Abscisic Acid and Chitosan Modulate Polyphenol Metabolism and Berry Qualities in the Domestic White-Colored Cultivar Savvatiano

During the last decade, several studies demonstrated the effect of biostimulants on the transcriptional and metabolic profile of grape berries, suggesting their application as a useful viticultural practice to improve grape and wine quality. Herein, we investigated the impact of two biostimulants—abscisic acid (0.04% w/v and 0.08% w/v) and chitosan (0.3% w/v and 0.6% w/v)—on the polyphenol metabolism of the Greek grapevine cultivar, Savvatiano, in order to determine the impact of biostimulants’ application in the concentration of phenolic compounds. The applications were performed at the veraison stage and the impact on yield, berry quality traits, metabolome and gene expression was examined at three phenological stages (veraison, middle veraison and harvest) during the 2019 and 2020 vintages. Results showed that anthocyanins increased during veraison after treatment with chitosan and abscisic acid. Additionally, stilbenoids were recorded in higher amount following the chitosan and abscisic acid treatments at harvest. Both of the abscisic acid and chitosan applications induced the expression of genes involved in stilbenoids and anthocyanin biosynthesis and resulted in increased accumulation, regardless of the vintage. Alterations in other phenylpropanoid gene expression profiles and phenolic compound concentrations were observed as well. Nevertheless, they were mostly restricted to the first vintage. Therefore, the application of abscisic acid and chitosan on the Greek cultivar Savvatiano showed promising results to induce stilbenoid metabolism and potentially increase grape defense and quality traits.


Introduction
Viticulture and winemaking are an indispensable part of the Greek national culture for over 4000 years. An area of more than 60,000 ha is covered with wine grapes, consisting of more than 90% domestic cultivars, namely Savvatiano, Assyrtiko (white-colored), Roditis (white-or pink-colored), Agiorgitiko and Xinomavro (red-colored). In particular, Savvatiano is the most widely planted domestic cultivar, covering an area of 10,370 ha and representing more than 16% of the Greek vineyard [1].

Climate Conditions
The experimental vineyard is located in a narrow valley at approximately 450 m altitude, in the area of Central Greece, 100 Km north-west of Athens ( Figure S1, Supplementary Materials). The Mediterranean climate of the Muses Valley is characterized by mild, wet winters and dry summers. Based on the meteorological data recorded at the nearby town of Askri, Viotia, monthly maximum temperatures were higher during the summer months (June-August) of the 2019 vintage (39.5 • C) compared to 2020 (37 • C) ( Figure 1A). In contrast, the total amount of rainfall in the same period was higher in 2020 (96.5 mm) compared to 2019 (37 mm) ( Figure 1A). Remarkably, during the first two weeks of September (i.e., the onset of the veraison to middle veraison stage) more days with a temperature of over 30 • C were recorded in 2020 than the same period of the 2019 vintage ( Figure 1B). As a consequence, the ripening period (i.e., veraison to harvest) of the 2020 vintage was shorter, due to the higher maximum temperatures.
performed in parallel to the berry quality traits, in order to elucidate the effects of abscisic acid and chitosan applications on the phenolic composition in the leading white-colored cultivar, Savvatiano. Further, we investigated the biostimulants' effect on the expression profile of genes encoding key enzymes in the phenylpropanoid pathway.

Climate Conditions
The experimental vineyard is located in a narrow valley at approximately 450 m altitude, in the area of Central Greece, 100 Km north-west of Athens ( Figure S1, Supplementary Materials). The Mediterranean climate of the Muses Valley is characterized by mild, wet winters and dry summers. Based on the meteorological data recorded at the nearby town of Askri, Viotia, monthly maximum temperatures were higher during the summer months (June-August)() of the 2019 vintage (39.5 °C) compared to 2020 (37 °C) ( Figure 1A). In contrast, the total amount of rainfall in the same period was higher in 2020 (96.5 mm) compared to 2019 (37 mm) ( Figure 1A). Remarkably, during the first two weeks of September (i.e., the onset of the veraison to middle veraison stage) more days with a temperature of over 30 °C were recorded in 2020 than the same period of the 2019 vintage ( Figure 1B). As a consequence, the ripening period (i.e., veraison to harvest) of the 2020 vintage was shorter, due to the higher maximum temperatures.

