Introduction

In recent years, edible berries have attracted much interest due to their high content and wide diversity of bioactive compounds with potential health benefits (Zhao 2007; Seeram 2012; Jimenez-Garcia et al. 2013; Nile and Park 2014). Berry fruits are widely consumed in fresh, dried, frozen forms or as processed products, including canned fruits, beverages, jams and yogurts. The most consumed berries are red raspberries (Rubus idaeus), strawberries (Fragaria x ananassa), blackberries (Rubus spp.), blueberries (Vaccinium corymbosum), black currants (Ribes nigrum), red currants (Ribes rubrum), chokeberries (Aronia melanocarpa), cranberries (Vaccinium macrocarpon), grapevines (Vitis vinifera L. and other Vitis species), bilberries (Vaccinium myrtillus L.), deerberries (Vaccinium stamineum L.), cowberries (Vaccinium vitis-idaea L.), passion fruits (Passiflora edulis) and tomatoes (Lycopersicon esculentum Mill.).

Among the wide variety of phytochemicals found in berries are stilbenes. These molecules occur within a limited group of plant families, which have the gene encoding the enzyme stilbene synthase (STS, EC 2.3.1.95). Biosynthesis occurs via the phenylalanine pathway, where phenylalanine ammonia lyase (PAL), cinnamate-4-hydroxylase (C4H), coumaroyl-CoA ligase (4CL) and STS play a core role in the synthesis. (Fig. 1) A transcriptional factor, Myb14, has been found to regulate the expression of STS (Holl et al. 2013). Chalcone isomerase (CHI) is responsible for the conversion of chalcone to flavanones. On the other hand, resveratrol O-methyltransferase (ROMT) is involved in the methylation of resveratrol. Stilbenes are biosynthesized and accumulate into lipid vesicles in the cytoplasm. Their content can be determined by many factors, including the cultivar, ripening stage, climatic conditions, soil type, agronomic management, storage conditions and postharvest management (Dixon and Paiva 1995; Castrejon et al. 2008). Due to the potential health benefits, stilbenes in edible fruits are of high interest. These compounds have demonstrated a wide range of biological and pharmacological activities, including anti-tumoural (Bai et al. 2010; Tsai et al. 2017), anti-viral (Nguyen et al. 2011), anti-inflammatory (Zhang et al. 2010), anti-atherogenic (Ramprasath and Jones 2010), anti-aging (Kasiotis et al. 2013) and neuroprotective (Lin and Yao 2006) effects.

Fig. 1
figure 1

Early stages of stilbene biosynthesis. PAL phenylalanine ammonia lyase, C4H cinnamate-4-hydroxylase, 4CL coumaroyl-CoA ligase, CHS chalcone synthase, STS stilbene synthase

Full size image

This review focuses on stilbenes as a specific class of non-flavonoid phenolic compounds present in edible berries. The aim of this review is to summarize the isolation and identification methods applied for stilbenes present in edible berries as well as the influence of external stimuli on the quantitative and qualitative composition of stilbenes in edible berries.

Molecular structures of stilbenes found in berry fruits

The stilbene structure is characterized by two aromatic rings linked by a double bond, of which the E isomer is the most common configuration. They can be found in berry fruits as monomers, dimers and more complex oligomers (Figs. 2, 3, 4, 5, 6, 7). According to a current paper, the most widely found monomeric stilbenes in berry fruits are E-resveratrol (1) and E-piceid (3) (Fig. 2). Their Z- and E-isomers mainly accumulate in the berry skin during all stages of development (Jeandet et al. 1991). E-isomer plant stilbenes may undergo several types of modifications, such as isomerisation, glycosylation, methoxylation, and oligomerization (Chong et al. 2009). Due to these modifications, different derivatives of stilbenes are formed in edible berries from dimers to hexamers (Figs. 2, 3, 4, 5, 6, 7). In plants, these metabolites generally accumulate in both free and glycosylated forms. Glycosylation of stilbenes could be involved in their storage, transport from the cytoplasm to the apoplasm and protection from peroxidative degradation (Morales et al. 1998). The oligomeric stilbenes are formed due to the oxidative coupling of E-resveratrol (1) or other monomeric stilbenes catalysed by peroxidase isoenzymes localized in the vacuole, cell wall and apoplast of grapevine cells (Ros Barcelo et al. 2003).

Fig. 2
figure 2

Molecular structures of E-stilbene monomers isolated from edible berries: 1: E-resveratrol (3,4′,5-trihydroxy-E-stilbene), 2: 3,5-O-dimethyl-E-resveratrol (E-pterostilbene), 3: E-resveratrol-3-O-β-D-glucopyranoside (E-piceid), 4: 3,3′,4,5-tetrahydroxy-E-stilbene (E-piceatannol), 5: E-piceatannol-3-O-β-D-glucopyranoside (E-astringin), 6: isorhapontin

Full size image
Fig. 3
figure 3

Molecular structures of Z-stilbene monomers isolated from edible berries: 7: Z-resveratrol, 8: Z-resveratrol-3-O-β-D-glucopyranoside (Z-piceid), 9: Z-piceatannol-3-O-β-D-glucopyranoside (Z-astringin)

