functional response of ulmus minor mill. to drought ...oa.upm.es/38556/1/meng_li.pdffunctional...

UNIVERSIDAD POLITÉCNICA DE MADRID

ESCUELA TÉCNICA SUPERIOR DE INGENIEROS DE MONTES,

FORESTAL Y DEL MEDIO NATURAL

FUNCTIONAL RESPONSE OF Ulmus minor Mill. TO

DROUGHT, FLOODING AND DUTCH ELM DISEASE

TESIS DOCTORAL

MENG LI

Ingeniero Agrónomo

2015

DEPARTAMENTO DE SISTEMAS NATURALES

ESCUELA TÉCNICA SUPERIOR

DE INGENIEROS DE MONTES, FORESTAL Y DEL MEDIO

NATURAL

FUNCTIONAL RESPONSE OF Ulmus minor Mill. TO

DROUGHT, FLOODING AND DUTCH ELM DISEASE

MENG LI

Ingeniero Agrónomo

DIRECTORES:

JESÚS RODRÍGUEZ CALCERRADA

Doctor en Biología

LUIS GIL SÁNCHEZ

Doctor Ingeniero de Montes

UNIVERSIDAD POLITÉCNICA DE MADRID

Tribunal nombrado por el Magfco. Y Excmo. Sr. Rector de la Universidad

Politécnica de Madrid, el día …. de ……………….. de 201…

Presidente: .…………………………………………………………………

Vocal: ………………………………………………………………………..

Vocal: ………………………………………………………………………..

Vocal: ………………………………………………………………………..

Secretario: …………………………………………………………………..

Suplente: ..…………………………………………………………………..

Suplente: ..…………………………………………………………………..

Realizado el acto de defensa y lectura de tesis el día …. de ……………….. de

201… en la E.T.S.I. Montes, Forestal y del Medio Natural

EL PRESIDENTE LOS VOCALES

EL SECRETARIO

Acknowledgements

I am pleased to have this opportunity to express my hearted gratitude to all the

people who have guided, helped and accompanied me through this long road of a

doctoral thesis, since the beginning, to this final stage.

I would like to thank my family first. Thanks to my parents; their concerned

callings from China every week made me feel the warmth of home...even if they have

never quite understand what I have been doing in Spain all these years where nobody

speaks Chinese. Thanks to my wife, who has always provided understanding,

motivation and love; sorry for having left you these past years; now we will meet soon.

I am thankful to my supervisor Luis Gil Sánchez, who trusted me and opened a

door to let me join this hard-working, efficient and friendly team. He gave me the

precious opportunity to work on a great project and an interesting Ph D topic. Sorry for

my terrible Castellano. “Your happiness is my happiness’ is an impressive slogan for a

supervisor!

I am thankful to Jesús Rodríguez Calcerrada; my second supervisor, and my best

Spanish friend. He has taught me to be a better person. His insightful and constructive

advices have improved my scientific capacity; his constant support, encouragement

and patience have cultivated my interest in science; his concern for and help to others,

consideration, sense of humor and positive thinking has helped to create a pleasant

working atmosphere.

I am thankful to Martín Venturas, Javier Cano, Juan Antonio Martín, Pilar Pita

and Rosana López for their sincere and professional assistance and suggestions to

improve my thesis manuscripts. Without their collaboration, you would not be now

reading the acknowledgments of this doctoral thesis.

I am thankful to Eva Miranda, Guillermo G. Gordaliza and Jorge Domínguez for

their selfless assistance. All the experimental preparations, maintenance, sampling and

measurements are tough works to which they have contributed. Moreover, with friends,

it is always more fun to work.

My sincere gratitude goes also to all members of the lab. It is pity that I cannot

speak Spanish yet. However, everyone has given me a hand without hesitation

whenever I have needed one. I am impressed by the diligence, passion and optimism

of this group whom I will never forget.

I am thankful to Ana Moliner (Secretary of PhD of “ETSI Agrónomos”), Angel

Álvarez (Associate Vice-Rector for International Relations of UPM), Carmen Collada

(Secretary of PhD of “ETSI Montes, Forestal y del Medio Natural”) and Ji Cailing

(Leader of Chinese Embassy Education Group in Spain). They have made many

administrative efforts to let me join this group successfully.

Finally, I am thankful to the cooperation between “Universidad Politécnica de

Madrid” and China scholarship council for providing me with the opportunity to have

this beautiful experience.

INDEX

RESUMEN…………………………………………………………………………………….

ABSTRACT…………………………………………………………………………………...

1. Introduction……………………………………………………………………………....

1.1 General introduction to European elm species…...……………………………..

1.1.1 European elm species – ecology and distribution..…............................

1.1.2 Elms in Spain……………………………………………………………….

1.1.3 A brief note on history of DED..…………………………………………..

1.1.4 Infection process…………………………………………………………….

1.1.5 Strategies to fight against DED…………………………………………….

1.2 General plant responses to abiotic and biotic stresses ………………………..

1.2.1 General plant responses to flooding………………………………………

1.2.2 General plant responses to drought………………………………………

1.2.3 General plant responses to fungus infection……………………………..

1.3 Thesis outline………………………………………………………………………..

1.4 References……………………………………………………………………………

2. Aims..………………………………………………………………………………..........

3. Greater resistance to flooding of seedlings of Ulmus laevis than Ulmus

minor is related to the maintenance of a more positive carbon balance……..

Introduction……………………………………………………………………………….

Material and methods…………………………………………………………………..

Results..………………………………………………………………………………….

Discussion……………………………………………..…………………………………

References……………………………………………………………………………….

4. Drought-induced massive xylem embolism incurs a variable depletion of

non-structural carbon reserves in dying seedlings of two tree species………

Introduction……………………………………………………………………………….

Material and methods…………………………………………………………………..

Results……………………………………………………………………………………

Discussion………………………………………………………………………….........

References……………………………………………………………………………….

I

III

1

3

3

7

7

8

9

10

10

12

14

18

19

26

31

35

37

42

52

57

65

69

71

76

92

97

5. Physiological and biochemical differences among Ulmus minor genotypes

showing a gradient of resistance to Dutch elm disease..……………………….

Introduction……….………………………………………………………………………

Material and methods…………………………………………………………………...

Results……………………………………………………………………………………

Discussion………………………………………………………………………………..

References……………………………………………………………………………….

6. General discussion..…………………………………………………………………....

References………………………………………………………………………………

.

7. Conclusions………………………………………………………….............................

104

108

111

118

130

136

145

152

154

I

RESUMEN

Ulmus minor es una especie arbórea originaria de Europa cuyas poblaciones han

sido diezmadas por el hongo patógeno causante de la enfermedad de la grafiosis. La

conservación de los olmos exige plantearse su propagación a través de plantaciones y

conocer mejor su ecología y biología. Ulmus minor es un árbol de ribera, pero

frecuentemente se encuentra alejado del cauce de arroyos y ríos, donde la capa

freática sufre fuertes oscilaciones. Por ello, nuestra hipótesis general es que esta

especie es moderadamente resistente tanto a la inundación como a la sequía.

El principal objetivo de esta tesis doctoral es entender desde un punto de vista

funcional la respuesta de U. minor a la inundación, la sequía y la infección por O.

novo-ulmi; los factores que posiblemente más influyen en la distribución actual de U.

minor. Con este objetivo se persigue dar continuidad a los esfuerzos de conservación

de esta especie que desde hace años se dedican en varios centros de investigación a

nivel mundial, ya que, entender mejor los mecanismos que contribuyen a la resistencia

de U. minor ante la inoculación con O. novo-ulmi y factores de estrés abiótico ayudará

en la selección y propagación de genotipos resistentes a la grafiosis.

Se han planteado tres experimentos en este sentido. Primero, se ha comparado

la tolerancia de brinzales de U. minor y U. laevis – otro olmo ibérico – a una inmersión

controlada con el fin de evaluar su tolerancia a la inundación y comprender los

mecanismos de aclimatación. Segundo, se ha comparado la tolerancia de brinzales de

U. minor y Quercus ilex – una especie típica de ambientes Mediterránea secos – a la

falta de agua en el suelo con el fin de evaluar el grado de tolerancia y los mecanismos

de aclimatación a la sequía. El hecho de comparar dos especies contrastadas

responde al interés en entender mejor cuales son los procesos que conducen a la

muerte de una planta en condiciones de sequía – asunto sobre el que hay una

interesante discusión desde hace algunos años. En tercer lugar, con el fin de entender

mejor la resistencia de algunos genotipos de U. minor a la grafiosis, se han estudiado

las diferencias fisiológicas y químicas constitutivas e inducidas por O. novo-ulmi entre

clones de U. minor seleccionados a priori por su variable grado de resistencia a esta

enfermedad.

En el primer experimento se observó que los brinzales de U. minor sobrevivieron

60 días inmersos en una piscina con agua no estancada hasta una altura de 2-3 cm

por encima del cuello de la raíz. A los 60 días, los brinzales de U. laevis se sacaron de

la piscina y, a lo largo de las siguientes semanas, fueron capaces de recuperar las

funciones fisiológicas que habían sido alteradas anteriormente. La conductividad

II

hidráulica de las raíces y la tasa de asimilación de CO2 neta disminuyeron en

ambas especies. Por el contrario, la tasa de respiración de hojas, tallos y raíces

aumentó en las primeras semanas de la inundación, posiblemente en relación al

aumento de energía necesario para desarrollar mecanismos de aclimatación a la

inundación, como la hipertrofia de las lenticelas que se observó en ambas especies.

Por ello, el desequilibrio del balance de carbono de la planta podría ser un factor

relevante en la mortalidad de las plantas ante inundaciones prolongadas.

Las plantas de U. minor (cultivadas en envases de 16 litros a media sombra)

sobrevivieron por un prolongado periodo de tiempo en verano sin riego; la mitad de las

plantas murieron tras 90 días sin riego. El cierre de los estomas y la pérdida de hojas

contribuyeron a ralentizar las pérdidas de agua y tolerar la sequía en U. minor. Las

obvias diferencias en tolerancia a la sequía con respecto a Q. ilex se reflejaron en la

distinta capacidad para ralentizar la aparición del estrés hídrico tras dejar de regar y

para transportar agua en condiciones de elevada tensión en el xilema. Más relevante

es que las plantas con evidentes síntomas de decaimiento previo a su muerte

exhibieron pérdidas de conductividad hidráulica en las raíces del 80% en ambas

especies, mientras que las reservas de carbohidratos apenas variaron y lo hicieron de

forma desigual en ambas especies.

Árboles de U. minor de 5 y 6 años de edad (plantados en eras con riego

mantenido) exhibieron una respuesta a la inoculación con O. novo-ulmi consistente

con ensayos previos de resistencia. La conductividad hidráulica del tallo, el potencial

hídrico foliar y la tasa de asimilación de CO2 neta disminuyeron significativamente en

relación a árboles inoculados con agua, pero solo en los clones susceptibles. Este

hecho enlaza con el perfil químico “más defensivo” de los clones resistentes, es decir,

con los mayores niveles de suberina, ácidos grasos y compuestos fenólicos en estos

clones que en los susceptibles. Ello podría restringir la propagación del hongo en el

árbol y preservar el comportamiento fisiológico de los clones resistentes al inocularlos

con el patógeno.

Los datos indican una respuesta fisiológica común de U. minor a la inundación,

la sequía y la infección por O. novo-ulmi: pérdida de conductividad hidráulica, estrés

hídrico y pérdida de ganancia neta de carbono. Pese a ello, U. minor desarrolla varios

mecanismos que le confieren una capacidad moderada para vivir en suelos

temporalmente anegados o secos. Por otro lado, el perfil químico es un factor

relevante en la resistencia de ciertos genotipos a la grafiosis. Futuros estudios

deberían examinar como este perfil químico y la resistencia a la grafiosis se ven

alteradas por el estrés abiótico.

III

ABSTRACT

Ulmus minor is a native European elm species whose populations have been

decimated by the Dutch elm disease (DED). An active conservation of this species

requires large-scale plantations and a better understanding of its biology and ecology.

U. minor generally grows close to water channels. However, of the Iberian riparian tree

species, U. minor is the one that spread farther away from rivers and streams. For

these reasons, we hypothesize that this species is moderately tolerant to both flooding

and drought stresses.

The main aim of the present PhD thesis is to better understand the functional

response of U. minor to the abiotic stresses – flooding and drought – and the biotic

stress – DED – that can be most influential on its distribution. The overarching goal is

to aid in the conservation of this emblematic species through a better understanding of

the mechanisms that contribute to resistance to abiotic and biotic stresses; an

information that can help in the selection of resistant genotypes and their expansion in

large-scale plantations.

To this end, three experiments were set up. First, we compared the tolerance to

experimental immersion between seedlings of U. minor and U. laevis – another

European riparian elm species – in order to assess their degree of tolerance and

understand the mechanisms of acclimation to this stress. Second, we investigated the

tolerance to drought of U. minor seedlings in comparison with Quercus ilex (an oak

species typical of dry Mediterranean habitats). Besides assessing and understanding U.

minor tolerance to drought at the seedling stage, the aim was to shed light into the

functional alterations that trigger drought-induced plant mortality – a matter of

controversy in the last years. Third, we studied constitutive and induced physiological

and biochemical differences among clones of variable DED resistance, before and

following inoculation with Ophiostoma novo-ulmi. The goal is to shed light into the

factors of DED resistance that is evident in some genotypes of U. minor, but not others.

Potted seedlings of U. minor survived for 60 days immersed in a pool with

running water to approximately 2-3 cm above the stem collar. By this time, U. minor

seedlings died, whereas U. laevis seedlings moved out of the pool were able to recover

most physiological functions that had been altered by flooding. For example, root

hydraulic conductivity and leaf photosynthetic CO2 uptake decreased in both species;

while respiration initially increased with flooding in leaves, stems and roots possibly to

respond to energy demands associated to mechanisms of acclimation to soil oxygen

deficiency; as example, a remarkable hypertrophy of lenticels was soon observed in

IV

flooded seedlings of both species. Therefore, the inability to maintain a positive carbon

balance somehow compromises seedling survival under flooding, earlier in U. minor

than U. laevis, partly explaining their differential habitats.

Potted seedlings of U. minor survived for a remarkable long time without irrigation

– half of plants dying only after 90 days of no irrigation in conditions of high vapour

pressure deficit typical of summer. Some mechanisms that contributed to tolerate

drought were leaf shedding and stomata closure, which reduced water loss and the risk

of xylem cavitation. Obviously, U. minor was less tolerant to drought than Q. ilex,

differences in drought tolerance resulting mostly from the distinct capacity to postpone

water stress and conduct water under high xylem tension among species. More

relevant was that plants of both species exhibited similar symptoms of root hydraulic

failure (i.e. approximately 80% loss of hydraulic conductivity), but a slight and variable

depletion of non-structural carbohydrate reserves preceding dieback.

Five- and six-year-old trees of U. minor (planted in the field with supplementary

watering) belonging to clones of contrasted susceptibility to DED exhibited a different

physiological response to inoculation with O. novo-ulmi. Stem hydraulic conductivity,

leaf water potential and photosynthetic CO2 uptake decreased significantly relative to

control trees inoculated with water only in DED susceptible clones. This is consistent

with the “more defensive” chemical profile observed in resistant clones, i.e. with higher

levels of saturated hydrocarbons (suberin and fatty acids) and phenolic compounds

than in susceptible clones. These compounds could restrict the spread of O. novo-ulmi

and contribute to preserving the near-normal physiological function of resistant trees

when exposed to the pathogen.

These results evidence common physiological responses of U. minor to flooding,

drought and pathogen infection leading to xylem water disruption, leaf water stress and

reduced net carbon gain. Still, seedlings of U. minor develop various mechanisms of

acclimation to abiotic stresses that can play a role in surviving moderate periods of

flood and drought. The chemical profile appears to be an important factor for the

resistance of some genotypes of U. minor to DED. How abiotic stresses such as

flooding and drought affect the capacity of resistant U. minor clones to face O. novo-

ulmi is a key question that must be contemplated in future research.

1

1. INTRODUCTION

Introduction

3

1. INTRODUCTION

1.1 General introduction to European elm species

1.1.1 European elm species – ecology and distribution

Elms are deciduous and semi-deciduous trees comprising more than 40

species, most of them widely distributed in the temperate region of the northern

hemisphere. Approximately 30 species are distributed in eastern Asia, 10 in America

and three in Europe, although the exact numbers are still a matter of controversy

(Hollingsworth et al. 2000).

Among the European elms, two sections of the genus are native to Europe. U.

minor Mill (field elm) is one of the three common native European species. It has the

highest polymorphism and belongs to the Ulmus section. Mature trees have a strong

capacity to grow root suckers but rarely reproduce sexually because of seed abortion

(López-Almansa 2004). It was named ‘minor’ due to its smaller leaf than other

European species. U. minor is predominantly distributed in southern Europe (Fig 1) and

commonly grows near rivers or on floodplains with a shallow water table that may lead

to periodic floodings and summer droughts. This species prefers to grow on calcareous

and alkaline soil but can also grow on acid soil (Collin et al. 2008; Venturas et al. 2014).

Before the appearance of Dutch elm disease (DED), U. minor was commonly

propagated and planted in urban areas due to its ornamental characteristics and was

capable of providing good quality timber (Martín et al. 2010). This situation

differentiates it from the other two European elm species, which mainly grow in native

woodlands (Gil et al. 2004). However, the two serious DED pandemics killed the vast

majority of mature U. minor trees, making this species one of the greatest casualties of

the disease (Brasier 2000).

Functional response of Ulmus minor Mill. to drought, flooding and Dutch elm disease

4

Fig. 1. The distribution of U. minor. This figure is obtained from Atlas Florae Europaeae (Jalas and

Suominen 1999) - Distribution of Vascular Plants in Europe.

The other species in the Ulmus section is U. glabra. It generally grows in

mountain areas and is widespread in the temperate zone of northern Europe and

mountain regions of southern Europe (Fig 2). Although U. glabra is a highly susceptible

species to DED, many trees in the north were not infected by the disease because of

their unattractiveness to the insect vector compared with other plant species in this

area (Collin et al. 2000). Unlike U. minor, this species rarely produces suckers and

does not resprout from stumps. As a result, its genetic resource is difficult to conserve

due to DED-induced loss combined with its weak ability to reproduce asexually.

However, it can widely reproduce sexually (Collin et al. 2000).

Introduction

5

Fig. 2. The distribution of U. glabra. This figure is obtained from Atlas Florae Europaeae (Jalas and

Suominen 1999) - Distribution of Vascular Plants in Europe.

The other section, Blepharocarpus, has a single species, U. laevis Pall

(European white elm). U. laevis is mainly distributed in eastern Europe, from the Ural

Mountains to eastern France and from southern Finland to the Caucasus and Bosnia,

crossing central and eastern Europe (Fig 3). U. laevis generally grows along rivers,

indicating it is one of the riparian elm species that can tolerate long term flooding

(Hollingsworth et al. 2000). This species has rarely been afflicted by DED due to its

lower attraction for the vector bark beetle compared with other elm species, such as U.

minor, although they grow together in Europe (Pajares 2004). However, the dry climate

and changes caused by humans, such as floodplain cultivation and urban development,

threaten its habitat (Collin 2003; Venturas et al. 2015). Therefore, protecting the habitat

of U. laevis could be as important for conserving this species as the efforts to combat

the threat of DED.


6

Fig. 3. The distribution of U. laevis. This figure is obtained from Atlas Florae Europaeae (Jalas and

Suominen 1999) - Distribution of Vascular Plants in Europe; more populations have been cited for

the species in Spain in more recent years (see Venturas et al. 2015)

Introduction

7

1.1.2 Elms in Spain

U. minor, U. laevis and U. glabra are all distributed in the Iberian Peninsula and

are all indigenous elm species in Spain. Although the elms in Spain have a similar

background to the other European elms, they also have their own particular genetic

characteristics, resistant features, method of breeding and conservation principle. As

previously mentioned, U. minor has been artificially planted throughout history for vine

grafting, ornamental purposes and timber utilization. It is therefore difficult to clearly

define its native distribution. As a riparian elm species that can also adapt to summer

drought, U. minor is distributed on floodplains with a shallow water table. In Spain, it

prefers to grow in the east due to the calcareous, alkaline soils, although the

Mediterranean climate allows its distribution throughout the Iberian Peninsula

(Venturas et al. 2014). In contrast, U. glabra, from the same section and also native to

Spain, is limited to the north and the Mediterranean mountains (Solla et al. 2000). U.

laevis was also genetically demonstrated to be a native species in the Iberian

Peninsula; in terms of soil preferences, the Spanish distribution of U. laevis is

associated with siliceous, moderately acid and moist soil prone to be temporarily

waterlogged (Venturas et al. 2015).

1.1.3 A brief note on history of DED

In the 20th century, a devastating fungal disease known as Dutch elm disease

(DED) attacked twice in Europe and North America, causing the death of millions of

elm trees. The first pandemic, noticed in 1920, was caused by the fungus Ophiostoma

ulmi. The second, which started in 1970 and continues today, was caused by

Ophiostoma novo-ulmi. This is a more aggressive and more common fungal species in

the current DED pandemic (Brasier 1991; Brasier 2000). Unfortunately, except for the

resistant Asian species, most elm species are susceptible to DED (Smalley and Guries

2000). Heavy losses of U. glabra occurred in northern Europe and the mountain forests

of southern Europe. Although U. laevis was able to avoid the DED threat because of its

lack of attraction for the vector bark beetle, it was demonstrated to be a highly

susceptible species to DED (Solla et al. 2005; Venturas et al. 2015). However, the

most serious damage was to U. minor, which suffered dramatic loss of mature trees in

forests and urban areas.


8

1.1.4 Infection process

The fungus is carried by the vector - elm beetles - or transmitted through grafted

roots. Three bark beetles, the native elm bark beetle (Hylurgopinus rufipes), the smaller

European elm bark beetle (Scolytus multistriatus) and the banded elm bark beetle (S.

schevyrewi), are responsible for fungus transmission (Pajares 2004). The general

process for DED infection starts when the beetles lay their eggs in dead or dying elm

trees affected by DED. The DED fungus produces sticky spores in the beetle galleries.

When the new beetles become adults and emerge from the infected elm tree, they also

carry spores. Then the adult beetles fly to other healthy elm trees, introducing the

fungus when they feed on the elm branches. In this vascular wilting disease, the fungus

is introduced into the vascular system of the tree during the feeding process, causing

infection. In the case of root infection, infected and healthy elm trees commonly grow

together, making it easy for their roots to graft and resulting in fungus transmission and

upwards spread from the roots (Karnosky 1979; Martín JA 2010) (details in Fig 4). The

infection induces vascular plugging, stops water and nutrition transport through the tree,

and results in wilting symptoms and even mortality (Newbanks 1983; Ouellette et al.

2004).

Figure 4. The cycle of Dutch elm disease. This figure is cited from D’Arcy (2000).

Introduction

9

1.1.5 Strategies to fight against DED

Because of the devastating damage by DED to European elms, many breeding

programs aim to find DED resistant elm clones and recover the elm population in every

European country (Heybroek 1993; Mittempergher and Santini 2004). Two important

international programs - EU RESGEN 78 project and EUFORGEN Noble Hardwoods

Network - were established among many collaborating European countries. RESGEN

focuses on evaluating the existing genetic resource collection and setting up long term

rational ex situ conservation of European elm resources. EUFORGEN recommends

dynamic genetic management according to genetic, ecological and geographic

information, to preserve, utilize and develop European elm resources sustainably in the

future (Collin et al. 2000).

As U. minor is one of the native elms, many studies have been conducted on

this species. Although it has shown very high susceptibility to DED and suffered huge

loss of populations, U. minor also has an excellent ability to grow root suckers (Collin et

al. 2000; Hollingsworth et al. 2000). This ability is beneficial for conservation of its

genetic resources, ease of cultivation and extensive planting. Its natural hybridization

with DED resistant elm species U. pumila, introduced from Asia, also accounts for the

interest of researchers in this species. Recent breeding programs have devoted a

continuous effort to the identification of resistant genotypes of U. minor (Collin and

Vernisson 2006; Franke et al. 2004; Martín et al. 2014) and several resistant genotypes

to DED have been propagated ex-situ in recent decades.

The two DED pandemics also occurred in Spain, although the first of these did

not have such a dramatic effect as it did in northern Europe, indicating the greater

resistance of Spanish elms in this case. The second DED pandemic resulted in high

mortality of U. minor in the Iberian Peninsula. This led to the establishment of the

Spanish program for the conservation and breeding of elms against DED, which aims

to preserve native genetic resources and breed individuals resistant to DED.

Preliminary work focused on creating resistant individuals by crossing with U. pumila, a

resistant Asian species introduced into Spain that commonly hybridizes with U. minor.

Some resistant elms have been observed. In the 1990s the Spanish program included

native U. minor and started to train native resistant clones for forest recovery and

ornamental use. Several U. minor clones with reliable resistant characteristics to the

aggressive DED pathogen O. novo-ulmi have been found (Martín et al. 2014). Large-


10

scale plantations with these clones are envisaged by the LIFE+ project “Restoration of

Iberian elms (Ulmus minor and U. laevis) in the Tagus River basin” as a primordial step

in the recovery of this species in its natural habitats.

As with other fungal diseases, DED is determined by the relationship between

fungus and host. Hence, the success of plantations depend obviously on the plant

material selected (and its resistance to DED), but also on micro-climatic and edaphic

conditions of chosen sites. It is necessary that more studies address U. minor

responses to abiotic and biotic factors from an eco-physiological perspective to better

discern optimal growing conditions and increase chances of plantation success. As

previously mentioned, the habitat of U. minor can experience periodic flooding and

drought. Both are classic abiotic stresses affecting plant growth, survival and

distribution. Therefore, research on the response to these factors is important to

understand the mechanisms of defense and survival of elm species. Moreover,

additional efforts are also needed for understanding causes of DED resistance. This is

a way for accelerating selection of suitable genotypes for artificial breeding and

regeneration.

1.2 General plant responses to abiotic and biotic stresses

1.2.1 General plant responses to flooding

Trees develop specific physiological, anatomical, biochemical and

morphological responses when encountering flooding. These responses are often

concurrent and interact with each other (Kozlowski 1997).

Photosynthetic net CO2 uptake rapidly decreases after flooding due to stomatal

closure and limitations to CO2 diffusion to chloroplasts (Gravatt and Kirby 1998). The

regulation of hormones such as abscisic acid (ABA) may be related to this (Kozlowski

1997; Pimenta et al. 2010), but stomatal closure is also related to reduced root

hydraulic conductivity. Hypoxia/anoxia limits the growth and function of submerged

roots, and limited root water uptake combined with continuous transpiration leads to

water deficit and stomata closure (Nicolás et al. 2005). After long periods of flooding,

non-stomatal limitations may also reduce photosynthetic net CO2 uptake and

photosynthetic capacity (Pezeshki 2001).

Interestingly, unlike drought stress, stomatal closure in many species is not

always associated with leaf water deficit during flooding. Anatomical and morphological

Introduction

11

adaptations can keep leaf water potential from decline (Pimenta et al. 2010). Water

transport is facilitated by the vigorous and highly porous root system of resistant

species and the formation of hypertrophied lenticels and adventitious roots. These

adaptations also supply oxygen to maintain the growth and function of submerged

roots. The formation of aerenchyma in the root cortex is a common response to

flooding, helping in transporting oxygen from the aerial shoot to submerged roots.

These morphological and anatomical responses are induced by hormonal signals from

submerged roots, such as ethylene (Glenz et al. 2006; Kozlowski 1997; Yamamoto et

al. 1995). The main detrimental effect of temporary flooding for trees is root hypoxia

(Colmer 2003). Soil oxygen deficiency inhibits root aerobic respiration, limiting normal

root function (Bailey-Serres and Voesenek 2008; Jaeger et al. 2009). Trees have to

shift their aerobic respiration to anaerobic fermentation. Because of low ATP synthesis

through glycolysis, more carbohydrates need to be consumed to meet energy

requirements (Jaeger et al. 2009).

All of this shows that flooding influences the water status and carbon reserves

of trees. Water transport, gas exchange and energy metabolism can be compromised,

which means suitable morphological, anatomical, physiological and biochemical

responses determine tree development and even survival under flooding.

According to their distribution, most Ulmus species prefer water-rich areas such

as sites close to rivers or streams or on floodplains (e.g., U. minor and U. laevis in

Europe, U. americana and U. rubra in North America, and U. davidiana and U.

parvifolia in Asia) (López-Almansa 2004). Flooding experiments on U. americana

exhibit many classic physiological, morphological and growth responses, such as

stomatal closure, ethylene production, formation of hypertrophied lenticels and

adventitious roots, and limited growth of leaves, stems and roots (Newsome et al.

1982). The anatomical and hydraulic features of U. laevis indicate it is a drought-

sensitive species, although it can show adaptations to waterlogged environments

(Venturas et al. 2013).


12

Fig. 5. General plant resistances to flooding.

1.2.2 General plant responses to drought

Trees develop physiological, anatomical and chemical responses to drought,

integrated at different organizational levels. Overall, responses aim at minimizing water

loss in dry soils, and tolerating a certain degree of water stress that inevitably arises

when there is an imbalance between water loss from transpiration and water uptake by

roots (Aroca et al. 2012). In conditions of disequilibrium between transpiration and

water uptake, xylem cavitation and embolism are commonly induced and result in

decreased hydraulic conductivity (Irvine et al. 1998; Plaut et al. 2012). Anatomical

characteristics of the hydraulic system control susceptibility to cavitation. In general,

compared to narrow-vessel species, those with wider vessels are more effective in

water transport capacity in well-watered conditions and more susceptible to embolism

when facing drought (Tyree and Sperry 1989; Hacke and Sperry 2001; Fonti et al.

2013).

In conditions of pervasive drought, massive xylem cavitation may cause the

failure of the hydraulic system and the death of a tree. Thus, the possession of an

efficient and/or deep root system that keeps absorbing water in water-deficient soil is

crucial for drought resistance (Leuzinger et al. 2005). Moreover, a major direct

response of trees to drought is to reduce transpiration via shedding of leaves and

closure of stomata. Upward transport of ABA from the roots leads directly to stomatal

closure (Tardieu and Davies 1993), which compromises photosynthetic net CO2 uptake.

