Physiological responses of ‘Italia’ grapevines infected with Esca pathogens‡

Copyright: © 2021 G.L. Bruno, M.P. Ippolito, F. Mannerucci, L. Bragazzi, F. Tommasi. This is an open access, peer-reviewed article published by Firenze University Press (http://www. fupress.com/pm) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


INTRODUCTION
The disease commonly called "Esca" is one of the longest recognized and most destructive diseases of grapevines (Vitis vinifera), and is associated with wood discolouration and decay, and sudden wilting of whole vines or individual vine arms within a few days (apoplexy or apoplectic symptoms), as well as leaf tiger stripe leaf symptoms (Ravaz, 1898;de Rolland, 1873;Marsais, 1923). Esca is now considered to be a complex of different diseases overlapping in the same vine or developing at different stages of a vine life. Esca-complex comprises brown wood streaking of grapevine cuttings, Petri disease, grapevine leaf stripe disease (GLSD, previously "young esca"), and white rot (which is at the origin of the name Esca). GLSD affects young and old vines, which show wood streaking and discolouration. The association between GLSD and white rot was described as a condition called "Esca proper". Within the Esca complex, white rot, and Esca proper can also show apoplectic symptoms (Surico, 2009;Mondello et al., 2018). Brown wood streaking, Petri disease and GLSD are also grouped under the name phaeotracheomycosis complex (Bertsch et al., 2012).
No pathogens have been isolated from leaves or berries of infected plants. Leaf tiger stripe and berry symptoms are considered linked to cultivar susceptibility, vine age, the microorganisms involved, pedoclimatic conditions and other physiological factors (Graniti et al., 2000;Surico et al., 2006;Bertsch et al., 2012;Guérin-Dubrana et al., 2013;Claverie et al., 2020). The substances originating in the discoloured woody tissues of affected trunks and branches contribute to development of the typical symptoms on leaves (Andolfi et al., 2011;Bertsch et al., 2012;Lecomte et al., 2012). These materials could be reaction products of the diseased wood, phytotoxic metabolites excreted by Esca-associated fungi, or a combination of these (Mugnai et al., 1999;. P. minimum and P. chlamydospora produce two naphthalenone pentaketides, scytalone and isosclerone, and exopolysaccharides including the α-glucan pullulan Tabacchi et al., 2000;Bruno and Sparapano 2006a;2006b;. Macro-and micro-nutrients also play roles in Esca complex symptom progression (Calzarano et al., 2009(Calzarano et al., , 2014Calzarano and Di Marco, 2018).
The aim of the present study was to gain insight into variations of physiological features of bleeding xylem sap and leaves of 'Italia' grapevine plants that were either healthy or naturally infected with P. minimum and P. chlamydospora showing brown wood streaking, or infected with P. minimum, P. chlamydospora and F. mediterranea showing brown wood streaking and white rot. Special emphasis was placed on the differences in hydrogen peroxide, lipid peroxidation, and antioxidant defence responses associated with the ascorbate-glutathione cycle. The changes in the physiology of diseased plants were assessed as possible factors in development of foliar symptoms.

Vineyard
A 20-year-old V. vinifera 'Italia' table grape vineyard (1600 vines) cropped in the countryside of Bari (Apulia, southern Italy) was used for sample collection. The vines, grafted onto 140-Ru rootstock, were trained by the Tendone system, and grown under irrigation in an alkaline clay soil. Since 2006, each vine had been under observation for symptom development of diseases within the Esca complex, i.e., foliar symptoms, or sudden wilting of GLSD or esca proper.

