The Role of Calcium in the Cell

Wall of Grape Berries

By Bradleigh James Hocking

A thesis submitted in fulfilment of the requirements for the

degree of Doctor of Philosophy

School of Agriculture, Food & Wine

Faculty of Science

The University of Adelaide

September 2015

ii

Table of contents

Abstract…………………………………………………………………………………………………………….i

Declaration….. ...................................................................................................................................... iii

Acknowledgements .............................................................................................................................. iv

List of abbreviations .............................................................................................................................. v

Chapter 1 Apoplastic calcium influences hormonal signalling, fruit water status and cell wall

composition during fruit ripening ........................................................................................................ 1

1.1 Introduction ............................................................................................................................ 1

1.2 Plant calcium uptake, delivery and storage ......................................................................... 2

1.3 Calcium and water relations in fruit...................................................................................... 7

1.4 Calcium-cell wall interactions during fruit development .................................................. 11

1.5 Calcium-hormone interactions during fruit development................................................. 18

1.6 Conclusion ............................................................................................................................ 21

Chapter 2 Varietal differences in berry physical properties, nutrient accumulation and pectin

distribution…. ...................................................................................................................................... 23

2.1 Introduction .......................................................................................................................... 23

2.2 Materials and Methods ......................................................................................................... 23

2.2.1 Bunch and berry sampling ..................................................................................................... 23

2.2.2 Microscopy ............................................................................................................................ 24

2.2.3 Histological staining ............................................................................................................... 25

2.2.4 Immmunofluorescent staining ................................................................................................ 25

2.2.5 Berry physical testing ............................................................................................................. 26

2.2.6 Apoplast fluid centrifuge method............................................................................................ 27

2.2.7 Nutrient analysis; low volume apoplast method ..................................................................... 28

2.2.8 Total soluble sugars, pH and titratable acidity methods ......................................................... 29

2.2.9 Apoplastic ion selective electrode (ISE), pH and calcium activity measurements .................. 29

2.2.10 Principal component analysis ................................................................................................ 30

2.3 Results .................................................................................................................................. 30

2.3.1 Berry histological staining ...................................................................................................... 30

2.3.2 Berry immunofluorescent staining .......................................................................................... 31

2.3.3 Berry physical properties ....................................................................................................... 31

2.3.4 Berry tissue nutrient composition ........................................................................................... 35

2.3.5 Principal component analysis ................................................................................................ 39

2.4 Discussion ............................................................................................................................ 41

2.4.1 Berry phenolics ...................................................................................................................... 41

2.4.2 Berry cell morphology ............................................................................................................ 41

2.4.3 Berry pectin distribution ......................................................................................................... 42

2.4.4 Berry physical changes .......................................................................................................... 44

2.4.5 Berry apoplast interactions .................................................................................................... 44

2.5 Conclusion ............................................................................................................................ 46

Chapter 3 Grapevine calcium nutrition ........................................................................................... 48

iii

3.1 Introduction .......................................................................................................................... 48

3.2 Materials and Methods ......................................................................................................... 48

3.2.1 Fruiting cuttings; Mullins method modifications ...................................................................... 48

3.2.2 Hydroponics system design ................................................................................................... 49

3.2.3 Biomechanical testing ............................................................................................................ 50

3.2.4 Leaf water potential ............................................................................................................... 50

3.2.5 Gas exchange measurements ............................................................................................... 50

3.2.6 Root hydraulic conductance .................................................................................................. 51

3.2.7 ICP-AES analysis .................................................................................................................. 51

3.2.8 Bunch reproductive measures ............................................................................................... 51

3.3 Results .................................................................................................................................. 52

3.3.1 Effect of calcium treatments upon calcium uptake ................................................................. 52

3.3.2 Shiraz berry development responses to modified calcium nutrition ....................................... 55

3.3.3 Berry class diversity ............................................................................................................... 62

3.3.4 Vine physiology...................................................................................................................... 63

3.3.5 Chenin Blanc nutrient uptake from veraison to late harvest ................................................... 65

3.4 Discussion ............................................................................................................................ 70

3.4.1 Calcium deficiency accelerates berry ripening ....................................................................... 70

3.4.2 Elevated calcium impairs berry development......................................................................... 72

3.4.3 Calcium impact on vine fruitset and reproductive indices ...................................................... 73

3.4.4 Chenin Blanc nutrient accumulation consistent with other varieties ....................................... 74

3.5 Conclusion ............................................................................................................................ 74

Chapter 4 Berry ripening physiology; calcium uptake and cell wall modification ...................... 76

4.1 Introduction .......................................................................................................................... 76

4.2 Materials and Methods ......................................................................................................... 76

4.2.1 Grapevine material ................................................................................................................ 76

4.2.2 Immunogold labelling and transmission electron microscopy ................................................ 76

4.2.3 Berry staining methods; FDA and PI ...................................................................................... 77

4.2.4 MATLAB/ImageJ image analysis ........................................................................................... 78

4.2.5 Berry electrical impedance spectroscopy .............................................................................. 79

4.2.6 Berry rehydration assay ......................................................................................................... 79

4.2.7 Biomechanical testing ............................................................................................................ 80

4.2.8 Botrytis wounding assay ........................................................................................................ 80

4.3 Results .................................................................................................................................. 81

4.3.1 Cell wall modifications from veraison to late harvest ............................................................. 81

4.3.2 Fresh berry staining ............................................................................................................... 86

4.3.3 Berry electrical impedance spectroscopy .............................................................................. 89

4.3.4 Berry physical changes .......................................................................................................... 91

4.3.5 Berry water relations .............................................................................................................. 93

4.3.6 Botrytis susceptibility ............................................................................................................. 94

4.4 Discussion ............................................................................................................................ 95

4.4.1 Effect of calcium on pectin distribution ................................................................................... 95

4.4.2 Effect of calcium on berry softening ....................................................................................... 98

iv

4.4.3 Effect of calcium on berry water relations .............................................................................. 994.4.4 Effect of calcium on cell vitality ............................................................................................ 1004.4.5 Effect of calcium on Botrytis susceptibility ........................................................................... 1034.5 Conclusion.......................................................................................................................... 104

Chapter 5 General Discussion and Conclusion............................................................................ 1055.1 Grapevine calcium nutrition .............................................................................................. 1055.2 Berry calcium physiology.................................................................................................. 1075.3 Implications of calcium nutrition for fruit quality ............................................................ 1075.4 Future perspectives and Conclusion ............................................................................... 110

Appendices ........................................................................................................................................ 112Appendix 1 .....................................................................................................................................

.....................................................................................................................................

.....................................................................................................................................112

Appendix 2 113Appendix 3 114Appendix 4 ..................................................................................................................................... 123Appendix 5 ..................................................................................................................................... 125Appendix 6 ..................................................................................................................................... 129Appendix 7 ..................................................................................................................................... 133Appendix 8 ..................................................................................................................................... 134

Reference list ..................................................................................................................................... 135

i

Abstract

Calcium has defined roles in plant signalling, water relations and cell wall interactions. Calcium nutrition

impacts fruit quality by facilitating developmental and stress response signalling, stabilising

membranes, and modifying cell wall properties through cross-linking of de-esterified pectins. The

importance of calcium in fruit development and ripening is reviewed, experimental work probing the

relationship between calcium nutrition and fruit development in grape berries is undertaken.

Relationships between calcium uptake and pectin modification were investigated in a survey of red,

white, and table grape varieties collected from two sites varying in calcium levels. Grapes harvested at

the Barossa site showed higher calcium concentrations within apoplastic fluid, skin and mesocarp

tissues than those from Waite. Chenin Blanc had higher apoplastic calcium content than other varieties.

Fluorescent immuno-labelling revealed de-esterified pectin localisation in the middle lamella of all

varieties with punctillate staining patterns observed in Grenache and Thompson Seedless. A negative

correlation between apoplastic pH and apoplastic calcium concentration was observed. Shiraz was the

only variety to demonstrate any significant difference between sites in apoplastic pH and apoplastic

calcium activity.

Effects of low and high calcium supply in grapevines were investigated. Low calcium grown Shiraz

showed early berry softening and onset of berry weight loss. High calcium grown Shiraz showed

delayed and asynchronous fruit development. Berry hydration assays indicated that early onset of berry

weight loss in low calcium grown berries was a result of higher post-veraison berry transpiration. High

calcium grown berries demonstrated lower berry water uptake rate pre-veraison, and lower berry

transpiration rates throughout development. Whole vine physiology was assessed in Chenin Blanc;

high calcium grown vines demonstrated reduced transpiration and net assimilation rates compared to

basal and low calcium grown vines.

An image analysis macro was developed for quantification of cell vitality (with fluorescein diacetate;

FDA) and pectin de-esterification (with propidium iodide; PI) staining patterns. Chenin Blanc maintained

higher PI staining in skin tissue than Shiraz throughout development; higher magnification imaging

revealed this staining to be localised to the epidermis and peripheral vasculature of Chenin Blanc

berries.

ii

Transmission electron microscopy demonstrated cuticle localisation of de-esterified pectin in Chenin

Blanc and Shiraz berries, particularly of low calcium grown berries; low levels of calcium-pectin cross-

linkages and high rates of berry transpiration result in increased movement of de-esterified pectin from

the epidermis into the cuticle. Shiraz cuticle de-esterified pectin levels increased throughout

development, indicating pectin solubilisation. Chenin Blanc showed strong de-esterified pectin labelling

in epidermal and hypodermal cell walls, consistent with patterns visualised using PI staining. Low

calcium grown Chenin Blanc berries showed a higher Botrytis infection rate than basal or high calcium

grown berries.

Differences in calcium accumulation and pectin modification contribute to varietal diversity in ripening

physiology. Berries supplied with low calcium are early softening and susceptible to shrivel and Botrytis

infection, whereas high calcium supply results in changes in vine physiology, including delayed and

asynchronous berry development.

iii

Declaration

I certify that this work contains no material which has been accepted for the award of any other degree

or diploma in my name in any university or other tertiary institution and, to the best of my knowledge

and belief, contains no material previously published or written by another person, except where due

reference has been made in the text. In addition, I certify that no part of this work will, in the future, be

used in a submission in my name for any other degree or diploma in any university or tertiary institution

without the prior approval of the University of Adelaide and where applicable, any partner institution

responsible for the joint award of this degree.

I give consent to this copy of my thesis, when deposited in the University Library, being made available

for loan and photocopying, subject to the provisions of the Copyright Act 1968.

The author acknowledges that copyright of published works within this thesis resides with the copyright

holder(s) of those works.

I also give permission for the digital version to be made available on the web, via the University’s digital

research repository, the Library Search and also through web search engines, unless permission has

been granted by the University to restrict access for a period of time.

…………………………….. …………….

Bradleigh James Hocking

iv

Acknowledgements

I thank my supervisors Associate Professor Matthew Gilliham, Associate Professor Rachel Burton, and

Professor Steve Tyerman, for their ongoing guidance and support throughout this project. Particular

thanks go to Matt for his persistence and encouragement. I greatly appreciate the time Matt has spent

reading and discussing this work.

I thank the University of Adelaide, Wine 2030, and the Farrer Memorial Trust, for their support of this

project in the form of resources and funding.

I thank my lab colleagues for sharing their enthusiasm and expertise. Particular thanks go to Wendy

Sullivan for experimental assistance and Johannes Scharwies for sharing his designs.

I thank the various others who have made significant contributions; Dr Cassandra Collins for her

viticultural advice and guidance in data analysis, Cameron Nowell for his assistance in image analysis,

Chris Fiebiger for kindly providing experimental material from his vineyard.

I thank my family for their unshakeable faith in me, and all of the things they have taught me. I thank

Asmini Athman for her wonderful support and encouragement, without which I would not have

completed this journey. I thank my friends for expressing their love of life in conversation and

celebration.

v

List of abbreviations

ABA Abscisic acid AGJ Artificial grape juice ANOVA Analysis of variance apo apoplast Ara Arabinan BNS Basal nutrient solution Cel Cellulose Ca2+ Calcium ion CEC Cation exchange capacity CI Coulure Index Cm Membrane capacitance cyt cytosol DAA Days after anthesis DW Dry weight FDA Fluorescein di-acetate GA Gibberellic acid Gal Galactan HCS High calcium solution HG Homogalacturonan IAA Indole acetic acid ICP-AES Inductively coupled plasma atomic emission spectroscopy IRGA Infra-red gas analyser ISE Ion selective electrode LCS Low calcium solution LGO Live green ovary MI Millerandage Index OGA Oligogalacturonide PCA Principal component analysis PG Polygalacturonase PI Propidium iodide PID PINOID PIN PIN-FORMED PM Plasma membrane PME Pectin methyl-esterase PMEI Pectin methyl-esterase inhibitor QTL Quantitative trait loci Re Extracellular resistance Rh Hydraulic resistance Ri Intracellular resistance RT Xylem hydraulic resistance RG-I Rhamnogalacturonan-I RG-II Rhamnogalacturonan-II SD Standard deviation TEM Transmission electron microscopy WAK Wall associated kinase XG Xyloglucan

1

Chapter 1 Apoplastic calcium influences hormonal signalling, fruit water status

and cell wall composition during fruit ripening

1.1 Introduction

Calcium has defined roles in signalling, water relations and cell wall interactions. Although significant

research into these individual roles in various tissues has been carried out, the holistic role that calcium

plays in fruit ripening is not well defined. This review will provide a brief overview of many of the

fundamental roles of calcium in plant physiology, working towards an improved understanding of how

these processes can interact to influence fruit ripening, and proposes future research directions to fill

our knowledge gaps.

Calcium is a phloem immobile element that accumulates in fruit via the xylem (White, 2012). During

ripening stages, xylem influx into most fruit is minimal (Lang, 1990; McCarthy and Coombe, 1999),

resulting in potential for localised calcium deficiencies. Therefore, the partitioning of calcium within fruit

is important for normal fruit development. How hydraulic conductance of fruit vascular components

(Drazeta et al., 2004; Castellarin et al., 2011; Mazzeo et al., 2013) and turgor pressure of tissues

(Thomas et al., 2008; Castellarin et al., 2011) can change throughout development due to solute

unloading, cellular expansion, and cell wall modification are important factors for determining the

distribution of calcium within fruit will be specifically discussed.

Calcium is a vital secondary messenger in a vast array of plant signalling pathways (Dodd et al., 2010).

Calcium is known to interact with gibberellic acid, auxin, and abscisic acid to regulate fruitset, initiation

of ripening, cell division, cell expansion, and fruit softening (Ferguson, 1984; Saure, 2005; Yu et al.,

2006). Additionally, hormonal regulation of cell expansion, cell wall modification, xylem development,

and sugar unloading from the phloem can affect calcium distribution within the fruit (Saure, 2005; de

Freitas et al., 2014).

The role of the cell wall in calcium signalling and fruit ripening is highlighted. The cell wall is a dynamic

structure that responds to both developmental and environmental stimuli by structural remodelling;

different cell types exhibit different cell wall compositions that perform specific roles within the plant

(Hoson, 1998). Cell wall modification can be triggered by a wide range of environmental stimuli and

2

stresses, including pathogen attack, light, and touch (Hoson, 1998; Seifert and Blaukopf, 2010). Cell

expansion is tightly controlled by selective secretion of components into the cell wall matrix and

subsequent modifications; regulation of cell turgor pressure is also postulated to be an important factor

in such growth (Cosgrove, 2005). The role of cytosolic calcium in signal transduction and cell wall

polysaccharide interactions indicates that it is a key factor in cell wall growth and remodelling

processes. As an important pool of calcium for cytosolic signalling, apoplastic calcium will not only have

an impact on the regulation of cell wall assembly, but also a direct impact through physical interaction

with the cell wall.

1.2 Plant calcium uptake, delivery and storage

Calcium plays two important roles in plants; as a secondary messenger in a myriad of signalling

networks and as a structural element in cell walls and membranes (White and Broadley, 2003). Calcium

ions (Ca2+) are taken up as a soil nutrient and transported primarily through the apoplast and xylem to

aerial tissues (Clarkson, 1984; White, 2001). Calcium demonstrates low phloem mobility; calcium

accumulation in aerial sink tissues such as fruit occurs primarily through the xylem (Rogiers et al., 2000;

Drazeta et al., 2004). Symplastic transport mechanisms for calcium and water (including transpiration)

also contribute to calcium accumulation in a species dependent manner (Gilliham et al., 2011).

The Casparian strip is a zone of suberin and/or lignin-enriched cell wall in the root endodermis that

forms a physical barrier to apoplastic movement of water and solutes into the transpiration stream

(Nawrath et al., 2013). An Arabidopsis (Arabidopsis thaliana) mutant esb1 that lacks expression of a

dirigent scaffold protein required for correct patterning of lignin deposition in the Casparian strip also

demonstrates ectopic deposition of suberin lamellae between the cell walls and plasma membrane of

root endodermal passage cells. This results in decreased transpiration and Ca2+ accumulation in the

leaf, presumably through a reduction in the level of water and Ca2+ movement into the root xylem

(Baxter et al., 2009; Hosmani et al., 2013). Another Arabidopsis mutant (sgn3) lacks a functional

Casparian strip due to knockout of a receptor-like kinase important for localising transmembrane

proteins along the endodermal Casparian strip domain (Pfister et al., 2014). Interestingly, the apoplastic

bypass route provided by the dysfunctional Casparian strip in sgn3 led to an increase in magnesium

accumulation (possibly due to increased uptake through the apoplastic pathway) and a decrease in

potassium accumulation (possibly due to loss of potassium from root stelar tissues); however

3

homeostasis of calcium appeared to be maintained. The authors hypothesized that other barriers to

calcium movement within the apoplast (e.g. interactions with cell walls) may be responsible for this

observation (Pfister et al., 2014).

Studies of radial ion transport using radioactive Ca2+ in onion (Allium cepa) roots showed no detectable

Ca2+ translocation through the root tips (which lack mature Casparian strips, suberin lamellae and

mature xylem vessels), low calcium translocation (0.87 nmol Ca2+ per mm root length) through the old

root zone containing both mature Casparian strips and suberin lamellae, and high calcium translocation

(2.67 nmol Ca2+ per mm root length) through the young root zone containing Casparian strips but no

suberin lamellae (Cholewa and Peterson, 2004). This supports the idea that Ca2+ primarily moves from

the extracellular space of the outer root into the stele via a symplastic route, utilising Ca2+ channels and

Ca2+-ATPase pumps to bypass the Casparian strip, but in some root zones becomes limited by the

presence of suberin lamellae (Cholewa and Peterson, 2004; Hayter and Peterson, 2004). Perturbing

the normal pattern of suberin deposition (e.g. in the esb1 and sgn3 mutants) can therefore affect plant

nutrient accumulation profiles. Compartmentation and hydraulic resistance in the apoplast also occurs

in many other tissues. Examples of this include; formation of pectin hydrogels in xylem conduit pit

membranes (Zwieniecki et al., 2001; Plavcova and Hacke, 2011; van Doorn et al., 2011), separation of

the leaf xylem from the leaf apoplast by bundle sheath cells, and separation of external surfaces of the

plant and the underlying apoplast by the cuticle (Nawrath et al., 2013). This effective compartmentation

of the apoplastic space highlights the importance of understanding physical transport barriers as well as

cellular transport mechanisms for controlling Ca2+ movement and utilisation.

The delivery of calcium to leaves is dependent on a number of interacting factors including leaf

hydraulic conductivity, transpiration rate, calcium transporter activity, calcium tissue compartmentation,

xylem architecture, and cell wall composition (Sattelmacher, 2001; Saure, 2005; Gilliham et al., 2011).

Cation exchange within the xylem plays an important role in calcium delivery. The cation exchange

capacity (CEC) for calcium of cell walls from different root and shoot tissues of Picea abies was

measured using transmission electron microscopy energy-dispersive microanalysis (Fritz, 2007). Wide

variation between root and shoot tissue CEC was observed, however CEC of the secondary cell wall of

xylem tracheids was consistently low ( ~24 meq/kg wall material) (Fritz, 2007), suggesting that the

composition of other zones within the xylem (e.g. pit membranes) may be more important for

determining calcium transport and buffering fluctuations in xylem calcium concentration. Factors

contributing to patterns of calcium delivery and distribution in aerial tissues include; rate of mass flow,

4

competition between ions for xylem vessel cell wall binding sites (including H+, making pH an important

factor), formation of lowly-soluble or in-soluble complexes (e.g. calcium oxalate) and water transport in

and out of the xylem (Franceschi and Nakata, 2005; Saure, 2005; Gilliham et al., 2011). It is important

to understand the relative contributions of these factors to the calcium nutrition of individual tissues.

The impact of transpiration upon delivery and the lack of phloem mobility ensure that calcium generally

accumulates to high concentrations in the leaf tissues, often in a cell-type specific manner (Conn and

Gilliham, 2010; Dayod et al., 2010; Gilliham et al., 2011). Monocots and dicots exhibit different cell-type

specific patterns of calcium and potassium accumulation in the leaf (Karley et al., 2000; Conn and

Gilliham, 2010). In leaves, it appears that plants of the order Poales (including grasses and the

economically important cereals) accumulate calcium primarily in the epidermal cells, whereas other

monocots and dicots accumulate calcium primarily in the mesophyll cells (Karley et al., 2000). This

profile has been observed to hold true in a large number of different species (R. Leigh, Pers. Comm.).

Calcium accumulation in the trichomes of some dicots has been observed, possibly as a way to

maintain the apoplastic calcium concentration surrounding guard cells in order to avoid inducing

stomatal closure (De Silva et al., 1996).

Regulated expression of a number of membrane ion transporters is responsible for mesophyll calcium

accumulation in Arabidopsis (Conn and Gilliham, 2010). Knockout of the vacuolar Ca2+/H+ antiporters

AtCAX1 and AtCAX3 resulted in lower mesophyll Ca2+ sequestration and higher apoplastic Ca2+, with

physiological impacts ranging from reduced stomatal aperture, stomatal conductance and CO2

assimilation to reduced cell wall extensibility and leaf growth rate (Conn et al., 2011). Calcium also

accumulates primarily in the mesophyll of Brassica rapa (Rios et al., 2012). Differences in leaf calcium

accumulation between tip-burn susceptible and resistant lines of Brassica oleracea were correlated with

expression changes in calcium transporters BoCAX1, BoACA4, and BoACA11 (Lee et al., 2013).

Expression QTL analysis of a mapping population of Brassica rapa under altered calcium supply

identified an ortholog of AtCAX1 (BraA.CAX1a). Lines carrying missense mutations in BraA.CAX1a

demonstrated changes in shoot calcium accumulation, indicating potential for the use of this gene as a

tool for the bio-fortification of calcium in Brassica crops (Graham et al., 2014). The importance of these

and other related calcium transport mechanisms in other tissues and species require further

investigation.

Regulation of calcium transport across the plasma membrane and organelle membranes is tightly

controlled by the expression pattern, interaction and post-transcriptional control of many Ca2+

5

transporters (Kudla et al., 2010). The major calcium transport mechanisms are represented in Figure 1.

Tight control of calcium transport is necessary in order to keep cytosolic calcium ([Ca2+]cyt) within the

range (~0.1-10 µM) required for signal transduction (Evans et al., 1991; White, 2000; Dodd et al.,

2010). Regulated fluctuations in [Ca2+]cyt form the “calcium signature” which is a major determining

factor in the specificity of downstream transcriptional and physiological responses (McAinsh and

Pittman, 2009; Dodd et al., 2010; Kudla et al., 2010). Different environmental stimuli create specific

calcium signatures (Dodd et al., 2010). Electrically-induced calcium transients with different amplitudes

and frequencies were shown to induce distinct patterns of gene expression (Whalley and Knight, 2013),

indicating that environmental stimuli can translate into specific expression profile changes in calcium

signalling components. There is considerable evidence indicating that Ca2+ signalling transients also

occur in compartments other than the cytosol, e.g. the nucleus, chloroplasts and the apoplast (Johnson

et al., 1995; Johnson et al., 2006; Tang et al., 2007; McAinsh and Pittman, 2009). However, these

mechanisms are not nearly as well characterised as the cytosolic pathways.

Figure 1. Major mechanisms involved in apoplastic and sub-cellular calcium partitioning. Reproduced

from (de Freitas and Mitcham, 2012).

Changes in apoplastic calcium can have impacts on the duration and amplitude of [Ca2+]cyt oscillations,

and ABA-dependent stomatal closure pathways, thereby affecting the dynamics of stomatal movements

(Allen et al., 2001; Webb et al., 2001). High apoplastic calcium ([Ca2+]apo) is known to close stomata in

6

vitro (Webb et al., 2001). It has also been shown that in planta manipulation of free [Ca2+]apo can affect

CO2 assimilation and transpiration rate (Conn et al., 2011). Components of an extracellular calcium

sensing pathway have been described; plastid localised CAS protein mediates stomatal closure in

response to changes in extracellular calcium (Han et al., 2003; Wang et al., 2012). The importance of

the extracellular calcium signalling pathway in optimising photosynthesis and water use has recently

been demonstrated; antisense CAS lines show reduced water use efficiency and photosynthetic

electron transport rate, due to reduced control of stomatal aperture and transcription of electron

transport components (Wang et al., 2014). Therefore, the supply of apoplastic calcium to fruit is likely to

have an impact on the signalling capacity and water relations of the fruit.

Both calcium deficiency and toxicity can affect the productivity of horticultural systems, and the quality

of the products. Common examples of the effects of calcium deficiency include; tipburn in leafy tissues,

blossom end rot in tomatoes, bitter pit in apples, and increased susceptibility to fruit cracking (White

and Broadley, 2003). These calcium deficiency disorders are demonstrated in Figure 2. Calcium

deficiency occurs due to an insufficient mobilisation of calcium from internal stores or a reduced supply

of calcium through the xylem due to low transpiration rates (White and Broadley, 2003). Calcium toxicity

occurs due to high concentrations of available calcium in the soil solution and can result in reduced

germination and growth rates, and the ectopic deposition of calcium oxalate crystals (White and

Broadley, 2003). Understanding of the mechanisms of plant resistance or susceptibility behind calcium

toxicity and deficiency are limited at this point.

7

Figure 2. Calcium disorders in horticultural crops: (a) cracking in tomato fruit; (b) tipburn in lettuce; (c)

calcium deficiency in celery; (d) blossom end rot in immature tomato fruit; (e) bitter pit in apples; (f) gold

spot in tomato fruit with calcium oxalate crystals (inset). Reproduced from White and Broadley (2003).

1.3 Calcium and water relations in fruit

The link between water and calcium transport becomes particularly apparent when looking at sink

tissues with relatively low transpiration rates, such as fruit. During later stages of fruit ripening in most

species, delivery of water, sugar, and basic nutritional inputs occurs largely via phloem influxes

(Drazeta et al., 2004; Rogiers et al., 2006b; Choat et al., 2009). Kiwifruit appears to be somewhat

different, with both phloem and xylem contributing to fruit hydration during late development; their

relative contributions are affected by environmental conditions (Clearwater et al., 2012). The low

phloem mobility of calcium can create a situation that leads to localised calcium deficiencies in fruit.

Examples of this include blossom end rot in tomatoes and bitter pit in apples. Calcium concentration

can have an impact upon water transport processes; aquaporins have been shown to be highly

sensitive to cytosolic calcium concentration (Gilliham et al., 2011). In order to address fruit calcium

nutrition disorders it is important to understand the physical and molecular pathways of water and

8

calcium delivery, and the impact of calcium signalling and cell wall interactions on transpiration and

water transport.

Fruit are strong sink organs; as such vascular flows into fruit can be substantial. Accumulation of ionic

nutrients into fruit is determined not only by water import rates, but also by their relative prevalence and

mobility in the phloem and xylem. Potassium and calcium are examples of two major ions that exhibit

very different pathways of vascular movement. In grapes, potassium is both xylem and phloem-mobile,

with concentrations in the phloem up to ten times that found in the xylem (Hocking, 1980). Grape berry

potassium accumulation occurs throughout berry development, reaching a maximum uptake rate during

early post-veraison with uptake continuing throughout ripening (Rogiers et al., 2006b). In contrast,

calcium is observed to be phloem-immobile and berry calcium content generally does not increase after

veraison (Rogiers et al., 2006a). The reduction in calcium import is a result of the large drop in xylem

hydraulic conductance into the berry post-veraison which is correlated with loss of cell vitality and berry

shrivel (Tyerman et al., 2004; Rogiers et al., 2006b; Tilbrook and Tyerman, 2009). A varietal survey

revealed genotype differences in the occurrence of cell vitality loss and berry shrivel in mature grapes

(Fuentes et al., 2010). The shift away from xylem water delivery during ripening also effectively buffers

the fruit against fluctuations in plant water status and water stress events that may affect the plant

during the ripening season (Thomas et al., 2006; Choat et al., 2009). Determining the hydraulic

pathways of ionic delivery to fruit is vital for understanding patterns of distribution and accumulation that

are important determinants of fruit quality.

Characterisation of changes in apoplastic and vacuolar solute composition supports the notion of a

switch from symplastic to apoplastic unloading of phloem solutes being linked to the loss of mesocarp

membrane integrity observed in some varieties during late ripening (Keller and Shrestha, 2014). The

table grape variety Concord maintains high apoplastic pH (relative to vacuolar pH) late into ripening; in

shrivel susceptible variety Merlot the pH difference between these compartments is reduced to zero

during late ripening, suggesting a loss of membrane viability in this variety (Keller and Shrestha, 2014).

The switch to apoplastic phloem unloading enables accumulation of high sugar levels in ripening fruit;

Merlot demonstrates a dramatic jump in apoplastic glucose and fructose concentrations during the

transition from red to ripe berries (Keller and Shrestha, 2014). The accumulation of sugars in the

apoplast activates cell wall localised invertases and hexose/proton transport pathways in berries

(Hayes et al., 2007). Loss of cell turgor, vitality, and membrane integrity in the locular tissues during

ripening may be related to the apoplastic unloading of solutes from the adjacent central vasculature

9

(Tyerman et al., 2004; Krasnow et al., 2008). However, berry cell death is normally observed after the

transition to apoplastic unloading. Additionally, cell membrane capacitance in the berry is maintained

through the cell death phase, indicating intact membranes (Caravia Bayer et al., 2015). This suggests

that rather than ‘cell death’, the loss of cell vitality often observed may actually represent an increase in

cell membrane porosity, with loss of membrane selectivity allowing distribution of some solutes (e.g.

sugars, ions, and perhaps the cell vitality stain fluorescein diacetate) into the apoplast. How the

accumulation of solutes in the apoplast and associated changes in cell turgor affect fruit water relations

requires further investigation.

Changes in the hydraulic resistance (Rh) of components of the bunch and berry vascular architecture

may account for some of the observed varietal differences in susceptibility to berry shrivel. By studying

the Rh of each component the contributions of particular structural and ionic variables to observed

changes in xylem flows may be identified (Tyerman et al., 2004; Choat et al., 2009; Mazzeo et al.,

2013) J. Scharwies, Pers. Comm.). These variables may include; physical barriers (i.e. pit membrane

porosity and xylem vessel diameter), structural changes (i.e. formation of pectin gels within the xylem),

cellular water permeability (i.e. through changes in temporal and spatial expression of aquaporins), and

ionic factors (i.e. solute driven changes in the osmotic potential of the cell wall).

In grapes, a decrease in mesocarp turgor coincides with the onset of veraison. However, after veraison

berry mesocarp turgor does not appear to respond to vine water deficit (Thomas et al., 2006). This

supports the notion that a switch to phloem water delivery reduces the link between water status of the

plant and fruit water relations, and that this relationship is dynamic through different stages of

development. When pre-veraison berries were physically boxed to restrict veraison associated cell

expansion, both sugar accumulation and the drop in mesocarp turgor pressure were delayed; when the

box was ventilated to allow transpiration, delayed sugar accumulation was not observed and the turgor

drop delay effect was moderated (Matthews et al., 2009). This suggests that fruit transpiration is

required to assist in phloem sugar loading into fruit (by removing excess water) (Lang and Thorpe,

1989), and that ripening related changes in mesocarp cell turgor pressures are linked to both rapid cell

expansion and sugar accumulation (Matthews et al., 2009). This function may act a surrogate for the

xylem water delivery that occurs in other plant tissues and earlier in fruit development. Calcium is

involved in the regulation of cell expansion and elongation during pollen tube growth through dynamic

pectin binding (Jiang et al., 2005; Rounds et al., 2011), signalling proteins and ion channels (Konrad et

al., 2011). This suggests that calcium may also be involved in the changes in cell turgor pressure and

10

cell expansion observed during the progression of fruit ripening. The relative contributions of turgor and

cell wall changes to fruit softening are still a major point of discussion; however it is clear that both

factors contribute through complex interactive pathways and feedback mechanisms to the onset and

development of ripening processes in fruit.

Fruit water relations and ripening-linked shifts in fruit hydraulic conductance vary between species.

Kiwifruit maintains positive water fluxes from both the phloem and xylem into the fruit throughout

development, with each pathway contributing approximately equally to the water balance (Clearwater et

al., 2012). However, when grown in high vapour pressure deficit conditions Actinidia chinensis var.

chinensis ‘Hort16A’ exhibits late ripening shrivel, similar to the phenomenon observed in Shiraz grapes.

This may be attributable to the high surface conductance and transpiration rate observed in this variety,

causing an imbalance between water delivery to the fruit and transpiration losses (Clearwater et al.,

2012). Additionally, kiwifruit does not accumulate sugars until late in the ripening phase; this difference

may explain its ability to maintain xylem flow from the plant into the fruit throughout development.

A study of kiwifruit xylem hydraulic resistance (RT) throughout development using pressure chamber

and flow meter techniques showed a general increase in RT during the second half of fruit development,

consistent with previous reports in grapevine and kiwifruit (Tyerman et al., 2004; Choat et al., 2009;

Mazzeo et al., 2013). However, the increase in RT began prior to ripening, indicating that decreasing

xylem inflows in kiwifruit may be attributable to increasing xylem hydraulic resistance (Mazzeo et al.,

2013). This contrasts to observations in grape, where increases in RT were observed after veraison,

therefore not accounting for drops in xylem flow rates into the berry (Tyerman et al., 2004; Choat et al.,

2009). The parallel use of pressure chamber and flow meter techniques (Mazzeo et al., 2013), and an

evaporative flux method (Clearwater et al., 2012), showed differences in the magnitude of resistance

measured depending on the methodology employed; the flow meter technique may underestimate

xylem resistance and subsequent calculations regarding berry hydraulic isolation and potential for

xylem backflow may be imprecise (Mazzeo et al., 2013). Despite difficulties in accurately and

consistently measuring hydraulic resistance, it is clear that differences in xylem composition can

contribute to fruit water relations.

Changes in cuticle composition can affect fruit mechanical properties and transpiration rate. Fruit

cuticles are typically thicker (but also more water permeable) than leaf cuticles; the scarcity of stomata

on fruit also suggests that cuticle composition is important for fruit water relations (Martin and Rose,

2014). A tomato cultivar (‘Delayed Fruit Deterioration’) with altered cuticle architecture was shown to

11

have low fruit transpiration and increased cell turgor pressure, leading to delayed softening despite

undergoing normal ripening related cell wall modifications (Saladie et al., 2007). Application of

gibberellins was also shown to increase cuticle thickness in tomato (Knoche and Peschel, 2007). In

grapes (cv. Riesling), a drop in the transpiration permeability of the cuticle occurs from pre-veraison to

post-veraison (Becker and Knoche, 2011), and this drop is strongly correlated with increased cuticle

deposition (Becker and Knoche, 2012). The composition of the cuticle changes throughout

development in cherry tomato (cv. Cascada); cuticle mass per unit fruit surface area increased rapidly

from 10 days after anthesis to reach a maximum 15 days after anthesis (subsequent increases in

cuticle thickness were attributed to reduced cuticle density) (Dominguez et al., 2008). Interestingly,

another study in the same cultivar looking at the cuticle mechanical properties found a shift from elastic

to predominantly viscoelastic behaviour from 10 to 15 days after anthesis. These changes in the cuticle

mechanical properties were correlated with the ratio of cutin:polysaccharide present; high ratios were

associated with cell enlargement growth stages, and lower ratios (approaching 1:1) were associated

with stages where cell expansion is minimal (i.e. early cell division and later ripening phases) (Espana

et al., 2014). As such, it is hypothesised that polysaccharides in the cuticle contribute elastic properties,

and cutin confers viscoelastic properties. It is clear that cuticle composition is an important variable in

determining both fruit physical properties and transpiration water losses.

1.4 Calcium-cell wall interactions during fruit development

The cell wall is composed of a diverse array of complex polysaccharides. In dicots, the primary cell wall

consists of cellulose microfibrils bound in a matrix of pectins and hemicelluloses. Cellulose is extruded

through the plasma membrane by cellulose synthase complexes, whereas pectins and hemicelluloses

undergo primary synthesis within the Golgi apparatus, and are transported to the cell surface where

further synthesis and modification may take place (Gendre et al., 2013). The matrix polysaccharides

are very diverse in their composition, with a variety of sugar residues, linkages and side chains present;

their synthesis and modification is therefore accomplished by a large number of genes (Burton et al.,

2010). Fruit softening is often attributed to changes in the composition of the cell wall. Cell wall

modifying enzymes activated at different stages of development, and under certain conditions (e.g.

heat, pH changes in the apoplast), are responsible for modification and degradation of cell wall

polysaccharides (Grignon and Sentenac, 1991; Brummell, 2006).

12

The chemical changes that occur in fruit cell walls during development include; modification of pectin

side chains, depolymerisation of pectins, and degradation of xyloglucan (a hemicellulose). Together

with other ripening related processes (such as the accumulation of solutes); this leads to a number of

physical and textural changes in fruit that can help us to delineate different types of fruit by their

ripening mechanisms. Some species (e.g. strawberry and plum) exhibit cell wall swelling during

ripening which results in a soft textured fruit, whereas other species (e.g. watermelon and apple) do not

exhibit swelling and maintain crisp textured fruit (Redgwell et al., 1997). Physical changes in fruit cell

walls are associated with ongoing modification and solubilisation of pectins. The importance of pectin

structure for determining the hydraulic and elastic properties of pectin gels has been examined mostly

in vitro; understanding the complexity of these cell wall interactions in planta requires further research.

Fruit cell walls contain a high proportion of pectin, a complex family of polysaccharides that are

structurally related by the occurrence of (1,4)-α-linked galacturonan in the backbone (commonly as

homogalacturonan, or as the rhamnose/galacturonan disaccharide repeat rhamnogalacturonan-I)

(Mohnen, 2008). The galacturonan residues of the backbone may be methyl-esterified or acetylated;

homogalacturonan is secreted into the cell wall in an esterified form (Willats et al., 2001). A wide variety

of linear and branched side chains are also observed, forming the pectin structural classes

rhamnogalacturonan-II, xylogalacturonan and apiogalacturonan. Rhamnogalacturonan-II is the most

complex pectin, it can include up to 12 different sugar residues and more than 20 different linkage

types. These have been reviewed previously (Mohnen, 2008; Burton et al., 2010). The distribution of

various cell wall components within a model cell wall is represented in Figure 3. The structural

complexity of pectin, enabled by the expression of a range of pectin synthesising and modifying

enzymes throughout development, implicates pectin in an array of potential interactions and functional

roles.

The prevalence of ionic and ester bonds between adjacent pectins play an important role in the

physical properties of fruit cell walls. These bond interactions influence the solubility of pectins. Suitable

ions for cross-linking adjacent pectins include calcium forming junctions between de-esterified

homogalacturonans and boron forming di-ester bonds between rhamnogalacturonan II units.

Associations between adjacent homogalacturonans ionically linked by calcium ions have been

characterised as forming an “egg-box” structure. Although this structure has been demonstrated in

pectin extracts (Tibbits et al., 1998), the diversity of side chains and modifications within the pectic

13

polysaccharides as well as the complexity of other cell wall components makes such interactions

difficult to characterise in planta.