Under severe and/or sustained drought, reduced Rubisco activity, electron transport

efficiency and chlorophyll content induce non-stomatal limitation to photosynthesis

Introduction

13

(Bota et al. 2004; Flexas and Medrano 2002; Rennenberg et al. 2006); ATP synthesis

and photophosphorylation can also be impaired (Tezara et al. 1999). Some drought-

tolerant species have a relatively higher carbon fixation capacity in drought conditions;

a higher capacity that partly comes from the ability to maintain turgor pressure and CO2

diffusion and fixation under drought (i.e. to reduce water stress under drought

conditions, see Aranda et al. 2002, 2015). Suppressed leaf formation and expansion

and induced defoliation also reduce net CO2 uptake; how this affects nonstructural

carbohydrate (NSC) reserves is less straightforward (Galiano et al. 2011). Reduced

carbon assimilation may result in reduced NSC, which can limit the metabolism and

growth of the tree. However, the down regulation of energy-demanding processes may

reduce respiratory needs and NSC consumption, in a way that reduced CO2 uptake

has no penalty in terms of NSC (see Atkin and Macherel 2009).

Drought acclimation limits are often exceeded by prolonged and/or severe

droughts, which are in fact one of the main causes of tree mortality and die-off (Prieto-

Recio et al.). Studies on tree responses to drought and the relationship between

drought and tree mortality are crucial for worldwide forest conservation under global

climate change (Allen et al. 2010). The study of McDowell et al. in 2008 has recently

renewed the interest on this topic. These authors discuss that hydraulic characteristics

and safety margins for water transport impinge on water regulation and gas exchange,

and hence on the way trees of different species die under drought. In general,

anisohydric species would be more prone to die from hydraulic failure, whereas

isohydric species that close the stomata at higher water potentials would die because

of carbon starvation. These theories have been insufficiently tested for them to be

confirmed, and many researchers have found that carbon starvation cannot be isolated

from the influences of hydraulic stress and biotic stress caused by insects and fungi

(Sala et al. 2010; McDowell 2011a, b).


14

Fig. 6. General plant resistances to drought.

There are relatively few studies that have addressed elm physiology in relation

to drought (but see for example Solla and Gil 2002). The hydraulic properties of three

European elm species in relation to xylem tension have recently been tested by

Venturas et al. (2013). These authors showed that all three species are sensitive to

xylem cavitation. However, U. minor shows anatomical and hydraulic characteristics

that make it more resistant than U. glabra and U. laevis to drought-induced xylem

cavitation. This suggests that U. minor could adapt to a drought environment better

than the other two species, but more studies are needed on U. minor drought

responses to better understand mechanisms of drought resistance.

1.2.3 General plant responses to fungal infection

The occurrence of a fungal infection is determined by the successful

establishment of a parasitic relationship between the fungus and the host plant. This

relationship is influenced by the properties of fungus, host and environment (Alvarez et

al. 2015). If any of these factors inhibits the infection process, the disease does not

occur. The infection process can be divided into three steps: pre-entry (find a suitable

host that can be parasitized), entry (break down barriers and penetrate into the host)

and colonization (absorb nutrition from host, establish the parasitic relationship) (Brown

and Ogle 1997). Pathogens can be divided into two groups, biotrophs and necrotrophs,

depending on the substrate requirements (biotrophs absorb nutrition from the living

Introduction

15

host cell, whereas necrotrophs kill the host cell to parasitize it). The interaction of plant

and pathogen may be antagonistic or symbiotic, or there may be no relationship

between them partly depending on the response of the plant to the pathogen (whether

or not the defenses activated by the plants are successful against the pathogen

infection) (Glazebrook 2005).

Once infection is established, plant physiology, anatomy and metabolism can

be affected. Infection of leaves directly limits the regulation of transpiration; vascular

infection blocks vessels and limits water transport; root infection reduces water

absorption. All these factors limit water relations, inducing stomatal closure (Ayres

1981; Manter and Kavanagh 2003; Melotto et al. 2008) and reducing photosynthesis.

Leaf defoliation by infection reduces the photosynthetic organs (Aldea et al. 2006;

Pinkard and Mohammed 2006). Fungal infection also reduces chlorophyll content from

leaf senescence and limits photochemical efficiency of PSII in chloroplasts, resulting in

a decrease in photosynthetic capacity (Oliveira et al. 2012). Respiration increases after

infection. Increased respiration indicates the change of metabolism in the host. This

can take place soon after pathogen inoculation in response to increased demand for

respiratory products to set up defense mechanisms (Landis and Hart 1972); as is the

case with oxidative burst – a resistance response that rapidly raises oxygen

consumption (Guest and Brown 1997). These responses alter normal plant function

and growth.

Plants have their own constitutive and induced methods to fight infection.

Constitutive methods are present before the challenge of the pathogen; as example,

physical barriers that prevent pathogen penetration, including the cuticle, cell walls, and

stomatal structure and function. Chemical defenses are also involved in plant

resistance to disease, to deprive the pathogen of the nutrition it requires (plant

defensins), or as toxins (Phytoanticipins) that limit its germination and development

directly by secreting or not secreting related chemical compounds (Guest and Brown

1997).

Once these constitutive defensive barriers are overcome by pathogens, the

recognition of pathogens by plant cell receptors of elicitor molecules leads to a complex

signaling cascade to activate the inducible defensive responses. Signaling hormones

such as salicylic acid (SA), jasmonic acid (JA) and ethylene (ET) play important roles in

activating induced defensive responses to pathogens (Thomma et al. 2001). These

signaling molecules may cooperate or antagonize each other to establish appropriate


16

interactions, specifically against pathogens. SA is mainly thought to mediate biotrophic

pathogens, whereas JA and ET have a defensive role against necrotrophic pathogenic

attacks (Anderson et al. 2004; Thomma et al. 2001). Some chemical defenses react

rapidly to prevent the spread of the pathogen by reinforcing the cell wall, inducing

hypersensitive cell death, or toxically killing the pathogen by phytoalexin accumulation.

Other reactions are activated relatively slowly as defense against secondary infection

and are wide-ranging, such as systemic acquired resistance. Induced defenses can be

divided into various types like i): toxic and anti-digestive chemical compounds such as

phenolic compounds, proteins (e.g. polyphenol oxidase), ii) induced defensive barriers

such as lignification and traumatic resin ducts, iii) deliberate cell death such as

hypersensitive cell death and systemic acquired resistance, and iv) indirect ecological

defense such as attracting predators of invading insects.

The severity of DED symptoms is determined by the interaction between the

fungus and the host elm tree. Research on responses to DED has been widely

reported for many elm species. Species with small vessels, which transport water less

efficiently than species with wider vessels, are considered to be more resistant to the

spread of fungus (Solla and Gil 2002). The formation of a barrier zone protects healthy

vessels from infected vessels (Martín et al. 2005). Induced biochemical compounds

such as phytoalexins are toxic compounds that limit fungus development (Duchesne et

al. 1985). Compared with anatomical and biochemical research, physiological

responses have been little studied, less so among genotypes of contrasted DED

resistance. The availability of genotypes of high resistance to DED obtained by the

Spanish Elm Breeding Program is a precious opportunity to shed light onto the

importance of functional (chemical and physiological) traits in DED resistance.

Introduction

17

Fig. 7. General plant defenses to fungus infection.


18

1.3 Thesis Outline

The above topics are addressed in this PhD thesis, as it can be seen in the

figure below. The overarching goal is to help in conservation of U. minor by better

understanding the mechanisms of response to the main abiotic and biotic factors that

shape its current distribution area.

Fig. 8. Scheme of thesis.

Introduction

19

1.4 Reference

Aldea M, Hamilton JG, Resti JP, Zangerl AR, Berenbaum MR, Frank TD, Delucia EH. 2006. Comparison of photosynthetic damage from arthropod herbivory and pathogen infection in understory hardwood saplings. Oecologia 149, 221-232.

Allen CD, Macalady AK, Chenchouni H, Bachelet D, McDowell N, Vennetier M, Kitzberger T, Rigling A, Breshears DD, Hogg EH, Gonzalez P, Fensham R, Zhang Z, Castro J, Demidova N, Lim JH, Allard G, Running SW, Semerci A, Cobb N. 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259, 660-684.

Alvarez G, Fernández M, Diez JJ. 2015. Ophiostomatoid fungi associated with declined Pinus pinaster stands in Spain. Forest System 24, 006

Anderson JP, Badruzsaufari E, Schenk PM, Manners JM, Desmond OJ, Ehlert C, Maclean DJ, Ebert PR, Kazan K. 2004. Antagonistic interaction between abscisic acid and jasmonate-ethylene signaling pathways modulates defense gene expression and disease resistance in Arabidopsis. The Plant Cell 16, 3460-3479.

Aranda I, Cano FJ, Gascó A, Cochard H, Nardini A, Mancha JA, López R, Sánchez-Gómez D. 2015: Variation in photosynthetic performance and hydraulic architecture across European beech (Fagus sylvatica L.) populations supports the case for local adaptation to water stress. Tree Physiology 35, 34-46.

Aranda I, Gil, L, Pardos JA. 2002. Physiological responses of Fagus sylvatica L. seedlings under Pinus sylvestris L. and Quercus pyrenaica Willd. overstories. Forest ecology and management 162, 153-164.

Aroca R, Porcel R, Ruiz-Lozano JM. 2012. Regulation of root water uptake under abiotic stress conditions. J Exp Bot 63, 43-57.

Atkin OK, Macherel D. 2009. The crucial role of plant mitochondria in orchestrating drought tolerance. Ann Bot 103, 581-597.

Ayres P. 1981. Responses of stomata to pathogenic microorganisms. stomatal Physiology Cambridge University Press Cambridge, pp 205-221.

Bailey-Serres J, Voesenek LA. 2008. Flooding stress: acclimations and genetic diversity. Annual review of plant biology 59, 313-339.

Bota J, Medrano H, Flexas J. 2004. Is photosynthesis limited by decreased Rubisco activity and RuBP content under progressive water stress? New phytologist 162, 671-681.

Brasier C. 1991. Ophiostoma novo-ulmi sp. nov., causative agent of current Dutch elm disease pandemics. Mycopathologia 115, 151-161.


20

Brasier C. 2000. Intercontinental Spread and Continuing Evolution of the Dutch Elm Disease Pathogens. In: Dunn C (ed) The Elms Springer US, pp 61-72.

Brown JF, Ogle H. 1997. Plant pathogens and plant diseases. Rockvale Publications

Collin E. 2003. EUFORGEN Technical Guidelines for Genetic Conservation and Use for European White elm (Ulmus laevis). Bioversity International, Rome.

Collin E, Bilger I, Eriksson G, Turok J. 2000. The Conservation of Elm Genetic Resources in Europe. In: Dunn C (ed) The Elms Springer US, pp 281-293.

Collin E, Gil L, Rusanen M, Ackzell L, de Aguiar A, Bohnens J, Diamandis S, Franke A, Harvengt L, Hollingsworth P, Jenkins G, Meier-Dinkel A, Mittempergher L, Musch B, Nagy L, Pinon J, Piou D, Rotach P, Santini A, Vanden Broeck A, Wolf H, Pâques M. 2008. Methods and progress in the conservation of elm genetic resources in Europe. 2008 13,

Collin E, Vernisson F. 2006. Conservation and utilisation of genetic resources of European elm species. Workshop on Genetics 18-22.

Colmer TD. 2003. Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant Cell And Environment 26, 17-36.

Duchesne L, Jeng R, Hubbes M. 1985, Accumulation of phytoalexins in Ulmus americana in response to infection by a nonaggressive and an aggressive strain of Ophiostoma ulmi. Canadian Journal of Botany 63, 678-680.

Flexas J, Medrano H. 2002. Drought‐ inhibition of Photosynthesis in C3 Plants:

Stomatal and Non‐stomatal Limitations Revisited. Annals of Botany 89, 183-189.

Fonti P, Heller O, Cherubini P, Rigling A, Arend M. 2013. Wood anatomical responses of oak saplings exposed to air warming and soil drought. Plant Biol (Stuttg) 15, 210-219.

Franke A, Rusanen M, de Aguiar A, Collin E, Harvengt L, Ackzell L, Bohnens J, Diamandis S. 2004. Methods and progress in the conservation of elm genetic resources in Europe. Investigación agraria. Sistemas y recursos forestales 13, 261-272.

Galiano L, Martínez‐ Vilalta J, Lloret F. 2011. Carbon reserves and canopy

defoliation determine the recovery of Scots pine 4 yr after a drought episode. New Phytologist 190, 750-759.

Gil L, Fuentes-Utrilla P, Soto A, Cervera MT, Collada C. 2004. Phylogeography: English elm is a 2,000-year-old Roman clone. Nature 431, 1053-1053.

Glazebrook J. 2005. Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol. 43, 205-227.

Introduction

21

Glenz C, Schlaepfer R, Iorgulescu I, Kienast F. 2006. Flooding tolerance of Central European tree and shrub species. Forest Ecology And Management 235, 1-13.

Gravatt DA, Kirby CJ. 1998. Patterns of photosynthesis and starch allocation in seedlings of four bottomland hardwood tree species subjected to flooding. Tree Physiology 18, 411-417.

Guest D, Brown J. 1997. Plant defences against pathogens. Plant pathogens and plant diseases 263-286.

Hacke UG, Sperry JS. 2001 Functional and ecological xylem anatomy. Perspectives in Plant Ecology, Evolution and Systematics 4, 97-115.

Heybroek HM. 1993. The Dutch elm breeding program. Dutch Elm Disease Research Springer, pp 16-25.

Hollingsworth PM, Hollingsworth ML, Coleman M. 2000. The European elms: Molecular markers, population genetics, and biosystematics. The Elms Springer, pp 3-20.

Irvine J, Perks MP, Magnani F, Grace J. 1998. The response of Pinus sylvestris to drought: stomatal control of transpiration and hydraulic conductance. Tree Physiology 18, 393-402.

Jaeger C, Gessler A, Biller S, Rennenberg H, Kreuzwieser J. 2009. Differences in C metabolism of ash species and provenances as a consequence of root oxygen deprivation by waterlogging. Journal of Experimental Botany 60, 4335-4345.

Jalas J, Suominen J. 1999. Atlas Florae Europaeae Databass.

Karnosky DF. 1979. Dutch elm disease: a review of the history, environmental implications, control, and research needs. Environmental Conservation 6, 311-322.

Kozlowski TT. 1997. Responses of woody plants to flooding and salinity. Tree Physiology 17, 490.

López-Almansa JC. 2004. Review. Reproductive ecology of riparian elms. Investigación agraria. Sistemas y recursos forestales 13, 17-27.

Landis WR, Hart JH. 1972. Physiological Changes in Pathogen-Free Tissue of Ulmus Americana Induced by Ceratocystis Ulmi. Phytopathology 62, 909-913.

Leuzinger S, Zotz G, Asshoff R, Korner C. 2005. Responses of deciduous forest trees to severe drought in Central Europe. Tree Physiology 25, 641-650.

Manter DK, Kavanagh KL. 2003. Stomatal regulation in Douglas fir following a fungal-mediated chronic reduction in leaf area. Trees - Structure and Function 17, 485-491.


22

Martín JA, F-UP, Gil L., Witzell J. 2010. Ecological factors in Dutch elm disease complex in Europe – a review. Ecological Bulletins 53, 209-224.

Martín JA, Solla A, Gil L, García-Vallejo MC. 2010. Phenological and histochemical changes of Ulmus minor due to root absorption of phenol: Implications for resistance to DED. Environmental and Experimental Botany 69, 175-182.

Martín JA, Solla A, Woodward S, Gil L. 2005. Fourier transform-infrared spectroscopy as a new method for evaluating host resistance in the Dutch elm disease complex. Tree Physiology 25, 1331-1338.

Martín JA, Solla A, Venturas M, Collada C, Domínguez J, Miranda E, Fuentes P, Burón M, Iglesias S, Gil L. 2014. Seven Ulmus minor clones tolerant to Ophiostoma novo-ulmi registered as forest reproductive material in Spain. iForest - Biogeosciences and Forestry e1-e9.

McDowell N, Pockman WT, Allen CD, Breshears DD, Cobb N, Kolb T, Plaut J, Sperry J, West A, Williams DG, Yepez EA. 2008. Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytol 178, 719-739.

McDowell NG. 2011. Mechanisms linking drought, hydraulics, carbon metabolism, and vegetation mortality. Plant Physiology 155, 1051-1059.

McDowell NG, Beerling DJ, Breshears DD, Fisher RA, Raffa KF, Stitt M. 2011. The interdependence of mechanisms underlying climate-driven vegetation mortality. Trends Ecol Evol 26, 523-532.

Melotto M, Underwood W, He SY. 2008. Role of stomata in plant innate immunity and foliar bacterial diseases. Annual Review of Phytopathology 46, 101.

Mittempergher L, Santini A. 2004. The history of elm breeding. Forest Systems 13, 161-177.

Newbanks D, Bosch A, Zimmermann MH. 1983. Evidence for xylem dysfunction by embolization in Dutch elm disease. Phytopathology 73, 1060-1063.

Newsome RD, Kozlowski TT, Tang ZC. 1982. Responses of Ulmus americana seedlings to flooding of soil. Canadian Journal of Botany 60, 1688-1695.

Nicolás E, Torrecillas A, Dell’Amico J, Alarcón J. 2005. The effect of short-term flooding on the sap flow, gas exchange and hydraulic conductivity of young apricot trees. Trees 19, 51-57.

Oliveira H, Sousa A, Alves A, Nogueira AJA, Santos C. 2012. Inoculation with Ophiostoma novo-ulmi subsp. americana affects photosynthesis, nutrition and oxidative stress in in vitro Ulmus minor plants. Environmental and Experimental Botany 77, 146-155.

Introduction

23

Ouellette G, Rioux D, Simard M, Cherif M. 2004. Ultrastructural and cytochemical studies of host and pathogens in some fungal wilt diseases: retro-and introspection towards a better understanding of DED. Forest Systems 13, 119-145.

Pajares J. 2004. Elm breeding for resistance against bark beetles. Forest Systems 13, 207-215.

Pezeshki SR. 2001. Wetland plant responses to soil flooding. Environmental And Experimental Botany 46, 299-312.

Pimenta JA, Bianchini E, Medri ME. 2010. Adaptations to flooding by tropical trees: morphological and anatomical modifications. Oecologia Australis 158-176.

Pinkard E, Mohammed C. 2006. Photosynthesis of Eucalyptus globulus with Mycosphaerella leaf disease. New Phytologist 170, 119-127.

Plaut JA, Yepez EA, Hill J, Pangle R, Sperry JS, Pockman WT, McDowell NG. 2012. Hydraulic limits preceding mortality in a pinon-juniper woodland under experimental drought. Plant, Cell and Environment 35, 1601-1617.

Prieto-Recio C, Martín-García J, Bravo F, Diez JJ. 2015. Unravelling the associations between climate, soil properties and forest management in Pinus pinaster decline in the Iberian Peninsula. Forest Ecology and Management.

Rennenberg H, Loreto F, Polle A, Brilli F, Fares S, Beniwal RS, Gessler A. 2006. Physiological responses of forest trees to heat and drought. Plant Biol (Stuttg) 8, 556-571.

Sala A, Piper F, Hoch G. 2010. Physiological mechanisms of drought-induced tree mortality are far from being resolved. New phytologist 186, 274-281.

Smalley EB, Guries RP. 2000. Asian elms: sources of disease and insect resistance. The Elms Springer, pp 215-230.

Solla A, Bohnens J, Collin E, Diamandis S, Franke A, Gil L, Buron M, Santini A, Mittempergher L, Pinon J, Broeck AV. 2005. Screening European elms for resistance to Ophiostoma novo-ulmi. Forest Science 51, 134-141.

Solla A, Burón M, Iglesias S, Gil L. 2000. Spanish Program for the Conservation and Breeding of Elms Against DED. In: Dunn C (ed) The Elms Springer US, pp 295-303.

Solla A, Gil L. 2002. Influence of water stress on Dutch elm disease symptoms in Ulmus minor. Canadian Journal of Botany 80, 810-817.

Tardieu F, Davies W. 1993. Integration of hydraulic and chemical signalling in the control of stomatal conductance and water status of droughted plants. Plant Cell Environ 16, 341-349.


24

Tezara W, Mitchell V, Driscoll S, Lawlor D. 1999. Water stress inhibits plant photosynthesis by decreasing coupling factor and ATP. Nature 401, 914-917.

Thomma BP, Penninckx IA, Broekaert WF, Cammue BP. 2001. The complexity of disease signaling in Arabidopsis. Curr Opin Immunol 13, 63-68.

Tyree MT, Sperry JS. 1989. Vulnerability of xylem to cavitation and embolism. Annual review of plant biology 40, 19-36.

Venturas M, Fernandez V, Nadal P, Guzman P, Lucena JJ, Gil L. 2014. Root iron uptake efficiency of Ulmus laevis and U. minor and their distribution in soils of the Iberian Peninsula. Front Plant Sci 5, 1-9.

Venturas M, Fuentes-Utrilla P, López R, Perea R, Fernández V, Gascó A, Guzmán P, Li M, Rodríguez-Calcerrada J, Miranda E, Domínguez J, González-Gordaliza G, Zafra E, Fajardo-Alcántara M, Martín JA, Ennos R, Nanos N, Lucena JJ, Iglesias S, Collada C, Gil L. 2015. Ulmus laevis in the Iberian Peninsula: a review of its ecology and conservation. iForest - Biogeosciences and Forestry 8, 135-142.

Venturas M, López R, Gascó A, Gil L. 2013. Hydraulic properties of European elms: xylem safety-efficiency tradeoff and species distribution in the Iberian Peninsula. Trees 27, 1691-1701.

Yamamoto F, Sakata T, Terazawa K. 1995. Physiological, morphological and anatomical responses of Fraxinus mandshurica seedlings to flooding. Tree Physiology 15, 713-719.

26

2. AIMS

Aims

28

2. AIMS

The overarching goal of the present PhD thesis is to continue past research

efforts on U. minor conservation. Overall, so far, these efforts have focused on the

identification of resistant genotypes, their propagation in situ, and the identification of

the factors responsible for their resistance. More research is needed and planned. The

future of U. minor conservation demands investigating the pathogenicity of O. novo-

ulmi, the resistance of U. minor to DED from genomic and functional approaches, and

the propagation of DED-resistant genotypes through large-scale plantations. Relatively

little information exists on U. minor physiology, less so in response to abiotic and biotic

stresses. Thus, this PhD thesis aims to extend existing knowledge of the resistance

and survival mechanisms of U. minor to flooding and drought, two classic abiotic

stresses that U. minor experience in its habitat, and to DED, the most serious biotic

stress that threatens this species. The information should prove useful for DED

resistance breeding and artificial regeneration of U. minor.

To achieve this, a series of experiments have been conducted with U. minor, in

comparison with other tree species (i.e. U. laevis and Q. ilex). Specific objectives and

main hypotheses of each experiment follow (with more details given in the introduction

of the respective chapters 3, 4 and 5):

a) To evaluate and understand mechanisms of flood resistance of U. minor, in

comparison with U. laevis.

To this end, two-year-old seedlings of these species were immersed in a pool

with running water for 60 days, and anatomical and physiological variables

related with carbon balance and water relations compared with well-watered

controls during and after seedlings’ immersion (to assess recovery from flooding).

The main hypothesis is that lower flood-tolerance of U. minor than U. laevis is

reflected in a higher impact of seedlings’ immersion on gas exchange and plant

hydraulics.

b) To evaluate and understand mechanisms of drought resistance of U. minor, in

comparison with Q. ilex.

To this end, two-year-old seedlings of these species cultivated in 16-l pots were

subjected to a long period of drought (i.e. no watering) and compared with well-


29

watered controls in terms of gas exchange, hydraulic conductivity and non-

structural carbohydrate (NSC) concentrations. The idea of comparing Q. ilex and

U. minor was to shed light into the arguable topic of drought-induced plant

mortality by comparing two species of highly contrasted drought resistance. The

main hypotheses are that lower drought-tolerance of U. minor than Q. ilex is

reflected in a higher impact of watering cessation on gas exchange and plant

hydraulics, and that xylem hydraulic failure has a primary role on unchaining

plant death.

c) To evaluate and understand mechanisms of DED resistance of several U. minor

genotypes.

To this end, five- (and six-) year-old saplings of six genotypes of U. minor of

variable DED resistance were inoculated with either O. novo-ulmi or water and

physiological and biochemical variables measured over the next 30 days. The

main hypothesis is that different DED resistance among clones is reflected in a

different physiological and biochemical profile either before or after inoculation.

31

3. GREATER RESISTANCE TO FLOODING OF SEEDLINGS OF Ulmus

laevis THAN Ulmus minor IS RELATED TO THE MAINTENANCE OF

A MORE POSITIVE CARBON BALANCE

Greater resistance to flooding of seedlings of Ulmus laevis than Ulmus minor is related to the maintenance of a more positive carbon balance

33

Greater resistance to flooding of seedlings of Ulmus laevis than Ulmus

minor is related to the maintenance of a more positive carbon balance

Meng Li, Rosana López, Martín Venturas, Pilar Pita, Guillermo G. Gordaliza, Luis Gil,

Jesús Rodríguez-Calcerrada

Published in Trees 29: 835-848


34

Abstract

Flooding affects plant physiology and development, ultimately determining

species’ ecology and distribution. Ulmus laevis Pallas and Ulmus minor L. are two

European riparian trees facing habitat degradation and Dutch elm disease. Here, we

have investigated the sensitivity to flooding of two-year-old seedlings of these species

to ascertain their level of tolerance in relation to future reforestations. Gas exchange of

leaves, stems and roots, hydraulic conductivity and growth were measured in a

controlled experiment. Seedlings of U. minor died by the 60th day of flooding, but not

those of U. laevis, which partly recovered physiological functions after 30 days of

adequate watering. Light-saturated net photosynthesis rate (Pn) and stomatal

conductance progressively declined after flooding started. Forty six days later, Pn was 2

and 3 times lower in flooded compared to control U. laevis and U. minor plants,

respectively; at this time, the percentage loss of root hydraulic conductivity increased

by 4-fold relative to control plants. Rates of respiration initially increased with flooding

in leaves, stems and roots, and then were similar in flooded and control plants.

Aerenchyma was not formed on either species but lenticels at the water line became

increasingly hypertrophied and could help in providing oxygen and sustaining

respiration. Whole-plant net carbon gain was 3 and 9 times lower in flooded than

control plants in U. laevis and U. minor, respectively. Our data suggest that the inability

to maintain a positive carbon balance somehow compromises seedling survival under

flooding, earlier in U. minor than U. laevis, partly explaining their differential habitats.

Keywords: waterlogging, plant mortality, anatomy, xylem cavitation, carbon budget


35

Introduction

Perpendicular to terrestrial water bodies, there is a marked environmental

gradient that shapes the zonation of riparian vegetation (Glenz et al. 2006). Many of

the environmental factors that vary along this gradient such as the depth of the

groundwater table and the duration and the frequency of floods have been shown to

play an important role in the distribution of tree species (Ferreira and Stohlgren 1999;

Buijse et al. 2002; Glenz et al. 2006). Thus, the ecophysiological response to

experimental flooding may help in explaining natural distribution of species (Ferreira et

al. 2009; Dat and Parent 2012), selecting adequate species for riverbank protection

(Evette et al. 2012), and determining species sensitivity to flooding (Piper et al. 2008)

Flooding alters soil physicochemical properties, since water occupies soil pores

which are otherwise filled with air (Pezeshki 2001). Because oxygen has approximately

10,000 times slower diffusivity in water than in air, reductions in oxygen concentration

and redox potential take place in waterlogged and flooded soils (Ponnamperuma 1972;

Parent et al. 2008). The availability of nutritional elements such as phosphorus

decreases, while CO2, sulfides, methane, ethylene and heavy metals such as

manganese are accumulated in the soil (Kozlowski 1997). These alterations in the soil

gas composition provoke a cascade of molecular to morphological changes that affect

growth and can eventually kill the plants (Kozlowski 1997; Striker 2012).

Soil oxygen deficiency and increased CO2 concentration can inhibit root aerobic

respiration and impede normal root functioning (Colmer 2003; Kreuzwieser et al. 2004;

Maček et al. 2005; Bailey-Serres and Voesenek 2008; Jaeger et al. 2009). Water

absorption and hydraulic conductivity can decline and cause water deficits in the plants

(Kozlowski 1997; Nicolás et al. 2005), partly due to suberization of roots, changes in

aquaporin expression and activity, and altered cambial growth (Else et al. 2001; Javot

and Maurel 2002; Tournaire-Roux et al. 2003; Islam and Macdonald 2004). However,

other studies report no symptoms of plant water deficit under flooding (Ismail and Noor

1996). The stomata play an important role in the regulation of plant water homeostasis

(Else et al. 2001; Aroca et al. 2012a). Induced by an accumulation of abscisic acid

((Bradford and Hsiao 1982; Jackson and Hall 1987; Castonguay et al. 1993; Zhang and

Zhang 1994) or other type of anti-transpirant compound (Else et al. 1996) stomatal

closure can occur, even in the absence of previous water deficits, and improve the

water status of flooded plants. Moreover, tolerant plants could maintain water uptake

under conditions of hypoxia/anoxia (McKee 1996; Striker 2012).


36

Photosynthesis will decline during flooding due to stomata closure and

reduction in CO2 diffusion to the chloroplasts (Kozlowski 1997; Gravatt and Kirby 1998b;

Li et al. 2007). Continued flooding may also reduce chlorophyll concentration,

carboxylation activity, photosynthetic electron transport efficiency and, therefore,

carbon fixation (Kozlowski 1997; Pezeshki 2001). This, together with the reduced

translocation of photosynthates from the leaves to the stems and roots (Yamamoto et

al. 1995; Angelov et al. 1996; Kozlowski 1997; Pezeshki and Santos 1998), can make

soluble sugars reach limiting concentrations for sustaining respiration of all plant

organs. Compared to photosynthetic rates, the response of respiration rates to flooding

has been less studied. With increasing oxygen deficiency, a shift may occur from

aerobic respiration to anaerobic fermentation of glucose (Kennedy et al. 1992;

Vartapetian and Jackson 1997). Fermentation produces 16 times less energy than

normal oxidative phosphorylation occurring in aerobic conditions (Vartapetian and

Jackson 1997; Dat and Parent 2012), so that a deficit in energy production can occur

and make plants more sensitive to flooding and other concurrent stresses or pathogens

(Munkvold and Yang 1995; Yanar et al. 1997). Maintaining a positive balance between

carbon uptake and release can thus be crucial for survival. Sometimes recently

assimilated photosynthates are insufficient to keep up with the metabolic activity of the

plant and a depletion of nonstructural carbohydrate reserves occurs (Anderson and

Pezeshki 2000; Parent et al. 2008). However, the down-regulation of energy-

demanding processes can attenuate the depletion of carbon reserves and ensure

glucose supply to fermentation over longer periods. Maintaining a positive carbon

balance favors plants’ recovery from flooding stress, with stored carbohydrates being

used in repairing or replacing damaged structures (Blom and Voesenek 1996; Blanke

and Cooke 2004).