Wood core sampling and vine characterization
To assess the presence of symptoms and fungi in the grapevine trunks, two wood cores per vine were taken with a 95% ethanol pre-sterilised Pressler increment borer at 30 and 110 cm above the ground level. All the 1600 vines were surveyed. To prevent further wood contamination, wood core sampling holes were disinfected with copper oxychloride solution (20% in water) and filled with 2.5% copper oxychloride in linseed oil. Each core was cut to fragments 1.5-2 cm in length. Slices were surface-sterilized for 1 min in 70% ethanol, soaked for 1-2 min in a sodium hypochlorite solution (3% active chlorine), and rinsed three times in sterile distilled water. The slices were then aseptically cut into pieces that were seeded onto 90 mm diam. Petri dishes (five per plate) containing agar media. Malt extract (2%) agar (MEA), MEA amended with 0.25% chloramphenicol (MEAC) and MEA amended with 0.25% benomyl (MEAB) were used as the isolation media. MEAC and MEAB were used as semi-selective media for detection, respectively, of phaeotracheomycotic and basidiomycete fungi. Inoculated plates were incubated at 25±1°C in the dark. The isolation frequency (IF) of each fungus taxon was calculated as IF = 100 × (N i /N t ), where N i was the number of wood-fragments from which the fungus was isolated and N t the total number of seeded wood tissue pieces.
At completion of characterizations of the vine wood, associated with leaves and berries symptoms surveyed during 12 years, 15 vines were selected, including five with brown wood streaking, five with brown wood streaking and white rot, and five healthy (symptomless) that did not show any foliar symptoms, and any indications of wood or berry alterations.

Bleeding xylem sap sampling and characterization
To evaluate bleeding sap quantity, viscosity, and ascorbate, glutathione and hormone concentrations, bleeding xylem sap was collected from five vines with brown wood streaking, five with brown wood streaking and white rot, and five healthy vines. Bleeding xylem sap was collected at the bud-break vine growth stage (Baggiolini, 1979;Wilcox et al., 2015). From each selected plant, four vine shoots were cut and the end of each spur was surface-treated with sodium hypochlorite solution (3% active chlorine), then with 95% ethanol, and then rinsed twice with sterile distilled water. The sap exuded during the first 15 min was discarded. A sterile plastic bottle covered with aluminium foil was secured at the end of each bleeding spur to collect the liquid over the following 4 d. All sap samples were filtered through 0.45 mm membranes (Millipore).
The dynamic viscosity (η x ) of each sap sample was calculated as η x =[(ρ s ×t x )/(ρ w ×t w )]×η w , where ρ s = sap density, ρ w = water density, η w = water dynamic viscosity (0.8937´10 -3 Poiseuille), t s = flow time of sap in sec, and t w = water flow time (sec). Measurements of t s and t w were carried at 25±0.1°C using an Ostwald glass capillary viscometer (Cannon-Fenske Instruments).
For each sap sample, 2 mL were lyophilized, and the resulting powder was treated with 5% metaphosphoric acid (6 mL). After centrifugation (20,000g, 15 min, 4°C), the supernatant was used for total ascorbate and total glutathione determinations following the method of Zhang and Kirkham (1996).
The physiological effects of growth regulator substances in the xylem sap were evaluated using the filter paper disk method (Zhao et al., 1992), with excised cucumber (Cucumis sativus) cotyledon root formation (auxin) and cucumber cotyledon expansion (kinetin) bioassays. Indole-3-acetic acid and 6-furfurylaminopurine were dissolved in 95% ethanol and tested in the range of 0.3-50.0 μg mL -1 , and 95% ethanol was used as a control.

Leaf sampling and characterization
To evaluate leaf fresh and dry weights, and areas, as well as total chlorophyll, hydrogen peroxide, lipid peroxidation, ascorbic acid, dehydroascorbic acid, reduced and oxidized glutathione concentrations and the activity of enzymes involved in ascorbate regeneration, leaves were collected from the selected 15 vines. Leaves (ten per vine) were randomly picked from each vine during the unfolded leaf, fruit setting, cluster closing, and bunch ripening vine growth stages (Baggiolini, 1979;Wilcox et al., 2015). At the cluster closing and bunch ripening vine growth stages of the diseased vines, symptomless and symptomatic leaves were sampled. The leaf petioles were removed, and the leaves were photographed.
