Overview of Kaolin Outcomes from Vine to Wine: Cerceal White Variety Case Study

: Kaolin protective e ﬀ ect was assessed in a white grapevine cultivar ‘Cerceal’ in ‘Alentejo’ Region (southeast Portugal) where plants face extreme conditions during the summer season. We addressed the hypothesis that kaolin e ﬀ ects lead to several changes in leaves, fruits, and wine characteristics on the primary and secondary metabolism. Results showed that kaolin reduces leaf temperature which provokes an improvement in physiological parameters such as net photosynthesis and water use e ﬃ ciency. This protection interferes with berry color, leaving them more yellowish, and an increase in phenolic compounds were observed in all fruit tissues (skin, seed, and pulp). Additionally, both berry and wine characteristics were strongly a ﬀ ected, with an increase of tartaric and malic acid and consequently high total acidity, while the sugar concentration decreased 8.9% in berries provoking a low wine alcohol level. Results also showed that kaolin induces high potassium, magnesium, and iron, and low copper and aluminum concentrations. Moreover, the control wine showed higher content of esters related with hostile notes whereas wine from kaolin treated vines presented higher content of esters associated with fruity notes. Overall, the results strengthen the promising nature of kaolin application as a summer stress mitigation strategy protecting grapevine plants and improving fruit quality and creating more balanced wines.


Introduction
Viticulture and winemaking promote economic, social, and environmental benefits, through trademark, rural income, employment, and tourism [1]. Nonetheless, currently the greatest challenge for the wine industry is climate change, essentially under Mediterranean conditions [2], due to projected shifts in precipitation and temperature. Among the Portugal's wine regions, the Alentejo Demarcated Region (southeast) stands out due to the more pronounced harsh climate conditions and where water scarcity is a major problem. The reoccurrence of combined environmental stresses poses a risk to the crop yield and quality. Extreme temperature (>35 • C) through the growing season, as occurs in grape Portuguese areas, can harshly damage leaf photosynthetic efficiency and berry metabolism [3]. Water deficit affects berry quality in a developmental manner [4]. Grape leaf and

Plant Material and Experimental Design
Samples were obtained from Cerceal, a white Vitis vinifera L., variety grafted on 1103P rootstock, located in a commercial vineyard "Herdade do Esporão" (38°22′48.1″ N, 07°33′38.4″ W), in southeast Portugal. The climate is of the Mediterranean-like type with dry and hot summers, moderate precipitation during the winter months, and dryness during the summer [25]. Records from a meteorological station, located 10 km away from the experimental vineyard, were collected. Vines were managed using the organic production mode. The white Cerceal cv. has a row with 200 m long (200 plants) in which two different conditions were set up: an experimental control (C; 100 plants) and another pulverized with kaolin (100 plants). The vines were 7 years old and were unilaterally cordon trained and pruned.

Foliar Leaf Temperature
Leaf temperature was measured with an infrared thermometer (Infratrace, model KM800S, Comark Ltd., Hertfordhire, UK) with a 15° field view, at veraison and maturation in the midday period. Measurements were performed under clear sunny days and on sun-exposed and fully expanded leaves at the middle of the shoots (usually between 8th and 11th nodes on the shoot axes). The average temperature of three randomly selected leaves (in eight plants) per treatment (3 × 8 = 24) was obtained by holding the thermometer at about 1 m above the foliar surface.

Physiological Parameters
Leaf gas exchange, chlorophyll a fluorescence, and OJIP test (open-source chlorophyll fluorometer based on the Kautsky induction curve) were obtained in both 2016 and 2017 years during

Plant Material and Experimental Design
Samples were obtained from Cerceal, a white Vitis vinifera L., variety grafted on 1103P rootstock, located in a commercial vineyard "Herdade do Esporão" (38 • 22 48.1" N, 07 • 33 38.4" W), in southeast Portugal. The climate is of the Mediterranean-like type with dry and hot summers, moderate precipitation during the winter months, and dryness during the summer [25]. Records from a meteorological station, located 10 km away from the experimental vineyard, were collected. Vines were managed using the organic production mode. The white Cerceal cv. has a row with 200 m long (200 plants) in which two different conditions were set up: an experimental control (C; 100 plants) and another pulverized with kaolin (100 plants). The vines were 7 years old and were unilaterally cordon trained and pruned.

Foliar Leaf Temperature
Leaf temperature was measured with an infrared thermometer (Infratrace, model KM800S, Comark Ltd., Hertfordhire, UK) with a 15 • field view, at veraison and maturation in the midday period. Measurements were performed under clear sunny days and on sun-exposed and fully expanded leaves at the middle of the shoots (usually between 8th and 11th nodes on the shoot axes). The average temperature of three randomly selected leaves (in eight plants) per treatment (3 × 8 = 24) was obtained by holding the thermometer at about 1 m above the foliar surface.

Physiological Parameters
Leaf gas exchange, chlorophyll a fluorescence, and OJIP test (open-source chlorophyll fluorometer based on the Kautsky induction curve) were obtained in both 2016 and 2017 years during the summer season (at veraison and maturation). In these times several field measurements were done.  (14:00-15:30). Net photosynthesis (P N ), stomatal conductance (g s ), transpiration rate (E), and intercellular CO 2 concentration/ambient CO 2 ratio (C i /C a ) were estimated according to von Caemmerer and Farquhar (1981). To eliminate the possible effects of air humidity and temperature on transpiration, the P N /g s ratio, rather than the P N /E ratio, was calculated to evaluate the intrinsic water-use efficiency (iWUE) [26].

