Molecular and Physiological Characterization of Grapevine Rootstock Adaptation to Drought

Université Bordeaux Segalen

Année 2011 Thèse n°1864

THÈSE

pour le

DOCTORAT DE L’UNIVERSITÉ BORDEAUX 2

Mention: Sciences, Technologie, Santé

Option: Biologie végétale

Présentée et soutenue publiquement

Le 19 Décembre 2011

Par Anthony PECCOUX

Né le 26 Mars 1980 à Annecy (Haute-Savoie, France)

Molecular and physiological characterization of grapevine rootstock adaptation to drought

Membres du Jury

Prof. Dr. C. LOVISOLO ...... University of Torino...... Président du jury Prof. Dr. G.R. CRAMER ...... University of Nevada...... Rapporteur Dr. T. SIMONNEAU ...... INRA Montpellier ...... Rapporteur Dr. M.-B. BOGEAT-TRIBOULOT. .INRA Nancy...... Examinateur Prof. Dr. S. DELROT...... University of Bordeaux...... Examinateur Dr. N. OLLAT...... INRA Bordeaux...... Co-directrice de thèse Prof. Dr. H.R. SCHULTZ ...... Geisenheim Research Center...... Co-directeur de thèse Université Bordeaux Segalen

Année 2011 Thèse n°1864

THÈSE

pour le

DOCTORAT DE L’UNIVERSITÉ BORDEAUX 2

Mention: Sciences, Technologie, Santé

Option: Biologie végétale

Présentée et soutenue publiquement

Le 19 Décembre 2011

Par Anthony PECCOUX

Né le 26 Mars 1980 à Annecy (Haute-Savoie, France)

Molecular and physiological characterization of grapevine rootstock adaptation to drought

Membres du Jury

“Les espèces qui survivent ne sont pas les espèces les plus fortes, ni les plus intelligentes, mais celles qui s’adaptent le mieux aux changements”

Charles DARWIN (1809-1882) Pour mes parents et ma famille Acknowledgements

We acknowledge the generous financial support of the "Studienstrukturprogramm, Doktorandenkolleg Hochschule RheinMain und Forschungsanstalt Geisenheim" of the State of Hesse, Germany. This project is also founded by grants from the Aquitaine region and CIVB, “Conseil Interprofessionnel des Vins de Bordeaux”, both are deeply acknowledged for their kind support.

I would also like to thank all the members of my PhD committee for reviewing this work and their critical insights and valuable suggestions.

This project was a collaboration between two laboratories in France and Germany. I am very grateful to my two supervisors for their guidance, support and advice during my PhD studies. I wish to express my deep gratitude to all the colleagues in these labs for their technical assistance, their suggestions, and critical reading and improvement of the manuscript. The many scientific interactions, both sharing and combining knowledge with colleagues, have contributed to the different approaches in this thesis, providing me a valuable and broad scientific background. These interactions were crucial to the advancement of my research and my progress as a scientist. I therefore wish to give special thanks to these collaborators. Hopefully we will have many more meaningful exchanges and collaborations in the future.

During this thesis, I met many amazing friends and I would like to thank all of them. We shared many scientific exchanges, and their friendship has made the daily life of my PhD more enjoyable and fulfilling.

Last but not least, I wish to show my tremendous gratitude towards my parents and my family for their daily support and for making me who I am. Abstract Climate change raises concerns about temporal and spatial water availability in many grape growing countries. The rapidly increasing world population and the scarcity of suitable land for agricultural food production, together with a changing climate, will increase competition with grape-producing areas for the use of land and resources. Consequently, other practices that can potentially improve water management of vineyards and water acquisition by grapevines need to be considered. Aside from canopy systems and their management, the choice of plant material is a key issue. Therefore, in the present work, the role of different rootstocks, regarding their tolerance to drought, was investigated for their potential effects on i) water uptake, ii ) water transport and iii ) shoot water use, using a combination of ecophysiological, modelling and transcriptomic approaches. Experiments were conducted under controlled conditions to decipher short and long term responses to drought of different rootstocks grafted with the same scion. An ecophysiological model was used to investigate the roles of rootstock genotypes in the control of stomatal aperture. Long-term steady state water-deficit conditions were used to examine the responses of i) whole plant growth, root anatomy and hydraulic properties and ii ) transcriptome remodelling in the roots. Our model showed that rootstock affect stomatal aperture of the grafted scion via coordinated processes between root traits, hydraulic signals and chemical signals. Stomatal conductance, transpiration rate and leaf-specific hydraulic conductance were higher and better maintained under well-watered and moderate water-deficit conditions in the drought-tolerant genotype (110 Richter) compared to the drought-sensitive one ( Vitis riparia cv. Gloire de Montpellier). We identified several genotype-specific parameters which play important roles, like root-related parameters, in the control of stomatal regulation. Additionally, root system architecture and root hydraulic properties are important constitutive traits identified between rootstocks. Long-term water-deficit induced genotype adaptive responses in the roots were evaluated. The drought-tolerant genotype exhibited a substantial shift in root tips xylem conduit diameter under moderate water-deficit while the drought-sensitive genotype did not respond. Transcriptomic analysis identified genotype-specific transcripts that are regulated by water- deficit levels. The comparison between stress levels and genotypes identified 24 significant genes in “treatment×genotype” interactions, most of them were involved in lipid metabolism and cell wall processes. These genes displayed genotype-specific water-deficit response curves. Protection against drought-induced oxidative damage was found to be an important mechanisms induced by the drought-tolerant rootstock, while the drought-sensitive one responds to water-deficit by modification of cell wall properties.

Keywords : grapevine, rootstock, Vitis spp. , roots, water stress, drought, transcriptomic, modelling, gas exchange, hydraulic signalling, chemical signalling

Address of the laboratory in France

UMR 1287 Ecophysiologie et Génomique Fonctionnelle de la Vigne (EGFV) Institut National de la Recherche Agronomique (INRA), UB2, UB1, ENITA Institut des Sciences de la Vigne et du Vin (ISVV) 210, Chemin de Leysotte F-33882 Villenave d’Ornon (France)

Address of the laboratory in Germany

Forschungsanstalt Geisenheim Fachgebiet Weinbau Von Lade Strasse 1 D-65366 Geisenheim (Germany)

Table of contents

List of abbreviations and frequently used symbols ...... I List of figures ...... II List of tables...... VII Chapter I. General introduction ...... - 8 - I.1 Context of the present studies...... - 13 - I.1.1 Global warming: new challenges for facing the future ...... - 13 - I.1.1.1 General considerations ...... - 13 - I.1.1.2 What about grapevine?...... - 14 - I.2 Objectives and outlines of the thesis...... - 15 - I.2.1 Objectives of the thesis ...... - 16 - I.2.2 Outline of the thesis...... - 17 - I.3 Paper publishing and conference communication ...... - 18 - I.3.1 Posters ...... - 18 - I.3.2 Oral communications ...... - 18 - I.3.3 Articles in peer-reviewed plant journals...... - 18 - Chapter II. The function of rootstocks in the adaptation of grapevine to drought: A review .- 19 - II.1 Root system architecture and rootstocks responses to drought ...... - 23 - II.1.1 Root system architecture...... - 24 - II.1.1.1 Methods for studying plant root systems ...... - 24 - II.1.1.2 Parameters affecting grapevine root architecture ...... - 25 - II.1.1.3 Developmental plasticity of roots under drought ...... - 26 - II.1.1.3.1 General relationships between root growth and water status ...... - 27 - II.1.1.3.2 Grapevine rootstocks plasticity to water status...... - 28 - II.1.1.3.3 Interactions with symbiotic organisms ...... - 30 - II.1.2 Drought-related physiological mechanisms in roots...... - 30 - II.1.2.1 Water-deficit perception ...... - 30 - II.1.2.2 Roles of plant hormones ...... - 30 - II.1.2.2.1 Some generalities...... - 30 - II.1.2.2.2 ABA sustain the root growth maintenance ...... - 31 - II.1.2.3 Accumulation of compatible solutes and osmotic adjustments...... - 32 - II.1.2.3.1 Sugars and polyalcohols ...... - 33 - II.1.2.3.2 Amino acids: glutamate and proline ...... - 33 - II.1.2.3.3 Amines: polyamines and glycine betaine...... - 35 - II.1.2.4 Morphological and cell wall modifications...... - 36 - II.1.2.5 Protection from oxidative damage ...... - 37 - II.1.2.6 Hydrotropism...... - 37 - II.1.2.7 Molecular pathways...... - 38 - II.1.2.7.1 Signal perception ...... - 40 - II.1.2.7.2 Signal transduction ...... - 40 - II.1.2.7.2.1 ABA biosynthesis and signal transduction ...... - 40 - II.1.2.7.2.2 MAPKinase pathways ...... - 42 - II.1.2.7.2.3 SNF-1-like kinases ...... - 42 - II.1.2.7.2.4 Phosphatases...... - 43 - II.1.2.7.2.5 Phospholipid signalling ...... - 43 - II.1.2.7.2.6 Calcium signalling...... - 44 - II.1.2.7.3 Transcriptional regulation of gene expression ...... - 44 - II.1.2.7.3.1 ABA-independent pathway ...... - 44 - II.1.2.7.3.2 ABA-dependent pathway ...... - 44 - II.1.2.7.4 Functional proteins: a role in stress-protection and adaptation ...... - 45 - II.1.2.7.5 Drought-induced transcriptomic and proteomic changes in roots...... - 45 - II.2 Root-to-shoot water transport, drought and grape rootstocks...... - 47 - II.2.1 Water transport: some generalities...... - 48 - II.2.1.1 Radial transport...... - 48 - II.2.1.1.1 The three pathways...... - 48 - II.2.1.1.2 Aquaporins and radial water transport ...... - 49 - II.2.1.2 Axial transport ...... - 50 - II.2.2 Regulation of water transport under water-deficit stress...... - 50 - II.2.2.1 Hydraulic architecture...... - 50 - II.2.2.1.1 Anatomical structure...... - 51 - II.2.2.1.2 Hydraulic conductivity and embolism induced-cavitation...... - 52 - II.3 Shoot growth and water use under water-deficit stress...... - 55 - II.3.1 Shoot growth and development...... - 55 - II.3.2 Regulation of leaf water potential and gas exchange...... - 56 - II.3.3 Root-to-shoot signalling...... - 57 - II.3.4 Water use efficiency ...... - 58 - II.4 Crop improvement under drought ...... - 59 - Chapter III. Control of stomata by rootstock-sourced signals under water stress: a model-based analysis in grapevine ...... - 61 - III.1 Introduction ...... - 62 - III.2 Materials and methods...... - 64 - III.2.1 Plant material and growth conditions ...... - 64 - III.2.2 Experimental setups...... - 65 - III.2.2.1 Experiment 1: drought cycle experiment...... - 65 - III.2.2.2 Experiment 2: steady-state drought experiment...... - 65 - III.2.3 Determination of leaf area and leaf plastochron index ...... - 66 - III.2.4 Estimation of root characteristics ...... - 66 - III.2.5 Leaf gas exchange measurements...... - 67 - III.2.6 Plant water relation measurements ...... - 67 - III.2.7 Xylem sap collection and abscisic acid (ABA) analysis...... - 68 - III.2.8 Model description and data analyses ...... - 68 - III.3 Results...... - 69 - III.3.1 Plant water status, gas exchange and ABA responses to gradual water-deficit...... - 69 - III.3.2 Physiological responses to gradual water-deficit at the single leaf-scale ...... - 71 - III.3.3 Scaling up from single leaf to canopy level...... - 73 - III.3.4 Model validation in long term water-deficit conditions...... - 75 - III.3.5 Contribution of genetic parameters to stomatal regulation...... - 76 - III.4 Discussion ...... - 77 - III.4.1 Rootstock-specific stomatal control under fluctuating water-deficit in grafted-scion ...... - 77 - III.4.2 Contribution of genetic parameters and water stress duration for stomatal regulation...... - 79 - III.4.3 Modelling to improve genetic selection of grape rootstocks ...... - 80 - III.5 Supporting informations...... - 82 - Chapter IV. Xylem development and hydraulic conductivity are differentially affected in roots of drought-sensitive and drought-tolerant grape rootstocks...... - 87 - IV.1 Introduction...... - 88 - IV.2 Materials and methods ...... - 89 - IV.2.1 Plant material and growth conditions ...... - 89 - IV.2.2 Experimental setup...... - 90 - IV.2.3 Plant and soil water relation measurement ...... - 90 - IV.2.4 Growth analysis: leaf area, biomass partitioning and growth rate...... - 91 - IV.2.4.1 Shoot growth and leaf area measurements...... - 91 - IV.2.4.2 Biomass partitioning and relative growth rate...... - 91 - IV.2.5 Root xylem anatomy ...... - 91 - IV.2.5.1 Sample collection ...... - 91 - IV.2.5.2 Sectioning and staining...... - 92 - IV.2.6 Root hydraulic conductivity ...... - 92 - IV.2.6.1 Theoretical hydraulic conductivity: the Hagen-Poiseuille equation ...... - 92 - IV.2.6.2 Measured hydraulic conductivity ...... - 93 - IV.2.7 Data analyses and statistics ...... - 93 - IV.3 Results...... - 93 - IV.3.1 Effects of soil water content on plant water status ...... - 93 - IV.3.2 Water-deficit effects on growth and biomass allocation ...... - 94 - IV.3.3 Root hydraulic properties ...... - 96 - IV.3.3.1 Drought-induced Shifts in xylem development ...... - 96 - IV.3.3.2 Root conductivity is genotype-dependent rather than stress-responsive...... - 99 - IV.4 Discussion ...... - 100 - IV.4.1 Plant water-status, growth and biomass ...... - 100 - IV.4.2 Water deficit effects on xylem conduits size and hydraulic conductivity ...... - 101 - IV.5 Supporting informations ...... - 103 - Chapter V. Long term steady-state drought-induced transcriptome changes in roots of drought- sensitive and drought-tolerant grape rootstocks ...... - 105 - V.1 Introduction ...... - 106 - V.2 Materials and Methods ...... - 107 - V.2.1 Genetic materials and growth conditions...... - 107 - V.2.2 Experimental setups and soil water relations...... - 108 - V.2.3 RNA extraction and microarray hybridization...... - 108 - V.2.4 Microarrays data analysis ...... - 109 - V.2.5 Transcripts abundance validation by qPCR...... - 110 - V.3 Results ...... - 111 - V.3.1 Differentially expressed genes increases with water-deficit levels...... - 112 - V.3.2 Number of significant transcripts across years, treatments and genotypes...... - 112 - V.3.3 Array performance validation...... - 114 - V.3.4 Water-deficit induced genotype-specific gene dose-responses...... - 115 - V.3.5 Gene ontology of water-deficit responsive genes...... - 118 - V.4 Discussion...... - 120 - V.4.1 Steady-state water deficits induce transcriptome rearrangements ...... - 120 - V.4.2 Molecular physiology of drought-responsive genes involved in T×G interactions ...... - 121 - V.4.2.1 Drought may induce lipid peroxidation in the roots of CS and 110R ...... - 121 - V.4.2.2 Water-deficit affects cell wall properties in the roots of RG...... - 123 - V.5 Supporting informations...... - 125 - Chapter VI. General conclusions and perspectives...... - 127 - VI.1 General conclusions ...... - 128 - VI.2 Perspectives ...... - 130 - References ...... - 132 -

List of abbreviations and frequently used symbols

A: Net CO 2 assimilation ABA : Abscisic acid ABA-GE : ABA-glucose ester ANOVA : Analysis of variance CS : Vitis vinifera cv. Cabernet Sauvignon cv. : Cultivar DNA : Deoxyribonucleic acid DPA : Dihydrophaseic acid E: Transpiration rate ESTs : Expressed Sequence Tags FACE : Free Air CO 2 Enrichment FDR : False discovery rate GO : Gene ontology gs: Stomatal conductance HWD : High water deficit Kh theo : Theoretical hydraulic conductivity Kh r: Root segment hydraulic conductivity Kleaf : leaf-specific hydraulic conductance LA : Leaf area LPI : Leaf plastochron index LWD : Low water deficit Mha : millions of hectares Mhl : millions of hectolitres MGt : Millardet et de Grasset MPSS : Massively Parallel Signature Sequencing Mqx : Millions of quintals (100 kg) MWD : Moderate water deficit PA : Phaseic acid leaf : Leaf water potential PD : Predawn leaf water potential PPFD : Photosynthetic photon flux density ppm : Parts per million qRT-PCR or qPCR : quantitative Real Time Polymerase Chain Reaction SAGE : Serial Analysis of Gene Expression SSR : Single sequence repeats (also referred to as microsatellites) RG : Vitis riparia cv. Gloire de Montpellier RGR : Relative growth rate RNA : Ribonucleic acid RNA-Seq : RNA sequencing or deep sequencing SWC : Soil water content VPD : Vapour pressure deficit WUE int : Intrinsic water use efficiency WUE inst : Instantaneous water use efficiency 110R : 110 Richter ( Vitis berlandieri ×Vitis rupestris )

I List of figures

Figure I.1: Drawing of grafted grapevine representing aboveground, belowground and the grafting point (here, the graft is represented by an Omega grafting)...... - 11 - Figure II.1: Schematic presentation of scion-rootstock interactions affecting whole grafted grapevine responses to water-deficit conditions (adapted from Chaves et al. , 2003). - 21 - Figure II.2: Displacement velocity as a function of distance from the root cap junction of primary roots of well-watered (WW) and water-stressed (WS) maize seedlings (cv. FR697). The inset shows the profile of displacement velocity for WW and WS soybean primary roots (cv. Magellan). Relative elongation rates (h%1 ) are obtained from the derivative of velocity with respect to position. Regions 1 to 3, as described in the text, are indicated. The green arrows indicate the comparisons between treatments within regions 1 and 2 and the comparison of WS region 2 with WW region 3 (from Yamaguchi & Sharp, 2010)...... - 28 - Figure II.3: Seasonal root production (± SE) for root systems of V. berlandieri ×V. rupestris cv. 1103P and V. riparia ×V. rupestris cv. 101-14MGt (season×root system interaction: P=0.002). Data represent total root length produced per cm2 of observational window over three month periods for the years, 2003-2005. Each season corresponded to the following months: Autumn, September-November (significance of difference between 1103P and 101-14Mgt, P=0.328); winter, December-February (P=0.009), spring, March- May (P=0.230), and summer, June-August (P=0.032). Significant differences are indicated by different letters (redraw from Bauerle et al. , 2008b)...... - 29 - Figure II.4: Schematic representation of predicted functions and interactions of water stress- responsive proteins/genes and associated metabolites in the cell wall and cytosol in region 1 (Figure II.2) of the elongation zone of the maize primary root. Major differences in response between regions 1 and 2 are indicated by shaded text. Non-italicized text and solid connecting arrows indicate responses that have been demonstrated in published studies; italicized text and dashed connecting arrows indicate responses that are hypothesized to occur but have not yet been demonstrated. Up-regulation under water stress is indicated by the short upward arrows; for ABA, proline and oxalate oxidase (OxO), the double arrows indicate that accumulation or activity in region 1 was greater than in region 2. *Flavonoid accumulation has not been determined in water-stressed (WS) maize roots; however, isoflavonoids were shown to accumulate in region 1 of the soybean primary root under water stress. The inset shows increased apoplastic reactive oxygen species (ROS) in the epidermis of region 1 in WS compared to well-watered

(WW) roots, as indicated by confocal microscopy of roots stained with H2 DCF (2 ಿ,7 ಿ- dichlorodihydrofluorescein, green fluorescence), a membrane-impermeable ROS indicator. The right-hand diagram illustrates a transverse view of the root surface and focal planes. The increase in apoplastic ROS in WS roots is consistent with the increases in abundance of superoxide dismutase (SOD) and oxalate oxidase, which contribute to H2 O 2 production, and peroxidases (POX), which can also contribute to ROS production • including the generation of hydroxyl radicals ( OH) from H2 O 2 in the presence of •% superoxide (O2 ) and/or reductant (e.g. NADH) (from Yamaguchi & Sharp, 2010). - 34 - Figure II.5: Transcriptional regulatory networks of cis -acting elements and transcription factors involved in drought-stress-responsive gene expression in Arabidopsis . Transcription factors controlling stress-inducible gene expression are shown in colored ellipses. Cis -acting elements involved in stress-responsive transcription are shown in boxes. Small filled circles reveal modification of transcription factors in response to stress signals for their activation, such as phosphorylation. Regulatory cascade of stress-

II ...... - 39 - Figure II.6: Model of the major ABA signalling pathway. PYR/PYL/RCAR, PP2C and SnRK2 form a signalling complex referred to as the ‘ABA signalosome’. (A) Under normal conditions, PP2C negatively regulates SnRK2 by direct interactions and dephosphorylation of multiple residues of SnRK2. Once abiotic stresses or developmental cues up-regulate endogenous ABA, PYR/PYL/RCAR binds ABA and interacts with PP2C to inhibit protein phosphatase activity. In turn, SnRK2 is released from PP2C-dependent regulation and activated to phosphorylate downstream factors, such as the AREB/ABF bZIP-type transcription factor or membrane proteins involving ion channels. (B) In contrast, the abi1-1-type mutated protein lacks PYR/PYL/RCAR binding, resulting in the constitutive inactivation of SnRK2, even in the presence of ABA, and strong insensitivity to ABA in the abi1-1 mutant (redraw from Umezawa et al. , 2010)...... - 41 - Figure II.7: Pathways for the movement of water and solutes in roots. The apoplastic path (a) refers to the flow around protoplasts. The symplastic component defines flow from cell to cell via plasmodesmata (b). On the third route (c), water and solutes have to cross cell membranes (two membranes per cell layer; transcellular path). The transcellular path is important for water, but is of minor importance for solutes. For water, pathways (b) and (c) cannot be separated experimentally to date. Therefore, they are summarized as a cell- to-cell path. It is usually assumed that, in roots, the Casparian bands in the exo- and endodermis completely interrupt apoplastic transport. Here, it is, however, indicated that there may be an apoplastic component of water flow across Casparian bands (from Steudle, 2000a)...... - 48 - Figure II.8: Conducting cells in the xylem. The xylem consists of two types of tracheary elements: tracheids and vessel elements (from Taiz & Zeiger, 2002)...... - 51 - Figure III.1: Soil and plant water relations during the drying phase (experiment 1). (a) Changes in SWC for the three rootstocks. Data are means (n = 10) ± SE. (b) Changes in the predawn leaf water potential ( PD ) for the three rootstocks. Data are means (n = 10) ± SE. There are statistically significant effects (p<0.001) of the days after treatment (DAT) but no interactions between DAT and rootstocks (a, b). (c) Non linear regression (see equation in Table S. 1: List of the equations which are used in the model. The details of variables, parameters, units and values are given in the Tables S2 and S3., R2 =0.9616; p<0.0001) between PD plotted as a function of SWC for the three rootstocks during the drying phase. Each point is an individual measurement...... - 70 - Figure III.2: Clustered heatmap of measured and calculated variables (n=10) during the drying phase for the three rootstocks. The values were normalized (observed value/mean value of well-watered plants), then log2 transformed to make the pixels square and subjected to hierarchical clustering. The dendogramm at the left of the matrix indicates the degree of similarity between variables (the larger is the distance in the dendogramm, the lower is the similarity). The columns of the matrix represent the day after treatment for each genotype (Cabernet Sauvignon, Vitis riparia and 110R) and the rows each variable ([ABA]: xylem sap abscisic acid concentration, PD : predawn leaf water potential, leaf : leaf water potential, WUEint : intrinsic water use efficiency, WUEinst : instantaneous water use efficiency, Ci: intracellular CO2 concentration, PA: xylem sap phaseic acid concentration, DPA: xylem sap dihydrophaseic acid concentration, ABA- GE: xylem sap ABA-glucose ester concentration, Kleaf : leaf-specific hydraulic

III 2 s ...- 71 - Figure III.3: SWC-response curves of leaf water potential (a), transpiration rate E (b), stomatal conductance gs (c), xylem sap abscisic acid concentration [ABA] (d), and relationship between gs and [ABA] (e) of measured (symbols) and estimated based on model runs (lines) traits for the three rootstocks during experiment 1. The measured values represent the mean ± SD of 5-10 samples. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock...... - 72 - Figure III.4: Daily measurements of photosynthetic photon flux density (PPFD, a,b) and vapour pressure deficit (VPD, c,d) measured during the drying cycle (experiment 1) at two different day (well-watered left panel and water-stressed right panel conditions). The lines represent the model simulation of leaf water potentials (e,f), root water potentials (g,h), transpiration rate E (symbols represent measured values [circle: Cabernet Sauvignon, triangle: Vitis riparia , square: 110R], mean (n=5) ± SE, i,j), stomatal conductance gs (k,i) and xylem sap abscisic acid concentration [ABA] (m,n) during the day for the three rootstocks under well-watered left panel and water-stressed right panel conditions. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock (i,j)...... - 74 - Figure III.5: Model validation between observed and simulated E (a) and gs (b) values using data from experiment 2 (see materials and methods) collected at different soil water content during two days on three rootstocks (mean ± SE, n=10). The lines represent the 1:1 relationship between observed and simulated values. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock. .- 75 - Figure III.6: Model validation taking into account the decrease of root length area (Table S. 3, RLA) parameter as a function of SWC for plants experiencing long-term water stress. Observed and simulated E (a) and gs (b) values using data from experiment 2 (see materials and methods) were collected at different soil water content during two days for the three rootstocks (mean ± SE, n=10). The lines represent the 1:1 relationship between observed and simulated values. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock...... - 76 - Figure III.7: Heatmap of sensitivity coefficient (SCs), which is defined as the ratio between the relative variation of the predicted variable and the relative variation of the parameter (± 20%). The SCs were computed to decipher the influence of 11 parameters (rows, meanings of parameter are given in Table S. 3) on the 6 functional variables (columns, i.e. leaf water potential [ leaf ], root water potential [ root ], transpiration rate [E], stomatal conductance [gs ] and ABA concentration in the xylem sap [ABA]) for the 3 rootstocks (columns, 110R, CS and RG). In order to simplify the figure, the effects of soil water content (SWC) on SCs were summarised by using median value of SCs calculated with different SWCs. The biological and physical processes regulated by parameters are indicated on the left of the heatmap...... - 77 - Figure IV.1: Relationship between soil water content (SWC) and predawn leaf water potential (PD ) measured for the two scion-rootstock pairs (CS onto RG, CS onto 110R). The SWC data correspond to the target values of each treatment ( i.e. CTL: control; LWD: low water-deficit; MWD: moderate water-deficit and HWD: high water-deficit). Each point represents the mean value ± standard deviation (SD) of 7-10 samples...... - 94 - Figure IV.2: Final leaf area (LA) per plant for V. vinifera cv. Cabernet Sauvignon (CS) grafted onto V. riparia cv. Gloire de Montpellier (RG) and 110 Richter (110R) measured at the end of each experiment in 2009 (a) and 2010 (b). CTL: control; LWD: low water- deficit; MWD: moderate water-deficit and HWD: high water-deficit. Each value is the

IV ...... - 94 - Figure IV.3: Effects of water-deficit (CTL: control; LWD: low water-deficit; MWD: moderate water-deficit and HWD: high water-deficit) on the relative growth rate (RGR) of whole plants for each scion/rootstock pair (CS onto RG, CS onto 110R) in 2009 (a) and 2010 (b). Mean (n=5) ± SE. Different letters indicate significant differences (p<0.05)...... - 95 - Figure IV.4: Biomass allocation to the different compartments presented for each scion/rootstock pair (RG for CS onto RG; 110R for CS onto 110R) and for each treatment. The biomass allocation to the different compartments was expressed as percentage of whole plant weight (the trunk was not included). Mean (n=10) ± SE.. - 96 - Figure IV.5: Light photomicrographs of root apical cross sections of Vitis riparia (RG, A,C) and 110 Richter (110R, B,D) under well-watered (CTL, A,B) and moderate water- stressed (MLD, C,D) conditions. The scale bar represents 0.1 mm...... - 97 - Figure IV.6: Absolute frequency distribution (a,b) and relative frequency (c,d) of conduits of different diameter size classes from root tips of a drought-sensitive (RG, triangle symbols, a,c) and a drought-tolerant (110R, square symbols, b,d) rootstock under well-watered (CTL, open symbols) and water-stressed (MWD, filled symbols) conditions. Relative theoretical hydraulic conductivity ( i.e. percent contribution of a diameter class to total hydraulic conductivity of the cross section) was calculated according to Eqn. 2 for each rootstock (RG, e, 110R, f). Data are averages of 4-5 replicates per treatment and genotypes (± SE)...... - 98 - Figure IV.7: Single root hydraulic conductivity (Khr ) at four levels of water supply for the drought-sensitive (RG, a) and the drought-tolerant (110R, b) rootstock (mean ± SD, n=10). Different letters indicate significant differences among treatments and rootstocks (p<0.05)...... - 99 - Figure V.1: Illustration of the long term steady-state drought treatments and sampling procedure used in this study. A. Development of Soil Water Content (SWC) over time for well-watered control (CTL) and three drought treatments: low water-deficit (LWD), moderate water-deficit (MWD) and high water-deficit (HWD). Years and genotypes were plotted together since no significant effect of year (p=0.726) and genotype (p=0.841) have been observed (three ways ANOVA) during the steady-state conditions (Days after treatment 11 to 25). Each point corresponds to the mean (n=6) ± SD of SWC collected at the same hour (8:00 am). Arrows indicate the sampling time of roots for RNA analyses. B. SWC values measured the morning (8 am) of root sampling for each treatment, year and rootstock. Each bar represents mean (n=3) ± SD. Treatments were significantly different (p<0.001) and different letters over the bars indicate statistically significant differences between treatments at 5% level using Tukey multiple comparison test...... - 111 - Figure V.2: Gene numbers regulated in response to water-deficit stress levels in the roots of the three rootstock genotypes differing by water stress tolerance. The Venn diagram presents the differentially expressed genes (p-value<0.05, false discovery rate (FDR) using Holm method to adjust the p-value, log2 fold change (FC) > 1) in the three water- stressed conditions (LWD, MWD and HWD) compared to well-watered condition (CTL) for each rootstock (n=6, A: CS, B: RG and C: 110R). The upregulated genes were showed in red (“Up” without underline) and downregulated were showed in blue (“Down” with underline) in the Venn diagram...... - 112 - Figure V.3: Number of significant genes by factors or factors interactions obtained from the three-ways ANOVA analysis (p-value<0.05, FDR p-value using Holm method). The 24

V - 113 - Figure V.4: Principal Components Analysis (PCA) of the transcripts whose significant T×G interactions are observed in ANOVA analysis and listed in Table V.2. The transcripts without Genoscope annotation or putative function were not included in the PCA, but the patterns were unchanged if they are included. (A) Spatial distribution of rootstocks for each treatment is drawn on the two-first principal components (PCs). Each points represent an individual biological replicate collected either in 2009 (n=3) or 2010 (n=3). (B) Correlation plots of transcripts for the first two PCs of PCA analysis...... - 116 - Figure V.5: Box plots of the log2 expression intensities of the 7 genes having significant T×G interactions (FDR p-value<0.1%) from ANOVA analysis (microarray). Each column of plots corresponds to one genotype ( i.e. CS, RG and 110R) and each line of plots corresponds to one gene ( i.e. AER, PBP, ADH, KIP, LOX, CAX and CYP). The grape genome 12x targets and putative function of each gene were presented in Table V.2. Each box corresponds to six biological replicates for each treatment ( i.e. CTL, LWD, MWD and HWD) collected either in 2009 (n=3) and 2010 (n=3). The box whiskers plots visualize the minimum (bottom of the whisker cap), the 25th percentiles (bottom border of the box), the median (line through the box), the mean (dotted line through the box), the 75th percentiles (top border of the box), and the maximum (end of the whisker cap) of the distribution. For each gene, different letters over the boxes indicate statistically significant differences between treatments and genotypes at 5% level using Tukey multiple comparison test...... - 117 - Figure V.6: Gene ontology (GO) analysis. For each GO category ( i.e. 1. Biological process [a-h], 2. Molecular function [i-p] and 3. Cellular component [q-x]), the sub-categories were selected according to the highest number of significant terms for all treatments and rootstocks. In each radial plot, the rootstocks CS, RG and 110R are represented by green, red and blue colours respectively. For each rootstock, the increasing levels of water- deficit ( i.e. LWD, MWD and HWD) are presented in counter-clockwise. The radial plots show both the mean value of the number of genes observed (n=6) for each treatment and rootstock (green, red and blue colour scales) and the dotted lines the number of gene expected. A Fisher’s exact test was used for assessing the overrepresentation or underrepresentation with a cut-off of p<0.005. A star (*) shows the significant difference...... - 119 - Figure VI.1: Summary of the main results obtained in the present work. The drought-sensitive (RG) and the drought-tolerant (110R) are compared according to their common and specific responses in both roots and shoots. The bold superscript numbers refer to chapter numbers as follow: 1 Chapter III, 2 Chapter IV and 3 Chapter V. The abbreviations used are: soil water content (SWC), soil water deficit (SWD), well-watered (WW), moderate water-deficit (MWD), high water-deficit (HWD), predawn leaf water potential ( PD ), leaf water potential ( leaf ), transpiration rate ( E), stomatal conductance ( gs), minimum stomatal conductance ( gsmin ), maximum stomatal conductance ( gsmax ), xylem sap ABA concentration ([ABA]), leaf-specific hydraulic conductance (Kleaf ), single root hydraulic conductivity (khr ) whole plant leaf area (LA), whole plant relative growth rate (RGR), root length area (RLA), gene ontology [GO] and several transcripts (2-alkenal reductase [AER], cation/H+ exchanger [CAX], kinase interacting protein [KIP], lipoxygenase [LOX], phosphatidylethanolamine binding protein [PBP])...... - 129 -

VI List of tables

Table I.1: Agronomic characteristics of important grapevine rootstocks (modified from Keller, 2010)...... - 12 - Table II.1: Overview of the most important rootstock genotypes used in the world including parentage, phylloxera resistance and water stress adaptation characteristics. The grey scale part showed rootstocks with high adaptation to water-deficit stress...... - 22 - Table IV.1: Average values (n=5) of the anatomical characteristics of apical roots of the two genotypes. Within a column, means that do not share the same letter are significantly different (p<0.05)...... - 97 - Table V.1: List of criterions used to qualify the alignment homology of sequence annotation. - 110 - Table V.2: List of transcripts whose significant T×G interactions are observed from ANOVA analysis (Figure V.2). The G12X target in 1st heading column refers to the Genoscope annotation (http://www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/) of grape genome (Jaillon et al. , 2007). The C12X target in the 2nd column refers to CRIBI annotation (http://genomes.cribi.unipd.it) and was given for information to complete transcript without prediction from Genoscope. The 3rd column shows the chromosome (Chr) position. All the putative functions have been controlled manually using BLASTP tools of NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The five right columns show accession number, species-referred, putative protein functions, query coverage and match E-value...... - 114 - Table V.3: Correlations of microarray (log2 expression intensity) and qPCR (log2 relative expression) data for 7 genes verified. All correlations were calculated using Spearman’s rank correlation coefficient. All the data points (i.e. each biological replicate, n=36) were used for the calculations of correlation...... - 115 -

VII Chapter I. General introduction

- 8 - Chapter I. General introduction

Grapes ( Vitis spp.) are the most important fruit crop in terms of area. In 2009, world vineyards reached a total surface of 7.66 Mha ( Castellucci, 2010 ) among which 7.437 Mha were harvested ( FAO, 2011b ). With regard to global production, grapes are the fourth fruit species (after bananas, apples and oranges, respectively) with 669.35 Mqx ( FAO, 2011b ) of which more than 70% are used for wine production, 27% consumed as fresh fruit (table grapes), 2% as dried fruit (raisins) and less than 1% was processed to grape juice or distilled (OIV, 2007 ). World wine production (other than juice and musts) in 2007 was 265.99 Mhl (lower than the average of the 2001-2005 period which produced 272.78 Mhl) among which 68% was made in Europe, 18% in America, 5% in Oceania, 5% in Asia and 4% in Africa (OIV, 2007 ). The international wine industry provides to millions of investors and hundreds of millions of consumers a far more fascinating product than its shares of global expenditure might suggest ( Anderson et al. , 2003a ). The historical connections with the developmental of human cultures and religions, combined with the importance of wine in the way of life of people and its benefits for health, the so-called “French paradox”, led to consider grapes and wines as unique among food and agricultural products ( Renaud & de Lorgeril, 1992 ; Unwin, 1996 ; Bisson et al. , 2002 ; Vidavalur et al. , 2006 ). Grapevine is cultivated all over the world in the northern hemisphere between 20° and 50° North latitudes, and in the southern hemisphere between 20° and 40° South latitudes. Botanically, the grapevine is a liana belonging to the Vitaceae family. Phylogenetic analysis based on the entire plastid genome places the Vitaceae into the earliest diverging lineage of rosids ( Jansen et al. , 2006 ). Grapes are mostly woody, tree-climbing vines, characterized by tendrils and inflorescences opposite the leaves in an alternate phyllolactic type and a pronounced acrotonic pattern of shoot growth ( Pratt, 1974 ). The family contains approximately 1000 species assigned to 17 genera, including Vitis [Tournefort], Cissus [Descoings], Ampelopsis [Michaux], Ampelocissus [Planchon], and Parthenocissus [Planchon] which are the most important ( Levadoux et al. , 1962 ; Galet, 2000 ; Keller, 2010 ). Grapes exhibit a tremendous genetic diversity which underlines the necessity of international collaboration and cooperation for germplasm collection, maintenance and preservation. The inventory of the world-wide Vitis species, varieties and genotypes grown in germplasm collections is recorded in two databases: i) European Vitis database ( http://www.eu-vitis.de/ ) and ii ) Vitis International Variety Catalogue ( http://www.vivc.de ) both maintained by the Julius Kühn institute and the Institute for grapevine breeding of Geilweilerhof in Germany (Maul et al. , 2011 ). All cultivated grapes belong either to the genus Muscadinia (2n=40 chromosomes) or the genus Vitis (2n=38 chromosomes). The former classification of Muscadinia and Euvitis in two sub-genera of the genus Vitis is outdated of favour among taxonomists ( Bouquet et al. , 2009 ; Keller, 2010 ). Despite no natural hybridization with those of the genus Vitis , the genetic potential of Muscadinia species, such as Muscadinia rotundifolia [Michaux], has been used to create interspecific crosses with enhanced diseases resistance ( Olmo, 1986 ; Walker et al. , 1994 ; Bouquet et al. , 2000 ; Barker et al. , 2005 ; Esmenjaud et al. , 2010 ). The genus Vitis consists of ~60 species and is often divided into two major groups: the Eurasian group and the American one which greatly differ in their useful agronomic traits

- 9 - Chapter I. General introduction

(Keller, 2010 ). All the species within this genus are inter-fertile to form interspecific hybrids and can be grafted onto each other ( Alleweldt & Possingham, 1988 ; Keller, 2010 ). Recently, the first molecular phylogeny based on plastid DNA markers, demonstrated the monophyly of the genus Vitis and the existance of three clades (American, Asian and European) that mostly mirror geographic distribution of the taxa ( Tröndle et al. , 2010 ; Péros et al. , 2011 ). The findings of these authors support a relatively recent and intense gene flow between East Asia and North America. The genetic diversity within the Asian clade is high, but it is low within North America and Europe, which suggests hybridization between cultivated grapevine and autochthonous accessions in the evolution of the genus Vitis ( Tröndle et al. , 2010 ; Péros et al. , 2011 ). The Eurasian group is dominated by one species of major agronomic interest, i.e. Vitis vinifera [Linné] ( V. vinifera L.) which is extensively used in wine industries (Alleweldt & Possingham, 1988 ; This et al. , 2006 ). Among V. vinifera L., the cultivated form, V. vinifera subsp. vinifera (or sativa ) and the wild form, V. vinifera subsp. silvestris [Gmelin] (or sylvestris ) still co-exist but differ in several traits ( This et al. , 2006 ). The domesticated form, including an estimated number of more than 10,000 cultivars, gave rise to the most renowned grape varieties cultivated today such as Chardonnay, Riesling, Pinot noir, Syrah, Merlot and Cabernet-Sauvignon ( Galet, 1988 ; Alleweldt et al. , 1990 ; Galet, 1990 ). This large genetic diversity has been probably driven by several processes such as sexual crossing, vegetative propagation and mutations during the evolution of cultivated grapes (for details see reviews of Arroyo-García et al. , 2006 ; This et al. , 2006 ; Bessis, 2007 ; Bouquet et al. , 2009 ; Pelsy, 2010 ; Terral et al. , 2010 ). However, the considerable number of varieties within V. vinifera is part of a complex network of close pedigree relationship that has been generated by crosses among elite cultivars ( Myles et al. , 2011 ). In fact, since the discovery of the origin of V. vinifera L. cv. Cabernet-Sauvignon ( Bowers & Meredith, 1997 ) based on the inheritance of nuclear SSR DNA markers, the parentage of many others cultivars such as Cornalin ( Vouillamoz et al. , 2003 ), Merlot ( Boursiquot et al. , 2009 ) and Sangiovese ( Di Vecchi Staraz et al. , 2007 ) has been dtermined (see also Vouillamoz & Grando, 2006 ; Cipriani et al. , 2010 ; Santana et al. , 2010 ). The old cultivar V. vinifera L. cv. Gouais blanc is the progenitor of a very large kin group of major wine grapes ( Bowers et al. , 1999 ; Hunt et al. , 2010 ), while some others ( e.g. Pinot blanc and Grenache blanc) are mutations of the black cultivars ( Kobayashi et al. , 2004a ; Walker et al. , 2006 ; 2007 ). Regarding the American group, Vitis native to America have been used in wine industries (e.g. V. labrusca L.), in breeding programs as crossing partners, or as rootstocks ( Galet, 1988 ; Keller, 2010 ). The American Vitis species have specific native habitats; most of them grow near a permanent source of water (canyons, riverbeds, alluvial soil, along streams and moist woody areas, etc.) but also in dry and rocky zones (rocky hills, dry hillsides, dunes, etc.). Their natural habitats generally reflect the adaptation of each species to environmental conditions ( Pongrácz, 1983 ; Galet, 1988 ; Morano & Walker, 1995 ; Padgett-Johnson et al. , 2003 ; Arrigo & Arnold, 2007 ). Following is an incomplete list of some of the American species: V. aestivalis [Michaux], V. arizonica [Engelmann], V. berlandieri [Planchon], V. californica [Bentham], V. candicans [Engelmann], V. cinerea [Englemann], V. cordifolia

- 10 - Chapter I. General introduction

[Michaux], V. doaniana [Munson], V. girdiana [Munson], V. labrusca [Linné], V. longii [Prince], V. monticola [Buckley], V. riparia [Michaux], V. rupestris [Scheele] ( Viala & Vermorel, 1901-1910 ; Pongrácz, 1983 ; Galet, 1988 ). The naturalised rootstock populations appear to reproduce sexually, and show genetic pools that are clearly distinct from V. vinifera ssp. silvestris . In some areas, their populations behave as invasive species ( Arrigo & Arnold, 2007 ). As the spread of viticulture followed the travels of European explorers, American grape species were brought back to Europe with their pests and diseases.

Aboveground Scion Vitis vinifera L.

Grafting point

Belowground Rootstocks Vitis spp.

Figure I.1: Drawing of grafted grapevine representing aboveground, belowground and the grafting point (here, the graft is represented by an Omega grafting).

Grapevines have been propagated from cuttings until the late 19th century when the soil born aphid phylloxera ( Daktulosphaira vitifoliae [Fitch]) destroyed V. vinifera L. vines grown on their own roots. Phylloxera wreaked havoc on European vineyards, forever changing the manner in which grapevines are grown. The damaging importation of grape phylloxera led to extensive research to save vineyards (for details on history, biology and management of this pest, see reviews by Galet, 1982 ; Pongrácz, 1983 ; Pouget, 1990 ; Granett et al. , 2001 ). Around 1870’s, Léopold Laliman and Gaston Bazille were the first ampelographers to observe that the roots of some American Vitis species were not damaged by phylloxera ( Pouget, 1990 ). Since then, much research has been done on American Vitis species ( Viala & Vermorel, 1901-1910 ). Many authors recently reviewed the characteristics of American Vitis species and grapevine rootstocks ( Pongrácz, 1983 ; Galet, 1988 ; Pouget, 1990 ; May, 1994 ; Galet, 1998 ; Cordeau, 2002 ; Dry, 2007 ). Nowadays, more than 80% of the vineyards around the world use grafted plants: a scion of V. vinifera L. grafted onto a rootstock of single American Vitis species or interspecific hybrids of Vitis species to combine desirable features of parentage. The majority of rootstocks used

- 11 - Chapter I. General introduction today are hybrids of three species: V. berlandieri , V. riparia and V. rupestris ( Galet, 1998 ; Whiting, 2005 ). Consequently, the cultivated grapevine is a combination of two genomes. The grafting point defines the interface between shoot (aboveground organs, i.e. V. vinifera L.) and root (belowground organs, i.e. rootstocks) (Figure I.1).

Phytophtoraresistance Scion fruitmaturation Ease Ease of Crowngall resistance Grafted Nematoderesistance Sandy soilSandy Susceptibility Floodingtolerance Acidso Clays Susceptibility Sa E linitytolerance Limetolerance bench grafting ase ofrootingase scion vigour oil tolerance oil iltolerance deficiency deficiency tolerance toMg K to

Rootstock Riparia Gloire N A Rupestris St George N D Rupestris du Lot D 420A MGt Y D 5BB Kober Y Y D SO4 Y N 8B 5C Teleki A 161-49 Couderc 99 Richter Y Y D 110 Richter Y Y D 1103 Paulsen N Y D 140 Ruggeri N Y D 44-53 Malègue Y N A 3309 Couderc N A 101-14 MGt A Schwarzmann Gravesac 1616 Couderc A Salt Creek (Ramsey) Dogridge Harmony Freedom Table I.1: Agronomic characteristics of important grapevine rootstocks (modified from Keller, 2010 ).

= Excellent; = High; = Medium; = Poor; = Low A: Advanced; D: Delayed; N: No; Y: Yes

Inaddition to phylloxera resistance (Table II.1), grape rootstock genotypes are also used for their ability to influence other major agronomic parameters (Table I.1). These include resistance to nematodes ( reviewed by Nicol et al. , 1999 ) or adaptability to different soil conditions such as high active lime content, low pH, salinity, and water logging ( Galet, 1998 ; Cordeau, 2002 ; Whiting, 2005 ). The edaphic conditions strongly influence the feedback of the whole grafted plant described above. Therefore, considerable research has been done in

- 12 - Chapter I. General introduction relation to the effect of rootstocks on water uptake (reviewed in Chapter II) and on mineral nutrition such as nitrogen ( Keller et al. , 2001a , 2001b ; Zerihun & Treeby, 2002 ; Holzapfel & Treeby, 2007 ), potassium ( Kodur et al. , 2010 ), phosphorus ( Grant & Matthews, 1996a , 1996b ), magnesium ( Garcia et al. , 2001a ), iron ( Bavaresco et al. , 1991 ; 1994 ; Jiménez et al. , 2007 ), sodium and chloride ( Gong et al. , 2010 ; Walker et al. , 2010 ; Gong et al. , 2011 ). The complex interactions between scion and rootstock have been widely studied, especially the effects of rootstock on shoot development and grape quality ( Ollat et al. , 2003 ; Tandonnet et al. , 2010 ). The rootstocks can deeply influence the traits of the scion such as bud fertility ( Huglin, 1958 ; Benz et al. , 2007 ), phenology ( Pongrácz, 1983 ; Whiting, 2005 ), leaf area and canopy development ( Paranychianakis et al. , 2004 ; Clingeleffer & Emmanuelli, 2006 ; Koundouras et al. , 2008 ), wood pruning weight ( Ezzahouani & Williams, 2005 ; Stevens et al. , 2008 ) and yield ( Jones et al. , 2009 ; Stevens et al. , 2010 ). Because grape and wine have a tremendous economic impact, the role of rootstock genotypes has been assessed for many fruit quality traits such as carbohydrates ( Ezzahouani & Williams, 1995 ), organic acids ( Rühl et al. , 1988 ; Garcia et al. , 2001b ), amino acids ( Huang & Ough, 1989 ; Treeby et al. , 1998 ) and secondary metabolites ( Walker et al. , 2000 ; Sampaio, 2007 ; Koundouras et al. , 2009 ). Unfortunately, our current understanding of grape root physiology is much more limited than shoots. Few studies have investigated the mechanisms of resistance and tolerance to abiotic stress such as drought (reviewed in Chapter II). Moreover, in a climate-change threatened world where rain and temperature will be increasingly erratic, the risks of abiotic and biotic stresses will probably be amplified raising pressure on agriculture. Therefore, the predicted climate change will be one of the major challenges for wine industry since the phylloxera crisis, and the introduction of powdery mildew ( Erysiphe necator [Schwein]) and downy mildew ( Plasmopara viticola [Berk. & Curtis] Berl. & de Toni]) in the 19 th century (Ollat et al. , 2011 ).

I.1 Context of the present studies

I.1.1 Global warming: new challenges for facing the future

I.1.1.1 General considerations Global warming is unequivocal: eleven of the last twelve years in the last state of the climate analysis (1995-2006) ranked among the twelve warmest years in the instrumental record since 1850 ( IPCC, 2007 ). Autumn 2006 and winter 2007 were extremely likely (>95%) to be the warmest in Europe since more than 500 years ( Luterbacher et al. , 2007 ). An observed increase of 0.74°C (for 1906-2005, with a range from 0.56 to 0.92°C) in global average temperature is probably due to anthropogenic factors, namely greenhouse gases

(GHG). Among GHG, the carbon dioxide (CO 2) is the most important. The atmospheric CO 2 concentration has increased since the pre-industrial period from 280 to 379 ppm in the year 2005 and an increase to approximately 700 ppm is expected by the end of the 21 st century. Continued GHG emissions at or above current rates would cause further warming and many changes in global climate. For the next two decades, a temperature increase of about 0.2°C

- 13 - Chapter I. General introduction per decade is projected; and at the end of the 21 st century, the increase of global temperature could fluctuate between 1.1 and 6.4°C depending on scenarios for GHG emissions ( IPCC, 2007 ). Human GHG emissions contributed significantly to precipitation trends during the 20 th century: precipitations increased significantly in the northern hemisphere mid-latitudes and in the southern hemisphere subtropics and deep tropics, but declined in the northern hemisphere subtropics and tropics ( IPCC, 2007 ; Zhang et al. , 2007 ; Bates et al. , 2008 ). There is still considerable uncertainty about the rates of change that can be expected, but it is clear that these changes will be increasingly obvious in important and tangible ways ( Karl & Trenberth, 2003 ). In the future, the rise in precipitation intensity and variability is projected to increase the risks of flooding and drought in many areas; and the surface proportion under extreme drought, globally, is predicted to increase over low and mid-latitudes by the end of the 21 st century ( Allen & Ingram, 2002 ; Milly et al. , 2005 ; Allan & Soden, 2008 ; Bates et al. , 2008 ). Recent global warming have already caused significant consequences for plant phenology (Menzel et al. , 2006 ; Menzel & Sparks, 2007 ) and physiological processes such as growth (Davies, 2007 ; Ziska & Bunce, 2007 ) and nutrition ( Kreuzwieser & Gessler, 2010 ; Lukac et al. , 2010 ). Moreover, there are signs that the range, the distribution of species ( Walther et al. , 2002 ; Parmesan & Yohe, 2003 ; Root et al. , 2003 ; Breshears et al. , 2005 ) or the agricultural production ( Olesen & Bindi, 2002 ; Maracchi et al. , 2005 ; Piao et al. , 2010 ) are already affected by climate change. The projected climate change will worsen these effects. Using the data from Free-Air CO 2 Enrichment (FACE) experiments, Leakey et al. ( 2009 ) reviewed the effects of elevated CO 2 concentration on plant physiology. Six conclusions were drawn from

FACE experiments: high CO 2 concentration i) stimulates photosynthetic carbon gain and biomass, ii ) improves nitrogen use efficiency, iii ) decreases water use at both leaf and canopy scale, iv ) stimulates dark respiration, v) the stimulation of carbon uptake in C4 plants is indirect and occurs only under water deficit and vi ) the increase in yield is less than expected

(Leakey et al. , 2009 ). However, the direct beneficial impact of rising atmospheric CO 2 ( Drake et al. , 2003 ) can be offset by other effects of climate change such as warming and altered precipitation patterns ( DaMatta et al. , 2010 ). Moreover, multiple environmental stresses induced by climate change will trigger major changes in physiological and molecular traits of plants ( Ahuja et al. , 2010 ) and subsequent shifts in their phenotype ( Nicotra et al. , 2010 ).

I.1.1.2 What about grapevine? The impacts of recent and long term climate changes on grapevine development and on fruit quality have been reviewed by several authors ( Schultz, 2000 ; Jones et al. , 2005 ; Holland & Smit, 2010 ; Jones & Webb, 2010 ; Mira de Orduña, 2010 ; Schultz, 2010 ; Schultz & Jones, 2010 ). The biggest challenges with climatic change in viticulture are the potential shift in the suitable grape cultivars planted, changes in regional wine styles and/or regional change in the suitability of certain grape areas. Thus, the concept of “terroir” ( See van Leeuwen & Seguin, 2006 ) is questioned and threatened by the potential impact of climate change ( Jones, 2006 ). White et al. ( 2009 ) raised doubts about the lifespan of the rigid concept of a single varietal or wine style in the next decades and argued in favour of flexibility to keep the concept of “terroir” alive. Currently, many regions appear to be at or near their optimum

- 14 - Chapter I. General introduction grapegrowing season temperature ( Schultz, 2000 ; Jones et al. , 2005 ). Several studies have already reported climate trends and its effects on various grapegrowing areas of the world such as in Australia ( Webb et al. , 2007 ; Petrie & Sadras, 2008 ; Sadras & Petrie, 2011 ), Europe ( Duchêne et al. , 2010 ; Malheiro et al. , 2010 ; Pieri, 2010 ) and the United States ( White et al. , 2006 ; Lobell et al. , 2007 ). Some regions that were once considered unsuitable for quality wine grape production may become suitable in the future ( Hall & Jones, 2009 ) such as mountain sites at elevations of about 1000 meters (Caffarra & Eccel, 2011 ). Predicted changes in the climate drivers like rising temperature and CO 2 concentration, change in levels of UV-B radiation and variation of precipitation patterns will deeply alter viticultural practices, grapevine physiology, grape biochemistry and wine quality (see excellent reviews of Schultz, 2000 ; Mira de Orduña, 2010 ). Additionally, the rapidly increasing world population and the scarcity of suitable land for agricultural food production together with a changing climate will ultimately put pressure on grape-producing areas for the use of land and the input of resources ( Schultz & Stoll, 2010 ). Unfortunately, the studies about the effects of climate change are rather fragmentary, and only a few combine simultaneously water deficit and high-temperature stress ( Edwards et al. ,

2011 ) or the interactive effects of elevated CO 2, temperature and water availability ( Salazar Parra et al. , 2010 ). The key regulatory mechanisms of grapevine physiology under modified environmental conditions (high CO 2, high temperature and water deficit), from the gene to the whole plant and vineyard level, must be investigated in order to help the mitigation of arising problems ( Schultz & Stoll, 2010 ). The breeding of new genotypes better adapted to fluctuating weather can be investigated, but requires the identification of pertinent genetic loci. This goal may be achieved by multidisciplinary research, involving ecophysiologists, molecular physiologists and geneticists ( see review by Ollat et al. , 2011 ). These authors described four strategies to sustain viticulture: i) modify the cultivation practices [ e.g. irrigation techniques ( Chaves et al. , 2010 ) or canopy management such as leaf-to-fruit ratio (Stoll et al. , 2010 )], ii ) adapt new oenological practices, iii ) select new cultivars and rootstocks and iv ) find new grapegrowing regions ( Ollat et al. , 2011 ).

I.2 Objectives and outlines of the thesis Currently, more than two billion people live in highly water-stressed areas because of the uneven distribution of renewable freshwater resources, while water scarcity is likely to affect up to two thirds of the world population over the next decade ( Oki & Kanae, 2006 ). Water scarcity is often intuitively associated with lack of drinking water, but it is mainly a result of insufficient water for agriculture ( Savenije, 2000 ). At a global level, agricultural water use accounts for 70% of the total water withdrawn ( FAO, 2011a ) and rising food demand and growing water scarcity will put increasing pressure on agriculture. In fact, crop yields are strongly reduced when plants are exposed to unfavourable conditions in the field ( Boyer, 1982 ). More than 80% of global agricultural land is rain-fed; in these regions, crop productivity depends solely on sufficient precipitation to meet evaporative demand and associated soil moisture distribution. The increase in the frequency of climate extremes may lower crop yields beyond the impacts of mean climate change ( Bates et al. , 2008 ). As rainfall

- 15 - Chapter I. General introduction patterns become more unpredictable as our climate evolves, plants will be subjected to increasing fluctuations in soil moisture availability. Therefore, the improvement of crop performance under water deficit has been challenging due to the complexity of the traits at the molecular and physiological level and the multitude of factors that influence the plant’s response. Sophisticated tools must be developed to monitor phenotype expression at the crop level to characterize variation among genotypes across a range of environments ( Sinclair, 2011 ). A combination of approaches using genetic engineering ( Collins et al. , 2008 ; Mittler & Blumwald, 2010 ; Osakabe et al. , 2011 ), genomic and bioinformatic tools ( Cushman & Bohnert, 2000 ; Tuberosa & Salvi, 2006 ; Mochida & Shinozaki, 2010 ) are now effective methods for the molecular understanding of gene function and regulatory networks involved in plant abiotic stress tolerance ( Takeda & Matsuoka, 2008 ).

I.2.1 Objectives of the thesis Climate change raises concerns about temporal and spatial water availability in many grape growing countries. In Europe, irrigation in many areas is not a sustainable way to counteract water scarcity because of the legal limits, the infra-structural problems and the increase of water demand. Consequently, other practices that can potentially improve water management of vineyards and water acquisition by grapevines need to be considered. Aside from canopy systems and their management, the choice of plant material is a key issue ( Ollat et al. , 2011 ). However, in many regions, such as Rheingau (Hessen region, Germany) or Bordeaux (Aquitaine region, France), the reputation of wines is based on the use of specific grape varieties, such as Riesling or Cabernet-Sauvignon. Thus, the potential for improving water use efficiency is limited, while progress with respect to plant water relations may be achievable primarily through the adaptation of rootstocks to environmental stresses (reviewed in Chapter II).

The research objectives of this study were to determine the mechanisms underlying adaptation of grapevine rootstocks to drought, using a combination of ecophysiological, modelling and transcriptomic approaches in controlled conditions. Most studies on the mechanisms of water stress adaptation in grapevines were focussed on the canopy level and largely excluded rootstocks as a factor of acclimation. Graft-induced rootstock-scion interactions may affect water flow in the soil-plant-atmosphere continuum and/or plant water status. In this context, the role of rootstock genotypes was investigated for their potential effects on i) water uptake, ii ) water transport and iii ) shoot water use under contrasting soil water availability. To this aim, experiments were conducted under controlled conditions to decipher short and long term responses to drought of contrasting genotypes. On the one hand, for the short term response, a drying phase protocol was established to investigate different hypotheses related to rootstock control of stomatal aperture via chemical, hydraulic or genetic traits. The data collected were incorporated into a model ( Tardieu & Davies, 1993 ) allowing to test these hypotheses. On the other hand, long term steady-state drought conditions were used to examine how different water stress levels affect i) the root anatomy and hydraulic

- 16 - Chapter I. General introduction properties and ii ) the root transcriptome of contrasting genotypes. The transcriptomic approach was the first one for grape rootstocks under drought conditions. This thesis should provide insight into the responses of the root system to water limitation. The genes differentially expressed in the rootstock in response to water stress may be used for future functional genomic studies, and for the selection of new rootstocks using marker- assisted selection ( Töpfer et al. , 2011 ).

I.2.2 Outline of the thesis The thesis is organised into four main sections which correspond to submitted articles or drafts in preparation. These sections cover the role of rootstocks in the adaptation of grafted grapevine to drought at different levels of organisation, from the whole plant to gene studies.

The sections are arranged as follows: Z Section 1 (i.e. Chapter II): Literature review. In this section, the roles of rootstocks in grafted-grape adaptation to water-deficit are reviewed, both from a physiological and molecular standpoint.

Z Section 2 (i.e. Chapter III): Modelling approach. Grape rootstocks may affect scion gas exchange via complex pathways involving different physiological processes ( Iacono et al. , 1998 ; Soar et al. , 2006b ; Koundouras et al. , 2008 ; Alsina et al. , 2011 ), but the mechanisms of stomatal regulation in grafted-grape under drought are poorly understood. We investigated the role of root characteristics, plant hydraulic conductivity (Kh) and chemical signals (ABA) in the responses of stomatal conductance (g s) and transpiration (E) to water deficit in drought-sensitive and drought-tolerant rootstocks using a mechanistic stomatal regulation model ( Tardieu & Davies, 1993 ).

Z Section 3 (i.e. Chapter IV): Root hydraulic properties. Rootstock genotypes may affect shoot water use via different root characteristics and root hydraulic properties ( de Herralde et al. , 2006 ; Alsina et al. , 2011 ) as well as different scion growth patterns ( Tandonnet et al. , 1999 ; Ollat et al. , 2003 ; Koundouras et al. , 2008 ; Tandonnet et al. , 2010 ). Therefore, several plant traits such as leaf area development, biomass partitioning, root xylem conduit development and hydraulic conductivity were investigated using drought-sensitive and drought-tolerant rootstocks under controlled drought conditions.

Z Section 4 (i.e. Chapter V): Transcriptomic analysis. At low water availability, plants evolve physiological and developmental adjustments that are, in part, driven by remodeling of the transcriptome. The reconfiguration of the transcriptome towards water limitation has been observed in various species ( Seki et al. , 2002 ; Degenkolbe et al. , 2009 ; Wilkins et al. , 2009 ; 2010 ) but not yet in the roots of grape rootstocks. The lack of molecular data in the roots of different grape rootstock genotypes under water deficit, the availability of the grape genome ( Jaillon et al. , 2007 ; Velasco et al. , 2007 ) and whole genome microarrays ( Bellin et al. , 2009 ; Polesani et al. , 2010 ) provided a

- 17 - Chapter I. General introduction good way to identify and integrate the function of candidate genes involved in natural genetic variation under drought ( Martinez-Zapater et al. , 2010 ).

I.3 Paper publishing and conference communication

I.3.1 Posters Pierre-François Bert, Jérôme Fernandez, Anthony Peccoux, Serge Delrot and Nathalie Ollat (2009) Growth and gene expression response of grapevine genotypes under osmotic stress. Poster presented at the COST 858 Viticulture Final Meeting “What's up in viticulture?” (ed S. Delrot), October 27-30, Bordeaux, France.

Anthony Peccoux, Brian Loveys, Philippe Vivin, Serge Delrot, Hans R. Schultz, Nathalie Ollat and Zhan Wu Dai (2011). Stomatal control by rootstock-sourced signals under water stress: a model-based analysis in grapevine. 2 nd International Plant Phenotyping Symposium “Toward plant phenotyping science: challenges and perspectives ”, September 5-7 th 2011, Forschungszentrum Jülich, Germany.

I.3.2 Oral communications Anthony Peccoux, Christian Kappel, François Barrieu, Serge Delrot, Hans R. Schultz and Nathalie Ollat (2011). Xylem development, hydraulic conductivity and gene expression are differentially affected in roots of drought-sensitive and drought-tolerant grapevine rootstocks. 6th International Symposium on Root Development: Adventitious, lateral and primary roots “Future directions in root research”, August 7-11 th , Amos, Canada.

I.3.3 Articles in peer-reviewed plant journals Anthony Peccoux, Brian Loveys, Philippe Vivin, Serge Delrot, Hans R. Schultz, Nathalie Ollat and Zhan Wu Dai. Control of stomata by rootstock-sourced signals under water stress: a model-based analysis in grapevine. Submitted to Plant, Cell and Environment in October 2011.

Anthony Peccoux, Christian Kappel, Hans R. Schultz, Nathalie Ollat and Serge Delrot. Long term steady-state drought-induced changes in grape rootstocks transcriptome. To be submitted to Plant Physiology before December 2011.

Anthony Peccoux, Bastien Golard, Serge Delrot, Hans R. Schultz and Nathalie Ollat. Xylem development and hydraulic conductivity are differentially affected in roots of drought- sensitive and drought-tolerant grapevine rootstocks. To be submitted to Planta in 2012.

Anthony Peccoux, Christian, Kappel, Serge Delrot, Hans R. Schultz and Nathalie Ollat. The function of rootstocks in the adaptation of grapevine to drought: A review. To be submitted to Annals of Botany in 2012.

- 18 -

Chapter II. The function of rootstocks in the adaptation of grapevine to drought: A review

- 19 - Chapter II. Literature review

Introduction A large proportion of vineyards in the world experience seasonal drought, where atmospheric (high VPD) and edaphic (soil) water deficits, together with high temperature and irradiance, exert large constraints on yield, grape and wine quality ( Chaves et al. , 2010 ; Cramer, 2010 ; Flexas et al. , 2010 ; Lovisolo et al. , 2010 ). Grapevines are well-adapted to arid and semi-arid climates. In these environments, grapevines are subjected either to a slow decrease in water availability during the growing season (edaphic water deficit) or short-term water stress (atmospheric water-deficit) ( Chaves et al. , 2003 ). Plants adapted to dry environments developed different strategies such as ( i) escaping water stress (short phenological cycle), ( ii ) avoiding water stress (reducing transpiration, increasing water uptake), ( iii ) maintaining growth under water stress through adaptative mechanisms, ( iv ) resisting to severe water depletion through survival mechanisms ( Tardieu, 2005 ; Verslues & Juenger, 2011 ). Grapevines appear to primarily rely on drought avoidance mechanisms ( Scienza, 1983 ; Chaves et al. , 2010 ). Water stress is not exclusively a negative phenomena since mild water-deficit can enhance grape quality for the production of red wines (Roby et al. , 2004 ; Chapman et al. , 2005 ; van Leeuwen et al. , 2009 ). Many agronomic practices have been used to induce mild water stress ( Chaves et al. , 2007 ; Chaves et al. , 2010 ; Lovisolo et al. , 2010 ). Since grapevines are mostly non-irrigated, especially in Europe, there may be a substantial risk of more frequent and more severe droughts in the near future due to climate change. One very simple way to overcome water stress without affecting quality and yield of grapevine plants could be the use of irrigation techniques ( see review of Chaves et al. , 2010 ). However, vineyard irrigation does not always represent a sustainable way to counteract low water availability, because of legal limits, infra-structure problems (access to water), the problems of successfully implementing irrigation systems on a larger scale and the growing competition for water as a resource. Moreover, the requirement of irrigation for crop production may induce soil salinity, another major abiotic stress ( Rengasamy, 2006 ). As an example, for the world’s allocation of water resources, 80% is already consumed by irrigated agriculture, and this level of consumption cannot be maintained in the future. Projected population growth will require more of the available water resources to be used for other purposes ( IPCC, 2007 ; Bates et al. , 2008 ). Indeed, the most realistic solution to this increased demand will be the reallocation of some of the water away from agriculture. Even a modest reallocation, such as reducing agriculture’s share to 70%, would increase the water available for other purposes by up to 50% ( Hamdy et al. , 2003 ). These new challenges for the international wine industry will require improvements in viticulture, which will be possible with a better understanding of grapevine genetic diversity and responses to environmental stresses ( Vivier & Pretorius, 2002 ). Genetic variability of grapevine rootstocks will undoubtedly play a fundamental role in the adaptation to future climate changes, especially to water shortage ( Walker, 1992 ), and will probably allow to breed new genotypes more appropriate to respond to drought scenarios ( Tardieu, 2005 ). Vitis is a genus in the Vitaceae family with a large range of species ( Pongrácz, 1983 ; Galet, 1988 ), and the choice of the plant material will be a sustainable strategy to overcome the increasing

- 20 - Chapter II. Literature review problem of water resource depletion in the future (Duchêne & Schneider, 2005 ; van Leeuwen et al. , 2009 ). Most of the American Vitis species cannot be employed alone as rootstock due to grafting incompatibility (low rooting capacity, low affinity with scions…). However, they can be used in breeding programs, where their adaptation to drought is reflected in the resulting hybrids, especially in crosses between V. berlandieri and V. rupestris (Table II.1). However, classification of the influence of rootstocks on water stress adaptation of the scion varies amongst the authors (Table II.1). To date, only one comprehensive comparison has been carried ( Carbonneau, 1985 ). Indeed, studies on the underlying mechanisms of water stress adaptation in grapevines deal mainly with eco-physiological, physiological and molecular responses of V. vinifera L. varieties and largely exclude rootstocks as a factor in acclimation. Water stress tolerance among rootstock genotypes varies ( Carbonneau, 1985 ), yet the physiological mechanisms are still unknown, with a surprising lack of molecular data on the roots of these important species.

