, recognized

70

mainly by quantitative differences. These included a transcriptome signature

71

of more pronounced gluconeogenesis in tolerant accessions, a response that

72

included stress-induced alternative splicing. This study provides organ-

73

specific molecular resolution of genetic variation in submergence responses

74

involving interactions between darkness and low oxygen constraints of

75

flooding stress and demonstrates that early transcriptome plasticity including

76

alternative splicing is associated with the ability to cope with a compound

77

environmental stress.

was

primarily

shoots.

conserved and

Root-specific

across

genotypes,

genotype-specific

transcriptome

although

responses

changes

flooding

were

included

also

marked

78 79


3

80 81

Introduction The environment that surrounds a plant changes constantly, often imposing

82

constraints

83

development. Flooding can have a dramatic impact on plant performance; and

84

while it occurs regularly in some natural ecosystems, it is usually disastrous in

85

controlled agricultural environments. Flooding restricts gas diffusion between

86

submerged organs and the surrounding aquatic environment. The limited

87

exchange of oxygen (O2) and carbon dioxide (CO2) slows down aerobic

88

respiration and photosynthesis (Mommer and Visser, 2005; Zabalza et al.,

89

2008). Turbid and muddy floodwaters restrict light penetration, further

90

compromising photoautotrophic generation of critical carbohydrates (Vervuren

91

et al., 2003). Finally, O2 deficient flooded soils often have a severely reduced

92

redox potential and accumulate toxic compounds, which limit root growth

93

(Armstrong and Armstrong, 2001).

on

metabolism

that

modify

vegetative

and

reproductive

94

Flooding is therefore a compound stress, imposing multiple constraints on

95

submerged plants. Despite this, marshes and river floodplains support a rich

96

diversity of plant life that display a gradient of flood tolerance traits and

97

responses (Van Eck et al., 2004; Voesenek et al., 2004). Studies on rice and

98

several wild species have identified two antithetical survival strategies,

99

dependent on the selection pressure of their natural flooding regime. An

100

escape response involving rapid shoot elongation allows plants to regain air

101

contact by forming a snorkel during shallow and prolonged floods (Voesenek

102

and Bailey-Serres, 2015). Deep or very short floods require a quiescent

103

strategy where a restriction of growth combined with conservation of energy

104

expenditure and reserve utilization promotes survival until the floods recede

105

(van Veen et al., 2014b). Fundamental knowledge of the genetic,

106

physiological and molecular regulation of these traits is not only of general

107

interest, but essential to improve the tolerance of many economically relevant

108

crops, most of which are very sensitive to floods (Voesenek et al., 2014). The

109

genetic and molecular regulation of flood adaptive strategies has been most

110

extensively studied in semi-aquatic flood tolerant species of the genera Oryza,

111

Rorippa and Rumex (Fukao et al., 2006; Hattori et al., 2009; Lee et al., 2009;

112

van Veen et al., 2013; Sasidharan et al., 2013; van Veen et al., 2014a; Narsai

113

et al., 2015).


4

114

The understanding of the flooding-induced low O2 and low energy signaling

115

networks, has also greatly benefited from studies on flood sensitive

116

Arabidopsis thaliana. These investigations have identified the main players in

117

energy and carbon signaling (Smeekens et al., 2010; Ljung et al., 2015) and

118

revealed whole plant and cell-type-specific transcriptional and translational

119

adjustments induced by low O2 stress (Mustroph et al., 2009; Juntawong et

120

al., 2014). Importantly, O2 dependent degradation of the group VII family of

121

ethylene response factors (ERF-VIIs) via the N-end rule pathway of protein

122

degradation has been identified as a molecular mechanism that translates O2

123

availability into transcriptional reprogramming (Licausi et al., 2011; Gibbs et

124

al., 2011; Weits et al., 2014). Recent studies have also revealed how this

125

molecular hypoxic response is highly regulated and fine tuned to maintain

126

cellular homeostasis during low O2 conditions (Gibbs et al., 2014; Giuntoli et

127

al., 2014; Gonzali et al., 2015).

128

Despite the progress in our understanding of flooding-induced signaling

129

pathways, much remains to be discovered regarding the molecular

130

mechanisms that cause variation in flooding tolerance across and within

131

species (Voesenek and Bailey-Serres, 2015). Variation in flooding responses

132

amongst natural plant populations is an important tool to identify the

133

underlying causal genes and processes (Xu et al., 2006; Magneschi et al.,

134

2009; Chen et al., 2010; Cambell et al., 2015). Despite their relative

135

intolerance to flooding stress, Arabidopsis accessions show considerable

136

variation in their tolerance to complete submergence (Vashisht et al., 2011).

137

Remarkably, this variation is not linked to differences in internal O2 content or

138

initial carbohydrate reserves, the two parameters generally considered to be

139

essential for surviving flooding events.

140

The majority of studies investigating molecular regulation of transcriptional

141

reprograming in response to changes in O2 availability in Arabidopsis have

142

relied on hypoxia and/or used agar-based seedling assays (Baena-González

143

et al., 2007; Branco-Price et al., 2008; Mustroph et al., 2009; Dennis et al.,

144

2009; Christianson et al., 2009; Banti et al., 2010). However, in natural

145

conditions, flooding results in a gradual decline in O2 levels and is often

146

accompanied by other physiological changes, such as a rapid build-up of the

147

gaseous hormone ethylene (Voesenek and Sasidharan, 2013). Furthermore,


5

148

flooding imposes distinct environmental constraints on the root and the shoot,

149

and thereby also elicits different physiological responses. Accordingly, an

150

exploration of shoot and root responses of flooded, soil-grown plants is more

151

relevant for understanding flooding stress as experienced in the field.

152

Here, we characterized the early molecular response to darkness and

153

flooding acclimation in eight different Arabidopsis genetic backgrounds

154

(Supplemental Table S1), varying in their tolerance to complete submergence,

155

using poly(A)+ mRNA-sequencing (mRNAseq). The use of soil-grown plants

156

subjected to submergence (in the dark), mimicked naturally flooded conditions

157

in a highly controlled way, and the inclusion of a darkness only (without

158

submergence) treatment, allowed us to simultaneously disentangle dark

159

effects from submergence effects (Lee et al., 2011; Vashisht et al., 2011).

160

Given the distinct carbohydrate and O2 status of the root and shoot (rosette)

161

tissues under these two stress conditions, these organs were analyzed

162

separately, and then each organ response was compared to the other. This

163

was performed for all eight accessions. Our data suggests an important role

164

for gluconeogenesis in short-term stress acclimation, which includes

165

alternative splicing of transcripts encoding key regulatory enzymes and

166

quantitative transcriptional differences between tolerant and intolerant

167

accessions. A conservative mode of energy and resource utilization via

168

metabolic reprogramming and constrained growth contributes towards

169

prolonged survival underwater. Shoot specific flooding-induced transcriptional

170

reprogramming was primarily growth related, whereas in the root mainly

171

plastidial and developmental processes changed. Our results provide insight

172

into the interactive and additive effects of the different elements of flooding

173

stress, present a detailed picture of early molecular events mediating stress

174

acclimation, and identify putative novel aspects of flooding tolerance.

175 176

Results

177 178

Early transcriptomic responses to flooding and darkness are largely

179

conserved amongst accessions

180

To identify early transcriptome modifications upon flooding and dark-

181

induced starvation, eight Arabidopsis genotypes (Cvi-0, Bay-0, Ita-0, Col(gl1),


6

182

Kas-1, Lp2-6, Ws-2, C24) (Supplemental Table S1) were exposed to light (air

183

+ light; AL), dark (air + darkness; AD) or complete submergence (submerged

184

+ darkness; SD) for 4 hours (Figure 1A). At this time point, the decline in O2

185

levels caused by submergence have stabilized in both the root ( 0.1) and also showed a

245

considerable response to the imposed stresses (Pmean response, adj. < 0.01, and

246

log2FC|1.6|) (Figure 2).

247

Typical overrepresented processes for conserved darkness and compound

248

effects in all accessions were related to downregulation of energetically

249

expensive cell wall construction, sulphur metabolism, starch biosynthesis


8

250

(shoots only) and secondary metabolism. Interestingly, sucrose and fructose

251

responses

252

overrepresented amongst upregulated genes in both the root and shoot. Not

253

surprisingly, the ‘response to absence of light’ category was also

254

overrepresented amongst the upregulated genes in both shoot and root in

255

dark and compound stress. We also identified ‘cellular response to iron ion’

256

overrepresentation amongst downregulated genes in the root in response to

257

the compound stress. The GO analyses further revealed changes associated

258

with nitrogen metabolism. This included the ‘nitrate transport’, ‘amino acid

259

transport’ and ‘leucine catabolic process’ terms that were overrepresented

260

amongst upregulated genes in the shoot. Together these results suggested a

261

fundamental change in the metabolic network in response to the applied

262

stresses.

and

trehalose

phosphate

synthase

activity

terms

were

263

Compared to the darkness and compound effects, a lower number of

264

significantly enriched GO terms were identified for submergence stress only.

265

The upregulated genes, as expected, included anaerobic metabolism, hypoxic

266

response and sucrose synthase activity. Other uniquely submergence

267

responsive GO categories were hormone related (ethylene, auxin, abscisic

268

acid (ABA)), indicating transcriptional regulation associated with these

269

hormonal cascades that is not activated by the more metabolically determined

270

darkness effects.

271

To

characterize

the

accession-dependent

responses,

GO

272

overrepresentation analysis was also performed on genes that varied in their

273

treatment response (Paccession*treatment,

274

The GO terms enriched amongst these genes encompassed a wide range of

275

categories. These were mostly associated with photosynthesis and

276

metabolism (lipids, amino acids and sulphur) and biotic defense. There was

277

some overlap in the GO enrichment categories between the conserved and

278

accession-dependent responses. This indicated a strong regulation of the

279

related processes but in varying levels of conservation amongst accessions.

adj.

< 0.05) (Supplemental Figure S3).

280 281

The compound stress response is an amplified darkness response in

282

the shoot


9

283

In nature, severe flooding often consists of submergence coupled with very

284

low light intensities. Here we investigated the relative contribution of darkness

285

and submergence towards the final compound flooding stress response. In

286

the shoot, the direct comparison of the compound and darkness response

287

showed a strong positive correlation (Figure 3A). The steep slope suggested

288

that the compound response was similar to the dark response but was

289

enhanced by the addition of submergence. A similar comparison of the

290

compound versus darkness response of the root also showed a strong

291

correlation. However, no additional effect of submergence on the global

292

transcriptomic response was observed, i.e. the compound and darkness effect

293

were of a similar magnitude (Figure 3A).

294

To further characterize the gene categories constituting the relationships

295

identified above, we grouped genes that were co-expressed, and are

296

therefore potentially members of same or similarly regulated gene pathways.

297

We used Weighted Gene Co-expression Network Analysis, WGCNA;

298

(Langfelder and Horvath, 2008) to perform a comparative analysis of gene

299

networks between the three conditions (AL, AD, SD). Fifteen and eight co-

300

expression modules were identified for the shoot and root, respectively, where

301

each module consists of genes that show largely similar expression patterns

302

across the different accessions and conditions (Figure 3B, Supplemental

303

Figure S4). GO term enrichment was subsequently investigated for the

304

identified modules (Supplemental Data, Sheet F and G).

305

For both the root and the shoot, two very large gene co-expression

306

modules were identified, namely R01, R02, S01 and S02 (Figure 3B). The

307

R01 and R02 module both showed a consistent change in expression upon

308

darkness (either an increase or decrease) in all accessions, but no change

309

upon submergence. However, R01 and R02 differed in the constitutive

310

expression levels of the accessions (Figure 3C). These were enriched in GO

311

terms related to metabolism such as glycolysis/gluconeogenesis, fatty acid

312

breakdown, acetylCoA and secondary metabolism (Glucosinolates and

313

isopentenyl pyrophosphate (IPP)/methylerythritol (MEP) pathway), but also

314

included sugar transport and signaling. Enrichment terms also indicated a role

315

for jasmonic acid and brassinosteroids in the root upon darkness and

316

compound stress (Figure 3C).


10

317

By comparison, the genes in module S01 were expressed similarly in all

318

accessions and only had a darkness response and no additional

319

submergence effect (Figure 3D). This module was enriched for GO categories

320

related to the photoperiod, lipid breakdown (in the peroxisome), protein

321

transport (required for peroxisome function), and sugar-mediated signaling

322

(Figure 3D, Supplemental Data, Sheet G). Gene expression patterns in the

323

other large shoot module, S02, demonstrated the amplified dark response by

324

submergence for the compound stress. Enriched GO terms included starch

325

and secondary metabolism. Furthermore, enrichment was found for the

326

processes of cell division and meristem function. No clear submergence-

327

specific module was identified in the shoot or the root (Supplemental Figure

328

S4), likely because of the relatively small number of genes affected by

329

submergence only (Figure 1C).

330 331

Root- and shoot-specific treatment-responsive genes are associated

332

with photosynthesis and growth regulation

333

Since the organ specific responses to the treatments were more distinct

334

than the response across accessions (Figure 1B), these differences were

335

further explored. First, DEGs that were dependent on the organ, i.e., genes

336

with a significant organ*treatment interaction, were identified (Porgan*treatment, adj.

337

< 0.05, Supplemental Figure S5A, Supplemental Data Sheet H). These organ-

338

dependent treatment responses were largely conserved across accessions

339

(Supplemental Figure S5B). Genes with an organ*treatment interaction

340

(Porgan*treatment, adj. < 0.05) and a significant treatment effect (Padj. < 0.05) in only

341

one organ for six or more accessions were identified and designated as either

342

root- or shoot-specific response genes (Figure 4A, red and blue dots,

343

respectively). The number of shoot-specific genes identified for the

344

compound, darkness and submergence effect were 340, 33 and 13,

345

respectively. Fewer root-specific genes were found: 59 and 48 for compound

346

and darkness, respectively. There were no root specific genes for the

347

submergence response. Clustering of the organ specific genes identified a

348

strong overlap between the three treatments (Figure 4B). The compound

349

response of the root-specific genes mirrored the darkness response, whereas


11

350

shoot-specific genes of the compound response also illustrated the

351

amplification of the darkness response by submergence.

352

There was a very strong overlap in shoot specific-genes between the

353

compound, darkness and submergence responses (Figure 4B). Among these

354

shoot specific genes were those involved in hormonal metabolism and

355

signaling, cell growth and cell wall modification (Supplemental Data, Sheet H).

356

For

357

DIOXYGENASE 4, catalyzing a crucial enzymatic step in ABA biosynthesis,

358

was

359

CARBOXYLATE SYNTHASE (ACS) and 1-AMINOCYCLOPROPANE-1-

360

CARBOXYLATE

361

GA2oxidase6) and cytokinin (CYTOKININ OXIDASE 3) metabolism enzymes

362

were upregulated. Downstream signaling components typical for auxin and

363

brassinosteroids were among the upregulated shoot-specific genes (SMALL

364

AUXIN UPREGULATED (SAUR) and SAUR-LIKE genes, and AUXIN-

365

REGULATED GENE INVOLVED IN ORGAN SIZE (ARGOS); BXR1-

366

SUPPRESSOR1 (BZS1), BR ENHANCED EXPRESSION 1 and the BZR1-

367

interacting GENERAL REGULATORY FACTOR 8 (GRF8)). Downstream

368

effector genes such as cell wall modifying enzymes with shoot-specific

369

regulation (in both directions) included six genes involved in pectin

370

esterification, two cell wall loosening EXPANSINs and eight XYLOGLUCAN

371

ENDOTRANSGLUCOSYLASE/HYDROLASEs.

example,

the

mRNA

levels

of

NINE-CIS-EPOXYCAROTENOID

downregulated, whereas ethylene (1-AMINO-CYCLOPROPANE-1-

OXIDASE

(ACO)),

gibberellin

(GA20oxidase

and

372

Several plant developmental control and light signaling genes were also

373

amongst the regulated shoot specific genes (Supplemental Data, Sheet H).

374

These included the genes SQUAMOSA PROMOTER-LIKE 11 responsible for

375

seedling to juvenile to adult stage transitions (Huijser and Schmid, 2011), but

376

also CLAVATA3/ESR-RELATED 16 (CLE16), CLE6 and CLAVATA2, that are

377

regulatory factors in shoot apical meristem activity (Gaillochet et al., 2015).

378

Amongst the light signaling genes were the negative regulators of

379

photomorphogensis B-BOX DOMAIN PROTEIN 18, SPA1-RELATED 3 and

380

FAR-RED ELONGATED HYPOCOTYL1 required for phyA signaling (Jigang

381

Li, 2011). The photoperiod-related gene FLOWERING bHLH 3 and circadian

382

clock gene PSEUDO-RESPONSE REGULATOR 9 were also amongst the

383

compound shoot specific DEGs.


12

384

In the root, the compound and dark responses were identical, and no

385

submergence root-specific genes were identified (Figure 4A and Figure 4B).

386

Interestingly, the root-specific upregulated genes consisted mainly of

387

chloroplast localized and photosynthesis related genes (Supplemental Data,

388

Sheet H). This included at least seven genes involved in photosystem

389

biosynthesis and maintenance, five additional proteins localized to the

390

chloroplast, one essential for chlorophyll biosynthesis and two involved in

391

photorespiration. Only a few root-specific downregulated mRNA were

392

identified, which included two nitrate transporters, and a MATE efflux protein.

393

In summary, mostly growth, developmental and hormonal regulatory gene

394

transcripts were stress-induced in the shoot, whilst chloroplast encoded and

395

photosynthesis associated genes dominated the root specific DEGs.

396 397

Induction of the core hypoxia gene set is organ independent only when

398

the darkness component is excluded

399

Previous studies identified 51 genes that were upregulated in Arabidopsis

400

seedlings upon hypoxic stress, regardless of organ or cell type (Mustroph et

401

al., 2009), and which are frequently used as core hypoxia response markers.

402

In soil grown plants, roots and shoots have distinct O2 profiles under both

403

control and submerged conditions. Soil grown roots of Arabidopsis are

404

constitutively hypoxic and upon submergence, internal O2 levels drop further

405

from 6 % to ~0 % pO2 KPa within 3 hours (Lee et al., 2011). Although the O2

406

dynamics of Arabidopsis leaf blades is unknown, the petiole goes from 17 %

407

to 6 % pO2 KPa upon submergence in the same time span. We investigated

408

the expression pattern of the 51 cell type-independent hypoxia-responsive

409

genes in the context of the severe and mild low O2 levels in the submerged

410

root and shoot, respectively (Figure 4A, green dots; Figure 4C).

411

A majority of core hypoxia genes were regulated in both shoots and roots

412

upon compound, darkness or submergence. However an organ-independent

413

hypoxia signature response, involving upregulation of most of the 51 genes,

414

was only observed for the submergence response (when the effects of

415

darkness were excluded) (Figure 4C). This submergence response was also

416

very similar in magnitude in the roots and shoots. In contrast, for the

417

compound response, 18 out of the 51 core hypoxia genes were classified as


13

418

shoot specifically regulated (Porgan*treatment,

419

accessions). Only few of the hypoxia marker genes were classified as root or

420

shoot-specific upon darkness. However, during darkness, the root had a

421

predominant down regulation of most core hypoxia genes, and in the shoot

422

several were dark upregulated in Cvi-0 and Ws-2 (Figure 4C). Interestingly, a

423

small subset was induced upon darkness in both organs (At4g27450,

424

At1g33055, At1g19530, At4g39675, At5g61440, At3g61060). These were

425

previously identified as induced by C-starvation (Usadel et al., 2008) and

426

include EXORDIUM LIKE-1 (Schröder et al., 2011). In conclusion, it is clear,

427

that for the compound response, the behavior towards darkness is an

428

important determinant of the difference between the shoot and root for these

429

cell-type independent hypoxia marker genes (Figure 4A and Figure 4C).

adj.

