Identification of Proteins Involved in Salinity Tolerance in Salicornia Bigelovii

Home , Kalidium, Salicornia, Salicornia bigelovii, Salicornia europaea

Identification of Proteins Involved in Salinity Tolerance in Salicornia bigelovii

Octavio Rubén Salazar Moya

In Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy of Science

King Abdullah University of Science and Technology, Thuwal, Kingdom of Saudi Arabia

November, 2017 2

EXAMINATION COMMITTEE PAGE

The dissertation of Octavio Rubén Salazar Moya is approved by the examination committee.

Committee Chairperson: Mark Tester Committee Members: Enrico Martinoia, Heribert Hirt, and Timothy Ravasi

Octavio Rubén Salazar Moya

ABSTRACT

With a global growing demand in food production, agricultural output must increase accordingly. An increased use of saline soils and brackish water would contribute to the required increase in world food production. Abiotic stresses, such as salinity and drought, are also major limiters of crop growth globally - most crops are relatively salt sensitive and are significantly affected when exposed to salt in the range of 50 to 200 mM NaCl. Genomic resources from plants that naturally thrive in highly saline environments have the potential to be valuable in the generation of salt tolerant crops; however, these resources have been largely unexplored.

Salicornia bigelovii is a plant native to Mexico and the United States that grows in salt marshes and coastal regions. It can thrive in environments with salt concentrations higher than seawater. In contrast to most crops, S. bigelovii is able to accumulate very high concentrations (in the order of 1.5 M) of Na+ and Cl- in its photosynthetically active succulent shoots. Part of this tolerance is likely to include the storage of Na+ in the vacuoles of the shoots, making S. bigelovii a good model for understanding mechanisms of Na+ compartmentalization in the vacuoles and a good resource for gene discovery.

In this research project, phenotypic, genomic, transcriptomic, and proteomic approaches have been used for the identification of candidate genes involved in salinity tolerance in S. bigelovii. The genomes and transcriptomes of 5

three Salicornia species have been sequenced. This information has been used to support the characterization of the salt-induced transcriptome of S. bigelovii shoots and the salt-induced proteome of various organellar membrane enriched fractions from S. bigelovii shoots, which led to the creation of organellar membrane proteomes. Yeast spot assays at different salt concentrations revealed several proteins increasing or decreasing yeast salt tolerance. This work aims to create the basis for Salicornia research by providing a genome, transcriptomes, and organellar proteomes, contributing to salinity tolerance research overall. We identified a set of candidate genes for salinity tolerance with the aim of shedding some light on the mechanisms by which this plant thrives in highly saline environments.

ACKNOWLEDGEMENTS

I would like to thank Prof. Mark Tester for accepting me in his lab when I was facing the unfortunate situation of my previous supervisor having to leave the university. Without his support, I would not have been able to continue my career in my field of interest. I am truly grateful to have received this lifeline. Of course,

I also appreciate all the time we spent together and the discussions that we had. I would also like to thank Dr. Sandra Schmöckel for her assistance and support throughout my Ph.D., she basically acted as my co-supervisor. All the discussions that we had, helped us shape the project. I thank the support of all the “Salties”, particularly to all the students whose conversations always helped to get a smile back into my face. I would like to thank Prof. Nina Fedoroff, she was the reason why I joined KAUST and she started the project with Salicornia. Dr. Suhail Khan was also involved at the start of the Salicornia project and we spent time working together. Dr. Alain Lesimple taught me and worked with me in analytical chemistry, even while he was facing a difficult time. Dr. Muppala Reddy was in charge of Salicornia growth in the greenhouse and helped me understand more about Salicornia. My undergraduate student José Otero, for spending all those long nights in the lab working with me. All the lab support team, who make all the work in the laboratory possible. All the friends that I made in KAUST were very important for me. DooLi Kim, for all her unconditional support during the hard times. Of course, to my committee members for all their time and input while 7

assessing my research project. I would like to thank KAUST for giving me the opportunity to study and fund my research. I would like to thank my family members, who have been always with me. My mother, who has always done her best to support me in any way possible. I would not have been able to do this Ph.D. without all the sacrifices that she has done during her life for me. Thank you, mother, you deserve all the credit. To my aunt, who basically has acted as a second mother throughout my life. You have always supported me and I am very happy that you have looked out for me throughout my life. I also hope this will encourage my cousin, Raúl, whatever you decide to do in life, give it everything. To all my friends back in my country, for their constant support, despite being afar, every time we meet is as if not a single day has passed. There are many people who assisted me and is impossible to mention every single person, but I am grateful to all of you.

TABLE OF CONTENTS

EXAMINATION COMMITTEE PAGE ...... 2

ABSTRACT ...... 4

ACKNOWLEDGEMENTS ...... 6

LIST OF ABBREVIATIONS ...... 12

LIST OF GENES AND PROTEINS ...... 17

LIST OF ILLUSTRATIONS ...... 20

LIST OF TABLES ...... 24

Chapter 1. Literature review ...... 27

1.1 Introduction ...... 27

1.2 Salinity tolerance mechanisms of land plants ...... 32

1.2.1 Shoot ion independent toxicity ...... 33

1.2.2 Shoot ion toxicity ...... 37

1.3 The role and mechanisms of Na+ accumulation in the vacuoles ...... 41

1.3.1 Ion transport across the tonoplast ...... 42

1.3.2 Proteins involved in Na+ accumulation in vacuoles...... 45

1.4 Salicornia as a model plant for understanding salinity tolerance ...... 56

1.4.1 Salinity tolerance in Salicornia ...... 58

1.4.2 Evidence of vacuolar Na+/H+ antiporter activity in Salicornia ...... 61

1.5 Aim of the project ...... 64

Chapter 2. Genome assembly and annotation of Salicornia species...... 66 9

2.1 Introduction ...... 66

2.2 Materials and methods ...... 69

2.2.1 Plant material, library preparation, and genome sequencing ...... 69

2.2.2 Read trimming and genome assembly ...... 71

2.2.3 Genome selection, completeness, and annotation ...... 72

2.3 Results and discussion ...... 73

2.3.1 Genome sequencing ...... 73

2.3.2 Genome assembly using Illumina reads ...... 76

2.3.3 Salicornia bigelovii genome assembly using SMRT sequencing and

hybrid assemblies ...... 88

2.4 Conclusion ...... 90

Chapter 3. Salicornia bigelovii phenotype under salinity ...... 92

3.1 Introduction ...... 92

3.2 Materials and methods ...... 94

3.2.1 Plant material and growth ...... 94

3.2.2 Plant ion content and phenotype ...... 95

3.3 Results and discussion ...... 96

3.3.1 Sodium and potassium distribution in Salicornia bigelovii plants ...... 96

3.3.2 Salicornia bigelovii growth at different NaCl concentrations...... 100

3.3.3 Salicornia bigelovii ion accumulation during growth at different NaCl

concentrations ...... 104

3.4 Conclusion ...... 117 10

Chapter 4. Transcriptomic and proteomic profiling of Salicornia bigelovii shoots under different salinities ...... 119

4.1 Introduction ...... 119

4.2 Materials and methods ...... 123

4.2.1 Plant material and growth ...... 123

4.2.2 RNA extraction and library preparation ...... 124

4.2.3 Transcriptome assembly, annotation, and differential expression

analysis ...... 125

4.2.4 Plant membrane extraction, fractionation, and quantitation ...... 127

4.2.5 Protein digestion and LC-MS analyses ...... 128

4.2.6 MS spectra analyses and protein quantitation ...... 129

4.2.7 Subcellular localization prediction ...... 130

4.2.8 Validation of subcellular compartmentalization prediction of pRoloc by

Western blot ...... 131

4.3 Results and discussion ...... 132

4.3.1 Transcriptomes assembly and annotation ...... 132

4.3.2 Gene differential expression analysis in response to salinity treatment in

Salicornia bigelovii ...... 137

4.3.3 Salicornia bigelovii membrane proteins differential abundance analysis

and generation of organellar membrane proteomes ...... 160

4.4 Conclusion ...... 184 11

Chapter 5. Selection and characterization of candidate genes related to salt tolerance in Salicornia bigelovii ...... 188

5.1 Introduction ...... 188

5.2 Materials and methods ...... 190

5.2.1 Gene selection...... 190

5.2.2 RNA extraction, and cloning of candidate genes into E. coli ...... 191

5.2.3 Yeast transformation, spot assays, and subcellular localization ...... 192

5.3 Results and discussion ...... 194

5.3.1 Gene selection for characterization ...... 194

5.3.2 Cloning and characterization in yeast ...... 207

5.3.3 Gene analysis ...... 224

5.4 Conclusion ...... 236

Chapter 6. Conclusion and future perspectives ...... 239

6.1 General conclusion and main findings ...... 239

6.2 Limitations of this study ...... 242

6.3 Perspectives and proposed future work ...... 245

References ...... 248

APPENDIX ...... 302

LIST OF ABBREVIATIONS

% percent

~ approximately

°C degrees Celsius

µg microgram(s)

µL microliter(s)

µM micromolar

3’ three prime, of a nucleic acid sequence

5’ five prime, of a nucleic acid sequence aa amino acid

ABA abscisic acid

ABySS assembly by short sequences

AP medium arginine phosphate medium

Arabidopsis Arabidopsis thaliana

B. vulgaris Beta vulgaris

BLAST basic local alignment search tool bp base pairs

BTP Bis-tris propane

BUSCO benchmarking universal single-copy orthologs

C. quinoa Chenopodium quinoa

Ca2+ calcium ion 13

CDD conserved domains database

3-[(3-cholamidopropyl)dimethyl-ammonio]-1-propane CHAPS sulfonate

Cl- chloride ion cm centimeter(s)

Cs+ cesium ion

CSM complete supplement mix d day(s)

DE differential expression

DM dry mass

DNA deoxyribonucleic acid

EC electric conductivity

EDX energy dispersive x-ray

EGFP enhanced green fluorescent protein

EMS ethyl methanesulfonate

ER endoplasmic reticulum

ESP exchangeable sodium percentage

Ex membrane potential

F Faraday’s constant

F fluorescence

FAO Food and Agriculture Organization of the United Nations 14

FASP filter aided sample preparation g gram(s) g gravity

Gb giga base pairs

H+ hydrogen ion

HRP Horseradish peroxidase

JA jasmonic acid

K potassium

K+ potassium ion kb kilo base pairs

KCL potassium chloride kDA kilo dalton(s)

KEGG Kyoto Encyclopedia of Genes and Genomes kg kilogram(s)

Km Michaelis constant

L liter(s)

LC liquid chromotagraphy

Mb mega base pairs min minute(s) mL milliliter(s)

ML maximum likelihood 15

mM millimolar mm millimolal mOsm milliosmole

MS mass spectrometry

MudPIT multidimensional protein identification technology mV millivolts

Na sodium

Na+ sodium ion

NaCl sodium chloride nm nanometer(s)

OD600 optical density at a wavelength of 600 nm

ORF open reading frame

PCA principal component analysis

PCR polymerase chain reaction pH potential of hydrogen

PPi pyrophosphate

QTL quantitative trait locus

R gas constant

Rb+ rubidium ion

RNA ribonucleic acid

ROS reactive oxygen species 16

s second(s)

S. bigelovii Salicornia bigelovii

S. brachaita Salicornia brachiata

S. europaea Salicornia europaea

S. oleracea Spinacia oleracea

Salicornia Salicornia bigelovii

SD medium synthetic defined medium

SDS Sodium dodecyl sulfate

SEM scanning electron microscope

SMRT single molecule real-time

T temperature

TAIR The Arabidopsis Information Resource

TFA trifluoroacetic acid

Tm melting temperature

TUE transpiration use efficiency

UDP uridine diphosphate

Vmax maximum rate of reaction wk weeks(s) x times z ion valence

LIST OF GENES AND PROTEINS

ABC abc transporter

ACA calcium ATPase

AMT ammonium transporter

AOX1 alternative oxidase 1

BOR boron transporter

CAAT cationic amino acid transporter

CAX Ca2+ / proton exchangers

CBL calcineurin B-like

CHX cation / hydrogen exchanger

CIPK CBL-interacting protein kinases

CLC chloride conducting channels

CNGC cyclic nucleotide gated channel

DIT dicarboxylate transporter

DUF domain of unknown function

ENT equilibrative nucleoside transporter

HHP heptahelical transmembrane protein

HKT high affinity potassium transporters

KAT highly selective inward-rectifying potassium channel

LAX auxin transporter-like protein

LRR leucine rich repeat 18

LTP lipid transfer protein

MATE multi drug and toxin extrusion

MRS magnesium transporter

MSL small conductance mechanosensitive channel-like

NHX sodium / hydrogen exchanger

OCT organic cation / carnitine transporter

OPT oligopeptide transporter

PIN1 pinformed 1

PIP plasma membrane intrinsic protein

POT potassium transporter

PQ PQ-loop repeat containing protein

PsbA photosystem II protein D1 precursor

RBOH respiratory burst homolog

Arginine Glycine Glycine (RGG) box-containing RNA- RGGA binding protein

SOS1 salt overly sensitive 1

STC sugar carrier protein

SUT sulfate transporter

SWEET Bidirectional sugar transporter SWEET

TIP tonoplast intrinsic protein

TMN9 transmembrane 9 superfamily 19

V-ATPase vacuolar ATPase

V-PPase vacuolar pyrophosphatase

YSL probable metal-nicotianamine transporter

LIST OF ILLUSTRATIONS

Figure 1. Projections of global production for maize, rice, wheat, and soybean... 28

Figure 2. Salt affected land in the world ...... 31

Figure 3. Plant responses to salinity ...... 33

Figure 4. Representation of Na+ influx and mobilization within a cell in response to an elevated Na+ concentration in the apoplast ...... 46

Figure 5. Salicornia bigelovii physiology ...... 57

Figure 6. Sodium accumulation in Salicornia europaea ...... 60

Figure 7. Na+ dependent dissipation of the proton gradient in vacuole vesicles from Salicornia bigelovii ...... 62

Figure 8. Genome assembly strategy using Illumina reads ...... 77

Figure 9. Relative Na content in top and bottom branches of S. bigelovii lines. ... 98

Figure 10. Na and K contents in upper shoots of S. bigelovii inbred lines ...... 99

Figure 11. Saliornia bigelovii growth under different salinities ...... 101

Figure 12. Growth of individual S. bigelovii plants under different salinities ..... 102

Figure 13. S. bigelovii shoot expansion upon NaCl treatment...... 103

Figure 14. S. bigelovii shoot sap osmolality and water content ...... 105

Figure 15. S. bigelovii Na accumulation in shoots ...... 107

Figure 16. S. bigelovii Na accumulation in roots ...... 108

Figure 17. S. bigelovii K accumulation in shoots ...... 110

Figure 18. S. bigelovii K accumulation in roots...... 111 21

Figure 19. S. bigelovii Na / K molality ratio...... 112

Figure 20. S. bigelovii Cl- accumulation in shoots ...... 114

Figure 21. S. bigelovii Cl- accumulation in roots ...... 115

Figure 22. Correlations between ion accumulation and water content in S. bigelovii plants ...... 116

Figure 23. Saliornia bigelovii growth under different salinities ...... 138

Figure 24. Hierarchical clustering of three replicates per treatment of differentially expressed genes in shoots of S. bigelovii plants ...... 143

Figure 25. Cell wall MapMan pathway for each of the treatments (transcriptomics)

...... 150

Figure 26. Biotic and abiotic stress MapMan pathway (transcriptomics) ...... 154

Figure 27. Receptor-like kinases MapMan pathway for plants treated with NaCl for 1 week (transcriptomics) ...... 156

Figure 28. Receptor-like kinases MapMan pathway for plants treated with NaCl for 6 weeks (transcriptomics) ...... 157

Figure 29. MapMan pathway for plants treated with 600 mM NaCl for 6 weeks

(transcriptomics) ...... 159

Figure 30. Separation of organellar membranes by their densities through ultracentrifugation...... 162

Figure 31. Sucrose % and specific gravity of each of the 12 density fractions .... 164

Figure 32. Protein concentration in µg/mL and specific gravity of each of the 12 density fractions ...... 165 22

Figure 33. Hierarchical clustering of three replicates per treatment of differentially abundant proteins in shoots of S. bigelovii plants ...... 170

Figure 34. MapMan pathway of cell wall precursors in plants treated with NaCl for 6 weeks (proteomics) ...... 173

Figure 35. MapMan pathway of cell wall precursors in plants treated with NaCl for 6 weeks (proteomics) ...... 174

Figure 36. Photosynthesis MapMan pathway in plants treated with NaCl for 6 weeks (proteomics) ...... 176

Figure 37. Photosynthesis MapMan pathway in plants treated with NaCl for 6 weeks (proteomics) ...... 177

Figure 38. Validation of marker abundance profiles generated with pRoloc by

Western blot ...... 180

Figure 39. Principal component analysis of protein predicted subcellular localization by pRoloc ...... 183

Figure 40. Expression profiles of the three different groups of selected proteins for characterization...... 198

Figure 41. Representation of AXT3 yeast strain ...... 213

Figure 42. Example of yeast spot assays ...... 215

Figure 43. Yeast spot assays of genes increasing salt tolerance (SD medium) .... 218

Figure 44. Yeast spot assays of genes increasing salt tolerance (AP medium) ... 219

Figure 45. Yeast spot assays of genes, without C terminal EGFP, increasing salt tolerance (SD medium)...... 220 23

Figure 46. Yeast spot assays of genes decreasing salt tolerance (SD medium) ... 221

Figure 47. Yeast spot assays of genes decreasing salt tolerance (AP medium) .. 222

Figure 48. Yeast spot assays of genes, without C terminal EGFP, decreasing salt tolerance (SD medium)...... 223

Figure 49. Secondary structure and subcellular localization of NHX proteins .. 225

Figure 50. Secondary structure and subcellular localization of HKT1 ...... 227

Figure 51. Secondary structure and subcellular localization of KAT2 ...... 229

Figure 52. Secondary structure and subcellular localization of Ol_pep_trans ... 230

Figure 53. Secondary structure and subcellular localization of PQ ...... 231

Figure 54. Secondary structure and subcellular localization of Eq_trans ...... 232

Figure 55. Secondary structure and subcellular localization of An_Unkn ...... 233

Figure 56. Secondary structure and subcellular localization of NacNeu3...... 234

Figure 56. Secondary structure and subcellular localization of NacNeu3...... 236

LIST OF TABLES

Table 1. Na+ and K+ content in plants with different NHX expression and NHX cation transport properties ...... 55

Table 2. Cation specificity for the Na+ dependent dissipation of the proton gradient in tonoplast vesicles from Salicornia bigelovii ...... 63

Table 3. Affinities and velocities of the vacuolar Na+/H+ antiporter activity in different species ...... 64

Table 4. Genome sequencing and sequence depth of coverage ...... 75

Table 5. Salicornia brachiata genome assemblies statistics ...... 82

Table 6. Salicornia europaea genome assemblies statistics ...... 83

Table 7. Salicornia brachiata genome protein prediction completeness assessment by BUSCO v3 ...... 84

Table 8. Salicornia europaea genome protein prediction completeness assessment by

BUSCO v3 ...... 85

Table 9. Comparison of genome protein prediction completeness assessment by

BUSCO v3 ...... 87

Table 10. Summary statistics for the transcriptome assemblies of S. bigelovii, S. brachiata, and S. europaea...... 135

Table 11. Completeness assessment of predicted proteins from the transcriptomes of S. bigelovii, S. brachiata, and S. europaea by BUSCO...... 136 25

Table 12. Summary statistics for the transcriptome assembly of S. bigelovii shoots for DE analysis...... 141

Table 13. Completeness assessment of transcripts and predicted proteins from the transcriptome of S. bigelovii shoots for DE analysis by BUSCO ...... 142

Table 14. Number of differentially expressed genes in shoots relative to 0 mM

NaCl...... 145

Table 15. Overrepresented MapMan bins for differentially expressed genes from shoots of S. bigelovii ...... 147

Table 16. Completeness assessment and comparison of the combined Salicornia dataset used for proteomics studies ...... 167

Table 17. Classification of proteins based on their differential abundances and their directionality across treatments identified with SCAFFOLD v4...... 169

Table 18. Overrepresented MapMan bins for differentially abundant proteins from shoots of S. bigelovii ...... 171

Table 19. Number of protein markers per organelle for Salicornia and Arabidopsis.

...... 179

Table 20. Organellar subcellular localization allocation by pRoloc of S. bigelovii proteins ...... 182

Table 21. Number of proteins in S. bigelovii shoot transcriptome with predicted transmembrane helices ...... 195

Table 22. Selected genes for characterization in yeast ...... 199 26

Table 23. Amplification and cloning of selected genes for characterization in yeast

...... 209

Table 24. Media and treatments used for yeast spot assays for all gene constructs

...... 214

Table 25. Genes affecting AXT3 growth under NaCl ...... 216

Chapter 1. Literature review

1.1 Introduction

Agriculture has been, and will continue to be, the primary source for the production of human nourishment. Unfortunately, food production and distribution has not been sufficient to feed the global population. Of the 7.2 billion people living on the planet, it is estimated that 842 million are undernourished

(UN-DESA, 2013).

The world population is expected to increase from currently 7.2 billion people to an estimated 9.6 billion people by 2050 and will increase mainly in developing countries (UN-DESA, 2013). In order to decrease undernourishment and cope with the growing population, agricultural production must increase accordingly. Highly productive agriculture demands good quality soil and large amounts of fresh water, e.g. wheat needs approximately 1350 liters of water per kg of grain produced (UNESCO, 2006). As a result, most of the freshwater resources are dedicated to agriculture, which uses approximately 70% of the freshwater resources of the world. Moreover, in developing countries, water can be employed for agriculture in up to 95% of its availability (FAO, 2007).

One way to increase agricultural production is by increasing the yield of crops. In the case of cereals it was postulated that the yearly increment in their production rate must be raised by 38% if we want to reach the demand projected 28

for 2050 (Tester & Langridge, 2010). However, how this yield increment can be achieved is more than questionable, mainly because yield increments have been constant over the past 50 years. Moreover, in the case of wheat, it is already very close to its theoretical maximum harvest index (Austin, 1982), thus limiting the room for improved crop yield. Similar cases are found for maize, rice, and soybean, where the projected yield increase falls behind the required production for 2050 (Fig. 1) (Ray et al., 2013).

Figure 1. Projections of global production for maize, rice, wheat, and soybean. Closed circles represent global yield for the period of 1961 – 2008. Solid lines represent the global projection for each crop and dashed lines represent the required global production if no additional lands are added to agriculture. Figure from: Ray et al. (2013). 29

A different approach would be to increase the area of agricultural lands.

The Food and Agriculture Organization of the United Nations (FAO) projects that food production must increase by 70% to meet the demand of 2050, which would correspond to an additional 130 million hectares of agricultural lands (FAO, 2009).

Furthermore, arable land per capita has decreased by almost 50% over the past 50 years and is expected to decrease even further by 2050 (FAOSTAT, 2006).

An alternative to increasing agricultural production without the necessity of the addition of arable lands, would be to decrease crop losses. Abiotic stresses are major contributors to crop yield losses. It was estimated that these factors annually decrease crop yield by 51-82% of their potential yield (Bray et al., 2000).

Of the different abiotic factors, drought and salinity contribute the most to crop losses, reducing yield globally by up to 50% (Bray et al., 2000). It is estimated that

10% of the total arable land on the planet is affected by drought and salinity

(Bartels & Sunkar, 2005) and every year 1-2% of irrigated lands are lost to soil erosion. In some scenarios, it is expected that salinity will affect up to 50% of the arable lands by 2050 (FAO, 1996).

Many crops are relatively sensitive to salinity, with most of them being severely affected by concentrations of 50 – 200 mM NaCl (sodium chloride)

(Munns & Tester, 2008). Some of the factors that have led to increased soil salinity include: bad agricultural practices such as over-irrigation, which leads to increased water tables salinizing the surface soils; irrigation with water containing 30

high amounts of salt, which leads to soil salinization once water is evaporated; and poor drainage, which increases the susceptibility to flooding and reduces salt leaching from the soil (FAO, 1985). It is estimated that about 1.1 Gha of land is affected by salt (Fig. 2) (Wicke et al., 2011). Furthermore, due to salinization of arable lands, the increase in agricultural production must come, at least in part, from the utilization of saline soils (Munns et al., 2012). Therefore, increasing the salinity tolerance of crops may not only diminish yield losses, but may also enable farmers to reclaim lands lost to soil erosion and salinization.

If salt tolerant crops could be developed which could grow with partially desalinized water or brackish water, agriculture’s fresh water constraints could be ameliorated, provided there was proper drainage and water management. This would not only mean the release of fresh water from agriculture for other applications (e.g. human consumption and industrial use); but, would also mean the potential access to virtually unlimited water resources. The identification and characterization of the mechanisms with which plants tolerate salinity will aid in the creation of crops that can thrive in saline environments, will reduce yield losses, and will build on the path of unlocking and recovering saline soils for agriculture and thus, increasing the overall agricultural output. 31

Figure 2. Salt affected land in the world. Lands are classified as: saline, with an electric conductivity (EC) between 2 - 16 and an exchangeable sodium percentage (ESP) of less than 15; sodic, with an EC of < 4 and an ESP between 15 – 40; and saline-sodic, with an EC > 4 and ESP > 15. Figure from: Wicke et al. (2011). 32

1.2 Salinity tolerance mechanisms of land plants

Sodium (Na) and potassium (K) are among the most abundant elements in the continental crust, ~2.4 % and ~2.1 % respectively (Hans Wedepohl, 1995).

However, only ~2 % of the K is considered to be easily available for crops (Zörb et al., 2014). Moreover, in the oceans, Na is ~27 times more abundant than K (Riley &

Tongudai, 1967). Both, Na and K are taken up by plants in their charged states: sodium cation (Na+) and potassium cation (K+) (Jacoby & Hanson, 1985, Maathuis

& Sanders, 1996, Tester & Davenport, 2003, Zhang et al., 2009a). It is interesting to note that, despite the greater availability of Na, K+ has been selected as the prevalent monovalent cation in the cytoplasm of most living cells (Benito et al.,

2014), with cytoplasmic levels of K+ being ~100 - 200 mM (Leigh & Jones, 1984) and of Na+ < 30 mM (Amtmann & Sanders, 1998, Tester & Davenport, 2003). It has been suggested that when the K+/Na+ ratio is altered in favor of Na+, plant cells experience Na+ toxicity (Maathuis & Amtmann, 1999). Plants have different levels and mechanisms of salinity tolerance (Munns & Tester, 2008, Roy et al., 2014), suggesting that salinity tolerance is not controlled by a single response but by a mix of different and probably independent processes. When exposed to salinity, plants are considered to undergo two main phases: shoot ion independent toxicity and shoot ion toxicity (Fig. 3). 33

Figure 3. Plant responses to salinity occur in two phases: Shoot ion independent (early); and shoot ion toxicity (later). Both phases have a detrimental effect on growth. Modified from: Munns and Tester (2008).

1.2.1 Shoot ion independent toxicity

The shoot ion independent toxicity phase is the first to appear, once a high enough concentration of Na+ surrounds the roots, leading to decreased shoot growth. This reduction in growth has been observed to be triggered by Na+, independently of the increased osmotic pressure of the soil during salinity (Termaat et al., 1985), implying that the Na+ ions themselves trigger a response leading to growth arrest. 34

Interestingly, shoot growth is reduced within minutes of salinity stress (Fricke,

2004), before Na+ can accumulate in the shoot, suggesting that there is signal traveling from roots to shoots early during salinity stress.

Long distance cellular signaling

It is not well understood how plants sense Na+; however, cellular signaling in response to Na+ has been linked with calcium (Ca2+) waves, which have been reported in Arabidopsis thaliana (Arabidopsis) seedlings when exposed to NaCl stress (Tracy et al., 2008, Choi et al., 2014, Schmöckel et al., 2015). These Ca2+ waves could interact with the CBL-CIPK network, where calcineurin B-like (CBL) proteins interact with CBL-interacting protein kinases (CIPK), activating transporters and regulating transcription to maintain ion homeostasis inside the cell (Xu et al., 2006, Liu et al., 2013, Sanyal et al., 2015). Interestingly, CBL proteins have also been observed to induce the production of reactive oxygen species (ROS) in Arabidopsis roots (Kimura et al., 2012, Drerup et al., 2013).

ROS waves have been reported mainly in response to wounding (Mittler et al., 2011). Yet, the disruption of the ROS inducible transcription factor SERF1 in rice leads to salinity susceptibility during the early response to NaCl stress

(Schmidt et al., 2013), suggesting that ROS might also be involved during salinity stress. Interestingly, Ca2+ waves move at a speed of 400 µm s-1 (Choi et al., 2014), whereas ROS waves move at a speed of 140 µm s-1 (Mittler et al., 2011). This could 35

propose for Ca2+ as the primary signal during salinity stress followed by a ROS response. Interestingly, Ca2+ waves in response to NaCl in Arabidopsis roots were slower when roots were treated the ROS scavenger ascorbate (Evans et al., 2016), suggesting ROS as support for Ca2+ waves propagation. It is not clear how these waves propagate from cell to cell (Choi et al., 2017). In the case of ROS waves, apoplastic diffusion of ROS could be hindered not only by cell walls but could be quenched by its antioxidative environment (Choi et al., 2017). ROS could be transported through the symplast. However, the respiratory burst homolog

(RBOH) protein, which is thought to be required for the propagation of ROS waves

(Evans et al., 2016), is not particularly abundant in the plasmodesmata (Fernandez-

Calvino et al., 2011). Alternatively, other signaling molecules such as Ca2+ could induce ROS. Ca2+ has been implicated in the induction of RBOH (Dubiella et al.,

2013, Kadota et al., 2015) leading to ROS. Ca2+ waves have been also considered to propagate via the symplast. However, modelling suggests that the speed of Ca2+ waves cannot be explained by simple diffusion (Evans et al., 2016), suggesting that

Ca2+ wave propagation might involve Ca2+ channels in the plasma membrane or endoplasmic reticulum (Gilroy et al., 2016).

Water uptake and transpiration

For roots to uptake water from the soil, roots must have a more negative water potential than the soil. To adjust their osmotic pressure, roots can use ions and/or 36

synthesize compatible solutes (Rodriguez et al., 1997, Hasegawa et al., 2000,

Munns et al., 2016).

Studies in which wheat and barley were exposed to increased salinity suggested that reduction of transpiration due to stomatal closure was a major factor contributing to decreased growth (Harris et al., 2010). Thus, transpiration use efficiency (TUE), defined as biomass produced per water transpired, becomes increasingly important during salt stress. Transpiration use efficiency is a complex trait, making it difficult to identify genes associated with it, nevertheless, TUE

QTLs were found in rice using genome wide association mapping (Al-Tamimi et al., 2016). It was observed that the highly salt tolerant plant quinoa had reduced stomatal density under saline conditions (Shabala et al., 2013). Reduced transpiration and reduced stomatal density was also observed in salt tolerant strawberries compared to a salt sensitive variety (Orsini et al., 2012). Congruently, overall photosynthesis is reduced during salinity stress (Sultana et al., 1999, Yeo et al., 1985, Maimaiti et al., 2014, He et al., 2016). This reduction in photosynthesis leads to an increased production of ROS, which are toxic to the cell. The increased production of ROS is countered by ROS scavenging enzymes such as peroxidases, dismutases, and superoxidases (Mittler et al., 2004, Asada, 2006).

1.2.2 Shoot ion toxicity

The second phase of salinity stress is ion toxicity, which is the consequence of an increased shoot ion concentration inside cells. It can start days or even weeks after exposure to salt (Munns & Tester, 2008), leading to premature leaf senescence.

This toxicity is a consequence of Na+ being detrimental to cells by: competing with K+ for enzyme activation sites (Blaha et al., 2000, Greenway &

Osmond, 1972); inhibiting protein synthesis by competing with K+ in the ribosomes and disrupting ribosomal structures (Flowers & Dalmond, 1992); interacting more strongly with water than K+, affecting water-protein interactions

(Collins, 1997, Hess & van der Vegt, 2009, Benito et al., 2014); and by Na+ accumulation in the chloroplasts leading to impaired photosynthesis (Müller et al.,

2014). Preventing Na+ from accumulating in the shoots is an important mechanism, to prevent damage to the photosynthetic machinery and sustain growth. In agreement, negative correlations between Na+ accumulation in leaf 3 and salinity tolerance have been found in tetraploid wheat (Munns & James, 2003).

Na+ exclusion from the shoots

It is considered that Na+ enters the roots through non selective cation channels

(Davenport & Tester, 2000, Demidchik & Tester, 2002). Once inside the plant, Na+ moves through the cortex into the endodermis and eventually reaches the transpiration stream. Roots must exclude ~98% of Na+ and Cl- to prevent toxicity 38

(Munns, 2005); hence, it is believed that roots must mobilize Na+ out of the plant and store it in the vacuoles to prevent Na+ from reaching the shoots (Munns &

Tester, 2008). Congruently, Lauchli et al. (2008) reported concentration gradients for Na+ and K+ along the roots of durum wheat. Higher concentrations of Na+ in the epidermal layer of the roots and K+ accumulation around the endodermis were also observed, suggesting for a mechanism preventing Na+ transport to the shoots while promoting K+ transport from the roots to the shoots.

At a cellular level, Na+ extrusion from cells into the apoplast has been commonly attributed to the plasma membrane Na+/H+ antiporter, SOS1 (Shi et al.,

2000). SOS1 expression is present, although at very low levels, in the root tip, root endodermis, and stele of shoots and roots (Shi et al., 2002). Downregulation of

SOS1 in Eutrema salsugineum (previously Thellungiella salsuginea) led to increased

Na+ in roots, suggesting a role for SOS1 in the prevention of Na+ accumulation in roots (Oh et al., 2009).

When Na+ influx is greater than Na+ efflux, Na+ consequently accumulates in the roots. This leads to Na+ reaching the xylem, where it can be either transported into the leaves or retrieved from the xylem. Na+ retrieval from the xylem has been attributed to the action of proteins from the high-affinity potassium transporters family (HKT). HKT proteins belonging to subclass I are transporters that can import Na+ (Davenport et al., 2007). Arabidopsis HKT1;1 has been shown to be involved in Na+ retrieval from the xylem, preventing Na+ 39

accumulation in leaves (Sunarpi et al., 2005, Davenport et al., 2007). Stelar specific

AtHKT1;1 expression in transgenic lines of the Arabidopsis C24 ecotype increased

Na+ concentration in roots and decreased it in shoots compared to the wild type

(Moller et al., 2009). Similarly, the introduction of the Nax2 locus from Triticum monococcum, containing, among other genes, an HKT (TmHKT1;5-A) which is the homolog in cereals of the Arabidopsis HKT1;1, into wheat, decreased shoot Na+ content and increased yield under highly saline conditions compared to near isogenic lines without this locus (Munns et al., 2012). This suggests the importance of HKT in shoot Na+ exclusion.

Tissue tolerance

Once Na+ has reached the shoots, it must be either transported to older leaves, transported out into the apoplast, or compartmentalized inside the cell in order to reduce toxicity. The ability of plants to cope with high concentrations of Na+ in the shoot is called tissue tolerance.

One mechanism of tissue tolerance, at the level of the whole shoot, is to redirect undesired Na+ to older tissues. It was observed that barley plants can target Na+ accumulation to older leaves via different ion import rates through the xylem and phloem, thus maintaining a higher K+/Na+ in young leaves (Wolf et al.,

1991). This protects the younger leaves, which are the most photosynthetically active tissues in the plant. Another mechanism of tissue tolerance consists of preventing Na+ from reaching the photosynthetically active leaf blade. OsHKT1;4 40

expression in rice leaf sheath has been negatively correlated with leaf blade Na+ accumulation (Cotsaftis et al., 2012). Congruently, downregulation of OsHKT1;4 led to an increased Na+ accumulation in the aerial part of the plant (Suzuki et al.,

2016).

Another mechanism of tissue tolerance is the extrusion of Na+ from the cell.

However, it can be intuitive to consider that when the concentration of Na+ in the leaves is high, extrusion of Na+ from the cell into the apoplast can become a redundant cycle - i.e. because extruded Na+ will re-enter the cell since Na+ cannot be transported out of the aerial part of the plant as in roots. Additionally, the volume of the apoplast is relatively small, limiting the amount of Na+ that it can hold. Hence, Na+ must be compartmentalized into unaffected organelles inside the cell. This is thought to be mainly taking place in the vacuole (Rains, 1972), where the highest amount of Na+ inside the cell has been found under saline conditions

(Lv et al., 2012b).

Vacuoles occupy the majority of the cellular volume and are involved in cellular ion homeostasis (Hasegawa, 2013). It has been proposed that a Na+/H+ antiporter is responsible for the translocation of Na+ into the vacuole at the expense of exporting H+ to the cytosol. This antiport activity was first measured in sugar beet (Blumwald & Poole, 1985) and was attributed to NHX1 proteins

(Apse et al., 1999). However, Barragan et al. (2012) showed that NHX1 and NHX2 41

proteins have equal affinity for K+ and for Na+, suggesting them to be K+ transporters since concentrations of K+ are much higher in the cytosol than of Na+.

The vacuolar Na+/H+ antiporter activity needs to be provided with sufficient H+ to be used as the driving force for Na+ import into the vacuole.

Congruent with this model, the introduction of an allele leading to an increased expression of the vacuolar H+ pyrophosphatase gene HVP10 in barley from a wild species, decreased shoot Na+ and increased shoot biomass under salinity stress

(Shavrukov et al., 2013), presumably by providing more H+ resourced for the vacuolar Na+/H+ antiporter. However, the gene encoding for the vacuolar

Na+/H+ antiporter remains to be identified. The mechanism and proteins involved in the vacuolar Na+/H+ antiport activity will be further discussed.

1.3 The role and mechanisms of Na+ accumulation in the vacuoles

Under high salinity conditions, Na+ will eventually accumulate to detrimental levels in the shoot. In order to avoid toxicity, Na+ can be accumulated in the vacuole. Vacuoles are organelles with diverse functions. They occupy up to 90% of the volume of the cell (Wink, 1993) and are involved in different processes, including: regulation of cell turgor (Reisen et al., 2003); protein lysis (Paris et al.,

1996); ion homeostasis (Gobert et al., 2007); and storage of proteins (Hoh et al.,

1995). Compartmentalization of Na+ into the vacuole would not only prevent ion toxicity in the cytosol but would aid in the creation of enough negative water 42

potential to maintain cell turgor (Hasegawa et al., 2000). Ion compartmentalization in the vacuoles has been reported in plants exposed to high salinity concentrations

(Harvey et al., 1981, Binzel et al., 1988) and is considered to be an important mechanism for salinity tolerance. The accumulation of Na+ into the vacuole has been associated with the action of three proteins: the Na+/H+ antiporter, the vacuolar adenosine triphosphatase (V-ATPase), and the vacuolar pyrophosphatase (V-PPase).

1.3.1 Ion transport across the tonoplast

Ions accumulate in the cytoplasm at different concentrations: Na+ and Cl- ions range from 1 to 30 mM; K+ from 100 to 200 mM; and free Ca2+ from 100 to 200 nM

(Binzel et al., 1988, Bush, 1995, Tester & Davenport, 2003). In contrast to uncharged solutes, which depend on concentration alone for directing their movement, the energetics of ion movements are affected by the difference on electrochemical potential across the membrane. The potential established across a membrane caused by the different concentrations of one ion on opposite sides of the membrane plus the difference in electrical potential across the membrane, can be calculated by the Nernst equation (Eq. 1). This equation also allows the calculation of ion concentrations based on a particular membrane potential if the ion were to be in electrochemical equilibrium. For a more realistic representation of the cell, the Goldman-Hodgkin-Katz equation takes into consideration several ions species

(Eq. 2). 43

푅푇 [푥0] 퐸푥 = × ln 푧퐹 [푥푖]

Equation 1. Nernst equation describing the direction of the net diffusion flux of an ion based on a membrane potential. Where Ex = membrane potential in mV; R = gas constant; T = temperature in Kelvin; z = ion valence; F = Faraday’s constant;

X0 = ion concentration outer side of the membrane; and Xi = ion concentration inner side of the membrane.

푛 + 푚 − ∑ 푃 + [푀 ] + ∑ 푃 − [퐴 ] 푅푇 푖 푀푖 푖 표푢푡 푗 퐴푖 푗 푖푛 퐸푥 = × ln ( 푛 + 푚 − ) 푧퐹 ∑ 푃 + [푀 ] + ∑ 푃 − [퐴 ] 푖 푀푖 푖 푖푛 푗 퐴푖 푗 표푢푡

Equation 2. Goldman-Hodgkin-Katz voltage equation describing the direction of the net diffusion flux taking into account different ions and their permeabilities, based on a membrane potential. Where Ex = membrane potential in mV; R = gas constant; T = temperature in Kelvin; z = ion valence; F= Faraday’s constant; PMi+ = ion permeability of ith cation; Mi+ = concentration of the ith cation; PAj- = ion permeability of jth anion; Aj- = concentration of the jth anion.

If free diffusion in cells were to occur based solely on membrane potentials, monovalent ions such as Na+ and K+ would accumulate in the cytoplasm at concentrations 100 to 1000 times greater than the observed concentrations (Nobel, 44

1991). However, ions cannot easily permeate through the membrane’s lipid bilayer and its influx and efflux into and out of the cell is mediated by transporters.

Three types of transporters exist: channels, pumps, and carriers (Sussman

& Harper, 1989). Na+ influx into the cell is considered to occur through nonselective cation channels (Tyerman & Skerrett, 1998, Davenport & Tester, 2000,

Pottosin & Dobrovinskaya, 2014). However, channels mediate a form of passive transport across a membrane. Considering the tonoplast to have an electrical potential of +40 mV relative to the cytosol, the Na+ concentration inside the vacuole would be around 4.8 folds lower than that of the cytosol based on the

Nernst equation (at 20 ºC). Nevertheless, higher concentrations of Na+ in the vacuole than in the cytosol have been observed in many plant species, such as

Suaeda maritima (Yeo & Flowers, 1986), Suaeda salsa, and Kalidium folium (Zhao et al., 2005, Flowers & Colmer, 2008).

If Na+ transport was only passive, the accumulation of high concentrations of Na+ in the vacuole would be impossible. However, Na+ accumulation in the vacuole against the tonoplast membrane potential has been repeatedly reported under high salinity conditions (Spanswick & Williams, 1964, Harvey et al., 1981,

Adams et al., 1992, Lv et al., 2012b), suggesting an active transport of Na+ into the vacuole. Several proteins have been proposed to be associated in the ion transport across the tonoplast, these proteins will be discussed in the following sections. 45

1.3.2 Proteins involved in Na+ accumulation in vacuoles

Accumulation of Na+ in the vacuole relies on active transport. This transport is believed to be directed by the vacuolar Na+/H+ antiporter. The activity of this antiporter relies on the availability of H+ to be used as the driving force for H+ translocation into the cytosol and Na+ into the vacuole (Blumwald & Poole, 1985).

Two proteins are considered to be involved in the supply of H+ to the vacuole: the vacuolar ATPase and the vacuolar pyrophosphatase (Fig. 4). 46

Figure 4. Representation of Na+ influx and mobilization within a cell in response to an elevated Na+ concentration in the apoplast. Na+ enters the cytosol through non-selective cation channels (a). Intracellular Na+ can be either exported out of the cell through proteins such as plasma membrane antiporters (b) or compartmentalized into the vacuole through vacuolar antiporters (c). The vacuolar antiport activity relies on the H+ gradient provided by the V-ATPase (d) and V-PPase (e). 47

1.3.2.1 The vacuolar Na+/H+ antiporter

It is considered that most ion fluxes in plant cells are dependent on the electrochemical transport of H+ across the bilayer membrane (Poole, 1978). In vacuoles, whose pH is more acidic than the cytoplasm, H+ are used in an antiport mechanism providing the necessary driving force for the transport of ions and solutes into the vacuole in exchange of transporting H+ into the cytoplasm (Tanner

& Caspari, 1996, Sze et al., 1999).

The vacuolar Na+/H+ antiporter follows the same mechanism and is localized in the tonoplast of the vacuole. Its activity was first described in Beta vulgaris (Blumwald & Poole, 1985, Blumwald & Poole, 1987). Blumwald and Poole

(1985) reported that the H+ dependent quenching of acridine orange loaded in vacuole enriched vesicles was reversed by the addition of Na+ to the medium; this increase in fluorescence was hypothesized to be caused by the Na+ dependent dissipation of the proton gradient. This observation suggested a Na+/H+ exchange.

Furthermore, the same study showed that this H+ translocation was membrane potential independent and that it obeyed the typical Michaelis Menten enzymatic kinetics (Blumwald & Poole, 1985), indicating that the reaction gets saturated, suggesting an active form of transport in the form of a Na+/H+ antiporter.

This vacuolar Na+/H+ antiporter activity was subsequently observed in many other plant species such as barley (Fan et al., 1989), sunflower (Ballesteros et 48

al., 1997), rice (Fukuda et al., 1998), and cell cultures of Populus euphratica (Silva et al., 2010). Differences in activities of these transporters across plants have been reported. The salt tolerant Plantago maritima showed an increased antiporter activity upon salt stress, whereas the salt sensitive Plantago media showed no antiporter activity after exposure to 50 mM NaCl (Staal et al., 1991). In Gossypium hirsutum and the halophyte Atriplex nummularia an increase in the vacuolar

Na+/H+ antiporter activity was shown under salinity stress and an additive effect of Na+ and K+ at saturating concentrations was observed, suggesting that different antiporters might be present in the tonoplast (Hassidim et al., 1990). Furthermore, increased Na+/H+ antiporter activities induced by increased salinity have been reported in halophytic species such as Thellungiella salsuginea (Vera-Estrella et al.,

2005), Mesembryanthemum crystallinum (Barkla et al., 2002), and Salicornia bigelovii

(Parks et al., 2002).

The vacuolar Na+/H+ antiporter requires H+ in order to translocate Na+ into the vacuole. Two different classes of proteins are considered to be involved in the provision of sufficient H+ required for the vacuolar Na+/H+ antiporter activity: V-

ATPase and V-PPase.

Vacuolar ATPase

H+ translocating ATPases hydrolyze ATP into ADP and one inorganic phosphate, releasing one H+ and translocating it to the other side of the membrane (Sze, 1984).

Interestingly, it was first hypothesized that this protein was an ion transporter due 49

to its increased activity in the presence of K+ and Na+ (Rungie & Wiskich, 1973,

Sze, 1980). The main function of H+ translocating ATPases is the provision of the necessary H+ to one side of the membrane in order to maintain the membrane potential and the electrochemical gradient of H+, which acts as the main driving force for the transport of ions and metabolites across the membrane.

V-ATPases have two main domains; the membrane integral domain V0; and the membrane peripheral domain Vi. In total, 13 subunits form a V-ATPase, and the genes encoding for these subunits range from 14 in Chlamydomonas to 54 in soybean (Schumacher & Krebs, 2010). V-ATPases can account for 33-35% of the total amount of proteins in the tonoplast (Klink et al., 1990). Mutants deficient in plasma membrane ATPases can be lethal in animals (Allan et al., 2005) and in plants (Strompen et al., 2005); although not always (Krebs et al., 2010).

Upregulation of these proteins have been reported in both halophytes and glycophytes under increased salinity conditions (Hanitzsch et al., 2007, Popova &

Golldack, 2007). Similarly, overexpression of V-ATPase subunits in plants have led to increased salinity tolerance based on yield (Baisakh et al., 2012) and biomass production (Wang et al., 2011, Dabbous et al., 2017), whereas downregulation leads to salinity hypersensitivity based on seedling survival (Batelli et al., 2007).

The increased salinity tolerance has been mainly attributed to the maintenance of the H+ gradient, provided by these enzymes for the Na+/H+ antiport activity. 50

Vacuolar pyrophosphatase

The V-PPase hydrolyzes pyrophosphate (PPi) into two inorganic phosphates, releasing one H+, which is translocated from the cytosol into the vacuole, maintaining an electrochemical gradient of H+ across the tonoplast (Rea & Poole,

1985). In contrast to V-ATPase, V-PPase has only been identified in plants, although plasma membrane PPases can be found in some bacteria (Zhen et al.,

1997, Drozdowicz & Rea, 2001). In contrast to the complex assembly of ATPases, the V-PPase consists of one single domain of 73 kDa (Tzeng et al., 1996). Two proteins have been found in Arabidopsis: AVP1, localized in the tonoplast

(Sarafian et al., 1992), and AVP2, localized in the Golgi apparatus (Shoji et al., 2010).

Overexpression of the Arabidopsis V-PPase (AtAVP1), leads to salinity and drought tolerance based on biomass and yield components (Gaxiola et al., 2001,

Pasapula et al., 2011, Schilling et al., 2014, Schilling et al., 2016). It was mainly considered that its role in salinity tolerance is by energizing the vacuolar Na+/H+ antiporter; however, Arabidopsis plants deficient of the V-PPase become unable to sustain heterotrophic growth, suggesting that the hydrolysis of pyrophosphate is also of importance for other cellular processes such as sucrose metabolism

(Ferjani et al., 2011). Additionally, plants overexpressing V-PPase have been observed to have greater Pinformed 1 (PIN1) auxin efflux proteins activity (Li et al., 2005) and increased root auxin content (Li et al., 2010). 51

Compatible solutes

A cellular response to stabilize water levels in the cytoplasm due to the vacuolar ion compartmentalization and maintain cell homeostasis is the production of compatible solutes. Increased Na+ accumulation in the vacuoles results in decreased vacuolar water potential. In order to maintain water levels in the cytosol and the other cellular compartments, cells must compensate by synthesizing organic compounds which act as osmolytes but that do not disturb the protein and enzymatic activities (Yancey et al., 1982). Such molecules are called compatible solutes.