Full size image
Fig. 4
figure 4

Molecular structures of stilbene dimers isolated from edible berries: 10: E- and Z-ε-viniferin, 11: E- and Z-δ-viniferin, 12: E- and Z-ω-viniferin, 13: scirpusin B, 14: parthenocissin A, 15: pallidol, 16: pallidol-3-O-glucoside, 17: ampelopsin D, 18: caraphenol B, 19: ampelopsin B

Full size image
Fig. 5
figure 5

Molecular structures of stilbene trimers isolated form edible berries: 20: amurensin B, 21: gnetin H, 22: vitisin E, 23: α-viniferin, 24: miyabenol C, 25: dividol A, 26: amurensin G, 27: ampelopsin G (wilsonol B), 28: wilsonol A

Full size image
Fig. 6
figure 6

Molecular structures of stilbene tetramers isolated form edible berries: 28: vitisin A, 29: vitisin B, 30: vitisin C, 31: hopeaphenol, 32: isohopeaphenol, 33: vaticanol C, 34: wilsonol C, 35: heyneanol A, 36: diviniferin B

Full size image
Fig. 7
figure 7

Molecular structure of stilbene hexamer isolated from edible berries: 37: chunganenol

Full size image

Isolation and identification of stilbenes in edible berries

The isolation and identification of stilbenes in edible berry extracts constitutes a complex procedure due to complex composition of the matrices in which their occur, their low concentration and structural complexity. Several strategies have been applied for the isolation and identification of stilbenes in edible berries. An overview of the preparation conditions, analytical methods and stilbene concentrations in accordance with the research objectives are presented in Table 1. In general, the extraction techniques applied in stilbene extraction are classified into two categories: conventional and green techniques. The conventional techniques involve soaking in solvent, which relies on the solubility of stilbenes from edible berries in the solvent at room or elevated temperature. These techniques consume a large volume of solvents and are usually time consuming. In contrast, the green extraction technique applies minimal volumes of solvent and requires a shorter time. It has been used as preparation procedure by Ehala et al. (2005) to isolate E-resveratrol from bilberry via microwave-assisted extraction. In conventional extraction techniques of stilbenes from edible berries solid–liquid extraction of lyophilized, air-dried, frozen or fresh samples with different solvents is applied (see Table 1). Unfortunately, there are no studies in which the influence of matrices (lyophilized, air-dried, frozen, fresh) on the presence of stilbenes was examined. From a quantitative point of view, the results obtained by different extraction solvents cannot be compared because solvents of different natures have different extractabilities. Basing on the analysed works (Table 1), the most often used solvent for the extraction process is methanol or the mixtures of methanol with other solvents. In the research conducted by Sun et al. (2006) and Romero et al. (2001) the influence of extraction solvents on the amount of extracted stilbenes was examined. Based on Sun et al. work (Table 1, entry no. 15), methanol acidified with 0.1% HCl was the best solvent to extract specific stilbenes from grape skins and seeds. In the research conducted by Romero et al. (2001) the influence of temperature and time of extraction on stilbenes content was examined. The highest extraction of E-resveratrol and piceid isomers was observed at 60 °C for 30 min with 80% ethanol. Z-Resveratrol was not detected in any conditions assayed. The longer time of extraction at 60 °C, the lower stilbenes content was measured, probably due to their degradation. It is only known that resveratrol and its glycon piceid are stable at 40 °C in the presence of ambient air (Prokop et al. 2006). However E-isomer is unstable in solution when exposed to light and readily isomerizes to the Z-form and other degradants (Jensen et al. 2010). Therefore, the preparation procedure of stilbenes isolation should be carried out in the dark due to the light sensitivity of double bond in stilbenes.

Table 1 Preparation conditions and analytical methods for stilbene separation and identification from edible berries
Full size table

Few studies have dealt with the influence of ultrasound on the resveratrol extraction efficiency from grapes (Burin et al. 2014; Babazadeh et al. 2017). The ultrasonication-assisted extraction of resveratrol showed more efficiency than the conventional solvent extraction with 80% ethanol at 60 °C for 30 min. The recovery of resveratrol increased by 24–30% compared with the conventional solvent extraction (Cho et al. 2006).

To decrease background noise in analytical techniques and improve the identification of stilbenes, the purification stage in the preparation procedure is very often applied. Typically, this purification utilizes an additional extraction with other solvent, often EtOAc (He et al. 2009a, b; Sun et al. 2006; Jiang et al. 2012), silica gel (Jiang et al. 2012) or C18 solid-phase extraction (Kiselev et al. 2017).

