Mechanisms of flood tolerance in wheat and rice Herzog.pdf · Mechanisms of flood tolerance in wheat and rice The role of leaf gas films during plant submergence Academic supervisor

U N I V E R S I T Y O F C O P E N H A G E N F A C U L T Y O F S C I E N C E F R E S H W A T E R B I O L O G I C A L L A B O R A T O R Y

PhD Thesis Max Herzog

Mechanisms of flood tolerance in wheat and rice The role of leaf gas films during plant submergence

Academic supervisor: Professor Ole Pedersen

Submitted on: 24 August 2017

ii

Name of department: Department of Biology Author: Max Herzog, Bsc. & Msc. Title and subtitle: Mechanisms of flood tolerance in wheat and rice

The role of leaf gas films during plant submergence Supervisor: Ole Pedersen, Professor Committee: Jens Borum, Associate Professor (Chair) Freshwater Biological Laboratory University of Copenhagen

Denmark Eric J. W. Visser, Assistant Professor Institute for Water and Wetland Radboud University Nijmegen

The Netherlands Margret Sauter, Professor Plant Developmental Biology and Plant Physiology Christian-Albrechts-Universität zu Kiel Germany

Submitted on: 24 August 2017 Cover photo: Submerged wheat during a flood in Denmark, June 2011. Photo: Ole

Pedersen

This thesis has been submitted to the PhD School of The Faculty of Science, University of

Copenhagen

iii

Preface

The current Ph.D. thesis represents the work of three years ranging from August 2014 to August

2017, of which one year was dedicated to teaching and course activities. The thesis was prepared

at the Freshwater Biology Laboratory at the University of Copenhagen, Denmark, and was

funded by the Villum Foundation. The aim of the project was to determine why wheat plants

perish upon submergence. Wheat can experience floods in environments ranging from

Scandinavian winter wheat fields to rice-wheat cropping systems in Asia. This thesis is based on

three manuscripts with me as first author (two published, one submitted for publication) and two

co-authored studies in the appendix. Emphasis was on the role of leaf gas films – air layers

surrounding submerged superhydrophobic leaves – due to their importance for rice and wild

wetland plant submergence tolerance. Hence, this thesis encompasses studies investigating the

role of leaf gas films on wheat (and rice) submergence tolerance, but also deals with other

aspects such as traits conferring waterlogging (soil flooding) tolerance. Although it may seem

overly optimistic, it is my hope that this and other work will one day result in commercially

available dry land crops with improved flood tolerance, allowing farmers to experience yield

stability in spite of a changing climate in the decades to come.

Copenhagen, Denmark, August 2017

Max Herzog

iv

Acknowledgements

I have many people to thank for enabling me to hand in this thesis. First of all I would like to

thank my supervisor Ole Pedersen for allowing me to join the project. I know that you have felt

responsible for luring me out of a permanent high school teaching position and onto the

uncertain waters of science, but I am very glad you did as the last three years have been a truly

great experience. I would also like to thank you for your understanding of the challenges in

combining science with family life, and for you and Anja to open your home for me and my

family. It is beyond what one can expect from a supervisor.

Special thanks go to Tim Colmer for invaluable scientific support, and for hosting me at the

University of Western Australia. Visiting Perth was a great experience for me and my family

which we will never forget. Thanks go out to other good people at UWA – ranging from my

neighbour Phill driving me to work, to Imran Malik cooking the most marvellous dinner. I would

also like to thank Gustavo Striker from Buenos Aires for a great collaboration – I hope to work

with you again someday.

When doing this Ph.D. little was more important than the daily interaction with all the friendly

people at FBL. Thank you Kathrine and Lars B. for sharing joys and frustrations of combining a

Ph.D. with parenthood. Thank you Anders W. for making me laugh at even the most hopeless

situations and/or datasets. Thank you Dennis for introducing me to the crazy world of science,

and always helping me find The Shed, tweezers and CO2 cylinders (etc.) at UWA. The biggest

thanks to Mikkel M.Ø. and Emil K. for invaluable shrimp tank consultancy, and for trying to

make me join the cross-fit team. It will never happen. Most sincere thanks to the good lab

technicians Ayoe and Anne for their patience. Thank you Lars I. for statistical advice (boy, you

must hear that a lot). Also thanks to Jens for all the advice on ‘røvballestatistik’, and all

remaining FBL staff and students for making FBL such a nice work place.

Thanks to my parents, sister and brothers; especially for asking why I wasn’t done yet from day

one. However, nobody could be more important to thank than my strong and beautiful wife

Lisbet. Thank you for encouraging me to follow my heart and take on this project and for

bearing with my wrinkled forehead or overly enthusiasm over drowned wheat plants. Thank you

for taking care of our wonderful children when days got long and supporting my every step.

v

Please grow old with me. Thank you Alfred for always making me want to leave work early to

read comic books and wrestle. Thank you Edith for just being, watching you grow is the best

thing in the world. If you read this one day I hope you will know that you mean more to me than

any piece of work will ever do – the two of you are the best I have ever made.

Max

Table of contents

PREFACE ........................................................................................................................... III

ACKNOWLEDGEMENTS ................................................................................................. IV

TABLE OF CONTENTS ...................................................................................................... 1

ABSTRACT ......................................................................................................................... 2

DANSK RESUMÉ ................................................................................................................ 3

THESIS INTRODUCTION ................................................................................................... 4

THESIS AIMS .................................................................................................................... 14

WORK INCLUDED IN THE CURRENT THESIS ............................................................... 15

CHAPTER 1: MECHANISMS OF WATERLOGGING TOLERANCE IN WHEAT – A REVIEW OF ROOT AND SHOOT PHYSIOLOGY ............................................................ 18

CHAPTER 2: PHYSIOLOGY, GENE EXPRESSION AND METABOLOME OF TWO WHEAT CULTIVARS WITH CONTRASTING SUBMERGENCE TOLERANCE .............. 38

CHAPTER 3: LEAF GAS FILMS CONTRIBUTE TO RICE (ORYZA SATIVA) SUBMERGENCE TOLERANCE DURING SALINE FLOODS .......................................... 66

CONCLUSION AND PERSPECTIVES FOR FUTURE RESEARCH ................................ 80

LITERATURE CITED IN INTRODUCTION, CONCLUSION AND OUTLOOK .................. 87

APPENDIX I: FLOOD TOLERANCE OF WHEAT – THE IMPORTANCE OF LEAF GAS FILMS DURING COMPLETE SUBMERGENCE ...................................................... 92

APPENDIX II: LEAF GAS FILM RETENTION DURING SUBMERGENCE OF 14 CULTIVARS OF WHEAT (TRITICUM AESTIVUM) ........................................................ 104

SUPPORTING INFORMATION TO CHAPTER 1, 2 AND 3 ............................................ 116

1

Abstract

Most crops are sensitive to excess water, and consequently floods have detrimental effects on

crop yields worldwide. In addition, global climate change is expected to regionally increase the

number of floods within decades, urging for more flood-tolerant crop cultivars to be released.

The aim of this thesis was to assess mechanisms conferring rice (Oryza sativa) and wheat

(Triticum aestivum) flood tolerance, focusing on the role of leaf gas films (see picture below)

during plant submergence.

Reviewing the literature showed that wheat germplasm holds genetic variation towards

waterlogging (soil flooding), and highlighted traits such as improved internal aeration of the root

system and short term anoxia tolerance of seminal roots as conferring tolerance. However,

further work on especially anoxia tolerance and genotype × environment interactions is required

in order to explore the available genetic resources. Experimental work assessed the physiologic,

metabolomic and genetic response of wheat subjected to complete submergence, documenting

contrasting submergence tolerance between two cultivars. While both cultivars displayed similar

leaf gas film retention times and carbohydrate consumption rates, results indicated that the

contrasting submergence tolerance could rather be governed by tolerance to radical oxygen

species or contrasting metabolic responses (other than carbohydrate consumption) to ethylene

accumulation. Manipulating leaf gas film presence affected wheat and rice submergence

tolerance such as plant growth and survival. However, leaf gas film retention times did not differ

between 14 winter wheat cultivars, and leaf gas films did not prevent significant leaf Na+ and Cl-

intrusion, and K+ loss, during rice submergence in saline water. Due to the significant salt

intrusion and low genetic variation in wheat gas film retention times, a future prominent role of

leaf gas films in improving (i) wheat submergence and (ii) rice salinity tolerance was not

generally supported.

Leaf gas film on a wheat leaf / gasfilm på et hvedeblad. Photo: Ole Pedersen.

2

Dansk resumé

Mange af menneskets vigtigste afgrøder er følsomme overfor oversvømmelser, hvilket årligt

medfører store globale udbyttetab. Derudover forventes klimaforandringer indenfor årtier at

medføre et øget antal oversvømmelser i mange landbrugsregioner. For at sikre den fremtidige

fødevareforsyning er der derfor behov for afgrøder med øget tolerance overfor abiotiske

stressfaktorer såsom oversvømmelser. Formålet med denne Ph.D.-afhandling var at undersøge

hvilke mekanismer der kan medføre øget oversvømmelsestolerance hos hvede (Triticum

aestivum) og ris (Oryza sativa), med fokus på betydningen af bladenes ’gasfilm’ – et tyndt

luftlag omkring superhydrofobe blade som dannes under neddykning (se billede på forrige side).

Et litteraturstudie over hvilke mekanismer der betinger hvedes tolerance overfor vandmættet jord

(waterlogging) viste at nogle sorter er relativt tolerante, og fremhæver bl.a. intern ilttransport via

luftvæv i adventivrødder, samt røddernes anoxi-tolerance som vigtige mekanismer.

Fortolkningen af resultater på tværs af globale regioner besværliggøres dog af interaktioner

imellem genotype og miljø. Derudover kræves en mere detaljeret viden om især anoxi-tolerance

førend disse ressourcer kan udnyttes i planteforædlingssammenhæng. Afhandlingens

eksperimentelle studier fokuserer på effekten af fuldstændig neddykning på fysiologi, gen-

ekspression og metabolisme hos ris og hvede. Heri dokumenteres bl.a. en signifikant forskel på

to hvedesorters oversvømmelsestolerance. Dog var tolerancen ikke som forventet betinget af

skuddets kulhydratomsætning som det er kendt fra ris, eller af gasfilmens levetid, men nok

nærmere regulering af fri iltradikaler, eller forskel i sorternes ethylen-sensitivitet. Manipulation

af gasfilmen påvirkede vækst og overlevelse i hvede (samt vækst i ris) under neddykning, men

gasfilms-levetiden blandt 14 neddykkede hvedesorter var meget ens. Derudover forhindrede

gasfilmen (ved en simuleret kystnær saltvands-oversvømmelse) ikke som ventet signifikant tab

af K+ eller indtrængen af Na+ og Cl- i ris-blade. Grundet denne signifikante salt-indtrængen hos

ris, samt den lave variation i hvedesorternes gasfilm-levetid, vurderes det ikke at gasfilmen vil

have afgørende betydning for fremtidig sikring af (i) oversvømmelsestolerance i hvede og (ii)

salttolerance i ris.

3

Thesis introduction and aims

Thesis introduction

Floods affect ecosystems ranging from desserts to river forelands and agricultural soils (Jackson

2004). Accounting for 43% of all natural disasters recorded within a decade, floods are the most

frequent natural disaster worldwide (de Guenni et al. 2005). This Ph.D.-thesis deals with the

detrimental effects of floods on crop production.

Floods occur when water input exceeds the rate of infiltration, evaporation or runoff leading to

flooding of the plant-soil system. Flash floods are rapidly occurring floods, often following brief

torrential rain in the catchment (Brammer 1990; Adhikari et al. 2010) or failing of dam or river

levees (NSSL 2017). Flash floods may be of short duration when waters run off quickly, but may

also result in longer, stagnant floods in lowland areas (Setter et al. 1987). Rainwater floods are

caused by heavy rain and/or poor soil drainage, while river floods occur when water levels rise

above the top of river banks (Brammer 1990; Adhikari et al. 2010). Coastal floods inundate land

areas when tides are higher than average, strong winds force sea water up river systems or sea

waters rise due to low atmospheric pressure. Floods may be seasonal (e.g., caused by monsoon

rains or snow melts) or more unexpected as due to more than average precipitation. Land use

changes such as catchment deforestation, farmland drainage causing faster runoff or excess

irrigation can also increase the risk of floods (Rienk et al. 2002).

Several estimates have been made as to how much land annually suffers from excess water.

Some commonly used estimates are that 10% of the global land area and 20% of Europe and the

Russian Federation is affected by severe soil drainage constrains (Setter & Waters 2003); 10% of

all irrigated farmland suffers from waterlogging (Jackson 2004); 16% of U.S. soils limit plant

production by being too wet (Boyer 1982); 15-20% of all wheat growing areas are affected by

waterlogging each year (Sayre et al. 1994) and > 35% of the global rice acreage is considered

flood prone (Bailey-Serres et al. 2012). Although the basis of such estimates is not always

entirely clear, they indicate that excess water is a global challenge to crop production. Attempts

to quantify such crop losses by reviewing historic insurance indemnity payouts in the U.S.

(Boyer 1982) showed that from 1939-1978 indemnity payout due to excess water (16% of all

payouts) was second only to drought (41% of all payouts). More recently, 2010-2016 floods

4


accounted for 25% of all indemnities, exceeding any other single stress (drought, heat, cold,

pests etc.) in 4 out of 7 years (Fig. 1). Payouts due to floods averaged US$2.4bn per year,

totalling US$19 billion (Fig. 1). These high indemnity payouts document that, in the U.S., floods

are one of the main factors limiting crop production.

Figure 1. a) number of floods characterised as disasters (see text for definition), as monitored by the International Disaster

Database at the University of Louvein in Belgium (Millennium Ecosystem Assessment 2005). b) Crop production loss due to

environmental stresses from 2010-2016, based on insurance indemnities paid to farmers. Values are from the U.S. Department of

Agriculture Risk Management Agency Cause of Loss Historical Data and grouped as follows: Floods (Excess

moisture/precipitation/rain or floods), drought (water deficit stress), temperature (cold winter, cold wet weather, freeze, frost,

heat, hot winds) and other stresses (e.g., decline in price, insects, pests, wildlife, hail, hurricanes, tornado, wind, snow and

lightning). For preceding years see Boyer (1982) and Bailey-Serres et al. (2012).

In some regions floods are expected to increase in frequency and severity within the next

decades due to global climate change (Parry et al. 2007). Global surveillance data indicate that

the number of floods characterised as disasters (10 or more people reported killed, 100 or more

people reported affected, international assistance was called or a state of emergency was

declared) has increased at a constant rate since the 1940’s (Fig. 1a). However, care should be

taken when interpreting these results as the number of reported floods has also increased due to

improved telecommunication and better coverage of global information (de Guenni et al. 2005).

As examples for the expected increase in floods, soil moisture in excess for European wheat

production (already considered a persistent problem in 5 out of 13 European environmental

zones) is expected to increase in the UK, the Netherlands and Denmark by 2060 (Olesen et al.

2011; Trnka et al. 2014). In addition, increased precipitation due to climate change has been

projected to double crop losses from $1.5 billion to $3 billion per year in the U.S.

5


(Rosenzweig et al. 2002), and rising sea levels are expected to more frequently inundate large

parts of Asia’s rice growing areas (Ali 1996; Wassmann et al. 2004; Sarwar & Khan 2007).

The Food and Agriculture Organization of the United Nations states that the world needs to

produce 70% more food by 2050 in order to feed a world population of 9 billion (FAO 2009). To

meet this goal, The World Bank Group argues that adapting Climate Smart Agriculture

(encompassing production of crops more tolerant to abiotic stress) especially in developing

countries will be necessary (Braimoh et al. 2016). An important step could be to improve crop

flood tolerance, minimizing yield losses when agricultural soils inevitably become flooded.

How floods affect terrestrial vegetation

Floods are one the most dramatic environmental changes terrestrial plants can experience. The

term “flooding” is used to describe excessively wet conditions encompassing “waterlogging”

(soil flooding) and “submergence” situations where the above ground organs are under water

(Sasidharan et al. 2017). In the following I will apply “waterlogging” when only the root zone is

flooded, “submergence” when all (or part of) the shoot is submerged and “flooding” when

discrimination is not necessary. The effects of waterlogging and/or submergence on terrestrial

plants have been thoroughly reviewed for anoxia tolerance (Gibbs & Greenway 2003; Greenway

& Gibbs 2003), waterlogging tolerance in wheat, barley and oats (Setter & Waters 2003),

ethylene and O2 signalling (Voesenek & Sasidharan 2013), underwater photosynthesis (Colmer

et al. 2011), root responses (Elzenga & van Veen 2010; Sauter 2013), adaptations to

submergence (Voesenek et al. 2006) and more general reviews (Bailey-Serres & Voesenek 2008;

Colmer & Voesenek 2009; Striker 2012). Thus, the following sections will only provide a brief

summary of the numerous challenges experienced by flooded terrestrial vegetation.

Waterlogging reduces soil O2 levels due to the 20-30 fold lower solubility and the 104 time

slower diffusion of O2 in water compared to that in air (Armstrong 1979). O2 is therefore quickly

consumed by roots and soil-microorganism when soil air-filled porosity is 10% or less

(Ponnamperuma 1984), resulting in severe hypoxic or even anoxic (absence of O2) conditions

(Fig. 2a). The rate and degree of which soils turn anoxic depend on a wide range of factors, e.g.,

soil composition, water flow through the soil profile, temperature, biological activity etc.

6


Waterlogged soils experience a wide range of electrochemical changes in addition to O2 decline

(Fig. 2b). A major change is the shift to low (200 to -400 mV) redox potentials (Ponnamperuma

1984). The change from oxidative to reducing environment is caused by anaerobe microbes

using oxidised soil components and organic matter as electron acceptors in their respiration,

reducing soils in thermodynamic sequence (Fig. 2b). The reduction of these compounds results

in the gradual disappearance of NO3-, Mn4+, Fe3+, SO4

2-, CO2 and increase in soluble NH4+,

Mn2+, Fe2+, S2-, H2S, CH4 (methane) and organic acids if waterlogging is prolonged

(Ponnamperuma 1984). Some of these reduced compounds may have phytotoxic effects, and can

enter roots and accumulate in addition to endogenously produced CO2 and ethylene.

Waterlogging often results in low nutrient uptake by terrestrial plants due to inadequate O2

supplies. Anaerobic energy metabolism produces some ATP, but since ATP production via

glycolysis renders only 2-4 mol ATP per mol hexose compared with 24-36 mol ATP in aerated

tissues, roots face energy shortage (Gibbs & Greenway 2003). Thus, during soil waterlogging the

Figure 2. Soil chemical changes upon waterlogging

(soil flooding). a) O2 partial pressures (kPa) with time of

waterlogging in commercial potting mix, measured

using four O2 micro sensors (OXF500PT, Pyrosience,

Aachen, Germany) in four pots (Rep# 1-4) at 4-5 cm

depth (own, unpublished data). Arrow indicates when

waterlogging was imposed. b) Electrochemical changes

in a Western Australia duplex soil (□; 5-15 °C) and a

Philippine heavy clay soil (●; 25-35 °C) with time of

waterlogging. From Setter and Waters (2003).

7


high energy demand of growth cannot be met and root growth of terrestrial plants is arrested,

leading to reduced soil exploration and reduced surface area for nutrient uptake (Elzenga & van

Veen 2010). Reduced proton motive force and less negative membrane potential resulting from

low ATP levels also decreases nutrient uptake per unit of root mass (Armstrong & Drew 2002;

Elzenga & van Veen 2010). Anoxic steles can reduce xylem loading, resulting in reduced

nutrient transport from roots to shoots (Gibbs et al. 1998; Colmer & Greenway 2011). In

addition, waterlogging also alters soil nutrient dynamics. While denitrification in waterlogged

soils causes losses of nitrate, P availability may increase as Fe is solubilised (Elzenga & van

Veen 2010).

A paradoxical response to waterlogging is leaf wilting (Striker 2012). Both structural and

functional constraints have been proposed as explanations for a disturbed water balance of shoots

of waterlogged plants. A reduced root:shoot ratio is common in dryland crops when waterlogged

(Huang et al. 1994b; Malik et al. 2001; Malik et al. 2002) and implies a lower water absorption

surface area of roots in relation to the transpiratory surface of leaves, which together with

impaired root hydraulic conductivity (Bramley & Tyerman 2010) can result in plant wilting,

particularly under high evaporative demand.

Cells may also be damaged by reactive O2 species (ROS) (Blokhina et al. 2003). When water

recedes and O2 re-enters the soil and plant tissues, the formation of ROS (e.g., superoxide,

hydroxyl radicals, singlet O2 and hydrogen peroxide) might especially damage cell membranes.

ROS production is a consequence of a low energy charge, high level of reducing equivalents and

saturation of the mitochondrial electron transport chain, favouring electron leakage to O2 upon

re-aeration (Blokhina et al. 2003). For details on nutrient uptake, water balance and ROS in

waterlogged wheat, see Chapter 1.

The waterlogging induced stressors described above result in reduced water transport and

nutrient uptake by the plant, reducing growth and yields. This is evident for wheat, with average

shoot biomass, root biomass and grain yields following waterlogging showing reductions of

33%, 62%, and 43% relative to drained controls (Fig. 3), respectively.

8


When floodwaters rise above soil levels, shoots will also become submerged adding additional

stressors to terrestrial plants (Voesenek et al. 2006). In terrestrial leaves in air, CO2 enters leaves

through open stomata, reaching chloroplasts through intracellular airspaces. Stomata are believed

to close upon submergence, forcing gasses to diffuse across leaf cuticles posing much higher

resistance to diffusion (Mommer et al. 2005). In addition, a thin diffusive boundary layers (DBL)

of water with no turbulence but only laminar flows parallel to the tissue surface form around

submerged leaves (Mommer et al. 2004; Pedersen et al. 2009), hampering CO2 uptake for

photosynthesis even further and resulting in the underwater photosynthesis (PN) to be merely 9%

of PN in air (mean of three species, Colmer et al. (2011)). The slow diffusion of gasses through

the DBL and across cuticles, together with light reduction by turbid floodwaters, also impedes

shoot uptake and/or production of O2 required for aerobe respiration.

S h o o t m a s s R o o t m a s s G r a in y ie ld0

5 0

1 0 0

1 5 0

Pe

rce

nta

ge

of

dra

ine

d c

on

tro

ls (

%) * * *

Adaptations in terrestrial wetland plants: what can be learnt from flood-tolerant plants?

From the above, one might gain the impression that vigorous plant growth in water saturated

soils or in water is not possible. However, nothing could be more wrong! Since

Figure 3. The effects on wheat of waterlogging for shoot dry mass (median = 67% of controls, n = 46), root dry mass (median

= 38% of controls, n = 46) and grain yield (median = 57% of controls, n = 206) as percentage of drained controls. This

summary figure was compiled from data extracted from peer-reviewed literature (see chapter 1 in this thesis). Boxes are 50%

of the observations with the median shown as the horizontal line and bars are 1 and 99 percentiles; outliers are shown as ●. *

denotes significant differences from 100% (Wilcoxon Signed Rank test, P < 0.0001). Data are from experiments where wheat

was waterlogged in soil (either in pots or in field situations) for 7-42 days without recovery (root and shoot mass) or for 4-120

d for grain yield (most experiments with recovery).

9


photosynthesizing macrophytes invaded land from marine habitats 400 million years ago, they

have (re)colonised marine environments (e.g., seagrasses such as Zostera), freshwaters (rooted

macrophytes) and wetlands with great success. Wetland plants show a range of adaptations,

allowing tropical wetland plants to exhibit primary production rates higher than any terrestrial

ecosystem (Kalff 2002). It is therefore intriguing to ask what can be learnt from these natural

wetland plants to possibly improve the flood tolerance of cultivated crops. In the following, I

will review the main adaptation conferring flood tolerance in wild wetland plants. Some parts of

the following text have been re-written from the unpublished introduction of my master’s thesis

(Herzog 2013).

Some species inhabiting flood prone areas are not truly flood-tolerant, but avoid flooding stress

by completing energy-demanding periods of their lifecycle (growing, flowering, seed-setting) in

dry periods, whereas flooding periods are survived by dormant life stages such as seeds, tubers

and bulbs (Blom & Voesenek 1996; Bailey-Serres & Voesenek 2008). Meanwhile, flood-tolerant

plants reduce flooding stress by a number of traits (Bailey-Serres & Voesenek 2008; Colmer &

Voesenek 2009) enabling them to maintain a high metabolism in spite of the lowered energy

harvest from underwater photosynthesis and risks of anoxia (Blom & Voesenek 1996). In the

following sections adaptations to enable waterlogging and submergence tolerance, respectively,

will be summarised.

Plant adaptations to waterlogging

As root tissues do not tolerate long term anoxia (Gibbs & Greenway 2003), acclimation to

waterlogging concentrates around supplying roots with O2. Waterlogging-tolerant species can

form airspaces (aerenchyma) in roots and shoots upon waterlogging, ensuring low-resistance gas

transport between plant organs. Many flood-tolerant species form large adventitious root systems

with high porosity after waterlogging (Blom & Voesenek 1996). Aerenchyma is constitutive in

rice (Oryza sativa), while in dryland crop species such as maize (Zea mays), wheat (Triticum

aestivum) and barley (Hordeum vulgare) aerenchyma does not form under drained conditions but

is induced upon flooding (Yamauchi et al. 2013). In these crop species, root aerenchyma is of the

lyzigenous type formed by programmed death and lysis of root cortical cells. Before reaching the

root apex however, O2 may readily be lost from roots to the soil, termed radial O2 loss (ROL)

(Colmer 2003).

10


In waterlogging-tolerant species, barriers to ROL in basal zones of the roots promote

longitudinal diffusion of O2 towards the root apex (Fig. 4), creating oxic conditions at these sites

(Colmer 2003). This enables deeper rooting in anoxic soils (Colmer 2003). Waterlogging-

sensitive species such as wheat only possess a weak ROL-barrier (Malik et al. 2011), so that O2

is lost along the whole root, thereby not reaching the root apex (Colmer & Voesenek 2009).

Aerenchyma formation takes time, so that roots are likely to experience O2 deficiency initially.

Tolerance to short term anoxia includes maintaining ATP production via glycolysis, ensuring

root functioning when aerobe respiration ceases (Bailey-Serres & Voesenek 2008). However,

noting that fermentation capacity was similar in anoxia-tolerant rice and intolerant soybean,

emphasis has been placed on the role of efficient energy utilization rather than ATP production

per se (Atwell et al. 2015). When growing in waterlogged soils, roots may face high

concentrations of various microelements such as Mn2+ and Fe2+. Studies have indicated that

tolerance towards high concentrations of these micronutrients is an important trait in crop

waterlogging tolerance (Setter et al. 2009; Shabala 2011), especially on acidic soils, with the

specific action of these metabolites still remaining elusive (Shabala 2011). Tolerance to ROS

requires an oxidative defence system. ROS can be detoxified by enzymes (e.g., superoxide

dismutase) and antioxidants (e.g., ascorbate and glutathione). The effect of O2 deficiency on the

antioxidant system is species-specific (Blokhina et al. 2003), with rice and Arabidopsis showing

upregulation of mRNA encoding enzymes involved in anaerobic metabolism and ROS

deprivation when subjected to low O2 (Bailey-Serres & Voesenek 2008).

11


Plant adaptations to submergence

Submergence regimes differ in depth and duration, requiring different plant responses which

have been categorised into two main strategies: the Low O2 Quiescence Syndrome (LOQS) and

the Low O2 Escape Syndrome (LOES) (Bailey-Serres & Voesenek 2008; Colmer & Voesenek

2009). In the quiescence syndrome, traits concentrate around “sitting through” the unfavourable

conditions. Energy (mainly from glycolysis) is spent on vital processes such as maintenance

instead of growth – resulting in preservation of substrates until the floodwater recedes (Bailey-

Serres & Voesenek 2008).

The LOQS differs markedly from the escape syndrome, where shoot elongation is a vital

response (Bailey-Serres & Voesenek 2008). When petioles, stems or leaves elongate, the plant

may be able to reestablish air contact by emerging above floodwaters. Shoot elongation

(alongside other acclimations to submergence) is initiated by ethylene, a volatile plant hormone

that accumulates in submerged plant tissues (Voesenek et al. 2006). A costly strategy, the escape

syndrome can result in plant death if energy stores are depleted before the plant emerges. This

has been illustrated in rice, were elongating cultivars show significantly lower submergence

tolerance than non-elongating cultivars (Ismail et al. 2013).

Figure 4. Scheme of two contrasting radial O2

loss (ROL) patterns in roots containing a weak

(a) and strong (b) barrier to ROL, resulting in

poor apex oxygenation in (a) as O2 is lost to

waterlogged and anoxic soils along the whole

root. In many waterlogging-tolerant wetland

plants waxy suberin depositions along the root

base act as a physical barrier to ROL, resulting

in an aerobic zone around the sensitive apex

(b). The leaking O2 detoxifies reduced soil

constituents, mineralises nurtients for plant

uptake and allows for deeper rooting in

waterlogged soils. Figure from Striker (2012).

12


Some wetland plants can also form new “aquatic” leaves upon submergence, reducing their

cuticle and cell wall thicknesses and reorienting chloroplasts towards the epidermis (Mommer et

al. 2005). In such leaves, the diffusive resistance to entry of CO2 and O2 is lowered, increasing

underwater photosynthesis and O2 uptake from floodwaters at night.

Underwater photosynthesis is also enhanced by leaf gas films (Colmer & Pedersen 2008;

Pedersen et al. 2009). Also termed “plant plastrons”, these ca. 50 µm thick micro layers of gas

(Fig. 5) surround superhydrophobic leaf surfaces when submerged (Colmer & Pedersen 2008;

Winkel et al. 2011). By enlarging the gas-water interface and possibly allowing stomata to

remain open, gas exchange between plant and water is enhanced (Verboven et al. 2014). Leaf

gas films have also shown to delay entry of NaCl from saline floodwater into submerged leaves

(Teakle et al. 2014). However, leaf gas film retention is time-limited as leaf hydrophobicity

declines with time of submergence (Teakle et al. 2014; Winkel et al. 2014). Screening 25 species

of terrestrial plants under controlled laboratory conditions showed that leaf gas film retention

time varied from 0 to more than 11 days, showing that gas film retention time is species specific

(Winkel et al. 2016). The reason for loss of leaf hydrophobicity during submergence, and

environmental factors affecting gas film collapse have not yet been resolved, but is of

importance in order for submerged terrestrial plants to “stay dry under water”.

13


Figure 5. Submerged wheat in (a) Western Australia and (b) Denmark. (c) Superhydrophobic wheat leaves causing water to form

perfectly spherical droplets on leaf surfaces and (d) a thin gas film to be retained on submerged leaves, resulting in a silvery

appearance. The gas bubble on (d) results from underwater photosynthesis. From Appendix I.

Thesis aims

The overall aim of this Ph.D.-thesis was to determine why wheat plants succumb during

submergence, focusing on the importance of gas films forming on submerged leaves. I attempted

to achieve this by

1) Reviewing the literature on wheat flood tolerance, highlighting important traits conferring

tolerance and suggesting areas for future research (Chapter 1)

2) Experimentally assessing the importance of leaf gas films during complete submergence of

wheat and also rice, with the use of rice enabling me to build upon previously published rice

work (Pedersen et al. 2009; Winkel et al. 2013; Winkel et al. 2014). In detail, (i) contrasting

submergence tolerance in two wheat cultivars, (ii) gas film effects on ion intrusion during

rice saline submergence, (iii) the effects of leaf gas film removal on wheat submergence

14


tolerance, (iv) variation in gas film retention times among 14 wheat cultivars and (v) reasons

for leaf gas film collapse during submergence were investigated.

My hopes are that these results will help to drive research towards improving crop flood

tolerance.

Work included in the current thesis

Chapter 1 - HERZOG, M., STRIKER, G. G., COLMER, T. D. & PEDERSEN, O. 2016.

Mechanisms of waterlogging tolerance in wheat – a review of root and shoot physiology. Plant,

Cell & Environment, 39, 1068-1086.

Commencing this study shortly after taking up my position as a Ph.D.-student allowed (forced)

me to get acquainted with the relevant literature at an early stage. We compiled data from

published studies into larger data-sets, allowing for meta-analysis of wheat responses to flooding

in regard of growth, yield, nutrient uptake, photosynthetic rates, root aerenchyma formation and

other physiological responses. Since research has focused on wheat waterlogging responses,

chapter 1 only deals with this flooding regime. We show that wheat germplasm holds significant

variation towards waterlogging, but that tolerance is affected by environmental factors (e.g.,

temperature, soil type) and experimental design (e.g., timing and duration of waterlogging). We

highlight the importance of shoot nitrogen deficiency as main contributor to reduced wheat shoot

growth upon waterlogging, while microelement toxicity appears to occur mainly in some acidic

soils. Mechanisms that seem to determine wheat waterlogging tolerance are seminal root short-

term anoxia tolerance (which needs to be specified), number of adventitious roots and amount of

aerenchyma within these roots. Some of the suggested areas for future research are genotypic

variations in ROS-tolerance, mechanisms of microelement tolerance, differences in efficiency of

energy use, phloem transport and ability of plants to recovery after waterlogging.

15


Chapter 2 – HERZOG, M., FUKAO, T., WINKEL, A., KONNERUP, D., LAMICHHANE, S.,

ALPUERTO, J.B., HASLER-SHEETAL, H., PEDERSEN, O. Physiology, gene expression and

metabolome of two wheat cultivars with contrasting submergence tolerance. Submitted to Plant

Physiology Focus Issue on Energy: Light and Oxygen Dynamics, August 1st 2017.

In this interdisciplinary study we document contrasting submergence tolerance in two wheat

cultivars with reputed differential waterlogging tolerance, as cultivar Jackson survived

submergence approximately 7 days longer than cultivar Frument. Physiological measurements

indicated that submergence induced leaf degradation (as indicated by leaf chlorophyll and leaf

porosity data) several days earlier in Frument than in Jackson. Meanwhile, shoot carbohydrate

levels did not seem to explain the varietal differences, although expression of genes related to

carbohydrate catabolism was occasionally higher in Frument. Metabolomic analysis revealed

that the two cultivars differed in concentration of the amino acid (and antioxidant) proline,

peroxidation marker malondialdehyde and chlorophyll brake down product phytol suggesting

that ethylene sensitivity or ROS deprivation might confer tolerance; however, more research is

required to elucidate these aspects.

Chapter 3 - HERZOG, M., KONNERUP, D., PEDERSEN, O., WINKEL, A., COLMER, T. D.

2017. Leaf gas films contribute to rice (Oryza sativa) submergence tolerance during saline

floods. Plant, Cell & Environment, doi 10.1111/pce.12873.

This experimental study was conducted during my three month change of scientific environment

at the University of Western Australia, Perth. Here we focus on the potential role of leaf gas

films as a physical gas barrier around rice leaves during submergence in saline waters, as shown

in Melilotus siculus. We hypothesised that entry of Na+ and Cl- and loss of K+ would be delayed

by the presence of leaf gas film during submergence, compared to plants with leaf gas films

experimentally removed. Surprisingly, we found that in spite of leaf gas film presence, rice

leaves lost substantial amounts of K+ and Na+ and Cl- entered tissues from the floodwater. The

results indicated that in rice some leaf-water interphase was present during submergence, and

also showed that leaf gas film removal diminished rice growth upon submergence. Analysing

leaf surfaces using scanning electron microscopy (SEM) did not reveal any clear causes of leaf

hydrophobicity decline in contrast to the results from Appendix II.

16


Work included in appendix

Appendix I - WINKEL, A., HERZOG, M., KONNERUP, D., FLOYTRUP, A. H.,

PEDERSEN, O. 2017. Flood tolerance of wheat – the importance of leaf gas films during

complete submergence. Functional Plant Biology, 44, 888-898.

In this study our group showed that experimentally removing leaf gas films resulted in

significantly shorter survival time upon complete submergence of a Danish wheat cultivar. Plants

retaining gas films survived complete submergence for 13 days, while plants brushed with a

detergent prior to submergence (preventing leaf gas film formation) only survived for 10 days.

Underwater PN at varying light intensities showed that light reflection by leaf gas film resulted in

a higher light compensation point, but that this effect was already overcome at 5% of full

sunlight. This study underlines that leaf gas film removal diminishes the submergence tolerance

of a dryland crop, but that leaf gas film retention time in this wheat cultivar was rather short (< 3

days).

Appendix II - KONNERUP, D., WINKEL, A., HERZOG, M., PEDERSEN, O. 2017. Leaf gas

film retention during submergence of 14 cultivars of wheat (Triticum aestivum). Functional

Plant Biology, 44, 877–887.

In this study our group investigated the retention of leaf gas films during submergence by 14

wheat cultivars. Compared to a wild wetland grass, gas film retention time was relatively short

due to loss of leaf hydrophobicity after 2 days of submergence. The investigated cultivars

showed very little variation in regard of gas film retention time, indicating that variation for this

trait in Danish wheat cultivars is low. SEM revealed that leaf cuticles were progressively covered

by an unidentified substance with time of submergence, correlating with leaf hydrophobicity

declines and loss of leaf gas film. The results open new questions as to why leaf gas films are

lost during submergence, and indicate that in order to obtain variation in leaf gas film retention

time for breeding efforts, more genetically diverse wheat genotypes should be screened.

17

Chapter 1: Mechanisms of waterlogging tolerance in wheat – a review of root and shoot physiology

Front page of Plant, Cell & Environment depicting partially and completely submerged winter

wheat during a winter flood in Germany. Photo: Ole Pedersen

18

Special Issue

Mechanisms of waterlogging tolerance in wheat – a review ofroot and shoot physiology

Max Herzog1,3, Gustavo G. Striker2,3, Timothy D. Colmer3 & Ole Pedersen1,3,4

1Freshwater Biological Laboratory, Department of Biology, University of Copenhagen, Universitetsparken 4, 3rd floor 2100Copenhagen, Denmark, 2IFEVA-CONICET, Facultad de Agronomía, Universidad de Buenos Aires, Avenida San Martín 4453 CPA1417 DSE Buenos Aires, Argentina, 3School of Plant Biology, Faculty of Science, The University of Western Australia, 35 StirlingHighway, Crawley, Perth 6009, Western Australia, Australia and 4Institute of Advanced Studies, The University of Western Australia,35 Stirling Highway, Crawley, Perth 6009, Western Australia, Australia

ABSTRACT

We review the detrimental effects of waterlogging on physiol-ogy, growth and yield of wheat.We highlight traits contributingto waterlogging tolerance and genetic diversity inwheat. Deathof seminal roots and restriction of adventitious root length dueto O2 deficiency result in low root:shoot ratio. Genotypes differin seminal root anoxia tolerance, but mechanisms remain to beestablished; ethanol production rates do not explain anoxia tol-erance. Root tip survival is short-term, and thereafter, seminalroot re-growth upon re-aeration is limited. Genotypes differ inadventitious root numbers and in aerenchyma formationwithin these roots, resulting in varying waterlogging tolerances.Root extension is restricted by capacity for internal O2 move-ment to the apex. Sub-optimal O2 restricts root N uptake andtranslocation to the shoots, with N deficiency causing reducedshoot growth and grain yield. Although photosynthesis de-clines, sugars typically accumulate in shoots of waterloggedplants. Mn or Fe toxicity might occur in shoots of wheat onstrongly acidic soils, but probably not more widely. Futurebreeding for waterlogging tolerance should focus on root inter-nal aeration and better N-use efficiency; exploiting the geneticdiversity in wheat for these and other traits should enableimprovement of waterlogging tolerance.

Key-words: adventitious roots; aerenchyma; flooding toler-ance; genotypic variation; micronutrient toxicity; nitrogen defi-ciency; O2 deficiency; recovery ability; root anoxia tolerance;wheat (Triticum aestivum).

INTRODUCTION

Waterlogging (soil flooding), due to high rainfall, irrigationpractices and/or poor soil drainage, annually affects large areasof farmlands worldwide, imposing major constraints on rootswith negative impacts on crop yields (Jackson 2004). Thisincludes wheat (Triticum aestivum), for which 15–20% of theannual crop suffers yield losses due to waterlogging (Sayreet al. 1994; Setter & Waters 2003). In the USA, annual cropinsurance payouts over the past 5 years due to floods totaledmore than US$2bn, second only to drought (US$3bn) as a

stress (U.S. Department of Agriculture 2015). Several otherregions, including those among the top 10 wheat producers(e.g. Europe and Pakistan), recently had severe floods inflictingcrop damage or loss (Bailey-Serres et al. 2012). Floods are ex-pected to increase as a consequence of climate change (Parryet al. 2007), and increased rainfall in some areas will adverselyimpact on wheat production (Dixon et al. 2009; Trnka et al.2014). Increased effort will be needed to breed wheat varietiesbetter adapted to the regionally prevailing abiotic stressfactors, for example, drought or waterlogging (Trnka et al.2014), to meet grain production needs of our increasing humanpopulation.

Waterlogging often results in anoxic (absence of O2) soils(Ponnamperuma 1972) and severe hypoxia or anoxia withinroots (Armstrong 1979). Even roots with aerenchyma, whichfacilitates internal O2 diffusion, including in wheat (Erdmann& Wiedenroth 1986; Huang et al. 1994a), will have tissuesthat become severely hypoxic (Armstrong 1979; Colmer &Greenway 2011; Kotula et al. 2015). The shift in O2-deficienttissues from aerobic respiration to the low ATP-yielding fer-mentation results in an ‘energy crisis’ (Gibbs & Greenway2003) and inhibition of root growth and functioning in trans-port of nutrients and water to the shoot (Jackson & Drew1984; Colmer & Voesenek 2009), and eventually death ofsome roots. In addition to O2 deficits per se, soil redox po-tential declines and Mn2+ and Fe2+ and organic acids canincrease in many soils (Ponnamperuma 1972). These canenter roots and accumulate in addition to endogenously pro-duced CO2 and ethylene (e.g. for CO2, Greenway et al. 2006).Cells may also be damaged by reactive oxygen species (ROS)(Blokhina et al. 2003). Thus, when subject to soil waterlogging,roots suffer O2 deficits as well as the additional conditionssummarized in the previous text; however, even O2 deficiencywithout the other soil chemical changes can exert severe stresson roots of dryland species, such as wheat, with consequencesfor the shoots (e.g. Trought & Drew 1980a).

Here, we review the physiological mechanisms conferringwaterlogging tolerance in wheat by examining root and shootresponses and adaptations to low O2 stress and also the effectsof the additional components of the ‘compound stress’ causedby soil waterlogging. We highlight genotypic variations whereapparent.Correspondence: M. Herzog; e-mail: [email protected]

© 2015 John Wiley & Sons Ltd 1

doi: 10.1111/pce.12676Plant, Cell and Environment (2016)

19

EFFECTS OF WATERLOGGING ON ROOTGROWTH AND FUNCTIONING

Wheat root systems are reduced in size owing to growth beingimpeded and also damage and decay of the existing root system(Fig. 1; references in Supporting Information Table S1). Theseminal root dry mass (DM) declines markedly, whereas newadventitious roots develop (Trought & Drew 1980a; Maliket al. 2001). The adventitious roots contain aerenchyma withthe associated internal O2 movement to the apex enablinggrowth, albeit to a limited distance, into anoxic soils. The ad-ventitious root growth does not fully compensate for loss ofseminal root DM (Colmer & Greenway 2011), and so medianrootDM is reduced to 38%of drained controls (Fig. 1). As rootgrowth is inhibited more than shoot growth, waterlogging re-duces the median root:shoot ratio of wheat from 0.4 to 0.2(Supporting Information Table S1). The large variation in re-sponse of wheat to soil waterlogging is evident in our meta-analysis of published data (Fig. 1 and Supporting InformationTable S1), which reflects waterlogging events of different dura-tions and depths, the ‘compound stress’ (i.e.O2 deficits andother changes in soil chemistry), different temperatures, soiltypes and plant ages/developmental stages and possible geno-typic differences in responses.

Environmental parameters influencing wheatwaterlogging responses

Cooler conditions result in less severe effects of waterloggingonwheat (Luxmoore et al. 1973; Trought&Drew1982). Lower

temperature results in slowerO2 depletion from the soil, slowerroot metabolism and slower shoot growth and thus less de-mand for water and nutrients (Trought &Drew 1982). In someconditions (e.g. low biological activity, low temperature andmass flow of water through the soil), soil anoxia may not occur(Setter & Waters 2003). One example of the effect of soil typeon the response of wheat to mid-winter waterlogging was yieldbeing reduced by 16% in clay soil comparedwith 7% in a sandysoil (Cannell et al. 1980), probably caused by a faster O2 deple-tion during waterlogging and slower return to oxic conditionsupon drainage in the clay, and as proposed by those authors ahigher denitrification for the clay (Cannell et al. 1984).

Depth of waterlogging affects the degree of plant damage;for example, when the water level was at 0, 10 and 20 cm belowsoil surface, tillering of wheat was reduced by 62%, 45% and24% and adventitious root main axes length per plant by73%, 58% and 39%, respectively (Malik et al. 2001). Seminalroot growth can also increase in an oxic soil zone above thewater-saturated anoxic soil; for example, seminal root DM ofwheat waterlogged to 10 cm below the soil surface increasedby 50% as compared with plants with waterlogging to the soilsurface (Malik et al. 2001). The depth and duration ofwaterlogging at some field sites have been described usingSEW30 (sum of excess water that occurs daily in the top30 cm soil layer; Sieben 1964). Support for such an approachfor wheat is that growth and yield progressively decreased withwaterlogging duration (Sharma & Swarup 1988; Malik et al.2002; Olgun et al. 2008; Yaduvanshi et al. 2012; Marti et al.2015), and reoccurring waterlogging periods can show additiveeffects (Belford 1981; Belford et al. 1985). However, limitationsof using SEW30 have been pointed out (McFarlane et al. 1989;Malik et al. 2001; Setter & Waters 2003) because it does nottake into account temperature, flooding frequencies, possibledegrees of shoot submergence, susceptibility at different devel-opmental stages and differential recovery responses. AlthoughSEW30 has limitations, with the addition of temperature data(suggested by Setter et al. 2009), it provides a method to quan-tify and integrate durations and depths of soil waterlogging.

Although low O2 stress is the major cause of the growth re-duction of wheat roots (Trought & Drew 1980c), waterloggingalso results in other changes in soils that can be detrimental forroots, such as higher concentrations of Fe2+, Mn2+, ethylene,CO2 and organic acids (Ponnamperuma 1972; Ponnamperuma1984). The soluble metal ions Fe2+ and Mn2+ can increase topotentially toxic levels (Setter et al. 2009). High concentrationsof CO2 may cause pH to decline in root cells (Greenway et al.2006) and high ethylene can inhibit root extension (Huanget al. 1997a). Data on the effects of organic acids on wheatare lacking, but adverse effects of organic acids on especiallyK+

fluxes in roots of two barley (Hordeum vulgare) varietiesdiffered in magnitude (Pang et al. 2007), so this aspect shouldbe investigated also in wheat.

In conclusion, in order to facilitate interpretation ofwaterlogging experiments, details on soil type, waterloggingduration and depth, temperature and when possible other pa-rameters such as soil O2 status and/or redox potential shouldbe provided. In the following subsections, we consider the di-rect effects of O2 deficits on wheat roots, with focus on anoxia

Figure 1. The effects on wheat of waterlogging for shoot dry mass(median= 67% of controls, n= 46, Supporting Information Table S1),root dry mass (median = 38% of controls, n= 46, SupportingInformation Table S1) and grain yield (median = 57% of controls,n= 206, Supporting Information Table S2) as percentage of drainedcontrols. This summary figure was compiled from data extracted frompeer-reviewed literature (data values, key experimental conditions andreferences are in Supporting Information Tables S1 and S2). Boxes are50% of the observations with the median shown as the horizontal linewithin the box, and bars are 1 and 99 percentiles; outliers are shown as●. **** Significant differences from 100% (Wilcoxon signed rank test,P< 0.0001). Data are from experiments where wheat was waterloggedin soil (either in pots or in field situations, water level �5 to +8 cmrelative to soil surface) for 7–42 d without recovery (root and shootmass) or for 4–120 d for grain yield (most experiments with recovery).

2 M. Herzog et al.

© 2015 John Wiley & Sons Ltd, Plant, Cell and Environment20

tolerance and aerenchyma formation and internal O2 move-ment. We consider physiological mechanisms contributing towaterlogging tolerance and where available highlight geno-typic variation in wheat.

O2 deficiency adversely affects the roots of wheat

The effect of severe root hypoxia on wheat has been evaluatedby growing plants in nutrient solutions bubbled with N2 or de-oxygenated andmade stagnant with 0.1% (w/v) agar to preventconvective flows (i.e.Wiengweera et al. 1997; Colmer &Greenway 2011). Wheat in N2-flushed (Trought & Drew1980c; Barrett-Lennard et al. 1988) or stagnant (Watkin et al.1998) nutrient solutions for 10 to 14d developed symptoms re-sembling those of plants in waterlogged soils (Trought &Drew1980a, 1980b; Malik et al. 2001; Malik et al. 2002). The similar-ities were as follows: (1) reduced seminal root growth, (2)death of seminal root apical meristems, (3) growth of adventi-tious roots to a restricted maximum length only, (4) lowerroot:shoot ratio, (5) increased root porosity, (6) non-structuralcarbohydrate accumulation in root and shoot tissues, (7) re-duced shoot nutrient concentrations and (8) early senescenceof the older leaves. Recovery of growth (see Recovery fromTransientWaterlogging andGrain Yield section) upon transferof plants back to aerated nutrient solutions also somewhatmimicked plant responses to soil drainage with O2 re-entry(Malik et al. 2001), as initially growth mainly took place in theroot system rather than in the shoot, tissue sugar levels de-clined and adventitious roots elongated in order to restore amore balanced root:shoot ratio (Huang et al. 1994a; Huang &Johnson 1995). The effects on wheat of N2 flushing in nutrientsolution lead to the conclusion byTrought&Drew (1980c) thatduring early stages of waterlogging, the inadequate supply ofO2 to the roots could explain the arrest of root growth and to-gether with reduced nutrient uptake caused by root dysfunc-tion limits shoot growth of wheat (Trought & Drew 1980c;Buwalda et al. 1988a).

Effects of O2 deficiency on root respiration

As the O2 in roots declines, at some point so will respiration;the O2 concentration at which respiration first declines hasbeen termed the ‘critical O2 pressure’ for respiration (COPR).COPR has been interpreted as being determined by the affinityof cytochrome oxidase for O2 and its rate of consumption ofO2

and the structure of roots, which determines the diffusion ofO2

to all consumption sites, that is, to mitochondria in all cells(Berry & Norris 1949; Armstrong & Beckett 2011a). However,in the case of pea and Arabidopsis, it has been suggested thatrespiration is down-regulated as O2 levels decline (Zabalzaet al. 2009), but this view has been challenged and the difficul-ties of external solution measurements discussed in light ofdiffusion limitations to O2 reaching interior tissues/cells(Armstrong&Beckett 2011a, 2011b). For root tissues ofwheat,O2 uptake rate was stable until reaching the COPR in the me-dium of 12.8 kPa for 1mm apical tips and of 7.2kPa for seg-ments from 2–4mm behind the apex (Thomson et al. 1989);there was no evidence for down-regulation of respiration.

The values from Thomson et al. (1989) are based on O2 uptakerates from solution and therefore overestimate cellular COPR(Armstrong 1979; Armstrong et al. 2009).

An alternative approach to determining COPR is the moni-toring of radial O2 loss (ROL) from intact roots in an O2-freemedium during manipulations of the shoot O2 concentrations(Armstrong & Gaynard 1976). Using this ROL-based ap-proach, COPRwas found to be 2.1kPa at the surface of the rootelongation zone (2–7mm behind apex) of wheat seminal roots(Thomson et al. 1990), 4.0kPa for wheat adventitious roots(Barrett-Lennard et al. 1988) and 2.4kPa for rice (Oryzasativa) adventitious roots (Armstrong & Gaynard 1976). Thehigher value for adventitious roots of wheat when comparedwith rice was suggested by the authors to be caused by lowerporosity in the apex of wheat roots (i.e. diffusional limitationresulting in higher COPR). It has also been hypothesized thatspecies with roots having large stelar diameters should havehigher COPRs due to a longer diffusion path (i.e. stele radius)to the innermost cells of this tissue of relatively low porosityand high rates of O2 consumption, so that higher O2 in the cor-tex would be needed to meet this demand (Armstrong et al.2009). Cross sections of wheat adventitious roots have propor-tional stelar areas of 18–20% (Huang et al. 1994b;Wiengweera& Greenway 2004) compared with 5% in those of rice(McDonald et al. 2002), therefore potentially contributing tothe explanation of differences in COPR between wheat andrice. Screening of wheat root stele proportions could reveal ifgenotypic variation exists that could then be followed up withmore detailed measurements to determine COPR of poten-tially contrasting genotypes.

Anoxia tolerance of wheat roots

Without O2 supply, respiration ceases and anaerobic energymetabolism produces someATP. Survival in anoxia varies fromhours to months for plant species and organs (Jackson&Drew1984), with wheat roots being able to re-grow after 24h ofanoxia when hypoxically pre-treated (Waters et al. 1991a;Mustroph & Albrecht 2007). Survival of root apices, or oflateral initials, would allow seminal roots to resume growthwhen O2 is again available as waterlogging recedes (Waterset al. 1991b; Goggin & Colmer 2005; Goggin & Colmer 2007;Mustroph & Albrecht 2007).

Survival of root meristems during anoxia has been assessedby resumption of root elongation upon return from anoxic toaerated solutions. Seminal roots of wheat without hypoxic ac-climation die within 9 h of anoxia: 85% of roots with ‘anoxicshock’ did not resume elongation upon re-aeration (Waterset al. 1991b). By contrast, exposure to hypoxia for 15–30h in-creased anoxia tolerance as 100% of seminal roots retainedtheir elongation capacity after 24h anoxia (Waters et al.1991b; Goggin & Colmer 2005), or for some genotypes evenup to 72h anoxia (Goggin & Colmer 2007). The use of ahypoxic pre-treatment mimics better the gradual O2 declinein waterlogged soils and avoids an ‘anoxic shock’ (Gibbs &Greenway 2003). During hypoxia, the activities of pyruvatedecarboxylase (PDC) and alcohol dehydrogenase (ADH) in-creased by 2-fold to 4-fold and 3.5-fold to 17-fold and the rate

Waterlogging tolerance of wheat 3

© 2015 John Wiley & Sons Ltd, Plant, Cell and Environment 21

of ethanol production (and thereby ATP generation) in thesubsequent anoxia increased 1.4–4 times (Waters et al.1991b). Further investigation is needed to understand wheatgenotypic variation for seminal root tip survival in anoxia, thephysiology of which is considered further in the followingsubsections.

Fermentation rates during anoxia

When suddenly exposed to anoxia, wheat seedling root tipssurvived for about 24h at 15 °C and only 10h at 25 °C (Waterset al. 1991a, 1991b). This low anoxia tolerance has beenlinked to the relatively low rate of ethanolic fermentationin wheat roots of 10 μmol ethanol g�1 FMh�1 in 5mm roottips (Waters et al. 1991b) and 3–5 μmol ethanol g�1 FMh�1

in the expanded root zone; in both cases, exogenous glucosewas provided as substrate (Waters et al. 1991b; Goggin &Colmer 2007). These rates are relatively low compared withthat of fermentation in 5mm root tips of maize (Zea mays)(19 μmol ethanol g�1 FMh�1, Saglio et al. 1988) and riceseedling shoots (22 μmol ethanol g�1 FMh�1, Meneguset al. 1991). However, differences between species maymerely be due to differences in protein concentrations, sothe relatively low fermentation rates in wheat need to beconfirmed on a protein basis (Gibbs & Greenway 2003).Moreover, anoxia tolerance is determined by factors in ad-dition to rates of ATP production during glycolysis linkedto ethanol production, including the efficient utilization ofthe limited amount of energy available during anoxia(Atwell et al. 2015). Similar ethanolic fermentation rates(on a protein basis) in species showing very different anoxiatolerances support this view (Atwell et al. 2015). Similarly,11 wheat genotypes did not differ for ethanolic fermentationrate (excised roots supplied with glucose) despite theseshowing large variation in anoxia tolerance measured asretention of seminal root elongation potential after 72 hanoxia (Goggin & Colmer 2007). Factors other than fer-mentation rates, for example, substrate supply, efficientuse of available energy or down-regulation of energy re-quirements, must therefore determine genotypic differencesin anoxia tolerance in wheat roots.

Wheat roots subjected to hypoxic pre-treatment, rather thanto anoxic shock, had 1.4-fold to 4-fold higher ethanol produc-tion rates and greater anoxia tolerance (Waters et al. 1991b).Ethanolic fermentation in wheat roots is initially limited bylow activity of PDC (Waters et al. 1991b; Albrecht et al.2004), and hypoxic pre-treatment increased PDC activity18-fold during 72h of hypoxia (Albrecht et al. 2004). The in-creased anoxia tolerance following hypoxic pre-treatmentcould reflect the improved capacity for ATP production inglycolysis linked to ethanol production, as well as several otheracclimations to low O2 (cf. Greenway & Gibbs 2003).

Substrate supply

Non-structural carbohydrates (i.e. sugars) accumulate in shoots(e.g. 2-fold at 3 d of soil waterlogging) and roots (e.g. 4-fold at6 d in N2-flushed solution) of wheat during soil waterlogging

or when in O2-deficient nutrient solutions (Barrett-Lennardet al. 1988; Waters et al. 1991b; Albrecht et al. 1993; Huang &Johnson 1995; Malik et al. 2001, 2002; Mustroph & Albrecht2003, 2007). During root hypoxia, fructans increased both inroots and shoots of 10-d-old wheat seedlings, which has beensuggested to be an energy-efficient carbohydrate storagemechanism (Albrecht et al. 1993; Albrecht & Biemelt 1998).However, views differ on the issue of substrate supply forfermentation in anoxic wheat roots. Sugar supplies were calcu-lated to be ample for 24h of fermentation, but apices of anoxia-shocked roots only survived up to 9h anoxia and exogenousglucose during anoxia resulted in five times higher retentionof elongation potential (and two to three times higher ethanolicfermentation); hypoxic pre-treated roots already had high tol-erance of 9 h anoxia, so there was little benefit of exogenousglucose to those roots during the same time period (Waterset al. 1991b). Loss to the medium of sugars and other metabo-lites, and the possibility of decreased phloem transport towardsthe tips, was suggested as reasons that exogenous glucose wasbeneficial despite the starting levels within the seminal roots(Waters et al. 1991b); such solute loss can be substantial forwheat roots in anoxia (Greenway et al. 1992). Sugar accumula-tion during a hypoxic pre-treatment, in addition to increasedPDC and ADH activities, may increase ethanolic fermentationrates during subsequent anoxia (Albrecht et al. 2004) and en-hance survival in anoxia, but even hypoxic pre-treated root tipsof wheat died after 48h anoxia (Mustroph & Albrecht 2007).

Sugar transport to wheat roots is restricted by anoxia. Sugartransport from the endosperm/shoot was reduced by 79–97%when whole seedlings were subjected to anoxia (Waters et al.1991a). Root sugar concentrations of seedlings exposed to72h anoxia decreased but did not differ significantly among11 genotypes (Goggin & Colmer 2007). Photosynthetically in-corporated 14C fed to shoots was detected in roots in anO2-freenutrient solution for 4 d (Wiedenroth & Poskuta 1981), butwhether 14C was at the root tips was not determined. In maizeseedlings, phloem unloading to the apex was severely inhibitedby anoxia (Saglio 1985). The decrease in sugar concentrationsin anoxic roots contrasts with the earlier-mentioned increasein sugars in hypoxic seedling roots. Interestingly, in the 20mmtips of adventitious roots of two varieties, which differ inwaterlogging tolerance and exposed to 21d of hypoxia, sugarswere substantially higher in the tolerant variety (Huang &Johnson 1995). Whether wheat genotypes differ in phloemtransport during hypoxia and the influence on waterloggingtolerance should be assessed (cf. Boru et al. 2003, but directmeasurements are needed), and comparisons should ensurethat tips are alive to avoid measurements on tissues that haveleaked cellular contents as a result of death.

Cytoplasmic acidosis

Anoxia results in declining cytosolic pH in wheat roots, as inother plants, and if prolonged, this may lead to cell death(Ishizawa 2014).Wheat seedling root tips died within 10h whensubject to anoxia at pH4 whereas survival was more than 90%at pH5 and 6, indicating that low external pH would increasecytoplasmic acidosis during anoxia (Waters et al. 1991a). The

4 M. Herzog et al.


cause for the initial pH drop has been suggested to result frominitial lactate production, being 1.5μmol g�1 FMh�1 in wheatroots (Thomson et al. 1989;Mustroph&Albrecht 2007), but lac-tate production/accumulation and cytoplasmic acidification arenot necessarily always correlated (Felle 2005).Moreover, maizeroots show lactate efflux and a higher cytoplasmic pH afterhypoxic pre-treatment (Xia &Roberts 1994). Similarly, hypoxicpre-treated wheat roots had 2-fold higher efflux of lactateduring the first 3 h of anoxia than when given anoxic shock(Mustroph & Albrecht 2007). However, the tissue concentra-tion of lactate was only 0.5μmolg�1 FM in entire excised wheatroots (Mustroph & Albrecht 2007), and although it would beexpected to be higher in root tips, whether lactate is responsiblefor the observed pH decline in wheat roots remains to bedetermined.Cytoplasmic pH decreases in anoxic shoots and root tips

of wheat (0.8 and 0.6 pH units within 2 h, respectively)were greater than decreases of 0.5 and 0.2 pH units inshoots and roots tips of anoxia-tolerant rice (Meneguset al. 1991; Kulichikhin et al. 2007; Ishizawa 2014). Wheatcytoplasmic pH continued to decline by another 0.2 pHunits so that pH was then 6.8 after 6 h anoxia in root tips(Kulichikhin et al. 2007) and 6.5 in wheat seedling shoots(Menegus et al. 1991), contrasting with rice root tips forwhich cytoplasmic pH stabilized at pH 7.15 during anoxia(Kulichikhin et al. 2007). After prolonged anoxia, loss ofregulation of cellular pH might be a consequence of theenergy shortage and reflect dying cells, rather than impairedpH regulation per se being the cause of death (Felle 2005;Atwell et al. 2015).

Root solute loss

Death causes cellular solute loss, but solutes may also belost well before death occurs, for example, cation effluxdue to membrane depolarization in anoxia. Wheat seedlingroots lost K+, amino acids and sugars when subjected to an-oxic shock and, to a lesser degree, when anoxia was pre-ceded by a hypoxic pre-treatment (Greenway et al. 1992).Plasma membrane depolarization leads to opening ofvoltage-gated ion channels (Ward et al. 2009), and K+ netloss has been documented for wheat seedling roots(Greenway et al. 1992; Goggin & Colmer 2007). Investiga-tion of genotypic variation in wheat for root K+ retentionshowed that the four varieties showing the most anoxiatolerance had higher tissue K+ concentrations than the twoanoxia-intolerant varieties (Goggin & Colmer 2007). Forwheat roots of 26-d-old plants subjected to severe hypoxia(0.23 kPa O2) for up to 10 d, K+ loss was due to membranedepolarization rather than increased membrane permeabil-ity, as membranes under anoxia were depolarized butremained impermeable to sorbitol (Buwalda et al. 1988b).In addition to these effects of anoxia (or severe hypoxia)on solute loss from roots, organic acids in anaerobic soilscan also result in membrane depolarization and K+ efflux,as demonstrated for barley roots in aerobic nutrient solu-tion (Pang et al. 2007).

Aerenchyma formation in roots of wheat

O2 movement within roots is largely determined by tissue po-rosity (gas volume per unit tissue volume; Armstrong 1979).Root porosity is an important trait contributing to waterloggingtolerance of wheat (e.g. Setter & Waters 2003) and other spe-cies (e.g. Justin & Armstrong 1987). The constitutive porosityof wheat roots (i.e. porosity in aerated conditions) is, like mostdryland species, relatively low (median values of 2.1% and5.2% for seminal and adventitious roots, respectively, Fig. 2,references in Supporting Information Table S3) and reflectsthe hexagonal pattern of cell packing in the root cortex(Trought & Drew 1980c; Xu et al. 2013) resulting in smallgas-filled intercellular spaces (cf. Armstrong 1979; Malik et al.2003). The waterlogging-induced increase in wheat root poros-ity results from formation of lysigenous aerenchyma in the cor-tex (Erdmann & Wiedenroth 1986; Huang et al. 1994b;Yamauchi et al. 2014b) resulting in 14.8% porosity (medianvalue) of adventitious roots formed uponwaterlogging (Fig. 2).

