Introduction

Grape (Vitis vinifera L.) from the Vitaceae family is one of the most valuable fruit crops cultivated in various regions of the world, including Asia, Europe and America, and has wide applications in the cosmetic, pharmaceutical, and food industries1. 80 million tons of grapes are produced every year, part of which is used as fresh grapes, juice and raisins2. A significant part of it is also used for winemaking as one of the most popular alcoholic beverages3. In the winemaking process, 0.2 kg of pomace is produced from each kilogram of grape fruit, of which 8 to 52% are seeds4. The amount of seed production from the winemaking process in the world is about 0.4 to 2.4 million tons of seeds per year5. Sustainable utilization of food wastes appears essential due to increasing economic value, reducing environmental issues, and providing valuable bioactive compounds6.

Grape seeds contain 7–20% oil, which consists of 90% unsaturated fatty acids (USFA) (linoleic acid and oleic acid) and 10% saturated fatty acids (SFA). It is also rich in phenolic compounds, flavonoids, procyanidins, catechins and tannins7. In addition to oil, grape seeds have high nutritional value due to their high content of lipids, vitamins, phytosterols, proteins, fiber and minerals, and biological effects such as anti-inflammatory, antifungal, antibacterial and antioxidant have been reported to improve health7,8. Grape seed extract and powder are used in the confectionery and bread industries and are added as an additive to biofilms (edible coatings) used to preserve food9,10.

Oil content and composition are influenced by cultivar and genotype, ripening time, environmental factors, and agronomic practices8,11. Studies of the effect of cultivar and cultivation region on olive oil content and composition showed that the fatty acid profile of the oil varies across cultivars and regions and is strongly influenced by these factors12. Variation in oil content and composition and phytochemical traits has been reported in different populations and cultivars of Xanthoceras sorbifolium13, Nigella sativa14, Ficus carica15, and Glycine max16.

Different cultivars can vary in oil content and its composition as well as nutritional values, therefore, efforts to identify superior cultivars with high oil content and composition and nutritional value are essential in the cosmetic-health and food industries17. The aim of this study was to investigate the oil content and composition, phenolic compounds and nutritional value of seeds of diverse grape cultivars in Iran in order to identify the best cultivar for producing raw materials required by the pharmaceutical and food industries.

Materials and methods

Plant material

The fruits of ten grape varieties (Rashmiri, Rashe, Mambrayma, Taefi, Ghazeozum, Champagne, Hosseini, Sahebi, Dastchin, and Rishbaba) were collected from Malekan City, East Azerbaijan Province, Iran. The fruits were harvested at the ripe stage. The varieties Hosseini and Rishbaba were harvested on July 12 and the Sahebi variety on September 18. The varieties Rashe, Mambrayma, Dastchin, Champagne, and Taefi were harvested on September 15, and the Ghazeozum variety was harvested on October 15. Prof. Ali Sonboli identified the different grape varieties and voucher specimens were deposited at the Shahid Beheshti University herbarium (MPH3241-MPH3251). The authors confirm that the necessary permissions to collect and cultivate the samples have been obtained. In addition, the present study complies with the IUCN Policy Statement on Research Involving Species at Risk of Extinction and the Convention on the Trade in Endangered Species of Wild Fauna and Flora. The geographical characteristics and physicochemical features of the soil at the collection site are presented in Table 1.

Table 1 Geographical positions, Climatic data, and soil properties of the cultivated region.
Full size table

Morphological features

Morphological and functional traits of ten grape cultivars were measured. Twenty fruits from each cultivar were randomly selected and various traits such as pericarp weight (g), fruit width (cm), fruit length (cm), fresh and dry seed weight (mg), and 1000 seed weight (g) were measured. Weight was measured using an electronic scale and digital caliper was used to measure fruit length and width18.