The majority of pectins occur in the middle lamella (outermost part of the extracellular matrix; where cell

junctions occur), with smaller amounts observed in the primary cell wall (Ferguson, 1984). Micro-

domain localisation of calcium in particular spaces is hypothesised to affect cell wall loosening and cell

separation, with potential impacts upon fungal infection (Roy 1995). This may be particularly prevalent

at three way cell junctions where turgor pressure is driving the separation of cells and the formation of

large intra-cellular spaces (Willats 2001). It has been suggested that the major physical effects of pectin

modification will therefore be in cell-cell adhesion rather than strength of the primary cell wall

(Ferguson, 1984). However, species and tissue differences in patterns of pectin deposition and

modification (through controlled expression of an array of cell wall modifying enzymes) indicate that the

situation may be much more complex.

14

Figure 3. Plant cell wall model. Primary cell walls are constructed of cellulose microfibrils (Cel),

crosslinked by hemicullose xyloglucan (XG), and a diverse array of pectins, adjacent to the plasma

membrane (PM). The predominant pectin backbones are homogalacturonan (HG) and

rhamnogalacturonan I (RGI); their complexity is increased by the occurrence of rhamnogalacturonan II

(RGII), arabinan (Ara), and galactan (Gal) side chains. The middle lamella is composed of mostly RGI

and HG, with adjacent de-esterified HG often cross-linked ionically by calcium (Ca2+). Reproduced from

Vorwerk et al. (2004).

Throughout fruit ripening, pectin de-esterification occurs by the action of pectin methyl-esterases

(PMEs). This exposes carboxyl residues which can be cross-linked by calcium. The level of PME

15

activity and Ca2+ availability within the apoplast has a direct impact on cell wall strength and expansion.

Studies in grapes suggest that PME expression begins before veraison and continues throughout

ripening (Barnavon et al., 2001; Schlosser et al., 2008). Mesocarp and skin tissues exhibit different

patterns of PME expression in grapes (Nunan et al., 1998; Schlosser et al., 2008; Lacampagne et al.,

2010). During the initiation of ripening a raft of cell wall modifying and hydraulic regulatory genes in

grape (including expansins EXP3 and EXPL, pectate lyase, a PME, and aquaporin PIP2;1) are

upregulated (Schlosser et al., 2008). This occurs initially in mesocarp; the delay in skin activation of

these genes suggests a role for the skin in moderating berry growth during ripening. The degree of

pectin de-esterification also varies between varieties (Ortega-Regules et al., 2008).

Plant and fungal PMEs have different modes of function; plant PMEs generally operate in a blockwise

manner, de-esterifying multiple homogalacturonan residues along a single chain, whereas fungal

PME’s operate in a non-blockwise manner (Willats et al., 2001). Patterning of pectin modification and

calcium binding may affect the binding and rate of pectin cleavage by polygalacturonases. Three

Botrytis isolates exhibited calcium inhibition of polygalacturonase activity. The calcium concentration

required to inhibit enzyme activity varied between isolates (Chardonnet et al., 2000). The pathogenicity

of these isolates also varied between four apple varieties, indicating that the interaction between the

pathogen cell-wall degrading enzymes and the composition of the fruit cell wall is important for

determining pathogenicity (Chardonnet et al., 2000). It has been demonstrated that the calcium content

of grape skin cell walls is negatively correlated with susceptibility to Botrytis enzymatic digestion

(Chardonnet and Doneche, 1995). Calcium infiltration reduced the level of pectin degradation by

Botrytis in grapes (Chardonnet et al., 1997), and reduced the level of decay in apples (Chardonnet et

al., 2000). The interaction of these factors contribute to the diversity of pectin profiles seen when

characterising cell wall composition from different species, varieties, tissues, developmental points, and

treatments, as well as influencing susceptibility to fungal pathogens.

The expression and activity patterns of cell wall modifying enzymes in grapes change throughout

development, as well as varying between varieties. In both Cabernet Sauvignon and Semillon,

polygalacturonase activity appears correlated with ABA levels, reaching maximum at veraison

(Deytieux et al., 2005). Polygalacturonase activity in skin was not detected, however transcripts of two

isoforms showed different expression patterns, with a common feature being strong expression late in

development (100 days after anthesis), indicating a variety of roles for the polygalacturonase family,

and a concerted role in cell wall disassembly at maturity (Deytieux-Belleau et al., 2008). Other research

16

has also indicated that transcript expression of polygalacturonases does not necessarily translate to

detectable enzyme activity (Nunan et al., 2001). In vitro calcium has an inhibitory effect on

polygalacturonase activity; drops in calcium concentration and availability following veraison may be

linked to increasing polygalacturonase activity at this time (Cabanne and Doneche, 2001). In skin

tissue, pectin methylesterase is present throughout ripening, with enzyme activity reaching a peak at

the beginning of veraison, decreasing sharply in the subsequent ten days, and increasing steadily

thereafter (Deytieux-Belleau et al., 2008). Low levels of polygalacturonase and pectate lyase activity

are observed during some stages of ripening; however, hormonal cues may regulate the targeted

expression of specific isoforms to produce the depolymerisation of pectin observed during ripening

(Nunan et al., 2001).

The combinatorial effects of pectin enzyme activity, apoplast pH and calcium concentration determine

various mechanical properties of pectin gels; compressive strength, water holding capacity, porosity

and elasticity (Tibbits et al., 1998; Willats et al., 2001; Ngouemazong et al., 2012). The strength of

calcium crosslinks is pH dependent, with the strongest bonds forming at apoplastic pH 6–7. Formation

and dissolution of pectin gels by calcium crosslinks is highly dependent on the level of de-esterification

(i.e. available carboxyl groups) and free calcium ion concentration (Tibbits et al., 1998). Gel swelling

can be observed during dissolution, due to both the osmotic pressure created by free carboxyl groups

in the pectin matrix (occurring when ionic strength is low), and the disassociation of calcium cross-

linked pectins. This can be expressed as the ratio of free calcium ions to carboxyl groups (i.e. if [Ca2+]

free: COO- <0.05 significant swelling is likely to occur). As gel dissolution and swelling occurs, the

breakdown of calcium crosslinks reduces the stiffness of the gel. Swelling of a gel is generally at

maximum around pH 3, which is also the pH at which calcium crosslinking and gel shear strength are at

a minimum (Tibbits et al., 1998). Some work has been done with pectin concentrations similar to those

observed in plant cell walls (films with ~30% pectin), demonstrating that pectin hydration status (or

degree of swelling) has a linear inverse relationship with tensile strength (Zsivanovits et al., 2004).

Additionally, the hydraulic properties and susceptibility to swelling of the pectin matrix are determined

by both the pectin composition and the ionic composition of the space (Zsivanovits et al., 2004).

Through understanding the pectin ion interaction effects on gel properties in vitro it is likely we will

advance our comprehension of fruit ripening processes.

Cell wall acidification promotes cell growth and expansion by displacing pectin bound Ca2+ through

protonation of pectin carboxyl groups. The pH of the apoplast may be affected by the pH of the xylem

17

sap when water delivery is high. Control of apoplast pH largely occurs through the activity of the

plasma-membrane localised H+-ATPase (this transport is regulated by auxin; covered in section 1.5).

Saline conditions initially decrease the leaf growth rate of a salt-sensitive maize cultivar by reducing H+-

ATPase activity, resulting in increased apoplastic pH; a tolerant hybrid cultivar showed none of these

effects (Pitann et al., 2009). This finding is contrasted by work measuring ion fluxes in bean leaf, which

showed an activation of H+ efflux from the mesophyll upon addition of NaCl (Shabala, 2000).

Amelioration of the effects of salinity on growth by high calcium has been observed; this may result

from interactive effects with plasma membrane transport proteins (such as H+/cation exchangers), or

through a reduction in the rate of Ca2+ displacement from pectin crosslinkages by Na+ and H+ ions

(Shabala, 2000). Additionally, cell wall localised expansins show optimal activity at low pH. They are

believed to act by reducing hydrogen bonding between primary cell wall components, allowing slippage

between adjacent polysaccharides and hence cell wall expansion (Sampedro and Cosgrove, 2005; Dal

Santo et al., 2013). Thus changes in apoplastic pH can have significant effects on the dynamics and

composition of the cell wall.

In addition to the de-esterification of pectins observed in fruit cell walls during ripening,

depolymerisation of pectins to shorter sub-units is also an important factor. By using a chelator (e.g.

CDTA) followed by size exclusion chromatograghy to extract and characterise ionically bound pectins,

the diversity in timing and degree of depolymerisation that occurs between species can be observed.

Some species show almost no change in pectin composition or solubilisation throughout development

(e.g. capsicum), whilst others (e.g. tomato) show high degrees of depolymerisation and high levels of

chelator soluble pectins (i.e. high levels of Ca2+ bound pectins) (Brummell, 2006). This

depolymerisation is achieved through the action of polygalacturonases and pectate lyases.

Polygalacturonases are absent or detected at very low levels in the fruit of some species; this may

account for some of the species differences in levels of pectin solubilisation. Additionally, it has been

demonstrated that reduced expansin activity (normally responsible for loosening of xyloglucan and

cellulose networks in the primary cell well during cell expansion) decreases solubilisation of primary cell

wall pectins, possibly through reduced access of polygalacturonases to their substrates in this space

(Brummell et al., 1999). This finding suggests that it is important to consider more than just transcript

levels or enzyme activity when considering potential for degradation of particular components; Ca2+

availability and activity of other cell wall enzymes may influence substrate accessibility.

18

Degradation of pectic homogalacturonan backbones generates short chain molecules known as

oligoglacturonides (OGAs). These have been implicated in pathogen defence signalling activation. This

role is carried out through OGA binding by the wall-associated kinase (WAK) family (Decreux and

Messiaen, 2005). It is of particular interest in fruit ripening, as the high pectin content, and the appeal of

ripening fruit to a variety of pathogens, suggests that functional OGAs may be quite prevalent in fruit.

Many factors influence the defence response eliciting capacity and specificity of OGAs, including;

calcium availability, length of OGA, degree of methyl-esterification and degree of acetylation (Decreux

and Messiaen, 2005; Vallarino and Osorio, 2012). The extracellular domain of Arabidopsis WAK1 binds

OGAs only in the presence of calcium and calcium crosslink forming conditions (Decreux and

Messiaen, 2005). Expression of a fruit-specific PME from cultivated strawberry in transgenic wild

strawberry resulted in a modified pattern of OGA esterification in the transgenic fruit. This change was

sufficient to constitutively activate defence responses in the transgenic, thereby increasing Botrytis

resistance (Osorio et al., 2008). A variety of evidence suggests that in addition to WAKs binding

specific OGAs during pathogen responses, WAKs also bind cell wall pectins during normal

development to regulate cell expansion (e.g. antisense reduction in WAK expression has been shown

to reduce cell size) reviewed in Kohorn and Kohorn (2012). It is apparent that the specificity of calcium-

pectin-WAK interactions may facilitate multiple signalling pathways important in pathogen defence

activation as well as during normal developmental control of cell expansion.

1.5 Calcium-hormone interactions during fruit development

Calcium is a known secondary messenger involved in hormone signalling. The various levels of control

exhibited in plant calcium relations combined with the complexity and concentration dependence of

hormone responses dictates that changes in calcium nutrition may create multiple plant hormonal

responses that are difficult to characterise. Although the physiological pathways and interactions of

plant hormones and calcium are still being uncovered, many hormone and calcium treatments are

already used for horticultural improvement. The role of plant hormones in fruit development and

ripening processes has been extensively reviewed (Ruan et al., 2012; McAtee et al., 2013; Osorio et

al., 2013; Wang and Ruan, 2013; Kumar et al., 2014; Leng et al., 2014). This section will therefore

articulate the current knowledge and gaps in our understanding of calcium and hormone interactions in

fruit.

19

Auxin has key roles in fruitset, cell division and cell expansion. These developmental pathways both

utilise calcium as a secondary messenger and affect patterns of calcium distribution. Fruitset and early

development are triggered by auxin synthesis in the ovules during fertilisation, which induces (GA)

synthesis, reviewed in Kumar et al. (2014). GA signalling in the pericarp of Arabidopsis fruit has been

demonstrated to activate a pathway degrading the growth inhibiting DELLA proteins (Fuentes et al.,

2012; Kumar et al., 2014). The relationship between GA and auxin is complex, with a GA independent

pathway for fruitset being demonstrated in tomato (Serrani et al., 2008; McAtee et al., 2013). High

levels of GA are commonly associated with rapid cell expansion, this has also been linked to low or

reduced calcium concentrations by disrupting calcium transport (Saure, 2005); the mechanism through

which this occurs requires further investigation.

Calcium acts as a secondary messenger downstream of auxin through the acid growth pathway. This

pathway has been demonstrated in Arabidopsis; auxin efflux from cells is facilitated by PIN-FORMED

(PIN) membrane proteins. PIN activity and targeted endocytotic transport of PIN are regulated by

PINOID (PID) protein kinase and PP2A phosphatase complex mediated phosphorylation (Fozard et al.,

2013). Extracellular auxin (possibly though the binding of ABP1) activates plasma membrane calcium

transport in wheat embryos, creating [Ca2+]cyt transients that activate the plasma membrane localised

H+-ATPase to reduce apoplastic pH. Lower apoplastic pH activates pH sensitive cell wall loosening

enzymes (Rober-Kleber et al., 2003; Shishova and Lindberg, 2010; Wang and Ruan, 2013). This proton

influx into the cell wall compartment also increases competition with calcium for binding sites on de-

esterified pectin, resulting in looser cell walls; as such higher levels of calcium can inhibit auxin-

activated acid growth (Ferguson, 1984). H+-ATPase transport also activates voltage dependant

potassium channels; potassium levels increase, causing osmotically driven movement of water into the

cell, increasing cell turgor pressure. The acid growth pathway is responsible for cell elongation and

expansion; it has been observed during growth of many tissues, including root tip growth, and

hormone-stimulated cell expansion.

Auxin is also involved in calcium uptake and distribution in fruit. Application of CME (an auxin transport

inhibitor) reduced calcium uptake into developing fruit of some tomato cultivars differing in susceptibility

to blossom end rot (Brown and Ho, 1993). This reduced calcium uptake may occur through modification

of cellular transport activity or perturbed cell expansion (disrupting xylem development). Calcium is also

involved in fruit basipetal auxin transport. CME induced reductions in basipetal IAA efflux were only

observed in tomato fruit grown under high salinity conditions where Ca2+ uptake was reduced (Brown

20

and Ho, 1993). In kiwifruit, light induction of higher levels of auxin-protecting hydroxycinnamic acids

decreased auxin degradation, resulting in increased calcium uptake (Montanaro et al., 2006). These

results suggests tomato susceptibility to blossom end rot may be determined not just by differences in

capacity for Ca2+ uptake and distribution, but also by related factors such as auxin transport and

metabolism, and rate of cell enlargement.

Fruit ripening processes typically involve ABA and ethylene signalling. Non-climacteric fruit show a

greater reliance upon ABA for initiation of ripening processes and do not demonstrate the same extent

of ethylene responsiveness as climacteric fruit. ABA signalling in Arabidopsis acts through a network of

calcium binding signal receptors including PP2C protein phosphatases ABI1 and ABI2 (Leung et al.,

1994; Allen et al., 1999), and an array of CBL (Pandey et al., 2004), CIPK (Kim et al., 2003), and CDPK

(Zhu et al., 2007) protein kinases. Concentration of ABA in grapes increases dramatically at the

beginning of veraison; it is possible that the drop in cell turgor that occurs at this time triggers increases

in ABA levels (Castellarin et al., 2011). There are varietal differences in the time during veraison (

measured as % colour change) at which maximal berry ABA is reached; Merlot (10 % colour change),

Cabernet Sauvignon (50% colour change), and Semillon (100% colour change) (Deytieux-Belleau et

al., 2008). The partitioning of ABA between mesocarp and skin shifts from 100% in mesocarp prior to

veraison, to approximately 40% in skin by maturity (Deytieux-Belleau et al., 2008). In non-climacteric

fruit (e.g. strawberries and grapes) several other factors have been identified as potential ripening

signal elements; auxin treatment of unripe fruit delays ripening (Boettcher et al., 2011) whilst reactive

oxygen species accumulate in grape berries at the onset of ripening (Pilati et al., 2014). The

transduction of these ripening triggers through the calcium signalling network suggests that calcium

also plays a role in sugar accumulation and fruit softening, evidence for these functions are examined

below.

Increased ABA levels at ripening leads to increased hexose accumulation through up-regulation of

hexose transporters and increased apoplastic invertase activity (Pan et al., 2005; Deluc et al., 2009;

Hayes et al., 2010). ABA activates sugar cell wall bound invertases at the initiation of grape ripening,

catalysing sucrose cleavage, decreasing the apoplastic sucrose concentration, and thereby allowing for

continued phloem unloading of sucrose into the berry apoplast (Pan 2005). Phloem unloading of sugars

becomes crucial for driving expansion, as well as efficiently maintaining the accumulation of sugars in

the fruit. It has been demonstrated that ABA activates the calcium dependent protein kinase (ACPK1) in

grape mesocarp through a complex mechanism involving influx of apoplastic calcium to the cytosol (Yu

21

et al., 2006). ACPK1 in turn activates plasma membrane H+-ATPase in the berry mesocarp, possibly

energising the cell for solute uptake (Yu et al., 2006). A transient decrease in calcium concentration is

observed in the apoplast of Vicia faba leaves following ABA treatment, providing further evidence for

apoplastic calcium as a transducer of ABA signalling (Felle et al., 2000). The timing of ABA

accumulation, metabolic responses and the drop in turgor varies between varieties with differing

ripening profiles (Deluc et al., 2009; Castellarin et al., 2011).

Endogenous hormone levels influence fruit softening by altering expression levels of enzymes that

modify cell turgor pressure, apoplast solute accumulation, and cell wall modification. Suppression of a

key enzyme involved in tomato ABA biosynthesis (9-cis-epoxycarotenoid dioxygenase) resulted in the

transcriptional down regulation of polygalacturonase, pectin methylesterase, expansin, and many other

cell wall modifying enzymes (Sun et al., 2012; Osorio et al., 2013). Application of gibberelic acid has

been shown to increase berry firmness and shelf life in the table grape variety Thompson Seedless

(Marzouk and Kassem, 2011). Conversely, in Cabernet Sauvignon berries treated with sucrose or

sucrose and ABA, a drop in berry firmness (as occurs at the onset of ripening in the field) was only

observed in the sucrose and ABA treated berries (Gambetta et al., 2010). The combined effect of ABA-

activation of sugar invertases and cell wall modifying enzymes in the apoplast is to simultaneously

reduce cell turgor pressure and loosen cell walls (Pan et al., 2005; Gambetta et al., 2010).

1.6 Conclusion

The complex nature of the multiple roles of calcium in fruit ripening highlights the requirement for further

research efforts towards understanding calcium nutrition for improvement of fruit quality outcomes in

horticultural systems. Changes in the polysaccharide composition of the cell wall are responsible for

some of the softening associated with fruit ripening. A range of modifications within the primary cell wall

and middle lamella cause both a loss of cellular structural integrity and weakened intercellular linkages.

Cell turgor pressure changes with solute accumulation during fruit ripening; similarly the rigidity of the

cell wall and maintenance of cell membrane integrity changes during ripening. As fruit cell walls

generally have high pectin content, the structural modification and depolymerisation of pectins is of

particular importance in fruit ripening. Increases in pectin de-esterification in the middle lamella during

ripening indicate that proper calcium nutrition is required for formation of calcium cross linkages and

maintenance of cell-cell adhesion. Localised calcium deficiencies have been implicated as the cause of

22

many significant fruit quality disorders, including blossom end rot, bitter pit, and tip burn. Through

furthering understanding of the combined effects of calcium nutrition upon fruit water relations, cell wall

properties, and developmental signalling, horticultural approaches to address these problems can be

more readily formulated.

In this study, grapevines will be used a model for plant and fruit calcium nutrition responses. As the

underlying relationships between calcium levels and fruit cell wall and physical properties have not

been elucidated, this study will determine if the indirect evidence suggesting that high calcium growth

leads to firmer cell walls is supported. Furthermore, the hypothesis that differences in berry calcium

uptake contribute to variety specific differences in susceptibility to physiological disorders will be tested.

This study will also attempt to overcome the difficulties of assessing nutritional responses by utilising a

hydroponics system that tightly controls nutritional inputs, allowing for manipulation of calcium supply to

vines. It is believed that this will assist in disentangling non-specific effects and allow the role of calcium

in vine water relations and fruit ripening to be clearly resolved.

23

Chapter 2 Varietal differences in berry physical properties, nutrient

accumulation and pectin distribution

2.1 Introduction

In grapes, cell wall properties are key determinants of grape and wine quality; extensive cell wall

disassembly occurs during ripening, this developmental stage is also associated with increased

pathogen susceptibility, berry splitting and berry shrivel (Cabanne and Doneche, 2002; Cantu et al.,

2008; Fuentes et al., 2010). Calcium nutrition impacts berry quality by facilitating stress signalling,

stabilising membranes, and imparting cell wall strength by cross-linking de-methylated pectins

(Cabanne and Doneche, 2003). A broad aim of this research is to understand how varietal differences

in calcium uptake and distribution may relate to the previously observed differences in maintenance of

berry hydraulic conductivity and cell vitality late into the season. Some varieties (e.g. Shiraz) show a

high propensity to lose mesocarp cell vitality and shrivel during late ripening, whereas other varieties

(e.g. Thompson Seedless and Chenin Blanc) are able to maintain mesocarp cell vitality and berry

integrity (Tilbrook and Tyerman, 2008; Fuentes et al., 2010). Varietal differences in mesocarp vitality

may be related to xylem hydraulic conductivity and capacity to re-distribute mineral nutrients within the

berry for maintenance of osmotically competent membranes (Rogiers et al., 2006a; Tilbrook and

Tyerman, 2008). In this chapter, the role of berry nutrition (particularly calcium nutrition and distribution

within the berry) and its relationship to berry physical qualities and maturity indicators will be

investigated between varieties differing in their susceptibility to berry shrivel. In addition, berry

properties of three varieties from two sites that differ in their calcium nutrition are studied in an attempt

to disentangle site-specific and variety-specific effects.

2.2 Materials and Methods

2.2.1 Bunch and berry sampling

Bunches were collected from the Coombe vineyard variety block (34°58’3” S, 138°37’59” E) located on

the Waite Campus of the University of Adelaide, and a commercial vineyard (34°29’57” S, 139°01’20”

E) located in Angaston in the Barossa Valley (courtesy of Chris Fiebiger) during the 2011-12 and 2012-

2013 seasons. Both vineyards had a vertically shoot positioned trellis system with all varieties planted

on own roots. Waite vineyard soil type is classified as Dr2.23 hard red duplex with clay subsoil

(Northcote, 1988). Row orientation is north-south and vines are planted at 1.8 m between vines and 3.0

24

m between rows under drip irrigation receiving 0.8 ML/ha of irrigation per season and pruned to 50

nodes per vine. Barossa vineyard soil type is classified as Uc2.21 bleached sand with deep calcareous

subsoil (Northcote, 1988). Row orientation is northwest-southeast and vines were planted at 3.0 m and

3.6 m between rows under drip irrigation receiving 0.25 ML/ha (Grenache) or 0.5 ML/ha (Shiraz and

Chenin Blanc) of irrigation per season and pruned to 30 nodes per vine. Varieties collected from the

Waite were Shiraz, Grenache, Chenin Blanc, Chardonnay, and Thompson Seedless. Varieties collected

from the Barossa were Shiraz, Grenache, and Chenin Blanc. Vines were selected randomly (avoiding

end of row cordons) following anthesis and two bunches per vine were tagged. Tagged bunches were

typically mid-sized (i.e. appeared to have good fruit set) and mid-vine (i.e. not exposed to excessive

sun). Picking of Barossa samples was timed to coincide with commercial harvest; Waite samples were

picked at a similar stage of maturity as Barossa samples, as determined by total soluble sugar content

using a digital refractometer. Picked bunches were placed in a cooler box pre-chilled with ice to

approximately 10 °C for transport to the laboratory. Berries selected for experimental analysis were

intact, mid-bunch berries that did not show any visible signs of physical damage, biological damage,

sunburn, or other anomalies.

2.2.2 Microscopy

Berry pieces (approximately 3 X 3 X 3 mm) were removed by cutting a triangular pattern into the berry

halfway up the side with a fresh razor blade. The resulting ‘pyramidal’ piece came away on the edge of

the blade. The piece was placed into TEM fixative (4% sucrose, 1 X PBS, 4% paraformaldehyde,

0.25% glutaraldehyde), where it was stored at 4 °C overnight (or at 4 °C for up to 2 weeks until

processing). Samples were rinsed with 1 X phosphate buffered saline (PBS) buffer at 4 °C (2-3

changes in 8 hours) to remove excess fixative. Ethanol dehydration was carried out using a series of

changes at 20 minutes each (3 X 70%, 3 X 90%, and 3 X 95%), followed by 3 X 100% dry ethanol

changes of 4 hours, 16 hours, and 8 hours each. The 100% dry ethanol was prepared using a 3A bead

size 8-12 mesh molecular sieve (Sigma Aldrich, Australia). Infiltration with hard grade LR resin

(Proscitech, Thuringowa, Australia) was carried out in a 50:50 mix of 100% ethanol/LR white resin

incubated overnight then with 3 X 100% resin incubations of 8 hours each. Resin-infiltrated samples

were positioned in size 1 gelatine capsules (Proscitech, Thuringowa, Australia) filled with fresh resin.

Berry pieces were consistently placed in the same orientation to achieve similar planes of view for cross

sectioning and ease of microtomy. Polymerisation was undertaken in an oven at 56 °C for 48 hours.

Resin blocks were trimmed with a razor blade to give a trapezoid face. Sections for light and

fluorescence microscopy (1 µm thick) were cut on a Leica EM UC6 ultra-microtome (Leica, Germany)

25

with glass knives. Sections were transferred to Polysine slides (Thermo Scientific, Massachusetts,

USA) for staining as detailed in the following section. Immunofluorescent stained slides were imaged

using a Zeiss Axio Imager M2 (Zeiss, Germany) as detailed in the figure legends.

2.2.3 Histological staining

Toluidine blue staining- Slides were placed on a hot plate (90-100 °C). Sections were covered with a

drop of toluidine blue O (Proscitech, Thuringowa, Australia) working solution (containing 5 mL of 10 g/L

toluidine blue O in 70% ethanol, and 45 mL of 10 g/L NaCl, pH 2.3), stained for 15–20 seconds, rinsed

well with RO water, and replaced on the hot plate to dry.

Methylene blue/basic fuchsin staining- Slides were placed on a hot plate (90–100 °C). Sections were

covered with a drop of Methylene blue solution (containing 0.13 g methylene blue, 0.02 g Azure II, 10

mL glycerol, 10 mL methanol, 30 mL 0.15 M phosphate buffer (pH 6.9), 50 mL distilled water:

Proscitech, Thuringowa, Australia) stained for 40 seconds, rinsed well with RO water, and replaced on

the hot plate to dry. Phosphate buffer (50 mL) was prepared by adding 28 mL 0.15 M Na2HPO4 to 22

mL 0.15 M NaH2PO4 to achieve pH 6.9. Sections were covered with a drop of basic fuchsin stain;

freshly prepared 0.125 g basic fuchsin (Proscitech, Thuringowa, Australia) and 0.125 g borax in 100 mL

distilled water. Sections were stained for 30 seconds, rinsed well with RO water, and replaced on the

hot plate to dry. Coverslips were sealed onto dried slides using Entellan™ new rapid mounting medium

according to manufacturer’s instructions (Merck, Darmstadt, Germany).

2.2.4 Immmunofluorescent staining

A PAP paraffin pen (Sigma Aldrich, Australia) was used to encircle sections on the slide, creating a well

for antibodies. Slides were placed in a slide holder filled with 1 X PBS for 5 minutes. The PBS was

replaced with 0.05 M glycine for 20 minutes, then with incubation buffer (1% BSA in 1 X PBS) twice for

10 minutes. Slides were removed and wick-dried on Kimwipes (Kimberly-Clark, Dallas, USA) and 50 µL

of primary antibody solution (1:50 dilution of Rat IgM LM19 or LM20, Plant Probes, Leeds, UK) was

added and the slide placed in a humidity chamber (plastic box lined with wet paper) for 1 hour. Primary

antibody was washed off with incubation buffer using a plastic Pasteur pipette, followed by 3 X 10

minute washes with incubation buffer in the slide box. Slides were removed and wick-dried on

Kimwipes, 50 µL of secondary antibody solution (1:200 dilution of Goat anti-rat IgG Alexa Fluor488, Life

Technologies, California, USA) was added, and the slide was placed in a humidity chamber (plastic box

lined with wet paper) for 2 hours, with the chamber wrapped in aluminium foil to protect the slides from

26

light. The secondary antibody was washed off with incubation buffer using a plastic Pasteur pipette,

followed by 3 X 10 minute washes with incubation buffer in the slide box. Coverslips received a drop of

90% glycerol before dried slides were inverted over them.

2.2.5 Berry physical testing

Physical testing of berries was undertaken using an Instron Model 5943 with proprietary Bluehill

software (Instron Corporation, Massacheusetts, USA). The Instron is a single column tabletop system

for performing precision tensile and compression materials testing.

2.2.5.1 Test protocols

For all Instron measurements, representative berries were taken from the bunch and basic physical

parameters measured prior to testing; berry weight with a digital balance, berry height (measured in the

distal to proximal axis, i.e. from pedicel to stylar end) and diameter (measured in mid-section of the

berry perpendicular to the distal-proximal axis) with a digital micrometer. For soft or shrivelled berries,

diameter was measured as the widest point (i.e. no compression from micrometer) in the axis that was

to be subsequently tested by the Instron. Berries were removed from the bunch with grape snips,

leaving a small section of pedicel (1-2 mm) still attached to prevent wounding at the pedicel end that

might affect test progress. Test protocols were designed around previously published test protocols for

berry strength testing (Rolle et al., 2012a) using the Bluehill software and available test probes.

For berry hardness tests a 50 N load cell and drill-chuck attachment was fitted. Whole berry

compression was tested using a 12.75 mm diameter flat faced fitting. Berries were positioned directly

under the fitting, and the test proceeded at a rate of 1 mm per second.

For skin hardness tests a 50 N load cell and drill-chuck attachment was fitted. Skin compression was

tested using a 2 mm diameter needle point fitting. Intact berries were positioned directly under the

fitting, and the test proceeded at a rate of 1 mm per second.

For skin thickness tests a 5 mm X 10 mm rectangular strip of skin tissue (composed of epidermal and

hypodermal cell layers) was cut on the berry using a razor blade. The strip was then carefully peeled off

using forceps. Excess fluids were removed by gently dabbing the strip on a Kimwipe. A 50 N load cell

and drill-chuck attachment was fitted. Skin thickness was tested using a 12.75 mm diameter flat faced

fitting. The skin strip was placed on a flat plate directly under the fitting, and the test proceeded at a rate

of 1 mm per second.

27

2.2.5.2 Instron data analysis

A method for manually calculating berry Young’s Modulus from raw Instron data was formulated. This

approach factored in berry dimensions and shape, unlike the automated calculations on the Bluehill

software. Briefly, raw test data and berry dimensions were copied into Microsoft Excel. A stress-strain

curve was generated, with compressive extension (mm) on the X-axis and compressive load (N) on the

Y-axis. The linear portion prior to maximum load and/or break was selected (consisting of at least 10

data points with a linear regression R2 value of >0.99). The linear regression slope and the berry

deformation, D (mm), over the linear portion were used to calculate the force applied, F (N). Poisson

ratio (v) was assumed to be 0.5. An example stress-strain curve is provided in Appendix 1. These

factors, together with the undeformed berry radius (R) were used to calculate Young’s Modulus, E

(MPa), as per Wada et al. (2009), according to the formula;

𝐸(𝑀𝑃𝑎) =3F(1 − 𝑣2)

√2𝑅𝐷

2.2.6 Apoplast fluid centrifuge method

Apoplastic fluids were collected from berries using a modified version of a previously published low-

speed centrifuge method (Wada et al., 2008). Individual berries were cut from the bunch with grape

snips, leaving the pedicel attached. A 3 mm slice was then removed from the distal end of the berry

using a fresh razor blade. The flat surface of the cut berry was then dabbed on a lint-free Kimwipe to

remove cellular fluids from cut cells. The berry was then placed on a stage for low speed centrifugation.

The stage was constructed from the modified cap and screw thread of a 10 mL tube. The top of the lid

and the section of the tube below the screw thread were removed using a scalpel. A square of nylon

mesh (30 mm X 30 mm) was then placed on top of the tube and the open top cap screwed down onto

the thread, producing a tight mesh filter screen. The cut berry was gently placed inside this filter, which

was then placed upright inside a 50 mL falcon tube (with cut berry surface resting against filter mesh

pointing downwards towards the conical base of the falcon tube. This filter method was settled upon

after trialling a variety of other materials for supporting the cut berry in the tube. Plastic ring supports

made from 2 mL Eppendorf tubes and lids were found to damage the berry circumference, producing

contamination of apoplastic extract with cellular fluids. The mesh screen ably distributed the force of

centrifugation across the cut surface; it also provides a greater surface area for wicking of apoplastic

extract away from the berry. Various centrifuge speeds and times were trialled, with it being necessary

28

to optimise these variables for berries of different maturity (i.e. veraison, post-veraison, over-ripe) and

variety (e.g. Shiraz, Thompson Seedless). Both potassium concentration and pH were used as

diagnostics for cellular fluid contamination of collected samples; pH was tested using a micro-pH probe,

pH of collected apoplast samples were compared to pH of juice squeezed from the same berry after

centrifugation. If significant cellular contamination of the apoplast samples was occurring, then the two

readings would be expected to be similar. The apoplast fluid obtained typically had a pH value 0.3-1.0

higher than the respective mixed cellular fluid (predominantly vacuolar fluid). Centrifuge times trialled

were 5 minutes and 20 minutes, with centrifuge forces of 100, 200, 400, and 1000 x g. Volumes

collected were typically 15-50 µL. Standard protocol was 5 minutes at 200g, yielding consistent

uncontaminated samples for most varieties post-veraison. However, Thompson Seedless failed to yield

sufficient volume for analysis using the standard protocol; the firmness of this table grape variety

throughout maturation meant that it was able to withstand greater centrifugation speeds without

suffering cell damage and contaminating apoplast samples. Consequently, for Thompson Seedless,

centrifugation for 5 minutes at 1000 g was used. Chenin Blanc also yielded very low volumes using the

standard protocol. Unlike Thompson Seedless, Chenin Blanc does soften during veraison and ripening,

and does show cell damage when centrifuged at higher speeds. In order to extract uncontaminated

apoplastic fluid from Chenin Blanc, samples were collected over longer centrifugation times (20 minutes

at 200 g). This still yielded reasonably low volumes, and individual berry samples were bulked together

to achieve sufficient volume for analysis.

2.2.7 Nutrient analysis; low volume apoplast method

A low volume protocol was designed for inductively coupled plasma atomic emission spectroscopy

(ICP-AES) analysis of apoplast fluid samples. An electronic dilutor-dispenser (Gilson model 402,

Wisconsin, USA) was used for measurement of apoplast sample volume and dilution with 2% nitric acid

to a final volume of 5 mL. After flushing the tube with diluent and drawing a 20 µL air gap, the diluter

was set to aspirate at 0.1 mL/min, the probe was inserted into the sample and aspirated until the whole

sample was drawn up, the volume was recorded (generally 20-50 µL), then diluted out into a fresh 10

mL tube with 2% nitric to a volume of 5 mL. This approach allowed for accurate measurement of initial

volumes and eliminated loss of sample. Due to the high level dilution in these samples, the ICP-AES

was cleaned prior to each run, and a low-drift method employed. This established a low background

level of detection and allowed for more confidence in assessing samples near the limit of quantification.

Initial tests were performed using 5 mL water to rinse nylon mesh squares and newly prepared filters

screens then performing ICP on the rinse solutions to test the contribution of these components to

29

background elemental detection. The contribution of new screens was negligible and triple rinsing used

screens in milliQ water and storing in 2% nitric acid was sufficient to remove any contamination and

allow reuse (data not shown).

2.2.8 Total soluble sugars, pH and titratable acidity methods

Bunch juice samples were collected by pressing 10-15 randomly selected intact berries from each

bunch in a small hand press. Pressed juice was clarified by centrifugation for 5 min at 3500 rpm. Total

soluble sugar measurements were taken from pressed juice samples (or individual berries where

appropriate) using a digital refractometer with automatic temperature correction. pH measurements of

bunch samples were taken using a digital pH probe. Individual berry juice was collected using a garlic

press. pH measurements from individual berries and apoplastic fluid samples were taken using a Micro

Combination pH electrode (Lazar Research Labs, California, USA). pH and titratable acidity (TA) were

also measured concurrently from bunch juice samples using a Crison Compact auto-titrator (Crison

Instruments, Barcelona, Spain). 10 mL of centrifuged grape juice was diluted with 40 mL of RO water

in plastic sample cups. The auto-titrator was calibrated at pH 4 and pH 7; outputs of initial pH, titre

volume, and final TA were recorded.

2.2.9 Apoplastic ion selective electrode (ISE), pH and calcium activity

measurements

Apoplast fluids were collected as per protocol described in 2.2.6. Micro-pH PHR-146B, calcium

microelectrode ion selective electrode (ISE) ISM-146A, and reference microelectrode DJM-146 (Lazar

Research Labs, Los Angeles, USA) were used to measure the pH and Ca2+ activity of individual berry

and apoplastic fluid samples. The pH of all samples was measured first, followed by Ca2+ activity. ISEs

were connected to a micro-voltmeter (Hanna Instruments H2211; Rhode Island, USA). It was observed

that the readings taken with the calcium microelectrode were prone to fluctuation by electrical

interference and proximity of the user and readings took several minutes to stabilise. A Faraday cage

placed around the electrodes eliminated interference and the influence of electrode drift was negated

by regular calibration. It was also found that the background composition of the solution being

measured (i.e. H2O or berry juice) had a large impact on the readings. This proved problematic for

calibrating the calcium ISE as the background of standards would ideally be similar to the samples

being measured. To address this, an ‘Artificial Grape Juice’ containing the major constituents of grape

juice with a range of calcium concentrations was created to serve as a background for ISE

measurements. Adjustments were made to ensure stability of the solution and eliminate tartrate

30

precipitation. Fresh solutions were made and calibrations carried out on each day of use. Artificial

grape juice composition was as follows; 555 mM D-(+)-Glucose, 555 mM D-Fructose, 26.6 mM cream

of tartar (KC4H5O6), 22.4 mM L-malic acid, 1.04 mM citric acid, 12.9 mM KH2PO4, 2 mM MgSO4, CaCl2

concentrations from 0-10 mM, pH adjusted to 3.2. Example calibration curves for 0-100 mM CaCl2

water based solutions and 0-5 mM CaCl2 artificial grape juice solutions are shown in Appendix 2.

2.2.10 Principal component analysis

Principal component analysis was performed using Unscrambler X (CAMO Software, Oslo, Norway).

Data was pre-processed to scale variables by the inverse of the standard deviation (producing a

standardized variance of 1.0 for all variables) (Esbensen, 2004). Mean centering was applied to the

principal components (Esbensen, 2004). Means were calculated from bunches from 10 replicate vines

per sample. Instron data represents mean values from measurement of five individual berries per

bunch. All other data represents pooled berry measurements, where five (ICP-AES data, apoplast pH,

apoplast Ca2+ activity) or ten (brix, pH, TA) berries from each bunch were pooled to give a single

measurement per bunch.

2.3 Results

2.3.1 Berry histological staining

Staining of grape berries with toluidine blue O can reveal details of phenolic distribution between cell

types and also highlight cell walls. Toluidine blue O stains phenolics bluish-green and cell walls

reddish-purple (Cadot et al., 2011). Shiraz demonstrated the strongest staining pattern for phenolics,

with cells of the outer epidermal layer displaying fine granulations and some spherical vacuolar

inclusions, hypodermal layers displaying large dense staining vacuolar inclusions, and outer mesocarp

cells displaying smaller dispersed inclusions (Figure 2.1A). Shiraz also had a much more rounded

epidermal cell shape than all other varieties surveyed. Grenache demonstrated additional layers of

epidermal cells showing small inclusions, lower staining intensity in hypodermal cell inclusions, and a

mix of small inclusions, granulation and uniform staining in outer hypodermal cells (Figure 2.1B).