GENETIC TRAITS OF SCION Vapor Pressure Deficit Temperature SHOOT WATER USE Light

Canopy structure and growth CO 2 Z Inhibition of leaf emergence and expansion, internode extension, tendril elongation, lateral shoot emergence and growth Z Reduced transpiration area (leaf area) WATER TRANSPORT Physiological processes Anatomical structure Z Root signal recognition (ABA, etc.) Z Grafting point hydraulic resistance Z Gas exchange responses (stomatal closure, etc.) Z Average vessel number and diameter Z Decreased carbon assimilation Hydraulic architecture Z Water potential ( ) responses Z Signal transport Z Water use efficiency responses Z Water transport pathways Z Metabolic acclimation Z Hydraulic conductivity responses, aquaporin responses Z Z Vulnerability to embolism Osmotic adjustment INTERACTIONS Z Multi-stress sensing Z Shoot-to-root signaling Z Root-to-shoot signaling Z Grafting compatibility GENETIC TRAITS OF ROOTSTOCK Z Assimilate partitioning WATE R UPTAKE Root system architecture Z Sustained root growth Z Increased root-to-shoot ratio Soil water Z Depth growth and branching Z Increased absorption area (fine root density) potential ( ) Physiological processes Z Plasticity to resources availability Z Interactions with symbiotic organisms Z Anatomical changes (suberization, etc.) Z Cell drought signaling SOIL PROPERTIES Z Hormones synthesis (ABA, etc.) Z Osmotic adjustment Z Structure, texture, nutrient availability Z Multi-stress sensing Z Water retention capacity Z Hydrotropism (growth toward soil moisture) Z Turgor maintenance

Figure II.1: Schematic presentation of scion-rootstock interactions affecting whole grafted grapevine responses to water-deficit conditions ( adapted from Chaves et al. , 2003 ).

- 21 - Chapter II. Literature review

Rootstocks Usual name Parentage Phylloxera resistance Water stress adaptation Riparia Gloire de Montpellier Riparia Gloire V. riparia Michaux 4,6,28 Very high 4,6,11,16,23 , High 20,22 Low 1,2,5,6,7,8,10,15,23 Grézot 1 G1 (Coude rc 1616× V. rupestris du Lot)×Ganzin 1 ( V. vinifera L. cv. Aramon× V. rupestris Ganzin) 4,11,28 Medium 4,6,11,23 , Low 4 Low 8,23 Foëx 34 École de Montpellier 34 EM V. riparia ×V. berlandieri 4,6,11,28 High 6,22,23 Medium 10,22 , Low 6,11,22 Millardet et de Grasset 420A 420A V. riparia ×V. berlandieri 4,6,11,28 High 6,11,20,22,23 Medium 3,5,22 , Low 1,2,7,8,11,23,24 , Very low 4,10 Kober-Téléki 5BB 5BB V. riparia ×V. berlandieri 4,6,11,28 High 6,20,22,23 Medium 1,3,5,22 , Low 2,4,6,11,15,24 Téléki 5C 5C V. riparia ×V. berlandieri 4,6,11,28 High 6,20,22,23 Medium 22 , Low 6,22,24 Couderc 1616 1616 Sol onis ( V. riparia ×V. longii )× V. riparia 4,6,11,28 High 4,6,11,20,22,23 Medium 6, Low 2,22,24 Rupestris du Lot (St. George) Rupestris V. rupestris Scheele 4,6,28 High 4,6,11,20,22,23 , Medium 4,22 Medium 4,8,22,23 , Low 1,2,5,6,11 Millardet et de Grasset 101-14 101-14 V. riparia ×V. rupestris 4,6,11,28 High 6,11,20,22,23 Medium 8,10,22,23 , Low 1,2,4,7,11,15,22 , Very low 24 Couderc 3309 3309 V. riparia Tomenteux× V. rupestris Martin 4,6,11,28 High 6,11,20,22,23 High 1, Medium 5,7,15,22 , Low 2,3,4,8,10,11,23 , Very low 24 Téléki-Fuhr Selection Oppenheim n°4 SO4 V. riparia ×V. berlandieri (Téléki n°4) 4,6,11,15 High 6,20,22,23 High 5,10 , Medium 7,15,22,23 , Low 1,2,4,8,24 , Very low 11 Téléki 8B 8B V. riparia ×V. berlandieri 4,6,11,28 High 6,23 Medium 23 , Low 11 Dog Ridge Dog Ridge V. rupestris×V. candicans 4,28 High 6,22 High 22 , Medium 22 , Low 8, Very low 10 Schwarzmann Schwarzmann V. riparia ×V. rupestris 4,9,22 Very high 16 , High 20,22 Medium 6,10,22 , Low 3,22 , Very low 24 Couderc 1613 1613 Solonis ( V. riparia ×V. longii )×Othello [Clinton ( V. riparia ×V. labrusca )× V. vinifera L. cv. Black Hamburg] 4,28 Medium 22 , Low 4,6 Medium 20,22,24 , Low 22,24 Couderc 161-49 161-49 V. riparia ×V. berlandieri 4,6,11,28 High 6,11,23 Medium 2,5,15,23 , Low 1,7,11,23 Kober-Téléki 125AA 125AA V. riparia ×V. berlandieri 6,11,28 High 6,18,23 Medium 17,23 Millardet et de Grasset 41B 41B V. vinifera L. cv. Chasselas× V. berlandieri 4,6,11,28 High 23 , Medium 4,6,11,20,23 High 2, Medium 1,4,5,8,15,23 , Low 7, Very low 10 Castel 216-3 216-3 1616 Couderc× V. rupestri s du Lot 4,6,28 High 4,6,23 Medium 2,6,11,23,24 Fercal INRA Bordeaux Fercal B.C. n°1B ( V. berlandieri Lafont n°9× V. vinifera L. cv. Ugni blanc)×31 Richter ( V. berlandieri Rességuier n°2× V. longii Novo-Mexicana) 25 High 6,11,23 , Medium 20,23 Medium 5,7,15,23 Gravesac INRA Bordeaux Gravesac Couderc 161-49×Couderc 3309 23,28 Very high 23 , High 20 Medium 7,11,15,23 Freedom Freedom 1613-59 (1613 Couderc×3306 Couderc)×Dog Ridge 26 High 20,22 , Medium 20 Medium 10,22 Harmony Harmony 1613 Couderc (Open Pollinated)×Dog Ridge (Open Pollinated) 4,28 Medium 8,22 , Low 4 High 10 , Medium 22 Foëx 333 École de Montpellier 333 EM V. vinifera L. cv. Cabernet-Sauvignon× V. berlandieri 4,6,11,15 High 11,23 , Medium 6,11,23 High 1,8,23 , Medium 2,4,11 , Low 5 Richter 99 99 R V. berlandieri Las Sorres× V. rupestris du Lot 4,6,11,28 High 6,11,22,23 Very high 10 , High 3,5,6,22,23,24 , Medium 1,2,4,6,8,11,15,23 Börner Börner V. riparia 183 Geisenheim× V. cinerea Arnold 14,16,28 Very high 14,16,24 High 12,17,21,24 Castel 196-17 196-17 1203 Couderc ( V. vinifera L. cv. Mourvèdre× V. rupestris Ganzin)× V. riparia Gloire de Montpellier 6,11,28 Medium 23 , Low 6,11 High 1,2,4,5,6,23 , Medium 2,11,15 Georgikon 28 Georgikon 28 Kobber 5BB× V. vinifera 28 High 19 High 13, 27 Malègue 44-53 44-53 V. riparia Grand Glabre×Malègue 144 ( V. cordifolia ×V. rupestris )4,6,11,15 High 4,11,23 Very high 5, High 1,2,23 , Medium 4,11 Ramsey Ramsey V. champinii Planchon 28 High 8,16,20,22 Very high 10,24 , High 22,24 , Medium 8,22 Paulsen 1103 1103 V. berlandieri Rességuier n°2× V. rupestris du Lot 4,6,11,28 High 6,20,22,23 Very high 10 , High 1,2,3,4,5,7,11,15,22,23,24 Paulsen 1447 1447 V. berlandieri Rességuier n°2× V. rupestris Martin 4,6,11,28 High 4,6,23 Very high 11 , High 5,6,23 Richter 110 110 R V. berlandieri Rességuier n°2× V. rupestris Martin 4,6,11,28 High 6,20,22,23 Very high 4,5,10,11,22 , High 1,2,7,8,11,15,23,24 Ruggeri 140 140 Ru V. berlandieri Rességuier n°2× V. rupestris du Lot 4,6,11,28 High 6,22,23 Very high 4,5,8,10,11,22,23,24 , High 2,3,7,15 , Medium 1 Table II.1: Overview of the most important rootstock genotypes used in the world including parentage, phylloxera resistance and water stress adaptation characteristics. The grey scale part showed rootstocks with high adaptation to water-deficit stress. The data were collected from 1Samson & Casteran ( 1971 ); 2Fregoni ( 1977 ); 3Fregoni et al. (1978 ); 4Pongrácz ( 1983 ); 5Carbonneau ( 1985 ); 6Galet ( 1988 ); 7Delas ( 1992 ); 8Southey ( 1992 ); 9Walker ( 1992 ); 10 Cirami et al. (1994 ); 11 Galet ( 1998 ); 12 Hajdu ( 1998 ); 13 Kocsis et al. (1998 ); 14 Schmid et al. (1998 ); 15 Cordeau ( 2002 ); 16 Kellow et al. (2002 ); 17 Schmid et al. ( 2003a ); 18 Schmid et al. ( 2003b ); 19 Tóth & Kocsis ( 2003 ); 20 Cousins ( 2005 ); 21 Schmid et al. ( 2005 ); 22 Whiting ( 2005 ); 23 Audeguin et al. (2007 ); 24 Dry (2007 ); 25 Laucou et al. ( 2008 ); 26 Garris et al. (2009 ); 27 Kocsis et al. ( 2009 ); 28 Maul et al. ( 2011 ).

- 22 - Chapter II. Literature review

In order to understand plant responses to drought, informations from the whole plant to the molecular level are required ( Chaves et al. , 2003 ; Cramer, 2010 ). The recent publications of grapevine genome sequence ( Jaillon et al. , 2007 ; Velasco et al. , 2007 ) and the growing availability of high throughput genomic tools (microarrays, ESTs, SAGE, MPSS, RNA-Seq, etc.) will allow a more comprehensive investigation of the molecular mechanisms underlying drought tolerance in grapevine ( Troggio et al. , 2008 ).

In this framework, we examined how grafting affects the scion-rootstock interactions and responses to drought, combining information from water uptake, transport and use. The chronology of this review is based on Figure II.1. The scion-rootstock interactions affecting the whole grafted-grape responses to drought have been divided into three functional categories acting together in a complex cross-talk: ( i) the genetic traits of rootstocks involved in water uptake, ( ii ) the coordination of anatomical structure and hydraulic architecture between the genotypes and ( iii ) the genetic traits of the scion controlling water use in the shoot (Figure II.1). The contribution of rootstocks in each functional category will be discussed in the following parts. First, the general features of grapevine root systems (rootstock) and physiological processes affecting water uptake will illustrate the strategies of adaptation to drought. Second, we summarise the key components of water transport that characterise whole-grapevine tolerance to water stress. Third, we will focus on the influence of rootstocks on scion water use under drought conditions.

II.1 Root system architecture and rootstocks responses to drought The root system has to serve several functions simultaneously: it is at the interface between the soil and plant within the soil-plant-atmosphere continuum. It anchors the trunk and shoots, stores carbohydrate reserves, and provides the surface area to exploit the water and nutrient resources of the soil ( Richards, 1983 ; Gregory, 2006 ; de Herralde et al. , 2010 ). In the soil-root interface, external factors can impose a large range of stresses, such as abiotic stresses including scarcity of water and nutrients. Individual plant genotypes can also assume particular characteristics within a given environment which is part of its so-called phenotypic plasticity ( Bradshaw, 1965 ; Nicotra & Davidson, 2010 ). Unfortunately, our current understanding of grapevine root plasticity and their regulation by the whole plant and environmental factors is limited ( Comas et al. , 2010 ). Root system architecture, especially the distribution of roots in the soil, can optimize water and nutrient uptake ( de Dorlodot et al. , 2007 ), and may provide a possible explanation for the abilities of different grapevine rootstocks to cope with water depletion ( Soar et al. , 2006b ).

The physiological and genetic processes controlling root growth, function and the complex formation of architecture have been reviewed by many authors. The intrinsic and environmental responses that regulate root growth and architecture have been of interest for the plant physiologists since many years ( see reviews of McCully, 1999 ; Malamy, 2005 ; Osmont et al. , 2007 ; Hodge et al. , 2009 ; Walter et al. , 2009 ; Ingram & Malamy, 2010 ).

- 23 - Chapter II. Literature review

II.1.1 Root system architecture Despite of the importance and the increasing interest of scientists on root research, many questions remain unanswered, since the majority of plant biologists have primarily focussed on the aboveground part of the plant ( Epstein, 2004 ). Root studies are difficult because in contrast to shoots, roots cannot be easily reached within the soil or separated from it ( Richards, 1983 ; Gregory, 2006 ). Thus, measurements of root characteristics are tedious, time consuming, and labour intensive. It is not surprising, therefore, that root characteristics are rarely a feature in active breeding targets, but this may change with new screening techniques.

II.1.1.1 Methods for studying plant root systems For 50 years, numerous methods for studying and examining root systems have been investigated: such as excavation methods, monolith-sampling methods, core-sampling methods, profile wall methods, glass wall methods (rhizotron), container methods and radioactive tracers ( for details see review of Böhm, 1979 ). More recently, several methods have been developed and improved to enhance the efficiency of roots studies: such as core and ingrowth core sampling methods, rhizotron and minirhizotron observations, isotopic techniques, geophysical imaging ( e.g. magnetic resonance imaging, electrical impedance spectroscopy, electrical resistivity tomography, neutron radiography and tomography, X-ray computed tomography) ( for review see Johnson et al. , 2001 ; Hagrey, 2007 ; Gregory et al. , 2009 ; Luster et al. , 2009 ; Neumann et al. , 2009 ). Furthermore, high-throughput tools have been developed for phenotyping root traits from plants growing mainly in rhizotrons and on agarose plate media ( Bengough et al. , 2004 ; Armengaud et al. , 2009 ; French et al. , 2009 ; Yazdanbakhsh & Fisahn, 2009 ; Chen et al. , 2011 ). Coupled with electronic devices, digital cameras and image analysis software ( Basu et al. , 2007 ; Chavarría-Krauser et al. , 2008 ; Iyer- Pascuzzi et al. , 2010 ; Lobet et al. , 2011 ), these techniques allow fast and accurate measurements of multiple root system parameters, including main and lateral root length, elongation rate, root number and diameter, root angle and curvature ( see review of Zhu et al. , 2011 ) both in 2D ( Le Bot et al. , 2010 ) or 3D ( Fang et al. , 2009 ; Clark et al. , 2011 ). However, these tools have not yet been used with grapevine to the best of our knowledge. As a relatively recent technique, developmental modelling is now another essential tool for understanding and predicting root growth, architecture, water and nutrient uptake ( Pierret et al. , 2007 ; Danjon & Reubens, 2008 ; Javaux et al. , 2008 ). The grapevine root system, similar to other crops, has received only limited attention compared to vegetative and reproductive organs. Recently, the methodologies used to study grapevine roots and their environment have been reviewed ( de Herralde et al. , 2010 ). The methods used follow a historic evolution and mainly began with excavation techniques throughout the past century ( Harmon & Snyder, 1934 ; Nappi et al. , 1985 ; Hunter, 1998 ). More recently, the most commonly used method in viticulture was the profile wall (for review see Van Zyl, 1988a ; Smart et al. , 2006 ). Multiple studies have been done with this method (Branas & Vergnes, 1957 ; Perry et al. , 1983 ; Swanepoel & Southey, 1989 ; Morlat & Jacquet, 1993 ; Araujo et al. , 1995 ; Morlat, 2008 ). Other techniques have also been employed under vineyard or greenhouse conditions. Among these were soil core and ingrowth core sampling

- 24 - Chapter II. Literature review

(Kirchhof et al. , 1991 ; Soar & Loveys, 2007 ; Wheaton et al. , 2008 ), monolith methods (Bassoi et al. , 2003 ; Slavtcheva & Pourtchev 2007 ), rhizotron and minirhizotron observation techniques ( Freeman & Smart, 1976 ; Bauerle et al. , 2008a ; 2008b ; Lehnart et al. , 2008 ) and electric tomography ( Goulet & Barbeau, 2006 ). Cultivation techniques in hydroponic (Tattersall et al. , 2007 ) and aeroponic containers ( Grechi et al. , 2007 ), trenches ( Zapata et al. , 2001 ) or pots ( Iacono et al. , 1998 ; Lovisolo & Schubert, 2006 ; Galmés et al. , 2007 ; Lovisolo et al. , 2008b ) have also been widely used in order to easily harvest roots for morphological, physiological, metabolomic, transcriptomic and proteomic analyses. Recently, a continuous recirculating drip system (RDS) has been designed for long term hydroponic grapevine culture ( Wheatley et al. , 2009 ). Since the end of 1990s, new devices and techniques have become available for coarse root architecture measurements, digitising and modelling, but these have mainly be used on other crops ( Danjon & Reubens, 2008 ; Fang et al. , 2009 ; Clark et al. , 2011 ).

II.1.1.2 Parameters affecting grapevine root architecture Differences in the horizontal and vertical distribution of rooting depth may have a genotypic origin ( Pongrácz, 1983 ; Galet, 1988 ). Across rootstock genotypes, Guillon ( 1905 ) observed different geotropism angles of adventitious roots which resulted in variations of the vertical and horizontal depth distribution. In addition to the genotype, soil properties and vineyard management techniques may also strongly impact the development of the root system (reviewed in Branas & Vergnes, 1957 ; Richards, 1983 ; Van Zyl, 1988a ; Smart et al. , 2006 ). Under normal pedoclimatic conditions, the grapevine root system shows two types of distribution. The first includes horizontal roots in the top soil layer (10 to 60 cm), mainly fine roots which are devoted to water and nutrient uptake under favourable aeration and temperature parameters. The second distribution type involves the deep vertical roots in the subsoil, mainly woody roots, which are primarily involved in water uptake ( Ionescu et al. , 1978 ; Schreiner, 2005 ). The physical properties (structure and texture, soil strength, bulk density, etc.) and the chemical composition ( Ionescu, 1978 ; Pourtchev 1995 ) of the soil have strong effects on grapevine root and shoot development ( Wheaton et al. , 2008 ). Root growth and branching occurr in non-compact soil layers containing high levels of nutrient. Root elongation tends to follow zones of soil with low mechanical resistance, and coarse grapevine roots are regularly observed in earthworm channels and diverse soil fissures ( Seguin, 1972 ; McKenry, 1984 ). Kaziev and Teimurov ( 2008 ) have shown that the distribution of the grapevine roots under dry farming conditions is affected by the increase in soil density (>1.5 g m -3 , root growth forced along fissures with a limited development), the decrease in porosity (<40%, roots are deformed and grow between stones), the soil texture, the nutrient availability, and percentage of humus and carbohydrates stored. Grapevines can have very deep roots, large depth distribution ( reviewed in Smart et al. , 2006 ) and often reache 6 meters depth under favourable soil conditions ( Branas & Vergnes, 1957 ; Seguin, 1972 ). To the best of our knowledge, the greatest rooting depth reported in the literature are the root systems of 90-year-old vines (Bulgarian cultivars) which developed to a depth ranging from 20 meters ( V. vinifera own-

- 25 - Chapter II. Literature review rooted) to 32 meters (grafted onto 41B rootstock) (Pourtchev 2003 ). Carneiro and Menezes- Sequeira ( 1989 ) reported that rootstocks Malègue 44-53, Castel 196-17 and Richter 99 had deeper roots compared to Selection Oppenheim n°4 (SO4), Ruggeri 140 and Paulsen 1103. Perry et al. ( 1983 ) observed different root architecture among four grape species: V. champini cv. Dogridge had the most extensive root system (highest root number, root biomass and deepest roots) while M. rotundifolia cv. Noble had the smallest root system (lowest root number, root biomass and shallow roots), V. vinifera cv. Barbera and V. labrusca cv. Concord being intermediate. Vineyard management practices ( e.g. application of herbicides, use of permanent cover crop, mechanical tillage, type and dose of fertilization) have been shown to greatly influence the root system distribution ( Kirchhof et al. , 1991 ; Morlat & Jacquet, 1993 ; Morlat & Jacquet, 2003 ; Morlat, 2008 ). In the presence of cover crops in the interrow, a relatively rapid redistribution of the grape root system can occur which depends on the competition for water and on the available soil depth ( Celette et al. , 2005, 2008 ). Grapevine cultivation techniques (See Smart et al. , 2006 ), such as canopy management ( Hunter & Le Roux, 1992 ; Hunter et al. , 1995 ; Comas et al. , 2000 ; 2005 ), plant number per hectare ( Archer & Strauss, 1985 , 1989 ; Hunter, 1998 ) and the scion-rootstock combination ( Harmon & Snyder, 1934 ; Oslobeanu, 1978 ; Tandonnet et al. , 2010 ), also greatly influence root system development of different grape rootstocks. Aside from edaphic conditions which have the greatest effect on root architecture, the root density, whether expressed as root biomass or total root number and the number of different sizes of roots per unit of soil volume, appears to be predominantly controlled by the rootstock (Branas & Vergnes, 1957 ; Perry et al. , 1983 ; Southey & Archer, 1988 ; Swanepoel & Southey, 1989 ; Williams & Smith, 1991 ). Morano and Kliewer ( 1994 ) found differences between rootstocks in both the total root and fine root numbers. In another experiment, Swanepoel and Southey ( 1989 ) observed that rootstocks exerted a substantial influence on both root density, number of fine roots and root distribution in a soil with favourable conditions ( see review of Southey & Archer, 1988 ). Smart et al. ( 2006 ) suggested that root characteristics other than root distribution may need to be considered in order to explain scion-rootstock interactions such as drought adaptation. In this context, the genetic influence of rootstocks on root density appears to be very interesting, since fine roots have a major importance in water uptake and play important roles in drought tolerance and recovery ( McCully, 1999 ).

II.1.1.3 Developmental plasticity of roots under drought Drought is not a single, simple stress but consist of multiple interactions with other stresses such as heat, soil strength, low nutrient availability and even hypoxia. The physical effects of drought on the root environments ( e.g. mechanical impedance) and adaptive responses of plants have been reviewed recently ( Bengough et al. , 2006 ; Whitmore & Whalley, 2009 ). Shishkova et al. ( 2008 ) defined two types of determinate root growth: constitutive, where determinacy is a natural part of root development, and non constitutive, where determinacy is usually induced by an environmental factor like water-deficit (root plasticity).

- 26 - Chapter II. Literature review

The mechanisms controlling root plasticity are not clear but they may be related to specific hormonal signals or to changes in carbohydrate allocation between root and shoot or within the root system ( Huang & Eissenstat, 2000 ). Water uptake capacity related to drought tolerance depends on root morphological characteristics ( e.g. root length density and distribution) and physiological properties ( e.g. viability, osmotic adjustment and hydraulic conductivity) ( Huang, 2000 ; Huang & Eissenstat, 2000 ).

II.1.1.3.1 General relationships between root growth and water status In dry conditions, water uptake by roots can result in soil drying below the rooting front, which restricts the maximum rooting depth due to the increase of mechanical impedance (Bengough, 1997 ). The growth of the root system and the elongation of individual roots are often limited by soil strength. Penetrometer pressures of 2-2.5 MPa or more are sufficient to impede root elongation significantly, but this can also depend on the stress history of the root as well as on the nature of the impedance ( Bengough et al. , 2006 ). High resolution measurements show a characteristic thigmotropic growth response when a root encounters an impenetrable obstacle. The mechanical stimuli induce an elevation of cytosolic Ca 2+ with a stimulus-specific signature, in addition to pH changes and production of reactive oxygen species (ROS) in the root ( Monshausen et al. , 2009 ; Monshausen & Gilroy, 2009 ). Leaf and root are competing for assimilates produced by the leaves and for the minerals and water taken up by roots. The development of both organs is coordinated and their relative size to each other varies dynamically in response to environmental cues ( Hsiao & Xu, 2000 ). Plant roots exhibit a high degree of physiological and morphological plasticity and respond to a wide variety of environmental stimuli (spatial and temporal fluctuation of resources in the soil) in ways that are presumably adaptive, especially in stressful environments ( see reviews of Huang, 2000 ; Huang & Eissenstat, 2000 ; Hodge et al. , 2009 ; Walter et al. , 2009 ). Considering the importance of the genotype, Ho et al. ( 2005 ) claim that plants which increase the density of their roots in the top soil are better adapted to low nutrient environments, while plants which allocate their root biomass to deeper soil layers better cope with water stress. They argue further that genotypes with a root architecture that allows for both shallow and deep root localization are highly plastic and better adapted to multiple resource limitations. Under water deficit conditions, some types of roots have the ability to maintain elongation at low water potentials (-1.6 MPa) that would be completely inhibitory to shoot growth. The physiological mechanisms have been studied extensively with protocols that allow precise and reproducible imposition of water deficits ( Sharp et al. , 2004 ; Ober & Sharp, 2007 ; Yamaguchi & Sharp, 2010 ). This research has taken advantage of a kinematic approach, that is, the study of spatial and temporal patterns of cell expansion within the tissue elongation zone ( Erickson & Silk, 1980 ; Walter et al. , 2009 ). The kinematic approach greatly facilitated the discovery of mechanisms involved in the control of root growth under water stress. The root elongation rates are fully maintained in the apical few millimeters but progressively decrease as cells are displaced further from the root apex ( Sharp & Davies, 1979 , 1985 ; Westgate & Boyer, 1985 ; Sharp et al. , 1988 ; Liang et al. , 1997 ). Figure II.2 illustrates the maintenance of elongation rate towards the root apex (region 1) in two different

- 27 - Chapter II. Literature review species under water-deficit stress. Additionally, two other contiguous regions (region 2 and 3), with distinct elongation characteristics depending on root water status, were identified (Figure II.2).

Figure II.2: Displacement velocity as a function of distance from the root cap junction of primary roots of well-watered (WW) and water-stressed (WS) maize seedlings (cv. FR697). The inset shows the profile of displacement velocity for WW and WS soybean primary roots (cv. Magellan). Relative elongation rates (h %1) are obtained from the derivative of velocity with respect to position. Regions 1 to 3, as described in the text, are indicated. The green arrows indicate the comparisons between treatments within regions 1 and 2 and the comparison of WS region 2 with WW region 3 ( from Yamaguchi & Sharp, 2010 ).

The mechanisms underlying sustained root growth under water stress include osmotic adjustment to allow partial turgor recovery and reestablishment of water potential gradients for water uptake and an increase in the loosening ability of the cell wall ( Hsiao & Xu, 2000 ). The root tip is the first organ to detect water stress. The transmission of some signal from the root tip to the decelerating region appears needed for the suppression of cell elongation in the decelerating region, without any effects on cell turgor ( Shimazaki et al. , 2005 ). This differential growth sensitivity between the growing regions of the root provides important clues to investigate of the mechanisms of root adaptation to water-deficit ( Sharp et al. , 2004 ; Ober & Sharp, 2007 ). These approaches will be detailed below (paragraph II.1.2).

II.1.1.3.2 Grapevine rootstocks plasticity to water status Grapevine root growth plasticity in response to soil moisture plays an important role to determine when and where roots capture resources ( Comas et al. , 2010 ). Both grafted and ungrafted grapevine roots strongly respond to soil moisture availability and irrigation techniques ( Nagarajah, 1987 ; Stevens & Douglas, 1994 ; Araujo et al. , 1995 ; Bassoi et al. ,

- 28 - Chapter II. Literature review

2003 ; Comas et al. , 2005 ). Grapevine (Cabernet-Sauvignon grafted onto Ramsey rootstock) established under sprinkler irrigation and then converted to drip irrigation had a significantly larger root system (compare to sprinkler throughout) and were more robust to cope with water deficit ( Soar & Loveys, 2007 ). Root growth of ungrafted grape species ( Vitis labruscana cv. Concord) can be greatly reduced ( Comas et al. , 2005 ), while in other ungrafted ( Gómez-del- Campo et al. , 2005 ) or grafted-grape species ( Van Zyl, 1988b ; Bauerle et al. , 2008b ), vines can maintain new root production during moderate to severe water-deficit.

Fall Winter Spring Summer b 20 ) -2 15 a a 10 a a (mm cm 5 a b a Root production Root 0 101-14 1103P 101-14 1103P 101-14 1103P 101-14 1103P New Root Production per Season Figure II.3: Seasonal root production (± SE) for root systems of V. berlandieri ×V. rupestris cv. 1103P and V. riparia ×V. rupestris cv. 101-14MGt (season×root system interaction: P=0.002). Data represent total root length produced per cm 2 of observational window over three month periods for the years, 2003-2005. Each season corresponded to the following months: Autumn, September-November (significance of difference between 1103P and 101-14Mgt, P=0.328); winter, December-February (P=0.009), spring, March-May (P=0.230), and summer, June-August (P=0.032). Significant differences are indicated by different letters ( redraw from Bauerle et al. , 2008b ).

Bauerle et al. ( 2008b ) studied root dynamics of Vitis vinifera cv. Merlot grafted onto two rootstocks (1103P and 101-14 MGt) that confer respectively high and low vigour to the shoot and high and low tolerance to water stress. The root system of 1103P seems to produce new roots preferentially during the summer dry season while the 101-14 MGt root system produce them during the cool wet winter season (Figure II.3). The 1103P exhibited a greater plasticity than 101-14 MGt, indicated by a greater preferential root production during the summer (Figure II.3). Interestingly, the two rootstocks thus exhibited different strategies of root growth in the deep soil layers. During the summer, the drought-tolerant rootstock grew a greater percentage of its total roots in deeper soil layers compare to the drought-sensitive rootstock. Bauerle et al. ( 2008b ) concluded that 1103P has a greater ability to proliferate roots during localized wetting events with a better chance of competing for ephemeral resources in a patchy environment, while 101-14 MGt seems to use a long-term strategy consisting in the growth of a large root system. These authors did not observe differences in root survival in response to soil moisture deficit despite differences in root diameter and rooting depth. Grape roots in deeper soil layers were found to have a longer lifespan than those in shallow soil. Anderson et al. ( 2003b ) pointed out that root lifespan of V. labruscana Bailey cv. Concord was significantly affected by soil depth (each cm soil depth reduces the risk of mortality by 0.3-0.7%), root diameter (coarse roots with a diameter greater than 0.4 mm had significantly reduced risk of mortality) and the phenological stage at the time of root birth (roots born at least 30 days after bloom and before harvest had significantly lower risk of

- 29 - Chapter II. Literature review mortality). For the 3309C rootstock, Volder et al. ( 2005 ) showed that high level of nutrient uptake and metabolic activity were associated with very young roots. In another experiment, Bauerle et al. ( 2008a ) showed that V. vinifera cv. Merlot grafted onto 101-14 MGt may extend the lifespan of roots in dry soil through redistribution of water from roots in wet soil to those in dry soil during periods of minimal transpiration such as night time.

II.1.1.3.3 Interactions with symbiotic organisms Root colonization with rhizobacteria may improve the drought tolerance of plants by producing glycine betaine, by favoring the production of auxin and ethylene, by modifying gene expression and the level of amino acids, and by increasing root growth and root cell membrane elasticity ( see review of Dimkpa et al. , 2009 ). Grapevine rootstocks ( Vitis spp.) and own-rooted V. vinifera L. are often colonized by Vesicular-Arbuscular-Mycorrhizal fungi ( Karagiannidis et al., 1997 ; Linderman & Davis, 2001 ; Schreiner, 2003 ; Van Rooyen et al., 2004 ) and this colonization may improve root growth, whole plant water status and indirectly the root lifespan ( Eissenstat et al., 2000 ; Augé, 2001 ; Schreiner et al., 2007 ). Among rootstocks, the presence of mycorrhizal fungus in the root system can delay drought symptoms, increase cytokinin concentration and improve plant water status ( Nikolaou et al. , 2003a ; 2003b ). The different responses to soil moisture between grape species and rootstocks may be due to different physiological and molecular strategies for coping with drought. Water-deficit tolerance is affected by the rootstock genotypes (Table II.1), but the physiological and molecular processes underlying grape root plasticity are poorly documented. The next paragraph will review studies related to water-deficit and root behaviours for grapevine, and other plant species.

II.1.2 Drought-related physiological mechanisms in roots Physiological research pertaining with the responses of roots to water stress has focussed on different areas including i) stress perception; ii ) the role of hormones; iii ) osmotic adjustment; iv ) anatomical changes and v) molecular mechanisms.

II.1.2.1 Water-deficit perception The root system is the primary site of perception of water-deficit and other stresses (Osmont et al. , 2007 ). The root tip meristematic activities apparently sense and signal information regarding soil conditions like moisture ( Aiken & Smucker, 1996 ; Nibau et al. , 2008 ). Plants may use this information to redirect root growth (tropisms) and reprogram the architecture of the root system ( Monshausen & Gilroy, 2009 ) as shown in low-phosphate media ( Svistoonoff et al. , 2007 ).

II.1.2.2 Roles of plant hormones

II.1.2.2.1 Some generalities Hormones play a central role in plant adaptation to the environment by mediating growth, development, nutrient allocation and source/sink relationships. Almost all known plant

- 30 - Chapter II. Literature review hormones affect the development of the root system, including auxin ( Fu & Harberd, 2003 ; Fukaki et al. , 2007 ; Laskowski et al. , 2008 ), cytokinins ( Bishopp et al. , 2011 ; Werner et al. , 2011 ), gibberellins ( Ubeda-Tomas et al. , 2008 ; Gou et al. , 2010 ), ethylene ( Ortega-Martinez et al. , 2007 ; Prasad et al. , 2010 ), brassinosteroids ( Müssig et al. , 2003 ; Mouchel et al. , 2006 ), strigolactones ( Koltai, 2011 ) and ABA ( Sharp & LeNoble, 2002 ; De Smet et al. , 2006 ). Although, ABA is the most studied stress-responsive hormone, hormonal cross-talk results in synergistic or antagonistic interactions between hormones that regulate plant responses to abiotic stress ( Peleg & Blumwald, 2011 ). This hormone-regulation of root developmental responses to both biotic and abiotic stresses downstream of the respective stimuli has been extensively reviewed ( Tanimoto, 2005 ; Fukaki & Tasaka, 2009 ; Péret et al. , 2009, 2009b ). After perception of water stress by the meristematic cells of the root, both local and systemic signals allow plants to develop an appropriate developmental strategy, and small noncoding RNAs seem to play a crucial role in posttranscriptional regulation of root development responses to environmental constraints ( Jovanovic et al. , 2007 ). The auxin signalling pathways appear to be particularly important in mediating environmental responses of root system architecture ( Osmont et al. , 2007 ; Wang et al. , 2009 ). Modelling of auxin afflux facilitators (PINs) showed that auxins transport from the shoot in Arabidopsis may be sufficient to generate a gradient that explain root growth maintenance in absence of auxin biosynthesis in the roots ( Grieneisen et al. , 2007 ). However, root development does not only depend on the auxin gradient, but also on four PLETHORA ( PLT ) homologues encoding AP 2-domain transcription factors ( Galinha et al. , 2007 ) and PLT activities are required for root cell specification and for maintenance of the root meristem (Aida et al. , 2004 ). In Arabidopsis , the repressing effects of DELLA proteins on root growth, mediated at least in part through a reduction in the levels of bioactive gibberellins, is beneficial for plant survival under environmental stress ( Achard et al. , 2006 ). These authors showed that wild type seedlings growing on saline substrate increased ethylene production and decreased growth, unlike mutants deficient in ABA signalling ( abi1-1 ) or lacking four of the five DELLA proteins. Recently, Tran et al. ( 2007 ) showed that homologues of the cytokinin receptor histidine kinases AHK1 act as a positive regulator of drought response through both ABA-dependent and ABA-independent signalling pathways.

II.1.2.2.2 ABA sustain the root growth maintenance Phytohormones, mainly ABA, ethylene and its precursor ACC (aminocyclopropane-1- carboxylic acid), are thought to play a role in root-to-shoot signalling in stressed plants. The response of xylem sap pH and ABA to drought stress is not consistent for all species or within the same species in different experiments. This may be explained by the fact that absolute concentration within the xylem sap varies with the flux rates of whole plant transpiration (Jackson, 1997 ). Under water-deficit, the root growth rate decreases and the roots synthesize and accumulate ABA. In a serie of classic hormone response experiments (using an inhibitor of ABA synthesis or mutants), it was demonstrated that primary root ABA accumulation is required for the maintenance of growth under water-deficit. Drought-induced accumulation of ABA towards the root apex played a crucial role for the maintenance of elongation in the root

- 31 - Chapter II. Literature review apex (region 1, Figure II.2) ( Ober & Sharp, 2007 ; Yamaguchi & Sharp, 2010 ). One part of this role of endogenous ABA was to limit ethylene production. As a result of this antagonism, ABA accumulation during water stress may often function to maintain rather than inhibit root growth, particularly towards the root apex ( Saab et al. , 1990, 1992 ; Spollen et al. , 2000 ; Sharp, 2002 ; Sharp & LeNoble, 2002 ; Sharp et al. , 2004 ). In addition, at low water potential, ABA plays a role in regulating the steady-state membrane potential in root cell elongation zone, and control homeostatic set-points for ion transport processes that shift in response to environmental changes ( Ober & Sharp, 2003 ). ABA also stimulates the elongation of the main root and the emergence of lateral roots in response to drought, and thus favouring developmental plasticity to changes in the soil environment ( De Smet et al. , 2006 ). Therefore, ABA has a significant impact on the final size and architecture of the root system. PERK4 , a gene encoding a member of A. thaliana proline- rich extension-like receptor kinase4, seems to play an important role at an early stage of ABA signalling, conducting to inhibition of cell elongation, and its effects are mediated by Ca 2+ (Bai et al. , 2009 ). The R2R3-type MYB transcription factor MYB96 regulates the ABA-auxin cross-talk during water stress response ( Seo et al. , 2009 ). ABA modulates the auxin signalling pathway by inducing GH3 genes, and ultimately reduces lateral root development, which provides an adaptive strategy for Arabidopsis under water stress ( Seo et al. , 2009 ). An increasing number of mutant studies in Arabidopsis have led to the identification of several genes that regulate lateral root formation and/or coordinate this process in response to environmental cues. Drought stress severely represses the emergence of lateral roots, although lateral root initiation is largely unaffected ( van der Weele et al. , 2000 ; Deak & Malamy, 2005 ; Xiong et al. , 2006 ). The Arabidopsis dig ( drought inhibition of lateral root ) mutants differentially respond to drought or ABA, indicating a close relationships between water stress, lateral root growth and whole-plant drought tolerance ( Xiong et al. , 2006 ).

II.1.2.3 Accumulation of compatible solutes and osmotic adjustments Water stress induces the accumulation of compatible solutes in the roots. This contributes to maintaining a low water potential favouring water uptake and the turgor pressure necessary for root growth. Although organic compounds ( i.e. sugars and amino acids) are the major constituents involved in the osmo-regulation of plant cells during water stress ( Morgan, 1984 ), inorganic ions may also play a role ( Roberts & Snowman, 2000 ). Compatible solutes are non- toxic molecules such as sugars, amino acids, polyalcohols or amines that accumulate under drought stress and function as osmoprotective compounds to maintain cell turgor and to stabilize cell proteins and structures ( Bartels & Sunkar, 2005 ; Seki et al. , 2007 ; Shulaev et al. , 2008 ; Szabados et al. , 2011 ). Water stress-induced osmotic adjustment has been reported in grapevine leaves ( Düring, 1984 ; Rodrigues et al. , 1993 ; Schultz & Matthews, 1993a ). Unfortunately, there is minimal evidence for osmo-regulation in grapevine roots in the literature. Moderate soil dehydration leads to a small but significant decrease of osmo-regulation in the apical 3 mm, but not in the other root parts of Kobber 5BB rootstock cultivar. By contrast, severe and rapid drying cycles caused a significant decline in osmotic potential in all root types ( Düring & Dry, 1995 ).

- 32 - Chapter II. Literature review

Grapevine roots were found to undergo osmotic adjustment by accumulating sugars and amino acids ( Düring & Dry, 1995 ). Recently, Rogiers et al. ( 2011 ) showed that own-rooted V. vinifera cvs. Grenache and Semillon accumulate sucrose in their roots grown under water- deficit. Root sucrose concentrations were inversely correlated to leaf and root water status and positively correlated with leaf xylem sap ABA concentration ( Rogiers et al. , 2011 ).

II.1.2.3.1 Sugars and polyalcohols A strong correlation between the accumulations of several sugars (sucrose, Raffinose Family-type Oligosaccharides [RFOs], galactinol and trehalose) and polyalcohols (mannitol, pinitol and inositol) were reported to play a role in drought stress tolerance ( Bartels & Sunkar, 2005 ; Szabados et al. , 2011 ). At low water potential (-1.6 MPa), maize roots accumulate a significant amount of hexoses and potassium for osmotic adjustment in the basal region of the growing zone, but these solutes account for little of the osmotic potential variation in the apical 2-3 mm ( Sharp et al. , 1990 ). The roles of sugars as sensors and signalling molecules in growth, development and environmental stresses are well recognised ( Rolland et al. , 2006 ; Smeekens et al. , 2010 ). Recently, sucrose was identified as an unexpected regulator of lateral root formation via root- to-shoot and shoot-to-root cross-talk during osmotic stress imposed on agar plates (MacGregor et al. , 2008 ). RFO sugars, such as raffinose, stachyose and galactinol play a role in the acquisition of drought tolerance. Overexpression of galactinol synthase in transgenic A. thaliana ( AtGolS2 ) induces accumulation of galactinol and raffinose and improved drought tolerance ( Taji et al. , 2002 ). Urano et al. ( 2009 ) showed that the accumulation of amino acids depends on ABA production, but the level of raffinose was independently regulated by ABA under dehydration stress. The disaccharide trehalose is also an important signalling molecule ( Paul et al. , 2008 ). Trehalose accumulation was observed in many plants under drought ( Iordachescu & Imai, 2008 ). The importance of trehalose in modulating plant stress responses was demonstrated by transgenesis. Trehalose has the ability to stabilise membranes and protect proteins in desiccated tissues ( Szabados et al. , 2011 ). However, its function is controversial: some trehalose metabolism mutants exhibit pleiotropic growth alterations, dwarfism and abnormal root structure ( Fernandez et al. , 2010 ). Recently, trehalose-6-phosphate was shown to regulate SnRK1 (Sucrose non-fermenting Related Kinase 1) activity ( Zhang et al. , 2009 ), which is a central regulator of sugar and ABA signalling in plants ( Jossier et al. , 2009 ). Accumulation of polyalcohols has been observed in several water-stressed plants (reviewed in Szabados et al. , 2011 ) and might increase drought tolerance by scavenging ROS and/or by stabilization of macromolecular structures ( Seki et al. , 2007 ).

II.1.2.3.2 Amino acids: glutamate and proline Water stress-induced proline accumulation as a compatible solute in osmotic adjustment has been associated with drought adaptation ( Szabados & Savouré, 2010 ). Proline is thought to play multifunctional roles, such as mediator of osmotic adjustment, stabilizer of subcellular structures, scavenger of ROS, activator of detoxification pathways and stress-related signals (Bartels & Sunkar, 2005 ; Seki et al. , 2007 ; Szabados et al. , 2011 ). Recently, using

- 33 - Chapter II. Literature review

Arabidopsis proline metabolism mutants, Sharma et al. ( 2011 ) showed that proline supply from the shoot and its catabolism in the root are essential for maintained growth under water- deficit. However, using a mutant deficient in proline dehydrogenase, Nanjo et al. ( 2003 ) demonstrated the toxicity of an excess of proline on plant growth.

Figure II.4: Schematic representation of predicted functions and interactions of water stress-responsive proteins/genes and associated metabolites in the cell wall and cytosol in region 1 (Figure II.2) of the elongation z one of the maize primary root. Major differences in response between regions 1 and 2 are indicated by shaded text. Non-italicized text and solid connecting arrows indicate responses that have been demonstrated in published studies; italicized text and dashed connecting arrows indicate responses that are hypothesized to occur but have not yet been demonstrated. Up-regulation under water stress is indicated by the short upward arrows; for ABA, proline and oxalate oxidase (OxO), the double arrows indicate that accumulation or activity in region 1 was greater than in region 2. *Flavonoid accumulation has not been determined in water-stressed (WS) maize roots; however, isoflavonoids were shown to accumulate in region 1 of the soybean primary root under water stress. The inset shows increased apoplastic reactive oxygen species (ROS) in the epidermis of region 1 in WS compared to well-watered (WW) roots, as indicated by confocal microscopy of roots stained with H 2DCF (2 a,7 a- dichlorodihydrofluorescein, green fluorescence), a membrane-impermeable ROS indicator. The right- hand diagram illustrates a transverse view of the root surface and focal planes. The increase in apoplastic ROS in WS roots is consistent with the increases in abundance of superoxide dismutase (SOD) and oxalate oxidase, which contribute to H 2O2 production, and peroxidases (POX), which can also contribute to ROS • production including the generation of hydroxyl radicals ( OH) from H 2O2 in the presence of superoxide •% (O 2 ) and/or reductant (e.g. NADH) ( from Yamaguchi & Sharp, 2010 ).

- 34 - Chapter II. Literature review

Proline deposition plays a major role in osmotic adjustment and contributed to maintain elongation of the root apex (Figure II.4) under water-deficit ( Voetberg & Sharp, 1991 ). At low water potential, high rates of proline deposition in the growing region of maize primary root require an increase of ABA ( Ober & Sharp, 1994 ). The proline precursors, glutamate and ornithine are not responsible for proline accumulation in the root growth zone at low water potential. Instead, an increase of proline transport to the root tip is the major source of proline accumulation in the root elongation zone ( Verslues & Sharp, 1999 ). In maize roots subjected to water stress, proline accumulation in the root tip is translocated from the endosperm. The accumulation is controlled in a precise manner proportionally to water potential ( Raymond & Smirnoff, 2002 ). By using a system of polyethylene glycol- infused agar plates and low-water potential response Arabidopsis mutants, LWR1 and LWR2 , were shown to mediated multiple aspects of cellular osmo-regulation and osmotic adjustments, through modification of proline accumulation, total solute and ABA contents ( Verslues & Bray, 2004 ). Sugar sensing may directly influence the efficiency of ABA-induced proline accumulation in water-stressed plants ( Verslues & Bray, 2006 ). Deak and Malamy ( 2005 ) pointed out that ABA and drought have similar and probably synergistic effects on lateral root growth, and they showed that the LATERAL ROOT DEVELOPMENT2 ( LRD2 ) gene is involved in the osmotic repression of lateral root formation. Glutamate is also important in water-stressed roots. Glutamate may be used for the synthesis of "-aminobutyrate acid (GABA) which is a non-protein amino acid. Glutamate decarboxylase which catalyses the conversion of glutamate to GABA is overexpressed in region 2 (Figure II.2) of water-stressed roots ( Yamaguchi & Sharp, 2010 ). GABA biosynthesis and GABA shunt may have important roles to prevent oxidative stress and participate in the regulation of cytosolic pH in water-stressed roots ( Bouché & Fromm, 2004 ).

II.1.2.3.3 Amines: polyamines and glycine betaine Amines such as polyamines (small polycations) and glycine betaine (an amphoteric quaternary amine) are also involved in drought stress responses ( Groppa & Benavides, 2008 ; Alcázar et al. , 2010a ; Szabados et al. , 2011 ). Genetic manipulation by overexpression/down-regulation of genes involved in the polyamine biosynthetic pathways enhanced drought tolerance in rice, thanks to the accumulation of polyamines (PAs), especially putrescine ( Capell et al. , 2004 ). Plants overexpressing the Arginine decarboxylase 2 ( ADC2 ) gene accumulated more putrescine in response to drought than wild-type ( Alcázar et al. , 2010b ). The roles of PAs in grape development and stress responses was recently reviewed ( Paschalidis et al. , 2009 ). Under osmotic stress, homeostasis of endogenous PAs differs between drought-sensitive and drought-tolerant V. vinifera genotypes. The intrinsic ABA signal induced an up-regulation of

PAs metabolism, which in turn generated ROS (H 2O2) and then triggered stress signalling (Toumi et al. , 2010 ). Grape rootstocks under salinity stress may respond by increasing PAs concentration ( Upreti & Murti, 2010 ). The function of glycine betaine (GB) for stress tolerance has been investigated in transgenic plants engineered to produce GB ( Sakamoto & Murata, 2002 ; Chen & Murata,

- 35 - Chapter II. Literature review

2008 ). Glycine betaine may play a role in stress tolerance by stabilizing the structure and activity of macromolecules as well as by maintaining the water balance between plant cell and environment. Under drought, GB had a strong protective effect on reproductive organs such as inflorescence apices and flowers ( Chen & Murata, 2008 ). In transgenic maize transformed with the betA gene (encoding choline dehydrogenase, a key enzyme in GB biosynthesis), the plants accumulating higher levels of GB, were more drought tolerant ( Quan et al. , 2004 ), which may have been due to a higher root biomass compared to wild-type ( Chen & Murata, 2008 ).

II.1.2.4 Morphological and cell wall modifications The structural and functional modifications induced by drought in root systems and the effects of environmental stress on the morphogenetic program of growth and differentiation within the whole plant have been described by Vartanian ( 1981 ). The roots of Arabidopsis respond to water depletion by developing specialized lateral structures characterized by short, tuberized hairless roots. These roots can withstand an extended drought period, but the root apical meristem can continue to function and generates a new functional root system upon rehydratation ( Vartanian et al. , 1994 ). Water stress induces ABA-mediated morphological changes of the root system ( Schnall & Quatrano, 1992 ). Exogenous application of ABA to well-watered rice ( Oryza sativa L.) leads to swelling, root hair formation and lateral root initiation ( Chen et al. , 2006 ). In that experiment, several ABA-induced or -repressed proteins were identified and the response of root morphogenesis changes to external ABA was Ca 2+ - dependent. Under severe water stress (-1.6 MPa), the roots of maize had lower diameters due to restriction of tangential and radial expansion rate of cells in both the stele and cortex ( Liang et al. , 1997 ), which was not related with the degree of alignment of either cortical microtubules or cellulose microfibrils ( Baskin et al. , 1999 ). Due to low turgor pressure, water-deficit stress markedly reduces cell expansion and cell growth ( Shao et al. , 2008 ). The effect of water stress on wall structure and properties have been intensively studied in maize roots ( Moore et al. , 2008 ). In the apical root (Figure II.4), the maintenance of elongation suggestes that longitudinal cell wall extensibility increases under water-deficit ( Yamaguchi & Sharp, 2010 ). In stressed-roots of maize, the activities of cell wall proteins, expansin and xyloglucan endotransglycosylase/hydrolase (XTH), were increased specifically in the apical region compared to well-watered roots (Figure II.4), suggesting that they may be necessary to maintain root elongation under drought ( Wu et al. , 1996, 2000 ; Wu & Cosgrove, 2000 ). This response seemed to depend on ABA accumulation (Wu et al. , 1994 ). Under water stress, the expression of expansin genes ( Exp1 , Exp5 and ExpB8 ) has been shown to be up-regulated in the apical region (Figure II.4) and down- regulated in the basal region of the elongation zone of roots. They were correlated with the increase and decrease of cell wall extensibility in these root regions, respectively. Thus, they help root cell elongation maintenance at low turgor pressure ( Wu et al. , 2001 ). The complex anatomical structure of roots results in a complex pattern of water flow across the different tissues via apoplastic, symplastic and transcellular pathways ( Steudle & Peterson, 1998 ; Steudle, 2000b ). Water stress-induced changes in the anatomy of root tissues,

- 36 - Chapter II. Literature review like the development of apoplastic barriers for water and ion flow or the decrease of root hydraulic conductivity through cell membranes (closure of water channels), play a role in minimizing water losses. During water stress, the formation of suberin lamellae in the endodermis and the exodermis (cell layer beneath the epidermis that contains Casparian bands) induces localized zones of high resistances to water flow in the root apoplast ( see review of Steudle, 2000b ). Increased suberization in roots as a consequence of water stress has been observed in many plants ( Enstone et al. , 2002 ). In non-grafted V. vinifera cvs. Chardonnay and Grenache, water stress may induce the differentiation of an exodermis and/or an endodermis with a lower apoplastic path conductance (water and soluted flow around protoplasts) due to a more abundant suberin deposits in root cell walls than in well-watered roots ( Vandeleur et al. , 2009 ). In non-grafted V. vinifera cv. Shiraz, water-stressed roots had more mature xylem elements closer to the root tip ( Mapfumo et al. , 1994 ). In the region between the tip and about 8 cm proximal to the root tip, there were no differences between the main root diameter for well-watered and stressed plants, while the diameter of segments farther from the tip was smaller in plants submitted to water stress vs. control plants. The percentage of fine lateral roots ( i.e. < 0.05 cm) was significantly higher in stressed plants. However, the length distribution of first order laterals did not differ between treatments, suggesting that in both main and lateral roots, water stress has a greater effect on diameter than on length ( Mapfumo et al. , 1994 ). Water stress seemed to decrease the proportion of roots occupied by the cortex whereas the extent of suberization increased with increasing distance from the root tip ( Mapfumo & Aspinall, 1994 ).

II.1.2.5 Protection from oxidative damage The production of Reactive Oxygen Species (ROS) such as superoxide, hydrogen peroxide and hydroxyl radicals often increases during drought ( Apel & Hirt, 2004 ). The formation of these toxic by-products of stress metabolism may cause oxidative damage ( Møller et al. , 2007 ). However, ROS also play a role in signal transduction during water-deficit (reviewed in Miller et al. , 2010 ; Demiral et al. , 2011 ). Protection from oxidative damage and the roles of apoplastic ROS in water-stressed roots were reviewed recently ( Yamaguchi & Sharp, 2010 ). These authors propose a complex and coordinated regulation, including co-regulation of ROS-scavenging proteins, proteinase inhibitors and antioxidant metabolites, to protect root elongation from oxidative damage under water-deficit. Moreover, they show that in water-stressed roots, apoplastic ROS play a role both in the maintenance of elongation in the apex and the inhibition of elongation in the basal region (Figure II.4).

II.1.2.6 Hydrotropism The ability of roots to sense and grow towards soil moisture gradients, i.e. hydrotropism, seems to be common among higher plant species (although not demonstrated yet for the grapevine root). It has been considered to play an important role both in drought avoidance and the formation of root architecture (see reviews of Takahashi, 1997 ; Eapen et al. , 2005 ; Miyazawa et al. , 2009b ; Miyazawa et al. , 2011 ). However, Cole and Mahall ( 2006 ) did not

- 37 - Chapter II. Literature review observe clear evidence for hydrotropic root behaviour under field conditions with a very strong moisture gradient. Hydrotropism is detected in roots via the root cap which perceives a water potential gradient ~ 0.5 MPa ( Jaffe et al. , 1985 ; Takahashi & Scott, 1993 ; Takahashi, 1994 ; Takano et al. , 1995 ; Hirasawa et al. , 1997 ; Eapen et al. , 2003 ). Both the root cap and the elongation zone have indispensable and functionally distinct roles in hydrotropism (Miyazawa et al. , 2008 ). The apoplastic-Ca 2+ and its influx through the plasmamembrane (Takano et al. , 1997 ; Eapen et al. , 2005 ), several hormones ( see review of Takahashi et al. , 2009 ) such as ABA ( Takahashi et al. , 2002 ; Eapen et al. , 2003 ; Ponce et al. , 2008 ) and auxin (Mizuno et al. , 2002 ; Takahashi et al. , 2002 ; Kaneyasu et al. , 2007 ), and the degradation of amyloplast starch in the root columella cells ( Takahashi et al. , 2003 ; Ponce et al. , 2008 ) are involved in hydrotropism signalling. Water stress seems to suppress gravitropism to promote root hydrotropism ( Takahashi et al. , 2003 ; Eapen et al. , 2005 ; Takahashi et al. , 2009 ). The understanding of hydrotropism was improved through mutants screening, such as nhr1 for no hydrotropic response1 ( Eapen et al. , 2003 ), MIZ1 ( Kobayashi et al. , 2007 ) and MIZ2 (Miyazawa et al. , 2009a ) for mizu-kussei1 and 2, respectively. MIZ1 was the first gene identified in Arabidopsis roots involved in hydrotropism independently of gravitropism. MIZ1 mRNA was up-regulated after exposure to osmotic stress and could contribute to drought avoidance by sensing moisture gradients in the early phase of hydrotropic response in roots (Kobayashi et al. , 2007 ). Like MIZ1, mizu-kussei2 , a unique allele of GNOM (encoding guanine-nucleotide exchange factor for ADP-ribosylation factor-type G proteins) has defects specifically in hydrotropic response, without affecting its response to gravistimulation and root elongation. The GNOM-mediated vesicular trafficking might play an essential role in root hydrotropism ( Miyazawa et al. , 2009a ). However, in a hydrotropic experiment, MIZ1 and MIZ2 were surprisingly not listed as hydrotropism-responsive genes ( Moriwaki et al. , 2010 ). Taniguchi et al. ( 2010 ) have shown that Arabidopsis phospholipase D#2 gene ( PLD #2) responds to drought via ABA signalling in the root cap. The inhibition of root hydrotropism in its mutant strongly suggests that drought-induced expression of PLD #2 accelerates root hydrotropism through the suppression of root gravitropism ( Taniguchi et al. , 2010 ). Recently, a transcriptomic analysis during the hydrotropic response showed that the hydrotropism- responsive genes are similar to those triggered by ABA, ROS or drought, but distinct from gravitropism-responsive genes, suggesting that ABA and water depletion are important steps in the signal transduction involved in root hydrotropism ( Moriwaki et al. , 2010 ).

II.1.2.7 Molecular pathways The reconfiguration of the transcriptome under water-deficit has been observed in various species, such as A. thaliana ( Seki et al. , 2002 ; Bray, 2004 ; Wilkins et al. , 2010 ), barley ( Guo et al. , 2009 ), rice ( Degenkolbe et al. , 2009 ) and poplar ( Street et al. , 2006 ; Bogeat-Triboulot et al. , 2007 ; Wilkins et al. , 2009 ). This progress at the molecular level, including gene expression, transcriptional regulation and signal transduction, has increased the understanding of the regulatory networks which control the water deficit responses ( Shinozaki et al. , 2003 ; Yamaguchi-Shinozaki & Shinozaki, 2006 ; Seki et al. , 2007 ; Shinozaki & Yamaguchi- Shinozaki, 2007 ; Kantar et al. , 2011 ). The regulatory networks include stress sensors,

- 38 - Chapter II. Literature review signalling pathways comprising a network of protein-protein reactions, transcription factors and promoters, and finally the output proteins or metabolites ( Bartels & Sunkar, 2005 ). This progress has been shown to be effective information for engineering drought tolerance in plants, such as Arabidopsis , tobacco, rice, tomato or wheat ( Vinocur & Altman, 2005 ; Umezawa et al. , 2006 ; Valliyodan & Nguyen, 2006 ; Varshney et al. , 2011 ). However, many challenges remain, such as the evaluation of natural genetic variations and association with specific gene functions ( Qin et al. , 2011 ). Several hundreds of genes induced by drought (or other abiotic stress because significant cross-talk exists) have been identified and broad variations in the timing of induction of these genes are known (Figure II.5).

Time course

Drought stress

ABA-independent pathway? NCED ABA-independent pathway

Rapid and Emerging ABA Response NAC HDZF DREB2 ABA-dependent NAC pathway CBF4/ HDZFR HDZFR DREB1D NACR

HB AREB/ABF DRE/CRT NACR ZF Slow and ABRE CE AP2/ERF Adaptive MYB MYC Response ? ? MYBR MYCR

Target stress-inducible genes

Stress tolerance

Figure II.5: Transcriptional regulatory networks of cis -acting elements and transcription factors involved in drought-stress-responsive gene expression in Arabidopsis . Transcription factors controlling stress- inducible gene expression are shown in colored ellipses. Cis -acting elements involved in stress-responsive transcription are shown in boxes. Small filled circles reveal modification of transcription factors in response to stress signals for their activation, such as phosphorylation. Regulatory cascade of stress- responsive gene expression is shown from top to bottom. Early and emergency responses of gene expression are shown in the upper part and late and adaptive responses in the bottom. Thick gray arrows indicate the major signalling pathways and these pathways regulate many downstream genes. Broken arrows indicate protein-protein interactions ( adapted from Yamaguchi-Shinozaki & Shinozaki, 2006 ).

Two groups of proteins induced during water-deficit stress emerged: i) proteins involved in further regulation of signal transduction and gene expression and ii ) proteins probably involved in stress tolerance ( Yamaguchi-Shinozaki & Shinozaki, 2006 ; Shinozaki & Yamaguchi-Shinozaki, 2007 ). In the first group, drought-induced gene expression was rapid

- 39 - Chapter II. Literature review and transient reached a maximum after several hours and then decreased. Most of these genes encode regulatory proteins such as several transcription factors (ABREs [ABA Response Elements], AP2/ERF [APETALA2/ethylene-responsive factor], DREB2 [Dehydration Response Element 2]), protein kinases ( i.e. MAPK [Mitogen-Activated Protein Kinase] cascades, SNF1-like kinase [Sucrose Non-Fermenting 1]), protein phosphatases ( e.g. PP2C [Protein Phosphatase 2C]), signalling molecules ( e.g. CDPKs [Calcium-Dependent Proteins Kinases] or SOS [Salt Overly Sensitive]). In the second group, gene expression slowly and gradually increased with stress. These genes mainly encode functional proteins such as Late Embryogenesis Abundant (LEA) proteins, Heat Shock Proteins (HSPs), chaperones; enzymes for osmolyte biosynthesis (sugars [RFOs, sucrose and trehalose], amino acids [ e.g. proline], polyamines, cyclitols and glycine betaine); and several detoxification enzymes ( e.g. aldehyde dehydrogenase or peroxiredoxins) ( reviewed by Shinozaki et al. , 2003 ; Bartels & Sunkar, 2005 ; Yamaguchi-Shinozaki & Shinozaki, 2006 ; Seki et al. , 2007 ; Shinozaki & Yamaguchi- Shinozaki, 2007 ).

II.1.2.7.1 Signal perception A change in the osmotic potential across the plasma membrane, caused by a drop of turgor pressure, is likely to be a major trigger of water-stress signal perception ( Shinozaki & Yamaguchi-Shinozaki, 1997 ). However, no plant molecule has truly been identified as osmosensor ( Bartels & Sunkar, 2005 ). Several candidates have been proposed such as AtHK1 [an Arabidopsis thaliana Histidine Kinase] ( Urao et al. , 1999 ), NtC7 [a membrane-located receptor-like protein] ( Tamura et al. , 2003 ) and Cre1 [cytokinin response 1] ( Reiser et al. , 2003 ). Recently, Wohlbach et al. ( 2008 ) showed that overexpression of AtHK1 increases drought resistance in Arabidopsis , while knockout mutants were less tolerant.

II.1.2.7.2 Signal transduction Plants react to water-deficit stimuli by initiating signal cascades which activate signal transduction ( Bartels & Sunkar, 2005 ).

II.1.2.7.2.1 ABA biosynthesis and signal transduction As described above, ABA plays crucial roles in adaptive responses to drought. Therefore, numerous studies on the ABA biosynthetic pathway, ABA signalling and degradation have been conducted. The molecular basis of ABA biosynthesis and catabolism was reviewed by Nambara and Marion-Poll ( 2005 ). Briefly, ABA biosynthesis is initiated by the cleavage of zeaxanthin, which is converted to violaxanthin by zeaxanthin epoxidase. Then, epoxycarotenoid cleavage by NCED (9- cis -epoxycarotenoid dioxygenase) to xanthoxin is the key point of ABA biosynthesis. Thereafter, xanthoxin is converted by a short-chain alcohol dehydrogenase into abscisic aldehyde, which is oxidized into ABA by an abscisic aldehyde oxidase. Regarding ABA catabolism, the oxidative pathway seems to be predominant. This way is initiated by 8’-hydroxylation via ABA 8’-hydroxylase, a cytochrome P450 CYP707A superfamily member. The most widespread and abundant ABA catabolites are phaseic acid (PA) and dihydrophaseic acid (DPA). Carboxyl and hydroxyl groups of ABA and its

- 40 - Chapter II. Literature review oxidative catabolites may also be conjugated to glucose. ABA glucose ester (ABA-GE) is the most abundant conjugate ( Nambara & Marion-Poll, 2005 ; Umezawa et al. , 2011 ). When plants are exposed to drought, ABA-GE is hydrolyzed by AtBG1, an Arabidopsis !- glucosidase, thus leading to an increase in the active ABA concentration. AtBG1-deficient mutants produced stress-sensitive phenotypes, such as defective stomatal movement and lower ABA levels ( Lee et al. , 2006 ). Moreover, ABA-GE plays an important role in long distance signalling under water-deficit stress ( Jiang & Hartung, 2008 ).