Oil extraction and fatty acid analysis

The crude oil content was assessed utilizing the maceration technique19. A sample of 1000 mg of dried and powdered seeds was combined with 10 ml of n-hexane and subjected to ultrasonication for a duration of 20 min (ultrasonication repeated three times). Then, the mixture was allowed to stand at a temperature of 23 ± 2 ˚C for 72 h and subsequently filtered using Whatman filter paper No. 1. The n-hexane solvent was then evaporated at ambient temperature. The extracted crude oil was stored in sealed, colored containers at − 18 ˚C until further evaluation. Fatty acid analysis is performed according to previously reported work20 as follows: 50 mg of the oil sample was mixed with a methanolic potassium hydroxide (KOH) solution at a ratio of 1:80 (w/v). The solution was prepared by dissolving 50 mg of KOH in 4 mL of methanol and added to the oil sample. The mixture was vortexed and centrifuged at 4000 rpm for 10 min. Fatty acids were first trans-esterified using KOH, followed by derivatization with boron trifluoride (BF₃) in methanol. This two-step process enabled the conversion of both free and esterified fatty acids into their corresponding methyl esters. Analyses were performed on an Agilent 7890 A Plus Gas Chromatograph (Agilent Technologies, Santa Clara, USA) equipped with a G4513A automatic liquid sampler and a flame-ionization detector (GC-FID). Separation was carried out on a 100-m capillary column (Agilent, CP-Sil 88 GC Columns, 100 m, 0.25 mm inner diameter, 0.20 μm thickness). Data acquisition and processing were performed using Agilent ChemStation software. The area percentages of each FAME were obtained. Finally, Peak identification was performed by comparing retention times with those of a commercial FAME standard mixture (GLC-462; Nu-Chek Prep, Elysian, MN, USA), which contains 28 known fatty acids. Due to the precise alignment of retention times between sample and standard chromatograms, peak identification was considered accurate without the need for calibration curves.

Quantification of phenolic compounds by HPLC

The extraction of phenolic compounds was performed according to the procedure outlined by Demir et al.21. In brief, 500 mg of powdered seeds underwent ultrasonication using an Elma S120H device from Germany, combined with 100 ml of a methanol/water solution (90:10, v/v) for a duration of 30 min. Subsequently, the mixture was centrifuged at 1,400 g for 10 min at a temperature of 4 ˚C using a Hettich Rotanta 460r centrifuge. The resulting extract was concentrated using a rotary evaporator at 35 ˚C. Finally, the extract was reconstituted in 1 ml of methanol and filtered through a 0.22 μm membrane.

Phenolic compounds were determined using a high-performance liquid chromatography (Knauer, Germany) photodiode array, with a Waters 2695 separations module equipped with a C18 column (250 × 4.6 mm) and a UV detector (Waters 2487). The mobile phase consisted of methanol (solvent A) and distilled water (solvent B). HPLC solvents both contained TFA (0.02%, v/v), and the flow rate was 0.5 ml/min. The gradient program was begun with 20%A: 80%B followed by 20:80 − 30:70 over 10 min, 30:70 − 50:50 over 10 min, 50:50 for 20 min, 50:50–100:0 over 2 min, 100:0 for 6 min, 100:0–20:80 over 7 min. The analysis was conducted under optimized conditions, with a 20 µl injection volume and a column temperature of 25 °C, to ensure accurate and reliable results. To detect phenolic compounds, the UV absorption peaks at 200–600 nm were investigated. Standard samples were used as references to identify compounds based on their characteristic retention times under equivalent analytical conditions. Pure standards of phenolic compounds were prepared from Sigma Aldric Co. (Buchs, Switzerland). Calibration curves were constructed by injecting standard mixture solutions at the seven concentrations of 2, 10, 50, 100, 250, 500, and 1000 ppm (Table 2). The results were presented in mg/g DW.

Table 2 Information of standard calibration curves for phenolic compounds.
Full size table

Evaluation of total phenol content (TPC) and flavonoid content (TFC)

Total phenolic compounds turned into quantified the usage of the approach described through Singleton et al.22,, which includes the Folin-Ciocalteu reagent. In summary, 25 µl methanolic extract (1000 ppm) and 125 µl Folin-Ciocalteu reagent, 100 µl sodium carbonate (7.5%) were taken in a test tube. The final volume was made up to 6 ml with distilled water. The solution was stored for 30 min in the dark. The absorbance was recorded at 765 nm using a spectrophotometer (Bio-Tek Instruments, Inc., USA). The results are expressed as mg gallic acid equivalents (GAE)/per g of dry weight (mg GAE/g DW).

Total flavonoids compounds were assayed based at the technique mentioned with the aid of Dewanto et al.23,, utilising aluminium chloride. Initially, 20 µl methanolic extract, 3.4 ml methanol (30%), 80 µl distilled water, 6 µl sodium nitrite (0.5 M), 6 µl aluminum chloride h (0.3 M) and 80 µl sodium hydroxide (1.0 M) was taken in a test tube and mixed well. The absorbance of the solution was determined against the reagent blank at 510 nm wavelength. The data were expressed as mg of quercetin equivalents (QE) per g of dry weight (mg QE/g DW).