Each leaf was weighed with a Sartorius BP 210S analytical balance (Data Weighing Systems, Inc.) to assess leaf fresh weight.
Leaf area was estimated using ImageJ open-source image-processing software (National Institutes of Health).
Leaf dry weight was measured by drying 100 mg of leaves for 20 min at 105°C using an infrared LP 16-M desiccator (Mettler-Toledo SpA). Leaf moisture was calculated as a percentage (%) of fresh weight.
Total chlorophyll concentration was verified by Harborne's method (1973) using leaf lamina samples (2 g each) and 80% acetone (16 mL) as extraction solvent in an ice bath. Absorbance at 645 and 663 nm were measured using a Beckman DU 640 spectrophotometer (Beckman Coulter Inc.).
Hydrogen peroxide content was determined as reported by Lee and Lee (2000), using 1 g of leaf lamina ground with 4 mL of sodium phosphate buffer (0.1 M; pH 6.5).
Lipid peroxidation was assessed as malondialdehyde (MDA) quantity in 200 mg of leaf lamina samples (Heath and Packer, 1968).

Statistical analyses
Data were subjected to general linear analysis of variance models using the SAS/STAT version 9.0 (SAS Institute Inc.). Normal distributions were assessed using the Shapiro-Wilk tests, and homoscedasticity was assessed using Bartlett's tests. Means were compared using Fisher's LSD test at P ≤ 0.05. Data of morphological and physiological features of the grapevine leaves were analysed for vine typology (brown wood streaking, brown wood streaking and white rot, or healthy), vine growth stages (unfolded leaf, fruit setting, cluster closing, or bunch ripening), symptoms on leaves (presence or absence), and their interactions.

Wood core and symptom examination
The wood core examinations showed that 562 vines had brown wood streaking, 176 had brown wood streaking and rotted wood in the trunks, and 862 vines were healthy. P. chlamydospora and P. minimum were always associated with brown wood streaking, while F. medi-terranea was isolated only from rotted wood (Table 1). Other micromycetes, including species of Penicillium, Alternaria, Paecilomyces, Trichoderma, Chaetomium, Cladosporium, and Paraconiothyrium, and sterile fungi, were also isolated from diseased vines. Species of Penicillium and Alternaria were the only fungi isolated from cores from the healthy vines.
During the survey, symptoms were recorded on leaves and berries of diseased vines ( Figure 1). The typical tiger stripe symptoms were first observed in July on vines with brown wood streaking and with brown wood streaking and rotted wood in the trunks. On plants also infected with F. mediterranea, trunk cracking was recorded. No apoplexy was viewed on selected vines throughout the survey.

Bleeding xylem sap characterization
Healthy vines discharged the lowest quantities of xylem sap. Both typologies of diseased vines discharged four-and five-fold more sap than the healthy vines, and the sap from these vines had the greatest dynamic viscosity coefficients, total ascorbic acid and glutathione concentrations, and auxin-like and kinetin activities ( Table 2).

Leaf characterization
A selection of leaves sampled from the 15 selected vines is illustrated in Figure 1G. Morphological and physiological features were strongly affected by the vine growth stages, vine typology, symptom development, and their interactions (Table 3). Only symptomatic leaves collected from vines with brown wood streaking and white rot at cluster closing stages showed a small decrease (29.6%) in fresh weight compared with symptomless leaves collected from the same vines. At bunch ripening, leaves from healthy vines and symptomless leaves from vines with brown wood streaking and white rot reached maximum fresh weight. Symptomless leaves of vines with brown wood streaking had reduced mean leaf fresh weight compared with healthy vines. Symptomatic leaves from vines with both disease typologies had leaves with fresh weights that were further 35-40% less.
The greatest leaf dry weight ( Figure 2B) was recorded at the bunch ripening phase from symptomatic leaves of vines with brown wood streaking and white rot.
No statistically significant differences were found in leaf moisture content for the different vine disease categories. Leaf moisture contents were in the range 90 to 95% of leaf fresh weight.