Chenin Blanc demonstrated mostly granular staining in the epidermis, uniform vacuolar staining in the

hypodermis, and uniform staining of select cells of the outer mesocarp (Figure 2.1C). Thompson

Seedless demonstrated very low phenolic staining by toluidine blue O with some granulations occurring

irregularly across various cell types (Figure 2.1D). Chardonnay had small vacuolar inclusions containing

31

phenolics in the epidermis, and uniform or granular staining in the hypodermis and outer mesocarp

(Figure 2.1E). Toluidine blue O staining also highlighted some differences between cell walls. Cell wall

thickness of Shiraz cells decreased from the periphery of the berry to the interior from thick walled

epidermal cells to moderately thick walled hypodermal cells to thin walled mesocarp cells (Figure 2.1A).

In contrast, Grenache had thick cell walls in epidermal, hypodermal, and mesocarp cell types (Figure

2.1B, G); Thompson Seedless had relatively thin walls surrounding all cell types (Figure 2.1D, I).

Methylene blue/basic fuchsin stains phenolics blue, cytoplasm pink, and cell walls and lipids red (Kraus

et al., 1998) The pattern of methylene blue-basic fuchsin phenolic staining and cell wall morphology

was consistent with that observed using toluidine blue O (Figure 2.1).

2.3.2 Berry immunofluorescent staining

Immunofluorescence microscopy using the pectin specific antibodies LM19 and LM20 demonstrated

that pectin was distributed throughout berry epidermal and hypodermal cell walls across all varieties

surveyed (Figure 2.2). LM19 (which binds de-esterified pectins) showed a diversity of staining patterns

between varieties. Shiraz, Chenin Blanc and Chardonnay showed a banded staining pattern, with

patches of intense staining observed in the middle lamella of lateral walls and at tri-cellular junctions

(Figure 2.2 A, C, E). Grenache and Thompson Seedless demonstrated a strong punctate pattern

throughout the walls of both epidermal and hypodermal cells (Figure 2.2B, D). LM20 (which binds highly

esterified pectins) showed less clear differences in staining pattern between varieties. Some areas of

stronger staining were observed in the middle lamella and transverse walls of hypodermal cells in a

number of varieties (Shiraz, Grenache, and Thompson Seedless) (Figure 2.2F, G, I).

2.3.3 Berry physical properties

Instron tests were used to analyse the hardness and elasticity of both whole berries and skins, as well

as to measure skin thickness. Shiraz showed low berry and skin hardness, as well as high elasticity of

berries and skins compared to other varieties (Figure 2.3A-D). Instron measurements indicated that

Grenache had significantly higher skin hardness and thicker skin than other varieties (Figure 2.3C, E).

These attributes may be related to the greater number of epidermal/hypodermal layers with thick cell

walls in Grenache (Figure 2.1B, G) compared to Shiraz, Chenin Blanc and Thompson Seedless berries

(Figure 2.1A, C, D, F, H, I). Chenin Blanc samples from the Waite vineyard showed greater berry-to-

berry variation in some measurements (e.g. skin thickness by Instron; Figure 2.3E). Thompson

Seedless had the hardest and least elastic berries of surveyed varieties (Figure 2.3A, B).

32

Figure 2.1. Representative images of varietal differences in grape cell morphology. Toluidine blue O (A-

E) and methylene blue/basic fuchsine (F- J) stained berry sections of Shiraz (A, F), Grenache (B, G),

Chenin Blanc (C, H), Thompson Seedless (D, I) and Chardonnay (E, J). Tissue fixed in LR resin and

visualised by light microscopy. Scale bars represent 50 µm. Similar sections were obtained from three

replicate berries from separate vines.

33

Figure 2.2. Representative images showing varietal differences in distribution of de-esterified and

esterified pectin epitopes. De-esterified homoglacturonan (LM19; A-E) and highly esterified

homogalacturonan (LM20; F-J) immuno-labelled sections of Shiraz (A, F), Grenache (B, G), Chenin

Blanc (C, H), Thompson Seedless (D, I), and Chardonnay (E, J) berries. Tissue fixed in LR resin and

visualised using immuno-fluorescence. Scale bars represent 50 µm). Similar sections were obtained

from three replicate berries from separate vines.

34

Figure 2.3. Instron measurement of berry physical characteristics reveals varietal differences. Separate

tests were performed for berry hardness (A), skin hardness (C), and skin thickness (E). Young's

modulus calculated using Instron software for berry (B), and skin (D) elasticity. Different colours indicate

different varieties (Shiraz; red, Grenache; purple, Chenin Blanc; yellow, Thompson Seedless; green,

Chardonnay; white).Mean + SD (n=10 replicate bunches, bunch means from 5 berries per bunch).

Columns with different notation are significantly different using one way analysis of variance and a

Tukey posthoc test at 95 % CI.

Interestingly, it showed low hardness in the skin (measured by maximum breaking energy using a

needle probe) (Figure 2.3C, D). Chardonnay had the second highest skin hardness after Grenache

0 .0 0

0 .0 2

0 .0 4

0 .0 6

0 .0 8

Be

rry

ha

rd

ne

ss

(J

)

b cb c

c

b

a

A

0 .0 0 0

0 .0 0 2

0 .0 0 4

0 .0 0 6

Sk

in h

ard

ne

ss

(J

)

b

c c

a

c

C

0 .0

0 .1

0 .2

0 .3

0 .4

Be

rry

Yo

un

g's

mo

du

lus

(N

mm

-1)

c

b

c

b

a

B

0 .0 0

0 .0 5

0 .1 0

0 .1 5

Sk

in Y

ou

ng

's m

od

ulu

s (

Nm

m-1

)b c

b

c

a a

D

Sk

in t

hic

kn

es

s (

M)

0

5 0

1 0 0

1 5 0

b

c

ba

b

E

S h ira z

G re n a c h e

C h e n in B la n c

T h o m p s o n S e e d le ss

C h a rd o n n a y

35

(Figure 2.3C) and moderately thick skin (Figure 2.5E). Similar to Grenache, Chardonnay was observed

to have multiple epidermal layers and relatively thick cell walls compared to other varieties (Figure 2.1).

2.3.4 Berry tissue nutrient composition

A number of analyses were carried out on extracted apoplast fluid to characterise the composition of

this compartment (Figure 2.4). Full ICP-AES data for apoplastic fluids, skin and mesocarp tissues, as

well as juice parameters of all varieties collected from the Waite vineyard are detailed in Table 2.1.

Shiraz showed the highest potassium concentration of all varieties in all tissues sampled (apoplast,

mesocarp, and skin) (Table 2.1). In Chenin Blanc, [Ca]apo and Ca2+ activity were relatively high (Figure

2.4A, C). Thompson Seedless demonstrated much higher [Ca]apo and Ca2+ activity than both Shiraz and

Grenache (Figure 2.4A, C). Both the mesocarp and apoplast of Thompson Seedless had a high

calcium concentration relative to other varieties (Table 2.1).

To assess the level of cellular breakdown (and contamination or dilution of apoplastic samples with

cellular fluid) the ratio of [K]mesocarp: [K]apo was calculated. No differences in this ratio were observed

between Thompson Seedless, Grenache and Shiraz (Table 2.1). The [K]mesocarp: [K]apo ratio in Chenin

Blanc from Waite vineyard was much lower than other varieties and also lower than Chenin Blanc from

the Barossa (Table 2.2), suggesting that some cellular breakdown may have been occurring.

Additionally, the variation in [K]apo, [Ca]apo, apoplastic Ca2+ activity, and apoplastic pH measurements

was high in Waite Chenin Blanc, suggesting an effect upon some berries but not others. Some Botrytis

infection was observed in Waite Chenin Blanc samples, and this may have been a factor.

Barossa samples showed higher calcium concentrations in apoplast fluids, mesocarp, and skin tissues

than Waite samples (Table 2.2). An exception to this was in Chenin Blanc, where the occurrence of

some Botrytis infection in Waite samples may have been responsible for elevated apoplast calcium

level, resulting in no significant difference being detected between sites. Additionally, the differences in

calcium levels between sites were not significant in Grenache apoplast and mesocarp samples,

although significant differences in the skin were observed (Table 2.2). This indicates that Grenache

either exhibits less site sensitivity in calcium uptake or greater translocation of calcium to the skin than

Shiraz. Potassium accumulation is usually correlated with sugar concentration over development;

Chenin Blanc demonstrated higher [K]mesocarp and sugar concentration in Barossa samples.

Interestingly, Barossa Shiraz demonstrated lower [K]mesocarp and higher sugar concentration than Waite

Shiraz. Sodium levels were also consistently higher in Barossa samples, indicating the soil and/or water

conditions here had an impact on uptake of sodium into the berry skins, mesocarp and apoplast. No

36

significant differences in berry juice pH within varieties were detected between sites. The only variety to

show any significant differences between sites in apoplastic pH or apoplastic Ca2+ activity was Shiraz,

with lower apoplastic pH and higher apoplastic Ca2+ activity in Barossa samples (Table 2.2).

Similar patterns were seen between Waite samples and Barossa samples with regards to skin and

berry physical properties (Table 2.2). At both sites, Grenache had the hardest berries and skins and the

highest Young’s Modulus (i.e. least elastic) berries and skins. Shiraz demonstrated the softest berries,

and also showed the lowest Young’s Modulus (i.e. most elastic) berries and skins. Differences in skin

thickness measured by the Instron were only apparent in Waite samples; the Botrytis infection of Waite

Chenin Blanc may have resulted in a lower calculated skin thickness.

Figure 2.4. Varietal differences in apoplastic ion composition from Waite vineyard berries. Total calcium

(A) and potassium (B) concentration (mM) in apoplastic fluid measured by ICP-AES. Free calcium ion

activity (mM Ca2+) (C) and pH (D) in apoplastic fluid measured with ISE. Different colours indicate

different varieties (Shiraz; red, Grenache; purple, Chenin Blanc; yellow, Thompson Seedless; green,

Chardonnay; white). No ICP-AES samples were collected for Chardonnay due to early ripening and

time constraints. Mean + SD. Columns with different notation are significantly different using one way

analysis of variance and a Tukey posthoc test at 95 % CI.

0

1

2

3

Ap

op

las

t c

alc

ium

co

nc

en

tra

tio

n (

mM

)

a

b b

a

A

0

1 0

2 0

3 0

4 0

5 0

Ap

op

las

t p

ota

ss

ium

co

nc

en

tra

tio

n (

mM

) a a

bb

B

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

Ap

op

las

t fr

ee

ca

lciu

m

(Ca

2+

) a

cti

vit

y (

mM

)

ccb c

a

a b

C

0

1

2

3

4

5

Ap

op

las

t p

H

a b ba bab

D

37

Shiraz Grenache Chenin Blanc Thompson Seedless Chardonnay

Mean SD Mean SD Mean SD Mean SD Mean SD

Berry juice parameters Brix 22.3 b 1.4 26.6 a 1.0 22.4 b 1.5 22.0 b 1.6 23.6 b 0.7

pH 3.23 b 0.13 3.40 a 0.09 3.25 b 0.12 3.19 b 0.13 3.19 b 0.07

TA 0.62 c 0.06 0.68b c 0.04 0.63 c 0.07 0.76 ab 0.10 0.84 a 0.07

Apoplast pH ISE pH 4.18 ab 0.12 4.26 a 0.15 4.11 ab 0.34 3.94 b 0.19 3.99 b 0.06

Apoplast Ca2+ activity ISE

(mM)

Ca2+ 0.60 c 0.18 0.91 bc 0.25 1.23 ab 0.60 1.65 a 0.58 0.64 c 0.22

Apoplast fluid ICP-AES

(mM)

B 0.66 a 0.21 0.41 b 0.09 0.59 ab 0.13 0.44 b 0.14

Ca 1.44 b 0.58 1.39 b 0.27 2.21 a 0.61 2.22 a 0.48

Mg 3.17 a 0.38 2.18 b 0.23 3.49 a 0.74 3.06 a 0.30

Na 0.61 ab 0.25 0.69 a 0.13 0.45 b 0.11 0.40 b 0.10

K 40.2 a 2.0 27.9 b 3.0 37.1 a 5.9 27.2 b 9.3

P 3.39 a 0.84 0.79 b 0.20 2.94 a 0.53 4.16 a 2.17

S 1.13 a 0.23 0.70 b 0.11 0.74 b 0.16 0.94 ab 0.57

Mesocarp tissue ICP-AES

(mM)

B 0.66 a 0.19 0.39 b 0.08 0.49 b 0.08 0.51 ab 0.09

Ca 1.64 b 0.41 1.28 bc 0.38 0.85 c 0.17 2.39 a 0.77

Mg 3.08 b 0.31 2.90 b 0.34 2.27 c 0.49 3.67 a 0.27

Na 0.76 a 0.38 0.68 a 0.09 0.35 b 0.10 0.28 b 0.18

K 55.1 a 2.6 42.7 b 3.9 39.7 b 2.6 37.8 b 9.7

P 5.04 b 0.63 2.17 c 0.27 4.21 b 0.54 6.88 a 2.30

S 2.51 a 0.27 1.77 b 0.32 1.60 b 0.18 1.91 b 0.56

Fe 0.030 b 0.003 0.045 b 0.005 0.026 b 0.006 0.208 a 0.048

Cu 0.025 a 0.002 0.007 c 0.002 0.016 b 0.003 0.019 b 0.005

Zn 0.007 a 0.001 0.007 a 0.002 0.004 b 0.001 0.007 a 0.002

Skin tissue ICP-AES (mM) B 2.83 a 0.90 1.84 b 0.28 1.87 b 0.38 2.35 ab 0.44

Ca 10.5 ab 1.9 12.2 a 2.1 7.8 c 1.8 8.1 bc 2.2

Mg 5.35 bc 1.57 8.50 a 0.72 4.93 c 0.17 6.42 b 1.08

Na 0.49 a 0.28 0.34 ab 0.07 0.15 b 0.03 0.39 ab 0.29

K 238 a 29 203 bc 15 213 ab 20 180 c 18

P 17.6 b 3.9 6.4 c 1.1 14.8 b 2.2 22.6 a 2.8

S 9.75 a 2.15 6.94 b 0.74 6.32 b 0.87 6.19 b 1.47

Fe 0.155 ab 0.043 0.198 a 0.023 0.110 b 0.015 0.171 ab 0.087

Cu 0.073 a 0.012 0.035 b 0.006 0.041 b 0.009 0.054 b 0.028

Zn 0.029 ab 0.005 0.030 ab 0.0047 0.017 b 0.005 0.044 a 0.041

[K]mesocarp:[K]apoplast 1.37 a 0.09 1.55 a 0.25 1.09 b 0.14 1.42 a 0.16

Table 2.1. Varietal differences in berry juice, apoplast, mesocarp and skin composition at harvest

maturity from Waite vineyard berries. No ICP-AES samples were collected for Chardonnay due to early

ripening and time constraints. Each bunch value determined as mean value from five individual berry

measurements, or single measurement of five or ten pooled berries, depending on variable (see

Materials and Methods). Bunch mean and SD from n=10 replicate vines. Columns with different

notation are significantly different using one way analysis of variance and a Tukey posthoc test

(P<0.05).

38

Sh

iraz

G

ren

ach

e

Ch

enin

Bla

nc

Site

1

Site

2

Sign

ific

ant

dif

fere

nce

(P

<0

.05

)

Site

1

Site

2

Sign

ific

ant

dif

fere

nce

(P

<0

.05

)

Site

1

Site

2

Sign

ific

ant

dif

fere

nce

(P

<0

.05

) M

ean

SD

M

ean

SD

M

ean

SD

M

ean

SD

M

ean

SD

M

ean

SD

Ber

ry ju

ice

pa

ram

ete

rs

Bri

x 2

2.3

1

.4

24

.0

0.8

*

2

6.6

1

.0

28

.3

0.9

*

2

2.4

1

.5

24

.5

1.5

*

pH

3

.24

0

.14

3

.28

0

.07

3.4

0

0.0

9

3.3

4

0.0

7

3

.25

0

.12

3

.35

0

.07

TA

0.6

2

0.0

5

0.9

2

0.0

5

*

0.6

8

0.0

4

0.6

9

0.0

5

0

.63

0

.07

0

.97

0

.07

*

Ap

op

last

pH

ISE

p

H

4.1

8

0.1

2

3.8

3

0.0

2

*

4.2

6

0.1

5

4.1

9

0.3

5

4

.12

0

.34

4

.10

0

.15

Ap

op

last

Ca2

+ act

ivit

y IS

E (m

M)

Ca2

+ 0

.60

0

.18

1

.72

0

.48

*

0

.91

0

.25

1

.01

0

.33

1.2

3

0.6

0

1.1

5

0.2

6

Ap

op

last

flu

id IC

P-A

ES (

mM

) B

0

.66

0

.21

0

.52

0

.08

*

0

.41

0

.09

0

.37

0

.07

0.5

9

0.1

3

0.5

8

0.0

8

Ca

1.4

4

0.5

8

2.1

6

0.6

7

*

1.3

9

0.2

7

1.8

1

0.5

0

2

.21

0

.61

2

.16

0

.33

Mg

3.1

7

0.3

8

2.9

1

0.2

7

2

.18

0

.22

1

.88

0

.34

3.4

9

0.7

3

3.2

2

0.3

6

Na

0.6

1

0.2

5

0.9

2

0.2

6

0

.69

0

.13

1

.29

0

.49

*

0

.45

0

.11

1

.29

0

.40

*

K

40

.2

2.0

2

7.2

3

.3

*

27

.9

3.0

2

5.5

4

.7

3

7.1

5

.9

32

.4

3.2

*

P

3.3

9

0.8

4

1.8

2

0.4

2

*

0.7

9

0.2

0

1.4

5

0.4

1

*

2.9

4

0.5

3

1.9

7

0.5

8

*

S 1

.12

9

0.2

27

1

.06

0

0.2

30

0.6

99

0

.10

6

0.8

39

0

.19

0

0

.73

8

0.1

63

1

.40

3

0.2

32

*

Mes

oca

rp IC

P-A

ES (

mM

) B

0

.66

0

.19

0

.53

0

.11

*

0

.39

0

.08

0

.47

0

.04

0.4

9

0.0

8

0.5

1

0.0

6

Ca

1.6

4

0.4

1

2.2

8

0.4

5

*

1.2

8

0.3

8

1.5

1

0.3

8

0

.85

0

.17

1

.33

0

.24

*

Mg

3.0

8

0.3

1

3.2

6

0.4

3

2

.90

0

.34

3

.01

0

.25

2.2

7

0.4

9

2.7

3

0.1

9

*

Na

0.7

6

0.3

7

1.4

0

0.5

4

*

0.6

8

0.0

9

1.8

0

0.8

9

*

0.3

5

0.1

0

1.4

4

0.5

3

*

K

55

.1

2.6

4

3.6

8

.8

*

42

.7

3.9

4

0.5

2

.8

3

9.7

2

.6

47

.7

5.7

*

P

5.0

4

0.6

3

4.2

5

0.9

7

*

2.1

7

0.2

7

3.9

8

0.6

0

*

4.2

1

0.5

4

4.0

2

0.6

9

S 2

.51

0

.27

2

.77

0

.27

1.7

7

0.3

2

2.0

6

0.2

0

1

.60

0

.18

2

.83

0

.39

*

Fe

0.0

30

0

.00

3

0.0

27

0

.00

5

0

.04

5

0.0

05

0

.04

0

0.0

05

0.0

26

0

.00

6

0.0

37

0

.00

4

*

Cu

0

.02

5

0.0

02

0

.01

4

0.0

02

*

0

.00

7

0.0

02

0

.01

1

0.0

01

*

0

.01

6

0.0

03

0

.01

1

0.0

02

*

Zn

0.0

07

0

.00

1

0.0

06

0

.00

2

0

.00

8

0.0

02

0

.00

5

0.0

02

*

0

.00

4

0.0

01

0

.00

6

0.0

02

*

Skin

tis

sue

ICP

-AES

(m

M)

B

2.8

3

0.9

0

1.8

1

0.3

9

*

1.8

4

0.2

8

1.9

3

0.2

6

1

.87

0

.38

1

.68

0

.22

Ca

10

.5

1.9

1

3.8

1

.7

*

12

.2

2.1

1

8.3

2

.9

*

7.8

1

.8

10

.3

2.1

*

Mg

5.4

1

.6

5.9

0

.8

8

.5

0.7

1

1.4

2

.0

*

4.9

0

.2

5.8

0

.8

Na

0.4

9

0.2

8

0.8

1

0.3

9

0

.34

0

.07

1

.29

0

.96

*

0

.14

0

.03

0

.96

0

.32

*

K

23

8

29

1

71

2

6

*

20

3

15

1

99

2

1

2

13

2

0

20

6

38

P

17

.6

3.9

2

4.1

5

.0

*

6.4

1

.1

17

.3

2.6

*

1

4.8

2

.2

16

.7

2.7

S 9

.75

2

.15

9

.71

1

.05

6.9

4

0.7

4

7.7

4

0.8

3

6

.32

0

.87

9

.31

1

.43

*

Fe

0.1

55

0

.04

3

0.1

65

0

.01

8

0

.19

8

0.0

23

0

.17

6

0.0

55

0.1

10

0

.01

5

0.1

46

0

.02

8

Cu

0

.07

3

0.0

12

0

.03

7

0.0

05

*

0

.03

5

0.0

06

0

.03

1

0.0

04

0.0

41

0

.00

9

0.0

29

0

.00

5

*

Zn

0.0

29

0

.00

5

0.0

17

0

.00

4

*

0.0

30

0

.00

5

0.0

18

0

.00

5

*

0.0

17

0

.00

5

0.0

15

0

.00

4

[K]m

eso

carp

:[K

]ap

op

last

1.3

7

0.0

9

1.6

3

0.4

1

1

.55

0

.25

1

.65

0

.41

1.0

9

0.1

4

1.4

8

0.1

6

*

Ber

ry h

ard

nes

s

J 0

.01

1

0.0

03

0

.01

6

0.0

02

*

0

.02

4

0.0

04

0

.02

4

0.0

03

0.0

14

0

.00

3

0.0

19

0

.00

3

*

Ber

ry Y

ou

ng’

s m

od

ulu

s

Nm

m-1

0

.13

3

0.0

37

0

.17

0

0.0

27

0.2

07

0

.02

7

0.3

52

0

.03

1

*

0.1

82

0

.03

2

0.2

53

0

.05

2

*

Skin

har

dn

ess

J

0.0

01

0

0.0

00

3

0.0

02

1

0.0

00

3

*

0.0

04

6

0.0

00

7

0.0

04

6

0.0

00

7

0

.00

10

0

.00

02

0

.00

22

0

.00

02

*

Skin

Yo

un

g’s

mo

du

lus

N

mm

-1

0.0

48

0

.00

5

0.0

51

0

.00

7

0

.08

6

0.0

97

0

.09

8

0.0

13

*

0

.05

9

0.0

10

0

.07

6

0.0

10

*

Skin

th

ickn

ess

µ

M

91

.5

6.5

9

2.5

9

.5

1

03

.5

3.2

9

4.9

5

.5

7

1.5

1

1.9

8

7.0

8

.7

*

Tab

le 2

.2. S

ite s

peci

fic d

iffer

ence

s in

ber

ry ju

ice,

apo

plas

t, m

esoc

arp,

and

ski

n co

mpo

sitio

n at

har

vest

mat

urity

. Com

paris

on o

f dat

a fr

om

Wai

te v

iney

ard

(Site

1)

and

Bar

ossa

vin

eyar

d (S

ite 2

). S

ite 1

dat

a is

rep

eate

d fr

om T

able

2.1

in o

rder

to h

ighl

ight

site

spe

cific

diff

eren

ces

in

nutr

ient

acc

umul

atio

n. E

ach

bunc

h va

lue

dete

rmin

ed a

s m

ean

valu

e fr

om fi

ve in

divi

dual

ber

ry m

easu

rem

ents

, or

sing

le m

easu

rem

ent o

f fiv

e

or te

n po

oled

ber

ries,

dep

endi

ng o

n va

riabl

e (s

ee M

ater

ials

and

Met

hods

). B

unch

mea

n an

d st

anda

rd d

evia

tion

from

n=

10 r

eplic

ate

vine

s.

Diff

eren

ces

wer

e te

sted

usi

ng tw

o w

ay a

naly

sis

of v

aria

nce

and

Sid

ak’s

pos

thoc

test

, * in

dica

tes

sign

ifica

nt d

iffer

ence

(P

<0.

05)

with

in a

varie

ty b

etw

een

site

s ac

cord

ing

to S

idak

’s p

osth

oc te

st.

39

2.3.5 Principal component analysis

Principal component analysis of the major berry variables gives a picture of the site and variety

differences in nutrient uptake and maturity, as well as demonstrating correlations between variables

(Figure 2.5 A). Shiraz samples separate strongly by site along PC-1, mostly due to differences in tissue

K and Ca levels. Grenache samples separated from all other samples on PC-2, due to the higher

maturity and skin calcium levels found in these samples. Instron measurements were included in the

PCA analysis (Figure 2.5 B) to examine relationships between berry and skin strength, berry maturity,

and tissue calcium levels. PC-1 explains 47% of the variation and is largely comprised of potassium

concentration measurements from mesocarp, skin and apoplast clustering together on one end of the

axis, and Instron measurements (berry/skin hardness and YM, skin thickness), berry maturity measures

(brix and pH), and skin calcium concentration at the other extreme. PC-2 explains 28% of the variation,

demonstrating a strong correlation between TA, [Ca2+]apo, [Ca2+]mesocarp, and apoplastic Ca2+ activity;

these measures demonstrate a strong negative correlation with apoplastic pH. Shiraz samples show

the strongest separation by site, with Barossa Shiraz showing high [Ca2+]apo and apoplastic Ca2+

activity. Grenache samples showed the least separation by site; this may be due to the consistently

high Instron values, maturity measures, and skin calcium concentration that Grenache demonstrated at

both sites (Table 2.2). Inclusion of boron measurements in PCA analysis (Figure 2.5 C) revealed that

boron concentrations in all tissues largely correlate with potassium concentrations, as would be

expected for nutrients that follow similar paths of phloem-fed berry uptake. Interestingly, TSS does not

correlate with potassium and boron concentrations, suggesting that independent cellular transport

mechanisms operate for accumulation of phloem mobile ions and sugars. Inclusion of sodium

measurements in PCA analysis (Figure 2.5 D) revealed that sodium concentration is very highly

correlated in all tissues; [Na+]mesocarp, [Na+]skin, and [Na+]apo, all clustered together. Inclusion of sodium

measurements in the analysis also provided very clear separation of varieties by site; indicating that

berry sodium uptake is highly site dependent.

40

A C

B D

Fig

ure

2.5.

Bi-p

lot o

f sco

res

(in b

lue)

and

var

iabl

e lo

adin

gs (

in r

ed)

for

prin

cipa

l com

pone

nts

of b

erry

var

ieta

l sur

vey

data

. PC

A p

erfo

rmed

usi

ng a

set

of

core

var

iabl

es (

juic

e m

easu

rem

ents

, tis

sue

calc

ium

and

pot

assi

um c

once

ntra

tions

) (A

), a

s w

ell a

s w

ith a

dditi

onal

Inst

ron

mea

sure

men

ts (

B),

bor

on

mea

sure

men

ts (

C),

and

sod

ium

mea

sure

men

ts (

D).

Dat

a re

pres

ents

sco

res

(in r

ed)

of 3

var

ietie

s (S

hira

z, G

rena

che

and

Che

nin

Bla

nc)

from

two

diffe

rent

si

tes

(site

1: W

aite

, site

2: B

aros

sa).

Var

iabl

e lo

adin

g (in

blu

e); a

po =

apo

plas

t, m

eso

= m

esoc

arp

(e.g

. pH

apo

= A

plop

last

pH

, fre

e C

a ap

o =

apo

plas

t fr

ee C

a2+ a

ctiv

ity, C

a ap

o =

apo

plas

t to

tal [

calc

ium

], C

a m

eso

= m

esoc

arp

tota

l [ca

lciu

m])

. Oth

er v

aria

bles

incl

ude;

K =

pot

assi

um, B

= b

oron

, Na

=

sodi

um, p

H =

ber

ry ju

ice

pH, T

A =

ber

ry ju

ice

titra

tabl

e ac

idity

, T

SS

= b

erry

juic

e T

SS

(°

Brix

), B

erry

YM

= b

erry

You

ng’s

mod

ulus

, Ber

ry h

ard

= b

erry

ha

rdne

ss, S

kin

hard

= s

kin

hard

ness

, Ski

n th

ick

= s

kin

thic

knes

s.

41

2.4 Discussion

2.4.1 Berry phenolics

In Shiraz grapes, the level of total phenolics in the skin has been observed to increase from veraison to

reach a maximum at 110-120 days after anthesis (Ristic and Iland, 2005). The composition of the cell

wall has an impact on extractability of an important group of pigmented phenolics called anthocyanins

(which are largely located in skin cell vacuoles), which subsequently impact wine colour (Ortega-

Regules et al., 2008). Indeed, skin mechanical properties have been linked to anthocyanin

extractability, with berries possessing harder skins demonstrating higher anthocyanin extractability

(Rolle et al., 2012b). When looking at histological sections of mature grapes, it would be expected to

see more phenolics (including anthocyanin) in the skin cells. High levels of skin phenolics were

detected in Shiraz berry tissue harvested at maturity and stained with toluidine blue O (Figure 2.1A).

Different phenolic staining patterns were observed between the epidermal, hypodermal, and outer

mesocarp cells of different varieties. Large inclusions were only found in the hypodermal cells of the

two red grape varieties, Shiraz and Grenache (Figure 2.1A, B). These red varieties also exhibited a

variety of small inclusions, granulations, and uniformly stained cells throughout the tissue (Figure 2.1A,

B). The white wine varieties Chenin Blanc and Chardonnay exhibited mostly uniformly staining cells in

the hypodermis and mesocarp, with some epidermal cells containing granular and small inclusions

(Figure 2.1C, E). This contrasts to the white table grape variety Thompson Seedless, which showed

very little staining in all cell types (Figure 2.1D). It has been shown that the occurrence of different

forms of phenolic deposits changes over the course of berry development and also in response to

water availability and environmental differences between sites (Cadot et al., 2011). As the samples

collected for this study were all from the same site and season, we can also conclude that the

occurrence of different phenolic forms varies greatly between varieties, with cells containing large

inclusions only occurring in red varieties, white wine varieties exhibiting cells containing mostly fine

granular or uniform staining, and the white table grape variety exhibiting very little phenolic staining.

2.4.2 Berry cell morphology

Varieties also showed differences in the morphology of epidermal, hypodermal and outer mesocarp cell

layers. Most noticeably, Shiraz demonstrated 1-2 layers of large, rounded epidermal cells, whereas

most other varieties had 1-2 layers (Chardonnay had 4 layers) of small, elongate epidermal cells

(Figure 2.1). A previous survey of table grape varieties found a correlation between the number and

thickness of epidermal and hypodermal cell layers and varietal resistance to Botrytis cinerea (Gabler et

42

al., 2003). This previous study also found correlations between cuticle and wax content and resistance,

but no relationship between phenolic content and Botrytis resistance (Gabler et al., 2003). Thompson

Seedless was identified as a variety highly susceptible to Botrytis, with relatively few epidermal and

hypodermal cell layers, and relatively low total skin thickness. Comparison of the skin cell morphology

of Thompson Seedless (Figure 2.1D, I) with the wine grape varieties revealed Thompson Seedless to

have the thinnest skins and the least epidermal and hypodermal cell layers. The diversity of

morphology in epidermal and hypodermal cell layers of the wine grapes may be one of many factors

determining their susceptibility to Botrytis. In the present study, some Botrytis infection was found in

Waite Chenin Blanc. However, many other factors are important in determining infection rates in the

field, so this should not be taken as an indicator of Chenin Blanc susceptibility to Botrytis per se.

2.4.3 Berry pectin distribution

Pectic polysaccharides are major components of grape berry cell walls; higher concentrations of the

basic pectin subunit galacturonic acid have been found in skin cell walls compared to mesocarp cell

walls (Vidal et al., 2001). However, mesocarp cell walls contain more soluble pectin than skin cell walls,

indicating that although pectins are more prevalent in the skin, they are also more tightly bound to the

cell wall matrix (Vidal et al., 2001). Pectin is secreted into the cell wall in an esterified form, the action of

pectin methyl esterases (PMEs) de-esterifies pectin, allowing cross linking of adjacent pectin by

calcium. By studying the patterns of pectin distribution in skin and outer mesocarp cell walls using

antibody labelling specific for esterified and de-esterified pectin, it is possible to unravel some of the

complex differences between grape varieties relating to changes in the properties of cell walls during

ripening.

LM19, which binds de-esterified pectin, was localised most strongly to the middle lamella in all varieties

(Figure 2.2). The disassembly of the cell wall matrix and weakening of intercellular bonds in the middle

lamella has previously been identified in ripening berries (Huang et al., 2005). The middle lamella of

fruit cell walls is believed to consist mostly of homogalacturonan and structural proteins, its relatively

simple polysaccharide composition compared to the primary cell wall suggests that the role of pectin

modification and binding in this space is crucial for determining cell adhesion (Brummell, 2006). The

strong punctate pattern of de-esterified pectin observed in the middle lamella of Grenache and

Thompson Seedless berries may indicate areas of high PME activity where adjacent cells are anchored

together by crosslinking of de-esterified pectin (Figure 2.2B, D). These areas may be plasmodesmatal

pit fields between adjacent cells. Pit fields of tomato fruit contain high levels of homogalacturonan

(Orfila and Knox, 2000). Another study identified high levels of PME epitope localisation to the

43

plasmodesmata in flax hypocotyl, suggesting that pectins in these domains are highly de-esterified

(Morvan et al., 1998). Similar results have been found in potato cortex and kiwifruit, where high levels

of Ca2+-pectin crosslinks were hypothesised to act as “spot-welds” protecting plasmodesmata from

damage during cell expansion (Bush and McCann, 1999; Sutherland et al., 1999). The more diffuse

banded pattern of de-esterified pectin distribution observed in Shiraz, Chenin Blanc, and Chardonnay

indicates that long strips of middle lamella are de-esterified along lateral cell walls and in tricellular

junctions, leading to cell-cell binding along the length of the cell boundaries (Figure 2.2A, C, E). Cellular

junctions have been previously identified as zones of high pectin de-esterification (Morvan et al., 1998;

Bush and McCann, 1999).

LM20 binds esterified pectin epitopes and differences between varieties were less clear using this

antibody tag. Some varieties (Shiraz, Grenache, and Thompson Seedless) showed slightly stronger

staining for LM20 (relative to the cell wall as a whole) in the middle lamella and transverse cell walls

(Figure 2.2F, G, I). This concurs with the high concentration of homogalacturonan found in the middle

lamella (Brummell, 2006; Hyodo et al., 2013); the preferential de-esterification of pectin observed in the

lateral walls of some varieties would logically leave relatively higher concentrations of esterified pectin

in the transverse walls. Particularly low amounts of esterified pectin staining was observed in

Chardonnay (Figure 2.2J), mostly in the transverse cell walls and as patches in the lateral primary cell

wall. This suggests that the large majority of pectin is de-esterified in this variety, leaving small areas

that have not been exposed to PME activity. The variety-specific nature of pectin distribution and

modification may help explain some ripening phenomena and deserves greater investigation.

Different patterns of pectin de-esterification have been observed using LM19 and LM20 antibodies on

tomato tissues (Hyodo et al., 2013). High pectin densities were observed in the skin, with the level of

de-esterification increasing throughout ripening. Increased expression of tomato PME1 in the skin

during the mature green stage is believed to be responsible for increased pectin de-esterification,

formation of Ca2+-pectin cross-links and maintenance of skin firmness (Brummell and Harpster, 2001;

Hyodo et al., 2013). The polygalacturonase β-subunit binds polygalacturonase PG2 to modify patterns

of pectin metabolism. Suppression of the polygalacturonase β-subunit therefore results in higher

amounts of polygalacturonide formation and increased softening during ripening phases. During normal

ripening, this polygalacturonase β-subunit is believed to suppress PG2 activity and pectin solubilisation

in the middle lamella, which leads to increased cell adhesion (Brummell and Harpster, 2001). Spatial

and temporal differences in pectin modification, degradation activity, and calcium localisation combine

to endow each tissue with different physical traits throughout development.

44

2.4.4 Berry physical changes

Grenache is a late ripening variety that is capable of attaining very high sugar levels during extended

on-vine maturation. The ability of Grenache to ripen slowly over long periods may be primarily

attributable to its skin traits, compared to Thompson Seedless where maintenance of mesocarp vitality

(Fuentes et al., 2010), membrane viability and high apoplastic calcium may play a role. The relatively

thick skins of Grenache, as observed by microscopy (Figure 2.1B) and measured using the Instron

(Figure 2.3E), as well as the high skin hardness (or skin break energy; Figure 2.3C), may contribute to

reduced water loss, even in the case of hydraulic isolation from the vine. Grenache has previously been

identified as a variety with high resistance to splitting due to the physical properties and skin thickness

of the berries (Giacosa et al., 2013). Another survey correlated higher skin break force of fresh berries

with slower berry dehydration rates off-vine (Giacosa et al., 2012). The table grape Thompson Seedless

may be an exception to this since it has low skin hardness (Figure 2.3C) and a relatively thin single

epidermal layer skin (Figure 2.1D, I) but good storage qualities. The low shrivel susceptibility of this

variety is reflected in the low elasticity (i.e. high firmness) measurements of berry and skin (Figure 2.3B,

D). The ability of Thompson Seedless to maintain cell vitality (Fuentes et al., 2010), membrane integrity,

hydraulic conductance into the berry (Tilbrook and Tyerman, 2009), low apoplastic pH, and high

apoplastic calcium, is likely to be a major contributing factor to the berry hardness (or firmness)

observed even during extended vine maturation.

2.4.5 Berry apoplast interactions

During grape development, the pH of berry pulp juice (composed of mainly vacuolar fluids) increases

from ~2.5 at green berry stage, to ~3.5 at harvest maturity, to ~4.0 when overripe. Berry apoplast pH is

typically higher (~4.5-5 during development), indicating a separation of compartments and regulation of

proton gradients between the apoplast, cytosol, and vacuole (Wada et al., 2008; Keller and Shrestha,

2014). In wine grapes maintenance of the proton gradient is seen to break down during late ripening as

pH values for apoplast extracts approach that of pulp juice (Wada et al., 2008; Keller and Shrestha,

2014). During berry development, active control of apoplast pH is primarily through plasma membrane

H+-ATPase activity. Auxin activated calcium transients are believed to promote this transport activity;

with the resulting decrease in apoplastic pH activating cell wall loosening and modifying enzymes

(Wang and Ruan, 2013). During late ripening or pathogen infection, extensive disassembly of the cell

wall may lead to destabilisation of the plasma membrane and/or loss of cell vitality. A more detailed

time course to pinpoint the drop in apoplastic pH and the onset of membrane porosity increases or cell

vitality loss would determine if this is the case.