Plants can acclimate to relatively long periods of flooding by modifying some

morphological and anatomical features. The hypertrophy of lenticels or the formation of

aerenchyma facilitate plant oxygenation and contribute to maintain root aerobic

respiration and water uptake under flooding (Yamamoto et al. 1995; Blom and

Voesenek 1996; Kozlowski 1997). The development of adventitious roots may replace

submerged roots when they become dysfunctional (Yamamoto et al. 1995; Glenz et al.

2006). Growth is another way plants have to acclimate to flooding stress, for example

by producing shoots emerging from the water line. The response of plant growth to

flooding is variable, with flooded plants showing either higher or lower growth than non-

flooded ones (Newsome et al. 1982b; Kozlowski 1997; Ye et al. 2003; Gérard et al.


37

2008). Often, phloem and xylem are affected differently by flooding, which can restrict

xylem growth but not stem diameter due to swelling or proliferation of parenchyma in

the phloem (Kozlowski 1997).

The area of distribution of riparian trees Ulmus minor L. and Ulmus laevis Pallas

expands from the Iberian Peninsula in the south of Europe to Russia and the

Scandinavian countries in the north. Dutch elm disease (DED) has decimated U. minor

populations throughout Europe, but long-term conservation programs have allowed the

propagation of Spanish DED-resistant U. minor genotypes which are planned to be

used in reforestation programs in the near future. This species grows in lowland

floodplains, but it is also found in drought-prone sites (Venturas et al. 2014b). Actually,

in the Iberian Peninsula, U. minor is the riparian tree most tolerant to water table

depletions (Castro 1997). U. laevis is less susceptible to DED than U. minor; however,

it faces the degradation of its habitat, typically the riparian deciduous forests suffering

long term floods (Collin et al. 2004; Venturas et al. 2014c). Assessing seedling

vulnerability to flooding is necessary if large-scale reforestations are to be made with

these species in flood-prone areas.

The aim of this study was to compare the sensitivity of seedlings of U. laevis

and U. minor to flooding. Based on their distribution in riverside forests, our working

hypothesis is that U. laevis is more tolerant to flooding than U. minor. To test this

hypothesis and the effects of flooding on plant function, we have compared the

ecophysiological response of two-year-old seedlings of both species to an experimental

flood lasting two months, followed by another month of normal watering to test for

potential recovery. We evaluated flood impacts on growth and anatomical factors, and

on physiological factors related with plant carbon balance (i.e. leaf net photosynthesis

and leaf, stem and root respiration) and water status (i.e. leaf water potential, and root

and stem hydraulic conductance), as all these aspects are key to plant survival.

Material and methods

Experimental design

This experiment was conducted with two-year-old seedlings of Ulmus laevis and

Ulmus minor, in the campus of the School of Forestry, Madrid, Spain (3º 43’ W, 40º 27’

N). Seeds from several trees were sown in 15-litres plastic pots containing a substrate

of peat and sand mixed 1 to 3 in volumetric proportions. Seedlings were grown under


38

partial shading and regular watering for the first year. In late spring of the second year,

30 healthy plants from each species were randomly selected and moved to another

plot with a pool with running water. Half of the plants were immersed in the pool, with

the basal 2-3 cm of the stem permanently covered by water (F plants), and the other

half were placed next to the pool and kept well watered throughout the experimental

period as controls (C plants). The pots in the pool were drilled in the upper part to allow

water to flow through.

The flooding period lasted for 60 days, from June 10th to August 10th 2013.

Plants of U. minor had died by the 60th day of flooding, but not those of U. laevis, which

were taken out of the pool and kept well-watered for 30 days to assess recovery from

flooding effects. We measured water temperature, pH and oxygen concentration twice

over the experimental course; values barely changed over time and were, on average,

23ºC, 8.1, and 7.8 mg O2/l, respectively. Average air temperature and relative humidity

were 25 ºC and 41 %, respectively, over the experimental course.

Leaf water potential at midday and gas exchange were measured repeatedly on

five plants per species and treatment, more frequently at the beginning of the

experiment and once per week afterwards. At days 12, 49 and 93 after the start of the

experiment, we measured respiration and hydraulic conductance in stems and coarse

roots. Each time plants were measured over two days, from 10.00 to 16.00 h. Plants

were moved from the pool to the laboratory in sets of four, left for 1-2 h at moderate

sunlight and approximately 25ºC (the same as the measurement temperature), and

then harvested. In addition, anatomical properties at the root collar were examined in

both species and treatments on days 12 and 49 of the experiment.

Leaf water potential

Leaf water potential was measured with a pressure chamber (PMS Instrument

Company, Albany, OR, USA) in the same leaves sampled for gas exchange

measurements. Leaves were cut and then immediately placed into the head of the

pressure chamber to be measured.

Gas exchange


39

Gas exchange was measured in fully-expanded leaves with a LI-6400 portable

gas exchange system (Li-Cor Inc., NE, USA), equipped with a 6-cm2 chamber coupled

to a LED light source (LI-6400-02B). Measurements were conducted between 10:00

and 13:00 h, at ambient air relative humidity, but artificially adjusted micro-

environmental conditions of 1500 mol m−2 s-1 photon flux density, 400 ppm CO2

concentration and 25ºC. Light-saturated net photosynthesis rate (Pn), stomatal

conductance to water vapor (gs), internal CO2 concentration (Ci) and transpiration rate

(E) were calculated for the exact leaf area included in the chamber. Dark respiration

rate (Rd) was then measured in the same leaves. Prior to measurements, leaves were

covered with aluminum foil paper for 30 minutes to prevent light-enhanced respiration

(Atkin et al. 1998). Area-based gas exchange rates were converted to mass units using

the mass per area of measured leaf.

Respiration in stems and coarse roots was measured in a conifer chamber (LI-

6400-05) attached to the portable gas exchange system. Measurements were

performed on severed fragments at 400 ppm CO2 concentration (using the LI-6400-01

CO2 injector), 25ºC temperature and 35-55% air relative humidity. All measurements

were completed within 10 minutes of cutting. Previous tests showed the rate of

respiration in stems and coarse roots was relatively stable soon after cutting (Fig. 1)


40

Fig. 1. Tests checking for temporal variation

of mass-based root and stem respiration. The

dashed line indicates the cutting time.

Measurements were completed within 10

minutes of cutting, at which time, relative to

all averaged previous values before cutting,

stem respiration had changed by 1.07%, -

11.61% and 0.97% in the 3 sampled trees,

respectively, and root respiration by -0.88%, -

2.55% and 2.11%.

Shifts in the rate of respiration of fine roots after cutting were faster and so we

decided not to measure fine roots. The two cut ends of stem and coarse root fragments

were sealed with putty (Terostat; Teroson, Heidelberg, Germany) to avoid CO2 release

through the open ends (Teskey and McGuire 2004) and minimize heterogeneity

between samples. The chamber foam gaskets were also covered with putty to prevent

leaks. The rate of respiration of stems and coarse roots was expressed per unit of

surface area and dry mass. Considering both organs as perfect cylinders, their curved

surface area was calculated from the diameter and height. The ratio of curved surface

area to oven-dry mass was used to convert area-based respiration rates to mass-

based rates.

Hydraulic conductance


41

Xylem hydraulic conductance was measured in the main stem and root of each

plant using a XYL’EM embolism-meter (Bronkhorst, Montigny les Cormeilles, France).

The stem and main root were cut under water to avoid air entering through the cut

surfaces. Then, root segments and current-year and two-year-old stem segments of

approximately 30 mm length were re-cut under water, removing the bark before

measuring. Initial hydraulic conductivity (Ki) was determined as the water mass flow per

pressure gradient across the length of segment, using a solution of 10 mM KCl and

1mM CaCl2 in deionized degassed water. After measuring Ki, segments were flushed

with the same solution at 0.18 MPa for 10 minutes to remove embolisms, and hydraulic

conductivity was measured again to obtain the maximum hydraulic conductivity through

the segment (Kmax). The percentage loss of hydraulic conductivity (PLC), was

calculated as: PLC (%) = 100*(Kmax – Ki)/Kmax. The cross-sectional area (S) and length

(L) of segments were also measured to calculate the specific maximum hydraulic

conductivity (Ksmax) as: Ksmax = Kmax*L/S.

Anatomy

Transverse sections of stems at the root collar were cut using a sliding

microtome. They were stained with an aqueous solution of 1% safranine and 1% alcian

blue, and mounted for optical microscopy. Photographs were taken with a digital

camera Canon EOS 600D with a LM microscopy adapter (MicroTech Lab). Images

were analyzed with WinCELL Pro (version 2004a; Regent Instruments Inc. Canada) for

measuring aerenchyma proportion, earlywood and latewood width, vessel frequency

(Vf, vessels mm-2), mean vessel area (Va, µm2) and percentage of grouped vessels

(PGV, %) in earlywood and latewood.

Growth

Five plants from each treatment and species were selected to measure main

stem diameter increment continuously from June 12th to July 7th with linear variable

displacement transducers (LVDTS, Model DF2.5, Solartron Metrology Bognor Regis

U.K.). The number of new leaves formed since the start of the experiment was also

registered 33 and 48 days after the start of the experiment.

Plant biomass distribution


42

Leaves, stems and roots of plants sampled for hydraulic conductivity were

oven-dried and weighted to obtain total leaf (LDM), stem (SDM) and root (RDM) dry

masses.

From data of plant biomass distribution, the instantaneous daytime whole plant

carbon balance (PCB) was calculated as:

PCB = LDM ∙ 𝑃n − SDM ∙ 𝑅s − RDM ∙ 𝑅r,

with Pn already including the respiratory carbon losses occurring through

photorespiration and mitochondrial respiration in the light.

Data analyses

This experiment was arranged as a completely randomized experimental design.

An independent sample t-test was used to compare means between the two treatments

(flooding and control), with 5 replicates (n=5) for each species respectively. Analyses

were performed with SPSS 17.0, with 95% confidence limits (p<0.05).

Results

Leaf water potential

Midday leaf water potential (md) of flooded U. laevis plants was significantly

higher than that of control plants from day 7 until just before the end of the flooding

period (Fig. 2). Thirty days after flooding ended, the values of md were similar to those

exhibited in previous days. However, no significant differences in md were found

between treatments because md of control plants increased, for unknown reasons. In

the case of U. minor there was no significant difference in md between flooded and

control plants until day 24, when flooded plants started to show significantly higher md

than controls.


43

Fig. 2. Time course of midday water potential (md) in plants of U. laevis and U. minor subjected to

regular watering (C; black symbols) or flooding (F; white symbols). Symbols are means ± standard

error (n=5). The dashed line indicates the end of flooding. Notice the different time scale on the X-

axis between species.

Gas exchange in leaves, stems and roots

Flooding caused a significant reduction in Pn and gs from day 17 on in both

species (Fig. 3). By day 46 after pot submersion, Pn and gs were approximately 2 and

3.5 times lower, respectively, in flooded than control plants of U. laevis, and 3 and 4.5

times lower, respectively, in flooded than control plants of U. minor. Then Pn and gs of

flooded U. laevis continued to decrease and reached a level approximately 4 times and

7 times lower than in control plants by day 60. Leaf transpiration (E) and internal CO2

concentration (Ci) developed similar trends over time as Pn and gs during the flooding

period (Fig. 3). This suggests the decline in Pn was caused by a reduction in gs in

response to flooding, which also caused reductions in E and Ci. Rates of Pn in U. laevis

plants moved out of the pool for 30 days were similar to those of control plants, gs and

E rates being significantly higher.


44

Fig. 3. Time course of leaf net photosynthesis (Pn), stomatal conductance to water vapor (gs),

transpiration (E) and internal CO2 concentration (Ci) in plants of U. laevis and U. minor subjected to

regular watering (C; black symbols) or flooding (F; white symbols). Symbols are means ± standard

error (n=5). The dashed line indicates the end of flooding. Notice the different time scale on the X-

axis between species.

Concerning respiration, significant differences between the two treatments in

leaf, stem and root respiration were found at the beginning of the experiment (day 11)

in U. laevis. Respiration of leaf, stem and root of flooded plants had 1.5 times, 2.1


45

times and 2 times higher values than control plants, respectively (Fig. 4). Flooded

plants of U. minor also had higher rates of root and stem respiration than control plants

by day 11, although in the case of the stem, the difference was only marginally

significant (p =0.11; Fig. 4).

Fig. 4. Leaf dark (Rl), stem (Rs) and root (Rr) respiration in three dates of the experimental course in

plants of U. laevis and U. minor subjected to regular watering (C; black symbols) or flooding (F;

white symbols). The last date corresponds to plants examined for recovery 30 days after the

treatment of flooding; recovery could not be examined in U. minor because flooded plants died.

Symbols are means ± standard error (n=5).

Whole-plant instantaneous net carbon gain was severely limited after 49 days of

flooding (Fig. 5). The balance between carbon gain by photosynthesis and carbon

losses through respiration was always positive in all plants, but it was close to zero in


46

flooded U. minor. Previously flooded plants of U. laevis did not reach the values of

controls at the end of the recovery period because, despite the recovery of

photosynthetic rates, flooded plants had barely produced any more leaves as

compared to control ones.

Fig. 5. Balance between foliage photosynthetic CO2 uptake and respiratory CO2 release via foliage,

stems and roots. Symbols are means ± standard error (n=5) for plants of U. laevis and U. minor

subjected to regular watering (C; black symbols) or flooding (F; white symbols). The last date

corresponds to plants examined for recovery 30 days after the treatment of flooding; it could not

be examined in U. minor because flooded plants died

Stem and root hydraulic conductance

Maximum hydraulic conductivity (Kmax) of root was similar in flooded and control

plants of both species (on average 7.3 kg m-1 s-1 MPa-1 for U. laevis and 9.8 kg m-1 s-1

MPa-1 for U. minor). The same lack of difference between treatments was found for

two-year-old shoots (on average, 5.9 kg m-1 s-1 MPa-1 for U. laevis and 7.5 kg m-1 s-1

MPa-1 for U. minor). However, one-year-old shoots of plants subjected to flooding for 49

days had half the Kmax of controls in both species. For U. laevis, Kmax was 2.3 kg m-1 s-1

MPa-1 in flooded plants and 4.3 kg m-1 s-1 MPa-1 in controls, although the difference was

not significant at p < 0.05; for U. minor, Kmax was 4.3 kg m-1 s-1 MPa-1 in flooded plants

and 8.9 kg m-1 s-1 MPa-1 in controls (p < 0.05) (Fig. 6).


47

Fig. 6. Maximum hydraulic conductivity (Kmax) of one-year-old shoots, two-year-old shoots and

roots of U. laevis and U. minor subjected to regular watering (C; black symbols) or flooding (F;

white symbols). The last date corresponds to plants examined for recovery 30 days after the

treatment of flooding; it could not be examined in U. minor because flooded plants died.

The percent loss of hydraulic conductivity (PLC) of the root was severely

affected by flooding, and was approximately 4 times higher in flooded than control


48

plants of both species (Fig. 7). No significant differences in PLC were found in stems

for flooded and control plants for either species, average values ranging from 16 to 26%

in one-year-old stems and from 49 to 60% in two-year-old stems.

Fig. 7. Percent loss of hydraulic conductivity in roots (PLCs) of U. laevis and U. minor subjected to

regular watering (C; black symbols) or flooding (F; white symbols). The last date corresponds to

plants examined for recovery 30 days after the treatment of flooding; it could not be examined in U.

minor because flooded plants died.

Stem morphology and anatomy

Lenticels in the main stem became increasingly hypertrophied in both elms, this

effect being visible as soon as one week after pot submersion (Fig. 8).


49

Fig. 8. Hypertrophied lenticels (above) and degradation of the stem (below) of U. minor at day 60th

.

Some anatomical changes were also observed with flooding. The width of the

latewood formed by U. laevis was significantly lower in flooded than control plants, but

no such difference was observed in U. minor. We did not find any difference in vessel

frequency between treatments on any species. However, flooded plants of U. minor

had bigger vessels, and these vessels did more often appear isolated rather than in

groups, as compared to control plants. These variables exhibited no difference

between treatments on U. laevis (Fig. 9). No differences in earlywood were observed

because it was already formed when the experiment started.


50

Fig. 9. Growth, mean vessel area (Valatewood) and percentage of grouped vessels (Pgvlatewood) in

latewood of stems at root collar in plants of U. laevis and U. minor subjected to regular watering or

flooding for 11 and 49 days.

Plant growth and survival

Stem diameter increment monitored daily with LVDTs was higher in flooded

than control plants of both species over the first month of the experiment (Fig. 10).


51

Fig. 10. Daily stem diameter increment in plants of U. laevis and U. minor subjected to regular

watering (C; black symbols) or flooding (F; white symbols). Symbols are means ± standard error

(n=5).

Over the same period, however, submerged plants of U. laevis and U. minor

produced, on average, 2 and 10 leaves, respectively, while control plants produced 15

and 54 leaves. Later on, submerged plants of U. laevis and U. minor did not produce

any more leaves while control plants of both species continued to produce new leaves.

Actually, the 5 plants of U. minor remaining in the pool at the end of the period of

submersion started to shed their leaves and eventually suffered stem dieback; outer

bark and phloem degraded forming a ring of visible xylem at the water line (Fig. 8)

which ultimately impeded carbohydrate translocation to the roots. These plants did not

recover once normal watering resumed, and in the year following the experiment, just

one of the five plants of U. minor subjected to flooding sprouted from the root. On the

contrary no plant of U. laevis died or shed leaves during the experiment or afterwards.

Total plant biomass was not significantly different between treatments, except in

U. laevis at the end of the experiment, when control plants had two times more

biomass than the ones that had been previously exposed to flooding (Fig. 11).


52

Fig. 11. Total plant dry mass of U. laevis and U. minor subjected to regular watering (C; black

symbols) or flooding (F; white symbols). Symbols are means ± standard error (n=5).

Discussion

Here we have evaluated the anatomical and functional alterations taking place

in seedlings of two Ulmus species over three months of experimental flooding and

potential recovery, to compare their susceptibility to flooding. Our results showed a

rapid and progressive decline of leaf stomatal conductance and net photosynthesis,

likely triggered by hormonal signals at the beginning of flooding and hydraulic

limitations later on. Leaf, stem and root respiration of plants exposed to flooding initially

increased, and then remained similar to values of control, well-watered plants. The

result was a clear decline in whole-plant net carbon gain, of higher magnitude in U.

minor than U. laevis. On continuation we further discuss these physiological responses

and the implications for flood resistance in these species.

Leaf stomatal conductance and transpiration declined, while stem and root

hydraulic conductivity were not affected within two weeks of exposure to flooding. This

suggests the initial decline in leaf gas exchange was induced by a hormonal signal that

attenuated leaf water deficit (Bradford and Hsiao 1982; Kozlowski 1997; Islam et al.

2003), at least during the first weeks of flooding. Hormones synthesized in the leaves

or transported from the roots to the leaves can promote stomata closure. This is the

case with ethylene (Islam et al. 2003; Blanke and Cooke 2004; Desikan et al. 2006)

and ABA (Else et al. 1996; Yanar et al. 1997; Loewenstein and Pallardy 1998),

although works of (Else et al. 1996; Else et al. 2001; Else et al. 2009) have found a


53

reduction of ABA delivery from roots together with stomatal closure and discussed

some unknown compounds other than ABA could participate in the flood-induced

closing of stomata. The decline in transpiration probably explained that leaf water

potential at midday was higher than in well watered plants. In the longer term, hydraulic

limitations to water transport may reduce stomatal conductance and transpiration

(Tyree and Sperry 1988). After 49 days of flooding, the percentage loss of conductivity

(PLC) of roots reached approximately 40% in U. laevis and 60% in U. minor, that is,

approximately four times higher than control plants. However, root hydraulic

conductivity was still within the range of stem hydraulic conductivity, suggesting root

cavitation was not limiting water supply to leaves. Anatomical traits did not provide a

straightforward explanation of differences in xylem susceptibility to cavitation. Flooded

U. minor plants had a higher proportion of isolated vessels than control plants, but the

relationship between vessel connectivity and hydraulic safety is not clear. Some

authors discuss that a greater number of connections allow for alternative pathways for

water transport (Yáñez-Espinosa et al. 2001), while others consider that embolism can

spread more easily through connected vessels (Johnson et al. 2014). In any case,

(Yáñez-Espinosa et al. 2001) also reported an increasing proportion of isolated vessels

as trees grew in sites of increasing duration of flooding.

The decline in stomatal conductance was the main factor limiting

photosynthesis in flooded plants, as it has been observed in similar studies (Kozlowski

1997; Gravatt and Kirby 1998b; Nicolás et al. 2005). Non stomatal limitations of the

photosynthetic capacity can occur in relation to reduced chlorophyll concentration,

carboxylation efficiency, or electron transport rate (Gravatt and Kirby 1998b; Herrera et

al. 2008; Polacik and Maricle 2013). Here, however, the internal concentration of CO2

remained generally lower in flooded plants, suggesting that photosynthesis was mostly

limited by the availability of CO2.

In contrast to leaf net CO2 assimilation, respiration rates of leaves, stems and

roots generally increased at the beginning of the flood and then remained unchanged

relative to controls. The initial increase in respiration, particularly in U. laevis, could

respond to the need of energy, carbon skeletons and reducing power to set up different

mechanism of acclimation to anaerobic conditions (Bailey-Serres and Voesenek 2008),

for example hypertrophication of lenticels, which involves increased phellogen activity

(Angeles et al. 1986). Although hypoxia may limit respiration (Bragina et al. 2004;

Bailey-Serres and Voesenek 2008; Polacik and Maricle 2013), we did not observe a

reduction of respiration in flooded relative to control plants later on. Firstly, the


54

threshold of oxygen concentration limiting mitochondrial respiration is low (Spicer and

Holbrook 2007). Secondly, sufficient oxygenation can be ensured through

hypertrophied lenticels (Newsome et al. 1982b; Kozlowski 1997; Lopez and Kursar

1999; Groh et al. 2002) or even cavitation and embolism of roots (Raven 1996; Else et

al. 2001; Sorz and Hietz 2005; Venturas et al. 2013). The formation of aerenchyma is a

common anatomical response of some plant species to flooding to improve oxygen

diffusion (Kozlowski 1997; Jackson 2002), but we did not observe it in either species.

Our discussion that respiration was barely affected by the flood is based on the

premise that a shift from aerobic respiration to fermentation, producing 3 times less

moles of CO2 per mol of glucose than aerobic respiration, would translate into a

reduction in measured CO2 efflux rates. Another methodological consideration is that

we measured root respiration at air ambient CO2 and O2 concentration, which could

result in the over-estimation of respiration rates that are actually occurring in the soil

(Marsden et al. 2008).

Compared to studies on photosynthetic carbon uptake, studies addressing the

effects of flooding on plant respiration are fewer. Here, the decline in photosynthesis

but not respiration resulted in an imbalance between the uptake and release of CO2 at

the plant level. It is remarkable that plant net carbon gain was very low and 9 times

lower in flooded than control plants for U. minor. These estimations would have

changed little had we accounted for fine root respiration, since fine root biomass was 4

times lower than coarse root biomass (data not shown). The imbalance between

photosynthesis and respiration could affect plant growth and survival. The fact that

latewood synthesis was negatively affected by flooding (or not at all in U. minor)

suggests that the increment in stem diameter observed with LVDTs over the first month

of flooding in both species was caused by swelling or a preferential allocation of carbon

to the bark (Yamamoto and Kozlowski 1987; Kozlowski 1997). While the ability to

maintain or enhance growth in flooding conditions has been related to flood tolerance

(Kozlowski 1997), a reduction in xylem growth implies reduced energy demand and

respiratory carbon losses, and thus can be beneficial for plants in overcoming floods.

Here, survival of U. minor flooded plants was ultimately compromised by the

degradation of bark and phloem. It is possible that ethylene participated in these

morphological alterations (He et al. 1996; Mergemann and Sauter 2000; Irfan et al.

2010), but also that shifts in carbon balance, allocation and metabolism were related to

the inability to maintain bark integrity under flooding (Vartapetian and Jackson 1997),

for example if phloem transport halted (Gravatt and Kirby 1998b; Ferner et al. 2012),


55

altering cell osmotic equilibrium at the water line, or if less carbon was available to

cope with water mold fungi abundant in flooded soils (Kozlowski 1997).

The ability to recover from flooding is a relevant aspect in the evaluation of

flooding tolerance (Anderson and Pezeshki 2000; Parolin 2001; Striker 2012).

Seedlings of U. laevis exhibited a remarkable capacity to recover normal plant function

after flooding: hydraulic conductivity and gas exchange were similar in flooded and

control plants. Hormonal and hydraulic factors limiting gas exchange disappeared

when soil gas conditions returned to normal. Since root maximum hydraulic

conductivity did not increase after flooding, it is possible that new vessels were grown

in the 40 day period between measurements or that xylem refilling occurred, as it

appears to happen after drought-induced xylem disruption (Canny 1998; Nardini et al.

2011). In this respect, the accumulation of soluble sugars in the parenchyma of stems

and roots possibly helps to stimulate xylem repair through water refilling (Ameglio et al.

2004; Zwieniecki and Holbrook 2009). Thus, the relatively high plant net carbon

balance in those U. laevis seedlings kept flooded for almost two months could help to

maintain a sufficient pool of carbohydrates and recover normal functionality afterwards.

In spite of complete recovery of leaf net photosynthesis, plant net carbon gain

continued to be significantly lower in plants previously flooded than in controls, due to

low leaf production during flooding in comparison with controls (Angelov et al. 1996;

Mielke et al. 2003). Such carry-over effect on carbon imbalance could compromise

survival if further stresses follow a first period of flooding. Actually, overcoming a period

of stress while maintaining carbohydrate reserves relatively high may benefit the plant

in coping with ensuing floods (Angelov et al. 1996), pathogen attacks (Kozlowski 1992)

or other abiotic stresses (Kozlowski 1997; Walters and Reich 1999; Flexas et al. 2006).

In conclusion, the above results suggest that U. laevis is more tolerant to

flooding than U. minor at the seedling stage. Seedlings of both species are able to

overcome short periods of flooding, but those of U. laevis maintain net photosynthesis

and whole-plant net carbon gain for longer periods of flooding and can recover

afterwards. These results are consistent with the distribution of both species; U. laevis

lives closer to the riverbank and could be more tolerant to flooding than U. minor, which

tends to occupy elevated areas away from the riverbank where the water table drops

during summer. This interpretation must be viewed with caution because experiments

with seedlings can differ from field experiments with adults (Parolin 2002; Glenz et al.

2006). Also, there are many environmental factors that interact with flooding in natural

conditions, such as time of flooding, light, and soil chemical properties (Streng et al.


56

1989; Kozlowski 1997; Kashem and Singh 2001; Voesenek et al. 2006). From a

practical viewpoint, however, it is useful to select species and genotypes for artificial

regeneration of riparian areas based on the ecophysiological response of seedlings to

controlled conditions of flooding. This study provides insights into the factors underlying

flood tolerance of two threatened species in the Iberian Peninsula for which

afforestation programs are recommended and planned for the near future. Our goal is

to determine most appropriate sites for planting both species, and this study constitutes

a first, necessary step towards this goal. It is still early to conclusively inform on best

sites for planting these species. Yet we can suggest that plantations with U. laevis,

particularly in the drought-prone South European regions, would benefit from choosing

locations in the close vicinity of the water channel due to the fast growth and flood

tolerance of this species. Lower growth and flood tolerance of U. minor seedlings

advices against planting this species as close to the water channel.

Acknowledgements

We are thankful to personnel from the fish farm of ETSI Montes (especially

Fernando Torrent and Victoriano Hernández) for kindly allowing the use of the

installations and monitoring pool water conditions. We also thank David Borrego for

making anatomical cuttings, Elena Zafra and Javier Cano for experimental preparations,

and Gerrie Seket for language revision.


57

References

Ameglio T, Decourteix M, Alves G, Valentin V, Sakr S, Julien JL, Petel G, Guilliot A, Lacointe A. 2004. Temperature effects on xylem sap osmolarity in walnut trees: evidence for a vitalistic model of winter embolism repair. Tree Physiology 24, 785-793.

Anderson PH, Pezeshki SR. 2000. The Effects of Intermittent Flooding on Seedlings of Three Forest Species. Photosynthetica 37, 543-552.

Angeles G, Evert RF, Kozlowski TT. 1986. Development of lenticels and adventitious roots in flooded Ulmus americana seedlings. Canadian Journal of Forest Research 16, 585-590.

Angelov MN, Sung S-JS, Doong RL, Harms WR, Kormanik PP, Black CC. 1996. Long- and short-term flooding effects on survival and sink – source relationships of swamp-adapted tree species. Tree Physiology 16, 477-484.

Aroca R, Porcel R, Ruiz-Lozano JM. 2012. Regulation of root water uptake under abiotic stress conditions. Journal of Experimental Botany 63, 43-57.

Atkin OK, Evans JR, Siebke K. 1998. Relationship between the inhibition of leaf respiration by light and enhancement of leaf dark respiration following light treatment. Functional Plant Biology 25, 437-443.

Bailey-Serres J, Voesenek LA. 2008. Flooding stress: acclimations and genetic diversity. Annual review of plant biology 59, 313-339.

Blanke M, Cooke D. 2004. Effects of flooding and drought on stomatal activity, transpiration, photosynthesis, water potential and water channel activity in strawberry stolons and leaves. Plant Growth Regulation 42, 153-160.

Blom CWPM, Voesenek LACJ. 1996. Flooding: the survival strategies of plants. Trends In Ecology & Evolution 11, 290-295.

Bradford KJ, Hsiao TC. 1982. Stomatal Behavior and Water Relations of Waterlogged Tomato Plants. Plant Physiology 70, 1508-1513.

Bragina TV, Ponomareva YV, Drozdova IS, Grinieva GM. 2004. Photosynthesis and Dark Respiration in Leaves of Different Ages of Partly Flooded Maize Seedlings. Russian Journal Of Plant Physiology 51, 342-347.

