Introduction

Grapes are an economically and agriculturally significant fruit, renowned for their diverse applications and health-promoting qualities [1, 2]. Among the vast array of grape cultivars, muscat and foxy and table grapes hold particular importance due to their unique flavor profiles and versatility [3]. Muscat grapes are celebrated for their intensely aromatic and sweet characteristics, making them a favored choice for winemaking, particularly in the production of dessert wines and Muscat varieties [4]. Foxy grapes, on the other hand, known as Vitis labrusca, are valued for their distinctive foxy flavor and are used extensively in the production of grape juice, jams, and jellies [5]. They have gained attention for their potential health benefits such as antioxidant properties [6]. Table grapes, namely the 'Italia' and 'Bronx Seedless' cultivars, have gained prominence not only for their exquisite taste [7], but also for their abundant phytochemical content, which encompasses phenolic acids, C13-Norisoprenoids; phytoalexins, and flavonoids. These bioactive compounds are not only responsible for the remarkable visual and gustatory appeal of grapes but have also garnered significant attention for their potential health benefits, making grapes a subject of great interest for both the health sciences and agriculture [8].

Grapes and their phytochemical composition have been extensively studied for their potential health benefits. Anthocyanins, a subgroup of flavonoids, are recognized for their role in reducing the risk of chronic diseases like cancer and heart disease [9]. They combat inflammation and oxidative stress, offering various health benefits when included in one's diet. Flavonoids and phenolic acids, including flavonols, flavones, anthocyanidins, and hydroxycinnamic acids, provide vibrant colors to grapes and add value due to their antioxidant properties. C13-Norisoprenoids such as damassene, β-Geranyl acetone, and β-Ionone contribute significantly to the distinctive aroma of grapes and grape-based products, notably wines, enhancing the sensory experience of consumers [9]. Phytoalexins, such as pterostilbene, piceid, and resveratrol are known for defending grapes against pathogens, activated in response to stress factors like UV radiation and fungal infections [10]. Research have noted the potential of resveratrol in conferring cardiovascular benefits, as well as its anti-inflammatory and antioxidant properties [11]. Although the health benefits of these compounds are well-documented, understanding the variation in the phytochemicals of grapes across different berry development stages is essential for several reasons [7]. The specific mix of phytochemicals in grapes may vary considerably owing to factors namely grape cultivar, environmental conditions, cultural treatments, and maturation stage [12]. This variation can significantly impact grape quality, flavor, and, by extension, the quality of grape-derived products like wine [13, 14]. Additionally, it is crucial for grape growers to optimize their production methods to ensure the highest possible levels of phytochemicals, thereby enhancing both the quality of their grapes and the potential health benefits for consumers [8].

'Italia' and 'Bronx Seedless' grape cultivars, esteemed for their succulent sweetness and versatility, have captivated the palates of many. However, while extensive research has explored various aspects of these grape varieties, there remains a conspicuous dearth of studies concerning the development of tannins within them. Phytoalexins, flavonoids, phenolic acids, C13-Norisoprenoids and anthocyanin, known for their significant role in taste, mouthfeel, and potential health benefits, are essential components of grape chemistry. Yet, surprisingly little is known about how tannin content evolves throughout the berry growth stages such as BBCH-77, BBCH-79, BBCH-81, BBCH-83, BBCH-85, and BBCH-89 of 'Italia' and 'Bronx Seedless' grapes. This paper investigates to address this research gap by delving into the intricate process of tannin development in these grape cultivars. It indicates the critical importance of understanding tannin dynamics across developmental stages, shedding light on their contributions to flavor and health-related attributes in these grapes.

