Introduction

The genus Allium has medicinal properties, is cultivated worldwide for use as a vegetable, ornamental, and spice, and is economically and medicinally important. Allium belongs to the Amaryllidaceae family of the order Asparagales. The genus Allium includes more than 920 species, among which onions, garlic, chives, leeks, and shallots are widely cultivated for their economic importance (Khandagale et al. 2020). Onion (Allium cepa) is among the most produced vegetables worldwide. In addition, it is considered one of the most important ingredients in almost every cuisine around the world. It has been used as a food and medicinal plant since ancient times due to its tremendous nutritional and medicinal properties (Mehrabi and Fazeli-Nasab 2012; Cramer et al. 2021; Sagar et al. 2022; Zhao et al. 2021). Onion is sometimes referred to as the “queen of the kitchen” due to its distinctive taste, unique and invaluable aroma, and various health benefits (Sagar et al. 2022). Garlic (Allium sativum L.), another important species of the genus Allium, is an important herbal, aromatic, and nutraceutical product known since ancient times. It is the second most widely distributed Allium species after onion (Manjunathagowda et al. 2017; Fazeli-Nasab et al. 2021).

Local cultivars constitute important starting materials to achieve the goals of plant breeding. However, the existing plant diversity is rapidly decreasing or even disappearing because of factors such as the preference for breeding cultivars over native germplasm due to economic concerns, natural disasters, uncontrolled urbanization and industrialization, and the consumption of plant diversity. As these threats became more visible, since the early nineteenth century many countries started to identify and protect their plant genetic resources and diversity levels (Balkaya et al. 2010). Plant genetic resources are valuable for producing food and other agricultural products, conservation of biodiversity, and sustainable development. At the same time, genetic resources are an important resource for adapting agriculture to ever-changing conditions such as global climate change and consumer preferences. But the loss of genetic resources is increasing at an astonishing rate, especially considering climate change (Hufford et al. 2019; Wang 2020; Anter 2023). Kelkit Valley, located in northeastern Türkiye, is one of the rare regions of the country that has not experienced serious pollution or damage to its natural structure because of diverse ecological features and the altitude differences in the basin (Dogan and Aslan, 2013).

Although many investigations on the biochemical characterization of onion and garlic genetic resources have been carried out to date, a comprehensive study in which both different genotypes and different organs are evaluated together in both species was not found. The hypothesis of this study (h1) is that the variation in the genetic resources of onion and garlic in the Kelkit Valley, which is a Turkish region with conservation of local diversity, will be significant both among genotypes and among the organs of these genotypes. This study has two main objectives. The first objective is to examine the variation among onion and garlic genotypes in the Kelkit Valley, which is rich in plant genetic diversity not yet suppressed by commercial cultivars, in terms of biochemical contents. The second objective is to observe the distribution of biochemical contents in onion and garlic genotypes among organs.

Materials and methods

Materials

A trip to the Kelkit Valley in the spring of 2021 was planned for the study. This valley, which is approximately 250 km long, serves as a crossroads between the Euro-Siberian and Irano-Turanian phytogeographic zones. Because of the merging of coastal and continental climates, these transition zones have distinct features. There are several settlements and small farms in these mountains, which have an average elevation of 1400–1500 m. Many villages and settlements are found in the valley’s floor section, between 280 and 650 m above sea level. During the first visit, the production areas to be sampled were determined, and 17 onion and 12 garlic genotypes were designated (Fig. 1). Samples were collected from the green sections of six plants in each group and transferred to the laboratory by cold chain for examination. During the harvest season, the same places were visited again, and six bulbs were collected from the same groups. Because garlic is a clonal plant and onion is an allogamous plant, different samples from the same population were collected at least 10 km apart to eliminate the chance of obtaining rametes from the same genotype. The experiment was planned according to the randomized plot design with 6 replications.