Ferric reducing activity power (FRAP) assay

The approach outlined via Benzie and strain24 turned into hired. The fresh FRAP reagent was prepared by mixing 300 mmol/l acetate buffer (pH 3.6), 10 mmol/l TPTZ in 40 mmol/l HCl, and 20 mmol/l FeCl3 at 10:1:1 (v/v/v). The reaction mixture was kept for 30 min at 37 °C. The absorption was then measured at 593 nm. A standard curve was generated using ferrous sulfate solution in concentrations ranging from 0.25 to 8 mmol dm − 3 (0.5–10 mg/ml). The result was expressed as µmol Fe+2 per g of DW.

Elemental analysis

Initially, the seed samples of the studied grape varieties were digested using the Başgel and Erdemoğlu25 method with hydrogen peroxide and pure nitric acid for 3 h at 140 °C. Then, mineral elements were evaluated by ICP-MS Agilent 7500 A series. The mineral content was determined using the values ​​confirmed in the reference sample received from NIST (Gaithersburg, MD, USA).

Proximate composition analysis

The analysis of the samples for ash, protein, fat, and fiber was conducted following the AOAC guidelines26. The ash content was quantified by subjecting the cultivars to high temperatures in a muffle furnace. Fat extraction was performed using a Soxhlet (SOX406, HANON Systems, South Korea) and its amount was determined. The micro-Kjeldahl method was employed to estimate protein levels, with the amount of protein calculated by multiplying the nitrogen content by a factor of 6.25. Fiber was derived following the digestion of samples with boiling solutions of diluted acid and alkali. The carbohydrate content was ascertained by deducting the combined weights of ash, protein, fiber, and fat from the overall dry matter. The energy value was achieved using the formula: energy value (kcal/100 g) = (4 × g of protein) + 9 × (g of fat) + (4 × g of carbohydrates)27.

Data analysis

All experiments conducted in this research were carried out in triplicate. The results obtained are presented as means ± standard deviation (SD). To assess significant disagreement among the cultivars regarding the measured traits, one-way analysis of variance (ANOVA) was performed using SAS 9.4. A post-hoc analysis was conducted employing Duncan’s test with a significance level set at p < 0.05. The biplot was generated using Origin software version 2022, while cluster analysis was performed utilizing the Euclidean distance coefficient and Ward’s method. Additionally, R software was utilized to create the correlation plot.

Results and discussion

Morphological characteristics

Significant diversity in morphological traits was observed among diverse Iranian grape cultivars. The coefficient of variation (CV %) for fruit width, number of seeds per fruit, and pericarp fresh weight were 57.04, 35.49, and 27.28%, respectively (Table 3). A high CV indicates higher variation in that trait, which allows for better selection for the breeder. The studied grape cultivars had significant differences in morphological traits (P < 0.05). The highest fresh (6.70 g) and dry (1.77 g) pericarp weights were observed in Dastchin cultivar, and the lowest fresh (1.68 g) and dry (0.44 g) pericarp weights were observed in Champagene cultivar. Dastchin cultivar had the highest fruit length (2.63 cm) and fruit width (2.23 cm), and the lowest were observed in Champagene cultivar. Rashe cultivar showed the highest fresh (78.00 mg) and dry (56.36) seed weights, while Mambrayma cultivar had the lowest fresh (36.73 mg) and dry (31.50 mg) seed weights. The highest number of seeds per fruit (3.36) was found in Rashmiri cultivar, although there was no significant difference with Rashe cultivar, but the minimum number of seeds per fruit (1.00) was observed in Mambrayma cultivar. The minimum (36.73 g) and maximum (78.00 g) thousand seed weights were obtained in Rasha and Mambrayma cultivars, respectively (Table 3). Grape fruit size and phenotypic variation as an inherent trait in all species is due to environmental, genetic interactions and their mutual effects28. The variation in fruit and seed traits of grape cultivars has been previously explained by other studies29,30,31, which is consistent with the results of this study. Differences in morphological and functional traits of different cultivars can be due to genetics, cultivar type, age, regional climatic conditions, and geographical region30.