Environmental stimuli A Developmental cues B ABA PYR/PYL/RCAR

P P a P PP2 C P bi 2 1 C P -1 P P P SnRK2 SnRK2 SnRK2

P P AREB/ABF Other Factors

ABRE Other responses Target genes (e.g. ion transport)

ABA Responses Stress responses Seed dormancy etc. Figure II.6: Model of the major ABA signalling pathway. PYR/PYL/RCAR, PP2C and SnRK2 form a signalling complex referred to as the ‘ABA signalosome’. (A) Under normal conditions, PP2C negatively regulates SnRK2 by direct interactions and dephosphorylation of multiple residues of SnRK2. Once abiotic stresses or developmental cues up-regulate endogenous ABA, PYR/PYL/RCAR binds ABA and interacts with PP2C to inhibit protein phosphatase activity. In turn, SnRK2 is released from PP2C- dependent regulation and activated to phosphorylate downstream factors, such as the AREB/ABF bZIP- type transcription factor or membrane proteins involving ion channels. (B) In contrast, the abi1-1-type mutated protein lacks PYR/PYL/RCAR binding, resulting in the constitutive inactivation of SnRK2, even in the presence of ABA, and strong insensitivity to ABA in the abi1-1 mutant ( redraw from Umezawa et al. , 2010 ).

Since the publication of Figure II.5, recent breakthroughs have led to the identification of multiple ABA receptors (see reviews of Cutler et al. , 2010 ; Raghavendra et al. , 2010 ; Umezawa et al. , 2010 ; 2011 ), giving new insights on the ABA-dependent network (Figure II.6). A series of heterotrimeric GTP-binding proteins (G-proteins) were identified to play a

- 41 - Chapter II. Literature review role in ABA signalling ( Perfus-Barbeoch et al. , 2004 ). On the one hand, a GCPR (canonical G-protein)-type protein (GCR2) was shown to bind ABA and suggested to be a plasma membrane-type ABA receptor ( Liu et al. , 2007 ), but several questions remain open concerning the role of GCR2 in ABA signalling ( Umezawa et al. , 2011 ). On the other hand, it was demonstrated that GTG1 and 2 positively regulate ABA response. These proteins were reported as G-protein-related ABA receptors ( Pandey et al. , 2009 ). Moreover, an ABA- binding protein (ABAR) which encodes a multisubunit Mg-chelatase H was reported as another ABA receptor and was functionally characterized ( Shen et al. , 2006 ). A recent study showed that ABAR antagonizes the WRKY transcription repressors to relieve ABA- responsive genes of inhibition ( Shang et al. , 2010 ). In 2009, a small protein family, named PYR1/PYLs/RCARs (PYrabactin Resistance 1/Pyrabactin resistance 1-Likes/Regulatory Component of ABA Receptors), has been proposed and validated as ABA receptor by two independent research groups ( Ma et al. , 2009 ; Park et al. , 2009 ). PYR/PYLs/RCAR was found to directly bind and regulate PP2C (see II.1.2.7.2.4). Together with SnRK2 (see II.1.2.7.2.3), a central signalling complex (ABA- PYR-PP2C-SnRK2) was responsible for ABA signal perception and transduction (Figure II.6) (see reviews of Cutler et al. , 2010 ; Raghavendra et al. , 2010 ; Umezawa et al. , 2010 ; 2011 ). The structural analyses of these proteins confirmed the mode of ABA binding in a ligand pocket surrounded by two open-lid loops of PYR/PYL/RCAR ( Melcher et al. , 2009 ; Miyazono et al. , 2009 ; Nishimura et al. , 2009 ; Santiago et al. , 2009 ; Yin et al. , 2009 ) which have been discriminated by their selectivity against ABA chirality or pyrabactin ( Melcher et al. , 2010 ; Peterson et al. , 2010 ). Recently, the ABA receptor VvPYL1 ( Li et al. , 2011b ) as well as other proteins ( i.e. RCARs and PP2Cs) involved in the ABA signaling cascade ( Boneh et al. ) were identified in grapevine.

II.1.2.7.2.2 MAPKinase pathways Protein phosphorylation is an important mechanism for controlling cellular functions in response to abiotic stress signals. The MAPK cascades were recognized as major signal transduction mechanisms and may regulate numerous processes, including abiotic stress and hormonal responses, through multiple signal transduction pathways ( reviewed by Bartels & Sunkar, 2005 ; Rodriguez et al. , 2010 ). Recently, a novel nuclear protein kinase, i.e. DSM1 [Drought Sensitive Mutant 1], has been shown to play a critical role in drought and oxidative stress resistance in rice by regulating ROS scavenging ( Ning et al. , 2010 ).

II.1.2.7.2.3 SNF-1-like kinases Another family of protein kinases is the SNF-1-like kinases, which are expressed in response to osmotic stress or ABA ( for review see Umezawa et al. , 2011 ). SNF-1 related proteins kinases have been classified into three families: SnRK1, SnRK2 and SnRK3. In Arabidopsis , SnRK1-type kinase (KIN10 and KIN11) plays a role as central integrator of transcription networks in plant stress and energy signalling ( Baena-González et al. , 2007 ). In Arabidopsis , nine of the ten SnRK2 are activated by osmotic stress ( Boudsocq et al. , 2004 ). Moreover, SnRK2 plays an important role in ABA signalling ( Yoshida et al. , 2002 ) and subclass II SnRK2s regulates some drought-responsive genes involving ABA-responsive

- 42 - Chapter II. Literature review element binding transcription factors (AREB/ABF) ( Mizoguchi et al. , 2010 ). Kobayashi et al. (2004b ) have proposed that the activation mechanisms of SnRK2 should differ between ABA and osmotic stress. Umezawa et al. ( 2004 ) demonstrated that knockout mutants of SRK2C exhibited drought hypersensitivity in their roots, suggesting that SRK2C is a positive regulator of drought tolerance in Arabidopsis roots. Moreover, these authors showed that transgenic plants overexpressing SRK2C displayed higher drought tolerance in concert with the upregulation of several drought-inducible genes. Recently, a negative regulatory mechanism of SnRK2s mediated by group A PP2C [type 2C protein phosphatase] have been demonstrated ( Umezawa et al. , 2009 ; Vlad et al. , 2009 ). PP2C can inactivate SnRK2s by direct dephosphorylation. ABA-activated SnRK2s is accompanied by phosphorylation of the kinase activation loop ( Umezawa et al. , 2009 ).

II.1.2.7.2.4 Phosphatases The action of protein kinases is counteracted by phosphatases providing modulation and reversibility of the phosphoregulatory mechanisms ( Bartels & Sunkar, 2005 ). Plants have many types of phosphatases: phosphoprotein serine/threonine phosphatases (PPases) and phosphotyrosine phosphatases (PTPases). PPases are classified into four groups: PP1, PP2A, PP2B and PP2C ( for details see Luan, 2003 ). The function of most of them has not yet been elucidated, but six PP2Cs belonging to clade A (ABI1, ABI2, HAB1, HAB2, AHG1 and AHG3) were shown to negatively regulate the ABA signalling pathways (see reviews of Hirayama & Shinozaki, 2007 ; Umezawa et al. , 2011 ). Until very recently, the molecular mechanisms were unclear up to the discovery of ABA receptors (see II.1.2.7.2.1), i.e. PYR1/PYL/RCAR-type ABA receptor. These proteins bind ABA and interact with ABA- related PP2Cs (ABI1 and ABI2) in an ABA-dependent manner and inhibits PP2C activity (Ma et al. , 2009 ; Park et al. , 2009 ). PP2Cs are also targets of various second messengers, such as phosphatidic acid (produced by phospholipase D) and ROS, with important roles in ABA response (reviewed in Cutler et al. , 2010 ; Raghavendra et al. , 2010 ; Umezawa et al. , 2010 ; 2011 ).

II.1.2.7.2.5 Phospholipid signalling Phospholipids play an important role in membrane biosynthesis and signal transduction. They are cleaved by phospholipases, which produces phospholipid-derived second messengers ( Bartels & Sunkar, 2005 ). Plant cells contain a variety of phospholipid-based signalling pathways, including phospholipase C (PLC), D (PLD), A1 (PLA1) and A2 (PLA2). Lipid signalling is well demonstrated in abiotic stress, where the signal production is fast (min) and plays an important role in osmotic stress and ABA responses ( Testerink & Munnik, 2005 ; Munnik & Vermeer, 2010 ; Testerink & Munnik, 2011 ). For example, phosphatidic acid inhibits the activity of ABI1 (ABA insensitive 1) (Zhang et al. , 2004 ), a PP2C that negatively regulates ABA responses (see II.1.2.7.2.4). PLD is involved in the regulation of diverse cellular processes in plants, such as ABA signalling, root growth and abiotic stress adaptation (Wang, 2005 ). Recently, a genome-wide and molecular analysis of the PLD gene family were reported for poplar and grapevine ( Liu et al. , 2010 ).

- 43 - Chapter II. Literature review

II.1.2.7.2.6 Calcium signalling Calcium functions as second messenger in osmotic stress ( Bartels & Sunkar, 2005 ) and ABA signalling ( Umezawa et al. , 2011 ). Calcium signalling involves Ca 2+ -dependent protein kinases (CDPKs) ( Harper et al. , 2004 ), calcineurin B-like proteins (CBLs) sensor and CBL- interacting protein kinases (CIPKs) ( Luan, 2009 ; Weinl & Kudla, 2009 ). An example was given by Zhu et al. ( 2007b ) who identified CPK4 and CPK11 (two homologues of CDPKs) as positive regulators of ABF1 and ABF4 (Abscisic acid responsive element-Binding Factor) in response to ABA, providing evidence that CDPKs may regulate ABA signalling. Recently, Xu et al. ( 2011 ) showed that the calmodulin-like gene OsMSR2 (a Ca2+ -binding protein) enhances drought tolerance and increases ABA sensitivity in Arabidopsis .

II.1.2.7.3 Transcriptional regulation of gene expression Molecular analyses have demonstrated the presence of both ABA-independent and ABA- dependent pathways (Figure II.5) in the transcriptional regulatory networks governing drought-inducible gene expression ( Yamaguchi-Shinozaki & Shinozaki, 2006 ).

II.1.2.7.3.1 ABA-independent pathway ABRE (ABA-responsive element) and DRE (dehydration responsive element)/CRT (C- RepeaT) are cis -acting elements involved in drought-inducible gene expression via ABA- independent and -dependent pathways ( Shinozaki & Yamaguchi-Shinozaki, 2007 ). Transcription factors belonging to the ERF/AP2 family binding to these DRE/CRT elements were isolated and termed CBF (C-repeat Binding Factor)/DREB1 and DREB2 ( Lata & Prasad, 2011 ). DREB genes are important plant transcription factors that regulate various stress- responsive gene expressions. They play a key role in providing tolerance to multiple stresses, generally in an ABA-independent manner ( Lata & Prasad, 2011 ). Interestingly, overexpression of CBF/DREB1 improved drought tolerance while DREB2 did not ( Shinozaki & Yamaguchi-Shinozaki, 2007 ). However, transcriptomic analysis revealed that DREB2A regulates the expression of many drought-inducible genes that do not respond to ABA, such as ERD1 (Early Response to Dehydration 1), ZHFD (zinc-finger homeodomain) and heat shock transcription factor 3 (AtHSFA3) ( Yamaguchi-Shinozaki & Shinozaki, 2006 ; Qin et al. , 2011 ). Moreover, NAC (an acronym for NAM [No Apical Meristem], ATAF1-2 and CUC2 [Cup-Shaped Cotyledon]) transcription factors were induced in both ABA-independent ( Tran et al. , 2004 ) and -dependent ( Fujita et al. , 2004 ) pathways (Figure II.5). NAC genes have been found to be involved in drought stress adaptation in Arabidopsis ( Tran et al. , 2004 ) and in rice ( Jeong et al. , 2010 ). In grape, three CBF/DREB ( VvCBF1 , VvCBF2 and VvCBF3 ) family members were identified. The expression levels of these three genes were higher in young compared to mature tissues and were shown to be induced in response to cold, water- deficit and exogenous ABA treatment ( Xiao et al. , 2006 ).

II.1.2.7.3.2 ABA-dependent pathway Water-deficit stress induces ABA synthesis and accumulation in plant cells. A host of ABA-responsive genes has been identified. Most of these genes contain a conserved, ABA- responsive cis -acting element named ABRE ( Qin et al. , 2011 ). This sequence is specifically

- 44 - Chapter II. Literature review recognized by AREB (ABA-responsive element-binding protein), a group of bZIP (basic domain leucine zipper) transcription factors ( Yamaguchi-Shinozaki & Shinozaki, 2006 ). With ABA, the ABA-bound receptor inactivates PP2C, which causes the SnRK2 to be in a phosphorylated state, allowing them to activate bZIP transcription factors, such as AREB1/ABF2 (ABA-responsive factor 2) by protein phosphorylation (Figure II.6). Once activated, AREB1/ABF2 promote downstream ABA-responsive gene transcription ( Qin et al. , 2011 ). The induction of the drought-inducible gene RD22 is mediated by ABA and requires protein biosynthesis. MYC (MYeloCytomatosis) and MYB (MYeloBlastosis) recognition sites in the RD22 promoter function as cis-acting elements in the dehydration-inducible expression of RD22. These two transcription factors are synthetized after ABA accumulation. Using transgenic plants overexpressing AtMYC2 and/or AtMYB2, Abe et al. ( 2003 ) showed that ABA-inducible gene expression of RD22 and AtADH1 (alcohol dehydrogenase 1) increased markedly. The ABA-responsive R2R3-type MYB transcription factor (MYB96) increases Arabidopsis drought resistance via ABA-auxin cross-talk and lateral root growth (Seo et al. , 2009 ). MYB96 is also involved in cuticular wax biosynthesis ( Seo et al. , 2011 ). More details on ABA-dependent transcriptional regulation are given in several recent reviews (Cutler et al. , 2010 ; Raghavendra et al. , 2010 ; Umezawa et al. , 2010 ; Qin et al. , 2011 ; Umezawa et al. , 2011 ).

II.1.2.7.4 Functional proteins: a role in stress-protection and adaptation Several functional proteins, such as Late Embryogenesis-Abundant (LEA) ( Rorat, 2006 ; Kantar et al. , 2011 ), Heat Shock Proteins (HSP), protease and proteinase inhibitors ( Bartels & Sunkar, 2005 ; Kantar et al. , 2011 ) are involved in water-deficit stress protection and adaptation. Transcriptomic analyses of transgenic Arabidopsis overexpressing DREB2A revealed that not only drought-responsive but also heat-shock-related genes are induced (Sakuma et al. , 2006 ). In two Vitis species ( V. vinifera and V. riparia ), Xiao and Nassuth (2006 ) observed that cold, drought and ABA treatments increased the expression of two similar dehydrin genes, which are members of the LEA protein group.

II.1.2.7.5 Drought-induced transcriptomic and proteomic changes in roots Advances in biotechnology such as microarrays ( Rensink & Buell, 2005 ; Sreenivasulu et al. , 2010 ), deep RNA sequencing ( Ozsolak & Milos, 2011 ) and other omics technologies (reviewed in Mochida & Shinozaki, 2010 ; Urano et al. , 2010 ) coupled with the availability of several plant genome sequences ( Feuillet et al., 2011 ) have improved the understanding of transcriptional behavior and root biology ( Benfey et al., 2010 ). High-resolution spatio-temporal gene expression of individual cell-types and developmental zones of the root led to the identification of new genes regulating lateral root initiation ( De Smet et al., 2008 ). These transcriptomic studies also showed genes co- regulation in specific tissues- or developmental zones ( Birnbaum et al., 2003 ) and they revealed fluctuating gene expression patterns depending on developmental stage ( Brady et al., 2007 ). Cell identity determines the pool of genes regulated, and cell type-specific processes seem a common target of the abiotic stress responses that allow the root to acclimatize or

- 45 - Chapter II. Literature review adapt to environmental stresses ( Dinneny et al., 2008 ). The large data sets of “omics” results pointed the way to a system biology analysis using bioinformatics ( Shinozaki & Sakakibara, 2009 ; Van Norman & Benfey, 2009 ). For example, co-expression gene network is a powerful approach to identify coordinated pathways ( Aoki et al., 2007 ; Usadel et al., 2009 ). It has been used to analyze plant responses to various stresses ( Less et al., 2011 ), including the response of drought-stressed roots ( Lorenz et al., 2011 ).

Recent advances elucidating the processes involved in root growth under water-deficit, by combining kinematic, transcriptomic and proteomic approaches, were reviewed by Yamaguchi and Sharp ( 2010 ). Spatial profiles of several interesting genes, such as genes coding for a plasma membrane proton pumping ATPase and a relatively rare proton pump activator, appeared to correlate with the profile of root elongation ( Bassani et al. , 2004 ). These authors also showed that cell wall metabolism and signal associated genes were strongly and exclusively expressed in the region of highest growth velocity (1-5 mm from the root cap) of well-watered plants and were also expressed most strongly in the 1-3 mm region from the root cap under water stress ( Bassani et al. , 2004 ). Fan et al. ( 2006 ) showed that inhibition of cell wall extensibility by water stress along successive zones of maize roots is related to the increases of transcripts of two genes involved in lignin biosynthesis, cinnamoyl- CoA reductase 1 and 2. Moreover, these authors observed progressive accumulation of wall- phenolic compounds in vascular areas of the stele. Phenolic compounds are potent scavengers of ROS and are involved in protection from oxidative damage, and they can be involved in water stress tolerance mechanisms. In V. vinifera roots, p-coumaric and ferulic acid levels rose significantly under water stress, while the levels of cafeic acid increased during the post- drought recovery period ( Weidner et al. , 2009 ).

By comparing transcript profiles of specific-regions of the maize root tip (Figure II.2), Poroyko et al. ( 2007 ) demonstrated that transcript populations were largely unaffected by water stress in region 1, correlating with the maintenance of elongation rate. In contrast, region 2 and region 3 profiles changed with water supply, particularly in region 2 where jasmonic acid biosynthesis and signalling appear to be important in stress sensing. Moreover, functions in cell expansion, cell wall biosynthesis, cell maturation and hormone biosynthesis are region-specific along the root tip ( Poroyko et al. , 2007 ). In maize primary roots, ROS, carbon metabolism and wall loosening proteins are involved in the maintenance of root tip growth while osmotic adjustment, regulation by ABA and changes in membrane transport participate in the inhibition of root elongation zone ( Spollen et al. , 2008 ). Large-scale proteomic analysis of the maize primary root elongation zone have identified 152 water deficit-responsive cell wall proteins categorized into groups associated with ROS metabolism, defence and detoxification, hydrolases, carbohydrates metabolism and other unknown processes ( Zhu et al. , 2007a ). Recently, proteomic analysis revealed region-specific regulation of phenylpropanoid metabolism and control of free iron in the root elongation zone (Yamaguchi et al. , 2010 ). Several enzymes related to isoflavonoid biosynthesis increased in region 1 (Figure II.2), correlating with a substantial increase of isoflavonoid content, and may

- 46 - Chapter II. Literature review contribute to growth maintenance. In contrast, caffeoyl-CoA O-methyltransferase, which is involved in lignin synthesis, was up-regulated in region 2 and may be related to the inhibition of growth in this region. Several proteins, like ferritin, whose abundance increased in both regions of water-stressed roots are related to protection from oxidative damage ( Yamaguchi et al. , 2010 ).

Since the publication of the whole genome sequences of Vitis vinifera L. cv. Pinot noir (Jaillon et al., 2007 ; Velasco et al., 2007 ), grape has become a model to study fleshy fruit in perennial plants. As a result, several transcriptomic approaches have been expanded (reviewed in Romieu et al. , 2011 ; Tillett & Cushman, 2011 ). Despite the importance of roots under water deficiency, grape root “omics” studies have been fairly limited compared to shoots and berries ( Cramer et al. , 2007 ; Tattersall et al. , 2007 ; Vincent et al. , 2007 ; Deluc et al. , 2009 ). To date, grape root transcriptomic tools were used to generate expressed sequence tags (ESTs) ( Goes da Silva et al. , 2005 ; Moser et al. , 2005 ), to profile microRNAs using oligonucleotide microarrays ( Mica et al., 2010 ) or to assess adventitious root development (Thomas & Schiefelbein, 2002 ; Thomas et al. , 2003 ). Because adventitious rooting is the most important way of grapevine propagation, the reader is referred to the excellent review of Li et al. ( 2009 ) who described the underlying signalling and molecular processes. Many endogenous and exogenous factors, such as Ca 2+ , sugars, auxin, polyamines, ethylene, nitric oxide and hydrogen peroxide, have been identified as signals and mediate auxin signal transduction during adventitious root development. During adventitious rooting, VvPRP1 and VvPRP2 , encoding a proline-rich protein and VvADF , encoding an actin depolymerising factor protein, are involved in the initiation of new roots on stem cuttings of grapevine, perhaps by reorganizing cytoskeleton and increasing cell wall plasticity to enable root meristem emergence ( Thomas & Schiefelbein, 2002 ; Thomas et al. , 2003 ). Recently, Tillett et al. ( 2011 ) identified tissue-specific abiotic responsive genes using curation and data mining of large-scale ESTs in grapevine. The analysis detected 135 genes with root-enriched expression patterns. Among these genes, qRT-PCR validation confirmed several root-specific genes: a tonoplast intrinsic protein (TIP1;4), a resveratrol O- methyltransferase (ROMT), a terpene synthase ((E,E)- -farnesene synthase), a cinnamyl- alcohol dehydrogenase, a flavonol O-glucosyltransferase and a MYB transcription factor-like gene ( Tillett et al. , 2011 ).

II.2 Root-to-shoot water transport, drought and grape rootstocks Water continuously flows from soil to atmosphere in the soil-plant-atmosphere continuum (Taiz & Zeiger, 2002 ). Under drought conditions, the efficiency water transport is dramatically affected. Plant traits such as anatomical structure ( Shao et al. , 2008 ), hydraulic constraints ( Steudle, 2000b ) and chemical signals ( Schachtman & Goodger, 2008 ) play a major role under water-deficit, affecting whole plant water transport.

- 47 - Chapter II. Literature review

II.2.1 Water transport: some generalities The movement of water is driven by a reduction in free energy. Water may move by diffusion, by bulk flow, or by a combination of these mechanisms. Water flow is a passive process and water moves in response to physical forces along a gradient of water potential, toward regions of low water potential or low free energy. Water moves from the soil through the plant via the xylem conduits to the atmosphere ( Taiz & Zeiger, 2002 ). Two main forces regulate the root water uptake rate, namely osmotic and hydrostatic forces. The hydrostatic force is generated by the transpiration stream, whereas the osmotic force is generated by the pressure (active transport of solutes or biosynthesis of new osmolytes). Water transport is divided into radial and axial transport ( Taiz & Zeiger, 2002 ).

II.2.1.1 Radial transport

II.2.1.1.1 The three pathways The radial water flow from the soil solution to the root xylem conduits involves three dynamically exchangeable paths (Figure II.7): apoplastic, symplastic and transcellular (Steudle & Peterson, 1998 ; Steudle, 2000a ).

Figure II.7: Pathways for the movement of water and solutes in roots. The apoplastic path (a) refers to the flow around protoplasts. The symplastic component defines flow from cell to cell via plasmodesmata (b). On the third route (c), water and solutes have to cross cell membranes (two membranes per cell layer; transcellular path). The transcellular path is important for water, but is of minor importance for solutes. For water, pathways (b) and (c) cannot be separated experimentally to date. Therefore, they are summarized as a cell-to-cell path. It is usually assumed that, in roots, the Casparian bands in the exo- and endodermis completely interrupt apoplastic transport. Here, it is, however, indicated that there may be an apoplastic component of water flow across Casparian bands ( adapted from Steudle, 2000a ).

- 48 - Chapter II. Literature review

The apoplastic path refers to water moving through the pores between the fibrils of the cell wall and through the intercellular spaces. The symplastic path defines water moving through the cytoplasm and through plasmodesmata between cells. Finally, the transmembrane path comprises water moving through the cytoplasm and the vacuoles crossing the plasma and vacuolar membranes (mainly plasma membranes). Since empirically the symplastic and transcellular paths cannot be discriminated, the sum of these two paths is called the cell-to- cell path ( Steudle & Peterson, 1998 ; Steudle, 2000a ). When the rate of transpiration is high, the apoplastic path will be preferred and the hydraulic resistance of the root will be low to facilitate water uptake; while under water stress, the apoplastic path is less used and the main route is the cell-to-cell path ( Steudle & Peterson, 1998 ; Steudle, 2000b ). In grapevine, a strong negative relationship, influenced by plant ontogeny, was observed between axial resistance and either root diameter or distance from the root tip. Water stress probably induced more lignified roots which would impede hydraulic conductivity and water uptake ( Mapfumo et al. , 1993 ). The axial resistance of root segments may not be uniform along roots. The main roots dominate lateral ones for water uptake under well-watered conditions, but under water stress, the lateral roots presumably become progressively more important for this process ( Mapfumo et al. , 1994 ). In the apical region of ungrafted grapevine roots, the major radial resistance to water flow is likely the large cortex. In the periphery of the vascular cylinder, the endodermis and the parenchyma cells are likely the dominant components of radial resistance in the basal regions ( Mapfumo & Aspinall, 1994 ).

II.2.1.1.2 Aquaporins and radial water transport Aquaporins are channel proteins present in the plasma and intracellular membranes of plant cells, where they facilitate the transport of water and/or small neutral solutes. The cell- to-cell path is mostly mediated by aquaporins ( Maurel et al. , 2008 ; 2009 ). Aquaporins are members of a large superfamily called major intrinsic proteins (MIPs). This superfamily is divided into four main sub-groups: tonoplast intrinsic proteins (TIPs), plasma membrane intrinsic proteins (PIPs), nodulin-26-like intrinsic membrane proteins (NIPs) and small basic intrinsic proteins (SIPs) ( Kaldenhoff & Fischer, 2006 ; Maurel et al. , 2008 ). Many aquaporins are expressed in roots and are involved in the control of water flow between the symplast and the apoplast ( Kaldenhoff et al. , 2008 ). Bramley et al. ( 2009 ) demonstrated the importance of examining root morphology and anatomy in assessing the role of aquaporins in root water transport. In grape, the expression of aquaporins is stronger in roots than in leaves, in the grape rootstock Richter 110 ( Baiges et al. , 2001 ; Galmés et al. , 2007 ). In four rootstock genotypes, the genes encoding VvPIP1;1 and VvTIP2;1 showed high expression levels in the roots and revealed many similarities between genotypes ( Fouquet, 2005 ). In V. vinifera Grenache and Chardonnay, two PIPs (VvPIP1;1 and VvPIP1;2) were functionally characterized in the root (Vandeleur et al. , 2009 ). The two genotypes displayed similar localized patterns of mRNA expression. A strong signal for both genes was detected in elongating cortical and vascular tissue of root tip longitudinal sections, especially in the root apex. Transversal sections at 30 and 50 mm from the root apex revealed that VvPIP1;2 expression occurred in vascular tissue

- 49 - Chapter II. Literature review adjacent to and between the xylem poles and also in the cortex, while VvPIP1;1 was not detected ( Vandeleur et al. , 2009 ). The xylem sap moves via a cell wall continuum in the apoplast. Thus, there is no direct involvement of aquaporins on water transport in xylem. However, the symplastic cells surrounding xylem vessels, the xylem parenchyma for example, could contribute to the regulation of xylem transport by variation of radial cell-to-cell water movement from perivascular tissues to the xylem. This, in turn, affects longitudinal xylem sap flow, especially during critical situations of water balance ( Kaldenhoff et al. , 2008 ). The role of aquaporins under water-deficit stress will be further discussed in section II.2.2.1.2.

II.2.1.2 Axial transport The axial transport of water along the xylem conduits to aerial parts does not contribute substantially to the resistance of water transport through the whole plant ( Tyree & Zimmermann, 2002 ). The transpiration of water from the soil through a vascular plant, and into the air, occurs passively. This mechanism is described by the cohesion-tension theory: water ascends plants in a metastable state under tension, i.e. with xylem pressure more negative than that of the vapour pressure of water. The loss of water by evapotranspiration reduces the pressure of the liquid water within the leaf, which generates a driving force. This reduced pressure pulls liquid water through a continuous water column, from the soil and up the xylem, causing a negative gradient pressure-driven water flow ( Tyree, 1997 ; Steudle, 2001 ). Despite controversies, the cohesion-tension theory is strongly supported by experimental evidences ( Tyree, 1997 ; Steudle, 2001 ; Tyree & Zimmermann, 2002 ). Moreover, the recent realization of the key features of the cohesion-tension mechanisms of transpiration in a synthetic system provides new supports for this theory (Wheeler & Stroock, 2008 ).

II.2.2 Regulation of water transport under water-deficit stress Large differences in canopy water supply between species can be attributed in part to differences in their “hydraulic equipment”, such as plant architecture and xylem traits which influence hydraulic conductivity and may cause hydraulic limits ( e.g. xylem cavitation) to leaf water supply under drought conditions ( Sperry et al. , 2002 ).

II.2.2.1 Hydraulic architecture In angiosperm tree species, the xylem includes three different tissue types that fulfil different functions: the vessels (dead cells) provide longitudinal water transport; the parenchyma contains physiologically active cells which function in carbohydrate storage and local radial transport; and the fibers provide mainly strength ( Poorter et al. , 2010 ). For example, hydraulic conductance depends both on the cross-sectional area occupied by vessels, and on the size and number of these conduits. According to the Hagen-Poiseuille law, wider vessels contribute to a larger hydraulic conductivity (the lumen conductivity increases with the fourth power of the lumen diameter) ( Sperry et al. , 2006 ) which, in turn, facilitates higher stomatal conductance, transpiration rate and photosynthesis ( Chouzouri & Schultz, 2005 ; Rodríguez-Gamir et al. , 2010 ). In contrast, smaller vessels provide a higher hydraulic safety because they are less prone to vessel implosion ( Hacke et al. , 2001 ) and cavitation (as small

- 50 - Chapter II. Literature review vessels have lower risk of air-seeding because of their smaller pit membrane area) ( Tyree & Sperry, 1989 ; Hacke et al. , 2006 ).

II.2.2.1.1 Anatomical structure Wood anatomy can be used to compare the hydraulic efficiency of species, and can also help to identify the hydraulic limits to water transport ( Pittermann, 2010 ). Long-distance water flow occurs through the lumens of non-living tracheary elements (vessels and tracheids) of the xylem and through the lateral pits that interconnect the tracheary elements ( Choat et al. , 2008 ; Nardini et al. , 2011 ). A vessel is composed of a series of vessel elements interconnected by perforations at the end walls (Figure II.8). Because vessels and tracheids are of finite length, the sap moves from vessel to vessel or from tracheid to tracheid through lateral pit pairs ( Tyree & Ewers, 1991 ).

Figure II.8: Conducting cells in the xylem. The xylem consists of two types of tracheary elements: tracheids and vessel elements ( from Taiz & Zeiger, 2002 ).

Grapevines have both long and wide vessels in their woody stem which can be in excess of 1 meter and 300 )m, respectively. Moreover, there is an obvious correlation between vessel diameter and vessel length: wider vessels being generally longer ( Zimmermann & Jeje, 1981 ; Chatelet et al. , 2006 ). Vessel dimorphism is characteristic of the liana-like Vitaceae . The vessels contain exclusively simple perforations and intervessel pits are predominantly

- 51 - Chapter II. Literature review scalariform or round in opposite positions ( Schweingruber et al. , 2011 ). Many vines and lianas have large number of narrow vessel elements (often fibriform vessel elements) in addition to some vessels with large diameters ( Carlquist, 1985 ; Schultz & Matthews, 1993b ). However, xylem conduit characteristics are also dependent on both V. vinifera ( Chouzouri & Schultz, 2005 ; Chatelet et al. , 2006 ) and rootstock cultivars ( Pongrácz & Beukman, 1970 ). The latter authors compared the anatomy of thirteen rootstock ( Vitis sp.) genotypes and one V. vinifera cultivar. Highly significant anatomical differences, such as size of the pith or width of xylem rays, were found between genotypes, especially in one-year-old roots ( Pongrácz & Beukman, 1970 ). Recently, Alsina et al. ( 2011 ) observed that the frequency distribution of vessel length classes differs between a drought-sensitive (101-14 MGt) and a drought-tolerant (1103P) rootstock. Additionally, average vessel length and maximum vessel length were significantly longer for 1103P than for 101-14MGt, while the number of vessels and the mean lumen diameter were not different. Another important feature for cultivated grape anatomical characteristics is the grafting process. The beginning of the graft process is characterized by callus formation and graft union is considered to be successful when several functional phloem and xylem connections have developed across the graft union ( reviewed by Pina & Errea, 2005 ). Different levels of compatibility between scions and rootstocks have been observed in several species, including grapevine ( Gökbayrak et al. , 2007 ). However, up to now, little information is available on anatomical plasticity, such as xylem structure and the accompanying shift in hydraulic conductivity of grafted plants, especially under water-deficit. Wet and warm environments tend to favour species with broad xylem conduits whereas cold or dry environments tend to favour species with narrow xylem conduits ( Tyree & Zimmermann, 2002 ). Obviously, water movement can be significantly restricted by the efficiency of the hydraulic system ( Tyree & Ewers, 1991 ). During periods of water deficit, xylem anatomy and hydraulic conductance are tightly linked, with a frequent decrease in xylem conduits diameter also in grapevines ( Lovisolo & Schubert, 1998 ). In grapevines, the average transectional xylem area is 35% lower in drought-stressed than in irrigated plants at all tested positions along the shoots ( Lovisolo & Schubert, 1998 ). Anatomical shifts which result into narrower xylem conduits under water stress may contribute to restrict water flow (Tyree & Ewers, 1991 ), but simultaneously reduce the vulnerability of xylem conduit to cavitation ( Gonçalves et al. , 2007 ; Bauerle et al. , 2011 ).

II.2.2.1.2 Hydraulic conductivity and embolism induced-cavitation Generally, stomatal conductance is considered as the major control point to limit water loss (Jones, 1992a ), while hydraulic resistance at the soil-rhizosphere interface ( North & Nobel, 1997a , 1997b ), root and stem xylem properties ( Tyree & Ewers, 1991 ), root-to-shoot ratio (Jackson et al. , 2000 ) can significantly influence water supply versus demand ( Sperry et al. , 2002 ). To improve the water status of the shoot and to respond to increasing evaporative demand, plants can either increase stomatal conductance or decrease root hydraulic resistance. The hydraulic resistance of the xylem is relatively small, but strongly depends on tissue type (Schultz & Matthews, 1993b ). Resistance can become important under water stress when high

- 52 - Chapter II. Literature review tensions causes cavitation and interrupts water flow between root and shoot ( Tyree & Sperry, 1988 ). When plants are exposed to drought, the water flow is modified. Several studies consistently report a decline of hydraulic conductivity under water stress ( Rieger, 1995 ; Sperry et al. , 2002 ; Trifilò et al. , 2004 ). Water channel (aquaporins) activity is likely to be involved in balancing the water relations of the plants during water deficit, but the actual roles of aquaporins during drought remain unclear. Some aquaporins are induced by ABA, some assist water movement to critical cell or organs of the plant or may be involved in the plant’s response to rewetting ( Vandeleur et al. , 2005 ). In tree species during the summer, fine root hydraulic conductivity and aquaporin contribution were found to cycle diurnally with peaks corresponding to the highest transpiration demand at midday. These adjustments may help to maintain the use of reliable water resources from deep soil layers ( McElrone et al. , 2007 ). Thus, plant hydraulic conductivity is partially regulated by aquaporin function ( Javot & Maurel, 2002 ; Maurel et al. , 2008 ), such as that of PIPs in the roots ( Javot et al. , 2003 ; Postaire et al. , 2010 ). ROS-dependent signalling mechanisms can trigger phosphorylation- dependent intracellular trafficking of PIPs in response to water stress. This may in part explain downregulation of root water transport ( Boursiac et al. , 2008 ). However, many studies report different patterns of aquaporin expression under drought, especially for PIPs. It is therefore difficult to find a common response of PIPs expression and protein abundance under water stress ( reviewed in Kaldenhoff et al. , 2008 ; Aroca et al. , 2011 ). Under water-deficit, another important constraint on hydraulic conductivity is cavitation- induced embolisms. Cavitation occurs when an air bubble of sufficient radius forms in a vessel containing water under tension. The bubble is gas-filled (water vapour and some air) and is inherently unstable. There is evidence that drought-induced cavitation occurs by air seeding ( i.e. an air bubble is sucked into a water-filled lumen from an adjacent air space) through pores in the intervessel pit membranes ( Tyree & Sperry, 1989 ; Cochard, 2006 ; Choat et al. , 2008 ; Delzon et al. , 2010 ). Under high tension (negative pressure) induced by water stress, the water column in the xylem may break, causing embolism formation and gas-filled vessels, which drastically reduce hydraulic conductance ( Lovisolo et al. , 2010 ). Vulnerability of xylem to cavitation and embolism may not directly be related to xylem diameter, since its pit membrane properties which largely determine the vulnerability to cavitation ( Choat et al. , 2008 ). However, because xylem size has a strong impact on hydraulic conductivity: a large embolized xylem vessel induces a considerable down-regulation of water flow ( Tyree & Sperry, 1989 ; Tyree & Zimmermann, 2002 ; Cochard, 2006 ). Research on water relations of grafted plants up to know has focussed on elucidating if scions or rootstocks plays the largest role in restricting water flow ( Cohen & Naor, 2002 ; Cohen et al. , 2007 ). The decrease in plant vigour resulting from increased hydraulic resistance induced by the graft point is widely accepted ( Atkinson et al. , 2003 ; Gonçalves et al. , 2007 ; Trifilò et al. , 2007 ; Tombesi et al. , 2010a , 2010b ). However, Clearwater et al. ( 2004 ) showed that leaf-area-specific conductance and stomatal conductance were both higher in kiwifruit vines grafted onto low-vigour rootstock compared to high vigour rootstock. Recently, Bauerle et al. ( 2011 ) observed a greater tolerance to water-deficit in Malus domestica grafted onto a

- 53 - Chapter II. Literature review high-shoot vigour (HSV) compared to low-shoot vigour (LSV) rootstocks. Under drought, the HSV combination led to reduced xylem vessel diameter, and displayed higher hydraulic resistance and smaller embolism than the LSV rootstock ( Bauerle et al. , 2011 ). For the Vitaceae , several studies showed that drought-induced a decreases of hydraulic conductivity in roots, shoot nodes, internodes and petioles ( reviewed by Lovisolo et al. , 2010 ). Grapevines have long- and large-diameter xylem vessels ( Zimmermann & Jeje, 1981 ) and stems are often reported to be highly susceptible to water stress-induced cavitation ( Schultz & Matthews, 1988a ; Lovisolo & Schubert, 1998 ; Schultz, 2003a ), but the roots seem more vulnerable to xylem cavitation than shoots ( Lovisolo & Schubert, 2006 ; Lovisolo et al. , 2008a ). A recent report ( Choat et al. , 2010 ) showed that V. vinifera stems are far less vulnerable to cavitation than reported previously and are predicted to suffer significant embolism only for xylem water potentials below -2.0 MPa, although the vulnerability may depend on cultivar or organ type ( Schultz, 2003a ; Lovisolo et al. , 2008b ). Plants have evolved mechanisms such as stomatal closure to prevent embolism ( Tyree & Sperry, 1989 ; Sperry et al. , 2002 ). Refilling of embolized vessels despite the presence of significant tension throughout the xylem was observed in V. labrusca cv. Concord using magnetic resonance imaging (MRI) ( Holbrook et al. , 2001 ). To reintegrate vessel functionality after drought, plants have developed different repair mechanisms, involving ABA ( Lovisolo et al. , 2008a ), PIP1 aquaporins ( Secchi & Zwieniecki, 2010 ), ABA/aquaporins interactions ( Kaldenhoff et al. , 2008 ) or sucrose ( Secchi & Zwieniecki, 2011 ). Accumulation of ABA in the leaves associated with stomatal closure may help gradual embolism repair by limiting the rate of transpiration during recovery ( Lovisolo et al. , 2008a ). Kaufmann et al. ( 2009 ) have used MRI to monitor xylem cavitation and refilling in maize roots. Recently, Brodersen et al. ( 2010 ) have used computed tomography to visualize in vivo embolism repair of V. vinifera stem internode. They showed that xylem refilling is a localized, coordinated and complex process with water droplets entering into vessel from cells surrounding the xylem. These authors confirmed previous observations ( Holbrook et al. , 2001 ) that hydraulic conductivity could be restored in spite of tensions in the bulk xylem. ( Brodersen et al. , 2010 ) Under water deficit, grape rootstock genotypes may influence whole grapevine hydraulic conductivity through the anatomical characteristics of the xylem vessels ( Pongrácz & Beukman, 1970 ; Alsina et al. , 2011 ), the resistance of the root and of the grafting union ( de Herralde et al. , 2006 ), the regulation of expression and activity of aquaporins ( Fouquet, 2005 ; Galmés et al. , 2007 ; Lovisolo et al. , 2008b ) and/or the susceptibility to embolism-induced cavitation ( Lovisolo et al. , 2008b ). In 2-year-old plants, a mercurial treatment reduced root hydraulic conductance much more in drought tolerant rootstock hybrids ( V. berlandieri ×V. rupestris : 140Ru, 775P, 1103P) than in other hybrids ( V. berlandieri ×V. riparia : SO4, 157-11, 420A and 5BB), suggesting a higher contribution of cell-to-cell pathways to total water transport in the former. Moreover, the drought-tolerant hybrids had higher root hydraulic conductance and lower vessel embolization during water stress, suggesting either lower vulnerability to embolism or higher repair efficiency ( Lovisolo et al. , 2008b ). Recently, Alsina et al. ( 2011 ) showed that the graft union represents a large proportion of the total root resistance to water transport, but no differences were found between a drought-sensitive (101-

- 54 - Chapter II. Literature review

14 MGt) and a drought-tolerant (1103P) rootstock grafted with the same scion ( V. vinifera cv. Merlot). However, the drought-tolerant genotype increased whole root hydraulic conductance and leaf specific hydraulic conductance during the summer dry period in contrast to the drought-sensitive genotype ( Alsina et al. , 2011 ).

II.3 Shoot growth and water use under water-deficit stress Plant water stress symptoms, such as the decline of shoot growth and gas exchange, appear during periods of water-deficit when the rate of supply of water from soil to plant falls below the water demands ( Neumann, 2011 ).

II.3.1 Shoot growth and development Plant growth and biomass allocation are the two most fundamental processes being remarkably influenced by water-deficit stress ( Hsiao, 1973 ; Chaves et al. , 2003 ). One of the first processes affected by water-deficit is leaf growth. Many physiological mechanisms have been investigated, including the temporal and spatial patterns of leaf growth ( reviewed by Granier & Tardieu, 2009 ) or the roles of ABA via hydraulic and non-hydraulic processes (reviewed by Tardieu et al. , 2010 ). On a global basis, drought limits plant growth and crop productivity more than any other single environmental factor ( Boyer, 1982 ). Although plant growth rates are generally reduced when soil water supply is limited, shoot growth is often more inhibited than root growth. In some cases, the absolute root biomass of plants in drying soil may increase relatively to that of well-watered controls ( Sharp & Davies, 1989 ). Several studies indicate that drought induces an increase of the root-to-shoot biomass ratio ( reviewed by Jackson et al. , 2000 ). Padilla et al. ( 2009 ) showed that plant subjected to water-deficit allocated proportionally more biomass to the roots, but no consistent differences were found in the responsiveness of seven Mediterranean shrub species. Inhibition of shoot growth in response to water deficit may therefore extend the period of soil water availability and plant survival and can be considered as an adaptive response ( Neumann, 2008 ). In grapevine, canopy development, growth of main and lateral shoots, shoot apical meristem, leaf emergence, leaf area and tendrils formation were all strongly affected by water stress ( Matthews et al. , 1987 ; Schultz & Matthews, 1988a ; Schultz & Matthews, 1988b ; Schultz & Matthews, 1993a ; Hardie & Martin, 2000 ; Gómez-del-Campo et al. , 2002 ). Lebon et al. ( 2006 ) showed that all developmental processes, including leaves number on the primary and lateral shoots, and their leaf area, were inhibited by water-deficit. Appearance rate of the leaves on the lateral shoots was particularly sensitive to water-deficit and was considered as a major determinant of leaf area adaptation under drought ( Lebon et al. , 2006 ). The sensitivity of the organs increased with organs ontogeny ( Schultz & Matthews, 1988b ) and was dependent on the branche order ( Pellegrino et al. , 2005 ; Lebon et al. , 2006 ). Leaf emergence rate on second order lateral branches was sensitive to moderate water-deficit whereas growth and vegetative development ( i.e. leaf emergence rate, final individual leaf area and internodes length) on first order branches were only affected under high water-deficit (Pellegrino et al. , 2005 ). Grape rootstocks can influence the developmental traits of the scion. In water limited environments, rootstocks influence leaf area ( Koundouras et al. , 2008 ; Alsina

- 55 - Chapter II. Literature review et al. , 2011 ), yield ( Stevens et al. , 2008 ; 2010 ; Alsina et al. , 2011 ) or pruning weight (Koundouras et al. , 2008 ; Stevens et al. , 2010 ).

II.3.2 Regulation of leaf water potential and gas exchange

The regulation of gas exchanges ( i.e. photosynthesis or CO 2 assimilation [ A], stomatal conductance [ gs] and transpiration [ E]) at the leaf level is a key factor for plant survival under drought conditions. An abundant literature shows that stomata respond to short-term modifications in the environment (soil water status, light, air humidity and vapour pressure deficit, temperature and CO 2 concentration). Stomata play a key role both in controlling CO 2 assimilation and water losses by transpiration, and thus modulating the rate of soil water depletion. Stomatal aperture is controlled by turgor pressure differences between the guard cells surrounding the pores and the bulk leaf epidermis. Water-deficit induces the decrease of gas exchange and leaf water potential ( reviewed in Jarvis, 1976 ; Jones, 1992b ; Buckley, 2005 ). Because guard cells regulate the aperture of stomatal pore through the integration of both endogenous hormonal stimuli and environmental signals, physiological and molecular players have been extensively studied ( see excellent reviews of Schroeder et al. , 2001 ; Sirichandra et al. , 2009 ; Kim et al. , 2010 ; Chaves et al. , 2011 ; Pinheiro & Chaves, 2011 ). Briefly, water shortage induces an increase in ABA levels. When perceiving ABA accumulation, guard cells reduce their turgor and volume via efflux of anions and potassium ions ( Kim et al. , 2010 ). In Arabidopsis , stomatal closure is driven by K + efflux mainly through GORK (Guard cell Outward Rectifying K +-channel) which is activated by membrane depolarization. A gene encoding the anion-conducting subunit of S-type anion channels, SLAC1 (SLow Anion Channel-associated 1), is likely to initiate the membrane depolarization ( Sirichandra et al. , 2009 ; Kim et al. , 2010 ). Moreover, the guard cell-expressed transmembrane ABC (ATP binding cassette) AtMRP5 (Multidrug Resistance Protein 5) is also involved in ABA-induced stomatal closure ( Kim et al. , 2010 ). In the presence of ABA, the phosphatase activity of the receptor is blocked (see II.1.2.7.2.4), then the activity of protein kinases OST1 (Open Stomata 1) and related SnRKs regulate several events leading to stomatal closure via Ca 2+ -dependent and Ca 2+ -independent way ( Chaves et al. , 2011 ). In this context, ABA plays a key role in stomatal aperture. A decrease in leaf/root water potential resulting from soil water-deficit is typically accompanied by ABA production in roots which is then transported to the guard cells via the xylem with the sap flow ( Davies & Zhang, 1991 ; Tardieu & Davies, 1993 ; Davies et al. , 2005 ). Moreover, this long-distance ABA signalling may also be interconnected with hydraulic signals that may trigger stomatal closure in response to drought ( Tardieu & Davies, 1993 ; Comstock, 2002 ; Christmann et al. , 2007 ), for example via xylem embolism ( Cochard et al. , 1996 ; Cochard et al. , 2002 ). Additionally, diurnal and seasonal time course measurements of plant water potentials have shown a substantial range among plant species in their regulation of transpiration, leading to the classification of isohydric and anisohydric species based on their water potential behaviour ( Stocker, 1956 cited in Tardieu & Simonneau, 1998 ). Isohydric plants strongly regulate stomatal aperture. Leaf water potential of these plants is stable during the afternoon,

- 56 - Chapter II. Literature review regardless of soil water availability or vapour pressure deficit. By contrast, leaf water potential of anisohydric plants drops with decreasing soil water-content or increasing vapour pressure deficit. The divergences in the stomatal control of isohydric and anisohydric plants seem due to differences in the perception of ABA and/or hydraulic signals ( Tardieu & Simonneau, 1998 ). The regulations of gas exchange and leaf water potential in grapevine have been reviewed recently ( Chaves et al. , 2010 ; Lovisolo et al. , 2010 ; Schultz & Stoll, 2010 ). Under drought, water flow is not sufficient to compensate water losses through evapotranspiration, which results in a midday to afternoon drecrease in leaf water potential, in turn inducing a decrease of gs and A. The decrease of photosynthesis involves both stomatal and non-stomatal factors. The midday to afternoon stomatal closure is often associated with chemical (increase of ABA or sap pH) and hydraulic signals. Non-stomatal limitations, such as photoinhibition or feedback inhibition through source-sink interactions may also be responsible ( Lovisolo et al. , 2010 ). The leaf water potential and gas exchange responses to water-deficit are genotype- dependent in grapevine. Some V. vinifera genotypes have been classified as isohydric or anisohydric, depending on the control of their stomatal behaviour ( Schultz, 2003a ; Soar et al. , 2006a ; Rogiers et al. , 2009 ; Vandeleur et al. , 2009 ). However, contradictory reports appear in the literature, and a strict classification of grape cultivars as isohydric or anisohydric seems inappropriate ( Chaves et al. , 2010 ; Lovisolo et al. , 2010 ). It seems plausible that genotype- specific stomatal responses to water-deficit vary according to the particular combination of rootstock, climate (temperature and vapour pressure deficit), and intensity and duration of stress ( Chaves et al. , 2010 ). Recently, Alsina et al. ( 2011 ) showed that the rootstock induces different stomatal regulations in V. vinifera . The drought-sensitive rootstock (101-14 MGt) conferred isohydric-like stomatal behaviour, while the drought-tolerant (1103P) conferred anisohydric-like stomatal behaviour. However, the effects of rootstock on scion gas exchanges seems to be also scion-specific ( Düring, 1994 ; Iacono et al. , 1998 ) and water stress-dependent ( Iacono et al. , 1998 ; Soar et al. , 2006b ; Koundouras et al. , 2008 ). Thus, the stomatal regulation induced by the rootstock via hydraulic and chemical signals warrants further investigations.

II.3.3 Root-to-shoot signalling Drought-induced root-to-shoot signalling involves both hydraulic and chemical signals (Davies & Zhang, 1991 ; Davies et al. , 2002 ; Dodd, 2005 ), whose relative contribution is still debated ( Augé & Moore, 2002 ; Comstock, 2002 ; Neumann, 2008 ; Tardieu et al. , 2010 ). Despite controversies, both root-borne signals likely work together in water limiting- conditions work ( Wilkinson & Davies, 2002 ; Wilkinson & Hartung, 2009 ; Tardieu et al. , 2010 ) and regulate leaf gas exchange and growth ( Tardieu & Davies, 1993 ; Christmann et al. , 2007 ; Thompson et al. , 2007 ; Parent et al. , 2009 ). Chemical signals present in the xylem sap, such as ABA, precursors of ABA, cytokinins, mineral compounds, ethylene precursors ( e.g. 1-aminocyclopropane-1-carboxylic acid) or pH play a role in root-to-shoot signalling (reviewed by Schachtman & Goodger, 2008 ). Among the chemical compounds, it is widely accepted that ABA regulate the stomatal conductance of water-stressed plants ( Dodd et al. ,

- 57 - Chapter II. Literature review

1996 ; Sauter et al. , 2001 ; Davies et al. , 2002 ; Wilkinson & Davies, 2002 ; Davies et al. , 2005 ). Xylem sap ABA concentration and stomatal conductance are often correlated ( Davies et al. , 1994 ; Dodd, 2005 ). Grafting studies have been used as a tool to investigate root-shoot interactions governing a range of physiological responses, such as the role of root-sourced ABA in water-stress induced stomatal closure ( reviewed by Dodd, 2005 ). For example, mutant or transgenic lines that are deficient in or over-produce a putative root-to-shoot signal can be used in reciprocal grafting to manipulate leaf or xylem sap hormonal composition. Recent studies using reciprocal grafting of wild-type plants and ABA-deficient mutants concluded that stomatal closure under water stress is driven by ABA biosynthesis in the leaves rather than long-distance ABA transport from the root ( Chen et al. , 2002 ; Holbrook et al. , 2002 ; Christmann et al. , 2007 ; Thompson et al. , 2007 ). Dodd et al. ( 2008a , 2008b ) demonstrated that in heterogeneous soil moisture, maintenance of water uptake from, and sap flow through the root is necessary to sustain ABA transport to the shoot via the xylem sap, showing the complexity of rootstock-mediated changes in shoot ABA status ( Dodd et al. , 2008a , 2008b ; Albacete et al. , 2009 ; Dodd et al. , 2009 ; Asins et al. , 2010 ). The role of ABA controlling stomatal aperture has been demonstrated in different grapevine cultivars, in xylem sap and leaf tissue (Loveys & Kriedemann, 1974 ; Loveys, 1984a , 1984b ; Lovisolo et al. , 2002 ; Pou et al. , 2008 ; Rodrigues et al. , 2008 ). A gradient in ABA concentration decreasing away from the shoot apex has been observed ( Soar et al. , 2004 ). This gradient might be associated with ABA biosynthesis in the leaves ( Soar et al. , 2006a ) or a pH gradient ( Rodrigues et al. , 2008 ; Li et al. , 2011a ) or hydraulic conductivity gradient ( Salleo et al. , 1985 ; Lovisolo & Schubert, 1998 ) along the stems ( reviewed by Lovisolo et al. , 2010 ). Other chemical signals might trigger ABA-induced stomatal closure, such as phaseic acid ( Loveys & Kriedemann, 1974 ), cytokinins ( Stoll et al. , 2000 ), or xylem sap pH ( Li et al. , 2011a ), but further investigations are needed to confirm these hypotheses (Lovisolo et al. , 2010 ). In grapevine, water-deficit triggers both hydraulic and chemical signals ( Lovisolo et al. , 2002 ; Rodrigues et al. , 2008 ). The regulation of stomatal aperture and transpiration by root and shoot hydraulic conductance, and by drought-induced cavitation have been shown for different grapevine cultivars (Lovisolo & Schubert, 1998 ; Schultz, 2003a ; Rogiers et al. , 2009 ). However, most of the studies largely excluded rootstocks as a factor of acclimation. Even when grape rootstocks are included in research, the hydraulic and chemical measurements are never combined together under controlled conditions to clearly decipher the role of the rootstocks in drought responses ( see for examples Soar et al. , 2006b ; Alsina et al. , 2011 ).

II.3.4 Water use efficiency Improving Water Use Efficiency (WUE) in viticulture will be an important issue under climate change. Genetic variability of WUE in grapes (reviewed in Flexas et al. , 2010 ; Schultz & Stoll, 2010 ) is partly due to rootstocks ( Iacono et al. , 1998 ; Koundouras et al. , 2008 ; Pou et al. , 2008 ). In grapevines, improving WUE with partial root drying techniques alters the balance between vegetative and reproductive development and is associated with some improvement of fruit quality ( Chaves et al. , 2007 ).

- 58 - Chapter II. Literature review

A wide range of definitions of WUE have been employed, with the basic definition of WUE at the plant scale being the ratio of the rate of biomass production to the rate of plant transpiration. At the leaf scale, WUE is often determined from gas exchange measurements, termed intrinsic WUE (ratio of A/gs) or instantaneous WUE (ratio of A/E) ( Jones, 2004 ). Another important method of WUE estimation is the analyses of 13 C carbon isotope discrimination ( 13 C), which is widely used for grapes ( Schultz & Stoll, 2010 ). Carbon isotope discrimination in C 3 plants is due to the fact that Rubisco discriminate the heavier isotope of carbon, so that the ratio of 13 C/ 12 C in dry matter is somewhat lower than the corresponding ratio for CO 2 in air ( Jones, 2004 ). Water-deficit affects carbon isotope discrimination by enriching the 13 C/ 12 C ratio ( i.e. less negative 13 C) ( Farquhar et al. , 1989 ). Flexas et al. ( 2010 ) reviewed the potential targets for improving WUE under drought, for example via mesophyll conductance to improve CO 2 availability for photosynthesis without increasing gs or via improvement of carboxylation efficiency of Rubisco. However, these authors mentioned that WUE and drought tolerance were not necessary linked, giving two examples from Arabidopsis studies. An Arabidopsis mutant overexpressing the plant nuclear factor Y (NF-Y) showed higher drought tolerance although WUE was not affected ( Nelson et al. , 2007 ). In contrast, Masle et al. ( 2005 ) reported that in the Arabidopsis ERECTA mutant, WUE is increased while drought tolerance is not improved. Recently, Nilson and Assmann (2010 ) identified GPA1 ( -subunit of the Arabidopsis heterotrimeric G protein) as a negative regulator of WUE via the control of gs and stomatal proliferation. Unfortunately, the roles of rootstocks in the regulation of WUE are poorly understood in grafted-grapevine.

II.4 Crop improvement under drought The advances in the understanding of plant responses to water stress provided the impetus for compiling up-to-date reviews on plant breeding for water-limited environments ( Blum, 2011 ). In his book, Blum ( 2011 ) discusses crop improvement for drought in terms of mechanisms of drought resistance ( e.g. plant constitutive traits and plant adaptive traits), genomics technology ( e.g. marker-assisted selection and transgenic technology into breeding programs) as well as phenotyping tools and genetic resources. Finally, the combination of plant water stress physiology, plant genetics and the knowledge of the target environment were used to design an appropriate plant ideotype to be used as guide in breeding for water- limited environments. High-throughput phenotyping tools are required to study the complex genetic architecture of physiological responses to abiotic factors ( Trontin et al. , 2011 ). Genetic analysis of tolerance traits may offer many candidate genes/loci for crop improvement to water stress (Ashraf, 2010 ; Roy et al. , 2011 ). The availability of several plant genome sequences ( Feuillet et al., 2011 ), including grapevine ( Jaillon et al. , 2007 ), in concert with phenotyping/modelling tools ( Tardieu & Tuberosa, 2010 ) have improved the understanding of transcriptional behavior and open new avenues in genetic improvement of grapevines to water-deficit. Breeding genotypes that are better adapted to drought is crucial to anticipate changes in climate ( Nicotra et al. , 2010 ; Nicotra & Davidson, 2010 ; Varshney et al. , 2011 ). The large number of quantitative trait loci (QTL) mapping studies have provided an abundance of DNA

- 59 - Chapter II. Literature review markers, thus improving the efficiency of conventional plant breeding via marker-assisted selection ( Collard & Mackill, 2008 ). Methods to dissect complex multigenic traits and integrate the genetic information ( Cooper et al. , 2009 ) into ecophysiological models ( Tardieu, 2003 ) have become more common, thereby allowing the identification of loci (genes or QTL) governing variability of traits involved in stress tolerance ( Tardieu & Tuberosa, 2010 ). These modelling approaches, referred as “gene-to-phenotype”, are promising to link phenotypic consequences to changes in genomic regions via stable associations with model coefficients (Hammer et al. , 2006 ). For examplethis approach has already been used to analyze genetic variability of maize leaf growth to temperature and water stress using QTL-based modelling (Reymond et al. , 2003 ; Chenu et al. , 2009 ). Recently, the role of rootstock controlling transpiration plasticity under water-deficit was investigated using functional QTL mapping (Marguerit, 2010 ). This study identified specific-QTL in root which is involved in the control of transpiration under water-deficit. Genomics research, by generating new tools such as functional molecular markers, has opened promising ways to improve plant breeding via genomic-assisted breeding ( Varshney et al. , 2005 ). Yet, despite several reports on the identification or even validation of QTL or markers for abiotic stress tolerance, breeding superior cultivars has had limited success due to several constraints ( Varshney et al. , 2011 ). These constraints responsible of this poor success are due to i) the nature of stress ( i.e. timing, duration, intensity), ii ) the traits measured (integrators over time of many processes or mechanisms) and iii ) our understanding of stress tolerance mechanisms limits our capacity to phenotype appropriate traits ( Varshney et al. , 2011 ). The use of next-generation sequencing data to perform expression QTL (eQTL) mapping ( Majewski & Pastinen, 2011 ) will improve the understanding of how gene expression variation contributes to phenotypic variation. It will help to elucidate how plants manage to adapt to a wide variety of environmental cues ( Ingvarsson & Street, 2011 ). In the future, genetically modified (GM) plants may be a good strategy to improve drought tolerance ( Pennisi, 2008 ), and manipulation of either functional or regulatory genes has already been used successfully in this respect ( Cominelli & Tonelli, 2010 ). However, this new biotechnology is debated, and its use faces hurdles due to environmental impacts, regulatory approvals, market adoption, and public acceptance. Grapevine genetic enginnering for genome analysis or plant breeding have been reviewed ( Bouquet et al. , 2003 ). These authors questioned the usefulness of transgenic grapevine and discussed the importance of grapevine transformation in functional genomic studies.

In this PhD thesis, we have explored some of the directions mentioned above, using ecophysiological, modelling and transcriptomic approaches, in order to investigate the role of rootstock in the whole grapevine responses to water-deficit stress. This review has highlighted that the roles of the rootstock genotypes in the plasticity of root xylem development, in the regulation of stomatal aperture and in the remodelling of root transcriptome under water- deficit stress have been poorly investigated. These research directions will be investigated in the following sections of this thesis.

- 60 -

Chapter III. Control of stomata by rootstock- sourced signals under water stress: a model-based analysis in grapevine

- 61 - Chapter III. Modelling approach

III.1 Introduction Grapevine is widely spread across xeric environments, and in some countries it is cultivated with little or no irrigation because of local legislations and restricted access to water. The projected climatic changes forecast longer dry periods during summer in maritime, Mediterranean- and temperate areas in Europe, which would increases the risk of drought in the near future ( Schultz, 2000 ; Bates et al. , 2008 ). This will enhance the constraints on grapevine physiology, yield and wine quality ( Mira de Orduña, 2010 ). Alternatively or complementary to irrigation, and a more sustainable strategy to improve or balance vineyard water status is the choice of adequate scion-rootstock combinations ( Soar et al. , 2006b ). Carbonneau ( 1985 ) carried out a comprehensive classification of drought tolerance among rootstocks based on grafted-scion stomatal behaviour. However, the mechanisms underlying the role of rootstocks in the adaptation of grapevine to drought are still largely unknown. Graft-induced cross-talk between the scion ( i.e. mainly Vitis vinifera L.) and the rootstock ( i.e. American Vitis species or interspecific hybrids) under water deficit can be divided into three functional categories acting together: ( i) the genetic traits of the rootstock involved in water uptake, ( ii ) the coordination of anatomical structure and hydraulic architecture between the two genotypes and ( iii ) the genetic traits of the scion controlling water use in the shoot. In spite of the strong effects of edaphic conditions on root architecture, the number and distribution of fine roots in the soil seem mainly dependent on rootstock genotype ( Southey & Archer, 1988 ; Smart et al. , 2006 ; Comas et al. , 2010 ). Several studies report the strong responses of grape roots to soil moisture ( Comas et al. , 2005 ; Soar & Loveys, 2007 ), where drought-tolerant grape rootstocks are able to sustain new root growth during patchy-wetting events, and new root production in the deeper soil layer in response to water depletion (Bauerle et al. , 2008b ). The ability of roots to proliferate in water-rich zones may help to tolerate soil moisture deficit through hydraulic lift during low transpiration periods ( Smart et al. , 2005 ; Bauerle et al. , 2008a ). Thus, these observations might at least partially explain the maintenance of transpiration by drought-tolerant rootstocks during the drying season ( Alsina et al. , 2011 ). Unfortunately, none of these studies addressed the role of chemical signals like abscisic acid (ABA) which has been shown to be involved in sustained root growth (See reviews of Sharp et al. , 2004 ; Yamaguchi & Sharp, 2010 ) and stomatal control ( Davies et al. , 2005 ) under low soil moisture. Drought-induced root-to-shoot signalling involves both chemical and hydraulic signals (Davies & Zhang, 1991 ; Davies et al. , 2002 ; Dodd, 2005 ) but the relative contributions of each process are still debated ( Augé & Moore, 2002 ; Comstock, 2002 ; Neumann, 2008 ; Tardieu et al. , 2010 ). In spite of controversies, it is likely that both types of signals, which are generated by roots in water limiting-conditions, co-exist and co-operate together ( Wilkinson & Davies, 2002 ; Wilkinson & Hartung, 2009 ) and are involved in the regulation of leaf gas exchange and growth ( Tardieu & Davies, 1993 ; Christmann et al. , 2007 ; Thompson et al. , 2007 ; Parent et al. , 2009 ; Tardieu et al. , 2010 ). Among chemical compounds ( reviewed by Schachtman & Goodger, 2008 ), ABA is widely recognized as a major player in the regulation of stomatal conductance ( gs) in water-stressed plants ( Dodd et al. , 1996 ; Sauter et al. , 2001 ;

- 62 - Chapter III. Modelling approach

Davies et al. , 2002 ; Wilkinson & Davies, 2002 ; Davies et al. , 2005 ). Xylem sap ABA concentration [ABA] and gs are often negatively correlated ( Davies et al. , 1994 ; Dodd, 2005 ). In heterogeneous soil moisture, maintenance of water uptake from, and sap flow through the root is necessary to sustain ABA transport to the shoot via the xylem sap ( Dodd et al. , 2008a , 2008b ). The role of ABA in stomatal aperture has been demonstrated for different grapevine cultivars, by analysis of either xylem sap or leaf tissue ( Loveys & Kriedemann, 1974 ; Loveys, 1984a , 1984b ; Lovisolo et al. , 2002 ). A gradient in ABA concentration decreasing away from the shoot apex has been observed ( Soar et al. , 2004 ) which might be associated with ABA biosynthesis in the leaves ( Soar et al. , 2006a ) or a pH ( Li et al. , 2011a ) or hydraulic conductivity gradient ( Salleo et al. , 1985 ; Lovisolo & Schubert, 1998 ) along the stems (reviewed by Lovisolo et al. , 2010 ). Surprisingly, in the past most studies excluded rootstocks as a factor involved in the regulation of water transport and gas exchange in water-stressed grapevine. Grape rootstocks may influence water flow in the soil-plant-atmosphere continuum through hydraulic properties such as anatomical characteristics of xylem vessels ( Pongrácz & Beukman, 1970 ; Alsina et al. , 2011 ), the hydraulic resistance of roots and the grafting point (de Herralde et al. , 2006 ), the expression and activity of aquaporins ( Galmés et al. , 2007 ) and/or the susceptibility to embolism-induced cavitation ( Lovisolo et al. , 2008b ). The effects of rootstock on scion gas exchange seem scion-specific ( Düring, 1994 ; Iacono et al. , 1998 ; Gibberd et al. , 2001 ) and water stress-dependent ( Iacono et al. , 1998 ; Soar et al. , 2006b ; Koundouras et al. , 2008 ). Soar et al. ( 2006b ) showed inverse relationship between xylem sap ABA and stomatal conductance in seven rootstock genotypes. Likewise, Alsina et al. (2011 ) observed distinct regulation of stomatal conductance and water potentials by rootstocks. They suggested that signal perception by the scion seems non-hydraulic and argued that root growth probably explains the maintenance of transpiration by the drought-tolerant rootstock. However, neither of these studies measured root characteristics ( Soar et al. , 2006b ) nor analysed the chemical signals ( Alsina et al. , 2011 ) in order to understand the role of rootstocks in drought tolerance. High-throughput phenotyping platforms combined with modelling ( Tardieu & Tuberosa, 2010 ) can be used as a powerful tool to determine which traits confer a fitness advantage when responding to abiotic stress, and to explain trait variability among genotypes ( Tardieu, 2003 ). Moreover, ecophysiological modelling can be used to simulate the responses of plants to environmental cues ( Tardieu, 2003 ) and in their extended form can integrate structural- functional traits ( Louarn et al. , 2008 ). Models can also help in the understanding of complex underlying mechanisms which control plant phenotype, as demonstrated with the virtual fruit model ( Génard et al. , 2010 ). Several models of stomatal conductance integrating the effects of water deficit have been developed ( reviewed by Damour et al. , 2010 ). Some of these models are able to calculate quantitative phenotypic traits from environmental inputs ( Tardieu, 2003 ). They allow to distinguish the effects of genetic traits, chemical signals, hydraulic signals and they have been used to explain the regulation of stomatal conductance under water-deficit (Tardieu & Davies, 1993 ; Tardieu et al. , 1993 ; Tardieu & Simonneau, 1998 ).