Fig. 1
figure 1

Some examples of the plant material used in the study

Full size image

Method

Onion and garlic bulbs and leaves were dried in an oven at 55 °C and 3 g of powdered sample were homogenized for 2 min with 25 ml of pure methanol and kept at + 4 °C for 16 h. The supernatant was collected from the samples centrifuged at 10,000 rpm for 20 min and stored at − 20 °C for analysis (Thaipong et al. 2006). Three different methods were used to determine the antioxidant capacity: copper (II) ion reducing antioxidant capacity determination (CUPRAC) method (Güçlü et al. 2006), 2,2-diphenyl-1-picrylhydrazil (DPPH) radical scavenging activity method (Thaipong et al. 2006), and ferric reducing antioxidant power method (FRAP) (Zhang et al. 2013). CUPRAC, DPPH, and FRAP activity were expressed as TEAC using the formula obtained from the graph for Trolox. The Folin-Ciocalteu method developed by Singleton and Rossi (1965) was used to determine the total amount of phenolic substances. The absorbance results of the samples were calculated as gallic acid equivalent mg/100 g dry matter (Thaipong et al. 2006). The amount of the ascorbic acid was determined by the method based on the reduction of 2,4-dinitrophenylhydrazine solution (Riemschneider et al. 1976). The method of Zhishen et al. (1999) was used to determine the total amount of flavonoids. The method of Ruch et al. (1989) was used to calculate hydrogen peroxide (H2O2) removal ability.

$${\text{\% H}}_{2} {\text{O}}_{2} {\text{Removal Activity }} = \, 1 \, {-} \, \left( {{\text{sample abs}}{\text{. at }}230{\text{ nm}}/{\text{control abs}}{\text{. at }}230{\text{ nm}}} \right) \times 100$$

The amount of soluble protein was determined using the method developed by Lowry et al. (1951). Carotenoid and chlorophyll amounts were determined for each sample according to the spectrophotometric method given by Porra et al. (1989). Chlorophyll and carotenoids were measured in a spectrophotometer device at wavelengths of 666, 653, and 470 nm and calculations were made according to the following formulas (Kirk and Allen 1965) (Fig. 2).

$${\text{Chlorophyll a }}\left( {\text{Cl a}} \right) = \, \left( {15.69.{\text{A}}666} \right) - \, \left( {7.340.A653} \right) \, \left( {3.2} \right)$$
$${\text{Chlorophyll b }}\left( {\text{Cl b}} \right) = \, \left( {27.05.A653} \right) \, - \, \left( {11.21.{\text{A}}666} \right) \, \left( {3.3} \right)$$
Fig. 2
figure 2

Laboratory images of chlorophyll and soluble protein analysis

Full size image

The significance of the difference between each mean result for bioactive compound analysis was determined by the analysis of variance. Post-hoc test (Tukey, HSD) and multiple means comparison were used to analyze the results between groups in case of significant differences (p < 0.05). Using the same program, the Pearson correlation test was applied to find the direct relationship between each independent variable. After standardizing the obtained data set, principal component analysis (PCA) was performed to understand which characteristic is more effective in explaining the variation. In addition, using these data, the similarity coefficients of the genotypes were found, and cluster analysis was performed by drawing a similarity dendrogram to observe the distribution.(Table 1).

Table 1 Onion (AC1-17) and Garlic (AS1-12) samples evaluated in the study and their locations
Full size table

Results

Summary statistics are presented in Table 2. According to the analysis of variance, the differences between the genotypes for all parameters were statistically significant (p < 0.05). The highest coefficient of variation was obtained for total flavonoids in onion bulbs (44.394%). The lowest coefficient of variation was found for the total dry matter content of garlic leaves (0.871%). As a result of the analysis of variance test, the effect of triple interaction (species x genotype x organ) on the differences between the means obtained for all parameters was significant.

Table 2 Summarized statistical data from biochemical analyses
Full size table

When the mean values including only the two species and their organs were compared, all results except DPPH and ascorbic acid values were higher in onion leaves than in onion bulbs. For garlic, leaves had higher values for hydrogen peroxide scavenging capacity, total phenolic matter, FRAP antioxidant capacity, CUPRAC antioxidant capacity, and total water-soluble protein content, while bulbs had higher values for the other 4 parameters.