Materials and methods

Plant material

The research was conducted using 'Italia' and 'Bronx Seedless' grape cultivars, which were 20 years old and grafted onto 5 BB rootstock. The study site was the Manisa Viticulture Research Institute in Türkiye, located at coordinates 27°23′57.36′′ E.L. and 38° 37′ 57.14′′ N.L. The vines were strategically arranged in a 2 × 3 m layout, utilizing a high trunk cordon trellis system. These spur-pruned cultivars typically exhibited 12–15 shoots per vine. The experiment was conducted following a randomized complete block design, incorporating three replicates, each consisting of six vines. For each treatment, eighteen basal clusters (two clusters per vine) were randomly selected, collected, and immediately weighed. Sample collections were executed randomly from the upper, middle, and lower sections of the grape clusters. Clusters were harvested six times between July 27 (referred to as BBCH-77, signifying the week preceding veraison) and August 28 (BBCH-89, the harvest period). For this study, Period I: Berries beginning to touch, Period II: Majority of berries touching, principal growth stage Period III: Ripening of berries Period IV:Beginning of ripening: berries begin to develop variety-specific colour, Period V: Berries developing colour, softening of berries, Period VI:Berries ripe for harvest (BBCH-77, BBCH-79, BBCH-81, BBCH-83, BBCH-85, and BBCH-89), corresponded to distinct developmental stages in berry maturation, including the commencement of berry touch, culmination of berry touch, initiation of berry color alteration, marked berry color transformation, berry softening, and optimal ripeness for harvest. To ensure standardized sampling, the protocol established by Lorenz et al. [15] was followed. Subsequently, the harvested grape clusters were promptly stored at 4 °C in the laboratory and subsequently preserved at − 80 °C for subsequent analytical procedures.

Determination of onological parameters in grape berry

A sample of 25 berries from each marked bunch was analyzed for total acidity and soluble solids, employing the methodology delineated by Kaya [14]. The berries were manually pressed, and the resulting musts were promptly assessed for total acidity and soluble solids content, measured in degrees oBrix. Soluble solids concentration was determined using a refractometer (Atago PR32, Tokyo, Japan) at a temperature of 22 °C, with the outcomes reported in degrees Brix. Total acidity was quantified through titration with 0.1 N NaOH until a pH of 8.2 was reached, with the results expressed in milliequivalents per liter (meq/l).

Determination of phytoalexins in grape berry

Phytoalexins were determined by slightly modifying the method developed by Jeandet et al. [16]. The process began by grinding the samples in a mortar with sand, utilizing 30 mL of a methanol–water mixture (8:2, v/v). Following the initial extraction, centrifugation was performed at 10,000 g for 15 min, separating the supernatant for further processing. Subsequently, the supernatant was subjected to a purification step using a Sep-Pak C18 cartridge (Waters, Milford, MA). The elution was accomplished using 30 mL of a methanol–water mixture (8:2, v/v), and the resulting eluate was evaporated to dryness while maintaining the temperature below 40 °C. To prepare the berry extracts for analysis, the dried eluate was reconstituted in 10 mL of methanol per gram of fresh weight and then meticulously filtered. During High-Performance Liquid Chromatography (HPLC) analysis, precisely 10 μL of each sample (equivalent to 1 mg of fresh weight) was injected into the system.

Determination of flavonoids in grape berry

The analysis of flavonoids from berry followed a method as detailed by Brossa et al. [17] with specific modifications. To summarize, 10 μL of each meticulously filtered extract was introduced into the Liquid Chromatography (LC) system, specifically the Acquity UPLC by Waters (Milford, MA, USA). Chromatographic separation was performed using a C18 column (3.5 μm; 100 × 2.1 Waters, Milford, MA, USA) maintained at a constant temperature of 42 °C and a flow rate of 0.3 mL/min. For the detection and quantification of flavonoids, the Electrospray Ionization (ESI) mode was employed, with positive ionization (UPLC-(ESI +)-MS/MS) for anthocyanins and negative ionization (UPLC-(ESI-)-MS/MS) for all other flavonoids. Spectra were recorded within the range of 100 to 900 atomic mass units per charge (amu/z). Additionally, UV/visible detection was conducted at 365 nm for flavonols and 520 nm for anthocyanins. To determine the concentrations of flavonols, proanthocyanidins, and anthocyanins, a comparative approach with external standards was utilized. Rutin (Quercetin-rutinoside), Quercetin (Quercetin-rhamnoside), Isoquercitin (Quercetin glucoside), and cyanidin-3-O-glucoside, acquired from Sigma–Aldrich Fluka (Madrid, Spain), along with malvidin-3-O-glucoside (Oenin), Isorhamnetin glucoside, and Procyanidin B1 from Extrasynthese (Genay, France) were used as calibration standards.