Table 3 Comparison of different morphological traits among Vitis vinifera cultivars.
Full size table

Oil content and fatty acid profile

The seed oil content of diverse grape cultivars is presented in Fig. 1. The oil content varied from 4.03 to 18.01% among the cultivars. The cultivars Hosseini, Ghazelozum, Sahebi, and Rashmiri had the highest oil content (18.01%, 14.66%, 13.12%, and 12.04%, respectively). The lowest oil content (4.03%) was observed in the cultivar Mambrayma. Consistent to the results of Beveridge et al.32, who investigated the oil content of grape seeds from eight Canadian wine varieties, revealing a range between 5.9% and 13.6%. Similarly, Sabir et al.33 assessed the oil content of Turkish grape seed oils across 21 varieties, which varied from 7.3 to 22.4%. Ovcharova et al.34 documented the seed oil content in Bulgarian grape varieties, noting white varieties yielded between 11.60% and 16.50% and red varieties yielded 15.70% and 13.90%. In a study conducted in China, Wen et al.35 found that the Merlot variety produced the maximum yield of grape seed oil at 15.92%, while Chardonnay had the lowermost yield at 13.71%. Furthermore, Harbeoui et al.36 reported that nine Tunisian grape varieties exhibited seed oil content ranging from 6.31 to 12.7%. This suggests a potential variation in oil content due to factors, such as geographical location, environmental conditions, genetic diversity, and cultivation practices37.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Histogram of seed oil content among Vitis vinifera L. cultivars. The mean comparisons were performed using the Duncan test at P ≤ 0.05 significant level. Means followed by the same letter(s) are not significantly different.

GC analyzed seed oil from diverse grape cultivars and the results are shown in Table 4. A representative chromatogram illustrating the fatty acid profiles across different cultivars is provided in Fig. 2. Eight fatty acids (palmitic acid, myristoleic acid, stearic acid, margaric acid, linoleic acid, oleic acid, arachidonic acid, arachidic acid,) have been identified in the oil of different cultivars, constituting 99.5–100% of the seed oil. The SFA and USFA ranged from 20.9% (in Taefi cultivar) to 74.8% (in Sahebi cultivar) and from 21.6% (in Sahebi cultivar) to 78.7% (in Taefi cultivar), respectively. The proportion of monounsaturated fatty acids (MUFA) ranged from 3.7% to. 16.8%, while the levels of polyunsaturated fatty acids (PUFA) fluctuated between 17.9% and 61.1%. Among the SFA examined, margaric acid and palmitic acid were predominant, with a concentration ranging from 1.6 to 33.2% and from 2.8 to 25.2%, respectively, particularly prevalent in the Rashmiri and e Hosseini cultivars. Additionally, linoleic acid, a significant PUFA, exhibited concentrations from 17.9 to 59.4%, with the highest levels recorded in the Taefi cultivar. Göktürk Baydar et al.38reported the linoleic, oleic, palmitic and stearic contents of grape seed oil as 63.33–71.37%, 16.15–21.63%, 7.42–10.24%, and 2.95–4.68%, respectively. Harbeoui et al.35, reported linoleic acid (64.77–75.37%) as the topmost fatty acid in the seed oil of Tunisian grape cultivars, which is consistent with our results. To et al.39 stated that MtWRI transcription factor homolog regulates fatty acid biosynthesis in M. orbicularis. Tang et al.40 also reported that overexpression of some genes and their mRNA levels are involved in oil accumulation and regulating of fatty acid biosynthesis in peanut seeds. The obtained variations in the fatty acid profile of the studied samples can be related to the transcription factors controlling the biosynthesis of these compounds.

Table 4 Fatty acid profiles of the studied Vitis vinifera cltivars.
Full size table
Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

A typical chromatogram of the fatty acids from Taefi (a), Dastchin (b), and Rashe (c) cultivars of Vitis vinifera L.