Berry Size
During the 2020 vintage, the berry size in both the control and biostimulant-treated vines increased significantly in all of the samples throughout ripening and at the harvest    (Table 1), compared to 2019. As mentioned above, the total amount of rainfall of July and August was higher in 2020 compared to the previous year ( Figure 1). During this period, the grape berries gather a significant amount of water which increases the berry weight, which could explain the differences observed between the two vintages. The results from the biostimulant treated vines showed that during the 2019 vintage the highest concentrations applied had a greater effect on grape berry weight. Specifically, the average weight of berries sprayed with the highest concentrations of ABA and CHT (2.08 g and 2.05 g at harvest, respectively) was significantly lower (p < 0.05) than the control (2.30 g) and berries treated with the low concentrations (2.55 g) at the harvest stage (Table 1). Significant differences between the treatments were observed in the berry size during the second vintage at the harvest, while during the ripening period no differences were observed. In contrast to the 2019 vintage, in the 2020 vintage the highest doses of ABA and CHT exhibited significantly increased berry weight than the control at the harvest stage (Table 1). The higher levels of weight/ berry in the 2020 compared to the 2019 vintage could be linked to the higher rainfall.

Total Soluble Solids
Among the most important berry quality traits is the grape juice composition, determined mainly by the total soluble solids (TSS; expressed in • Brix) and the titratable acidity (TA; expressed in tartaric acid g/L). The grape berries in both of the vintages exhibited a constant increase in the total soluble solids during maturation (Table 1). The TSS concentration in the control vines in almost all of the sampling dates of the 2019 vintage was lower Plants 2022, 11, 1648 5 of 21 than in the 2020 vintage, implying that the high temperatures observed during July and August of 2019 ( Figure 1A) could have a negative impact on the subsequent sugar accumulation (Table 1). Biostimulant application induced variations in the TSS concentration in both of the vintages. The highest concentrations of ABA and CHT resulted in higher • Brix at the harvest stage in the 2019 vintage than the untreated and lowest concentrations, while no significant differences were observed between the controls and the berries treated with the lower elicitor concentrations (Table 1). However, during the 2019 ripening period no differences were recorded between the treatments and the control berries. On the other hand, the ABA and CHT treatments during the 2020 vintage resulted in decreased TSS concentrations, compared to the control in all of the sampling dates (Table 1), indicating that the elicitors' application, together with the high temperatures of September 2020 (Figure 1), had a negative impact on TSS.

pH and Total Acidity
Statistically significant differences in the titratable acidity were observed between the controls and treated berries at the first two phenological stages of both of the vintages (Table 1), with the highest CHT dose providing a higher value of TA in the veraison and middle veraison sampling dates. In contrast, both the ABA and CHT treated berries exhibited a lower level of TA, compared to the control vines, at the harvest stage in 2019 (Table 1), while the TA level in both of the treatments was higher than the controls at the harvest stage in the following vintage (Table 1). Similarly to TA, differences in the pH levels were observed during the ripening period in both of the vintages, while the variations between the treated and control berries at the harvest stage concurred with the TA results. Specifically, the pH level was increased in the treated berries, with the highest dose of ABA or CHT, at the harvest of 2019, and decreased at the harvest of 2020, compared to the untreated ones (control) ( Table 1).

