1 Introduction

Grapes are one of the major fruits grown worldwide. According to a report from the Food and Agriculture Organization of the United Nations (FAO), 75 million tons of grapes are produced annually, of which about 50% are consumed as wine, 33% as table grapes, and the rest consumed in a dry form (FAO 2022). Grapes have considerable economic value. They are used for wine, juice, and jams, and their seeds are used in supplemental food (Tu et al. 2016).

As the berries mature, the chlorophyll content decreases and the skin undergoes a green to veraison transformation, finally becoming black-blue or black-purple (Giovanelli and Brenna 2007). Anthocyanin accumulates in the skin cells in the veraison phase and the polyphenol content increases during maturation (Deytieux et al. 2007; Giuffrè et al. 2013; Koyama et al. 2017; Versari et al. 2001). Grapes with red skin, including wild grapes, have abundant polyphenol components, such as flavonoids, anthocyanin, and stilbene compounds (Ahn et al. 2015b; Flamini et al. 2013; Li et al. 2017; Salehi et al. 2019; Shi et al. 2014). Polyphenols are functional substances for humans with many benefits, such as anti-inflammatory, anticancer, liver function improvement, cardiovascular protective, and antioxidant effects (Guthrie et al. 2017; He et al. 2010a, b; Leifert and Abeywardena 2008; Mollica et al. 2021; Rice-Evans and Packer 2003; Smoliga et al. 2011; Vislocky and Fernandez 2010). Stilbenes, a type of polyphenol, were originally called phytoalexins, which is a secondary metabolite produced by plants exposed to abiotic and biotic stress (Ahuja et al. 2012; Hart 1981; Kim et al. 2018; Kim et al. 2021; Schmidlin et al. 2008).

Grape (Vitis spp.) belongs to Vitaceae, and the genus (Vitis) is classified mainly into two subgenera, Euvitis (2n = 38) and Muscadinia (2n = 40) (Reisch et al. 2012). Ampelopsis (porcelain berry), a perennial creeping vine belonging to Vitaceae, has the same number of chromosomes as the subgenus Muscadania (Karkamkar et al. 2010). There are approximately 95 species of Ampelopsis that are mainly native to Northeast Asia, such as Korea, Japan, and China, where they grow fast in warm climates (Emerine 2011; Ōi et al. 1965; Soejima and Wen 2006).

All the organs of Ampelopsis, such as leaves, fruits, stems, and roots, have medicinal use because of their excellent effect in recovering liver function and reducing diseases in human bodies (Rhim and Choi 2010). In addition, Ampelopsis has antioxidant effects, such as antibacterial and anti-inflammatory effects, and shows resistance to powdery mildew (Erysiphe necator) (Cadle-Davidson et al. 2011; Chang et al. 2017; Choi and Rhim 2010; Choi et al. 2013).

The skin of Ampelopsis mature berries is pale-yellow, which is different from the dark purple skins of grapes. During ripening, the skin of porcelain berry develops from green to purple, blue, sky blue, and finally white. The polyphenol content of the berries of wild grapes native to Korea, such as V. amurensis, V. coignetiae, V. flexuosa, and V. thunbergia, have been studied (Ahn et al. 2015b; Kim et al. 2006; Kwon et al. 2019; Sim et al. 2019). On the other hand, there are few reports on the polyphenol content in berries and genetic characteristics in accessions of Ampelopsis with various skin colors. Therefore, this study examined the expression of genes related to stilbene compound synthesis, skin coloration, and scavenging of reactive oxygen species (ROS) in Ampelopsis. In addition, the flavonoids, anthocyanin, and resveratrol contents during the ripening stages were analyzed.

2 Materials and methods

2.1 Plant materials and extracts

The leaves and berries of 14 Ampelopsis accessions, V. coignetiae ‘VC-1’, and V. labruscana ‘Super Hamburg’ (SH) grown in germplasm vineyards at Yeungnam University were used in this study. The leaves of Ampelopsis were harvested in May, and the grapes were harvested at several ripening stages from July to September (Fig. 1). Samples were stored frozen at − 80 °C until analyzed.