Phenolic compound profile

Phenolic compounds have traditionally been considered in preventing and treating many diseases as antioxidants due to their important biologically active materials that are good for human health41. The main phenolic compounds of the seed extracts of different grape cultivars were gallic acid, quercetin, p-coumaric acid, vanillic acid, caffeic acid, and kaempferol (Fig. 3). The maximum level of gallic acid (1.53 mg/g DW) was obtained in the Mambrayma cultivar, while the minimum level (0.23 mg/g DW) was obtained in the Rishbaba cultivar. The maximum levels of caffeic acid, quercetin and kaempferol were obtained in the Rashmiri cultivar, 0.07, 0.54 and 0.23 mg/g DW, respectively. The mininun levels of quercetin (0.09 mg/g DW) and kaempferol (0.02 mg/g DW) were obtained in the Hosseini cultivar. The level of vanillic acid varied between 0.02 and 0.15 mg/g DW, with the Maximum and lowermost levels being in the Taefi and Dastchin cultivars, respectively. Rasha cultivar had the highest (0.12 mg/g DW) p-coumaric acid and Hosseini and Rishbaba cultivars had the lowest (0.01 mg/g DW). The compounds p-coumaric acid, caffeic acid, vanillic acid, ferulic acid, kaempferol and quercetin in grape seed extract have been expressed by other studies42,43. The degree of variation in phenolic compounds of the different grape cultivars studied may be due to the type of cultivar or genetics, extraction and analysis method, location of seed preparation and climatic factors44.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

The phenolic compound profiles of Vitis vinifera L. cultivars.

TPC, TFC and antioxidant activity

The figure shows the TPC, TFC and antioxidant activity content of seed extracts of different grape cultivars. There was a significant disagreement among the cultivars in terms of the mentioned traits (p < 0.05). The maximum TPC (8.83 mg GAE/g DW) and TFC (4.64 mg QE/g DW) was obtained in Rashmiri cultivar, followed by Rasha cultivar with the maximum TPC (8.79 mg GAE/g DW) and TFC (3.15 mg QE/g DW). The lowermost TPC (2.21 mg GAE/g DW) and TFC (0.97 mg QE/g DW) were obtained in Hosseini and Ghazelozum cultivars, respectively (Fig. 4a, b). Krasteva et al.43 reported the TPC of seeds of different grape cultivars to be between 70.06 and 111.2 mg GAE/g DW. Castro-Lopeza et al.45 reported the TPC of seed extracts of different grape varieties as 107.8–10.7 mg GAE/g DW. In addition, in another study, the TPC of seed extracts of different grape varieties was determined as 8.79–11.27 mg GAE/g46. The TFC of seed extracts of different grape varieties was reported as 40.05–52.01 mg QE/g DW in one study44 and in another study, as 6.9–1.25 mg QE/g DW47. The results of evaluating the antioxidant activity of seed extracts of different cultivars by FRAP method showed that the maximum antioxidant activity was determined in Rashmiri (59.63 µmol Fe2+/g DW) and Rasha (49.95 µmol Fe2+/g DW) cultivars, and the lowest (7.17 µmol Fe2+/g DW) was related to Ghazelozum cultivar (Fig. 4c). Seed extracts of different grape cultivars showed different antioxidant activity, which is consistent with the study of Guaita and Bosso48 and Ghouila et al.49. Based on the results, seed extracts that contained more compounds also had higher antioxidant power. Polyphenolic compounds can capture free radicals due to their chemical structure and form complexes with metal ions. Therefore, these compounds showed good antioxidant activity. A positive and significant correlation among phenolic compounds and antioxidant activity has been expressed in other studies50. There are some methods to produce polyphenolic compounds in plants and various mechanisms for their distribution across different plant structures. Genetics, environmental conditions, climatic conditions, and the solvent used in extraction are some factors that influence the level of phenolic and flavonoid compounds derived from plants51.

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Histogram of total phenol content (a), total flavonoid content (b), and antioxidant activity (c) among Vitis vinifera L. cultivars. The mean comparisons were performed using the Duncan test at P ≤ 0.05 significant level. Means followed by the same letter(s) are not significantly different. .

Proximate composition

Figure 5 shows the approximate composition of seeds including carbohydrates, protein, fat, fiber and energy value of different grape varieties. Significant disagreement were observed between different grape cultivars in terms of approximate composition content of seeds. The carbohydrate varied from 13.23 g/100 g DW in Hosseini to 19.37 g/100 g DW in Rashmiri. The seed protein content of different varieties varied from 6.24 to 9.72 g/100 g DW, with the lowest and maximum being determined in Ghazeozum and Rishbaba cultivars, respectively. Hosseini cultivar showed the highest (15.40 mg/100 g DW) fat content and Mambrayma cultivar showed the lowest (9.62 mg/100 g DW) fat content. The maximum (42.87 mg/100 g DW) and lowest (38.41 mg/100 g DW) fiber content was determined in Mambrayma and Rishbaba cultivars, respectively. The maximum energy value was achieved in Rashmiri cultivar (231.96 Kcal/100 g), followed by Rasha (230.63 Kcal/100 g), Ghazelozum (221.55 Kcal/100 g), and Hosseini (217.68 Kcal/100 g). In previous research, the protein, lipid, carbohydrate and fiber content of grape seeds were reported to be 8.7–9.8, 10.45–16.73, 18.2–19.8 and 40.2–43.7 g/100 g, respectively52,53. Nikolidaki et al.54 reported the protein and lipid content of grape seeds to be 2.2% and 0.43%, respectively. Fiber, which plays a substantial role in the prevention of cardiovascular diseases, gastrointestinal diseases, diabetes and some cancers, is about 9 times higher in this study than the amount found in the US Food Database (4.5 g/100 g)55. The fiber content of grape seeds cultivated in Morocco was reported by Kalili et al.56 to be 13.93 g/100 g.