Chlorophyll a Fluorescence Analysis and OJIP Test
Chlorophyll a fluorescence emission was measured at morning (09:00-10:30) and midday (14:00-15:30) on fully expanded leaves in both developmental stages with a Pulse Amplitude Modulation Fluorometer (mini-PAM, Photosynthesis Yield Analyzer; Walz, Effeltrich, Germany), using two scripts: (i) in the first script, the measurements were done on well sun exposed leaves. In this procedure, after a 35 s exposure to actinic light (1450 µmol m −2 s −1 ), light-adapted steady-state fluorescence yield (F s ) was averaged, followed by exposure to saturating pulse light (6000 µmol m −2 s −1 ) for 0.6 s to establish F m '. The sample was then shaded for 5 s with a far-red light source to determine F 0 . (ii) In a second script, using the dark leaf clip (DLC-8), the same leaf portion used in first script was immediately dark acclimated for 30-45 min. After this, the maximum photochemical efficiency of PSII was given by where F 0 corresponds to the minimum fluorescence level excited by very low intensity of measuring light to keep PSII reaction centers open, and F m corresponds to the maximum fluorescence level elicited by a pulse of saturating light (6000 µmol m −2 s −1 ) which closes all PSII reaction centers. From these measurements, several fluorescence attributes were calculated [27,28]: photochemical quenching (qP = (F m ' − F s )/(F m ' − F 0 ')), non-photochemical quenching (NPQ = (F m − F m ')/F m '), and efficiency of electron transport as a measure of the quantum effective efficiency of PSII (Φ PSII = ∆F/F m ' = (F m ' − F s )/F m '). The photosynthetic electron transport rate was estimated as ETR (µmol m −2 s −1 ) = (∆F/F m ') × PPFD × 0.5 × 0.84, where PPFD is the photosynthetic photon flux density incident on the leaf, 0.5 is the factor that assumes equal distribution of energy between the two photosystems, and the leaf absorbance used was 0.84 because it is the most common value for C 3 plants [27]. The analysis of the fluorescence transients using the JIP test was performed according to our previous study [20].

Pests Control
To study the kaolin pesticide effect against Panonychus ulmi (Koch, 1836) and Scaphoideus titanus (Ball, 1932), 10 leaves of the upper third of canopy in 10 contiguous plants were observed and the percentage of leaves with symptoms were registered at veraison stage in 2017. The incidence was expressed as the percentage (%) of affected leaves.

Fruit Surface Color Index and Biometric Parameters
Berry color was determined by a spectrophotometer Konica Minolta Sensing's CM-2500c portable (Minolta Corp., Osaka, Japan) in 30 berries per treatment (three measurements were made around the equatorial belt of each berry) in the maturation stage, in 2017. This color system, evaluated lightness, L* (0, black-100, white), chroma, C*ab (0, achromatic), and hue angle, h ab (0, red-90, yellow-180, green-270, blue) as previously described [29]. Fruit biometric parameters were measured (mm) and weighed (mg) individually in situ in triplicate (n = 30 per treatment and in each stage). The width and height were measured to calculate the average of grape berry radius in order to estimate the absolute volume (mm 3 ) of each one. Both color and fruit biometric parameters were obtained in the maturation stage in 2017.

Phenolic Compounds and Antioxidant Activity
The total phenolic contents from whole fruit in both veraison and maturation stages and from skin, pulp, and seed of the fruit in maturation stage were determined by the Folin-Ciocalteu method [30] during 2017. All samples were lyophilized and macerated with liquid nitrogen and MeOH 70% was used for the phenolic compound extraction. Briefly, the extract was added, 20 µL sample (4 mg/mL) or gallic acid standards in MeOH, 90 µL distilled H 2 O, and 10 µL of Folin-Ciocalteau reagent solution. After 6 min, 80 µL of 7% Na 2 CO 3 was added and mix gently. The reaction mixture was kept in dark for 2 h and its absorbance was measured at 750 nm in microplate. Total phenolics was expressed as mg gallic acid equivalents per gram of extract (mg g −1 DW) (y = 1.076x + 0.059, R 2 = 0.99987).
The aluminum chloride (AlCl 3 ) complex method at 510 nm was used for the quantification of the total flavonoids content of extracts [31] and was expressed as mg of catechin equivalents per gram of extract (mg CAE g −1 DW) (y = 2.7814x + 0.0477, R 2 = 0.9993). The ortho-diphenols content was estimated according to the colorimetric method based on a complex reaction with sodium molybdate dehydrate at 370 nm [32]. The results were expressed as mg of gallic acid equivalents per gram of extract (mg GAE g −1 DW) (y = 3.173x + 0.166, R 2 = 0.9983). The condensed tannins contents were determined according to the vanillin-HCl assay [33] at 500 nm. The results were expressed as mg of catechin equivalents per gram of extract (mg CAE g −1 DW) (y = 1.874x + 0.0624, R 2 = 0.99708).
The radical scavenging activity on ABTS radical was evaluated by the method of Trolox equivalent antioxidant capacity assay at 734 nm [31], and was done in the same samples used for total phenols. ABTS were expressed as mg of trolox equivalents per gram of extract (mg TE g −1 DW) (y = −0.3004x + 0.1968, R 2 = 0.99969). The radical scavenging activity on the DPPH radical was evaluated [31] being previously adapted to microplates [34]. DPPH were expressed as mg of trolox equivalents per gram of extract (mg TE g −1 DW) (y = −0.1442x + 0.1039, R 2 = 0.99904).