< 0.05 in six or more

430 431

Conserved alternative splicing events indicate an additional layer of

432

regulation in the adaptation to compound, darkness and submergence

433

stresses.

434

By using mRNAseq as a platform we were able to investigate transcriptome

435

reconfiguration at the mRNA isoform level corresponding to variations in

436

mRNA splicing events. These events can include exon skipping, mutually

437

exclusive (alternative) exon usage and alternative donor and acceptor splice

438

sites that can alter protein coding and untranslated regions, all of which are

439

generally termed as alternative splicing (AS). Another event is intron retention

440

(IR), which involves the retention of introns in the mature mRNA. IR events

441

that result in an open reading frame that is upstream of an intron junction

442

typically target transcripts for nonsense mediated decay and are therefore

443

unstable mRNA isoforms (Kazan, 2003). We focused on splice site selection

444

and IR variants similar to the method described in Chang et al. (2014), by

445

characterizing the relative increase or decrease in specific variants between

446

samples. More specifically, for each accession, we characterized AS and IR

447

induced by the three treatments (compound, darkness, submergence); and for

448

each of the three conditions (AL, AD, SD) we characterized the relative

449

variant usage between the eight accessions.

450

A considerable level of both IR and AS events was identified between the 8

451

accessions, which was largely independent of the treatment (Supplemental


14

452

Figure S6A and S6C). 1819 and 1014 IR events were treatment independent

453

(|log2FC| > 1, Padj. < 0.01) in the root and shoot respectively (Supplemental

454

Figure S6B). For AS, 2061 and 1798 treatment independent root and shoot

455

events (|log2FC| > 1, Padj. < 0.01) were identified (Supplemental Figure S6D).

456

The consistency and strong overlap in AS and IR across the three conditions

457

(AL, AD and SD) indicated that these are robust differences between the

458

accessions. However, with respect to the acclimative responses to darkness,

459

compound and submergence stress, the treatment-induced splicing events

460

were of more interest (Figure 5). While 1214 and 2122 genes with IR events

461

(|log2FC| > 1, Padj. < 0.01, in at least one accessions) were found in the root

462

and shoot, corresponding numbers for treatment induced AS events (|log2FC|

463

> 1, Padj. < 0.01, in at least one accessions) were 210 (root) and 2471 genes

464

(shoot). For both AS and IR events, the overlap between accessions was

465

minimal (Figure 5B and Figure 5D). However, the genes with conserved

466

splicing behaviour across the accessions were of interest as robust examples

467

of darkness and submergence-induced post-transcriptional regulation. Indeed,

468

the 167 and 63 genes in the shoot and root, that showed IR in 5 or more

469

accessions, showed consistent behavior across the accessions and,

470

depending on the gene, IR was favoured either upon the stress condition (AD

471

and SD) or under air light conditions (Supplemental Figure S7 and

472

Supplemental Figure S8).

473

Compared to IR, fewer treatment-dependent conserved AS events were

474

identified, with 15 and 31 genes displaying AS in five or more accessions for

475

roots and shoots, respectively (Figure 6). For these genes this additional

476

aspect of transcriptome reconfiguration in response to compound, darkness or

477

submergence stress would not only affect mRNA stability, as is the case for

478

IR, but could potentially also lead to altered protein function, localization or

479

influence other post-transcriptional processes such as translational efficiency.

480

These regulatory processes would be in addition to the differences we already

481

observed in the total transcript abundance of these genes between the

482

treatments (Supplemental Figure S9). To verify the validity of the observed AS

483

patterns, independent qRT-PCR analyses using Cvi-0 and C24 accessions as

484

representatives was done. We tested 6 genes that in addition to showing

485

distinct AS patterns, also showed strong transcriptional regulation. All the six


15

486

genes tested, confirmed the mRNAseq-based evidence of AS (Figure 7). This

487

also revealed that AS began within a few hours of stress and persisted at

488

elevated levels over 48 h of compound, darkness and submergence

489

treatments (Figure 7). For most genes tested, the increase in splice variant

490

isoform(s) occurred rapidly and then declined somewhat (e.g., ROPGEF11,

491

At1g52240; Figure 7). Several of the 46 genes with conserved AS events

492

could have an important role in acclimation to the imposed stress.

493

For instance in the root, ROP (RHO OF PLANTS) GUANINE NUCLEOTIDE

494

EXCHANGE FACTOR 11 (ROPGEF11, At1g52240) preferentially produced a

495

short transcript over a longer transcript isoform under darkness and

496

compound stress, with total transcript abundance elevated by both stresses

497

(Figure 6 and 7). The shorter isoform lacks the Rop Nucleotide Exchanger

498

domain (PRONE, PF03759), which for ROPGEF11 is implicated in

499

phytochrome interactions in the regulation of root development (Shin et al.,

500

2010), and instead contains a dynein light chain domain (PF01221) (Figure

501

8A). Another interesting gene was ARABIDOPSIS RESPONSE REGULATOR

502

1 (ARR1), a cytokinin receptor (Kieber and Schaller, 2014) that was darkness-

503

induced and accumulated an alternatively spliced variant in the shoot and root

504

(Figure 6 and Figure 8B). Although AS of the pre-mRNA of ARR1 results in

505

transcript isoforms that encode two distinct proteins, both contain all known

506

conserved domains of the receptor, but differ in their carboxy termini. The

507

isoform variant of ARR1 that is elevated in darkness also has a shorter and

508

distinct 3’ untranslated region, which could influence interaction with RNA

509

binding proteins that alter mRNA stability or translation.

510

Among the genes displaying AS was a relatively large group of genes

511

associated with metabolic functions (Figure 6). Many of these were

512

differentially regulated by either darkness or submergence. These included

513

two FRUCTOSE BISPHOSPHATE ALDOLASEs (FBA1 and FBA5), a

514

SUCROSE

515

DEHYDROGENASE 2. Also of relevance was PEROXISOMAL NAD-MALATE

516

DEHYDROGENASE 2 (PMDH2), which is required for redox balance during

517

fatty acid breakdown in the peroxisome (Pracharoenwattana et al., 2007).

518

PMDH2 AS upon darkness and submergence favored an enzyme form with

519

increased activity (Figure 8C). Of particular interest were the AS patterns of

6-PHOSPHATE

PHOSPHORYLASE

and

GLUTAMATE


16

520

PYRUVATE ORTHOPHOSPHATE DIKINASE (PPDK, At4g15530) and

521

LYSINE-KETOGLUTARATE

522

DEHYDROGENASE (LKR/SDH, At4g33150). PPDK is a single-copy gene

523

that is induced by low O2 in a variety of species (Huang et al., 2008). Of the

524

five PPDK transcript isoforms, the shortest transcript (At4g15530.2) preferably

525

and progressively accumulated upon darkness and submergence (Figure 7

526

and Figure 8D). This transcript encodes the cytosol-localized form of PPDK

527

(Parsley and Hibberd, 2006) that is suggested to be important during amino

528

acid remobilization and senescence (Taylor et al., 2010), and for

529

gluconeogenesis (Eastmond et al., 2015). Interestingly, several other genes

530

involved in gluconeogenesis were quantitatively up or downregulated in

531

response to the compound and darkness treatments (Supplemental Figure

532

S10). Of these, only PPDK showed pronounced upregulation across

533

accessions under submergence.

REDUCTASE/SACCHAROPINE

534

LKR/SDH (At4g33150) encodes a bifunctional enzyme catalyzing the first

535

two steps of lysine catabolism. While the LKR component of the protein works

536

in the lysine catalytic direction, the SDH component has bidirectional

537

enzymatic activity (Zhu et al., 2002). Besides being strongly upregulated by

538

darkness and compound stress in roots and shoots (Supplemental Figure S9),

539

AS of LKR/SDH was evident in the root (Figure 6, Figure 7 and Figure 8E),

540

with the longer transcript being favored under both conditions (Figure 7E).

541

The long transcript results in a protein with both LKR and SDH activity,

542

whereas the short transcript only has SDH activity (Zhu et al., 2002).

543

In the shoot, a relatively large amount of AS occurred in transcripts related

544

to photorespiration (GLYCERATE KINASE, GLYCOLATE OXIDASE 1), light

545

capture (PHOTOSYSTEM 1 LIGHT HARVESTING COMPLEX GENE 1, a

546

putative

547

QUENCHING 1 and 4 (NPQ1 and 4)), CO2 sensing (BETA CARBONIC

548

ANHYDRASE 1 AND 4 (BCA1 and BCA4)) and plastid development (F-box

549

family protein, PLASTID REDOX INSENSITIVE 2 (PRIN2), TRANSLOCON

550

AT THE OUTER ENVELOPE MEMBRANE OF CHLOROPLAST). These AS

551

variants do not necessarily lead to distinctions in the encoded protein, but

552

modify untranslated regions of the mRNA (Figure 8F and Figure 8G), and

553

hence could influence other post-transcriptional processes. Intriguingly, both

cytochrome

b6f

complex

subunit,

NON

PHOTOCHEMICAL


17

554

beta carbonic anhydrase loci (BCA1 and BCA4) displayed differences in the 5’

555

UTR between the treatments (Figure 8H and Figure 8I). The isoforms

556

preferentially accumulating in darkness and submergence encode N-

557

terminally truncated proteins that retain enzymatic activity but lack the

558

sequences responsible for specific subcellular targeting (BCA1 to the

559

chloroplast and BCA4 to the plasma membrane) (Fabre et al., 2007).

560

Altogether these data demonstrate that AS can serve as a post-transcriptional

561

control point that impacts the accumulation, location and activity of a number

562

of proteins that regulate carbon flux.

563 564

Natural variation in submergence tolerance is associated with relatively

565

minor transcriptomic differences in a group of putative “tolerance

566

genes”

567

A previous study that used an identical experimental set up showed

568

considerable variation in the tolerance to complete submergence in the dark

569

(compound stress) among 86 Arabidopsis accessions (Vashisht et al., 2011).

570

This variation could not be ascribed to difference in anatomy, decline in organ

571

O2 content, or initial carbon resources. Based on that study, three accessions

572

profiled here were classified as submergence sensitive (Cvi-0, Bay-0 and Ita-

573

0), and three as tolerant (Lp2-6, Ws-2, C24). The tolerant accessions also

574

performed better when their survival under submerged conditions (SD) was

575

compared to their survival under darkness (AD) (submergence stress). Based

576

on this prior study, genes that responded differently to compound and

577

submergence stress were identified by comparing the tolerant with the

578

sensitive accessions (Ptolerance*treatment, adj.. < 0.05, Supplemental Data Sheet K).

579

In this way, 33 and five potential ‘tolerance genes’ were identified in the shoot

580

for

581

Supplemental Data Sheet K). A larger number of potential tolerance genes

582

were identified for the root (47 compound and 43 submergence). Although this

583

relatively large number of potential tolerance genes could be responsible for

584

differential tolerance among these accessions, the magnitude of the

585

difference was not always large (Figure 9B).

compound

and

submergence

stress,

respectively

(Figure

9A,

586

Interesting shoot ‘potential tolerance genes’ with a stronger overall increase

587

in expression in tolerant genotypes included the PPi utilizing and


18

588

gluconeogenic enzyme PPDK (AT4G15530) and a gene encoding a natural

589

antisense RNA (AT4G20362) to a RAB GTPase homolog (AT4G20360). Also

590

predominantly upregulated in tolerant genotypes was plant DEFENSIN 1.2b

591

(AT2G26020). Additionally the shoot potential tolerance-genes included

592

several growth- and cell wall-associated genes such as XTH4, an auxin

593

responsive GH3 family protein and EXPANSIN A16, which were more

594

induced in the sensitive accessions (Supplemental data, Sheet K). This

595

suggested a more conserved growth response in the tolerant accessions in

596

our treatment conditions. To assess whether this was indeed true, petiole

597

elongation rates of a sensitive (Cvi-0) and tolerant (C24) accession were

598

measured as a marker for shoot growth (Supplemental Figure S11). The

599

tolerant C24 had a greater reduction in petiole elongation rates (relative to AL)

600

compared to Cvi-0 in both light and dark submerged conditions. In the dark,

601

Cvi-0 petiole elongation rates were similar to control plants. In contrast, in

602

C24, petiole elongation rates were reduced to less than 33% of control (AL)

603

rates.

604

The ‘potential tolerance genes’ from the root were of a different nature

605

compared to the shoot, and no overlap in gene composition was found

606

between the two organs. For instance FERRIC REDUCTION OXIDASE 4 and

607

5 (FRO4 and FRO5, AT5G23980 and AT5G23990), which play an important

608

role in the uptake of iron and copper from the soil (Jain et al., 2014; Bernal et

609

al., 2012), were identified. Additionally a vacuolar iron transporter and a metal

610

transporter

611

AT5G59520, ZRT/IRT-LIKE PROTEIN 2) were classified as ‘potential

612

tolerance genes’. All of these genes had stronger downregulation in the

613

tolerant accessions, especially upon submergence. Another root potential

614

tolerance gene, LOW PHOSPHATE ROOT1 (AT1G23010) is involved in

615

sensing and signaling of low Pi availability in the root in an iron dependent

616

manner (Svistoonoff et al., 2007; Müller et al., 2015), and was upregulated in

617

the sensitive and downregulated in the tolerant accessions (Supplemental

618

Data, sheet K).

(At3g25190,

VACUOLAR

IRON

TRANSPORTER-LIKE

5;

619

Interestingly, several of the ‘potential tolerance genes’ identified here have

620

been previously identified as commonly hypoxia regulated through the plant

621

kingdom (Mustroph et al., 2010). This was especially the case in the root, with


19

622

tolerance group dependent regulation upon the compound stress for

623

HYPOXIA UNKNOWN PROTEIN 37 (HUP37, AT2G41730), SIMILAR TO

624

RCD ONE 5 (SRO5, AT5G62520), an unknown protein (AT3G23170) and

625

CALMODULIN-LIKE 38 (AT1G76650), recently shown to be a calcium-

626

regulated cytosolic RNA binding protein (Lokdarshi et al., 2016). A close

627

homolog of SRO5, namely SRO4 (AT3G47720) was also identified as a

628

‘tolerance gene’ in the root. For the shoot compound stress this category of

629

genes included PPDK (AT4G15530) and PYRUVATE DECARBOXYLASE 1

630

(AT4G33070).

631 632

Discussion

633

Anatomical and physiological features contribute little towards the large

634

variation in the tolerance to darkness and submergence observed in A.

635

thaliana accessions (Vashisht et al., 2011). For this reason, the short-term

636

transcriptomic acclimation was investigated in a selection of eight accessions

637

under highly controlled conditions that closely mimic natural flooding events in

638

the field. Furthermore, the experimental design made it possible to

639

disentangle the effects of darkness from the responses caused by reduced

640

gas diffusion (i.e., O2, CO2 and ethylene) in the underwater environment. This

641

allowed us to identify conserved processes in relation to the naturally

642

occurring stress, whilst simultaneously identifying accession and tolerance

643

specific processes. The systematic comparison of the shoot and root

644

transcriptomic adjustments of eight accessions to flooding and starvation

645

stress revealed robust conservation in a response, which encompasses

646

specific transcript isoform production, but also contained subtle and possibly

647

significant distinctions between flooding tolerant and intolerant accessions.

648

Our results emphasize that understanding plant adaptation to flooding

649

requires consideration of its compound nature, which often includes reduction

650

in light or even complete darkness in combination with a severe reduction of

651

gas exchange.

652 653

Alternative

654

transcriptional response to darkness and submergence

metabolic

reserve

mobilization

as

a

coordinated


20

655

Darkness and submergence both cause similar physiological changes in

656

affected plants, including carbohydrate depletion, utilization of alternative

657

carbon sources and chlorophyll degradation. However, Arabidopsis has an

658

amazing capacity to buffer its metabolism to unexpected darkness (Graf et al.,

659

2010). An unexpected early night leads to appropriate reductions in the rate of

660

starch breakdown. This adjustment allows the available carbohydrate

661

reserves to last throughout the longer-than-expected night and thus prevents

662

starvation-related transcriptional responses. However, our dataset revealed

663

transcriptome reconfigurations typical of starvation responses already within 4

664

hours. This suggests that imposing darkness or submergence stress during

665

approximately the first half of the light period under short day conditions is

666

taxing on the buffering capacity of Arabidopsis. The metabolic buffering

667

capacity has been shown to require the circadian clock and targets starch

668

breakdown (Graf et al., 2010). Consistently, in our analysis starch breakdown

669

and biosynthesis were predominately down regulated upon darkness, an

670

effect that was even stronger when coupled with submergence. However,

671

while the clock and circadian machinery were also affected by darkness, there

672

was no additional effect of submergence, in line with the large contribution of

673

light cues in entraining the circadian rhythm (Hsu and Harmer, 2014).

674

Instead of starch, the transcriptional activation of fatty acid and amino acid

675

breakdown

676

(intracellular degradation and recycling of cellular components). Arabidopsis

677

mutants defective in autophagy are highly sensitive to submergence (Chen et

678

al. 2015). Several of these genes involved in fatty acid and amino acid

679

breakdown are crucial to maintain energy status and performance during

680

stress conditions or in non-photosynthetic developmental stages. This was

681

shown in mutant studies on LKR/SDH (lysine breakdown) and PMDH2 (fatty

682

acid beta oxidation) in heterotrophic germinating seeds (Pracharoenwattana

683

et al., 2007; Angelovici et al., 2010). Interestingly, for both these genes,

684

specific transcript isoforms were preferentially induced upon compound,

685

darkness and submergence stress. For LKR/SDH, the preferentially induced

686

longer transcript favors lysine breakdown (Tang et al., 2002), and for PMDH2,

687

isoforms with the intact enzymatic domain preferentially accumulated,

688

suggesting increased enzymatic activity upon the imposed stresses. Several

was

observed,

indicating

the

occurrence

of

autophagy


21

689

other genes encoding key metabolic enzymes of the amino acid and fatty acid

690

breakdown pathways, displayed AS upon compound, darkness and

691

submergence (Supplemental Figure S12). This suggests that AS provides an

692

additional layer of regulation that could have a significant impact on

693

metabolite fluxes during these stress conditions.

694

A strong upregulation of transcripts of glyoxylate pathway enzymes was also

695

observed. The glyoxylate pathway shortcuts a part of the TCA cycle (from

696

isocitrate to succinate) thereby preventing the loss of two CO2 molecules and

697

thus preserving fixed carbon. This suggests that protein and fatty acid

698

breakdown is not necessarily being utilized in respiration and energy

699

production, but also is used for sugar biosynthesis. The activation of key steps

700

of the gluconeogenic pathway, involving PEP CARBOXYKINASE (PCK) and

701

PPDK, upon compound, darkness and submergence further points towards

702

the utilization and mobilization of alternative carbon resources (Supplemental

703

Figure S12). PPDK regulation was of special interest since the cytosolic

704

transcript isoform that was preferentially upregulated has been shown to

705

increase nitrogen mobilization when overexpressed (Taylor et al., 2010).