45

Chenin Blanc at both sites demonstrated higher [Ca2+]apo and apoplast Ca2+ activity than the red

varieties Shiraz and Grenache (Figure 2.4 A, C). Chenin Blanc had mesocarp calcium concentration

that was lower than other varieties, indicating that the observed high apoplastic calcium may have been

a result of increased cell wall degradation and/or increased allocation of berry calcium into the apoplast

rather than higher overall berry calcium uptake (Table 2.1). Previously, knockout of Arabidopsis

vacuolar calcium transporters CAX1 and CAX3 has been shown to result in higher apoplastic calcium

(Conn et al., 2011). Reduced expression of vacuolar calcium transporters in Chenin Blanc may explain

the observed pattern of calcium distribution. Some Botrytis infection was observed in Chenin Blanc

Waite samples. Breakdown of cell wall components and membrane leakage caused by fungal enzymes

may help to account for the high level of apoplastic calcium and low ratio of [K]mesocarp: [K]apo observed

in these samples (Table 2.2). Further investigation would be required to confirm whether this was a

Botrytis effect or simply that the Waite Chenin Blanc samples had a different pattern of potassium and

calcium allocation compared to the other samples. Botrytis susceptibility testing of Shiraz and Chenin

Blanc berries was carried out to probe the effect of different levels of calcium nutrition on cell wall

composition (Chapter 5). It has been previously established that differences in endogenous cell wall

disassembly can affect Botrytis susceptibility in grapes (Cantu et al., 2008), and that varieties differ in

activity levels of a variety of cell wall enzymes, translating to dynamic changes in cell wall composition

throughout development (Nunan et al., 2001; Ortega-Regules et al., 2008; Zoccatelli et al., 2013).

Thompson Seedless also exhibited high [Ca]apo and Ca2+ activity, although it also had a high [K]mesocarp:

[K]apo and a low apoplastic pH. These relationships were further probed using principal component

analysis.

The pH of the apoplast fluid is normally lower than that of bulk fluids obtained from fruit (composed of

predominantly vacuolar fluid) (Ruan et al., 1996; Almeida and Huber, 1999; Wada et al., 2006). It has

been observed in tomato that apoplastic pH decreases during ripening, approaching that of the bulk

pericarp and locule tissue extract (Almeida and Huber, 1999). As the fruit ripens the compartmentation

between apoplast and symplast begins to break down; this may be due to changes in membrane

permeability (Lang and During, 1991; Almeida and Huber, 1999). Whether this drop is a trigger for the

breakdown of compartmentation or a result of it remains unclear. The decrease in apoplast pH in

tomato occurs simultaneously with the cessation of xylem calcium import and increasing PME activity in

the apoplast; low pH in the apoplast would also serve to increase competition between free Ca2+ and H+

ions for binding of de-esterified pectin. These factors may result in a negative correlation between

apoplast pH and available Ca2+ in the apoplast. This relationship becomes apparent in grapes when

PCA analysis is performed on the present data; apoplastic pH and apoplastic Ca2+ activity are strongly

46

negatively correlated, as shown in Figure 2.5A. Regression analysis of apoplastic pH and apoplastic

Ca2+ activity data from all varieties across both sites confirmed this negative correlation (data not

shown).

Whilst the impacts of unaccounted for and potentially confounding variables between the two sites may

reduce the possibility of drawing strong conclusions about site-specific effects, some interesting

observations were made. Shiraz was the only variety to show any significant difference between sites in

apoplastic pH and apoplastic Ca2+ activity (Table 2.2). Barossa Shiraz had the lowest apoplastic pH

and the highest apoplastic Ca2+ activity measured. Both Waite and Barossa Shiraz had higher

mesocarp calcium concentrations than Grenache and Chenin Blanc from the same site. The high

calcium conditions at the Barossa site may have exceeded the capacity for Shiraz to redistribute xylem-

derived calcium from the apoplast to the vacuole or to bind calcium in the cell wall, resulting in much

higher levels of free Ca2+ in the apoplast compared to Waite Shiraz. Potentially, Shiraz exhibits cell

walls with a low cation exchange capacity; under high calcium conditions, high levels of calcium

accumulate in the vacuole through the action of Ca2+-H+ exchangers and H+-ATPase activated calcium

channels in the PM and the tonoplast. This may result in accumulation of H+ occurs in the apoplast,

acidifying the cell wall, and increasing binding competition with pectin bound Ca2+ and further

increasing free Ca2+ levels. Conversely, the relatively low apoplastic pH may indicate that plasma

membrane and tonoplast breakdown is already occurring, and that collected apoplastic fluids are

contaminated with vacuolar calcium.

2.5 Conclusion

Clear differences in skin cell morphology between red, white and table grape varieties were revealed by

light microscopy. Varieties with high skin break energy (e.g. Grenache) generally have higher [Ca]skin

than varieties with weaker skin (e.g. Thompson Seedless). Thompson Seedless exhibits the hardest

grapes, highest [Ca]mesocarp and highest [Ca]apo, probably related to maintenance of hydraulic

conductivity and cell vitality (Tilbrook and Tyerman, 2009; Fuentes et al., 2010). In samples where

some Botrytis infection was observed, [Ca]apo was difficult to estimate accurately, possibly due to cell

wall degradation and loss of cell vitality. These results demonstrate that differences between grape

varieties in Ca2+ uptake, cell wall composition, pectin modification, membrane porosity and/or cell

vitality can be linked to physical and qualitative traits (Rogiers et al., 2006a; Tilbrook and Tyerman,

2009; Fuentes et al., 2010). Interesting differences between sites indicate that some varieties do

47

respond differently to calcium nutrition; however a more controlled system is required to test these

impacts. Further studies to ascertain the effects of different levels of calcium nutrition on bunch quality

and cell wall properties including pectin modification are required to understand these relationships.

48

Chapter 3 Grapevine calcium nutrition

3.1 Introduction

The effect of varying grapevine calcium nutrition on vine function has not been previously explored. By

designing a system in which calcium availability is manipulated, whilst minimising potential confounding

changes in availability of other nutrients, the impact of differing calcium supply on the vine was

investigated. By studying the effects of calcium upon grapevine growth and berry development, it

becomes possible to identify the physiological and developmental signs indicating that calcium nutrition

may be sub-optimal. This may provide opportunities for ameliorative steps to improve fruit quality. In

this chapter, the design and optimisation of a hydroponics system for probing vine calcium relations

using a modified fruiting cutting method will be described. The effects of high and low calcium supply

were investigated using this system. Changes in petiole nutrient profiles, vine phenology, root and

shoot function, reproductive performance, and berry development were all assessed. Shiraz was

studied as it is highly susceptible to berry shrivel; Chenin Blanc was also studied as it has a very low

susceptibility to berry shrivel. New data is presented here of the petiole nutrient accumulation of Chenin

Blanc under different calcium regimes.

3.2 Materials and Methods

3.2.1 Fruiting cuttings; Mullins method modifications

Cuttings were collected from Waite Campus Vineyard (Shiraz) and Barossa Valley Vineyard (Chenin

Blanc). Cuttings were 40-70 cm length and contained 4-7 nodes. An angled cut was made at the distal

end to indicate orientation. Cuttings were stored in sealed plastic bags in a dark cool room (4 °C) for up

to 12 months.

Fruiting cuttings were produced according to a modified version of the Mullins protocol (Mullins and

Rajasekaran, 1981). The modifications to this method were published in Baby et al. (2014); Appendix

3). Briefly, when planting stored cuttings, a fresh cut was made 2-3 cm from the proximal end, and the

fresh cut end was dipped in Clonex Green Hormone Gel for softwood cuttings (Growth Technology Ltd.,

Devon, England) to a depth of ~5 cm, including the first node. These cuttings were planted to a depth of

10-15 cm in a heated sand bed located in a dark cool room (4 °C). Heat bed temperature at cutting end

depth was maintained at 25-28 °C. Heat beds were watered twice a week to maintain moisture content

in the sand. Callus and root development in the heat beds continued for 6-8 weeks. This enabled

49

extensive root development and rapid establishment of bud burst and aerial development when

transferred to glasshouses or growth rooms. Rooted cuttings were transferred into pots and filled with

perlite.

3.2.2 Hydroponics system design

Full details of modifications to Mullins method with specific reference to the use of 100% perlite media

and automated hydroponics system design are detailed in Appendix 3.

3.2.2.1 Perlite growth media

In order to tightly control the nutrient inputs in the hydroponic system, use of 100% perlite media was

pursued as an alternative to 50:50 perlite:vermiculite mix. When subjected to a mildly acidic growth

solution (i.e. pH 5.6), perlite:vermiculite media releases additional Na, Mg, and Mn through cation

exchange (Baby et al., 2014). By comparison, 100% perlite is relatively inert when used as the growth

media (Baby et al., 2014).

3.2.2.2 Nutrient solution design

Hydroponic solutions were based on a modified ¼ strength Hoagland’s mix used for hydroponic growth

of Arabidopsis (Conn et al., 2013). A variety of solutions were designed with the aim of adjusting

available calcium whilst keeping other ion availabilities within the solution relatively constant (and within

the recommended ranges for grapevine nutrition). This was achieved using Visual MINTEQ software

(KTH Royal Institute of Technology, Stockholm, Sweden), that uses an equilibrium speciation model.

Solutions designed were; Low Calcium Solution (LCS), Basal Nutrient Solution (BNS), High Calcium

Solution (HCS). See Appendix 4 for solution compositions.

3.2.2.3 Hydroponics growth issues

The timer circuits used allowed for 15 minute watering intervals on a 24 hour cycle. Concerns about the

lower water holding capacity of perlite (compared to a perlite/vermiculite mix) led to a more intensive

watering schedule being trialled initially. Cuttings were watered for 15 minutes every hour. Although this

watering schedule did not appear to affect the growth of Shiraz, Chenin Blanc, and Thompson

Seedless; Grenache suffered from root rot and epinasty. Grenache is colloquially known as a variety

that does not like to have “wet feet”; this trait was evident in this system. Grenache demonstrated

reduced leaf and bunch turgor and reduced leaf angle over the growing period, eventually drying out

and senescing. Upon inspection, the root systems of affected Grenache vines were brown and

50

decaying. Excess watering had caused root rot, with aerial symptoms being a result of reduced

capacity for water uptake. Watering schedules were adjusted to eliminate this problem. It was observed

that one 15 minute watering interval a day was sufficient to saturate the media for an established root

system, and sufficient moisture was still available for no obvious signs of water stress to be present

after 24 hours. Whilst Thompson Seedless vines appeared to grow healthily in this system, many

bunches failed to set fruit and shrivelled after flowering. The use of gibberellic acid (GA) to encourage

fruitset and increase berry size in Thompson Seedless is a common commercial practice; the unreliable

fruitset of this variety make it unsuitable for these trials. If further work was to be carried out on

Thompson Seedless fruit in hydroponics, the use of GA sprays would need to be investigated.

3.2.3 Biomechanical testing

Harpenden Skinfold calipers HSK-B1 (British Indicators, West Sussex, UK) were used to measure berry

deformability, and subsequently calculate berry Young’s Modulus. A single spring set applying a 110 g

load (i.e. 1.078 N) was used. Due to the lack of stress/strain resolution using this method relative to

Instron methods, Young’s Modulus calculations assumed that during caliper compression the

stress/strain curve was linear in its entirety.

3.2.4 Leaf water potential

A Scholander pressure chamber was used to measure midday leaf water potential. Measurements

were taken between 11:00 am and 2:00 pm. Mature and undamaged leaves were cut from the vine with

a fresh razor blade. The cut petiole was tightly secured protruding through the lid seal of the pressure

chamber. Pressure from a CO2 cylinder was applied and a pressure measurement taken upon visual

observation of sap appearing on the surface of the cut petiole. Two leaves were measured per vine.

3.2.5 Gas exchange measurements

Leaf gas exchange measurements were carried out using an ADC LCpro-SD Infra-red gas analyser

(IRGA) fitted with a small leaf chamber (ADC Bioscientific Limited, Hertfordshire, England). Light

intensity (as photosynthetically active radiation, Qset) was set at 1000 µmol m-2 s-1. CO2 was set at 400

ppm with a CO2 cartridge inserted. Air supply unit flow rate was set at 200 µmol s-1. Fully mature,

undamaged, unshaded leaves were selected; the leaf chamber was clamped halfway along the leaf

adjacent to the mid vein. Three leaves per vine were measured. After attaching to the leaf, sufficient

time was allowed for the chamber conditions to stabilise (usually 2-4 minutes) prior to taking

measurements. This was determined by plotting transpiration rate (E) against time on the IRGA display;

51

after four subsequent stable intervals (with the display updating at 15 second intervals) were observed,

the measurement was taken.

3.2.6 Root hydraulic conductance

Root hydraulic conductance measurements were taken on the root system of decapitated plants with a

hydraulic conductance flow meter (HCFM) (Dynamax, Houston, TX, USA) as detailed in Vandeleur et

al. (2009) and Vandeleur et al. (2014). Briefly, the transient method was used whereby pressure was

ramped and flow into the root system was recorded simultaneously. The protocol was performed as

rapidly as possible to minimise the reduction in conductance that occurs after decapitation of the shoot

system (Vandeleur et al., 2014). The linear region of flow versus pressure was used to calculate

hydraulic conductance. Three consecutive measurements were taken and the average conductance

obtained per root sample. Hydraulic conductance was then normalized by dividing the conductance by

the total root dry weight. The soil was washed from the roots before drying at 60°C for 48 hours.

3.2.7 ICP-AES analysis

Leaf petioles of Chenin Blanc were dried at 70°C for 48 hours. Dried petioles were then ground in an

ESSA LM1-P pulverising mill (FLSmidth, Welshpool, Western Australia). Ground samples were

analysed using ICP-AES by Waite Analytical Services (Urrbrae, South Australia). Elemental

concentrations in dry petiole samples are expressed as % dry weight.

3.2.8 Bunch reproductive measures

Fruitset, Coulure Index and Millerandage Index of Shiraz were calculated as previously described in Dry

et al. (2010), formulas used for the calculations are reproduced below:

Fruitset (%) = (total berry number per bunch / number of flowers per inflorescence) X 100

Coulure Index = 10 – [(no. of seeded berries per bunch + no. of seedless berries per bunch + no. of

LGOs* per bunch) X 10 / no. of flowers per inflorescence]

Millerandage Index = 10 – [(no. of seeded berries per bunch X 10) / (no. of seeded berries per bunch +

no. of seedless berries per bunch + no. of LGOs per bunch)]

*LGO = live green ovary

52

Briefly, developing bunches were enclosed in mesh bags prior to flowering. After completion of

flowering, the bags were removed and flower caps counted. At harvest, berries were divided into berry

classes and counted. The following classes were used; live green ovary (stunted growth, no seeds, less

than 4 mm diameter), seedless berry (no seeds, greater than 4 mm diameter), green berry (contains 1

or more seeds, no red colouration), veraison berry (contains 1 or more seeds, 0 - 75 % berry red

colouration), ripe berry (contains 1 or more seeds, 75-100 % berry red colouration, no visible shrivel),

shrivelled berry (contains 1 or more seeds, 100 % berry red colouration, visible signs of berry shrivel).

3.3 Results

3.3.1 Effect of calcium treatments upon calcium uptake

Vine mineral nutrient composition of Chenin Blanc vines grown hydroponically in Basal Nutrient

Solution (BNS), High Calcium Solution (HCS), and Low Calcium Solution (LCS) was characterised by

ICP-AES analysis of petioles at harvest. The results showed that these treatments pushed calcium

levels outside of the normal range. As calcium nutrition is poorly understood in grapevines, calcium

levels required for deficiency or toxicity were previously undefined. Table 3.1 demonstrates the full suite

of plant mineral nutrients analysed relative to previously published recommended ranges in petioles at

anthesis (Robinson et al., 1997) and petiole measurements at harvest (Smith et al., 2014).

Calcium nutritional requirements of vines are not particularly well documented, with a speculated

adequate range at flowering of 1.2-2.5 % dry weight (DW) (Robinson et al., 1997), and a measured

mean concentration in Chardonnay increasing from 1.35 % DW at flowering to 2.9 % DW at harvest

(Smith et al., 2014). Mean calcium levels observed in these treatments at harvest were as follows; 0.75

% DW in LCS, 3.2 % DW in BNS, and 3.55 % DW in HCS (Figure 3.9A). Whilst the BNS and HCS

levels are above the “adequate” levels at flowering, they are not much higher than that measured in

Chardonnay at harvest, in line with normal increases in leaf petiole calcium concentration over the

course of the growing season. However, the LCS levels measured at harvest are much lower than the

adequate levels suggested at flowering, indicating that these vines were operating under a calcium

deficit throughout the growing season, with minimal calcium being stored in leaf petioles.

Phosphorus is considered adequate in the range 0.25-0.5 % DW. Phosphorus has been shown to

decrease over the growing season in Chardonnay from 0.7 % DW at flowering to 0.2 % DW at harvest.

53

LCS demonstrated adequate mean phosphorous levels at 0.47 % DW, whereas BNS and HCS mean

phosphate levels were high at 0.76 % DW and 1.37 % DW, respectively.

The adequate range for potassium is 1.8-3.0 % DW at flowering. The concentration in Chardonnay at

harvest was 2.3 % DW. The lack of significant differences between treatment means and the closeness

to recommended range (LCS 2.77, BNS 3.15, HCS 2.75 % DW) suggests that potassium levels are

sufficient and well-regulated in this experiment.

Sodium is considered toxic at >5000ppm (i.e. >0.5 % DW). Mean sodium levels in LCS, BNS, and HCS

petioles at harvest were 0.91, 0.62, and 0.25 % DW, respectively. The high levels of sodium in LCS and

BNS may be a concern for growth, or may be indicative of other factors at play; high soil calcium can

reduce sodium uptake, low soil calcium may result in poor root membrane selectivity and high sodium

accumulation. Excess sodium may be loaded into leaves prior to abscission for removal from the plant.

In the LCS treatment, the low availability of calcium in the nutrient solution may have resulted in

increased root uptake of sodium as a surrogate divalent cation for maintenance of osmotic equilibrium.

A similar relationship may be responsible for the higher magnesium levels observed in LCS.

Magnesium is considered adequate above 0.4 % DW. All samples exceeded this lower limit.

Two other ions will also be considered. Manganese is considered adequate in the range of 30-60ppm

(i.e. 0.003-0.006 % DW), and toxic at >500ppm (i.e. >0.05 % DW). All treatments exceeded the target

range; however none reached ‘toxic’ levels. Boron is considered adequate at 35-70ppm (i.e. 0.0035-

0.007 % DW), and toxic at >100ppm (i.e. >0.01 % DW). All treatments boron levels fell within the range

0.002-0.0026 % DW, which puts them borderline deficit/marginal, but similar across treatments.

Unfortunately, Shiraz petioles were not collected in this experiment. Chenin Blanc is therefore used as

a representative variety in this respect. Analysis of Shiraz petioles would have provided a further insight

into the calcium responses of this variety and possibly mitigated concerns about chloride levels in the

high calcium treated vines.

54

Tab

le 3

.1. M

iner

al n

utrie

nt c

ompo

sitio

n of

Che

nin

Bla

nc p

etio

les

at h

arve

st m

atur

ity. P

etio

le a

naly

sis

was

car

ried

out b

y IC

P-A

ES

on

bulk

ed s

ampl

es o

f 6 p

etio

les

per

vine

, exp

ress

ed a

s m

g/kg

on

a dr

y w

eigh

t bas

is. C

hard

onna

y m

ean

petio

le n

utrie

nt c

ompo

sitio

n at

har

vest

ada

pted

from

Sm

ith e

t al.

(201

4). R

ange

of p

etio

le m

iner

al n

utrie

nt

com

posi

tions

for

petio

les

at fl

ower

ing

adap

ted

from

Rob

inso

n et

al.

(199

7). M

ean

and

SD

from

n =

4 r

eplic

ate

vine

s. *

indi

cate

s si

gnifi

cant

diff

eren

ce fr

om B

NS

con

trol

usi

ng o

ne w

ay a

naly

sis

of v

aria

nce

and

a D

unne

tt’s

mul

tiple

com

paris

on p

osth

oc te

st a

t 95

% C

I (P

<0.

05

C

hen

in B

lan

c p

eti

ole

ICP

at

har

vest

C

har

do

nn

ay a

t h

arve

st.

Smit

h e

t al

. (2

01

4)

Gen

eral

pe

tio

le n

utr

ien

t co

mp

osi

tio

n r

ange

at

flo

wer

ing.

R

ob

inso

n e

t al

. (1

99

7)

Low

cal

ciu

m

solu

tio

n

Bas

al n

utr

ien

t so

luti

on

H

igh

cal

ciu

m

solu

tio

n

Def

icie

nt

Mar

gin

al

Ad

equ

ate

Hig

h

Toxi

c

Mea

n

SD

Mea

n

SD

Mea

n

SD

Mea

n

% D

W

P

0.4

7

0.0

4

0.7

6

0.1

2

1.3

7*

0.2

8

~0.2

7 <0

.2

0.2

-0.2

4 0

.25

-0.5

0 >0

.50

-

K

2.7

7 0

.64

3.1

5 0

.71

2.7

5 0

.42

~2.4

9 <1

.0

1.0

-1.7

1

.8-3

.0

-

Mg

2.1

1*

0.2

4 0

.79

0.0

7 0

.58

0.1

4 ~1

.23

<0.3

0

.3-0

.39

>0.4

-

Ca

0.7

5*

0.0

9 3

.20

0.2

9 3

.55

0.5

6 ~3

.0

- -

1.2

-2.5

-

Cl

1.2

2*

0.0

5 1

.77

0.2

2 2

.55

* 0

.31

- -

- -

- -

mg/

kg

Fe

53

.2

13

.6

41

.5

7.8

5 4

6.0

1

4.4

~4

8

- -

>30

-

Mn

4

20

62

.7

36

5 4

6.5

1

00

* 2

2.6

~3

85

<2

0

20

-29

30

-60

- >5

00

B

23

.0

1.8

3 2

6.0

2

.83

20

.2*

0.5

0 ~6

7

<25

2

6-3

4 3

5-7

0 7

1-1

00

Zn

50

.5

10

.8

85

.0

30

.8

54

.0

10

.5

~10

0

<15

1

6-2

5 >2

6

- -

Na

91

25

* 2

46

0 6

22

5 6

29

25

45

* 6

32

~28

60

- -

- -

>50

00

55

3.3.2 Shiraz berry development responses to modified calcium nutrition

Vines grown under high calcium conditions showed a range of differences in growth and vine

performance. Two-way analysis of variance with Tukey’s multiple comparison test was used to test the

significance of differences observed between time points and treatments; the following results summary

states the key differences observed at a significance level of P < 0.05. HCS berry weights were lower

than BNS from veraison to harvest; at 108 DAA, berry weight loss and shrivel in LCS berries resulted in

berry weights that were not significantly different to HCS berry weights (Figure 3.1A). Onset of berry

weight loss occurred earlier in LCS than in BNS, with an average weight loss between 93 DAA and 108

DAA in LCS of 24%. During the same period, BNS exhibited an average weight gain of 9%. The berry

weight loss and shrivel observed in LCS was so severe that at the last time point (123 DAA) it was not

possible to collect sufficient juice for total soluble solids (TSS) and osmolality measurements. Berry

weight change in HCS bunches was difficult to characterise; the erratic profile of berry weight gain

observed in Figure 3.1A was the result of asynchronous ripening. Some HCS berries ripened rapidly

between 78 and 93 DAA and shrivelled in subsequent stages, whilst other berries were still undergoing

veraison and ripening.

Accumulation of TSS was delayed in HCS bunches, onset of veraison occurred approximately 4 weeks

after BNS and LCS grown vines (Figure 3.1B). Between 78 and 93 DAA HCS berries showed a rapid

accumulation of sugars as they progressed quickly through veraison. The TSS in LCS are seen to

increase rapidly from 93 DAA to 108 DAA due to berry shrivel and water loss causing concentration of

sugars. At 123 DAA HCS berries show large standard deviation for TSS and osmolality due to the

asynchronicity of ripening observed; high calcium bunches contained berries at all stages of

development, including green, veraison, ripe, and shrivelled. The osmolality of berries showed the

same trends between treatments as TSS, indicating that the high and low calcium treatments were not

a major influence on the osmolality of the berries with sugar accumulation dominating (Figure 3.1C).

Calculation of sugar content per berry revealed that sugar uptake ceased in LCS berries between 93

and 108 DAA and in BNS berries between 108 to 123 DAA, whilst still increasing in HCS berries from

108 to 123 DAA (Figure 3.1D).

56

Figure 3.1. Berry development parameters of Shiraz grown under different calcium regimes. High

(HCS), basal (BNS), and low (LCS) calcium grown berry weight (A), total soluble sugars (B), osmolality

(C), and berry sugar content (D). Mean ± SD. N = 14-30 individual berries per sample date per

treatment.

Differences were also observed between treatments in the time taken to reach developmental time

points such as anthesis and veraison. The distribution of anthesis and veraison dates was studied

using the date of the first vine to reach that particular stage as a zero point. Anthesis was measured at

50% cap fall or 50% visibly swollen ovaries (whichever came first). No significant differences in anthesis

dates were found between Shiraz treatments in two consecutive years (Figure 3.2A). In both years,

high calcium treated vines demonstrated later veraison occurrence (measured at 50% of berries

coloured per bunch) than BNS vines (Figure 3.2B). In 2013, HCS treated vines had bunches that

showed much greater variation both within bunches and between vines (i.e. lower synchronicity) in

veraison timing than other treatments. Although the later veraison period for HCS treated vines may

suggest a simple delay in berry development, it would be more accurate to say that veraison in HCS

vines was both delayed and asynchronous, with some berries ripening on a similar timeframe as control

berries and others up to a month later. In 2014, LCS vines had earlier onset of veraison than BNS

D A A

Be

rry

we

igh

t (g

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5L C S

B N S

H C S

D A A

TS

S (

°B

rix

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

1 0

2 0

3 0

D A A

Os

mo

lali

ty (

mm

ol/

Kg

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0

A B

C

D A A

Su

ga

r c

on

ten

t (g

/be

rry

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .0

0 .2

0 .4

0 .6

D

57

vines. Due to the similarities observed in date of anthesis occurrence, the gap between individual

bunch anthesis and veraison dates (Figure 3.2C) reflects similar trends to veraison date (Figure 3.2B).

Figure 3.2. Distribution of anthesis and veraison occurrence in Shiraz bunches grown under different

calcium regimes in two consecutive years (2013 and 2014). Anthesis date distribution (as days from the

date of first observation) (A). Veraison date distribution (as days from the date of first observation) (B).

Anthesis to veraison period calculated as difference in days between observations of these two

developmental points (C). Whiskers indicate data minimum and maximum. * indicates significant

difference of treatment mean from BNS mean (P<0.05) using two way analysis of variance with

Dunnett's posthoc test. n = 12-40 individual vines per treatment per year.

An

the

sis

da

te d

istr

ibu

tio

n

(da

ys

fro

m f

irs

t o

bs

erv

ati

on

)

S h ira z 2 0 1 3 S h ira z 2 0 1 4

0

5

1 0

1 5

2 0

B N S

H C S

L C S

An

the

sis

to

ve

ra

iso

n

pe

rio

d (

da

ys

)

S h ira z 2 0 1 3 S h ira z 2 0 1 4

4 0

6 0

8 0

1 0 0

1 2 0

*B N S

H C S

L C S*

*

A

B

C

Ve

ra

iso

n d

ate

dis

trib

uti

on

(da

ys

fro

m f

irs

t o

bs

erv

ati

on

)

S h ira z 2 0 1 3 S h ira z 2 0 1 4

0

1 0

2 0

3 0

4 0

5 0

*

B N S

H C S

L C S

*

*

58

The treatment effects upon berry ripening were visible in Shiraz bunches; early onset of shrivel in LCS

bunches, and asynchronous and delayed ripening in HCS bunches (Figure 3.3C and 4.3D), relative to

BNS grown bunches (Figure 3.3B) was observed.

Measurements of reproductive performance were taken on whole bunches harvested at the first signs

of shrivel occurring (corresponding to 113 DAA on average). Bunch weights of HCS vines were

significantly lower than BNS or LCS (Figure 3.4A). Flower cap counts were not significantly different

between treatments (Figure 3.4B). Similarly, no significant differences were found between treatments

for number of seeds per berry (Figure 3.4C). Fruitset was lower in HCS vines (Figure 3.4D). Large

variation was observed in Millerandage (Figure 3.4E) and Coulure (Figure 3.4F) indices, however no

significant differences between treatments were observed in these measures. Similar changes in bunch

development and composition were observed in Chenin Blanc, however no measurements were taken.

Figure 3.3. Calcium treatment effect upon ripening of Shiraz bunches. Low calcium treatment (A), basal

nutrient solution (B), and high calcium treatment (C) at 93 days after anthesis. Diversity of berry classes

harvested from a single high calcium grown bunch (D).

59

Figure 3.4. Bunch reproductive indexes of Shiraz grown under different calcium regimes. Bunch weight

(A), flower cap count per bunch (B), seed count per berry (C), Fruitset % (D), Millerandage Index (E),

Coulure Index (F). Mean ± SD. * indicates significant difference using one way analysis of variance with

Dunnett's posthoc test (P<0.05), except 4.3D which used a Fischer LSD posthoc test on Log

transformed data. n = 7-9 individual bunches per treatment.

The consistency of berry ripeness within bunches at harvest varied between treatments. This was

quantified by counting the prevalence of designated berry classes within harvested bunches. It is

interesting to note that, although not statistically significant, both LCS and HCS had slightly higher

flower cap counts than BNS (Figure 3.4B), and that these treatments also exhibited higher rates of

flower abortion (Figure 3.5A). The distribution of different berry classes are represented as; berry class

Bu

nc

h w

eig

ht

(g)

L C S B N S H C S

0

1 0 0

2 0 0

3 0 0

* B N S

H C S

L C S

Co

ulu

re

In

de

x

L C S B N S H C S

0

2

4

6

8

1 0

Mil

lera

nd

ag

e I

nd

ex

L C S B N S H C S

0

1

2

3

4

Flo

we

r c

ap

co

un

t

L C S B N S H C S

0

2 0 0

4 0 0

6 0 0

8 0 0

A B

C D

E F

No

. s

ee

ds

/be

rry

L C S B N S H C S

0

1

2

3

Lo

g1

0(F

ru

its

et

%)

L C S B N S H C S

0 .0

0 .5

1 .0

1 .5

2 .0

*

60

count per bunch (Figure 3.5B), and berry class incidence as a % of total berry count (Figure 3.5C). In

terms of total berry numbers per bunch, HCS had the lowest, and LNS had the highest (Figure 3.5B).

All treatments showed 15-25 % of berries as live green ovaries (Figure 3.5C). The majority of berries

(>70 % of total berry count) on LCS bunches were shrivelled (Figure 3.5C). The predominant berry

class in BNS (also >70 % of total berry count) was (i.e. berries which were fully coloured but did not

show signs of shrivel). Interestingly, HCS bunches showed higher rates of berry shrivel than BNS, as

well as higher rates of green and veraison berry incidence (Figure 3.5C).

61

Figure 3.5. Flower and berry development of Shiraz grown under different calcium regimes and

harvested at 114 DAA. Flower fate; incidence of mature berries (ripe/shrivelled), stunted berries

(LGO/seedless/green/veraison), and aborted flowers calculated as % of total flowers per bunch at time

of harvest (A). Incidence of individual berry classes, presented as berry numbers per bunch (B) and

berry class incidence as % of total berry count (C).

62

3.3.3 Berry class diversity

A number of measures were taken on individual berries from Shiraz bunches divided by berry class.

Mean berry weight was measured in each berry class; due to the classing system, it was unsurprising

that the mean berry weight of shrivelled LCS and HCS berry classes were lower than that of their

respective ripe berry classes (Figure 3.6A). TSS levels increased according to berry class (Figure

3.6B). LCS berries at veraison, ripe, and shrivelled stages showed higher berry deformability than BNS

and HCS berries at the same stage (Figure 3.6C). Berry Young’s Modulus was generally lower (i.e.

more elastic) in the more mature berry classes; LCS berries were more elastic at all maturity levels

(Figure 3.6D). Seed counts per berry showed no discernible differences between treatments or berry

maturity class (Figure 3.6E).

63

Figure 3.6. Measures of berry variability within Shiraz bunches grown in high, basal and low calcium

conditions and harvested at 114 DAA, divided by berry class. Berry weight (A), total solubles sugars

(°Brix) (B), berry deformability (C), berry elasticity (skin fold caliper calculations; D), seeds per berry

(E). Mean ± SD. Replicate numbers vary due to scarcity/prevalence of some berry classes in some

treatments.

3.3.4 Vine physiology

HCS treated Shiraz and Chenin Blanc vines had lower transpiration rate (E), net assimilation rate (A),

stomatal conductance (Gs), and sub-stomatal CO2 (Ci) than BNS vines (Figure 3.7A-D). LCS grown

Chenin Blanc vines had higher A and Gs than BNS vines (Figure 3.7B, C). Midday leaf water potential

Be

rry

we

igh

t (g

)

G re e n V e ra is o n R ip e S h r iv e lle d

0

1

2

3

B N S

H C S

L C S

TS

S (

°B

rix

)

G re e n V e ra is o n R ip e S h r iv e lle d

0

1 0

2 0

3 0

4 0

5 0

Be

rry

de

form

ati

on

(% d

iam

ete

r)

G re e n V e ra is o n R ip e S h r iv e lle d

0

1 0

2 0

3 0

4 0

5 0

Be

rry

Yo

un

g's

mo

du

lus

(N

mm

-1)

G re e n V e ra is o n R ip e S h r iv e lle d

0 .0 1

0 .1

1

1 0

Se

ed

s p

er b

erry

G re e n V e ra is o n R ip e S h r iv e lle d

0

1

2

3

A B

C D

E

64

of HCS vines was more negative than BNS in both varieties (Figure 3.7E). No significant differences

between treatments were identified in root hydraulic conductance (Figure 3.7F).

Figure 3.7. Whole vine physiology measures of Shiraz grown in high and basal calcium conditions and

Chenin Blanc grown in high, basal and low calcium conditions. Infra-red gas analyser used to measure

transpiration rate (A), net assimilation rate (B), stomatal conductance (C), sub-stomatal CO2 (D).

Pressure bomb measurements of leaf water potential (E). HCFM measurements of root hydraulic

conductance (F). Mean ± SD. * indicates significant difference from BNS (P<0.05) using unpaired t-test

(Shiraz), and one way analysis of variance with Dunnett's posthoc test (Chenin Blanc). n = 3 - 4

individual vines per treatment, 4 leaves per vine were selected for IRGA measurements, 2 leaves per

vine were selected for leaf water potential measurements.

Tra

ns

pir

ati

on

ra

te E

(mo

l m

-2 s

-1)

S h ira z C h e n in B la n c

0

1

2

3

4

5

*

*

Ne

t a

ss

imil

ati

on

ra

te A

(

mo

l m

-2 s

-1)

S h ira z C h e n in B la n c

0

5

1 0

1 5

2 0

B N S

L C S

H C S

*

*

*

Sto

ma

tal

co

nd

uc

tan

ce

of

H2

O g

s

(mo

l m

-2 s

-1)

S h ira z C h e n in B la n c

0 .0

0 .1

0 .2

0 .3

0 .4

0 .5

*

*

*

Su

b-s

tom

ata

l C

O2

cI

(vp

m)

S h ira z C h e n in B la n c

0

1 0 0

2 0 0

3 0 0

4 0 0

v p m = v o lu m e s p e r m illio n v o lu m e s o f a ir

**

Le

af

wa

ter p

ote

nti

al

le

af

(MP

a)

S h ira z C h e n in B la n c

-1 .5

-1 .0

-0 .5

0 .0

*

*

Ro

ot

hy

da

uli

c c

on

du

cta

nc

e

(K

25

;g

DW

(K

g s

ec

-1M

pa

-1)

g-1

)

S h ira z C h e n in B la n c

-2 .01 0 -9

0

2 .01 0 -9

4 .01 0 -9

6 .01 0 -9

8 .01 0 -9

n s n s

A B

C D

E F

65

3.3.5 Chenin Blanc nutrient uptake from veraison to late harvest

Chenin Blanc grown in BNS was sampled throughout development to study calcium accumulation

patterns. Berry weight and osmolality steadily increased from veraison at ~55 DAA to 120 DAA,

showing no signs of shrivel or weight loss (Figure 3.8A, B). Berry juice pH started increasing at ~70

DAA to reach ~3.6 at 120 DAA (Figure 3.8C). Total soluble sugars increased from veraison (Figure

3.8D).

Measurement of berry ionic composition using ICP-AES demonstrated differences between phloem-

immobile and phloem-mobile nutrient accumulation. Total berry calcium content was stable from

veraison to late harvest; regression analysis revealed that the slope did not significantly vary from zero

(P=0.654) (Figure 3.8E).The increase in berry weight during this period resulted in decreasing whole

berry calcium concentration (Figure 3.8F). This contrasts to phloem mobile potassium, which was seen

to increase in both total berry content (Figure 3.8G) and concentration (Figure 3.8H) throughout

development. As TSS and ICP-AES data was not collected on the same berries, osmolality was used to

test for a correlation with berry [K+] instead. A positive correlation between berry osmolality and berry

[K+] was observed; regression analysis revealed that the slope significantly varied from zero

(P<0.0001).

66

Figure 3.8. Chenin Blanc berry development from veraison to harvest. Individual berry measurements

of berry weight (A), osmolality (B), pH (C), total soluble sugars (°Brix; D), total berry Ca2+ (E), berry

[Ca2+] (F), total berry K+ (G), berry [K+] (H). Days after anthesis (DAA); anthesis taken as date of 50%

capfall.

4 0 6 0 8 0 1 0 0 1 2 0

0

1

2

3

D A A

Be

rry

we

igh

t (g

)

4 0 6 0 8 0 1 0 0 1 2 0

0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

D A A

Os

mo

lali

ty (

mm

ol/

Kg

)

4 0 6 0 8 0 1 0 0 1 2 0

2 .0

2 .5

3 .0

3 .5

4 .0

D A A

pH

4 0 6 0 8 0 1 0 0 1 2 0

0

1 0

2 0

3 0

D A A

TS

S (

°B

rix

)

4 0 6 0 8 0 1 0 0 1 2 0

0 .0

0 .1

0 .2

0 .3

0 .4

0 .5

D A A

To

tal

be

rry

Ca

2+

(m

g)

4 0 6 0 8 0 1 0 0 1 2 0

0

2

4

6

8

D A A

Be

rry

[C

a2

+]

(mM

)

4 0 6 0 8 0 1 0 0 1 2 0

0

2

4

6

8

1 0

D A A

To

tal

be

rry

K+

(m

g)

4 0 6 0 8 0 1 0 0 1 2 0

0

5 0

1 0 0

1 5 0

D A A

Be

rry

[K

+]

(mM

)

A B

C D

E F

G H

67

Petiole ionic composition of hydroponically grown Chenin Blanc was measured at harvest using ICP-

AES. This revealed many differences in petiole nutrient accumulation between treatments. Interestingly,

although calcium levels in the hydroponic solutions were vastly different, only petioles from LCS vines

showed significant difference in calcium concentration from BNS petioles (Figure 3.9A). Phosphate

levels in HCS petioles were significantly higher than BNS (Figure 3.9B). LCS had lower chloride, and

HCS higher chloride than BNS (Figure 3.9C), possible due to differences in chloride concentration in

hydroponics solutions. No significant differences in potassium were observed (Figure 3.9D). LCS had

higher sodium, and HCS lower sodium, than BNS (Figure 3.9E). Petiole magnesium levels in LCS vines

were much higher than in BNS vines (Figure 3.9F). Petioles from HCS vines had lower manganese and

boron than from BNS vines (Figure 3.9G, H).

Basic berry juice parameters and ICP-AES data for Chenin Blanc whole berries, mesocarp, and skin

are detailed in Table 3.2.

68

Figure 3.9. Chenin Blanc petiole ion composition at harvest. Concentration of nutrient ions in dried

petiole samples as determined by ICP-AES analysis. All units are % dry weight; calcium (A),

phosphorus (B), chloride (C), potassium (D), sodium (E), magnesium (F), manganese (G), boron (H). *

indicates significant difference from BNS using one-way ANOVA with Dunnett's multiple comparison

test (P<0.05). n = 4 vines per treatment, ICP-AES carried out bulked samples of 6 petioles per vine.