Total Soluble Proteins
The total soluble proteins were obtained from whole fruit in both veraison and maturation stages during 2017 and were extracted using an extraction buffer containing phosphate (Fisher Scientific, Loughborough, UK) of pH 7.5 mixed with EDTA (ethylenediaminetetraacetic acid) (Panreac, Barcelona, Spain). The work solution included the extraction buffer described above, PMSF (Phenylmethanesulfonyl fluoride) (Panreac, Barcelona, Spain), PVP (Polyvinylpyrrolidone) (Sigma, St. Louis, MO, USA) according to the method of Bradford (1976) [35] at 595 nm. All the absorbances of this work were determined using PowerWave XS2 microplate scanning spectrophotometer (Bio-Tek Instruments, Winooski, VT, USA). Total soluble proteins were expressed as mg bovine serum albumin equivalents per gram of extract (mg BSAE g −1 DW) (y = 14.029x + 0.6234, R 2 = 0.981).

Brix and Acidity Parameters
• Brix were measured (n = 30 per treatment), at maturation in 2017, using an ATAGO digital refractometer (CO., LTD., Tokyo, Japan). The Brix scale or degrees • Brix is numerically equal to the percent of sugar and other dissolved solids in the solution [36].
The physico-chemical parameters of grapes, such as pH and total acidity, and wine, such as alcohol degree and total acidity, were analyzed according to the OIV [37] methodologies. The tartaric and malic acid were measured enzymatically (Miura One, TDI S.A.).

Trace Elements Quantification
Some element quantifications were obtained in 2017 in the grape berries at maturation and in the wine must, during fermentation process according to [38]. In the wine samples obtained from vines under different treatments (C and Kl) only the Al (aluminum) was quantified. Prior to the analyses each sample was vigorously shaken. An aliquot (0.5 g) of sample was weighed directly into the digestion vessels. The digestion was performed by adding HNO 3 (1.0 mL) and H 2 O 2 (5.0 mL) to each sample. The mixture was left at room temperature with a marble (preventing evaporation) for 24 h, and afterwards the marbles were removed, and the samples left overnight at room temperature. After this period, the sample was heated using a block heater at 50 • C during 1 h followed by 100 • C during 1 h (temperature at which the release of Nitric Oxide brown fumes starts), 120 • C during 1 h, and finally left overnight at 155 • C (usual time needed to obtain a clear digestion mixture), or until the solution was clear, with a glass marble on the top of the culture tube (to avoid drying before digestion and sample charring). After this period the glass marbles were removed, and the contents were dried at 155 • C. After cooling to room temperature, 10.0 mL of HNO3 matrix solution (1.5 mL of acid to 1000 mL of water) was added to the digested samples and stirred. Some of the solutions were diluted in order to allow the determination of the respective metals. All samples were analyzed in triplicate. The copper (Cu) and potassium (K) were determined by flame atomic emission spectrometry and calcium (Ca), magnesium (Mg), iron (Fe), and zinc (Zn) metals were analyzed by flame atomic absorption spectrometry (Thermo Scientific ICE 3000). The Al were analyzed by atomic absorption spectrometry in graphite furnace (Unicam 939 AA spectrometer, GF90 furnace). Each run of samples was preceded by calibration using aqueous mixed standards prepared in HNO 3 (1.0 M). For this purpose, five different dilutions of standards were used, besides the blank, with the range of concentrations being selected according to the expected concentrations of the elements of interest.

Analysis of Wine Volatile Compounds by HS-SPME-GC-MS
Solid phase microextraction (SPME) was used for the extraction of volatile compounds from grapevine berries at maturation and wine from 2017 samples. One milliliter of sample was measured to a 20 mL headspace vial (La-Pha-Pack ® , Langerwehe, Germany) and was capped with a white PTFE silicone septum (Specanalitica). The SPME operating conditions were extraction temperature 40 • C for 40 min, rotating speed 100 rpm, and desorption time 10 min at 250 • C. Analysis were carried out in a GCMS-QP2010 Plus (Shimadzu ® , Kyoto, Japan) equipped with an AOC-5000 autosampler (Shimadzu ® ). A divinylbenzene/carboxen/polydimethylsiloxane (DVB/Car/PDMS) fiber (SUPELCO Analytical, Bellefonte, PA, USA) was used for headspace SPME sampling. For the analysis a capillary column Sapiens-5-MS (Teknokroma), 30 m, 0.25 mm (IS), 0.25 µm (film thickness) was used. The working conditions were as follows: injector temperature: 250 • C, injection mode: splitless during 1.5 min, detector temperature: 250 • C. High-purity helium (≥99.999%) was used as the carrier gas, column oven temperature was kept at 40 • C for 5 min, increased to 170 • C at a rate of 5 • C min −1 , 230 • C at 30 • C min −1 , and maintained for 4 min, then was raised to 300 • C at 30 • C min −1 and maintained for 2 min; carrier gas (He) with a flow of 2.00 mL min −1 . In MS interface temperature was 250 • C and ion source temperature was 250 • C. Mass spectra were acquired in electron ionization (EI) mode at 70 eV. in a m/z range between 29 and 300 with a scan speed of 555 scans s −1 . The compounds were identified using the mass spectra libraries, NIST 21, 27, 107, 147, and Wiley 229.
Firstly, the peak areas data of all compounds were extracted from the chromatograms and used to build the full data matrix from 'Cerceal' wines consisting of six observations (two treatments of wine samples, each one by three replicates) and 51 variables (volatile components).