Fig. 1
figure 1

Coloration of grape berries in the developmental stages. YB: young berry, RB: ripening berries

The methanol extracts of berry skins were used to measure the flavonoid and resveratrol contents. One gram (fresh weight) of skin samples was homogenized and extracted in 4 mL of 80% methanol for three minutes in the dark. After centrifugation at 25,000 × g for 20 min, the supernatant was filtered using a 0.45-µm syringe filter (PTFE filter media, Whatman, USA) and stored at − 20 °C until analyzed (Kwon et al. 2019).

2.2 Measurement of flavonoid and anthocyanin contents

The flavonoid content was analyzed following the method of Moreno et al. (2000) with slight modification. One hundred microliters of 10% aluminum nitrate, 100 μL of 1 M potassium acetate, and 4.3 mL of ethanol were added to 1 mL of the sample extract and left at room temperature for 50 min. The absorbance was measured at 415 nm using a UV spectrophotometer (Multimode Microplate Reader, SPARK, TECAN). A standard curve was drawn using quercetin (Sigma, St. Louis, MO, USA), and the total flavonoid content was calculated from the standard curve. The total flavonoid content was indicated as mg quercetin equivalent (QE) per g of fresh weight (FW).

Anthocyanin content was analyzed by spectrophotometry (Jang et al. 2006; Kwon 2012). The skin (1 g, FW) was sonicated in 20 mL of 1% HCl-methanol for 30 min and incubated at room temperature for 2 h. The extract was filtered using filter paper (Whatman. No. 2) and 5 mL of methanol was added to the final extract. The extract was then filtered through a 0.45-µm syringe filter (PTFE filter media, Whatman, USA), and its absorbance was measured at 530 nm with a UV spectrophotometer. The total anthocyanin content was calculated as follows (Jang et al. 2006):

$${\text{Total}}\;{\text{anthocyanin}}\;{\text{content }}\left( {\% ,\;{\text{w}}/{\text{w}}} \right) \, = {\text{ OD}} \times {1}00/{\text{W}} \times { 1}00 \, \times { 1}/{65}.{1}$$

(OD: absorbance, W: sample amount (mg), 100: dose 65.1: absorbance coefficient).

2.3 Measurement of resveratrol contents

The absorbance of the 80% methanol extract from the sample was measured at 310 nm using a UV spectrophotometer (Multimode Microplate Reader, SPARK, TECAN). The resveratrol content was calculated from a standard curve generated using 1–50 μg resveratrol.

2.4 RNA isolation and reverse transcription quantitative PCR (RT-qPCR)

RNA was isolated from the leaves, young berries, ripe berries, and skin of Ampelopsis accessions, ‘VC-1’, and ‘Super Hamburg’ according to the method described by Chang et al. (1993). A Nano Drop spectrophotometer (NABI, UV spectrophotometer, Korea) was used to measure the RNA concentration of 500 ng μL−1. cDNA was synthesized from 500 ng RNA using the GoScriptTM Reverse Transcription System (Promega, Madison, USA). qPCR (CFX96™ Real-Time System, BioRad, Foster City, CA, USA) was performed using the cDNA obtained as the template and SYBR Premix Ex Taq (TaKaRa Bio Inc., Osaka, Japan). The reactions were subjected to one cycle of 95 ℃ for 30 s, followed by 40 cycles of 95 ℃ for 5 s and 60 ℃ for 30 s. The beta-actin gene (AB372563) was used as a control, and the transcript levels were calculated using a standard curve. Table 1 lists the nucleotide sequences of the primers used for the expression of the genes related to the synthesis of stilbene compounds, coloration, and reactive oxygen. Gene-specific primers were designed using Primer3 (http://frodo.wi.mit.edu/prime r3). All qPCR experiments were performed in triplicate.

Table 1 Primer sequences of genes for RT-qPCR

2.5 Statistical analysis

Statistical analysis was performed using the SPSS program (ANOVA, Duncan’s multiple range test). Statistical analysis of the expression of the related genes was carried out; p values < 0.05 were considered significant.