The ability of seminal roots of wheat to form aerenchyma is re-lated to age and/or developmental stage (Supporting Information

Figure 2. Porosity (%, gas volume per unit root volume) in seminaland adventitious roots, and aerenchyma (%, cross-sectional area) inadventitious roots, of wheat grown in aerated or drained (control) orO2-deficient [2 studies in waterlogged soils and 12 studies in N2-flushed/deoxygenated nutrient solutions; waterlogging (WL)] conditions. Thissummary figure was compiled from data extracted from peer-reviewedliterature (data values, key experimental conditions and references arein Supporting Information Table S3). Boxes are 50% of theobservations with the median shown as the horizontal line within thebox, and bars are 1 and 99 percentiles; for porosity of seminal roots andaerenchyma of adventitious roots, the numbers of observations wereinsufficient for box-whiskers plots so each observation is shown with●.Letters denote significant differences between control and WL (one-way ANOVAwith Tukey’s multiple comparison test on log-transformeddata, P< 0.05, n= 15–17 for porosity of adventitious roots and 5–7 forseminal roots); n.s. denotes no significant difference between drainedand WL adventitious root aerenchyma (Mann–Whitney test,P= 0.556). Medians are 2.1% and 5.3% porosity for control and WLseminal roots, 5.2% and 14.8% porosity for control and WLadventitious roots, and 5.5% and 20% aerenchyma for control andWLadventitious roots, respectively.



Table S3). Seedling (5- to 7-d-old) seminal roots form aeren-chyma whereas older seminal roots lose this capacity (e.g. rootsof 18- to 36-d-old plants; Thomson et al. 1990, 1992; Xu et al.2013).More specifically, porosity of short, young seminal roots in-creased from 4–6% in aerated solution to up to 12% when instagnant conditions whereas porosity of long, old seminal roots(>100mm) hardly increased and was 3–7% in both conditions(Thomson et al. 1990). Interestingly, for adventitious roots in aer-ated solution with exogenous ethylene, shorter (100–200mm)roots formed aerenchyma whereas longer (>300mm) roots didnot (Huang et al. 1997a). In addition to this possible decrease inresponsiveness to ethylene in older/longer roots limiting aeren-chyma formation, when in an anaerobic medium, the distal por-tions of long roots of low porosity might not receive internalO2, which would result in damage or even death of tissues alsopreventing the opportunity for aerenchyma to form.

In contrast to seminal roots, adventitious roots form moreaerenchyma, and these newly produced roots are able toelongate at least to some extent into waterlogged soil orO2-deficient solutions (Thomson et al. 1992; Malik et al.2002). Adventitious root constitutive porosity ranges from3% to 7%, and during waterlogging, the porosity increases upto 11–20% by formation of aerenchyma depending on the ge-notype and O2 deficiency duration and treatment method(Thomson et al. 1990, 1992; Huang et al. 1997a; Wiengweeraet al. 1997; Malik et al. 2001, 2002, 2003 and Supporting Infor-mation Table S3). Waterlogging-tolerant genotypes had rootporosities of 20%, 14% and 11%, and waterlogging-sensitivegenotypes had root porosities of 8%and 6%, after 20d in occa-sionally N2-flushed nutrient solution (root types not stated,Boru et al. 2003), showing that genotypic variation for aeren-chyma formation is present in wheat. The importance of thistrait in potentially conferring waterlogging tolerance was illus-trated in a field study using 12 genotypes, as yield as percent ofcontrols during waterlogging was positively correlated withaerenchyma percentage of the mid-cortex (Setter 2000).

Aerenchyma formation in roots of wheat, like lysigenousaerenchyma in other species, results from the degradation ofcortical cells (Erdmann et al. 1986; Huang et al. 1994b; Maliket al. 2001) via programmed cell death (PCD) as described forwheat (Jiang et al. 2010) and more generally (Evans 2004;Shiono et al. 2008; Takahashi et al. 2014). Interestingly, exposureto O2 deficiency of only the apical portion of adventitious rootsis enough to trigger the development of aerenchyma along theentire main axis (Malik et al. 2003). As ethylene signallingtriggers lysigenous aerenchyma formation (Steffens & Sauter2014), including for wheat (Huang et al. 1997a; Yamauchi et al.2014a, 2014b), possible transport of 1-aminocyclopropane-1-carboxylic acid (ACC) and/or ethylene movement from thehypoxic tip back along the roots could underpin this response(Malik et al. 2003). Exogenous ethylene (1 to 5μLL�1) resultedin increased porosity of wheat roots in aerated nutrient solutionfrom <5% to 18% (Huang et al. 1997a). The promoting effectof ethylene on aerenchyma formation was greatest under highdoses (5μLL�1), more pronounced for newly formed shortroots (10–20 cm length) than pre-existing longer roots (30 cmlength) and of higher magnitude in a waterlogging-tolerantvariety than in a sensitive one (Huang et al. 1997a).

Details onmechanisms of aerenchyma formation and associ-ated signalling in roots have been reviewed (Steffens & Sauter2014). Recent work of interest in wheat, using seedling seminalroots, highlights the role of ROS as part of PCD in aerenchymaformation (Xu et al. 2013; Yamauchi et al. 2014b). ROS accu-mulation starts in the mid-cortex cells where PCD begins, ac-companied with up-regulation of some genes encoding forROS-producing enzymes (e.g.NADPH oxidase) and down-regulation of ROS-detoxifying enzymes (e.g. catalase (CAT);Xu et al. 2013). Pre-treatment of 5-d-old wheat seedlings withACC increased aerenchyma as well as the expression of threegenes encoding respiratory burst oxidase homolog (Yamauchiet al. 2014a, 2014b), which act by generating ROS, and anNADPH oxidase inhibitor partially suppressed the ACC-induced response.

Ethylene triggers other plant acclimations to flooding in ad-dition to aerenchyma formation (reviewed by Sasidharan &Voesenek 2015); of interest here is the initiation of adventitiousroots. The number of adventitious roots per wheat plant is typ-ically reduced by hypoxia or waterlogging but proportionallyless than the number of tillers per plant; therefore, adventitiousroot number per tiller increases (Huang et al. 1997b;Malik et al.2001, 2002). Greater adventitious root number per tiller pre-sumably assists the shoots to cope with the waterlogging-induced restrictions impacting on their roots (Malik et al.2003). Wheat shows genotypic variation for the formation ofadventitious roots, as 21 d of hypoxia (5% O2) decreased thenumber of adventitious roots in two waterlogging-sensitivevarieties by 17% and 37% from aerated controls, whereas thisincreased by 82% in a waterlogging-tolerant variety (Huanget al. 1994a). Wheat adventitious rooting in response toethylene is dependent on the genotype and ethylene concen-tration (Huang et al. 1997a). Production of adventitious rootsby a waterlogging-tolerant variety increased by 17% at allethylene concentrations (0.1, 1.0 and 5.0μLL�1), whereas ina sensitive variety, only the lowest ethylene concentration pro-moted these roots (Huang et al. 1997a).

Internal O2 supply determines root growth inwaterlogged soils

Root growth is arrested by soil waterlogging as a lack of soil O2

to support root respiration means a low energy supply cannotmeet the high energy demand of growth (Elzenga & van Veen2010). Internal O2 movement from shoots and along roots viaaerenchyma can then support root extension (Armstrong1979), but only a limited amount of O2 can reach the root api-ces. The restricted capacity for internal O2 movement withinroots of wheat was demonstrated by the increased adventitiousand seminal root extension rates when O2 around shoots wasraised from 21kPa to 80–100kPa and thus increasing internalO2 diffusion into and along the roots (Thomson et al. 1990;Wiengweera & Greenway 2004). Maximum lengths of wheatadventitious roots in waterlogged soil (11–14 cm, Thomsonet al. 1992; Malik et al. 2001) are limited by internal O2 supplyto the apex owing to low to moderate porosity of tissues, highrespiratory rates and/or high root ROL to the rhizosphere, all

6 M. Herzog et al.


of which in concert determine the length that roots can growwhen reliant on internal O2 movement to the apex (Armstrong1979). The ‘Root-Length Model’ by Armstrong (1979) hasbeen used to predict the maximum length of wheat roots inan anoxic medium and reliant on internal O2; for seminal roots(3.4% porosity) in N2-flushed nutrient solution and for adven-titious roots (22% porosity) in waterlogged soil, the lengthsachieved were 85% and 77% of those predicted by the model(Thomson et al. 1990, 1992; Malik et al. 2001). By comparison,rice roots reached the predicted lengths, which could be due todifferences between the two species in ROL from the basalzones of roots (high in wheat and low in rice; Colmer 2003);ROL is assumed to be zero in themodel.Wheat roots did reachthe predicted lengths in stagnant nutrient solution where ROLis lower than in soil (Watkin et al. 1998; McDonald et al. 2001).Root extension rates can initially be maintained in O2-defi-

cient media, but as O2 concentration at the root tip decreases,extension declines at the critical O2 pressure (COPE) and even-tually ceases (Armstrong & Webb 1985). Monitoring root ex-tension rates and the O2 concentration just behind the rootapex (2–7mm) when in anoxic medium, while manipulatingO2 levels around the shoot and thus at the root apex owing tointernal O2 movement along the roots, enables determinationof the COPE (Armstrong & Webb 1985). In wheat, COPE of5- to 7-d-old seminal roots grown in semi-stagnant solutionand transferred to deoxygenated agar medium for the COPEmeasurements was 2.68 kPa at the root surface (Thomsonet al. 1990), being 7-fold to 15-fold higher than similarly deter-mined COPE in rice roots (0.2–0.4 kPa, Armstrong & Webb1985). These contrasting values of COPE in wheat and riceare similar to the differences between these two species inCOPR (see Effects of O2 Deficiency on Root Respirationsection), so that the declines in root extension presumablyresult from diminishing respiratory activity below the COPR(Thomson et al. 1990). The higher COPR and COPE values inwheat roots than for rice might be related to insufficient O2

reaching the apex due to the low tissue porosity owing tohexagonal cell packing in wheat roots in contrast to higher tis-sue porosity of cubic packing of cells in rice roots.Studies of wheat root lengths as affected by waterlogging

showed a significant genotype× treatment interaction for rootlength density (five genotypes, Hayashi et al. 2013) and for lon-gest adventitious root length (seven genotypes, Dickin et al.2009), but unfortunately, root porosity and/or ROL were notmeasured. Maximal length was greater for adventitious rootsof higher porosity for two varieties compared in stagnantnutrient solution (Watkin et al. 1998). Waterlogging for 17din sterilized sand reduced total length of adventitious roots to64% of control in a variety with 30% cortical aerenchymaand to 42% of control in one with 19% cortical aerenchyma(Huang et al. 1994b). Seminal root lengths after 7 and 50dwaterlogging did not show a significant treatment× genotypeinteraction (seven genotypes, Dickin et al. 2009; 6 genotypes,Haque et al. 2012), which contrast with the genotypic variationfor adventitious roots.Aerenchyma formation substantially increases internal O2

diffusion, while consumption of O2 along the diffusion pathby respiration and ROL both decrease the O2 at the root tip

(Armstrong 1979). Roots of many waterlogging-tolerant spe-cies, such as rice, develop a barrier to ROL in the outer cortex(Armstrong 1979; Colmer 2003). By contrast, roots of wheathave substantial ROL from basal zones (Malik et al. 2003;Malik et al. 2011; Alamri et al. 2013), which decreases rootgrowth (Thomson et al. 1992). An approach aimed at improv-ing the tolerance of wheat to waterlogging is the transfer ofthe barrier to ROL from a wetland wild relative, Hordeummarinum, to wheat by wide hybridization and amphiploid pro-duction (Malik et al. 2011). Two H.marinum–wheat amphi-ploids had tight ROL barriers and two only moderate ROLbarriers (Malik et al. 2011). This work demonstrates the poten-tial to target specific traits from more stress-tolerant wild rela-tives of wheat, but the amphiploids had low fertility and thuslow grain production, so further breeding would be requiredto potentially produce a more waterlogging-tolerant and com-mercially viable (high-yielding) wheat.

How does O2 deficiency affect wheat rootfunctioning?

Nutrient uptake

Arrest in root elongation upon waterlogging is pronounced inwheat, thereby leading to a reduced soil exploration and re-duced surface area for uptake of nutrients (Elzenga & vanVeen 2010). In addition, nutrient uptake by wheat is decreasedper unit of root mass; for example, P uptake by seminal rootswas 0–16% of aerated controls and by adventitious roots itwas 31–73% (Barrett-Lennard et al. 1988; Kuiper et al. 1994;Wiengweera & Greenway 2004). The higher maintenanceof ion uptake by the adventitious than seminal roots in O2-deficient solution presumably is owing to the greater capacityfor internal aeration of the adventitious roots (Colmer &Greenway 2011).

Reduced ion transport capacity in O2-deficient roots low inenergy is likely due to reduced proton motive force and lessnegative membrane potential (Elzenga & van Veen 2010).The proton motive force is generated by plasma membraneH+ ATPases, but activity of these H+ ‘pumps’ is reduced whenadenylate energy charge declines (Armstrong & Drew 2002;Elzenga & van Veen 2010). In addition to reduced capacityfor nutrient uptake by root cells, transport from roots to shootscan also be reduced by the effect of an anoxic stele to reducexylem loading (Gibbs et al. 1998; Colmer & Greenway 2011).Cells in dense stelar tissues can experience more severe O2 de-ficiency than epidermal and cortical cells, as has been illus-trated by microelectrode profiling, for example, in excisedmaize roots (Gibbs et al. 1998) or intact barley adventitiousroots (Kotula et al. 2015). Such conditions would also be ex-pected in wheat roots and could reduce ion release from xylemparenchyma cells into xylem vessels (Colmer & Greenway2011; Kotula et al. 2015).

Indeed, soil waterlogging reduces the nutrient concentra-tions in the shoots of wheat (Trought & Drew 1980b; Sharma& Swarup 1988; Steffens et al. 2005). Similarly, shoot nutrientconcentrations are decreased when wheat is grown inN2-flushed or in stagnant nutrient solutions (Trought & Drew



1980c; Wiengweera & Greenway 2004); these results from nu-trient solutions support that root hypoxia restricting respirationcan reduce root nutrient uptake and/or xylem transport (Drew1983). In addition, waterlogging also alters nutrient dynamicsin soils. P availability may increase as Fe is solubilized (Elzenga& van Veen 2010). Denitrification causing losses of nitrate canoccur in waterlogged soils, including soils planted with wheat(Cannell et al. 1980; Trought & Drew 1982; Belford et al.1985; Webster et al. 1986; Hamonts et al. 2013), although toour knowledge, the amount of N lost has not been quantifiedfor wheat fields.

Root hydraulic conductivity

Structural and functional constraints in O2-deficient roots havebeen proposed as explanations for a disturbed water balance ofshoots of plants in waterlogged conditions. A reduced root:shoot ratio is common in wheat (and other dryland cereals)when waterlogged (Huang et al. 1994b; Malik et al. 2001; Maliket al. 2002) (Supporting Information Table S1) and implies alower water absorption surface area of roots in relation to thetranspiratory surface of leaves, which together with impairedroot hydraulic conductivity (Bramley & Tyerman 2010) can re-sult in plant wilting, particularly under high evaporative de-mand. A detailed study on water transport in wheat rootsshowed that short-term hypoxia (30min at 4 kPa) reduced hy-draulic conductivity of root cortical cells by 45% and tran-siently reduced the conductivity of entire roots, possibly dueto decreased opening of aquaporins (Bramley et al. 2010). Dur-ing long-term waterlogging events, root cell death, xylemblockage and changes in root anatomy, for example, for wheat,cell wall lignification (Erdmann et al. 1986) and decreased xy-lem diameter in adventitious roots (Huang et al. 1994b), couldeach affect root hydraulic conductivity (Bramley & Tyerman2010). Leaf water potential declined for two wheat varieties af-ter 17d of waterlogging from �0.54 and �0.79MPa to �1.02and �1.10MPa (Huang et al. 1994b). However, leaf waterpotential in six wheat genotypes was unaffected after 7 d inhypoxic nutrient solution (Huang et al. 1994a). Decreasesin stomatal conductance of wheat in waterlogged soil or inO2-deficient nutrient solution can be substantial; 18% to 60%of controls (Trought & Drew 1980a; Huang et al. 1994b;Musgrave 1994; Malik et al. 2001; Zheng et al. 2009; Li et al.2011; Shao et al. 2013; Wu et al. 2014) and as discussed in thePhotosynthesis Decreases due to Feedback fromAccumulatedSugars section can be related to reduced growth and possiblefeedback of high leaf sugars and not necessarily adverse plantwater relations.

Summary of effects of O2 deficiency on roots ofwheat

Soil waterlogging severely inhibits seminal root growth andnutrient uptake, whereas new adventitious roots containingaerenchyma grow and enable some nutrient uptake, albeitreduced in comparison with O2-sufficient roots. Anoxiatolerance and survival of seminal roots would presumably

benefit wheat subjected to short transient waterlogging; thefew available data indicate some genotypic variation in seminalroot survival during up to 3d of anoxia. Ethanolic fermentationenables production of some ATP, which would be of impor-tance, but anoxia tolerance appears to be related to additionalaspects of metabolism, which requires further elucidation.Wheat genotypes differ in both the numbers of adventitiousroots formed and amount of aerenchyma within these roots.Aerenchyma enhances internal movement of O2 into andalong the roots, but O2 becomes limited at the apex of rela-tively short roots due to moderate porosity and high rates ofROL from basal zones. The low porosity but high O2 demandin the stele and root tips could limit both growth and nutrienttransport. Importantly, apices of many adventitious roots cansurvive at least for several days after root extension stops, so re-covery growth can occur when transient waterlogging recedes.

IMPACT OF WATERLOGGING ON SHOOTS

Having considered in the previous text the direct effects ofsoil waterlogging on wheat roots, in this section, we examinethe consequences for shoots in terms of growth, photosyn-thesis, sugar and nutrient status and possible toxicity ofmicroelements.

Waterlogging reduces wheat shoot growth

Waterlogging generally reduces shoot growth of wheat, andmedian shoot DM is reduced to 67% of drained controls(Fig. 1). The high variability in the reduction of shoot DM inFig. 1 reflects different conditions of waterlogging: (1) depth,(2) duration, (3) temperature, (4) soil type, (5) mineral nutri-tion regime, (6) plant age/developmental stage and (7) geno-type (key experimental conditions and references are inSupporting Information Table S1). The reduction of shootgrowth results from less tillering and reduced rates of leafgrowth and smaller leaf size. Nitrogen (N) deficiency is onelikely cause of the reduced tillering and slower growth (Belfordet al. 1985; Robertson et al. 2009). The reduced rate of leaf areaproduction, as well as early senescence of basal leaves (Trought& Drew 1980a; Malik et al. 2001; Malik et al. 2002), would re-duce the area available for light interception.

Genotypic variation for shoot growth is evident for wheat inexperiments with side-by-side comparisons in waterlogged soil,for example, shoot DM 70% and 40% of controls in a tolerantand sensitive variety, respectively (Huang et al. 1994b), and inhypoxic nutrient solution, for example, after 21d, shoot DMof 46–70% of controls in six varieties (Huang et al. 1994a). Ascreening of 37 genotypes revealed shoot growth range of27–63% of drained controls on an acidic soil waterlogged for49d (Setter et al. 2009). The yield of wheat when waterloggedhas been shown to correlate with shoot DM, which is alsoclosely related to plant tiller number (Musgrave & Ding 1998;Collaku & Harrison 2002; Hayashi et al. 2013).

Shoot DM of wheat in waterlogged soil can increase duringthe first few days above that of drained controls as a result ofaccumulation of photosynthates (Trought & Drew 1980a;Supporting Information Table S1). This accumulation of sugars

8 M. Herzog et al.


in the shoots of wheat (Trought & Drew 1980a, 1980c; Barrett-Lennard et al. 1988; Albrecht et al. 1993; Huang & Johnson1995; Malik et al. 2002) means that sugar production exceedsconsumption/export, which is likely the consequence of ahypoxic root system demanding less substrate due to cessationof growth and/or decreased ability for phloem transport in thehypoxic roots.

Photosynthesis decreases due to feedback fromaccumulated sugars

Waterlogging reduces net photosynthesis (PN) in wheat, al-though with significant variations in both severity (Fig. 3a;references in Supporting Information Table S4) and also timeof onset. We will consider separately short-term and longer-term responses to soil waterlogging of PN in wheat, as thecauses of these declines in PN likely differ. Possible causes ofthe initial decline in PN could be stomatal closure limiting inter-nal CO2 concentration (Ci), but negative feedback from carbo-hydrate accumulation could also decrease PN within the firstdays. In the longer term, N deficiency could also decrease leafphotosynthetic capacity. Numerous studies have documentedthat sugars accumulate in wheat shoots, stems and roots whenwaterlogged (Trought & Drew 1980a; Malik et al. 2001, 2002)or when in hypoxic solutions (Trought & Drew 1980c;Barrett-Lennard et al. 1988; Albrecht et al. 1993; Huang &Johnson 1995; Mustroph & Albrecht 2003, 2007), indicatingthat reductions in PN during short-term waterlogging are notlikely to be the cause of reductions in wheat shoot growth, astissue sugars remain high.

The different mechanisms leading to reduced PN are illus-trated by correlating ΔPN [PN(waterlogged)�PN(control)]against Δ stomatal conductance [gS(waterlogged)� gS(control)],during short-term (1–3 d) and long-term (7–21 d)waterlogging of wheat (Fig. 4; references in Supporting Infor-mation Table S5). PN was significantly correlated with gS bothin short-term (r=0.79) and long-term waterlogging (r=0.46);however, the correlation was weaker during the long-termwaterlogging. Short-term effects of waterlogging on PN couldbe due to ‘physiological drought’ in waterlogging-sensitivespecies, which can even wilt, and gs would decrease toconserve water (Bramley & Tyerman 2010), resulting also indecreased intercellular CO2 (Ci) and lower PN. However,experiments on wheat (Malik et al. 2001; Wu et al. 2014) andother species (see Else et al. 2009) concluded that Ci increaseduponwaterlogging; that is, PN was not limited by gs. Our calcu-lations of Ci for wheat during waterlogging on the basis ofpublished PN and gs data (Fig. 3a,b) show that during short-term (1–3 d) waterlogging, Ci is maintained at close to that inleaves of controls (with some exceptions where values declinedown to 80% of controls) and Ci increased during thelong-term waterlogging to above control levels (304 and252μmolCO2mol�1, respectively), supporting that in mostcases, the reductions in PN for wheat in waterlogged soil werenot the result of reduced Ci. Thus, declines in gs were likely aresponse to increased Ci. During long-term waterlogging,factors such as decreases in chlorophyll (Fig. 3c) or other com-ponents of the photosynthetic apparatus as a result of Ndeficiency and/or negative feedback from carbohydrate accu-mulation would be likely causes of reduced PN, although insome situations, possible damage to leaves potentially from

Figure 3. Net photosynthesis (PN, a), intercellular CO2 (Ci, b) and total chlorophyll (c), in youngest leaf or flag leaf of wheat in waterlogged soil, aspercent of drained controls; PN (n = 65) during 1–34 d of waterlogging, Ci (n= 23 and 24) and chlorophyll (n= 17 and 36) during short-term (1–3 d) orlong-term (7–21 d) soil waterlogging. This summary figure was compiled from data extracted from peer-reviewed literature (data values, keyexperimental conditions and references are in Supporting Information Table S4). Boxes are 50% of the observations with the median shown as thehorizontal line within the box, and bars are 1 and 99 percentiles. Median values of Ci of controls and short-term and long-termwaterlogging were 251,252 and 304μmolmol�1, respectively. Ci was estimated as Ci = 385μmolmol�1 (as atmospheric CO2)� (PN/gs) (Long & Bernacchi 2003). Differentletters indicate significant differences between medians of short-term versus long-term waterlogging (Mann–Whitney test, P< 0.05). ** Significantdifferences from 100% (Wilcoxon signed rank test, P< 0.01).



ROS or phytotoxins (Fe2+ or Mn2+) might also contribute.Interestingly, high nutrient application reduced the negativeeffect of waterlogging on PN in two experiments (Huanget al. 1994b; Wu et al. 2014). The positive effect on PN of im-proved mineral nutrition supports that leaf N deficiency inconcert with feedback from increased sugar concentrations re-presses PN during waterlogging (Paul & Driscoll 1997). Whenmeasuring the effect of waterlogging on PN, Ci values shouldbe provided in order to aid interpretations.

Declines in PN when wheat is waterlogged are highlyvariable (Fig. 3a), and differences in environmental factors(e.g. temperature and evaporative demand) would contrib-ute to this large variation. Studies designed to evaluategenotypic differences showed less variation, such as the fol-lowing: (1) Musgrave (1994) was unable to detect awaterlogging × genotype interaction in eight genotypes,and (2) waterlogging-tolerant varieties only showed slightlyhigher PN rates than sensitive varieties, that is, PN relativeto drained controls was 87% versus 76% and 89 versus86% in two pairs of waterlogging-tolerant and waterlogging-sensitive varieties, respectively (Huang et al. 1994b; Zhenget al. 2009).

Waterlogging adversely affects shoot nutrientlevels in wheat

One of the first visible signs of waterlogging stress is acceler-ated leaf chlorosis (Fig. 3c) preceding early leaf senescence(Trought & Drew 1980a; Samad et al. 2001). The earlyyellowing of basal leaves of wheat during waterlogging hasbeen linked to the remobilization of N from old to new leaves(Trought & Drew 1980b; Stieger & Feller 1994a). Our meta-analysis of previously published data (references in SupportingInformation Table S6) on shoot nutrient concentrations forwheat inwaterlogged soils highlights N deficiency as a key issue(Fig. 5a). Decreases in shoot nutrient concentrations withwaterlogging duration are due to both decreased nutrient accu-mulation in the shoot (less uptake and transfer from the roots)and continued shoot growth, which results in some ‘dilution’ ofthe nutrients already present through increased shoot size.

The impact of nutrient deficiency on wheat in waterloggedsoil, or hypoxic nutrient solutions, has been tested by increas-ing the availability of nutrients, and reviewed elsewhere(Colmer & Greenway 2011). Although shoot growth respondspositively to increased nutrient supply, DM remained lowerthan aerated controls even when wheat was grown inN2-flushed solutions with high nutrient concentrations(Barrett-Lennard et al. 1988), so other factors (e.g. regulationof the root:shoot ratio via hormonal changes in the plant) mustalso limit shoot growth of these plants with impaired rootgrowth (Colmer & Greenway 2011).

Possible toxicity to shoots of microelements

Observations of responses to soil waterlogging of wheat geno-types characterized as waterlogging-tolerant in Mexico (VanGinkel et al. 1991; Boru et al. 2001) but not in Australia(Condon 1999; Setter et al. 1999) raised questions of how soilchemistry affects waterlogging tolerance in wheat (Setter et al.2009; Yaduvanshi et al. 2012). Some authors have argued for in-creased evaluation of ion toxicity tolerances and breeding fortolerance to Fe2+ and Mn2+ to improve growth of wheat insome waterlogged soils (Setter et al. 2009; Shabala 2011;Khabaz-Saberi et al. 2012). Increased concentrations of Fe2+

and Mn2+ in the shoots of wheat during waterlogging have beenreported (Stieger & Feller 1994a; Khabaz-Saberi et al. 2005;

Figure 4. Changes in photosynthesis (ΔPN) versus changes instomatal conductance (ΔgS) upon waterlogging of wheat for short(1–3 d, n= 26, a) and long (7–21 d, n= 44, b) durations. This summaryfigure was compiled from data extracted from peer-reviewed literature;the numbers next to each data point refer to the data source (datavalues, key experimental conditions and references are in SupportingInformation Table S5). Spearman rank correlation coefficients andP values are presented in each panel.

10 M. Herzog et al.


Setter et al. 2009; Khabaz-Saberi et al. 2010a; Khabaz-Saberiet al. 2012), but others have failed to confirm microelementtoxicity (diagnosed by shoot concentrations) prevailing inwheat during waterlogging (Trought & Drew 1980a,1980b; Stieger & Feller 1994b; Steffens et al. 2005). Shoot

concentrations of other microelements either showed negli-gible response to waterlogging (Cu, Zn and B) or, to ourknowledge, have not been investigated for wheat (Ding &Musgrave 1995; Khabaz-Saberi et al. 2005; Setter et al.2009). In the case of strongly acidic soils that become

Figure 5. Shoot nutrient concentrations in wheat during soil waterlogging (WL). (a) N (n= 35) and K (n= 11) and (b) P (n= 11), Mg (n= 6) and Ca(n= 11) in wheat (28 genotypes) after waterlogging in 11 different soils. In (a) and (b) boxes are 50%of the observations with themedian shown as thehorizontal line within the box, and bars are 1 and 99 percentiles. Red horizontal lines display the critical whole shoot value for nutrient deficiency inwheat at Feeke’s scale 5–7; plant age was 23–155 d at sampling (Snowball & Robson 1991; Reuter 1997). The median concentration of N [15.8mg g�1

dry mass (DM)] is significantly lower than the critical deficiency level (37mg g�1 DM) (Wilcoxon signed rank test, *P< 0.05; ****P< 0.0001). Dataare from experiments in pots of soil or in field situationswith waterlogging periods of 7–49 d and shoots sampled duringwaterlogging (recovery periodsnot included). In (c), shoot concentrations ofmicronutrientsMn (n= 65, median = 135, grey line; these median values were significantly lower than thetoxicity threshold, P< 0.0001) and Fe (n= 65, median= 209, grey line) of wheat in drained soil and after waterlogging for 2–86 d in 16 different soils.Closed symbols indicate experiments conducted on acidic soils (pHCaCl2< 4.8), open symbols indicate pHCaCl2> 4.8. Different letters denote significantdifferences (P< 0.05) between shootmicronutrient concentrations in controls andWL conditions (Mann–Whitney test). The blue horizontal line displaysthe critical value forMn toxicity in whole shoots, 37 d after sowing (Reuter 1997). Plant agewas 26–70 d at sampling. No critical value for Fe toxicity couldbe found for wheat, but upper sufficiency level is 100 and 300mgkg�1DM forwhole shoots and young leaves, respectively (Mills et al. 1996;Reuter 1997).Mn concentration median is significantly lower than critical levels for toxicity (Wilcoxon signed rank test, P< 0.05). This summary figure was compiledfrom data extracted from peer-reviewed literature (data values, key experimental conditions and references are in Supporting Information Table S6).



waterlogged, genotypic differences in Al tolerance are ofimportance for growth of wheat (Khabaz-Saberi & Rengel2010; Khabaz-Saberi et al. 2012).

Our meta-analysis of published shoot concentrations of Mnand Fe for wheat inwaterlogged versus drained soils shows thatshoot Mn concentrations did not increase significantly fromdrained controls (Fig. 5c; references in Supporting InformationTable S6); some studies even found decreases in shoot Mn(Trought & Drew 1980b; Steffens et al. 2005; Khabaz-Saberi& Rengel 2010) contrasting with others that showed Mn accu-mulation (Khabaz-Saberi et al. 2005; Setter et al. 2009). Mn andFe accumulated to high levels mainly when wheat became wa-terlogged in strongly acidic soils (pHCaCl2 = 4.1–4.8).

Possible adverse effects of microelements in shoots might beassessed from tissue concentrations where critical values fortoxicity have been established for tissues and species (Reuter1997). For wheat, Reuter (1997) lists critical values for Mn tox-icity from several sources (range of 160–1100mgkg�1 DM) butno critical value for Fe toxicity in wheat. The wide range ofMncritical toxicity values highlights the uncertainty of this ap-proach. In our meta-analysis, we tested published shoot Mnconcentrations against a toxicity threshold of 380mgkg�1

DM, which is the toxicity value (Reuter 1997) for 37-d-oldplants (Mn concentrations in Fig. 5c are from 26- to 70-d-oldplants) and this value according to Reuter (1997) also summa-rizes the array of published tissue tests available for wheat andhence is a suitable benchmark to diagnose shoot nutrient disor-ders at mid-late tillering stage. Our analysis shows that the Mntoxicity threshold is only reached on very few occasions, whichwere for wheat when waterlogged on strongly acidic soils, andthat the median of shoot Mn concentrations is significantlylower than the toxicity threshold (Fig. 5c). Interestingly, Setteret al. (2009) considered 100mgFekg�1 DM to be toxic forwheat, but the basis of this value was unclear, as Reuter(1997) does not list a Fe toxicity value for wheat. Thus, shootcritical values for toxicity of Fe in wheat need to be establishedin order to further evaluate the shoot Fe concentrations com-piled in Fig. 5c.

A strong case for the importance of microelement toxicitiesin wheat during 40d of waterlogging on strongly acidic soils(pHCaCl2 = 4.1–4.3) was that varieties tolerant to high Fe andMn (and also Al) had 32–53% higher relative shoot DMcompared with regular varieties (Khabaz-Saberi & Rengel2010) and subsequently also yielded better (Khabaz-Saberiet al. 2012). However, the better shoot growth of themicroelement-tolerant varieties on waterlogged acidic soilsin Khabaz-Saberi et al. (2012) contrasts with the samemicroelement-tolerant and microelement-intolerant varietiesevaluated in waterlogged acidic soils for which no differencewas found (Setter et al. 2009). Here, Mn-tolerant Norquay(Khabaz-Saberi et al. 2010b) waterlogged for 49d in astrongly acidic soil (pHCaCl2 of 4.7) showed similar intermedi-ate waterlogging tolerance as Mn-intolerant Columbus (seefig. 2 in Setter et al. 2009), so Mn tolerance does not necessar-ily determine waterlogging tolerance even on acidic soil, orperhaps, other toxicities (e.g.Al or Fe) were affecting shootgrowth in this study (Setter et al. 2009). That shoot Feconcentration during waterlogging can be a poor indicator of

microelement stress is reflected by Fe tolerant varieties havinghigher Fe shoot concentrations than Fe-intolerant ones(Khabaz-Saberi & Rengel 2010; Khabaz-Saberi et al. 2012).

In summary, microelement toxicity likely impacts on wheatwhenwaterlogged in some acidic soils, but the shootmicronutri-ent data available indicate that the problem is secondary to themain effect of O2 deficiency impeding root growth. This conclu-sion is limited by the apparent lack of reliable shoot Fe criticalconcentrations for toxicity in wheat. Furthermore, whethertolerance to high soil microelements during waterlogging isgoverned by ‘exclusion’ from the roots and shoots and/or by‘tissue tolerance’ of high concentrations within cells cannot beconcluded at present and requires additional research.

Summary

Waterlogging reduces shoot growth of wheat, reflecting theconstraints on roots. The large variation in growth reductions(Fig. 1) results from variable conditions (see EnvironmentalParameters Influencing Wheat Waterlogging Responsessection), but also genotypic differences in tolerance ashighlighted within some experiments. Waterlogging results inshoot N deficiency (Fig. 5a), which likely contributes to re-duced growth. Tissue sugars often increase, despite reductionsin PN and reduced leaf area, as sugar consumption declinesowing to slow root and shoot growth. Ci in leaves of water-logged wheat is often similar to, or even exceeds, levels indrained controls, indicating non-stomatal limitations determinerates of PN, including possible down-regulation by high sugars.Direct adverse effects ofmicroelements on shoot tissues appearto occur only for wheat in some acidic soils when waterlogged.

RECOVERY FROM TRANSIENT WATERLOGGINGAND GRAIN YIELD

Recovery of wheat seminal and adventitious rootgrowth following waterlogging

Most experiments investigating the effects of waterlogging onplants have not included a recovery period (Malik et al. 2002;Striker 2012), with a few exceptions and also several evalua-tions of crop grain yields following transient waterlogging.Recovery of wheat involves allocation of carbon to roots afterwaterlogging (Malik et al. 2001, 2002) and hypoxia (Arakiet al. 2012b) for preferential root growth to reestablish a root:shoot ratio typical of plants in drained soil (fig. 7 in Malik et al.2002). This preferential resource allocation to root growthwould be a major reason explaining the reduced shoot growthfollowing a period of waterlogging (Malik et al. 2002; Robertsonet al. 2009). The bulk of the recovery growth is by adventitiousroots (main axes and laterals) that resume extension, asseminal root apices (and much of the other tissues of seminalroots) can die within 3–4 d of waterlogging (Malik et al. 2002).

Recovery of the seminal root growth after 3d of waterloggingresulted in structural DM during 25d of recovery reaching amodest one-third to one-half of that of drained controls, andthere was no recovery growth of seminal roots after 14d ofwaterlogging (Malik et al. 2002). This was related to the death

12 M. Herzog et al.


of the apical meristems at the end of the low porosity seminalroots, and the low ability to form new lateral roots from the sur-viving basal portions of the seminal roots, as illustrated also bypruning of distal portions of seminal roots of 13- to 20-d-oldwheat in aerated nutrient solution (Malik et al. 2002). Bycontrast, new laterals did form for 16-d-old wheat after 10dgrowth in N2-flushed nutrient solution and return to aerated so-lution (Barrett-Lennard et al. 1988). As temperature and plantage were similar, the difference could be due to quiescence oflateral initials during the period ofO2 deficiency, whichwas thensubsequently released, or to the different genotypes used and isan interesting parameter to study further in a greater number ofwheat genotypes. A study comparing seminal root lateral for-mation in 11 genotypes after 72h of anoxia showed that numberof laterals per cm main root varied from 0.28 to 2.61, and thatmost laterals were formed on roots that had lost their elongationpotential during anoxia, presumably reflecting the loss of apicaldominance, and thus correlated negatively with anoxia andwaterlogging tolerance (Goggin & Colmer 2007). Recoverygrowth was also evident when six wheat genotypes were grownin hypoxic (5%O2) nutrient solutions for 21d followed by 7d ofre-aeration; although total length of laterals was reduced signif-icantly in all genotypes during the hypoxia, this parameter re-covered to 80% to 100% of control levels after 7 d re-aeration(Huang et al. 1994a). Recovery growth, however, did not occurfor two wheat genotypes in stagnant agar solution for 14d; roottips became flaccid and brown, and the seminal root FM did notincrease upon re-aeration for 7d (Watkin et al. 1998). Thegenerally poor re-growth of the seminal roots of wheat high-lights the importance of adventitious root growth after watersubsides. This was further supported by Kuiper et al. (1994)showing that adventitious roots’ contribution to nutrient uptakeincreased dramatically during hypoxia and recovery, comparedwith the seminal root system.In contrast to seminal roots losing their ability to re-grow

within days of waterlogging, the tips of adventitious rootsremained alive and resumed extension upon re-aeration after21 d of waterlogging (Malik et al. 2002), 14d in stagnantdeoxygenated agar (Watkin et al. 1998) or 10d in N2-flushednutrient solution (Barrett-Lennard et al. 1988), most likelydue to a continued aeration of the roots tips via aerenchymaconnected to the shoot (see Internal-O2 Supply DeterminesRoot Growth in Waterlogged Soils section). The continuedgrowth/survival of the adventitious roots resulted in a greaterproportion of root mass than the seminal roots after 14d ofwaterlogging, and this difference was even more pronouncedafter 14 d of recovery (Trought & Drew 1980a).During waterlogging or nutrient solution hypoxia, sugars

(especially fructans) accumulate in wheat shoots and roots(see Substrate Supply section in Anoxia Tolerance of WheatRoots section). When transferred back to aerated conditionsfrom O2-deficient nutrient solutions, wheat roots (measuredfor adventitious roots or for entire root systems) showedincreased respiration rates compared with both continuouslyaerated controls and the previous anoxic period (Albrecht &Wiedenroth 1993, 1994; Huang & Johnson 1995; Boru et al.2003; Araki et al. 2012b); for example, respiration of 20mm ad-ventitious root tips increased in one variety to 32% above

controls following 21d of hypoxia (Huang & Johnson 1995).The increase in respiration of roots upon return to aerated con-ditions is most likely due to increased biosynthesis, supportedby consumption of the stored carbohydrates (Albrecht et al.1993). Much of this substrate seems to be used upon root re-growth (Albrecht & Wiedenroth 1993; Malik et al. 2002), asroot relative growth rates can be 30–40% higher after hypoxiacompared with plants in continuous aeration/well-drained con-ditions (Barrett-Lennard et al. 1988; Malik et al. 2001). Photo-synthetic rates of wheat were able to recover within 3–14dupon soil drainage and re-entry of O2 into the root zone (Maliket al. 2001; Shao et al. 2013), but recovery of shoot growthmightbe limited by N supply (see Nutrient Uptake section in HowDoes O2 Deficiency Affect Wheat Root Functioning section).

Reactive oxygen species defense systems in wheatroots during re-aeration

When water recedes and O2 re-enters the soil and root tissues,the formation of ROS (e.g. superoxide, hydroxyl radicals,singlet oxygen and hydrogen peroxide) might especiallydamage cell membranes. ROS production is a consequence ofa low energy charge, high level of reducing equivalents andsaturation of the mitochondrial electron transport chain,favouring electron leakage to O2 upon re-aeration (Blokhinaet al. 2003). ROS can be detoxified by defense mechanisms in-cluding enzymes (e.g. superoxide dismutase (SOD)) and anti-oxidants (e.g. ascorbate (ASA) and glutathione (GSH)). Theeffect of O2 deficiency on the antioxidant system is species-specific (Blokhina et al. 2003); in the following paragraph, theresponses in wheat roots are discussed. The recovery of exten-sion of adventitious roots near their maximum lengths (i.e.withvery low O2 at the tips during waterlogging) indicates that, ifROS increased during re-aeration, any damage was either con-trolled or repaired as growth resumed.

In wheat seedling roots subject to re-aeration upon 3–8d ofO2 deficiency, reduced glutathione (GSH) and ascorbate(ASA) became oxidized to dehydroascorbate (DHA) andoxidized glutathione (GSSG), pointing to increased oxidativestress (Albrecht & Wiedenroth 1994; Biemelt et al. 1998;Blokhina et al. 2000) as did excised roots subjected to cyclesof 16 h anoxia/8h re-aeration during the third period of anoxia,defined as a GSH:total glutathione factor below 0.9 (Goggin &Colmer 2005). Duration of O2 deprivation prior to re-aerationinfluences the degree of ROS production and plant responsebecause of the following: (1) root H2O2 levels upon re-aerationincreasedwith duration (2, 4 and 8d) of the previous anoxic pe-riod (Biemelt et al. 2000), (2) three cycles of 16h anoxia werenecessary for the glutathione pool to become significantly re-duced in roots (Goggin &Colmer 2005) and (3) lipid peroxida-tion in roots occurred after 6 d of anoxia (Albrecht &Wiedenroth 1994) in contrast to rather short (16 h) anoxia pe-riods (Goggin & Colmer 2005). Wheat roots exposed toprolongedO2 deficiency showed reduced, or at best similar, ac-tivity of ROS scavenging and antioxidant replenishing enzymes(glutathione reductase, GR; ascorbate peroxidase, APX; CAT;SOD) during the anoxia period (Albrecht &Wiedenroth 1994;



Biemelt et al. 1998, 2000; Goggin & Colmer 2005), with someincreased activities (e.g.APX and GR activities) during re-aeration (Biemelt et al. 2000; Goggin & Colmer 2005). Theseexperiments used sudden shifts from anoxia back to aeration,whereas soil O2 might increase more gradually upon drainage,for example, 10 d for redox potential to increase to ≥400mV(Thomson et al. 1992; Malik et al. 2001). So, whether the sud-den changes in controlled experiments possibly cause higherROS formation than might occur under a slower change inO2 supply should be evaluated. Hypoxic pre-treatment priorto anoxia can induce theROS defense system, such as a 2.5-foldincrease in total glutathione whereas it declined to 50%during anoxia (Goggin & Colmer 2005). To our knowledge,ROS defense systems have not been compared among asubstantial set of wheat genotypes when exposed to low O2

stress and/or re-aeration, so potential variation in this aspectremains to be assessed.

Effects of transient waterlogging on grain yield ofwheat

Waterlogging tolerance is by agronomic definition a highgrain yield under waterlogged conditions or following a tran-sient waterlogging and could refer to absolute yields or be rel-ative to control plants/plots in non-waterlogged conditions.Our meta-analysis of data on the effect of waterlogging onyield of wheat shows a median yield depression to 57% ofcontrols (Fig. 1, references in Supporting Information TableS2). In some cases, a low grain yield reduction due towaterlogging (15–27% reduction of drained controls) canmerely reflect low yield potential under drained conditions(e.g. 3 of 15 varieties tested by Collaku & Harrison 2002),making such genotypes of low priority for growers but poten-tially valuable for breeding if the tolerance has a physiologicalbasis (Collaku & Harrison 2002). Because shoot growth is pri-marily affected through reduced tillering, waterlogging re-duces the number of spikes per square metre (Watson et al.1976; Cannell et al. 1984; Belford et al. 1985). However, tillerproduction during recovery may alleviate this effect ifwaterlogging does not occur too late and N is available forrecovery growth (Cannell et al. 1984; Robertson et al. 2009). Be-sides a reduced number of spikes per squaremetre, waterloggingcan also reduce grain number per spike and individual grainweight, all these reducing grain yield (Araki et al. 2012a; deSan Celedonio et al. 2014).

The timing of waterlogging determines which yield compo-nents are affected themost inwheat.Waterlogging at early veg-etative stages reduces tillering and hence spikes per plant,waterlogging at anthesis stage can reduce numbers of floretsand hence grains per spike andwaterlogging during grain fillingcan reduce weight of individual grains (Belford et al. 1985;Sayre et al. 1994; Araki et al. 2012a; Powell et al. 2012; Shaoet al. 2013; de San Celedonio et al. 2014; Marti et al. 2015). Asimilar pattern of plant stage and the yield component affectedhas been reported for the effect of drought stress on wheat(Powell et al. 2012). Short waterlogging periods of 1–3dcan result in long-term detrimental effects on both growth

(Malik et al. 2002) and yield of wheat (Sharma & Swarup1988; Melhuish et al. 1991). However, under some conditions(4–12 °C), wheat can recover from prolonged waterlogging(42 to 80d) with yields achieving 82–96% of controls (Cannellet al. 1980; Belford 1981; Belford et al. 1985). Temperature orsoil type differences might explain these huge differences inresponses of wheat between some studies (see EnvironmentalParameters Influencing Wheat Waterlogging Responsessection).

Wheat might suffer N deficiency during waterlogging, whichif it persists, would also be detrimental to recovery growth andgrain yield (Watson et al. 1976; Cannell et al. 1980; Setter 2000;Robertson et al. 2009). Additional N fertilizer afterwaterlogging reduced grain yield loss by 20% for wheat in potsof soil (Robertson et al. 2009). However, if roots have sufferedsubstantial dieback, then N application after waterloggingmight not result in substantial yields (Setter 2000). Foliarspraying of N might not be practical in most situations, butwheat that received foliar N during 6d of waterlogging afteranthesis had smaller declines in grain number per spike andin individual seed weight (Wu et al. 2014).

Yield responses depend on the developmental stage whenwaterlogging occurs. For wheat, waterlogging was mostdamaging during pre-emergence, killing seeds and/or very youngseedlings (Setter & Waters 2003). Effects of waterlogging onseeds or emerging seedlings are a result of complete submer-gence, as water near the soil surface could inundate all tissuesof seeds/seedlings during emergence, including the coleop-tiles; for discussion of submergence versus waterlogging, seeBailey-Serres & Voesenek (2008) and Colmer & Voesenek(2009). In contrast to rice, wheat seeds cannot germinate un-der anoxic conditions (Perata et al. 1992). Inability of anoxicwheat seeds to break down starch into sugars contributes totheir lack of germination; provision of exogenous sugars re-sulted in 84% germination but still only enabled a fewmillimetres of root elongation and no coleoptile growth in an-oxia (Perata et al. 1992).

After germination and emergence, the two other stages atwhich waterlogging is most detrimental for wheat yields areat the seedling stage and at anthesis (Setter & Waters 2003).Waterlogging periods of 15 to 20d imposed on two varietiesin five consecutive periods (three vegetative and two reproduc-tive stages) along the crop life cycle with drainage and recoveryafter each waterlogging period showed the highest yield penal-ties occurred when waterlogging was applied during stem elon-gation to anthesis (de San Celedonio et al. 2014). However, thestage at which yield of wheat is most affected seems to dependon a combination of variety and environmental conditions. Thehighest yield reductionwas found at post-anthesis waterloggingdue to impaired grain filling caused by earlier leaf senescenceand shortening of the grain filling period (Araki et al. 2012a),rather than at anthesis as described by de San Celedonio et al.(2014). Environmental effects causing higher waterlogging sen-sitivity at later than at early developmental stages could behigher temperatures leading to faster O2 depletion, a higherevaporative demand during the warmer late-season conditionsand plant factors such as large reductions in root:shoot ratio thatcould occur in these older plants.

14 M. Herzog et al.


Interestingly, a significant genotype by waterloggingtiming interaction was found when analysing yield reduc-tions in 16 spring wheat varieties waterlogged at five dif-ferent developmental stages for 28–42d. A remarkableexample was for a Chinese genotype, which achieved grainyield of only 18% of control when waterlogged for 28d atfirst node to mid-boot stage, but grain yield was 93% ofcontrol when waterlogged for the same duration at anthesisto grain filling stage; in several other varieties, yield reduc-tions at these two stages were similar (e.g. 49% and 54%of controls in one variety, Sayre et al. 1994). The contrastingresponses were proposed to reflect genotype breeder selec-tion for better adapted materials to the different onsets ofwaterlogging that prevail in different climates, for example,late-season waterlogging in Southern China and Argentinacompared with earlier waterlogging events in Mexico (Sayreet al. 1994). Chinese and Argentinean varieties better toler-ated waterlogging at reproductive stages (93%, 84% and82% of controls) compared with Mexican varieties thatwere superior at early stages but showed large reductionsduring reproductive stage waterlogging (66–88% of con-trols). Even though selection specifically for waterloggingtolerance may not have been conducted, tolerance mayhave been incorporated in the germplasm by selection foryield in the specific environments/agroecological systems.Similarly, it has been suggested that waterlogging tolerancein winter wheats may have been indirectly selected for whenbreeding varieties in the UK for increased winter hardiness(Dickin et al. 2009).Grain quality of wheat impacted by waterlogging has been

considered in few studies. Waterlogging can decrease grainprotein content, consistent with the earlier described shootN deficiency (Fig. 5a), and thus affect processing quality(Olgun et al. 2008; Ashraf 2013), although this also variedbetween genotypes (Setter 2000). Individual grain weight,also a quality parameter, has been shown to decreasesignificantly (single grain weight 17% of controls) whenwaterlogging occurred during late developmental stages, suchas during grain filling (Araki et al. 2012a; de San Celedonioet al. 2014).

Summary

During recovery periods, growth mainly takes places in ad-ventitious roots as root:shoot ratio is restored, while seminalroot re-growth is poor due to death of apical meristems andlow formation of laterals. Root re-growth is fuelled both bysugars accumulated during waterlogging and by subsequentrecovery of photosynthesis. The high substrate demand bythe root system hampers shoot growth, eventually reducinggrain yield. Shoot N deficiency may also cause a decline ingrain yield and quality. Wheat cultivars differ in theirwaterlogging tolerance according to their growth stages. Thisis of importance to cultivation in fields prone to flooding atdifferent times, and for design of screening and breedingstrategies aimed at improvement of waterlogging tolerancein wheat.

CONCLUSIONS AND OUTLOOK

Root growth and physiology in wheat are adversely affected bysoil waterlogging, the magnitude of which depends on geno-type, plant developmental stage and the prevailing environ-mental conditions (especially soil type and temperature).Genotypic variation is evident for some root responses to lowO2, such as seminal root survival in short-term anoxia, adventi-tious rooting and amount of aerenchyma and recovery growthof lateral roots post-anoxia. Anoxia tolerance of genotypes isnot related simply to ethanolic fermentation rates; further re-search should evaluate energy requirements for maintenanceof anoxic roots and whether differences in efficiency of energyuse contribute to anoxia tolerance in wheat. Ability for sugartransport via phloem to root apices during hypoxia and anoxiashould also be assessed. Genotypes differ both in the numbersof adventitious roots formed and the porosity (i.e. amount ofaerenchyma) and thus internal movement of O2 into and alongthe roots. Even for these acclimated adventitious roots, O2 be-comes limiting at the root tip when relatively short, as porosityis moderate and O2 is consumed in respiration and lost radiallyto the rhizosphere, so that growth ceases. Energy-dependention transport is inhibited, resulting in a restricted nutrient up-take and translocation to the shoots. Tissue sugars often in-crease, despite declines in photosynthesis, as demand forsugars is markedly reduced when growth is inhibited. N defi-ciency in shoots contributes to reduced growth duringwaterlogging, and would slow recovery growth. Mn or Fetoxicity can impact on wheat when waterlogged in somestrongly acidic soils. Following waterlogging, adventitious rootgrowth resumes and lateral roots possibly emerge from some ofthe remaining seminal roots. Ability to recover upon drainagefollowing waterlogging is important for yield.

Identification of quantitative trait loci (QTL) associated withwaterlogging tolerance in wheat (e.g.Ballesteros et al. 2014)opens up avenues, with further work, for possible marker-assisted selection and discovery of underlying genes. For sub-mergence tolerance of rice, a very different situation to soilwaterlogging of wheat with shoots in air, identification of thegroup VII ethylene response factor (ERFVII) transcription fac-tors, and more specifically the SUB1A locus, has transformedrice breeding for submergence tolerance (Mackill et al. 2012).Such transcription factors regulating coordinated expressionsof stress-responsive genes, if identified also for roots in water-logged conditions, might be deployed for crop breeding(Bailey-Serres et al. 2012). Recent work using RNA interfer-ence has modulated in barley an ERFVII that is a substrate ofthe N-end rule pathway, with these transgenic plants showingimproved waterlogging tolerance (Mendiondo et al. 2015).

We suggest three priority areas for research on traits thatcould contribute to breeding of more waterlogging-tolerantwheat varieties: (1) internal aeration of new adventitious roots,(2) N deficiency of the shoot and (3) recovery ability followingtransient waterlogging. Progress in research on root aeration incrops is worthwhile to highlight. Identification of a QTL forroot porosity (as a surrogate for aerenchyma) in barley(Broughton et al. 2015) is of interest to explore also for wheat.Moreover, microarray analyses of laser micro-dissected root



tissues of maize (Rajhi et al. 2011) and rice (Shiono et al. 2015)have identified candidate genes for aerenchyma (maize) andROL barrier induction (rice). Use of wild relatives, such asH.marinum for wheat (Malik et al. 2011) andZea nicaraguensisfor maize (Mano et al. 2007; Abiko et al. 2012; Mano & Omori2013), to introduce some traits (e.g. root constitutive aeren-chyma and a ROL barrier) could improve waterlogging toler-ance, and these wide-crosses also provide novel resources forgene discovery. Knowledge of genome sequences (e.g. wheat,IWGSC 2014) will aid translation of findings to wheat. The Ndeficiency in wheat owing to waterlogging could also be ad-dressed both by seeking improved adventitious root nutrientuptake during waterlogging and possibly also targeting N-useefficiency in wheat, which is also being sought for other reasons(e.g. Jackson& Ismail 2015). Finally, much remains to be learntregarding recovery of root growth following anoxia andwaterlogging, and whole-plant recovery more generally, bothof which show some apparent genotypic variation in wheat.

ACKNOWLEDGMENTS

T.D.C. acknowledges support from the Australian ResearchCouncil (DP120101482). We thank the UWA Institute ofAdvanced Studies for hosting O.P. as Professor-At-Large andfor supporting G.S. with a Stay-fellowship to visit UWA. M.H.was supported by a PhD fellowship from theVillumFoundation.

REFERENCES

Abiko T., Kotula L., Shiono K., Malik A.I., Colmer T.D. & Nakazono M. (2012)Enhanced formation of aerenchyma and induction of a barrier to radial oxygenloss in adventitious roots ofZea nicaraguensis contribute to its waterlogging tol-erance as compared with maize (Zea mays ssp. mays). Plant, Cell & Environ-ment 35, 1618–1630.

Alamri S.A., Barrett-Lennard E.G., Teakle N.L. & Colmer T.D. (2013) Improve-ment of salt and waterlogging tolerance in wheat: comparative physiology ofHordeum marinum-Triticum aestivum amphiploids with their H.marinumand wheat parents. Functional Plant Biology 40, 1168–1178.

Albrecht G. & Biemelt S. (1998) A comparative study on carbohydrate reservesand ethanolic fermentation in the roots of two wetland and non-wetland spe-cies after commencement of hypoxia. Physiologia Plantarum 104, 81–86.

Albrecht G., Kammerer S., Praznik W. & Wiedenroth E.M. (1993) Fructan con-tent of wheat seedlings (Triticum aestivum L.) under hypoxia and following re-aeration. New Phytologist 123, 471–476.

Albrecht G., Mustroph A. & Fox T.C. (2004) Sugar and fructan accumulationduring metabolic adjustment between respiration and fermentation underlow oxygen conditions in wheat roots. Physiologia Plantarum 120, 93–105.

Albrecht G. &Wiedenroth E.M. (1993) Is long-term hypoxia met by the Pasteureffect in roots of wheat seedlings? Proceedings of the Royal Society ofEdinburgh Section B (Biological Sciences) 102, 407–412.

Albrecht G. & Wiedenroth E.M. (1994) Protection against activated oxygen fol-lowing re-aeration of hypoxically pretreated wheat roots – the response of theglutathione system. Journal of Experimental Botany 45, 449–455.

Araki H., Hamada A., Hossain M.A. & Takahashi T. (2012a) Waterlogging atjointing and/or after anthesis in wheat induces early leaf senescence and im-pairs grain filling. Field Crops Research 137, 27–36.

Araki H., Hossain M.A. & Takahashi T. (2012b) Waterlogging and hypoxia havepermanent effects on wheat root growth and respiration. Journal of Agronomyand Crop Science 198, 264–275.

Armstrong W. (1979) Aeration in higher plants. In Advances in Botanical Re-search (ed Woolhouse H.W.) 7, pp. 225–332. Academic Press, London.

Armstrong W. & Beckett P.M. (2011a) Experimental and modelling data contra-dict the idea of respiratory down-regulation in plant tissues at an internal [O2]substantially above the critical oxygen pressure for cytochrome oxidase. NewPhytologist 190, 431–441.

Armstrong W. & Beckett P.M. (2011b) The respiratory down-regulation debate.New Phytologist 190, 276–278.

ArmstrongW.&DrewM. (2002) Root growth andmetabolism under oxygen de-ficiency. Plant roots: The Hidden Half 3, 729–761.

Armstrong W. & Gaynard T. (1976) The critical oxygen pressures for respirationin intact plants. Physiologia Plantarum 37, 200–206.

ArmstrongW. &Webb T. (1985) A critical oxygen pressure for root extension inrice. Journal of Experimental Botany 36, 1573–1582.

Armstrong W., Webb T., Darwent M. & Beckett P.M. (2009) Measuring andinterpreting respiratory critical oxygen pressures in roots. Annals of Botany103, 281–293.

AshrafM. (2013) Stress-induced changes in wheat grain composition and quality.Critical Reviews in Food Science and Nutrition 54, 1576–1583.

Atwell B.J., Greenway H. & Colmer T.D. (2015) Efficient use of energy inanoxia-tolerant plants with focus on germinating rice seedlings. NewPhytologist 206, 36–56.

Bailey-Serres J., Lee S.C. & Brinton E. (2012) Waterproofing crops: effectiveflooding survival strategies. Plant Physiology 160, 1698–1709.

Bailey-Serres J. &VoesenekL.A.C.J. (2008) Flooding stress: acclimations and ge-netic diversity. Annual Review of Plant Biology 59, 313–339.

Ballesteros D., Mason R.E., Addison C., Andrea A.M., Nelly A.M.,Subramanian N., … Miller R. (2014) Tolerance of wheat to vegetative stagesoil waterlogging is conditioned by both constitutive and adaptive QTL.Euphytica 201, 329–343.

Barrett-Lennard E.G., Leighton P.D., Buwalda F., Gibbs J., Armstrong W.,Thomson C.J. & Greenway H. (1988) Effects of growing wheat in hypoxic nu-trient solutions and of subsequent transfer to aerated solutions. 1. Growth andcarbohydrate status of shoots and roots.Australian Journal of Plant Physiology15, 585–598.

Belford R.K. (1981) Response of winter-wheat to prolonged waterlogging underoutdoor conditions. Journal of Agricultural Science 97, 557–568.

Belford R.K., Cannell R.Q. & Thomson R.J. (1985) Effects of single andmultiplewaterloggings on the growth and yield of winter wheat on a clay soil. Journal ofthe Science of Food and Agriculture 36, 142–156.

Berry J.L. &NorrisW.E. Jr. (1949) Studies of onion root respiration I. Velocity ofoxygen consumption in different segments of root at different temperatures asa function of partial pressure of oxygen. Biochimica et Biophysica Acta 3,593–606.

Biemelt S., Keetman U. & Albrecht G. (1998) Re-aeration following hypoxia oranoxia leads to activation of the antioxidative defense system in roots of wheatseedlings. Plant Physiology 116, 651–658.

Biemelt S., Keetman U., Mock H.P. & Grimm B. (2000) Expression and activityof isoenzymes of superoxide dismutase in wheat roots in response to hypoxiaand anoxia. Plant, Cell & Environment 23, 135–144.

Blokhina O.B., Virolainen E. & Fagerstedt K.V. (2003) Antioxidants, oxidativedamage and oxygen deprivation stress: a review.Annals of Botany 91, 179–194.

Blokhina O.B., Virolainen E., Fagerstedt K.V., Hoikkala A., Wähälä K. &Chirkova T.V. (2000) Antioxidant status of anoxia-tolerant and -intolerant plantspecies under anoxia and reaeration. Physiologia Plantarum 109, 396–403.

BoruG., van Ginkel M., KronstadW.E. & Boersma L. (2001) Expression and in-heritance of tolerance to waterlogging stress in wheat. Euphytica 117, 91–98.

Boru G., van Ginkel M., Trethowan R.M., Boersma L. & Kronstad W.E. (2003)Oxygen use from solution by wheat genotypes differing in tolerance towaterlogging. Euphytica 132, 151–158.

Bramley H., Turner N.C., Turner D.W. & Tyerman S.D. (2010) The contrastinginfluence of short-term hypoxia on the hydraulic properties of cells and rootsof wheat and lupin. Functional Plant Biology 37, 183–193.

BramleyH.&Tyerman S.D. (2010)Rootwater transport underwaterlogged con-ditions and the roles of aquaporins. InWaterlogging Signalling and Tolerance inPlants (eds Mancuso S. & Shabala S.). Springer, Berlin Heidelberg.

Broughton S., Zhou G., Teakle N.L., Matsuda R., Zhou M., O’Leary R.A.,Colmer T.D. & Li C. (2015)Waterlogging tolerance is associated with root po-rosity in barley (Hordeum vulgare L.).Molecular Breeding 35, 1–15.

Buwalda F., Barrettlennard E.G., GreenwayH.&Davies B.A. (1988a) Effects ofgrowing wheat in hypoxic nutrient solutions and of subsequent transfer to aer-ated solutions. 2. Concentrations and uptake of nutrients and sodium in shootsand roots. Australian Journal of Plant Physiology 15, 599–612.

Buwalda F., Thomson C.J., Steigner W., Barrett-Lennard E.G., Gibbs J. &Greenway H. (1988b) Hypoxia induces membrane depolarization and potas-sium loss from wheat roots but does not increase their permeability to sorbitol.Journal of Experimental Botany 39, 1169–1183.

Cannell R.Q., BelfordR.K.,GalesK., Dennis C.W.&PrewR.D. (1980)Effects ofwaterlogging at different stages of development on the growth and yield ofwinter-wheat. Journal of the Science of Food and Agriculture 31, 117–132.

16 M. Herzog et al.


Cannell R.Q., Belford R.K., Gales K., Thomson R.J. & Webster C.P. (1984)Effects of waterlogging and drought on winter wheat and winter barley grownon a clay and a sandy loam soil. Plant and Soil 80, 53–66.

Collaku A. & Harrison S.A. (2002) Losses in wheat due to waterlogging. CropScience 42, 444–450.

Colmer T.D. (2003) Long-distance transport of gases in plants: a perspective oninternal aeration and radial oxygen loss from roots. Plant, Cell & Environment26, 17–36.