Statistical Analysis
Statistical analyses were performed with SPSS 20.0 software. After testing for ANOVA assumptions (homogeneity of variances with the Levene's mean test, and normality with the Kolmogorov-Smirnov test), statistical differences among stages and treatments were evaluated by two-way factorial ANOVA, followed by the post hoc Tukey's test and in some cases only within the stage one-way ANOVA was done. Significant differences were considered for p < 0.05 and *** p < 0.001, ** p < 0.01, * p < 0.05 represent significant differences. Absence of superscript indicates no significant difference between treatments. Values are presented as mean ± standard deviation (SD). Regarding volatile compounds a hierarchical cluster analysis (HCA) combined with the heatmap visualization was applied for the dataset using the MetaboAnalyst 3.0 (web software, The Metabolomics Innovation Centre (TMIC), Edmonton, AB, Canada). The area of each variable was auto scaled. The significance of the compounds detected in samples were compared between control and kaolin treated samples, through a two-sided Mann-Whitney test (using the SPSS software 20.0 (IBM, New York, NY, USA).

Physiological and Pests Control Changes under Kaolin Application
Kaolin particle film showed several physiological effects. Regarding Figure 2, leaf temperature, which can be considered the first bound between weather conditions and plant health, was positively affected by Kl application in veraison and maturation in both years.
Agronomy 2020, 10, x FOR PEER REVIEW 7 of 25 Metabolomics Innovation Centre (TMIC), Edmonton, AB, Canada). The area of each variable was auto scaled. The significance of the compounds detected in samples were compared between control and kaolin treated samples, through a two-sided Mann-Whitney test (using the SPSS software 20.0 (IBM, New York, NY, USA).

Physiological and Pests Control Changes under Kaolin Application
Kaolin particle film showed several physiological effects. Regarding Figure 2, leaf temperature, which can be considered the first bound between weather conditions and plant health, was positively affected by Kl application in veraison and maturation in both years. Values are presented as mean ± SD (n = 24 per treatment). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same stage of the season, and * represents significant differences between stages within the same treatment (p < 0.05). Absence of superscript indicates no significant differences.
The kaolin treated leaves showed decreases of 8.2% and 9.3% and 6.4% and 10.4% in leaf temperature in veraison and maturation stages of 2016 and 2017, respectively. Regarding leaf gas exchange parameters (Table 1)   Values are presented as mean ± SD (n = 24 per treatment). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same stage of the season, and * represents significant differences between stages within the same treatment (p < 0.05). Absence of superscript indicates no significant differences.
The kaolin treated leaves showed decreases of 8.2% and 9.3% and 6.4% and 10.4% in leaf temperature in veraison and maturation stages of 2016 and 2017, respectively. Regarding leaf gas exchange parameters (Table 1) the results showed low values of stomatal conductance (g s ), net photosynthesis (P N ), and intrinsic water use efficiency (iWUE) in the maturation stage in both years and periods of the day (morning and midday), comparing with the veraison stage. The Kl treatment leads to a significantly high P N and iWUE and low C i /C a ratio in both stages and periods in 2016 and 2017.
Relating to g s , the results only showed differences in the morning period in 2017 with high values in the Kl-treated plants. The transpiration rate (E) only presents a significant high value in the Kl-treated plants in the maturation stage in the morning period in 2016. Table 2 presents parameters related to chlorophyll a fluorescence data. Regarding 2016, only differences were observed in the midday period, except for the ETR whose values were higher in the kaolin treated plants. Overall, kaolin application boosts F v /F m , Φ PSII , qP, and ETR values, while leads to a decrease in the F 0 and NPQ values. Table 1. Gas exchange parameters, namely transpiration rate (E, mmol m −2 s −1 ), stomatal conductance (g s , mmol m −2 s −1 ), photosynthesis net (P N, µmol m −2 s −1 ), intrinsic water use efficiency (iWUE, µmol mol −1 ), and ratio of intercellular to atmospheric CO 2 concentration (C i /C a ), at morning and midday periods in veraison and maturation stages in kaolin and control leaves (n = 10). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same period of the day (morning/midday) and stage of the season (veraison/maturation). * represents significant differences between stages of the season within the same period of the day (p < 0.05). Absence of superscript indicates no significant differences.  Table 2. Chlorophyll a fluorescence parameters, namely basal fluorescence (F 0 ), maximum quantum efficiency of photosystem II (F v /F m ), effective PSII efficiency (Φ PSII ), photochemical quenching (qP), electron transport rate (ETR, µmol e − m −2 s −1 ), and non-photochemical quenching (NPQ) at morning and midday periods in veraison and maturation stages in kaolin and control leaves (n = 10). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same period of the day (morning/midday) within the stage of the season (veraison/maturation), and * represents significant differences between stages of the season within the same period of the day (p < 0.05). Absence of superscript indicates no significant differences.  Figure 3 shows the higher values of Kl treated plants in parameters related to the OJIP test, specifically the efficiency of energy conservation in the electron transport (Ψ 0 ) and performance index (PI ABS ) in both years and stages in morning and midday periods.