706

Furthermore, PPDK was more strongly upregulated upon compound stress in

707

the tolerant accession C24 than the sensitive Cvi-0, a pattern that persisted

708

over time (Figure 7). Besides their function in C4 photosynthesis, PCK and

709

PPDK have been studied primarily in the context of their gluconeogenic role in

710

reserve mobilization in germinating seeds (Penfield et al., 2004; Delgado-

711

Alvarado et al., 2007; Malone et al., 2007; Eastmond et al., 2015), and around

712

vein tissue where they also display high activity (Hibberd and Quick, 2002;

713

Brown et al., 2010). Furthermore, PPDK and PCK are upregulated in

714

submerged Rumex acetosa (van Veen et al., 2013), anoxic rice coleoptiles

715

(Narsai et al., 2009) and waterlogged Arabidopsis roots (Hsu et al., 2011).

716

In line with previously identified roles and functions we hypothesize that

717

under compound, darkness and submergence stress, the enzymes PCK and

718

PPDK could occupy a key function in fuelling starving plant organs with

719

reserves alternative to starch, possibly by redirecting energy rich metabolites

720

from source leaves to sink meristems and roots. An additional adaptive

721

benefit of utilizing alternative resources to starch is that the maintenance and

722

upkeep associated with high protein levels and organelles (Amthor, 2000) can


22

723

be minimized. Thus sacrificing older leaves, minimizing the requirements of

724

young leaves and concentrating resources in the meristems might provide a

725

useful strategy to persist under adverse flooded conditions.

726

A common observation in studies profiling metabolic changes upon flooding

727

is an increase in the levels of certain amino acids. This has been documented

728

for instance, in submerged rice shoots (Barding et al., 2013), anoxic rice

729

coleoptiles (Narsai et al., 2009) and waterlogged poplar and Lotus japonicus

730

(Kreuzwieser et al., 2009; Rocha et al., 2010). Indeed, in senescing leaves,

731

increased protein breakdown and amino acid catabolism coincides with

732

increased amino acid content (Hildebrandt et al., 2015; Watanabe et al.,

733

2013). Similarly, in petioles of submerged R. acetosa plants a large increase

734

in free ammonia was observed upon submergence, suggesting an increased

735

amino acid breakdown (van Veen et al., 2013). Further metabolic evidence for

736

amino acid catabolism comes from Arabidopsis mutants defective in energy

737

starvation signaling. In these lines, a similar suite of catabolic and

738

gluconeogenic genes were regulated as observed here, including PPDK, and

739

a subsequent altered metabolic profile was observed (Hartmann et al., 2015).

740

However, it is difficult to discern metabolic fluxes from transcriptomic and

741

metabolomic data except when start or end products of specific routes are

742

quantified. Indeed, Rocha et al. (2010) and Antonio et al. (2015) provide a

743

model based on isotope flux determination, specific to low O2 availability

744

induced by waterlogging, where the intertwining of nitrogen metabolism and

745

the TCA cycle potentially doubles ATP production relative to glycolysis alone,

746

when the mitochondrial electron transport chain is compromised. This

747

requires pyruvate to be funneled to alanine, to prevent pyruvate-induced

748

respiration, the blocking of TCA cycle at succinate dehydrogenase (also

749

downregulated in our study (Supplemental Figure S12), and activation of the

750

GABA shunt. These, however could be fundamentally different processes

751

than what is observed under our experimental conditions, given the bulk of the

752

transcriptomic changes we observed occurred in response to darkness in air-

753

grown plants, where the advantages of these adaptations to low O2 would be

754

less apparent.

755

Although we assign a significant role to PPDK in resource mobilization, its

756

relevance during hypoxic conditions has been previously attributed to its role


23

757

in the low O2-induced switch to PPi dependent glycolysis (Huang et al. 2008;

758

Mustroph et al., 2014b). The primary advantage of PPi utilizing enzymes such

759

as PPDK is the conservation of ATPs and the yield of additional ATPs for

760

each sugar molecule going through glycolysis. This could be essential for

761

survival during low O2 conditions when the electron transport chain, which

762

provides the bulk of ATP, is hampered. However, the activation of PPDK

763

already under darkness strongly favors a role in gluconeogenesis, since under

764

these conditions the ATP gain is almost negligible. Although PPi utilizing

765

enzymes do provide a more energetically favorable route during anaerobic

766

metabolism, whether these pathways are preferred during low O2 conditions is

767

now under scrutiny. Analyses of sucrose synthase mutants under hypoxic

768

conditions suggest that, despite hampered performance under flooding stress,

769

a major portion of the carbon provided for glycolysis is still generated by the

770

ATP-dependent invertase and not via the PPi-linked sucrose synthase route,

771

at least in Arabidopsis thaliana (Santaniello et al. 2014). In the context of the

772

role of PPDK as well, a reassessment might be required for its precise role

773

during hypoxic glycolysis.

774

The observation that, upon submergence the shoot shows an amplification

775

of the darkness response, underscores the knowledge that plants tightly

776

adjust their metabolism to suit their environmental conditions. Additionally, we

777

observed that the transcriptional changes identified here are typical of carbon

778

and energy starvation, which requires resource mobilizations. In addition to

779

this, differential regulation of translation is another regulatory control point

780

under hypoxia and in darkness (Branco-Price et al., 2008; Pal et al., 2013;

781

Juntawong et al., 2014). Our studies also suggest that AS might play an

782

important role in this response as an additional layer of regulation in the

783

coordinated mobilization of existing and alternative reserves to endure

784

starvation conditions and prolong underwater survival (Supplemental Figure

785

S10).

786 787

Organ

788

responses

specific

transcriptome

reconfiguration

and

O2 dependent

789

Previous studies have shown that following low O2 stress and submergence,

790

the root and shoot transcriptomes are reconfigured in a distinct manner


24

791

underscoring variation between these organs in cues and protective

792

mechanisms (Ellis et al., 1999; Mustroph et al., 2009; Lee et al., 2011;

793

Mustroph et al., 2014a). The differences between root and shoot in their

794

transcriptome responses to darkness and submergence signals can be

795

attributed to several factors, including the autotrophic and heterotrophic

796

nature of the shoot and root, respectively, different cellular identities and

797

composition, distinct physiological functions and varying O2 profiles.

798

Endogenous O2 levels are primarily determined by a balance between the

799

internal production/consumption and the rate of inward diffusion from the

800

surrounding environment (soil, air or water). In Arabidopsis roots, O2 levels

801

drop from 6% to 0 % and in the petiole from 17 % to 6 % pO2 KPa (Lee et al.,

802

2011) within 3 hours (O2 levels in the lamina are unknown). The

803

transcriptomic profiling of both organs using mRNAseq allowed for a detailed

804

investigation of organ-specific responses to darkness, submergence and the

805

role of O2 herein.

806

The large-scale differences between the organs were typified by a higher

807

number of DEGs and stronger gene expression fold-changes in the shoot,

808

than the root. A possible explanation for the greater responsiveness of the

809

shoot is that roots continuously habituate a dark environment under control

810

(AL) conditions, meaning that the transition to dark treatment would have had

811

a relatively smaller impact. The fact that Arabidopsis roots are non-

812

photosynthetic and constitutively a sink may also contribute to their less

813

dramatic responsiveness. More striking, however, was the lack of an

814

amplification of the darkness responses in the compound transcriptome

815

behavior in the root. This could also reflect the existing sink-source

816

relationship between the root and shoots; wherein roots typically dependent

817

on the shoot for carbon resources were already maximally starved after the

818

dark treatment, whereas the shoot had more reserves to buffer the response.

819

Despite the fewer transcriptional changes observed in the root, several

820

processes were identified as root-specific. Plants have several sensing and

821

signaling mechanisms to detect changes in redox and maintain redox

822

homeostasis, which is important in all aspects of plant growth and

823

development. Interestingly, genes of this category were prevalent among the

824

root specific genes such as thioredoxins and rubredoxin. Furthermore, many


25

825

photosynthesis related genes were amongst the root-specific genes (e.g.,

826

photosystem I and II proteins). The increased expression of these genes and

827

several other chloroplast-associated genes in the roots indicates the presence

828

of chloroplasts in this organ, likely in the cortex (Dinneny et al., 2008). While

829

root greening has been described before, it is known to occur only in the

830

presence of a light signal (Usami et al., 2004; Kobayashi et al., 2012).

831

Elevated transcripts encoding photosynthesis-chloroplast associated proteins

832

in roots was reported previously in hypoxic/flooded seedlings of Arabidopsis

833

and Rorippa (Sasidharan et al., 2013; Chang et al., 2012). In contrast to this

834

work, the roots in these studies were at some point exposed to light signals.

835

Sugar starvation and salt stress are also reported to induce photosynthesis-

836

associated genes in roots (Sheen, 1990; Baena-González et al., 2007;

837

Dinneny et al., 2008). It has been speculated that this might be triggered by

838

reactive oxygen species (ROS) generated during the stress and with a

839

potential role in ROS amelioration (Dinneny et al., 2008). The relevance of the

840

expression of these genes in a root specific manner upon darkness and

841

compound stress in the absence of a light signal remains intriguing. The

842

regulation of ROS production is likely a relevant function in stressed roots.

843

However, whether this is the case here and what the underlying mechanism

844

is, remains to be determined.

845

Unlike the root, the shoot responded differently to darkness and

846

submergence. Shoot-specific genes upregulated by these conditions across

847

the accessions were associated with growth, senescence and oxidative

848

stress, all elements of the underwater response. Although there was an

849

upregulation of transcripts of genes associated with growth and growth-

850

associated hormonal signaling (GA, ABA), in the shoot, the petiole elongation

851

response to dark and compound stress was varied across accessions

852

(Supplemental Figure S11) (Vashisht et al., 2011). Considering that whole

853

shoots were sampled here, the involvement of these shoot-specific genes in

854

mediating changes in leaf expansion or perhaps hyponasty cannot be ruled

855

out. Nevertheless, the shoot core gene set reflects growth regulation and

856

extensive regulation of cell wall modifying proteins and growth regulatory

857

hormones. This likely is reflective of specific growth strategies that are an


26

858

important mechanism to deal with both flooding (van Veen et al., 2014b) and

859

low light conditions (Gommers et al., 2013).

860

Interestingly, amongst the compound shoot-specific genes, was a subset of

861

core hypoxia genes. Although previous studies have established their cell-

862

type independent hypoxia upregulation, the shoot-specific regulation here was

863

not surprising (Mustroph et al., 2009). Oxygen measurements on soil-grown

864

plants in an identical set-up have revealed that despite being in well-aerated

865

soils, these soil-grown roots were already hypoxic (~6% pO2). Considering the

866

already hypoxic conditions of roots under control conditions, it can be

867

speculated that constitutive expression of these genes is associated with

868

acclimation to hypoxic conditions. Unlike seedlings grown on vertical agar

869

plates that experience a normoxic to hypoxic transition, in our system,

870

submerged roots transition from hypoxic to severely hypoxic conditions (Lee

871

et al., 2011). Interestingly, it has been shown that low O2 in the root tip only is

872

sufficient to activate low O2 responsive genes throughout the entire root

873

(Mugnai et al., 2012). In active meristems, hypoxia is a common event during

874

periods of high activity (Van Dongen and Licausi, 2015). Interestingly,

875

darkness caused a significant repression of the core hypoxia genes in the

876

root. This suggests that although similar physiological responses are triggered

877

during submergence and darkness, associated with starvation conditions,

878

transcriptomic responses are prioritized to adapt to starvation in the presence

879

of O2. The core hypoxia signature including the inefficient fermentative mode

880

of energy generation would then be a wasteful mode of energy generation

881

under carbon limiting conditions.

882

Clearly, in the final compound response the behaviour in response to

883

darkness largely determines the difference between the shoot and root for

884

these cell type-independent hypoxia-responsive genes (Figure 4C). However,

885

overall a shift to severe or mild low O2 levels did not lead to a different

886

response when the effects of darkness were disregarded and only

887

submergence-induced hypoxia was considered. Even the magnitude of the

888

core hypoxia gene upregulation was similar for shoots and roots (Figure 4A

889

and Figure 4C). Remarkably, these distinct transcriptomic reconfigurations of

890

the shoot and root systems were highly conserved across the eight

891

accessions of Arabidopsis.


27

892 893

Natural variation in submergence tolerance

894

Natural variation in stress responses can be exploited to identify molecular

895

processes and components that regulate stress responses in a differential

896

way and therefore determine tolerance. Previous studies established

897

significant variation in flooding tolerance of six accessions used in this study

898

at the level of whole plant survival (Vashisht et al., 2011). This allowed us to

899

classify the accessions into two tolerance groups and correlatively identify

900

potential tolerance genes based on altered transcript accumulation or

901

distinctions in AS. Interestingly, the potential tolerance genes identified in the

902

root had no overlap with those from the shoot, further underscoring the

903

distinct physiological states and functions of these two organs. Despite the

904

lack of overlap between root and shoot potential tolerance genes, both

905

included members of the core hypoxia responsive gene set.

906

Interesting shoot potential tolerance genes that have previously not been

907

implicated in flooding survival included the gluconeogenic enzyme PPDK,

908

which was highly upregulated in the tolerant accessions. Targeted qPCR

909

analyses confirmed this trend over a 48h period (Figure 7) and revealed a

910

much stronger upregulation in the tolerant (C24) accession. Previous studies

911

have suggested a role for the PPi utilizing PPDK in mobilizing protein stores

912

(Huang et al., 2008) and in facilitating nitrogen remobilization in senescing

913

leaves (Taylor et al., 2010). Furthermore, it has a key position in a metabolic

914

network we identified as being important during starvation and submergence

915

(Supplemental Figure S12). It further stresses the importance of efficient

916

alternative reserve mobilization during energy limiting conditions. Future

917

biochemical and metabolic studies are necessary to determine if upregulation

918

of a cytosolic PPDK enhances the utilization of non-carbohydrate stores to

919

enhance energy production and long-term survival.

920

Another interesting shoot-specific potential tolerance gene was an antisense

921

to a RAB GTPase homolog (AT4G20360) with organellar translation

922

elongation factor (EF) activity. Previous studies have shown that Arabidopsis

923

seedlings exposed to hypoxia drastically limit translation as a means to curb

924

energy expenditure (Branco-Price et al., 2008; Juntawong et al., 2014;

925

Sorenson and Bailey-Serres, 2014). The upregulation of this antisense RNA


28

926

in tolerant accessions could serve to function to limit EF-Tu synthesis, thereby

927

limiting overall levels of plastid or mitochondrial mRNA translation.

928

Several shoot potential tolerance genes that were induced only in the

929

sensitive accessions had functions in growth- and cell wall-remodeling,

930

implying more dampened growth responses in the tolerant genotypes.

931

Consistently, we found that petiole elongation rates were significantly lower in

932

submerged plants (relative to control (AL) grown plants) of a tolerant

933

accession (C24). In contrast, a sensitive accession (Cvi-0) maintained control

934

petiole growth rates under compound stress (SD) conditions (Supplemental

935

Figure S11). Taken together, this suggests that tolerance in Arabidopsis can

936

be attributed to a conservative mode of energy utilization and efficient

937

carbohydrate management resulting in prolonged underwater survival.

938

Tolerance is a complex phenomenon, especially in the case of a compound

939

stress like flooding. Several aspects come into play, such as the

940

environmental conditions and the physiological state of the plant before,

941

during and after submergence. Our data suggests that tolerant Arabidopsis

942

accessions have restricted shoot growth and exhibit conservative and

943

alternative resource utilization, involving specific stress-induced metabolic

944

readjustments. This is likely an important factor influencing tolerance in the

945

shoot, whilst in the roots tolerance appears to involve genes related to

946

hypoxia and development. This suggests that for achieving tolerance, different

947

alterations may be required in the root and shoot.

948

The tolerance to starvation stress is another factor that interacts with low O2

949

stress to influence the final outcome of tolerance. As observed here, some

950

accessions (i.e., Ws-2 and Cvi-0) showed largely overlapping dark and

951

submergence transcriptomes. This suggests that the tolerance of Ws-2 could

952

partly be due to its ability to withstand starvation stress. Similarly, the

953

sensitivity of Cvi-0 is likely linked to its poor performance under dark

954

conditions (Vashisht et al., 2011). Accordingly, Cvi-0 petiole growth rates in

955

the dark (AD) equaled control (AL) rates. This would likely result in a faster

956

depletion of existing energy and carbohydrates reserves under stress

957

conditions and hasten plant demise.

958

Conclusions


29

959

The current upsurge in the number of global flooding events underscores the

960

importance of understanding tolerance mechanisms and plant responses to

961

flooding stress. Knowledge of the basis of variation in stress tolerance is also

962

critical for developing more stress resistant crops for environments

963

experiencing unexpected floods. This work details transcript abundance and

964

AS alterations in soil-grown vegetative-stage Arabidopsis rosettes that was

965

surprisingly conserved across a set of eight diverse genotypic backgrounds.

966

Contrasting this conservation was the distinct transcriptomic reconfiguration of

967

the shoot and root across the accessions, reflecting each organ’s anatomical

968

and

969

developmental plasticity as a result of the stress. We showed that alternative

970

selection of splice sites provides an additional layer of molecular regulation to

971

fine tune the response to flooding and starvation stress. Our study also

972

reveals that tolerance in Arabidopsis is related to the ability to restrict shoot

973

growth and exhibit conservative and alternative resource utilization, involving

974

specific stress-induced metabolic readjustments.

physiological

identity

and

highlighted

unique

metabolic

and

975 976

Materials and Methods

977

Plant material and growth conditions. Seeds of the studied accessions

978

(Bay-0: N22633, C24: N22620, Col(gl1): N3879, Cvi-0: N22614, Ita-0: N1244,

979

Kas1: N22638, Lp2-6: N22595, Ws-2: N1601) were obtained from Nottingham

980

Arabidopsis Stock Centre (NASC, UK). They were sown in a soil : perlite (1 :

981

2) mixture and cold (4°C) stratified in the dark for 4 days. Germination

982

occurred at 20°C, 160 μmol m-2 s-1 photosynthetically active radiation (PAR)

983

(9 h photoperiod) and 70% relative humidity. At the two-leaf stage seedlings

984

were transplanted, one seedling per pot (70 ml) with the soil : perlite (1 : 2)

985

mixture enriched with

986

Netherlands) and 0.14 mg of slow-release fertilizer (Osmocote ‘plus mini’;

987

Scotts Europe bv, Heerlen, the Netherlands) per pot. Each pot received 25 ml

988

of nutrient solution of the composition described by (Vashisht et al., 2011).

989

The soil was covered with black mesh with a small hole for the seedling to

990

grow through. The mesh prevented floating of soil material during

991

submergence experiments. Plants were grown in climate-controlled chambers

992

(20°C, 160 μmol m-2 s-1 PAR (9 h photoperiod) and 70% relative humidity).

0.14 mg MgOCaO (17%; Vitasol BV, Stolwijk, the


30

993

Plants were automatically watered each day at the start of the photoperiod.

994

Submergence

995

developmental stage of 10 leaves.

996

Experimental conditions. Plants were completely submerged in plastic tubs

997

(60 x 40 x 27 cm) filled to the brim with tap water and allowed to acclimatize

998

overnight. Both darkness only and darkness and submergence treatment took

999

place in the same conditions as plant growth, but with the lights off (in

1000

complete darkness). Experiments were started two hours after the start of the

1001

photoperiod. Tissue harvest was done with a low intensity green safe light.

1002

Samples for air light, air darkness and submergence darkness were harvested

1003

after 4 hours of treatment. Roots and shoots were harvested separately and

1004

the hypocotyl region (region between shoot base and at the beginning of first

1005

lateral root) was left out. The experiment was performed individually three

1006

times. Each time, five biological replicates, each including a pool of five

1007

individual plants were sampled and the tissues were flash frozen in liquid

1008

nitrogen. All samplings were completed within 30 minutes minimizing effects

1009

of circadian rhythms, and plants for each treatment were harvested

1010

simultaneously at the two chambers (light and dark chamber).