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Histogram of proximate composition for ten cultivars of Vitis vinifera L.

Mineral composition

There was a significant difference among different grape cultivars in terms of seed mineral content (p < 0.05). The amount of mineral elements in the seeds of different cultivars is shown in the Fig. 6. Macro elements such as potassium, phosphorus and calcium and micro elements such as magnesium and iron were identified in the seeds of diverse cultivars. The highest amount of potassium was found in the Champagne cultivar (8.34 mg/g dry weight) followed by Rashmiri (7.32 mg/g dry weight), mamabrayma (6.61 mg/g dry weight) and Rasha (6.40 mg/g dry weight). The lowest amount (3.82 mg/g dry weight) of potassium was found in the Taefi cultivar. The amount of phosphorus varied from 2.34 mg/g dry weight in the Sahebi cultivar to 3.74 mg/g dry weight in the Champagne cultivar. The cultivars Mambrayma, Rashmiri and Rasha had the maximum calcium amount with 2.31, 2.17 and 2.04 mg/g DW respectively and the lowest calcium content (1.05 mg/g DW) was observed in the cultivar Hosseini. The highest magnesium and iron content was determined in the cultivar Champagne. The lowermost magnesium amount (0.23 mg/g DW) was in the cultivar Rishbaba and the cultivars Sahebi, Taefi and Rishbaba had the lowest iron content (0.01 mg/g dry weight). The diversity of mineral elements in the seeds of Turkish grape cultivars (B, Al, Co, Ca, Cr, Mo, K, Fe, Mg, Na, Mn, S, P, Se, S, and Zn) showed that the content of potassium, calcium, magnesium, sodium, phosphorus and S in Turkish grape seed samples was generally very high57. In addition, Vrcek et al.58 reported that the uttermost microelements in grape seeds were iron, copper, zinc, manganese and aluminum. Several factors such as soil type, pH, the amount of mineral compounds in the soil and the plant’s ability to absorb can affect the amount of mineral compounds in the plant59. This study showed that grape seeds, due to their appropriate nutrients, could play a substantial role in the diet and prevent various diseases.

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Histogram of mineral composition for ten cultivars of Vitis vinifera L.

Correlation, cluster analysis and PCA

The outcomes of the analysis concerning the interrelations among traits, as determined by the Pearson correlation coefficient, are illustrated in the accompanying Fig. 7. Positive communication are depicted in blue, while negative correlations are indicated in red. Notably, a positive and statistically significant relationship was observed between fat and carbohydrates with energy. Conversely, oil content demonstrated a negative and statistically significant relationship with fiber, phosphorus, calcium, magnesium, iron, gallic acid, and p-coumaric acid. Additionally, a significant positive relationship was identified among oil content, fat, and energy. The dry weight of seeds had a positive and significant correlation with fiber, potassium, calcium, magnesium, iron, and fresh seed weight. Furthermore, carbohydrates, energy, calcium, TPC, TFC, gallic acid, and caffeic acid all showed a positive and significant correlation with antioxidant activity. Numerous research have documented a significant positive association among antioxidant activity and phenolic compounds, aligning with the findings of this investigation50,60.

Fig. 7
Fig. 7The alternative text for this image may have been generated using AI.
Full size image

Pearson correlation representation of morphological traits, seed oil content, proximate and mineral composition of Vitis vinifera L. cultivars.