Surface areas of leaves sampled at unfolded leaf and  fruit set stages did not show any statistically significant differences, while the leaves of healthy vines were significantly smaller than those of diseased plants during the other two vine growth stages ( Figure 2C). At cluster closing stage, the surface areas of symptomless leaves from diseased vines were 2-fold greater than for leave from the healthy vines. At bunch ripening, symptomatic and symptomless leaves of diseased vines had leaf surface areas that were three times greater than leaves from the healthy vines. During the four sampling times, total chlorophyll concentration in healthy vines was in the range of 160-165 mg g -1 leaf fresh weight ( Figure 3A). At the cluster closing and bunch ripening stages, diseased plants had less total chlorophyll. At cluster closing, in comparison with healthy vines, symptomless leaves from vines with brown wood streaking had 20% less total chlorophyll. This loss in symptomatic leaves was greater than 40%. Symptomless leaves from vines with brown wood streaking and white rot had 31% less total chlorophyll compared with healthy vines, and symptomatic leaves had a further 36% less total chlorophyll.
During the assessed four vine growth stages, healthy vines showed the least H 2 O 2 concentrations ( Figure 3B). Symptomless leaves collected from vines with brown wood streaking had increased H 2 O 2 content compared with the healthy vines. Symptomatic leaves collected during cluster closing and bunch ripening had 20% more H 2 O 2 than healthy vines. Leaves from vines with brown wood streaking and white rot, had more H 2 O 2 than healthy vines at all four growth stages. Symptomatic leaves collected during cluster closing or bunch ripening had 3% more H 2 O 2 than symptomless ones.
MDA concentrations were different between healthy vines and the two categories of disease at the four vine growth stages ( Figure 3C). Healthy vines reached minimum MDA contents in the unfolded leaf stage, and the greatest MDA concentration was occurred during bunch ripening. A similar trend was recorded in symptomless leaves from vines with brown wood streaking, but the final MDA concentrations were greater than in healthy vines. Symptomless leaves from vines with brown wood streaking and white rot, compared to healthy vines, had greater MDA content at all the sampling times. At the cluster closing and bunch ripening stages, symptomatic leaves from the same vines showed, respectively, further 43% and 11% increases in MDA contents.
During the four sampling times, ascorbic acid concentrations in healthy vines were always greater than in leaves from diseased vines ( Figure 4A). At each assessed growth stage, ascorbic acid contents were reduced by up to 78% in leaves collected from vines with brown wood    streaking, and 91% in leaves from vines with brown wood streaking and white rot. Leaves from healthy vines showed similar levels of dehydroascorbic acid during the four assessed growth stages ( Figure 4B). In the unfolded leaf and fruit set stages, leaves collected from vines with brown wood streaking and white rot showed the greatest dehydroascorbic acid concentrations.
The reduced glutathione contents of leaves collected from healthy vines were greater than in leaves collected from diseased plants ( Figure 4C).
The oxidized glutathione amounts ( Figure 4D) in symptomless leaves collected from vines with brown wood streaking, during the four considered growth stages, were approx. 38% less than in leaves from the healthy vines.
The lowest redox states (Table 4) were recorded in all symptomatic leaves tested.
No significant differences were detected between healthy and diseased vines in ascorbate peroxidase activity ( Figure 5A) during the four growth stages.
Dehydroascorbate reductase activity ( Figure 5B) in leaves from healthy vines and symptomless diseased vines reached minima in the unfolded leaf growth stage and were greatest at the bunch ripening stage.
Ascorbate free radical reductase activity ( Figure 5C) in leaves from healthy vines was the least at the unfolded leaf stage, increased during the next two growth stages, and reached a maximum at the bunch ripening stage. Compared to healthy vines, at the cluster closing and bunch ripening stages, leaves from diseased vines had reduced ascorbate free radical reductase activities.