Many analytical methods with various detection techniques are reported for the separation and identification of individual stilbenes in edible berries, such as liquid chromatography-tandem mass spectrometry (LC–MS/MS) (Vrhovsek et al. 2012), liquid chromatography-mass spectrometry (LC–MS) (Može et al. 2011), liquid chromatography with dual detection by a photodiode array and quadrupole time-of-flight mass spectrometry (LC-PDA-QTOF/MS) (Samoticha et al. 2017), high-pressure liquid chromatography-mass spectrometry (HPLC–MS) (Bavaresco et al. 2002; Jiang et al. 2012; Kiselev et al. 2017), high-pressure liquid chromatography with diode array detection (HPLC–DAD) (Bavaresco et al. 2002; Sun et al. 2006; Vilanova et al. 2015; Guerrero et al. 2010a, 2016), high-pressure liquid chromatography with UV detection (HPLC–UV) (Vincenzi et al. 2013; Kawakami et al. 2014; He et al. 2009a), ultra-high-pressure liquid chromatography quadrupole time-of-flight mass spectrometry (UHPLC/QTOF/MS) (Flamini et al. 2016; De Rosso et al. 2016), ultra-performance liquid chromatography with dual detection by a photodiode array and fluorescence detectors (UPLC/DAD/FL) (Samoticha et al. 2017), ultra-performance liquid chromatography with dual detection by a diode array and tandem quadrupole mass spectrometry (UPLC/DAD/TQD) (Guerrero et al. 2010a), gas chromatography-mass spectrometry (GC–MS) (Ragab et al. 2006; Viñas et al. 2009, 2011), gas chromatography–mass spectrometry with selected ion monitoring (GC–MS SIM) (Rimando and Cody 2005), gas–liquid chromatography with flame ionization detection (GLC-FID) (Moriartry et al. 2001), capillary electrophoresis (CE) (Ehala et al. 2005) and high-speed counter-current chromatography (HSCC) (He et al. 2009b). Various methods are used for analyses of stilbenes contents in the edible berries are different which could also contribute to the observed variability in the published results. However, the most commonly used methods for stilbenes analysis in different edible berries is normal- and reverse-phase liquid chromatography connected to a diode array detector (DAD) or mass spectrometry (MS). To improve the identification of stilbenes the ultra-performance LC (UPLC) technique coupled with QTOF-MS has been used due to higher resolution and sensitivity of analysis (Flamini et al. 2016) (Table 1, entry no. 24, 26, 27).

Gas chromatography coupled with mass spectrometry (GC–MS) has also been applied for the analyses of stilbenes (Ragab et al. 2006; Rimando and Cody 2005; Viñas et al. 2009, 2011). Prior to GC–MS, the derivatization of hydroxy groups in stilbenes was performed in order to reduce polarity and increase volatility, and simultaneously, thermal stability of metabolites. Stilbenes derivatization was based on silylation reactions by means of N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) or N-methyl-N-(trimethylsilyl)-trifluoroacetamide (MSTFA) (Rimando and Cody 2005; Ragab et al. 2006; Viñas et al. 2009, 2011). However, due to the lower detectability of this technique only mono-stilbenes were determined such as E- and Z-resveratrol, E- and Z-piceid, pterostilbene and piceatannol.

Two strategies have been applied to identify stilbenes. In the case of known compounds, the identification was based on comparison of their retention times and MS or MS/MS data with those of standards. For unknown stilbenes, the characterization was performed by IR, MS, UV–Vis and NMR methods.

It is well known that the highest concentrations of stilbenes are in berries seeds and skins (Sun et al. 2006; Babazadeh et al. 2017). The amount of E-resveratrol in grape skin is approximately three times higher than in pulp (Babazadeh et al. 2017). Among the analysed edible berries, the highest concentration of stilbenes was found in the seeds of passion fruit (Passiflora edulis) (Kawakami et al. 2014) by applying 80% EtOH as the solvent for extraction. However, the highest concentration of E-resveratrol in the skin has been found in table grapes (Vitis vinifera L.) (Ragab et al. 2006) by using extraction with ethyl acetate at 70 °C.

Influence of external stimuli on the presence of stilbenes in grapes

Because stilbenes are secondary metabolites in edible berries, their quantitative and qualitative composition depends on many factors, including the cultivar, genotype, type of soil, climatic conditions, developmental stage, agronomic management, storage conditions (time, temperature) and postharvest treatments. In addition, there are other external stimuli (stress factors) that activate defence mechanisms in fruits responsible for the accumulation of stilbene as phytoalexins. The defence mechanism may be induced by abiotic elicitors such as UV irradiation, ozone, ultrasonication, methyl jasmonate, chitosan and visible light or biotic elicitors such as Aspergillus carbonarius and Botrytis cinerea. Recent review has discussed the influence of external stimuli on resveratrol synthesis in grapes (Hasan and Bae 2017). Various abiotic and biotic stress conditions have significant influence on the quantitative and qualitative composition of stilbenes in grapes, which are presented in Table 2.

Table 2 Influence of abiotic and biotic external stimuli on the presence of stilbenes in grapes
Full size table

Abiotic preharvest treatments of grape

Preharvest UV-C treatment of the Crimson seedless variety was applied daily for 3 days before harvest (Table 2, no. 1) and resulted in an increased concentration of stilbenes (Guerrero et al. 2016). The maximum content of E-resveratrol and E-piceatannol was achieved 24 h after each daily treatment. However, the ε-viniferin concentration was maximal at 48 and 72 h. In the case of E-piceid, Z-piceid and ω-viniferin, the maximum concentration was achieved at 72 h. The maximal contents of E-resveratrol, Z-piceid, E-piceid, E-piceatannol, ε-viniferin, ω-viniferin, isohopeaphenol and stilbenoids were 12-, 9-, 5-, 4-, 7-, 4-, 3- and 4-fold increased over those in the control sample. After daily periodic preharvest treatment of berries, the E-resveratrol content increased 83-fold in comparison with 18-fold growth for a single UV-C irradiation over the initial concentration. The first and the second treatments significantly increased the stilbene content, but the third daily treatment might have been important for maintaining their concentration. First, the biosynthesis of E-resveratrol is induced, and then the compound is glycosylated to E-piceid, which, under UV-C irradiation from the daily preharvest treatment, is transformed into Z-piceid.