Metabolic Changes in Grape Berries
The grape berries were collected from the control and treated vines at three phenological stages (veraison, middle veraison and harvest) and a metabolic profiling was performed by UPLC-MS. A total of 30 metabolites were detected and identified by UPLC-MS in the berry samples ( Table 2), comprising of six amino acids, four phenolic acids, four stilbenoids, six flavonols, eight flavan-3-ols and two anthocyanins di-OH.
The PCA score plots of the berries at the middle veraison and harvest stages showed no discrimination between the control and treated samples, suggesting that ABA temporarily impacted the berry metabolism at the veraison stage, with no visible metabolic changes at the middle veraison and harvest stages ( Figure S3, Supplementary Materials). As well as for ABA treatments, the PCA score plots at middle veraison and harvest showed no discrimination between the control and CHT-treated berries ( Figure S4, Supplementary Materials).

Univariate Statistics
The application of the biostimulants did not affect the total amino acid level at any of the three stages during the vintage of 2019. However, the ABA and CHT applications resulted in significantly higher concentrations of total amino acids at the middle veraison stage of 2020 ( Figure 3A; Figures  bolic changes at the middle veraison and harvest stages ( Figure S3, Supplementary Materials). As well as for ABA treatments, the PCA score plots at middle veraison and harvest showed no discrimination between the control and CHT-treated berries ( Figure S4, Supplementary Materials).

Univariate Statistics
The application of the biostimulants did not affect the total amino acid level at any of the three stages during the vintage of 2019. However, the ABA and CHT applications resulted in significantly higher concentrations of total amino acids at the middle veraison stage of 2020 ( Figure 3A; Figures     The phenolic acids represent an important group of compounds, with a key role in the white varieties. In this study, four phenolic acids were detected by UPLC-MS, namely gallic acid, coutaric acid, caftaric acid and fertaric acid. A consistent decrease in the total phenolic acids' concentration in the grape berries was recorded over the ripening period, while the ABA and CHT treatments did not lead to any significant change ( Figure 3D; Figures S5D and S6D, Supplementary Materials). It is worth mentioning that the biostimulant applications resulted in a reduced total phenols level compared to the controls in the 2019 vintage, while the opposite effect was observed in the following year. However, these differences were not statistically significant ( Figure 3D Figure 3E). At the other two phenological stages, no differences between the treated and control vines were recorded in both of the vintages, except for the significantly lower concentration measured at the harvest of 2020 in those grapes treated with the higher dose of CHT ( Figures S5E and S6E, Supplementary Materials).
Another important group of phenolic compounds detected by the UPLC-MS was the stilbenoids, namely piceid, E-resveratrol, E-piceatannol and E-ε-viniferin. The concentration of stilbenoids in the control berries was increased during ripening, peaking at the middle veraison stage ( Figure 3F; Figures S5F and S6F, Supplementary Materials). Although no significant differences were observed between the treated and untreated vines at the veraison stage of 2019, the highest ABA treatment showed a significantly higher (i.e., two to three-fold) concentration of stilbenoids, compared to the controls and the other treatments, at the same phenological stage of 2020 ( Figure 3F). The effect of the ABA and CHT application was negligible at the middle veraison stage of both of the vintages ( Figure S5F, Supplementary Materials), while it was evident at the harvest stage. Remarkably, the lowest dose of the ABA treatment resulted in a significantly increased total stilbenoids level in both of the vintages, while the higher dose CHT treatment exhibited a similar effect in the 2019 vintage ( Figure S6F, Supplementary Materials).
Taken together, the UPLC-MS results indicate that most of the changes in the phenolic composition due to the different biostimulant applications were observed during the veraison stage in both of the vintages. In general, the ABA and CHT treatments resulted in higher stilbenoid and anthocyanin levels and, on the other hand, in a decreased flavan-3-ols concentration. However, the effect of the treatments on the phenolic compounds' concentrations was found to be mostly dependent on the dose and the vintage.