3 Results and discussion

3.1 Flavonoid and anthocyanin contents in the skin of grape berries at different ripening stages

The total flavonoid content in Ampelopsis accessions was analyzed and compared with the contents of two grape strains during ripening (Fig. 2A). There were no significant differences in skin flavonoid content among the Ampelopsis accessions (1.35–2.32 mg QE g−1) and other grapes strains (1.09–2.52 mg QE g−1) at the young berry developmental stage. However, as ripening developed, differences were observed between the Ampelopsis accessions and grape strains, and the skin color of berries changed. In the ripe berry 1 (RB1) stage, Ampelopsis accessions had purple skin and a flavonoid content of 8.07–9.67 mg QE g−1, while the flavonoid contents of ‘VC-1’ (V. coignetiae) and 'Super Hamburg' were 1.66 and 3.29 mg QE g−1, respectively. There were significant difference in flavonoid content among Ampelopsis accessions and grape strains in the RB1 stage. All Ampelopsis accessions showed a higher flavonoid content than the two grape strains in the RB2–RB3 stages. In the RB2 stage, the Ampelopsis accession 'YG10075' showed the highest flavonoid content (9.54 mg QE g−1), while the lowest flavonoid content (2.23 mg QE g−1) was observed in 'VC-1'. In the RB3 stage, however, there were no significant differences in flavonoid content between Ampelopsis accessions and 'VC-1', which showed the lowest content (4.34 mg QE g−1). In the final ripening stage, RB4, although 'VC-1' showed the highest berry flavonoid content (6.45 mg QE g−1), there were no significant differences between Ampelopsis accessions and grape strains (5.04 mg to 6.45 mg QE g−1) except for the Ampelopsis accession 'YG10062' (3.29 mg QE g−1). In the RB1 to RB2 stages, the berry skin of Ampelopsis accessions was purple and blue colored. On the other hand, the skin of 'VC-1' and 'Super Hamburg' was light pink and reddish colored. The predicted flavonoid content in Ampelopsis accessions was higher in the early stage due to the change in berry skin color from purple to light yellow during ripening. In addition, despite the white berry skin color of Ampelopsis accessions, there were no significant differences in the flavonoid content between Ampelopsis accessions and 'VC-1' or 'Super Hamburg', which have black-purple berries in the final ripening stage. The total berry flavonoid content of all development stages was higher in Ampelopsis accessions than in 'VC-1' and 'Super Hamburg'. The Ampelopsis accession with the highest flavonoid content (9.67 mg QE g−1) was 'YG10075'.

Fig. 2
figure 2

Total flavonoid (TF) content a, total anthocyanin (TA) content b in the skin of berries at different developmental stages for Ampelopsis spp. ‘YG10042’, ‘YG10075’, and ‘YG10062’, as well as Vitis coignetiae ‘VC-1’ and V. labruscana ‘Super Hamburg’ (SH). YB: young berries, RB: ripening berries at four different coloration stages. Vertical bars represent the standard error of the means (n = 3)

The total anthocyanin content in the Ampelopsis accessions and grape strains (‘VC-1’ and ‘Super Hamburg’) in the ripening stages was compared (Fig. 2B). In the YB stage, there were no significant differences in anthocyanin content among Ampelopsis accessions, 'VC-1' and 'Super Hamburg' (0.027–0.033%). However, from the RB2 stage, there was a dramatic increase in the anthocyanin content in 'VC-1' and 'Super Hamburg', which have a dark purple skin color, unlike that in Ampelopsis accessions. The anthocyanin content in Ampelopsis accessions throughout ripening was 0.027–0.266%, while the flavonoid content in 'VC-1' reached 1.16% in the RB4 stage. Although there were no significant differences in flavonoid content among Ampelopsis accessions in the final ripening stage (RB4), there was variation in the anthocyanin content among the Ampelopsis accessions, with contents of 0.266% in 'YG10062', 0.221% in 'YG10075', and 0.157% in 'YG10042' in the RB1 to RB2 stages.

A difference was noted in the flavonoid and anthocyanin contents of V. vinifera ‘Merlot’ and ‘Vranec’, which are red varieties, and ‘Smederevka’ and ‘Chardonnay’, which are white varieties (Ivanova et al. 2011). Similarly, there was a difference between Ampelopsis (white) and grapes (black-purple) in this study. Yang et al. (2009) found a flavonoid content of 0.97–1.67 mg QE g−1 (FW) and anthocyanin content of 1.01–2.39 mg QE g−1 (FW) in Vitis hybrid skins (dark purple). The anthocyanin contents of ‘Sheridan’ and ‘Muscat Bailey A’ (MBA) was 1.6% and 1.2%, respectively (Jang et al. 2006). The results from this study are consistent with previous reports.