The maximum E-resveratrol concentration detected in grape after UV-C treatment depends on its initial concentration, which correlates with the developmental stage of berries (Guerrero et al. 2010a, b). In another work (Table 2, no. 2), UV-C light preharvest treatment was applied on different days before grape ripeness to establish the optimum application day to reach the maximum E-resveratrol concentration. Due to UV-C irradiation, the highest E-resveratrol concentration in Red Globe grapes was achieved 3 days before harvest and was 46-fold higher than that observed in the non-treated control sample. However, at harvest, the E-resveratrol content was only 8.8 times higher than that in the control sample. UV-C treatment of berries 1 day before harvest resulted in 26 times higher E-resveratrol concentrations (Table 2, no. 2). The maximum E-resveratrol and ε-viniferin content was achieved when the UV-C dose was approximately 10,000 J/m2. Both the dose and the application method, in terms of output power and exposure time, are key factors determining the final stilbene content. Treatment with an output power and exposure time of 1040 W and 5 min, respectively, was selected as the most suitable condition for the UV-C treatment.

According to the literature, UV-C irradiation of edible berries effectively induces stilbene biosynthesis (Liu et al. 2010; Wang et al. 2010; Crupi et al. 2013). However, these studies only focused on berries at veraison and ripening. It is known that the biosynthesis and accumulation of stilbenes in berries after UV-C treatment depends on the developmental stages of fruits. From genomic analyses, it has been deduced that there are two main stages during grape development that are sensitive to UV-C irradiation (Pilati et al. 2007), namely, before and after veraison. “Before veraison” constitutes the restructuring phase of cell metabolism, characterized by an up-regulation of genes associated with hormone signalling and transcription. “After veraison” is characteristic of fruit ripening, whereas ‘veraison’ is characterized by an oxidative burst and antioxidant regulation. Therefore, it has been speculated that ‘veraison’ may be the most sensitive stage in which to apply UV-C treatment.

Under natural conditions, before veraison, the E-resveratrol content was very low in ‘Beihong’ (V. vinifera × V. amurensis) berries (Wang et al. 2015). From veraison to maturity, the E- and Z-piceid contents increased. UV-C treatment significantly stimulated the biosynthesis of E-resveratrol and E-piceid. The response of berries to UV-C irradiation was also related to berry development. Among the six developmental stages, the stage at 55 DAA (days after anthesis, 2 weeks before veraison) was the most sensitive to UV-C treatment. The contents of E-resveratrol and E-piceid increased 292- and 11-fold, respectively (Table 2, no. 3). Along with developmental factors, the sensitivity of resveratrol synthesis to UV-C irradiation gradually declined, which may be associated with the regulation of STS by the Myb14 promoter. STS expression was the highest when the berries were exposed to UV-C irradiation at 55 DAA, which may explain why resveratrol accumulated during this developmental stage. The expression of the Myb14 promoter reached maximum levels 12 h before STS. These results suggest that Myb14 expression may play an important role in the transcriptional regulation of resveratrol biosynthesis induced by UV-C irradiation.

Abiotic postharvest treatments of grape

Postharvest treatment with UV-C light has been proposed as a valuable method to increase the stilbenes content in the grape berries (Langcake and Pryce 1977; Douillet-Breuil et al. 1999; Versari et al. 2001; Petit et al. 2009; Yin et al. 2016).

Postharvest treatment of grape varieties with UV-C resulted in higher concentrations of stilbenes, such as E-resveratrol, piceatannol, viniferins and pterostilbene (Table 2, no. 4, 5) (Adrian et al. 2000; Guerrero et al. 2010a). Differences in concentration after UV-C irradiation depended on the variety and campaign, but not on the grape subspecies. Each variety seemed to be influenced to a different degree by the climate. Thus, the same variety behaved in a different way in each campaign, and climate could determine the final concentration of stilbenes.

The highest accumulation of resveratrol (tenfold) in irradiated Napoleon grapes was achieved using the following combination of parameters: irradiation power, 510 W; irradiation time, 30 or 60 s; irradiation distance, 40 cm; and elapsed days, 3 (Table 2, no. 6). Therefore, controlled UV irradiation parameters are useful as a simple postharvest treatment to increase the resveratrol concentration in Napoleon grapes (Cantos et al. 2001).

To achieve the highest possible stilbene accumulation, the interactive effects of storage time, temperature and UV-C irradiation on the stilbene content in postharvest Red globe table grapes were investigated (Table 2, no. 7) (Crupi et al. 2013). During storage, both cold storage and UV-C doses of 3 min raised the contents of Z- and E-piceid, achieving 90 and 34 μg/g in skin, respectively, which was approximately threefold higher than those in control berry samples. Similar results were found in Napoleon table grapes (Cantos et al. 2001). Cold storage in combination with UV irradiation increased the piceid concentration more than cold storage alone. Also Cho et al. (2012) reported that it is possible to enrich resveratrol content in harvested grapes by modulating cell metabolism with UV treatment and storage conditions. Storage temperature had an effect on time-delayed resveratrol biosynthesis after removal of the UV irradiation. A larger amount of resveratrol was formed when UV-treated grapes were stored at higher temperature.