Determination of phenolic acids in grape berry

Phenolic acids were determined by slightly modifying the method developed by Pantelić et al. [18]. The quantification of polyphenols involved the utilization of advanced analytical instrumentation and precise separation techniques. An ultra-high performance liquid chromatography (UHPLC) system, specifically the Dionex Ultimate 3000 UHPLC equipped with a diode array detector (DAD), in conjunction with a mass spectrometer (TSQ Quantum Access Max triple-quadrupole with heated electrospray ionization (HESI)) by ThermoFisher Scientific (Basel, Switzerland), was employed for this purpose. The analytical separation was achieved using a Syncronis C18 column from ThermoFisher Scientific, located in Bremen, Germany. The mobile phase for gradient elution consisted of two components: a 0.1% aqueous formic acid solution (A) and 100% acetonitrile (B). The elution process followed a specific gradient pattern: starting at 5% B for 2.0 min, it then increased from 5 to 95% B over the course of the next 10 min (from 2.0 to 12.0 min). Subsequently, a swift transition occurred from 95 to 5% B in 0.1 min (from 12.0 to 12.1 min), and the mobile phase was maintained at 5% B for the subsequent 3 min. Each individual phenolic compound was quantified through a direct comparison with commercially available standards. The outcomes of the analysis were expressed as concentrations in milligrams per kilogram (mg/kg) of berry samples.

Determination of C13-Norisoprenoids in grape berry

C13-Norisoprenoids were determined by slightly modifying the method developed by Bindon et al. [19]. A highly specialized solid-phase microextraction (SPME) technique was devised for the meticulous analysis of C13-norisoprenoids in grape samples. The procedure commenced with the placement of a 2-g sample of homogenized grape berries into a 10 mL glass vial. These berry homogenates were then judiciously diluted with 500 μL of deionized water to ensure the formation of a homogenate suspension, followed by manual pH adjustment to pH 1. To facilitate accurate quantification, deuterated analogues of C13-Norisoprenoids were introduced as internal standards. These deuterated standards, specifically [2H3] β-ionone and [2H4] β-damascenone for β-damascenone, and d8-naphthalene for TDN, were sourced from the Australian Wine Research Institute (AWRI). Subsequently, 10 μL of a 5 μg/mL solution of the deuterated standard mix was added to the grape samples. The vials containing the samples were then hermetically sealed with PTFE/silicone septa (Supelco). The sealed vials were transferred to a heating block for 1 h at 100 °C and subsequently cooled to 40 °C. The SPME sampling process was carried out manually using a 65 μm polydimethylsiloxane/divinylbenzene (PDMS/DVB) fiber (Supelco). The fiber was exposed at a constant level above the sample surface for 1 h. For the subsequent SPME–GC–MS analysis, a Hewlett-Packard HP6890 gas chromatograph equipped with a 30-m fused silica SGE BP20 column (0.25 mm i.d. and 0.25 μm film thickness) was employed. The SPME fiber containing adsorbed headspace volatiles was manually transferred to the GC injection port for 1 min. The splitless/split injection port was heated to 220 °C. Ultrahigh-purity helium served as the carrier gas, maintaining a constant flow rate of 1 mL/min, with a column head pressure of 6.98 psi. The temperature program was initiated at 40 °C for 4 min and was then ramped up to 240 °C at a rate of 12 °C/min. The temperature was held at 240 °C for 5 min, resulting in a total runtime of 25.67 min. The GC instrument was seamlessly linked to a Hewlett-Packard HP5972 mass selective detector. For precise quantitation of C13-Norisoprenoids using selective ion monitoring, specific mass-to-charge ratios (m/z) were used, including m/z 69 for β-damascenone, m/z 73 for [2H4]β-damascenone, m/z 177 for β-ionone, m/z 180 for [2H3]β-ionone, m/z 136 for TDN, and m/z 157 for d8-naphthalene.

Determination of anthocyanin in grape berry

Anthocyanins were determined by slightly modifying the method developed by Yousef et al. [20]. The analysis of anthocyanins within grape berry was executed with a high-performance liquid chromatography (HPLC) system, specifically the Agilent 1100 Series system (Agilent, Waldbronn, Germany). This system was equipped with a diode array detector (DAD) (G1315B) and coupled to an LC/MSD Trap VL (G2445C VL) ESI–MS/MS system. Data processing was managed through an Agilent ChemStation (version B.01.03) data-processing station. Mass spectral data were further processed using the Agilent LC/MS Trap software (version 5.3). Anthocyanin extracts, in volumes of 10 μL, were meticulously injected into a reversed-phase column, namely the Zorbax Eclipse XDB-C18 (2.1 mm × 150 mm; 3.5 μm particle; Agilent, Germany). The column temperature was precisely maintained at 40 °C. The separation method employed was based on a previously established technique, with slight adjustments to enhance the separation of the detected anthocyanins. The mobile phase utilized for separation consisted of a mixture of water, acetonitrile, and formic acid, with two solvent compositions: 88.5% water, 3% acetonitrile, and 8.5% formic acid (solvent A); and 41.5% water, 50% acetonitrile, and 8.5% formic acid (solvent B). The separation gradient employed was as follows: 97% A and 3% B for 8 min; 70% A and 30% B over 20 min; 50% A and 50% B within 6 min; 0% A and 100% B over 4 min, maintained for 2 min, followed by a return to initial conditions in 6 min. A conditioning period of 8 min between injections was incorporated. The flow rate was set at 0.19 mL/min.