The most common compatible solutes are glycine betaine, proline, and sugars, yet other osmoprotectants exist. Betaine, proline, and sugars have been seen to accumulate under increased salinity conditions in leaves of Elaeagnus oxycarpa (Maimaiti et al., 2014) and tomato leaves (Rivero et al., 2014), among others. Roles other than maintaining the osmotic potential in the cytoplasm have been also reported for compatible solutes: They have been associated with protection of extrinsic proteins of the Photosystem II complex from NaCl induced disassociation (Murata et al., 1992); with the provision of protection against ROS under NaCl stress (Shabala et al., 2012); and with the maintenance of membrane integrity (Quan et al., 2004). Nevertheless, the accumulation of compatible solutes leading to increased salinity tolerance in plants has yet to be shown (Negrão et al.,

2017). Interestingly, in many cases, a lower accumulation of compatible solutes is 52

found in the more salt tolerant relative compared to the more salt sensitive species

(Gharbi et al., 2017). Similarly, in the case of barley, varieties with increased salt tolerance have been shown to accumulate less compatible solutes (Chen et al.,

2007). It has been suggested that the sole use of compatible solutes to compensate for changes in osmotic pressure would be too energetically expensive for the plant

(Munns & Gilliham, 2015), which could indicate that salt tolerant species might use ions as osmolytes to compensate for osmotic changes outside of the cell. This strategy is thought to be employed by many halophytes (Flowers & Colmer, 2008).

NHX as the vacuolar Na+/H+ antiporter?

For many years the activity of the vacuolar Na+/H+ antiporter was described, but the identity of the protein responsible for this activity remained unknown. Apse et al. (1999) reported the first putative vacuolar Na+/H+ antiporter, named Na+/H+ proton exchanger 1 or NHX1. In that study, overexpression of NHX1 in

Arabidopsis lead to an increase in the Na+ dependent dissipation of the proton gradient under salt stress and increased the salinity tolerance of plants. Similarly, overexpression of AtNHX1 or its homologs in other plant species resulted in increased salinity tolerance based on plant dry weight (Wu et al., 2004), plant fresh weight (Chen et al., 2008), seedling survival (Brini et al., 2007), and plant growth

(Mishra et al., 2014). Such results seemed to support NHX as the vacuolar Na+/H+ antiporter. Further studies revealed a total of 8 NHX genes in Arabidopsis, divided 53

into two categories: intracellular proteins, class I (AtNHX1-4) localized in the tonoplast (Yokoi et al., 2002, Pardo et al., 2006) and class II (AtNHX5-6) localized in the endosome (Brett et al., 2005a, Bassil et al., 2011); and proteins localized in the plasma membrane, AtNHX7/SOS1 and AtNHX8 (Shi et al., 2000, An et al.,

2007).

However, further studies provided evidence for a different role for NHX1-

2. It was reported that Arabidopsis NHX1 and NHX2 have similar affinities for both Na+ and K+ (Venema et al., 2002, Barragan et al., 2012), or even greater for K+ as in the case of AtNHX1 expressed in yeast (Yamaguchi et al., 2005) (Table 1).

Considering that K+ concentrations in the cytosol are 10 times greater than Na+ concentrations, the role of NHX1-2 as Na+/H+ antiporters seems unlikely, and it has been hypothesized that instead has the role in planta of a vacuolar K+/H+ antiporter.

In line with this thought, compared to the wild type, the overexpression of the tomato antiporter LeNHX2 in tomato resulted in: an increased Na+/H+ antiport activity; an even greater K+/H+ antiport activity; increased salinity tolerance; and higher K+ but similar Na+ content under saline conditions (Huertas et al., 2013).

Similarly, tomato plants overexpressing Arabidopsis NHX1 showed increased vacuolar (Na+|K+)/H+ antiport activities, higher K+ uptake, increased K+ but not

Na+ content in the vacuoles of shoots and roots (Table 1), higher K+ content in xylem sap, roots, and shoots, increased salinity tolerance, and more severe K+ 54

starvation symptoms under K+ deficiency conditions compared to the wild type

(Leidi et al., 2010).

All the previous data suggested that K+ homeostasis was the main contributor to the increased salinity tolerance. Indeed, NHX1-2 proteins seem to be able to transport Na+ into the vacuoles; however, their selectivity for K+ and the abundance of this cation in the cytosol favors a K+/H+ rather than a Na+/H+ antiporter role in planta. Several other studies have associated NHX2 with K+ homeostasis (Rodríguez-Rosales et al., 2008, Xu et al., 2013, McCubbin et al., 2014).

Congruently, Arabidopsis nhx1/nhx2 knock-down double mutants had higher levels of cytosolic K+ (Table 1) and were more susceptible to K+ starvation but not more susceptible to salinity compared to wild type plants (Barragan et al., 2012). 55

Table 1. Na+ and K+ content in plants with different NHX expression and NHX cation transport properties. Organism of expression Vacuolar cation content (SEM-EDX counts) Leaf Root K+ Na+ K+ Na+ 150 mM NaCl1 Palisade Cortex WT Tomato 0.15* 0.75* 0.86 1.24* 35S::AtNHX1 Tomato 0.24* 0.36* 1.95 0.63* Stele Stele WT Tomato 0.13* 0.67 0.36* 1.60* 35S::AtNHX1 Tomato 0.44* 0.73 2.99* 1.08*

Cytosolic cation content (mM) Control Root epidermis K+ Membrane potential (mV) WT2 Arabidopsis 75 ± 14 -122 ± 14 Δnhx1/Δnhx2 Arabidopsis 112 ± 17 -132 ± 14 WT3 Tomato 75 ± 5 ~ -150 35S::LeNHX2 Tomato 43 ± 11 ~ -150 Antiporter activity

Km (mM) Vmax K+ Na+ K+ Na+ AtNHX11 Tomato 20.9 ± 14.7 22 ± 12.8 96.5 ± 19.9 116 ± 20 AtNHX14 Yeast 12.2 24.3 7.42 4.50 AtNHX22 Arabidopsis 35 35 - -

1(Leidi et al., 2010) * Significant difference between lines at a 0.05 p-value 2(Barragan et al., 2012) 3 (Huertas et al., 2013) 4(Yamaguchi et al., 2005) 56

It would seem that the increased salinity tolerance in plants overexpressing

NHX, is due to a greater uptake of K+, allowing maintenance of a higher K+/Na+ in the cells. Interestingly, K+ levels in the cytosol have been seen to decline only when K+ content in the vacuoles reached a certain threshold (Walker et al., 1996,

Leigh, 2001), thus storage of K+ in the vacuoles by an antiporter might contribute to the maintenance of K+ pools in the cells. Considering all the evidence and the fact that K+ levels are greater than Na+ levels in the cytosol, a K+/H+ antiport activity for NHX1-2 appears more likely. However, this then leaves the question: which protein is responsible for the vacuolar Na+/H+ antiporter activity?

Addressing this question is one of the major aims of this thesis.

1.4 Salicornia as a model plant for understanding salinity tolerance

Salicornia is a genus in the order Caryophyllales, the family Amaranthaceae

(formerly Chenopodiaceae), and the subfamily Salicornioideae. The genus

Salicornia comprises annual leafless plants with succulent stems leading to fruiting spikes (Fig. 5) that grow in salt marshes and coastal regions (Glenn et al., 1991).

Salicornia can grow in up to twice the salt concentration of seawater, making it one of the most salt tolerant of all plant species (Lv et al., 2012b, Ma et al., 2013, Ventura

& Sagi, 2013). The Salicornia genus karyotype consists of x = 9, with species having mainly two chromosome sets, comprising: diploid 2n = 18; or tetraploid 2n = 36 57

(Dalby, 1962). Its leafless stem and increased succulence are adaptive traits to help plants thrive in their saline environments (Delf, 1912).

Figure 5. Salicornia bigelovii physiology. Plants consist of succulent photosynthetic leafless stems which mature into seed-producing spikes. Salicornia bigelovii can grow directly in seawater. Photo source: https://www.ambientebio.it

Salicornia bigelovii is of particular interest due to its greater size, shoot thickness, seed yield, and seed size compared to other members of the same genus.

Salicornia bigelovii is a tetraploid plant (2n = 36), originally from the north of Mexico and the United States (Rivers & Weber, 1971b, Kadereit et al., 2007). The Salicornia life cycle usually takes from 8 to 10 months and field trials of Salicornia bigelovii showed that flowering and seed maturation occurs always in the same month

(mid-June) regardless of the sowing date (Glenn et al., 1991), suggesting that light and temperature play a crucial factor in the completion of its life cycle. 58

1.4.1 Salinity tolerance in Salicornia

Salicornia is considered to be a euhalophyte. Euhalophytes are plants that, in most of cases, need 50 - 200 mM salts for their optimal growth (Flowers et al., 1986).

Different germination behaviors have been seen in different Salicornia species at several salt concentrations. Generally low amounts of salt increase germination

(Ungar, 1967, Khan et al., 2000). In the case of Salicornia bigelovii, germination rate reached its maximum at seawater concentrations at temperatures between 4 – 15

ºC but, salinity reduced germination at higher temperatures (Rivers & Weber,

1971b).

Salicornia bigelovii has an optimal growth at around 200 mM NaCl (Webb,

1966, Ayala & Oleary, 1995, Brown et al., 1999, Ohori & Fujiyama, 2011). This is probably due to the use of Na+ ions as an energetically cheap osmoticum to maintain a high osmotic pressure of cells, helping in the maintenance of succulence

(Pardo & Quintero, 2002). Because of its growth in saline environments, Salicornia accumulates high amounts of salts in the shoots. But how does it avoid toxicity?

In order to avoid Na+ toxicity, many plants store Na+ ions in old leaves. The old leaves accumulating high concentrations of ions will wither and die, but will be replaced by photosynthetically active new leaves (Wolf et al., 1991). Other special adaptations exist in plants that thrive in saline environments (halophytes).

One of these adaptations is the formation of salt bladders. Salt bladders consist of 59

modified trichomes that accumulate salt in the shoots which break and release the salt once it reaches its full capacity (Flowers et al., 1977), thus preventing ion accumulation in photosynthetically active cells. However, Salicornia neither sheds leaves nor possesses salt glands as many other halophytes, suggesting that most of the Na+ that reaches the shoots is maintained in the shoot. Thus, Salicornia must have high tissue tolerance.

Shoot Na+ content in Salicornia growing at optimal conditions of 200 mM

NaCl reached concentrations of 133 mg.g-1 DW and showed an increase in shoot growth of 158% compared to plants grown at 5 mM NaCl (Ayala & Oleary, 1995).

Conversely, barley plants (e.g. cv Morex), one of the most salt tolerant cereal crops, grown at 200 mM NaCl reached shoot Na+ concentrations of 57 mg.g-1 (DW) and displayed an 80% fresh weight reduction compared to non-saline conditions

(Nguyen et al., 2013). Since enzymatic activities inside the cells of halophytes, such as Atriplex and Salicornia, and non-halophytes seem to be equally sensitive to salt concentration (Ball & Anderson, 1986, Greenway & Osmond, 1972), other mechanisms must take place in Salicornia which prevent cell ion toxicity.

Succulence is another adaptation of halophytes in which cells increase their size by retaining water, diluting Na+ and presumably allowing for its higher accumulation in the vacuole (Jennings, 1968). Salicornia displays this attribute. In

Salicornia europaea, accumulation of Na+ mainly takes place in the shoots, and Na+ is predominantly stored in the vacuoles of the endodermal cells of shoots (Fig. 6) 60

(Lv et al., 2012b), suggesting that Na+ compartmentalization into the vacuole is one of the main mechanisms in the prevention of Na+ toxicity in Salicornia.

Figure 6. Sodium accumulation in Salicornia europaea. a) Na+ content in shoots and roots of Salicornia europaea. b) Na element subcellular distribution in shoot cortex cells of

Salicornia europaea by TEM–X-ray microanalysis treatment with 200 mM NaCl for 21 days.

Figure from: Lv et al. (2012b). 61

1.4.2 Evidence of vacuolar Na+/H+ antiporter activity in Salicornia

Despite being highly salt tolerant, very few studies exist investigating the salinity tolerance mechanisms in Salicornia. Ayala et al. (1996) compared plasma membrane (PM-ATPase) and vacuolar ATPase (V-ATPase) activities in Salicornia bigelovii plants grown at 5 or 200 mM NaCl. Both ATPases showed increased activities in plants grown at 200 mM NaCl. However, this was not due to higher levels of ATPase protein.

In the same study, plants grown at 200 mM NaCl displayed a better growth and had higher Na+ concentrations in the shoots than those grown at 5 mM NaCl

(Ayala et al., 1996). It was found that in vesicles from plants grown at 200 mM

NaCl, V-ATPase was stimulated by Cl- and moderately reduced by the addition of

K+ in vitro, whereas the PM-ATPase showed an opposite behavior. Interestingly, the addition of Na+ in vitro did not stimulate PM-ATPase activity. In vitro assays in vacuolar enriched vesicles from plants grown at 5 mM NaCl, showed that V-

ATPase activity was greatly stimulated by the addition of 200 mM NaCl but not to KCl or Cl-BTP, indicating that the increase in the activity was caused by Na+ and not Cl-.

In a similar study, Parks et al. (2002) studied the V-PPase activity in

Salicornia bigeloviii grown at 5 and 200 mM NaCl. The hydrolysis of pyrophosphate increased 1.5- to 2-fold and V-PPase abundance measured by western blot 62

targeting a hydrophilic loop of the Arabidopsis V-PPase was higher in plants grown at 200 mM NaCl. However, the addition of 50 mM NaCl to vacuolar vesicles inhibited the V-PPase activity by 20% while other cations such as K+, rubidium

(Rb+), and cesium (Cs+) increased it. In the same study, the vacuolar Na+/H+ antiporter activity was indirectly measured by the Na+-dependent dissipation of the proton gradient. The results showed that there is an increased activity in plants grown at 200 mM NaCl in comparison to those grown at 5 mM (Fig. 7).

Figure 7. Na+ dependent dissipation of the proton gradient in vacuole vesicles from

Salicornia bigelovii with the addition of 50 or 4 mM K+ in the assay in plants grown at: 200 mM NaCl with 50 mM KCl ; 200 mM NaCl with 4 mM KCl ; and 5 mM NaCl with 50 mM KCl . Figure from: Parks et al. (2002).

Moreover, the vacuolar Na+/H+ antiporter activity seemed to be K+ independent (Table 2). Interestingly, the affinity and transport rate was higher 63

than for other reported species: rice (Barkla et al., 1995, Fukuda et al., 1998) and M. crystallinum (Fukuda et al., 1998) (Table 3).

Intriguingly, no differences were found in the Na+/H+ antiport activity in

Salicornia dolichostachya grown at 10 mM and 200 mM NaCl (Katschnig et al., 2014).

The discrepancy between the two studies could be due to differences between species, differences in the duration of the treatment, or differences in developmental stages. Alternatively, this antiport activity could be induced after a certain Na+ threshold, in this case of at least 10 mM NaCl.

Table 2. Cation specificity for the Na+ dependent dissipation of the proton gradient in tonoplast vesicles from Salicornia bigelovii grown in 200 mM NaCl. Values from Parks et al. (2002)

Cation*(Cl) Concentration mM Exchange activity (Δ%F min-1)

LiCl 1 14.7 ± 1.3 5 19.0 ± 5 NaCl 1 30 5 80 ± 0 KCl 1 or 5 No detectable activity RbCl 1 or 5 No detectable activity CsCl 1 or 5 No detectable activity BTPCl 1 or 5 No detectable activity

Table 3. Affinities and velocities of the vacuolar Na+/H+ antiporter activity in different species based on the Na+ dependent dissipation of the proton gradient.

Salicornia bigelovii1 M. crystallinum2 Rice3

Affinity mM (Km) 3.8 51 10

Vmax (Δ%F min-1 mg-1 3214 1007 2160 protein)

1 (Parks et al., 2002) 2 (Barkla et al., 1995, Fukuda et al., 1998) 3 (Fukuda et al., 1998)

1.5 Aim of the project

The aim of this project is to identify proteins of Salicornia bigelovii which are involved in salinity tolerance, with specific focus on the identification of proteins related to the compartmentalization of Na+ into the vacuole, e.g. a vacuolar

Na+/H+ antiporter. However, other proteins involved in salt tolerance might also be discovered. The project will also provide a better understanding of the responses of S. bigelovii to high concentrations of Na+ and Cl- in its shoots.

Additionally, the project aims to generate genomic and proteomics resources that will facilitate the development of future Salicornia research.

The identification of proteins involved in salt tolerance in S. bigelovii will hopefully contribute to the understanding of plant stress responses, and provide new genes for crop improvement, either by conventional breeding, by the identification of homologs in other species, or by genetic modification. 65

To achieve these goals, the following steps were taken:

1. The genome sequencing, assembly, and annotation of species from

the Salicornia genus.

2. The physiological and ionic characterization of S. bigelovii treated

with different concentrations of NaCl.

3. The transcriptomic and proteomic responses of S. bigelovii to

different concentrations of NaCl, allowing the identification of

candidate proteins involved in salt tolerance.

4. The generation of organellar membrane proteomes of S. bigelovii,

allowing the identification of tonoplast proteins and the further

selection of genes for characterization.

5. The characterization of selected candidate genes of S. bigelovii

involved in salt tolerance using the heterologous expression system,

yeast.

Chapter 2. Genome assembly and annotation of Salicornia species

Genomic resources are fundamental for research. A genome serves as the basis to enable the development of further studies - e.g. gene discovery, natural variation, plant adaptation to the environment, population genetics, and evolutionary analyses, among others. In this chapter, the generation of draft reference genomes is described for three different Salicornia species: Salicornia bigelovii, Salicornia brachiata, and Salicornia europaea.

2.1 Introduction

Genomic resources from plants that naturally thrive in highly saline environments have the potential to be valuable in the generation of salt tolerant crops; however, these resources have been largely unexplored. Only a handful of salt tolerant plants have been sequenced and made publicly available, such as Eutrema salsugineum (formerly Thellungiella halophila) (Yang et al., 2013), Schrenkiella parvula

(formerly Thullengiella parvula) (Dassanayake et al., 2011), Chenopodium quinoa

(quinoa) (Jarvis et al., 2017), and Mesembryanthemum crystallinum, which has been assembled and is expected to become available in the near future (Yim et al., 2017)

Eutrema salsugineum and Schrenkiella parvula are closely related to the model plant Arabidopsis thaliana. Both, Eutrema salsugineum and Schrenkiella parvula are able to survive NaCl concentrations of 500 mM, whereas Arabidopsis thaliana is severely affected between 150 – 200 mM NaCl. These two salt tolerant 67

species seem to be able to maintain lower Na+ and higher K+ in their leaves compared to their salt sensitive relative Arabidopsis thaliana (Orsini et al., 2010), suggesting that their salt tolerance might be in part related to their capacity to exclude Na+ from the photosynthetically active leaves.

Quinoa is capable of growing in highly saline conditions, in the range of

150 – 750 mM NaCl (Orsini et al., 2011). Its tolerance mechanisms are not well understood. It was initially suggested that the epidermal bladder cells present on its stem and leaves might function as salt bladders (Bonales-Alatorre et al., 2013).

However, only a modest increase in Na+ has been found in epidermal bladder cells upon salt treatment (Orsini et al., 2011), suggesting different mechanisms are driving its salt tolerance. Quinoa genotypes with increased salt tolerance accumulated less Na+ in its leaves under salinity treatment than less tolerant genotypes (Shabala et al., 2013). In contradictory reports, K+ was found to be increased upon salt treatment in quinoa leaves (Shabala et al., 2013) or unaltered

(Wilson et al., 2002). The discrepancies might be due to differences in salt treatments and genotypes used. Nevertheless, quinoa seems to have a high level of regulation of K+ homeostasis, as K+ is commonly decreased in the shoots upon salt treatment (Jha et al., 2010, Jaarsma et al., 2013). Mesembryanthemum crystallinum is a halophytic plant able to cope with high concentrations of NaCl. Its primary tolerance mechanism is the storage of Na+ in its epidermal cell bladders that act as salt bladders capable of accumulating Na+ concentrations exceeding 1 M (Adams et al., 1998). 68

It is indubitable that these genomes have the potential to greatly contribute to the understanding of salt tolerance in plants; nevertheless, different genomic resources are required as different salt tolerance mechanisms are expected to be used by different plant species. Salicornia are succulent annual leafless plants that belong to the order Caryophyllales, the family Amaranthaceae, and the subfamily

Salicornioideae. The Amaranthaceae family holds several salt tolerant plant species such as the Suaeda genus and quinoa. Salicornia grows in sea shores and salt marshes (Glenn et al., 1991) and has an optimum growth at 200 mM NaCl

(Ayala & Oleary, 1995) (Chapter 3). In contrast to some of the aforementioned plants, Salicornia can accumulate Na+ and Cl- at concentrations higher than 1 M in their photosynthetically active shoots (Lv et al., 2012b), while sustaining growth, without the use of salt bladders. This suggests that cells must either exclude or compartmentalize high concentrations of Na+ in compartments such as the vacuole. This tolerance mechanism is referred to as Na+ tissue tolerance (Munns &

Tester, 2008, Roy et al., 2014, Munns et al., 2016), posing Salicornia as an ideal candidate to understand cellular mechanisms of shoot Na+ tolerance in plants.

Unfortunately, there are not enough genomics resources available to allow the genomic study of any Salicornia species.

The Salicornia genus consists of plants with 9 chromosomes (x = 9) and have been identified as diploid (2n = 18) e.g. S. europaea (Jefferies et al., 1981, Kadereit et al., 2007), or tetraploid (2n = 36) (Dalby, 1962, Kadereit et al., 2007) e.g. S. bigelovii 69

(Kadereit et al., 2007). Salicornia has been considered to be a taxonomically challenging, as its relatively simple anatomy makes it problematic to distinguish between different species (Kadereit et al., 2007). Moreover, Salicornia names have been used loosely and sometimes interchanged, rendering the classification of species and subspecies difficult.

In this chapter, the genome assembly and annotation is described of three

Salicornia species: Salicornia bigelovii Torr.; endemic to Mexico and the United

States; Salicornia brachiata, endemic to India; and Salicornia europaea, endemic to

Europe, northern Africa, the Middle East, and central Asia. Using Illumina sequencing, different genomes were created with three different assemblers:

ABySS (Simpson et al., 2009), SOAPdenovo2 (Luo et al., 2012), and Platanus

(Kajitani et al., 2014). The best assemblies derived from each assembler were combined using Metassembler (Wences & Schatz, 2015), resulting in the reference genomes of S. brachiata and S. europaea. A rough draft was created for S. bigelovii by combining Illumina and SMRT Sequencing by PacBio using different hybrid assembling methods.

2.2 Materials and methods

2.2.1 Plant material, library preparation, and genome sequencing

S. brachiata seeds were collected by Dr. Muppala Reddy (KAUST) from dried plants naturally growing in the coastal areas of the west cost of Gujarat, India. S. bigelovii seeds were kindly provided by Dr. Edward Glenn of the University of

Arizona, USA. S. europaea seeds were kindly provided by Dr. Moshe Sagi of the

Blaustein Institute for Desert Research (BIDR). All Salicornia plants were grown in a greenhouse at King Abdullah University of Science and Technology (KAUST),

Saudi Arabia, under constant temperatures with a day night/cycle of 25/20 °C and a relative humidity of 50 ± 10%.

Total genomic DNA was extracted from the succulent fleshy stems of four month-old Salicornia plants using a DNeasy® Plant Maxi Kit (Qiagen) following the manufacturer’s instructions. The purity and concentration of the DNA was evaluated via NanoDrop (Thermo Fisher Scientific, USA) and Quibit 2.0

Fluorometer (Invitrogen, USA). For all three species, genome shotgun libraries were prepared with TruSeq DNA sample preparation kit v2 (Illumina, USA); PCR-

Free libraries were prepared with TruSeq DNA PCR Free LS Sample Prep Kit

(Illumina, USA); and Mate pair libraries with insert sizes of 3, 5, and 8 kb were prepared with Nextera Mate Pair Gel-Plus Sample Preparation system (Illumina,

USA). All Illumina libraries were prepared by Dr. Suhail Khan (KAUST) following the manufacturer’s instructions. Genome sequencing was carried out in a

Hiseq2000 system (Illumina Inc. USA) under paired-end mode with 101 bp read length by Core Labs (KAUST). For S. bigelovii, a library for single molecule real time (SMRT) sequencing (Pacific Biosciences, USA) was prepared by extracting 71

high molecular weight DNA with Qiagen Genomic DNA kit following the Tissue protocol. The resulting DNA was used for SMRTbell™ Library Preparation in about 20 µg per library. 20 SMRT Cells were run in a PacBio RS II system by Core

Labs (KAUST).

RNA samples from shoots and roots of four month-old plants of the three

Salicornia species were extracted with TRIzol™ with RNA MiniPrep kit (Zymo

Research, USA). RNA quality was determined with a Bioanalyzer 2100 using an

RNA 600 Nano kit (Agilent, USA). High quality of RNA was used to prepare

RNASeq libraries with TruSeq RNA Library Prep Kit v2 (Illumina, USA) according to the manufacturer’s instructions. RNASeq libraries were sequenced in 9 lanes of

Hiseq2000 system (Illumina, USA) under paired-end mode and 101 bp read length.

2.2.2 Read trimming and genome assembly

Illumina Genome shotgun, PCR-Free, Mate pair, and RNASeq reads had adapters removed and trimmed with Trimmomatic (Bolger et al., 2014) and Trim Galore!

(https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/), and quality was verified with FastQC

(https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). The trimmed reads, accounted for a sequence depth of coverage for S. brachiata of 114x for

Genome shotgun, 33x for PCR-Free, and 30x for Mate pair libraries. For S. europaea of 111x Genome shotgun, 18x for PCR-Free, and 30x for Mate pair libraries. And 72

for S. bigelovii 102x for Genome shotgun and 20x for PCR-Free libraries. SMRT sequencing reads were corrected and trimmed using Canu v1.5 (Koren et al., 2017), and depending on the trimming parameters, accounted for 2 - 8x coverage. The genomes of S. brachiata and S. europaea were assembled using the trimmed Illumina reads with three different assemblers and tested with different k-mers. Assemblies with ABySS (Simpson et al., 2009) and SOAPdenovo2 (Luo et al., 2012) used k- mers in odd numbers of sizes ranging from 61 – 87. Assemblies with Platanus

(Kajitani et al., 2014) used k-mers of 33, 41, 51, 61, 71, and 81. Genome assemblies were all gap closed with GapCloser from SOAPdenovo2. The best assemblies resulting from each assembler were merged together using Metassembler (Wences

& Schatz, 2015) using two of the three mate pair libraries at a time. All the different permutations of mate libraries were tested and the best merged assemblies were combined in another round of Metassembler. The genome of S. bigelovii was assembled either using only SMRT sequencing reads using Canu v1.5 and

SMARTdenovo (https://github.com/ruanjue/smartdenovo) or by hybrid assemblies using Illumina and SMRT sequencing reads with SPAdes (Bankevich et al., 2012) and DBG2OLC (Ye et al., 2016).

2.2.3 Genome selection, completeness, and annotation

Selection of the best genomes generated from different assemblers was done using two parameters: assembly contiguity and protein prediction completeness. 73

Assembly contiguity was assessed by statistics such as number of scaffolds, N50, and maximum scaffold length using BBTools (http://jgi.doe.gov/data-and- tools/bbtools/). Repeat families were de novo identified with RepeatModeler

(http://www.repeatmasker.org/RepeatModeler/) and each genome was masked with RepeatMasker (http://www.repeatmasker.org/) using the RepeatModeler library. Masked genomes were used to predict genes with BRAKER1 (Hoff et al.,

2016) using in silico normalized RNASeq reads by Trinity (Grabherr et al., 2011).

Reads were mapped to the genome with Tophat (Trapnell et al., 2009) as hints for intron-exon boundaries. Genome protein prediction completeness was assessed using BUSCO v3 (Simão et al., 2015) against the Embryophyta dataset and the genome with the best annotation was selected as the final assembly.

2.3 Results and discussion

2.3.1 Genome sequencing

The DNA was extracted from four month-old S. bigelovii, S. brachiata, and S. europaea plants and used for library preparation for Illumina sequencing. Three different types of libraries were generated: Genome shotgun, PCR-Free, and Mate pair libraries with insert sizes of 3, 5, and 8 kb. Illumina sequencing was carried out in a HiSeq2000 system under paired-end mode and read length of 101 bp.

Illumina reads were quality assessed and adapter trimmed using Trimmommatic

(Bolger et al., 2014) and Trim Galore! 74

(https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/). The precise genome sizes of S. bigelovii, S. brachiata, and S. europaea are unknown.

However, Kaligarič et al. (2008) sampled the Gulf of Trieste for different Salicornia plants and identified 2 classes of DNA through flow cytometry, one class belonging to diploid plants and another to tetraploids. 2C-values corresponding to diploid plants ranged from 1.069 to 1.090 pg and tetraploid plants ranged from

2.131 to 2.235 pg. This would be equivalent to one set of chromosomes having 0.534

– 0.545 pg of DNA or a genome size of 522 – 533 Mb. Assuming a set of chromosomes (x) of 530 Mb, would result in a predicted sequence depth of coverage of 122x, 177x, and 159x for S. bigelovii, S. brachiata, and S. europaea, respectively (Table 4). Different reports suggest that certain Salicornia species might be primarily autogamous (Dalby, 1962, Fernández-Illescas et al., 2011), particularly diploid species, while others seem to be allogamous (Fernández-

Illescas et al., 2011), predominantly tetraploid (Ferguson, 1964). Since Salicornia are non-domesticated wild species, allelic variation might add difficulties to its genome assembly, particularly to the tetraploid S. bigelovii.

Table 4. Genome sequencing and sequence depth of coverage

Salicornia bigelovii Salicornia brachiata Salicornia europaea

Ploidy Tetraploid (4x) Unknown Diploid (2x)

Estimated genome size (1 set of chromosomes) 522 - 533 Mb1 522 - 533 Mb1 522 - 533 Mb1

Illumina read bp after trimming | coverage 65,474,201,901 | 122 x 94,617,982,640 | 177 x 85,091,039,160 | 159 x

Genome shotgun libraries | coverage 54,508,694,929 | 102 x 60,581,145,149 | 114 x 58,927,506,281 | 111 x

PCR-Free libraries | coverage 10,965,506,972 | 20 x 17,626,958,411 |33 x 9,857,584,879 | 18 x

Mate pair libraries | coverage N/A 16,409,879,080 | 30 x 16,305,948,000 |30 x

SMRT sequencing (PacBio) 1,593,594,833 | 3 x N/A N/A

Total sequence depth coverage 125 x 177 x 159 x

1 Estimation from Kaligarič et al. (2008) 76

2.3.2 Genome assembly using Illumina reads

The genomes of S. brachaita and S. europaea were assembled with three different assemblers and using Genome shotgun, PCR-Free, and Mate pair libraries.

Genome assemblies were generated using ABySS (Simpson et al., 2009),

SOAPdenovo2 (Luo et al., 2012), and Platanus (Kajitani et al., 2014). Assemblies generated with ABySS and SOAPdenovo2 were created with odd k-mers ranging in size from 61 – 87. Due to the high requirement of memory allocation, assemblies generated with Platanus, which is an assembler designed for highly heterozygous diploid species, were created with k-mers 33, 41, 51, 61, 71, and 81. Gaps in the assemblies were closed with GapCloser from SOAPdenovo2. The resulting genome assemblies ranged from 500 – 600 Mb in size, with N50 from 50 – 170 kb and were highly fragmented ranging from 360,000 – 2,000,000 scaffolds.

Because different assemblers and assemblies have different strengths and biases, the best assemblies generated from each different assembler were merged into one with Metassembler (Wences & Schatz, 2015) (Fig. 8). Metassembler identifies conserved regions between assemblies by mapping mate pair reads with

Nucmer (Kurtz et al., 2004), extending scaffolds and discarding those that were only found in one assembly, which were likely erroneously generated. The selection of the best assemblies from each assembler was based on three different criteria: Best assembly statistics; best annotation; and a combination of both. The best assemblies generated from Metassembler where then merged into one by 77

Figure 8. Genome assembly strategy using Illumina reads.

another round of Metassembler to obtain the final assembly.

2.3.2.1 Genome statistics and protein prediction

In order to select the best assemblies from each individual assembler, assembly statistics were measured with BBTools (http://jgi.doe.gov/data-and- tools/bbtools/) and compared across assemblies generated with the same assembler. The best assemblies from each assembler were then merged with

Metassembler using two of the three mate pair libraries at a time. Each mate pair library permutation was tested and the best assembly was selected.

Genome N50 is not always well correlated with annotation completeness

(Simão et al., 2015), for this reason another criterion for the selection of each assembly was based on gene prediction completeness. Genes and their resulting proteins were predicted for each assembly using BRAKER1 (Hoff et al., 2016). For protein prediction, de novo repeat libraries were generated with RepeatModeler

(http://www.repeatmasker.org/RepeatModeler/) and masked using

RepeatMasker (http://www.repeatmasker.org/). RNASeq reads were in silico normalized using Trinity (Grabherr et al., 2011) and mapped to the genomes using

Tophat (Trapnell et al., 2009). The masked genomes were fed into BRAKER1, which calls GeneMark-ET (Lomsadze et al., 2014) and AUGUSTUS (Stanke & Waack,

2003) to create gene prediction models using RNASeq mapped reads to the genome for gene prediction and as hints for intron – exon boundaries. Proteins derived from 79

the predicted genes by BRAKER1 were assessed for completeness using BUSCO v3 (Simão et al., 2015) against the Embryophyta protein dataset (Appendix Tables

3 and 4).

2.3.2.2 Genome selection and annotation

The best assemblies generated from each individual assembler in terms of assembly statistics, annotation, and a combination of both, were merged using

Metassembler. The merged assemblies had improved genome assembly statistics for both S. brachiata (Table 5) and S. europaea (Table 6). In the case of S. brachiata, the best individual assembly statistics were found in genomes generated with

ABySS. Resulting in the smallest scaffold number of 300,000, an N50 of 173 kb, and the longest scaffold of 2.3 Mb. It is noteworthy to mention that these statistics were not found in any single genome assembly, but each trait was found in different assemblies.

After a first round of Metassembler, all the genome assembly statistics improved (Table 5). The number of scaffolds was reduced from over a million, in the case of Platanus, to just below 9,000 scaffolds. The longest scaffold improved from 2.3 to 2.5 Mb and the N50 from 173 kb to 254 kb, all in one genome. The best assemblies originated by merging 3 different ABySS assemblies with

Metassembler. The merged assemblies were then combined with another round of

Metassembler, resulting in the final S. brachiata assembly. The statistics of the final 80

genome assembly improved moderately, decreasing the scaffold number to 5,230, increasing the genome N50 to 302 kb, slightly decreasing the longest scaffold from

2.53 to 2.51 Mb, but decreasing the gap percentage of the genome (Table 5). The full statistics of S. brachiata final assembly can be found in Appendix Table 1.

Interestingly, in the case of S. europaea, the best individual genome assembly statistics were found in the genomes generated by SOAPdenovo2 and not from

ABySS, as found for S. brachiata. The shortest scaffold number was just below

400,000, the best N50 was of 195 kb, and the longest scaffold of 2.3 Mb (Table 6).

Similarly to S. brachiata, S. europaea assemblies derived from Metassembler, had improved genome assembly statistics. The number of scaffolds was greatly reduced up to 6,063, the N50 to 285 kb, and the longest scaffold slightly improved to 2.34 Mb. The best assemblies generated with Metassembler, by combining the best assemblies from each assembler (SOAPdenovo2, ABySS, Platanus), were combined with Metassembler to produce the final assembly of S. europaea. The best assembly had a slightly lower longest scaffold of 1.92 Mb; however, the scaffold number was further reduced to 3,996 while increasing the total assembled base pairs and increasing the N50 to 367 Mb (Table 6). The full statistics of S. europaea final assembly can be found in Appendix Table 2.

Genome assembly statistics greatly improved by the merging different assemblies with Metassembler. Nevertheless, the utility of a genome resides on the completeness of its annotation. RNASeq in silico normalized reads mapped 62 % and 80 % to S. brachiata and S. europaea final assemblies, respectively. This number 81

of mapped reads was very similar for all genome assemblies. Genome annotations for each genome assembly were produced with BRAKER1 and assessed with

BUSCO v3. The number of proteins predicted ranged from 26,000 to 33,000 for both S. brachiata and S. europaea genome assemblies. Similarly to assembly statistics, protein prediction was improved in the genomes generated from Metassembler, for both S. brachiata (Table 7) and S. europaea (Table 8). The selection of

Metassembler genomes to be combined and that of the final assembly, were primarily based on the completeness of its annotation.

To compare the completeness of the final genome assemblies, the protein sequences of three Caryophyllales were downloaded: Beta vulgaris (Dohm et al.,

2014), with 29,088 predicted proteins; Chenopodium quinoa V1 (Jarvis et al., 2017), with 44,776 predicted proteins; and Spinacia oleracea (Xu et al., 2017), with 23,689 predicted proteins. Arabidopsis thaliana proteins were also included and were downloaded from TAIR10 dataset comprising 35,386 proteins. Proteins from each species were assessed with BUSCO v3 using the Embryophyta dataset. BUSCO compares protein and nucleotide sequences to a dataset that aims to comprise the core genes for a given lineage. The genomes of Beta vulgaris and Chenopodium quinoa have been sequenced to a chromosome level and may serve as a very thorough comparison for genome annotation within the Caryophyllales order.

Comparable genome annotation completeness was found for both S. brachiata and

S. europaea when compared to the other sequenced Caryophyllales (Table 9). 82

Table 5. Salicornia brachiata genome assemblies statistics Assembled bases N50 Longest scaffold Assembler Scaffolds Gap % Genome % > 50 kb (Mb) (kb) (Mb)

Platanus k-mer 71 544 1,040,064 61 2.30 2.25 53.19

Soapdenovo2 k-mer 596 527,017 140 1.47 4.50 71.00 77

Abyss k-mer 79 547 361,286 173 1.28 5.98 75.39

Best statistics Metssembler (Ab85 481 9,735 254 2.48 5.37 89.64 Sp77 Pl71) Mate: 5_8kb Best statistics and annotation Metassembler (Ab71 477 9,827 229 1.83 5.81 88.94 Sp65 Pl51) Mate: 2_5kb Abyss Metassembler gap closed (83-85-71) 469 9,468 242 2.53 5.03 90.41 Mate: 2_8kb proteins (28609) Metassembler assemblies combined: 468 5,230 302 2.51 3.55 93.55 Salicornia brachiata final assembly

Table 6. Salicornia europaea genome assemblies statistics Assembled bases N50 Longest scaffold Assembler Scaffolds Gap % Genome % > 50 kb (Mb) (kb) (Mb)

Platanus 33 506 868,063 68 1.334 2.56 55.25

Soapdenovo2 K-mer_79 582 458,909 157 1.218 4.76 76.42

Abyss k-mer 75 561 504,652 83 1.17 7.24 61.99

Best statistics Metassembler (Ab75 521 6,063 285 2.34 4.38 93.61 Sp71 Pl61) Mate: 2_8kb Best annotation Metassembler (Ab73 526 6,331 283 1.61 4.54 93.24 Sp67 Pl33) Mate: 2_5kb Best statistics or annotation Metassembler 508 7,040 233 2.18 3.86 91.72 (Ab71 Sp65 Pl51) Mate: 2_8kb Metassembler assemblies combined: Salicornia 530 3996 367 1.92 4.29 95.89 europaea final assembly 84

Table 7. Salicornia brachiata genome protein prediction completeness assessment by BUSCO v3 against the Embryophyta dataset Complete single Complete Genome (proteins) Complete Fragmented Missing copy duplicated Platanus 71 k-mer gap closed 1256 1135 (78.8 %) 121 (7.4 %) 78 (5.4 %) 106 (7.4%) (32228) (87.2 %) Soapdenovo2 77 k-mer gap closed 1237 1100 (76.4 %) 137 (9.5 %) 92 (6.4 %) 111 (7.7 %) (31687) (85.9 %) 1270 Abyss 79 k-mer gap closed (29092) 1138 (79 %) 132 (9.2 %) 76 (5.3 %) 94 (6.5 %) (88.2 %) Best statistics Metassembler gap 1271 closed (Ab75 Sp71 Pl61) Mate: 1152 (80 %) 119 (8.3 %) 79 (5.5 %) 90 (6.2 %) (88.3 %) 5_8kb proteins (28471) Best statistics or annotation 1284 Metassembler gap closed (Ab85 1161 (80.6 %) 123 (8.5 %) 63 (4.4 %) 93 (6.5 %) (89.1 %) Sp79 Pl33) 2_5kb proteins (29497) Abyss Metassembler gap closed 1291 1155 (80.2 %) 136 (9.4 %) 64 (4.4 %) 85 (6 %) (83-85-71) Mate: 2_8kb proteins (89.6 %) (28609) Metassembler assemblies 1314 1184 (82.2 %) 130 (9.0 %) 50 (3.5 %) 76 (5.3 %) combined: Salicornia brachiata final (91.2 %) assembly (32116) 85

Table 8. Salicornia europaea genome protein prediction completeness assessment by BUSCO v3 against the Embryophyta dataset

Genome (proteins) Complete Complete single copy Complete duplicated Fragmented Missing

Platanus 33 k-mer 1267 (88 %) 1155 (80.2 %) 112 (7.8 %) 79 (5.5 %) 94 (6.5 %) gapclosed (27383) Soapdenovo2 K-mer 79 1286 (89.3 %) 1165 (80.9 %) 121 (8.4 %) 67 (4.7 %) 87 (6.0 %) gapclosed (29650) Abyss k-mer_75 gap 1271 (88.3 %) 1152 (80 %) 119 (8.3 %) 73 (5.1 %) 96 (6.6 %) closed (29427) Best_annotation Meta (Ab73 Sp67 Pl33) Mate: 1293 (89.8 %) 1164 (80.8 %) 129 (9 %) 61 (4.2 %) 86 (6 %) 2_5kb proteins (28241) Best_statistics Meta (Ab75 Sp71 Pl61) Mate: 1295 (89.9 %) 1168 (81.1 %) 127 (8.8 %) 61 (4.2 %) 84 (5.9 %) 5_8kb proteins (27959) Best statistics or annotation Metassembler 1287 (89.4 %) 1162 (80.7 %) 125 (8.7 %) 64 (4.4 %) 89 (6.2 %) (Ab71 Sp65 Pl51) Mate: 2_8kb (28357) Metassembler assemblies combined: Salicornia 1307 (90.8 %) 1183 (82.2 %) 124 (8.6 %) 54 (3.8 %) 79 (5.4 %) europaea final assembly (29086) 86

Arabidopsis thaliana annotation had a nearly perfect completeness score with

BUSCO v3. This is expected, as it is arguably the most studied plant species. An appropriate comparison should be assessed from the closer related species.

Arabidopsis thaliana had 99.5 % of the protein set, Beta vulgaris 93.1 %, S. brachiata

91.2 %, S. europaea 90.8 %, Chenopodium quinoa 90.6 %, and Spinacia oleracea 90.5 %.

The results suggest that the annotation of the genomes of S. brachiata and S. europaea are of a similar completeness to the previously published genomes of the

Caryophyllales order.

After assessing the completeness of the genome protein prediction, the protein models generated from BRAKER1 were passed to MAKER (Campbell et al., 2014) for protein annotation. Following the MAKER pipeline, proteins were annotated by BLAST against SwissProt and protein domains were analyzed with

InterProScan (Quevillon et al., 2005). InterProScan searches for protein signatures and serves as a link between different protein databases e.g. Pfam, CDD, and

KEGG. Transmembrane domains in proteins were searched with TMHMM (Krogh et al., 2001). The annotation of the genomes has been completed and the genomes are ready to be released.

Table 9. Genome protein prediction completeness assessment by BUSCO v3 against the Embryophyta dataset

Genome (proteins) Complete Complete single copy Complete duplicated Fragmented Missing

Arabidopsis thaliana (35386) 1433 (99.5 %) 1093 (75.9 %) 340 (23.6 %) 3 (0.2 %) 4 (0.3 %)

Beta vulgaris (29088) 1341 (93.1 %) 1200 (83.3 %) 141 (9.7 %) 35 (2.4 %) 64 (4.5 %)

Chenopodium quinoa (44776) 1305 (90.6 %) 366 (25.4 %) 939 (65.2 %) 56 (3.9 %) 79 (5.5 %)

Spinacia oleracea (23689) 1303 (90.5 %) 1145 (79.5 %) 158 (10.9 %) 60 (4.2 %) 77 (5.3 %)

Salicornia brachiata (32116) 1314 (91.2 %) 1184 (82.2 %) 130 (9.0 %) 50 (3.5 %) 76 (5.3 %)

Salicornia europaea (29086) 1307 (90.8 %) 1183 (82.2 %) 124 (8.6 %) 54 (3.8 %) 79 (5.4 %) 88

2.3.3 Salicornia bigelovii genome assembly using SMRT sequencing and hybrid assemblies

Illumina sequencing of S. bigelovii resulted in 122x sequence depth of coverage of

Genome shotgun and PCR-Free libraries (Table 4). Unfortunately, previously generated Mate pair libraries were not successful. Genome assembly of S. bigelovii using only Illumina reads, resulted in millions of very small scaffolds, which were too small for protein prediction. To compensate for this, 20 cells of SMRT sequencing were generated, resulting in 40 x coverage. Because S. bigelovii is an undomesticated and tetraploid, which could be cross-pollinated (Fernández-

Illescas et al., 2011), coverage requirements might increase depending on its heterozygosity (VanBuren et al., 2015).

2.3.3.1 Genome assembly using SMRT sequencing

It has been suggested that 20 x coverage of SMRT sequencing could be sufficient for de novo assembly and outperform any hybrid assembly (Koren et al., 2017).

SMRT sequencing reads were corrected and trimmed using Canu v1.5 (Koren et al., 2017) and assembled with either Canu v1.5 or SMARTdenovo

(https://github.com/ruanjue/smartdenovo) following the strategy as described in Schmidt et al. (2017), as SMARTdenovo does not perform error correction.

Using default parameters in Canu, the final assembly was only 15 Mb. Using the 89

same trimmed reads, the final assembly using SMARTdenovo resulted in a 16 Mb assembly. One of the reasons why the assemblies were so small, could be the fact that coverage was greatly reduced after read trimming from 40x to 3x.

Alternatively, Canu was run with parameters suggested for low coverage and heterozygous species e.g. increasing the read error rate. Read coverage decreased from 40x to 15x and the assembled genome resulted in 202 Mb. SMARTdenovo was run with the 15x trimmed reads and generated an assembly of 85 Mb. Since low coverage might be compromising genome assembly, untrimmed reads were used in SMARTdenovo, which have been shown to generate good assemblies with uncorrected reads (Giordano et al., 2017). This resulted in an assembly of 51.9 Mb.

The short assemblies generated with Canu and SMARTdenovo, suggest that increased coverage is needed for a non-hybrid de novo assembly.

2.3.3.2 Hybrid genome assembly using SMRT and Illumina sequencing

Low coverage SMRT sequencing reads were used to scaffold contigs generated from

Illumina sequencing. Assemblers such as SPAdes (Bankevich et al., 2012) and DBG2OLC

(Ye et al., 2016) are capable of scaffold using long reads. Genome assembling with

SPADES resulted in an assembly of 672 Mb with medium sized scaffolds, which were insufficient for protein prediction. DBG2OLC can use contigs generated from any assembler and scaffold them using long reads; its success has been demonstrated with the recently assembled apple genome (Daccord et al., 2017). Contigs generated with ABySS 90

and the previously mentioned genome assembly from SPAdes were used for DBG2OLC scaffolding. The assembly using ABySS resulted in 436 Mb, whereas the assembly using

SPAdes resulted in 655 Mb. However, only 29 and 31 % of RNASeq reads were able to be mapped with Tophat to the genome of DBG2OLC-ABySS and DBG2OLC-SPAdes, respectively. Moreover, protein prediction with BRAKER1 and subsequent completeness assessment with BUSCO v3 for the DBG2OLC-ABySS assembly, resulted in only 23.8 % completeness. Furthermore, masking of the genome for repetitive elements in the assembly with RepeatMasker, resulted in up to 80 % masking. Suggesting, that the sequence depth of coverage of SMRT sequencing reads, might not be sufficient to assemble the full genome of S. bigelovii causing the assemblers to generate bubbles in the assembly leading to repetitive scaffolding. Alternatively, sequence quality or sample contamination could hamper genome assembly. Sequencing quality was assessed with FASTQC and was discarded as a possible source for the resulting short assemblies. Contamination with prokaryotic and human DNA was assessed with BBSplit from BBTools, resulting in very little observed contamination in the samples. Suggesting that the main factor limiting the assembly of S. bigelovii was sequence depth of coverage.

2.4 Conclusion

The genomes of the halophytes S. brachiata and S. europaea have been assembled.

The methodology of combining different assemblies with Metassembler was shown to be successful and greatly decreased the number of scaffolds from hundreds of thousands to around 5,000 scaffolds using Illumina sequencing alone.