Principal component Aanalysis (PCA) is a widely utilized statistical technique in metabolomics aimed at reducing dataset dimensionality while highlighting significant variance and patterns61. It transforms correlated variables into a smaller set of orthogonal variables known as principal components (PCs), which serve as axes in a new coordinate system where the first PC captures the greatest variance. PCA results are often visualized through “scores plots,” which display sample similarities and differences, and “loadings plots,” which identify the variables contributing to sample separation. As an unsupervised method, PCA finds application across various research fields, including agricultural science62. In this research, PCA was conducted focusing on oil content and composition, phenolic compounds, as well as mineral and proximate composition across various grape cultivars. Table 5 illustrates the relationships among the cultivars and the studied traits. The analysis revealed that the first eight PC accounted for 98.42% of the variance among the cultivars. Specifically, PC1, PC2, and PC3 contributed 26.52%, 19.82%, and 16.08% of the total variance, respectively. The first component was predominantly influenced by traits such as calcium, magnesium, carbohydrates, potassium, phosphorus, iron, FRAP, seed dry weight, moisture content, fiber, TFC, caffeic acid, and oil content. The second component was primarily affected by quercetin, kaempferol, oleic acid, myristoleic acid, p-coumaric acid, gallic acid, proteins, margaric acid, fruit length, and palmitic acid. The third component highlighted traits such as energy, ash content, fat content, stearic acid, seed fresh weight, arachidonic acid, and TFC with the highest coefficients (Table 4). This analysis serves to elucidate the key differentiating factors among the populations under study. However, according to the PCA and despite the same cultivation conditions, the similarity of the phytochemical composition values among the cultivar, and the difference in the composition values in the cultivars are more influenced by genetic factors and less related to their geographical origin. The results can provide useful information for breeding program and exploitation of these potential compounds in the field of food, pharmaceutical, cosmetic and health industries63.

Table 5 Eigenvalues of the principal component axes from the multiple regression analysis of the studied parameters in Vitis vinefera cultivars.
Full size table

The results of the cluster analysis conducted on various cultivars based on oil content and composition, phenolic compounds, as well as mineral and proximate composition are presented in Fig. 8. The cluster analysis revealed that the cultivars could be classified into four primary groups. The Rasheh cultivar was allocated to the first group (I), while the Rashmiri and Mambrayma cultivars were grouped together in the second group (II). The third group (III) comprised the Dastchin, Champagne, Hosseini, and Ghazelozum cultivars. Lastly, the fourth group (IV) included the Rishbaba, Taefi, and Sahebi cultivars. A biplot generated from PC1 and PC2 illustrated the diversity among the cultivars and corroborated their classification into four distinct groups (Fig. 9). In the study of Shakuri et al.41, PCA was used to evaluate different ecotypes of Iranian oak, in which the studied phytochemical and morphological traits were classified into the first two principal components, which together accounted for 68.53% of the total variance. In another study on Medicago species, morphological and phytochemical traits were classified into the first two principal components using PCA, which accounted for 78.06% of the total variance63. This biplot reinforced that the findings related to oil content and composition, phenolic compounds, as well as mineral and proximate composition were largely consistent with those derived from the cluster analysis.

Fig. 8
Fig. 8The alternative text for this image may have been generated using AI.
Full size image

Cluster analysis of Vitis vinifera L. cultivars based on morphological traits, seed oil content, proximate and mineral composition using Euclidean distances.

Fig. 9
Fig. 9The alternative text for this image may have been generated using AI.
Full size image

Bi-plot graph for the first and second principal components based on morphological traits, seed oil content, proximate and mineral composition for ten cultivars of Vitis vinifera L.

Conclusion

The current investigation revealed significant variation among Iranian grape cultivars regarding their morphological characteristics, mineral and proximate composition, as well as phytochemical properties, including phenolic acids and fatty acids. This research provides new insights into the fatty acid profiles of grape seeds and the oils extracted from them, along with detailed information on the mineral and nutritional composition and phenolic amount of seeds from Iranian grape varieties. Among the cultivars examined, Hosseini, Ghazelozum, Sahebi, and Rashmiri exhibited the highest levels of oil content. Linoleic and margaric acids were identified as the uttermost fatty acids present in these cultivars. Additionally, the seeds of the Iranian grape varieties demonstrated a commendable level of minerals, fats, proteins, and carbohydrates. Furthermore, the TPC, TFC, and phenolic compounds in the seeds significantly contributed to their antioxidant potential. The high genetic diversity among the studied grape cultivars allows us to use superior genotypes in future breeding programs to achieve proper cultivars with desirable growth parameters, high adaptability, and rich secondary metabolite contents.