- 63 - Chapter III. Modelling approach

The objective of the present work was to identify genetic differences related to the regulation of scion gas exchange in response to water-deficit between different rootstocks. Drought-sensitive and drought-tolerant rootstocks grafted with the same scion were subjected to water limitation. Structural and functional variables were evaluated ( i.e. root characteristics, water potentials, leaf gas exchanges, [ABA], and others) and were integrated into a mechanistic model of stomatal regulation ( Tardieu & Davies, 1993 ; Tardieu & Simonneau, 1998 ). The model was used to test different hypotheses with respect to the control of stomatal aperture (chemical signals, hydraulic signals and other genetic traits) exerted by rootstocks under water-deficit stress.

III.2 Materials and methods

III.2.1 Plant material and growth conditions The grapevine genotypes Vitis vinifera L. cv. Cabernet Sauvignon (CS, clone 15), Vitis riparia cv. Gloire de Montpellier (RG, clone 1030) and the hybrid Vitis berlandieri ×Vitis rupestris cv. 110 Richter (110R, clone 756) were used in this experiment. Mechanical omega grafting was performed on the following scion/rootstock combinations, including homografting: CS onto CS, CS onto RG and CS onto 110R. These rootstocks were selected with regard to their drought tolerance: RG is classified as susceptible to drought, 110R is classified as highly resistant and CS was used as a control ( Carbonneau, 1985 ). After callusing and rooting, the grafted plants were grown for one year under semi-controlled conditions in a greenhouse in 7 L pots (wrapped below with a film to avoid loss of soil). The pots were filled with exactly 1 kg of gravel for drainage and 5 kg of dry soil (clay: 155.7 ± 12.2 g kg -1 , silt: 98.3 ± 3.8 g kg -1 , sand: 745.7 ± 12.7 g kg -1 , total nitrogen: 1.5 ± 0.1 g kg -1 , total carbon: 15.1 ± 2.1 g kg -1 , organic matter: 26 ± 3.7 g kg -1 , C/N ratio: 10 ± 0.8, pH water: + -1 6.2 ± 0.1, pH KCl: 5.5 ± 0.1, total CaCO 3: 0, cation exchange capacity: 10.8 ± 0.7 cmol kg , -1 -1 -1 P2O5: 0.04 ± 0.01 g kg , K 2O: 0.065 ± 0.004 g kg , MgO: 0.15 ± 0.02 g kg , CaO: 3.3 ± 0.2 g kg -1 ). During the first year, the plants were trained to a single shoot topped when the stem reached 1.8 m length. The lateral shoots were initially removed, then left to develop after topping. The plants were supplied with a nutrient solution (see below for the composition) several times a day. During the second year, the plants were grown in a semi-controlled environment greenhouse equipped with a cooling system and 150 balances (CH15R11 CHAMPII, Obaus GmbH, Nänikon, Switzerland) with automatic recording of pot weight (Sadok et al. , 2007 ). The plants of experiment 1 (see experimental setups thereafter) were trained to two shoots and all the lateral shoots were removed throughout the experiment. In experiment 2, plants were trained to a single shoot but otherwise treated as in experiment 1. The photosynthetic photon flux density (PPFD) close to the canopy level was measured with two Quantum Sensors (LI-190, LI-COR, Nebraska, USA). The dry and wet temperatures were measured with two psychrometers, and all the data were recorded every 15 min with a data logger (Campbell 21X, Campbell Scientific Inc., Loughborough, U.K.) over the entire experimental period. The greenhouse relative humidity (RH) and vapour pressure deficit (see

- 64 - Chapter III. Modelling approach climate data in supporting information, Figure S. 1) were calculated using the procedure described by Allen et al. ( 1998 ).

III.2.2 Experimental setups The data were acquired over the course of two experiments, a drought cycle experiment (experiment 1) and a steady-state drought experiment (experiment 2). Before budburst, the pots of the two experiments were wrapped with thick polyethylene bags to prevent any water loss by direct evaporation from the soil. Each pot was watered, and after a 12 h draining period, weighed on a balance with a resolution of 1 gram. The individual weight of each pot was determined and recorded automatically every 15 min. After 4 or 5 days, the pot weight became stable and the value was recorded in order to calculate soil saturated water content

(%s).

III.2.2.1 Experiment 1: drought cycle experiment When the plants reached budburst ( growth stage E-L 4, Coombe, 1995 ), the pots were first irrigated with water to maintain soil water content (SWC) between 85 to 95% of %s until the 6th leaf emerged on the two shoots. Then, ten plants of each scion/rootstock pair were selected for homogeneity of shoot development. These plants were irrigated daily to maintain SWC between 85 to 95% of %s with the same volume (200 mL) of nutrient solution (2.5 mM KNO 3,

0.25 mM MgSO 4.7H 2O, 0.62 mM NH 4NO 3, 1 mM (NH 4)H 2PO 4, 9.1 )M MnCl 2.4H 2O, 46.3

)M H 3BO 3, 2.4 )M ZnSO 4.H 2O, 0.5 )M CuSO 4, 0.013 )M (NH 4)6Mo 7O24 .4H 2O and iron was supplied as 8.5 mg L -1 of Fe-EDDHA 5.9%) plus water (volume dependent of pot weight water loss, ~400 mL) until the beginning of the experiment, i.e. when each shoot reached 15 leaves. One week before the beginning of the drying phase, the plants were irrigated to 100% of %s (average SWC = 0.32 ± 0.04 kg H 2O/kg soil) every day. The irrigation was then stopped during 6 days (SWC reached on average a minimum of 0.097 ± 0.004 kg H 2O/kg soil). During these stages, the physiological and chemical measurements described below were conducted daily on each plant.

III.2.2.2 Experiment 2: steady-state drought experiment After budburst, the pots were first irrigated with water to maintain SWC between 85 to th 95% of %s until the 6 leaf emerged on the single shoot. Then, 25 plants of each scion/rootstock pair were selected for homogeneity of shoot development and randomly attributed to four different treatments. All the pots were irrigated daily to recover %s to 100% with the same volume of nutrient solution described above plus water. Before the beginning of the experiment, four different theoretical targets of SWC: 100% of %s (SWC = 0.32 ± 0.04 kg H 2O/kg soil) for the control, ~70% (SWC = 0.23 ± 0.03 kg H 2O/kg soil), ~55% (SWC =

0.18 ± 0.02 kg H 2O/kg soil) and ~40% (SWC = 0.14 ± 0.02 kg H 2O/kg soil) of SWC for the water-limited conditions were determined. In order to reach the target values of SWC at the same time for each treatment, water supply was decreased step by step during one week. When the SWC values of each treatment were achieved, the pots were irrigated twice a day with water in order to maintain SWC values under steady-state conditions during 3 weeks. During these steady-state drought conditions, leaf gas exchange was measured three times (on

- 65 - Chapter III. Modelling approach

3 different days) the last week of stress (see section below). The dataset was used to validate the model described thereafter.

III.2.3 Determination of leaf area and leaf plastochron index In order to non-destructively estimate the canopy area, leaves (n = 278) of different size were collected from the same scion/rootstock pairs and from additional plants not included in these two experiments. Maximal leaf length (L l) and maximal leaf width (L w) were measured, leaves were then scanned (Canon MP600, Canon Inc., Japan) and leaf area was determined using ImageJ (ImageJ, U.S. NIH, Maryland, USA). The linear equation (Eqn 1, n=529, 2 2 R =0.9912) was used to estimate LA (cm ) from measurements of L l (cm) and L w (cm) as described by Montero et al. ( 2000 ).

2 LA ( cm ) # .0 6161 L l Lw ! .0 5087 (Eqn. 1)

The day before the start of experiment 1, the leaf size (L l and L w) was determined with a ruler for every leaf of each plant. The area of each leaf removed for the water potential measurements was taken into account for the calculation of the total daily leaf area. In experiment 2, the total leaf area was determined twice a week.

To quantify the physiological age, the leaf plastochron index (LPI) was derived from L l of sequentially measured leaves along the stem ( Schultz & Matthews, 1988b ) and used to select leaves with the same LPI to make comparisons of gas exchange at the same developmental stage.

III.2.4 Estimation of root characteristics In order to estimate the root characteristics ( i.e. root length density (RLD), root length area (RLA) and mean root diameter), the entire root system of 5 plants for each scion-rootstock pair was recovered (experiment 2). These plants were grown under well-watered conditions and were destroyed before the beginning of experiment 1. The roots were removed cautiously from the soil volume (5000 cm 3), washed using a low pressure water flow and collected on several sieves with an extra sieve of 0.2 mm placed at the outflow to make sure that no fine root material was lost. After cleaning, roots were stored in 15% ethanol (v/v) at 4°C for later analysis ( Böhm, 1979 ). In order to avoid overlapping during scanning procedure, all roots were cut into 1-2 cm long segments and the entire root system was scanned by successive sub-sample analysis. The sub-sample root segments were spread onto a transparent tray into a thin layer of water. An image (400 dpi) was acquired using a flatbed A3 scanner (Epson expression 10000XL Pro, Seiko Epson Corporation, Japan) and analysed with WinRhizo software (Regent Instrument Inc., Canada) procedure ( Bouma et al. , 2000 ). The measurements involved total root length, average root diameter, length and area as a function of different root diameter classes. RLD was calculated as the ratio between root length (cm) and soil volume (cm 3) and RLA was calculated as the ratio between root length (cm) and root area (cm 2).

- 66 - Chapter III. Modelling approach

III.2.5 Leaf gas exchange measurements On each scion-rootstock pairs (n=10), leaf gas exchange was measured daily between 10.00 am and 1.00 pm during the drought cycle (experiment 1), with a portable open-system infra-red gas exchange apparatus (GFS-3000, Heinz Walz GmbH, Effeltrich, Germany). In experiment 2, the measurements were conducted three times (on 3 different days) the last week of stress between 10.00 am and 1.00 pm for each treatment (control, ~ 70, 55 and 40% of %s) and each genotype combination (n=5). In each experiment and for each biological replicate, one leaf with the same developmental stage (LPI ± 2) was clamped into gas exchange cuvette with an exposed leaf surface area of 8 cm 2. The cuvette microclimate was set to saturating PPFD of 2000 )mol m -2 s -1 at the adaxial -1 leaf surface, CO 2 concentration to 380 )mol mol , relative humidity to 60% and air temperature to 25°C. The air flow was fed through the chamber at 750 mL min -1 . Each leaf sample was allowed to adapt to the cuvette conditions for 2 min and the recording started when stable values were observed by the real-time monitoring panel.

Gas exchange parameters, i.e. net CO 2 assimilation ( A), stomatal conductance ( gs), intracellular CO 2 concentration ( Ci) and transpiration rate ( E) were calculated using the equations of von Caemmerer & Farquhar ( 1981 ). Water use efficiency was determined from the single leaf gas exchange measurements, as the ratio of A to gs, termed intrinsic water use efficiency (WUE int ) or A to E, termed instantaneous water use efficiency (WUE inst ) ( Jones, 2004 ). In addition to single leaf gas exchange measurements, the whole plant transpiration was determined from the pot water loss (using pot weight data). The transpiration rate per unit of leaf area was calculated as the ratio of whole plant transpiration to total leaf area of individual plant. Additionally, light response curves were established on each genotype combination (experiment 1) grown under well-watered conditions (n=3) using leaves with the same physiological age (LPI 10 ± 2). The cuvette microclimate was set as described above. The PPFD on the adaxial leaf surface was initially set to high light (>1500 )mol m -2 s -1 ). Then, PPFD was decreased as follows: 1250, 1000, 750, 500, 350, 250, 200, 150, 100, 60, 40, 20 and 0 )mol m -2 s -1 . Before recording the gas exchange values at each PPFD, the leaf was allowed to adapt to the prevailing light condition for 10 min.

III.2.6 Plant water relation measurements Water potentials were determined in experiment 1 with a pressure chamber (SAM Précis 2000, Gradignan, France) equipped with an electronic pressure gauge with an accuracy of 0.001 MPa ( Scholander et al. , 1965 ; Turner, 1988 ). On each scion-rootstock pairs (n=10), predawn water potential ( PD ) measurements were conducted between 03.00 and 05.00 am every day during the drying period. Leaf water potential ( leaf ) was measured daily between 10.00 am to 02.00 pm immediately following leaf gas exchange measurements. Just prior to excision with a razor blade, each leaf was wrapped in a plastic bag ( Turner & Long, 1980 ).

E Kleaf # (Eqn. 2) ! ( leaf ! PD )

- 67 - Chapter III. Modelling approach

Leaf-specific hydraulic conductance ( Kleaf ) was calculated as the ratio of transpiration rate

(E) to differences in water potential across the soil ( PD ) to leaf ( leaf ) pathway (Eqn 2) according to Clearwater et al. ( 2004 ).

III.2.7 Xylem sap collection and abscisic acid (ABA) analysis

After recording leaf , an overpressure of 0.4 MPa was applied to the same leaf and xylem sap was collected in a microcentrifuge tube. A sap volume (20-30 )l) was transferred in a second tube before being snap frozen in liquid nitrogen. The remaining sap was used to measure the pH using a micro-electrode (A 157 1M-BNC-ID, SCHOTT Instrument, SI Analytics GmbH, Mainz, Germany). Samples were stored at -80°C prior to analysis, and then freeze-dried. [ABA] and ABA-related metabolites ( i.e. ABA-glucose ester (ABA-GE), phaseic acid (PA) and dihydrophaseic acid (DPA)) abundance in xylem sap were assayed by liquid chromatography/mass spectrometry (Agilent 6410 Triple Quadrupole LC-MS/MS with Agilent 1200 series HPLC, Agilent Technologies Inc., Santa Clara, U.S.A.) using a stable isotope dilution assay. The dry samples of xylem sap were dissolved in 30µL 10% acetonitrile (v/v) containing 0.05% acetic acid (v/v). This acetonitrile solution also contained the deuterated internal standards D3-7’,7’,7’-PA, D3-7’,7’,7’-DPA, D5-4,5,8’,8’,8’-ABA-GE and D6-3’,5’,5’,7’,7’,7’-ABA, all at a concentration of 100 pg/µL. The column used was a Phenomenex C18(2) 75mm×4.5mm×5µm and column temperature was set at 40°C. The solvents used were nanopure water and acetonitrile, both added with 0.05% acetic acid (v/v). Samples were eluted with a linear 15 min gradient starting at 10% acetonitrile (v/v) and ending with 90% acetonitrile (v/v). Compounds were identified by retention times (DPA= 7.25-7.75 mins, ABA-GE= 8.25- 8.75 mins, PA= 9.0-9.5 mins and ABA= 10.5-11.0 mins) and multiple reaction monitoring mass-to-charge ratio ( m/z ) for parent and product ions of native (DPA= 281/284, ABA-GE= 425/430, PA= 279/282 and ABA= 263/269) and deuterated internal standards (DPA= 171/174, ABA-GE= 263/268, PA= 139/142 and ABA= 153/159) ( Speirs et al. , 2010 ).

III.2.8 Model description and data analyses The model used in the present work was developed by Tardieu and co-workers ( Tardieu & Davies, 1993 ; Tardieu & Simonneau, 1998 ). This model combines the effect of environmental variables ( i. e. from the soil and the atmosphere) sensed by the plant and is able to simulate five physiological variables largely controlled by the plant: g s, leaf , root water potential ( r), water flux through the plant (Jw) and [ABA] which are solved by five equations ( Tardieu, 2003 ). The model and data analyses were executed using R software ( R Development Core Team , 2010 ). The details of equations used in the simulations are described in Table S. 1. The effect of PPFD on gs was additionally considered in the present work. Light-response curves of gs normalized for each genotype (Figure S. 2) were included in the calculation of gs (Table S. 1, Eqn. 3 and 9). The functional variables of the model (Table S. 2) were obtained from the measurements conducted in experiment 1, yet the maximal soil hydraulic conductivity

(Ks max ) and r were estimated based on the set of data described above. The equations 2 and

- 68 - Chapter III. Modelling approach

4 (Table S. 1) were combined to estimate a, b and Rp at the same time. The method was transform ed to a nonlinear least-squares minimization problem that minimizes the difference between observed and calculated values. This methodology used an R software implementation from the package soma ( Zelinka, 2004 ) of the Self-Organizing Migrating Algorithm (SOMA) ( Senkerik et al. , 2010 ). The approach of this stochastic optimization algorithm can be applied to any cost-minimization problem with bounded parameter space, and is robust to local minima. The five unknowns were solved using a nonlinear least-squares minimization problem to find ABA, leaf and r that minimize the transpiration rates (J w) calculated through the equations (Table S. 1) using BB package ( Varadhan & Gilbert, 2009 ) in R software. The param eters values (Table S. 3) were estimated for each genotype and differences in the param eters were evaluated using the extra sum-of-square F-test of Fisher ( Motulsky &

Christopoulos, 2003 ). The ability of the model to predict gs and E under different water- deficit conditions was tested using the dataset from experiment 2. The goodness of fit of the model was evaluated using the root mean squared error (RMSE) and relative root mean square error (RRMSE) to compare the mean difference between simulated and observed results ( Kobayashi & Salam, 2000 ). A local sensitivity analysis was used to study the role of input variables and param eters in the model using normalized sensitivity coefficients. The sensitivity coefficients, defined as the relative variation of the output variable, were calculated by shifting one parameter by ± 20% while keeping the remaining constant (default values) (Monod et a l. , 2006 ). All statistical analyses were performed using R software ( R Development Core Team, 2010 ). The residual sum of squares variance of nonlinear reg ression was calculated according to the F-test of Fisher at the 5% level of significance to discriminate the fitted curves between genotypes. Two-way analyses of variance (ANOVA) were performed to access the effects of treatment and genotypes on measured variables. When interactions occurred, Tukey’s multiple comparison procedure was executed at the 5% level of significance. When the assumption of normality and equal variance were violated, a Kruskal-Wallis one way ANOVA was performed. Dunn’s multiple comparisons at the 5% level of significance were performed if interactions were significant. The genetic effects on model parameters were tested using F-test of Fisher at the 5% level of significance.

III.3 Results

III.3.1 Plant water status, gas exchange and ABA responses to gradual water-deficit The progress of water deficit during experiment 1 was monitored through measurements of

SWC and PD . Results did not statistically differ among the three rootstocks at any given day

(Figure III.1a & b), with PD decreasing on average down to –0.97 MPa for RG and to –1.04 MPa for CS and 110R at the end of drying phase, i.e. after 6 days of treatment (DAT). The relationship between PD and SWC was tested for all rootstock genotypes (Figure III.1c). The

- 69 - Chapter III. Modelling approach curve f itted to all data pooled was not statistically different ( F-test of Fisher) from the curves of the three genotypes, so the result of the global fitting was used in the model (Figure III.1c).

0.35 0.0 0.0 aCabernet Sauvignon b -0.2 c 0.30 Vitis riparia -0.2 110R -0.4 0.25 -0.4 O/kgsoil) -0.6 2 (MPa) -0.6 (MPa) 0.20 -0.8 PD -0.8 PD -1.0 (kgH $ 0.15 $ -1.0 -1.2 0.10 -1.4

SWC SWC -1.2 0123456 0123456 0.08 0.13 0.18 0.23 0.28 0.33 Days after treatment Days after treatment SWC (kg H O/kg soil) 2

Figure III.1: Soil and plant water relations during the drying phase (experiment 1). (a) Changes in SWC for the three rootstocks. Data are means (n = 10) ± SE. (b) Changes in the predawn leaf water potential (PD ) for the three rootstocks. Data are means (n = 10) ± SE. There are statistically significant effects (p<0.001) of the days after treatment (DAT) but no interactions between DAT and rootstocks (a, b). (c) Non linear regression (see equation in Table S. 1: List of the equations which are used in the model. The details of variables, parameters, units and values are given in the Tables S2 and S3., R 2=0.9616; p<0.0001) between PD plotted as a function of SWC for the three rootstocks during the drying phase. Each point is an individual measurement.

Impacts of water-deficit on many physiological processes in co-ordinated patterns are presented in a hierarchical clustering heatmap (Figure III.2) of normalized and log 2- transformed variables measured over the drying period and created using standard statistical algorithms ( see the methods in Eisen et al. , 1998 ). The cluster analysis identified the presence of continuous patches of colour representing three m ain groups of variables that share similar temporal patterns when water-deficit increased gradually (Figure III.2). The normalized variables in the first group ( i.e. [ABA] and

PD ) increased with a 2-fold (4 DAT) to 8-fold (6 DAT) during the progression of water- deficit. It means that non-normalized [ABA] increased while PD decreased. However, there were no statistical differences between rootstocks at any given day, even though log 2-change of [ABA] remained unchanged until 4 DAT for 110R compared to the other 2 rootstocks

(Figure III.2). In the second group of variables ( i.e. leaf , WUE int , WUE inst , sap pH, Ci, PA and DPA), leaf followed the same pattern as PD in the first group. Both WUE int and

WUE inst reached their maximum at 5 DAT with a 4-fold up-regulation, but xylem sap pH was not affected by the stress (Figure III.2). More interestingly, PA and DPA were significantly affected by the genotype (p<0.001) and the progression of water-deficit (p<0.01 and p<0.001, respec tively for PA and DPA) (Figure III.2). For the two last days of stress, the DPA level was statistically higher (p<0.001) in CS than in RG and 110R, but it did not differ between

RG and 110R. The third group of variables ( i.e. ABA-GE, Kleaf , A, E and gs) followed patterns which were opposite to the first group, with a 2-fold (4 DAT) to 8-fold (6 DAT) decrease during the progression of soil water stress (Figure III.2). However, in the drought- rd tolerant rootstock 110R Kleaf increased (Figure III.2, ~2-fold log 2 change) only the 3 DAT, then dropped as described above. Kleaf of 110R was significantly higher (p<0.05) compared to

- 70 - Chapter III. Modelling approach the drought-sensitive rootstock RG. Regarding other variables in the third group, there was no difference b etween rootstocks at any given day (Figure III.2).

Color Key

-3 -1 1 3 Value Cabernet-Sauvignon Vitis riparia 110R %PD [ABA]

%l eaf DPA WUE int WUE inst PA Sap pH

Ci ABA-GE A Kl eaf E gs 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6

Days after treatment

Figure III.2: Clustered heatmap of measured and calculated variables (n=10) during the drying phase for the three rootstocks. The values were normalized (observed value/mean value of well-watered plants), then log 2 transformed to make the pixels square and subjected to hierarchical clustering. The dendogramm at the left of the matrix indicates the degree of similarity between variables (the larger is the distance in the dendogramm, the lower is the similarity). The columns of the matrix represent the day after treatment for each genotype (Cabernet Sauvignon, Vitis riparia and 110R) and the rows each variable ([ABA]: xylem sap abscisic acid concentration, PD : predawn leaf water potential, leaf : leaf water potential, WUE int : intrinsic water use efficiency, WUE inst : instantaneous water use efficiency, Ci: intracellular CO 2 concentration, PA: xylem sap phaseic acid concentration, DPA: xylem sap dihydrophaseic acid concentration, ABA-GE: xylem sap ABA-glucose ester concentration, Kleaf : leaf- specific hydraulic conductance, A: net CO 2 assimilation, E: transpiration rate, gs: stomatal conductance). III.3.2 Physiological responses to gradual water-deficit at the single leaf- scale The effects of gradual decrease of SWC on leaf water potential, gas exchange and [ABA] were presented in Figure III.3. The results for leaf showed a strong decrease with SWC

(Figure III.3a) while [ABA] increased (Figure III. 3d) when SWC fell below 0.15 kg H 2O/kg soil. There was no significant difference between genotypes ( Figure III.3d) although [ABA]

- 71 - Chapter III. Modelling approach

of drought-tolerant 110R was lower than for the other 2 rootstocks at 0.15 kg H 2O/kg soil.

Likewise, a non rootstock-dependent relationship was observed between gs and [ABA] (Figure III.3e). Stomatal conductance was closely related to [ABA] which was roughly proportional to soil water-deficit ( Figure III.3d & e) regardless of the rootstock. In contrast, the rootstock genotypes exhibited different patterns of E and gs at a given SWC value (Figure III.3 b & c). Com pared to graft combinations with CS (moderately drought-tolerant) and RG (drought- sensitive), E and gs of CS grafted onto the drought-tolerant 110R were higher under well- watered conditions and remained at a higher level until SWC dropped below 0.18 kg H 2O/kg soil (Figure III.3b & c).

-0.4 b180 c a 3.0 ) )

-0.6 -1 -1 150 s

s 2.5 -2 -0.8 -2 2.0 120 O m O m 2 -1.0 2 (MPa) 1.5 90

leaf RMSE=0.16 RMSE=0.27 RMSE=14.9

$ -1.2 RRMSE=8.7% 1.0 RRMSE=12.5% 60 RRMSE=12.5% RMSE=0.06 RMSE=0.09 (mmol H RMSE=6.9 (mmol H

RRMSE=10.4% s RRMSE=13.1% -1.4 E RRMSE=13.1% 30 RMSE=0.08 0.5 g RMSE=0.21 RMSE=10.8 RRMSE=9.4% RRMSE=9.8% RRMSE=9.2% -1.6 0.0 0 0.10 0.15 0.20 0.25 0.30 0.10 0.15 0.20 0.25 0.30 0.10 0.15 0.20 0.25 0.30

SWC (kg H 2O/kg soil) SWC (kg H 2O/kg soil) SWC (kg H 2O/kg soil) 210 490 d RMSE=48.4 e Cabernet Sauvignon

RRMSE=31.1% ) 180 -1 420 RMSE=41.1

s Vitis riparia )

RRMSE=30.2% -2 150 -1 350 RMSE=36.6 110R RRMSE=43.5%

O m 120

280 2

(ng mL (ng 210 90

60 ABA 140 (mmol H s

70 g 30

0 0 0.10 0.15 0.20 0.25 0.30 0 70 140 210 280 350 420 490

SWC (kg H O/kg soil) ABA (ng mL -1 ) 2

Figure III.3: SWC-response curves of leaf water potential (a), transpiration rate E (b), stomatal conductance gs (c), xylem sap abscisic acid concentration [ABA] (d), and relationship between gs and [ABA] (e) of measured (symbols) and estimated based on model runs (lines) traits for the three rootstocks during experiment 1. The measured values represent the mean ± SD of 5-10 samples. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock.

The model generates quantitative phenotypic traits ( e.g. gs, E, [ABA], Figure III.3) from edaphic (SWC), atmospheric (PPFD and VPD) and genetic specific inputs. All the parameters in the present work were estimated using the data from experiment 1. The list of parameters estimated in the simulation is summarized in Table S. 3. Parameters related to soil properties

(e.g. %s, Ks max , Table S. 3) were common in the three genotypes (Table S. 3). In contrast, other parameter values were specific for a given genotype ( e.g. gsmin , Table S. 3). These parameters were estimated together from measurements conducted in experiment 1 ( i.e. water

- 72 - Chapter III. Modelling approach

potentials, gs and E using the gas exchange analyser and [ABA]). Some parameters (RLD, d and r, Table S. 3) were determined experimentally. Between rootstocks, differences between param eters were observed (Table S. 3), which may explain the genotype-dependent dose- response curves (Figure III.3).

The SWC-response curves of leaf (Figure III.3a), E (Figure III.3b), gs (Figure III.3c),

[ABA] (Figure III.3d) and the relationship between gs and [ABA] (Figure III.3e) were accurately estimated for each rootstock. The RMSE and RRMSE values, quantifying the goodness of fit, were satisfactory for most of the model runs (Figure III.3a, b, c & e), and acceptable for [ABA] (Figure III.3d inset) despite higher vari ability. Furthermore, the parameters estimated in model runs based on experiment 1 indicated specific-genetic values between rootstocks (Table S. 3). The estimated gsmax ( in Table S. 3) was significantly higher for the drought-tolerant 110R (210 mmol m -2 s -1 ) than for the other 2 rootstocks (168 and 155 mmol m -2 s -1 , respectively for CS and RG). This confirmed the observations (Figure III.3b & c) of different strategies in the control of shoot water use induced by the rootstocks.

Later during water deficit, when the SW C level decreased below 0.18 kg H 2O/kg soil, similar restrictions were observed for all rootstocks (Figure III.3b & c).

III.3.3 Scaling up from single leaf to canopy level The ability of the model to describe the responses of canopy to different water availability throughout a day was assessed. Parameter values described in above section and climate data (PPFD and VPD) from two days (DAT 1 and DAT 5) with contrasted water availability in experiment 1 were inputted into the model to predict daily fluctuation of leaf (Figure III.4e & f), root (Figure III.4g & h), E (Figure III.4i & j), gs (Figure III.4k & i) and [ABA] (Figure III.4m & n). The m odel was first validated by comparing measured and predicted whole plant transpiration (Figure III.4i & j). The model provided a broadly correct sim ulation of whole plant transpiration. The adjustment quality was very good for each rootstock in the absence of water-deficit (Figure III.4i, RMMSE = 0.13, 0.18 and 0.22, respectively for 110R, CS and RG). It reproduced the pattern of m easurements under water stress, although simulated transpiration was underestimated (Figure III.4j, RMMSE = 0.86, 0.86 and 0.89, respectively for 110R, CS and RG). Considering the validation of canopy transpiration, the m odel was further used to simulate the other variables related to plant water status. During the day, regardless of treatment and genotype, leaf (Figure III.4e & f) and root (Figure III.4g & h) decreased to a minimum value around 3 pm and increased thereafter. Stomatal conductance (Figure III.4k & l) displayed diurnal variations with contrasted p atterns between well-watered (Figure III.4k) and water- stressed (Figure III.4l) conditions. The simulated [ABA] changed with time of day (Figure III.4m & n), was always higher under water stress (Figure III.4n), and inversely related to transp iration (Figure III.4i & j).

- 73 - Chapter III. Modelling approach

Well-watered Water-stressed

) 1200 1200 -1 a b s -2 900 900

molm 600 600 ' (

300 300

0 0 PPFD c d 1.5 1.5

1.0 1.0 (KPa)

0.5 0.5 VPD

0.0 0.0 e f -0.2 -0.2 -0.4 -0.4 (MPa) -0.6 -0.6

leaf -0.8 -0.8 $ -1.0 -1.0 g h -0.2 -0.2 -0.4 -0.4 (MPa) -0.6 -0.6 &

-0.8 -0.8 root $ -1.0 -1.0

) i RMSE=11.1 j RMSE=20.5 -1 160 RRMSE=18.1% RRMSE=86.1% 160 h RMSE=19.5 -2 RMSE=8.2 120 RRMSE=21.5% RRMSE=88.8% 120 RMSE=25.9 O m O RMSE=8.5 2 80 RRMSE=12.9% RRMSE=85.5% 80

(gH 40 40 E 0 0

k l ) 160 160 -1 s -2 120 120

80 80

40 40 (mmol m s g 0 0 m n ) 600 600 -1 Cabernet Sauvignon 450 Vitis riparia 450

(ng mL (ng 300 110R 300

150 150 [ABA] 0 0 0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21

Time of day

Figure III.4: Daily measurements of photosynthetic photon flux density (PPFD, a,b) and vapour pressure deficit (VPD, c,d) measured during the drying cycle (experiment 1) at two different day (well-watered left panel and water-stressed right panel conditions). The lines represent the model simulation of leaf water potentials (e,f), root water potentials (g,h), transpiration rate E (symbols represent measured values [circle: Cabernet Sauvignon, triangle: Vitis riparia , square: 110R], mean (n=5) ± SE, i,j), stomatal conductance gs (k,i) and xylem sap abscisic acid concentration [ABA] (m,n) during the day for the three rootstocks under well-watered left panel and water-stressed right panel conditions. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock (i,j).

Under well-watered conditions, the drought-tolerant 110R showed higher g s (Figure III.4k) and E (Figure III.4i), especially compared to the drought-sensitive RG. For RG, leaf (Figure

- 74 - Chapter III. Modelling approach

III.4e) and root (Figure III.4g) were relatively stable during the afternoon. The lower gs observed in RG (Figure III.4k) therefore allowed leaf to remain relatively constant due to less open stomata, which balanced the evaporative demand. Stomatal conductance ( gs) from 9 am to 8 pm showed large difference between rootstocks whereas [ABA] was fairly similar (Figure III.4m). Under water-stressed conditions, the difference betw een genotypes were lower (Figure III.4j,l), except for [ABA] (Figure III.4n). The simulation showed that the drought-tolerant 110R exhibited the lowest leaf (Figure III.4f) and root (Figure III.4h), the highest E (Figure III.4j) and the lowest [ABA] (Figure III .4n).

III.3.4 Model validation in long term water-deficit conditions The fitness of the model was tested with the same genotypes using data collected during an independent experiment (experiment 2, see material and methods) which was not used for parameterization. The comparisons between observed and simulated E (Figure III.5a) and gs (Figure III.5b) at the single leaf level showed that the m odel often overestimated these variables, especially at low water availability.

3.5 200 ) ) Cabernet Sauvignon (CS) Cabernet Sauvignon (CS) -1 -1 a b

Vitis riparia (RG) s Vitis riparia (RG) s

3.0 -2 -2 110R 110R 1:1 150 1:1 O m O O m O

2.5 2 2

2.0 100 (mmol H (mmol H 1.5 s

1.0 50 CS: RMSE=0.66; RRMSE=41.8% CS: RMSE=37.1; RRMSE=41.8% 0.5 RG: RMSE=0.56; RRMSE=38.4% RG: RMSE=30.4; RRMSE=37.4% 110R: RMSE=1.18; RRMSE=80.7% 110R: RMSE=63.7; RRMSE=78% Simulated E Simulated 0.0 g Simulated 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 50 100 150 200

Observed E (mmol H O m -2 s -1 ) Observed g (mmol H O m -2 s -1 ) 2 s 2

Figure III.5: Model validation between observed and simulated E (a) and gs (b) values using data from experiment 2 (see materials and methods) collected at different soil water content during two days on three rootstocks (mean ± SE, n=10). The lines represent the 1:1 relationship between observed and simulated values. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock.

Considering possible differences between datasets used for model parameterisation (experiment 1, short term water-deficit) and datasets used for model validation (long term water-deficit), others factors may be involved in the regulation of gas exchange that have not been taken into account during model development (root growth plasticity, cavitation, etc). Considering that root growth is affected by long term water-deficit, the model prediction was largely improved (Figure III.6) by including a function describing the decrease of RLA 6.0932×SWC param eter in function of SWC (RLA stress = RLA×0.139e ).

- 75 - Chapter III. Modelling approach

3.5 200 ) ) Cabernet Sauvignon (CS) Cabernet Sauvignon (CS) -1 -1 a b

Vitis riparia (RG) s Vitis riparia (RG) s

3.0 -2 -2 110R 110R 1:1 150 1:1 O m O O m O

2.5 2 2

2.0 100 (mmol H (mmol H 1.5 s

1.0 50 CS: RMSE=0.33; RRMSE=20.9% CS: RMSE=21.9; RRMSE=24.6% 0.5 RG: RMSE=0.3; RRMSE=20.4% RG: RMSE=19.7; RRMSE=24.3% 110R: RMSE=0.5; RRMSE=34% 110R: RMSE=25.9; RRMSE=31.7% Simulated E Simulated 0.0 g Simulated 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 50 100 150 200

Observed E (mmol H O m -2 s -1 ) Observed g (mmol H O m -2 s -1 ) 2 s 2

Figure III.6: Model validation taking into account the decrease of root length area (Table S. 3, RLA) parameter as a function of SWC for plants experiencing long-term water stress. Observed and simulated E (a) and gs (b) values using data from experiment 2 (see materials and methods) were collected at different soil water content during two days for the three rootstocks (mean ± SE, n=10). The lines represent the 1:1 relationship between observed and simulated values. The root mean square error (RMSE) and relative root mean square error (RRMSE) are indicated for each rootstock.

III.3.5 Contribution of genetic parameters to stomatal regulation A sensitivity analysis was performed using normalized sensitivity coefficients (Figure III.7) to assess the role of model parameters in the response of physiological variables ( i.e .

leaf , root , gs and [ABA]) and rootstocks. With respect to rootstocks, the pattern of sensitivity did not differ between genotypes, except for RG with E and gs which substantially responded to parameters ! and ABA a. The parameters , ABA a, RLA and Ks max , which are related to stomatal control, ABA synthesis, root characteristics and soil hydraulic conductivity, respectively, have remarkable effects on the profile of each variable. A 20% variation in ( i.e. maximum stomatal conductance) led to the same direction of changes in leaf , root , E and gs but had an opposite effect on [ABA] (Figure III.7).

ABA biosynthesis seem s to be mainly affected by the ABA a parameter, which is the rate of ABA synthesis in the root system (see eqn. 4 Table S. 1). When this parameter was increased by 20%, [ABA] increased in the s ame range while E and gs decreased, especially for RG

(Figure III.7). When shifting RLA and Ks max , E and gs followed in the same direction while

leaf , root and [ABA] differed in their response. In addition, soil properties (Ks max ) as well as root characteristics (RLA) had a strong impact on [ABA] (Figure III.7). The virtual profile was m odestly impacted by changes in other parameters. Overall, a shift in a single parameter could strongly affect a plant trait, water status and chemical signals ( i.e. , ABA a, RLA and

Ks max ).

- 76 - Chapter III. Modelling approach

Color Key

-1 0 1 Value

gs min * Stomatal control ) (

ABA a ABA synthesis ABA b

Plant hydraulic conductivity Rp d Root characteristics r RLA

Soil hydraulic conductivity ks max CS CS CS CS CS RG RG RG RG RG 110R 110R 110R 110R 110R %leaf %root E gs ABA

Figure III.7: Heatmap of sensitivity coefficient (SCs), which is defined as the ratio between the relative variation of the predicted variable and the relative variation of the parameter (± 20%). The SCs were computed to decipher the influence of 11 parameters (rows, meanings of parameter are given in Table S. 3) on the 6 functional variables (columns, i.e. lea f water potential [ leaf ], root water potential [ root ], transpiration rate [E], stomatal conductance [g s] and ABA concentration in the xylem sap [ABA]) for the 3 rootstocks (columns, 110R, CS and RG). In order to simplify the figure, the effects of soil water content (SWC) on SCs were summarised by using median value of SCs calculated with different SWCs. The biological and physical processes regulated by parameters are indicated on the left of the heatmap. III.4 Discussion

III.4.1 Rootstock-specific stomatal control under fluctuating water-deficit in grafted-scion Plants respond to water stress through numerous adjustments of physiological and molecular traits ( Ahuja et al. , 2010 ). These adjustments depend on stress duration, and on the extent and rate of water deficit ( Bray, 1997 ). In the present study, the response of leaf gas exchange and ABA metabolites ( i.e. PA and DPA) to SWC differed significantly between Cabernet Sauvignon scions grafted onto different rootstocks. These results confirm the influence of rootstock genotypes on the gas exchange behaviour of grafted scion which had been previously observed under well-watered conditions ( Düring, 1994 ) and at low water availability ( Carbonneau, 1985 ; Iacono et a l. , 1998 ; Soar et al. , 2006b ). CS grown either on their own roots (moderately drought tolerant) or

- 77 - Chapter III. Modelling approach

grafted onto 110R (highly drought tolerant) showed high E and gs under well-watered conditions and moderate water-deficit whereas CS on RG (low degree of drought tolerance) exhibited the lowest gs and E (Figure III.3b & c). These results contrast with those of Gibberd et al. ( 2001 ) who observed no significant effect of rootstock on scion gas exchange when plants were kept well-w atered. However, in our study, when more stressful conditions were reached (SWC < 0.15 kg H 2O/kg soil, PD > -0.6 MPa, Figure III.3b & c and Figure III.4j & i), all rootstocks tended to have equivalent rates of gas exchange. A gradual increase of

[ABA] with decreasing SWC (Figure III.3d) and a single relationship between [ABA] and gs (Figure III.3e) were observed with no difference between rootstocks. Moreover, the transient decrease in [ABA] from 6-10 am regardless of SWC and rootstock has also been observed by other authors ( Tardieu & Simonneau, 1998 ). These findings show that grapevine stomata respond to ABA when the soil dries (Fig. 2 & Figs. 3c & d; Loveys & Düring, 1984 ; Correia et al. , 1995 ), as well as when leaf ( Fig. 3a; Schultz, 2003a ) and hydraulic conductance (Fig. 2; Salleo et al. , 1985 ; Alsina et al. , 2011 ) decrease. Water-deficit induced no change in xylem sap pH (Figure III.2) as observed by Li et al. ( 2011a ) for a Vitis hybrid. O ur findings confirm the observation on numerous plant species that stomatal control depends on soil and that ABA contributes to this control (Fig. 3; Davies & Zhang, 1991 ; Tardieu & Simonneau, 1998 ; Davies et al. , 2002 ; Davies et al. , 2005 ). Additionally, Figure III.4 supports the conclusion that root water uptake, sap flow and [ABA] are closely related ( Dodd et al. , 2008a , 2008b ), although whole-plant hydraulic conductance and root-to-shoot flow of ABA see med to be independently affected by water stress in grapevine ( Lovisolo et al. , 2002 ).

Under well-watered conditions, higher E and gs (Figure III.3b & c and Figure III.4i & k) when CS was grafted onto the drought-tolerant 110R, associated with the fact that [ABA] did not differ between genotypes and remained nearly constant during the day (Figure III.4m), suggests that hydraulic signals played a role in regulating stom atal aperture of the other genotypes. Given that the evaporative demand was similar for all genotypes, these observations, in combination with the measurements of root characteristics (Table S. 3), indicate that water uptake and water supply to the scion differed between rootstocks. Under well-watered and m oderate water-deficit conditions, higher RLA and higher root density (parameter d, Table S. 3) were the most likely reasons why 110R provided higher hydraulic supply to the scion (as shown by the increase of Kleaf 3 DAT, Figure III.2). This resulted in higher plant E (Figure III.3b and Figure III.4i & k) than for the other rootstock genotypes. Aquaporin activities are likely to be involved in balancing water relations during water-deficit (Vandeleur et al. , 2005 ). Interestingly, the genes expression of several PIP and TIP aquaporins have shown up-regulation at early stage of water-deficit in the roots of 110R (Galmés et al. , 2007 ), which may contribute to sustain hydraulic supply to the scion under moderate water stress. Under water-deficit, stomatal aperture was mainly controlled by [ABA] (Figure III.3d & e) regardless of genotype. Overall, the rootstock-specific sto matal control in grafted-scion is likely to be driven by a co-ordinated process including root traits (RLA and d parameters, Table S. 3) and hydraulic supply under non-limiting conditio ns which became associated with [ABA] under stress.

- 78 - Chapter III. Modelling approach

III.4.2 Contribution of genetic parameters and water stress duration for stomatal regulation The values of parameters estimated in the present work (Table S. 3) were often in agreement with those found in the literature, reported mainly in earlier studies of Tardieu and co-workers ( Tardieu et al. , 1992a ; Tardieu & Davies, 1992 ; Tardieu et al. , 1992b ; Tardieu & Davies, 1993 ; Tardieu et al. , 1993 ). However, several genetic param eters differed due to model specificities and experimental conditions. The leaf boundary layer conductance (g a) was estimated for each genotype, thereafter kept constant because no genetic difference was observed. Although this value is higher than those measured in the vineyard ( Daudet et al. , 1998 ), it is in agreement with others ( Brenner & Jarvis, 1995 ), maybe due to greenhouse conditions and the presence of a cooling system ( Martin et a l. , 1999 ). The residual stomatal -2 -1 conductance to H 2O vapour (g smin ) was in the same range as the value (15.23 mmol m s ) determined by Schultz ( 2003b ) for other grape cultivars. Regarding resistance to water flow

(R p), our estimations agreed reasonably well with those determined for other grape rootstocks (de Herralde et al. , 2006 ; Alsina et al. , 2011 ). Our model parameters (Table S. 3) and sensitivity analysis (Figure III.7) highlighted the im portance of soil properties ( i.e. soil hydraulic conductivity [Ks max ]), as well as root characteristics for [ABA], E, gs, leaf and root . Soil physical properties may affect plant responses to drying soil ( Whitmore & Whalley, 2009 ) such as ABA signalling and indirectly stom atal regulation ( Tardieu et al. , 1992c ; Dodd et al. , 2010 ) and our model simulation showed that an increase of Ks max induced a decrease in [ABA] and an increase in leaf , root

(less negative), E and gs or vice versa . However, roots also exhibit a high degree of physiological and morphological plasticity to environmental stimuli ( Osmont et al. , 2007 ; Walter et al. , 2009 ). Changes in root distribution induce modifications in the soil volume subject to water uptake, thus causing soil moisture heterogeneity ( Sharp & Davies, 1985 ). When soil moisture is heterogeneous, this may be associated with higher root biomass in drying soil which can inhibit leaf growth and increase leaf ABA concentration ( Martin- Vertedor & Dodd, 2011 ). Additionally, hydraulic lift (internal redistribution of water) from roots in wet soil to those in dry soil may extend the lifespan of roots and help to support short- term drought resistance in grape ( Smart et al. , 2005 ; Bauerle et al. , 2008a ). During the sensitivity analysis conducted for this work, it was found that an increase in the RLA param eter (Figure III.7) caused a decrease in [ABA] and an increase in E and gs. The significantly higher RLA values (Table S. 3) found for the drought-tolerant 110R m ay explain its higher E and gs observed at high and moderate SWC compared to the other genotypes (Figure III.3a & b). It provides information on the rela tive importance of the root system during water-deficit. The ability of drought-tolerant rootstocks both to generate new roots in deep soil layers ( Bauerle et al. , 2008b ) and to increase whole root system hydraulic conductance during summ er drought ( Alsina et al. , 2011 ) may be partly responsible for the maintenance of E compared to drought-sensitive rootstocks. When model parameters estimated from a short-term drought cycle (experiment 1, Figure

III.1) were used to simulate E and gs in a long-term steady-state drought trial (Figure III.5), the predicted variables were always overestimated at moderate and high water-deficit for all

- 79 - Chapter III. Modelling approach genotypes. This unsuccessful validation using different stress conditions highlights several weaknesses of the m odel. Some factors involved in the regulation of gas exchange were not taken into account such as i) root dynamics and turnover after long stress duration, ii ) water- deficit induced cavitation and embolism repair and iii ) roles of chemical signals other than ABA. Despite growth maintenance of specific root zones ( Sharp et al. , 2004 ) and deeper root growth ( Sharp & Davies, 1985 ) in drying soil, total root biomass and the lifespan and growth of fine roots m ay be depressed by water-deficit ( Meier & Leuschner, 2008 ; Olesinski et al. , 2011 ). Additionally, soil drying induces several physical constraints (increased bulk density and com pactness), impacting root growth in a number of interacting ways (soil oxygen content and nutrient availability) ( Whitmore & Whalley, 2009 ). In our long-term drought experim ent (experiment 2), only gas exchange data and previously estimated plant parameters were used in the prediction. Root growth dynamics were neglected because the model did not include root growth in the equation. However, when the RLA parameter was decreased

(Figure III.6), the prediction of E an d gs at moderate and high water-deficit was improved in all genotypes, suggesting that a decline in active root quantity may have played a role during long stress periods. An improvement would be to develop a new module taking into account the time-dependent dynamic of root growth, in order to integrate root plasticity and ontogeny in the control of stomatal aperture under water-deficit. These kinds of models have been developed either to simulate 3D-root water uptake (Doussan et al. , 2006 ; Javaux et al. , 2008 ; Draye et al. , 2010 ) by combining soil and root h ydraulic properties, or to study root growth (Dupuy et al. , 2010 ) and functional architecture ( Pierret et al. , 2007 ). The second possible reason for the discrepancy between model prediction and observed values of gs and E may have been embolism-induced cavitation which can occur under water- deficit ( Tyree & Sperry, 1989 ). It may strongly down-regulate hydraulic conductivity in grape rootstocks ( Lovisolo et al. , 2008b ) and is closely linked to stomatal control in several species includ ing grapevine ( Salleo et a l. , 2000 ; Cochard et al. , 2002 ; Sperry et al. , 2002 ; Schultz, 2003a ; Arango-Velez et al. , 2011 ). So far no parameter related to cavitation has been integ rated in the model but recent quantification of losses in hydraulic conductivity as a function of water potential may allow to do so in the future ( Zufferey et al. , 2011 ). The third limitation of the model is the possible role of chemical signals other than ABA, such as ABA conjugates ( i.e. ABA-GE as quantified in the current study) or others in the regulation of leaf stomatal conductance ( Schachtman & Goodger, 2008 ). However, these have not been sufficiently studied and quantified to include them into the model and ABA seems to play the dominant role ( Schachtman & Goodger, 2008 ).

III.4.3 Modelling to improve genetic selection of grape rootstocks Breeding genotypes that are better adapted to drought is crucial to mitigate changes in climate ( Varshney et a l. , 2011 ). Yet, despite several reports on the identification and validation of QTL or markers for abiotic stress tolerance, the breeding of superior cultivars has had limited success. Varshney et al. ( 2011 ) identified several constraints responsible of this poor success such as i) the nature of stress ( i.e. timing, duration, intensity), ii ) the traits measured (integrators over time of many processes or mechanisms), iii ) and our limited

- 80 - Chapter III. Modelling approach understanding of stress tolerance m echanisms reduces our capacity to phenotype appropriate traits. A modelling approach allows the dissection of processes, and may help to identify adequate traits involved in drought responses. Methods to dissect complex multigenic traits and integrate the genetic information ( Cooper et al. , 2009 ) into ecophysiological models (Tardieu, 2003 ) have gained importance, thereby allowing the identification of loci (genes or quantitative trait loci [QTL]) governing variability of traits involved in stress tolerance (Tardieu & Tuberosa, 2010 ). This modelling approach, referred as “g ene-to-phenotype” is promising to link phenotypic consequence to changes in genomic regions via stable associations with model coefficients ( Hammer et al. , 2006 ). This approach has already been used to analyze genetic variability of m aize leaf growth in response temperature and water stress ( Reymond et al. , 2003 ; Chenu et a l. , 2009 ). In the present work, the model allowed the assessment of genetic responses and the identification of genotype-dependent parameters in different grafted-grape cultivars. It should be used to analyse a larger genetic variability in different scion-rootstock pairs. Overall, this model and the phenotyping platform allow the identification of genotype-dependent parameters, and of co-ordinated processes between root traits, hydraulic and chemical signals controlling stomatal aperture in grafted-grapevines.

In conclusion, the model correctly simulated the concomitant reductions in leaf-specific gs and E and increase of [ABA] observed for all scion-rootstock pairs. Moreover, it successfully reproduced the patterns observed, which showed that gs and E were higher under well- watered conditions and better maintained at moderate soil water-deficit in the drought-tolerant genotype than in the other genotypes. On the basis of model analyses, the genotype-specific stomatal regulation was associated with constitutive plant traits (root parameters) and interrelated co-operation between hydraulic and ABA regulation. Under well-watered and moderate stress conditions, the higher gs and E observed in the drought-tolerant 110R than in the other genotypes were dissected by the model. The sensitivity analysis demonstrated that root parameters strongly influenced stomatal regulation, thus suggesting that higher gs and E observed in the drought-tolerant 110R are related to root traits.

- 81 - Chapter III. Modelling approach

III.5 Supporting informations

1200 A

) 1000 -1 s -2 800

mol m mol 600 '

400 PPFD ( PPFD 200

0 34 B 32

Air temperature (°C) temperature Air 20

18 3.0 C 2.5

2.0

1.5

VPD (KPa) VPD 1.0

0.5

0.0 180 181 182 183 184 185 186 Julian day

Figure S. 1: (A) Photosynthetic Photon Flux Density (PPFD), (B) air temperature and (C) Vapour Pressure Deficit (VPD) measured every 15 min during the drying phase. Mean (n=2) ± SD.

- 82 - Chapter III. Modelling approach

Cabernet Sauvignon Vitis riparia 110R A B C 1.0

0.8 s

0.6

0.4 Normalized g

0.2

0.0 0 00 00 00 0 00 00 00 0 00 00 00 5 10 15 5 10 15 5 10 15

-2 -1 PPFD ( 'mol m s )

Figure S. 2: Light-response curves as a function of PPFD of normalized stomatal conductance ( gs) (observed value/maximum value) for Cabernet Sauvignon (A), Vitis riparia (B) and 110 Richter (C) rootstocks. Each point is a mean of 3 measurements made on different plants (n=3).

- 83 - Chapter III. Modelling approach

Equations

s+n ! G, / - c pa VPDga Jw1 # (1) (-s / +"g a /gs , .

! leaf Jw2 # r (2) R p

g ssat # gsmin / exp +-ABA. ! exp + leaf ,, (3)

! ABA -ABA . # a r (4) J w3 / ABAb

s ! r Jw4 # (5) Rsp

ln d 2 r/ 2 R # + , (6) sp 4 4 Ks +% , RLA

2 I 1 m Y p F V L m L C A !Ar S L G C A !Ar S W L 1 ks +% , # ks D T J1 ! 1!D T Z m # 1! (7) max DA ! A T G DA ! A T W n E s r U L G E s r U W L KL H X [L

1 C 1 S n m DC % s ! %r S T DD T !1T DE % ! %r U T E U 1 # m # 1! (8) s n

C C c SS D C n S T g s # gssat Da / bD1! expD! TTT (9) E E E b UUU

Table S. 1: List of the equations which are used in the model. The details of variables, parameters, units and values are given in the Tables S2 and S3.

- 84 - Chapter III. Modelling approach

Related Variables Label Unit equation

-2 -1 Water flow (1, 2, 4, 5) Jw mmol m s

-2 -1 Photosynthetic photon flux density (1, 9) n )mol m s

Vapour Pressure Deficit (1) VPD KPa

-2 -1 Stomatal conductance (1, 9) gs mmol m s

Root water potential (2, 4, 5) r MPa

Leaf water potential (2, 3) leaf MPa

-2 -1 Stomatal conductance at saturated light (3, 9) gssat mmol m s

Xylem sap ABA concentration (3, 4) [ABA] $g )l-1 Soil water potential ~ Predawn leaf (5, 8) s MPa water potential ( PD ) Resistance to water flow in the soil and (5, 6) R mmol m -2 MPa -1 s -1 in the interface sp Soil hydraulic conductivity at the mean (6, 7) Ks( %) mmol MPa -1 s -1 soil water content

Soil water content at any time (7, 8) % Kg H 2O/Kg soil

Table S. 2: List of the functional variables which are used in the model. The related equations (see Table S. 1), the labels and the units are given for each variable. The units have been transformed in the program for calculation.

- 85 - Chapter III. Modelling approach

Related Genetic values Parameters Labels Units equations CS RG 110R Density of dry air -a Kg m -3 1.2 -1 -1 Specific heat capacity of air cp J Kg °C 1013 Latent heat of vaporisation (1) ( J Kg-1 2.45×10 -6 Psychrometric constant " KPa °C -1 0.063 -1 Leaf boundary layer conductance ga m s 0.097 -1 Resistance to water flow in the plant (2) Rp MPa s m 0.063 0.025 0.034 Minimum stomatal conductance g mmol m -2 s -1 19.1 13.2 18.8 (3) smin Maximum stomatal conductance mmol m -2 s -1 167.6 155.1 209.9 ! )l $g-1 -0.005 -0.012 -0.006 (3) MPa -1 -1.151 -0.115 -1.164 Fitted parameters mmol $g m -2 s -1 )l- ABA 170 207 164 (4) a 1 -2 -1 ABA b mmol m s 0.211 0.398 0.444 Half the mean distance between d mm 8.13 8.15 7.11 neighbouring root (6) Mean radius of the root r mm 0.193 0.215 0.23 Root Length Area RLA m m -2 788 744.7 845 Soil water content at field water % 0.32 capacity s (7, 8) Kg H O/Kg soil Soil water content when 2 % 0.056 transpiration ratio = 0.1 r Maximum soil hydraulic (7) Ks m2 s -1 MPa -1 1.22×10 -10 conductivity max (7) p No unit 0.262 (7, 8) n No unit 2.141 (8) MPa -1 4.474 Fitted parameters a No unit 0.072 0.198 0.131 (9) b No unit 0.889 0.743 0.87 c No unit 0.004 0.006 0.004 Table S. 3: List of the parameters estimated in the parameterization. The related equations (see Table S. 1), the labels and the units are given for each parameter. The estimated values are given for each rootstock genotypes. The units have been transformed in the program for calculation.

- 86 -

Chapter IV. Xylem development and hydraulic conductivity are differentially affected in roots of drought-sensitive and drought-tolerant grape rootstocks

- 87 - Chapter IV. Plant growth and hydraulic properties

IV.1 Introduction Plant growth and biomass allocation are two fundamental processes influenced by water deficit stress ( Hsiao, 1973 ; Chaves et al. , 2003 ). One of the first processes affected by water deficit is leaf growth. Many physiological mechanisms potentially affecting leaf in response to water stress have been investigated, including the temporal and spatial patterns of leaf growth ( reviewed by Granier & Tardieu, 2009 ) or the roles of ABA via hydraulic and non- hydraulic processes ( reviewed by Tardieu et al. , 2010 ). The allocation of biomass to different plant organs depends on species, ontogeny and environmental conditions ( Poorter & Nagel, 2000 ). The effect of water-deficit on the fraction of biomass alloca ted to leaves is small compared to the effect on relative growth rate ( Poorter & Nagel, 2000 ). Differences in relative growth rate (RGR) and its compone nts between higher plants have been related to different adaptation patterns to the environment ( Lambers & Poorter, 1992 ). However, few authors have investigated the causes of variation in RGR under water-deficits ( Poorter & Nagel, 2000 ; Galmés et al. , 2005 ). Poorter and Nagel ( 2000 ) attributed the decrease in RGR to drop in the net assimilation rate (NAR, daily net rate of carbon acquisition), and to a lesser extent to a decrease in specific leaf area (SLA). Comparing eight Mediterranean species, Galmés et al. (2005 ) showed that decreases in RGR caused by water-deficit was m ainly explained by decreases in SLA, but strongly depended on the species and their growth form. In woody perennial plants, the decrease in RGR was also accompanied by decreases in NAR ( Galmés et al. , 2005 ). Although plant growth rates are generally reduced when soil water supply is limited, shoot growth is often more inhibited than root growth ( Sharp & Davies, 1989 ). In som e cases, the absolute root biomass of plants in dry soil may increase relative to that of well-watered controls ( Sharp & Davies, 1989 ). Biomass allocation is known to shift from leaves to roots under drought conditions ( Poorter & Nagel, 2000 ). Shoot growth inhibition in response to water lim itation may extend the period of water availability in the soil by limiting the transpiring leaf area ( Neumann, 2008 ). These shifts could be seen as adaptive mechanisms to water-deficit periods, as they enable the plant to capture more of the resources that most strongly limit plant growth ( Poorter & Nagel, 2000 ). Few studies have examined the possible links between growth and biomass allocation associated with rootstock hydraulic conductance in grafted-plants ( i.e. Solari et al. , 2006 ). The studies on water relations of grafted plants have focussed on partitioning of the hydraulic path to decipher which of the scion or rootstock play the largest ro le in restricting water flow (Cohen & Naor, 2002 ; Cohen et al. , 2007 ). The effects of grafting on the size of the xylem conduits and hydraulic properties have been mainly investigated in fruit trees such as apple (Atkinson et al. , 2003 ; Bauerle et al. , 2011 ), kiwifruit ( Clearwater et al. , 2004 ), olive ( Gascó et al. , 2007 ; Trifilò et al. , 2007 ), peach ( Solari & DeJong, 2006 ) and sweet cherry ( Gonçalves et al. , 2007 ) using rootstocks conferring different scion growth potential. Regarding grape rootstocks, the specific genotype may influence whole grapevine hydraulic conductivity through the anatomical characteristics of xylem vessels ( Pongrácz & Beukman, 1970 ; Alsina et al. , 2011 ), the resistance of roots and/or grafting point ( de Herralde et al. , 2006 ) and/or the susceptibility to embolism-induced cavitation ( Lovisolo et al. , 2008b ). Alsina et al. ( 2011 )

- 88 - Chapter IV. Plant growth and hydraulic properties showed that the graft union represents a large proportion of the total root resistance to water transport, but no differences were found between drought-sensitive and drought-tolerant rootsto cks grafted with the same scion. Additionally, the hydraulic conductivity of root segments (Kh r) increased from wet to dry season regardless of the rootstock ( Alsina et al. , 2011 ). Generally, stom atal conductance is considered to be the major control point for limiting water loss, while hydraulic resistance at the soil-rhizosphere interface, root and stem xylem properties, and root-to-shoot ratio can also significantly limit water supply versus demand (Sperry et al. , 2002 ). When plants are exposed to drought, the water flow is modified. Several studies consistently report a decline in hydraulic conductivity during water stress ( Rieger, 1995 ; Sperry et al. , 2002 ; Trifilò et al. , 2004 ). Although xylem hydraulic conductivity (Kh) and modifications of xylem conduits have been studied in grapevine, mainly in shoots, little information is available on anatomical plasticity of xylem conduits in different rootstocks under water-deficit. Furthermore, it is known that rootstocks alter dry matter partitioning and the vigour of the scion ( Tandonnet et al. , 2010 ), but few studies have investigated biomass allocation within different grafted grape combinations under water-deficit ( Qi et al. , 2007 ). The objectives of this study were i) to investigate growth and biomass allocation in both rootstock and scion portions of different genotype combinations under contrasting water supply, ii) to investigate root xylem plasticity theoretical (Kh theo using Hagen-Poiseuille law) and directly measured hydraulic conductivity under water-deficit.

IV.2 Materials and methods

IV.2.1 Plant material and growth conditions Vitis vinifera L. cv. Cabernet Sauvignon (CS, clone 15) was hetero-grafted onto a drought- sensitive ( Vitis riparia cv. Gloire de Montpellier [RG], clone 1030) or a drought-tolerant (Vitis berlandieri x Vitis rupestris cv. 110 Richter [110R], clone 756) rootstock ( Carbonneau, 1985 ). After callusing and rooting, the grafted plants were grown for one year in a greenhouse in 7 L pots (wrapped below with a film to avoid loss of soil). The pots were filled with exactly 1 kg of gravel for drainage and 5 kg of dry soil (clay=155.7±12.2 g kg -1 , silt=98.3±3.8 g kg -1 , sand=745.7±12.7 g kg -1 ). Soil analyses showed total nitrogen=1.5±0.1 g kg-1 , total carbon=15.1±2.1 g kg -1 , organic matter=26±3.7 g kg -1 , C/N ratio=10±0.8, pH water=6.2±0.1, + -1 pH KCl=5.5±0.1, CaCO 3=0, cation exchange capacity=10.8±0.7 cmol kg , P 2O5=0.04±0.01 -1 -1 -1 -1 g kg , K 2O=0.065±0.004 g kg , MgO=0.15±0.02 g kg , CaO=3.3±0.2 g kg ). During the first year, the plants were trained to a single shoot topped when the stem reached 1.8 m length. The lateral shoots were initially removed, then left to develop after topping. The plants were supplied with a nutrient solution (2.5 mM KNO 3, 0.25 mM MgSO 4.7H 2O, 0.62 mM

NH 4NO 3, 1 mM (NH 4)H 2PO 4, 9.1 )M MnCl 2.4H 2O, 46.3 )M H 3BO 3, 2.4 )M ZnSO 4.H 2O, -1 0.5 )M CuSO 4, 0.013 )M (NH 4)6Mo 7O24 .4H 2O and iron was supplied as 8.5 mg L of Fe- EDDHA 5.9%) several times a day (total~600mL). During the second year, one-year-old plants were grown in a semi-controlled environment greenhouse equipped with a cooling system and 150 balances (CH15R11 CHAMPII, Obaus

- 89 - Chapter IV. Plant growth and hydraulic properties

Gm bH, Nänikon, Switzerland) automatically recording pot weight ( Sadok et al. , 2007 ). Before budburst, the pots were wrapped with thick polyethylene bags to prevent any water loss by direct evaporation from the soil. Each pot was watered, and after a 12 h of drainage period, weighed on a 1 gram resolution balance. The individual weight of each pot was determined and recorded automatically every 15 min. After 4 or 5 days, the pot weight became stable and the value was recorded in order to calculate soil saturated water content

(%s). The plants were trained to a single shoot and all the lateral shoots were removed throughout the experiment. The photosynthetic photon flux density (PPFD) close to the canopy level was measured with two Quantum Sensors (LI-190, LI-COR, Nebraska, USA). The dry and wet temperatures were measured with two psychrometers, and all the data were recorded every 15 min with a data logger (Campbell 21X, Campbell Scientific Inc., Loughborough, U.K.) over the entire experimental period (Figure S. 3). VPD was calculated from the temperature difference between dry and wet bulb thermometers. The equations to calculate VPD were described by Allen et al. ( 1998 ).