The methodology not only reduced the scaffold number, but increased the 91

contiguity of the scaffolds and improved genome annotation. It is noteworthy that the performance of different assemblers varied greatly between species, even if these are closely related species. Suggesting, that assemblers are susceptible to data biases and different assemblers should be used for sequencing projects to identify the most appropriate assembly. In terms of the annotation quality, the genomes of S. brachiata and S. europaea are as complete as those of other published

Caryophyllales, even of those assembled to chromosome level. Unfortunately, the genome assembly of S. bigelovii was of poor quality, probably due to a lack of coverage of good quality reads, as published plant genomes with long reads have at least 40x coverage of corrected reads. Low quality and contamination were discarded as possible sources of error during assembly. It would be recommended to either increase the coverage of SMRT sequencing long reads or generate

Illumina mate pair libraries. This was probably exacerbated by the likely high level of heterozygosity and tetraploidy of S. bigelovii. The assembled genomes of S. brachiata and S. europaea, represent the first of the Salicornioideae subfamily and the first euhalophytes. They set the basis that will enable future research in these plants and hoping to aid in the understanding of the salt tolerance that these plants possess. Previous research projects selected S. bigelovii as a target for domestication of halophytic plant species (personal communication with Prof. Emerita Nina

Fedoroff, KAUST). For this reason, it was decided to focus subsequent experiments on S. bigelovii. 92

Chapter 3. Salicornia bigelovii phenotype under salinity

Salinity tolerance is a complex trait and plants have different strategies to cope with high concentrations of NaCl. Among some strategies, plants have developed mechanisms to extrude sodium ions (Na+) from the roots, have a tight regulation of Na+ import into the stele to prevent Na+ translocation to the shoots, sequester

Na+ into the stems, accumulate Na+ in older leaves, extrude Na+ from the cells in the shoots, or sequester Na+ into the vacuoles. In this chapter, the basic characterization of the phenotypic and ionic responses of Salicornia bigelovii to increased concentrations of NaCl is presented.

3.1 Introduction

S. bigelovii is a halophyte that grows in seashores and salt marshes and displays enhanced growth upon addition of NaCl, with optimum growth reported to occur at approximately 200 mM NaCl (Webb, 1966, Ayala & Oleary, 1995, Brown et al.,

1999, Ohori & Fujiyama, 2011, Kong & Zheng, 2014). Land plants that are able to grow in seawater are rare and it is considered that < 500 species can grow under these conditions (Flowers et al., 2010). This makes Salicornia an interesting genus for the study of salinity tolerance. Different seed germination has been observed for different Salicornia species at several salt concentrations - generally, low amounts of salt increase germination (Ungar, 1967, Khan et al., 2000). However, in the case of S. bigelovii, temperature has been reported to also play an important role, as germination was maximal 93

at sea level NaCl concentrations at temperatures of 4 and 15 °C, but lower NaCl concentrations were preferred at 26 °C (Rivers & Weber, 1971a). Germination of

Salicornia europaea was observed to be optimal at 150 to 300 mM NaCl (Muscolo et al., 2014).

The effects of supplementation of the growth medium with Na+, K+, and Cl- were studied in S. europaea (Lv et al., 2012a). It was found that the addition of Na+ had a greater effect than K+ or Cl- on cell expansion, succulence, photosynthesis, and stomatal opening. Greater concentrations of Na+ than K+ in the medium were also found to be beneficial for S. bigelovii (Yamada et al., 2016). Scanning electron microscopy, coupled with X-ray microanalyses of shoot cross-sections, revealed that Na was more abundant in spongy mesophyll cells than in the vasculature or palisade cells of both Salicornia brachiata (Reddy et al., 1993) and S. europaea (Lv et al., 2012b). Interestingly, Cl was similarly abundant in both palisade and spongy mesophyll cells, whereas K was more abundant in palisade cells of S. brachiata

(Reddy et al., 1993), suggesting a preference for higher K concentrations in photosynthetically active cells.

Despite ion accumulation in S. bigelovii being measured in some studies, very few have measured its growth and ion accumulation in soil. Similarly, very little is known about its natural variation in terms of Na and K accumulation and the relationships between ion and water contents. In this chapter, the phenotypic responses and ion accumulation of S. bigelovii plants under different salt 94

concentrations are described. To assess possible Na partitioning in shoots, Na distribution was assessed between older and younger branches and across different S. bigelovii inbred lines. Water content, ion accumulation, and their correlations in plants treated with 0, 50, 200, and 600 mM NaCl were analyzed.

3.2 Materials and methods

3.2.1 Plant material and growth

Salicornia bigelovii seeds were provided by Dr. E. Glenn of the Environmental

Research Laboratory, University of Arizona, Tucson, USA. Plants showing contrasting morphologies and displaying agronomic traits of interest, such as harvest index, canopy compactness and biomass, were inbred for three generations (Supplementary Fig. 1 and 2). Seeds derived from the inbred populations were sown in Metro-Mix® 360 soil (SunGro Horticulture, USA) contained in plastic trays and watered daily to 70% soil water holding capacity.

Plants were grown in a glasshouse in KAUST under natural irradiance and kept at a constant temperature of 28/24 °C day/night with 65% relative humidity. S. bigelovii inbred lines were grown for 4 months, from May – September 2014 and were irrigated twice with half sea water. S. bigelovii plants of one inbred line were grown during the months of May – July 2015. Five weeks after sowing, S. bigelovii plants were treated with 0, 5, 50, 200, or 600 mM NaCl. Shoots were harvested at midday at 1 and 6 weeks after treatment for ion analyses. 95

3.2.2 Plant ion content and phenotype

Upper and lower branches of different inbred lines of Salicornia were harvested from 4-month-old plants and used for ion analyses. Similarly, whole shoots and roots of S. bigelovii were harvested 1 and 6 weeks after salt treatment. Samples were rinsed with MilliQ water and surface water removed with paper towels prior to fresh mass measurements. Samples were then dried in an oven at 70 ºC for 3 d to determine the dry mass. Water content was calculated as (fresh mass – dry mass)

/ dry mass. 10 mg of dried tissue were mixed with 1.5 mL of 500 mM nitric acid, heated at 80 ºC for 3 h, and used for Na and K content analyses by flame photometry (Flame photometer model 420, Sherwood Scientific Ltd, United

Kingdom) and Cl- content (Chloride analyser model 926, Sherwood Scientific Ltd,

United Kingdom). Flame photometry measures the total element in a sample, not distinguishing between ionic and non-ionic forms; hence, these measurements will be referred to as “Na” and “K” and not as “Na+” and “K+”. In contrast, the Choride analyser uses titration with silver nitrate to measure chloride in its ionic form, “Cl-

“. Shoot diameter imaging and measurements were obtained by finely slicing S. bigelovii shoots with a surgical blade and examining them under a light microscope

(AZ100 Multizoom, Nikon, Japan). Shoot area cross sections were measured with

NIS-Elements BR software (Nikon). Shoot sap osmolality was obtained by crushing S. bigelovii shoots and centrifuging at 10,000 rcf for 2 minutes. 10 µL of the supernatant was measured with a VAPRO® (ELITechGroup). For statistical analyses, three samples per treatment were taken into consideration while 96

comparing different treatments and statistical analyses were done with OriginPro

2015 (OriginLab).

3.3 Results and discussion

3.3.1 Sodium and potassium distribution in Salicornia bigelovii plants

Salicornia bigelovii (within this chapter referred as Salicornia) is a wild species and very little is known about its natural variation and salinity tolerance. To assess if differences in Na accumulation exist within a plant and across different plants, Na and K content were measured for different inbred “lines”. These lines were derived from three generations of inbreeding of selected plants with contrasting phenotypes in KAUST (Appendix Figs. 1 and 2), derived from a population provided by Dr. E. Glenn of the Environmental Research Laboratory, University of Arizona, Tucson, USA. While the plants had a homogenous phenotype, their level of genetic heterozygosity was not measured.

S. bigelovii have been reported to accumulate high concentrations of NaCl in its shoots (Ayala & Oleary, 1995). One mechanism by which plants can cope with increased levels of salinity in their shoots, is by storing Na+ in older leaves, preventing the accumulation of Na+ in the young photosynthetically active leaves

(Yeo & Flowers, 1982, Parida & Das, 2005, Munns & Tester, 2008, Di Gioia et al.,

2013). To see if S. bigelovii possesses such tolerance mechanism, Na content was measured in the upper and lower branches of Salicornia. No differences were 97

found in Na distribution in Salicornia plants (Fig. 7), suggesting that salinity tolerance in Salicornia is not related to a re-distribution of Na within the shoot, but to a mechanism that is conserved across the whole aerial part of the plant.

Salt stress affects the balance between the Na+/K+ inside the cells (Maathuis

& Amtmann, 1999). Under favorable conditions, K+ concentration is typically 100 to 200 mM (Leigh & Jones, 1984), whereas Na+ is maintained below 20 mM

(Amtmann & Sanders, 1998). Under salt stress, Na+ accumulation can increase

Na+/K+, leading to Na+ toxicity (Maathuis & Amtmann, 1999). To assess if

Salicornia might have differences in Na and K accumulation which might lead to different strategies for salt tolerance within different lines of Salicornia, Na and K concentrations were measured in the upper branches of Salicornia inbred lines (Fig.

10). All plants showed 10 to 20 times greater accumulation of Na+ than of K+ in its shoots. It can therefore be concluded that Na exclusion from the shoots and low

Na/K are not the main mechanisms of salt tolerance in Salicornia.

100

40 Upper branch Lower branch

30 Na content percentage content Na

0 1 2 3 6 7 9 10 11 12 13 15 16 17 21 22 24 25 28 29 31 33 34 35 36 37 38 45 48 Salicornia line

Figure 9. Relative Na content in top and bottom branches of S. bigelovii lines. Values are shown as percentages of the sum of the total Na found in both tissues. Upper branches are shown in orange, lower branches are shown in blue. Error bars are mean confidence intervals at α = 0.05, n = 3. 99

10000 1200 9000 1000 8000 7000 800

6000

(drymass)

1 1 mass) (dry 5000 600 - 4000 400 3000

2000 µmolK* g µmolNa* g 200 1000 0 0 13 33 28 36 21 38 12 48 24 2 17 31 45 9 37 16 29 15 3 25 35 26 34 10 11 8 1 50 7 6 43 49 22 42 44 14 Mean Na content Salicornia bigelovii line Mean K content

Figure 10. Na (orange) and K (purple) contents in upper shoots of S. bigelovii inbred lines. Error bars are mean confidence intervals at α = 0.05, n = 3. 100

3.3.2 Salicornia bigelovii growth at different NaCl concentrations

To assess the phenotypic responses of Salicornia to salt, Salicornia were grown in soil under different NaCl concentrations: 0, 5, 50, 200, and 600 mM. Plants were grown in soil for 5 weeks prior to salt treatment. Plants treated with higher concentrations of NaCl already displayed enhanced growth one week after treatment (Figs. 11a and 12a). Plants treated with 200, and 600 mM NaCl started developing lateral branches and turned into a slight lighter green color when compared to those plants grown under 0 and 5 mM NaCl (Fig. 12a). Differences in growth became more evident six weeks after treatment (Figs. 11b and 12b). Plants treated with 50, 200, and 600 mM were taller than those treated with 0 and 5 mM

NaCl. Plants treated with 600 mM NaCl had stems that were were shorter, but thicker, than those treated with 200 mM NaCl. Cross-sections of Salicornia shoots showed that stem diameter and area increased as plants were grown in higher salt concentrations (Figs. 13 a and b). The increased stem thickness seemed not to be due to a greater cell number but to an apparent enlargement of cells based on visual inspection of cross-section images. However, proper measurements are required to conclude this. 101

Figure 11. Saliornia bigelovii growth under different salinities. Plants were grown without the addition of NaCl for 5 weeks prior the addition of 0, 5, 50, 200, or 600 mM NaCl. a) Plants 1 week after treatment; b) Plants 6 weeks after treatment. 102

Figure 12. Growth of individual S. bigelovii plants under different salinities. Plants were grown without the addition of NaCl for 5 weeks prior the addition of 0, 5, 50, 200, or 600 mM NaCl. a) Plants 1 week after treatment; b) Plants 6 weeks after treatment. 103

Figure 13. S. bigelovii shoot expansion. a) shoot cross-sections and b) shoot cross-section area of plants treated with 0, 5, 50, 200, or 600 mM NaCl. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s LSD α = 0.05 and are indicated as letters above the bar graphs.

104

3.3.3 Salicornia bigelovii ion accumulation during growth at different NaCl concentrations

Salicornia growth increased with greater concentrations of NaCl in the soil, with an optimum at 200 mM NaCl (Fig. 11), in agreement with previous reports for

Salicornia bigelovii (Webb, 1966, Ayala & Oleary, 1995, Brown et al., 1999, Ohori &

Fujiyama, 2011, Kong & Zheng, 2014). In order to analyze water and ion status in

Salicornia, measurements of water content, shoot sap osmolality, Na, K, and Cl were carried out in plants treated for 6 weeks with 0, 50, 200, and 600 mM NaCl.

Salicornia shoots had increased osmolality with higher NaCl treatment (Fig.

14 a). This means, that Salicornia shoots had an increased concentration of solutes as NaCl treatment increased. Water content in both shoots and roots increased with NaCl treatments (Figs. 14 b and c). This is in agreement with the observed change in shoot cross-section area with salinity (Fig. 13), as it is likely that the increased water content led to cell expansion and thus, increased stem cross- sectional area. Interestingly, a greater increase in water content was observed in shoots than in roots in response to NaCl. Roots only displayed altered water content when treated with 600 mM NaCl. This increment in water content in response to salt treatment is different to what is observed for glycophytes in which water content is normally either decreased or maintained (Gama et al., 2009,

Vysotskaya et al., 2010, Negrão et al., 2017). This suggests that Salicornia might either need NaCl as an osmolyte for proper water transport into the shoots or the 105

increased water content could be used as a strategy to dilute NaCl below toxic levels.

Figure 14. S. bigelovii shoot sap osmolality and water content. a) Salicornia shoot sap osmolality; b) shoot water content; and c) root water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s LSD α

= 0.05 and are indicated as letters above the bar graphs. 106

Na was accumulated at greater concentrations in plants exposed to higher salt treatments (Figs. 15 and 16). In shoots, when Na is measured relative to the total dry mass of the shoot, Na is significantly accumulated at different amounts in each treatment (Fig. 15 a). However, when Na is determined based on the water content, the concentration of Na is not significantly different between 50 and 200 mM NaCl treated plants (Fig. 15 b). This suggests that Salicornia is accumulating extra water to maintain Na concentration at a level which might be optimal for its development.

In roots, as opposed to shoots, both Na content and molality (mol/kg), expressed as milimolal (mm), increased with all treatments (Fig. 16). This is probably due to the fact that water content did not vary in roots, except in 600 mM

NaCl treated plants, removing the diluting factor. In all treatments, shoots had greater concentration of Na than roots. Noteworthy, the shoot Na concentration at the 0 mM NaCl treatment was over 250 mm. This accumulation is 20 – 40 times greater than that of Arabidopsis plants without the supplementation of NaCl; and already doubled or equaled the Na in Arabidopsis leaves when treated with 100 mM NaCl, depending on the ecotype (Jha et al., 2010). This suggests that Salicornia either has very limited Na exclusion capability or has higher Na transport into the shoots compared to Arabidopsis. Taking into account the high Na concentrations at 0 mM NaCl treatment, the increased water content with NaCl, and optimal 107

growth at about 200 mM NaCl, would suggest that Salicornia utilizes Na as an osmolyte.

Figure 15. S. bigelovii Na accumulation in shoots. a) Salicornia Na content relative to shoot dry mass; b) Salicornia Na molality i.e. Na relative to shoot water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s

LSD α = 0.05 and indicated as letters above the bar graphs. 108

Figure 16. S. bigelovii Na accumulation in roots. a) Salicornia Na content relative to root dry mass; b) Salicornia Na molality i.e. Na relative to root water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s

LSD α = 0.05 and indicated as letters above the bar graphs.

K followed the opposite trend to Na. With increased salt treatment, K content decreased in both shoots and roots (Figs. 17 and 18). This is a typical 109

outcome of salt treatments and a similar trend has been reported for Salicornia

(Ayala & Oleary, 1995). Na+ is thought to enter the cells through non-selective cation channels (Davenport & Tester, 2000, Demidchik & Tester, 2002). Since Na+ and K+ have a relatively similar effective (Benito et al., 2014) and hydrated radii

(Volkov et al., 1997), K+ transport might be hampered by the presence of high concentrations of Na+. In fact, several K+ transporters can also transport Na+

(Benito et al., 2014); thus, increased Na+ in the medium might lead to a decreased

K+ uptake.

Plants typically contain ~10 times more K+ than Na+ in the absence of NaCl

(Leigh & Jones, 1984, Amtmann & Sanders, 1998). The ratio between Na and K

(Na/K) in Salicornia was in favor of K in roots and similar to Arabidopsis plants

(Jha et al., 2010), except for the 600 mM NaCl treatment, where Na was 1.5 - 2 times more abundant (Fig. 19 b). This is strikingly different to Salicornia shoots, which accumulated greater concentrations of Na than K (Fig. 19 a), even at the 0 mM

NaCl treatment, where the Na/K ratio was 1.9. At its optimum growth in 200 mM

NaCl, Na in shoots was 10 times more abundant than K and in 600 mM NaCl treated plants, Na was 25 times more abundant (Fig. 19 a). This indicates that there must be a very efficient ion transport in the shoot cells of Salicornia, allowing K+ in the cytosol while compartmentalizing the high concentrations of Na+ into the vacuole to prevent toxicity. 110

Figure 17. S. bigelovii K accumulation in shoots. a) Salicornia K content relative to shoot dry mass; b) Salicornia K molality i.e. K relative to shoot water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s

LSD α = 0.05 and indicated as letters above the bar graphs.

111

Figure 18. S. bigelovii K accumulation in roots. a) Salicornia K content relative to root dry mass; b) Salicornia K molality i.e. K relative to root water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s LSD α

= 0.05 and indicated as letters above the bar graphs.

112

Figure 19. S. bigelovii Na / K molality ratio. a) Shoots; b) roots. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s LSD α

= 0.05 and indicated as letters above the bar graphs. Note the very different scale for the y-axis between the two panels.

Cl accumulation was measured to see if its accumulation is comparable to

Na accumulation in NaCl treated plants. Chloride was measured using silver titration, which detects chloride anions (Cl-). Both Cl- content and molality in shoots and roots (Fig. 20 and 21) followed the same pattern as Na accumulation 113

(Fig. 15 and 16). Cl- and Na molality values in shoots were almost identical, with:

Cl-, 260 mm, 504 mm, 560 mm, and 653 mm; and Na, 262 mm, 468 mm, 538 mm, and 710 mm, for plants treated with 0, 50, 200, and 600 mM NaCl, respectively.

This indicates that both Na and Cl- are being accumulated at similar concentrations in the shoots and roots of Salicornia. Cl- toxicity is not well understood, but appears to vary greatly with plant species (Munns & Tester, 2008). Most plants can tolerate Cl- tissue concentrations of 400 mM (Munns & Tester, 2008). However, some plants species, such as Glycine max, appear to be more affected by Cl- than by Na+ (Luo et al., 2005). Salt tolerance in Vitis vinifera (Sykes, 1992), Lotus spp.

(Teakle et al., 2006), and Citrus spp. (Romero-Aranda et al., 1998, Storey & Walker,

1998) have also been correlated with Cl- exclusion from the shoots. Cl- concentrations observed in Salicornia shoots are very high and indicates that

Salicornia possesses high Cl- tolerance, in addition to high Na+ tolerance.

Na+ and Cl- can be used by plants as osmolytes to maintain water balance in the tissues and has been suggested as a strategy that could be employed by halophytes (Yeo, 1983, Flowers et al., 2010). However, a tight regulation must exist for these ions as excess accumulation in the cytoplasm could lead to toxicity. It is likely, that these ions are stored in the vacuole, while compatible solutes are used in the cytoplasm to maintain water balance (Flowers et al., 1977, Flowers & Colmer,

2008). This could hold true, as Salicornia cells are highly vacuolated. Na molality is strongly positively correlated with water content, and Cl- molality, while it is 114

negatively correlated with K molality (Fig. 22). This supports the notion that both

Na+ and Cl- are involved in water translocation to the shoots.

Figure 20. S. bigelovii Cl- accumulation in shoots. a) Salicornia Cl- content relative to shoot dry mass; b) Salicornia Cl- molality i.e. Cl- relative to shoot water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s

LSD α = 0.05 and indicated as letters above the bar graphs.

115

Figure 21. S. bigelovii Cl- accumulation in roots. a) Salicornia Cl- content relative to root dry mass; b) Salicornia Cl- molality i.e. Cl- relative to shoot water content. Error bars are mean confidence intervals at α = 0.05, n = 3. Mean differences were tested with Fisher’s

LSD α = 0.05 and indicated as letters above the bar graphs.

116

Figure 22. Correlations between ion accumulation and water content in S. bigelovii plants treated with 0, 50, 200, or 600 mM

NaCl. Correlations were 2-tailed Pearson correlations with α = 0.01. 117

3.4 Conclusion

Salicornia bigelovii growth was shown to improve with the addition of NaCl with an optimum of ~ 200 mM NaCl. Plants treated with at least 50 mM NaCl displayed improved growth even 1 week after treatment and started developing lateral branches, showing the importance of NaCl for the development of Salicornia. Six weeks after treatment, plants exposed to 50, 200, and 600 mM NaCl showed improved growth and almost doubled the size of those plants grown under 0 and

5 mM NaCl. Shoot thickness also increased with salt treatment, with plants exposed to 600 mM NaCl showing an increase in shoot diameter and cross- sectional area. The increase did not appear to be due to a greater cell number but to cell expansion. Interestingly, and in contrast to glycophytes, Salicornia’s water content increased with salt treatment and this probably led to the observed cell expansion. Differently to crops, Salicornia accumulated very high concentrations of both Na and Cl- in its shoots. Even when exposed to 0 mM NaCl, Salicornia plants accumulated Na in concentrations 20 - 40 times of that accumulated by

Arabidopsis (Jha et al., 2010), suggesting that Salicornia might promote, or at least not impede as much, Na transport into the shoots. Under 600 mM NaCl treatment,

Salicornia shoots accumulated around 700 mm of Na, a concentration which would not be tolerated by most land plants. Sodium was equally distributed across the aerial part of the plant, suggesting Salicornia does not preferentially accumulate Na in older branches as a salinity tolerance mechanism to prevent toxicity in young and photosynthetically active tissue. This was observed across 118

different inbred lines, which also accumulated ~ 10 times more Na than K in their shoots, suggesting that the tolerance in Salicornia as a species is not related to Na exclusion from the shoots or high K accumulation. Shoot osmolality increased in plants exposed to higher salt treatments. This increase is probably due to the increase in Na and Cl ions in the shoots. K decreased with salt treatment and Na/K was maintained in favor of K in Salicornia roots, except in 600 mM treated plants.

Na molality was strongly positively correlated with water content and Cl- molality while negatively correlated with K molality. Supporting the hypothesis of the importance of Na and Cl- for water uptake in Salicornia shoots. Interestingly,

Na/K was always greater than 1 for every treatment in Salicornia shoots and Na was more than 20 times more abundant than K in plants treated with 600 mM NaCl.

This is noteworthy as most plants prefer to have K concentrations ~10 times higher than Na. Nevertheless, K+ could still be predominant in the cytosol of Salicornia shoot cells, having both Na+ and Cl- stored in the vacuoles and used as osmolytes for water uptake while maintaining compatible solutes in the cytosol. This would require a tight regulation of ion specific transporters that would allow the compartmentation of these toxic ions. Alternatively, proteins that prevent NaCl toxicity in the cytosol could exist. This high accumulation of ions in the shoots makes Salicornia an ideal model to investigate Na transporters and proteins involved in salt tolerance. These proteins could prove valuable for the understanding of salt tolerance and could be introduced to current crops to improve their growth under saline soils. 119

Chapter 4. Transcriptomic and proteomic profiling of Salicornia bigelovii shoots under different salinities

Salicornia bigelovii displays enhanced growth upon addition of NaCl to the medium, as previously discussed in Chapter 3. Despite high concentrations of both Na+ and Cl- being accumulated in its shoots, S. bigelovii manages to prevent ion toxicity. This requires the tight regulation of ion specific transporters in the shoots. In this chapter, the identification of proteins involved in salinity tolerance of S. bigelovii is addressed through differential expression analyses and proteomics studies of shoots in S. bigelovii plants treated with different concentrations of NaCl.

Particular emphasis is put on transporter proteins, particularly to those predicted to target the vacuole. An organellar membrane proteome is created from membrane enriched density fractions allowing spatial separation of proteins and their subsequent organellar allocation.

4.1 Introduction

As discussed in Chapter 3, Salicornia bigelovii plants have an optimum growth at approximately 200 mM NaCl (Webb, 1966, Ayala & Oleary, 1995, Brown et al.,

1999, Ohori & Fujiyama, 2011, Kong & Zheng, 2014) and is considered to be one of the most salt tolerant plant species on the planet (Flowers & Colmer, 2015). Its growth at optimum saline conditions results in the high accumulation of both Na+ and Cl- in its shoots reaching concentrations of around 500 mm, but can reach 120

almost 800 mm when grown under 600 mM NaCl in soil (Chapter 3). This is accompanied by high Na+/K+ > 10 - 20 in favor of Na+ (Chapter 3). Ion toxicity is considered to begin once higher concentrations of Na+ than K+ are present in the cytoplasm (Amtmann & Sanders, 1998, Davenport et al., 2007). The high concentrations of Na+ and Cl- ions in the shoots of S. bigelovii would be toxic for most land plants (Flowers et al., 2010) and are considered to be used as osmolytes within the cells (Flowers et al., 2010). Nevertheless, these ions are thought to be compartmentalized in the vacuole to prevent toxicity in the cytoplasm (Flowers et al., 1986, Munns & Tester, 2008).

The compartmentalization of Na+ in vacuoles was considered to be carried out by Na+/H+ antiporter proteins of the NHX family (Apse et al., 1999, Yokoi et al., 2002, Pardo et al., 2006). However, NHX transporters were later identified to primarily function as K+ transporters in planta (Rodríguez-Rosales et al., 2008,

Leidi et al., 2010, Barragan et al., 2012, Xu et al., 2013, McCubbin et al., 2014). The question remains open of which protein is responsible for the Na+/H+ antiporter activity in the vacuoles. Plants of the Salicornia genus are succulent plants that are able to enlarge their cells with an increased water accumulation, presumably diluting Na+ and compartmentalizing it in its vacuoles (Jennings, 1968).

Accordingly, Salicornia europaea plants predominantly accumulated Na+ in vacuoles of shoot endodermal cells (Lv et al., 2012b), posing Salicornia as an ideal candidate to study and identify proteins related to Na+ transport and salinity 121

tolerance as a whole. Nevertheless, very few genetic studies have been carried out in any Salicornia species.

Fan et al. (2013a) studied the transcriptomic responses of one month-old S. europaea plants treated with 200 mM NaCl after 3 h, 12 h, 24 h, 3 d, and 7 d after salt treatment. They concluded that genes related to cell wall and lignin biosynthesis were downregulated upon salt treatment in roots (Fan et al., 2013a).

In shoots, chlorophyll biosynthesis genes were downregulated, but genes related to photosystems I and II were upregulated (Fan et al., 2013a). Photosystem II activity measured by chlorophyll fluorescence remained unaltered upon salt treatment (Fan et al., 2013a). In a different study, Ma et al. (2013) analyzed the differential expression of genes expressed in shoots of two month-old S. europaea plants treated with 200 mM NaCl for 3 d. They found the upregulation of an NHX4 homolog when exposed to salt treatment. miRNAs differential expression was analyzed in one month-old S. europaea plants treated with 200 mM NaCl for 0 h,

12 h, and 7 d after treatment (Feng et al., 2015), resulting in the majority of the miRNAs targeting transcription factors. Wang et al. (2009a) made a comparative proteomic analysis by 2-dimensional gel electrophoresis of shoots of three week- old S. europaea plants treated with 0, 200, 600, and 800 mM NaCl for three weeks.

They found 196 differentially abundant spots out of 1880 and were able to identify

111 proteins by mass spectrometry. Among these proteins, choline monooxygenase, needed for the biosynthesis of the osmoprotectant glycinebetaine

(Sakamoto & Murata, 2000, Tabuchi et al., 2005), was more abundant under salt 122

treatment (Wang et al., 2009a). Congruently, glycinebetaine was observed to be greatly accumulated in S. europaea shoots and its accumulation increased with salt treatment (Moghaieb et al., 2004). Furthermore, overexpression of S. europaea choline monooxygenase increased tobacco salt tolerance (Wu et al., 2010).

Despite some studies addressing differential expression of genes and proteins in S. europaea, very little emphasis has been given to transporters.

Moreover, publicly available data on transcripts is scarce and incomplete.

Furthermore, 2-dimensional gel electrophoresis analysis is not suitable for the identification of membrane proteins, as their solubility is very limited (Tan et al.,

2008, Rabilloud & Lelong, 2011). This restricted previous proteomics studies in S. europaea, underrepresenting membrane proteins. In this chapter, the transcriptomes of S. bigelovii, S. brachiata, and S. europaea are assembled and annotated. The differential expression analysis of shoots of S. bigelovii treated plants with 0, 50, 200, and 600 mM NaCl is analyzed, with emphasis on transporter proteins. A gel free proteomics approach is used for the differential abundance analysis of shoot membrane enriched fractions of different densities and under different salt treatments. The differential abundance of marker proteins per fraction is used to allocate membrane proteins to subcellular compartments, creating the first organellar membrane proteome for S. bigelovii. Na+ transporters and differentially expressed genes encoding for transporters were selected for further characterization (Chapter 5). Similarly, differentially abundant proteins in 123

response to salinity treatment and allocated to the tonoplast of the vacuole were selected for further analyses (Chapter 5).

4.2 Materials and methods

4.2.1 Plant material and growth

The plant material used only for transcriptome assembly is the same as described in Chapter 2. Two different RNASeq experiments were conducted. The first, aimed to generate the transcriptomes of S. bigelovii, S. brachiata, and S. europaea. The second was designed to perform differential expression analysis in shoots of S. bigelovii. The plant material for the first RNASeq experiment consisted in S. bigelovii seeds that were kindly provided by Dr. Edward Glenn of the University of Arizona, USA. S. brachiata seeds were collected by Dr. Muppala Reddy (KAUST) from dried plants naturally growing in the coastal areas of the west cost of Gujarat,

India. S. europaea seeds were kindly provided by Dr. Moshe Sagi of the Blaustein

Institute for Desert Research (BIDR). All Salicornia plants were grown in a greenhouse at King Abdullah University of Science and Technology (KAUST),

Saudi Arabia in 2012, under constant temperatures with a day night/cycle of

25/20 °C and a relative humidity of 50 ± 10%.

The plant material utilized in the second RNASeq experiment for gene differential expression analysis of S. bigelovii shoots, was the same as described in Chapter 3. 124

S. bigelovii seeds were provided by Dr. E. Glenn of the Environmental Research

Laboratory, University of Arizona, Tucson, USA. Plants showing contrasting morphologies and displaying agronomic traits of interest, such as harvest index, canopy compactness and biomass, were inbred for three generations

(Supplementary Fig. 1 and 2). Seeds from one inbred line were sown in Metro-

Mix® 360 soil (SunGro Horticulture, USA) contained in plastic trays and watered daily to 70% soil water holding capacity. Plants were grown in a glasshouse in

KAUST under natural irradiance and kept at a constant temperature of 28/24 °C day/night with 65% relative humidity during the months of May – July 2015. Five weeks after sowing, S. bigelovii plants were treated with 0, 50, 200, or 600 mM NaCl.

Shoots were harvested at midday at 1 and 6 weeks after treatment for RNA extraction and 6 wk after treatment for membrane preparations.

4.2.2 RNA extraction and library preparation

For the generation of reference transcriptomes, RNA samples from shoots and roots of four month-old plants of the three Salicornia species were previously extracted and processed by Dr. Suhail Khan. RNA was extracted with TRIzol™ with RNA MiniPrep kit (Zymo Research, USA), following the manufacturer’s protocol. RNA quality was determined with a Bioanalyzer 2100 using an RNA 600

Nano kit (Agilent, USA). High quality of RNA was used to prepare RNASeq libraries with TruSeq RNA Library Prep Kit v2 (Illumina, USA) according to the 125

manufacturer’s instructions. RNASeq libraries were sequenced in 9 lanes of

Hiseq2000 system (Illumina, USA) under paired-end mode and 101 bp read length at Core Labs (KAUST), resulting in more than 1000x coverage per plant species.

For gene differential expression analysis of shoots of S. bigleovii plants treated with 0, 50, 200, or 600 mM NaCl for 1 and 6 weeks, RNA was extracted with TRIzol™ with RNA MiniPrep kit (Zymo Research, USA), following the manufacturer’s protocol. RNA quality was determined with a Bioanalyzer 2100 using an RNA 600 Nano kit (Agilent, USA) using only RNA with RIN values greater to 8. Libraries for RNAseq were prepared with NEBNext® Ultra™

Directional RNA Library Prep Kit for Illumina® (New England BioLabs, USA) for three replicates per treatment resulting in 24 libraries with a mean insert size of

300 bp. DNA quality was determined with a high sensitivity DNA kit (Agilent,

USA) using a Bioanalyzer 2100. Libraries were barcoded and sequenced in two lanes of Illumina HiSeq™ 2000 generating ~270 million paired end reads of 2x101 bp, resulting in a sequence depth of coverage of 24x per library and a total of 576x combined coverage.

4.2.3 Transcriptome assembly, annotation, and differential expression analysis

RNASeq reads were quality checked with FASTQC

(https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and adapter 126

trimmed using Trimmomatic v0.36 (Bolger et al., 2014). The reads were in silico normalized with a set coverage of 30 and 50x and de novo assembled using Trinity v.2.06 (Haas et al., 2013). In silico normalization has been suggested to aid in the identification of lowly abundant transcripts without decreasing the assembly of highly abundant transcripts, while reducing RAM usage (Haas et al., 2013) and is now default in the latest version of Trinity (2.4.0). The quality of the assembly was measured by mapping all the reads, together and per library, back to the transcriptome assembly using Bowtie 2 (Langmead & Salzberg, 2012). The transcriptomes generated with Trinity were annotated following the Trinotate pipeline (Haas et al., 2013). Open reading frames (ORFs) were predicted with

TransDecoder (https://transdecoder.github.io/) and predicted proteins with a minimum length of 50 aa, transcripts were annotated by BLASTX and predicted proteins by BLASTP using UniRef90 and Swiss-Prot as databases. Transmembrane topology was predicted with tmHMM (Krogh et al., 2001) and pfam domains (Finn et al., 2016) were identified in predicted proteins with HMMER (http://hmmer.org/).

Transcriptome completeness was assessed by BUSCO v.3 (Simão et al., 2015) against the Viridiplantae dataset. The shoot transcripome of S. bigelovii was used for differential expression analysis. Transcripts with Viridiplantae BLAST hits had their redundancy removed with CD-HIT-EST (Fu et al., 2012), resulting in a final set of 37,646 trinity genes comprising 78,085 transcripts. Differential expression

(DE) analysis was done within the Trinity package (Haas et al., 2013). Transcript abundance was measured with RSEM (https://github.com/deweylab/RSEM) and 127

compared across all treatments and time points with three biological replicates.

DE genes were obtained with DESeq2 (Love et al., 2014) at a false discovery rate of 0.001 and a minimum 2-fold change. DE genes were clustered by hierarchical clustering of log2 transformed expression and displayed using MeV v.4.9 (Howe et al., 2011). Transcripts were mapped to MapMan bins with Mercator (Lohse et al., 2014). Plant cellular responses were explored by analyzing the DE genes with

MapMan (Thimm et al., 2004). Overrepresentation of MapMan bins were tested with Wilcoxon Rank Sum test and corrected for multiple testing with Benjamini

Hochberg with p value < 0.05.

4.2.4 Plant membrane extraction, fractionation, and quantitation

Membrane preparations were obtained for S. bigelovii shoots treated with 0, 50, 200, or 600 mM NaCl for 6 weeks with three biological replicates per treatment and processed similarly as described in Dunkley et al. (2006) with all steps done at 4 °C.

100 – 120 g of shoot tissue was blended in homogenization buffer (5 % glycerol,

100 mM KCl, 0.5 % PVP, 10 mM EDTA, 2 mM PMSF, 10 % sucrose, 50 mM HEPES, pH 7.8) and osmotically adjusted with sucrose to shoot SAP osmolality measured with a VAPRO® vapor pressure osmometer (ELITechGroup Biomedical Systems).

Homogenate was filtered through three layers of cheesecloth and spun at 10,000 x g for 30 min to sediment debris and intact organelles. The supernatant containing the microsomal fraction was transferred to a new tube and spun at 100,000 x g for 128

1 h. The pellet was resuspended in 2 mL resuspension buffer (5 % glycerol, 50 mM

KCl, 10 % sucrose, 25 mM HEPES, pH 7.8) and layered on top of a 24 mL continuous sucrose density gradient (20 – 45 %). Gradients were spun at 100,000 x g for 14 h. Fractions were recovered each 2 mL with an Auto Densi-Flow II (Buchler

Instruments) resulting in 12 different density fractions. Fraction density was measured with a digital refractometer (Brix/RI-Chek, Reichert). To release cytosolic proteins trapped inside microsomes, fractions were diluted 3 times in 160 mM Na2CO3, left on ice for 30 min, and spun at 100,000 x g for 1 h. The pellet was washed with 160 mM Na2CO3 and spun again at 100,000 x g for 1 h. The resultant pellet was resuspended in 1 mL of storage buffer (7 M urea, 2 M thiourea, 5 mM

Tris-HCl, pH 7.1) and stored at -80 °C. Protein concentrations were measured by

Bradford assay (Bradford, 1976).

4.2.5 Protein digestion and LC-MS analyses

Proteins for each fraction were trypsin digested based on filter aided sample preparation (Wisniewski et al., 2009). 30 µg of protein were incubated with 100 mM DTT for 1 h at 37 °C. Samples were loaded into a Vivacon 500 hydrosart 30 kDa filter (Sartorius) and spun at 14,000 x g for 15 min. 100 µL of 50 mM iodoacetamide were added to the column, incubated for 20 min, and spun at 14,000 x g for 10 min. 100 µL of 8 M urea, 100 mM Tris-HCl, pH 7.8 were added and the sample was spun at 14,000 x g for 15 min. This step was repeated twice. 40 µL of 129

trypsin 1:100 (Sequencing grade modified trypsin, Promega) in 50 mM ammonium bicarbonate were added to the sample and incubated in a wet chamber at 37 °C overnight. Samples were spun at 14,000 x g for 10 min. 100 µL of 50 mM ammonium bicarbonate were added to the column and spun at 14,000 x g for 15 min. 50 µL of 0.5 M NaCl were added to the column and spun at 14,000 x g for 15 min followed by the addition of 50 µL ammonium bicarbonate and another centrifugation step at 14,000 x g for 15 min. Filtered peptides were acidified by the addition of 2 % trifluoroacetic acid (TFA) and desalted with Sep-Pak C18 cartridges (Waters Scientific). Samples were lyophilized, resuspended in 15 µL 3 % acetonitrile 0.1 % TFA, and quantified with Nanodrop measurement at 280 nm wavelength. 1 µg of peptides for label-free LC/MS analyses were loaded to an

Ultimate 3000 UHPLC/Q-Exactive system (Thermo Scientific) with an Acclaim

PepMap RSLC 75 μm × 15 cm nanoViper column and run for 1 h.

4.2.6 MS spectra analyses and protein quantitation

Mass spectra peak files were compared against a combined set of available S. europaea sequences (Ma et al., 2013) together with our own generated Salicornia transcriptome assemblies of S. bigelovii, S. brachiata, and S. europaea with MASCOT v. 2.4 (Perkins et al., 1999). Database searching parameters included peptide tolerance of ± 10 ppm, fragment tolerance of ± 0.5 Da, carbamidomethyl modification, and reverse sequences as decoy. MASCOT files were loaded into 130

SCAFFOLD 4 (Searle, 2010) for protein quantitation and differential abundance analyses. For differential abundance analyses, MASCOT files were loaded as

Multidimensional Protein Identification Technology (MudPIT) with three replicates per treatment. Protein identification required a 95 % peptide identity and minimum of 2 assigned peptides per protein. Protein abundances were normalized by total spectrum counts. Differential protein abundances were tested at a significance of p ≤ 0.05 and corrected for multiple testing with Hochberg-

Benjamini correction (Benjamini & Hochberg, 1995). Differentially abundant proteins were analyzed for pathway and overrepresentation with MapMan

(Thimm et al., 2004, Usadel et al., 2009).

4.2.7 Subcellular localization prediction

Protein subcellular localization was predicted with pRoloc (Gatto et al., 2014,

Breckels et al., 2016) and MSnbase (Gatto & Lilley, 2012) R packages. To create a

Salicornia organelle marker set, all protein sequences of S. bigelovii, S. brachiata, S. europaea derived from transcriptome assemblies and published data (Ma et al.,

2013) were combined and its redundancy reduced by CD-HIT (Fu et al., 2012) resulting in the Salicornia protein dataset. The Salicornia protein dataset was blasted against 542 Arabidopsis organelle markers retrieved from pRolocdata

(Dunkley et al., 2006, Ferro et al., 2010, Trotter et al., 2010, Nikolovski et al., 2012,

Groen et al., 2014, Nikolovski et al., 2014). Positive hits with Arabidopsis markers 131

were in silico predicted for subcellular compartmentalization with mGOASVM

Plant V2 (Wan et al., 2012). Proteins whose predicted subcellular localization by mGOASVM Plant V2 was congruent with their Arabidopsis organelle marker blast hit, were assigned as Salicornia organelle markers. To predict for protein subcellular localization based on protein abundance profiles in the 12 density fractions, protein abundances per fraction of samples coming from shoots of S. bigelovii plants treated with 200 mM NaCl were analyzed with pRoloc. pRoloc was run under a supervised maximum likelihood (ML) analysis with a support vector machines analysis with 100 iterations and using the generated Salicornia organelle marker set as reference.

4.2.8 Validation of subcellular compartmentalization prediction of pRoloc by Western blot

The predicted subcellular localization by pRoloc was validated with western blots against organelle markers. 5 µg of membrane protein per density fraction were used for western blot analyses using antibodies for different organelles (polyclonal antibodies generated in rabbit, Agrisera): Chloroplast PsbA (catalog number AS05

084) 1:10000 working dilution; Plasma membrane H+ ATPase (catalog number

AS07 260) 1:1000 working dilution; and Tonoplast V-ATPase (catalog number

AS07 213) 1:2000 working dilution. Signal was detected by chemiluminescence using WesternBreeze® Chemiluminescent Kit (Invitrogen) using a secondary 132

antibody which was horse reddish peroxidase (HRP) coupled anti-rabbit generated in goat. Western blot profiles for different subcellular localization marker proteins were compared against protein profiles generated with pRoloc.

4.3 Results and discussion

4.3.1 Transcriptomes assembly and annotation

Gene differential expression analyses and shotgun proteomics require a reference transcriptome and proteome, respectively. However, genomic resources of any

Salicornia species are very limited. In terms of trancriptomic information, very few studies have been published (Fan et al., 2013b, Ma et al., 2013). Only one transcriptome is publicly available and it corresponds to two month-old S. europaea plants treated with 200 mM NaCl for 3 days (Ma et al., 2013). This transcriptome, from this point forward referred as Ma_SAEU, consists of 162,969 transcripts and proteins were predicted using TransDecoder (https://transdecoder.github.io/) resulting in 110,851 predicted proteins (Table 10). Transcripts were annotated by

BLASTX and predicted proteins by BLASTP using UniRef90 and Swiss-Prot as reference databases. The completeness of the transcriptome was assessed by

BUSCO v.3 (Simão et al., 2015) against the Viridiplantae dataset (Table 11). Only

42.6 % of the proteins were identified as complete and 32.3 % as missing, indicating that this transcriptome is significantly incomplete. 133

De novo transcriptome assemblies for S. bigelovii, S. brachiata, and S. europaea were generated by extracting RNA of shoots and roots of 4 month-old

Salicornia plants. A total sequence depth of coverage of > 1000x per Salicornia species was generated, if considering a similar number to Arabidopsis of ~27,000 genes with an average transcript length of 1.7 kb (TAIR10 release).

De novo generated transcriptome assemblies consisted of: S. bigelovii assembly with 548,477 genes, 776,897 transcripts, and 531,223 predicted proteins;

S. brachiata assembly with 289,918 genes, 399,965 transcripts, and 292,468 predicted proteins; and S. europaea assembly with 203,448 genes, 291,565 transcripts, and

212,359 predicted proteins. More details on the statistics of the transcriptome assemblies can be found in Table 10. Proteins were annotated in the same way as previously described for Ma_SAEU. The proteomes generated were used as reference for shotgun proteomics of S. bigelovii. The completeness of each assembly was assessed by BUSCO v3 (Simão et al., 2015) against the Viridiplantae protein dataset (Table 11). Complete transcripts are challenging to obtain through

RNASeq and it is expected to obtain fragmented transcripts (Steijger et al., 2013), which would lead to fragmented proteins. Overall, complete proteins ranged from

78.9 to 86.8 % per proteome and missing proteins ranged from 6.9 to 9.8 %. Despite

S. bigelovii having about twice the amount of proteins predicted than S. brachiata or S. europaea, it was the assembly with the lowest percentage of complete predicted proteins with 78.9 %. Conversely, S. europaea had the highest percentage of complete proteins with 86.8 %, despite having the lowest number of predicted 134

proteins of the three species. One explanation for this could be the level of heterozygosity of the plants. S. bigelovii is a tetraploid plant (Kadereit et al., 2007) and allelic variation could pose difficulties to the program performing the assembly, particularly when working with undomesticated species. S. europaea is a diploid species (Jefferies et al., 1981, Kadereit et al., 2007), which would present a lower degree of allelic variation, making its assembly easier. Additionally, a tendency has been observed for Salicornia in which diploid plants seem to be primarily autogamous (Dalby, 1962, Fernández-Illescas et al., 2011), while tetraploid seem to be predominantly allogamous (Ferguson, 1964). This increases the possibility that S. bigelovii has greater allelic variation than S. europaea.

135

Table 10. Summary statistics for the transcriptome assemblies of S. bigelovii, S. brachiata, and S. europaea Ma_SAEU1 S. bigelovii S. brachiata S. europaea Number of genes n/a 548,477 289,918 203,448 Number of transcripts 162,969 776,897 399,965 291,565 Transcript N50 (bp) 574 998 1,660 1,849 Average transcript length (bp) 488 638 844 954 Number of predicted proteins 110,851 531,223 292,468 212,359 1. Transcript sequences from: Ma et al. (2013) 136

Table 11. Completeness assessment of predicted proteins from the transcriptomes of S. bigelovii, S. brachiata, and S. europaea by BUSCO v3 against the Embryophyta dataset.

Complete single Complete Transcriptome (proteins) Complete Fragmented Missing copy duplicated

Ma_SAEU1 (110,851) 613 (42.6 %) 334 (23.2 %) 279 (19.4 %) 362 (25.1 %) 465 (32.3 %)

4.3.2 Gene differential expression analysis in response to salinity treatment in Salicornia bigelovii

To examine the transcriptomic responses of S. bigelovii to salinity, 5 week-old S. bigelovii plants, derived from the third generation of inbreeding, were grown in soil and treated with different concentrations of NaCl: 0, 50, 200, and 600 mM (Figs.

11 and 23) (same plant material as Chapter 3). Plants treated with NaCl displayed increased growth and accumulated Na+ and Cl- ions at high concentrations in its shoots. The phenotypic and ionic characterization of these plants is explored in

Chapter 3. RNA samples were collected from shoots of plants 1 and 6 weeks after salt imposition, resulting in a sequence depth of coverage of 24x per library and a total of 576x combined coverage.

De novo transcriptome assembly was done with Trinity v2.1.1 (Haas et al.,

2013), on 50x coverage in silico normalized reads and assembled with directionality enabled. This resulted in 279,319 trinity genes and 482,771 transcripts. 93 % of the normalized reads mapped back to the transcriptome assembly, indicating a good

138

Figure 23. Saliornia bigelovii growth under different salinities. Plant material is the same as in Chapter 3. Plants were grown without the addition of NaCl for 5 weeks prior the addition of 0, 50, 200, or 600 mM NaCl. a) Plants 1 week after treatment; b)

Plants 6 weeks after treatment. 139

assembly. Not normalized reads mapped from 84 – 91 % to the transcriptome, depending on the sample. Assembly statistics can be seen in Table 12.

Transcriptome annotation was done as previously described for S. bigelovii, S. brachiata, and S. europaea. Protein prediction resulted in 269,895 predicted proteins with length greater than 50 amino acids.

The annotation revealed low levels of contamination in the experiment. Out of the 279,319 genes, 50 were mammalian, 65 prokaryotic, and 15 viral (Table 12), resulting in a 0.04 % of contamination. Proteome completeness was assessed with

BUSCO v3 (Simão et al., 2015) against the Viridiplantae dataset (Table 13). Despite derived from only shoot transcripts, this proteome appeared to be slightly more complete (Table 13) than the one generated from roots and shoots (Table 11), increasing its completeness from 78.9 % to 81.5 % while reducing the number of missing proteins from 9.8 % to 8.4 %. One explanation for this would be that shoot transcripts came from plants treated with different concentrations of NaCl, which could lead to the expression of a broader variety of genes. Another explanation could be that the plants used for the generation of a reference transcriptome using roots and shoots came from the first generation of inbred lines, whereas the transcriptome for shoot differential expression analysis derived from plants coming from the third generation of inbreeding. This difference in inbreeding generation could affect heterozygosity and decrease allelic variation, thus, decreasing the complexity for transcriptome de novo assembly. 140

Trinity produces hundreds of thousands of transcripts. To reduce assembly errors and chimeras, only transcripts that had a BLAST hit with Viridiplantae were selected, resulting in 156,084 transcripts. To reduce redundancy, transcripts with

Viridiplantae BLAST hits were clustered with CD-HIT-EST (Fu et al., 2012), resulting in 78,085 transcripts belonging to 37,646 trinity genes. To assess if the reduction in transcripts affected the completeness of the dataset, transcript completeness was assessed with BUSCO v3 against the Viridiplantae database

(Table 13). When all 482,771 transcripts were evaluated, it resulted in a completeness of 85.1%. When the 78,085 non-redundant subset of transcripts was assessed, it resulted in 84% completeness. This reduction in completeness is minimal and the redundancy removed dataset was used as the reference for differential expression analysis.