Gene Expression Analysis
Recent results have highlighted that biostimulants cause alterations in the grape berry transcriptome [13][14][15]49,50]. In the present study, the effect of biostimulant applications on the gene expression was examined by targeted RT-qPCR analysis of the berry samples collected at three different stages (veraison, middle veraison and harvest) during the 2019 and 2020 vintages.
We initially investigated the expression profile of the phenylalanine ammonia lyase (VviPAL) and cinnamate 4-hydrolase (VviC4H) genes that encode the first two enzymes of the phenylpropanoid pathway [51]. The expression analysis of the VviPAL gene in the control vines showed an upregulation after veraison, followed by a slight decrease at the harvest stage during the vintage of 2019. However, the transcript level remained constant throughout berry maturation in 2020 (Figure 4; Figure S7A The stilbene synthase gene (VviSTS), initiating the stilbenes' biosynthesis [52], exhibited a dramatical upregulation throughout ripening, mainly in the first vintage. During the following year, the increase was also evident, but was observed only at the harvest stage ( Figure 4; Figure S7C, Supplementary Materials). All of the biostimulant treatments in the 2019 vintage resulted in increased VviSTS transcript accumulation at the first phenological stages. Furthermore, the application of lower doses of ABA and CHT had the same effect on the expression level in the 2020 vintage ( Figure 4; Figure S7C, Supplementary Materials).
We further examined the expression patterns of the genes belonging to the biosynthetic pathway of flavonoids. The expression of the flavonol synthase gene (VviFLS) that encodes for the enzyme catalyzing the flavonol biosynthesis [53] showed a declining trend during berry maturation (Figure 4; Figure S8A, Supplementary Materials). The same trend-and also a lower expression level-was observed in most of the biostimulant treatments, except for the lower doses of ABA and CHT that led to VviFLS upregulation at the veraison stage of 2019 and 2020, respectively ( Figure 4; Figure S8A, Supplementary Materials). The expression of the UDP-glucose-flavonoid 3-O-glycosyltransferase gene (VviUFGT), encoding for the critical step in anthocyanin biosynthesis [54], was constant throughout maturation in the 2019 vintage, while it was gradually increased at the harvest stage in the following year (Figure 4; Figure S8B The stilbene synthase gene (VviSTS), initiating the stilbenes' biosynthesis [52], exhibited a dramatical upregulation throughout ripening, mainly in the first vintage. During the following year, the increase was also evident, but was observed only at the har-