Stage
in the Kl-treated plants. The transpiration rate (E) only presents a significant high value in the Kltreated plants in the maturation stage in the morning period in 2016. Table 2 presents parameters related to chlorophyll a fluorescence data. Regarding 2016, only differences were observed in the midday period, except for the ETR whose values were higher in the kaolin treated plants. Overall, kaolin application boosts Fv/Fm, ΦPSII, q P, and ETR values, while leads to a decrease in the F0 and NPQ values. Figure 3 shows the higher values of Kl treated plants in parameters related to the OJIP test, specifically the efficiency of energy conservation in the electron transport (Ψ0) and performance index (PIABS) in both years and stages in morning and midday periods. Values are presented as mean ± SD (n = 10). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same period of the day (morning/midday) within the stage of the season (veraison/maturation), and * represents significant differences between stages of the season within the same period of the day (p < 0.05). Absence of superscript indicates no significant differences.
The results revealed that the incidence of Panonychus ulmi was reduced by 10% in leaves sprayed with Surround WP ® (Kl) compared to control fruit (C-20%; Kl-10%) ( Figure S1). Furthermore, Scaphoideus titanus incidence was also significantly reduced by 30%, after Kl sprays (C-30%; Kl-0%). Values are presented as mean ± SD (n = 10). Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same period of the day (morning/midday) within the stage of the season (veraison/maturation), and * represents significant differences between stages of the season within the same period of the day (p < 0.05). Absence of superscript indicates no significant differences.

Kaolin Effects on Fruit Quality
In white grapes varieties it is predictable that skin berry color changes during grape maturation, from an initial green to a yellow/brownish at maturation. The Kl effects on the color skin parameters in the maturation stage were evaluated and results are shown in Table 3. Accordingly, Kl leads to a decrease in a* and b* parameters, i.e., a rise of green and a little decline of yellow color, giving thus more yellowish coloring to the skin compared to the brownish coloring to the skin in control fruits. Additionally, a decrease was observed in the chroma values (C*ab) and no differences were obtained in the lightness (L*). Relevant differences existed in hue (h ab ), being higher in fruits from Kl treated plants. Furthermore, C*ab is positively correlated with b*, independently from treatment, showing correlation coefficients (r) of 1.00 and 0.988, for Kl and control treatments, respectively (all Pearson's correlations were significant at p < 0.05, data not shown). Usually, hue (h ab ) showed an inverse correlation with a*. Our results are in agreement with this tendency with r = −0.992 for Kl and −0.991 for control treatments (all correlations were significant at p < 0.05). Table 3. Kaolin effect on colorimetric parameters analyzed for all grape skin color variants and • Brix of fruits picked in the maturation period, and total acidity (g L −1 of tartaric acid), pH, tartaric and malic acid concentration (g L −1 ) of fruits picked in the veraison and maturation stages, in 2017. Values are presented as mean ± SD. Significant differences were presented between treatments and considered for p < 0.05 and *** p < 0.001, ** p < 0.01, * p < 0.05 represent significant differences; ns indicates no significant differences.

Kaolin
Control Significance Kaolin application shows influence in the • Brix and acidity parameters (Table 3). A decrease of 8.9% was observed in the • Brix of fruits from Kl treated plants. As expectable, a decrease in the total acidity as well as in the tartaric and malic acids concentration from veraison to maturation stage were observed. However, Kl berries presented 16.6% and 8.2% higher total acidity in veraison and maturation stage, respectively. Regarding pH in Kl fruits, the results showed an increase between veraison and maturation stage, the treated fruits being 4.3% and 4.7% more acidic than control ones, respectively. Both tartaric and malic acids decreased from veraison to maturation stage. In the veraison stage Kl fruits presented 20.6% more tartaric acid and 22.3% less malic acid concentration. However, in the maturation stage the tartaric acid concentrations remained 16.2% higher in Kl fruits. The malic acid concentration decreased 24.9% in Kl berries compared to the fruits from the control plants.