Glutathione reductase activity ( Figure 5D) in leaves from healthy vines was least during the unfolded leaf stage and reached a maximum at bunch ripening. Compared to healthy vines, leaves from diseased vines had reduced glutathione reductase activity. No significant changes in glutathione reductase activity were detected in symptomless compared to symptomatic leaves from diseased vines. DISCUSSION In this study, in a 20-year-old 'Italia' vineyard surveyed since 2006 for symptoms of diseases within the Esca complex, we selected vines with brown wood streaking, vines with brown wood streaking and white rot, and healthy vines. The fungus isolation procedure confirmed the presence of P. minimum and P. chlamydo- spora in grapevine wood with brown wood streaking, and F. mediterranea was always associated with white rotted tissues. No Botryosphaeriaceae fungi or other grapevine trunk disease pathogens were isolated from diseased vines. No symptoms on foliage, wood, or berries were observed on healthy vines. Vines with brown wood streaking and vines with brown wood streaking and white rot showed 'tiger-stripe' symptoms on leaves, and spots, shrivelling and wilt of berries. These observations agreed with those previously described for vines affected by Esca complex pathogens Mondello et al., 2018).
The present study recorded differences in the contents of bleeding xylem sap and leaves between healthy and diseased vines. Bleeding of xylem sap is a process that characterizes grapevines and many other perennial plants as an effect of positive root pressure that transports water upward. Bleeding occurs in the springtime because increasing soil temperatures stimulated root pressure. Water fills the xylem vessels, dissolves, and pushes out air bubbles formed during the winter and restores xylem activity (Sperry et al., 1987). Cavitation reduces sap flux density and sap surface tension and induces xylem dysfunction (Hammond-Kosack and Jones, 2015). To bypass obstructions or cavitation, and restore vertical water conductivity, plants respond by producing new xylem conduits or refill cavitated vessels (Nardini et al., 2008).
Presence of F. mediterranea and its wood-degrading action could alter xylem conductivity, and thus, the quantity of bleeding sap. As in previous studies (Bruno Table 4. Ascorbate and glutathione redox states in symptomless or symptomatic grapevine leaves collected from 'Italia' grapevines during the vine growth stages of unfolded leaves (UL), fruit set (FS), cluster closing (CC) or bunch ripening (BR), from healthy vines (HV), or vines with brown wood streaking (BWSV) or brown wood streaking and white rot (BWSWRV). and Sparapano, 2006b;, diseased vines here analysed bled more abundantly than healthy vines. Viscosity is the capacity of a fluid layer to run with an adjacent layer. In the present study, the dynamic viscosity coefficient increased from healthy vines to the vines with brown wood streaking or with white rot and brown wood streaking. These results suggest that substances produced by fungal pathogens, and molecules resulting from cell component degradation by pathogen lytic enzymes, added to vine response molecules (such as phenols, tannins, flavonoids), could affect dynamic viscosity coefficients, and, thus, the xylem sap flow.
The presence of several plant hormones in host xylem sap has been assessed in herbaceous and woody plants, including grapevine (Niim and Torikata, 1978). In the present study the greatest auxin-and kinetin-like activities were detected in diseased vines. Auxin activity increased when the vines showed white rot symptoms associated with F. mediterranea.
This study also demonstrated that grapevine leaf surface area, leaf fresh and dry weights, and chlorophyll contents varied according to the host growth stages, but were significantly affected by the behaviour of the pathogens inside the woody tissues and, consequently, by altered physiological functions. Symptomatic leaves always had the least fresh and dry weights, and total chlorophyll concentrations. Diseased plants had physiological dysfunctions related to photosynthesis (i.e., reduced photosynthetic pigments) similar to those reported for Esca affected 'Cabernet Sauvignon' and 'Merlot' grapevines (Christen et al., 2007). Decreased gas exchange and chlorophyll fluorescence, and repression of photosynthesis-related genes have been detected for presymptomatic leaves of Esca-affected vines (Magnin-Robert et al., 2011). Chlorophyll decline leads to decreased photosynthesis efficiency, organic carbon production, host growth, and general plant health. Symptomatic leaves showed further reductions in fresh and dry weights associated with lamina necrosis and wilt. The main aetiological agents of the Esca complex produced phytotoxic metabolites involved, in vitro and in planta, with symptom development on leaves Sparapano, 2006a, 2006b;Luini et al., 2010;Andolfi et al., 2011). Chlorophyll decline could also explain the decrease in leaf weight because of low photosynthesis efficiency. Activation of plant defence mechanisms possibly modified host sugar metabolism, moving towards production of new molecules (Jeandet et al., 2002) and reducing carbohydrates used for plant growth and reproduction.