Statistical analysis

In this study, all descriptive and inferential analyses were conducted using the statistical package integrated within R Studio. To comprehensively assess the influence of cultivar (two levels), phenological stage (six levels), and their potential interactions on the levels of phytoalexins, flavonoids, phenolic acids, and C13-Norisoprenoids and anthocyanins, a rigorous analysis of variance (ANOVA) was executed, again leveraging the capabilities of the statistical package within R Studio. The statistical model encompassed all main effects and interaction effects, with a subsequent assessment of the data for conformity with normality assumptions. Four distinct models were meticulously constructed to evaluate the primary effects of both cultivar and phenological stages on the concentrations of phytoalexins, flavonoids, phenolic acids, and C13-Norisoprenoids and anthocyanins. In instances where statistical significance was identified through ANOVA, post hoc analysis was conducted using Tukey's test, a widely recognized and applied method for the comprehensive exploration of differences among multiple groups. To gain further insights and visualize the interrelationships among various variables, Principal Component Analyses (PCAs) were performed. This analytical technique, employed for phytoalexins, flavonoids, phenolic acids, and C13-Norisoprenoids and anthocyanins, was executed using ggplot2 within R Studio. The application of PCA aids in the reduction of multidimensional data into a more interpretable form, allowing for the identification of underlying patterns and trends within complex datasets.

Results

The weight of the 'Italia' grapes exhibited a consistent increase across all periods, starting from 0.70 g/berry in Period I to 8.14 g/berry in Period VI. In contrast, the 'Bronx Seedless' cultivar had a lighter weight, beginning at 0.38 g/berry in Period I and culminating at 3.61 g/berry in Period VI. The width and length of the 'Italia' grapes were consistently greater than the 'Bronx Seedless' cultivar across all periods. In Period I, the width for 'Italia' was 12.27 mm, while for 'Bronx Seedless', it was 8.97 mm. By Period VI, the width for 'Italia' grapes had increased to 21.62 mm compared to 16.48 mm for 'Bronx Seedless'. The length of the berries showcased a similar trend. The TSS values, indicative of the sugar content, displayed an increasing trend for both cultivars. 'Italia' grapes exhibited a rise from 2.90 in Period I to 16.80 in Period VI. The 'Bronx Seedless' cultivar followed a similar trend but with slightly lower values, from 2.80 in Period I to 17.40 in Period VI. 'Italia' grapes began with a higher TA value of 34.44 g/L in Period I, which declined to 6.51 g/L by Period VI. In contrast, 'Bronx Seedless' started at 29.52 g/L and decreased to 5.39 g/L over the same periods. The maturity index, which is a ratio of TSS to TA, showed an upward trend for both grape cultivars. The 'Italia' grapes exhibited an increase from 0.84 in Period I to 25.81 in Period VI. Similarly, 'Bronx Seedless' grapes displayed a rise from 0.98 to 32.28 over the respective periods (Fig. 1).

Fig. 1
figure 1

Berry weight, width and length,TSS,TA and maturity index values of table grapes ('Italia' and 'Bronx Seedless') harvested in Periods I, II, III, IV, V, and VI. (Period I: Berries beginning to touch, Period II: Majority of berries touching, Principal growth stage Period III: Ripening of berries Period IV:Beginning of ripening: berries begin to develop variety-specific colour, Period V: Berries developing colour, 85: Softening of berries, Period VI:Berries ripe for harvest)