1011

Petiole elongation measurements. Plants of Cvi-0 and C24 were grown as

1012

described above and when they reach the 10 leaf developmental stage

1013

subjected the same experimental conditions. From a homogenous set of

1014

plants the leaf blades of the 7th developed leaf were marked with a pink dye.

1015

Petiole lengths were measured using a digital caliper before and after 72h of

1016

treatment on the same petiole.

1017

RNA extraction and sequencing. Plant tissue was ground with a mortar and

1018

pestle, after which the RNA was extracted using the RNeasy plant RNA

1019

isolation kit (Qiagen). DNA was removed via on-column DNAase digestion

1020

using the RNAase-Free DNase kit (Qiagen). For RNA sequencing, for each

1021

treatment, RNA samples consisted of RNA pooled from biological replicates

1022

that showed consistent results in terms of marker gene expression and petiole

1023

elongation response to submergence. Library preparation and sequencing (on

1024

a HiSeq 2000) was done commercially (Macrogen (www.macrogen.com)). All

1025

treatments for each accession per organ type were bar coded in the same

experiments

were

performed

after

plants

reached


a

31

1026

sequencing reaction to allow multiplexing of three samples per lane. Single

1027

end reads of 50 bp length were obtained.

1028

Quality control and read mapping. All the sequenced libraries had the

1029

Phred quality score ranging between 30-40 indicating 99.9% of base call

1030

accuracy. Therefore all reads were mapped to the TAIR10 Arabidopsis

1031

Columbia-0 genome by using tophat2 with bowtie2 (Kim et al., 2013) and

1032

allowing two mismatches. Only single hits were used for further analysis. The

1033

number of reads mapping to the exons, introns, and splice variant identifying

1034

gene regions (VIGRs; see MM splicing) were determined with the R packages

1035

genomic ranges (Lawrence et al., 2013), Rsamtools (Morgan et al., 2013). For

1036

the exons, only reads that had no overlap with non-exonic regions (i.e. introns

1037

or intergenic; IntersectionStrict) were counted, whereas for intron and VIGRs

1038

counts,

1039

(IntersectionNotEmpty).

1040

Differential expression analysis. Differential expression analysis was done

1041

with generalized linear modeling approaches of the R package edgeR

1042

(Robinson and Oshlack, 2010). Where no degrees of freedom were available

1043

a common dispersion of 0.08 was used, which is realistic for controlled

1044

experiments with genetically identical organisms and is conservative

1045

compared to common dispersion estimates that were assessed by known

1046

housekeeping genes in the dataset (0.02-0.06). Additionally, only genes with

1047

more than 1 RPKM in at least one sample were included. Differential

1048

expression upon treatment for each accession and organ was done with a

1049

model including all three conditions. Genes that responded differently in the

1050

shoot compared to the root were assessed in a full factorial model for each

1051

accession and treatment. The overall response across accessions (mean

1052

response) to the treatments was assessed in a paired design correcting for

1053

baseline differences amongst genotypes (i.e. an additive model with no

1054

interaction), which was done for each treatment and organ separately using

1055

tagwise dispersions. An analogous approach was taken for the organ

1056

dependent mean response, but with the added factor of the organ*treatment

1057

interaction.

1058

Genotype dependent treatment responses were determined with a full

1059

factorial model and with testing for accession*treatment effects. Here also

overlap

with

neighboring

genomic

regions

was

permitted


32

1060

organs and treatments were analyzed separately, and with the manual

1061

common dispersion parameter. Genes showing differentiation between

1062

tolerant and intolerant accessions were identified by contrasting the sensitive

1063

genotypes (Cvi-0, Bay-0, Ita-0) against the tolerant genotypes (Lp2-6, Ws-2,

1064

C24). Here the overall response of tolerant genotypes was tested for

1065

significant difference from the overall response of the sensitive genotypes.

1066

Where the overall responses of each group were determined by a paired

1067

design without an interaction term, subsequently tagwise dispersion was

1068

used.

1069

Gene ontology overrepresentation was assessed with the GOseq Rpackage

1070

(Young et al., 2010), which incorporates genelength biases. Multi-Dimensional

1071

Scaling was done with the edgeR package (Robinson and Oshlack, 2010).

1072

Weighted Gene Co-expression Network Analysis (WGCNA). WGCNA was

1073

used to calculate co-expressed gene modules (Langfelder and Horvath,

1074

2008). Since gene expression shows distinct patterns in roots and shoots,

1075

they were analyzed separately in the clustering analysis. The raw count data

1076

were filtered with RPKM method yielding 17525 and 15550 genes for roots

1077

and shoots, respectively. Library size normalization was done with the edgeR

1078

package (Robinson and Oshlack, 2010) and the data was transformed in

1079

limma package with “voom” function (Ritchie et al., 2015) in order to enable

1080

usage of WGCNA package designed for microarrays. The clustering was

1081

performed with default settings and soft thresholds of 4 and 9 were used for

1082

roots and shoots, respectively. As a representative of each module, an Eigen

1083

gene was calculated as the first PC axis of the gene expression patterns in

1084

that module. For each gene within a module, a module membership score

1085

was computed based on the similarity of the gene to this Eigen gene.

1086

Alternative Splicing. Estimation of alternative splicing (AS) and intron

1087

retention (IR) due to treatment or genotype effects was based on existing

1088

TAIR10 Col-0 annotation, using a method analogous to (Chang et al., 2014).

1089

Genomic regions which provide information regarding variant use, i.e.

1090

genomic regions which are transcribed in one variant but not in the alternative

1091

variant, were identified with the R package GenomicRanges (Lawrence et al.,

1092

2013). Reads mapping to these variant identifying gene regions (VIGRs) were

1093

counted

allowing

for

overlap

with

neighboring

genomic


regions 33

1094

(IntersectionNotEmpty). Similarly, reads mapping to unambiguous intron

1095

regions were counted to assess IR.

1096

Expected reads for each intron and VIGR where determined assuming that

1097

splice variant use and intron use does not change upon treatment or between

1098

genotypes. For instance, the expected reads for a treatment in a particular

1099

genotype would be (VIGRAL + VIGRAD + VIGRSD) / (ExonAL + ExonAD +

1100

ExonSD) * ExonSD. In case of multiple VIGRs in a single gene, they were

1101

calculated independently, whereas introns were grouped as one unit. Only

1102

introns and VIGRs with an average read count of more than 12 were included.

1103

The magnitude of AS and IR was determined by the ratio of the observed and

1104

expected reads, and subsequently log2 transformed so that AS and IR equals

1105

0 when observed and expected reads have equal values. Significance was

1106

estimated with a chi square test (sum((O-E)^2/E)) and Benjamini Hochberg

1107

corrected for multiple testing. Treatment dependent AS and IR was assessed

1108

separately for genotypes and organs. Genotype dependent AS and IR was

1109

assessed separately for each condition and organ. The maximum difference

1110

in splice variant usage is the highest log2(obs/exp) minus the lowest

1111

log2(obs/exp).

1112

RT-qPCR. From RNA extracted with the RNeasy plant RNA isolation (Qiagen)

1113

kit and treated with DNase (Qiagen), cDNA was made by reverse transcription

1114

(SuperScript® III Reverse Transcriptase, invitrogen) with random hexamers

1115

and including RNase inhibitor (ThermoScientific). qRT-PCR was performed in

1116

a Viia7TM Real/Time PCR system (ThermoScientific) using iTaq universal

1117

SYBR Green Supermix (Bio-Rad) in 5 µL reaction mixtures with gene-specific

1118

primers and five reference genes (Supplemental Table S4).

1119

Accession numbers

1120

The raw sequencing files from RNA sequencing are available in the

1121

ArrayExpress

1122

number E-MTAB-4730.

database

(www.ebi.ac.uk/arrayexpress)

under

accession

1123 1124 1125 1126

Supplemental Material. The following supplemental data accompanies this

1127

manuscript.


34

1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144

Supplemental Figure S1. Compound and darkness responses of the eight accessions in root and shoot tissues. Supplemental Figure S2. Number of genes with an accession dependent treatment response. Supplemental Figure S3. GO overrepresentation of genes that vary in their response across accessions. Supplemental Figure S4. Weighted Gene Co-expression Network Analysis (WGCNA) of shoots and roots. Supplemental Figure S5. Genes with an organ dependent response to the treatments. Supplemental Figure S6. Overview of intron retention (IR) and alternative splicing (AS) across genotypes and upon treatments. Supplemental Figure S7. The conserved responses in IR upon treatment in the root. Supplemental Figure S8. The conserved responses in IR upon treatment in the shoot. Supplemental Figure S9. Change in total transcript abundance of genes with

1145

evidence

1146

submergence.

1147 1148 1149 1150

of

alternative

splicing

upon

compound,

darkness

and

Supplemental Figure S10. Genes encoding important enzymatic steps of gluconeogenesis and the glyoxylate pathway. Supplemental Figure S11. Change in petiole growth rate upon different combination of darkness and submergence.

1151

Supplemental

1152

transcriptionally regulated by compound, darkness and submergence.

1153

Supplemental Table S1. Variation in tolerance to complete submergence in

1154

the dark of the 8 accessions used in this study.

1155

Supplemental Table S2. Summary statistics of Illumina sequencing of the

1156

mRNAseq libraries and subsequent mapping to the TAIR10 Arabidopsis

1157

thaliana genome.

Figure

S12.

Schematic

simplification

of

pathways


35

1158

Supplemental Table S3. Correlation statistics of the response of an

1159

individual genotype compared to the mean responses of all eight genotypes

1160

Supplemental Table S4. Primers used for qRT-PCR analyses of transcript

1161

abundance.

1162

Supplemental Data Set: Containing differential expression data from the

1163

RNAseq dataset for all the different comparisons investigated in the current

1164

study. (SupplementalData.xlsx)

1165 1166


36

1167 1168

Acknowledgements

1169

We would like to acknowledge Johanna Kociemba, Ankie Ammerlaan, Rob

1170

Welschen and Judith Koerselman for practical assistance; Reed Sorenson

1171

and Marcel van Verk for advice on mRNAseq data analyses.

1172 1173 1174 1175 1176

FIGURE LEGENDS

1177

submergence stress in eight Arabidopsis accessions.

1178

A. Schematic representation of the experimental setup, light cycle and

1179

treatments used. Arabidopsis seedlings were grown until the 10-leaf stage (9h

1180

photoperiod (Zeitgeber time (ZT): ZT0-ZT9). Plants then remained in control

1181

(air light; AL) conditions, or were transferred to submerged (dark) (SD) or dark

1182

(air dark; AD) conditions 2 h after photoperiod initiation (ZT2). Shoot and root

1183

material from Arabidopsis seedlings exposed to AL, AD and SD conditions for

1184

4h were harvested at ZT6, and used for mRNAseq. Black bars indicate

1185

darkness. Double ended arrows indicate datasets compared to deduce the

1186

compound stress (AL vs. SD), darkness (AL vs. AD) and submergence (AD

1187

vs. SD) differentially expressed genes.

1188

B. Multidimensional scaling (MDS) of all mRNAseq libraries, and of the shoot

1189

or root only, with distances based on the pairwise top 500 genes differing in

1190

fold change. AL = air light, AD = air darkness, SD = submergence darkness.

1191

C. Number of differentially expressed genes (DEGs) in shoots and roots (Padj

1192

< 0.05) in response to the compound, darkness and submergence stress.

1193

D. The compound, darkness and submergence response of individual

1194

accessions compared to the weighted average response of all eight

1195

accessions. DEGs (Padj. < 0.05) counted in 1C are depicted in the

1196

corresponding graph.

Figure 1. Transcriptional responses to compound, darkness and

1197 1198

Figure 2. Gene Ontology terms overrepresented in the conserved

1199

compound, darkness and submergence responses of the eight

1200

accessions.


37

1201

Overrepresentation was determined for genes where the average response

1202

across accessions was |mean Log2FC| >1.6 and Padj. < 0.01, and where

1203

variation in the response between accessions was absent (Paccession*treatment, adj.

1204

> 0.1). GO terms with a Padj. < 0.01 are shown.

1205 1206

Figure 3 Cumulative effects of darkness and submergence and WGCNA

1207

clustering.

1208

A. Comparison of the compound and dark responses of roots and shoots.

1209

Mean responses to treatments are plotted for both the x and y variable.

1210

Responses for individual accessions can be found in Supplemental Figure S1.

1211

The black dotted line represents y = x, the blue solid line is the regression of

1212

the data for the root (y = -0.02 + 0.93*x, r2 = 0.88) and shoot (y = -0.06 +

1213

1.23*x, r2 = 0.84). .

1214

B. Gene co-expression modules and their sizes as identified by a Weighted

1215

Gene Co-expression Network Analysis (WGCNA). Gene-modules show

1216

similar expression patterns across the three treatments and eight accessions.

1217

Roots (R) and shoots (S) were analyzed separately. Not all genes included in

1218

the analysis could be placed in a module of co-expressed genes and these

1219

were unplaced.

1220

C and D. Mean and variation centered RPKM values of the largest two root

1221

(C) and shoot (D) co-expression modules identified by WGCNA. Top 12% of

1222

the genes with the highest module-membership-score are shown. The blue

1223

and red lines reflect the trends of the gene with the strongest positive and

1224

negative correlation to the mean module behavior. The remaining modules

1225

are visualized in Supplemental Figure S4. To the right of each module are

1226

representative terms related to the identified enriched GO terms. The

1227

complete GO analysis is in the Supplemental Data Sheet F and G.

1228 1229

Figure 4. Organ-specific transcriptome reconfiguration and low oxygen

1230

responsive genes.

1231

A. Scatterplot showing the mean response of DEGs which show organ-

1232

dependent regulation by the treatment. Grey genes are those with a variable


38

1233

organ specific response (Porgan*treatment,

1234

Black

1235

(Porgan*treatment, adj. < 0.05 in 6, 7 or 8 accessions). Blue (shoot-specific genes)

1236

are genes that have a robust organ*treatment (black) but also a robust

1237

treatment response in the shoot (Padj. < 0.05 in 6, 7, or 8 accessions) but not a

1238

robust treatment response in the same direction in the root (Padj. < 0.05 in 6,

1239

7, or 8 accessions). Red circles (root-specific genes) are genes similar to blue

1240

but in roots. Green dots depict the cell-type-independent hypoxia-responsive

1241

genes (51 core hypoxia gene set) as defined by Mustroph et al. (2009).

1242

B. Clustering of the organ specific behavior for compound, darkness and

1243

submergence responses identified in A. The log2FC of root- and shoot-

1244

specific genes (red and blue circles in A) is shown. The accessions are

1245

ordered from left to right for each organ and treatment: Cvi-0, Bay-0, Ita-0,

1246

Col(gl), Kas-1, Lp2-6, Ws-2, C24. Yellow indicates up and cyan down

1247

regulation. Genes shown and fold change values are in the supplemental

1248

data, sheet H.

1249

C. Log2FC of the 51 core hypoxia genes. The accessions are ordered as: Cvi-

1250

0, Bay-0, Ita-0, Col(gl), Kas-1, Lp2-6, Ws-2, C24 (left to right for each organ

1251

and treatment). Yellow indicates up and blue down regulation.

represents

genes

that

adj.

show

< 0.05 in at least 1 accession). robust

organ-specific

behaviour

1252 1253

Figure 5. Alternative splicing (AS) and intron retention (IR) upon

1254

compound, darkness and submergence stress.

1255

A. Histograms that shows the magnitude of IR (Calculated as the maximum

1256

difference in IR) upon the treatments for each accession.

1257

B. The number of genes in the shoot (green) and root (gray), that show IR

1258

upon treatment (compound, dark and submergence), and their overlap

1259

between accessions (|log2(obs/exp)| > 1 and Padj. < 0.01).

1260

C. Histograms that shows the magnitude of AS (Calculated as the maximum

1261

difference in AS) upon the treatments for each accession.

1262

D. The number of genes in the shoot (green) and root (gray), that are

1263

alternatively spliced upon treatment (compound, dark and submergence), and

1264

their overlap between accessions (|log2(obs/exp)| > 1 and Padj. < 0.01).


39

1265 1266

Figure 6. The conserved compound, darkness and submergence

1267

induced alternative splicing.

1268

For each variant identifying gene region (VIGR) the deviation from the

1269

expected read counts if no change in variant usage upon treatment would

1270

take place (log2(obs/exp)) is shown for roots and shoots. Yellow indicates

1271

more, and cyan less reads than expected for a specific VIGR. All genes that

1272

are alternatively spliced (|log2(obs/exp)| > 1 and Padj. < 0.01) in five or more

1273

genotypes are shown. For each condition the genotypes are ordered from left

1274

to right: Cvi-0, Bay-0, Ita-0, Col(gl), Kas-1, Lp2-6, Ws-2 and C24.

1275 1276

Figure 7. RT-qPCR validation of a selection of genes alternatively

1277

spliced upon compound, darkness and submergence.

1278

Fold change of transcript abundance in the accessions Cvi-0 and C24 upon

1279

compound, darkness and submergence of six genes that were identified as

1280

alternatively spliced upon the treatment in a conserved manner across

1281

accessions. Primers were used that either amplified all transcript variants (red

1282

lines), or selectively amplify only specific variants (blue and green lines). Root

1283

or shoot tissue was analyzed depending on the organ in which alternative

1284

splicing was identified. Details for each gene are as follows. LYSINE-

1285

KETOGLUTARATE

1286

(AT4G33150, LKR/SDH, root): blue line is AT4G33150.1 and AT4G33150.2;

1287

RHO GUANYL-NUCLEOTIDE EXCHANGE FACTOR 11 (AT1G52240,

1288

ROPGEF11,

1289

DEHYDROGENASE 2 (AT5G07440, GDH2, root): blue is AT5G07440.1 and

1290

AT5G07440.2

1291

ERYTHRONATE-4-P DEHYDROGENASE (AT1G75180, E4PDH, shoot): blue

1292

line is AT1G75180.2 and AT1G75180.3; FRUCTOSE-BISPHOSPHATE

1293

ALDOLASE 1 (AT2G21330, FBA1, shoot): blue line is specific for

1294

AT2G21330.2; PYRUVATE ORTHOPHOSPHATE DIKINASE (AT4G15530,

1295

PPDK, shoot): blue line represents all transcripts, excluding AT4G15530.2

root): and

REDUCTASE/SACCHAROPINE

blue green

line line

is

is

DEHYDROGENASE

AT1G52240.1;

AT5G07440.1

and

GLUTAMATE AT5G07440.3;

1296 1297

Figure 8. Coverage plots of a selection of genes that are alternative

1298

spliced in response to the imposed treatments.


40

1299

Coverage plots were normalized to the maximum read depth, or a percentage

1300

of maximum read depth. Gene models (introns – line; exons – fat box; thin

1301

box – UTR) are shown below ordered from the first model (.1) onwards and

1302

are depicted from left to right in the 5’ to 3’ direction. AL is air light, AD is air

1303

darkness and SD is submergence darkness. Red or green lines indicate the

1304

location of intact protein domains specific to certain transcript isoforms

1305

A. ROPGEF11, ROP (RHO OF PLANTS) GUANINE NUCLEOTIDE

1306

EXCHANGE FACTOR 11, AT1G52240, in Bay-0. Red domain - dynein light

1307

chain domain (PF01221), green domain - PRONE (PF03759)

1308

B. ARR1, ARABIDOPSIS RESPONSE REGULATOR 1, AT3G16857, in Bay-

1309

0.