After UV-C postharvest irradiation, all of the red grape varieties in each terroir increased their resveratrol, piceatannol and viniferin contents (Table 2, no. 8) (Fernández-Marín et al. 2013). The stilbene content was different depending on the variety and the terroir. Cabra was the terroir where the varieties achieved the highest induction capacity (2.02 μg/g per day after UV-C irradiation), especially the Syrah variety. This is in agreement with previous research in which Syrah increased its stilbene content more than the other thirteen varieties studied (Guerrero et al. 2010a). However, the highest increase in the resveratrol and piceatannol contents in the Syrah variety was from the Jerez terroir, which amounted to 2.7 and 6.1 times, respectively, in comparison to those in the untreated berries.

With regard to piceatannol and viniferins, higher concentrations were found in varieties that achieved higher resveratrol levels because resveratrol has been proposed as the precursor of the other stilbenes (Coutos-Thévenot et al. 2001). Thus, it could be concluded that resveratrol determines the tendency for the synthesis of other stilbenes and the amount of total stilbenes.

Moreover, varieties reached the highest level of resveratrol induction in different periods of time depending on the terroir. The day of maximum concentration (dm) was constant for Cabernet sauvignon (dm = 7). Pinot noir showed the same dm in all terroirs (dm = 6), except for Cadiar, where it was delayed 1 day (dm = 7). However, in Syrah and Merlot varieties, the dm changed depending on the terroir, ranging from 4 to 7 days.

Treatment of the postharvested Pinot Noir grape berries with UV-C irradiation increased the E-resveratrol, ε-viniferin, piceid and Z-resveratrol contents 355, 4.8, 2.4 and 0.7 times, respectively, in comparison with those in the control sample (Table 2, no. 9) (Suzuki et al. 2015). Transcriptome analysis revealed that 238 genes were up-regulated more than fivefold in grape berry skin by UV-C treatment. Enrichment analysis of the gene ontology terms showed that genes encoding stilbene synthase were enriched in the up-regulated genes.

Two different wavelengths were used for the UV treatment of red grapes: 302.1 nm for the resonant wavelength and 300.0 nm for the non-resonant wavelength (Table 2, no. 10) (Jiménez Sánchez et al. 2007). Four sets of irradiation times were selected for each of the two different wavelengths: 15, 30, 45 and 60 min. The use of photons of resonant energy to produce absorption through the real electronic states of the molecule significantly increased the absorption yield, producing an important effect on the photoinduced E-resveratrol level in the grapes. The enhancement was optimal for 45 min of irradiation at 302.1 nm. The samples were prepared for analysis immediately after irradiation to detect the direct effect of resonant elicitation.

The combination of UV-C treatment with a chitosan coating and incubation was investigated (Table 2, no. 11) (Freitas et al. 2015). The concentration of E-resveratrol in red table grapes with a 0.5% chitosan coating treated with UV-C irradiation, incubated at 20 °C and then stored for 5 days under refrigeration was approximately 5 and 2.5 times higher than that in control and UV-C treated grapes with a 0.5% chitosan coating, respectively. After 8 days of storage at 4 °C, treated berries also showed a 2.9 times higher content of E-resveratrol and lower susceptibility to fungal decay than control grapes.

The individual and combined effects of CaCl2 and ultraviolet light on the biosynthesis of resveratrol in berry skins were investigated (Table 2, no. 12) (Wang et al. 2013). CaCl2 application had no effect on the E-piceid content and little influence on the Z-piceid content, but it resulted in a 0.5 times higher content of E-resveratrol compared with that in the control sample. UV-C and UV-C irradiation with CaCl2 treatment increased E-resveratrol and E-piceid biosynthesis and accumulation in berry skins. This accumulation continued until 13 days after cold storage. Then, the E-resveratrol content slightly decreased but still remained at high levels at the end of storage, while the E-piceid content increased continuously until the end of storage. In contrast, the Z-piceid content after UV-C and UV-C/CaCl2 treatments decreased by 0.6 and 1.4 times, respectively, compared with that in the control sample.

UV-C and UV-C/CaCl2 treatments significantly stimulated the expression of PAL, C4H, 4CL and STS, which are related to the biosynthesis of E-resveratrol. Moreover, the expression levels of these genes in the UV-C/CaCl2 combination treatment were higher than those in the UV-C irradiation treatment. The expression of PAL, C4H, 4CL, and STS in both treatments reached a maximal content at 12 h after initiating the treatment and then declined rapidly to approach the control level at the end of the experiment. CaCl2 treatment alone did not modify the expression of any of these genes (Wang et al. 2013).

Increased accumulation of E-resveratrol in grape skin by ultrasonication treatment has also been investigated (Table 2, no. 13) (Hasan and Baek 2013). A significantly higher amount of E-resveratrol over that in the control sample was observed following ultrasonication treatment. Ultrasonic treatment for 5 min followed by 6 h of incubation induced the highest levels of E-resveratrol, with amounted to a level that was 6.7 times higher than that in the control sample. However, the treatment did not lead to an increase in the maintenance of the level of E-resveratrol. When the amount of E-resveratrol was measured again in fruits incubated for another 6 h, it had decreased drastically to the level in samples treated by 5 min of ultrasonication without any incubation.