Full size image

The contents of various phytoalexins, flavonoids, phenolic acids, and C13-Norisoprenoids in the table grapes from the 'Italia' and 'Bronx Seedless' cultivars, harvested across six periods (I to VI), are detailed in Table 1. The impact of the cultivar on the contents of these compounds was highly significant for most of the analyzed compounds (p < 0.001). Sampling time also exhibited a profound influence on the concentrations of these phytochemicals, with most showing a highly significant variation (p < 0.001). In the analysis of phytochemical compositions across different grape cultivars, 'Italia' and 'Bronx Seedless' displayed distinct profiles. The 'Italia' cultivar exhibited higher concentrations of resveratrol (8.86 µg/g), pterostilbene (1.71 µg/g), piceid (3.06 µg/g), viniferin (6.30 µg/g), anthocyanidins (63.5 mg/100 g), flavonols (18.8 mg/100 g), flavonoids (6.30 mg/100 g), flavanones (18.4 mg/100 g), catechins (5.24 mg/100 g), hydroxycinnamic acids (14.8 mg/100 g), hydroxybenzoic acids (3.32 mg/100 g), β-damascenone (26.8 µg/L), geranyl acetone (108.6 µg/L), and β-ionone (27.2 µg/L). Conversely, 'Bronx Seedless' was characterized by slightly lower levels of these compounds, with notable concentrations of resveratrol (8.20 µg/g), pterostilbene (1.34 µg/g), piceid (3.64 µg/g), viniferin (5.45 µg/g), anthocyanidins (54.8 mg/100 g), flavonols (16.2 mg/100 g), flavones (5.45 mg/100 g), flavanones (16.0 mg/100 g), catechins (4.68 mg/100 g), hydroxycinnamic acids (14.7 mg/100 g), hydroxybenzoic acids (2.67 mg/100 g), β-damascenone (26.5 µg/L), geranyl acetone (91.7 µg/L), and β-ionone (26.2 µg/L). The statistical tests (ST) conducted on the data indicated the significance of these changes, with p-values less than 0.05 across all compounds, many of which were much lower, indicating strong statistical significance. Particularly, resveratrol, piceid, viniferin, anthocyanidins, flavonols, and β-ionone presented p-values of less than 2e-16, highlighting the robustness of the observed trends over time. The longitudinal study on cultivar revealed significant variations in phytochemical content over six distinct periods, indicating progressive increases in the concentrations of various compounds. The data presents a clear trend of augmentation across all measured phytochemicals from Period I to Period VI. Specifically, resveratrol concentrations escalated from 5.96 µg/g in Period I to 11.59 µg/g in Period VI, marking a notable increase in its presence. Similarly, pterostilbene rose from 1.15 to 1.96 µg/g, and piceid from 2.49 to 4.35 µg/g over the same intervals. Viniferin, anthocyanidins, flavonols, flavones, flavanones, catechins, hydroxycinnamic acids, and hydroxybenzoic acids also demonstrated significant increases, with anthocyanidins notably rising from 44.6 mg/100 g in Period I to 75.9 mg/100 g in Period VI. The trend continued with the volatile compounds β-damascenone, geranyl acetone, and β-ionone, which also showed substantial increases in concentration through the periods. β-damascenone levels increased from 20.0 to 34.2 µg/L, and geranyl acetone from 75.5 L to 128.4 µg/L (Table 1).

Table 1 Phytoalexins, flavonoids, phenolic acids, C13-norisoprenoids content (mg/kg) of table grapes ('Italia' and 'Bronx Seedless') harvested in Periods I, II, III, IV, V, and VI
Full size table