Colmer T.D. & Greenway H. (2011) Ion transport in seminal and adventitiousroots of cereals during O2 deficiency. Journal of Experimental Botany 62, 39–57.

ColmerT.D.&VoesenekL.A.C.J. (2009) Flooding tolerance: suites of plant traitsin variable environments. Functional Plant Biology 36, 665–681.

Condon A. (1999) Assessing potential sources of improved waterlogging toler-ance in bread wheat. In Proceedings 11th Australian Plant Breeding Confer-ence (Adelaide, 19–23 April 1999), pp. 166.

Dickin E., Bennett S. & Wright D. (2009) Growth and yield responses of UKwheat cultivars to winter waterlogging. The Journal of Agricultural Science147, 127–140.

Ding N.C. & Musgrave M.E. (1995) Relationship between mineral coating onroots and yield performance of wheat under waterlogging stress. Journal of Ex-perimental Botany 46, 939–945.

Dixon J., Braun H.J., Kosina P. & Crouch J.H. (2009) Wheat Facts and Futures2009. CIMMYT, Mexico, DF.

DrewM.C. (1983) Plant injury and adaptation to oxygen deficiency in the root en-vironment – a review. Plant and Soil 75, 179–199.

Else M.A., Janowiak F., Atkinson C.J. & Jackson M.B. (2009) Root signals andstomatal closure in relation to photosynthesis, chlorophyll a fluorescence andadventitious rooting of flooded tomato plants. Annals of Botany 103, 313–323.

Elzenga J.T.M. & van Veen H. (2010)Waterlogging and plant nutrient uptake. InWaterlogging Signalling and Tolerance in Plants (edsMancuso S.& Shabala S.),pp. 23–35. Springer, Berlin Heidelberg.

ErdmannB., Hoffmann P. &WiedenrothE.M. (1986)Changes in the root-systemof wheat seedlings following root anaerobiosis. 1. Anatomy and respiration inTriticum aestivum L. Annals of Botany 58, 597–605.

Erdmann B. & Wiedenroth E. (1986) Changes in the root system of wheatseedlings following root anaerobiosis 2.Morphology and anatomy of evolutionforms. Annals of Botany 58, 607–616.

Evans D.E. (2004) Aerenchyma formation. New Phytologist 161, 35–49.Felle H.H. (2005) pH regulation in anoxic plants. Annals of Botany 96, 519–532.Gibbs J. & Greenway H. (2003) Mechanisms of anoxia tolerance in plants. I.Growth, survival and anaerobic catabolism. Functional Plant Biology 30, 1–47.

Gibbs J., Turner D.W., Armstrong W., Darwent M.J. & Greenway H. (1998) Re-sponse to oxygen deficiency in primary maize roots I. Development of oxygendeficiency in the stele reduces radial solute transport to the xylem. AustralianJournal of Plant Physiology 25, 745–758.

Goggin D.E. & Colmer T.D. (2005) Intermittent anoxia induces oxidative stressin wheat seminal roots: assessment of the antioxidant defence system, lipid per-oxidation and tissue solutes. Functional Plant Biology 32, 495–506.

Goggin D.E. & Colmer T.D. (2007) Wheat genotypes show contrasting abilitiesto recover from anoxia in spite of similar anoxic carbohydrate metabolism.Journal of Plant Physiology 164, 1605–1611.

Greenway H., Armstromg W. & Colmer T.D. (2006) Conditions leading to highCO2 (>5 kPa) in waterlogged–flooded soils and possible effects on root growthand metabolism. Annals of Botany 98, 9–32.

Greenway H. & Gibbs J. (2003) Mechanisms of anoxia tolerance in plants. II.Energy requirements for maintenance and energy distribution to essentialprocesses. Functional Plant Biology 30, 999–1036.

GreenwayH.,Waters I.&Newsome J. (1992)Effects of anoxia on uptake and lossof solutes in roots of wheat.Australian Journal of Plant Physiology 19, 233–247.

Hamonts K., Clough T.J., Stewart A., Clinton P.W., Richardson A.E.,Wakelin S.A., … O’Callaghan M. (2013) Effect of nitrogen and waterloggingon denitrifier gene abundance, community structure and activity in the rhizo-sphere of wheat. FEMS Microbiology Ecology, 83, 568–584.

Haque M.E., Oyanagi A. & Kawaguchi K. (2012) Aerenchyma formation in theseminal roots of Japanese wheat cultivars in relation to growth under water-logged conditions. Plant Production Science 15, 164–173.

Hayashi T.,YoshidaT., Fujii K.,Mitsuya S., Tsuji T., OkadaY.,…Hayashi E. (2013)Maintained root length density contributes to the waterlogging tolerance incommon wheat (Triticum aestivum L.). Field Crops Research 152, 27–35.

Huang B.R.& Johnson J.W. (1995) Root respiration and carbohydrate status of 2wheat genotypes in response to hypoxia. Annals of Botany 75, 427–432.

Huang B.R., Johnson J.W., Box J.E.&NeSmithD.S. (1997a) Root characteristicsand hormone activity of wheat in response to hypoxia and ethylene. CropScience 37, 812–818.

Huang B.R., Johnson J.W. & NeSmith D.S. (1997b) Responses to root-zone CO2

enrichment and hypoxia of wheat genotypes differing in waterlogging toler-ance. Crop Science 37, 464–468.

Huang B.R., Johnson J.W., NesmithD.S. & Bridges D.C. (1994a) Root and shootgrowth of wheat genotypes in response to hypoxia and subsequent resumptionof aeration. Crop Science 34, 1538–1544.

Huang B.R., Johnson J.W., Nesmith S. & Bridges D.C. (1994b) Growth, physio-logical and anatomical responses of 2 wheat genotypes to waterlogging andnutrient supply. Journal of Experimental Botany 45, 193–202.

Ishizawa K. (2014) Intracellular pH regulation of plant cells under anaerobicconditions. In Low-Oxygen Stress in Plants – Oxygen Sensing and AdaptiveResponses to Hypoxia (eds van Dongen J.T. & Licausi F.), pp. 59–74. Springer,Wien.

Jackson M.B. (2004) The impact of flooding stress on plants and crops.JacksonM.B. &DrewM.C. (1984) Effects of flooding on growth andmetabolismof herbaceous plants. In Flooding and Plant Growth (ed Kozlowski T.T.),pp. 47–128. Academic Press, San Diego.

Jackson M.B. & Ismail A.M. (2015) Electrons, water and rice fields: plant re-sponse and adaptation to flooding and submergence stress. AoBPlants 7,plv078. DOI:10.1093/aobpla/plv078.

Jiang Z., Song X., Zhou Z., Wang L., Li J., Deng X. & Fan H. (2010) Aeren-chyma formation: programmed cell death in adventitious roots of winterwheat (Triticum aestivum) under waterlogging. Functional Plant Biology37, 748–755.

Justin S.H.F.W. & Armstrong W. (1987) The anatomical characteristics of rootsand plant response to soil flooding. New Phytologist 106, 465–495.

Khabaz-Saberi H., Barker S.J. & Rengel Z. (2012) Tolerance to ion toxicities en-hances wheat (Triticum aestivum L.) grain yield in waterlogged acidic soils.Plant and Soil 354, 371–381.

Khabaz-Saberi H.&Rengel Z. (2010)Aluminum,manganese, and iron toleranceimproves performance of wheat genotypes in waterlogged acidic soils. Journalof Plant Nutrition and Soil Science 173, 461–468.

Khabaz-Saberi H., Rengel Z., Wilson R. & Setter T.L. (2010a) Variation for tol-erance to high concentration of ferrous iron (Fe

2+) in Australian hexaploid

wheat. Euphytica 172, 275–283.Khabaz-Saberi H., Rengel Z., Wilson R. & Setter T.L. (2010b) Variation of toler-ance to manganese toxicity in Australian hexaploid wheat. Journal of PlantNutrition and Soil Science 173, 103–112.

Khabaz-Saberi H., Setter T.L. & Waters I. (2005) Waterlogging induces high totoxic concentrations of iron, aluminum, and manganese in wheat varieties onacidic soil. Journal of Plant Nutrition 29, 899–911.

Kotula L., Clode P.L., Striker G.G., Pedersen O., Läuchli A., Shabala S. &Colmer T.D. (2015) Oxygen deficiency and salinity affect cell-specific ion con-centrations in adventitious roots of barley (Hordeum vulgare).NewPhytologist.DOI:10.1111/nph.13535.

Kuiper P.J.C., Walton C.S. & Greenway H. (1994) Effect of hypoxia on ion up-take by nodal and seminal wheat roots. Plant Physiology and Biochemistry32, 267–276.

Kulichikhin K.Y., Aitio O., Chirkova T.V. & Fagerstedt K.V. (2007) Effect of ox-ygen concentration on intracellular pH, glucose-6-phosphate and NTP contentin rice (Oryza sativa) and wheat (Triticum aestivum) root tips: in vivo

31P-NMR

study. Physiologia Plantarum 129, 507–518.Li C., Jiang D., Wollenweber B., Li Y., Dai T. & CaoW. (2011)Waterlogging pre-treatment during vegetative growth improves tolerance to waterlogging afteranthesis in wheat. Plant Science 180, 672–678.

Long S.P. & Bernacchi C.J. (2003) Gas exchange measurements, what can theytell us about the underlying limitations to photosynthesis? procedures andsources of error. Journal of Experimental Botany 54, 2393–2401.

LuxmooreR., Fischer R.& Stolzy L. (1973) Flooding and soil temperature effectson wheat during grain filling. Agronomy Journal 65, 361–364.

Mackill D.J., Ismail A.M., Singh U.S., Labios R.V. & Paris T.R. (2012) Develop-ment and rapid adoption of submergence-tolerant (Sub1) rice varieties.Advances in Agronomy 115, 299.

Malik A.I., Colmer T.D., Lambers H. & Schortemeyer M. (2001) Changes inphysiological and morphological traits of roots and shoots of wheat in responseto different depths of waterlogging.Australian Journal of Plant Physiology 28,1121–1131.

Malik A.I., Colmer T.D., Lambers H. & Schortemeyer M. (2003) Aerenchymaformation and radial O2 loss along adventitious roots of wheat with only theapical root portion exposed to O2 deficiency. Plant, Cell & Environment 26,1713–1722.

Malik A.I., Colmer T.D., Lambers H., Setter T.L. & Schortemeyer M. (2002)Short-term waterlogging has long-term effects on the growth and physiologyof wheat. New Phytologist 153, 225–236.



MalikA.I., IslamA.K.M.R.&Colmer T.D. (2011)Transfer of the barrier to radialoxygen loss in roots ofHordeummarinum to wheat (Triticum aestivum): evalu-ation of four H.marinum–wheat amphiploids. New Phytologist 190, 499–508.

Mano Y. &Omori F. (2013) Flooding tolerance in interspecific introgression linescontaining chromosome segments from teosinte (Zea nicaraguensis) in maize(Zea mays subsp. mays). Annals of Botany 112, 1125–1139.

ManoY.,Omori F., TakamizoT.,KindigerB.,BirdR.M., LoaisigaC.&TakahashiH.(2007) QTLmapping of root aerenchyma formation in seedlings of amaize × rareteosinte “Zea nicaraguensis” cross. Plant and Soil 295, 103–113.

Marti J., SavinR.& Slafer G.A. (2015)Wheat yield as affected by length of expo-sure to waterlogging during stem elongation. Journal of Agronomy and CropScience 201, 473–486.

McDonald M.P., Galwey N.W. & Colmer T.D. (2002) Similarity and diversity inadventitious root anatomy as related to root aeration among a range ofwetlandand dryland grass species. Plant, Cell & Environment 25, 441–451.

McDonaldM.P., Galwey N.W., Ellneskog-Staam P. & Colmer T.D. (2001) Evalu-ation ofLophopyrum elongatum as a source of genetic diversity to increase thewaterlogging tolerance of hexaploid wheat (Triticum aestivum). NewPhytologist 151, 369–380.

McFarlane D., Barrett-Lennard E. & Setter T. (1989) Waterlogging: a hiddenconstraint to crop and pasture production in southern regions of Australia. Pa-per presented at the Proceedings of the 5th AustralianAgronomy Conference.

Melhuish F.M., Humphreys E., Muirhead W.A. & White R.J.G. (1991) Floodirrigation of wheat on a transitional red-brown earth .1. Effect of duration ofponding on soil-water, plant-growth, yield and N-uptake. Australian Journalof Agricultural Research 42, 1023–1035.

Mendiondo G.M., Gibbs D.J., Szurman-Zubrzycka M., Korn A., Marquez J.,Szarejko I., … Maluszynski M. (2015) Enhanced waterlogging tolerance inbarley by manipulation of expression of the N-end rule pathway E3 ligasePROTEOLYSIS6. Plant Biotechnology Journal. DOI:10.1111/pbi.12334.

Menegus F., Cattaruzza L., MattanaM., Beffagna N. &Ragg E. (1991) Responseto anoxia in rice and wheat seedlings: changes in the pH of intracellular com-partments, glucose-6-phosphate level, and metabolic rate. Plant Physiology95, 760–767.

Mills H.A., Jones J.B. &Wolf B. (1996) Plant Analysis Handbook II: A PracticalSampling, Preparation, Analysis, and Interpretation Guide. Micro–macroPublishing, Athens.

Musgrave M.E. (1994) Waterlogging effects on yield and photosynthesis in eightwinter wheat cultivars. Crop Science 34, 1314–1318.

Musgrave M.E. & Ding N. (1998) Evaluating wheat cultivars for waterloggingtolerance. Crop Science 38, 90–97.

MustrophA. &Albrecht G. (2003) Tolerance of crop plants to oxygen deficiencystress: fermentative activity and photosynthetic capacity of entire seedlingsunder hypoxia and anoxia. Physiologia Plantarum 117, 508–520.

Mustroph A. & Albrecht G. (2007) Fermentation metabolism in roots of wheatseedlings after hypoxic pre-treatment in different anoxic incubation systems.Journal of Plant Physiology 164, 394–407.

Olgun M., Metin K.A., Cemal A.M. & Caglar A. (2008) The effect ofwaterlogging inwheat (T. aestivumL.).ActaAgriculturae Scandinavica: SectionB, Soil & Plant Science 58, 193–198.

Pang J., Cuin T., Shabala L., ZhouM.,MendhamN.& Shabala S. (2007) Effect ofsecondary metabolites associated with anaerobic soil conditions on ion fluxesand electrophysiology in barley roots. Plant Physiology 145, 266–276.

Parry M.L., Canziani O.F., Palutikof J.P., van der Linden P.J. & Hanson C.E.(2007) Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribu-tion of Working Group II to the Fourth Assessment Report of the Intergovern-mental Panel on Climate Change. IPCC, Cambridge.

Paul M.J. & Driscoll S.P. (1997) Sugar repression of photosynthesis: the role ofcarbohydrates in signalling nitrogen deficiency through source:sink imbalance.Plant, Cell & Environment 20, 110–116.

Perata P., Pozueta-Romero J., Akazawa T. & Yamaguchi J. (1992) Effect ofanoxia on starch breakdown in rice and wheat seeds. Planta 188, 611–618.

Ponnamperuma F. (1984) Effects of flooding on soils. In Flooding and PlantGrowth (ed Kozlowski T.T.), pp. 9–45. Academic Press, London.

Ponnamperuma F.N. (1972) The chemistry of submerged soils. In Advances inAgronomy (ed Brady N.C.), pp. 29–96. Academic Press.

Powell N., Ji X.M., Ravash R., Edlington J. & Dolferus R. (2012) Yield stabilityfor cereals in a changing climate. Functional Plant Biology 39, 539–552.

Rajhi I., Yamauchi T., Takahashi H., Nishiuchi S., Shiono K., Watanabe R., …Mliki A. (2011) Identification of genes expressed in maize root cortical cellsduring lysigenous aerenchyma formation using laser microdissection and mi-croarray analyses. New Phytologist 190, 351–368.

Reuter D. (1997) Plant Analysis: An Interpretation Manual. CSIRO publishing,Collingwood.

RobertsonD., ZhangH., Palta J.A., Colmer T.&TurnerN.C. (2009)Waterloggingaffects the growth, development of tillers, and yield of wheat through asevere, but transient, N deficiency. Crop and Pasture Science 60, 578–586.

Saglio P.H. (1985) Effect of path or sink anoxia on sugar translocation in roots ofmaize seedlings. Plant Physiology 77, 285–290.

Saglio P.H., Drew M.C. & Pradet A. (1988) Metabolic acclimation to anoxia in-duced by low (2–4 kPa partial pressure) oxygen pretreatment (hypoxia) in roottips of Zea mays. Plant Physiology 86, 61–66.

SamadA., Meisner C.A., SaifuzzamanM.&GinkelM.v. (2001)Waterlogging toler-ance. InApplication of Physiology inWheatBreeding (edsReynolds J.I.O.-M.M.P.& McNab A.), pp. 136–144. CIMMYT, Mexico, DF, Mexico.

de San Celedonio R., Abeledo L.G. & Miralles D. (2014) Identifying the criticalperiod for waterlogging on yield and its components in wheat and barley. Plantand Soil 378, 265–277.

Sasidharan R. & Voesenek L.A.C.J. (2015) Ethylene-mediated acclimations toflooding stress. Plant Physiology 169, 3–12.

Sayre K., Van Ginkel M., Rajaram S. & Ortiz-Monasterio I. (1994) Tolerance towaterlogging losses in spring breadwheat: effect of time of onset on expression.Annual Wheat Newsletter 40, 165–171.

Setter T., Burgess P., Waters I. & Kuo J. (1999) Genetic diversity of barley andwheat for waterlogging tolerance in Western Australia. Paper presented atthe Proceedings of the 9th Australian Barley Technical Symposium.Melbourne, Australian Barley Technical Symposium Inc.

Setter T.L. (2000) Farming systems for waterlogging prone sandplain soils of thesouth coast: Grains Research and Development Corporation: GRDC projectnumber DAW292: a research project of Agriculture Western Australia/researchsupervisor T.L. Setter. Agriculture Western Australia, South Perth, W.A.

Setter T.L. & Waters I. (2003) Review of prospects for germplasm improvementfor waterlogging tolerance in wheat, barley and oats. Plant and Soil 253, 1–34.

Setter T.L., Waters I., Sharma S.K., Singh K.N., Kulshreshtha N., YaduvanshiN.P.S., … Ram P.C. (2009) Review of wheat improvement for waterloggingtolerance in Australia and India: the importance of anaerobiosis and elementtoxicities associated with different soils. Annals of Botany 103, 221–235.

Shabala S. (2011) Physiological and cellular aspects of phytotoxicity tolerance inplants: the role of membrane transporters and implications for crop breedingfor waterlogging tolerance. New Phytologist 190, 289–298.

ShaoG.C., Lan J.J., Yu S.E., Liu N., Guo R.Q. & SheD.L. (2013) Photosynthesisand growth of winter wheat in response to waterlogging at different growthstages. Photosynthetica 51, 429–437.

Sharma D.B. & Swarup A. (1988) Effects of short-term flooding on growth, yieldand mineral composition of wheat on sodic soil under field conditions. Plantand Soil 107, 137–143.

ShionoK., Takahashi H., Colmer T.D.&NakazonoM. (2008)Role of ethylene inacclimations to promote oxygen transport in roots of plants in waterloggedsoils. Plant Science 175, 52–58.

Shiono K., Yamauchi T., Yamazaki S., Mohanty B., Malik A.I., Nagamura Y.,…Nishizawa N.K. (2015)Microarray analysis of laser-microdissected tissues indi-cates the biosynthesis of suberin in the outer part of roots during formation of abarrier to radial oxygen loss in rice (Oryza sativa). Journal of ExperimentalBotany 65, 4795–4806.

Sieben W. (1964) Het Verband tussen ontwatering en opbrengst bij de jongezavelgronden in de Noordoostpolder (Relationship between drainage condi-tions and crop yield for young light clay soils in the Nordost Polder). TjeenkWillink V: Zwolle, The Netherlands.

Snowball K. & Robson A.D. (1991)Nutrient Deficiencies and Toxicities in Wheat:A Guide for Field Identification. CIMMYT, Mexico, D.F.

Steffens B. & Sauter M. (2014) Role of ethylene and other plant hormones in or-chestrating the responses to low oxygen concentrations. In Low-Oxygen Stressin Plants Oxygen Sensing andAdaptive Responses toHypoxia (eds vanDongenJ.T. & Licausi F.), pp. 117–132. Springer, Wien.

Steffens D., Hutsch B.W., Eschholz T., Losak T. & Schubert S. (2005) Waterlogging may inhibit plant growth primarily by nutrient deficiency rather thannutrient toxicity. Plant, Soil and Environment 51, 545–552.

Stieger P.A. & Feller U. (1994a) Nutrient accumulation and translocation inmaturing wheat plants grown on waterlogged soil. Plant and Soil 160, 87–95.

Stieger P.A. & Feller U. (1994b) Senescence and protein remobilisation in leaves ofmaturing wheat plants grown on waterlogged soil. Plant and Soil 166, 173–179.

Striker G.G. (2012) Time is on our side: the importance of considering a recoveryperiod when assessing flooding tolerance in plants. Ecological Research 27,983–987.

Takahashi H., Yamauchi T., Colmer T. & Nakazono M. (2014) Aerenchyma for-mation in plants. InLow-Oxygen Stress in Plants Oxygen Sensing and AdaptiveResponses to Hypoxia (eds van Dongen J.T. & Licausi F.), pp. 247–265.Springer, Wien.

18 M. Herzog et al.


The International Wheat Genome Sequencing Consortium (IWGSC) (2014) Achromosome-based draft sequence of the hexaploid bread wheat (Triticumaestivum) genome. Science 345, 1251788.

Thomson C.J., ArmstrongW.,Waters I. &GreenwayH. (1990) Aerenchyma for-mation and associated oxygen movement in seminal and nodal roots of wheat.Plant, Cell & Environment 13, 395–403.

ThomsonC.J.,Atwell B.J. &GreenwayH. (1989) Response of wheat seedlings tolow O2 concentrations in nutrient solution. 1. Growth, O2 uptake and synthesisof fermentative end-products by root segments. Journal of ExperimentalBotany 40, 985–991.

Thomson C.J., Colmer T.D., Watkin E.L.J. & Greenway H. (1992) Tolerance ofwheat (Triticum aestivum cvs Gamenya and Kite) and triticale (Triticosecalecv. Muir) to waterlogging. New Phytologist 120, 335–344.

Trnka M., Rotter R.P., Ruiz-Ramos M., Kersebaum K.C., Olesen J.E., Zalud Z.& SemenovM.A. (2014)Adverse weather conditions for Europeanwheat pro-duction will become more frequent with climate change. Nature ClimateChange 4, 637–643.

Trought M.C.T. & Drew M.C. (1980a) Development of waterlogging damage inwheat seedlings (Triticum aestivum L). 1. Shoot and root-growth in relationto changes in the concentrations of dissolved-gases and solutes in the soil solu-tion. Plant and Soil 54, 77–94.

TroughtM.C.T. &DrewM.C. (1980b) The development of waterlogging damagein wheat seedlings (Triticum aestivum L.). 2. Accumulation and redistributionof nutrients by the shoot. Plant and Soil 56, 187–199.

TroughtM.C.T. &DrewM.C. (1980c) The development of waterlogging damagein young wheat plants in anaerobic solution cultures. Journal of ExperimentalBotany 31, 1573–1585.

Trought M.C.T. & Drew M.C. (1982) Effects of waterlogging on young wheatplants (Triticum aestivum L.) and on soil solutes at different soil temperatures.Plant and Soil 69, 311–326.

U.S. Department of Agriculture (2015) Risk Management Agency – Cause ofLoss. United States Department of Agriculture, Washington.

Van Ginkel M., Rajaram S. & Thijssen M. (1991) Waterlogging in wheat: germ-plasm evaluation and methodology development. In The 7th Regional WheatWorkshop for Eastern, Central, and Southern Africa (ed Mwangi D.G.T.W.).CIMMYT, Najuru, Kenya.

Ward J.M., Mäser P. & Schroeder J.I. (2009) Plant ion channels: gene families, phys-iology, and functional genomics analyses.Annual Review of Physiology 71, 59–82.

Waters I., Kuiper P.J.C., Watkin E. & Greenway H. (1991a) Effects of anoxia onwheat seedlings. 1. Interaction between anoxia and other environmental-factors. Journal of Experimental Botany 42, 1427–1435.

Waters I., Morrell S., Greenway H. & Colmer T.D. (1991b) Effects of anoxia onwheat seedlings. 2. Influence of O2 supply prior to anoxia on tolerance toanoxia, alcoholic fermentation, and sugar levels. Journal of ExperimentalBotany 42, 1437–1447.

Watkin E.L.J., Thomson C.J. & Greenway H. (1998) Root development andaerenchyma formation in two wheat cultivars and one triticale cultivar grownin stagnant agar and aerated nutrient solution. Annals of Botany 81, 349–354.

WatsonE.R., Lapins P. &BarronR.J.W. (1976)Effect of waterlogging on growth,grain and straw yield of wheat, barley and oats. Australian Journal of Experi-mental Agriculture 16, 114–122.

Webster C.P., Belford R.K. & Cannell R.Q. (1986) Crop uptake and leachinglosses of 15N labelled fertilizer nitrogen in relation to waterlogging of clayand sandy loam soils. Plant and Soil 92, 89–101.

WiedenrothE.& Poskuta J. (1981) The influence of oxygen deficiency in roots onCO2 exchange rates of shoots and distribution of

14C-photoassimilates of wheatseedlings. Zeitschrift für Pflanzenphysiologie 103, 459–468.

Wiengweera A. & Greenway H. (2004) Performance of seminal and nodal rootsof wheat in stagnant solution: K

+and P uptake and effects of increasing O2

partial pressures around the shoot on nodal root elongation. Journal ofExperimental Botany 55, 2121–2129.

Wiengweera A., Greenway H. & Thomson C.J. (1997) The use of agar nutrientsolution to simulate lack of convection in waterlogged soils. Annals of Botany80, 115–123.

Wu J.-D., Li J.-C., Wei F.-Z., Wang C.-Y., Zhang Y. & Sun G. (2014) Effects of ni-trogen spraying on the post-anthesis stage of winter wheat under waterloggingstress. Acta Physiologiae Plantarum 36, 207–216.

Xia J.-H. & Roberts J.K. (1994) Improved cytoplasmic pH regulation, increasedlactate efflux, and reduced cytoplasmic lactate levels are biochemical traitsexpressed in root tips of whole maize seedlings acclimated to a low-oxygen en-vironment. Plant Physiology 105, 651–657.

Xu Q.T., Yang L., Zhou Z.Q., Mei F.Z., Qu L.H. & Zhou G.S. (2013) Process ofaerenchyma formation and reactive oxygen species induced by waterlogging inwheat seminal roots. Planta 238, 969–982.

Yaduvanshi N.P.S., Setter T.L., Sharma S.K., Singh K.N. & Kulshreshtha N.(2012) Influence of waterlogging on yield of wheat (Triticum aestivum), redoxpotentials, and concentrations of microelements in different soils in India andAustralia. Soil Research 50, 489–499.

Yamauchi T.,Abe F., KawaguchiK.,Oyanagi A.&NakazonoM. (2014a)Adven-titious roots of wheat seedlings that emerge in oxygen-deficient conditions haveincreased root diameters with highly developed lysigenous aerenchyma. PlantSignaling & Behavior 9, e28506.

Yamauchi T., Watanabe K., Fukazawa A., Mori H., Abe F., Kawaguchi K., …Oyanagi A. (2014b) Ethylene and reactive oxygen species are involved in rootaerenchyma formation and adaptation of wheat seedlings to oxygen-deficientconditions. Journal of Experimental Botany 65, 261–273.

Zabalza A., Van Dongen J.T., Froehlich A., Oliver S.N., Faix B., Gupta K.J., …Schmälzlin E. (2009) Regulation of respiration and fermentation to controlthe plant internal oxygen concentration. Plant Physiology 149, 1087–1098.

Zheng C., Jiang D., Liu F., Dai T., Jing Q. & Cao W. (2009) Effects of salt andwaterlogging stresses and their combination on leaf photosynthesis, chloroplastATP synthesis, and antioxidant capacity in wheat. Plant Science 176, 575–582.

Received 21 August 2015; received in revised form 03 November 2015;accepted for publication 08 November 2015

SUPPORTING INFORMATION

Additional Supporting Information may be found in the onlineversion of this article at the publisher’s website.

Table S1. Wheat growth from studies where wheat was grownin soil under waterlogged (water at soil surface unless other-wise stated) and drained conditions and recording both shootand root dry mass (DM), allowing for root:shoot ratio calcula-tion. Values are % reduction from controls [1-(waterlogged/drained)]. When multiple nutrient-levels were used we referto the lowest. Wl=waterlogged, DAS=Days after sowing.

Table S2.Grain yield reductions from studies where wheat wasgrown in soil under waterlogged (water at soil surface unlessotherwise stated) and drained conditions. Values are % reduc-tion from controls [1-(waterlogged/drained)]. When multiplenutrient-levels were used we refer to the lowest.WL=waterlogged, DAS=Days after sowing.Table S3. Summary of average root porosity (POR) oraerenchyma (AER) in wheat roots of different varieties, plantage, growing media to induce hypoxia, hypoxia/waterloggingduration, type of root (seminal and/or adventitious-nodal).Table S4. Reductions in photosynthetic rates calculated fromstudies where wheat was grown in soil under waterlogged (waterat soil surface unless otherwise stated) and drained conditions.Values are % reduction from controls [1-(waterlogged/drained)].All nutrient treatments were included. WL=waterlogged,DAS=Days after sowing, DAA=days after anthesis.Table S5. Summary of values/range for photosynthesis (PN)and stomatal conductance (gs) in wheat under control andwaterlogging conditions used for calculations in Figure 6. In-formation on waterlogging (WL) duration, variety, PN, gs,phenological stage during measurements and source are pro-vided. The numbers in brackets identify the bibliographicsource of each point in Figure 4.Table S6. Shoot nutrient levels of macronutrients N, P, K,Mg, Ca, (mg g�1 DM) and microelements Mn, Fe, Zn, Al(mg kg�1 DM) after soil waterlogging.



Chapter 2: Physiology, gene expression and metabolome of two wheat cultivars with contrasting

submergence tolerance

Wheat subject to complete submergence in a constant temperature room at the University of

Copenhagen, October 2016. Photo: Max Herzog.

38

Submergence tolerance in two wheat cultivars

Physiology, gene expression and metabolome of two wheat cultivars with contrasting submergence tolerance

Max Herzog, Takeshi Fukao, Anders Winkel, Dennis Konnerup, Suman Lamichhane, Jasper Benedict Alpuerto, Harald Hasler-Sheetal & Ole Pedersen.

The Freshwater Biological Laboratory, Department of Biology, University of Copenhagen, 2100 Copenhagen, Denmark (M.H., A.W., D.K., O.P.); Department of Crop and Soil Environmental Sciences, Virginia Tech, Blacksburg, Virginia 24061, USA (T.F., S.L., J.A.); Nordcee, Department of Biology, University of Southern Denmark, 5230 Odense, Denmark (H.H.) and VILLUM Center for Bioanalytical Sciences, University of Southern Denmark, 5230 Odense, Denmark (H.H.)

Abstract

Global climate change is projected to regionally increase floods on agricultural soils, reducing wheat (Triticum

aestivum) yields. Responses of wheat to complete submergence are not well understood as research has focused on

waterlogging (soil flooding). The aim of this study was to characterize the phenotypic responses of two wheat

genotypes (cultivars) to complete submergence and to determine traits conferring submergence tolerance. 18-day-old

wheat cv. Frument and Jackson grown in pots were completely submerged for up to 19 days while assessing responses

in physiology, gene expression and shoot metabolome (metabolomics). Results revealed 50% mortality after 9.3 and

15.9 days of submergence in Frument and Jackson, respectively, and 3-fold higher relative growth rate in Jackson

during 8 days of submergence and 25 days of recovery. Frument displayed faster leaf degradation as evident from

metabolomic fingerprinting and earlier declines in leaf chlorophylla and leaf tissue porosity. Surprisingly, shoot soluble

carbohydrates, starch and individual sugars declined to similar low levels in both cultivars by day 5, in spite of Frument

overexpressing genes encoding sucrose and fructan degrading enzymes relative to Jackson. Frument showed accelerated

leaf chlorosis and higher levels of phytol relative to Jackson, suggesting higher ethylene sensitivity in the intolerant

cultivar as seen in rice (Oryza sativa). The lipid peroxidation marker malondialdehyde increased from day 12 in

Frument but not in Jackson, indicating higher reactive O2 species (ROS) inflicted damage in Frument. Our study

suggests that ethylene sensitivity and ROS deprivation could be mechanisms that determine the vast differences in flood

tolerance.

Introduction

Wheat (Triticum aestivum) faces flooding on estimated 15-20% of its cropping area each year, reducing

growth and yield (Sayre et al., 1994; Setter and Waters, 2003), where waterlogging reduces average grain

yield by 43% (Herzog et al., 2016). In the USA, 2016 insurance payouts due to floods totaled US$217mn,

39


which was 3.4-fold higher than payouts due to droughts and more than any other stressor alone (U.S. Risk

Management Agency, 2016). Floods are expected to increase in the coming decades in major wheat

production areas due to climate change, increasing demand for the development of more flood-tolerant

wheat cultivars (Trnka et al., 2014). The terms “flooding” and “waterlogging” are often used

interchangeably to describe excessively wet conditions, but in the following we apply “waterlogging” only

when the root zone is flooded, and “submergence” when, in addition to the root system, all (or part of) the

aboveground organs are under water (c.f. Sasidharan et al., 2017).

Waterlogging reduces soil O2 levels due to the relatively low solubility of O2 in water and the 10,000-fold

slower gas diffusion in water compared to air (Armstrong, 1979). O2 is quickly consumed by roots and soil

microorganisms, resulting in severe hypoxic or even anoxic conditions in waterlogged soils (Ponnamperuma,

1984). Root O2 deficiency hampers root growth and function of wheat (Trought and Drew, 1980). In addition

to O2 deficiency, various potentially toxic compounds such as Mn2+, Fe2+, H2S and organic acids can

accumulate to phytotoxic levels in waterlogged soils (Ponnamperuma, 1984). When floodwaters recede and

tissues are re-oxygenated, reactive O2 species (ROS) such as superoxide radicals, hydrogen peroxide or

singlet O2 may form and damage cell structures such as membranes, proteins and nucleic acids (Blokhina et

al., 2003). When floodwaters rise above soil level, shoot tissues become submerged further increasing stress

(Colmer and Voesenek, 2009). Availability of O2 and CO2 for aerobic respiration and photosynthesis,

respectively, becomes limited during submergence, as the gasses must overcome diffusional resistances of

the cuticle as well as the diffusive boundary layers (Mommer et al., 2005). Submerged terrestrial vegetation

may therefore experience an ‘energy crisis’ due to low carbohydrate production and the low energy harvest

in anaerobic glycolysis (Gibbs and Greenway, 2003).

Plants growing in flood-prone areas, i.e. terrestrial wetland plants, display traits that confer flood tolerance

(Colmer and Voesenek, 2009). A key trait is ‘internal aeration’ via interconnected air spaces enabling

internal O2 gas phase diffusion to submerged organs such as roots. O2 reaching submerged tissues may

originate from floodwaters (Winkel et al., 2013), endogenously from underwater photosynthesis (Winkel et

al., 2011; Pellegrini et al., 2017) or ‘snorkeling’ (Voesenek and Blom, 1989; Colmer and Voesenek, 2009;

Herzog and Pedersen, 2014) by shoot tissues protruding into the atmosphere. Moreover, roots of most

terrestrial wetland plants form a barrier to radial O2 loss (ROL) reducing O2 loss to anoxic soils along the

roots base, ensuring adequate root tip O2 for root growth (Armstrong, 1979; Colmer, 2003). The gaseous

phytohormone ethylene is known to induce such adaptations in both wild wetland plants (Visser et al., 1996),

wheat (Huang et al., 1997) and rice (Oryza sativa) (Shiono et al., 2008).

40


Breeding for flood-tolerant crops has resulted in high yielding rice cultivars able to withstand 14 days of

submergence (Ismail et al., 2013). While wheat waterlogging tolerance could have been enhanced by

breeding for winter hardiness in UK cultivars (Dickin et al., 2009), waterlogging tolerance is to our

knowledge not specifically selected for in wheat breeding, and thus waterlogging or submergence-tolerant

wheat cultivars are yet to be released. Several studies have documented variation in waterlogging tolerance

in wheat germplasm (root traits: Dickin et al. (2009); Hayashi et al. (2013); shoot biomass: Huang et al.

(1994); Hayashi et al. (2013); grain yield: Van Ginkel (1991); Sayre et al. (1994); Collaku and Harrison

(2002)), relating to seminal root short term anoxia tolerance (allowing seminal roots to resume growth

following reaeration) and formation of porous adventitious roots improving root O2 supply and thereby

nutrient uptake (Herzog et al., 2016).

Meanwhile, traits conferring submergence tolerance in wheat have not yet been documented. Submergence

of wheat has been observed to occur (Musgrave and Ding, 1998; Winkel et al., 2017) and to decrease yields

relative to waterlogging (Samad et al., 2001). However, the extent of partial or complete submergence of

wheat has to our knowledge not been estimated, but would be a prominent risk during early stages of

development when plants are still small (e.g., prolonged seedling stage in winter wheat). Hence, the aim of

this study was to assess wheat genotype variations in submergence tolerance, and to determine which traits

confer submergence tolerance. Two wheat cultivars with different waterlogging tolerance were subjected to

complete submergence in a pot experiment while assessing plant growth, plant survival, metabolites

(metabolomics), gene expression and a range of physiologic parameters in order to identify possible traits

conferring submergence tolerance. We hypothesized that submergence tolerance would be related to

phenotype specific carbohydrate consumption as seen in rice (Das et al., 2005), prompting us to evaluate

shoot levels of soluble carbohydrates, starch, expression of genes related to carbohydrate degradation and

metabolites of the primary energy metabolism.

Results

Wheat cv. Frument and Jackson exhibit contrasting submergence tolerance

Growth of wheat cultivars Frument and Jackson was strongly impaired by complete submergence, even

when provided 14-25 days recovery (Fig. 1). Survival and shoot relative growth rate (RGR) declined with

time of submergence, but interestingly differed significantly between cultivars (Fig. 1). Cultivar RGR

already differed significantly after 8 days of submergence being 0.06 d-1 and 0.02 d-1 in Jackson and

Frument, respectively (compared to 0.1 d-1 in drained controls), relating to Jackson and Frument shoot

biomass at 24% and 13% of controls in air, respectively. Frument shoot RGR continued to decline faster than

41


Jackson and thus two-way time × cultivar ANOVA detected significant time, cultivar and interaction effects.

Importantly, when comparing cultivar growth during submergence assessed without a recovery period (i.e.,

harvesting shots directly after 8 days of submergence), both cultivars exhibited similar slightly negative RGR

(Fig. S1).

Figure 1. Survival and growth of 18-day-old wheat cultivars Frument and Jackson during 0-19 day complete submergence and a following 14-25 day recovery period. (A) Photos illustrating growth and survival of Jackson and Frument wheat cultivars following complete submergence. All plants were 51 days old, but had been subject to varying lengths of submergence (0-19 days) and recovery (14-25 days, depending on the duration of submergence) when the photos were taken. After 19 days of complete submergence no plants survived, hence this time point is not shown. (B) Survival of wheat cultivars with time of submergence. Symbols represent survived plants/number of desubmerged plants after 0-19 days of submergence (n = 4-8) of Frument (open symbols) and Jackson (closed symbols). The blue and green lines represent the central tendency of a logistic model fitted to Frument and Jackson survival data, respectively. Shaded areas are 95% confidence intervals. The vertical dashed line indicates 50% survival. Survival was defined as the presence of turgid, green leaf material following recovery. (C) Shoot RGR of wheat cultivars Frument (open symbols) and Jackson (closed symbols) during submergence and a following recovery period. Plant RGR was calculated from initial biomass (at the start of submergence) and biomass when all plants were 51 days old, but had been subject to varying lengths of submergence (0-19 days) and recovery (14-25 days). Two-way ANOVA showed significant time, cultivar and interaction effects (P < 0.0001). * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05). Values are means (± SE, n = 4-8). Shoot DM of Frument and Jackson controls in air at the end of recovery (0 days of submergence) were not significantly different (t-test, P = 0.499, n = 6-7).

Plant survival also decreased significantly with the duration of complete submergence (Fig. 1, P < 0.05,

Wald Chi-square test). Frument succumbed to submergence approximately seven days earlier than Jackson,

evident from the time of submergence resulting in 50% mortality (LT50) of 9.3 and 15.9 days in Frument and

Jackson, respectively (Fig. 1). Submergence duration resulting in 100% mortality also differed by seven days

42


(12 and 19 days in Frument and Jackson, respectively). Non-overlapping 95% confidence intervals of the

modeled survival curves support that cultivars differed significantly in submergence tolerance (Fig. 1). In the

following sections, we describe how this contrasting submergence tolerance is phenotypically reflected in

physiologic, genetic and metabolic responses of the two cultivars.

Submergence induces faster leaf degradation in Frument than Jackson

Superhydrophobic wheat leaves retain a gas film when submerged (Raskin and Kende, 1983; Konnerup et

al., 2017), and leaf gas films enhance wheat submergence tolerance (Winkel 2017). We therefore assessed if

leaf gas film thickness during submergence differed between cultivars Frument and Jackson (Fig. 2). In both

cultivars, leaf gas film thickness declined from initial 10-24 µm to below the 3 µm detection limit by day 5

and did not recover afterwards. Consequently, leaf gas film thickness did not differ significantly between

cultivars according to two-way time × cultivar ANOVA.

Figure 2. Gas film thickness (A), FV/FM (B), chlorophylla (C) and leaf tissue porosity (D) of Frument (open symbols) and Jackson (closed symbols) wheat cultivars with time of submergence. The leaf sampled was the youngest, fully expanded leaf at time of submergence (3rd leaf). Round symbols symbolize untreated controls in air at the end of the treatment period (day 16). Values are means (± SE, n = 4), except (A) where means are ± SD (n = 4). * indicate significant difference between cultivars at single time points (Sidak’s multiple comparisons test, P < 0.05). (A) Two-way ANOVA on Ln-transformed data only showed a significant time effect (P < 0.0001) and no significant differences between cultivars in post-hoc tests. (B) Two-way repeated measures ANOVA showed a significant time effect only (P < 0.0001). (C) Two-way ANOVA showed significant time, cultivar and interaction effects (P < 0.0001). (D) Two-way ANOVA showed significant time, cultivar and interaction effects (P < 0.0001).

0 5 1 0 1 50 .5

0 .6

0 .7

0 .8

D a y s o f s u b m e rg e n c e

FV

/FM

B

0 5 1 0 1 5 2 0

0

1 0

2 0

3 0

4 0


Le

af

ga

s f

ilm t

hic

kn

es

s (

µm

)

A

J a c k s o n s u b m e rg e d

F ru m e n t s u b m e rg e d

J a c k s o n c o n tro l

F ru m e n t c o n tro l

0 5 1 0 1 5 2 0

0

5

1 0

1 5

2 0

2 5


Ch

loro

ph

yll a

(m

g g

-1 D

M)

* **

*

C

0 5 1 0 1 5 2 0

0

5

1 0

1 5

2 0

2 5


Le

af

po

ros

ity

(%

)

* *

D

43


Leaf gas film disappearance was followed by indications of submergence induced leaf damage. FV/FM ratios

measured on day 0, 5, 8 and 12 on the youngest fully expanded leaf at the time of submergence indicated

increasing damage to PSII in submerged plants (Fig. 2), with FV/FM ratios declining from initial 0.79 to 0.65-

0.67 on day 12. However, two-way ANOVA detected significant time but no cultivar effect (P = 0.2702).

Underwater photosynthesis (PN) and dark respiration (RD) rates were measured on days 0, 2, 8 and 16 of

submergence (Fig. S2). In both cultivars, PN remained at initial 1.8-2.3 µmol O2 m-2 s-1 until day 2, but

approached zero on day 8. Leaf RD on day 2 was reduced to half of initial -0.45 to -0.47 µmol O2 m-2 s-1 in

both genotypes. Frument respiration rates on day 2 and 8 were 26-32% higher than Jackson but these

differences were not statistically significant (Sidak’s multiple comparisons test, P > 0.05) and for PN and RD

two-way ANOVA only detected significant time-effects.

In contrast to the above-mentioned factors where both cultivars responded relatively similar to submergence

stress, tissue porosity and leaf chlorophylla in the youngest fully expanded leaf at the time of submergence

declined significantly faster in Frument than in Jackson. Chlorophylla concentrations remained at initial 9.7-

11.0 mg g-1 dry mass (DM) in both cultivars until day 5, before declining faster in Frument than in Jackson

(Fig. 2) resulting in almost three-fold higher chlorophylla in Jackson on day 12 (Sidak’s multiple

comparisons test, P < 0.0001). Tissue porosity remained at initial 20% for the first 8 days of submergence in

both cultivars, until water began to infiltrate Frument leaves (Fig. 2). In contrast, Jackson leaf porosity did

not decline until day 16. Therefore, the two-way ANOVA detected significant time, cultivar and interaction

effects in both leaf chlorophylla and leaf tissue porosity.

Shoot carbohydrates decline to similarly low levels in Frument and Jackson

Submergence tolerance of rice has been linked to reduced underwater elongation of leaves and internodes,

resulting in lower carbohydrate consumption and lower mortality in non-elongating genotypes (Das et al.,

2005). In order to evaluate if the contrasting submergence tolerance observed in this study was also related to

differences in cultivar elongation and carbohydrate consumption, we measured shoot length and shoot

carbohydrate concentrations during submergence.

44


Figure 3. The sum of ethanol and water soluble carbohydrates (A) determined using the anthrone-method, and fructose (B), glucose (C) and sucrose (D) determined by GC-MS in shoots of wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence. Round symbols symbolize untreated controls in air at the end of the treatment period (day 16). In (A) the entire shoot except for the 3rd leaf (used for other physiological measurements, see Fig. 2) was homogenized and used for analysis. Two-way time × cultivar ANOVA on Ln-transformed data showed significant time (P < 0.0001), cultivar (P = 0.0396) and interaction (P < 0.0001) effects. Time explained 91% of the variation compared to cultivar (0.5%) and interaction (5%) effects. In (B), (C) and (D) entire shoots were sampled and two-way time × cultivar ANOVA showed significant time, cultivar and interaction effects; time; time and interaction effects, respectively. * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05), values are means (± SE, n = 3-4).

Initial concentrations of combined ethanol and water soluble carbohydrates were significantly higher (28%)

in Frument than Jackson (Sidak’s multiple comparisons test, P < 0.05), but on day 2 the abundance of total

soluble carbohydrates was similar in the two cultivars likely due to higher sugar consumption in Frument

(Fig. 3). Soluble carbohydrate concentrations remained at similar levels until day 12 and 14, when they

declined further in Frument resulting in shoots of Jackson containing ~30% more soluble carbohydrates,

coinciding with leaf disintegration as indicated by loss in leaf porosity. At the end of treatment (day 16),

soluble carbohydrate levels in both cultivars had declined to 21-31% of initials. The periodically higher

soluble carbohydrate consumption by Frument resulted in ANOVA detecting significant time, cultivar and

interaction effects with factor cultivar explaining 0.5% of the variation. Shoot fructose, glucose and sucrose

determined using metabolomics had declined to equally low levels in both cultivars by day 2 (Fig. 3). By day

5 these sugars had declined to ~10% of initial values and did not recover. Interestingly, cultivars did not

differ significantly at any submergence time point in any of these three sugars. The significant decline, as

well as the only minor cultivar differences, was also evident in other sugars (tagatose, trehalose, kestose,

0 5 1 0 1 5 2 0

5 0 0

1 0 0 0


Eth

an

ol+

wa

ter

so

ub

le s

ug

ars

(µm

ol

he

xo

se

eq

. g

-1 D

M)

0

* J a c k s o n s u b m e rg e d

F ru m e n t s u b m e rg e d



A

* *

0 5 1 0 1 5 2 00

2 0 0 0

4 0 0 0

6 0 0 0g lu c o s e


No

rma

lize

d A

rea

(a

rbit

rary

un

its

)

C

0 5 1 0 1 5 2 00

5 0 0 0

1 0 0 0 0

1 5 0 0 0

2 0 0 0 0s u c r o s e


No

rma

lize

d A

rea

(a

rbit

rary

un

its

)

D

*

0 5 1 0 1 5 2 00

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0f r u c t o s e


No

rma

lize

d A

rea

(a

rbit

rary

un

its

)

B

*

45


ribose, myo-isonitol, maltose and glucose-6-phosphate), while 1,6-anhydro-glucose first declined from day 8

(Fig. S3).

Quantifying soluble carbohydrates using the colorimetric reactions with anthrone indicated that shoots

continue to possess sugar reserves at 20-30% of initial levels, while metabolomics data showed that

especially glucose and sucrose decline to very low levels. The anthrone method is a popular method due to

its simplicity and sensitivity, but has the disadvantage of a large difference in the color intensity produced by

different types of sugars (Pontis, 2017). Combined with the arbitrary units of the metabolomics method this

makes comparisons with absolute numbers from the anthrone analysis (Fig. 3) challenging. In addition,

carbohydrates stored in sheath and stem tissues, which were also harvested, would conceal declines to very

low levels in leaf tissues.

Submergence also caused starch levels to decline (Fig. S4). On day 2, Jackson and Frument contained 31%

and 21% of initial starch concentrations, respectively, remaining at similar levels until day 8. From day 12-

16 starch concentrations started to increase, especially in Frument. The increasing starch concentration at

such late time points is surprising, as carbohydrate production would be insignificant considering the low PN.

We suggest that the increasing starch levels reflect increasing leaf blade disintegration and detachment from

the shoot, causing leaf sheaths (acting as the main wheat carbohydrate storage organ, Scofield et al. (2009))

to make up most of the tissue sample resulting in seemingly higher starch concentrations. Leaf disintegration

was more severe and occurred earlier in Frument, likely explaining why starch reached highest

concentrations in this cultivar.

F ru m e n t J a c k s o n0

2 0

4 0

6 0

Sh

oo

t le

ng

th (

cm

)

a a a A A A

In it ia ls ( d a y 0 )

c o n tro ls in a ir ( d a y 8 )

S u b m e rg e d (d a y 8 )

Figure 4. Shoot elongation during 8 days of submergence in Frument and Jackson wheat cultivars. 18-day-old wheat plants were completely submerged in artificial floodwater for 8 days (except controls in air which were not submerged). Shoot length was measured on initials (day 0 of submergence, open bars) and after 8 days under drained (gray bars) and completely submerged conditions (closed bars, no recovery period). Values are means (± SE, n = 4). One-way ANOVA showed no significant treatment effects (P > 0.05). Different letters indicate significant difference between treatments within each cultivar (Tukey’s multiple comparisons test, P > 0.05).

46


Neither Frument nor Jackson shoots elongated during submergence, evident from submerged and control

plants exhibiting similar shoot length after 8 days of treatment (Fig. 4). Shoot lengths measured on day 2 and

5 confirmed that shoots did not elongate compared to drained controls (data not shown). Thus, the slight

difference in cultivar carbohydrate consumption was seemingly not caused by differences in elongation

response.

In conclusion, submergence-intolerant cultivar Frument displayed higher carbohydrate consumption at the

early time points (0-2 days), but otherwise, levels of soluble carbohydrates and starch were similar between

cultivars. Frument experienced faster chlorophylla and tissue porosity loss, while responses in leaf gas film

thickness, FV/FM, PN, RD and shoot length were similar amongst the two cultivars.

Figure 5. Multivariate analysis (A) and selected metabolites with time of submergence (B, C) in entire shoots of wheat cultivars Frument and Jackson during 16 days of complete submergence. (A) Principal component analysis (PCA) scores plots showing PC1 vs. PC2 (top) and PC1 vs. PC3 (bottom) of 1211 metabolite entities in shoots of submerged Frument (squares) and Jackson (triangles) wheat cultivars, the color legend identifying time points. Only metabolites that passed the quality control filter are included. Day 16 controls in air clustered with the initials (data not shown). (B) Phytol in wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence. Two-way time × cultivar ANOVA on Ln-transformed data showed significant time, cultivar and interaction effects. (C) Malondialdehyde (MDA) in wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence. Two-way time × cultivar ANOVA on Ln-transformed data showed significant time, cultivar and interaction effects. In (B) and (C) * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05), values are means (± SE, n = 4). Round symbols symbolize untreated controls in air at the end of the treatment period (day 16).

47


Metabolomic fingerprinting reveals accelerated metabolomic changes in submerged Frument

A total of 1211 out of 74,359 mass spectral features passed our quality controls filters (present in all samples

of at least one group; in 80% of the quality control samples and with a coefficient of variance < 35%)

following LC-MS analysis of Frument and Jackson shoot tissues. The 1211 metabolites were used for

metabolomic fingerprinting without further annotation since the high number of reproducibly detected

metabolites suggests good and robust metabolome coverage (Hasler-Sheetal et al., 2016; Lindahl et al.,

2017). To visualize the metabolic changes due to submergence stress, we conducted a principal component

analysis (PCA) showing clear treatment and cultivar related clustering of the samples (Fig. 5). PC1 explained

38.5% of the variance separating days of submergence, while PC2 and PC3 explained 25.3% and 7.21% of

the variance, respectively, with especially PC3 separating the two cultivars. Within 2 days of submergence,

both genotypes had moved relative to the initials (day 0) with Frument moving furthest along PC1 and PC2,

indicating a stronger metabolic response in this submergence-intolerant cultivar. Frument metabolome

continued to shift faster along especially PC1, resulting in Jackson day 8 and Frument day 5 clustering in the

PC1 vs. PC2 plot (Fig. 5). By the end of the experiment, Frument day 14 and day 16 had moved furthest

along PC1 and PC2, clearly separating these severely degraded shoots. A VENN-diagram of the 1211

metabolites showed that the metabolic changes separating treatments in the PCA plots were driven by

changes in virtually all metabolites, with time significantly affecting 1152 entities and approximately half of

the metabolites showing significant time, cultivar and time × cultivar interaction effects (Fig. S5).

Annotation of metabolic entities allowed for monitoring changes in amino acids, sugars, phytol (all GC-MS)

and MDA (LC-MS). Phytol results from the initial step of enzymatic chlorophyll hydrolysis by

chlorophyllase (Matile et al., 1999), with two-way ANOVA showing significant time, cultivar and

interaction effects (Fig. 5). In both cultivars phytol increase coincided with chlorophylla decline in the

youngest fully expanded leaf (Fig. 2) by day 5 (Frument) and day 12 (Jackson). Malondialdehyde (MDA) is

considered a useful indicator of lipid peroxidation (Hodges et al., 1999), as measured using thiobarbituric

acid-reactive-substances (TBARS) for assessing oxidative stress in wheat and rice during submergence or

hypoxia (Albrecht and Wiedenroth, 1994; Li et al., 2011; Alpuerto et al., 2016). In our study, ANOVA

showed significant time, cultivar and interaction effects for shoot MDA (Fig. 5). MDA in Jackson remained

close to the initial levels while in Frument MDA increased from day 12, resulting in final levels 10 times

higher than initials and 11 times higher than Jackson. However, it should be noted that Frument survival

rates declined before MDA increased.

The metabolome analyis also revealed that 12 out of 17 measured amino acids (Asn, Gln, Ile, Leu, Lys, Met,

Phe, Pro, Thr, Thp, Tyr, Val) increased in shoots during the first 12 days of submergence, with only 5

48


showing decreasing or unchanged levels in both cultivars (Ala, Asp, Glu, Ser, Gly; Fig. S6). During the first

12 days of submergence, Frument never had lower and occasionally significantly higher amino acid

concentrations compared to Jackson, but on day 14 and 16 all amino acid levels in Frument had dropped

below Jackson levels. E.g., Pro levels were significant 5-fold higher in Jackson than Frument from day 12-16

(Sidak’s multiple comparisons test, P < 0.05; Fig. S6). From day 14 to 16 amino acids in Jackson also

generally declined, indicating that Jackson shoot tissues were increasingly degraded by that final time point.

Meanwhile, levels of GABA did not differ between cultivars (data not shown).

Frument exhibits higher expression of genes encoding carbohydrate degrading enzymes than Jackson

The mRNA levels of 13 genes associated with carbohydrate degradation (fructan exohydrolases, kestose

exohydrolase, sucrose synthases and α-amylases) were quantified relative to the initial (day 0) levels in

Frument using quantitative real-time PCR (qRT-PCR) on days 0, 2, 5, 8 and 12 (Fig. 6). While three genes

encoding fructan and kestose exohydrolases were down-regulated by submergence or remained unchanged in

both cultivars (6&1-FEH, 6-FEH, 6-KEHw2), two genes encoding fructan exohydrolases were significantly

upregulated in Frument compared to Jackson (1-FEHw1, 1-FEHw3) at several time points. Similar patterns

were observed in sucrose synthase genes: SUS3 and SUS11 were downregulated in both cultivars, while

SUS4 and SUS5 were expressed at significantly higher levels in Frument than Jackson at several time points.

The expression of α-amylase genes generally increased with time, but in contrast to genes associated with

fructan and sucrose degradation, only one out of four α-amylase genes was expressed slightly (13-19%)

higher in Frument than in Jackson (α-AMY4-1, Fig. S7). In conclusion, during submergence Frument

expressed genes encoding sucrose and fructan degrading enzymes at significantly higher levels than Jackson.

49


Figure 6. Relative mRNA levels of genes associated with carbohydrate degradation. Genes are fructan exohydrolases (1FEHw1, 1FEHw3, 6-FEH, 6&1-FEH), kestose exohydrolase (6-KEHw2) and sucrose synthases (SUS3, SUS4, SUS5, SUS11), for details on primers and annealing temperatures see Table S1. Transcripts of representative genes were quantified in shoots of wheat cultivars Frument (open bars) and Jackson (closed bars) exposed to submergence for 0-12 days by quantitative real-time PCR. The relative level of each mRNA was calculated by comparison with initial (day 0) Frument. Values are means (± SE, n = 3). * indicate significant differences between the cultivars (Student’s t-test, P < 0.05).

Discussion

Comparing submergence tolerance in two wheat cultivars revealed contrasting survival, growth, physiologic

and metabolic responses during the 19 days treatment period and following 14-25 days of recovery. Frument

showed accelerated structural leaf degradation, leaf chlorosis, metabolic response and elevated MDA levels

compared to Jackson, while shoot carbohydrate consumption rates only differed during the initial 2 days of

submergence. In the following sections we discuss these findings in relation to especially rice submergence

responses.

Compared to growth during waterlogging reported in the literature, submergence induced higher growth

reductions with wheat shoot biomass reductions to 29-31% of controls following 14 days of waterlogging

(Malik et al., 2001; Robertson et al., 2009) and to 2-6% of controls following 14 days submergence in this

study. The larger growth penalty due to submergence resulted from negative shoot RGR during submergence

(-0.03 d-1 during 8 days, Fig. S1) compared to positive shoot RGR of 0.09 d-1 during 14 days of waterlogging

(Malik et al., 2001) and would also have been aggravated by senesced shoot material (Winkel et al., 2017)

0 2 5 8 1 20

2

4

6

8

1 F E H w 1

Re

lativ

e m

RN

A l

eve

l J a c k s o n

F ru m e n t

**

**

*

0 2 5 8 1 20

1

2

3

4

1 F E H w 3

Re

lativ

e m

RN

A l

eve

l

*

**

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

6 -F E H

Re

lativ

e m

RN

A l

eve

l

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

2 .0

6 & 1 -F E H

Re

lativ

e m

RN

A l

eve

l

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

6 -K E H w 2R

ela

tive

mR

NA

le

vel

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

S U S 3

Re

lativ

e m

RN

A l

eve

l

0 2 5 8 1 20

1

2

3

4

S U S 4

Re

lativ

e m

RN

A l

eve

l

* * ** *

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

2 .0

S U S 5

Re

lativ

e m

RN

A l

eve

l

**

** *

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

S U S 1 1

Re

lativ

e m

RN

A l

eve

l

50


impeding recovery growth. Moreover, while waterlogging does not lead to death of entire wheat plants

(Herzog et al., 2016), mortality resulting from the additional stresses caused by complete submergence

(Colmer and Voesenek, 2009) was observed following submergence for 8 days in the present study and 10

days in Winkel et al. (2017).

Comparing growth and survival rates from the current study with those of other submerged grass species

revealed that especially Jackson can survive relatively long but faces severe shoot biomass reductions.

Following 12 days of complete submergence and recovery, shoot biomass of tropical grasses Chloris gayana

and Panicum coloratum were 21-54% of the controls (Imaz et al., 2013), compared to shoot biomass of 9%

of the controls in Jackson after 12 days of submergence and 21 days of recovery. No mortality was reported

in C. gayana and P. coloratum following treatment whereas mortality in Frument was 100% after 12 days.

Rice submergence tolerance is very cultivar dependent. Intolerant IR42 displayed 70%, 50% and 0% survival

after 8, 10 and 12 days of winter submergence (temperature 22-24 °C), respectively, and FR13A exhibited

90% survival following 12 days of complete submergence (Das et al., 2009). In the present study, Frument

showed survival rates resembling submergence intolerant rice cultivar IR42 (Fig. 1). Meanwhile, survival in

Jackson following 12 days of submergence was 100% compared to 90% in submergence-tolerant rice

cultivar FR13A (Das et al., 2009). However, shoot biomass was severely reduced in wheat (5-17% of

controls in Frument and Jackson) compared to rice (44-60% of controls in IR42 and FR13A) after 10 days of

submergence and recovery (Das et al., 2009). The relatively large shoot biomass reductions in wheat

compared to rice, C. gayana and P. coloratum could reflect the in comparison low waterlogging tolerance of

wheat (Nishiuchi et al., 2012; Imaz et al., 2013) resulting in lower submergence tolerance as well. Possible

traits conferring higher flooding tolerance in rice, C. gayana and P. coloratum are root porosities > 35%

compared to 13-22% in wheat (Colmer, 2003; Imaz et al., 2013) and a strong barrier to ROL in rice but not

found in wheat (Colmer, 2003). It should be noted that complete submergence of winter wheat at low

temperatures during winter dormancy is likely to result in less detrimental effects as shown for waterlogging

(Luxmoore et al., 1973; Trought and Drew, 1982).

Frument displayed faster leaf degradation than Jackson as evident from earlier leaf lamina chlorophylla and

leaf tissue porosity declines (Fig. 2). In rice, leaf chlorosis upon submergence is triggered by ethylene

accumulation (Jackson et al., 1987), leading to lower photosynthesis (Smith et al., 1988; Winkel et al.,

2014). Submergence-tolerant rice cultivars (FR13A, M202(Sub1)) had lower activity and expression of

chlorophyllase due to lower ethylene sensitivity than intolerant cultivars (IR42, M202) resulting in more

severe leaf chlorosis in the latter upon submergence (Smith et al., 1988; Ella et al., 2003; Fukao et al., 2006;

51


Panda et al., 2008). This was linked to the ethylene-driven expression of SUB1A, an ethylene response factor

(ERF) transcription factor which limits gibberellin-mediated elongation growth promoted by ethylene (Xu et

al., 2006; Bailey-Serres et al., 2012). The earlier leaf chlorosis and higher levels of the chlorophyll

degradation product phytol (Matile et al., 1999) in Frument could be indicative of higher ethylene sensitivity

or production (shown to vary between Jackson and waterlogging sensitive cultivar Bayles, Huang et al.

(1997)) in Frument; however, further studies are needed to clarify this aspect.

Lower chlorophylla concentrations in Frument relative to Jackson did not result in a corresponding decline in

underwater PN, indicating that light harvest was not the rate-limiting step in Frument PN. Although it is

possible that low chlorophylla in Frument resulted in lower PN during recovery (as seen in two rice cultivars

with different chlorophyll levels after de-submergence, Alpuerto et al. (2016)), this was not assessed in the

present study. Other factors limiting underwater PN could be damage to the photosynthetic apparatus as

indicated by FV/FM ratios of 0.65-0.67 by day 12 (Fig. 2). Indeed, wheat underwater PN at diagnostic high

external CO2 (2500 µM) did indicate damage to the photosynthetic apparatus after 4 days of complete

submergence when rates declined to ~25% of initials (Konnerup et al., 2017). In submergence intolerant rice

cultivar IR42, FV/FM declined to similarly low levels as in both wheat cultivars (~0.70) after 8 days of

complete submergence, while remaining high in tolerant FR13A. Meanwhile, FV/FM did not separate rice

cultivars M202 and M202(Sub1) during 3 days of submergence (Alpuerto et al., 2016) or Jackson and

Frument in the present study. CO2 limitations caused by leaf gas film loss by day 5 would also hamper PN

(Verboven et al., 2014; Konnerup et al., 2017) but leaf gas film retention times did not differ between

cultivars in the present study (Fig. 2) or in the 14 wheat cultivars studied by Konnerup et al. (2017).

