Plant material and open field treatments

The experiment was carried out in 2012 on three varieties of Vitis vinifera (Monastrell, Syrah and Merlot) grafted onto 1103 Paulsen and planted in 2002 in Bullas (Murcia, south-east Spain). The experimental plot consisted of vines trained to a vertically shoot positioned (VSP) spur pruned cordon, with six 2-bud spurs (12 nodes). The distance between rows was 3 m with 1.25 m between vines. The rows were oriented in an N-NW to S-SE direction. Plants were drip irrigated.

All treatments were applied to three replicates. A completely randomized experimental design was set up consisting of three replicates of 10 vines for each treatment. The field trial involved the application of two elicitors: BTH ([benzo-(1,2,3)-thiadiazole-7-carbothioic acid S-methyl ester], Sigma Aldrich, St. Louis, USA) at a concentration of 0.3 mM; and MeJ (methyl jasmonate; Sigma Aldrich, St. Louis, USA) at a concentration of 10 mM. The protocol used to apply the different treatments was described previously (Gómez-Plaza et al., 2017). The concentrations were chosen according to the literature (Iriti et al., 2004; Vezzulli et al., 2007). Samples were harvested at optimum maturity. For analysis, five mature clusters per plant were randomly collected at harvest from treated and untreated grapevines of each treatment and replicate and immediately transported to the laboratory and frozen at -20 °C until analysis.

Vinifications

All the vinifications were made in triplicate in 50-L stainless steel tanks using 45 kg of grapes. Before alcoholic fermentation (AF) started, total acidity was corrected to 5.5 g/L and selected yeasts were added (10 g of dry yeast/100 kg of grapes, Laffort, DSM, Servian, France). The fermentative pomace contact period was 10 days. All the vinifications were conducted at 25±1 °C. Throughout the fermentation pomace contact period, the cap was punched down twice a day and the temperature and must density were recorded. At the end of this period, the wines were pressed at 1.5 bar in a 75-L tank membrane press. Free-run and press wines were combined and stored at room temperature. Wine oenological parameters were analyzed in triplicate at the end of AF.

Physicochemical determination in grapes at harvest

    Grape analysis involved the traditional flesh measurements. Total soluble solids (°Brix) were measured using a digital refractometer (Atago RX-5000). Titratable acidity was measured using an automatic titrator (Metrohm, Herisau, Switzerland) with 0.1 N NaOH. The methodology used to carry out these analyses is described in OIV (2014).

Analysis of anthocyanins and flavonols in grapes and wines

Grapes were peeled with a scalpel, and the skins were stored at -20 °C until analysis. Samples (2 g) were immersed in methanol (40 mL) in hermetically closed tubes and placed on a stirring plate at 150 rpm and 25 °C. After 2 h, the methanolic extracts were filtered (0.45 µm) and analyzed by high-performance liquid chromatography (HPLC). Samples of wines were filtered through the 0.45-µm nylon filters. The samples were directly analyzed by HPLC according to Bautista-Ortín et al. (2005).

Waters Empower Pro software was used to process the data obtained. Compounds were identified by comparison of the UV spectra recorded with the diode array detector and those reported in the literature. In addition, HPLC–MS analysis was made to confirm the identity of each peak. An LC-MSD-Trap VL-01036 liquid chromatograph-ion trap mass detector (Agilent Technologies, Santa Clara, CA) equipped with electrospray ionization (ESI) was used. Elution was performed in the HPLC analysis conditions described by Bautista-Ortín et al. (2005). The heated capillary and voltage were maintained at 350 °C and 4 kV, respectively. Mass scans were measured from 100 to 800 m/z. Mass spectrometry data were acquired in the negative ionization mode and processed using Data Analysis 2.1 LC/MSD Trap software (Agilent). For quantification, DAD chromatograms were extracted at 520 nm (anthocyanins) using malvidin 3-O-glucoside chloride (Extrasynthèse, Genay, France) as external standard, and at 360 nm (flavonols) using quercetin (Sigma Aldrich) as external standard.