In this study, the statistical analysis (C) reinforced the significance of these differences, with p-values indicating a highly significant difference for most anthocyanins, except for peonidin-3-O-glycoside and malvidin-3-O-glycoside-acetyl where the differences were not statistically significant. Specifically, delphinidin-3-O-glycoside, cyanidin-3-O-glycoside, petunidin-3-O-glycoside, and malvidin-3-O-glycoside-p-coumaryl showed extremely significant differences (p < 0.05), highlighting the distinct anthocyanin profiles between 'Italia' and 'Bronx Seedless' grapes. The comparative analysis of anthocyanin profiles in 'Italia' and 'Bronx Seedless' grape cultivars indicated significant variances in the concentrations of specific anthocyanins, with 'Italia' generally displaying higher levels. Delphinidin-3-O-glycoside concentrations were found to be higher in 'Italia' (5.84 mg/L) compared to 'Bronx Seedless' (4.08 mg/L). Similarly, cyanidin-3-O-glycoside and petunidin-3-O-glycoside levels were also greater in 'Italia', recorded at 3.18 mg/L and 9.67 mg/L respectively, against 2.63 mg/L and 8.15 mg/L in 'Bronx Seedless'. While the concentration of peonidin-3-O-glycoside showed no significant difference between the two cultivars, malvidin-3-O-glycoside, a prominent anthocyanin, was notably higher in 'Italia' (12.3 mg/L) compared to 'Bronx Seedless' (10.3 mg/L). Moreover, 'Italia' exhibited a significantly higher concentration of peonidin-3-O-glycoside-acetyl (83.8 mg/L) and malvidin-3-O-glycoside-p-coumaryl (4.78 mg/L) when compared to 'Bronx Seedless', which had concentrations of 59.2 mg/L and 3.88 mg/L, respectively. However, no significant difference was observed in the levels of malvidin-3-O-glycoside-acetyl between the two cultivars. The statistical significance of these findings, supported by p-values below 0.05 for most compounds, confirmed the robustness of the observed trends, although the peonidin-3-O-glycoside and malvidin-3-O-glycoside-p-coumaryl did not show significant changes, highlighting variability in the biosynthesis and accumulation of different anthocyanins within the cultivar. The longitudinal analysis of cultivar over six periods has revealed a distinct and significant evolution in its phytochemical and anthocyanin composition, demonstrating the dynamic nature of grape biochemistry over time. Delphinidin-3-O-glycoside, cyanidin-3-O-glycoside, and petunidin-3-O-glycoside levels showed a marked increase, with concentrations rising progressively across the study periods. Delphinidin-3-O-glycoside levels increased from 3.74 mg/L in Period I to 6.36 mg/L in Period VI, and petunidin-3-O-glycoside from 6.55 to 11.66 mg/L, indicating significant growth in these anthocyanins. Conversely, peonidin-3-O-glycoside concentrations remained relatively stable. Malvidin-3-O-glycoside, a key anthocyanin, exhibited a strong upward trend, moving from 8.50 to 14.46 mg/L. Similarly, peonidin-3-O-glycoside-acetyl and malvidin-3-O-glycoside-acetyl concentrations saw significant increases, reflecting the grape's changing biochemical landscape. However, malvidin-3-O-glycoside-p-coumaryl showed a minor reduction over time (Table 2).

Table 2 Anthocyanin content (%) of table grapes ('Italia' and 'Bronx Seedless') harvested in Periods I, II, III, IV, V, and VI
Full size table

Figure 2 displays the PCA biplots of berries, segmented by cultivars. The biplots illustrate the relationships and variance among various phytochemical components such as phytalexins, flavonoids, phenolic acids, c13-norisoprenoids, and anthocyanin content. Regarding phytalexins, the first principal component (Dim1) accounts for 96.2% of the variance, and the second (Dim2) accounts for 2.7%. Pterostilbene, piceid, viniferin, and resveratrol are closely aligned with the positive end of Dim1. Their cos2 values are high, indicating a strong correlation with this principal component (Fig. 2A, B). For flavonoids, Dim1 explains 93.9% of the variance, while Dim2 explains 3.8%. The flavonoids such as anthocyanidin-flavones, flavones, flavonols, and catechins are prominently situated on the positive side of Dim1, showing a strong association. The cos2 values suggest a robust representation in the PCA (Fig. 2C, D). For anthocyanin content, Dim1 and Dim2 capture 70.8% and 18.0% of the variance, respectively. Multiple anthocyanins, such as peonidin-3-O-glycoside, delphinidin-3-O-glycoside, among others, are dispersed throughout the plot (Fig. 2E, F). Considering phenolic acids, Dim1 and Dim2 account for 70.6% and 20.9% of the variance, respectively. Hydroxybenzoic acids and hydroxycinnamic acids are the two dominant phenolic acids displayed. The former is closer to the positive side of Dim1, while the latter leans towards the positive side of Dim2 (Fig. 2G, H). Regarding C13-norisoprenoids, representing 84.1% and 14.1% of the variance, Dim1 and Dim2 show β-Damascenone and Ionone closely associated with the positive side of Dim1, while geranyl acetone shows a positive correlation with Dim2 (Fig. 2L, M).

