Study of polyphenolic compounds in wines and different parts of the grapevine

Two experiments were conducted during this research. One was intended to study polyphenolic compounds in parts of the grapevine (Figure 1). The other was focused on the fermentation of wine with the addition of the individual parts of the plant (Figure 2).

Figure 1.

Figure 2.

The Souvignier gris grapes were separated into their individual parts: seeds, skins, flesh and clusters. On the day of harvesting, leaves were also collected. The individual parts (from five bunches) were freeze-dried for 12 hr in a freeze-drier (model BK-FD18P vertical freeze-dryer; Biobase: Hangzhou, China). The seeds were then ground in a coffee grinder, and the leaves, stems and skin were crushed in a mortar.

A sample (1 g) of each of the single homogenised mixtures was weighed into a tube. A total of three tubes were filled for each variant (seeds, leaves, stems and skin).

The extraction was carried out using a pre-prepared methanol solution, prepared as follows: 175 mL of methanol (99.9%) and 0.025 mL of acetic acid (99.8%) were measured out into a container, and the total volume was made up to 250 mL with the addition of distilled water. Then, 10 mL of this methanol solution was added to the sample tube and left to shake for 30 min (model ES-60+ incubator shaker; Miulab, Hangzhou, China) in an ultrasonic environment at a temperature of 50°C. This was then placed in a centrifuge for 4 min at 3000× g (model 5702 R; Eppendorf, Hamburg, Germany). Samples were collected for further experimental measurements.

The Souvignier gris grape variety (Czech Republic; village: Lednice; vineyard: Na valtické) was used in the experiment. It was harvested on 21 October 2021. After pressing, the juice was left to settle naturally for 24 hr. Subsequently, the must was divided into fermentation vessels and was twice fermented using active dried wine yeast (Institut Oenologique de Champagne, Cedex, France). Each experimental variant was repeated three times. To each fermentation vessel, the following fresh parts of the grapevine were added: the leaves, stems, seeds and skins, with the last variant as the control. The concentration of each part was 72 g · L^-1 seeds, 250 g · L^-1 skins, 36 g · L^-1 stems and 10 g · L^-1 leaves. The quantities were calculated from the original quantity of grapes that were required to produce 1 L of must. After fermentation, samples were taken, and then the experiment was terminated (Table 1).

The basic parameters (sugar content, titratable acidity and pH) of the resulting wine or must were determined using an Alpha Fourier-transform infrared spectroscopy (FTIR) analyser (Bruker, Karlsruhe, Germany) using the attenuated total reflection sampling technique. Prior to the first measurement, the spectrometer was thoroughly rinsed with deionised water and the background levels were determined using a blank sample (deionised water). For the analyses, 1 mL samples were taken using a syringe; 0.5 mL was used to rinse the system, and the remaining 0.5 mL was analysed three times. The measured values were evaluated automatically using the Opuswin software system (Opuswin Technology and Systems, Lagos, Nigeria).

Each variant was prepared in triplicate and each sample was measured three times. The arithmetic mean with standard deviation was calculated from the values using STATISTICA 14 statistical software (Tibco, Palo Alto, CA 94304, USA). Further, one-way analysis of variance (ANOVA) was used to evaluate the experimental data, with significant differences between the means set at p ≤ 0.05. Tukey’s honestly significant difference (HSD) post-hoc test was used to determine similar and dissimilar levels of factors (groups). The results are presented as tables in the ‘Results and Discussion’ section.

The raw material used in the experiment was Souvignier gris. It was divided into the following parts: seeds, skins, pulp and stems. On the day of harvesting, leaves were also added to these samples. The raw material was dried and divided into samples as per the methodology described. After sample preparation, the levels of the individual polyphenols were measured by LC-MS. There were three replicas of each variant, and each variant was subsequently measured three times. The results were averaged and the standard deviation was calculated (see Table 3).

Significant differences were measured in the experimental extracts from different parts of the grapevine. Caftaric acid and quercetin-3-glucoside were the most abundant compounds found in the leaves (Table 3). The values of these two compounds were considerably higher than for any of the others. They were followed by kaempferol-3-D-glucoside and coutaric acid. Myricetin, ethyl coumarate, ethyl caffeate and isorhamnetin were found to have the lowest levels of the polyphenols measured. As for the skins, the main polyphenol found was quercetin-3-glucoside (Table 2). In the case of the skins, a substantially greater difference was seen between the concentrations of the most abundant polyphenols. Caftaric acid was found at the second highest concentration. Similar results were found in the experiment by Ostroukhova et al. (2016). In their experiment, quercetin, kaempferol and their glycosides dominated in the leaves (Ostroukhova et al., 2016). No value was measurable for the polyphenols myricetin, vanillic acid, syringic acid, ethyl caffeate, ethyl coumarate and piceatannol. In the seeds, two substances predominated, namely catechin and epicatechin; they only displayed minimal differences in concentration. It is worth mentioning that procyanidin B1 had the third highest concentration. However, this is not the case with the stems, wherein procyanidin B1 was the most dominant. Once again, for the stems, we can see a considerable difference between the polyphenols. Catechin was the second most abundant and the third was trans-resveratrol. Similar results were presented by Souquet et al. (2000), who conducted similar experimental work and measured the extracts. They confirmed that the seed was the main source of catechins and epicatechins and reported that stems came in second place. For comparison with the merlot variety, the value of catechins was 60 mg · kg^-1 (Souquet et al., 2000). From the results, it can be seen that the main source of trans-resveratrol is the stem (Table 1). Ethyl coumarate was not measured in any of the parts of the grapevine tested. A study by Rothwell et al. (2005) claimed that the largest source of quercetin and its compounds is the stem. This claim was not confirmed by our results (Rothwell et al., 2005).