In grape skin, the expression levels of resveratrol synthase increased directly in response to 5 min of ultrasonication treatment and were then maintained during 12 h of incubation. This suggests that the accumulation of E-resveratrol in grape in response to ultrasonication occurs in a time-dependent manner based on the induction of the resveratrol synthase gene.

Ozone was also found to stimulate the synthesis of stilbenes in grape berries (Sarig et al. 1996; Palou et al. 2002). Continued and intermittent ozone treatments (2 ppm) were applied during the storage of 3 varieties of table grapes at 5 °C for 72 days (Table 2, no. 14) (Cayuela et al. 2010). The continuous presence of ozone in the storage atmosphere inhibited the biosynthesis of resveratrol, whereas intermittent treatment with O3 stimulated its biosynthesis. It was found that the treatment consisting of intermittent shocks of 8 ppm O3 had the strongest influence on table grapes (Artés-Hernández et al. 2003). It was also found that the grapes stored in an atmosphere containing 0.1 ppm O3 exhibited higher concentrations of resveratrol than those kept in air. Mild ozone treatments, such as those using ozonized water, which are effective in the control of microbial growth, did not induce stilbenoid accumulation in grapes (Gonzalez Ureña et al. 2003). Continuous exposure of the berries to a high concentration of O3 could reduce the content of antioxidant compounds such as resveratrol produced by the plant as a defensive metabolite against oxidative stress.

The influence of the combination of ozone and UV-treatment on stilbenes content was also investigated (González-Barrio et al. 2006). The results showed that UV-C was generally much more efficient (shorter treatment time) in inducing resveratrol content than ozone. However, with regard to total stilbenoids accumulated in the grape skin, the ozone treatment with the highest concentration and the longest time led to higher stilbenoid content than the UV-C irradiation. Also the viniferin content accumulated was threefold higher than that induced by the UV-C treatment. However, UV-C light treatment resulted in less damage to grape tissues than ozone gas.

Different gaseous treatments have been applied because of their efficacy in ensuring the quality of Napoleon table grapes (Table 2, no. 15) (Artés-Hernández et al. 2003). The E-resveratrol content increased up to threefold its value sampled at harvest in O3 shock-treated clusters and up to twofold for other treatments, probably because these treatments induced a faster biosynthesis of this compound. The increases were more marked when clusters were stored at 15 °C, which is in agreement with previous reports, where control and UV-treated cv. Napoleon table grapes increased their resveratrol content after the grapes were transferred to 15 °C (Cantos et al. 2000). However, O3-treated clusters that underwent shock treatment increased their E-piceid content by 11 and 4.5 times compared to that sampled at harvest.

The accumulation of stilbenes and the expression of genes related to their synthesis in ‘Campbell Early’ and ‘Kyoho’ grapes were investigated by irradiating the harvested grapes with four different light sources for 48 h (Table 2, no. 16) (Ahn et al. 2015). The total concentrations of five stilbene derivatives at 24 h after irradiation differed in response to different light sources and cultivars. The accumulation of stilbenes in the skins of two grape cultivars and the expression of PAL and STS1 genes were induced under mainly red and blue LED light. The amount of stilbenes tended to be higher in blue and red light-treated grapes than in the grape berries treated with white fluorescent or purple light. Among the stilbenes tested, E-resveratrol was present in ‘Campbell Early’ berries treated with blue, red, purple and fluorescent light at levels 7.4-, 6.2-, 2.8- and 1.8-fold higher than the levels in the control sample, respectively. The expression of PAL, CHS, CHI, STS1, STS12 and ROMT genes was differently induced in response to irradiation with different light sources in both grape cultivars. The mRNA levels of PAL and STS1 were higher than those of CHS, CHI, STS12 and ROMT in the two grape berries. The results indicated that red and blue LEDs induced the accumulation of stilbenes and the expression of genes related to their syntheses in grape berries.

Accumulation of E-resveratrol in post-harvested grapes by dry nitrogen treatment was also investigated (Table 2, no. 17) (Jiménez et al. 2007). E-resveratrol content in the grape berries increased with the duration of the dry nitrogen treatment up to 24 h (twofold). For longer treatment, only a slight enhancement was observed. However, shorter treatments kept higher content of E-resveratrol during several days with no appreciable damage of the grape berries regarding to their organoleptic quality.

It was also observed that storage conditions influence the stilbene concentration (Table 2, no. 18). Grapes stored under mild conditions (20 °C) favoured E-resveratrol synthesis, whereas cold storage (4 °C) decreased its concentration. Cold storage apparently inhibited E-resveratrol synthesis after harvest, and its concentration slowly decreased. This may be explained by the action of peroxidase enzymes and the subsequent formation of ε-viniferin. In non-UV-C irradiated grapes, the E-resveratrol content increased by nearly 14 times when after 3 days of storage at low temperature, the grapes were stored under mild storage conditions.