Fig. 2
figure 2figure 2

PCAs biplot of berries colored by Cultivars. All phytoalexins, flavonoids, phenolic acids, C13-norisoprenoids and anthocyanin content are demonstrated

Full size image

Figure 3 illustrates the hierarchical clustering heatmap of grape berry components across various cultivars and periods. This heatmap demonstrates the relative concentrations of specific phytochemicals in the grape samples. The phytochemical components (X-axis) are clustered at the bottom of the heatmap. Distinct clusters emerge, suggesting some similarities and dissimilarities between these components. For instance, cyanidin-3-glycoside and malvidin-3-glycoside appear closely related, as do flavones, piceid, and catechins, implying similar patterns of concentration across samples. The grape samples (Y-axis), labeled with their respective cultivars and periods, are vertically clustered on the right. It's evident that certain samples, like the Period IV 'Italia' series and Period V 'Italia' series, exhibit similar phytochemical compositions, as indicated by their close clustering. In contrast, samples from different periods or cultivars, such as the Period I 'Bronx Seedless' series and Period VI 'Bronx Seedless' series, show more pronounced differences. Regarding Heatmap intensity, the deep red patch in the middle indicated that compounds like anthocyanin-flavones and peonidin-3-glycoside had higher concentrations in samples like Period IV 'Italia' and Period V 'Italia'. Conversely, compounds like hydroxybenzoic acids were found to have lower concentrations in these same samples, as shown by the deep blue color (Fig. 3).

Fig. 3
figure 3

Conducting a heatmap analysis that scrutinizes numerous components, from phytoalexins, flavonoids, phenolic acids, C13-norisoprenoids and anthocyanins are demonstrated

Full size image

Discussion

The findings in this study, show changes in grape weight, size, TSS, TA, and maturity index during ripening, are in complete alignment with well-established knowledge in grape production and ripening (Fig. 1). These observations correspond to previous research, affirming the consistency of the results with established patterns in viticulture. As supported by references in the field [6, 14], the increase in berry size, rising TSS values, declining acidity, and the improvement in the maturity index are all indicative of the typical ripening process. These parameters serve as crucial indicators for determining the optimal grape harvest time, a fundamental factor in ensuring the production of high-quality wines and grape-based products. The study presented provides a comprehensive analysis of the phytochemical composition of grapes from the 'Bronx Seedless' and 'Italia' cultivars harvested across six different berry development periods. The examination of various phytochemical compounds, namely C13-norisoprenoids, phenolic acids, phytoalexins, anthocyanins, and flavonoids sheds light on the intricate interplay between grape cultivars and sampling time (Tables 1 and 2). In order to better understand the implications and significance of these findings, we critically evaluated our findings by discussing their alignment and differences with previous studies. One of the most noteworthy results of the current paper is the substantial influence of the cultivars on the content of these phytochemical compounds. This observation aligns with previous studies in the field. Indeed, a study by Roxana et al. [21] and Ali et al. [22] investigated the phytochemical profiles of different grape cultivars and found that distinct cultivars exhibited significant variations in flavonoid and phytoalexin compounds. This is consistent with the results of highly significant differences in phenolic acids, C13-norisoprenoids, and flavonoids phytoalexins between the 'Italia' and 'Bronx Seedless' cultivars. Environmental factors, genetic diversity, and the activation of specific metabolic pathways contribute significantly to these variations among different grape cultivars. The intricate interplay of these variables has been a subject of interest in various studies in this field. It's important to note that genetic diversity significantly shapes the phytochemical profiles of grape cultivars. Different grape cultivars have unique genetic characteristics that affect how various phytochemical compounds accumulate and are synthesized. This is consistent with the findings of He et al. [23] and Jeandet et al. [10], who discovered notable distinctions in phytoalexins and flavonoids among different grape cultivars, indicating the genetic basis for these differences. Moreover, environmental factors, including climate, vineyard management practices, and soil composition, also contribute to variations in fruit phytochemistry. In a study by Harb et al. [24], the impact of climate on the production of specific phytochemicals was emphasized, illustrating that different grape cultivars respond differently to environmental conditions, resulting in varying compound compositions. This highlights the importance of taking the grape-growing environment into account when understanding the composition of these compounds.