K

% D

W

L C S B N S H C S

0

1

2

3

4

5

N a

% D

W

L C S B N S H C S

0 .0

0 .5

1 .0

1 .5

*

*

M g

% D

W

L C S B N S H C S

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5 *

M n

% D

W

L C S B N S H C S

0 .0 0

0 .0 2

0 .0 4

0 .0 6

*

B

% D

W

L C S B N S H C S

0 .0 0 0

0 .0 0 1

0 .0 0 2

0 .0 0 3

0 .0 0 4

*

C l

% D

W

L C S B N S H C S

0

1

2

3

4

*

*

A B

C D

E F

G H

C a%

DW

L C S B N S H C S

0

1

2

3

4

5

*

P

% D

W

L C S B N S H C S

0 .0

0 .5

1 .0

1 .5

2 .0

*

69

Low calcium solution Basal nutrient solution High calcium solution

Mean SD Mean SD Mean SD

Berry juice

parameters

Brix 19.58 3.316 19.86 1.964 17.83 1.402

pH 3.705 0.1718 3.602 0.07004 3.106** 0.2951

Ca2+ activity

ISE (mM)

0.1263* 0.02626 0.2985 0.05225 0.797**** 0.1736

Whole grape

ICP-AES

(mg/kg)

B 0.4862 0.1276 0.5302 0.1070 0.2454** 0.01311

Ca 1.110**** 0.08731 4.055 0.8201 10.04**** 0.5533

Mg 7.207** 0.1634 4.486 0.4821 5.608**** 0.2266

Na 1.227 0.4028 1.533 0.9438 0.3164* 0.01979

K 106.8 13.27 92.08 8.088 100.4 3.218

P 11.62 2.190 11.95 0.6457 27.36**** 1.612

S 4.889 0.2473 5.167 1.327 6.165 0.3524

Fe 0.1886 0.02609 0.1757 0.07184 0.1589 0.07644

Mn 0.02369 0.004854 0.02320 0.004911 0.02925 0.001620

Zn 0.01413 0.002013 0.01536 0.002772 0.02054* 0.002043

Mesocarp

tissue ICP-AES

(mg/kg)

B 0.3267 0.04836 0.3821 0.09942 0.1904** 0.02270

Ca 0.2607* 0.01920 0.6618 0.1334 1.604*** 0.3028

Mg 4.281*** 0.1975 2.385 0.5069 3.378* 0.4210

Na 1.273 0.3498 1.611 0.5720 0.2959** 0.04700

K 58.25 15.67 60.11 4.898 60.11 4.430

P 6.820 1.845 6.360 0.4367 13.08**** 1.071

S 2.872 0.2117 3.657 0.8121 3.680 0.1222

Fe 0.1742 0.07272 0.1683 0.05213 0.1565 0.05540

Mn 0.008656 0.002136 0.007543 0.001739 0.009967 0.001903

Cu 0.006913 0.002031 0.008926 0.001480 0.006325 0.0008032

Zn 0.009965 0.001238 0.01158 0.001001 0.008683** 0.0008661

Skin tissue

ICP-AES

(mg/kg)

B 1.088 0.3350 1.377 0.3131 0.4781** 0.08148

Ca 1.945**** 0.2569 5.608 0.5292 13.97**** 0.4555

Mg 9.715** 1.468 6.549 0.4013 5.690 0.7387

Na

K 390.7* 44.47 305.0 31.42 370.2* 26.11

P 30.99 9.013 32.28 9.646 102.5**** 8.890

S 10.45 1.496 9.669 2.836 12.24 1.121

Fe 0.2013 0.03363 0.2193 0.03950 0.2193 0.07980

Mn 0.06337 0.01553 0.05743 0.006415 0.05915 0.003566

Cu 0.01601 0.007563 0.01914 0.002881 0.01571 0.0002091

Zn 0.02429 0.009513 0.03471 0.004552 0.02704 0.0004426

Table 3.2. Ionic composition of Chenin Blanc whole grape, mesocarp, and skin tissues at harvest

maturity. Brix, pH and calcium activity measurements carried out on juice from 10 bulked berries per

bunch. Whole grape, mesocarp, and skin analysis was carried out on five individual berries per bunch

expressed as a mean, or as a single measurement of five or ten pooled berries, expressed as mg/kg on

a fresh weight basis. Mean and SD from n = 4 replicate bunches. Columns with asterisks are significantly

different from BNS control using one way analysis of variance and a Dunnett’s multiple comparison

posthoc test at 95 % CI (* P=0.05-0.01, ** P=0.01-0.001, *** P=0.001-0.0001, **** P <0.0001).

70

3.4 Discussion

The differences in berry ripening phenology between vines grown under different calcium regimes

indicate that calcium contributes to fruit development through previously identified roles as a signalling

element and cell wall component. Although calcium has been identified as a factor in several fruit

disorders, few previous studies have focused on the effects of calcium nutrition levels on fruit

development. Response to changes in calcium nutrition were investigated in two varieties; Shiraz is an

economically important variety that demonstrates susceptibility to the late ripening disorder berry

shrivel, whereas even under late harvest conditions, Chenin Blanc maintains berry cell vitality and does

not succumb to berry shrivel.

Petiole analysis of Chenin Blanc provided interesting insights into the effects of the three nutritional

regimes on ion accumulation and vine nutrition. Whilst petiole sampling for field vine nutritional analysis

is typically carried out at flowering, the results presented here give a snapshot of a later developmental

phase (i.e. at harvest). This time point may reveal more about the long term effects of the treatments,

as the vines have been exposed for longer, allowing greater accumulation of “excess ions” into the leaf

prior to abscission. Petiole concentration of some ions has been shown to change from flowering to

harvest (Smith et al., 2014).

3.4.1 Calcium deficiency accelerates berry ripening

Previous studies examining the effects of low calcium supply on fruit development are scarce, with most

focused on localised calcium deficiency disorders such as blossom end rot in tomatoes and bitter pit in

apples (White and Broadley, 2003). Whilst many studies (and indeed horticultural practices) have

utilised direct application of calcium to leaves and fruit to alleviate calcium deficiency disorders (often

as calcium chloride or calcium nitrate sprays), the results have been inconsistent (Saure, 2005). It is

broadly accepted that calcium applied to leaves does not increase fruit calcium. Direct uptake of

calcium from the fruit surface is also problematic, affected by variety and developmental stage (Saure,

2005). Additionally, high levels of surface calcium can cause skin injury. Although soil calcium is

generally thought of as sufficient in most agricultural soils, the prevalence of localised calcium deficit

disorders indicate that plant calcium interactions affecting calcium uptake and translocation often result

in low calcium supply situations that are detrimental to fruit development.

The research undertaken here demonstrates that low or deficient calcium supply in Shiraz results in

accelerated ripening through changes to the physical properties and water relations of berries. Early

71

onset of berry weight loss and shrivel were the clearest signs that low calcium supply had perturbed

normal berry functions (Figure 3.1). The causes for this are believed to be related to calcium-cell wall

interactions and water transport in the berries (investigated further in Chapter 4). This cannot be

explained by general water stress on the vine caused by low calcium supply, as indicated by the leaf

gas exchange data, leaf water potentials and root hydraulic conductance.

There were no signs that low calcium supply in Shiraz affected fruitset or reproductive indices relative to

basal nutrient supplied vines (Figure 3.4). In bunches harvested at an approximation of commercial

harvest point (114 DAA), the majority of berries in low calcium grown bunches had progressed into the

shrivelled berry class, indicating that calcium deficient growth shortens the developmental window of

grapes or exacerbates cell vitality loss and berry shrivel. Very few berries from the BNS treatment were

shrivelled at this point, with the dominant berry class being ripe (Figure 3.5).

Low calcium grown Chenin Blanc vines had higher leaf net assimilation rate and stomatal conductance

than basal grown vines, but no significant difference in transpiration rate, sub-stomatal CO2, leaf water

potential, or root hydraulic conductance (Figure 3.7). Calcium plays an important role in leaf stomatal

movement. Antisense knockout of the extracellular calcium sensor CAS exhibited increased stomatal

aperture and decreased stomatal response to water stress, leading to higher transpiration and stomatal

conductance (Wang et al., 2014). The higher transpiration rate may serve to increase delivery of

calcium to the leaf; the relationship between leaf calcium supply and transpiration has been previously

reviewed in Gilliham et al. (2011).The reduction in sensitivity to extracellular calcium observed in CAS

knockouts mimics the response to low calcium supply observed in grapevines. Additionally, the

increased net assimilation rate may advance the ripening schedule of berries, with sugar signalling

linked to regulation of cell division and expansion (Wang and Ruan, 2013).

By studying individual berry qualities in particular berry classes of Shiraz from all three treatments we

were able to distinguish developmental differences from other influences. Conclusions about some

classes and treatments are speculative due to low berry numbers. Importantly, in all berry classes, low

calcium berries showed higher levels of deformation and lower Young’s Modulus values than basal

nutrient berries (Figure 3.6).

Previous research has demonstrated that grapevines have the capacity to mobilise calcium stores in

the roots and trunks to supply aerial tissues throughout the growing season (Conradie, 1981). The

delivery mechanisms for this calcium are as yet unknown; however it is clear that under these

72

conditions, movement from other tissues did not enable sufficient calcium transport into berries after

veraison to ameliorate the low calcium supply.

3.4.2 Elevated calcium impairs berry development

Some studies have directly applied calcium sprays throughout development in an attempt to modify

post-harvest storage traits. Direct application of calcium chloride sprays (as calcium-EDTA) at pre-

veraison stage to the table grape variety Italia increased skin, mesocarp, and seed calcium

concentrations, and increased berry firmness and breaking force. Post-veraison calcium treatments

delivered less consistent results (Ciccarese et al., 2013). Spray treatment of Thompson Seedless with

CaCl2 at pea-size and veraison resulted in increased berry firmness and berry adherence strength,

whilst simultaneously increasing sugar accumulation (Marzouk and Kassem, 2011). However, that

study had minimal replication and all treatments appeared to increase sugar accumulation with various

shifts in harvest ripeness compared to control vines. Calcium infiltration has been shown to delay fruit

softening in Golden Delicious apples, possibly due to strengthening of cell-walls (Sams and Conway,

1984). It is clear that variations in calcium response occur at the species and even variety level, as well

as being dependent upon stage of fruit development at the time of application. Exogenously applied

calcium does not produce consistent changes in fruit physiology and observed changes are not

necessarily representative of soil calcium nutritional responses.

Both sodium and chloride ions affect grapevine growth under saline conditions, with different rootstocks

exhibiting different capacities for exclusion of these ions (Walker et al., 2014). Although the effects of

salinity on plants are often characterised in terms of the osmotic and ionic effects of sodium, the

injurious effects of chloride are more prominent in grapevines, particularly in the context of Australian

viticulture, where chloride is the major anion in irrigation water (Downton, 1977; Teakle and Tyerman,

2010). Work to understand the molecular mechanisms of chloride exclusion in grapevine is currently

being undertaken, with several chloride transporters being characterised (Henderson et al., 2014). The

level of chloride in the high calcium hydroponics treatment may have exceeded levels for normal

growth. Petiole chloride concentrations at flowering from a variety of field sites were measured in own-

rooted Chardonnay (0.9-1.22 % DW), and own-rooted Shiraz (0.98-1.43 % DW) (Walker et al., 2010). In

this study, Chenin Blanc petioles were sampled at harvest, with 1.22 % DW in LCS, 1.77 % DW in BNS,

and 2.55 % DW in HCS (Figure 3.9). The later sampling date (at harvest rather than flowering) may

account for some of the extra chloride accumulation. Excess potassium, sodium and chloride ions can

be retained in senescing leaves for removal from the plant (Downton, 1977). Although, attempts were

made to limit chloride supply to the vines, the high levels of petiole chloride observed (particularly in the

73

high calcium treatment) may indicate that chloride toxicity is responsible for some of the treatment

effects observed.

Berry development of high calcium grown vines was delayed; lower berry weights, brix, and osmolarity

were observed at every time point (Figure 3.1). This delayed development translated into lower bunch

weights at harvest, which can be at least partially attributed to lower fruitset, higher CI, resulting in lower

berry numbers per bunch (Figure 3.4). At an approximation of commercial harvest point (114 DAA), the

high calcium bunches exhibited a mixture of berry classes, including shrivelled, ripe, and significant

numbers of green berries (Figure 3.5). HCS grown Shiraz also had lower berry weight and TSS than

BNS in all berry classes (Figure 3.6). Grapevine irrigation with saline water can result in reduced berry

weight (Prior et al., 1992) , delayed sugar accumulation and onset of veraison (Hawker and Walker,

1978), indicating that some of the effects observed here may be a result of the high chloride content of

the HCS solution. Shiraz berry deformability and elasticity of HCS was similar to BNS at all berry

maturity levels (Figure 3.6).

The effects of high calcium treatment on Shiraz and Chenin Blanc vine physiology included reductions

in transpiration rate, net assimilation rate, stomatal conductance, sub-stomatal CO2, and leaf water

potential (Figure 3.7). This suggests that the high levels of calcium present in leaves of HCS grown

vines may reduce transpiration rates through perturbed stomatal responses or ionic effects upon water

transport. The reduced net assimilation rate may well explain the lower rate of sugar accumulation

observed in HCS berries.

3.4.3 Calcium impact on vine fruitset and reproductive indices

Although significant differences in individual measures of reproductive performance were minimal, a

cumulative effect produced very different bunch characteristics. Neither Coulure Index (the proportion

of aborted/un-fertilised flowers) nor Millerandage Index (the proportion of seedless berries and live

green ovaries in a bunch) showed significant differences between HCS and BNS Shiraz vines in this

study. However, there were indicators of reduced reproductive performance, including reduced bunch

weight and lower fruitset in high calcium Shiraz vines (Figure 3.4). Repeating the experiment with

higher replicate numbers would help to elucidate this. The effects of high calcium upon flowering,

fruitset and berry development are similar to salinity responses. For instance, in navel oranges irrigated

with excess sodium chloride, lower flower numbers, lower fruitset, and slower/delayed fruit growth and

sugar accumulation was observed (Howie and Lloyd, 1989). Differences in veraison synchronicity

observed in high calcium grown fruit are carried out by differences in berry sugar accumulation rates

74

and bunch composition at harvest. The lack of significant differences between treatments with regards

to reproductive measures may be an artefact of the fruiting cutting system, or the relatively low bunch

numbers used in these measurements. Additionally, as bud formation occurs in the previous season

(prior to treatment), treatment effects on some reproductive measures may be mitigated.

3.4.4 Chenin Blanc nutrient accumulation consistent with other varieties

Calcium accumulation in grapes appears consistent across varieties that differ in ripening phenology.

Previous work has identified calcium as a predominantly xylem mobile element in Shiraz grapes

(Rogiers et al., 2000; Rogiers et al., 2006a). The majority of calcium accumulation in grapes occurs

prior to veraison, with total berry calcium reaching a maximum at veraison. The continued growth of

berries post-veraison leads to a reduced concentration of calcium in the skin and mesocarp tissues.

Some evidence suggests that during later phases of ripening berry mesocarp calcium is mobilised and

transported to the skin (Cabanne and Doneche, 2003). The lack of post-veraison calcium accumulation

has been observed in a number of varieties. Varieties differ in susceptibility to loss of mesocarp cell

vitality and berry shrivel (Fuentes et al., 2010), this may be caused by differences in post-veraison

calcium import. To test this hypothesis, the nutrient accumulation of Chenin Blanc, a variety that shows

very low susceptibility to loss of cell vitality loss and berry shrivel, was studied. These results

demonstrated that Chenin Blanc had a similar calcium accumulation profile to previously surveyed

varieties Shiraz (Rogiers et al., 2006b) and Grenache Noir (Etchebarne et al., 2009). Although the

uptake of calcium into Chenin Blanc berries throughout development does not appear to differ from

other varieties, translocation of calcium between tissues and cell wall binding of calcium is a potential

factor warranting further investigation.

3.5 Conclusion

It has been demonstrated that low calcium supply can cause a number of changes in Shiraz and

Chenin Blanc grape development and vine physiology. Whilst some of the osmotic effects of low

calcium supply may be moderated by increased uptake of other divalent cations (e.g. sodium and

magnesium), sufficient calcium is still required to carry out specific roles in cellular signalling and

modifying cell wall properties. Specifically, low calcium supply resulted in increased vine photosynthesis

and stomatal conductance, accelerated fruit ripening, early onset of berry shrivel, higher berry

deformation, and higher berry elasticity. Changes in berry cell wall properties are unlikely to affect leaf

75

photosynthesis and stomatal conductance, suggesting that these changes were a result of perturbed

cellular signalling or other calcium interactions in the leaf. As fruit cell wall changes and signalling are

interconnected, it is difficult to determine whether effects observed in the fruit are direct or indirect

responses to deficient calcium supply.

The high calcium treatment applied resulted in an array of changes in Shiraz berry and vine physiology.

Some of these changes may be attributable to high chloride levels in the solution; however levels of

calcium were so far above the norm as to be expected to exert an influence. Changes in high calcium

grown berry development included; lower fruitset, lower bunch weights, delayed (measured by reduced

berry weights, brix, and osmolarity) and asynchronous (observed in the diversity of berry classes

present at harvest) berry ripening. High calcium grown vines also exhibited lower transpiration,

photosynthesis, stomatal conductance, sub-stomatal CO2, and leaf water potential. The combination of

developmental defects and reduced vine performance is a result of high calcium (and chloride) levels

perturbing hormonal developmental signalling and stomatal operation, as well as the ionic effects which

affects water transport. Interactions of calcium with berry cell walls may also play a role in reducing

berry cell expansion and delaying ripening. The relationships between berry calcium, cell walls, and

berry water relations will be further investigated in Chapter 4.

Investigation of the post-veraison nutrient accumulation of Chenin Blanc berries demonstrated similar

patterns to previously studied varieties; no obvious indication was discovered that the observed low

susceptibility of Chenin Blanc to berry cell vitality loss and shrivel (Fuentes et al., 2010) was due to

differences in berry calcium uptake. However, the increased berry shrivel observed in low calcium

grown Shiraz indicated that processes affecting the distribution and utilisation of calcium within the

berry may still exert an influence on membrane viability, susceptibility to cell vitality loss and shrivel.

Low or high calcium supply was shown to affect the uptake and translocation of other nutrient ions,

particularly other divalent cations and the chloride counter-ion.

76

Chapter 4 Berry ripening physiology; calcium uptake and cell wall modification

4.1 Introduction

The link between cell wall modification and ripening has been explored in a number of fruits (Brummell,

2006). Although it has been shown that calcium interacts with de-methyl esterified pectin to change the

physical properties of the cell wall (Willats et al., 2001), how exactly this process affects ripening in fruit

is unknown. Changes in calcium nutrition will not just affect cell wall properties of ripening fruit, but also

water relations (Gilliham et al., 2011) and transduction of cellular signals (including those of ripening

related phytohormones) (Yu et al., 2006). In this chapter, the effects of changes in calcium nutrition on

grape berry ripening will be investigated. Vines were grown in hydroponics and subjected to three

different nutritional regimes; high calcium solution (HCS), low calcium solution (LCS), and basal nutrient

solution (BNS). The impact of these treatments on aspects of berry physiology (including ripening

phenology, water relations, and cell wall modification) and fruit quality traits (such as berry shrivel and

Botrytis cinerea susceptibility) were investigated.

4.2 Materials and Methods

4.2.1 Grapevine material

Fruiting cuttings were produced according to the Mullins method modifications described in Baby et al.

(2014) (Appendix 3) and elaborated on in Chapter 3.2.1 and 3.2.2. At least 24 biological replicate vines

were grown per treatment. For time course experiments, 3 biological replicate vines were sampled per

treatment per time point; other experiments comprised 3-4 biological replicate vines per treatment. The

number of technical replicates (i.e. berries sampled from each biological replicate vine) varied from

depending on the experiment; technical replicate numbers are noted in figure legends.

4.2.2 Immunogold labelling and transmission electron microscopy

Berry sections were prepared as per Chapter 2.2.2, except for changes to ethanol dehydration. Ethanol

dehydration was carried out using a series of changes at 20 minutes each (2 X 30%, 2 X 50%, 2 X

70%, 2 X 90%, and 2 X 95%), followed by 3 X 100% dry ethanol changes of 4 hours, 16 hours, and 8

hours each.

Immunolabelling was undertaken for TEM of berry sections. Sections for TEM (70-90 nm thick) were cut

on a Leica EM UC6 ultra-microtome (Leica, Germany) using a diamond knife. Sections were transferred

77

to 100 bar nickel grids coated with 2% collodion in amyl acetate (Electron Microscopy Sciences,

Pennsylvania, USA). Grids were incubated on 50 µL drops of 0.05 M glycine for 20 minutes (drops

were placed on parafilm with a Pasteur pipette and grids inverted onto the drops). This was followed by

3 X 10 minute washes on drops of incubation buffer (1 X PBS, 0.2% AURION BSA-c) (AURION,

Wageningen, Netherlands). Grids were blotted and wick-dried on filter paper after the third wash. Grids

were then inverted onto 20 µL drops of primary antibody solution (LM19 and LM20 diluted 1/20 in

incubation buffer) on parafilm for 90 minutes, followed by 3 X 10 minute washes on drops of incubation

buffer (1 X PBS, 0.2% AURION BSA-c). Grids were blotted and wick-dried on filter paper after the third

wash. Grids were then inverted onto 20 µL drops of gold conjugate solution (Goat Anti-Rat IgM diluted

1/30 in incubation buffer) on parafilm for 90 minutes, followed by 3 X 10 minute washes on 50 µL drops

of incubation buffer (1 X PBS, 0.2% AURION BSA-c). Grids were blotted and wick-dried on filter paper

after the third wash then washed on 50 µL drops of distilled water twice for 5 mins, blotted after the

second wash, and stored face-up in a petri dish. Prior to microscopy, grids were stained with uranyl

acetate by inverting them on drops of 4% aqueous uranyl acetate solution for 15 minutes, followed by

three rinses on water drops, then wick-dried on filter paper and air-dried.

TEM microscopy was carried out on a Philips CM100 with Megaview II imaging system (Philips,

Eindhoven, Netherlands) at Adelaide Microscopy Waite Campus.

4.2.3 Berry staining methods; FDA and PI

Staining methods on fresh berries were utilised to capture developmental changes in cell vitality and

cell wall composition. Fluorescein diacetate (FDA), a cell vitality stain, was used for this purpose. FDA

staining method was modified from Fuentes et al. (2010). Stock solution of FDA (4.8 mM) was made up

in acetone on the day of staining. Sucrose solutions spanning the range of osmolality observed in the

berries were made up. FDA stock solution was added to each sucrose solution for a final FDA

concentration of 4.8 µM. Propidium iodide (PI) was used to stain de-esterified pectins in fresh berries. A

stock solution of 2 mg/mL (3 mM) PI was made up in milliQ water. PI stock solution was added to FDA

sucrose solution for a final concentration of 30 µM. Individual berries were cut in half longitudinally with

a ‘Diplomat’ double-edged razor blade (Personna, USA) following removal of the pedicel. Efforts to

avoid seeds when cutting were made by illuminating the berry underside using an LED torch to show

seed location. Only berries that were cut cleanly through with, other than the cut face, no or negligible

tissue damage or deformation were used for staining. Juice was expressed from one half of the berry

by squeezing. TSS was measured using a digital refractometer and osmolality using a Wescor 5500B

Vapour Pressure Osmometer (Wescor Inc. Utah, USA). The other berry half was carefully positioned in

78

a holding rack, 100-150 µL of PI/FDA/sucrose solution of appropriate osmolality (within 100 mmol/Kg)

was pipetted onto the cut berry surface. Berries were stained for 2 hours in darkness before imaging.

Shorter staining times (e.g. 15 minutes) also produced good staining patterns. However, no loss of

staining intensity was observed using longer stain times, and this allowed a greater number of berries

to be stained and imaged in each experiment. After staining, berries were imaged using a Leica

MZFLIII Stereo Fluorescence Dissecting Microscope with Fluo III camera unit (Leica, Wetzlar,

Germany). Sequential imaging on each half berry was undertaken at 1X magnification; FDA staining

with GFP Plant filter, PI staining with dsRed2 filter, and combination FDA/PI with wide pass GFP Plus

filter, a bright field (BF) image was also taken for edge detection in image analysis software.

Microscope settings were optimised for maximum contrast of stains to background and for balance of

stain intensity in GFP Plus images (FDA typically stained more strongly than PI). Fluorescence image

exposure time was 596.6 ms, BF exposure time was 10 ms. Check colour/colour correction was set to

100%/0% (i.e. red emphasis) to balance the lower fluorescent signal strength of red PI fluorescence

against the stronger green FDA fluorescent signal.

4.2.4 MATLAB/ImageJ image analysis

A Matlab macro developed by Fuentes et al. (2010) was used initially for determining berry area, %

berry area FDA stained, and % berry area PI stained. The development of a more detailed staining

analysis macro was undertaken in order to capture the tissue specific co-staining patterns of PI. The

ImageJ platform (ImageJ, NIH, USA) was used. The macro utilised the combined FDA/PI images

captured with wide pass GFP Plus filter to detect berry edges, to manually exclude seed areas, and

quantify levels of PI and FDA staining in skin and mesocarp. Skin tissue was determined as a band of

nominal width (30 pixels) around the circumference of the berry; this band contained the epidermal and

hypodermal tissues in their entirety, as well as some outer mesocarp cells and peripheral vasculature.

The following outputs were extracted from individual images; berry area, berry area (- seeds), % berry

area FDA stained, % berry area PI stained, % berry area FDA/PI co-stained, skin (epidermis and

hypodermis) area, % skin area FDA stained, % skin area PI stained, and % skin area FDA/PI co-

stained. Assistance with scripting and de-bugging the macro was generously provided by Cameron

Nowell (Monash Institute of Pharmaceutical Sciences, Melbourne, Australia). The ImageJ Macro script

is provided in Appendix 5.

79

4.2.5 Berry electrical impedance spectroscopy

Berry impedance was measured using an impedance analyser (Bode 100 Vector Network Analyser,

OMICRON Lab, Hong Kong) attached to two silver/silver chloride 8 mm disc electrodes (SDR Scientific,

Sydney, Australia). The electrodes were calibrated with an open circuit, a short circuit, and a 49.6 Ω

resistor according to the manufacturer’s instructions. Berries were prepared by slicing thin sections

(approximately 3 mm diameter) with a razor blade from opposite sides of the berry on the axis

perpendicular to the pedicel end to blossom end axis. A drop of 0.1 M KCl was applied to both cut

surfaces before being placed on the bottom electrode. The top electrode was lowered using a

calibrated micrometer screw until it contacted the berry surface. The electrical path distance (i.e. the

distance between the electrodes) was recorded from the micrometer screw. The impedance was

measured at 100 frequencies across a range from 1 Hz to 1 MHz. After impedance measurement, berry

weight and total soluble solids were measured.

Modelling of berry electrical circuits to determine Re, Ri, and Cm was performed as per Caravia Bayer et

al. (2015) from the Bode impedance analyser raw data.

4.2.6 Berry rehydration assay

A custom assay to measure the water uptake, transpiration rate, and berry weight change of individual

berries in vitro was carried out as per the design published in detail by Scharwies (2013). Single berries

with pedicel attached were cut from the bunch under degassed milliQ water; the pedicel was then cut

again underwater with a sharp razor to provide a clean undamaged interface for water transport. The

cut pedicel was then inserted underwater into the flared end of an artificial pedicel; a 40 mm long piece

of 2 mm external diameter X 1 mm internal diameter polyethylene tubing pre-filled with degassed milliQ

water. The junction of pedicel and artificial pedicel tubing was dabbed dry with absorbent paper towel

and sealed with adhesive free silicone glue (3140 RTV Coating, Dow Corning Corp, Missouri, USA).

Care was taken to exclude air bubbles from the tube at all stages. The open end of the artificial pedicel

was placed into a water reservoir constructed from a modified pipette tip box filled with degassed milliQ

water, and left overnight for the glue to dry. The following morning berry height and berry diameter were

measured with digital calipers. Berry holders were constructed of a 5 mm moulded blue tube (formed

from Xantopren L Blue polysiloxane impression material; Hereaus Kulzer, Hanau, Germany) glued to

the cap of 2 mL microcentrifuge tube with a 2 mm hole drilled through it. The microcentrifuge tube was

fixed into a falcon tube lid with a hole drilled in the top. Berry and artificial pedicel weight, berry holder

weight and assembled weight were measured using an OHAUS Explorer (OHAUS, New Jersey, USA)

80

precision analytical digital balance (increments in 0.0001 gram graduations). The assembled berry

holder was then screwed into a custom transpiration chamber (constructed of a transparent plastic

container with falcon tube threads glued onto holes in the container lid, filled with 140 g of dried silica

gel, holding up to 24 berries). This chamber seeks to maintain a constant relative humidity of 20%;

temperature was maintained at 20 °C. After 24 hours, the berry holder assembly was unscrewed from

the transpiration chamber and the following measurements were taken; assembled weight, berry holder

weight, berry and tube weight, berry TSS. Any berries which showed signs of poor seal between

pedicel and tubing (e.g. water coming out of artificial pedicel junction, air bubbles in tubing), were

excluded from analysis. Berry volume and surface area was calculated from berry diameter and height,

approximating berry shape as a prolate sphere. Transpiration rate (µmol m-2 s-1), water uptake rate

(mmol m-2 s-1), and net flux rate (rate of net berry water loss or gain; µmol m-2 s-1) were then calculated.

4.2.7 Biomechanical testing

Physical testing of berries was undertaken using an Instron materials testing machine (as per Chapter

2.2.5) and skin fold calipers (as per Chapter 3.2.3).

4.2.8 Botrytis cinerea infection assay

An infection assay was developed with and without wounding to determine the rate of Botrytis cinerea

infection spread and the impact of Ca nutrition, based on the mycelial plug method described in

Deytieux-Belleau et al. (2009). The infection assay without wounding was intended to estimate the rate

of infection spread on intact berries, including penetration of the cuticle and epidermal cell walls by

mycelia (e.g. as would occur when an uninfected berry comes into contact with an infected berry). The

use of mycelial plugs rather than spore culture droplets provided a high level of active mycelia and a

nutrient source. This allowed assessment of the rate of mycelial spread through the cell wall rather than

a measure of the rate of spore germination and pathogen development on the berry surface.

Fresh Botrytis cinerea culture plates were prepared by scraping spores across potato dextrose agar

plates (40 g/L H2O, autoclaved at 121 °C for 20 minutes) 24 hours prior to its application to fruit. Using

a 4 mm hole punch, mycelial plugs were stamped out on the leading edge of mycelial growth from the

fresh plates. Assays were carried out in 12-well white ceramic plates; mycelial plugs were carefully

inverted onto berries (with or without wounding). Wounding consisted of a single hole penetrating the

berry skin and underlying the mesocarp using a syringe needle. The mycelial plug was placed over the

puncture wound. A negative control was prepared using agar plugs from uninfected plates. Progression

81

of infection spread was scored visually every 24 hours, with the percentage of berry surface area

discoloured (browning) due to infection classified into the following classes; 0 = 0%, 1 = 1-25%, 2 = 26-

50%, 3 = 51-75%, 4 = 76-100% of total berry surface area.

4.3 Results

4.3.1 Cell wall modifications from veraison to late harvest

Immunolabelling and TEM of pectin epitopes in grapes revealed interesting differences between BNS,

HCS, and LCS treatments in Shiraz and Chenin Blanc at 100 DAA. Immunolabelled gold particles can

be observed as black dots in TEM images. In the epidermal cells, de-esterified pectin (labelled with

LM19) was found predominantly in the cuticle of Shiraz epidermal sections (Figure 4.1A). Chenin Blanc

exhibited much lower cuticle labelling intensity of LM19 compared to Shiraz, with labelling occurring

predominantly in the epidermal cell wall (Figure 4.1B). LM19 cuticle labelling was slightly higher in the

low calcium treated Chenin Blanc than basal and high calcium treated Chenin Blanc (Appendix 6.1). In

contrast, fully esterified pectin (labelled with LM20) was not found in the cuticle; instead its distribution

in the epidermal zone was clearly restricted to the cell walls (Figure 4.2). The pattern with LM20 was

consistent across all treatments and varieties.

Distribution of pectin epitopes in the hypodermal layers was also studied at 100 DAA in both varieties.

De-esterified pectin was distributed throughout the hypodermal cell walls, with areas of stronger LM19

labelling in the middle lamella (Appendix 6.1D-F, J-L). Esterified pectin labelling with LM20 was evenly

distributed throughout the hypodermal cell walls, with no pattern of stronger staining in the middle

lamella (Appendix 6.2D-F, J-L). Slightly stronger labelling of both LM19 and LM20 in the hypodermis

was observed in Chenin Blanc compared to Shiraz; no discernible differences were observed between

treatments.

The LM19 epitope was used to study changes in epidermal de-esterified pectin localisation in Shiraz

throughout development. In early post-veraison (75 DAA) Shiraz LM19 was found predominantly in the

outer epidermal cell wall, with much lower labelling intensity observed in the cuticle (Figure 4.3A). By

harvest point (100 DAA) the de-esterified pectin label was observed much more strongly in the cuticle

(Figure 4.1A). This trend continued til a later harvest date (118 DAA), with higher staining intensity

found in the cuticles (Figure 4.3B). High calcium grown samples appeared to have slightly higher LM19

labelling intensity in the outer cell wall at 118 DAA, with cuticle labelling mostly confined to the inner

82

cuticle layer (Appendix 6.3H). The distribution of esterified pectin (LM20) epitope was largely

constrained to the epidermal cell wall throughout development (Appendix 6.4). However, some low level

cuticle labelling was observed in the low calcium grown late harvest sample (Appendix 6.4F).

83

Figure 4.1 TEM images of Shiraz and Chenin Blanc berry epidermal and hypodermal tissue grown

under basal conditions at 100 DAA immunogold labelled with LM19. LM19 binds de-esterified pectin

Shiraz (A) and Chenin Blanc (B) demonstrate differences in labelling pattern. Immunogold labelling

present as small black dots, examples are indicated by white arrows. cu indicates cuticle. p indicates

primary cell wall. Multiple berries (at least 3) were examined for each treatment. Representative images

are shown. Scale bars represent 2 µm.

84

Figure 4.2 TEM images of Shiraz and Chenin Blanc berry epidermal and hypodermal tissue grown

under basal conditions at 100 DAA immunogold labelled with LM20. LM20 binds esterified pectin.

Shiraz (A) and Chenin Blanc (G) demonstrate similar labelling patterns. Immunogold labelling present

as small black dots, examples are indicated by white arrows. cu indicates cuticle. p indicates primary

cell wall. Multiple berries (at least 3) were examined for each treatment. Representative images are

shown. Scale bars represent 2 µm.

85

Figure 4.3 TEM images of Shiraz berry epidermal tissue grown under basal calcium conditions at post-

veraison (75 DAA; A), and late harvest (118 DAA; B) immunogold labelled with LM19. LM19 binds de-

esterified pectin. Immunogold labelling present as small black dots, examples are indicated by white

arrows. cu indicates cuticle. p indicates primary cell wall. Multiple berries (at least 3) were examined for

each treatment and time point. Representative images are shown. Scale bars represent 2 µm.

86

4.3.2 Fresh berry staining

Changes in Shiraz FDA staining throughout development were very similar in whole berry, skin and

mesocarp tissues, with a decrease in staining intensity occurring after veraison. Representative images

of stained Shiraz and Chenin Blanc berries grown in BNS at veraison (58 DAA) and late harvest (110

and 114 DAA, respectively) are presented (Figure 4.5). Closer inspection of the images reveals that the

difference between Chenin Blanc and Shiraz skin staining patterns detected by image analysis is due to

prominent staining of the Chenin Blanc peripheral vasculature. Example images demonstrate the

intensity of PI staining of epidermis and peripheral vasculature of Chenin Blanc relative to Shiraz in

berries sampled from BNS grown vines (Figure 4.6).

As the mesocarp occupies the vast majority of cut berry surface area, it is unsurprising that such close

resemblance is seen between mesocarp and whole berry staining patterns. Both BNS and LCS grown

Shiraz exhibit a rapid drop in mesocarp FDA staining from 100% at 49 DAA to ~75% at 63 DAA;

veraison occurred at ~55 DAA (Figure 4.7E). Subsequent decreases in FDA staining intensity were

more gradual in BNS than in LCS. HCS Shiraz demonstrated a later drop in FDA staining, consistent

with the delay in veraison. Interestingly, Chenin Blanc did not exhibit the same drop in FDA staining

intensity concurrent with veraison, instead maintaining at least 90% mesocarp staining intensity

throughout development (Figure 4.7E). Levels of FDA staining were lower in skin tissue than in

mesocarp across all time points (Figure 4.7C).

BNS and LCS grown Shiraz exhibited similar changes in PI staining intensity throughout development;

PI staining intensity of BNS Shiraz mesocarp decreased from 60% at 35 DAA to 16% at 64 DAA (Figure

4.7F). HCS Shiraz PI staining patterns were different; staying between 65 - 82% until 78 DAA, a rapid

drop to 28% occurred at 93 DAA (Figure 4.7F). Chenin Blanc exhibited a pattern of steadily decreasing

mesocarp PI staining from 75% at 64 DAA to 32% at 108 DAA. Visual observations of Chenin Blanc PI

images indicated that there were much higher levels of PI staining in the skin tissues than was

observed in Shiraz; this was supported by the data from the image analysis. PI skin staining was

consistently higher in BNS Chenin Blanc (32-42%) than BNS Shiraz (maximum 11% during the same

period) (Figure 4.7D). Unlike Shiraz, Chenin Blanc exhibited no discernible decrease in skin PI staining

intensity throughout development.

87

Fig

ure

4.5.

Rep

rese

ntat

ive

imag

es o

f sta

inin

g pa

ttern

s of

fres

h be

rrie

s. B

errie

s sa

mpl

ed fr

om v

ines

gro

wn

hydr

opon

ical

ly in

bas

al

nutr

ient

sol

utio

n. S

hira

z at

ver

aiso

n (5

8 D

AA

; A-C

) an

d ha

rves

t (11

0 D

AA

; D-F

). C

heni

n B

lanc

at v

erai

son

(58

DA

A; G

-I)

and

harv

est

(114

DA

A; J

-L).

Com

bine

d P

I and

FD

A s

tain

ing

(A, D

, G, J

), P

I sta

inin

g (B

, E, H

, K),

FD

A s

tain

ing

(C, F

, I, L

). S

cale

bar

rep

rese

nts

5

mm

.

88

Figure 4.6. Representative images demonstrating FDA and PI staining of berry skin and peripheral

vasculature. Berries sampled from vines grown hydroponically in basal nutrient solution. Panels

demonstrate combined PI and FDA staining (A, D, G), PI staining (B, E, H), and FDA staining (C, F, I)

of Shiraz at 30.6 Brix (A-C), and Chenin Blanc at 28.7 °Brix (D-F) and 23.5 °Brix (G-I). Panels G-I are

provided as further demonstration of the high levels of peripheral vascular staining observed in Chenin

Blanc. Scale bar represents 500 µm.

89

Figure 4.7. Cell vitality and pectin staining of Shiraz and Chenin Blanc fresh berries. Fluorescein

diacetate (A, C, E) and propidium iodide (B, D, F) fluorescence as % of; whole berry area (A, B), skin

area (C, D), and mesocarp area (E, F). n = 3 vines per treatment, 3-5 berries per vine, per time point.

Mean ± SD.

4.3.3 Berry electrical impedance spectroscopy

Using electrical modelling, components representing extracellular resistance (Re), intracellular

resistance (Ri), and cell membrane capacitance (Cm) were extracted from berry resistance/reactance

data of berries sampled at 95 DAA. No significant difference in Re was observed between treatments.

However, a significant difference between varieties was observed, with Chenin Blanc demonstrating

higher extracellular resistance than Shiraz (Figure 4.8A). Significant differences in Ri were found

D A A

FD

A s

tain

(%

be

rry

are

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

D A A

PI

sta

in (

% b

erry

are

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

D A A

FD

A s

tain

(%

sk

in a

re

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

D A A

PI

sta

in (

% s

kin

are

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

D A A

FD

A s

tain

(%

pu

lp a

re

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

S h ira z L C S

S h ira z B N S

S h ira z H C S

C h e n in B la n c B N S

D A A

PI

sta

in (

% p

ulp

are

a)

3 5 4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

2 0

4 0

6 0

8 0

1 0 0

A B

C D

E F

90

between treatments, but not between varieties (Figure 4.8B). However, this may be due to differences

in sugar content; intracellular resistance is positively correlated with sugar levels (Caravia Bayer et al.,

2015). A correlation between TSS and Ri in these samples was characterised (R2=0.6011) in Appendix

7.