After 60 days of withering, Raboso Piave and Corvina grape samples showed an evident increase in the content of most stilbenes (Table 2, no. 18) (De Rosso et al. 2016). Raboso Piave had a statistically significant increase in E-resveratrol, pallidol, and E-ε-viniferin. In Corvina grape, E-resveratrol, piceatannol, E-astringin, E-piceid, pallidol, pallidol glucoside, E-miyabenol C, and ε-viniferin increased significantly. In general, these findings are in agreement with those reported in a study of Aleatico grape withering (Mencarelli et al. 2010).

Combined abiotic pre- and postharvest treatments of grape

Preharvest red grapes (Syrah Vitis vinifera L.) treated with methyl jasmonate (MEJA) showed a significant increase in E-resveratrol and piceatannol contents compared to those in the control sample (piceatannol not detected) (Table 2, no. 19) (Fernández-Marín et al. 2014). The highest concentrations of piceatannol and E-resveratrol were found at harvest. Larronde et al. (2003) described an E-resveratrol increase (ninefold) in Cabernet sauvignon berries when treated with MEJA vapours 15 days after veraison. However, berries rapidly lose the capacity to respond to MEJA with ripening. Vezzulli et al. (2007) found that MEJA treatment of the Barbera grape variety improved the E-resveratrol and ε-viniferin contents in an accumulative manner. Thus, it seems that there is an effect of MEJA on the stilbene concentration in grapes.

Harvested red grapes were treated with UV-C irradiation and stored for 4 days (Table 2, no. 19) (Fernández-Marín et al. 2014). On the 4th day, control red grapes showed a significantly lower E-resveratrol content than grapes treated with UV-C, and UV-C samples a lower E-resveratrol content than MEJA samples, whereas MEJA-UV-C achieved the highest E-resveratrol content. In fact, UV-C irradiation has been described as a stronger stress than MEJA, causing higher E-resveratrol induction in grape, but a longer grape storage period was required (Fernández-Marín et al. 2012).

The accumulation of piceatannol, isorhapontigenin, and ε-viniferin were also induced (Table 2, no. 19). At harvest, piceatannol was found in MEJA grapes, but it was absent in control samples. After 2 days of storage, piceatannol was found in every batch, and isorhapontigenin was found in the MEJA and MEJA-UV-C batches in concentrations that were not significantly different. The trend of these stilbenes was similar to that of E-resveratrol. The control batch showed the lowest concentration, in contrast to MEJA-UV-C, which showed the highest concentration after 4 days of storage.

The effectiveness of the chitosan treatment of table grapes (Autumn Black and B36-55 variety) in combination with UV-C irradiation to determine the E-resveratrol concentration in grape berry skins was investigated (Table 2, no. 20) (Romanazzi et al. 2006). Grape berries were sprayed in the vineyard with 1% chitosan and then harvested daily for 5 days. Immediately after harvest, they were inoculated with B. cinerea. In cv. Autumn Black and selection B36-55, E-resveratrol was not detected in control berries or berries with a chitosan coating. In berries exposed to UV-C irradiation alone and in berries treated with 1% chitosan coating and later exposed UV-C, E-resveratrol was found. The berries treated with the combination of chitosan and UV-C irradiation contained approximately 11 times more E-resveratrol than the control grapes and 0.2 times more than those only irradiated with UV-C. Z-Resveratrol, E-piceid, and Z-piceid were not detected in any of the samples.

Biotic postharvest treatments of grape

Infection of the postharvested Palomino fino grapes with Botrytis cinerea led to a domination of piceid over resveratrol (Table 2, no. 21) (Roldán et al. 2003). At the early stage of fungal development, the content of piceid increased more than onefold in fruits over that in the control sample, while the level of resveratrol decreased more than onefold. At more advanced infection stages, the piceid and resveratrol contents increased 3.5-fold and 0.4-fold, respectively. According to the literature, resveratrol synthesis occurs in the skin of edible berries (Sotheeswaran and Pasupathy 1993). Thus, an accumulation of resveratrol in the skin after infection by B. cinerea could be expected. However, the results showed that even at the early stage of pathogen development, a sharp decrease in the resveratrol content occurred. This could be because before pathogen exposure, resveratrol synthesis is stimulated. The resveratrol already present in the skin is used by the fruit as part of its defence mechanism. This phytoalexin would become a building block of stilbenes such as ε- and δ-viniferins, which are more toxic to the pathogen. For this reason, the piceid and resveratrol contents decreased. In the second stage of pathogen development, an accumulation of resveratrol occurred until a certain level was reached, and it was maintained during external stress. These results are in accordance with the literature, which showed that in grape berry skins infected by powdery mildew, the resveratrol and piceid isomers were considerably increased and the degree of infection was positively related to their stilbene content (Romero et al. 2001).