The experiments carried out in the present study have shown increases in leaf surface area in diseased plants. These results are similar to those where growth regulator activities have been measured for bleeding xylem sap from diseased vines. Hormone-like substances with auxin activity, produced by P. minimum, P. chlamydospora and especially by F. mediterranea, could contribute to host cell hyperplasia, hypertrophy, and leaf lamina expansion.
The most prominent features of plant responses to pathogens and, in general, against stresses, is the 'oxidative burst', i.e. the rapid increase in the cellular concentration of Reactive Oxygen Species (ROS) and mainly H 2 O 2 (De Gara et al., 2003;Torres et al., 2009). In the present study, leaves collected from healthy vines had the lowest H 2 O 2 concentrations during all the considered growth stages. In leaves of diseased plants, H 2 O 2 increased about 3-fold in symptomless leaves compared to healthy ones, regardless of growth stage, and reached greatest amounts in symptomatic plants. This evidence suggests strong correlation between the metabolic activity of pathogens and H 2 O 2 production and accumulation in host leaves. H 2 O 2 in leaves of diseased plants may act as an antimicrobial to counteract pathogens, or as a strengthener of cell wall polymers, a promoter of phytoalexin synthesis, or in triggering of programmed cell death. However, infected vines failed their defence upgrades and eventually suffered the effects of oxidative stress. H 2 O 2 increased oxidative stress and damaged integrity and functionality of cell membranes (Pérez et al., 2002), by lipid peroxidation of unsaturated fatty acids. The lipid peroxidation levels clearly showed changes in cell membranes. Levels of MDA, a product of lipid peroxidation, were correlated to membrane damage (Heath and Packer, 1968;Soares et al., 2019).
Plants produce ROS-scavenging mechanisms under biotic and abiotic stresses. Enzymes, including superoxide dismutase, catalase, peroxidase, ascorbate peroxidase, and glutathione reductase (Zhang and Kirkham, 1996;Lee and Lee, 2000), and non-enzymatic antioxidants such as tocopherols, ascorbic acid, and glutathione (Noctor and Foyer, 1998), functioned as ROS detoxifiers. Ascorbic acid is considered a key molecule for H 2 O 2 elimination. Ascorbic acid reacts with H 2 O 2 directly or by ascorbate peroxidase, a Class I heme-peroxidase that uses ascorbic acid as an electron donor and is the main peroxidase involved in H 2 O 2 detoxification (Asada, 1999). Monodehydroascorbate reductase, dehydroascorbate reductase and reduced glutathione regenerate ascorbic acid. Glutathione controls the redox state in plant cells under abiotic and biotic stresses, and this compound is involved in ascorbic acid regeneration through the Ascorbate-Glutathione cycle (Noctor and Foyer, 1998;Asada, 1999;Mittler, 2002;Hung et al., 2005). If the Ascorbate-Glutathione cycle operated well, ascorbic acid content and ascorbate peroxidase activity increased in host leaves of infected vines as expected (De Gara et al., 2003;Hung et al., 2005). However, leaves of diseased vines, during all the four growth assessed, showed ascorbic acid, and reduced glutathione concentrations that were less than in healthy vines. This trend was also confirmed for total ascorbate presence in bleeding xylem sap. In contrast, presence of pathogens in the grapevine trunks stimulated total glutathione contents.