Katalinić et al. (2010) reported that the flavonoid content of a red grape variety (V. vinifera) was 2.99 mg QE g−1 (FW), which was lower than the results from this study. Kwon et al. (2019) and Zeghad et al. (2019) showed that the flavonoid contents of V. vinifera and V. coignetiae were 9.5–14.37 mg QE g−1 (FW), which was 1.6–4.4 times higher than the results from the present study. The anthocyanin content in the skins of wild grapes during the harvest season was 0.2–2.96 mg g−1 depending on the genetic background and environmental conditions (Revilla et al. 2010). The flavonoid content in the leaves of A. megalophylla was also different depending on the harvest time (Yang et al. 2019). Kim et al. (2019) and Kwon et al. (2019) reported that the variation in the flavonoid and anthocyanin contents in grapes resulted from differences in the fruit growing region, environment, developmental stages, and genetic characteristics.

3.2 Resveratrol content in the skin of grape berries at different developmental stages

The total resveratrol content in Ampelopsis accessions was analyzed and compared with that in grape strains (‘VC-1’ and ‘Super Hamburg’) according to berry development and ripening stages (Fig. 3B). In the YB stage, the highest resveratrol content (30.3 μg g−1) was found in ‘Super Hamburg’; however, there was no significant difference between ‘Super Hamburg’ and ‘VC-1’. In the RB1 to RB3 stages, the highest resveratrol content was found in 'Super Hamburg'. ‘VC-1’ had the lowest resveratrol content until the RB2 stage; however, there was a sharp increase in resveratrol content in ‘VC-1’ from the RB3 stage, and it had the highest content at the RB4 stage (70.4 μg g−1). Ampelopsis accessions generally had slightly lower resveratrol contents than the grape strains. Among the Ampelopsis accessions, the highest content (48.5 μg g−1) was found in 'YG10075'.

Fig. 3
figure 3

Determination of the ultraviolet spectrum of resveratrol a and total resveratrol content in the skin of berries at different developmental stages for Ampelopsis spp. ‘YG10042’, ‘YG10075’, and ‘YG10062’ as well as Vitis coignetiae ‘VC-1’ and V. labruscana ‘Super Hamburg’ (SH) b. YB, young berry; RB, ripening berries at four different coloration stages. The vertical bars represent the standard error of the means (n = 3)

Resveratrol, the most representative compound among the stilbenes, has been found in 72 plants (Acquaviva et al. 2002) and exists in trans- and cis- forms. Trans-resveratrol shows high biological activity (Kiselev 2011). Resveratrol is abundant in red wine (Roldán et al. 2003). Similarly, the results of this study showed a higher content in 'VC-1' and 'Super Hamburg', which have dark purple skin. The resveratrol content was less than 10 µg g−1 (FW) in V. flexuosa, V. coignetiae, and ‘Gaeryangmeoru’ (Vitis sp.) (Kwon et al. 2019). The resveratrol content was 10 µg g−1 (FW) at the final harvest stage in V. quinquangularis, which is a type of Chinese wild grape (Li et al. 2017). Kwon (2012) reported a significant variation in resveratrol content at berry developmental stages, which showed an approximately 32-fold difference in the young berry period and more than a 14-fold difference in the harvesting period compared to the content in this study.

On the other hand, the resveratrol content was 20–500 µg g–1 (FW) in the berry skins of 21 Vitis red varieties (Vincenzi et al. 2013) and was 264–507 µg g−1 in the leaves of Ampelopsis sinica (Miq.) (Jin et al. 2014). As mentioned previously, the variation in the resveratrol content resulted from the differences in genetic background, developmental stages, and the change in environmental conditions for vine and fruit growing.