Analysis of tannins in grapes and wines

The seeds and skins of 10 berries were separated from the mesocarp and rinsed with distilled deionized water. Seeds and skins were extracted separately in hermetically closed falcon tubes with 10 mL 2:1 acetone/water at room temperature for 24 h on an orbital shaker at 200 rpm. The extraction was carried out in the dark. The extract was concentrated in a CentriVap concentrator (Labconco, United States) to remove acetone and re-dissolved in 2 mL methanol in a volumetric flask. Skin and seed proanthocyanidins were determined and analyzed according to a previously described method (Kennedy and Jones, 2001; Busse-Valverde et al., 2010).

For wines, the samples were prepared by an optimization of the method described by Pastor del Río and Kennedy (2006). Wine (5 mL) was evaporated in a CentriVap concentrator (Labconco, United States), re-dissolved in 3 mL of water, and then passed through a C18-SPE column (1g Waters), previously activated with 10 mL of methanol followed by 10 mL of water. The cartridge was washed with 20 mL of water and the tannins were eluted with 10 mL of methanol, evaporated in a CentriVap and re-dissolved in 1 mL of methanol.

Proanthocyanidin cleavage products were estimated using their response factors relative to (+)-catechin (Sigma Aldrich, Spain), which was used as the quantitative standard. The results of these analyses enabled the total proanthocyanidin content, the apparent mean degree of polymerization (mDP), and the percentage of each constitutive unit to be obtained.

Sensory analysis

    Wines (elaborated in 2012) were subjected to a triangular sensory test, for which ten members associated with the project were trained over several weeks. For this analysis, three samples were presented, two of which were identical. Samples were presented in random order in coded, clear 125 mL official glasses. Each taster selected the sample that he/she considered different (forced election method) and the sample that he/she preferred.

Statistical data treatments

Significant differences among grapes and wines and for each variable were assessed by analysis of variance (ANOVA) using Statgraphics 5.0 Plus. The Duncan test was used to separate the means (p<0.05) when the ANOVA test was significant.

Physicochemical analysis of grapes at harvest

The effects of the different treatments on the physicochemical characteristics of the control and treated grapes were determined (Table 1). In general, treated grapes were very similar and the effect of the treatments was very small. Differences in berry weight were only observed in Syrah grapes, which were heavier after both treatments. In the other two varieties, the treatments had no effect on grape size. Previous studies (Ruiz-García et al., 2013b) found that the application of BTH, MeJ, BTH+MeJ and ABA (abscisic acid) to Monastrell grapes had no effect on berry weight.

The BTH-treated grapes from Merlot and the MeJ-treated grapes from Monastrell had higher sugar content than the corresponding control grapes, and higher total acidity values were observed in treated Monastrell grapes and MeJ-treated Syrah grapes when compared to control grapes. By contrast, Villango et al. (2015) measured lower acidity in Syrah grapes treated with a product obtained from yeast cell wall than in control grapes.

Anthocyanins in grapes and wines

    Anthocyanins are natural colorants present in the skin of red grapes, and they play a key role in the organoleptic characteristics of wines (Lesschaeve and Noble, 2005). The biosynthesis of these compounds can be influenced by exogenous elicitors such as ABA, BTH, and BTH/MeJ mixture (Ruiz-García et al., 2012; Ruiz-García et al., 2013a; Ruiz-García et al., 2013b).

    The anthocyanin content of grapes and wines for the Syrah, Merlot and Monastrell varieties used in this research is shown in Table 2. The influence of the two treatments studied differed in grapes and wines as a function of the variety, although for all three varieties, anthocyanin concentrations were always higher in grapes than in wines.