Leaf lamina tissue porosity decreased on day 12 in Frument and on day 16 in Jackson, thereby coinciding

with the time point when plant survival in both cultivars reached 0% upon recovery. Leaf tissue porosity

decline in submerged terrestrial plants has been interpreted as indicating structural leaf degradation (Winkel

et al., 2014; Konnerup et al., 2017), i.e. loss of leaf hydrophobicity, cuticle deterioration, solute leakage and

cell turgor loss allowing water to infiltrate intercellular gas filled spaces, but the sequence of events leading

to porosity decrease remains to be assessed. In our experience, leaves that have lost porosity do not recover

upon de-submergence resulting in senesced shoot tissues (Herzog et al., 2017; Konnerup et al., 2017; Winkel

et al., 2017).

52


Frument had significantly higher levels of mRNA encoding sucrose and fructan degrading enzymes (Fig. 6),

explaining why the initially higher levels of soluble carbohydrates (Fig. 3) and sucrose (Fig. 3) in Frument

reached similar levels in both cultivars by day 2 of submergence. Although Frument continued to

overexpress sucrose synthases until day 12 compared to Jackson, sucrose remained at similar levels in the

two cultivars as they had already approached zero by day 5. In addition, the patterns of shoot starch, glucose

and fructose declines were similar in the two cultivars (Fig. 3 and Fig. S4). Shoot soluble carbohydrates

differed initially (day 0), but cultivars had very similar shoot carbohydrate concentrations until day 12 of

submergence, thereby not indicating that different carbohydrate consumption rates explain the contrasting

submergence tolerance in Frument and Jackson. The similar carbohydrate consumption patterns in these two

wheat cultivars contrast with rice where submergence-tolerant cultivars maintained significantly higher

levels of starch and soluble carbohydrates than intolerant cultivars during submergence (Das et al., 2005;

Winkel et al., 2014).

PCA showed that Frument metabolites changed faster relative to Jackson (Fig. 5), indicating that the

Frument metabolome was generally more affected by submergence than Jackson. The higher levels of free

amino acids in Frument than Jackson shoot tissues (up until shoot disintegration) resemble amino acid

accumulation to higher levels in submergence intolerant rice cultivar M202 than tolerant M202(Sub1) during

3 days of submergence (Barding et al., 2013; Alpuerto et al., 2016). However, the current study does not

support the suggestion in those two studies that, at least during prolonged submergence, differences in

cultivar amino acid concentrations are linked to difference in carbohydrate consumption. Alternative

explanations could be higher protein degradation and/or lower protein synthesis in Frument. During low O2,

pyruvate can be converted to Ala and prevent carbon loss to ethanol and lactic acid, but in the present study,

Ala did not accumulate in the shoot tissues as seen in submerged rice (Barding et al., 2013; Alpuerto et al.,

2016). Accumulation of amino acids, especially Pro which accumulated in Jackson but not in Frument, may

also serve as osmoprotectants compensating for the loss of soluble carbohydrates during submergence or

anoxia (Magneschi and Perata, 2009; Alpuerto et al., 2016). Moreover, Pro is also considered a powerful

antioxidant (Verbruggen and Hermans, 2008). In the present study, MDA (a measure of cell membrane

damage by ROS) indicated that Frument experienced significantly higher levels of lipid peroxidation than

Jackson from day 12-16, while MDA levels in Jackson remained low even on day 16 when survival rates had

started to decrease. ROS damage during submergence could occur when O2 is low during the night and high

during the day as seen in rice field floodwaters where diurnal pO2 ranged from 5-19 kPa (Winkel et al.,

2013).

53


Conclusion

Wheat cultivars showed contrasting submergence tolerance as evident from significantly higher survival and

growth in Jackson than Frument following submergence and recovery. A clear relationship between

submergence tolerance and steady-state level of shoot carbohydrates was not evident. However, the rate of

initial carbohydrate consumption linked with expression of genes encoding sucrose and fructan catabolism

enzymes was associated with submergence tolerance. Metabolomics analysis confirmed that cultivars

experienced similarly rapid declines in shoot sugar levels, and revealed accumulation of most amino acids.

Thus, cultivar Jackson tolerates longer periods of low shoot carbohydrate levels than Frument. Metabolic

fingerprinting revealed that the Frument metabolome changed faster upon submergence than Jackson. These

metabolic changes under submergence were in accordance with faster leaf senescence end deterioration as

evident from leaf chlorophylla and leaf tissue porosity data. However, leaf chlorosis did not result in lower

underwater PN in Frument relative to Jackson. Elevated levels of MDA indicated that Frument experienced

higher levels of ROS-inflicted membrane damage at the end of the submergence period. Greater

accumulation of proline in Jackson may partly contribute to the suppression of lipid peroxidation during

submergence but further studies monitoring other antioxidant metabolites and enzymes are required to

evaluate the ROS detoxification mechanism in tolerant and intolerant cultivars.

Materials and Methods

Plant culture

Seeds of wheat (Triticum aestivum L., cv. ‘Frument’ and ‘Jackson’) were imbibed for three hours in aerated

0.5 mM CaSO4 and germinated for 48 hours in Petri dishes on wet paper towels in darkness at 20 °C.

Cultivars Jackson and Frument were used due to their contrasting waterlogging tolerance, assuming that

waterlogging and submergence tolerance could be linked. Jackson has been reported as waterlogging and

hypoxia tolerant (Huang and Johnson, 1995; Huang et al., 1997) while Frument emerged as waterlogging

intolerant when screening Danish (Frument, Jensen, Mariboss) and international reference cultivars (Jackson,

Chara, Nishikazekumogi) for waterlogging tolerance in preliminary pot experiments.

Three germinated seeds were sown at 10 mm depth in each of 190 round pots with drainage holes (height,

120 mm; diameter, 90 mm) filled with substrate (specified below) and irrigated with deionized (DI) water. In

order to obtain pots containing either three plants (providing sufficient plant material for a range of

measurements, see below) or one plant (allowing for growth and survival analysis with minimum pot

54


effects), part of the seedlings were thinned to one per pot before treatment. To prevent roots from exiting

through the drainage holes and into the aerated flood water, the bottom of the pots were sealed with two

layers of landscape fabric (Plantex, DuPont Ltd., Hertfordshire, UK) fitted to the bottom with silicone. No

roots were seen exiting the drainage holes throughout the experiment. Plants were grown September-October

2016 in Copenhagen, Denmark in a glasshouse (daytime temperature 14-25 °C, night time temperature 14-19

°C, relative humidity 20-70%) before moving to a constant temperature room for treatments.

Substrate in the pots consisted of 20 mm of washed sand at the bottom, a commercial potting mix with a dry

matter content of 55-75 g l-1, conductivity 3.0-5.0 mS cm-1 and pH 6.0 (Pindstrup Substrate no. 2, Pindstrup

Mosebrug A/S, Ryomgaard, Denmark) and a 20 mm layer of washed sand on top to reduce the flux of soil

derived nutrients into the floodwater upon submergence. Each pot received 1g Osmocote slow release

fertilizer (Osmocote Bloom, Everris, Geldermalsen, The Netherlands) that contained (by % mass): N, 12; P,

7.0; K, 18.0; Mg, 1.5; Fe, 0.35; Mn, 0.05; Cu, 0.045; Mo, 0.017; Zn, 0.013; B, 0.01. In a recent study using

identical pots and substrate, O2 disappeared from the soil matrix within 6-22 hours of soil flooding (Winkel

et al., 2017). To control powdery mildew shoots were sprayed with a 2 g L-1 sulfur solution (ECOstyle

Svampefri, ECOstyle A/S, Odense, Denmark) 9, 19 and 39 days after imbibition and Flexity (Metratenon;

0.15 g L-1) 15 and 41 days after imbibition.

Experimental design and treatments

The study consisted of two treatments (‘completely submerged’ and ‘controls in air’) × 2 wheat cultivars ×

4-8 replicates × 0-19 days of treatment in a 2 × 2 × 7 factorial design. Four glass aquariums (length × width

× height, 800 mm × 400 mm × 500 mm) filled with submergence solution (composition given below) in a

constant temperature room (20 °C, relative humidity 40-89%) served as tanks for submergence. Light (PPFD

of approximately 450 µmol m-2 s-1 at canopy level in the filled tanks, day/night cycle 12 h/12 h) was

provided by two light panels (AkvaStabil Effektline AL 39x2, AkvaStabil, Haderslev, Denmark) per tank

each holding two TL5 39 W fluorescent tubes. Drained controls in air grew in the same constant temperature

room as the submerged plants and received the same PPFD, but from a different light source (Gavita Pro

LEP 300, Gavita Holland BV, Aalsmeer, the Netherlands). The submergence solution was a modified Smart

and Barko (1985) ‘artificial floodwater’ solution containing (in mM): CaCl2.2H2O, 0.62; MgSO4.7H2O, 0.28;

KHCO3, 2.0. The submergence solution was continuously bubbled with air and was replaced by fresh

solution eight days after start of treatment.

Treatments commenced 18 days after imbibition when all plants had a fully expanded third leaf. 128 pots

containing one or three plants of each genotype were fully submerged by placing pots into aquariums filled

with submergence solution. Plants were randomly assigned to the four tanks, making sure that an even

55


distribution allowed for harvest of replicates from each tank at all time points. Additional plants subject to

drained conditions served as ‘controls in air’. One pot holding three plants was harvested from each tank

after 0, 2, 5, 8, 12, 14 and 16 days of submergence. One of three plants from each pot (n = 4) was used for

underwater net photosynthesis, dark respiration, gas film thickness, leaf tissue porosity, chlorophylla and

shoot carbohydrates measurements while the remaining two plants were harvested for gene expression and

metabolomic analysis, respectively. For growth and survival analysis, 1-2 pots holding one plant each were

desubmerged from each tank for recovery (n = 4-8) after 0, 8, 10, 12, 14, 16 and 19 days of submergence.

Details on measurements are given in the following sections.

Growth parameters

Pots holding one plant were moved to empty aquariums with light panels on top and watered with DI water

for 14-25 days of recovery (Striker, 2012) following submergence. Recovery duration depended on the

preceding submergence event: plants submerged for only 8 days received a longer recovery period (25 days

recovery) than plants submerged for 19 days (14 days recovery) in order to harvest equally old plants 51

days after imbibition. Plants were scored as ‘survived’ (green leaf tissue present) or ‘dead’ (no green leaf

tissue present) and shoots were oven-dried at 60 °C for 48 h before weighing the DM. RGR were calculated

as RGR = (lnW2-lnW1)/(t2-t1), where W1 and W2 are the initial and final weight, respectively, and t1 and t2

are the initial and final time (days), respectively. In order to determine shoot growth during submergence

without a recovery period, four replicates were harvested immediately after eight days of submergence and

DM recorded. Shoot length was measured on submerged and control plants randomly selected from each

tank after 2, 5 and 8 days of treatment.

Underwater net photosynthesis and dark respiration measurements

Underwater net photosynthesis and dark respiration by lamina segments were measured using the principles

described by Pedersen et al. (2013). In brief, for each replicate leaf (n = 4), one lamina segment of

approximately 25 mm length was taken half way up the blade of the youngest fully expanded (third) leaf at

the time of submergence. Leaf tips of the youngest fully expanded leaf had prior to submergence been

marked with a permanent marker to enable identification of leaves throughout the experiment. Glass cuvettes

(approx. 28 mL) contained individual lamina segments in incubation medium (see submergence solution

above for composition) and two glass beads for mixing as the cuvettes rotated on a wheel within an

illuminated water bath (PN) or in darkness (RD), at 20 °C.

56


The dissolved O2 concentration in the PN incubation medium was initially set at approximately 50% of air

equilibrium, by bubbling in 1:1 volumes of N2 and air; this reduced build-up of excess O2 during the

incubation that might otherwise have resulted in photorespiration. For RD, the incubation medium was

initially adjusted to air equilibrium by purging with air to obtain RD at non-limiting external O2

concentrations (Colmer and Pedersen, 2008). Initial concentrations of 200 µM CO2 in both solutions (PN and

RD) was obtained by adjusting pH to 7.35 after adding 2.2 mmol KHCO3 L-1 solution. 200 µM CO2 is

considered an environmentally relevant CO2 concentration (Colmer et al., 2011) and allowed direct

comparison with other studies on PN in submerged wheat (Konnerup et al., 2017; Winkel et al., 2017). The

final solution consequently contained 200 µM CO2, 2 mM HCO3- and 2.2 mM K+ with an alkalinity of 2.0

mM H+ equivalents L-1 (Stumm and Morgan, 1996).

Following incubation with photosynthetically active radiation inside the vials of 1000 µmol photons m–2 s–1

provided by a light source (Gavita Pro LEP 300, Gavita Holland BV, Aalsmeer, Netherlands) for 60 to 90

min (PN) or 120-180 min (RD), dissolved O2 concentrations in the cuvettes were measured using a mini O2

optode (OP-MR, Unisense A/S, Aarhus, Denmark), connected to an optode meter (MicroOptode meter,

Unisense A/S, Aarhus, Denmark) that was calibrated at 20.6 kPa (air bubbled DI water at 20 °C, 283.9 µmol

O2 L-1) and at 0 kPa (DI water with ascorbate and KOH at 20 °C). Dissolved O2 concentrations in cuvettes

prepared and incubated in the same way as described above, but without leaf tissues, served as blanks. The

projected area of leaf segments was measured by scanning the segments (bizhub C454e, Konica Minolta,

Tokyo, Japan), analyzed digitally using ImageJ (Schneider et al., 2012), frozen at -20 °C, freeze-dried (Type

1102, Martin Christ Gefriertrocknungsanlagen GmbH, Osterode, Germany) and DM recorded.

Chlorophyll fluorescence

Maximum photochemical quantum yield of PS II (FV/FM) was measured halfway up of the blade of the

youngest fully expanded leaf at the time of submergence. Measurements were performed on the same four

replicates of each cultivar subject to submergence or serving as controls in air control on day 0, 5, 8 and 12.

Chlorophyll fluorescence was measured using a chlorophyll fluorometer (Junior-PAM, Heinz Walz GmbH,

Effeltrich, Germany) following 20 min of dark acclimation.

Chlorophyll concentration

The lamina of the third leaf at the time of submergence (consisting of freeze-dried leaf segments from gas

film thickness, tissue porosity and gas exchange measurements) was homogenized by cutting into max 2 × 2

mm pieces using scissors. Chlorophyll was extracted for 24 h in 96% ethanol at 20 °C in darkness,

57


centrifuged at 4500 rpm for 3 min and A656 and A750 measured (Shimadzu UV-1800, Shimadzu Corp., Kyoto,

Japan). Chlorophylla concentrations were calculated using equations of Mackinney (1941).

Leaf gas film thickness and tissue porosity

The leaf gas film volume (Winkel et al., 2014) and tissue porosity (gas-filled volume per unit tissue volume)

was measured using the “buoyancy method” (Raskin, 1983; Thomson et al., 1990) on 70 mm segments of

the youngest fully expanded (third) leaf lamina at time of submergence. The leaf segment area was measured

and DM recorded as described above. Mean gas film thickness was calculated by dividing gas film volume

(mm3) with the two-sided area (mm2) as wheat leaves possess gas films on both the adaxial and abaxial sides

(Konnerup et al., 2017).

Shoot carbohydrate assays

The entire shoot remaining after excision of the third leaf, which had been used for leaf gas film thickness,

tissue porosity and gas exchange measurements (described above), was excised below water, rinsed in DI

water, blotted dry on paper towels, placed into perforated aluminium foil bags, flash-frozen in liquid nitrogen

and stored at -80 °C. Shoots were freeze-dried and homogenized in a 2 mL Eppendorf tube using two metal

beads for 20 sec on a mini bead-beater (MiniBead Beater, BioSpec Products Inc., Bartlesville, OK, USA).

Shoot soluble sugars and starch were analyzed following Alpuerto et al. (2016). Ethanol soluble

carbohydrates were extracted by incubating 20 mg of ground tissue in 1 mL 80% (v/v) ethanol at 80 °C for

20 min. After centrifugation (10 min at 20,800 g) the supernatant was removed and the extraction was

repeated twice more. For extraction of water soluble carbohydrates the remaining pellet was re-suspended in

1 mL DI water and incubated at 80 °C for 20 min. After centrifugation (10 min at 20,800 g) the supernatant

was removed and the extraction was repeated once more. Extracts containing ethanol and water-soluble

carbohydrates were pooled in separate pre-weighed Eppendorf tubes and weighed for determination of

extract volumes. Ethanol and water-soluble sugars were measured using the anthrone-method with glucose

as the standard (Pontis, 2017). 50-200 µL extract was mixed with 200-50 µL ethanol to a final volume of 250

µL, mixed with 2.5 mL 0.2 % (w/v) anthrone solution in 96% H2SO4, incubated at 100 °C for 10 min, rapidly

cooled and A620 measured on a spectrophotometer (Shimadzu UV-1800, Shimadzu Corp., Kyoto, Japan). For

starch determination the pellet remaining from the extraction of soluble sugars was dried under vacuum and

re-suspended in 1 mL of water containing 10 units of heat-resistant α-amylase. After incubation at 95 °C for

30 min, the suspension was mixed with 25 µL 1M sodium citrate (adjusted to pH 4.8) and five units of

amyloglucosidase. After incubation at 55°C for 1 h, the reaction mixture was centrifuged (30 min at 20,800

g) and glucose content in the supernatant (100-200 μL) was quantified by the anthrone method described

above. Complete degradation of starch into glucose was confirmed by coloring test reaction mixtures with

58


Lugol’s iodine. Cultivars and time points were spread across independent extraction and measurement

procedures to avoid systematic errors.

Quantitative RT-PCR

One entire shoot from a pot of initially three plants (of which the first was used for measurement described

above) was harvested as described for “Shoot carbohydrate assays”. Shoots were freeze-dried for 72 h before

shipment on silica gel and with ice packs for gene expression analysis in Blacksburg, VA, USA. Freeze-

drying has been used in different plant (Román et al., 2012) and animal (Wu et al., 2012) species without

loss of RNA quality or integrity. RNA extraction, cDNA synthesis and qRT-PCR were performed as

described in Fukao and Bailey-Serres (2008). Total RNA was extracted using the RNeasy plant mini kit

(Qiagen, Hilden, Germany). Genomic DNA was eliminated by on-column DNase treatment using a

manufacturer’s protocol. Single-stranded cDNA was synthesized from 2 µg of total RNA using SuperScript

IV reverse transcriptase and oligo dT primer (Thermo Scientific, Waltham, MA, USA). qRT-PCR was

conducted in a 15 µL reaction using iTaq Universal SYBR Green Supermix (Bio-Rad, Hercules, CA, USA)

in the CFX Connect real-time PCR detection system (Bio-Rad). Amplification specificity was validated by

melt-curve analysis at the end of each PCR experiment. Relative transcript abundance was calculated by the

comparative CT method (Livak and Schmittgen, 2001). TaRP15, RNA polymerases I, II, and III, 15-kD

subunit (TC265122) were used as a reference gene (Xue et al., 2008). Primer sequences and annealing

temperatures used for qRT-PCR are listed in Table S1.

Metabolomics

One entire shoot was harvested as described for “Quantitative RT-PCR”, stored at -80°C and homogenized

in liquid N2 using mortar and pestle. Metabolites were extracted and analyzed as described in Hasler-Sheetal

et al. (2016) with slight modifications. In brief, 50 mg homogenized shoot material was extracted (2 min in

an ultrasound bath followed by 15 min on a thermo shaker both at 4°C) in 1 ml methanol/acetonitrile/water

[4:4:2] at -20°C (spiked with 0.4 mg/l 13C6 Sorbitol and Reserpine as internal standards). After centrifugation

(19,000 g for 5 min) the metabolites in the supernatants were analyzed by gas chromatography quadrupole

time of flight mass spectrometry (GC-MS; 7200 GC QTOF MS) and liquid chromatography quadrupole time

of flight mass spectrometry (LC-MS; 1290LC, 6530 QTOF MS) (both Agilent Technologies, Santa Clara,

CA, USA), following Hasler-Sheetal et al. (2015); Hasler-Sheetal et al. (2016) and Yonny et al. (2017) with

slight modifications.

59


Data analysis

Data were analyzed with GRAPHPAD PRISM version 7.02 (GraphPad Software, La Jolla, CA, USA) and R

(R Core Team, 2014) for Windows statistical software. A significance level of P < 0.05 was used for all

analyses. Normality of distributions was confirmed by visual inspections and Shapiro-Wilk normality test,

and variance homogeneity using F-test or Brown-Forsythe test (P > 0.05). Data of leaf gas film thickness,

soluble carbohydrates, starch and GC-MS and LC-MS metabolites requiring transformation were ln-

transformed in order to improve variance homogeneity. Transformations were unsuccessful in improving

variance homogeneity in rates of underwater PN, but as sample sizes were equal making ANOVA robust to

unequal variances (Prophet Statguide, 1997; Graphpad Software Inc., 2013), we considered application of

ANOVA on untransformed data appropriate. Data sets were subject to two-way (time × cultivar) fixed factor

ANOVA and when significant main effects were found Sidak’s multiple comparisons post-hoc test was

performed. Chlorophyll fluorescence data were subject to two-way repeated measures ANOVA. Shoot

length measurements of each genotype were subject to one-way ANOVA with Tukey’s multiple

comparisons test. Shoot RGR during 8 days of submergence (no recovery) and relative mRNA levels were

analyzed using unpaired t-tests.

Using the amount of plants surviving upon submergence and recovery as the response variable (coded as 0,

dead or 1, survived), the difference in cultivar survival was tested using a linear logistic regression model, a

Generalized Linear Model (GLM) with binomial error structure and logit link function. Duration of

submergence was used as the explanatory variable and survival was analyzed for both cultivars separately.

When fitting the model we used Firth-type penalized likelihood estimation instead of maximum likelihood

estimation due to otherwise inflated standard errors caused by near separation into 0 and 100% survival in

Jackson (Heinze and Schemper, 2002).

List of Supplemental Material

Table S1. Primer sequences and annealing temperatures used in qRT-PCR.

Fig. S1. Shoot RGR of wheat cultivars Frument and Jackson during 8 days of submergence (no recovery

period).

Fig. S2. Leaf underwater photosynthesis and dark respiration of wheat cultivars Frument and Jackson with

time of submergence.

60


Fig. S3. GC-MS determined sugars in shoots of wheat cultivars Frument and Jackson with time of

submergence.

Fig. S4. Starch in shoots of wheat cultivars Frument and Jackson with time of submergence.

Fig. S5. Metabolites detected by LC-MS in shoots of wheat cultivars Frument and Jackson during

submergence, presented as Venn diagram for time, cultivar and time × cultivar interaction effects.

Fig. S6. GC-MS determined amino acids in shoots of wheat cultivars Frument and Jackson with time of

submergence.

Fig. S7. Relative mRNA levels of α-amylases in shoots of wheat cultivars Frument and Jackson with time of

submergence.

Acknowledgements

The authors would like to thank Timothy D. Colmer for discussing experimental procedures and data

interpretation, Lars Iversen for statistical support and Victoria C. Herskov for helping with analyses of

chlorophyll. M.H., D.K. and A.W. were supported by PhD and postdoctoral fellowships from the Villum

Foundation.

Literature cited

Albrecht G, Wiedenroth EM (1994) Protection against activated oxygen following re-aeration of hypoxically pretreated wheat roots - the response of the glutathione system. Journal of Experimental Botany 45: 449-455

Alpuerto JB, Hussain RMF, Fukao T (2016) The key regulator of submergence tolerance, SUB1A, promotes photosynthetic and metabolic recovery from submergence damage in rice leaves. Plant, Cell & Environment 39: 672-684

Armstrong W (1979) Aeration in higher plants. In HW Woolhouse, ed, Advances in Botanical Research, Vol 7. Academic Press, London, pp 225-332

Bailey-Serres J, Lee SC, Brinton E (2012) Waterproofing crops: effective flooding survival strategies. Plant Physiology 160: 1698-1709

Barding GA, Jr., Beni S, Fukao T, Bailey-Serres J, Larive CK (2013) Comparison of GC-MS and NMR for metabolite profiling of rice subjected to submergence stress. Journal of Proteome research 12: 898-909

Blokhina O, Virolainen E, Fagerstedt KV (2003) Antioxidants, oxidative damage and oxygen deprivation stress: a review. Annals of Botany 91: 179-194

Collaku A, Harrison SA (2002) Losses in wheat due to waterlogging. Crop Science 42: 444-450

61


Colmer TD (2003) Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant, Cell & Environment 26: 17-36

Colmer TD, Pedersen O (2008) Underwater photosynthesis and respiration in leaves of submerged wetland plants: gas films improve CO2 and O2 exchange. New Phytologist 177: 918-926

Colmer TD, Voesenek LACJ (2009) Flooding tolerance: suites of plant traits in variable environments. Functional Plant Biology 36: 665-681

Colmer TD, Winkel A, Pedersen O (2011) A perspective on underwater photosynthesis in submerged terrestrial wetland plants. AoB Plants 2011: plr030; doi:010.1093/aobpla/plr1030

Das KK, Panda D, Sarkar RK, Reddy JN, Ismail AM (2009) Submergence tolerance in relation to variable floodwater conditions in rice. Environmental and Experimental Botany 66: 425-434

Das KK, Sarkar RK, Ismail AM (2005) Elongation ability and non-structural carbohydrate levels in relation to submergence tolerance in rice. Plant Science 168: 131-136

Dickin E, Bennett S, Wright D (2009) Growth and yield responses of UK wheat cultivars to winter waterlogging. The Journal of Agricultural Science 147: 127-140

Ella ES, Kawano N, Yamauchi Y, Tanaka K, Ismail AM (2003) Blocking ethylene perception enhances flooding tolerance in rice seedlings. Functional Plant Biology 30: 813-819

Fukao T, Bailey-Serres J (2008) Submergence tolerance conferred by Sub1A is mediated by SLR1 and SLRL1 restriction of gibberellin responses in rice. Proceedings of the National Academy of Sciences 105: 16814-16819

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

Gibbs J, Greenway H (2003) Mechanisms of anoxia tolerance in plants. I. Growth, survival and anaerobic catabolism. Functional Plant Biology 30: 1-47

Graphpad Software Inc. (2013). What to do when data fail tests for homogeneity of variance [Online]. Available: http://www.graphpad.com/support/faqid/1007/ [Accessed 14.06.2017].

Hasler-Sheetal H, Castorani MC, Glud RN, Canfield DE, Holmer M (2016) Metabolomics reveals cryptic interactive effects of species interactions and environmental stress on nitrogen and sulfur metabolism in seagrass. Environmental Science & Technology 50: 11602-11609

Hasler-Sheetal H, Fragner L, Holmer M, Weckwerth W (2015) Diurnal effects of anoxia on the metabolome of the seagrass Zostera marina. Metabolomics 11: 1208-1218

Hayashi T, Yoshida T, Fujii K, Mitsuya S, Tsuji T, Okada Y, Hayashi E, Yamauchi A (2013) Maintained root length density contributes to the waterlogging tolerance in common wheat (Triticum aestivum L.). Field Crops Research 152: 27-35

Heinze G, Schemper M (2002) A solution to the problem of separation in logistic regression. Statistics in Medicine 21: 2409-2419

Herzog M, Konnerup D, Pedersen O, Winkel A, Colmer TD (2017) Leaf gas films contribute to rice (Oryza sativa) submergence tolerance during saline floods. Plant, Cell & Environment: In press

Herzog M, Pedersen O (2014) Partial versus complete submergence: snorkelling aids root aeration in Rumex palustris but not in R. acetosa. Plant, Cell & Environment 37: 2381-2390

Herzog M, Striker GG, Colmer TD, Pedersen O (2016) Mechanisms of waterlogging tolerance in wheat – a review of root and shoot physiology. Plant, Cell & Environment 39: 1068-1086

Hodges DM, DeLong JM, Forney CF, Prange RK (1999) Improving the thiobarbituric acid-reactive-substances assay for estimating lipid peroxidation in plant tissues containing anthocyanin and other interfering compounds. Planta 207: 604-611

Huang B, Johnson JW, Box JE, NeSmith DS (1997) Root characteristics and hormone activity of wheat in response to hypoxia and ethylene. Crop Science 37: 812-818

Huang BR, Johnson JW (1995) Root respiration and carbohydrate status of 2 wheat genotypes in response to hypoxia. Annals of Botany 75: 427-432

62

http://www.graphpad.com/support/faqid/1007/


Huang BR, Johnson JW, Nesmith S, Bridges DC (1994) Growth, physiological and anatomical responses of 2 wheat genotypes to waterlogging and nutrient supply. Journal of Experimental Botany 45: 193-202

Imaz JA, Giménez DO, Grimoldi AA, Striker GG (2013) The effects of submergence on anatomical, morphological and biomass allocation responses of tropical grasses Chloris gayana and Panicum coloratum at seedling stage. Crop and Pasture Science 63: 1145-1155

Ismail AM, Singh US, Singh S, Dar MH, Mackill DJ (2013) The contribution of submergence-tolerant (Sub1) rice varieties to food security in flood-prone rainfed lowland areas in Asia. Field Crops Research 152: 83-93

Jackson MB, Waters I, Setter TL, Greenway H (1987) Injury to rice plants caused by complete submergence; a contribution by ethylerie (ethene). Journal of Experimental Botany 38: 1826-1838

Konnerup D, Winkel A, Herzog M, Pedersen O (2017) Leaf gas film retention during submergence of 14 cultivars of wheat Triticum aestivum. Functional Plant Biology: In press

Li C, Jiang D, Wollenweber B, Li Y, Dai T, Cao W (2011) Waterlogging pretreatment during vegetative growth improves tolerance to waterlogging after anthesis in wheat. Plant Science 180: 672-678

Lindahl A, Saaf S, Lehtio J, Nordstrom A (2017) Tuning metabolome coverage in reversed phase LC-MS metabolomics of MeOH extracted samples using the reconstitution solvent composition. Analytical Chemistry 89: 7356-7364

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods 25: 402-408

Luxmoore R, Fischer R, Stolzy L (1973) Flooding and soil temperature effects on wheat during grain filling. Agronomy Journal 65: 361-364

Mackinney G (1941) Absorption of light by chlorophyll solutions. Journal of Biological Chemistry 140: 315-322

Magneschi L, Perata P (2009) Rice germination and seedling growth in the absence of oxygen. Annals of Botany 103: 181-196

Malik A, Colmer TD, Lambers H, Schortemeyer M (2001) Changes in physiological and morphological traits of roots and shoots of wheat in response to different depths of waterlogging. Australian Journal of Plant Physiology 28: 1121-1131

Matile P, Hörtensteiner S, Thomas H (1999) Chlorophyll degradation. Annual Review of Plant Biology 50: 67-95

Mommer L, Pons TL, Wolters-Arts M, Venema JH, Visser EJW (2005) Submergence-induced morphological, anatomical, and biochemical responses in a terrestrial species affect gas diffusion resistance and photosynthetic performance. Plant Physiology 139: 497-508

Musgrave ME, Ding N (1998) Evaluating wheat cultivars for waterlogging tolerance. Crop Science 38: 90-97 Nishiuchi S, Yamauchi T, Takahashi H, Kotula L, Nakazono M (2012) Mechanisms for coping with

submergence and waterlogging in rice. Rice 5: 1-14 Panda D, Sharma SG, Sarkar RK (2008) Chlorophyll fluorescence parameters, CO2 photosynthetic rate and

regeneration capacity as a result of complete submergence and subsequent re-emergence in rice (Oryza sativa L.). Aquatic Botany 88: 127-133

Pedersen O, Colmer TD, Sand-Jensen K (2013) Underwater photosynthesis of submerged plants – recent advances and methods. Frontiers in Plant Science 4: 1-19

Pellegrini E, Konnerup D, Winkel A, Casolo V, Pedersen O (2017) Contrasting oxygen dynamics in Limonium narbonense and Sarcocornia fruticosa during partial and complete submergence. Functional Plant Biology: In press

Ponnamperuma F (1984) Effects of flooding on soils. In TT Kozlowski, ed, Flooding and plant growth. Academic Press, London, pp 9-45

Pontis HG (2017) Chapter 1 - determination of carbohydrates metabolism molecules. In Methods for analysis of carbohydrate metabolism in photosynthetic organisms. Academic Press, Boston, pp 3-27

Prophet Statguide (1997). Do your data violate one-way ANOVA assumptions? [Online]. Northwestern University, Chicago, IL. Available:

63


http://www.basic.northwestern.edu/statguidefiles/oneway_anova_ass_viol.html#Unequal%20population%20variances [Accessed 14.7. 2017].

R Core Team (2014) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria

Raskin I (1983) A method for measuring leaf volume, density, thickness, and internal gas volume. Hortscience: A Publication of the American Society for Horticultural Science 8: 698-699

Raskin I, Kende H (1983) How does deep-water rice solve its aeration problem. Plant Physiology 72: 447-454

Robertson D, Zhang H, Palta JA, Colmer T, Turner NC (2009) Waterlogging affects the growth, development of tillers, and yield of wheat through a severe, but transient, N deficiency. Crop and Pasture Science 60: 578-586

Román B, González-Verdejo CI, Peña F, Nadal S, Gómez P (2012) Evaluation of different pulverisation methods for rna extraction in squash fruit: Lyophilisation, cryogenic mill and mortar grinding. Phytochemical Analysis 23: 622-626

Samad A, Meisner CA, Saifuzzaman M, Ginkel Mv (2001) Waterlogging tolerance. In MP Reynolds, JI Ortiz Monasterio, A McNab, eds, Application of Physiology in Wheat Breeding. CIMMYT, Mexico, DF, Mexico, 2001, pp 136–144

Sasidharan R, Bailey-Serres J, Ashikari M, Atwell BJ, Colmer TD, Fagerstedt K, Fukao T, Geigenberger P, Hebelstrup KH, Hill RD, Holdsworth MJ, Ismail AM, Licausi F, Mustroph A, Nakazono M, Pedersen O, Perata P, Sauter M, Shih M-C, Sorrell BK, Striker GG, van Dongen JT, Whelan J, Xiao S, Visser EJW, Voesenek LACJ (2017) Community recommendations on terminology and procedures used in flooding and low oxygen stress research. New Phytologist: In press

Sayre K, Van Ginkel M, Rajaram S, Ortiz-Monasterio I (1994) Tolerance to waterlogging losses in spring bread wheat: effect of time of onset on expression. Annual Wheat Newsletter 40: 165-171

Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9: 671-675

Scofield GN, Ruuska SA, Aoki N, Lewis DC, Tabe LM, Jenkins CLD (2009) Starch storage in the stems of wheat plants: localization and temporal changes. Annals of Botany 103: 859-868

Setter TL, Waters I (2003) Review of prospects for germplasm improvement for waterlogging tolerance in wheat, barley and oats. Plant and Soil 253: 1-34

Shiono K, Takahashi H, Colmer TD, Nakazono M (2008) Role of ethylene in acclimations to promote oxygen transport in roots of plants in waterlogged soils. Plant Science 175: 52-58

Smart MR, Barko JW (1985) Laboratory culture of submersed freshwater macrophytes on natural sediments. Aquatic Botany 21: 251-263

Smith P, Kupkanchanakul K, Emes M, Cutter E (1988) Changes in fluorescence and photosynthesis during submergence of deepwater rice. International Deepwater Rice Workshop. 26-30 Oct 1987, Bangkok (Thailand)

Striker GG (2012) Time is on our side: the importance of considering a recovery period when assessing flooding tolerance in plants. Ecological Research 27: 983-987

Stumm W, Morgan JJ (1996) Aquatic chemistry: chemical equilibria and rates in natural waters, Ed 3. John Wiley & Sons, New York

Thomson CJ, Armstrong W, Waters I, Greenway H (1990) Aerenchyma formation and associated oxygen movement in seminal and nodal roots of wheat. Plant, Cell & Environment 13: 395-403

Trnka M, Rotter RP, Ruiz-Ramos M, Kersebaum KC, Olesen JE, Zalud Z, Semenov MA (2014) Adverse weather conditions for European wheat production will become more frequent with climate change. Nature Climate Change 4: 637-643

Trought MCT, Drew MC (1980) Development of waterlogging damage in wheat seedlings (Triticum-aestivum L). 1. Shoot and root-growth in relation to changes in the concentrations of dissolved-gases and solutes in the soil solution. Plant and Soil 54: 77-94

64

http://www.basic.northwestern.edu/statguidefiles/oneway_anova_ass_viol.html#Unequal%20population%20variances



Trought MCT, Drew MC (1982) Effects of waterlogging on young wheat plants (Triticum-aestivum L) and on soil solutes at different soil temperatures. Plant and Soil 69: 311-326

U.S. Risk Management Agency (2016). Cause of loss historical data files [Online]. Risk Management Agency, United States Department of Agriculture. Available: http://www.rma.usda.gov/data/cause.html [Accessed 2017].

Van Ginkel M, S. Rajaram & M. Thijssen. (1991) Waterlogging in wheat: Germplasm evaluation and methodology development. DG Tanner, W Mwangi, eds, The 7th Regional Wheat Workshop for Eastern, Central, and Southern Africa. CIMMYT, Najuru, Kenya

Verboven P, Pedersen O, Ho QT, Nicolai BM, Colmer TD (2014) The mechanism of improved aeration due to gas films on leaves of submerged rice. Plant, Cell & Environment 37: 2433-2452

Verbruggen N, Hermans C (2008) Proline accumulation in plants: a review. Amino Acids 35: 753-759 Visser EJW, Cohen JD, Barendse GWM, Blom CWPM, Voesenek LACJ (1996) An ethylene-mediated

increase in sensitivity to auxin induces adventitious root formation in flooded Rumex palustris Sm. Plant Physiology 112: 1687-1692

Voesenek LACJ, Blom CWPM (1989) Growth responses of Rumex species in relation to submergence and ethylene. Plant, Cell & Environment 12: 433-439

Winkel A, Colmer TD, Ismail AM, Pedersen O (2013) Internal aeration of paddy field rice (Oryza sativa) during complete submergence – importance of light and floodwater O2. New Phytologist 197: 1193-1203

Winkel A, Colmer TD, Pedersen O (2011) Leaf gas films of Spartina anglica enhance rhizome and root oxygen during tidal submergence. Plant, Cell & Environment 34: 2083-2092

Winkel A, Herzog M, Konnerup D, Floytrup AH, Pedersen O (2017) Flood tolerance of wheat – the importance of leaf gas films during complete submergence. Functional Plant Biology: In press

Winkel A, Pedersen O, Ella E, Ismail AM, Colmer TD (2014) Gas film retention and underwater photosynthesis during field submergence of four contrasting rice genotypes. Journal of Experimental Botany 65: 3225–3233

Wu Y, Wu M, Zhang Y, Li W, Gao Y, Li Z, Wang Z, Lubec G, Zhang C (2012) Lyophilization is suitable for storage and shipment of fresh tissue samples without altering RNA and protein levels stored at room temperature. Amino Acids 43: 1383-1388

Xu K, Xu X, Fukao T, Canlas P, Maghirang-Rodriguez R, Heuer S, Ismail AM, Bailey-Serres J, Ronald PC, Mackill DJ (2006) Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature 442: 705-708

Xue G-P, McIntyre CL, Jenkins CL, Glassop D, van Herwaarden AF, Shorter R (2008) Molecular dissection of variation in carbohydrate metabolism related to water-soluble carbohydrate accumulation in stems of wheat. Plant Physiology 146: 441-454

Yonny ME, Rodríguez Torressi A, Nazareno MA, Cerutti S (2017) Development of a novel, sensitive, selective, and fast methodology to determine malondialdehyde in leaves of melon plants by ultra-high-performance liquid chromatography-tandem mass spectrometry. Journal of Analytical Methods in Chemistry: In press

65

http://www.rma.usda.gov/data/cause.html

Chapter 3: Leaf gas films contribute to rice (Oryza sativa) submergence tolerance during saline floods

Rice leaves retaining leaf gas films on the submerged leaf sections. Photo: Ole Pedersen.

66

Original Article

Leaf gas films contribute to rice (Oryza sativa) submergencetolerance during saline floods

Max Herzog1,2, Dennis Konnerup1,2, Ole Pedersen1,2,4, Anders Winkel1,2 & Timothy David Colmer2,3

1Freshwater Biological Laboratory, Department of Biology, University of Copenhagen, Universitetsparken 4, 3rd floor, 2100Copenhagen, Denmark, 2School of Plant Biology, The University of Western Australia, 35 Stirling Highway, Crawley 6009, WesternAustralia, Australia, 3Institute of Agriculture, The University of Western Australia, 35 Stirling Highway, Crawley 6009, WesternAustralia, Australia and 4Institute of Advanced Studies, The University of Western Australia, 35 Stirling Highway, Crawley 6009,Western Australia, Australia

ABSTRACT

Floods and salinization of agricultural land adversely impactglobal rice production. We investigated whether gas films onleaves of submerged rice delay salt entry during saline submer-gence. Two-week-old plants with leaf gas films (+GF) or withgas films experimentally removed (�GF) were submerged inartificial floodwater with 0 or 50mM NaCl for up to 16d. Gasfilms were present>9d onGFplants after which gas films werediminished. Tissue ion analysis (Na+, Cl� and K+) showed thatgas films caused some delay of Na+ entry, as leaf Na+ concen-tration was 36–42% higher in �GF leaves than +GF leaveson days 1–5. However, significant net uptakes of Na+ andCl�, andK+ net loss, occurred despite the presence of gas films,indicating the likely presence of some leaf-to-floodwater con-tact, so that the gas layer must not have completely separatedthe leaf surfaces from the water. Natural loss and removal ofgas films resulted in severe declines in growth, underwater pho-tosynthesis, chlorophylla and tissue porosity. Submergence wasmore detrimental to leaf PN and growth than the additional ef-fect of 50mMNaCl, as salt did not significantly affect underwa-ter PN at 200μM CO2 nor growth.

Key-words: flooding; leaf Cl�; leaf K+; leaf Na+; plant submer-gence tolerance; salinity; salt intrusion.

INTRODUCTION

Floods annually affect large areas of farmlands worldwide andcause severe crop losses when plants become submerged(Jackson 2004). Crop damage is mainly caused by the ham-pered gas exchange between plants and floodwater becauseof a 104-fold slower gas diffusion and low solubility of O2 in wa-ter compared with that in air (Armstrong 1979; Voesenek et al.2006). Paddy field rice is adapted to growth in anoxic soils andtherefore is tolerant to soil waterlogging and even partial shootsubmergence (Colmer et al. 2014; Kirk et al. 2014). However,only a few days of complete submergence can lead to severedamage and death of rice (Das et al. 2009), but with important

differences among rice genotypes (Ismail et al. 2013). The re-stricted gas exchange impedes respiration and photosynthesis(also because of low light) in submerged shoots (Mommer &Visser 2005) while the consumption of soluble carbohydrates(Setter et al. 1997) further depletes tissue sugars and energy(if shoots elongate). The factors described previouslycontribute to damage during floods, together with atdesubmergence water deficits (Setter et al. 2010) and oxidativestress (Bailey-Serres & Voesenek 2008) that can result infurther damage and even death.

Floodwaters may contain NaCl, and salinity is a major im-pediment to increasing global rice production (Negrão et al.2011) as rice is a salt-sensitive species. For rice with shoots inair, salinity above 30mM NaCl results in yield decreases by12% for each ~10mM NaCl increase (Grieve et al. 2012).Salinity imposes both an osmotic stress on the plant becauseof high solute concentrations outside cells, as well as ion-specific stresses caused by high Na+ and Cl� concentrations inplant tissues (Munns & Tester 2008; Negrão et al. 2011). Theneed to improve rice salinity and submergence tolerance is fur-ther urged by climate change causing rising seawater levels andlower river flows, leading to seawater inundation of large ricegrowing regions such as the Vietnamese Mekong Delta(Wassmann et al. 2004). A second example is that the salinityaffected areas in Bangladesh increased from about 83 millionha in 1973 to 106 million ha in 2009 (Sinha et al. 2014). Thus, ef-forts are being made to combine submergence tolerance andsalinity tolerance in the so-called climate-smart rice (DeOcampo et al. 2013; IRRI 2016).

Rice leaves are surrounded by a gas film (initial averagethickness of 50–62μm, Pedersen et al. 2009; Winkel et al.2013; Winkel et al. 2014) for up to 6d during submergence inthe field (Winkel et al. 2014). Presence of such gas films delayedsalt entry into submerged leaves of Melilotus siculus (Teakleet al. 2014), but this is the only study to have evaluated this ef-fect. Leaf gas films have been shown to enhance underwaterphotosynthesis of rice, dark respiration, root pO2 and growth,by greatly enhancing gas exchange between leaves and flood-water (Pedersen et al. 2009; Winkel et al. 2013; Verbovenet al. 2014; Winkel et al. 2014), thereby contributing to rice sub-mergence tolerance. Our main objective was to test the effectof the presence of leaf gas films on salt entry into submerged

Correspondence: M. Herzog, Freshwater Biological Laboratory,Department of Biology, University of Copenhagen, Universitetsparken4, 3rd floor, 2100 Copenhagen, Denmark. E-mail: [email protected]

© 2016 John Wiley & Sons Ltd 1

doi: 10.1111/pce.12873Plant, Cell and Environment (2017)

67

rice; we hypothesized that Na+ and Cl� entry, and K+ loss,would be delayed by the presence of leaf gas films acting asan ‘insulating’ physical barrier between each leaf and salinefloodwater.

MATERIALS AND METHODS

Plant culture

Seeds of rice (Oryza sativa L. var. Amaroo) were germinatedfollowingMongon et al. (2014). Dehulled seeds (i.e. caryopses)were washed with dilute sodium hypochlorite (0.1%) for 30 s,rinsed in deionized (DI) water and then imbibed in aerated0.5mM CaSO4 for 3 h. The seeds were placed on a plastic meshfloating on a 10% strength nutrient solution (for chemical com-position, see below) in darkness. After 4 d, the seedlings weretransferred to a 25% strength nutrient solution and exposedto light. Seven days after imbibition, the seedlings weretransplanted to 4L plastic pots with perforated lids (eightplants per pot) containing 100% strength nutrient solution.Plants were held individually in each of the eight holes in thelids using polyethylene foam, and the pots were covered withaluminium foil to exclude light from the root system. Elevendays after imbibition (2–3d before submergence), roots re-ceived a hypoxic pretreatment by flushing the nutrient solutionwith N2 gas for 5min. On the day before submergence, plantswere transferred to 2.2 L pots (four plants per pot) containing100% strength nutrient solution, with additional 2.5mM

NH4NO3, made stagnant with 0.1% (w/v) agar and previouslydeoxygenated by flushing overnight with N2 gas.

The composition of the nutrient solution at 100% strengthwas as follows: KNO3, 3.75mM; NH4NO3, 0.625mM (plus2.5mM NH4NO3 when stagnant agar was used); KH2PO4,0.2mM; MgSO4.2H2O, 0.40mM; Na2O3Si.9H2O, 0.10mM;CaSO4.2H2O, 1.5mM; KCl, 100μM; H3BO3, 50μM; MnSO4.H2O, 4.0μM; ZnSO4.7H2O, 4.0μM; CuSO4.5H2O, 1.0μM;Na2MoO4.2H2O, 1.0μM; NiSO4.7H2O, 2.0μM; and Fe-EDTA,50μM. The solution also contained 2.5mMMES buffer, and thepH was adjusted to 6.5 using KOH. At 7–8d after imbibition,one dose of FeSO4.7H2O was added to each 4L pot to a finalconcentration of 5.0μM to avoid any iron deficiency in theseedlings. The nutrient solution in the pots was replaced withfresh solution every 6d during the entire experiment andtopped up with DI water as required to replace water con-sumed in transpiration. Plants were kept in a naturally lit,temperature-controlled (30/25 °C day/night) phytotron duringOctober to November 2015 in Perth, Western Australia. Lightin the phytotron was 741μmol photons m�2 s�1 at midday evenon a cloudy day.

Experiment 1 – tissue ions in +GF and �GF plantsduring 16d submergence in non-saline and saline(50mM NaCl) artificial floodwaters

Plants were grown in two batches staggered with time, owing tothe limited number of cylinders in the submergence systems (de-scribed next). Submergence treatments commenced 13–15d af-ter imbibition when all plants had a visible fourth leaf collar.

The setup of the submergence system has been described previ-ously (Pedersen et al. 2009; Teakle et al. 2014). In short, the 2.2Lpots each containing four plants were randomly transferred to12L clear Perspex cylinders filled with either saline (50mM

NaCl) or non-saline (0mM NaCl) submergence solution. Thebasal submergence solution (artificial floodwater) containedthe following: CaSO4, 2.0mM; MgSO4, 0.25mM; and KHCO3,2.0mM. The root medium in all cases was non-saline.

Cylinders filled with saline or non-saline submergence solu-tion were connected to two separate, identical lines of aquar-ium pumps and ultraviolet (UV)-filters (JBL AquaCristalUV-C; JBL GmbH & Co. KG, Neuhofen, Germany). In eachsystem of nine Perspex cylinders per line, a pH controller(JBL CO2/pH Control; JBL GmbH & Co. KG, Neuhofen,Germany) connected to a cylinder with pressurized CO2

maintained free CO2 at 200μM by referring to the relevantpH set points for non-saline (pH7.3, Mackereth et al. 1979)and saline (pH7.1, Pierrot et al. 2006) water. Dark plastic cov-ered the lids of the pots and the bottom and basal sides of thecylinders, excluding light from entering basal portion of eachcylinder that contained the plastic pots with the nutrient solu-tion. Rubber-covered weights weighed down the pot in eachcylinder. Plants grown in identical pots with nutrient solutionand rubber weights were placed in empty cylinders (i.e. con-taining air) and with plastic mesh near the top of each cylin-der (see below for the reason this mesh was neededespecially for the submerged plants), serving as ‘emergent’controls with shoots in air. Light in the water-filled cylinderswas 863μmol photons m�2 s�1 at midday on a cloudy day,which was above values in air (see section ‘Plant culture’);this could be caused by filled cylinders acting as a lensthereby focusing light onto the light sensor (Walz US-SQS/L; Heinz Walz GmbH, Effeltrich, Germany).

Before submergence, plants were either untreated, thusretaining clearly visible leaf gas films upon submergence(+GF), or the entire shoot was brushed with 0.1% (v/v)Triton X-100 (Colmer & Pedersen 2008; Pedersen et al.2009; Winkel et al. 2013) preventing leaf gas film formationwhen submerged (�GF). Shoots treated with 0.1% TritonX-100 were rinsed using a separate batch of submergence so-lution prior to insertion of these plants into the cylinders.New leaves formed during the submergence period werebrushed with 0.1% Triton X-100 and rinsed with a separatebatch of submergence solution, when plants were raised outof the tanks for this process every 2d. Plastic mesh held20mm below the water surface within each cylinderprevented leaf emergence into the air above the water whenshoots elongated following submergence. The submergencetreatment lasted 9d for plants treated with 0.1% Triton X-100 (without leaf gas films) and 16d for plants initiallyretaining leaf gas films. The shorter treatment period of the0.1% Triton X-100 treated plants was due to the beginningof some disintegration of the leaves after 9 d of submergence(observed during a pilot experiment). Plants were harvestedon days 0, 1, 2, 5, 9 and 16 of the submergence treatment.The youngest fully expanded leaves at time of submergence(leaf 4) and leaf 3 were excised and used for further analysis(the entire third leaf blade was used for tissue ion analysis;

2 M. Herzog et al.


Leaf gas film thickness and tissue porosity

The leaf gas film volume and tissue gas-filled porosity weremeasured using the ‘buoyancy method’ (Raskin 1983;Thomson et al. 1990) on 50mm segments of the fourth leafaccording toWinkel et al. (2013) at room temperature. The leafsegment area was measured as described for ‘Underwater netphotosynthesis’, frozen at �20 °C, freeze-dried and DMrecorded. Mean gas film thickness was calculated by dividinggas film volume (mm3) with the two-sided area (mm2).

Tissue ion concentrations

In order to retrieve sufficient tissue for ion concentration anal-ysis, the entire third leaf and the remaining ~30–80mm of thefourth leaf were excised from submerged plants and rinsedfor 5–10 s in DI water. Leaf Na+, Cl� and K+ concentrationswere determined following Munns et al. (2010). In short,oven-dried (60 °C) leaf samples were extracted in 2.5–5mL0.5mM HNO3 for 2 d at 25 °C. Extracts were diluted withMilli-Q water as required and analysed for Na+ and K+

(Jenway PFP7 Flame Photometer, Jenway, Essex, UK) andCl� (Slamed Chloridometer CHL 50, Slamed ING GmbH,Frankfurt, Germany). The reliability of these analyses was con-firmed by taking a reference plant sample (ASPACno. 85)withknown ionic composition through the same procedures.

Chlorophyll concentration

The freeze-dried leaf segments (from underwater PN measure-ments) were each homogenized in a 2mL Eppendorf tubeusing two metal beads for 10 s on a mini bead-beater (MiniBead Beater; BioSpec Products Inc., Bartlesville, OK, USA).Chlorophyll was extracted for 24h in 96% ethanol, centrifugedat 9000 rpm for 3min and chlorophylla absorbancemeasured at656 and 750nm on a spectrophotometer (Shimadzu UV-1800;Shimadzu Corp., Kyoto, Japan). Chlorophylla concentrationswere calculated using equations of Mackinney (1941).

Scanning electron microscopy

Leaf segments were frozen immediately after sampling andthen freeze-dried. Samples were gold-coated in a sputter coaterfor 90 s and then analysed using a scanning electronmicroscope(FEI Inspect S; FEI Company, Hillsboro, OR, USA) at highvacuum mode, 12.5kV and 500–7000×magnification. Forcloser examination of wax platelets, samples were alsoanalysed with a field emission scanning electron microscope(JEOL JSM-6335F, JEOL Ltd., Peabody, MA, USA) at7.0kVand 27–45.000×magnification.

Experiment 2 – the effect of pO2 on leaf ionconcentrations

To evaluate the effect of O2 supply to submerged leaves on tis-sue ion net uptake or loss, excised leaves were subject to 24hincubation in darkness in saline water with pO2 set to five dif-ferent levels (described next). Plants were grown to the same

Salt intrusion into leaves of submerged rice 3

the fourth leaf was used for measuring underwater photosyn-thesis, leaf gas film thickness, leaf tissue porosity, scanning electron microscopy, chlorophyll concentration and tissue ion analysis). Details of measurements are given next.

Growth

Plants were harvested for dry mass (DM) measurements on days 0, 9 and 16 of submergence treatments. Plants were sepa-rated into shoot and roots and oven dried at 60 °C for 48 h be-fore weighing. As the experiment consisted of two different batches of plants, we calculated relative growth rates RGR = (lnW2 � lnW1)/(t2 � t1) for growth comparisons, where W1

and W2 are the initial and final weight (g), respectively, and t1

and t2 are the initial and final time (days), respectively.Recovery was assessed following desubmergence after 9 d of

submergence. Four pots each containing four plants were desubmerged and placed in empty Perspex cylinders. After 10 d with shoots again in air, the plants were scored for survival, dead and living shoot tissues were separated, samples were dried at 60 °C for 48 h, and DM was recorded.

Underwater net photosynthesis

Underwater net photosynthesis (PN) was measured following the approach described in Pedersen et al. (2013). Leaf segments were incubated in a defined medium (described next) for a known time in closed transparent glass vials with gentle mixing and held at a constant temperature in light [photosynthetically active radiation (PAR) given next], after which the O2 evolu-tion (PN) by the leaf segments was measured against a blank vial lacking leaf segments. Four replicate leaves (youngest fully expanded at the time of submergence from four different plants) were taken from each of the two treatments (non-saline or saline submergence). Leaf segments of 10 mm in length (projected area ~50 mm2) were excised from the top third of the lamina. Underwater PN was measured at 30 °C using 25 mL glass vials with two glass beads added to provide mixing as the vials were held on a ‘turning wheel’ during incubation with PAR inside the vials of 1000 μmol photons m�2 s�1 pro-vided from a vertically positioned light-emitting diode lamp (Valoya R300 NS1; Valoya Ltd., Helsinki, Finland) providing 94% of PAR with a colour temperature of 4800°K. Measurements were performed during the same time of day (1000–1400 h) on all days.Following incubations of known duration (90–120 min), dis-

solved O2 concentration in each vial was measured using an O2 optode (Unisense OP-MR; Unisense A/S, Aarhus, Denmark) connected to an optode meter (Unisense Micro-Optode meter). The optode was calibrated at 30.0 °C in water at air equilibrium (20.6 kPa O2) and in anoxic water (0.0 kPa O2) containing 100 mM sodium ascorbate and 100 mM NaOH. Projected area of each individual leaf segment was measured using digital photos and analysis in ImageJ (Schneider et al. 2012). Samples were then immediately frozen at �20 °C, freeze-dried and DM recorded.


age and developmental stage as in experiment 1 (15-day-old-plants). The youngest fully expanded leaf was excised andeither treated with 0.1% (v/v) Triton X-100 in non-saline sub-mergence solution (and rinsed with non-saline submergencesolution, �GF) or left untreated as controls (+GF). The sub-mergence solution composition was as described for experi-ment 1. The cut end was sealed using Vaseline. One treated(�GF) and one control (+GF) leaf were placed pairwise in four250mL conical flasks for each pO2 treatment containing salinesubmergence solution (basal submergence solution plus 50mM

NaCl) as described previously for experiment 1; each leaf wasweighed-down under the solution by a plastic-coated paperclip. Two flow controllers (Bronkhorst High-Tech B.V. serieswith B.V. E-5700 power supply; Bronkhorst High-Tech B.V.,Ruurlo, the Netherlands) connected to a pressurized N2 cylin-der and an air pump were used to adjust pO2 in the submer-gence solution to 0.01, 0.46, 1.59, 3.16 and 20.23 kPa O2.Leaves were incubated in the dark for 24h at 25 °C. After incu-bation, leaves were visually inspected for presence/absence ofleaf gas films and then rinsed and analysed for tissue ion con-centrations as described previously. Ion uptake rates werecalculated using initial tissue ion concentrations from the sameleaf type sampled from plants in experiment 1; these initialconcentrations were then subtracted from the final concentra-tions and divided by the incubation time (24h).

Data analysis

Data were analysed with GRAPHPAD PRISM version 6.07(GraphPad Software, La Jolla, CA, USA), SYSTAT version12.02 (Systat Software Inc., San Jose, CA, USA) and SPSS

version 22 (SPSS Inc., Chicago, IL, USA) for Windows statis-tical software. Normally distributed data were analysed usingtwo-way or three-way ANOVA; data requiring transformationsare specified next. Variance homogeneity was confirmed byvisual inspections of residual plots and Levene’s test forvariance homogeneity (P> 0.05). Correlations (Fig. 4 andSupporting Information Figs S4 and S5) were analysed by cal-culating non-parametric Spearman rank correlation coeffi-cients because of lack of bivariate data normality andrelationships being non-linear. Significance level of P< 0.05was used for all analyses. For ANOVA analyses, a post hocSidak or Tukey test was performed when significant effectswere found.

Leaf gas film thickness, chlorophylla and leaf porosity data(Fig. 1) were analysed using two-way ANOVA with ‘time’ and‘salt’ as fixed factors (days 1–16 of treatments). Measurementsperformed on day 0 (initials) were excluded, and leaf gas filmthickness and leaf porosity datawere log and square-root trans-formed, respectively, in order to improve variance homogene-ity. Leaf porosity variances were, however, still significantlydifferent (Levene’s test, P=0.033), but as sample sizes wereequal making ANOVA robust to unequal variances (ProphetStatguide 1997; Graphpad Software Inc. 2013) and after visualinspection of residual plots, we considered application ofANOVA on transformed data appropriate.

Tissue ion concentrations from experiment 1 (Fig. 2) wereanalysed using two-way ANOVA with ‘time’ and ‘gas film’ as

Figure 1. Gas film thickness (a), chlorophylla (b) and tissue porosity(c) of the youngest fully expanded leaf with time of submergence of ricein water (containing basal ions, see Methods) with 0mM NaCl(squares) or 50mM NaCl (circles) for plants with leaf gas films (+GF,open symbols) or treated with 0.1%Triton X-100 and without gas films(�GF, closed symbols). The +GF and �GF plants are from differentbatches; hence, these have separate initials. Roots were in non-salinenutrient solution. In (a), two-way ANOVA on log-transformed data(days 1–16) showed a significant time × salt interaction (P< 0.0001). *denotes significant difference (Sidak’s multiple comparisons test,P< 0.05). Gas film thickness on plants that had been brushed with0.1% Triton X-100 remained between 0.0 and 0.4 μm throughout theexperiment, and gas film thickness on initials and emergent controls(i.e. in both cases, leaves from shoots in air were submerged andimmediately measured) at end of treatment (day 16) did not differsignificantly (t-test, P= 0.332; data not shown). In (b), two-waytime × salt ANOVA (days 1–16) showed significant salt and time effectsin both +GF and �GF treatments (P< 0.01). * denotes significantdifference at 0 and 50mM NaCl, within each GF treatment (Sidak’smultiple comparisons test, P< 0.05). In (c), two-way time × salt ANOVA

on square root-transformed data (days 1–16) showed significant timeeffect for +GF (P= 0.0001) and significant time (P< 0.0001) and salt(P= 0.0017) effect for �GF.* denotes significant difference between 0and 50mM NaCl within GF treatments (Sidak’s multiple comparisonstest, P< 0.05). Values are means (�SE, n = 3–4).

4 M. Herzog et al.


fixed factors (days 1–9 of treatments). Initials were excluded,andNa+ and Cl� concentrations were log-transformed in orderto improve variance homogeneity. Underwater PN (Fig. 3) wasanalysed separately for 2500 and 200μM free CO2, resulting ina three-way ANOVA with ‘time’, ‘salt’ and ‘gas film’ as fixed

factors. RGR (Fig. 5) and amount of dead shoot DM (Fig. 6)were analysed using two-way ANOVA with ‘salt’ and ‘gas film’as fixed factors. Tissue ions from experiment 2 (Fig. 7 andSupporting Information Fig. S6) were analysed using two-wayANOVA with ‘gas film’ and ‘pO2’ as fixed factors. Shoot lengthand tiller number (Supporting Information Table S1) wereanalysed using one-way ANOVA and non-parametric Kruskal–Wallis test, respectively, as transformation of tiller numberwas unable to ensure variance homogeneity.

RESULTS

To investigate the effects of gas films on leaves of rice sub-merged in saline water, we compared tissue ion concentrationsand other parameters for plants retaining leaf gas films orwhere the gas films had been experimentally removed.

Figure 2. LeafNa+ (a), Cl� (b) andK+ (c) concentrationswith time ofsubmergence of rice in water (containing basal ions, seeMethods) with0mM NaCl (squares) or 50mM NaCl (circles) for plants with leaf gasfilms (+GF, open symbols) or treated with 0.1% Triton X-100 andwithout gas films (�GF, closed symbols). Samples were the entire thirdand part of the fourth leaf (see Methods). Roots were in non-salinenutrient solution. * denotes significant difference between means of+GF and�GFwithin anNaCl treatment (Sidak’s multiple comparisonstest, P< 0.05). The GF had no significant effect on leaf K+ or log-transformed Cl� concentrations during submergence in 50mM NaClusing two-way time ×GFANOVA (days 1–9, P> 0.05), but time did(P< 0.0001). Two-way ANOVA on log-transformed Na+ concentrations(days 1–9) showed significant GF (P< 0.0001) and time (P< 0.0001)effects. The +GF and �GF plants are from different batches; hence,these have separate initials. On day 16, ion concentrations in emergentcontrols in air (roots in non-saline nutrient solution) did not differsignificantly from initials (leaf tissue ion concentrations of emergentcontrols were 180, 137 and 792μmol Na+, Cl� and K+ g�1 DM,respectively, P> 0.05, Tukey’s multiple comparisons test, data notshown). Values are means (�SE, n= 3–4 except Na+ on day 5 at 0mM

NaCl (�GF) where n = 1 because of a sampling error).