3.3 Analysis of gene expression related to berry ripening

RT-qPCR was performed for the genes encoding phenylalanine ammonia-lyase (PAL), chalcone synthase (CHS), chalcone isomerase (CHI), flavonoid-3′-hydroxylase (F3H), and dihydroflavonol reduction (DFR), which are involved in the coloration of berry skins (Fig. 4). Expression of the PAL gene was generally higher in the Ampelopsis accessions than in 'VC-1' and 'Super Hamburg'. In the RB4 stage, however, the highest expression was observed in 'YG10062', followed by 'VC-1', and the lowest expression was noted in 'Super Hamburg'. Expression of the CHS gene was high in Ampelopsis accessions compared to in grape strains at the RB1 stage, and the highest expression was observed in 'YG10062'. On the other hand, the expression of CHS decreased in the Ampelopsis accessions while it increased in 'VC-1' and 'Super Hamburg' from the RB2 stage to the RB4 stage. Expression of the CHI, F3’H, and DFR genes decreased after the increase in the RB1 stage in the Ampelopsis accessions, while it showed a continuously increasing pattern in ‘VC-1’ and ‘Super Hamburg’.

Fig. 4
figure 4

Expression of the genes related to skin coloration in berries at different developmental stages for Ampelopsis spp. ‘YG10042’, ‘YG10075’, and ‘YG10062’ as well as Vitis coignetiae ‘VC-1’ and V. labruscana ‘Super Hamburg’. YB: young berry, RB: ripening berries at four different coloration stages. PAL, phenylalanine ammonia-lyase; CHS, chalcone synthase; CHI, chalcone isomerase; F3’H, flavonoid-3′-hydroxylase; and DFR, dihydroflavonol reduction. Vertical bars represent the standard error of the means (n = 3)

PAL is involved in the phenylpropanoid pathway. PAL induces the accumulation of various secondary metabolites, including flavonoid and stilbene compounds, in plants (Chen et al. 2006; Huang et al. 2010). CHS is involved in the first step of chalcone and flavonoid synthesis through the phenylpropanoid pathway (Flamini et al. 2013). Then, CHI catalyzes the stereospecific cyclization of chalcone to flavonone (Forkmann and Heller 1999; Luan et al. 2013; Wang et al. 2012). F3'H catalyzes hydroxylation at the 3' position of the naringenin structure formed by CHI to generate the eridictyol structure (Bogs et al. 2006; Flamini et al. 2013; Li et al. 2017). DFR catalyzes the reduction of anthocyanin-based compounds from dihydroflavonols to leucoanthocyanidins (2010a; b; Sparvoli et al.1994). The expression of the PAL gene was reported to increase as ripening progressed in red grapes (Boss et al. 1996a; Lijavetzky et al. 2012), and the DFR gene was induced in petunia using CaMV, resulting in a change in petal color by the anthocyanin content (Chu et al. 2015). Genes, such as PAL, CHS, CHI, F3'H, and DFR are involved in flavonoid accumulation in the petal tissue (Van der Krol et al. 1990) as well as grape berry skin pigmentation (Castellarin et al. 2006; Flamini et al. 2013; Katayama-Ikegami et al. 2016; Sparvoli et al. 1994). The increase in PAL gene expression in the RB4 stage of Ampelopsis accessions with white berry skin color was associated with the induction of various metabolites and the induction of expression related to the berry skin color. The expression of genes including PAL, CHS, CHI, F3’H, and DFR also increased (Boss et al. 1996b; Luan et al. 2013; Ramazzotti et al. 2008) as the anthocyanin content increased during skin ripening. Boss et al. (1996b) reported a variation in the expression of the genes related to coloration between the white and red varieties of V. vinifera. Similarly, the results showed that the expression of genes increased in the RB4 stage in 'VC-1' and 'Super Hamburg', while the expression decreased after a slight increase in the RB1 stage in Ampelopsis accessions.

In terms of the expression of DFR and related genes in Ampelopsis accessions, the expression decreased significantly after an increase in the RB1 stage. Furthermore, the expression of genes related to skin coloration increased with accumulation of anthocyanin in the berries.