The highest total phenolic content (36.291 GAE mg/g) was obtained from the bulbs of garlic genotype AS-2 (Table 3). However, there was no significant distribution between the organs of both species in terms of this trait. In fact, the lowest value was obtained from the leaves of another garlic genotype (AS-1) (15.338). This resulted in a higher coefficient of variation for the total phenolic content of garlic bulbs (28.362%). The lowest coefficients of variation among the parameters examined in the study were observed for FRAP and CUPRAC antioxidant values. The leaves of onion genotypes AC-1 and AC-3 had the highest FRAP values (14.982 and 14.280, respectively). The lowest FRAP value was found in the bulbs of garlic genotype AS-6 (10.947). Interestingly, the leaves of the same genotype had the highest hydrogen peroxide scavenging capacity value in the study. The highest CUPRAC antioxidant values were measured in the leaves of onion genotypes AC-3, AC-4, and AC-12 (0.142).

Table 3 Total phenolic matter and antioxidant analysis results in the analyzed samples
Full size table

In general, homogeneous results were observed among the genotypes within each species x organ group, while only the leaves of onion genotypes AC-10 and AC-11 had significantly lower CUPRAC content than the others (0.086 and 0.087, respectively). In terms of DPPH antioxidant capacity, the highest (genotype AC-11, 16.935) and lowest (genotype AC-2, 8.514) results were obtained in the leaves of onion genotypes (Table 3). In general, the variation in both bulbs and leaves of onion genotypes was higher than in garlic genotypes. The highest and lowest values for total water-soluble protein were also measured in onion genotypes. The highest value for this parameter was measured in the leaves of AC-1 onion genotype and the lowest value was measured in the bulbs of AC-14 onion genotype (Table 3).

The values of total water-soluble protein content in garlic leaves were very close to each other and the coefficient of variation was lower than others. The extract with the lowest ascorbic acid content was isolated from the leaves of onion genotype AC-17. The ascorbic acid content in the bulbs of this genotype was also lower than in the bulbs of many other onion genotypes. In general, the ascorbic acid level in garlic bulbs was higher than in the other sample groups. Moreover, the variation in the bulbs of both species was larger than in the leaves. The maximum hydrogen peroxide scavenging capacity (%) was found in the leaves of garlic genotype AS-6 when the biochemical contents of the bulbs and leaves of each genotype of both species were evaluated individually. The lowest values for this parameter were measured in the leaves and bulbs of genotype AC-10 (41.595 and 35.973%, respectively). The most remarkable aspect of H2O2 scavenging capacity was that the variation in garlic bulbs was much lower than in onion bulbs (5.696 and 12.245%, respectively) (Table 4). The most remarkable thing about the total dry matter content is that the results for the leaves of garlic genotypes were very close to each other and the coefficient of variation was very low (0.871%). The other noteworthy point is that leaves had higher dry matter content than bulbs in onion and bulbs had higher dry matter content than leaves in garlic (Table 4). This ratio was 0.63 (bulb/leaf) for onion and 4.00 (bulb/leaf) for garlic. The highest dry matter content was measured in the bulbs of genotypes AS-8 and AS-11 (40.114 and 42.376%, respectively). The highest total flavonoid content was found in the bulbs of garlic genotypes AS-2, AS-3, and AS-10 (219.00, 219.952, and 220.904, respectively). The highest mean value in the species/organ groupings belonged to the leaves of the bulbs (184.686).

Table 4 Results for H2O2 scavenging capacity, total soluble protein, ascorbic acid, total dry matter, and total flavonoids
Full size table