Buijse AD, Coops H, Staras M, Jans LH, Van Geest GJ, Grift RE, Ibelings BW, Oosterberg W, Roozen FCJM. 2002. Restoration strategies for river floodplains along large lowland rivers in Europe. Freshwater Biology 47, 889-907.

Canny M. 1998. Applications of the compensating pressure theory of water transport. American Journal Of Botany 85, 897.


58

Castonguay Y, Nadeau P, Simard RR. 1993. Effects of flooding on carbohydrate and ABA levels in roots and shoots of alfalfa. Plant Cell And Environment 16, 695-702.

Castro EB. 1997. Los bosques ibéricos: una interpretación geobotánica. GeoPlaneta, Editorial, SA

Collin E, Rusanen M, Ackzell L, Bohnens J, Aguiar Ad, Diamandis S, Franke A, Gil L, Harvengt L, Hollingsworth P, Jenkins G, Meier-Dinkel A, Mittempergher L, Musch B, Nagy L, Pâques M, Pinon J, Piou D, Rotach P, Santini A, Broeck AV, Wolf H. 2004. Methods and progress in the conservation of elm genetic resources in Europe. Invest Agrar: Sist Recur For 13, 261-274.

Colmer TD. 2003. Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant Cell And Environment 26, 17-36.

Dat JF, Parent C. 2012. Differential responses in sympatric tree species exposed to waterlogging. Tree Physiology 32, 115-118.

Desikan R, Last K, Harrett-Williams R, Tagliavia C, Harter K, Hooley R, Hancock JT, Neill SJ. 2006. Ethylene-induced stomatal closure in Arabidopsis occurs via AtrbohF-mediated hydrogen peroxide synthesis. The Plant Journal 47, 907-916.

Else MA, Coupland D, Dutton L, Jackson MB. 2001. Decreased root hydraulic conductivity reduces leaf water potential, initiates stomatal closure and slows leaf expansion in flooded plants of castor oil (Ricinus communis) despite diminished delivery of ABA from the roots to shoots in xylem sap. Physiologia Plantarum 111, 46-54.

Else MA, Janowiak F, Atkinson CJ, Jackson MB. 2009. Root signals and stomatal closure in relation to photosynthesis, chlorophyll a fluorescence and adventitious rooting of flooded tomato plants. Annals Of Botany 103, 313-323.

Else MA, Tiekstra AE, Croker SJ, Davies WJ, Jackson MB. 1996. Stomatal Closure in Flooded Tomato Plants Involves Abscisic Acid and a Chemically Unidentified Anti-Transpirant in Xylem Sap. Plant Physiol 112, 239-247.

Evette A, Balique C, Lavaine C, Rey F, Prunier P. 2012. Using ecological and biogeographical features to produce a typology of the plant species used in bioengineering for riverbank protection in europe. River Research and Applications 28, 1830-1842.

Ferner E, Rennenberg H, Kreuzwieser J. 2012. Effect of flooding on C metabolism of flood-tolerant (Quercus robur) and non-tolerant (Fagus sylvatica) tree species. Tree Physiology 32, 135-145.

Ferreira CS, Piedade MTF, Franco AC, Gonçalves JFC, Junk WJ. 2009. Adaptive strategies to tolerate prolonged flooding in seedlings of floodplain and upland


59

populations of Himatanthus sucuuba, a Central Amazon tree. Aquatic Botany 90, 246-252.

Ferreira LV, Stohlgren TJ. 1999. Effects of river level fluctuation on plant species richness, diversity, and distribution in a floodplain forest in Central Amazonia. Oecologia 120, 582-587.

Flexas J, Bota J, Galmés J, Medrano H, Ribas-Carbó M. 2006. Keeping a positive carbon balance under adverse conditions: responses of photosynthesis and respiration to water stress. Physiologia Plantarum 127, 343-352.

Gérard B, Alaoui-Sossé B, Badot P-M. 2008. Flooding effects on starch partitioning during early growth of two oak species. Trees 23, 373-380.

Glenz C, Schlaepfer R, Iorgulescu I, Kienast F. 2006. Flooding tolerance of Central European tree and shrub species. Forest Ecology And Management 235, 1-13.

Gravatt DA, Kirby CJ. 1998. Patterns of photosynthesis and starch allocation in seedlings of four bottomland hardwood tree species subjected to flooding. Tree Physiology 18, 411-417.

Groh B, Hubner C, Lendzian KJ. 2002. Water and oxygen permeance of phellems isolated from trees: the role of waxes and lenticels. Planta 215, 794-801.

He C-J, Morgan PW, Drew MC. 1996. Transduction of an ethylene signal is required for cell death and lysis in the root cortex of maize during aerenchyma formation induced by hypoxia. Plant Physiology 112, 463-472.

Herrera A, Tezara W, Marin O, Rengifo E. 2008. Stomatal and non-stomatal limitations of photosynthesis in trees of a tropical seasonally flooded forest. Physiologia Plantarum 134, 41-48.

Irfan M, Hayat S, Hayat Q, Afroz S, Ahmad A. 2010. Physiological and biochemical changes in plants under waterlogging. Protoplasma 241, 3-17.

Islam MA, Macdonald SE. 2004. Ecophysiological adaptations of black spruce ( Picea mariana ) and tamarack ( Larix laricina ) seedlings to flooding. Trees - Structure and Function 18, 35-42.

Islam MA, MacDonald SE, Zwiazek JJ. 2003. Responses of black spruce (Picea mariana) and tamarack (Larix laricina) to flooding and ethylene. Tree Physiology 23, 545-552.

Ismail MR, Noor KM. 1996. Growth and physiological processes of young starfruit (Averrhoa carambola L.) plants under soil flooding. Scientia Horticulturae 65, 229-238.

Jackson MB. 2002. Long‐distance signalling from roots to shoots assessed: the

flooding story. J. Exp. Bot 53, 175-181.


60

Jackson MB, Hall KC. 1987. Early stomatal closure in waterlogged pea plants is mediated by abscisic acid in the absence of foliar water deficits. Plant Cell And Environment 10, 121-130.

Jaeger C, Gessler A, Biller S, Rennenberg H, Kreuzwieser J. 2009. Differences in C metabolism of ash species and provenances as a consequence of root oxygen deprivation by waterlogging. Journal of Experimental Botany 60, 4335-4345.

Javot H, Maurel C. 2002. The role of aquaporins in root water uptake. Annals Of Botany 90, 301-313.

Johnson DM, Brodersen CR, Reed M, Domec JC, Jackson RB. 2014. Contrasting hydraulic architecture and function in deep and shallow roots of tree species from a semi-arid habitat. Annals Of Botany 113, 617-627.

Kashem MA, Singh BR. 2001. Metal availability in contaminated soils: I. Effects of floodingand organic matter on changes in Eh, pH and solubility of Cd, Ni andZn. Nutrient Cycling In Agroecosystems 61, 247-255.

Kennedy RA, Rumpho ME, Fox TC. 1992. Anaerobic metabolism in plants. Plant Physiol 100, 1-6.

Kozlowski TT. 1992. Carbohydrate sources and sinks in woody plants. Botanical Review 58, 107-222.

Kozlowski TT. 1997. Responses of woody plants to flooding and salinity. Tree Physiology 17, 490.

Kreuzwieser J, Papadopoulou E, Rennenberg H. 2004. Interaction of Flooding with Carbon Metabolism of Forest Trees. Plant Biology 6, 299-306.

Li M, Yang D, Li W. 2007. Leaf gas exchange characteristics and chlorophyll fluorescence of three wetland plants in response to long-term soil flooding. Photosynthetica 45, 222-228.

Loewenstein NJ, Pallardy SG. 1998. Drought tolerance, xylem sap abscisic acid and stomatal conductance during soil drying: a comparison of canopy trees of three temperate deciduous angiosperms. Tree Physiology 18, 431-439.

Lopez OR, Kursar TA. 1999. Flood tolerance of four tropical tree species. Tree Physiology 19, 925-932.

Maček I, Pfanz H, Francetič V, Batič F, Vodnik D. 2005. Root respiration response to high CO2 concentrations in plants from natural CO2 springs. Environmental And Experimental Botany 54, 90-99.

Marsden C, Nouvellon Y, Epron D. 2008. Relating coarse root respiration to root diameter in clonal Eucalyptus stands in the Republic of the Congo. Tree Physiology 28, 1245-1254.


61

McKee KL. 1996. Growth and physiological responses of neotropical mangrove seedlings to root zone hypoxia. Tree Physiology 16, 883-889.

Mergemann H, Sauter M. 2000. Ethylene induces epidermal cell death at the site of adventitious root emergence in rice. Plant Physiology 124, 609-614.

Mielke MS, de Almeida A-AF, Gomes FP, Aguilar MAG, Mangabeira PAO. 2003. Leaf gas exchange, chlorophyll fluorescence and growth responses of Genipa americana seedlings to soil flooding. Environmental And Experimental Botany 50, 221-231.

Munkvold GP, Yang XB. 1995. Crop damage and epidemics associated with 1993 floods in Iowa. Plant Disease 79, 95-101.

Nardini A, Lo Gullo MA, Salleo S. 2011. Refilling embolized xylem conduits: Is it a matter of phloem unloading? Plant Science 180, 604-611.

Newsome RD, Kozlowski TT, Tang ZC. 1982. Responses of Ulmus americana seedlings to flooding of soil. Canadian Journal of Botany 60, 1688-1695.

Nicolás E, Torrecillas A, Dell’Amico J, Alarcón J. 2005. The effect of short-term flooding on the sap flow, gas exchange and hydraulic conductivity of young apricot trees. Trees 19, 51-57.

Parent C, Capelli N, Berger A, Crèvecoeur M, Dat JF. 2008. An Overview of Plant Responses to Soil Waterlogging. plant stress 2, 20-27.

Parolin P. 2001. Morphological and physiological adjustments to waterlogging and drought in seedlings of Amazonian floodplain trees. Oecologia 128, 326-335.

Parolin P. 2002. Submergence tolerance vs. escape from submergence: two strategies of seedling establishment in Amazonian floodplains. Environmental And Experimental Botany 48, 177-186.

Pezeshki SR. 2001. Wetland plant responses to soil flooding. Environmental And Experimental Botany 46, 299-312.

Pezeshki SR, Santos MI. 1998. Relationships among Rhizosphere Oxygen Deficiency, Root Restriction, Photosynthesis, and Growth in Baldcypress (Taxodium Distichum L.) Seedlings. Photosynthetica 35, 381-390.

Piper F, Zuniga-Feest A, Rojas P, Alberdi M, Corcuera LJ, Lusk CH. 2008. Responses of two temperate evergreen Nothofagus species to sudden and gradual waterlogging: relationships with distribution patterns. Revista Chilena De Historia Natural 81, 257-266.

Polacik KA, Maricle BR. 2013. Effects of flooding on photosynthesis and root respiration in saltcedar (Tamarix ramosissima), an invasive riparian shrub. Environmental And Experimental Botany 89, 19-27.


62

Ponnamperuma FN. 1972. The Chemistry of Submerged Soils. Advances in Agronomy 24, 29-96.

Raven JA. 1996. Into the Voids: The Distribution, Function, Development and Maintenance of Gas Spaces in Plants. Annals Of Botany 78, 137-142.

Sorz J, Hietz P. 2005. Gas diffusion through wood: implications for oxygen supply. Trees 20, 34-41.

Spicer R, Holbrook NM. 2007. Effects of carbon dioxide and oxygen on sapwood respiration in five temperate tree species. Journal of Experimental Botany 58, 1313-1320.

Streng DR, Glitzenstein JS, Harcombe PA. 1989. Woody Seedling Dynamics in an East Texas Floodplain Forest. Ecological Monographs 59, 177-204.

Striker GG. 2012. Flooding Stress on Plants: Anatomical, Morphological and Physiological Responses. Botany

Teskey RO, McGuire MA. 2004. CO2 transported in xylem sap affects CO2 efflux from Liquidambar styraciflua and Platanus occidentalis stems, and contributes to observed wound respiration phenomena. Trees 19, 357-362.

Tournaire-Roux C, Sutka M, Javot H, Gout E, Gerbeau P, Luu DT, Bligny R, Maurel C. 2003. Cytosolic pH regulates root water transport during anoxic stress through gating of aquaporins. Nature 425, 393-397.

Tyree MT, Sperry JS. 1988. Do woody plants operate near the point of catastrophic xylem dysfunction caused by dynamic water stress? Answers from a model. Plant Physiology 88, 0574-0580.

Vartapetian BB, Jackson MB. 1997. Plant Adaptations to Anaerobic Stress. Annals Of Botany 79, 3-20.

Venturas M, Fernandez V, Nadal P, Guzman P, Lucena JJ, Gil L. 2014a. Root iron uptake efficiency of Ulmus laevis and U. minor and their distribution in soils of the Iberian Peninsula. Front Plant Sci 5, 1-9.

Venturas M, Fuentes-Utrilla P, López R, Perea R, Fernández V, Gascó A, Guzmán P, Li M, Rodríguez-Calcerrada J, Miranda E, Domínguez J, González-Gordaliza G, Zafra E, Fajardo-Alcántara M, Martín JA, Ennos R, Nanos N, Lucena JJ, Iglesias S, Collada C, Gil L. 2014b. Ulmus laevis in the Iberian Peninsula: a review of its ecology and conservation. iForest



63

Voesenek LA, Colmer TD, Pierik R, Millenaar FF, Peeters AJ. 2006. How plants cope with complete submergence. New phytol 170, 213-226.

Walters MB, Reich PB. 1999. Low-light carbon balance and shade tolerance in the seedlings of woody plants: do winter deciduous and broad-leaved evergreen species differ? New Phytologist 143, 143-154.

Yáñez-Espinosa L, Terrazas T, López-Mata L. 2001. Effects of flooding on wood and bark anatomy of four species in a mangrove forest community. Trees 15, 91-97.

Yamamoto F, Kozlowski TT. 1987. Regulation by auxin and ethylene of responses of Acer negundo seedlings to flooding of soil. Environmental And Experimental Botany 27, 329-340.

Yamamoto F, Sakata T, Terazawa K. 1995. Physiological, morphological and anatomical responses of Fraxinus mandshurica seedlings to flooding. Tree Physiology 15, 713-719.

Yanar Y, Lipps PE, Deep IW. 1997. Effect of Soil Saturation Duration and Soil Water Content on Root Rot of Maize Caused by Pythium arrhenomanes. Plant Disease 81, 475-480.

Ye Y, Tam NFY, Wong YS, Lu CY. 2003. Growth and physiological responses of two mangrove species (Bruguiera gymnorrhiza and Kandelia candel) to waterlogging. Environmental And Experimental Botany 49, 209-221.

Zhang J, Zhang X. 1994. Can early wilting of old leaves account for much of the ABA accumulation in flooded pea plants? Journal of Experimental Botany 45, 1335-1342.

Zwieniecki MA, Holbrook NM. 2009. Confronting Maxwell's demon: biophysics of xylem embolism repair. Trends In Plant Science 14, 530-534.

65

4. DROUGHT-INDUCED MASSIVE XYLEM EMBOLISM INCURS A

VARIABLE DEPLETION OF NON-STRUCTURAL CARBON

RESERVES IN DYING SEEDLINGS OF TWO TREE SPECIES

Drought-induced massive xylem embolism incurs a variable depletion of non-structural carbon reserves in dying seedlings of two tree species

67

Drought-induced massive xylem embolism incurs a variable depletion of non-

structural carbon reserves in dying seedlings of two tree species

Jesús Rodríguez-Calcerrada, Meng Li, Rosana López, Francisco Javier Cano, Jacek

Oleksyn, Owen K. Atkin, Pilar Pita, Ismael Aranda, Luis Gil

Submitted to New phytologist


68

Abstract

Combining hydraulic- and carbon-related measurements helps to untangle the

causes of plant mortality under drought.

Here we have measured hydraulic conductivity of stems and roots, and gas

exchange and non-structural carbohydrate (NSC) concentrations of leaves, stems and

roots of seedlings of two resprouting species of contrasted drought resistance – Ulmus

minor and Quercus ilex – exposed to either drought or well-watered conditions.

With increasing water stress, less pronounced declines were observed in leaf,

stem and root respiration than leaf net photosynthetic CO2 uptake (Pn). Daytime whole-

plant carbon gain was negative below -3 and -5 MPa midday xylem water potential in U.

minor and Q. ilex, respectively. Relative to controls, seedlings showing symptoms of

shoot dieback suffered ~80% loss of hydraulic conductivity, and little reductions in NSC

concentrations in U. minor. Higher drought-induced depletion of root NSC reserves in

this species was not related to higher limitations to Pn, but to higher root respiration and

premature leaf shedding.

Differences in species’ drought resistance relied on the ability to maintain water

transport during drought, rather than tolerating severe losses of conductivity. Root

hydraulic failure impinged on shoot dieback and elicited further alterations precluding

resprouting without root NSC reserves being apparently limiting for respiration.

Key words: plant mortality, carbon starvation, cavitation, die-back, forest decline, holm

oak, field elm.


69

Introduction

There has long been interest in understanding the causes of drought-induced

plant mortality (e.g. Parker & Patton, 1975; Tyree & Sperry, 1989; Pedersen, 1998).

Recent widespread phenomena of forest mortality and die-off associated to drought

(e.g. Allen et al., 2010) have fueled research on physiological alterations preceding

plant death by water stress. McDowell et al. (2008) proposed a conceptual framework

of the functional mechanisms underlying plant mortality under drought. Two interrelated

mechanisms of mortality were adduced by these authors: hydraulic failure – massive

xylem cavitation occurring when soil water content does not meet transpiration demand

– and carbon starvation – insufficient carbon substrates to meet respiratory demand.

The capacity of a given species to put in place mechanisms that minimize water loss or

contribute to overcoming water-deficit stress will determine the extent and pace to

which xylem water transport and non-structural carbohydrates (NSC) are affected by

drought and cause plant death in the last. Based on this reasoning, McDowell et al.,

(2008) suggested that isohydric species, with a marked decline in stomatal

conductance to water vapor (gs) and net photosynthetic carbon dioxide (CO2) uptake

(Pn) in response to drought would be more prone to die from carbon starvation; by

contrast, anisohydric species that maintain higher transpiration rates and xylem

tensions during drought episodes would be more prone to die from hydraulic failure.

Subsequent studies have reported that safety margins for hydraulic transport are not

necessarily lower in anisohydric species (Plaut et al., 2012) and emphasized that

hydraulic failure and carbon starvation are tightly interrelated (Sala et al., 2010;

McDowell 2011a, b; Galvez et al., 2013). More studies are needed to understand how

drought-induced physiological dysfunctions lead to plant mortality, and eventually

generalizing the way those interactions vary among plant functional types.

Declines in leaf gs are a common response of most plant species when

challenged with increasing soil drought. Early stomata closure is an adaptive

mechanism partly triggered by hormones (e.g. abscisic acid; Tardieu & Davies 1993)

that reduces water loss and prevents xylem cavitation (Plaut et al., 2012). However,

stomata have a limited capacity to regulate water loss. Sustained drought exacerbates

the imbalance between water transport and transpiration (through stomata and

cuticles), causing increasing xylem tension and cavitation, and further declines in gs

and Pn. Importantly, hydraulic limitations to Pn are just one way through which plant

water relations and carbon dynamics are coupled. For example, a given percentage

loss of hydraulic conductivity can prompt a variable penalty to NSC, depending on


70

existing NSC pools preceding drought, and the impact of drought on diffusional

limitations to Pn, growth, and organ-specific respiration rates (Sala et al., 2010 and

2012; McDowell 2011b; Dietze et al., 2014). This can explain that fatal loss of hydraulic

conductivity takes place at variable declines of NSC, with recent studies reporting

either increased (Galvez et al., 2011; Anderegg et al., 2012), decreased (Galiano et al.,

2011; Adams et al., 2013; Mitchell et al., 2013; Dickman et al., 2014) or unchanged

(Gruber et al., 2012; Anderegg & Anderegg 2013) concentrations of NSC in plants

exposed to severe drought relative to previous rainy periods or well-watered control

plants. Moreover, at the organ-scale, drought-induced variations in NSC concentrations

are conditioned by the impacts of water stress on NSC export from source to sink

tissues. The few studies on phloem transport available suggest that phloem transport

fails at mild to moderate levels of drought (Sevanto et al., 2014; Woodruff, 2014); given

this, there is an urgent need to examine how sink activities, such as mitochondrial

respiration (R) rates and NSC pools in all plant organs (leaves, stems and roots) are

affected by drought among species to understand the role of NSC in surviving drought.

While great attention has been paid to assess how stomatal regulation of transpiration

and Pn affects NSC reserves during drought, less has been paid on the assessment of

plant R.

The response of the hydraulic system to drought is more consistent than that of

NSC: xylem hydraulic conductivity almost invariably decreases, well beyond 50% of

maximum capacity in severely water-stressed plants. The ability to maintain the

integrity of the hydraulic system in response to xylem water tension varies across

species, reflecting inter-specific differences in tolerance to drought and habitat

preferences (Pockman & Sperry 2000; Bartlett et al., 2013; López et al., 2013).

However, tolerance of plant species to hydraulic failure is a more conserved feature;

that is, while the xylem tension at which 50% of the hydraulic conductivity is lost is

related to drought resistance, the percent loss of conductivity (PLC) at which plants die

is generally high and largely invariant among species. Angiosperms of variable drought

resistance can often recover normal physiological function (or resprout) after severe

drought episodes have caused almost a complete loss of stem hydraulic conductivity

(i.e. 60 to almost 100% PLC; Davis et al., 2002; Vilagrosa et al., 2003; Barigah et al.,

2013; Nardini et al., 2013; Urli et al., 2013). Such studies show that hydraulic failure

plays a major role in drought-induced dieback. Importantly, however, most studies

addressing the question of plant mortality during drought have so far ignored root

hydraulic conductivity and lacked a concomitant examination of both water- and

carbon-related aspects (but see Anderegg et al., 2012; Anderegg & Anderegg, 2013;


71

Galvez et al., 2013; Mitchell et al., 2013; Pratt et al., 2014); such information is needed

to more fully elucidate the contribution of hydraulic failure and/or carbon starvation in

contributing to drought-induced plant mortality.

The objective of our study was to compare the impact of drought on functional

features of seedlings of two broadleaved tree species of contrasted drought resistance:

Ulmus minor L. and Quercus ilex L. Both species have common broad distribution

ranges and coexist in some areas of South Western Europe. However, Q. ilex is very

tolerant of summer drought, whereas U. minor only grows at sites near rivers where the

underlying water table provides greater moisture availability. By submitting seedlings to

a sustained drought treatment, we expected that the functional alterations preceding

dieback would vary between species in relation to their resistance to drought. We

addressed two main questions: (1) how differences in species’ drought resistance

affect responses of whole-plant net carbon gain, hydraulic conductivity, and NSC levels

of source and sink organs to increasing drought severity; (2) do plants of each species

succumb at similar levels of xylem cavitation and NSC depletion (if any), or do plants of

the more drought-resistant Q. ilex die at higher levels of xylem cavitation and/or with

distinct levels of NSC reserves?

Material and methods

Experimental design

Seeds from several mother trees of U. minor and Q. ilex were sown in 16-dm3

pots. The substrate was a mixture of sieved peat TKS2 (Floragard Vertriebs GmbH,

Oldenburg, Germany) and sand in 3:1 volume proportions, enriched with 1 g dm-3 of

slow-release fertilizer (Osmocote Pro 5-6 months, 17N+11P+10K+2MgO+trace

elements; Everris International B.V., The Netherlands). All pots were filled to the same

weight of 14 kg. Seedlings were grown outdoors under optimal water conditions for the

first year. By mid-summer of the second year, irrigation was stopped to half of plants

and rainfall excluded by a 3-m height transparent rain-out shelter. Two groups of plants

were established: one intended to monitoring survival and another one to make

destructive physiological and biochemical measurements in different plants over time.

Plants from each group were then randomly assigned to either a treatment of drought

(D) or regular watering (C). Stem diameter at the base at the beginning of the second

year (before imposing drought) was 6.3 ± 0.2 and 6.4 ± 0.3 cm for C and D plants of U.

minor and 3.6 ± 0.1 and 3.5± 0.2 cm for C and D plants of Q. ilex. Three complete


72

blocks were established under the rain-out shelter in a randomized block split-plot

design, with treatment plots established within each replicate block, and species

randomly allocated to positions within each plot.

The drought treatment was applied by withholding watering; to ensure a slow

decline of soil water content and a progressive imposition of water stress, plants were

grown in big pots under neutral shade clothing reducing 40% sunlight. Although the use

of pots almost inevitably alter plant responses to drought (e.g. Anderegg & Anderegg,

2013), our data from final harvests indicated plant biomass to pot volume ratios were

1.2 and 0.4 g dm-3 for U. minor and Q. ilex, respectively, which are below the 2 g dm-3

limit suggested by Poorter et al., (2012) to minimize pot size artefacts on experimental

treatments.

At repeated intervals after stopping watering, five plants per treatment of either

U. minor or Q. ilex were sampled for physiological and biochemical traits. At sampling

days, leaf water potential was measured before dawn, and immediately afterwards dark

respiration was measured in a nearby leaf. Two-three hours after sunrise, we

measured leaf net photosynthetic CO2 uptake and midday water potential. Thereafter,

pots were moved to a laboratory with natural irradiance and 25ºC air temperature, and

plants were harvested over approximately 3-4 h for measuring stem and root

respiration and hydraulic conductivity, alternately between treatments. Additional

fragments of stems and coarse roots were cut, weighed, and instant-frozen in liquid

nitrogen for further biochemical analyses. Roots, stems and leaves were separated and

oven dried.

Mortality

In the group of plants intended to monitor survival, stem dieback (i.e. dried-out

stems that easily crumbled) was considered as a surrogate of plant mortality. Plants

affected by dieback were then immediately re-watered to assess root sprouting and

how reliable stem dieback was as a surrogate of plant mortality.

Water status


73

The relative water content (RWC) of plant organs (leaf, stem and roots) was

calculated as:

RWC =𝐹𝑀 − 𝐷𝑀

𝑆𝑀 − 𝐷𝑀∙ 100;

where, for each organ, FM is the fresh mass at the time of harvesting, DM is the

mass after oven-drying for 3 days, and SM is the mass after rehydration overnight.

Values of 100% root RWC or higher were found for some plants, suggesting overnight

sample rehydration led to tissue oversaturation in these cases.

Predawn (pd) and midday (md) water potential were also measured in one leaf

per plant (occasionally in one or two extra leaves), using a pressure chamber (PMS

Instrument Co., Albany, OR, USA). Pressure values over 8 MPa were beyond the limits

of the pressure chamber and could not be measured. Stem xylem water potential at

midday (md,stem) in a subset of plants was estimated by measuring water potential in a

non-transpiring leaf, covered with aluminum paper for one hour (Zhang et al., 2013).

md,stem was then calculated for all plants based on the linear regressions between

md,stem and md (Q. ilex: md,stem = 1.34 + 1.21md, r2= 0.95, n=36; U. minor: 0.66 +

1.12md, r2= 0.98, n=33).

Stem diameter growth

Short-term fluctuations in diameter were continuously measured with linear

variable displacement transducers (LVDTs; Model DF2.5, Solartron Metrology Bognor

Regis, UK) in five plants per species and treatment.

Leaf gas exchange

Leaf gas exchange was measured with an IRGA system (Li-6400, Li-Cor INC.,

Lincoln, NE, USA) coupled to a broadleaf chamber (6400-02B LED Light Source, Li-

Cor INC., Lincoln, NE, USA). Measurements were made at 400 ppm CO2 concentration,

24.9 ºC (± 0.4 ºC standard deviation; SD) average air temperature (corresponding to an

average leaf temperature of 26.4 ± 1.0 ºC), ambient air relative humidity of 45.5 ±

5.6 %, and either 1200 or 0 mol m-2 s-1 photosynthetic photon flux density (PPFD) for

net photosynthetic CO2 uptake (Pn) or respiratory CO2 efflux (Rleaf) measurements,


74

respectively. Area-based values were expressed on a leaf dry mass basis using values

of leaf mass per area.

Stem and root respiration

Fragments of 5-9 cm length/3-11 mm diameter were cut from the stem, near its

base, and the main pivotal root at its most upper part. Root samples were gently

cleaned of soil particles using moistened paper towel. Stem and root CO2 efflux was

measured in a 6400-05 conifer chamber covered with foil paper using the Li-6400

IRGA system. The two cut ends of fragments were sealed with terostat putty (Teroson,

Heidelberg, Germany) to avoid CO2 release through the cut surfaces and minimize

sampling noise. Terostat putty was also applied all around the gaskets of the chamber

to prevent CO2 leakage. We assumed CO2 storage was minimal in sampled fragments

and that CO2 efflux reflected the activity of their living cells (Teskey et al., 2008).

Fragments were in the chamber for 5-10 min at 24.8 ± 0.3 ºC air temperature until CO2

release was constant. Previous tests showed that CO2 release rate changed by 8.7 and

2.7 % in stems and roots, respectively, within the hour following equilibration time,

suggesting a negligible effect of cutting in respiratory activity. Rates of stem (Rstem) and

root (Rroot) respiration were expressed per unit dry mass from the relationship between

the curved surface area and dry mass. The curved surface area was computed from

the height and diameter of the cylindrical samples.

Not knowing the CO2 concentration in the soil, we decided to measure root

respiration (Rroot) at the same CO2 concentration than leaves and stems (400 ppm); it is

possible that in vivo Rroot was lower due to inhibition of mitochondrial activity by high

CO2 concentration in the soil (e.g. Maček et al., 2005).

Whole-plant net carbon gain

Instantaneous net carbon gain (PCG) was calculated during day time by

subtracting CO2 losses from woody organs (estimated by multiplying stem and root dry

masses by their mass-specific rates of respiration) to foliage net CO2 uptake (estimated

by multiplying mass-based leaf Pn, already accounting for leaf respiration in light

conditions, by total leaf dry mass).

𝑃𝐶𝐺 = 𝐿𝐷𝑀𝑖 ∙ 𝑃𝑛,𝑖 − 𝑆𝐷𝑀 ∙ 𝑅𝑠𝑡𝑒𝑚 − 𝑅𝐷𝑀 ∙ 𝑅𝑟𝑜𝑜𝑡


75

In this equation the subscript i refers to Pn of current- or previous-year leaves. A

45% reduction of Pn in previous- relative to current-year leaves was assumed for the

evergreen Q. ilex based on data from Rodríguez-Calcerrada et al., (2012).