Gene differential expression analysis was performed with Trinity (Haas et al., 2013). RSEM (https://github.com/deweylab/RSEM) was used to estimate gene and isoform expression from RNASeq reads mapped to the non-redundant subset of the transcriptome with Bowtie 2 (Langmead & Salzberg, 2012). Gene expression correlation for each replicate per treatment was >0.95 (Appendix Figs. 3 - 11), indicating good repeatability. Gene differential expression was analyzed with

DESeq2 (Love et al., 2014) at a false discovery rate of 0.001 and a minimum 2-fold change in expression. Of the 37,646 genes, 4,152 were differentially expressed between at least two treatments. Hierarchical clustering of the differentially expressed genes was done with MeV v.4.9 (Howe et al., 2011) (Fig. 24). 141

Table 12. Summary statistics for the transcriptome assembly of S. bigelovii shoots for DE analysis. Viridiplantae BLAST All Viridiplantae redundancy removed (CD-HIT-EST) hits Number of genes 279,319 46,468 37,646 Number of transcripts 482,771 156,084 78,085 Transcript N50 (bp) 1,232 2,235 1,940 Average transcript length 740 1,561 1,369 (bp) Number of predicted 269,895 n/a n/a proteins Contamination (genes) Mammals 50 (.017 %) Prokaryotes 65 (.023 %) Virus 15 (.005 %)

142

Table 13. Completeness assessment of transcripts and predicted proteins from the transcriptome of S. bigelovii shoots for DE analysis by BUSCO v3 against Viridiplantae database.

Complete single Complete Transcriptome (transcripts) Complete Fragmented Missing copy duplicated 1225 All (482,771) 289 (20.1 %) 936 (65.0 %) 110 (7.6 %) 105 (7.3 %) (85.1 %) 1218 Viridiplantae BLAST hits (156,084) 288 (20.0 %) 930 (64.6 %) 110 (7.6 %) 112 (7.8 %) (84.6 %) Viridiplantae redundancy 1209 629 (43.7 %) 580 (40.3 %) 113 (7.8 %) 118 (8.2 %) removed (CD-HIT-EST) (78,085) (84.0 %)

Proteome (proteins)

1173 All (269,895) 286 (19.9 %) 887 (61.6 %) 145 (10.1 %) 122 (8.4 %) (81.5 %)

143

Figure 24. Hierarchical clustering of three replicates per treatment of differentially expressed genes in shoots of S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl for 1 and 6 weeks. 4,152 genes were differentially expressed.

144

Plants that were not treated with any NaCl clustered together regardless of their age (Fig. 24), suggesting that gene expression responses to NaCl are greater than those of growth responses in soil without any addition of NaCl, over the time course of 5 weeks. This could also be associated with the limited growth that plants displayed without the addition of NaCl to the medium, relative to plants treated with NaCl (Figs. 12 and 23). Plants exposed to NaCl for 6 weeks, clustered together and within the replicates of their own treatment, except for one replicate of plants treated with 600 mM NaCl, which clustered with the 200 mM NaCl treatment.

Similarly, plants treated with NaCl for 1 week clustered together, except for plants treated with 600 mM NaCl. Plants treated with 600 mM NaCl for 1 week clustered independently to any time point and treatment and had the highest number of differentially expressed genes with 1,786 (Table 14), relative to the 0 mM NaCl treatment. Nevertheless, 6 weeks after treatment, plants responded similarly to those treated with 50 and 200 mM NaCl (Fig. 24, Table 14, Appendix Table 5). This suggests that plants adapted to the high concentrations of NaCl. Plants treated with 600 mM NaCl for 6 weeks had only 37 differentially expressed genes relative to 50 mM treated plants and 5 to plants treated with 200 mM NaCl for 6 weeks

(Appendix Table 5). Of the 5 differentially expressed genes between plants treated with 200 and 600 mM NaCl for 6 weeks, four were downregulated and one upregulated in the 600 mM NaCl treatment. Of the downregulated genes, two were uncharacterized proteins, one was involved in vesicle trafficking from the

Golgi to the plasma membrane, and another one was involved in specific O- 145

acetylation of cell wall polymers. The only gene that was upregulated was a phosphatase involved in the regulation of cAMP levels inside the cell. cAMP has been known to activate ion channels such as the cyclic nucleotide gated channel

(CNGC) (Leng et al., 1999, Lemtiri-Chlieh & Berkowitz, 2004). CNGCs have been shown to transport ions such Na+ and K+. While some channels such as AtCNGC2 are able to discriminate between both ions and transport K+ specifically (Leng et al., 1999, Hua et al., 2003), others can transport both (Hua et al., 2003). Controlling the regulation of these channels might be important during salt stress. The number of differentially expressed genes for the other treatments ranged from 133 to 320, when compared to their respective 0 mM NaCl treatment and time point. The number of differentially expressed genes between each treatment can be found in

Appendix Table 5.

Table 14. Number of differentially expressed genes in shoots relative to 0 mM NaCl. Time point 1 week after treatment 6 weeks after treatment 50 mM 200 mM 600 mM 50 mM 200 mM 600 mM Up 127 51 828 165 120 168 Down 116 82 958 155 102 151 Total 243 133 1786 320 222 319

146

To analyze plant responses to the different salt treatments, the differentially expressed genes were mapped to cellular metabolic pathways and biological processes using MapMan (Thimm et al., 2004). MapMan classifies genes into bins.

Each bin corresponds to a specific cellular process and each bin can have sub-bins, allowing the mapping into more specific processes. One of the advantages of

MapMan relative to Gene Ontology or similar classifiers is that MapMan was designed specifically for plants (Thimm et al., 2004, Usadel et al., 2009). MapMan pathways are derived from TAIR annotations (Huala et al., 2001), Gene ontology

(Ashburner et al., 2000), KEGG (Kanehisa & Goto, 2000), and TIGR (Lee et al., 2005).

Since MapMan is designed for Arabidopsis, a reference database linking genes to

MapMan bins for S. bigelovii had to be created. The set of genes used for differential expression analysis was mapped into MapMan bins with Mercator (Lohse et al.,

2014). Genes differentially expressed were analyzed with MapMan and PageMan

(Usadel et al., 2006).

Overrepresentation of MapMan bins was tested with Wilcoxon Sum of

Ranks test and corrected for multiple testing with Benjamini Hochberg correction with a p value cutoff of 0.05 (Table 15). All plants treated with NaCl showed few and very similar overrepresentation of MapMan bins except for 600 mM NaCl 1 week after treatment, which showed a greater number of overrepresented bins. S. bigelovii plants treated with NaCl had an increased number of genes being upregulated related to cell wall and cell wall modification (Table 15, Fig 25). This was observed for every treatment and time point. 147

Table 15. Overrepresented MapMan bins for differentially expressed genes from shoots of S. bigelovii. Overrepresentation was tested with Wilcoxon Rank Sum test and corrected for multiple testing with Benjamini Hochberg with p value < 0.05.

Bin Description Elements p value Treatment: 1 week 50 mM NaCl 10 cell wall 25 1.9E-8 10.7 cell wall modification 9 6.8E-4 10.5 cell wallcell wall proteins 6 0.04 Treatment: 1 week 200 mM NaCl 10 cell wall 13 6.1E-6 10.7 cell wall modification 6 0.01 Treatment: 1 week 600 mM NaCl 10 cell wall 83 0.0 30.2 signalling receptor kinases 135 1.36E-14 31 cell 41 1.4E-4 10.5 cell wall cell wall proteins 12 1.9E-5 31.1 cell organisation 27 2.5E-4 10.7 cell wall modification 19 2.6E-5 28.99 DNA unspecified 19 3.1E-6 29.2 protein synthesis 47 4.1E-4 30 signalling 170 6.1E-10 30.2.17 signalling receptor kinases DUF 26 93 6.6E-29 35.2 not assigned unknown 636 6.9E-53 26.12 misc peroxidases 9 8.2E-4 31.1.1 cell organisation cytoskeleton 16 9.6E-5 10.6 cell wall degradation 18 0.001 29 protein 158 0.001 10.5.1 cell wall cell wall proteins AGPs 8 0.001 10.5.1.1 cell wall cell wall proteins AGPs AGP 8 0.001 11 lipid metabolism 37 0.003 148

Bin Description Elements p value 31.1.1.2 cell organisation cytoskeleton mikrotubuli 12 0.003 28 DNA 47 0.003 35.1 not assigned no ontology 104 0.004 30.2.3 signalling receptor kinases leucine rich repeat III 8 0.01 10.1 cell wall precursor synthesis 10 0.01 27.3 RNA regulation of transcription 62 0.01 misc protease inhibitor/seed storage/lipid transfer 26.21 7 0.01 protein (LTP) family protein 26 misc 99 0.01 33 development 28 0.01 34 transport 89 0.02 34.19 transport Major Intrinsic Proteins 6 0.02 29.2.4 protein synthesis elongation 11 0.02 27 RNA 85 0.02 10.2 cell wall cellulose synthesis 17 0.02 17 hormone metabolism 42 0.02 26.4 misc beta 1,3 glucan hydrolases 8 0.02 26.2 misc UDP glucosyl and glucoronyl transferases 11 0.02 cell wall degradation mannan-xylose-arabinose- 10.6.2 6 0.03 fucose transport metabolite transporters at the 34.9 6 0.03 mitochondrial membrane 29.2.1.2 protein synthesis ribosomal protein eukaryotic 20 0.03 33.99 development unspecified 25 0.03 13.1 amino acid metabolism synthesis 10 0.04 31.1.1.1 cell organisation cytoskeleton actin 4 0.04 31.1.1.1 cell organisation cytoskeleton actin Actin 4 0.04 1 20.2.99 stress abiotic unspecified 5 0.04 13 amino acid metabolism 14 0.04 149

Bin Description Elements p value Treatment: 6 weeks 50 mM NaCl 10 cell wall 19 0.03 34 transport 24 0.03 30.2 signalling receptor kinases 18 0.03 30.2.17 signalling receptor kinases DUF 26 9 0.03 Treatment: 6 weeks 200 mM NaCl 10 cell wall 25 0.001 10.7 cell wall modification 6 0.04 Treatment: 6 weeks 600 mM NaCl 10 cell wall 23 3.9E-4 30.2 signalling receptor kinases 21 0.002 10.7 cell wall modification 7 0.005

This is congruent with the observation that plants treated with NaCl have increased growth and water content compared to plants grown without additional NaCl

(Figs. 11,12, 14, and 23). Congruently, PageMan ontologies revealed overrepresentation of major intrinsic proteins, particularly plasma membrane intrinsic proteins (PIPs) and tonoplast intrinsic proteins (TIPs) for plants treated with 600 mM for 7 days (Appendix Figure 11). These proteins are water channels which allow water flow through the membranes (Johansson et al., 2000). As cells expand due to increased water content, cell walls must be extended and remodeled. Most of the differentially expressed genes, were mapped to the pathway of biotic and abiotic stress (Figs. 26, 27, and 28). In this pathway, no overrepresentation of genes involved in pathogen recognition were found for any 150

Figure 25. Cell wall MapMan pathway for each of the treatments. Squares represent differentially expressed genes at the corresponding treatment relative to plants grown without the addition of NaCl. Red squares represent downregulation, blue squares upregulation.

treatment. Cell wall remodeling genes were overrepresented and overexpressed in all treatments. Jasmonic acid (JA) and abscisic acid (ABA) related genes, were predominantly downregulated in all salt treatments. Conversely, auxin and ethylene responsive genes were upregulated in response to salt, except in the case of plants treated with 600 mM NaCl for 6 weeks, where auxin related genes were 151

downregulated. These hormones have all been linked to salt stress (Ryu & Cho,

2015) and the cross-talk across and timing of their accumulation is key for a salt- response. ABA is considered to be one of the main regulators of drought responses in plants (Daszkowska-Golec, 2016, Sah et al., 2016). It is needed for root elongation under low water potential conditions (Sharp et al., 1994), while restricting ethylene production (Tan et al., 1989, Sharp et al., 1994). It also controls guard cells by promoting closure upon stress (Mittler & Blumwald, 2015, Riemann et al., 2015). One explanation for the downregulation of genes related to ABA could be that plants treated with NaCl have more water in the shoots than plants grown without the addition of NaCl; thus, removing its need for increased water uptake and stomata closure. Conversely, plants grown in the absence of NaCl might be water stressed, promoting the biosynthesis and response to ABA. Auxins play a major role in plant development and its concentration have been observed to be reduced during salt stress (Dunlap & Binzel, 1996). They are also crucial for root architecture during salt stress (Wang et al., 2009b). The increased auxin related genes in shoots of S. bigelovii plants treated with NaCl could be related to their increased growth. Contrarily, plants grown in 600 mM NaCl for 6 weeks might be experiencing salt stress and might require the downregulation of auxin related genes.

Peroxidases were upregulated in all NaCl treatments and were overrepresented in plants treated with 600 mM NaCl for 1 week (Fig. 26).

Peroxidases are required for detoxification of reactive oxygen species (ROS) 152

153

Figure 26 (cont.)

154

Figure 26 (cont.)

Figure 26. Biotic and abiotic stress MapMan pathway for plants treated with: a) 50 mM for 1 week; b) 200 mM for 1 week; c) 600 mM for 1 week; d) 50 mM for 6 weeks; e) 200 mM for 6 weeks; and f) 600 mM for 6 weeks. Squares represent differentially expressed genes at the corresponding treatment relative to plants grown without the addition of NaCl. Red squares represent downregulation, blue squares upregulation. 155

(Mittler et al., 2004, Asada, 2006) which are produced through photorespiration and under salt stress (Hossain & Dietz, 2016). ROS waves have been observed in response to salt and are suggested to help in signal transduction in response to salt

(Evans et al., 2016, Choi et al., 2017). ROS are toxic to the cell, oxidizing cellular components leading to cellular damage, premature senescence, and necrosis

(Møller et al., 2007, Hossain & Dietz, 2016). Increased peroxidases would help in the scavenging of ROS, preventing cellular damage.

Signaling was one of the MapMan bins with the highest number of genes allocated to it. Signaling was primarily overrepresented in plants treated with 600 mM NaCl for 1 week, but was also abundant in plants treated with NaCl for 6 weeks (Fig. 26). This may suggest that changes in signaling responses are required for salt response and adaptation, particularly once ions have been accumulated in shoots. Leucine rich repeat (LRR) receptor-like kinases were overrepresented and overexpressed in plants treated with 600 mM NaCl for 1 week (Fig. 26). LRR receptor-like kinases are involved in signal transduction and have been associated with plant development, biotic stress, and abiotic stress (Torii, 2004). Interestingly, the most abundant receptors in salt treated plants, predominantly in plants treated with 600 mM NaCl for 1 week and NaCl treated plants for 6 weeks (Figs. 27 and

28), were of the domain of unknown function 26 (DUF26) family. This domain has been renamed as Salt stress response/antifungal. It was first characterized through apoplastic proteomics of rice exposed to 200 mM NaCl for 1 ,3, and 6 hours and named OsRMC (Zhang et al., 2009b). This protein was already about 156

Figure 27. Receptor-like kinases MapMan pathway for plants treated with NaCl for 1 week. Squares represent differentially expressed genes at the corresponding treatment relative to plants grown without the addition of NaCl. Red squares represent downregulation, blue squares upregulation. 157

Figure 28. Receptor-like kinases MapMan pathway for plants treated with NaCl for 6 weeks. Squares represent differentially expressed genes at the corresponding treatment relative to plants grown without the addition of NaCl. Red squares represent downregulation, blue squares upregulation. 158

ten times more abundant 6 hours after salt imposition. Knockdown lines of

OsRMC in rice, displayed normal growth when grown without NaCl, but displayed reduced growth when treated with 100 mM NaCl (Zhang et al., 2009b).

However, when plants were exposed to higher concentrations of NaCl, of 150 and

200 mM for 4 days and allowed to recover, OsRMC knockdown lines had increased survival rates and decreased membrane lipid peroxidation compared to wild type (Zhang et al., 2009b). In another study, knockdown of OsRMC led to increased susceptibility to JA and reduced shoot growth when treated with JA

(Jiang et al., 2007). S. bigelovii plants treated with NaCl had downregulation of several DUF26 receptor-like kinase, predominantly in plants treated with 600 mM

NaCl for 1 week and plants treated with NaCl for 6 weeks (Figs. 27 and 28). This evident downregulation could help in the survival of S. bigelovii plants with high concentrations of Na+ and Cl- in its shoots. This downregulation of DUF26 receptor-like kinases could work in conjunction with the downregulation of genes related with JA, preventing the reduced effect seen in rice of knockdown lines of

OsRMC upon the addition of JA (Jiang et al., 2007). Nevertheless, further studies are needed to understand the function of these receptors and its interplay with hormones.

Plants treated with 600 mM NaCl for 1 week showed particular responses not observed for any other treatment. In particular, an overrepresentation and upregulation of transporter proteins at the mitochondrial membrane and 159

Figure 29. MapMan pathway for plants treated with 600 mM NaCl for 6 weeks. a) Mitochondrial transport. b) RNA and protein synthesis. Squares represent differentially expressed genes at the corresponding treatment relative to plants grown without the addition of NaCl. 160

mitochondrial ATP synthesis proteins was observed. This increased ATP synthesis could suggest that the cell is at the demand of sufficient energy to cope with the necessary processes required for salt tolerance. Additionally, there was an upregulation in proteins related to the secretory pathway and ribosomal proteins in the chloroplasts. Proteins related to post-translational modification and protein degradation were both upregulated and downregulated, suggesting that proteins might go through specific modifications when cells are exposed to high concentrations of NaCl.

4.3.3 Salicornia bigelovii membrane proteins differential abundance analysis and generation of organellar membrane proteomes

RNASeq data is useful as it measures mRNA abundance. However, not all transcripts are translated into proteins and different proteins have different turnover rates and half-lives (Nelson & Millar, 2015, Li et al., 2017). Similarly, some transcripts are more stable than others and can be translated into proteins several times (Narsai et al., 2007). Furthermore, transcript and protein abundances have not been well correlated (Maier et al., 2009, Vogel & Marcotte, 2012, Bai et al., 2015), presumably due to the aforementioned reasons. Since proteins are the machinery of the cells, it was decided to assess protein abundances in shoots of S. bigelovii treated with 0, 50, 200, and 600 mM NaCl for 6 weeks. Because S. bigelovii accumulates high concentrations of both Na+ and Cl- in its shoots, transporters 161

must play a key role in the salt tolerance of this plant. Therefore, it was decided to focus on membrane proteins.

It is considered that Na+ is primarily accumulated in the vacuoles to prevent

Na+ toxicity in the cytoplasm (Flowers et al., 1986, Munns & Tester, 2008). For this reason, particular emphasis was put on the identification of tonoplast proteins responsive to salt treatment. To achieve this, a separation of organellar membranes must be obtained. Organellar membranes can be separated based on their densities by ultracentrifugation (Hodges et al., 1972, Yang & Murphy, 2013). Nevertheless, organelles have similar densities and vesicle trafficking across different organelles add to the difficulties in their separation (Dunkley et al., 2006). Thus, membrane separation by ultracentrifugation does not result in pure isolates but results in membrane fractions enriched for a specific organelle (Sadowski et al., 2006). The abundances of each protein in each fraction can be correlated to known organellar markers and assigned to a subcellular compartment based on its abundance profile

(Dunkley et al., 2004, Gatto et al., 2014, Breckels et al., 2016).

S. bigelovii microsomal fractions were obtained from 100 g of fresh shoots from plants treated with 0, 50, 200, and 600 mM NaCl for 6 weeks and in triplicates.

Microsomal fractions were loaded on top of a continuous sucrose density gradient from 20 - 45 % (Fig. 30). Membranes were separated by their isopycnic points by ultracentrifugation 100,000 x g, resulting in the recovery of 12 fractions with different densities (Fig. 31). Samples were carbonate washed to remove any remaining water-soluble proteins trapped within the microsomes 162

Figure 30. Separation of organellar membranes by their densities through ultracentrifugation. a) Representation of the separation of organellar membranes through a continuous sucrose density gradient from 20 – 45 % through ultracentrifugation. b) Continuous sucrose density gradient of S. bigelovii shoot microsomal fraction after ultracentrifugation. c) The recovered 12 fractions from the continuous sucrose density gradient, each enriched with membranes of different organelles. 163

(Dunkley et al., 2006). The specific gravities of the different fractions were measured with a digital refractometer (Brix/RI-Chek, Reichert). Specific gravity is defined as the relative density of one substance to another, in this case sucrose to water. The specific gravities for the density fractions ranged from 1.08 – 1.19 and should cover all the different organelles based on the specific gravities of

Arabidopsis organellar membranes (Yang & Murphy, 2013) (Figs. 31 and 32). The sucrose % of each fraction, from which specific gravity was calculated, was very consistent across samples from each fraction and varied by a maximum 0.6 % sucrose (Fig. 31a) or in their specific gravity by a maximum of 0.004 g/cm3 (Fig.

31a and 32a). Protein concentration was measured by Bradford assay (Bradford,

1976) and varied greatly across fractions. The least dense fraction (1) had the lowest protein concentration, whereas fraction number 10 had the highest concentration (Fig. 32a). This is expected as it is considered that differences in membrane densities are primarily caused by differences in the amount of membrane proteins (Chang et al., 2008).

Once the density fractions were obtained, proteins in each fraction were identified by shotgun proteomics. Despite membrane proteins accounting for ~30 % of the total proteins (Wallin & Heijne, 1998, Tan et al., 2008, Wiśniewski et al., 2009), membrane proteins are often underrepresented in proteomics studies due to their hydrophobicity leading to sedimentation (Tan et al., 2008, Rabilloud, 2009, Møller et al., 2011). Denaturing solvents such as urea, thiourea, SDS, and CHAPS are more appropriate for the solubilization of these proteins (Tan et al., 2008); however, 164

Figure 31. a) Sucrose % and specific gravity of each of the 12 density fractions recovered from shoots of S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl for 6 weeks. b) Specific gravity of different organellar membranes from Arabidopsis thaliana

(Yang & Murphy, 2013). Specific gravity is defined as the substance density relative to water.

165

Figure 32. a) Protein concentration in µg/mL and specific gravity of each of the 12 density fractions recovered from shoots of

S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl for 6 weeks. b) Specific gravity of different organellar membranes from Arabidopsis thaliana (Yang & Murphy, 2013). Specific gravity is defined as the substance density relative to water.

166

these solvents are not compatible with conventional proteomics approaches such as 2-dimensional gel electrophoresis (Tan et al., 2008). Moreover, conventional protein digesting enzymes are susceptible to highly denaturing environments, adding another difficulty to membrane proteomic studies. To overcome this, proteins from the 12 density fractions were digested through a filter aided sample preparation (FASP) approach in a gel-free system (Wisniewski et al., 2009), which has been shown to be suitable for transmembrane proteins (Wisniewski et al., 2009,

Tanca et al., 2014). In short, it consists of solubilizing membrane proteins in SDS, urea or thiourea, loading them into a filter where proteins are digested in non- denaturing conditions, followed by centrifugation releasing the digested peptides ready to be identified by mass spectrometry.

For protein identification, the assembled proteomes of S. bigelovii, S. brachiata, and S. europaea were combined into one dataset and had its redundancy reduced by CD-HIT. The combined dataset had better completeness than any of the individual proteomes when assessed with BUSCO v3 against the Viridiplantae dataset (Table 16). Resulting in a 94.3 % of complete proteins, while the best individual proteome had only 86.8 % (Table 16). Proteins were identified with

MASCOT v. 2.4 (Perkins et al., 1999) and SCAFFOLD v4 (Searle, 2010) was used for protein quantitation. For protein quantitation, all fractions were loaded as

MudPit (Washburn et al., 2001) creating one sample composed of all the fractions per treatment. 2,521 were identified in at least one of the 4 treatments.

167

Table 16. Completeness assessment and comparison of the combined Salicornia dataset used for proteomics studies against the proteomes of S. bigelovii, S. brachiata, and S. europaea by BUSCO v3 against the Embryophyta dataset.

Complete single Complete Fragmente Transcriptome (proteins) Complete Missing copy duplicated d

Ma_SAEU1 (110,851) 613 (42.6 %) 334 (23.2 %) 279 (19.4 %) 362 (25.1 %) 465 (32.3 %)

1136 S. bigelovii (531,223) 309 (21.5 %) 827 (57.4 %) 162 (11.3 %) 142 (9.8 %) (78.9 %) 1231 S. brachiata (292,468) 631 (43.8 %) 600 (41.7 %) 105 (7.3 %) 104 (7.2 %) (85.5 %) 1250 S. europaea (212,359) 637 (44.2 %) 613 (42.6 %) 91 (6.3 %) 99 (6.9 %) (86.8 %) 1173 S. bigelovii shoots (269,895) 286 (19.9 %) 887 (61.6 %) 145 (10.1 %) 122 (8.4 %) (81.5 %) Combined Salicornia dataset 1358 75 (5.2 %) 1283 (89.1 %) 26 (1.8 %) 56 (3.9 %) (929,540) (94.3 %) 1. Transcript sequences from: Ma et al. (2013) 168

To identify proteins that are salt responsive, proteins were classified based on their differential abundance across the 4 salt treatments (Table 17). 960 proteins were differentially abundant in at least one of the treatments with a significance of p ≤

0.05 and corrected for multiple testing with Hochberg-Benjamini correction

(Benjamini & Hochberg, 1995) (Table 17). The highest number of differentially abundant proteins was found in the 0 and 50 mM treatments when compared to

200 and 600 mM NaCl. 384 were more abundant in in the lower salt treatments and 277 more abundant in the higher salt treatments. 108 proteins were less abundant in plants growing without the addition of NaCl compared to plants treated with any salt treatment and 63 proteins were more abundant in plants treated with 600 mM NaCl (Table 17).

Hierarchical clustering of the differentially abundant proteins showed a clear clustering between replicates from each treatment and between low and high salt treatments (Fig. 33). Two main clusters were generated with plants treated with 0 and 50 mM NaCl on one side and plants treated with 200 and 600 mM NaCl clustering on the other side of the tree. This indicated, that at a membrane protein level, plants treated with higher concentrations of NaCl respond to NaCl in a very similar way. Additionally, plants treated with 0 and 50 mM NaCl behaved, in terms of membrane proteins abundances, very similarly. Since the optimum growth of S. bigelovii is observed at 200 mM NaCl, it is possible that plants treated with 0 and 50 mM NaCl are responding to the low concentrations of NaCl.

169

Table 17. Classification of proteins based on their differential abundances and their directionality across treatments identified with SCAFFOLD v4. 0 mM 50 mM 200 mM 600 mM Number of proteins up up down down 384 down down up up 277 up down down down 108 down down down up 63 down up down down 42 down up up up 36 up up up down 16 down up up down 13 down down up down 7 up down up up 7 up down down up 3 up up down up 2 up down up down 2 Total: 960

Similarly to MapMan enrichment analysis (Thimm et al., 2004, Usadel et al.,

2009) of the differentially expressed genes, MapMan enrichment was assessed for the differentially abundant proteins. The combined Salicornia dataset had its proteins assigned to MapMan bins through Mercator (Lohse et al., 2014). MapMan analysis was carried out with proteins differentially abundant between 0-50 mM and 200-600 mM treatments, as the majority of the differentially abundant proteins arose from this pattern (Table 17). The significantly overrepresented MapMan bins can be found in Table 18. 170

Figure 33. Hierarchical clustering of three replicates per treatment of differentially abundant proteins in shoots of S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl for 6 weeks. 960 proteins had significantly different abundances. Blue, reduced abundance;

Yellow, increased abundance. 171

Table 18. Overrepresented MapMan bins for differentially abundant proteins from shoots of S. bigelovii with increased or decreased abundance between 0-50 mM when compared to 200-600 mM NaCl treatments. Overrepresentation was tested with Wilcoxon Rank Sum test and corrected for multiple testing with Benjamini Hochberg with p value < 0.05.

Bin Description Elements p value 1 PS 23 2.3E-4 34 transport 81 7.2E-7 29.3.3 protein targeting chloroplast 20 0.001 9 mitochondrial electron transport / ATP synthesis 39 0.002 31.1 cell organisation 23 0.005 31.1.1 cell organisation cytoskeleton 21 0.005 1.3 PS calvin cycle 8 0.007 31.1.1.2 cell organisation cytoskeleton microtubuli 14 0.008 29.2 protein synthesis 19 0.010 mitochondrial electron transport / ATP synthesis 9.1 NADH-DH 20 0.011 27 RNA 10 0.017 mitochondrial electron transport / ATP synthesis 9.1.2 NADH-DH localisation not clear 15 0.017 29.2.1 protein synthesis ribosomal protein 18 0.020 transport ABC transporters and multidrug 34.16 resistance systems 21 0.020 29.5.7 protein degradation metalloprotease 4 0.020 4 glycolysis 6 0.024 4.1 glycolysis cytosolic branch 6 0.024 1.1 PS light reaction 15 0.029 19 tetrapyrrole synthesis 6 0.039 29.2.1.1 protein synthesis ribosomal protein prokaryotic 8 0.039 1.3.6 PS Calvin cycle aldolase 4 0.039 172

Bin Description Elements p value 4.1.10 glycolysis cytosolic branch aldolase 4 0.039 4.3 glycolysis unclear/dually targeted 4 0.039 4.3.10 glycolysis unclear/dually targeted aldolase 4 0.039 1.1.1 PS light reaction photosystem II 6 0.039 29.5.11. protein degradation ubiquitin proteasome 7 0.044 20 29.5.11 protein degradation ubiquitin 10 0.047

The most overrepresented terms were Photosynthesis and, unsurprisingly,

Transport. Whilst at first glance no overrepresentation of cell wall related genes was identified, which were most abundant and shared a common response in gene expression in plants treated with NaCl. Pathway analysis revealed overrepresentation and overexpression of proteins involved in the production of

UDP-D-Glucose, which is the building block of cellulose (Saxena & Brown, 2005), and UDP-L-rhamnose (Fig. 34), which is required for the biosynthesis of cell wall components such as rhamnogalacturonan-I and II (Oka et al., 2007), which are two of the three major pectic polysaccharides in primary cell walls (Ridley et al., 2001).

Consistent with gene expression analysis, receptor-like kinases with the

DUF26 domain (Fig. 28), were downregulated in plants treated with higher concentrations of NaCl (Fig. 35 a). Similarly, proteins in the secretory pathway and ribosomal proteins in the plastids were more abundant in high salt treatments 173

(Figs. 29 b and 35 c). Proteins related to post-translational modification and protein degradation were both upregulated and downregulated (Fig. 35 c).

Figure 34. MapMan pathway of cell wall precursors in plants treated with NaCl for 6 weeks. Squares represent Log2 fold change in abundance of proteins in 200-600 mM when compared to 0-50 mM treatments.

174

Figure 35. MapMan pathway of cell wall precursors in plants treated with NaCl for 6 weeks. a) Receptor-like kinases; b)

Photosynthesis and mitochondria electron transport; c) Synthesis of proteins and regulation. Squares represent Log2 fold change in abundance of proteins in 200-600 mM when compared to 0-50 mM treatments. 175

Proteins related to photosynthesis and mitochondria electron transport were more abundant in plants treated with high concentrations of NaCl (Fig. 35 b). Whereas proteins related to ascorbate and glutathione were less abundant (Fig. 35 b).

Photosynthesis related proteins were more abundant in plants treated with high concentrations of NaCl, having different processes upregulated such as biosynthesis of chlorophyll by tetrapyrroles, proteins involved in the light chain reactions of photosynthesis, and the Calvin cycle (Fig. 35 b). At the thylakoid level, plants treated with higher NaCl concentrations had increased abundance in proteins related to photosystems I and II, cytochromes, and ATP synthase, suggesting that the whole photosynthetic apparatus is more active in plants treated with 200 and 600 mM NaCl compared to plants treated with 0 and 50 mM

NaCl (Fig. 36). This is congruent with the improved growth of S. bigelovii plants upon the addition of NaCl (Chapter 3, Fig. 23).

Proteins related to electron transport in ATP synthesis in mitochondria were more abundant in plants treated with higher concentrations of NaCl, except for AOX1 (Fig. 37). AOX1 has been associated with oxidative stress and is considered a marker for oxidative stress with the cell (Vanlerberghe, 2013).

Conversely, proteins related to metabolite transport in the mitochondrion were less abundant in plants exposed to NaCl (Fig. 37). The overall responses of S. bigelovii to higher concentrations of NaCl (200 and 600 mM) is showing increased 176

Figure 36. Photosynthesis MapMan pathway in plants treated with NaCl for 6 weeks. Squares represent Log2 fold change in abundance of proteins in 200-600 mM when compared to 0-50 mM treatments. 177

Figure 37. Photosynthesis MapMan pathway in plants treated with NaCl for 6 weeks. Squares represent Log2 fold change in abundance of proteins in 200-600 mM when compared to 0-50 mM treatments. 178

abundance in proteins related to ATP synthesis, photosynthetic machinery, and cell wall biosynthesis.

To investigate organelle specific responses and have a better understanding of the subcellular compartmentalization of membrane proteins in S. bigelovii, an organellar membrane proteome was generated for S. bigelovii plants treated with

200 mM NaCl by correlating the abundances of each protein across the 12 density fractions, to that of organelle marker proteins (Trotter et al., 2010) with the R packages pRoloc and pRolocdata (Gatto et al., 2014, Breckels et al., 2016). Marker proteins for S. bigelovii were identified by homology searches by BLAST to

Arabidopsis organelle marker proteins retrieved from pRolocdata and originated from previous studies (Dunkley et al., 2006, Ferro et al., 2010, Trotter et al., 2010,

Nikolovski et al., 2012, Groen et al., 2014, Nikolovski et al., 2014). The best BLAST hit of each marker with a protein from the combined Salicornia dataset were selected as candidate markers. In silico subcellular localization prediction of the candidate marker was obtained with mGOASVM Plant V2 (Wan et al., 2012). To assess the predictor’s specificity, all organellar Arabidopsis marker proteins were in silico predicted for their subcellular localization. 86.7% of the proteins were accurately predicted to their experimentally observed subcellular localization.

Candidate marker proteins for Salicornia whose predicted subcellular localization coincided with the subcellular localization of their BLAST hit with Arabidopsis markers and which were relatively abundant, were used to create the Salicornia 179

organellar marker set (Table 19). Golgi markers were not included in the analysis as the abundances of Golgi markers were low and noisy.

Table 19. Number of protein markers per organelle for Salicornia and Arabidopsis. Salicornia Marker Arabidopsis Marker Plasma membrane 31 60 ER membrane 21 42 Tonoplast 16 18 Thylakoid 36 56 Mitochondrion 64 103 Total 168 279

S. bigelovii organellar membrane proteome was generated by correlating the relative abundances of proteins with unknown subcellular localization with the protein abundance profiles of the predicted Salicornia organelle marker proteins in each density fraction (Fig. 38a). To validate the abundance profiles of marker proteins, western blots with antibodies against marker proteins were done on the

12 density fractions (Fig. 38b). The protein abundance profiles generated with pRoloc from the Salicornia organelle marker set, followed the same pattern as that of western blots against organelle marker proteins. This indicates that Salicornia marker protein profiles generated with pRoloc, follow the real pattern of protein enrichment found in each density fraction. Moreover, it validates the approach for 180

Figure 38. Validation of marker abundance profiles generated with pRoloc by Western blot. a) Relative protein abundance profiles identified of organelle marker proteins on the 12 density fractions by pRoloc. b) Western blot analyses of organelle marker proteins on 5 µg of proteins from each of the 12 density fractions. 181

protein subcellular localization allocation with pRoloc and validates the proteomics shotgun methodology and its execution. Plasma membrane proteins were more abundant in the least dense fractions (Fig. 38). This was unexpected as

Arabidopsis plasma membrane has a specific gravity between 1.14 -1.17 g /cm3

(Yang & Murphy, 2013), which would correspond somewhere between fractions 8

– 10. The other organelles had equivalent densities to those of Arabidopsis organelles (Yang & Murphy, 2013). Tonoplast proteins were found to be more abundant between fractions 4 – 6 with a specific gravity between 1.10 – 1.12 g /cm3.

Thylakoids were more abundant in the denser fractions with a specific gravity between 1.15 -1. 19 g /cm3. Mitochondria followed a very similar pattern to thylakoids and were both most abundant at the denser fractions. Similarly, ER membrane proteins followed a very similar pattern to plasma membrane and was most abundant at the least dense fractions (Fig. 38).

Protein subcellular localization prediction for proteins with unknown subcellular localization was generated with the support vector machine algorithm and 100 iterations. A total of 880 proteins were allocated to one subcellular localization with high confidence (Table 20). Most of the allocated proteins corresponded to Mitochondrion and plasma membrane. Thylakoid proteins were the least abundant; nevertheless, due to similar densities between mitochondrion and thylakoids, it is likely that several proteins allocated to the mitochondrion are localized in the thylakoids. Visualization of protein profiles was generated with a

PCA (Fig. 39). Proteins were separated in three main groups (Fig. 39). One group 182

Table 20. Organellar subcellular localization allocation by pRoloc of S. bigelovii proteins

Low confidence High confidence

Plasma membrane 401 253

ER 208 101

Mitochondrion 635 327

Thylakoid 192 48

Tonoplast 324 151

Unknown - 880

Total 1760 1760

comprised of plasma membrane and ER proteins, one group of thylakoid and mitochondrion proteins, and one group of tonoplast proteins. The separation makes it difficult to distinguish between proteins localized to the ER and plasma membrane and between proteins localized to the thylakoids and mitochondrion.

Serendipitously, tonoplast proteins were clearly separated from the other organelles, allowing for the identification of tonoplast proteins. Some of these proteins predicted to be localized to the tonoplast were selected for further characterization (Chapter 5).

183

Figure 39. Principal component analysis of protein predicted subcellular localization by pRoloc. The size of the spot represents confidence values of localization. Tonoplast: yellow, markers as open triangles; Plasma membrane: green, markers as squares; ER: red, markers as circles; Thylakoid: purple, markers as open squares; Mitochondrion: blue, markers as open circles. 184

4.4 Conclusion

The transcriptomes of S. bigelovii, S. brachiata, and S. europaea were assembled and annotated. The transcriptomes were of greater quality when compared to the only published transcriptome of S. europaea (Ma et al., 2013). S. bigelovii shoot transcriptome had 0.045 % of contamination with prokaryotic, mammalian, and viral sequences, resulting in a clean assembly. The transcriptome was curated by maintaining proteins with Viridiplantae BLAST hits and reducing its redundancy with CD-HIT-EST, resulting in 37,646 genes and 78,085 transcripts. Gene differential expression analysis of shoots of S. bigelovii treated with 0, 50, 200, or

600 mM NaCl for 1 and 6 weeks, resulted in 4,152 differentially expressed genes.

Samples clustered according to their treatments and time points, except for plants treated with 600 mM NaCl for 1 week, which formed a monophyletic clade independent of all the other treatments. After 6 weeks of treatment, plants exposed to 600 mM NaCl had similar expression profiles to those treated with 200 mM

NaCl, indicating that plants had already adapted to the high concentrations of Na+ and Cl-. Pathway and abundance analysis of the differentially expressed genes with MapMan, revealed a common overrepresentation and upregulation of genes related to cell wall biosynthesis and modification for all salt treatments and time points when compared to plants grown without the addition of NaCl. This could be related to the enhanced growth that plants displayed when treated with NaCl.

Interestingly, plants treated with 600 mM NaCl for 1 week showed a high overrepresentation and downregulation of receptor DUF26 (also called Salt stress 185

response / antifungal). DUF26 knockdown lines in rice, did not show any phenotype under control conditions, had decreased salt tolerance at mild levels of salinity, but displayed improved tolerance and recovery from severe salt treatments (Zhang et al., 2009b). It is possible that the downregulation of these receptors was needed for the survival of S. bigelovii plants treated plants with 600 mM NaCl for one week. Plants exposed to NaCl for 6 weeks had less differentially expressed DUF26 proteins, but were also downregulated. No overrepresentation of genes related to apoptosis or cell death were found in plants treated with 600 mM NaCl for 1 week. This may suggest, that despite facing a high concentration of NaCl and displaying different expression patterns compared to other treatments, plants were not stressed to the extent where cell death is activated.

The membrane proteome of S. bigelovii was generated from a continuous sucrose density gradient with 12 density fractions enriched with membranes of different organelles from shoots of plants treated with 0, 50, 200, and 600 mM NaCl for 6 weeks. Shotgun proteomics by filter aided sample preparation of the 12 different fractions resulted in the identification of 2,521 proteins. 960 proteins were differentially abundant in at least one of the treatments. Clustering analysis revealed a clear grouping between samples from the same treatment and a tree bifurcation with 0 and 50 mM NaCl treatments clustering together and 200 and

600 mM clustering on the other side of the tree. Most of the differentially abundant proteins followed the pattern of either increasingly abundant in 0 and 50 mM treatments with respect to 200 and 600 mM NaCl treatments, or a reduction in 186

abundances between 0 and 50 mM with respect to 200 and 600 mM NaCl treatments. MapMan analysis of 0 and 50 mM vs 200 and 600 mM proteins abundances, revealed increased protein abundances for key enzymes in the production of cellulose and pectins for cell wall biosynthesis. Plants treated with

200 and 600 mM NaCl had increased abundance in proteins involved in photosynthesis and in the electron chain reaction and ATP production in the mitochondria. This might suggest that, at a cellular level, plants treated with higher concentrations of NaCl had improved energy production and cell growth.

Similarly to gene differential expression analysis, proteins of the DUF26 family were less abundant in plants treated with high concentrations of NaCl. The first organellar membrane proteomes for S. bigelovii were generated by comparing the abundance profiles of marker proteins across the 12 density fractions. Marker abundance profiles were validated by Western blot, showing the same pattern with in silico abundance profiles, validating the whole shotgun proteomics approach. 880 proteins were allocated to the plasma membrane, ER, tonoplast, thylakoid, or mitochondrion. Plasma membrane in S. bigelovii had a lower specific gravity compared to that of Arabidopsis and showed a similar abundance profile with the ER. Similarly, thylakoids and mitochondria had similar abundance profiles. Tonoplasts separated nicely from the other organelles, allowing for a high confidence in the identification of tonoplast proteins. Both, transcriptomic and proteomic analyses will serve as the basis for the identification of candidate proteins involved in salinity tolerance in S. bigelovii. The selection and 187

characterization of some candidate proteins involved in salt tolerance in S. bigelovii is discussed in Chapter 5.

188

Chapter 5. Selection and characterization of candidate genes related to salt tolerance in Salicornia bigelovii

Saicornia bigelovii responds to different concentrations of NaCl at a transcriptomic and proteomic level (Chapter 4). As a general response, genes and proteins related to cell wall modification, photosynthesis, and ATP synthesis are more abundant in plants treated with higher concentrations of NaCl. In this chapter, the selection and characterization of candidate proteins involved in salt tolerance in S. bigelovii is discussed. Emphasis is put on transmembrane proteins, particularly to those likely to be localized to the tonoplast.

5.1 Introduction

Gene differential expression studies result in identification of thousands of differentially expressed genes. Whilst a general response to treatments can be often observed, the high number of genes varying in expression renders the identification of key genes related to the study of interest challenging (Han et al.,

2015). Several RNASeq studies have tried to address salt tolerance in different plant species, but generally result in the report of broad responses such as ABA signaling and oxidative stress response (Wang et al., 2015, Li et al., 2016, Zhang et al., 2016, Luo et al., 2017). The high number of differentially expressed genes has greatly limited the translation from RNASeq studies to increased salt tolerance in 189

plants. In the case of Salicornia, no RNAseq studies have been translated to further characterization and increased salt tolerance of plants. However, choline monooxygenase was identified through a proteomics study in Salicornia europaea

(Wang et al., 2009a) and its expression in tobacco increased its salt tolerance (Wu et al., 2010).

Very few plant ion transporter proteins have been shown to be important for salt tolerance (Roy et al., 2014). Most notably, salt overly sensitive 1 (SOS1) has been described as a plasma membrane Na+/H+ antiporter and has been associated with Na+ extrusion from cells (Shi et al., 2000). Its downregulation in Eutrema salsugineum resulted in increased Na+ accumulation in roots (Oh et al., 2009).

Proteins of the sodium / hydrogen exchanger (NHX) family, were first identified by Apse et al. (1999) and were thought to be responsible for the vacuolar Na+/H+ antiporter activity in plants. Nevertheless, these proteins were later described to have a primary role as a K+/H+ in planta (Rodríguez-Rosales et al., 2008, Leidi et al., 2010, Barragan et al., 2012, Xu et al., 2013, McCubbin et al., 2014). High affinity potassium transporter (HKT) proteins, have been associated with retrieval of Na+ from the xylem, reducing Na+ accumulation in leaves (Sunarpi et al., 2005,

Davenport et al., 2007, Plett et al., 2010). Proteins of the cation / hydrogen exchanger (CHX) family have been implicated in K+ and pH homeostasis (Zhao et al., 2008, Chanroj et al., 2011, Chanroj et al., 2012). Loss of function of Glycine max

GmCHX1 led to salt sensitivity (Qi et al., 2014). Some cyclic nucleotide gated 190

channels (CNGC) have been shown to be permeable to Na+ and have been suggested as possible pathways for Na+ uptake (Deinlein et al., 2014, Hua et al.,

2003). Downregulation of AtCNGC10 resulted in reduced biomass upon salt stress

(Guo et al., 2008).

In this chapter, differentially expressed genes and differentially abundant proteins in shoots of S. bigelovii in response to different NaCl treatments (Chapter

4) are selected for further characterization in yeast. Particular emphasis is put on transmembrane proteins annotated as ion transporters and channels, and to proteins likely to be localized to the tonoplast. Candidate genes are cloned into yeast and screened for salt tolerance, resulting in the identification of salt related genes.

5.2 Materials and methods

5.2.1 Gene selection

Three different sets of genes were selected for characterization in yeast derived from the transcriptomic and proteomics studies described in Chapter 4. Genes upregulated upon NaCl treatment, genes downregulated upon NaCl treatment, and genes conventionally associated with NaCl stress regardless of their expression profile. A total of 63 genes were selected for characterization. Primers to amplify coding regions of each candidate gene were designed with and without 191

stop codons with Primer3Plus (Untergasser et al., 2007) (Appendix table 6). Their secondary structures were predicted with Protter (Omasits et al., 2014).

5.2.2 RNA extraction, and cloning of candidate genes into E. coli

RNA was extracted from shoots of S. bigelovii plants grown in 200 mM NaCl using

TRIzol™ with RNA MiniPrep kit (Zymo Research, USA), following the manufacturer’s protocol. RNA quality was assessed using agarose gel electrophoresis. cDNA was prepared from the extracted, good quality RNA with

SuperScript™ II (Invitrogen, USA), following the manufacturer’s protocol. Genes were amplified by PCR using a high fidelity KOD DNA polymerase (Toyobo,

Japan) and ranging from 30 – 35 cycles depending on the transcript’s abundance.

Amplified genes were A’ tailed by 30 min of extension with Taq DNA polymerase

(Thermo Scientific), creating a one adenine overhang at the end of the PCR product.

These fragments were cloned into pCR™8/GW/TOPO® (Invitrogen, USA) and transformed into TOP10 or MACH strains of E. coli, following the manufacturer’s protocol. Plasmids from positive colonies were extracted by alkaline extraction

(Birnboim & Doly, 1979). Correct insert insertion and orientation were assessed by restriction enzyme digestion and agarose gel electrophoresis. Genes whose cloning resulted in only inverted transcripts, were cloned directionally using pENTR™/D-TOPO® (Invitrogen, USA) with amplified PCR products with a primer containing the 4 required nucleotides for directional cloning CACC and 192

proceeded to transformation, plasmid extraction, and fragment insertion validation as described for pCR™8/GW/TOPO® clones. Since S. bigelovii is a tetraploid undomesticated species, several isoforms per gene might exist. Entry clones were sequenced and compared with the reference transcriptome. Genes that did not match the sequence were re-amplified and cloned. If the same polymorphisms were observed in subsequent amplifications, they were considered natural allelic variants and proceeded to further analysis. In the case two or more alleles were cloned, alleles sharing equal sequences with S. brachiata or S. europaea were selected as it was assumed that conserved alleles across species would be more likely to be functional.

5.2.3 Yeast transformation, spot assays, and subcellular localization

Validated entry clones were recombined with Gateway™ LR Clonase™ II

(Invitrogen, USA) into pAG426GAL-ccdB-EGFP. pAG426GAL-ccdB-EGFP was a gift from Susan Lindquist (Addgene plasmid # 14203). Correct inserts were validated with restriction enzyme digestion and agarose gel electrophoresis.

Plasmids with correct inserts were then sequenced to exclude possible mutations.

Plasmids were transformed into the salt sensitive AXT3 (Δena1-4::HIS3

Δnha1::LEU2 Δnhx1::TRP1) yeast strain by the lithium acetate method (Daniel

Gietz & Woods, 2002). Yeast spot assays were carried out in synthetic defined (SD) medium: 1.7 g/L BD™ Difco™ yeast nitrogen base without amino acids and 193

ammonium sulfate, 5 g/L ammonium sulfate, 0.77 g/L complete supplement mix

(CSM) Drop-out – uracil (Formedium), adjusted to pH 5.6 with KOH, and supplemented with either 2 % glucose or galactose; or in arginine phosphate (AP) medium (Rodríguez-Navarro & Ramos, 1984): 10 mM L-arginine, 8 mM phosphoric acid, 2 mM MgSO4, 0.2 mM CaCl2, 20 µg/L biotin, 2 mg/L calcium pantothenate, 2 µg/L folic acid, 400 µg/L niacin, 200 µg/L riboflavin, 10 mg/L inositol, 200 µg/L p-aminobenzoic acid, 400 µg/L ZnSO4, 400 µg/L pyridoxine hydrochloride, 400 µg/L thiamine hydrochloride, 40 µg/L CuSO4, 500 µg/L

H3BO3, 100 µg/L KI, 200 µg/L FeCl3, 200 µg/L Na2MoO4, 1mM KCl, 0.77 g/L complete supplement mix (CSM) Drop-out – uracil (Formedium), adjusted to pH

6.5 with L-arginine, and supplemented with either 2 % glucose or galactose.