Veraison
Results showed that although fruit height was not affected by foliar Kl application, neither differences were observed between stages, the diameter was modified, also leading a different fruit volume (Table 4) in the veraison stage. Therefore, the fruit diameter and volume in this stage were significantly higher in kaolin-treated plants, these fruits being approximately 8.6% wider and with 14.3% more volume. In the maturation stage no differences were observed.
In Table 4 some fruit quality parameters related with secondary metabolism and primary metabolism (protein content) are presented. Generally, the secondary metabolism decreased from veraison to maturation stage and the protein content increased. The Kl application does not have an effect on the total phenols and tannins concentration in the veraison stage (Table 4). However, the flavonoids, ortho-diphenols, and protein content as well as DPPH presented lower values than control fruits and ABTS showed a higher value in the same stage. From veraison to maturation the Kl effect changes the tendency. At maturation stage the fruits from Kl treated plants showed higher total phenols (+11.6%), ortho-diphenols (+1.83%), tannins (+6.45%), and protein content (+9.42%). Regarding antioxidant activity obtained by ABTS and DPPH methods the Kl fruits also presented higher values (+5.36% and 12.7%, respectively) in this stage (Table 4). Deepening the study, Table 5 shows the same secondary metabolism parameters and the antioxidant activity in different berry tissues, namely seed, skin, and pulp. Independent of the treatment, our results showed, as expected, higher content of phenols, flavonoids, and tannins and also higher ABTS values in the seeds. However, the skin presented the higher content of ortho-diphenols as well as higher DPPH value (antioxidant activity). Kl treated plants presented berries with higher content of flavonoids (+10.3%), ortho-diphenols (+32.5%), tannins (+27.3%), and DPPH value (+25.8%) in the seeds; higher content of total phenols (+13.1%), flavonoids (+25.5%), ortho-diphenols (+29.4%), tannins (+7.46%), and ABTS value (+11.0%) in the skin; higher content of total phenols (+39.7%), flavonoids (+41.1%), ortho-diphenols (+13.8%), and tannins (+14.4%) in the pulp. Table 4. Kaolin application effect on height (mm), diameter (mm), and volume (mm 3 ) of berries, and phenols (mg g −1 DW), flavonoids (mg g −1 DW), ortho-diphenols (mg g −1 DW), tannins (mg g −1 DW), protein (mg g −1 DW), and antioxidant activity (mg g −1 DW; obtained by ABTS and DPPH methods) of whole fruits at the veraison and at maturation stages in 2017. Values are presented as mean ± SD. Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same stage of the season (veraison/maturation) and * represents significant differences between stages of the season within the same treatment (p < 0.05). Absence of superscript indicates no significant differences. Noticeably, the mineral analysis also showed that mature berries and must from vines treated with kaolin had a significantly lower quantity of aluminum (Al) and copper (Cu) and high quantity of potassium (K), iron (Fe), magnesium (Mg), and zinc (Zn) than berries from the control vines (Table 6).

Kaolin Application Consequences in White 'Cerceal' Wine
As Kl is an aluminum silicate, we quantified the aluminum concentration in the wine (Table 7). Accordingly, the Al content is significantly lower (−12.9%) in wine of Kl treated plants than in the control one. The alcohol degree was also lower in Kl wine (3.57%) and an increase in total acidity, malic and tartaric acid were obtained (+16.3%, 11.1%, and 7.08% in Kl samples, respectively).
We also evaluated the wine volatile compounds ( Table 8). The heatmap ( Figure 4) shows a graphical representation of the chromatographic data (presented in Table 8) achieved for the 51 volatile components, allowing a rapid visual evaluation of the wine's volatile profiles. The chromatic scale of the heatmap allows access the relative amount of each volatile component (from dark blue, minimum, to dark red, maximum). Whereas the dendrogram (Figure 4) built from the HCA is an exploratory tool that reveals two clusters corresponding to the two types of wines, i.e., control and kaolin. From the 51 volatile components detected, 53% (corresponding to 27 components) exhibited differences statistically significant between both types of wines (differences corresponding to p < 0.05). Table 5. Phenols, flavonoids, ortho-diphenols and tannins, and antioxidant activity (ABTS and DPPH) (mg g −1 DW) in the seeds, skins, and pulps of fruits from Kl treated and untreated (control) plants.
Values are presented as mean ± SD. Significant differences were presented between treatments and considered for p < 0.05 and *** p < 0.001, ** p < 0.01, * p < 0.05 represent significant differences; ns indicates no significant differences.  Table 7. Aluminum concentration (μg g −1 ), total acidity (g L −1 of tartaric acid), pH, alcohol degree (%, v/v), and tartaric and malic acid concentration (g L −1 ) of wine of white 'Cerceal' cv. from Kl treated and untreated (control) grapevines. Values are presented as mean ± SD. Significant differences are presented between treatments and considered for p < 0.05 and *** p < 0.001, ** p < 0.01, * p < 0.05 represent significant differences. We also evaluated the wine volatile compounds ( Table 8). The heatmap ( Figure 4) shows a graphical representation of the chromatographic data (presented in Table 8) achieved for the 51 volatile components, allowing a rapid visual evaluation of the wine's volatile profiles. The chromatic scale of the heatmap allows access the relative amount of each volatile component (from dark blue, minimum, to dark red, maximum). Whereas the dendrogram (Figure 4) built from the HCA is an exploratory tool that reveals two clusters corresponding to the two types of wines, i.e., control and kaolin. From the 51 volatile components detected, 53% (corresponding to 27 components) exhibited differences statistically significant between both types of wines (differences corresponding to p < 0.05). wines under study: control and kaolin treated grapevines, which reveals the distinction among wines. The content of each compound was illustrated through different colors (from dark blue, minimum, to dark red, maximum). Dendrogram for the HCA results using Ward's cluster algorithm to the data set was also included. Differences corresponding to p < 0.05 were considered significant and were marked with *. wines under study: control and kaolin treated grapevines, which reveals the distinction among wines. The content of each compound was illustrated through different colors (from dark blue, minimum, to dark red, maximum). Dendrogram for the HCA results using Ward's cluster algorithm to the data set was also included. Differences corresponding to p < 0.05 were considered significant and were marked with *. Table 6. Mineral composition of white grape berries and must in different stages of 'Cerceal' cv. from kaolin treated and untreated (control) plants. Values are presented as mean ± SD. Different lowercase letters represent significant differences between treatments (control vs. kaolin), in the same stage of the season (veraison/maturation) and * represents significant differences between stages of the season within the same treatment (p < 0.05). Absence of superscript indicates no significant differences. N.D. means that concentration was so low that it was not detected.