The current study highlights one of its most significant findings: the considerable impact of two grape cultivars on phytochemical compound content. This observation corresponds with previous studies in the field. Indeed, studies by Yang et al. [25] and Katalinić et al. [26] delved into the phytochemical profiles of various grape cultivars, revealing significant differences in flavonoids and phytoalexins among them. This is consistent with our results identification of marked distinctions in phenolic acids, flavonoids, phytoalexins, and C13-norisoprenoids between the 'Bronx Seedless' and 'Italia' varieites. These disparities arise from genetic diversity, environmental factors, and the activation of specific metabolic pathways in different grape varieites. Moreover, the study emphasizes the noteworthy influence of sampling time on grape phytochemical concentrations, a temporal variation well-documented in previous research. Indeed, Gómez-Plaza et al. [27] investigated how grape harvest timing affects phenolic compounds and found that the levels of flavonoids and resveratrol closely correlate with the maturation stage of berry. Our results align with this perspective, revealing substantial variations in phytochemical content across the six berry development stages (Table 1 and 2). This shows the significance of choosing the optimal harvest time to attain the desired phytochemical profiles in grapes for specific applications. Our findings indicate a significant interaction between grape cultivars and berry development stages, particularly for specific compounds like piceid, while for others, it is not statistically significant. This phenomenon has also been explored in earlier studies. Indeed, Teixeira et al. [28] noted the interaction between climate conditions and grape varieties, revealing that specific grape cultivars showed unique responses to environmental factors, leading to variations in the levels of specific compounds, including resveratrol derivatives. The observed interactions for certain compounds indicate the complex nature of grape chemistry and the need for a nuanced approach to understanding the factors influencing phytochemical content, while most compounds in the current study did not exhibit a statistically significant interaction. Regarding differences in anthocyanin content between two grape varieties, the current study reveals significant disparities in anthocyanin content between the 'Bronx Seedless' and 'Italia' grape cultivars, with 'Bronx Seedless' consistently exhibiting lower percentages of anthocyanins. Costa et al. [29] highlighted significant differences in anthocyanin content among various grape cultivars, and they found different levels of anthocyanins among these cultivars; Additionally, some were found to exhibit higher anthocyanin content than others. This highlights the vital role of selecting specific grape cultivars when trying to increase the concentration of specific phytochemicals in grape-derived products. Since different grape cultivars have unique anthocyanin biosynthesis pathways, these disparities in anthocyanin content may be attributed to genetic factors.

The use of advanced multivariate techniques such as PCA biplots and hierarchical clustering heat maps visually represented the complex relationships and distribution of phytochemical components between two grape cultivars and berry development stages. PCA biplots provided a comprehensive overview of the interrelationships between phytochemical components in the data set. In our results, the strong correlations of certain compounds with major components in PCA biplots indicate their importance in characterizing grape phytochemical composition (Fig. 2). These compounds play an important role in explaining the differences and similarities between grape cultivars. Hierarchical clustering heatmaps provide additional insights (Fig. 3). This method aided in recognizing trends and patterns in the dataset, ultimately providing a clearer understanding of the relationships among two grape cultivars, berry components, and berry development stages. By visually grouping similar samples based on the concentrations of phytochemical components, heatmaps helped identify clusters of samples that share similar compositional characteristics. In support of the utilization of hierarchical clustering heatmaps in exploring complex datasets and PCA, Pietrafesa et al. [30] Kaya [31] a emphasized the importance of these methods in their study of multivariate analysis of phytochemical data in plant sciences. Additionally, our PCA and Heatmap results are consistent with previous research reports [32, 33].

Conclusion

Current results revealed significant changes in the phytochemical composition of both grape cultivars across different berry development stages. We meticulously tracked the phytochemical and physical transformations in 'Bronx Seedless' and 'Italia' grape cultivars across six distinct berry development stages. Resveratrol, pterostilbene, flavonoids, anthocyanins, catechins, and aromatic compounds exhibited remarkable fluctuations. 'Bronx Seedless' grapes consistently exhibited lower percentages of various anthocyanins compared to 'Italia', underscoring the importance of both the grape cultivars and the berry development stages. These compounds, known for their contributions to grape quality, color, and potential health benefits, were influenced by both the grape cultivars and the berry berry development stages, was of paramount importance owing to their role in grape nutritional and coloration value. Our analysis encompassed various phytochemical compounds, namely C13-norisoprenoids, phytoalexins, flavonoids, phenolic acids, and shedding light on the intricate chemistry underlying grape development. In essence, this study indicated the intricate interplay between grape development, phytochemical composition, and cultivar choice. These findings are invaluable for grape growers, winemakers, and researchers seeking to optimize grape-based products and enhance their understanding of grape phytochemistry, contributing to the ongoing advancement of the viticultural and agricultural industries.