Figure 3. Underwater net photosynthesis (PN, youngest fullyexpanded leaf) with time of submergence of rice in water (containingbasal ions, see Methods) with 200 (a) or 2500 (b) μM CO2 and 0(squares) or 50 (circles) mM NaCl for plants with leaf gas films (+GF,open symbols) or treated with 0.1%Triton X-100 and without gas films(�GF, closed symbols). Roots were in non-saline nutrient solution.Three-way time × salt ×GF ANOVA performed for 200 and 2500 μMCO2, respectively, showed significant salt effect only at 2500 μM CO2

(P= 0.002). There was a significant time ×GF interaction at both 200and 2500μM CO2 (P< 0.01).* denotes significant difference between 0and 50mM NaCl within the same CO2 and GF treatment (Sidak’smultiple comparisons test, P< 0.05). Values are means (�SE, n= 3–4).



Experiment 1 had non-saline and saline (50mM NaCl,~5dSm�1) artificial floodwater, a level that allowed salinityto be imposed in one step without an ‘osmotic shock’ and50mM NaCl resulted in substantial Na+ entry into the shootof rice at the early seedling stage (var. Amaroo; Kurniasihet al. 2013); this same variety was used in the present study. Ex-periment 2 investigated whether O2 status affects ion net up-take or loss by incubating excised leaves of rice in saline(50mMNaCl) artificial floodwater at a range of pO2. In the firstsection followed, we describe the retention time of gas filmsand changes in leaf tissue Na+, Cl� and K+ in plants with intactgas films (+GF) compared with plants with gas films removedby brushing with dilute Triton X-100 (�GF) immediately priorto submergence in artificial floodwater with 0 and 50mM NaCl(experiment 1).We then report on leaf chlorophylla, leaf poros-ity, underwater PN and growth of submerged plants (experi-ment 1). Finally, we describe the effects of varying pO2 on netuptake or loss of ions by submerged leaves with or withoutgas films (experiment 2).

Gas films: retention duration and influence on leaftissue ions during saline submergence

Rice leaves retained a clearly visible gas film when submerged.Initially, mean gas film thickness was 25μm (Fig. 1). Gas filmthickness declined to 18μm (0mM NaCl) and 13μm (50mM

NaCl) during the first 5 d of submergence, followed by earlierloss of gas films in saline water (after day 5 and before day 9)than in non-saline water (after day 9 and before day 16). Itshould be noted that other studies report initial rice leaf gasfilm thickness ranging from 50 to 62μm in five genotypes (Pe-dersen et al. 2009; Winkel et al. 2013; Winkel et al. 2014), thatis, twice as thick as in the present study (Fig. 1). As one of thesestudies was performed on var. Amaroo (of similar age as inpresent study), this difference should not necessarily beinterpreted as sign of genotypic variation but could result fromenvironmental conditions (e.g. temperature during measure-ments, which differed with 10 °C) or different leaf sectionsused.

Adaxial sides of leaves showed similar macro-structures,micro-structures and nano-structures considered responsiblefor leaf hydrophobicity (grooves, papillae and wax platelets,respectively) prior to and after loss of gas films (SupportingInformation Figs S1 and S2), with the exception of wax plate-lets located on papillae that showed some slight changes. Thesewax platelets appeared to bemore rounded after loss of leaf hy-drophobicity at 50mM NaCl compared with the initials(Supporting Information Fig. S2b,c). From day 9, leaves wereincreasingly covered by filaments, possibly from filamentousepiphytic algae (Supporting Information Fig. S1).

To evaluate the effect of gas film removal on tissue ions, wemeasured Na+, Cl� and K+ in the third leaf and part of thefourth leaf sampled from various plants at 4–5 time points dur-ing 9–16d of submergence. Na+ uptake was substantial fromday 1 even in leaves with a gas film, with tissue Na+ concentra-tion having increased 4.6-fold (+GF and gas films still present)on day 5 relative to initials. Leaf Na+ increased even more(36–42% greater in �GF than +GF plants on days 1, 2 and 5;Fig. 2) when plants had their gas films removed prior to sub-mergence in saline water. Consequently, two-way ANOVA

showed significant gas film (and time) effects for Na+ (see cap-tion of Fig. 2). Tissue Na+ increased to 1111μmol Na+ g�1 DMin �GF plants on day 9, but on this single time point, Na+ wasnot significantly higher than the 981μmol Na+ g�1 DM mea-sured in plants initially possessing a gas film (note that gas filmswere no longer present at this sampling time).When tissueNa+

is expressed as a concentration in tissue water (SupportingInformation Fig. S3) rather than on a DM basis, Na+ accumu-lated to 175mM in �GF plants and 172mM in +GF plants onday 5 at 50mM NaCl (not significantly different).

Surprisingly, experimental removal of gas films did not sig-nificantly affect leaf tissue Cl� or K+ concentrations of sub-merged plants at 50mM NaCl (Fig. 2). On day 5 when gasfilms were still present on the +GF plants, both +GF and�GF leaves only retained 33% of initial K+ concentrations.K+ loss resulted in minimum tissue K+ of 215μmolK+ g�1

DM (27mM in tissue water, �GF, day 9; Supporting Informa-tion Fig. S3). Cl� concentrations had increased threefold in

Figure 4. Underwater photosynthesis (PN, youngest fully expandedleaf) at 200 and 2500μM CO2 of rice leaves retaining leaf gas filmssubmerged for 1–16 d in water (containing basal ions, see Methods)with 0 (circles) or 50 (squares) mM NaCl, plotted against thecorresponding leaf gas film thickness. Roots were in non-saline nutrientsolution. r values from non-parametric Spearman rank correlationanalysis, * denoting levels of significance (levels of P> 0.05, P ≤ 0.05,P ≤ 0.01, P ≤ 0.001 or P ≤ 0.0001 are denoted by n.s., *, **, ***, ****,respectively): 200μM CO2, r= 0.7279****; 2500 μM CO2,r= 0.7476****. Values are means as presented in Figs 1 and 3.

6 M. Herzog et al.


both +GF and�GF leaves on day 5, with maximum tissue Cl�

of 719μmol Cl� g�1 DM on day 9 (+GF, 124mM in tissuewater; Supporting Information Fig. S3). Consequently, two-way ANOVA on leaf Cl� and K+ concentrations only showed

significant time effects, contrasting to the additional gas film ef-fect found for leaf Na+ concentrations (see caption of Fig. 2).

When submerged in non-saline water, leaf tissue ionsremained similar to initial levels, the only exception beingCl� in +GF plants, where at the end of the treatment the tissueions had increased almost twofold above the initial values.Hence, for plants submerged in non-salinewater,�GF resultedin significantly lower tissue Cl� in leaves. For control plantswith shoots in air, leaf ion concentrations remained at initialvalues throughout the experiment (see caption of Fig. 2).

In conclusion, leaf gas films were present for at least 5 d(saline) and 9d (non-saline) submergence and significantlydelayed Na+ uptake, but not that of Cl�, and also did notprevent substantial K+ loss during submergence in saline water.In the following sections, we describe the effects of NaCl andsubmergence on the other physiological parameters measuredfor leaves and on plant growth and recovery upondesubmergence.

Removal of leaf gas films accelerates chlorophylladegradation

Leaf chlorophylla declined several days earlier when sub-merged plants had their gas films removed (Fig. 1). For

Figure 5. Relative growth rates (RGR) during 9 d submergence (top panel) and the following 10 d recovery period (bottom panel) in roots (a, c) andshoots (b, d) of rice plants retaining leaf gas films (+GF, open bars) or treated with 0.1%Triton X-100 and without leaf gas films (�GF, closed bars) inwater (containing basal ions, see Methods) with 0 or 50mM NaCl. Roots were in non-saline nutrient solution. Letters denote significant differencesbetween columns (Tukey’s multiple comparisons test, P< 0.05). In all four datasets, two-way salt ×GFANOVA only detected significant effects of gasfilm (P< 0.01) but not salt (P ≥ 0.2598). The +GF and �GF were batches staggered with time, and the emergent control columns are therefore themean RGR of these two columns. Values are means (�SE, n= 4).

Figure 6. Dead shoot tissue as percent of total shoot dry mass of riceafter submergence in water (containing basal ions, see Methods) with 0or 50mM NaCl and a following 10 d recovery period of plants with leafgas films (+GF, open bars) or treated with 0.1% Triton X-100 andwithout gas films (�GF, closed bars). Roots were in non-saline nutrientsolution. Letters denote significant difference between means using apost hoc Tukey’s multiple comparisons test (P< 0.05). Two-waysalt ×GF ANOVA showed significant salt ×GF interaction (P= 0.0014).Values are means (�SE, n = 4).



example, after 5 d of submergence at 50mM NaCl, �GF leafchlorophylla declined to 33% of initial levels, while plantsretaining a gas film increased in chlorophyll by 28%. For plantsinitially with gas films, loss of gas films after 9 d of submergencewas followed by declines in chlorophylla levels to as low as 21%of initials after 16d of submergence. The 50mM NaCl

decreased leaf chlorophylla even further, as evident from a sig-nificant time× salt effect in both gas film treatments. However,when comparing single time points, the effect of salt was notsubstantial and with one exception (day 2, +GF), remained sta-tistically insignificant in post hoc tests.

In spite of significant negative (Na+, Cl�) and positive (K+)correlations found when plotting chlorophylla concentrationsagainst tissue ions concentrations (Supporting InformationFig. S4), a causal relationship between these two factors is notnecessarily present, because of the different time points atwhich leaves were sampled. For example, leaf chlorophylladeclined with time of submergence regardless of salinity treat-ment (Fig. 1); this decline with time could then have resulted inlower chlorophylla at later time points (where leaf Na+ and Cl�

were at their highest in the saline treatment).We therefore sug-gest that chlorophylla degradation is mainly caused by theeffects of submergence per se, as also prevailing during non-saline submergence (see Discussion).

Leaf porosity is affected by salinity and gas filmremoval

Gas film removal caused leaf porosity to decline substantiallywithin the first 5 d of submergence (mean tissue porosity of2.0% after 5 d in non-saline water; Fig. 1), compared withleaves with intact gas films under the same conditions (meantissue porosity of 10.4%, same as initials). The adverse effectsof gas film removal were stronger for plants in saline than innon-saline solution (0.0 and 11.7% porosity in �GF and +GFplants on day 5, respectively); therefore, two-way ANOVA

(time× salt) showed a significant salt effect for �GF plantsonly. Thus, +GF plants maintained higher leaf porosity duringsubmergence until the gas films were lost naturally.

Gas films enhance underwater gas exchange

Gas films enhanced underwater PN at 200μM CO2 (Fig. 3). Atthis CO2 concentration, PN rates of �GF leaves were 15% of+GF leaves (when first submerged in non-saline water). Thepositive effect of leaf gas films on underwater PN was main-tained for 5 d and then declined by day 9 as gas films dimin-ished, with PN of +GF plants reduced to 16% (non-saline)and 7% (saline) of initials on day 16. Results of three-wayANOVA (time× salt ×GF) reflected this decrease over time withsignificant time×gas film interactions at both 200 and 2500μMCO2. Elevating CO2 to 2500μM closed the gap between +GFand �GF leaves, as now, gas film removal only reduced PN to72% of leaves with intact gas films. Increasing external CO2

partially alleviated the negative effect of not possessing gasfilms on CO2 uptake, that is, increasing external CO2 can over-come the higher resistance of CO2 uptake in leaves with no gasfilms. Plotting PN against gas film thickness (Fig. 4) revealedsignificant positive correlations both at low (r=0.73) and high(r=0.75) CO2, underlining the positive effect of gas films onunderwater PN.

The NaCl treatment only affected PN significantly at highCO2 and when gas films were removed. Under these

Figure 7. Leaf Na+ (a), Cl� (b) and K+ (c) net uptake or loss of ayoungest fully expanded leaf of rice submerged in 50mM NaCl (thesolution also contained basal ions, see Methods) for 24 h in the darkwith gas film (+GF, open symbols) or treated with 0.1% Triton X-100and without gas films (�GF, closed symbols). During incubation, thesubmergence solution was maintained at 0.01, 0.46, 1.59, 3.16 and20.23 kPa O2. * denotes significant difference between mean leaf ionuptake rates (Sidak’s multiple comparisons test, P< 0.05). Two-wayGF× pO2 ANOVA showed a significant effect of GF on leaf Na+

(P= 0.0345) andK+ (P= 0.0091) uptake, although the overall effect wasrelatively small and on most time points not significant. pO2 had asignificant effect on K+ loss (P< 0.0001), as ion loss increased severelywhen pO2 was below 1.6 kPa. For Cl�, two-way ANOVA showed asignificant GF× pO2 interaction (P= 0.0050). Values are means (�SE,n= 4). Tissue ion concentrations after 24 h incubation are displayed inSupporting Information Fig. S6.

8 M. Herzog et al.


conditions, PN rates were 25% lower on average when subjectto NaCl (significant on days 1, 5 and 9). Thus, three-wayANOVA (time× salt ×GF) performed at both high and lowCO2 only showed a significant salt effect at high CO2. WhenplottingPN against tissue ion concentrations (Supporting Infor-mation Fig. S5), and excluding from the analysis late timepoints where leaves were severely damaged by submergence(characterized by low leaf chlorophyll and low leaf porosityvalues, see caption in Supporting Information Fig. S5), only tis-sue Na+ and PN at 2500μM CO2 showed a significant negativecorrelation (r=�0.61). As no significant NaCl effect was de-tected for PN at 200μM CO2 (P≥ 0.2094), the adverse impacton growth of submergence alone was significant whereas the50mM NaCl treatment during submergence had little addi-tional effect on growth (see next section on growth analysis)in spite of PN at high CO2 revealing some damage to thephotosynthetic apparatus.

Gas film removal significantly reduces rice growthwhen submerged

Rice shoots elongated during 9d of submergence and were52% longer than controls in air (average of NaCl and GFtreatments, Table S1). Meanwhile, submergence severelyinhibited tillering, as mean tiller numbers after 9 d of submer-gence across NaCl and GF treatments were only 1–1.75 insubmerged plants compared with 4–4.5 in controls with shootsin air (Table S1).Complete submergence over 9 d reduced root RGR more

than the reduction in shoot RGR, and gas film removal causedfurther reductions to growth of plants when submerged ineither non-saline or saline water (Fig. 5). During 9d of submer-gence in non-saline water, +GF plants maintained root andshoot RGR at 27 and 61% of controls in air, respectively, whileremoval of gas film resulted in root and shoot RGR of 10 and45% of controls in air. Plants that had gas films removed alsoshowed reduced RGR during the recovery period afterdesubmergence (root and shoot RGR to 60 and 55% of con-trols in air). By contrast, plants with intact gas films had rootand shoot RGR of 98 and 84% when desubmerged relativeto controls in air.NaCl had a tendency to further decrease RGR in �GF

plants (e.g. root RGR during submergence in saline water;Fig. 5a), but two-way ANOVA (salt ×GF) only detected signifi-cant gas film and no salinity effects during submergence andrecovery. Nonetheless, a significant salt ×GF interaction wasfound when analysing the amount of shoot tissue (% of totalshoot DM) that had senesced and was scored as dead afterthe recovery period (Fig. 6), that is, plants subjected to bothdilute Triton X-100 brushing and 50mM NaCl had lower func-tioning leaf area for further growth than plants only subjectedto one of these treatments.In conclusion, gas film removal had a profound effect on rice

RGR during complete submergence in non-saline water andthe following recovery period. On the other hand, 50mMNaClin the submergence solution only had limited additional effecton growth.

Experiment 2 – the effect of pO2 on leaf ionconcentrations

To separate the effects of gas films acting as a possible physicalbarrier to ion uptake/loss, and gas films resulting in higher leafO2 status that could affect energy status and thus energy-dependent ion transport (potentially impacting K+ retentionand Na+ and Cl� ‘exclusion’ from leaves), excised leaves withand without gas films were incubated in submergence solutionwith 50mM NaCl and pO2 ranging from 0.01 to 20.3 kPa(Fig. 7). Low pO2 did not seem to diminish gas films as clearlyvisible gas films were observed on all +GF leaves following the24h incubation.

Gas film removal resulted in 12%higher final tissueNa+ con-centration across the pO2 range tested (Supporting Informa-tion Fig. S6); however, the difference was only significant at0.46 kPa O2. The little difference between +GF and �GF leafNa+ concentration was due to a high Na+ influx in both cases:average net Na+ uptake across the pO2 range tested was7.0μmol Na+ g�1 DM h�1 in +GF and 8.7μmol Na+ g�1 DMh�1 in �GF leaves (Fig. 7), so the leaf gas films acted only asa weak physical barrier as Na+ entry into +GF leaves wassubstantial.

Gas films on leaves acting as a rather weak barrier to tissueion fluxes were confirmed by changes in tissue K+. LoweringpO2 to 0.01 kPa resulted in a severe loss of tissue K+ regardlessof gas film presence (Fig. 7), with tissue K+ decreasing to 21%(�GF) and 25% (+GF) of leaves incubated at 20.3kPa O2.Nonetheless, gas films did have a significant overall effect ontissue K+ according to two-way ANOVA (see caption of Fig. 7).

Interestingly, leaf Cl� showed a significant GF×pO2 interac-tion, as removal of gas films decreased tissue Cl� by 8–16%relative to leaves retaining a gas film at 1.59–20.23 kPa O2,and increased tissue Cl� by 20% at 0.01 and 0.46 kPa(Supporting Information Fig. S6). Although these differencesin Cl� concentrations at single pO2 levels were not significantin post hoc tests, it should be noted that in experiment 1, pres-ence of gas films also tended to result in higher tissue Cl� ondays 5 and 9 in both non-saline (significant on day 9) and saline(not significant) water compared with leaves without gas films.These coinciding observations from two separate experimentsseem to indicate a complex interaction between leaf gas filmsand Cl� uptake and resulting tissue concentrations.

In conclusion, this second experiment with varying pO2 con-firmed that gas films on leaves of rice apparently only act as aweak physical barrier to ion uptake/loss; a leaf-water interface,allowing for substantial K+ net loss and Na+ net uptake musthave been present during the 24h of submergence, despitethe gas films being visibly present.

DISCUSSION

Gas film presence had the expected beneficial effects on plantgrowth,PN, leaf chlorophylla, leaf porosity and shoot tissue sur-vival for rice during submergence in saline (50mM NaCl) wa-ter. However, the results did not support our initialhypothesis of rice leaf gas films acting as a strong physicalbarrier to ion uptake (Na+, Cl�) or loss (K+) during



submergence in saline water. Although gas film removal signif-icantly increased Na+ uptake by plants submerged in artificialfloodwater containing 50mM NaCl, Na+ and Cl� accumulationand K+ loss were substantial even in leaves possessing gas films(Fig. 2). Gas films acting as a rather weak physical barrier toNa+ entry were confirmed in a separate experiment (Fig. 7and Supporting Information Fig. S6), where removal of gasfilms resulted in a 12% increase in leaf Na+ concentration.

Gas film retention time in the present experiment was up to6d longer than in the field (Winkel et al. 2014), indicating thatturbid floodwaters may accelerate gas film loss. Interestingly,the gas films were retained longer by plants in the non-salinetreatment as compared with those in the saline submergence(Fig. 1). Loss of gas films with time of submergence was not as-sociated with clear structural changes of the surface of the leafcuticle, except for wax platelets on papillae appearing morerounded (Supporting Information Figs S1 and S2). This lossof gas films without significant changes in cuticle surface struc-ture was unexpected as leaf hydrophobicity is known to berelated to the amount of wax platelets on the cuticle surface(see Neinhuis and Barthlott, 1997 for characterization of leafhydrophobicity and the relationship with cuticle nano-structure). The recent description of the gene ‘OsHSD1’responsible for synthesis of rice epicuticular wax compounds(Zhang et al. 2016) adds new perspectives to further exploregas film retention during submergence.

Uptake of Na+ and Cl� and loss of K+ indicatesleaf-to-water contact even with gas films present

Leaves of rice with gas films when submerged in 50mM NaClfor 9 d had tissue ion concentrations (μmol g�1 fresh mass) ofNa+ 145, Cl� 106 and K+ 38 (data not shown), respectively,which compare with maximum concentrations of Na+ 94 andCl� 141 and minimumK+ 52 in the coleoptiles of rice seedlingssubmerged for 42–186h at 50mM NaCl (Kurniasih et al. 2013).The coleoptiles emerged from seeds under water and lackedgas films (Kurniasih et al. 2013). The substantial entries ofNa+ and Cl� for rice leaves with gas films when submerged insaline water at 72 and 93%of those for leaves without gas films(Fig. 2) contrast with the reduced ion entry into leaves of M.siculus with gas films of only 51 and 44% of the amounts with-out gas films (during the first 24 h of complete submergence,Teakle et al. 2014). These species differences in ion entry couldbe due to contrasting leaf morphology resulting in distinctthree-dimensional (3D) structures of the gas films. The 3Dtomograms of submerged Spartina anglica leaves with gas filmsindicated that along the leaf ridges, approximately 20% of theridge surface is in direct contact with water (Lauridsen et al.2014). Rice also possesses plicate leaves (Wu et al. 2011) wherethe majority of the external gas volume is present in the deepgrooves between each ridge running parallel along leaves,and presumably areas along the ridges of leaves of rice mustalso have some direct contact with the floodwater. With timeof submergence, and with possible declines in surface hydro-phobicity, these exposed patches are likely to grow in sizeresulting in increasingly larger interfaces (i.e. areas of direct

contact) between floodwater and the leaf surface. In contrastto rice and S. anglica,M. siculus does not possess plicate leaves(see photo of submerged leaf in Teakle et al. 2014), likelyresulting in much less variation in 3D structure (and thickness)of the gas film across the leaf surface. Consequently, we suggestthat the differences in ion entry observed between rice (thisstudy) and M. siculus (Teakle et al. 2014) are due to 3D struc-tural differences in the gas layer forming the interface betweencuticle and floodwater, with likely more direct leaf-to-watercontact in submerged rice than inM. siculus.

Presence of significant direct leaf-to-water contact for riceleaves with gas films was supported by substantial loss of K+

from leaves submerged in water at low pO2 (Fig. 7). K+ loss

during severe hypoxia or anoxia is, in the short-term, causedby depolarization of plasma membranes (Buwalda et al.1988), leading to opening of voltage-gated ion channels (Wardet al. 2009), and during longer periods can result from deterio-ration or damage to membranes, as described for wheat roots(Buwalda et al. 1988; Greenway et al. 1992; Goggin & Colmer2007). Meanwhile, the leaf K+ loss observed when submergedin saline solution at air-equilibrium pO2 (Figs 2 and 7) is mostlikely caused by high external Na+ known to induce K+ efflux(Shabala et al. 2006; Britto et al. 2010).

While removal of leaf gas films increased Na+ uptake byleaves of rice submerged at 50mMNaCl, Cl� uptake was muchmore similar in �GF and +GF leaves (Fig. 2). In addition, onsome occasions (submerged in non-saline water in experiment1 and at high pO2 in experiment 2), tissue Cl� concentrationwas higher in +GF than in�GF leaves.We suggest that this dif-ference in tissue Na+ and Cl� concentrations could be causedby these ions entering the leaf in different ways: Na+ is likelyto enter leaves down an electrochemical gradient, while Cl�

has to be actively taken up because of its negative charge. Ricecoleoptiles submerged in 50mM NaCl showed peak uptake ofCl� during the initial 42–114h of submergence (Kurniasihet al. 2013), and such Cl� uptake can be a more rapid and lessenergy-demanding means to maintain cell turgor or volumethan production of organic solutes (Raven 1985; Oren 1999).Energy available for Cl� influxes via H+�Cl� symports andassociated H+ATPase activity required to maintain the H+

gradient across the plasma membrane (Teakle & Tyerman2010) is likely to be higher in +GF leaves because of highersugar levels andO2 uptake compared with�GF leaves (Peder-sen et al. 2009; Winkel et al. 2013). The significant pO2×GFinteraction on leaf Cl� concentrations in experiment 2 furthersupports that Cl� uptake is altered by leaf energy status.Indeed, ion net fluxes even in the anoxia-tolerant coleoptileof rice seedlings are substantially reduced in anoxia ascompared with aerated conditions for seedlings submergedin 50mM NaCl (Kurniasih et al. 2016). Tracer experimentsare needed to separate the roles of ion influx or efflux onchanges in net uptake rates and leaf ion concentrations ofsubmerged rice.

Rice subjected to 50mM NaCl in the root medium but withshoots in air in several earlier experiments accumulated higherleaf tissue Na+ and Cl� concentrations (Yeo & Flowers 1982;1984; 1985; 1986) than those in the shoot tissues of rice duringcomplete submergence in 50mM NaCl (present study and

10 M. Herzog et al.


Kurniasih et al. 2013), possibly because of continuous iontransport to leaves via the transpiration stream of plants withshoots in air. Rice with roots in 50mM NaCl and with shootsin air contained leaf Na+ at 1996–2280 and Cl� at 1770–2122μmol g�1 DM after 7–10d of treatment (Yeo & Flowers1984; 1985), compared with the leaf Na+ at 980 and Cl� at719μmol g�1 DM in +GF plants after 9 d submergence inthis present experiment. In some experiments with roots ofrice in saline solutions and with shoots in air, leaf (Yeoet al. 1999) and shoot (Flowers & Yeo 1981; Yeo & Flowers1984; 1986; Yeo et al. 1999; Kavitha et al. 2012) Na+ and Cl�

concentrations were similar to concentrations in the presentstudy. Differences in shoot Na+ concentration between riceplants in different experiments can result from variations inNa+ ‘exclusion’ ability amongst genotypes (Yeo & Flowers1986; Yeo et al. 1999), relative humidity (Yeo & Flowers1984), leaf age (Yeo et al. 1985) and treatment duration(Flowers & Yeo 1981).

Gas films enhance underwater PN and delay leaftissue degradation of submerged rice

Leaf gas films have beneficial effects on underwater PN of riceboth in non-saline (Pedersen et al. 2009;Winkel et al. 2014) andsaline (present study, Fig. 3) submergence. 3D diffusionmodel-ling has demonstrated the enhanced leaf-floodwater gas ex-change by leaves with gas films if stomata remain at leastpartially open during submergence (Verboven et al. 2014).The significant adverse effect of gas film removal on growth(Fig. 5) is in accordance with a previous study of submergedrice (Pedersen et al. 2009).In addition to lowering PN, removal of leaf gas films leads to

earlier leaf chlorophylla degradation of submerged leaves(Fig. 1). This contrasts with a previous study where no differ-ence in total leaf chlorophyll between +GF and –GF plantswas found during 7d submergence in the field (Winkel et al.2013). We suggest that the lack of decline in chlorophylla in�GF plants could be due to Winkel et al. (2013) sampling theyoungest fully developed leaf at all time points and not beyond7d. In a subsequent experiment, Winkel et al. (2014) observedno chlorophyll decline until day 7 (second youngest fully ex-panded leaf, submerged for 13d in the field, +GF only) consis-tent with declines after 9 d in the present study. Yeo andFlowers (1983) established the leaf Na+ concentrationsassociated with a 50% loss of chlorophyll (LC50) for nine ricegenotypes with shoots in air. However, for plants in the presentstudy, chlorophylla degradation was mainly caused by durationof submergence (explaining 76% of the variation inchlorophylla according to ANOVA) rather than leaf Na+ concen-tration (5%of the variation), and chlorophylla degradation wassevere even in leaves of plants submerged in non-saline water,so we refrained from calculating a LC50 in the present study.Leaf senescence is a common feature of submerged rice andhas been associated with the accumulation of ethylene causingchlorophyll degradation (Jackson et al. 1987; Ella et al. 2003).Leaf hydrophobicity has previously been suggested as an ad-

aptation to prevent adverse effects of salt spray on leaves of

some coastal plants (Ahmad & Wainwright 1976; McNeillyet al. 1987). Variation in leaf wettability and leaf Na+ retention(upon spraying with or immersion into water containing500mM NaCl) was linked to distributions of three Agrostisstolinefera ecotypes growing in sheltered inland habitats, sea-water spray-zone or salt marshes (Ahmad & Wainwright1976). The inland ecotype showed high wettability because oflower contact angles and shorter epicuticular waxes than theecotypes in the salt-spray and salt marsh zones, resulting in 16times higher Na+ retention on the surface of leaves after 5 s im-mersion into saline water. Ahmad and Wainwright (1976) sug-gested that differences in adaxial and abaxial sides for leafhydrophobicity in spray-zone plants but not in salt marsh plantscould be an adaption to episodic inundations of the marshplants, as inundation would affect both sides of the leaf in con-trast to salt-spray. However, another low salt marsh plant fromthe intertidal zone (S. anglica) being hydrophobic only on theadaxial leaf side (Winkel et al. 2011) is not in support of two-sided leaf hydrophobicity as a general adaptation to salt watersubmergence, but like for rice with two-sided leaf gas films(Winkel et al. 2013), the one-sided leaf gas films on submergedS. anglica also benefit internal O2 status both during the dayand at night.

CONCLUSIONS

Leaf gas films contribute to rice submergence tolerance by im-proving underwater gas exchange, growth, internal aerationand plant sugar levels (Pedersen et al. 2009; Winkel et al.2013). This study found that during submergence in saline wa-ter (50mMNaCl), gas films diminished earlier than for leaves infreshwater and that rice plants possessing leaf gas films main-tained higher levels of underwater PN, more growth duringsubmergence and recovery, greater proportion of survivingshoot biomass and better maintained leaf porosity andchlorophylla. Submergence was more detrimental to leaf PN

than the additional effect of 50mM NaCl. However, gas filmson leaves of rice delayed Na+ entry to a much smaller degreecompared with leaves of M. siculus, which was likely due to3D structural differences in the gas layers on these two specieswith probable greater leaf-to-water contact for rice. Rice hasplicate leaves, like S. anglica, for which the ridges have some di-rect contact with surrounding water even when the gas film ispresent (Lauridsen et al. 2014). Varying pO2 had no effect onleaf Na+ net uptake, suggesting that the observed delay ofNa+ uptake in +GF leaves should be attributed to gas filmsacting as a physical barrier rather than from the possibleinfluence of altered O2 supply and potential improved leafenergy status.

ACKNOWLEDGMENTS

T.D.C. acknowledges support from the Australian ResearchCouncil (DP120101482). We thank the UWA Institute ofAdvanced Studies for hosting O.P. as Professor-at-Large. M.H., D.K. and A.W. were supported by PhD and postdoctoralfellowships from the Villum Foundation.



REFERENCES

Ahmad I. &Wainwright S.J. (1976) Ecotype differences in leaf surface propertiesof Agrostis stolonifera from salt marsh, spray zone and inland habitats. NewPhytologist 76, 361–366.

Armstrong W. (1979) Aeration in higher plants. In Advances in BotanicalResearch (ed Woolhouse H.W.), pp. 225–332. Academic Press, London.

Bailey-Serres J. & Voesenek L.A.C.J. (2008) Flooding stress: acclimations andgenetic diversity. Annual Review of Plant Biology 59, 313–339.

Britto D.T., Ebrahimi-Ardebili S., Hamam A.M., Coskun D. & Kronzucker H.J.(2010)

42K analysis of sodium-induced potassium efflux in barley: mechanism

and relevance to salt tolerance. New Phytologist 186, 373–384.Buwalda F., Thomson C.J., Steigner W., Barrett-Lennard E.G., Gibbs J. &Greenway H. (1988) Hypoxia induces membrane depolarization and potas-sium loss from wheat roots but does not increase their permeability to sorbitol.Journal of Experimental Botany 39, 1169–1183.

Colmer T.D., ArmstrongW.,GreenwayH., Ismail A.M., KirkG.J.D. &Atwell B.J. (2014) Physiological mechanisms of flooding tolerance in rice: transient com-plete submergence and prolonged standing water. In Progress in Botany: Vol.75 (eds Lüttge U., Beyschlag W. & Cushman J.), pp. 255–307. Springer, BerlinHeidelberg.

Colmer T.D.& PedersenO. (2008)Underwater photosynthesis and respiration inleaves of submerged wetland plants: gas films improve CO2 and O2 exchange.New Phytologist 177, 918–926.

DasK.K., PandaD., SarkarR.K., Reddy J.N.& Ismail A.M. (2009) Submergencetolerance in relation to variable floodwater conditions in rice. Environmentaland Experimental Botany 66, 425–434.

De Ocampo M., Zantua R.E., Egdane J.A. & Ismail A.M. (2013) Pyramiding ofsalinity and submergence tolerance into IR64 background through marker-assisted selection (MAS). In 7th International Rice Genetics Symposium. Inter-national Rice Research Institute, Manila, Philippines.

Ella E.S., Kawano N., Yamauchi Y., Tanaka K. & Ismail A.M. (2003) Blockingethylene perception enhances flooding tolerance in rice seedlings. FunctionalPlant Biology 30, 813–819.

Flowers T.J. & Yeo A.R. (1981) Variability in the resistance of sodium chloridesalinity within rice (Oryza sativa L.) varieties. New Phytologist 88, 363–373.

Goggin D.E. & Colmer T.D. (2007) Wheat genotypes show contrasting abilitiesto recover from anoxia in spite of similar anoxic carbohydrate metabolism.Journal of Plant Physiology 164, 1605–1611.

Graphpad Software Inc. 2013. What to do when data fail tests for homogeneity ofvariance [Online]. Available: http://www.graphpad.com/support/faqid/1007/[Accessed 14.07.2016].

Greenway H., Waters I. & Newsome J. (1992) Effects of anoxia on uptake and lossof solutes in roots of wheat. Australian Journal of Plant Physiology 19, 233–247.

Grieve C.M., Grattan S.R. & Maas E.V. (2012) Plant salt tolerance. Agriculturalsalinity assessment and management. 2nd edition. ASCE Manual and Reportson Engineering Practice 71, 405–459.

IRRI. 2016. Climate-smart rice [Online]. Manila, Phillipines: International RiceResearch Institute. Available: http://irri.org/resources/publications/brochures/climate-smart-rice [Accessed 23.7.2016].

Ismail A.M., Singh U.S., Singh S., Dar M.H. &Mackill D.J. (2013) The contribu-tion of submergence-tolerant (Sub1) rice varieties to food security in flood-prone rainfed lowland areas in Asia. Field Crops Research 152, 83–93.

Jackson M.B. 2004. The impact of flooding stress on plants and crops [Online].Available: http://www.plantstress.com/Articles/waterlogging_i/waterlog_i.htm[Accessed 14.07.2016].

Jackson M.B., Waters I., Setter T.L. & Greenway H. (1987) Injury to rice plantscaused by complete submergence; a contribution by ethylerie (ethene). Journalof Experimental Botany 38, 1826–1838.

Kavitha P.G., Miller A.J., MathewM.K. & Maathuis F.J.M. (2012) Rice cultivarswith differing salt tolerance contain similar cation channels in their root cells.Journal of Experimental Botany 63, 3289–3296.

KirkG.J.D., GreenwayH., Atwell B.J., IsmailA.M.&ColmerT.D. (2014)Adap-tation of rice to flooded soils. In Progress in Botany: Vol. 75 (eds Lüttge U.,Beyschlag W. & Cushman J.), pp. 215–253. Springer, Berlin Heidelberg.

Kurniasih B., Greenway H. & Colmer T.D. (2013) Tolerance of submerged ger-minating rice to 50–200mM NaCl in aerated solution. Physiologia Plantarum149, 222–233.

Kurniasih B., Greenway H. & Colmer T.D. (2016) Energetics of acclimation to NaClby submerged, anoxic rice seedlings.Annals of Botany, doi: 10.1093/aob/mcw189.

Lauridsen T., Glavina K., Colmer T.D., Winkel A., Irvine S., Lefmann K. &Pedersen O. (2014) Visualisation by high resolution synchrotron X-ray phasecontrast micro-tomography of gas films on submerged superhydrophobicleaves. Journal of Structural Biology 188, 61–70.

Mackereth F.J.H., Heron J. & Talling J.F. (1979) Water analysis: some revisedmethods for limnologists. Freshwater Biological Association, Ambleside, Cum-bria, UK.

Mackinney G. (1941) Absorption of light by chlorophyll solutions. Journal ofBiological Chemistry 140, 315–322.

McNeilly T., Ashraf M. & Veltkamp C. (1987) Leaf micromorphology of sea cliffand inland plants of Agrostis stolonifera L., Dactylis glomerata L. and Holcuslanatus L. New Phytologist 106, 261–269.

MommerL. &Visser E.J.W. (2005)Underwater photosynthesis in flooded terres-trial plants: a matter of leaf plasticity. Annals of Botany 96, 581–589.

Mongon J., Konnerup D., Colmer T.D. & Rerkasem B. (2014) Responses of riceto Fe2+ in aerated and stagnant conditions: growth, root porosity and radialoxygen loss barrier. Functional Plant Biology 41, 922–929.

MunnsR.&TesterM. (2008)Mechanisms of salinity tolerance.Annual Review ofPlant Biology 59, 651–681.

Munns R., Wallace P.A., Teakle N.L. & Colmer T.D. (2010) Measuring solubleion concentrations (Na+, K+, Cl�) in salt-treated plants. In Plant Stress Toler-ance (ed Sunkar R.), pp. 371–382. Humana Press, Totowa.

Negrão S., Courtois B., Ahmadi N., Abreu I., Saibo N. & Oliveira M.M. (2011)Recent updates on salinity stress in rice: from physiological to molecular re-sponses. Critical Reviews in Plant Sciences 30, 329–377.

Neinhuis C. & Barthlott W. (1997) Characterization and distribution of water-repellent, self-cleaning plant surfaces. Annals of Botany 79, 667–677.

OrenA. (1999) Bioenergetic aspects of halophilism.Microbiology andMolecularBiology Reviews 63, 334–348.

Pedersen O., Colmer T.D. & Sand-Jensen K. (2013) Underwater photosynthesisof submerged plants – recent advances and methods. Frontiers in Plant Science4, 140.

Pedersen O., Rich S.M. & Colmer T.D. (2009) Surviving floods: leaf gas films im-prove O(2) and CO(2) exchange, root aeration, and growth of completely sub-merged rice. The Plant Journal 58, 147–156.

Pierrot D., Lewis E. &Wallace D.W.R. (2006) MS Excel program developed forCO2 system calculations. In ORNL/CDIAC-105a. Carbon Dioxide Informa-tion Analysis Center, Oak Ridge National Laboratory, U.S. Department ofEnergy, Oak Ridge, Tennessee.

Prophet Statguide. 1997.Do your data violate one-way ANOVA assumptions? [On-line]. Northwestern University, Chicago, IL. Available: http://www.basic.north-western.edu/statguidefiles/oneway_anova_ass_viol.html#Unequal%20population%20variances [Accessed 14.7. 2016].

Raskin I. (1983) A method for measuring leaf volume, density, thickness, and in-ternal gas volume. Hortscience: A Publication of the American Society forHorticultural Science 8, 698–699.

Raven J.A. (1985) Tansley review no. 2. Regulation of pH and generation of os-molarity in vascular plants: a cost-benefit analysis in relation to efficiency of useof energy, nitrogen and water. New Phytologist 101, 25–77.

Schneider C.A., Rasband W.S. & Eliceiri K.W. (2012) NIH Image to ImageJ:25 years of image analysis. Nature Methods 9, 671–675.

Setter T.L., Bhekasut P. & Greenway H. (2010) Desiccation of leaves after de-submergence is one cause for intolerance to complete submergence of the ricecultivar IR 42. Functional Plant Biology 37, 1096–1104.

Setter T.L., Ellis M., Laureles E.V., Ella E.S., Senadhira D., Mishra S.B. & DattaS. (1997) Physiology and genetics of submergence tolerance in rice. Annals ofBotany 79, 67–77.

Shabala S., Demidchik V., Shabala L., Cuin T.A., Smith S.J., Miller A.J. &Newman I.A. (2006) Extracellular Ca

2+ameliorates NaCl-induced K

+loss

from Arabidopsis root and leaf cells by controlling plasma membrane K+-permeable channels. Plant Physiology 141, 1653–1665.

Sinha D.D., Singh A.N. & Singh U.S. (2014) Site suitability analysis for dissemi-nation of salt-tolerant rice varieties in southern Bangladesh. The InternationalArchives of Photogrammetry, Remote Sensing and Spatial Information Sciences40, 961.

Teakle N.L., Colmer T.D. & Pedersen O. (2014) Leaf gas films delay salt entryand enhance underwater photosynthesis and internal aeration of Melilotussiculus submerged in saline water. Plant, Cell & Environment 37, 2339–2349.

Teakle N.L. & Tyerman S.D. (2010) Mechanisms of Cl(-) transport contributingto salt tolerance. Plant, Cell & Environment 33, 566–589.

Thomson C.J., ArmstrongW.,Waters I. &Greenway H. (1990)Aerenchyma for-mation and associated oxygen movement in seminal and nodal roots of wheat.Plant, Cell & Environment 13, 395–403.

Verboven P., Pedersen O., Ho Q.T., Nicolai B.M. & Colmer T.D. (2014) Themechanism of improved aeration due to gas films on leaves of submerged rice.Plant, Cell & Environment 37, 2433–2452.

Voesenek L.A.C.J., Colmer T.D., Pierik R., Millenaar F.F. & Peeters A.J. (2006)How plants cope with complete submergence. New Phytologist 170, 213–226.

12 M. Herzog et al.


http://www.graphpad.com/support/faqid/1007/

http://irri.org/resources/publications/brochures/climate-smart-rice

http://irri.org/resources/publications/brochures/climate-smart-rice

http://www.plantstress.com/Articles/waterlogging_i/waterlog_i.htm

http://doi.org/10.1093/aob/mcw189




Ward J.M., Mäser P. & Schroeder J.I. (2009) Plant ion channels: gene families,physiology, and functional genomics analyses. Annual Review of Physiology71, 59–82.

WassmannR.,HienN.X.,HoanhC.T.&TuongT.P. (2004) Sea level rise affectingthe Vietnamese Mekong Delta: water elevation in the flood season and impli-cations for rice production. Climatic Change 66, 89–107.

Winkel A., Colmer T.D., Ismail A.M. & Pedersen O. (2013) Internal aeration ofpaddy field rice (Oryza sativa) during complete submergence – importance oflight and floodwater O2. New Phytologist 197, 1193–1203.

Winkel A., Colmer T.D. & Pedersen O. (2011) Leaf gas films of Spartina anglicaenhance rhizome and root oxygen during tidal submergence. Plant, Cell & En-vironment 34, 2083–2092.

Winkel A., Pedersen O., Ella E., Ismail A.M. & Colmer T.D. (2014) Gas film re-tention and underwater photosynthesis during field submergence of four con-trasting rice genotypes. Journal of Experimental Botany 65, 3225–3233.

Wu D., Wang J.N., Wu S.Z., Chen Q.D., Zhao S., Zhang H. & Jiang L. (2011)Three-level biomimetic rice-leaf surfaces with controllable anisotropic sliding.Advanced Functional Materials 21, 2927–2932.

YeoA.R., Caporn S.J.M. & Flowers T.J. (1985) The effect of salinity upon photo-synthesis in rice (Oryza sativaL.): gas exchange by individual leaves in relationto their salt content. Journal of Experimental Botany 36, 1240–1248.

Yeo A.R., Flowers S.A., Rao G., Welfare K., Senanayake N. & Flowers T.J.(1999) Silicon reduces sodium uptake in rice (Oryza sativa L.) in saline condi-tions and this is accounted for by a reduction in the transpirational bypass flow.Plant, Cell & Environment 22, 559–565.

Yeo A.R. & Flowers T.J. (1982) Accumulation and localisation of sodium ionswithin the shoots of rice (Oryza sativa) varieties differing in salinity resistance.Physiologia Plantarum 56, 343–348.

YeoA.R. & Flowers T.J. (1983) Varietal differences in the toxicity of sodium ionsin rice leaves. Physiologia Plantarum 59, 189–195.

YeoA.R. & Flowers T.J. (1984) Nonosmotic effects of polyethylene glycols uponsodium transport and sodium-potassium selectivity by rice roots.Plant Physiol-ogy 75, 298–303.

YeoA.R. & Flowers T.J. (1985) The absence of an effect of the Na+/Ca

2+ratio on

sodium chloride uptake by rice (Oryza sativa L.). New Phytologist 99, 81–90.Yeo A.R. & Flowers T.J. (1986) Salinity resistance in rice (Oryza sativa L.) and apyramiding approach to breeding varieties for saline soils. Functional Plant Bi-ology 13, 161–173.

Zhang Z., Cheng Z.-j., Gan L., Zhang H., Wu F.-q., Lin Q.-b. &Wan J.-m. (2016)OsHSD1, a hydroxysteroid dehydrogenase, is involved in cuticle formation andlipid homeostasis in rice. Plant Science 249, 35–45.

Received 15 August 2016; received in revised form 16 November 2016;accepted for publication 20 November 2016

SUPPORTING INFORMATION

Additional Supporting Information may be found in the onlineversion of this article at the publisher’s web-site:

Figure S1. Scanning electron microscopy micrographs of leafsurfaces of rice subject to 0–16d submergence in non-saline(0mM NaCl) or saline (50mM NaCl) water (containing basalions see Methods). Each leaf is shown at 7000×magnification(left) showing stomata horizontal field width= 36.6μm and500×magnification (right) horizontal field width= 512μm.Figure S2. Scanning electron microscopy micrographs showingwax platelets on leaf surface after loss of leaf hydrophobicityand leaf gas film disappearance (a, plants had been submergedin water with 50mM NaCl for 9 d), and wax platelets on papil-lae before (b, prior to submergence) and after (c, plants hadbeen submerged in water with 50mM NaCl for 9 d) loss of leafhydrophobicity and leaf gas film disappearance. Leaves areshown at 27.000× (a) and 45.000× (b, c) magnification (hori-zontal field width= 3.9 and 2.5μm, respectively).Figure S3. Leaf Na+ (a), Cl� (b) and K+ (c) concentrations inthe tissue water (mM) with time of submergence in water (con-taining basal ions, see Methods) with 0mM NaCl (squares) or50mM NaCl (circles) for rice plants with leaf gas films (+GF,

open symbols) or treated with 0.1% Triton X-100 and withoutgas films (�GF, closed symbols). Samples were the entire thirdand part of the fourth leaf (see Methods). Roots were in non-saline nutrient solution. Values are means [�SE, n=3–4 exceptNa+ on day 5 at 0mM NaCl (�GF) where n=1 due to a sam-pling error].Figure S4.Correlations among leaf Na+ (a), Cl� (b) and K+ (c)concentrations (data from Fig. 2) and corresponding leafchlorophylla concentrations (data from Fig. 1b) after submer-gence of rice in water (containing basal ions, seeMethods) with50mMNaCl. Leaves were either left untreated, thus retaining aleaf gas film (+GF, open symbols) or treated with 0.1% TritonX-100 (�GF, closed symbols). Roots were in non-saline nutri-ent solution. r values from non-parametric Spearman rank cor-relation analysis, * denoting levels of significance (levels ofP> 0.05, P≤ 0.05, P≤ 0.01, P≤ 0.001 or P≤ 0.0001 are denotedby n.s., *, **, ***, ****, respectively): Na+ r=�0.5084**; Cl�

r=�0.3658*; K+ r=0.4664**.Figure S5.Correlations among leaf Na+ (a), Cl� (b) and K+ (c)concentrations (data from Fig. 2) with corresponding underwa-ter PN (μmol O2 m�2 s�1) at 2500μM CO2 and 50mM NaCl(data from Fig. 3b). Leaves were either left untreated, thusretaining a leaf gas film (+GF) or treated with 0.1% Triton X-100 (�GF). Roots were in non-saline nutrient solution, andshoots were submerged in water containing NaCl treatmentsand basal ions (see Methods). r values from non-parametricSpearman rank correlation analysis, * denoting levels of signif-icance (levels of P> 0.05, P≤ 0.05, P≤ 0.01, P≤ 0.001 orP≤ 0.0001 are denoted by n.s., *, **, ***, ****, respectively):Na+ r=�0.6078**; Cl� r=�0.3243 n.s.; K+ r=0.1923 n.s. † de-notes points excluded from the correlation analysis to preventleaf deterioration with time of submergence to draw the corre-lation. Points were excluded when leaf both porosity andchlorophylla was <4.5% and 8.3mgg�1 DM, respectively, asfor day 16 (+GF) and days 5 and 9 (�GF). At 200μM, freeCO2 all correlations were not significant (Spearman rank cor-relation analysis, P> 0.05, data not shown).Figure S6. Leaf Na+ (a), Cl� (b) and K+ (c) concentrations of ayoungest fully expanded leaf of rice submerged in 50mM NaCl(containing also basal ions, see Methods) for 24 h in the darkwith gas films (+GF, open symbols) or treatedwith 0.1%TritonX-100 and without gas films (�GF, closed symbols). During in-cubation, the submergence solution was maintained at 0.01,0.46, 1.59, 3.16 and 20.23 kPa O2. * denotes significant differ-ence between ion concentrations (Sidak’s multiple compari-sons test, P< 0.05). Two-way GF×pO2 ANOVA showed asignificant effect ofGF on leafNa+ (P=0.0345) andK+ concen-trations (P=0.0091). pO2 had a significant effect onK

+ concen-trations (P< 0.0001). For Cl�, two-way ANOVA showed asignificant pO2×GF interaction (P=0.0050). Values are means(�SE, n=4).Table S1. Shoot length and number of tillers of rice plants sub-merged in water (containing basal ions, see Methods) with0mMNaCl or 50mMNaCl with leaf gas films (+GF) or treatedwith 0.1%TritonX-100 and without gas films (�GF). +GF and–GF plants are from different batches; hence, these have sepa-rate emergent (shoots in air) controls. Roots were in non-salinenutrient solution. Letters denote significant difference(P< 0.05) between means (�SE, n=4) according to one-wayANOVAwith Sidak’s multiple comparisons post hoc test (shootlength) or non-parametric Kruskal–Wallis with Dunn’s multi-ple comparisons test (number of tillers).



Conclusion, perspectives and outlook

Conclusion and perspectives for future research

In the work of this thesis, genotypic variation towards waterlogging (Chapter 1) and

submergence tolerance (Chapter 2) in wheat was established. Traits conferring waterlogging

tolerance (internal aeration, seminal root short term anoxia tolerance), as well as mechanisms in

need of further elucidation (e.g., anoxia tolerance, micronutrient toxicity, wheat root stele

proportions, phloem transport to root tips and downregulation of energy requirements) were

highlighted. In order to clarify the role of microelement toxicity in determining waterlogging

tolerance, toxicity thresholds for especially Fe2+ should be determined. Also, the location of

microelements within shoot tissues (using X-ray microanalysis) would be of interest in order to

determine their potential phototoxic effects. The effect of microelements on root function could

be clarified using the same approach; earlier work indicated that upon waterlogging Fe formed

plaques around roots but did not enter the symplast (Ding & Musgrave 1995).

Chapter 1 also highlighted that seminal root anoxia tolerance correlated to waterlogging

tolerance, but was not governed by fermentation capacity or root substrate supply. A suggested

follow-up on these exciting results could be to assess the ROS levels in these varieties, or

investigate if energy use is more efficient in some cultivars such as regulation of cytoplasmic pH

(Greenway & Gibbs 2003). In order to breed for more waterlogging-tolerant crops, the possible

locations of QTL for root aerenchyma formation and ROL barrier in wheat or wild relatives (as

in barley, Broughton et al. (2015)) could be of use (see Outlook for waterproofing crops – where

are we at?). The N deficiency in wheat owing to waterlogging could also be addressed by

seeking improved adventitious root nutrient uptake during waterlogging and possibly also

targeting N-use efficiency in wheat.

Chapter 2 showed that wheat submergence tolerance differed significantly between two

cultivars, however further work is needed to pinpoint traits conferring tolerance as our

hypothesis that shoot carbohydrate levels could explain varietal differences was not supported by

the results. The reputed waterlogging tolerance difference between the two cultivars suggests

that root traits should be further studied, or that other traits such as ROS-deprivation or ethylene

sensitivity could have caused the contrasting submergence tolerance. An approach similar to

rice, where rice cultivars with contrasting submergence tolerance were subjected to submergence

80


following pre-treatment with an ethylene inhibitor (1-methyl cyclopropene) could be used to

evaluate this aspect (Ella et al. 2003).

Experimental removal of leaf gas film decreased wheat and rice submergence tolerance (Chapter

3, Appendix 1). However, during a saline submergence of rice, significant intrusion of ions into

leaf tissues was recorded in spite of leaf gas film presence, indicating presence of a leaf-water

interface and thus contrasting with delayed salt intrusion into leaves of M. siculus retaining a leaf

gas film (Chapter 3). This raises questions about the structure of leaf gas films in these two

contrasting species. X-ray phase contrast micro-tomography used to visualise leaf gas films on

Spartina anglica indicated that leaf-water interfaces could exist on ridges of the plicate leaves;

rice also possesses plicate leaves. An experimental setup where rice leaves with gas films are

submerged into a solution containing compounds adhering to the leaf surface could reveal the

presence of such an interface. Moreover, clarifying if the “unidentified substance” covering

submerged leaves is a biofilm could be attempted using staining-techniques for biofilms.

Meanwhile, genotypic variation for leaf gas film retention time was not evident in Danish

cultivars (Appendix II), and contrasting submergence tolerance in two wheat cultivars was not

related to leaf gas film thickness or retention time (Chapter 2). Unless significant variation in gas

film retention time is detected in more distinct wheat genotypes or wheat relatives, a prominent

role of leaf gas films in wheat flood tolerance improvement is unlikely. When assessing the

ecological significance of leaf gas films in plant communities, a correlation between presence of

leaf gas films and the distribution of plant species along a flood gradient could not be established

(Winkel et al. 2016). This seems to indicate that in general, leaf gas film formation during

submergence is merely a side-effect of having a hydrophobic cuticle (for a number of reasons not

related to submergence) rather than for improved gas exchange upon submergence per se.

Nevertheless, gas films enhance aeration and fitness for some species experiencing short and

frequent submergence scenarios such as S. anglica growing in the intertidal zone (Winkel et al.

2011). Clarifying the precise pathways of cuticle wax synthesis and their limitations during

submergence could possibly also be of interest for the field of biomimicry, where scientist are

partly successful in mimicking natural superhydrophobic surfaces, but failing in maintaining this

characteristic for longer durations (Latthe et al. 2012).

81


Outlook for waterproofing crops – where are we at?

Given the genetic variation to especially submergence tolerance in rice (Das et al. 2005) and

waterlogging tolerance in wheat (Chapter 1), it becomes clear that excess water can be tolerated

by some crop genotypes. While the production of cultivars tolerant towards drought has been

relatively successful for major crops such as maize (La Rovere et al. 2010; Cooper et al. 2014),

the production of waterlogging-tolerant cultivars is still lagging behind as to my knowledge apart

from rice no other commercial flood-tolerant cultivars have been produced. This might seem

surprising considering the high flood-induced losses in crop production (Fig. 1), and the amount

of research conducted on crop flood tolerance concluding that genetic variation for breeding

efforts is present (Van Ginkel 1991; Musgrave & Ding 1998; Boru et al. 2001; Samad et al.

2001; Setter & Waters 2003; Collaku & Harrison 2005; Hayashi et al. 2013). In order to provide

an overview of the status and challenges of producing flood-tolerant cultivars, the following

sections review the current state in breeding for major crop (rice, wheat, soybean and maize)

flood tolerance.

Rice: Serving as a promising example for the production of flood-tolerant crops, submergence

tolerance was introduced into high-yielding rice cultivars in 2006 (Xu et al. 2006).

Submergence-tolerant cultivars able to survive 14 days of complete submergence (Fig. 6) and

with good agronomic traits were within 3 years of release grown by 4 million famers in India,

Laos, Philippines and Bangladesh (Ismail et al. 2013). In the previous 28 years, several flood-

resistant cultivars had been produced by conventional backcross-breeding, but had inferior grain

quality and were therefore not welcomed by farmers (Septiningsih et al. 2009). The successful

introduction of the Sub-1 locus conferring submergence tolerance but without significantly

affecting plant morphology was achieved using marker-assisted back-crossing (Septiningsih et

al. 2009; Iftekharuddaula et al. 2011). Submergence tolerance is conferred by inhibiting shoot

elongation and carbohydrate depletion upon submergence, resulting in yields twice as high in

submergence-tolerant compared to in intolerant cultivars (Das et al. 2005; Mackill et al. 2012).

Current efforts are to improve seedling anoxia tolerance in order to promote direct seeding

practices.

82


Figure 6. Pictures illustrating differential growth responses in wheat, rice, maize, soybean and/or their wild relatives upon flood

treatments. a) Waterlogging-tolerant breeding-line (96W639-D4-13, left) and commercial wheat variety Cascades (right) after 6

weeks of waterlogging 3 weeks after sowing. From Setter and Waters (2003). b) Submergence-tolerant wheat cultivar Swarna-

sub1 (right) and intolerant cultivar Swarna (left) after 10 days of submergence. Source: International Rice Research Institute. c)

Flood-intolerant maize Mi29 (left) compared to wild maize relative Zea nicaraguagensis (right) during a flood. From Omori et

al. (2016). d) Soybean (G. max, left) compared to wild relative (G. soja, right) upon flooding in University of Missouri field

evaluations. From Valliyodan et al. (2016).

Wheat: The use of marker-assisted selection (MAS) in producing flood-tolerant wheat has been

described as a great prospect for almost two decades now (Setter 2000; Setter & Waters 2003).

However, to my knowledge no commercial cultivars with a reputed high waterlogging tolerance

have been released using this or any other breeding approach, in spite of the notion that “three

promising lines are now in the final stages of breeding evaluations” (Setter & Waters 2003).

Apparently, the failure in developing a waterlogging-tolerant cultivar in this particular case was

the spreading of yellow rust to Australia to which the promising lines held no resistance (Pers.

Com., wheat breeder Robin Wilson, 2017).

83


Several studies have reported QTLs for waterlogging tolerance in wheat (Yu & Chen 2013;

Ballesteros et al. 2014; Yu et al. 2014) measuring shoot and root biomass, root length, tiller

number, chlorophyll content, plant height and other agronomical traits and detecting 34, 36 and

48 QTLs for waterlogging tolerance, respectively, across the genome. Such generic approaches

have been criticised as insufficient, as given the breadth of QTL location, undesirable genes will

inevitably be transferred during breeding attempts to move desirable genes into elite germplasm

(Shabala 2011). Instead, more specific mechanism conferring waterlogging tolerance should be

targeted, but this is often not achieved due to screening mostly taking place in the field where

techniques used in studies on waterlogging tolerance are not applicable (Shabala 2011).

The use of Hordeum marinum as donor of a ROL-barrier to wheat amphiploids has often been

referred to as a promising way to enhance flooding tolerance of wheat (Malik et al. 2011).

However, results from a recent study challenged the potential of this approach since disomic

chromosome addition lines did not possess a ROL barrier, suggesting that the genes for this trait

are not located on one single chromosome (Konnerup 2016). Hopes are also that the much more

advanced stage of rice genomics (owing its status as model genetic monocot) allows for the

exploitation of rice genomics information for wheat improvement (Anderson 2003).

The use of molecular markers for ion toxicity/waterlogging tolerance still warrants hope to

produce flood-tolerant wheat with good agronomic traits (Pers. Com., wheat breeder Robin

Wilson, 2017), especially with the use of next-generation breeding tools such as high-throughput

and cost-effective mapping and phenotyping becoming available for breeding (Valliyodan et al.

2016). However, future research should seek to clarify the role of micronutrient toxicity on a

range of soils in order to aid breeding efforts, as some divergence seems to exist as to which

traits mainly determine waterlogging tolerance.

Maize: Attempts to produce waterlogging-tolerant maize cultivars rely on the high waterlogging

tolerance in the wild maize relative teosinte (Zea nicaraguensis). Z. nicaraguensis grows in

lowland areas of Nicaragua that are frequently flooded during the rainy season and exhibits

adaptations such as constitutive aerenchyma formation, a ROL-barrier and formation of

adventitious roots at the soil surface. Researchers have followed a breeding strategy plan for a

decade now (Mano & Omori 2015), focusing on differentiating factors determining flooding

tolerance and analyzing each component separately, rather than selecting germplasm and

performing genetic analysis under field conditions. This approach was chosen due to the

84


observations that the flood tolerance ranking of varieties was inconsistent between different

researchers, and also difficult to repeat under field conditions due to interaction with

environmental factors (Omori et al. 2016). Work has resulted in determination of flood tolerance

trait QTL’s (mentioned above) in Z. nicaraguensis, and the production of near isogenic maize

lines possessing one or more QTL’s for the given traits using marker assisted selection (Mano &

Omori 2015; Omori et al. 2016). The chromosome location of a QTL for ROL-barrier has also

been identified (Watanabe et al. 2017), and one of the next steps will be to pyramid these traits

into maize. Expectations are that we will see the release of a flood-tolerant F1 hybrid within a

few years (Omori et al. 2016).

Soybean (Glycine max): Soybean germplasm has been screened for waterlogging tolerance,

showing that flood-intolerant lines lost twice as much yield as the flood-tolerant lines

(Valliyodan et al. 2016). Comparing two genotypes with contrasting waterlogging tolerance

indicated that development of aerenchyma and adventitious roots occurred faster in the flood-

tolerant genotype (Valliyodan et al. 2014).

Several QTLs have been mapped for soybean flood tolerance. Some QTLs were not repeatable

across years, or were tightly linked with flowering time (Githiri et al. 2006). The tight link

between flowering time and flood tolerance could be explained by a longer recovery period

when flowering is delayed (Githiri et al. 2006). Several other QTLs have been associated with

resistance to root fungal disease Phytophtora sojae (Cornelious et al. 2005; Valliyodan et al.

2016). Apparently, a breeding program by the University of Missouri has developed three

flooding lines through MAS exhibiting yields 90% of commercial checks under drained

conditions, but outperforming commercial checks during flooding by up to 1 ton/hectare

(Valliyodan et al. 2016). In addition, wild soybean (Glycine soja) is claimed to exhibit excellent

flood tolerance, outperforming G. max (Fig. 6) and thereby indicating potential for utilizing wild

relatives as in maize. However, these results and/or traits conferring higher tolerance in G. soja

have not yet been identified (Valliyodan et al. 2016).

In conclusion, research in dryland crop flood tolerance is ongoing, but as the production of

submergence-tolerant rice cultivars by IRRI (more than three decades of breeding efforts,

Septiningsih et al. (2009)) and the production of drought-tolerant maize by CIMMYT (four

decades of breeding efforts, estimated 2007-2016 research costs of $76 million, La Rovere et al.

(2010)) indicate, long term commitment and investment by foundations, governments or the

85


private sectors are necessary. Since commercial breeders are not inclined to cover costs to run

field trials for flood tolerance, globally dispersed and short term research projects without

sufficient infrastructure can contribute with knowledge on plant flood responses, but are unlikely

to result in major breakthroughs.

86

Literature cited

Literature cited in introduction, conclusion and outlook

Adhikari P., Hong Y., Douglas K.R., Kirschbaum D.B., Gourley J., Adler R. & Brakenridge G.R. (2010) A digitized global flood inventory (1998–2008): compilation and preliminary results. Natural Hazards, 55, 405-422.

Ali A. (1996) Vulnerability of Bangladesh to climate change and sea level rise through tropical cyclones and storm surges. Water, Air, and Soil Pollution, 92, 171-179.

Anderson J.A. (2003) Plant genomics and its impact on wheat breeding. Blackwell Publishing, Oxford, UK.

Armstrong W. (1979) Aeration in higher plants. In: Advances in Botanical Research (ed H.W. Woolhouse), pp. 225-332. Academic Press, London.

Armstrong W. & Drew M. (2002) Root growth and metabolism under oxygen deficiency. In: Plant Roots: The Hidden Half (eds Y. Waisel, A. Eshel, & U. Kafkaf), pp. 729-761. Marcel Dekker, New York & Basel.

Atwell B.J., Greenway H. & Colmer T.D. (2015) Efficient use of energy in anoxia‐tolerant plants with focus on germinating rice seedlings. New Phytologist, 206, 36-56.

Bailey-Serres J., Lee S.C. & Brinton E. (2012) Waterproofing crops: effective flooding survival strategies. Plant Physiology, 160, 1698-1709.

Bailey-Serres J. & Voesenek L.A.C.J. (2008) Flooding stress: Acclimations and genetic diversity. Annual Review of Plant Biology, 59, 313-339.

Ballesteros D., Mason R.E., Addison C., Andrea Acuña M., Nelly Arguello M., Subramanian N., . . . Tucker D. (2014) Tolerance of wheat to vegetative stage soil waterlogging is conditioned by both constitutive and adaptive QTL. Euphytica, 201, 329–343.