Kaolin Modulates Grapevine Plant Physiology and Pest Control
The kaolin particle film was initially developed for suppression of arthropod pests and diseases by its repellent effect [39,40]. Our results confirm that Kl spraying reduces the incidence of pests in grapevines (Supplementary data), reinforcing their repellent benefit as reported in other crops e.g., apple, pear, and olive trees [40][41][42]. Increased plant productivity results from insect control with Kl has also been documented [42,43]. This work, as well as a previous one done by our group in red grape cultivars in the Douro valley (NE Portugal) [44] reinforce that the treatment of grapevine leaves with the inert clay mineral kaolin increases also physiological capacity of plants [44]. The clay particles protect leaves from excessive radiation leading to a lower leaf temperature ( Figure 2) and thus better iWUE, extremely linked to the lower ABA accumulation [14], reducing potentially damage triggered by visible and ultraviolet radiation and therefore decreases heat stress and sunburn injury [45]. Grapevine Kl plants under better temperature and irradiation conditions showed higher P N values (Table 1). In control leaves nonstomatal limitations to photosynthesis were evident, as revealed by the decrease of P N and iWUE and the increase of C i /C a ratio (Table 1), either through CO 2 diffusion and carboxylation efficiency [46], and/or photochemical perturbations. At this level, the decrease in P N under saturating light conditions was associated with a reduction of Φ PSII , F v /F m , and ETR (Table 2). Additionally, linking the decrease of F v /F m with the increase of F 0 , these results suggest that the photoprotective capacity of these leaves was surpassed and photoinhibitory damage in the PSII occurred [47], as seen previously in plant reaction to high temperature and water stress [48]. The Kl photoprotection induction is also supported by the higher qP and lower NPQ values, which suggest an effective radiative and nonradiative dissipation of the excess energy, avoiding the photosystem damage by oxidation [49].

Kaolin Boosts Grapevine Fruit Quality
It is known that during grape berry development a complex series of physicochemical modifications, such as changes in size, color, chemical composition, and flavor occurs [50]. Temperature influences both cell division and enlargement and in excess, such as in our study (>35 • C), reduce growth rate and size. However, solar radiation is also crucial for berry growth [51], which could explain that Kl treatment induces 11.2% lower increase in berry diameter than control ones, from veraison to maturation stages, supposedly due to more shaded fruits provoked by a high number of healthy leaves that exist because of the leaf Kl protection from sunburn. White berries from V. vinifera varieties are the consequence of the inactivation of anthocyanin biosynthesis, thus that grape green-yellowish color is mainly linked to catabolic pathways instead of specific pigment accumulation [52]. During the maturation process a degradation of carotenoids and chlorophyll pigments occurs [53] being, according C*ab coordinates, the yellowish color of the fruits related with the b* value. In spite of the high b* value of control fruits, both Kl and control berries have a positive b* (Table 3) and, consequently, yellowish color. However, the negative a* value of the Kl fruits (greener) compared with the positive one (more reddish) obtained for control berries, reveals that Kl fruits are more yellowish compared with the brownish-yellow of control ones. This difference could be related with the low chlorophylls/carotenoids ratio, and the ripening characteristic oxidative burst, promoting the appearance of the yellowish color of the fruits [53]. During the berry ripening stage, pH should increase, mainly related with the decline of tartaric and malic acid [54]. These acids decrease and were observed in berries of both treatments, resulting in a decrease of overall total acidity (Table 3). Tartaric acid concentration decreases from veraison to maturation stage, contrary to other studies reporting that their content remains relatively constant in the grape berry and is not related to climatic conditions [55]. Inversely, malic acid concentration in grapes depends of several factors, such as climate conditions, especially irradiance [55] and temperature [56], as the most important ones. The positive protection of Kl treatment leads to berries with higher tartaric and malic acid concentration (Table 3). Probably, this is due to the sun protection of fruits by the healthy leaves and their shade effects on lower acid degradation. It is known that higher level of organic acids (particularly tartaric acid) is a positive characteristic of grapevine varieties in warm climates, such in this studied region and others most threatened by climate change [57]. For this reason, Kl treatment showed potential to produce well balanced wines avoiding the intensive need of acidification of must/wine. As referred before, organic acids (malic and tartaric acids) and sugars concentration showed opposite behavior [54] as we observed in our study. In fact, Kl fruits have more tartaric and malic acid and lower sugar concentration than control berries (Table 3). These results in white grapevine fruits, as previous ones in red berries [10,17,21], reinforce that Kl treatment boosts, mainly in the mature grape berry (maturation period), the quantities of phenolic compounds, including total phenolics and tannins, leading also to an augment of antioxidant activity (Table 4). This happened because in response to kaolin there is a global stimulation of phenylpropanoid and flavonoid-flavonol pathways at the gene expression and/or protein activity levels [21]. This fact should have major implications in fruit and wine quality, while protecting plant against abiotic stress. An analysis focused on secondary metabolism in the different fruit tissues showed as predictable that the major quantity of phenolics are found in the seed, after in the skin, and only a little percentage in the pulp [58]. However, we observed that Kl provokes an increase of total phenols, flavonoids, tannins, and ortho-diphenols compared with the control (Table 5). Amongst polyphenols, ortho-diphenols are known as the most significant in relation to their antioxidant activity which is related with hydrogen donation, i.e., their capacity to improve radical stability by forming an intra-molecular hydrogen bond between the hydrogen of their hydroxyl group and their phenoxyl radicals [59]. Grape berry mineral content is also important and involved in wine chemical composition. Among the several minerals present in grape berries, potassium (K + ) usually represents the most abundant cation because is accumulated during the entire period (pre-and post-veraison) with a huge rise at the beginning of ripening [60]. The Kl use leads to an enhance of K + . Looking from a technological point of view, this effect is beneficial because K + influences the pH of musts and wines and thereby their chemical and microbiological stability, in addition to the perception of wine flavor [61]. Taking into account that the factors that affect K + accumulation include their soil availability and weather conditions effects, this study reinforces the Kl protective capacity against summer stress, since the pedoclimatic conditions were similar for all plants under study. This high accumulation is reflected in the must when the Kl application also boosts their content (Table 6). Magnesium is considered to be a phloem-mobile element and its amount increased during ripening but at different rates depending on the response of berry to vine water relations [62]. This finding allows us to highlight the Kl capacity on the WUE avoiding water stress, which enabled the increase the quantity of Mg ++ in Kl fruits and must. With magnesium, other mineral elements such as calcium and copper also play a role in osmotic balance. In spite of copper being an essential micronutrient for all living organisms, including humans [63], elevated copper concentrations in grapevine can cause oxidative spoilage leading to browning of white wine as well as haze formation [64], which will hardly happen in the Kl must when the copper quantity is much lower than the control one. Of the several questions that winemakers have about Kl application, the most frequent one refers to aluminum. As Kl is an aluminum silicate (Al 2 Si 2 O 5 (OH) 4 ) there was a fear that high quantities of this metal would be found in grapes and wines whose plants were sprayed with it. However, these results confirm that the powdered aluminum was not absorbed into the fruits, nor to the musts (Table 6), which is reflected in the wine (Tables 7 and 8), and even its concentrations are even lower than those of the control vines. One possible explanation is that the kaolin that falls to the soil may change the pH of the soil, making it less acidic, leading to less solubility of Al in the soil and consequently less absorption of Al by the vines.