    For the Syrah variety, the highest concentrations of anthocyanins were found in BTH-treated grapes and in BTH and MeJ wines. In contrast, Portu et al. (2015) found that the foliar application of MeJ in Tempranillo induced anthocyanin synthesis in grapevines, increasing the concentration of several individual anthocyanins as well as the total content in both grapes and wines. More specifically, Table 2 also shows the results obtained for the individual anthocyanins in grapes and wines for both treatments applied to Syrah. Compared to the control samples, both treatments increased total 3-monoglucosides in grapes (by up to 27% for BTH and 22% for MeJ) and in wines (by up to 10% for BTH and 8% for MeJ) except cyanidin-3-glucoside in BTH-treated grapes and peonidin-3-glucoside in MeJ wines. In the case of acetylated compounds, both BTH and MeJ increased the petunidin-3-acetyl-glucoside content in grapes, while only MeJ increased the cyanidin-3-acetyl-glucoside content, and in wines, only BTH increased the concentration of cyanidin-3-acetyl-glucoside and peonidin-3-acetyl-glucoside. As regards the total content of acetylated compounds, an increase of 22% was observed in grapes when plants were sprayed with MeJ. Coumarates (cyanidin-3-coumaryl-glucoside and peonidin-3-coumaryl-glucoside) only increased in wines following BTH treatment but the total coumarate content remained unaltered.

    The results of the anthocyanin analysis in Merlot grapes and wines are shown in the second part of Table 2. In this variety, the highest concentrations of anthocyanins were found in MeJ-treated grapes (up to 14% increase) followed by BTH-treated grapes (up to 8%), and only a slight increase was observed in BTH wines (up to 3%). Previous studies demonstrated an increase in the anthocyanin content of Merlot grapes after BTH application, accompanied by increased resistance to Botrytis attack (Iriti et al., 2005).

    As can be seen, both treatments (MeJ and BTH) increased most individual anthocyanins in grapes (glucosides, acetates and coumarates) except malvidin-3-acetate and coumarate. In MeJ-treated Tempranillo grapes, Portu et al. (2016) reported an increase in the content of delphinidin-, cyanidin-, petunidin- and peonidin-3-glucoside, besides peonidin-3-acetyl-glucoside and cyanidin-3-coumaryl-glucoside. In the wines of our experiment, glucoside concentrations increased when grapes were treated with MeJ, acetates increased with both treatments, and coumarates increased when grapes were treated with BTH.

    Finally, for the Monastrell variety, the highest concentrations of anthocyanins were found in MeJ-treated grapes (up to 25% increase) and in MeJ and BTH wines (up to 51% and 10% increase, respectively). The influence of variety on the effect of both treatments was evident. Ruiz-García et al. (2013a) previously reported that the impact of MeJ was clone-dependent. In another study carried out over two years (2009 and 2010), the same authors reported an increase in skin anthocyanin content in preharvest BTH- and MeJ-treated Monastrell grapes (Ruiz-García et al., 2013a). The results obtained pointed to higher total monomeric anthocyanin values in the 2009 wines than in the corresponding 2010 wines. No differences due to treatments were observed in 2009, whereas the anthocyanin concentration was higher in the wines made from MeJ-treated grapes in 2010 (Ruiz-García et al., 2013a). Finally, Ruiz-García et al. (2013b) also observed a statistically significant increase in the anthocyanin concentration of Monastrell grapes and wines when BTH+MeJ was applied.

The results for individual anthocyanins in Monastrell grapes and wines are shown in the third part of Table 2. As can be seen, almost all individual anthocyanins increased in grapes as a result of both treatments, although the increase in total anthocyanins was only significant in MeJ-treated grapes. Many authors have described the effect of MeJ treatment on anthocyanin accumulation in plants: e.g. strawberry fruits (Pérez et al., 1997), Vitis vinifera (Zhang et al., 2002), and tulip leaves (Saniewski et al., 2003). Our results showed that MeJ had a stronger effect in grapes than BTH, since tri-OH forms (delphinidin, petunidin and malvidin) did not increase when grapes were treated with BTH. However, in wines, both treatments produced higher levels of glucosides and acetylated anthocyanins compared to the control samples, but not higher levels of coumarate anthocyanins (except delphinidin and peonidin in BTH wines and malvidin and peonidin in MeJ wines).