Blokhina O., Virolainen E. & Fagerstedt K.V. (2003) Antioxidants, oxidative damage and oxygen deprivation stress: a review. Annals of Botany, 91, 179-194.

Blom C.W.P.M. & Voesenek L.A.C.J. (1996) Flooding: the survival strategies of plants. Trends in Ecology & Evolution, 11, 290-295.

Boru G., van Ginkel M., Kronstad W.E. & Boersma L. (2001) Expression and inheritance of tolerance to waterlogging stress in wheat. Euphytica, 117, 91-98.

Boyer J.S. (1982) Plant productivity and environment. Science, 218, 443-448. Braimoh A., Emenanjo I., Rawlins M.A., Heumesser C. & Zhao Y. (2016) Climate-smart

agriculture indicators. World Bank Group, Washington D.C. Brammer H. (1990) Floods in Bangladesh: Geographical background to the 1987 and 1988

floods. The Geographical Journal, 156, 12-22. Broughton S., Zhou G., Teakle N., Matsuda R., Zhou M., O’Leary R., . . . Li C. (2015)

Waterlogging tolerance is associated with root porosity in barley (Hordeum vulgare L.). Molecular Breeding, 35, 1-15.

Collaku A. & Harrison S.A. (2005) Heritability of waterlogging tolerance in wheat. Crop Science, 45, 722-727.

Colmer T.D. (2003) Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant, Cell & Environment, 26, 17-36.

Colmer T.D. & Greenway H. (2011) Ion transport in seminal and adventitious roots of cereals during O2 deficiency. Journal of Experimental Botany, 62, 39-57.

Colmer T.D. & Pedersen O. (2008) Underwater photosynthesis and respiration in leaves of submerged wetland plants: gas films improve CO2 and O2 exchange. New Phytologist, 177, 918-926.

87

Literature cited

Colmer T.D. & Voesenek L.A.C.J. (2009) Flooding tolerance: suites of plant traits in variable environments. Functional Plant Biology, 36, 665-681.

Colmer T.D., Winkel A. & Pedersen O. (2011) A perspective on underwater photosynthesis in submerged terrestrial wetland plants. AoB Plants, 2011, plr030; doi:010.1093/aobpla/plr1030.

Cooper M., Gho C., Leafgren R., Tang T. & Messina C. (2014) Breeding drought-tolerant maize hybrids for the US corn-belt: discovery to product. Journal of Experimental Botany, 65, 6191-6204.

Cornelious B., Chen P., Chen Y., de Leon N., Shannon J.G. & Wang D. (2005) Identification of QTLs underlying water-logging tolerance in soybean. Molecular Breeding, 16, 103-112.

Das K.K., Sarkar R.K. & Ismail A.M. (2005) Elongation ability and non-structural carbohydrate levels in relation to submergence tolerance in rice. Plant Science, 168, 131-136.

de Guenni L.B., Cardoso M., Goldammer J., Hurtt G., Mata L.J., Ebi K., . . . Valdes J. (2005) Findings of the condition and trends working group: Regulation of natural hazards: Floods and fires. Millennium Ecosystem Assessment, Washington, DC.

Ding N.C. & Musgrave M.E. (1995) Relationship between mineral coating on roots and yield performance of wheat under waterlogging stress. Journal of Experimental Botany, 46, 939-945.

Ella E.S., Kawano N., Yamauchi Y., Tanaka K. & Ismail A.M. (2003) Blocking ethylene perception enhances flooding tolerance in rice seedlings. Functional Plant Biology, 30, 813-819.

Elzenga J.T. & van Veen H. (2010) Waterlogging and plant nutrient uptake. In: Waterlogging Signalling and Tolerance in Plants (eds S. Mancuso & S. Shabala), pp. 23-35. Springer, Berlin.

FAO (2009) Food and Agriculture Organization (FAO) of the United Nations: How to feed the world in 2050. Food and Agriculture Organization (FAO) of the United Nations, Rome, Italy.

Gibbs J. & Greenway H. (2003) Mechanisms of anoxia tolerance in plants. I. Growth, survival and anaerobic catabolism. Functional Plant Biology, 30, 1-47.

Gibbs J., Turner D.W., Armstrong W., Darwent M.J. & Greenway H. (1998) Response to oxygen deficiency in primary maize roots. I. Development of oxygen deficiency in the stele reduces radial solute transport to the xylem. Australian Journal of Plant Physiology, 25, 745-758.

Githiri S., Watanabe S., Harada K. & Takahashi R. (2006) QTL analysis of flooding tolerance in soybean at an early vegetative growth stage. Plant Breeding, 125, 613-618.

Greenway H. & Gibbs J. (2003) Mechanisms of anoxia tolerance in plants. II. Energy requirements for maintenance and energy distribution to essential processes. Functional Plant Biology, 30, 999-1036.

Hayashi T., Yoshida T., Fujii K., Mitsuya S., Tsuji T., Okada Y., . . . Yamauchi A. (2013) Maintained root length density contributes to the waterlogging tolerance in common wheat (Triticum aestivum L.). Field Crops Research, 152, 27-35.

Herzog M. (2013) Masters’s thesis (30 ECTS): Snorkeling in two Rumex species – effects of partial and complete submergence on internal aeration. Masters’s thesis, University of Copenhagen, Denmark.

Iftekharuddaula K.M., Newaz M.A., Salam M.A., Ahmed H.U., Mahbub M.A.A., Septiningsih E.M., . . . Mackill D.J. (2011) Rapid and high-precision marker assisted backcrossing to introgress the SUB1 QTL into BR11, the rainfed lowland rice mega variety of Bangladesh. Euphytica, 178, 83-97.

88

Literature cited

Ismail A.M., Singh U.S., Singh S., Dar M.H. & Mackill D.J. (2013) The contribution of submergence-tolerant (Sub1) rice varieties to food security in flood-prone rainfed lowland areas in Asia. Field Crops Research, 152, 83-93.

Jackson M.B. 2004. The impact of flooding stress on plants and crops [Online]. Available: http://www.plantstress.com/Articles/waterlogging_i/waterlog_i.htm [Accessed 14.07.2017].

Kalff J. (2002) Limnology: Inland water ecosystems. Prentice Hall, Upper Saddle River, NJ. La Rovere R.K., Abdoulaye G., Dixon T., Mwangi J., Guo W. & Z Banziger M. (2010) Potential

impact of investments in drought tolerant maize in Africa. CIMMYT, Mexico DF. Latthe S.S., Gurav A.B., Maruti C., Shridhar & Vhatkar R.S. (2012) Recent progress in

preparation of superhydrophobic surfaces: a review. Journal of Surface Engineered Materials and Advanced Technology, 2012, 76-94.

Mackill D.J., Ismail A.M., Singh U.S., Labios R.V. & Paris T.R. (2012) Development and rapid adoption of submergence-tolerant (Sub1) rice varieties. Advances in Agronomy, 11, 299-352.

Malik A.I., Islam A.K.M.R. & Colmer T.D. (2011) Transfer of the barrier to radial oxygen loss in roots of Hordeum marinum to wheat (Triticum aestivum): evaluation of four H. marinum–wheat amphiploids. New Phytologist, 190, 499-508.

Mano Y. & Omori F. (2015) Flooding tolerance in maize (Zea mays subsp. mays) F1 hybrids containing a QTL introgressed from teosinte (Zea nicaraguensis). Euphytica, 205, 255-267.

Millennium Ecosystem Assessment (2005) Ecosystems and Human Well-being: Synthesis. Millennium Ecosystem Assessment, Washington, DC.

Mommer L., Pedersen O. & Visser E.J.W. (2004) Acclimation of a terrestrial plant to submergence facilitates gas exchange under water. Plant, Cell & Environment, 27, 1281-1287.

Mommer L., Pons T.L., Wolters-Arts M., Venema J.H. & Visser E.J.W. (2005) Submergence-induced morphological, anatomical, and biochemical responses in a terrestrial species affect gas diffusion resistance and photosynthetic performance. Plant Physiology, 139, 497-508.

Musgrave M.E. & Ding N. (1998) Evaluating wheat cultivars for waterlogging tolerance. Crop Science, 38, 90-97.

NSSL. 2017. Flood types [Online]. National Oceanic and Atmospheric Administration. Available: http://www.nssl.noaa.gov/education/svrwx101/floods/types/ [Accessed 29.5 2017].

Olesen J.E., Trnka M., Kersebaum K.C., Skjelvåg A.O., Seguin B., Peltonen-Sainio P., . . . Micale F. (2011) Impacts and adaptation of European crop production systems to climate change. European Journal of Agronomy, 34, 96-112.

Omori F., Tamaki H., Mitsuhashi S. & Takahashi W. (2016) DNA marker-assisted selection approach for developing flooding-tolerant maize. Japan Agricultural Research Quarterly, 50, 175-182.

Parry M.L., Canziani O.F., Palutikof J.P., van der Linden P.J. & Hanson C.E. (2007) Climate change 2007: Impacts, adaptation and vulnerability. Contribution of working group ii to the fourth assessment report of the intergovernmental panel on climate change. IPCC, Cambridge, UK.

Pedersen O., Rich S.M. & Colmer T.D. (2009) Surviving floods: leaf gas films improve O(2) and CO(2) exchange, root aeration, and growth of completely submerged rice. The Plant Journal, 58, 147-156.

Ponnamperuma F. (1984) Effects of flooding on soils. In: Flooding and plant growth (ed T.T. Kozlowski), pp. 9-45. Academic Press, London.

89

http://www.plantstress.com/Articles/waterlogging_i/waterlog_i.htm

http://www.nssl.noaa.gov/education/svrwx101/floods/types/

Literature cited

Rienk R., Ploeg D., Machulla G., Hermsmeyer D., Ilsemann J. & Gieska M. (2002) Changes in land use and the growing number of flash floods in Germany. Paper presented at the Agricultural Effects on Ground and Surface Waters: Research at the Edge of Science and Society, Wageningen, Germany, October 2000.

Rosenzweig C., Tubiello F.N., Goldberg R., Mills E. & Bloomfield J. (2002) Increased crop damage in the US from excess precipitation under climate change. Global Environmental Change, 12, 197-202.

Samad A., Meisner C.A., Saifuzzaman M. & Ginkel M.v. (2001) Waterlogging tolerance. In: Application of Physiology in Wheat Breeding (eds M.P. Reynolds, J.I. Ortiz Monasterio, & A. McNab), pp. 136–144. CIMMYT, Mexico DF.

Sarwar G. & Khan M. (2007) Sea level rise. A threat to the coast of Bangladesh. Internationales Asienforum, 38, 375-398.

Sasidharan R., Bailey-Serres J., Ashikari M., Atwell B.J., Colmer T.D., Fagerstedt K., . . . Voesenek L.A.C.J. (2017) Community recommendations on terminology and procedures used in flooding and low oxygen stress research. New Phytologist, 214, 1403-1407.

Sauter M. (2013) Root responses to flooding. Current Opinion in Plant Biology, 16, 282-286. Sayre K., Van Ginkel M., Rajaram S. & Ortiz-Monasterio I. (1994) Tolerance to waterlogging

losses in spring bread wheat: effect of time of onset on expression. Annual Wheat Newsletter, 40, 165-171.

Septiningsih E.M., Pamplona A.M., Sanchez D.L., Neeraja C.N., Vergara G.V., Heuer S., . . . Mackill D.J. (2009) Development of submergence-tolerant rice cultivars: the Sub1 locus and beyond. Annals of Botany, 103, 151-160.

Setter T.L. (2000) Farming systems for waterlogging prone sandplain soils of the south coast. Grains Research Development Corporation, Agriculture Western Australia, Perth, W.A.

Setter T.L., Kupkanchanakul T., Kupkanchanakul K., Bhekasut P., Wiengweera A., Greenway H., . . . Greenway H. (1987) Concentrations of CO2 and O2 in floodwater and in internodal lacunae of floating rice growing at 1–2 metre water depths. Plant, Cell & Environment, 10, 767-776.

Setter T.L. & Waters I. (2003) Review of prospects for germplasm improvement for waterlogging tolerance in wheat, barley and oats. Plant and Soil, 253, 1-34.

Setter T.L., Waters I., Sharma S.K., Singh K.N., Kulshreshtha N., Yaduvanshi N.P.S., . . . Cakir M. (2009) Review of wheat improvement for waterlogging tolerance in Australia andIndia: the importance of anaerobiosis and element toxicities associated with differentsoils. Annals of Botany, 103, 221-235.

Shabala S. (2011) Physiological and cellular aspects of phytotoxicity tolerance in plants: the role of membrane transporters and implications for crop breeding for waterlogging tolerance. New Phytologist, 190, 289-298.

Striker G.G. (2012) Flooding stress on plants: Anatomical, morphological and physiological responses. In: Botany (ed J. Mworia), pp. 3-28. InTech.

Teakle N.L., Colmer T.D. & Pedersen O. (2014) Leaf gas films delay salt entry and enhance underwater photosynthesis and internal aeration of Melilotus siculus submerged in saline water. Plant, Cell & Environment, 37, 2339-2349.

Trnka M., Rotter R.P., Ruiz-Ramos M., Kersebaum K.C., Olesen J.E., Zalud Z. & Semenov M.A. (2014) Adverse weather conditions for European wheat production will becomemore frequent with climate change. Nature Climate Change, 4, 637-643.

Valliyodan B., Van Toai T.T., Alves J.D., de Fatima P.G.P., Lee J.D., Fritschi F.B., . . . Nguyen H.T. (2014) Expression of root-related transcription factors associated with flooding tolerance of soybean (Glycine max). International Journal of Molecular Sciences, 15, 17622-17643.

90

Literature cited

Valliyodan B., Ye H., Song L., Murphy M., Shannon J.G. & Nguyen H.T. (2016) Genetic diversity and genomic strategies for improving drought and waterlogging tolerance in soybeans. Journal of Experimental Botany, 68, 1835-1849.

Van Ginkel M., S. Rajaram & M. Thijssen. (1991) Waterlogging in wheat: Germplasm evaluation and methodology development. In: The 7th Regional Wheat Workshop for Eastern, Central, and Southern Africa (eds D.G. Tanner & W. Mwangi). CIMMYT, Najuru, Kenya.

Verboven P., Pedersen O., Ho Q.T., Nicolai B.M. & Colmer T.D. (2014) The mechanism of improved aeration due to gas films on leaves of submerged rice. Plant, Cell & Environment, 37, 2433-2452.

Voesenek L.A.C.J., Colmer T.D., Pierik R., Millenaar F.F. & Peeters A.J. (2006) How plants cope with complete submergence. New Phytologist, 170, 213-226.

Voesenek L.A.C.J. & Sasidharan R. (2013) Ethylene – and oxygen signalling – drive plant survival during flooding. Plant Biology, 15, 426-435.

Wassmann R., Hien N.X., Hoanh C.T. & Tuong T.P. (2004) Sea level rise affecting the Vietnamese Mekong Delta: Water elevation in the flood season and implications for rice production. Climatic Change, 66, 89-107.

Watanabe K., Takahashi H., Sato S., Nishiuchi S., Omori F., Malik A.I., . . . Nakazono M. (2017) A major locus involved in the formation of the radial oxygen loss barrier in adventitious roots of teosinte Zea nicaraguensis is located on the short-arm of chromosome 3. Plant, Cell & Environment, 40, 304-316.

Winkel A., Colmer T.D., Ismail A.M. & Pedersen O. (2013) Internal aeration of paddy field rice (Oryza sativa) during complete submergence – importance of light and floodwater O2. New Phytologist, 197, 1193-1203.

Winkel A., Colmer T.D. & Pedersen O. (2011) Leaf gas films of Spartina anglica enhance rhizome and root oxygen during tidal submergence. Plant, Cell & Environment, 34, 2083-2092.

Winkel A., Pedersen O., Ella E., Ismail A.M. & Colmer T.D. (2014) Gas film retention and underwater photosynthesis during field submergence of four contrasting rice genotypes. Journal of Experimental Botany, 65, 3225–3233.

Winkel A., Visser E.J., Colmer T.D., Brodersen K.P., Voesenek L.A., Sand‐Jensen K. & Pedersen O. (2016) Leaf gas films, underwater photosynthesis and plant species distributions in a flood gradient. Plant, Cell & Environment, 39, 1537-1548.

Xu K., Xu X., Fukao T., Canlas P., Maghirang-Rodriguez R., Heuer S., . . . Mackill D.J. (2006) Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature, 442, 705-708.

Yamauchi T., Shimamura S., Nakazono M. & Mochizuki T. (2013) Aerenchyma formation in crop species: A review. Field Crops Research, 152, 8-16.

Yu M. & Chen G.-Y. (2013) Conditional QTL mapping for waterlogging tolerance in two RILs populations of wheat. SpringerPlus, 2, 245.

Yu M., Mao S.-l., Chen G.-y., Liu Y.-x., Li W., Wei Y.-m., . . . Zheng Y.-l. (2014) QTLs for waterlogging tolerance at germination and seedling stages in population of recombinant inbred lines derived from a cross between synthetic and cultivated wheat genotypes. Journal of Integrative Agriculture, 13, 31-39.

91

Appendix I: Flood tolerance of wheat – the importance of leaf gas films during complete submergence

Photo: Ole Pedersen

92

Flood tolerance of wheat – the importance of leaf gas filmsduring complete submergence

Anders WinkelA,B, Max HerzogA, Dennis KonnerupA, Anja Heidi FloytrupA

and Ole PedersenA

AThe Freshwater Biological Laboratory, Department of Biology, University of Copenhagen,Universitetsparken 4, 3rd floor, 2100 Copenhagen, Denmark.

BCorresponding author. Email: [email protected]

Abstract. Submergence invokes a range of stressors to plants with impeded gas exchange between tissues andfloodwater being the greatest challenge. Many terrestrial plants including wheat (Triticum aestivum L.), possesssuperhydrophobic leaf cuticles that retain a thin gas film when submerged, and the gas films enhance gas exchangewith the floodwater. However, leaf hydrophobicity is lost during submergence and the gas films disappear accordingly.Here, we completely submerged wheat (with or without gas films) for up to 14 days and found that plants with gas filmssurvived significantly longer (13 days) than plants without (10 days). Plants with gas films also had less dead tissuefollowing a period of recovery. However, this study also revealed that reflections by gas films resulted in a higher lightcompensation point for underwater net photosynthesis for leaves with gas films compared with leaves without (IC = 52 vs35mmol photons m–2 s–1 with or without gas films, respectively). Still, already at ~5% of full sunlight the beneficial effectof gas films overcame the negative under ecologically relevant CO2 concentrations. Our study showed that dryland cropsalso benefit from leaf gas films during submergence and that this trait should be incorporated to improve flood toleranceof wheat.

Additional keywords: air film, flooding tolerance, hydrophobicity, underwater photosynthesis, underwater respiration,water repellent.

Received 7 November 2016, accepted 23 March 2017, published online 26 April 2017

Introduction

Torrential rains can result in overland floods that inundateterrestrial plants (Vervuren et al. 2003). With the currentprojection on climate changes, the frequency of such floodingevents is predicted to increase (Parry et al. 2006) and alreadynow flooding results in great annual losses in crop production(Bailey-Serres et al. 2012). Complete submergence invokes arange of stressors to terrestrial plants of which impeded gasexchange between tissues and floodwater poses the greatestchallenge (Armstrong 1979). Diffusion of gases occurs 10 000-fold faster in air compared with water so that in completelysubmerged plants, uptake from the floodwater of CO2 forphotosynthesis in light and O2 for aerobic respiration indarkness is strongly diffusion limited (Pedersen and Colmer2012). In order to sustain internal aeration, wetland plantshave porous tissues where aerenchyma forms a series ofinterconnected air channels enabling internal gas phasediffusion of both CO2 and O2 (Justin and Armstrong 1987).Another common trait of roots of wetland plants is a barrier toradial O2 loss (ROL) (Colmer 2003). The ROL barrier greatlyreduces radial O2 loss to the anoxic waterlogged soils andthereby enables aeration of the root tips so that the root cancontinue to grow (Waters et al. 1989).

The slow diffusion of gases in water also greatly restrictsunderwater net photosynthesis (PN) in aquatic plants (Madsenand Sand-Jensen 1991; Maberly and Madsen 2002) and insubmerged terrestrial wetland plants (Colmer et al. 2011).Terrestrial wetland plants lack most of the leaf acclimationtraits to reduce the resistance to CO2 uptake, of which themost important traits are those that reduce the resistancecaused by the diffusive boundary layers (Madsen et al. 1993).Nevertheless, as the CO2 concentration of floodwaters is oftenseveral fold above air equilibrium of 10–20mmolm–3 (Colmeret al. 2011), most terrestrial plants show some capacity forunderwater photosynthesis although rates are generally greatlyreduced compared with those in air (Colmer et al. 2011; Winkelet al. 2011, 2013, 2016). However, some terrestrial wetlandplants possess superhydrophobic leaves and retain a thinsilvery gas film on their leaves when submerged. Leafsuperhydrophobicity was first established as a self-cleansingmechanism (Barthlott and Neinhuis 1997; Neinhuis andBarthlott 1997), but the leaf gas films enable species withsuperhydrophobic cuticles to continue photosynthesising underwater at rates up to 27% (rice, Pedersen et al. 2009) and 28%(Hordeum marinum Huds., Pedersen et al. 2010) of those in airat ecologically relevant CO2 concentrations in the floodwater.

CSIRO PUBLISHING

Functional Plant Biology, 2017, 44, 888–898http://dx.doi.org/10.1071/FP16395

Journal compilation � CSIRO 2017 www.publish.csiro.au/journals/fpb

93

mailto:[email protected]

Leaf gas films have been shown to enhance submergencetolerance of completely submerged wetland plants. Leaf gasfilms lower the apparent resistance to gas exchange with thefloodwater several fold (Colmer and Pedersen 2008; Pedersenet al. 2009; Winkel et al. 2013; Verboven et al. 2014) andalthough the exact mechanism has yet to be revealed, the leafgas films resemble the ‘physical gills’ of some diving aquaticinsects (Pedersen and Colmer 2012). In a study of rice,completely submerged plants with leaf gas films intactcontinued to grow for 7 days at rates similar to those in air,whereas plants with gas films removed ceased growth underwater (Pedersen et al. 2009). Also H. marinum, a wild relativeto barley, continued to grow when completely submerged whenthe floodwater was enriched with dissolved CO2 (200mmolm–3)and the leaf gas films remained intact throughout the7 days of submergence (Pedersen et al. 2010). In addition toenhancing underwater PN and thus carbohydrate productionduring submergence, leaf gas films have also been shown toaid internal aeration of belowground tissues in darkness ofsubmerged terrestrial plants (Pedersen et al. 2009; Winkel et al.2011, 2013), and so the trait is crucial for submergence toleranceof terrestrial wetland plants. More recently, it has beenshown that also dryland plants possessing superhydrophobicleaf cuticles benefit from gas film formation when submerged(Winkel et al. 2016). In fact, rates of both O2 uptake in darknessand CO2 uptake and O2 production in light did not differbetween five species of wetland plants and nine species of

dryland plants that had leaf gas films when submerged(Winkel et al. 2016).

The wetland crop rice has been shown to greatly benefitfrom leaf gas films when submerged (Pedersen et al. 2009), butsome dryland crops also form gas films under water e.g. wheat(Fig. 1), barley and oats (Raskin and Kende 1983). Moreover,dryland crops such as wheat can experience submergence atvarious stages throughout their life cycle (Fig. 1) althoughcomplete inundation is more likely at the seedling stage(Herzog et al. 2016b) but submergence tolerance of wheat hasso far not been examined. This is in stark contrast to the vastliterature on responses to waterlogging of wheat that revealssome tolerance to waterlogging. A recent review showed thatacross 17 studies, the tissue porosity of adventitious rootsincreased from 5.2 under drained conditions to 14.8% (medianvalues) when waterlogged (Herzog et al. 2016b), indicating thatalso wheat can improve internal aeration by facilitating gasphase O2 transport to the growing root tips.

In the present study we tested the importance of leaf gasfilms for survival of completely submerged wheat (6–7 fullyexpanded leaves) by experimentally removing the gas films bybrushing the leaves before submergence with a dilute detergent.The plants were submerged up to 14 days and allowed aminimum of 14 days of recovery before they were scored deador alive and final biomass measured. During submergence,tissues were sampled to enable assessment of thickness ofleaf gas films, tissue porosity of the lamina and chlorophyll

(a) (b)

(c) (d)

Fig. 1. A field flood event inWestern Australia resulting in complete submergence of wheat (a) and a similar event inDenmark (b) where wheat is submerged late in the growth cycle. In (c), water forms perfectly spherical droplets on thesurfaces of the superhydrophobic leaf cuticle (contact angle = 155.9� 0.7�; mean� s.e., n= 8; see also Konnerup et al.(2017) for contact angle of additional 14 types of genotypes of wheat) and when submerged (d), a thin gas film isretained on leaf surfaces with a silvery appearance. The gas bubble (d) results from underwater photosynthesis.All photos are of wheat (Triticum aestivum) and in the case of (c) and (d), of the cultivar ‘JB Asano’.

Flood tolerance of wheat Functional Plant Biology 889

94

concentration. Plants were also grown to enable mechanisticstudies of the leaf gas films so that CO2 response curves andlight response curves of underwater PN could be establishedfor wheat. We hypothesised that plants with initially intact gasfilms would survive complete submergence better than plantswhere the gas films had been removed, and that plants with gasfilms would maintain tissue integrity longer. Mechanistically,we hypothesised that leaf gas films would enhance CO2 uptakeand thus underwater PN within ecologically relevant CO2

concentrations in the floodwater, but also that the reflection oflight by gas films (as indicated by the silvery appearance) wouldbe disadvantageous as limiting light levels, a novel aspect thathas not previously been considered.

Materials and methodsPlant material

For mechanistic studies of CO2 and light response of leafsegments, seeds of wheat (Triticum aestivum L., cv. ‘JBAsano’) were sown directly into a commercial potting mix(Pindstrup substrate no. 2, Pindstrup Mosebrug A/S) withOsmocote ‘Vegetable’ (N : P : K; 15.3 : 2.4 : 5.9) in 250mLplastic pots. Following germination, seedlings were thinned sothat each pot contained only one plant, and plants were grownin a glasshouse during August 2012 to an age of 4–5 weeks(daytime temperature 17�28�C; night-time temperature12�18�C).

For growth and survival experiments, seeds of the samecultivar were sown directly into a commercial potting mix(Pindstrup substrate no. 2, Pindstrup Mosebrug A/S) withOsmocote (N : P : K; 15.3 : 2.4 : 5.9) in 600mL pots with 3 cmof river sand in the bottom and on 3 cm of sand on top to reducethe flux of soil derived nutrients into the floodwater uponsubmergence. The beakers had drainage holes covered withtwo layers of fine mesh in order to constrain all roots to thepot. Following germination, seedlings were thinned so that eachpot contained only one plant, and plants were grown in aglasshouse during July–August 2016 to an age of 17 days(daytime temperature 17�32�C; night time temperature16�26�C) experiments ended 28 days later at which pointplants were 45 days old.

Growth and survival during submergence or waterlogging

Wheat occasionally becomes submerged also in well drainedsoils (Fig. 1) so survival during waterlogging or completesubmergence was assessed over a 14 day treatment periodfollowed by at least 2 weeks of recovery (Striker 2012). Priorto submergence, one set of plants (leaves and leaf sheaths) werebrushedwith 0.1%TritonX-100 (v/v) to remove hydrophobicity.Then, 160 individual 17-day-old plants were divided into fiveaquariums (32 in each, half with gas film removed) andcompletely submerged in artificial floodwater which contained(in mol m–3; Ca2+, 0.50; Mg2+, 0.25; Cl–, 1.00; SO4

2–, 0.25,with an alkalinity of 2molm–3 achieved by adding 200.2 g ofKHCO3 m–3) in a glasshouse with ambient light (PAR=11–1151mmol photons m–2 s–1; mid-day range, 1100–1500 hours). Floodwater O2 and CO2 concentrations weremaintained at air equilibrium (constant partial pressures) by

bubbling with atmospheric air throughout the experimentresulting in flow velocities of 0.5 to 3 cm s–1 dependent ondistance to air stone. Water temperature ranged from 18.8([O2] = 290; [CO2] = 16mmolm–3) to 28.9�C ([O2] = 240;[CO2] = 12mmolm–3) during the submergence period with amedian value of 22.8�C ([O2] = 269; [CO2] = 14mmolm–3. Aplastic mesh was placed below the surface to prevent elongatingshoots to restore atmospheric contact so that plants were alwayskept submerged by a minimum of 2 cm of water. At each timepoint, 10 individuals were desubmerged and left to drain (fivewith gas films initially and five with gas films removed). Another20 individuals were waterlogged in artificial floodwater and werede-waterlogged after 5, 10 and 14 days. Furthermore 10 plantswere kept under drained conditions with shoots in air (halfbrushed with Triton X-100, half unbrushed). Fv/Fm wasmeasured daily to ensure that there was no adverse effect onphotosynthesis in air of brushing with the detergent (see Fig S1,available as Supplementary Material to this paper). Survival wasscored after plants were 45 days old and live plant material wasseparated fromdeadandboth fractionswere freeze-dried toobtaindry mass (DM). With this arrangement, plants submerged for14 days had 2 weeks of recovery before scoring survival. Uponde-submergence shoot length, number of leaves and tillers wererecorded.

Leaf gas film thickness and porosity

Leaf gas film thickness and leaf porosity were measured usingthe buoyancy method (Raskin 1983; Pedersen et al. 2009). Thebuoyancy of 5 cm long leaf segments from the mid lamina of theyoungest fully expanded leaf was measured in de-ionised waterwith a 4 digit balance mounted with a hook underneath beforeand after brushing with 0.1% Triton X-100. Leaf gas filmthickness was calculated as gas volume (m3) divided by two-times the projected area (m2); see Colmer and Pedersen (2008)for details. Leaf segments were then placed under vacuum for atleast three times 5min until all internal gas volumes wereinfiltrated by water, and then buoyancy was assessed again.Leaf gas film thickness and leaf porosity were measuredimmediately after complete submergence (initials), after 24 hand thereafter each day until the thickness of the gas layer wasbelow detection limit (~1mm) and for leaf porosity also after 6, 9and 14 days of submergence.

Leaf chlorophyll concentration

Chlorophyll concentration was measured on the middle portionof what was the youngest fully expanded leaf at day 17 (day 0,start of treatment); at that time the wheat plants had 5–6 leaves.Leaf tips of the youngest fully expanded leaf were marked witha permanent marker to enable identification of leaves throughoutthe experiment. After harvest, leaf segments were freeze-driedfor 48 h, then cut into small pieces (less than 2mm2). Chlorophyllwas extracted using 96.5% ethanol at 20�C and incubated for24 h in darkness and then absorbance was measured at 665, 775and 649 nm on a spectrophotometer (UV-160A, Shimadzu).Chlorophyll concentrations were calculated using equations byWintermans and De Mots (1965).

890 Functional Plant Biology A. Winkel et al.

95

Underwater net photosynthesis (CO2 and light response)Underwater net photosynthesis (PN) by the lamina segmentswas measured using the method described by Winkel et al.(2013). In brief, for each replicate leaf (n= 4), one laminasegment of ~25mm length was taken half way up the blade ofthe youngest fully expanded leaf. Glass cuvettes (25mL)contained individual lamina segments in incubation mediumand two glass beads for mixing as these rotated on a wheelwithin an illuminated water bath, at 25�C. For CO2 responsemeasurements, PAR was 1000mmol photons m–2 s–1 inside thecuvettes measured using a spherical PAR sensor (US-SQS/L,Walz Heinz GmbH). Leaves were either brushed with a dilutedetergent (0.1% Triton X-100) to remove hydrophobicity(without gas films) (Raskin 1983; Colmer and Pedersen 2008)or transferred directly into the artificial floodwater (see below)and served as controls (with gas films).

The artificial floodwater serving as incubation mediumcontained (in mol m–3) Ca2+, 0.50; Mg2+, 0.25; Cl–, 1.00;SO4

2–, 0.25. The dissolved O2 concentration in the incubationmedium was set at ~50% of air equilibrium, by bubbling in 1 : 1volumes of N2 and air before adjustment of pH to obtain thedesired CO2 concentrations; this avoided build-up of excess O2

during the measurements that might otherwise have resulted inphotorespiration as described for submerged rice leaves (Setteret al. 1989). Dissolved CO2 treatments were imposed by addingspecific concentrations of KHCO3 to the incubation mediumwith pH adjusted to pH 9.00–6.17 to provide a range of CO2

concentrations from 3 to 2800mmolm–3 of constant carbonatealkalinity at 2molm–3 (Stumm and Morgan 1996).

Following60–90min incubation, dissolvedO2concentrationsin the cuvettes were measured using a Clark-type O2

minielectrode (OX-500, Unisense A/S), connected to a pAmeter (Multimeter, Unisense A/S) and were calibrated at 20.6kPa (air bubbled de-ionised water at 25�C) and at 0 kPa (de-ionised water with ascorbate and KOH at 25�C). Dissolved O2

concentrations in cuvettes prepared and incubated in the sameway as described above, but without leaf tissues, served asblanks. Projected area of leaf segments was measured using alaser leaf area meter (CI-202, CID Inc.).

Underwater PN as a function of light was measured asabove but at a CO2 concentration of 200mmolm–3. PAR inthe cuvettes was manipulated by varying the distance of thelight source (GP Plus 600W EL, Philips) or wrapping thecuvettes in neutral shading mesh resulting in a PAR gradientfrom 25 to 1000mmol photons m–2 s–1 inside the cuvettesmeasured as above. Cuvettes wrapped in aluminium foil servedas measurements for uptake of O2 in the dark under the samemixing conditions. Vials in dark and at low irradiance (25 to100mmol photons m–2 s–1) were incubated up to 180min inorder to achieve sufficient differences in pO2 between thesetreatments and blanks. For dark O2 uptake, the incubationmedium was initially set to air equilibrium to simulate O2 uptakein fully aerated floodwater.

Photosynthesis in air

In order to benchmark rates of underwater photosynthesis withrates in air, PN in air by intact leaves (youngest fully expanded)with ambient (390mmolmol–1) CO2 was measured at 25�C and

PAR of 1000mmol photons m–2 s–1, using a flow through leafcuvette connected to an infrared gas analyser (LCi-SD, ADC).Measurements were taken from five replicate plants.

Soil pO2

Soil pO2 was measured at the onset of waterlogging orsubmergence using 4 optical O2 needle sensors (OXF500PT,Pyroscience) connected to a four-channel meter (Firesting,Pyroscience). The optodes were positioned half way in betweenthe stem of the plant and the rim of the pot at ~5 cm depth andthe signal was logged every 1min. using OxygenLogger(Pyroscience). Prior to measurements, O2 optodes werecalibrated at 20.6 kPa (air bubbled de-ionised water at 25�C)and at 0 kPa (de-ionised water with ascorbate and KOH at25�C). Upon waterlogging or submergence, O2 disappearedfrom the soil matrix within 6–22 h.

Data analysisData was graphed and analysed using GraphPad Prism 6.01 or7.0; statistical tests used are given in table or figure captions.All tests were conducted at a probability level of 0.05 andmeans� s.e. are reported except in Table 1 where 95%confidence limits are also shown: 1- or 2-way ANOVAs wereused as appropriate and ANOVA details are shown in Tables S1and S2 available as Supplementary Material to this paper. ATukey or Sidak post hoc test was applied to analyse individualtime points. For survival data, a log rank (Mantel Cox) test wasused to test if survival differed with and without leaf gas filmsand median survival time is reported. Relative growth rateswere calculated assuming logarithmic growth or mortality inthe time interval. CO2 vs underwater PN was fitted usingMichaelis-Menten kinetics and light vs underwater PN kineticswere fitted to a Jassby and Platt (1976) photosynthesis function.Resistance to CO2 uptake was calculated as 1/initial slope ofthe CO2 response curve and the photosynthetic light useefficiency (Fi) was calculated as the initial slope of the lightresponse curve.

Results

Survival during complete submergence

The importance of leaf gas films for whole plant survival ofcomplete submergence was tested by comparing control plantswith plants where leaf cuticle hydrophobicity had beenexperimentally reduced so that gas films did not form whensubmerged. Plants were de-submerged at various time pointsranging from 6–14 days and scored dead or alive after a recoveryperiod of at least 14 days. Survival was defined as new growthafter de-submergence and all plants were scored 28 days afterthe beginning of the submergence event (day 0) regardless ofhow long they had been submerged.

Plants with intact gas films survived 3 days longer thanplants with their gas films removed. Control plants with leafgas films initially intact survived 13 days of completesubmergence but had a mortality rate of 80%, whereas after12 days of submergence the mortality rate was 0% (all plantssurvived, Fig. 2). In contrast, plants with leaf gas films initiallyremoved all survived 9 day of submergence but 80% had diedafter 10 days and none survived 11 days. The 3 day difference


96

in survival between control and treatment was statisticallysignificant.

Leaf gas film thickness, leaf tissue porosity and leaf tissuechlorophyll concentration

During the submergence experiment, leaf gas film thicknessand leaf tissue porosity were measured initially and after 1, 2,

3 and 4 days of submergence, whereas porosity was alsomeasured again after 6, 9 and 14 days of submergence. Gasfilm thickness of control plants rapidly declined so that 50% waslost already during the first 24 h of submergence (mean gasfilm thickness = 12mm) and after 3 days of submergence, thethickness was below detection limit and indistinguishablefrom plants brushed with Triton X-100 (Fig. 3a). Leaf tissueporosity remained constant between treatments for the first6 days of submergence with no significant effect of treatment(leaf gas film present or absent) but days of submergence (time)did show a significant effect (P< 0.05, 2-way ANOVA withSidak post hoc test); porosity values varied between 13 and21% (Fig. 3b). After 9 days of submergence, control plants(initially with intact leaf gas films) had significantly higherleaf tissue porosity than plants with leaf gas films initiallyremoved; 14 and 0% respectively (Fig. 3b). After 14 days ofsubmergence, water had infiltrated the leaf tissues and tissueporosity approached 0% (Fig. 3b). Overall, 2-way ANOVA(time� gas film) showed a significant effect of time and alsosignificant interaction effect indicating that the effect of timediffered for control plants and plants with gas films initiallyremoved (see Table S1 and caption of Fig. 3). Leafchlorophyll concentration followed the same pattern as leaftissue porosity with significant negative effects of time andinteraction between time and gas film, but with little effect ofpresence or absence of leaf gas films on early time points (seeTable S1 and caption of Fig. 3).

Relative growth rates, dead shoot mass and shootelongation

Relative growth rate (RGR) was calculated as relative increase inshoot DM from day 0 (beginning of the submergence event)to day 28 (end of recovery period) assuming exponential growth

Table 1. Underwater CO2 and light kinetics of photosynthesis for wheat (Triticum aestivum, cultivar‘JB Asano’)

Value With gas film Without gas film(mean ± s.e.) (mean ± s.e.)

CO2 kineticsUnderwater PN (mmol O2 m

–2 s–1) at 200mM CO2A 2.28 ± 0.1 0.387 ± 0.02

Underwater PN (mmol O2 m–2 s–1) at 18mM CO2

A 0.22 ± 0.004 0.16 ± 0.01Predicted Pmax (mmol O2 m

–2 s–1)B 5.45 ± 0.333 1.62 ± 0.123Air PN (mmol O2 m

–2 s–1)F 21.6 ± 1.7 naResistance to CO2 uptake (s m

–1)C 92 880 1 662 000

Light kineticsPredicted Pmax at 200mM CO2 (mmol O2 m

–2 s–1)B 2.61 ± 0.07 0.764 ± 0.02IC (Light compensation point, mmol photons m–2 s–1)D 52 (45–63) 35 (21–68)a (Light use efficiency, mol O2 mol photon–1)E 0.014 ±<0.001 0.017 ± 0.001RD (dark respiration. mmol O2 m

–2 s–1)A 0.061 ± 0.05 0.046 ± 0.02

APhotosynthetic rates obtained from day 0 (initials, Fig. 5).BPredicted maximum photosynthetic value calculated by fitting a Michaelis-Menten model to CO2 response-curve (Fig. 5) or a Jassby and Platt (1976) model to light response curve (Fig. 6).

CThe reciprocal value of the initial slope where linear in the CO2 response curve (Fig. 5).DIntersection of light response-curve with x-axis where y= 0 (Fig. 6).ESlope of linear curve at values between 0 and 50mmol photons m–2 s–1 in light response curve (Fig. 6).FMeasured on 37-day-old plants in air at 258C with ambient CO2. Light compensation point is given with 95%confidence intervals.

100

75

50

25

0 2 4 6 8 10

Days of submergence

Per

cent

sur

viva

l

12 14 160

With gas film Without gas film

Fig. 2. Survival of completely submerged wheat (Triticum aestivum,cultivar ‘JB Asano’) with or without leaf gas films. Plants were submergedin artificial floodwater and scored dead or alive following a minimum of14 days of recovery under drained conditions. The median survival time forplants with intact gas films was 13 days versus 10 days for plants where theleaf hydrophobicity had been removed (no gas film formation when underwater) with a dilute detergent before submergence. Data points aremean� s.e. (n = 5). The survival time is significantly different between thetwo treatments (P= 0.005; Mantal-Cox test).


97

ormortality in the period (Hunt 1982).Hence, the recovery periodranged from a minimum of 14 days for plants de-submergedafter 14 days to a maximum of 28 days (control plants in air

never submerged). RGR decreased significantly for plants thathad been submerged for 6 days but with no significant effect ofpresence or absence of leaf gas films (RGR had declined25–30% in plants with or without gas film compared withcontrols in air after 6 days of submergence; Table S1 andcaption of Fig. 4). For plants submerged for 9 days, there wasa clear but statistically insignificant trend towards control plants(initially with gas films) maintaining a higher RGR. However,after 10, 11 and 12 days of submergence, control plants hadsignificantly higher RGR than plants with gas films removedbefore submergence. At these time points, plants with gas filmsinitially removed showed negative RGR, i.e. the shoot massafter recovery was lower than at the time of submergence.After 13 and 14 days of submergence, there was no longer anysignificant difference between treatments since plants thatinitially had gas films had also died (Fig. 4a). To sum up, leafgas film presence resulted in higher growth when submergenceexceeded 6 days up until death (after 13 days). Similarly,waterlogging also significantly affected RGR, resulting in rates24% lower than the drained controls in air (t-test, P < 00001,n= 10) after 14 days of waterlogging.

At the final harvest after the recovery period, plants wereseparated into living or dead shoot tissues. Controls in air (neversubmerged but one set of plants brushed with Triton X-100) had6.5 and 7.3% dead shoot tissues, respectively, 28 days afteronset of the submergence event (plants were 45 days old).Plants submerged for 6 days had significantly more dead tissuethan controls in air mainly due to time of submergence and withonly little effect of presence or absence of gas films (Fig. 4b,Table S2). Control plants (initiallywith leaf gasfilms) submergedfor 10–12 days had significantly less dead shoot tissues thanplants with leaf gas films initially removed (Fig. 4b; plants withgas film initially removed had all died and thus consisted of alldead shoot tissues).

When plants were de-submerged at various time pointsduring the experiment, shoot length was measured. There wasno significant elongation of the shoot during the 14 days ofcomplete submergence (shoot height at time of submergenceversus shoot height after 14 days of submergence, t-test,P= 0.20, n= 5).

Mechanistic function of leaf gas films – CO2 responseof underwater PNUnderwater PN was measured at dissolved CO2 concentrationsfrom 3 to 2800mmolm–3 in the artificial floodwater atPAR= 1000mmol photons m–2 s–1 (Fig. 5). Underwater PN

increased steeply in a linear fashion at CO2 concentrationsbelow 400mmolm–3 (Fig. 5b) and reached saturation at~1000mmolm–3 with an estimated Pmax of 5.48mmol O2 m–2

s–1 for leaf segments with intact gas film (Fig. 5a, Michaelis-Menten model). In contrast, the estimated Pmax of leafsegments without gas films were significantly lower at only1.62mmol O2 m–2 s–1. At 200 mmol CO2 m–3, an ecologicalrelevant floodwater CO2 concentration (Colmer et al. 2011),underwater PN was 6-fold higher for leaf segments with gasfilms than those without (Fig. 5b). The survival experiment in thepresent studywas conducted at air equilibriumofCO2 (~18mmolCO2 m–3) and under these conditions, the underwater PN of

30

20

10

0

30

20

10

0

15

10

5

0 2 4 6 8 10

Days of submergence

(a)

(b)

(c)

Gas

film

thic

knes

s (µ

m)

Leaf

por

osity

(%

gas

vol

ume)

Chl

orop

hyll a

(m

g g–1

DM

)

12 14 160


Fig. 3. Leaf gas film thickness (a), leaf porosity (b) and leaf chlorophyll aconcentration (c) of wheat (Triticum aestivum, cultivar ‘JB Asano’) duringcomplete submergence. At day 0, plants were 17 days old, and half of theplants had the leaf hydrophobicity removed (no gas film formation whenunder water) with a dilute detergent before submergence. Data pointsare mean� s.e. (n= 5). Gas film thickness decreased significantly withtime of submergence (1-way ANOVA) for plants with intact gas filmsbefore submergence. Leaf porosity and leaf chlorophyll a also decreasedsignificantly with time of submergence and in both cases there was asignificant time� gas film interaction (see details of ANOVA output inTable S1, available as Supplementary Material to this paper).


98

control plants (initially with leaf gas films intact) was 0.22mmolO2m

2 s–1� 0.01 or 32%higher than thePN of plantswith leaf gasfilms initially removed (0.16mmol O2 m2 s–1� 0.004). Forcomparison, photosynthesis in air also at PAR= 1000mmolphotons m–2 s–1 and with ambient CO2 (390mmolmol–1) was21.6� 1.7mmol O2 m

–2 s–1 (mean� s.e., n= 4).The reciprocal of the initial slope of underwater PN vs

dissolved CO2 (Fig. 5b) indicates the apparent resistance toCO2 uptake by the leaf segments. Linear regression analysis ofunderwaterPN in the range from3 to 389mmolCO2m

–3 revealed

an apparent resistance of 92 884 s m–1 and 1 659 541 s m–1 forleaf segments with and without gas films respectively (Table 1).

Mechanistic function of leaf gas films – light responseof underwater PNUnderwater PN was also measured at a range of PAR with 200mmolCO2m

–3 in the artificialfloodwater (Fig. 6).UnderwaterPN

increased linearly at PAR <200mmol photons m–2 s–1 (Fig. 6b)and light saturationwas reached at ~200mmol photonsm–2 s–1 forleaveswith gasfilms andat less than100mmolphotonsm–2 s–1 forleaves with no gas films (Fig. 6b). The Jassby and Platt (1976)model predicted a Pmax of 2.61 and 0.76mmol O2 m

–2 s–1 for leafsegments with or without gas films, respectively. Rates ofunderwater PN at low PAR for leaf segments with gas filmwere significantly lower than for leaf segments without gas

0 2 4 6 8 10

Days of submergence

Days of recovery

(a)

(b)

RG

R (

day–1

)D

ead

shoo

t DM

(% o

f tot

al s

hoot

DM

)

12 14 16

28

0.2

0.1

100

75

50

25

0

0

–0.1

26 24 22 20 18 16 14


Fig. 4. Relative growth rate (a, RGR) and percent of dead shoot mass(b) with time of complete submergence for wheat (Triticum aestivum, cultivar‘JB Asano’). At day 0, plants were 17 days old, and half of the plants had theleaf hydrophobicity removed (no gasfilm formationwhen underwater) with adilute detergent before submergence. RGR was calculated from shoot massof 17-day-old plants (day 0) compared with shoot mass of 45-day-old plantswith 0 to 14 days of complete submergence and 14 to 28 days of recovery(second horizontal axis). At harvest (day 28), shoot tissue was separated intodead or alive tissue. Data points are mean� s.e. (n= 5). The shoot mass ofcontrol plants (never submerged) was 2.65� 0.07 and 2.11� 0.15 (P< 0.05,t-test), for plants brushed with a dilute detergent or brushed with DI waterrespectively. Plants with gas film had significantly higher RGR and lessdead shoot tissue (see Table 2S, available as Supplementary Material to thispaper, for details of ANOVA analyses; * denotes significantly different timepoints according to a Sidak test).

0 50 100

PN (

µmol

O2

m–2

s–1

)

Dissolved CO2 (mmol m–3)

150 200

0 1000 2000 3000

8

6

4

2

0

3

2

1

0


(a)

(b)

Fig. 5. Underwater net photosynthesis (PN) versus dissolved CO2

(a) 0–3000mmolm–3; (b) 0–200mmolm–3, of leaf segments of wheat(Triticum aestivum, cultivar ‘JB Asano’) with or without leaf gas films.During incubation (1–3 h), the leaf segments were exposed to 1000mmolphotons m–2 s–1 and PN was measured as O2 produced. (a) Each dataset wasfitted to aMichaelis-Mentenmodel (Km= 198and733mmolm–3 for segmentswith andwithout gas film, respectively,Vmax = 5.48 and 1.62mmolO2m

–2 s–1

for segments with and without gas film, respectively, r2 = 0.82 and 0.80for segments with and without gas film respectively). Data points aremean� s.e. (n= 4).


99

film up to between 50 and 100mmol photons m–2 s–1 (Fig. 6b).The light compensation point was 52mmol photons m–2 s–1 forleaves with gas films and 35m–2 s–1 for leaves without,indicating less light captured by leaves with gas films at lowlight intensities also confirmed by the lower photosyntheticefficiency (Fi) by leaves with gas films (0.01394mol O2 molphoton–1) compared with leaves with no gas film (0.0167mol O2

mol photon–1). However, already at around 100mmol photonsm–2 s–1 (~5% of full daylight) the improved gas exchange ofleaves with gas films overcame the negative effect of lightreflection by gas films.

Dark O2 uptake (aerobic respiration, RD) of leafsegments was significantly higher for leaf segments with gasfilm(s) present compared with segments without gas films

demonstrating that even at atmospheric equilibrium of O2

(258mmolm–3 at 25�C), the apparent resistance was too highto saturate RD for leaf segments without gas film (RD = 0.061 and0.046mmol O2 m

–2 s–1 for segments with or without gas filmsrespectively).

Discussion

This study demonstrates a beneficial effect of leaf gas films onwhole-plant survival during complete submergence; plantswith intact gas films at the time of submergence survivedsignificantly longer than plants that were manipulated so thatgas films did not form upon submergence. Early signs ofsenescence such as loss of leaf porosity and chlorophyll adegradation were both correlated with gas film retention time.Moreover, this study also revealed that the reflection by thesilvery gas layers on the leaf surfaces can be disadvantageousat low light, and below we discuss these novel findings in thecontext of superhydrophobic cuticles as a trait that confers floodtolerance to terrestrial wetland plants and also to the drylandcrop wheat.

Growth and survival

In the present study we found that the wheat cultivar ‘JB Asano’survived 13 days of complete submergence with intact leaf gasfilms at the time of submergence. This is almost a week shorterthan 10 rice cultivars that all survived more than 17 daysalthough with severe damages in the intolerant genotypes(Singh et al. 2009). In contrast, when leaf hydrophobicity ofthe wheat cultivar ‘JB Asano’ had been manipulated so that gasfilms did not form upon submergence, survival was only 10 days.Even though the leaf gas films were lost after 3 days ofsubmergence (Fig. 3a, see below), this trait extended thesurvival time by 3 days (Fig. 2a), indicating that the beneficialeffect of the leaf gas film in the beginning of the submergenceevent carried some benefit that enabled the plants to survive at alater stage. It is already well established that leaf gas filmsenhance gas exchange (CO2 and O2 during the day and O2

during the night) of completely submerged plants (Colmer andPedersen 2008;Winkel et al. 2016). Also in the present study, wefound that underwater PN was significantly higher for plantswith intact gas films, even at atmospheric equilibrium of CO2

(~18mmolm–3 free CO2) which is in the lower range of CO2

concentrations found in overland floodwaters (Colmer et al.2011). Similarly, underwater RD at air equilibrium of O2 wasalso significantly higher in the presence of leaf gas films(Fig. 6b). In a previous study, the beneficial effect of leaf gasfilms on dark uptake of O2 was shown to help maintaining arelatively constant RD down to 30–50% of air equilibrium ofO2, but in the case of e.g. Phragmites australis and Phalarisarundinacea underwater RD was also not saturated atatmospheric levels of pO2 in leaf segments without gas film(Colmer and Pedersen 2008). Hence, we suggest that the higherunderwater PN in combination with the better O2 supply indarkness would have resulted in a better carbohydrate balancewith production during the day and conservation during thenight via strictly aerobic metabolism, supported by the studyof Pedersen et al. (2009), which showed higher shoot sugarconcentration in plants with gas film compared with plants

PN (

µmol

O2

m–2

s–1

)

PAR (µmol photons m–2 s–1)

0 250 500 750 1000

0 25 50 75 100

1.0

0.5

0

–0.5

–1.0

3

2

1

–1

0


(a)

(b)

Fig. 6. Underwater net photosynthesis (PN) vs light (PAR=0–1000mmolphotons m–2 s–1, (a) 0–100mmol photons m–2 s–1, (b) of leaf segments ofwheat (Triticum aestivum, cultivar ‘JB Asano’) with or without leaf gasfilms. During incubation leaf segments were exposed to 200 mmol CO2 m

–3

and PN was measured as O2 produced. (a) Each dataset was fitted to a Jassbyand Platt (1976) model (Pmax = 2.61 and 0.76mmol O2 m

–2 s–1 for segmentswith and without gas film, respectively, a = 0.014 and 0.017mol O2 mol–1

photons for segments with and without gas film, respectively, IC = 48 and35mmol photonsm–2 s–1 for segmentswith andwithout gasfilm, respectively,r2 = 0.97 and 0.93 for segments with and without gas film respectively. Datapoints are mean� s.e. (n= 4).


100

without gasfilms after 7 days of submergence.Moreover, it is alsoknown from rice that root O2 of completely submerged plants isstrongly correlated with incident light and thus underwater PN

(Winkel et al. 2013), so even though rates of underwater PN aremodest comparedwith photosynthesis in air, O2 produced in lightby plants with gas films enhances internal aeration of the rootsystem (Pedersen et al. 2009) and a healthy root system is crucialfor survival following submergence or waterlogging (Herzoget al. 2016b). As a consequence, we are unable to pinpointthe positive effects of leaf gas films resulting in better survivalduring complete submergence, i.e. carbohydrates produced inunderwater photosynthesis and/or better internal aeration due tophotosynthetically produced O2 and/or O2 taken up from thefloodwater in darkness.

Plants that had leaf gas films during the first 3 days ofsubmergence had significantly higher values of leaf porosity,leaf chlorophyll a concentration and RGR whereas the amountof dead shoot tissue was less than in plants that had leafhydrophobicity manipulated before submergence (Figs 3, 4).As plants started reacting to submergence stress, leafchlorophyll a declined rapidly and the leaves lost structuralintegrity allowing water to infiltrate intercellular gas spaces; inour experience, leaves that have started to lose chlorophyll aand tissue porosity, do not recover when de-submerged (Herzoget al. 2016a). In a similar study by Herzog et al. (2016a) onrice, where leaf hydrophobicity had also been manipulatedbefore submergence, leaf chlorophyll a and porosity alsodeclined faster in plants with no gas films at the time ofsubmergence. Likewise, the loss of leaf chlorophyll a fromcontrol plants with intact gas films resembled the time courseof two rice genotypes (IR42 and Swarna, sensitive tosubmergence) in a field flood trial where leaves also had lost50% of the initial Chlorophyll on day 8–9 and with a somewhatparallel decline in leaf porosity (Winkel et al. 2014). Leafporosity as well as leaf chlorophyll a concentration followedthe exact same pattern for both controls with intact gasfilms at thetime of submergence and plants with gas films removed, butshifted by 3 days, and therefore aligned well with the patternobserved for survival.

Gas films and gas exchange

The wheat cultivar ‘JB Asano’ retained leaf gas films for the first3 days of submergence but then some changes in the leaf cuticleresulted in loss of hydrophobicity and gas films were no longerpresent. A screening study on gas film retention time in 14genotypes of wheat, showed very little genotypic variation inthis trait and in all cases, gas films disappeared after 3–4 days ofsubmergence (Konnerup et al. 2017). Only the wild wetlandplant, Glyceria fluitans, retained a superhydrophobic cuticlefor almost a week (Konnerup et al. 2017), and studies on rice(Colmer et al. 2009; Winkel et al. 2014) and H. marinum (seabarley grass, a wild relative to barley) (Pedersen et al. 2010) alsoshowed that leaf gas films can be retained for up to 1 week ofsubmergence. At present, however, it is not known why thecuticle loose hydrophobicity with time of submergence butone study showed that leaves of submerged wheat werecovered by an unidentified substance covering the cuticle(Konnerup et al. 2017) whereas in another study, the leaves of

rice did not develop such a layer but nevertheless lost leafhydrophobicity (Herzog et al. 2016a).

Leaf gas films on wheat enabled higher underwater PN

under environmentally relevant CO2 concentrations, and theeffect of gas films on CO2 uptake was significant both in termsof efficiency (Fig. 5b, initial slope) and capacity (Table 1). Wefound that rates of underwater PN were several fold higher inleaf segments with gas film at all except for the lowest CO2

concentrations (50mmolm–3 and below). The beneficial effectof leaf gas films on underwater PN has previously been shownfor other terrestrial wetland plants such as rice (Pedersen et al.2009) and P. australis (Colmer and Pedersen 2008). We notedthat underwater PN of wheat at low CO2 concentrations(~20mmolm–3) in this study was similar to that of P. australis(0.22 vs 0.14mmol O2 m–2 s–1 respectively), whereas rice hasbeen shown to obtain higher rates (up to 1.07mmol O2 m

–2 s–1,Winkel et al. 2013). Although underwater PN of wheat is only afraction of the rate in air (21.6mmol O2 m–2 s–1), we havepreviously shown that such low photosynthetic rates can becrucial for aeration of roots in rice (Winkel et al. 2013). Ifwheat possesses a similar gas pathway between roots andshoot as indicated by Malik et al. (2003), underwaterphotosynthesis in the presence of gas films could play animportant role in internal aeration, even at low external CO2

concentrations.Wheat showed the same pattern of CO2 response of

underwater PN as P. australis (Colmer and Pedersen 2008)where leaf segments without gas film never intercepted thecurve of segments with gas film even at very high external CO2

concentrations. We speculate that something restricts CO2 andO2 exchange across the cuticle of both wheat and P. australisand perhaps result in higher rates of photorespiration. In thepresent study, the resistance to CO2 uptake was 1.6� 106 s m–1

for leaf segments without leaf gas film compared with theresistance of P. australis (0.5� 106 s m–1, Colmer and Pedersen2008) and rice (0.3� 106 s m–1, Pedersen et al. 2009). Moreoverin rice, the disadvantage of the missing gas film could bealleviated by high external CO2 concentrations (Pedersen et al.2009). Such alleviation does not take place in wheat or inP. australis indicating an even more crucial role of leaf gasfilms for gas exchange in completely submerged wheat.

The role of leaf gas films in light use efficiency (Fi) duringunderwater PN has not previously been evaluated. In accordancewith our initial hypothesis, we found that gas films reflect lightand therefore are disadvantageous at low light levels and thereflection results in a higher light compensation point (IC,Fig. 6b). We found that leaves without gas films had slightlyhigher light use efficiency than leaves with leaf gas films (0.017and 0.014mol O2 mol–1 incident photons in leaves withoutand with gas films, respectively) but in both cases Fi was inthe lower range of those previously reported for aquaticangiosperms (0.012–0.058mol O2 mol–1 incident photons;Frost-Christensen and Sand-Jensen (1992)). In the formerstudy, however, a much higher CO2 concentration was used(500mmolm–3) in order to saturate underwater PN whereas inthe present study, we used a more environmentally relevant CO2

concentration of 200mmolm–3 (Colmer et al. 2011) and thelower CO2 concentration could subsequently have resulted in alower light use efficiency. In conclusion, the light use efficiency


101

during underwater photosynthesis of this dryland crop is notlargely different from many amphibious wetland plants.

The present study shows that gas films forming onsuperhydrophobic leaf cuticles during submergence is a strongtrait resulting in longer survival during complete submergence inthe dryland crop, wheat. At present, the genetic background forcuticle structure is poorly understood but a recent study on ricesuggest that at least one gene is involved in the deposition of thenumerouswax platelets (Zhang et al. 2016)which is an importantstructural feature resulting in superhydrophobicity. At present, itis not yet knownwhy leaf cuticles loose hydrophobicity over timewhen submerged but it happens in both wetland and drylandplants (Neinhuis and Barthlott 1997; Winkel et al. 2014, 2016;Konnerup et al. 2017). It is possible that with better geneticunderstanding of features involved in the 3D structure of thecuticle that more flood tolerant crops can be developed if thegenes involved in cuticle maintenance can be overexpressedand this could lead to longer leaf gas film retention times andpossibly increase the duration that wheat can survive completesubmergence.

Acknowledgements

We acknowledge Timothy Colmer for constructive discussion onexperimental design, Oliver Mørk for helping growing and maintainingplants as well as laboratory work and Victoria C Herskov for help withanalyses of chlorophyll. The study is supported by a grant (WheatSUB) fromthe Villum Foundation.

References

Armstrong W (1979) Aeration in higher plants. Advances in BotanicalResearch 7, 225–332. doi:10.1016/S0065-2296(08)60089-0

Bailey-Serres J, Lee SC, Brinton E (2012) Waterproofing crops: effectiveflooding survival strategies. Plant Physiology 160, 1698–1709.doi:10.1104/pp.112.208173

Barthlott W, Neinhuis C (1997) Purity of the sacred lotus, or escape fromcontamination in biological surfaces.Planta 202, 1–8. doi:10.1007/s004250050096

Colmer TD (2003) Aerenchyma and an inducible barrier to radial oxygenloss facilitate root aeration in upland, paddy and deep-water rice (Oryzasativa L.). Annals of Botany 91, 301–309. doi:10.1093/aob/mcf114

Colmer TD, Pedersen O (2008) Underwater photosynthesis and respirationin leaves of submerged wetland plants: gas films improve CO2 and O2

exchange. New Phytologist 177, 918–926. doi:10.1111/j.1469-8137.2007.02318.x

Colmer TD, Vos H, Pedersen O (2009) Tolerance of combined submergenceand salinity in the halophytic stem-succulent Tecticornia pergranulata.Annals of Botany 103, 303–312. doi:10.1093/aob/mcn120

Colmer TD, Winkel A, Pedersen O (2011) A perspective on underwaterphotosynthesis in submerged terrestrial wetland plants. AoB Plants 2011,plr030. doi:10.1093/aobpla/plr030

Frost-Christensen H, Sand-Jensen K (1992) The quantum efficiency ofphotosynthesis in macroalgae and sumberged angiosperms. Oecologia91, 377–384. doi:10.1007/BF00317627

Herzog M, Konnerup D, Pedersen O, Winkel A, Colmer TD (2016a) Roleof leaf gas films in improving rice (Oryza sativa) submergencetolerance during saline floods. Plant, Cell & Environment doi:10.1111/pce.12873

Herzog M, Striker GG, Colmer TD, Pedersen O (2016b) Mechanisms ofwaterlogging tolerance in wheat – a review of root and shoot physiology.Plant, Cell & Environment 39, 1068–1086. doi:10.1111/pce.12676

Hunt R (1982) ‘Plant growth curves. The functional approach to plant growthanalysis.’ (Edward Arnold Ltd.: London)

Jassby AD, Platt T (1976) Mathematical formulation of the relationshipbetween photosynthesis and light for phytoplankton. Limnology andOceanography 21, 540–547. doi:10.4319/lo.1976.21.4.0540

Justin S, Armstrong W (1987) The anatomical characteristics of roots andplant-response to soil flooding. New Phytologist 106, 465–495.doi:10.1111/j.1469-8137.1987.tb00153.x

Konnerup DW, Winkel A, Herzog M, Pedersen O (2017) Leaf gas filmretention during submergence of 14 cultivars of wheat (Triticumaestivum). Functional Plant Biology. doi:10.1071/FP16401

Maberly SC, Madsen TV (2002) Freshwater angiosperm carbonconcentrating mechanisms: processes and patterns. Functional PlantBiology 29, 393–405. doi:10.1071/PP01187

Madsen TV, Sand-Jensen K (1991) Photosynthetic carbon assimilation inaquatic macrophytes. Aquatic Botany 41, 5–40. doi:10.1016/0304-3770(91)90037-6

Madsen TV, Sand-Jensen K, Beer S (1993) Comparison of photosyntheticperformance and carboxylation capacity in a range of aquaticmacrophytes of different growth forms. Aquatic Botany 44, 373–384.doi:10.1016/0304-3770(93)90078-B

Malik AI, Colmer TD, Lambers H, Schortemeyer M (2003) Aerenchymaformation and radial O2 loss along adventitious roots of wheat withonly the apical root portion exposed to O2 deficiency. Plant, Cell andEnvironment 26, 1713–1722. doi:10.1046/j.1365-3040.2003.01089.x

Neinhuis C, Barthlott W (1997) Characterization and distribution of water-repellent, self-cleaning plant surfaces. Annals of Botany 79, 667–677.doi:10.1006/anbo.1997.0400

Parry ML, Canziani OF, Palutikof JP, Van Der Linden PJ, Hanson CE(2006) ‘Climate change 2007: impacts, adaptation and vulnerability.’(Cambridge University Press: Cambridge, UK)

Pedersen O, Colmer TD (2012) Physical gills prevent drowning of manywetland insects, spiders and plants. The Journal of Experimental Biology215, 705–709. doi:10.1242/jeb.065128

Pedersen O, Rich SM, Colmer TD (2009) Surviving floods: leaf gas filmsimprove O2 and CO2 exchange, root aeration, and growth of completelysubmerged rice. The Plant Journal 58, 147–156. doi:10.1111/j.1365-313X.2008.03769.x

PedersenO,MalikAI,ColmerTD(2010)Submergence tolerance inHordeummarinum: dissolved CO2 determines underwater photosynthesis andgrowth. Functional Plant Biology 37, 524–531. doi:10.1071/FP09298

Raskin I (1983) A method for measuring leaf volume, density, thickness,and internal gas volume. HortScience 18, 698–699.