The results from the present study allow development of the hypothesis that P. minimum and P. chlamydospora, or these fungi in association with F. mediterranea, can affect host antioxidant defences based on both glutathione and ascorbic acid. This change could be correlated with an unbalanced host oxidative state, damage to membrane integrity and appearance leaf necrosis symptoms.
Leaves of diseased vines showed significant decreases in redox state, and shift of ascorbic acid and glutathione towards oxidized forms. Ascorbate and glutathione redox states provide reliable estimation of cellular oxidative stress (Munné-Bosch and Alegre, 2003). In the present study, diseased vines were more stressed than healthy vines. To explain this change in physiological status, we suggest that the cause was the metabolic complex produced by F. mediterranea, P. minimum and P. chlamydospora Sparapano, 2006a, 2006b;. The accumulation of resveratrol, benzoic acid derivatives and flavonols as host defence compounds (Amalfitano et al., 2000;Jeandet et al., 2002;Bruno and Sparapano, 2006b;Calzarano et al., 2016Calzarano et al., , 2017aCalzarano et al., , 2017b were also suggested as causes. Phenols or flavonoids contribute to ascorbic acid oxidation in the scavenging of H 2 O 2 in grape leaves (Yamasaki et al., 1997), i.e., phenoxy or flavonoxy radicals accept electrons from ascorbic acid and produce the monodehydroascorbate radical (Pérez et al., 2002).
To prevent oxidative stress, following the glutathione-ascorbate metabolic pathway, ascorbate peroxidase reduced H 2 O 2 to water converting ascorbic acid to monodehydroascorbate that spontaneously disproportionate into ascorbic and dehydroascorbic acids. Using reduced glutathione, dehydroascorbate reductase reduced dehydroascorbic acid to ascorbate and produced oxidized glutathione. Finally, glutathione reductase reduced oxidized glutathione using NADPH as an electron donor (Asada, 1999). Under our conditions, the activities of enzymes regenerating ascorbic acid, also active in diseased as well as healthy vines, did not show any marked differences. This implied that ascorbate peroxidase, dehydroascorbate reductase, ascorbate free radical reductase, and glutathione reductase made a non-significant contribution to increasing ascorbic acid regeneration in diseased vines. Therefore, our data on ascorbate peroxidase, the key-enzymes of Ascorbate-Glutathione cycle, were in contrast with the low concentrations in cv Sultanina protoplasts (Papadakis et al., 2001) and the absence in 'Sultana' leaves (Pérez et al., 2002). This difference could be due to grape varieties, stress considered, plant materials and assay procedure applied.
In conclusion, the results of the present study suggest that F. mediterranea, P. minimum and P. chlamydospora interfere with several morphological, physiological, and biochemical functions in 'Italia' grapevines. Alterations affected bleeding xylem sap and leaves. Flux, dynamic viscosity, and growth regulator activity distinguished between bleeding xylem sap of vines infected with P. minimum and P. chlamydospora and those infected with P. minimum, P. chlamydospora and F. mediterranea. Leaf surface area, fresh, and dry weights, chlorophyll, hydrogen peroxide contents, lipid peroxidation, and redox states were altered in leaves of all assessed diseased vines. The presence of F. mediterranea in wood tissues of infected vines further debilitated the host physiological status. These alterations were detected in symptomatic leaves and, at low intensity, in symptomless leaves of diseased vines. These deleterious effects marked a presymptomatic stage, for irreversible changes inducing symptoms appearance. In diseased vines, low concentrations of ascorbic acid, reduced glutathione, and moderate levels of dehydroascorbic acid and oxidized glutathione were also associated with increased amounts of H 2 O 2 and MDA, and considerable oxidative stress. Under these conditions, scavenging enzymes were not able to sufficiently restore the balance between ROS and the antioxidants managing host stress conditions. The stresses caused by oxidative unbalance increased lipid peroxidation of unsaturated fatty acids of host membranes, damaged membrane integrity, and contributed to cell death and development of leaf symptoms. The present study has indicated that Ascorbate-Glutathione cycle is likely to be involved in grapevine susceptibility to fungi associated with the Esca complex.