Discussion
Muses Valley is characterized by extinguishing topographical and meteorological features that have favored grapevine cultivation since the ancient times. Although the grapevine is considered to be a drought-resilient species traditionally covering semiarid areas, the environmental factors, such as the climatic variability, influence grape and wine quality. The different weather conditions between the two years at the location of the vineyard influenced the maturation process; therefore, the vintage effect was evident in the grape berries' physiochemical characteristics during the harvest stage. For instance, the higher amount of rainfall during the summer months of 2020 resulted in an increased berry size compared to the previous vintage. Furthermore, differences in berry weight were observed between the treatments, as well as the treated vines that received the highest doses producing significantly larger grape berries. On the other hand, the treatment effect on the berry size was evident under the moderate weather conditions of 2019; hence, the highest dose of biostimulants decreased the berry weight. Vintage-dependent variations were also observed in the grape juice components. However, whilst biostimulant application was found to have a limited effect on the berry sugar content in other studies [38,56], we observed that the highest doses of ABA and CHT resulted in an increased TSS level in 2019, while both of the doses of biostimulants caused a decreased TSS level at the harvest of 2020. On the contrary, berries from the treated vines exhibited a lower TA level than those from the untreated vines at the harvest stage of 2019, while the opposite effect was observed in the following vintage.
It is well-established that biostimulants can increase the deposit of phenolic compounds [29][30][31][32]. However, the outcome of the application is dependent on the environmental conditions [45]. Herein, we showed that the ABA and CHT treatments increased the stilbenoid and anthocyanin concentrations, regardless of the vintage. Although anthocyanins are known to determine the red color of grapes and wines, recent studies demonstrated the existence of small amounts in the white grape berries [57] and wines [58]. The white color of grapes was initially explained by mutations in the VviMYBA1 and VviMYBA2 genes, which regulate the VviUFGT expression [59]. However, the existence of several other MYB-type transcription factors that can modulate flavonoid biosynthesis [60], and the identification of VviUFGT in transcriptomic studies in white-colored cultivars [51,61], imply more complex regulatory mechanisms.
In contrast to the effect of the stilbenoids and anthocyanins, the positive effect on the other phenolic compounds (i.e., flavonols) was restricted to the first vintage. The increased berry size, observed during the second vintage, possibly led to the dilution of the phenolic compounds, so that the differences between the treatments could not be distinguished. Furthermore, the increased daily maximum temperatures observed during the ripening period of 2020 could have a negative impact on the phenolic compounds. It is known that high temperatures decrease the accumulation of the anthocyanin, phenolic and volatile compounds, and it has been suggested that specific secondary metabolites (e.g., kaempferol and flavonols in general) could be indicators of fruit quality losses associated with warming [62]. The beneficial effect of exogenously applied biostimulants on grape composition can be achieved during moderate climate conditions [63], while the effect is limited under a hot and dry climate [56]-similar to the local weather conditions observed at the experimental vineyard during the ripening period of the second vintage.
Although Savvatiano is a white grape variety, this is the first time that small amounts of anthocyanins were detected in all of the treatments. Although still controversial, it is believed that the presence of these anthocyanins in white wines could provoke the pinking phenomenon [64].
Principal component analyses showed that the biostimulant applications had a strong impact on the grape berry composition at veraison, while no discrimination between the control and treated samples was observed in the following phenological stages. Veraison is the crucial stage in grape berry development. The initiation of berry softening and sugar accumulation, together with the transcriptional reprogramming and alterations in secondary metabolites' biosynthesis, will lead to the exquisite characteristics of any distinct grapevine cultivar. During recent years, several studies have shown that the critical time period for biostimulant application is the week before or after veraison [12,15,36,40].
Considering that the grape phenolic composition depends on the transcriptional changes induced by environmental conditions, it was challenging to complement our study with an expression analysis of the target genes encoding the key enzymes of the phenylpropanoid biosynthetic pathway. Recently, accumulated results from transcriptomic studies revealed the differential expression of the phenylpropanoid pathway genes, due to biostimulant applications [12][13][14][15]49]. Herein, we showed that the expression level of VviPAL and VviC4H was nearly constant during the grape berry maturation in untreated plants, while the VviSTS exhibited a significant increase throughout ripening and the VviFLS expression was found to decline towards harvest. An increase was also observed in the VviUFGT gene expression at the harvest stage of 2020. The ABA and CHT applications resulted in increased VviPAL and VviUFGT expression levels, respectively, while both of the treatments resulted in an early upregulation of the VviSTS gene expression. However, in certain cases the treatment effect on the gene expression was also dependent on the vintage. For instance, the VviC4H and VviFLS and expression level were increased in the treated plants compared to the controls at the veraison stage of the 2019 vintage, while no difference was observed the following year.
Remarkably, the VviUFGT and VviSTS expression was positively affected by the biostimulants' application in both of the vintages, which concurs with the increased concentration of anthocyanins and stilbenoids. Our results indicate that the early upregulation of these two genes at the veraison stage led to higher levels of the corresponding compounds at harvest, contributing to an improved grape berry quality. The positive effect of ABA and CHT on anthocyanin biosynthesis and accumulation has been reported in many studies [13,15,35,38,65] and similar results have been obtained with stilbenoids. More specifically, the treatment of grape berries with ABA [66] and CHT [67], as well as the elicitation of cell suspension cultures [68][69][70], resulted in the upregulation of stilbene synthase genes, accumulation of the encoded enzymes and a significant increase in the stilbenoid content [71]. Considering the important role of stilbenoids in plants' defense mechanisms, and their numerous benefits to human health, these findings can be useful in future applications.