Yeast spot assays were done with three different colonies per construct. In short, to normalize growth, yeast transformed with pAG426GAL-gene-EGFP were grown in SD or AP medium with 2 % glucose for two days at 30 °C. 200 µL of the pre-grown yeast were transferred into 1.8 mL of medium with 2 % glucose and grown for 16 hours at 30 °C. The OD600 was measured and each sample was adjusted to an OD600 of 0.6 by diluting with medium without any carbon source.

For each sample, four 10-fold dilutions were prepared. 10 µL per sample were spotted in each SD or AP plate. SD plates consisted in SD medium with 2 % agar and supplemented with either 0, 100 mM, 150 mM, 175 mM, and 200 mM NaCl or

800 mM KCl. SD plates were supplemented with 2 % galactose for all treatments and with 2 % glucose for plates with 0 mM NaCl as control. AP plates consisted of 194

AP medium with 2 % agar and supplemented with either 0, 20 mM, 30 mM, 40 mM, and 60 mM NaCl. AP plates were supplemented with 2 % galactose for all treatments and with 2 % glucose for plates with 0 mM NaCl as control. Plates were grown at 30 °C for 2 – 5 days. Subcellular localization of proteins was analyzed by

EGFP fusions in the C’ terminus of each protein and observed with a confocal laser scanning microscope (LSM 710, Zeiss, Germany). Images were acquired with Zen

2009 image software (Zess, Germany) and analyzed with Fiji image analysis software (Schindelin et al., 2012).

5.3 Results and discussion

5.3.1 Gene selection for characterization

The general responses in gene differential expression analysis and differential protein abundances in response to different salt treatments in S. bigelovii were discussed in Chapter 4. Based on these studies, a number of genes were selected for further characterization in yeast. Protein selection was based on different criteria. The first criterion was to focus on transmembrane proteins, preferentially in those with several transmembrane helices (Table 21), as transporters are likely to play a key role in the extrusion of Na+ from the cytosol. Transmembrane helices can be often confused for signal peptides (Krogh et al., 2001); thus, the high number of proteins predicted to have 1 transmembrane helix (Table 21).

195

Table 21. Number of proteins in S. bigelovii shoot transcriptome with predicted transmembrane helices with tmHMM1

Transmembrane helices Number of proteins 1 25692 2 6676 3 1820 4 1311 5 801 6 707 7 459 8 327 9 233 10 314 11 188 12 153 13 51 14 38 15 13 16 6 17 3 22 2 27 1 33 1 1. (Krogh et al., 2001)

Proteins with 9 – 13 transmembrane helices were given priority, as cation / H+ antiporters such as NHX1 and SOS1 are reported to have 12 transmembrane helices (Yamaguchi et al., 2003, Sato & Sakaguchi, 2005, Shi et al., 2000). However, in other species such as poplar, NHX proteins have been predicted to have 10 196

transmembrane helices (Tian et al., 2017), suggesting that variation across species exist. Nevertheless, some proteins with fewer transmembrane helices were also selected. Channels are considered to play an important role in cellular Na+ uptake

(Davenport & Tester, 2000, Demidchik & Tester, 2002). Channels such as cyclic nucleotide gated channels and aquaporins such as Plasma membrane Intrinsic

Proteins (PIPs) and Tonoplast Intrinsic Proteins (TIPS), have been reported to have

6 transmembrane helices (Kaplan et al., 2007, Afzal et al., 2016, Shivaraj et al., 2017).

Therefore, proteins with around 6 transmembrane helices were also selected.

Another criterion for protein selection was their transcript expression and protein abundance at different salt treatments. Because S. bigelovii requires relatively high concentrations of NaCl for its correct development (Figs. 11 and 23), it is unclear which response would the plant have to salt. S. bigelovii could expect high concentrations of NaCl as it grows at the sea shores and might induce the expression of proteins related to Na+ transport upon Na+ deficiency. Alternatively, it could induce the expression of these genes only in response to NaCl.

Furthermore, cell specific induction could take place. Proteins involved in Na+ accumulation could be present in the cortex, preventing Na+ from reaching the photosynthetically active chlorenchyma. Higher accumulation of Na+ in cortical non-chlorenchymatic cells compared to chlorenchymatic cells was observed in S. brachiata (Reddy et al., 1993) and S. europaea (Lv et al., 2012b). Hence, two different protein groups were generated, those comprised of proteins that were more abundant upon salt treatment and those that were less abundant upon salt 197

treatment. Proteins were further selected by focusing on those whose description contained keywords such as: ion transport, sodium, chloride, antiporter, vacuole and also focusing in uncharacterized proteins.

A third group of proteins was generated based on the assumption that S. bigelovii might be expecting NaCl; thus, having unaltered expression of key transporters. Proteins conventionally associated with salt tolerance such as SOS1 and NHXs were selected for characterization regardless their expression patterns.

Similarly, proteins such as CHXs, chloride conducting ion channels (CLCs), and cation (Ca2+) / proton exchangers (CAX) were also selected for further characterization. Examples of the three different groups can be found in Figure 40.

Additionally, a few proteins without transmembrane helices but with markedly expression patterns were selected for characterization. A total of 63 genes from S. biglovii were selected for characterization. The list of all selected proteins can be found in Table 22.

198

Figure 40. Expression profiles of the three different groups of selected proteins for characterization. Examples of genes from each group are shown in: a) genes or proteins overexpressed upon salt treatment; b) downregulated upon salt treatment; c) genes commonly associated with salt tolerance regardless of their expression pattern. Gene expression measured as Transcripts per Million (TPM) normalized by Trimmed Mean of

M-values (TMM). Error bars are mean confidence intervals at α = 0.05, n = 3. 199

Table 22. Selected genes for characterization in yeast. Length Transcriptome Gene alias Blast Description pFam # TM Expression (bp) S. bigelovii Sodium/hydrogen Sodium/hydrogen NHX1 NHX1_ARATH 1608 11 Constant shoots exchanger 1 exchanger family S. bigelovii Sodium/hydrogen Sodium/hydrogen NHX2 NHX2_ARATH 1683 12 Constant shoots exchanger 2 exchanger family S. bigelovii Sodium/hydrogen Sodium/hydrogen NHX3 NHX2_ARATH 1563 12 Constant shoots + roots exchanger 2 exchanger family S. bigelovii Sodium/hydrogen Sodium/hydrogen NHX6 NHX6_ARATH 1593 12 Constant shoots + roots exchanger 6 exchanger family Vacuolar S. bigelovii Sodium/calcium CAX2 CAX2_ORYSJ cation/proton 1395 10 Constant shoots + roots exchanger protein exchanger 2 Vacuolar S. bigelovii Sodium/calcium CAX3 CAX3_ARATH cation/proton 1338 10 Constant shoots exchanger protein exchanger 3 S. bigelovii Cation/H+ antiporter Sodium/hydrogen CHX4 CHX4_ARATH 2406 10 Constant shoots 4 exchanger S. bigelovii Sodium/hydrogen CHX18 CHX18_ARATH Cation/H+ antiporter 2403 7 Constant shoots + roots exchanger 200

S. bigelovii Sodium/hydrogen CHX20 CHX20_ARATH Cation/H+ antiporter 2571 9 Constant shoots + roots exchanger S. bigelovii Sodium/calcium Sodium/calcium NCL1 D4QD79_DIACA 1785 10 Constant shoots exchanger protein exchanger protein S. bigelovii Chloride channel Voltage gated CLC_C1 CLCC_ARATH 2343 10 Constant shoots protein CLC-c chloride channel S. bigelovii Chloride channel Voltage gated CLC_C2 CLCC_ARATH 2346 10 Constant shoots protein CLC-c chloride channel S. bigelovii Chloride channel Voltage gated CLC_B CLCB_ARATH 2406 8 Constant shoots + roots protein CLC-b chloride channel S. bigelovii Chloride channel Voltage gated CLC_D CLCD_ARATH 2396 9 Constant shoots protein CLC-d chloride channel S. bigelovii Anoctamin-like Calcium-activated CA_CLC CACLC_ARATH 1980 8 Constant shoots protein chloride channel S. brachiata Cation-chloride Amino acid CCC1 CCC1_ARATH 2781 11 Constant shoots + roots cotransporter 1 permease S. bigelovii Major intrinsic Tip1_3 TIP13_ARATH Aquaporin TIP1-3 762 6 Constant shoots + roots protein S. brachiata Major intrinsic Pip2_1 PIP27_MAIZE Aquaporin PIP2-7 870 6 Constant genome (old) protein S. bigelovii Cyclic nucleotide- Down in Gated_N CNGC1_ARATH Ion transport protein 2127 5 shoots gated ion channel 1 salt 201

S. brachiata shoots + roots Cyclic nucleotide- Cyclic nucleotide- Down in (5') – Gated_N2 CNGC4_ARATH >2kb? ≥6 gated ion channel 4 binding domain salt S. bigelovii shoots (3') S. bigelovii Potassium K+ potassium Down in POT7 POT7_ARATH 2538 13 shoots transporter 7 transporter salt Sulfate transporter S. bigelovii Sulfate transporter Down in Sulf_1_3 SUT13_ARATH N-terminal domain 1944 10 shoots 1.3 salt with GLY motif S. bigelovii Cationic amino acid Amino acid Down in Catam_5 CAAT5_ARATH 1743 13 shoots transporter 5 permease salt S. bigelovii COPI associated Down in Pr_Unkn 636 4 shoots protein salt S. bigelovii Magnesium CorA-like Mg2+ Down in Mg2 MRS21_ARATH 1320 2 shoots transporter MRS2-1 transporter protein salt S. bigelovii Domain of unknown Down in DUF21_Pr Y4424_ARATH DUF21 1458 3 shoots function DUF21 salt Putative calcium- S. bigelovii Cation Down in Ca_ATPase ACA11_ARATH transporting ATPase 3075 8 shoots transporter/ATPase salt 11 202

S. bigelovii Major Facilitator Down in Mon_Sac2 MSSP2_ARATH MSSP2_ARATH 2211 11 shoots Superfamily salt S. bigelovii ABC transporter C Down in ABC4 AB4C_ARATH ABC transporter 4506 17 shoots family member 4 salt S. bigelovii ABC transporter C Down in ABC2 AB2C_ARATH ABC transporter 4908 14 shoots family member 2 salt S. bigelovii ABC transporter C Down in ABC9 AB9C_ARATH ABC transporter 4578 12 shoots family member 9 salt Major facilitator S. bigelovii Down in MHC1 A0A061ECX0_THECC superfamily protein Nodulin-like 1860 13 shoots + roots salt isoform 1 Transmembrane S. bigelovii Eukaryotic Down in Punch ACFR1_ARATH ascorbate 699 6 shoots cytochrome b561 salt ferrireductase 1 S. bigelovii Potassium channel Down in KAT2 KAT2_ARATH Ion channel 2160 4 shoots KAT2 salt Vacuolar S. bigelovii Sodium/calcium Down in CAX1 CAX3_ARATH cation/proton 1365 10 shoots + roots exchanger protein salt exchanger 1 S. bigelovii Sodium transporter Cation transport Down in HKT1 HKT1_ARATH 1614 9 shoots + roots HKT1 protein salt 203

S. bigelovii Sodium/hydrogen SOS1 NHX7_ARATH Salt overly sensitive 1 3480 12 Up in salt shoots exchanger family S. bigelovii Cationic amino acid Amino acid CatAm CAAT2_ARATH 1971 14 Up in salt shoots transporter 2 permease Transmembrane 9 S. bigelovii Endomembrane Endo_9 TMN9_ARATH superfamily member 1914 10 Up in salt shoots protein 9 Organic S. bigelovii Major Facilitator Carnit_3 OCT3_ARATH cation/carnitine 1572 10 Up in salt shoots Superfamily transporter 3 S. bigelovii Protein of unknown DUF 1323 7 Up in salt shoots function (DUF3537) S. bigelovii Unkn 1236 8 Up in salt shoots S. bigelovii Oligopeptide oligopeptide Ol_pep_trans OPT4_ARATH 2259 16 Up in salt shoots transporter 4 transporter protein Heptahelical S. bigelovii Haemolysin-III HPP1 HHP1_ARATH transmembrane 1080 6 Up in salt shoots related protein 1 S. bigelovii Sugar carrier protein Sugar (and other) Sug_trans STC_RICCO 1575 13 Up in salt shoots C transporter 204

Sodium:sulfate S. brachiata Dicarboxylate symporter Na_sulf DIT1_SPIOL 1683 13 Up in salt shoots + roots transporter 1 transmembrane region Sugar efflux S. bigelovii Bidirectional sugar transporter for Sweet SWET2_ARATH 705 7 Up in salt shoots transporter SWEET2 intercellular exchange Probable metal- S. bigelovii oligopeptide Ol_nuc_trans YSL7_ARATH nicotianamine 2082 14 Up in salt shoots transporter protein transporter YSL7 Transmembrane S. bigelovii Auxin transporter- Na_amin_aux LAX4_MEDTR amino acid 1470 10 Up in salt shoots like protein 4 transporter protein S. bigelovii MATE efflux family Mate MATE9_ARATH MatE 1692 12 Up in salt shoots + roots protein 9 S. bigelovii Probable boron Bor2 BOR2_ARATH HCO3- transporter 2163 10 Up in salt shoots transporter 2 S. bigelovii Mechanosensitive ion Mechanosensitive Mec_chan MSL8_ARATH 2748 6 Up in salt shoots + roots channel protein 8 ion channel 205

Probable vacuolar S. bigelovii PQ YPQ3_YEAST amino acid PQ-loop 1017 3 Up in salt shoots transporter Equilibrative S. bigelovii Nucleoside Eq_trans ENT1_ARATH nucleotide 1236 11 Up in salt shoots transporter transporter 1 Ammonium S. bigelovii Ammonium Amm_trans AMT11_SOLLC transporter 1 1524 9 Up in salt shoots Transporter member 1 S. bigelovii W_Unkn 993 1 Up in salt shoots Putative Transmembrane S. bigelovii Prot_am_trans F6H4D0_VITVI uncharacterized amino acid 1278 10 Up in salt shoots protein transporter protein S. bigelovii Protein of unknown Salt_Unkn 825 4 Up in salt shoots function (DUF4079) S. bigelovii An_Unkn 2583 0 Up in salt shoots S. bigelovii Salt tolerance protein Hyaluronan / Salt_tol2 Q711N3_BETVU 1074 0 Up in salt shoots + roots 2 mRNA binding 206

Sodium-coupled Transmembrane S. brachiata NacNeu3 S38A3_HUMAN neutral amino acid amino acid 1389 11 Up in salt shoots + roots transporter 3 transporter S. bigelovii Major intrinsic Tip2_1 TIP21_ARATH Aquaporin TIP2-1 741 7 Up in salt shoots protein S. bigelovii Major intrinsic Pip3 PIP1_ATRCA Aquaporin PIP-type 846 6 Up in salt shoots protein Sodium/hydrogen TAIR Arab_SOS1 NHX7_ARATH Salt overly sensitive 1 3438 12 n/a exchanger family 207

5.3.2 Cloning and characterization in yeast

The 63 selected genes were amplified by PCR to be cloned into pENTR™/D-

TOPO® or pCR™8/GW/TOPO® entry clones in E. coli. Primer sequences to amplify each gene can be found in Appendix Table 6. Two different kinds of entry clones were required. The non-directional cloning of pCR™8/GW/TOPO® was first attempted for all genes. However, some genes proved difficult to be cloned and the orientation of the inserts was always inverted, suggesting that these genes might be detrimental for E. coli growth. Those genes that were not possible to be cloned in the right orientation were then cloned using directional cloning into pENTR™/D-TOPO®. Of the 63 genes, nine could not be amplified by PCR and 18 did not give colonies with the correct insert, even with directional cloning (Table

23). Among the genes that did not give correct inserts was S. bigelovii SOS1. When using non-directional cloning, SOS1 inserts were always inverted. Few colonies were obtained while using directional cloning; however, these colonies had mutations leading to early stop codons. These mutations were not observed while sequencing SOS1 PCR products prior cloning, suggesting that S. bigelovii SOS1 is toxic to E. coli and either low copy erroneous amplicons are favored or E. coli generates mutations in the plasmid. To observe if this was a SOS related case,

Arabidopsis SOS1 was cloned into E. coli using directional cloning. Few colonies were generated, but no mutations were found in its sequence. It could be that S. bigelovii SOS1 is more active or toxic to E. coli than Arabidopsis SOS1. Alternatively, 208

the insert was sequenced in 95 % of its full sequence, it could be that a mutation is present in the unsequenced region. However, GFP fusions with Arabidopsis SOS1 resulted in fluorescence, suggesting that the insert is functional. Other genes were also not able to be cloned into E. coli (Table 23).

The cloned genes were then recombined into the galactose inducible vector pAG426GAL-ccdB-EGFP. Correct recombination events were verified by restriction enzyme digestions and sequencing. Vectors were then cloned into the yeast strain AXT3 (Δena1-4::HIS3, Δnha1::LEU2, Δnhx1::TRP1). AXT3 is a salt sensitive yeast mutant lacking Na+ extrusion and compartmentalizing related genes (Fig. 41) (Quintero et al., 2000). AXT3 has knockouts of the plasma membrane Na+ pumps ENA 1-4, the plasma membrane Na+|K+ / H+ antiporter

NHA1, and the endosomal Na+|K+ / H+ antiporter NHX1 (Quintero et al., 2000).

A total of 46 different genes were cloned into the yeast strain AXT3.

Additionally, two alleles were cloned for 5 different genes: Bor2, NacNeu3, NHX3,

PQ, and Salt_tol2. The different alleles were found while sequencing plasmids from different colonies. Interestingly, 1 of the 2 alleles appeared to be more conserved compared to gene sequences of S. brachiata and S. europaea, while the other allele appeared to diverge. This could be a consequence of a lower selective pressure on some copies of a gene, as S. bigelovii is a tetraploid species, while S. brachiata and S. europaea are thought to be diploid. The tetraploidy of S. bigelovii might allow for gene mutations and might help in the process of gene evolution

(Van de Peer et al., 2017). 209

Table 23. Amplification and cloning of selected genes for characterization in yeast. Amplified Gene alias Cloned pFam Length (bp) # TM Expression by PCR CAX3 yes yes Sodium/calcium exchanger protein 1338 10 Constant CCC1 yes yes Amino acid permease 2781 11 Constant CHX18 yes yes Sodium/hydrogen exchanger 2403 7 Constant CHX4 yes yes Sodium/hydrogen exchanger 2406 10 Constant CLC_C2 yes yes Voltage gated chloride channel 2346 10 Constant CLC_D yes yes Voltage gated chloride channel 2396 9 Constant NHX1 yes yes Sodium/hydrogen exchanger family 1608 11 Constant NHX2 yes yes Sodium/hydrogen exchanger family 1683 12 Constant NHX3 yes yes Sodium/hydrogen exchanger family 1563 12 Constant NHX6 yes yes Sodium/hydrogen exchanger family 1593 12 Constant Pip2_1 yes yes Major intrinsic protein 870 6 Constant Tip1_3 yes yes Major intrinsic protein 762 6 Constant Ca_ATPase yes yes Cation transporter/ATPase 3075 8 Down in salt Catam_5 yes yes Amino acid permease 1743 13 Down in salt DUF21_Pr yes yes Domain of unknown function DUF21 1458 3 Down in salt Gated_N yes yes Ion transport protein 2127 5 Down in salt Gated_N2 yes yes Cyclic nucleotide-binding domain >2kb ≥6 Down in salt KAT2 yes yes Ion channel 2160 4 Down in salt 210

Mg2 yes yes CorA-like Mg2+ transporter protein 1320 2 Down in salt Pr_Unkn yes yes COPI associated protein 636 4 Down in salt Punch yes yes Eukaryotic cytochrome b561 699 6 Down in salt Sulf_1_3 yes yes Sulfate transporter N-terminal domain with GLY motif 1944 10 Down in salt HKT1 yes yes Cation transport protein 1614 9 Down in salt Arab_SOS1 yes yes Sodium/hydrogen exchanger family 3438 12 n/a An_Unkn yes yes 2583 0 Up in salt Bor2 yes yes HCO3- transporter 2163 10 Up in salt Carnit_3 yes yes Major Facilitator Superfamily 1572 10 Up in salt CatAm yes yes Amino acid permease 1971 14 Up in salt DUF yes yes Protein of unknown function (DUF3537) 1323 7 Up in salt Eq_trans yes yes Nucleoside transporter 1236 11 Up in salt HPP1 yes yes Haemolysin-III related 1080 6 Up in salt Mate yes yes MatE 1692 12 Up in salt Na_amin_aux yes yes Transmembrane amino acid transporter protein 1470 10 Up in salt Na_sulf yes yes Sodium:sulfate symporter transmembrane region 1683 13 Up in salt NacNeu3 yes yes Transmembrane amino acid transporter 1389 11 Up in salt Ol_nuc_trans yes yes oligopeptide transporter protein 2082 14 Up in salt Ol_pep_trans yes yes oligopeptide transporter protein 2259 16 Up in salt Pip3 yes yes Major intrinsic protein 846 6 Up in salt PQ yes yes PQ-loop 1017 3 Up in salt Prot_am_trans yes yes Transmembrane amino acid transporter protein 1278 10 Up in salt 211

Salt_tol2 yes yes Hyaluronan / mRNA binding 1074 0 Up in salt Salt_Unkn yes yes Protein of unknown function (DUF4079) 825 4 Up in salt Sug_trans yes yes Sugar (and other) transporter 1575 13 Up in salt Sweet yes yes Sugar efflux transporter for intercellular exchange 705 7 Up in salt Tip2_1 yes yes Major intrinsic protein 741 7 Up in salt Unkn yes yes 1236 8 Up in salt CA_CLC yes no Calcium-activated chloride channel 1980 8 Constant CAX2 yes no Sodium/calcium exchanger protein 1395 10 Constant CHX20 no no Sodium/hydrogen exchanger 2571 9 Constant CLC_B yes no Voltage gated chloride channel 2406 8 Constant CLC_C1 yes no Voltage gated chloride channel 2343 10 Constant NCL1 no no Sodium/calcium exchanger protein 1785 10 Constant ABC2 no no ABC transporter 4908 14 Down in salt ABC4 yes no ABC transporter 4506 17 Down in salt ABC9 yes no ABC transporter 4578 12 Down in salt MHC1 no no Nodulin-like 1860 13 Down in salt Mon_Sac2 yes no Major Facilitator Superfamily 2211 11 Down in salt POT7 no no K+ potassium transporter 2538 13 Down in salt CAX1 yes no Sodium/calcium exchanger protein 1365 10 Down in salt Amm_trans no no Ammonium Transporter 1524 9 Up in salt Endo_9 no no Endomembrane protein 1914 10 Up in salt Mec_chan no no Mechanosensitive ion channel 2748 6 Up in salt 212

SOS1 yes no Sodium/hydrogen exchanger family 3480 12 Up in salt W_Unkn no no 993 1 Up in salt 213

Figure 41. Representation of AXT3 yeast strain. AXT3 has knockouts in the plasma membrane sodium pumps ENA 1-4, the plasma membrane Na+|K+ / H+ antiporter NHA1, and the endosomal Na+|K+ / H+ antiporter NHX1. Figure from: Cyert and Philpott (2013).

Yeast spot assays were carried out for all the different genes and tested in different treatments (Table 24). Two different media were tested, the minimal media synthetic defined (SD) at pH 5.6 and arginine phosphate (AP) at pH 6.5.

Media at different pH were tested as it has been reported that some genes provide tolerance in yeast at lower medium pH while others at higher (BañueIos et al.,

1998). An example of how yeast spot assays typically looked can be found in

Figure 42. Genes were first tested with 3’ GFP fusions to observe the subcellular 214

localization of the proteins using confocal microscopy. Genes showing a phenotype were then also tested without any tag to reduce likelihood that function was compromised by the tag.

Table 24. Media and treatments used for yeast spot assays for all gene constructs.

Treatment Carbon source Salt concentration SD Medium pH 5.6 Control (no gene expression) 2 % Glucose Control (gene expression no salt) 2 % Galactose Salt: NaCl 100 mM 2 % Galactose 100 mM NaCl Salt: NaCl 150 mM 2 % Galactose 150 mM NaCl Salt: NaCl 175 mM 2 % Galactose 175 mM NaCl Salt: NaCl 200 mM 2 % Galactose 200 mM NaCl Salt: KCl 800 mM 2 % Galactose 800 mM KCl AP medium pH 6.5 Control (no gene expression) 2 % Glucose Control (gene expression no salt) 2 % Galactose Salt: NaCl 20 mM 2 % Galactose 20 mM NaCl Salt: NaCl 30 mM 2 % Galactose 30 mM NaCl Salt: NaCl 40 mM 2 % Galactose 40 mM NaCl Salt: NaCl 60 mM 2 % Galactose 60 mM NaCl

Growth under glucose, which inhibits gene expression of the transformed gene, resulted in an even growth across all transformants indicating that the construct itself does not influence yeast growth. Growth under galactose typically resulted in even growth; however, some genes displayed a detrimental phenotype when expressed in yeast. In particular, expression of HKT1 severely affected growth, with yeast seeming to die even without the addition of NaCl. Transformants 215

responded differently to different concentrations of NaCl. Out of 46 different genes that were cloned in yeast, 11 appeared to either increase or decrease salt tolerance in the yeast strain AXT3 (Table 25). Four genes increased salt tolerance and seven genes decreased salt tolerance.

Figure 42. Example of yeast spot assays. a) Growth under glucose (no gene expression); b) growth under galactose (gene expression); c) growth under galactose and under mild

NaCl concentration; d) growth under galactose and under high NaCl concentration. 216

Table 25. Genes affecting AXT3 growth under NaCl. Gene alias Effect in salt tolerance Description KAT2 Increase (+ + +) Potassium channel KAT2 Hyaluronan / mRNA Salt_tol2 Increase (+ + +) binding Sodium/hydrogen NHX3 Increase (+ +) exchanger Ol_pep_trans Increase (+) Oligopeptide transporter An_Unkn Decrease (- - -) Unknown HKT1 Decrease (- - -) Sodium transporter HKT1 Sodium/hydrogen NHX1 Decrease (- - -) exchanger Sodium/hydrogen NHX2 Decrease (- - -) exchanger Equilibrative nucleotide Eq_trans Decrease (- -) transporter Sodium-coupled neutral NacNeu3 Decrease (- -) amino acid transporter PQ Decrease (- -) PQ-loop, Probable vacuolar amino acid transporter

217

Of the genes that increased salt tolerance in AXT3 yeast, Salt_tol2 and KAT2 showed the highest increase in tolerance, allowing yeast to grow in conditions which would otherwise kill it (Figs. 43, 44, and 45). KAT2 has the highest identity with a potassium channel, this is supported by the observation that expression of

KAT2 led to hypersensitivity of AXT3 when grown in high concentrations of KCl

(Fig 43). Yeast expressing NHX3 and Ol_pep_trans showed increased salt tolerance, particularly at less acidic pH (Fig. 44). NHX3 without a C terminal EGFP fusion showed a greater increase in salt tolerance than its tagged counterpart (Figs. 43 and 45). In the case of KAT2, no correct entry clones were found when cloned with a stop codon. However, its increase in NaCl tolerance and decreased KCl tolerance suggest the tagged protein is fully functional.

Several genes decreased salt tolerance in AXT3 yeast (Figs. 46, 47, and 48).

HKT1 was particularly detrimental for yeast growth, even under conditions without salt. As with HKT1, An_Unkn decreased yeast growth, particularly under saline conditions at acidic pH (Figs. 46 and 48), under less acidic pH it reduced yeast growth even without the addition of NaCl (Fig. 47). Contrarily to what was observed for NHX3 (Figs. 43, 44, and 45), NHX1 and NHX2 decreased salt tolerance in AXT3 (Figs. 46, 47, and 48). PQ, Eq_trans, and NacNeu3 reduced salt tolerance to a lesser extent. NacNeu3 showed greater growth reduction at lower pH (Figs. 46 and 48) and almost no growth reduction at pH 6.5 (Fig. 47). Most of the genes decreasing NaCl tolerance also decreased KCl tolerance, suggesting that either they are involved in Na+, K+, or Cl- homeostasis or they affect growth under stress. 218

Figure 43. Yeast spot assays of genes increasing salt tolerance. Assays were done in SD medium pH 5.6 and supplemented with 2 % galactose for gene induction. Genes without a stop codon were cloned in pAG426GAL-ccdB-EGFP.

219

Figure 44. Yeast spot assays of genes increasing salt tolerance. Assays were done in AP medium pH 6.5 and supplemented with 2 % galactose for gene induction. Genes without a stop codon were cloned in pAG426GAL-ccdB-EGFP.

220

Figure 45. Yeast spot assays of genes increasing salt tolerance. Assays were done in SD medium pH 5.6 and supplemented with 2 % galactose for gene induction. Genes with stop codons were cloned in pAG426GAL-ccdB-EGFP.

221

Figure 46. Yeast spot assays of genes decreasing salt tolerance. Assays were done in SD medium pH 5.6 and supplemented with 2 % galactose for gene induction. Genes without stop codons were cloned in pAG426GAL-ccdB-EGFP. 222

Figure 47. Yeast spot assays of genes decreasing salt tolerance. Assays were done in AP medium pH 6.5 and supplemented with 2 % galactose for gene induction. Genes without stop codons were cloned in pAG426GAL-ccdB-EGFP.

223

Figure 48. Yeast spot assays of genes decreasing salt tolerance. Assays were done in SD medium pH 5.6 and supplemented with 2 % galactose for gene induction. Genes with stop codons were cloned in pAG426GAL-ccdB-EGFP.

224

5.3.3 Gene analysis

Yeast spot assays in the salt sensitive AXT3 yeast strain revealed 11 genes that affect yeast fitness upon growth in NaCl or KCl (Table 25). A small description of these genes can be found below.

NHX genes

Different phenotypes were observed across the different genes with highest identity to NHX proteins. NHX1 and NHX2 reduced salt tolerance while NHX3 increased salt tolerance in AXT3 yeast. Expression of these genes was unaltered across treatments. S. bigelovii NHX1 protein had 50 % identity with Arabidopsis thaliana NHX1 and 87 % identity with Beta vulgaris NHX1, as the best hit. S. bigelovii

NHX2 had 72 % identity with A. thaliana NHX2 and 91 % identity with B. vulgaris

NHX2. In the case of S. bigelovii NHX3, it had a 55 % identity with A. thaliana NHX2 and 70 % identity with Spinacia oleracea NHX2. The three proteins contained the sodium / hydrogen exchanger domain and are considered to be vacuolar proteins.

Their secondary structures can be found in Figures 49, 50, and 51. The number of transmembrane helices vary depending on the predictor. In the case of tmHMM,

11 transmembrane helices are predicted for NHX1 (Tables 22 and 23), while Protter predicted 13 (Fig. 49), A. thaliana NHX1 has been described to have 12 (Yamaguchi et al., 2003, Sato & Sakaguchi, 2005). NHX1 and NHX2 are predicted to have the N terminus at the inside of the organelle while having the C terminus in the cytosol.

NHX3 have both N and C termini at the inside of the organelle (Fig. 49 a). 225

Figure 49. Secondary structure and subcellular localization of NHX proteins. Predicted secondary structure with Protter of: a) NHX1; c) NHX2; and e) NHX3. Subcellular localization observed by EGFP fusions in AXT3 of: b) NHX1; d) NHX2; and f) NHX3. 226

The C terminus of NHX proteins is considered to play a role in the regulation of these transporters (Yamaguchi et al., 2003) and in the case of Arabidopsis NHX1, a calmodulin has been observed to bind to its C terminus, affecting its transporter activity (Yamaguchi et al., 2005). NHX1 and NHX2 are predicted to have their N and C termini at different sides of the membrane, while NHX3 has both termini on the same side, which suggests differences in regulation across these proteins.

NHX1 and NHX2 appear to be localized to endosomes, while NHX3 subcellular localization is different to establish, as it seemed to localize to the ER, vacuole, and cytosol in AXT3 yeast (Fig. 49). The different subcellular localization of these proteins could explain the phenotypic differences in respect to salt tolerance, as different subcellular compartments would respond differently to ion accumulation.

HKT1

HKT proteins belong to the superfamily of K+ transporters, but class 1 HKT proteins are considered to primarily transport Na+ (Mäser et al., 2002, Ali et al.,

2016). Expression of HKT1 in S. bigelovii decreased upon salt treatment. Expression of HKT1 in yeast resulted in toxicity, particularly under stress conditions. The S. bigelovii HKT1 protein has a 41 % identity with A. thaliana HKT1 and 78 % identity with B. vulgaris HKT1. Interestingly, no expression of HKT1;1 was detected in

Salicornia dolichostachya and was suggested that S. dolichostachya might not require

HKT (Katschnig et al., 2015). In this study, HKT1 expression was most abundant 227

in treatments without NaCl; nevertheless, its expression was abundant throughout treatments (Fig. 40). The inability to detect HKT in S. dolichostachya might be due to primer design, as they were based on sequences from other Amaranthaceae, and S. bigelovii HKT1 shared 78 % identity with that of B. vulgaris. S. bigelovii HKT1 was predicted to have 9 transmembrane helices (Fig. 50 a).

Figure 50. Secondary structure and subcellular localization of HKT1. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3.

Subcellular localization of HKT1 in yeast is difficult to define, as HKT1 is toxic to yeast, but confocal microscope images suggest that it predominantly 228

localizes to the plasma membrane (Fig. 50 b). The toxicity of S. bigelovii HKT1 in yeast might be due to an excess of ion transport. Expression of Arabidopsis

HKT1;1 in yeast resulted in cell death upon salt NaCl treatment (Jin et al.,

2015). Both, knockout and overexpression of Arabidopsis HKT1, led to decreased salt tolerance (Ali et al., 2016). However, expression of HKT1 under its native promoter recovered its tolerance (Ali et al., 2016). This indicates that the expression of a protein under the right conditions and cell types is crucial

(Moller et al., 2009).

KAT2

KAT2 is a plasma membrane highly selective inward-rectifying Shaker-like K+ potassium channel (Lebaudy et al., 2010). KAT proteins have been commonly associated with guard cells. They are considered to form heterodimers with

KAT1 and play a key role in stomatal opening (Pilot et al., 2001, Lebaudy et al.,

2010). S. bigelovii KAT2 has a 52 % identity with A. thaliana KAT2 and 68 % identity with S. oleracea KAT1-like protein. KAT2 expression in S. bigelovii increased upon salt treatment. Expression of KAT2 in AXT3 greatly increased

NaCl tolerance but inhibited growth upon KCl treatment (Figs. 43 and 44).

Increased K+ uptake could reduce the effects of Na+ toxicity by increasing K+

/ Na+ ratio. Congruently, expression of Arabidopsis KAT1 in yeast have led to increased salt tolerance (Jin et al., 2015). S. bigelovii KAT2 was predicted to have six transmembrane helices with both N and C termini at the same side of the 229

membrane (Fig 51 a). The subcellular localization of KAT2 appears to primarily be at specific regions of the plasma membrane and ER (Fig 51 b).

Figure 51. Secondary structure and subcellular localization of KAT2. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3.

Ol_pep_trans

Ol_pep_trans has high similarity with oligopeptide transporter proteins

(OPT). Ol_pep_trans has a 73 % identity with Arabidopsis oligopeptide transporter 4 and 89 % identity with B. vulgaris oligopeptide transporter 4.

Oligopeptide transporters are associated with transport of peptides into the cytosol (Lubkowitz, 2011). Arabidopsis OPT4 has been described as a proton / 230

olipeptide transporter (Osawa et al., 2005). OPTs have 12 – 14 transmembrane helices (Koh et al., 2002). S. bigelovii Ol_pep_trans was predicted to have 16 transmembrane helices, with both N and C termini at the same side of the membrane (Fig. 52 a). Its subcellular localization appeared to be predominantly in plasma membrane, ER, and tonoplast (Fig. 52 b).

Figure 52. Secondary structure and subcellular localization of Ol_pep_trans. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3.

S. bigelovii PQ protein has a 52 % identity with Arabidopsis PQ-loop repeat family protein and 77 % with Chenopodium quinoa uncharacterized protein. PQ- loop proteins are not well characterized in plants, but are characterized for having a PQ amino acid doublet within two transmembrane domains (Saudek, 231

2012). PQ-loop proteins in humans have been shown to function as cationic amino acid symporters such as cystine / H+ symporters in lysosomes (Kalatzis et al., 2001, Jézégou et al., 2012). S. bigelovii PQ was upregulated upon salt treatment. It is predicted to have four transmembrane helices, with both N and

C termini at the same side of the membrane (Fig. 53 a). It appeared to localize in AXT3, predominantly to the plasma membrane and ER (Fig. 53 b).

Figure 53. Secondary structure and subcellular localization of PQ. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in AXT3.

Eq_trans

S. bigelovii Eq_trans has a 61 % identity with Arabidopsis equilibrative nucleotide transporter 1 (ENT1) and 86 % identity with B. vulgaris ENT1. 232

Proteins of this family mediate the transport of purine and pyrimidine nucleosides (Wormit et al., 2004). Arabidopsis ENT1 is localized to the tonoplast (Bernard et al., 2011). Overexpression of Arabidopsis ENT1 led to reduced growth, presumably by depleting adenoside storage in the vacuole

(Bernard et al., 2011). S. bigelovii Eq_trans was upregulated upon salt treatment and was predicted to be localized to the tonoplast (Appendix Table 7). It was predicted to have 11 transmembrane helices with N and C termini at opposite sides of the membrane (Fig. 54 a). Eq_trans appeared to predominantly localize to the plasma membrane, ER, and a part of the tonoplast (Fig. 54 b).

Figure 54. Secondary structure and subcellular localization of Eq_trans. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3.

233

An_Unkn

An_Unkn has no significant hit with any Arabidopsis protein, but has 62 % identity with C. quinoa probable serine threonine kinase and 51 % identity with

S. oleracea FK506-binding protein 5. Differently to the aforementioned C. quinoa and S. oleracea proteins, An_Unkn has a truncated K+ dependent Na+ / Ca2+ exchanger domain. It does not have any predicted transmembrane helices (Fig.

55 a) and was downregulated upon salt treatment. Its expression in AXT3 yeast led to cell death; however, differently to HKT1, this effect was dependent on

NaCl and KCl when yeast was grown in pH 5.6. At the less acidic pH 6.5,

An_Unkn caused cell toxicity even without NaCl or KCl in the media (Figs. 46,

47, and 48). Its subcellular localization appeared to be cytosolic; however, it was also abundant at specific places in the plasma membrane and surrounding the ER and possibly the tonoplast (Fig. 55 b). This raises the possibility that this protein could bind to membrane proteins or membranes.

Figure 55. Secondary structure and subcellular localization of An_Unkn. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3. 234

NacNeu3

NacNeu3 has 66 % identity with Arabidopsis putative amino acid transporter and 90 % identity with C. quinoa amino acid transporter AVT6A-like. Search in

Conserved Domain Database (Marchler-Bauer et al., 2017) revealed that

NacNeu3 has a SLC5-6-like_sbd superfamily domain. SLC5 proteins cotransport Na+ and glucose (Wright, 2013) whereas SLC6 proteins utilize Na+ or Cl- to cotransport substrates (Bala et al., 2013). NacNeu3 expression was decreased upon salt treatment and was predicted to localize to the tonoplast

(Appendix Table 7). Its expression in yeast resulted in decreased salt tolerance.

NacNeu3 is predicted to have 11 transmembrane helices with both termini at different sides of the membrane (Fig. 56 a). It appears to be localized to the tonoplast and ER in yeast (Fig. 56 b).

Figure 56. Secondary structure and subcellular localization of NacNeu3. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3. 235

Salt_tol2

Salt_tol2 together with KAT2, were the proteins which showed a greater effect on yeast fitness upon salt treatment than the other genes (Figs. 43, 44, and 45).

Two different alleles of Salt_tol2 were cloned and tested in yeast. Both alleles provided equal levels of salt tolerance to AXT3 yeast (data not shown).

Salt_tol2 has 36 % identity with Arabidopsis Hyaluronan / mRNA binding family protein (AtRGGA) and 84 % identity with B. vulgaris plasminogen activator inhibitor 1 RNA-binding protein, also known as salt tolerance protein

2. Overexpression of Arabidopsis AtRGGA in Arabidopsis led to increased salt and drought tolerance, whereas knockout of AtRGGA led to salt sensitivity

(Ambrosone et al., 2015). It was shown that AtRGGA binds to RNA, primarily to non-poly A RNA, suggesting that AtRGGA might bind to ribosomal RNA

(Ambrosone et al., 2015). Moreover, its ortholog in potato was observed to be upregulated upon drought stress (Ambrosone et al., 2011). S. bigelovii Salt_tol2 expression and protein were more abundant upon salt treatment. It has no transmembrane helices (Fig. 57 a) and is localized to the cytosol in yeast (Fig.

57 b). 236

Figure 57. Secondary structure and subcellular localization of NacNeu3. a) Predicted secondary structure with Protter; b) Subcellular localization observed by EGFP fusions in

AXT3.

5.4 Conclusion

Genes derived from the transcriptomic and proteomic analyses of S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl, discussed in Chapter 4, were selected for characterization in yeast. 65 different genes were selected, out of which 46 were cloned into yeast. The 46 different genes were tested by yeast spot assays under different NaCl and KCl treatments in the salt sensitive AXT3 yeast strain, allowing for the identification of 11 proteins affecting yeast growth. Of the 11 proteins, 4 improved growth upon NaCl treatment (Figs. 43,

44, and 45) and 7 led to reduced growth (Figs. 46, 47, and 48). NHX proteins displayed contrasting phenotypes. NHX1 and NHX2 expression were toxic to yeast, particularly under salt treatments. NHX3 increased yeast NaCl 237

tolerance. Interestingly, NHX1 and NHX2 appeared to localize to endosomes

(Fig. 49), whilst NHX3 seemed to localize to the ER, cytosol, and vacuole (Fig.

49). Additionally, NHX1 and NHX2 are predicted to have their N and C termini on different sides of the membrane, whereas NHX3 is predicted to have both on the same side. NHX proteins have been observed to be regulated at their C terminus (Yamaguchi et al., 2005), this difference in C terminal localization in

NHX3 might lead to different regulatory mechanisms than those in NHX1 and

NHX2. The potassium channel KAT2 greatly increased NaCl tolerance in yeast, but similarly reduced KCl tolerance, presumably by allowing the accumulation of excessive K+ in the cell. HKT1, which was previously reported to not be present in

Salicornia dolichostachya (Katschnig et al., 2015), was found to be abundant in S. bigelovii. Its expression in yeast led to cell toxicity, even in treatments without the addition of any salt. Similar was the case of An_Unkn; however, its detrimental effect on yeast growth appeared to be dependent on NaCl and KCl or upon growth at less acidic pH. An_Unkn seemed to be toxic for AXT3 yeast upon stress. A PQ- loop (PQ), a likely equilibrative nucleoside transporter (Eq_trans), and a possible sodium / amino acid transporter (NacNeu3) reduced yeast growth upon NaCl and

KCl treatment. In the case of NacNeu3, growth at high pH (6.5) reduced its detrimental effect. The proton / oligopeptide transporter (Ol_pep_trans) moderately increased salt tolerance. Salt_tol2, a possible RNA or ribosome binding protein, greatly increased salt tolerance in yeast. However, its sequence has low 238

identity with any characterized protein and further studies are needed to understand its function.

It is important to keep in consideration that fusion at the C’ terminus of proteins might affect their subcellular localization (Froissard et al., 2006, Cagnac et al., 2007, Cagnac et al., 2010) and N’ terminal fusions would be required to confirm their subcellular localization. Moreover, their subcellular localization should be confirmed in planta. Additionally, C’ terminal fusions have been suggested to lead to protein retention in the ER (Froissard et al., 2006). This could explain why most proteins in this study appeared to partially localize to the ER.

Proteins leading to cell death upon salt treatment might still increase salt tolerance in plants, as cell specific expression is key for salt tolerance. E.g. the accumulation of Na+ in stems or non-chlorenchymatic cells might increase plant salt tolerance by preventing Na+ from reaching photosynthetically active cells. Proteins that did not show any effect on yeast growth upon NaCl treatment, could still be important for NaCl tolerance, but might require posttranslational modifications. This study identified several proteins which affect salt tolerance in yeast. This serves as the basis for further studies, which should assess the effect of these genes in planta and possibly to investigate the mechanisms by which some might confer salt tolerance.

239

Chapter 6. Conclusion and future perspectives

6.1 General conclusion and main findings

This study had two main aims: 1) The generation of genomic resources that will enable future research in the Salicornia genus; and 2) The identification of possible mechanisms and proteins that might be involved in the salt tolerance of S. bigelovii, in particular, to the remarkable ability of the plant to tolerate accumulation of extremely high concentrations of salt in the shoot. In order to generate resources that will allow scientists to carry out further research in Salicornia, the genomes of

S. brachiata and S. europaea were assembled and annotated, becoming the first genomes of the Salicornioideae subfamily and of any euhalophytes to have been sequenced. The genomes were relatively small in size, having 468 Mb and 530 Mb of assembled sequence for S. brachiata and S. europaea, respectively (Chapter 2). The annotations of the genomes showed the composition to be in par with other reported genomes of the Amaranthaceae family: Chenopodium quinoa, Spinacia oleracea, and Beta vulgaris. Some of the genomes of the Amaranthaceae family have been assembled to chromosome level, suggesting a high-quality annotation of the

Salicornia genomes, despite their fragmented status.

S. bigelovii plants displayed improved growth upon addition of NaCl to the medium. Plants exposed to higher concentrations of NaCl exhibited enhanced growth and greater shoot diameter (Chapter 3). Plants accumulated in their shoots 240

similar concentrations of Na+ and Cl- at levels above those found in seawater. K+ concentration decreased with NaCl treatment, and shoots, but not roots, accumulated higher concentrations of Na+ than of K+, even under treatments without the addition of any NaCl. Water content also increased in plants treated with NaCl, suggesting that S. bigelovii plants might require Na+ and Cl- to be used as osmolytes for water uptake in the shoots. Additionally, the increase in cell size

(or, more particularly, vacuole size) might also be used to dilute to some degree accumulated Na+ and Cl- in the cell. Gene differential expression analysis was performed in shoots of S. bigelovii plants treated with 0, 50, 200, and 600 mM NaCl to identify transcriptional responses to NaCl (Chapter 4). As a general response,

ATP synthesis, and cell wall biosynthesis and modification were upregulated upon salt treatment. A family of widely uncharacterized receptors named Domain of unknown function 26 (DUF26, also called Salt stress response / antifungal) was downregulated upon salt treatment, particularly when plants were exposed to 600 mM NaCl for 1 week.

Because S. bigelovii accumulates high concentrations of Na+ in its shoots, tight control of intracellular Na+ transport must exist. To have a better picture of which transporter proteins might be involved in salt tolerance in S. bigelovii, a membrane enriched organellar proteome was generated in shoots of S. bigelovii plants treated with 0, 50, 200, or 600 mM NaCl for 6 weeks (Chapter 4). The proteome was generated through a continuous sucrose density gradient, separating organellar membranes based on their isopycnic points. The 241

transcriptomes of S. bigelovii, S. brachiata, and S. europaea were generated and used as reference for the proteomics studies. Protein differential abundance analyses revealed similar responses to those found in gene differential expression analysis.

Plants treated with higher concentrations of NaCl showed overrepresentation and increased abundance of proteins involved in ATP synthesis, cell wall biosynthesis, and photosynthesis. Reduced abundance of DUF26 receptors in these plants was also found. Membrane proteins were predicted for their subcellular localization and revealed a good separation and identification of tonoplast proteins.

Unexpectedly, the tonoplast had a higher relative density when compared to other plant species and the plasma membrane appeared to be less dense. This was corroborated by Western blot analyses of the different density fractions from the continuous sucrose density gradient. It might be that the tonoplast in S. bigelovii is particularly protein rich, leading to an increased density.

Based on the transcriptomics and proteomics analyses, 63 genes were selected for further characterization in yeast (Chapter 5). A total of 11 proteins were identified to affect salt tolerance in the yeast strain AXT3, four increased salt tolerance and seven decreased it. NHX1 and NHX2 homologs in S. bigelovii decreased salt tolerance while NHX3 increased it. Confocal microscopy images revealed differences in their subcellular localization. NHX1 and NHX2 appeared to be localized to endosomes, whilst NHX3 seemed to localize to the ER and tonoplast. This difference in subcellular localization could explain the contrasting phenotypes. HKT1 was suggested not to be expressed in Salicornia dolichostachya 242

(Katschnig et al., 2015); nevertheless, its expression was high in S. bigelovii. HKT1 expression in yeast resulted in cell toxicity, similar results were observed when

Arabidopsis HKT1 was expressed in yeast (Jin et al., 2015), suggesting that S. bigelovii HKT1 might have a similar activity. Two proteins greatly increased salt tolerance in yeast: the potassium transporter, KAT2; and an uncharacterized protein with some similarity to AtRGGA, which is thought to bind to RNA and whose expression increases salt tolerance in Arabidopsis (Ambrosone et al., 2015).