IV.2.2 Experimental setup Data were acquired over the course of two independent experiments run during two years. After budburst, the pots were first irrigated with water to maintain soil water content (SWC) th between 85 to 95% of %s until the 6 leaf emerged on the single shoot. Then, 25 plants of each scion/rootstock pair were selected for homogeneity of shoot development and randomly attributed to four different water supply treatments and one treatment used to harvest plants onset of the experiment. All pots were irrigated daily to recover %s to 100% with the same volume of nutrient solution (200 mL) plus water (~400 mL, depending on pot weight water loss) until each stem had produced the 15 th leaf. Then, four different theoretical targets of

SWC were determined: 100% of %s (SWC = 0.32 ± 0.04 kg H 2O/kg soil) for the control

(CTL), ~70% of %s (SWC = 0.23 ± 0.03 kg H 2O/kg soil) for low water-deficit (LWD), ~55% of %s (SWC = 0.18 ± 0.02 kg H 2O/kg soil) for moderate water-deficit (MWD) and ~40% of %s

(SWC = 0.14 ± 0.02 kg H 2O/kg soil) for high water-deficit (HWD). In order to reach the target values of SWC at the same time for each treatment, water supply was decreased step by step during one week. When the SWC values of each treatment were achieved, the pots were irrigated twice a day in order to maintain theoretical target values of SWC under steady-state conditions for 2 weeks. The individual weight of each pot was monitored and recorded automatically every 15 min during the 2 weeks to continuously estimate SWC.

IV.2.3 Plant and soil water relation measurement To assess for plant water status, water potential was determined with a pressure chamber (SAM Précis 2000, Gradignan, France) equipped with an electronic pressure gauge with an accuracy of 0.001 MPa ( Turner, 1988 ). During each of the two years of experiments, ten additional plants of each scion/rootstock pair (n=10) were assigned to predawn leaf water potential ( PD ) monitoring. For these plants, each pot was irrigated to reach %s and then no more water was supplied during 6 days. PD measurements were conducted every day

- 90 - Chapter IV. Plant growth and hydraulic properties

between 03.00 am and 05.00 am. The pot weight was recorded concomitant to PD in order to establish a relationship between the target values of SWC and PD for each scion/rootstock pair.

IV.2.4 Growth analysis: leaf area, biomass partitioning and growth rate

IV.2.4.1 Shoot growth and leaf area measurements Maximal leaf length (L) and maximal leaf width (W) were recorded twice a week on all rootstock-scion pairs for each treatment (n=5) in both experiments and two independent years. In order to estimate canopy area non-destructively, leaves (n = 529) of different sizes were collected from all treatments and genotypes at the end of each experiment. Maximal leaf length (L, cm) and maximal leaf width (W, cm) were measured. Leaves were then scanned (Canon MP600, Canon Inc., Japan) and leaf area was measured using ImageJ (ImageJ, U.S. NIH, Maryland, USA) software. Linear equation (Figure S. 4, y=0.6161x-0.5087, R 2=0.9912) was used to estimate whole plant leaf area (LA, m 2) from measurement of leaf length and width as described by Montero et al. ( 2000 ).

IV.2.4.2 Biomass partitioning and relative growth rate Five plants per rootstock-scion pair were harvested before the beginning of water stress. At the end of each experiment, all plants ( i.e. 5 plants per treatment and rootstock-scion pair) were also harvested. Each plant was divided into four organ types: main shoot (including petioles), leaf laminae, trunk and roots. Dry weight (DW) of each organ was determined after drying for 96 h in an air-forced oven at 70°C. Biomass allocation to plant organs was estimated using the following ratios, LWR (leaf DW/whole plant DW excluding trunk, %) for leaves, SWR (stem DW/whole plant DW excluding trunk, %) for stem and RWR (root DW/whole plant DW excluding trunk, %) for roots. Additionally, the whole plant RGR (Eqn. 1, g g -1 d -1 ) together with its components net assimilation rate (NAR, g m -2 d -1 ), specific leaf area (SLA, leaf area/leaf DW, m 2 g -1 ) and LWR were calculated according to Hunt et al. ( 2002 ):

C 1 S C (W S C 1 S C (W S LA LW D TD T # D T D T (Eqn. 1) E W U E (t U E LA U E (t U LW W RGR NAR SLA LWR

where t is time, W is total DW per plant excluding the trunk, LA is total leaf area per plant and LW is total leaf DW per plant.

IV.2.5 Root xylem anatomy

IV.2.5.1 Sample collection Measurements of root xylem anatomy were carried out on each rootstock-scion pair for CLT and MWD treatments only in the first year (2009). Main adventitious root segments (0.5 cm length) were collected at 25 cm distance from the trunk and fixed for 2 hours in FAA

- 91 - Chapter IV. Plant growth and hydraulic properties

(form alin:acetic acid:[ethyl] alcohol, 2% formaldehyde, 5% acetic acid, 63% ethanol and 30% distilled water, v/v) ( Bogeat-Triboulot et al. , 2007 ). Then, the root segments were dehydrated in a graded water:ethanol series and stored in 70% ethanol at room temperature in screw-top glass bottles until use ( Chaffey, 2002 ).

IV.2.5.2 Sectioning and staining Before sectioning, the root segments were rehydrated in a graded water:ethanol series (50%, 30% and distilled water for 15 min each). Then, 20 )m cross-sections were cut with a vibrating blade microtome (MICROM HM 650 V, Microm International GmbH, Walldorf, Germany). After bleaching and rinsing with distilled water, the sections were stained for 1 min with toluidine blue (0.05% [w/v] in 0.1M sodium acetate buffer at pH 5.8) ( Bogeat- Triboulot et al. , 2007 ) and wet mounted using a drop of distilled water between glass microscope slide and cover glass. Images were taken with a digital camera (RTKE Spot camera, Diagnostic Instruments Inc., Sterling Heights, USA) mounted onto a light microscope (Axiophot, Carl Zeiss, Toronto, Canada). Xylem conduits (tracheids or vessel elements) were measured and counted using image processing software (Image-Pro Plus, MediaCybernetics Inc., Bethesda, USA). Several root xylem-related traits were recorded, such as number of conduits, xylem area of each conduit, maximal diameter, minimal diameter and mean diameter of each xylem conduit as well as the area and diameter of the stele (central cylinder).

IV.2.6 Root hydraulic conductivity

IV.2.6.1 Theoretical hydraulic conductivity: the Hagen-Poiseuille equation From the root xylem-related trait measurements, the theoretical hydraulic conductivity

(Kh theo ) was calculated using the Hagen-Poiseuille equation. Since xylem conduits are rarely circular in cross-section ( Tyree & Ewers, 1991 ) and are better described as ellipsoids ( Tyree

& Zimm ermann, 2002 ), we calculated Kh theo according to the Hagen-Poiseuille equation (Eqn. 2) for elliptical vessel ( Lewis & Boose, 1995 ): F n a 3 b3 Kh theo # B 2 2 (Eqn. 2) 64 E i#1 a / b 4 Where Kh theo is the theoretical hydraulic conductivity of conduits in a cross-section (m MPa -1 s -1 ); $ is the viscosity of water at 20°C (1.002 e -9 MPa s); n is the number of conduits in a cross-section; a and b are maximal and minimal diameter (m) of the xylem conduit. Eqn. 2 shows that hydraulic efficiency is proportional to the conduit diameter raised to the fourth power, so a modest increase in conduit diameter leads to a significant gain in transport capacity ( Tyree et al. , 1994 ; Tyree & Zimmermann, 2002 ; Pittermann, 2010 ). In Eqn. 2, the solute concentration of xylem sap is thought negligible and does not influence viscosity ( $) of the fluid while temperature has an effect on viscosity ( Tyree & Zimmermann, 2002 ). Therefore, the dynamic viscosity of water ( $) at 20°C (1.002 e -9 MPa s) was used for calculations ( Zwieniecki et al. , 2001 ).

- 92 - Chapter IV. Plant growth and hydraulic properties

IV.2.6.2 Measured hydraulic conductivity

Hydraulic conductivity of individual adventitious roots (Kh r) was measured using a method described by Schubert et al. ( 1995 ). This method was used to investigate the effects of shoot orientation ( Schubert et al. , 1999 ) and water stress ( Lovisolo & Schubert, 1998 ) on 4 -1 -1 grapevine stem hydraulic conductivity. In these studies, Kh r (m MPa s ) was measured immediately after excision of 0.2 m long root segment. This length was chosen to avoid overestimation of conductivity based on the results of previous studies ( Lovisolo & Schubert, 1998 ; Schubert et al. , 1999 ).

Root sectio ns were sealed into a pressure chamber filled with degassed and deionised water. The sample was then pressurized during 2 min at 0.3 MPa m -1 to eliminate embolisms and 5 min at 0.1 MPa m -1 to allow steady state flow rates ( Lovisolo & Schubert, 1998 ). The water flow ( Q, m 3 s -1 ) was measured by collecting three replicates of exudates into a pre-weighed tube containing cotton wool placed over the protruding sample to minimize evaporation ( Schultz, 2003a ) over 2 min at a constant pressure (0.1 MPa m -1 ) ( Lovisolo & -1 Schubert, 1998 ). Kh r was calculated relating Q to the driving force (MPa m ) across the root segment ( Schultz, 2003a ).

IV.2.7 Data analyses and statistics The data were analyzed using R software ( R Development Core Team, 2010 ) with the agricolae package. All data were subjected either to two-way analysis of variance (ANOVA) or a three-way ANOVA to assess for the effects of treatment and genotype, or the effects of treatment, genotype and year, respectively. When interactions occurred, Tukey’s multiple comparison procedure was executed at the 5% level of significance. When the assumption of normality and equal variance failed, a non-parametric Kruskal- Wallis test was performed. Dunn’s multiple comparisons at the 5% level of significance were performed if interactions were significant. When the year was shown to have no significant effect, the data collected in experiments over two years were combined.

IV.3 Results

IV.3.1 Effects of soil water content on plant water status Each scion-rootstock pair was submitted to 14 days of steady-state water-deficit at four SWC levels ( i.e. CTL: control; LWD: low water-deficit; MWD: moderate water-deficit and HWD: high water-deficit).

The relationship between the four SWC levels and PD are shown in Figure IV.1. Results did not statistically differ between rootstocks at any given SWC, but PD was significantly

(p<0.01) different between SWC conditions (Figure IV.1). PD decreased on average to -0.24 MPa for LWD, to -0.44 MPa for MWD and to -0.74 MPa for HWD (Figure IV.1).

- 93 - Chapter IV. Plant growth and hydraulic properties

SWC (kg H 2O/kg soil) 0.12 0.16 0.20 0.24 0.28 0.32 0.0 CTL LWD -0.2

MWD -0.4 (MPa)

HWD

PD -0.6

% 110R -0.8 RG (SWC) %PD = -7.74e R2= 0.987, p<0.001 -1.0

Figure IV.1: Relationship between soil water content (SWC) and predawn leaf water potential ( PD ) measured for the two scion-rootstock pairs (CS onto RG, CS onto 110R). The SWC data correspond to the target values of each treatment ( i.e. CTL: control; LWD: low water-deficit; MWD: moderate water-deficit and HWD: high water-deficit). Each point represents the mean value ± standard deviation (SD) of 7-10 samples.

IV.3.2 Water-deficit effects on growth and biomass allocation Gradual increases of water-deficit induced proportional deceases of LA regardless of rootstock genotypes (Figure IV.2).

5200 a RG b 5200 110R ab a 4550 ab 4550 c ab 3900 abc a ab 3900 ) bc ab ab 2 bc bc c c bc 3250 c 3250 2600 2600 (cm 1950 1950 LA 1300 1300 650 650 0 0 CLT LWD MWD HWD CLT LWD MWD HWD Treatments Figure IV.2: Final leaf area (LA) per plant for V. vinifera cv. Cabernet Sauvignon (CS) grafted onto V. riparia cv. Gloire de Montpellier (RG) and 110 Richter (110R) measured at the end of each experiment in 2009 (a) and 2010 (b). CTL: control; LWD: low water-deficit; MWD: moderate water-deficit and HWD: high water-deficit. Each value is the mean of five biological replicates (n=5) ± standard error (SE). Vertical bars that do not share the same letter are significantly different (p<0.05).

The year (p<0.001) as well as treatment (p<0.001) had significant effects on LA. The treatments showed more pronounced effects ( i.e. higher decreases in LA) in 2009 (Figure

- 94 - Chapter IV. Plant growth and hydraulic properties

IV.2a) than in 2010 (Figure IV.2b) and there were significant interactions between year and treatment (p=0.01). CLT was significantly different from HWD conditions both in 2009 (Figure IV.2a) and in 2010 (Figure IV.2b), regardless of rootstocks. In 2009 (Figure IV.2a), LA showed higher decreased under water deficit for 110R than for RG, a lthough no statistical difference was observed between rootstocks. In 2010 (Figure IV.2b), the difference between rootstocks was less pronounced than for 2009 (Figure IV.2a). Results showed no significant interaction between treatm ent and rootstock (p=0.465) as well as between year, treatment and rootstock (p=0.380). Statistical analyses of whole plant RGR (Figure IV .3) showed that both years (p=0.003), treatm ents (p<0.001) and rootstocks (p<0.001) had significant effects. Increasing water-deficit reduced RGR both in 2009 (Figure IV.3a) and 2010 (Figure IV.3b), but the effects were more pronounced in 2010 and there was a significant interaction between year and treatm ent (p<0.001). Additionally, the year significantly affected the response of the rootstocks and there was a significant interaction between year and rootstock (p<0.001). The effects of rootstocks were higher in 2010 (Figure IV.3b) than in 2009 (Figure IV.3a).

25 25 a RG b a 110R a 20 20 ab ab ab ab ab abc ) bb b -1 15 15 bc d b bc -1 cd 10 10 (g g (g d

5 5 Whole plant dry mass RGR Whole plant dry 0 0 CTL LWD MWD HWD CTL LWD MWD HWD

Treatments Figure IV.3: Effects of water-deficit (CTL: control; LWD: low water-deficit; MWD: moderate water- deficit and HWD: high water-deficit) on the relative growth rate (RGR) of whole plants for each scion/rootstock pair (CS onto RG, CS onto 110R) in 2009 (a) and 2010 (b). Mean (n=5) ± SE. Different letters indicate significant differences (p<0.05).

In 2009, rootstock RGR was not significantly different at the same level of water deficit (Figure IV.3a). Between levels of water supply, RGR of 110R at H WD was significantly different to CTL, while the RGR of RG was not significantly affected by the treatment despite a noticeable decrease with water-stress (Figure IV .3a). In 2010, the drought-tolerant 110R had higher RGR’s com pared to the drought-sensitive RG regardless of water deficit. However, the two rootstocks differed significantly for the MWD and HWD treatments (Figure IV.3b). Regardless of genotype, RGR decreased for the proportional to water supply (Figure IV.3b). RGR patterns observed between year, treatm ent and genotype were not consistent, but water-

- 95 - Chapter IV. Plant growth and hydraulic properties deficit decreased RGR and the drought-tolerant 110R seem ed to maintain higher RGR’s compared to RG.

Pooling all data across the two experimental years showed that water-deficit and rootstocks had no significant effect on biomass allocation to leaves, stem and roots (Figure IV.4).

100 (%)

RG Leaf 110R Leaf 60 RG Stem 110R Stem RG Root 40 110R Root

20 Biomass allocation Biomass 0 CTL LWD MWD HWD

Treatments Figure IV.4: Biomass allocation to the different compartments presented for each scion/rootstock pair (RG for CS onto RG; 110R for CS onto 110R) and for each treatment. The biomass allocation to the different compartments was expressed as percentage of whole plant weight (the trunk was not included). Mean (n=10) ± SE.

For all treatments and rootstocks, allocation to the roots represented ~50%, and the allocation to stems and leaves represented ~30% and ~20% of the total biomass (excluding trunk), respectively (Figure IV.4). Biomass allocation followed the same trends for the genotypes, treatm ents and organs, albeit water-deficit tended to reduce biomass production overall (Figure IV.4).

IV.3.3 Root hydraulic properties

IV.3.3.1 Drought-induced Shifts in xylem development

The relationship between the anatomical characteristics of rootstocks, Kh theo and Kh r was investigated for different levels of water-deficit (CTL and MWD). The histological study showed consistent differences in root anatomy between the drought-sensitive RG (Figure IV.5A,C) and the drought-tolerant 110R (Figure IV .5B,D) rootstocks, and strong differential modifications of this anatom y for these genotypes depending on water supply (CLT, Figure IV.5A;B and MWD, Figure IV.5C,D).

- 96 - Chapter IV. Plant growth and hydraulic properties

Figure IV.5: Light photomicrographs of root apical cross sections of Vitis riparia (RG, A,C) and 110 Richter (110R, B,D) under well-watered (CTL, A,B) and moderate water-stressed (MLD, C,D) conditions. The scale bar represents 0.1 mm.

The drought-tolerant 110R showed higher conduit (xylem vessels and tracheids) numbers compared to the drought-sensitive RG regardless of water supply (Table IV.1), but the rootsto cks did not differ in total conduit area, stele area as well as conduit-to-stele area ratio (Table IV.1).

Total conduit area Stele area Conduit-to-stele area ratio Treatment Genotype Conduit number (cm 2×10 5) (cm 2) (×10 6) RG 28.8 a 1.9 a 15.2 a 2.92 a CTL 110R 40.3 bc 2.2 a 18.2 a 1.77 b RG 33.5 ab 1.6 a 13.3 a 2.05 ab MWD 110R 55.4 c 1.8 a 15.6 a 2.79 ab Table IV.1: Average values (n=5) of the anatomical characteristics of apical roots of the two genotypes. Within a column, means that do not share the same letter are significantly different (p<0.05).

When xylem was characterized by conduit dimensions, the results showed that the frequency distribution of xylem conduit diameters differed between rootstocks and was

- 97 - Chapter IV. Plant growth and hydraulic properties modified by water stress (Figure IV.6a,b,c&d). Regardless of treatment and rootstock, the narrow conduit size class 10-20 )m had the largest number of conduit (Figure IV.6a, b), except for RG under CLT (Figure IV.6a). None of the roots had conduits exceeding 70 )m diam eter (Figure IV.6a,b). Regardless of water supply, a significant increase (p<0.05) of conduit number in the size class 10-20 )m was observed for 110R under MWD (Figure IV.6b). This increase was much larger as the one observed for RG (Figure IV.6a), yet both were significant (p <0.05) compared to CTL (Figure IV.6a&b). RG 110R 40 40 a b 35 CTL CTL 35 MWD MWD 30 30

25 25

20 20

15 15

10 10 (number of conduits) Frequency per class Frequency 5 5

0 0 c d 60 60

50 50

40 40

30 30 (% of(% total) 20 20

Frequency distribution Frequency 10 10

0 0 e f 50 50

40 40

30 30

(% of(% total) 20 20

10 10

Relative hydraulic conductivity hydraulic Relative 0 0 0 0 0 0 0 0 0 -10 -20 -30 -40 -50 -60 -70 -1 -2 -3 -4 -5 -6 -7 0 10 20 30 40 50 60 0 10 20 30 40 50 60

Xylem conduit diameter classes ( 'm) Figure IV.6: Absolute frequency distribution (a,b) and relative frequency (c,d) of conduits of different diameter size classes from root tips of a drought-sensitive (RG, triangle symbols, a,c) and a drought- tolerant (110R, square symbols, b,d) rootstock under well-watered (CTL, open symbols) and water- stressed (MWD, filled symbols) conditions. Relative theoretical hydraulic conductivity ( i.e. percent contribution of a diameter class to total hydraulic conductivity of the cross section) was calculated according to Eqn. 2 for each rootstock (RG, e, 110R, f). Data are averages of 4-5 replicates per treatment and genotypes (± SE).

- 98 - Chapter IV. Plant growth and hydraulic properties

Most conduits (>80% of the total amount) are concentrated in size classes between 10 and 40 )m (Figure IV.6c,d), regardless of treatment and rootstock. Under we ll-watered condition ~38% of all conduits were in size class 20-30 )m for RG (Figure IV.6c) and ~34% in size class 10-20 )m for 110R (Figure IV.6d). Under water-deficit (MWD), the highest relative frequency distribution per class was ~43% (Figure IV.6c) and ~57% (Figure IV.6d) in size class 10-20 )m, respectively for RG and 110R. Under CTL conditions, most of the conduction was through 20-50 )m conduits for both rootstock genotypes (Figure IV.6e&f) with a higher percentage through classes 30-50 )m fo r

110R (Figure IV.6f) as compared to RG (Figure IV. 6e). Under MWD conditions, Kh theo was mainly through 30-60 )m conduits for RG (Figure IV.6e) and through 10-40 )m conduits for

110R (Figure IV.6f). Regardless of diameter class, the Kh theo distribution of RG was little affected by water stress (Figure IV.6e). In contrast, there was a substantial shift in conduction to clas ses 10-20 )m (~25% of total) and 20-30 )m (~44% of total) for 110R (Figure IV.6f) when water was lim iting. This was significantly different (p<0.05) from RG. Thus, the major contribution to total Kh theo under water-deficit involved narrow conduits in the drought- tolerant 110R while it involved wider conduits in the drought-sensitive RG rootstock.

IV.3.3.2 Root conductivity is genotype-dependent rather than stress-responsive

Accompanying Kh theo , Kh r was assessed on root segments of 0.2 m length with a pressure chamber under all levels of water deficit (Figure IV.7). The data collected during the two independent experim ents were combined because years had no significant effect (p=0.19).

The results show that root hydraulic conductivity (Kh r) was significantly affected by genotype (p<0.001) while water status had no or only minor effects which were not significant (p=0.141). RG 110R 6 6 a 6 b 5 5 a

) x 10 x ) a -1 4 a 4 s

-1 a 3 3 MPa 4 2 2 (m b r 1 b b 1 b Kh 0 0 CTL LWD MWD HWD CTL LWD MWD HWD

Figure IV.7: Single root hydraulic conductivity (Kh r) at four levels of water supply for the drought- sensitive (RG, a) and the drought-tolerant (110R, b) rootstock (mean ± SD, n=10). Different letters indicate significant differences among treatments and rootstocks (p<0.05).

Kh r of the drought-tolerant 110R (Figure IV.7b) was always lower compared to values observed for the drought-sensitive RG (Figure IV.7 a). Hydraulic conductivity was in average

3-fold lower for 110R than for RG. Water-deficit tended to decrease Kh r for 110R, especially under HWD (Figure IV.7b), although there was no statistical difference between treatm ent.

- 99 - Chapter IV. Plant growth and hydraulic properties

The genotype-dependent rather than stress-responsive root hydraulic conductivity (Figure

IV.7) suggests that Kh r is a constitutive trait of the rootstock which was unaffected by modification of the water supply under this experimental design.

IV.4 Discussion In spite of their wide contrast in water-deficit adaptation ( Carbonneau, 1985 ), the scion- rootstock responses to changes in water availability are poorly understood. The present study is based on the contrasted drought sensitivity of two rootstocks grafted with the same scion and investigates the physiological changes induced by the rootstock genotypes.

IV.4.1 Plant water-status, growth and biomass

In the present experiments, PD dropped gradually when SWC decreased (Figure IV.1 ) but no difference in this response between genotypes was observed which indicate that strictly controlled gradients of SWC si milarly affect water status in both genotypes (Figure IV.1). In grapevine, leaf expansion, tendril elongation, and internode extension were inhibited by water-d eficits and the sensitivity of these organs depended on ontogeny ( Schultz & Matthews, 1988b ). Vegetative growth responses to water-deficit follow a hierarchical order and sensitivity which was found to increase to increase with the degree of branching ( Pellegrino et al. , 2005 ; Lebon et al. , 2006 ). In grapevines, leaf emergence rate on second order lateral branches is sensitive to moderate water deficit whereas growth and vegetative development (i.e. leaf emergence rate, final individual leaf area and internode length) on first order branches are only affected under strong water deficit ( Pellegrino et al. , 2005 ). In the present experiments, one-year-old plants were grown with a single shoot without branches and each stem had at least 15 leaves before the water supply was limited. Final LA was depressed as water-deficits developed, but only LA measured under HWD was significantly affected but there was no significant effect of the rootstock (Figure IV.2a&b). It may be hypothesized that water deficit implementation on well-developed shoots diminished possible effects of the rootstock. The fact that LA was significantly reduced only under HWD suggests that growth and vegetative development are less sensitive to LWD and MWD, similar to first order lateral branches ( Pellegrino et al. , 2005 ). Additionally, decreased LA in LWD and MWD relative to CTL, although not significant (Figure IV.2), suggests that the growth of older leaves developed before the onset of stress was com pletely inhibited during water deficit while growth was maintained to some extend in younger leaves ( Schultz & Matthews, 1988b ). The m echanisms that determine the different sensitivities of shoot growth to water deficit are complex and not well understood, although hormones likely play an important regulatory roles ( Sharp, 2002 ) of most interest is the role of ABA, which may antagonize ethylene production to maintain rather than to inhibit shoot growth ( Sharp, 2002 ; Sharp & LeNoble, 2002 ). However, ABA can affect plant metabolism and water transfer via multiple interconnected mechanisms ( Tardieu et al. , 2010 ). It can affect leaf growth via non-hydraulic and hydraulic effects which can both repress or promote leaf growth depending on environm ental conditions ( Tardieu et al. , 2010 ).

- 100 - Chapter IV. Plant growth and hydraulic properties

Water deficit and rootstock did not affect biom ass partitioning to different organs (Figure IV.4) whereas plant RGR was reduced and this reduction seem ed genotype-dependent (Figure IV.3b), despite inconsistent responses between experiments (Figure IV.3). Water deficit has been reported to have lim ited effects on biomass allocation compared to plants grown under different light or nutrient levels ( Poorter & Nagel, 2000 ). The causes of RGR variations due to drought m ay be ascribed to changes of a wide variety of parameters related to morphology and physiology which together can form a cluster of correlated traits ( Lambers & Poorter, 1992 ; Wright et al. , 2004 ). In the present study, the decrease in RGR (Figure IV.3) regardless of year and genotype could be explained by a decrease in SLA and NAR. In 2010, higher levels of and slower decreases in R GR in response to water deficit for 110R (Figure IV.3b) were related to higher S LA (203 and 160 cm 2 g -1 in average, respectively in CTL and HWD conditions) and higher NAR values (0.56 and 0.23 g m -2 d -1 in average, respectively in CTL and HWD conditions) as compared to the drought-sensitive RG (SLA: 194 and 155 cm 2 g -1 ; NAR: 0.47 and 0.14 g m -2 d -1 in average, respectively in CTL and HWD conditions). These results confirm findings by Galmés et al. ( 2005 ) with other woody perennials.

IV.4.2 Water deficit effects on xylem conduits size and hydraulic conductivity A variety of physiological and molecular mechanisms induced by water deficit are adopted by plants to prevent fatal dysfunction. Plants prevent fatal drops in Kh by adopting one or more of the three strategies: i) embolism avoidance, based on the stomatal control of xylem pressure, ii) embolism reversal, based on refilling mechanisms of gas-filled conduits and iii) production of new xylem, based on cambial activity ( Nardini et al. , 2011 ). Water deficit increased conduit number, although not significantly, in both genotypes (Table IV.1) suggesting a regulation of conduit production under water-deficit. Nevertheless, the num ber of xylem conduits in the roots was also a constitutive trait between genotypes (Table IV.1) with a larger number for the drought-tolerant 110R than for the drought-sensitive RG. Anatom ical constitutive differences between Vitis rootstocks have been reported ( Pongrácz & Beukman, 1970 ). We found greater xylem vessel diameter adjustm ent under MWD in roots of 110R (Figure IV.6b) compared to roots of RG (Figure IV.6a). Under stress, the number of narrow conduits (10-20 )m) were significantly increased in the 110R rootstock compared to well-watered conditions (Figure IV.6b) and to RG (Figure IV.6a), thus resulting in higher contribution of water flow through narrow conduits for 110R (Figure IV.6f). These results suggest greater plasticity and increased hydraulic safety for the drought-tolerant genotype (Hacke et al. , 2006 ). Water deficit induced changes in xylem development have already been reported for roots ( Mapfumo & Aspinall, 1994 ) and stems ( Lovisolo & Schubert, 1998 ) of V. vinifera but not yet for Vitis sp. rootstocks. Shifts in xylem conduit dimensions in response to water-deficit have been reported in other species such as in olive ( Trifilò et al. , 2007 ) and apple trees ( Bauerle et al. , 2011 ). Anatomical shifts resulting in narrower xylem conduits may contribute to restrict the water flow ( Tyree & Ewers, 1991 ), and reduce xylem conduit vulnerab ility to cavitation ( Bauerle et al. , 2011 ). However, vulnerability of xylem conduits to

- 101 - Chapter IV. Plant growth and hydraulic properties cavitation is not directly related to xylem size, as the pit m embrane properties seem to determine the vulnerability to cavitation ( Choat et al. , 2008 ). Yet size distribution may strongly impact the hydraulic conductivity since a large embolized xylem vessel induces a considerable down-regulation of water flow ( Tyree & Sperry, 1989 ; Tyree & Zimmermann, 2002 ; Cochard, 2006 ). Species that are able to adjust and op timize their hydraulic architecture in response to environmental changes have a greater resistance to cavitation ( Tyree & Zimmer mann, 2002 ) and may better tolerate soil moisture deficit ( Fonti et al. , 2010 ). The substantial shift in roots anatomy for the drought-tolerant 110R (Figure IV.6b ) may be a mechanism to avoid cavitation, since grapevine roots have been shown to be more vulnerable than shoots ( Lovisolo & Schubert, 2006 ; Lovisolo et al. , 2008a ). Hydraulic conductivity adjustments and lower xylem conduit embolization during water stress have been already reported in drought-tolerant rootstocks compare to other sensitive hybrids ( Lovisolo et al. , 2008b ). Moreover, it was suggested that drought-tolerant m ay have a higher contribution of cell-to- cell pathway to total water transport ( Lovisolo et al. , 2008b ). Aquaporin activities are likely to be involved in the balance of water relations during water deficit ( Vandeleur et al. , 2005 ), but the actual roles of aquaporins during drought remain unclear, due to the different regulation patterns ( Kaldenhoff et al. , 2008 ; Aroca et al. , 2011 ) which have also been observed in grape rootstocks ( Fouquet, 2005 ; Galmés et al. , 2007 ). Aquaporins ( Martre et al. , 2002 ) together with ABA ( Lovisolo et al. , 2008a ) play a significant role during recovery from water-deficit and this may also be an important trait of a drought-tolerant rootstock.

We found that Kh r is lower in the drought-tolerant (Figure IV.7b) than in the drought- sensitive (Figure IV.7a) genotype regardless of water supply. Together with xylem conduit adjustm ent, lower Kh r may represent an advantage under water stress, reducing the rate of water depletion ( Tyree & Zimmermann, 2002 ). Similar to the present study (Figure IV.7b), de Herrald e et al. ( 2006 ) showed that root hydraulic resistance is higher in 110R than in other rootstocks, regardless of water supply. These authors also found a lower hydraulic resistances in the trunk of V. vinifera cv. Tempranillo grafted onto 110R than onto another rootstock (SO4), where the grafting point presented the highest resistance to water flow regardless of genotype ( de Herralde et al. , 2006 ).

The present results give som e clues on how grapevines grafted onto different genotypes deal with water deficit. Different growth responses were found between rootstocks, especially with regard to RGR. However, further investigations are needed to understand the underlying mechanisms for the observed patterns of RGR. Particularly, a careful analysis taking into account time-series measurements of RGR would help to further elucidate the developmental regulation of growth at different levels of water supply ( Poorter & Nagel, 2000 ). The plasticity of root xylem conduit number and size and the constitutive low hydraulic conductivity in the drought-tolerant genotype contribute to explain the drought tolerance conferred by rootstocks to the grafted scion. Recently, the discovery of an ionic effect as a potential mechanism involved in Kh regulation ( Jansen et al. , 2011 ; Nardini et al. , 2011 ) point towards other possibilities for control of Kh by rootstocks. Although most

- 102 - Chapter IV. Plant growth and hydraulic properties measurements of the hydraulic properties of plants are destructive and time consuming, new techniques based on non-destructive imaging such as nuclear magnetic resonance ( Holbrook et al. , 2001 ; Van As, 2007 ), high resolution computed tomography ( Brodersen et al. , 2010 ; 2011 ) or synchrotron X-ray sources ( Kim & Lee, 2010 ) have allowed in vivo visualization of water flow in xylem conduits, as well as embolism repair ( Nardini et al. , 2011 ). Future investigations using these techniques may contribute to a more complete understanding of hydraulic differences between rootstock genotypes.

IV.5 Supporting informations

Year 2009 Year 2010

) A 1500 Steady-state water-deficit B Steady-state water-deficit 1500 -1 s

-2 1200 1200

900 900 mol m mol ' 600 600

300 300 PPFD ( PPFD 0 0 31 C D 31 27 27 23 23 19 19 15 15 11 11 Air temperature (°C) temperature Air 7 7 F E 2.0 2.0

1.6 1.6

1.2 1.2

0.8 0.8 VPD (KPa) VPD 0.4 0.4

0.0 0.0 4 9 4 9 4 9 4 9 6 1 6 1 6 1 6 1 6 12 12 13 13 14 14 15 15 12 13 13 14 14 15 15 16 16

Day of year Figure S. 3: Climatic data record in the greenhouse in 2009 (A,C,E) and 2010 (B,D,F). The PPFD, i.e. Photosynthetic photon Flux Density (A,B), the air temperature (C,D) and VPD, i.e. Vapour Pressure Deficit (F,E) were recorded every 15 minutes (average of measurements collected every 10 seconds) using a datalogger (Campbell 21X, Campbell Scientific Ltd., Leicestershire, UK). Data are mean values of two sensors.

- 103 - Chapter IV. Plant growth and hydraulic properties

300

y = 0.6161x - 0.5087 250 n = 529 r2 = 0.9912

) 200 2

150 (cm

LA 100

0 0 50 100 150 200 250 300 350 400 450 2 L*W (cm ) Figure S. 4: Relationship between LA (leaf area) measured using image analysis (ImageJ software) and LW (maximal leaf length [L, cm] ! maximum leaf width [W, cm]) measured during the two years of experiments for Cabernet Sauvignon grafted onto two rootstocks (RG and 110R). The two genotypes were plotted together because no rootstock effect was observed. Each point is an individual measurement (n=529).

- 104 -

Chapter V. Long term steady-state drought-induced transcriptome changes in roots of drought-sensitive and drought-tolerant grape rootstocks

- 105 - Chapter V. Transcriptomic analysis

V.1 Introduction Plant responses to water deficit depend on the extent and rate of water loss, as well as on the duration of water shortage ( Bray, 1997 ). At low water availability, plants evolve physiological and developm ental adjustments that are, in part, driven by remodelling of the transcriptome. Advances in biotechnology such as microarrays ( Rensink & Buell, 2005 ; Sreenivasulu et al. , 2010 ), deep RNA sequencing ( Ozsolak & Milos, 2011 ) and other omics technologies ( Urano et al. , 2010 ) coupled with the availability of several plant genome sequences ( Feuillet et al. , 2011 ) have furthered the understanding of transcriptional behaviour and root biology ( Benfey et al. , 2010 ). The reconfiguration of the transcriptome under water- deficit has been observed in various species, such as Arabidopsis thaliana ( Seki et al. , 2002 ; Bray, 2004 ; Wilkins et al. , 2010 ), barley ( Guo et al. , 2009 ), rice ( Degenkolbe et al. , 2009 ) and poplar ( Street et al. , 2006 ; Bogeat-Triboulot et al. , 2007 ; Wilkins et al. , 2009 ). This progress at molecular level, including gene expression, transcriptional regulation and signal transduction, has improved the understanding of the regulatory networks which control the water-deficit responses ( Bray, 2004 ; Yamaguchi-Shinozaki & Shinozaki, 2006 ; Seki et al. , 2007 ). It also proved useful for engineering drought tolerance in plants ( Umezawa et al. , 2006 ; Valliyodan & Nguyen, 2006 ). Several hundreds of genes induced by drought through both ABA-independent and ABA-dependent pathways have been identified ( Yamaguchi- Shinozaki & Shinozaki, 2006 ; Qin et al. , 2011 ). The adaptation of roots is poorly understood, although root growth is often less sensitiv e than shoot growth to water-deficit ( Sharp & Davies, 1989 ; Ober & Sharp, 2007 ). Progress in understanding the physiological mechanisms underlying elongation maintenance of the root tip (~4 mm) under water-deficit have been made with maize, using a precise and reproducible protocol (reviewed in Sharp et al. , 2004 ; Yamaguchi & Sharp, 2010 ). These mechanisms involve osmotic adjustment to allow partial turgor recovery and reestablishment of water potential gradient, loosening of cell wall ability, ABA accumulation and membrane hyperpolarization. Recent advances from transcriptomic ( Poroyko et al. , 2007 ; Spollen et al. , 2008 ) and proteomic ( Zhu et al. , 2007a ; Yamaguchi et al. , 2010 ) analyses along different zones of primary roots of maize and soybean provide new insights into the coordination of growth under water-deficit. The transcriptome, proteome and metabolome are controlled in a root-zone specific manner and involved in several physiological processes such as protection from oxidative damage, modification of cell wall properties, regulation of phenylpropanoid and amino acid metabolisms ( reviewed in Yamaguchi & Sharp, 2010 ). Cell identity determines the pool of genes that is regulated under abiotic stress ( Dinneny et al. , 2008 ). Drought not only differentially induces transcriptome expression patterns along the root axis, but also growth stage-, organ-, stress duration- and genotype-dependent responses ( Kreps et al. , 2002 ; Wilkins et al. , 2009 ; Cohen et al. , 2010 ; Lorenz et al. , 2011 ; Stolf-Moreira et al. , 2011 ; Wang et al. , 2011 ). Bioinformatics tools may help to decipherer this complexity of transcriptome remodelling. Large sets of “om ics” data pave the way to a system biology approaches using bioinformatics ( Van Norman & Benfey, 2009 ). For example, co-expression gene network is a

- 106 - Chapter V. Transcriptomic analysis powerf ul approach to identify coordinated pathways ( Aoki et al. , 2007 ; Usadel et al. , 2009 ), and has been used to analyze the response of drought-stressed roots ( Lorenz et al. , 2011 ). Behind the dissection of such complex traits, recent progress in phenotyping, including high- throughput methods such as quantifying and controlling soil water-deficit ( Granier et al. , 2006 ; Sadok et al. , 2007 ), allows robust comparison of drought-related physiological and molecular responses across genotypes ( Yano & Tuberosa, 2009 ; Tardieu & Tuberosa, 2010 ). Since the publication of the whole genom e sequences of Vitis vinifera L. cv. Pinot noir (Jaillon et al. , 2007 ; Velasco et al. , 2007 ), grape has become a model to study fleshy fruit in perennial plants. Despite the importance of roots under water deficiency, grape root studies have been f airly limited compared to berries ( Deluc et al. , 2009 ; Grimplet et al. , 2009 ). To date, grape root transcriptomic tools were used to generate expressed sequence tags (ESTs) (Moser et al. , 2005 ), to profile microRNAs ( Mica et al. , 2010 ) or to assess adventitious root development ( Thomas & Schiefelbein, 2002 ; Thomas et al. , 2003 ). Recently, 135 genes exhibiting specific root-enriched expression patterns were identified using curation and data mining of large-scale ESTs ( Tillett et al. , 2011 ). However, most of these studies focussed on the root of V. vinifera , although more than 90% of vineyards worldwide use a scion of V. vinifera grafted onto a rootstock of single American Vitis species or interspecific hybrids. Grape rootstocks are known to have contrasted water stress tolerance ( Carbonneau, 1985 ), but there is a su rprising lack of molecular investigation on the roots of grafted vines. The present work aim s to study the transcriptome responses induced by different levels of water-deficit in contrasted grape rootstock genotypes grafted with the same scion. A strictly controlled water-deficit stress was applied using a phenotyping platform in order to reveal the molecular regulation of rootstocks under long-term steady-state stress. Transcriptomic analyses were performed in growing root apices in order to assess rootstocks responses to water-deficit.

V.2 Materials and Methods

V.2.1 Genetic materials and growth conditions In the present study, data were acquired over the course of two independent experiments during two years. One-year-old Vitis vinifera L. cv. Cabernet-Sauvignon (CS, clone 15) grafted onto drought-sensitive ( V. riparia cv. Gloire de Montpellier clone 1030, i.e. RG), drought-tolerant ( V. berlandieri ×V. rupestris cv. 110 Richter clone 756, i.e. 110R) rootstocks and homografted (CS, clone 15) were selected with regard to their conferred drought tolerance to the scion ( Carbonneau, 1985 ). Before starting each experiment, the grafted plants were grown for one year in a greenhouse in 7 L pots (pitted below with a film to avoid soil loss) filled with exactly 1 kg of gravel for drainage and 5 kg of dry soil ( Table S. 4) to favour the development of a dense root system. The plants were irrigated daily with the same amount of nutrient solution (~600 mL, Table S. 5), trained to a single shoot without lateral shoots and topped when the stem reached 1.8 m long. At the beginning of each experim ent, the dormant plants were moved in a semi-controlled environment greenhouse equipped with 150 balances (CH15R11 CHAMP II, Ohaus GmbH,

- 107 - Chapter V. Transcriptomic analysis

Nänikon, Switzerland) with a resolution of 1 gram . The pots of each dormant plant were wrapped with thick polyethylene bags to prevent any water loss by direct evaporation from the soil. Before budburst, each pot was watered, drained for 12 h and the weight was measured and recorded automatically every 15 min. After 4-5 days, the pot weight became stable and the value was recorded to calculate field water ( %s).

V.2.2 Experimental setups and soil water relations After budburst, the pots were first irrigated with water to maintain soil water content th (SWC) between 85 to 95% of %s until the 6 leaf emerged on the single shoot. Then, 20 plants of each scion/rootstock pair were selected for homogeneity of shoot development and randomly dispatched to four different water supply treatments. All the pots were irrigated daily to recover %s to 100% with the same volume (200 mL) of nutrient solution described above plus water (~400 mL, depending of pot weight water loss). Before the beginning of stress, four different theoretical targets of soil water content (SWC) were determined: 100% of %s (SWC = 0.32 ± 0.04 kg H 2O/kg soil) for the control (CTL), ~70% of %s (SWC = 0.23 ±

0.03 kg H 2O/kg soil) for low water-deficit (LWD), ~55% of %s (SWC = 0.18 ± 0.02 kg

H2O/kg soil) for moderate water-deficit (MWD) and ~40% of %s (SWC = 0.14 ± 0.02 kg

H2O/kg soil) for high water-deficit (HWD). When all the plants reached 15 leaves, water supply was decreased step by step during one week in order to reach at the same time the target value of SWC for each treatment. Then, the pots were irrigated twice a day (9:00 am and 8:00 pm) to maintain % under steady-state conditions during 14 days. Throughout each experiment, the SWC was measured by weighting each pot and the amount of water for irrigation was determined to recover the desired SWC. During the two years of independent experim ents, root biological samples (n=5) were harvested for each treatment between 8:00 and 9:00 am after 14 days of water-deficit. In less than 1 min, about ten 2-cm-long apices were sampled in the whole root system, flash frozen in liquid nitrogen and stored at –80°C prior to transcriptome analysis.

V.2.3 RNA extraction and microarray hybridization All collected samples were quickly frozen in liquid nitrogen, ground to a fine powder and stored at –80°C until use. High-quality total RNA was extracted from 100 mg of starting root tissue using the Spectrum™ Plant Total RNA Kit (Sigma-Aldrich, St. Louis, MO, USA) following the manufacturer’s protocol. RNA extraction was followed by DNase I treatment (RQ1 RNase-Free DNase, Promega Corporation, Madison, USA), acid phenol:chloroform:isoamyl alcohol mix (25:24:1) treatment and a RNA clean-up procedure using Nucleospin ® RNA Clean-up XS Kit (Macherey-Nagel, Düren, Germany) following the manufacturer’s protocol. Total RNA quality (A 260 /A 280 and A 260 /A 230 ratio greater than 1.8) was checked using a micro-spectrophotometer (NanoDrop 200C, NanoDrop products, Wilmington, USA) and RNA integrity (28s/18s ribosomal RNA ratio greater than 1.4 and RNA integrity number (RIN) greater than 7) using a lab-on-chip (2100 Bioanalyzer, Agilent Technologies, Santa Clara, USA). For each sample, 10 )g of total RNA was reverse

- 108 - Chapter V. Transcriptomic analysis transc ribed (SuperScript ® One-Cycle cDNA Kit, Invitrogen Corporation, Carlsbad, USA), labeled and hybridized to the NimbleGen 12x135K whole grape genome microarrays 090918_Vitus_exp_HX12 (Roche, NimbleGen Inc., Madison, USA) according to the manufacturer’s protocol. Arrays were scanned with the MS 200 microarray scanner coupled with the NimbleScan V2.6 software (Roche NimbleGen). The grape whole-genom e array includes 118,015 probe sets representing 29,582 transcripts including 4 probes per transcript. NimbleGen probe design includes 60 mer oligos which are designed on the last 1500 bp on the 3’-end of transcripts (Roche, NimbleGen Inc., Madison, USA). The chip probe design is based on the 12X genome assembly ( Jaillon et al. , 2007 ) using the grapevine V1 gene model prediction from CRIBI ( http://genomes.cribi.unipd.it/ ). The chip probe design is available online at the following address: http://ddlab.sci.univr.it/FunctionalGenomics/ . The hybridizations were perform ed using 72 NimbleGen arrays which represent 72 individual plant samples ( i.e. 3 biological replicates per year [2009, 2010], per treatment [CTL, LWD, MWD and HWD] and per rootstock [CS, RG and 110R]). All the samples were hybridized following a complete randomized factorial design to take into account the variability across the glass-slide microarrays for broader statistical inference ( Yang & Speed, 2002 ).

V.2.4 Microarrays data analysis Microarray data analyses were performed using R ( R Development Core Team, 2010 ) and R/Bioconductor tools ( Gentleman et al. , 2004 ). Microarray quality controls were performed using the arrayQualityMetrics package ( Kauffmann et al. , 2009 ). Expression intensities were background corrected, quantile-normalized and summarized using the rma function of the oligo package ( Carvalho & Irizarry, 2010 ). Differentially expressed genes were identified using the limma package ( Smyth, 2004 , 2005 ) for all water-stressed conditions against the well-watered reference of the corresponding genotype. Genes with absolute log 2 fold changes >1 and Holm ( Holm, 1979 ) corrected p-values (False Discovery Rate, FDR) below 0.05 were considered significant. A three-way analysis of variance (ANOVA) was implemented in R ( R Development Core Team, 2010 ) to investigate the effects of factors ( i.e. year, genotype and treatment) and factor interactions on the number of transcripts with normalized log 2 expression intensities. Transcripts with Holm FDR p-values below 0.05 were considered significant. Probe sequences were m apped against the 12X version of the grapevine genome ( Jaillon et al. , 2007; http://www.cns.fr/vitis ) using the blast program ( Altschul et al. , 1997 ). Gene models were aligned against the UnirRef100_2011_07 database ( Suzek et al. , 2007 ) using the blast program of UniProt website ( Jain et al. , 2009 ; The UniProt Consortium, 2011; http://www.uniprot.org ). Annotation was done using the qualifiers described in Table V.1. Annotations for genes with significant Treatm ent×Genotype (T×G) interactions identified by ANOVA analyses were verified manually by blasting them against NCBI databases (http://blast.ncbi.nlm.nih.gov/Blast.cgi ).

- 109 - Chapter V. Transcriptomic analysis

Criterions Qualifiers > 50% alignment identity homologue to > 70% alignment identity similar to <= 70% alignment identity weakly similar to > 98% hit coverage complete <= 98% hit coverage partial Table V.1: List of criterions used to qualify the alignment homology of sequence annotation.

A principal component analysis (PCA) was performed to analyse gene expression patterns.

The PCA of normalized log 2 expression intensities of transcripts was implemented in R using the ade4 package ( Dray & Dufour, 2007 ). The principal components (PCs) were then plotted for each biological replicates (n =6). Gene ontology (GO) ( Ashburner et al. , 2000 ) terms were assigned to all predicted proteins of the 12X version of the grapevine genome ( Jaillon et al. , 2007 ) using InterProscan (Zdobnov & Apweiler, 2001 ) version 4.8 with data from Interpro ( Hunter et al. , 2009 ) version 32. Enrichment analysis of GO was performed using the R/Bioconductor topGO package (Alexa et al. , 2006 ) in R ( R Development Core Team, 2010 ).

V.2.5 Transcripts abundance validation by qPCR Seven genes with significant T×G interaction (p-value<0.1%) were chosen for validation by qPCR. Reverse transcription was performed from each sample from 2 µg of purified RNA using the Moloney Murine Leukemia virus reverse transcriptase (M-MLV RT, Promega Corporation, Madison, USA) according to the manufacturer’s instructions. The cDNA obtained was diluted (1/20, v/v) in distilled RNase free water and qPCR was performed using the CFX96 Real-Time PCR Detection system (Bio-Rad, Hercules, USA). Ten µl reaction mixes were prepared, including 5 µl of iQ™ SYBR Green Supermix (Bio-Rad), 0.2 µM of each primer and 2 µl of diluted root genotype-specific cDNA. Gene transcripts were quantified upon normalization to VvActin and VvEF1 G according to Vandesompele et al. (2002 ). Only the biological samples from year 2009 (36 samples from each treatm ent and rootstock combinations) were tested in technical duplicates. Specific oligonucleotide primer pairs were designed with Beacon Designer 7 software (Premier Biosoft International, Palo Alto, USA). Melt-curve analysis was controlled at the end of each qPCR run to confirm the specificity of the amplification. The efficiency of each primer pair was measured on a PCR product serial dilution during each run. Primer sequences used in qPCR experiments are listed in Table S. 6. Correlation between the microarray and qPCR results for this gene set was then perform ed for each experiment, and the statistical significance of the correlations determined.

For the microarray, the data input into the correlation analysis was the log 2 ratio value of expression intensity. For qPCR, we used the log 2 ratio value reported by qPCR from all replicates. Prior to performing correlation analyses, the data were tested for normality using the Shapiro-Wilk test. Because the data was not normally distributed, Spearman rank correlation was used as non-parametric equivalent of Pearson’s correlation calculation ( Morey et al. , 2006 ).

- 110 - Chapter V. Transcriptomic analysis

V.3 Results The transcriptomic responses of root apices to different levels of water-deficits induced by withholding water were investigated during two years (2009 and 2010). By quantifying and controlling water-deficit by physical soil variable ( i.e. SWC) with the balance platform (Sadok et al. , 2007 ), four reproducible long term steady-state levels of water-deficits were applied to different Vitis sp. (CS, RG and 110R) (Figure V.1A) and with the same patterns during the two independent years of experiment. The three genotypes experienced a similar degree of water-deficit between year and treatment (Figure V.1B), allowing a robust com parison of their root transcriptome rearrangements.

A B CS 0.32 a RG 0.32 110R 0.28 0.28

O/kg soil) 0.24 0.24 2

0.20 c 0.20 (kg H d 0.16 CTL 0.16 LWD SWC 0.12 MWD 0.12 HWD 0.08 0.08 9 9 0 9 9 -3 0 3 6 9 12 15 18 21 24 0 0 0 01 00 -20 -20 2 -20 2 Days after treatment D- D- CTL CTL-2010 W W LWD L MWD MWD-2010H HWD-2010

Treatment/Year Figure V.1: Illustration of the long term steady-state drought treatments and sampling procedure used in this study. A. Development of Soil Water Content (SWC) over time for well-watered control (CTL) and three drought treatments: low water-deficit (LWD), moderate water-deficit (MWD) and high water- deficit (HWD). Years and genotypes were plotted together since no significant effect of year (p=0.726) and genotype (p=0.841) have been observed (three ways ANOVA) during the steady-state conditions (Days after treatment 11 to 25). Each point corresponds to the mean (n=6) ± SD of SWC collected at the same hour (8:00 am). Arrows indicate the sampling time of roots for RNA analyses. B. SWC values measured the morning (8 am) of root sampling for each treatment, year and rootstock. Each bar represents mean (n=3) ± SD. Treatments were significantly different (p<0.001) and different letters over the bars indicate statistically significant differences between treatments at 5% level using Tukey multiple comparison test.

This experimental design, i.e. three individual plants as biological replicates collected in each independent year in four contrasted water treatments (Figure V.1), enabled to attach a robust statistical significance to specific contrasts ( Yang & Speed, 2002 ). It makes possible the identification of genes with significant dif ferential transcript accumulation between genotypes in response to long term water-deficit levels. Additionally, the precise control and monitoring of soil water deficit associated with genotypes exhibiting contrasting tolerance to drought was an efficient means to identify divergences and similarities in transcriptome profiles.

- 111 - Chapter V. Transcriptomic analysis

V.3.1 Differentially expressed genes increases with water-deficit levels The genotype-specific global changes in gene expression induced by water-deficit levels were compared to determine if a common set of genes was up- or down-regulated (Figure V.2). In total there were 1441, 1032 and 784 up-regulated genes and 1863, 1525 and 1078 down-regulated genes respectively for CS (Figure V.2A), RG (Figure V.2B) and 110R (Figure V.2C). The comparison between treatments revealed 164, 78 and 113 genes up- regulated and 212, 15 and 38 genes down-regulated respectively for CS (Figure V.2A), RG (Figure V.2B) and 110R (Figure V.2C) that were induced under all wate r-stressed conditions. CS RG 110R A B C LWD MWD LWD MWD LWD MWD

48 31 105 5 1 19 2 2 19 41 32 151 0 0 9 3 1 11 164 78 113 27 212 366 7 15 277 16 38 57 21 258 1 263 3 59 700 645 784 1148 1237 964

Up Up Up HWD HWD HWD Down Down Down

Figure V.2: Gene numbers regulated in response to water-deficit stress levels in the roots of the three rootstock genotypes differing by water stress tolerance. The Venn diagram presents the differentially expressed genes (p-value<0.05, false discovery rate (FDR) using Holm method to adjust the p-value, log2 fold change (FC) > 1) in the three water-stressed conditions (LWD, MWD and HWD) compared to well- watered condition (CTL) for each rootstock (n=6, A: CS, B: RG and C: 110R). The upregulated genes were showed in red (“Up” without underline) and downregulated were showed in blue (“Down” with underline) in the Venn diagram.

Regardless of genotypes, the levels of water-deficit induced an increased number of differentially expressed genes, with strong rearrangements for the severe stress (HWD condition) which induced ~ 700 up-regulated genes and ~ more than 1000 down-regulated genes (Figure V.2). The comparison between genotype profiles shows that CS (Figure V.2A) induced a higher num ber of differentially expressed genes under LWD and MWD conditions compared to RG (Figure V.2B) and 110R (Figure V.2 C), suggesting that CS perceived low and m oderate SWC as more stressful than the other genotypes. Severe water-deficits (HWD conditions) induced the same type of changes among genotypes (Figure V.2).

V.3.2 Number of significant transcripts across years, treatments and genotypes Microarray data have been interrogated using three-ways ANOVA to identify the number of significant genes affected by the different factors ( i.e. year, treatment and genotype) or by factor interactions (Figure V.3).

- 112 - Chapter V. Transcriptomic analysis

Year 4150

Treatment 5647

Genotype 6956

Year×Treatment 1160

Year×Genotype 58

Treatment×Genotype 24

Year×Treatment×Genotype 3 Factororfactor interaction 0 10 20 30 40 50 60 00 00 00 00 00 00 12 24 36 48 60 72

Gene count Figure V.3: Number of significant genes by factors or factors interactions obtained from the three-ways ANOVA analysis (p-value<0.05, FDR p-value using Holm method). The 24 significant genes in “Treatment×Genotype” (T×G) interactions were listed in Table V.2.

Analyses of variance allowed the identification of 4150, 5647 and 6956 significant genes, respectively for year, treatment and genotype and 1160, 58, 24 and 3 significant genes, respectively for the following factors interactions: Year×Treatment, Year×Genotype, Treatment×Genotype (T×G) and Year×Treatment×Genotype (Figure V.3). These findings indica te that the greatest source of variation in the transcriptome of samples was the effects of genotypes, following by the treatment and the year (Figure V.3). Surprisingly, the interaction of drought treatm ents and genotypes reveals only 24 genes which were stable between the 6 biological replicates.

Therefore, this pool of genes draws our attention and pointed out further investigations. The sequence names based on Vitis vinifera genome sequence G12X ( Jaillon et al. , 2007 ) both from Genoscope ( http://www.genoscope.cns.fr/externe/Genom eBrowser/Vitis/ ) and CRIBI ( http://genomes.cribi.unipd.it ) annotations, the regions of chrom osomes, the FDR p- value of the interaction from ANOVA results as well as the putative function of this pool of genes are listed in Table V.2. Among these pools of genes, the putative functions have been controlled m anually and annotated using BLASTP tools of NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi ). The pool of genes listed in Table V.2 showed that four genes have no predicted protein function (Table V.2).

- 113 - Chapter V. Transcriptomic analysis

T×G Best match Query G12X target C12X target Chr Species Putative function E-valu e p-value (Accession) coverage

Ar temisia GSVIVT01038233001 VIT_05s0094g01230 5 2.8e -9 ACN65116.1 2-alkenal reductase (AER) 90% 1e -132 annua Ri cinus Phosphatidylethanolamine GSVIVT01030332001 VIT_08s0058g00120 8 2.4e -8 XP_002530493.1 98% 9e -88 communis binding protein (PBP) Ri cinus Alcohol dehydrogenase, GSVIVT01025012001 VIT_06s0004g04320 6 2.9e -6 XP_002520078.1 90% 1e -147 communis putative (ADH) Solanum Putative kinase interacting GSVIVT01029368001 VIT_17s0053g00590 17 1.4e -4 AAU90326.1 67% 2e -94 demissum protein, identical (KIP) Populus GSVIVT01015854001 VIT_03s0017g01970 3 1.6e -4 XP_002323361.1 Predicted protein 99% 0.0 trichocarpa Camellia GSVIVT01032029001 VIT_13s0064g01480 13 1.9e -4 ADO51752.1 Lipoxygenase (LOX) 92% 0.0 sinensis Ricinus Vacuolar cation/proton GSVIVT01033962001 VIT_08s0007g02240 8 5.8e -4 XP_002528100.1 99% 0.0 communis exchanger 1a (CAX) Po pulus GSVIVT01027558001 VIT_15s0048g01470 15 9.8e -4 XP_002301305.1 Cytochrome P450 (CYP) 82% 0.0 trichocarpa GSVIVT01010053001 VIT_18s0001g15730 18 1.3e -3 CBI19984.3 V. vinifera Unnamed protein product 98% 1e -44 Ri cinus Ci nnamoyl-CoA reductase, GSVIVT01027603001 VIT_15s0048g00980 15 1.8e -3 XP_002526355.1 98% 9e -165 communis putative (CCR) Ricinus Desacetoxyvindoline 4- GSVIVT01027751001 VIT_05s0049g00350 5 1.9e -3 XP_002529304.1 99% 1e -114 communis hydroxylase (D4H) Glutathione S-transferase No prediction VIT_05s0049g01100 5 2.1e -3 AAG34804.1 Glycine max 96% 6e -101 (GST) Pr unus O-methyltransferase-like GSVIVT01027448001 VIT_15s0048g02490 15 2.8e -3 ADB85561.1 94% 4e -113 mume protein (OMT) BRASSINOSTEROID Ricinus INSENSITIVE 1-associated No prediction VIT_18s0086g00200 18 4.6e -3 XM_002517904.1 50% 3e -177 communis receptor kinase 1 precursor(BAK) No prediction VIT_05s0094g00300 5 1.1e -2 Q7XB39 V. vinifera Class IV Chitinase 92% 1e -105 Anthranilate N- Ricinus GSVIVT01024610001 VIT_06s0004g07650 6 1.7e -2 XP_002517738.1 benzoyltransferase protein 94% 2e -166 communis (HCBT) GSVIVT01010781001 VIT_05s0102g00660 5 2.6e -2 CBI36681.3 V. vinifera Unnamed protein product 99% 0.0 Ricinus GSVIVT01016653001 VIT_09s0002g00140 9 2.8e -2 AAV66577.1 Lipase (LIP) 99% 0.0 communis Populus GSVIVT01003226001 VIT_01s0182g00150 1 3.3e -2 XP_002312260.1 Pho1-like protein (PHO1) 97% 0.0 trichocarpa Corchorus Progesterone 5- !-reductase GSVIVT01037879001 VIT_03s0091g00450 3 3.4e -2 ADG46029.1 99% 0.0 capsularis (POR) GSVIVT01011698001 VIT_01s0011g05390 1 3.8e -2 CBI26862.3 V. vinifera Unnamed protein product 99% 8e -124 Arabidopsis mRNA-decapping enzyme- GSVIVT01033895001 VIT_08s0007g02870 8 3.9e -2 NP_563814.1 99% 3e -13 thaliana like protein (DCP) Receptor protein kinase Ricinus GSVIVT01033522001 VIT_08s0007g06250 8 4.0e -2 XP_002532341.1 CLAVATA1, putative 99% 0.0 communis (CLV1) Astragalus Universal Stress Protein GSVIVT01031228001 VIT_14s0060g01300 14 5.0e -2 ABB13620.1 93% 2e -65 sinicus (USP) Table V.2: List of transcripts whose significant T×G interactions are observed from ANOVA analysis (Figure V.2). The G12X target in 1 st heading column refers to the Genoscope annotation (http://www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/ ) of grape genome ( Jaillon et al. , 2007 ). The C12X target in the 2 nd column refers to CRIBI annotation ( http://genomes.cribi.unipd.it ) and was given for information to complete transcript without prediction from Genoscope. The 3 rd column shows the chromosome (Chr) position. All the putative functions have been controlled manually using BLASTP tools of NCBI database ( http://blast.ncbi.nlm.nih.gov/Blast.cgi ). The five right columns s how accession number, species-referred, putative protein functions, query coverage and match E-value.

V.3.3 Array performance validation Quantitative PCR (qPCR) was used to validate treatment-specific expression patterns of 7 transcripts whose T×G interactions were highly significant (FDR p-value<0.001, the 7 first listed genes, Table V.2) in microarray data. The correlation between microarray and qPCR are presented in Table V.3. Regardless of genes, all the pair of variables ( i.e. microarray and qPCR) showed positive correlation coefficients with p-values below 0.001, indicating that microarray and qPCR

- 114 - Chapter V. Transcriptomic analysis results increase togeth er in the same direction (Table V.3). Therefore, all tested genes exhibited significant correlation between m icroarray and qPCR results (rs>0.6, p<0.001, n=36), thus validating the microarray data (Table V.3).

Genes ID (12X target) Gene names Spearman’s Rho (rs) p-values

GSVIVT01038233001 AER 0.887 2e -7

GSVIVT01030332001 PBP 0.769 2e -7

GSVIVT01025012001 ADH 0.669 5.07e -6

GSVIVT01029368001 KIP 0.648 1.69e -5

GSVIVT01032029001 LOX 0.685 1.08e -6

GSVIVT01033962001 CAX 0.735 2e -7

GSVIVT01027558001 CYP 0.67 4.63e -6

Table V.3: Correlations of microarray (log 2 expression intensity) and qPCR (log 2 relative expression) data for 7 genes verified. All correlations were calculated using Spearman’s rank correlation coefficient. All the data points ( i.e. each biological replicate, n=36) were used for the calculations of correlation.

V.3.4 Water-deficit induced genotype-specific gene dose-responses The expression patterns of significant annotated genes in T×G interaction (Table V.2) were studied via PCA (Figure V.4) to have a global view on the main changes that occu rred during water stress between genotypes. The first two PCs accounted for ~51% of the total variance in transcripts expression (Figure V.4). The first PC accounted f or 28% of the variation and was explained by the presence of LIP, AER, PBP, CLV1, HCBT, ADH and USP transcripts, which are involved in lipid metabolism, oxidoreductase activity and responses to stress (Figure V.4). The second PC accounted for 23% of the variation and was m ainly represented by expressions of CYP, KIP, OMT, POR, DCP, PHO and LOX which are involved in membrane metabolism, kinase activity, lignin biosynthesis, phosphate transport, oxidation-reduction and oxylipin biosynthetic processes (Figure V.4B). When all biological replicates were plotted in relation to the first two PCs (Figure V.4A), it appeared that the three genotypes differed each other in relation to water- deficit levels. The drought-sensitive RG and the moderately drought-tolerant CS were m ainly separated by PC2 due to the expressions of CYP, KIP, D4H, OMT, PHO and LOX (Figure V.4A&B). The drought-tolerant was discriminated from other genotypes by PC1 via the expressions of LIP, AE R, PBP and CAX (Figure V.4A&B).

To understand the expression patterns of the drought-responsive genes between the different genotypes, log 2 expression intensities of the first 7 transcripts (FDR p-value<0.1%, Table V.2) were plotted for each water-d eficit condition (Figure V.5). These transcripts

- 115 - Chapter V. Transcriptomic analysis showed both genotype-specific dose-response to water-deficit (Figure V.5C, F, G & K) and dose-responses shared at least by two rootstocks (Figure V.5M&O, Q&R, S&U).

CS-CTL ACS-LWD B CS-MWD CS-HWD CS RG-CTL RG-LWD RG-MWD CYP RG-HWD 110R-CTL CCR 110R-LWD D4H 110R-MWD 110R-HWD KIP CLV1

HCBT LIP AER POR ADH PC2 (22.8%) PC2 (22.8%) PC2 PBP USP CAX DCP LOX PHO 110R OMT

RG -1.0 -0.5 0.0 0.5 1.0 -4 -2 0 2 4

-4 -2 0 2 4 -1.0 -0.5 0.0 0.5 1.0 PC1 (27. 85%) PC1 (27. 85%) Figure V.4: Principal Components Analysis (PCA) of the transcripts whose significant T×G interactions are observed in ANOVA analysis and listed in Table V.2. The transcripts without Genoscope annotation or putati ve function were not included in the PCA, but the patterns were unchanged if they are included. (A) Spatial distribution of rootstocks for each treatment is drawn on the two-first principal components (PCs). Each points represent an individual biological replicate collected either in 2009 (n=3) or 2010 (n=3). (B) Correlation plots of transcripts for the first two PCs of PCA analysis.

For the drought-tolerant 110R, the expressions of AER and PBP genes (Table V.2) were up-regulated (Figure V.5C&F) in response to gradual increase of soil water-deficit. Under HWD conditions, AER (Figure V.5C) and PBP (Figure V.5F) expressions were significantly higher com pared to CS (Figure V.5A&D, respectively for AER and PBP) and RG (Figure V.5B&E, respectively for AER and PBP) regardless of stress levels. For CS, ADH gene was up-regulated in response to decreasing SW C (Figure V.5G) while it was unaffected for RG (Figure V.5H) and 110R (Figure V.5I). Interestingly, ADH expression for CS in C TL condition (Figure V.5G) shared the same level as 110R (Figure V.5I) whereas RG (Figure V.5H) showed always the same expression than CS in MWD and HWD conditions (Figure V.5G). CS and 110R do not differ in the KIP expression profiles (Figure V.5J&L), while RG down-regulated this gene in response to water stress (Figure V.5K). Moreover, CS and 110R up-regulated LOX (Figure V.5M&O, respectively for CS and 110R) and down-regulated CYP (Figure V.5S&U, respectively for CS and 110R) expression profiles in response to water- deficit levels, while these two genes do not respond to stress for R G (Figure V.5N&T, respec tively for LOX and CYP). Regarding CAX expression patterns, RG (Figure V.5Q ) and 110R (Figure V.5R) signif icantly up-regulated this transcript in HWD condition compared to CTL, while CAX profile remains unaffected in all water supply conditions for CS (Figure V.5P ).