Abbreviations LDM, SDM and RDM refer to total leaf, stem, and root dry masses,

respectively. The fresh mass of material sampled for biochemistry was annotated and

its dry mass estimated from the linear regression between fresh and dry masses for

each organ (r2 = 0.90 – 0.94).

Concentration of non-structural carbohydrates

Frozen plant materials were oven dried at 75 ºC for two days and then ground

to fine powder. The concentration of NSC in leaves (current- and previous-year leaves

separately for Q. ilex), stems and roots was measured using modified protocols

described by Haissig & Dickson (1979) and Hansen & Møller (1975). Soluble sugars

(SS) were extracted in a solution of methanol, chloroform and water (12:5:3 by volume)

and incubated with anthrone reagent (Sigma A 1631). The solution was measured

within 30 min of incubation at 625 nm wavelength in a spectrophotometer (UV–1700

PharmaSpec, UV-Visible Spectrophotometer, Shimadzu, Japan). Starch contained in

the insoluble material was transformed into glucose with amyloglucosidase (Sigma

10115-1G-F). Glucose was then oxidized with the peroxidase-glucose oxidase complex

(Sigma P7119), and measured at 450 nm within 30 min. Additional details on the

protocols can be found in Oleksyn et al., (2000). The concentration of NSC for each

organ was calculated by summing the concentrations of SS and starch. Concentrations

at the whole-plant level were calculated as the sum of the weighted averages for each

organ, which were calculated by multiplying the concentration of SS, starch or NSC of

each organ by the respective fraction of dry mass (i.e. organ dry mass per unit plant

dry mass). Similarly, the weighted average of NSC in current- and previous-year leaves

was used to calculate leaf NSC concentration for Q. ilex.

Hydraulic conductance of shoots and roots

Native embolism was determined using a XYL’EM embolism-meter (Bronkhorst,

Montigny-Les-Cormeilles, France). The main stem and root were cut under water to

minimize air entering through the cut surfaces; thereafter, current-year and two-year-

old stem segments and root segments 50 mm long were re-cut under water and the

edges were trimmed using a razor blade. Excising the samples under native tension

was unlikely to have induced xylem cavitation (Venturas et al., 2015). Bark was


76

removed also under water before fitting the segment to water-filled tubing attached on

the upstream end to a solution reservoir containing 10 mM KCl and 1mM CaCl2 in

deionized degassed water. Initial hydraulic conductance (ki) was determined as the

water mass flow per pressure gradient through the segment under low pressure (6

kPa). Segments were then flushed with the same solution at 0.18 MPa for 10 minutes

to remove embolisms, and hydraulic conductance was measured again to obtain the

maximum hydraulic conductance through the segment (kmax). Percentage loss of

conductance (PLC) was determined for each segment as: PLC (%) = 100*(kmax –

ki)/kmax.

Statistical analyses

Means between C and D treatments were compared at each date with t-tests,

or Welch tests when variances were unequal. When examining the relationship

between two variables, linear and non-linear models were fitted and compared based

on their root mean square error (RMSE), retaining the one with the lowest RMSE. The

statistical significance and coefficient of determination (r2) was calculated for linear

models. The performance of non-linear models, however, was estimated by the r2 of

the linear relationship between predicted and observed values (Ratkowsky, 1990;

Martin-StPaul et al., 2014). Pearson correlations were calculated among NSC

concentrations and water status indicators, even for cases of slightly skewed (non-

normal) data distributions (Edgell & Noon, 1984). The comparison of slopes of

relationships between species was made by testing the interaction term between the

species factor and the predictor variable. Variables were log-transformed (log x, or log

(x+k) for negative values) to linearize exponential relationships. Analyses were

conducted with the software Statistica 7.1 (StatSoft, Tulsa, OK, USA).

Results

Mortality

From the group of plants intended to monitoring survival, no plant subjected to

regular watering died over the course of the experiment; by contrast, 74% of U. minor

and 17% of Q. ilex plants subjected to water withholding suffered shoot dieback (Fig. 1).

Of these, only one plant of U. minor resprouted in the next spring. No attacks of insects

or pathogens were apparent during the experiment.


77

Fig. 1. Percentage of seedlings exposed to water withholding showing stem dieback as an

indicator of plant mortality.

Water status

Leaf predawn (pd) and midday (md) water potential, and midday stem xylem

water potential (md,stem) declined as drought progressed; all were increasingly lower in

D than C plants after water cessation (Fig. 2).


78

Fig. 2. Progression of leaf predawn water potential (pd; above), and midday leaf (md; below,

circles) and stem (md,stem; below, triangles) water potentials in seedlings of Quercus ilex and

Ulmus minor. Asterisks mark significant differences between means (±SE; n= 3-7) of control and

droughted plants (filled symbols) as *, p < 0.05 and **, p < 0.01. Notice the different X-axes for each

species.

Q. ilex exhibited a larger gradient of water potential between md and pd (i.e.

absolute value of md minus pd) than U. minor. However, this gradient decreased with

increasing drought in both species, from 1.3 to 0.2 MPa in U. minor and from 2.1 to 0.3

MPa in Q. ilex (Fig. 3).


79

Fig. 3. Relationship between leaf predawn (pd) and midday (md) water potentials for Quercus ilex

and Ulmus minor seedlings.

Leaf, stem and root relative water content (RWC) also declined relative to C

plants as drought progressed (Fig. 4).


80

Fig. 4. Progression of leaf, stem and root relative water content (RWC) in seedlings of Quercus ilex

and Ulmus minor. Asterisks mark significant differences between means (±SE; n= 3-7) of control

and droughted plants (filled symbols) as *, p < 0.05; **, p < 0.01; and ***, p < 0.001. Notice the

different X-axes for each species.

The RWC of each organ and water potentials (pd, md and md,stem) exhibited

strong, positive correlations (Table. 1).


81

Table 1 Pearson product-moment correlation coefficients of water status indicators leaf predawn

water potential (pd), midday stem (md,stem) water potential and organ relative water content (RWC)

for seedlings of Quercus ilex and Ulmus minor. All correlations are significant at p < 0.001.

Q. ilex pd md, stem RWCleaf RWCstem RWCroot

pd 1.00 0.98 0.89 0.94 0.92

md, stem 0.98 1.00 0.87 0.93 0.94

RWCleaf 0.89 0.87 1.00 0.92 0.91

RWCstem 0.94 0.93 0.92 1.00 0.93

RWCroot 0.92 0.94 0.91 0.93 1.00

U. minor pd md, stem RWCleaf RWCstem RWCroot

pd 1.00 0.97 0.97 0.86 0.92

md, stem 0.97 1.00 0.95 0.80 0.87

RWCleaf 0.97 0.95 1.00 0.86 0.93

RWCstem 0.86 0.80 0.86 1.00 0.93

RWCroot 0.92 0.87 0.93 0.93 1.00

Some of the plants sampled for physiology exhibited visible symptoms of

dieback - i.e. stem tips becoming dried, wilting leaves of Q. ilex turning brown, and

most leaves shed in U. minor (herein, we refer to these as ‘dying plants’). The time

taken to reach this state differed between the two species: 94 days for Q. ilex and 50

for U. minor. pd in dying plants of Q. ilex (n= 5) was lower than the limit of the pressure

bomb (-8 MPa) and -5.2 to -5.3 MPa in those of U. minor (n=4). Moreover, organ RWC

was below 50% in all but the leaves of two dying plants; controls at this time had pd > -

0.75 MPa and RWC > 85%.

Stem diameter growth

In the first weeks of the experiment U. minor seedlings exhibited higher growth

than those of Q. ilex; well-watered plants then exhibited a trend of reducing stem

diameter growth as summer moved forward towards autumn (Fig. 5).


82

Fig. 5. Weekly growth in seedlings of Quercus ilex and Ulmus minor exposed to drought or control

well-watered conditions. Asterisks mark significant differences between means (±SE; n= 4-5) of

control and droughted plants (filled symbols) as *, p < 0.05.

In the case of D plants, growth stopped 28 days after the onset of the treatment

in U. minor, and it exhibited increasingly negative values (i.e. stem shrinkage) over the

following 20 days, down to 150 m week-1 shrinkage by day 50th. Thereafter stem

shrinkage gradually decreased, until, finally, stem diameter did not vary at all.

Seedlings of Q. ilex sustained growth for 90 days after water withholding, after which

time the same pattern than U. minor of diameter shrinkage was observed. Common to

both species was the observation that stem shrinkage started when stem RWC

decreased below approximately 75% (see Fig. 4 and Fig. 5).

Leaf gas exchange

Defoliation increased abruptly in U. minor as stem RWC and md,stem dropped

below 80% and -2 MPa, respectively (Fig. 6). This created difficulties in quantifying

rates of gas exchange of water-stressed leaves of this species.


83

Fig. 6. Leaf shedding per plant expressed as a percentage of total foliage versus midday stem

water potential (md,stem) and stem relative water content (RWC) in Ulmus minor plants exposed to

water withholding. Sigmoid models are fitted to data. No leaf shedding was observed in control

plants of this species or control and droughted plants of Quercus ilex.

Leaf stomatal conductance to water vapor (gs) per unit area was 60% higher in

U. minor than Q. ilex under non-limiting water conditions (i.e. averaged for all C plants;

Fig. 7).


84

Fig. 7. Response of leaf stomatal conductance to water vapor (gs; above) and net CO2 uptake (Pn;

below) to changes in leaf relative water content for seedlings of Quercus ilex and Ulmus minor.

Exponential models are fitted to droughted plants (filled symbols), including those with incipient

symptoms of dieback (“Dying”). Notice the different Y-axes for each species.

Leaf net photosynthetic CO2 uptake (Pn) per unit area was 20% higher in U.

minor than Q. ilex in C plants, and differences reached 200% when values were

expressed per unit leaf dry mass (Fig. 7). Both variables declined steeply in D plants as

leaf RWC decreased down to 80%, and more flatly afterwards (Fig. 7). The rates of

decline of Pn and gs were similar in both species, as indicated by the non-significant

interactions between leaf RWC and species (both p > 0.15). Leaf dark respiration (Rleaf)

per unit area in U. minor was half that in Q. ilex at non-limiting water conditions (data

not shown); however, it was 40% higher in the former species when expressed per unit

dry mass (Fig. 8).


85

Fig. 8. Response of leaf (Rleaf; upper), stem (Rstem; middle) and root (Rroot; bottom) respiration to

changes in the relative water content of the respective organs for seedlings of Quercus ilex and

Ulmus minor. Models are fitted to droughted plants (filled symbols), including those with incipient

symptoms of dieback (“Dying”).

Rleaf declined with decreasing leaf RWC, less steeply than Pn. The rate of

decline of this variable was similar in both species (interaction term, p > 0.15), so that

mass-based Rleaf was higher in U. minor than Q. ilex also during the drought.

Stomatal conductance was minimal (i.e. <10 mmol m-2 s-1) and Pn was negative

(-4.3 ± 1.3 nmol g-1s-1; i.e. net CO2 efflux under saturating light) in dying plants of both

species (Fig. 7); however, mass-based Rleaf in dying plants was higher for U. minor


86

than Q. ilex (3.3 ± 0.9 vs 1.1 ± 0.2 nmol g-1s-1; p =0.025; Fig. 8). At the same time, C

plants of U. minor and Q. ilex exhibited values of Pn of 203 ± 42 and 59 ± 5 nmol g-1s-1,

and values of Rleaf of 9.2 ± 1.7 and 5.1 ± 0.7 nmol g-1s-1, respectively. Scaling Rleaf to

total foliage, differences between dying plants of both species were less evident: 6.5 ±

1.0 nmol s-1 for U. minor and 3.7 ± 0.8 nmol s-1 (p=0.067; data not shown).

Stem and root respiration

Respiration per unit root dry mass (Rroot) was on average 50% higher in U.

minor than Q. ilex at non-limiting water conditions (p<0.001). Rroot declined slightly in D

plants of Q. ilex as root RWC decreased, but not in D plants of U. minor (interaction

term, p=0.033; Fig. 8). Variability in Rroot was high, especially for U. minor. Contrary to

respiratory flux in leaves and roots, respiration per unit stem dry mass (Rstem) was 40%

lower in U. minor than Q. ilex at non-limiting water conditions (p < 0.001). In D plants,

this variable exhibited a significant decline with decreasing stem RWC, of similar

magnitude in both species (interaction term, p > 0.15; Fig. 8).

Dying plants of both species exhibited similar average values of Rstem (1.0 ± 0.2

nmol g-1s-1), but U. minor had three times higher Rroot than Q. ilex (1.0 ± 0.3 vs 2.9 ± 0.3

nmol g-1s-1; Fig. 8). At this time, for well-watered C plants, Rstem was 1.5 ± 0.3 nmol g-1s-

1 in U. minor and 2.3 ± 0.3 nmol g-1s-1 in Q. ilex, whereas Rroot was 2.1 ± 0.3 in U. minor

and 1.9 ± 0.5 nmol g-1s-1 in in Q. ilex. Scaling Rstem and Rroot to the organ-level by taking

into account stem and root biomass, respectively, differences between dying plants of

both species were more evident: total stem respiration was 4.5 times higher in U. minor

than Q. ilex (6.5 ± 3.1 vs 1.4 ± 0.4 nmol s-1; p= 0.039) and total root respiration was

also 8 times higher in U. minor than Q. ilex (27.8 ± 5.4 vs 3.5 ± 0.2 nmol s-1; p=0.002;

data not shown).

Whole-plant net carbon gain

Averaged across dates, instantaneous whole-plant net carbon gain during

daytime (PCG) was 5 times higher in U. minor than Q. ilex under well-watered

conditions (p < 0.001; Fig. 9), in spite of 3 times higher woody (stem + roots) biomass

in the former species (data not shown).


87

Fig. 9. Relationship between the instantaneous rate of whole-plant net carbon gain during day time

and midday stem water potential (md,stem) for seedlings of Quercus ilex and Ulmus minor. Sigmoid

models are fitted to droughted plants (filled symbols), including those with incipient symptoms of

dieback (“Dying”). Notice the different Y- and X-axes for each species.

In response to drought, PCG declined and reached negative values below -3

MPa in U. minor and -5 MPa in Q. ilex (Fig. 9), indicating carbon losses from leaf

photorespiration and leaf and woody mitochondrial respiration exceeded foliage carbon

uptake at different thresholds of water stress according to species. All dying plants

exhibited negative PCG values while, at the same time, well-watered C plants exhibited

positive PCG values: 525 ± 102 nmol plant-1s-1 in U. minor and 162 ± 25 nmol plant-1s-1

in Q. ilex.

Concentration of non-structural carbohydrates

Leaf, stem and root non-structural carbohydrates (NSC) concentrations in U.

minor were higher than in Q. ilex under well-watered conditions (all p < 0.001; Fig. 10)


88

Fig. 10. Response of leaf (upper), stem (middle) and root (bottom) concentrations of nonstructural

carbohydrates (NSC; as % of dry matter) to changes in midday stem water potential (md,stem) for

seedlings of Quercus ilex and Ulmus minor. Models are fitted to droughted plants (filled symbols),

including those with incipient symptoms of dieback (dying). Notice the different Y- and X-axes for

each species.

Soluble sugars (SS) were similar in the two species, whereas starch

concentrations were two times higher in the roots of U. minor than Q. ilex (25.7 vs

12.4%, respectively; p < 0.001), 9 times higher in the stems (5.8 vs 0.6%; p < 0.001)

and 10 times higher in the leaves (4.2 vs 0.4%; p < 0.001). The concentrations of SS

increased with increasing water stress in all plant organs except the stems of U. minor

(Table 2).


89

Table 2 Pearson product-moment correlation coefficients of logarithmically transformed data of leaf, stem,

root and plant non-structural carbohydrate (NSC) concentrations, and water status indicators leaf predawn

water potential (pd), midday stem (md,stem) water potential and organ relative water content (RWCi) for

seedlings of Quercus ilex and Ulmus minor. Values highlighted in bold indicate a significant correlation at p

< 0.05. SS = soluble sugars.

Q. ilex

U. minor

Leaf SS -0.63 -0.61 -0.46 -0.58 -0.48 -0.37

Leaf starch 0.15 0.15 0.08 0.57 0.73 0.51

Leaf NSC -0.63 -0.60 -0.46 0.26 0.53 0.31

Stem SS -0.66 -0.52 -0.56 0.13 0.15 0.02

Stem starch -0.23 -0.28 -0.20 0.26 0.33 0.14

Stem NSC -0.66 -0.52 -0.56 0.13 0.15 0.02

Root SS -0.56 -0.39 -0.29 -0.63 -0.62 -0.56

Root starch 0.49 0.49 0.38 0.65 0.65 0.52

Root NSC 0.31 0.35 0.30 0.55 0.56 0.40

Plant SS -0.68 -0.53 -0.57 -0.59

Plant starch 0.51 0.49 0.60 0.60

Plant NSC 0.15 0.17 0.49 0.50

pd md,stem RWCi pd md,stem RWCi

These increases were accompanied by significant decreases in starch

concentrations with increasing water stress in the roots of both species and the leaves

of U. minor. Considering the sum of SS and starch (i.e. total NSC), with increasing

water stress, NSC declined in the root of both species, but it increased in leaves and

stems of Q. ilex and declined in leaves of U. minor (Fig. 10; Table 2. Expressed at the

whole plant level, there was a trend for NSC to decline with water stress in U. minor but

not in Q. ilex (Table 2).

Dying plants of U. minor exhibited significantly lower NSC concentrations in the

roots and at the whole-plant level than controls, but the decline was not significant in

dying plants of Q. ilex (Figs. 10 and 12).


90

Fig. 12. Comparison of hydraulic and carbon-

related traits in seedlings of Quercus ilex and

Ulmus minor either kept well-watered (controls)

or exposed to severe drought and showing

symptoms of shoot dieback (“Dying” plants).

Data are means (±SE; n= 4-5) of midday stem

water potential (md,stem), root percentage loss of

hydraulic conductance (PLC), root concentration

of nonstructural carbohydrates (NSC) and

weighted-average whole-plant NSC concentration.

Asterisks mark significant differences between

means of control and dying plants as *, p < 0.05;

**, p < 0.01 and ***, p < 0.001. Different letters

mark significantly different means after Tukey

HSD tests.


91

Despite their declining NSC values, dying plants of U. minor still had twice as

much NSC concentrations in the roots and at the whole-plant level than dying plants of

Q. ilex.

Stem and root hydraulic conductance

The mean percentage loss of hydraulic conductance (PLC) caused by native

embolism in well-watered plants varied between 8-12% in stems and roots of Q. ilex

and 18-19% in stems and roots of U. minor, respectively. The PLC in stems and roots

of D plants increased as xylem water potential decreased (Fig. 11).

Fig. 11. Response of percentage loss of hydraulic conductance in stems (above) and roots (below)

to changes in midday stem water potential (md,stem) for seedlings of Quercus ilex and Ulmus minor.

Sigmoid models are fitted to droughted plants (filled symbols). Notice the different scales in X-axes

for each species.

Despite considerable variation among plants, it was noticeable that the stems

and roots of U. minor had a greater loss of hydraulic conductance with decreasing

md,stem than those of Q. ilex. The values of md,stem at which PLC was 50% (P50) in

stems and roots were -6.7 and -4.2 MPa for Q. ilex and -3.4 and -2.6 MPa for U. minor,

respectively.


92

In dying plants, root PLC was 75-80% (Figs. 11 and 12), and stem PLC was

lower in Q. ilex (60%) than U. minor (90%) at p =0.011. The actual specific hydraulic

conductivities of stems and roots of dying plants was, respectively, 0.49 ± 0.20 and

0.47 ± 0.20 Kg m-1 s-1 MPa-1 for Q. ilex, and 0.11 ± 0.03 and 0.17 ± 0.04 Kg m-1 s-1

MPa-1 for U. minor (means not significantly different between species at p < 0.05).

Discussion

Effects of drought on water transport, gas exchange and NSC dynamics

Cessation of irrigation provoked a strong down-regulation of transpiration by

stomatal closure in Q. ilex and U. minor. The decline in gs resulted in the gradient of

diurnal water potential (|md –pd|) declining as drought progressed, similarly in both

species, even though Q. ilex always exhibited higher |md –pd| – a consistent result

with its lower vulnerability to cavitation compared to U. minor (i.e. lower P50 in Q. ilex).

Stomata closure did not prevent increases in plant water stress in response to the

imposed drought, with the onset of drought being faster in U. minor due to its higher

foliage area. The increasing xylem tension caused the progressive embolization of

vessels, more steeply in roots than stems, as has been reported for other tree species

(e.g. Maherali et al., 2006). Below -2 MPa md,stem, leaf shedding increased markedly in

U. minor, a strategy to avoid runaway embolism occurring in some species in response

to increasing xylem tension and loss of conductivity (Fig. 6; Vilagrosa et al., 2003;

Pineda-García et al., 2013). Eventually, the maintenance of the drought treatment

caused severe hydraulic failure in some plants showing visible symptoms of decay, in

agreement with previous studies reporting dramatic reductions in xylem hydraulic

conductivity during experimental or natural lethal droughts (Davis et al., 2002;

Vilagrosa et al., 2003; Barigah et al., 2013; Urli et al., 2013).

The increasing levels of water stress and xylem tension provoked marked

reductions in Pn in both species, but lower and more variable changes in respiration

and NSC reserves between species. Respiration is often neglected in studies of plant

mortality. Although drought sometimes does not affect or even stimulates respiration

(see Atkin & Macherel, 2009, and Atkin et al., 2015 for aridity vs Rleaf relationships

across species), reductions in respiration rates of leaves (e.g. Rodríguez-Calcerrada et

al., 2011), stems (Rodríguez-Calcerrada et al., 2014) and roots (e.g. Bryla et al., 1997)

are commonly observed in response to short-term water deficits. These reductions are


93

generally lower than those in Pn, with the consequence that leaf (Rodríguez-Calcerrada

et al., 2012), and more so whole-plant, net carbon gain (PCG; Fig. 9) decrease with

drought. The balance between Pn and plant respiration impacts on carbon reserves.

However, predictions of carbon reserves from gas exchange must be taken with

caution. Leaf NSC concentrations may not vary (Piper, 2011) or even increase (Table 1;

Ayub et al., 2011) in response to drought despite decreasing Pn / Rleaf ratios, as a result

of drought-induced declines on carbon sink demand and/or SS export.

It is frequently discussed that drought-induced decreases in Pn can result in

respiration and growth becoming carbon limited. Here, based on the levels of SS in all

plant organs, it seems unlikely that the decline in plant respiration were originated by a

decline in the availability of respiratory substrates. Rather, drought-mediated declines

in respiration were more likely the result of a metabolic down-regulation, which reduces

the demand for respiratory energy (ATP), and in turn limits the availability of adenylates

(ADP) for respiration (Atkin & Macherel, 2009). The downregulation of energy

demanding processes during drought could be vital for survival because of the positive

consequences for NSC pools. In this sense, the reduction in secondary growth in both

species occurred prior to NSC reduction, suggesting it was related to water limitations

to cell expansion and division (Hsiao et al., 1976; Körner, 2003; Sala et al., 2012; Klein

et al., 2014). Higher avoidance of dehydration and xylem resistance to cavitation in Q.

ilex would have resulted in the ~40-day longer period of positive stem growth after

stopping watering compared to U. minor. Whereas, lower bark tissues in Q. ilex (R.

López, unpublished data) would have contributed to the ~40-day shorter period of stem

shrinkage compared to U. minor, given the importance of bark in storing water (Zweifel

et al., 2000; Pfautsch et al., 2015).

The concentrations of SS, starch and NSC varied with increasing water stress

differently among organs and species. One consistent response was that SS in all plant

organs tended to increase with increasing water stress, partly at the expense of starch.

This is a frequent response to water stress related with osmoregulation in leaves

(Rodríguez-Calcerrada et al., 2011; Warren et al., 2011; Warren et al., 2012; Duan et

al., 2013; Mitchell et al., 2013; Dickman et al., 2014), although in some cases the

opposite pattern has been found, i.e. accumulation of starch in water-stressed

conditions (Ayub et al., 2011; Anderegg et al., 2012; Adams et al., 2013). Data on roots

and stems are scarcer, sometimes pointing to the same conclusion that SS account for

an increasing proportion of NSC under drought (Rodríguez-Calcerrada et al., 2014;

Aguadé et al., 2015), but not in other times (Galvez et al., 2011, 2013). The decline in


94

root NSC reserves with increasing water stress was likely associated to continued root

respiration, particularly in U. minor, and limitations to phloem transport under drought

(Klein et al., 2014; Sevanto et al., 2014). Similarly, differences between species in

foliage and stem respiration and phloem sensitivity to water stress could explain the

opposite trend of NSC aboveground exhibited by the two species.

Physiological dysfunctions in dying plants

The imposed continued drought affected some plants that showed incipient

symptoms of dieback. It was remarkable that root PLC values were similarly high (75-

80%) in dying seedlings of both species, although these values were reached at

significantly different xylem tensions: -5.2 MPa for U. minor and less than -8 MPa for Q.

ilex (Fig. 12). This supports previous findings of plant death at similar and high PLC

values in angiosperms of contrasting vulnerability to cavitation and drought resistance

(Vilagrosa et al., 2003; Barigah et al., 2013; Nardini et al., 2013; Urli et al., 2013).

Based on these results, we suggest that root hydraulic failure has a leading role in

shoot dieback. In resprouters, severe hydraulic failure and shoot dieback may not

necessarily lead to plant death (Vilagrosa et al., 2003; Landhäusser and Lieffers 2012;

Galvez et al 2013); however, it would still be a primary factor in mortality because for a

long time shoot dieback would limit the replenishment of NSC reserves and hence

chances of producing new shoots (Dietze et al., 2014; Pratt et al., 2014). Both U. minor

and Q. ilex are species potentially capable of resprouting when the aerial part is lost;

here, however, only one seedling of U. minor that had experienced stem dieback

resprouted in the spring of the following growing season. Thus, differences in root NSC

between dying plants of each species did not translate into a different capacity to

resprout, a result that can be interpreted in different ways. First, it is possible that high

levels of root embolism prevented resprouting by impeding starch mobilization or water

supply to dormant buds. Second, the importance of NSC reserves for survival should

not be merely seen in relation to concentrations or pool sizes; rather, NSC reserves

need to be considered in relation to respiratory energy demands, that is, in relation to

how much the reduction in NSC is limiting for maintaining basic metabolic functions in a

given environment. In this sense, higher root biomass and Rroot in U. minor could make

it more demanding in NSC. When considering this point, we note that plants sampled

for NSC showed incipient symptoms of stem dieback, not complete dieback. NSC

reserves could change in the few days running from incipient to complete plant dieback;

i.e. reserves could diminish in both species further and down to a point – not


95

necessarily the same for each species – where they could no longer meet Rroot demand

to restore production of new shoots (see Fig. 13).

Fig. 13 Illustrative schematics summarizing some key functional alterations preceding drought-

induced mortality in plants of Q. ilex (a) and U. minor (b). The imposed treatment of drought causes

increasing xylem tension (xil) and percent loss of hydraulic conductivity (PLC), faster in U. minor

due to its lower ability to avoid dehydration and maintain water transport under xylem tension than

Q. ilex. Coincidently in both species, PLC in roots of dying plants (i.e. showing incipient symptoms

of shoot dieback) reaches ~80%, although at higher xylem tension in Q. ilex. Net photosynthetic

CO2 uptake (Pn) declines over time by an effect of drought to near-zero values in dying plants; rates

of respiration (R), however, decline more slowly and are still positive in dying plants. Lack of

drought effect on root R, together with high root biomass and specific rates of R in U. minor result

in larger depletions of nonstructural carbohydrates in the roots (and whole plants) of U. minor than

Q. ilex. No symptoms of carbon starvation are evident in dying plants (NSC > R demand). However,

in the few days running from incipient to complete dieback, root NSC could have declined to

values no longer enough to sustain R (i.e. hydraulically-driven metabolic failure by carbon

starvation; encircled area). Alternatively, water supply restriction to dormant buds caused by

massive root embolism could have impaired their sprouting (i.e. hydraulically-driven metabolic

failure by water starvation; squared area).

Some studies have reported that species exposed to prolonged periods of near-

zero Pn suffer greater declines in NSC than those exposed to shorter periods

(McDowell et al., 2008; Plaut et al., 2012; Woodruff et al., 2014). In our study, despite

Q. ilex having endured water stress restrictions to Pn for longer than U. minor, drought-

induced NSC declines relative to control plants were only apparent in U. minor. One

factor contributing to this was the shedding of leaves in U. minor at the onset of

moderate levels of water stress, when Pn was still positive. Shedding leaves during

drought periods can provide a mechanism to avoid plant dehydration, but also could

increase the likelihood of plants dying from carbon starvation. Another contributing

factor was the higher size of faster-growing U. minor, which together with high rates of

respiration even under severe drought resulted in high NSC consumption by woody

respiration. We emphasize that the depletion of NSC reserves during drought does not


96

only depend on hydraulically-driven constraints on carbon uptake, e.g. derived from

stomatal regulation or leaf shedding in plants relatively vulnerable to xylem cavitation,

but also on the extent of respiratory demands during drought.

Concluding statements

Imposition of soil drought on seedlings of Q. ilex – a species from dry to semi-

arid environments – and U. minor – a riparian tree – revealed that xylem cavitation

causing ~ 80% loss of root hydraulic conductivity was a common factor preceding

drought-induced mortality in both species. This result, together with similar results for

other angiosperms suggest xylem hydraulic failure is a primary physiological

dysfunction triggering death by drought. The similar level of root hydraulic failure in

dying seedlings of both species occurred at double the time since drought imposition

and 2.5 MPa lower xylem water potential in Q. ilex than U. minor, suggesting

differences in drought resistance between species rely on the ability to postpone xylem

tension and maintain water transport under high xylem tension, rather than tolerating

severe losses of conductivity. Moreover, the common hydraulic failure in dying

seedlings of both species was associated to a variable depletion of NSC reserves

between species. Compared to Q. ilex, early leaf shedding together with high plant

biomass and specific rates of respiration in water-stressed seedlings of U. minor

resulted in larger declines of NSC reserves before death relative to well-watered plants.