3.4 Expression of genes related to stilbene synthesis and ROS scavenging in Ampelopsis

RT-qPCR was performed for genes related to the synthesis of stilbene compounds, coloration, and ROS scavenging in each organ of 14 Ampelopsis accessions (Fig. 5). In the leaves, the expression of the ROMT gene was highest in ‘YG10045’ and lowest in ‘YG10075’. The STS1 gene expression was highest in ‘YG10043’, followed by in ‘YG10042’ and ‘YG10060’, while its expression was low in other accessions of Ampelopsis. Expression of the PPO gene was highest in 'YG10060' and 'YG-Songni4' but lowest in 'YG10064'. In young berries, there were no significant differences in the expression of the ROMT gene among accessions of Ampelopsis. Expression of the STS1 gene was highest in ‘YG10075’ and ‘YG10062’, with 4.5 times higher expression than that in ‘YG10060’. Expression of the PPO gene was highest in ‘YG10060’. In ripe berries, the highest expression of the ROMT gene was in 'YG10057', the highest expression of the PPO gene was in ‘Ab SV 36–1’, and the highest expression of the STS1 gene was in ‘YG10064’, with 9.8 times higher expression than that in ‘YG10075’. Overall, expression of genes associated with stilbene compound synthesis was highest in the leaves of 'YG10045', in the young berries of ‘YG10075’, and in the ripe berries of ‘YG10057’.

Fig. 5
figure 5

Expression of genes in leaves and berries of Ampelopsis accessions. a Stilbene synthesis-related genes: ROMT, resveratrol O-methyltransferase; STS1, stilbene synthase; and PPO, polyphenol oxidase. b Skin coloration-related genes: PAL, phenylalanine ammonia-lyase; CHS, chalcone synthase; CHI, chalcone isomerase; F3’H, flavonoid-3′-hydroxylase; and DFR, dihydroflavonol reductase. C Genes related to scavenging of reactive oxygen species: CAT, catalase and SOD, superoxide dismutase. The different letters indicate the significant differences according to a Duncan test (p < 0.05)

In the leaves, 'YG11030' and 'YG-Songni4' showed the highest expression of the PAL gene, followed by 'Ab SV 36–1' and 'YG10062'. High expression of the CHS gene was observed in 'YG10064' and 'YG-Songni4', and high expression of the CHI gene was observed in ‘YG10045’, and ‘YG10064’. In the young berries, high expression of the PAL gene was observed in ‘YG10075’, high expression of the CHI gene was observed in 'YG10075', and high expression of the CHS gene was observed in 'Ab SV 36–1', 'YG11030', and 'YG10075', which had approximately 6.5 times higher expression than the accessions. In ripe berries, high expression of the PAL gene was noted in 'YG10043' and 'YG-Songni4', high expression of the CHS gene was observed in ‘YG10044’, and high expression of the CHI gene was observed in 'YG-Songni4'.

Stilbene compounds are present in various plants, such as peanuts, grapevines, and pine trees (Choi 2011; Kiselev et al. 2017). Three molecules of malonyl-CoA and one molecule of p-coumaroyl-CoA are directly synthesized as resveratrol by stilbene synthase (STS) (Austin et al. 2004). STS is also involved in the formation of other stilbenes (Viret et al. 2018). For this reason, many studies on the STS gene have been conducted (He et al. 2016; Kiselev et al. 2017; Parage et al. 2012). Resveratrol, a stilbene compound, is methylated to pterostilbene by resveratrol O-methyltransferase (ROMT) (Schmidlin et al. 2008). Of resveratrol, viniferins are formed through oxidation by polyphenol oxidase (PPO) (Dry and Robinson 1994; Pezet 1998). The resveratrol content and the expression of STS, ROMT, and PPO genes increases under abiotic and biotic stress in grapes (Ahn et al. 2015a; Kiselev et al. 2017, 2019; Wang et al. 2018).