To determine the similarity of a total of 29 genotypes belonging to two different species, cluster analysis was performed using Euclidean similarity coefficients (Fig. 3). To obtain more detailed results, three different data sets were used in cluster analysis: A: biochemical contents of leaves only, B: biochemical contents of bulbs only, and C: biochemical contents of both leaves and bulbs. Cluster analysis according to the biochemical contents of the leaves of both species revealed two main groups. Eight garlic and 4 onion genotypes were in the first main cluster; 13 onion and 4 garlic genotypes were in the second main cluster. Both garlic and onion genotypes were found in subgroups within these main groups. Therefore, it was not possible to distinguish between onion and garlic genotypes based only on the biochemical content of the leaves. Cluster analysis based on the biochemical contents of bulbs also revealed two main groups. In this grouping, 7 garlic and 9 onion genotypes were included in the first main cluster; and 5 garlic and 8 onion genotypes were included in the second cluster. In the cluster analysis based on bulbs, within the second main cluster, genotypes belonging to both species were distributed in separate subclusters. Only genotype AC-12 broke this pattern. In the dendrogram obtained by processing bulb and leaf data together, two main groups were obtained. In the sub-groups of the second main group, onions and garlics clustered separately, just as in the bulb samples. Correlation analyses were carried out to examine the relationship between the results of the analyses performed on the bulbs and leaves for each species (Table 5). The highest positive correlation coefficient was observed between total water-soluble protein in the leaves and bulbs (0.760). In the same species, the largest negative correlation coefficient was measured between total dry matter content in leaves and FRAP antioxidant capacity in leaves (− 0.689). Only 7 correlation coefficients were statistically significant. In the correlation analysis for garlic genotypes, the highest correlation coefficient was found between FRAP antioxidant capacity in leaves and total phenolic content in bulbs (0.752). The highest negative correlation coefficient was determined between total dry matter content in leaves and DPPH antioxidant capacity in bulbs (− 0.791).

Fig. 3
figure 3

Cluster analysis dendrogram. A Using only leaf extracts, B Using only bulb extracts C Using leaf and bulb extracts

Full size image
Table 5 Pearson correlation coefficients for biochemical analysis of onion and garlic accessions*
Full size table

The degree of efficiency of all data obtained in explaining the variation in the gene pool was analyzed with principal component analysis (PCA) (Table 6). For this purpose, all data were first standardized. Principal component analysis revealed that the eigenvalues of five principal components were greater than "1". These five principal components explained 66.724% of the total variation. Parameters with eigenvalues |0.3| or greater in each principal component were effective in explaining the variation in the entire gene pool. In this context, five parameters were included in PC1, which explained 24.774% of the total variation. Three of these parameters are related to leaves (FRAP, ascorbic acid, total dry matter content) and two are related to bulbs (FRAP and total dry matter content). In the first three principal components explaining almost half of the total variation, FRAP, ascorbic acid, total dry matter, and total soluble protein had the highest eigenvalues for both bulbs and leaves. Among the principal components with eigenvalues greater than 1, all analysis parameters for bulbs were >|0.3| at least once. This result indicates a high contribution to explaining variation. However, in the analysis of leaves, the parameters for hydrogen peroxide scavenging capacity and total flavonoid content did not reach this value for any of the subprincipal components (PC1-PC8).

Table 6 Eigenvalues and percentage of variation for each principal component (PC)*
Full size table

Continuing principal component analysis, the distributions of both genotypes and parameters were examined on the coordinate plane. While the origin point on the coordinate plane is the area with least variation, the close positioning of the samples indicates that there are many similarities between them.

When the distribution of genotypes on the basic coordinate plane is examined according to the first two principal components, all garlic genotypes, except AS-2, are distributed homogeneously in the 4th region (+ X, − Y) (Fig. 4). Most onion genotypes are located in the 1st (+ X, + Y) and 2nd (− X, + Y) regions, and are generally clustered along the axis lines. The tendency for different species to have positions is more consistent with principal coordinate analysis than with cluster analysis. The clear species separation obtained in coordinate analysis could not be observed in all three dendrograms (based on leaves, bulbs, and leaves + bulbs) obtained in cluster analysis. AC-7, AS-2, AC-9 and AC-10 genotypes were located in the 3rd region (− X, − Y). The distributions of these four genotypes on the dendrograms were compared with the distributions on the coordinate plane. The grouping based on leaf data and the grouping resulting from basic coordinate analysis were parallel to each other. In the dendrogram drawn according to the results obtained from the bulb extracts, these four genotypes are distributed differently.