Raskin I, Kende H (1983) How does deep water rice solve its aerationproblem. Plant Physiology 72, 447–454. doi:10.1104/pp.72.2.447

Setter TL,Waters I,Wallace I, Bekhasut P, GreenwayH (1989) Submergenceof rice. I. Growth and photosynthetic response to CO2 enrichment offloodwater. Australian Journal of Plant Physiology 16, 251–263.doi:10.1071/PP9890251

Singh S,Mackill DJ, Ismail AM (2009) Responses of SUB1 rice introgressionlines to submergence in the field: yield and grain quality. Field CropsResearch 113, 12–23. doi:10.1016/j.fcr.2009.04.003

Striker G (2012) Time is on our side: the importance of considering arecovery period when assessing flooding tolerance in plants.Ecological Research 27, 983–987. doi:10.1007/s11284-012-0978-9

StummW,Morgan JJ (1996) ‘Chemical equilibria and rates in naturalwaters.’(John Wiley & Sons Inc.: New York)

Verboven P, Pedersen O, Ho QT, Nicolai BM, Colmer TD (2014) Themechanism of improved aeration due to gas films on leaves ofsubmerged rice. Plant, Cell & Environment 37, 2433–2452.

Vervuren PJA, Blom CWPM, de Kroon H (2003) Extreme floodingevents on the Rhine and the survival and distribution of riparian plantspecies. Journal of Ecology 91, 135–146. doi:10.1046/j.1365-2745.2003.00749.x


102

dx.doi.org/10.1016/S0065-2296(08)60089-0

dx.doi.org/10.1104/pp.112.208173

dx.doi.org/10.1007/s004250050096

dx.doi.org/10.1007/s004250050096

dx.doi.org/10.1093/aob/mcf114

dx.doi.org/10.1111/j.1469-8137.2007.02318.x

dx.doi.org/10.1111/j.1469-8137.2007.02318.x

dx.doi.org/10.1093/aob/mcn120

dx.doi.org/10.1093/aobpla/plr030

dx.doi.org/10.1007/BF00317627

dx.doi.org/10.1111/pce.12873

dx.doi.org/10.1111/pce.12873

dx.doi.org/10.1111/pce.12676

dx.doi.org/10.4319/lo.1976.21.4.0540

dx.doi.org/10.1111/j.1469-8137.1987.tb00153.x

dx.doi.org/10.1071/FP16401

dx.doi.org/10.1071/PP01187

dx.doi.org/10.1016/0304-3770(91)90037-6

dx.doi.org/10.1016/0304-3770(91)90037-6

dx.doi.org/10.1016/0304-3770(93)90078-B

dx.doi.org/10.1046/j.1365-3040.2003.01089.x

dx.doi.org/10.1006/anbo.1997.0400

dx.doi.org/10.1242/jeb.065128

dx.doi.org/10.1111/j.1365-313X.2008.03769.x

dx.doi.org/10.1111/j.1365-313X.2008.03769.x

dx.doi.org/10.1071/FP09298

dx.doi.org/10.1104/pp.72.2.447

dx.doi.org/10.1071/PP9890251

dx.doi.org/10.1016/j.fcr.2009.04.003

dx.doi.org/10.1007/s11284-012-0978-9

dx.doi.org/10.1046/j.1365-2745.2003.00749.x

dx.doi.org/10.1046/j.1365-2745.2003.00749.x

Waters I, Armstrong W, Thompson CJ, Setter TL, Adkins S, Gibbs J,Greenway H (1989) Diurnal changes in radial oxygen loss andethanol-metabolism in roots of submerged and non-submerged riceseedlings. New Phytologist 113, 439–451. doi:10.1111/j.1469-8137.1989.tb00355.x

Winkel A, Colmer TD, Pedersen O (2011) Leaf gas films of Spartina anglicaenhance rhizome and root oxygen during tidal submergence. Plant, Cell& Environment 34, 2083–2092. doi:10.1111/j.1365-3040.2011.02405.x

Winkel A, Colmer TD, Ismail AM, Pedersen O (2013) Internal aerationof paddy field rice (Oryza sativa) during complete submergence –

importance of light and floodwater O2. New Phytologist 197,1193–1203. doi:10.1111/nph.12048

Winkel A, Pedersen O, Ella E, Ismail AM, Colmer TD (2014) Gas filmretention and underwater photosynthesis during field submergence of

four contrasting rice genotypes. Journal of Experimental Botany 65,3225–3233. doi:10.1093/jxb/eru166

Winkel A, Visser EJ, Colmer TD, Brodersen KP, Voesenek LA, Sand-JensenK, Pedersen O (2016) Leaf gas films, underwater photosynthesis andplant species distributions in a flood gradient. Plant, Cell & Environment39, 1537–1548. doi:10.1111/pce.12717

Wintermans JFGM, De Mots A (1965) Spectrophotometric characteristicsof chlorophylls a and b and their phenophytins in ethanol. Biochimica etBiophysica Acta 109, 448–453. doi:10.1016/0926-6585(65)90170-6

Zhang Z, Cheng Z-J, Gan L, Zhang H,Wu F-Q, Lin Q-B,Wang J-L,Wang J,Guo X-P, Zhang X, Zhao Z-C, Lei C-L, Zhu S-S, Wang C-M, Wan J-M(2016) OsHSD1, a hydroxysteroid dehydrogenase, is involved in cuticleformation and lipid homeostasis in rice. Plant Science 249, 35–45.doi:10.1016/j.plantsci.2016.05.005


www.publish.csiro.au/journals/fpb

103

dx.doi.org/10.1111/j.1469-8137.1989.tb00355.x

dx.doi.org/10.1111/j.1469-8137.1989.tb00355.x

dx.doi.org/10.1111/j.1365-3040.2011.02405.x

dx.doi.org/10.1111/nph.12048

dx.doi.org/10.1093/jxb/eru166

dx.doi.org/10.1111/pce.12717

dx.doi.org/10.1016/0926-6585(65)90170-6

dx.doi.org/10.1016/j.plantsci.2016.05.005

Appendix II: Leaf gas film retention during submergence of 14 cultivars of wheat (Triticum

aestivum)

hydrophobicity. Photo: Dennis Konnerup

Water drop on a superhydrophobic leaf surface. The angle between water drop and leaf

surface is used to quantify leaf hydrophobicity. Photo: Dennis Konnerup

104

Leaf gas film retention during submergence of 14 cultivarsof wheat (Triticum aestivum)

Dennis KonnerupA,B, Anders WinkelA, Max HerzogA and Ole PedersenA

AFreshwater Biological Laboratory, Department of Biology, University of Copenhagen, Universitetsparken 4,3rd floor, 2100 Copenhagen, Denmark.

BCorresponding author. Email: [email protected]

Abstract. Flooding of fields after sudden rainfall events can result in crops being completely submerged. Some terrestrialplants, including wheat (Triticum aestivum L.), possess superhydrophobic leaf surfaces that retain a thin gas film whensubmerged, and the gas films enhance gas exchange with the floodwater. However, the leaves lose their hydrophobicityduring submergence, and the gas films subsequently disappear. We tested gas film retention time of 14 different wheatcultivars and found that wheat could retain the gas films for a minimum of 2 days, whereas the wild wetland grassGlyceriafluitans (L.) R.Br. had thicker gasfilms and could retain its gasfilms for aminimumof 4 days. Scanning electronmicroscopyshowed that the wheat cultivars and G. fluitans possessed high densities of epicuticular wax platelets, which couldexplain their superhydrophobicity. However,G. fluitans also had papillae that contributed to higher hydrophobicity duringthe initial submergence and could explain whyG. fluitans retained gas films for a longer period of time. The loss of gas filmswas associated with the leaves being covered by an unidentified substance.We suggest that leaf gas film is a relevant trait touse as a selection criterion to improve the flood tolerance of crops that become temporarily submerged.

Additional keywords: air film, flooding, underwater photosynthesis, wettability.

Received 10 November 2016, accepted 31 January 2017, published online 23 March 2017

Introduction

Floods are expected to increase as a consequence of climatechange (Parry 2007) and increased rainfall will affect wheat(Triticum aestivum L.) production negatively in some areas(Trnka et al. 2014). It has been estimated that 15–20% of the70million ha of wheat sown each year is affected negatively bywaterlogging (Sayre et al. 1994; Setter and Waters 2003).Therefore, an increased effort is needed to breed wheatcultivars that are better adapted to the regionally abiotic stressin the future climate scenarios, where some areas will experienceincreased rainfall and some areas increased drought (Trnkaet al. 2014). Most flooding studies refer to the waterloggingscenario and use ‘flooding’ and ‘waterlogging’ interchangeably.According to Bailey-Serres et al. (2012), the term ‘flooding’ isused to describe the inundation by water of all or part of a plant,‘waterlogging’ to describe flooding of the root system and‘submergence’ to describe the situation where most or allaerial tissue is under water, thus we will use this terminologyin the present paper.

Waterlogging results in reduction of plant growth as thesoil becomes anoxic and in addition various potentially toxiccompounds such as Mn2+, Fe2+, S2- and carboxylic acids canaccumulate (Ponnamperuma 1984) with phytotoxic effects onroots (Shabala 2011). Furthermore, nutrient uptake and thegrowth and yield of wheat are severely reduced whenwaterlogging occurs and roots suffer O2 deficits (Colmer andGreenway 2011; Herzog et al. 2016b).

Root growth in anoxic or severely hypoxic soils depends ontheO2 that is availablevia diffusion fromabovegroundparts of theplant through aerenchyma (Armstrong 1979; Colmer 2003). Thegas-filled channels of aerenchyma and other porous tissuesprovide an internal pathway with low resistance for O2

diffusion within roots, enhancing O2 movement to the root tipswhen in anoxic substrates (Armstrong 1979). A second featurethat enhances the movement of O2 to the root tip, in addition toroot gas-filled porosity, is the formation of a barrier to radial O2

loss (ROL) in the basal part of the roots. This feature occurs inroots of many wetland plants and diminishes ROL to therhizosphere and can also restrict phytotoxin entry, whereas indryland specieswithout the barrier,mostO2 is lost at the basal rootzone so that lessO2 reaches the root tip (Armstrong 1979; Colmer2003). An approach aimed at improving the waterloggingtolerance of wheat is the transfer of the barrier to ROL from awild relative, Hordeum marinum Huds., to wheat by widehybridisation (Malik et al. 2011). This work has resulted in thesuccessful development of amphiploids with a barrier to ROL butwith lower yield, so further breeding is needed to producewaterlogging and commercially viable cultivars (Konnerupet al. 2017).

Submergence is a more severe type of flooding as it involveswaterlogging of the root systemand the shoot being submerged infloodwater. During submergence, the ~10 000-fold reduction indiffusion of gases in floodwaters limits the availability of O2

and CO2 for aerobic respiration and net photosynthesis (PN),

CSIRO PUBLISHING

Functional Plant Biology, 2017, 44, 877–887http://dx.doi.org/10.1071/FP16401

Journal compilation � CSIRO 2017 www.publish.csiro.au/journals/fpb

105

mailto:[email protected]

respectively (Armstrong 1979). Plants can photosynthesise underwater provided that both the light andCO2 levels are sufficient. Inair, CO2 enters the tissue via open stomata, but under water, thestomata are assumed to close (Mommer and Visser 2005).Therefore, CO2 first has to overcome the resistance caused bythe aqueous diffusive boundary layer via slow moleculardiffusion and then subsequently cross the cuticle, which addssignificantly to the total resistance to CO2 uptake (Mommer andVisser 2005). Some terrestrial wetland plants acclimate tofacilitate gas exchange under water by the production of new‘semi-aquatic’ leaves that are thin, as well as having reducedcuticles and rearrangement of chloroplasts closer to theepidermis,all resulting in lower resistance to CO2 diffusion to chloroplasts(Mommer et al. 2005; Mommer and Visser 2005).

Many terrestrial plants have superhydrophobic leaves thatresult in the formation of thin gas films when immersed intowater (Colmer and Pedersen 2008). The leaf gas films have beenshown to increase underwater gas exchange and thus CO2

entry to sustain underwater PN during the day and improvedinternal aeration during the night (Pedersen et al. 2009; Winkelet al. 2013). Thus, retention and persistence of gas films by plantsare beneficial during submergence, but factors involved in thedegradation of leaf gas films during prolonged submergencerequires further study. In rice, gas films were retained onsubmerged leaves for 4–6 days depending on genotype and theloss of gas films were strongly linked to a steep decline inunderwater PN (Winkel et al. 2014). In addition to providingthe leaves with the ability to form gas films, superhydrophobicleaf surfaces arehypothesised tobeanadaptation for self-cleaningand facilitate water to roll off leaves in air when it rains to preventcovering of leaves by a film of water (Neinhuis and Barthlott1997). Awater layer on a leaf surface would reduce gas exchangeand thus PN, and also increase the risk of bacteria and fungiinfecting leaves (Koch and Barthlott 2009).

In wheat, superhydrophobic leaf surfaces have also beendemonstrated to form gas films when submerged, andexperiments on the cultivar ‘JB Asano’ showed that plants withgas films survived complete submergence significantly longer(13 days) than plants without (10 days), which could be highlyrelevant to improve survival of temporarily submerged wheatduring field flood events (Winkel et al. 2017). This promptedus to investigate if there is any diversity of gas film retention indifferentwheat cultivars andwhat occurswhen the leaves lose theirhydrophobicity and subsequently the gas films. Thus, in thepresent study, we tested the gas film retention times duringsubmergence of 14 cultivars of wheat, which were submergedfor up to 10 days. Based on these results, we chose three wheatcultivars for more in-depth mechanistic studies consisting ofunderwater PN, hydrophobicity measured as contact angles, leafchlorophyll and scanning electron microscopy (SEM) withGlyceria fluitans included as a flooding tolerant model species.

Materials and methodsPlant material and growth conditionsFourteen wheat (Triticum aestivum L.) cultivars were usedin the study. Thirteen of the cultivars were commonly used inagriculture in northern Europe with no information onwaterlogging tolerance and one was a waterlogging tolerant

cultivar (Jackson) (Huang and Johnson 1995; Huang et al.1997). Nine of the cultivars (Mariboss, Jensen, Nakskov,Torp, Gedser, Ohio, Albert, Elixir, Ambition) were suppliedby Nordic Seed and four cultivars (Hereford, Substance,Benchmark and Sheriff) by Sejet Plant Breeding. Furthermore,floating sweet-grass (Glyceria fluitans (L.) R.Br.), a wild grassspecies which is tolerant to waterlogging and temporarysubmergence was included in the study as a flooding tolerantmodel species (Mony et al. 2010).

The experiment was conducted in a constant temperatureroom at 15�C. Seeds (three per pot) were sown in plastic pots(diameter 60mm, height 90mm) with substrate. The substrateconsisted of a 10mm layer of washed sand at the bottom, then a70mm layer of soil with a dry matter content of 55–75 gL–1,conductivity 3.0–5.0mS cm–1 and pH 6.0 (Pindstrup Substrateno. 2, Pindstrup Mosebrug A/S) and a 10mm layer of washedsand on top; the top sand layer served to reduce nutrient releasefrom the soil to the floodwater during submergence. Pellets ofslow-release fertilizer (OsmocoteBloom)were implanted into thesoil and each pot received 0.55 g fertiliser that contained (by %mass): N, 12; P, 7.0; K, 18.0; Mg, 1.5; Fe, 0.35; Mn, 0.05; Cu,0.045; Mo, 0.017; Zn, 0.013; B, 0.01.G. fluitanswas propagatedby using small plants taken from tillers and potted as wheat i.e.same soil and fertilizer. The pots, which had drainage holes at thebottom, were placed on trays with 10mm DI water, so they hadaccess to sufficient water. Following germination, the seedlingswere exposed to light (12 h photoperiod) using an artificialfull spectrum light source (Gavita Pro LEP 300, GavitaHolland BV) providing photosynthetically active radiation(PAR) of ~200mmol photons m–2 s–1 at the pot surface.

Experimental design and submergence conditions

Submergence treatments commenced when plants were 16 daysold (2.0–2.5 leaf stage). One of the three seedlings in each potwas removed, so that each pot contained two seedlings ofsimilar size. The plants were submerged in glass aquariums(length�width� height, 0.8m� 0.4m� 0.5m) with artificialfloodwater consisting of a modified Smart and Barko (1985)solution containing (in mM): CaCl2*2H2O, 0.62; MgSO4*7H2O,0.28; NaHCO3, 1.0; KCO3, 1.2. In each aquarium, therewas a pHcontroller (JBL CO2/pH control) connected to a cylinder withpressurised CO2 keeping the pH at 7.35 by bubbling with CO2

when necessary. This maintained the free CO2 at 200mM, whichis an environmentally relevant concentration (Colmer et al. 2011)and the solution had an alkalinity of 2.0mM (Mackereth et al.1978). The floodwater was circulated through UV filters (JBLAquaCristal UV-C) in order to prevent algal growth in theaquariums. The plants were submerged for 0 (non-submergedinitials), 1, 2, 4, 7 or 10 days, and control plants in air werealso measured on day 10. All measurements were performed onthe youngest fully expanded leaf at the time of submergence (the1st leaf).

The study consisted of two experiments (experiment 1 and 2).In experiment 1, the gas film retention timesmeasured as gas filmthickness during submergence of 14 different wheat cultivarswere assessed in order to choose cultivars for experiment 2. Inexperiment 2, gasfilm thickness andmore in-depthmeasurementsconsisting of underwaterPN, contact angle, chlorophyll and SEM

878 Functional Plant Biology D. Konnerup et al.

106

were undertaken on three of the 14 cultivars from experiment1with contrasting retention times, andG. fluitanswas included asa flooding tolerant model species.

Gas film thickness and leaf porosity

Gas film volume was measured by determining buoyancy of leafsegments (60–80mm in length) before and after gasfilm removal.The gas films were removed by brushing the leaves with a dilutesolution (0.05% v/v) of Triton X-100 (Raskin and Kende 1983;Colmer andPedersen2008). The leaf segmentswere thenvacuuminfiltrated with water and buoyancy was determined again toenable calculation of tissue porosity (gas-filled volume per unittissue volume) using the method described by Raskin (1983) andthe equations as modified by Thomson et al. (1990). Projectedarea wasmeasured for each leaf segment using digital photos andimage analysis in ImageJ (Schneider et al. 2012). Mean gas filmthicknesswas calculated by dividing gas film volume (mm3)withthe two-sided area (mm2), as the leaves of both wheat andG. fluitans possessed gas films on both the adaxial and abaxialsides. In the present study, the detection limit of gasfilm thicknesswas~2mm,and thereforemeasurements resulting in values below2mm were classified as ‘gas films absent’.

Leaf surface wettability

Leaf surface wettability was assessed by measuring the contactangle of a 5-mm3 droplet of water on the leaf surface (Brewer andSmith 1997), held flat using double-sided tape. Droplets wereapplied to the adaxial leaf surface, and photographed at� 90magnification using a horizontally positioned digital microscopecamera (Dino-Lite AM4013MZTL, IDCP) and the contact anglewas determined by image analysis in ImageJ. The wettability ofthe leaf surfaces could be divided into four classes defined bytheir contact angle according to Koch and Barthlott (2009):superhydrophilic (contact angle <10�), hydrophilic (contactangle 10�-90�), hydrophobic (contact angle 90–150�) andsuperhydrophobic (contact angle >150�).

Net photosynthesis (PN) under water

Underwater PN was measured as net O2 production by leafsegments. The leaf segments were excised from the middlethird of the lamina and were incubated in a defined medium(see later in this section) for a known time in closed transparentvials with gentle mixing and held at a constant temperature afterwhich the O2 evolution by the tissue was measured. UnderwaterPN rates (n= 4) were measured at 20�C using 25mL glass vialswith two glass beads added to ensure mixing as the vials rotatedaccording to the method of Pedersen et al. (2013) withphotosynthetically active radiation (PAR) inside the vials of1000mmol photons m–2 s–1 provided by a light source (GavitaPro LEP 300, Gavita Holland BV) with a horizontal light beam.The basal incubation medium was based on the general purposeculture medium described by Smart and Barko (1985) andcontained (in mM) CaCl2.2H2O, 0.62; MgSO4*7H2O, 0.28. Toprepare artificial floodwater with a final concentration of 200 or2500mM CO2 and an alkalinity of 2.0mM (mostly bicarbonateand carbonate), we addedKHCO3 at 2.2 or 4.5mM in the generalpurpose medium. We subsequently added known volumes of0.1M HCl to convert the desired portion of the HCO3

- into CO2,

resulting in pH values of 7.4 and 6.35 for the 200 and 2500mMCO2 respectively (Mackereth et al. 1978). The dissolved O2

concentration in the incubation medium was set at 50% of airequilibrium (mixing solutions bubbled with N2 and air in 1 : 1volumes); this procedure was applied to reduce effects ofphotorespiration (Pedersen et al. 2013). Vials without leafsegments served as blanks.

Following incubations of known duration (60–120min), thedissolved O2 concentration in each vial was measured using amicrooptode (OP-430, fibre diameter 430mm, Unisense A/S)connected to an optode meter (MicroOptode meter, UnisenseA/S). Values of O2were logged on a computer using the softwareprovided by Unisense (Sensor Trace Suite ver. 2.7.0). Projectedarea was measured as described for ‘gas film thickness’ for eachindividual leaf segment and underwater PN rates were calculatedper one-sided leaf area. The leaf segments were then immediatelyfrozen at�20�C and later freeze-dried for subsequent analysis ofchlorophyll.

Chlorophyll

Chlorophyll a concentration was measured in the leaf segmentsused for underwaterPN.The freeze-dried leaf segmentswere eachhomogenised in a 2mL Eppendorf tube using 2 metal beads for10 s on a mini bead-beater (BioSpec Products Inc. Mini BeadBeater). Chlorophyll was extracted for 24 h in 96% ethanol,centrifuged at 8000g for 3min and chlorophyll a absorbancemeasured at 656 and 750 nm on a spectrophotometer (ShimadzuUV-1800). Chlorophyll a concentrations were calculated usingequations by Mackinney (1941).

SEM and analyses of SEM

Leaf segments were frozen immediately after sampling andthen freeze-dried. Samples were gold coated in a sputter coaterfor 90 s and then analysedusing afield emission scanning electronmicroscope (JEOL JSM-6335F) at 7.0 kV and �10 000magnification or a scanning electron microscope (FEI InspectS) at 12.5 kV and �500 magnification. As the submergenceresulted in the wax platelets on the leaves being covered by anunidentified substance, the images were analysed using ImageJto determine the percentage of the leaf surface which was stillexposed after 0 (non-submerged initials), 1, 2, 4, 7 and 10 dayssubmergence.

Statistical analyses

Data were analysed by analysis of variance (ANOVA) usingType III sum of squares with the software Statgraphics XVIcenturion version 16.1.11 (StatPoint Inc.). Multiple post hoccomparisons of means were performed using the Tukey’s HSDprocedure at the 0.05 significance level. Data were tested forhomogeneity of variance by Levene’s test. If necessary,logarithmic or square root transformations were performed toensure homogeneity of variance, but for clarity all data arepresented as untransformed. Two-tailed Pearson correlationsbetween underwater PN at the two CO2 concentrations and gasfilm thickness were performed using GraphPad Prism 6(GraphPad Software Inc.).

Gas film retention during submergence Functional Plant Biology 879

107

Results

Gas film retention

In the gas film screening experiment (experiment 1), all thewheatcultivars had gas films on both sides of the leaves whensubmerged. The initial gas film thicknesses were in the rangefrom15 to 33mmand declinedwith time of submergence (Fig. 1).At day 2, the gas film thicknesses were 2 to 8mm, and by the4th day, gas films had declined to <2mm indicating that the gasfilms had been lost since the detection limit was defined to be2mm. The results of a 2-way ANOVA of the 14 tested cultivarsshowed that there was a significant main effect of time ofsubmergence (P < 0.0001), but no effect of cultivar and withno significant cultivar� time of submergence interaction. Thisdemonstrated that all the cultivars lost their gas films in a similarpattern with time of submergence. In the mechanistic experimenton the focus species (experiment 2), the three wheat cultivarsAlbert, Jensen and Jackson were again included and comparedwith G. fluitans. For these mechanistic studies we selected thecultivar with thickest gas film (Jensen), the one with thinnest gasfilm (Albert) and Jackson because this cultivar has been reportedto be waterlogging tolerant (Huang and Johnson 1995; Huanget al. 1997). Furthermore, we included G. fluitans in these in-depth studies as this species has been reported to be floodingtolerant (Mony et al. 2010) and our field observations had shownthat it possessed gas films when submerged under naturalconditions. For the wheat cultivars, the initial gas filmthickness was 24–25mm and as in experiment 1, the gas filmswere lost by day 4 (Fig. 2a). ForG.fluitans, the initial gasfilmwas36mmand it wasmaintained at this thickness until day 2, where itstarted to decline and was lost by day 7. In a 2-way ANOVAanalysis, therewere significant effects of cultivar/species and timeof submergence and also a significant interaction (Table 1). The

post hoc test showed that the three wheat cultivars were similar atall time points, whereas G. fluitans had thicker gas films thanwheat at day 0, 1, 2 and 4.

Leaf surface wettability

Leaf surfacewettability assessed as contact angleswere152–157�

initially demonstrating that all wheat cultivars and alsoG. fluitanshad superhydrophobic leaves before submergence (Fig. 2b). Theleaves of G. fluitans maintained superhydrophobic properties(contact angle >150�) until day 2, whereas the three wheatcultivars had lost the superhydrophobicity already on day 1with contact angles of 130–140�. However, by day 4 theleaves of all plants had lost their hydrophobicity and werehydrophilic. The 2-way ANOVA showed significant effects ofcultivar/species, timeof submergence andcultivar/species� timeof submergence interaction as the leaves of G. fluitans weresuperhydrophilic at day 7 and day 10 with contact anglesclose to 0�.

The initial (Day 0) superhydrophobicity of the leaf surfaces inboth wheat andG. fluitanswas caused by a dense arrangement ofthree-dimensional epicuticular waxes (Fig. 3). In addition, theleaf surfaces of G. fluitans had papillae and a plicate structure,which could explain their thicker gas films and longer lastinghydrophobicity. The declines in gas film thickness andhydrophobicity measured as contact angles were associatedwith the leaves being covered by a layer of an unidentifiedsubstance observed using SEM (Fig. 3). The wax crystalsseemed intact underneath the layer, but the three wheatcultivars were partly covered by the unidentified substanceon day 2 and almost completely covered on day 4 (Figs 2c, 3).The leaves of G. fluitans showed a delayed response and werecompletely covered on day 7. A plot of gas film thickness vs

40Jensen Mariboss

ElixirAmbition

AlbertJackson

Gedser SheriffNakskovTorpOhio

BenchmarkHerefordSubstance

30

20

10

0

40

Gas

film

thic

knes

s (µ

m)

Days of submergence

30

20

10

00 2 4 6 8 10 0 2 4 6 8 10

Fig. 1. Leaf gas film thickness of 14 cultivars of 16 days old wheat (Triticum aestivum) with timeof submergence (means� s.e., n = 4). At day 10, control plants in air had a gas film thickness inthe range of 55–142% of the initials at day 0.


108

contact angle showed a distinct pattern with many pointsclustered around contact angles of 150� when gas films werepresent (Fig. 2d). Contact angles above 150� defined the leafsurfaces as superhydrophobic and the plot indicates thatsuperhydrophobicity is necessary to have gas films thickerthan 20mm. Hydrophobic leaves (90� < q <150�) could retain a

thin gas film below 20mm, and hydrophilic leaves (q <90�) didgenerally not have gas films except a few points withG. fluitans.

Net photosynthesis (PN) under water

Underwater PN was determined at two different CO2

concentrations (200 and 2500mM). At 200mM CO2, PN is

40

50(a) (b)

(c)

(d )

200

150

100

50

0

30

20

10

100

80

60

40

20

0

00

Gas

film

thic

knes

s (µ

m)

Con

tact

ang

le (

°)

Days of submergence

Days of submergence Contact angle (°)

Days of submergence

Gas

film

thic

knes

s (µ

m)

Per

cent

of l

eaf e

xpos

ed

2 4 6 8 10

0 2 4 6 8 10 0 50 100 150 200

0 2 4 6 8 10

40

50

60

30

20

10

0

JensenAlbertJacksonG. fluitans

Fig. 2. (a) Leaf gasfilm thickness, (b) leaf surfacewettability determined as contact angle, (c) percent of leaf surface exposed i.e. not coveredby unidentified substance (Fig. 3) and (d) gas film thickness vs contact angle of three cultivars (Jensen, Albert and Jackson) of wheat (Triticumaestivum) andGlyceria fluitanswith time of submergence (means� s.e., n= 4). At day 10, control plants in air had a gas film thickness in therange of 56–89%, contact angle 97–100% and percent leaf exposed 99–100% of the initials at day 0.

Table 1. Results of two-way ANOVA (F-ratios and P-values) showing the effects of cultivar/species and time of submergence (0, 1, 2, 4, 7 or 10 days)on gas film thickness, leaf porosity, contact angle, chlorophyll a and photosynthesis (PN) of three wheat (Triticum aestivum) cultivars (Jensen, Albert

and Jackson) and Glyceria fluitansData are shown in Figs 2, 4 and 6.

Parameter Cultivar/species

Submergencetime

Cultivar/species�submergence time

Data inFig. number

F-ratio P-value F-ratio P-value F-ratio P-value

Gas film thickness 60.3 <0.0001 610.5 <0.0001 15.0 <0.0001 2aContact angle 11.7 <0.0001 290.1 <0.0001 8.5 <0.0001 2bPN at 200mM CO2 230.6 <0.0001 441.0 <0.0001 6.7 <0.0001 4aPN at 2500mM CO2 1.4 0.2389 242.2 <0.0001 13.0 <0.0001 4bLeaf porosity 62.0 <0.0001 3.7 0.0052 5.4 <0.0001 6aChlorophyll a 11.0 <0.0001 25.1 <0.0001 1.4 0.1883 6b


109

Day 1

Day 0

Day 2

Day 4

Day 10

Day 7

Jensen Albert Jackson G. fluitans

Day 0

5 µm

200 µm

Fig. 3. (Continued )


110

limited by CO2 because of diffusion resistance from the bulkmedium into the leaf and presence of a gas film is thereforebeneficial as it reduces the gas exchange resistance of submergedleaves (Pedersen et al. 2009;Verboven et al. 2014). The highCO2

concentration of 2500mM was used to assess changes in thecapacity for underwater PN with time of submergence as thisconcentration of CO2 was expected to saturate underwater PN

regardless of whether gas films were present or not. The 2-wayANOVA showed that there were significant effects of cultivar/species and time of submergence and also a significant interactionat 200mMCO2 (Fig. 4a; Table 1). The initial PN rates at 200mMCO2 were 3.7–4.3mmol O2 m

–2 s–1 for the three wheat cultivarsand 5.2mmol O2 m–2 s–1 for G. fluitans. With time ofsubmergence, the PN rates of the wheat cultivars declinedfaster than G. fluitans so that at day 4, the wheat cultivars hadrates in the range 0.3–0.5mmol O2 m

–2 s–1, whereas G. fluitanshad a mean rate of 1.9mmol O2 m

–2 s–1. However, at 2500mMCO2 there was no main effect of cultivar/species but a significanteffect of time of submergence and an interaction effect (Fig. 4b;Table 1). The initial PN rates 2500mM CO2 were in the range7.1–8.1mmol O2 m–2 s–1 with no difference between cultivar/species and all rates declined with time of submergence.However, at day 4, the PN rates of the three wheat cultivarshad declined to 2.3–2.9mmol O2 m

–2 s–1 and G. fluitans had asignificantly higher rate of 7.0mmol O2 m

–2 s–1.Correlation analyses were used to evaluate relationships

between underwater PN at the two different CO2 concentrationsand gas film thickness (Fig. 5). Only data points with ‘gas filmpresent’ (gas film thickness >2mm) were used to ensure that thecorrelations mainly assessed the effects of gas film thicknessand not potential damages to the photosynthetic apparatusoccurring after the gas films had been lost. There were positiverelationships between underwater PN and gas film thickness at200mMCO2for all cultivars/species (Fig.5a),whereasat2500mMCO2 therewere no significant correlations betweenPN andgasfilmthickness (Fig. 5b). This indicates that at the high CO2

concentration, the gas films were less important for the capacityfor underwater PN as the CO2 was not limiting and thereforethe beneficial effects of the gas films on gas exchange did notenhance PN.

Leaf porosity and chlorophyll

Initial leaf porosities were significantly higher inG. fluitanswith29% whereas the wheat cultivars had values of 15–18% porosity(Fig. 6a; Table 1). However, the porosity of G. fluitans declinedduring the submergence until day 7, where it was similar to thevalues of the wheat cultivars in the range of 16–18%. Leafconcentrations of chlorophyll a were initially 14–15mg g–1

and generally declined during submergence so at day 10, thevalues were 7–11mg g–1 (Fig. 6b; Table 1). The chlorophyll ofG. fluitans declined faster than the wheat cultivars but at day 10,

there were no significant differences between the cultivars/species.

Discussion

Gas film retention times during complete submergence werestudied in 14 wheat cultivars and compared with the wetlandplantG. fluitans. Previous studies have demonstrated that leaf gasfilms facilitate gas exchange with the floodwater by decreasingthe resistance across the cuticle, thereby contributing to wetlandand dryland crop submergence tolerance (Colmer and Pedersen2008; Pedersen et al. 2009; Pedersen and Colmer 2012). Leaf gasfilms provide immediate benefits upon submergence, whereassome terrestrial wetland plants acclimate to submergence byproducing new leaves which are thin and have reducedcuticles to decrease the resistance to CO2 and O2 diffusion

Fig. 3. Scanning electron microscopy (SEM)micrographs of adaxial leaf surfaces of three cultivars (Jensen, Albert and Jackson) of wheat (Triticum aestivum)and Glyceria fluitans with time of submergence (0, 1, 2, 4, 7 or 10 days). The scale bar is 5mm in all rows except the top row where the scale bar is 200mm.Larger papillae on the leaf surface of G. fluitans at the low magnification in the top row are indicated by white arrows. In addition, the leaf surface of G. fluitanswas also covered by smaller papillae, and the leaves had a plicate shape.

0

2

4

PN (

µmol

O2

m–2

s–1

)

6

8

10200 µM CO2

2500 µM CO2

12

(a)

(b)

0

0 2 4

Days of submergence

6 8 10

2

4

6

8

10

12

Jensen

Albert

Jackson

G. fluitans

Fig. 4. Underwater net photosynthesis (PN) measured at (a) 200 and (b)2500mM free CO2 of three cultivars (Jensen, Albert and Jackson) of wheat(Triticum aestivum) and Glyceria fluitans with time of submergence(means� s.e., n= 4). At day 10, control plants in air had PN in the rangeof 82–88% (at 200mMfree CO2) and 110–150% (at 2500mMfreeCO2) of theinitials at day 0.


111

(Mommer et al. 2005) making this strategy suitable for wetlandplants that can be exposed to prolonged inundation.

The present study shows that wheat retains leaf gas films forat least 2 days when submerged (Fig. 1). The 14 tested cultivarsall showed a similar pattern when gas films were lost, and we didnot find cultivars that had remarkably longer gas film retentiontimes than other cultivars. Although there were no obviousdifferences in gas film retention times between the cultivars,we selected three of the cultivars formore in-depth studies of theirphysiology when submerged. Interestingly, our results showedthat G. fluitans had thicker gas films and could retain its gasfilms significantly longer than the wheat cultivars (Fig. 2a). Therelationship between gas film thickness and hydrophobicitymeasured as contact angles showed that hydrophobicity was avery good proxy for whether the leaves had gas films or not(Fig. 2d).

Contact angle measurements are routinely used to characterisethewettability of plant surfaces (KochandBarthlott 2009).Ahigh

contact angle is related to surfaces on which a water dropletforms a spherical shape and the actual contact between the dropletand the surface is very small compared with wettable surfaces.Surfaces with contact angles of >150� are classified as beingextremely water-repellent or superhydrophobic (Koch andBarthlott 2009), so in the present study all wheat cultivars andG. fluitans possessed these properties at the beginning of thesubmergence event (Fig. 2b). Plant leaves are covered by aprotective layer called the cuticle consisting of cutin andintegrated and superimposed waxes which can result inhydrophobic properties. Plant waxes embedded into the cuticleare called intracuticular waxes and waxes superimposed on thesurface are called epicuticular waxes (Koch and Ensikat 2008).Epicuticular waxes are crystalline and can be seen on the surfaceas wax crystals with different shapes when investigatedwith SEM (Fig. 3). The classification of waxes is based onmicromorphological features according to Barthlott et al.(1998). Following this terminology, waxes are termed plateletswhen flat crystals are connected with their narrow side to thesurface like we observed in the present study in both wheat andG. fluitans (Fig. 3). This three-dimensional structure and thechemical composition of the wax crystals contribute to the

0

2

4

6

8

10

12

00 10

Gas film thickness (µm)

20 30 40 50 60

2

4

6

8

10

12

PN (

µmol

O2

m–2

s–1

)

(a)

(b)

200 µM CO2

2500 µM CO2

Jensen

Albert

Jackson

G. fluitans

Fig. 5. Underwater net photosynthesis (PN) measured at (a) 200 and (b)2500mM CO2 versus gas film thickness of three cultivars (Jensen, Albertand Jackson) of wheat (Triticum aestivum) and Glyceria fluitans. Thedetection limit of gas film thickness was ~2mm so only points with ‘gasfilm present’ i.e. gas film >2mm were plotted. Pearson correlation analyses(two-tailed) of underwater PN versus gas film thickness showed: For 200mMCO2: Jensen P= 0.0030 (r2 = 0.60); Albert P< 0.0001 (r2 = 0.77); JacksonP = 0.0113 (r2 = 0.49); G. fluitans P= 0.0199 (r2 = 0.35). For 2500mM CO2:Jensen P= 0.2072 (r2 = 0.17); Albert P= 0.6735 (r2 = 0.02); JacksonP = 0.0646 (r2 = 0.30); G. fluitans P= 0.6949 (r2 = 0.01).

40

30

20

10

0

20

10

Leaf

por

osity

(%

)C

hl a

(m

g g–1

)

15

5

00 2

Days of submergence

4 6 8 10

Jensen

Albert

Jackson

G. fluitans

(a)

(b)

Fig. 6. (a) Leaf porosity and (b) leaf chlorophyll a concentration of threecultivars (Jensen, Albert and Jackson) of wheat (Triticum aestivum) andGlyceria fluitans with time of submergence (means� s.e., n= 4). At day 10,control plants in air had leaf porosities and leaf chlorophyll a concentrationsin the range of 85–116% and 82–102%, respectively, of the initials at day 0.


112

hydrophobic nature of the cuticles (Koch et al. 2008). Besidesproviding the leaves of some plantswith a hydrophobic surface sothey can form gas films when submerged, the cuticle also playsother important roles for plants. These roles include minimisingwater loss by transpiration, protecting against UV-radiation andself-cleaning effects as a strategy to remove surface bacteria andspores of fungi (Koch and Ensikat 2008). In wheat, it has beendemonstrated by Stosch et al. (2007), that the superhydrophobicleaf cuticles function as protection against Blumeria graminis, afungus that causes powdery mildew. The study compared intactleaves with leaves that had been wiped to remove theirsuperhydrophobicity, and it showed that intact leaves had asignificantly better removal of conidia from their surfaces.This could explain why wheat as a dryland crop could benefitfrom superhydrophobic leaves.

The SEM micrographs showed that all three wheatcultivars and G. fluitans possessed high densities of plateletswhich explained their superhydrophobicity (Fig. 3). However,G. fluitans also had papillae that would contribute to higherhydrophobicity during the initial submergence, and this couldexplain why G. fluitans retained gas films for at least 4 days,whereas the wheat cultivars only maintained a gas film for atleast 2 days. Neinhuis and Barthlott (1997) investigated thesurface structures of over 200 water-repellent plant species,and concluded that most of them possessed hierarchicalsurface structures formed by convex papillose epidermal cellsand a very dense arrangement of three-dimensional epicuticularwaxes of different shapes.Wetting of such hierarchical structuresis minimised because air is trapped in cavities of the convex cellarrangements, and the hierarchical roughness enlarges the water-air interface while the solid-water interface is reduced (Bhushanand Jung 2008). The ability of superhydrophobic plant surfaceshas been studied in detail for lotus (Nelumbo nucifera), whereit has been reported that its persistent superhydrophobicity isattributed to the robustness of its leaf papillae in combinationwith their high density (Ensikat et al. 2011). Another remarkableplant in terms of hydrophobicity is Salvinia, which has beendemonstrated to retain a gas film up to 17 days under water(Cerman et al. 2009). This extreme hydrophobicity is causedby eggbeater-shaped trichomes that are up to 2mm long andsuperimposed by a layer of small hydrophobic wax crystals(Barthlott et al. 2010). In the present study, we showed thatG. fluitans possessed papillae which most likely contributed toits ability to retain a gas film for 4 days, and since this species isoften found in wetland habitats, it is possibly an adaption totemporary submergence.

We showed in this study that the loss of gas films is associatedwith the leaves being covered by an unidentified substance. Toour knowledge, this is the first time this phenomenon has beenreported, but in a similar submergence study on rice, the loss ofhydrophobicity was not associated with such depositions ofunidentified substances (Herzog et al. 2016a). Therefore, atthis point in time we can only speculate about what the causescould be. First of all, the question is whether the substance causedthe gas film to disappear, or the substance formed on the leafsurfaces after thegasfilmhadbeen lost. In another study,Brassicaoleracea, Eucalyptus gunnii and Tropaeolum majus withdifferent types of surface waxes, were cultivated at threedifferent relative air humidity levels (20–30, 40–75 and 98%)

(Koch et al. 2006). Plants grown at the highest humidity of 98%haddecreased totalwaxmass andwax crystal density and showedincreased leaf surface wettability, but the exact reason for thisdifference could not be determined. Koch et al. (2006) speculatedthat the differences in wax amounts at different air humidity werea consequence of different rates of wax transport from the siteof synthesis through the aqueous cell wall to the exterior of thecuticle. However, it is not known how this transport occursbut it has been suggested that a co-transport of wax moleculeswith water serving as a solvent could be responsible forthe movement of the wax molecules from the cells to thesurface of the cuticles (Neinhuis et al. 2001). Thus, at an airhumidity of 98% the water and wax diffusion through the cuticlewould be significantly reduced since the driving force is close tozero. In our study it is also possible that the submergence ofthe plants resulted in lower wax movement and regeneration,because the transpiration would be greatly reduced when theplants became completely submerged.

Epicuticular wax crystals are formed by a process called self-assembly, where organic solvents are excreted from the plantsand then crystallise on the cuticle surface. This self-assemblyresults in wax crystals but the shape and composition ofthese depend on several factors such as temperature, solventand substrate (Koch and Ensikat 2008). We speculate that theself-assembly of wax crystals does not function properly whenthe plants are submerged as the transport of wax componentsto the cuticle surface might be impaired and only some of thecomponents reach the surface. This could result in a layer ofunidentified substance on the cuticle instead of new wax crystalsand could cause the leaves to lose their hydrophobicity andsubsequently the gas films. An alternative explanation couldbe that the unidentified substance formed on leaves after thegas films had disappeared, and that the substance originates fromthe floodwater, as organic compounds or biofilm in the water. Toanswer these questions, further investigations have to beconducted.

Leaf gas films have been shown to increase underwater gasexchangeand thusCO2entry to sustain higher rates ofPN (ColmerandPedersen 2008; Pedersen et al. 2009). In the present study,PN

rates were measured at two CO2 concentrations. At 200mMCO2,underwater PN is limited by CO2 owing to the high resistance todiffusion from the bulk medium into the submerged leaf(Pedersen et al. 2009). Therefore, presence of gas films wouldbe beneficial at low CO2 concentrations, whereas PN ratesmeasured at 2500mM CO2 would not be limited by CO2.Correlation analyses showed that at 200mM CO2 there weresignificant positive relationships between gas film thickness andPN rates, whereas there were no correlations at 2500mM CO2

(Fig. 5). The declines with time in underwater PN at 200mMCO2

were probably due to loss of gas films as there was a strong linkbetween decreased gas film thickness and steep declines inunderwater PN during the initial 4 days of submergence forthe three wheat cultivars and also G. fluitans (Figs 2a and 4a).However, at day 4 the three wheat cultivars also showed a steepdecline at the ‘non-limiting’ CO2 concentrations indicatingthat the photosynthetic apparatus had been damaged (Fig. 4b).By contrast, G. fluitans still seemed to have well-functioningphotosynthesis at day 4. At 200mM CO2, G. fluitans hadsignificantly higher PN compared with the wheat cultivars at


113

all submergence durations indicating that it was generally lesslimited byCO2.Wenote thatG.fluitans had similarPN rates at thetwo CO2 concentrations at days 7 and 10, which showed that itwas not limited by CO2 even after the gas films had been lostsuggesting that it had developed a more permeable cuticle(Fig. 4), a phenomenon observed also in new aquatic leaves ofRumex palustris (Mommer et al. 2005).

We looked at leaf porosity and chlorophyll concentrationto investigate whether these factors could explain the declinesin PN rates during submergence (Fig. 6). Interestingly, thewheat cultivars did not show any sign of water infiltrationof the tissue which would be an indication of structuraldegradation, whereas G. fluitans showed declines in leafporosity until it reached a level of 16–20% corresponding to thewheat cultivars. Concentrations of chlorophyll did show somedeclines but only to around50%of the initial levels indicating thatother components of the photosynthetic apparatus had beencompromised. For comparison, Winkel et al. (2014) found thatporosity and chlorophyll concentrations of various rice cultivarsdeclined more drastically during a 13-day submergence periodexcept in the submergence tolerant cultivar FR13A. This ricestudy was conducted at 28�30�C whereas the present study wasconducted at 15�C which could probably explain why weobserved a slower degradation of leaf tissue and chlorophyll.Indeed, a submergence study on the wheat cultivar ‘JB Asano’performed at around 23�C showed a reduction in leaf chlorophylla by 90% after 14 days of submergence (Winkel et al. 2017)indicating a strong effect of temperature on tissue degradationduring submergence stress.

In summary, the present study found little diversity in the gasfilm retention of 14 different wheat cultivars, whereasG. fluitanshad thicker gas films and could retain its gas films significantlylonger than the wheat cultivars. The SEM micrographs showedthat thewheat cultivars andG. fluitans possessed high densities ofplatelets which could explain their superhydrophobicity.However, G. fluitans also had papillae that would contribute tohigher hydrophobicity during the initial submergence and couldexplain why G. fluitans retained gas films for 4 days, and thewheat cultivars only maintained gas films for 2 days. We haveshown that the loss of gasfilms is associatedwith the leaves beingcovered by an unidentified substance, and to identify the originof this substance, further investigations have to be conducted. Inconclusion, the leaf gas films provide benefits upon submergenceand as such is a relevant trait to use as a selection criterion toimprove the flood tolerance of both wetland and dryland cropsthat become temporarily submerged after sudden rainfall events.

Acknowledgements

The study was supported by the Villum Foundation via the grantWheatSUB.

References

Armstrong W (1979) Aeration in higher plants. Advances in BotanicalResearch 7, 225–332. doi:10.1016/S0065-2296(08)60089-0

Bailey-Serres J, Lee SC, Brinton E (2012) Waterproofing crops: effectiveflooding survival strategies. Plant Physiology 160, 1698–1709.doi:10.1104/pp.112.208173

BarthlottW, Neinhuis C, Cutler D, Ditsch F,Meusel I, Theisen I,Wilhelmi H(1998) Classification and terminology of plant epicuticular waxes.

Botanical Journal of the Linnean Society 126, 237–260. doi:10.1111/j.1095-8339.1998.tb02529.x

Barthlott W, Schimmel T, Wiersch S, Koch K, Brede M, Barczewski M,Walheim S, Weis A, Kaltenmaier A, Leder A (2010) The Salviniaparadox: superhydrophobic surfaces with hydrophilic pins for airretention under water. Advanced Materials 22, 2325–2328. doi:10.1002/adma.200904411

Bhushan B, Jung YC (2008) Wetting, adhesion and friction ofsuperhydrophobic and hydrophilic leaves and fabricated micro/nanopatterned surfaces. Journal of Physics Condensed Matter 20,225010. doi:10.1088/0953-8984/20/22/225010

Brewer CA, Smith WK (1997) Patterns of leaf surface wetness for montaneand subalpine plants. Plant, Cell & Environment 20, 1–11. doi:10.1046/j.1365-3040.1997.d01-15.x

Cerman Z, Striffler BF, Barthlott W (2009) Dry in the water: thesuperhydrophobic water fern Salvinia – a model for biomimeticsurfaces. In ‘Functional surfaces in biology’. (Ed. SN Gorb) pp. 97–111.(Springer: Berlin)

Colmer TD (2003) Long-distance transport of gases in plants: a perspectiveon internal aeration and radial oxygen loss from roots. Plant, Cell &Environment 26, 17–36. doi:10.1046/j.1365-3040.2003.00846.x

Colmer TD, Greenway H (2011) Ion transport in seminal and adventitiousroots of cereals during O2 deficiency. Journal of Experimental Botany62, 39–57. doi:10.1093/jxb/erq271

Colmer TD, Pedersen O (2008) Underwater photosynthesis and respirationin leaves of submerged wetland plants: gas films improve CO2 and O2

exchange. New Phytologist 177, 918–926. doi:10.1111/j.1469-8137.2007.02318.x

Colmer TD, Winkel A, Pedersen O (2011) A perspective on underwaterphotosynthesis in submerged terrestrial wetland plants. AoB Plants 2011,plr030. doi:10.1093/aobpla/plr030

Ensikat HJ, Ditsche-Kuru P, Neinhuis C, Barthlott W (2011)Superhydrophobicity in perfection: the outstanding properties of thelotus leaf. Beilstein Journal of Nanotechnology 2, 152–161. doi:10.3762/bjnano.2.19

Herzog M, Konnerup D, Pedersen O, Winkel A, Colmer TD (2016a) Leafgas films contribute to rice (Oryza sativa) submergence tolerance duringsaline floods. Plant, Cell & Environment doi:10.1111/pce.12873

Herzog M, Striker GG, Colmer TD, Pedersen O (2016b) Mechanismsof waterlogging tolerance in wheat – a review of root and shootphysiology. Plant, Cell & Environment 39, 1068–1086. doi:10.1111/pce.12676

Huang B, Johnson JW (1995) Root respiration and carbohydrate status oftwo wheat genotypes in response to hypoxia. Annals of Botany 75,427–432. doi:10.1006/anbo.1995.1041

Huang B, Johnson JW, Box JE, NeSmith DS (1997) Root characteristicsand hormone activity of wheat in response to hypoxia and ethylene. CropScience 37, 812–818. doi:10.2135/cropsci1997.0011183X003700030020x

Koch K, Barthlott W (2009) Superhydrophobic and superhydrophilicplant surfaces: an inspiration for biomimetic materials. PhilosophicalTransactions of the Royal Society of London A: Mathematical, Physicaland Engineering Sciences 367, 1487–1509. doi:10.1098/rsta.2009.0022

Koch K, Ensikat H-J (2008) The hydrophobic coatings of plant surfaces:epicuticular wax crystals and their morphologies, crystallinity andmolecular self-assembly. Micron (Oxford, England) 39, 759–772.doi:10.1016/j.micron.2007.11.010

Koch K, Hartmann KD, Schreiber L, Barthlott W, Neinhuis C (2006)Influences of air humidity during the cultivation of plants on waxchemical composition, morphology and leaf surface wettability.Environmental and Experimental Botany 56, 1–9. doi:10.1016/j.envexpbot.2004.09.013

Koch K, Bhushan B, Barthlott W (2008) Diversity of structure, morphologyand wetting of plant surfaces. Soft Matter 4, 1943–1963. doi:10.1039/b804854a


114

dx.doi.org/10.1016/S0065-2296(08)60089-0

dx.doi.org/10.1104/pp.112.208173

dx.doi.org/10.1111/j.1095-8339.1998.tb02529.x

dx.doi.org/10.1111/j.1095-8339.1998.tb02529.x

dx.doi.org/10.1002/adma.200904411

dx.doi.org/10.1002/adma.200904411

dx.doi.org/10.1088/0953-8984/20/22/225010

dx.doi.org/10.1046/j.1365-3040.1997.d01-15.x

dx.doi.org/10.1046/j.1365-3040.1997.d01-15.x

dx.doi.org/10.1046/j.1365-3040.2003.00846.x

dx.doi.org/10.1093/jxb/erq271

dx.doi.org/10.1111/j.1469-8137.2007.02318.x

dx.doi.org/10.1111/j.1469-8137.2007.02318.x

dx.doi.org/10.1093/aobpla/plr030

dx.doi.org/10.3762/bjnano.2.19

dx.doi.org/10.3762/bjnano.2.19

dx.doi.org/10.1111/pce.12873

dx.doi.org/10.1111/pce.12676

dx.doi.org/10.1111/pce.12676

dx.doi.org/10.1006/anbo.1995.1041

dx.doi.org/10.2135/cropsci1997.0011183X003700030020x

dx.doi.org/10.1098/rsta.2009.0022

dx.doi.org/10.1016/j.micron.2007.11.010

dx.doi.org/10.1016/j.envexpbot.2004.09.013

dx.doi.org/10.1016/j.envexpbot.2004.09.013

dx.doi.org/10.1039/b804854a

dx.doi.org/10.1039/b804854a

Konnerup D, Malik AI, Islam AKMR, Colmer TD (2017) Evaluation ofroot porosity and radial oxygen loss of disomic addition lines ofHordeummarinum in wheat. Functional Plant Biology 44, 400–409. doi:10.1071/FP16272

Mackereth FJH, Heron J, Talling JF (1978) Water analysis: some revisedmethods for limnologists. Freshwater Biological Association PublicationNo. 36. (Springer-Verlag: Berlin)

Mackinney G (1941) Absorption of light by chlorophyll solutions. TheJournal of Biological Chemistry 140, 315–322.

Malik AI, Islam AKMR, Colmer TD (2011) Transfer of the barrier to radialoxygen loss in roots of Hordeum marinum to wheat (Triticum aestivum):evaluation of fourH.marinum–wheat amphiploids.NewPhytologist 190,499–508. doi:10.1111/j.1469-8137.2010.03519.x

Mommer L, Visser EJW (2005) Underwater photosynthesis in floodedterrestrial plants: a matter of leaf plasticity. Annals of Botany 96,581–589. doi:10.1093/aob/mci212

Mommer L, Pons TL, Wolters-Arts M, Venema JH, Visser EJW (2005)Submergence-induced morphological, anatomical, and biochemicalresponses in a terrestrial species affect gas diffusion resistance andphotosynthetic performance. Plant Physiology 139, 497–508.doi:10.1104/pp.105.064725

Mony C, Mercier E, Bonis A, Bouzille J-B (2010) Reproductive strategiesmay explain plant tolerance to inundation: a mesocosm experimentusing six marsh species. Aquatic Botany 92, 99–104. doi:10.1016/j.aquabot.2009.10.011

Neinhuis C, Barthlott W (1997) Characterization and distribution of water-repellent, self-cleaning plant surfaces. Annals of Botany 79, 667–677.doi:10.1006/anbo.1997.0400

Neinhuis C, Koch K, Barthlott W (2001) Movement and regeneration ofepicuticular waxes through plant cuticles. Planta 213, 427–434.doi:10.1007/s004250100530

Parry ML (2007) ‘Climate change 2007-impacts, adaptation andvulnerability: Working Group II contribution to the fourth assessmentreport of the IPCC.’ (Cambridge University Press: Cambridge, UK)

Pedersen O, Colmer TD (2012) Physical gills prevent drowning of manywetland insects, spiders and plants. Journal of Experimental Biology215, 705–709. doi:10.1242/jeb.065128

Pedersen O, Rich SM, Colmer TD (2009) Surviving floods: leaf gas filmsimprove O2 and CO2 exchange, root aeration, and growth of completelysubmerged rice. The Plant Journal 58, 147–156. doi:10.1111/j.1365-313X.2008.03769.x

Pedersen O, Colmer TD, Sand-Jensen K (2013) Underwater photosynthesisof submerged plants - recent advances and methods. Frontiers in PlantScience 4, 140. doi:10.3389/fpls.2013.00140

Ponnamperuma FN (1984) Effects of flooding on soils. In ‘Floodingandplantgrowth’. (Ed.TKozlowski)pp.9–45. (AcademicPress:NewYork)

Raskin I (1983) A method for measuring leaf volume, density, thickness,and internal gas volume. HortScience 18, 698–699.

Raskin I,KendeH(1983)Howdoesdeepwater rice solve its aerationproblem.Plant Physiology 72, 447–454. doi:10.1104/pp.72.2.447

Sayre KD, Van Ginkel M, Rajaram S, Ortiz-Monasterio I (1994) Toleranceto waterlogging losses in spring bread wheat: effect of time of onset onexpression. Annual Wheat Newsletter 40, 165–171.

Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ:25 years of image analysis. Nature Methods 9, 671–675. doi:10.1038/nmeth.2089

Setter TL, Waters I (2003) Review of prospects for germplasm improvementfor waterlogging tolerance in wheat, barley and oats. Plant and Soil 253,1–34. doi:10.1023/A:1024573305997

Shabala S (2011) Physiological and cellular aspects of phytotoxicitytolerance in plants: the role of membrane transporters and implicationsfor crop breeding for waterlogging tolerance. New Phytologist 190,289–298. doi:10.1111/j.1469-8137.2010.03575.x

Smart RM, Barko JW (1985) Laboratory culture of submersed freshwatermacrophytes on natural sediments. Aquatic Botany 21, 251–263.doi:10.1016/0304-3770(85)90053-1

Stosch AK, Solga A, Steiner U, Oerke E-C, Barthlott W, Cermann Z (2007)Efficiency of self-cleaning properties in wheat (Triticum aestivum L.).Journal of Applied Botany and Food Quality 81, 49–55.

Thomson CJ, Armstrong W, Waters I, Greenway H (1990) Aerenchymaformation and associated oxygen movement in seminal and nodal rootsof wheat. Plant, Cell & Environment 13, 395–403. doi:10.1111/j.1365-3040.1990.tb02144.x

Trnka M, Rötter RP, Ruiz-Ramos M, Kersebaum KC, Olesen JE, Žalud Z,Semenov MA (2014) Adverse weather conditions for European wheatproduction will become more frequent with climate change. NatureClimate Change 4, 637–643. doi:10.1038/nclimate2242

Verboven P, Pedersen O, Ho QT, Nicolai BM, Colmer TD (2014)The mechanism of improved aeration due to gas films on leaves ofsubmerged rice. Plant, Cell & Environment 37, 2433–2452.

Winkel A, Colmer TD, Ismail AM, Pedersen O (2013) Internal aerationof paddy field rice (Oryza sativa) during complete submergenceimportance of light and floodwater O2. New Phytologist 197, 1193–1203.doi:10.1111/nph.12048

Winkel A, Pedersen O, Ella E, Ismail AM, Colmer TD (2014) Gas filmretention and underwater photosynthesis during field submergence offour contrasting rice genotypes. Journal of Experimental Botany 65,3225–3233. doi:10.1093/jxb/eru166

Winkel A, Herzog M, Konnerup D, Fløytrup AH, Pedersen O (2017)Flood tolerance of wheat – the importance of leaf gas films duringcomplete submergence. Functional Plant Biology 44, 000–000.


www.publish.csiro.au/journals/fpb

115

dx.doi.org/10.1071/FP16272

dx.doi.org/10.1071/FP16272

dx.doi.org/10.1111/j.1469-8137.2010.03519.x

dx.doi.org/10.1093/aob/mci212

dx.doi.org/10.1104/pp.105.064725

dx.doi.org/10.1016/j.aquabot.2009.10.011

dx.doi.org/10.1016/j.aquabot.2009.10.011

dx.doi.org/10.1006/anbo.1997.0400

dx.doi.org/10.1007/s004250100530

dx.doi.org/10.1242/jeb.065128

dx.doi.org/10.1111/j.1365-313X.2008.03769.x

dx.doi.org/10.1111/j.1365-313X.2008.03769.x

dx.doi.org/10.3389/fpls.2013.00140

dx.doi.org/10.1104/pp.72.2.447

dx.doi.org/10.1038/nmeth.2089

dx.doi.org/10.1038/nmeth.2089

dx.doi.org/10.1023/A:1024573305997

dx.doi.org/10.1111/j.1469-8137.2010.03575.x

dx.doi.org/10.1016/0304-3770(85)90053-1

dx.doi.org/10.1111/j.1365-3040.1990.tb02144.x

dx.doi.org/10.1111/j.1365-3040.1990.tb02144.x

dx.doi.org/10.1038/nclimate2242

dx.doi.org/10.1111/nph.12048

dx.doi.org/10.1093/jxb/eru166

Supporting Information to chapter 1, 2 and 3

116

Supporting Information chapter 1

Supporting Information to chapter 1

Table S1. Wheat growth from studies where wheat was grown in soil under waterlogged (water at soil surface unless otherwise stated) and drained conditions and recording both shoot and root dry mass (DM), allowing for root:shoot ratio calculation. Values are % reduction from controls [1-(waterlogged/drained)]. When multiple nutrient-levels were used we refer to the lowest. Wl=waterlogged, DAS=Days after sowing.

Variety Soil type and temperature

Treatment Reduction in shoot DM (%) relative to drained controls

Reduction in root DM (%) relative to drained controls

Root:shoot ratio Reference

Jaypee and USG3209

Sandy loam soil. 20/18 °C.

Wl for 28 d from 24 DAS in pots in greenhouse.

Jaypee: 28%. USG3209: 31%.

Jaypee: 37%. USG3209: 21%.

Controls: 0.16/0.13 (Jaypee/USG3209). Wl: 0.14/0.15 (Jaypee/USG3209).

Ballesteros et al. (2014)

Monad Silt loam soil mixed with sand. 22/10 °C.

Wl for 22 d from 84 DAS in pots in climate chamber.

14% 55% Controls: 0.7 Wl: 0.37

Hamonts et al. (2013)

Bobwhite SH 9826 Granular soil and volcanic top-soil. 20/15 °C.

Wl for 7 d 3 cm above soil level from 5 DAS in pots in greenhouse.

12% 40% Controls: 0.54 Wl: 0.37

Haque et al. (2011)

Bobwhite, Norin 61, Shiroganekomugi, Chikugoizumi, Minaminokomugi, Kinuhime

Granular soil and volcanic top-soil. 17-23°C.

Wl for 7 d 3 cm above soil level from 5 DAS in pots in greenhouse.

-4-18% (variety range). 7-40% (variety range). - Haque et al. (2012)

Al, Mn and Fe tolerant and intolerant varieties (names not stated)

4 strongly acidic soils. 20/15 °C.