However, final NSC levels were still relatively high and twice as high as in dying plants

of Q. ilex, and they did not seem to confer any differential advantage for resprouting.

Collectively, the results suggest that root metabolic failure impeding resprouting is

initiated by root hydraulic failure at a time when there is no sign of carbon starvation,

which, however, may ensue once the stems die back, as respiratory rates continue to

be high at severe water stress (see Fig. 13). Mechanistic studies are needed to

determine which levels of NSC limit basic metabolic functions, such as osmotic

adjustment, resprouting capacity, or detoxification of reactive oxygen species derived

from oxidative stress during dehydration.

Acknowledgements

We acknowledge the assistance of Guillermo G. Gordaliza, Jorge Domínguez,

Roberto Salomón and Ricardo Álvarez-Flores during experimental preparations. We

thank Roma Zytkoviak for measurements of carbohydrates, and Jean-Marc Limousin,

Serge Rambal, Nicolas Martin-StPaul, Martin Venturas and José Carlos Miranda for

helpful comments on results. J. R-C was supported by a “Juan de la Cierva” fellowship


97

from the Spanish Ministry of Economy and Competitiveness. This study was funded by

the project OLMOS (AGL2012-35580). The support of the Australian Research Council

to O.K.A. (DP1093759 and CE140100008) is acknowledged.

References

Adams HD, Germino MJ, Breshears DD, Barron-Gafford GA, Guardiola-Claramonte M, Zou CB, Huxman TE. 2013. Nonstructural leaf carbohydrate dynamics of Pinus edulis during drought-induced tree mortality reveal role for carbon metabolism in mortality mechanism. New Phytologist 197, 1142-1151.

Aguadé D, Poyatos R, Gómez M, Oliva J, Martínez-Vilalta J. 2015. The role of defoliation and root rot pathogen infection in driving the mode of drought-related physiological decline in Scots pine (Pinus sylvestris L.). Tree Physiology 35, 229-242.

Allen CD, Macalady AK, Chenchouni H, Bachelet D, McDowell N, Vennetier M, Kitzberger T, Rigling A, Breshears DD, Hogg EH (Ted) et al., 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259, 660-684.

Anderegg WRL, Berry JA, Smith DD, Sperry JS, Anderegg LDL, Field CB. 2012. The roles of hydraulic and carbon stress in a widespread climate-induced forest die-off. Proceedings of the National Academy of Sciences, USA 109, 233-237.

Anderegg WRL, Anderegg LDL. 2013. Hydraulic and carbohydrate changes in experimental drought-induced mortality of saplings in two conifer species. Tree Physiology 33, 252-260.

Atkin OK, Macherel D. 2009. The crucial role of plant mitochondria in orchestrating drought tolerance. Annals of Botany 103, 581–597.

Atkin OK, Bloomfield KJ, Reich PB, Tjoelker MG, Asner GP, Bonal D, Bönisch G, Bradford MG, Cernusak LA, Cosio EG et al., 2015. Global variability in leaf respiration in relation to climate, plant functional types and leaf traits. New Phytologist 206, 614-636.

Ayub G, Smith RA, Tissue DT, Atkin OK. 2011. Impacts of drought on leaf respiration in darkness and light in Eucalyptus saligna exposed to industrial-age atmospheric CO2 and growth temperature. New Phytologist 190, 1003-1018.

Barigah TS, Charrier O, Douris M, Bonhomme M, Herbette S, Améglio T, Fichot R, Brignolas F, Cochard H. 2013. Water-stress induced xylem hydraulic failure is a causal factor of tree mortality in beech and poplar. Annals of Botany 112, 1431-1437.

Bartlett MK, Scoffoni C, Sack L. 2012. The determinants of leaf turgor loss point and prediction of drought tolerance of species and biomes: a global meta-analysis. Ecological Letters 15, 393–405.


98

Bryla DR, Bouma TJ, Eissenstat DM 1997. Root respiration in citrus acclimates to temperature and slows during drought. Plant, Cell & Environment 20, 1411-1420.

Davis SD, Ewers FW, Sperry JS, Portwood KA, Crocker C, Adams GC. 2002. Shoot dieback during prolonged drought in Ceanothus (Rhamnaceae) chaparral of California: a possible case of hydraulic failure. American Journal of Botany 89, 820-828.

Dickman LT, McDowell NG, Sevanto S, Pangle RE, Pockman WT. 2014. Carbohydrate dynamics and mortality in a piñon-juniper woodland under three future precipitation scenarios. Plant, Cell & Environment 38, 729-739.

Dietze MC, Sala A, Carbone MS, Czimczik CI, Mantooth JA, Richardson AD, Vargas R. 2014. Nonstructural carbon in woody plants. The Annual Review of Plant Biology 65, 667-687.

Duan H, Amthor JS, Duursma RA, O'Grady AP, Choat B, Tissue DT. 2013. Carbon dynamics of eucalypt seedlings exposed to progressive drought in elevated [CO2] and elevated temperature. Tree Physiology 33, 779-792.

Edgell SE, Noon SM. 1984. Effect of violation of normality on the t test of the correlation coefficient. Psychological Bulletin 95, 576-583.

Galiano L, Martínez-Vilalta J, Lloret F. 2011. Carbon reserves and canopy defoliation determine the recovery of Scots pine 4 yr after a drought episode. New Phytologist 190, 750-759.

Galvez DA, Landhäusser SM, Tyree MT. 2011. Root carbon reserve dynamics in aspen seedlings: does simulated drought induce reserve limitation? Tree Physiology 31, 250-257.

Galvez DA, Landhäusser SM, Tyree MT. 2013. Low root reserve accumulation during drought may lead to winter mortality in poplar seedlings. New Phytologist 198, 139-148.

Pineda-García F, Paz H, Meinzer FC. 2013. Drought resistance in early and late secondary successional species from a tropical dry forest: the interplay between xylem resistance to embolism, sapwood water storage and leaf shedding. Plant, Cell & Environment 36, 405-418.

Gruber A, Pirkebner D, Florian C, Oberhuber W. 2012. No evidence for depletion of carbohydrate pools in Scots pine (Pinus sylvestris L.) under drought stress. Plant Biology 14, 142-148.

Haissig BE, Dickson RE. 1979. Starch measurement in plant tissue using enzymatic hydrolysis. Physiologia Plantarum 47, 151-157.

Hansen J, Møller I. 1975. Percolation of starch and soluble carbohydrates from plant tissue for quantitative determination with anthrone. Analytical Biochemistry 68, 87-94.


99

Hsiao TC, Acevedo E, Fereres E, Henderson DW. 1976. Stress metabolism – water stress, growth, and osmotic adjustment. Philosophical Transactions of the Royal Society B: Biological Sciences 273, 479-500.

Klein T, Hoch G, Yakir D, Körner C. 2014. Drought stress, growth and nonstructural carbohydrate dynamics of pine trees in a semi-arid forest. Tree Physiology 34, 981-992.

Körner C. 2003. Carbon limitation in trees. Journal of Ecology 91, 4-17.

Landhäusser SM, Lieffers VJ. 2012. Defoliation increases risk of carbon starvation in root systems of mature aspen. Trees 26, 653-666.

López R, López de Heredia U, Collada C, Cano FJ, Emerson BC, Cochard H, Gil L. 2013. Vulnerability to cavitation, hydraulic efficiency, growth and survival in an insular pine (Pinus canariensis). Annals of Botany 111, 1167-1179.

Maček I, Pfanz H, Francetič V, Batič F, Vodnik D. 2005. Root respiration response to high CO2 concentrations in plants from natural CO2 springs. Environmental and Experimental Botany 54, 90-99.

Maherali H, Moura CF, Caldeira MC, Willson CJ, Jackson RB. 2006. Functional coordination between leaf gas exchange and vulnerability to xylem cavitation in temperate forest trees. Plant, Cell & Environment 29, 571-583.

Martin-StPaul NK, Longepierre D, Huc R, Delzon S, Burlett R, Joffre R, Rambal S, Cochard H. 2014. How reliable are methods to assess xylem vulnerability to cavitation? The issue of “open vessel” artifact in oaks. Tree Physiology 34, 894-905.

McDowell N, Pockman WT, Allen CD, Breshears DD, Cobb N, Kolb T, Plaut J, Sperry J, West A, Williams DG, Yepez EA. 2008. Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytologist 178, 719-739.

McDowell NG. 2011a. Mechanisms linking drought, hydraulics, carbon metabolism, and vegetation mortality. Plant Physiology 155, 1051-1059.

McDowell NG, Beerling DJ, Breshears DD, Fisher RA, Raffa KF, Stitt M. 2011b. The interdependence of mechanisms underlying climate-driven vegetation mortality. Trends in Ecology and Evolution 26, 523-532.

Mitchell PJ, O’Grady AP, Tissue DT, White DA, Ottenschlaeger ML, Pinkard EA. 2013. Drought response strategies define the relative contributions of hydraulic dysfunction and carbohydrate depletion during tree mortality. New Phytologist 197, 862-872.

Nardini A, Battistuzzo M, Savi T. 2013. Shoot desiccation and hydraulic failure in temperate woody angiosperms during an extreme summer drought. New Phytologist 200, 322-329.


100

Oleksyn J, Zytkowiak R, Karolewski P, Reich PB, Tjoelker MG. 2000. Genetic and environmental control of seasonal carbohydrate dynamics in trees of diverse Pinus sylvestris populations. Tree Physiology 20, 837-847.

Parker J, Patton RL. 1975. Effects of drought and defoliation on some metabolites in roots of black oak seedlings. Canadian Journal of Forest Research 5, 457-463.

Pedersen BS. 1998. The role of stress in the mortality of Midwestern oaks as indicated by growth prior to death. Ecology 79, 79-93.

Pfautsch S, Renard J, Tjoelker MG, Salih A. 2015. Phloem as capacitor: radial transfer of water into xylem of tree stems occurs via symplastic transport in ray parenchyma. Plant Physiology 167, 963-971.

Piper FI. 2011. Drought induces opposite changes in the concentration of non-structural carbohydrates of two evergreen Nothofagus species of differential drought resistance. Annals of Forest Science 68, 415-424.

Plaut JA, Yepez EA, Hill J, Pangle R, Sperry JS, Pockman WT, McDowell NG. 2012. Hydraulic limits preceding mortality in a piñon-juniper woodland under experimental drought. Plant, Cell & Environment 35, 1601-1617.

Pockman WT, Sperry JS. 2000. Vulnerability to xylem cavitation and the distribution of Sonoran desert vegetation. American Journal of Botany 87, 1287-1299.

Poorter H, Bühler J, van Dusschoten D, Climent J, Postma JA. 2012. Pot size matters: a meta-analysis of the effects of rooting volume on plant growth. Functional Plant Biology 39, 839-850.

Pratt RB, Jacobsen AL, Ramirez AR, Helms AM, Traugh CA, Tobin MF, Heffner MS, Davis SD. 2014. Mortality of resprouting chaparral shrubs after a fire and during a record drought; physiological mechanisms and demographic consequences. Global Change Biology 20, 893-907.

Ratkowsky DA. 1990. Handbook of nonlinear regression models. Marcel Dekker, Inc., New York and Basel, 241 p.

Rodríguez-Calcerrada J, Shahin O, del Rey MC, Rambal S. 2011. Opposite changes in leaf dark respiration and soluble sugars with drought in two Mediterranean oaks. Functional Plant Biology 38, 1004-1015.

Rodríguez-Calcerrada J, Limousin J-M, Martin-StPaul NK, Jaeger C, Rambal S. 2012. Gas exchange and leaf aging in an evergreen oak: causes and consequences for leaf carbon balance and canopy respiration. Tree Physiology 32, 464-477.

Rodríguez-Calcerrada J, Martin-StPaul NK, Lempereur M, Ourcival J-M, del Rey MC, Joffre R, Rambal S. 2014. Stem CO2 efflux and its contribution to ecosystem CO2 efflux decrease with drought in a Mediterranean forest stand. Agricultural and Forest Meteorology 195-196, 61-72.


101

Sala A, Piper F, Hoch G. 2010. Physiological mechanisms of drought-induced tree mortality are far from being resolved. New Phytologist 186, 274-281.

Sala A, Woodruff DR, Meinzer FC. 2012. Carbon dynamics in trees: feast or famine? Tree Physiology 32, 764-775.

Sevanto S, McDowell NG, Dickman LT, Pangle R, Pockman WT. 2014. How do trees die? A test of hydraulic failure and carbon starvation hypotheses. Plant, Cell & Environment 37, 153-161.

Tardieu F, Davies WJ. 1993. Integration of hydraulic and chemical signaling in the control of stomatal conductance and water status of droughted plants. Plant, Cell & Environment 16, 341-349.

Teskey RO, Saveyn A, Steppe K, McGuire MA (2008) Origin, fate and significance of CO2 in tree stems. New Phytologist 177, 17-32.

Tyree MT, Sperry JS. 1989. Vulnerability of xylem to cavitation and embolism. Annual Review of Plant Physiology and Plant Molecular Biology 40, 19-38.

Urli M, Porté AJ, Cochard H, Guengant Y, Burlett R, Delzon S. 2013. Xylem embolism threshold for catastrophic hydraulic failure in angiosperm trees. Tree Physiology 33, 672-683.

Venturas M, MacKinnon ED, Jacobsen AL, Pratt RB. 2015. Excising stem samples under water at native tension does not induce xylem cavitation. Plant, Cell & Environment 38, 1060-1068.

Vilagrosa A, Bellot J, Vallejo VR, Gil-Pelegrín E. 2003. Cavitation, stomatal conductance, and leaf dieback in seedlings of two co-occurring Mediterranean shrubs during an intense drought. Journal of Experimental Botany 54, 2015-2024.

Warren CR, Aranda I, Cano FJ. 2011. Responses to water stress of gas exchange and metabolites in Eucalyptus and Acacia spp. Plant, Cell & Environment 34, 1609-1629.

Warren CR, Aranda I, Cano FJ. 2012. Metabolomics demonstrates divergent responses of two Eucalyptus species to water stress. Metabolomics 8, 186-200.

Woodruff DR. 2014. The impacts of water stress on phloem transport in Douglas-fir trees. Tree Physiology 34, 5–14.

Woodruff DR, Meinzer FC, Marias DE, Sevanto S, Jenkins MW, McDowell NG. 2014. Linking nonstructural carbohydrate dynamics to gas exchange and leaf hydraulic behavior in Pinus edulis and Juniperus monosperma. New Phytologist 206, 411-421.


102

Zhang Y-J, Meinzer FC, Qi J-H, Goldstein G, Cao K-F. 2013. Midday stomatal conductance is more related to stem rather than leaf water status in subtropical deciduous and evergreen broadleaf trees. Plant, Cell & Environment 36, 149-158.

Zweifel R, Item H, Häsler R. 2000. Stem radius changes and their relation to stored water in stems of young Norway spruce trees. Trees 15, 50-57.

104

5. PHYSIOLOGICAL AND BIOCHEMICAL DIFFERENCES AMONG

Ulmus minor GENOTYPES SHOWING A GRADIENT OF RESISTANCE

TO DUTCH ELM DISEAS

Physiological and biochemical differences among Ulmus Minor genotypes showing a gradient fo resistance to Dutch elm disease

106

Physiological and biochemical differences among Ulmus minor genotypes

showing a gradient of resistance to Dutch elm disease

Meng Li, Rosana López, Martín Venturas, Juan Antonio Martín, Guillermo G

Gordaliza, Luis Gil, Jesús Rodríguez-Calcerrada

Submitted to Forest pathology


107

Summary

Dutch elm disease (DED) spread across Europe and North America in the

20th century killing most natural elm populations. Today, breeding programs aim at

identifying, propagating and studying elm clones resistant to DED. Here we have

compared the physiology and biochemistry of 6 genotypes of Ulmus minor of variable

DED resistance. Leaf gas exchange, water potential, stem hydraulic conductivity and

biochemical status were studied in 5-year-old trees of AB-AM2.4, M-DV2.3, M-

DV2×M-CC1.5 and M-DV1 and 6-year-old trees of VA-AP38 and BU-FL7 before and

after inoculation with Ophiostoma novo-ulmi. Leaf water potential and net

photosynthesis rates declined, while the percentage loss of hydraulic conductivity

(PLC) increased after the inoculation in susceptible trees. By the 21st day, leaf

predawn and midday water potential, stomatal conductance to water vapor and net

photosynthesis rates were lower, and PLC was higher in trees of susceptible (S)

genotypes inoculated with the pathogen than in control trees inoculated with water,

whereas no significant treatment effect was observed on these variables in the

resistant (R) genotypes. Fourier transform infrared spectroscopy analyses revealed a

different biochemical profile for branches of R and S clones. R clones showed higher

absorption peaks that could be assigned to phenolic compounds, saturated

hydrocarbons, cellulose and hemicellulose than S clones. The differences were more

marked at the end of the experiment than at the beginning, suggesting that R and S

clones responded differently to the inevitable wounding from inoculation and

repeated sampling over the experimental course. We hypothesize that a weak

activation of the defense system in response to experimental wounding can

contribute to the susceptibility of some genotypes to O. novo-ulmi. In turn, the decline

in shoot hydraulic conductivity and leaf carbon uptake caused by the infection further

exacerbates tree susceptibility to the fungus.


108

Introduction

In the twentieth century, two pandemics of Dutch elm disease (DED) spread

across Europe and North America. The first, detected in 1920, was caused by the

fungus Ophiostoma ulmi, whereas the second, which started in 1970 and is still

ongoing, was caused by the more aggressive Ophiostoma novo-ulmi (Brasier, 1991).

Both pandemics devastated elm populations; for example, in Europe, more than 90%

of mature Ulmus minor Mill. trees were lost to DED (Brasier and Buck 2001).

The pathogen causing DED is transmitted by several species of bark beetles,

mostly Scolytus and Hylurgopinus (family Curculionidae). These insects feed on twig

phloem where fungal spores are deposited and germinate. The fungal hyphae grow

into the outer vessels of xylem where they can spread upwards, downwards and

laterally through vessel pits, and sporulate as budding cells (yeast-like phase).

Several are the mechanisms by which DED affects its host. The fungus, once

inside xylem vessels, produces cell-wall degrading enzymes such as xylanases,

pectinases or glucosidases (Binz and Canevascini 1996; Przybył et al. 2006). Early

research on pathogenicity of O. ulmi suggested that the hydrophobin cerato-ulmin

was a wilt toxin that affected elm physiology by reducing transpiration and increasing

leaf respiration (Takai 1974; Richards 1993). However, further research showed that

cerato-ulmin-disrupted mutants of O. novo-ulmi were also pathogenic (Bowden et al.

1996). Cerato-ulmin is now considered a parasitic fitness factor which contributes to

the transmission of the disease by protecting spores from desiccation and increasing

their adherence to bark beetles (Temple et al. 1997; Bernier et al. 2015). Molecules

secreted by the fungus also enhance xylem vulnerability to cavitation by reducing the

sap surface tension and by degrading inter-vessel pit membranes (Newbanks et al.

1983; Solla and Gil 2002a; Ouellette et al. 2004). Xylem dysfunction is caused mainly

by induced embolism (Newbanks et al. 1983). Nevertheless, losses in hydraulic


109

conductivity are also due to vessel occlusion by tyloses and gels produced by the

tree to prevent fungal spreading as well as by substances secreted in the sap by the

fungus (Ouellette et al. 2004). As a consequence of reduced shoot hydraulic

conductivity, water supply to leaves ceases, leaf water potential declines, and the

stomata close to conserve the water homeostasis (Aldea et al. 2006; Venturas et al.

2014). Hydraulic failure has interactive impacts on tree carbon metabolism. The most

immediate impact is the reduction of leaf photosynthesis because of impaired water

and CO2 availability to mesophyll cells; but the fungal infection may also limit the

photosynthetic capacity because of the reduction of chlorophyll content and

photochemical efficiency of PSII in chloroplasts (Oliveira et al. 2012). Less

straightforward effects of fungal invasion on the tree carbon balance might arise from

a potential increase of respiration related to the upregulation of the defense system

(Landis and Hart 1972; Richards 1993). In the longer term, massive fungal spread

can cause massive embolism of the hydraulic system and dieback (Thomas et al.

2002; Wullschleger et al. 2004; Jacobsen et al. 2012). New shoots can be produced

from the roots, but eventually, sustained hydraulic and carbon limitations may

compromise the ability of the tree to resprout (Goodsman et al. 2013).

Elm trees have a variable resistance to DED. Breeding programs have

devoted a continuous effort towards the identification of resistant genotypes of U.

minor (Franke et al. 2004; Collin and Vernisson 2006; Martín et al. 2015), and

several genotypes of marked resistance to DED have been propagated ex-situ over

the last decades. This plant material has provided an excellent opportunity to

investigate the aspects underpinning DED resistance. For example, today we know

that elms with narrower and scattered vessels tend to be more resistant to DED than

those with larger xylem vessels or larger vessel groups (Solla and Gil 2002a), as

these later characteristics favor hydraulic conductance, fungus spreading and xylem

cavitation. Compared to anatomical studies, fewer studies have evaluated the

importance of physiological and biochemical characteristics for DED resistance.


110

Martín et al. (2005; 2008a) have compared the biochemical profiles of elm species

and genotypes of contrasted DED resistance with Fourier transform-infrared (FT-IR)

spectroscopy analysis. Actually, the use of FT-IR fingerprinting has proved useful for

detecting chemical changes after fungal infections (Dorado et al. 2001; Pandey and

Pitman 2003; Martín et al. 2005; 2008a). Moreover, research on water relations and

gas exchange in response to pathogen infection (e.g., Meinzer et al. 2004: Domec et

al. 2013) has rarely been conducted in relation to DED (Oliveira et al. 2012; Urban

and Dvořák 2014; Venturas et al. 2014).

Plants have constitutive phenological, physical and chemical barriers against

fungus infection. Early flushing can be a phenological barrier against DED

(Ghelardini and Santini 2009); physical barriers involve some properties of the plant

surfaces that prevent fungus penetration, such as hard cuticles or small stomatal

pores; whereas chemical barriers are mainly ‘phytoanticipins’, antimicrobial

compounds that are present in plants before fungal infection, or produced after

infection from pre-existed constituents (Vanetten et al. 1994; Iriti and Faoro 2009),

including saponins, cyanogenic glycosides and glucosinolates (Bowyer et al. 1995;

Osbourn 1996; Bidart-Bouzat and Kliebenstein 2008; Zagrobelny et al. 2008). Once

constitutive barriers fail, the fungus is recognized by plant cell receptors that elicit a

complex signaling cascade that eventually induces a defense response. It is worth

noting that sometimes the defense response is induced not by the pathogen itself,

but rather by concomitant factors such as wounding (Steele et al. 1998; Nagy et al.

2000; León et al. 2001; Garcia et al. 2015). Compartmentalization is a mechanism by

which trees isolate the pathogen in a localized area (Shigo and Tippett 1981; Pearce

1996). In the case of DED, the formation of suberized tyloses in xylem vessels

restricts the vertical spread of the fungus and produces the isolation of healthy

vessels from infected ones (Ouellette et al. 2004); more so in resistant than

susceptible clones from Ulmus x Hollandica (U. minor x U. glabra) (Elgersma 1973).

One of the most effective barrier against DED is probably the wall 4 of the CODIT


111

model (Shigo and Tippet 1981), constituted by suberized or lignified parenchyma

cells formed the novo by the cambium, which prevents the radial spread of the

pathogen from earlywood vessels towards the cambial cells (Ouellete et al. 2004;

Martín et al. 2005). However, compartmentalization may be insufficient for trees to

prevent fungal infections; rather, this response is usually integrated within other

biochemical responses involving the production of secondary metabolites such as

mansonones (elm phytoalexins; Duchesne et al. 1992; Ouellette and Rioux 1992;

Vanetten et al. 1994; Grayer and Kokubun 2001; Witzell and Martín 2008). Once

trees are infected, there may be a negative feedback of fungus spreading on

subsequent tolerance. Genotypes capable of maintaining relatively high rates of

hydraulic conductance and net CO2 assimilation after infection could have a larger

pool of non-structural carbohydrates, and thus more respiratory substrates to

synthesize defense compounds and new tissue after shoot dieback (Kozlowski 1992).

On the contrary, a rapid closure of stomata upon infection could slow down the

dispersion of pathogens through xylem sap and allow for the tree to restrict the

damages by compartmentalizing or killing the pathogen.

We hypothesized that contrasting levels of DED resistance are related to

differences in physiological and chemical profiles between clones, either prior or

subsequent to the infection. Thus, we have compared the physiology and

biochemistry of clones of U. minor of variable DED resistance before and after an

experimental inoculation with an aggressive strain of O. novo-ulmi. Clones of high,

intermediate and low resistance to DED were selected from those clonally

propagated by the Spanish Elm Breeding Program. Our objectives were to

investigate 1) if there are constitutive differences in leaf physiological and shoot

biochemical properties among these clones; 2) how fast and to what extent does the

inoculation with O. novo-ulmi affect their physiology and biochemistry; and 3) what

are the causes and implications of any potentially different response.


112

Materials and methods

Plant material

Twenty-four U. minor trees representing 6 genotypes (4 trees/genotype) were

selected for the experiment. The genotype selection was based on previous

evaluations of susceptibility to inoculation with Ophiostoma novo-ulmi performed by

the Spanish Elm Breeding Program. In this program those 6 genotypes (M-DV2.3-

‘Dehesa de Amaniel’, AB-AM2.4, BU-FL7, M-DV2×M-CC1.5, VA-AP38 and M-DV1)

were inoculated in 2011 and 2012. This way, 2 genotypes were of high resistance to

the disease, showing under 30% leaf wilting and branch dieback after inoculation

with O. novo-ulmi (“R” genotypes AB-AM2.4, M-DV2.3), 2 of medium resistance (30-

75% wilting and dieback; “M” genotypes BU-FL7, M-DV2×M-CC1.5) and 2 of low

resistance, showing over 75% leaf wilting soon after the inoculation (“S” genotypes

VA-AP38, M-DV1).

Experimental design

Trees were planted at the forest breeding center of Puerta de Hierro (Madrid,

Spain; 3º45’ N, 40º 27’ E; 595 m a.s.l.) in spring 2010, and maintained with

supplementary watering during summers to avoid water-deficit stress. Plants were

divided into 2 blocks, each including the 6 genotypes randomly planted at 1×1m

spacing and the two treatments applied in 2013: inoculation with O. novo-ulmi

(hereafter inoculated trees) and inoculation with distilled water (hereafter controls).

Trees were 5-year-old for AB-AM2.4, M-DV2.3, M-DV2×M-CC1.5 and M-DV1 and 6-

year-old for VA-AP38 and BU-FL7 when treatments were applied, as it is

recommended that U. minor trees older than 4 years are used when studying

resistance to DED (Solla et al. 2005).


113

The inoculation took place in the first week of May 2013, when leaves had

almost fully expanded. A 0.1 ml of spore suspension (106 conidia ml-1) was

inoculated by making a horizontal incision in the stem at approximately 10-15 cm

from the ground with a sharpened, sterilized razor blade reaching the xylem. The

Ophiostoma novo-ulmi Brasier ssp. americana Z - BU1 was used for inoculation. The

isolate was collected from DED infected trees in Bubierca (Zaragoza, Spain) in 2009.

Control trees were also cut and inoculated with distilled water in the same way.

Inoculations were made on a sunny day to make sure the solutions were effectively

taken up by trees.

Sampling for leaf and branch physiology was conducted on days 0, 7, 14, 21,

and 30 after the inoculations while sampling for branch biochemistry was conducted

on days 0 and 21. To decide which leaves or branches to sample, we first evaluated

the health status of the tree. Either dark-green or yellowing leaves were selected

according to the predominant status of tree foliage (healthy or yellowing); i.e. a

yellowing leaf was sampled when trees had > 50% of leaves yellowing. Similarly,

branches of similar diameter (approximately 0.8-1.0 cm) were sampled for hydraulic

conductance and chemistry studies according to the predominant health status of the

tree, i.e. branches had some yellowing leaves when most tree leaves were yellowing.

The idea behind the non-random sampling was to have a realistic picture of the

physiological and chemical changes that were taking place after the inoculation at the

tree level (see Domec et al. 2013). Withered leaves or branches containing withered

leaves were never sampled.

Gas exchange

Leaf gas exchange was measured with a LI-6400 portable gas exchange

system (Li-Cor Inc., NE, USA) equipped with a 6 cm2 chamber coupled to a LED light

source (LI-6400-02B) at the lower tree crown. Measurements were typically


114

conducted from 10:00 to 13:00 h under ambient air relative humidity, and artificially

adjusted micro-environmental conditions of 800 mol m−2 s-1 photon flux density, 400

ppm CO2 concentration, and 25ºC air temperature. Net photosynthesis rate (Pn),

stomatal conductance to water vapor (gs), internal CO2 concentration (Ci) and

transpiration rates (E) were recorded. Dark respiration rate (Rd) was subsequently

measured in the same leaf, but before, the leaf was covered with aluminum foil paper

for 30 minutes to prevent light-enhanced respiration (Atkin et al. 1998). Area-based

gas exchange rates were expressed per unit leaf dry mass after measuring the mass

of the leaf area enclosed in the cuvette.

Water potential

Predawn and midday leaf water potentials were measured with a pressure

chamber (PMS Instrument Company, Albany, OR, 142 USA) on the same days as

gas exchange and hydraulic conductance measurements.


Xylem hydraulic conductance was measured using a XYL’EM embolism-meter

(Bronkhorst, Montigny les Cormeilles, France). Stem segments from one year old

branches were selected to avoid the cold induced embolism from previous winter.

Branches from the upper tree crown (>0.75 m long) were cut under water, to avoid

air entering through the cut surface. They were placed with their cut end in a bucket

containing water, covered with double black plastic bags, and transported to the

laboratory (<10 min) where measurements were performed immediately. Their bark

was carefully removed from the end to be connected to the conductivity apparatus,

and both segment ends were shaved with a new razor. Initial hydraulic conductivity

(Ki) was measured with a solution of 10 mM KCl and 1 mM CaCl2 in deionized

degassed water at 2-3×10-3 MPa pressure. After measuring Ki, segments were


115

flushed with the same solution at 0.18 MPa for 10 minutes to remove native

embolisms and to obtain the maximum hydraulic conductivity (Kmax). The percentage

loss of hydraulic conductivity (PLC) was then calculated as: PLC (%) = 100 × (Kmax –

Ki) / Kmax. Initial stem specific conductivity (Ks) and maximum stem specific

conductivity (Ksmax) were calculated by dividing Ki and Kmax, respectively, by the

xylem cross-section area. All the leaves on the branches above the sampled

segment were collected and then scanned in order to know the total leaf area on the

branch (AL). Then, the specific leaf hydraulic conductivity (KL) was obtained by

dividing initial hydraulic conductivity by the total leaf area on the branch.