Infection with Aspergillus carbonarius at ripening induced in the grapes an approximately onefold increase in E-resveratrol and resveratrol dimers, such as ω-viniferin, E-ε-viniferin, caraphenol and δ-viniferin, and resveratrol trimers, such as α-viniferin and E-miyabenol C (Table 2, no. 22) (Flamini et al. 2016). However, infection with this pathogen decreased the E-piceid and E-astringin concentrations by approximately onefold in grapes. The total stilbene content in grapes infected with A. carbonarius increased 60% over that in uninfected berries. Glycosylation of stilbenes is involved in storage, transport from the cytoplasm to the apoplasm and protection from peroxidative degradation. Subsequent storage in vacuoles may protect plant cells from potentially their toxic effects. In susceptible cultivars, such as Gamay, Gamaret, Pinot Noir and Chasselas, resveratrol was found to be either glycosylated or present in very low concentrations (Pezet et al. 2004). Studies of grapevine varieties resistant to P. viticola reported that E-resveratrol is probably rapidly oxidized into toxic stilbenes, such as ε- and δ-viniferins, while in susceptible cultivars, it is rapidly glycosylated into less toxic piceid (Pezet et al. 2004). In the case of infected Negro Amaro grape, the content of E-resveratrol increased, but the content of glycosylated stilbenes decreased, which confirms the low susceptibility of this cultivar against A. carbonarius infection.

In the research conducted by Bavaresco et al. (2003) the influence of fungal infection with A. japonicus, A. ochraceus, A. fumigatus and A. carbonarius on the stilbenes content in the grape berries was examined. All tested fungi, except A. fumigatus, significantly increased E-resveratrol concentration over the control sample, while E-piceid content was not affected. Among tested fungi, only A. ochraceus significantly elicited the grape berries to synthesize piceatannol.

Combined abiotic and biotic postharvest treatments of grape

In another study, phenyl propanoid metabolism induction by abiotic (UV-C) and biotic elicitors (fungal infection—ochratoxigenic Aspergillus) in undamaged and damaged Napoleon table grapes was investigated (Table 2, no. 23) (Selma et al. 2008). In addition, the effect of the sequence of elicitors on the content of stilbenes and storage time was analysed. The contents of E-resveratrol in non-inoculated and inoculated grapes were similar and maintained at the same level during 5 days of storage. Treatment of undamaged or damaged grapes with UV-C and fungal infection increased the content of E-resveratrol 6 times (7 or 4 times for undamaged and damaged berries, respectively, in the case of infection and subsequent UV-C treatment) compared with that in non-inoculated samples, after 5 days of storage. In addition, the E-resveratrol content did not increase in damaged grapes compared to that in undamaged grapes either at day 0 or after storage. These results suggest that E-resveratrol was not elicited by ochratoxigenic Aspergillus. However, in another study, it has been shown that E-resveratrol accumulates in A. carbonarius inoculated grapes at veraison time but not during ripening (Bavaresco et al. 2003). Moreover, E-resveratrol formation decreases from veraison to ripening in berries, and this formation is elicited by biotic (ochratoxigenic Aspergillus) and abiotic (UV-C) factors (Jeandet et al. 1991). However, the induction capacity for E-resveratrol was sevenfold, which is higher than the value of 4.4-fold obtained in another study (Cantos et al. 2002). The accumulation of E-resveratrol was significantly enhanced by increasing the storage time and by UV-C treatment, as well as by the combination of both treatments.

Inoculation with ochratoxigenic Aspergillus did not specifically elicit the accumulation of E-piceid (Table 2, no. 23). However, there was a significant increase in the induction of E-piceid during storage, and this induction was particularly faster in damaged berries than in undamaged grapes. Therefore, E-piceid induction in UV-C untreated grapes could be a consequence of the total microbial contamination rather than a specific response to ochratoxigenic Aspergillus. On the other hand, in undamaged berries, the E-piceid content increase in response to microbial infection was faster in UV-C treated berries than in untreated berries. The E-piceid content in damaged grapes after treatment with UV-C and fungal infection increased onefold.

UV-C treatment induced the biosynthesis of E-resveratrol as a result of an increased transcription of genes encoding stilbene synthase. Fungal infection induced not only the transcription of stilbene synthase but also the transcription of genes encoding glycosyltransferases, which transform E-resveratrol to E-piceid.

Conclusions

Edible berries are a rich source of nutritive and bioactive compounds, including stilbenes, among which E-resveratrol, piceatannol, Z/E-piceid and viniferin are the most common. The extraction techniques applied in stilbene isolation can be classified into two categories: green techniques and conventional techniques. Conventional techniques use a large volume of solvent, especially methanol or ethanol and their aqueous mixtures or mixtures with other solvents. Green techniques are not as common as conventional techniques but allow minimalization of solvent usage.

Among the analysed edible berries, the highest concentration of stilbenes has been found in passion fruit (Passiflora edulis). However, the highest concentration of E-resveratrol in the skin has been found in table grapes (Vitis vinifera L.). The quantitative and qualitative composition of stilbenes in edible berries depends on many environmental factors. However, the application of various abiotic and biotic external stimuli may increase the production of stilbenes in grapes. Several different elicitors, as inducers of secondary metabolite stilbenes, were analysed: UV irradiation, visible light, ultrasonication, signalling compounds and fungicides. Among them, UV treatment was the most effective method to enhance stilbene derivative production. The enhancement of the stilbene content depended on the UV wavelength, UV intensity, duration of irradiation and developmental stage of the berry fruit. The STS expression responsible for stilbene production was the highest when the berries were exposed to UV-C irradiation at 2 weeks before veraison. In general, the effect of biotic stressors, such as Aspergillus carbonarius and Botrytis cinerea, was found to be not as efficient as the effect of abiotic elicitors.