1310

C.

1311

AT5G09660, in Bay-0. Red domain – NAD binding domain, green domain -

1312

malate dehydrogenase, alpha/beta C-terminal domain

1313

D. PPDK, PYRUVATE ORTHOPHOSPHATE DIKINASE, AT4G15530, in Bay-

1314

0

1315

E. LKR/SDH, LYSINE-KETOGLUTARATE REDUCTASE/SACCHAROPINE

1316

DEHYDROGENASE, AT4G33150, in Bay-0. Red domain – LKR activity,

1317

green domain – SDH activity (Zhu et al. 2002)

1318

F. NPQ1, NON-PHOTOCHEMICAL QUENCHING 1, AT1G08550, in Bay-0.

1319

G. PRIN2, PLASTID REDOX INSENSITIVE 2, AT1G10522 in Bay-0.

1320

H and I. BCA1 and BCA4, BETA CARBONIC ANHYDRASE 1 and 4,

1321

AT3G01500 and AT1G70410, in Lp2-6.

PMDH2,

PEROXISOMAL

NAD-MALATE

DEHYDROGENASE

2,

1322 1323

Figure 9. Flooding tolerance dependent compound, darkness and

1324

submergence responses.

1325

A. Number of DEGs in the shoot and root where the treatment response

1326

depends on the tolerance group (Ptolerance*treatment,

1327

tolerance classification is based on Vashisht et al. (2011). Sensitive

1328

accessions are Cvi-0, Bay-0 and Ita-0 and tolerant accessions are Lp2-6, Ws-

1329

2 and C24.

1330

B. Tolerance specific DEGs (from A) that show distinct responses between

1331

the flooding tolerant (teal circles) and sensitive (yellow circles) accessions and

1332

their deviation from the average treatment response (for root and shoot

adj.

< 0.05). Flooding


41

1333

compound and submergence treatments). Within each panel, left graphs

1334

show DEGs with fold-change higher in the tolerant accessions and the right

1335

shows DEGs with a fold-change higher in the sensitive accessions.

1336 1337

LITERATURE CITED

1338 1339

Amthor, J (2000) The McCree–de Wit–Penning de Vries–Thornley respiration paradigms: 30 years later. Ann Bot 86: 1–20

1340 1341 1342

Angelovici R, Fait A, Fernie AR, and Galili G (2010) A seed high-lysine trait is negatively associated with the TCA cycle and slows down Arabidopsis seed germination. New Phytol 189: 148–159

1343 1344 1345

Antonio C, Papke C, Rocha M, Diab H, Limami AM, Obata T, Fernie AR, and van Dongen, JT (2015) Regulation of primary metabolism in response to low oxygen availability as revealed by carbon and nitrogen isotope redistribution. Plant Physiol: 1–34

1346 1347

Armstrong J and Armstrong W (2001) An overview of the effects of phytotoxins on Phragmites australis in relation to die-back. Aquatic Botany 69: 251–268

1348 1349

Baena-González E, Rolland F, Thevelein JM, and Sheen J (2007) A central integrator of transcription networks in plant stress and energy signaling. Nature 448: 938–942

1350 1351 1352

Banti V, Mafessoni F, Loreti E, Alpi A, and Perata P (2010) The Heat-inducible transcription factor HsfA2 enhances anoxia tolerance in Arabidopsis. Plant Physiol 152: 1471–1483

1353 1354 1355

Barding GA, Béni S, Fukao T, Bailey-Serres J, and Larive CK (2013) Comparison of GCMS and NMR for metabolite profiling of rice subjected to submergence stress. J Prot Res 12: 898–909

1356 1357 1358 1359

Bernal M, Casero D, Singh V, Wilson GT, Grande A, Yang H, Dodani SC, Pellegrini M, Huijser P, Connolly EL, Merchant SS, and Krämer U (2012) Transcriptome sequencing identifies SPL7-regulated copper acquisition genes FRO4/FRO5 and the copper dependence of iron homeostasis in Arabidopsis. Plant Cell 24: 738–761

1360 1361 1362

Branco-Price C, Kaiser KA, Jang CJH, Larive CK, and Bailey-Serres J (2008) Selective mRNA translation coordinates energetic and metabolic adjustments to cellular oxygen deprivation and reoxygenation in Arabidopsis thaliana. Plant J 56: 743–755

1363 1364 1365

Brosché M, Merilo E, Mayer F, Pechter P, Puzõrjova I, Brader G, Kangasjärvi J, and Kollist H (2010) Natural variation in ozone sensitivity among Arabidopsis thaliana accessions and its relation to stomatal conductance. Plant Cell Environ 33: 914–925

1366 1367 1368 1369 1370

Brown NJ, Palmer BG, Stanley S, Hajaji H, Janacek SH, Astley HM, Parsley K, Kajala K, Quick WP, Trenkamp S, Fernie AR, Maurino VG, Hibberd JM (2010) C4 acid decarboxylases required for C4 photosynthesis are active in the mid-vein of the C3 species Arabidopsis thaliana and are important in sugar and amino acid metabolism. Plant J 61: 122–133


42

1371 1372 1373 1374

Campbell MT, Proctor CA, Dou Y, Schmitz AJ, Phansak P, Kruger, GR, Zhang C, Walia H (2015) Genetic and molecular characterization of submergence response identifies Subtol6 as a major submergence tolerance locus in maize PLoS ONE, 10, e0120385 doi:101371/journalpone0120385

1375 1376

Chang CY, Lin WD, and Tu SL (2014) Genome-wide analysis of heat-sensitive alternative splicing in Physcomitrella patens. Plant Physiol 165: 826–840

1377 1378 1379

Chang R, Jang CJH, Branco-Price C, Nghiem P, and Bailey-Serres J (2012) Transient MPK6 activation in response to oxygen deprivation and reoxygenation is mediated by mitochondria and aids seedling survival in Arabidopsis. Plant Mol Biol 78: 109–122

1380 1381 1382

Chen X, Pierik R, Peeters AJM, Poorter H, Visser EJW, Huber H, de Kroon H, and Voesenek LACJ (2010) Endogenous abscisic acid as a key switch for natural variation in flooding-induced shoot elongation. Plant Physiol 154: 969–977

1383 1384 1385

Chen L, Liao B, Qi H, Xie LJ, Huang L, Tan WJ, Zhai N, Yuan LB, Zhou Y, Yu, LJ, Chen QF, Shu W, and Xiao, S (2015) Autophagy contributes to regulation of the hypoxia response during submergence in Arabidopsis thaliana. Autophagy 11: 2233–2246

1386 1387 1388

Christianson JA, Wilson IW, Llewellyn DJ, and Dennis ES (2009) The low-oxygeninduced NAC domain transcription factor ANAC102 affects viability of Arabidopsis seeds following low-oxygen treatment. Plant Physiol 149: 1724–1738

1389 1390 1391

Delgado-Alvarado A, Walker RP, and Leegood RC (2007) Phosphoenolpyruvate carboxykinase in developing pea seeds is associated with tissues involved in solute transport and is nitrogen-responsive. Plant Cell Environ 30: 225–235

1392 1393 1394

Dennis ES, Pogson BJ, and Finnegan EJ (2009) VERNALIZATION INSENSITIVE 3 (VIN3) is required for the response of Arabidopsis thaliana seedlings exposed to low oxygen conditions 2009 Plant J 59:576-587

1395 1396 1397

Dinneny JR, Long TA, Wang JY, Jung JW, Mace D, Pointer S, Barron C, Brady SM, Schiefelbein J, and Benfey PN (2008) Cell identity mediates the response of Arabidopsis roots to abiotic stress. Science 320: 942–945

1398 1399 1400

Eastmond PJ, Astley HM, Parsley K, Aubry S, Williams BP, Menard GN, Craddock CP, Nunes-Nesi A, Fernie AR, and Hibberd JM (2015) Arabidopsis uses two gluconeogenic gateways for organic acids to fuel seedling establishment. Nature Comm 6: 6659

1401 1402

Ellis MH, Dennis ES, and Peacock WJ (1999) Arabidopsis roots and shoots have different mechanisms for hypoxic stress tolerance. Plant Physiol 119: 57–64

1403 1404 1405

Fabre N, Reiter IM, Becuwe-Linka N, Genty B, and Rumeau D (2007) Characterization and expression analysis of genes encoding alpha and beta carbonic anhydrases in Arabidopsis. Plant Cell Environ 30: 617–629

1406 1407 1408

Fukao T, Xu K, Ronald PC, and Bailey-Serres, J (2006) A variable cluster of ethylene rsponse factor-like genes regulates metabolic and developmental acclimation responses to submergence in Rice. Plant Cell 18: 2021–2034

1409 1410 1411

Fukao T, Yeung E, and Bailey-Serres J (2011) The submergence tolerance regulator SUB1A mediates crosstalk between submergence and drought tolerance in Rice. Plant Cell 23: 412–427


43

1412 1413

Gaillochet, C, Daum, G, and Lohmann, JU (2015) O Cell, Where Art Thou? The mechanisms of shoot meristem patterning Curr Opin Plant Biol 23: 91–97

1414 1415 1416

Gibbs DJ, Lee SC, Isa NM, Gramuglia S, Fukao T, Bassel GW, Correia CS, Corbineau F, Theodoulou FL, Bailey-Serres J, and Holdsworth MJ (2011) Homeostatic response to hypoxia is regulated by the N-end rule pathway in plants. Nature 479: 415–418

1417 1418 1419 1420 1421

Gibbs DJ, Isa NM, Movahedi M, Lozano-Juste J, Mendiondo GM, Berckhan S, Marin-de la Rosa N, Conde JV, Correia CS, Pearce SP, Bassel GW, Hamali B, Talloji P, Tome DFA, Coego A, Beynon J, Alabadi D, Bachmair A, Leon J, Gray JE, Theodoulou FL, Holdsworth MJ (2014) Nitric oxide sensing in plants Is mediated by proteolytic control of Group VII ERF transcription factors. Mol Cell 53: 369-379

1422 1423 1424

Giuntoli B, Lee SC, Licausi F, Kosmacz M, Oosumi T, van Dongen JT, Bailey-Serres J, and Perata P (2014) A trihelix DNA binding protein counterbalances hypoxia-responsive transcriptional activation in Arabidopsis. PLoS Biol 12: e1001950

1425 1426

Gommers CMM, Visser EJW, St Onge KR, Voesenek LACJ, and Pierik R (2013) Shade tolerance: when growing tall is not an option. Trends Plant Sci 18: 65–71

1427 1428 1429

Gonzali, S, Loreti, E, Cardarelli, F, Novi, G, Parlanti, S, Pucciariello, C, Bassolino, L, Banti, V, Licausi, F, and Perata, P (2015) Universal stress protein HRU1 mediates ROS homeostasis under anoxia Nature Plants 1: 15151

1430 1431 1432

Graf A, Schlereth A, Stitt M, and Smith AM (2010) Circadian control of carbohydrate availability for growth in Arabidopsis plants at night. Proc Natl Acad Sci USA 107: 9458– 9463

1433 1434 1435 1436

Hartmann L, Pedrotti L, Weiste C, Fekete A, Schierstaedt J, Gottler J, Kempa S, Krischke M, Dietrich K, Mueller MJ, Vicente-Carbajosa J, Hanson J, Droge-Laser W (2015) Crosstalk between Two bZIP signaling pathways orchestrates salt-induced metabolic reprogramming in Arabidopsis roots. Plant Cell 27: 2244–2260

1437 1438 1439 1440

Hattori Y, Nagai K, Furukawa S, Song XJ, Kawano R, Sakakibara H, Wu J, Matsumoto T, Yoshimura A, Kitano H, Matsuoka M, Mori H, Ashikari M (2009) The ethylene response factors SNORKEL1 and SNORKEL2 allow rice to adapt to deep water. Nature 460: 1026–1030

1441 1442

Hibberd JM and Quick WP (2002) Characteristics of C4 photosynthesis in stems and petioles of C3 flowering plants. Nature 415: 451–454

1443 1444

Hildebrandt TM, Nesi AN, Araújo WL, and Braun HP (2015) Amino acid catabolism in plants. Molecular Plant: 1–35

1445 1446 1447

Hsu FC, Chou MY, Peng HP, Chou SJ, and Shih MC (2011) Insights into hypoxic systemic responses based on analyses of transcriptional regulation in Arabidopsis. PLoS ONE 6: e28888

1448 1449

Hsu PY and Harmer SL (2014) Wheels within wheels: the plant circadian system. Trends in Plant Sci 19: 240–249

1450 1451

Huang S, Colmer TD, and Millar AH (2008) Does anoxia tolerance involve altering the energy currency towards PPi? Trend Plant Sci 13: 221–227


44

1452 1453

Huijser P and Schmid M (2011) The control of developmental phase transitions in plants. Development 138: 4117–4129

1454 1455 1456

Usami, T, Mochizuki, N, Kondo, M, Nishimura, M, and Nagatani, A (2004) Cryptochromes and phytochromes synergistically regulate Arabidopsis root greening under blue light Plant Cell Physiol 45:1798–1808

1457 1458

Jain A, Wilson GT, and Connolly EL (2014) The diverse roles of FRO family metalloreductases in iron and copper homeostasis. Front Plant Sci 5: 100

1459 1460

Li J, Li G, Wang H, Deng XW (2011) Phytochrome Signaling Mechanisms. The Arabidopsis Book, American Society of Plant Biologists. 9:e0148

1461 1462 1463

Juntawong P, Girke T, Bazin J, and Bailey-Serres J (2014) Translational dynamics revealed by genome-wide profiling of ribosome footprints in Arabidopsis. Proc Natl Acad Sci USA 111: E203–12

1464 1465

Kazan K (2003) Alternative splicing and proteome diversity in plants: the tip of the iceberg has just emerged. Trends Plant Sci 8: 468–471

1466

Kieber JJ and Schaller GE (2014) Cytokinins. The Arabidopsis Book 12: e0168

1467 1468 1469

Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, and Salzberg SL (2013) TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 14: R36

1470 1471 1472

Kobayashi K, Baba S, Obayashi T, Sato M, Toyooka K, Keranen M, Aro EM, Fukaki H, Ohta H, Sugimoto K, and Masuda T (2012) Regulation of root greening by light and Auxin/Cytokinin signaling in Arabidopsis. Plant Cell 24: 1081–1095

1473 1474 1475

Kreuzwieser J, Hauberg J, Howell KA, Carroll A, Rennenberg H, Millar AH, and Whelan J (2009) Differential response of gray poplar leaves and roots underpins stress adaptation during hypoxia. Plant Physiol 149: 461–473

1476 1477

Langfelder P and Horvath S (2008) WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9: 559

1478 1479 1480

Lawrence M, Huber W, Pagès H, Aboyoun P, Carlson M, Gentleman R, Morgan MT, and Carey VJ (2013) Software for computing and annotating genomic ranges. PLoS Comput Biol 9: e1003118

1481 1482 1483

Lee SC, Mustroph A, Sasidharan R, Vashisht D, Pedersen O, Oosumi T, Voesenek LACJ, and Bailey-Serres J (2011) Molecular characterization of the submergence response of the Arabidopsis thaliana ecotype Columbia. New Phytol 190: 457–471

1484 1485 1486

Lee KW, Chen PW, Lu CA, Chen S, Ho THD, and Yu SM (2009) Coordinated responses to oxygen and sugar deficiency allow rice seedlings to tolerate flooding. Sci Signaling 2: ra61–ra61

1487 1488 1489

Licausi F, Kosmacz M, Weits DA, Giuntoli B, Giorgi FM, Voesenek LACJ, Perata P, and van Dongen JT (2011) Oxygen sensing in plants is mediated by an N-end rule pathway for protein destabilization. Nature 479: 419–422

1490

Ljung K, Nemhauser JL, and Perata, P (2015) New mechanistic links between sugar and


45

1491

hormone signalling networks. Curr Opin Plant Biol 25: 130–137

1492 1493 1494

Lokdarshi A, Conner WC, McClintock C, Li T, and Roberts D (2015) Arabidopsis CML38, a calcium sensor that localizes to ribonucleoprotein complexes under hypoxia stress. Plant Physiol 170: 1046-1059

1495 1496 1497 1498

Magneschi L, Kudahettige RL, Alpi A, and Perata P (2009) Comparative analysis of anoxic coleoptile elongation in rice varieties: relationship between coleoptile length and carbohydrate levels, fermentative metabolism and anaerobic gene expression. Plant Biol 11: 561–573

1499 1500 1501 1502

Malone S, Chen ZH, Bahrami AR, Walker RP, Gray JE, and Leegood RC (2007) Phosphoenolpyruvate Carboxykinase in Arabidopsis: Changes in gene expression, protein and activity during vegetative and reproductive development. Plant Cell Physiol 48: 441–450

1503 1504

Mommer L and Visser EJW (2005) Underwater photosynthesis in flooded terrestrial plants: A matter of leaf plasticity. Ann Bot 96: 581–589

1505 1506 1507

Morgan M, Pages H, Obenchain V, and Haydon N (2013) Rsamtools: Binary alignment (BAM), FASTA, variant call (BCF) and tabix file import. R package version 1.22.0 http://bioconductor.org/packages/release/bioc/html/Rsamtools.html.