Flavonols in grapes and wines

The application of BTH or MeJ has previously been shown to be an interesting option for increasing the levels of grape phenolic compounds (Portu et al., 2015). Flavonols are very close to anthocyanins in the biosynthetic pathway - indeed, they share most of the pathway - so that any increase in enzyme activity upstream of the flavonoid biosynthetic pathway may also affect the concentration of these compounds (Ruiz-García et al., 2012). The results obtained for grape and wine flavonols for the three varieties studied are summarized in Table 3. Neither myricetin-3-glucoside nor quercetin-3-glucoside was detected in grapes but both were detected in wines. The results obtained for Syrah are shown in Table 3 (first part). As can be seen, statistically significant differences were found for total flavonols in BTH-treated grapes (increases of up to 36%) but not in the corresponding wines. As regards individual flavonols, the differences in the Syrah variety were generally more significant in grapes than in wines. In BTH-treated grapes, statistical differences from control levels were found for all the compounds analyzed, except kaempferol-3-galactoside, while in MeJ-treated grapes statistically significant differences from the control sample were found in quercetin-3-glucuronide, kaempferol-3-galactoside, kaempferol-3-glucoside and syringetin-3-glucoside. De la Peña-Moreno et al. (2010) reported an increase in the levels of quercetin and kaempferol in red raspberry fruits exposed to MeJ postharvest. By contrast, in wines, only slight differences between treatments were observed when compared to control samples: statistically significant differences were found in quercetin-3-glucoside in both treatments and in kaempferol-3-galactoside in BTH wines.

The results for flavonols in the Merlot variety are shown in Table 3 (second part). As can be seen, there were significant differences in grape total flavonol content between the control and the treatments, control grapes obtaining higher values for this parameter. However, statistically significant differences from the control existed in the wines elaborated with BTH-treated grapes. In agreement with our results, Ruiz-García et al. (2012) found that the concentration of flavonols in wine made from Monastrell showed only small differences due to grape treatments (MeJ and BTH), the differences only being significant for the wines made from BTH-treated grapes in one year (2009). Here, none of the treatments increased the level of total or individual flavonols in grapes at harvest, but statistical differences from control levels were found in total flavonols in BTH wines and in some of the individual compounds in both treatments. In BTH wines, all compounds except kaempferols (3-galactoside and 3-glucoside) and isorhamnetin increased in concentration compared with control wines, while in MeJ wines increases were only observed in the case of myricetin-3-glucoside, quercetin-3-glucoside and quercetin-3-glucuronide.

    The results for flavonols in the Monastrell variety are shown in Table 3 (third part). As can be seen, there were no significant differences in total flavonol content between the control and the treated grapes, but statistically significant differences from control values were found for the wines made from MeJ-treated grapes. As regards individual flavonols, few differences could be observed in grapes at harvest, with only myricetin-3-glucoside showing a higher concentration in MeJ-treated grapes than in control grapes, and isorhamnetin-3-glucoside showing a higher concentration for both treatments. In wines, statistical differences were found for all individual flavonols (except myricetin-3-galactoside and kaempferol-3-glucoside) in MeJ wines, and for some compounds (kaempferol-3-glucoside and galactoside and syringetin-3-glucoside) in BTH wines. Similarly, Ruiz-García et al. (2012) observed that control wines and MeJ wines had similar flavonol content. In a two-year experiment in which MeJ and BTH were applied preharvest to Monastrell grapes, Ruiz-García et al. (2013a) showed an increase in the flavonol content of treated grapes at harvest. On the other hand, these authors also reported differences between years for Monastrell grapes, suggesting that years with lower temperatures and higher humidity may provide suitable conditions for pathogen development and consequently help elicitor-treated plant react more efficiently.

The year this experiment was run was very dry and warm: only 130 mm of rain were measured from May to July and 53 mm from July to September, with maximum temperatures of 37, 37, and 32°C in July, August and September, respectively (data not shown; from Murcia Agricultural Information System).

Tannins in grapes and wines

    Tannins are an important component of wine quality contributing to both perceived mouthfeel and long-term colour stability. Seeds contain a higher concentration of tannins than skin and a higher proportion of galloylated units, whereas skin tannins have a higher mDP than seed tannins.