Of the signaling molecules of the defense responses, plants produce ROS when stressed by the environment, UV irradiation, and pathogens (Han et al. 2009; Hu et al. 2008; Pitzschke et al. 2006; Shah et al. 2001). ROS include radicals, such as superoxide anion (O2), hydroxyl radical (•OH), and hydrogen peroxide (H2O2) (Sharma et al. 2012). Although ROS production results from the general physical response to stress (Biswas et al. 2020), excessive ROS production causes damage to DNA and proteins (Gill and Tuteja 2010). On the other hand, ROS are important indicators of the defense response of plants caused by biotic and abiotic stress (Fujita et al. 2006). SOD and CAT scavenge ROS in the plants (Caverzan et al. 2016; Gill and Tuteja 2010; Sofo et al. 2015). SOD is involved in changing the superoxide anion (O2) to hydrogen peroxide (H2O2). CAT plays a role in changing H2O2 to water (H2O) in various metabolic processes in plants (Hasanuzzaman et al. 2012).

In Ampelopsis leaves, the highest expression of the CAT gene was observed in ‘YG10075’ and ‘YG-Songni4’ and the highest expression of the SOD gene was observed in ‘YG10045’ and ‘YG10064’. In Ampelopsis ripe berries, the highest expression of the CAT gene was observed in 'YG10044' and the highest expression of the SOD gene was observed in 'YG10044' and 'YG-Songni4'. In the leaves of Ampelopsis, the expression levels of genes associated with ROS were highest in 'YG10045' and lowest in 'YG10044', while they were highest in 'YG10075', and lowest in 'YG10043' in young berries.

As the expression patterns of investigated genes varied, even in the same grape variety (Kiselev et al. 2017), these results also showed a variation in expression depending on the accession, organ, and developmental stage of the grape and vines. There was also a difference in the expression of genes related to coloration depending on the harvesting time in the leaves of Ampelopsis megalophylla (Yang et al. 2019). To select the most meaningful Ampelopsis accession, the values were quantified by scoring on a scale of 1 to 10 points for the expression of 10 genes (Table 2). Overall, the leaves of 'YG10045' showed the best expression of genes, while the young berries of ‘YG10075’ and ripe berries of ‘YG-Songni4’ showed the best expression pattern. Based on the scoring of the characteristics of Ampelopsis accessions, 'YG-Songni4' was the most valuable genetic resource, with the best gene expression related to the coloration of the skin, stilbene compound synthesis, and ROS scavenging activity in their leaves and ripening berries.

Table 2 Comprehensive analysis for selection of Ampelopsis accessions based on expression of genes related with synthesis of stilbenic compounds

Among the wild grapes native to Korea, there are few reports investigating the characteristics of Ampelopsis as a genetic resource for grape breeding. In this study, there was genetic diversities among the Ampelopsis accessions. Moreover, the accessions of Ampelopsis were selected to be used as genetic resources in grape breeding programs. In addition to the characteristics investigated in this study, as domestic native grapes show excellent activity in human bodies (Ahn et al. 2015b; Kim et al. 2006; Kwon et al. 2019), further research on various Korean native grapevines will be needed to identify and select varieties with desirable characteristics.

4 Conclusion

This study examined the content of polyphenols and their related gene expression in the skins of Ampelopsis accessions, V. coignetiae ‘VC-1’, and V. labruscana ‘Super Hamburg’ throughout development and ripening stages. The flavonoid contents were higher in Ampelopsis accessions compared to in grape strains. Among the Ampelopsis accessions, the highest content of flavonoid was 9.67 mg QE g−1 (FW) in ‘YG10075’. ‘VC-1’ had the highest anthocyanin content with 1.2%, followed by Ampelopsis ‘YG10062’ with 0.27%. 'VC-1' had the highest resveratrol content of 70.4 μg g−1, followed by Ampelopsis 'YG10075' at 48.5 μg g−1. As ripening developed, gene expression involved in skin coloration increased in ‘VC-1’ and ‘Super Hamburg’, while it decreased in 'YG10042', 'YG10075', and 'YG10062'. Gene expression levels related to stilbene compound synthesis, skin coloration, and ROS were highest in the leaves of ‘YG10045’, the young berries of ‘YG10075’, and the ripe berries of ‘YG-Songni4’. The expression of the genes related to fruit development and ROS scavenging showed different patterns depending on the Ampelopsis Ampelopsis, organ, and developmental stage of the berries. Based on expression of genes related with stilbene synthesis, 'YG-Songni4' was identified as a useful genetic resource of Ampelopsis, which could be useful in further studies and for breeding new varieties of grapes.