- 116 - Chapter V. Transcriptomic analysis

CS RG 110R CTL LWD MWD HWD CTL LWD MWD HWD CTL LWD MWD HWD 14.5 14.5 A B C a 13.0 b 13.0 11.5 b 11.5 10.0 10.0

AER c cd cd cd 8.5 cd cd cd cd 8.5 d 7.0 7.0 12.5 D E F a 12.5 11.0 11.0 b 9.5 9.5

PBP bc bc c bc bc 8.0 c c bc c bc 8.0 6.5 6.5

11.5 G ab H I 11.5 ab a abc abc 10.0 bc 10.0 cd de ADH 8.5 e e 8.5 e e 7.0 7.0 ab 12.4 J a K L ab 12.4 ab ab ab abc abc ab 11.4 bc c 11.4 KIP 10.4 d 10.4

9.4 9.4 15.6 M a N O 15.6 abc ab ab abc abc Expression Intensity Expression bc cd bc 2 14.3 cd 14.3 de LOX 13.0 13.0 e Log 11.7 11.7

15.3 P Q ab R a 15.3 bcde bcd ab ab cde abc 14.5 cde de cde 14.5 e CAX 13.7 13.7

12.9 12.9

13.1 S a T U 13.1 bc ab 11.5 bcd def 11.5 cde cdef 9.9 9.9 CYP f f ef 8.3 f f 8.3 6.7 6.7 CTL LWD MWD HWD CTL LWD MWD HWD CTL LWD MWD HWD

Treatments Figure V.5: Box plots of the log 2 expression intensities of the 7 genes having significant T×G interactions (FDR p-value<0.1%) from ANOVA analysis (microarray). Each column of plots corresponds to one genotype ( i.e. CS, RG and 110R) and each line of plots corresponds to one gene ( i.e. AER, PBP, ADH, KIP, LOX, CAX and CYP). The grape genome 12x targets and putative function of each gene were presented in Table V.2. Each box corresponds to six biological replicates for each treatment ( i.e. CTL, LWD, MWD and HWD) collected either in 2009 (n=3) and 2010 (n=3). The box whiskers plots visualize the minimum (bottom of the whisker cap), the 25 th percentiles (bottom border of the box), the median (line through the box), the mean (dotted line through the box), the 75 th percentiles (top border of the box), and the maximum (end of the whisker cap) of the distribution. For each gene, different letters over the boxes indicate statistically significant differences between treatments and genotypes at 5% level using Tukey multiple comparison test.

- 117 - Chapter V. Transcriptomic analysis

V.3.5 Gene ontology of water-deficit responsive genes In order to detect physiological significance of the drought-responsive transcripts between genotypes, an enrichment of GO terms was performed (Figure V.6). The frequency of annotations in each GO term for the drought-responsive transcripts for each genotype was compared with the frequency of annotations in each GO term for the whole genome in CTL condition. Several GO terms were significantly (F-test of Fisher, p<0.05) enriched in genes that had accumulated in response to water-deficit levels in a genotype-specific manner (Figure V.6). Regarding the biological process ontology (Figure V.6a-h), transcripts included in “sucrose metabolic process” (Figure V.6a) were over-represented in HWD for RG and 110R, but not for CS roots. In the other collections of processes, the tran scripts appear to be significantly over-represented under LWD and MWD conditions for at least two rootstocks (Figure V.6b,c,e&f). Interestingly, “response to abiotic stim ulus” was specifically enriched for 110R roots in LWD and MWD conditions (Figure V.6d). This category included genes such as dehydrin (DHN, co mplete to V. vinifera , GSVIVT01018878001, VIT_04s0023g02480), galactinol synthase (GolS, similar to Coffea arabica , GSVIVT01013763001, VIT_01s0127g00470) and a protein phosphatase 2C (PP2C, weakly similar to Oryza sativa , GSVIVT01028698001, VIT_16s0050g02680), which were found to be up-regulated for 110R roots under LWD (log 2 FC=4.1, 2.5 and 2.9, p<0.001, respectively for DHN, GolS and PP2C) and MWD (log 2 FC=4.3, 3 and 4.3, p<0.001, respectively for DHN, GolS and PP2C) conditions. Transcripts included into “Oxidation-reduction process” (Figure V.6g) were significan tly enriched for CS roots in all drought conditions, while only in HWD condition for 110R roots. The up-regulation of LOX (Figure V.6M& O) and down-regulation of CYP (Figure V.6M&O) in response to water-deficit levels in CS and 110R have contributed to the enrichment of this sub-category for these genotypes. In the molecular function ontology (Figure V.6i-p) , the “iron ion binding” (Figure V.6i), “electron carrier activity” (Figure V.6j), “oxidoreductase activity” (Figure V.6k&m) and “transition m etal ion binding” (Figure V.6p) were significantly enriched in response to all drought con ditions for CS and 110R roots under MWD and/or HWD conditions. These categories included such genes as CYP, LOX and D4H which have been found to be up- regulated (LOX, Figure V.5M&O) or down-regulated (CYP, Figure V.5M&O; DH4: log 2 FC=-0.73 to -1.1 for 110R roots, respectively under LWD to HWD conditions). Concerning the cellular component ontology (Figure V.6q-x), “extracellular region” (Figure V.6q) and apoplast (Figure V.6x) were significantly enriched in response to HWD conditions f or all rootstocks. The “cell wall” GO category was significantly over-represented for RG roots under MWD and HWD conditions (Figure V.6r). This category includes genes such as !-ex pansin 2 (EXPB, weakly similar to Solanum tuberosum , GSVIVT01005917001,

VIT_00s0309g00090) which was up-regulated for RG roots under MWD (log 2 FC=2.9, p<0.01) and HWD (log 2 FC=4.3, p<0.001) conditions, while some other expansin such as expansin-like B1 (EXLB, weakly similar to Arabidopsis thaliana , GSVIVT01017505001,

VIT_09s0002g08510) were down-regulated in these conditions (log 2 FC=-3.3 and -4.4, p<0.001, respectively for MWD and HWD).

- 118 - Chapter V. Transcriptomic analysis

Figure V.6: Gene ontology (GO) analysis. For each GO category ( i.e. 1. Biological process [a-h], 2. Molecular function [i-p] and 3. Cellular component [q-x]), the sub-categories were selected according to the highest number of significant terms for all treatments and rootstocks. In each radial plot, the rootstocks CS, RG and 110R are represented by green, red and blue colours respectively. For each rootstock, the increasing levels of water-deficit (i.e. LWD, MWD and HWD) are presented in counter- clockwise. The radial plots show both the mean value of the number of genes observed (n=6) for each treatment and rootstock (green, red and blue colour scales) and the dotted lines the number of gene expected. A Fisher’s exact test was used for assessing the overrepresentation or underrepresentation with a cut-off of p<0.005. A star ( *) shows the significant difference.

- 119 - Chapter V. Transcriptomic analysis

Som e other GO categories such as “integral to membrane” (Figure V.6t), “intrinsic to membrane” (Figure V.6v), and “membrane part” (Figure V.6) were over-represented under MWD and/or HWD for CS and 110R roots.

V.4 Discussion Water availability is a key process influencing plant growth and development. In the present study, the strict monitoring of SWC with a balance platform ( Sadok et al. , 2007 ) enables to induce and control different steady-state water deficit conditions (Figure V.1), allowing th e comparison of transcriptome rearrangements between genotypes. Such ecophysiological techniques, combined with quantitative trait loci ( Reymond et al. , 2003 ; Welcker et al. , 2007 ; 2011 ) or molecular and transcriptomic approaches ( Bogeat-Triboulot et al. , 2007 ; Parent et al. , 2009 ; Wilkins et al. , 2009 ; Cohen et al. , 2010 ), have been used to compared genotype responses to water-deficit stress. Since the publication of V. vinifera genome sequence ( Jaillon et al. , 2007 ; Velasco et al. , 2007 ), grapevine has become the model species for fleshy fruit trees ( Troggio et al. , 2008 ). However, “omics” studies in response to abiotic stresses in the roots of Vitis sp. have been fairly limited compared to shoots ( Cramer et al. , 2007 ; Tattersall et al. , 2007 ; Vincent et al. , 2007 ) or berries ( Castellarin et al. , 2007 ; Deluc et al. , 2009 ; Grimplet et al. , 2009 ). Molecular tools were used to generate expressed sequence tags in V. vinifera roots grown under different abiotic stress conditions ( Goes da Silva et al. , 2005 ; Tillett et al. , 2011 ). To the best of our knowledge, no studies have been performed on the roots of different grafted Vitis sp. genotypes grown under different water supply conditions.

V.4.1 Steady-state water deficits induce transcriptome rearrangements Experimental design in microarray analyses is essential for statistical assessment, especially replication ( Pavlidis et al. , 2003 ), to provide reliable conclusions ( Yang & Speed, 2002 ). In the present study, two independent experim ents, using a total of six plants as biological replicates in each water supply condition, provide a statistical power in the interpretation of the large-scale genes expression data. Regardless of genotypes, differentially expressed genes increased when SWC decreased (Figure V.2). Transcriptional profiling showed that LWD and M WD conditions induced less than 0.6% of the genes being differentially up- or down-regulated (Figure V.2). Under HWD conditions, 2.3% (645 genes, Figure V.2B) to 2.8% (784 genes, Figure V.2C) of differentially expressed genes were up-regulated and 3.5% (964 genes, Figure V.2C) to 4.5% (1237 genes, Figure V.2B) were down-regulated. Interestingly, the drought-tolerant 110R induced the highest up-regulation and the lowest down-regulation (Figure V.2C) of differentially expressed genes com pared to the drought-sensitive genotype RG (Figure V.2B). These results are consistent with several other analyses conducted in A. thaliana ( Seki et al. , 2002 ; Bray, 2004 ; Wilkins et al. , 2010 ), loblolly pine ( Lorenz et al. , 2011 ), poplar ( Bogeat-Triboulot et al. , 2007 ; Wilkins et al. , 2009 ; Cohen et al. , 2010 ) and V. vinifera ( Cramer et al. , 2007 ; Tattersall et al. , 2007 ) where 1-16% of the genes are differentially expressed in responses to water- deficit. Nevertheless, the identification of differentially expressed genes was strongly

- 120 - Chapter V. Transcriptomic analysis influenced by the rate and severity of water deficit, the m ode of water stress induction ( e.g. tissue desiccation, polyethylene glycol, etc.) ( Kreps et al. , 2002 ; Bray, 2004 ; Huang et al. , 2008 ), the genotypes and plant organs ( Bogeat-Triboulot et al. , 2007 ; Wilkins et al. , 2009 ; Cohen et al. , 2010 ; Wilkins et al. , 2010 ). Additionally, the method used to identify genes may induce strong variations among microarray studies ( Shi et al. , 2008 ) and the selection of differentially expressed genes using FC ranking together with a non-stringent p-value cut-off (e.g. p<0.05) was an effective procedure to avoid false positive genes ( Shi et al. , 2006 ). The present study used statistical methods ( i.e. FDR p-value<0.05) in conjunction with log 2 FC>1 to identify robust transcripts in response to water deficit conditions, following MAQC recommendations ( Shi et al. , 2006 ; 2010 ).

V.4.2 Molecular physiology of drought-responsive genes involved in T×G interactions In the present study, long-term steady-state water deficits have been imposed in order to identify few candidate genes that might play a role in drought acclimation. The significant genes identified in T×G interactions (Table V.2) may be involved in long-term adaptive responses to water-deficit stress, as the expression patterns of these transcripts depends on the water deficit intensity (Figure V.5). The regulation of these genes follo wed the increase of water-deficit in a dose- and genotype-dependent manner (Figure V.4 & Figure V.5). However, transcripts abundances are not sufficient to prove their roles in drought adaptation due to translational and post-translational regulation: transcripts and proteins abundance are not always co rrelated ( Kawaguchi et al. , 2004 ; Böhmer & Schroeder, 2011 ). Additionally, the number of drought-responsive genes are known to strongly depend on the timing of water stress induction ( Yamaguchi-Shinozaki & Shinozaki, 2006 ; Qin et al. , 2011 ), with the highest number observed shortly (hours) after drought induction ( Kreps et al. , 2002 ), followed by slow and adaptive regulations (hours to days) of gene expression ( Yamaguchi-Shinozaki & Shinozaki, 2006 ).

V.4.2.1 Drought may induce lipid peroxidation in the roots of CS and 110R Among the putative genes identified in T×G interactions (Table V.2), 2-alkenal reductase (AER, Figure V.5C), phosphatidylethanolamine binding protein (PBP, Figure V.5F), alcohol dehydrogenase (ADH, Figure V.5G) and lipoxygenase (LOX, Figure V.5M&O) were gradually up-regulated in response to water-deficit intensity in the roots of CS and 110R genotypes. Additionally, GO term s related to “iron ion binding” (Figure V.6i) and oxidation- reduction metabolism (Figure V.6g,j&k) were also significantly enriched in response to water-d eficit intensity in the roots of CS and 110R. Drought is often accompanied by oxidative stress ( Moran et al. , 1994 ), inducing the production of reactive oxygen species (ROS) such as hydroxyl radicals (OH •), which cause oxidative damage of membrane lipids 2+ and proteins ( Møller et al. , 2007 ). A metal/H 2O system, including Fe ions, is very efficient to generate OH • ( Apel & Hirt, 2004 ; Rorat, 2006 ) which corroborates with enrichment of GO terms (Figure V.6g,i,j&k) observed in this study. Oxidation of polyunsaturated fatty acids by ROS leads to the f ormation of lipids peroxides ( Porter et al. , 1995 ; Bartels, 2001 ). Formation

- 121 - Chapter V. Transcriptomic analysis of oxidized polyenoic fatty acids, co llectively called oxylipins, is one of the main reactions in lipid alterations. Enzymatic synthesis of oxylipins is initiated by incorporation of oxygen into a fatty acid molecule, catalyzed by the activities of 9- and 13-LOXs ( Feussner & Wasternack, 2002 ; Liavonchanka & Feussner, 2006 ) or in the presence of singlet oxygen or ROS ( Bartels, 2001 ; Durand et al. , 2009 ; Hamberg, 2011 ). This initial fatty acid oxidation forms a fatty acid hydroperoxide, which is further modified by an array of enzymatic activities to give rise to a multitude of metabolites, including alkenal and aldehyde compounds ( Mosblech et al. , 2009 ). Eighteen LOX-like sequence were identified in the grape genome by phylogenetic analysis, including the LOX in the present study (Table V.2) which was a 13-LOX type II named VvLOXJ ( Podolyan et al. , 2010 ). The degradation of lipid peroxides induces the formation of cytotoxic 2-alkenals and oxenes (aldehyde and ketone designated as reactive carbonyls) ( Mano et al. , 2002 ). NADPH- dependent reductases, such as 2-alkenal reductase (AER) found in A. thaliana ( Mano et al. , 2002 ), were involved in detoxification of lipid peroxide-derived reactive carbonyls in plants (Mano et al. , 2005 ; Yamauchi et al. , 2011 ). Aldo-keto reductase improved tolerance to drought through enhanced reduction of reactive aldehyde ( Oberschall et al. , 2000 ; Hideg et al. , 2003 ). Up-regulation of AER in 110R roots (Figure V.5C) suggests that related-enzymes might function to reduce aldehyde produced under water-deficit conditions. Significant up-regulation of PBP observed during water deficit in the roots of 110R (Figure V.5E) is known to play a role in phospholipid biosynthesis, which are involved in m embrane biosynthesis and signal transduction during drought ( Munnik & Meijer, 2001 ; Testerink & Munnik, 2005 ; Munnik & Vermeer, 2010 ; Testerink & Munnik, 2011 ). Phospholipids of cell plasm a membranes, such as phosphatidylethanolamine, increase during water-deficit (Norberg & Liljenberg, 1991 ; Larsson et al. , 2006 ) and may play a role in the flexibility of the membrane ( Larsson et al. , 2006 ). Moreover, PBP genes are involved in various other biological functions such as the regulation of signalling pathways, the adjustment of growth cessation, dormancy and desiccation tolerance ( reviewed by Karlgren et al. , 2011 ). Therefore, up-regulation of PBP in response to water-deficit in 110R roots (Figure V.5F), suggests a role in phospholipid signalling and/or growth cessation. Additionally, significant enrichm ent of GO “response to abiotic stimulus” observed in the roots of 110R (Figure V. 6d), in part due to transcripts such as dehydrins (DNHs), galactinol synthase (GolS) and protein phosphatase (PP2C) m ay contribute in the protection from oxidative damage, synthesis of osmoticum and lipid signalling. Studies have implicated members of DHNs ( Rorat, 2006 ) and GolS (Nishizawa et al. , 2008 ) families in protection from oxidative stress, acting as osmoprotectants by ROS-scavenging. In addition to fundamental role in ABA pathway ( Park et al. , 2009 ), PP2Cs are also known to be targets of second messengers, such as phosphatidic acid and ROS ( Umezawa et al. , 2010 ). Up-regulation of ADH in response to water-deficit in CS roots (Figure V.5G) has been also reported under drought in several plants such as A. thaliana ( Seki et al. , 2002 ), Medicago sativa ( Irigoyen et al. , 1992 ), Nicotiana tabacum ( Frazier et al. , 2011 ), Zea mays ( Kato- Noguchi, 2000 ), Phaseolus sp. ( Micheletto et al. , 2007 ) and Populus sp. ( Street et al. , 2006 ; Bogeat-Triboulot et al. , 2007 ; Xiao et al. , 2009 ). Plant ADH gene family has been implicated

- 122 - Chapter V. Transcriptomic analysis in anaerobic ( i.e. flooding stress) and aerobic fermentation ( i.e. carbon-balance), in LOX pathway ( reviewed by Strommer, 2011 ) and in the response to a wide range of stresses, including drought and ABA ( de Bruxelles et al. , 1996 ). Using post transcriptional gene silencing based RNAi approach, Senthil-Kumar et al. ( 2010 ) showed that N. tabacum RNAi knockdown lines have higher susceptibility to drought compared to wild type. In maize doubly null ADH mutants, Peters and Frenkel ( 2004 ) showed that lipid peroxidation increased during cold stress com pared to wild type, suggesting that ADH is involved in acetaldehyde detoxification which has been initiated from lipid peroxidation. To summarize these results, it may be hypothesized that water-deficit induced oxidative stress and ROS production in the roots of CS and 110R, leads to lipid peroxidation which generates substrates for LOX activation (Figure V. 5M&O). The two rootstocks then use detoxification process, involving A ER (Figure V.5C) and ADH (Figure V.5G), respectively for CS and 110R, as protective m echanism against drought-induced oxidative damage.

V.4.2.2 Water-deficit affects cell wall properties in the roots of RG A putative kinase interacting protein (KIP, Figure V.5K) and a cation/H + exchanger (CAX, Figure V.5G) were down- and up-regulated in response to water-deficit respectively, in the roots of RG. Plants contain large num bers of protein kinase ( e.g. 1027 and 1429 in the genome of Arabidopsis and rice respectively) which play a key role in regulating nearly all aspect of cellular processes, including stress responses ( Ding et al. , 2009 ). For the vast majority of kinases, little is known of their functions and few have been functionally characterized, e.g. some mitogen-activated protein kinases ( Rodriguez et al. , 2010 ), SNF1- like kinases ( Umezawa et al. , 2011 ) or Ca 2+ -dependent protein kinases ( Harper et al. , 2004 ) which are involved in abiotic stress and hormonal responses, through multiple signal transduction pathways. CAX genes play an important role in the regulation of diverse transport processes ( Shigaki & Hirschi, 2006 ; Manohar et al. , 2011 ). The characterization of Arabidopsis CAX knockout mutants identified a variety of phenotypes such as sensitivity to abiotic stresses, especially salt stress ( Barkla et al. , 2008 ; Manohar et al. , 2011 ). Regarding water-deficit stress, little information is available on CAX regulation, but these transporters might play a role in the calcium-mediated signal-transduction cascades ( Hong-Bo et al. , 2008 ). More interestingly, gene ontology analysis reveals significant enrichment of terms related to “cell wall organization or biogenesis” (Figure V.6h) and “cell wall” (Figure V.6r) for the roots of RG under water-deficit. These categories include genes such as expansins which are involved in cell modification ( Cosgrove, 2000 , 2005 ; Sampedro & Cosgrove, 2005 ). Under water-d eficit, expansin activity and extractable expansin proteins increased in the apical roots of maize compared with well-watered roots ( Wu et al. , 1996 ; Wu & Cosgrove, 2000 ), correlating with the maintenance of elongation and the increase of cell wall extensibility of apical roots ( Yamaguchi & Sharp, 2010 ). In the present study, expansin-related transcripts were both up- and dow n-regulated in the roots of RG under water-deficit. Wu et al. ( 2001 ) showed that expansin genes are up-regulated in the apical 5 mm and down-regulated in the 5- 10 mm region of water-stressed maize roots. In the present work, 2-cm-long root apices were collected, the results of Wu et al. ( 2001 ) provide an explanation on the contrasted patterns of

- 123 - Chapter V. Transcriptomic analysis expansins regulation observed in the roots of RG. Specific-enrichment of GO terms together with expansins up-regulation suggests that water-deficit strongly influence the cell wall properties of the drought-sensitive RG rootstock.

To the best of our knowledge, this study is the first transcriptomic analysis in the roots of different grape rootstocks under water-deficit stress. The comparison between stress levels and genotypes identified 24 significant genes in “Treatment×Genotype” interactions. These genes displayed genotype-specific gene dose-responses, most of them were involved in lipid metabolism and cell wall process. This study provides new insights into the physiological mechanisms of grapevine rootstock adaptation to drought, and offers robust candidate genes and biological markers to the scientific community for functional genomic studies.

- 124 - Chapter V. Transcriptomic analysis

V.5 Supporting informations

Elements Concentration ± SD Clay 155.7 ± 12.2 g kg -1 Silt 98.3 ± 3.8 g kg -1 Sand 745.7 ± 12.7 g kg -1 Total Nitrogen 1.5 ± 0.1 g kg -1 Total Carbon 15.1 ± 2.1 g kg -1 Organic Matter 26 ± 3.7 g kg -1 C/N ratio 10 ± 0.8

pH H 2O 6.2 ± 0.1 pH KCl 5.5 ± 0.1

Total CaCO 3 0 Cation Exchange Capacity (CEC) 10.8 ± 0.7 cmol + kg -1 -1 P2O5 0.04 ± 0.01 g kg -1 K2O 0.065 ± 0.004 g kg MgO 0.15 ± 0.02 g kg -1 CaO 3.3 ± 0.2 g kg -1 Table S. 4: Concentration of elements in soil samples collected in 2009 (n=3) and 2010 (n=3) and expressed as mean ± standard deviation (SD).

Elements Concentration

KNO 3 2.5 mM

MgSO 4.7H 2O 0.25 mM

NH 4NO 3 0.62 mM

(NH 4)H 2PO 4 1 mM

MnCl 2.4H 2O 9.1 )M

H3BO 3 46.3 )M

ZnSO 4.H 2O 2.4 )M

CuSO 4 0.5 )M (NH ) Mo O .4H 4 6 7 24 2 0.013 )M O Fe-EDDHA 5.9% 8.5 mg L -1 Table S. 5: Concentration of macro- and micro-elements in the nutrient solution.

- 125 - Chapter V. Transcriptomic analysis

Genes ID (12X target) Genes names Primer sequence (5’->3’) CTTGCATCCCTCAGCACCTT GSVIVT01026580001 Actin TCCT GTGGACAATGGATGGA CAAGAGAAACCATCCCTAGCTG GSVIVT01023227001 EF1 G TCAATCTGTCTAGGAAAGGAAG GGGAAGAAGTCAGAAACAAGCAGGTG GSVIVT01038233001 AER TTCAT CAGCATAGCCACCCTGAATG CAACGAAGGAAGATTACCAAGAA GSVIVT01030332001 PEBP CGATGTCCTGAACCACGA GCTT GATGGCACTTCCAGGATGTC GSVIVT01025012001 ADH TACCCAGCCCATGTGGAGCAAC GTTGGAGGAGATGACGACGAAGGAAG GSVIVT01029368001 KIP TAGAGAGAAGCCAGGCAGGGT GAG TCGGTTAAGGCTGTCGTTA GSVIVT01032029001 LOX CATC AAGAGGCTTCGTAATCC TGAAGGAGTCGTCAGTGCTTAATAGG GSVIVT01033962001 CAX GCCAAAGCCTGCATATGTCAAATAGC AGACCAAACCTTTTCCGGCAGGAC GSVIVT01027558001 CYP GTTCCGCCAATGAGCCGATGC Table S. 6: Primers used for qPCR validation of transcript accumulation patterns. The gene ID in the first heading column refers to the grape genome annotation ( Jaillon et al., 2007; ht tp://www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/ ).

- 126 -

Chapter VI. General conclusions and perspectives

- 127 - Chapter VI. General conclusions and perspectives

VI.1 General conclusions

Predicted changes in the climate drivers like rising temperatures and CO 2 concentrations, together with variations of precipitation patterns will deeply alter viticultural practices, grapevine physiology, grape biochemistry and wine quality ( Schultz, 2000 ; Mira de Orduña, 2010 ). Climate change raises concerns about temporal and spatial water availability in m any grape growing countries ( Bates et al. , 2008 ). In Europe, irrigation in many areas is not a sustainable way to counteract water scarcity because of the legal limits, the infra-structural problems and the increase of water demand. Additionally, the rapidly increasing world population and the scarcity of suitable land for agricultural food production together with a changing climate will ultimately put pressure on grape-producing areas for the use of land and the input of resources ( Schultz & Stoll, 2010 ). Consequently, other practices that can potentially improve water m anagement of vineyards and water acquisition by grapevines need to be considered. Aside from canopy systems and their management, the choice of plant material is a key issue ( Ollat et al. , 2011 ).

Most prev ious studies on the mechanisms of water stress adaptation in grapevines focussed on the canopy level and largely excluded rootstocks as a factor of acclimation. Graft-induced rootstock-scion interactions may affect water flow in the soil-plant-atmosphere continuum and/or plant water status ( Chapter II). Therefore, in the present work, the role of rootstock genotypes was investigated for their potential effects on i) water uptake, ii ) water transport and iii ) shoot water use, using a combination of ecophysiological, modelling and transcriptomic approaches. Experiments were conducted under controlled conditions to decipher short and long term responses to drought of contrasting genotypes. An ecophysiological model was used to investigate the roles of rootstock genotypes in the control of stomatal aperture ( Chapter III). Long term steady-state water-deficit conditions were used to exam ine the responses of i) whole plant growth, root anatomy and hydraulic properties ( Chapter IV) and ii ) remodelling of roots transcriptome ( Chapter V). A summary of the major results obtained through these different experiments is presented in Figure VI.1.

The m odelling approach showed that rootstock affect stomatal aperture of the grafted scion via co-ordinated processes between root traits, hydraulic and chemical signals (Chapter III).

Stom atal conductance (gs), transpiration rate ( E) and leaf-specific hydraulic conductance

(K leaf ) were higher and better maintained under well-watered and moderate water-deficit conditions in the drought-tolerant genotype compared to the drought-sensitive one (Chapter III, Figure VI.1). The model identified several genotype-specific param eters which play important role, like root-related parameters, in the control of stomatal regulation ( Chapter III, Figure VI.1). Root system architecture ( i.e. root length area [RLA] and root density [model parameter d], Chapter III) and root hydraulic properties ( i.e. single root hydraulic conductivity

[kh r] and xylem conduit numbers) are important constitutive traits identified between rootstocks. Such traits may explain the difference in growth ( i.e. RGR, Chapter IV) and scion gas exchange ( Chapter III) observed between rootstock-scion pairs in the present wor k.

- 128 - Chapter VI. General conclusions and perspectives

Common shoot responses Z 1 PD dropped with decreasing SWC Z 1 gs responses to [ABA] signal Z 1 Restriction of E and gs under HWD Z Decrease of LA in function of SWD 2

V. vinifera cv. Cabernet Sauvignon Shoot-specific responses Shoot-specific responses grafted onto RG grafted 110R Z Z 1 Lower leaf (less open stomata) Higher leaf under WW 1 under WW Z 1 Higher K leaf under MWD Z 1 Lower K leaf under MWD Z No response of E and gs to model Z 1 E and gs respond to model parameters ! and ABA a parameters ! (stomatal regulation) Z Higher E and g under WW and and ABA (ABA synthesis) 1 s a MWD 1 Z Lower E and g under WW and s Z Higher g and g 1 MWD 1 smin smax Z Maintain higher RGR in response Z Lower g and g 1 smin smax to SWD 2 Z Lower RGR in response to SWD 2

Drought-sensitive rootstock Drought-tolerant rootstock V. riparia cv. Gloire de Montpellier V. berlandieri V. rupestris cv. 110 Richter (RG ) (110R )

Root-specific responses of Root-specific responses of Controlled levels of 110R soil water-deficit RG Z Higher root density (model parameter d) and lower RLA 1 Z Lower root density (model parameter d) and lower RLA 1 Z Higher number of xylem conduits regardless of SWD levels 2 Z Lower number of xylem conduits regardless of SWD levels 2 Z Drought-induced shift in xylem conduits frequency distribution 2 Z No shift in xylem conduits frequency distribution induced by Z Largest number of xylem conduits 2 2 drought Common root responses in size class 10-20 )m under MWD Z Higher Kh regardless of SWD 2 Z Lower Kh regardless of SWD 2 r Z No xylem conduits > 70 )m in root tips 2 r Z Higher number of down-regulated Z Lower number of down-regulated transcripts and lower number of up- transcripts and higher number of up- regulated transcripts 3 regulated transcripts 3 Z Down-regulation of KIP and up- Z No response to SWD levels of KIP regulation of CAX in response to and CAX genes 3 SWD levels 3 Z Up-regulation of AER, PBP and Z No response to SWD levels of LOX genes in responses to SWD 3 AER, PBP and LOX genes 3 Z Protective mechanisms against Z Specific GO enrichment related to drought-induced oxidative damage cell wall properties involving through detoxification process Expansins gene family 3 involving AER 3

Figure VI.1: Summary of the main results obtained in the present work. The drought-sensitive (RG) and the drought-tolerant (110R) are compared according to their common and specific responses in both roots and shoots. The bold superscript numbers refer to chapter numbers as follow: 1Chapter III, 2Chapter IV and 3Chapter V. The abbreviations used are: soil water content (SWC), soil water deficit (SWD), well- watered (WW), moderate water-deficit (MWD), high water-deficit (HWD), predawn leaf water potential (PD ), leaf water potential ( leaf ), transpiration rate ( E), stomatal conductance ( gs), minimum stomatal conductance ( gsmin ), maximum stomatal conductance ( gsmax ), xylem sap ABA concentration ([ABA]), leaf- specific hydraulic conductance (K leaf ), single root hydraulic conductivity (kh r) whole plant leaf area (LA), whole plant relative growth rate (RGR), root length area (RLA), gene ontology [GO] and several transcripts (2-alkenal reductase [AER], cation/H+ exchanger [CAX], kinase interacting protein [KIP], lipoxygenase [LOX], phosphatidylethanolamine binding protein [PBP]).

- 129 - Chapter VI. General conclusions and perspectives

Long-term water-deficit seems to affect root dynamics (Chap ter III) and functioning ( Chapter IV and Chapter V). Stress-induced adaptive responses in the roots are genotype- specific. The drought-tolerant genotype exhibits a substantial shift in root tips xylem conduit diameter under moderate water-deficit while the drought-sensitive genotype does not respond ( Chapter IV, Figure VI.1). Such responses suggest adaptive m echanisms against drought- induced cavitation, thus improving tolerance to water-deficit stress. Additionally, remodelling of root transcriptome in response to water-deficit ( Chapter V) differs between rootstocks (Figure VI.1). Transcriptomic analysis identified genotype-specific transc ripts that are regulated by water-deficit levels. The comparison between stress levels and genotypes identified 24 significant genes in “Treatment×Genotype” interactions ( Chapter V, Figure VI.1). Protection against drought-induced oxidative dam age seemed important mechanisms induced by the drought-tolerant rootstock, while the drought-sensitive respond to water-deficit by modification of cell wall properties ( Chapter V).

VI.2 Perspectives Genetic improvement and breeding new rootstock genotypes better adapted to fluctuating weather may contribute to the long-term strategies to aid in the mitigation of climate change. Advances in our understanding of the key regulatory mechanisms of grafted-grape physiology to water-limited environment, from the gene to whole plant, should build an integrative knowledge base and provide information on the genetic determinism and numerous targets for genetic improvement. The model approach developed in the present work can be applied in future studies focusing on the genetic determinism on the regulation of scion gas exchange induced by the rootstock under water-deficit stress. The availability of several genetic linkage maps from intra- or inter-specific Vitis crosses ( Doligez et al. , 2006 ; Lowe & Walker, 2006 ; Marguerit et al. , 2009 ) in concert with whole grape genome ( Jaillon et al. , 2007 ) and phenotyping tools used in the present work ( Sadok et al. , 2007 ) open new avenues in this direction. The use of next-generation sequencing (RNA sequencing, RNA-seq) to perform expression QTL (eQTL) mapping ( Majewski & Pastinen, 2011 ) combined with phenotyping methods, both at rootstock and scion levels, m ay be used to link molecular data to physiological data. These new biotechnologies will help the understanding of how gene expression variation contributes to phenotypic variation and will facilitate comprehensive investigations of the molecular mechanisms underlying drought adaptation and tolerance ( Ingvarsson & Street, 2011 ). An eQTL approach will be performed in the segregating population ( Marguerit et al. , 2009 ) using gene candidates (French projected VITSEC, funded by “agence national de la recherche [ANR]” and coordinated by our lab, F. Barrieu). The results will provide a functional characterization of QTL region s involved in water-deficit adaptation and will contribute to the identification of molecular markers of drought tolerance.

In the present work, the root system was shown to be an important genetic characteristic between rootstocks affecting both scion gas exchange ( Chapter III) and hydraulic properties ( Chapter IV). Knowledge on root system architecture dynam ic and functioning is limited,

- 130 - Chapter VI. General conclusions and perspectives largely due to the difficulty of observing and phenotyping roots. In the future, a more detailed examination of the root system in different scion-rootstock pairs is required. New experiments using in vitro micro-grafting growing in different gel-media combined with a platform for automatic phenotyping root traits ( e.g. Iyer-Pascuzzi et al. , 2010 ) would be effective to investigate the responses of a large number of plants. They will allow to associate root traits under different stress conditions imposed in the growing media to QTLs. Additionally, the determination of root hydraulic behaviour could be investigated simultaneously using nondestructive visualization of water flow and stress-induced cavitation (e.g. Van As, 2007 ; Brodersen et al. , 2010 ). Such techniques will expand the number of functionally root traits and will lead to broader knowledge regarding the genetic mechanisms underlying rootstocks adaptation to water-deficit and other abiotic stresses.

Genotype-specific genes identified under water-deficit levels ( Chapter V) provided robust candidate genes for future functional genom ic studies. Despite that transcriptomic is highly informative, integrating data into a molecular network offers a global view of genome function and allows the comparison of genotypes responses to water-deficit by identification of hubs with divergence and similarity. A weighted gene co-expression network ( Zhang & Horvath, 2005 ) is under construction in R using WGCNA package ( Langfelder & Horvath, 2008 ) to provide more genotype-specific candidate genes involved in global genom e regulation under water-deficit. In complement to transcriptomic analysis, proteomic, enzymatic and/or metabolomic approaches are required to take into account translational and post-translational regulation. In the present study, water-deficit may induce oxidative stress and reactive oxygen species (ROS) production, leading to lipid peroxidation and detoxification process. Measurements of several ROS enzymatic activity such as peroxidase are now possible with different assay kits. As an essential piece of evidence, metabolic profiling should substantiate transcriptomic analyses. In the potential role of transcripts in the lipid metabolism discussed in this work, it is worthwhile to focus on lipid-derived compounds. Analysis of more than 150 fatty acids and oxylipins can now be performed in parallel (possible collaboration with the lab of Prof. R. Lessire, CNRS UMR 5200). These measurements should be performed in our samples to gain further knowledge and demonstrate in vivo activity of such compounds. Several candidate genes provided in this work should be used in future functional genomic screening to identify and underline their roles in grape development and adaptation to abiotic stress. Grapevine transformation is difficult but possible ( reviewed by Bouquet et al. , 2003 ) and has been used with success in functional characterization ( e.g. Tesniere et al. , 2006 ). The methods are based on construction of transgenes and transformation to obtain transgenic plant from embryogenesis callus, cell suspensions or hairy roots. These available biotechnologies could be developed by using some candidate genes from the present study to foster our understanding of Vitis sp. root biology.

- 131 -

References

- 132 - References

Abe H., Urao T., Ito T. et al. (2003). Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptional activators in abscisic acid signaling. Plant Cell 15 , 63-78. Achard P., Cheng H., De Grauwe L. et al. (2006). Integration of plant responses to environmentally activated phytohormonal signals. Science 311 , 91-94. Ahuja I., de Vos R.C.H., Bones A.M. & Hall R.D. (2010). Plant molecular stress responses face climate change. Trends Plant Sci. 15 , 664-674. Aida M., Beis D., Heidstra R. et al. (2004). The PLETHORA genes mediate patterning of the Arabidopsis root stem cell niche. Cell 119 , 109-120. Aiken R.M. & Smucker A.J.M. (1996). Root system regulation of whole plant growth. Annu. Rev. Phytopathol. 34 , 325-346. Albacete A., Martínez-Andújar C., Ghanem M.E. et al. (2009). Rootstock-mediated changes in xylem ionic and hormonal status are correlated with delayed leaf senescence, and increased leaf area and crop productivity in salinized tomato. Plant Cell Environ. 32 , 928-938. Alcázar R., Altabella T., Marco F. et al. (2010a). Polyamines: molecules with regulatory functions in plant abiotic stress tolerance. Planta 231 , 1237-1249. Alcázar R., Planas J., Saxena T. et al. (2010b). Putrescine accumulation confers drought tolerance in transgenic Arabidopsis plants over-expressing the homologous Arginine decarboxylase 2 gene. Plant Physiol. Biochem. 48 , 547-552. Alexa A., Rahnenführer J. & Lengauer T. (2006). Improved scoring of functional groups from gene expression data by decorrelating GO graph structure. Bioinformatics 22 , 1600-1607. Allan R.P. & Soden B.J. (2008). Atmospheric warming and the amplification of precipitation extremes. Science 321 , 1481-1484. Allen M.R. & Ingram W.J. (2002). Constraints on future changes in climate and the hydrologic cycle. Nature 419 , 224-232. Allen R.G., Pereira L.S., Raes D. & Smith M. (1998). Crop evapotranspiration - Guidelines for computing crop water requirements - FAO Irrigation and drainage paper 56 . FAO - Food and Agriculture Organization of the United Nations, Rome, Italy, 328 p. Alleweldt G. & Possingham J.V. (1988). Progress in grapevine breeding. Theor. Appl. Genet. 75 , 669-673. Alleweldt G., Spiegel-Roy P. & Reisch B. (1990). Grapes ( Vitis ). Acta Hortic. 290 , 291-327. Alsina M.M., Smart D.R., Bauerle T. et al. (2011). Seasonal changes of whole root system conductance by a drought-tolerant grape root system. J. Exp. Bot. 62 , 99-109. Altschul S.F., Madden T.L., Schäffer A.A. et al. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25 , 3389- 3402. Anderson K., Norman D. & Wittwer G. (2003a). Globalisation of the world's wine markets. World Economy 26 , 659-687. Anderson L.J., Comas L.H., Lakso A.N. & Eissenstat D.M. (2003b). Multiple risk factors in root survivorship: a 4-year study in Concord grape. New Phytol. 158 , 489-501. Aoki K., Ogata Y. & Shibata D. (2007). Approaches for extracting practical information from gene co-expression networks in plant biology. Plant Cell Physiol. 48 , 381-390. Apel K. & Hirt H. (2004). Reactive oxygen species: metabolism, oxidative stress, and signal transduction. Annu. Rev. Plant Biol. 55 , 373-399. Arango-Velez A., Zwiazek J.J., Thomas B.R. & Tyree M.T. (2011). Stomatal factors and vulnerability of stem xylem to cavitation in poplars. Physiol. Plant. 143 , 154-165.

- 133 - References

Araujo F., Williams L.E., Grimes D.W. & Matthews M.A. (1995). A comparative study of young ‘Thompson Seedless’ grapevines ( Vitis vinifera L.) under drip and furrow irrigation. I. Root and soil water distributions. Sci. Hortic. 60 , 235-249. Archer E. & Strauss H.C. (1985). Effect of plant density on root distribution of three-year- old grafted 99 Richter grapevines. S. Afr. J. Enol. Vitic. 6, 25-30. Archer E. & Strauss H.C. (1989). The effect of plant spacing on the water status of soil and grapevines. S. Afr. J. Enol. Vitic. 10 , 49-58. Armengaud P., Zambaux K., Hills A. et al. (2009). EZ-Rhizo: integrated software for the fast and accurate measurement of root system architecture. Plant J. 57 , 945-956. Aroca R., Porcel R. & Ruiz-Lozano J.M. (2011). Regulation of root water uptake under abiotic stress conditions. J. Exp. Bot. In press, http://dx.doi.org/10.1093/jxb/err266 . Arrigo N. & Arnold C. (2007). Naturalised Vitis rootstocks in Europe and consequences to native wild grapevine. PLoS ONE 2, e521. Arroyo-García R., Ruiz-García L., Bolling L. et al. (2006). Multiple origins of cultivated grapevine ( Vitis vinifera L. ssp. sativa ) based on chloroplast DNA polymorphisms. Mol. Ecol. 15 , 3707-3714. Ashburner M., Ball C.A., Blake J.A. et al. (2000). Gene ontology: tool for the unification of biology. Nat. Genet. 25 , 25-29. Ashraf M. (2010). Inducing drought tolerance in plants: recent advances. Biotechnol. Adv. 28 , 169-183. Asins M., Bolarín M., Pérez-Alfocea F. et al. (2010). Genetic analysis of physiological components of salt tolerance conferred by Solanum rootstocks. What is the rootstock doing for the scion? Theor. Appl. Genet. 121 , 105-115. Atkinson C.J., Else M.A., Taylor L. & Dover C.J. (2003). Root and stem hydraulic conductivity as determinants of growth potential in grafted trees of apple ( Malus pumila Mill.). J. Exp. Bot. 54 , 1221-1229. Audeguin L., Boursiquot J.M., Charmont S. et al. (2007). Description des variétés de porte-greffes. In: Catalogue des variétés et clones de vigne cultivés en France. 2 ème édition (eds Institut Français de la Vigne et du vin (IFVV), INRA, Montpellier SupAgro & Viniflhor), pp. 327-359. CBE Production, Montpellier, France. Augé R.M. (2001). Water relations, drought and vesicular-arbuscular mycorrhizal symbiosis. Mycorrhiza 11 , 3-42. Augé R.M. & Moore J.L. (2002). Stomatal response to nonhydraulic root-to-shoot communication of partial soil drying in relation to foliar dehydration tolerance. Environ. Exp. Bot. 47 , 217-229. Baena-González E., Rolland F., Thevelein J.M. & Sheen J. (2007). A central integrator of transcription networks in plant stress and energy signalling. Nature 448 , 938-942. Bai L., Zhang G., Zhou Y. et al. (2009). Plasma membrane-associated proline-rich extensin- like receptor kinase 4, a novel regulator of Ca 2+ signalling, is required for abscisic acid responses in Arabidopsis thaliana . Plant J. 60 , 314-327. Baiges I., Schäffner A.R. & Mas A. (2001). Eight cDNA encoding putative aquaporins in Vitis hybrid Richter-110 and their differential expression. J. Exp. Bot. 52 , 1949-1951. Barker C.L., Donald T., Pauquet J. et al. (2005). Genetic and physical mapping of the grapevine powdery mildew resistance gene, Run1 , using a bacterial artificial chromosome library. Theor. Appl. Genet. 111 , 370-377. Barkla B.J., Hirschi K.D. & Pittman J.K. (2008). Exchangers man the pumps: functional interplay between proton pumps and proton-coupled Ca 2+ exchangers. Plant Signal. Behav. 3, 354-356. Bartels D. (2001). Targeting detoxification pathways: an efficient approach to obtain plants with multiple stress tolerance? Trends Plant Sci. 6, 284-286.

- 134 - References

Bartels D. & Sunkar R. (2005). Drought and salt tolerance in plants. Crit. Rev. Plant Sci. 24 , 23 - 58. Baskin T.I., Meekes H.T.H.M., Liang B.M. & Sharp R.E. (1999). Regulation of growth anisotropy in well-watered and water-stressed maize roots. II. Role of cortical microtubules and cellulose microfibrils. Plant Physiol. 119 , 681-692. Bassani M., Neumann P. & Gepstein S. (2004). Differential expression profiles of growth- related genes in the elongation zone of maize primary roots. Plant Mol. Biol. 56 , 367- 380. Bassoi L.H., Hopmans J.W., de Castro Jorge L.A. et al. (2003). Grapevine root distribution in drip and microsprinkler irrigation. Sci. Agr. 60 , 377-387. Basu P., Pal A., Lynch J.P. & Brown K.M. (2007). A novel image-analysis technique for kinematic study of growth and curvature. Plant Physiol. 145 , 305-316. Bates B.C., Kundzewicz Z.W., Wu S. & Palutikof J.P. (2008). Climate change and water. Technical paper of the Intergovernmental Panel on Climate Change . IPCC Secretariat, Geneva, Switzerland, 210 p. Bauerle T.L., Richards J.H., Smart D.R. & Eissenstat D.M. (2008a). Importance of internal hydraulic redistribution for prolonging the lifespan of roots in dry soil. Plant Cell Environ. 31 , 177-186. Bauerle T.L., Smart D.R., Bauerle W.L. et al. (2008b). Root foraging in response to heterogeneous soil moisture in two grapevines that differ in potential growth rate. New Phytol. 179 , 857-866. Bauerle T.L., Centinari M. & Bauerle W.L. (2011). Shifts in xylem vessel diameter and embolisms in grafted apple trees of differing rootstock growth potential in response to drought. Planta 234 , 1045-1054. Bavaresco L., Fregoni M. & Fraschini P. (1991). Investigations on iron uptake and reduction by excised roots of different grapevine rootstocks and a V. vinifera cultivar. Plant Soil 130 , 109-113. Bavaresco L., Fregoni M. & Perino A. (1994). Physiological aspects of lime-induced chlorosis in some Vitis species. 2. Genotype response to stress conditions. Vitis 34 , 233-234. Bellin D., Ferrarini A., Chimento A. et al. (2009). Combining next-generation pyrosequencing with microarray for large scale expression analysis in non-model species. BMC Genomics 10 , 555. Benfey P.N., Bennett M. & Schiefelbein J. (2010). Getting to the root of plant biology: impact of the Arabidopsis genome sequence on root research. Plant J. 61 , 992-1000. Bengough A.G. (1997). Modelling rooting depth and soil strength in a drying soil profile. J. Theor. Biol. 186 , 327-338. Bengough A.G., Gordon D.C., Al-Menaie H. et al. (2004). Gel observation chamber for rapid screening of root traits in cereal seedlings. Plant Soil 262 , 63-70. Bengough A.G., Bransby M.F., Hans J. et al. (2006). Root responses to soil physical conditions; growth dynamics from field to cell. J. Exp. Bot. 57 , 437-447. Benz M.J., Anderson M.M., Williams M.A. & Wolpert J.A. (2007). Viticultural performance of three Malbec clones on two rootstocks in Oakville, Napa valley, California. Am. J. Enol. Vitic. 58 , 262-267. Bessis R. (2007). Evolution of the grapevine ( Vitis vinifera L.) imprinted by natural and human factors. Can. J. Bot. 85 , 679-690. Birnbaum K., Shasha D.E., Wang J.Y. et al. (2003). A gene expression map of the Arabidopsis root. Science 302 , 1956-1960. Bishopp A., Benková E. & Helariutta Y. (2011). Sending mixed messages: auxin-cytokinin crosstalk in roots. Curr. Opin. Plant Biol. 14 , 10-16.

- 135 - References

Bisson L.F., Waterhouse A.L., Ebeler S.E. et al. (2002). The present and future of the international wine industry. Nature 418 , 696-699. Blum A. (2011). Plant breeding for water-limited environments . Springer, New York, USA, 255 p. Bogeat-Triboulot M.-B., Brosche M., Renaut J. et al. (2007). Gradual soil water depletion results in reversible changes of gene expression, protein profiles, ecophysiology, and growth performance in Populus euphratica , a poplar growing in arid regions. Plant Physiol. 143 , 876-892. Böhm W. (1979). Methods of studying root systems (vol. 33). Springer-Verlag, Berlin- Heidelberg-New York, 188 p. Böhmer M. & Schroeder J.I. (2011). Quantitative transcriptomic analysis of abscisic acid- induced and reactive oxygen species-dependent expression changes and proteomic profiling in Arabidopsis suspension cells. Plant J. 67 , 105-118. Boneh U., Biton I., Zheng C. et al. Characterization of potential ABA receptors in Vitis vinifera . Plant Cell Rep. In press, http://dx.doi.org/10.1007/s00299-011-1166-z . Bouché N. & Fromm H. (2004). GABA in plants: just a metabolite? Trends Plant Sci. 9, 110-115. Boudsocq M., Barbier-Brygoo H. & Laurière C. (2004). Identification of nine sucrose nonfermenting 1-related protein kinases 2 activated by hyperosmotic and saline stresses in Arabidopsis thaliana . J. Biol. Chem. 279 , 41758-41766. Bouma T., Nielsen K. & Koutstaal B. (2000). Sample preparation and scanning protocol for computerised analysis of root length and diameter. Plant Soil 218 , 185-196. Bouquet A., Danglot Y., Bongiovanni M. et al. (2000). Breeding rootstocks resistant to Grape fanleaf virus spread, using Vitis ×Muscadinia hybridization. Acta Hortic. 528 , 523-532. Bouquet A., Torregrosa L. & Chatelet P. (2003). Grapevine genetic engineering: tool for genome analysis or plant breeding method? Which future for transgenic vines? AgBiotechNet 5, ABN 116 , 1-10. Bouquet A., Torregrosa L., Iocco P. & Thomas M.R. (2009). Grapes. In: Compendium of transgenic crop plants. Volume 4: Transgenic temperate fruits and nuts (eds C. Kole & T.C. Hall), pp. 189-231. John Wiley & Sons Ltd, New York, USA. Boursiac Y., Boudet J., Postaire O. et al. (2008). Stimulus-induced downregulation of root water transport involves reactive oxygen species-activated cell signalling and plasma membrane intrinsic protein internalization. Plant J. 56 , 207-218. Boursiquot J.-M., Lacombe T., Laucou V. et al. (2009). Parentage of Merlot and related winegrape cultivars of southwestern France: discovery of the missing link. Aust. J. Grape Wine Res. 15 , 144-155. Bowers J., Boursiquot J.-M., This P. et al. (1999). Historical genetics: the parentage of Chardonnay, Gamay, and other wine grapes of northeastern France. Science 285 , 1562-1565. Bowers J.E. & Meredith C.P. (1997). The parentage of a classic wine grape, Cabernet Sauvignon. Nat. Genet. 16 , 84-87. Boyer J.S. (1982). Plant productivity and environment. Science 218 , 443-448. Bradshaw A.D. (1965). Evolutionary significance of phenotypic plasticity in plants. In: Adv. Genet. (eds E.W. Caspari & J.M. Thoday), pp. 115-155. Academic Press inc., London. Brady S.M., Orlando D.A., Lee J.-Y. et al. (2007). A high-resolution root spatiotemporal map reveals dominant expression patterns. Science 318 , 801-806. Bramley H., Turner N.C., Turner D.W. & Tyerman S.D. (2009). Roles of morphology, anatomy, and aquaporins in determining contrasting hydraulic behavior of roots. Plant Physiol. 150 , 348-364.

- 136 - References

Branas J. & Vergnes A. (1957). Morphologie du système radiculaire de la vigne. Prog. Agric. Vitic. 74 , (3-4) 29-32, (27-28) 94-104, (109-110) 122-129, (111-112) 147-158, (113) 173-183, (114-115) 203-209. Bray E.A. (1997). Plant responses to water deficit. Trends Plant Sci. 2, 48-54. Bray E.A. (2004). Genes commonly regulated by water-deficit stress in Arabidopsis thaliana . J. Exp. Bot. 55 , 2331-2341. Brenner A.J. & Jarvis P.G. (1995). A heated leaf replica technique for determination of leaf boundary layer conductance in the field. Agric. For. Meteorol. 72 , 261-275. Breshears D.D., Cobb N.S., Rich P.M. et al. (2005). Regional vegetation die-off in response to global-change-type drought. Proc. Natl. Acad. Sci. 102 , 15144-15148. Brodersen C.R., McElrone A.J., Choat B. et al. (2010). The dynamics of embolism repair in xylem: in vivo visualizations using high-resolution computed tomography. Plant Physiol. 154 , 1088-1095. Brodersen C.R., Lee E.F., Choat B. et al. (2011). Automated analysis of three-dimensional xylem networks using high-resolution computed tomography. New Phytol. 191 , 1168- 1179. Buckley T.N. (2005). The control of stomata by water balance. New Phytol. 168 , 275-292. Caffarra A. & Eccel E. (2011). Projecting the impacts of climate change on the phenology of grapevine in a mountain area. Aust. J. Grape Wine Res. 17 , 52-61. Capell T., Bassie L. & Christou P. (2004). Modulation of the polyamine biosynthetic pathway in transgenic rice confers tolerance to drought stress. Proc. Natl. Acad. Sci. 101 , 9909-9914. Carbonneau A. (1985). The early selection of grapevine rootstocks for resistance to drought conditions. Am. J. Enol. Vitic. 36 , 195-198. Carlquist S. (1985). Observations on functional wood histology of vines and lianas: vessel dimorphism, tracheids, vasicentric tracheids, narrow vessels, and parenchyma. Aliso 11 , 139-157. Carneiro L.C. & Menezes-Sequeira L. (1989). Distribuição radical de porta-enxertos: contributo para o seu estudo. Ciência e Técnica Vitivinícola 8, 87-95. Carvalho B.S. & Irizarry R.A. (2010). A framework for oligonucleotide microarray preprocessing. Bioinformatics 26 , 2363-2367. Castellarin S.D., Matthews M.A., Di Gaspero G. & Gambetta G.A. (2007). Water deficits accelerate ripening and induce changes in gene expression regulating flavonoid biosynthesis in grape berries. Planta 227 , 101-112. Castellucci F. (2010). World statistics: OIV report on the state of the vitiviniculture world market in 2009 . Paper presented at the 8 th General Assembly of the OIV, Tbilisi, Georgia. Organisation Internationale de la Vigne et du Vin (OIV). Celette F., Wery J., Chantelot E. et al. (2005). Belowground interactions in a vine ( Vitis vinifera L.)-tall fescue ( Festuca arundinacea Shreb.) intercropping system: water relations and growth. Plant Soil 276 , 205-217. Celette F., Gaudin R. & Gary C. (2008). Spatial and temporal changes to the water regime of a Mediterranean vineyard due to the adoption of cover cropping. Eur. J. Agron. 29 , 153-162. Chaffey N. (2002). Wood microscopical techniques. In: Wood formation in trees: cell and molecular biology techniques (ed N. Chaffey), pp. 17-40. Taylor & Francis Group, New York, USA. Chapman D.M., Roby G., Ebeler S.E. et al. (2005). Sensory attributes of Cabernet Sauvignon wines made from vines with different water status. Aust. J. Grape Wine Res. 11 , 339-347.

- 137 - References

Chatelet D.S., Matthews M.A. & Rost T.L. (2006). Xylem structure and connectivity in grapevine ( Vitis vinifera ) shoots provides a passive mechanism for the spread of bacteria in grape plants. Ann. Bot. 98 , 483-494. Chavarría-Krauser A., Nagel K.A., Palme K. et al. (2008). Spatio-temporal quantification of differential growth processes in root growth zones based on a novel combination of image sequence processing and refined concepts describing curvature production. New Phytol. 177 , 811-821. Chaves M.M., Maroco J.P. & Pereira J.S. (2003). Understanding plant responses to drought - from genes to the whole plant. Funct. Plant Biol. 30 , 239-264. Chaves M.M., Santos T.P., Souza C.R. et al. (2007). Deficit irrigation in grapevine improves water-use efficiency while controlling vigour and production quality. Ann. Appl. Biol. 150 , 237-252. Chaves M.M., Zarrouk O., Francisco R. et al. (2010). Grapevine under deficit irrigation: hints from physiological and molecular data. Ann. Bot. 105 , 661-676. Chaves M.M., Miguel Costa J. & Madeira Saibo N.J. (2011). Recent advances in photosynthesis under drought and salinity. Adv. Bot. Res. 57 , 49-104. Chen C.-W., Yang Y.-W., Lur H.-S. et al. (2006). A novel function of abscisic acid in the regulation of rice ( Oryza sativa L.) root growth and development. Plant Cell Physiol. 47 , 1-13. Chen G., Lips S.H. & Sagi M. (2002). Biomass production, transpiration rate and endogenous abscisic acid levels in grafts of flacca and wild-type tomato (Lycopersicon esculentum ). Funct. Plant Biol. 29 , 1329-1335. Chen T.H.H. & Murata N. (2008). Glycinebetaine: an effective protectant against abiotic stress in plants. Trends Plant Sci. 13 , 499-505. Chen Y.L., Dunbabin V.M., Diggle A.J. et al. (2011). Development of a novel semi- hydroponic phenotyping system for studying root architecture. Funct. Plant Biol. 38 , 355-363. Chenu K., Chapman S.C., Tardieu F. et al. (2009). Simulating the yield impacts of organ- level quantitative trait loci associated with drought response in maize: a "gene-to- phenotype" modeling approach. Genetics 183 , 1507-1523. Choat B., Cobb A.R. & Jansen S. (2008). Structure and function of bordered pits: new discoveries and impacts on whole-plant hydraulic function. New Phytol. 177 , 608-626. Choat B., Drayton W.M., Brodersen C. et al. (2010). Measurement of vulnerability to water stress-induced cavitation in grapevine: a comparison of four techniques applied to a long-vesseled species. Plant Cell Environ. 33 , 1502-1512. Chouzouri A. & Schultz H.R. (2005). Hydraulic anatomy, cavitation susceptibility and gas- exchange of several grapevine cultivars of different geographic origin. Acta Hortic. 689 , 325-332. Christmann A., Weiler E.W., Steudle E. & Grill E. (2007). A hydraulic signal in root-to- shoot signalling of water shortage. Plant J. 52 , 167-174. Cipriani G., Spadotto A., Jurman I. et al. (2010). The SSR-based molecular profile of 1005 grapevine ( Vitis vinifera L.) accessions uncovers new synonymy and parentages, and reveals a large admixture amongst varieties of different geographic origin. Theor. Appl. Genet. 121 , 1569-1585. Cirami R., Furkaliev J. & Radford R. (1994). Summer drought and vine rootstocks. Australian and New Zealand Grapegrower and Winemaker 366 , 145. Clark R.T., MacCurdy R.B., Jung J.K. et al. (2011). Three-dimensional root phenotyping with a novel imaging and software platform. Plant Physiol. 156 , 455-465. Clearwater M.J., Lowe R.G., Hofstee B.J. et al. (2004). Hydraulic conductance and rootstock effects in grafted vines of kiwifruit. J. Exp. Bot. 55 , 1371-1382.

- 138 - References

Clingeleffer P.R. & Emmanuelli D.R. (2006). An assessment of rootstocks for Sunmuscat (Vitis vinifera L.): a new drying variety. Aust. J. Grape Wine Res. 12 , 135–140. Cochard H., Bréda N. & Granier A. (1996). Whole tree hydraulic conductance and water loss regulation in Quercus during drought: evidence for stomatal control of embolism? Ann. For. Sci. 53 , 197-206. Cochard H., Coll L., Le Roux X. & Ameglio T. (2002). Unraveling the effects of plant hydraulics on stomatal closure during water stress in walnut. Plant Physiol. 128 , 282- 290. Cochard H. (2006). Cavitation in trees. C. R. Physique 7, 1018-1026. Cohen D., Bogeat-Triboulot M.-B., Tisserant E. et al. (2010). Comparative transcriptomics of drought responses in Populus: a meta-analysis of genome-wide expression profiling in mature leaves and root apices across two genotypes. BMC Genomics 11 , 630. Cohen S. & Naor A. (2002). The effect of three rootstocks on water use, canopy conductance and hydraulic parameters of apple trees and predicting canopy from hydraulic conductance. Plant Cell Environ. 25 , 17-28. Cohen S., Naor A., Bennink J. et al. (2007). Hydraulic resistance components of mature apple trees on rootstocks of different vigours. J. Exp. Bot. 58 , 4213-4224. Cole E.S. & Mahall B.E. (2006). A test for hydrotropic behavior by roots of two coastal dune shrubs. New Phytol. 172 , 358-368. Collard B.C.Y. & Mackill D.J. (2008). Marker-assisted selection: an approach for precision plant breeding in the twenty-first century. Phil. Trans. R. Soc. B 363 , 557-572. Collins N.C., Tardieu F. & Tuberosa R. (2008). Quantitative trait loci and crop performance under abiotic stress: where do we stand? Plant Physiol. 147 , 469-486. Comas L.H., Eissenstat D.M. & Lakso A.N. (2000). Assessing root death and root system dynamics in a study of grape canopy pruning. New Phytol. 147 , 171-178. Comas L.H., Anderson L.J., Dunst R.M. et al. (2005). Canopy and environmental control of root dynamics in a long-term study of Concord grape. New Phytol. 167 , 829-840. Comas L.H., Bauerle T.L. & Eissenstat D.M. (2010). Biological and environmental factors controlling root dynamics and function: effects of root ageing and soil moisture. Aust. J. Grape Wine Res. 16 , 131-137. Cominelli E. & Tonelli C. (2010). Transgenic crops coping with water scarcity. New Biotechnol. 27 , 473-477. Comstock J.P. (2002). Hydraulic and chemical signalling in the control of stomatal conductance and transpiration. J. Exp. Bot. 53 , 195-200. Coombe B.G. (1995). Adoption of a system for identifying grapevine growth stages. Aust. J. Grape Wine Res. 1, 104-110. Cooper M., van Eeuwijk F.A., Hammer G.L. et al. (2009). Modeling QTL for complex traits: detection and context for plant breeding. Curr. Opin. Plant Biol. 12 , 231-240. Cordeau J. (2002). Création d’un vignoble. Greffage de la vigne et porte-greffes. Élimination des maladies à virus . (2nd ed.). Édition Féret, Bordeaux, France, 182 p. Correia M.J., Pereira J.S., Chaves M.M. et al. (1995). ABA xylem concentrations determine maximum daily leaf conductance of field-grown Vitis vinifera L. plants. Plant Cell Environ. 18 , 511-521. Cosgrove D.J. (2000). New genes and new biological roles for expansins. Curr. Opin. Plant Biol. 3, 73-78. Cosgrove D.J. (2005). Growth of the plant cell wall. Nat. Rev. Mol. Cell Biol. 6, 850-861. Cousins P. (2005). Evolution, genetics, and breeding: viticultural applications of the origins of our rootstocks. In: Grapevine rootstocks: current use, research and application (eds P. Cousins & R.K. Striegler), pp. 1-7. Proceedings of the 2005 Rootstock

- 139 - References

Sym posium. Mid-America Viticulture and Enology Center, Southwest Missouri State University, Mountain Grove Campus, Osage Beach, Missouri, USA. Cramer G.R., Ergül A., Grimplet J. et al. (2007). Water and salinity stress in grapevines: early and late changes in transcript and metabolite profiles. Funct. Integr. Genomics 7, 111-134. Cramer G.R. (2010). Abiotic stress and plant responses from the whole vine to the genes. Aust. J. Grape Wine Res. 16 , 86-93. Cushman J.C. & Bohnert H.J. (2000). Genomic approaches to plant stress tolerance. Curr. Opin. Plant Biol. 3, 117-124. Cutler S.R., Rodriguez P.L., Finkelstein R.R. & Abrams S.R. (2010). Abscisic acid: emergence of a core signaling network. Annu. Rev. Plant Biol. 61 , 651-679. DaMatta F.M., Grandis A., Arenque B.C. & Buckeridge M.S. (2010). Impacts of climate changes on crop physiology and food quality. Food Res. Int. 43 , 1814-1823. Damour G., Simonneau T., Cochard H. & Urban L. (2010). An overview of models of stomatal conductance at the leaf level. Plant Cell Environ. 33 , 1419-1438. Danjon F. & Reubens B. (2008). Assessing and analyzing 3D architecture of woody root systems, a review of methods and applications in tree and soil stability, resource acquisition and allocation. Plant Soil 303 , 1-34. Daudet F.A., Silvestre J., Ferreira M.I. et al. (1998). Leaf boundary layer conductance in a vineyard in Portugal. Agric. For. Meteorol. 89 , 255-267. Davies W.J. & Zhang J. (1991). Root signals and the regulation of growth and development of plants in drying soil. Annu. Rev. Plant Physiol. Plant Mol. Biol. 42 , 55-76. Davies W.J., Tardieu F. & Trejo C.L. (1994). How do chemical signals work in plants that grow in drying soil? Plant Physiol. 104 , 309-314. Davies W.J., Wilkinson S. & Loveys B. (2002). Stomatal control by chemical signalling and the exploitation of this mechanism to increase water use efficiency in agriculture. New Phytol. 153 , 449-460. Davies W.J., Kudoyarova G. & Hartung W. (2005). Long-distance ABA signaling and its relation to other signaling pathways in the detection of soil drying and the mediation of the plant’s response to drought. J. Plant Growth Regul. 24 , 285-295. Davies W.J. (2007). Responses of plant growth and functioning to changes in water supply in a changing climate. In: Plant Growth and Climate Change (eds J.I.L. Morison & M.D. Morecroft), pp. 96-117. Blackwell Publishing Ltd, Oxford, UK. de Bruxelles G.L., Peacock W.J., Dennis E.S. & Dolferus R. (1996). Abscisic acid induces the alcohol dehydrogenase gene in Arabidopsis. Plant Physiol. 111 , 381-391. de Dorlodot S., Forster B., Pagès L. et al. (2007). Root system architecture: opportunities and constraints for genetic improvement of crops. Trends Plant Sci. 12 , 474-481. de Herralde F., del Mar Alsina M., Aranda X. et al. (2006). Effects of rootstock and irrigation regime on hydraulic architecture of Vitis vinifera L. cv Tempranillo. J. Int. Sci. Vigne Vin 40 , 133-139. de Herralde F., Savé R., Aranda X. et al. (2010). Grapevine roots and soil environment: growth, distribution and function. In: Methodologies and results in grapevine research (eds S. Delrot, H. Medrano, E. Or, L. Bavaresco & S. Grando), pp. 1-20. Springer, Heidelberg - New York. De Smet I., Zhang H., Inzé D. & Beeckman T. (2006). A novel role for abscisic acid emerges from underground. Trends Plant Sci. 11 , 434-439. De Smet I., Vassileva V., De Rybel B. et al. (2008). Receptor-like kinase ACR4 restricts formative cell divisions in the Arabidopsis root. Science 322 , 594-597. Deak K.I. & Malamy J. (2005). Osmotic regulation of root system architecture. Plant J. 43 , 17-28.