1508 1509

Mugnai S, Azzarello E, Baluska F, and Mancuso S (2012) Local root apex hypoxia induces NO-mediated hypoxic acclimation of the entire root. Plant Cell Physiol 53: 912–920

1510 1511 1512

Mustroph A, Barding GA, Kaiser KA, Larive CK, and Bailey-Serres, J (2014a) Characterization of distinct root and shoot responses to low-oxygen stress in Arabidopsis with a focus on primary C- and N-metabolism. Plant Cell Environ 37: 2366–2380

1513 1514 1515

Mustroph, A, Hess, N, and Sasidharan, R (2014b) Hypoxic energy metabolism and PPi as an alternative energy currency. In Plant Cell Monographs, Plant Cell Monographs (Springer Vienna: Vienna), pp 165–184

1516 1517 1518 1519

Mustroph A, Lee SC, Oosumi T, Zanetti ME, Yang H, Ma K, Yaghoubi-Masihi A, Fukao T, and Bailey-Serres J (2010) Cross-Kingdom comparison of transcriptomic adjustments to low-oxygen stress highlights conserved and plant-specific responses. Plant Physiol 152: 1484–1500

1520 1521 1522 1523

Mustroph A, Zanetti ME, Jang CJH, Holtan HE, Repetti PP, Galbraith DW, Girke T, and Bailey-Serres J (2009) Profiling translatomes of discrete cell populations resolves altered cellular priorities during hypoxia in Arabidopsis. Proc Natl Acad Sci USA 106: 18843–18848

1524 1525 1526

Müller J, Toev T, Heisters M, Teller J, Moore KL, Hause G, Dinesh DC, Bürstenbinder K, and Abel S (2015) Iron-dependent callose deposition adjusts root meristem maintenance to phosphate availability. Dev Cell 33: 216–230

1527 1528 1529

Narsai, R, Edwards, JM, Roberts, TH, Whelan, J, Joss, GH, and Atwell, BJ (2015) Mechanisms of growth and patterns of gene expression in oxygen-deprived rice coleoptiles Plant J 82: 25–40

1530 1531

Narsai, R, Howell, KA, Carroll, A, Ivanova, A, Millar, AH, and Whelan, J (2009) Defining core metabolic and transcriptomic responses to oxygen availability in rice embryos and


46

1532

young seedlings Plant Physiol 151: 306–322

1533 1534 1535 1536

Pal SK, Liput M, Piques M, Ishihara H, Obata T, Martins MCM, Sulpice R, Van Dongen JT, Fernie A, Yadav UP, Lunn JE, Usadel B, and Stitt M (2013) Diurnal changes of polysome loading track sucrose content in the rosette of wild-type Arabidopsis and the starchless pgm mutant. Plant Physiol 162: 1246–1265

1537 1538 1539

Parsley K and Hibberd JM (2006) The Arabidopsis PPDK gene is transcribed from two promoters to produce differentially expressed transcripts responsible for cytosolic and plastidic proteins. Plant Mol Biol 62: 339–349

1540 1541 1542 1543

Penfield S, Rylott EL, Gilday AD, Graham S, Larson TR, and Graham IA (2004) Reserve mobilization in the Arabidopsis endosperm fuels hypocotyl elongation in the dark, is independent of abscisic acid, and requires PHOSPHOENOLPYRUVATE CARBOXYKINASE1. Plant Cell 16: 2705–2718

1544 1545 1546

Pracharoenwattana I, Cornah JE, and Smith SM (2007) Arabidopsis peroxisomal malate dehydrogenase functions in beta-oxidation but not in the glyoxylate cycle. Plant J 50: 381–390

1547 1548 1549

Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, and Smyth GK (2015) Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43: e47–e47

1550 1551

Robinson MD and Oshlack A (2010) A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol 11: R25

1552 1553 1554 1555

Rocha M, Licausi F, Araujo WL, Nunes-Nesi A, Sodek L, Fernie AR, and van Dongen JT (2010) Glycolysis and the tricarboxylic acid cycle are linked by alanine aminotransferase during hypoxia induced by waterlogging of Lotus japonicas. Plant Physiol 152: 1501– 1513

1556 1557 1558 1559

Sasidharan R, Mustroph A, Boonman A, Akman M, Ammerlaan AMH, Breit T, Schranz ME, Voesenek LACJ, and van Tienderen PH (2013) Root transcript profiling of two Rorippa species reveals gene clusters associated with extreme submergence tolerance. Plant Physiol 163: 1085–1086

1560 1561

Schroder F, Lisso J, and Mussig C (2011) EXORDIUM-LIKE1 promotes growth during low carbon availability in Arabidopsis. Plant Physiol 156: 1620–1630

1562 1563

Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell 2: 1027– 1038

1564 1565 1566

Shin DH, Cho MH, Kim TL, Yoo J, Kim JI, Han YJ, Song PS, Jeon JS, Bhoo SH, and Hahn TR (2010) A small GTPase activator protein interacts with cytoplasmic phytochromes in regulating root development. J Biol Chem 285: 32151–32159

1567 1568

Smeekens S, Ma J, Hanson J, and Rolland, F (2010) Sugar signals and molecular networks controlling plant growth. Curr Opin Plant Biol 13: 274–279

1569 1570 1571

Sorenson R and Bailey-Serres, J (2014) Selective mRNA sequestration by OLIGOURIDYLATE-BINDING PROTEIN 1 contributes to translational control during hypoxia in Arabidopsis. Proc Natl Acad Sci USA 111: 2373–2378


47

1572 1573 1574

Svistoonoff S, Creff A, Reymond M, Sigoillot-Claude C, Ricaud L, Blanchet A, Nussaume L, and Desnos T (2007) Root tip contact with low-phosphate media reprograms plant root architecture. Nat Genet 39: 792–796

1575 1576 1577 1578

Tang G, Zhu X, Gakiere B, Levanony H, Kahana A, and Galili G (2002) The bifunctional LKR/SDH locus of plants also encodes a highly active monofunctional lysineketoglutarate reductase using a polyadenylation signal located within an intron. Plant Physiol 130: 147–154

1579 1580 1581 1582

Taylor L, Nunes-Nesi A, Parsley K, Leiss A, Leach G, Coates S, Wingler A, Fernie AR, and Hibberd JM (2010) Cytosolic pyruvate,orthophosphate dikinase functions in nitrogen remobilization during leaf senescence and limits individual seed growth and nitrogen content. Plant J 62: 641–652

1583 1584 1585

Usadel B, Blasing OE, Gibon Y, Retzlaff K, Hohne M, Gunther M, and Stitt M (2008) Global transcript levels respond to small changes of the carbon status during progressive exhaustion of carbohydrates in Arabidopsis rosettes. Plant Physiol 146: 1834–1861

1586 1587

van Dongen JT, and Licausi F (2015) Oxygen sensing and signaling. Ann Rev Plant Biol 66: 345–367

1588 1589 1590

Van Eck WHJM, van de Steeg HM, Blom CWPM, and de Kroon H (2004) Is tolerance to summer flooding correlated with distribution patterns in river floodplains? A comparative study of 20 terrestrial grassland species. Oikos 107: 393–405

1591 1592 1593 1594

van Veen H, Akman M, Jamar DCL, Vreugdenhil D, Kooiker M, van Tienderen P, Voesenek LACJ, Schranz ME, and Sasidharan R (2014a) Group VII Ethylene Response Factor diversification and regulation in four species from flood-prone environments. Plant Cell Environ 37:2421-32

1595 1596 1597 1598

van Veen H, Mustroph A, Barding GA, Vergeer-van Eijk M, Welschen-Evertman RAM, Pedersen O, Visser EJW, Larive CK, Pierik R, Bailey-Serres J, Voesenek LACJ, and Sasidharan R (2013) Two Rumex species from contrasting hydrological niches regulate flooding tolerance through distinct mechanisms. Plant Cell 25: 4691–4707

1599 1600 1601

van Veen H, Vashisht D, Voesenek LACJ, and Sasidharan R (2014b) Different survival strategies amongst plants to cope with underwater conditions. In Plant Cell Monographs, Plant Cell Monographs (Springer Vienna: Vienna), pp 329–349

1602 1603 1604 1605

Vashisht D, Hesselink A, Pierik R, Ammerlaan JMH, Bailey-Serres J, Visser EJW, Pedersen O, van Zanten M, Vreugdenhil D, Jamar DCL, Voesenek LACJ, and Sasidharan R (2011) Natural variation of submergence tolerance among Arabidopsis thaliana accessions. New Phytol 190: 299–310

1606 1607

Vervuren P, Blom C, and de Kroon H (2003) Extreme flooding events on the Rhine and the survival and distribution of riparian plant species. J Ecology 91: 135–146

1608 1609

Voesenek LACJ and Bailey-Serres J (2015) Flood adaptive traits and processes: an overview. New Phytol 206: 57–73

1610 1611

Voesenek LACJ and Sasidharan R (2013) Ethylene--and oxygen signalling--drive plant survival during flooding. Plant Biol (Stuttg) 15: 426–435

1612

Voesenek LACJ, Rijnders JHGM, Peeters AJM, van de Steeg HM, and de Kroon H


48

1613 1614

(2004) Plant hormones regulate fast shoot elongation under water: from genes to communities. Ecology 85: 16–27

1615 1616

Voesenek LACJ, van Veen H, and Sasidharan R (2014) Learning from nature: the use of non-model species to identify novel acclimations to flooding stress. AoB Plants 6

1617 1618 1619 1620

Watanabe M, Balazadeh S, Tohge T, Erban A, Giavalisco P, Kopka J, Mueller-Roeber B, Fernie AR, and Hoefgen R (2013) Comprehensive dissection of spatiotemporal metabolic shifts in primary, secondary, and lipid metabolism during developmental senescence in Arabidopsis. Plant Physiol 62:1290-1310

1621 1622 1623

Weits DA, Giuntoli B, Kosmacz M, Parlanti S, Hubberten HM, Riegler H, Hoefgen R, Perata P, van Dongen JT, and Licausi F (2014) Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nature Comm 5: 3425

1624 1625 1626

Xu K, Xu X, Fukao T, Canlas P, Maghirang-Rodriguez R, Heuer S, et al (2006) Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature 442: 705–708

1627 1628

Young MD, Wakefield MJ, Smyth GK, and Oshlack A (2010) Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol 11: R14

1629 1630 1631 1632

Zabalza A, van Dongen JT, Froehlich A, Oliver SN, Faix B, Gupta KJ, Schmalzlin E, Igal M, Orcaray L, Royuela M, and Geigenberger P (2008) Regulation of respiration and fermentation to control the plant internal oxygen concentration. Plant Physiol 149: 1087– 1098

1633 1634 1635 1636

Zhu X, Tang G, and Galili G (2002) The activity of the Arabidopsis bifunctional lysineketoglutarate reductase/saccharopine dehydrogenase enzyme of lysine catabolism is regulated by functional interaction between its two enzyme domains. J Biol Chem 277: 49655–49661

1637


49

A

air light (AL)

air dark. (AD)

subm. dark. (SD)

submergence

darkness

compound ZT0

ZT2

harvest ZT6

ZT9

MDS y-axis

B

2

C24

1.5 0

2

4

C24

Kas-1

2

1

0

root Lp2-6

Lp2-6 Col(gl) Ws-2 Col(gl) Kas-1 Bay-0

Ws-2

Kas-1

Bay-0

1

Ita-0

Ita-0

1

C24

C24

Cvi-0

2

Kas-1

Kas-1

3

0

Ita-0 Col(gl) Col(gl) Cvi-0 Bay-0 Cvi-0 Bay-0 C24

1.0

2

Ws-2

1

Cvi-0

2

2

1

0

1

MDS X axis

shoot

1500 1000

500 1000 1500

root

1000 500 0 500 1000 1500

compound darkness submergence

D response to stress of individual accessions (log2FC)

upreg. downreg.

0

Lp2-6

Ws-2 Ita-0

Ita-0 Cvi-0 Col(gl) Bay-0

0.5

4

1500

upreg.

0.5

2

Lp2-6

Lp2-6 Ws-2

1.0

1

0

downreg.

shoot

1.5

0

500

no. differentially expressed genes

shoot - AL shoot - AD shoot - SD

1

AL AD &SD

C

root - AL root - AD root - SD

24 C 2 sW 6 2Lp 1 sKa 1) l (g ol C -0 Ita 0 yBa -0 vi

C

average response to stress accros accesion (log2FC)

Figure 1. Transcriptional responses to compound, darkness and submergence stress in eight Arabidopsis accessions. A. Schematic representation of the experimental setup, light cycle and treatments used. Arabidopsis seedlings were grown until the 10-leaf stage (9h photoperiod (Zeitgeber time (ZT): ZT0-ZT9). Plants then remained in control (air light; AL) conditions, or were transferred to submerged (dark) (SD) or dark (air dark; AD) conditions 2 h after photoperiod initiation (ZT2). Shoot and root material from Arabidopsis seedlings exposed to AL, AD and SD conditions for 4h were harvested at ZT6, and used for mRNAseq. Black bars indicate darkness. Double ended arrows indicate datasets compared to deduce the compound stress (AL vs. SD), darkness (AL vs. AD) and submergence (AD vs. SD) differentially expressed genes. B. Multidimensional scaling (MDS) of all mRNAseq libraries, and of the shoot or root only, with distances based on the pairwise top 500 genes differing in fold change. AL = air light, AD = air darkness, SD = submergence darkness. C. Number of differentially expressed genes (DEGs) in shoots and roots (Padj < 0.05) in response to the compound, darkness and submergence stress. D. The compound, darkness and submergence response of individual accessions Downloadedresponse from www.plantphysiol.org on accessions. May 16, 2016 - Published www.plant.org compared to the weighted average of all eight DEGsby(P < 0.05) adj. Copyright © 2016 American Society of Plant Biologists. All rights reserved. counted in 1C are depicted in the corresponding graph.

comp dark subm comp dark subm


shoot

root comp dark subm comp dark subm

root


shoot

GO:0010200 response to chitin

GO:0019252 starch biosynthetic process

GO:0009753 response to jasmonic acid

GO:0000023 maltose metabolic process

GO:0009862 systemic acquired resistance, SA mediated signaling

GO:0009744 response to sucrose

GO:0002679 respiratory burst involved in defense response

GO:0009750 response to fructose

GO:0010583 response to cyclopentenone

GO:0016157 sucrose synthase activity

GO:0009543 chloroplast thylakoid lumen

GO:0003825 trehalose-phosphate synthase activity

GO:0009535 chloroplast thylakoid membrane

GO:0010264 myo-inositol hexakisphosphate biosynthetic process

GO:0010224 response to UV-B

GO:0043085 positive regulation of catalytic activity

GO:0009646 response to absence of light

GO:0042546 cell wall biogenesis

GO:0009411 response to UV

GO:0009831 plant-type cell wall modification

GO:0048573 photoperiodism, flowering

GO:0009828 plant-type cell wall loosening

GO:0009718 anthocyanin-containing compound biosynthetic process

GO:0044036 cell wall macromolecule metabolic process

GO:0009813 flavonoid biosynthetic process

GO:0016759 cellulose synthase activity

GO:0080003 thalianol metabolic process

GO:0033946 xyloglucan-specific endo-beta-1,4-glucanase activity

GO:0009693 ethylene biosynthetic process

GO:0009807 lignan biosynthetic process

GO:0009723 response to ethylene

GO:0008194 UDP-glycosyltransferase activity

GO:0071369 cellular response to ethylene stimulus

GO:0016757 transferase activity, transferring glycosyl groups

GO:0009684 indoleacetic acid biosynthetic process

GO:0016740 transferase activity

GO:0009733 response to auxin

GO:0001666 response to hypoxia

GO:0009738 abscisic acid-activated signaling pathway

GO:0071456 cellular response to hypoxia

GO:0009408 response to heat

GO:0009061 anaerobic respiration

GO:0080167 response to karrikin

GO:0000103 sulfate assimilation

GO:0071398 cellular response to fatty acid

GO:0080103 4-methylthiopropyl glucosinolate S-oxygenase activity

GO:0035556 intracellular signal transduction

GO:0019761 glucosinolate biosynthetic process

GO:0003700 sequence-specific DNA binding TF activity

GO:0010439 regulation of glucosinolate biosynthetic process

GO:0016556 mRNA modification

GO:0010438 cellular response to sulfur starvation

GO:0010310 regulation of hydrogen peroxide metabolic process

GO:0009973 adenylyl-sulfate reductase activity

GO:0019825 oxygen binding

GO:0015706 nitrate transport

GO:0071732 cellular response to nitric oxide

GO:0015112 nitrate transmembrane transporter activity

GO:0004497 monooxygenase activity

GO:0010167 response to nitrate

GO:0006949 syncytium formation

GO:0006865 amino acid transport

GO:0010089 xylem development

GO:0006552 leucine catabolic process GO:0016020 membrane

-log10(Padj.)

GO:0042538 hyperosmotic salinity response GO:0071281 cellular response to iron ion GO:0030418 nicotianamine biosynthetic process

0

5

overrepresentation amongst: upregulated genes downregulated genes

GO:0030410 nicotianamine synthase activity

Figure 2. Gene Ontology terms overrepresented in the conserved compound, darkness and submergence responses of the eight accessions. Overrepresentation was determined for genes where the average response across accessions was |mean Log2FC| >1.6 and Padj. < 0.01, and where variation in the response between accessions was absent (Paccession*treatment, adj. > 0.1). GO terms with a Padj. < 0.01 are shown.


R01

C

scaled RPKM values

average compound response (log2FC)

A

air light

acetyl-CoA metabolism glycolysis / gluconeogenesis brassinosteroid biosynthesis ER stress / golgi organization

air dark

subm dark

R02

photoperiod sugar transport and responses nitrate transport and responses fatty acid beta oxidation IPP/MEP pathway glucosinolate metabolism chloroplast organization jasmonic acid responses

average dark response (log2FC) air light

air dark

subm dark

C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0 C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0 C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0

B

D

S01

photoperiod, circadian clock peroxisome protein import fatty acid beta oxidation nucleotide metabolism sugar mediated signalling

root shoot

scaled RPKM values

no. genes

6000

4000

2000

0

air light

air dark

subm dark

S02

unplaced S15 S14 S13 S12 S11 S10 S09 S08 S07 S06 S05 S04 S03 S02 S01

unplaced R08 R07 R06 R05 R04 R03 R02 R01

starch and maltose metabolism IPP/MEP pathway glucosinolate metabolism carotenoid biosynthesis meristem & cytokinesis chloroplast organization cell wall organization

air light

air dark

subm dark

C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0 C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0 C24 Ws-2 Lp2-6 Kas-1 Col(gl) Bay-0 Ita-O Cvi-0

Figure 3 Cumulative effects of darkness and submergence and WGCNA clustering. A. Comparison of the compound and dark responses of roots and shoots. Mean responses to treatments are plotted for both the x and y variable. Responses for individual accessions can be found in Supplemental Figure 1. The black dotted line represents y = x, the blue solid line is the regression of the data for the root (y = -0.02 + 0.93*x, r2 = 0.88) and shoot (y = -0.06 + 1.23*x, r2 = 0.84). B. Gene co-expression modules and their sizes as identified by a Weighted Gene Co-expression Network Analysis (WGCNA). Gene-modules show similar expression patterns across the three treatments and eight accessions. Roots (R) and shoots (S) were analyzed separately. Not all genes included in the analysis could be placed in a module of co-expressed genes and these were unplaced. C and D. Mean and variation centered RPKM values of the largest two root (C) and shoot (D) co-expression modules identified by WGCNA. Top 12% of the genes with the highest module-membership-score are shown. The blue and red lines reflect the trends of the gene with the strongest positive and negative correlation to the mean module behavior. The remaining modules are visualized in Supplemental Figure 4. To the right of each module are representative terms related to the identified enriched GO terms. The complete GO analysis is in the Supplemental Data Sheet F and G. Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

mean response in the shoot (log2FC)

A

8

compound

8

darkness

8

4

4

4

0

0

0

4

-4

4

8

-8

8

8

4

0

4

8

-8

-4

0

4

8

8

mean response in the root (log2FC)

subm

Dark

Compound

B

comp. S R

dark. S R

subm. S R

C

compound root shoot

darkness root shoot

submergence

4

0

4

8

variable organ*treatment interaction conserved organ*treatment interaction shoot specific response root specific response core hypoxia genes

submergence root shoot AT4G27450 AT1G33055 AT1G19530 AT4G39675 AT5G61440 AT3G61060 AT2G29870 AT3G43190 AT1G26270 AT1G55810 AT1G72330 AT5G42200 AT3G17860 AT5G26200 AT1G74940 AT4G32840 AT1G63090 AT5G54960 AT1G17290 AT4G22780 AT5G44730 AT5G45340 AT5G02200 AT5G58070 AT1G72940 AT3G23170 AT5G47060 AT5G47910 AT4G17670 AT2G34390 AT5G62520 AT3G27220 AT2G19590 AT4G33560 AT1G35140 AT1G76650 AT3G23150 AT5G15120 AT3G02550 AT5G39890 AT2G47520 AT3G10040 AT1G77120 AT4G24110 AT4G10270 AT1G43800 AT4G33070 AT5G66985 AT2G16060 AT5G10040 AT2G17850

Aluminium induced protein with YGL and LRDR motifs Unknown protein Unknown protein Unknown protein Aypical CYS HIS rich thioredoxin 5 Phloem protein 2-A13 Aquaporin-like superfamily protein Sucrose synthase 4 Phosphatidylinositol 3- and 4-kinase family protein Uridine kinase-like 3 Alanine aminotransferase 2 RING/U-box superfamily protein Jasmonate insensitive 3 Mitochondrial substrate carrier family protein Protein of unknown function (DUF581) Phosphofructokinase 6 Phloem protein 2-A11 Pyruvate decarboxylase 2 Alanine aminotransferase ACT domain repeat 7 Haloacid dehalogenase-like hydrolase (HAD) superfamily protein CYP707A3, ABA 8'-hydroxylase Far-red-elongated hypocotyl1-like Temperature-induced lipocalin Toll-Interleukin-Resistance (TIR) domain-containing protein Unknown protein Protein of unknown function (DUF581) Respiratory burst oxidase homologue D Protein of unknown function (DUF581) NOD26-like intrinsic protein 2;;1 Similar to RCD one 5 Galactose oxidase/kelch repeat superfamily protein ACC oxidase 1 Wound-responsive family protein Exordium like 1 Calmodulin like 38 Ethylene response 2 Plant cysteine oxidase 1 LOB domain-containing protein 41 Plant cysteine oxidase 2 Hypoxia Responsive ERF 2 Hypoxia Response Attenuator alcohol dehydrogenase 1 Unknown protein Wound-responsive family protein Floral transition at the mersitem 1 Pyruvate decarboxylase 1 Unknown protein Hemoglobin 1 Unknown protein Rhodanese/Cell cycle control phosphatase superfamily protein

Figure 4. Organ-specific transcriptome reconfiguration and low oxygen responsive genes. A. Scatterplot showing the mean response of DEGs which show organ-dependent regulation by the treatment. Grey genes are those with a variable organ specific response (Porgan*treatment, adj. < 0.05 in at least 1 accession). Black represents genes that show robust organ-specific behaviour (Porgan*treatment, adj. < 0.05 in 6, 7 or 8 accessions). Blue (shoot-specific genes) are genes that have a robust organ*treatment (black) but also a robust treatment response in the shoot (Padj. < 0.05 in 6, 7, or 8 accessions) but not a robust treatment response in the same direction in the root (Padj. < 0.05 in 6, 7, or 8 accessions). Red circles (root-specific genes) are genes similar to blue but in roots. Green dots depict the cell-type-independent hypoxia-responsive genes (51 core hypoxia gene set) as defined by Mustroph et al. (2009). B. Clustering of the organ specific behavior for compound, darkness and submergence responses identified in A. The log2FC of root- and shoot-specific genes (red and blue circles in A) is shown. The accessions are ordered from left to right for each organ and treatment: Cvi-0, Bay-0, Ita-0, Col(gl), Kas-1, Lp2-6, Ws-2, C24. Yellow indicates up and cyan down regulation. Genes shown and fold change values are in the supplemental data, sheet H. C. Log2FC of the 51 core hypoxia genes. The accessions are ordered as: Cvi-0, Bay-0, Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Ita-0, Col(gl), Kas-1, Lp2-6, Ws-2, C24 (left to right for each organ and treatment). Yellow Copyright © 2016 American Society of Plant Biologists. All rights reserved. indicates up and blue down regulation.