Wl for 42 d from 21 DAS in pots in phytotron.

9-22% (variety range). 45-84% (variety range). - Khabaz-Saberi et al. (2012)

Cascades Top of sandy surfaced duplex soil. 20/15 °C.

Wl for 14 d from 21 DAS in pots in growth chamber.

39% 75% Controls: 0.4 Wl: 0.19

Malik et al. (2001)

Cascades Top soil of sandy surfaced duplex soil. 20/15 °C.

Wl for 28 d 21 DAS in pots in growth chamber.1

44% after 14d wl. 72% after 28 d wl.

64% after 14 d wl. 86% after 28 d wl.

Controls: 0.33 Wl 14d: 0.22

Malik et al. (2002)

Cascades

Top of sandy surfaced duplex soil. 20/15 °C.

Wl for 21 d from 21 DAS in pots in growth chamber.

38% 74% Controls: 0.58 Wl: 0.25

Malik et al. (2003)

Chinese spring Sandy duplex top soil. 20/15 °C.

Wl for 42 d 11 DAS in pots in phytotron.

78% 91% Controls: 0.4 Wl: 0.17

Malik et al. (2011)

117


Chinese spring Fine grey sand topsoil. 20/15 °C.

Wl for 21 d from 11 DAS in pots in phytotron.

84% 89%

Controls: 0.4 Wl: 0.21

McDonald et al. (2001b)

Wyalkatchem Yellow sand:sandy duplex top soil mixture. 20/10 °C.

Wl for 14 d from 22 DAS in pots in greenhouse.

22%

68%


Robertson et al. (2009)

Cappelle Deprez Sandy soil collected from field. 14 °C.

Wl for 14 d 11 DAS in cylinders. 54% 81% Controls: 0.64 Wl: 0.25

Trought and Drew (1980a)


Wl for 14 d 11 DAS in cylinders. 40%

72%


Trought and Drew (1980b)

1 Authors measured structural dry mass rather than dry mass; not included in Figure 1.

118


Table S2. Grain yield reductions from studies where wheat was grown in soil under waterlogged (water at soil surface unless otherwise stated) and drained conditions. Values are% reduction from controls [1-(waterlogged/drained)]. When multiple nutrient-levels were used we refer to the lowest. WL=waterlogged, DAS=Days after sowing.

Variety Soil type and temperature Treatment Yield reduction (% reduction from controls)

Reference

Non declared Sandy loam. Field conditions, mean soil temperature 4-11 °C.

WL for 20-120 d at seedling, tillering and/or stem elongation stage in cylinders in the field.

7-19% depending on growth stage and duration.

Belford (1981)

Non declared Calcerous clay soil. Field conditions, mean soil temperature 3-11 °C.

WL for 4, 21 or 42 d at pre-emergence, tillering or stem elongation stage in lysimeters.

4% at pre-emergence, 12% at tillering and 6% at stem elongation (n. s.).

Belford et al. (1985)

Non declared Sandy loam and clay soil. Field conditions mean soil temperature 4-12 °C.

WL for 16, 42 or 6 d early winter, mid-winter or spring in lysimeters in the field.

0% (light yields) to 15% (heavy yields).

Cannell et al. (1980)

Pioneer 2643, Pioneer 2691, LA 87167, Savannah, Terral LA 422, Coker 9663, Florida 304, FFR502W, and Jaypee.

Silt loam. Field conditions. WL for 35 d from 3-4 leaf stage in cone-tainers in the field under rain-shelter.

15-60% in 9 varieties (44% mean).

Collaku and Harrison (2002)

Claire, Deben, Xi-19 Fertile alluvial loam. Field conditions, mean air temperature 3-11 °C.

WL for 44-58 d from 64-93 DAS in lysimeters in the field in two consecutive cropping seasons.

20 and 24% in two consecutive years.

Dickin and Wright (2008)

Claire, Deben, Hereward, Riband and Xi-19. Fertile alluvial loam. Field conditions, mean air temperature 3-10 °C.

WL for 77-86 d from 61-78 DAS in the field. Water level 10 cm below soil surface.

9% (mean of two years and 5 varieties).

Dickin et al. (2009)

Coker 916, Coker 9733, Coker 9877, Florida 302, McNair 1003, Pioneer 2551, Buckshot's DS 2368, Coker 9105, Florida 303, Savannah, Terral 101, and Terral 877, LA8564 A8O-3-1-X, LA861 A18-4-1-X.

Silty clay. Field conditions. WL for entire growth period from 28 DAS in pots in open greenhouse.

28, 43 and 49% in three consecutive years (variety mean).

Ding and Musgrave (1995)

Peck, UJO77296, Hill 81, Gallahad, Birch 75, Lawson, Birch 41, Braemar Velvet, M4195, Quarrion, Kellalac, Matong, Meering.

Loam over clay duplex soil. Field conditions.

Intermittent WL for 30 d or more from mid- to late-tillering in the field.

-3-74% (variety range). Gardner and Flood (1993), summarized by Setter and Waters (2003)

Nishikazekomugi, Iwainodaichi, Shiroganekomugi, Norin 61, Uniculm.

Andosoil. Field conditions, 3-23 °C during 2005-2010 growing seasons.

WL for up to 49 d from 120 DAS (jointing stage till maturity) in the field. Water level 0-5 cm below soil surface.

96-27% (variety and year range, see table 2 in reference).

Hayashi et al. (2013)

Al, Mn and Fe tolerant and intolerant varieties, only names of reference varieties stated.

4 strongly acidic soils. 20/15 °C. WL for 42 d from 21 DAS in pots in phytotron.

30%/50% Al-T/Al-I 21%/41% Mn-T/Mn-I 16%/47% Fe-T/Fe-I

Khabaz-Saberi et al. (2012)

Yangmai 9 Clay soil. Field conditions. WL for 7 d from 7 DAA in pots in the 22% Li et al. (2011)

119


field. Water level 1-2 cm above soil surface.

Yecora 70 Silt loam. Soil temperatures at 5, 15 and 25 °C.

WL for 30d from 72 DAS in pots in water bath. Flooded to a depth of 5 cm.

21% at 5 °C 23% at 15 °C 73% at 25 °C

Luxmoore et al. (1973)

Yecora 70 Clay loam. Field conditions, mean soil temperature 17 °C.

WL for 10-20 d from 101 DAS (anthesis) in the field. Flooded to a depth of 5 cm.

10 d: 1-2 % 20d: 16%

Luxmoore et al. (1973)

Oxley One well drained and one waterlogging-prone duplex soil. Field conditions, 9-16 °C.

WL occurred on the poorly drained duplex soil by excess rainfall. Crops sown at three different dates.

58% reduction in well drained to poorly drained soil (mean of three sowing dates).

McDonald and Gardner (1987)

Coker 9877 Silty clay. Field conditions, mean soil temperature 25 °C.

WL occurred on poorly drained plots. 51%

Musgrave (1994)

Pioneer 2551, Florida 302, McNair 1003, Rosen, Coker 9877, Coker 916, Coker 9733, Coker 9323.

Fertilized river silt. Field conditions, mean soil temperature 17-25 °C in three years.

WL for entire growth period from 14 DAS in pots in unheated greenhouse.

37-45% (variety range). Musgrave (1994)

Florida 303, Terral 877, Coker 9105, McNair 1003, Coker 9877, LA8564 A80-3-1-X.

Silty clay. Field conditions. WL for 120 d from stand establishment in the field under rain shelter.

9-69% in 6 varieties (45% mean).

Musgrave and Ding (1998)

Florida 303, Terral 877, Coker 9105, Buckshots DS 2368, LA8564 A80-3-1-X, LA 861 A18-4-1-X, Savannah, Terral 101.

River silt in pots. Field conditions. WL for entire growth period from 14 DAS in pots in greenhouse.

43% and 28% in two consecutive years.

Musgrave and Ding (1998)

Karasu-90 Loamy textured soil. Field conditions. WL for 5, 10, 15, 20, 25 or 50 d from flowering in PVC containers in the field.

4, 21, 38, 49, 74 and 90% (5, 10, 15, 20, 25 and 50 d).

Olgun et al. (2008)


WL for 14 d from 22 DAS in glasshouse.

32% Robertson et al. (2009)

Ducula-1, Ducula-2, Ducula-3, Ducula-4, Seri 82, Pato Blanco, BR34, PF8442, Mikn Yang#11, Zhen 7843, WR89-3420, WR89-3246, 46 WR Norin, Tinamou, Vee/Myna, Prl/Sara

Coarse sandy clay. Field conditions. WL for 48 d (15 d after emergence to mid boot) or 28 d (1st node stage, mid boot, anthesis, grain filling) in the field. Water level 3-8 cm above soil surface.

68, 53, 70, 59, 26% at 5 different growth stages (mean of 16 varieties). Min. reduction 7%, max. 96%.

Sayre et al. (1994)

HD 2009 Poorly permeable sodic soil. Field conditions.

WL for 2, 4 and 6 d from 25 DAS in the field.

18, 29 and 47% (2, 4 and 6 d).

Sharma and Swarup (1988)

Arina Unspecified field soil. Field conditions. WL from anthesis through maturity in pots embedded in the field.

39%. Stieger and Feller (1994)

Norin 6, Siroganegomugi Field soil. Field conditions, soil temperature 7-22 °C.

WL for 15 d at tillering, elongation, booting or ripening stages in the field.

9% (mean of different growth stages, range 7-10%).

Suh (1978)

120


Yangmai 9, Yumai 34 Clay soil. Field conditions. WL from 7 DAA for 15 d in pots. Water level 1-2 cm above soil surface.

32% and 34% (variety 1 and 2).

Tan et al. (2008)

Non declared Top soil collected from upper, middle and lower valley slope (clay content increasing). Field conditions.

WL for 42 d from tillering stage. 57% (average of 3 soil types).

Watson et al. (1976)

Non declared Sandy loam and clay. Field conditions. WL in January for 21-42 d and in one year also 21 d in May, in lysimeters in three cropping seasons.

11 % (sand, 21 d, year 1). -3% (clay, 42 d, year 2). 17% (clay, 63 d, year 3).

Webster et al. (1986)

HD-2009, KRL-3-4 Two sodic soils (pH 8.5 and 9.2). Field conditions, mean daily temperature 17 °C.

WL for 14 d from 21 DAS in the field. Water level 5 cm above soil surface.

HD-2009: 16%/34% (pH 8.5/pH 9.2). KRL-3-4: 6%/14% (pH 8.5/pH 9.2).

Yaduvanshi et al. (2012)

121


Table S3. Summary of average root porosity (POR) or aerenchyma (AER) in wheat roots of different varieties, plant age, growing media to induce hypoxia, hypoxia/waterlogging duration, type of root (seminal and/or adventitious-nodal).

Variety Plant age (days) Growing media for inducing hypoxia

Treatment duration (days)

Root type Root porosity (POR) or aerenchyma (AER) under aerated conditions

Root porosity (POR) or aerenchyma (AER) under hypoxic/waterlogged conditions

Reference

Chinese Spring 11 d Stagnant nutrient solution

27-29 d Adventitious 6% POR (segments of 100mm from tip)

19% POR (segments of 100mm from tip)

Alamri et al. (2013)

Gamenya 16 d N2 flushed nutrient solution

10 d Seminal Seminal Adventitious Adventitious

Not stated Not stated Not stated Not stated

POR 6% at 0-50 mm from root tip POR 3% in proximal segment POR 9% at 0-50 mm from root tip POR 14% in proximal segments

Barrett-Lennard et al. (1988)

Jackson Coker 9835

14 d 14 d

Hypoxic nutrient solution (1 kPa O2)

21 d Adventitious Adventitious

7.5% POR ( segments of root axes) 7.0% POR ( segments of root axes)

21.5% POR (segments of root axes) 10.7% POR ( segments of root axes )

Huang and Johnson (1995)

Bayles Savannah

14 d 14 d

Waterlogged sand with influx of ½ strength Hoagland’s nutrient solution

15 d 15 d

Seminal Adventitious Seminal Adventitious

Aerenchyma absent in root cortex Aerenchyma absent in root cortex Aerenchyma absent in root cortex Aerenchyma absent in root cortex

12%1 AER starting at 50mm from tip 19%1AER at 10mm of tip 12% - 28%1 AER at 10mm and 300mm from tip 30%1 AER at 10mm of tip

Huang et al. (1994)

Cascades 21 d Waterlogged sandy soil

14 d 28 d2

Adventitious Adventitious

4% to 8% POR along the root 2% to 4% POR along the root

12% to 13% POR along the root 20% to 12% POR along the root

Malik et al. (2001)

Cascades 21 d Waterlogged sandy soil

21 d Adventitious 3%-5% POR along the root 16%-20% POR along the root Malik et al. (2003)


21 d Adventitious 3% POR (entire root) 15.9% POR (entire root) Malik et al. (2011)


21 d Seminal Adventitious

2.1 POR (root segments 40-50mm) 5.7 POR (root segments 40-50mm)

2.7 POR (root segments 40-50mm) 21.6 POR (root segments 40-50mm)

McDonald et al. (2001a)

122


1 Values correspond to aerenchyma proportion of the cortical cross sectional area (stele area was excluded), not plotted in Figure 2. 2 Plotted in Figure 3. 3 Nutrient solution with 20 μM 1-aminocyclopropanecarboxylic acid (ACC)– not plotted in Figure 2.


21 d Adventitious Adventitious Seminal Adventitious

Not stated Not stated POR 2% POR 3.35%

AER 32% 10mm below root/shoot AER 37.3% 50 mm behind root tip POR 1.65% POR 17.66%

McDonald et al. (2001b)

Gamenya

25 d Hypoxic nutrient solution (0.22 kPa O2)


0.8% POR (root segments until 80mm from tip) 3.4% POR (root segments until 80mm from tip)

5.3% POR (root segments until 80mm from tip) 12.6% POR (root segments until 80mm from tip)

Thomson et al. (1992)

Gamenya 5-7 d 18-23 d 18-23 d 23-28 d

Stagnant nutrient solution

4-6 d 5,7-10 d 5,7-10 d 5 d

Seminal Seminal Adventitious Adventitious

3.4% POR (entire root) 4.8% POR (entire root) 4.5% POR (segment 0-50 mm from tip) and 6.8% POR (>50 mm) 6.3% to 8.2% POR (entire root)

12% POR (entire root) 6% POR (entire root) 10.6% POR (segment 0-50 mm from tip) and 15.3% POR (>50 mm) 6% to 9% POR (entire root)

Thomson et al. 1990

Gamenya Kite

14 d 14 d

Stagnant nutrient solution

14 d Adventitious Adventitious

Not informed Not informed

11.6% AER at 50 mm from tip 13.7% AER at 50 mm from tip

Watkin et al. (1998)

Gamenya 16 d Stagnant nutrient solution


POR 1.3% POR 5.2%

POR 3% POR 14.8%, AER 22.3%

Wiengweera et al. (1997)

Huamai 8 5 d Roots submerged in distilled water

3 d Seminal Aerenchyma absent in root cortex. 4% AER at 10 mm from tip and 17-18% AER at 50 mm from tip

Xu et al. (2013)

Bobwhite line SH98 26 9 d Stagnant nutrient solution

7 d Adventitious 0% AER at 10 mm from tip and 11%2 AER at 50 mm from tip.

14% AER at 10 mm from tip and 20%2 AER at 50 mm from tip.

Yamauchi et al. (2014a)

Bobwhite line SH98 26 7 d Stagnant nutrient solution3

7 d

Seminal Aerenchyma absent in root cortex at all positions examined from tip in the 1st, 2nd and 3rd seminal root.

3% AER at 10mm from tip and 4% AER for the 1st seminal root. ca. 2% AER at 10mm and 30mm from tip for the 2nd and 3rd seminal root.

Yamauchi et al. (2014b)

123


Table S4. Reductions in photosynthetic rates calculated from studies where wheat was grown in soil under waterlogged (water at soil surface unless otherwise stated) and drained conditions. Values are % reduction from controls [1-(waterlogged/drained)]. All nutrient treatments were included. WL=waterlogged, DAS=Days after sowing, DAA=days after anthesis.

Genotype Soil type and temperature Treatment Leaf used for measurements PN reduction (% reduction from control levels)

Reference

Monad Silt loam soil mixed with sand. 22/10 °C.

WL for 22 d from 84 DAS in pots in climate chamber.

Flag leaf 22 d after WL, 1400 µmol photons m-2 s-1.

9% with additional N supply. 8% without N supply.

Hamonts et al. (2013)

Nishikazekomugi, Iwainodaichi

Andosol. Field conditions, monthly mean air temperature 4-23 °C in three consecutive cropping seasons.

WL for 31-34 d from jointing stage to maturity in the field. Water level 0-5 cm below soil surface.

Flag leaf, 1400 µmol photons m-2 s-1

2008: 0%. 2009: No change for 22d, then 23% and 51%. 2010: Close to controls for 11-16 d, then 62 and 71 %.


Uniculm As above. As above. Flag leaf, 1400 µmol photons m-2 s-1

2008: No change for 8d, then 48% (22d). 2009: 70%. 2010: 63%.


Haruyutaka Clay loam. Field condition, temperature range 0-28 °C during cropping season.

WL for 14 d before anthesis (2008) or throughout grain filling (2009).

Flag leaves around anthesis, mid-day at sunny weather.

Not within 12 d WL, then 85% after 25 d of WL.

Hossain et al. (2011)

Daichinominori As above. As above. As above. Not within 12 d WL, then 65% after 17 d of WL.

Hossain et al. (2011)

Savannah Sterilized fine sand. 20/15 °C. WL for 17 d from 14 DAS with half (HS) or full strength (FS) Hoagland solution in pots in growth chamber. Water level 2 cm above soil surface.

Youngest fully expanded leaf, mid-day.

FS, HS: Drop by 13% within first 3 d, then stable.

Huang et al. (1994)

Bayles As above. Same. Same. HS: Drop by 24% within first 3 d, then stable. FS: PN 17% lower than HS.

Huang et al. (1994)

Yangmai 9 Clay soil. Field conditions. WL for 7 d from 7 DAA in pots in the field. Water level 1-2 cm above soil surface.

Flag leaf, 1400 µmol photons m-2 s-1.

51% Li et al. (2011)

Cascades Top soil of sandy surface duplex soil. 20/15 °C.

WL for 14 d from 21 DAS in pots in growth chamber.

Youngest fully expanded leaf, 2000 µmol m–2 s–1.

25% decrease within 24 hours, 100% decrease after 5 d in 3rd leaf (84% in 4th leaf).

Malik et al. (2001)

Pioneer 2551, Florida Fertilized river silt. Mean soil WL for entire growth period Flag leaf at saturating light. 32% drop in 1992, n.s. in 1990. Musgrave (1994)

124


302, McNair 1003, Rosen, Coker 9877, Coker 916, Coker 9733, Coker 9323.

temperature 17-25 °C in three cropping seasons.

from 14 DAS in pots in unheated greenhouse.

Treatment x variety interaction n.s. in both years.

Yangmei 14 Clay soil. Field conditions, mean temperature 15.3 °C.

WL for 1-3 d at tillering, milky and booting stages in lysimeters in the field. Water level 0.5-1 cm above soil surface.

Top fully expanded leaves, 1000-1100 µmol photons m-2 s-1

12-13% reduction at tillering, n.s. at booting and milky stages after 1 day of WL.

Shao et al. (2013)

Cappelle Deprez Sandy soil collected from field. 14 °C. WL for 14 d from 11 DAS in cylinders.

Not stated. 11% after 8 d, 35% between d 8-15.

Trought and Drew (1980a)

Yannong 19 Alluvial top. Field conditions. WL for 6 d from 3 DAA in pots in the field. Water level 0.5-1 cm above soil surface.


15%. 7% with foliar N spray.

Wu et al. (2014)

Huamai 17 Clay soil. Field conditions. WL for 5 d from 7 DAA in pots in the field. Water level 1-2 cm above soil surface.


14% after 3 d WL (n.s.). Zheng et al. (2009)

Yangmai 12 As above. As above. Flag leaf, 1200 µmol photons m-2 s-1.

11% after 3 d WL (n.s.). Zheng et al. (2009)

125


Table S5. Summary of values/range for photosynthesis (PN) and stomatal conductance (gs) in wheat under control and waterlogging conditions used for calculations in Figure 6. Information on waterlogging (WL) duration, variety, PN, gs, phenological stage during measurements and source are provided. The numbers in brackets identify the bibliographic source of each point in Figure 4.

WL duration Variety

PN (µmol m-2s-1) Value or range

gs (mol m-2s-1) Value or range

Phenological stage during

measurements Reference

Control Waterlogging Control Waterlogging Measurements at early waterlogging ( 1 to 3 days)

1d Yangmai 14 24.5 – 25.20 22.63 – 20.33 0.38 – 0.45 0.26 – 0.32 Tillering Shao et al. (2013) [3] 1d Yangmai 14 21.23 – 21.73 17.03 – 20.70 0.32 – 0.35 0.21– 0.28 Jointing-booting Shao et al. (2013) [3] 1d Yangmai 14 20.73 – 23.77 18.55 – 21.20 0.29 – 0.46 0.22 – 0.46 Onset of flowering Shao et al. (2013) [3] 1d Yangmai 14 14.88 – 16.23 13.20 – 13.98 0.25 – 0.26 0.22 – 0.25 Milky stage Shao et al. (2013) [3] 1d Yangmai 9 18.27 17.70 0.27 0.26 Around anthesis Li et al. (2011) [8] 1d Cascades 23.34 17.73 0.38 0.24 Tillering Malik et al. (2001) [6] 1d Bayles 20.27 – 21.76 24.08 – 24.82 0.87 – 0.88 0.97 – 1.03 Tillering Huang et al. (1994) [7] 1d Savannah 20.81 – 21.46 22.99 – 23.63 0.86 – 0.87 1.0 – 1.1 Tillering Huang et al. (1994) [7] 2d Cascades 23.06 9.38 0.521 0.136 Tillering Malik et al. (2001) [6] 2d Yangmai 9 15.13 14.03 0.242 0.216 Around anthesis Li et al. (2011) [8] 3d Huaimai 17 17.5 15.0 0.20 0.203 Anthesis Zheng et al. (2009) [5] 3d Yangmai 12 19.0 17.0 0.24 0.21 Anthesis Zheng et al. (2009) [5] 3d Cascades 20.51 5.56 0.315 0.077 Tillering Malik et al. (2001) [6] 3d Bayles 20.9 – 22.74 19.45 – 21.86 0.775 – 0.827 0.873 – 0.935 Tillering Huang et al. (1994) [7] 3d Savannah 20.09 – 22.33 19.18 – 20.51 0.93 – 1.0 0.863 – 0.913 Tillering Huang et al. (1994) [7] 3d Nishikazekomugi (2010) 20.74 19.7 0.303 0.298 Jointing-booting Hayashi et al. (2013) [9] 3d Uniculm (2010) 20.70 20.3 0.304 0.294 Jointing-booting Hayashi et al. (2013) [9] 3d Iwainodaichi (2010) 19.3 19.5 0.305 0.305 Jointing-booting Hayashi et al. (2013) [9]

Measurements at late waterlogging ( 7 to 21 days) 7d Yannong 19 13.22 – 14.17 11.25 – 13.02 0.268 – 0.282 0.244 – 0.265 Post-anthesis Wu et al. (2014) [2] 7d Monad 23.5 18.0 – 21.3 0.29 0.16 – 0.18 Non declared Hamonts et al. (2012) [4] 7d Yangmai 9 15.05 7.68 0.219 0.086 Around anthesis Li et al. (2011) [8] 7d Bayles 18.08 – 18.48 13.34 – 16.96 0.71 – 0.78 0.55 – 0.68 Tillering Huang et al. (1994) [7] 7d Savannah 16.5 – 16.68 13.78 – 15.77 0.814 – 0.879 0.66 – 0.71 Tillering Huang et al. (1994) [7]

126


8d Cascades 19.71 2.39 0.416 0.076 Tillering Malik et al. (2001) [6] 10d Cascades 18.96 3.7 0.393 0.096 Tillering Malik et al. (2001) [6] 10d Nishikazekomugi (2009/10) 16.8 – 20.89 15.81 – 19.7 0.278 – 0.292 0.281 – 0.284 Jointing-booting Hayashi et al. (2013) [9] 10d Uniculm (2011) 21.15 19.1 0.308 0.279 Jointing-booting Hayashi et al. (2013) [9] 10d Iwainodaichi (2010) 19.5 19.1 0.926 0.926 Jointing-booting Hayashi et al. (2013) [9] 11d Bayles 16.57 – 17.65 14.75 – 16.16 0.82 – 0.90 0.66 – 0.72 Tillering Huang et al. (1994) [7] 11d Savannah 17.04 – 18.56 13.98 – 16.15 0.88 – 0.94 0.44 – 0.56 Tillering Huang et al. (1994) [7] 12d Cascades 19.39 3.04 0.461 0.094 Tillering Malik et al. (2001) [6] 13d Bayles 16.96 – 17.28 11.81 – 15.27 0.665 – 0.743 0.356 – 0.423 Tillering Huang et al. (1994) [7] 13d Savannah 15.73 12.67 – 13.83 0.766 – 0.836 0.68 – 0.70 Tillering Huang et al. (1994) [7] 13d Nishikazekomugi (2009) 16.6 13.9 0.28 0.264 Jointing-booting Hayashi et al. (2013) [9] 13d Uniculm (2009) 22.3 21.7 0.29 0.274 Jointing-booting Hayashi et al. (2013) [9]

14d Pioneer 2551, Coker 916, Coker 9223, Florida 302, McNair 1003, Coker 9733, Coker 9877, Rosen

9.04 – 10.36 6.17 – 9.3 0.135 – 0.157 0.105 – 0.135 Estimatively near anthesis Musgrave (1994) [1]

15d Bayles 17.12 – 18.08 12.21 – 13.1 0.741 – 0.777 0.298 – 0.386 Tillering Huang et al. (1994) [7] 15d Savannah 15.48 – 17.79 12.74 – 14.15 0.712 – 0.802 0.490 – 0.632 Tillering Huang et al. (1994) [7] 15d Nishikazekomugi (2010) 21.19 19.25 0.300 0.298 Jointing-booting Hayashi et al. (2013) [9] 15d Uniculm (2010) 20.98 16.16 0.299 0.247 Jointing-booting Hayashi et al. (2013) [9] 15d Iwainodaichi (2010) 20.4 17.7 0.269 0.282 Jointing-booting Hayashi et al. (2013) [9] 18d Nishikazekomugi (2010) 21.94 18.2 0.298 0.224 Jointing-booting Hayashi et al. (2013) [9] 18d Uniculm (2009/10) 20.82 – 21.12 10.15 – 19.53 0.286– 0.296 0.128 – 0.273 Jointing-booting Hayashi et al. (2013) [9] 18d Iwainodaichi (2010) 21.1 12.5 0.308 0.229 Jointing-booting Hayashi et al. (2013) [9] 21d Uniculm (2009) 11.61 3.84 0.0121 0.078 Jointing-booting Hayashi et al. (2013) [9] 21d Iwainodaichi (2009) 16.3 16.15 0.272 0.147 Jointing-booting Hayashi et al. (2013) [9] 21d Nishikazekomugi (2009) 19.8 18.45 0.31 0.28 Jointing-booting Hayashi et al. (2013) [9]

127


Table S6. Shoot nutrient levels of macronutrients N, P, K, Mg, Ca, (mg g-1 DM) and microelements Mn, Fe, Zn, Al (mg Kg-1 DM) after soil waterlogging.

Variety Soil type (if stated) and temperature (day/night or mean).

Treatment Nutrient levels in entire shoot or leaves after waterlogging (macronutrients: mg g-1 DM, microelements: mg Kg-1 DM)

Reference

Cappelle Deprez Sandy loam and clay soil. Field conditions, mean soil temperature 4 °C.

Wl for 42 d (midwinter wl) 90 DAS sowing in the field. Whole shoot sampled.

N: 35/30 (sand/clay). K: 21/16 (sand/clay).

Cannell et al. (1980)

Claire, Deben, Hereward, and Xi-19 Fertile alluvial loam. Field conditions, mean air temperature 2.5-9.7 °C.

Wl in the field for 77 and 86 d from 78 and 61 DAS. Whole shoot sampled.

N: 32, 34, 33, 33 (4 varieties). Dickin et al. (2009)

Tamarin-Rock, Sunco, Krichauf, Sunvale, Qal2000

Three strongly acidic soils (pHCaCl=4.1-4.2). 20/15 °C.

Wl for 40 d from 21 DAS in pots in glasshouse. Whole shoot sampled.

Large data set – see table 3 in reference. Khabaz-Saberi and Rengel (2010)

Brookton, Cascades, Chara, Ducula-4, HD2329, Savannah

Strongly acidic soil (pHCaCl=4.5), potting mix or neutral clay soil. 14 °C.

Wl in pots for 49 d from 21 DAS in growth cabinet. Whole shoot sampled.

N: 11-20 (soil type range, separate values plotted). P: 2-3 (soil type range, separate values plotted). K: 19-121 (soil type range, separate values plotted). Mg: 0.9-1.8 (soil type range, separate values plotted). Ca: 0.5-5 (soil type range, separate values plotted). Mn: 58-445 (soil type range, separate values plotted). Fe: 43-390 (soil type range, separate values plotted). Zn: 11-67 (soil type range, separate values plotted).

Khabaz-Saberi et al. (2005)

Mn, Fe and Al tolerant (T) and intolerant (I) lines, only names of reference varieties stated.

4 strongly acidic soils (pHCaCl=4.2-4.5). 20/15 °C.

Wl for 42 d from 21 DAS in in pots glasshouse. Whole shoot sampled.

Large data set – see table 3 in reference. Khabaz-Saberi et al. (2012)


Wl for 14 d from 21 DAS in pots in growth chamber. Leaves sampled.

N: 35 Malik et al. (2001)


Wl for 28 d 21 DAS in pots in growth chamber. Green leaves taken from tillers sampled.

N: 29 Malik et al. (2002)


Wl for 14 d from 22 DAS (tillering) in glasshouse. Whole shoot sampled.

N: 15 Robertson et al. (2009)

Camm, Cascades, Westonia Three strongly acidic soils Wl in pots for 49 d from 21 DAS in the field. Mn: 14-146 Setter et al. (2009)

128


(pH 4.4-4.8). Field conditions.

Whole shoot sampled. Fe:186-344 Al: 40-263

Camm, Cascades, Westonia Potting mix. Field conditions.

Wl in pots for 49 d from 21 DAS in the field. Whole shoot sampled.

Mn: 38-53 Fe: 49-98 Al: 37-93

Setter et al. (2009)

Thassos Alluvial topsoil and brown subsoil. Field conditions.

Wl in pots for 15 d from 13 DAS in pots in the field. Whole shoot sampled.

N: 38/29 (topsoil/subsoil). P: 4.9/3.3 (topsoil/subsoil). K: 28/49 (topsoil/subsoil). Mg: 1.4/2 (topsoil/subsoil). Ca: 5.8/6.2 (topsoil/subsol). Mn: 27.5/35.1 (topsoil/subsoil). Fe: 89.7/70 (topsoil/subsoil). Zn: 28.5/21.4 (topsoil/subsoil).

Steffens et al. (2005)


Wl for 14 d from 11 DAS in growth cabinet. Whole shoot sampled.

N: 8 P: 3.5 K: 10.6 Mg: 0.4 Ca: 2.5



Wl for 15 d from 11 DAS (2 leave stage) in growth cabinet. Whole shoot sampled after 2 and 15 d wl1.

N: 28.2/8.7 (2/15d). P: 2.6/0.7 (2/15d). K: 32.2/10.1 (2/15d). Ca: 3.8/2.7 (2/15d). Mn: 122.5/14.3 (2/15d). Fe: 27.6/25.1 (2/15d).


Cappelle Deprez Sandy soil collected from field. Grown at 14 °C; wl at 6, 10, 14 or 18 °C soil temperature.

Wl for 14 d from 9 DAS (2 leave stage) at different temperatures in plastic cylinders in growth cabinet. Whole shoot sampled.

N: 10.6-14.3 (temperature range). P: 0.8-1.1 (temperature range). K: 9.1-12.2 (temperature range). Ca: 2.6-4.0 (temperature range).

Trought and Drew (1982)

1 Only 15 d plotted in Figure 5

129


Alamri S.A., Barrett-Lennard E.G., Teakle N.L. & Colmer T.D. (2013) Improvement of salt and waterlogging tolerance in wheat: comparative physiology of Hordeum marinum-Triticum aestivum amphiploids with their H. marinum and wheat parents. Functional Plant Biology, 40, 1168-1178.

Ballesteros D., Mason R.E., Addison C., Andrea Acuña M., Nelly Arguello M., Subramanian N., . . . Tucker D. (2014) Tolerance of wheat to vegetative stage soil waterlogging is conditioned by both constitutive and adaptive QTL. Euphytica, 201, 329–343.

Barrett-Lennard E.G., Leighton P.D., Buwalda F., Gibbs J., Armstrong W., Thomson C.J. & Greenway H. (1988) Effects of growing wheat in hypoxic nutrient solutions and of subsequent transfer to aerated solutions. 1. Growth and carbohydrate status of shoots and roots. Australian Journal of Plant Physiology, 15, 585-598.

Belford R.K. (1981) Response of winter-wheat to prolonged waterlogging under outdoor conditions. Journal of Agricultural Science, 97, 557-568.

Belford R.K., Cannell R.Q. & Thomson R.J. (1985) Effects of single and multiple waterloggings on the growth and yield of winter wheat on a clay soil. Journal of the Science of Food and Agriculture, 36, 142-156.

Cannell R.Q., Belford R.K., Gales K., Dennis C.W. & Prew R.D. (1980) Effects of waterlogging at different stages of development on the growth and yield of winter-wheat. Journal of the Science of Food and Agriculture, 31, 117-132.

Collaku A. & Harrison S.A. (2002) Losses in wheat due to waterlogging. Crop Science, 42, 444-450. Dickin E., Bennett S. & Wright D. (2009) Growth and yield responses of UK wheat cultivars to winter

waterlogging. The Journal of Agricultural Science, 147, 127-140. Dickin E. & Wright D. (2008) The effects of winter waterlogging and summer drought on the growth

and yield of winter wheat (Triticum aestivum L.). European Journal of Agronomy, 28, 234-244.

Ding N.C. & Musgrave M.E. (1995) Relationship between mineral coating on roots and yield performance of wheat under waterlogging stress. Journal of Experimental Botany, 46, 939-945.

Gardner W.K. & Flood R.G. (1993) Less waterlogging damage with long season wheats. Cereal Research Communications, 21, 337-343.

Hamonts K., Clough T.J., Stewart A., Clinton P.W., Richardson A.E., Wakelin S.A., . . . Condron L.M. (2013) Effect of nitrogen and waterlogging on denitrifier gene abundance, community structure and activity in the rhizosphere of wheat. FEMS Microbiology Ecology, 83, 568-584.

Haque M.E., Kawaguchi K. & Komatsu S. (2011) Analysis of proteins in aerenchymatous seminal roots of wheat grown in hypoxic soils under waterlogged conditions. Protein and Peptide Letters, 18, 912-924.

Haque M.E., Oyanagi A. & Kawaguchi K. (2012) Aerenchyma formation in the seminal roots of japanese wheat cultivars in relation to growth under waterlogged conditions. Plant Production Science, 15, 164-173.

Hayashi T., Yoshida T., Fujii K., Mitsuya S., Tsuji T., Okada Y., . . . Yamauchi A. (2013) Maintained root length density contributes to the waterlogging tolerance in common wheat (Triticum aestivum L.). Field Crops Research, 152, 27-35.

Hossain M.A., Araki H. & Takahashi T. (2011) Poor grain filling induced by waterlogging is similar to that in abnormal early ripening in wheat in Western Japan. Field Crops Research, 123, 100-108.

Huang B.R. & Johnson J.W. (1995) Root respiration and carbohydrate status of 2 wheat genotypes in response to hypoxia. Annals of Botany, 75, 427-432.

130


Huang B.R., Johnson J.W., Nesmith S. & Bridges D.C. (1994) Growth, physiological and anatomical responses of 2 wheat genotypes to waterlogging and nutrient supply. Journal of Experimental Botany, 45, 193-202.

Khabaz-Saberi H., Barker S.J. & Rengel Z. (2012) Tolerance to ion toxicities enhances wheat (Triticum aestivum L.) grain yield in waterlogged acidic soils. Plant and Soil, 354, 371-381.

Khabaz-Saberi H. & Rengel Z. (2010) Aluminum, manganese, and iron tolerance improves performance of wheat genotypes in waterlogged acidic soils. Journal of Plant Nutrition and Soil Science, 173, 461-468.

Khabaz-Saberi H., Setter T.L. & Waters I. (2005) Waterlogging induces high to toxic concentrations of iron, aluminum, and manganese in wheat varieties on acidic soil. Journal of Plant Nutrition, 29, 899-911.

Li C., Jiang D., Wollenweber B., Li Y., Dai T. & Cao W. (2011) Waterlogging pretreatment during vegetative growth improves tolerance to waterlogging after anthesis in wheat. Plant Science, 180, 672-678.

Luxmoore R., Fischer R. & Stolzy L. (1973) Flooding and soil temperature effects on wheat during grain filling. Agronomy Journal, 65, 361-364.

Malik A., Colmer T.D., Lambers H. & Schortemeyer M. (2001) Changes in physiological and morphological traits of roots and shoots of wheat in response to different depths of waterlogging. Australian Journal of Plant Physiology, 28, 1121-1131.

Malik A.I., Colmer T.D., Lambers H. & Schortemeyer M. (2003) Aerenchyma formation and radial O2 loss along adventitious roots of wheat with only the apical root portion exposed to O2 deficiency. Plant, Cell & Environment, 26, 1713-1722.

Malik A.I., Colmer T.D., Lambers H., Setter T.L. & Schortemeyer M. (2002) Short-term waterlogging has long-term effects on the growth and physiology of wheat. New Phytologist, 153, 225-236.

Malik A.I., Islam A.K.M.R. & Colmer T.D. (2011) Transfer of the barrier to radial oxygen loss in roots of Hordeum marinum to wheat (Triticum aestivum): evaluation of four H. marinum–wheat amphiploids. New Phytologist, 190, 499-508.

McDonald G.K. & Gardner W.K. (1987) Effect of waterlogging on the grain-yield response of wheat to sowing date in southwestern victoria. Australian Journal of Experimental Agriculture, 27, 661-670.

McDonald M.P., Galwey N.W. & Colmer T.D. (2001a) Waterlogging tolerance in the tribe Triticeae: the adventitious roots of Critesion marinum have a relatively high porosity and a barrier to radial oxygen loss. Plant, Cell & Environment, 24, 585-596.

McDonald M.P., Galwey N.W., Ellneskog-Staam P. & Colmer T.D. (2001b) Evaluation of Lophopyrum elongatum as a source of genetic diversity to increase the waterlogging tolerance of hexaploid wheat (Triticum aestivum). New Phytologist, 151, 369-380.

Musgrave M.E. (1994) Waterlogging effects on yield and photosynthesis in eight winter wheat cultivars. Crop Science, 34, 1314-1318.

Musgrave M.E. & Ding N. (1998) Evaluating wheat cultivars for waterlogging tolerance. Crop Science, 38, 90-97.

Olgun M., Metin Kumlay A., Cemal Adiguzel M. & Caglar A. (2008) The effect of waterlogging in wheat (T. aestivum L.). Acta Agriculturae Scandinavica: Section B, Soil & Plant Science, 58, 193-198.

Robertson D., Zhang H., Palta J.A., Colmer T. & Turner N.C. (2009) Waterlogging affects the growth, development of tillers, and yield of wheat through a severe, but transient, N deficiency. Crop and Pasture Science, 60, 578-586.

Sayre K., Van Ginkel M., Rajaram S. & Ortiz-Monasterio I. (1994) Tolerance to waterlogging losses in spring bread wheat: effect of time of onset on expression. Annual Wheat Newsletter, 40, 165-171.

131


Setter T.L. & Waters I. (2003) Review of prospects for germplasm improvement for waterlogging tolerance in wheat, barley and oats. Plant and Soil, 253, 1-34.

Setter T.L., Waters I., Sharma S.K., Singh K.N., Kulshreshtha N., Yaduvanshi N.P.S., . . . Cakir M. (2009) Review of wheat improvement for waterlogging tolerance in Australia and India: the importance of anaerobiosis and element toxicities associated with different soils. Annals of Botany, 103, 221-235.

Shao G.C., Lan J.J., Yu S.E., Liu N., Guo R.Q. & She D.L. (2013) Photosynthesis and growth of winter wheat in response to waterlogging at different growth stages. Photosynthetica, 51, 429-437.

Sharma D.B. & Swarup A. (1988) Effects of short-term flooding on growth, yield and mineral composition of wheat on sodic soil under field conditions. Plant and Soil, 107, 137-143.

Steffens D., Hutsch B.W., Eschholz T., Losak T. & Schubert S. (2005) Water logging may inhibit plant growth primarily by nutrient deficiency rather than nutrient toxicity. Plant Soil and Environment, 51, 545-552.

Stieger P.A. & Feller U. (1994) Nutrient accumulation and translocation in maturing wheat plants grown on waterlogged soil. Plant and Soil, 160, 87-95.

Suh H.-S. (1978) Studies on the wet-injury resistance of wheat and barley varieties. IV. Effect of Excess-Moisture in the Soil on the Growth of wheat, six row and two row barley at Various Stages. Korean Journal of Crop Science, 23, 26–31.

Tan W., Liu J., Dai T., Jing Q., Cao W. & Jiang D. (2008) Alterations in photosynthesis and antioxidant enzyme activity in winter wheat subjected to post-anthesis water-logging. Photosynthetica, 46, 21-27.

Thomson C.J., Colmer T.D., Watkin E.L.J. & Greenway H. (1992) Tolerance of wheat (Triticum aestivum cvs Gamenya and Kite) and triticale (Triticosecale cv. Muir) to waterlogging. New Phytologist, 120, 335-344.

Trought M.C.T. & Drew M.C. (1980a) Development of waterlogging damage in wheat seedlings (Triticum-aestivum L). 1. Shoot and root-growth in relation to changes in the concentrations of dissolved-gases and solutes in the soil solution. Plant and Soil, 54, 77-94.

Trought M.C.T. & Drew M.C. (1980b) The development of waterlogging damage in wheat seedlings (Triticum-aestivum L). 2. Accumulation and redistribution of nutrients by the shoot. Plant and Soil, 56, 187-199.

Trought M.C.T. & Drew M.C. (1982) Effects of waterlogging on young wheat plants (Triticum-aestivum L) and on soil solutes at different soil temperatures. Plant and Soil, 69, 311-326.

Watkin E.L.J., Thomson C.J. & Greenway H. (1998) Root development and aerenchyma formation in two wheat cultivars and one triticale cultivar grown in stagnant agar and aerated nutrient solution. Annals of Botany, 81, 349-354.

Watson E.R., Lapins P. & Barron R.J.W. (1976) Effect of waterlogging on growth, grain and straw yield of wheat, barley and oats. Australian Journal of Experimental Agriculture, 16, 114-122.

Webster C.P., Belford R.K. & Cannell R.Q. (1986) Crop uptake and leaching losses of 15N labelled fertilizer nitrogen in relation to waterlogging of clay and sandy loam soils. Plant and Soil, 92, 89-101.

Wiengweera A., Greenway H. & Thomson C.J. (1997) The use of agar nutrient solution to simulate lack of convection in waterlogged soils. Annals of Botany, 80, 115-123.

Wu J.-D., Li J.-C., Wei F.-Z., Wang C.-Y., Zhang Y. & Sun G. (2014) Effects of nitrogen spraying on the post-anthesis stage of winter wheat under waterlogging stress. Acta Physiologiae Plantarum, 36, 207-216.

Xu Q.T., Yang L., Zhou Z.Q., Mei F.Z., Qu L.H. & Zhou G.S. (2013) Process of aerenchyma formation and reactive oxygen species induced by waterlogging in wheat seminal roots. Planta, 238, 969-982.

Yaduvanshi N.P.S., Setter T.L., Sharma S.K., Singh K.N. & Kulshreshtha N. (2012) Influence of waterlogging on yield of wheat (Triticum aestivum), redox potentials, and concentrations of microelements in different soils in India and Australia. Soil Research, 50, 489-499.

132


Yamauchi T., Abe F., Kawaguchi K., Oyanagi A. & Nakazono M. (2014a) Adventitious roots of wheat seedlings that emerge in oxygen-deficient conditions have increased root diameters with highly developed lysigenous aerenchyma. Plant Signaling & Behavior, 9, e28506.

Yamauchi T., Watanabe K., Fukazawa A., Mori H., Abe F., Kawaguchi K., . . . Nakazono M. (2014b) Ethylene and reactive oxygen species are involved in root aerenchyma formation and adaptation of wheat seedlings to oxygen-deficient conditions. Journal of Experimental Botany, 65, 261-273.

Zheng C., Jiang D., Liu F., Dai T., Jing Q. & Cao W. (2009) Effects of salt and waterlogging stresses and their combination on leaf photosynthesis, chloroplast ATP synthesis, and antioxidant capacity in wheat. Plant Science, 176, 575-582.

133


Supporting Information to Chapter 2 Table S1. Primer sequences and annealing temperatures used for qRT-PCR of genes displayed in Fig. 6 and Fig. S7.

Enzyme name Gene name ID Forward Reverse

Annealing temperature

(oC) Reference

fructan exohydrolase 1-FEHw1 AJ516025 CACTAGTCCTGAAATTCACGGG GCTCACCGAGTCTTTTAACTAC 60 Meguro-Maoka and Yoshida (2016)

fructan exohydrolase 1-FEHw3 AJ564996 AATGTGGAGAAGGGTTGGAG GGCTATTTTCTTTCCTGCTG 60 Sharbatkhari et al. (2016)

fructan 6-exohydrolase 6-FEH

AM075205 CCTCGCCATTGTACACATCG TCCGGACGCGTAGCCAAG 60

Meguro-Maoka and Yoshida (2016)

fructan 6&1-exohydrolase 6&1-FEH AB089269

CATTGATGAGATCAACGGCTCAG

GCAACTACTTCAAAAGTTCAAGAGCAG 60 Meguro-Maoka and Yoshida (2016)

6-kestose exohydrolase 6-KEHw2 AB089271 TGAAGTAGACGCCCACGAGC CATAAAGTATGACGAGACTAAGATGGT 60

Meguro-Maoka and Yoshida (2016)

sucrose synthase SuS3 TC247088 AAGGCCTCCACTCTGCTTGTT CCTGTGAAATCTTGGTCCAGTG 57 Xue et al. (2008)

sucrose synthase SuS4 TC272798 CTGGAACCCCTGCTTGATTTC ACATCATCACATGCCCCTTGT 57 Xue et al. (2008)

sucrose synthase SuS5 TC267682 AGTACACGTCCCCTGCTAAACTGT TCTCCTTCCCCCGATGACA 59 Xue et al. (2008)

sucrose synthase SuS11 TC248447 CGAAAGAAGGGTGACTTGGATTT GCAGCGCCTAGGACGTAAAC 57 Xue et al. (2008)

alpha-amylase α-Amy2-1 CK207286 CAAAATTACCAAAGCAGCTCTACGAG ACTTTTACATGGAGGAAGTACTAAATCGTAC 57 Barrero et al. (2013) alpha-amylase α-Amy2-8 X13576 CAGCCAGTCAGCCAATCAC CATCATCATGTTGTACCACCC 55 Wu et al. (2013) alpha-amylase α-Amy2-54 X13580 CAGCCAGTCAGCCAATCAT AGCACAGGTTGCCGTTCTT 57 Wu et al. (2013) alpha-amylase α-Amy4-1 CJ574929 CGGCATGTGGAGACAGATATG TGCGGTAATCTGGTGTGGTGTAC 55 Barrero et al. (2013) RNA polymerase I, II, III 15kD subunit RP15 TC265122 GCACACGTGCTTTGCAGATAAG GCCCTCAAGCTCAACCATAACT 60 Xue et al. (2008)

134


Figure S1. Shoot RGR of wheat cultivars Frument and Jackson during submergence calculated from initial biomass (day 0 of submergence) and biomass after 8 days of submergence (no recovery period). n.s. indicates no significant difference between means (t-test, P = 0.8600). Values are means (± SE, n = 4).

F ru m e n t Ja c k s o n

-0 .0 5

0 .0 0

0 .0 5

0 .1 0

RG

R (

d-1

)

n .s .

135


Figure S2. (A) Leaf underwater photosynthesis (PN) and (B) dark respiration (RD) of wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence at 200 µM free CO2. The leaf sampled was the youngest, fully expanded leaf at the time of submergence (3rd leaf). Values are means (± SE, n = 4). Two-way time × cultivar ANOVA only showed a significant time effect (P < 0.0001) for both PN and RD and no significant cultivar effects. * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P > 0.05). Round symbols symbolize untreated controls in air at the end of the treatment period (day 16).

0 2 4 6 8 1 0 1 2 1 4 1 6-1

0

1

2

3

4


PN

(µ

mo

l O

2 m

-2 s

-1)

F r u m e n t s u b m e rg e d




A

0 2 4 6 8 1 0 1 2 1 4 1 6-0 .6

-0 .4

-0 .2

0 .0


RD

(µ

mo

l O

2 m

-2 s

-1)

F r u m e n t s u b m e rg e d




B

136


Figure S3. Trajectory plots from GC-MS analysis representing the average normalized relative peak areas of sugars in shoots of wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence. Round symbols symbolize untreated controls in air at the end of the treatment period (day 16). The entire shoot was homogenized and used for analysis. * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05), values are means (± SE, n = 4).

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0m a lto s e*

*

0 5 1 0 1 5 2 00

2 0 0 0 0

4 0 0 0 0

6 0 0 0 0

8 0 0 0 0

1 0 0 0 0 0ta g a to s e

*

0 5 1 0 1 5 2 00

2 0 0 0

4 0 0 0

6 0 0 01 ,6 -a n h y d ro -g lu c o s e

*

*

0 5 1 0 1 5 2 00

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0g lu c o s e -6 -p h o s p h a te

0 5 1 0 1 5 2 00

5 0 0 0

1 0 0 0 0

1 5 0 0 0

2 0 0 0 0k e s to s e*

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0m y o - is o n ito l


0 5 1 0 1 5 2 00

5 0 0 0

1 0 0 0 0

1 5 0 0 0r ib o s e


*

*

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0tre h a lo s e


*

*

No

rma

lize

d A

rea

(a

rbit

rary

un

its

)


F ru m e n t c o n tro lJ a c k s o n s u b m e rg e dF ru m e n t s u b m e rg e d

137


Figure S4. Starch in shoots of wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence. The entire shoot except for the 3rd leaf (used for other physiological measurements, see Fig. 2) was sampled. Round symbols symbolize untreated controls in air at the end of the treatment period (day 16). Two-way time × cultivar ANOVA on Ln-transformed data showed significant time (P < 0.0001) and cultivar (P < 0.001) effects; time explained 78% of the variation and cultivar 5%. * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05), values are means (± SE, n = 3-4).

0 5 1 0 1 5 2 0

1 0 0

2 0 0

3 0 0

4 0 0


Sta

rch

(µm

ol

he

xo

se

eq

. g

-1 D

M)

0

*F ru m e n t s u b m e rg e dJ a c k s o n s u b m e rg e d

F ru m e n t c o n tro lJ a c k s o n c o n tro l

138


Figure S5. Metabolites detected by LC-MS in shoots of wheat cultivars Frument and Jackson during 0-16 days of complete submergence, selected by two-way ANOVA (P < 0.05) and presented as Venn diagram for time, cultivar and time × cultivar interaction effects. LC-MS detected 74,359 mass spectral features out of which 1211 passed our quality controls filters (present in all samples of at least one group; in 80% of the quality control samples and with a CV < 35%).

139


Figure S6. Trajectory plots from GC-MS analysis representing the average normalized relative peak areas of amino acids in entire shoots of wheat cultivars Frument (open symbols) and Jackson (closed symbols) with time of submergence (for legend see Fig. S3). Round symbols symbolize untreated controls in air at the end of the treatment period (day 16). * indicate significant difference between cultivars (Sidak’s multiple comparisons test, P < 0.05), values are means (± SE, n = 4). Levels of Ser also declined for both cultivars (data not shown).

0 5 1 0 1 5 2 00

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0A s p *

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0I le

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0G ln *

0 5 1 0 1 5 2 00

5 0 0 0

1 0 0 0 0

1 5 0 0 0

2 0 0 0 0

2 5 0 0 0G ly

*

*

*

*

0 5 1 0 1 5 2 00

2 0 0 0

4 0 0 0

6 0 0 0

8 0 0 0A la

*

0 5 1 0 1 5 2 00

5 0 0 0

1 0 0 0 0

1 5 0 0 0

2 0 0 0 0A s n

*

**

0 5 1 0 1 5 2 00

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0G lu

0 5 1 0 1 5 2 00

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0L e u

*

0 5 1 0 1 5 2 00

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0L y s

*

*

0 5 1 0 1 5 2 00

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0M e t

*

0 5 1 0 1 5 2 00

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0P ro *

* *

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0T h r

0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0T h p *

*

0 5 1 0 1 5 2 00

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0V a l

*

0 5 1 0 1 5 2 00

5 0 0 0 0

1 0 0 0 0 0

1 5 0 0 0 0P h e

**

*

No

rma

lize

d A

rea

(a

rbit

rary

un

its

)

0 5 1 0 1 5 2 00

2 0 0 0 0

4 0 0 0 0

6 0 0 0 0

8 0 0 0 0T y r

*

*

*

140


Figure S7. Relative mRNA levels of α-amylases (α-AMY2-1, α-AMY2-8, α-AMY2-54, α-AMY4-1) in shoots of wheat cultivars Frument (open bars) and Jackson (closed bars) with time of submergence. For details on primers and annealing temperatures see Table S1. Transcripts of representative genes were quantified in shoots exposed to submergence for 0-12 days by qRT-PCR. The relative level of each mRNA was calculated by comparison with initial (day 0) Frument. Values are means (± SE, n = 3). * indicate significant differences between the cultivars (Student’s t-test, P < 0.05).

References in Table S1

Barrero JM, Mrva K, Talbot MJ, White RG, Taylor J, Gubler F, Mares DJ (2013) Genetic, hormonal, and physiological analysis of late maturity α-amylase in wheat. Plant Physiology 161: 1265-1277

Meguro-Maoka A, Yoshida M (2016) Analysis of seasonal expression levels of wheat fructan exohydrolase (FEH) genes regulating fructan metabolism involved in wintering ability. Journal of Plant Physiology 191: 54-62

Sharbatkhari M, Shobbar Z-S, Galeshi S, Nakhoda B (2016) Wheat stem reserves and salinity tolerance: molecular dissection of fructan biosynthesis and remobilization to grains. Planta 244: 191-202

Wu M, Wang F, Zhang C, Xie Y, Han B, Huang J, Shen W (2013) Heme oxygenase-1 is involved in nitric oxide-and cGMP-induced α-Amy2/54 gene expression in GA-treated wheat aleurone layers. Plant Molecular Biology 81: 27-40

Xue G-P, McIntyre CL, Jenkins CL, Glassop D, van Herwaarden AF, Shorter R (2008) Molecular dissection of variation in carbohydrate metabolism related to water-soluble carbohydrate accumulation in stems of wheat. Plant Physiology 146: 441-454

0 2 5 8

12

0

2

4

6

8

1 0

α -A M Y 2 -1

Re

lativ

e m

RN

A l

eve

l

F ru m e n tJ a c k s o n

0 2 5 8

12

0

2

4

6

8

1 0

α -A M Y 2 -8

Re

lativ

e m

RN

A l

eve

l *

0 2 5 8 1 20

2

4

6

α -A M Y 2 -5 4


Re

lativ

e m

RN

A l

eve

l *

0 2 5 8 1 20 .0

0 .5

1 .0

1 .5

2 .0

α -A M Y 4 -1


Re

lativ

e m

RN

A l

eve

l* *

*

141


Supporting Information to chapter 3

Figure S1. Scanning electron microscopy (SEM) micrographs of leaf surfaces of rice subject to 0-16 d

submergence in non-saline (0 mM NaCl) or saline (50 mM NaCl) water (containing basal ions, see

Methods). Each leaf is shown at 7000 × magnification (left) showing stomata, horizontal field width

(HFW) = 36.6 µm, and 500 × magnification (right), HFW = 512 µm.

142


Figure S2. Scanning electron microscopy (SEM) micrographs showing wax platelets on leaf surface

after loss of leaf hydrophobicity and leaf gas film disappearance (a, plants had been submerged in

water with 50 mM NaCl for 9 d), and wax platelets on papillae before (b, prior to submergence) and

after (c, plants had been submerged in water with 50 mM NaCl for 9 d) loss of leaf hydrophobicity and

leaf gas film disappearance. Leaves are shown at 27.000 × (a) and 45.000 × (b, c) magnification (HFW

= 3.9 µm and 2.5 µm, respectively).

143


0 5 1 0 1 5 2 00

1 0 0

2 0 0

3 0 0

4 0 0

T im e (d )

Le

af

K+

(mM

)

c ) + G F (0 m M N a C l)

+ G F (5 0 m M N a C l)

-G F (0 m M N a C l)

-G F (5 0 m M N a C l)

0 5 1 0 1 5 2 00

1 0 0

2 0 0

3 0 0

4 0 0

T im e (d )

Le

af

Na

+(m

M)

a)

0 5 1 0 1 5 2 00

1 0 0

2 0 0

3 0 0

4 0 0

T im e (d )

Le

af

Cl-

(mM

)

b )

Figure S3. Leaf Na+ (a), Cl- (b) and K+ (c) concentrations in the tissue water (mM) with time of

submergence in water (containing basal ions, see Methods) with 0 mM NaCl (squares) or 50 mM

NaCl (circles) for rice plants with leaf gas films (+GF, open symbols) or treated with 0.1% Triton X-

100 and without gas films (-GF, closed symbols). Samples were the entire 3rd and part of the 4th leaf

(see Methods). Roots were in non-saline nutrient solution. Values are means (± SE, n = 3-4 except Na+

on day 5 at 0 mM NaCl (-GF) where n = 1 due to a sampling error).

144


0 5 1 0 1 5 2 00

5 0 0

1 0 0 0

1 5 0 0

L e a f N a + (µ m o l g -1 D M )

Ch

loro

ph

yll

a

(mg

g-1

DM

)

-G F

+ G Fa)

0 5 1 0 1 5 2 00

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

L e a f C l- (µ m o l g -1 D M )

Ch

loro

ph

yll

a (

mg

g-1

DM

)

b )

0 5 1 0 1 5 2 00

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

L e a f K + (µ m o l g -1 D M )

Ch

loro

ph

yll

a (

mg

g-1

DM

)

c )

Figure S4. Correlations between leaf Na+ (a), Cl- (b) and K+ (c) concentrations (data from Fig. 2) and

corresponding leaf chlorophylla concentrations (data from Fig. 1b) after submergence of rice in water

(containing basal ions, see Methods) with 50 mM NaCl. Leaves were either left untreated thus

retaining a leaf gas film (+GF, open symbols) or treated with 0.1% Triton X-100 (-GF, closed

symbols). Roots were in non-saline nutrient solution. r values from non-parametric Spearman rank

correlation analysis, * denoting levels of significance (levels of P > 0.05, P ≤ 0.05, P ≤ 0.01, P ≤ 0.001

or P ≤ 0.0001 are denoted by n.s., *, **, ***, ****, respectively): Na+ r = -0.5084**; Cl- r = -0.3658*;

K+ r = 0.4664**.

145


0 5 0 0 1 0 0 0 1 5 0 0

0

5

1 0

1 5

L e a f N a + (µ m o l g -1 D M )

PN

(µ

mo

l O

2 m

-2 s

-1)

-G F+ G F

†

†

†

†

†

†

††† †

††

a)

0 5 0 0 1 0 0 0 1 5 0 0

0

5

1 0

1 5

L e a f C l- (µ m o l g -1 D M )

PN

(µ

mo

l O

2 m

-2 s

-1)

-G F+ G F

†

†

†

†

†

†

††† ††

b )

0 5 0 0 1 0 0 0 1 5 0 0

0

5

1 0

1 5

L e a f K + (µ m o l g -1 D M )

PN

(µ

mo

l O

2 m

-2 s

-1) + G F

-G F

† †††

†

†

†

†

†

†

††

c )

Figure S5. Correlations between leaf Na+ (a), Cl- (b) and K+ (c) concentrations (data from Fig. 2) with

corresponding underwater PN (µmol O2 m-2 s-1) at 2500 µM CO2 and 50 mM NaCl (data from Fig.

3b). Leaves were either left untreated thus retaining a leaf gas film (+GF) or treated with 0.1% Triton

X-100 (-GF). Roots were in non-saline nutrient solution and shoots were submerged in water

containing NaCl treatments and basal ions (see Methods). r values from non-parametric Spearman

rank correlation analysis, * denoting levels of significance (levels of P > 0.05, P ≤ 0.05, P ≤ 0.01, P ≤

0.001 or P ≤ 0.0001 are denoted by n.s., *, **, ***, ****, respectively): Na+ r = -0.6078**; Cl- r = -

0.3243 n.s.; K+ r = 0.1923 n.s. † denotes points excluded from the correlation analysis to prevent leaf

deterioration with time of submergence to draw the correlation. Points were excluded when leaf both

porosity and leaf chlorophylla were < 4.5% and 8.3 mg g-1 DM, respectively, as for d 16 (+GF) and d 5

146


and 9 (-GF). At 200 µM free CO2 all correlations were not significant (Spearman rank correlation

analysis, P > 0.05, data not shown).

Table S1. Shoot length and number of tillers of rice plants submerged in water (containing basal ions,

see Methods) with 0 mM NaCl or 50 mM NaCl with leaf gas films (+GF) or treated with 0.1% Triton

X-100 and without gas films (-GF). +GF and –GF plants are from different batches, hence these have

separate emergent (shoots in air) controls. Roots were in non-saline nutrient solution. Letters denote

significant difference (P < 0.05) between means (± S.E., n = 4) according to one-way ANOVA with

Sidak’s multiple comparisons post-hoc test (shoot length) or non-parametric Kruskal-Wallis with

Dunn’s multiple comparisons test (number of tillers).

Emergent controls 0 mM NaCl 50 mM NaCl

+GF -GF +GF -GF +GF -GF

Shoot length

(cm)

51.8 ±

3.3a

32.5 ±

2.6c

72 ± 1.7b 60.8 ± 6.6ad 67.1 ± 2.9bd 57.1 ± 4.9a

Number of

tillers

4.5 ± 0.6a 4 ± 0a 1.5 ± 0.6 ab 1.75 ± 0.5 ab 1.5 ± 0.6 ab 1 ± 0b

147


0 1 2 3 40

1 0 0

2 0 0

3 0 0

4 0 0

5 0 0

2 0 2 2

O 2 (k P a )

Na

+(µ

mo

l g

-1 D

M)

*a)

+ G F

-G F

0 1 2 3 40

1 0 0

2 0 0

3 0 0

4 0 0

5 0 0

2 0 2 2

O 2 (k P a )

Cl-

(µm

ol

g-1

DM

)

b )

0 1 2 3 40

5 0 0

1 0 0 0

1 5 0 0

2 0 2 2

O 2 (k P a )

K+

(µm

ol

g-1

DM

)

c )

Figure S6. Leaf Na+ (a), Cl- (b) and K+ (c) concentrations of a youngest fully expanded leaf of rice

submerged in 50 mM NaCl (containing also basal ions, see Methods) for 24 hours in the dark with gas

films (+GF, open symbols) or treated with 0.1% Triton X-100 and without gas films (-GF, closed

symbols). During incubation the submergence solution was maintained at 0.01, 0.46, 1.59, 3.16, and

20.23 kPa O2. * denotes significant difference between ion concentrations (Sidak’s multiple

comparisons test, P < 0.05). Two-way GF × pO2 ANOVA showed a significant effect of GF on leaf

Na+ (P = 0.0345) and K+ concentrations (P = 0.0091). pO2 had a significant effect on K+

concentrations (P < 0.0001). For Cl- two-way ANOVA showed a significant pO2 × GF interaction (P

= 0.0050). Values are means (± SE, n = 4).

148

Documents

Mechanisms of flood tolerance in wheat and rice Herzog.pdf · Mechanisms of flood tolerance in wheat and rice The role of leaf gas films during plant submergence Academic supervisor