- 140 - References

Degenkolbe T., Do P., Zuther E. et al. (2009). Expression profiling of rice cultivars differing in their tolerance to long-term drought stress. Plant Mol. Biol. 69 , 133-153. Delas J.J. (1992). Criteria used for rootstock selection in France. In: Rootstock seminar: a worldwide perspective (eds J.A. Wolpert, M.A. Walker & E. Weber), pp. 1-14. American Society for Enology and Viticulture, Davis, USA. Deluc L.G., Quilici D.R., Decendit A. et al. (2009). Water deficit alters differentially metabolic pathways affecting important flavor and quality traits in grape berries of Cabernet Sauvignon and Chardonnay. BMC Genomics 10 , 212. Delzon S., Douthe C., Sala A. & Cochard H. (2010). Mechanism of water-stress induced cavitation in conifers: bordered pit structure and function support the hypothesis of seal capillary-seeding. Plant Cell Environ. 33 , 2101-2111. Demiral T., Turkan I. & Sekmen A.H. (2011). Signalling strategies during drought and salinity, recent news. Adv. Bot. Res. 57 , 293-317. Di Vecchi Staraz M., Bandinelli R., Boselli M. et al. (2007). Genetic structuring and parentage analysis for evolutionary studies in grapevine: kin group and origin of the cultivar Sangiovese revealed. J. Am. Soc. Hortic. Sci. 132 , 514-524. Dimkpa C., Weinand T. & Asch F. (2009). Plant-rhizobacteria interactions alleviate abiotic stress conditions. Plant Cell Environ. 32 , 1682-1694. Ding X., Richter T., Chen M. et al. (2009). A rice kinase-protein interaction map. Plant Physiol. 149 , 1478-1492. Dinneny J.R., Long T.A., Wang J.Y. et al. (2008). Cell identity mediates the response of Arabidopsis roots to abiotic stress. Science 320 , 942-945. Dodd I. (2005). Root-to-shoot signalling: assessing the roles of ‘up’ in the up and down world of long-distance signalling in planta. Plant Soil 274 , 251-270. Dodd I.C., Stikic R. & Davies W.J. (1996). Chemical regulation of gas exchange and growth of plants in drying soil in the field. J. Exp. Bot. 47 , 1475-1490. Dodd I.C., Egea G. & Davies W.J. (2008a). Abscisic acid signalling when soil moisture is heterogeneous: decreased photoperiod sap flow from drying roots limits abscisic acid export to the shoots. Plant Cell Environ. 31 , 1263-1274. Dodd I.C., Egea G. & Davies W.J. (2008b). Accounting for sap flow from different parts of the root system improves the prediction of xylem ABA concentration in plants grown with heterogeneous soil moisture. J. Exp. Bot. 59 , 4083-4093. Dodd I.C., Theobald J.C., Richer S.K. & Davies W.J. (2009). Partial phenotypic reversion of ABA-deficient flacca tomato ( Solanum lycopersicum ) scions by a wild-type rootstock: normalizing shoot ethylene relations promotes leaf area but does not diminish whole plant transpiration rate. J. Exp. Bot. 60 , 4029-4039. Dodd I.C., Egea G., Watts C.W. & Whalley W.R. (2010). Root water potential integrates discrete soil physical properties to influence ABA signalling during partial rootzone drying. J. Exp. Bot. 61 , 3543-3551. Doligez A., Adam-Blondon A., Cipriani G. et al. (2006). An integrated SSR map of grapevine based on five mapping populations. Theor. Appl. Genet. 113 , 369-382. Doussan C., Pierret A., Garrigues E. & Pagès L. (2006). Water uptake by plant roots: II – Modelling of water transfer in the soil root-system with explicit account of flow within the root system – Comparison with experiments. Plant Soil 283 , 99-117. Drake B.G., Gonzàlez-Meler M.A. & Long S.P. (2003). More efficient plants: a consequence of rising atmospheric CO 2? Annu. Rev. Plant Physiol. Plant Mol. Biol. 48 , 609-639. Dray S. & Dufour A.-B. (2007). The ade4 package: implementing the duality diagram for ecologists. J. Stat. Softw. 22 , 1-20.

- 141 - References

Draye X., Kim Y., Lobet G. & Javaux M. (2010). Model-assisted integration of physiological and environmental constraints affecting the dynamic and spatial patterns of root water uptake from soils. J. Exp. Bot. 61 , 2145-2155. Dry N. (2007). Grapevine rootstocks: selection and management for South Australian vineyards . Lythrum Press (Adelaide), Stepney, Australia, 85 p. Duchêne E. & Schneider C. (2005). Grapevine and climatic changes: a glance at the situation in Alsace. Agron. Sustain. Dev. 25 , 93-99. Duchêne E., Huard F., Dumas V. et al. (2010). The challenge of adapting grapevine varieties to climate change. Clim. Res. 41 , 193-204. Dupuy L., Gregory P.J. & Bengough A.G. (2010). Root growth models: towards a new generation of continuous approaches. J. Exp. Bot. 61 , 2131-2143. Durand T., Bultel-Poncé V., Guy A. et al. (2009). New bioactive oxylipins formed by non- enzymatic free-radical-catalyzed pathways: the phytoprostanes. Lipids 44 , 875-888. Düring H. (1984). Evidence for osmotic adjustment to drought in grapevines ( Vitis vinifera L.). Vitis 23 , 1-10. Düring H. (1994). Photosynthesis of ungrafted and grafted grapevines: effects of rootstock genotype and plant age. Am. J. Enol. Vitic. 45 , 297-299. Düring H. & Dry P.R. (1995). Osmoregulation in water stressed roots: responses of leaf conductance and photosynthesis. Vitis 34 , 15-17. Eapen D., Barroso M.L., Campos M.E. et al. (2003). A no hydrotropic response root mutant that responds positively to gravitropism in Arabidopsis . Plant Physiol. 131 , 536-546. Eapen D., Barroso M.L., Ponce G. et al. (2005). Hydrotropism: root growth responses to water. Trends Plant Sci. 10 , 44-50. Edwards E.J., Smithson L., Graham D.C. & Clingeleffer P.R. (2011). Grapevine canopy response to a high-temperature event during deficit irrigation. Aust. J. Grape Wine Res. 17 , 153-161. Eisen M.B., Spellman P.T., Brown P.O. & Botstein D. (1998). Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. 95 , 14863-14868. Eissenstat D.M., Wells C.E., Yanai R.D. & Whitbeck J.L. (2000). Building roots in a changing environment: implications for root longevity. New Phytol. 147 , 33-42. Enstone D.E., Peterson C.A. & Ma F. (2002). Root endodermis and exodermis: structure, function, and responses to the environment. J. Plant Growth Regul. 21 , 335-351. Epstein E. (2004). Plant biologists need to get back to their roots. Nature 430 , 829-829. Erickson R.O. & Silk W.K. (1980). The kinematics of plant growth. Sci. Am. 242 , 134-151. Esmenjaud D., Van Ghelder C., Voisin R. et al. (2010). Host suitability of Vitis and Vitis- Muscadinia material to the nematode Xiphinema index over one to four years. Am. J. Enol. Vitic. 61 , 96-101. Ezzahouani A. & Williams L.E. (1995). The influence of rootstock on leaf water potential, yield, and berry composition of Ruby Seedless grapevines. Am. J. Enol. Vitic. 46 , 559- 563. Ezzahouani A. & Williams L.E. (2005). Performance of Dattier de Beyrouth and Alphonse Lavallée grapevines on eight rootstocks under dry-land conditions. J. Int. Sci. Vigne Vin 39 , 91-94. Fan L., Linker R., Gepstein S. et al. (2006). Progressive inhibition by water deficit of cell wall extensibility and growth along the elongation zone of maize roots is related to increased lignin metabolism and progressive stelar accumulation of wall phenolics. Plant Physiol. 140 , 603-612.

- 142 - References

Fang S., Yan X. & Liao H. (2009). 3D reconstruction and dynamic modeling of root architecture in situ and its application to crop phosphorus research. Plant J. 60 , 1096- 1108. FAO (2011a). AQUASTAT: water use . Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. http://www.fao.org/nr/aquastat , accessed May, 2011. FAO (2011b). FAOSTAT: crops production . Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. http://faostat.fao.org , accessed May, 2011. Farquhar G.D., Ehleringer J.R. & Hubick K.T. (1989). Carbon isotope discrimination and photosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40 , 503-537. Fernandez O., Béthencourt L., Quero A. et al. (2010). Trehalose and plant stress responses: friend or foe? Trends Plant Sci. 15 , 409-417. Feuillet C., Leach J.E., Rogers J. et al. (2011). Crop genome sequencing: lessons and rationales. Trends Plant Sci. 16 , 77-88. Feussner I. & Wasternack C. (2002). The lipoxygenase pathway. Annu. Rev. Plant Biol. 53 , 275-297. Flexas J., Galmés J., Gallé A. et al. (2010). Improving water use efficiency in grapevines: potential physiological targets for biotechnological improvement. Aust. J. Grape Wine Res. 16 , 106-121. Fonti P., von Arx G., García-González I. et al. (2010). Studying global change through investigation of the plastic responses of xylem anatomy in tree rings. New Phytol. 185 , 42-53. Fouquet R. (2005). Les aquaporines de vigne: identification, études d'expression en conditions de contraintes abiotiques et approches de caractérisation fonctionnelle . PhD Thesis, University of Bordeaux 1, Bordeaux, 124 p. Frazier T., Sun G., Burklew C. & Zhang B. (2011). Salt and drought stresses induce the aberrant expression of microRNA genes in tobacco. Mol. Biotechnol. 49 , 159-165. Freeman B.M. & Smart R.E. (1976). Research note: a root observation laboratory for studies with grapevines. Am. J. Enol. Vitic. 27 , 36-39. Fregoni M. (1977). Irrigation des vignobles: problèmes de physiologie, de biochimie, de génétique, d'agronomie, de technologie et d'économie. Bull. OIV 50 , 3-19. Fregoni M., Scienza A. & Miravalle R. (1978). Évaluation précoce de la résistance des porte-greffes à la sécheresse. In: Génétique et amélioration de la vigne, II e Symposium international sur l'amélioration de la vigne, Bordeaux, 14-18 juin 1977 (ed INRA), pp. 287-295. Institut National de la Recherche Agronomique, Paris. French A., Ubeda-Tomas S., Holman T.J. et al. (2009). High-throughput quantification of root growth using a novel image-analysis tool. Plant Physiol. 150 , 1784-1795. Fu X. & Harberd N.P. (2003). Auxin promotes Arabidopsis root growth by modulating gibberellin response. Nature 421 , 740-743. Fujita M., Fujita Y., Maruyama K. et al. (2004). A dehydration-induced NAC protein, RD26, is involved in a novel ABA-dependent stress-signaling pathway. Plant J. 39 , 863-876. Fukaki H., Okushima Y., Tasaka M. & Kwang W.J. (2007). Auxin-mediated lateral root formation in higher plants. Int. Rev. Cytol. 256 , 111-137. Fukaki H. & Tasaka M. (2009). Hormone interactions during lateral root formation. Plant Mol. Biol. 69 , 437-449. Galet P. (1982). Phylloxera. In: Les maladies et parasites de la vigne, Tome 2: les parasites animaux (ed P. Galet), pp. 1059-1313. Imprimerie du "Paysan du Midi", Montpellier. Galet P. (1988). Cépages et vignobles de France. Tome1: Les vignes Américaines . (2nd ed.). Imprimerie Déhan, Montpellier, 553 p.

- 143 - References

Galet P. (1990). Cépages et vignobles de France. Tome2: L'ampélographie Française . (2nd ed.). Imprimerie Déhan, Montpellier, 400 p. Galet P. (1998). Grape varieties and rootstock varieties . Oenoplurimedia, Chaintré, France, 315 p. Galet P. (2000). Précis de viticulture . JF Impression, Saint-Jean de Védas, France, 602 p. Galinha C., Hofhuis H., Luijten M. et al. (2007). PLETHORA proteins as dose-dependent master regulators of Arabidopsis root development. Nature 449 , 1053-1057. Galmés J., Cifre J., Medrano H. & Flexas J. (2005). Modulation of relative growth rate and its components by water stress in Mediterranean species with different growth forms. Oecologia 145 , 21-31. Galmés J., Pou A., Alsina M.M. et al. (2007). Aquaporin expression in response to different water stress intensities and recovery in Richter-110 ( Vitis sp.): relationship with ecophysiological status. Planta 226 , 671-681. Garcia M., Gallego P., Daverède C. & Ibrahim H. (2001a). Effect of three rootstocks on grapevine ( Vitis vinifera L.) cv. Négrette, grown hydroponically. I. Potassium, calcium and magnesium nutrition. S. Afr. J. Enol. Vitic. 22 , 101-103. Garcia M., Ibrahim H., Gallego P. & Puig P. (2001b). Effect of three rootstocks on grapevine ( Vitis vinifera L.) cv. Négrette, grown hydroponically. II. Acidity of musts and wines. S. Afr. J. Enol. Vitic. 22 , 104-106. Garris A., Cousins P., Ramming D. & Baldo A. (2009). Parentage analysis of Freedom rootstock. Am. J. Enol. Vitic. 60 , 357-361. Gascó A., Nardini A., Raimondo F. et al. (2007). Hydraulic kinetics of the graft union in different Olea europaea L. scion/rootstock combinations. Environ. Exp. Bot. 60 , 245- 250. Génard M., Bertin N., Gautier H. et al. (2010). Virtual profiling: a new way to analyse phenotypes. Plant J. 62 , 344-355. Gentleman R., Carey V., Bates D. et al. (2004). Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5, R80. Gibberd M.R., Walker R.R., Blackmore D.H. & Condon A.G. (2001). Transpiration efficiency and carbon-isotope discrimination of grapevines grown under well-watered conditions in either glasshouse or vineyard. Aust. J. Grape Wine Res. 7, 110-117. Goes da Silva F., Iandolino A., Al-Kayal F. et al. (2005). Characterizing the grape transcriptome. Analysis of expressed sequence tags from multiple Vitis species and development of a compendium of gene expression during berry development. Plant Physiol. 139 , 574-597. Gökbayrak Z., Söylemezoglu G., Akkurt M. & Çelik H. (2007). Determination of grafting compatibility of grapevine with electrophoretic methods. Sci. Hortic. 113 , 343-352. Gómez-del-Campo M., Ruiz C. & Lissarrague J.R. (2002). Effect of water stress on leaf area development, photosynthesis, and productivity in Chardonnay and Airén grapevines. Am. J. Enol. Vitic. 53 , 138-143. Gómez-del-Campo M., Baeza P., Ruiz C. & Lissarrague J.R. (2005). Effects of water stress on dry matter content and partitioning in four grapevine cultivars ( Vitis vinifera L.). J. Int. Sci. Vigne Vin 39 , 1-10. Gonçalves B., Correia C.M., Silva A.P. et al. (2007). Variation in xylem structure and function in roots and stems of scion–rootstock combinations of sweet cherry tree (Prunus avium L.). Trees-Struc. Func. 21 , 121-130. Gong H., Blackmore D.H. & Walker R.R. (2010). Organic and inorganic anions in Shiraz and Chardonnay grape berries and wine as affected by rootstock under saline conditions. Aust. J. Grape Wine Res. 16 , 227-236.

- 144 - References

Gong H., Blackmore D., Clingeleffer P. et al. (2011). Contrast in chloride exclusion between two grapevine genotypes and its variation in their hybrid progeny. J. Exp. Bot. 62 , 989-999. Gou J., Strauss S.H., Tsai C.J. et al. (2010). Gibberellins regulate lateral root formation in Populus through interactions with auxin and other hormones. Plant Cell 22 , 623-639. Goulet E. & Barbeau G. (2006). Contribution of soil electric resistivity measurements to the studies on soil/grapevine water relations. J. Int. Sci. Vigne Vin 40 , 57-69. Granett J., Walker M.A., Kocsis L. & Omer A.D. (2001). Biology and management of grape phylloxera. Annu. Rev. Entomol. 46 , 387-412. Granier C., Aguirrezabal L., Chenu K. et al. (2006). PHENOPSIS, an automated platform for reproducible phenotyping of plant responses to soil water deficit in A rabidopsis thaliana permitted the identification of an accession with low sensitivity to soil water deficit. New Phytol. 169 , 623-635. Granier C. & Tardieu F. (2009). Multi-scale phenotyping of leaf expansion in response to environmental changes: the whole is more than the sum of parts. Plant Cell Environ. 32 , 1175-1184. Grant R.S. & Matthews M.A. (1996a). The influence of phosphorus availability and rootstock on root system characteristics, phosphorus uptake, phosphorus partitioning, and growth efficiency. Am. J. Enol. Vitic. 47 , 403-409. Grant R.S. & Matthews M.A. (1996b). The influence of phosphorus availability, scion, and rootstock on grapevine shoot growth, leaf area, and petiole phosphorus concentration. Am. J. Enol. Vitic. 47 , 217-224. Grechi I., Vivin P., Hilbert G. et al. (2007). Effect of light and nitrogen supply on internal C:N balance and control of root-to-shoot biomass allocation in grapevine. Environ. Exp. Bot. 59 , 139-149. Gregory P.J. (2006). Plant roots: growth, activity and interactions with the soil . Blackwell Publishing, Garsington Road, Oxford, UK, 328 p. Gregory P.J., Bengough A.G., Grinev D. et al. (2009). Root phenomics of crops: opportunities and challenges. Funct. Plant Biol. 36 , 922-929. Grieneisen V.A., Xu J., Maree A.F.M. et al. (2007). Auxin transport is sufficient to generate a maximum and gradient guiding root growth. Nature 449 , 1008-1013. Grimplet J., Wheatley M.D., Jouira H.B. et al. (2009). Proteomic and selected metabolite analysis of grape berry tissues under well-watered and water-deficit stress conditions. Proteomics 9, 2503-2528. Groppa M.D. & Benavides M.P. (2008). Polyamines and abiotic stress: recent advances. Amino Acids 34 , 35-45. Guillon J.-M. (1905). Étude générale de la vigne: historique, les vignobles et les crus, anatomie et physiologie, sol et climat . Masson et Cie (Éditeurs), Paris, 451 p. Guo P., Baum M., Grando S. et al. (2009). Differentially expressed genes between drought- tolerant and drought-sensitive barley genotypes in response to drought stress during the reproductive stage. J. Exp. Bot. 60 , 3531-3544. Hacke U.G., Sperry J.S., Pockman W.T. et al. (2001). Trends in wood density and structure are linked to prevention of xylem implosion by negative pressure. Oecologia 126 , 457-461. Hacke U.G., Sperry J.S., Wheeler J.K. & Castro L. (2006). Scaling of angiosperm xylem structure with safety and efficiency. Tree Physiol. 26 , 689-701. Hagrey S.A.a. (2007). Geophysical imaging of root-zone, trunk, and moisture heterogeneity. J. Exp. Bot. 58 , 839-854. Hajdu E. (1998). Climate tolerance in the background of reliable vine yield. Acta Hortic. 473 , 83-91.

- 145 - References

Hall A. & Jones G.V. (2009). Effect of potential atmospheric warming on temperature-based indices describing Australian winegrape growing conditions. Aust. J. Grape Wine Res. 15 , 97-119. Hamberg M. (2011). Stereochemistry of hydrogen removal during oxygenation of linoleic acid by singlet oxygen and synthesis of 11( S)-deuterium-labeled linoleic acid. Lipids 46 , 201-206. Hamdy A., Ragab R. & Scarascia-Mugnozza E. (2003). Coping with water scarcity: water saving and increasing water productivity. Irrig. and Drain. 52 , 3-20. Hammer G., Cooper M., Tardieu F. et al. (2006). Models for navigating biological complexity in breeding improved crop plants. Trends Plant Sci. 11 , 587-593. Hardie W.J. & Martin S.R. (2000). Shoot growth on de-fruited grapevines: a physiological indicator for irrigation scheduling. Aust. J. Grape Wine Res. 6, 52-58. Harmon F.N. & Snyder E. (1934). Grape root distribution studies. Proc. Amer. Soc. Hort. Sci. 32 , 370-373. Harper J.F., Breton G. & Harmon A. (2004). Decoding Ca 2+ signals through plant protein kinases. Annu. Rev. Plant Biol. 55 , 263-288. Hideg É., Nagy T., Oberschall A. et al. (2003). Detoxification function of aldose/aldehyde reductase during drought and ultraviolet-B (280–320 nm) stresses. Plant Cell Environ. 26 , 513-522. Hirasawa T., Takahashi H., Suge H. & Ishihara K. (1997). Water potential, turgor and cell wall properties in elongating tissues of the hydrotropically bending roots of pea (Pisum sativum L.). Plant Cell Environ. 20 , 381-386. Hirayama T. & Shinozaki K. (2007). Perception and transduction of abscisic acid signals: keys to the function of the versatile plant hormone ABA. Trends Plant Sci. 12 , 343- 351. Ho M.D., Rosas J.C., Brown K.M. & Lynch J.P. (2005). Root architectural tradeoffs for water and phosphorus acquisition. Funct. Plant Biol. 32 , 737-748. Hodge A., Berta G., Doussan C. et al. (2009). Plant root growth, architecture and function. Plant Soil 321 , 153-187. Holbrook N.M., Ahrens E.T., Burns M.J. & Zwieniecki M.A. (2001). In vivo observation of cavitation and embolism repair using magnetic resonance imaging. Plant Physiol. 126 , 27-31. Holbrook N.M., Shashidhar V.R., James R.A. & Munns R. (2002). Stomatal control in tomato with ABA-deficient roots: response of grafted plants to soil drying. J. Exp. Bot. 53 , 1503-1514. Holland T. & Smit B. (2010). Climate change and the wine industry: current research themes and new directions. J. Wine Res. 21 , 125-136. Holm S. (1979). A simple sequentially rejective multiple test procedure. Scand. J. Statist. 6, 65-70. Holzapfel B.P. & Treeby M.T. (2007). Effects of timing and rate of N supply on leaf nitrogen status, grape yield and juice composition from Shiraz grapevines grafted to one of three different rootstocks. Aust. J. Grape Wine Res. 13 , 14-22. Hong-Bo S., Li-Ye C. & Ming-An S. (2008). Calcium as a versatile plant signal transducer under soil water stress. BioEssays 30 , 634-641. Hsiao T.C. (1973). Plant responses to water stress. Annu. Rev. Plant Physiol. 24 , 519-570. Hsiao T.C. & Xu L.-K. (2000). Sensitivity of growth of roots versus leaves to water stress: biophysical analysis and relation to water transport. J. Exp. Bot. 51 , 1595-1616. Huang B. (2000). Role of root morphological and physiological characteristics in drought resistance of plants In: Plant-Environment Interactions (ed R.E. Wilkinson), pp. 39-64. CRC Press, New York, USA.

- 146 - References

Huang B. & Eissenstat D.M. (2000). Root plasticity in exploiting water and nutrient heterogeneity. In: Plant-Environment Interactions (ed R.E. Wilkinson), pp. 111-132. CRC Press, New York, USA. Huang D., Wu W., Abrams S.R. & Cutler A.J. (2008). The relationship of drought-related gene expression in Arabidopsis thaliana to hormonal and environmental factors. J. Exp. Bot. 59 , 2991-3007. Huang Z. & Ough C.S. (1989). Effect of vineyard locations, varieties, and rootstocks on the juice amino acid composition of several cultivars. Am. J. Enol. Vitic. 40 , 135-139. Huglin P. (1958). Recherche sur les bourgeons de la vigne. Initiation florale et développement végétatif. Ann. Amélior. Plantes 8, 113-272. Hunt H.V., Lawes M.C., Bower M.A. et al. (2010). A banned variety was the mother of several major wine grapes. Biology Letters 6, 367-369. Hunt R., Causton D.R., Shipley B. & Askew A.P. (2002). A modern tool for classical plant growth analysis. Ann. Bot. 90 , 485-488. Hunter J.J. & Le Roux D.J. (1992). The effect of partial defoliation on development and distribution of roots of Vitis vinifera L. cv. Cabernet Sauvignon grafted onto rootstock 99 Richter. Am. J. Enol. Vitic. 43 , 71-78. Hunter J.J., Ruffner H.P., Volschenk C.G. & Le Roux D.J. (1995). Partial defoliation of Vitis vinifera L. cv. Cabernet Sauvignon/99 Richter: effect on root growth, canopy efficiency, grape composition, and wine quality. Am. J. Enol. Vitic. 46 , 306-314. Hunter J.J. (1998). Plant spacing implications for grafted grapevine. I. Soil characteristics, root growth, dry matter partitioning, dry matter composition and soil utilisation. S. Afr. J. Enol. Vitic. 19 , 25-34. Hunter S., Apweiler R., Attwood T.K. et al. (2009). InterPro: the integrative protein signature database. Nucleic Acids Res. 37 , D211-D215. Iacono F., Buccella A. & Peterlunger E. (1998). Water stress and rootstock influence on leaf gas exchange of grafted and ungrafted grapevines. Sci. Hortic. 75 , 27-39. Ingram P.A. & Malamy J.E. (2010). Root system architecture. Adv. Bot. Res. 55 , 75-117. Ingvarsson P.K. & Street N.R. (2011). Association genetics of complex traits in plants. New Phytol. 189 , 909-922. Ionescu E.M., Budan C., Calistru G.H. et al. (1978). Particularités du développement de l'appareil radiculaire de la vigne induites par le spécifique écologique du sol . Paper presented at the 1 st International Symposium on Ecology of Grapevine, Constanta, Romania. pp. 209-218. Ionescu E.M. (1978). Quelques aspects relatifs à l'influence de certains facteurs pédoécologiques sur l'appareil radiculaire de la vigne . Paper presented at the 1 st International Symposium on Ecology of Grapevine, Constanta, Romania. pp. 219-226. Iordachescu M. & Imai R. (2008). Trehalose biosynthesis in response to abiotic stresses. J. Integr. Plant Biol. 50 , 1223-1229. IPCC (2007). Climate change 2007: synthesis report. Contribution of working Groups I, II and III to the fourth assessment report of the Intergovernmental Panel on Climate Change . IPCC, Geneva, Switzerland, 104 p. Irigoyen J.J., Emerich D.W. & Sánchez-Díaz M. (1992). Phosphoenolpynivate carboxylase, malate and alcohol dehydrogenase activities in alfalfa ( Medicago sativa ) nodules under water stress. Physiol. Plant. 84 , 61-66. Iyer-Pascuzzi A.S., Symonova O., Mileyko Y. et al. (2010). Imaging and analysis platform for automatic phenotyping and trait ranking of plant root systems. Plant Physiol. 152 , 1148-1157. Jackson M. (1997). Hormones from roots as signals for the shoots of stressed plants. Trends Plant Sci. 2, 22-28.

- 147 - References

Jackson R.B., Sperry J.S. & Dawson T.E. (2000). Root water uptake and transport: using physiological processes in global predictions. Trends Plant Sci. 5, 482-488. Jaffe M.J., Takahashi H. & Biro R.L. (1985). A pea mutant for the study of hydrotropism in roots. Science 230 , 445-447. Jaillon O., Aury J.-M., Noel B. et al. (2007). The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla. Nature 449 , 463-467. Jain E., Bairoch A., Duvaud S. et al. (2009). Infrastructure for the life sciences: design and implementation of the UniProt website. BMC Bioinformatics 10 , 136. Jansen R., Kaittanis C., Saski C. et al. (2006). Phylogenetic analyses of Vitis (Vitaceae) based on complete chloroplast genome sequences: effects of taxon sampling and phylogenetic methods on resolving relationships among rosids. BMC Evol. Biol. 6, 32. Jansen S., Gortan E., Lens F. et al. (2011). Do quantitative vessel and pit characters account for ion-mediated changes in the hydraulic conductance of angiosperm xylem? New Phytol. 189 , 218-228. Jarvis P.G. (1976). The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Phil. Trans. R. Soc. B 273 , 593-610. Javaux M., Schroder T., Vanderborght J. & Vereecken H. (2008). Use of a three- dimensional detailed modeling approach for predicting root water uptake. Vadose Zone J. 7, 1079-1088. Javot H. & Maurel C. (2002). The role of aquaporins in root water uptake. Ann. Bot. 90 , 301-313. Javot H., Lauvergeat V., Santoni V. et al. (2003). Role of a single aquaporin isoform in root water uptake. Plant Cell 15 509–522. Jeong J.S., Kim Y.S., Baek K.H. et al. (2010). Root-specific expression of OsNAC10 improves drought tolerance and grain yield in rice under field drought conditions. Plant Physiol. 153 , 185-197. Jiang F. & Hartung W. (2008). Long-distance signalling of abscisic acid (ABA): the factors regulating the intensity of the ABA signal. J. Exp. Bot. 59 , 37-43. Jiménez S., Gogorcena Y., Hévin C. et al. (2007). Nitrogen nutrition influences some biochemical responses to iron deficiency in tolerant and sensitive genotypes of Vitis . Plant Soil 290 , 343-355. Johnson M.G., Tingey D.T., Phillips D.L. & Storm M.J. (2001). Advancing fine root research with minirhizotrons. Environ. Exp. Bot. 45 , 263-289. Jones G.V., White M.A., Cooper O.R. & Storchmann K. (2005). Climate change and global wine quality. Clim. Change 73 , 319-343. Jones G.V. (2006). Climate and terroir: impacts of climate variability and change on wine. In: Fine wine and terroir - The geoscience perspective (eds R.W. Macqueen & L.D. Meinert), pp. 203-216. Geoscience Canada Reprint Series Number 9, Geological Association of Canada, St. John's, Newfoundland, Canada. Jones G.V. & Webb L.B. (2010). Climate change, viticulture, and wine: challenges and opportunities. J. Wine Res. 21 , 103-106. Jones H. (2004). What is water use efficiency? In: Water use efficiency in plant biology (ed M.A. Bacon), pp. 27-41. Blackwell Publishing Ltd ; CRC Press, Oxford, UK ; Boca Raton, USA. Jones H.G. (1992a). Plants and microclimate: a quantitative approach to environmental plant physiology . (Second Edition ed.). Cambridge University Press, Cambridge, UK, 428 p. Jones H.G. (1992b). Stomata. In: Plants and microclimate: a quantitative approach to environmental plant physiology (ed H.G. Jones), pp. 131-162. Cambridge University Press, Cambridge, UK.

- 148 - References

Jones T.H., Cullis B.R., Clingeleffer P.R. & Rühl E.H. (2009). Effects of novel hybrid and traditional rootstocks on vigour and yield components of Shiraz grapevines. Aust. J. Grape Wine Res. 15 , 284-292. Jossier M., Bouly J.-P., Meimoun P. et al. (2009). SnRK1 (SNF1-related kinase 1) has a central role in sugar and ABA signalling in Arabidopsis thaliana . Plant J. 59 , 316-328. Jovanovic M., Lefebvre V., Laporte P. et al. (2007). How the environment regulates root architecture in dicots. Adv. Bot. Res. 46 , 35-74. Kaldenhoff R. & Fischer M. (2006). Aquaporins in plants. Acta Physiol. 187 , 169-176. Kaldenhoff R., Ribas-Carbo M., Flexas Sans J. et al. (2008). Aquaporins and plant water balance. Plant Cell Environ. 31 , 658-666. Kaneyasu T., Kobayashi A., Nakayama M. et al. (2007). Auxin response, but not its polar transport, plays a role in hydrotropism of Arabidopsis roots. J. Exp. Bot. 58 , 1143- 1150. Kantar M., Lucas S.J. & Budak H. (2011). Drought stress: molecular genetics and genomics approaches. Adv. Bot. Res. 57 , 445-493. Karagiannidis N., Velemis D. & Stavropoulos N. (1997). Root colonization and spore population by VA-mycorrhizal fungi in four grapevine rootstocks. Vitis 36 , 57-60. Karl T.R. & Trenberth K.E. (2003). Modern global climate change. Science 302 , 1719- 1723. Karlgren A., Gyllenstrand N., Källman T. et al. (2011). Evolution of the PEBP gene family in plants: functional diversification in seed plant evolution. Plant Physiol. 156 , 1967- 1977. Kato-Noguchi H. (2000). Osmotic stress increases alcohol dehydrogenase activity in maize seedlings. Biol. Plant. 43 , 621-624. Kauffmann A., Gentleman R. & Huber W. (2009). arrayQualityMetrics - a bioconductor package for quality assessment of microarray data. Bioinformatics 25 , 415-416. Kaufmann I., Schulze-Till T., Schneider H.U. et al. (2009). Functional repair of embolized vessels in maize roots after temporal drought stress, as demonstrated by magnetic resonance imaging. New Phytol. 184 , 245-256. Kawaguchi R., Girke T., Bray E.A. & Bailey-Serres J. (2004). Differential mRNA translation contributes to gene regulation under non-stress and dehydration stress conditions in Arabidopsis thaliana . Plant J. 38 , 823-839. Kaziev R. & Teimurov A. (2008). Characteristics of development of the dry farmed grape root system depending on growing conditions. Russ. Agr. Sci. 34 , 379-382. Keller M., Kummer M. & Vasconcelos M.C. (2001a). Reproductive growth of grapevines in response to nitrogen supply and rootstock. Aust. J. Grape Wine Res. 7, 12-18. Keller M., Kummer M. & Vasconcelos M.C. (2001b). Soil nitrogen utilisation for growth and gas exchange by grapevines in response to nitrogen supply and rootstock. Aust. J. Grape Wine Res. 7, 2-11. Keller M. (2010). The science of grapevines: anatomy and physiology . Academic Press, Elsevier Inc., San Diego, USA, 377 p. Kellow A.V., McDonald G., Corrie A.M. & Heeswijck R. (2002). In vitro assessment of grapevine resistance to two populations of phylloxera from Australian vineyards. Aust. J. Grape Wine Res. 8, 109-116. Kim H.K. & Lee S.J. (2010). Synchrotron X-ray imaging for nondestructive monitoring of sap flow dynamics through xylem vessel elements in rice leaves. New Phytol. 188 , 1085-1098. Kim T.-H., Böhmer M., Hu H. et al. (2010). Guard cell signal transduction network: 2+ advances in understanding abscisic acid, CO 2, and Ca signaling. Annu. Rev. Plant Biol. 61 , 561-591.

- 149 - References

Kirchhof G., Blackwell J. & Smart R. (1991). Growth of vineyard roots into segmentally ameliorated acidic subsoils. Plant Soil 134 , 121-126. Kobayashi A., Takahashi A., Kakimoto Y. et al. (2007). A gene essential for hydrotropism in roots. Proc. Natl. Acad. Sci. 104 , 4724-4729. Kobayashi K. & Salam M.U. (2000). Comparing simulated and measured values using mean squared deviation and its components. Agron. J. 92 , 345-352. Kobayashi S., Goto-Yamamoto N. & Hirochika H. (2004a). Retrotransposon-induced mutations in grape skin color. Science 304 , 982. Kobayashi Y., Yamamoto S., Minami H. et al. (2004b). Differential activation of the rice sucrose nonfermenting1-related protein kinase2 family by hyperosmotic stress and abscisic acid. Plant Cell 16 , 1163-1177. Kocsis L., Lehoczky E., Bakonyi L. et al. (1998). New lime and drought tolerant grape rootstock variety. Acta Hortic. 473 , 75-82. Kocsis L., Varga Z. & Pernesz G.Y. (2009). Introduction of a lime and drought tolerant rootstock variety. Acta Hortic. 827 , 465-470. Kodur S., Tisdall J.M., Tang C. & Walker R.R. (2010). Accumulation of potassium in grapevine rootstocks ( Vitis ) as affected by dry matter partitioning, root traits and transpiration. Aust. J. Grape Wine Res. 16 , 273-282. Koltai H. (2011). Strigolactones are regulators of root development. New Phytol. 190 , 545- 549. Koundouras S., Tsialtas I.T., Zioziou E. & Nikolaou N. (2008). Rootstock effects on the adaptive strategies of grapevine ( Vitis vinifera L. cv. Cabernet-Sauvignon) under contrasting water status: leaf physiological and structural responses. Agric. Ecosyst. Environ. 128 , 86-96. Koundouras S., Hatzidimitriou E., Karamolegkou M. et al. (2009). Irrigation and rootstock effects on the phenolic concentration and aroma potential of Vitis vinifera L. cv. Cabernet Sauvignon grapes. J. Agric. Food Chem. 57 , 7805-7813. Kreps J.A., Wu Y., Chang H.-S. et al. (2002). Transcriptome changes for Arabidopsis in response to salt, osmotic, and cold stress. Plant Physiol. 130 , 2129-2141. Kreuzwieser J. & Gessler A. (2010). Global climate change and tree nutrition: influence of water availability. Tree Physiol. 30 , 1221-1234. Lambers H. & Poorter H. (1992). Inherent variation in growth rate between higher plants: a search for physiological causes and ecological consequences. Adv. Ecol. Res. 23 , 187- 261. Langfelder P. & Horvath S. (2008). WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559. Larsson K.E., Nyström B. & Liljenberg C. (2006). A phosphatidylserine decarboxylase activity in root cells of oat ( Avena sativa ) is involved in altering membrane phospholipid composition during drought stress acclimation. Plant Physiol. Biochem. 44 , 211-219. Laskowski M., Grieneisen V.A., Hofhuis H. et al. (2008). Root system architecture from coupling cell shape to auxin transport. PLoS Biol. 6, e307. Lata C. & Prasad M. (2011). Role of DREBs in regulation of abiotic stress responses in plants. J. Exp. Bot. 62 , 4731-4748. Laucou V., Boursiquot J.M., Lacombe T. et al. (2008). Parentage of grapevine rootstock Fercal finally elucidated. Vitis 47 , 163-167. Le Bot J., Serra V., Fabre J. et al. (2010). DART: a software to analyse root system architecture and development from captured images. Plant Soil 326 , 261-273.

- 150 - References

Leakey A.D.B., Ainsworth E.A., Bernacchi C.J. et al. (2009). Elevated CO 2 effects on plant carbon, nitrogen, and water relations: six important lessons from FACE. J. Exp. Bot. 60 , 2859-2876. Lebon E., Pellegrino A., Louarn G. & Lecoeur J. (2006). Branch development controls leaf area dynamics in grapevine ( Vitis vinifera ) growing in drying soil. Ann. Bot. 98 , 175- 185. Lee K.H., Piao H.L., Kim H.-Y. et al. (2006). Activation of glucosidase via stress-induced polymerization rapidly increases active pools of abscisic acid. Cell 126 , 1109-1120. Lehnart R., Michel H., Löhnertz O. & Linsenmeier A. (2008). Root dynamics and pattern of Riesling on 5C rootstock using minirhizotrons. Vitis 47 , 197-200. Less H., Angelovici R., Tzin V. & Galili G. (2011). Coordinated gene networks regulating Arabidopsis plant metabolism in response to various stresses and nutritional cues. Plant Cell 23 , 1264-1271. Levadoux L., Boubals D. & Rives M. (1962). Le genre Vitis et ses espèces. Ann. Amélior. Plantes 12 , 19-44. Lewis A.M. & Boose E.R. (1995). Estimating volume flow rates through xylem conduits. Am. J. Bot. 82 , 1112-1116. Li B., Feng Z., Xie M. et al. (2011a). Modulation of the root-sourced ABA signal along its way to the shoot in Vitis riparia ×Vitis labrusca under water deficit J. Exp. Bot. 62 , 1731-1741. Li G., Xin H., Zheng X.F. et al. (2011b). Identification of the abscisic acid receptor VvPYL1 in Vitis vinifera . Plant Biol. In press, http://dx.doi.org/10.1111/j.1438- 8677.2011.00504.x . Li S.-W., Xue L., Xu S. et al. (2009). Mediators, genes and signaling in adventitious rooting. Bot. Rev. 75 , 230-247. Liang B.M., Sharp R.E. & Baskin T.I. (1997). Regulation of growth anisotropy in well- watered and water-stressed maize roots. I. Spatial distribution of longitudinal, radial, and tangential expansion rates. Plant Physiol. 115 , 101-111. Liavonchanka A. & Feussner I. (2006). Lipoxygenases: occurrence, functions and catalysis. J. Plant Physiol. 163 , 348-357. Linderman R.G. & Davis E.A. (2001). Comparative response of selected grapevine rootstocks and cultivars to inoculation with different mycorrhizal fungi. Am. J. Enol. Vitic. 52 , 8-11. Liu Q., Zhang C., Yang Y. & Hu X. (2010). Genome-wide and molecular evolution analyses of the phospholipase D gene family in Poplar and Grape. BMC Plant Biol. 10 , 117. Liu X., Yue Y., Li B. et al. (2007). A G protein-coupled receptor is a plasma membrane receptor for the plant hormone abscisic acid. Science 315 , 1712-1716. Lobell D.B., Cahill K.N. & Field C.B. (2007). Historical effects of temperature and precipitation on California crop yields. Clim. Change 81 , 187-203. Lobet G., Pagès L. & Draye X. (2011). A novel image-analysis toolbox enabling quantitative analysis of root system architecture. Plant Physiol. 157 , 29-39. Lorenz W.W., Alba R., Yu Y.-S. et al. (2011). Microarray analysis and scale-free gene networks identify candidate regulators in drought-stressed roots of loblolly pine ( P. taeda L.). BMC Genomics 12 , 264. Louarn G., Lecoeur J. & Lebon E. (2008). A three-dimensional statistical reconstruction model of grapevine ( Vitis vinifera ) simulating canopy structure variability within and between cultivar/training system pairs. Ann. Bot. 101 , 1167-1184.

- 151 - References

Loveys B.R. & Kriedemann P.E. (1974). Internal control of stomatal physiology and photosynthesis. I. Stomatal regulation and associated changes in endogenous levels of abscisic and phaseic acids. Aust. J. Plant Physiol. 1, 407-415. Loveys B.R. (1984a). Abscisic acid transport and metabolism in grapevine ( Vitis vinifera L.). New Phytol. 98 , 575-582. Loveys B.R. & Düring H. (1984). Diurnal changes in water relations and abscisic acid in field-grown Vitis vinifera cultivars. II. Abscisic acid changes under semi-arid conditions. New Phytol. 97 , 37-47. Loveys B.R. (1984b). Diurnal changes in water relations and abscisic acid in field-grown Vitis vinifera cultivars. III. The influence of xylem-derived abscisic acid on leaf gas exchange. New Phytol. 98 , 563-573. Lovisolo C. & Schubert A. (1998). Effects of water stress on vessel size and xylem hydraulic conductivity in Vitis vinifera L. J. Exp. Bot. 49 , 693-700. Lovisolo C., Hartung W. & Schubert A. (2002). Whole-plant hydraulic conductance and root-to-shoot flow of abscisic acid are independently affected by water stress in grapevines. Funct. Plant Biol. 29 , 1349-1356. Lovisolo C. & Schubert A. (2006). Mercury hinders recovery of shoot hydraulic conductivity during grapevine rehydration: evidence from a whole-plant approach. New Phytol. 172 , 469-478. Lovisolo C., Perrone I., Hartung W. & Schubert A. (2008a). An abscisic acid-related reduced transpiration promotes gradual embolism repair when grapevines are rehydrated after drought. New Phytol. 180 , 642-651. Lovisolo C., Tramontini S., Flexas J. & Schubert A. (2008b). Mercurial inhibition of root hydraulic conductance in Vitis spp. rootstocks under water stress. Environ. Exp. Bot. 63 , 178-182. Lovisolo C., Perrone I., Carra A. et al. (2010). Drought-induced changes in development and function of grapevine ( Vitis spp.) organs and in their hydraulic and non-hydraulic interactions at the whole-plant level: a physiological and molecular update. Funct. Plant Biol. 37 , 98-116. Lowe K. & Walker M. (2006). Genetic linkage map of the interspecific grape rootstock cross Ramsey ( Vitis champinii ) × Riparia Gloire ( Vitis riparia ). Theor. Appl. Genet. 112 , 1582-1592. Luan S. (2003). Protein phosphatases in plants. Annu. Rev. Plant Biol. 54 , 63-92. Luan S. (2009). The CBL–CIPK network in plant calcium signaling. Trends Plant Sci. 14 , 37-42. Lukac M., Calfapietra C., Lagomarsino A. & Loreto F. (2010). Global climate change and tree nutrition: effects of elevated CO 2 and temperature. Tree Physiol. 30 , 1209-1220. Luster J., Göttlein A., Nowack B. & Sarret G. (2009). Sampling, defining, characterising and modeling the rhizosphere-the soil science tool box. Plant Soil 321 , 457-482. Luterbacher J., Liniger M.A., Menzel A. et al. (2007). Exceptional European warmth of autumn 2006 and winter 2007: Historical context, the underlying dynamics, and its phenological impacts. Geophy. Res. Lett. 34 , L12704. Ma Y., Szostkiewicz I., Korte A. et al. (2009). Regulators of PP2C phosphatase activity function as abscisic acid sensors. Science 324 , 1064-1068. MacGregor D.R., Deak K.I., Ingram P.A. & Malamy J.E. (2008). Root system architecture in Arabidopsis grown in culture is regulated by sucrose uptake in the aerial tissues. Plant Cell 20 , 2643-2660. Majewski J. & Pastinen T. (2011). The study of eQTL variations by RNA-seq: from SNPs to phenotypes. Trends Genet. 27 , 72-79.

- 152 - References

Malamy J.E. (2005). Intrinsic and environmental response pathways that regulate root system architecture. Plant Cell Environ. 28 , 67-77. Malheiro A.C., Santos J.A., Fraga H. & Pinto J.G. (2010). Climate change scenarios applied to viticultural zoning in Europe. Clim. Res. 43 , 163-177. Mano J.i., Torii Y., Hayashi S.-i. et al. (2002). The NADPH:quinone oxidoreductase P1- #- crystallin in Arabidopsis catalyzes the ,!-hydrogenation of 2-alkenals: detoxication of the lipid peroxide-derived reactive aldehydes. Plant Cell Physiol. 43 , 1445-1455. Mano J.i., Belles-Boix E., Babiychuk E. et al. (2005). Protection against photooxidative injury of tobacco leaves by 2-alkenal reductase. Detoxication of lipid peroxide-derived reactive carbonyls. Plant Physiol. 139 , 1773-1783. Manohar M., Shigaki T. & Hirschi K.D. (2011). Plant cation/H + exchangers (CAXs): biological functions and genetic manipulations. Plant Biol. 13 , 561-569. Mapfumo E., Aspinall D., Hancock T. & Sedgley M. (1993). Xylem development in relation to water uptake by roots of grapevine ( Vitis vinifera L.). New Phytol. 125 , 93- 99. Mapfumo E. & Aspinall D. (1994). Anatomical changes of grapevine ( Vitis vinifera L. cv. Shiraz) roots related to radial resistance to water movement. Aust. J. Plant Physiol. 21 , 437-447. Mapfumo E., Aspinall D. & Hancock T.W. (1994). Growth and development of roots of grapevine ( Vitis vinifera L.) in relation to water uptake from soil. Ann. Bot. 74 , 75-85. Maracchi G., Sirotenko O. & Bindi M. (2005). Impacts of present and future climate variability on agriculture and forestry in the temperate regions: Europe. Clim. Change 70 , 117-135. Marguerit E., Boury C., Manicki A. et al. (2009). Genetic dissection of sex determinism, inflorescence morphology and downy mildew resistance in grapevine. Theor. Appl. Genet. 118 , 1261-1278. Marguerit E. (2010). Déterminisme génétique des réponses au déficit hydrique de la transpiration et de la croissance, induites par le porte-greffe, chez la vigne. Approche intégrée de génétique quantitative et d'écophysiologie . PhD thesis, University of Bordeaux 2, 228 p. Martin-Vertedor A.I. & Dodd I.C. (2011). Root-to-shoot signalling when soil moisture is heterogeneous: increasing the proportion of root biomass in drying soil inhibits leaf growth and increases leaf abscisic acid concentration. Plant Cell Environ. 34 , 1164- 1175. Martin T.A., Hinckley T.M., Meinzer F.C. & Sprugel D.G. (1999). Boundary layer conductance, leaf temperature and transpiration of Abies amabilis branches. Tree Physiol. 19 , 435-443. Martinez-Zapater J.M., Carmona M.J., Diaz-Riquelme J. et al. (2010). Grapevine genetics after the genome sequence: challenges and limitations. Aust. J. Grape Wine Res. 16 , 33-46. Martre P., Morillon R., Barrieu F. et al. (2002). Plasma membrane aquaporins play a significant role during recovery from water deficit. Plant Physiol. 130 , 2101-2110. Masle J., Gilmore S.R. & Farquhar G.D. (2005). The ERECTA gene regulates plant transpiration efficiency in Arabidopsis. Nature 436 , 866-870. Matthews M.A., Anderson M.M. & Schultz H.R. (1987). Phenologic and growth responses to early and late season water deficits in Cabernet franc. Vitis 26 , 147-160. Maul E., Töpfer R., Eibach R. et al. (2011). Vitis International Variety Catalogue (VIVC) . Julius Kühn-Institut - Federal Research Centre for Cultivated Plants, Quedlinburg (Germany). Institute for Grapevine Breeding - Geilweilerhof, Siebeldingen (Germany). http://www.vivc.de/index.php , accessed May 2011.

- 153 - References

Maurel C., Verdoucq L., Luu D.-T. & Santoni V. (2008). Plant aquaporins: membrane channels with multiple integrated functions. Annu. Rev. Plant Biol. 59 , 595-624. Maurel C., Santoni V., Luu D.-T. et al. (2009). The cellular dynamics of plant aquaporin expression and functions. Curr. Opin. Plant Biol. 12 , 690-698. May P. (1994). Using grapevine rootstocks: the Australian perspective . Winetitles, Adelaide, 62 p. McCully M.E. (1999). Roots in soil: unearthing the complexities of roots and their rhizospheres. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50 , 695-718. McElrone A.J., Bichler J., Pockman W.T. et al. (2007). Aquaporin-mediated changes in hydraulic conductivity of deep tree roots accessed via caves. Plant Cell Environ. 30 , 1411-1421. McKenry M.V. (1984). Grape root phenology relative to control of parasitic nematodes. Am. J. Enol. Vitic. 35 , 206-211. Meier I.C. & Leuschner C. (2008). Genotypic variation and phenotypic plasticity in the drought response of fine roots of European beech. Tree Physiol. 28 , 297-309. Melcher K., Ng L.-M., Zhou X.E. et al. (2009). A gate-latch-lock mechanism for hormone signalling by abscisic acid receptors. Nature 462 , 602-608. Melcher K., Xu Y., Ng L.-M. et al. (2010). Identification and mechanism of ABA receptor antagonism. Nat. Struct. Mol. Biol. 17 , 1102-1108. Menzel A., Sparks T.H., Estrella N. et al. (2006). European phenological response to climate change matches the warming pattern. Global Change Biol. 12 , 1969-1976. Menzel A. & Sparks T. (2007). Temperature and plant development: phenology and seasonality. In: Plant Growth and Climate Change (eds J.I.L. Morison & M.D. Morecroft), pp. 70-95. Blackwell Publishing Ltd, Oxford, UK. Mica E., Piccolo V., Delledonne M. et al. (2010). Correction: High throughput approaches reveal splicing of primary microRNA transcripts and tissue specific expression of mature microRNAs in Vitis vinifera. BMC Genomics 11 , 109. Micheletto S., Rodriguez-Uribe L., Hernandez R. et al. (2007). Comparative transcript profiling in roots of Phaseolus acutifolius and P. vulgaris under water deficit stress. Plant Sci. 173 , 510-520. Miller G., Suzuki N., Ciftci-Yilmaz S. & Mittler R. (2010). Reactive oxygen species homeostasis and signalling during drought and salinity stresses. Plant Cell Environ. 33 , 453-467. Milly P.C.D., Dunne K.A. & Vecchia A.V. (2005). Global pattern of trends in streamflow and water availability in a changing climate. Nature 438 , 347-350. Mira de Orduña R. (2010). Climate change associated effects on grape and wine quality and production. Food Res. Int. 43 , 1844-1855. Mittler R. & Blumwald E. (2010). Genetic engineering for modern agriculture: challenges and perspectives. Annu. Rev. Plant Biol. 61 , 443-462. Miyazawa Y., Sakashita T., Funayama T. et al. (2008). Effects of locally targeted heavy- ion and laser microbeam on root hydrotropism in Arabidopsis thaliana . J. Radiat. Res. 49 , 373-379. Miyazawa Y., Takahashi A., Kobayashi A. et al. (2009a). GNOM-mediated vesicular trafficking plays an essential role in hydrotropism of Arabidopsis roots. Plant Physiol. 149 , 835-840. Miyazawa Y., Ito Y., Moriwaki T. et al. (2009b). A molecular mechanism unique to hydrotropism in roots. Plant Sci. 177 , 297-301. Miyazawa Y., Yamazaki T., Moriwaki T. & Takahashi H. (2011). Root tropism: its mechanism and possible functions in drought avoidance. Adv. Bot. Res. 57 , 349-375.

- 154 - References

Miyazono K.-i., Miyakawa T., Sawano Y. et al. (2009). Structural basis of abscisic acid signalling. Nature 462 , 609-614. Mizoguchi M., Umezawa T., Nakashima K. et al. (2010). Two closely related subclass II SnRK2 protein kinases cooperatively regulate drought-inducible gene expression. Plant Cell Physiol. 51 , 842-847. Mizuno H., Kobayashi A., Fujii N. et al. (2002). Hydrotropic response and expression pattern of auxin-inducible gene, CS-IAA1 , in the primary roots of clinorotated cucumber seedlings. Plant Cell Physiol. 43 , 793-801. Mochida K. & Shinozaki K. (2010). Genomics and bioinformatics resources for crop improvement. Plant Cell Physiol. 51 , 497-523. Møller I.M., Jensen P.E. & Hansson A. (2007). Oxidative modifications to cellular components in plants. Annu. Rev. Plant Biol. 58 , 459-481. Monod H., Naud C. & Makowski D. (2006). Uncertainty and sensitivity analysis for crop models. In: Working with dynamic crop models. Evaluation, analysis, parameterization, and applications (eds D. Wallach, D. Makowski & J.W. Jones), pp. 55-99. Elsevier, Oxford, UK. Monshausen G.B., Bibikova T.N., Weisenseel M.H. & Gilroy S. (2009). Ca 2+ regulates reactive oxygen species production and pH during mechanosensing in Arabidopsis roots. Plant Cell 21 , 2341-2356. Monshausen G.B. & Gilroy S. (2009). The exploring root - root growth responses to local environmental conditions. Curr. Opin. Plant Biol. 12 , 766-772. Montero F.J., de Juan J.A., Cuesta A. & Brasa A. (2000). Nondestructive methods to estimate leaf area in Vitis vinifera L. HortScience 35 , 696-698. Moore J.P., Vicré-Gibouin M., Farrant J.M. & Driouich A. (2008). Adaptations of higher plant cell walls to water loss: drought vs desiccation. Physiol. Plant. 134 , 237-245. Moran J.F., Becana M., Iturbe-Ormaetxe I. et al. (1994). Drought induces oxidative stress in pea plants. Planta 194 , 346-352. Morano L. & Kliewer W.M. (1994). Root distribution of three grapevine rootstocks grafted to Cabernet Sauvignon grown on a very gravelly clay loam soil in Oakville, California. Am. J. Enol. Vitic. 45 , 345-348. Morano L.D. & Walker M.A. (1995). Soils and plant communities associated with three Vitis species. Am. Midl. Nat. 134 , 254-263. Morey J., Ryan J. & Van Dolah F. (2006). Microarray validation: factors influencing correlation between oligonucleotide microarrays and real-time PCR. Biol. Proced. Online 8, 175-193. Morgan J.M. (1984). Osmoregulation and water stress in higher plants. Annu. Rev. Plant Physiol. 35 , 299-319. Moriwaki T., Miyazawa Y. & Takahashi H. (2010). Transcriptome analysis of gene expression during the hydrotropic response in Arabidopsis seedlings. Environ. Exp. Bot. 69 , 148-157. Morlat R. & Jacquet A. (1993). The soil effects on the grapevine root system in several vineyards of the Loire valley (France). Vitis 32 , 35-42. Morlat R. & Jacquet A. (2003). Grapevine root system and soil characteristics in a vineyard maintained long-term with or without interrow sward. Am. J. Enol. Vitic. 54 , 1-7. Morlat R. (2008). Long-term additions of organic amendments in a Loire valley vineyard on a calcareous sandy soil. II. Effects on root system, growth, grape yield, and foliar nutrient status of a Cabernet franc vine. Am. J. Enol. Vitic. 59 , 364-374. Mosblech A., Feussner I. & Heilmann I. (2009). Oxylipins: structurally diverse metabolites from fatty acid oxidation. Plant Physiol. Biochem. 47 , 511-517.

- 155 - References

Moser C., Segala C., Fontana P. et al. (2005). Comparative analysis of expressed sequence tags from different organs of Vitis vinifera L. Funct. Integr. Genomics 5, 208-217. Motulsky H. & Christopoulos A. (2003). Fitting models to biological data using linear and nonlinear regression. A practical guide to curve fitting . GraphPad Software Inc., San Diego CA, USA, 351 p. Mouchel C.F., Osmont K.S. & Hardtke C.S. (2006). BRX mediates feedback between brassinosteroid levels and auxin signalling in root growth. Nature 443 , 458-461. Munnik T. & Meijer H.J.G. (2001). Osmotic stress activates distinct lipid and MAPK signalling pathways in plants. FEBS Lett. 498 , 172-178. Munnik T. & Vermeer J.E.M. (2010). Osmotic stress-induced phosphoinositide and inositol phosphate signalling in plants. Plant Cell Environ. 33 , 655-669. Müssig C., Shin G.-H. & Altmann T. (2003). Brassinosteroids promote root growth in Arabidopsis. Plant Physiol. 133 , 1261-1271. Myles S., Boyko A.R., Owens C.L. et al. (2011). Genetic structure and domestication history of the grape. Proc. Natl. Acad. Sci. 108 , 3530-3535. Nagarajah S. (1987). Effects of soil texture on the rooting patterns of Thompson seedless vines on own roots and on Ramsey rootstock in irrigated vineyards. Am. J. Enol. Vitic. 38 , 54-59. Nambara E. & Marion-Poll A. (2005). Abscissic acid biosynthesis and catabolism. Annu. Rev. Plant Biol. 56 , 165-185. Nanjo T., Fujita M., Seki M. et al. (2003). Toxicity of free proline revealed in an Arabidopsis T-DNA-tagged mutant deficient in proline dehydrogenase. Plant Cell Physiol. 44 , 541-548. Nappi P., Jodice R., Luzzati A. & Corino L. (1985). Grapevine root system and VA mycorrhizae in some soils of Piedmont (Italy). Plant Soil 85 , 205-210. Nardini A., Salleo S. & Jansen S. (2011). More than just a vulnerable pipeline: xylem physiology in the light of ion-mediated regulation of plant water transport. J. Exp. Bot. 62 , 4701-4718. Nelson D.E., Repetti P.P., Adams T.R. et al. (2007). Plant nuclear factor Y (NF-Y) B subunits confer drought tolerance and lead to improved corn yields on water-limited acres. Proc. Natl. Acad. Sci. 104 , 16450-16455. Neumann G., George T. & Plassard C. (2009). Strategies and methods for studying the rhizosphere-the plant science toolbox. Plant Soil 321 , 431-456. Neumann P.M. (2008). Coping mechanisms for crop plants in drought-prone environments. Ann. Bot. 101 , 901-907. Neumann P.M. (2011). Recent advances in understanding the regulation of whole-plant growth inhibition by salinity, drought and colloid stress. Adv. Bot. Res. 57 , 33-48. Nibau C., Gibbs D.J. & Coates J.C. (2008). Branching out in new directions: the control of root architecture by lateral root formation. New Phytol. 179 , 595-614. Nicol J.M., Stirling G.R., Rose B.J. et al. (1999). Impact of nematodes on grapevine growth and productivity: current knowledge and future directions, with special reference to Australian viticulture. Aust. J. Grape Wine Res. 5, 109-127. Nicotra A.B. & Davidson A. (2010). Adaptive phenotypic plasticity and plant water use. Funct. Plant Biol. 37 , 117-127 Nicotra A.B., Atkin O.K., Bonser S.P. et al. (2010). Plant phenotypic plasticity in a changing climate. Trends Plant Sci. 15 , 684-692. Nikolaou N., Angelopoulos K. & Karagiannidis N. (2003a). Effects of drought stress on mycorrhizal and non-mycorrhizal Cabernet-Sauvignon grapevine, grafted onto various rootstocks. Exp. Agric. 39 , 241-252.

- 156 - References

Nikolaou N.A., Koukourikou M., Angelopoulos K. & Karagiannidis N. (2003b). Cytokinin content and water relations of 'Cabernet Sauvignon' grapevine exposed to drought stress. J. Hortic. Sci. Biotechnol. 78 , 113-118. Nilson S.E. & Assmann S.M. (2010). The *-subunit of the Arabidopsis heterotrimeric G protein, GPA1, is a regulator of transpiration efficiency. Plant Physiol. 152 , 2067- 2077. Ning J., Li X., Hicks L.M. & Xiong L. (2010). A raf-like MAPKKK gene DSM1 mediates drought resistance through reactive oxygen species scavenging in rice. Plant Physiol. 152 , 876-890. Nishimura N., Hitomi K., Arvai A.S. et al. (2009). Structural mechanism of abscisic acid binding and signaling by dimeric PYR1. Science 326 , 1373-1379. Nishizawa A., Yabuta Y. & Shigeoka S. (2008). Galactinol and raffinose constitute a novel function to protect plants from oxidative damage. Plant Physiol. 147 , 1251-1263. Norberg P. & Liljenberg C. (1991). Lipids of plasma membranes prepared from oat root cells: effects of induced water-deficit tolerance. Plant Physiol. 96 , 1136-1141. North G.B. & Nobel P.S. (1997a). Drought-induced changes in soil contact and hydraulic conductivity for roots of Opuntia ficus-indica with and without rhizosheaths. Plant Soil 191 , 249-258. North G.B. & Nobel P.S. (1997b). Root-soil contact for the desert succulent Agave deserti in wet and drying soil. New Phytol. 135 , 21-29. Ober E.S. & Sharp R.E. (1994). Proline accumulation in maize ( Zea mays L.) primary roots at low water potentials. I. Requirement for increased levels of abscisic acid. Plant Physiol. 105 , 981-987. Ober E.S. & Sharp R.E. (2003). Electrophysiological responses of maize roots to low water potentials: relationship to growth and ABA accumulation. J. Exp. Bot. 54 , 813-824. Ober E.S. & Sharp R.E. (2007). Regulation of root growth responses to water deficit. In: Advances in molecular breeding toward drought and salt tolerant crops (eds M.A. Jenks, P.M. Hasegawa & S.M. Jain), pp. 33-53. Springer, Dordrecht, The Netherlands. Oberschall A., Deák M., Török K. et al. (2000). A novel aldose/aldehyde reductase protects transgenic plants against lipid peroxidation under chemical and drought stresses. Plant J. 24 , 437-446. OIV (2007). World vitivinicultural statistics 2007: structure of the world vitivinicultural industry in 2007 . Organisation Internationale de la Vigne et du Vin (OIV), Paris, France, 68 p. Oki T. & Kanae S. (2006). Global hydrological cycles and world water resources. Science 313 , 1068-1072. Olesen J.E. & Bindi M. (2002). Consequences of climate change for European agricultural productivity, land use and policy. Eur. J. Agron. 16 , 239-262. Olesinski J., Lavigne M.B. & Krasowski M.J. (2011). Effects of soil moisture manipulations on fine root dynamics in a mature balsam fir ( Abies balsamea L. Mill.) forest. Tree Physiol. 31 , 339-348. Ollat N., Tandonnet J.P., Bordenave L. et al. (2003). Vigour conferred by rootstock: hypotheses and direction for research. Bull. OIV 76 , 581-595. Ollat N., Fernandez L., Romieu C. et al. (2011). Multidisciplinary research to select new cultivars adapted to climate changes . Paper presented at the Proceedings of the 17 th International Symposium GiESCO, 29th August - 2nd September 2011, Asti - Alba (CN), Italy. Olmo H.P. (1986). The potential role of ( vinifera × rotundifolia ) hybrids in grape variety improvement. Experientia 42 , 921-926.

- 157 - References

Ortega-Martinez O., Pernas M., Carol R.J. & Dolan L. (2007). Ethylene modulates stem cell division in the Arabidopsis thaliana root. Science 317 , 507-510. Osakabe Y., Kajita S. & Osakabe K. (2011). Genetic engineering of woody plants: current and future targets in a stressful environment. Physiol. Plant. 142 , 105-117. Oslobeanu M. (1978). Some aspects of the interaction within scion-rootstock biosystem with grapevine . Paper presented at the 1 st International Symposium on Ecology of Grapevine, Constanta, Romania. pp. 227-236. Osmont K.S., Sibout R. & Hardtke C.S. (2007). Hidden branches: developments in root system architecture. Annu. Rev. Plant Biol. 58 , 93-113. Ozsolak F. & Milos P.M. (2011). RNA sequencing: advances, challenges and opportunities. Nat. Rev. Genet. 12 , 87-98. Padgett-Johnson M., Williams L.E. & Walker M.A. (2003). Vine water relations, gas exchange, and vegetative growth of seventeen Vitis species grown under irrigated and nonirrigated conditions in California. J. Am. Soc. Hortic. Sci. 128 , 269-276. Padilla F., Miranda J., Jorquera M. & Pugnaire F. (2009). Variability in amount and frequency of water supply affects roots but not growth of arid shrubs. Plant Ecol. 204 , 261-270. Pandey S., Nelson D.C. & Assmann S.M. (2009). Two novel GPCR-type G proteins are abscisic acid receptors in Arabidopsis . Cell 136 , 136-148. Paranychianakis N.V., Aggelides S. & Angelakis A.N. (2004). Influence of rootstock, irrigation level and recycled water on growth and yield of Soultanina grapevines. Agric. Water Manag. 69 , 13-27. Parent B., Hachez C., Redondo E. et al. (2009). Drought and abscisic acid effects on aquaporin content translate into changes in hydraulic conductivity and leaf growth rate: a trans-scale approach. Plant Physiol. 149 , 2000-2012. Park S.-Y., Fung P., Nishimura N. et al. (2009). Abscisic acid inhibits type 2C protein phosphatases via the PYR/PYL family of START proteins. Science 324 , 1068-1071. Parmesan C. & Yohe G. (2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature 421 , 37-42. Paschalidis K.A., Moschou P.N., Aziz A. et al. (2009). Polyamines in grapevine: an update. In: Grapevine Molecular Physiology and Biotechnology (ed K.A. Roubelakis- Angelakis), pp. 207-228. Springer, Netherlands. Paul M.J., Primavesi L.F., Jhurreea D. & Zhang Y. (2008). Trehalose metabolism and signaling. Annu. Rev. Plant Biol. 59 , 417-441. Pavlidis P., Li Q. & Noble W.S. (2003). The effect of replication on gene expression microarray experiments. Bioinformatics 19 , 1620-1627. Peleg Z. & Blumwald E. (2011). Hormone balance and abiotic stress tolerance in crop plants. Curr. Opin. Plant Biol. 14 , 290-295. Pellegrino A., Lebon E., Simonneau T. & Wery J. (2005). Towards a simple indicator of water stress in grapevine ( Vitis vinifera L.) based on the differential sensitivities of vegetative growth components. Aust. J. Grape Wine Res. 11 , 306-315. Pelsy F. (2010). Molecular and cellular mechanisms of diversity within grapevine varieties. Heredity 104 , 331-340. Pennisi E. (2008). The blue revolution, drop by drop, gene by gene. Science 320 , 171-173. Péret B., De Rybel B., Casimiro I. et al. (2009). Arabidopsis lateral root development: an emerging story. Trends Plant Sci. 14 , 399-408. Perfus-Barbeoch L., Jones A.M. & Assmann S.M. (2004). Plant heterotrimeric G protein function: insights from Arabidopsis and rice mutants. Curr. Opin. Plant Biol. 7, 719- 731.