Experimental Design
The experiment was conducted during two growing seasons (2019 and 2020) in a commercial vineyard in Muses Valley (Askri, Viotia; 38 • 19 30 N, 23 • 05 37 E, at an elevation of 450 m) in Central Greece ( Figure S1, Supplementary Materials), planted with the Vitis vinifera L. Savvatiano. The vines were more than 50 years old and pruned as bush vines. The vineyard was located on a deep loamy soil and was managed according to standard agronomical practices of the region, without irrigation. The number and timing of the viticultural practices (i.e., plant protective applications) were similar for all of the treatments. The experiment was conducted in a randomized block design, with all of the treatments applied in three replicates, using 10 vines for each replication ( Figure S2, Supplementary Materials). More than three grapevine plants in each row between the different treatments were considered as the buffer parts of the vineyard and were not sampled. The vines were sprayed with an aqueous solutions of 0.04% w/v (low dose) or 0.08% w/v (high dose) abscisic acid (s-abscisic acid 10.4% w/v, Protone SL, Hellafarm, Peania, Greece; ABA treatment) and 0.3% w/v (low dose) or 0.6% w/v (high dose) chitosan (chitosan hydrochloride 3% w/w, Project One, Phytorgan S.A., Kifisia, Greece; CHT treatment). Aquascope (Hellafarm, Greece) and Tween 80 (Sigma-Aldrich, St. Louis, MI, USA) were used as wetting agents for the ABA and CHT treatments, respectively. None of the sprayed vines served as the control (control treatment). For the ABA treatment, spraying was performed at the grape zone (clusters only) at the veraison stage (81-85, according to the BBCH scale) [72], and a second and a third application were performed 3 and 6 days after the first application (Table S1, Supplementary Materials). The CHT applications were carried out on the whole vine canopy (leaves and clusters) at the veraison stage with a second and a third application performed 7 and 14 days after the first application (Table S1, Supplementary Materials).

Sampling and Physicochemical Determination in Grapes
The grapes were harvested on 29 September 2019 and 24 September 2020 (Table S1, Supplementary Materials). In order to follow the ripening process of the grape berries and define the optimum harvest date, sampling was carried out once every week. At all of the sampling points, 50 berries from each experimental treatment were randomly collected and the berries' fresh weight was determined. The grape maturity level was monitored weekly by measuring the Total Soluble Solids content ( • Brix), Titratable Acidity and pH, according to the official methods from the Compendium of International Methods of Wine and Must Analysis [73]. The grape berries were collected at three maturity level points: at veraison, just before the first treatment application; at middle veraison and at the optimum pulp maturity stage. A total of 50 berries of each replicate were collected randomly. The samples were transported on dry ice and were carefully stored at −80 • C until required for analysis.