6.2 Limitations of this study

The results of this study serve as a basis for further Salicornia research. However, there are limitations to this work. The genome of S. bigelovii was not able to be assembled. This could be in part due to the inherent difficulty in assembling tetraploid genomes. However, the major factors affecting the assembly were the failure to generate mate pair libraries and the low coverage in SMRT sequencing.

This could be addressed by increasing the sequencing depth of coverage of SMRT sequencing or, alternatively, generating new mate pair libraries. Probably the most cost efficient solution would be the generation of mate pair libraries and proceed with a hybrid assembly using DBG2OLC or a similar assembler, similar to how the apple genome was assembled (Daccord et al., 2017).

The organellar membrane proteomics study was fairly successful; nevertheless, the identification of transmembrane proteins is always limited due 243

to the nature of the technique. Since the identification of proteins rely on the digestion of proteins into smaller peptides, transmembrane proteins have the limitation that transmembrane regions are not exposed for digestion, reducing the availability of cleavage sites to be recognized by the protease to only regions external to the membrane. Despite being considered that membrane proteins account for ~30 % of total proteins (Wallin & Heijne, 1998, Tan et al., 2008,

Wiśniewski et al., 2009), membrane proteomic studies usually identify about 500 to 1500 proteins per fraction (Dunkley et al., 2006, Komatsu, 2008, Ferro et al., 2010,

Trotter et al., 2010, Nikolovski et al., 2012, Groen et al., 2014, Nikolovski et al., 2014,

Smolders et al., 2015). In this study, 2,521 proteins were identified in total, with an average of ~1,200 proteins per fraction, depending on the density fraction. For this reason, it is advisable to couple proteomic and transcriptomic studies.

The number of candidate genes that can be tested in plants must be small.

A prior characterization, such as the yeast spot assays presented in this work, allow for a quick characterization of several genes, enabling the selection of few candidates for further characterization. However, despite yeast having a relatively similar organellar structure to plants, it is still a different organism. Differences in cell sorting signals sometimes affect subcellular localization (Chao & Etzler, 1994).

Similarly, not all protein regulatory elements are expected to be conserved between yeast and plants, which might compromise protein activity. Additionally, in the case of salt tolerance, yeast is highly salt tolerant, which calls for the need to use salt sensitive mutants. In some cases, these mutants have been shown to affect 244

protein sorting (Brett et al., 2005b). For this reason, a lack of increased salt tolerance in yeast, does not necessarily mean that the protein would not help with the stress in planta. Additionally, subcellular localization was only studied with C terminal

EGFP fusions in yeast. In some cases, the fusion of tags to the C terminus have been shown to affect protein sorting (Cagnac et al., 2007, Cagnac et al., 2010) and it is advisable to compare the localization with N terminal fusions. Nonetheless, yeast is one of the best systems to screen high number of genes currently available.

Despite the aforementioned limitations of the study, this work provides a useful basis for future research. One of the reasons why research of euhalophytes has been scarce is the lack of available genomic resources. This study generated resources that will enable scientists to further develop research in Salicornia. The generation of the genomes of S. brachiata and S. europaea will also enable comparative studies between euhalophytes and glycophytes, which might shed light into salt tolerance mechanisms. The genomes, together with the transcriptomes and organellar membrane proteomes of S. bigelovii, provide tools that will help scientists identify genes of interest and proceed with their characterization. Moreover, this study revealed several candidate proteins that might improve salt tolerance in plants.

245

6.3 Perspectives and proposed future work

Research in Salicornia has been scarce, as genomic resources were limited; hence, the research in Salicornia is still in its infancy and there is still a lot more to be studied. With the data generated in this study, research in Salicornia can be carried out properly. As future work, it would be important to compare the genomes of S. brachiata and S. euroapea against the genomes of other members of the

Amaranthaceae, such as S. oleracea. This might result in the identification of increased numbers of proteins of certain families in Salicornia, which would be interesting to analyze if they play a role in the high salt tolerance of Salicornia. It has been reported that Na+ is predominantly found in non-chlorenchymatic cortical cells in Salicornia (Reddy et al., 1993, Lv et al., 2012b). A cell-type specific gene differential expression analysis in shoots of S. bigelovii under different NaCl treatments separating the stele, cortical non-chlorenchymatic cells, and photosynthetically active chlorenchymatic cells at the periphery of the shoots, would provide valuable information on salt tolerance mechanisms in S. bigelovii.

The association of proteins correlating with Na+ distribution in shoots could allow the identification of candidate proteins involved in salt tolerance. Additionally, a phosphoproteomic study of membrane proteins in Salicornia would be a good complement to the proteomics experiment presented in this study, looking for post-translational modifications which might regulate protein activity. Further characterization of these proteins would be required to validate their function. 246

Similarly, the characterization in planta of the proteins identified in this study is needed. Subcellular localization of these proteins could be identified in onion cells, tobacco, or in stable transformation in Arabidopsis and /or rice. It is advisable to generate both N and C termini fluorescent tag fusions to confirm their subcellular localization. Arabidopsis stable lines could be generated expressing these proteins under their native or constitutive promoters followed by NaCl tolerance assays.

Proteins increasing salt tolerance in Arabidopsis should be characterized in detail and could be later transformed to crops such as wheat and rice and salt tolerance measured in the transformed lines. It would also be important to generate a transformation protocol in Salicornia. Gene knockdowns in Salicornia plants could lead to the identification of key proteins involved in salt tolerance. Until now, calli have been generated from seeds and shoots of S. bigelovii; however, callus regeneration was not achieved (data not shown). The development of a proper regeneration protocol for Salicornia is required. Another interesting study would be the generation of a mutant population of Salicornia through induced mutations with ethyl methanesulfonate (EMS) or radiation. This could help in the identification of important genes related to salt tolerance. In a different type of research, Salicornia could be tested as a possible agent for the bioremediation of salinized soils, which could recover lost lands due to overirrigation. Additionally,

Salicornia can be studied for reasons other than salinity. Some Salicornia species grow at high temperatures, which could allow the study of heat stress tolerance.

Salicornia has been recently introduced into the European market for food, which 247

might increase the attention put into it. The resources generated in this study will allow further research in this very interesting plant.

248

REFERENCES

1. Adams P, Nelson DE, Yamada S, et al., 1998. Growth and development of

Mesembryanthemum crystallinum (Aizoaceae). New Phytologist 138, 171-

90.

2. Adams P, Thomas JC, Vernon DM, Bohnert HJ, Jensen RG, 1992. Distinct

cellular and organismic responses to salt stress. Plant Cell Physiology, 1215-

23.

3. Afzal Z, Howton CT, Sun Y, Mukhtar SM, 2016. The Roles of Aquaporins

in Plant Stress Responses. Journal of Developmental Biology 4.

4. Al-Tamimi N, Brien C, Oakey H, et al., 2016. Salinity tolerance loci revealed

in rice using high-throughput non-invasive phenotyping. Nature

Communications 7, 13342.

5. Ali A, Raddatz N, Aman R, et al., 2016. A Single Amino-Acid Substitution

in the Sodium Transporter HKT1 Associated with Plant Salt Tolerance.

Plant Physiology 171, 2112-26.

6. Allan AK, Du J, Davies SA, Dow JaT, 2005. Genome-wide survey of V-

ATPase genes in Drosophila reveals a conserved renal phenotype for lethal

alleles. Physiological Genomics 22, 128-38.

7. Ambrosone A, Batelli G, Nurcato R, et al., 2015. The Arabidopsis RNA-

Binding Protein AtRGGA Regulates Tolerance to Salt and Drought Stress.

Plant Physiology 168, 292. 249

8. Ambrosone A, Costa A, Martinelli R, et al., 2011. Differential gene

regulation in potato cells and plants upon abrupt or gradual exposure to

water stress. Acta Physiologiae Plantarum 33, 1157-71.

9. Amtmann A, Sanders D, 1998. Mechanisms of Na+ Uptake by Plant Cells.

In: Callow JA, ed. Advances in Botanical Research. Academic Press, 75-112.

(Volume 29.)

10. An R, Chen Q-J, Chai M-F, et al., 2007. AtNHX8, a member of the

monovalent cation:proton antiporter-1 family in Arabidopsis thaliana,

encodes a putative Li+/H+ antiporter. The Plant Journal 49, 718-28.

11. Apse MP, Aharon GS, Snedden WA, Blumwald E, 1999. Salt tolerance

conferred by overexpression of a vacuolar Na+/H+ antiport in Arabidopsis.

Science 285, 1256-8.

12. Asada K, 2006. Production and scavenging of reactive oxygen species in

chloroplasts and their functions. Plant Physiology 141, 391-6.

13. Ashburner M, Ball CA, Blake JA, et al., 2000. Gene Ontology: tool for the

unification of biology. Nat Genet 25, 25-9.

14. Austin RB, 1982. Crop characteristics and the potential yield of wheat.

Journal of Agricultural Science 98, 447-53.

15. Ayala F, Oleary JW, 1995. Growth and physiology of Salicornia bigelovii Torr

at suboptimal salinity. International Journal of Plant Sciences 156, 197-205. 250

16. Ayala F, Oleary JW, Schumaker KS, 1996. Increased vacuolar and plasma

membrane H+-ATPase activities in Salicornia bigelovii Torr in response to

NaCl. Journal of Experimental Botany 47, 25-32.

17. Bai Y, Wang S, Zhong H, et al., 2015. Integrative analyses reveal

transcriptome-proteome correlation in biological pathways and secondary

metabolism clusters in A. flavus in response to temperature. 5, 14582.

18. Baisakh N, Ramanarao MV, Rajasekaran K, et al., 2012. Enhanced salt stress

tolerance of rice plants expressing a vacuolar H+-ATPase subunit c1

(SaVHAc1) gene from the halophyte grass Spartina alterniflora Löisel. Plant

Biotechnology Journal 10, 453-64.

19. Bala PA, Foster J, Carvelli L, Henry LK, 2013. SLC6 Transporters: Structure,

Function, Regulation, Disease Association and Therapeutics. Molecular

aspects of medicine 34, 197-219.

20. Ball MC, Anderson JM, 1986. Sensitivity of Photosystem II to NaCl in

relation to salinity tolerance - Comparative studies with thylakoids of the

salt tolerant mangrove, Avicennia marina, and the salt sensitive pea, Pisum

sativum. Australian Journal of Plant Physiology 13, 689-98.

21. Ballesteros E, Blumwald E, Donaire JP, Belver A, 1997. Na+/H+ antiport

activity in tonoplast vesicles isolated from sunflower roots induced by NaCl

stress. Physiologia Plantarum 99, 328-34. 251

22. Bankevich A, Nurk S, Antipov D, et al., 2012. SPAdes: A New Genome

Assembly Algorithm and Its Applications to Single-Cell Sequencing.

Journal of Computational Biology 19, 455-77.

23. Bañueios MA, Sychrová H, Bleykasten-Grosshans C, Souciet J-L, Potier S,

1998. The Nhal antiporter of Saccharomyces cerevisiae mediates sodium and

potassium efflux. Microbiology 144, 2749-58.

24. Barkla BJ, Vera-Estrella R, Camacho-Emiterio J, Pantoja O, 2002. Na+/H+

exchange in the halophyte Mesembryanthemum crystallinum is associated

with cellular sites of Na+ storage. Functional Plant Biology 29, 1017-24.

25. Barkla BJ, Zingarelli L, Blumwald E, Smith J, 1995. Tonoplast Na+/H+

antiport activity and its energization by the vacuolar H+-ATPase in the

halophytic plant Mesembryanthemum crystallinum L. Plant Physiology 109,

549-56.

26. Barragan V, Leidi EO, Andres Z, et al., 2012. Ion exchangers NHX1 and

NHX2 mediate active potassium uptake into vacuoles to regulate cell turgor

and stomatal function in Arabidopsis. Plant Cell 24, 1127-42.

27. Bartels D, Sunkar R, 2005. Drought and salt tolerance in plants. Critical

Reviews in Plant Sciences 24, 23-58.

28. Bassil E, Ohto M-A, Esumi T, et al., 2011. The Arabidopsis intracellular

Na+/H+ antiporters NHX5 and NHX6 are endosome associated and

necessary for plant growth and development. The Plant Cell Online 23, 224-

39. 252

29. Batelli G, Verslues PE, Agius F, et al., 2007. SOS2 promotes salt tolerance in

part by interacting with the vacuolar H+-ATPase and upregulating its

transport activity. Molecular and Cellular Biology 27, 7781-90.

30. Benito B, Haro R, Amtmann A, Cuin TA, Dreyer I, 2014. The twins K+ and

Na+ in plants. Journal of Plant Physiology 171, 723-31.

31. Benjamini Y, Hochberg Y, 1995. Controlling the False Discovery Rate - a

Practical and Powerful Approach to Multiple Testing. Journal of the Royal

Statistical Society Series B-Methodological 57, 289-300.

32. Bernard C, Traub M, Kunz HH, Hach S, Trentmann O, Möhlmann T, 2011.

Equilibrative nucleoside transporter 1 (ENT1) is critical for pollen

germination and vegetative growth in Arabidopsis. Journal of Experimental

Botany 62, 4627-37.

33. Binzel ML, Hess FD, Bressan RA, Hasegawa PM, 1988. Intracellular

compartmentation of ions in salt adapted tobacco cells. Plant Physiology 86,

607-14.

34. Birnboim HC, Doly J, 1979. A rapid alkaline extraction procedure for

screening recombinant plasmid DNA. Nucleic Acids Research 7, 1513-23.

35. Blaha G, Stelzl U, Spahn CMT, Agrawal RK, Frank J, Nierhaus KH, 2000.

Preparation of functional ribosomal complexes and effect of buffer

conditions on tRNA positions observed by cryoelectron microscopy. Rna-

Ligand Interactions Pt A 317, 292-309. 253

36. Blumwald E, Poole RJ, 1985. Na+/H+ antiport in isolated tonoplast vesicles

from storage tissue of Beta vulgaris. Plant Physiology 78, 163-7.

37. Blumwald E, Poole RJ, 1987. Salt tolerance in suspension cultures of sugar

beet: Induction of Na+/H+ antiport activity at the tonoplast by growth in

salt. Plant Physiology 83, 884-7.

38. Bolger AM, Lohse M, Usadel B, 2014. Trimmomatic: a flexible trimmer for

Illumina sequence data. Bioinformatics 30, 2114-20.

39. Bonales-Alatorre E, Shabala S, Chen Z-H, Pottosin I, 2013. Reduced

Tonoplast Fast-Activating and Slow-Activating Channel Activity Is

Essential for Conferring Salinity Tolerance in a Facultative Halophyte,

Quinoa. Plant Physiology 162, 940-52.

40. Bradford MM, 1976. A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein-dye

binding. Anal Biochem 72, 248-54.

41. Bray EA, J. B-S, E. W, 2000. Responses to abiotic stresses. In. Biochemistry

and Molecular Biology of Plants. Rockville, Md: American Society of Plant

Physiologists, 1158-203. (B. B. Buchanan, W. Gruissem, Jones RL, eds.)

42. Breckels LM, Holden SB, Wojnar D, et al., 2016. Learning from

Heterogeneous Data Sources: An Application in Spatial Proteomics. PLoS

Comput Biol 12, e1004920. 254

43. Brett CL, Donowitz M, Rao R, 2005a. Evolutionary origins of eukaryotic

sodium/proton exchangers. American Journal of Physiology - Cell Physiology

288, C223-C39.

44. Brett CL, Tukaye DN, Mukherjee S, Rao R, 2005b. The Yeast Endosomal

Na+(K+)/H+ Exchanger Nhx1 Regulates Cellular pH to Control Vesicle

Trafficking. Molecular Biology of the Cell 16, 1396-405.

45. Brini F, Hanin M, Mezghani I, Berkowitz GA, Masmoudi K, 2007.

Overexpression of wheat Na+/H+ antiporter TNHX1 and H+-

pyrophosphatase TVP1 improve salt and drought stress tolerance in

Arabidopsis thaliana plants. Journal of Experimental Botany 58, 301-8.

46. Brown JJ, Glenn EP, Fitzsimmons KM, Smith SE, 1999. Halophytes for the

treatment of saline aquaculture effluent. Aquaculture 175, 255-68.

47. Bush DS, 1995. Calcium regulation in plant cells and its role in signaling.

Annual Review of Plant Physiology and Plant Molecular Biology 46, 95-122.

48. Cagnac O, Aranda-Sicilia MN, Leterrier M, Rodriguez-Rosales M-P,

Venema K, 2010. Vacuolar Cation/H+ Antiporters of Saccharomyces

cerevisiae. The Journal of Biological Chemistry 285, 33914-22.

49. Cagnac O, Leterrier M, Yeager M, Blumwald E, 2007. Identification and

Characterization of Vnx1p, a Novel Type of Vacuolar Monovalent

Cation/H+ Antiporter of Saccharomyces cerevisiae. Journal of Biological

Chemistry 282, 24284-93. 255

50. Campbell MS, Holt C, Moore B, Yandell M, 2014. Genome Annotation and

Curation Using MAKER and MAKER-P. Current protocols in bioinformatics /

editoral board, Andreas D. Baxevanis ... [et al.] 48, 4.11.1-4..39.

51. Chang J, Ruiz V, Vancura A, 2008. Purification of Yeast Membranes and

Organelles by Sucrose Density Gradient Centrifugation. In: Vancura A, ed.

Membrane Trafficking. Totowa, NJ: Humana Press, 141-9.

52. Chanroj S, Lu Y, Padmanaban S, et al., 2011. Plant-Specific Cation/H+

Exchanger 17 and Its Homologs Are Endomembrane K+ Transporters with

Roles in Protein Sorting. Journal of Biological Chemistry 286, 33931-41.

53. Chanroj S, Wang G, Venema K, Zhang M, Delwiche C, Sze H, 2012.

Conserved and Diversified Gene Families of Monovalent Cation/H+

Antiporters from Algae to Flowering Plants. Frontiers in Plant Science 3.

54. Chao QM, Etzler ME, 1994. Incorrect Targeting of Plant Vacuolar Lectins in

Yeast. Journal of Biological Chemistry 269, 20866-71.

55. Chen L-H, Zhang B, Xu Z-Q, 2008. Salt tolerance conferred by

overexpression of Arabidopsis vacuolar Na+/H+ antiporter gene AtNHX1

in common buckwheat (Fagopyrum esculentum). Transgenic Research 17, 121-

32.

56. Chen Z, Cuin TA, Zhou M, Twomey A, Naidu BP, Shabala S, 2007.

Compatible solute accumulation and stress-mitigating effects in barley

genotypes contrasting in their salt tolerance. Journal of Experimental Botany

58, 4245-55. 256

57. Choi W-G, Miller G, Wallace I, Harper J, Mittler R, Gilroy S, 2017.

Orchestrating rapid long-distance signaling in plants with Ca2+, ROS and

electrical signals. The Plant Journal 90, 698-707.

58. Choi W-G, Toyota M, Kim S-H, Hilleary R, Gilroy S, 2014. Salt stress

induced Ca2+ waves are associated with rapid, long-distance root-to-shoot

signaling in plants. Proceedings of the National Academy of Sciences 111, 6497-

502.

59. Collins KD, 1997. Charge density-dependent strength of hydration and

biological structure. Biophysical Journal 72, 65-76.

60. Cotsaftis O, Plett D, Shirley N, Tester M, Hrmova M, 2012. A Two-Staged

Model of Na+ Exclusion in Rice Explained by 3D Modeling of HKT

Transporters and Alternative Splicing. PLOS ONE 7, e39865.

61. Cyert MS, Philpott CC, 2013. Regulation of Cation Balance in Saccharomyces

cerevisiae. Genetics 193, 677.

62. Dabbous A, Ben Saad R, Brini F, et al., 2017. Over-expression of a subunit

E1 of a vacuolar H+-ATPase gene (Lm VHA-E1) cloned from the halophyte

Lobularia maritima improves the tolerance of Arabidopsis thaliana to salt

and osmotic stresses. Environmental and Experimental Botany 137, 128-41.

63. Daccord N, Celton J-M, Linsmith G, et al., 2017. High-quality de novo

assembly of the apple genome and methylome dynamics of early fruit

development. Nat Genet 49, 1099-106. 257

64. Dalby DH, 1962. Chromosome number, morphology and breeding

behaviour in the British Salicorniae. Watsonia 5, 150-62.

65. Daniel Gietz R, Woods RA, 2002. Transformation of yeast by lithium

acetate/single-stranded carrier DNA/polyethylene glycol method. In:

Guthrie C, Fink GR, eds. Methods in Enzymology. Academic Press, 87-96.

(350.)

66. Dassanayake M, Oh D-H, Haas JS, et al., 2011. The genome of the

extremophile crucifer Thellungiella parvula. Nat Genet 43, 913-8.

67. Daszkowska-Golec A, 2016. The Role of Abscisic Acid in Drought Stress:

How ABA Helps Plants to Cope with Drought Stress. In: Hossain MA, Wani

SH, Bhattacharjee S, Burritt DJ, Tran L-SP, eds. Drought Stress Tolerance in

Plants, Vol 2: Molecular and Genetic Perspectives. Cham: Springer

International Publishing, 123-51.

68. Davenport RJ, Muñoz-Mayor A, Jha D, Essah PA, Rus ANA, Tester M, 2007.

The Na+ transporter AtHKT1;1 controls retrieval of Na+ from the xylem in

Arabidopsis. Plant, Cell & Environment 30, 497-507.

69. Davenport RJ, Tester M, 2000. A weakly voltage-dependent, nonselective

cation channel mediates toxic sodium influx in wheat. Plant Physiology 122,

823-34.

70. Deinlein U, Stephan AB, Horie T, Luo W, Xu G, Schroeder JI, 2014. Plant

salt-tolerance mechanisms. Trends Plant Sci 19. 258

71. Delf EM, 1912. Transpiration in succulent plants. Annals of Botany 26, 409-

42.

72. Demidchik V, Tester M, 2002. Sodium fluxes through nonselective cation

channels in the plasma membrane of protoplasts from Arabidopsis roots.

Plant Physiology 128, 379-87.

73. Di Gioia F, Signore A, Serio F, Santamaria P, 2013. Grafting Improves

Tomato Salinity Tolerance through Sodium Partitioning within the Shoot.

Hortscience 48, 855-62.

74. Dohm JC, Minoche AE, Holtgrawe D, et al., 2014. The genome of the

recently domesticated crop plant sugar beet (Beta vulgaris). Nature 505, 546-

75. Drerup MM, Schlücking K, Hashimoto K, et al., 2013. The calcineurin B-like

calcium sensors CBL1 and CBL9 together with their interacting protein

kinase CIPK26 regulate the Arabidopsis NADPH oxidase RBOHF.

Molecular Plant 6, 559-69.

76. Drozdowicz YM, Rea PA, 2001. Vacuolar H+ pyrophosphatases: from the

evolutionary backwaters into the mainstream. Trends in Plant Science 6, 206-

11.

77. Dubiella U, Seybold H, Durian G, et al., 2013. Calcium-dependent protein

kinase/NADPH oxidase activation circuit is required for rapid defense

signal propagation. Proceedings of the National Academy of Sciences 110, 8744-

9. 259

78. Dunkley TPJ, Hester S, Shadforth IP, et al., 2006. Mapping the Arabidopsis

organelle proteome. Proceedings of the National Academy of Sciences of the

United States of America 103, 6518-23.

79. Dunkley TPJ, Watson R, Griffin JL, Dupree P, Lilley KS, 2004. Localization

of Organelle Proteins by Isotope Tagging (LOPIT). Molecular & Cellular

Proteomics 3, 1128-34.

80. Dunlap JR, Binzel ML, 1996. NaCl reduces indole-3-acetic acid levels in the

roots of tomato plants independent of stress-induced abscisic acid. Plant

Physiology 112, 379-84.

81. Evans MJ, Choi W-G, Gilroy S, Morris RJ, 2016. A ROS-assisted Calcium

Wave Dependent on AtRBOHD and TPC1 Propagates the Systemic

Response to Salt Stress in Arabidopsis Roots. Plant Physiology.

82. Fan P, Nie L, Jiang P, et al., 2013a. Transcriptome Analysis of Salicornia

europaea under Saline Conditions Revealed the Adaptive Primary Metabolic

Pathways as Early Events to Facilitate Salt Adaptation. PLOS ONE 8,

e80595.

83. Fan PX, Nie LL, Jiang P, et al., 2013b. Transcriptome analysis of Salicornia

europaea under saline conditions revealed the adaptive primary metabolic

pathways as early events to facilitate salt adaptation. PLOS ONE 8.

84. Fan TW, Higashi RM, Norlyn J, Epstein E, 1989. In vivo 23Na and 31P NMR

measurement of a tonoplast Na+/H+ exchange process and its 260

characteristics in two barley cultivars. Proceedings of the National Academy of

Sciences 86, 9856-60.

85. Fao, 1985. Irrigation water management: Introduction to irrigation. In.

(2015.)

86. Fao, 1996. World Food Summit. In.:

http://www.fao.org/wfs/index_en.htm. (2013.)

87. Fao, 2007. Water at a Glance The relationsip between water, agriculture,

food security and poverty.

http://www.fao.org/nr/water/docs/waterataglance.pdf.

88. Fao, 2009. How to Feed the World in 2050.

http://www.fao.org/fileadmin/templates/wsfs/docs/expert_paper/Ho

w_to_Feed_the_World_in_2050.pdf.

89. Faostat, 2006. In.: http://faostat.fao.org/site/291/default.aspx. (2013.)

90. Feng J, Wang J, Fan P, et al., 2015. High-throughput deep sequencing reveals

that microRNAs play important roles in salt tolerance of euhalophyte

Salicornia europaea. BMC Plant Biology 15, 63.

91. Ferguson IK, 1964. Notes on the stigma morphology and flowering

behaviour in British Salicorniae. Wastonia 6, 25-7.

92. Ferjani A, Segami S, Horiguchi G, Muto Y, Maeshima M, Tsukaya H, 2011.

Keep an eye on PPi: The vacuolar type H+ pyrophosphatase regulates

postgerminative development in Arabidopsis. The Plant Cell Online 23,

2895-908. 261

93. Fernandez-Calvino L, Faulkner C, Walshaw J, et al., 2011. Arabidopsis

Plasmodesmal Proteome. PLOS ONE 6, e18880.

94. Fernández-Illescas F, Nieva FJJ, De Las Heras MÁ, Muñoz-Rodríguez AF,

2011. Dichogamy in Salicornieae species: establishment of floral sex phases

and evaluation of their frequency and efficacy in four species. Plant

Systematics and Evolution 296, 255-64.

95. Ferro M, Brugiere S, Salvi D, et al., 2010. AT_CHLORO, a Comprehensive

Chloroplast Proteome Database with Subplastidial Localization and

Curated Information on Envelope Proteins. Molecular & Cellular Proteomics

9, 1063-84.

96. Finn RD, Coggill P, Eberhardt RY, et al., 2016. The Pfam protein families

database: towards a more sustainable future. Nucleic Acids Research 44,

D279-D85.

97. Flowers TJ, Colmer TD, 2008. Salinity tolerance in halophytes. New

Phytologist 179, 945-63.

98. Flowers TJ, Colmer TD, 2015. Plant salt tolerance: adaptations in

halophytes. Annals of Botany 115, 327-31.

99. Flowers TJ, Dalmond D, 1992. Protein synthesis in halophytes: The

influence of potassium, sodium and magnesium in vitro. Plant and Soil 146,

153-61. 262

100. Flowers TJ, Galal HK, Bromham L, 2010. Evolution of halophytes:

multiple origins of salt tolerance in land plants. Functional Plant Biology 37,

604-12.

101. Flowers TJ, Hajibagheri MA, Clipson NJW, 1986. Halophytes.

Quarterly Review of Biology 61, 313-37.

102. Flowers TJ, Troke PF, Yeo AR, 1977. The mechanism of salt tolerance

in halophytes. Annual Review of Plant Physiology 28, 89-121.

103. Fricke W, 2004. Rapid and tissue-specific accumulation of solutes in

the growth zone of barley leaves in response to salinity. Planta 219, 515-25.

104. Froissard M, Belgareh-Touzé N, Buisson N, Desimone M, Frommer

WB, Haguenauer-Tsapis R, 2006. Heterologous expression of a plant uracil

transporter in yeast: Improvement of plasma membrane targeting in

mutants of the Rsp5p ubiquitin protein ligase. Biotechnology Journal 1, 308-

20.

105. Fu LM, Niu BF, Zhu ZW, Wu ST, Li WZ, 2012. CD-HIT: accelerated

for clustering the next-generation sequencing data. Bioinformatics 28, 3150-

106. Fukuda A, Yazaki Y, Ishikawa T, Koike S, Tanaka Y, 1998. Na+/H+

antiporter in tonoplast vesicles from rice roots. Plant and Cell Physiology 39,

196-201.

107. Gama PBS, Tanaka K, Eneji AE, Eltayeb AE, Siddig KE, 2009. Salt-

Induced Stress Effects on Biomass, Photosynthetic Rate, and Reactive 263

Oxygen Species-Scavenging Enzyme Accumulation in Common Bean.

Journal of Plant Nutrition 32, 837-54.

108. Gatto L, Breckels LM, Wieczorek S, Burger T, Lilley KS, 2014. Mass-

spectrometry-based spatial proteomics data analysis using pRoloc and

pRolocdata. Bioinformatics 30, 1322-4.

109. Gatto L, Lilley KS, 2012. MSnbase-an R/Bioconductor package for

isobaric tagged mass spectrometry data visualization, processing and

quantitation. Bioinformatics 28, 288-9.

110. Gaxiola RA, Li J, Undurraga S, et al., 2001. Drought and salt tolerant

plants result from overexpression of the AVP1 H+ pump. Proceedings of the

National Academy of Sciences 98, 11444-9.

111. Gharbi E, Martínez J-P, Benahmed H, et al., 2017. Phytohormone

profiling in relation to osmotic adjustment in NaCl-treated plants of the

halophyte tomato wild relative species Solanum chilense comparatively to

the cultivated glycophyte Solanum lycopersicum. Plant Science 258, 77-89.

112. Gilroy S, Białasek M, Suzuki N, et al., 2016. ROS, Calcium, and

Electric Signals: Key Mediators of Rapid Systemic Signaling in Plants. Plant

Physiology 171, 1606-15.

113. Giordano F, Aigrain L, Quail MA, et al., 2017. De novo yeast genome

assemblies from MinION, PacBio and MiSeq platforms. Scientific Reports 7,

3935. 264

114. Glenn EP, Oleary JW, Watson MC, Thompson TL, Kuehl RO, 1991.

Salicornia Bigelovii Torr.: an oilseed halophyte for seawater irrigation.

Science 251, 1065-7.

115. Gobert A, Isayenkov S, Voelker C, Czempinski K, Maathuis FJM,

2007. The two-pore channel TPK1 gene encodes the vacuolar K+

conductance and plays a role in K+ homeostasis. Proceedings of the National

Academy of Sciences 104, 10726-31.

116. Grabherr MG, Haas BJ, Yassour M, et al., 2011. Full-length

transcriptome assembly from RNA-Seq data without a reference genome.

Nat Biotech 29, 644-52.

117. Greenway H, Osmond CB, 1972. Salt responses of enzymes from

species differing in salt tolerance. Plant Physiology 49, 256-&.

118. Groen AJ, Sancho-Andres G, Breckels LM, Gatto L, Aniento F, Lilley

KS, 2014. Identification of Trans-Golgi Network Proteins in Arabidopsis

thaliana Root Tissue. Journal of Proteome Research 13, 763-76.

119. Guo K-M, Babourina O, Christopher DA, Borsics T, Rengel Z, 2008.

The cyclic nucleotide-gated channel, AtCNGC10, influences salt tolerance

in Arabidopsis. Physiologia Plantarum 134, 499-507.

120. Haas BJ, Papanicolaou A, Yassour M, et al., 2013. De novo transcript

sequence reconstruction from RNA-seq using the Trinity platform for

reference generation and analysis. Nature Protocols 8, 1494-512. 265

121. Han Y, Gao S, Muegge K, Zhang W, Zhou B, 2015. Advanced

Applications of RNA Sequencing and Challenges. Bioinformatics and Biology

Insights 9, 29-46.

122. Hanitzsch M, Schnitzer D, Seidel T, Golldack D, Dietz K-J, 2007.

Transcript level regulation of the vacuolar H+-ATPase subunit isoforms

VHA-a, VHA-E and VHA-G in Arabidopsis thaliana. Molecular Membrane

Biology 24, 507-18.

123. Hans Wedepohl K, 1995. The composition of the continental crust.

Geochimica et Cosmochimica Acta 59, 1217-32.

124. Harris BN, Sadras VO, Tester M, 2010. A water-centred framework

to assess the effects of salinity on the growth and yield of wheat and barley.

Plant and Soil 336, 377-89.

125. Harvey DR, Hall JL, Flowers TJ, Kent B, 1981. Quantitative ion

localization within Suaeda maritima leaf mesophyll cells. Planta 151, 555-60.

126. Hasegawa PM, 2013. Sodium (Na+) homeostasis and salt tolerance of

plants. Environmental and Experimental Botany 92, 19-31.

127. Hasegawa PM, Bressan RA, Zhu J-K, Bohnert HJ, 2000. Plant cellular

and molecular responses to high salinity. Annual Review of Plant Physiology

and Plant Molecular Biology 51, 463-99.

128. Hassidim M, Braun Y, Lerner HR, Reinhold L, 1990. Na+/H+ and

K+/H+ antiport in root membrane vesicles isolated from the halophyte

Atriplex and the glycophyte Cotton. Plant Physiology 94, 1795-801. 266

129. He Y, Chen Y, Yu CL, et al., 2016. Photosynthesis and yield traits in

different soybean lines in response to salt stress. Photosynthetica 54, 630-5.

130. Hess B, Van Der Vegt NFA, 2009. Cation specific binding with

protein surface charges. Proceedings of the National Academy of Sciences of the

United States of America 106, 13296-300.

131. Hodges TK, Leonard RT, Keenan TW, Bracker CE, 1972. Purification

of an Ion-Stimulated Adenosine-Triphosphatase from Plant Roots -

Association with Plasma-Membranes. Proceedings of the National Academy of

Sciences of the United States of America 69, 3307-&.

132. Hoff KJ, Lange S, Lomsadze A, Borodovsky M, Stanke M, 2016.

BRAKER1: Unsupervised RNA-Seq-Based Genome Annotation with

GeneMark-ET and AUGUSTUS. Bioinformatics 32, 767-9.

133. Hoh B, Hinz G, Jeong BK, Robinson DG, 1995. Protein storage

vacuoles form de novo during pea cotyledon development. Journal of Cell

Science 108, 299-310.

134. Hossain MS, Dietz K-J, 2016. Tuning of Redox Regulatory

Mechanisms, Reactive Oxygen Species and Redox Homeostasis under

Salinity Stress. Frontiers in Plant Science 7, 548.

135. Howe EA, Sinha R, Schlauch D, Quackenbush J, 2011. RNA-Seq

analysis in MeV. Bioinformatics 27, 3209-10. 267

136. Hua B-G, Mercier RW, Leng Q, Berkowitz GA, 2003. Plants Do It

Differently. A New Basis for Potassium/Sodium Selectivity in the Pore of

an Ion Channel. Plant Physiology 132, 1353-61.

137. Huala E, Dickerman AW, Garcia-Hernandez M, et al., 2001. The

Arabidopsis Information Resource (TAIR): a comprehensive database and

web-based information retrieval, analysis, and visualization system for a

model plant. Nucleic Acids Research 29, 102-5.

138. Huertas R, Rubio L, Cagnac O, et al., 2013. The K+/H+ antiporter

LeNHX2 increases salt tolerance by improving K+ homeostasis in

transgenic tomato. Plant, Cell & Environment 36, 2135-49.

139. Jaarsma R, De Vries RSM, De Boer AH, 2013. Effect of Salt Stress on

Growth, Na(+) Accumulation and Proline Metabolism in Potato (Solanum

tuberosum) Cultivars. PLOS ONE 8, e60183.

140. Jacoby B, Hanson JB, 1985. Controls on 22Na+ Influx in Corn Roots.

Plant Physiology 77, 930-4.

141. Jarvis DE, Ho YS, Lightfoot DJ, et al., 2017. The genome of

Chenopodium quinoa. Nature 542, 307-12.

142. Jefferies RL, Davy AJ, Rudmik T, 1981. Population Biology of the Salt

Marsh Annual Salicornia Europaea agg. Journal of Ecology 69, 17-31.

143. Jennings DH, 1968. Halophytes succulence and sodium in plants: A

unified theory. New Phytologist 67, 899-&. 268

144. Jézégou A, Llinares E, Anne C, et al., 2012. Heptahelical protein

PQLC2 is a lysosomal cationic amino acid exporter underlying the action of

cysteamine in cystinosis therapy. Proceedings of the National Academy of

Sciences 109, E3434–E43.

145. Jha D, Shirley N, Tester M, Roy SJ, 2010. Variation in salinity

tolerance and shoot sodium accumulation in Arabidopsis ecotypes linked

to differences in the natural expression levels of transporters involved in

sodium transport. Plant Cell and Environment 33, 793-804.

146. Jiang J, Li J, Xu Y, et al., 2007. RNAi knockdown of Oryza sativa root

meander curling gene led to altered root development and coiling which

were mediated by jasmonic acid signalling in rice. Plant, Cell & Environment

30, 690-9.

147. Jin Y, Jing W, Zhang Q, Zhang W, 2015. Cyclic nucleotide gated

channel 10 negatively regulates salt tolerance by mediating Na+ transport

in Arabidopsis. Journal of Plant Research 128, 211-20.

148. Johansson I, Karlsson M, Johanson U, Larsson C, Kjellbom P, 2000.

The role of aquaporins in cellular and whole plant water balance. Biochimica

et Biophysica Acta (BBA) - Biomembranes 1465, 324-42.

149. Kadereit G, Ball P, Beer S, et al., 2007. A taxonomic nightmare comes

true: Phylogeny and biogeography of glassworts (Salicornia L.,

Chenopodiaceae). Taxon 56, 1143-70. 269

150. Kadota Y, Shirasu K, Zipfel C, 2015. Regulation of the NADPH

Oxidase RBOHD During Plant Immunity. Plant and Cell Physiology 56, 1472-

80.

151. Kajitani R, Toshimoto K, Noguchi H, et al., 2014. Efficient de novo

assembly of highly heterozygous genomes from whole-genome shotgun

short reads. Genome Research 24, 1384-95.

152. Kalatzis V, Cherqui S, Antignac C, Gasnier B, 2001. Cystinosin, the

protein defective in cystinosis, is a H+‐driven lysosomal cystine transporter.

The EMBO Journal 20, 5940.

153. Kaligarič M, Bohanec B, Simonovik B, Šajna N, 2008. Genetic and

morphologic variability of annual glassworts (Salicornia L.) from the Gulf

of Trieste (Northern Adriatic). Aquatic Botany 89, 275-82.

154. Kanehisa M, Goto S, 2000. KEGG: Kyoto Encyclopedia of Genes and

Genomes. Nucleic Acids Research 28, 27-30.

155. Kaplan B, Sherman T, Fromm H, 2007. Cyclic nucleotide-gated

channels in plants. FEBS Letters 581, 2237-46.

156. Katschnig D, Bliek T, Rozema J, Schat H, 2015. Constitutive high-

level SOS1 expression and absence of HKT1;1 expression in the salt-

accumulating halophyte Salicornia dolichostachya. Plant Science 234, 144-54.

157. Katschnig D, Jaarsma R, Almeida P, Rozema J, Schat H, 2014.

Differences in proton pumping and Na+/H+ exchange at the leaf cell

tonoplast between a halophyte and a glycophyte. AoB PLANTS 6. 270

158. Khan MA, Gul B, Weber DJ, 2000. Germination responses of

Salicornia rubra to temperature and salinity. Journal of Arid Environments 45,

207-14.

159. Kimura S, Kaya H, Kawarazaki T, et al., 2012. Protein

phosphorylation is a prerequisite for the Ca2+-dependent activation of

Arabidopsis NADPH oxidases and may function as a trigger for the

positive feedback regulation of Ca2+ and reactive oxygen species. Biochimica

et Biophysica Acta - Molecular Cell Research 1823, 398-405.

160. Klink R, Haschke HP, Kramer D, Lüttge U, 1990. Membrane

particles, proteins and ATPase activity of tonoplast vesicles of

Mesembryanthemum crystallinum in the C-3 and CAM state. Botanica Acta

103, 24-31.

161. Koh S, Wiles AM, Sharp JS, Naider FR, Becker JM, Stacey G, 2002.

An Oligopeptide Transporter Gene Family in Arabidopsis. Plant Physiology

128, 21-9.

162. Komatsu S, 2008. Plasma membrane proteome in Arabidopsis and

rice. Proteomics 8, 4137-45.

163. Kong Y, Zheng YB, 2014. Potential of Producing Salicornia bigelovii

Hydroponically as a Vegetable at Moderate NaCl Salinity. Hortscience 49,

1154-7. 271

164. Koren S, Walenz BP, Berlin K, Miller JR, Bergman NH, Phillippy AM,

2017. Canu: scalable and accurate long-read assembly via adaptive k-mer

weighting and repeat separation. Genome Research.

165. Krebs M, Beyhl D, Görlich E, et al., 2010. Arabidopsis V-ATPase

activity at the tonoplast is required for efficient nutrient storage but not for

sodium accumulation. Proceedings of the National Academy of Sciences 107,

3251-6.

166. Krogh A, Larsson B, Von Heijne G, Sonnhammer ELL, 2001.

Predicting transmembrane protein topology with a hidden markov model:

application to complete genomes. Journal of Molecular Biology 305, 567-80.

167. Kurtz S, Phillippy A, Delcher AL, et al., 2004. Versatile and open

software for comparing large genomes. Genome Biology 5.

168. Langmead B, Salzberg SL, 2012. Fast gapped-read alignment with

Bowtie 2. Nat Methods. 9.

169. Lauchli A, James RA, Huang CX, Mccully M, Munns R, 2008. Cell-

specific localization of Na+ in roots of durum wheat and possible control

points for salt exclusion. Plant Cell and Environment 31, 1565-74.

170. Lebaudy A, Pascaud F, Véry A-A, et al., 2010. Preferential KAT1-

KAT2 Heteromerization Determines Inward K+ Current Properties in

Arabidopsis Guard Cells. Journal of Biological Chemistry 285, 6265-74. 272

171. Lee Y, Tsai J, Sunkara S, et al., 2005. The TIGR Gene Indices:

clustering and assembling EST and known genes and integration with

eukaryotic genomes. Nucleic Acids Research 33, D71-D4.

172. Leidi EO, Barragán V, Rubio L, et al., 2010. The AtNHX1 exchanger

mediates potassium compartmentation in vacuoles of transgenic tomato.

The Plant Journal 61, 495-506.

173. Leigh RA, 2001. Potassium homeostasis and membrane transport.

Journal of Plant Nutrition and Soil Science 164, 193-8.

174. Leigh RA, Jones RGW, 1984. A HYPOTHESIS RELATING

CRITICAL POTASSIUM CONCENTRATIONS FOR GROWTH TO THE

DISTRIBUTION AND FUNCTIONS OF THIS ION IN THE PLANT CELL.

New Phytologist 97, 1-13.

175. Lemtiri-Chlieh F, Berkowitz GA, 2004. Cyclic adenosine

monophosphate regulates calcium channels in the plasma membrane of

Arabidopsis leaf guard and mesophyll cells. Journal of Biological Chemistry

279, 35306-12.

176. Leng Q, Mercier RW, Yao W, Berkowitz GA, 1999. Cloning and First

Functional Characterization of a Plant Cyclic Nucleotide-Gated Cation

Channel. Plant Physiology 121, 753-61.

177. Li J, Yang H, Ann Peer W, et al., 2005. Arabidopsis H+ -PPase AVP1

Regulates Auxin-Mediated Organ Development. Science 310, 121-5. 273

178. Li L, Nelson CJ, Trosch J, Castleden I, Huang SB, Millar AH, 2017.

Protein Degradation Rate in Arabidopsis thaliana Leaf Growth and

Development. Plant Cell 29, 207-28.

179. Li S, Fan C, Li Y, et al., 2016. Effects of drought and salt-stresses on

gene expression in Caragana korshinskii seedlings revealed by RNA-seq.

BMC Genomics 17, 200.

180. Li ZG, Baldwin CM, Hu Q, Liu H, Luo H, 2010. Heterologous

expression of Arabidopsis H plus -pyrophosphatase enhances salt tolerance

in transgenic creeping bentgrass (Agrostis stolonifera L.). Plant Cell and

Environment 33, 272-89.

181. Liu L-L, Ren H-M, Chen L-Q, Wang Y, Wu W-H, 2013. A protein

kinase, calcineurin B-like protein-interacting protein kinase9, interacts with

calcium sensor calcineurin B-like protein3 and tegulates potassium

homeostasis under low potassium stress in Arabidopsis. Plant Physiology

161, 266-77.

182. Lohse M, Nagel A, Herter T, et al., 2014. Mercator: a fast and simple

web server for genome scale functional annotation of plant sequence data.

Plant Cell and Environment 37, 1250-8.

183. Lomsadze A, Burns PD, Borodovsky M, 2014. Integration of mapped

RNA-Seq reads into automatic training of eukaryotic gene finding

algorithm. Nucleic Acids Research 42, e119-e. 274

184. Love MI, Huber W, Anders S, 2014. Moderated estimation of fold

change and dispersion for RNA-seq data with DESeq2. Genome Biology 15,

550.

185. Lubkowitz M, 2011. The Oligopeptide Transporters: A Small Gene

Family with a Diverse Group of Substrates and Functions? Molecular Plant

4, 407-15.

186. Luo Q, Yu B, Liu Y, 2005. Differential sensitivity to chloride and

sodium ions in seedlings of Glycine max and G. soja under NaCl stress.

Journal of Plant Physiology 162, 1003-12.

187. Luo R, Liu B, Xie Y, et al., 2012. SOAPdenovo2: an empirically

improved memory-efficient short-read de novo assembler. GigaScience 1,

18-.

188. Luo Y, Reid R, Freese D, et al., 2017. Salt tolerance response revealed

by RNA-Seq in a diploid halophytic wild relative of sweet potato. Scientific

Reports 7, 9624.

189. Lv S, Nie L, Fan P, et al., 2012a. Sodium plays a more important role

than potassium and chloride in growth of Salicornia europaea. Acta

Physiologiae Plantarum 34, 503-13.

190. Lv SL, Jiang P, Chen XY, Fan PX, Wang XC, Li YX, 2012b. Multiple

compartmentalization of sodium conferred salt tolerance in Salicornia

europaea. Plant Physiology and Biochemistry 51, 47-52. 275

191. Ma JB, Zhang MR, Xiao XL, et al., 2013. Global transcriptome

profiling of Salicornia europaea L. shoots under NaCl treatment. PLOS ONE

192. Maathuis FJM, Amtmann A, 1999. K + Nutrition and Na + Toxicity:

The Basis of Cellular K + /Na + Ratios. Annals of Botany 84, 123-33.

193. Maathuis FJM, Sanders D, 1996. Mechanisms of potassium

absorption by higher plant roots. Physiologia Plantarum 96, 158-68.

194. Maier T, Güell M, Serrano L, 2009. Correlation of mRNA and protein

in complex biological samples. FEBS Letters 583, 3966-73.

195. Maimaiti A, Yunus Q, Iwanaga F, Mori N, Tanaka K, Yamanaka N,

2014. Effects of salinity on growth, photosynthesis, inorganic and organic

osmolyte accumulation in Elaeagnus oxycarpa seedlings. Acta Physiologiae

Plantarum 36, 881-92.

196. Marchler-Bauer A, Bo Y, Han L, et al., 2017. CDD/SPARCLE:

functional classification of proteins via subfamily domain architectures.

Nucleic Acids Research 45, D200-D3.

197. Mäser P, Hosoo Y, Goshima S, et al., 2002. Glycine residues in

potassium channel-like selectivity filters determine potassium selectivity in

four-loop-per-subunit HKT transporters from plants. Proc Natl Acad Sci U S

A 99. 276

198. Mccubbin T, Bassil E, Zhang S, Blumwald E, 2014. Vacuolar Na+/H+

NHX type antiporters are required for cellular K+ homeostasis, microtubule

organization and directional root growth. Plants 3, 409-26.

199. Mishra S, Behura R, Awasthi J, et al., 2014. Ectopic overexpression of

a mungbean vacuolar Na+/H+ antiporter gene (VrNHX1) leads to increased

salinity stress tolerance in transgenic Vigna unguiculata L. Walp. Molecular

Breeding 34, 1345-59.

200. Mittler R, Blumwald E, 2015. The Roles of ROS and ABA in Systemic

Acquired Acclimation. The Plant Cell 27, 64.

201. Mittler R, Vanderauwera S, Gollery M, Van Breusegem F, 2004.

Reactive oxygen gene network of plants. Trends in Plant Science 9, 490-8.

202. Mittler R, Vanderauwera S, Suzuki N, et al., 2011. ROS signaling: the

new wave? Trends in Plant Science 16, 300-9.

203. Moghaieb REA, Saneoka H, Fujita K, 2004. Effect of salinity on

osmotic adjustment, glycinebetaine accumulation and the betaine aldehyde

dehydrogenase gene expression in two halophytic plants, Salicornia

europaea and Suaeda maritima. Plant Science 166, 1345-9.

204. Møller ALB, Witzel K, Vertommen A, et al., 2011. Plant Plasma

Membrane Proteomics: Challenges and Possibilities. In: Ivanov AR,

Lazarev AV, eds. Sample Preparation in Biological Mass Spectrometry.

Dordrecht: Springer Netherlands, 411-34. 277

205. Møller IM, Jensen PE, Hansson A, 2007. Oxidative Modifications to

Cellular Components in Plants. Annual Review of Plant Biology 58, 459-81.