1200

shoot

no. genes

1600 1200

800

Cvi-0 Bay-0 Ita-0 Col(gl) Kas-1 Lp2-6 Ws-2 C24

800 400 400 0

0

root shoot

1600 1200 800 400 0

8a

cc

8a

6a

7a

cc

7a

cc

6a

5a

cc

5a

4a

cc

4a

3a

cc

3a

2a

cc

2a

cc

1a

0 0 0 1 1 1 > 1 .25 .25-0 .5-0. .75-1 -1.25 .25-1 .5-1. .75-2 2 .5 75 .5 75

0-0

1a

0 0 0 1 1 1 > 1 .25 .25-0 .5-0. .75-1 -1.25 .25-1 .5-1. .75-2 2 .5 75 .5 75

0-0

B no. genes with IR

root

A

magnitude of intron retainment (log2FC)

C

2000

root

shoot

D

1500

no. genes with AS

no. genes

1500 1000 1000 500

500 0

0-0

0

0-0

80 40 0

cc

cc

cc

cc

cc

cc

cc

0 0 0 1 1 1 > 1 .25 .25-0 .5-0. .75-1 -1.25 .25-1 .5-1. .75-2 2 .5 75 .5 75 magnitude of alternative splicing (log2FC)

120

cc

0 0 0 1 1 1 > 1 .25 .25-0 .5-0. .75-1 -1.25 .25-1 .5-1. .75-2 2 .5 75 .5 75

2400 2300

Figure 5. Alternative splicing (AS) an intron retention (IR) upon compound, darkness and submergence stress. A. Histograms that shows the magnitude of IR (Calculated as the maximum difference in IR) upon the treatments for each accession. B. The number of genes in the shoot (green) and root (gray), that show IR upon treatment (compound, dark and submergence), and their overlap between accessions (|log2(obs/exp)| > 1 and Padj. < 0.01). C. Histograms that shows the magnitude of AS (Calculated as the maximum difference in AS) upon the treatments for each accession. D. The number of genes in the shoot (green) and root (gray), that are alternatively spliced upon treatment (compound, dark and submergence), and their overlap between accessions (|log2(obs/exp)| > 1 and Padj. < 0.01).


air light

root

air dark

subm dark

air light AT4G33150 lysine-ketoglutarate reductase/ saccharopine dehydrogenase bifunctional enzyme (LKR/SDH)

shoot air dark

subm dark AT3G49430 SER/ARG-rich protein 34A (SRP34A) AT1G75180 Erythronate-4-phosphate dehydrogenase AT1G12440 A20/AN1-like zinc finger AT3G16857 response regulator 1 (ARR1)

AT2G35840 Sucrose-6P-phosphohydrolase (SPP) AT3G16857 response regulator 1 (ARR1) AT5G03360 DC1 domain-containing protein AT1G51690 protein phosphatase 2A 55 kDa regulatory subunit AT1G58190 receptor like protein 9 (RLP9) AT1G54100 aldehyde dehydrogenase 7B4 AT2G41430 dehydration-induced protein (ERD15) AT1G12440 A20/AN1-like zinc finger family protein AT3G15010 RNA-binding (RRM/RBD/RNP motifs) AT1G72370 40s ribosomal protein SA AT3G15353 metallothionein 3 AT1G52240 RHO guanyl-nucleotide exchange factor 11 (ROPGEF11)

AT2G30600 BTB/POZ domain-containing protein

AT4G28300; Protein of unknown function (DUF1421) AT3G61870; unknown protein AT2G22990 sinapoylglucose 1 (SNG1)

AT1G18390 Leaf Rust 10 locus RLK-Like 1.1 (LRK10L1) AT5G02600 Nuclear-Enriched Phloem Companion Cell Gene 6 (NPCC6) AT1G08550 non-photochemical quenching 1 (NPQ1) AT1G70410 beta carbonic anhydrase 4 (BCA4) AT1G10522; plastid redox insensitive 2 (PRIN2) AT1G54100 aldehyde dehydrogenase 7B4 (ALDH7B4) AT2G21330 fructose-bisphosphate aldolase 1 (FBA1) AT1G13930 Involved in response to salt stress AT1G16880 Act Domain Repeats 11 (ACT11)

AT2G16365 F-box family protein AT2G41640; Glycosyltransferase family 61 protein AT1G60000 RNA-binding (RRM/RBD/RNP motifs) AT1G80380 glycerate kinase (C2 cycle)

AT5G07440 glutamate dehydrogenase 2 (GDH2) AT5G09660 peroxisomal NAD-malate dehydrogenase 2 (PMDH2)

-2.0

log2(obs/exp)

2.0

AT1G44575; non-photochemical quenching 4 (NPQ4) AT5G05000 translocon at outer envelope membrane of chloroplasts 34 AT2G26500 cytochrome b6f complex subunit (petM), putative AT2G46330; arabinogalactan protein 16 (AGP16) AT3G01500 beta-carbonic anhydrase 1 (BCA1) AT3G14420 glycolate oxidase 1 (GOX1) AT3G54890 photosystem I light harvesting complex gene 1 (LHCA1) AT4G15530 pyruvate orthophosphate dikinase (PPDK) AT4G22540 OSBP (oxysterol binding protein)-related protein 2A AT4G26530 Fructose-bisphosphate aldolase 5 (FBA5)

Figure 6. The conserved compound, darkness and submergence induced AS For each variant identifying gene region (VIGR) the deviation from the expected read counts if no change in variant usage upon treatment would take place (log2(obs/exp)) is shown for roots and shoots. Yellow indicates more, and cyan less reads than expected for a specific VIGR. All genes that are alternatively spliced (|log2(obs/exp)| > 1 and Padj. < Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016For American of Plant Biologists. All rights reserved. 0.01) in five or more genotypes are shown. eachSociety condition the genotypes are ordered from left to right: Cvi-0, Bay-0, Ita-0, Col(gl), Kas-1, Lp2-6, Ws-2 and C24.

LKR/SDH, root

compound

darkness

100

ROPGEF11, root

20

50

0

12

24

36

48

compound

0

0

GDH2, root

24

36

48

0

0

30

12

24

36

48

36

48

36

48

36

48

36

48

36

48

submergence 20

20 20 10 0

10

10

0

12

24

36

48

0

0

12

24

36

48

0

20

10

10

10

12

24

36

48

compound

3

0

0

12

24

36

48

darkness

4

12

24

submergence

20

0

0

darkness

20

0

E4PDH, shoot

12 darkness

compound

0

0

12

24

submergence

3 2

2 2

1

1 0

FBA1, shoot

FC of all transcript isoforms FC of specific transcript isoform(s) FC of specific transcript isoform(s) Cvi-0 C24

40

50

0

0

12

24

36

48

0

0

compound

1.0

12

24

36

48

darkness

0

0

12

24

submergence

1.0

1.0 0.5

0

0.5

0.5

0

12

24

36

48

0

0

compound PPDK, shoot

submergence

150

100

12

24

36

48

darkness

0

12

24

submergence

15

20

40

0

10 10

20

0

5

0

12

24

36

48

0

0

12

24

36

48

0

0

12

24

time (hours)

Figure 7. RT-qPCR validation of a selection of genes alternatively spliced upon compound, darkness and submergence. Fold change of transcript abundance in the accessions Cvi-0 and C24 upon compound, darkness and submergence of six genes that were identified as alternatively spliced upon the treatment in a conserved manner across accessions. Primers were used that either amplified all transcript variants (red lines), or selectively amplify only specific variants (blue and green lines). Root or shoot tissue was analyzed depending on the organ in which alternative splicing was identified. Details for each gene are as follows. LYSINE-KETOGLUTARATE REDUCTASE/SACCHAROPINE DEHYDROGENASE (AT4G33150, LKR/SDH, root): blue line is AT4G33150.1 and AT4G33150.2;; RHO GUANYL-NUCLEOTIDE EXCHANGE FACTOR 11 (AT1G52240, ROPGEF11, root): blue line is AT1G52240.1;; GLUTAMATE DEHYDROGENASE 2 (AT5G07440, GDH2, root): blue is AT5G07440.1 and AT5G07440.2 and green line is AT5G07440.1 and AT5G07440.3;; ERYTHRONATE-4-P DEHYDROGENASE (AT1G75180, E4PDH, shoot): blue line is AT1G75180.2 and AT1G75180.3;; FRUCTOSE-BISPHOSPHATE ALDOLASE 1 (AT2G21330, FBA1, shoot): blue line is specific for AT2G21330.2;; PYRUVATE ORTHOPHOSPHATE DIKINASE (AT4G15530, PPDK, shoot): blue line represents all transcripts, excluding AT4G15530.2 Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

A. ROPGEF11, root AL

D. PPDK, shoot AL

G. PRIN2, shoot AL

AD

AD

AD

SD

SD

SD

B. ARR1, root AL

E. LKR/SDH, root AL

H. BCA1, shoot AL

AD

AD

AD

SD

SD

SD

C. PMDH2, shoot AL

F. NPQ1, shoot AL

I. BCA4, shoot AL

AD

AD

AD

SD

SD

SD

Figure 8. Coverage plots of a selection of genes that are alternative spliced in response to the imposed treatments. Coverage plots were normalized to the maximum read depth, or a percentage of maximum read depth. Gene models (introns – line;; exons – fat box;; thin box – UTR) are shown below ordered from the first model (.1) onwards and are depicted from left to right in the 5’ to 3’ direction. AL is air light, AD is air darkness and SD is submergence darkness. Red or green lines indicate the location of intact protein domains specific to certain transcript isoforms. A. ROPGEF11, ROP (RHO OF PLANTS) GUANINE NUCLEOTIDE EXCHANGE FACTOR 11, AT1G52240, in Bay-0. Red domain - dynein light chain domain (PF01221), green domain - PRONE (PF03759) B. ARR1, ARABIDOPSIS RESPONSE REGULATOR 1, AT3G16857, in Bay-0. C. PMDH2, PEROXISOMAL NAD-MALATE DEHYDROGENASE 2, AT5G09660, in Bay-0. Red domain – NAD binding domain, green domain - malate dehydrogenase, alpha/beta C-terminal domain D. PPDK, PYRUVATE ORTHOPHOSPHATE DIKINASE, AT4G15530, in Bay-0 E. LKR/SDH, LYSINE-KETOGLUTARATE REDUCTASE/SACCHAROPINE DEHYDROGENASE, AT4G33150, in Bay-0. Red domain – LKR activity, green domain – SDH activity (Zhu et al. 2002) F. NPQ1, NON-PHOTOCHEMICAL QUENCHING 1, AT1G08550, in Bay-0. G. PRIN2, PLASTID REDOX INSENSITIVE 2, AT1G10522 in Bay-0. H. and I. BCA1 and BCA4, BETA CARBONIC ANHYDRASE 1 and 4, AT3G01500 and AT1G70410, in Lp2-6.


" -(

!"#$%&!&'

,( +( *( )( (

2/ '1 0 / ."

2/ '1 0 /

."

3""4 '6""4

5&98;48"!$73"/$;9&3;%&$3&'0"!'&$?

Figure 9. Flooding tolerance dependent compound, darkness and submergence responses. A. Number of DEGs in the shoot and root where the treatment response depends on the tolerance group (Ptolerance*treatment, adj. < 0.05). Flooding tolerance classification is based on Vashisht et al. (2011). Sensitive accessions are Cvi-0, Bay-0 and Ita-0 and tolerant accessions are Lp2-6, Ws-2 and C24. B. Tolerance specific DEGs (from A) that show distinct responses between the flooding tolerant (teal circles) and sensitive (yellow circles) accessions and their deviation from the average treatment response (for root and shoot compound and submergence treatments). Within each panel, left graphs show DEGs with fold-change higher in the tolerant accessions and the right shows DEGs with a fold-change higher in the sensitive accessions.