Fig. 4
figure 4

Distribution of the genotypes in the principal coordinates

Full size image

According to principal component analysis, the distribution of analyzed characteristics on the coordinate plane is more homogeneous than the genotype distribution (Fig. 5). No analysis parameters were located too close to each other, except for CUPRAC and total flavonoid in bulbs. The most striking feature of this distribution is the absence of grouping between bulb and leaf parameters. In the principal component analysis, FRAP in leaves (FRAP L), ascorbic acid in leaves (VC L), total dry matter in leaves (TDM L), FRAP in bulbs (FRAP B), and total dry matter in bulbs (TDM B), which are in the first sub-principal component with the highest variation and whose eigenvalue is greater than |0.3|, are located in the extreme regions on the coordinate plane. This reflects the effectiveness of these parameters in explaining variation in the onion and garlic gene pool.

Fig.5
figure 5

Distribution of the investigated parameters in principal coordinates (abbreviations as in Table 2)

Full size image

Discussion

Tahmas Kahyaoğlu (2021) reported that the average dry matter content in the bulbs of garlic grown in Türkiye was 38.80%; total phenolic content was 23.32, total flavonoid content was 8.63, DPPH antioxidant capacity was 47.87%, and FRAP antioxidant capacity was 0.45%. In the same study, total phenolic content was 10.34, total flavonoid content was 2.89, DPPH antioxidant capacity was 24.57%, and FRAP antioxidant capacity was 0.13% in dried above-ground plant residues left by farmers in the fields during harvest. Total phenolic concentrations in different kinds of garlic and Allium ursinum genotypes were compared by Mahmutovic et al. in 2013. According to their findings, A. ursinum, spring garlic, and fall garlic leaves had the highest concentrations of phenolic compounds. The spring-garlic bulb, autumn-garlic bulb, and A. ursinum bulb all had lower phenolic contents. However, in our study, the values for bulbs and green leaves were much closer to each other. This is assumed to be because fresh leaves were used in our study and dried leaves were used in Tahmas Kadıoğlu’s study. Nuutila et al. (2003) compared the bulbs and stems of one garlic and two onion cultivars in terms of various biochemical contents. At the end of the study, they reported that the stems (green parts) had much higher total phenolic content than the bulbs in both species; DPPH antioxidant capacity was higher in onion bulbs, while garlic stems had higher content. For both parameters, the data obtained from onions were higher than from garlic. In our study, the total phenolic content of leaves was higher for both species. However, DPPH content was higher in bulbs in both species. In addition, in our study, no statistical superiority of one species over the other could be determined, and samples with high values were emphasized according to species-organ interaction.

In a study conducted by Chaudhry et al. (2024) about drought and salinity stress in different onion genotypes, FRAP levels were between 8 and 13 mmol/g in non-stressed plants. These values are compatible with the values obtained from the bulbs in our study (12.536 mmol/g). Rekowska and Skupien (2009) investigated the effect of various cultural practices on the ascorbic acid content of winter garlic leaves and bulbs. They reported contents of 13.9 and 99.8 mg/100 g in the bulbs and leaves of the control group plants, respectively. The result for bulbs was similar to the average for all genotypes in our study (12.656 mg/100 g). However, our result was considerably lower than the result for leaves (10,000 mg/100 g). Bibi et al. (2022) compared the phenolic and flavonoid content in the leaves and bulbs of nine different onion cultivars. At the end of the study, similar to our results, the genotype-organ interaction was found to be significant and leaves of some genotypes and bulbs of others were found to be rich in the studied properties. Astaneh et al. (2018) reported the total phenolic matter content in garlic fresh leaves was 27.5 in the group without any treatment. Benkeblia (2005) compared the biochemical contents of four types of onions (green, yellow, red, and purple) and garlic extracts. Garlic (49 mg/100 g) had the greatest concentration, whereas green onions (30 mg/100 g) had the lowest. Other samples had contents of 34.7, 44, and 47.3 mg/100 g each for yellow, red, and purple onion bulbs, respectively. In our study, the total phenolic content in the leaves of the onions was higher than in the bulbs. The bulb values for garlic were higher than those of onions. In a study conducted by Selvan et al. (2018) on different species, total phenol levels in garlic cloves were reported as 34.2 GAE mg/g. These values are higher than the average values obtained from both bulbs and leaves of garlic genotypes in our study. (22.30 and 22.557 GAE mg/g, respectively).