Boso et al. (2019) performed an experiment in which they focused on flavanols in different parts in Albarino and Mencía varieties of grapes. The Albarino variety had the highest level of flavanols in the stems, followed by the seeds. With regard to polyphenols in the stems, this variety had the highest level of catechins, followed by procyanidin B1; in our results, they came in the other way around. Their findings for the seeds were remarkably similar to ours, with catechin and epicatechin being the most abundant. A match was also found for procyanidin B1 in the skins; it was at the highest level in both experiments (Boso et al., 2019).

Jara-Palacios et al. (2014) did similar research, and we have confirmed their results. They performed their work on the Zalema grape variety. They measured 31 polyphenolic compounds in different parts of the fruit (seeds, flesh, stems and skins) through rapid resolution LC (RRLC)/MS. They reported that flavanols were the most abundant phenolics found, with concentrations in the range of 121–613 mg · 100 g^-1 dry weight, followed by flavanols (8–146 mg · 100 g^-1) and phenolic acids (9-27 mg · 100 g^-1). The highest levels of flavanols were found in the seeds (613 mg · 100 g^-1), followed by the stems (348 mg · 100 g^-1) and the pomace (282 mg · 100 g^-1), while the skins had the lowest concentration (122 mg · 100 g^-1). In contrast, the flavanols were most abundant in the skins and pomace (146 and 144 mg · 100 g^-1, respectively) with no significant differences between them, whereas the seeds were low in these compounds (8 mg · 100 g^-1). The derivatives of phenolic acid were minority compounds in the four by-products. The main flavanols in the by-products were the quercetin glycosides, with quercetin 3-O-glucoside and quercetin 3-O-glucuronide accounting for 32%–42% and 35%–37% of the total flavanol content, respectively. Both compounds were more abundant in the pomace and skins (51–58 mg · 100 g^-1) than in the stem (25–27 mg · 100 g^-1) or seed extracts (around 3 mg · 100 g^-1) (Jara-Palacios et al., 2014).

The Souvigner gris variety was used for this experiment. The grapes were handpicked and processed. After pressing, the must was allowed to settle naturally, and then the individual parts of the plant were added. These were the skins, seeds, leaves and stems. The must had not fermented and fermentation started spontaneously. Fermentation lasted for 14 days. After fermentation, samples were taken, and the individual polyphenolic compounds were subsequently measured by LC-MS. Three replicates of each variant were produced, and each variant was measured three times. The results were averaged, and the standard deviation was calculated. In addition to the above variants, measurements were also taken for the wine as a control (Table 4).

According to Table 4, for the variant with the addition of skins, caftaric acid was found to be the most abundant, followed by coutaric acid. The levels of kaempferol, myricetin, ethyl coumarate, piceatannol, astringin and procyanidin C were insignificant. Very similar results were obtained for the variant that had leaves added to the must. Again, here, the most dominant polyphenolic compound was caftaric acid. No detectable levels of concentration were measured for myricetin, kaempferol, isorhamnetin, syringic acid, piceatannol and procyanidin C. It is interesting to note that for the variant that contained the skins, quercetin-3-glucoside and caftaric acid were dominant in the extracts, but in the wine, this dominance was not apparent. Caftaric acid is still the most abundant substance, but it is not at as high a level as it was in the extracts. In terms of the results, we can say that there are only three compounds that were not present: kaempferol, isorhamnetin and piceatannol. For the version with added seeds, on the other hand, the situation is different. According to Tables 3 and 4, it is clear that the main compounds in both the experimental extract and the experimental wine are similar. They were mainly catechin, epicatechin, procyanidin B1 and procyanidin B2. The level of catechin found was several times greater than was found in the other variants. Kaempferol and piceatannol are two polyphenolic substances that were not measured in this variant. In the stems, the substance with the highest concentration was caftaric acid. There is a considerable difference between the concentrations of the most abundant substance versus the second most abundant. As was the case with the extracts, catechin and procyanidin B1 were the predominant substances found. For some substances, such as myricetin, kaempferol, isorhamnetin and piceatannol, no detectable levels were found. These findings confirm a study by Puskás et al. (2010), which reported that the addition of seeds increases the total content of polyphenolic substances. According to that study, the addition of 50% of the picked stems during processing increases the number of phenolic compounds by 60%(Puskás et al., 2010).

In 1999, Sun et al. (1999) conducted an experiment where they attempted to establish the percentage of polyphenolic compounds that could be released into the wine from the solid parts of the plant. They found that between 45% and 50% of the total catechins entered the wine from the seeds. We agree with their results regarding the release of polyphenolic compounds from the skins. The skins contain much lower levels of these substances, but they enter the wine to a much greater extent than from the stems. Our study confirms these results (Sun et al., 1999).

Another interesting study was done by Revilla et al. (1997). They found that the extraction rate of catechins and procyanidins can be considered to be linear during the first 7 days after the start of maceration. In the case of merlot, the catechin and procyanidin contents of the Me-2 wine was approximately 25% higher than in the Me-1 wine. The maceration took the same length of time (92 hr), but the quantity of seeds in contact with the must was 25% higher for the Me-2 wine in comparison to the Me-1 wine. The addition of more pomace led to a higher total phenolic content in the free run wines from the Me-B, Me-C and Fr-B trials than in the control merlot and frankinja wines. As expected, the total content of phenolics in the pressed wines (Me-4 and Fr-3) was higher than in the corresponding free run wines (Me-2 and Me-3 for Merlot and Fr-2 for Frankinja). The total levels of phenolics in the coupage wines (Me-5 and Fr-5) were fairly similar to what was expected given the total phenolic content of the wines (Revilla et al., 1997).