RNA Extraction and Gene Expression Analysis by RT-qPCR
The grape berries without seeds were ground to powder with liquid nitrogen, and the RNA was extracted by the method of Reid et al. [85]. Briefly, approximately 1 g of ground tissue was extracted with a buffer containing 300 mM Tris-HCl (pH 8.0), 25 mM EDTA, 2 M NaCl, 2% (w/v) CTAB, 2% (w/v) PVPP and 0.05% (w/v) spermine at 65 • C for 15 min, mixed thoroughly with an equal volume of chloroform: isoamyl alcohol (24:1) and centrifuged. The step was repeated for aquatic phase, the RNA was precipitated with 0.6 volumes isopropanol and 0.1 volumes sodium acetate at −20 • C overnight, centrifuged and finally dissolved in 100 µL ddH2O. The RNA samples were treated with DNAse I (Takara Bio, Shiga, Japan) and further purified using phenol: chloroform: isoamyl alcohol (25:24:1), followed by ethanol precipitation. The RNA quantity and quality were determined using a NanoDrop ND-1000 Spectrophotometer (Thermo Fisher Scientific Inc., Wilmington, DE, USA) and verified by 0.8% agarose gel electrophoresis. The reverse transcription was performed with 2 µg RNA using SMART MMLV-Reverse Transcriptase (Takara Bio, Shiga, Japan) and oligo (dT) primer (Eurofins Genomics, Ebersberg, Germany). The synthesized cDNA was five-fold diluted and PCR conditions were optimized for primers corresponding to the selected genes from various metabolic pathways (listed in Table 3). The samples were further diluted and quantitative PCR reactions were performed in the PikoReal Real-Time PCR System (Thermo Fisher Scientific, Vantaa, Finland) using the KAPA SYBR FAST qPCR Master Mix (KAPA Biosystems, Cape Town, South Africa) and applying the following cycler conditions: 2 min at 50 • C, 2 min at 95 • C, followed by 40 cycles of 15 s at 95 • C, 30 s at 62 • C, 30 s at 72 • C. All of the quantitative PCR reactions were performed as triplicates and the melting curve analysis was performed at the end of each reaction to confirm primer specificity. The quantification of gene expression was performed according to the 2 −∆Ct method and elongation factor 1a (VviEF1a) was used as the reference gene for data normalization.

Statistical Analysis
All of the values are presented as the mean standard deviation. The statistical analyses were performed using Statgraphics Centurion application (version 1.0.1.C)(Virginia, USA). The significance of the results was determined with an unpaired t-test or one-way ANOVA with Tukey's test. A multivariate statistical data analysis (MVA) of the samples was performed with SIMCA P+ version 15 (Umetrics AB, Umeå, Sweden), after mean centering all of the variables and scaling unit-variance. The metabolic variables affected by ABA and CHT treatments were revealed through the principal component analysis (PCA), applied as the unsupervised MVA method. The heatmaps were created using the Perseus software (version 1.5.3.2) and the 2 −∆Ct values from gene expression analysis.

Meteorological Data
The climatic data were pooled from the National Observatory of Athens' Automatic Network [87].

Conclusions
Biostimulant application is a useful viticultural practice to improve the grape and wine quality, especially in the challenging era of climate change. However, the optimization of the application time and the concentration of the compound applied are critical parameters for the outcome of the approach. In the present study, abscisic acid and chitosan were found to induce the expression of genes involved in anthocyanins and stilbenoids' biosynthesis, and to enhance their accumulation in a white-color cultivar under the Greek vineyard conditions. Further alterations in the other phenylpropanoid gene expression profiles and the phenolic compound concentrations were also observed, but they were mostly dependent on the vintage. More extensive and in-depth research studies would be necessary in order to elucidate the grapevine responses to various biostimulants and their specific impact on each cultivar metabolic profile. The acquired knowledge, combined with the grape response mechanisms to the environmental conditions during ripening, will be beneficial to viticulturists and winemakers in order to better exploit the distinguishing quality characteristics of a certain cultivar.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants11131648/s1, Table S1: Date of biostimulant applications during two growing seasons, Figure S1: Map of Central Greece showing the location of the experimental vineyard; Figure S2: Single vineyard in the Muses Valley and the Experimental Design; Figure S3: Unsupervised classification using PCA on metabolomic data from grape berries treated with ABA; Figure S4: Unsupervised classification using PCA on metabolomic data from grape berries treated with CHT; Figure  Funding: This research has been co-financed by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH-CREATE-INNOVATE (project code: T1EDK-04200 MU-SA). This work was supported by the Region-Centre Val de Loire (France) grants to CEPATLAS, PRESTO and INNOCOSM programs, as well as by the French Embassy in Athens-Greece with the program "Programme de bourses pour séjour scientifique de haut niveau (SSHN) en France 2020-2021".
Institutional Review Board Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors (pending privacy and ethical considerations).