Santas (2008) reported the FRAP value as 16.1 ± 0.10 mM Trolox/g in the Calçot de Valls onion cultivar, while Lisanti et al. (2015) reported the FRAP value as 9.19 ± 2.54 mM Trolox/g in the Rossa di Toscana cultivar. In our study, the average FRAP value for all onion samples was 12.99 (mM Trolox/g), the highest value was 14.98 (mm Trolox/g), and the lowest value was 11.73 (mm Trolox/g). This is consistent with our results. Frond et al. (2019) found that the average CUPRAC content of red onions was 0.113 when they analyzed antioxidants in vegetables including red chicory, purple sweet potato, black carrot, eggplant, and red onion. In our study, the lowest CUPRAC value was 0.086 mM Trolox/g and the highest value was 0.125 mM Trolox/g in onion leaves. These results are consistent with the study by Frond et al. (2019). Koca et al. (2016) reported that the DPPH radical scavenging activity of yellow onions was higher than that of white onions (42.25 µm Trolox/g and 11.57 µm Trolox/g, respectively). In our study, the average DDPH level in onion bulbs was 11.717 µm Trolox/g, and that of onion leaves was 10.176 µm Trolox/g. These results are lower than those found by Koca et al. in onions with yellow skins but parallel to those with white skins. Kumar et al. (2007) found the amount of water-soluble protein in onion bulbs was in the range of 0.98–1.04 g/100 g according to drought conditions. These results are consistent with the results obtained in our study.

Dinu et al. (2023) conducted various biochemical analyses on five different garlic and three different chive genotypes. As a result of the study, the total flavonoid level in garlic was between 223.50 and 353.58 mg quercetin/g, and the ascorbic acid level was between 10.46 and 15.68 mg/100 g. These values are compatible with the values obtained in our study. Researchers also reported that the bioactive profile of garlic depends on the age of the morphological organ and bioactive concentration decreases with maturity. In the study by Liguori et al. (2017), the average amount of ascorbic acid in the bulbs of five different onion genotypes was reported as 12,724 mg/100 g. In our study, the average ascorbic acid in the bulbs of 17 onion genotypes was 10.729 mg/100 g. These results are partially compatible, and the difference between them may be due to genotype and environmental differences. Mikulewicz et al. (2019) reported an average ascorbic acid level of 6.29 mg/100 g and a total dry matter level of 9.58% in onion bulbs grown without any treatment. While the ascorbic acid level was lower than our results, the total dry matter content is consistent with our results. Lemma et al. (2022) compared garlic bulbs and leek leaves in terms of various biochemical parameters. As a result of the study, the DDPH level in garlic bulbs was 17.25 µm Trolox/g, and the ascorbic acid level was 20.77 mg/100 g. These values are partially compatible with the highest DPPH level obtained in our study but above the ascorbic acid level (12.656 mg/100 g).

Conclusion

Türkiye is known to have a wide diversity of plant genetic resources due to its different ecological sub-regions, its location on major migration routes, and its long history of agriculture. In this study, local onion, and garlic genotypes were investigated in the Kelkit Valley, a protected region of Türkiye that has not yet been exposed to commercial vegetable cultivars. The diverse health effects and nutraceutical properties of garlic and onions can be attributed to a variety of bioactive compounds, particularly polyphenolic substances with strong antioxidant properties. Hitherto, research was undertaken to investigate the bioactive profiles of garlic and onion, which vary between genotypes. However, this study investigated whether this variation among genotypes also could vary according to the plant parts consumed.

The results indicated that there was high diversity in biochemical content among the onion and garlic genotypes. In addition, the results showed that genotype × plant organ interaction was significant for all the parameters examined. The parameters of FRAP in leaves, ascorbic acid in leaves, total dry matter in leaves, FRAP in bulbs, and total dry matter in bulbs proved to be more effective in explaining all variation in biochemical characterization research conducted in allium gene pools. According to these results, the Kelkit Valley may be an important source for future onion and garlic cultivar development. In addition, according to the results obtained, it is recommended that organ differences should be taken into consideration when evaluating the biochemical contents of onion and garlic in future research.

In conclusion, genotype differences in garlic and onion have a significant impact on chemical composition; therefore, breeding work and cultivar development should be carried out mainly focusing on the quality of the final product, in addition to climatic requirements and market needs.