Figure 4.8. Modelling of resistance and capacitance components of Chenin Blanc and Shiraz grapes

across calcium treatments, harvested at 95 DAA. Components include extracellular resistance (Re; A),

intracellular resistance (Ri; B), and cellular membrane capacitance (Cm; C). n = 3-4 vines per

treatment, 3 berries per vine. Mean ± SD.

Re

(

)

C h e n in B la n c S h ira z

0

1 0 0 0 0

2 0 0 0 0

3 0 0 0 0

B N S

L C S

H C S

H C S V e ra is o n

H C S G re e n

Ri

( )

C h e n in B la n c S h ira z

0

1 0 0

2 0 0

3 0 0

4 0 0

B N S

L C S

H C S

H C S V e ra is o n

H C S G re e n

Cm

(F

)

C h e n in B la n c S h ira z

0

1 .01 0 -9

2 .01 0 -9

3 .01 0 -9

4 .01 0 -9

B N S

L C S

H C S

H C S V e ra is o n

H C S G re e n

A

C

B

91

4.3.4 Berry physical changes

Berry biomechanical properties were investigated using both an Instron materials testing machine and

skinfold calipers to determine berry hardness, deformability, and elasticity. Young’s modulus of

elasticity was calculated both manually from the raw data (Figure 4.9A), and automatically by the

Instron (Figure 4.9B). The Instron automatic calculations do not take into account differences in size

and shape of berries, and were therefore somewhat inaccurate. However, they do demonstrate similar

trends to what is observed using manual calculations (Figure 4.9A). Differences between time points

and treatments were calculated using two way analysis of variance with Tukey’s multiple comparison

test (P<0.05). All treatments demonstrated very similar berry elasticity pre-veraison (49 DAA). The

elasticity of BNS and LCS berries increased significantly (i.e. lower Young’s Modulus values) from 49 to

64 DAA. This berry elasticity increase was delayed in HCS berries, concurrent with the delay in

reaching veraison (Figure 4.9A, B, C). By 93 DAA, elasticity levels of HCS berries were once again

similar to BNS and LCS berries. Significant increases in berry deformability (measured with skinfold

calipers) after veraison were observed in BNS and LCS berries; this increase was also delayed in HCS

berries (Figure 4.9D). LCS berries exhibited a larger increase in berry deformability than BNS or LCS

from 93 DAA to 108 DAA, concurrent with onset of shrivel in LCS berries at this time (Figure 4.9D).

Maximum berry compressive load of HCS berries was on average slightly lower than BNS and LCS

berries; this is probably due at least in part to the smaller berry size (Figure 4.9E). Both BNS and LCS

berries showed a rapid increase in skin maximum compressive load from 43 DAA to 64 DAA, with

subsequent decreases. At 64 and 93 DAA HCS berries had much significantly lower skin maximum

compressive load than the other treatments (Figure 4.9F). HCS berries showed slow but steady

increases in skin maximum compressive load throughout development, reaching similar levels to BNS

berries at 93 DAA.

92

Figure 4.9. Physical properties of Shiraz from veraison to late harvest grown under different calcium

regimes. High (HCS), basal (BNS), and low (LCS) grown measurements of; berry elasticity (manual

calculation from Instron data; A), berry elasticity (Instron automatic calculation; B), berry elasticity

(skinfold caliper calculation; C), berry deformability (D), berry maximum compressive load (E), skin

maximum compressive load (F). Instron measurements were taken from 49-93 DAA (days after

anthesis). Skinfold caliper measurements were taken from 49-123 DAA. No data was collected from

LCS treatment at 123 DAA due to the severe shrivel observed in these berries, making measurements

impractical. n = 3 replicate vines per treatment per timepoint, 5 berries per vine. Mean ± SD.

D A A

Be

rry

Yo

un

g's

Mo

du

lus

(in

str

on

ma

nu

al

ca

lc)

(MP

a)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

5

1 0

1 5

2 0

2 5

D A A

Be

rry

Yo

un

g's

Mo

du

lus

(in

str

on

au

to c

alc

) (M

Pa

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .0

0 .5

1 .0

1 .5

L C S

B N S

H C S

D A A

Be

rry

ma

xim

um

co

mp

re

ss

ive

lo

ad

(J

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .0 0

0 .0 1

0 .0 2

0 .0 3

0 .0 4

D A A

Sk

in m

ax

imu

m

co

mp

re

ss

ive

lo

ad

(J

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .0 0 0

0 .0 0 2

0 .0 0 4

0 .0 0 6

0 .0 0 8

A

C

E F

D

D A A

Be

rry

de

form

ati

on

(% d

iam

ete

r s

kin

fold

ca

lip

ers

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0

1 0

2 0

3 0

D A A

Be

rry

Yo

un

g's

Mo

du

lus

(sk

info

ld c

ali

pe

rs

) (M

Pa

)

4 9 6 4 7 8 9 3 1 0 8 1 2 3

0 .1

1

1 0

1 0 0

B

93

4.3.5 Berry water relations

Berry transpiration assays at pre-veraison (49 DAA), post-veraison (78 DAA), and harvest maturity (108

DAA) in Shiraz were performed. Transpiration rate of HCS berries was significantly lower than BNS

berries at all three time points. Transpiration rate of LCS berries was significantly higher than BNS

berries at 78 DAA (Figure 4.10A). HCS berries exhibited significantly lower water uptake than BNS

berries at 49 DAA, no differences were discernible post-veraison (Figure 4.10B). LCS berries exhibited

more negative net flux rates (i.e. greater net water loss) than BNS berries in this assay at 49 and 78

DAA (Figure 4.10C).

Comparisons of Shiraz and Chenin Blanc berry water relations at harvest maturity (108 DAA) were

made. LCS Chenin Blanc berries had significantly higher transpiration and water uptake rates than all

other samples (Figure 4.11A, B). BNS grown Chenin Blanc berries had relatively low transpiration and

water uptake rates. Although no significant net water balance differences were observed between

treatments in either variety, BNS Chenin Blanc berries had significantly less net water loss than all

three Shiraz treatments (Figure 4.11C).

Figure 4.10. Berry rehydration assay of Shiraz berries. Transpiration rate (A), water uptake rate (B), and

net flux rate (C), of individual Shiraz berries grown in high, basal and low calcium conditions was

measured. Rates were calculated from berry weight loss and berry water uptake after 24 hours in

custom transpiration chambers. n = 3 replicate vines per treatment per time point, 3-6 berries sampled

per vine. * indicates significant difference from BNS using two way ANOVA with Dunnett post-hoc test

(p<0.05).

D A A

Tra

ns

pir

ati

on

ra

te

(mm

ol

m-2

s-1

)

4 9 7 8 1 0 8

0 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

0 .2 5

S h ira z B N S

S h ira z L C S

S h ira z H C S

*

*

* *

D A A

Wa

ter u

pta

ke

ra

te

(mm

ol

m-2

s-1

)

4 9 7 8 1 0 8

0 .0

0 .1

0 .2

0 .3

*

D A A

Ne

t fl

ux

ra

te

(

mo

l m

-2s

-1)

4 9 7 8 1 0 8

-1 0 0

-5 0

0

5 0

* *

*

A B C

94

Figure 4.11. Berry rehydration assay of Shiraz and Chenin Blanc berries at 108 DAA. Transpiration rate

(A), water uptake rate (B), and net flux rate (C), of individual berries grown in high, basal and low

calcium conditions was measured. Rates were calculated from berry weight loss and berry water

uptake after 24 hours in custom transpiration chambers. n = 3-4 replicate vines per treatment, 3-6

berries sampled per vine. Mean + SD. Letters indicate significant difference using two way ANOVA with

Tukey's post-hoc test (p<0.05).

4.3.6 Botrytis cinerea susceptibility

No treatment effect upon initial susceptibility to infection (i.e. time taken to show initial signs of infection)

was observable in Chenin Blanc berries at the resolution offered by this assay; all berries showed initial

signs of infection within the first 24 hours, with or without wounding (Figure 4.12). However, a treatment

effect was observed over the full time course of the assay in both intact (Figure 4.12A) and wounded

(Figure 4.12B) berries when a two way ANOVA was performed (P<0.05). Infection had spread further in

LCS berries than in both BNS and HCS berries at all test points after initial infection; this suggests that

the rate of infection spread is higher in the low calcium berries than those with basal or high calcium

levels.

Botrytis cinerea assays were also carried out in Shiraz berries; although these results indicate similar

patterns to Chenin Blanc, a limited number of replicates prevent statistical analysis (Appendix 8).

Tra

ns

pir

ati

on

ra

te

(mm

ol

m-2

s-1

)

S h ira z C h e n in B la n c

0 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

0 .2 5

a

bb c

b cc

d

L C S

B N S

H C S

Wa

ter u

pta

ke

ra

te

(mm

ol

m-2

s-1

)

S h ira z C h e n in B la n c

0 .0 0

0 .0 5

0 .1 0

0 .1 5 a

b

b c

b

cc

Ne

t fl

ux

ra

te

(

mo

l m

-2s

-1)

S h ira z C h e n in B la n c

-1 0 0

-8 0

-6 0

-4 0

-2 0

0

a cb cc

a b

c

a

A B C

H o u rs a fte r in n o c u la t io n

Infe

cti

on

sta

ge

2 4 4 8 7 2 9 6

0

1

2

3

4

B

C h e n in L C S

C h e n in B N S

C h e n in H C S

+ w o u n d in g

H o u rs a fte r in n o c u la t io n

Infe

cti

on

sta

ge

2 4 4 8 7 2 9 6

0

1

2

3

4

A

N o in n o c u la tio n c o n tro l

C h e n in L C S

C h e n in B N S

C h e n in H C S

- w o u n d in g

95

Figure 4.12. Botrytis cinerea assay of Chenin Blanc berries grown under different calcium regimes.

Assay was carried out over 96 hours, with inspection of berries every 24 hours after Botrytis cinerea

inoculation. Assay was performed on intact (A) and wounded (B) berries. Berries were scored by visual

assessment of percentage berry surface area discoloured/infected; 0 = 0%, 1 = 1-25%, 2 = 26-50%, 3 =

51-75%, 4 = 76-100%. Mean + SD. n = 4 replicate vines, 6 berries per vine.

4.4 Discussion

In this study the importance of calcium availability for determining berry properties has been examined.

The complex nature of calcium’s role in fruit ripening will be addressed with specific focus on the

impacts of calcium nutrition on pectin distribution, fruit softening, fruit water relations and pathogen

susceptibility.

4.4.1 Effect of calcium on pectin distribution

Calcium is able to cross link de-esterified pectin, modifying cell wall properties (Willats et al., 2001).

However, pectin de-esterification without formation of calcium cross-links can lead to increased

solubilisation of pectins and irregular distribution of solubilised fragments (Decreux and Messiaen,

2005; Brummell, 2006). Interestingly, in this study de-esterified pectin was found in the cuticle of Shiraz

and Chenin Blanc berries. Low calcium growth resulted in higher levels of de-esterified pectin in the

cuticle. Reduced calcium supply resulted in multiple changes in pectin modification and solubilisation,

as well as modifying berry water relations. The lower degree of ionic cross-linking in low calcium grown

fruit would presumably allow greater access of polygalacturonases (PGs) to the pectin backbone. The

de-polymerisation of the pectin backbone creates smaller fragments of soluble pectin (Brummell, 2006).

The disassembly of pectin structures in xylem vessels may have contributed to the relatively high berry

transpiration rate (Figure 4.10 and 4.11) of low calcium berries, resulting in movement of soluble pectin

fragments out of the epidermal cell wall and into the cuticle. In Shiraz berries grown under high calcium,

de-esterified pectin is retained in the cell wall (rather than cuticle). The high levels of available calcium

may cross-link de-esterified pectins and limit access of PGs to the pectin backbone. High calcium

supply also reduced berry transpiration. These factors combine to result in low solubilisation of pectin,

increasing retention of pectin within the epidermal cell wall with minimal distribution into the cuticle.

Previous characterisation of the tomato cuticle showed presence of pectin, hemicellulose, and cellulose

(Lopez-Casado et al., 2007). High levels of pectin methyl esterases (PME) have been observed in the

96

outer pericarp tissue of tomato at the breaker stage; the combined action of PMEs during breaker

stage, and PGs during ripening lead to a reduction in total insoluble pectin within the cell wall (Blumer

et al., 2000). As pectin from the primary cell wall of epidermal cells is de-esterified by PMEs, the

release of hydrogen ions locally reduces the pH, stimulating polygalacturonase and pectate lyase

activity (Bosch and Hepler, 2005). Polygalacturonase activity in Sauvignon Blanc and Semillon

increased during ripening (Cabanne and Doneche, 2001). Additionally, cell wall modifying enzyme

expression changes throughout development. In grape mesocarp PME expression increases from

stage II of berry development (veraison), with subsequent increases in levels of pectate lyase

expression (Schlosser et al., 2008). These expression changes were somewhat delayed in exocarp

tissue, suggesting that berry growth during ripening may be controlled and/or limited by cell wall

loosening and expansion in the exocarp tissue. Fragmentation and solubilisation of long chain pectins

during ripening caused by this change in PME activity may lead to distribution of pectin labelling into

ectopic locations such as the cuticle.

Studies in tomato demonstrated that the presence of polysaccharides in the cuticle resulted in higher

elastic modulus and higher maximum breaking stress in cuticles from both mature green and red ripe

fruit, relative to cuticles stripped of their polysaccharide content (Lopez-Casado et al., 2007). These

results indicate that the presence of polysaccharides (in this case, a mixture of cellulose, hemicellulose

and pectin) in the cuticle contribute to the stiffening of this matrix, however the contribution of each

polysaccharide component is difficult to determine. A comprehensive timeline of changes in tomato

cuticle mechanical properties is provided in Espana et al. (2014). Differences between two varieties

(‘Cascada’ and ‘Moneymaker’) were observed from green mature to red ripe growth stages, with both

showing a decrease in viscoelastic strain, whilst elastic strain remained relatively constant. Elastic

modulus of Cascada generally increased throughout development, reaching a maximum at red ripe. It

was noted that the ratio of cutin to polysaccharides in the cuticle decreased over development,

although no further insights into the contribution of polysaccharides to the mechanical properties were

offered.

A survey of the chemical composition and mechanical properties of the cuticle in 27 Japanese

persimmon (Diospyros kaki) cultivars revealed a negative correlation between polysaccharide content

and maximum strain, and a positive correlation between polysaccharide content and elastic modulus

(Tsubaki et al., 2012). This indicates that in Japanese persimmon cuticles, polysaccharides serve to

increase the stiffness, reducing the maximum breaking strain. Although these studies looked broadly at

the effect of cuticle polysaccharide content on cuticle biomechanical properties, the impact of the

97

presence of particular cell wall components (e.g. de-esterified pectin fragments) was not elucidated.

Additionally, the impact of cuticle composition changes on water permeability and pathogen

susceptibility were not investigated.

In the work presented here, it was observed that Chenin Blanc maintains higher levels of de-esterified

pectin in the skin throughout ripening than Shiraz (Figure 4.7D). Varietal differences in expression of

PMEs create differences in the distribution of de-esterified pectin in the skin. Propidium iodide (PI)

stains de-esterified pectin; fresh berry staining with PI was performed to study distribution of de-

esterified pectin. Chenin Blanc showed much higher levels of PI staining in the skin throughout

development; closer examination revealed strong staining in both the epidermal cell walls and the

underlying peripheral vasculature (Figure 4.6). High levels of de-esterified pectin in the berry

vasculature may affect water movement and distribution through the berry, contributing to low shrivel

susceptibility in Chenin Blanc. Indeed, under low calcium conditions, Chenin Blanc exhibited higher

berry transpiration and water uptake rates than basal or high calcium Chenin Blanc, or all treatments of

Shiraz (4.11A, B). Both basal and high calcium grown Chenin Blanc exhibited lower net water flux than

low calcium Chenin Blanc (Figure 4.11C). This may indicate that the higher levels of de-esterified pectin

in Chenin Blanc vasculature affect this variety’s ability to moderate berry water relations; under low

calcium, high berry transpiration and water uptake are observed, when sufficient calcium is present,

berry transpiration and water uptake is reduced.

High resolution TEM of hypodermal cell walls revealed the esterified pectin epitopes were distributed

fairly evenly throughout the primary cell walls. De-esterification of pectin was more highly observed in

the middle lamella and tri-cellular junctions of apples, consistent with the postulated role of pectin

modification in cell-cell adhesion (Roy et al., 1995; Willats et al., 2001).

Previous work has looked at some of the chemical composition changes in berry cell walls during

ripening. Nunan et al. (1998) indicated that berry galacturonan (the main pectin residue) content

increases during ripening, with the level of de-esterification decreasing slightly pre-veraison then

remaining relatively constant thereafter. This work also found that the level of type I arabinogalactan (a

pectic side chain) as a mol % of total cell wall extract decreased from 21% pre-veraison (44 DAA) to 4%

in ripe (114 DAA) berries (Nunan et al., 1998). Comparison of Thompson Seedless and NN107 (a

variety which demonstrates reduced softening) revealed that NN107 had higher levels of

homogalacturonan (including de-esterified homogalacturonan) at late ripening stages, correlated with

higher PME activity (Balic et al., 2014). These results indicate that a range of pectin modifications may

be important in ripening and softening processes.

98

In two apple cultivars, both de-esterified and esterified pectin were localised to the middle lamella (Ng

et al., 2013). A decrease in the molecular weight of chelator soluble pectin in Scifresh was postulated to

be related to the increased cell expansion/larger cell size observed in this cultivar (Ng et al., 2013). In

apples, PME activity has been shown to increase during growth and decrease during ripening. The

firmer variety, Scifresh also exhibited greater cell-cell adhesion and occurrence of esterified pectin in

tricellular junctions. In homogalacturonan with a high degree of esterification, gels can form through

hydrogen bonds and hydrophobic interactions (Braccini et al., 2005). The higher level of PME activity

and pectin de-esterification in Royal Gala may have allowed for greater access to the pectin matrix for

pectin degrading pectin lyases and polygalacturonases. Their action would decrease the amount of

pectin in the walls, resulting in rapid softening.

Studies in grapes demonstrated that the activities of polygalacturonases was inversely correlated to

calcium content; enzyme inhibition was also observed in vitro in the presence of 1 mM calcium

(Cabanne and Doneche, 2001). The mechanism of this inhibition may be direct binding of the enzyme

or binding of homogalacturonan restricting enzyme access to the pectin substrate. In tomatoes, PME

activity in skin tissues has been linked with calcium binding and increased cell-cell adhesion (Hyodo et

al., 2013). Pectin de-esterification in skin and mesocarp tissues increased throughout development.

Total calcium content of alcohol insoluble residue cell wall extracts was much higher in skin tissue than

mesocarp. Skin cell wall calcium content increased throughout development, until decreasing slightly

when overripe. Mesocarp cell wall calcium content remained fairly constant throughout development. It

is clear that a complex relationship exists between pectin, PME activity, calcium availability, and pectin

degrading enzyme activity. The differential tissue localisation of pectin modifying enzymes and calcium

facilitates developmental changes in cell wall properties, cell-cell adhesion, water permeation through

the cell wall and the cuticle, and ostensibly fruit softening.

4.4.2 Effect of calcium on berry softening

Shiraz berry elasticity was shown to increase in all treatments concurrent with veraison; this increase in

berry elasticity was delayed in HCS berries, consistent with the delayed onset of veraison in this

treatment (Figure 4.9A, B, C). Berry deformability increased at veraison in all treatments; a large

subsequent increase in LCS berry deformability was indicative of the early onset of berry shrivel

occurring in these bunches (Figure 4.9D). Skin properties changed throughout development and were

affected by calcium levels. A large increase in skin maximum compressive load was observed at

veraison in BNS and LCS berries, followed by slight reductions in later ripening (Figure 4.9F). In

contrast, HCS showed a gradual increase in skin maximum compressive load, reaching similar

99

elasticity levels to BNS berries at 93 DAA. Both cell wall properties and cell turgor contribute to

biomechanical properties of the berry (Redgwell et al., 1997; Matthews et al., 2009). By utilising two

different test protocols, differences between whole berry and skin properties were elucidated here. The

effect of high calcium upon skin physical properties may further confirm that calcium nutrition affects

cell wall modification and pectin distribution. Potentially, high calcium leads to higher calcium-pectin

cross linking, restricting access of cell wall degrading enzymes, delaying softening related changes in

cell wall architecture, and reducing elasticity and skin maximum compressive load.

One unusual table grape variety (NN107) demonstrates an increase in berry firmness during late

ripening, and shows higher PME activity and calcium content than Thompson Seedless (Balic et al.,

2014). The authors hypothesised that in NN107, PME activity during late ripening stages increases the

prevalence of calcium-pectin crosslinkages, reducing cell wall access of polygalacturonases,

consequently increasing firmness. This is supported by previous work which showed a negative

correlation between calcium content and polygalacturonase activity in Sauvignon Blanc and Semillon

grapes (Cabanne and Doneche, 2001). Tomato fruit with transgenically suppressed expression of

expansin gene Exp1 showed increased firmness during ripening; pectin depolymerisation was

substantially decreased in these fruit (Brummell et al., 1999). The degree of pectin de-esterification and

depolymerisation varies widely between species, as such the role of calcium in softening processes will

also differ (Brummell, 2006).

It has also been shown that some measurements of fruit elasticity or berry firmness can be highly

correlated with cell turgor pressure and berry water relations (Matthews et al., 2009). The impact of

calcium nutrition on berry water relations is discussed below.

4.4.3 Effect of calcium on berry water relations

Hydrostatic gradient from the xylem in the pedicel to the terminal xylem vessels in the berry is required

for effective xylem flow into the berry. It has previously been hypothesised that as the berry develops,

physical disruption of xylem connectivity and/or loss of the hydrostatic gradient into the berry, causes a

reduction, if not cessation, of xylem flow into the berry around the time of veraison (Chatelet et al.,

2008). The results presented here suggest that cell wall modifications in the berry vasculature and

epidermis also impact upon water delivery and distribution in the berry. Higher levels of de-esterified

pectin present in the peripheral vasculature of Chenin Blanc berries may enable formation of pectin

gels when sufficient calcium is present, reducing water distribution to the epidermis and subsequent

water loss through the berry surface. In basal and high calcium treatments, Chenin Blanc demonstrated

100

lower net water loss than the respective treatments in Shiraz (Figure 4.11). In low calcium, Chenin

Blanc demonstrated much higher berry water uptake and berry transpiration rates than basal or high

calcium treatments, however the greater net water loss exhibited in these berries resulted in reduced

berry weight and shrivel at later stages. That this pattern was not followed in Shiraz suggests that both

pectin de-esterification and calcium concentration in the peripheral vasculature play interacting roles in

regulating berry water relations.

It has been shown previously that kiwifruit transpiration rate decreased rapidly from growth stage II,

concurrent with cessation of calcium accumulation (Montanaro et al., 2006). This may suggest that

cumulative fruit transpiration is a very good predictor for calcium accumulation in fruit (Montanaro et al.,

2012). However, changing growth conditions would disrupt this relationship as transpiration is not the

only factor affecting calcium accumulation. For instance, in apricot fruit, bagging of fruit to prevent fruit

transpiration resulted in a 55% reduction in fruit calcium, suggesting that 45% accumulates

independent of transpiration (Montanaro et al., 2010). Modelling of kiwifruit transpiration rates indicated

that transpiration varies diurnally and decreases throughout the growing season; relative humidity, air

temperature, and skin conductance were significant factors, whereas wind speed and light were not

(Montanaro et al., 2012). The model suggests that reducing relative humidity and increasing

temperature in the orchard may result in increased fruit calcium content. Similar work in grapes also

demonstrated diurnal patterns of berry volume increases and water loss, implicating xylem backflows

and surface transpiration as the two sources of water loss from berries (Lang and Thorpe, 1989).

4.4.4 Effect of calcium on cell vitality

Using FDA as a cell vitality stain and analysing staining patterns with ImageJ, FDA staining intensity of

Shiraz berries was observed to decrease at veraison in skin and mesocarp, with high calcium treatment

delaying the drop in staining intensity (Figure 4.7C, E). FDA staining intensity in Chenin Blanc berries

was maintained throughout development (Figure 4.7C, E). FDA staining of fresh berries has been used

previously to determine cell vitality (Krasnow et al., 2008; Tilbrook and Tyerman, 2008). FDA is

membrane permeable, and upon entering the cytoplasm, esterases cleave acetate groups to produce

the membrane impermeable fluorescein, which emits a bright fluorescent signal within the cytoplasm.

For determination of cell vitality this technique relies on the presence of active esterases limited to the

cytoplasm of vital cells and the maintenance of membrane impermeability to fluorescein. Recent work

has indicated that membrane integrity is maintained even during periods of “cell death”, this suggests

that observations of cell death or vitality loss may in fact be due to increases in membrane porosity

(Caravia Bayer et al., 2015). With this in mind, it should be noted that these (and previous)

101

observations of vitality loss using FDA may rather be indicators of increases in plasma membrane

porosity. This technique has indicated that normally developing berries maintain high levels of

mesocarp cell vitality up to commercial harvest dates, whereas shrivelling berries showed a rapid loss

of cell vitality (Krasnow et al., 2008). Varietal differences in cell vitality were observed at a tissue level

between the table grape Thompson Seedless, and the shrivel-susceptible wine grape variety Shiraz

(Tilbrook and Tyerman, 2008). A large varietal survey suggested that there was a high correlation

between levels of cell vitality loss and susceptibility to shrivel (Fuentes et al., 2010). A bilinear model

has been used to describe cell death dynamics (Bonada et al., 2013b). Key factors used were; date of

onset of rapid cell death, rates of cell death before and after onset of rapid cell death. Two

environmental factors were proposed to affect cell death; high temperature can advance the onset of

rapid cell death and the rate of rapid cell death, water stress can increase the rate of rapid cell death

(Bonada et al., 2013a; Bonada et al., 2013b). These relationships were not consistent between

varieties; Chardonnay did not demonstrate a detectable period of rapid cell vitality loss under control

conditions, and although heat treated Chardonnay did demonstrate rapid cell vitality loss there was no

concomitant increase in the rate of shrivel (Bonada et al., 2013b). The lack of a linear relationship

between shrivel and FDA staining in Chardonnay (Bonada et al., 2013b) may indicate that; (a) the FDA

staining intensity methodology used is not a reliable indicator for cell vitality, or (b) berry physiological

factors other than cell vitality are also important for determining potential for berry shrivel.

In Krasnow et al. (2008) assessment of cell viability was calculated as fluorescent area (determined

using a binary threshold and image analysis software) as a percentage of total berry area. Changes in

cell size and membrane permeability are potentially confounding issues in this methodology.

Throughout berry development the permeability of membranes and the relative size of cytosol to

vacuole changes greatly. Both of these factors will affect the intensity of FDA staining in berries. As

such, the resolution of images being analysed and the threshold method for image analysis will

contribute to the determination of “cell vitality”. A more diffuse staining pattern, as apparent in post

veraison mesocarp cells (Figure 4.5), may result in a determination of non-vital tissue. However, this

may occur as a result of increasing membrane permeability during ripening and irregular distribution of

esterases and/or FDA into the apoplast and vacuole, rather than a drop in cell vitality. Furthermore, as

the cells expand rapidly post veraison, the relative contribution of the vacuole to the surface area of cut

berries increases, meaning that cell vitality calculated as a percentage of total surface area may be

skewed by these normal physiological changes. Higher resolution imaging of “non-vital” cells revealed

that FDA staining is still present, albeit more diffuse and with lower intensity than “vital” cells in adjacent

tissues (Figure 4.6). The issues presented here suggest that results of FDA staining and image

102

analysis must be carefully interpreted, and drops in staining intensity are not necessarily representative

of drops in cell vitality.

The counter-stain (propidium iodide) used to confirm cell vitality loss in Tilbrook and Tyerman (2008)

stains both DNA and cell walls. As PI exhibits a fluorescence shift upon DNA binding, specific

microscope filter setups can help distinguish staining of these two components. Predominantly cell wall

staining of PI was observed by Krasnow et al. (2008), and as such PI was subsequently discounted as

a useful vitality stain using their methodology. The imaging setup used in the work presented in this

chapter (see Chapter 4.2.4 for details) captured PI staining of the cell wall with minimal interference

from DNA binding (Figure 4.6). This was determined by the absence of the punctate pattern typical of

nuclear DNA staining.

Electrical impedance spectroscopy has been used to model relative contributions of various

components of plant tissue to water and ionic relations of that tissue (Ando et al., 2014; Caravia Bayer

et al., 2015). The application of this technique to grapes allows for estimation of the resistance to

electrical current offered by extracellular and intracellular pathways, as well as the cellular membrane

capacitance. This methodology utilises modelling of reactance (i.e. changes in impedance as a function

of changes in frequency), as it relates to the preferential passage of high frequency currents through

cellular membranes (intracellular) and the passage of low frequency currents through the apoplast

(extracellular). Chenin Blanc demonstrated higher extracellular resistance in all treatments than Shiraz.

The structure and composition of the extracellular space (such as cell wall thickness, hydration of the

cell wall, and ionic composition) determine extracellular resistance. Potentially, formation of pectin gels

with low hydration in Chenin Blanc may be responsible for the observed difference. Additionally, Shiraz

may exhibit ion leakage from cells, increasing the solute content of the apoplast and decreasing the

resistance. This is supported by the observation that Chenin Blanc had significantly higher cellular

membrane capacitance (Cm; an indicator of membrane integrity) than Shiraz, however no significant

differences were found between treatments in this component (Figure 4.8C). Differences in Cm may be

due to differences in berry size (measured as distance between electrodes) however no correlation was

found between these two variables (Appendix 7). Differences in cell size may contribute to Cm

differences; berries with smaller cells would contain proportionally more membrane in the electrical path

than similarly sized berries with larger cells. Looking more closely at cell and tissue morphology may

help to determine the cause of these differences.

103

4.4.5 Effect of calcium on Botrytis cinerea susceptibility

The cell wall forms a significant barrier to pathogens. Many pathogens secrete cell wall degrading

enzymes to aid infection; as such the composition of fruit cell walls is important in determining pathogen

susceptibility (Lionetti et al., 2012). In the final part of this study, the impact of calcium nutrition on

Chenin Blanc berry susceptibility to the fungal pathogen Botrytis cinerea was examined. The results

demonstrated that low calcium growth led to increased spread of Botrytis cinerea infection (Figure

4.12). Most previous studies have focused on the effect of external calcium applications upon fruit

Botrytis cinerea susceptibility, and this is believed to be the first study demonstrating increased

susceptibility associated with reduced calcium nutrition.

Calcium applications to bunches at pre- and post-veraison phases were found to decrease occurrence

of Botrytis cinerea post-harvest, also increasing berry firmness (Ciccarese et al., 2013). The pre-

veraison treatments were found to be particularly effective; much lower levels of calcium accumulation

(and uptake from sprays) were observed after veraison. Infiltration of grapes with calcium prior to

Botrytis cinerea infection assays did not affect levels of infection, however changes in pectin

fractionation suggest that there is an underlying impact on the progression of the infection (Chardonnet

et al., 1997). Calcium pre-treatment reduced levels of water soluble pectin and chelator soluble (i.e.

calcium bound) pectin whilst increasing levels of acid soluble pectin in Botrytis cinerea infected berries

compared to uninfected controls (Chardonnet et al., 1997). This suggests that calcium cross-linkage of

pectin serves to protect long chain water insoluble pectins from enzymatic degradation, resulting in

preservation of the cell wall integrity, and reduced formation of water soluble and weakly bound short

chain chelator soluble pectins. It should also be noted that the levels of both water soluble and chelator

soluble pectin fractions were higher in both calcium treated and untreated infected berries than they

were in calcium treated and untreated uninfected berries.

Pectin has two modes of plant protection; (1) esterified pectin forms a barrier to pathogen degradation,

and (2) short soluble fragments of de-esterified pectin (oligogalacturonides; often released by pathogen

enzymatic action) elicit plant defence responses. De-esterified pectin is more susceptible to

degradation by pathogen enzymes (Lionetti et al., 2012). As such, some research has focused on

blocking the action of pathogen enzymes by expression of inhibitors; overexpression of pectin methyl

esterase inhibitors in Arabidopsis increased resistance to Botrytis cinerea (Lionetti et al., 2007),

constitutive expression of a polygalacturonase inhibiting protein (VvPGIP1) in tobacco increased

resistance to Botrytis cinerea (Nguema-Ona et al., 2013). Some research has suggested that perturbed

cell wall modelling processes can produce oligogalacturonides that constitutively activate plant defence

104

responses, priming them against pathogen attack (Curvers et al., 2010). In this study, the increased

rate of Botrytis cinerea infection progress in low calcium grown fruit may be attributable to the low level

of calcium crosslinkages, allowing increased accessibility of the pectin matrix to cell wall degrading

enzymes, offering less resistance to pathogen infection.

4.5 Conclusion

Distribution of pectin within the cell wall varies with a number of factors. Calcium nutrition has a

multitude of effects on cell wall biology in fruit development. Through cross-linking of de-esterified

pectins, calcium can affect the degradation and distribution of pectins. High calcium nutrition resulted in

retention of de-esterified pectin within the primary cell wall of outer epidermal cells. Low calcium

nutrition resulted in higher net water loss from berries and distribution of solubilised de-esterified pectin

into the berry cuticle. The role of pectin gel formation in xylem vessels is also discussed; levels of

pectin de-esterification in berry peripheral vasculature are higher in Chenin Blanc than in Shiraz. This

suggests that Chenin Blanc may exert greater control over water transport in the berry during ripening,

explaining its low susceptibility to berry shrivel. The formation of calcium crosslinked pectin gels in

xylem vessels may also explain the differences in berry transpiration and water uptake rates observed

under different calcium growth regimes. Use of fresh berry staining techniques revealed PI as a useful

tool for studying de-esterified pectin distribution, and highlighted some reliability issues with FDA as a

quantifiable cell vitality stain. The role of calcium in pathogen susceptibility was also investigated; low

calcium nutrition led to higher Botrytis cinerea susceptibility, due to increased pathogen degradation of

the pectin matrix. This study has highlighted the importance of understanding pectin-calcium

interactions and their impact upon grape berry development and physiology.

105

Chapter 5 General Discussion and Conclusion

5.1 Grapevine calcium nutrition

Calcium is a plant macronutrient with a variety of important roles in plant physiology. The impact of

calcium nutrition upon grapevines is poorly understood, soil calcium availability is not often obviously

limiting or detrimental to plant growth. However, identification of sub optimal calcium nutrition as a key

factor in several fruit disorders (e.g. blossom end rot in tomatoes and bitter pit in apples) suggests that

a better understanding of calcium nutrition in grapevines may allow for improved horticultural practices

and fruit quality outcomes in wine and table grape production. The aims of this thesis were to examine

the effect of variety on calcium uptake and to study the response of vines to different calcium supply

levels. Differences in calcium accumulation and response to high calcium soils were examined in

several varieties. As grapevines do not usually show obvious signs of calcium toxicity or deficiency in

vineyard soils, it was also necessary to perturb normal calcium nutrition to elicit physiological responses

that elucidate the impact of calcium on vine growth.

Varietal differences in berry calcium uptake and distribution, cell morphology, cell wall composition, and

biomechanical properties were investigated at two sites purportedly differing in calcium availability.

Chenin Blanc demonstrated relatively high levels of calcium in apoplastic fluid (Figure 2.4). Analysis of

berry skin, mesocarp, and apoplast fluid composition revealed higher concentration of calcium in

samples from the Barossa site than from the Waite site. Differences between varieties were observed in

their response to different growing conditions. Grenache demonstrated a significant difference between

sites in skin calcium concentration, but not in mesocarp calcium concentration, suggesting that this

variety has low site sensitivity with regards to calcium uptake, or a high capacity for translocation of

calcium to the skin. Shiraz was the only variety surveyed that demonstrated any significant difference

between sites in apoplastic pH and apoplastic Ca2+ activity, Waite berries had higher apoplast pH and

lower apoplast Ca2+ activity than Barossa berries (Table 2.2). A negative correlation between apoplast

pH and apoplast Ca2+ availability was also demonstrated across the full dataset. Availability of calcium

in the cell wall determines the level of cross linking that occurs between adjacent chains of de-esterified

pectin (Figure 2.5). Differences between varieties in basic cell wall morphology, as well as patterns of

de-esterified pectin distribution within the cell wall, were observed (Figure 2.1 and 2.2). Punctate pectin

de-esterification in the middle lamella of Grenache and Thompson Seedless may indicate the location

of plasmodesmatal pit fields, where pectin crosslinkages are important for cell-cell adhesion in these

varieties. Chenin Blanc demonstrated high levels of de-esterified pectin in the epidermis and peripheral

106

vasculature of the berry (Figure 4.6 and 4.7), possibly contributing to this varieties low shrivel

susceptibility.

A hydroponics system was designed to subject fruiting cuttings of grapevines to abnormally high and

low levels of calcium nutrition. The impact of these treatments upon vine physiology and development

were examined. Deficit calcium supply in Chenin Blanc increased net assimilation rate and stomatal

conductance (Figure 3.7). It was also observed to accelerate berry ripening in Shiraz through changes

in cell wall modification and berry water relations. This resulted in an early onset of berry weight loss

and berry shrivel (Figure 3.1). Under an approximation of a commercial harvest schedule (i.e. at ~114

DAA) this level of calcium nutrition would result in very low yields, high sugars, and poor fruit quality

due to extensive shrivel (Figure 3.5). Low calcium supply affects the physical properties of berries;

when compared to berries at the same stage of development grown under basal calcium conditions low

calcium berries showed higher deformability and elasticity (Figure 3.6). Whilst the low calcium treatment

and the hydroponics system used in these experiments may not be directly applicable to field

conditions, these results do serve to further our understanding of the impact of reduced or deficit

calcium nutrition on vine function and berry development.

Elevated calcium supply in Chenin Blanc reduced vine transpiration rate, net assimilation rate, stomatal

conductance, and sub-stomatal CO2 (Figure 3.7). Berry ripening in high calcium grown Shiraz vines

was delayed and asynchronous; high calcium bunches showed, on average, later veraison date

occurrence, as well as a wider range of occurrence dates (Figure 3.2). High calcium bunches had lower

bunch weights (Figure 3.4), smaller berries, and lower TSS (Figure 3.1). Interestingly, when harvested

at 114 DAA, these bunches demonstrated a wide variety of berry class occurrence; both pre-veraison

green berries and overripe shrivelling berries were prevalent in these bunches (Figure 3.5). Petiole

nutrient analysis revealed that the high calcium treatment did indeed increase calcium uptake. Chloride

content was also increased and this is a potentially confounding issue; some of the effects were similar

to that observed when grapevines are grown under saline conditions (Hawker and Walker, 1978).

Both the field and controlled environment approaches used in this study have their drawbacks and

limitations. Whilst the field-based study was an important starting point for understanding the varietal

diversity in grapevine calcium relations, it is also difficult to make strong conclusions where many

environmental variables may have an influence, particularly when comparing different sites. The

reductionist approach using hydroponics, fruiting cuttings and glasshouse growth allowed for

elimination of some of these variables and elucidation of the physiological responses of the grapevine

to calcium nutrition. It is important to note that the responses of established, field grown vines to similar

107

treatments may differ. However, these results do provide an understanding of underlying physiology

around which viticultural practices modifying calcium nutrition for optimal vine health and grape quality

may be developed.

5.2 Berry calcium physiology

The importance of calcium in normal plant nutrition and development has been established. Substantial

evidence is presented here that calcium plays crucial and complex roles in fruit physiology through its

influence on fruit water relations, cell wall modification and phytohormonal signalling.