Biochemistry

Stem segments collected for biochemistry analyses were immediately frozen in

liquid nitrogen after cutting and later stored at -80ºC until required. Frozen branch

samples were lyophilized and milled into a fine powder to conduct Fourier transform-

infrared (FT-IR) spectroscopy analysis (600–4000 cm-1). An infrared

spectrophotometer FT-IR Perkin–Elmer 1600 (Perkin Elmer Inc, USA) was used to

obtain the absorption spectra of each sample. Each pellet was prepared with 240 mg

KBr and 3 mg of sample powder ground into an agate mortar uniformly, and then

pressed in a vacuum micro-environment using a pulling press (Perkin Elmer Inc, USA)

specially constructed for the preparation of KBr pellets for IR spectroscopy and pre-

adjusted at 10 ×104N. Three measurements were made per sample for obtaining the

final IR spectrum. Spectra were displayed at a resolution of 4 cm-1 at the absorbance

mode from 4000 to 600 cm−1. Information on the relationship between absorption

peaks, chemical assignments and tentative compound groups is given in Table 1.


116

Table 1. List of the most characteristic FT-IR bands and proposed structural assignments of the spectra from Ulmus minor samples.

Wavelenght (cm-1) Assignment Tentative functional groups Reference

3394 O–H stretching Carbohydrates and glycoconjugates Dorado et al. (2001), Pandey and Pitman (2003)

2922 C–H stretching Saturated hydrocarbons (fatty acids, suberin) Dorado et al. (2001), Pandey and Pitman (2003)

1738 C=O stretching Fatty acids, aliphatic and aromatic carbonyl-

containing compounds

Séné et al. (1994), Dorado et al. (2001)

1622 C=O, C–N, and C=C stretching Lignin, proteins, peptidoglycans Séné et al. (1994), Dorado et al. (2001), Pandey and

Pitman (2003)

1514 C=C stretching Aromatic ring in lignin Chen and McClure (2000), Dorado et al. (2001), Pandey

and Pitman (2003)

1427 C–H deformation Lignin, phenolic compounds Chen and McClure (2000), Pandey and Pitman (2003),

Séné et al. (1994)

1374 C–H deformation Cellulose and hemicellulose Pandey and Pitman (2003)

1318 C–H vibration Cellulose Pandey and Pitman (2003)

1248 C–O–H deformation and C–O stretching Phenolic compounds Séné et al. (1994)

1161 C–O–C vibration Cellulose Kacurakova et al. (2000), Pandey and Pitman (2003)

1108 C–O and C–C stretching Pectin Kacurakova et al. (2000)

1054 C–O stretching Cellulose, hemicellulose and pectin Kacurakova et al. (2000), Pandey and Pitman (2003)

888 C–H deformation Cellulose, hemicellulose and pectin Kacurakova et al. (2000), Pandey and Pitman (2003)

<825 C-H and O-H bending; O-O, O-OH and P-O-C

stretching, CH2 rocking; N-H wagging and

twisting

Polysaccharides (glucose, maltose, starch); cell wall

glycoconjugates; lipid peroxides; aromatic containing

compounds; phospholipids; nucleic acids; lipids;

nitrogen containing compounds

Pavlovic and Brandao (2003), Chalmers and Griffiths

(2002), Stuart (2004), Coates (2000)


117

Statistical analyses

Physiological variables were analyzed by repeated measures ANOVA to test

the main effects of time, inoculation treatment and susceptibility to DED, and their

interaction effects on data means. Measurement day was considered as the

repeated-measures factor, whereas genotype was nested into the DED susceptibility

factor. At each day, t-tests – or Welch tests when variances were significantly

different – were used to compare means between the two infection treatments, for

each group of DED susceptibility (high, intermediate and low). Control trees of each

group were compared by one-way ANOVAs before the inoculation to test for potential

constitutive differences among them. Some variables were log10-transformed so that

the test requirements of normality and homogeneity of variance were observed.

These analyses were performed with SPSS 17.0, with 95% confidence limits (p <

0.05).

FT-IR spectra were processed using OPUS 2.1 Software. For each spectrum,

a total of 19 peaks were measured. The base-line of each peak was first added and

then the height of the peak, from the maximum absorbance value to the peak

baseline, was measured. Analysis of variance was used to compare means of peak

heights among tree groups of different DED susceptibility, with Tukey post-hoc tests

further employed to separate significantly different groups. A principal component

analysis (PCA) of the 19 peak heights per sample was also used to explore for

potential differences in the broad biochemical profiles of clones with different DED

susceptibility and between treatments. The scatter plot of PC1 and PC2 scores

helped identifying any clustering in samples; loadings of original variables were also

shown in the plots to visually inspect the influence of specific peaks on data

scattering. Furthermore, the relationship between branch chemical profile (quantified

through the PC1) and the susceptibility to DED of each genotype (quantified through


118

the percentage of wilting after inoculation) was analyzed via linear regression

analysis. PCA analyses were made with SPSS 17.0.

Results

Wilting status

The results of wilting after inoculation with O. novo-ulmi of this experiment

confirmed the contrasted susceptibility among genotypes observed in the previous

susceptibility test (Table 2): wilting progressed faster in S than R clones. First

symptoms of wilting were found on day 14, and by day 21 more than 50% of tree

foliage exhibited wilting in S clones. Clones M and R also showed visible symptoms

of wilting, but to a lower extent (8-20%). Wilting increased to approximate 90, 50, and

20% in clones S, M and R, respectively, on day 30 (Table 2). Six months after

inoculation, 37.5 % of the S trees were dead, whereas all M and R clones were alive.

One susceptible tree of M-DV1 was infested on day 21.


119

Table 2. Evolution of wilting percentage for 6 genotypes of variable susceptibility to DED after inoculation with O. novo-ulmi (I) or water (C; controls).

Wilting leaves (%)

Genotype Previous DED susceptibility characterization Day 1 Day 14 Day 21 Day 30

I C I C I C I C

M-DV1 Susceptible (S) 0 0 5 ± 5 0 67.5 ± 7.5 12.5 ± 12.5 90 30 ± 30

VA-AP38 Susceptible (S) 0 0 12.5 ± 7.5 0 55 ± 35 2.5 ± 2.5 87.5 ± 7.5 10 ± 10

M-DV2*M-CC1.5 Intermediate (M) 0 0 5 ± 5 0 20 2.5 ± 2.5 72.5 ± 12.5 10 ± 10

BU-FL7 Intermediate (M) 0 0 0 0 20 0 30 ± 10 0

AB-AM2.4 Resistant (R) 0 0 0 0 7.5 ± 2.5 0 20 0

M-DV2.3 Resistant (R) 0 0 0 0 15 ± 5 0 22.5 ± 2.5 0


120

Water potential

No constitutive significant differences in pd (p = 0.43) or md (p = 0.22) were

observed among S, M and R clones before the inoculation. Afterwards, there were

clear differences in both variables among treatments and clones (Table 3).


121

Table 3. Repeated-measures ANOVA testing the effects of time, treatment, susceptibility to DED and the interactions of treatment × susceptibility, time × treatment,

and time × treatment × susceptibility on means of leaf and shoot physiological variables: predawn water potential (pd), midday water potential (md), leaf net

photosynthesis (Pn), stomatal conductance to water vapor (gs), transpiration rate (E), dark respiration (Rd), internal CO2 concentration (Ci), leaf mass per area

(LMA), initial specific stem hydraulic conductivity (Ks), maximum specific stem hydraulic conductivity (Ksmax), specific leaf hydraulic conductivity (KL) and

percentage loss of hydraulic conductivity (PLC).

Time Treatment Susceptibility Treatment ×Susceptibility Time × Treatment Time × Treatment ×Susceptibility

pd 0.00 0.00 0.00 0.00 0.00 0.04

md 0.00 0.00 0.43 0.00 0.00 0.02

Pn 0.00 0.01 0.05 0.66 0.00 0.56

gs 0.00 0.03 0.00 0.47 0.09 0.87

E 0.00 0.01 0.01 0.56 0.04 0.99

Rd 0.00 0.32 0.12 0.59 0.25 0.17

Ci 0.00 0.85 0.03 0.23 0.66 0.23

LMA 0.00 0.37 0.08 0.03 0.08 0.88

KL 0.65 0.86 0.95 0.42 0.36 0.57

Ks 0.08 0.64 0.23 0.94 0.01 0.40

Ksmax 0.01 0.96 0.08 0.56 0.18 0.29

PLC 0.10 0.26 0.55 0.93 0.00 0.38


122

Both pd and md were barely affected by the inoculation treatment in R

genotypes (Fig. 1). However, the response of M and S clones was similar and more

intense; differences between treatments were appreciable at day 21, when for example

control trees of M clones had 2 MPa higher md than M inoculated trees. Differences

between treatments persisted on day 30 although they were of lower magnitude and

statistical significance (Fig. 1).

Fig. 1. Time course of predawn (pd) and midday (md) water potentials of resistant (circles),

intermediate (squares) and susceptible (triangles) clones (each including 2 genotypes) subjected

to inoculation with O. novo-ulmi (solid symbols) or water (controls; open symbols). Symbols are

means ± standard error (n = 4). Differences between inoculation treatments are based on t-tests (p

< 0.05, *; p < 0.01, **; p < 0.001, ***).

Gas exchange

No constitutive significant differences in gs (p = 0.97), Pn (p = 0.92) or Rd (p =

0.42) were observed among S, M and R clones before the inoculation. Both gs and Pn

tended to decline after inoculation with O. novo-ulmi, although significant differences

between control and inoculated trees were only found in S clones (Fig. 2; Table 3).

The inoculated S trees had 3 and 10 times lower Pn than controls on days 21 and 30,


123

respectively. In the case of Rd, the only significant difference between inoculated and

control trees was found for S clones on day 30, when Rd was 2 times higher in the

inoculated trees (Fig. 2).


124

Fig. 2. Time course of leaf mass-based net photosynthesis (Pn), dark respiration (Rd) and stomatal

conductance to water vapor (gs) of resistant (circles), intermediate (squares) and susceptible

(triangles) clones (each including 2 genotypes) subjected to inoculation with O. novo-ulmi (solid

symbols) or water (controls; open symbols). Symbols are means ± standard error (n = 4).

Differences between inoculation treatments are based on t-tests (p < 0.05, *; p < 0.01, **; p < 0.001,

***).

Both gs and Pn in inoculated trees declined exponentially in response to

increasing water stress (Fig. 3). The decline appeared to be steeper in R clones, but

the fact that pd did not fall to values as low as in the other clones precluded a reliable

comparison of stomata sensitivity to water stress.


125

Fig. 3. Relationship between leaf predawn water potential (pd) and leaf mass-based stomatal

conductance to water vapor (gs) and net photosynthesis (Pn) in resistant (R; circles), intermediate

(M; squares) and susceptible (S; triangles) clones (each including 2 genotypes) with inoculation.


No constitutive significant differences in KL (p = 0.55), Ks-max (p = 0.51) or PLC (p

= 0.61) were observed among S, M and R clones before the inoculation. However, all

these variables were significantly affected by O. novo-ulmi in S clones. Significant

differences in KL were observed on day 21, when inoculated S trees had 2 times lower

water supply to leaves than controls. PLC was approximately more than 2 times higher

in S inoculated plants than in S controls on day 21. No significant differences between

treatments were observed for R and M clones for these variables. In addition, on day

21 Ks-max was significantly lower for S inoculated clones than controls (Fig. 4).


126

Fig. 4. Evolution in time of percent loss of hydraulic conductivity (PLC), maximum xylem specific

conductivity (Ks-max) and leaf-specific hydraulic conductivity (KL) of resistant, intermediate and

susceptible clones (each including 2 genotypes) subjected to inoculation with O. novo-ulmi (solid

bars) or water (controls; open bars). Symbols are means ± standard error (n = 4).

Biochemistry

Results of PCA of FT-IR spectral peaks indicated that clones separated

according to their susceptibility to DED. Including only control trees in the analysis, R

clones tended to occupy the positive side of the PC1 and S clones the negative, left

side (Fig. 5a). The same analysis applied to data of day 21 showed a clearer

separation between R and S clones along the PC1 (Fig. 5b). The absorption peaks

that weighted more on PC1 (and thus on the separation of susceptibility groups)

appeared at 3394 (O–H, N–H and C–H stretching), 2922 (C–H aliphatic stretching),


127

1738 (C=O stretching of ester groups), 1514 (aromatic C=C skeletal stretching), 1374

cm-1 (C–H symmetric deformation), and 1248 (C–O–C stretching) cm-1 (see Fig. 5 c, d

and Table 1). When plotting the average wilting percentage of clones (as a surrogate

of tree susceptibility to DED) against the average scores of PC1 (as a surrogate of

shoot chemical profile), a trend for S clones to have negative PC1 scores and R clones

to have positive PC1 scores was evident at day 0 and significant at day 21 (Fig. 5 e, f),

supporting the observation that the biochemical profile differs among clones according

to their susceptibility to DED. Considering individual absorption peaks, only that at

3394 cm-1 at day 0 and that at 2922 cm-1 at day 21 were significantly different among

clones (highest in R and lowest in S in both cases; Fig. 5 g, h).


128

Fig. 5. (a, b) PCA scatter plots of FT-IR spectroscopy data from branches of resistant (R; circles;

AB-AM2.4 and M-DV2.3 pooled), intermediate (M; squares; BU-FL7 and M-DV2×M-CC1.5 pooled)

and susceptible (S; triangles; VA-AP38 and M-DV1 pooled) clones on days 0 and 21 since

inoculation; (c, d) weight of absorption peaks (see table 1 for details on related chemical

compounds) in PCA on days 0 and 21 since inoculation; (e, f) relationship between average wilting

percentage of clones after inoculation (as a surrogate of tree susceptibility to DED) and the


129

average PC1 scores (as a surrogate of shoot chemical profile); (g, h) comparison of peak heights

at 3394 cm-1

(related with absorption of carbohydrates and glycoconjugates; day 0) and 2922 cm-1

(related with saturated hydrocarbons; day 21) among resistant (R), intermediate (M) and

susceptible (S) clones (each including 2 genotypes). Different letters indicate significant

differences at P ＜ 0.05 after Tukey’s HSD test in g) and h). A half black and white symbol in b)

refers to one susceptible tree with wilting leaves. All data are from control (c; open symbols) trees

inoculated with water at day 0.

In response to the inoculation with O. novo-ulmi, no clear distinction in the

biochemical profile was observed between inoculated and controls on any date; only a

light separation was visible for S clones at day 21 (Fig. 6).

Fig. 6. (a, b, c) PCA scatter plots of FT-IR spectroscopy data from branches of resistant (R; circles;

AB-AM2.4 and M-DV2.3 pooled), intermediate (M; squares; BU-FL7 and M-DV2×M-CC1.5 pooled)

and susceptible (S; triangles; VA-AP38 and M-DV1 pooled) clones on day 21 since inoculation with

O. novo-ulmi (i; solid symbols) or water (c; open symbols).

Pooling data of infected and control trees to analyze differences in absorption

peaks among clones, we observed significantly higher peaks at 2922, 1374, 1248 and

888 cm-1 in R than S clones, with M clones showing intermediate values (Fig. 7).


130

Fig. 7. Comparison of absorption peak heights at 2922, 1374, 1248 and 888 cm-1

wavelengths

among resistant (R), intermediate (M) and susceptible (S) clones (each including 2 genotypes) on

day 21 after inoculation. Different letters indicate significant differences at P ＜ 0.05. Trees

inoculated with O. novo-ulmi or water (controls) were pooled based on lack of chemical differences

between them (see Fig. 6).

Discussion

Thousands of U. minor genotypes have been screened for DED resistance in the

last decades within the Spanish Elm Breeding Program. Thanks to this continuous

effort, today, seven U. minor genotypes – including the M-DV2.3 genotype studied here

– have been registered as resistant clones by the Spanish Administration and are

planned to be used in reforestations in their natural habitats (Martín et al. 2015). Indeed,

here, clones M-DV2.3 and AB-AM2.4 showed a remarkable resistance to DED. These

clones exhibited a different biochemical profile than more susceptible ones and their

basic physiological functions – tree water status, shoot hydraulic conductivity and leaf

gas exchange – were not affected for 30 days since the inoculation with the pathogen

O. novo-ulmi.

By combining carbon-related and hydraulic measurements we delineated the

sequence of functional changes occurring in U. minor trees when inoculated with O.


131

novo-ulmi. Susceptible clones started to show physiological alterations as soon as 14

days after the inoculation; shoot hydraulic conductivity decreased relative to control

trees, resulting in lower supply of water to individual leaves, lower water potentials, and

lower rates of stomatal conductance and net CO2 assimilation. The reduction in

hydraulic conductivity and water potential is consequence of pathogen infection,

something observed in some vascular tree diseases (Duniway 1973), and that has

previously been reported for DED (Newbanks et al. 1983; Urban and Dvořák 2014).

This can be due to induced embolism or to direct vessel blockage by the fungus and

compounds it secretes. The accumulation of gels and tyloses can also block xylem

vessels of U. minor trees in response to O. novo-ulmi (Ouellette et al. 2004) and

explain the associated decline in water transport capacity. Actually, the significant

differences observed in susceptible clone’s hydraulic parameters 21 days after

inoculation indicate that both induced embolism and vessel blockage contribute to

losses in conductivity. If increased PLC was only due to induced embolism, Ksmax for

inoculated and control plants would have been equal after flushing. However, Ksmax was

reduced to half that of control plants. On the other hand, the high reduction in native Ks,

which lead to 85% PLC, cannot be explained only by vessel blockage, supporting

Newbanks et al. (1983) induced embolism hypothesis. It is to be noted that U. minor is

highly susceptible to drought-induced cavitation (Venturas et al. 2013, 2014) and any

conductivity losses may lead to run-away cavitation. These restrictions in leaf water

potential caused a stomatal limitation to photosynthesis, as suggested by the similar

leaf internal concentration of CO2 in infected and control susceptible trees (data not

shown). However, other studies have reported that fungal infections can limit leaf

photochemical efficiency and photosynthetic capacity (Santos et al. 2005; Oliveira et al.

2012). Contrarily, an increase in respiration can take place soon upon pathogen

inoculation in response to an increased demand of respiratory products to set up

defense mechanisms (Landis and Hart 1972). Leaf dark respiration increased in

inoculated susceptible trees relative to controls, but only in senescing leaves late after

inoculation, which does not point to a defense response but to a senescence-related

peak (Tetley and Thimann 1974). It is possible that leaf water stress, which typically

causes a reduction in mitochondrial respiration rates (e.g. Rodríguez-Calcerrada et al.

2011), offset any defense-induced increase in respiration.

Physiological traits measured here did not show any difference indicative of

constitutive resistance among clones. However, the fact that resistant clones

maintained near-optimal tree water status, shoot hydraulic conductivity, foliage area,

and leaf gas exchange for one month upon inoculation with O. novo-ulmi feeds forward


132

on the capacity to overcome the fungal infection. Maintaining foliage area and net CO2

assimilation ensures that non-structural carbohydrate reserves follow normal seasonal

dynamics. On the contrary, susceptible clones suffering a progressive loss of water

transport capacity and CO2 assimilation after inoculation will likely suffer a progressive

depletion of non-structural carbohydrate reserves, more so if respiratory carbon losses

remain unchanged or increase, as it seems to occur (see above). In support of this, i)

higher levels of starch were found in the resistant U. pumila than the susceptible U.

minor and also in resistant U. minor genotypes relative to susceptible ones after

inoculation with O. novo-ulmi (Martín et al. 2008a); ii) photosynthesis, and the

concentrations of soluble sugar and starch decreased after O. novo-ulmi inoculation in

in vitro cultures of U. minor (Oliveira et al. 2012); iii) whereas leaf gas exchange and

carbon and nutrient accumulation were not affected by O. novo-ulmi in mature resistant

‘Dodoens’ elm (U. glabra ‘Exoniensis’ × U. wallichiana) hybrids (Ďurkovič et al. 2012).

The shortage of carbon reserves, in turn, could limit the synthesis of defense

compounds and compromise the capacity to recover from shoot dieback (McDowell et

al. 2008). While maintaining near-optimal water transport and carbon balance may help

resistant trees to surmount the fungal infection, it remains unclear why the water

transport is barely affected by the pathogen in these trees, while it is fast and severely

restricted in others. One possibility that has been previously discussed is that clones

with bigger vessels (Solla and Gil 2002b; Venturas et al. 2014) or bordered pits (Martín

et al. 2009) are more prone to cavitate upon fungal inoculation or to facilitate the

spread of fungal toxins and propagules. However, only the R clone M-DV2.3 showed

certain anatomical properties (low proportion of large vessels) that could be related

with resistance (Unpublished data). Another possibility is that resistant clones are able

to restrict fungal spreading to areas close to the inoculation without producing gels and

tyloses that block xylem vessels. The possession or rapid induction of certain defense

compounds plays an important role in tree resistance to fungal infection (Duchesne et

al. 1992; Grayer and Kokubun 2001; Witzell and Martín 2008). Through FT-IR

spectroscopy analysis, here we explored the possibility that resistant clones displayed

a broad chemical profile (before or after the inoculation) that might have conducted to a

capacity to defend against fungal infection without compromising water transport

capacity.

Before the inoculation, resistant clones tended to separate chemically from

susceptible ones. Resistant clones scored higher in the first component of the PCA

positively related with carbohydrates, suberin, fatty acids, cellulose and hemicellulose,

and phenolic compounds (Fig. 7). An analysis of individual absorption peaks indicated


133

that the peak at 3394 cm-1, was higher in resistant than susceptible clones. This

intense peak, although rather unspecific, could be mainly assigned to carbohydrate and

glycoconjugate biochemical compounds (Table 1). As mentioned above, a higher pool

of non-structural carbohydrates could lead to a better capacity to stop fungal infections,

due to the need of carbon substrates for the secondary metabolism (Goodsman et al.

2013; Iriti and Faoro 2009). The separation among clones using control trees was

verified on day 21. Actually, the trend for more resistant clones to score higher in the

PC1 was more evident and significant at p < 0.05, which has led us to hypothesize that

repeated wounding – at stem base for water inoculation and tree top for hydraulic and

biochemical sampling – enhanced the capacity to chemically defense against fungal

infection. There are numerous examples of wound-induced defense activation (León et

al. 2001; Nagy et al. 2000; Smith 2015; Steele et al. 1998; Vek et al. 2013), but this

possibility has never been suggested in the case of DED. The resistant clones had a

higher peak at 2922 cm-1 than susceptible clones, band characteristic of fatty acids and,

in particular, of suberin-like substances (Martín et al. 2008b). Suberin is typically

induced by wounding, at least locally, to isolate the injured parts of the tree (Biggs 1987;

Pollard et al. 2008) and was reported to be induced in elms by chemical stress (Martín

et al. 2008b; 2010). Suberin is an important cell wall phenolic polymer involved in

preventing pathogen infection through the “compartmentalization of decay in trees”

(CODIT; Shigo 1984). As a hydrophobic waxy substance, suberin could also maintain

xylem functionality by reducing the occurrence of cavitation (Pearce 1996; Franke and

Schreiber 2007). Furthermore, both fatty acids and their derivatives jasmonic acid and

oxylipins have been well reported in wound responses (Weber 2002; Tumlinson and

Engelberth 2008; Upchurch 2008) and can induce systemic resistance against fungal

infections (Kachroo and Kachroo 2009). In elms, soil application of carvacrol induced

the accumulation of 16C and 18C fatty acids in xylem tissues and increased the plant

tolerance to O. novo-ulmi (Martín et al. 2012).

Constitutive (phytoanticipins) or induced (phytoalexins) phenolic compounds

are common and widespread secondary metabolites that play a role in defense. After

pooling branches of control and infected trees, which surprisingly barely varied

throughout the experimental course, it was observed that resistant clones had higher

absorption peaks at 1248 cm-1, band that could be associated to C–O–H deformation

and C–O stretching in phenolic compounds (Séné et al. 1994), and at 1374 and 888

cm-1, bands which could be mainly assigned to cellulose and hemicellulose

(Kacurakova et al. 2000; Pandey and Pitman 2003). Phenolic compounds play a

double role as antimicrobial and signaling compounds (Cvikrova et al. 2006;

http://en.wikipedia.org/wiki/Hydrophobic


134

Hammerschmidt 2005; Lattanzio et al. 2006; Witzell and Martín 2008). For example,

the accumulation of phenolic compounds is necessary for suberin biosynthesis

(Kolattukudy 1981; Pollard et al. 2008) and inhibits growth and sporulation of O. novo-

ulmi (Martín et al. 2007; ,2010; Rioux and Ouellette 1991). Cellulose is a component of

the cell wall and was found to be in higher amounts in the resistant U. pumila than the

susceptible U. minor, and in more resistant genotypes of U. minor inoculated with O.

novo-ulmi (Martín et al. 2008a). Again, the fact that differences among clones in height

of peaks at 1374, 1248 and 888 cm-1 were not observed at day 0, but at day 21 after

repeated sampling injury to trees, suggests that a chemical response favorable to cope

with O. novo-ulmi was induced by wounding, and that this response was of higher

extent in resistant than susceptible clones. In line with this, it has been observed that

wounding promotes the accumulation of phenolic compounds (Kemp and Burden 1986;

Klepzig et al. 1995), whereas insect attacks can prompt the reinforcement of the cell

walls with more cellulose and reduce cell susceptibility to subsequent fungal attacks

(Furstenberg-Hagg et al. 2013; Mohebby 2005). This idea is consistent with the

transmission of DED by bark beetles, which cause hundreds of small wounds in

branches when feeding or laying eggs (Peacock et al. 1981; Webber 2004).

In conclusion, trees of clones M-DV2.3 and AB-AM2.4 not inoculated with O.

novo-ulmi had a different biochemical profile than those of more DED-susceptible

clones VA-AP38 and M-DV1. In particular, FT-IR spectroscopy data suggest that

resistant clones M-DV2.3 and AB-AM2.4 had higher pools of saturated hydrocarbons

(suberin and fatty acids), phenolic compounds, cellulose, and hemicellulose. All these

compounds are well known for playing direct or indirect roles in defense, which

suggests that part of the resistance of M-DV2.3 and AB-AM2.4 to DED is related with a

better capacity to restrict the fungal spreading. Because differences among clones

were observed after 21 days of repeated destructive sampling, we hypothesize that

wounding induced the activation of chemical defense mechanisms more in resistant

than susceptible clones. Future experiments must be specifically planned to

corroborate this hypothesis. In any case, the higher capacity of resistant clones to

restrict pathogen infection is reflected in the maintenance of shoot hydraulic

conductivity, and leaf water status and gas exchange, which feed forward on the

capacity of these clones to cope with the fungus and/or recover from eventual

damages.

Acknowledgments


135

We thank Paz Andrés for assistance and guidance in the use of the infrared

spectrophotometer FT-IR Perkin–Elmer 1600. We also thank Pedro Perdiguero and

Carmen Collada for helpful discussions. Funding was provided by the project OLMOS

(AGL2012-35580).


136

References

Aldea M, Hamilton JG, Resti JP, Zangerl AR, Berenbaum MR, Frank TD, Delucia EH. 2006. Comparison of photosynthetic damage from arthropod herbivory and pathogen infection in understory hardwood saplings. Oecologia 149, 221-232.

Atkin OK, Evans JR, Siebke K. 1998. Relationship between the inhibition of leaf respiration by light and enhancement of leaf dark respiration following light treatment. Functional Plant Biology 25, 437-443.

Bernier L, Aoun M, Bouvet GF, Comeau A, Dufour J, Naruzawa ES, Nigg M, Plourde KV. 2015. Genomics of the Dutch elm disease pathosystem: are we there yet? iForest - Biogeosciences and Forestry 8, 149-157.

Bidart-Bouzat MG, Kliebenstein DJ. 2008. Differential levels of insect herbivory in the field associated with genotypic variation in glucosinolates in Arabidopsis thaliana. Journal of Chemical Ecology 34, 1026-1037.

Biggs AR. 1987. Occurrence and Location of Suberin in Wound Reaction Zones in Xylem of 17 Tree Species. Phytopathology 77, 718-725.

Binz T, Canevascini G. 1996. Xylanases from the Dutch elm disease pathogens Ophiostoma ulmi and Ophiostoma novo-ulmi. Physiological and Molecular Plant Pathology 49, 159-175.

Bowden CG, Smalley E, Guries RP, Hubbes M, Temple B, Horgan PA. 1996. Lack of association between cerato-ulmin production and virulence in Ophiostoma novo-ulmi. Molecular Plant-Microbe Interactions 9, 556-564.

Bowyer P, Clarke B, Lunness P, Daniels M, Osbourn A. 1995. Host range of a plant pathogenic fungus determined by a saponin detoxifying enzyme. Science 267, 371-374.

Brasier CM. 1991: Ophiostoma novo-ulmi sp. nov., causative agent of current Dutch elm disease pandemics. Mycopathologia 115, 151-161.

Brasier CM, Buck KW. 2001. Rapid Evolutionary Changes in a Globally Invading Fungal Pathogen (Dutch Elm Disease). Biological Invasions 3, 223-233.

Chalmers JM, Griffiths PR. 2002. Handbook of vibrational spectroscopy. Wiley, Chichester, UK

Chen M, Mcclure JW. 2000. Altered lignin composition in phenylalanine ammonia-lyase-inhibited radish seedlings: implications for seed-derived sinapoyl esters as lignin precursors. Phytochemistry, 53, 365-370.

Choat B, Gambetta GA, Wada H, Shackel KA, Matthews MA. 2009. The effects of Pierce's disease on leaf and petiole hydraulic conductance in Vitis vinifera cv. Chardonnay. Physiologia plantarum, 136, 384-394.


137

Coates J. 2000. Interpretation of infrared spectra, a practical approach. Encyclopedia of analytical chemistry.

Collin E, Vernisson F. 2006. Conservation and utilisation of genetic resources of European elm species. Workshop on Genetics 18-22.

Cvikrova M, Mala J, Hrubcova M, Eder J. 2006. Soluble and cell wall-bound phenolics and lignin in Ascocalyx abietina infected Norway spruces. Plant Science 170, 563-570.