206. Moller IS, Gilliham M, Jha D, et al., 2009. Shoot Na+ exclusion and

increased salinity tolerance engineered by cell type-specific alteration of

Na+ transport in Arabidopsis. Plant Cell 21, 2163-78.

207. Müller M, Kunz H-H, Schroeder JI, Kemp G, Young HS, Neuhaus

HE, 2014. Decreased capacity for sodium export out of Arabidopsis

chloroplasts impairs salt tolerance, photosynthesis and plant performance.

The Plant Journal 78, 646-58.

208. Munns R, 2005. Genes and salt tolerance: bringing them together.

New Phytologist 167, 645-63.

209. Munns R, Gilliham M, 2015. Salinity tolerance of crops - what is the

cost? New Phytologist 208, 668-73.

210. Munns R, James RA, 2003. Screening methods for salinity tolerance:

a case study with tetraploid wheat. Plant and Soil 253, 201-18.

211. Munns R, James RA, Gilliham M, Flowers TJ, Colmer TD, 2016.

Tissue tolerance: an essential but elusive trait for salt-tolerant crops.

Functional Plant Biology 43, 1103-13.

212. Munns R, James RA, Xu B, et al., 2012. Wheat grain yield on saline

soils is improved by an ancestral Na+ transporter gene. Nature Biotechnology

30, 360-U173. 278

213. Munns R, Tester M, 2008. Mechanisms of salinity tolerance. Annual

Review of Plant Biology 59, 651-81.

214. Murata N, Mohanty PS, Hayashi H, Papageorgiou GC, 1992.

Glycinebetaine stabilizes the association of extrinsic proteins with the

photosynthetic oxygen evolving complex. FEBS Letters 296, 187-9.

215. Muscolo A, Panuccio MR, Piernik A, 2014. Ecology, Distribution and

Ecophysiology of Salicornia Europaea L. In: Khan MA, Böer B, Öztürk M,

Al Abdessalaam TZ, Clüsener-Godt M, Gul B, eds. Sabkha Ecosystems:

Volume IV: Cash Crop Halophyte and Biodiversity Conservation. Dordrecht:

Springer Netherlands, 233-40.

216. Narsai R, Howell KA, Millar AH, O'toole N, Small I, Whelan J, 2007.

Genome-Wide Analysis of mRNA Decay Rates and Their Determinants in

Arabidopsis thaliana. The Plant Cell 19, 3418-36.

217. Negrão S, Schmöckel SM, Tester M, 2017. Evaluating physiological

responses of plants to salinity stress. Annals of Botany 119, 1-11.

218. Nelson CJ, Millar AH, 2015. Protein turnover in plant biology. 1,

15017.

219. Nguyen V, Ribot S, Dolstra O, Niks R, Visser RF, Van Der Linden

CG, 2013. Identification of quantitative trait loci for ion homeostasis and

salt tolerance in barley (Hordeum vulgare L.). Molecular Breeding 31, 137-52. 279

220. Nikolovski N, Rubtsov D, Segura MP, et al., 2012. Putative

Glycosyltransferases and Other Plant Golgi Apparatus Proteins Are

Revealed by LOPIT Proteomics. Plant Physiology 160, 1037-51.

221. Nikolovski N, Shliaha PV, Gatto L, Dupree P, Lilley KS, 2014. Label-

Free Protein Quantification for Plant Golgi Protein Localization and

Abundance. Plant Physiology 166, 1033-43.

222. Nobel PS, 1991. Physiochemical and environmental plant

physiology. Academic Press, New York.

223. Oh D-H, Leidi E, Zhang Q, et al., 2009. Loss of halophytism by

interference with SOS1 expression. Plant Physiology 151, 210-22.

224. Ohori T, Fujiyama H, 2011. Water deficit and abscisic acid

production of Salicornia bigelovii under salinity stress. Soil Science and Plant

Nutrition 57, 566-72.

225. Oka T, Nemoto T, Jigami Y, 2007. Functional analysis of Arabidopsis

thaliana RHM2/MUM4, a multidomain protein involved in UDP-D-

glucose to UDP-L-rhamnose conversion. Journal of Biological Chemistry 282,

5389-403.

226. Omasits U, Ahrens CH, Müller S, Wollscheid B, 2014. Protter:

interactive protein feature visualization and integration with experimental

proteomic data. Bioinformatics 30, 884-6. 280

227. Orsini F, Accorsi M, Gianquinto G, et al., 2011. Beyond the ionic and

osmotic response to salinity in Chenopodium quinoa: functional

elements of successful halophytism. Functional Plant Biology 38, 818-31.

228. Orsini F, Alnayef M, Bona S, Maggio A, Gianquinto G, 2012. Low

stomatal density and reduced transpiration facilitate strawberry adaptation

to salinity. Environmental and Experimental Botany 81, 1-10.

229. Orsini F, D'urzo MP, Inan G, et al., 2010. A comparative study of salt

tolerance parameters in 11 wild relatives of Arabidopsis thaliana. Journal of

Experimental Botany 61, 3787-98.

230. Osawa H, Stacey G, Gassmann W, 2005. ScOPT1 and AtOPT4

function as proton-coupled oligopeptide transporters with broad but

distinct substrate specificities. Biochemical Journal 393, 267.

231. Pardo JM, Cubero B, Leidi EO, Quintero FJ, 2006. Alkali cation

exchangers: roles in cellular homeostasis and stress tolerance. Journal of

Experimental Botany 57, 1181-99.

232. Pardo JM, Quintero FJ, 2002. Plants and sodium ions: keeping

company with the enemy. Genome Biology 3.

233. Parida AK, Das AB, 2005. Salt tolerance and salinity effects on plants:

a review. Ecotoxicology and Environmental Safety 60, 324-49.

234. Paris N, Stanley CM, Jones RL, Rogers JC, 1996. Plant cells contain

two functionally distinct vacuolar compartments. Cell 85, 563-72. 281

235. Parks GE, Dietrich MA, Schumaker KS, 2002. Increased vacuolar

Na+/H+ exchange activity in Salicornia bigelovii Torr. in response to NaCl.

Journal of Experimental Botany 53, 1055-65.

236. Pasapula V, Shen G, Kuppu S, et al., 2011. Expression of an

Arabidopsis vacuolar H+-pyrophosphatase gene (AVP1) in cotton improves

drought- and salt tolerance and increases fibre yield in the field conditions.

Plant Biotechnology Journal 9, 88-99.

237. Perkins DN, Pappin DJC, Creasy DM, Cottrell JS, 1999. Probability-

based protein identification by searching sequence databases using mass

spectrometry data. Electrophoresis 20, 3551-67.

238. Pilot G, Lacombe BT, Gaymard F, et al., 2001. Guard Cell Inward K+

Channel Activity inArabidopsis Involves Expression of the Twin Channel

Subunits KAT1 and KAT2. Journal of Biological Chemistry 276, 3215-21.

239. Plett D, Safwat G, Gilliham M, et al., 2010. Improved Salinity

Tolerance of Rice Through Cell Type-Specific Expression of AtHKT1;1.

PLOS ONE 5, e12571.

240. Poole RJ, 1978. Energy coupling for membrane transport. Annual

Review of Plant Physiology 29, 437-60.

241. Popova OV, Golldack D, 2007. In the halotolerant Lobularia maritima

(Brassicaceae) salt adaptation correlates with activation of the vacuolar H+-

ATPase and the vacuolar Na+/H+ antiporter. Journal of Plant Physiology 164,

1278-88. 282

242. Pottosin I, Dobrovinskaya O, 2014. Non-selective cation channels in

plasma and vacuolar membranes and their contribution to K+ transport.

Journal of Plant Physiology 171, 732-42.

243. Qi X, Li M-W, Xie M, et al., 2014. Identification of a novel salt

tolerance gene in wild soybean by whole-genome sequencing. 5, 4340.

244. Quan R, Shang M, Zhang H, Zhao Y, Zhang J, 2004. Engineering of

enhanced glycine betaine synthesis improves drought tolerance in maize.

Plant Biotechnology Journal 2, 477-86.

245. Quevillon E, Silventoinen V, Pillai S, et al., 2005. InterProScan:

protein domains identifier. Nucleic Acids Research 33, W116-W20.

246. Quintero FJ, Blatt MR, Pardo JM, 2000. Functional conservation

between yeast and plant endosomal Na+/H+ antiporters. FEBS Letters 471,

224-8.

247. Rabilloud T, 2009. Membrane proteins and proteomics: Love is

possible, but so difficult. Electrophoresis 30, S174-S80.

248. Rabilloud T, Lelong C, 2011. Two-dimensional gel electrophoresis in

proteomics: A tutorial. Journal of Proteomics 74, 1829-41.

249. Rains DW, 1972. Salt transport by plants in relation to salinity.

Annual Review of Plant Physiology 23, 367-&.

250. Ray DK, Mueller ND, West PC, Foley JA, 2013. Yield Trends Are

Insufficient to Double Global Crop Production by 2050. PLOS ONE 8,

e66428. 283

251. Rea PA, Poole RJ, 1985. Proton-translocating inorganic

pyrophosphatase in red beet (Beta vulgaris L.) tonoplast vesicles. Plant

Physiology 77, 46-52.

252. Reddy MP, Sanish S, Iyengar ERR, 1993. Compartmentation of ions

and organic compounds in Salicornia brachiata Roxb. Biologia Plantarum 35,

547.

253. Reisen D, Leborgne-Castel N, Özalp C, Chaumont F, Marty F, 2003.

Expression of a cauliflower tonoplast aquaporin tagged with GFP in

tobacco suspension cells correlates with an increase in cell size. Plant

Molecular Biology 52, 387-400.

254. Ridley BL, O'neill MA, Mohnen D, 2001. Pectins: structure,

biosynthesis, and oligogalacturonide-related signaling. Phytochemistry 57,

929-67.

255. Riemann M, Dhakarey R, Hazman M, Miro B, Kohli A, Nick P, 2015.

Exploring Jasmonates in the Hormonal Network of Drought and Salinity

Responses. Frontiers in Plant Science 6.

256. Riley JP, Tongudai M, 1967. The major cation/chlorinity ratios in sea

water. Chemical Geology 2, 263-9.

257. Rivero RM, Mestre TC, Mittler R, Rubio F, Garcia-Sanchez F,

Martinez V, 2014. The combined effect of salinity and heat reveals a specific

physiological, biochemical and molecular response in tomato plants. Plant,

Cell & Environment 37, 1059-73. 284

258. Rivers WG, Weber DJ, 1971a. Influence of Salinity and Temperature

on Seed Germination in Salicornia-Bigelovii. Physiologia Plantarum 24, 73-&.

259. Rivers WG, Weber DJ, 1971b. The influence of salinity and

temperature on seed germination in Salicornia bigelovii. Physiologia

Plantarum 24, 73-5.

260. Rodríguez-Navarro A, Ramos J, 1984. Dual system for potassium

transport in Saccharomyces cerevisiae. Journal of Bacteriology 159, 940-5.

261. Rodríguez-Rosales MP, Jiang X, Gálvez FJ, Aranda MN, Cubero B,

Venema K, 2008. Overexpression of the tomato K+/H+ antiporter LeNHX2

confers salt tolerance by improving potassium compartmentalization. New

Phytologist 179, 366-77.

262. Rodriguez HG, Roberts J, Jordan WR, Drew MC, 1997. Growth,

water relations, and accumulation of organic and inorganic solutes in roots

of maize seedlings during salt stress. Plant Physiology 113, 881-93.

263. Romero-Aranda R, Moya JL, Tadeo FR, Legaz F, Primo-Millo E,

Talon M, 1998. Physiological and anatomical disturbances induced by

chloride salts in sensitive and tolerant citrus: beneficial and detrimental

effects of cations. Plant, Cell & Environment 21, 1243-53.

264. Roy SJ, Negrão S, Tester M, 2014. Salt resistant crop plants. Current

Opinion in Biotechnology 26, 115-24. 285

265. Rungie JM, Wiskich JT, 1973. Salt stimulated adenosine

triphosphatase from smooth microsomes of turnip. Plant Physiology 51,

1064-8.

266. Ryu H, Cho Y-G, 2015. Plant hormones in salt stress tolerance. Journal

of Plant Biology 58, 147-55.

267. Sadowski PG, Dunkley TPJ, Shadforth IP, et al., 2006. Quantitative

proteomic approach to study subcellular localization of membrane

proteins. Nat. Protocols 1, 1778-89.

268. Sah SK, Reddy KR, Li J, 2016. Abscisic Acid and Abiotic Stress

Tolerance in Crop Plants. Frontiers in Plant Science 7, 571.

269. Sakamoto A, Murata N, 2000. Genetic engineering of glycinebetaine

synthesis in plants: current status and implications for enhancement of

stress tolerance. Journal of Experimental Botany 51, 81-8.

270. Sanyal SK, Pandey A, Pandey GK, 2015. The CBL–CIPK signaling

module in plants: a mechanistic perspective. Physiologia Plantarum 155, 89-

108.

271. Sarafian V, Kim Y, Poole RJ, Rea PA, 1992. Molecular cloning and

sequence of cDNA encoding the pyrophosphate-energized vacuolar

membrane proton pump of Arabidopsis thaliana. Proceedings of the National

Academy of Sciences 89, 1775-9. 286

272. Sato Y, Sakaguchi M, 2005. Topogenic Properties of Transmembrane

Segments of Arabidopsis thaliana NHX1 Reveal a Common Topology Model

of the Na+/H+ Exchanger Family. The Journal of Biochemistry 138, 425-31.

273. Saudek V, 2012. Cystinosin, MPDU1, SWEETs and KDELR Belong to

a Well-Defined Protein Family with Putative Function of Cargo Receptors

Involved in Vesicle Trafficking. PLOS ONE 7, e30876.

274. Saxena IM, Brown RM, 2005. Cellulose Biosynthesis: Current Views

and Evolving Concepts. Annals of Botany 96, 9-21.

275. Schilling RK, Marschner P, Shavrukov Y, et al., 2014. Expression of

the Arabidopsis vacuolar H+-pyrophosphatase gene (AVP1) improves the

shoot biomass of transgenic barley and increases grain yield in a saline field.

Plant Biotechnology Journal 12, 378-86.

276. Schilling RK, Tester M, Marschner P, Plett DC, Roy SJ, 2016. AVP1:

One Protein, Many Roles. Trends in Plant Science 22, 154-62.

277. Schindelin J, Arganda-Carreras I, Frise E, et al., 2012. Fiji: an open-

source platform for biological-image analysis. Nat Meth 9, 676-82.

278. Schmidt MHW, Vogel A, Denton A, et al., 2017. Reconstructing The

Gigabase Plant Genome Of Solanum pennellii Using Nanopore Sequencing.

bioRxiv.

279. Schmidt R, Mieulet D, Hubberten H-M, et al., 2013. SALT-

RESPONSIVE ERF1 regulates reactive oxygen species–dependent signaling 287

during the initial response to salt stress in rice. The Plant Cell Online 25, 2115-

31.

280. Schmöckel SM, Garcia AF, Berger B, Tester M, Webb AaR, Roy SJ,

2015. Different NaCl-Induced Calcium Signatures in the Arabidopsis

thaliana Ecotypes Col-0 and C24. PLOS ONE 10, e0117564.

281. Schumacher K, Krebs M, 2010. The V-ATPase: small cargo, large

effects. Current Opinion in Plant Biology 13, 724-30.

282. Searle BC, 2010. Scaffold: A bioinformatic tool for validating

MS/MS-based proteomic studies. Proteomics 10, 1265-9.

283. Shabala L, Mackay A, Tian Y, Jacobsen S-E, Zhou D, Shabala S, 2012.

Oxidative stress protection and stomatal patterning as components of

salinity tolerance mechanism in quinoa (Chenopodium quinoa). Physiologia

Plantarum 146, 26-38.

284. Shabala S, Hariadi Y, Jacobsen S-E, 2013. Genotypic difference in

salinity tolerance in quinoa is determined by differential control of xylem

Na+ loading and stomatal density. Journal of Plant Physiology 170, 906-14.

285. Sharp RE, Wu YJ, Voetberg GS, Saab IN, Lenoble ME, 1994.

Confirmation That Abscisic-Acid Accumulation Is Required for Maize

Primary Root Elongation at Low Water Potentials. Journal of Experimental

Botany 45, 1743-51. 288

286. Shavrukov Y, Bovill J, Afzal I, et al., 2013. HVP10 encoding V-PPase

is a prime candidate for the barley HvNax3 sodium exclusion gene:

evidence from fine mapping and expression analysis. Planta 237, 1111-22.

287. Shi H, Ishitani M, Kim C, Zhu J-K, 2000. The Arabidopsis thaliana

salt tolerance gene SOS1 encodes a putative Na+/H+ antiporter. Proceedings

of the National Academy of Sciences 97, 6896-901.

288. Shi H, Quintero FJ, Pardo JM, Zhu J-K, 2002. The putative plasma

membrane Na+/H+ antiporter SOS1 controls long-distance Na+ transport in

plants. The Plant Cell Online 14, 465-77.

289. Shivaraj SM, Deshmukh RK, Rai R, Bélanger R, Agrawal PK, Dash

PK, 2017. Genome-wide identification, characterization, and expression

profile of aquaporin gene family in flax (Linum usitatissimum). 7, 46137.

290. Shoji S, Yoichi N, Masa-H. S, Masayoshi M, 2010. Quantification,

organ-specific accumulation and intracellular localization of type II H+-

pyrophosphatase in Arabidopsis thaliana. Oxford University Press 51, 11.

291. Silva P, Facanha AR, Tavares RM, Geros H, 2010. Role of tonoplast

proton pumps and Na+/H+ antiport system in salt tolerance of Populus

euphratica Oliv. Journal of Plant Growth Regulation 29, 23-34.

292. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov

EM, 2015. BUSCO: assessing genome assembly and annotation

completeness with single-copy orthologs. Bioinformatics 31, 3210-2. 289

293. Simpson JT, Wong K, Jackman SD, Schein JE, Jones SJM, Birol İ, 2009.

ABySS: A parallel assembler for short read sequence data. Genome Research

19, 1117-23.

294. Smolders K, Lombaert N, Valkenborg D, Baggerman G, Arckens L,

2015. An effective plasma membrane proteomics approach for small tissue

samples. 5, 10917.

295. Spanswick RM, Williams EJ, 1964. Electrical potentials and Na, K,

and Cl concentrations in the vacuole and cytoplasm of Nitella translucens.

Journal of Experimental Botany 15, 193-200.

296. Staal M, Maathuis FJM, Elzenga JTM, Overbeek JHM, Prins HBA,

1991. Na+/H+ antiport activity in tonoplast vesicles from roots of the salt-

tolerant Plantago maritima and the salt-sensitive Plantago media. Physiologia

Plantarum 82, 179-84.

297. Stanke M, Waack S, 2003. Gene prediction with a hidden Markov

model and a new intron submodel. Bioinformatics 19, ii215-ii25.

298. Steijger T, Abril JF, Engstrom PG, et al., 2013. Assessment of

transcript reconstruction methods for RNA-seq. Nat Meth 10, 1177-84.

299. Storey R, Walker RR, 1998. Citrus and salinity. Scientia Horticulturae

78, 39-81.

300. Strompen G, Dettmer J, Stierhof Y-D, Schumacher K, Jürgens G,

Mayer U, 2005. Arabidopsis vacuolar H+-ATPase subunit E isoform 1 is 290

required for Golgi organization and vacuole function in embryogenesis. The

Plant Journal 41, 125-32.

301. Sultana N, Ikeda T, Itoh R, 1999. Effect of NaCl salinity on

photosynthesis and dry matter accumulation in developing rice grains.

Environmental and Experimental Botany 42, 211-20.

302. Sunarpi, Horie T, Motoda J, et al., 2005. Enhanced salt tolerance

mediated by AtHKT1 transporter-induced Na+ unloading from xylem

vessels to xylem parenchyma cells. The Plant Journal 44, 928-38.

303. Sussman MR, Harper JF, 1989. Molecular biology of the plasma

membrane of higher plants. Plant Cell and Environment 1, 953-60.

304. Suzuki K, Yamaji N, Costa A, et al., 2016. OsHKT1;4-mediated Na+

transport in stems contributes to Na+ exclusion from leaf blades of rice at

the reproductive growth stage upon salt stress. BMC Plant Biology 16, 22.

305. Sykes SR, 1992. The Inheritance of Salt Exclusion in Woody Perennial

Fruit Species. Plant and Soil 146, 123-9.

306. Sze H, 1980. Nigericin-stimulated ATPase activity in microsomal

vesicles of tobacco callus. Proceedings of the National Academy of Sciences 77,

5904-8.

307. Sze H, 1984. H+-translocating ATPases of the plasma membrane and

tonoplast of plant cells. Physiologia Plantarum 61, 683-91. 291

308. Sze H, Li X, Palmgren MG, 1999. Energization of plant cell

membranes by H+-pumping ATPases: regulation and biosynthesis. The

Plant Cell Online 11, 677-89.

309. Tabuchi T, Kawaguchi Y, Azuma T, Nanmori T, Yasuda T, 2005.

Similar Regulation Patterns of Choline Monooxygenase,

Phosphoethanolamine N-Methyltransferase and S-Adenosyl-l-Methionine

Synthetase in Leaves of the Halophyte Atriplex nummularia L. Plant and Cell

Physiology 46, 505-13.

310. Tan, Zhi Y, Thimann KV, 1989. The roles of carbon dioxide and

abscisic acid in the production of ethylene. Physiologia Plantarum 75, 13-9.

311. Tan S, Tan HT, Chung MCM, 2008. Membrane proteins and

membrane proteomics. Proteomics 8, 3924-32.

312. Tanca A, Abbondio M, Pisanu S, Pagnozzi D, Uzzau S, Addis MF,

2014. Critical comparison of sample preparation strategies for shotgun

proteomic analysis of formalin-fixed, paraffin-embedded samples: insights

from liver tissue. Clinical Proteomics 11, 28.

313. Tanner W, Caspari T, 1996. Membrane transport carriers. Annual

Review of Plant Physiology and Plant Molecular Biology 47, 595-626.

314. Teakle NL, Real D, Colmer TD, 2006. Growth and ion relations in

response to combined salinity and waterlogging in the perennial forage

legumes Lotus corniculatus and Lotus tenuis. Plant and Soil 289, 369. 292

315. Termaat A, Passioura JB, Munns R, 1985. Shoot turgor does not limit

shoot growth of NaCl-affected wheat and barley. Plant Physiology 77, 869-

72.

316. Tester M, Davenport R, 2003. Na+ tolerance and Na+ transport in

higher plants. Annals of Botany 91, 503-27.

317. Tester M, Langridge P, 2010. Breeding technologies to increase crop

production in a changing world. Science 327, 818-22.

318. Thimm O, Bläsing O, Gibon Y, et al., 2004. mapman: a user-driven

tool to display genomics data sets onto diagrams of metabolic pathways

and other biological processes. The Plant Journal 37, 914-39.

319. Tian F, Chang E, Li Y, Sun P, Hu J, Zhang J, 2017. Expression and

integrated network analyses revealed functional divergence of NHX-type

Na+/H+ exchanger genes in poplar. Scientific Reports 7, 2607.

320. Torii KU, 2004. Leucine-Rich Repeat Receptor Kinases in Plants:

Structure, Function, and Signal Transduction Pathways. International

Review of Cytology 234, 1-46.

321. Tracy FE, Gilliham M, Dodd AN, Webb AA, Tester M, 2008. NaCl-

induced changes in cytosolic free Ca2+ in Arabidopsis thaliana are

heterogeneous and modified by external ionic composition. Plant Cell and

Environment 31, 1063-73.

322. Trapnell C, Pachter L, Salzberg SL, 2009. TopHat: discovering splice

junctions with RNA-Seq. Bioinformatics 25, 1105-11. 293

323. Trotter MWB, Sadowski PG, Dunkley TPJ, Groen AJ, Lilley KS, 2010.

Improved sub-cellular resolution via simultaneous analysis of organelle

proteomics data across varied experimental conditions. Proteomics 10, 4213-

324. Tyerman SD, Skerrett IM, 1998. Root ion channels and salinity.

Scientia Horticulturae 78, 175-235.

325. Tzeng CM, Yang CY, Yang SJ, et al., 1996. Subunit structure of

vacuolar proton-pyrophosphatase as determined by radiation inactivation.

Biochemical Journal 316, 143-7.

326. Un-Desa, 2013. United Nations Department of Economic and Social

Affairs. World population prospects: The 2012 Revision.

http://esa.un.org/wpp/.

327. Unesco, 2006. Facts and figures about water foorprints and virtual

water. In. (2015.)

328. Ungar IA, 1967. Influence of salinity and temperature on seed

germination. The Ohio Journal of Science 67, 120-3.

329. Untergasser A, Nijveen H, Rao X, Bisseling T, Geurts R, Leunissen

JA, 2007. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids

Res 35, W71-4.

330. Usadel B, Nagel A, Steinhauser D, et al., 2006. PageMan: An

interactive ontology tool to generate, display, and annotate overview

graphs for profiling experiments. BMC Bioinformatics 7, 535-. 294

331. Usadel B, Poree F, Nagel A, Lohse M, Czedik-Eysenberg A, Stitt M,

2009. A guide to using MapMan to visualize and compare Omics data in

plants: a case study in the crop species, Maize. Plant Cell and Environment

32, 1211-29.

332. Van De Peer Y, Mizrachi E, Marchal K, 2017. The evolutionary

significance of polyploidy. Nat Rev Genet 18, 411-24.

333. Vanburen R, Bryant D, Edger PP, et al., 2015. Single-molecule

sequencing of the desiccation-tolerant grass Oropetium thomaeum. Nature

527, 508-11.

334. Vanlerberghe GC, 2013. Alternative Oxidase: A Mitochondrial

Respiratory Pathway to Maintain Metabolic and Signaling Homeostasis

during Abiotic and Biotic Stress in Plants. International Journal of Molecular

Sciences 14, 6805-47.

335. Venema K, Quintero FJ, Pardo JM, Donaire JP, 2002. The Arabidopsis

Na+/H+ exchanger AtNHX1 catalyzes low affinity Na+ and K+ transport in

reconstituted liposomes. Journal of Biological Chemistry 277, 2413-8.

336. Ventura Y, Sagi M, 2013. Halophyte crop cultivation: The case for

Salicornia and Sarcocornia. Environmental and Experimental Botany 92, 144-

53.

337. Vera-Estrella R, Barkla BJ, García-Ramírez L, Pantoja O, 2005. Salt

stress in Thellungiella halophila activates Na+ transport mechanisms required

for salinity tolerance. Plant Physiology 139, 1507-17. 295

338. Vogel C, Marcotte EM, 2012. Insights into the regulation of protein

abundance from proteomic and transcriptomic analyses. Nat Rev Genet 13,

227-32.

339. Volkov AG, Paula S, Deamer DW, 1997. Two mechanisms of

permeation of small neutral molecules and hydrated ions across

phospholipid bilayers. Bioelectrochemistry and Bioenergetics 42, 153-60.

340. Vysotskaya L, Hedley PE, Sharipova G, et al., 2010. Effect of salinity

on water relations of wild barley plants differing in salt tolerance. AoB

PLANTS 2010, plq006-plq.

341. Walker DJ, Leigh RA, Miller AJ, 1996. Potassium homeostasis in

vacuolate plant cells. Proceedings of the National Academy of Sciences 93,

10510-4.

342. Wallin E, Heijne GV, 1998. Genome-wide analysis of integral

membrane proteins from eubacterial, archaean, and eukaryotic organisms.

Protein Science 7, 1029-38.

343. Wan SB, Mak MW, Kung SY, 2012. mGOASVM: Multi-label protein

subcellular localization based on gene ontology and support vector

machines. BMC Bioinformatics 13.

344. Wang J, Li B, Meng Y, et al., 2015. Transcriptomic profiling of the salt-

stress response in the halophyte Halogeton glomeratus. BMC Genomics 16,

169. 296

345. Wang L, He X, Zhao Y, Shen Y, Huang Z, 2011. Wheat vacuolar H+-

ATPase subunit B cloning and its involvement in salt tolerance. Planta 234,

1-7.

346. Wang XC, Fan PX, Song HM, Chen XY, Lil XF, Li YX, 2009a.

Comparative proteomic analysis of differentially expressed proteins in

shoots of Salicornia europaea under different salinity. Journal of Proteome

Research 8, 3331-45.

347. Wang Y, Li K, Li X, 2009b. Auxin redistribution modulates plastic

development of root system architecture under salt stress in Arabidopsis

thaliana. Journal of Plant Physiology 166, 1637-45.

348. Washburn MP, Wolters D, Yates JR, 2001. Large-scale analysis of the

yeast proteome by multidimensional protein identification technology. Nat

Biotech 19, 242-7.

349. Webb KL, 1966. NaCl effects on growth and transpiration in

Salicornia bigelovii a salt-marsh halophyte. Plant and Soil 24, 261-8.

350. Wences AH, Schatz MC, 2015. Metassembler: merging and

optimizing de novo genome assemblies. Genome Biology 16, 207.

351. Wicke B, Smeets E, Dornburg V, et al., 2011. The global technical and

economic potential of bioenergy from salt-affected soils. Energy &

Environmental Science 4, 2669-81. 297

352. Wilson C, Read JJ, Abo-Kassem E, 2002. EFFECT OF MIXED-SALT

SALINITY ON GROWTH AND ION RELATIONS OF A QUINOA AND A

WHEAT VARIETY. Journal of Plant Nutrition 25, 2689-704.

353. Wink M, 1993. The plant vacuole: A multifunctional compartment.

Journal of Experimental Botany 44, 231-46.

354. Wiśniewski JR, Zougman A, Mann M, 2009. Combination of FASP

and StageTip-Based Fractionation Allows In-Depth Analysis of the

Hippocampal Membrane Proteome. Journal of Proteome Research 8, 5674-8.

355. Wisniewski JR, Zougman A, Nagaraj N, Mann M, 2009. Universal

sample preparation method for proteome analysis. Nature Methods 6, 359-

U60.

356. Wolf O, Munns R, Tonnet ML, Jeschke WD, 1991. The role of the stem

in the partitioning of Na+ and K+ in salt treated barley. Journal of

Experimental Botany 42, 697-704.

357. Wormit A, Traub M, Flörchinger M, Neuhaus H e, Möhlmann T,

2004. Characterization of three novel members of the Arabidopsis thaliana

equilibrative nucleoside transporter (ENT) family. Biochemical Journal 383,

19-26.

358. Wright EM, 2013. Glucose transport families SLC5 and SLC50.

Molecular aspects of medicine 34, 183-96. 298

359. Wu C-A, Yang G-D, Meng Q-W, Zheng C-C, 2004. The cotton

GhNHX1 gene encoding a novel putative tonoplast Na+/H+ antiporter

plays an important role in salt stress. Plant and Cell Physiology 45, 600-7.

360. Wu S, Su Q, An LJ, 2010. Isolation of choline monooxygenase (CMO)

gene from Salicornia europaea and enhanced salt tolerance of transgenic

tobacco with CMO genes. Indian Journal of Biochemistry & Biophysics 47, 298-

305.

361. Xu C, Jiao C, Sun H, et al., 2017. Draft genome of spinach and

transcriptome diversity of 120 Spinacia accessions. 8, 15275.

362. Xu J, Li H-D, Chen L-Q, et al., 2006. A protein kinase, interacting with

two calcineurin B-like proteins, regulates K+ transporter AKT1 in

Arabidopsis. Cell 125, 1347-60.

363. Xu Y, Zhou Y, Hong S, et al., 2013. Functional characterization of a

wheat NHX antiporter gene TaNHX2 that encodes a K+/H+ exchanger.

PLOS ONE 8.

364. Yamada M, Kuroda C, Fujiyama H, 2016. Function of sodium and

potassium in growth of sodium-loving Amaranthaceae species. Soil Science

and Plant Nutrition 62, 20-6.

365. Yamaguchi T, Aharon GS, Sottosanto JB, Blumwald E, 2005.

Vacuolar Na+/H+ antiporter cation selectivity is regulated by calmodulin

from within the vacuole in a Ca2+ and pH dependent manner. Proceedings of

the National Academy of Sciences of the United States of America 102, 16107-12. 299

366. Yamaguchi T, Apse MP, Shi H, Blumwald E, 2003. Topological

analysis of a plant vacuolar Na+/H+ antiporter reveals a luminal C terminus

that regulates antiporter cation selectivity. Proceedings of the National

Academy of Sciences 100, 12510-5.

367. Yancey P, Clark M, Hand S, Bowlus R, Somero G, 1982. Living with

water stress: evolution of osmolyte systems. Science 217, 1214-22.

368. Yang H, Murphy A, 2013. Membrane Preparation, Sucrose Density

Gradients and Two-phase Separation Fractionation from Five-day-old

Arabidopsis seedlings. Bio-protocol 3.

369. Yang R, Jarvis DE, Chen H, et al., 2013. The Reference Genome of the

Halophytic Plant Eutrema salsugineum. Frontiers in Plant Science 4, 46.

370. Ye C, Hill CM, Wu S, Ruan J, Ma Z, 2016. DBG2OLC: Efficient

Assembly of Large Genomes Using Long Erroneous Reads of the Third

Generation Sequencing Technologies. 6, 31900.

371. Yeo AR, 1983. Salinity Resistance - Physiologies and Prices.

Physiologia Plantarum 58, 214-22.

372. Yeo AR, Caporn SJM, Flowers TJ, 1985. The effect of salinity upon

photosynthesis in rice (Oryza sativa L.): gas exchange by individual leaves

in relation to their salt content. Journal of Experimental Botany 36, 1240-8.

373. Yeo AR, Flowers TJ, 1982. Accumulation and Localization of

Sodium-Ions within the Shoots of Rice (Oryza-Sativa) Varieties Differing in

Salinity Resistance. Physiologia Plantarum 56, 343-8. 300

374. Yeo AR, Flowers TJ, 1986. Ion transport in Suaeda maritima: Its

relation to growth and implications for the pathway of radial transport of

ions across the root. Journal of Experimental Botany 37, 143-59.

375. Yim WC, Lim SD, Wone BWM, et al. Genome Sequencing of the

Common Ice Plant (Mesembryanthemum crystallinum L.). Proceedings of the

Plant and Animal Genome Conference (PAG XXIV), 2017.

376. Yokoi S, Quintero FJ, Cubero B, et al., 2002. Differential expression

and function of Arabidopsis thaliana NHX Na+/H+ antiporters in the salt

stress response. The Plant Journal 30, 529-39.

377. Zhang F, Zhu G, Du L, et al., 2016. Genetic regulation of salt stress

tolerance revealed by RNA-Seq in cotton diploid wild species, Gossypium

davidsonii. 6, 20582.

378. Zhang J-L, Flowers TJ, Wang S-M, 2009a. Mechanisms of sodium

uptake by roots of higher plants. Plant and Soil 326, 45.

379. Zhang L, Tian L-H, Zhao J-F, Song Y, Zhang C-J, Guo Y, 2009b.

Identification of an Apoplastic Protein Involved in the Initial Phase of Salt

Stress Response in Rice Root by Two-Dimensional Electrophoresis. Plant

Physiology 149, 916-28.

380. Zhao J, Cheng N-H, Motes CM, et al., 2008. AtCHX13 Is a Plasma

Membrane K+Transporter. Plant Physiology 148, 796. 301

381. Zhao K-F, Fan H, Song J, et al., 2005. Two Na+ and Cl−

hyperaccumulators of the Chenopodiaceae. Journal of Integrative Plant

Biology 47, 311-8.

382. Zhen RG, Kim EJ, Rea PA, 1997. The molecular and biochemical basis

of pyrophosphate energized proton translocation at the vacuolar

membrane. In: R.A. Leigh DS, Callow JA, eds. Advances in Botanical Research.

Academic Press, 297-337. (25.)

383. Zörb C, Senbayram M, Peiter E, 2014. Potassium in agriculture –

Status and perspectives. Journal of Plant Physiology 171, 656-69.

302

APPENDIX

303

Appendix Figure 1. Phenotypic variation across different Salicornia bigelovii inbred lines. 304

Appendix Figure 2. Differences in flowering time a) and growth and selection b) of Salicornia bigelovii inbred lines in the greenhouse.

Selected plants were bagged in order to prevent cross pollination. 305

Appendix Figure 3. Replicates comparisons for the treatment 0 mM NaCl after 1 week. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

306

Appendix Figure 4. Replicates comparisons for the treatment 50 mM NaCl after 1 week. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

307

Appendix Figure 5. Replicates comparisons for the treatment 200 mM NaCl after 1 week. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

308

Appendix Figure 6. Replicates comparisons for the treatment 600 mM NaCl after 1 week. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

309

Appendix Figure 7. Replicates comparisons for the treatment 0 mM NaCl after 6 weeks. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

310

Appendix Figure 8. Replicates comparisons for the treatment 50 mM NaCl after 6 weeks. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

311

Appendix Figure 9. Replicates comparisons for the treatment 200 mM NaCl after 6 weeks. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

312

Appendix Figure 10. Replicates comparisons for the treatment 600 mM NaCl after 6 weeks. a) Mapped fragments to the transcriptome; b) heatmap of Person’s correlation of counts per million (CPM) per gene; c) Pairwise MA plots (x-axis: mean log(CPM), y-axis log(fold_change); d) Pairwise comparisons of replicate log(CPM) values. Data points more than 2-fold different are highlighted in red.

313

Appendix Figure 11. PageMan ontology overrepresentation of differentially expressed genes from plants treated with 0, 50, 200, and 600 mM NaCl for 1 and 6 weeks.

Downregulation in red, upregulation in blue. 314

Continuation of Appendix Figure 11. PageMan ontology overrepresentation of differentially expressed genes from plants treated with 0, 50, 200, and 600 mM NaCl for 1 and 6 weeks. Downregulation in red, upregulation in blue. 315

317

Appendix Table 1. Genome assembly statistics of Salicornia brachiata final assembly. A = 0.3281 C = 0.1719 G = 0.1718 T = 0.3281 N = 0.0355 GC = 0.3438 Main genome scaffold total: 5,230 Main genome contig total: 15,179 Main genome scaffold sequence 468.596 Mb total: Main genome contig sequence 451.946 Mb 3.553% gap total: Main genome scaffold N/L50: 429/302.404 kb Main genome contig N/L50: 2141/63.821 kb Main genome scaffold N/L90: 1602/70.528 kb Main genome contig N/L90: 7219/17.069 kb Max scaffold length: 2.531 Mb Max contig length: 448.447 kb Number of scaffolds > 50 KB: 1,879 % main genome in scaffolds > 50 93 55% kb: Number of Number of Total Scaffold Total Contig Scaffold Contig Minimum Scaffold Length Scaffolds Contigs Length Length Coverage All 5,230 15,179 468,595,861 451,945,863 0.96 250 5,230 15,179 468,595,861 451,945,863 0.96 500 5,208 15,157 468,587,033 451,937,035 0.96 1 kb 4,960 14,906 468,393,749 451,744,069 0.96 2 5 kb 4,219 14,133 467,124,713 450,477,966 0.96 5 kb 3,353 13,185 463,919,120 447,297,620 0.96 10 kb 2,695 12,396 459,482,987 442,996,263 0.96 25 kb 2,262 11,718 452,554,079 436,506,169 0.96 50 kb 1,879 10,970 438,381,433 423,130,492 0.97 100 kb 1,338 9,424 399,475,173 386,193,516 0.97 250 kb 575 5,975 274,676,929 266,359,375 0.97 500 kb 184 2,740 139,251,456 135,773,196 0.98 1 Mb 29 663 37,953,728 37,234,106 0.98 2 5 Mb 2 86 5,054,876 4,990,599 0.99 318

Appendix Table 2. Genome assembly statistics of Salicornia europaea final assembly. A = 0.3267 C = 0.1735 G = 0.1735 T = 0.3263 N = 0.0429 GC = 0.347 Main genome scaffold total: 3,996 Main genome contig total: 26,457 Main genome scaffold sequence 530.794 Mb total: Main genome contig sequence 507.999 Mb 4.295% gap total: Main genome scaffold N/L50: 415/367.561 kb Main genome contig N/L50: 3946/39.845 kb Main genome scaffold N/L90: 1467/100.25 kb Main genome contig N/L90: 13152/10.716 kb Max scaffold length: 1.927 Mb Max contig length: 271.153 kb Number of scaffolds > 50 kb: 1,895 % main genome in scaffolds > 50 95.89% kb: Number of Number of Total Scaffold Total Contig Scaffold Contig Minimum Scaffold Length Scaffolds Contigs Length Length Coverage All 3,996 26,457 530,794,157 507,999,138 0.96 250 3,996 26,457 530,794,157 507,999,138 0.96 500 3,734 26,195 530,692,448 507,897,429 0.96 1 kb 3,519 25,978 530,537,905 507,743,043 0.96 2 5 kb 3,161 25,608 529,959,919 507,166,462 0.96 5 kb 2,890 25,307 528,950,753 506,170,356 0.96 10 kb 2,587 24,906 526,799,344 504,164,263 0.96 25 kb 2,223 24,249 520,751,236 498,563,973 0.96 50 kb 1,895 23,333 508,969,031 487,452,956 0.96 100 kb 1,472 21,384 478,215,480 458,501,542 0.96 250 kb 693 14,935 350,131,997 336,575,500 0.96 500 kb 275 8,461 205,782,816 198,278,555 0.96 1 Mb 38 1,835 47,686,111 46,225,487 0.97

319

Appendix Table 3. Protein prediction completeness of Salicornia brachiata genomes using BUSCO v3. Genome (# of proteins) Complete BUSCO's Complete single copy Complete Duplicated Fragmented Missing Platanus Platanus 33k-mer gap closed (30127) 1266 (88 %) 1134 (78.8 %) 134 (9.2 %) 70 (4.9 %) 104 (7.3 %) Platanus 41k-mer gap closed (28595) 1273 (88.4 %) 1149 (79.8 %) 124 (8.6 %) 71 (4.9 %) 96 (6.7 %) Platanus 51k-mer gap closed (28422) 1261 (87.5 %) 1138 (79 %) 123 (8.5 %) 75 (5.2 %) 104 (7.3 %) Platanus 61k-mer gap closed (31703) 1258 (87.3 %) 1144 (79.4 %) 114 (7.9 %) 81 (5.6 %) 101 (7.1 %) Platanus 71k-mer gap closed (32228) 1256 (87.2 %) 1135 (78.8 %) 121 (7.4 %) 78 (5.4 %) 106 (7.4%) SOAPdenovo2 Soapdenovo2 63k-mer gap closed 1241 (86.2 %) 1093 (75.9 %) 148 (10.3 %) 94 (6.5 %) 105 (7.3 %) (31828) Soapdenovo2 65k-mer gap closed 1244 (86.4 %) 1106 (76.8 %) 138 (9.6 %) 90 (6.3 %) 106 (7.3 %) (31987) Soapdenovo2 67k-mer gap closed 1250 (86.8 %) 1105 (76.7 %) 145 (10.1 %) 83 (5.8 %) 107 (7.4 %) (32274) Soapdenovo2 69k-mer gap closed 1240 (86.2 %) 1088 (75.6 %) 152 (10.6 %) 91 (6.3 %) 109 (7.5 %) (32130) Soapdenovo2 73k-mer gap closed 1237 (85.9 %) 1102 (76.5 %) 135 (9.4 %) 104 (7.2 %) 99 (6.9 %) (31901) Soapdenovo2 75k-mer gap closed 1245 (86.5 %) 1103 (76.6 %) 142 (9.9 %) 97 (6.7 %) 98 (6.8 %) (31452) Soapdenovo2 77k-mer gap closed 1237 (85.9 %) 1100 (76.4 %) 137 (9.5 %) 92 (6.4 %) 111 (7.7 %) (31687) Soapdenovo2 79k-mer gap closed 1237 (85.9 %) 1107 (76.9 %) 130 (9 %) 97 (6.7 %) 106 (7.4 %) (31655) Soapdenovo2 81k-mer gap closed 1239 (86 %) 1103 (76.6 %) 136 (9.4 %) 96 (6.7 %) 105 (7.1 %) 320

(32157) Soapdenovo2 83k-mer gap closed 1223 (85 %) 1091 (75.8 %) 132 (9.2 %) 108 (7.5 %) 109 (7.5 %) (32403) ABySS Abyss 61 k-mer gap closed (29570) 1270 (88.2 %) 1127 (78.3 %) 143 (9.9 %) 72 (5 %) 98 (6.8 %) Abyss 69 k-mer gap closed (29555) 1274 (88.4 %) 1141 (79.2 %) 133 (9.2 %) 75 (5.2 %) 91 (6.4 %) Abyss 71 k-mer gap closed (29512) 1276 (88.6 %) 1145 (79.5 %) 131 (9.1 %) 71 (4.9 %) 93 (6.5 %) Abyss 77 k-mer gap closed (29497) 1263 (87.7 %) 1131 (78.5 %) 132 (9.2 %) 78 (5.4 %) 99 (6.9 %) Abyss 79 k-mer gap closed (29092) 1270 (88.2 %) 1138 (79 %) 132 (9.2 %) 76 (5.3 %) 94 (6.5 %) Abyss 81 k-mer gap closed (29524) 1275 (88.5 %) 1140 (79.2 %) 135 (9.4 %) 68 (4.7 %) 97 (6.7 %) Abyss 83 k-mer gap closed (29571) 1287 (89.4 %) 1156 (80.3 %) 131 (9.1 %) 67 (4.7 %) 86 (5.9 %) Abyss k-mer_85 gap closed (29721) 1265 (87.9 %) 1150 (79.9 %) 115 (8 %) 80 (5.6 %) 95 (6.5%) Metassembler assemblies Best_annotation Meta gap closed 1278 (88.8 %) 1147 (79.7 %) 131 (9.1 %) 59 (4.1 %) 103 (7.1 %) (Ab83 Sp75 Pl41) 2_5kb proteins (29853) Best_annotation Meta gap closed 1276 (88.6 %) 1161 (80.6 %) 115 (8 %) 67 (4.7 %) 97 (6.7 %) (Ab83 Sp75 Pl41) 2_8kb proteins (29344) Best_annotation Meta gap closed 1275 (88.6 %) 1153 (80.1 %) 122 (8.5 %) 66 (4.6 %) 99 (6.8 %) (Ab83 Sp75 Pl41) 5_8kb proteins (29548) Best_statistics Meta gap closed (Ab75 1241 (86.2 %) 1125 (78.1 %) 116 (8.1 %) 67 (4.7 %) 132 (9.1 %) Sp71 Pl61) 2_5kb proteins (29429) Best_statistics Meta gap closed (Ab75 1269 (88.1 %) 1146 (79.6 %) 123 (8.5 %) 76 (5.3 %) 95 (6.6 %) Sp71 Pl61) 2_8kb proteins (29612) 321

Best_statistics Meta gap closed (Ab75 1271 (88.3 %) 1152 (80 %) 119 (8.3 %) 79 (5.5 %) 90 (6.2 %) Sp71 Pl61) 5_8kb proteins (28471) Extra Meta gap closed (Ab85 Sp79 1284 (89.1 %) 1161 (80.6 %) 123 (8.5 %) 63 (4.4 %) 93 (6.5 %) Pl33) 2_5kb proteins (29497) Extra Meta gap closed (Ab85 Sp79 1292 (89.7 %) 1169 (81.2 %) 123 (8.5 %) 60 (4.2 %) 88 (6.1 %) Pl33) 2_8kb proteins (29152) Extra Meta gap closed (Ab85 Sp79 1275 (88.5 %) 1144 (79.4 %) 131 (9.1 %) 71 (4.9 %) 94 (6.6 %) Pl33) 5_8kb proteins (29453) Best_statistics 2 Meta gap closed (Ab85 1271 (88.3 %) 1152 (80 %) 119 (8.3 %) 72 (5 %) 97 (6.7 %) Sp77 Pl71) 2_5kb proteins (29363) Best_statistics 2 Meta gap closed (Ab85 1278 (88.8 %) 1160 (80.6 %) 118 (8.2 %) 76 (5.3 %) 86 (5.9 %) Sp77 Pl71) 2_8kb proteins (28711) Best_statistics 2 Meta gap closed (Ab85 1277 (88.7 %) 1169 (81.2 %) 108 (7.5 %) 70 (4.9 %) 93 (6.4 %) Sp77 Pl71) 5_8kb proteins (29218) Abyss alone gap closed (83-85-71 k- 1284 (89.2 %) 1168 (81.1 %) 116 (8.1 %) 65 (4.5 %) 91 (6.3 %) mer) 2_5kb proteins (28891) Abyss alone gap closed (83-85-71 k- 1291 (89.6 %) 1155 (80.2 %) 136 (9.4 %) 64 (4.4 %) 85 (6 %) mer) 2_8kb proteins (28609) Abyss alone gap closed (83-85-71 k- 1286 (89.3 %) 1158 (80.4 %) 128 (8.9 %) 69 (4.8 %) 85 (5.9 %) mer) 5_8kb proteins (28377) Best_combined Meta gap closed 2_5kb 1243 (77.4 %) 1114 (77.4 %) 129 (9 %) 62 (4.3 %) 135 (9.3 %) proteins (28562) Best_combined Meta gap closed 2_8kb 1281 (89 %) 1160 (80.6 %) 121 (8.4 %) 63 (4.4 %) 96 (6.6 %) proteins (28965) Best_combined Meta gap closed 5_8kb 1278 (88.8 %) 1153 (80.1 %) 125 (8.7 %) 64 (4.4 %) 98 (6.8 %) proteins (28934) Best_combined2 Meta gap closed 1290 (89.6 %) 1164 (80.8 %) 126 (8.8 %) 60 (4.2 %) 90 (6.2 %) 2_5kb proteins (28780) 322