- 158 - References

Péros J.-P., Berger G., Portemont A. et al. (2011). Genetic variation and biogeography of the disjunct Vitis subg. Vitis (Vitaceae). J. Biogeogr. 38 , 471-486. Perry R., Lyda S. & Bowen H. (1983). Root distribution of four Vitis cultivars. Plant Soil 71 , 63-74. Peters J.S. & Frenkel C. (2004). Relationship between alcohol dehydrogenase activity and low-temperature in two maize genotypes, Silverado F1 and Adh1 –Adh2 – doubly null. Plant Physiol. Biochem. 42 , 841-846. Peterson F.C., Burgie E.S., Park S.-Y. et al. (2010). Structural basis for selective activation of ABA receptors. Nat. Struct. Mol. Biol. 17 , 1109-1113. Petrie P.R. & Sadras V.O. (2008). Advancement of grapevine maturity in Australia between 1993 and 2006: putative causes, magnitude of trends and viticultural consequences. Aust. J. Grape Wine Res. 14 , 33-45. Piao S., Ciais P., Huang Y. et al. (2010). The impacts of climate change on water resources and agriculture in China. Nature 467 , 43-51. Pieri P. (2010). Climate change and viticulture: the main impacts. In: Climate change, agriculture and forests in France: simulations of the impacts on the main species. The Green Book of the CLIMATOR project (2007-2010) (eds N. Brisson & F. Levrault), pp. 213-224. ADEME, Angers, France. Pierret A., Doussan C., Capowiez Y. et al. (2007). Root functional architecture: a framework for modeling the interplay between roots and soil. Vadose Zone J. 6, 269- 281. Pina A. & Errea P. (2005). A review of new advances in mechanism of graft compatibility- incompatibility. Sci. Hortic. 106 , 1-11. Pinheiro C. & Chaves M.M. (2011). Photosynthesis and drought: can we make metabolic connections from available data? J. Exp. Bot. 62 , 869-882. Pittermann J. (2010). The evolution of water transport in plants: an integrated approach. Geobiol. 8, 112-139. Podolyan A., White J., Jordan B. & Winefield C. (2010). Identification of the lipoxygenase gene family from Vitis vinifera and biochemical characterisation of two 13- lipoxygenases expressed in grape berries of Sauvignon Blanc. Funct. Plant Biol. 37 , 767-784. Polesani M., Bortesi L., Ferrarini A. et al. (2010). General and species-specific transcriptional responses to downy mildew infection in a susceptible ( Vitis vinifera ) and a resistant ( V. riparia ) grapevine species. BMC Genomics 11 , 117. Ponce G., Rasgado F.A. & Cassab G.I. (2008). Roles of amyloplasts and water deficit in root tropisms. Plant Cell Environ. 31 , 205-217. Pongrácz D.P. & Beukman E.F. (1970). Comparative root anatomy of Vitis rootstocks. Agroplantae 2, 83-94. Pongrácz D.P. (1983). Rootstocks for grape-vines . David Philip Publisher (Pty), Cape town, South Africa, 150 p. Poorter H. & Nagel O. (2000). The role of biomass allocation in the growth response of plants to different levels of light, CO 2, nutrients and water : a quantitative review. Aust. J. Plant Physiol. 27 , 595-607. Poorter L., McDonald I., Alarcón A. et al. (2010). The importance of wood traits and hydraulic conductance for the performance and life history strategies of 42 rainforest tree species. New Phytol. 185 , 481-492. Poroyko V., Spollen W.G., Hejlek L.G. et al. (2007). Comparing regional transcript profiles from maize primary roots under well-watered and low water potential conditions. J. Exp. Bot. 58 , 279-289.

- 159 - References

Porter N., Caldwell S. & Mills K. (1995). Mechanisms of free radical oxidation of unsaturated lipids. Lipids 30 , 277-290. Postaire O., Tournaire-Roux C., Grondin A. et al. (2010). A PIP1 aquaporin contributes to hydrostatic pressure-induced water transport in both the root and rosette of Arabidopsis. Plant Physiol. 152 , 1418-1430. Pou A., Flexas J., del Mar Alsina M. et al. (2008). Adjustments of water use efficiency by stomatal regulation during drought and recovery in the drought-adapted Vitis hybrid Richter-110 ( V. berlandieri × V. rupestris ). Physiol. Plant. 134 , 313-323. Pouget R. (1990). Histoire de la lutte contre le phylloxéra de la vigne en France . INRA/OIV, Paris, 157 p. Pourtchev P. (1995). Détermination de certains indices agrobiologiques du système racinaire de la vigne. Riv. Vitic. Enol. 48 , 63-70. Pourtchev P. (2003). To the problem of depth of root system penetration of grapevine. Soil Science Agrochemistry and Ecology 38 , 47-52. Prasad M.E., Schofield A., Lyzenga W. et al. (2010). Arabidopsis RING E3 ligase XBAT32 regulates lateral root production through its role in ethylene biosynthesis. Plant Physiol. 153 , 1587-1596. Pratt C. (1974). Vegetative anatomy of cultivated grapes: a review. Am. J. Enol. Vitic. 25 , 131-150. Qi W., Guan X.-q., Li E.-m. et al. (2007). Effects of partial rootzone drying on the growth of Vitis vinifera cv. Malvasia grafted on different rootstocks. Agric. Sci. China 6, 567- 572. Qin F., Shinozaki K. & Yamaguchi-Shinozaki K. (2011). Achievements and challenges in understanding plant abiotic stress responses and tolerance. Plant Cell Physiol. 52 , 1569-1582. Quan R., Shang M., Zhang H. et al. (2004). Engineering of enhanced glycine betaine synthesis improves drought tolerance in maize. Plant Biotechnol. J. 2, 477-486. R Development Core Team (2010). R: A language and environment for statistical computing . R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051- 07-0, http://www.R-project.org/ , accessed May, 2011. Raghavendra A.S., Gonugunta V.K., Christmann A. & Grill E. (2010). ABA perception and signalling. Trends Plant Sci. 15 , 395-401. Raymond M.J. & Smirnoff N. (2002). Proline metabolism and transport in maize seedlings at low water potential. Ann. Bot. 89 , 813-823. Reiser V., Raitt D.C. & Saito H. (2003). Yeast osmosensor Sln1 and plant cytokinin receptor Cre1 respond to changes in turgor pressure. J. Cell Biol. 161 , 1035-1040. Renaud S. & de Lorgeril M. (1992). Wine, alcohol, platelets, and the French paradox for coronary heart disease. Lancet 339 , 1523-1526. Rengasamy P. (2006). World salinization with emphasis on Australia. J. Exp. Bot. 57 , 1017- 1023. Rensink W.A. & Buell C.R. (2005). Microarray expression profiling resources for plant genomics. Trends Plant Sci. 10 , 603-609. Reymond M., Muller B., Leonardi A. et al. (2003). Combining quantitative trait loci analysis and an ecophysiological model to analyze the genetic variability of the responses of maize leaf growth to temperature and water deficit. Plant Physiol. 131 , 664-675. Richards D. (1983). The grape root system. Hortic. Rev. 5, 127-168. Rieger M. (1995). Offsetting effects of reduced root hydraulic conductivity and osmotic adjustment following drought. Tree Physiol. 15 , 379-385.

- 160 - References

Roberts S.K. & Snowman B.N. (2000). The effects of ABA on channel-mediated K + transport across higher plant roots. J. Exp. Bot. 51 , 1585-1594. Roby G., Harbertson J.F., Adams D.A. & Matthews M.A. (2004). Berry size and vine water deficits as factors in winegrape composition: anthocyanins and tannins. Aust. J. Grape Wine Res. 10 , 100-107. Rodrigues M.L., Chaves M.M., Wendler R. et al. (1993). Osmotic adjustment in water stressed grapevine leaves in relation to carbon assimilation. Aust. J. Plant Physiol. 20 , 309-321. Rodrigues M.L., Santos T.P., Rodrigues A.P. et al. (2008). Hydraulic and chemical signalling in the regulation of stomatal conductance and plant water use in field grapevines growing under deficit irrigation. Funct. Plant Biol. 35 , 565-579. Rodríguez-Gamir J., Intrigliolo D.S., Primo-Millo E. & Forner-Giner M.A. (2010). Relationships between xylem anatomy, root hydraulic conductivity, leaf/root ratio and transpiration in citrus trees on different rootstocks. Physiol. Plant. 139 , 159-169. Rodriguez M.S.C., Petersen M. & Mundy J. (2010). Mitogen-activated protein kinase signaling in plants. Annu. Rev. Plant Biol. 61 , 621-649. Rogiers S.Y., Greer D.H., Hutton R.J. & Landsberg J.J. (2009). Does night-time transpiration contribute to anisohydric behaviour in a Vitis vinifera cultivar? J. Exp. Bot. 60 , 3751-3763. Rogiers S.Y., Holzapfel B.P. & Smith J.P. (2011). Sugar accumulation in roots of two grape varieties with contrasting response to water stress. Ann. Appl. Biol. 159 , 399-413. Rolland F., Baena-Gonzalez E. & Sheen J. (2006). Sugar sensing and signaling in plants: conserved and novel mechanisms. Annu. Rev. Plant Biol. 57 , 675-709. Romieu C., Kappel C. & Delrot S. (2011). Functional genomics: closed system approaches for transcriptome analysis. In: Genetics, genomics and breeding of grapes (eds A.-F. Adam-Blondon, J.M. Martínez-Zapater & C. Kole), pp. 270-298. Science Publishers, CRC Press, Taylor & Francis group, New York, USA. Root T.L., Price J.T., Hall K.R. et al. (2003). Fingerprints of global warming on wild animals and plants. Nature 421 , 57-60. Rorat T. (2006). Plant dehydrins - Tissue location, structure and function. Cell. Mol. Biol. Lett. 11 , 536-556. Roy S.J., Tucker E.J. & Tester M. (2011). Genetic analysis of abiotic stress tolerance in crops. Curr. Opin. Plant Biol. 14 , 232-239. Rühl E.H., Clingeleffer P.R., Nicholas P.R. et al. (1988). Effect of rootstocks on berry weight and pH, mineral content and organic acid concentrations of grape juice of some wine varieties. Aust. J. Exp. Agric. 28 , 119-125. Saab I.N., Sharp R.E., Pritchard J. & Voetberg G.S. (1990). Increased endogenous abscisic acid maintains primary root growth and inhibits shoot growth of maize seedlings at low water potentials. Plant Physiol. 93 , 1329-1336. Saab I.N., Sharp R.E. & Pritchard J. (1992). Effect of inhibition of abscisic acid accumulation on the spatial distribution of elongation in the primary root and mesocotyl of maize at low water potentials. Plant Physiol. 99 , 26-33. Sadok W., Naudin P., Boussuge B. et al. (2007). Leaf growth rate per unit thermal time follows QTL-dependent daily patterns in hundreds of maize lines under naturally fluctuating conditions. Plant Cell Environ. 30 , 135-146. Sadras V.O. & Petrie P.R. (2011). Climate shifts in south-eastern Australia: early maturity of Chardonnay, Shiraz and Cabernet Sauvignon is associated with early onset rather than faster ripening. Aust. J. Grape Wine Res. 17 , 199-205. Sakamoto A. & Murata N. (2002). The role of glycine betaine in the protection of plants from stress: clues from transgenic plants. Plant Cell Environ. 25 , 163-171.

- 161 - References

Sakuma Y., Maruyama K., Qin F. et al. (2006). Dual function of an Arabidopsis transcription factor DREB2A in water-stress-responsive and heat-stress-responsive gene expression. Proc. Natl. Acad. Sci. 103 , 18822-18827. Salazar Parra C., Aguirreolea J., Sánchez-Díaz M. et al. (2010). Effects of climate change scenarios on Tempranillo grapevine ( Vitis vinifera L.) ripening: response to a combination of elevated CO 2 and temperature, and moderate drought. Plant Soil 337 , 179-191. Salleo S., Lo Gullo M.A. & Oliveri F. (1985). Hydraulic parameters measured in 1-year-old twigs of some Mediterranean species with diffuse-porous wood: changes in hydraulic conductivity and their possible functional significance. J. Exp. Bot. 36 , 1-11. Salleo S., Nardini A., Pitt F. & Lo Gullo M.A. (2000). Xylem cavitation and hydraulic control of stomatal conductance in Laurel ( Laurus nobilis L.). Plant Cell Environ. 23 , 71-79. Sampaio T.L.B. (2007). Using rootstocks to manipulate vine physiological performance and mediate changes in fruit and wine composition . PhD Thesis, Oregon State University, USA, 257 p. Sampedro J. & Cosgrove D.J. (2005). The expansin superfamily. Genome Biol. 6, 242. Samson C. & Casteran P. (1971). Techniques de multiplication de la vigne. In: Sciences et techniques de la vigne. Tome 2: Culture, pathologie, défense sanitaire de la vigne (eds J. Ribéreau-Gayon & E. Peynaud), pp. 3-34. Dunod, Paris, France. Santana J.C., Heuertz M., Arranz C. et al. (2010). Genetic structure, origins, and relationships of grapevine cultivars from the Castilian plateau of Spain. Am. J. Enol. Vitic. 61 , 214-224. Santiago J., Dupeux F., Round A. et al. (2009). The abscisic acid receptor PYR1 in complex with abscisic acid. Nature 462 , 665-668. Sauter A., Davies W.J. & Hartung W. (2001). The long-distance abscisic acid signal in the droughted plant: the fate of the hormone on its way from root to shoot. J. Exp. Bot. 52 , 1991-1997. Savenije H.H.G. (2000). Water scarcity indicators; the deception of the numbers. Phys. Chem. Earfh (B) 25 , 199-204. Schachtman D.P. & Goodger J.Q.D. (2008). Chemical root to shoot signaling under drought. Trends Plant Sci. 13 , 281-287. Schmid J., Sopp E. & Rühl E.H. (1998). Breeding rootstock varieties with complete phylloxera resistance. Acta Hortic. 473 , 131-138. Schmid J., Manty F. & Rühl E.H. (2003a). Experience with phylloxera tolerant and resistant rootstocks at different vineyard sites. Acta Hortic. 617 , 85-93. Schmid J., Manty F. & Rühl E.H. (2003b). Utilizing the complete phylloxera resistance of Vitis cinerea Arnold in rootstock breeding. Acta Hortic. 603 , 393-400. Schmid J., Manty F., Huber L. et al. (2005). Experience with rootstock varieties in Germany. In: Grapevine rootstocks: current use, research and application (eds P. Cousins & R.K. Striegler), pp. 14-24. Proceedings of the 2005 Rootstock Symposium. Mid-America Viticulture and Enology Center, Southwest Missouri State University, Mountain Grove Campus, Osage Beach, Missouri, USA. Schnall J.A. & Quatrano R.S. (1992). Abscisic acid elicits the water-stress response in root hairs of Arabidopsis thaliana . Plant Physiol. 100 , 216-218. Scholander P.F., Bradstreet E.D., Hemmingsen E.A. & Hammel H.T. (1965). Sap pressure in vascular plants: negative hydrostatic pressure can be measured in plants. Science 148 , 339-346.

- 162 - References

Schreiner R., Tarara J. & Smithyman R. (2007). Deficit irrigation promotes arbuscular colonization of fine roots by mycorrhizal fungi in grapevines ( Vitis vinifera L.) in an arid climate. Mycorrhiza 17 , 551-562. Schreiner R.P. (2003). Mycorrhizal colonization of grapevine rootstocks under field conditions. Am. J. Enol. Vitic. 54 , 143-149. Schreiner R.P. (2005). Spatial and temporal variation of roots, arbuscular mycorrhizal fungi, and plant and soil nutrients in a mature Pinot noir ( Vitis vinifera L.) vineyard in Oregon, USA. Plant Soil 276 , 219-234. Schroeder J.I., Allen G.J., Hugouvieux V. et al. (2001). Guard cell signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52 , 627-658. Schubert A., Restagno M., Novello V. & Peterlunger E. (1995). Effects of shoot orientation on growth, net photosynthesis, and hydraulic conductivity of Vitis vinifera L. cv. Cortese. Am. J. Enol. Vitic. 46 , 324-328. Schubert A., Lovisolo C. & Peterlunger E. (1999). Shoot orientation affects vessel size, shoot hydraulic conductivity and shoot growth rate in Vitis vinifera L. Plant Cell Environ. 22 , 197-204. Schultz H.R. & Matthews M.A. (1988a). Resistance to water transport in shoots of Vitis vinifera L.: relation to growth at low water potential. Plant Physiol. 88 , 718-724. Schultz H.R. & Matthews M.A. (1988b). Vegetative growth distribution during water deficits in Vitis vinifera L. Aust. J. Plant Physiol. 15 , 641-656. Schultz H.R. & Matthews M.A. (1993a). Growth, osmotic adjustment, and cell-wall mechanics of expanding grape leaves during water deficits. Crop Sci. 33 , 287-294. Schultz H.R. & Matthews M.A. (1993b). Xylem development and hydraulic conductance in sun and shade shoots of grapevine ( Vitis vinifera L.): evidence that low light uncouples water transport capacity from leaf area. Planta 190 , 393-406. Schultz H.R. (2000). Climate change and viticulture: a European perspective on climatology, carbon dioxide and UV-B effects. Aust. J. Grape Wine Res. 6, 2-12. Schultz H.R. (2003a). Differences in hydraulic architecture account for near-isohydric and anisohydric behaviour of two field-grown Vitis vinifera L. cultivars during drought. Plant Cell Environ. 26 , 1393-1405. Schultz H.R. (2003b). Extension of a Farquhar model for limitations of leaf photosynthesis induced by light environment, phenology and leaf age in grapevines ( Vitis vinifera L. cvv. White Riesling and Zinfandel). Funct. Plant Biol. 30 , 673-687. Schultz H.R. (2010). Climate change and viticulture: research needs for facing the future J. Wine Res. 21 , 113-116. Schultz H.R. & Jones G.V. (2010). Climate induced historic and future changes in viticulture J. Wine Res. 21 , 137-145. Schultz H.R. & Stoll M. (2010). Some critical issues in environmental physiology of grapevines: future challenges and current limitations. Aust. J. Grape Wine Res. 16 , 4- 24. Schweingruber F.H., Börner A. & Schulze E.-D. (2011). Atlas of stem anatomy in herbs, shrubs and trees (vol. 1). Springer, Berlin Heidelberg, 495 p. Scienza A. (1983). Adattamento genetico della vite allo stress idrico. Vignevini 10 , 27-39. Secchi F. & Zwieniecki M.A. (2010). Patterns of PIP gene expression in Populus trichocarpa during recovery from xylem embolism suggest a major role for the PIP1 aquaporin subfamily as moderators of refilling process. Plant Cell Environ. 33 , 1285- 1297. Secchi F. & Zwieniecki M.A. (2011). Sensing embolism in xylem vessels: the role of sucrose as a trigger for refilling. Plant Cell Environ. 34 , 514-524.

- 163 - References

Seguin G. (1972). Répartition dans l'espace du système radiculaire de la vigne. Comptes Rendus des Séances de l'Academie des Sciences 274 (D) , 2178-2180. Seki M., Narusaka M., Ishida J. et al. (2002). Monitoring the expression profiles of 7000 Arabidopsis genes under drought, cold and high-salinity stresses using a full-length cDNA microarray. Plant J. 31 , 279-292. Seki M., Umezawa T., Urano K. & Shinozaki K. (2007). Regulatory metabolic networks in drought stress responses. Curr. Opin. Plant Biol. 10 , 296-302. Senkerik R., Zelinka I., Davendra D. & Oplatkova Z. (2010). Utilization of SOMA and differential evolution for robust stabilization of chaotic Logistic equation. Comp. Math. Appl. 60 , 1026-1037. Senthil-Kumar M., Hema R., Suryachandra T.R. et al. (2010). Functional characterization of three water deficit stress-induced genes in tobacco and Arabidopsis : an approach based on gene down regulation. Plant Physiol. Biochem. 48 , 35-44. Seo P.J., Xiang F., Qiao M. et al. (2009). The MYB96 transcription factor mediates abscisic acid signaling during drought stress response in Arabidopsis . Plant Physiol. 151 , 275- 289. Seo P.J., Lee S.B., Suh M.C. et al. (2011). The MYB96 transcription factor regulates cuticular wax biosynthesis under drought conditions in Arabidopsis . Plant Cell 23 , 1138-1152. Shang Y., Yan L., Liu Z.-Q. et al. (2010). The Mg-chelatase H subunit of Arabidopsis antagonizes a group of WRKY transcription repressors to relieve ABA-responsive genes of inhibition. Plant Cell 22 , 1909-1935. Shao H.-B., Chu L.-Y., Jaleel C.A. & Zhao C.-X. (2008). Water deficit stress induced anatomical changes in higher plants. C. R. Biol. 331 , 215-225. Sharma S., Villamor J.G. & Verslues P.E. (2011). Essential role of tissue-specific proline synthesis and catabolism in growth and redox balance at low water potential. Plant Physiol. 157 , 292-304. Sharp R.E. & Davies W.J. (1979). Solute regulation and growth by roots and shoots of water-stressed maize plants. Planta 147 , 43-49. Sharp R.E. & Davies W.J. (1985). Root growth and water uptake by maize plants in drying soil. J. Exp. Bot. 36 , 1441-1456. Sharp R.E., Silk W.K. & Hsiao T.C. (1988). Growth of the maize primary root at low water potentials. I. Spatial distribution of expansive growth. Plant Physiol. 87 , 50-57. Sharp R.E. & Davies W.J. (1989). Regulation of growth and development of plants growing with a restricted supply of water. In: Plants under stress: biochemistry, physiology and ecology and their application to plant improvement (eds H.G. Jones, T.J. Flowers & M.B. Jones), pp. 71-94. Cambridge University Press, Cambridge, UK. Sharp R.E., Hsiao T.C. & Silk W.K. (1990). Growth of the maize primary root at low water potentials. II. Role of growth and deposition of hexose and potassium in osmotic adjustment. Plant Physiol. 93 , 1337-1346. Sharp R.E. & LeNoble M.E. (2002). ABA, ethylene and the control of shoot and root growth under water stress. J. Exp. Bot. 53 , 33-37. Sharp R.E. (2002). Interaction with ethylene: changing views on the role of abscisic acid in root and shoot growth responses to water stress. Plant Cell Environ. 25 , 211-222. Sharp R.E., Poroyko V., Hejlek L.G. et al. (2004). Root growth maintenance during water deficits: physiology to functional genomics. J. Exp. Bot. 55 , 2343-2351. Shen Y.-Y., Wang X.-F., Wu F.-Q. et al. (2006). The Mg-chelatase H subunit is an abscisic acid receptor. Nature 443 , 823-826.

- 164 - References

Shi L., Reid L.H., Jones W.D. et al. (2006). The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat. Biotechnol. 24 , 1151-1161. Shi L., Perkins R.G., Fang H. & Tong W. (2008). Reproducible and reliable microarray results through quality control: good laboratory proficiency and appropriate data analysis practices are essential. Curr. Opin. Biotechnol. 19 , 10-18. Shi L., Campbell G., Jones W.D. et al. (2010). The MicroArray Quality Control (MAQC)-II study of common practices for the development and validation of microarray-based predictive models. Nat. Biotechnol. 28 , 827-838. Shigaki T. & Hirschi K.D. (2006). Diverse functions and molecular properties emerging for CAX cation/H + exchangers in plants. Plant Biol. 8, 419-429. Shimazaki Y., Ookawa T. & Hirasawa T. (2005). The root tip and accelerating region suppress elongation of the decelerating region without any effects on cell turgor in primary roots of maize under water stress. Plant Physiol. 139 , 458-465. Shinozaki K. & Yamaguchi-Shinozaki K. (1997). Gene expression and signal transduction in water-stress response. Plant Physiol. 115 , 327-334. Shinozaki K., Yamaguchi-Shinozaki K. & Seki M. (2003). Regulatory network of gene expression in the drought and cold stress responses. Curr. Opin. Plant Biol. 6, 410-417. Shinozaki K. & Yamaguchi-Shinozaki K. (2007). Gene networks involved in drought stress response and tolerance. J. Exp. Bot. 58 , 221-227. Shinozaki K. & Sakakibara H. (2009). Omics and bioinformatics: an essential toolbox for systems analyses of plant functions beyond 2010. Plant Cell Physiol. 50 , 1177-1180. Shishkova S., Rost T.L. & Dubrovsky J.G. (2008). Determinate root growth and meristem maintenance in Angiosperms. Ann. Bot. 101 , 319-340. Shulaev V., Cortes D., Miller G. & Mittler R. (2008). Metabolomics for plant stress response. Physiol. Plant. 132 , 199-208. Sinclair T.R. (2011). Challenges in breeding for yield increase for drought. Trends Plant Sci. 16 , 289-293. Sirichandra C., Wasilewska A., Vlad F. et al. (2009). The guard cell as a single-cell model towards understanding drought tolerance and abscisic acid action. J. Exp. Bot. 60 , 1439-1463. Slavtcheva T. & Pourtchev P. (2007). Effect of training on root system development of cv. Merlot grapevines. Acta Hortic. 754 , 221-226. Smart D.R., Carlisle E., Goebel M. & Núñez B.A. (2005). Transverse hydraulic redistribution by a grapevine. Plant Cell Environ. 28 , 157-166. Smart D.R., Schwass E., Lakso A. & Morano L. (2006). Grapevine rooting patterns: a comprehensive analysis and a review. Am. J. Enol. Vitic. 57 , 89-104. Smeekens S., Ma J., Hanson J. & Rolland F. (2010). Sugar signals and molecular networks controlling plant growth. Curr. Opin. Plant Biol. 13 , 273-278. Smyth G.K. (2004). Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat. Appl. Genet. Molec. 3, Article 3, 26 p. Smyth G.K. (2005). limma: Linear Models for Microarray Data. In: Bioinformatics and computational biology solutions using R and bioconductor (eds R. Gentleman, V.J. Carey, W. Huber, R.A. Irizarry & S. Dudoit), pp. 397-420. Springer, New York, USA. Soar C.J., Speirs J., Maffei S.M. & Loveys B.R. (2004). Gradients in stomatal conductance, xylem sap ABA and bulk leaf ABA along canes of Vitis vinifera cv. Shiraz: molecular and physiological studies investigating their source. Funct. Plant Biol. 31 , 659-669. Soar C.J., Speirs J., Maffei S.M. et al. (2006a). Grape vine varieties Shiraz and Grenache differ in their stomatal response to VPD: apparent links with ABA physiology and gene expression in leaf tissue. Aust. J. Grape Wine Res. 12 , 2-12.

- 165 - References

Soar C.J., Dry P.R. & Loveys B.R. (2006b). Scion photosynthesis and leaf gas exchange in Vitis vinifera L. cv. Shiraz: mediation of rootstock effects via xylem sap ABA. Aust. J. Grape Wine Res. 12 , 82-96. Soar C.J. & Loveys B.R. (2007). The effect of changing patterns in soil-moisture availability on grapevine root distribution, and viticultural implications for converting full-cover irrigation into a point-source irrigation system. Aust. J. Grape Wine Res. 13 , 1-13. Solari L.I. & DeJong T.M. (2006). The effect of root pressurization on water relations, shoot growth, and leaf gas exchange of peach ( Prunus persica ) trees on rootstocks with differing growth potential and hydraulic conductance. J. Exp. Bot. 57 , 1981-1989. Solari L.I., Johnson S. & DeJong T.M. (2006). Hydraulic conductance characteristics of peach ( Prunus persica ) trees on different rootstocks are related to biomass production and distribution. Tree Physiol. 26 , 1343-1350. Southey J.M. & Archer E. (1988). The effect of rootstock cultivar on grapevine root distribution and density. In: The grapevine root and its environment. Technical communication n° 215 (ed J.L. Van Zyl), pp. 57-73. Department of agriculture and water supply, Stellenbosch. Southey J.M. (1992). Grapevine rootstocks performance under diverse conditions in South Africa. In: Rootstock seminar: a worldwide perspective (eds J.A. Wolpert, M.A. Walker & E. Weber), pp. 27-51. American Society for Enology and Viticulture, Davis, USA. Speirs J., Loveys B. & Binney A. (2010). Optimizing irrigation and response to abiotic stress through development of biochemical, physiological and molecular markers of vine performance . CSIRO Plant Industry, Glen Osmond, Australia, 126 p. Sperry J.S., Hacke U.G., Oren R. & Comstock J.P. (2002). Water deficits and hydraulic limits to leaf water supply. Plant Cell Environ. 25 , 251-263. Sperry J.S., Hacke U.G. & Pittermann J. (2006). Size and function in conifer tracheids and angiosperm vessels. Am. J. Bot. 93 , 1490-1500. Spollen W., Tao W., Valliyodan B. et al. (2008). Spatial distribution of transcript changes in the maize primary root elongation zone at low water potential. BMC Plant Biol. 8, 32. Spollen W.G., LeNoble M.E., Samuels T.D. et al. (2000). Abscisic acid accumulation maintains maize primary root elongation at low water potentials by restricting ethylene production. Plant Physiol. 122 , 967-976. Sreenivasulu N., Sunkar R., Wobus U. & Strickert M. (2010). Array platforms and bioinformatics tools for the analysis of plant transcriptome in response to abiotic stress. In: Plant stress tolerance. Methods in molecular biology (ed R. Sunkar), pp. 71-93. Humana Press Inc., Springer, New York, USA. Steudle E. & Peterson C. (1998). How does water get through roots? J. Exp. Bot. 49 , 775- 788. Steudle E. (2000a). Water uptake by plant roots: an integration of views. Plant Soil 226 , 45- 56. Steudle E. (2000b). Water uptake by roots: effects of water deficit. J. Exp. Bot. 51 , 1531- 1542. Steudle E. (2001). The cohesion-tension mechanism and the acquisition of water by plant roots. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52 , 847-875. Stevens R.M. & Douglas T. (1994). Distribution of grapevine roots and salt under drip and full-ground cover microjet irrigation systems. Irrig. Sci. 15 , 147-152. Stevens R.M., Pech J.M., Gibberd M.R. et al. (2008). Effect of reduced irrigation on growth, yield, ripening rates and water relations of Chardonnay vines grafted to five rootstocks. Aust. J. Grape Wine Res. 14 , 177-190.

- 166 - References

Stevens R.M., Pech J.M., Gibberd M.R. et al. (2010). Reduced irrigation and rootstock effects on vegetative growth, yield and its components, and leaf physiological responses of Shiraz. Aust. J. Grape Wine Res. 16 , 413-425. Stolf-Moreira R., Lemos E., Carareto-Alves L. et al. (2011). Transcriptional profiles of roots of different soybean genotypes subjected to drought stress. Plant Mol. Biol. Report. 29 , 19-34. Stoll M., Loveys B. & Dry P. (2000). Hormonal changes induced by partial rootzone drying of irrigated grapevine. J. Exp. Bot. 51 , 1627-1634. Stoll M., Lafontaine M. & Schultz H.R. (2010). Possibilities to reduce the velocity of berry maturation through various leaf area to fruit ratio modifications in Vitis vinifera L. cv. Riesling. Prog. Agric. Vitic. 127 , 68-71. Street N.R., Skogström O., Sjödin A. et al. (2006). The genetics and genomics of the drought response in Populus . Plant J. 48 , 321-341. Strommer J. (2011). The plant ADH gene family. Plant J. 66 , 128-142. Suzek B.E., Huang H., McGarvey P. et al. (2007). UniRef: comprehensive and non- redundant UniProt reference clusters. Bioinformatics 23 , 1282-1288. Svistoonoff S., Creff A., Reymond M. et al. (2007). Root tip contact with low-phosphate media reprograms plant root architecture. Nat. Genet. 39 , 792-796. Swanepoel J. & Southey J.M. (1989). The influence of rootstock on the rooting pattern of the grapevine. S. Afr. J. Enol. Vitic. 10 , 23-28. Szabados L. & Savouré A. (2010). Proline: a multifunctional amino acid. Trends Plant Sci. 15 , 89-97. Szabados L., Kovács H., Zilberstein A. & Bouchereau A. (2011). Plants in extreme environments: importance of protective compounds in stress tolerance. Adv. Bot. Res. 57 , 105-150. Taiz L. & Zeiger E. (2002). Plant Physiology . (3 rd edition ed.). Sinauer Associates, Sunderland, 690 p. Taji T., Ohsumi C., Iuchi S. et al. (2002). Important roles of drought- and cold-inducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana . Plant J. 29 , 417-426. Takahashi H. & Scott T.K. (1993). Intensity of hydrostimulation for the induction of root hydrotropism and its sensing by the root cap. Plant Cell Environ. 16 , 99-103. Takahashi H. (1994). Hydrotropism and its interaction with gravitropism in roots. Plant Soil 165 , 301-308. Takahashi H. (1997). Hydrotropism: the current state of our knowledge. J. Plant Res. 110 , 163-169. Takahashi H., Miyazawa Y. & Fujii N. (2009). Hormonal interactions during root tropic growth: hydrotropism versus gravitropism. Plant Mol. Biol. 69 , 489-502. Takahashi N., Goto N., Okada K. & Takahashi H. (2002). Hydrotropism in abscisic acid, wavy, and gravitropic mutants of Arabidopsis thaliana . Planta 216 , 203-211. Takahashi N., Yamazaki Y., Kobayashi A. et al. (2003). Hydrotropism interacts with gravitropism by degrading amyloplasts in seedling roots of Arabidopsis and radish. Plant Physiol. 132 , 805-810. Takano M., Takahashi H., Hirasawa T. & Suge H. (1995). Hydrotropism in roots: sensing of a gradient in water potential by the root cap. Planta 197 , 410-413. Takano M., Takahashi H. & Suge H. (1997). Calcium requirement for the induction of hydrotropism and enhancement of calcium-induced curvature by water stress in primary roots of pea, Pisum sativum L. Plant Cell Physiol. 38 , 385-391. Takeda S. & Matsuoka M. (2008). Genetic approaches to crop improvement: responding to environmental and population changes. Nat. Rev. Genet. 9, 444-457.

- 167 - References

Tamura T., Hara K., Yamaguchi Y. et al. (2003). Osmotic stress tolerance of transgenic tobacco expressing a gene encoding a membrane-located receptor-like protein from tobacco plants. Plant Physiol. 131 , 454-462. Tandonnet J.-P., Ollat N., Neveux M. & Renoux J.-L. (1999). Effect of three levels of water supply on the vegetative and reproductive development of Merlot and Cabernet Sauvignon grapevines. Acta Hortic. 493 , 301-307. Tandonnet J.-P., Cookson S.J., Vivin P. & Ollat N. (2010). Scion genotype controls biomass allocation and root development in grafted grapevine. Aust. J. Grape Wine Res. 16 , 290-300. Taniguchi Y., Taniguchi M., Tsuge T. et al. (2010). Involvement of Arabidopsis thaliana phospholipase D #2 in root hydrotropism through the suppression of root gravitropism. Planta 231 , 491-497. Tanimoto E. (2005). Regulation of root growth by plant hormones - Roles for auxin and gibberellin. Crit. Rev. Plant Sci. 24 , 249-265. Tardieu F., Bruckler L. & Lafolie F. (1992a). Root clumping may affect the root water potential and the resistance to soil-root water transport. Plant Soil 140 , 291-301. Tardieu F. & Davies W.J. (1992). Stomatal response to abscisic acid is a function of current plant water status. Plant Physiol. 98 , 540-545. Tardieu F., Zhang J. & Davies W.J. (1992b). What information is conveyed by an ABA signal from maize roots in drying field soil? Plant Cell Environ. 15 , 185-191. Tardieu F., Zhang J., Katerji N. et al. (1992c). Xylem ABA controls the stomatal conductance of field-grown maize subjected to soil compaction or soil drying. Plant Cell Environ. 15 , 193-197. Tardieu F. & Davies W.J. (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. Tardieu F., Zhang J. & Gowing D.J.G. (1993). Stomatal control by both [ABA] in the xylem sap and leaf water status: a test of a model for droughted or ABA-fed field- grown maize. Plant Cell Environ. 16 , 413-420. Tardieu F. & Simonneau T. (1998). Variability among species of stomatal control under fluctuating soil water status and evaporative demand: modelling isohydric and anisohydric behaviours. J. Exp. Bot. 49 , 419-432. Tardieu F. (2003). Virtual plants: modelling as a tool for the genomics of tolerance to water deficit. Trends Plant Sci. 8, 9-14. Tardieu F. (2005). Plant tolerance to water deficit: physical limits and possibilities for progress. Comptes Rendus Geosciences 337 , 57-67. Tardieu F., Parent B. & Simonneau T. (2010). Control of leaf growth by abscisic acid: hydraulic or non-hydraulic processes? Plant Cell Environ. 33 , 636-647. Tardieu F. & Tuberosa R. (2010). Dissection and modelling of abiotic stress tolerance in plants. Curr. Opin. Plant Biol. 13 , 206-212. Tattersall E.A.R., Grimplet J., Deluc L. et al. (2007). Transcript abundance profiles reveal larger and more complex responses of grapevine to chilling compared to osmotic and salinity stress. Funct. Integr. Genomics 7, 317-333. Terral J.-F., Tabard E., Bouby L. et al. (2010). Evolution and history of grapevine ( Vitis vinifera ) under domestication: new morphometric perspectives to understand seed domestication syndrome and reveal origins of ancient European cultivars. Ann. Bot. 105 , 443-455. Tesniere C., Torregrosa L., Pradal M. et al. (2006). Effects of genetic manipulation of alcohol dehydrogenase levels on the response to stress and the synthesis of secondary metabolites in grapevine leaves. J. Exp. Bot. 57 , 91-99.

- 168 - References

Testerink C. & Munnik T. (2005). Phosphatidic acid: a multifunctional stress signaling lipid in plants. Trends Plant Sci. 10 , 368-375. Testerink C. & Munnik T. (2011). Molecular, cellular, and physiological responses to phosphatidic acid formation in plants. J. Exp. Bot. 62 , 2349-2361. The UniProt Consortium (2011). Ongoing and future developments at the Universal Protein Resource. Nucleic Acids Res. 39 , D214-D219. This P., Lacombe T. & Thomas M.R. (2006). Historical origins and genetic diversity of wine grapes. Trends Genet. 22 , 511-519. Thomas P. & Schiefelbein J. (2002). Cloning and characterization of an actin depolymerizing factor gene from grape ( Vitis vinifera L.) expressed during rooting in stem cuttings. Plant Sci. 162 , 283-288. Thomas P., Lee M.M. & Schiefelbein J. (2003). Molecular identification of proline-rich protein genes induced during root formation in grape ( Vitis vinifera L.) stem cuttings. Plant Cell Environ. 26 , 1497-1504. Thompson A.J., Andrews J., Mulholland B.J. et al. (2007). Overproduction of abscisic acid in tomato increases transpiration efficiency and root hydraulic conductivity and influences leaf expansion. Plant Physiol. 143 , 1905-1917. Tillett R.L., Ergül A., Albion R.L. et al. (2011). Identification of tissue-specific, abiotic stress-responsive gene expression patterns in wine grape ( Vitis vinifera L.) based on curation and mining of large-scale EST data sets. BMC Plant Biol. 11 , 86. Tillett R.L. & Cushman J.C. (2011). Vitis functional genomics: open systems for transcriptome analysis. In: Genetics, genomics and breeding of grapes (eds A.-F. Adam-Blondon, J.M. Martínez-Zapater & C. Kole), pp. 235-269. Science Publishers, CRC Press, Taylor & Francis group, New York, USA. Tombesi S., Johnson R.S., Day K.R. & DeJong T.M. (2010a). Interactions between rootstock, inter-stem and scion xylem vessel characteristics of peach trees growing on rootstocks with contrasting size-controlling characteristics. AoB Plants 2010 , plq013. Tombesi S., Johnson R.S., Day K.R. & DeJong T.M. (2010b). Relationships between xylem vessel characteristics, calculated axial hydraulic conductance and size- controlling capacity of peach rootstocks. Ann. Bot. 105 , 327-331. Töpfer R., Hausmann L. & Eibach R. (2011). Molecular breeding. In: Genetics, genomics and breeding of grapes (eds A.-F. Adam-Blondon, J.M. Martínez-Zapater & C. Kole), pp. 160-185. Science Publishers, CRC Press, Taylor & Francis group, New York, USA. Tóth H. & Kocsis L. (2003). Level of phylloxera resistance of two new rootstocks hybrids: Georgikon 28 and Georgikon 121. Acta Hortic. 603 , 739-743. Toumi I., Moschou P.N., Paschalidis K.A. et al. (2010). Abscisic acid signals reorientation of polyamine metabolism to orchestrate stress responses via the polyamine exodus pathway in grapevine. J. Plant Physiol. 167 , 519-525. Tran L.-S.P., Nakashima K., Sakuma Y. et al. (2004). Isolation and functional analysis of Arabidopsis stress-inducible NAC transcription factors that bind to a drought- responsive cis -element in the early responsive to dehydration stress 1 promoter. Plant Cell 16 , 2481-2498. Tran L.-S.P., Urao T., Qin F. et al. (2007). Functional analysis of AHK1/ATHK1 and cytokinin receptor histidine kinases in response to abscisic acid, drought, and salt stress in Arabidopsis . Proc. Natl. Acad. Sci. 104 , 20623-20628. Treeby M.T., Holzapfel B.P., Walker R.R. & Nicholas P.R. (1998). Profiles of free amino acids in grapes of grafted Chardonnay grapevines. Aust. J. Grape Wine Res. 4, 121- 126.

- 169 - References

Trifilò P., Raimondo F., Nardini A. et al. (2004). Drought resistance of Ailanthus altissima : root hydraulics and water relations. Tree Physiol. 24 , 107-114. Trifilò P., Lo Gullo M.A., Nardini A. et al. (2007). Rootstock effects on xylem conduit dimensions and vulnerability to cavitation of Olea europaea L. Trees-Struc. Func. 21 , 549-556. Troggio M., Vezzulli S., Pindo M. et al. (2008). Beyond the genome, opportunities for a modern viticulture: a research overview. Am. J. Enol. Vitic. 59 , 117-127. Tröndle D., Schröder S., Kassemeyer H.-H. et al. (2010). Molecular phylogeny of the genus Vitis (Vitaceae) based on plastid markers. Am. J. Bot. 97 , 1168-1178. Trontin C., Tisné S., Bach L. & Loudet O. (2011). What does Arabidopsis natural variation teach us (and does not teach us) about adaptation in plants? Curr. Opin. Plant Biol. 14 , 225-231. Tuberosa R. & Salvi S. (2006). Genomics-based approaches to improve drought tolerance of crops. Trends Plant Sci. 11 , 405-412. Turner N.C. & Long M.J. (1980). Errors arising from rapid water loss in the measurement of leaf water potential by the pressure chamber technique. Aust. J. Plant Physiol. 7, 527-537. Turner N.C. (1988). Measurement of plant water status by the pressure chamber technique. Irrig. Sci. 9, 289-308. Tyree M.T. & Sperry J.S. (1988). Do woody plants operate near the point of catastrophic xylem dysfunction caused by dynamic water stress? Answers from a model. Plant Physiol. 88 , 574-580. Tyree M.T. & Sperry J.S. (1989). Vulnerability of xylem to cavitation and embolism. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40 , 19-36. Tyree M.T. & Ewers F.W. (1991). The hydraulic architecture of trees and other woody plants. New Phytol. 119 , 345-360. Tyree M.T., Davis S.D. & Cochard H. (1994). Biophysical perspectives of xylem evolution: is there a tradeoff of hydraulic efficiency for vulnerability to dysfunction? IAWA J. 15 , 335-360. Tyree M.T. (1997). The Cohesion-Tension theory of sap ascent: current controversies. J. Exp. Bot. 48 , 1753-1765. Tyree M.T. & Zimmermann M.H. (2002). Xylem structure and the ascent of sap . (2nd ed.). Springer, Berlin, 304 p. Ubeda-Tomas S., Swarup R., Coates J. et al. (2008). Root growth in Arabidopsis requires gibberellin/DELLA signalling in the endodermis. Nat. Cell Biol. 10 , 625-628. Umezawa T., Yoshida R., Maruyama K. et al. (2004). SRK2C, a SNF1-related protein kinase 2, improves drought tolerance by controlling stress-responsive gene expression in Arabidopsis thaliana . Proc. Natl. Acad. Sci. 101 , 17306-17311. Umezawa T., Fujita M., Fujita Y. et al. (2006). Engineering drought tolerance in plants: discovering and tailoring genes to unlock the future. Curr. Opin. Biotechnol. 17 , 113- 122. Umezawa T., Sugiyama N., Mizoguchi M. et al. (2009). Type 2C protein phosphatases directly regulate abscisic acid-activated protein kinases in Arabidopsis . Proc. Natl. Acad. Sci. 106 , 17588-17593. Umezawa T., Nakashima K., Miyakawa T. et al. (2010). Molecular basis of the core regulatory network in ABA responses: sensing, signaling and transport. Plant Cell Physiol. 51 , 1821-1839. Umezawa T., Hirayama T., Kuromori T. & Shinozaki K. (2011). The regulatory networks of plant responses to abscisic acid. Adv. Bot. Res. 57 , 201-248.

- 170 - References

Unwin T. (1996). Wine and the vine: an historical geography of viticulture and the wine trade . Routledge, Taylor & Francis group, New York, USA, 390 p. Upreti K. & Murti G. (2010). Response of grape rootstocks to salinity: changes in root growth, polyamines and abscisic acid. Biol. Plant. 54 , 730-734. Urano K., Maruyama K., Ogata Y. et al. (2009). Characterization of the ABA-regulated global responses to dehydration in Arabidopsis by metabolomics. Plant J. 57 , 1065- 1078. Urano K., Kurihara Y., Seki M. & Shinozaki K. (2010). `Omics' analyses of regulatory networks in plant abiotic stress responses. Curr. Opin. Plant Biol. 13 , 132-138. Urao T., Yakubov B., Satoh R. et al. (1999). A transmembrane hybrid-type histidine kinase in Arabidopsis functions as an osmosensor. Plant Cell 11 , 1743-1754. Usadel B., Obayashi T., Mutwil M. et al. (2009). Co-expression tools for plant biology: opportunities for hypothesis generation and caveats. Plant Cell Environ. 32 , 1633- 1651. Valliyodan B. & Nguyen H.T. (2006). Understanding regulatory networks and engineering for enhanced drought tolerance in plants. Curr. Opin. Plant Biol. 9, 189-195. Van As H. (2007). Intact plant MRI for the study of cell water relations, membrane permeability, cell-to-cell and long distance water transport. J. Exp. Bot. 58 , 743-756. van der Weele C.M., Spollen W.G., Sharp R.E. & Baskin T.I. (2000). Growth of Arabidopsis thaliana seedlings under water deficit studied by control of water potential in nutrient-agar media. J. Exp. Bot. 51 , 1555-1562. van Leeuwen C. & Seguin G. (2006). The concept of terroir in viticulture J. Wine Res. 17 , 1- 10. van Leeuwen C., Tregoat O., Choné X. et al. (2009). Vine water status is a key factor in grape ripening and vintage quality for red Bordeaux wine. How can it be assessed for vineyard management purposes? J. Int. Sci. Vigne Vin 43 , 121-134. Van Norman J.M. & Benfey P.N. (2009). Arabidopsis thaliana as a model organism in systems biology. Wiley Interdiscip. Rev. Syst. Biol. Med. 1, 372-379. Van Rooyen M., Valentine A. & Archer E. (2004). Arbuscular mycorrhizal colonisation modifies the water relations of young transplanted grapevines ( Vitis ). S. Afr. J. Enol. Vitic. 25 , 37-42. Van Zyl J.L. (1988a). The grapevine root and its environment. Technical communication n° 215 . Department of agriculture and water supply, Stellenbosch, South Africa, 146 p. Van Zyl J.L. (1988b). Response of grapevine roots to soil water regimes and irrigation systems. In: The grapevine root and its environment. Technical communication n° 215 (ed J.L. Van Zyl), pp. 30-43. Department of agriculture and water supply, Stellenbosch, South Africa. Vandeleur R., Niemietz C., Tilbrook J. & Tyerman S. (2005). Roles of aquaporins in root responses to irrigation. Plant Soil 274 , 141-161. Vandeleur R.K., Mayo G., Shelden M.C. et al. (2009). The role of plasma membrane intrinsic protein aquaporins in water transport through roots: diurnal and drought stress responses reveal different strategies between isohydric and anisohydric cultivars of grapevine. Plant Physiol. 149 , 445-460. Vandesompele J., De Preter K., Pattyn F. et al. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3, research0034.0031-0034.0011. Varadhan R. & Gilbert P. (2009). BB: an R package for solving a large system of nonlinear equations and for optimizing a high-dimensional nonlinear objective function. J. Stat. Softw. 32 , 1-26.

- 171 - References

Varshney R.K., Graner A. & Sorrells M.E. (2005). Genomics-assisted breeding for crop improvement. Trends Plant Sci. 10 , 621-630. Varshney R.K., Bansal K.C., Aggarwal P.K. et al. (2011). Agricultural biotechnology for crop improvement in a variable climate: hope or hype? Trends Plant Sci. 16 , 363-371. Vartanian N. (1981). Some aspects of structural and functional modifications induced by drought in root systems. Plant Soil 63 , 83-92. Vartanian N., Marcotte L. & Giraudat J. (1994). Drought rhizogenesis in Arabidopsis thaliana (Differential responses of hormonal mutants). Plant Physiol. 104 , 761-767. Velasco R., Zharkikh A., Troggio M. et al. (2007). A high quality draft consensus sequence of the genome of a heterozygous grapevine variety. PLoS ONE 2, e1326. Verslues P.E. & Sharp R.E. (1999). Proline accumulation in maize ( Zea mays L.) primary roots at low water potentials. II. Metabolic source of increased proline deposition in the elongation zone. Plant Physiol. 119 , 1349-1360. Verslues P.E. & Bray E.A. (2004). LWR1 and LWR2 are required for osmoregulation and osmotic adjustment in Arabidopsis . Plant Physiol. 136 , 2831-2842. Verslues P.E. & Bray E.A. (2006). Role of abscisic acid (ABA) and Arabidopsis thaliana ABA-insensitive loci in low water potential-induced ABA and proline accumulation. J. Exp. Bot. 57 , 201-212. Verslues P.E. & Juenger T.E. (2011). Drought, metabolites, and Arabidopsis natural variation: a promising combination for understanding adaptation to water-limited environments. Curr. Opin. Plant Biol. 14 , 240-245. Viala P. & Vermorel V. (1901-1910). Ampélographie. Traité générale de viticulture (vol. 1- 7). Masson et C ie , Éditeurs, Paris. Vidavalur R., Otani H., Singal P.K. & Maulik N. (2006). Significance of wine and resveratrol in cardiovascular disease: French paradox revisited. Exp. Clin. Cardiol. 11 , 217-225. Vincent D., Ergül A., Bohlman M.C. et al. (2007). Proteomic analysis reveals differences between Vitis vinifera L. cv. Chardonnay and cv. Cabernet Sauvignon and their responses to water deficit and salinity. J. Exp. Bot. 58 , 1873-1892. Vinocur B. & Altman A. (2005). Recent advances in engineering plant tolerance to abiotic stress: achievements and limitations. Curr. Opin. Biotechnol. 16 , 123-132. Vivier M.A. & Pretorius I.S. (2002). Genetically tailored grapevines for the wine industry. Trends Biotechnol. 20 , 472-478. Vlad F., Rubio S., Rodrigues A. et al. (2009). Protein phosphatases 2C regulate the activation of the Snf1-related kinase OST1 by abscisic acid in Arabidopsis . Plant Cell 21 , 3170-3184. Voetberg G.S. & Sharp R.E. (1991). Growth of the maize primary root at low water potentials. III. Role of increased proline deposition in osmotic adjustment. Plant Physiol. 96 , 1125-1130. Volder A., Smart D.R., Bloom A.J. & Eissenstat D.M. (2005). Rapid decline in nitrate uptake and respiration with age in fine lateral roots of grape: implications for root efficiency and competitive effectiveness. New Phytol. 165 , 493-502. von Caemmerer S. & Farquhar G.D. (1981). Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153 , 376-387. Vouillamoz J., Maigre D. & Meredith C.P. (2003). Microsatellite analysis of ancient alpine grape cultivars: pedigree reconstruction of Vitis vinifera L. 'Cornalin du Valais'. Theor. Appl. Genet. 107 , 448-454. Vouillamoz J.F. & Grando M.S. (2006). Genealogy of wine grape cultivars: 'Pinot' is related to 'Syrah'. Heredity 97 , 102-110.

- 172 - References

Walker A.R., Lee E. & Robinson S. (2006). Two new grape cultivars, bud sports of Cabernet Sauvignon bearing pale-coloured berries, are the result of deletion of two regulatory genes of the berry colour locus. Plant Mol. Biol. 62 , 623-635. Walker A.R., Lee E., Bogs J. et al. (2007). White grapes arose through the mutation of two similar and adjacent regulatory genes. Plant J. 49 , 772-785. Walker M.A. (1992). Future directions for rootstock breeding. In: Rootstock seminar: a worldwide perspective (eds J.A. Wolpert, M.A. Walker & E. Weber), pp. 60-68. American Society for Enology and Viticulture, Davis, USA. Walker M.A., Wolpert J.A. & Weber E. (1994). Viticultural characteristics of VR hybrid rootstocks in a vineyard site infected with grapevine fanleaf virus. Vitis 33 , 19-23. Walker R.R., Read P.E. & Blackmore D.H. (2000). Rootstock and salinity effects on rates of berry maturation, ion accumulation and colour development in Shiraz grapes. Aust. J. Grape Wine Res. 6, 227-239. Walker R.R., Blackmore D.H. & Clingeleffer P.R. (2010). Impact of rootstock on yield and ion concentrations in petioles, juice and wine of Shiraz and Chardonnay in different viticultural environments with different irrigation water salinity. Aust. J. Grape Wine Res. 16 , 243-257. Walter A., Silk W.K. & Schurr U. (2009). Environmental effects on spatial and temporal patterns of leaf and root growth. Annu. Rev. Plant Biol. 60 , 279-304. Walther G.R., Post E., Convey P. et al. (2002). Ecological responses to recent climate change. Nature 416 , 389-395. Wang D., Pan Y., Zhao X. et al. (2011). Genome-wide temporal-spatial gene expression profiling of drought responsiveness in rice. BMC Genomics 12 , 149. Wang X. (2005). Regulatory functions of phospholipase D and phosphatidic acid in plant growth, development, and stress responses. Plant Physiol. 139 , 566-573. Wang Y., Li K. & Li X. (2009). Auxin redistribution modulates plastic development of root system architecture under salt stress in Arabidopsis thaliana . J. Plant Physiol. 166 , 1637-1645. Webb L.B., Whetton P.H. & Barlow E.W.R. (2007). Modelled impact of future climate change on the phenology of winegrapes in Australia. Aust. J. Grape Wine Res. 13 , 165-175. Weidner S., Karolak M., Karama 4 M. et al. (2009). Phenolic compounds and properties of antioxidants in grapevine roots ( Vitis vinifera L.) under drought stress followed by recovery. Acta Soc. Bot. Pol. 78 , 97-103. Weinl S. & Kudla J. (2009). The CBL-CIPK Ca 2+ -decoding signaling network: function and perspectives. New Phytol. 184 , 517-528. Welcker C., Boussuge B., Bencivenni C. et al. (2007). Are source and sink strengths genetically linked in maize plants subjected to water deficit? A QTL study of the responses of leaf growth and of Anthesis-Silking Interval to water deficit. J. Exp. Bot. 58 , 339-349. Welcker C., Sadok W., Dignat G. et al. (2011). A common genetic determinism for sensitivities to soil water deficit and evaporative demand: meta-analysis of quantitative trait loci and introgression lines of maize. Plant Physiol. 157 , 718-729. Werner T., Nehnevajova E., Kollmer I. et al. (2011). Root-specific reduction of cytokinin causes enhanced root growth, drought tolerance, and leaf mineral enrichment in Arabidopsis and tobacco. Plant Cell 22 , 3905-3920. Westgate M.E. & Boyer J.S. (1985). Osmotic adjustment and the inhibition of leaf, root, stem and silk growth at low water potentials in maize. Planta 164 , 540-549.

- 173 - References

Wheatley M.D., Tattersall E.A.R., Tillett R.L. & Cramer G.R. (2009). An expanded clay pebble, continuous recirculating drip system for viable long-term hydroponic grapevine culture. Am. J. Enol. Vitic. 60 , 542-549. Wheaton A.D., McKenzie B.M. & Tisdall J.M. (2008). Management to increase the depth of soft soil improves soil conditions and grapevine performance in an irrigated vineyard. Soil Tillage Res. 98 , 68-80. Wheeler T.D. & Stroock A.D. (2008). The transpiration of water at negative pressures in a synthetic tree. Nature 455 , 208-212. White M.A., Diffenbaugh N.S., Jones G.V. et al. (2006). Extreme heat reduces and shifts United States premium wine production in the 21st century. Proc. Natl. Acad. Sci. 103 , 11217-11222. White M.A., Whalen P. & Jones G.V. (2009). Land and wine. Nature Geosci. 2, 82-84. Whiting J.R. (2005). Grapevine rootstocks. In: Viticulture: volume 1 - Resources (eds P.R. Dry & B.G. Coombe), pp. 167-188. Winetitles Pty Ltd, Ashford, Australia. Whitmore A.P. & Whalley W.R. (2009). Physical effects of soil drying on roots and crop growth. J. Exp. Bot. 60 , 2845-2857. Wilkins O., Waldron L., Nahal H. et al. (2009). Genotype and time of day shape the Populus drought response. Plant J. 60 , 703-715. Wilkins O., Bräutigam K. & Campbell M.M. (2010). Time of day shapes Arabidopsis drought transcriptomes. Plant J. 63 , 715-727. Wilkinson S. & Davies W.J. (2002). ABA-based chemical signalling: the co-ordination of responses to stress in plants. Plant Cell Environ. 25 , 195-210. Wilkinson S. & Hartung W. (2009). Food production: reducing water consumption by manipulating long-distance chemical signalling in plants. J. Exp. Bot. 60 , 1885-1891. Williams L.E. & Smith R.J. (1991). The effect of rootstock on the partitioning of dry weight, nitrogen and potassium, and root distribution of Cabernet Sauvignon grapevines. Am. J. Enol. Vitic. 42 , 118-122. Wohlbach D.J., Quirino B.F. & Sussman M.R. (2008). Analysis of the Arabidopsis histidine kinase ATHK1 reveals a connection between vegetative osmotic stress sensing and seed maturation. Plant Cell 20 , 1101-1117. Wright I.J., Reich P.B., Westoby M. et al. (2004). The worldwide leaf economics spectrum. Nature 428 , 821-827. Wu Y., Spollen W.G., Sharp R.E. et al. (1994). Root growth maintenance at low water potentials: increased activity of xyloglucan endotransglycosylase and its possible regulation by abscisic acid. Plant Physiol. 106 , 607-615. Wu Y., Sharp R.E., Durachko D.M. & Cosgrove D.J. (1996). Growth maintenance of the maize primary root at low water potentials involves increases in cell-wall extension properties, expansin activity, and wall susceptibility to expansins. Plant Physiol. 111 , 765-772. Wu Y. & Cosgrove D.J. (2000). Adaptation of roots to low water potentials by changes in cell wall extensibility and cell wall proteins. J. Exp. Bot. 51 , 1543-1553. Wu Y., Thorne E.T., Sharp R.E. & Cosgrove D.J. (2001). Modification of expansin transcript levels in the maize primary root at low water potentials. Plant Physiol. 126 , 1471-1479. Xiao H. & Nassuth A. (2006). Stress- and development-induced expression of spliced and unspliced transcripts from two highly similar dehydrin 1 genes in V. riparia and V. vinifera . Plant Cell Rep. 25 , 968-977. Xiao H., Siddiqua M., Braybrook S. & Nassuth A. (2006). Three grape CBF/DREB1 genes respond to low temperature, drought and abscisic acid. Plant Cell Environ. 29 , 1410- 1421.

- 174 - References

Xiao X., Yang F., Zhang S. et al. (2009). Physiological and proteomic responses of two contrasting Populus cathayana populations to drought stress. Physiol. Plant. 136 , 150- 168. Xiong L., Wang R.-G., Mao G. & Koczan J.M. (2006). Identification of drought tolerance determinants by genetic analysis of root response to drought stress and abscisic acid. Plant Physiol. 142 , 1065-1074. Xu G.-Y., Rocha P., Wang M.-L. et al. (2011). A novel rice calmodulin-like gene, OsMSR2 , enhances drought and salt tolerance and increases ABA sensitivity in Arabidopsis. Planta 234 , 47-59. Yamaguchi-Shinozaki K. & Shinozaki K. (2006). Transcriptional regulatory networks in cellular responses and tolerance to dehydration and cold stresses. Annu. Rev. Plant Biol. 57 , 781-803. Yamaguchi M. & Sharp R.E. (2010). Complexity and coordination of root growth at low water potentials: recent advances from transcriptomic and proteomic analyses. Plant Cell Environ. 33 , 590-603. Yamaguchi M., Valliyodan B., Zhang J. et al. (2010). Regulation of growth response to water stress in the soybean primary root. I. Proteomic analysis reveals region-specific regulation of phenylpropanoid metabolism and control of free iron in the elongation zone. Plant Cell Environ. 33 , 223-243. Yamauchi Y., Hasegawa A., Taninaka A. et al. (2011). NADPH-dependent reductases involved in the detoxification of reactive carbonyls in plants. J. Biol. Chem. 286 , 6999-7009. Yang Y.H. & Speed T. (2002). Design issues for cDNA microarray experiments. Nat. Rev. Genet. 3, 579-588. Yano M. & Tuberosa R. (2009). Genome studies and molecular genetics - from sequence to crops: genomics comes of age. Curr. Opin. Plant Biol. 12 , 103-106. Yazdanbakhsh N. & Fisahn J. (2009). High throughput phenotyping of root growth dynamics, lateral root formation, root architecture and root hair development enabled by PlaRoM. Funct. Plant Biol. 36 , 938-946. Yin P., Fan H., Hao Q. et al. (2009). Structural insights into the mechanism of abscisic acid signaling by PYL proteins. Nat. Struct. Mol. Biol. 16 , 1230-1236. Yoshida R., Hobo T., Ichimura K. et al. (2002). ABA-activated SnRK2 protein kinase is required for dehydration stress signaling in Arabidopsis . Plant Cell Physiol. 43 , 1473- 1483. Zapata C., Magné C., Deléens E. et al. (2001). Grapevine culture in trenches: root growth and dry matter partitioning. Aust. J. Grape Wine Res. 7, 127-131. Zdobnov E.M. & Apweiler R. (2001). InterProScan – an integration platform for the signature-recognition methods in InterPro. Bioinformatics 17 , 847-848. Zelinka I. (2004). SOMA - self-organizing migrating algorithm. In: New optimization techniques in engineering (eds G.C. Onwubolu & B.V. Babu), pp. 167-217. Springer, Berlin Heidelberg, Germany. Zerihun A. & Treeby M.T. (2002). Biomass distribution and nitrate assimilation in response to N supply for Vitis vinifera L. cv. Cabernet Sauvignon on five Vitis rootstock genotypes. Aust. J. Grape Wine Res. 8, 157-162. Zhang B. & Horvath S. (2005). A general framework for weighted gene co-expression network analysis. Stat. Appl. Genet. Molec. 4, Article 17, 37 p. Zhang W., Qin C., Zhao J. & Wang X. (2004). Phospholipase D 1-derived phosphatidic acid interacts with ABI1 phosphatase 2C and regulates abscisic acid signaling. Proc. Natl. Acad. Sci. 101 , 9508-9513.

- 175 - References

Zhang X., Zwiers F.W., Hegerl G.C. et al. (2007). Detection of human influence on twentieth-century precipitation trends. Nature 448 , 461-465. Zhang Y., Primavesi L.F., Jhurreea D. et al. (2009). Inhibition of SNF1-related protein kinase1 activity and regulation of metabolic pathways by trehalose-6-phosphate. Plant Physiol. 149 , 1860-1871. Zhu J., Alvarez S., Marsh E.L. et al. (2007a). Cell wall proteome in the maize primary root elongation zone. II. Region-specific changes in water soluble and lightly ionically bound proteins under water deficit. Plant Physiol. 145 , 1533-1548. Zhu J., Ingram P.A., Benfey P.N. & Elich T. (2011). From lab to field, new approaches to phenotyping root system architecture. Curr. Opin. Plant Biol. 14 , 310-317. Zhu S.-Y., Yu X.-C., Wang X.-J. et al. (2007b). Two calcium-dependent protein kinases, CPK4 and CPK11, regulate abscisic acid signal transduction in Arabidopsis . Plant Cell 19 , 3019-3036. Zimmermann M.H. & Jeje A.A. (1981). Vessel-length distribution in stems of some American woody plants. Can. J. Bot. 59 , 1882-1892. Ziska L.H. & Bunce J.A. (2007). Plant responses to rising atmospheric carbon dioxide. In: Plant Growth and Climate Change (eds J.I.L. Morison & M.D. Morecroft), pp. 17-47. Blackwell Publishing Ltd, Oxford, UK. Zufferey V., Cochard H., Ameglio T. et al. (2011). Diurnal cycles of embolism formation and repair in petioles of grapevine ( Vitis vinifera cv. Chasselas). J. Exp. Bot. 62 , 3885- 3894. Zwieniecki M.A., Melcher P.J. & Holbrook N.M. (2001). Hydraulic properties of individual xylem vessels of Fraxinus americana . J. Exp. Bot. 52 , 257-264.

- 176 -