Positive Effect of Kaolin Application in White Wine Balance
One of the main challenges for the wine industry is climate change, because it has an enormous effect on vine phenology and physiology and consequently on yield and grape composition. Among climate change-related consequences, the advanced maturation times and temperatures are the most significant because these lead to an increase in grape sugar concentrations, which provokes high wine alcohol and lower acidities, especially in warm regions [65]. White varieties with highest level of tartaric and malic acid are suggested to be of great interest for breeding new cultivars. These are the principal organic acids in wine representing 70-90% of total grape acidity [66]. Their concentration is one of the most key quality characteristics of grapes for wine production, and has an essential impact on the color, flavor, and stability of wine [67]. Relating to white wines from warm regions, malic acid can even have a positive effect on wine balance. In such cases malolactic fermentation may result in significant changes of the wine aromatic profile and in increase of lactic and buttery characteristics and a decrease in fruity characteristics [68]. Looking to our results, Kl pulverization should be also used in white cultivars in areas with severe summer conditions due their capacity to trigger high total acidity (higher malic and tartaric acid) and lower alcohol degree (Table 7) which are characteristics well appreciated by the new wine consumers. After analysis of volatile compounds, it is possible to see that, in general, the control wine exhibited higher content of esters with C 11 to C 18 and decanoic acid, which if present in an amount higher than their aroma threshold, may contribute with unpleasant notes, such as wax, soap, or fatty [69,70]. On the other hand, in general, wine from kaolin treated grapevines presented higher content of esters with <C 12 , which are associated with fruity notes [69]. Additionally, control wine presents higher content of vitispirane, a C 13 -norisoprenoid. As observed for several fruits, including grape berries, during ripening diverse reactions modulated by enzymes may occur, namely degradation processes, such as carotenoids cleavage resulting in formation of norisoprenoids [71,72]. Hydroxylated C 13 -norisoprenoids often occur in plants as glycosides and can be liberated from these by enzymatic or acid hydrolysis and then transformed into aroma compounds, such as vitispirane. The statistically significant differences observed between both types of wines in the content of vitispirane may infer the impact of kaolin treatment of grapevines on the carotenoids cleavage. Vitispirane is associated with camphor and eucalyptus notes [73], but as its odor threshold is relatively high (800 µg/kg, wine) [68]. Further research is needed, namely sensorial analysis assays, to properly evaluate the impact of the observed statistically significant differences related with wine volatile profiles on the aroma characteristics of control wine and wine from kaolin treated grapevines.

Conclusions
We showed that kaolin application triggered an improvement in plant physiology, especially under conditions of abiotic stress, and can also be considered as an alternative to synthetic pest control. Moreover, kaolin application significantly influences the grape fruit metabolome in a way that provides berries with high phenolic compounds, tartaric and malic acids, total acidity, and lower sugar content. Besides, it is essential to reinforce that a good influence was observed in wine having higher acidity and lower alcohol levels and seems to have improved the aroma. In sum, foliar kaolin application in grapevine leaves shows great potential as a summer stress mitigation strategy because it clearly impacts on berry and wine quality as a result of many molecular and biochemical changes in key primary/secondary metabolic pathways.