    The tannin levels of Syrah grapes (skin and seeds) and wines are shown in Table 4. Statistically significant differences from the control values were detected for total tannins in skin and seeds from MeJ-treated grapes and in wine from both treatments. Gómez-Plaza et al. (2017) also found higher concentrations of tannins in Syrah grapes treated with MeJ at harvest. In our study, both treatments (BTH and MeJ) also had an effect on mDP in the skin, seeds and wines. The percentage of galloylation was higher in BTH grape skins and in both BTH and MeJ wines. The percentage of tannin galloylation affects both the bitterness and astringency of wines (Lesschaeve and Noble, 2005). An increase in maceration time may explain the higher percentages of galloylation in higher quality wines (greater extraction of seed tannins, which are more galloylated) (Busse-Valverde et al., 2010). However, other winemaking techniques, such as the use of macerating enzymes or prefermentative cold maceration, could also increase the percentage of galloylation of wine tannins (He et al., 2010), although the maceration time was similar in all our wines.

    Regarding the composition of tannins in general and the terminal units in particular, in the Syrah variety (Table 4-first part), differences from control levels were more pronounced in skin and wine tannins than in seed tannins. As regards the terminal units, only (+)-catechin showed statistically significant differences from control levels for both treatments in skin tannins, but higher differences were found in the extension units: (+)-catechin, (-)-epicatechin and epicatechin-3-O-gallate content increased with respect to the control samples for both treatments. In seeds, only (+)-catechin content increased in both treatments and in wines, (-)-epicatechin content increased in both treatments and epicatechin-3-O-gallate content increased in the MeJ treatment.

    As regards the Merlot variety, statistical analysis identified significant differences from the control for total tannins in skin, seeds and wines in the MeJ treatment and only in seeds in the BTH treatment (Table 4-second part). The increase of mDP in skin, seeds and wines was only observed with the BTH treatment, while the MeJ treatment only affected the mDP of the skin tannins. As regards terminal units, only the percentage of (+)-catechin and epicatechin-3-O-gallate increased (for both treatments) in skin, (+)-catechin in seeds following BTH treatment, and epicatechin-3-O-gallate in seeds of MeJ-treated grapes and MeJ wines. Considering the extension units, Table 4 points to an increase in MeJ skin and wine epicatechin, BTH and MeJ seed epicatechin-3-O-gallate, and BTH skin and wine epigallocatechin. This last compound is a prodelphinidin and only appears in grape skin and, therefore, in wines made with these skins. Both (+)-catechin and (-)-epicatechin reached a higher percentage in wines following MeJ treatment.

    Finally, in the Monastrell variety (Table 4-third part), the differences were evident in skin, seeds and wines. Total tannins only increased over control levels in the seeds of BTH-treated grapes and in the wines of both treatments. Also, an increase in mDP was observed in skin and wine tannins for both treatments but only in seeds following BTH treatment. The percentage of galloylation was enhanced only in the skin of MeJ-treated grapes. An increase in the terminal units (-)-epicatechin and epicatechin-3-O-gallate was observed in BTH skins and wines and an increase in the extension unit epigallocatechin gallate was observed in MeJ skins and wines. The increases in the different tannin compounds resulted in an increase in the total tannin content of the wines made in both treatments.

    As has been seen, responses to the treatments depended on the variety. Other authors have mentioned a similar dependence on variety: for example, Vitalini et al. (2011) in a study of the effect of BTH and chitosan on the melatonin content of grapes, and Gómez-Plaza et al. (2017) in a study of the effects of elicitors applied during the ripening period in Syrah, Merlot and Monastrell grapes.

Sensory analysis

    To determine whether the results observed in the differently treated wines could be detected at sensory level, the wines were tested using a triangular test (Table 5). Again, the effect that the treatments had on the sensory analysis depended on the variety. In Syrah wines, no clear difference or preference for treated wines was observed. However, in the case of Merlot, the wines elaborated with BTH-treated grapes were clearly differentiated and preferred. Finally, in Monastrell wines, both treatments could be differentiated and most panellists preferred the MeJ wine to the control, while the opposite was observed for the BTH treatment.