Best_combined2 Meta gap closed 1291 (89.7 %) 1159 (80.5 %) 132 (9.2 %) 57 (4 %) 92 (6.3 %) 2_8kb proteins (29541) Best_combined2 Meta gap closed 1282 (89.1 %) 1147 (79.7 %) 135 (9.4 %) 61 (4.2 %) 97 (6.7 %) 5_8kb proteins (29469) Best_combined_abyss Meta gap closed 1317 (91.4 %) 1187 (82.4 %) 130 (9.0 %) 50 (3.5 %) 73 (5.1 %) 2_5kb proteins (32164) Best_combined_abyss Meta gap closed 1315 (91.3 %) 1193 (82.8 %) 122 (8.5 %) 52 (3.6 %) 73 (5.1 %) 2_8kb proteins (31579) Best_combined_abyss Meta gap closed 1318 (91.6 %) 1186 (82.4 %) 132 (9.2 %) 53 (3.7 %) 69 (4.7 %) (not full) 5_8kb proteins (32256) Best_combined_abyss Meta gap closed 1316 (91.4 %) 1187 (82.4 %) 129 (9.0 %) 52 (3.6 %) 72 (5.0 %) (complete) 5_8kb proteins (33134)

Appendix Table 4. Protein prediction completeness of Salicornia europaea genomes using BUSCO v3. Genome (# of proteins) Complete BUSCO's Complete single copy Complete Duplicated Fragmented Missing Platanus Platanus 33 k-mer gap closed 1267 (88 %) 1155 (80.2 %) 112 (7.8 %) 79 (5.5 %) 94 (6.5 %) (27383) Platanus 41 k-mer gap closed 1266 (87.9 %) 1148 (79.7 %) 118 (8.2 %) 70 (4.9 %) 104 (7.2 %) (27624) Platanus 51 k-mer gap closed 1254 (87 %) 1144 (79.4 %) 110 (7.6 %) 78 (5.4 %) 108 (7.6 %) (28019) Platanus 61 k-mer gap closed 1257 (87.3 %) 1148 (79.7 %) 109 (7.6 %) 69 (4.8 %) 114 (7.9 %) (27579) Platanus 71 k-mer gap closed 1241 (86.2 %) 1135 (78.8 %) 106 (7.4 %) 87 (6 %) 112 (7.8 %) (28858) 323

SOAPdenovo2 Soapdenovo2 K-mer_61 gap closed 1270 (88.2 %) 1152 (80 %) 118 (8.2 %) 74 (5.1 %) 96 (6.7 %) (30356) Soapdenovo2 K-mer_63 gap closed 1279 (88.8 %) 1155 (80.2 %) 124 (8.6 %) 68 (4.7 %) 93 (6.5 %) (29949) Soapdenovo2 K-mer_67 gap closed 1286 (89.3 %) 1159 (80.5 %) 127 (8.8 %) 68 (4.7 %) 86 (6.0 %) (30395) Soapdenovo2 K-mer_69 gap closed 1286 (89.3 %) 1155 (80.2 %) 131 (9.1 %) 63 (4.4 %) 91 (6.3 %) (29169) Soapdenovo2 K-mer_71 gap closed 1286 (89.3 %) 1164 (80.8 %) 122 (8.5 %) 62 (4.3 %) 92 (6.3 %) (29558) Soapdenovo2 K-mer_73 gap closed 1278 (88.8 %) 1152 (80 %) 126 (8.8 %) 70 (4.9 %) 92 (6.3 %) (29519) Soapdenovo2 K-mer_77 gap closed 1280 (88.9 %) 1161 (80.6 %) 119 (8.3 %) 67 (4.7 %) 93 (6.4 %) (29839) Soapdenovo2 K-mer_79 gap closed 1286 (89.3 %) 1165 (80.9 %) 121 (8.4 %) 67 (4.7 %) 87 (6.0 %) (29650) ABySS Abyss k-mer_69 gap closed (29811) 1271 (88 3 %) 1152 (80 %) 119 (8.3 %) 74 (5.1 %) 95 (6.6 %) Abyss k-mer_71 gap closed (29622) 1262 (87 6 %) 1138 (79 %) 124 (8.6 %) 79 (5.5 %) 99 (6.9 %) Abyss k-mer_73 gap closed (29616) 1270 (88 2 %) 1163 (80.8 %) 107 (7.4 %) 76 (5.3 %) 94 (6.5 %) Abyss k-mer_75 gap closed (29427) 1271 (88 3 %) 1152 (80 %) 119 (8.3 %) 73 (5.1 %) 96 (6.6 %) Abyss k-mer_77 gap closed (29881) 1268 (88 1 %) 1160 (80.6 %) 108 (7.5 %) 71 (4.9 %) 101 (7 %) Abyss k-mer_79 gap closed (30028) 1263 (87 7 %) 1153 (80.1 %) 110 (7.6 %) 78 (5.4 %) 99 (6.9 %) Abyss k-mer_83 gap closed (29912) 1264 (87 7 %) 1151 (79.9 %) 113 (7.8 %) 79 (5.5 %) 97 (6.8 %) 324

Abyss k-mer_85 gap closed (29721) 1265 (87.9%) 1150 (79.9 %) 115 (8 %) 80 (5.6 %) 95 (6.5 %)

Metassembler

Best_annotation Meta gap closed 1293 (89.8 %) 1164 (80.8 %) 129 (9 %) 61 (4.2 %) 86 (6 %) (Ab73 Sp67 Pl33) 2_5kb proteins (28241) Best_annotation Meta gap closed 1287 (89.4 %) 1170 (81.3 %) 117 (8.1 %) 63 (4.4 %) 90 (6.2 %) (Ab73 Sp67 Pl33) 2_8kb proteins (27711) Best_annotation Meta gap closed 1291 (89.6 %) 1158 (80.4 %) 133 (9.2 %) 61 (4.2 %) 88 (6.2 %) (Ab73 Sp67 Pl33) 5_8kb proteins (28198) Best_statistics Meta gap closed 1285 (89.3 %) 1153 (80.1 %) 132 (9.2 %) 67 (4.7 %) 88 (6 %) (Ab75 Sp71 Pl61) 2_5kb proteins (27207) Best_statistics Meta gap closed 1287 (89.3 %) 1157 (80.3 %) 130 (9 %) 66 (4.6 %) 87 (6.1 %) (Ab75 Sp71 Pl61) 2_8kb proteins (28140) Best_statistics Meta gap closed 1295 (89.9 %) 1168 (81.1 %) 127 (8.8 %) 61 (4.2 %) 84 (5.9 %) (Ab75 Sp71 Pl61) 5_8kb proteins (27959) Extra Meta gap closed (Ab85 Sp79 1293 (89.8 %) 1164 (80.8 %) 129 (9 %) 61 (4.2 %) 86 (6 %) Pl33) 2_5kb proteins (28241) Extra Meta gap closed (Ab85 Sp79 1287 (89.4 %) 1162 (80.7 %) 125 (8.7 %) 64 (4.4 %) 89 (6.2 %) Pl33) 2_8kb proteins (28357) Extra Meta gap closed (Ab85 Sp79 1278 (88.7 %) 1154 (80.1 %) 124 (8.6 %) 68 (4.7 %) 94 (6.6 %) Pl33) 5_8kb proteins (27673) Best_annotation Meta gap closed 1292 (89.8 %) 1173 (81.5 %) 119 (8.3 %) 54 (3.8 %) 94 (6.4 %) (Ab73 Sp67 Pl33) 2_5kb proteins 325

(30120)

Best_annotation Meta gap closed 1292 (89.7 %) 1179 (81.9 %) 113 (7.8 %) 59 (4.1 %) 89 (6.2 %) (Ab73 Sp67 Pl33) 2_8kb proteins (30332) Best_annotation Meta gap closed 1294 (89.8 %) 1171 (81.3 %) 123 (8.5 %) 63 (4.4 %) 83 (5.8 %) (Ab73 Sp67 Pl33) 5_8kb proteins (29692) Best_statistics Meta gap closed 1290 (89.6 %) 1168 (81.1 %) 122 (8.5 %) 69 (4.8 %) 81 (5.6 %) (Ab75 Sp71 Pl61) 2_5kb proteins (30225) Best_statistics Meta gap closed 1300 (90.3 %) 1170 (81.3 %) 130 (9 %) 58 (4 %) 82 (5.7 %) (Ab75 Sp71 Pl61) 2_8kb proteins (30209) Best_statistics Meta gap closed 1303 (90.4 %) 1180 (81.9 %) 123 (8.5 %) 58 (4 %) 79 (5.6 %) (Ab75 Sp71 Pl61) 5_8kb proteins (28850) Extra Meta gap closed (Ab85 Sp79 1247 (86.6 %) 1124 (78.1 %) 123 (8.5 %) 70 (4.9 %) 123 (8.5 %) Pl33) 2_5kb proteins (30233) Extra Meta gap closed (Ab85 Sp79 1282 (89.1 %) 1153 (80.1 %) 129 (9 %) 65 (4.5 %) 93 (6.4 %) Pl33) 2_8kb proteins (29622) Extra Meta gap closed (Ab85 Sp79 1280 (88.9 %) 1149 (79.8 %) 131 (9.1 %) 61 (4.2 %) 99 (6.9 %) Pl33) 5_8kb proteins (29803) Combined Meta gap closed (Annot, 1267 (88.0 %) 1152 (80.0 %) 115 (8.0 %) 55 (3.8 %) 118 (8.2 %) stats, extra) 2_5kb proteins (30631) Combined Meta gap closed (Annot, 1300 (90.3 %) 1173 (81.5 %) 127 (8.8 %) 58 (4 %) 82 (5.7 %) stats, extra) 2_8kb proteins (30165) Combined Meta gap closed (Annot, 1307 (90.8 %) 1183 (82.2 %) 124 (8.6 %) 54 (3.8 %) 79 (5.4 %) stats, extra) 5_8kb proteins (29086) 326

Appendix Table 5. Matrix of the number of differentially expressed genes between treatments. 1wk 0mM 1wk 200mM 1wk 50mM 1wk 600mM 6wk 0mM 6wk 200mM 6wk 50mM 6wk 600mM 1wk 0mM 0 133 243 1786 49 575 289 436 1wk 200mM 133 0 4 345 247 223 38 132 1wk 50mM 243 4 0 403 553 368 35 130 1wk 600mM 1786 345 403 0 1797 1288 508 1099 6wk 0mM 49 247 553 1797 0 222 320 319 6wk 200mM 575 223 368 1288 222 0 100 5 6wk 50mM 289 38 35 508 320 100 0 37 6wk 600mM 436 132 130 1099 319 5 37 0 327

Appendix Table 6. Primers used to amplify the selected genes for characterization.

Transcriptome, gene|CDS Forward Tm (°C) Reverse Tm (°C) Product Stop ID, alias length (bp) codon

S. bigelovii shoots ATGGCAGAAATGAATGCTAC 59.6 CTACACTTTAGAAGTCGATTGC 57.5 2406 + TRINITY_DN85516_c0_g1 AA TG _i4|m.51400 (CHX4)

CACTTTAGAAGTCGATTGCTGT 60.3 2403 - TG

S. bigelovii shoots ATGTTGTCACAATTGAGCTCT 58.0 CTATGTTCTGTCTAGCAAATTGT 59.4 1683 + TRINITY_DN91644_c0_g1 CT CG _i4|m.225611 (NHX2)

TGTTCTGTCTAGCAAATTGTCG 58.1 1680 -

S. bigelovii shoots ATGGCAGCATCTCGAATTGA 61.7 TCAAGGAGCTTGGCGGA 61.7 3480 + TRINITY_DN95767_c2_g1 _i2|m.374937 (SOS1)

ATGGCAGCATCTCGAATTG 59.8 AGGAGCTTGGCGGAAAGT 60.0 3477 -

TCAAGGAGCTTGGCGGAAA 64.2 3480 + 328

S. bigelovii shoots CACCATGGCAGCATCTCGAA 70.4 n/a TRINITY_DN95767_c2_g1 TTGA _i2|m.374937 (SOS1) Directional

S. bigelovii shoots CACCATGGCAGCATCTCGAA 69.0 n/a TRINITY_DN95767_c2_g1 TTG _i2|m.374937 (SOS1) Directional

S. bigelovii shoots CTGATGTATCTGGGGTCTTGA 57.6 TGAGCATTCTCATATCTTGCA 57.0 1844 n/a TRINITY_DN95767_c2_g1 _i2|m.374937 (SOS1) middle primers

S. bigelovii shoots CGCTGAGAGGTCTACCTTGAAT 59.9 1904 n/a TRINITY_DN95767_c2_g1 _i2|m.374937 (SOS1) middle primers

S. bigelovii shoots ATGTTTGGAGTAGCAAAGAT 56.8 CTAATGCTCATGAATTGGATCT 58.7 1608 + TRINITY_DN96584_c0_g2 AGTG G _i6|m.351156 (NHX1)

ATGCTCATGAATTGGATCTG 56.0 1605 - 329

S. bigelovii shoots + roots ATGCAACTCCAACAAGACC 56.0 TTAGAGAGAAAGTTTCCAAGCT 54.5 1614 + transcriptome TRINITY_DN234792_c3_g 1_i1|m.1072578 (HKT1)

ATGCAACTCCAACAAGACCT 57.2 GAGAGAAAGTTTCCAAGCTTTG 57.4 1611 -

S. bigelovii shoots + roots ATGGCTGTAAACATTACAAA 59.7 TTAAACTTGGTTGGCAGCATCT 61.0 2571 + transcriptome CATCA TRINITY_DN230280_c0_g 1_i3|m.602759 (CHX20)

AACTTGGTTGGCAGCATCTTG 62.5 2568 -

S. bigelovii shoots + roots ATGGCTCATAACGGTACTTCT 57.0 TTATTCTGTTTCAGTCCCATCA 57.2 2403 + transcriptome T TRINITY_DN239295_c1_g 1_i2|m.337270 (CHX18)

TTCTGTTTCAGTCCCATCAAC 57.6 2400 -

S. bigelovii shoots + roots ATGGCAGCAGCGATTGAA 61.5 TCAATTACGACTACTGTAGACT 55.4 1593 + transcriptome TCTC TRINITY_DN242112_c3_g 1_i10|m.568547 (NHX6) 330

TCAATTACGACTACTGTAGACT 59.1 1593 + TCTCTTC

ATTACGACTACTGTAGACTTCT 61.2 1590 - CTTCTTGG

ATTACGACTACTGTAGACTTCT 56.2 1590 - CTTCTT

S. bigelovii shoots + roots ATGGGTTTAGAATCAGATGA 58.7 TTAGTAGCCTGAATCTGCAGTT 57.4 1563 + transcriptome AGC T TRINITY_DN231579_c6_g 1_i1|m.1049335 (NHX3)

GTAGCCTGAATCTGCAGTTTG 57.6 1560 -

S. bigelovii shoots ATGGGTTCTACTGAGGTATGG 59.2 TTAAGTCCTCATCCAACGAGA 57.4 1338 + TRINITY_DN78708_c0_g1 G _i9|m.435462 (CAX3)

AGTCCTCATCCAACGAGATTTG 60.5 1335 -

S. bigelovii shoots + roots ATGGAGTCAAGTAATACACA 55.9 TCAGATGACAATGTTGATGGA 57.4 1365 + transcriptome AGAAG TRINITY_DN237578_c1_g 3_i5|m.878895 (CAX1) 331

GATGACAATGTTGATGGATGT 55.6 1362 -

S. bigelovii shoots + roots ATGTATTGTATCTGGTGTAAT 55.9 TTACTCTTCTTTTTTGTCATTATG 54.7 1395 + TRINITY_DN93112_c3_g2 TGTGT TAC _i2|m.270346 (CAX2)

CTCTTCTTTTTTGTCATTATGTA 54.8 1392 - CA

S. bigelovii shoots ATGAAAATCCTAAAAATCCC 55.4 TTAGGTGGGGATGAGGAT 55.1 1788 + TRINITY_DN97007_c1_g1 TAA _i9|m.336857 (NCL1)

GGTGGGGATGAGGATGT 56.3 1785 -

ATGAAAATCCTAAAAATCCC 57.8 GGTGGGGATGAGGATGTG 59.7 1785 - TAAAA

S. bigelovii shoots ATGGATCCACCTAATGAGAT 56.4 TTAATGCTTATGCGGCTTTA 56.4 2343 + TRINITY_DN95140_c1_g1 G _i1|m.732722 (CLC-C1)

ATGCTTATGCGGCTTTATCT 56.7 2340 -

S. bigelovii shoots ATGAATGATCATGAAAATGA 57.0 TTACTTCACTGATTCTGCAGGA 57.6 1983 + TRINITY_DN92389_c1_g1 ACA _i2|m.242575 (CACLC) 332

CTTCACTGATTCTGCAGGATT 57.0 1980 -

S. bigelovii shoots ATGGATAAAGAAAGAGAGG 55.5 CTACTTGTGGATTTCGAGGTG 57.3 2349 + TRINITY_DN92475_c0_g2 AAAA _i1|m.245250 (CLC-C2)

CTTGTGGATTTCGAGGTG 54.9 2346 -

S. bigelovii shoots + roots ATGGAAGCAACTGTAGAATC 56.0 TCATTTCTGTTTTTTGGATCTTG 58.3 2409 + transcriptome AC TRINITY_DN241560_c2_g 2_i1|m.885324 (CLC-B)

TTTCTGTTTTTTGGATCTTGC 57.0 2406 -

S. bigelovii shoots ATGTTGTCAAACCATTTGCA 57.4 CTATATATTATTTTGCACCAGA 57.9 2382 + TRINITY_DN97770_c0_g7 AGACC _i5|m.316941 (CLC-D)

TATATTATTTTGCACCAGAAGA 56.6 2379 - CC

S. brachiata shoots + roots ATGAATGTTGATACCGGCAC 62.4 CTACGTGAACAAGGTAACCAC 60.7 2784 + TRINITY_DN52191|c1_g3 AA ATCT _i3|m.178547 (CCC1) 333

ATGAATGTTGATACCGGCAC 58.3 CTACGTGAACAAGGTAACCAC 57.8 2784 + A

CGTGAACAAGGTAACCACATC 58.0 2781 -

CGTGAACAAGGTAACCACATC 60.3 2781 - TT

(SOS1) Arabidopsis ATGACGACTGTAATCGACGC 58.8 TCATAGATCGTTCCTGAAAACG 59.2 3441 +

TAGATCGTTCCTGAAAACGATT 59.2 3438 - T

(SOS1) Arabidopsis CACCATGACGACTGTAATCG 67.9 n/a directional ACGC

(SOS1) Arabidopsis check GCATGTTTTATGCTGCATTTG 59.2 GAACCAATGCACTTTCCTGC 60.6 1707 n/a in the middle

S. bigelovii shoots ATGGGTATTGCAGAATCAGG 59.4 CTACGCCAAGGTGTACTGTGA 58.9 1917 + TRINITY_DN87897_c0_g2 A _i6|m.147698 (CatAm)

CGCCAAGGTGTACTGTGAGC 61.9 1914 -

S. bigelovii shoots ATGGATGTTCAACGAACATCT 59.9 TCAATCTATTTTCACAGAAGAG 60.0 1914 + TRINITY_DN93623_c0_g2 G TAAATGA _i2|m.675398 (Endo9) 334

ATCTATTTTCACAGAAGAGTAA 59.9 1911 - ATGAGC

S. bigelovii shoots ATGACTGATTCATCGCCGTT 60.5 CTAATGGGTTGCCCAGAAGC 61.9 1572 + TRINITY_DN94883_c1_g1 _i4|m.718597 (Carnit3)

ATGGGTTGCCCAGAAGCT 60.5 1569 -

S. bigelovii shoots ATGGGAGATGATGGAACCC 59.5 TCAGAAACCAACTGTTTTTCCA 59.6 1323 + TRINITY_DN87967_c1_g1 _i3|m.151326 (DUF)

GAAACCAACTGTTTTTCCAAGT 57.3 1320 -

S. bigelovii shoots ATGTCAACAAGTTTGCAGCTT 59.0 TTAAACATCATAGACAGAAGT 59.6 2127 + TRINITY_DN96894_c5_g1 T GAAATCA _i3|m.688165 (GatedN)

AACATCATAGACAGAAGTGAA 57.0 2124 - ATCA

S. bigelovii shoots ATGCCCCTGACCAGTCAAG 61.1 CTACGAGGACGAAAGTTTCATC 61.5 1236 + TRINITY_DN92807_c0_g1 TG _i1|m.642934 (Unkn)

CGAGGACGAAAGTTTCATCTG 59.9 1233 - 335

S. bigelovii shoots ATGGGTGGTTCCTTGGAACT 60.6 CTAGTGTATAGGGCAACCATCA 58.2 2259 + TRINITY_DN87554_c1_g1 A _i3|m.143622 (Ol_pep_trans)

GTGTATAGGGCAACCATCAAC 60.5 2256 - C

S. bigelovii shoots ATGTGTTCTGCAAGAAACAG 60.7 TCAGCACCCCACAAGGTC 60.7 1080 + TRINITY_DN89547_c1_g1 AGG _i3|m.184089 (HPP1)

GCACCCCACAAGGTCTCTGT 62.5 1077 -

S. bigelovii shoots ATGCCGGCAGTAGGAATTG 61.0 TCAATCAACATTCTTTCCATATT 58.5 1575 + TRINITY_DN90119_c0_g1 TATT _i1|m.561325 Sugar transporter, modified protein based on alignment (Sug_trans)

ATCAACATTCTTTCCATATTTAT 59.8 1572 - TGTTG 336

S. brachiata shoots + roots ATGGCGTCAATGGCTCTG 60.8 TCACCATAATCCAATGGCCT 60.2 1683 + TR94368|c4_g2_i2|m.6445 20 (Na_Sulf)

CCATAATCCAATGGCCTTCC 61.4 1680 -

S. bigelovii shoots ATGGTTTCTTCTCTCCAAATT 60.8 TTATGAATAGGAAACAATGAG 60.1 705 + TRINITY_DN99928_c1_g1 GC AGGTTC _i4|m.260967 SWEET (Sweet)

TGAATAGGAAACAATGAGAGG 58.7 702 - TTC

S. bigelovii shoots ATGGACGAAGAGAAGGCG 58.9 TTAAGATTTCAACCTAGGAACA 60.3 2082 + TRINITY_DN94257_c0_g2 ACTTG _i1|m.315705 (Ol_nuc_trans)

AGATTTCAACCTAGGAACAACT 59.6 2079 - TGA

S. bigelovii shoots ATGGTGACGCAGAAGCAAG 60.0 TCAGTGTTGGGTCGAATTTG 59.5 1470 + TRINITY_DN91452_c0_g1 _i1|m.218596 (Na_amin_aux) 337

GTGTTGGGTCGAATTTGGTG 61.2 1467 -

S. bigelovii shoots + roots ATGTGTAAACAAAGGGGCTC 59.5 TCAAACAAGTAAAGGTTGATT 57.6 1692 + transcriptome A AATATTATT TRINITY_DN238619_c3_g 1_i1|m.305201 (Mate)

Really low tm... let's see AACAAGTAAAGGTTGATTAAT 50.7 1689 - ATT

S. bigelovii shoots ATGGAAGAAACTTTTTTCCCC 59.4 TTAACTTGAAGATGATGCATCA 58.8 2163 + TRINITY_DN100001_c0_g T AA 1_i8|m.757753 (Bor2)

ACTTGAAGATGATGCATCAAA 59.1 2160 - ATT

S. bigelovii shoots + roots ATGGAGGTAAATCAAGGAAG 60.7 TCATCTGGCACAAGTAAGCC 58.9 2748 + transcriptome CC TRINITY_DN240904_c0_g 1_i1|m.1026992 (Mec_chan)

TCTGGCACAAGTAAGCCAGTT 59.9 2745 - 338

S. brachiata shoots + roots ATGCTATGTATTAGAATGACG 58.5 TCAGAACTCAAAGTCGTCGTG 59.1 >2kb? + for 5' AGTAGTG TR93735|c9_g2_i3|m.5873 69 / S. bigelovii shoots for 3’ TRINITY_DN90306_c5_g2 _i1|m.199236 (GatedN2)

GAACTCAAAGTCGTCGTGATTT 58.4 -

S. bigelovii shoots ATGAAGACGAAGACGATAAA 60.2 TTAACCACTATGTGCAACGATG 59.5 1017 + TRINITY_DN93438_c2_g1 CAGAG T _i1|m.282234 (PQ)

ACCACTATGTGCAACGATGTGA 61.4 1014 -

S. bigelovii shoots ATGGCGGATGAAGATGGC 62.0 TCAAACCATGTATGTCATGCC 59.3 2538 + TRINITY_DN93055_c1_g1 _i2|m.265746 (POT7)

AACCATGTATGTCATGCCAA 57.2 2535 -

S. bigelovii shoots ATGGAAAAGGAGGAAATGG 59.8 TCAAGCCTGTTCGACAACTA 57.1 1944 + TRINITY_DN93228_c1_g4 AA _i2|m.272438 (Sulf1_3) 339

AGCCTGTTCGACAACTACTAGG 61.0 1941 - G

S. bigelovii shoots ATGGTTACAACTACACCAAC 57.7 TTAAGCAGGTTCAGAAGTTGCA 60.1 1743 + TRINITY_DN84661_c0_g1 TGAA _i1|m.498853 (CatAm-5)

AGCAGGTTCAGAAGTTGCAA 58.6 1740 -

S. bigelovii shoots ATGAGGAAGAATCAAAAAG 58.0 TCAAACTTCATTTTCAATTAGC 59.3 636 + TRINITY_DN87160_c0_g1 AAGG AAA _i7|m.535190 (Pr-Unkn)

AACTTCATTTTCAATTAGCAAA 58.2 633 - GC

S. bigelovii shoots ATGGCAGAGATAAAAGAGCC 59.7 TTACAACGGCATAAGCCTTTT 58.8 1320 + TRINITY_DN88171_c0_g3 C _i1|m.542960 (Mg2)

CAACGGCATAAGCCTTTTGT 60.1 1317 -

S. bigelovii shoots ATGGGAGTCGAAGGAAACAC 59.0 TCAGAAGACCCAAAACCAAGA 59.7 1236 + TRINITY_DN92158_c0_g1 _i1|m.615041 (Eq_trans)

GAAGACCCAAAACCAAGACA 57.6 1233 - 340

S. bigelovii shoots ATGTCATCACCACCACCGC 62.9 TTAGGCAGTAGTGGGCCCA 61.6 1524 + TRINITY_DN93502_c4_g1 _i1|m.670595 (Amm_trans)

GGCAGTAGTGGGCCCACT 61.1 1521 -

S. bigelovii shoots ATGGGGAAGTTTTTGAGAGG 58.1 TTAGTAGCCACCACCTCCAT 57.1 993 + TRINITY_DN86456_c0_g1 _i2|m.126405 (W_Unkn)

GTAGCCACCACCTCCATATCC 60.6 -

S. bigelovii shoots ATGGGGTTTGAGAAGAAGAG 59.7 TCAAGCTTTGGAGCTGAAAA 58.8 1278 + TRINITY_DN94204_c0_g2 C _i3|m.315384 (Prot_am_trans)

AGCTTTGGAGCTGAAAATCTC 57.8 1275 -

S. bigelovii shoots ATGGCTGCCACACTCTCC 59.8 TCAAGGCCAGTTAGTGAATTGA 59.7 825 + TRINITY_DN85424_c1_g1 _i1|m.511375 (Salt_unkwn) 341

AGGCCAGTTAGTGAATTGAAG 59.8 822 - C

S. bigelovii shoots ATGGCTGCTGAAGTTCCTGT 59.9 TTACAATTCTCCCTTACTAGCAT 58.5 2583 + TRINITY_DN83233_c5_g1 CA _i1|m.97557 (An_Unkn)

CAATTCTCCCTTACTAGCATCA 59.3 2580 - GTT

S. bigelovii shoots + roots ATGGCAACGACAAATCCTTT 59.4 TCAGTTCGCACCCAAGC 59.5 1074 + transcriptome TRINITY_DN237891_c1_g 1_i1|m.610336 (Salt_tol_2)

GTTCGCACCCAAGCACG 62.5 1071 -

S. bigelovii shoots ATGCCTTATTCTTTTATTGCTG 58.9 TCAATTATGTCTTGTTTTCCTTG 59.4 4578 + TRINITY_DN99969_c0_g2 C G _i3|m.262656 (ABC9)

ATTATGTCTTGTTTTCCTTGGAT 58.5 4575 - G 342

S. bigelovii shoots ATGCAGGTGGTAAATGCAGT 58.1 TCATTTAGTTGAAGCAGGAACA 59.7 1458 + TRINITY_DN98804_c3_g3 TC _i1|m.284214 (DUF21Pr)

TTTAGTTGAAGCAGGAACATCA 59.7 1455 - TC

S. bigelovii shoots ATGGAGAAAAAGCTACTGAA 58.6 CTAGACCTGCTCTGGACCAGA 59.6 3075 + TRINITY_DN98123_c1_g1 GGA _i1|m.328100 (Ca_ATPase)

GACCTGCTCTGGACCAGAAG 60.0 3072 -

S. bigelovii shoots ATGAGTGCACCGGTATTAGTT 59 TCAATTGTTCTTGTCATCAGTAG 58.3 2211 + TRINITY_DN94604_c1_g1 G G _i7|m.331923 (Mon_Sac2)

ATTGTTCTTGTCATCAGTAGGG 59.5 2208 - TG

S. bigelovii shoots ATGTCTTCAAATTCTTGGATA 60.6 TTACAGACCAGATGAGCGATTT 58.9 4506 + TRINITY_DN99130_c1_g2 ATCTCA _i1|m.586789 (ABC4)

CAGACCAGATGAGCGATTTG 59.4 4503 - 343

S. bigelovii shoots + roots ATGGCTTTCAAGCCATTGGT 61.8 CTACAGATCCTCTCGACTTTCC 59.9 4908 + TRINITY_DN241851_c1_g A 2_i5|m.1180310 (ABC2)

CAGATCCTCTCGACTTTCCA 57.9 4905 -

S. bigelovii shoots + roots ATGTTAGTCACAAAGCATTTG 60 TTAAGTACGCGGTTTTGGTTTT 59.9 1860 + TRINITY_DN236148_c2_g ATCA 2_i2|m.945373 (MCH1)

AGTACGCGGTTTTGGTTTTC 59.1 1857 -

S. brachiata shoots + roots ATGACGACTGTGGGTAATAC 61 TCAACTAGTTTCGGAATTCTTCT 60.2 1389 + TR54513|c9_g3_i7|m.3804 CG TG 53 (NaCNeu3)

ACTAGTTTCGGAATTCTTCTTGA 57.8 1386 - A

S. bigelovii shoots ATGGTGGTAAAGGCGACTC 57.6 TCAAATAGGAGTATAGCTATGG 58.4 699 + TRINITY_DN89335_c3_g1 TCATC _i2|m.555405 (Punch)

AATAGGAGTATAGCTATGGTCA 59.5 696 - TCAGTCT 344

S. bigelovii shoots ATGTCACTTTCTTGTACAAAA 57.5 TATGAAAAATAGATGGTCCCC 58.3 2160 - TRINITY_DN99839_c2_g1 ACG A _i2|m.576774 (KAT2)

S. brachiata genome CACCatgacgaaggacatggaagc 64.0 GCATTGCTTCTGAATGATCCAA 62.4 870 - g39524.t1 (Pip2_1)

S. bigelovii shoots CACCATGTCGAAGGAAGTGA 59.2 ATTGGTGGGGTTGCTTCG 61.9 846 - TRINITY_DN95605_c0_g2 TGAGTG _i8|m.750781 (PIP3)

S. bigelovii shoots CACCATGGCTATTGCTTTTGG 58.8 GTATTCATTGGTGAGAGGGACA 58.9 741 - TRINITY_DN96163_c7_g4 AAGA _i3|m.724639 (TIP2_1)

S. bigelovii shoots + roots CACCATGCCGATCAACAGAA 57.8 AGCAGAGAGTCTTTGGTAGTCA 59.2 762 - TRINITY_DN229280_c1_g TAGC GTT 1_i1|m.682218 (TIP 1_3)

. 345

Appendix Table 7. Differentially abundant proteins in tonoplast enriched fraction. Protein description Accession Quantitative Profile 0 50 200 600 Number mM mM mM mM

VPS18_DANRE RecName: Full=Vacuolar protein sorting- Old_SABI_m. 0 mM high, 200 mM high, 11 9 8 3 associated protein 18 homolog; E:7e-174 len:988 type:complete 307860 50 mM high, 600 mM low PTR9_ARATH RecName: Full=Protein NRT1/ PTR FAMILY Old_SABI_m. 0 mM high, 200 mM high, 10 12 7 2 5.10; E:8e-152 len:417 type:3prime_partial 531516 50 mM high, 600 mM low VHAA2_ARATH RecName: Full=V-type proton ATPase Old_SABI_m. 0 mM high, 200 mM high, 4 3 4 0 subunit a2; E:0 len:817 type:complete 608248 50 mM high, 600 mM low PTR2_ARATH RecName: Full=Protein NRT1/ PTR FAMILY New_SABI_ 0 mM high, 200 mM high, 43 46 47 31 8.3; E:0 len:584 type:complete m.185598 50 mM high, 600 mM low VA714_ARATH RecName: Full=Vesicle-associated membrane Old_SABR_m 0 mM high, 200 mM low, 93 83 56 41 protein 714; E:7e-139 len:221 type:complete .782086 50 mM high, 600 mM low ERDL6_ARATH RecName: Full=Sugar transporter ERD6-like Old_SAEU_m 0 mM high, 200 mM low, 32 30 14 14 6; E:0 len:485 type:complete .159197 50 mM high, 600 mM low ERDL5_ARATH RecName: Full=Sugar transporter ERD6-like Old_SAEU_m 0 mM high, 200 mM low, 18 16 10 7 5; E:0 len:475 type:complete .251789 50 mM high, 600 mM low OE64C_ARATH RecName: Full=Outer envelope protein 64, Old_SAEU_m 0 mM high, 200 mM low, 57 58 16 23 chloroplastic; E:0 len:591 type:complete .602661 50 mM high, 600 mM low VPS41_SOLLC RecName: Full=Vacuolar protein sorting- Old_SAEU_m 0 mM high, 200 mM low, 60 54 44 32 associated protein 41 homolog; E:0 len:958 type:complete .622345 50 mM high, 600 mM low 346

VHAA3_ARATH RecName: Full=V-type proton ATPase Old_SAEU_m 0 mM high, 200 mM low, 969 908 766 706 subunit a3; E:0 len:846 type:complete .86626 50 mM high, 600 mM low ZFY26_MOUSE RecName: Full=Zinc finger FYVE domain- New_SABI_ 0 mM high, 200 mM low, 20 20 10 6 containing protein 26; E:2e-25 len:2514 type:complete m.762650 50 mM high, 600 mM low CAAT9_ARATH RecName: Full=Cationic amino acid New_SABI_ 0 mM high, 200 mM low, 50 48 37 35 transporter 9, chloroplastic; E:0 len:574 type:complete m.764525 50 mM high, 600 mM low PDI52_ARATH RecName: Full=Protein disulfide-isomerase 5- Old_SABI_m. 0 mM high, 200 mM low, 17 14 6 3 2; E:3e-162 len:374 type:5prime_partial 1071921 50 mM high, 600 mM low AB2C_ARATH RecName: Full=ABC transporter C family Old_SABI_m. 0 mM high, 200 mM low, 768 708 570 378 member 2; E:0 len:1636 type:complete 1180310 50 mM high, 600 mM low AB14C_ARATH RecName: Full=ABC transporter C family Old_SABI_m. 0 mM high, 200 mM low, 13 9 1 0 member 14; E:2e-100 len:416 type:internal 432745 50 mM high, 600 mM low OP24A_ARATH RecName: Full=Outer envelope pore protein Old_SABI_m. 0 mM high, 200 mM low, 53 47 12 15 24A, chloroplastic; E:3e-98 len:242 type:complete 746874 50 mM high, 600 mM low VA711_ARATH RecName: Full=Vesicle-associated membrane Old_SABI_m. 0 mM high, 200 mM low, 17 17 15 10 protein 711; E:8e-129 len:220 type:complete 890941 50 mM high, 600 mM low MCH1_YARLI RecName: Full=Probable transporter MCH1; Old_SABI_m. 0 mM high, 200 mM low, 104 96 83 62 E:7e-06 len:620 type:complete 945373 50 mM high, 600 mM low PLST4_ARATH RecName: Full=Plastidic glucose transporter Old_SABI_m. 0 mM high, 200 mM low, 42 56 17 18 4; E:0 len:553 type:complete 946918 50 mM high, 600 mM low TLDC1_DANRE RecName: Full=TLD domain-containing Old_SABR_m 0 mM high, 200 mM low, 32 29 15 9 protein 1; E:3e-08 len:537 type:complete .212929 50 mM high, 600 mM low 347

SSUH2_HUMAN RecName: Full=Protein SSUH2 homolog; Old_SABR_m 0 mM high, 200 mM low, 71 57 39 35 E:5e-28 len:427 type:complete .491770 50 mM high, 600 mM low VCL1_ARATH RecName: Full=Protein VACUOLELESS1; E:0 Old_SABR_m 0 mM high, 200 mM low, 45 36 28 15 len:844 type:complete .521928 50 mM high, 600 mM low A0A068UJ45_COFCA SubName: Full=Coffea canephora Old_SABR_m 0 mM high, 200 mM low, 75 59 37 35 DH200=94 genomic scaffold, scaffold_31 .563458 50 mM high, 600 mM low ECO:0000313|EMBL:CDP08204.1; E:0 len:1460 type:complete LAL5_ARATH RecName: Full=MATE efflux family protein New_SABI_ 0 mM high, 200 mM low, 5 2 1 1 LAL5; E:1e-126 len:308 type:3prime_partial m.313562 50 mM high, 600 mM low ACA11_ARATH RecName: Full=Putative calcium- New_SABI_ 0 mM high, 200 mM low, 101 101 43 32 transporting ATPase 11, plasma membrane-type; E:0 len:1025 m.328100 50 mM high, 600 mM low type:complete MSSP2_ARATH RecName: Full=Monosaccharide-sensing New_SABI_ 0 mM high, 200 mM low, 363 347 283 236 protein 2; E:0 len:737 type:complete m.331923 50 mM high, 600 mM low NPC1_PIG RecName: Full=Niemann-Pick C1 protein; E:0 New_SABI_ 0 mM high, 200 mM low, 448 368 273 230 len:1279 type:complete m.360789 50 mM high, 600 mM low SYP52_ARATH RecName: Full=Syntaxin-52; E:2e-91 len:232 New_SABI_ 0 mM high, 200 mM low, 12 10 7 5 type:complete m.495821 50 mM high, 600 mM low TLC1_SOLTU RecName: Full=Plastidic ATP/ADP- New_SABI_ 0 mM high, 200 mM low, 15 21 3 1 transporter; E:0 len:627 type:complete m.571274 50 mM high, 600 mM low AB4C_ARATH RecName: Full=ABC transporter C family New_SABI_ 0 mM high, 200 mM low, 149 105 59 32 member 4; E:0 len:1502 type:complete m.586789 50 mM high, 600 mM low 348

I1M1G2_SOYBN SubName: Full=Uncharacterized protein New_SABI_ 0 mM high, 200 mM low, 29 24 17 12 ECO:0000313|EnsemblPlants:GLYMA13G29020.1; E:4e-26 m.736056 50 mM high, 600 mM low len:93 type:5prime_partial DIT1_SPIOL RecName: Full=Dicarboxylate transporter 1, New_SABI_ 0 mM high, 200 mM low, 78 83 40 44 chloroplastic; E:0 len:544 type:5prime_partial m.739420 50 mM high, 600 mM low B9I494_POPTR SubName: Full=Uncharacterized protein Ma_SAEU_m 0 mM high, 200 mM low, 85 67 44 39 ECO:0000313|EMBL:EEE96442.2; E:4e-106 len:233 .121011 50 mM high, 600 mM low type:complete K4BTJ3_SOLLC SubName: Full=Uncharacterized protein Ma_SAEU_m 0 mM high, 200 mM low, 164 148 117 104 ECO:0000313|EnsemblPlants:Solyc04g072060.2.1; E:2e-139 .131608 50 mM high, 600 mM low len:206 type:complete C4MN96_9CARY SubName: Full=Peroxisomal ascorbate Ma_SAEU_m 0 mM high, 200 mM low, 85 98 40 38 peroxidase ECO:0000313|EMBL:ACI15209.1; E:0 len:288 .77536 50 mM high, 600 mM low type:complete XYLL3_ARATH RecName: Full=D-xylose-proton symporter- New_SABI_ 0 mM high, 200 mM low, 97 106 43 45 like 3, chloroplastic; E:0 len:575 type:complete m.201592 50 mM high, 600 mM low PDI54_ORYSJ RecName: Full=Protein disulfide isomerase-like New_SABI_ 0 mM high, 200 mM low, 14 14 2 3 5-4; E:0 len:483 type:complete m.218101 50 mM high, 600 mM low GLPT4_ARATH RecName: Full=Putative glycerol-3- New_SABI_ 0 mM high, 200 mM low, 65 44 24 19 phosphate transporter 4; E:0 len:530 type:complete m.240225 50 mM high, 600 mM low AB9C_ARATH RecName: Full=ABC transporter C family New_SABI_ 0 mM high, 200 mM low, 94 60 32 14 member 9; E:0 len:1526 type:complete m.262656 50 mM high, 600 mM low 349

Y4424_ARATH RecName: Full=DUF21 domain-containing New_SABI_ 0 mM high, 200 mM low, 304 273 203 152 protein At4g14240; E:0 len:486 type:complete m.284214 50 mM high, 600 mM low CML27_ARATH RecName: Full=Probable calcium-binding Old_SAEU_m 0 mM high, 200 mM low, 18 9 6 4 protein CML27; E:6e-52 len:174 type:complete .264410 50 mM low, 600 mM low AB5C_ARATH RecName: Full=ABC transporter C family Old_SAEU_m 0 mM high, 200 mM low, 9 7 5 3 member 5; E:0 len:1517 type:complete .417000 50 mM low, 600 mM low VP13A_DICDI RecName: Full=Putative vacuolar protein New_SABI_ 0 mM high, 200 mM low, 259 150 85 76 sorting-associated protein 13A; E:2e-55 len:4317 type:complete m.789925 50 mM low, 600 mM low POT7_ARATH RecName: Full=Potassium transporter 7; E:0 Old_SABI_m. 0 mM high, 200 mM low, 85 50 23 14 len:846 type:complete 1085834 50 mM low, 600 mM low VATG_CITLI RecName: Full=V-type proton ATPase subunit Old_SABI_m. 0 mM high, 200 mM low, 26 20 19 16 G; E:3e-47 len:111 type:complete 246243 50 mM low, 600 mM low SYP52_ARATH RecName: Full=Syntaxin-52; E:1e-111 len:235 Old_SABI_m. 0 mM high, 200 mM low, 89 65 46 37 type:complete 667730 50 mM low, 600 mM low RB1BV_BETVU RecName: Full=Ras-related protein RAB1BV; Old_SABR_m 0 mM high, 200 mM low, 17 16 12 12 E:7e-156 len:216 type:complete .645915 50 mM low, 600 mM low D4QD79_DIACA SubName: Full=Sodium/calcium exchanger New_SABI_ 0 mM high, 200 mM low, 191 163 141 136 protein ECO:0000313|EMBL:BAI94501.1; E:0 len:431 m.336829 50 mM low, 600 mM low type:internal MRS21_ARATH RecName: Full=Magnesium transporter New_SABI_ 0 mM high, 200 mM low, 39 25 18 10 MRS2-1; E:0 len:440 type:complete m.542960 50 mM low, 600 mM low 350

GRXC1_ARATH RecName: Full=Glutaredoxin-C1; E:3e-46 New_SABI_ 0 mM high, 200 mM low, 4 3 3 2 len:124 type:complete m.67234 50 mM low, 600 mM low W9RJA1_9ROSA SubName: Full=Vesicle-associated Ma_SAEU_m 0 mM high, 200 mM low, 184 140 108 90 membrane protein 713 ECO:0000313|EMBL:EXB80720.1; E:8e- .244764 50 mM low, 600 mM low 132 len:221 type:complete E6NU59_JATCU SubName: Full=JHL18I08.3 protein Ma_SAEU_m 0 mM high, 200 mM low, 18 12 6 7 ECO:0000313|EMBL:BAJ53169.1; E:0 len:456 .266307 50 mM low, 600 mM low type:5prime_partial CALM_SPIOL RecName: Full=Calmodulin; E:6e-103 len:150 New_SABI_ 0 mM high, 200 mM low, 106 92 88 72 type:complete m.126585 50 mM low, 600 mM low SHGR6_ARATH RecName: Full=Protein SHOOT New_SABI_ 0 mM high, 200 mM low, 2 3 0 0 GRAVITROPISM 6; E:0 len:1715 type:complete m.275575 50 mM low, 600 mM low OPT7_ARATH RecName: Full=Oligopeptide transporter 7; Old_SAEU_m 0 mM low, 200 mM high, 1 5 7 4 E:0 len:767 type:complete .447580 50 mM high, 600 mM high GUN25_ARATH RecName: Full=Endoglucanase 25; E:0 New_SABI_ 0 mM low, 200 mM high, 8 28 32 24 len:620 type:complete m.749953 50 mM high, 600 mM high D7TPV8_VITVI SubName: Full=Putative uncharacterized Ma_SAEU_m 0 mM low, 200 mM high, 2 7 9 6 protein ECO:0000313|EMBL:CBI32531.3; E:2e-123 len:248 .219341 50 mM high, 600 mM high type:internal DIM_PEA RecName: Full=Delta(24)-sterol reductase; E:0 New_SABI_ 0 mM low, 200 mM high, 6 14 9 9 len:567 type:complete m.540368 50 mM high, 600 mM low 351

ENT1_ARATH RecName: Full=Equilibrative nucleotide New_SABI_ 0 mM low, 200 mM high, 16 26 28 17 transporter 1; E:0 len:412 type:complete m.615041 50 mM high, 600 mM low Q8L5K5_GOSHI SubName: Full=Fiber cell elongation protein Ma_SAEU_m 0 mM low, 200 mM high, 329 398 360 264 ECO:0000313|EMBL:ACO51065.1; E:0 len:567 type:complete .109307 50 mM high, 600 mM low PSD1A_ARATH RecName: Full=26S proteasome non-ATPase Old_SAEU_m 0 mM low, 200 mM high, 5 5 17 22 regulatory subunit 1 homolog A; E:0 len:998 type:complete .429772 50 mM low, 600 mM high LACS2_ARATH RecName: Full=Long chain acyl-CoA New_SABI_ 0 mM low, 200 mM high, 13 36 63 41 synthetase 2; E:0 len:436 type:complete m.772929 50 mM low, 600 mM high No_Blast_hit len:305 type:internal Old_SABI_m. 0 mM low, 200 mM high, 0 3 9 9 379849 50 mM low, 600 mM high No_Blast_hit len:309 type:internal Old_SABI_m. 0 mM low, 200 mM high, 0 1 2 2 379859 50 mM low, 600 mM high No_Blast_hit len:482 type:3prime_partial Old_SABI_m. 0 mM low, 200 mM high, 2 1 10 8 379870 50 mM low, 600 mM high VILI2_ARATH RecName: Full=Villin-2 Old_SABR_m 0 mM low, 200 mM high, 0 0 5 2 ECO:0000303|PubMed:10631247; E:0 len:985 type:complete .616190 50 mM low, 600 mM high S36A1_HUMAN RecName: Full=Proton-coupled amino acid New_SABI_ 0 mM low, 200 mM high, 4 12 20 14 transporter 1; E:5e-40 len:426 type:complete m.315384 50 mM low, 600 mM high PH1_ARATH RecName: Full=Pleckstrin homology domain- New_SABI_ 0 mM low, 200 mM high, 0 0 4 3 containing protein 1; E:6e-64 len:148 type:complete m.66690 50 mM low, 600 mM high 352

A0A059DFC8_EUCGR SubName: Full=Uncharacterized Ma_SAEU_m 0 mM low, 200 mM high, 0 0 3 1 protein ECO:0000313|EMBL:KCW88920.1; E:6e-178 len:329 .17850 50 mM low, 600 mM high type:complete M1APH7_SOLTU SubName: Full=Uncharacterized protein Ma_SAEU_m 0 mM low, 200 mM high, 35 43 50 44 ECO:0000313|EnsemblPlants:PGSC0003DMT400027312; E:2e- .258415 50 mM low, 600 mM high 168 len:368 type:5prime_partial S38A3_HUMAN RecName: Full=Sodium-coupled neutral Old_SAEU_m 0 mM low, 200 mM high, 1 7 14 5 amino acid transporter 3; E:7e-36 len:427 type:internal .472958 50 mM low, 600 mM low INT1_ARATH RecName: Full=Inositol transporter 1; E:0 Old_SABR_m 0 mM low, 200 mM low, 11 18 9 6 len:500 type:complete .232691 50 mM high, 600 mM low AMT12_SOLLC RecName: Full=Ammonium transporter 1 New_SABI_ 0 mM low, 200 mM low, 3 9 2 0 member 2; E:0 len:502 type:complete m.306583 50 mM high, 600 mM low CCD_CROSA RecName: Full=Carotenoid 9,10(9',10')-cleavage New_SABI_ 0 mM low, 200 mM low, 188 232 189 151 dioxygenase; E:0 len:558 type:complete m.609942 50 mM high, 600 mM low DNJ10_ARATH RecName: Full=Chaperone protein dnaJ 10; New_SABI_ 0 mM low, 200 mM low, 10 9 13 17 E:2e-120 len:354 type:complete m.467702 50 mM low, 600 mM high