Low calcium growth resulted in changes in levels of pectin modification and distribution of de-esterified

pectin in Shiraz and Chenin Blanc. Higher levels of de-esterified pectin were observed in the cuticles of

low calcium grown fruit (Chapter 4.3.1). These fruit also exhibited a higher berry transpiration rate

(Chapter 4.3.5). Due to the lower availability of calcium in the apoplast of low calcium grown fruit, lower

levels of pectin-calcium crosslinkage are believed to arise, resulting in greater access to the cell wall

structure for pectin solubilising polygalacturonases and pectate lyases. Additionally, reduced formation

of calcium cross-linked pectin gels in xylem vessel pit membranes increases berry transpiration. These

effects combine to produce berries with a high deformability and susceptibility to berry shrivel (Chapter

4.3.4).

High calcium grown Shiraz did not demonstrate occurrence of any de-esterified pectin in the cuticle,

indicating that the high levels of available calcium in the cell wall effectively cross-linked de-esterified

pectins and limited the activity of polygalacturonases and pectate lyases. The effect of calcium on cell

wall properties was also supported by results of biomechanical tests; high calcium berries

demonstrated later increases in berry elasticity and skin strength relative to basal and low calcium

grown berries. High calcium grown berries also had reduced berry transpiration, supporting the role of

calcium in xylem vessel gel formation for moderation of water relations (Chapter 4.3.5).

5.3 Implications of calcium nutrition for fruit quality

Understanding the role of calcium in fruit development helps to address ripening disorders (e.g. berry

shrivel in Shiraz grapes), tissue localised calcium deficiencies (e.g. blossom end rot in tomatoes, bitter

108

pit in apples), and pathogen susceptibilities (e.g. Botrytis cinerea). Improved understanding of the

calcium nutritional requirements of plants may also aid in optimising fruit quality outcomes.

Emerging evidence suggests that the cell wall composition of intervessel pit membranes in the xylem

play an important role in determining fruit water relations and resistance to plant vascular diseases. Ion

binding changes the properties of pectic gels at the pit membranes, forming an important regulatory

point for xylem hydraulic conductance. Evidence for this hypothesis has been reviewed in van Doorn et

al. (2011) and Nardini et al. (2011). Potassium and sodium ions affect the hydration status of the gel

through their interactions with negatively charged pectin carboxyl residues, resulting in gel shrinkage

(and transition to an amorphous solid), increasing the pore size of pit membranes and hydraulic

conductance of the xylem (Nardini et al., 2011). This effect is often decreased or suppressed by the

presence of the divalent cation calcium, presumably due to crosslinking of pectins resulting in gel

formation (Nardini et al., 2007; Cochard et al., 2010). Grapevine genotype differences in pit membrane

polysaccharide composition affect susceptibility to Pierce’s disease. The pathogen responsible for

Pierce’s disease (Xylella fastidiosa) spreads through the vascular system of the grapevine, secreting

polygalacturonases and endoglucanases to degrade the cell walls of pit membranes. Resistant

varieties showed high levels of esterified homogalacturonan (Sun et al., 2011). Esterified

homogalacturonan is more resistant to PG degradation, forming a barrier at the pit membrane to the

spread of the pathogen. These effects demonstrate the importance of cell wall composition and calcium

distribution for plant water relations and pathogen resistance.

Calcium accumulation in tomato fruit has been shown to be dependent on rates of xylem sap flow,

influenced by transpiration and growth rates (Ho et al., 1993; de Freitas et al., 2014). The strength of

other calcium sinks in the plant can affect calcium accumulation in the tomato, and lead to calcium

related physiological disorders such as blossom end rot (Ho and White, 2005). ABA treatment of whole

plants reduces transpiration, root water uptake and calcium uptake into the leaves, whilst increasing

stem water potential (de Freitas et al., 2014). This may be due to continuation of photosynthesis and

phloem mediated solute accumulation. This continued solute accumulation could decrease the fruit

apoplastic solute potential and fruit total water potential, enabling increased xylem flows into the fruit

from the stem and accumulation of xylem-mobile solutes such as calcium (de Freitas et al., 2014).

An ABA deficient tomato mutant (sitiens; which exhibits Botrytis cinerea resistance) exhibit a lower

degree of epidermal cell wall pectin de-esterification, reduced cuticle thickness, and increased cuticle

permeability, when compared to wild type (Asselbergh et al., 2007; Curvers et al., 2010). The

consequent reduction in Botrytis cinerea susceptibility of sitiens may be as a result of; (a) plant

109

detection of defective cuticle, prompting constitutive expression of chitinases and beta-glucosidases

into the cell wall, enabling rapid release of fungal elicitors upon infection, and/or (b) a lower level of de-

esterification in sitiens cell walls providing a source of more bio-active oligogalacturonide (OGA)

elicitors upon infection; thereby producing a more rapid and effective response to pathogen attack

(Curvers et al., 2010). The level of esterification in OGAs is one of several factors that determine their

activity and specificity in triggering plant responses; it has been shown that the level of de-esterification

in strawberry OGAs contributes to their capacity to elicit defence responses (Osorio et al., 2008).

Although the exact mechanism of Botrytis cinerea resistance in sitiens is unknown, it is clear that the

interaction between epidermal cell wall derived pectins and the cuticle, either as defence signalling

oligogalacturonides or as structural components, is important. In addition to the processes controlling

deposition of cutin into the cuticle, processes affecting polysaccharide solubilisation and movement into

the cuticle, such as pectin de-esterification and calcium crosslinking, will modify fruit mechanical

properties and pathogen susceptibility.

GA treatment of tomatoes leads to increased occurrence of blossom end rot while treatment with GA

biosynthesis inhibitor prohexadione-calcium eliminated blossom end rot (de Freitas et al., 2012). GA

treated tomatoes showed increased expression of CAX and Ca-ATPase genes and reduced apoplastic

Ca2+, whereas GA inhibitor treated fruit showed higher pericarp calcium levels and an increased

number of functional fruit xylem vessels (de Freitas et al., 2012). GA-induced expression of CAX and

Ca-ATPase genes results in depletion of the apoplastic calcium pool, possibly below the critical

concentration required for membrane stabilisation and moderation of cell expansion through pectin-

Ca2+ crosslinks in the cell wall. In addition to the localised drop in calcium concentration, rapidly

expanding tissue (such as the blossom end of tomatoes) may further impede normal fruit calcium

distribution due to cellular intrusions causing obstruction or breakage of xylem vessels (Drazeta et al.,

2004; de Freitas et al., 2012). Apogee-treated fruit showed increased numbers of functional xylem

vessels; the impact of GA on xylem differentiation and development modifies normal pathways for

calcium distribution (Saure, 2005). Additionally, GA triggers increased cuticle deposition (Knoche and

Peschel, 2007), potentially modifying fruit water relations and calcium uptake. All of these factors

provide possible linkages between GA responses and changes in calcium localisation leading to

blossom end rot.

The apple industry has significant interest in increasing fruit sugar content and post-harvest shelf life;

applications of exogenous calcium have been proposed as one way to achieve this. However, the

relationship between calcium and sugar accumulation is complex and many factors appear to affect the

110

effectiveness of this strategy including soil calcium availability, timing of spray/application, apple variety,

tree calcium status and in planta interactions with other ions (e.g. boron) (Lu et al., 2013).

Levels of apple tree shading also have a complex effect, with conflicting reports as to whether fruit

calcium uptake is increased or decreased with shading (Chen et al., 1997; de Freitas et al., 2013).

Calcium accumulation is also reduced by progressive breakdown of xylem connectivity as the result of

growth related damage, potentially increasing occurrence of bitter pit disorder (Drazeta et al., 2004). In

contrast, dye studies in post-veraison grapes indicate that xylem not only remains relatively intact, but

also continues to develop and mature (Chatelet et al., 2008).

Melons suffer from a water-soaking condition that has been linked to calcium deficiency. It has been

hypothesised that depletion of apoplastic calcium supply can lead to insufficient pectin crosslinks in the

middle lamella of the mesocarp, resulting in water-soaked tissue (Madrid et al., 2004; Nishizawa et al.,

2004). It is also suggested that in climacteric fruits, depletion of unbound calcium may, through

secondary signalling mechanisms, lead to increased ethylene production, promoting ripening

mechanisms, however water-soaking appears to develop in calcium deficient conditions regardless of

ethylene status (Nishizawa et al., 2004).

5.4 Future perspectives and Conclusion

The delayed and asynchronous bunch development observed in high calcium grown vines suggests

that the calcium treatment disrupted several aspects of vine physiology. By studying the levels of

hormone accumulation and expression patterns of ripening pathway components (especially calcium

binding proteins) around ripening, the trigger for ripening initiation in these bunches could be

elucidated. It is possible that calcium transduction of hormonal developmental signals for ripening

initiation is perturbed. ABA levels in high calcium bunches may differ between berries. Alternatively, the

high calcium levels and pectin gel formation in xylem vessels may selectively limit vascular flows within

the bunch, resulting in reduced and variable rates of water uptake, reduced berry size, and delayed

ripening.

This study has uncovered high and low calcium levels at which physiological deficiency and toxicity

traits can be observed. Re-formulation of hydroponics solutions to achieve these effects whilst

moderating unwanted effects of other ion imbalances may allow for a clearer observation of the

individual effect of calcium. This is particularly relevant in the high calcium solution, where the necessity

111

of high chloride levels to achieve high calcium activity may have had a confounding effect upon the

results.

Whilst this study characterised patterns of pectin epitope distribution in grape tissues and within the cell

wall, further characterisation of the cell wall would improve understanding of the varietal differences and

developmental processes in this space. This would include confirmation of patterns of pectin de-

esterification through cell wall analysis. Additionally, characterisation of expression and activity profiles

of pectin methyl esterases and polygalacturonases under different calcium regimes would help to

understand the drivers of berry softening.

Calcium localisation at a sub-cellular level using SEM-X-ray microanalysis may allow for confirmation of

paths of calcium movement through the berry. Particularly, microscopy studies of calcium distribution

on the cell wall and xylem vessels would further comprehension of the interactions between calcium

nutrition, cell wall processes, and berry water relations.

Given the results of the hydroponic grapevine trials, investigation of the effects of calcium fertilisation

under field conditions could be rewarding. Whilst this study has demonstrated that changes in levels of

calcium nutrition affect calcium uptake and various aspects of berry physiology, further studies would

be required to understand how manipulation of calcium nutrition in the field may affect fruit quality

outcomes. These trials would potentially involve calcium fertilisation of a variety of soil types to

understand the dynamics of calcium uptake. After establishing that calcium fertilisation increases berry

calcium uptake, further focus on berry development, berry softening, occurrence of berry shrivel, and

pathogen susceptibility could be investigated.

112

Appendices

Appendix 1

Example stress-strain curve of berry hardness test from Instron materials testing machine. Slope of

regression line over linear portion prior to break used to calculate berry Young’s Modulus according to

the formula presented in Chapter 2.2.5.

y = 3.8495x - 19.504 R² = 0.9975

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

Stre

ss (

N)

Strain (mm)

Instron raw data

Linear portion

113

Appendix 2

Standard background [Ca2+] mM [Ca2+] M pCa (-LOG([Ca2+]) mV

H2O 100 0.1 1 47

10 0.01 2 14.2

5 0.005 2.30103 9.6

1 0.001 3 -6.5

0.25 0.00025 3.60206 -20

0 0 -59.6

Artificial grape juice (AGJ)

5 0.005 2.30103 1

4 0.004 2.39794 -3.2

3 0.003 2.522879 -8.1

2 0.002 2.69897 -15

1 0.001 3 -22.2

0.5 0.0005 3.30103 -28.4

0.25 0.00025 3.60206 -37.8

0 0 -62

Calcium selective electrode calibration with CaCl2 in H2O and CaCl2 in artificial grape juice (AGJ) (A)

and calibration curves (B) as outlined in Chapter 2.2.9.

y = -0.0345x + 2.2719 R² = 0.9875

0

0.5

1

1.5

2

2.5

3

3.5

4

-60 -40 -20 0 20 40 60

pC

a

mV

CaCl2

AGJ

Linear (AGJ)

B

A

114

Appendix 3

115

Baby, T., Hocking, B., Tyerman, S.D., Gilliham, M. and Collins, C. (2014) Modified Method for Producing Grapevine Plants in Controlled Environments. American Journal of Enology and Viticulture, v. 65 (2), pp. 261-267, June 2014

NOTE: This publication is included in the print copy of the thesis

held in the University of Adelaide Library.

It is also available online to authorised users at:

http://dx.doi.org/10.5344/ajev.2014.13121

123

Appendix 4

Grapevine Basal Nutrient Solution (BNS)

Macronutrients MW

g added to make 75L sol. Final conc (mM)

NH4NO3 80

6 1

KNO3 101.1

10.6155 1.4

KCl 74.55

8.386875 1.5

Ca(NO3)2•4H20 236.1

38.94 2.2

MgSO4•7H20 246.5

18.4875 1

KH2PO4 136.1

4.083 0.4

NaCl 58.44

8.766 2

NaOH TO pH 5.6 0 0.05

Micronutrients MW g added to make 1L stock mL stock added to make 75L sol. Final conc (uM)

NaFe(III)EDTA 367.1 96.6 9.375 35

H3BO3 61.8 9.27 9.375 20

MnCl2•4H20 197.9 7.425 9.375 5

ZnSO4•7H20 287.5 25.875 9.375 12

CuSO4•5H20 249.7 2.25 9.375 1.2

Na2MoO3 242 0.18375 9.375 0.1

Final Conc of ion mM Activity (mM) Final Conc of ion mM

K 3.3 2.8815 Fe-EDTA-2 0.035

Ca 2.2 1.2108 Mn 0.005

Mg 1 0.582 Zn 0.012

NH4 1 0.87611 Cu 0.0012

Cl 3.51 3.0692 Mo 0.0001

NO3 6.8 5.9545

SO4 1.013 0.48568

H2PO4- 0.4 0.3177

Na 2.085 1.8264

Grapevine Low Calcium Nutrient Solution (LCS)

Macronutrients MW

g added to make 75L sol. Final conc (mM)

NH4NO3 80

6 1

KNO3 101.1

22.7475 3

CaCl2 147

3.3078285 0.3

MgSO4•7H20 246.5

14.79 0.8

KH2PO4 136.1

3.572625 0.35

Mg(NO3)2.6H20 256.4

5.769225 0.3

HNO3 1M solution nitric acid 150mL 2

NaOH TO pH 5.6 0 1.91

Micronutrients MW g added to make 1L stock mL stock added to make 75L sol. Final conc (uM)

NaFe(III)EDTA 367.1 96.6 9.375 35

H3BO3 61.8 9.27 9.375 20

MnCl2•4H20 197.9 7.425 9.375 5

ZnSO4•7H20 287.5 25.875 9.375 12

CuSO4•5H20 249.7 2.25 9.375 1.2

Na2MoO3 242 0.18375 9.375 0.1

Final Conc of ion mM Activity (mM) Final Conc of ion mM

K 3.35 2.985 Fe 0.035

Ca 0.3 0.178 Mn 0.005

Mg 1.1 0.674 Zn 0.012

NH4 1 0.893 Cu 0.0012

Cl 0.61 0.545 Mo 0.0001

NO3 6.6 5.908

SO4 0.813 0.465

H2PO4- 0.35 0.292

Na 1.945 1.737

124

High Calcium Nutrient Solution (HCS)

Macronutrients MW

g added to make 75L sol. Final conc (mM)

CaCl2 147

88.20876 8

Ca(NO3)2•4H20 236.1

70.83 4

MgSO4•7H20 246.5

14.79 0.8

KH2PO4 136.1

37.76775 3.7

NH4Cl 53.49

4.01175 1

CaSO4.2H20 172.2

12.91275 1

MgCl2.6H2O 203.3

12.198 0.8

NaCl 58.44

8.766 2

NaOH pH5.6

0 0.435

Micronutrients MW g added to make 1L stock mL stock added to make 75L sol. Final conc (uM)

NaFe(III)EDTA 367.1 96.6 9.375 35

H3BO3 61.8 9.27 9.375 20

MnCl2•4H20 197.9 7.425 9.375 5

ZnSO4•7H20 287.5 25.875 9.375 12

CuSO4•5H20 249.7 2.25 9.375 1.2

Na2MoO3 242 0.18375 9.375 0.1

Final Conc of ion mM Activity (mM) Final Conc of ion mM

K 3.7 3.0592 Fe 0.035

Ca 13 3.1407 Mn 0.005

Mg 1.6 0.72438 Zn 0.012

NH4 1 0.8337 Cu 0.0012

Cl 20.61 17.08 Mo 0.0001

NO3 8 6.6354

SO4 1.813 0.62319

H2PO4- 3.7 0.079511

Na 2.521 2.0907

125

Appendix 5

//This macro is designed for quantifying staining levels of Propidium Iodide (PI) and Fluorescein di-acetate (FDA) in grapes. //Cam Edit - Store the name of the original image in nameStore variable and the path into dirStore nameStore=getTitle; dirStore=getInfo("image.directory"); //Cam Edit - Store file separator in fs variable to make code cross platform (PC/Mac/Linux) complatible for file saving operations fs=File.separator; //waitForUser("Select BERRY image"); run("Properties...", "channels=1 slices=1 frames=1 unit=mm pixel_width=0.0131579 pixel_height=0.0131579 voxel_depth=0.0131579 frame=[0 sec] origin=0,0"); run("Set Measurements...", "area mean standard min area_fraction limit display redirect=None decimal=3"); rename("Berry"); setTool("polygon"); waitForUser("Select seed 1 for exclusion"); run("Fit Spline"); run("Create Mask"); selectWindow("Mask"); rename("Seed 1"); selectWindow("Berry"); run("Select None"); waitForUser("Select seed 2 for exclusion"); run("Fit Spline"); run("Create Mask"); selectWindow("Mask"); rename("Seed 2"); imageCalculator("Add create", "Seed 1", "Seed 2"); selectWindow("Result of Seed 1"); rename("Seeds"); run("8-bit"); //cam edit - close seed 1 and seed 2 images selectWindow("Seed 1"); close(); selectWindow("Seed 2"); close(); selectWindow("Berry"); run("Select None"); run("Duplicate...", "title=[Split]"); selectWindow("Berry"); run("8-bit"); run("Gaussian Blur...", "sigma=8"); setAutoThreshold("Default dark"); run("Threshold..."); setThreshold(6, 255); selectWindow("Berry"); selectWindow("Threshold"); waitForUser("Select BERRY threshold. Range 1-12."); //setOption("BlackBackground", true); run("Convert to Mask");

126

run("Fill Holes"); run("Analyze Particles...", "size=50-Infinity circularity=0.00-1.00 show=Masks"); rename("BerryArea"); //cam edit - invert LUT to stop binary being made backwards run("Invert LUT"); setThreshold(1,255); run("Measure"); resetThreshold(); run("Duplicate...", "title=[Erode]"); run("Options...", "iterations=30 count=1 black edm=Overwrite do=Erode"); imageCalculator("Subtract create", "BerryArea","Erode"); selectWindow("Result of BerryArea"); rename("SkinArea"); //setOption("BlackBackground", true); setThreshold(1,255); run("Measure"); resetThreshold(); selectWindow("BerryArea"); imageCalculator("Subtract create", "BerryArea", "Seeds"); selectWindow("Result of BerryArea"); rename("BerryNoSeeds"); setThreshold(1,255); run("Measure"); resetThreshold(); selectWindow("Split"); run("Split Channels"); selectWindow("Split (green)"); run("Profile Plot Options...", "width=450 height=200 minimum=0 maximum=0 auto-close vertical list"); makeRectangle(0, 0, 50, 400); waitForUser("Select berry transect"); run("Plot Profile"); selectWindow("Plot Values"); //Cam Edit - save plot profiles to excel files based on orignal name with suffix of FDA or PI added saveAs("Text", dirStore+fs+nameStore+" - FDA Profile.xls"); selectWindow("Plot Values"); run("Close"); selectWindow("Split (red)"); run("Restore Selection"); run("Plot Profile"); selectWindow("Plot Values"); saveAs("Text", dirStore+fs+nameStore+" - PI Profile.xls"); selectWindow("Plot Values"); run("Close"); selectWindow("Split (blue)"); close(); selectWindow("Split (red)"); run("Select None"); run("Duplicate...", "title=[PI]"); run("8-bit");

127

setAutoThreshold("Default dark"); run("Threshold..."); selectWindow("PI"); setThreshold(20,255); selectWindow("Threshold"); waitForUser("Select PI threshold. Range 6-24."); run("Convert to Mask"); imageCalculator("AND create", "PI","BerryNoSeeds"); selectWindow("Result of PI"); rename("BerryPI"); //setOption("BlackBackground", true); //run("Convert to Mask"); setThreshold(1,255); run("Measure"); resetThreshold(); imageCalculator("AND create", "SkinArea","PI"); selectWindow("Result of SkinArea"); rename("SkinPI"); //setOption("BlackBackground", true); setThreshold(1,255); run("Measure"); resetThreshold(); selectWindow("Split (green)"); run("Select None"); run("Duplicate...", "title=[FDA]"); run("8-bit"); setAutoThreshold("Default dark"); run("Threshold..."); selectWindow("FDA"); setThreshold(8,255); selectWindow("Threshold"); waitForUser("Select FDA threshold. Range 6-12."); run("Convert to Mask"); imageCalculator("AND create", "FDA", "BerryNoSeeds"); selectWindow("Result of FDA"); rename("BerryFDA"); //setOption("BlackBackground", true); //run("Convert to Mask"); setThreshold(1,255); run("Measure"); resetThreshold(); imageCalculator("AND create", "SkinArea","FDA"); selectWindow("Result of SkinArea"); rename("SkinFDA"); //setOption("BlackBackground", true); //run("Convert to Mask"); setThreshold(1,255); run("Measure"); resetThreshold(); imageCalculator("AND create", "BerryFDA","BerryPI"); selectWindow("Result of BerryFDA"); rename("BerryCo-stain"); //setOption("BlackBackground", true); //run("Convert to Mask"); setThreshold(1,255); run("Measure");

128

resetThreshold(); imageCalculator("AND create", "SkinFDA","SkinPI"); selectWindow("Result of SkinFDA"); rename("SkinCo-stain"); //setOption("BlackBackground", true); //run("Convert to Mask"); setThreshold(1,255); run("Measure"); resetThreshold(); //Cam Edit - save are information to excel files based on original name with suffix of FDA or PI added run("Close All"); selectWindow("Results"); saveAs("Text", dirStore+fs+nameStore+" - Area Info.xls"); selectWindow("Results"); run("Close");

129

Appendix 6

Appendix 6.1 TEM images of Shiraz and Chenin Blanc berry epidermal and hypodermal tissue grown

under low, basal, and high calcium conditions at 100 DAA immunogold labelled with LM19. LM19 binds

de-esterified pectin with immunogold labelling present as small black dots. Due to print resolution

limitations, viewing of staining patterns on a monitor may be preferable. Shiraz (A-F) and Chenin Blanc

(G-L) demonstrate differences in labelling pattern between epidermal (A-C, G-I) and hypodermal (D-F,

J-L) tissue when grown in LCS (A, D, G, J), BNS (B, E, H, K), and HCS (C, F, I, L). cu indicates cuticle.

p indicates primary cell wall. m indicates middle lamella. Multiple berries (at least 3) were examined for

each treatment. Representative images are shown. Scale bars represent 2 µm.

130

Appendix 6.2 TEM images of Shiraz and Chenin Blanc berry epidermal and hypodermal tissue grown

under low, basal, and high calcium conditions at 100 DAA immunogold labelled with LM20. LM20 binds

esterified pectin with immunogold labelling present as small black dots. Due to print resolution

limitations, viewing of staining patterns on a monitor may be preferable. Shiraz (A-F) and Chenin Blanc

(G-L) demonstrate differences in labelling pattern between epidermal (A-C, G-I) and hypodermal (D-F,

J-L) tissue when grown in LCS (A, D, G, J), BNS (B, E, H, K), and HCS (C, F, I, L). cu indicates cuticle.

p indicates primary cell wall. m indicates middle lamella. Multiple berries (at least 3) were examined for

each treatment. Representative images are shown. Scale bars represent 2 µm.

131

Appendix 6.3 TEM images of Shiraz berry epidermal tissue grown under low, basal, and high calcium

conditions at post-veraison (75 DAA), harvest (100 DAA), and late harvest (118 DAA) immunogold

labelled with LM19. LM19 binds de-esterified pectin with immunogold labelling present as small black

dots. Due to print resolution limitations, viewing of staining patterns on a monitor may be preferable.

100 DAA images are repeated from Figure 4.1 for ease of comparison. Berry tissues from vines grown

in LCS (A, C, F), BNS (B, D, G), and HCS (E, H). cu indicates cuticle. p indicates primary cell wall.

Multiple berries (at least 3) were examined for each treatment and time point. Representative images

are shown. Scale bars represent 2 µm.

132

Appendix 6.4 TEM images of Shiraz berry epidermal tissue grown under low, basal, and high calcium

conditions at post-veraison (75 DAA), harvest (100 DAA), and late harvest (118 DAA) immunogold

labelled with LM20. LM20 binds esterified pectin with immunogold labelling present as small black dots.

Due to print resolution limitations, viewing of staining patterns on a monitor may be preferable. 100 DAA

images are repeated from Figure 4.2 for ease of comparison. Berry tissues from vines grown in LCS (A,

C, F), BNS (B, D, G), and HCS (E, H). cu indicates cuticle. p indicates primary cell wall. Multiple berries

(at least 3) were examined for each treatment and time point. Representative images are shown. Scale

bars represent 2 µm.

133

Appendix 7

Relationships between electrical and physical properties of berries. Intracellular resistance and

brix (A). Cellular membrane capacitance and berry length (electrical path distance between

probes) (B). Linear regression lines and R2 values shown.

y = 27.606x - 152.84 R² = 0.6011

0.00E+00

5.00E+01

1.00E+02

1.50E+02

2.00E+02

2.50E+02

3.00E+02

3.50E+02

4.00E+02

4.50E+02

0 5 10 15 20

Intr

ace

llula

r re

sist

ance

Ri (Ω

)

Brix

A

Series1

Linear (Series1)

y = 1E-12x + 3E-09 R² = 4E-06

0.00E+00

5.00E-10

1.00E-09

1.50E-09

2.00E-09

2.50E-09

3.00E-09

3.50E-09

4.00E-09

4.50E-09

5.00E-09

0 2 4 6 8 10 12 14

Ce

llula

r m

em

bra

ne

cap

acit

ance

Cm

(F)

Path distance (mm)

B

Series1

Linear (Series1)

134

Appendix 8

Appendix 8. Botrytis cinerea assay of Shiraz berries grown under different calcium regimes. Assay was

carried out over 96 hours, with inspection of berries at 48, 72, and 96 hours after Botrytis cinerea

inoculation. Assay was performed on intact (A) and wounded (B) berries. Berries were scored by visual

assessment of percentage berry surface area discoloured/infected; 0 = 0%, 1 = 1-25%, 2 = 26-50%, 3 =

51-75%, 4 = 76-100%. Mean + SD. 3 berries from each of 3 (BNS), 1 (LCS), or 4 (HCS) vines.

H o u rs a fte r in n o c u la t io n

Infe

cti

on

sta

ge

4 8 7 2 9 6

0

1

2

3

4

5

A

H o u rs a fte r in n o c u la t io n

Infe

cti

on

sta

ge

4 8 7 2 9 6

0

1

2

3

4

5

S h ira z B N S

S h ira z L C S

S h ira z H C S

B

135

Reference list

Allen, G.J., Chu, S.P., Harrington, C.L., Schumacher, K., Hoffman, T., Tang, Y.Y., Grill, E., and

Schroeder, J.I. (2001). A defined range of guard cell calcium oscillation parameters encodes

stomatal movements. Nature 411, 1053-1057.

Allen, G.J., Kuchitsu, K., Chu, S.P., Murata, Y., and Schroeder, J.I. (1999). Arabidopsis abi1-1 and

abi2-1 phosphatase mutations reduce abscisic acid-induced cytoplasmic calcium rises in guard

cells. Plant Cell 11, 1785-1798.

Almeida, D.P.F., and Huber, D.J. (1999). Apoplastic pH and inorganic ion levels in tomato fruit: A

potential means for regulation of cell wall metabolism during ripening. Physiologia Plantarum

105, 506-512.

Ando, Y., Mizutani, K., and Wakatsuki, N. (2014). Electrical impedance analysis of potato tissues during

drying. Journal of Food Engineering 121, 24-31.

Asselbergh, B., Curvers, K., Franca, S.C., Audenaert, K., Vuylsteke, M., Van Breusegem, F., and Hofte,

M. (2007). Resistance to Botrytis cinerea in sitiens, an abscisic acid-deficient tomato mutant,

involves timely production of hydrogen peroxide and cell wall modifications in the epidermis.

Plant Physiology 144, 1863-1877.

Baby, T., Hocking, B., Tyerman, S.D., Gilliham, M., and Collins, C. (2014). Modified method for

producing grapevine plants in controlled environments. American Journal of Enology and

Viticulture 65, 261-267.

Balic, I., Ejsmentewicz, T., Sanhueza, D., Silva, C., Peredo, T., Olmedo, P., Barros, M., Verdonk, J.C.,

Paredes, R., Meneses, C., Prieto, H., Orellana, A., Defilippi, B.G., and Campos-Vargas, R.

(2014). Biochemical and physiological study of the firmness of table grape berries. Postharvest

Biology and Technology 93, 15-23.

Barnavon, L., Doco, T., Terrier, N., Ageorges, A., Romieu, C., and Pellerin, P. (2001). Involvement of

pectin methyl-esterase during the ripening of grape berries: partial cDNA isolation, transcript

expression and changes in the degree of methyl-esterification of cell wall pectins.

Phytochemistry 58, 693-701.

136

Baxter, I., Hosmani, P.S., Rus, A., Lahner, B., Borevitz, J.O., Muthukumar, B., Mickelbart, M.V.,

Schreiber, L., Franke, R.B., and Salt, D.E. (2009). Root suberin forms an extracellular barrier

that affects water relations and mineral nutrition in Arabidopsis. Plos Genetics 5, 12.

Becker, T., and Knoche, M. (2011). Water movement through the surfaces of the grape berry and its

stem. American Journal of Enology and Viticulture 62, 340-350.

Becker, T., and Knoche, M. (2012). Deposition, strain, and microcracking of the cuticle in developing

'Riesling' grape berries. Vitis 51, 1-6.

Blumer, J.M., Clay, R.P., Bergmann, C.W., Albersheim, P., and Darvill, A. (2000). Characterization of

changes in pectin methylesterase expression and pectin esterification during tomato fruit

ripening. Canadian Journal of Botany-Revue Canadienne De Botanique 78, 607-618.

Boettcher, C., Harvey, K., Forde, C.G., Boss, P.K., and Davies, C. (2011). Auxin treatment of pre-

veraison grape (Vitis vinifera L.) berries both delays ripening and increases the synchronicity of

sugar accumulation. Australian Journal of Grape and Wine Research 17, 1-8.

Bonada, M., Sadras, V., Moran, M., and Fuentes, S. (2013a). Elevated temperature and water stress

accelerate mesocarp cell death and shrivelling, and decouple sensory traits in Shiraz berries.

Irrigation Science 31, 1317-1331.

Bonada, M., Sadras, V.O., and Fuentes, S. (2013b). Effect of elevated temperature on the onset and

rate of mesocarp cell death in berries of Shiraz and Chardonnay and its relationship with berry

shrivel. Australian Journal of Grape and Wine Research 19, 87-94.

Bosch, M., and Hepler, P.K. (2005). Pectin methylesterases and pectin dynamics in pollen tubes. Plant

Cell 17, 3219-3226.

Braccini, I., Rodriguez-Carvajal, M.A., and Perez, S. (2005). Chain-chain interactions for methyl

polygalacturonate: Models for high methyl-esterified pectin junction zones. Biomacromolecules

6, 1322-1328.

Brown, M.M., and Ho, L.C. (1993). Factors affecting calcium-transport and basipetal IAA movement in

tomato fruit in relation to blossom-end rot. Journal of Experimental Botany 44, 1111-1117.

Brummell, D.A. (2006). Cell wall disassembly in ripening fruit. Functional Plant Biology 33, 103-119.

137

Brummell, D.A., and Harpster, M.H. (2001). Cell wall metabolism in fruit softening and quality and its

manipulation in transgenic plants. Plant Molecular Biology 47, 311-340.

Brummell, D.A., Harpster, M.H., Civello, P.M., Palys, J.M., Bennett, A.B., and Dunsmuir, P. (1999).

Modification of expansin protein abundance in tomato fruit alters softening and cell wall

polymer metabolism during ripening. Plant Cell 11, 2203-2216.

Burton, R.A., Gidley, M.J., and Fincher, G.B. (2010). Heterogeneity in the chemistry, structure and

function of plant cell walls. Nature Chemical Biology 6, 724-732.

Bush, M.S., and Mccann, M.C. (1999). Pectic epitopes are differentially distributed in the cell walls of

potato (Solanum tuberosum) tubers. Physiologia Plantarum 107, 201-213.

Cabanne, C., and Doneche, B. (2001). Changes in polygalacturonase activity and calcium content

during ripening of grape berries. American Journal of Enology and Viticulture 52, 331-335.

Cabanne, C., and Doneche, B. (2002). Purification and characterization of two isozymes of

polygalacturonase from Botrytis cinerea. Effect of calcium ions on polygalacturonase activity.

Microbiological Research 157, 183-189.

Cabanne, C., and Doneche, B. (2003). Calcium accumulation and redistribution during the development

of grape berry. Vitis 42, 19-21.

Cadot, Y., Chevalier, M., and Barbeau, G. (2011). Evolution of the localisation and composition of

phenolics in grape skin between veraison and maturity in relation to water availability and some

climatic conditions. Journal of the Science of Food and Agriculture 91, 1963-1976.

Cantu, D., Vicente, A.R., Greve, L.C., Dewey, F.M., Bennett, A.B., Labavitch, J.M., and Powell, A.L.T.

(2008). The intersection between cell wall disassembly, ripening, and fruit susceptibility to

Botrytis cinerea. Proceedings of the National Academy of Sciences of the United States of

America 105, 859-864.

Caravia Bayer, L., Collins, C., and Tyerman, S. (2015). Electrical impedance of berries correlates with

decreasing cell vitality during ripening in Shiraz. Australian Journal of Grape and Wine

Research (submitted).

138

Castellarin, S.D., Gambetta, G.A., Wada, H., Shackel, K.A., and Matthews, M.A. (2011). Fruit ripening

in Vitis vinifera: spatiotemporal relationships among turgor, sugar accumulation, and

anthocyanin biosynthesis. Journal of Experimental Botany 62, 4345-4354.

Chardonnet, C., and Doneche, B. (1995). Relation between calcium content and resistance to

enzymatic digestion of the skin during grape ripening. Vitis 34, 95-98.

Chardonnet, C., Lhyvernay, A., and Doneche, B. (1997). Effect of calcium treatment prior to Botrytis

cinerea infection on the changes in pectic composition of grape berry. Physiological and

Molecular Plant Pathology 50, 213-218.

Chardonnet, C.O., Sams, C.E., Trigiano, R.N., and Conway, W.S. (2000). Variability of three isolates of

Botrytis cinerea affects the inhibitory effects of calcium on this fungus. Phytopathology 90, 769-

774.

Chatelet, D.S., Rost, T.L., Shackel, K.A., and Matthews, M.A. (2008). The peripheral xylem of

grapevine (Vitis vinifera). 1. Structural integrity in post-veraison berries. Journal of Experimental

Botany 59, 1987-1996.

Chen, K., Hu, G.Q., and Lenz, F. (1997). Training and shading effects on vegetative and reproductive

growth and fruit quality of apple. Gartenbauwissenschaft 62, 207-213.

Choat, B., Gambetta, G.A., Shackel, K.A., and Matthews, M.A. (2009). Vascular function in grape

berries across development and its relevance to apparent hydraulic isolation. Plant Physiology

151, 1677-1687.

Cholewa, E., and Peterson, C.A. (2004). Evidence for symplastic involvement in the radial movement of

calcium in onion roots. Plant Physiology 134, 1793-1802.

Ciccarese, A., Stellacci, A.M., Gentilesco, G., and Rubino, P. (2013). Effectiveness of pre- and post-

veraison calcium applications to control decay and maintain table grape fruit quality during

storage. Postharvest Biology and Technology 75, 135-141.

Clarkson, D.T. (1984). Calcium-transport between tissues and its distribution in the plant. Plant Cell and

Environment 7, 449-456.

139

Clearwater, M.J., Luo, Z.W., Ong, S.E.C., Blattmann, P., and Thorp, T.G. (2012). Vascular functioning

and the water balance of ripening kiwifruit (Actinidia chinensis) berries. Journal of Experimental

Botany 63, 1835-1847.

Cochard, H., Herbette, S., Hernandez, E., Holtta, T., and Mencuccini, M. (2010). The effects of sap

ionic composition on xylem vulnerability to cavitation. Journal of Experimental Botany 61, 275-

285.

Conn, S., and Gilliham, M. (2010). Comparative physiology of elemental distributions in plants. Annals

of Botany 105, 1081-1102.

Conn, S.J., Gilliham, M., Athman, A., Schreiber, A.W., Baumann, U., Moller, I., Cheng, N.H.,

Stancombe, M.A., Hirschi, K.D., Webb, A.a.R., Burton, R., Kaiser, B.N., Tyerman, S.D., and

Leigh, R.A. (2011). Cell-specific vacuolar calcium storage mediated by CAX1 regulates

apoplastic calcium concentration, gas exchange, and plant productivity in Arabidopsis. Plant

Cell 23, 240-257.

Conn, S.J., Hocking, B., Dayod, M., Xu, B., Athman, A., Henderson, S., Aukett, L., Conn, V., Shearer,

M.K., Fuentes, S., Tyerman, S.D., and Gilliham, M. (2013). Protocol: optimising hydroponic

growth systems for nutritional and physiological analysis of Arabidopsis thaliana and other

plants. Plant Methods 9.

Conradie, W.J. (1981). Seasonal uptake of nutrients by Chenin Blanc in sand culture: II. Phosphorus,

potassium, calcium and magnesium. South African Journal of Enology and Viticulture 2, 7-13.

Cosgrove, D.J. (2005). Growth of the plant cell wall. Nat Rev Mol Cell Biol 6, 850-861.

Curvers, K., Seifi, H., Mouille, G., De Rycke, R., Asselbergh, B., Van Hecke, A., Vanderschaeghe, D.,

Hofte, H., Callewaert, N., Van Breusegem, F., and Hofte, M. (2010). Abscisic acid deficiency

causes changes in cuticle permeability and pectin composition that influence tomato resistance

to Botrytis cinerea. Plant Physiology 154, 847-860.

Dal Santo, S., Vannozzi, A., Tornielli, G.B., Fasoli, M., Venturini, L., Pezzotti, M., and Zenoni, S. (2013).

Genome-wide analysis of the expansin gene superfamily reveals grapevine-specific structural

and functional characteristics. Plos One 8.

140

Dayod, M., Tyerman, S.D., Leigh, R.A., and Gilliham, M. (2010). Calcium storage in plants and the

implications for calcium biofortification. Protoplasma 247, 215-231.

De Freitas, S.T., Do Amarante, C.V.T., Dandekar, A.M., and Mitcham, E.J. (2013). Shading affects

flesh calcium uptake and concentration, bitter pit incidence and other fruit traits in

"Greensleeves" apple. Scientia Horticulturae 161, 266-272.

De Freitas, S.T., Jiang, C.Z., and Mitcham, E.J. (2012). Mechanisms involved in calcium deficiency

development in tomato fruit in response to gibberellins. Journal of Plant Growth Regulation 31,

221-234.

De Freitas, S.T., Mcelrone, A.J., Shackel, K.A., and Mitcham, E.J. (2014). Calcium partitioning and

allocation and blossom-end rot development in tomato plants in response to whole-plant and

fruit-specific abscisic acid treatments. Journal of Experimental Botany 65, 235-247.

De Freitas, S.T., and Mitcham, E.J. (2012). "Factors involved in fruit calcium deficiency disorders," in

Horticultural Reviews. (New York, USA: John Wiley & Sons, Inc.), 107-146.