Domec JC, Rivera LN, King JS, Peszlen I, Hain F, Smith B, Frampton J. 2013. Hemlock woolly adelgid (Adelges tsugae) infestation affects water and carbon relations of eastern hemlock (Tsuga canadensis) and Carolina hemlock (Tsuga caroliniana). New Phytologist 199, 452-463.

Dorado J, Almendros G, Field JA, Sierra-Alvarez R. 2001. Infrared spectroscopy analysis of hemp (Cannabis sativa) after selective delignification by Bjerkandera sp at different nitrogen levels. Enzyme and Microbial Technology 28, 550-559.

Duchesne L, Hubbes M, Jeng R. 1992. Biochemistry and molecular biology of defense reactions in the xylem of angiosperm trees. In: Defense Mechanisms of Woody Plants Against Fungi: Springer, pp. 133-146.

Duniway JM. 1973. Pathogen-Induced Changes in Host Water Relations. Phytopathology 63, 458-466.

Ďurkovič J, Čaňová I, Lagaňa R, Kučerová V, Moravčík M, Priwitzer T, Urban J, Dvořák M, Krajňáková J. 2012. Leaf trait dissimilarities between Dutch elm hybrids with a contrasting tolerance to Dutch elm disease. Annals of Botany

Elgersma DM. 1973. Tylose formation in elms after inoculation with Ceratocystis ulmi, a possible resistance mechanism. Netherlands Journal of Plant Pathology 79, 218-220.

Franke A, Rusanen M, de Aguiar A, Collin E, Harvengt L, Ackzell L, Bohnens J, Diamandis S. 2004. Methods and progress in the conservation of elm genetic resources in Europe. Investigación agraria. Sistemas y recursos forestales 13, 261-272.

Franke R, Schreiber L. 2007. Suberin-a biopolyester forming apoplastic plant interfaces. Current Opinion in Plant Biology 10, 252-259.

Furstenberg-Hagg J, Zagrobelny M, Bak S. 2013. Plant Defense against Insect Herbivores. International Journal of Molecular Sciences 14, 10242-10297.

Garcia T, Gutierrez J, Veloso J, Gago-Fuentes R, Diaz J. 2015. Wounding induces local resistance but systemic susceptibility to Botrytis cinerea in pepper plants. J Plant Physiol 176, 202-209.


138

Ghelardini L, Santini A. 2009. Avoidance by early flushing: a new perspective on Dutch elm disease research. iForest - Biogeosciences and Forestry 2, 143-153.

Goodsman DW, Lusebrink I, Landhausser SM, Erbilgin N, Lieffers VJ. 2013. Variation in carbon availability, defense chemistry and susceptibility to fungal invasion along the stems of mature trees. New Phytologist 197, 586-594.

Grayer RJ, Kokubun T. 2001. Plant-fungal interactions: the search for phytoalexins and other antifungal compounds from higher plants. Phytochemistry 56, 253-263.

Hammerschmidt R. 2005. Phenols and plant-pathogen interactions: The saga continues. Physiological and Molecular Plant Pathology 66, 77-78.

Iriti M, Faoro F. 2009. Chemical diversity and defence metabolism: how plants cope with pathogens and ozone pollution. International Journal of Molecular Sciences 10, 3371-3399.

Jacobsen AL, Roets F, Jacobs SM, Esler KJ, Pratt RB. 2012: Dieback and mortality of South African fynbos shrubs is likely driven by a novel pathogen and pathogen-induced hydraulic failure. Austral Ecology 37, 227-235.

Kachroo A, Kachroo P. 2009. Fatty Acid-derived signals in plant defense. Annual Review of Phytopathology 47, 153-176.

Kacurakova M, Capek P, Sasinkova V, Wellner N, Ebringerova A. 2000. FT-IR study of plant cell wall model compounds: pectic polysaccharides and hemicelluloses. Carbohydr Polym, 43, 195-203

Kemp MS, Burden RS. 1986. Phytoalexins and stress metabolites in the sapwood of trees. Phytochemistry 25, 1261-1269.

Klepzig K, Kruger E, Smalley E, Raffa K. 1995. Effects of biotic and abiotic stress on induced accumulation of terpenes and phenolics in red pines inoculated with bark beetle-vectored fungus. Journal of Chemical Ecology 21, 601-626.

Kolattukudy P. 1981. Structure, biosynthesis, and biodegradation of cutin and suberin. Annual Review of Plant Physiology 32, 539-567.

Kozlowski TT. 1992. Carbohydrate sources and sinks in woody plants. Botanical Review 58, 107-222.

Landis WR, Hart JH. 1972. Physiological Changes in Pathogen-Free Tissue of Ulmus Americana Induced by Ceratocystis Ulmi. Phytopathology 62, 909-913.

Lattanzio V, Lattanzio VM, Cardinali A. 2006. Role of phenolics in the resistance mechanisms of plants against fungal pathogens and insects. Phytochemistry: Advances in research 661, 23-67.


139

León J, Rojo E, Sánchez‐Serrano JJ. 2001. Wound signalling in plants. Journal of

Experimental Botany 52, 1-9.

Martín JA, Solla A, Woodward S, Gil L. 2005. Fourier transform-infrared spectroscopy as a new method for evaluating host resistance in the Dutch elm disease complex. Tree Physiol, 25, 1331-1338.

Martín JA, Solla A, Woodward S, Gil L. 2007. Detection of differential changes in lignin composition of elm xylem tissues inoculated with Ophiostoma novo-ulmi using Fourier transform-infrared spectroscopy. For Pathol, 37, 187-191.

Martín JA, Solla A, Coimbra MA, Gil L. 2008a. Metabolic fingerprinting allows discrimination between Ulmus pumila and U. minor, and between U. minor clones of different susceptibility to Dutch elm disease. Forest Pathology 38, 244-256.

Martín JA, Solla A, Domingues MR, Coimbra MA, Gil L. 2008b. Exogenous phenol increase resistance of Ulmus minor to Dutch elm disease through formation of suberin-like compounds on xylem tissues. Environmental and Experimental Botany 64, 97-104.

Martín JA, Solla A, Esteban LG, de Palacios P, Gil L. 2009. Bordered pit and ray morphology involvement in elm resistance to Ophiostoma novo-ulmi. Canadian Journal of Forest Research 39, 420-429.

Martín JA, Solla A, Gil L, García-Vallejo MC. 2010. Phenological and histochemical changes in Ulmus minor due to root absorption of phenol: implications for resistance to DED. Environmental and Experimental Botany, 69, 175- 182.

Martín JA, Solla A, Garcia-Vallejo MC, Gil L. 2012. Chemical changes in Ulmus minor xylem tissue after salicylic acid or carvacrol treatments are associated with enhanced resistance to Ophiostoma novo-ulmi. Phytochemistry, 83, 104-109.

Martín JA, Solla A, Venturas M, Collada C, Domínguez J, Miranda E, Fuentes-Utrilla P, Burón M, Iglesias S, Gil L. 2015. Seven Ulmus minor clones tolerant to Ophiostoma novo-ulmi registered as forest reproductive material in Spain. iForest-Biogeosciences and Forestry, 8, 172-180.

McDowell N, Pockman WT, Allen CD, Breshears DD, Cobb N, Kolb T, Plaut J, Sperry J, West A, Williams DG. 2008. Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytologist 178, 719-739.

Meinzer FC, Woodruff DR, Shaw DC. 2004. Integrated responses of hydraulic architecture, water and carbon relations of western hemlock to dwarf mistletoe infection. Plant, Cell and Environment 27, 937-946.

Mohebby B. 2005. Attenuated total reflection infrared spectroscopy of white-rot decayed beech wood. International Biodeterioration & Biodegradation 55, 247-251.


140

Newbanks D, Bosch A, Zimmermann MH. 1983. Evidence for xylem dysfunction by embolization in Dutch elm disease. Phytopathology, 73, 1060-1063.

Nagy NE, Franceschi VR, Solheim H, Krekling T, Christiansen E. 2000. Wound-induced traumatic resin duct development in stems of Norway spruce (Pinaceae): anatomy and cytochemical traits. American Journal of Botany 87, 302-313.

Oliveira H, Sousa A, Alves A, Nogueira AJA, Santos C. 2012. Inoculation with Ophiostoma novo-ulmi subsp. americana affects photosynthesis, nutrition and oxidative stress in in vitro Ulmus minor plants. Environmental and Experimental Botany 77, 146-155.

Osbourn A. 1996. Saponins and plant defence-a soap story. Trends in Plant Science 1, 4-9.

Ouellette G, Rioux D. 1992. Anatomical and physiological aspects of resistance to Dutch elm disease. In: Defense mechanisms of woody plants against fungi: Springer, pp. 257-307.

Ouellette G, Rioux D, Simard M, Cherif M. 2004. Ultrastructural and cytochemical studies of host and pathogens in some fungal wilt diseases: retro-and introspection towards a better understanding of DED. Forest Systems 13, 119-145.

Pandey KK, Pitman AJ. 2003. FTIR studies of the changes in wood chemistry following decay by brown-rot and white-rot fungi. International Biodeterioration & Biodegradation 52, 151-160.

Pavlovic S, Brandao PRG. 2003. Adsorption of starch, amylose, amylopectin and glucose monomer and their effect on the flotation of hematite and quartz. Miner Eng, 16, 1117-1122.

Peacock J, Cuthbert R, Lanier G. 1981. Deployment of Traps in a Barrier Strategy to Reduce Populations of the European Elm Bark Beetle, and the Incidence of Dutch Elm Disease. In: Mitchell E (ed) Management of Insect Pests with Semiochemicals Springer US, pp 155-174.

Pearce RB. 1996. Antimicrobial defences in the wood of living trees. New Phytologist 132, 203-233.

Pollard M, Beisson F, Li Y, Ohlrogge JB. 2008. Building lipid barriers: biosynthesis of cutin and suberin. Trends in Plant Science 13, 236-246.

Pouzoulet J, Pivovaroff AL, Santiago LS, Rolshausen PE. 2014. Can vessel dimension explain tolerance toward fungal vascular wilt diseases in woody plants? Lessons from Dutch elm disease and esca disease in grapevine. Front Plant Sci 5, 253.

Przybył K, Dahm H, Ciesielska A, Moliński K. 2006. Cellulolytic activity and virulence of Ophiostoma ulmi and O. novo-ulmi isolates. Forest Pathology 36, 58-67.


141

Richards W. 1993. Cerato-ulmin: A Unique Wilt Toxin of Instrumental Significance in the Development of Dutch Elm Disease. In: Sticklen M, Sherald J (eds) Dutch Elm Disease Research Springer New York, pp 89-151.

Rioux D, Ouellette GB. 1991. Barrier zone formation in host and nonhost trees inoculated with Ophiostoma ulmi. I. Anatomy and histochemistry. Canadian Journal of Botany 69, 2055-2073.

Rodríguez-Calcerrada J, Shahin O, del Carmen del Rey M, Rambal S. 2011. Opposite changes in leaf dark respiration and soluble sugars with drought in two Mediterranean oaks. Functional Plant Biology 38, 1004.

Santos C, Fragoeiro S, Phillips A. 2005. Physiological response of grapevine cultivars and a rootstock to infection with Phaeoacremonium and Phaeomoniella isolates: an in vitro approach using plants and calluses. Scientia Horticulturae 103, 187-198.

Sene CF, Mccann MC, Wilson RH, Grinter R. 1994. Fourier-transform Raman and Fourier-transform infrared spectroscopy (an investigation of five higher plant cell walls and their components). Plant Physiology, 106, 1623-1631.

Shigo A, Tippett JT. 1981. Compartmentalization of American Elm Tissues Infected by Ceratocystis-Ulmi. Plant Disease 65, 715-718.

Shigo AL. 1984. Compartmentalization: a conceptual framework for understanding how trees grow and defend themselves. Annual Review of Phytopathology 22, 189-214.

Smith KT. 2015. Compartmentalization, Resource Allocation, and Wood Quality. Current Forestry Reports 1, 8-15.

Solla A, Gil L. 2002a. Influence of water stress on Dutch elm disease symptoms in Ulmus minor. Canadian Journal of Botany 80, 810-817.

Solla A, Gil L. 2002b. Xylem vessel diameter as a factor in resistance of Ulmus minor to Ophiostoma novo-ulmi. Forest Pathology 32, 123-134.

Solla A, Martín J, Ouellette G, Gil L. 2005. Influence of plant age on symptom development in Ulmus minor following inoculation by Ophiostoma novo-ulmi. Plant Disease 89, 1035-1040.

Steele CL, Crock J, Bohlmann J, Croteau R. 1998. Sesquiterpene synthases from grand fir (Abies grandis). Comparison of constitutive and wound-induced activities, and cDNA isolation, characterization, and bacterial expression of delta-selinene synthase and gamma-humulene synthase. Journal of Biological Chemistry 273, 2078-2089.

Stuart BH. 2004. Infrared Spectroscopy: Fundamentals and Applications. Wiley,

Chichester, UK


142

Takai S. 1974. Pathogenicity and cerato-ulmin production in Ceratocystis ulmi. Nature 252, 124-126.

Temple B, Horgen PA, Bernier L, Hintz WE. 1997. Cerato-ulmin, a hydrophobin secreted by the causal agents of Dutch elm disease, is a parasitic fitness factor. Fungal Genetics and Biology 22, 39-53.

Tetley RM, Thimann KV. 1974. The metabolism of oat leaves during senescence I. Respiration, carbohydrate metabolism, and the action of cytokinins. Plant Physiology 54, 294-303.

Thomas FM, Blank R, Hartmann G. 2002: Abiotic and biotic factors and their interactions as causes of oak decline in Central Europe. Forest Pathology 32, 277-307.

Tumlinson J, Engelberth J. 2008. Fatty Acid-Derived Signals that Induce or Regulate Plant Defenses Against Herbivory. In: Schaller A (ed) Induced Plant Resistance to Herbivory Springer Netherlands, pp 389-407.

Upchurch RG. 2008. Fatty acid unsaturation, mobilization, and regulation in the response of plants to stress. Biotechnology Letters 30, 967-977.

Urban J, Dvořák M. 2014. Sap flow-based quantitative indication of progression of Dutch elm disease after inoculation with Ophiostoma novo-ulmi. Trees, 28, 1599-1605.

Vanetten HD, Mansfield JW, Bailey JA, Farmer EE. 1994. Two classes of plant antibiotics: Phytoalexins versus" Phytoanticipins". The Plant Cell 6, 1191.

Vek V, Oven P, Humar M. 2013. Phenolic extractives of wound-associated wood of beech and their fungicidal effect. International Biodeterioration & Biodegradation 77, 91-97.

Venturas M, López R, Gascó A, Gil L. 2013. Hydraulic properties of European elms: Xylem safety-efficiency tradeoff and species distribution in the Iberian Peninsula. Trees, 27, 1691-1701.

Venturas M, López R, Martín J, Gascó A, Gil L. 2014. Heritability of Ulmus minor resistance to Dutch elm disease and its relationship to vessel size, but not to xylem vulnerability to drought. Plant Pathology 63, 500-509.

Webber J. 2004. Experimental studies on factors influencing the transmission of Dutch elm disease. Forest Systems 13, 197-205.

Weber H. 2002. Fatty acid-derived signals in plants. Trends in Plant Science 7, 217-224.


143

Witzell J, Martín JA. 2008. Phenolic metabolites in the resistance of northern forest trees to pathogens — past experiences and future prospects. Canadian Journal of Forest Research 38, 2711-2727.

Wullschleger SD, Mclaughlin SB, Ayres MP. 2004: High-resolution analysis of stem increment and sap flow for loblolly pine trees attacked by southern pine beetle. Canadian Journal of Forest Research 34, 2387-2393.

Zagrobelny M, Bak S, Møller BL. 2008. Cyanogenesis in plants and arthropods. Phytochemistry 69, 1457-1468.

145

6. GENERAL DISCUSSION

General discussion

147

6. GENERAL DISCUSSION

The overarching goal of this PhD thesis is to better understand the resistance of

U. minor to factors of abiotic and biotic stress that are and will continue putting at risk

the future of this species. DED is the main factor that has decimated most elm

populations in Europe and that has received the largest attention in research programs

in the past. Active efforts have been made in Europe – particularly Spain, Italy, United

Kingdom and The Netherlands – towards finding DED-resistant elm clones and

understanding the biological bases of such resistance. One substantial reward to these

efforts has been the finding of several U. minor clones highly resistant to DED. In Spain,

seven resistant clones have been included in the National Catalogue of Plant Basic

Materials for Production of Qualified Forest Reproductive Materials in 2014 by the

Spanish Ministry of Agriculture, Food and Environment (BOE-A-2014-1353). This is a

good starting point towards propagating elms at a large scale and recover the

importance that this species had not many years ago all throughout Europe. But it is

essential that research efforts continue to fully understand the genetic and functional

bases of elm resistance to DED, as it is not yet clear why only few genotypes have

proved resistant to the disease. Phenological, chemical and wood anatomical traits

play a role in resistance (Martín et al. 2008; Martín et al. 2010; Solla and Gil 2002),

which, however, is apparently a complex response integrating different levels of the

plant’s metabolism. Here, we examined to what extent susceptible and resistant clones

to DED of U. minor differed in physiological and chemical traits before and after

inoculation with O. novo-ulmi. We concluded that resistant clones had a more

defensive chemical profile than susceptible clones – a difference apparently enhanced

by wounding – which could have restricted fungal spreading after inoculation and so

helped to maintain water transport and leaf gas exchange similar to control trees not

inoculated with the pathogen.

Future studies need to combine measurements at different organizational levels

to eventually identify genes of DED resistance and molecular markers that may

accelerate the detection of elm genotypes resistant to DED. Perdiguero et al. (In

preparation) have studied gene expression levels in clones of contrasted susceptibility

to DED inoculated with O. novo-ulmi or water – analogously as to our experiment in

Chapter 5. They have found that a consistent group of gen transcripts significantly

changed in response to O. novo-ulmi exclusively in resistant genotypes; 157 genes in

AB-AM2.4, 88 genes in MDV2.3 and 18 genes in both genotypes. Notable increases in

level of transcripts of LRR (Leucine rich repeat), MYB (myeloblastosis) and BHLH


148

(basic helix-loop-helix) transcription were identified in response to O. novo-ulmi. The

results suggest that U. minor tolerance to O. novo-ulmi is related to the expression of

these differential genes.

The future of U. minor conservancy requires also evaluating species tolerance

to other potential factors of stress, such as drought and flooding. These factors may

limit the survival of the few elms that are still alive and the success of large-scale

plantations with DED-resistant clones projected for the future. One surprising remark

we noticed at the beginning of this PhD thesis was the scarceness of basic information

on physiology of U. minor in relation to drought or flooding. It is worth citing the studies

of Solla (Solla and Gil 2002) and Venturas (Venturas et al. 2014) in recent years in

relation to drought, while no study before ours (see Chapter 3) had addressed U. minor

responses to a simulated or real flooding events. The response of U. minor to drought

and flood had some aspects in common. Indeed, both factors impose a water stress on

plants, which respond in both cases by closing the stomata to preventing any further

decline in leaf water potential at the cost of net photosynthetic CO2 uptake. In fact, the

response to the inoculation with O. novo-ulmi caused a similar response in the leaves

of susceptible clones VA-AP38 and M-DV1. Xylem cavitation caused by the spreading

of the fungus resulted in a decline of xylem water transport to the leaves, and thus

stomatal conductance and net photosynthetic CO2 uptake, as it happened after plants

were exposed to drought and flood for some weeks. Another common response in all

cases was the shedding of leaves. Stomata closure was not enough to cope with

increasing water stress provoked by either spreading of O. novo-ulmi, water shortage,

or water excess and leaves started to fall to prevent runaway embolism, which, had it

occurred, would have demanded the construction of new wood or shoots. Xylem water

potential at which leaves fell in the drought experiment was -5.2 MPa (Chapter 4); this

corresponded to a value of midday leaf water potential of -4.3 MPa, which was similar

to values of midday leaf water potential at which leaves were observed to fall in the

other experiments (Chapter 3 and 5).

One remarkable observation in control trees of all three experiments not

subjected to any stress treatment was the relatively high values of loss of conductivity.

Trees exposed to well-watered conditions and no fungal inoculation reached in some

cases losses of conductivity as high as 45% of maximal capacity. And the same was

true for plants of U. laevis monitored together with those of U. minor in chapter 3, in

agreement with a previous study by Venturas et al. (2013) reporting that xylem

vulnerability to cavitation is relatively high in both species. Somehow, high diurnal

General discussion

149

peaks in transpiration and net CO2 uptake that render these trees highly competitive

occur at the risk of hydraulic failure. And this would be consistent with a rapid decline in

transpiring surface upon reductions in water transport caused by abiotic and biotic

stresses.

To the best of our knowledge mass-specific rates of respiration in leaves and

woody organs of U. minor had not been previously measured. They varied from 0.6 to

27.8 nmol g-1s-1 in leaves, 0.5 to 2.7 nmol g-1s-1 in stems and 1.5 to 3.5 nmol g-1s-1 in

roots. While all three stress agents tested here (O. novo-ulmi, drought and flood)

invariably limited net photosynthetic CO2 uptake, respiration exhibited a more variable

response; it progressively declined by an effect of water-deficit stress (Chapter 4),

increased rapidly by an effect of water-excess stress to remain constant later on

(Chapter 3), and was little affected by the inoculation with O. novo-ulmi (Chapter 5).

These responses are mostly consistent with literature and agree with the idea that

stress-induced variations in respiration rates respond to variations in the demand of

respiratory products (Atkin and Macherel 2009). Drought stops growth and phloem

upload and gear down many other metabolic processes and, hence, ATP turnover

does not proceed at enough speed as to supply mitochondria with the same ADP as in

normal, non-stress conditions, and respiration rates in leaves and stems decline. On

the contrary, flooding seems to demand an increase in energy to cope with alterations

in soil aeration – for example the hypertrophy of lenticels at the root collar – so that

respiration rates of all organs increase soon after the change in soil conditions. Finally,

inoculation with O. novo-ulmi did barely affect leaf respiration, despite the clear water

stress the pathogen caused in leaves of DED-susceptible U. minor clones, suggesting

some counter effects of pathogen infection per se and water stress on final rates of

respiration.

The distribution of U. minor and common knowledge based on empirical

evidences and observations of survival relative to other species led us to hypothesize

that U. minor was moderately resistant to drought and flood. The results supported this

hypothesis. First, moderate resistance to drought relied on the capacity of seedlings to

reduce transpiration under soil water shortage by means of stomata closure and leaf

shedding. This made possible that 2-year-old seedlings were able to survive in 16-l

pots without watering during summer for approximately 90 days – time at which 50%

mortality was reached. Second, although xylem vulnerability was high, particularly

relative to drought-resistant Q. ilex, it was lower than that of other riparian species. The

plotting of midday xylem water potential vs percentage loss of stem- and root-specific


150

hydraulic conductivity (PLC) resulted in values of P50 (xylem tension causing 50%

decline in PLC) of -3.4 and -2.6 MPa, respectively. Values of P50 reported for U. laevis,

U. glabra, Populus trichocarpa×deltoides, Salix spp. were -0.4 MPa, -0.5 MPa, -1.7

Mpa, -1.8 MPa and -1.4 MPa, respectively (Cochard et al. 2007; Melcher and

Zwieniecki 2013; Plavcová and Hacke 2012; Venturas et al. 2013). Xylem resistance to

cavitation in U. minor also contributed to postpone death by drought, which took place

below -5 MPa midday xylem water potential, and above 80% root PLC. Third, drought-

induced attenuation in stem growth was high in U. minor, which likely contributed to

maintain carbon reserves relatively high even at severe drought conditions. Similarly,

resistance to flood relied on the reduction in transpiration when root water uptake was

limited by soil hypoxia. This response prevented leaf water potential to fall and kept

root PLC below lethal values (i.e. around 60% after 60 days of pot immersion,

assuming lethal PLC is > 80% either the origin of water stress is shortage or excess in

the soil). However, stomatal limitations to net photosynthetic CO2 uptake could affect

NSC reserves (not measured in the flooding experiment) and contribute to phloem

degradation of the stem at the water line, which was the factor that ultimately killed

plants of U. minor.

Comparatively with other riparian tree species, it is the moderate flood and

drought resistance which confer U. minor the ability to grow from near the water

channel to areas far from the channel, but with some influence of the water table.

Faster growing species making a prolific use of water resources (e.g. U. laevis) can

outcompete U. minor in the close vicinity of the water channel, but not far from the

channel where the water table suffers strong seasonal variations, and a moderate

degree of drought tolerance is needed to survive the summer.

Here we have focused on the single effects of DED, drought and flood on U.

minor. Future experiments should also investigate plant responses to interacting

factors, for example to subsequent cycles of flood (during spring) and drought (during

summer), or to inoculation with O. novo-ulmi in drought- and/or flood-stressed plants.

During the course of this PhD thesis, we have started a new experiment in which plants

of U. minor, U. laevis, and U. glabra – the three elm species native to the Iberian

Peninsula – have been subjected to repeated drought cycles during the summers of

2014 and 2015; these plants are planned to be inoculated with O. novo-ulmi in the

spring of 2016. The objectives are to compare drought acclimation responses among

species and assess how acclimation to drought affects subsequent propagation of O.

novo-ulmi. It is possible that reduced size of vessels induced by drought favor the

General discussion

151

capacity of trees to control fungal propagation (Solla and Gil 2002), but also that

drought reduces NSC reserves and the response of the secondary metabolism to cope

with the pathogen (McDowell et al. 2008). Moreover, it is also possible that species-

specific tolerance to drought results in distinct effects of drought on tree anatomy and

chemical profile, and thus DED progression upon O. novo-ulmi inoculation among the

three species.

Future experiments should also aim at comparing resistance to drought, alone

or in combination with other stress factors, among the catalogued Spanish DED-

resistant genotypes, in order to identify more appropriate genotypes for and increase

the effectiveness and efficiency of future plantations. Extreme climatic events are

expected to be more frequent as a consequence of Global Change (Allen and

Breshears 1998); for example, droughts will be more frequent and intense in the west

Mediterranean basin by the middle of this century (Lindner et al. 2010). This makes

even more important that large-scale plantations consider within-species drought

resistance variability. Within-species monitoring of stress resistance together with the

identification and eventual propagation of new genotypes resistant to DED will provide

a keystone in U. minor conservation. In the long run, the establishment of some few

planted genotypes resistant to both DED and other stresses will translate in the sexual

reproduction among them in natural conditions and the growing expansion of many

more genotypes to areas previously occupied by this species.


152

Reference

Allen CD, Breshears DD. 1998. Drought-induced shift of a forest–woodland ecotone: rapid landscape response to climate variation. Proceedings of the National Academy of Sciences 95, 14839-14842.

Atkin OK, Macherel D. 2009. The crucial role of plant mitochondria in orchestrating drought tolerance. Ann Bot 103, 581-597.

Cochard H, Casella E, Mencuccini M. 2007. Xylem vulnerability to cavitation varies among poplar and willow clones and correlates with yield. Tree Physiology 27, 1761-1767.

Lindner M, Maroschek M, Netherer S, Kremer A, Barbati A, Garcia-Gonzalo J, Seidl R, Delzon S, Corona P, Kolström M, Lexer MJ, Marchetti M. 2010. Climate change impacts, adaptive capacity, and vulnerability of European forest ecosystems. Forest Ecology and Management 259, 698-709.

Martín JA, Solla A, Domingues MR, Coimbra MA, Gil L. 2008. Exogenous phenol increase resistance of Ulmus minor to Dutch elm disease through formation of suberin-like compounds on xylem tissues. Environmental and Experimental Botany 64, 97-104.

Martín JA, Solla A, Gil L, García-Vallejo MC. 2010. Phenological and histochemical changes of Ulmus minor due to root absorption of phenol: Implications for resistance to DED. Environmental and Experimental Botany 69, 175-182.

Melcher PJ, Zwieniecki MA. 2013. Functional analysis of embolism induced by air injection in Acer rubrum and Salix nigra. Front Plant Sci 4,

Plavcová L, Hacke UG. 2012. Phenotypic and developmental plasticity of xylem in hybrid poplar saplings subjected to experimental drought, nitrogen fertilization, and shading. Journal of Experimental Botany

Solla A, Gil L. 2002. Influence of water stress on Dutch elm disease symptoms in Ulmus minor. Canadian Journal of Botany 80, 810-817.


Venturas M, López R, Martín J, Gascó A, Gil L. 2014. Heritability of Ulmus minor resistance to Dutch elm disease and its relationship to vessel size, but not to xylem vulnerability to drought. Plant Pathology 63, 500-509.

154

7. CONCLUSIONS

Conclusions

156

7. CONCLUSIONS

1. Seedlings of U. minor are moderately tolerant to flooding, although less tolerant

than those of U. laevis, in agreement with the distribution of both species in riverside

forests.

2. The inability to maintain a positive carbon balance somehow compromises

seedling survival under flood conditions, earlier in U. minor than U. laevis.

3. Bark degradation at water line ultimately causes flood-induced dieback.

4. Seedlings of U. minor are moderately tolerant to drought: stomata closure and

leaf shedding contributes to reduce water loss and xylem tension, whereas water

transport is maintained at relatively high xylem tension, compared to other riparian tree

species.

5. Differences in drought resistance between U. minor than Q. ilex rely on the ability

to postpone xylem tension and maintain water transport under high xylem tension,

rather than tolerating severe losses of hydraulic conductivity.

6. Hydraulic failure is a primary factor in drought-induced shoot dieback of U. minor

and Q. ilex; there is no evidence of carbon starvation at this time, but it might follow

shoot dieback and impede root resprouting.

7. The impact of drought on NSC concentrations depends more on regulation of

plant respiration than stomatal behaviour under water stress.

8. DED-resistant genotypes of U. minor possess a biochemical profile more

consistent with a higher capacity to cope with O. novo-ulmi than more susceptible

genotypes.

9. Wounding could enhance the capacity of DED-resistant genotypes to cope with O.

novo-ulmi.

10. Reduced hydraulic conductivity and photosynthetic CO2 uptake caused by O.

novo-ulmi feeds forward on susceptibility to O. novo-ulmi.

11. Young plants of U. minor exhibit similar functional responses to the stresses of

flooding, drought and O. novo-ulmi infection.

functional response of ulmus minor mill. to drought ...oa.upm.es/38556/1/meng_li.pdffunctional...

Documents