Parsed Citations Amthor, J (2000) The McCree-de Wit-Penning de Vries-Thornley respiration paradigms: 30 years later. Ann Bot 86: 1-20 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Angelovici R, Fait A, Fernie AR, and Galili G (2010) A seed high-lysine trait is negatively associated with the TCA cycle and slows down Arabidopsis seed germination. New Phytol 189: 148-159 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Antonio C, Papke C, Rocha M, Diab H, Limami AM, Obata T, Fernie AR, and van Dongen, JT (2015) Regulation of primary metabolism in response to low oxygen availability as revealed by carbon and nitrogen isotope redistribution. Plant Physiol: 1-34 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Armstrong J and Armstrong W (2001) An overview of the effects of phytotoxins on Phragmites australis in relation to die-back. Aquatic Botany 69: 251-268 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Baena-González E, Rolland F, Thevelein JM, and Sheen J (2007) A central integrator of transcription networks in plant stress and energy signaling. Nature 448: 938-942 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Banti V, Mafessoni F, Loreti E, Alpi A, and Perata P (2010) The Heat-inducible transcription factor HsfA2 enhances anoxia tolerance in Arabidopsis. Plant Physiol 152: 1471-1483 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Barding GA, Béni S, Fukao T, Bailey-Serres J, and Larive CK (2013) Comparison of GC-MS and NMR for metabolite profiling of rice subjected to submergence stress. J Prot Res 12: 898-909 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Bernal M, Casero D, Singh V, Wilson GT, Grande A, Yang H, Dodani SC, Pellegrini M, Huijser P, Connolly EL, Merchant SS, and Krämer U (2012) Transcriptome sequencing identifies SPL7-regulated copper acquisition genes FRO4/FRO5 and the copper dependence of iron homeostasis in Arabidopsis. Plant Cell 24: 738-761 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Branco-Price C, Kaiser KA, Jang CJH, Larive CK, and Bailey-Serres J (2008) Selective mRNA translation coordinates energetic and metabolic adjustments to cellular oxygen deprivation and reoxygenation in Arabidopsis thaliana. Plant J 56: 743-755 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Brosché M, Merilo E, Mayer F, Pechter P, Puzõrjova I, Brader G, Kangasjärvi J, and Kollist H (2010) Natural variation in ozone sensitivity among Arabidopsis thaliana accessions and its relation to stomatal conductance. Plant Cell Environ 33: 914-925 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Brown NJ, Palmer BG, Stanley S, Hajaji H, Janacek SH, Astley HM, Parsley K, Kajala K, Quick WP, Trenkamp S, Fernie AR, Maurino VG, Hibberd JM (2010) C4 acid decarboxylases required for C4 photosynthesis are active in the mid-vein of the C3 species Arabidopsis thaliana and are important in sugar and amino acid metabolism. Plant J 61: 122-133 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Campbell MT, Proctor CA, Dou Y, Schmitz AJ, Phansak P, Kruger, GR, Zhang C, Walia H (2015) Genetic and molecular characterization of submergence response identifies Subtol6 as a major submergence tolerance locus in maize PLoS ONE, 10, e0120385 doi:101371/journalpone0120385 Chang CY, Lin WD, and Tu SL (2014) Genome-wide analysis of heat-sensitive alternative splicing in Physcomitrella patens. Plant Physiol 165: 826-840 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Chang R, Jang CJH, Branco-Price C, Nghiem P, and Bailey-Serres J (2012) Transient MPK6 activation in response to oxygen deprivation and reoxygenation is mediated by mitochondria and aids seedling survival in Arabidopsis. Plant Mol Biol 78: 109-122 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Chen X, Pierik R, Peeters AJM, Poorter H, Visser EJW, Huber H, de Kroon H, and Voesenek LACJ (2010) Endogenous abscisic acid as a key switch for natural variation in flooding-induced shoot elongation. Plant Physiol 154: 969-977 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Chen L, Liao B, Qi H, Xie LJ, Huang L, Tan WJ, Zhai N, Yuan LB, Zhou Y, Yu, LJ, Chen QF, Shu W, and Xiao, S (2015) Autophagy contributes to regulation of the hypoxia response during submergence in Arabidopsis thaliana. Autophagy 11: 2233-2246 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Christianson JA, Wilson IW, Llewellyn DJ, and Dennis ES (2009) The low-oxygen-induced NAC domain transcription factor ANAC102 affects viability of Arabidopsis seeds following low-oxygen treatment. Plant Physiol 149: 1724-1738 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Delgado-Alvarado A, Walker RP, and Leegood RC (2007) Phosphoenolpyruvate carboxykinase in developing pea seeds is associated with tissues involved in solute transport and is nitrogen-responsive. Plant Cell Environ 30: 225-235 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Dennis ES, Pogson BJ, and Finnegan EJ (2009) VERNALIZATION INSENSITIVE 3 (VIN3) is required for the response of Arabidopsis thaliana seedlings exposed to low oxygen conditions 2009 Plant J 59:576-587 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Dinneny JR, Long TA, Wang JY, Jung JW, Mace D, Pointer S, Barron C, Brady SM, Schiefelbein J, and Benfey PN (2008) Cell identity mediates the response of Arabidopsis roots to abiotic stress. Science 320: 942-945 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Eastmond PJ, Astley HM, Parsley K, Aubry S, Williams BP, Menard GN, Craddock CP, Nunes-Nesi A, Fernie AR, and Hibberd JM (2015) Arabidopsis uses two gluconeogenic gateways for organic acids to fuel seedling establishment. Nature Comm 6: 6659 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ellis MH, Dennis ES, and Peacock WJ (1999) Arabidopsis roots and shoots have different mechanisms for hypoxic stress tolerance. Plant Physiol 119: 57-64 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Fabre N, Reiter IM, Becuwe-Linka N, Genty B, and Rumeau D (2007) Characterization and expression analysis of genes encoding alpha and beta carbonic anhydrases in Arabidopsis. Plant Cell Environ 30: 617-629 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Fukao T, Xu K, Ronald PC, and Bailey-Serres, J (2006) A variable cluster of ethylene rsponse factor-like genes regulates metabolic and developmental acclimation responses to submergence in Rice. Plant Cell 18: 2021-2034 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Fukao T, Yeung E, and Bailey-Serres J (2011) The submergence tolerance regulator SUB1A mediates crosstalk between submergence and drought tolerance in Rice. Plant Cell 23: 412-427 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gaillochet, C, Daum, G, and Lohmann, JU (2015) O Cell, Where Art Thou? The mechanisms of shoot meristem patterning Curr Opin Plant Biol 23: 91-97 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gibbs DJ, Lee SC, Isa NM, Gramuglia S, Fukao T, Bassel GW, Correia CS, Corbineau F, Theodoulou FL, Bailey-Serres J, and Holdsworth MJ (2011) Homeostatic response to hypoxia is regulated by the N-end rule pathway in plants. Nature 479: 415-418 Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gibbs DJ, Isa NM, Movahedi M, Lozano-Juste J, Mendiondo GM, Berckhan S, Marin-de la Rosa N, Conde JV, Correia CS, Pearce SP, Bassel GW, Hamali B, Talloji P, Tome DFA, Coego A, Beynon J, Alabadi D, Bachmair A, Leon J, Gray JE, Theodoulou FL, Holdsworth MJ (2014) Nitric oxide sensing in plants Is mediated by proteolytic control of Group VII ERF transcription factors. Mol Cell 53: 369-379 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Giuntoli B, Lee SC, Licausi F, Kosmacz M, Oosumi T, van Dongen JT, Bailey-Serres J, and Perata P (2014) A trihelix DNA binding protein counterbalances hypoxia-responsive transcriptional activation in Arabidopsis. PLoS Biol 12: e1001950 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gommers CMM, Visser EJW, St Onge KR, Voesenek LACJ, and Pierik R (2013) Shade tolerance: when growing tall is not an option. Trends Plant Sci 18: 65-71 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Gonzali, S, Loreti, E, Cardarelli, F, Novi, G, Parlanti, S, Pucciariello, C, Bassolino, L, Banti, V, Licausi, F, and Perata, P (2015) Universal stress protein HRU1 mediates ROS homeostasis under anoxia Nature Plants 1: 15151 Graf A, Schlereth A, Stitt M, and Smith AM (2010) Circadian control of carbohydrate availability for growth in Arabidopsis plants at night. Proc Natl Acad Sci USA 107: 9458-9463 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hartmann L, Pedrotti L, Weiste C, Fekete A, Schierstaedt J, Gottler J, Kempa S, Krischke M, Dietrich K, Mueller MJ, VicenteCarbajosa J, Hanson J, Droge-Laser W (2015) Crosstalk between Two bZIP signaling pathways orchestrates salt-induced metabolic reprogramming in Arabidopsis roots. Plant Cell 27: 2244-2260 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hattori Y, Nagai K, Furukawa S, Song XJ, Kawano R, Sakakibara H, Wu J, Matsumoto T, Yoshimura A, Kitano H, Matsuoka M, Mori H, Ashikari M (2009) The ethylene response factors SNORKEL1 and SNORKEL2 allow rice to adapt to deep water. Nature 460: 1026-1030 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hibberd JM and Quick WP (2002) Characteristics of C4 photosynthesis in stems and petioles of C3 flowering plants. Nature 415: 451-454 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hildebrandt TM, Nesi AN, Araújo WL, and Braun HP (2015) Amino acid catabolism in plants. Molecular Plant: 1-35 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hsu FC, Chou MY, Peng HP, Chou SJ, and Shih MC (2011) Insights into hypoxic systemic responses based on analyses of transcriptional regulation in Arabidopsis. PLoS ONE 6: e28888 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Hsu PY and Harmer SL (2014) Wheels within wheels: the plant circadian system. Trends in Plant Sci 19: 240-249 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Huang S, Colmer TD, and Millar AH (2008) Does anoxia tolerance involve altering the energy currency towards PPi? Trend Plant Sci 13: 221-227 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Huijser P and Schmid M (2011) The control of developmental phase transitions in plants. Development 138: 4117-4129 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Downloaded from www.plantphysiol.org May Cryptochromes 16, 2016 - Published www.plant.org synergistically regulate Usami, T, Mochizuki, N, Kondo, M, Nishimura, M, and Nagatani, Aon (2004) andbyphytochromes Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Arabidopsis root greening under blue light Plant Cell Physiol 45:1798-1808 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Jain A, Wilson GT, and Connolly EL (2014) The diverse roles of FRO family metalloreductases in iron and copper homeostasis. Front Plant Sci 5: 100 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Li J, Li G, Wang H, Deng XW (2011) Phytochrome Signaling Mechanisms. The Arabidopsis Book, American Society of Plant Biologists. 9:e0148 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Juntawong P, Girke T, Bazin J, and Bailey-Serres J (2014) Translational dynamics revealed by genome-wide profiling of ribosome footprints in Arabidopsis. Proc Natl Acad Sci USA 111: E203-12 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kazan K (2003) Alternative splicing and proteome diversity in plants: the tip of the iceberg has just emerged. Trends Plant Sci 8: 468-471 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kieber JJ and Schaller GE (2014) Cytokinins. The Arabidopsis Book 12: e0168 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, and Salzberg SL (2013) TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 14: R36 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kobayashi K, Baba S, Obayashi T, Sato M, Toyooka K, Keranen M, Aro EM, Fukaki H, Ohta H, Sugimoto K, and Masuda T (2012) Regulation of root greening by light and Auxin/Cytokinin signaling in Arabidopsis. Plant Cell 24: 1081-1095 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Kreuzwieser J, Hauberg J, Howell KA, Carroll A, Rennenberg H, Millar AH, and Whelan J (2009) Differential response of gray poplar leaves and roots underpins stress adaptation during hypoxia. Plant Physiol 149: 461-473 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Langfelder P and Horvath S (2008) WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9: 559 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Lawrence M, Huber W, Pagès H, Aboyoun P, Carlson M, Gentleman R, Morgan MT, and Carey VJ (2013) Software for computing and annotating genomic ranges. PLoS Comput Biol 9: e1003118 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Lee SC, Mustroph A, Sasidharan R, Vashisht D, Pedersen O, Oosumi T, Voesenek LACJ, and Bailey-Serres J (2011) Molecular characterization of the submergence response of the Arabidopsis thaliana ecotype Columbia. New Phytol 190: 457-471 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Lee KW, Chen PW, Lu CA, Chen S, Ho THD, and Yu SM (2009) Coordinated responses to oxygen and sugar deficiency allow rice seedlings to tolerate flooding. Sci Signaling 2: ra61-ra61 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Licausi F, Kosmacz M, Weits DA, Giuntoli B, Giorgi FM, Voesenek LACJ, Perata P, and van Dongen JT (2011) Oxygen sensing in plants is mediated by an N-end rule pathway for protein destabilization. Nature 479: 419-422 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title Downloaded from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Ljung K, Nemhauser JL, and Perata, P (2015) New mechanistic links between sugar and hormone signalling networks. Curr Opin Plant Biol 25: 130-137 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Lokdarshi A, Conner WC, McClintock C, Li T, and Roberts D (2015) Arabidopsis CML38, a calcium sensor that localizes to ribonucleoprotein complexes under hypoxia stress. Plant Physiol 170: 1046-1059 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Magneschi L, Kudahettige RL, Alpi A, and Perata P (2009) Comparative analysis of anoxic coleoptile elongation in rice varieties: relationship between coleoptile length and carbohydrate levels, fermentative metabolism and anaerobic gene expression. Plant Biol 11: 561-573 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Malone S, Chen ZH, Bahrami AR, Walker RP, Gray JE, and Leegood RC (2007) Phosphoenolpyruvate Carboxykinase in Arabidopsis: Changes in gene expression, protein and activity during vegetative and reproductive development. Plant Cell Physiol 48: 441-450 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mommer L and Visser EJW (2005) Underwater photosynthesis in flooded terrestrial plants: A matter of leaf plasticity. Ann Bot 96: 581-589 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Morgan M, Pages H, Obenchain V, and Haydon N (2013) Rsamtools: Binary alignment (BAM), FASTA, variant call (BCF) and tabix file import. R package version 1.22.0 http://bioconductor.org/packages/release/bioc/html/Rsamtools.html. Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mugnai S, Azzarello E, Baluska F, and Mancuso S (2012) Local root apex hypoxia induces NO-mediated hypoxic acclimation of the entire root. Plant Cell Physiol 53: 912-920 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mustroph A, Barding GA, Kaiser KA, Larive CK, and Bailey-Serres, J (2014a) Characterization of distinct root and shoot responses to low-oxygen stress in Arabidopsis with a focus on primary C- and N-metabolism. Plant Cell Environ 37: 2366-2380 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mustroph, A, Hess, N, and Sasidharan, R (2014b) Hypoxic energy metabolism and PPi as an alternative energy currency. In Plant Cell Monographs, Plant Cell Monographs (Springer Vienna: Vienna), pp 165-184 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mustroph A, Lee SC, Oosumi T, Zanetti ME, Yang H, Ma K, Yaghoubi-Masihi A, Fukao T, and Bailey-Serres J (2010) Cross-Kingdom comparison of transcriptomic adjustments to low-oxygen stress highlights conserved and plant-specific responses. Plant Physiol 152: 1484-1500 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Mustroph A, Zanetti ME, Jang CJH, Holtan HE, Repetti PP, Galbraith DW, Girke T, and Bailey-Serres J (2009) Profiling translatomes of discrete cell populations resolves altered cellular priorities during hypoxia in Arabidopsis. Proc Natl Acad Sci USA 106: 1884318848 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Müller J, Toev T, Heisters M, Teller J, Moore KL, Hause G, Dinesh DC, Bürstenbinder K, and Abel S (2015) Iron-dependent callose deposition adjusts root meristem maintenance to phosphate availability. Dev Cell 33: 216-230 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Narsai, R, Edwards, JM, Roberts, TH, Whelan, J, Joss, GH, and Atwell, BJ (2015) Mechanisms of growth and patterns of gene expression in oxygen-deprived rice coleoptiles Plant J 82: 25-40 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title


Narsai, R, Howell, KA, Carroll, A, Ivanova, A, Millar, AH, and Whelan, J (2009) Defining core metabolic and transcriptomic responses to oxygen availability in rice embryos and young seedlings Plant Physiol 151: 306-322 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Pal SK, Liput M, Piques M, Ishihara H, Obata T, Martins MCM, Sulpice R, Van Dongen JT, Fernie A, Yadav UP, Lunn JE, Usadel B, and Stitt M (2013) Diurnal changes of polysome loading track sucrose content in the rosette of wild-type Arabidopsis and the starchless pgm mutant. Plant Physiol 162: 1246-1265 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Parsley K and Hibberd JM (2006) The Arabidopsis PPDK gene is transcribed from two promoters to produce differentially expressed transcripts responsible for cytosolic and plastidic proteins. Plant Mol Biol 62: 339-349 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Penfield S, Rylott EL, Gilday AD, Graham S, Larson TR, and Graham IA (2004) Reserve mobilization in the Arabidopsis endosperm fuels hypocotyl elongation in the dark, is independent of abscisic acid, and requires PHOSPHOENOLPYRUVATE CARBOXYKINASE1. Plant Cell 16: 2705-2718 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Pracharoenwattana I, Cornah JE, and Smith SM (2007) Arabidopsis peroxisomal malate dehydrogenase functions in beta-oxidation but not in the glyoxylate cycle. Plant J 50: 381-390 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, and Smyth GK (2015) Limma powers differential expression analyses for RNAsequencing and microarray studies. Nucleic Acids Res 43: e47-e47 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Robinson MD and Oshlack A (2010) A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol 11: R25 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Rocha M, Licausi F, Araujo WL, Nunes-Nesi A, Sodek L, Fernie AR, and van Dongen JT (2010) Glycolysis and the tricarboxylic acid cycle are linked by alanine aminotransferase during hypoxia induced by waterlogging of Lotus japonicas. Plant Physiol 152: 15011513 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Sasidharan R, Mustroph A, Boonman A, Akman M, Ammerlaan AMH, Breit T, Schranz ME, Voesenek LACJ, and van Tienderen PH (2013) Root transcript profiling of two Rorippa species reveals gene clusters associated with extreme submergence tolerance. Plant Physiol 163: 1085-1086 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Schroder F, Lisso J, and Mussig C (2011) EXORDIUM-LIKE1 promotes growth during low carbon availability in Arabidopsis. Plant Physiol 156: 1620-1630 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell 2: 1027-1038 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Shin DH, Cho MH, Kim TL, Yoo J, Kim JI, Han YJ, Song PS, Jeon JS, Bhoo SH, and Hahn TR (2010) A small GTPase activator protein interacts with cytoplasmic phytochromes in regulating root development. J Biol Chem 285: 32151-32159 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Smeekens S, Ma J, Hanson J, and Rolland, F (2010) Sugar signals and molecular networks controlling plant growth. Curr Opin Plant Biol 13: 274-279 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Downloaded Only Author and Title from www.plantphysiol.org on May 16, 2016 - Published by www.plant.org Copyright © 2016 American Society of Plant Biologists. All rights reserved.

Sorenson R and Bailey-Serres, J (2014) Selective mRNA sequestration by OLIGOURIDYLATE-BINDING PROTEIN 1 contributes to translational control during hypoxia in Arabidopsis. Proc Natl Acad Sci USA 111: 2373-2378 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Svistoonoff S, Creff A, Reymond M, Sigoillot-Claude C, Ricaud L, Blanchet A, Nussaume L, and Desnos T (2007) Root tip contact with low-phosphate media reprograms plant root architecture. Nat Genet 39: 792-796 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Tang G, Zhu X, Gakiere B, Levanony H, Kahana A, and Galili G (2002) The bifunctional LKR/SDH locus of plants also encodes a highly active monofunctional lysine-ketoglutarate reductase using a polyadenylation signal located within an intron. Plant Physiol 130: 147-154 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Taylor L, Nunes-Nesi A, Parsley K, Leiss A, Leach G, Coates S, Wingler A, Fernie AR, and Hibberd JM (2010) Cytosolic pyruvate,orthophosphate dikinase functions in nitrogen remobilization during leaf senescence and limits individual seed growth and nitrogen content. Plant J 62: 641-652 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Usadel B, Blasing OE, Gibon Y, Retzlaff K, Hohne M, Gunther M, and Stitt M (2008) Global transcript levels respond to small changes of the carbon status during progressive exhaustion of carbohydrates in Arabidopsis rosettes. Plant Physiol 146: 18341861 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

van Dongen JT, and Licausi F (2015) Oxygen sensing and signaling. Ann Rev Plant Biol 66: 345-367 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Van Eck WHJM, van de Steeg HM, Blom CWPM, and de Kroon H (2004) Is tolerance to summer flooding correlated with distribution patterns in river floodplains? A comparative study of 20 terrestrial grassland species. Oikos 107: 393-405 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

van Veen H, Akman M, Jamar DCL, Vreugdenhil D, Kooiker M, van Tienderen P, Voesenek LACJ, Schranz ME, and Sasidharan R (2014a) Group VII Ethylene Response Factor diversification and regulation in four species from flood-prone environments. Plant Cell Environ 37:2421-32 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

van Veen H, Mustroph A, Barding GA, Vergeer-van Eijk M, Welschen-Evertman RAM, Pedersen O, Visser EJW, Larive CK, Pierik R, Bailey-Serres J, Voesenek LACJ, and Sasidharan R (2013) Two Rumex species from contrasting hydrological niches regulate flooding tolerance through distinct mechanisms. Plant Cell 25: 4691-4707 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

van Veen H, Vashisht D, Voesenek LACJ, and Sasidharan R (2014b) Different survival strategies amongst plants to cope with underwater conditions. In Plant Cell Monographs, Plant Cell Monographs (Springer Vienna: Vienna), pp 329-349 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Vashisht D, Hesselink A, Pierik R, Ammerlaan JMH, Bailey-Serres J, Visser EJW, Pedersen O, van Zanten M, Vreugdenhil D, Jamar DCL, Voesenek LACJ, and Sasidharan R (2011) Natural variation of submergence tolerance among Arabidopsis thaliana accessions. New Phytol 190: 299-310 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Vervuren P, Blom C, and de Kroon H (2003) Extreme flooding events on the Rhine and the survival and distribution of riparian plant species. J Ecology 91: 135-146 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Voesenek LACJ and Bailey-Serres J (2015) Flood adaptive traits and processes: an overview. New Phytol 206: 57-73 Pubmed: Author and Title CrossRef: Author and Title


Google Scholar: Author Only Title Only Author and Title

Voesenek LACJ and Sasidharan R (2013) Ethylene--and oxygen signalling--drive plant survival during flooding. Plant Biol (Stuttg) 15: 426-435 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Voesenek LACJ, Rijnders JHGM, Peeters AJM, van de Steeg HM, and de Kroon H (2004) Plant hormones regulate fast shoot elongation under water: from genes to communities. Ecology 85: 16-27 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Voesenek LACJ, van Veen H, and Sasidharan R (2014) Learning from nature: the use of non-model species to identify novel acclimations to flooding stress. AoB Plants 6 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Watanabe M, Balazadeh S, Tohge T, Erban A, Giavalisco P, Kopka J, Mueller-Roeber B, Fernie AR, and Hoefgen R (2013) Comprehensive dissection of spatiotemporal metabolic shifts in primary, secondary, and lipid metabolism during developmental senescence in Arabidopsis. Plant Physiol 62:1290-1310 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Weits DA, Giuntoli B, Kosmacz M, Parlanti S, Hubberten HM, Riegler H, Hoefgen R, Perata P, van Dongen JT, and Licausi F (2014) Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nature Comm 5: 3425 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Xu K, Xu X, Fukao T, Canlas P, Maghirang-Rodriguez R, Heuer S, et al (2006) Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature 442: 705-708 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Young MD, Wakefield MJ, Smyth GK, and Oshlack A (2010) Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol 11: R14 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Zabalza A, van Dongen JT, Froehlich A, Oliver SN, Faix B, Gupta KJ, Schmalzlin E, Igal M, Orcaray L, Royuela M, and Geigenberger P (2008) Regulation of respiration and fermentation to control the plant internal oxygen concentration. Plant Physiol 149: 1087-1098 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title

Zhu X, Tang G, and Galili G (2002) The activity of the Arabidopsis bifunctional lysine-ketoglutarate reductase/saccharopine dehydrogenase enzyme of lysine catabolism is regulated by functional interaction between its two enzyme domains. J Biol Chem 277: 49655-49661 Pubmed: Author and Title CrossRef: Author and Title Google Scholar: Author Only Title Only Author and Title