De Silva, D.L.R., Hetherington, A.M., and Mansfield, T.A. (1996). Where does all the calcium go?

Evidence of an important regulatory role for trichomes in two calcicoles. Plant Cell and

Environment 19, 880-886.

Decreux, A., and Messiaen, J. (2005). Wall-associated kinase WAK1 interacts with cell wall pectins in a

calcium-induced conformation. Plant and Cell Physiology 46, 268-278.

Deluc, L.G., Quilici, D.R., Decendit, A., Grimplet, J., Wheatley, M.D., Schlauch, K.A., Merillon, J.M.,

Cushman, J.C., and Cramer, G.R. (2009). Water deficit alters differentially metabolic pathways

affecting important flavor and quality traits in grape berries of Cabernet Sauvignon and

Chardonnay. Bmc Genomics 10.

Deytieux-Belleau, C., Geny, L., Roudet, J., Mayet, V., Doneche, B., and Fermaud, M. (2009). Grape

berry skin features related to ontogenic resistance to Botrytis cinerea. European Journal of

Plant Pathology 125, 551-563.

Deytieux-Belleau, C., Vallet, A., Doneche, B., and Geny, L. (2008). Pectin methylesterase and

polygalacturonase in the developing grape skin. Plant Physiology and Biochemistry 46, 638-

646.

141

Deytieux, C., Geny, L., and Doneche, B. (2005). "Relation between hormonal balance and

polygalacturonase activity in grape berry," in Proceedings of the 5th International Postharvest

Symposium, Vols 1-3, eds. F. Mencarelli & P. Tonutti. (Leuven 1: International Society

Horticultural Science), 163-170.

Dodd, A.N., Kudla, J., and Sanders, D. (2010). The language of calcium signalling. Annual Review of

Plant Biology 61.

Dominguez, E., Lopez-Casado, G., Cuartero, J., and Heredia, A. (2008). Development of fruit cuticle in

cherry tomato (Solanum lycopersicum). Functional Plant Biology 35, 403-411.

Downton, W.J.S. (1977). Influence of rootstocks on accumulation of chloride, sodium and potassium in

grapevines. Australian Journal of Agricultural Research 28, 879-889.

Drazeta, L., Lang, A., Hall, A.J., Volz, R.K., and Jameson, P.E. (2004). Causes and effects of changes

in xylem functionality in apple fruit. Annals of Botany 93, 275-282.

Dry, P.R., Longbottom, M.L., Mcloughlin, S., Johnson, T.E., and Collins, C. (2010). Classification of

reproductive performance of ten winegrape varieties. Australian Journal of Grape and Wine

Research 16, 47-55.

Esbensen, K.H. (2004). Multivariate Data Analysis - In Practice. Oslo, Norway: CAMO Process.

Espana, L., Heredia-Guerrero, J.A., Segado, P., Benitez, J.J., Heredia, A., and Dominguez, E. (2014).

Biomechanical properties of the tomato (Solanum lycopersicum) fruit cuticle during

development are modulated by changes in the relative amounts of its components. New

Phytologist 202, 790-802.

Etchebarne, F., Ojeda, H., and Deloire, A. (2009). "Grape berry mineral nutrition composition in relation

to vine water status & leaf area/fruit ratio," in Grapevine Molecular Physiology & Biotechnology,

ed. K.A. Roubelakis-Angelakis. 2 ed: Springer Science & Business Media B.V.).

Evans, D.E., Briars, S.A., and Williams, L.E. (1991). Active calcium-transport by plant-cell membranes.

J Exp Bot 42, 285-303.

Felle, H.H., Hanstein, S., Steinmeyer, R., and Hedrich, R. (2000). Dynamics of ionic activities in the

apoplast of the sub-stomatal cavity of intact Vicia faba leaves during stomatal closure evoked

by ABA and darkness. Plant Journal 24, 297-304.

142

Ferguson, I.B. (1984). Calcium in plant senescence and fruit ripening. Plant Cell and Environment 7,

477-489.

Fozard, J.A., King, J.R., and Bennett, M.J. (2013). Modelling auxin efflux carrier phosphorylation and

localization. Journal of Theoretical Biology 319, 34-49.

Franceschi, V.R., and Nakata, P.A. (2005). "Calcium oxalate in plants: Formation and function," in

Annual Review of Plant Biology. (Palo Alto: Annual Reviews), 41-71.

Fritz, E. (2007). Measurement of cation exchange capacity (CEC) of plant cell walls by X-ray

microanalysis (EDX) in the transmission electron microscope. Microscopy and Microanalysis

13, 233-244.

Fuentes, S., Ljung, K., Sorefan, K., Alvey, E., Harberd, N.P., and Ostergaard, L. (2012). Fruit growth in

Arabidopsis occurs via DELLA-dependent and DELLA-independent gibberellin responses.

Plant Cell 24, 3982-3996.

Fuentes, S., Sullivan, W., Tilbrook, J., and Tyerman, S. (2010). A novel analysis of grapevine berry

tissue demonstrates a variety-dependent correlation between tissue vitality and berry shrivel.

Australian Journal of Grape and Wine Research 16, 327-336.

Gabler, F.M., Smilanick, J.L., Mansour, M., Ramming, D.W., and Mackey, B.E. (2003). Correlations of

morphological, anatomical, and chemical features of grape berries with resistance to Botrytis

cinerea. Phytopathology 93, 1263-1273.

Gambetta, G.A., Matthews, M.A., Shaghasi, T.H., Mcelrone, A.J., and Castellarin, S.D. (2010). Sugar

and abscisic acid signaling orthologs are activated at the onset of ripening in grape. Planta 232,

219-234.

Gendre, D., Mcfarlane, H.E., Johnson, E., Mouille, G., Sjodin, A., Oh, J., Levesque-Tremblay, G.,

Watanabe, Y., Samuels, L., and Bhalerao, R.P. (2013). Trans-Golgi network localized

ECHIDNA/Ypt interacting protein complex is required for the secretion of cell wall

polysaccharides in Arabidopsis. Plant Cell 25, 2633-2646.

Giacosa, S., Torchio, F., Segade, S.R., Caudana, A., Gerbi, V., and Rolle, L. (2012). Varietal

relationship between skin break force and off-vine withering process for winegrapes. Drying

Technology 30, 726-732.

143

Giacosa, S., Torchio, F., Segade, S.R., Gaiotti, F., Tomasi, D., Lovat, L., Vincenzi, S., and Rolle, L.

(2013). Physico-mechanical evaluation of the aptitude of berries of red wine grape varieties to

resist the compression in carbonic maceration vinification. International Journal of Food

Science and Technology 48, 817-825.

Gilliham, M., Dayod, M., Hocking, B.J., Xu, B., Conn, S.J., Kaiser, B.N., Leigh, R.A., and Tyerman, S.D.

(2011). Calcium delivery and storage in plant leaves: exploring the link with water flow. Journal

of Experimental Botany 62, 2233-2250.

Graham, N.S., Hammond, J.P., Lysenko, A., Mayes, S., Lochlainn, S.O., Blasco, B., Bowen, H.C.,

Rawlings, C.J., Rios, J.J., Welham, S., Carion, P.W.C., Dupuy, L.X., King, G.J., White, P.J.,

and Broadley, M.R. (2014). Genetical and comparative genomics of Brassica under altered Ca

supply identifies Arabidopsis Ca-transporter orthologs. Plant Cell 26, 2818-2830.

Grignon, C., and Sentenac, H. (1991). pH and ionic conditions in the apoplast. Ann Rev of Plant Phys

42, 103-128.

Han, S.C., Tang, R.H., Anderson, L.K., Woerner, T.E., and Pei, Z.M. (2003). A cell surface receptor

mediates extracellular Ca2+ sensing in guard cells. Nature 425, 196-200.

Hawker, J.S., and Walker, R.R. (1978). Effect of sodium chloride on growth and fruiting of Cabernet

Sauvignon vines. American Journal of Enology and Viticulture 29, 172-176.

Hayes, M.A., Davies, C., and Dry, I.B. (2007). Isolation, functional characterization, and expression

analysis of grapevine (Vitis vinifera L.) hexose transporters: differential roles in sink and source

tissues. Journal of Experimental Botany 58, 1985-1997.

Hayes, M.A., Feechan, A., and Dry, I.B. (2010). Involvement of abscisic acid in the coordinated

regulation of a stress-inducible hexose transporter (VvHT5) and a cell wall invertase in

grapevine in response to biotrophic fungal infection. Plant Physiology 153, 211-221.

Hayter, M.L., and Peterson, C.A. (2004). Can Ca2+ fluxes to the root xylem be sustained by Ca2+-

ATPases in exodermal and endodermal plasma membranes? Plant Physiology 136, 4318-

4325.

144

Henderson, S.W., Baumann, U., Blackmore, D.H., Walker, A.R., Walker, R.R., and Gilliham, M. (2014).

Shoot chloride exclusion and salt tolerance in grapevine is associated with differential ion

transporter expression in roots. Bmc Plant Biology 14.

Ho, L.C., Belda, R., Brown, M., Andrews, J., and Adams, P. (1993). Uptake and transport of calcium

and the possible causes of blossom-end rot in tomato. Journal of Experimental Botany 44, 509-

518.

Ho, L.C., and White, P.J. (2005). A cellular hypothesis for the induction of blossom-end rot in tomato

fruit. Annals of Botany 95, 571-581.

Hocking, P.J. (1980). The composition of phloem exudate and xylem sap from tree tobacco (Nicotiana

glauca Grah.). Annals of Botany 45, 633-643.

Hosmani, P.S., Kamiya, T., Danku, J., Naseer, S., Geldner, N., Guerinot, M.L., and Salt, D.E. (2013).

Dirigent domain-containing protein is part of the machinery required for formation of the lignin-

based Casparian strip in the root. Proceedings of the National Academy of Sciences of the

United States of America 110, 14498-14503.

Hoson, T. (1998). Apoplast as the site of response to environmental signals. Journal of Plant Research

111, 167-177.

Howie, H., and Lloyd, J. (1989). Response of orchard 'Washington Navel' orange, Citrus sinensis (L) to

saline irrigation water. 2. Flowering, fruit set and fruit growth. Australian Journal of Agricultural

Research 40, 371-380.

Huang, X.M., Huang, H.B., and Wang, H.C. (2005). Cell walls of loosening skin in post-veraison grape

berries lose structural polysaccharides and calcium while accumulate structural proteins.

Scientia Horticulturae 104, 249-263.

Hyodo, H., Terao, A., Furukawa, J., Sakamoto, N., Yurimoto, H., Satoh, S., and Iwai, H. (2013). Tissue

specific localization of pectin-Ca2+ cross-linkages and pectin methyl-esterification during fruit

ripening in tomato (Solanum lycopersicum). Plos One 8.

Jiang, L.X., Yang, S.L., Xie, L.F., Puah, C.S., Zhang, X.Q., Yang, W.C., Sundaresan, V., and Ye, D.

(2005). VANGUARD1 encodes a pectin methylesterase that enhances pollen tube growth in

the Arabidopsis style and transmitting tract. Plant Cell 17, 584-596.

145

Johnson, C.H., Knight, M.R., Kondo, T., Masson, P., Sedbrook, J., Haley, A., and Trewavas, A. (1995).

Circadian oscillations of cytosolic and chloroplastic free calcium in plants. Science 269, 1863-

1865.

Johnson, C.H., Shingles, R., and Ettinger, W. (2006). "Regulation and role of calcium fluxes in the

chloroplast," in The structure and function of plastids, eds. R. Wise & J. Hoober. (Dordrecht:

Springer), 403-416.

Karley, A.J., Leigh, R.A., and Sanders, D. (2000). Differential ion accumulation and ion fluxes in the

mesophyll and epidermis of barley. Plant Physiology 122, 835-844.

Keller, M., and Shrestha, P.M. (2014). Solute accumulation differs in the vacuoles and apoplast of

ripening grape berries. Planta 239, 633-642.

Kim, K.N., Cheong, Y.H., Grant, J.J., Pandey, G.K., and Luan, S. (2003). CIPK3, a calcium sensor-

associated protein kinase that regulates abscisic acid and cold signal transduction in

Arabidopsis. Plant Cell 15, 411-423.

Knoche, M., and Peschel, S. (2007). Gibberellins increase cuticle deposition in developing tomato fruit.

Plant Growth Regulation 51, 1-10.

Kohorn, B.D., and Kohorn, S.L. (2012). The cell wall-associated kinases,WAKs, as pectin receptors.

Frontiers in Plant Science 3.

Konrad, K.R., Wudick, M.M., and Feijo, J.A. (2011). Calcium regulation of tip growth: new genes for old

mechanisms. Current Opinion in Plant Biology 14, 721-730.

Krasnow, M., Matthews, M., and Shackel, K. (2008). Evidence for substantial maintenance of

membrane integrity and cell viability in normally developing grape (Vitis vinifera L.) berries

throughout development. Journal of Experimental Botany 59, 849-859.

Kraus, J.E., De Sousa, H.C., Rezende, M.H., Castro, N.M., Vecchi, C., and Luque, R. (1998). Astra

blue and basic fuchsin double staining of plant materials. Biotechnic & Histochemistry 73, 235-

243.

Kudla, J., Batistic, O., and Hashimoto, K. (2010). Calcium signals: the lead currency of plant information

processing. Plant Cell 22, 541-563.

146

Kumar, R., Khurana, A., and Sharma, A.K. (2014). Role of plant hormones and their interplay in

development and ripening of fleshy fruits. Journal of Experimental Botany 65, 4561-4575.

Lacampagne, S., Lambert, C., Belleau-Deytieux, C., L'hyvernay, A., Doneche, B., and Geny, L. (2010).

"Expression, activity and cellular localization of pectin methylesterase in grape berry skin

during ripening," in Vi International Postharvest Symposium, eds. M. Erkan & U. Aksoy.

(Leuven 1: Int Soc Horticultural Science), 1057-1062.

Lang, A. (1990). Xylem, phloem and transpiration flows in developing apple fruit. Journal of

Experimental Botany 41, 645-651.

Lang, A., and During, H. (1991). Partitioning control by water potential gradient- Evidence for

compartmentation breakdown in grape berries. Journal of Experimental Botany 42, 1117-1122.

Lang, A., and Thorpe, M.R. (1989). Xylem, phloem and transpiration flows in a grape - application of a

technique for measuring the volume of attached fruits to high-resolution using Archimedes

principle. Journal of Experimental Botany 40, 1069-1078.

Lee, J., Park, I., Lee, Z.W., Kim, S.W., Baek, N., Park, H.S., Park, S.U., Kwon, S., and Kim, H. (2013).

Regulation of the major vacuolar Ca2+ transporter genes, by intercellular Ca2+ concentration

and abiotic stresses, in tip-burn resistant Brassica oleracea. Molecular Biology Reports 40,

177-188.

Leng, P., Yuan, B., and Guo, Y. (2014). The role of abscisic acid in fruit ripening and responses to

abiotic stress. Journal of Experimental Botany 65, 4577-4588.

Leung, J., Bouvierdurand, M., Morris, P.C., Guerrier, D., Chefdor, F., and Giraudat, J. (1994).

Arabidopsis ABA response gene ABI1 - features of a calcium-modulated protein phosphatase.

Science 264, 1448-1452.

Lionetti, V., Cervone, F., and Bellincampi, D. (2012). Methyl esterification of pectin plays a role during

plant-pathogen interactions and affects plant resistance to diseases. Journal of Plant

Physiology 169, 1623-1630.

Lionetti, V., Raiola, A., Camardella, L., Giovane, A., Obel, N., Pauly, M., Favaron, F., Cervone, F., and

Bellincampi, D. (2007). Overexpression of pectin methylesterase inhibitors in Arabidopsis

restricts fungal infection by Botrytis cinerea. Plant Physiology 143, 1871-1880.

147

Lopez-Casado, G., Matas, A.J., Dominguez, E., Cuartero, J., and Heredia, A. (2007). Biomechanics of

isolated tomato (Solanum lycopersicum L.) fruit cuticles: the role of the cutin matrix and

polysaccharides. Journal of Experimental Botany 58, 3875-3883.

Lu, Y.Q., Liu, H.P., Wang, Y., Zhang, X.Z., and Han, Z.H. (2013). Synergistic roles of leaf boron and

calcium during the growing season in affecting sugar and starch accumulation in ripening apple

fruit. Acta Physiologiae Plantarum 35, 2483-2492.

Madrid, R., Valverde, M., Alcolea, V., and Romojaro, F. (2004). Influence of calcium nutrition on water

soaking disorder during ripening of Cantaloupe melon. Scientia Horticulturae 101, 69-79.

Martin, L.B.B., and Rose, J.K.C. (2014). There’s more than one way to skin a fruit: formation and

functions of fruit cuticles. Journal of Experimental Botany 65, 4639-4651.

Marzouk, H.A., and Kassem, H.A. (2011). Improving yield, quality, and shelf life of Thompson seedless

grapevine by preharvest foliar applications. Scientia Horticulturae 130, 425-430.

Matthews, M.A., Thomas, T.R., and Shackel, K.A. (2009). Fruit ripening in Vitis vinifera L.: possible

relation of veraison to turgor and berry softening. Australian Journal of Grape and Wine

Research 15, 278-283.

Mazzeo, M., Dichio, B., Clearwater, M.J., Montanaro, G., and Xiloyannis, C. (2013). Hydraulic

resistance of developing Actinidia fruit. Annals of Botany 112, 197-205.

Mcainsh, M.R., and Pittman, J.K. (2009). Shaping the calcium signature. New Phytologist 181, 275-

294.

Mcatee, P., Karim, S., Schaffer, R., and David, K. (2013). A dynamic interplay between phytohormones

is required for fruit development, maturation, and ripening. Frontiers in Plant Science 4.

Mccarthy, M.G., and Coombe, B.G. (1999). Is weight loss in ripening grape berries cv. Shiraz caused

by impeded phloem transport? Australian Journal of Grape and Wine Research 5.

Mohnen, D. (2008). Pectin structure and biosynthesis. Current Opinion in Plant Biology 11, 266-277.

Montanaro, G., Dichio, B., and Xiloyannis, C. (2010). Significance of fruit transpiration on calcium

nutrition in developing apricot fruit. Journal of Plant Nutrition and Soil Science 173, 618-622.

148

Montanaro, G., Dichio, B., Xiloyannis, C., and Celano, G. (2006). Light influences transpiration and

calcium accumulation in fruit of kiwifruit plants (Actinidia deliciosa var. deliciosa). Plant Science

170, 520-527.

Montanaro, G., Dichio, B., Xiloyannis, C., and Lang, A. (2012). Fruit transpiration in kiwifruit:

environmental drivers and predictive model. Aob Plants.

Morvan, O., Quentin, M., Jauneau, A., Mareck, A., and Morvan, C. (1998). Immunogold localization of

pectin methylesterases in the cortical tissues of flax hypocotyl. Protoplasma 202, 175-184.

Mullins, M.G., and Rajasekaran, K. (1981). Fruiting cuttings- revised method for producing test plants of

grapevine cultivars. American Journal of Enology and Viticulture 32, 35-40.

Nardini, A., Gasco, A., Trifilo, P., Lo Gullo, M.A., and Salleo, S. (2007). Ion-mediated enhancement of

xylem hydraulic conductivity is not always suppressed by the presence of Ca2+ in the sap.

Journal of Experimental Botany 58, 2609-2615.

Nardini, A., Salleo, S., and Jansen, S. (2011). More than just a vulnerable pipeline: xylem physiology in

the light of ion-mediated regulation of plant water transport. Journal of Experimental Botany 62,

4701-4718.

Nawrath, C., Schreiber, L., Franke, R.B., Geldner, N., Reina-Pinto, J.J., and Kunst, L. (2013).

Apoplastic diffusion barriers in Arabidopsis. The Arabidopsis book / American Society of Plant

Biologists 11.

Ng, J.K.T., Schroder, R., Sutherland, P.W., Hallett, I.C., Hall, M.I., Prakash, R., Smith, B.G., Melton,

L.D., and Johnston, J.W. (2013). Cell wall structures leading to cultivar differences in softening

rates develop early during apple (Malus x domestica) fruit growth. Bmc Plant Biology 13.

Ngouemazong, D.E., Jolei, R.P., Cardinaels, R., Fraeye, I., Van Loey, A., Moldenaers, P., and

Hendrickx, M. (2012). Stiffness of Ca2+-pectin gels: combined effects of degree and pattern of

methylesterification for various Ca2+ concentrations. Carbohydrate Research 348, 69-76.

Nguema-Ona, E., Moore, J.P., Fagerstrom, A.D., Fangel, J.U., Willats, W.G.T., Hugo, A., and Vivier,

M.A. (2013). Overexpression of the grapevine PGIP1 in tobacco results in compositional

changes in the leaf arabinoxyloglucan network in the absence of fungal infection. Bmc Plant

Biology 13.

149

Nishizawa, T., Kobayashi, T., and Aikawa, T. (2004). Effect of calcium supply on the physiology of fruit

tissue in 'Andesu' netted melon. Journal of Horticultural Science & Biotechnology 79, 500-508.

Northcote, K.H. (1988). "Soils and Australian Viticulture," in Viticulture - Resources in Australia, eds.

B.G. Coombe & P.R. Dry. 1 ed (Underdale, Australia: Australian Industrial Publishers Pty. Ltd.),

61-90.

Nunan, K.J., Davies, C., Robinson, S.P., and Fincher, G.B. (2001). Expression patterns of cell wall-

modifying enzymes during grape berry development. Planta 214, 257-264.

Nunan, K.J., Sims, I.M., Bacic, A., Robinson, S.P., and Fincher, G.B. (1998). Changes in cell wall

composition during ripening of grape berries. Plant Physiology 118, 783-792.

Orfila, C., and Knox, J.P. (2000). Spatial regulation of pectic polysaccharides in relation to pit fields in

cell walls of tomato fruit pericarp. Plant Physiology 122, 775-781.

Ortega-Regules, A., Ros-Garcia, J.M., Bautista-Ortin, A.B., Lopez-Roca, J.M., and Gomez-Plaza, E.

(2008). Changes in skin cell wall composition during the maturation of four premium wine grape

varieties. Journal of the Science of Food and Agriculture 88, 420-428.

Osorio, S., Castillejo, C., Quesada, M.A., Medina-Escobar, N., Brownsey, G.J., Suau, R., Heredia, A.,

Botella, M.A., and Valpuesta, V. (2008). Partial demethylation of oligogalacturonides by pectin

methyl esterase 1 is required for eliciting defence responses in wild strawberry (Fragaria

vesca). Plant Journal 54, 43-55.

Osorio, S., Scossa, F., and Fernie, A.R. (2013). Molecular regulation of fruit ripening. Frontiers in Plant

Science 4.

Pan, Q.H., Li, M.J., Peng, C.C., Zhang, N., Zou, X., Zou, K.Q., Wang, X.L., Yu, X.C., Wang, X.F., and

Zhang, D.P. (2005). Abscisic acid activates acid invertases in developing grape berry.

Physiologia Plantarum 125, 157-170.

Pandey, G.K., Cheong, Y.H., Kim, K.N., Grant, J.J., Li, L.G., Hung, W., D'angelo, C., Weinl, S., Kudla,

J., and Luan, S. (2004). The calcium sensor calcineurin B-Like 9 modulates abscisic acid

sensitivity and biosynthesis in Arabidopsis. Plant Cell 16, 1912-1924.

Pfister, A., Barberon, M., Alassimone, J., Kalmbach, L., Lee, Y., Vermeer, J.E.M., Yamazaki, M., Li,

G.W., Maurel, C., Takano, J., Kamiya, T., Salt, D.E., Roppolo, D., and Geldner, N. (2014). A

150

receptor-like kinase mutant with absent endodermal diffusion barrier displays selective nutrient

homeostasis defects. Elife 3.

Pilati, S., Brazzale, D., Guella, G., Milli, A., Ruberti, C., Biasioli, F., Zottini, M., and Moser, C. (2014).

The onset of grapevine berry ripening is characterized by ROS accumulation and

lipoxygenase-mediated membrane peroxidation in the skin. Bmc Plant Biology 14.

Pitann, B., Schubert, S., and Muehling, K.H. (2009). Decline in leaf growth under salt stress is due to an

inhibition of H+-pumping activity and increase in apoplastic pH of maize leaves. Journal of Plant

Nutrition and Soil Science 172, 535-543.

Plavcova, L., and Hacke, U.G. (2011). Heterogeneous distribution of pectin epitopes and calcium in

different pit types of four angiosperm species. New Phytologist 192, 885-897.

Prior, L.D., Grieve, A.M., and Cullis, B.R. (1992). Sodium chloride and soil texture interactions in

irrigated field grown Sultana grapevines .I. Yield and fruit quality. Australian Journal of

Agricultural Research 43, 1051-1066.

Redgwell, R.J., Macrae, E., Hallett, I., Fischer, M., Perry, J., and Harker, R. (1997). In vivo and in vitro

swelling of cell walls during fruit ripening. Planta 203, 162-173.

Rios, J.J., Lochlainn, S.O., Devonshire, J., Graham, N.S., Hammond, J.P., King, G.J., White, P.J.,

Kurup, S., and Broadley, M.R. (2012). Distribution of calcium (Ca) and magnesium (Mg) in the

leaves of Brassica rapa under varying exogenous Ca and Mg supply. Annals of Botany 109,

1081-1089.

Ristic, R., and Iland, P.G. (2005). Relationships between seed and berry development of Vitis vinifera

L. cv Shiraz: Developmental changes in seed morphology and phenolic composition. Australian

Journal of Grape and Wine Research 11, 43-58.

Rober-Kleber, N., Albrechtova, J.T.P., Fleig, S., Huck, N., Michalke, W., Wagner, E., Speth, V.,

Neuhaus, G., and Fischer-Iglesias, C. (2003). Plasma membrane H+-ATPase is involved in

auxin-mediated cell elongation during wheat embryo development. Plant Physiology 131, 1302-

1312.

151

Robinson, J.B., Treeby, M.T., and Stephenson, R.A. (1997). "Fruits, vines and nuts", in: Plant Analysis:

An Interpretation Manual. (eds.) D. Reuter & J.B. Robinson. 2nd ed. (Melbourne: CSIRO

Publishing).

Rogiers, S.Y., Greer, D.H., Hatfield, J.M., Orchard, B.A., and Keller, M. (2006a). Mineral sinks within

ripening grape berries (Vitis vinifera L.). Vitis 45, 115-123.

Rogiers, S.Y., Greer, D.H., Hatfield, J.M., Orchard, B.A., and Keller, M. (2006b). Solute transport into

Shiraz berries during development and late-ripening shrinkage. American Journal of Enology

and Viticulture 57, 73-80.

Rogiers, S.Y., Keller, M., Holzapfel, B.P., and Virgona, J.M. (2000). Accumulation of potassium and

calcium by ripening berries on field vines of Vitis vinifera (L) cv. Shiraz. Australian Journal of

Grape and Wine Research 6, 240-243.

Rolle, L., Siret, R., Segade, S.R., Maury, C., Gerbi, V., and Jourjon, F. (2012a). Instrumental texture

analysis parameters as markers of table-grape and winegrape quality: a review. American

Journal of Enology and Viticulture 63, 11-28.

Rolle, L., Torchio, F., Ferrandino, A., and Guidoni, S. (2012b). Influence of wine-grape skin hardness

on the kinetics of anthocyanin extraction. International Journal of Food Properties 15, 249-261.

Rounds, C.M., Lubeck, E., Hepler, P.K., and Winship, L.J. (2011). Propidium iodide competes with Ca2+

to label pectin in pollen tubes and Arabidopsis root hairs. Plant Physiology 157, 175-187.

Roy, S., Gillen, G., Conway, W.S., Watada, A.E., and Wergin, W.P. (1995). Use of secondary ion mass

spectrometry to image (44)calcium uptake in the cell walls of apple fruit. Protoplasma 189, 163-

172.

Ruan, Y.L., Patrick, J.W., Bouzayen, M., Osorio, S., and Fernie, A.R. (2012). Molecular regulation of

seed and fruit set. Trends in Plant Science 17, 656-665.

Ruan, Y.L., Patrick, J.W., and Brady, C.J. (1996). The composition of apoplast fluid recovered from

intact developing tomato fruit. Australian Journal of Plant Physiology 23, 9-13.

Saladie, M., Matas, A.J., Isaacson, T., Jenks, M.A., Goodwin, S.M., Niklas, K.J., Ren, X., Labavitch,

J.M., Shackel, K.A., Fernie, A.R., Lytovchenko, A., O'neill, M.A., Watkins, C.B., and Rose,

152

J.K.C. (2007). A reevaluation of the key factors that influence tomato fruit softening and

integrity. Plant Physiology 144, 1012-1028.

Sampedro, J., and Cosgrove, D.J. (2005). The expansin superfamily. Genome Biology 6.

Sams, C.E., and Conway, W.S. (1984). Effect of calcium infiltration on ethylene production, respiration

rate, soluble polyuronide content, and quality of Golden Delicious apple fruit. Journal of the

American Society for Horticultural Science 109, 53-57.

Sattelmacher, B. (2001). Tansley review no. 22 - The apoplast and its significance for plant mineral

nutrition. New Phytol 149, 167-192.

Saure, M.C. (2005). Calcium translocation to fleshy fruit: its mechanism and endogenous control.

Scientia Horticulturae 105, 65-89.

Scharwies, J. (2013). Water transport in grape berry and pre-harvest berry dehydration. M. Ag. Sc.,

University of Adelaide.

Schlosser, J., Olsson, N., Weis, M., Reid, K., Peng, F., Lund, S., and Bowen, P. (2008). Cellular

expansion and gene expression in the developing grape (Vitis vinifera L.). Protoplasma 232,

255-265.

Seifert, G.J., and Blaukopf, C. (2010). Irritable walls: the plant extracellular matrix and signaling. Plant

Physiology 153, 467-478.

Serrani, J.C., Ruiz-Rivero, O., Fos, M., and Garcia-Martinez, J.L. (2008). Auxin-induced fruit-set in

tomato is mediated in part by gibberellins. Plant Journal 56, 922-934.

Shabala, S. (2000). Ionic and osmotic components of salt stress specifically modulate net ion fluxes

from bean leaf mesophyll. Plant Cell and Environment 23, 825-837.

Shishova, M., and Lindberg, S. (2010). A new perspective on auxin perception. Journal of Plant

Physiology 167, 417-422.

Smith, J.P., Schmidtke, L.M., Muller, M.C., and Holzapfel, B.P. (2014). Measurement of the

concentration of nutrients in grapevine petioles by attenuated total reflectance Fourier

transform infrared spectroscopy and chemometrics. Australian Journal of Grape and Wine

Research 20, 299-309.

153

Sun, L., Sun, Y.F., Zhang, M., Wang, L., Ren, J., Cui, M.M., Wang, Y.P., Ji, K., Li, P., Li, Q., Chen, P.,

Dai, S.J., Duan, C.R., Wu, Y., and Leng, P. (2012). Suppression of 9-cis-epoxycarotenoid

dioxygenase, which encodes a key enzyme in abscisic acid biosynthesis, alters fruit texture in

transgenic tomato. Plant Physiology 158, 283-298.

Sun, Q.A., Greve, L.C., and Labavitch, J.M. (2011). Polysaccharide compositions of intervessel pit

membranes contribute to Pierce's disease resistance of grapevines. Plant Physiology 155,

1976-1987.

Sutherland, P., Hallett, L., Redgwell, R., Benhamou, N., and Macrae, E. (1999). Localization of cell wall

polysaccharides during kiwifruit (Actinidia deliciosa) ripening. International Journal of Plant

Sciences 160, 1099-1109.

Tang, R.H., Han, S.C., Zheng, H.L., Cook, C.W., Choi, C.S., Woerner, T.E., Jackson, R.B., and Pei,

Z.M. (2007). Coupling diurnal cytosolic Ca2+ oscillations to the CAS-IP3 pathway in

Arabidopsis. Science 315, 1423-1426.

Teakle, N.L., and Tyerman, S.D. (2010). Mechanisms of Cl- transport contributing to salt tolerance.

Plant Cell and Environment 33, 566-589.

Thomas, T.R., Matthews, M.A., and Shackel, K.A. (2006). Direct in situ measurement of cell turgor in

grape (Vitis vinifera L.) berries during development and in response to plant water deficits.

Plant Cell and Environment 29, 993-1001.

Thomas, T.R., Shackel, K.A., and Matthews, M.A. (2008). Mesocarp cell turgor in Vitis vinifera L.

berries throughout development and its relation to firmness, growth, and the onset of ripening.

Planta 228, 1067-1076.

Tibbits, C.W., Macdougall, A.J., and Ring, S.G. (1998). Calcium binding and swelling behaviour of a

high methoxyl pectin gel. Carbohydrate Research 310, 101-107.

Tilbrook, J., and Tyerman, S.D. (2008). Cell death in grape berries: varietal differences linked to xylem

pressure and berry weight loss. Functional Plant Biology 35, 173-184.

Tilbrook, J., and Tyerman, S.D. (2009). Hydraulic connection of grape berries to the vine: varietal

differences in water conductance into and out of berries, and potential for backflow. Functional

Plant Biology 36, 541-550.

154

Tsubaki, S., Ozaki, Y., Yonemori, K., and Azuma, J.-I. (2012). Mechanical properties of fruit-cuticular

membranes isolated from 27 cultivars of Diospyros kaki Thunb. Food Chemistry 132, 2135-

2139.

Tyerman, S.D., Tilbrook, J., Pardo, C., Kotula, L., Sullivan, W., and Steudle, E. (2004). Direct

measurement of hydraulic properties in developing berries of Vitis vinifera L. cv Shiraz and

Chardonnay. Australian Journal of Grape and Wine Research 10, 170-181.

Vallarino, J.G., and Osorio, S. (2012). Signaling role of oligogalacturonides derived during cell wall

degradation. Plant signaling & behavior 7, 1447-1449.

Van Doorn, W.G., Hiemstra, T., and Fanourakis, D. (2011). Hydrogel regulation of xylem water flow: an

alternative hypothesis. Plant Physiology 157, 1642-1649.

Vandeleur, R.K., Mayo, G., Shelden, M.C., Gilliham, M., Kaiser, B.N., and Tyerman, S.D. (2009). The

role of plasma membrane intrinsic protein aquaporins in water transport through roots: Diurnal

and drought stress responses reveal different strategies between isohdyric and anisohydric

cultivars of grapevine. Plant Physiology 149, 445-460.

Vandeleur, R.K., Sullivan, W., Athman, A., Jordans, C., Gilliham, M., Kaiser, B.N., and Tyerman, S.D.

(2014). Rapid shoot-to-root signalling regulates root hydraulic conductance via aquaporins.

Plant Cell and Environment 37, 520-538.

Vidal, S., Williams, P., O'neill, M.A., and Pellerin, P. (2001). Polysaccharides from grape berry cell

walls. Part I: tissue distribution and structural characterization of the pectic polysaccharides.

Carbohydrate Polymers 45, 315-323.

Vorwerk, S., Somerville, S., and Somerville, C. (2004). The role of plant cell wall polysaccharide

composition in disease resistance. Trends in Plant Science 9, 203-209.

Wada, H., Matthews, M.A., and Shackel, K.A. (2009). Seasonal pattern of apoplastic solute

accumulation and loss of cell turgor during ripening of Vitis vinifera fruit under field conditions.

Journal of Experimental Botany 60, 1773-1781.

Wada, H., Shackel, K., and Matthews, M.A. (2006). Apoplastic solute accumulation accounts for berry

cell turgor loss before veraison in Vitis vinifera. American Journal of Enology and Viticulture 57,

387A-388A.

155

Wada, H., Shackel, K.A., and Matthews, M.A. (2008). Fruit ripening in Vitis vinifera: apoplastic solute

accumulation accounts for pre-veraison turgor loss in berries. Planta 227, 1351-1361.

Walker, R.R., Blackmore, D.H., and 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. Australian Journal of Grape and Wine

Research 16, 243-257.

Walker, R.R., Blackmore, D.H., Clingeleffer, P.R., and Emanuelli, D. (2014). Rootstock type determines

tolerance of Chardonnay and Shiraz to long-term saline irrigation. Australian Journal of Grape

and Wine Research 20, 496-506.

Wang, L., and Ruan, Y.L. (2013). Regulation of cell division and expansion by sugar and auxin

signaling. Frontiers in Plant Science 4.

Wang, W.H., Chen, J., Liu, T.W., Chen, J., Han, A.D., Simon, M., Dong, X.J., He, J.X., and Zheng, H.L.

(2014). Regulation of the calcium-sensing receptor in both stomatal movement and

photosynthetic electron transport is crucial for water use efficiency and drought tolerance in

Arabidopsis. Journal of Experimental Botany 65, 223-234.

Wang, W.H., Yi, X.Q., Han, A.D., Liu, T.W., Chen, J., Wu, F.H., Dong, X.J., He, J.X., Pei, Z.M., and

Zheng, H.L. (2012). Calcium-sensing receptor regulates stomatal closure through hydrogen

peroxide and nitric oxide in response to extracellular calcium in Arabidopsis. Journal of

Experimental Botany 63, 177-190.

Webb, A.a.R., Larman, M.G., Montgomery, L.T., Taylor, J.E., and Hetherington, A.M. (2001). The role

of calcium in ABA-induced gene expression and stomatal movements. Plant Journal 26, 351-

362.

Whalley, H.J., and Knight, M.R. (2013). Calcium signatures are decoded by plants to give specific gene

responses. New Phytologist 197, 690-693.

White, P.J. (2000). Calcium channels in higher plants. BBA-Biomembranes 1465, 171-189.

White, P.J. (2001). The pathways of calcium movement to the xylem. J. Exp. Bot. 52, 891-899.

156

White, P.J. (2012). "Chapter 3 - Long-distance Transport in the Xylem and Phloem," in Marschner's

Mineral Nutrition of Higher Plants (Third Edition), ed. P. Marschner. (San Diego: Academic

Press), 49-70.

White, P.J., and Broadley, M.R. (2003). Calcium in plants. Annals of Botany 92, 487-511.

Willats, W.G.T., Orfila, C., Limberg, G., Buchholt, H.C., Van Alebeek, G., Voragen, A.G.J., Marcus,

S.E., Christensen, T., Mikkelsen, J.D., Murray, B.S., and Knox, J.P. (2001). Modulation of the

degree and pattern of methyl-esterification of pectic homogalacturonan in plant cell walls -

Implications for pectin methyl esterase action, matrix properties, and cell adhesion. Journal of

Biological Chemistry 276, 19404-19413.

Yu, X.C., Li, M.J., Gao, G.F., Feng, H.Z., Geng, X.Q., Peng, C.C., Zhu, S.Y., Wang, X.J., Shen, Y.Y.,

and Zhang, D.P. (2006). Abscisic acid stimulates a calcium-dependent protein kinase in grape

berry. Plant Physiology 140, 558-579.

Zhu, S.Y., Yu, X.C., Wang, X.J., Zhao, R., Li, Y., Fan, R.C., Shang, Y., Du, S.Y., Wang, X.F., Wu, F.Q.,

Xu, Y.H., Zhang, X.Y., and Zhang, D.P. (2007). Two calcium-dependent protein kinases, CPK4

and CPK11, regulate abscisic acid signal transduction in Arabidopsis. Plant Cell 19, 3019-

3036.

Zoccatelli, G., Zenoni, S., Savoi, S., Dal Santo, S., Tononi, P., Zandona, V., Dal Cin, A., Guantieri, V.,

Pezzotti, M., and Tornielli, G.B. (2013). Skin pectin metabolism during the postharvest

dehydration of berries from three distinct grapevine cultivars. Australian Journal of Grape and

Wine Research 19, 171-179.

Zsivanovits, G., Macdougall, A.J., Smith, A.C., and Ring, S.G. (2004). Material properties of

concentrated pectin networks. Carbohydrate Research 339, 1317-1322.

Zwieniecki, M.A., Melcher, P.J., and Holbrook, N.M. (2001). Hydrogel control of xylem hydraulic

resistance in plants. Science 291, 1059-1062.